THE MODERN

ASPHALT PAVEMENT

CLIFFORD RICHARDSON
ft

M. AM. Soc. C. E.; PROPRIETOR NEW YORK TESTING LABORATORY

SECOND EDITION, REVISED AND ENLARGED
THIRD THOUSAND

NEW YORK

JOHN WILEY & SONS

LONDON CHAPMAN & HALL, LIMITED

1912

V r

CLIFFORD RICHARDSON

THE SCIENTIFIC
ROBERT DRUMMONO AND COMPANY
BROOKLYN, N. Y.

PREFACE TO FIRST EDITION.

THE present work being designed for a rather wide class of
readers necessarily includes a large collection of data in regard
to the chemistry of asphalt and the technology of the industry,
which is of interest only to civil engineers, asphalt experts and
those who have made a special study of the subject. The general
reader is recommended to omit Chapters III to VIII, XII, and
most of XVI, or to confine his attention to the resumes of them
which are presented at the end of each. The property holder and
taxpayer will find the conclusions which will prove of most interest
to him in Chapters I, II, and XIV, Summaries of III, XIII, XVI,
and XVII to XXIV 1 relating to the construction of pavements and
the causes of their deterioration. The relative merits of various
asphalts for paving purposes are given in a compact form at the
end of Chapter XIV. The importance of the action of water on
asphalt in Chapter XXIII 2 The resume at the end of each Chapter
will generally furnish the reader at a glance with an opportunity
of determining whether the details in that Chapter appeal to his
interest and intelligence.

THE AUTHOR.

NEW YORK, February 18, 1905.

1 Second Edition, Chapter XXV. 2 Second Edition, Chapte' XXIV.

iii

248445

PREFACE TO SECOND EDITION.

THE very flattering reception given to the first edition of
this work, resulting in an exceedingly large sale within the last
three years, has pointed to the very general interest in the
subject treated of, by the persons for whom the book was
especially intended. In consequence of the very considerable
developments in the industry, it has seemed desirable, after the
exhaustion of the second impression of the first edition, to revise
the manuscript completely, and to bring it up to date. This has
been done with the results given in the accompanying pages.

The writer is indebted to Mr. C. N. Forrest for a revision of the
matter regarding the methods of analysis which are described,
and for many other suggestions.

TABLE OF CONTENTS.

INTRODUCTION 1

PART I.
THE FOUNDATION AND INTERMEDIATE COURSE.

CHAPTER

I. THE FOUNDATION OR BASE 3

II. THE INTERMEDIATE COURSE . - 19

PART II.

THE MATERIALS CONSTITUTING THE ASPHALT SURFACE

MIXTURE.

III. THE MINERAL AGGREGATE 29

IV. FILLER, OR DUST 87

V. THE NATURE OF THE HYDROCARBONS WHICH CONSTITUTE THE

NATIVE BITUMENS 97

VI. CHARACTERIZATION AND CLASSIFICATION OF THE NATIVE

BITUMENS 110

PART III.
NATIVE BITUMENS IN USE IN THE PAVING INDUSTRY.

VII. DIFFERENTIATION AND CHARACTERIZATION OF THE NATIVE

BITUMENS 115

VIII. PETROLEUMS 127

IX. THE SOLID BITUMENS 147

vii

Viii TABLE OF CONTENTS.

CHAPTER PAGB

X. INDIVIDUAL ASPHALTS 156

XI. SOLID NATIVE BITUMENS WHICH ARE NOT ASPHALT 208

XII. ASPHALTIC SANDS AND LIMESTONES 221

XIII. RESIDUAL PITCHES, OR SOLID BITUMENS DERIVED FROM AS-

PHALTIC AND OTHER PETROLEUMS 256

XIV. COMPARISON OF VARIOUS NATIVE ASPHALTS AND THEIR RELA-

TIVE MERITS FOR PAVING PURPOSES 278

PART IV.
TECHNOLOGY OF THE PAVING INDUSTRY.

XV. REFINING OF SOLID BITUMENS 291

XVI. SURFACE MIXTURES 313

XVII. ASPHALTIC OR BITUMINOUS CONCRETE 375

XVIII. ASPHALT BLOCKS 389

XIX. THE PROCESS OF COMBINING THE CONSTITUENTS INTO A SUR-
FACE MIXTURE 395

PART V.

HANDLING OF BINDER AND SURFACE MIXTURE ON THE

STREET.

XX. THE STREET 412

PART VI.
THE PHYSICAL PROPERTIES OF ASPHALT SURFACES.

XXI. RADIATION, EXPANSION, CONTRACTION, AND RESISTANCE TO

IMPACT 425

PART VII.

SPECIFICATIONS FOR AND MERITS OF ASPHALT PAVEMENT.

XXII. SPECIFICATIONS 435

XXIII. THE MERITS OF THE MODERN SHEET-ASPHALT PAVEMENT ... 455

XXIV. ACTION OF WATER ON ASPHALT PAVEMENTS 460

TABLE OF CONTENTS.

PART VIII.

CAUSES OF THE DEFECTS IN AND THE DETERIORATION OF
ASPHALT SURFACES.

CHAPTER PAGE

XXV. DEFECTS IN AND DETERIORATION OF ASPHALT PAVEMENTS . . 471
XXVI. MAINTENANCE OF ASPHALT PAVEMENTS 504

PART IX.
CONTROL OF WORK.

XXVII. INSTRUCTIONS FOR COLLECTING AND FORWARDING TO THE
LABORATORY SAMPLES OF MATERIALS IN USE IN CON-
STRUCTING ASPHALT PAVEMENTS 509

XXVIII. METHODS EXPLOYED IN THE ASPHALT-PAVING INDUSTRY FOR
THE CHEMICAL AND PHYSICAL EXAMINATION OF THE

MATERIALS OP CONSTRUCTION 519

XXIX. SOLVENTS 589

XXX. EQUIPMENT OF A LABORATORY FOR CONTROL OF ASPHALT

WORK 595

APPKNDIX. 599

THE MODERN ASPHALT PAVEMENT.

INTBODUCTION.

THE object of this work is to demonstrate the nature of asphalt
pavements and the causes of defects in them, to bring about im-
provement in the methods of their construction, and to show how
this can be done.

During an extended experience in the asphalt paving industry,
which has included the inspection of the construction of asphalt
pavements on behalf of a large city and the technical supervision
of the work of several prominent companies which contract to
lay them, it has been forced upon the attention of the writer that
engineers, and others who are interested in obtaining the best
results, have not been made sufficiently acquainted with the
technology of the industry and with the importance of some of
the engineering details involved to enable them to differentiate,
at the time that the pavement is being laid, or even on its com-
pletion, between good, bad, or medium work. Cities have, con-
sequently, been obliged to rely on the statements and good faith
of contractors, with the result that many asphalt pavements have
eventually proved unsatisfactory, although when completed they
were, to all outward appearance, of good quality a condition which
might have been readily avoided either by an intelligent super-
vision of the materials in use and the manner of handling them,
or by a change in the form of construction.

It is proposed, therefore, to describe in the following pages
the forms of construction which have been shown by experience

2 THE MODERN ASPHALT PAVEMENT.

to be the most satisfactory, the character of the materials enter-
ing into the composition of asphalt pavements, the most refined
methods used in the industry at the present day and the reasons
which have led to their adoption, in order that engineers and
others who are responsible for the supervision and character of
such work may be able to distinguish between that of good and
that of inferior quality. To this will be added specifications for
asphalt pavements to meet various environments and use, and
something as to their maintenance and the causes of their deterio-
ration.

The conclusions which are advanced are the results of twenty
years' experience in the industry by the writer with pavements in
over one hundred cities in the United States and in several in
England, Scotland, and France, involving the construction of
between twenty and thirty million yards of surface.

PART I.

THE FOUNDATION AND INTERMEDIATE COURSE.

CHAPTER I.
THE FOUNDATION OR BASE.

THE modern asphalt pavement in its perfected state is the
evolution of thirty years of experiment and experience. It seems
unnecessary here to give a history of the origin of this form of
pavement on the Continent of Europe, or of the earliest experi-
ments in the United States by De Smedt with artificial mixtures
of sand and asphalt, as this information is readily available in numer-
ous publications. It is sufficient to take the subject up at as late
a date as 1894, when the first successful effort was made to place
the industry on a rational basis as distinguished from the rule-of-
thumb methods previously in vogue, and to follow it down to the
most recent practice.

An asphalt pavement consists essentially of a foundation or
support for the surface which is to carry the traffic, itself supported
by the soil, and a surface consisting of a mineral aggregate cement-
ed together with asphalt to protect the base from wear and disinte-
gration, between ;svhich is commonly interposed either a course of
broken .stone coated with bitumen, known as binder, or some sub-
stitute for it, such as a cushion or separate course of the surface
material, or a paint coat of bitumen dissolved in naphtha. These
three elements of the pavement will be considered in turn.

4 THE MODERN ASPHALT PAVEMENT.

The Subsoil Base. As the foundation of a pavement, of any
kind, is placed upon the soil and supported by the latter, it is a
matter of vital importance that this latter support should be
adequate, and it will only prove adequate if the subsoil is not
subject to displacement from settlement or frost and is thoroughly
drained.

With sandy soils which are well compacted and which, from
their nature, are well drained and dry there is no difficulty in the
preparation of a satisfactory subgrade. Trenches in such soils, if
they occur, can be solidly refilled with the aid of water. If the
subgrade is a true sand, as in the neighborhood of seabeaches, it
may be necessary, in order to compact the surface, to spread a
course of gravel between the sand and the base in order to be able
to roll it properly.

With clay or heavy soils it is much more difficult to prepare
a satisfactory subgrade, especially if this is likely to be subjected
to the action of frost and if the original soil has been much dis-
turbed by trenches, or if fills occur. If the subgrade consists
of the original soil in situ, the only consideration necessary is its
proper drainage, unless there are soft spots due to local causes,
in which case they must be excavated and removed, with the sub-
stitution of firm for the softer material. The satisfactory back
filling of trenches in a heavy soil is a difficult matter. The use
of water is a disadvantage in such work. Heavy soils absorb
and hold it tenaciously and thus prevent thorough compaction,
final settlement only taking place after the completion of the
pavement. Trenches in such a soil should be carefully and slowly
tamped in thin layers.

The proper drainage of heavy clay soils is an essential feature
in the construction of a satisfactory pavement, especially where
these soils are apt to be thrown or cracked by frost in very cold
climates, a condition which may be illustrated by that occurring
in Manitoba, where cracks frequently open in the ground in
winter, from four to six inches wide, and which would cause cor-
responding cracks in the asphalt surface were not some provision
made against it. For this purpose a form of construction has
been evolved which has proved quite successful by providing sat-
isfactory drainage and not laying the hydraulic concrete foundation

THE FOUNDATION. 5

in direct contact with the subsoil. Upon the subsoil clean sand
and gravel are spread and rolled to a depth of three inches,
and upon this the hydraulic concrete foundation is laid. At the
same time, at intervals of twenty-five feet, trenches are cut
in the subsoil to a depth of six inches, and filled with coarse
broken stone; these cross-drains being connected with similar
trenches, containing coarse broken stone, under the curb, which
are graded to catch-basins for the removal of water. In such a
climate tile drains cannot be used successfully, the author is
informed by the City Engineer of Winnipeg, because the ground
is generally frozen when the surface-water first begins to drain
away and this water filling the tile often freezes and bursts it.
The provisions in use in Winnipeg seem to be an ideal way of
treating heavy soils in cold climates and, although absolutely
necessary in such a location, are extremely desirable where any
soil of such a description is found. Further details in regard
to this method will appear in the chapter on "Specifications." l

When a street is terraced and the roadway is lower than the
adjacent property the greatest precaution should be taken to
prevent the seepage from higher levels from working down between
the soil and the foundation, or between the foundation and surface.
For this purpose drainage 6hould be provided below or along the
curb, and at times ia the subsoil itself.

The most serious subgrade to encounter as a support for a
foundation is marshy or swampy land, fills made on the latter for
the purpose of raising the grade, or even fills on ordinary soils where
sufficient time has not elapsed to bring about final settlement and
ultimate compaction. Where such conditions are unavoidable
good practice calls for the use of a sufficiently strong hydraulic
concrete foundation to bridge over irregular settlement and to
distribute the load over weak points.

The cause of much of the deterioration in asphalt surfaces
and in otner pavements is due to a neglect of such precautions
in regulating the subgrade or in providing a foundation of a
character to bridge over defects in the latter.

1 Page 435.

6 THE MODERN ASPHALT PAVEMENT.

It seems hardly necessary to state that any subsoil should be
thoroughly compacted by a heavy roller of broad tread, and that
any weak portions revealed by the use of such roller should be
removed by treatment in an appropriate way.

The Foundation. Foundations of most varied character have
been used in the construction of pavements, including broken stone
with or without a coating of more or less bitumen or coal-tar, mac-
adam, old cobblestone pavement, an old surface of granite blocks
or blocks turned and reset, old brick or asphalt-block surfaces, and
hydraulic concretes of natural or Portland cement of varying thick-
ness. Each of these forms has been more or less successful under
different conditions, consideration being given to economy, to
local environment, and to the traffic to be carried.

Bituminous Foundation. The so-called bituminous founda-
tion possesses no advantage save, in some cases, that of economy.
It has been almost entirely abandoned as a support for asphalt
surfaces. It is a relic of the days when hydraulic cement was a
much more expensive article than at the present time. As
generally constructed it consists of six or more inches of broken
stone passing a two or two and one-half-inch ring and not
containing any particles passing a one or one and one-half-inch
ring. Stone of such uniform size contains a large volume of
voids, forty per cent or over, and does not compact well. Were
it the run of the crusher, the foundation would be far more satis-
factory. Under the roller much of the stone is often lost in the
subsoil before the required thickness is attained. The coating of
bitumen applied to the surface of the foundation is of little or no
advantage. Enough cannot be used to fill the voids, as if this is
done the excess will be drawn up into the surface by a hot sun
and destroy or soften the latter while the cost would also be
prohibitive.

Additional disadvantages of such a foundation are that it pos-
sesses no rigidity or stability and consequently responds at once to
any settlement or weakness of the subsoil; that it is porous and
allows the free movement of water and gas, and that the binder and
surface cannot be readily removed from it for renewal without
its destruction.

THE FOUNDATION. 7

Excellent asphalt pavements have been constructed with a
bituminous foundation where the subsoil was firm and resistant
and the travel light, as is the case in many residence streets, but
surfaces equally good have been laid with broken stone alone.
The cost of renewal of the surface of such pavements is, however,
high, as has already been shown.

Macadam Foundation. Old macadam has been used success-
fully in several instances as a foundation for asphalt pavement. It
possesses many advantages over broken-stone. In macadam the
voids in the stone are well filled with finer particles and it has
received its ultimate compression under travel. It has not a
coating of bitumen, so that the binder and surface coats are readily
removed for renewal. It possesses the defect that the grade of
the macadam can be altered but slightly without serious disturb-
ance of its bond, and in replacing the pavement over trenches the
stone becomes far less of a support than the original macadam.
It, of course, presents the merit of economy. An excellent example
of an asphalt pavement with a foundation of this class is to be seen
on Broadway, above Fifty-ninth Street, in New York City, and on
Michigan Avenue, to the south of Congress Street, in Chicago,
111. The former street has been trenched to a large extent, and
repairs have resulted very satisfactorily. The latter has a few
cracks owing to the action of frost.

Other Old Pavement as Foundation. Old cobblestone and
asphalt-block pavements form an excellent foundation for asphalt
pavements if the height of curb shown is sufficient and the amount
of traffic permits. They should not be used if resetting is necessary
except on very favorable soil and with the expectation of very
moderate use.

Old granite-block pavements have been very extensively used
for supporting asphalt surfaces, especially in New York City,
and their value for this purpose and the defects which result there-
from are well illustrated there. Granite blocks laid in sand on
the soil have been found very satisfactory on cross-town residence
streets, but have been most unsatisfactory on streets like First
Avenue, where the subsoil is soft, fills are frequent, and partly on
land below high-water mark, in bringing the street originally to

S THE MODERN ASPHALT PAVEMENT.

grade. These blocks could be seen to move under the steam-roller
when the binder course was laid, and the asphalt surface is, in con-
sequence, in constant need of repairs. The only foundation which
would prove satisfactory under these conditions would be, con-
sidering the heavy travel on the street, the best form of Portland-
cement concrete to a depth of at least eight inches.

Granite blocks on concrete make an excellent foundation for
an asphalt surface, if not reset, but where the grade necessitates
taking them up and replacing them on their broad sides the result
is not satisfactory except on residence streets. When relaid they
are not rigid, but have a tendency to rock under heavy travel.
Such a foundation supports the asphalt surface on Broadway and
several avenues in New York City, and has not been entirely
successful. The turned blocks were opened to travel for some time
to bed them thoroughly, and any loose ones reset. The binder
and surface were then laid directly on the blocks. The vibration
on, these blocks, especially along the rail, is nevertheless large.
A better form of construction would be to grout the blocks, after
turning, with Portland cement, and keep traffic off of them until
the grout is set. The lesson is that turned granite blocks should
not be used as a foundation for asphalt pavements on streets of
heavy travel, even when supported by Portland-cement concrete,
and much less so on soil alone, as has recently been done on
Fourth Avenue in New York City.

Construction of this description inevitably results in deteriora-
tion of the best asphalt surfaces, with the results that the defects
are attributed to the asphalt and not to the foundation where they
really originate. As a matter of fact no foundation is suitable for
a street of heavy travel except one of Portland-cement concrete
of sufficient depth and strength to carry the load imposed upon the
surface with perfect rigidity, and it is equally true, as determined
by years of careful observation, that ninety per cent of the defects
in asphalt pavements in such cities as New York, where the sur-
face mixture is of standard quality, are due to the insufficiency
of the foundation.

Old brick pavements have served as a support for asphalt sur-
faces with entire satisfaction. They have received the full traffic
of the street to be resurfaced and are therefore well compacted.

THE FOUNDATION. 9

Such pavements, as their surfaces become too uneven for use, will,
in the future, be largely renewed in this way.

Hydraulic-Concrete Foundation. Hydraulic concrete, if prop-
erly proportioned, made with good cement and a well-graded
aggregate, well mixed and put in place satisfactorily and in good
weather, is the ideal foundation. Unfortunately these conditions
are not always met. To discuss the possible variations and de-
ficiencies in detail would be to write an elaborate treatise on the
subject of concrete. It will suffice, however, to point out the chief
merits and defects which appear most strongly in its use as a
foundation for pavements.

The proportions of the different constituents, from which the
contractor cannot depart, are sometimes injudiciously prescribed,
and usually from motives of economy. In an eastern city a
concrete foundation is specified which is to consist of one (1) part
of Portland cement, four (4) parts of sand, frwe (5) p.arts of gravel,
and five (5) parts of stone. While such a foundation may have
sufficient strength to support a pavement carrying moderate travel,
it is nevertheless porous and permits of the free movement of water
and gas from below, both of which act on asphalt under such
circumstances. The most favorable proportions which the writer
has observed for a concrete for heavily travelled streets are those
which were prescribed for Fifth Avenue, in New York City, in 1896.
These were: one (1) part of Portland cement, three (3) parts of
sand, two (2) or three (3) parts of gravel, and four (4) or five (5)
parts of broken stone. If the stone were large the larger amount
of gravel was used and the reverse. The gravel in the preceding
concrete is of the greatest aid in filling the voids in the stone and
in facilitating the compaction of the concrete when rammed. Stone
with sand alone is very apt to bind, bridge, and resist compaction,
while the voids are so large as to leave at times a portion of them
unfilled with mortar.

The foundation thus constructed on Fifth Avenue has a depth
of at least seven inches and is absolutely rigid. The asphalt sur-
face placed on this is subject to no vibration and has shown no
deterioration due to this cause in twelve years, while the same
surface mixture supported only by the granite blocks on a soft
soil in First Avenue was in far from good condition in a vear.

10 THE MODERN ASPHALT PAVEMENT.

This contrast between the two avenues with their different founda-
tions is, therefore, most instructive and points to the fact that the
primary consideration in an asphalt pavement is the foundation
which supports it. Without a rigid one the best of materials and
workmanship in the remainder of the pavement will go for naught.

Ordinary practice as regards the aggregate in a hydraulic
concrete provides for broken stone all of which "will pass in any
direction through a revolving circular screen having holes two and
one-half (2^) inches in diameter and be retained by a screen hav-
ing holes one (1) inch in diameter." 1 This may be good prac-
tice, but is certainly not the best.

The grading of the broken stone is an important consideration
in adjusting the relations of the constituents in a well-proportioned
concrete not only in a foundation for asphalt pavements, but in its use
far every purpose. Fortunately this is. rapidly becoming recognized.
As has already been mentioned, the voids in broken stone passing
a two-inch ring and not passing an inch and one-half or a
one-inch ring are very large in volume, and it has been, and
generally is, the practice to attempt to fill them with a mortar of
one part of cement and three parts of sand, in the case of
Portland, or two in the case of natural cement. This is not
economical in more ways than one. It is much better to reduce
the voids by using the broken stone as it comes from the crusher
in well-assorted sizes and with smaller voids, the screenings passing
a quarter- or three-eighth-inch screen only being removed, on
account of its tendency to segregate and because it should actually
be considered as sand, as will appear later, or to add gravel where
it is available or where economy demands the separation of the
inch stone for use in binder. The mortar then goes much farther,
is not in such large masses, and the concrete is rammed and com-
pacts with much greater ease. Under such circumstances, while
the proportion of cement and sand should not be extended beyond
one of the former to three of the latter, corresponding to the
relation of the volume of the cement to the voids in the average
concrete sand, the proportion of stone to the mortar may be largely

1 Manhattan (New York) Specifications, 1901, paragraph 12.

THE FOUNDATION. 11

extended beyond that which may be safely allowed for stone of
uniform size and large voids.

Where good gravel is available, containing particles of suf-
ficiently large size, a concrete made with this material without the
use of crushed stone may be as equally satisfactory or even preferable
to one constructed with stone alone or with a mixture of stone and
gravel. A provision for such concrete is now contained in the speci-
fications of some of our cities.

Where gravel occurs mixed with the requisite proportion of
sand such a natural mineral aggregate can be employed in the
manner in which Thames ballast is used in London, England, and
with the most satisfactory results. The occurrence of such deposits
in the United States is exceptional.

Another fortunate thing in recent practice is the recognition
of the facts that a concrete the mortar in which is wet enough to
almost quake under the tamper gives the most satisfactory results,
since the slight early loss in strength due to the water excess is
more than made up by the improved and thorough compaction
attained. Dry concrete is no longer regarded as good prac-
tice.

In the early days of the asphalt-paving industry hydraulic
concrete was usually mixed by hand labor on boards. To-day
much of this work is very satisfactorily and much more cheaply
done with the use of mixers driven by power. Of these there are
a number of successful types now on the market, and, from the
author's observation, they are strongly to be recommended where
the extent of the work will justify their being employed.

Concrete Sand. It is usually specified that sand in use in
concrete shall be clean, coarse, sharp, and free from loam and dirt.
The degree of coarseness is generally somewhat indefinitely expressed.
Concrete sand as a rule contains but a small percentage of grains
finer than will pass a fifty-mesh sieve. It may, however, con-
tain to advantage a considerable portion of fine gravel. In this
connection it may be remarked, however, that the permeability
of a concrete is greater the coarser the sand, and that the presence
of some fine grains is not undesirable. A small amount of loam
or clay in sand is not injurious if it is not present in a lumpy con-

12 THE MODERN ASPHALT PAVEMENT.

dition. Many pit sands which have been rejected on account of
the presence of loam make excellent concrete.

Crusher Screenings. Sand has been defined as the detritus
of rock, smaller than gravel and larger than silt. Under such a
definition the screenings from the crushing of rock for the pro-
duction of broken stone is sand and may be used as such in concrete.
Its use for such purposes has attracted very considerable attention
recently, and the results obtained with it have been most successful.
The strength of the concrete in which sand is replaced by screenings
is always equal to and in many cases in excess of that made with
natural sand. It has been used, and pronounced a desirable
material, in the concrete of the Buffalo breakwater, 1 in the Man-
chester and Liverpool (England) water-works, 2 in the Jerome
Park Reservoir in New York City, in masonry construction on the
C. M. & St. P. R. R., and in the concrete on the Water-power
Sections Nos. 1, 2, and 3 of the Chicago Drainage Canal. It has
been tested by many engineers in the laboratory and found to
produce concrete exceeding or equalling in strength that made with
sand. An excellent resume of the availability of this material
will be found in "The Cement Age," Vol. l,.No. 3, page 5, August
1904, and in a publication of the Producers' Supply Company,
entitled "Crushed Stone and its Uses/' pages 100, 103, and 107,
Chicago, 1904.

It has been successfully used in the concrete foundation for asphalt
pavements in several cities, and its use for this and other purposes
is rapidly increasing. The author made the following statement
in connection with the use of crusher screenings as a witness before
the Aqueduct Commissioners of the City of New York when this
was objected to by the Merchants' Association of the city:

"The advantages are that the particles in the crusher screen-
ings are better graded in size, and in consequence those screenings
have a smaller volume of voids or unfilled cavities in them. As
a result a definite volume of Portland cement will go farther
towards filling those voids than with sand, where the particles

1 Eng. News, Sept. 11, 1902.

2 Hill, Institution of Civil Engineers, London, 1896; Deacon, ibid.

THE FOUNDATION. 13

are more uniform in size and the volume of the voids large. . . . "
There can be no question that crusher screenings are preferable
to many natural sands when the rock in which they originate is
of desirable quality.

Character of the Hydraulic Cement in Use. The character of
the hydraulic cement in use in concrete foundations is as important
as any constituent of the pavement. If it is defective in any way,
the result will be shown in the surface. In one case one of the
most prominent surfaces in the country became cracked across
the street at wide intervals two years after it was laid, and in three
years the surface was noticeably raised at these points. On open-
ing the pavement the cracks in the surface were found to be due
to cracks in the concrete formed during the first two years, and
the elevation of the surface, which occurred later, to the subse-
quent expansion of the cement, which in this way pushed one
portion of the foundation upward and over the other, although the
cement was a Portland and one which responded satisfactorily to
all short-time tests for constancy of volume. A similar exper-
ience was met with for several years with the natural cements of
western New York, and it was generally necessary, where one
brand was used, to remove the surface and cut out the expanded
portion after a few years in order to bring the surface of the
pavement to grade.

In the middle West serious troubles due to the character of
the natural cement in use were often met with befoie Portland
cement became available. The natural cements of that part of the
country are not always reliable or uniform and are especially un-
suited for use in cold weather, as they fail to set when the tem-
perature approaches freezing. The writer has frequently seen
hydraulic foundations which have acquired no bond, either from
their inferior quality alone or because of use in cold weather. Such
a foundation is open and porous and allows water to reach and
disintegrate the asphalt surface. It frequently cracks after a firm
set has taken place, and these cracks are eventually repeated in the
asphalt surface, as can be seen in the accompanying illustration,
Fig. 1 . The evident conclusion is that the use of natural cement
in concrete for the foundation of asphalt pavements should be

FIG. 1.

THE FOUNDATION 15

abandoned, although it would not be justifiable to suppose that
none made from this cement or even the majority of it is poor.
That under the first asphalt pavement of any area, on Pennsyl-
vania Avenue, Washington, D. C., was constructed with natural
cement from the Potomac Valley, and during thirty years gave
entire satisfaction. Foundations containing the Rosendale cem-
ents have proved equally good, but all the natural cements of
this description attain their strength so slowly that an unfortu-
nately long period must elapse before they will safely sustain a heavy
roller suitable for compressing the binder and surface, and in this
way the completion of the pavement is delayed. It is therefore
much better to avoid using natural cements, and the substitution
of Portland cement has become the very general practice.

All hydraulic cement in use in the construction of asphalt
pavements should be tested before it is allowed to go into the
work, and should meet the requirements which the local engineer
believes to be reasonable. The Committee on Uniform Tests of
Cement of the American Society of Civil Engineers has recom-
mended methods which should bring about greater uniformity
in testing cements, and their use is suggested. 1

A similar committee of the American Society for Testing Mate-
rials has recommended reasonable specifications for cement which
can be adopted if they meet with the approval of the engineer. 2

Lateral Support. Closely related in importance to the char-
acter of the foundation of the pavement is that of the lateral sup-
port which the surface receives.

It is quite as well proved by experience that more defects
in asphalt surfaces are due, proportionally to the area involved,
to the lack of this than to weak foundation and, often, to all other
causes.

The lateral support should be as rigid as 'in the case of the
foundation, and unfortunately it is not always so about manholes,
water- and gas-boxes, at headers where the surface ends, and espe-
cially against rails. Vibration about manholes, boxes, etc., can be

1 Proceedings Am. Soc. C. E., 1903, 29, No. 1.

1 Report of Committee C on Standard Specifications for Cement. Pre-
sented at Annual Meeting, 1904, June 17.

16 THE MODERN ASPHALT PAVEMENT.

avoided by providing heavy castings with a broad base and set-
ting them upon a proper foundation in Portland cement a suffi-
ciently long period before laying the surface to prevent them from
being loosened by a blow from the roller. Headers should be
sufficiently heavy to hold the surface up and resist the impact
of traffic.

The construction of a street car-track the rails and sleepers
of which shall be sufficiently free from vibration to form a sup-
port for an immediately adjacent asphalt surface is a most diffi-
cult matter and one that is rarely successfully carried out, espe-
cially when trolley-cars of the size and weight of those in use
to-day are to 'be considered.

Experience has shown that construction involving the use of
a very heavy girder-rail placed upon ties, which, together with
the rail, are embedded in concrete from the base of the former
to the height of the adjoining foundation of the pavement, is the best.
Such construction will be, however, of little value if traffic is
allowed over the rail before the concrete has had time to set thor-
oughly. In cases where the soil is very heavy and the drainage
is bad crushed stone used as ballast may often prove more satis-
factory than hydraulic concrete, as affording better drainage.
Where mud forms, owing to poor drainage, and works into cracks
between the asphalt surface and the concrete the result is very
disastrous.

Another form of rail construction which has met with con-
siderable approval is the placing of the rail upon a hydraulic con-
crete beam extending its entire length. It is possible that this
may be desirable where carefully carried out, but, in the author's
experience, where a girder-rail of sufficiently heavy type is to be
used no advantage is derived commensurate with the expense, and
the possibility of vibration is not lessened.

If vibration still takes place in a rail, even with the best form
of construction, and this is rarely absent with heavy trolley-cars,
a triple row of the best paving-blocks, or bricks, laid with broken
joints parallel to the rail should be placed against it, bedded in
cement, and well grouted, depressing the base sufficiently for
this purpose. Header and stretcher construction is most faulty.
The asphalt toothing is then a point of weakness.

THE FOUNDATION. 17

Vibration of the rail will eventually destroy an immediately
adjoining surface not only by breaking the bond between the
particles of the surface, but by admitting water and mud after
the first fracture has taken place.

All the defects which are due to weakness in the foundation and to
the lack of lateral support involve not only an expense to the
contractor during the guarantee period which he must consider
hi his bids after a study of the form of construction specified by
the city, but will also prove an additional cost to the city when
it takes over the maintenance of the street. Economy in the cost
of the pavement in this direction may not prove true economy
in the end.

While it is not, of course, necessary that a needlessly expen-
sive foundation should be provided for an asphalt pavement, it will
eventually prove cheaper if a good margin of safety in this direc-
tion is allowed. An asphalt surface is no stronger than its weak-
est part.

SUMMARY.

It appears that better concrete can be made of graded
broken stone than of stone of uniform size, that the addi-
tion of gravel is an improvement, that a concrete consisting of
gravel alone as a substitute for broken stone will often prove sat-
isfactory, that crusher screenings are an excellent substitute for
natural sand, that Portland cement is infinitely preferable to
natural cement, that the greatest care should be used that the
pavement should have a proper lateral as well as vertical support,
and that the greatest attention should be paid to the rigidity of
railroad-track construction.

It seems, therefore, that while an asphalt pavement of the
best quality can be constructed only when all its elements foun-
dation, binder or its substitute, and surface are of the highest degree
of perfection, refinement in the character of the binder and sur-
face is thrown away if the subsoil is not satisfactorily drained and
if the foundation of the pavement is not sufficiently strong to carry
the traffic to which the surface is subjected with entire rigidity

18 THE MODERN ASPHALT PAVEMENT.

and is not sufficiently impervious to protect the surface from the
action of water and illuminating-gas.

In addition a well-constructed foundation is a matter of
economy, as it should last for all time and will only require
resurfacing at intervals, whereas an inferior one must eventually
be renewed by one properly constructed.

CHAPTER II.
THE INTERMEDIATE COURSE.

IN the early days of the asphalt-paving industry a thicker
wearing surface was hi use than to-day. That of 1876 on Pennsyl-
vania Avenue in Washington, D. C., and most of those laid in the
following fifteen years were two and one-half niches thick. These
surfaces were laid in two courses, and are thus described in an
old specification of a Washington contractor in 1878: x

"The asphalt (surface mixture), having been prepared in the
manner thus indicated, is laid on the foundation in two coats.

"The first coat of one-half inch thickness, called protecting
coat, might be laid richer in asphaltic cement, and may be consoli-
dated simply by rolling with iron or stone rollers weighing about
1000 pounds or half a ton.

"On this first asphalt coat is then carefully spread with iron
rakes the final finishing coat," etc., etc.

It is evident from this that the one-half-inch coat of surface
mixture was laid for no other purpose, at this tune, than to pro-
tect the rather friable hydraulic concrete of natural cement from being
broken up by hauling the final surface mixture over it. It will
be noted that it is suggested to lay the first coat of material richer
in asphalt cement.

In 1884 the specifications which the city itself adopted were
evidently based on those of 1878, but the wording was somewhat
changed, "pavement mixture" replacing "asphalt" in the first
paragraph quoted, and " cushion coat" for "protective coat," with
some other minor alterations, in the second, the thickness of the

1 Rept. of Com. D. C. f 1878, 292.

20 THE MODERN ASPHALT PAVEMENT.

latter remaining one-half inch "after being consolidated by a
roller," while it is to contain, specifically, "from two to four per
cent more asphaltic cement" than the "surface coat."

In the interval mentioned the term "protective coat" which
casts some reflection on the character of the foundation has, therefore,
been changed to "cushion coat." The greater richness of the cush-
ion has been retained.

In the specifications for 1886-87 no mention is made of a pro-
tective or cushion coat. It is provided that the surface mixture
will be "carefully spread, in such a manner as to give a uniform
and regular surface and to such depth as, after having received
its ultimate compression of 40 per cent, to have a thickness of 2J
inches."

The cushion coat was temporarily abandoned in that year. The
reason for this is instructive, as showing the defects of this method
of construction. Surfaces laid with a thickness of two and one-
half inches the lower part of which consisted of a protective
or cushion coat richer in bitumen were liable to serious displace-
ment under travel, with the result that the surface became very
wavy and uncomfortable to drive over, an experience met with in
other cities as well, and which subjected such asphalt pavements
to unfavorable comment. The excessive thickness and the richer
cushion coat permitted not only of this displacement of the sur-
face, but also allowed its movement on the foundation under
the impact of wheels of vehicles, when once the waves were formed,
especially where travel, as in streets with car-tracks, was confined
in one direction.

The change in the specifications in 1886-87 was intended to
avoid this by doing away with the cushion and compacting the
entire surface at once. It was an improvement, but for some reason
the provision for a cushion coat one or two per cent richer in asphalt
appeared again in the specifications for 1887-88. No asphalt
pavements were laid in Washington during this fiscal year, as the
city was compelled, by provisions in the act making appropriations
for the purpose, not to go beyond a limit in price for such pavement
for which the contractors refused to lay asphalt. A return was,
therefore, made to coal-tar with disastrous results, the only gain

THE INTERMEDIATE COURSE. 21

being that when asphalt surfaces were again laid in 1888 the ex-
cessive thickness of the surface was reduced and a course of broken
stone coated with coal-tar or asphalt was introduced which had
been an element of the so-called distillate pavements of the inter-
mediate period. This gain was a distinct one for the asphalt-
paving industry all over the country, as it did away with the dis-
placement of the thicker surface under traffic. No general return
to the original method of construction without this course has been
made since that time except in two or three cities, and there with
less unsatisfactory results than formerly 'owing to the more careful
grading of the mineral aggregate in modern mixtures and their
greater stability.

The Binder Course. The binder course, as has been said, is
an inheritance from the days of coal-tar pavements with broken
stone foundations, and its use in combination with an asphalt sur-
face was the result of an attempt to improve upon the unsatis-
factory distillate pavement, so called, laid in Washington in 1887,
by substituting an asphalt for a coal-tar surface, leaving the
foundation and binder unchanged. Eventually a hydraulic con-
crete foundation was substituted for the broken stone and the re-
sult was the modern form of construction.

The binder course was evidently the result of an attempt to
close the large openings in the broken-stone foundation by a course
of finer stone in order to prevent the loss of the more expensive
surface mixture by its compression into the voids of the former
and with no idea of preventing displacement in the surface. That
it accomplished this was only detected when it was noticed that
its use as a matter of economy in reducing the thickness of the
asphalt surface from two and one-half to one and one-half inches
produced this desired result. From Washington in 1888-89 the
binder course rapidly spread over the country and proved suc-
cessful. In its original form it consisted of "clean broken stone,
thoroughly screened, not exceeding one and one-fourth (1^) inches
in the largest dimension and No. 4 coal-tar paving cement."

The coal-tar was soon replaced by an asphaltic cement and
the broken stone in some cities by a smaller stone passing an inch
ring with the grit and finer material removed.

From a binder constructed in this way there has been little

22 THE MODERN ASPHALT PAVEMENT.

departure for many years, although recently the possibility of
some improvement in this course has become very evident. The
original course was one and one-half inches thick when compacted,
a depth quite necessary with one and one-fourth inch stone. With
finer stone and for economy inch binder has frequently been speci-
fied, but there can be little or no bond to such a thickness and
its use in this way is plainly poor practice.

It is generally specified that the binder stone shall pass a one
and one-fourth- or one-inch screen and contain not more than
a certain percentage of fine material. This is a great mistake,
as in the case of hydraulic concrete, since the more fine material
the stone contains, up to the point where the voids in the large
particles are filled, the more compact and desirable the binder is.
At the same time a binder with much fine material requires a
larger amount of asphalt cement and is consequently more expen-
sive. If the contractor is willing to assume this extra expense
no objection can be raised to such a practice.

The amount of asphalt cement necessary to coat satisfactorily
a binder of clean stone free from grit and dust will vary with the
character of the stone and the nature of the asphalt cement. With
Hudson River limestone or trap three per cent of bitumen is
sufficient, and this is represented by that amount of an asphalt
cement composed of pure bitumen or four per cent of one made
with Trinidad asphalt. With the softer limestones of the middle
West, a higher percentage is occasionally necessary. The exact
amount can only be determined by experiment. It should not be
sufficient to run off of the hot stone or too little to give a bright
glossy coat. An excess may result disastrously, as it will collect
inevitably in pools or spots where the binder has been taken from
the bottom of the truck in which it has been hauled to the street,
and the excess collecting at these points may be drawn up by a
hot summer sun and soften the surface of the pavement or even
appear in mass thereon, as has happened in one or two instances
in a western city.

On the other hand, a slight and well distributed excess may
prove of decided benefit on streets having little or no traffic, where
the surface would be apt to crack ordinarily. It has been found

THE INTERMEDIATE COURSE.

that in such a case the surface is slowly enriched by the excess,
and is thus preserved.

An example of this enrichment was observed in an Alcatraz
surface laid on a very rich binder in a western city in 1889. A
specimen of the surface was analyzed several years after it was
laid, after separating it into top and bottom sections. The results
were as follows :

Section

Bitu-

Passim

5 Mesh.

men.

200

100

Top

9.8

12.2

Duplicate.

Bottom. .
Duplicate.

9.8

10.3
10.7

11.2

11.7
11.3

10
10

32
33

32
31

1
1

The bottom of the pavement carries, on an average, seven-tenths
per cent more bitumen than the top, and that this is no accident
in mixing appears from the uniformity of the sand grading in
all the samples of the different sections.

The consistency of the asphalt cement hi use in binder should
be softer than that in the surface for several reasons. The ordinary
binder is a very open material which permits the volatilization of
oil from the asphalt cement by the heat of the stone, especially
if the stone is accidentally too hot and the haul to the work a
long one, with the result that the cement becomes much hardened
and more brittle. In the second place the tendency to the rupture
of the bond between the fragments of binder stone is much less
with an asphalt cement of soft than of hard consistency.

Good practice leads to the use of a cement for binder which
is twenty or more points softer, by the penetration machine,
than that hi the surface.

The actual temperature of the binder as it is laid on the street
should be no greater than is necessary to make it possible to. rake
it out. It may be much colder than an asphalt surface mixture.

Recent experience has shown that there are defects in such
a binder which are due to the fact that the voids are unfilled and

THE MODERN ASPHALT PAVEMENT.

the course lacks stability and solidity. Such defects have been
manifested in two ways for many years. If binder is not laid with
great attention to the character of the asphalt cement which covers
the stone and binds it together it soon loses its bond under heavy
traffic and, the stone itself having but little supporting power,
the asphalt surface goes to pieces. If, on the other hand, the
binder stone itself is not a strong one it is frequently crushed by
the weight of heavy traffic and the surface, losing its support,
either goes to pieces or the crushed binder is forced into it irregu-
larly, thus causing a decided displacement which eventually results
in disintegration.

In individual cases the surface has been observed to have been
driven by traffic into the voids in the binder without displacement,
with the result that the thickness of the pavement has been much
reduced, although in this condition it is a much more rigid mass
than as it was first constructed.

The binder previously described has consisted of stone prac-
tically or largely of one size, three-quarters to one and one-half
inches in the largest diameter, as appears from the following an-
alyses:

Test number ....

69978

70804

70854

71102

74893

Bitumen . .

54%

44%

3.8%

36%

3.5%

Filler

4 IT

2.21 n ~

2.4 1 KA

1.5 | , e

Sand

21 o r^- 8

12]5 ) 16 ' 6

7.5J 9 ' 7

3.0 / 5A

3.0 r 4 - 5

Stone:

Passing \" sieve

5.81
13.6 67 8

8.71
46.8 79Q

18.0]
52.0 ) 86 5

13.51
51.5 L 10

49.51

10.0 L 20

1" "

41.4 F 67 - 8

23.5 [' y ' U

16.5 j

26.0 f y >u

32.5 j

Retained 1" "

7.0 J

0.0 J

100.0

It will be seen from the preceding results that the percentage
of bitumen which binder will carry depends largely upon the
amount of fine material which it contains, binder No. 69978 with
26.8 per cent of fine material holding 5.4 per cent of bitumen, while
those made from cleaner stone where the fine material does not
exceed 5 per cent, carry less than 4 per cent of bitumen. If the

THE INTERMEDIATE COURSE. 25

stone in use is not screened, but contains all the finer particles
coming from the crusher, the binder will be more satisfactory.

Asphaltic Concrete Binder. The weakness of the ordinary
open-binder course, where subjected to heavy traffic, can be
avoided by filling the voids in the material with fine stone or grit
and the remaining voids, after this addition, with sand or a mineral
aggregate corresponding in grading to that of a standard surface
mixture.

Such a binder has given most excellent results in supporting
an asphalt surface on an ordinary foundation, such as turned or
reset blocks or alongside of poorly constructed street-railway
tracks, such as those on Broadway in New York City, being hi
itself perfectly rigid.

A good example of this form of construction on a hydraulic
concrete foundation may be seen on Court Street in Boston, from
Washington Street to the old Court House, and on Kilby Street
in the same city, between State and Central Streets. A photograph
of a sawn section of such pavement is shown in Fig. 16.

The manner in which a binder of this type, which is generally
known hi the industry as close or compact binder, is turned out at
the plant is much the same as that used in preparing an as-
phaltic concrete to be used as a wearing surface, with the exception
that it contains no filler,. and will be described in a later chap-
ter. It will be of interest, however, to give some examples of the
proportions in which the various components have been employed
in actual practice during the year 1907, together with the results
of analyses of the finished binder. (See Table on page 26.)

Close binder of the Boston type is probably more desirable than
that produced in New York, owing to the larger percentage of
three-quarter inch stone which it contains and its consequent
greater stability, but as the stone at that point is not separated
into two sizes after heating, but drawn from one bin, there is
probably a greater amount of segregation in the mixture than in
that turned out at New York, where stone of each size is separated
and weighed out in definite proportions. Nevertheless, a very
satisfactory asphaltic concrete for the binder course has been
turned out in Boston in that way, with the exception that where

THE MODERN ASPHALT PAVEMENT.

New York.

Boston.

Plant 1.

Plant 2.

Coarse stone

480 lbs. = 54.6%
200 u 22.7
150 " 17.0

50 " 5.7

480 lbs. = 53.3%
202 " 22.4
150 " 16.7
68 " 7.6

} 885 Ibs. =73.4%

250 " 20.8
70 " 5.8

Fine stone

Sand

Trinidad asphalt cement
Bermudez

880 Ibs. 100.0

900 Ibs. 100.0

1205 Ibs. 100.0

ANALYSES:

Bitumen soluble in CS S .
Passing 200 mesh screen

5,%

5.2%
6.4

4.8%
4.2

' 10 " "

29.2

29.0

26.4

* " "

1.8

2.0

i inch

10.0

7.0

4.4

l if, rt

23.2

25.8

24.4

t 3 tl

13.6

17.8

32.2

t n

7.4

6.8

2.8

Retained 1 "

4.8

0.0

100.0

there is segregation of the finer material, it sometimes happens
that there will be a displacement of the surface of the finished
pavement at that point.

On this account, there are one or two points which experience
has shown must be guarded against. The fine material should
not be at all in excess of an amount sufficient to fill the voids in
the stone, since, if this is the case, too smooth a surface will be
obtained, with resulting displacement of the wearing surface. An
excess of bitumen must also be avoided for the same reason.

In the construction of an asphalt pavement with an asphaltic
concrete binder course, the surface should be applied to the binder
before it has become entirely cold, in order to obtain a satisfactory
bond. It is not possible to do satisfactory work where a large area
of close binder is laid on one day and covered with surface on the
next. There should be an alternation of the binder and surface
courses,, for example, no more binder should be laid in the begin-

THE INTERMEDIATE COURSE. 27

ning of a day than can be covered with surface during the remain-
der of the available working hours.

Where old surface mixture is available and facilities are at
hand for softening this by means of heat or by grinding it in a
disintegrator, such material can be used quite as satisfactorily
for filling the voids in an ordinary binder as new sand and filler,
thus reducing very much the cost of a concrete binder course.
The percentage of asphalt and its consistency in such a case will,
of course, be regulated by the amount already present in the old
material.

The form of construction involving the use of a compact binder
naturally increases the cost of the pavement to a certain extent,
but the advantages gained more than make up for this, since its
life will be extended to a degree more than sufficient to bring the
cost per year below that of one in which the old open form of binder
is used, especially on streets of heavy travel. In the author's
opinion it is the most important advance that has been made in
the asphalt paving industry since the evolution of the rationally
graded surface mixture in 1896. Its use is specified in Kansas
City and Omaha, as can be seen in the very excellent specifications
of the former city, which are reprinted in the appendix.

Paint-coat. Some of the defects in a pavement due to an
open binder can, perhaps, also be removed by abandoning it entirely
and substituting therefor a so-called paint course which con-
sists of an asphalt cement of suitable consistency dissolved in
benzine, 62 B., and then applied with a brush or squeegee to the
surface of the hydraulic foundation, which should be made, if this
coat is used, of Portland cement, or else floated with a mortar of
this cement, and should have a comparatively smooth surface.
The coating should be bright and glossy, but not sticky, and it
must be carefully protected from becoming dirty. It cannot
be applied successfully to a concrete that is in the slightest degree
damp, as the adhesion is then imperfect. If well done, and this
requires some skill and experience, the surface mixture applied
directly to this coat will be cemented firmly to it and any displace-
ment in the surface on the foundation will be prevented if the
mixture itself is stable.

28 THE MODERN ASPHALT PAVEMENT.

The first use of such a coat as a substitute for binder was made
in a town in Ohio in 1896, where an asphalt surface was laid on
an old brick pavement, the grade of which did not permit of the
use of a binder course. The adhesion of the surface to the brick
was afterwards found, on making cuts for water and gas connec-
tions, to be so strong that the upper portions of the brick were
torn away with the asphalt surface. On one or two streets in
New York on which very heavy coal trucks are constantly passing,
and where the binder course was frequently crushed, a similar
construction on a Portland-cement base was successful.

Specifications for the use of paint coat are as follows:
"Upon the surface of the foundation of the Portland cement
concrete, a paint course shall be applied to bind or tie the surface
course to the foundation. For this purpose, the surface of the
cement concrete shall, in its preparation, be well rammed, so that
mortar shall come to the top, and it shall be made so smooth that
no depression shall exist of a depth of more than three-eighths (f )
of an inch. The paint for use shall consist of 62 B. naphtha and
any satisfactory asphalt cement free from mineral matter, and
of a consistency such as will allow the penetration, at 77 degrees
Fahrenheit, of a No. 2 needle weighted with 100 grams, of not
more than three (3) millimeters, and not less than two (2).
The asphalt cement shall be dissolved while soft and warm, in the
naphtha in such proportions that the resulting paint shall give a
glossy surface after evaporation of the latter, but at the same
time can be applied so as to form as thin a coating as possible.
The proportions will vary, depending upon the temperature at
which the paint is made, but shall be about 240 pounds of asphalt
cement to 50 gallons, or one barrel, of naphtha,

"The concrete foundation shall be carefully swept and thorough-
ly cleaned of all foreign matter. The paint coat shall only be
applied to it when the latter is absolutely dry and free from the
slightest dampness, as otherwise it will not adhere. It should be
used in such quantity that fifty (50) gallons will cover from 350
to 4QO square yards of the concrete surface.

"No more of the surface of the foundation shall be painted than
can be covered with asphalt surface mixture within a few hours after

THE INTERMEDIATE COURSE. 29

the application. Under no circumstances shall the paint coat be
allowed to become in any way dirty, nor shall the surface mixture
be permitted to be applied to such a coat more than five hours after
the painting has been done.

"Owing to the inflammability of naphtha, the paint shall be
prepared at a distance from all fire or flame and shall be applied to
the surface of the concrete with the same precautions."

SUMMARY.

In the preceding chapter it appears that the use of a so-called
cushion coat that is to say, the application of the surface mixture
to the foundation in two courses instead of one is usually unsat-
isfactory and has generally been abandoned. The ordinary open-
binder course has been shown to be defective, owing to its lack of
stability, on heavy-traffic streets and the substitution for it of a
compact binder has been recommended, or, where economy is
desired, the use of a paint-coat to tie the surface mixture to the
foundation.

An open binder course of this description will no doubt con-
tinue to be very generally an element in the construction of the
majority of asphalt pavements which are subjected to only mod-
erate traffic.

PART II.

THE MATERIALS CONSTITUTING THE ASPHALT
SURFACE MIXTURE.

CHAPTER III.
THE MINERAL AGGREGATE.

THE asphalt surface, which directly carries the traffic and
which is intended to withstand the wear and tear of the same and
the action of the elements, is composed of a mineral aggregate
and an asphalt cement, that is to say, it is an asphalt mortar
or concrete.

The mineral aggregate consists of sand, in exceptional cases
also of stone, and a fine mineral dust or filler.

The asphalt cement consists of a native hard asphalt, or some
hard residue from an asphaltic oil or maltha, softened to the
proper consistency by some heavy petroleum oil, generally the
residual product of the distillation of petroleum.

Before considering surface mixture as a whole the constituents
which enter into its composition must be examined individually
and the variations which are met with in them noted.

The Mineral Aggregate. Sand Sand is the detritus of rock,
consisting of particles smaller than gravel and larger than silt,
and produced either by natural causes such as weathering and
water action or by the hand of man in crushing rocks mechanically.

Natural sand is the detritus, generally, of crystalline rocks

THE MINERAL AGGREGATE. 31

and commonly water borne and water worn, in which quartz
usually predominates, although calcareous sands, and those com-
posed entirely of feldspar or largely of other silicates, are known.

Artificial sand consists of the particles, produced in the process
of crushing rocks, which are of corresponding size to those which
make up natural sands.

Sand is the principal constituent of asphalt pavements, and
as such demands careful attention and study. Mr. A. W. Dow has
remarked in a paper before the Society of Municipal Improvements
in 1898:

"As sand is 90 per cent of the pavement, why should it not
be the most important ingredient to consider; and when a pave-
ment is at fault, why should it not be more responsible than the
asphalt which now bears the brunt of all failures?"

As a matter of fact it is now pretty well known that, even
if all the other constituents of an asphalt surface mixture are
of the best, the wearing surface will not prove a success unless
the sand is suitable for the purpose.

In the early days of the asphalt-paving industry but little
attention was given to the subject and the sand hi use was what-
ever happened to be the most available at the particular locality
where work was being done. Later, opinions varied as to whether
a coarse or a fine sand was more desirable, and there was a vibration
from one to the other, together with equally wide variations in
the consistency of the asphalt cement. In 1890 we find an expert
of that day stating that he is "pretty well convinced that sand
and matter that passes the 60 mesh ought not to enter the mix-
ture"; and again in 1892, having examined "ten old pavements
that have withstood wear," saying: "taking these results [of his
analyses] on the face of it, it is observed that in general the sand
used was on the fine side." No definite conclusions were drawn
at that time as to what a desirable sand was.

To-day we are better informed as to the best sand for a good
surface mixture, but unfortunately we know too little in regard
to the cause of the varying character of the particles composing
the quartz sand which is used. We have not been able to tell
why a certain Missouri River sand produces such a mushy mixture

32 THE MODERN ASPHALT PAVEMENT.

and is so unsatisfactory that its use has had to be abandoned, or
why a Platte River sand is possessed of peculiarities seen in that
from no other river.

Difference in the shape of the grain and in the character of
its surface are the probable causes, and these characteristics of
a sand are, therefore, probably next in importance to the composi-
tion and size of the grains in determining its suitability for paving
purposes.

Sorby, 1 who has studied the subject of sands carefully, has
classified them as follows:

" 1. Normal, angular, fresh-formed sand, such as has been
derived almost directly from the breaking up of granite or schistose
rocks.

" 2. Well-worn sand in rounded grams, the original angles being
completely lost and the surfaces looking like fine-ground glass.

" 3. Sand mechanically broken into sharp angular chips, show-
ing a glassy fracture.

" 4. Sand having the grains chemically corroded, so as to pro-
duce a peculiar texture of the surface, differing from that of worn
grains or crystals.

" 5. Sand in which the grains have perfectly crystalline outline,
in some cases undoubtedly due to the deposition of quartz upon
rounded or angular nuclei of ordinary non-crystalline sand."

In the paving industry all these sands have been met with,
but grains of several kinds not mentioned by Sorby are frequently
found. From an examination of several hundred sands from dif-
ferent localities the writer has been able to classify them, according
to their source, by peculiarities of composition, by the shape, and
by the surface of the grains, as follows:

Classification of Sand.
Source:

Commercially.
1. Beach sand.
Seashore.
Lakeshore.

Q. J. GeoL Soc., 1880, 36, 58.

THE MINERAL AGGREGATE. 33

2. River sand.

3. Bank sand.

4. Sand derived from soft sandstone.

5. Artificial sand.

Or with especial reference to their physical origin.

1. Beach sand.

Marine tidal action and sorting.
Lakeshore storm action and sorting.

2. Alluvial sand.

Subaqueous, recent.

Stream.

Lake.

Bank or pit deposits.
Glacial, stream, lake, etc.

3. ^Eolian sand.

Dune.

Volcanic.

4. From sandstone.

5. Crushed stone.

Composition:

Silica.
Quartz.
Hard clear.
Soft cloudy.
Ferruginous.
Silicates.

Shales and schists.
Feldspar.

Hornblende, Pyroxenes.
Calcareous.

Limestone.
Carbonates.
Shell.
Coral.

34 THE MODERN ASPHALT PAVEMENT.

Mixed composition.

Various kinds of quartz.
Quartz and silicates.
Quartz and carbonates.
Quartz and shell.

Shape:

Irregular.

Sharp angles.

Rounded angles.
Oval.

Worn by water action.
Round.

River.

Glacial.

Rock.
Crystalline.

Surface :

Sharp, original or fractured surface, not at all or little

worn.

Slightly worn on edges.
Smooth and polished "soft sand."
Smooth and with surface like ground glass.
Covered with cementing material.
Acted upon chemically.
Porous, coral sand, limestone sand, shells.

Size of grains :
Uniform.

Particles distributed in size, well graded.
Quicksand.

These different classes of sand may be described with special
reference to their use in the asphalt industry.

Beach Sands. Seashore. These are little used because as a
rule they are so sorted by currents of more or less uniform hydraulic
value that they are too much of one size. For example, a sand
found on Rockaway Beach, Long Island, is made up of grains
passing the following sieves:

THE MINERAL AGGREGATE. 35

Passing 200-mesh sieve 0%

100- " " 7

80- " " 32

50- " " 57

40- " " 2

30- " ". 1

" 20- " " 1

" 10- " "

100

It appears that 89 per cent of all the particles in the sand are
of 50- and 80-mesh size. The tidal currents are such that par-
ticles of smaller size are washed away while the larger ones
have been left behind in the movement of the beach sand to its
present location.

As far as character of the grain is concerned beach sands often,
and in fact in most cases, could not be improved upon. Fig. 2,
No. 1.

The most remarkable seabeach sands in the United States
are found on the eastern coast of Florida. They consist, on the
beaches of the northern part of the State, of pure white quartz
grains which have a fresh and angular fracture. Further south,
as at Lake Worth Inlet, they are made up of quartz and shell
fragments of about the same size. The grains are much coarser
owing to local conditions. Fig. 2, No. 2.

An explanation of the presence of quartz sand at a point so
far distant from any rock formations which contain this material
can only be arrived at by assuming that this mass of sand has
been transported down the coast by tidal action and ocean currents.

No. 30516 Lake Worth Inlet; bar sand; about 6 miles from Palm Beach,

Florida.

" 30534 St. Augustine; north beach; from dunes 10 to 15 feet high.
' ' 30535 " " " ' ' along high-water line ; river side.

" 30536" " " " " " " ocean side.

" 30547 St. John's Bluff.
" 30546 Mayport, Duval County; from 1 mile south of Mayport, \ mile

from ocean.
" 30548 Mayport, Duval County; from sand-dunes 10 feet high.

THE MODERN ASPHALT PAVEMENT.

Test number

30516

30534

30535

30536

30547

30546

30548

Passing 200-mesh

100-

80-

50-

40-

30-

20-

10-

100

T3 rk 4- ! r ^J -| f\ <(

AC/

1 1 C7

2 no/

xvetaineQ lu-

.4y ( -

l.l/p

U/o

^rklnHlo in TTP1

Ar QO7

1K.C7

2707

QO7

1 9O7

ooiuDie in xiui. . .

*O . \J /o

b/o-

o /O

1-^/0

The above sands are all pure quartz with the exception of
sample No. 30516 from Lake Worth Inlet, which contains 45.9
per cent shell detritus.

The beach sands of Cuba, to the west of Havana, are composed
entirely of small shell fragments, while those on the south coast
to the east of Santiago are either coral or, in Daiquiri Bay, largely
of hard silicates, the particles being transparent and of the rich
color of hornblende.

Seabeach sands are reputed to be far from sharp, but among
many recently examined for their suitability for use in the paving
industry most of them have proved sharper than river or bank
sand. Fig. 2, No. 1.

Beach Sands. Lakeshore. Another form of beach sand is
found on the shores of the numerous lakes near some of our large
cities, the great lakes in the North, and Lake Pontchartrain in
the South.

The assorting of the particles composing lakeshore sands is
accomplished largely by the movement of water produced by
storms, a more complicated one usually than is presented on the
seabeaches, although not as powerful as a rule. Tidal action is
of course absent. In many localities the force of the waves or
of the induced currents are so small as to permit of the deposition
of very fine sand or of that in which the particles are very well
graded in size. In other cases there is a great similarity between
lake and seabeach sands. The remarkable variation in the size

FIG. 2. Sand Grains-

THE MODERN ASPHALT PAVEMENT.

of lakebeach sand, even on beaches within a few miles of each
other, is illustrated by the following examples:

Lake Michigan.

Lake Erie.

Kenosha, Wis.
1899.

Chicago, 111.
1897.

Sandusky,
Ohio.

Pas

sing 200-me
100-
80-
50-
40-
30-
20-
10-

'sh sie
t

ve
t

16
52
13
3
2
4

100

10%
68
15
3
3
1

100

o%
11

23
24
32

7
2
1

100

In the Kenosha sand, although the uniformity of grade is not
carried as far as was the case on Rockaway Beach, 52 per cent
of the particles are of a size to pass a sieve of 50 meshes to the
inch. And, again, we have finer sand from near Chicago of still
greater uniformity. On the other hand, near Sandusky a lake
sand is available which is of quite varied size of grains. Here the
sorting of the sand particles has been limited and the grading
is satisfactory for use in asphalt surface mixtures without modi-
fication. Where lake sands are of too uniform size two or more
sources of supply may be used and mixed in suitable proportions.
Other typical beach sands from Lakes Michigan, Erie, and Ontario,
which are in the writers' collection, sift as follows :

Milwaukee
Beach.

White Fish
Bay, Wis.

Lake
Ontario.
1894.

Lake
Pontchar-
train.

Lake Erie.

1892.

Passing 200-mesh. . .

2%.

100- . . .

80- ...

50- ...

40- ...

30- ...

20- ...

10- . ..

100

THE MINERAL AGGREGATE. 39

The very considerable variations in the size of these sands
from different sources make it possible, however, by mixing those
of different sizes, to produce a sand of any grading that may be
desired for a surface mixture, and that, too, without great labor.

A peculiarity of lakebeach sand is the rapidity with which
all the sand, which may be of most desirable character for asphalt
work, may be removed from any particular locality by a violent
winter storm and its place taken by a sand of quite different grad-
ing. This is of common occurrence between Chicago and Mil-
waukee, and no doubt elsewhere, so that the fact that a suitable
sand can be found at a particular point during any one working
year does not mean that the sand will be of the same grading
another year, especially if the storms of an intervening winter
have been heavy. The variation in the available sand from year
to year in this way makes a decided difference in the character
of the asphalt mixture turned out at different times in cities which
are dependent on such a source of supply.

Lakebeach sands, originating at old lake levels, may, like
alluvial sands, be found at times in banks or pits where changes
of lake levels, which are so frequent in geological time, have left
them above the elevation of, and at times far distant from, the
present water level. For commercial uses these cannot be dis-
tinguished from beach sands of more recent origin.

Alluvial Sands include all those which have been moved by
and deposited from running water as distinguished from beach
sands. They may be found to-day in the beds of streams or along
their shores and in banks and pits where they have been left by
running water in past geological times. Alluvial sands may
also include those originating in glacial streams and occurring
both in banks and pits and as reassorted by recent water action.
There are also deposits in lakes from streams flowing into them
which need not enter into our consideration, being rarely so avail-
able as to be of technical importance.

River Sands. As river sands are conveniently considered
those which are found in the beds of streams or on their beaches,
and which are still largely subject to the action of water. They
are oftener found at the concave side of some bend or at a place

40 THE MODERN ASPHALT PAVEMENT.

where the current makes a change in direction or loses its force.
Geikie describes the deposition of river sand as follows:

"While the main upper current is making a more rapid sweep
round tho opposite bank, under currents pass across to the inner
side of the curve and drop their freight of loose detritus, which,
when laid bare in dry weather, forms the familiar sand-bank or
shingle-beach. Again, when a river well supplied with sediment
leaves mountainous ground where its course has been rapid and
enters a region of level plain, it begins to drop its burden on the
channel."

feiver sands are usually obtained by dredging and are thus
distinguished from bank or pit sand, which are worked from dry
deposits. They are most varied in character and form an impor-
tant part of the supply in use in asphalt paving. In each river
they seem to have distinguishing peculiarities, and in no two cities
having their source of sand in a river bottom is the supply of
the same character, while for any one city the supply may vary
in character from year to year.

This may be due to two causes : to the different nature of the
rock formations from which the sand found in different rivers
is derived and to the different physical conditions to which the
debris from these formations has been exposed, resulting in pecu-
liarities of shape, size, surface, etc.

River Sand at Kansas City and Washington, D. C., etc. River
sands are or have been in use in Washington from the Potomac, in
Kansas City from the Missouri and from the Kansas, in Omaha
from the Platte, and in St. Louis from the Mississippi and Missouri.
No two of the sands resemble each other in their behavior in an
asphalt surface mixture. The peculiarities which each one shows
can be described, but as yet it is impossible to show why they
all differ so much in their adaptability to making a desirable
surface mixture.

In one of these cities for many years river sands were in use
in the surface mixture without the fine bank sand now mixed
with it. The river sands would not carry a sufficient percentage
of asphalt and made a surface which marked badly under traffic,
perhaps more from their coarseness than from other reasons, but
which, in comparison with the coarse Washington mixtures of

THE MINERAL AGGREGATE.

Potomac River sand, which do not mark in the same way, shows
a decided difference in character, not due to the size of the particles
of which it is composed.

Two typical surface mixtures from the two cities will illustrate
the difference due to the sand. A street in the western city marks
up more than most of the pavements in that town, that is to say,
badly. The average mixture laid in Washington in 1894 with
a straight Potomac River sand scarcely marked at all. The com-
position of these two mixtures is as follows:

Kansas
City, 1893.

Washington,
1894.

Bitumen.
Passing 2

" 1
tt

tt
tt
tt
tt
tt

9.9%
9.0
6.3
5.4
36.3
10.8
8.2
6.5
7.6

100.0

10.9%
9.9
1.5
3.7
16.1
28.9
20.7
5.9
2.4

100.0

00-mesh sie
00-
80-
50-
40-
30-
20- " '
10- " '

It would be natural to expect that the Washington mixture
with the higher percentage of bitumen and coarser sand would
be the softer and mark more than the western mixture, but that
is not the case. With a grading not far different and apparently
less favorable, the Potomac sand will carry 1 per cent more bitu-
men than that in use in the West and still not mark in hot weather.
The possible difference in sands from different rivers is well illus-
trated in this case.

As striking differences are to be seen between the mixtures
made with river sand in two western cities, which may be denoted
No. 1 and No. 2. It is not difficult to get sands in either city which
can be properly graded to our present accepted standard and
which, it would be supposed, from all appearances would make
equally excellent asphalt surface mixtures. On making the mix-
tures, however, it is found that the river sand at city No. 1 would
not hold the usual amount of asphalt cement and that the varia-
tions in amount from one box of mixture to another was so great
as to make any uniformity in working impossible.

THE MODERN ASPHALT PAVEMENT.

River Sand at City No. i. In 1896 an attempt was made to
devise a satisfactory mixture for work in this city. A coarse
sand was obtained from one river and a fine sand from another.
They had the following mesh composition:

Passing 200-mesh sieve

19%

" 100- "

" 80- "

" 50- " ....

" 40- "

" 30- "

" 20- " "

" 10- " "

100

These sands were combined in such proportions as to make a
suitable grading and asphalt cement added until the paper test 1
showed a suitable amount. The mixed sand would hold at
the most but 142 pounds of asphalt cement to the 9-foot box of
material, and would frequently carry only 126 pounds, and yet
the mixture was very sloppy, where a New York mixed sand,
weighing about the same per cubic foot, would carry over 160
pounds and stand up firmly.

The grading of the western sand and that from New York,
for comparison, was as follows:

City No. 1.

New York.

Passing 200-mesh sieve

" 100- " "

" 80- '" " ...

" 50- " "

" 40- " "

" 30- " "

a 20- " "

" 10- " lt

Weight per 9-foot box, Ibs

100

846

100
875

" cubic foot, Ibs

A C per box, Ibs

126-142

163

Per cent Trinidad A. C. in mixture

13.0-14.0

15 7

Pages 352-356.

THE MINERAL AGGREGATE.

It is impossible at present to explain the difference between
the two sands, but it must be one of shape and surface of the grains
rather than of volume per cent of voids, there being no great
difference between them in this respect.

The use of these sands was abandoned for the reasons which
have been given, although the work done with the mixture made
at that time has been fairly satisfactory. Such a mixture required
too much watching owing to the rapid changes in proportions
which were necessary. The experience has proved, however, very
instructive and has shown that many mushy mixtures do not
prove as bad under traffic as they look when hot, but may give
good service ; and that the asphalt cement with such s^nd may be
held at a point as shown by the paper stain, which with sand
from other sources would be dangerous.

Later the sands in use in this city were both taken from the
same river, one being a coarse sand and the other finer. They
have been carefully selected by the yard foreman, who. has gone
out with his sieves on the dredge and taken only sand of a certain
grade.

Typical specimens of these river sands sift as follows:

Coarse.

Fine.

1896.

99.

1898.

99.

Passing 200-mesh sieve
." 100- '

20%
24

17%
40

25%
42

11 80- "

50-

40-

" 30- "

" 20-

" 10-

100

Retained on 10-mesh sieve

44 THE MODERN ASPHALT PAVEMENT.

And the mixed sands as coming from the hopper:

Passing mesh. . .
1898

200
9%

100

14%

80
30%

28%

10
3%

-100%

1899

6
10

10
9

25
25

36
9

9
16

3
9

= 100
= 100

13
5
17

7
13

18
18
17

27
18

15
21
15

9
9
9

6
9

4
4
4

= 100
= 100
= 100

The sands in 1898 and 1899 carried from 154 to 165 pounds
of asphalt cpment to the 9-foot box and are more satisfactory than
those previously in use on account of their greater regularity.
The later mixtures have averaged in composition as follows, which
may be compared with that made formerly with the unsatisfactory
sand:

Earlier Sand.

Later Sand.
First Year.

Later Sand.
Second Year.

Biti
Pas

jmen .

9.9%
14.3
9.4
13.3
33.0
10.3
6.1
2.6
1.1

100.0

11.3%
13.2
9.7
14.7
37.2
5.7
3.8
2.7
1.7

100.0

10.4%
13.0
10.0
9.0
25.0
15.0
8.0
6.0
4.0

100.4

sing 200-me
100-
80-
50-
40-
30-
20-
10- '

jsh sieve

t i

The second mixture of the year 1899, was the most satisfactory
of any made up to that time, but even here undesirable features
are to be found, which will be considered later.

The peculiarities of these sands are, however, very instructive
if at the same time very trying in the asphalt business.

River Sand at City No. 2. In city No. 2 the river sand presents
a most desirable grading, as shown by the following sifting of some
in use in July, 1899:

THE MINERAL AGGREGATE

200-mesh sieve
100-

80-

50-

40-

30-

20-

10-

3%
26
12
39
14

100

2%
19
19
41
12

100

This sand is peculiar, however, in that in making a mixture
according to our practice it is found that if the asphalt cement
is added in amount only sufficient to stain the paper * to the
same degree as with the sands in other cities, the bitumen hi the
mixture does not exceed 9 per cent. With a larger amount of
asphalt cement, sufficient to yield 10 per cent of bitumen in the
mixture, the latter is very sloppy.

This peculiarity might lie either in the heavier volume weight
of the sand and the smaller voids or in some characteristic of the
surface of the grain which would prevent the usual amount of
asphalt cement from adhering to it. It has been found, however,
on the use of this same sand in a neighboring city that a satis-
factory surface containing as much as 10 per cent of bitumen can
be laid by making the mixture as rich as is necessary for this figure,
and disregarding the usual indications of richness and overfill-
ing of the voids. The finished surface does not appear to be exces-
sively soft in summer and wears well. This would point to the
correctness of the idea that the peculiarity of this sand lies in its
surface and perhaps in its shape. Nothing peculiar in either of
these respects, as far as can be seen under the microscope, can be
discovered, the sand being a nearly pure quartz, having a ground-
glass surface with rounded angles. Fig. 2, No. 3.

In considering the subject of mixtures these peculiar sands
will be referred to again. 2

1 This paper test will be described later, pages 352-356, 514. It indicates
the amount of asphalt the sand will carry.

2 Page 351.

46 THE MODERN ASPHALT PAVEMENT.

The river sands of the United States in the Mississippi Valley
seem, therefore, to be possessed of peculiar properties which we
are unable as yet to account for, and it is necessary to handle
them in the paving industry in a different way from other sands,
or else to reject them.

Bank or Pit Sands. Bank or pit sands are deposits of sand
which have been laid down in their present position by various
agencies in past geological times, as distinguished from the river
and beach sands, which have been described and which are the
results of the recent assorting of detritus by water action, or by
the reasserting of bank sands under changed conditions, which
is quite possible, as on the north shore of Long Island, where the
glacial bank sands are often reasserted by water action into modern
beach sands.

Bank sands are of the most varied derivation river, beach,
glacial, seolian, etc. including all possible sources of origin. On
the Hudson are found banks of river sand, as at Croton ; on Long
Island banks of glacial sand, as at Cow Bay, and in Sioux City,
Iowa, banks of seolian sand in the loess.

Bank sands grade in size from fine gravel or coarse concrete
sand to the impalpably fine ones which are found in those wind-
blown deposits of a large portion of the West, called loess, and
which are often almost entirely a sharp sand composed of quartz
particles fine enough to pass a 200-mesh sieve.

To the paving industry the bank sands offer sources of sup-
plies which are more varied in the size of the particles of which
the sand is made up than the river and beach sands of recent
origin, and are oftener to be obtained of that degree of fineness
which has been found to be such an essential feature in our modern
mixtures, that is to say, of 80- and 100-mesh size. The varied
grading of different bank sands, not including the coarser con-
crete supplies which are not used in the asphalt industry for sur-
face mixture, is illustrated by the following characteristic speci-
mens (see table, p. 47).

These bank sands, it will be seen, admit of combinations of two
or more in such a way as to attain any required grading. Fortu-
nately no bank sands are met with which present any of the pecu-

THE MINERAL AGGREGATE.

liarities of the river sands of the Mississippi River Valley,
all carry asphalt well and make good mixtures.

They

New York Supply, 1899.

Boston Supply.

Passing Sieve
of Meshes.

Cow Bay.

Corbin's
Bank,
Steinway.

Delagoa
Bay
Ballast.

Braintree,
Mass.

Canton,
Mass.

200

12%

100

Buffalo
Supply,
Attica, N. Y.

Elmira
Supply.

Utica Supply.

Lafayette
Supply.

Toronto,
Ontario,
Supply.

200

32%

29%

100

Kent, England, Glacial.

Louisville Bank.

Sioux City.

White.

Yellow.

200

tr.

48%

99.5%

100

tr.

100

100.0

48 f HE MODERN ASPHALT PAVEMENT.

bank satids are unsuitable for use in the surface mixture,
however, owing to the presence of too much clay or loam, or to a
surface on the grain which is more or less covered with a ferru-
ginous cement, existing in the original rock from which the sand
is derived, or with argillaceous matter, to neither of which surfaces
does asphalt adhere satisfactorily. The former peculiarity is of
the commonest occurrence, but probably only becomes serious
when the amount of clay or loam is too large to be taken care of
as dust, or is in such a form as to ball up in the sand-heating
drums, or not to mix with the asphalt cement properly. A loamy
tempering sand has been in use successfully in an Ohio River
city for several years. The latter difficulty is typical of the red
sands of New Jersey, which have a coating of iron oxide firmly
adherent to their surfaces, and the sands found associated with
the London gravels which are similarly but not so distinctively
coated. From the latter sand a coating of asphalt cement would
wash off in a few weeks when exposed to the weather, destroying
the surface mixture made with it. The red sands of New Jersey
may possibly be used without danger; that from Rutherford has
been as a tempering sand, but they do not look attractive and
are suspicious.

There are no other noticeable peculiarities of bank sands to
be mentioned which, as far as is known, render them unsuited
for surface mixtures, except the presence of too much 200-mesh
material which consists of sand grains of that size and not dust.
Sand of this size is, as a rule, disadvantageous in a mixture and
makes it mushy and liable to push or shove under traffic. The
peculiarities of mixtures containing much 200-mesh sand will be
discussed later.

Quicksands. Any of the preceding sands is often called quick-
sand if it is very fine. Quicksands are really of that peculiar
nature only when they consist of particles largely finer than will
pass a sieve of 200 meshes to the inch and consequently having
a small hydraulic value. When such sands have their voids filled
or more than filled with water they are unstable and mobile. There
is no reason why a coarse sand should not at the same time be a
quicksand if it is supported and its voids more than filled by a

THE MINERAL AGGREGATE.

force of water of sufficient head. It has often been stated that
quicksands consist of uniform sized and round particles, but recent
examination of some material of this description 1 has shown that
they generally consist of sharp grains and are often well graded.
Several quicksands have been examined by the writer with the fol-
lowing results, which are characteristic of such material :

QUICKSANDS BOSTON, 1897; WORCESTER, 1900.

Test No. 11541. Boston Neponset Valley Sewer. Nat. Contr Co.
it it 11542 " " " " " " "

" " 11544 " lt " ll " " "

" " 30723. Worcester Green Street. H. P. Eddy.

" " 30724 ' ' " "

it QQ725 " " " it u

Finest-ground limestone, all passing 200 mesh.

Xos. 11541, 11542, and 30725 are clean sands, grains all sharp. Nos.
11544, 30723, and 30724 contain a small amount of clay, less than 1 per cent,
not subsiding entirely in one week.

Test Nc

Voids in
pacte(
Weight
foot o
pounc

Sieve.

200
100
80
50
40
30
20
10

hot com-
i sand. . .
per cubic
same in

s . .

11541
29.3%
117.2

11542

11544
40.2%
99.1

65.5%
13.7
17.8
2.0
1.0

30723
36.7%
103.8

47.2%
19.6
11.2
13.0
4.0
3.0
1.0
1.0

30724
34.7%
106.1

11.6%
11.4
9.0
39.0
12.0
10.0
3.0
3.0
1.0

30725

Finest-
ground
Lime-
stone.

39.3%
100.4

63.5%
17.7
18.8

Diameter,
Milli-
meters.

.035
.065
.09
.17
.23
.31
.50
.67
1.00
2.00

Greater
than 2.

19.2%
7.9
18.9
34.0
11.0
7.0
1.0
1.0

H.2%
14.2
19.6
22.0
8.0
16.0
4.0
2.0
2.0

79.7%
9.5
9.8
1.0

1.0

100.0

100.0
2.0%

1CO.O

100.0

1 Landreth, Wm. B., The Improvement of a Portion of the Jordan Level
of the Erie Canal. Trans. Am. Soc. C. E., 1900, 43, 596.

50 THE MODERN ASPHALT PAVEMENT.

It is apparent that some quicksands are quite widely graded,
others consist to a very large extent of uniform particles smaller
than .035 millimeter in diameter (.0014 inch) and even finer than
the finest dust in Portland cement. That they are all composed
of sharp grains, contain traces only of clay or none, have an ex-
traordinary fineness, in one case greater even than that of the
best ground limestone, are their astonishing characteristics.

The angularity of these small particles is explained by their
small hydraulic . value and consequent freedom from attrition,
as shown by some experiments of Daubree, quoted by Geikie,
Text-book of Geology, third edition, page 385. He says:

"In the series of experiments already referred to, Prof. Daubree
made fragments of granite and quartz to slide over each other
in a hollow cylinder partially filled with water and rotating on
its axis with a mean velocitiy of .80 to 1 metre in a second. He
found that after the first 25 kilometers (about 15J English miles)
the angular .fragments of granite had lost T 4 of their weight,
while in the same distance fragments already well rounded had
not lost more than -j-J-^ to J}Q. The fragments rounded by this
journey of 25 kilometers in a cylinder could not be distinguished
either in form or general aspect from the natural detritus of a river-
bed. A second product of these experiments was an extremely
fine impalpable mud, which remained suspended in the water
several days after cessation of the movement. During the pro-
duction of this fine sediment the water, even though cold, was
found in a day or two to have acted chemically upon the granite
fragments. After a journey of 160 kilometers, 3 kilograms
(about 6J pounds avoirdupois) yielded 3.3 grams (about 50
grains) of soluble salts, consisting chiefly of silicate of potash.
A third product was an extremely fine angular sand consisting
almost wholly of quartz, with scarcely any feldspar, nearly the
whole of the latter mineral having passed into the state of clay.
The sand grains as they are continually pushed onward over each
other upon the bottom of a river become rounded as the large
pebbles do. But a limit is placed to this attrition by the size
and specific gravity of the grains. As a rule the smaller particles
suffer proportionately less loss than the larger, since the friction

THE MINERAL AGGREGATE. 51

on the bottom varies directly as the weight and therefore as the
cube of the diameter, while the surface exposed to attrition varies
as the square of the diameter. Mr. Sorby, in calling attention to
this relation, remarks that a grain y 1 ^ of an inch in diameter
would be worn ten times as much as one T ^o of an inch in diameter,
and a pebble 1 inch in diameter would be worn relatively more by
being drifted a few hundred yards than a sand grain y^ of an
inch in diameter would be by being drifted for a hundred miles.
So long as the particles are borne along in suspension, they will
not abrade each other, but remain angular. Prof. Daubree found
that the milky tint of the Rhine at Strasburg in the months of
July and August was due, not to mud, but to a fine angular sand
(with grains about -$ millimeter in diameter) which constitutes
TTJ irinr of the total weight of the water. Yet this sand had
travelled in a rapidly flowing tumultuous river from the Swiss
mountains, and had been tossed over waterfalls and rapids in its
journey. He ascertained also that sand grains with a mean diam-
eter of T V mm. will float in feebly agitated water, so that all
sand of finer grains must remain angular. The same observer has
noticed that sand composed of grains with a mean diameter of
mm. and carried along by water moving at a rate of 1 metre per
second is rounded and loses about y o^W f its weight in every
kilometer travelled."

These remarks explain some of the characteristics of the quick-
sands which have been described.

So-called quicksands consisting largely of particles of 100-
and 80-mesh size form one of the most valuable sand supplies which
can be used in the paving industry. They are known as temper-
ing sands, and when mixed with the ordinary sand produce a
grading which is more satisfactory than that of any sand deficient
in such fine particles.

Glacial Sand. Such sands are found on the north shore of
Long Island and are largely used in the paving industry in the
city of New York. Fig. 2, No. 4.

In those parts of the country which were covered with the
ice sheet during the Glacial Period a large part of the beach, lake,
and river sands may consist of glacial material reasserted by

THE MODERN ASPHALT PAVEMENT.

more recent water action, but this is, of course, not true of sands
from regions south of the terminal moraine, nor is it probably
the case with the sands found in our western rivers, which are
of very recent origin. This may account for the fact that the
sands from the Mississippi and Missouri Rivers and their tribu-
taries are so different from many other river sands.

Sands Derived from Sandstones. Supplies of sand are to be
found at times which are obtained by grinding and breaking down
loose sandstones of little coherence. These sands are largely used
in glass-making and are usually very fine quartz. They have
been offered in several cities for paving purposes. Two samples
sifted as follows:

Passing 200-mesh sieve

' ' 100- ' '

" 80- "

" 50- "

" 40- "

" 30- "

" 20- "

" 10- "

100

They have never been utilized in the paving industry.

Artificial Sands. The finer material produced in crushing
rock for the purpose of obtaining broken stone for concrete when
screened to a proper size is a sand. It differs essentially from
the natural sands in that it has not been subjected to weathering
or attrition and consequently is sharp and has a rough surface.
Fig. 2, No. 5. It is generally well graded through different
sizes and has low voids. Specimens of such a sand are represented
by the screenings from the crushing of granite, gneiss, limestone,
and trap in various parts of the East, which sift as follows:

Test No. 68417. Crushed gneiss screenings from Jerome Park Reservoir,

New York.

" " 66563. Trap-rock screenings, Nyack, N. Y.
" " 62082. Limestone screenings, Muskegon, Mich.
" " 64840. " " Owosso, Mich.

" " 69721. " " Harrisburg, Pa.

" " Washington, D. C.

THE MINERAL AGGREGATE.

Tes
Pas

Ret

t number

68417

6.5%
7.5
5.0
12.5
12.5
8.0
10.0
8.0
17.0
3.5
9.5
0.0

66563

10.5%
7.5
2.5
7.0
4.0
5.0
5.5
15.5
34.0
8.5
0.0

J o.o

62082

17%
3

64840

7%
5
4
8
5
5
6
18
42

69721

28%
12
7
13
9
9
11
11

21%
14
8
11
8
6
8
19
5 l

100

sing 200-rm
100-
80-
50-
40-
30-
20-
10-
1-incl

1 *- "
1- "
ained on 1-

;sh sie

inch s

ve
ie\

28 l

100.0

100

1 Retained on 10 mesh.

These screenings are of excellent character for use in hydraulic
concrete and have also been successfully made part of the mineral
aggregate of surface mixtures, as the finer material which they
contain is very desirable in localities where the particles of the
same size are not found in the native sand.

The screenings from the softer limestones of the middle and
far West are not so desirable as a substitute for native sand, and
no attempt has been made to use them in an asphalt-surface mix-
ture as sand. When ground they form a most desirable filler, as
they readily become impregnated with asphalt.

Purchase of Sands. Iu this connection it may be well to state
that no sand is desirable for use in an asphalt-surface mixture
which would be considered suitable for use in lime or cement mortar.
It is therefore often difficult to explain to sand-dealers the kind
of sand needed in the asphalt industry and more often difficult
to find it, as there is no demand for such sand from others and
consequently little inducement to open up or develop pits or
supplies of the proper kind. It is well usually to say, in asking a
sand-dealer for sands for surface mixtures, that one wishes a sand
that is too "soft" for any other use, and that one which is suitable
for mortar would be of no value for pavements.

Composition. The composition of a sand, as long as the grains
are hard, cannot seriously affect its availability for use in an asphalt-

54 THE MODERN ASPHALT PAVEMENT.

surface mixture or be a cause of defects in it. Soft-grained sand
should be rejected when this is possible. The grains of a very large
majority of all the sands in use in the asphalt industry, in fact
of almost all natural sands, are composed of quartz, the silicates,
feldspar, etc., being decomposed and removed from the detritus
of the original rock by weathering or water action. The character
of the quartz may vary largely, however. It may be a clear, trans-
parent, hard quartz, a softer cloudy-white quartz, or an even softer
ferruginous one. The two latter suffer more from attrition, have
round forms and dead surfaces. Rarely sands are found which
consist of a large proportion of silicates or of shales, although a
small amount of magnetite and the harder pyroxenes are to be
detected in most sands. Potomac River sand often carries 3 per
cent of magnetite, and that from Siboney beach, which has been
used in Santiago, Cuba, must consist fully half of hard silicates
such as hornblende and similar minerals, which are, however, not
unsuitable for paving purposes. A sand formed by weathering
of a granite rich in feldspar on Cape Neddick in Maine is largely
made up of coarse particles of feldspar, but it is not of any prac-
tical importance.

The sand found in the Mohawk and Hudson River Valleys
from Poughkeepsie to Geneva consists largely of small oval and
flat particles of shale, although some quartz is present. These
sands are, of course, comparatively soft, but good work for light
traffic has been done with them.

Calcareous sands are rather unusual, except those derived
from shells and the detritus of coral. Limestones weather out
or dissolve too rapidly in water to permit of the formation of
sand, although they are found to a certain extent in admixture
with quartz sands at many points.

Coral and shell sands are common in southern latitudes, as
in Cuba and Bermuda, for instance. They are the softest sands
that are met, usually crumbling under moderate pressure. Few
particles finer than 80-mesh size are found in the coral sands,
as these are readily washed away and dissolved. The shell sands
of Cuba are firmer than the coral sands.

Mixed sands composed of quartz and silicates, and quartz

THE MINERAL AGGREGATE.

and carbonates, occur. Such include the more evenly propor-
tioned Siboney beach sand and the shale sand of Northern New
York. The lake sands often contain considerable carbonates
in the form of shells, and a small amount is found in almost all.
Determinations of lime in several supplies, made in 1896, gave
the following results:

Source of Sand.

Per Cent CaO.

Buffalo lake

9 3%

New York bank

2 2

St Louis river

Kansas City river

Boston bank

Their presence has no injurious influence on the sand as far
as its use in asphalt surface mixture is concerned. Asphalt cements
probably adhere better to limestone than to silica or silicates, and
in this way it may be desirable.

It is sometimes of interest to determine how much or what
percentage of a sand is in a condition soluble in strong acid, such
as hydrochloric. The amount found in the sands mentioned
above, which were examined for lime, was as follows:

Source of Sand.

Per Cent
Soluble.

Per Cent Iron
Oxide and
Aluminum.

St Louis river

2.2%

Kansas City river

2.4

1.4

New York bank

3.4

2.3

Boston bank

3.6

2.4

Buffalo lake

12.9

1.5

In the ordinary run of sand this is small and of no importance.
In some of the sands, like the red sands of New Jersey, it is indic-
ative of a weakness in the material.

Sands may often carry admixtures of substances which cannot
be considered as part of the original material from which they
have been derived but which are adventitious in some way or

56 THE MODERN ASPHALT PAVEMENT.

other. Clay and loam are the commonest substances and their
presence is accounted for in two ways. Either they are inti-
mately mixed with the sands in the deposits, as in the fine bank
sand mentioned as being used for tempering purposes, or they
exist in strata covering or adjacent to the sand and are unavoid-
ably mixed with it in collecting the latter. If not adherent to
the grain a small amount will act merely like the dust added to
the sand before the asphalt cement, but if the clay or loam balls
in the drums and is not screened out it may prove injurious.
A clean sand is in any case probably more desirable, although
satisfactory results have been obtained with many loamy ones.
Organic matter in the shape of vegetable debris is sometimes
found in sand. It is usually removed in screening the hot sand
as it comes from the drums. If this is not possible and the amount
remaining is excessive the sand should be rejected.

Shape of Sand Grains. The shape of the grains of which sands
are composed is quite varied.

Irregular Grains. Fresh detritus from the original rock
is generally irregular in shape and with sharp angles unless it is
derived from a rock composed of grains which have already existed
as sand. Sands formed of irregular grains are not common, but
they are often found with the grain quite irregular in shape but
the angles somewhat rounded by water or glacial action. The
sands from the north shore of Long Island are of this class. Fig. 2,
No. 4, shows the peculiarity of the grain. They are of very irregu-
lar shape with re-entrant angles, but are slightly rounded with
loss of the sharpness of the original fragment.

Crystalline Grains. Sands containing crystals are mentioned
by Sorby. They have been met with by the author but once in
the United States. In a sand supply used in Atlanta, Ga., some
years ago, there could be detected with a glass quite a large pro-
portion of well-shaped quartz crystals some of which were twins.
Fig. 2, No. 6, shows the irregular weathered crystals forming this
sand.

Such sands are undesirable for an asphalt surface mixture for
reasons too obvious to require mention.

Oval Grains are worn much more smooth by continued water,

THE MINERAL AGGREGATE. 57

tidal or glacial action, the original angles being mostly or entirely
lost. Tidal action has a peculiar tendency to produce grains with
one diameter longer than the other, as particles with this peculiarity
arrange themselves with their longer diameter in the direction of
the force of the waves and are then worn still more into this shape.
Seabeach sands are supposed on this account to be far from sharp,
but that on the Florida beaches is markedly so. Fig. 2, No. 1.

Round Grains. Round-grained sands are not uncommon. They
are oftener found in river, glacial, and aeolian sands, which are
worn by being rolled over and over and polished against each
other like the well-known spheres of quartz prepared by the
Japanese. Fig. 2, No. 7. It is probable that a very large per-
centage of sand is composed of the somewhat irregular, rounded
grains.

The shape of the grains of a sand has a marked influence, when
combined with their size and grading, upon the character of the
asphalt surface mixture made with them, the closeness with which
they can be packed together depending to a very considerable
degree on their shape. Round sands and oval sands can be com-
pressed much more readily than sharp ones, owing to the smaller
friction between the smooth surfaces, although they may not
on this account be found to pack as closely eventually. Whether
the shape of the grains in the sands in use hi paving mixtures has
any effect sufficient to account for the cracking of some surfaces
more than others is a question for investigation. It may so
influence the character of the voids in a mineral aggregate as to
do so. This will be considered later. Mixtures made with round-
grained sands are of course less stable and mark more easily in
summer than those made with sharp sand, since round particles
move much more readily over one another than sharp ones ; but,
on the other hand, with plenty of filler this tendency can be neu-
tralized, while the round-grained sands can be packed much more
readily and closely and with smaller voids and the resulting sur-
face can, in this way, be made denser.

Surface of Sand. The character of the surfaces of sand grains
is very different, much more so than would appear from mere
ocular examination. Under the microscope sand grains are found,

58 THE MODERN ASPHALT PAVEMENT.

as shown in our classifications, with the surfaces of the original
material of which the grains are composed, or with the surfaces
derived from fracture of this material. There are surfaces polished
by attrition and water action, surfaces like ground glass, Fig. 2,
No. 7, originating in the same way, surfaces coated with the
cementing material originally uniting the grains into a sandstone,
surfaces acted upon chemically, and the porous surfaces of lime-
stone and coral grains. The different kinds of surfaces behave
quite differently toward asphalt cement. The porous limestone
surfaces absorb it and it, of course, adheres very firmly. To
the quartz surfaces the bitumen adheres in most cases well, but
in others only slightly, being readily washed off with water. Some
quartz grains will carry a heavy coat of asphalt cement, others
but a small and thin one, as can be detected by examining with
a glass mixtures made with different sands.

These peculiarities can be explained by the difference in the
capacity of surfaces of different character for retaining or adsorb-
ing bitumen, the film adhering being thicker in one case than in
another. They have a very decided effect upon the character
of different mixtures and may well be the cause of cracking in
surfaces made with certain kinds of sand. As has been already
remarked, a sand from the London gravels has a surface such that
asphalt cement would not adhere to it in the presence of the damp-
ness of a London fog, so that it would be found in the gutters
after a rain, washed quite clean and free from bitumen.

That even water has strong chemical effect on the surface of
sands, thus altering its character, can be seen from Daubree's
experiments already alluded to. Geikie quotes him as follows:

"Daubree endeavored to illustrate the chemical action of
rivers upon their transported pebbles by exposing angular frag-
ments of feldspar to prolonged friction in revolving cylinders of
sandstone containing distilled water. He found that they under-
went considerable decomposition, as shown by the presence of
silicate of potash, rendering the water alkaline. Three kilograms
of feldspar fragments made to revolve in an iron cylinder for a
period of 192 hours, which was equal to a journey of 460 kilo-
meters (287 miles), yielded 2.720 kilograms of mud, while the

THE MINERAL AGGREGATE. 59

5 litres of water in which they were kept moving contained 12.60
grams of potash, or 2.52 grams per litre."

Of course quartz grains are not attacked like the feldspar,
and it is for this reason that in sands resulting from the decomposi-
tion of rocks containing feldspar none of the latter remains and
the grains of which it is composed are all quartz, hornblende, etc.

From what has been said in the preceding paragraphs the very
variable character of sand, apart from the size of the grain, will
be readily understood. All of these characteristics demand care-
ful consideration in the selection of sands for successful asphalt
surface mixtures. But more important even than these considera-
tions is the size of the grains whicl^ go to make up any sand.

Size of Sand Grains. The size of the sand grains in an asphalt
pavement, that is to say their average diameter, is of the greatest
importance, as will be found in studying the subject of surface
mixtures. Sands occur hi nature in which are found every size
grain from the impalpably fine quartz of silt to fine gravel. In a
standard sheet asphalt surface it has been found generally prefer-
able to have no sand grains larger than 2 millimeters in diameter,
passing a 10-mesh sieve made of wire .027" in diameter, or smaller
than .17 millimeter, which pass a sieve of 100 meshes to the inch,
made of wire .0043" in diameter. It is always possible to exclude
the coarser grams, but it is not so easy to get rid of the material
which is too fine.

Sands are differentiated into various sizes by means of sieves
somewhat arbitrarily made, but so selected for use in the asphalt
industry that the average diameter of the particles shall bear a
definite relation to each other. The finest sieve in use, 200 meshes
to the inch, made of wire .00235" in diameter, passes particles of
all degrees of fineness up to a diameter of .10 to .083 millimeter.
The coarsest parti les passed by this sieve are plainly sand and are
generally round and usually undesirable. The next sieve in
use has 100 meshes to the lineal inch and is made of wire .0043"
in diameter. It passes particles having a diameter of about
.17 millimeter, or double that of those passing the 200-mesh sieve.
The next, made of wire .00575" with 80 meshes, passes particles
of a diameter of .23 to .24 millimeter, three times that of the 200;

THE MODERN ASPHALT PAVEMENT.

the next, made of wire .0090" in diameter, 50 meshes, particles
of about .37 millimeter, or four times the finest sieve. The 40-
mesh sieve, made of wire .01025", however, passes at a jump to a
particle six times the diameter; the 30, made of wire .01375"
in diameter, to one of eight times; the 20, made of wire .01650"
in diameter, to one of about twelve, and the 10, made of wire
.027" in diameter, to one of about twenty to twenty-five times
the diameter of the largest grain passing the finest sieve. A more
minute differentiation than this of the size of the sand grains
would be burdensome and of no advantage. For this reason
sieves of 150, 90, 70, and 60 meshes to the inch are not used in the
paving industry. As an example of the use of these sieves the
great difference between the size of the particles of a sand suitable
for hydraulic concrete and that suitable for an asphalt surface
mixture will serve and is seen to be quite marked. In the former
a coarse sand is sought, in the latter a fine one.

Hydraulic
Concrete Sand.

Paving Sand.

Passin

g 200-mesh sieve ...

Trace

1%
2
,17
18
24
22
16

100

17%
17
30
13
10
8
5

100

100- ' "

80- ' "

50- ' "

40- ' "

30- ' "

20- ' "

10- " "

It will be observed in the above siftings that the results are
Btated in percentages of the material passing the various sieves.
This method of statement is much to be preferred to that in which
the percentages remaining on the different sieves are given, and for
two reasons. In the first place it enables one to judge of the
size of the grains from the diameter of the meshes and the size
of the particles passed by such a mesh, which cannot be done
from the percentage remaining on the sieve. In the second
place, as will appear from the description of the method of making
a sifting in Chapter XXVIII, it is much more easy to remove the

THE MINERAL AGGREGATE.

fine material which passes the 200-mesh sieve by attrition of the
coarser grains one upon another and upon the cloth of the sieve
than to separate out the coarser grains first and afterwards weigh
the remaining 200-mesh material, especially as there is apt to
be a loss of this material during the process of sieving, which
makes no difference if it is sifted out first and the amount deter-
mined by loss.

The sieves which are used for the sifting are carefully made in
large lots by one firm from the same lot or roll of cloth woven
for the purpose, so that several sets shall be so nearly alike as to
give uniform results even in different hands.

COMPARISON OF THE SIEVING OF TWO SAMPLES OF SAND
ON DIFFERENT SETS OF SIEVES IN DIFFERENT LABO-
RATORIES.

Sample number

Siftings

Lab. No. 1

Lab. No. 2

Tab. No. 1

Lab. No. 2

Passing 200-mesh sieve

15%

10%

10.5%

100-

18.9

80-

10.5

50-

34.7

" 40-

11.5

30-

6.3

" 20-

3.2

" 10-

3.1

100

98.7

While the agreement in the siftings obtained with sieves made
in this way is as concordant as could be expected a similar agree-
ment will not be found among sieves obtained from different
sources, especially in the case of the finer ones with 80, 100, and
200 meshes to the linear inch. In explanation of the difference
in the character of the cloth woven by the different manufacturers
Messrs. Howard & Morse, of Brooklyn, N. Y., who manufacture
the sieves in use by the author, have the following to say in reply
to certain questions which were propounded to them.

In reply to the inquiry as to where the cloths in use for making
these sieves are manufactured they state that:

62 THE MODERN ASPHALT PAVEMENT.

"Wire cloth of iron, steel, brass, or copper, from the coarsest
to No. 100 mesh, is made in this country, while the finer meshes
are imported from France, Germany, and Scotland. We think
the Scotch cloth is the best. Even 100 mesh can be imported at
a lower cost than the rate paid our weavers will allow it to be
produced. . . . We do not believe any American manufacturer could
be induced to make the necessary outlay to produce cloth finer
than 120, unless the field for its usefulness was very much enlarged,
as the imported cloth at a much lower cost seems to answer every
general purpose for which such cloth is used. Moreover it might
be necessary to import the workmen/ 7

As regards the process of manufacture they say:

"Beginning with wire, say, &" in diameter, the mill draws
down to smaller sizes until too hard to safely draw smaller; it
is then annealed, when, its ductility being restored, it is drawn
down finer, and then reannealed, and the process repeated until
the requisite size is obtained and it is given its final annealing,
to render it fit for fabrication into cloth. The wire is delivered
by the mill to the wire-cloth manufacturers in skeins, which are
rewound on spools according to the mesh required. Usually the
process is as follows:

"Take for instance 100 mesh. We desire to put on a 'warp'
say 36"X300 feet. This will make 3 'cuts' each 100 feet long.
We estimate the weight, allowing for waste and 'thrums/ and
taking a little over one-half the total weight, we divide this as
equally as we may among the 100 spools, being careful that none
are under weight. The spools are placed in a rack as closely as
convenient; the wires from each spool are led through a device
which prevents their crossing (and serves other purposes of a
nature somewhat complicated to explain) to the 'back-beam of
loom. 7 The 100 wires are what we term an inch. They are
tied together at the 'end and hook over a bolt-head in a slot which
runs lengthwise of the beam, which in our looms is about 5'
circumference and 52" long. For 36" width we hook our first
inch on peg 18" from centre. For 300 feet we would need to put on
60 'rounds,' i.e., the beam (which is really a heavy hollow cylinder)
is turned 60 times, and the 'lease' wire separating the contigu-

THE MINERAL AGGREGATE. 63

ous wires alternately above and below is put in place and the
beam turned up one more revolution to allow for distance from
back-beam to face of loom.

"The inch is then secured to another peg, which is firmly
secured in the partition between the inches, these partitions form-
ing 'grooves ' in which the wires lay, and the bunch of 100 wires
is then cut off and the second groove receives its 60 rounds, and so
continued till the 36 grooves or inches are filled. This is a slow,
tedious process.

"However careful may be the windings of the spools, whatever
device may be used for spool racks, yet the wire will catch and
break, and it is necessary to repair every break before another
round is turned up on the back-beam. If the spools run too freely,
the wire comes off too fast, and a * bight ' will draw into a kink
and snap even with comparatively heavy wires, or the bight will lay
across other wires and catching may snap a dozen or more. How-
ever, the warp being on the back-beam, one inch is taken off the
peg, and the wires, being separated by the ' lease/ are passed each
one in the order in which it lies in the grooves, first through the
'gears/ or 'heddles/ and next through the 'reed.' The heddles
consist of two frames about 8 to 10 inches high by the width of
the loom in length, secured in a variety of ways; to these frames
are attached twisted wires with an eye in the centre. For 100 mesh
each frame must contain at least 50 of these twisted wires to each
mesh. The 100 wires we have loosed from the back-beam are passed
in their exact order alternately through back and front gear, or
rather 50 are passed through the eyes of the back-gear and the other
50 passed between the wires of the back-gear through the eyes of
the front.

"These gears being operated by treadles, it is evident that
when the back-gear is down and the front one is up from the normal
line in which wires would tend to lie, that a ' shade, ' or shed, is
formed, every alternate wire being in the upper, while the others
are in the lower shade, or shed. The shade at the gears may be
3", while at the face of the cloth it is nil, and in front of the reed
when swung back as far as the lay will carry it may be 2". The
lay, or ' beater, ' is hung overhead and is provided with a groove

64 THE MODERN ASPHALT PAVEMENT.

in what is known as the bottom shell, and also in the top shell,
which is removable and adjustable vertically to suit reeds of various
heights, which are so placed in the shells as to have free lateral
play, but very little in any other direction. All the inches being
successively passed through the gears and reed, they are properly
fastened to a ' sacking ' which leads from the face of the reed around
a ' breast-beam ' down to the ' cloth-beam/ on which it is wound up
as the cloth is made.

"The reed consists of a series of ' splits ' made of flat steel of pecul-
iar temper. In a 100-mesh reed they would probably be about
&" wide, and the thickness of each split would be equal to a trifle
less than the space between the wires of the cloth. They are
compacted into reeds by a process of lacing, which must be very
particularly done, as the split must stand square to plane of cloth,,
parallel, and evenly spaced, the spaces being a trifle more than the
thickness of the wire which passes through them and there must
be exactly 100 in each and every inch. The warp being already
to commence weaving, the weaver, stepping on the treadle, opens
his shade and throws his shuttle through, catching it on the other
side of the piece, the lay is brought up one blow and he changes
his treadle and gives a second blow to place the shot. After throw-
ing 100 shots and giving 200 blows he has completed 1 inch, when
he proceeds to count it and thus discover whether he is driving the
' woof, ' or ' filling, ' up too hard or too lightly to place 100 trans-
verse wires in 1 inch.

"Until fairly started his warp-wires will be constantly break-
ing in fine cloth, and it is a constant contest of patience, with
unavoidable delays, until the last shot is thrown, though always
worse at the beginning and gradually diminishing. After 2 or 3
inches have been made the weaver gets the ' blow 'necessary to secure
the required fineness in the mesh, and many become very expert and
exact; that is, we thought it was exact until Mr. Richardson called
our attention to many inaccuracies and defects. Being as good
(perhaps better) than similar work from other factories, it sold
and we heard no complaint of inequalities until this cement testing
question brought us face to face with a different problem."

The size of the wire with which any cloth is made will of course

THE MINERAL AGGREGATE.

have a decided influence on the width of the meshes of the cloth
for any given number per lineal inch. Messrs. Howard & Morse
state that while cloth of the same mesh can be made of many
different sizes of wire, as shown for the coarser sieves by the ordinary
trade catalogues, the difference is more theoretical than practical
when we go beyond the 60- or 70-mesh sieves. For the four finer
sieves in use in the asphalt-paving industry the diameter of the
wire, the mesh, and the space between the meshes are intended to
be as follows:

Mesh.

Diam. of
Wire.

Space.

No. 50
" 80
" 100
" 200

No. 35 O. E. gauge wire. . .
" 38 " " " ...

" 40 " " *';

" 42JB. & S. wire

.02
.0125
.01
.005

.009
.00575
.0043
00235

.011
.00675
.0055
00265

In reply to the question as to why the ordinary cloth is much
more regular in the number of meshes to the inch in the one direc-
tion than it is in the other the makers of the sieves say:

"If the reed be exact the cloth must have the proper number
of holes one way, as it is governed by the reed. The reason that
cloth sometimes has fewer holes the other way is that it is governed
by the blow given by the weaver. If he can pass coarse cloth
under the eye of the inspector, he gains the few missing shots in
each inch and the same number of blows may in a day's work gain
him 5 to 10 per cent more pay, but it is not so often the intention
of the weaver so to deceive. A warp of 100 may be put on the loom
in December and not be out until the following June. It is slow
work. Consider the effects of various temperatures and other
causes which may affect a man's disposition meanwhile. Too gay
or cheerful, you would be obliged to check his blow which would
drive cloth too fine. In brisk, cool weather cloth would be driven
up finer than in warm, uncomfortable weather. Again, a fresh
start in the morning means better cloth than that made in the
later hours of the day. We have been accustomed to pass a coarse-
ness not exceeding 5 per cent; i.e., we would accept 80X76 as a
square mesh. Again, the wire is not even in either temper or size.

66 THE MODERN ASPHALT PAVEMENT.

Hard wire or wire a trifle larger than it should be will not 'go
to place' with the same blow that soft, proper gauge wire would
require. We select all wire as carefully as possible, and though
a great difference is not common in a single skein, yet the writer
has gauged a skein of brass -wire which has shown a full size dif-
ference when gauged at both ends. Not being wire-drawers, we are
at a loss to account for this. It seems almost impossible that a
die should be worn so perceptibly in drawing less than a mile of
wire, and yet one end of the skein may be round while the other
is elliptical in cross-section. These causes and others all tend
to an irregularity of mesh which shows that the weaver is not
entirely at fault.

"To eliminate any question of nicety of touch and skill on the
part of the weaver, fortunes have been sunk in experiments to pro-
duce an automatic loom; but the other causes remain, and though
they affect the accuracy to a less degree in a power loom, yet
they have a strong influence. We have power looms that will
make the cloth exact, but cannot use them for anything finer than
20 or 30 mesh, and with only a medium-weight wire. Another cause
that may affect the mesh is the inequality of 'temper/ or in
other words, speaking of other metal than steel, we should per-
haps say, 'malleability/ hardness, or softness, but we have come
to use the word ' temper/ however incorrect it may be, in reference
to all metals used in fabrication of wire cloth.

"The degree of heat to which wire is subjected in the anneal-
ing process may vary with the different charges; more than
this, it may vary with the heat applied to the different skeins in
a single charge, and more troublesome still, it may be hotter on
one side of the skein than on the other. This makes serious in-
equalities in the 'temper' and in consequence a variation in the
mesh of the cloth in which it is to be used."

Messrs. Howard & Morse also add as a reason for the great
variability in the sieves offered by the trade:

"That each wire-cloth manufacturer has ideas of his own
as to what the trade requires: for instance, he may use 48 reed
for 50 mesh and have his cloth driven up to 44 the other way,
so that he will furnish you on your call for 50 mesh a piece 48X44,

THE MINERAL AGGREGATE. 67

while the manufacturer who gave you 50X50 on your call gives
you a cloth that costs him more, both for material and labor than
the other.

"Again, certain trades, notably the paper trade, requires cloth
not driven up square, and 48x38 passes for 50 mesh, 68x52 for
70 mesh, and if this cloth passes through other channels, as a dealer's
hands, he will sell a piece to any transient customer, say 58X46.
and call it 60 mesh, and with entire innocence, for he bought it for
60 mesh.

"Mill strainer cloth is another irregularity. It is made of
very light wire and not driven up with any approach to accu-
racy. In fact the low price obtained for it prohibits care in its
manufacture. It is either woven by boys or on a power loom,
which, as explained above, will not insure accuracy in fine meshes."

The Committee on Uniform Tests of Portland Cement of
the American Society of Civil Engineers at one time hoped that
all sieves in use in the testing of cement might be constructed
of wire the diameter of which should be one-half the width of the
opening. If sieves could be constructed on this plan it would
no doubt be very desirable, but in response to an inquiry as to
whether this could be done the manufacturers make the following
statement :

"To make fine brass cloth with the diameter of wire one-half
the width of opening would result in a flimsy fabrication which in
use would give you more unsatisfactory results than you now
attain.

"The individual wires would be of very little use and not only
break very easily, but would push to one side or the other; two con-
tiguous wires crowding each the neighboring wire would separate
and give space four times the size natural to the mesh. We find
that in order to make a fairly rigid cloth it is necessary that the
diameter of wire be about 80 per cent of space size, or practically
as shown in table on page 65. To make the lighter wire would
increase cost, though we presume that is of minor importance, and
yet it must be considered. It would be difficult to make this per-
fectly clear perhaps. To take a sample case of frequent occur-
rence: Our customers know that in the coarse grades of cloth

68 THE MODERN ASPHALT PAVEMENT. .

for a certain mesh the price diminishes as diameter of wire
decreases, and this is true up to about 60 mesh. We make this
from stock of No. 36 O. E. gauge wire (.0075"). They order No. 37
in hope of obtaining it at a lower price per square foot. The
weight of 60 of No. 37 is 75 per cent as great as the weight of 60 of
36 ; and yet material for 37 costs about 40 per cent more per pound
than 36. This difference becomes greater as we advance to the
finer sizes. No. 200 mesh is made of .00235" wire (as near as the
micrometer gauge will show it). The finest wire the writer ever
saw was silver .002" in diameter, and this was shown as a curiosity
rather than of any practical use.

"It may be that the ductility of brass is sufficient to make
it practicable to draw it to .0017", but we doubt it, and at
what expense? 105,263' to one pound, i.e., to draw one pound
.0017" brass wire about 20 miles of wire must pass through the
dies. This is getting down to fine work. It means about sixteen
days' work to one pound, and the finer the wire the more slowly
it must be drawn. We do not mean weight, for that is evident,
but as regards length. Imagine, too, the making of the die. Can
one expect it to be round, square, or true to any regular shape,
or exactly accurate with regard to size?

"Take even our 80 mesh, the wire wherein is .00575" in diame-
ter, nearly 3 times the diameter of the wire you specify for the
200 mesh, 11 J times as large in area of cross-section, consequently
11 times as heavy in a given length, and contemplate the skill
required to make a hole in a die which shall be round and with
an exact diameter smaller than the hole in 100-mesh cloth. Con-
sider the care necessary in drawing this wire, constant watching
to note when the die is worn too large, and the whole manipulation
until wire is woven into cloth and put into sieves, and there will
be apparent reasons sufficient for the inaccuracies noted by you
from time to time.

"Up to 100 mesh we can make a cloth as accurate as any one
in the trade; beyond that we cannot control it. We will write
parties in Glasgow in a few days, and if we can learn any-
thing of interest to you will be glad to communicate it to
you.

THE MINERAL AGGREGATE. 69

''We are willing to put on a short warp, say 36"X 100 feet each
Nos. 50, 80 and 100 mesh, guaranteeing that it shall be driven
up square, i.e., the 50 to between 48 and 50, and the 80 between
77 and 80, and the ICO to between 97 and 101, the wire carefully
selected to conform to size given in answer to No. 5 (see table
on page 65), within 2 per cent of diameter, provided you will
agree to find sale for same, at list price net, within one year of
completion.

"We are aware that we are undertaking a hard piece of work.
The delays will be expensive; we shall expect to pay more than
the usual price for the wire; every skein will need to be gauged
at both ends, and if long, in the centre as well; much of the wire
will have to be discarded; the mill contests our claims of inaccuracy;
the cloth will have to be carefully and constantly watched, and
the supervision will be onerous, but we are desirous of giving
you all the satisfaction possible, though naturally without pecuniary
loss to ourselves, and we therefore do not consider the whole
cost of supervision in naming the above price. "

The preceding explanation will enable the reader to grasp the
difficulty of obtaining satisfactory sieves more readily than any-
thing that could be said by the author, and it will show the care
which is used in order to obtain, for use in the asphalt industry,
sieves which are at least uniform among themselves and which
can be considered as standards, at least arbitrarily if not abso-
lutely.*

That a better grade of 2CO-mesh cloth can be obtained which
is much more regular than the average supply can be seen from
the accompanying illustration, Fig. 3, where the good can be dis-
tinguished from the poor cloth without difficulty, and where it
can be seen that the lack of regularity is due to the manner in
which the weaver pushes up the wire.

Until 1907, the 200-mesh cloth was twilled, but since that
time Messrs. Howard & Morse have produced a plain woven cloth,
which in some ways is more satisfactory as far as the number of
mashes per inch is concerned, but will probably not wear as well
as the twilled material. Three counts of this cloth recently made,
resulted as follows :

70 THE MODERN ASPHALT PAVEMENT.

Warp. Woof.

First piece 200 197

Second piece 200 197

Third piece 200 198

while the 100-mesh cloth made at the same time with great care
counted

Warp. Woof

First piece 100 98

Second piece 100 102

These counts are much more satisfactory than those which
were formerly obtained.

Good Cloth. Poor Cloth.

FIG. 3.

The sieves are generally known, as has appeared in what has
been said, by the number of meshes which they show to the linear
inch. For strictly scientific purposes they would be more properly
identified by the diameter of the largest particle which each sieve
would pass, and these diameters have already been given for the
sieves which are in use in the asphalt industry. For the purpose
of determining this diameter Mr. Allen Hazen adopted a method
which he describes as follows: 1

twenty-fourth Annual Report of the State Board of Health of Mas-
sachusetts, 1892, 541.

THE MINERAL AGGREGATE. 71

"It can be easily shown by experiment that when a mixed
sand is shaken upon a sieve the smallest particles pass first, and
as the shaking is continued larger and larger particles pass, until
the limit is reached, when almost nothing will pass. The last and
largest particles passing are collected and measured, and they
represent the separation of that sieve. The size of separation of
a sieve bears a tolerably definite relation to the size of the mesh,
but the relation is not to be depended upon, owing to the irregu-
larities in the meshes and also to the fact that the finer sieves
are woven on a different pattern from the coarser ones, and the
particles passing the finer sieves are somewhat larger in proportion
to the mesh than is the case with the coarser sieves. For these
reasons the sizes of the sand grains are determined by actual
measurements regardless of the size of the mesh of the sieve. . . .

"The sizes of the sand grains can be determined in either of
two ways: from the weight of the particles or from micrometer
measurements. For convenience the size of each particle is con-
sidered to be the diameter of a sphere of equal volume. When
the weight and specific gravity of a particle are known, the diameter
can be readily calculated. The volume of a sphere is fad 3 , and
is also equal to the weight divided by the specific gravity. With
the Lawrence materials, the specific gravity is uniformly 2.65
within very narrow limits, and we have

Solving for d we obtain the formulae

when d is the diameter of a particle in millimeters and w its weight
in milligrams. As the average weight of particles when not too
small can be determined with precision, this method is very accu-
rate, and altogether the most satisfactory for particles above
.10 millimeter; that is, for all sieve separations. For the finer

72 THE MODERN ASPHALT PAVEMENT.

particles the method is inapplicable, on account of the vast num-
ber of particles to be counted in the smallest portion which can
be accurately weighed, and in these cases the sizes are determined
by micrometer measurements. As the sand grains are not spher-
ical or even regular in shape, considerable care is required to ascer-
tain the true mean diameter. The most accurate method is to
measure the long diameter and the middle diameter at right angles
to it, as seen by a microscope. The short diameter is obtained
by a micrometer screw, focusing first upon the glass upon which
the particle rests and then upon the highest point to be found.
The mean diameter is then the cube root of the product of the three
observed diameters. The middle diameter is usually about equal
to the mean diameter, and can generally be used for it, avoiding
the troublesome measurement of the short diameters.

"The sizes of the separations of the sieves are always deter-
mined from the very last sand which passes through in the course
of an analysis, and the results so obtained are quite accurate."

Voids in Sand. Sand consists of particles of such shapes that
they cannot be packed in any space without leaving intervals
between the grains which are not filled. These spaces are known
as voids and their volume and the size of the spaces are very impor-
tant in characterizing different sands. The amount of volume
of the voids and their size are dependent on the shape and variety
in the size of the grains, upon the way they are arranged, and upon
the degree to which they are compacted. If the sand grains were
perfect spheres it can be regularly calculated what the percentage
of voids in any volume will be. Dr. G. F. Becker, of the United
States Geological Survey, has made this calculation for me and
states the results as follows:

"Suppose, first, that 8 spheres of radius r are so arranged that
the centre of each is at one corner of a cube and that the edge
of the cube is 2r. Then one-eighth of each sphere will be included
within the cube, and the total of the spherical matter in the cube
will be just one sphere. In this case the voids will be

THE MINERAL AGGREGATE.

"Now shift the four spheres forming the lower layer into this
order

which amounts to changing the angles made by the edges of the
cube without any change in length. Also bring the centres of
the upper layer of spheres over the point marked "x". Then
the cube is distorted into an oblique prism of which this is a plan.

"It still includes just one sphere.
face of the prism will be

The area of the lower BUT-

74 THE MODERN ASPHALT PAVEMENT.

and the height of the prism is the height of a tetrahedron formed
by the centres of four spheres when three are placed in contact
in one plane and the fourth is laid upon them. This height, and
thus the whole volume of the prism, is

and the interstitial space is

- = 1 ^==0.2595

4v / 2r 3 3\/2

"By diminishing all lines in one direction in the same propor-
tion the system of spheres becomes a system of ellipsoids. Since
the spheres and the interstitial spaces are distorted in the same
manner, it is evident that a system of equal and similarly oriented
ellipsoids may also be packed so as to leave only 25.95 per cent
of voids. But if any ellipsoid is differently oriented from the rest,
the voids will be greater."

With the spheres or regularly oriented ellipsoids packed as
closely as possible the voids should therefore be 25.95 per cent,
but it is not possible to pack small grains like sand in this regular
way. They are jumbled together irregularly, and experiment has
shown that perfect spheres like small bird shot when shaken
and tamped until they are as compact as it is practical to make
them with the devices at our command still contain voids of 32 per
cent, or 6 per cent more than theory.

With spheres of quartz of similar size it would probably be
impossible to compact them to the same extent, while if the material
is merely poured loosely into the space which is to be filled and no
attempt is made to attain greater compaction the voids will be
found to be very much larger.

THE MINERAL AGGREGATE. 75

It is important therefore before going further to consider the
conditions under which determinations of voids are usually made
and to decide upon a uniform method, one which is the best to
use in obtaining results of both absolute and relative value. There
are several difficulties to be met with in arriving at the ultimate
practical compaction of sand aside from the impossibility of bring-
ing small particles of any size and shape into the closest possible
juxtaposition.

Determination of the Voids in Sand. The usual method of
determining voids in sand, gravel, stone, etc., is to fill a vessel
of known volume with the material in the condition under which
it is desired to find the voids and then to pour in water until the
space between the grains is filled, thus determining the voids
from the relation of the volume of water to the volume occupied
by the material examined. This method is satisfactory with
coarse substances like gravel, but not when grains smaller than
those which will pass a 30-mesti sieve are present. One reason
why it is not satisfactory with fine material is because air is liable
to be entangled between the grains and not to be replaced by
water. According to another method a definite volume of the
compacted material is poured into a measured volume of water.
The increase in volume is that of the material and the difference is
that of the voids. This is a more satisfactory way if no fine material
or dust is present, which has a tendency to float in water or to
obscure the meniscus.

It is better, therefore, to determine the specific gravity of the
material of which the grains are composed and then calculate
the voids from the weight of a definite volume of sand. For
instance, if 100 cubic centimeters of quartz sand weighs 168
grammes and the specific gravity of the grains is 2.65 the voids
may be found by dividing the weight by the gravity multiplied
by 100 and subtracting the results from 100, the result in this case
being 100-168-^2.65 = 36.6 which is the per cent of voids by vol-
ume in this sand.

Where the ultimate practical compaction of fine material is
sought, some further precautions must be taken, as this cannot be
accomplished at all at ordinary room temperatures, as a film of

THE MODERN ASPHALT PAVEMENT.

adsorbed aqueous vapor adheres to each grain, which keeps them
apart. With dust or 200-mesh material alone, air also is often
imprisoned in the mass and increases the voids. If, however, the
sand or dust is heated above the temperature of boiling water these
difficulties are removed and the fine grains compact as well as
coarser ones. As examples, the voids found 'in a sand and in a
ground limestone when determined with hot and cold materials
respectively were :

VOLUME OF VOIDS.

Cold.

Hot.

Sand
Dust.

34.2
56.6

33.3
38.0

In each determination the means employed for compaction
were the same, but only when the material was hot was this com-
plete, especially with the fine dust. The degree of compaction to
which a substance such as sand is subjected of course influences
the extent of the voids. The relation between those in a mass
loosely poured into a cylinder and slightly shaken down but
not compacted and those in a thoroughly compacted mass can
be seen from the following determinations:

VOIDS.

Loosely
Compacted.

Thoroughly
Compacted.

Long Island sand coarse
Buffalo fine
Chicago

41.7
45.4
42.2

33.7
37.9
35.5

New Orleans very coarse

37.6

29.6

Kansas City very fine

4G 9

36.0

The voids in sands and mineral aggregates should of course
be determined in a state of thorough compaction, as this is the
condition in which these materials are or should be found in a
finished asphalt surface.

THE MINERAL AGGREGATE.

Weight per Cubic Foot. The voids being known in any sand
or aggregate, and the gravity of the particles (for all practical
purposes for quartz sand this may be taken as 2.65), the weight
per cubic foot is calculated and is usually stated together with
per cent of voids, as it is a factor of some importance hi consider-
ing the relations of bitumen to the aggregate in certain surface
mixtures, sand being frequently taken by volume 9 cubic feet
and as an indication of possible density of the finished surface.

The very considerable variations in the voids and weight per
cubic foot in various unmixed sands examined in 1894 and later
are seen in the following tables, pages 77 and 78.

VOLUME WEIGHTS AND VOIDS IN SANDS.

Wt. per
Cu Ft.
Loose.

Voids.

Wt. per
Cu. Ft.
Compact.

Voids.

Ballast very fine .

45 4

102 7

37 9

Buffalo No* 3

97 1

41 2

110 4

33 2

New Orleans Prophet Island coarse. . .
' ' screened.
" lakeshore

103.4
100.9
95 1

37.6
39.1
42 5

116.4
113.6
109 1

29.6
33.3
34

1 ' ' ' Jordan River much loam.
Omaha . .

82.4
98 6

51.2
40 4

95.3
109 1

42.4
34 i

London coarse

94 9

42 6

107 8

34 8

fine

86 6

44 9

99 9

39 6

Yonkers coarse

94 6

42 8

106 6

35 6

" fine

92 9

43 8

106 5

35 6

Chicago. .

95 6

42 2

107 7

Boston.

43 3

106 4

35 7

New York

94.4

43.0

108 3

34 6

Kansas City fine

46 9

105 9

' coarse

101 7

38 5

112 9

31 7

Altoona

92 4

44 1

105 1

37 7

Youngstown

92 4

44 1

105

36 5

Niagara Falls No. 1

92.1

44 3

107 6

n (( <i 2

89.4

99 8

Louisville

91.9

44.5

101 6

38 6

Fort Wayne

91 4

44 7

108 8

34 3

Washington ... .

89 2

46 1

100 1

39 5

Harrisburg (much coal)

83.4

41 3

Voids as Affected by Size and Shape of Particles and by their
Uniformity. With the method of determining the voids in sands
in a uniform and satisfactory way in mind, the peculiarities found
in the aggregation of particles of mineral matter of different shape
and size may be considered.

Voids in Sand Consisting of Particles of Uniform Size. If the

THE MODERN ASPHALT PAVEMENT .

particles of a sand are all of uniform size, or nearly so, the voids
present under our normal conditions will depend upon the shape
of the particles alone. Sand the grains of which are perfect
spheres would have, as has been noted for fine shot, voids of
32 per cent, theory being 25.95, whereas irregular fragments,
such as are found in crushed stone, the crushed quartz of the
sand-paper manufacturer, contain practically about 44 per cent,
but if the particles are uniform their absolute size has no influence
on the volume of the voids they show. As examples, the following

GRADING, POUNDS PER CUBIC FOOT, AND VOIDS IN VARIOUS
SANDS OF 1899.

Test
No.

Source.

Passing Mesh.

Total.

Retained
on 10.

Lbs.
per
Cu.Ft.

4
1

200

100

22693
22694
22768
22769
22789
22790

22802
22803
22827

22917
22925

22967
22968

Paterson, Sand Hill
No. 16

3
4
3
2

23
17
2

6
2

22
3
2

9
12
15
3
2

24
50
5

20
2

15
51
13

6
14
23
3
3

18
7

12
2

28
17

36
42
56
19
53

28
13

12
16
43

18
3
28
31

8
2
26

10
18

8
2
13

13
8

18
6

5
2

18
3

9
2

= 100%
= 100%
= 100%
= 100%
= 100%

= 100%
= 100%
= 100%

= 100%
= 100%

= 100%
= 100%
= 100%,

10%

104.7
102.2
104.7
97.7
99.7

103.4
102.2
107.2

107.2
112.1

105.9
99.7
104.7

36.6
38.1
36.6
39.6
39.6

37.4
36.1
35.1

35.1

32.1

35.8
39.6
36.6

Paterson, Sand Hill
No 17

Scranton, Temper-
ing No 1.

Scranton, Temper-
ing No. 2
Louisville No. 1 ,
coarse

Louisville No. 1 ,
fine

Saginaw,fine bank.
Sag. River
La Fayette No. 9,
Wagner bank. . . .
Kansas City, Kaw
River, coarse. . . .
Washington No. 3,
crusher dust

Chicago, fine sand. .
' ' coarse sand

determinations of voids in crushed quartz of uniform but very
varied size will sieve:

THE MINERAL AGGREGATE.

Crushed Quartz very Sharp.

Volume Per
Cent of
Voids.

Passing 6-mesh sieve not passing 10. . .
" 20- " " " " 30. . .
" 90- " " " " 100. . .

43.3
43.4
44.2

These materials represent a very coarse sand, a sand of size
in use in concrete, and a very fine sand but all of the particles
of very uniform size. They all contain, when compacted hot,
the same volume of voids. Without compaction material of
this description will, however, vary hi proportion to the fineness
of the particles; a coarse broken stone of uniform size will have
47 per cent of voids, while a similar fine sand will often have over
50, owing to the fact that the finer material will not naturally
assume the same degree of compaction as the coarser material.

The crushed quartz, the voids in which, when compacted,
have just been given, consists of extremely sharp particles with
the angles of the original fracture unworn. In sands the grains
are more or less round as has been shown, and hi consequence
they pack more readily and closely. If sands are taken and sepa-
rated into portions the particles of each one of which are of nearly
the same size that is to say, will pass a certain sieve but be
retained on one of the next smaller size and the voids are deter-
mined for each of these, it will be found that there is a very con-
siderable variation hi the voids for the same sized particles of
different sands and also for the different sizes, and that with all
of them the voids are smaller hi volume than was the case with
the sharp particles of crushed quartz. Following are illustrations;

THE MODERN ASPHALT PAVEMENT.
SIFTING.

Source.

Passing Mesh.

Total.

200

100

Buffalo Canada lakeshore
Omaha Platte River, 1897
Chicago fine lake, 1897. . .
Detroit fine lake, 1897. . . .
Kansas City fine river,
1897 '

17
5
2
2

10
10
32

24
13
74
10

15
12
29

15
34
23
23

8
17

24
31
1
59

44
27
19

4
3

4
4

2
2

= 100%
= 100%
= 100%
= 100%

= 100%
= 100%
= 100%

6
15

1
12
1

7
1

"9"

Long Island bank, 1897. . . .
Buffalo Attica fine bank. .

See also tables on pages 81 and 82.

The crushed quartz of about the 100-mesh size contains 44.2
per cent of voids. The 100-mesh particles of the fine Attica sand
showed but 42.9 per cent and the same sized grains of the other
sands much less, until those from the coarse Buffalo supply con-
tained but 34.5 per cent.

Voids in Sand of Varying Sized Particles or Grains. When the
grains in a sand vary in size the voids are at once reduced in volume
by the fitting of the smaller particles into the spaces between the
larger ones, the voids at the same tune becoming smaller in size
as well as in volume.

THE MINERAL AGGREGATE.

VOIDS.

Source.

Loose Hot.

Entire.

Passing Mesh.

200

100

Buffalo Canada lake l

40.3%
35.8
46 3

'48.'6%

41.8%
43.7
43.6
45.8
45.9
49.0
45.7

38.3%
42.6
46.6
44.0
44.6
48.1
47.5

42.0%
42.9
46.8
41.3
42.9
47.3
47.1

Omaha Platte River, 1897

Chicago fine lake, 1897

Detroit fine lake, 1897

41 1

Kansas City fine river, 1897. . . .
Long Island bank, 1897
Buffalo Attica, fine bank, 1897

41.0
39.2

49.8
50.1
48.0

Source.

Tamped Hot.

Entire.

Passing Mesh.

200

100

Buffalo Canada lake *

34.6%
33.5
39 2

'44:5%

34.5%
38.4
38.4
39.6
40.2
41.8
42.9

32.8%
37.5
40.2
39.1
40.1
43.4
41.3

36.5%
37.5
42.3
37.5
38.8
42.4
41.4

Omaha Platte River, 1897.
Chicago fine lake, 1897 .

Detroit fine lake, 1897.

35 4

Kansas City fine river, 1897. . . .
Long Island bank, 1897
Buffalo Attica, fine bank, 1897. .

35.3
33.0
32.0

42.7
42.1
37.7

1 1.33 per cent magnetite.

When the various sized particles, the voids in which taken sepa-
rately are shown in the preceding table, are combined in the pro-
portions found in nature, the voids are then, with one or two ex-
ceptions, much reduced. This is conspicuous in the Attica sand,
which as a whole shows 32 per cent of voids, where its 200 mesh
has 37.7 per cent and the coarser particles over 40 per cent.

The greatest reduction will of course occur where there are
enough fine particles present of a size small enough to fit between
the larger ones for instance, dust with sand, sand with gravel,
and gravel with broken stone. It was found that by adding dust

82 THE MODERN ASPHALT PAVEMENT.

VOLUME WEIGHT OF HOT SAND, POUNDS PER CUBIC FOOT.

Source.

Loose Hot.

Tamped Hot.

Entire

Passing Mesh.

Entire

Passing Mesh.

100

Buffalo Canada lake ! . . . .
Omaha Platte River, 1897
Chicago fine lake, 1897. . .
Detroit fine lake, 1897
Kansas City fine river,
1897

98.5

93.'5
97.2

97.4
100.3

99.4

96.0
92.9
93.1
89.5

89.3
84.2

89.6

101.8
94.6
88.1
92.5

91.5

85.5

86.7

95.6
94.2
87.8
96.9

94.3
87.0

87.3

108.0
109.7
100.3
106.6

106.8
110.8

112.2

108.2
101.6
101.6
99.7

98.7
96.1

94.2

110.9
103.1
96.7
100.6

99.1
94.0

96.9

104.8
103.1
95.3
103.2

101.1
96.1

96.8

Long Island bank, 1897. .
Buffalo Attica, fine bank,
1897

1 1.33 per cent magnetite.

in continually increasing portions to a sand with 35.5 per cent
of voids, the percentage of voids in the mixture was gradually
reduced to a certain point corresponding to the voids to be filled,
but on further addition they were again increased, as shown by
the following figures. This point was reached when the dust
amounted to 41 per cent of the sand, an amount greater than the
voids; but this is due to the fact that the sand grains were un-
avoidably separated to a very considerable extent by the dust and
the voids consequently increased.

WEIGHTS AND VOIDS IN NEW YORK SAND WITH VARIOUS
PERCENTAGES OF 200-MESH DUST.

Weight per
Cubic Foot.

Voids.

Original sand, compacted hot

106

35 5

12 . 4% of dust

114

16 7" " " . i

116

29 1

20 6' " " .

120 2

26 6

24 2 ' " "

127

30 4' " " .

130

21 2

36.0' " "

132 5

41.7' " "

133.1

19 7

50 0' " "

114 6

24 7

THE MINERAL AGGREGATE.

When the densest of these sand and dust mixtures is added to
a gravel with voids of 35.1 per cent in amount sufficient to fill the
voids in the latter they are further reduced to about 12.1 per cent,
and the aggregate weighs 144.8 pounds per cubic foot as compared
to 164.1 pounds for solid quartz.

WEIGHT PER CUBIC FOOT AND VOIDS IN CRUSHED FLINT,
GRADED LIKE THE AVERAGE SAND IN SEVERAL CITIES,
COMPARED WITH THE LOCAL SANDS OF THESE CITIES
OF THE SAME GRADING, WITH NO 200-MESH MATERIAL,
WITH 13 PER CENT OF 200-MESH FLINT AND 13 PER CENT
OF 200-MESH DUST. SPECIFIC GRAVITY OF FLINT = 2. 65.

Without 200-
Mesh Material.

With 200 Flint.

With 200 Dust.

Wt. per
Cu. Ft.

Voids.

Wt. per
Cu. Ft.

Voids.

Wt. per
Cu. Ft.

Voids.

New York 1898:
Local

109.6
105.5

109.1
103.4

111.9
103.6

104.5
99.6

110.4
103.0

113.3
105.3

111.1
107.6

110.6
104.3

106.1
109.0

109.6
100.1

107.8
104.5

34.1
38.1

34.6
37.4

31.7
37.1

36.0
39.5

32.5
37.6

30.9
36.2

31.6
34.8

32.6
36.8

36.5
34.0

33.9
39.4

35.2
36.7

115.6
109.0

113.9
104.9

116.1
108.3

110.9
107.0

115.0
106.2

118.6
106.6

117.1
112.6

115.6
109.4

114.5
113.5

115.9
103.9

111.3
106.5

30.5
34.0

32.3
36.4

29.1
34.4

32.1
35.2

29.9
35.7

27.6
34.2

28.2
31.8

29.6
33.7

31.4
31.8

30.1
37.1

33.1
35.5

118.9
110.9

120.4
107.4

122.2
111.8

115.7
112.2

122.4
113.5

124.5
113.5

123.5
116.0

121.0
112.4

119.0
118.1

120.5
108.6

119.4
113.9

26.5
32.9

27.9
34.9

25.4
32.3

29.1
32.0

25.3
31.0

24.0
31.2

24.1
29.7

26.3
31.9

28.1
26.5

27.3
34.2

28.2
31.0

Flint

Chicago 1898 :
Local . ...

Flint

St. Louis 1899:
Local

Flint

Louisville 1899 :
Local

Flint

Kansas City 1898:
Local

Flint

Omaha 1899:
Local

Flint

Trenton 1898:
Local

Flint

Paterson 1899:
Local

Flint

Washington 1899 :
Local

Flint

Buffalo 1899:
Local. .

Flint

Philadelphia 1899:
Local

Flint .

THE MODERN ASPHALT PAVEMENT.

Sharp as Compared with Rounded Sand. If particles of the
crushed quartz of sizes corresponding to those found in natural
sand are combined in the proportions found in the latter, and the
voids in each determined, the results are a striking illustration
of the difference in the degree to which compaction can be carried
with sand made up of sharp and rounded particles. (See preceding
table, page 83)

As in the case of the single-sized particles the sharp sands do not
compact as well as those with rounded particles, and as the greatest
possible compaction is desirable, it seems that a rounded sand is
more suitable for the construction of asphalt pavements than a
sharp one. Experience has shown that this is the case. The
particles should, however, be rounded and not round, as in the
latter case they would move too easily on one another and give
the pavement a tendency to displacement under traffic. It will
not do, however, to draw too general conclusions from a determina-
tion of voids in a sand alone. Small voids are desirable, but may
at the same tune occur in sands which are unsuitable for use in a
surface mixture. For example, sands too coarse to permit of
being employed may have a smaller volume of voids than a finer
or more suitable sand.

Passing Mesh.

Total.

Wt per
Cu. Ft.

Voids.

200

100

25
30

Trenton

11
22

14
14

= 100%
= 100%

111.1
107.8

31.6
35.2

Philadelphia.

Here the coarse sand has the smaller volume of voids, which is
quite often the case, but the size of the voids is too large and the
sand is consequently undesirable. It is therefore necessary to
consider the grading of a sand as well as its voids in judging it.

Grading of Sands. The proper grading for an asphalt mix-
ture is seldom found in a single sand, but it can be arranged by
mixing two or more containing particles of different sizes. The
character of the sands which have been used in mixtures in various
cities is illustrated by the following examples.

THE MINERAL AGGREGATE.
SAND GRADING VARIOUS CITIES.

Cities.

Passing Mesh.

Total.

200

100

Chicago fine, 1896

10
2
2
32

2
2
2
6

14
2
17
31
2
1

68
15
1
33
1
25
19
14
13
1
26
4
40
39
9
6

15
17
4
13
2
29
19
26
14
1
14
22
30
21
36
10

5
52
53
18
36
36
41
49
31
48
38
28
10
8
49
41

2
9
25
3
32
4
12
6
20
46
6
19
1
1
3
19

2'
10
1
17
3
3
2
10
3
2
10
1

2
3

9
1
2
1
4
1

l6'
1

2
3
2

= 100%
= 100%

=100%

= 100%
= 100%
= 100%
= 100%
= 100%

= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%

' ' medium

Louisville river

< bar.

Milwaukee coarse beach

< < White Fish Bay

Omaha single sand, 1899

Shelby single sand 1899

Boston single sand, 1899

St Louis coarse, 1897 . . .

< fine, 1897

' river, coarse .*....

__ foe

Buffalo bank fine

' lake fine

' " coarse .

The preceding data show the very great variations in the charac-
ter of different sands, not only as to the size and shape of the
grains, which are of the utmost importance in preparing a surface
mixture, but also in the character of the surface of the grains. It
does not seem too much to say, therefore, that the selection of a
suitable sand or the combination of two or more in a proper way is
as important a detail in producing a satisfactory asphalt pavement
as the regulating of any of the other constituents. The difficulties
to be met with are numerous and demand experience in meeting
them.

Stone. Stone is sometimes used as a part of the mineral
aggregate, and this use has grown of late years. When it forms the
chief portion of the wearing surface the latter is known as an
asphaltic concrete. This material will be considered in a later
part of this volume. 1

SUMMARY.

In this chapter the very varied character of the sand which is
the chief constituent of the mineral aggregate of an asphalt surface

1 Page 375.

86 THE MODERN ASPHALT PAVEMENT.

mixture has been shown, and that skill is necessary in selecting this
material for the preparation of a satisfactory mixture. Sands differ
according to their origin and the source from which they are
obtained. They differ as regards the size of the grains of which
they are made up, the relative proportion of different sized grains
which are present, the shape of the grains, the character of the
surface and of the material of which they are composed, and
the proportion of the voids or vacant spaces between the grains
when compacted. They also vary in their volume weight, a
matter of importance where materials are mixed by weight and
not by volume.

Altogether there seems to be nothing more important for the
construction of a satisfactory asphalt surface mixture than a
thorough understanding of the peculiarities of the various sands
and of their adaptability to the purpose for which they are used.

CHAPTER IV.
FILLER, OR DUST.

A FILLER, or dust, is made a part of the mineral aggregate of
asphalt surfaces for the purpose of rendering the surface more
dense, so that it will be less acted upon by water, and less liable to
interior displacement or movement. Its presence in a surface
mixture may be looked at in much the same way as that of the
finer clay particles in a soil. In fact soil physics, as treated by
the agricultural physicists, is in many directions instructive as
applied to surface mixtures, which are aggregates of small particles
like soils and contain more or less bitumen as the soils do more or
less water.

In regard to the different parts played by fine and coarse
particles in a soil, Whitney remarks : l

"In a symmetrical arrangement of the grains in a soil con-
taining 47.64 per cent 2 by volume of empty space, each grain will
touch the surface of six adjacent grains. There is a certain
amount of surface attraction between these particles.

" If the grains are large they still only touch at six points, and
the weight of the grains is sufficient to overcome this slight sur-
face attraction. A lump of wet sand will fall apart as it dries,
for it is bound together by the contracting power of the film of
water which surrounds it, and when this is removed by evapora-
tion the weight of the grains is sufficient to overcome the surface
attraction of the relatively large and heavy particles and they
fall apart.

1 Bulletin No. 4, U. S. Weather Bureau, 1892, 27.

Whitney is, of course, wrong in assuming this volume; the voids, a-*
has been shown on page 74, might be 25.95 per cent.

88 THE MODERN ASPHALT PAVEMENT.

"If the grains are very small, like grains of clay, the surface
attraction of the grains is sufficient to bind the mass into a com-
pact lump when dry; for while there are still only six points of
contact for any one grain, there are many other grains and so many
more points of contact in a given weight of material. If the size
of the grains was still further reduced to molecular proportions
the mass would assume the hardness and rigidity of a single grain
of sand or clay."

These facts apply as well to asphalt-surface mixtures as to
soils, and explain the parts which a filler plays.

This, however, has been little understood heretofore in the
paving industry.

Dust in the Earlier Days. In the earlier days of the industry,
and even as late as 1885, the Washington specifications for asphalt
pavements read that the mixture shall contain 12 to 15 per cent
of carbonate of lime and that "the powdered carbonate of lime
will be of such a degree of fineness that 16 per cent by weight
of the entire mixture for the pavement shall be an impalpable
powder of limestone and the whole of it shall pass a No. 26 screen.
The sand will be of such size that none of it will pass a No. 80
screen and the whole shall pass a No. 20 screen." As a matter
of fact, very little real dust, 200 mesh, was put in the mixture
and it was a question even as late as 1893 whether dust con-
tributed in any way to improve it.

That it is of the greatest value, especially in surface exposed
to heavy traffic, is now known. The difference in the penetra-
tion, ductility, and resistance to stress of the same bitumen with
and without filler can be readily shown. 1 Filler, therefore, enables
us to use a softer cement than otherwise would be the case and
thus make an asphalt surface less liable to mark in hot, less brittle
in cold weather, and far less liable to internal displacement. In
the earlier asphalt pavements where Trinidad asphalt was the
only cementing material in use this contained in itself so much
fine inorganic matter which was a natural filler that it was a great
help to the surface before the necessity for the presence of a high
percentage of dust was understood.

1 See page 373

FILLER, OR DUST. 89

The use of a filler is well illustrated in the laying of coal-tar
walks in England, where very soft coal-tar is mixed with all the
slaked lime it will hold, often more than 50 per cent by weight,
and this mixture used as a cement with sand. The filler makes
it possible to use a tar so soft that it will not crack in winter, while
preventing its marking excessively in summer.

Varieties of Filler. Numerous kinds of mineral matter, ground
to a more or less fine powder, have been used from time ta time
in asphalt mixtures as a filler. These include

Limestone, Natural hydraulic cement,

Hydraulic limestone, Portland hydraulic cement,

Trap rock, Clay,

Volcanic, Ground shale,

Marl, Dust-collector dust,

Silica, Ground waste, lime from beet-

Caustic or slaked lime, sugar factories.

Ground Limestone has been used far more than any other and
was the original material employed by De Smedt. There is prob-
ably nothing better than this, unless it be Portland cement, for
heavy-traffic streets. It is a desirable material, as asphalt cement
adheres to it firmly and does so by being absorbed by it to a cer-
tain extent.

Ground hydraulic limestone has also been used where it could
be conveniently and cheaply obtained from the cement manu-
facturers. It is, no doubt, as suitable for its purpose as the simple
carbonate.

Ground Shale. In the manufacture of shale bricks the shale is
first ground to a powder which is often extremely fine and in con-
sequence suitable for use as a filler. Such a shale dust is available
in the State of Washington, 91 per cent passing a 200-mesh screen
and 79 per cent remaining suspended in water for fifteen seconds.
Ground Clay or loam free from organic matter could also be
used in a similar way. A large part of the natural filler in Trinidad
asphalt is clay, and on this account it was thought that clay might
eventually prove the most desirable filler, as owing to its peculiar
surface and porosity it will absorb bitumen much more satisfac-
torily than any of the ground rock fillers, but it has been found to
have such a small volume weight and to be so light and fluffy that

90 THE MODERN ASPHALT PAVEMENT.

a large part of the clay is blown away in an open mixer and it can
only be used successfully in a tightly-closed one, which is rarely
available. In this connection the studies of Dr. A. S. Cushman,
of the Office of Public Roads, U. S. Department of Agriculture,
on 'The Nature of Clay" and on "The Adsorption of Solids by
Rock Powders" are of great interest.

Ground Waste Lime from Beet-sugar Factories. In the process
of defecating the diffusion liquors obtained in the extraction of
sugar from beets large quantities of caustic lime in the form of
cream of lime are used, which is subsequently removed from the
sugar solution by nitration. This when dried and ground has been
suggested for use as a filler and employed to a small extent in
California and Michigan. As far as fineness is concerned it is a
satisfactory material, but it contains many impurities such as
organic matter in the form of sugar and organic acids combined
with lime. Analysis shows that it has the following composition:

COMPOSITION OF DRIED AND POWDERED BEET-SUGAR
FILTER-PRESS CAKE.

Calcium carbonate 78 .0%

Free lime 1.0

Lime combined with organic matter 5.0

Magnesia carbonate 2.0

Alkalies 2

Iron and alumina 2.6

Sulphuric, phosphoric, and oxalic acids 2.4

Sugar 5.0

Organic not sugar 2.1

98.3

It is to a certain extent an open question whether the organic
matter will prove deleterious to the surface mixture and the free
lime likewise. Experience alone can prove the availability of
this material as a filler.

Ground Marl has served of late years as a filler in those cities
near the marl-beds of Ohio and Michigan. It gave fairly satis-
factory results, but its disadvantage lies in its low volume weight,
in consequence of which it is readily blown away on mixing it with
sand, and its use has been discontinued.

Ground Silica. Ground sand and trap-rock have been largely
used in the work in New York. It is questionable if it is desirable

FILLER, OR DUST. , 91

as a substitute for limestone, as asphalt does not adhere to it as
well and it cannot be ground as fine. It is a filler, however, and
successful results have been obtained with it.

As between ground limestone and silica or silicate dusts, experi-
ments of Mr. F. P. Smith, formerly of the Alcatraz Asphalt Co.,
have shown that the former enables a mixture made with it to
resist water action better than the silica filler, and this can be
readily understood for the reason that has been given, namely,
that bitumen will adhere to the former much more firmly than to
the latter by being partly absorbed by it.

Caustic and Slaked Lime. These fillers have only been used
experimentally. They are largely employed in coal-tar work.
No peculiarities have been noticed in the small amount of work
done with them, but in the laboratory cylinders of surface mixture
containing caustic lime expanded badly on immersion in water.
It would probably not be desirable to experiment further with
their use.

Natural Hydraulic Cement. This material began to be used
as a filler in cases where limestone dust was not available. How-
ever, of late years its use has been abandoned, as it has been observed
that surfaces laid with this material as a filler have cracked more
than where limestone was the ground material. It seems to
possess the property, perhaps owing to the presence of free lime,
of hardening the asphalt cement very rapidly. If it is necessary to
use such a filler the cement should be at least 20 points softer than
would be the case with other materials.

Portland Cement. This is a material which, for some reason
not yet satisfactorily explained, gives the best results as a filler in
asphalt surfaces, especially on streets of heavy traffic or where
the surface is subject to the action of water. Its desirability
may be due to its capacity for adsorbing a thick film of bitu-
men, but it cannot with certainty be attributed to its hydraulic
properties.

A cylinder of open Trinidad asphalt surface mixture, made up in
Washington, D. C., in 1894, half of which contained limestone
dust as a filler and the other Portland cement, showed the most
striking contrast in its appearance after nearly six years' immersion

92 THE MODERN ASPHALT PAVEMENT.

in water, the portion containing Portland cement being still hard
and firm, while the ordinary limestone mixture was much more
strongly acted upon.

The slight extra cost of Portland cement is more than made
up by the improvement in the character of the surface, where
especially trying conditions are to be met, and its use is highly to
be commended.

Fine Grinding of the Filler. The material which is of value in a
filler is the impalpable dust, much finer than the particles merely
passing a 200-mesh sieve. Sandy particles of dust of about the
200-mesh size are probably of somewhat greater value than the
200-mesh rounded particles of an ordinary sand, which are at
times distinctly disadvantageous in a mixture, as they are
sharper. Larger particles do not differ from sand grains of the
same size.

It is important, therefore, in securing a filler that it should
contain as much real dust as possible. If there is only 45 per
cent of this material, twice as much must be used as if it contained
90 per cent. In the former case the sand must be heated to a
much higher temperature to take care of so much cold material,
while in the other, as a matter of economy, the smaller bulk to be
handled to accomplish the same object is an important con-
sideration.

As it is difficult to find a desirable filler on the market, dust
should be ground by paving companies themselves. It can be
turned out with a tube-mill 85 to 95 per cent fine. There is no
question but that the production and use of such dust will pay
if for no other reason than to do away with the excessive cooling
of the mixture caused by the addition of the large quantities of
cold, coarse material to the sand which are necessary to obtain
a sufficient amount of true filler.

More Refined Methods of Examining Filler. Up to the present
point fillers have chiefly been spoken of, as to their fineness, accord-
ing to the amount which will pass a 200-mesh sieve, the finest
wire sieve that is made. As has been said, the material passing
this sieve may be much of it sand smaller than .10 mm. in diam-
eter, and very little of it may be true dust or filler of a diameter

FILLER, OR DUST. 93

smaller than .025. The difference in character of the two sizes
is readily seen on inspection.

In judging the value of a filler it is desirable to determine the rela-
tive amount of these materials, coarse material, and the impalpable
dust. As there are no finer sieves than the 200-mesh, this can only
be done by elutriation, or washing with water, the coarser grains
settling out rapidly and the finer more slowly. The manner of
doing this is as follows:

Five grams of the dust to be examined are placed in a beaker
about 120 mm. high, holding about 600 c.c. The beaker is nearly
filled with distilled water, at a temperature of exactly 68 F.,
and agitated with an air-blast until the dust and water are thor-
oughly mixed, taking care not to produce cyclonic currents hi
the latter. On stopping the blast the liquid is allowed to stand
exactly 15 seconds and the water above the sediment immediately
decanted without pouring off any of the latter. ^This washing
is repeated three times. The sediment is then washed out into
a dish, dried, and weighed. The loss hi weight represents that
portion which may be considered as dust free from sand. The
washing must be done with distilled water and at a definite tem-
perature.

This method can also be used with hydraulic cements or mate-
rials acted upon by water, since the finer portion is the only part
acted upon, while the coarser part, which is recovered and weighed,
is not acted upon at all.

The differentiation of the particles not subsiding in 15 seconds
can be carried further, if desired, by reagitating the decanted
liquid and allowing the sedimentation to go on for 1 minute, 30
minutes, 1 hour, and so on. For ordinary purposes this is unneces-
sary. 1 The size of the particles obtained by elutriation can be
measured in the same way as that of the particles passed by
the finer sieves, as described by Hazen. The size of these
particles among ordinary fillers will be found in the following
table:

1 For further details, see Hazen, 24th Annual Report Massachusetts State
Board of Health, 1892, 541.

THE MODERN ASPHALT PAVEMENT.

VOLUME WEIGHT OF DUST.

Test number

75803

75804

75805

75806

71076

75791

Dust

Lime-

Trap

Port

Clav

Marl

Vol-

stone

rock

cement

canic

84 0%

81%

74%

93%

92%

100%

" 100- "

14.0

80- "

2.0

" 50- "

Elutriation test not set-

tled in 15 seconds

71.3%

70.3%

56.7%

87.8%

80.3%

98.2%

Pounds per cubic foot. . . .

113.7

112.3

123.5

78.0

63.4

A number of dusts from various parts of the country have been
differentiated and the results are as follows:

SIZE OF PARTICLES IN VARIOUS FILLERS.

Test No.

Character.

Per Cent
Passing
200.

Per Cent
on 200.

30915

Limestone.

30963

Silica.

30267

Limestone.

30578

30715

30716

30766

Silica.

30606

Marl.

In these fillers, as supplied for use, there was present
the percentages of particles passing a 200-mesh sieve shown in
the preceding table.

This 200-mesh material consists of the following sized par-
ticles:

FILLER, OR DUST.

Average

Test No.

Time of

Size of

Less than

30015

30963

30276

30578

30715

30716

30766

30606

Difference and

loss

001 mm.

4 4

3 8

2 7

11 9

3 2

2.4

16 hours

.0025 '

2.8

4.2

2 "

.0075 '

5.0

2.9

4.9

5.4

3.1

3.2

3.0

7.7

30 minutes. . .

.025

51.3

42.4

55.1

67.4

32.5

23.9

23.3

66.5

1 minute

.050

17.7

9.3

15.0

12.9

24.8

20.1

25.6

13.0

15 seconds. . . .

.080

18.8

40.2

21.2

13.9

36.9

40.9

44.9

10.4

100.0

Actual dust in

original ma-
terial small-

er than

.050 "

78.0

55.6

71.7

80.1

30.3

39.0

36.9

80.5

As showing the variation in the amount of material which
is not ground even fine enough to pass a 200-mesh sieve the first
data given may be examined. The coarse material varies from
4 per cent in high-grade fillers to 52 per cent in an inferior article.
Separating out and rejecting this coarse material as of no value
greater than that of sand the finer grams were separated by elu-
triation into the particles of the sizes given.

It seems fair to consider that particles smaller than .050 mm.
in average diameter are the only portions of a filler to be con-
sidered as true dust, and it will be seen that of the entire mate-
rial in several instances only 30-40 per cent is dust and the remain-
der sand.

A good filler should contain at least 60 per cent of its weight
of actual dust and preferably over 70 per cent.

An examination of a filler or even a mixture in this way is
very serviceable in revealing the actual amount of fine dust which
either may conHH. In a mixture examined in New York the
actual amount of dust of different sizes is shown in the following
analysis:

96 THE MODERN ASPHALT PAVEMENT.

TEST NO. 30954.
Bitumen 11 .0%

Settling in,
Time.

Size of
Particles
Less than

Passing 200. ..

Difference,

or loss. .

8%....

2 hours. . .

.0075mm.

30 minutes.

.025 "

2.5

1 1

1 minute..

.050 "

4.2

15 seconds.

.080 "

10.2

18.0

100. ..

80. ..

50. .

40. ..

30. ..

2.0

20. .

10..

2.0

200 Mesh

on 100

Per Cent

Basis.

4-2%

1.6
13.8
23.8
56.6

100.0

Sand 77.6%

Dust 5.4

Trinidad A. C 17.0

100.0

The dust was the finest-ground limestone of the composition
given in the preceding table. It contained, as used, 81.2 per
cent of particles not subsiding in 15 seconds. From the amount
of dust in use it is readily calculated that no more than 4.3 per
cent of material which acts as a filler would be expected in the
mixture. 7.9 per cent is actually found, the excess over the cal-
culated 4.3 per cent being due to the fine material derived from
that in the Trinidad asphalt. The small percentage of real dust
in some of our mixtures is therefore striking.

SUMMARY.

The character of the filler or finely ground inorganic matter
which enters into the composition of the mineral aggregate of
an asphalt surface has been shown in the preceding chapter to
be a matter of very considerable importance. Impalpably fine
mineral matter of various kinds can be satisfactorily used, but
it should be as fine as possible, and for construction of an asphalt

FILLER, OR DUST. 97

pavement on streets of heavy travel Portland cement should alone
be used as a source of supply.

The intelligent use of filler in an asphalt surface mixture demands
further careful consideration from a physical point of view, and
the investigations which have been carried on in regard to rock
powders in the Office of Public Roads, U. S. Department of
Agriculture, have thrown much light upon this subject.

CHAPTER V,

THE NATURE OF THE HYDROCARBONS WHICH CONSTITUTE
THE NATIVE BITUMENS.

ASPHALT cement is the distinctive feature of an asphalt pave-
ment. It serves to bind the mineral aggregate together and
enables it to carry traffic without displacement.

It consists of some native asphalt or other hard native bitu-
men, fluxed with some soft bitumen in the form of dense petro-
leum oil or maltha, or of some hard residue from the distillation
or treatment of asphaltic petroleum, softened to a proper con-
sistency in the same way.

Asphalt cement is, therefore, native bitumen and its intelli-
gent consideration necessitates a knowledge of and the differentia-
tion of the various native bitumens of which it may be made up.

The hydrocarbons, or compounds of hydrogen and carbon,
which when mixed in varying proportions constitute the sub-
stances which are known as bitumen, belong to different series,
so called, which are characterized by the relative proportion of
hydrogen and carbon atoms which they contain and by their
structure or the relation of the carbon atoms to one another in
space. A short explanation in regard to the structure of the
various hydrocarbons of these series is necessary for an intelli-
gent understanding of their properties as affecting the character
of the bitumen in use in asphalt* pavements.

Chain Hydrocarbons. The carbon atom is characterized
chemically as being quadrivalent; that is to say, it possesses four
affinities, bonds, or links by means of which it may be said to
combine with atoms of other elements. The hydrogen atom
has but one bond and is univalent.

NATURE OF THE HYDROCARBONS. 99

If a carbon atom combines with all the hydrogen atoms that
it is capable of, its four bonds must each be linked with a hydrogen
atom, and the resulting molecule will consist of one atom of car-
bon and four of hydrogen, which can be represented by the symbol
CH 4 , in which C stands for one carbon atom and H 4 for four hydro-
gen atoms. In this substance or compound, which is known
as methane, or marsh-gas, we have the carbon atom saturated
as to its affinities, or bonds, with hydrogen atoms. It cannot
combine with any other atom except by replacing with it one
or more of the hydrogen atoms. It is, therefore, known as a
saturated hydrocarbon.

If two, three, or more atoms of carbon are combined in the
same way to form a molecule having 2 or 3, etc., in its com-
position, these carbon atoms are themselves linked or bonded
together by one or more of the bonds of each atom, so that we
may have:

One carbon atom with its four affinities or bonds which may
be represented thus:

Two carbon atoms joined by one affinity of each thus?

44-

Or by two affinities of each thus:

In the case of two carbon atoms joined by one affinity each,
there are six bonds remaining to unite with hydrogen. The result-
ing compound with hydrogen in this case is represented by the
symbol or formula C2H 6 . It is a saturated hydrocarbon in the
same way that CH 4 , with one atom of carbon, is, because, while
two bonds out of the eight of the two carbon atoms are necessarily

100 THE MODERN ASPHALT PAVEMENT.

joined in linking the carbon atoms together, all the remaining
available affinities are satisfied by hydrogen. In the same way
with three atoms of carbon in the molecule we have the carbon
atoms linked to each other, so that eight affinities out of
twelve remain to be saturated by hydrogen thus:

H H H
H C C C H

One atom is, of course, linked to two others, and so two affini-
ties of this atom are not available for combining with hydrogen.
This saturated hydrocarbon is represented by the symbol CsHg.

As the carbon atoms increase in number there is found to be
a regular increase in those of hydrogen, so that the compounds of
this saturated nature become what is called a homologous series,
differing by one carbon and two hydrogen atoms from the
following and preceding.

CH4, C2Hg, CaHg, CnH.2n+2>

If in any of these simple chain hydrocarbons, which owing
to the simplicity of their constitution are known as normal hydro-
carbons, one of the hydrogen atoms is supposed to be removed, a
group is left with one free affinity, or if two hydrogens or more
are removed, with two or more affinities. These imaginary groups
of atoms with different affinities are known as radicals. They
can combine with other similar radicals or with other elements,
such as the halogens, or with acid radicals. Thus we may have

Saturated. Methyl. Methylene. Methenyl.

If different hydrocarbon radicals are substituted for hydrogen
in other hydrocarbons new hydrocarbons result. In this way
hydrocarbons are produced which have the same composition
or number of carbon and hydrogen atoms as in the normal hydro-
carbon, but a different structure. For pentane, therefore, C 5 H 12
there may be three forms:

NATURE OF THE HYDROCARBONS. 101

CH 3 CH2 CH2 CH2 CH 3
CH 3 CH CHa H 3
CH 3

CH 3
H 3 C C CH 3

CH 3

Here the straight chain is converted into one with one or more
radicals known as side-chains.

These hydrocarbons are denominated normal pentane, iso-
pentane, and tetramethylmethane, the latter being methane in
which the four hydrogen atoms are substituted by methyl groups.
They are also known as isomers, since they contain the same num-
ber of carbon and hydrogen atoms; that is to say, have the same
percentage composition but a different structure and different
physical properties.

With similar hydrocarbons in which the carbon atoms are
greater hi number the possible variations in structural arrange-
ment are much more numerous, and it can be readily seen that
the number of different paraffin e hydrocarbons is enormous.

These compounds of carbon and hydrogen illustrate what
is meant by a series of hydrocarbons, which is, in this case, a satu-
rated series known as the paraffine, limit, or chain series, since
the carbon atoms are represented as linked in the form of a chain.
It is the series which makes up the greater part of ordinary Penn-
sylvania and Ohio petroleum and the residuum made from these
oils.

In this series, the carbon being combined with as much hydro-
gen as possible, there is the largest percentage of hydrogen and
the smallest percentage of carbon found in any hydrocarbons
of a given number of carbon atoms. For marsh-gas, CH 4 , it
is 75 per cent carbon and 25 per cent hydrogen, a proportion
gradually diminishing as the number of carbon atoms increases.
For example, for C 30 H 6 2 it is carbon 85.31, hydrogen 14.69.

102 THE' MODERN ASPHALT PAVEMENT.

Unsaturated Hydrocarbons. When fhe carbon atoms in a
hydrocarbon do not combine with all the hydrogen atoms they
might, the remaining affinities are satisfied in joining the carbon
atoms together, in addition to the single bond found in the satu-
rated series. The linking of the carbon atoms is then doubled
and the hydrocarbons may be represented thus:

H H H H

C=C C=C C H

Owing to the double bond, two affinities which in the unsatu-
rated series were combined with hydrogen, are now linked with
each other and a new series is determined in which the hydrogen
atoms number always twice the carbon atoms. The affinities of
the carbon are not entirely satisfied with hydrogen, and the hydro-
carbons are known as unsaturated hydrocarbons. As the rela-
tion of carbon to hydrogen is constant the percentage composition
of all the hydrocarbons of the C n H 2w series is carbon 85.71, hydro-
gen 14.29, the amount of hydrogen being always less than in any
of the hydrocarbons of the saturated series containing the same
number of carbon atoms.

In this series, which is known as the Olefine Hydrocarbons,
but two of the carbon affinities are joined by a double bond. More
of these affinities may be joined in this way, resulting in other
series represented by the general formula C n H 2n _2, C n H2 n -4,
C n H 2n -6, etc., in which the per cent of hydrogen is still less.

The hydrocarbons of these series, it is plain, are even more
unsaturated.

Hydrocarbons in general are divided, therefore, into those
which are saturated and those which are unsaturated, the former
being stable and the latter reactive and very susceptible to change,
combining with or being converted into other hydrocarbons by
the action of sulphuric acid and other reagents. The saturated
can be separated from the unsaturated hydrocarbons by strong
sulphuric acid, and this will be found to be a very important means
of differentiating the oils and the solid bitumens among them-

-
NATURE OF THE HYDROCARBONS. 103

selves, by determining the relative proportions of these two classes
of hydrocarbons which they contain.

Cyclic Hydrocarbons. In the preceding hydrocarbons the
carbon atoms have been imagined as being linked in the form
of a chain of more or less simplicity. It can readily be imagined
that the normal chain can be bent into the form of a circle so
that the carbon atoms at the ends may be united with each other
by one of each of their three affinities. In this way a ring is formed,
each carbon atom of which has only two affinities unsaturated,
but which possesses no double bond when these affinities are all
satisfied with hydrogen, so that although its general formula is
C n H 2n , the same as that of the unsaturated olefines, they are
saturated hydrocarbons.

Owing to reasons which it is unnecessary to go into in this
place the carbon atoms in such a ring do not exceed seven hi num-
ber, as above that they would be quite unstable and could not
exist. The most- stable rings are those of five and six atoms, and
hydrocarbons with this number are the foundation or source of
many of the solid native bitumens. Their structure may be
represented as follows:

CH2 CH-K CH 2 CH 2 -H 2

>CH 2 or C 5 H 10 | or C 6 H 12

CH 2 CH 2 / CH 2 CH 2 CH 2

Pentamethylene. Hexamethylene.

The carbon-hydrogen groups of which they are made up are
the groups or radicals known as methylene.

For this reason the hydrocarbons are known as a class as the
polymethylenes, pentamethylene being the hydrocarbon of five
groups, hexamethylene the one of six. The generic name of
naphthenes is also applied to the series, having been used to desig-
nate the polymethylenes occurring in Russian petroleum before
their structure was elucidated. They are distinguished by the
fact that, although not as stable as the paraffine hydrocarbons,
they still possess the stability of saturated compounds and are
unacted upon by strong sulphuric acid.

In these polymethylenes, as in the normal chain hydrocarbons,
one or more of the hydrogen atoms can be substituted by radicals

104 THE MODERN ASPHALT PAVEMENT.

like methyl. We have, for instance, methylpentamethylene, in
which one of the hydrogen atoms of one of the methylene groups
in pentamethylene is substituted by CH 3 the methyl group:

<j\

>CH CH 3
CH 2 -CH/

It will be noted that this hydrocarbon has the same formula
as hexamethylene and differs from it only in structure. They
are isomers.

More complicated chains can exist, as where the radical propyl
CaHr or others replace the methyl radical and the possibilities in
number and isomerism is again immense.

The more complicated single-ring polymethylenes with side-
chains are more reactive than the simple naphthenes.

Unsaturated Cyclic Hydrocarbons. Corresponding to these
so-called cyclic saturated hydrocarbons, hi which the carbon
atoms are only united with each other by one bond, unsaturated
cyclic hydrocarbons exist in which double bonds occur. The
most familiar hydrocarbon of this type is benzol, derived from
coal-tar, which has the folio whig structure:

/N C-H

II '

C H

This forms a new series known as the benzol or aromatic series,
the general formula for which is C n H 2n -6-

These hydrocarbons occur in a greater or less degree in all
petroleums, at least among the more volatile portions, and are
particularly prominent in California and Russian petroleum.

NATURE OF THE HYDROCARBONS. 105

Where the number of double bonds is fewer than in the benzol
ring other series of hydrocarbons are formed, known as the hydro-
aromatic series, the hydrocarbons of which, in then* constitution,
are between the saturated polymethylenes and the aromatic com-
pounds. The terpenes are members of this series, but they are
not found hi the solid native bitumens used in pavements. Hexa-
hydrobenzol is the same thing as hexamethylene and is a saturated
hydrocarbon. Tetrahydrobenzol is an unsaturated hydrocarbon
corresponding in the cyclic series to the olefines of the chain hydro-
carbons.

In all of these aromatic and hydrated aromatic hydrocarbons,
as well as in the saturated polymethylenes, any or all of the hydro-
gen atoms may be substituted by paramne or olefine radicals,
thus making it possible to form a vast number of new hydro-
carbons containing side-chains, of which toluol, or methyl benzol,
is a type in the aromatic series, as was methyl pentamethylene
in the polymethylene series.

C CH3 CH2 CIi2\

II I \C-H-CHs

H C C H CHjj CH/

Methy Ipen tame thy lene .

Methylbenaol.

Dicyclic Hydrocarbons. The cyclic hydrocarbons thus far
considered have consisted of but one ring. Dicyclic and poly-
cyclic hydrocarbons are also known to exist in which two or more
rings may be united by having carbon atoms in common, as in
the case of naphthalene, CioH 8 , the result of the condensation
of two benzol rings:

106 THE MODERN ASPHALT PAVEMENT.

H H

-H

or of three rings, as in anthracene;
H H H

i i A

/\ /\ /\

H C C C C H

I II II II orC 14 H 10

, H C C C C H

\/ v V

C C C

I I I

H H H

The latter substance may also be considered as consisting of
two benzol rings united by two methenyl radicals.

The benzol rings may also be united by free affinities or by
one or more methylene radicals without common carbon atoms,
as in diphenyl, CoH. 5 CeHs, as dibenzyl, CoH. 5 CH 2 CH 2 CeHs,
stilbene, C 6 H 5 CH=:CHC 6 H5, or as tolane, C 6 H 5 C=C C 6 H 5 .

These hydrocarbons are mentioned, not from their immediate
interest in connection with the bitumens, as they only occur in
coal-tar, but as showing the infinite variation in structure, which
is possible.

Polycyclic Polymethylenes. In the polymethylene series
bicyclic and pqlycyclic hydrocarbons also exist, in which two or
more rings have common carbon atoms, but instead of two carbon
atoms being common to both rings, as in naphthalene, three such
are found in the bicyclic polymethylenes, forming what is known
as a bridge structure. This is illustrated by two hydrocarbons

NATURE OF THE HYDROCARBONS. 107

prepared synthetically by Rabe and Weilinger 1 and having the
following structure :

CH 2 C (CH 3 ) CH 2 CH 2 OH

I I -CH 3

CH 2 CH 2 CH 2 CH 2 CH.CH 3 CH.CH

I I -CH 3

CH 2 -CH CH 2 CH 2 -CH CH 2

Methyl-bicyclo-ndhane. Isopropyl-methyl-bicyclo-nonane.

The latter hydrocarbon has probably been found by Coates
in Louisiana and by Mabery in Santa Barbara, Cal., petroleum, 2
as can be seen from the close correspondence in the physical and
other characteristics of the synthetic and native hydrocarbons.

Synthetic. Louisiana. California,

Specific gravity 0.8646 0.8629 0.8621

Refractive index 1 .460 1 .4692 1 .4687

Molecular refraction:

Found for C 13 H 24 57.677 57.93 58.05

Calculated 57.737

Boiling-point, 755 mm 232 235

28 mm 132 60 mm. 150

Ultimate composition:

Carbon 86.48-86.56 86.58 86.68

Hydrogen 13.28-13.15 13.42 13.62

Calculated:

Carbon 86.67

Hydrogen 13 .33

The hydrocarbons of the heavier asphaltic petroleums are,
therefore, probably bicyclic or bridge compounds in those of lower
boiling point, and polycyclic in the higher ones.

In the lightest oils from Trinidad asphalt a hydrocarbon has
been isolated which has the same formula as the preceding ones,
but differs from them in some of its characteristics, as can be seen
from the data on the next page.

This hydrocarbon, while in some respects like those prepared
by synthesis and from the California and Louisiana petroleums, is

J Berichte der deutschen chemischen Gesellschaft, 1904, 37, 1667-1675.
2 Coates, J. Am. Chem. Soc., 1906, 384. Mabery, Proc. Am. Acad., 1904,
40, 340.

108 THE MODERN ASPHALT PAVEMENT.

HYDROCARBON FROM TRINIDAD ASPHALT.

Specific gravity 8690

Refractive index 1 .4721

Molecular refraction, found 57 . 98

' l calculated 57 . 737

Boiling-point, 30 mm 170-180

Ultimate composition:

Carbon 86 .85

Hydrogen 13 .34

sharply differentiated from them by its much higher boiling point,
and cannot, consequently, be of the same structure. The as-
phalt hydrocarbons differ, therefore, from those found in the
petroleums, but apparently are of a similar nature, and the low-
est ones at least of the series, are probably substituted bicyclic or
bridge polymethylenes, the higher being polycyclic. The idea
previously advanced by Markownikoff, Mabery and the author
that they were probably two rings united by polymethylene groups,
is seen to be untenable.

Hydrocarbon Derivatives. Hitherto hydrocarbons only have
been described as constituents of the native bitumens, but there
are other substances entering into their composition which con-
tain, in addition to carbon and hydrogen, sulphur, nitrogen, and
more rarely oxygen. They consist of cyclic compounds contain-
ing an atom of sulphur or nitrogen in the carbon ring, compounds
which are, in asphalt, dicyclic and polycyclic, and oxygen deriva-
tives, probably polycyclic phenols, together with oxidation products
of the hydrocarbons. As these constitute but a minor portion
of native bitumen they will not be described in detail here. They
can be readily separated from the hydrocarbons by appropriate
reagents, but have not been closely studied. They are also found
in various dense petroleums.

The nature and structure of the hydrocarbons and their deriv-
atives have been entered into with some detail since the relative
proportion of the various series which are present in any petro-
leum or solid bitumen has a strong influence on its characteristics,
and, more especially, the relation of saturated to unsaturated
hydrocarbons and of the paraffines to polymethylenes, these con-
siderations being, of course, quite apart from the relative amounts
of liquid and solid substances, malthenes and asphaltenes, upon

NATURE OF THE HYDROCARBONS. 109

which the consistency of the bitumen, but not its chemical char-
acteristics, are based.

The saturated hydrocarbons, especially those of the paraffine
series which are found in the residues from the distillation of
Pennsylvania petroleum and of those from Ohio, Kentucky, Kansas,
and similar oils, are most stable. They are not readily attacked
by strong acids, alkalies, or water They form by far the largest
part of the residuums derived from these petroleums. The satu-
rated hydrocarbons of the polymethylene series found hi some
residuums as well as in the solid bitumens are not attacked by
acids or water, but are readily condensed by the abstraction of
hydrogen under certain other conditions. The polymethylene
hydrocarbons or the petroleums containing them are in this way
the primary source of all asphalts. No asphalt can originate in
nature in a paraffine oil, but all polymethylene oils leave an as-
phaltic residue on weathering or on evaporation or distillation
with heat.

No solid native bitumen suitable for paving purposes is known
which contains paraffine, while the relative proportions of satu-
rated and unsaturated hydrocarbons in them may be very variable.

The characterization of heavy oils or of solid bitumens and
their differentiation is, therefore, arrived at by determining by
appropriate analytical methods and by treatment with reagents
the relative proportion of the malthenes and asphaltenes present,
the proportions of saturated to unsaturated hydrocarbons in
the malthenes, and the characteristics of all these classes of bitu-
mens. The asphaltenes are probably composed entirely of unsat-
urated or unstable compounds.

SUMMARY.

For a thorough understanding of the nature of the native bitu-
mens the constitution of the various hydrocarbons of which they
may be composed has been outlined in the preceding chapter.
This involves some knowledge of chemistry and may, therefore,
be somewhat unintelligible to the general reader, but the state-
ments here presented are entirely necessary in any treatise which
aims at being at all complete in its consideration of the subject
of the native bitumens and of asphalt paving mixtures.

CHAPTER VI.

CHARACTERIZATION AND CLASSIFICATION OF THE NATIVE

. BITUMENS.

IN a recent article on the " Bitumens of Cuba " the author has
shown that while there have been numerous attempts to define the
nature of bitumen, and to characterize and classify the various forms,
none of them has been satisfactory, and that this has been plainly
due to the fact that it is only recently that a sufficient number
of deposits have been studied in their native environment, and
in the laboratory, by methods which were sufficiently developed
to reveal anything as to the constitution of the harder forms. For
example, the fact that hard bitumen in the form of asphalt con-
sists of cyclic polymethylenes of two or more rings, of the con-
densation products of such hydrocarbons and of their derivatives,
and that this form of bitumen is without doubt the result of the
metamorphism of cyclic petroleums by natural causes has only
been made apparent within the last few years.

This lack of data to serve as a basis of comparison and char-
acterization of species, and as an aid to the close definition of
what bitumen is, and how its various forms can be differentiated,
has been largely supplied, as far as the harder forms maltha,
asphalt, gilsonite, grahamite, albertite, etc. are concerned, by
the examination in the author's laboratory of several hundred
occurrences of these materials, scattered over the greater portion
of the United States and Canada, the West Indies, and the northern
coast of South America. Our knowledge of the nature of various
forms of petroleum has also been greatly extended by the work
of C. F. Mabery, C. E. Coates, E. O'Neill, and numerous con-
tinental chemists, and that of natural gas by F. C. Phillips and
others.

There is, of course, a vast field still open for research, but it

110

NATIVE BITUMENS. Ill

is believed that the presentation of the subject here given is based
on more complete evidence than anything heretofore attempted.

What is Bitumen? The most rational way of approaching
the question appears to be to present the definitions and character-
ization of this class of materials as a whole, and then of the par-
ticular forms as they may be differentiated by the available evi-
dence; that is to say, to put the results which have been reached
before the reader, and then to show how these have been arrived
at from the data and evidence available.

As a beginning, bitumen and pyrobitumen must be defined:

Native Bitumens and Pyrobitumens. Native bitumens con-
sist of a mixture of native hydrocarbons and their derivatives,
which may be gaseous, liquid, a viscous liquid or solid, but, if
solid, melting more or less readily on the application of heat, and
soluble in turpentine, chloroform, bisulphide of carbon, similar
solvents, and in the malthas or heavy asphaltic oils. Natural
gas, petroleum, maltha, asphalt, grahamite, gilsonite, ozocerite,
etc., are bitumens. Coal, lignite, wurtzilite, albertite, so-called
indurated asphalts, are not bitumens, because they are not soluble
to any extent in the usual solvents for bitumen, nor do they melt
at comparatively low temperatures nor dissolve in heavy asphaltic
oils. These substances, however, on destructive distillation give
rise to products which are similar to natural bitumens, and they
have been on this account defined by T. Sterry Hunt as " pyro-
bitumens," which differentiates them very plainly from the true
bitumens. They usually contain oxygen, whereas the true bitu-
mens, as a rule, do so to only a limited extent.

There is, of course, no sharp dividing line between bitumens
and pyrobitumens, as the former are gradually metamorphosed
by tune and exposure to varied environment into the latter.

Classifications of Bitumens. Among the bitumens there are
such variations in physical attributes and chemical composition
that they may be differentiated as follows:
BITUMENS:

GAS.

Natural hydrocarbon gases.
Marsh-gas.

112 THE MODERN ASPHALT PAVEMENT.

PETROLEUM.
Paraffine-oils.

Consisting of hydrocarbons and their derivatives, the
lower members of which belong entirely to the
paraffine series and have the general formula C n H 2n + 2 .

(1) Those containing C n H 2n+2 hydrocarbons up to
C2gH58 with but traces of sulphur derivatives : Penn-
sylvania, West Virginia, Kentucky, Kansas, Colorado,
etc.

(2) Those containing C n H 2n +2 hydrocarbons up to
CnH 2 4 and above that C n H 2n and C n H 2n -2 poly-
methylenes with considerable amounts of sulphur
derivatives: Ohio, Canada.

Cyclic Oils.

Consisting principally of polymethylene hydrocarbons
of the series C n H 2n , C n H 2n _ 2 + C n H 2n _ 4 , together
with a certain amount of unsaturated hydrocarbons
and their derivatives.

(1) Stable polymethylenes, consisting largely of naph-
thenes, C n H 2n : Russian oils.

(2) Less stable polymethylenes together with consider-
able amounts of unsaturated hydrocarbons and
their nitrogen and sulphur derivatives, and leaving
an asphaltic residue on distillation: California.

Oils of Mixed Composition.

Semi-asphaltic oils composed largely of stable paraffine
and polymethylene hydrocarbons not readily attacked
by sulphuric acid: Texas.
MALTHA.

Known also as mineral tar, brea, and chapapote.

Originating from polymethylene petroleums alone.
Transition products between oil and asphalt.
SOLID BITUMENS.

Consisting largely of paraffine hydrocarbons.

Ozocerite, hatchettite, etc.

Consisting of unsaturated cyclic hydrocarbons.
Ter -penes, fossil resins, amber, etc.

NATIVE BITUMENS. 113

Derived from or originating in polymethylene petroleums,
the more volatile components consisting of di- or tricyclic
saturated hydrocarbons.
Asphalts.

Asphalt. Numerous varieties: Trinidad, Venezuela, Cali-
fornia, Cuba.
Glance pitch.
Consisting largely of cyclic hydrocarbons attacked by strong

sulphuric acid, but which otherwise are stable.
Manjak.

Yielding high fixed carbon and a black powder.
Gilsonite.

Yielding average fixed carbon and a brown powder, .the
more volatile components resembling sticky oleo resins
rather than hydrocarbons found in the asphalts.
Grahamite.

Consisting of hydrocarbons almost entirely insoluble in
naphtha and yielding a higher percentage of fixed carbon
on ignition. Melting with difficulty.
The grahamites rapidly shade into pyrobitumens.
PYROBITUMENS:

Practically insoluble in chloroform or heavy petroleum hydro-
carbons.

Derived from petroleum,
Albertite, with varieties called nigrite, etc.
Wurtzilite.

Derived from direct metamorphoses of vegetable growth.
Anthracite.
Bituminous coal.
Lignite.
Peat (?).

Of the bitumens, as we have seen, the hard ones and the
oils enter into the composition of paving cements and
must be considered individually.

SUMMARY.

The author's classification of the native bitumens and those
of other writers are unfortunately founded on an empirical basis

114 THE MODERN ASPHALT PAVEMENT.

to too great a degree to admit of their being satisfactory to every
one. Such classifications can be regarded as mere steps toward
a final conclusion which can only be arrived at after years of
investigation of the subject. The author's classification is pre-
sented for what it is worth, and there will be no hesitation in
modifying it in the future in the light of any additional informa-
tion which may become available which is based on facts and
not on mere opinion or theory. For a thorough understanding
of the character of the native bitumens it is advisable that the*
general reader should acquaint himself with the peculiarities 01
the different classes which it has been possible to differentiate,
the one from the other, in order that an intelligent understanding
may be arrived at of the very variable nature of the bitumens in
use in the asphalt paving industry.

The mere minute differences in the various bitumens, upon
which their differentiation has been based, will be made plain in
the following pages.

PAET III.

NATIVE BITUMENS IN USE IN THE PAVING
INDUSTRY.

INTRODUCTION.

IN the light of the preceding classification of the native bitumens
and our knowledge of the various series of hydrocarbons of which
they are composed the characteristics of the fluxes, the asphalts,
and other solid native bitumens in use in the paving business
may now be taken up.

CHAPTER VII.

DIFFERENTIATION AND CHARACTERIZATION OF THE NATIVE

BITUMENS.

ALL the native bitumens are such complicated mixtures of
various hydrocarbons and their derivatives that it is impossible
to separate them completely into their individual constituents
and to differentiate and characterize them in this way. Recourse
must, therefore, be had to the determination of their physical
properties and to attempts, more or less successful, to separate
the proximate constituents of which they are made up into various
classes according to their solubility and behavior towards reagents,
supplemented by the determination of the amount of fixed carbon
which they yield on ignition and their ultimate composition.

Physical Properties. The physical properties which are of

value in characterizing the bitumens are:

115

116 THE MODERN ASPHALT PAVEMENT.

SOLID BITUMENS.

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F. Original substance, dry

' ' " " Pure bitumen

Color of powder or streak

Lustre

Structure

Fracture.

Hardness, original substance

Odor

Softens

Flows

Consistency, penetration at 78 F

The chemical characteristics of interest are:

CHEMICAL CHARACTERISTICS.

Original substance

Loss, 212 F., 1 hour

Dry substance

Loss, 325 F., 7 hours

Character of residue

Consistency, penetration of residue at 78 F

Loss, 400 F. , 7 hours (fresh sample)

Character of residue

Consistency, penetration of residue at 78 F.

Bitumen soluble in CS 2 , air temperature

Inorganic or mineral matter

Difference . .

Malihenes:

Bitumen soluble in 88 naphtha, air temperature . . ,

This is per cent of total bitumen

Per cent of soluble bitumen removed by H 2 SO 4 ....
Per cent of total bitumen as saturated hydrocarbons.

Bitumen soluble in 62 naphtha

This is per cent of total bitumen

Carbenes:

Bitumen insoluble in carbon tetrachloride, air tem-
perature

Bitumen yields on ignition :

Fixed carbon. . .

Sulphur

Ultimate composition

NATIVE BITUMENS. 117

FLUXES.

PHYSICAL PROPERTIES.

Specific gravity, dried at 212 F., 78 F./78 F.

Flows, cold test

Color

Odor. . .

Under microscope

Flashes, F., N. Y. State oil-tester '.

Viscosity P.R.R. pipette at - F. . ,

CHEMICAL CHARACTERISTICS.

Original substance

Loss, 212 F., 1 hour or until dry

Dry substance

Loss, 325 F., 7 hours

Character of residue

Penetration of residue at 78 F. .

Loss, 400 F., 7 hours (fresh sample).

Character of residue

Penetration of residue at 78 F. .

Bitumen soluble in CS 2 , air temperature.

Inorganic or mineral matter

Difference . .

Bitumen insoluble in 88 naphtha, air temperature,

pitch

Per cent of soluble bitumen removed by H.jSO 4

Per cent of total bitumen as saturated hydrocarbons

Per cent of solid paraffines

Fixed carbon

Ultimate composition.

Specific Gravity. For the physical characteristics it may be
said that the specific gravity of the solid bitumens as they are
originally found in nature will depend very largely upon the per-
centage of mineral matter which they contain. One like Trinidad
lake asphalt, containing 37 per cent of mineral matter, will have
a specific gravity of 1.40, while an extremely pure bitumen, like
gilsonite, will have a specific gravity of 1.04.

118 THE MODERN ASPHALT PAVEMENT.

Where the pure bitumen is extracted from the native material
the density will not, as a rule, vary very widely. For the asphalts
it will lie between 1.03 to 1.07.

Ozocerite is the only solid bitumen which has a specific gravity
below 1.00, .912, while grahamite, albertite, and glance pitch
exceed a density of 1.09.

The density of the residual pitches, at least of those met in
the paving industry, is usually quite similar to that of the
bitumens of the native asphalts, not rising above 1.1, unless in
their preparation they have been carried to a very high tem-
perature, nor falling below 1.0 if they are at all solid. Some of
the condensed oils, such as Byerlyte, Pittsburg flux, and blown
asphaltic oil, have a density below 1.0 and can be recognized by
the fact that they float on water.

The specific gravity of the various bitumens is, therefore, of
some considerable interest.

A summary of some of the data collected by the author is
given in the following table:

Substance. Specific Gravity.

Trinidad lake refined asphalt 1 . 4000

land " " 1.4196

Bermudez refined asphalt 1900 1 .0823

11 1903 1.0575

Maracaibo refined asphalt. . 1 . 0667

Cuban (Bejucal) asphalt 1 .3050

Mexico Tamesi River asphalt (dry) 1 .1180

11 chapapote asphalt (dry) 1 .0450

California La Patera 1 . 3808

" Standard, refined 1 .0627

Utah gilsonite firsts 1 .0433

" - " seconds 1 .0457

Grahamite Indian Territory, Ten Mile Creek. . 1 . 1916

Colorado, Middle Park 1 . 1600

Egyptian glance pitch ' 1 . 0970

Manjak 1 .0844

Ozocerite Utah .9123

Albertite Nova Scotia 1 .0790

" Utah 1 .0990

" Cuba 1 .2040

Wurtzilite Utah 1 .0556

Kentucky, Grayson Co. seepage 0.9783

NATIVE BITUMENS. 119

Substance. Specific Gravity.

Utah, Soldier Creek extracted bitumen 1 .2000

" Grand Co. extracted bitumen 1 .0370

"D" grade Calif orina, carefully prepared 1 .0622

carelessly " .... 1.0887

Asph. O.&Ref. Co.... 1.0770

Beaumont, Texas, oil asphaltic residue 1 .0803

Baku pitch from Russian petroleum 1 . 1098

Pittsburg flux .9879

Ventura flux 1 .0199

Byerlyte, paving 1 .0230

roofing 0.9070

Hydroline "B" 1 .0043

Color of Powder or Streak. Where the solid native bitumens
are sufficiently hard to permit of their being powdered or to make
a streak upon porcelain, the color of the powder or streak may
be of some value in differentiating them. For example, the powder
of refined Trinidad lake asphalt is of a bluish-black color, whereas
that of Trinidad land asphalt is distinctly brown. Most of the
asphalts give a powder of either a dull-black or brownish-black
color, but gilsonite is distinguished by its extreme brittleness and
the fact that the powder is of an extremely light-brown color.

Lustre. All the native bitumens are lustrous if pure, with
the exception of ozocerite. The residual pitches are, of course,
lustrous. In the presence of mineral matter the lustre is more
or less diminished, depending upon the amount of the latter.

Structure. The structure of the native bitumens is in many
cases very characteristic. To begin with, it is either uniform
and homogeneous in every part or the reverse. In crude Trinidad
lake asphalt we note the presence of gas cavities and of emul-
sified water. In some California asphalts large particles of brec-
ciated shale are scattered through the native asphalt, which occurs
in veins. On the other hand, gilsonite is of an extremely uniform
structure except where the material approaches the vein wall,
where it at times takes on a columnar structure due to weather-
ing. Glance pitch and manjak are also of extremely uniform
structure. The same thing may be said in regard to many refined
asphalts in which the lack of homogeneity has been removed by
melting. The structure of the residual pitches is, of course, quite

120 THE MODERN ASPHALT PAVEMENT.

homogeneous, except where they may have been coked to a cer-
tain degree by excessive heating.

Fracture. The fracture of the native solid bitumens in many
cases is as characteristic as the structure. Almost all grahamites,
although homogeneous in structure, have a peculiar fracture which
distinguishes them from all the other solid bitumens. It has
been described as a hackley or pencilated fracture, which, perhaps,
covers it sufficiently. It is an irregular fracture and shows no
evidence of a purely lustrous surface, as in the fracture of gilsonite.
The fracture of crude Trinidad asphalt is quite irregular, that of
gilsonite conchoidal and highly lustrous, while that of many refined
asphalts is only semi-conchoidal.

Hardness. The hardness of the native bitumens in the form
in which they originally occur may be stated in terms of Mohr's
scale. Where the pure bitumen is softer than 1 of this scale, it
may be stated in terms of one of the various penetration machines.

Odor. The odor of most of the native bitumens is character-
istic at ordinary temperatures. The asphalts have in general
an asphaltic odor, but some of them, such as that from near the
Gulf of Maracaibo, in Venezuela, are characteristically rank. Gil-
sonite has scarcely any perceptible odor, while the residual pitches
have a peculiar oily odor. On heating, stronger odors are com-
monly developed which are recognizable to one accustomed to
them, but the nature of which is difficult to describe in print.

Softening and Flowing Points. The native bitumens possess
no melting-point. It can be stated that they are in a melted
condition at such and such a temperature, but since they are
made up of a mixture of hydrocarbons and their derivatives it
is impossible for them to have a true melting-point, such as that
of ice or any definite compound. In cooling a mass of water in
which a thermometer is immersed from any temperature, say 50 F.,
to a point below freezing and representing this on a system of
coordinates, the time being denoted by the abscissae and temper-
ature by the ordinates, a curve will be developed which at the
point of freezing, while the water is being converted into ice, is
broken by a straight line which denotes the tune during which
the liquid is becoming converted to ice. If any native bitumens

NATIVE BITUMENS. 121

are melted and cooled in the same way no definite break corre-
sponding to any definite freezing-point is detected. It is, therefore,
impossible for us to speak of the melting-point of a bitumen, but
we may determine in any empirical way the point at which any
solid bitumen softens and again when it flows, as specified in the
author's method given in Chapter XXVIII. The determinations
of this nature given in the following pages were made in this way.
Chemical Characteristics. It will be noted that in the differen-
tiation of the bitumens into classes by means of solvents, certain
names have been applied to the various classes of hydrocarbons.
In the early days of the study of the behavior of solvents towards
native bitumens, the various hydrocarbons and their derivatives
were differentiated, according to their solubility in naphtha, into
classes to which the names " Petrolene " and " Asphaltene,"
terms used by Boussingault in his earliest investigations, were
applied. These terms were open to the objection that it led per-
sons not thoroughly acquainted with the chemistry of the natural
hydrocarbons to believe that petrolene and asphaltene were
definite compounds, which, of course, is no more the case than to
assume that the oil known under the name kerosene is a definite
compound. The author, therefore, changed the designation to
" Petrolenes " and " Asphaltenes " as more plainly indicating
that the differentiation was merely one of classes. More recently
it has seemed possible to carry the differentiation still further,
since it has been found that the solubility of the native bitumens
in cold carbon tetrachloride is not in all cases the same as in bisul-
phide of carbon or chloroform and that solubility or insolubility
in this medium can be added to those previously employed for
this purpose. Peckham has also shown that chloroform dissolves
hydrocarbons which are not soluble hi carbon disulphide, and
this solvent may be added to our list, or even oil of turpentine
if thought necessary. The author's idea in regard to the future
differentiation of the native bitumens would involve the applica-
tion of the term " Malthenes," to the bitumens soluble hi naphtha
in place of the term " Petrolenes," stating the specific gravity
of the naphtha used as a solvent, as this class of bitumens bears a
great resemblance to the malthas; reserving the term " Petrolenes "

122 THE MODERN ASPHALT PAVEMENT.

for those hydrocarbons which are volatile at 325 F. in 7 hours,
according to the author's method, these hydrocarbons being com-
paratively light oils resembling ordinary petroleum. He would
characterize as " Asphaltenes " those hydrocarbons and their
derivatives which are soluble in cold carbon tetrachloride and
as " Carbenes " those not soluble in cold carbon tetrachloride,
but soluble in carbon disulphide. This differentiation has not
been carried out in the author's work to any great extent in the
past, but in many cases in the following tables the percentage of
the bitumen insoluble in cold carbon tetrachloride will be found
to prove of great interest.

In the determination of the amount of fixed carbon which any
native bitumen will yield when heated to a high temperature in
the absence of oxygen, after the manner proposed for making the
same determination in coal, data are obtained which are of great
interest as showing the relative proportion of carbon and hydrogen
in the bitumen under examination. In the case of the parafnne
hydrocarbons of the formula C n H 2n +2 no fixed carbon is left on
ignition, while the amount increases with each diminution in the
proportion of hydrogen to carbon, until in grahamite as much
as 50 per cent is found, where the relation of carbon to hydrogen
is as 8 to 1.

The ultimate composition of the bitumens is, of course, of
interest, but for general purposes the information derived from
this determination seldom repays the time and care necessary.

The methods employed in arriving at the results which are
presented in the following tables are given hi a later chapter of
this book, and reference must be made to it for the details of the
process and for a thorough understanding of what each deter-
mination may mean. 1

It will be found on examining these methods that the results
obtained by their use are in no sense absolute determinations,
but when each of the asphalts is treated in quite the same way
as the others they are of great value relatively and for purposes
of comparison and differentiation. In regard to certain of the
data some explanation will be necessary.

1 Page 519.

NATIVE BITUMENS. 123

Bitumen Soluble in Bisulphide of Carbon, Air Temperature. This
determination shows the amount of bitumen which is soluble in cold
bisulphide of carbon. If hot bisulphide of carbon, chloroform, or
turpentine and in some exceptional cases hot carbon tetrachloride
are used as a solvent, a slightly larger percentage would probably be
found but, owing to the difficulties in maintaining uniform conditions
for extraction at other than ordinary temperatures and for other
reasons which it is unnecessary to specify here, cold bisulphide
of carbon has been used and the results obtained are more satis-
factory for comparative purposes than would otherwise be the
case.

Inorganic Matter. The determination of inorganic or mineral
matter represents the residue remaining on the ignition of the
native bitumen in a muffle at such a temperature as to remove
all the carbon. In certain cases the amount obtained may be
less than that originally present, owing to the volatilization of
alkalies or sulphuric acid, but it is sufficiently accurate for pur-
poses of comparison.

Insoluble or Difference. On adding together the percentages
of bitumen soluble in carbon disulphide and of inorganic matter
obtained on ignition, the sum will seldom amount to 100.0. The
difference which is stated as such in our analyses has for many
years been considered as organic matter not bitumen. This may
be true in exceptional cases, but recent investigations 1 have shown
that it is not at all so in many bitumens. For example, in Trinidad
asphalt it has been found to consist of the water of combination of
the clay which the material contains and some inorganic salts which
are volatilized on ignition. The amount of organic matter is
extremely small. In other cases, it may consist to a considerable
extent of grass and twigs, as in the seepages which have run
out over sod. On the whole, therefore, it seems desirable not to
describe it by any definite name, but merely as an undetermined
difference.

1 Proc. Am. Soc. Test. Mat., 1906, 6, 509.

124 1HE MODERN ASPHALT PAVEMENT.

Loss at 825 F. in 7 hours. The hydrocarbons lost at 325 F.,
which the author has proposed to denominate " Petrolenes," as a
class, will vary in amount enormously according to the conditions
under which the heating is carried out. When these are accu-
rately defined, however, as in our methods, the relative loss is
an important indication in differentiating any two bitumens.

Malthenes. It is a well-known fact that the percentage of
malthenes or bitumen soluble in naphtha will vary according to
the solvent used, and in the case of naphtha if its density is low,
according to the method by which the solvent is applied. In
the determinations given in the table naphthas of a definite density,
88 and 62 Beaume, have been allowed to act on the native bitu-
men, in as finely a comminuted condition as possible, in the cold
for a definite length of time and the residue washed quite clean
with the same solvent. Had the solvent been applied warm or
in a continuous percolation apparatus the figures would have
been higher but would not have been constant, since the lighter
hydrocarbons in the solvent would have been gradually volatilized
and their solvent power slowly increased. For comparative pur-
poses the method in use is more satisfactory than any other, but
the results themselves are of no value as an absolute differentiation
of the native bitumen into two definite classes of materials.

Action of Strong Sulphuric Acid on the Hydrocarbons. The
results presented showing the action of strong sulphuric acid on
the hydrocarbons soluble in 88 naphtha are only of value because
all of them are carried out according to a definite and arbitrary
method. If this were varied the results would also vary. It is
important to know that strong sulphuric acid has a somewhat
different action on the hydrocarbons of a solid bitumen if it is
allowed to act on an 88 or 62 naphtha solution of them. In
the 62 naphtha solution the action is apparently much less than
in the solution of the solvent of lighter density.

Bitumen Insoluble in Carbon Tetrachloride, Air Temperature.
The amount of bitumen soluble in cold carbon tetrachloride, in all
the normal asphalts, is practically the same as that soluble in
carbon disulphide, but in certain native bitumens hydrocarbons are

NATIVE BITUMENS. 125

found which are insoluble in this medium. This differentiates these
bitumens grahamite, for example from the true asphalts, the
insoluble bitumen forming a class of hydrocarbons or of their
derivatives to which the name of " Carbenes " has been applied.
Such bitumens have evidently been much more metamorphosed
by weathering or otherwise than the true asphalts, or have orig-
inated in a different series of hydrocarbons.

In residual pitches, at tunes some of the bitumen is found
which is insoluble in cold carbon tetrachloride, and this is evi-
dently due to the severe treatment which the material has suffered
in the course of its production at very high temperatures. A
determination of the amount is only valuable as an indication
of the care which has been used in the preparation of such pitches.
In the best asphaltic residues from California petroleum the per-
centage of " Carbenes " has been found to vary from 7 to less
than one-half of 1 per cent.

Fixed Carbon. The amount of fixed carbon which any solid
native bitumen will yield will depend, as in the case of coal, upon
the way in which the material is ignited. All the deteYmina-
tions given have been made by following the 'scheme suggested
by the Committee of the American Chemical Society on the Analysis
of Coal, and are, therefore, strictly comparable.

With these facts in view the analyses of the native bitumens
which are to be presented will be of interest for the purpose of
comparing the characteristics of the various materials, but the
chemical data must not be looked upon as being absolute in any
case.

SUMMARY.

By determining the physical characteristics of any native
bitumen, its specific gravity, its color hi a powdered condition,
its lustre, structure, fracture, hardness, odor, softening point,
and consistency, by differentiating the bitumen into various classes
of hydrocarbons by means of solvents and by observing certain
other chemical characteristics, such as the ultimate composition,
the amount of fixed carbon left on ignition, and the extent to
which the hydrocarbons of which it is composed are acted upon

126 THE MODERN ASPHALT PAVEMENT.

by strong sulphuric acid, it is quite possible to characterize and
classify the various native bitumens in such a way as to make it
possible to recognize them without difficulty and without con-
fusing one with another. The methods for making these deter-
minations appear in Chapter XXVIII.

CHAPTER VIII.
PETROLEUMS.

IN the asphalt paving industry the petroleums are of interest
because the heavier hydrocarbons or residuum which remains
on distilling off the lighter portion of the oil are used as a flux
for the solid native bitumens. The character of these residuums
and fluxes reflects, of course, the nature of the petroleum from
which they have been made, the paraffine oils yielding a residuum
of one kind, the asphaltic oils one of another, and the mixed par-
affine and asphaltic oils, such as those from the Beaumont field
and elsewhere in Texas and from Oklahoma and part of Kansas,
one of quite another character.

For a more definite knowledge of the proximate composition
of various petroleums, beyond that which has been given in the
classification, reference must be made to the publications of those
who have devoted their tune to a study of this subject, among
whom may be named Mabery, Young, Markownikoff, and Engler.
Unfortunately the publications of these investigators are widely
scattered and appear nowhere hi condensed form. Their general
conclusions have been included hi the author's classification
of petroleums.

Malthas. The malthas are viscous liquid natural bitumens cor-
responding hi consistency to that of the artificial residuums or usually
denser. They are only rarely of a suitable character for use as a
flux, owing to the fact that on heating they are generally rapidly
converted into a harder material by the loss of volatile hydro-
carbons and condensation of the molecule. It was due to this
fact that the early pavements laid with Alcatraz asphalt were
not successful. The flux in use was a natural maltha derived
from the Carpinteria sands, which, while of the proper consistency

127

128 THE MODERN ASPHALT PAVEMENT.

as prepared, was rapidly converted into- a solid bitumen on pro-
longed heating.

The bitumen in the Kentucky sands is much more of the nature
of a maltha than of an asphalt, and it is on this account that these
materials are unsatisfactory for paving purposes in addition to
the fact that they contain usually a much too small percentage
of bitumen.

For the above reasons the native malthas are rarely used in
the preparation of asphalt cement and. never with successful
results.

On page 129 are given some examples of the characteristics of
malthas from various parts of the world.

It appears that the malthas may, when dry but not otherwise
altered, have a specific gravity either less or greater than water.
They all volatilize a very considerable amount of light oils at
325 F., and most of them large amounts at 400 F. In the case
qf four out of nine of those cited the residue after heating to only
325 F. was hard enough to permit a determination of their con-
sistency with the penetration-machine. Under the same circum-
stances a paraffine residuum would still remain a flux, as would
the best asphaltic petroleum residuums. They frequently contain
notable percentages of the asphaltenes and all become pitch after
heating to a constant weight at 400 F.

The Residuums of Petroleums Used as Fluxes in Softening
the Solid Native Bitumens. In order to bring the solid native
bitumens to such a consistency as will make them available for use
as a paving-cement it is generally necessary to flux them with
some other softer bitumen. The flux in use for this purpose is
uniformly a heavy residuum prepared by the removal of the lighter
portions of petroleum by distillation. These residues naturally
vary in character in the same way that the petroleums do from
which they have been derived, and that the petroleums are
very variable, depending upon the series of hydrocarbons of which
they are composed, has already been made evident. The oils from
which residuums or fluxes are prepared for use in the United States
are the paraffine petroleums from the Eastern, Ohio, Kentucky,
Kansas, Oklahoma, and Colorado fields, the asphaltic petro-

PETROLEUMS.

129

Characteristics of Malthas.

Test No.

30374.

California,
Sunset
District.

30116.

California,
McKitrick
District.

25129.

Cuba,
Hato
Nuevo.

50912.

Cuba,

Matanzas.

Loss 212 F

270 F.

.9884
12.4%

16.1%

7.76%

11.0%
325 F.

1.0029
8-9%

13.2%

22.8%

32.1%
16

17.6%
8.1%
8.6%

245 F.

.9867
7.0%

28.0%
40

DRY SUBSTANCE.

Specific gravity, 78 F./78 F.
Loss 325 F 7 hours . . .

.9445

5.8%
38

Penetration of residue at 78F.
Loss 400 F , 7 hours

Penetration of residue

Loss 325 F to constant wt

Penetration of residue

Loss, 400 F. to constant wt .

Bitumen insoluble in 88
naphtha, pitch

6.7%

25.5%
15.7%

Bitumen insoluble in 62
naphtha pitch

Bitumen yields on ignition :
Fixed carbon ....

Test No.

Characteristics of Malthas.

39556.

Venezuela,
Peder-
nales.

10106.

Venezuela,
Mene
River.

63846.

Texas,
Austin.

30283.

Trinidad,
Boodoo-
shingh.

60487.

Trinidad,
Mara-
bella.

Loss, 212 F

9.7%

9.0%

1.5%

18.5%

5.2%

Flash-point

DRY SUBSTANCE.

Specific gravity 78 F./78 F.
Loss, 325 F., 7 hours
Penetration of residue at 78F.
Loss, 400 F , 7 hours . . .

1.032

4.6%
75
11 1%

8^5%
5 9%

.974
7.1%

17 0%

10^%
hard

6- '4%
32
10 3%

Penetration of residue

Loss, 325 F to constant wt. . .

Loss, 400 F. to constant wt . .
Penetration of residue

31.0%
13

Bitumen insoluble in 88
naphtha, pitch

9.4%

3.5%

40.4%

Bitumen insoluble in 62
naphtha, pitch

25 9%

Bitumen yields on ignition:
Fixed carbon

7 0%

10.0%

* Too large to read.

130 THE MODERN ASPHALT PAVEMENT.

leums from California and the petroleum of mixed character
from Texas containing both paraffine and asphaltic hydrocarbons.
The residues from paraffine petroleum usually carry a very con-
siderable amount of paraffine scale, which has a decided influence
on their character. The residue from Texas petroleum from the
Beaumont field, now nearly exhausted, contained only a small
amount of paraffine hydrocarbons and not more than 1 per cent
of paraffine scale. The supply of flux of this character is now very
small, and is all produced at Chaison, Texas. The greater part
of the Texas flux on the market in 1907 was produced from pipe
line oil from the various Texas and Oklahoma fields. This petro-
leum is necessarily of a very mixed character and the flux pre-
pared from it has contained very considerable amounts of paraffine
hydrocarbons and scale, the latter reaching at times as high
as six per cent. It is far less satisfactory as an asphaltic flux than
that prepared from the original Beaumont oil, and in some respects
is not a suitable substitute for it. For use with ordinary asphalts
it is, however, as good for paving purposes, although perhaps
not quite as stable at high temperatures. It must be used as a
paraffine flux. The California residuum or flux is composed
almost entirely of complicated polymethylenes and contains notable
proportions of aromatic hydrocarbons, nitrogenous bases, and
phenols, and is characterized by its very great density.

Residuums from the same kind of petroleum may, on the
other hand, vary very largely among themselves, according to
the care with which they have been prepared. It is quite as
necessary, therefore, to determine whether a flux has been care-
fully distilled as it is to know the character of the petroleum from
which it has been produced.

In making a comparative study of the available fluxes it will
also be necessary to determine whether there is a preference in
favor of one over another as an actual solvent for the solid bitu-
mens, and to determine their stability and such other properties
as may point to their adaptability for use in the preparation of
an asphalt cement.

The fluxes now in use in the paving industry may be con-
sidered as consisting of the three classes which have been men-

PETROLEUMS.

131

tioned above, and their character may be taken up according to
such a classification.

Paraffine Petroleum Residuum. Paraffine petroleum residuum
is the form of flux originally used in the asphalt paving industry
in the United States. It is also known as petroleum tar and was
originally a by-product remaining after the distillation of crude
Pennsylvania paraffine petroleum for the production of illumi-
nating-oil. It was the flux used by De Smedt in the early days
of the paving industry for imparting a proper consistency to Trini-
dad asphalt in the production of paving-cement. Its use has
been largely continued up to the present time, but its character
has been greatly modified and improved. It is no longer a by-
product, but is especially prepared for the purpose.

In the early days of the industry the residuum used in the
preparation of asphalt cement was of most varied character.
Among available records it is found that in the summer of 1888
several residuums in use had the following properties:

Flash.

Gravity.

230 F.
203
226
392
428
176

22 B.
24
25
27
22

.9240
.9130
.9070
.8950
.9240

The result of making a cement with oil flashing at 176 F.
would be that much of it would be easily volatilized at the tem-
perature of melted asphalt cement, 325 F., and that the con-
sistency would of course be most unstable.

In 1889 the oils hi use had the following characteristics:

Source.

Gravity.

Flash.

Volatile, 400 F.

Eagle Refining Co

22.8 B.

416 F.

14.40%

MaToney Oil (To

21.7

361

14.33

Jenney Mfg Co . .

20 8

271

16.80

132 THE MODERN ASPHALT PAVEMENT.

As late as 1892 the oils were still variable:

Source.

Gravity.

Flash.

Volatile, 400 F.

Solar . .

21 B.

417 F

4 06%

Jenney

18.5

235

11 20

National

20.0

295

15 94

Whiting

22.2

415

5 39

Continental (Denver)

23.8

345

15 73

Crew Levick

20 4

320

17 36

In 1891 the Standard Oil Company undertook to prepare a
heavy oil especially for paving purposes which should be fur-
nished to those who were particular as to its character and were
willing to pay for a better quality. It has had the following
average composition since 1896:

PARAFFINE RESIDUUMS. AVERAGE AND EXTREMES IN
COMPOSITION FOR FOUR YEARS.

Year.'

Volatile,
400 F.

Extremes.

Specific
Gravity.

Extremes.

Flash.

Extremes.

1896

4.7%

2.6- 8.8

.9313

.9204-. 9351

430 F.

397-476

1897

6.1

1.4-12.3

.9302

.9219-. 9383

420

393-456

1898

5.1

2.9- 8.8

.9327

.9206-. 9397

432

392-455

1899

3.8

2.0- 6.4

.9331

.9295-. 9376

442

416-460

Consistency after heating : Flows slowly at about 78 F.

The great uniformity of the supply is apparent. Oil of the
old character is still in use by careless contractors, as it is cheap.
Oils of this description which have come under my observation
have the following characteristics:

Specific
Gravity.

Flash.

Volatile,
400 F., 7 Hours.

.9100

280 F.

23.9%

.9197

330

17.3

.8829

260

27.2

.9222

286

12.6

The best paraffine residuum from Ohio petroleum and some of
the Kansas and Colorado fields, which has been in use during the

PETROLEUMS.

133

past six or seven years has commended itself in quality from the
fact that it is carefully prepared for the paving industry by dis-
tillation with steam agitation without cracking that is to say,
decomposition of the oil that it has a high flash point, and on
this account contains little oil volatile at the temperatures at
which asphalt cement is maintained in a melted condition, and
that it is uniform. The possible disadvantages of such dense
residuum, if they are such, in comparison with the lighter and
more volatile oils are, that more of this oil must be used to pro-
duce a cement of given penetration or consistency and that at
comparatively low temperature asphalt cements made with it
harden more than when lighter oils are used, owing to the separa-
tion of paraffine scale.

The only advantage of the lighter form of residuum, how-
ever, is the one just mentioned, that it and the cement prepared
from it do not harden or solidify as much in winter tempera-
tures.

Typical Paraffine Fluxes of 1907. In the following table are
given the characteristics of various paraffine fluxes which were
available and in use in the paving industry in the year 1907.
Many of the paraffine fluxes, analyses of which were given in the
original edition of this work, are no longer available, and there is
no assurance that the supply for any one year can be duplicated
the following one.

Paraffine

Solar Ref.

S O Co

Date shipment received ...

Lima, Ohio
10-28-07

Whiting,
Ind.
10-30-07

Neodesha,
Kan.
11-10-07

100537

100610

100446

g p< gj. Beaum6 at 78 F

18.4

20 5

21 2

gp gj. actual at 78 F .

943

930

926

Flash point degrees F

455

425

435

Volatile 7 hours at 212 F. .

Volatile in 7 hours, 325 F. (dry sample) . .
Residue at 78 F

.2
Crystalline

1.2
Crystalline

.6
Crystalline

Bit insol 88 naphtha pitch . . .

Slow flow
2 7

Quick flow
3 4

Quick flow
2 6

Per cent sol. bit. rem. by H 2 SO4

24 9

24 7

25 9

Paraffine scale

6.4

7.9

8 9

Fixed carbon

2.8

1.9

1.8

134

THE MODERN ASPHALT PAVEMENT.

The petroleum residuum used in the work carried out under the
author's supervision is furnished under the following specifications:

Specifications for Paraffine Flux, 1907-1908. "This oil or flux
shall consist of the heavier or higher boiling portions of any par-
affine petroleum. It shall have a specific gravity of between .942
and .924 (18.6-21.5 B.) at 78 F.

"It shall be free from water and decomposition products, shall
contain not more -than 2% of bitumen insoluble in 88 naphtha,
not more than 10% of material recovered when one gram of the
flux is treated as for the determination of paraffine scale, accord-
ing to the method described in "The Modern Asphalt Pavement/'
John Wiley & Sons, and shall not volatilize more than 5% at
325 F. in 7 hours."

It appears from the preceding table and from our specifications
that a residuum from a paraffine petroleum, such as used in the
asphalt industry at the present day, if it is in the highest degree
desirable, should have a specific gravity of .93, equivalent to a
density of 21.0 B., a flash point of 400 F., should volatilize but
a small amount at 325 F. under certain conditions which are im-
posed, should be practically completely soluble in carbon disul-
phide, and to the extent of 95 per cent in 88 naphtha. The par-
affine scale should be less than 10 per cent, and the amount of
fixed carbon which is obtained on ignition not over 4 per cent.

It will be noted in the preceding analyses that the percentage
of paraffine scale varies, in the samples examined, from 6.4 to
8.9 per cent. In other residuums even more paraffine scale has
been found, as can be seen from the following determinations
which were made some years ago in the author's laboratory.

Manufacturer.

Paraffine.

Craig Oil Co , Milwaukee .

17.6%

Crew Levick Co , Philadelphia

8.7

American Petroleum Product Co., Find-
lay, Ohio

12.3

Scofield, Shurmer & Teagle, Indianapo-
lis I n d

7 1

Standard Oil Co

14.5

Wilburine Oil Co Brooklyn

33 3

Standard Oil Co , thin oil

9 1

PETROLEUMS. 135

The smaller the amount of paraffine scale that is present the
more desirable the flux, since a substance of this nature which
becomes solid at low temperatures cannot be advantageous in a
paving cement. That paraffine residuum is an extremely stable
oil appears from the fact that about 75 per cent of the hydrocar-
bons of which it is composed are not attacked in 88 naphtha
solution by concentrated sulphuric acid, a fact which is confirmed
by exposing such a residuum to water for many years, when no
action is found to have taken place. 1

The characteristics of asphalt cement made with paraffine
residuums of various densities, and the refutation of the claim
that it is not a satisfactory solvent for asphalts and is unsuited
for the purpose for which it is used hi the paving industry, will
be taken up later when asphalt cements are under consideration.
It is merely necessary to state here that of such a residuum as
has been described from 18 to 22 pounds must be used with every
100 pounds of refined Trinidad lake asphalt in order to produce
a cement of proper consistency for paving purposes.

Kansas Petroleum Residuum. Within the last two years
a flux has appeared on the market which is prepared from the oils
coming from the Kansas fields through pipe lines to the East.
This flux is very peculiar in character. It possesses largely the
characteristics of a paraffine, and some of those of an asphaltic
flux. It is peculiar in that, although it carries but 6.7% of par-
affine scale, it leaves a solid, cheesy residuum amounting to 85%
of the original oil on heating at 400 F., while, on being main-
tained at that temperature for a considerable period of time, it
decomposes with the separation of a granular insoluble material,
and with the evolution of gas. On this account it is looked
upon by the author as more or less unsuitable. It has been used
to a very large extent, however, and apparently works satis-
factorily at ordinary temperatures with Trinidad and Bermudez
asphalt. It is to-day one of the largest sources of supply of flux

1 Whipple & Jackson, The Action of Water on Asphalt. Engineering
Record, March 17, 1900, 41.

136 THE MODERN ASPHALT PAVEMENT.

in the paving industry and its value can only be shown by service
tests. In the meantime, it must be classed as a paraffine
residuum.

Its characteristics are shown in the following table:

Manufacturer Standard Oil Company,

Brooklyn, N. Y.

Received from '. Maurer, N. J.

Date shipment received 5-10-08

Test number 2857-M.

Specific gravity Beaume" at 78 F 19.4

Specific gravity Actual at 78 F .937

Flash point, degrees F 485

Volatile in 7 hours at 212 F .2%

Volatile in 7 hours at 325 F. (dry sample) .6%

Residue at 78 F Medium slow flow crys-
talline

Volatile in 7 hours at 400 F. (dry sample) 2.0%

Residue at 78 F Medium flow crystalline

Bitumen insoluble in 88 naphtha, pitch 2.7%

Per cent soluble bitumen removed by H 2 SO 4 23 . 4

Paraffine scale . 2.5

Fixed carbon 3.5

California Asphaltic Petroleum Residuum. The petroleums
of California are characterized by the fact that the residue left on
distillation, if the latter is carried sufficiently far, is a solid bitu-
men resembling asphalt. The oil is said, on this account, to have
an asphaltic base. If the distillation is suspended at a point
where the residue does not solidify on cooling but remains liquid,
like a heavy and dense natural maltha, the material known as
California flux is obtained which has been in use in the paving
industry to a very considerable extent on the Pacific Coast and
to but a small extent elsewhere. Such residuums, as found on
the market in 1907, have the characteristics given in the table on
page 137.

Before discussing the characteristics of these residuums it will

PETROLEUMS.

137

be necessary to consider what takes place in the process of their
manufacture. The petroleum is distilled in cylindrical stills in
the usual manner with some steam agitation, the temperature
being eventually carried up to about 600 F. or higher, and main-
tained there until a sample is slightly hea,vier than water when
poured into it. It is very evident that in this process the petro-
leum is subjected to a very severe treatment and it is not diffi-
cult to determine at the plant where such flux is produced that
very decided cracking goes on. That such cracking takes place
can also be shown by heating a small portion of the original petro-

CALIFORNIA ASPHALTIC PETROLEUM RESIDUUM.

68489

69607

"No. 2"

"G" grade

PHYSICAL PROPERTIES.

Specific gravity, dried at 212 F., 78 F./78 F..
Flashes F N Y State oil-tester .

1.002
354 F.

1.006 1
376 F *

CHEMICAL CHARACTERISTICS.

Dry substance :
Loss 325 F , 7 hours

5.9%

3 2%

Character of residue

smooth

soft

oft

Loss, 400 F., 7 hours (fresh sample)

16.7%

17.3%

Character of residue

smooth

Penetration of residue at 78 F

soft

Bitumen soluble in CSj, air temperature

99 9%

99 7%

Difference

Inorganic or mineral matter

Bitumen insoluble in 88 naphtha, air temper-
ature, pitch

100.0

7.6%

100.0

7.7%

Per cent of soluble bitumen removed by ILjSO^ .
Per cent of total bitumen as saturated hydrocar-
bons

48.3
47 9

54.9
41.9

Fixed carbon

'Extremes 1.018-.993

2 Extremes 430-350.

3 Extremes 5.50-.83.

leum in a glass dish in an oven at not over 400 F., when the lighter
portions all volatilize without cracking and the residue recovered

138 THE MODERN ASPHALT PAVEMENT.

is found to be of the same consistency but much larger in amount
than that obtained by the industrial process. It is further not,
moreover, surprising that cracking should take place very readily
with California petroleums, since it is known that they are com-
posed of the unstable polycyclic polymethylenes of a high degree
of molecular aggregation. 1

In California flux, therefore, we have one which is of a much
greater density than that derived from paraffine petroleum, or
even that derived from Beaumont, Texas, asphaltic oil. It origi-
nates in an unstable petroleum and has been subjected to very
severe treatment, and is, therefore, partially cracked. This is
apparent in some of the determinations given in the preceding
table, where the amount of oils volatile in seven hours at 400 F.
is found to be much larger than would be the case with a stand-
ard paraffine residuum under similar treatment. This loss, about
17 per cent, must be due to the volatilization of light oils produced
by cracking.

Of the components of these California fluxes only 40 to 50 per
cent consist of saturated hydrocarbons as compared to 70 or 80 per
cent found in paraffine residuum. This points with great prob-
ability to the conclusion that such fluxes will harden with age
and exposure much more rapidly than the more stable paraffine
fluxes.

It will be noted that the residue after heating the California
flux to 400 F. is still soft. This is a property which is abso-
lutely essential and differentiates these fluxes from the natural
malthas, which, as has appeared, usually become converted into
hard pitches on heating for any length of time to a high tem-
perature, and it is this property which makes it possible to use
the modern California residuums as a flux.

The fixed carbon which the California residuums yield on
ignition is larger than that found in the paraffine residuum, as
would be expected from the character of the hydrocarbons of
which it is composed, those in the California oil containing a

1 J. Soc. Chem. Ind., 1900, 19, 123.

PETROLEUMS. 139

very considerably larger percentage of carbon than those found in
the eastern residuums.

In drawing specifications for a California asphaltic- flux it
should be provided that it should remain soft after heating for
seven hours at 400 F. The specifications which the author has
proposed for use in work under his directions on the Pacific slope
are as follows:

Specifications for " G " Grade California Flux. " California
flux, known as ' G ' Grade Flux, should be a residue from the
distillation of California petroleum, with steam agitation, at a
temperature not above 620 F.

" It shall have the following characteristics:

" It shall be soluble in carbon disulphide to the extent of 99
per cent and in 88 naphtha to the extent of 90 per cent.

" It shall be free from water, shall not flash below 350 F.
in a New York State oil-tester, and shall have a density of not
less than .98, 12.9 B., or more than 1.050, 9.3 B., at 25 C. when
referred to water at the same temperature.

" It shall volatilize not more than 5 per cent of oil when heated
for seven hours at 325 F., according to the method employed in
the New York Testing Laboratory.

" The residue from heating the oil in the same way to 400 F.
for seven hours shall be a soft flux not hard enough to give a pen-
etration of less than 150 with the Bowen penetration machine.

" It shall not yield more than 6 per cent of fixed carbon on igni-
tion. Under the microscope, beneath a cover-glass, it shall appear
free from insoluble or suspended matter."

One of the most important characteristics of a California flux
to be noted from an industrial point of view is that, owing to
its great density, more than twice as much of it is required to
soften the solid native bitumens as of a paraffme residuum
or of one of the semi-asphaltic nature produced from Texas oil.
For example, with Trinidad asphalt 51 pounds of a California
flux are often necessary to make a cement of normal penetration
where no more than 22 pounds of paraffine residuum are used.

The disadvantages to be met with in the use of a California
flux or the defects in its character have been presented in the

140 THE MODERN ASPHALT PAVEMENT.

preceding paragraphs. Aside from this the flux presents certain
advantages and desirable properties which cannot be equalled
in any other softening agent, and on this account makes it of
great value in certain problems in the paving industry. With
an asphalt such as that from La Patera, California, or the Bejucal
mine in Cuba, a satisfactory paving material could not be made
were it not possible to supply the deficiencies of malthenes in
these hard bitumens by means of those present in a California
flux. The use of the material in this way was well illustrated
in the Alcatraz XX asphalt, which was formerly on the market.
Sixty per cent of La Patera asphalt was mixed with 40 per cent of
dense California residuum and the resulting product was a bitumen
which contained asphaltenes and malthenes in normal propor-
tions, and which, when made with care and uniformity, proved a
desirable material. Great uniformity in its manufacture was not
possible, however, and the defects inherent therein will be consid-
ered when the study of asphalt cements is taken up.

As in the case of paraffine residuums, so with the California
fluxes : in the early days of the industry they were not at all care-
fully prepared, and even to-day many of them are found on the
market which are too badly cracked to be desirable. They can,
with care, however, be prepared with a very considerable degree
of uniformity, as can be seen from the extremes given in the table
on page 137.

As an example of an unsatisfactory flux the following will
serve:

TEST NO. 69012.

Specific gravity, 78 F./78 F 9815

Loss, 212 F 8%

" 325 F., 7 hours 8.0

" 400 F. " " (fresh sample) 16.2

Penetration of 400 F. residue 43

Bitumen insoluble in 88 naphtha, air temp. 9.0%
Fixed carbon 6.0

In this flux the specific gravity is low, with the result that
there is a large loss of volatile matter at 325 F., while the residue
after heating to 400 F. is a solid bitumen, showing that the flux
is not a stable one and therefore undesirable.

PETROLEUMS.

141

Fluxes from Beaumont and Similar Oils. Petroleum of the
character of that found in the well-known Beaumont field in
Texas is usually considered to have an asphaltic base, but, as is
not so well known, it also contains a very considerable propor-
tion of paraffine hydrocarbons, or hydrocarbons which are ex-
tremely stable, as shown by then* behavior with sulphuric acid, 1
the loss on treatment of the crude oil with sulphuric acid being
only 39 per cent as compared with 30 per cent for an Ohio oil
and a still larger amount for the asphaltic petroleums of Cali-
fornia. The result is that the residuum or flux prepared from
Beaumont petroleum possesses some very desirable properties, as
revealed by the following determinations:

BEAUMONT, TEXAS, FLUX.

69330

66364

Oualitv

light

heavy

PHYSICAL PROPERTIES.

Specific gravity, dried at 212 F., 78 F./78 F. .
Flashes F N Y State oil-tester

.9565
395 F

.9735
418 F.

CHEMICAL CHARACTERISTICS.

Dry substance :
Loss 325 F 7 hours

4.3%

smooth

Penetration of residue at 78 F

soft

Loss 400 F 7 hours (fresh sample)

14.5%

6.2%

Character of residue

smooth

soft

Bitumen soluble in CS air temperature

99.8%

99.6%

Difference .

Bitumen insoluble in 88 naphtha, air temper-
ature, pitch

100.0
2.5%

100.0

4.8%

Per cent of soluble bitumen removed by H.jSO 4 . .
Per cent of total bitumen as saturated hydrocar-
bons

25.4
72.8

20.9
79.4

1.0

1.7

3.0

3.5

1 J. Soc. Chem. Ind., 1901, 20, 690.

142 THE MODERN ASPHALT PAVEMENT.

When the preceding results are compared with those which
have been given as representing the character of the paraffine
residuums and of California fluxes, it will be seen that the density
of these oils is much higher than that of the true paraffine residuums,
but much lower than that of the California fluxes, and that the
percentage of total bitumen which is present as saturated hydro-
carbons is between 70 and 80 per cent as compared to 40 and
50 per cent in the latter form of flux. This must be a very desir-
able property and one which is due probably to the presence
of an appreciable amount of paraffine hydrocarbons and of a
large proportion of stable polymethylenes. The investigations
of Mabery and the author have shown that the polymethylenes
belong to the C n H 2n , C n H 2n -2, and C n H 2n _4 series. That par-
affine hydrocarbons are present, and that some of them are of
high molecular weight, is revealed by the fact that the residuum
contains 1 per cent of paraffine scale. As prepared for use as a
flux the residuum is much denser than ordinary paraffine flux,
its specific gravity being .95 to .96 as compared with .93 for the
latter. In other respects, when very carefully prepared, it is not
essentially different in its physical properties. It sometimes con-
tains a slightly larger proportion of light oils, volatile at 325 F.,
but the same proportions of the lighter flux and hard asphalt
are necessary as in the case of paraffine residuum. The denser
form, with a gravity of .97, is only used with asphalts that are
deficient in malthenes, such as Trinidad land asphalt.

The above conclusions only hold true when the residuum is
carefully prepared and some is found on the market which,
like the less carefully distilled paraffine residuum of the earlier
years of the industry, is not satisfactory. If carefully prepared,
however, it is without doubt the most desirable flux which is
available to-day for the purpose for which it is used.

Specifications for this residuum may read as follows:

"This oil or flux shall consist of the heavier or higher boil-
ing hydrocarbons of Texas oil containing not more than 3 per
cent of paraffine scale, as determined according to the method
described in "The Modern Asphalt Pavement," John Wiley
& -Sons. It shall be of a character suitable for fluxing

PETROLEUMS. 143

grahamite and other hard bitumens in such a way that, when
forty (40) parts of grahamite and sixty (60) parts of the flux are
melted together at 400 F., the resulting product shall not, after
24 hours in an ice box, contain free oil or have a granular or short
structure, due to the presence of paraffine. It shall have a gravity
not less than .97 at 78 F., shall not volatilize more than five per
cent when heated at 400 F. for seven (7) hours, and shall not
flash below 400 F. in a closed oil tester, New York State pattern.

"It shall be free from water and decomposition products, con-
tain not more than 3% of bitumen insoluble in 88 naphtha,
and shall not volatilize more than 5 per cent when heated for
seven (7) hours at a temperature of 325 F."

As has already been said, the supply of flux of this description
is nearly exhausted at the present time, owing to the decline of
the Beaumont field and the mixing of petroleums from various
fields in pipe lines, many of which contain a large proportion of
paraffine oil.

Other Fluxes. While the fluxes which have been previously
described are those which are actually hi use hi the industry in
the United States, others are found on the Continent of Europe
which, although not available for use in this country, would be
entirely satisfactory if this were the case. Among these may
be mentioned residuum from the distillation of Russian petroleum.
This is free from paraffine scale and consists of very stable hydro-
carbons, and with it an especially desirable asphalt cement could
be made, if it were reduced to a proper density and uniformity.

In France residuum from the distillation of shale oils is avail-
able and is largely used in the fluxing of asphalts for use hi mastic.
Such an oil has the properties shown in table on page 144.

It will be noticed that this oil consists very largely of unstable
hydrocarbons, as would be expected in one which is the product
of the distillation of shales, a process in which cracking must go
on to a certain extent, and contains a very considerable quantity
of paraffine scale.

Wax Tailings. Wax tailings, or still wax, is a thick, yellow-
brown buttery product at ordinary temperature, which melts
to a thin liquid at about 175 F. It is the product of the destruc-

144

THE MODERN ASPHALT PAVEMENT.

SHALE-OIL RESIDUUM FROM FRANCE. GOUDRON DE
SCHISTE D'AUTUN.

65402

72630

PHYSICAL PROPERTIES

Specific gravity, dried at 212 F., 78 F./78 F. .
Flashes F N Y State oil-tester

.9849
215 F

.9894
355 F.

CHEMICAL CHARACTERISTICS.

Dry substance :
Loss, 325 F , 7 hours

7.6%

5.0%

Character of residue . .

soft

Loss 400 F 7 hours (fresh sample)

26.4%

10.0%

Character of residue

soft

Bitumen soluble in CS air temperature

99 7%

99 8%

Difference

Inorganic or mineral matter .

trace

Bitumen insoluble in 88 naphtha, air temper-
ature pitch

100.0

6 4%

100.0

7 3%

Per cent of soluble bitumen removed by H 2 SO 4 . .
Per cent of total bitumen as saturated hydro-
carbons

53.5
43.6

56.2
43.5

Per cent of solid paraffine

4 4

5 2

Fixed carbon

3.0

5.0

live distillation of paraffine petroleum residuum, the residue in
the still being coked.

Owing to the method of preparation the product found on
the market is extremely variable in character. The results of the
examination of several lots in the author's laboratory are given
in the table on p. 145.

It appears from the data given that, as has been said,
wax tailings are extremely variable in character. They often
carry a large amount of water and vary in specific gravity and
flash point. Usually they contain but little volatile oil, although
on heating at 400 F. they become converted, as a rule, into a
solid substance of various degrees of consistency. Wax tailings
are almost absolutely pure bitumen, and it is surprising to find
that, although they are the result of destructive distillation, they

PETROLEUMS.
WAX TAILINGS.

145

Test number .

30245

55285

64439

64587

64916

Original material:
Loss, 212 F., until dry. .
* ' 325 F , 7 hours. . .

7.0%

le 2%i

2.8% l

15.5%'

Residue pen. at 78 F . .

Dry material:
Specific gravity 78 F./
78 F

1 02

58
1 0794

1 1445

1 0994

Flashes, F

298 F.

240 F.

340 F

410 F

Loss, 325 F., 7 hours
Residue pen at 78 F. . .

5.9%
83

3.0%
58

2.1%
soft

Loss, 400 F. , 7 hours (fresh
.sample).
"-Residue. pen. at 78 F.

13.3%
brittle

4.8%
9

4-6%
50

B ; aimen soluble in CS,. .-. . .
^iflference

99. 9% 2

99.8%
2

ID organic or mineral matter.

Bitumen sol. in 88 naphtha

83 1%

97 2%

100.0
96 7%

100.0
98 1<&

This is per cent of total bitu-
men

96 9

98 4

Bitumen sol in 62 naphtha

98 9%

99 6%

This is per cent of total bitu-
men

99 1

99 7

88 naphtha sol. bitumen :
Per cent not removed by
H 2 SO 4 .

49 3%

50 0%

55 0%

Unacted on by H^O^SOa. . .
Paraffine scale

3.7%

0.4

1.1

2.3

Fixed carbon

5.5

4 1

3 4

Bitumen insoluble in cold
carbon tetrachloride

0.0

Contains a large amount of watr.

2 Unacted on by HaSO,, 79.9%.

contain 50 per cent of saturated hydrocarbons unacted upon by
strong sulphuric acid, although fuming sulphuric acid destroys
them entirely. Although produced from a residuum carrying
considerable paraffine scale, mere traces of this material are found
in the tailings. The amount of fixed carbon which they yield

146 THE MODERN ASPHALT PAVEMENT.

is low, as would be expected in the case of a substance derived
from paraffine petroleum. Wax tailings, owing to the lack of
uniformity in the material, are of no interest in the paving indus-
try, but are used to a very considerable extent in insulating and
other bituminous compounds.

SUMMARY.

From the preceding data it appears that there are three
classes of fluxes available commercially in the United States for
the softening of native solid bitumens in the preparation of
asphalt cement for paving purposes: paraffine residuums, as-
phaltic residuums of California, and the semi-asphaltic residuums
of Texas, and from the Oklahoma and Kansas fields, the Texas
being the most desirable of all and probably sufficiently superior
to the good paraffine residuums to justify an additional expendi-
ture for its use in work of the highest grade, the reasons for which
have appeared in the preceding pages.

The standard residuums from paraffine oil will, however, con-
tinue to be used over a large area of country where they can be
obtained at considerably lower prices than other fluxes and where
the conditions to be met are such that they are entirely satisfac-
tory.

CHAPTER IX.
THE SOLID BITUMENS.

WITH the solid bitumens, consisting largely of paraffine hydro-
carbons, such as ozocerite, hatchettite, etc., the paving industry
has nothing to do, nor is it interested in the terpenes, fossil resins,
amber, etc., which are composed largely of unsaturated cyclic
compounds. Our attention must be at once turned, therefore,
to the solid bitumens which are of commercial importance in the
paving industry, especially the asphalts.

The Asphalts. The asphalts industrially include all the solid
native bitumens which are in use in the paving and other industries.
Specifically, true asphalt is sharply differentiated from several of
the bitumens which are used under this designation, such as
gilsonite and grahamite. In the table which follows the physical
properties and proximate composition of the more prominent
asphalts, using the term in its industrial sense, which are or have
been used in the paving industry, are given.

Until recently but little has been known of the nature of
asphalt beyond the fact that it is a native bitumen. Boussingault's
investigation of the viscid bitumen and asphalt of Pechelbron, so
often quoted, threw no light on the question from the point of view
of modern chemistry, as he merely separated the material into two
portions, one more volatile than the other, and both, without
doubt, more or less decomposed by the heat to which they were
subjected, and consisting of mixtures of various hydrocarbons and
their derivatives. Warren, who revised the subject of the hydro-
carbons for Dana's Mineralogy, states that the following " classes
of ingredients " are present in asphalt (see page 150) :

147

148

THE MODERN ASPHALT PAVEMENT.

PHYSICAL PROPERTIES AND PROXIMATE

Test number

63260

36721

liituinen

Trinidad lake

Trinidad land

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F. Original sub-
stance, dry .

refined.
1 40

refined.
1 4196

Blue black

Brown black

Lustre

Dull

Structure ;

Homogeneous

Fracture

Semi-

Hardness original substance

conchoidal
2

Odor

Asphaltic

Softens

180 F.

188 F

Flows

190 F

198 F

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance:
Loss, 325 F , 7 hours

1 1%

1 0%

Character of residue

Smooth

Blistered

Loss 400 F 7 hours (fresh sample) .

4 0%

3 0%

Character of residue .

Blistered

Bitumen soluble in CS 2 , air temperature

56 53%

54 1%

Inorganic or mineral matter

36 50

Difference . . ...

6 97

7 9

Malthenes :
Bitumen soluble in 88 naphtha, air tem-
perature

100.0
35 6%

100.0
33 5%

This is per cent of total bitumen
Per cent of soluble bitumen removed by
H 2 SO 4 .

63.1
61 3

61.9
64 8

Per cent of total bitumen as saturated hy-
drocarbons

24 4

21 8

Bitumen soluble in 62 naphtha

41.7%

38 2%

This is per cent of total bitumen

73.9

70.6

Carbenes:
Bitumen insoluble in carbon tetrachloride,
air temperature

Bitumen more soluble in carbon tetra-
chloride, air temperature

1 3%

Bitumen yields on ignition:
Fixed carbon

108%

12.9%

Sulphur

0.2%

5.0%

1 These bitumens are not strictly asphalts, as appears in the text, but may

THE SOLID BITUMENS.
COMPOSITION OF THE MORE IMPORTANT ASPHALTS.

149

44412
Bermudez
refined, 1900

67753
Bermudez
refined, 1903

22220
Cuban,
Bejucal. 1

13541

Califorina,
La Patera.

13601
California,
Standard. 1

66923
Mara-
caibo.

1.0823
Black
Bright
Uniform
Semi-
conchoidal
Soft
Asphaltic
170 F
180 F.
22

1.0575
Black
Bright
Uniform
Semi-
conchoidal
Soft
Asphaltic
160 F.
170 F.
26

1.305
Red-brown
Dull
Compact
Semi-
conchoidal
2
Asphaltic
230 F.
240 F.

1.3808
Black
Dull
Uniform
Irregular

Asphaltic
260 F.
300 F.

1.0627
Black
Dull
Uniform
Semi-
conchoidal
Soft
Asphaltic
170 F.
180 F.
0-27

1.0638
Black
Bright
Uniform
Semi-
conchoidal
Soft
Asphaltic
200 F.
210 F.
20

3.0%
Smooth

4.4%
Smooth

.88%
Cracked

1-5%
Shrunken

6.6%
Smooth

2,7%
Blistered

8.2%
Wrinkled

9-5%
Shrunken

1-5%
Wrinkled

2.5%
Shrunken

19.9%
Blistered

4.7%
Much blist.

95.0%
2.5
2.5

96.0%
2.0
2.0

75.1%
21.4
3.5

49.3%
48.6
2.1

89.8%
6.8
3.4

96.8%
1.8
1.4

100.0

62.2%
65.4

69.1%
71.9

32.4%
43.1

21.6%
43.8

53.4%
59.4

45.7%
47.2

62.4

67.3

60.5

81.4

51.9

46.4

24.4

23.4

17.0

8.1

28.6

25.3

69.2%

72.8

0.1%

75.9%
79.0

1.1%

39.6%
52.7

%:l %

60.0%
66.8

0.3%

51.5%
53.2

17 5%

1 6%

1 7%

13.4%
4.0%

14.0%

25.0%

8-3%

14.9%
6.2%

8.0%

18.0%

be considered as such in their relation to the asphalt paving industry.

150 THE MODERN ASPHALT PAVEMENT.

WARREN'S CHARACTERIZATION OF ASPHALTUM.*

"A. Oils vaporizable at or about 100 or below; sparingly
present if at all.

"B. Heavy oils, probably of the Pittoliumor Petrolene groups;
vaporizable between 100 and 250, constituting some-
times 85 per cent of the mass.

"C. Resins soluble in alcohol.

"D. Solid asphalt-like substance or substances, soluble in ether
and not in alcohol; black, pitch-like, lustrous in fracture;
15 to 85 per cent.

"E. Black or brownish substance or substances, not soluble in
either alcohol or ether; similar to D in color and appear-
ance (Kersten); brown and ulmin-like (Volckel); 1 to
75 per cent.

"F. Nitrogenous substances; often as much as corresponds to
1 or 2 per cent of nitrogen."

He also defines asphalt as " a mixture of different hydrocar-
bons parts of which are oxygenated."

In the light of our present information neither the classifica-
tion of the proximate constituents nor the definition of asphalt
is satisfactory.

A. Hard asphalts and but few malthas seldom contain any
oils vaporizable at 100. Such hydrocarbons cannot be present
in any amount in a hard asphalt, as the material would then have
the properties of a maltha.

B. Heavy oils are undoubtedly the chief constituents of
asphalt and as such require the most careful study and differ-
entiation to determine their nature. Recent investigations have
shown that these oils are a mixture of saturated and unsaturated
hydrocarbons of di- and polycyclic series and of their sulphur
and nitrogen derivatives, the sulphur derivatives being particu-
larly characteristic of asphalt. This class of oils is that which
we name to-day malthenes.

C. Resins are not present in asphalt. The oils are slightly
soluble in alcohol, but the soluble portion is not similar to resin
in its behavior towards reagents.

1 Descriptive Mineralogy, Dana, th Edition, 1896, 1017.

THE SOLID BITUMENS. 151

D. The substances soluble in ether are soft and resemble mal-
thas. They are the same substances mentioned under B. They
usually amount to from 60 to 75 per cent of the asphalt.

E. The bitumen not soluble in ether (petroleum naphtha is
now commonly used instead of ether as a solvent) is a hard sub-
stance which does not melt without decomposition but is soluble
in the malthenes, the predominating constituent of asphalt, or
in heavy asphaltic oils. Together with the malthenes it con-
stitutes the pure bitumen of asphalts. It contains the larger
part of the sulphur compounds present and seems to owe its hard-
ness to this fact. It is known as asphaltenes.

F. Nitrogenous substances are found both in the malthenes
and in the hard bitumens of class E. In the malthenes the nitro-
gen derivatives have been identified as hydrocyclic bases.

As has already been said, there is a decided difference in the
use of the term asphalt in an industrial and specific sense.

From investigations carried on in the author's laboratory
within the last seven years, the results of which have been largely
published elsewhere, 1 it is possible to characterize true asphalt more
closely and to differentiate it specifically from other solid bitumens.

Asphalt may be defined, industrially, as a mineral pitch,
found in nature in a more or less solid state, melting at a tempera-
ture in the neighborhood of that of boiling water, and miscible in all
proportions with heavy petroleum oils or fluxes to form a viscous
cementing material which is in use in the paving and in other
industries.

More specifically asphalt is a bitumen found in nature, orig-
inating in certain types of petroleum, of various degrees of
consistency, but generally solid, softening at a temperature in
the neighborhood of 100 degrees Centigrade, and consisting of a
mixture of various saturated and unsaturated hydrocarbons and
their derivatives, a part of which are soluble in naphtha, very
viscous liquids, largely saturated hydrocarbons unattacked by
strong sulphuric acid in such a solvent; while those insoluble in
naphtha are soluble in carbon disulphide and carbon tetrachloride,

1 On the Nature and Origin of Asphalt, Long Island City, N. Y., 1898.

152 THE MODERN ASPHALT PAVEMENT.

brittle solids which do not melt by themselves without decom-
position on the application of heat.

The naphtha soluble bitumen is known, conventionally, as a
class, as "Malthenes," while the bitumen insoluble in naphtha, but
soluble in carbon disulphide, has been called "Asphaltenes." The
malthenes consist in part, depending upon the hardness of the bitu-
men, of from 20 to 50 per cent of saturated di- or polycyclic poly-
rnethylenes of the series of C n H 2n _ 2 ; C n H 2n _ 4 , C n H 2n _ 8 , etc., the
lowest member found in Trinidad asphalt being C^H^ boiling at
165 C. under a pressure of 30 mm.; in part of unsaturated hydro-
carbons, which can be separated from the polymethylenes by
strong sulphuric acid with which they combine readily, the nature
of which is not so well understood; of sulphur compounds separated
in the same way which, on isolation, are found to be the same as
those occurring in Ohio, Canadian, and California petroleum and
nitrogen derivatives, perhaps bearing the same relation to the poly-
methylenes that chinolin does to benzol. The structure of these sat-
urated hydrocarbons is given on page 105. The asphaltenes
are probably unsaturated hydrocarbons or their derivatives, as
they are all strongly acted upon by concentrated sulphuric acid.
The molecules of which they consist must be highly condensed and
have a very high molecular weight. Of their structure we know
nothing. The asphaltenes contain the greater part of the sul-
phur present in asphalts, and they are, as a rule, characterized
by the presence of very considerable amounts of it, the larger
the amount the harder the asphalt. 1 Normal asphalts yield about
from 10 to 15 per cent of fixed carbon on ignition, a fact which
enables us to differentiate them by this characteristic alone from
many of the other solid bitumens.

Differentiation of the Asphalts among Themselves. Considered
as pure bitumens, the asphalts vary quite largely in character among
themselves, depending upon the nature of the bitumen from which
they have been derived, the extent to which they have been met-
amorphosed by their environment, and the resulting difference
in the relative proportions of malthenes and asphaltenes, and

1 On the Nature and Origin of Asphalt, Long Island City, N. Y., 1898.

THE SOLID BITUMENS. 153

of saturated and unsaturated hydrocarbons which are present.
Those asphalts which have undergone the most complete meta-
morphism contain the largest portion of the asphaltenes and are
the harder. A high percentage of sulphur also works in the same
direction. Some actual variations in the amount of the total
bitumen soluble in cold 88 naphtha in asphalts of different degrees
of hardness are as follows:

No. 22220. Hard asphalt, Bejucal, Cuba 43.1%

" 13541. " " La Patera, Cal 43.8

Medium hard, Trinidad 65.0

Softer asphalt, Bermudez 69:0

The following differences have been noted in the amount of
sulphur found in the pure bitumen from the same localities:

No. 13541. La Patera, Cal. . 6.20%

" 22220. Cuban, Bejucal 8.28

" 63260. Trinidad Lake 6.16

" 44412. Venezuela, Bermudez 3 .93

The asphalts can also be differentiated by their physical prop-
erties, such as consistency, which is particularly valuable, by
their melting-point, color, and specific gravity, though in the form
of pure bitumen they are not very variable in the latter respect,
and by other minor differences.

Asphalt Associated with Mineral Matter. In the preceding
paragraphs, asphalt has been considered as a more or less pure
bitumen, but it is more often than not associated with mineral
matter of one kind or another. This fact has resulted in a classi-
fication of this material on this ground alone, according to the
nature of the mineral matter. Such a classification may be of
some industrial value, although having nothing to do with the
character of the asphalt itself. At best it is artificial. The fol-
lowing, based on the writer's investigation of a large number of
asphalts from all over the world, is suggested:

Asphalt. 1. Impregnating compact, amorphous limestones
to the extent of less than 16 per cent.

2. Impregnating limestones, partially crystalline and mixed
with silica or silicates, to the same extent.

154 THE MODERN ASPHALT PAVEMENT.

3. Impregnating fossiliferous limestones to the same extent.

4. Impregnating shales or schists.

5. Impregnating hard sandstones.

6. Mixed with silica and clay to a fixed and uniform extent
at the source where the asphalt originates.

7. Mixed with sands by percolation into beds of the latter in
place.

8. Mixed with soil and organic matter of vegetable origin
where the effusion of tar springs have hardened on exposure.

Type 1 is found on the continent of Europe, in Sicily, Val de
Travers, Seyssel, and hi the United States but to a limited extent
in Utah.

Type 2 is found in Oklahoma, near Ravia and elsewhere.

Type 3 occurs near Dougherty in Oklahoma, the rock being
a member of the lower coal measures, according to Eldridge, and
in the neighboring Buckhorn District.

Type 4 is found in Ventura County, California.

Type 5 is found in the same locality.

Type 6 is unique, Trinidad Lake Asphalt.

Type 7 includes large deposits of sand in Kentucky and Cali-
fornia, which are saturated, in situ, with a rather soft asphalt.

Type 8 is found wherever exudations of bitumen have spread
over the soil and become hardened by exposure. They are com-
mon in Kern County, California.

It appears, therefore, that the character of the mineral matter,
with which an asphalt is mixed naturally, must be considered
as well as the nature of the bitumen itself, in forming an opinion
of its availability for paving purposes.

With the information which has been presented in the preced-
ing pages it is now possible to consider individually each of the
native bitumens, known generically or specifically as asphalts,
and afterwards to make a comparative study of their availability
in the paving industry.

THE SOLID BITUMENS. 155

SUMMARY.

Asphalt is a term which may be used either industrially 01
specifically; industrially to cover all the solid native bitumens
used in the paving industry and specifically to include only such
as melt on the application of heat, at about the temperature of
boiling water, are equally soluble in carbon disulphide and car-
bon tetrachloride and to a large extent in 88 naphtha, those
hydrocarbons soluble in naphtha consisting to a very consider-
able degree of saturated hydrocarbons, yielding about 15 per cent
of fixed carbon and containing a high percentage of sulphur.
Under this definition it can be seen that grahamite is not an asphalt,
since it is not largely soluble in naphtha and yields a very high
percentage of fixed carbon on ignition. It is also less soluble in
carbon tetrachloride than in carbon disulphide. Gilsonite is not
an asphalt, since the saturated hydrocarbons contained in the
naphtha solution are very small in amount and quite different
in character from those found in asphalt.

CHAPTER X.
INDIVIDUAL ASPHALTS.

Trinidad Lake Asphalt. In considering the asphalts indi-
vidually, it will be best to examine in detail the characteristics
of the one in regard to which the most is known and with which
the most successful pavements have been laid and then, to com-
pare others with it. Trinidad lake asphalt is, of course, the
one referred to. The location of the deposit and the manner of
its occurrence may be summarized as follows: 1

" The island of Trinidad lies off the north coast of South
America, between 10 and 11 of latitude and 61 and 62 of longi-
tude (Fig. 4). It is bounded on the north by the Caribbean Sea,
on the east by the Atlantic, on the south by a narrow channel,
into which flow the waters of the northern and most westerly
mouths of the Orinoco, and on the west by the Gulf of Paria,
the two latter bodies of water separating it from the mainland
of Venezuela. . . .

" While there are deposits of pitch scattered all over the island,
the only ones of commercial importance are those situated on
La Brea Point, in the wards of La Brea and Guapo, in the county
of St. Patrick, on the western shore of the island. They are
about 28 miles in an air line from Port of Spain, the seat of govern-
ment, the chief harbor and only port of entrance, and lie on the
north shore of the southwestern peninsula, the point upon which
they are situated being apparently preserved from destruction
by the sea, which is elsewhere rapidly wearing away the coast
by the bituminous deposits which exist along the shore and even

1 Report of the Inspector of Asphalt and Cements, Eng. Dept., District
of Columbia, 1892. On the Nature and Origin of Asphalt, Long Island City,
N. Y., 1898.

156

INDIVIDUAL ASPHALTS.

157

some distance from it, and which from their toughness resist the
action of the waves better than the soft rocks of this region. The
pitch deposits are found scattered over the point, but can be divided
conveniently into two classes, according to their source.

" The main deposit is a body of pitch known as the Pitch
Lake, situated at the highest part of the point.

" Between this and the sea, and more especially toward La
Brea, are other deposits, covered more or less and mixed with soil.

FIG. 4.

" The pitch from these sources is classed as ' lake pitch ' and
' land pitch/

" By far the largest amount of pitch is found hi the pitch lake,
originally nearly a circular area of 127 acres, the surface of which,
in 1894, was 138.5 feet above sea level. From the lake the ground
falls away on all sides, except, perhaps, a slight ridge to the east
and southeast. In fact it seems plain that this deposit lies in
the crater of a large mud volcano which has filled up with pitch."

It appeared, when first examined by the author in 1891, "as
a flat, gently sloping mound, wooded over a large portion, open
savanna elsewhere, and toward the northeast merely grassed over.

158 THE MODERN ASPHALT PAVEMENT.

" On the west its slopes toward the sea are gentle for some
distance,, but then more abrupt. On the north, toward La Brea
Point, the reverse is the case, and a ravine runs, with a small
stream, quite to the village, this slope being very scantily covered
by a growth of coarse grass near the lake, becoming more bushy
farther on, while the other slopes are well wooded, with magni-
ficient palms near the lake, forming a beautiful band or border
around it, within which is a grassy zone of about 100 to 200 feet
or more in width."

Since then the removal of large quantities of pitch has quite
changed the surroundings, owing to the cutting off of the wood
and other alterations resulting from the operations incident to
the exportation of asphalt.

In 1893 a series of borings were made upon the lake by the
New Trinidad Lake Asphalt Company.

" A boring at the center of the lake was carried to a depth
of 135 feet, the entire distance being through pitch, which, as
far as ocular evidence goes, has the same character as that at
the surface. It was impossible to carry the boring deeper, as
the movement of the pitch had so inclined the tube one foot
in six which formed the lining, that it had to be abandoned.
It then gradually toppled over and was engulfed. Nothing has
been seen of it since. The result was sufficient, however, to show
the great depth of the crater and the uniformity of the pitch.
The depth attained was within a few feet not more than three
and a half of sea level, and yet we do not know how much deeper
the pitch may extend. The borings on the north side of the lake
about 1000 feet from the centre, and 100 feet from the edge, was
in pitch of the usual character for 75 feet, showing a very steep
slope to the sides of the crater. At 80 feet a layer of fine white
sand was met for a few feet, and then asphalt was again encount-
ered. At 90 feet sand mixed with asphalt was struck, and this
continued to a depth of 150 feet.

" Further borings, made at some distance from the lake, gave
results near the surface which were similar to those found at the
deeper levels at the edge of the lake. Sand, mixed with asphalt
here and there, was the common material, while at a depth of

INDIVIDUAL ASPHALTS. 159

80 feet on the southern side of the lake, and about 80 feet south
of the road, and between 1200 and 1300 fee,t from the centre of
the lake, a very hard asphaltic sandstone was found.

" All the evidence thus goes to show that the sides of the crater
are of sand or sandstone, more or less impregnated with bitumen,
the sandstone being no doubt the rock of the hillside toward the
south, against which the crater has been built up.

" From the borings it was thus learned for the first time how
enormous the deposit was, and the idea that the mound was really
a crater seemed to be confirmed. It is, nevertheless, hard to
realize that there is at this point, 138 feet above the sea, a bowl-
like depression over 2300 feet across, and over 135 feet deep,
reaching below the sea level, and filled with a uniform mass of
pitch, which must amount to over 9,000,000 tons."

Crude Trinidad asphalt is an extremely uniform mixture
or emulsion of gas, water, fine sand, clay and bitumen which are
found in the following proportions:

Bitumen 39 .3%

Mineral matter 27 .2

Water of hydration of clay 3.3

Water and gas, loss at 325 F 29 .0

98.8%

The material is of the greatest uniformity in composition, as
it has been found that specimens collected at intervals of 100 feet
and at a depth of 1 foot over the surface of the deposit and to a
depth of 135 feet at the centre all have the same composition as
appears from the following table.

The amount of uncombined water does not appear among the
constituents in 'these determinations as, in order to avoid any
possibility of change, it was removed by air-drying at the lake
in the course of the 'preparation of the samples for transportation
to the United States. Frequent examinations of the crude asphalt
in a fresh condition have shown, however, the water to be as con-
stant in amount as the other constituents.

In this deposit there are, therefore, many millions of tons of
asphalt of a highly uniform composition, not only as far as the

160

THE MODERN ASPHALT PAVEMENT.

AVERAGE COMPOSITION OF TRINIDAD LAKE ASPHALT IN

CIRCLES.

Bitumen

by CS 2 .

Mineral
Matter
on
Ignition.

Differ-
ence
Undeter-
mined. *

Soluble
in

Naptha.i

Total
Bitumen
thus
Soluble.

Circle 2 : 200 feet from centre. .
4: 400 ' ' . .
6: 600 ' ' . .
" .8: 800 ' ' . .
" 10: 1000 ' ' . .
" 12: 1100 ' ' . .
General Average

55.02%
54.99
54.84
54.66
54.78
54.62
54.92

35.41%
35.40
35.49
35.56
35.44
35.45
35.46

9.57%
9.61
9.67
9.78
9.78
9.93
9.72

31.83%
31.63
31.85
31.67
31.58
31.77
31.72

57.85%
57.55
58.26
57.97
57.64
57.51
57 79

Circle 14 : 1400 feet from centre. .

53.86

36.38

9.76

30.52

56.66

AVERAGE COMPOSITION OF TRINIDAD LAKE PITCH FROM
THE BORING.

From surface to 135 feet in depth.

54.66%

35.90%

9.44%

31.53%

57.67%

> This naphtha possessed less solvent power than usual.

* Water of clay, not lost on refining, bitumen held by clay and inorganic salts, volatil-
ized on ignition.

proportions of bitumen and mineral matter are concerned, but
also in the relation of the malthenes to the asphaltenes.

Constitution of Crude Trinidad Asphalt. The constitution of
the material forming the great deposit found in the Pitch Lake
in the Island of Trinidad has only been explained very recently.
In the course of the ordinary routine analysis of the substance
from which the water has been removed, it is found that the
summation of the percentages of bitumen soluble in cold carbon
disulphide and of the mineral matter obtained by ignition does
not amount to 100, as is seen in the above table. It has been the
custom for many years to call the difference between this sum
and 100 "organic matter not bitumen." Recent investigations
of the author l have shown that this is an entirely incorrect state-
ment, as the asphalt contains practically no organic matter not

J The Proximate Composition and Physical Structure of Trinidad Asphalt,
Proc. Am. Soc. Test. Materials, 1906, 6, 509.

INDIVIDUAL ASPHALT. 161

bitumen, and that the undetermined matter or. difference can
be readily explained as follows:

Water and Gas. If a weighed amount of crude Trinidad
asphalt, immediately after being taken from the deposit, is ground
to a fine powder and exposed to the air for twenty-four hours, it
will lose about 29 per cent of water and a small additional amount
on further exposure in vacuo over sulphuric acid, this latter
small amount being very probably water of crystallization of the
salts which are present. The total loss averages 29 per cent,
with but very small variation, in the fresh pitch taken several
inches below the surface, although on exposure to the air for
some time a very considerable proportion of this amount may be
lost, as in the case of storage of the crude material.

At the same time with the water the gas which it contains
in solution, consisting of a mixture of carbon dioxide and hydrogen
sulphide, is lost. Its percentage must be extremely small by weight.

Having determined the average amount of water in the fresh
pitch with some degree of accuracy it will be convenient to con-
duct the further investigation of the pitch on the dried or refined
material.

Mineral Matter. The residue of mineral matter obtained
on the ignition of refined Trinidad asphalt in a muffle averages,
as the result of fifteen determinations of representative samples,
36.5 per cent corresponding to 25.9 per cent in the original crude
material, with extremes of 36 and 36.8 per cent. As this igni-
tion has taken place at a temperature approaching 800 degrees
Centigrade, there is every reason to believe that considerable
inorganic matter has been volatilized, as is often found to be the
case in the preparation of the ash of vegetable material. In the
latter case the addition of tricalcium phosphate prevents such loss
and it seemed that this mode of procedure might be applicable
to the asphalt. When the latter was ignited in the presence of
the phosphate a residue greater by 2 per cent was obtained, 38.5
per cent, and one which more correctly represents the percent-
age of anhydrous inorganic matter present in the asphalt, showing
that 2 per cent is volatilized in the ordinary course of ignition,
thus accounting for that amount of the ' 'undetermined matter."

162

THE MODERN ASPHALT PAVEMENT.

The mineral matter on examination is found by elutriation to
consist to a large extent, 30 to 40 per cent, of clay and, of course,
its water of hydration, which may be properly regarded as a part
of the mineral matter originally present, is lost on ignition. If
the residue after the extraction of the asphalt with solvents is
heated to a temperature of 340 degrees Centigrade, it loses from

4 to 6 per cent, the determination at such a temperature
of course not being one capable of being made with great accuracy.
A certain amount of this loss is plainly due to the volatilization
or destruction of bitumen held by the clay present, as shown by
its condensation on the side of a glass tube in which the heating
is conducted, but as this amount, as will appear later, cannot
exceed 1 per cent, the loss due to the presence of water in the clay
must reach at least 4.5 per cent. The total loss on ignition of
bitumen and water in the clay must, however, amount to about

5 per cent and be included in the "undetermined matter." When
this per cent is added to the 2 per cent of inorganic matter vol-
atilized at 800 degrees Centigrade, the entire percentage of
undetermined matter obtained by the usual method of analysis
is accounted for, so that it appears that Trinidad asphalt really
contains practically no organic matter not bitumen.

In order *to determine whether this assumption is correct,
analyses have been made of mixtures of an extremely pure bitumen,
gilsonite, with clay and other fine inorganic materials, the mixture
being prepared by melting the *bitumen, stirring being kept
up during cooling, and the cold brittle mixture being then ground
to a fine powder in order to obtain a uniform mixture for
analysis. The results of an examination of such mixtures by
the routine methods are as follows:

Bitumen sol-
uble in Car-
bon Disul-
phide.

Mineral
Residue on
Ignition.

Undeter-
mined.

62 3%

28 8%

8 9%

' ' " ignited

59.7

34.1

6 2

72.3

22.9

4 8

' ' ' ' ignited

77.9

17.9

4 2

Portland cement ignited

82 6

15 8

1 6

Trinidad mineral residue, ignited

64.4

34.3

1.3

INDIVIDUAL ASPHALTS.

163

It appears from the preceding results that all the artifical
mixtures of bitumen and fine mineral matter apparently contain,
as the result of the analyses, a certain amount of undetermined
matter or organic matter not bitumen, although from the nature
of these mixtures nothing of this kind can be present. The amount
varies from 8.9 per cent in the case of the unignited Fullers earth,
to 1.3 per cent for the ignited mineral residue of Trinidad asphalt.
The larger percentage is due to the presence of water of hydration
in the Fullers earth, while the smaller percentage shows that the
mineral matter in Trinidad refined asphalt is capable of retaining
1.3 per cent of bitumen, either by absorption, adsorption, or both.
Our assumptions in regard to the true composition of Trinidad
asphalt are, therefore, correct and its actual composition should
be stated as follows:

Crude Trinidad
Asphalt.

Refined
Trinidad
Asphalt.

Water and gas ... .

29 0%

Bitumen soluble in cold carbon disulphide .

56 5%

Bitumen retained by mineral matter
Mineral matter, on ignition with tricalcium
phosphate . . .

.3
27 2

38 5

Water of hydration, clay, and silicates

3 3

4 2

98.8%

99.5%

Refined Trinidad Lake Asphalt. Refined Trinidad lake asphalt,
which is dried with steam and agitated with steam, has the charac-
teristics given in the table on pages 164 and 165.

Refined Trinidad Lake asphalt is of a homogeneous structure
and uniform composition except in so far as the percentage of
mineral matter and consequently of bitumen may vary, being
greater in one sample than another, through sedimentation caused
by the lack of agitation during cooling. It possesses none of the
emulsion structure seen in the crude material. It has a dull
conchoidal fracture, no lustre, a blue-black color in powder and

164

THE MODERN AFPHALT PAVEMENT.

REFINED TRINIDAD LAKE ASPHALT. (AVERAGE COMPO-
SITION.)

Test number 63260

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry. ... 1 .40

Color of powder Blue-black

Lustre Dull

Structure Homogeneous

Fracture. Semi-conchoidal

Hardness, original substance 2

Odor Asphaltic

Softens 180 F.

Flows 190 F.

Penetration at 78 F 7

CHEMICAL CHARACTERISTICS.

Dry substance :

Loss, 325 F., 7 hours 1.1%

Character of residue, Smooth

Loss, 400 F., 7 hours (fresh sample) 4.0%

Character of residue Blistered

Bitumen soluble in CS 2 , air temperature '. 56 . 53%

Difference undetermined 6 . 97

Inorganic or mineral matter 36 . 50

> 100.00
Malthenes :

Bitumen soluble in 88 naphtha, air temperature 35, 6%

This is per cent of total bitumen 63 . 1

Per cent of soluble bitumen removed by H 2 SO 4 61. 3

Per cent of total bitumen as saturated hydrocarbons. ... 24 . 4

Bitumen soluble in 62 naphtha 41 . 7%

This is per cent of total bitumen 73 . 9

Carbenes:

Bitumen more soluble in carbon tetrachloride, air tem-
perature, than in bisulphide of carbon 1.3%

Bitumen yields on ignition:

Fixed carbon. 10.8

Sulphur 6.2

INDIVIDUAL ASPHALTS.

165

REFINED TRINIDAD LAKE ASPHALT.
COMPOSITION.)

(EXTREMES IN

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry
Softens

1.370
170 F.

1.405
180 F.

Flows

180 F.

190 F.

122%

90%

CHEMICAL CHARACTERISTICS.

Dry substance : *
Loss 325 F 7 hours

1.0%

1.5%

Loss 400 F 7 hours (fresh sample)

4 8%

3 5%

Bitumen soluble in CS

54.0%

57.0%

Malthenes :
Per cent of total bitumen soluble in 88 naphtha,
air temperature

63.0

68. 1

Per cent of total bitumen soluble in 62 naphtha . .
Bitumen yields on ignition:

73.0
11.0

77.0
10.0

1 Percentage depends upon the character of solvent naptha and method of treatment.

an asphaltic odor. It softens and flows below the temperature of
boiling water, but becomes liquid only at much higher tempera-
tures, about 300 F. Its density is high as compared with most
other bitumens, 1.40, owing to the mineral matter which it contains.

Refined Trinidad asphalt differs slightly in composition from
the crude material, or rather from this material after the removal,
without the aid of heat, of the water which it contains, owing to
the volatilization, at the higher temperature used in refining, of
certain constituents not driven off at the temperature of boiling
water.

The Mineral Matter in Trinidad Asphalt. The mineral matter
in Trinidad lake asphalt is as much a fixed constituent of the
material as the bitumen. It is not accidental or adventitious as
it would then vary in amount. It has evidently become mixed
with the bitumen under fixed and invariable conditions at some
subterraneous point where the bitumen originates and meets an
environment that results hi the production of the asphalt itself.

It is in an extremely fine state of subdivision. On sifting
the particles are found to be of the size given below:

166 THE MODERN ASPHALT PAVEMENT.

Passing 200-mesh sieve 08 mm. 89.8%

" 100- " " 17 " 8.0

" 80- " " 20 " 2.2

100.0

And on elutriation of the 200-mesh material:

Subsiding in 15 seconds, smaller than .08 mm. 24.3%

" " 1 minute, " " .05 " 13.1

" "30 minutes, " " .025 " 46.7

After " " " " .0075 " 15.9

100.0

Under the microscope it is found to be composed principally
of quartz with clay and the residue of the salts from the mineral
water originally emulsified with the crude bitumen. The quartz con-
sists of very sharp flakes, Fig. 2, No. 8, and their appearance leads
one to believe that it has originally been in solution under enormous
pressure in the thermal water, from which it has separated on
the release of the pressure or on cooling, and has finally flown
into fragments on further reduction of the pressure to that of
the atmosphere. The presence of the clay is more difficult to
explain. It is sufficient to know from the point of view of the
engineer that the mineral matter is extremely fine, is most inti-
mately mixed with the bitumen by nature and is consequently
the most perfect form of filler, far exceeding anything which can
be artificially mixed with a purer asphalt, and as such a most
desirable constituent.

The clay, when it has been freed by treatment with acid from
the oxide of iron which colors it the characteristic flesh-pink color
of the ash of Trinidad asphalt and which is derived from the iron
salts in solution in the thermal water, is a pure white, impalpable
powder which amounts to not more than one-third of the entire
mineral matter.

The mineral matter also contains constituents, other than the
iron, derived from the thermal water, principally sulphates and
chlorides of soda, but as the entire amount of salts in the thermal
water is less than 2 per cent, these so-called soluble salts can amount
to no more than 1 per cent of the refined asphalt and probably

INDIVIDUAL ASPHALTS.

167

much less, as a part is volatilized in refining and a part rendered
insoluble. Actual extraction of the refined asphalt when most
thoroughly carried out yields only two-tenths of 1 per cent of
soluble salts.
> The composition of the mineral matter is as follows:

Soluble in Acid.

Insoluble.

Total.

Silica, SiO 2

70.64

Alumina A1 2 O-

7.38

9.66

17.04

Ferric oxide Fe 2 O 3 *

6.30

1.32

7.62

Lime CaO

0.46

0.24

0.70

0.11

0.79

0.90

Soda, Na-jO

1.56

Potassium K . ...

Sulphuric oxide, SO 3

0.22

17.35

82.65

100.00

1 FeO not determined.

Some of the mineral matter is so impalpably fine that it will
not separate from a solution of melted or dried Trinidad pitch
in any of the usual solvents even after months of standing and
many hours' treatment in a centrifugal. It will pass also through
the finest filters. It has been thought by Peckham and others, on
this account, to be chemically combined with the organic com-
pounds of the asphalt, but the author has found that by con-
tinued swinging of a solution of the asphalt hi carbon disulphide
in a centrifugal it can be so far reduced that it amounts to but
2 per cent, and when the bitumen thus purified is dissolved in
chloroform or bisulphide of carbon and passed through a biscuit
filter all the mineral matter is removed, leaving an absolutely
pure bitumen. This is not surprising, since the smaller amount
of mineral matter finally removed is a ferruginous clay and could
not possibly be combined with the organic matter in a chemical
way, although it no doubt is in a state of close physical combina-
tion. In this connection the conclusions of Dr. Allerton S. Cush-
man of the Office of Public Roads of the U. S. Department oi
Agriculture on the porosity of clay particles is of interest.

168

THE MODERN ASPHALT PAVEMENT.

The very finest mineral matter which is separated from Trini-
dad lake asphalt has the following composition:

ANALYSIS OF FINEST MINERAL MATTER.

Insoluble
in HC1.

Soluble.

Total.

SiO 2 . .

32.36

6.74
1 40

33.64

11 74

40.38
13 14

CaO

3 20

3 65

MgO .

1 40

1 83

K 2 O

1 18

Na 2 O

.53

SO.. .

7.16

41.29

58.85

100.23

The Bitumen of Trinidad Lake Asphalt. The bitumen of Trin-
idad lake asphalt amounts to about 56 per cent of the refined
material. It is a lustrous black pitch like all pure bitumens. It
has a specific gravity of 1.06 to 1.07 and retains in suspension very
persistently amounts of the finest mineral matter of the asphalt, it
being only possible, as has been shown, to remove this by passing its
solution in bisulphide of carbon through a biscuit-filter of the Pasteur
type. In this way it can be obtained in an absolutely pure form.
In this condition it has the composition given in table on p. 169.

This bitumen is characterized by the large percentage of sul-
phur which it contains, and the presence of nitrogen. There are
apparently no oxygen derivatives present in the pure bitumen,
or they occur in very minute amounts, that is to say, they are
insoluble in chloroform or carbon disulphide.

It is characterized also by its great stability or lack of liability
to change or volatilize at high temperatures. When 20 grams
of the materials are heated in a glass crystallizing dish to 325 F.,
according to the method described in Chapter XXVIII, it loses
but 1 per cent, and only 4 per cent when exposed to a tempera-
ture of 400 F. for 7 hours.

INDIVIDUAL ASPHALTS. 169

TOTAL BITUMEN IN TRINIDAD LAKE ASPHALT.

Preparation.

IV.

Average.

82.59

81.95

82.44

82.33

10.74

10.51

10.81

10.69

6.04

6.54

5.90

6.16

0.51

0.92

1.00

0.81

99.88

99.92

100.15

99.99

Of the total bitumen about 63 per cent, when the process of
extraction is carried out according to the method described in
the chapter referred to, is in the form of hydrocarbons of the
class known as malthenes, soluble in 88 B. naphtha, having a
density of .994. The malthenes are soft and exceedingly sticky,
like maltha. When they are treated with strong sulphuric acid,
according to the author's method, all but 39 per cent prove to
be unsaturated hydrocarbons which readily enter into combina-
tion with acid. This means that but 24 to 25 per cent of the
total bitumen in the asphalt consists of saturated hydrocarbons.
On ignition these malthenes yield 6.3 per cent of fixed carbon.
This is about the same percentage that is found for the denser
California fluxes and for the lighter natural malthas.

The malthenes can be fractioned in vacuo into hydrocarbons
of different boiling-points, and from them can also be obtained
the nitrogen and sulphur derivatives corresponding to those found
by Mabery in the sulphur petroleums, all of which have been
described bv the author elsewhere. 1

It is sufficient to state here that the saturated hydrocarbon of
lowest boiling-point and molecular weight, which has been sepa-
rated from the malthenes of Trinidad lake asphalt, has the follow-
ing properties:

Boiling-point at 30 mm 165 C.

Specific gravity at 25 C. . . 8576

Refractive index 1 .4650

Carbon 86.85%

Hydrogen 13 . 34

1 On the Nature and Origin of Asphalt, Long Island City, N. Y., 1898.

170

THE MODERN ASPHALT PAVEMENT.

Such a hydrocarbon would correspond to one of the CnH2n-2
series having the formula CisH24, which contains 86.67 per cent
of carbon and 13.33 per cent of hydrogen. That this is the formula
has been confirmed by determinations of the molecular, weight
of the substance, and its molecular refraction. Of the consti-
tution of the hydrocarbons of asphalt something has been said
in a preceding chapter.

The bitumen of Trinidad asphalt which is insoluble in 88
naphtha is of the class known as the asphaltenes, according to our
purely arbitrary classification; the relative proportion of the two
forms of bitumen, asphaltenes and malthenes, being dependent upon
the nature of the solvent used, so that any information derived from
a determination of the percentages of the two classes of hydro-
carbons will be purely relative as compared with other bitumens
which have been examined by exactly the same methods and with
the same solvents. Trinidad asphalt contains a very small amount
of bitumen which is soluble in chloroform, carbon tetrachloride,
and turpentine, and which is not soluble in carbon disulphide, as
shown by Peckham. 1

The asphaltenes are hard, brittle bitumens which do not melt
but only intumesce on heating, and in this respect, as well as in
the percentage of fixed carbon which they yield, 25.8 per cent.,
correspond closely with the softer grahamites. The asphaltenes
are soluble in the heavy asphaltic oils.

The ultimate composition of the malthenes and asphaltenes
in Trinidad lake asphalt is as follows, as compared with that of
the total bitumen previously given:

Malthenes.

Asphaltenes.

Pure Bitumen.

Carbon

84 6

82 33

Hydrogen

11 3

7 8

10 69

Sulphur . .

2 9

10 9

6 16

Nitrogen

99.4

100.7

99.99

Am. J. Science, 1896, [4], 151, 193.

INDIVIDUAL ASPHALTS. 171

The saturated hydrocarbons in the malthenes have the follow-
ing composition:

Specific gravity 976

Carbon 86.40

Hydrogen 12.70

Sulphur 45

Nitrogen 07

99.62

From these figures it appears that the sulphur derivatives
are largely contained in the asphaltenes and in but relatively
small amounts in the malthenes, while they are almost completely
removed from the latter by treatment with strong sulphuric acid.
From the ultimate composition of the saturated hydrocarbons
contained in the malthenes it is evident that these belong to a
series in which the number of hydrogen atoms is considerably
below twice the carbon atoms, that is to say, they must be di-
or polycyclic polymethylenes, and very similar to those which
are found in Texas, California, and other asphaltic oils.

With the aid of the above data some insight may be gained
of the character of the bitumen of Trinidad lake asphalt. In
other asphalts and solid bitumens the proportion of malthenes
to asphaltenes may be greater or less, while the amount of satur-
ated hydrocarbons which they contain may vary. ' If we accept
the bitumen of Trinidad lake asphalt as our standard and refer
others to it, by making the same determinations of their character-
istics as has been done with the type bitumen, it is possible to
differentiate them more or less satisfactorily.

Trinidad Land Asphalt. Of the Trinidad land asphalt deposits
the author wrote, in 1892, as follows:

" La Brea Point consists of a mass of hardened pitch deposits
and reefs extending some distance into the gulf and along the
shore in both directions. The deposits are found in greater or
less abundance at all points between the shore and the lake, and
directly along the line of the road, over an area estimated at a
thousand acres or more. Two feet or more of soil cover the deposit
at some distance from the lake, but near it the thickness diminishes
and at places bare pitch is found.

172 THE MODERN ASPHALT PAVEMENT.

" On the point the pitch of the reefs is hard and resonant and
has no cementitious value. The nearer the deposits are to the
lake, however, the softer they become.

" The incline from the lake to the gulf, a distance of three-
quarters of a mile, is at first about one in twenty-five, gradually
diminishing to the shore. Near the edge of the lake there is now
a rank growth of grass, followed by shrubs and trees after passing
the forks of the road. In the village, cultivated land is found,
and large pits filled with stagnant water, from which pitch has
been excavated. Except very near the lake, the pitch excavated
from the land deposits is of a very different appearance from that
taken from the lake, and it is also of several kinds.

" The conchoidal masses removed from the lake, as I have
said, contain large gas cavities, and in appearance and somewhat
in consistency resemble a black Swiss cheese. On this account
the land pitch most nearly resembling this is known as ' cheese
pitch.' It occurs in different degrees of porosity and life. In
addition, land pitch is found in solid masses scarcely to be dis-
tinguished from refined asphalt, and this is known as ' iron pitch/
Pitch, known as ' cokey pitch/ from having been coked by the
burning of the brush over its surface, and the chocolate and friable
alteration products which have originated from atmospheric
action and disintegration, are also recognized."

Again in another place in the same report:

" In past times the pitch very probably continued to collect
(in the lake) until it overflowed the rim of the crater, in many
directions, and thus perhaps became the source of many of the
land pitch deposits now found from the end of the lake to the sea."

It has also been claimed that the land pitch has reached its
present position by being forced up through the soil from the
same source from which the lake derives its material.

For the purpose of looking into the nature of the material
from the point of view of its suitability for the asphalt paving
industry this is immaterial. It is the nature and characteristics
of the land asphalt deposits as they are available commercially
which are of interest at this point.

Trinidad land asphalt differs from that found in the lake deposit

INDIVIDUAL ASPHALTS. 173

as the result of such changes as have been brought about by its
having been buried under soil and exposed to the action of ground-
water and aging for many centuries, that is to say, it is a very
much weathered material. The two materials are undoubtedly
derived from the same original subterranean source. The land
asphalt may, as has been said, have reached its present position
either by overflow from the lake or by intrusion into its present
position in the soil directly from the point of origin. In either
case no land asphalt has been found which has not been altered
in its nature owing to its present or past environment, so that
it differs essentially from the lake material.

Land asphalt is very variable in character, depending upon the
length of time during which it has been subjected to weathering. It
is much less cheesy than lake asphalt, that is to say, it contains
a smaller number of gas cavities, and is harder. Some of it has
been converted by brush fires into a hard compact pitch without
gas cavities, resembling refined asphalt which, from its hardness,
is known as iron pitch. Some portion of the land asphalt has
been converted even to coke. These two latter forms are of no
interest to the paving industry and are carefully removed from
the material which is collected for this purpose. Further weather-
ing of the material from the action of soil-water converts it into
a substance of a chocolate color, which is very friable. Where
the weathering is carried to its ultimate conclusion only the fer-
ruginous mineral matter is left, of a bright-red color, as is evident
on the beach; where the asphalt is subjected to the continuous
action of sea water it becomes very hard but is not weathered
to any further degree. That is to say, water containing the salts
found in sea water does not act upon it. This is important in
connection with the claims that soluble salts cause disintegration
of Trinidad asphalt.

Much of the asphalt found in the land deposits plainly orig-
inated in the so-called lake and is now found mingled with the
soil after it has run over the rim of the lake. In consequence,
the land asphalt nearest the lake is much less weathered than
that at a distance. The difference can be seen from the following
analyses of specimens collected near the lake and at intervals

174

THE MODERN ASPHALT PAVEMENT.

between it and the shore. For comparison an analysis of lake
asphalt is given:

AVERAGE COMPOSITION OF LAKE PITCH, DRIED IN VACUO,
KEARNEY COLLECTION.

Bitumen
Soluble CS 2 .

Mineral
Matter.

Difference
Undeter-
mined.

Bitumen
Soluble
Petroleum
Naphtha.

Total
Bitumen
Soluble in
Naphtha.

Average

54.25%

36.51%

9.24%

35.41%

65.27%

AVERAGE COMPOSITION OF LAND PITCH, DRIED IN VACUO,

KEARNEY COLLECTION.
Ei^ht specimens from Lot C. , near the lake.

Average

54.03%

36.49%

9.48%

33.02%

Four specimens from Crown Land Lots adjoining C.

Average.

53.81%

36.62%

9.57%

32.29%

Five specimens from east of road, middle ground.

Average

52.31%

37.80%

9.89%

31.25%

61.11%

60.01%

59.74%

Seven specimens from Village Lots, near the Gulf.

Average
General average.

52.27%
53.10

37.73%
37.16

10.01%
9.74

31-42%
31.99

60.12%
60.14

It is apparent that the effect of weathering has been com-
paratively small in the deposits closely adjoining the lake, but
that, as we proceed further on toward the shore, the change is
more marked and is particularly evidenced by the decrease in
bitumen present, and decrease in the percentage of malthenes.
The lake pitch, so-called, has, with the naphtha used as a solvent,
65.3 per cent of its bitumen soluble in naphtha, while just outside
the lake the land pitch has only 61.1 per cent, and further on only
59.7 per cent soluble. This may seem a small difference, but it

INDIVIDUAL ASPHALTS. 175

is evidence of a large change. In glance pitch, examined in the
same way, 24 per cent only of the entire bitumen was found to
be soluble in naphtha, in lake pitch 65.3 per cent. Land pitch
may, therefore, be inferred to be partly converted from lake to
glance pitch.

The composition of an average commercial refined land
asphalt as compared with the average lake material is well shown
in the table on page 176.

Land asphalt being so dependent upon its environment for
its character, it is hardly possible to determine its average com-
position. The extremes which are met with in recent commer-
cial supplies are of interest (see table, page 177).

From these figures it appears that refined Trinidad land
asphalt of good quality is differentiated from the lake supply by its
higher specific gravity, owing to the rather larger amount of mineral
matter which it contains, by a higher softening or melting-point,
and somewhat lower percentage of bitumen and, in consequence of
these facts, a much greater hardness at all temperatures.

Naturally the percentage of malthenes is smaller in the land
than in the lake asphalt and that of fixed carbon slightly higher.

The ultimate composition of the pure bitumen of land asphalt as
compared with that of lake is given in the second table on page 177.

The weathering of the bitumen has, therefore, produced essen-
tial changes in the composition of the material, there having
been a loss of sulphur, probably due to its elimination as hydro-
gen sulphide, and an increase in carbon, due to the same cause
and to the elimination of hydrogen as water during the process of
condensation.

These differences, though small in themselves, are indicative
of the fact that the weathered land asphalt is not the same in
its character as the fresh lake material. In themselves they
would not amo'unt to a great deal unless confirmed by actual
results obtained in the use of the material industrially. As a
matter of fact, such confirmation is not lacking if land is used in
the same way as lake asphalt, that is to say, if fluxed to an asphalt
cement with ordinary paraffine petroleum residuum. In the
preparation of such an asphalt cement the striking differences

176

V THE MODERN ASPHALT PAVEMENT.
REFINED TRINIDAD LAND ASPHALT.

63260

36721

Lake R. A.

Land R A

PHYSICAL PROPERTIES.

Specific gravity, 78 F,/78 F., original sub-
stance dry . . .

1 40

1 .4196

Color of powder or streak

Blue-black

Brown-black

Lustre

Dull

Homogeneous

Semi-

Odor

conchoidal
Asphaltic

Softens

180 F.

188 F

Flows . ....

190 F

198 F

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance:
Loss, 325 F , 7 hours

1 1%

1 0%

Character of residue

Smooth

Blistered

Loss, 400 F. , 7 hours (fresh sample)

4 0%

3 0%

Character of residue

Blistered

Bitumen soluble in CS 2 , air temperature
Difference

56.53%
6 97

54.1%
7 9

Inorganic or mineral matter

36 50

Malthenes:
Bitumen soluble in 88 naphtha, air tem-
perature

100.0
35 6%

100.0
33 5%

This is per cent of total bitumen

63.1

61 9

Per cent of soluble bitumen removed by
H 2 SO 4

61.3

64.8

Per cent of total bitumen as saturated hy-
drocarbons

24 4

21 8

Bitumen soluble in 62 naphtha

41 7%

38 2%

This is per cent of total bitumen

73.9

70.6

Carbenes:
Bitumen more soluble in carbon tetra-

1 3%

Bitumen yields on ignition:
Fixed carbon

108%

129%

Sulphur

6.2%

5.0%

INDIVIDUAL ASPHALTS.

177

REFINED TRINIDAD LAND ASPHALT. (EXTREMES IN
COMPOSITION.)

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance,
drv

1.400

1.450

Softens

188 F.

220 F.

Flows

198 F.

230 F.

Flow in per cent of Trinidad lake = 100%

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 air temperature

83%
55 0%

15%
52 0%

Difference .

7.5

9.5

Inorganic or mineral matter

37.5

38.5

Per cent of total bitumen soluble in 88 naphtha,
air temperature

100.0
63

100.0
52.0

Per cent of soluble bitumen removed by H-jSO. . . .
Per cent of total bitumen soluble in 62 naphtha . .

Bitumen yields on ignition:

62.0
71.0

12.9

64.8
60.0

14.0

COMPARISON OF ULTIMATE COMPOSITION OF PURE
BITUMEN, TRINIDAD LAND AND LAKE ASPHALT.

Land.

Lake.

Carbon

83 7

82.33

Hydrogen

10 8

10.69

Sulphur

5.1

6.16

Nitrogen

0.5

0.81

100.1

99.99

between lake and land asphalt are brought out by the fact that
where 20 pounds of a good paraffine residuum is sufficient to make
a suitable asphalt cement when added to each 100 pounds of lake
asphalt, as much as 30 or more pounds of the same residuum are
required to make a cement of the same consistency with land
asphalt. Extended experience with surfaces laid with these two
asphalt cements has shown that after 3 or 4 years service the
surface laid with the land asphalt begins to show signs of deterio-
ration and often at even shorter periods. Our practical results,
therefore, confirm those obtained in the laboratory even more
strikingly than might be expected.

178 THE MODERN ASPHALT PAVEMENT.

Land asphalt, as has been seen, contains a very much weathered
and hardened bitumen, much more hardened than one would be
led to believe from mere analytical results and only shown by
the necessity for the use of half as much again of flux in making a
cement, and by the fact that when cylinders of the two asphalts
of the same size are placed upon a corrugated brass plate and
exposed to a high temperature, at an angle of 45, the land asphalt
flows from but 20 to 50 per cent as far in a given time as is the
case with the lake asphalt. This is, of course, due to the absence
of the softer hydrocarbons of the malthene series which, in the
lake asphalt, are very susceptible to increase of temperature and
consequently occasion the more rapid flow of the latter. The
absence of these malthenes in a paving cement is a serious deficiency.

The recent advent of the heavy asphaltic oils as fluxes makes
it possible, however, to supply to a certain extent the deficiencies
in the bitumen of land asphalt by the addition of a suitable amount
of such flux, instead of producing the required softness with a
great excess of paraffine residuum, thus forming an unbalanced
cement. Pavements made with such an asphalt cement have
been fairly satisfactory, but are, unfortunately, uneven in char-
acter owing to the lack of uniformity of the original land asphalt,
and to the large amount of skill which must be used in combining
the different ingredients and to the care which the production of
a uniform asphalt cement from such materials presents.

Bermudez Asphalt. The occurrence of a deposit of asphalt
in what has been called the Bermudez " Pitch Lake," in the State
of Sucre, formerly Bermudez, in Venezuela, Fig. 4, has been
described by the author in another place as follows: 1

" From the mouth of the Orinoco, the northeastern coast of
Venezuela, which faces Trinidad, is low and consists of vast man-
grove swamps, through which run deep tidal estuaries. That por-
tion forming part of the State of Bermudez extends inland for many
miles. It lies on the opposite side of the Gulf of Paria from Trini-
dad. About 30 miles in an air-line from the coast the asphalt
deposit, known as the Bermudez Pitch Lake, is found at the point
where a northern range of foot hills comes down to the swamp.

1 On the Nature and Origin of Asphalt, Long Island City, N. Y., 1898,

INDIVIDUAL ASPHALTS. 179

The Guanaco River, a branch of the San Juan, one of the large
canos or estuaries of this region, at about 65 miles in its winding
course, from its mouth, runs within 3 miles of the deposit, but
it is 5 or 6 miles to a suitable wharfage site. On the other hand,
towards the north a road runs to the hills and to the village of
Guaryquen. These are the means of communication with the
deposit. The so-called lake is situated between the edge of the
swamp and the foot hills in what might be termed a savanna. It
is an irregular-shaped surface with a width of about a mile and a
half from north to south and about a mile east and west. Its area
is a little more than 900 acres, and it is covered with vegetation,
high rank grass, and shrubs, 1 to 8 feet high, with groves of large
moriche palms, called morichales. One sees no dark expanse of
pitch on approaching it as at the Trinidad pitch lake, and except
at certain points where soft pitch is welling up, nothing of the
kind can be found. The level of the surface of the deposit does
not vary more than 2 feet and is largely the same as that of the
surrounding swamps. In the rainy season it is mostly flooded
and at all times very wet, so that any excavation will fill up with
water. These conditions make it difficult to get about upon it
or to excavate pitch easily.

" It is readily seen that this deposit is a very different one from
that in the pitch lake of Trinidad. It seems to be in fact merely an
overflow of soft pitch from several springs over this large expanse
of savanna and one which has not the depth or uniformity of that at
Trinidad.

" Being on a level with the mangrove swamps and with foot hills
on its other side, any large amount of asphalt could hardly be held
in position here, as in the old crater in Trinidad, but would burst
out into the swamp and be lost, and, as far as borings have been
made, they seem to indicate but a small depth anywhere as com-
pared with that of the Trinidad lake.

" At different points there is at most a depth of 7 feet of mate-
rial, while the deepest part of the soft maltha is only 9 feet and
the average of pitch below the soil and coke only 4 feet. At points
there* is not more than 2 feet of pitch, and in the morichales or
palm groves it is often 5 feet below the surface. At several points,

180 THE MODERN ASPHALT PAVEMENT.

scattered over the surface, are areas of soft pitch, or pitch that
is just exuding from springs. The largest area is about 7 acres
in extent and of irregular shape. This has little or no vegetation
upon it, and from the constant evolution of fresh pitch is raised
several feet above the level of the rest of the deposit. This soft
asphalt has become hardened at the edges, but when exposed
to the sun is too soft to walk upon. The material is of the nature
of a maltha and it is evidently the source of all the asphalt in the
lake, from these exudations the pitch having spread in every
direction, so that no great depth of pitch is found even at this point.

" A careful examination of the surroundings shows that in
one respect there is a resemblance between the point of evolution
of the soft pitch at the Bermudez and at the Trinidad lakes. Gas
is given off in considerable quantities at both places, and in both
cases consists partly, at least, of hydrogen sulphide. At the
Bermudez lake I was unable to determine whether it was accom-
panied by carbonic dioxide, but the odor of hydrogen sulphide
was strong.

" The consistency of the soft pitch at the centre of the Ber-
mudez lake is much thinner than that of the Trinidad lake. It
will run like a heavy tar and does not evolve gas in the same rapid
way or harden as quickly after collection. It therefore does not
retain the gas which is generated in it, nor does the deposit as a
whole do so to the same extent as the Trinidad pitch. Where,
however, the surface of the soft pitch has toughened by exposure
to the sun and air and where gas is given off beneath it, it is often
raised in dome-like protuberances, the beehives which were spoken
of by early visitors to the Trinidad lake. These have a thin wall of
pitch and are filled with gas which readily burns, and have been
seen two feet or more in height and 18 inches in diameter. They
are, of course, found only near the soft spots.

" Although the pitch at the Bermudez lake is too soft to entangle
and hold permanently the gas which is given off, where the pitch
of medium consistency is covered with water it does not escape
so readily, and thus often raises in the pools of water a mushroom-
like growth of pitch by the reduction of the gravity of the friass
from the included gases. These mushrooms correspond completely,

INDIVIDUAL ASPHALTS. 181

except in size, with those described by Manross as existing at the
Trinidad lake when he visited it. It seems, therefore, that we
have to-day several of the phenomena represented at the Venezue-
lan lake which the hand of man has destroyed at Trinidad.

" There is, however, no evidence of the same simultaneous
boiling up of water with the fresh soft pitch that has been deter-
mined at the Trinidad lake, but that there is none at all is not
certain, as at the tune I visited the locality heavy rains were fall-
ing which prevented the detection of a small amount. It seems,
however, improbable, as the soft pitch contains little or no water
and the traces found in the samples collected are probably derived
from rain.

" Hardening of the Main Mass of Pitch. The soft pitch, after
it exudes at the centre of the Bermudez lake, undoubtedly hardens
slowly on exposure, but the condition of the surface of the main
mass, which is very hard and rough, and of the harder borders
of the soft spots is due to other causes also.

" The edges of the areas of soft asphalt are covered here and
there with masses of glance pitch and with black and brittle cin-
ders or coke, and which seem to have been produced from the
maltha by fire. This is evidently the case, since the rank growth of
grass which is very dry in the dry season is particularly adapted
for a rapid and intense combustion. Such fires have been even
recently started intentionally and accidentally and to them are
due the condition of the present surface of the deposit and the
character of much of the pitch.

" The general surface of the lake is very irregular and hard.
There are many very narrow : and irregular channels or depres-
sions from a few inches to 4 feet deep, filled with water, and not
being easily distinguished, one often falls into them. At the foot
of the growth of grass and shrubs are ridges of pitch mingled with
soil and decayed vegetation, which have been plainly coked and
hardened by fires of the nature which have been mentioned. When
this hardened material which forms only a crust is removed, asphalt
of a kind suitable for paving is found. The crust is from 1% to
2 feet in depth and very firm, while the asphalt underneath would
not begin to sustain the weight which that of the Trinidad pitch

182 THE MODERN ASPHALT PAVEMENT.

lake does easily. There are breaks in the crust here and there
through which soft pitch exudes as has been described.

" It appears, therefore, that the Bermudez deposit owes its
existence to the exudation of a large quantity of soft maltha,
which is still going on and which has spread over a great extent;
that this has hardened spontaneously in the sun, and has also,
by the action of fire, been converted over almost the entire sur-
face into a cokey crust of some depth, beneath which the best
material lies and that here and there are scattered masses of glance
pitch produced in a similar way from less violent action of heat.
There is no evidence of a general movement and mingling of the
mass of this deposit in any way that would produce a uniformity
of composition as seen in the Trinidad pitch lake, although there
is a certain amount of gas evolved at the soft spots where maltha
exudes and some gas cavities are found in the general mass of
the pitch beneath the crust."

The original Bermudez pitch, as it exudes at the soft spots,
contains no mineral matter or water and is consequently an
extremely pure bitumen. It has the following ultimate com-
position:

Carbon 82.88%

Hydrogen 10 .79

Sulphur 5.87

Nitrogen . . . .75

100.29

As in the case of the bitumen of Trinidad asphalt, sulphur plays
an important role in its composition, but in the asphalt as it is
used commercially much of this sulphur has disappeared, having
been evolved as hydrogen sulphide, the industrial material con-
taining less than 4 per cent.

Exposure of the soft maltha to the sun, after its evolution
and chemical changes, hardens the material somewhat. The com-
mercial supply has, however, been largely altered, owing to the
fact that the rank vegetable growth which extends over the 900
or 1000 acres of the deposit has frequently been set on fire, either
accidentally or intentionally, with the production of such a degree
of heat as to convert much of the material on the surface to coke

INDIVIDUAL ASPHALTS.

183

and that below to a harder form of bitumen than that of the orig-
inal maltha. It is, of course, very evident that this conversion
has not been uniform and that the product taken from the lake
cannot, on this account, be in itself uniform. In a collection of
some forty samples taken in 1896, the following extremes of com-
position were found:

EXTREMES IN THE COMPOSITION OF CRUDE BERMUDEZ

ASPHALT.

Highest.

Lowest.

CRUDE SUBSTANCE.

Loss 212 F 4 days

46 20%

10 72%

' ' 400 F 7 hours

13.60

4 72

DRIED SUBSTANCE.

Specific gravity 78 F /78 F

1 075

1 005

Loss 400 F 7 hours

16 05

5 81

Softens

170 F

140 F.

Flows *

lg8F.

135 F.

Bitumen soluble in CS^ air temperature

98 52%

90 65%

Difference

6 45

Inorganic or mineral matter

3 65

Bitumen soluble, 88 naphtha, air temperature. . . .
This is per cent of total bitumen

73.05
76.55

63.40

67.78

The asphalt as it occurs in the deposit holds from 46 to 10 per
cent of water and 3.6 to .5 per cent of mineral matter. These
substances as well as the undetermined material, amounting
to from 6 to .6 per cent, must be adventitious, and are in part, at
least, derived from the vegetation with which it comes in contact
and, although some of the organic matter is due to the conversion
of part of the bitumen to coke by fires, the greater portion con-
sists of roots of grasses and shrubs which penetrate the asphalt
with ease.

The wide variation in the character of the material in differ-
ent parts of the deposit is therefore made evident by the above
figures. The high percentage of undetermined matter is not, as
in Trinidad asphalt, due to the presence of clay or volatile inor-
ganic salts.

There are other evidences of this variation derived from the

184

THE MODERN ASPHALT PAVEMENT.

examination of various cargoes of Bermudez asphalt which have
been brought to this country for commercial use.

Crude.

After Drying.

Year.

Water.

Bitumen

Soluble
in CS 2 .

Difference
Undeter-
mined.

Inorganic or
Mineral
Matter.

1898

15.8%

95.7%

2.3%

2.0%

1899

19.8

92.5

4.3

3.0

1901

95.0

2.5

sample

23 9

95 7

2 8

1.5

1902 { Poor

< i

53 2

89.2

9.0

1.8

1 QO** / G d

1903 \ Poor

sample

< <

29.9
33.3

97.0
95.5

1.5
2.9

1.5
1.6

Refined Bermudez Asphalt. When the crude Bermudez asphalt,
as it is taken from the deposit, is melted and dried it becomes the
refined Bermudez asphalt of commerce and it is, of course, vari-
able in character like the crude material it is derived from. The
loss of light oils in the softer material has a tendency to bring all
the refined asphalt more nearly to a uniform condition than would
be expected. There are, however, decided differences not only
as regards its physical characteristics but also chemically.

The variation in the consistency is well shown by the relative
length to which various lots of this asphalt will flow on an inclined
plane at temperatures above the softening point. For several
cargoes in this way the following data were obtained:

PER CENT OF FLOW.

Original material in use in Washington, D. C. ,

1893 100.0%

Importation of 1895 73 . 5

" " 1898 73.0

" " March, 1899 50.0

" " May " 41.6

" " June " 73.3

" " 1903 125.0

From the preceding figures it appears that while the first mate-
rial brought to this country was quite soft, the refined asphalt
became harder in subsequent years and recently is again softer,

INDIVIDUAL ASPHALTS. 185

owing to the fact that the crude asphalt has been collected of
late at points nearer the maltha springs.

In consequence of this every lot of Bermudez asphalt must be
handled in its own peculiar way, and in this respect it compares
unfavorably with Trinidad lake asphalt which is, as has been
seen, extremely uniform.

The percentage of bitumen in the refined material varies as
well as the consistency, ranging from 93 to 97 per cent, but will
usually average 95 per cent.

The data given in the tables on pages 186 and 187 show the
physical properties and approximate chemical composition of refined
Bermudez asphalt available on the market in 1900 and 1903.

Bermudez asphalt is a comparatively pure bitumen and conse-
quently possesses the lustre of such material instead of the dull
fracture of Trinidad bitumen which is due to the presence of
mineral matter. It has a uniform structure with here and there
small particles of vegetable organic matter. The fracture at low
temperature is somewhat conchoidal and the consistency, as shown
on the Bowen penetration machine, ranges from 22 to 26. The
specific gravity of the material is 1.08, somewhat higher than that
of the pure bitumen of Trinidad asphalt, on account of the presence
of a certain amount of mineral matter. The hydrocarbons compos-
ing the bitumen consist to a very considerable extent of such as
are volatile at 400 F., or even at 325 F. In this respect it is
markedly different from Trinidad lake asphalt. Bermudez asphalt
differs from Trinidad lake asphalt in containing a larger percent-
age of malthenes, which accounts for its greater softness. As
this percentage becomes increased the softer is the consistency
of the asphalt, as appears from the fact that the softer refined
material of 1903 contains 71.9 per cent of malthenes, where that
of 1900 contained but 65.4 per cent.

The percentage of saturated hydrocarbons unacted upon by
sulphuric acid in Bermudez asphalt is about the same as that
found in Trinidad lake asphalt, so that the bitumens in the two
asphalts do not differ essentially in this respect.

The undetermined matter not of a bituminous nature in Ber-
mudez is, as a rule, very much smaller than in Trinidad asphalt and

186

THE MODERN ASPHALT PAVEMENT.
REFINED BERMUDEZ ASPHALT.

Test number

44412

67753

Year

1900

1903

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original sub-
stance dry

1 0823

1 0575

Color of powder or streak

Black

Lustre

Bright

Structure

Uniform

Fracture

Semi-

Hardness, original substance

conchoidal

Soft

conchoidal

Soft

Odor

Asphaltic

Softens

170 F

160 F

Flows

180 F

170 F

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance :
Loss 325 F , 7 hours

3 0%

4 4%

Character of residue

Smooth

Loss, 400 F., 7 hours (fresh sample)

<8 2%

9 5%

Character of residue

Wrinkled

Shrunken

Bitumen soluble in CS 2 , air temperature
Difference . . >

95.0%
2 5

96.0%
2

Inorganic or mineral matter.

2 5

Malthenes:
Bitumen soluble in 88 naphtha, air tem-
perature

100.0

62 2%

100.0
69 1%

This is per cent of total bitumen
Per cent of soluble bitumen removed by
H,SO 4

65.4
62 4

71.9
67 4

Per cent of total bitumen as saturated hy-
drocarbons

24 4

23 4

Bitumen soluble in 62 naphtha

69.2%

75.9%

This is per cent of total bitumen

72 8

Carbenes :
Per cent bitumen insoluble in carbon
tetrachloride, air temperature

0.1%

1.1%

Bitumen yields on ignition: "
Fixed carbon

13.4%

14.0%

Sulphur

4 0%

INDIVIDUAL ASPHALTS.

187

EXTREMES IN THE COMPOSITION OF REFINED BERMUDEZ

ASPHALT.

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance,
dry

1.085

1.057

Flow in per cent of average

41.6%

178%

CHEMICAL CHARACTERISTICS.

Dry substance:
Loss 325 F 7 hours

5.5%

1.5%

Loss 400 F 7 hours (fresh sample)

9 5%

4 5%

Bitumen soluble in CS air temperature

93.0%

96.8%

Difference

5.0

1.4

2.0

1.8

Malthenes:
Per cent total bitumen soluble in 88 naphtha,
air temperature

100.0
65.0

100.0
73.0

Per cent total bitumen soluble in 62 naphtha. .

Bitumen yields on ignition :
Fixed carbon

72.0
14.0

83.0
13.5

of an entirely different origin. In Bermudez asphalt it is derived
almost entirely from vegetable organic matter in the shape of grasses
and twigs with which the pitch has become contaminated. As a
result of this, when the asphalt is made into a cement with a flux
and maintained in a melted condition for a considerable period
of time, this settles out on the bottom of the melting-tank as the
gummy material which has been mentioned by the persons not
further investigating its nature. If this gummy material is
extracted with carbon disulphide, the vegetable nature of the
material will be revealed at once.

If the ultimate composition of the bitumen of Bermudez asphalt
where it exudes into the lake and that of the pure Trinidad bitu-
men are compared, it will be noticed that they agree very closely
in their composition (see table, page 188).

The percentage of fixed carbon which Bermudez refined asphalt
yields on ignition is much higher than that found in Trinidad

188

THE MODERN ASPHALT PAVEMENT

COMPARISON OF ULTIMATE COMPOSITION OF PURE BITUMEN,
BERMUDEZ AND TRINIDAD ASPHALT.

Bermudez,
Pure Bitumen.

Trinidad,
Pure Bitumen.

Carbon

82 88%

82 33%

Hydrogen

10 79

10 69

Sulphur

5.87

6 16

Nitrogen

.75

100.29

99.99

asphalt and it is an amount which is usually characteristic of all
the native bitumens which are distinctly asphaltic in their nature.

The amount of sulphur is greater in the softer than hi the
harder varieties.

From the preceding data the two most important asphalts
in use in the paving industry may be compared with the following
results:

Trinidad asphalt is one which carries a very considerable amount
of mineral matter which acts as a filler. Bermudez asphalt is a
nearly pure bitumen. Trinidad asphalt is very stable at high
temperatures and but little susceptible to change. Bermudez
asphalt volatilizes an appreciable amount of light oils at high
temperatures and hardens very rapidly. Bermudez asphalt con-
tains a larger percentage of malthenes than Trinidad and on this
account is more susceptible to temperature changes. Finally,
Trinidad asphalt is a substance of fixed and extremely uniform
composition, while Bermudez asphalt is most variable in this re-
spect, material from different parts of the deposit showing great lack
of uniformity in both its physical and chemical properties. The
relative merits of the two materials from an industrial point of view
will be considered later when surface mixtures are under discussion.

Maracaibo Asphalt. The asphalt used in the paving industry,
known as Maracaibo asphalt, is put upon the market by the United
States and Venezuela Company. It is found in the State of Zulia,
west of the Gulf of Maracaibo, on the river Limon, about 50 miles
west from the City of Maracaibo, as shown on the accompanying
map, Fig. 5. The principal deposit is known as that of Inciarte.

INDIVIDUAL ASPHALTS.

189

It is an exudation, from maltha springs, of
gathered up and crudely refined, after which
the river to the village of Toas, at the opening
into the Gulf, from which point it is shipped to
The material resembles and possesses many
tics of the crude Bermudez asphalt, but it is

bitumen which is
it is floated down
of Maracaibo Lake
the United States,
of the characteris-
distinguished from

eflnery^ __

Maracaibo
LaPaz Deposit
STAT.E OF ZULIA

V E N yJE Z

MAMA CALBO

FlG. 5.

it by having a markedly rank odor suggestive of unsaturated
hydrocarbons and of sulphur derivatives. This odor may, how-
ever, be due somewhat to cracking which has taken place during
refining, since in the analysis of the material numbered 66923
seventeen per cent of the bitumen is in the form of carbenes
insoluble in cold carbon tetrachloride, while in the samples col-
lected in 1904 less than two per cent is in this form.

Five analyses of the crudely refined material made itt. th
York Testing Laboratory have resulted as follows:

190

THE MODERN ASPHALT PAVEMENT.

MARACAIBO

Test number 60380

Date sample received 6-19-02

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 .0784

Color of powder or streak Black

Lustre Bright

Structure Uniform-

homogeneoui

Fracture Semi-

conchoidal

Hardness, original substance 1

Odor Strong

Softens 280 F.

Flows 300 F.

Penetration at 78 F 8

CHEMICAL CHARACTERISTICS.

Original substance :

Loss, 212 F., 1 hour Trace

Dry substance :

Loss, 325 F., 7 hours S.3% 1

Character of residue ....

Loss, 400 F., 7 hours (fresh sample)

Character of residue ....

Bitumen soluble in CS 2 , air temperature 92 . 2%

Difference 6.3

Inorganic or mineral matter 1.5

100.0

Malthenes :

Bitumen soluble in 88 naphtha, air temperature 45 . 8%

This is per cent of total bitumen 49 . 7

Per cent of soluble bitumen removed by H 2 SO 4 ....

Per cent of total bitumen as saturated hydrocarbons ....

Bitumen soluble in 62 naphtha 49. 7%

This is per cent of total bitumen 53 . 5

Carbenes :

Bitumen insoluble in carbon tetrachloride, air temperature. ....

Bitumen yields on ignition:

Fixed carbon ^ 19.0%

Vegetable organic matter ....

1 Loss determined in open dish temperature 325 to 350 F.

INDIVIDUAL ASPHALTS.

191

ASPHALT.

61047

66923

72214

72215

8-20-02

10-19-03

8-25-04

1.0634
Black
Bright
Uniform

1.0638
Black
Bright
Uniform

1.0660
Black
Bright
Uniform

1.0621
Black
Bright
Uniform

Semi-
conchoidal
Soft
Strong
220 F.
230 F.
25

Semi-
conchoidal
Soft
Strong
200 F.
210 F.
20

Semi-
conchoidal
Soft
Strong
215 F.
230 F.
27

Semi-
conchoidal
Soft
Strong
195 F.
210 F.
26

Trace

4.5%'

2.7%
Blistered

1.5%
Smooth

.":::.: :

4-7%
Much blistered

5.8%
Blistered

6.0%
Blistered

94.0%
4.5%
1.5

96.8%
1.4
1.8

92.2%
2.0
5.8

94.3%
2.1
3.6

100.0

54.5%
57.9

45.7%
47.2
46.4
25.3

53.4%
57.9
48.5
25.5

53.9%
57.2
49.2
29.1

59.5%
63.3

51.5%
53.2

17. 5% 2

1.5%

1.3%

15.0%

18.0%

17.0%

16.9%

8.0%

2 Duplicate 17.8%.

192 THE MODERN ASPHALT PAVEMENT.

It appears from the preceding data that Maracaibo asphalt,
like that from the Bermudez lake, contains considerable vegetable
organic matter. On re-refining in the laboratory it has a density
corresponding to the pure bitumen of Trinidad lake asphalt, that
is to say, somewhat less than that of Bermudez, and a very con-
siderable degree of purity, from 92 to 97 per cent. The refined
material is soft enough to be indented with the finger-nail. Apart
from the preceding characteristics it differs in other respects from
Bermudez asphalt and from other asphalts with which we are
acquainted. Its softening-point is not only higher than that of
Bermudez asphalt but even that of Trinidad. It contains a very
small percentage of malthenes, which might be expected from its
high softening-point, but not from its consistency. The percent-
age of saturated hydrocarbons found in the malthenes is 25 to
29 per cent. The percentage of fixed carbon obtained on ignition
is higher than that found in the normal asphalts, reaching 18 per
cent. On heating it to a temperature not above 325 F. gas is
evolved showing that the bitumen is unstable and in a state of
change. What effect these peculiarities may produce in the
asphalt pavements constructed with it time and experience alone
can tell.

Cuban Asphalts. Many different forms of native bitumens are
found in the island of Cuba but the deposits are of such small
extent in any one place that they are of no great commercial inter-
est, except that they are imported in small amounts for the manu-
facture of varnishes and in one or two instances for paving purposes.
Solid bitumen, in the form of grahamite, has been mined in
the Provinces of Pinar del Rio and Havana, but no attempts have
been made to utilize these in the paving industry. In the neigh-
borhood of the village of Bejucal, 18 miles south of Havana, there
are several mines of bitumen of an asphaltic nature, one of which
has been worked on a commercial scale and utilized in the United
States in the construction of pavements in Washington, D. C. The
bitumen is more or less variable. A specimen of it had the follow-
ing composition:

INDIVIDUAL ASPHALTS.

193

ASPHALT FROM BEJUCAL DISTRICT, CUBA.

Test number 22220

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 .305

Color of powder or streak Red-brown

Lustre Dull

Structure Compact

Fracture Semi-conchoidal

Hardness 2

Odor Asphaltic

Softens 230 F

Flows 240 F

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance:

Loss, 325 F., 7 hours .88%

Character of residue Cracked

Loss, 400 F., 7 hours (fresh sample) 1 .50%

Character of residue Wrinkled

Bitumen soluble in CS 2 , air temperature 75. 1%

Difference 3.5

Inorganic or mineral matter 21.4

100.0

Malthenes:

Bitumen soluble in 88 naphtha, air temperature 32 . 4%

This is per cent of total bitumen 43 . 1

Per cent of soluble bitumen removed by H.5SO 4 60 . 5

Per cent of total bitumen as saturated hydrocarbons 17.0

Bitumen soluble in 62 naphtha 39.6%

This is per cent of total bitumen 52 . 7

Carbenes :

Bitumen more soluble in carbon tetrachloride, air temper-
ature 1.6%

Bitumen yields on ignition:

Fixed carbon 25.0%

Sulphur 8.3%

194 THE MODERN ASPHALT PAVEMENT.

In general appearance the Bejucal asphalt resembles the Trini-
dad refined material. It contains about 21 per cent of mineral
matter in the form of silica and silicates and 75 per cent of bitumen.
The specific gravity of the asphalt corresponds to a mixture of
bitumen and mineral matter in these proportions. It has a very
high softening point. The percentage of its total bitumen in the
form of saturated hydrocarbons as revealed by the action of sul-
phuric acid on the malthenes soluble in 88 naphtha is smaller
than that in Trinidad asphalt. It contains a large percentage
of sulphur and yields a high percentage of fixed carbon, larger
than that usually found in any of the asphalts. In this respect
it is more closely allied to the grahamites than to the asphalts,
and it may perhaps eventually be necessary to classify it with
the latter form of bitumen. As might be expected from its
extreme hardness, the percentage of malthenes is only about two-
thirds as much as that found in Trinidad and Bermudez asphalt,
and this necessitates the use of a heavy asphaltic flux in the prepa-
ration of a satisfactory asphalt cement from this material.

In the neighborhood of the Bejucal mine are several others,
partial examinations of which have been made giving the results
tabulated on page '195.

Mexican Asphalt. Native bitumen in the shape of maltha
and in a more or less solid form is of frequent occurrence in Mexico,
especially along the coast of the Gulf of Mexico, in the States of
Tamaulipas and Vera Cruz. In the neighborhood of Tampico
and Tuxpan attempts have been made for many years to work
the large effusions of maltha which occur there. At none of these
deposits, however, is there a sufficient amount of bitumen avail-
able to make the material of commercial importance. Lots of
several hundred tons have, however, been collected and shipped
to the United States and used in pavements, so that it may be
a matter of interest to determine what the character of the bitu-
men is.

Asphalt Effusions on the Tamesi River. At about 45 miles from
Tampico and 25 miles from Los Esteros, a station on the Me.
& Gulf R.R., there are large tar springs. From these effusions
some hundreds of tons of asphalt have been collected from time

INDIVIDUAL ASPHALTS.

195

DEPOSITS IN THE NEIGHBORHOOD OF THE BEJUCAL
MINE, CUBA.

Test number

22221

Name of mine

"Angelo

"Raboul"

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original sub-
stance dry

Elmira"
1 348

1.306

Color of powder or streak

Dark brown

Brown

Lustre

None

Uniform

Brecciated

Semi-

Conchoidal

Hardness original substance. .

conchoidal
3

Softens

245 F.

240 F.

Flows

270 F.

250 P.

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Original substance:
Loss, 212 F , 1 hour

3.2%

.4%

Bitumen soluble in CS 2 air temperature

68 6%

73 0%

Difference ... .... . .

3 5

Inorganic or mineral matter

27 9

Malthenes:
Per cent of total bitumen soluble in 88
naphtha air temperature

100.0
36 6%

100.0
49 3%

Per cent of total bitumen soluble in 62

64.5%

Carbenes:
Bitumen insoluble in carbon tetrachloride,
air temperature

14.3%

Bitumen yields on ignition:
Fixed carbon

17 4%

to time and shipped to the United States. The material is not ot
uniform composition, samples collected in 1899 and examined in
the author's laboratory having the characteristics given in the first
table on page 196 and in the table on page 197.

It appears that this bitumen although usually originally quite
hard, as shown by the penetration at 78 F., loses a large amount
of volatile matter on heating and becomes converted into a pitch

196

THE MODERN ASPHALT PAVEMENT.

FROM TAMESI RIVER, MEXICO.

SAMPLES COLLECTED AT DEPOSIT.

Test number

28075

28076

28077

28078

Bitumen soluble in CS a , air temp . .
Difference . .

99.5%
0.0

68.3%
6.8

59.7%
6.9

68.1%

Inorganic or mineral matter

24 9

33 4

29 5

Loss, 230 F., until dry
Loss, 325 F., 7 hours additional. .
Loss, 400 F., 5 hours additional. .

2.83%
17.30
8.42

13.50%
7.46
3.36

4.37%
6.34
2.42

9.66%
13.75
9.68

Residue after 325 and 400

Pitch

Penetration of original material
at 78 F

in all cases. The instability of the material, as revealed by this
fact, would necessitate the heating of this bitumen until all the
volatile portion was removed before it could be used for paving
purposes satisfactorily. For this reason, as well as on account
of its great lack of uniformity and the small extent of the avail-
able supply, it will not, probably, play a very important part in
the asphalt paving industry.

Deposits at Chijol. At a locality known as Chijol, 25 miles
from Tampico, on the Mex. Cent. R.R., and 3 miles distant from
the latter, asphalt effusions have been worked to a limited extent.
A sample of this material has the following characteristics:

TEST NO. 28082.

Loss, 250 F., until dry 12.70%

Penetration at 78 F. (original substance) 66

DRY SUBSTANCE.

Loss, 325 F., 7 hours 13.42%

Residue after heating Pitch

Loss, 400 F., 5 hours 7.24%

Residue after heating Pitch

Bitumen soluble in CS 2 , air temperature. ... 91 . 1%

Difference 1.7

Inorganic or mineral matter 7.2

100.0

INDIVIDUAL ASPHALTS.

197

FROM TAMESI RIVER, MEXICO.

SAMPLES SHIPPED TO NEW YORK.

44312

51471

51470

Specific gravity, 78 F /78 F (original)

1 211

1 0385

*" ' ' ' ' " (dry) ...'..

1 118

Flashes, F

308 F

Loss, 212 F., until dry

15.0%

20.1%

10 0%

Loss on refining water

15 o%

20 1%

1 ' " ' ' impurities

22 2

Total loss

42 3%

Penetration of refined substance at 78 F. . .

REFINED SUBSTANCE.

Loss, 325 F., 7 hours .

1 5%

1 8%

4 8%

Penetration of residue at 78 F

Loss, 400 F., 7 hours (fresh sample)

4 3%

5 5%

8 9%

Penetration of residue at 78 F

Pitch

Loss, 325 F., 21 hours. .

3 4%

it no <

10 4%

Loss, 400 F , 21 hours .

7 4%

" " 28 "

16 9%

Bitumen soluble in CS a , air temperature ....
T^ifference ...

89.1%
1 8

71.5%
8 3

99.0%

Inorganic or mineral matter

9 1

20 2

Mai t henes :
Bitumen soluble in 88 naphtha, air
temperature

100.0
49 5%

100.0

48 8%

100.0
73 9<

This is per cent of total bitumen soluble. .
Bitumen soluble in 62 naphtha

55.6
57 0%

68.2
56 2%

74.6
83 8%

This is per cent of total bitumen soluble. .

Bitumen yields on ignition:
Fixed carbon

64.0
16 1%

78.7
10 4%

84.6
12 6%

It will be noted that this bitumen is very similar to the purer
form of that found along the Tamesi River; that is to say, it loses
large quantities of volatile matter on heating and becomes con-
verted into a pitch. For the same reasons, as in the case of the
previous bitumen, it will not prove of any importance in the paving

198 THE MODERN ASPHALT PAVEMENT.

industry, although no doubt a certain proportion of both of these
materials could be incorporated with other and more satisfactory
asphalts if it were a matter of economy to do so.

Deposits in the Neighborhood of Tuxpan. 54 miles from Tuxpan
and 9 miles from the Tuxpan River are found large effusions of
asphalt, identified under the name of the Santa Theresa deposits,
attempts to develop which have been made for many years, and
by many individuals, and with but little success from a com-
mercial point of view.

A sample of the material examined in the author's laboratory
had the following characteristics:

FROM DEPOSIT AT TUXPAN, MEXICO.

TEST No. 28083.
Loss, 230 F., until dry 15.30%

Penetration at 78 F. (original substance) 76

DRY SUBSTANCE.

Loss, 325 F., 7 hours 12.48%

Residue after heating Pitch

Loss, 400 F., 5 hours additional 6 . 84%

Residue after heating Pitch

Bitumen soluble in CS 2 , air temperature ... 90 . 3%

Difference 3.1

Inorganic or mineral matter 6.6

100.0

This bitumen is, it will be seen, quite similar to those found
near Tampico.

Deposits at Chapapote. Effusions of asphalt which are iden-
tified under the above name occur 15 miles from Timberdar, the
head of navigation of the Tuxpan River. These deposits have
been worked by the Mexcian Asphalt Company, who packed the
material in bags and made an effort to float it down the river.
Some of the bitumen thus exported was examined by the author,
giving the results tabulated on page 199.

From these data it appears that both soft and hard bitu-
men are found on the Chapapote Ranch, but that the former
hardens very rapidly on heating, like other Mexican bitumens,

INDIVIDUAL ASPHALTS.
FROM DEPOSIT AT CHAPAPOTE, MEXICO.

199

42226

28080

Specific gravity at 78 F./78 F. (original) . ,

1 0343

' " " " (dry)..

1.045

Color

Black

Lustre .

Shining

Structure . .

Massive

Fracture

Conchoidal

2 +

Readily

Softens . ....

258 F

Flows.

272 F

Penetration at 78 F. (dry)

140

Loss, 220 F., 2 hours

11.4%

DRY SUBSTANCE.

Loss 325 F , 7 hours . .

4 7%

Residue after heating, penetration at 78 F

Loss 400 F 7 hours (fresh sample)

13 0%

Residue after heating, penetration at 78 F.

Bitumen soluble in CS 2 ^ air temperature

99 0%

99 2%

Inorganic or mineral matter. . . .

Malthenes :
Bitumen soluble in 88 napntha, air temperature
This is per cent of total bitumen

100.0

74.1%
74 8

100.0

Bitumen soluble in 62 naphtha .

83 5%

This is per cent of total bitumen

84 3

Bitumen yields on ignition :
Fixed carbon

13 3%

21 5%

and becomes converted into a pitch. The material of this descrip-
tion which has been imported into the United States has been
used by mixing it with other more desirable asphalts, or, where
used alone, has made a more or less unsatisfactory pavement.

Malthas and solid bitumens from various other deposits in
Mexico have been examined in the author's laboratory but the
preceding are sufficient to illustrate the general character of the
material found in that country. From information available it
hardly seems possible that any of these deposits can ever furnish

200 THE MODERN ASPHALT PAVEMENT.

a reliable commercial supply. They have been mentioned in this
place merely to bring out this fact and to show the character of
the material that is available.

La Patera, California, Asphalt. In Santa Barbara County, Cali-
fornia, and about 9J miles in an air-line west of the City of Santa
Barbara, a vein or intrusion of asphalt in the shales of that neigh-
borhood was worked for several years in the early nineties. Its
geologic environment has been described by Eldridge. 1
The material was used in the production of an asphalt cement which
attracted much attention at that time, and was known as Alca-
traz XX. Although the vein is now exhausted and the mine
abandoned the bitumen is of some interest as being typical and
illustrative of the hardest type of asphalt. In its best days it
never yielded more than 70 tons in a day and generally not more
than 30 or 40, the entire production in the 5 years that it was
worked being less than 30,000 tons.

La Patera crude asphalt is a mixture of bitumen with the min-
eral matter of the adjoining shale, which is composed of sand
and clay. Its fracture resembles in some respects Trinidad lake
asphalt, the material being filled with small gas cavities due to
the imprisonment of gas which has been evolved at an early stage
in the existence of the material in the same way that takes place
in Trinidad pitch. It differs from the latter in being very hard
and brittle, not softening below 250 F. Material taken from
the mine in 1894 contained 59 per cent of bitumen but in 1896
this fell to 55 per cent and in 1897 it had fallen to 49 per cent.
The latter material had the physical properties and proximate
composition given in the table on page 201.

From these figures it appears that the density corresponds to
the percentage of mineral matter and bitumen which it contains.
As has already been said, the softening point is very high. It of
course, being so hard a material, loses but little on heating for a
length of time at high temperature. The percentage of malthenes
is, of course, very low and corresponds to that found in the Cuban
asphalt. The percentage of hydrocarbons unacted on by sulphuric
acid is low, lower even than in the Bejucal, Cuban, bitumen, and
1 The Asphalt and Bituminous Rock Deposits of the U. S., 1901, 442.

INDIVIDUAL ASPHALTS. 201

LA PATERA, CALIFORNIA, ASPHALT.

Test number 13541

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 . 3808

Color of powder or streak Black

Lustre Dull

Structure Uniform

Fracture Irregular

Hardness, original substance 2

Odor Asphalti*

Softens 260 P

Flows 300 P

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance :

Loss, 325 F., 7 hours 1.5%

Character of residue , Shrunke*

Loss, 400 F., 7 hours (fresh sample) 2.5%

Character of residue Shrunken

Bitumen soluble in CS a , air temperature 49.3%

Difference 2.1

Inorganic or mineral matter 48.6

100.0

Malthenes:

Bitumen soluble in 88 naphtha, air temperature 21 .6%

This is per cent of total bitumen 43 . 8

Per cent of soluble bitumen removed by H^Oj 81 . 4

Per cent of total bitumen as saturated hydrocarbons 8.1

Bitumen soluble in 62 naphtha 26.7%

This is per cent of total bitumen 54 . 1

Carbenes:

Bitumen more soluble in carbon tetrachloride, air temperature 1 . 7%

Bitumen yields on ignition :

Fixed carbon 14.9%

Sulphur 6.2%

the lowest found in any asphalt. It yields about 15 per cent of
fixed carbon on ignition and is, therefore, a true asphalt and in no

202 THE MODERN ASPHALT PAVEMENT.

way allied to the grahamites. The percentage of sulphur which
it contains is the same as that found in Trinidad asphalt.

An asphalt of this description can only be used in combination
with a dense residuum of an asphaltic petroleum. In the early
days of the Alcatraz Company attempts were made to produce a
paving cement by combining La Patera asphalt with the natural
maltha found in the Carpinteria sands. Pavements made with
this material went to pieces very rapidly, and it is not difficult,
in the light of our present knowledge, to explain why this was so.
The Carpinteria maltha hardens and becomes a pitch very rapidly
on heating, with the result that it is impossible to guarantee that
an asphalt cement made with it should have a proper consistency
in an asphalt surface. Later on a heavy asphaltic petroleum
residuum obtained from a petroleum produced at Summerland
was used with much more satisfactory results, and some excellent
pavements were laid with a cement prepared in this way, but
the proportion of the La Patera asphalt to the oil, 40 to 60, was
so small that it could be better regarded as an amendment to
qualities lacking in the oil rather than as an asphaltic cement per se.

Asphalt on the More Ranch, Santa Barbara County, California.
This deposit of asphalt is of importance not on account of its size,
but because the addition of perhaps a shovelful of it to each barrel
of residual pitch from California petroleum has been used as a
basis for the statement that the latter contains a native solid
bitumen. The deposit is found on the seashore about 6 miles
to the west of Santa Barbara. It has been described by Mr. J. D.
Whitney in his report on the Geological Survey of California,
Geology, I, 132, by Peckham in the American Jour, of Science,
(2), 48, 368, and by Eldridge.

The shore here consists of an exposed cliff, 75 to 80 feet high,
of sandy clay which is quite soft and easily weathered. It is
much fissured and in these fissures the asphalt is found either
in the shape of kidneys or veins. It can be seen at various points
along the face of the cliff, where it has been exposed by the action
of the waves. In places the wall rock is mixed in in fragments
with the bitumen and in others it is a homogeneous material.
The amount of bitumen found at any point is, therefore, very

INDIVIDUAL ASPHALTS.

203

variable, as can be seen from the accompanying analyses. A few
hundred tons have been taken out annually for many years and
sold along the Pacific Coast. When the author examined the
deposit in 1897 a kidney was being worked about 15 feet deep
and about 12 feet broad, which illustrated the appearances of the
material; larger masses than this are seldom found. It is evi-
dent, therefore, that the asphalt available at this point is not of
commercial importance. The composition of the material is as
follows:

ANALYSES OF ASPHALT FROM MORE RANCH, SANTA
BARBARA COUNTY, CALIFORNIA.

Test No. 13383. From mine, collected December, 1897.
" " 13536. Supply ready for shipment on wharf.
" " 13539. Stringer in tunnel from pit.

DRY SUBSTANCE.

13383

13536

13539

Bitumen by CS 2 air temperature . . .

38 3%

40 1%

48 ^^

3.4

1.4

*0 . O /0

1 2

58 3

58 5

50 5

Malthenes :
Per cent of total bitumen soluble in 88
naphtha air temperature

100.0
63 2

100.0
63 3

100.0
59 2

Loss of crude at 212 F., 1 hour

1.4%

It is evident that the asphalt contains too little bitumen to
melt readily without the aid of a flux. The pure bitumen extracted
from the asphalt is much harder than that obtained from Trinidad
lake asphalt, flowing but 69 per cent as far as the latter on a cor-
rugated plate at high temperature.

All attempts to' utilize this material, except locally, have been
made purely for advertising purposes in connection with the use
of residual pitches, where specifications demanded the use of a
native solid bitumen.

Standard Asphalt. In the western part of Kern County, in
the first tier of foot-hills on the coast range, forming the western
boundary of the central valley of California, at a point called

204

THE MODERN ASPHALT PAVEMENT.

Asphalto, at the end of a branch of the Southern Pacific Railroad
from Bakersfield, an asphalt mine was in existence in the nineties
which was worked by the Standard Asphalt Company of Cali-
fornia. The company originally endeavored to work certain
superficial overflows of bitumen upon the surface of the ground,
but the material proving to be of no value a shaft was sunk upon
a vein which penetrated the shales at this point, in the manner
which has been described by Eldridge. 1 Some of the material
was obtained also by running tunnels. It is of interest in this
place merely to determine the character of the bitumen which
was obtained.

The crude material was found in different degrees of purity
and containing from 54 to 91 per cent of bitumen, as appears
from the following analyses:

FROM DEPOSIT OF STANDARD ASPHALT COMPANY, CALIFORNIA.

Test number

13391

13589

13593

13594

Bitumen bv CS,, air temperature .
Difference

80.6%
7.7

No. 1
90.5%
0.0

No. 2

87.9%
3.1

54.3%
6.2

78.7%
4.0

Inorganic or mineral matter ....

11 7

9 5

39 5

17 3

Malthenes :
Bitumen soluble in 88 naphtha,
air temperature

100.0
46

100.0
49 8%

100.0

43 1%

100.0
31 0%

100.0
41 o%

This is per cent of total bitumen .
Loss of crude at 212 F for 1 hour

57.1

55.0

5.7%

49.0
14.9%

57.1

5.8%

52.8
5 9%

Bitumen yields on ignition:
Fixed carbon

7 3

The crude material was a compact homogeneous brownish
bitumen, very much resembling gilsonite in its outward appear-
ance, but being very much softer. Much of it, although showing
no outward evidence of so doing, contained an appreciable per
cent of water which, together with a certain amount of gas, is
evolved on heating to 100 C. In some cases the loss reached

*The Asphalt and Bituminous Rock Deposits of the United States, 1901,

449.

INDIVIDUAL ASPHALTS.

205

REFINED STANDARD ASPHALT, CALIFORNIA.

Test number. 13601

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, diy 1 .0627

Color of powder or streak. Black

Lustre Dull

Structure Uniform

Fracture Semi-conchoidal

Hardness Soft

Odor Asphalt ic

Softens 170 F.

Flows 180 F,

Penetration at 78 F. to 27

CHEMICAL CHARACTERISTICS.

Dry substance :

Loss, 325 F., 7 hours 6.6%

Character of residue Smooth

Loss, 400 F., 7 hours (fresh sample) 19.9%

Character of residue Blistered

Bitumen soluble in CS, air temperature 89 .8%

Difference 3.4

Inorganic or mineral matter. 6.8

100.0

Malt henes :

Bitumen soluble in 88 naphtha, air temperature 53 . 4%

This is per cent of total bitumen 59 . 4

Per cent of soluble bitumen removed by H.jSO 4 51 .9

Per cent of total bitumen as saturated hydrocarbons. .... 28.6

Bitumen soluble in 62 naphtha 60 .0%

This is per cent of total bitumen 66 .8

Carbenes :

Bitumen insoluble in carbon tetrachloride, ah* temperature . 3%

Bitumen yields on ignition:

Fixed carbon. 8.0%

as high as 16 per cent. The run of the mine would average above

80 per cent of bitumen with 10 per cent of mineral matter and

5 per cent of moisture and gas. This bitumen is more particu*

206 THE MODERN ASPHALT PAVEMENT.

larly characterized and differentiated from ordinary asphalts by
the fact that it yields only 7 to 8 per cent of fixed carbon, where
the asphalts and gilsonite contain 14 to 15 per cent.

For the purpose of preparing the crude material for use in
the paving industry it was melted with the addition of about 30
per cent of a dense asphaltic flux. The resulting product was
quite hard and was further fluxed for the purpose of making an
asphalt cement. This material was in use to a considerable extent
in the middle West before 1900, but the mine became exhausted
and it is no longer available.

The refined material had the characteristics tabulated on p. 205.

The refined Standard asphalt was a rather pure bitumen carry-
ing but 6.8 per cent of mineral matter with 90 per cent of bitu-
men. The percentage of malthenes was smaller than that found
in Trinidad and Bermudez asphalts, but the softening point was
lower than would have been expected in such a case. As has
been said the fixed carbon which this material yields is very low.
Except for this its general outward resemblance to gilsonite would
seem to point to the fact that the bitumen from the Standard
mine must be closely allied to it. It differs from it, however,
in that in gilsonite the percentage of the total bitumens present
as saturated hydrocarbons is very much smaller. This may be
due, however, to the fact that in gilsonite metamorphism has gone
much further than in the case of the bitumen from the Standard
mine. Although none of this bitumen is available for paving
purposes at the present day, its character has been shown
because of its uniform structure and resemblance in certain
respects to gilsonite.

Good pavements were constructed, with the Standard bitumen
where it was properly handled, but in many cases the surface
failed to give satisfaction owing to lack of skill in its use.

Other Deposits of Solid Bitumen in California. In addition
to the two abandoned deposits of asphalt which have been described,
namely, those at the La Patera and Standard mines, there are
numerous others scattered throughout Lower California, descrip-
tions of which will be found in the Eldridge report on "The Asphalt
and Bituminous Rock Deposits of the United States." None of

INDIVIDUAL ASPHALTS. 207

them have proved, although development has been attempted in
many cases, to be of the slightest commercial importance, as the
material available at any one point is too small to pay for mining
it, owing to the fact that the bitumen is found in fissures or veins
in the shales, which always pinch out at a very moderate depth,
due to the pressure exerted by the superimposed strata, and is
often mixed with such a large proportion of the mineral matter
from the vein walls, at times in the shape of brecciated masses
scattered through the bitumen, that it is extremely difficult to
handle. The only interest to the paving industry hi these deposits
lies in the fact that minute percentages of them have at times
been added to the solid residues from asphaltic oils hi order to
substantiate the claim that the latter contained native solid bitu-
mens. As a paving material none of them has ever amounted
to anything nor will any of them ever do so.

SUMMARY.

In the preceding pages the characteristics of the asphalts which
are or have been available to any commercial extent are given.
The supply of Trinidad asphalt is extremely large in amount,
uniform in character, and much more stable than any other, owing
to the character of the hydrocarbons of which it is composed.
Bermudez asphalt, for the same reason, is much more liable to
change. Maracaibo asphalt differs essentially from all others in
several respects. The data in regard to many minor deposits
illustrate the very considerable variation which occurs in material
included under the specific designation asphalt.

CHAPTER XI.

SOLID NATIVE BITUMENS WHICH ARE NOT ASPHALT.

IT appears in our classification of native bitumens that several
solid native bitumens exist which, from their peculiar character-
istics, cannot be included among the asphalts. These include
gilsonite, grahamite, manjak, and glance pitch. The two former
are t-he only ones which are of any interest in the paving industry.

Gilsonite. The occurrence of gilsonite in Utah and Colorado
is thoroughly described in the report of Eldridge, which has been
frequently referred to. It is only necessary in the present place
to consider its physical properties and proximate chemical com-
position.

Gilsonite is one of, if not the purest, forms of solid bitumen
found in nature. It is practically entirely soluble in carbon di-
sulphide. It varies to a certain extent in hardness in the various
deposits and veins in which it occurs. The extremes in composi-
tion in the present commercial supply appear in the following
table.

GILSONITE.

PHYSICAL PROPERTIES.

Test number

92603

80343

Specific gravity, 78 F./78 F., original sub-
stance dry

1.044

1 049

Color of powder or streak . .

Brown

Lustre

Lustrous

Structure

Homogeneous

P racture

Sub-conchoidal

ilnrdness original substance .

Softens

260 F.

300 F.

Flows

275 F.

325 F

Penetration at 78 F

206

SOLID NATIVE BITUMENS, NOT ASPHALT.

CHEMICAL CHARACTERISTICS.

209

Dry substance:
Loss 325 F , 7 hours

.9%

2.3%

Smooth

Intumesces

Loss 400 F , 7 hours

1.2

4.0

Smooth

Intumesces

Bitumen soluble in CSj, air temperature ....
Inorganic or mineral matter

99.0%
.0

99.9%
.1

1.0

Malthenes:
Bitumen soluble in 88 naphtha, air temp.
This is per cent of total bitumen

100.0

47.2
47.7

100.0
15.9%
15.9

Per cent of soluble bitumen removed by
H 2 SO 4

87.7

71.8

Per cent of total bitumen as saturated
hydrocarbons .

5.9

4.5

67.4%

30.3%

This is per cent of total bitumen . . .

68.1

30.4

Carbenes:
Bitumen insoluble in carbon tetrachloride,

.0%

Bitumen yields on ignition:
Fixed carbon .

13.0%

13.4%

Gilsonite is known in the trade in two forms, firsts and seconds,
the firsts being the highest-grade material in large lumps unac-
companied by powder, while the seconds are that part of the
mineral which occurs nearest the vein walls, and in fragmentary
particles and are made up of the less attractive product of the
mine. The difference in these two grades of gilsonite is one largely
of appearance rather than quality when taken from the same vein.

Gilsonite is the purest native bitumen with which we are
acquainted, the best varieties containing 99.5 per cent of bitumen,
and with but traces of matter not of a bituminous nature. The
bitumen is equally soluble in cold carbon tetrachloride and carbon
disulphide, thus differentiating it from grahamite and some of
the residual pitches.

Gilsonite is more variable when taken from different deposits
and at different depths.

210 THE MODERN ASPHALT PAVEMENT.

The density of this bitumen is somewhat smaller than that
of the asphalts and it has a much higher softening point, as might
be expected from the fact that it is brittle and readily reduced
to a reddish-brown powder, the latter characteristic alone differ-
entiating it from the other native bitumens which give a
much blacker powder. As would be expected in such a brittle
material the percentage of malthenes is low, only 48 per cent
of the total bitumen being soluble in 88 naphtha. The amount
will vary, however, in gilsonite from different veins and from
different parts of the vein, weathered material at times contain-
ing but 14 per cent, while in the best it may rise to over 70 per
cent.

The hydrocarbons composing the malthenes of gilsonite are
entirely different in character from those found in the asphalts.
They are almost entirely composed of unsaturated hydrocarbons
attacked by strong sulphurs acid, and this fact differentiates gil-
sonite completely from asphalt. The hydrocarbons unattacked by
dilute sulphuric acid are extremely viscous, sticky, and resinous
and absolutely different from those found in any other native bitu-
men, and there seems to be good ground for the inference that the
other hydrocarbons composing gilsonite are likewise quite different
from those occurring in the asphalts. A close study of these
hydrocarbons will be of great interest, but our information at
present available is sufficient to justify us in placing gilsonite
in a class by itself among the native bitumens. Gilsonite is char-
acterized by yielding the same percentage of fixed carbon on
ignition that is found in the asphalts. This is not what would be
expected from a consideration of the proximate composition of
the material, which would lead us to suppose that the percentage
would be higher. In material which is much weathered a higher
percentage is actually found, reaching in one instance 26 per
cent, and corresponding, of course, to a smaller percentage of
naphtha soluble bitumen in the material, although the relation
between fixed carbon and malthenes is by no means a constant
one. Gilsonite is readily soluble in the heavy asphaltic residues
from California and Texas petroleums and, when mixed with

SOLID NATIVE BITUMENS, NOT ASPHALT. 21 1

this in the proper proportion, makes a material which is extremely
rubbery and more or less elastic and ductile.

Gilsonite in the Paving Industry. Gilsonite has been used
very successfully and to a very considerable extent in the paving
industry, the peculiar rubbery nature of the cement produced
from it with a heavy asphaltic flux, and the fact that this cement
is less susceptible to temperature changes than those made with
the ordinary asphalts, making it a very desirable material. It
was not available for use, at least in the East, until the advent
of the Texas asphaltic flux in 1902, as the paraffine fluxes will
not cut Gilsonite satisfactorily, the result being a very short
and granular substance which has no cementing power, while
the California fluxes were too expensive. Gilsonite has wide
applications in many industries. Successful pavements have been
laid with it which have been in use for years, and it is an impor-
tant constituent of varnishes, insulations, and waterproofing com-
pounds.

Grahamite. Grahamite is a brittle black bitumen, rarely of
compact structure, which does not melt readily but merely intu-
mesces on heating to high temperatures. It occurs hi veins rather
widely disseminated, but never in large amounts. Its physical
properties and chemical constitution differentiate it from all
other solid bitumens. Its structure is distinguished by what
has been called a hackly or pencillated fracture produced appar-
ently by the working of the brittle bitumen, induced by the move-
ment of the vein wall. At times there is a grosser columnar
structure. As types of this material, that found in Oklahoma
in the Ten Mile Creek district, in the Choctaw Nation, and in
Ritchie County, West Virginia, where it was originally discovered,
will serve. These characteristics are shown by the analyses
given on page 212.

Grahamite is an almost entirely pure bitumen soluble in carbon
disulphide, naphthalene, and dead oil, but it is differentiated from
the asphalts and gilsonite by the fact that it is almost entirely insol-
uble in naphtha, even of 62 density, and to the extent of 55.0 per

212

THE MODERN ASPHALT PAVEMENT.
GRAHAMITE.

Test number

68940

75637

Location

West

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original sub-
stance dry

1.171

Virginia
1.137

Color of powder or streak

Black

Dull

Uniform

Fracture

Hackly

friable
Irregular

Hardness original substance

Brittle

Odor

None

Softens

Intumesces

t (

< <

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance:
Loss 325 F 7 hours

+ .1%

Loss 400 F 7 hours (fresh sample)

+ .5%

Bitumen soluble in CS, air temperature .

94 1%

97.8%

Difference

Inorganic or mineral matter

5 7

2.1

Malthenes:
Bitumen soluble in 88 naphtha, air temp
This is per cent of total bitumen

100.0

.4%
.4

100.0

3.3%
3.37

Per cent of soluble bitumen removed by
H^SO

25.0

Per cent of total bitumen as saturated
hydrocarbons

.32

Bitumen soluble in 62 naphtha

.7%

3.4%

This is per cent of total bitumen

Carbenes :
Bitumen insoluble in carbon tetrachloride,
air temperature . .

68.7%

3.47
55.0%

Bitumen insoluble in hot carbon tetra-
chloride

48.6

1.3%

Bitumen yields on ignition:

53.3%

41.0%

Ultimate composition:

86.56%

8.68

1.79

2.97

100.00

SOLID NATIVE BITUMENS, NOT ASPHALT. 213

cent to 80.6 per cent in cold carbon tetraehloride. It is also differ-
entiated from the asphalts and gilsonites by the fact that it yields
from 30 to 50 per cent of fixed carbon on ignition. Grahamite,
although not soluble in the lighter oils, is readily dissolved by
the denser or semi-asphaltic fluxes, and in this condition forms a
rubbery material quite similar to that produced in the same way
with gilsonite. A small cylinder of it when bent upon itself will
rapidly return to its original form. It lacks ductility, that is to
say, it is very short when a cylinder of it is drawn out.

A similar deposit of grahamite is found in Middle Park, Colo-
rado. This material is inaccessible and of no commercial impor-
tance. It has the composition given on page 214.

It is apparent from the results of the analyses of the 3 gra-
hamites that the different deposits vary in the degree to which
the molecule has been condensed, as shown by the percentage
of bitumen insoluble hi cold carbon tetraehloride.

It has also been found that grahamite can be divided into
two classes, those containing sulphur and those containing oxygen.

As has been said, numerous deposits of grahamite are found
in Cuba, Mexico, Trinidad, and elsewhere, but they are of no
commercial importance as far as the asphalt paving industry is
concerned, although of great interest from a purely scientific
point of view. These will be described by the writer in another
place.

Grahamite has a number of industrial uses. Successful pave-
ments have been constructed with it in combination with gilsonite
and heavy asphaltic fluxes. Mastic made with it is more resistant
to grease and oil than the ordinary type, and it has been made
a constituent of varnishes, of rubber substitutes, and of filler
for brick and stone blocks. When cut with heavy asphaltic flux
it is even more rubbery and elastic than gilsonite under the same
circumstances, and less susceptible to heat.

Glance Pitch and Manjak. These bitumens are of little or
no interest in connection with the paving industry, but they must
be mentioned here in order to complete our description of the
solid native bitumens.

Glance pitch is a material which is quite widely distributed

214 THE MODERN ASPHALT PAVEMENT.

GRAHAMITE FROM MIDDLE PARK, COLORADO.
Test number 19162

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 .160

Color of powder or streak Black

Lustre Dull

Structure Uniform-
homogeneous

Fracture Columnar,

scaly

Hardness, original substance 3

Odor .* . None

Softens Intumesces

Flows

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 , air temperature 98 .2%

Difference 1.7

Inorganic or mineral matter .1

100.0

Malthenes:

Per cent total bitumen soluble in 88 naphtha, air tem-
perature. .8%

Carbenes:

Bitumen insoluble in carbon tetrachloride, air temperature 80 . 6%

Bitumen yields on ignition'

Fixed carbon 47.4%

Ultimate composition:

Carbon 85.97%

Hydrogen 7.65

Sulphur .93

Difference (oxygen?) 5. 45

100.00

over the world, although the best supplies come from the East,
Syria, and the Dead Sea.

Manjak is found only in the island of Barbadoes. A bitumen
is shipped from Trinidad under the name of manjak, but this

SOLID NATIVE BITUMENS, NOT ASPHALT. 215

material is really a grahamite and not a true manjak, as it does
not melt and has all the properties of the latter bitumen.

The characteristics of these materials are shown by the follow-
ing analyses:

EGYPTIAN GLANCE PITCH.

Test number 14145

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 .097

Color of powder or streak Black

Lustre Lustrous

Structure Brittle-
uniform

Fracture Conchoidal

Hardness, original substance 2

Softens 250 F.

Flows 260 F.

Penetration at 78 F

CHEMICAL CHARACTERISTICS

Bitumen soluble in CS 2 , air temperature 99 . 7%

Difference .2

Inorganic or mineral matter. .1

100.0

Malthenes:

Bitumen soluble in 88 naphtha, air temperature 23 . 5%

This is per cent of total bitumen 23.6

Per cent of soluble bitumen removed by H^Of 72 .

Bitumen soluble in 62 naphtha 36.9%

This is per cent of total bitumen 37.0

Carbenes :

Bitumen insoluble in carbon tetrachloride, air temperature 0.1%

Bitumen yields on ignition:

Fixed carbon 15.0%

Ultimate composition:

Sulphur 8.52%

Carbon 80.87

Hydrogen 10.42

Nitrogen .19

100.00

216 THE MODERN ASPHALT PAVEMENT.

BARBADOES MANJAK.
Test number 14143

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 .0844

Color of powder or streak Dark brown

Lustre Lustrous

Structure Uniform

Fracture Conchoidal

Hardness, original substance 1

Softens 230 F.

Flows 250 F.

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 , air temperature 99 . 2%

Difference .5

Inorganic or mineral matter .3

100.0

Malthenes:

Bitumen soluble in 88 naphtha, air temperature 26 .9%

This is per cent of total bitumen 27.0

Per cent of soluble bitumen removed by HjSO 4 75 .0

Bitumen soluble in 62 naphtha 40 . 4%

This is per cent of total bitumen 40 . 7

Carbenes :

Bitumen insoluble in carbon tetrachloride, air temperature 1.2%

Bitumen yields on ignition :

Fixed carbon 25.0

It will be noted that glance pitch is a very brittle material,
of a higher density and much higher melting-point than asphalt,
of great purity and containing but a very small percentage of
malthenes. It has evidently originated in the very complete
hardening of asphalt either by natural causes or, exceptionally,
by its exposure to heat in one way or another.

Manjak resembles it in many respects, but is distinguished
from it by being more closely related to grahamite, on account
of the higher percentage of fixed carbon which it contains, that
obtained from glance pitch being an amount normal to asphalt,

SOLID NATIVE BITUMENS, NOT ASPHALT. 217

while that from manjak approaches that obtained from grahamite.
It is, however, differentiated from grahamite by the fact that it
actually melts, instead of intumescing only, and dissolves com-
pletely in cold carbon tetrachloride.

In both of these solid bitumens there is a very small propor-
tion of stable hydrocarbons unattacked by sulphuric acid.

Ozocerite. Ozocerite is a solid bitumen the principal supply
of which is found hi Galicia. A small amount of it is also found
in Utah, in Emery and Uintah Counties. The hydrocarbons of
which it is composed are solids, resembling paraffine scale. When
purified it is known as ceresin and is used for the adulteration of
beeswax, and as a substitute for paraffine scale, which it is superior
to on account of its high melting-point. As it is a paraffine com-
pound it is of no interest in the paving industry.

PYROBITUMENS.

Albertite. Albertite is an extremely brittle and lustrous
material. It was first described by Wetherill. 1 It " occurs fill-
ing an irregular fissure in rocks of the Subcarboniferous Age in
Nova Scotia." It has since been found in Cuba, Mexico, Okla-
homa, and Utah. It is not a true bitumen, but a very small
part of it being soluble in the usual solvents for that substance.
It yields a very high percentage of fixed carbon. Analyses of
albertites from various localities which have been examined in
the author's laboratory are tabulated on pages 218, 219.

Its ultimate composition is, for the Nova Scotia material,

Carbon 85 . 53%

Hydrogen 13.20

Sulphur 1 .20

Nitrogen 42

Albertite is usually quite free from mineral matter, but in the
case of that found in Mexico it contains 22.6 per cent and a rather
larger portion of bitumen soluble in carbon disulphide than in
that found elsewhere.

The material is of no importance in the paving industry.

1 Trans. Am. Phil. Soc., Phila., 1852, 353.

218

THE MODERN ASPHALT PAVEMENT.

ANALYSES OF

Nova
61486

Scotia
7834

1.075
Black

Lustrous

Homogeneous
Smooth

Test number

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original sub-
stance, dry

Color

Black

Lustrous

Homogeneous
Smooth

Lustre

Structure

Odor

None
Intumesces

Softens

Flows

Penetration at 78 F . ...

CHEMICAL CHARACTERISTICS.

Original substance :
Loss, 212 F., until dry

Dry substance :
Bitumen soluble in CS 2 , air temperature. . . .
Difference

9.0%
91.0
.2

100.2

5.9%
94.1
.0

100.0
1.5%

Inorganic or mineral matter

Malthenes:
Bitumen soluble in 88 naphtha, air tem-
perature

This is per cent of total bitumen

Bitumen yields on ignition :

39.0%

29.8%
1-2%

Sulphur

Wurtzilite. Wurtzilite is a hard lustrous pyrobitumen,
slightly elastic in thin fragments, which is found in Uintah County,
Utah. It does not fuse at high temperatures, but a process has
been devised for fluxing it with heavy malthas by gradually crack-
ing it at high temperatures. It is practically insoluble in car-
bon disulphide and heavy residuum.

SOLID NATIVE BITUMENS, NOT ASPHALT.
ALBERTITE.

219

Utah.

Mexico.

Cuba.

Oklahoma.

19187

30495

36326

62989

74995

Bright

Dull

sample

1.092

1.099

1.204

Black

Brown-

Black

black

Partly
lustrous

Dull

Lustrous

Homogeneous

Irregular

Semi-
conchoidal

Brittle

None

Does not

Intumesces

intumesce

- O o

2.25%

.2%

.1%

5.6%

3.4%

11.9%

* /%j

Trace

1.6%

6.8%

94.2

96.4

61.9

98.9%

87.7

71.2

26.2

1.1

10.7

22.0

100.0

Trace

3 2%

.0%

** /o
23.4

v* /%}

37.0%

40.4%

39.0%

53.0%

33.6%

54.2%

1.06

The amount available is too small to make it of any impor-
tance in the paving industry, and this and the preceding pyro-
bitumen have been merely mentioned to complete our illustra-
tion of the various types which are found in nature. Some deter-
minations of its characteristics resulted as follows:

220

THE MODERN ASPHALT PAVEMENT.
WURTZILITE.

Test number.

15270

31724

72684

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original

substance, dry

1 0544

1 0490

1 0639

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 , air temperature

12 8%

6 7%

Bitumen yields on ignition :
Fixed carbon

8 8%

5 2%

8.3%

SUMMARY.

The several solid native bitumens which are not asphalts are
shown to have interesting characteristics and some valuable
properties.

Grahamite, like gilsonite, can be fluxed with asphaltic oils at
very high temperature and in such form makes a very rubbery
material of value in the paving industry as well as for paint and
varnish and for waterproofing. The use of Grahamite in the
construction of pavements is limited by the fact that it may be
utilized only at high temperatures with certain modifications and
that the supply is extremely small.

The other solid bitumens are of interest only in connection
with the manufacture of varnishes and for insulating purposes;
they do not offer inducements towards introducing them into
the paving industry.

CHAPTER XII.
ASPHALTIC SANDS AND LIMESTONES.

Kentucky. Sands are found in Carter and Boyd Counties in
the northeastern part of Kentucky and in the counties of Brecken-
ridge, Grayson, Edmonson, Warren, and Logan in the western
part of the State. The geological relations of these sands and
the manner of their occurrence is described in great detail by
Eldridge. 1

In the present place it will only be necessary to show the nature
of the sands and that of the bitumen with which they are impreg-
nated in order to determine their availability for paving purposes.

In a general way it may be said that the sands are all com-
posed of loose grains which fall to pieces on the extraction of the
bitumen and are in no case sandstone. The bitumen impregna-
ting the sand is not a solid one, but consists of a maltha which
pulls out to a long thread at ordinary temperature or, in rare
instances, after extraction has a penetration as low as 60. The
bitumen from the Green River sand which was extracted with
carbon disulphide had the following characteristics:

BITUMEN EXTRACTED FROM BITUMINOUS SAND FROM THE
GREEN RIVER DEPOSIT, KENTUCKY.

Penetraton at 78 F 60

Per cent of total bitumen soluble in 88 naphtha .... 65 . 4
Per cent of soluble bitumen acted upon by HjSO 4 . . 15.0

Bitumen contains soft paraffines 2 . 6%

Yields on ignition :

Fixed carbon 15 .C%

It appears from the preceding figures that the bitumen of the
Kentucky sands is semi-asphaltic in that it yields an amount ef

1 The Asphalt and Bituminous Rock Deposits of the United States, 1901.

221

222

THE MODERN ASPHALT PAVEMENT.

fixed carbon corresponding to that found in the asphalts, but at
the same time contains some soft paraffine scale, and is largely
made up of stable hydrocarbons which are not attacked by sul-
phuric acid. As a rule, the bitumen is, however, too soft to be
suitable for use as a paving cement until the volatile oils have
been driven off by heating. When the bitumen is heated, how-
ever, it is, as in the case of the California Carpinteria sands, rapidly
converted into a hard pitch. The Kentucky sands contain on the
average about 6.5 per cent of bitumen of the nature which has been
described. In exceptional cases it reaches 13 per cent. The
characteristics of the sands in particular localities are as follows:

Carter County. A deposit of bituminous sand occurs on Soldier
Creek in Carter County, as described by Eldridge. Samples of
this sand analyzed in the author's laboratory in 1898 and 1900
had the following composition:

BITUMINOUS SAND FROM SOLDIER CREEK, CARTER COUNTY,

KENTUCKY.

Test number

10681

33896

Bitumen soluble in

cs, .

1898

8 2%

1900
9 1%

Passing 200-mesh s

ieve

4.5

3 9

" 100-

1 1

21.2

35.0

" 80-

" 50-

27 6

" 40-

3 5

" 30-

2 8

20-

1.2

0.0

100.0

The extracted bitumen is a soft maltha flowing slowly at 78 F.,
and hardening rapidly on heating, with a loss of 12 per cent.

Breckenridge County. The sandstones of Breckenridge County
are worked by the Breckenridge Asphalt Company and lie, accord-
ing to Eldridge, two miles south of Garfield, in a bed 14 feet thick,
the lower 7 or 8 feet being much more enriched than the upper
portion. Specimens examined in the author's laboratory had
the following composition:

ASPHALT1C SANDS AND LIMESTONES.

223

BITUMINOUS SAND FROM DEPOSIT OF BRECKENRIDGE AS-
PHALT COMPANY, BRECKENRIDGE COUNTY, KENTUCKY.

TEST No. 9264.

Richer Portion.

Poorer Portion.

Bitumen soluble in CS 2
Passing 200-mesh sieve . .

7.7%
7.2

4.3%
10.6

" 100- " ' '

26.6

80- " "

26.0

" 50- " "

tt 4Q_ H tt

29.4
2.5

ft QA It tt

0.4

tt 20- " "

0.2

100.0

The extracted bitumen is a soft maltha hardening on heating,
as is the case with bitumens from other Kentucky sands.

Grayson County. Asphaltic sands are found in Grayson County
in the neighborhood of Leitchfield, which Eldridge describes under
the designation " Schillinger Prospects " and " Breyfogle Quarries."
At the latter point sands impregnated with bitumen and seepages
of a gummy consistency are found. The sands are of varying
degrees of richness and the seepages of different degrees of hard-
ness. The results of analyses of specimens of the materials which
were formerly mined at this point are given in the following tables
and in one on p. 225:

ASPHALTIC SANDS, GRAYSON COUNTY, KENTUCKY.

Test number

41187

41277

41188

49577

Bitumen soluble in CS 2

10%

13%

13.7%

Passing 200-mesh sieve

4.3

' < 100- ' '

4.0

" 80- " ....

6.0

" 50- "

34.0

" 40- "

24.0

" 30- "

8.0

" 20- "

3.0

" 10- " ....

3.0

Retained on 10-mesh sieve ....

100

100.0

224

THE MODERN ASPHALT PAVEMENT.

BITUMEN IMPREGNATING MINERAL MATTER, GRAYSON
COUNTY, KENTUCKY.

Test number

41189

41190

Specific gravity, 78 F./78 9 F

1.282

1.769

Color

Black

Lustre

Shining

Structure

Massive

Fracture

Irregular

Hardness

Odor

Asphaltic

Loss 220 F 1 hour .

7.0%

1.5%

Bitumen soluble in CS 2 , air temperature
Difference

62 9 1 %

30.0%
4 9

iziorganic or mineral matter

28 4

65 1

Malthenes:
Bitumen soluble 88 naphtha, air tempera-
ture

100.0
50.8%

100.0
21.9%

This is per cent of total bitumen

70 3

Bitumen soluble 62 naphtha, air tempera-
ture

54 3%

29 7%

This is per cent of total bitumen

87.0

fiitumen yields on ignition:
Fixed carbon

12 0%

Penetration of extracted bitumen at 78 F. .

It will be noticed that in the case of the sand some of it is
quite rich in bitumen and other parts of it quite poor. The sand
grains are extremely coarse, the majority of them being of 50-
and 40-mesh size in one instance, and larger than 30 in another.
Such a sand grading alone would make this material unsuitable
for use in an asphalt surface.

The mixture of loose mineral matter and asphalt contains a
large percentage of bitumen which is very hard in consistency
but is hardly asphaltic in nature, as the amount of fixed carbon
which it yields is only 12 per cent.

The seepages vary in consistency from that of a mere maltha
to that of a bitumen having a penetration of only 35. In the

ASPHALTIC SANDS AND LIMESTONES.

225

SEEPAGES, GRAYSON COUNTY, KENTUCKY.

41192

41277

Rnprifip trravitv 78 F /78 F

9783

Loss 220 F , 1 hour

26.8%

19.2%

Dry substance :
Loss 325 F , 7 hours

4.1%

' ' , 400 F. , " " (fresh sample)

12.4

Character of residue after 325 F

Too soft for

" " " " 400 F

penetration.
22

Bitumen soluble in CS 2 , air temperature

89.8%
0.3

88.6%
5.5

Inorganic or mineral matter

9.9

5.9

Malthenes:
Bitumen soluble in 88 naphtha, air tem-
perature

100.0

74 5%

100.0
54.0%

This is per cent of total bitumen

61.2

Bitumen soluble in 62 naphtha, air tem-
perature

81 0%

CO 8%

This is per cent of total bitumen

90 2

68 7

Bitumen yields on ignition :
Fixed carbon . . .

15.6%

Penetration of extracted bitumen at 78 F. . .

latter case the material seems to be more asphaltic, as it yields
15.6 per cent of fixed carbon.

A deposit identified to the author as being from Ferry's quarry,
in Grayson County, has the following characteristics:

BITUMINOUS SAND FROM FERRY'S QUARRY, GRAYSON

COUNTY, KENTUCKY.

TEST No. 7218.

Bitumen soluble in CS 2 7.7%

Passing 200-mesh sieve 17.7

" 100- " " 34.5

" 80- " " 36.4

" 50- " " . 3.7

100.0

EXTRACTED BITUMEN.

Bitumen soluble in 88 naphtha, air temperature 71 .4%

Loss, 400 F., 7 hours 18.6

Penetration at 78 F. . . 81

226

THE MODERN ASPHALT PAVEMENT.

This material is composed of an extremely fine sand, quite
different from that found at the Breyfogle quarry, and contains
between 7 and 8 per cent of bitumen, which, after heating for 7
hours at 400 F. ; loses 18.6 per cent. This bitumen, before heat-
ing, has a penetration of 81.

Edmonson County. Eldridge states that in Edmonson County
there are many deposits of asphaltic sandstone and small tar
springs, none of which he considers to be of any value. They have
not been examined in the author's laboratory.

Warren County. Along the Green River, in Warren County,
are deposits of bituminous sand which have been worked for
paving purposes, the characteristics of which are that they con-
sist of quartz sand impregnated with from 6 to 9 per cent of very
soft maltha.

At Youngs Ferry the Green River Asphalt Company has opened
a quarry the sand from which has the following compositions:

BITUMINOUS SAND FROM DEPOSIT OF GREEN RIVER ASPHALT
COMPANY, WARREN COUNTY, KENTUCKY.

Test number

61873

6.1%
14.9
48.0
19.0
12.0
0.0
0.0
0.0
0.0

27606

8.8%
11.3
10.0
23.0
37.1
8.0
.9
.9
.0

Bitumen solub
Passing 200-me
100-
80-
'" 50-
40-
30-
20- '
10- '

Loss, 212 F.,

einC

;sh sie

t (
i <

L houi

i
t

100.0

100.0
1.5%

EXTRACTED BITUMEN.

Penetration at 78 F Too soft, pulls to a thread

Fixed carbon 11.8%

The first column represents the average composition of that
which has been taken out for paving purposes. It is very evi-
dent that the small percentage of bitumen in this sand, aside

ASPHALTIC SANDS AND LIMESTONES. 227

from the fact that it is not of a strictly asphaltic nature, as it
yields but 11.8 per cent of fixed carbon, would make it impossible
to produce a satisfactory surface mixture with it without some
amendment.

Near Brownsville the Rock Creek Natural Asphalt Company
has worked another sand which is coarser than that found at
Youngs Ferry, as shown by the following analysis:

BITUMINOUS SANDSTONE NEAR BROWNSVILLE, KENTUCKY.
TEST No. 61363.

Bitumen soluble in CS 2 6.6%

Passing 200-mesh sieve 12.4

100-
80-
50-
40-
30-
20-
10-

7.0

3.0

18.0

30.0

13.0

6.0

4.0

100.0

This material, like that from the Youngs Ferry deposit, is
not sufficiently rich hi bitumen to make it of any value.

Oozings of this bitumen, collected in drill-holes, had the com-
position given in table on page 228 under test No. 40894.

The bitumen in the sands at this point is a very soft one,
having a penetration of 193. It hardens rapidly on heating,
the residue after 7 hours at 400 F. having a penetration of only
12. Such an unstable bitumen, aside from the unsatisfactory
grading of the sand, makes this sand unavailable for the produc-
tion of a satisfactory pavement.

Logan County. Eldridge states: "The deposits of bituminous
sandstone in Logan County lie in its northern half. ... A single
quarry of the Standard Asphalt Company has been opened about
5 miles northeast of Russellville."

The average material from the base of the quarry has the com-
position given in table on page 228, test No. 19938.

The bitumen which oozes from the sand has been examined
with the results given in table on page 229.

228 THE MODERN ASPHALT PAVEMENT.

OOZINGS OF BITUMEN FROM NEAR BROWNSVILLE, KENTUCKY.

TEST No. 40894.
Loss, 212 F., until dry 22.8%

DRY SUBSTANCE.

Loss, 325 F., 7 hours 11.5%

Residue after heating to 325 F Penetration = 55

Loss, 400 F. for 7 hours (fresh sample) 13 . 1%

Residue after heating to 400 F Penetration = 12

Bitumen soluble in CS 2 , air temperature 94 . 4%

Difference .5

Inorganic or mineral matter 5.1

100.

Malthenes :

Bitumen soluble in 88 naphtha, air tem-
perature 75.6%

This is per cent of total bitumen. .... 80 .0

Bitumen soluble in 62 naphtha, air tem-
perature 82.6%

This is per cent of total bitumen 88 .

EXTRACTED BITUMEN.

Penetration a 78 F = 193

BITUMINOUS SANDSTONE, LOGAN COUNTY, KENTUCKY.
TEST No. 19938.

Bitumen soluble in CS 2 7.8%

Passing 200-mesh sieve 6.2

" 100- " " 27.0

" 80- " " 31.0

" 50- " " .. 25.0

40- " " 2.0

" 30- " " 1.0

100.0
Loss at 212 F *.5%

These results show that this material, like the bitumen in
the sands from other parts of Kentucky, is a maltha which
hardens on heating to a very brittle substance, and on that account
is not suitable for paving purposes.

ASPHALTIC SANDS AND LIMESTONES. 229

BITUMEN OOZING FROM SAND, LOGAN COUNTY, KENTUCKY.
TEST No. 21264.

Loss, 212 F., until dry 16.3%

Dry substance:

Loss, 325 F., 7 hours 5.5%

Residue after heating to 325 F Too soft for

penetration

Loss, 400 F. , 7 hours 3.8%

Residue after heating to 400 F Penetra-
tion =20
Bitumen soluble in CS 2 , air temperature. . . 62.8%

Difference 7.7

Inorganic or mineral matter 29 . 5

100.0

Malthenes :

Bitumen soluble in 88 naphtha, air tem-
perature 35 . 4%

This is per cent of total bitumen 56 . 4

Bitumen soluble in 62 naphtha, air tem-
perature 50 . 6%

This is per cent of total bitumen 80 . 6

Importance of the Kentucky Bituminous Sands for the Paving
Industry. From the preceding data it is very evident that no
satisfactory asphalt pavements can be constructed from any of
the bituminous sands available in Kentucky for two reasons.
In the first place the bitumen is a maltha which has no stability
and hardens very much on exposure to high temperatures. In
the second place it is not present in sufficient amount to cement
the sand grains together satisfactorily, and finally, the sand itself
is seldom graded in a way to form a satisfactory mineral aggregate.
It is always deficient in filler. It may be possible, for light traffic
streets, to make an asphalt surface mixture from a Kentucky
sand by the addition of some hard bitumen and a satisfactory
amount of filler, but when this is done it would generally be found
to have been a matter of economy not to have used the bituminous
sand at all, but to have started with a suitable local sand and have
combined this with a proper asphalt cement and a good filler.

Actual experience with asphalt pavements constructed with

230 THE MODERN ASPHALT PAVEMENT.

Kentucky material has confirmed all these conclusions, and it is
safe to say that the sooner the attempt to work these deposits is
abandoned the less money will be sunk.

Oklahoma. Deposits of bitumen in various forms are found
widely scattered over this State. Among them are several
which consist of bituminous sands. Although none of them
is of any great value in the paving industry, it will be of interest
here to show what their composition is in order that vain
attempts may not be made to utilize them at great financial
loss.

Limestones saturated with bitumen are also found in the
immediate neighborhood of the bituminous sands, and as attempts
have been made to utilize these in conjunction with the sands
they will be described at the same time.

In what Eldridge denominates the Buckhorn District, in the
region east of the Washita River and in the neighborhood of Rock
Creek, numerous quarries of bituminous sands and limestones have
been opened by different individuals and companies and some of the
material has been utilized in the construction of street pavements.

The material from the Ralston quarry, about 2 miles west-
northwest of Schley and 8 miles northeast of Dougherty, has the
following composition. This material is still further described
by Eldridge. 1

FROM RALSTON QUARRY, NEAR SCHLEY AND DOUGHERTY,

OKLAHOMA.

TEST No. 11602.

Bitumen soluble in CS 2 5.0%

Passing 200-mesh sieve 9.7

" 100- " lt 41.0

" 80- " " 29.0

50- " " 11.0

40- " " 2.0

30- " " 1.0

" 20- " " 1.0

" 10- " " 3

100.0
1 The Asphalt and Bitumimous Deposits of the U. S., 1901, 294.

ASPHALTIC SANDS AND LIMESTONES.

231

EXTRACTED BITUMEN.

Extracted bitumen : a soft maltha, consistency of residuum.

Loss, 325 F. , 7 hours 5.96%

" 400 F., 5 " (fresh sample) . 9.98%

Residue after heating to 400 F Pulls to a long

thin thread
and pene-
trates 76.

The Gilsonite Roofing and Paving Company's mines of bitu-
minous limestone and asphaltic sands are found in Sections 21,
22, and 23, Range R.3.E. The Rock Creek Natural Asphalt
Company own and have somewhat developed several deposits
of bituminous sands and limestone rocks north of the preceding
deposits, in the Buckhorn District.

The sands which have been developed to the greatest extent
and used by the Rock Creek Natural Asphalt Company have the
following composition:

BITUMINOUS SAND FROM BUCKHORN DISTRICT, OKLAHOMA.

Test number

30481

12.2%
1.8
29.0
26.0
30.0
1.0

100.0

69086

11-1%
12.9
48.0
23.0
5.0
0.0

100.0
Soft

Bitumen soluble in
Passing 2CO-mesh s
" 100- "
80- "
50- "
40- "

Extracted bitumen

CS,

icve

< <

i (

( t

at 78 F

This sand is fine and somewhat variable in grading, the bitu-
men which it contains is a soft maltha, although it varies some-
what in accordance with the extent to which it has been weathered.
The material has been combined with the limes tqnes, a description
of which follows, and fairly successful pavements have resulted
from the combination in Kansas City, Mo.

The principal bituminous limestone quarries of the Gilsonite
Roofing and Paving Company are known as No. 2 and No. 4, the
former being found at the southeast end of the so-called Buck-

232

THE MODERN ASPHALT PAVEMENT.

horn District, while the No. 4 quarry or mine is at the western end
of the District about 1 mile west of Schley and 7 miles northeast
of Dougherty. The composition of these limestones is as follows:

BITUMINOUS LIMESTONE FROM BUCKHORN DISTRICT,
OKLAHOMA.
TEST No. 69084.

LIME ROCK NO. 4.

Bitumen soluble in CS 2 4 . 3%

Carbonates and organic matter 86 . 8

Mineral matter insoluble in HC1 8.9

Extracted bitumen penetrates at 78 F. .

LIME ROCK NO. 2.

100.0
60

Test number ....

69085

57870

Bitumen soluble in CS 2

12 1%

13 1%

Carbonates and organic matter

76 5

81 6

Mineral matter insoluble in HC1

11 4

5 3

100.0

Extracted bitumen penetrates at 78 F. . .

In thin section under the microscope it is seen that these lime-
stones differ entirely in their structure from those found on the
Continent of Europe and which have been utilized so largely for
the construction of pavements. The mineral matter in the latter
consists entirely of the remains of marine animal life that are very
uniformly impregnated with bitumen. The limestones from
Oklahoma on the other hand, contain a very consider-
able proportion of hard crystalline calcite which is not impreg-
nated at all with bitumen. On this account the latter do not
compare favorably with the rock asphalts of Europe. 1

These limestones, as has been said previously, have been com-
bined with the sand rock to make a very satisfactory paving sur-
face, the proportions in use being J No. 2 lime rock, J No. 4 lime
rock, and J. sand rock. The sand rock supplies the flux necessary
for the hard bitumen in the limestone, the latter having a pene-

1 See page 252.

ASPHALTIC SANDS AND LIMESTONES. 233

tration of only 60 to 65, before heating, as it occurs in nature
and only 20 after that operation. Such a mixture has the follow-
ing composition:

SURFACE MIXTURE MADE FROM BITUMINOUS SAND AND LIME

ROCKS FROM OKLAHOMA.

TEST No. 48231.

Bitumen soluble in CS 2 8.2%

Passing 200-mesh sieve 18.8

100-
80-
50-
40-
30-
20-
10-

9.0

18.0

16.0

4.0

3.0

8.0

6.0

Retained on 10-mesh sieve 9.0

100.0

It will be noticed that the bitumen in this mixture is lower
than in a sand mixture of the same grading and yet it has been
shown by experience that it is a satisfactory one. This is prob-
ably due to the fact that the film of asphalt on the lime rock is
not necessarily as thick as that upon the sand grains and for this
reason the percentage of bitumen which this mineral aggregate
will carry is smaller than when the latter is of a silicious nature.
Although fairly satisfactory pavements have been made with
these materials it is not probable that they will prove of any impor-
tance in the paving industry as the supply as turned out is too
small to permit of obtaining a requisite quantity of uniform quality
and because the greatest skill is necessary in so handling the mate-
rial as to make it possible to put it down with the bitumen of a
proper state of consistency, as this changes very readily on being
heated in the slightest degree to too high temperature.

Bruns)i*ick District. The District which has been named by
Eldridge the " Brunswick District " lies on the Brunswick Creek
immediately north of Rock Creek, to which it is a tributary, 4 miles
northeast of Dougherty. The deposits here resemble those found in
the Buckhorn District. They have been worked to a certain extent
industrially but are probably of no great commercial interest.

234

THE MODERN ASPHALT PAVEMENT.

Analyses of the bituminous products obtained there, made in
1898, resulted as follows:

FROM BRUNSWICK DISTRICT, OKLAHOMA.

Test No. 18656 and 18657. Fossilif erous limestone, impregnated with bitumen.
" " 18662. Bituminous sand, Kirby mine.

' 18667. as shipped.
" " 18668. " " " "

Test number

18656
1.6%

18657

3.1%
3.9
1.0
1.0
11.0
12.0
30.0
31.0
7.0

18662

11.3%
1.7
29.0
45.0
13.0
0.0
0.0
0.0
0.0

18667

9.3%
1.7
36.0
40.0
13.0
0.0
0.0
0.0
0.0

18668

8.6%
1.4
20.0
42.0
28.0
0.0
0.0
0.0
0.0

Bitumen soluble in CS 2

Passing 200-mesh sieve

100- " "

80- " "

50-

40-

30-

20-

" 10-

100.0

Materials received from the same locality in 1903 had the
following composition :

Test number

63279

63280

63283

Bitumen soluble in CS 2

4 9%

6 8%

2 3%

Carbonates ....

89 1

86 4

94 1

Mineral matter insoluble in HC1

6.0

6.8

3.6

100.0

REMARKS: No. 63279. Hard compact limestone with free bitumen in small

seams, somewhat crystalline.

" 63280. Same as No. 63279, containing more seams and larger
ones and the latter being filled with a considerable
seepage of free bitumen, aside from that impregnat-
the rock.

" 63283. Same as No. 63279, unevenly impregnated with bitu-
men, the seams carrying free material, although to
no such extent as No. 63280.

It would seem from the above data that there can be no
question that the lime rock will not bear the cost of transportation.

Arbuckle Mountains. Many deposits of bituminous sands are
found in the neighborhood of the Arbuckle Mountains, some of

ASPHALTIC SANDS AND LIMESTONES. 235

which have been examined by the author. That occurring south-
east of Woodford, close to the Henryhouse Creek, and known as
the Sneider Deposit, has the following composition:

BITUMINOUS SAND FROM SNEIDER DEPOSIT, ARBUCKLE

MOUNTAINS DISTRICT, OKLAHOMA.

TEST No. 30478.

Bitumen soluble in CS 2 11 . 1%

200-mesh sieve 8.9

100- " " 75.0

80- " " 2.0

50-
40-
30-
20-
10-

2.0
1.0
0.0
0.0
0.0

100.0

The quarry is described in detail by Eldridge. 1
Attempts have been made to extract the bitumen from this sand,
industrially and the material obtained has the following properties:

EXTRACTED BITUMEN FROM BITUMINOUS SANDS FROM

ARBUCKLE MOUNTAINS DISTRICT, OKLAHOMA.

TEST No. 30474.

Loss, 212 F., 1 hour 0.1%

Residue after heating to 212 F Too soft for pene-
tration

DRY SUBSTANCE.

Loss, 325 F., 7 hours 3.5%

Residue after heating to 325 F Penetration = 110

Bitumen soluble in CS 2 , air temperature 68 . 7%

Difference 1.5

Inorganic or mineral matter 29 . 8

100.0
Malthenes:

Bitumen soluble in 88 naphtha , air temp. . . 57 . 3%

This is per cent of total bituman 83 . 4

Bitumen soluble in 62 naphtha, air temp. . . 62 . 6%

This is per cent of total bitumen 91.1

Character of extracted bitumen Soft at 78 F.

The character of this material and the cost of obtaining it
will probably exclude it from any commercial application.

1 Asphalt and Bituminous Rock Deposits, 1901, 316.

236 THE MODERN ASPHALT PAVEMENT.

The Elk Asphalt Company has a similar sand of the following
composition :

BITUMINOUS SAND FROM ELK ASPHALT COMPANY DEPOSIT,

OKLAHOMA.
TEST No. 30483.

Bitumen soluble in CS 2 8.7%

Passing 200-mesh sieve 10 .3

" 100- " " 7.0

" 80- " " 10.0

" 50- " " 34.0

" 40- " " 12.0

" 30- " " 4.0

" 20- " " 4.0

" 10- " " 6.0

Retained on 10-mesh sieve 4.0

100.0
Character of extracted bitumen = Soft maltha.

From this sand a bitumen has been extracted having the follow-
ing composition:

BITUMEN EXTRACTED FROM ELK ASPHALT COMPANY'S

DEPOSIT OF BITUMINOUS SAND, OKLAHOMA.

TEST No. 30475.

Loss, 212 F., 1 hour 0.2%

Consistency of residue after heating Too soft for

penetration

DRY SUBSTANCE.

Loss, 325 F., 7 hours 4.2%

Consistency of residue after heating Too soft for

penetration

Bitumen soluble in CS 2 , air temperature . . .- 88 . 0%

Difference 3.1

Inorganic or mineral matter 8.9

100.0
Malthenes :

Bitumen soluble in 88 naphtha, air temperature. . . 79 . 1%

This is per cent of total bitumen 89 . 9

Bitumen soluble in 62 naphtha, air temperature . . 85.4%
This is per cent of total bitumen 96 . 6

Extracted bitumen Too soft for

penetration

ASPHALTIC SANDS AND LIMESTONES.

237

Upon this material the same remarks may be made as in the
case of the Sneider bitumen.

Near Emet, in the same neighborhood, bituminous sands are
found which have the following characteristics:

DEPOSIT OF BITUMINOUS SAND NEAR EMET, OKLAHOMA-
TEST No. 30477.

Bitumen soluble in CS 2 10.4%

Passing 200-mesh sieve 2.6

100-
80-
50-
40-
30-
20-
10-

2.0

5.0

63.0

16.0

1.0

0.0

100.0

Consistency of extracted bitumen = . . . . Soft maltha

This sand, like the others, is impregnated with a soft maltha.

In the Quapaw Reservation a sand occurs which contains
from 16 to 18 per cent of bitumen, the sand grains having the
following grading :

BITUMINOUS SAND, QUAPAW RESERVATION, OKLAHOMA.

Test number

30479

30141

Bitumen soluble in CS 2
Passing 200-mesh sieve. . .

18.0%
29

16.5%
37 3

" 100- " "

10 1

" 80- " "

" 50- " "

19.0

" 40- " "
" 30- " "
tt 20- " "

7.0
3.0
2

10.0

.7
7

" 10- " "

Retained on 10-mesh sieve. .

9.0
100.0

.0
100.0

This sand apparently contains some organic matter not of a
bituminous nature.

Bituminous Limestone at Ravia. Just south of the Washita
River and Tishomingo, at Ravia, is a large deposit of bituminous
limestone. This contains 7.1 per cent of bitumen with varia-

238

THE MODERN ASPHALT PAVEMENT.

tions between 2.3 and 13.2 per cent. The bitumen is a rather
dense maltha having a penetration of 210 on extraction. The
limestone does not break down on extraction with solvents and
in thin section under the microscope is shown to be of uneven
texture containing crystals of calcite which are not impregnated
with bitumen. The rock is not made up, as in the case of the
Continental asphaltic limestones, of the remains of marine organ-
isms. Owing to this fact and the small percentage of bitumen
in the rock it can be of no commercial interest. Several samples
from the face of the mine give the following results on analyses:

BITUMINOUS LIMESTONE FROM RAVIA, OKLAHOMA.

Test number .

67316

67317

67318

67319

67320

67321

Bitumen by CS 2
Mineral matter insoluble
in HC1

10.8%
18.5

7.3%
15 5

7.0%
14 3

3-4%
30 8

9.9%
11 3

9.6%
17 9

Mineral matter soluble in
HC1

69.9

75.8

73 2

63 7

77 9

71.3

Difference

1 4

5 5

2 1

1 2

100.0

The Ravia rock is typical of all American asphaltic limestones
and for this reason it is hardly to be believed that any of them
possess the same desirable features as those which are mined in
Europe.

Other deposits in the Arbuckle Mountain region are of much
the same character as those which have been described and are
of no commercial interest.

Eldridge states that at Wheeler there is one of the largest
oil seepages in the United States. The character of the maltha
at this point is very much the same as that which has been extracted
from the sandstone. It is, of course, impossible to collect it in
sufficient quantity to be employed in the paving industry.

Five miles northwest of Ardmore a stratum of bituminous sand-
stone is found having a dip which is nearly vertical. Many attempts
have been made to utilize this material by boiling it with water,
but they have all been failures and have resulted in the loss of
considerable capital in the same way that has been the case else-

ASPHALTIC SANDS AND LIMESTONES.

239

where in Oklahoma. The best of the rock available at this point
has the following composition:

BITUMINOUS SANDSTONE NEAR ARDMORE, OKLAHOMA.

47443

9.6%
12.4
50.0
25.0
2.0
1.0
0.0
0.0
0.0

100.0

47444

11.8%
1.2
5.0
16.0
59.0
6.0
1.0
0.0
0.0

100.0

Bitumen soluble
Passing 200-mesl
100- "
80- "
50- "
40- "
30- "
20- "
" 10- "

in CS 2

ti sieve

< <

Penetration of extracted bitumen at 78 F. =44. Too soft for pen.
In the preparation of the bitumen by extraction it is impossible
to remove all the fine material and the maltha is much hardened
in the process of boiling out the water. A sample of the so-called
refined material contained 77.8 per cent of bitumen which had a
penetration of 75.

Grahamite in Oklahoma. There are several occurrences
of grahamite in Oklahoma, the chief one being in what Eldridge
has denominated the Tenmile Creek District. This has been
described in detail under the heading " Grahamite. " 1 At
another point, near Loco, in Section 36, T.2.S., R.4.W., gra-
hamite has been found in a network of veins which is remarkable
as containing from 23.6 per cent to 2.4 per cent of pyrites, evi-
dently introduced by infiltration. No commercial supply of
this material is available.

Eldridge has classified these materials as " Asphaltites " and
regards that found at the so-called Moulton mine as closely resem-
bling albertite. He has named it Impsonite. The author sees
no reason to use the word Asphaltite as applied to these bitu-
mens as its original use was quite different. It is a fact that the
weathered bitumen at the Moulton deposit on the surface resembles
albertite to a considerable extent, being only slightly soluble in

1 See page 21L

240 THE MODERN ASPHALT PAVEMENT.

the ordinary solvents for bitumen. As the material has been
taken out deeper down in the vein it is found to be entirely soluble
in carbon disulphide, to yield a percentage of fixed carbon which
is characteristic of grahamite and in every way to resemble closely
the type grahamite originally described from Ritchie County,
West Virginia. It cannot, therefore, be properly assigned a new
name, Impsonite.

The Value of the Deposits of Oklahoma in Relation
to the Paving Industry. From what has been said in our de-
scription of the Oklahoma bituminous deposits it is evident
that the only conclusion that can be drawn in regard to them
is that they are of little industrial interest with the exception,
perhaps, of the grahamite. The deposits, although large in amount,
taken as a whole are individually small and moreover, far from
being uniform in their character, they contain too little bitumen
and this bitumen is not sufficiently asphaltic in its character.
It is very improbable that any return will ever be obtained for
the amount of money that has been spent in attempting to develop
them.

Texas. Bitumen is found in Texas impregnating limestone in
Burnet and Uvalde Counties and mixed with sand in Montague
County.

The latter deposits are near St. Jo. They are of no commercial
interest and resemble in many respects those found in
Oklahoma.

In Burnet County, Eldridge states, the deposit consists of
Cretaceous limestones at Post Mountain, near the town of Bur-
net, which are impregnated with from 4 to 8 per cent of bitumen,
mostly with the latter amount. The bitumen is soft and sticky,
penetrating 240 on extraction. The quantity of asphalt bearing
rock is stated to be limited.

In Uvalde County the bituminous material is found 18 to 25
miles west of Uvalde in the region of the Anacacho Mountains.
The only deposit which has been worked to any considerable
extent is a peculiar limestone which Eldridge described as being
" an assemblage of minute organisms together with a conspicuous
proportion of crystalline calcite. Molluscan remains, often of

ASPHALTIC SA1S T DS AND LIMESTONES. 241

large size, are also present. Through the mass of rock there is
a high per cent of interstitial spaces, which in some instances
may even exceed the solid portions. In addition to the inter-
stitial spaces, properly so-called, are cavities produced by the
removal of the molluscan remains. . . ." The bitumen partially
fills these cavities and also impregnates the limestone to a certain
extent but the voids are never completely filled. On account of
the nature of the rock and its great lack of homogeneity the mate-
rial is not satisfactory for paving purposes, as is generally the
case when calcite is present. The rock carries about 12 to 15 per
cent of a peculiar bitumen, attempts to extract which were made
at one tune. Although it is now of no commercial value the
character of the bitumen may be noted with interest.

AsphaUic Rock from Litho-Carbon Company. An average sample
of the rock, Test No. 7293, as worked, contained 12.8 per cent of
bitumen, the mineral matter consisting of 87.0 per cent of lime-
stone and 1.2 per cent of silicates insoluble in the acid. The bitu-
men extracted from this had the following characteristics:

ASPHALTIC ROCK FROM LITHO-CARBON COMPANY.

Softens 160 F.

Flows 170 F.

Per cent of bitumen soluble in 88 naphtha. . . 54 . 5
Fixed carbon 18.0

ULTIMATE COMPOSITION.

Carbon 80 . 3%

Hydrogen 10 . 1

Sulphur 9.8

The character of this bitumen is quite different from that in
asphalt found elsewhere owing to the fact that, notwithstanding
its consistency, it has a high percentage of fixed carbon and
contains a larger amount of sulphur than is found in any asphalt
of the same consistency. On account of these peculiar proper-
ties it was known, at the time that an attempt was made to put
it on the market, as Litho Carbon or Gum Asphalt. There is,
of course, no reason to call it a gum, except from the fact that
it might be employed as a substitute for some of the resins known
as gums in the varnish trade. No native bitumen can possibly
be regarded as a gum.

242

THE MODERN ASPHALT PAVEMENT.

A fine grained bituminous limestone is also found on the Smythe
Ranch, " about 20 miles a little south of west from Uvalde and
4 or 5 miles south of the quarry of the Uvalde Asphalt Company."
It is quite different in character from that which has previously
been described. A specimen of the material examined in the
author's laboratory had the following characteristics:

BITUMINOUS LIMESTONE FROM SMYTHE RANCHE, TEXAS.
TEST No. 22070.

Bitumen soluble in CS 2 12.2%

Limestone 87 . 8

Sand insoluble in HC1. ., 1.2

100.0

EXTRACTED BITUMEN.

Consistency Hard-friable

Softens 240 F.

Flows 250 F.

Fixed carbon 16 .9%

Bituminous sandstones are also found in Uvalde County and
consist of an extremely fine sand, the larger portion passing the
100-mesh sieve, impregnated with a bitumen yielding a large
amount of fixed carbon and, therefore, similar to that found in
the limestone.

BITUMINOUS SANDSTONE, UVALDE COUNTY, TEXAS. '

22071

22072

Bitumen soluble in CS 2

9.8%

8.1%

Sand

90.2

91.9

Per cent of the sand insoluble
in HC1

100.0
96.5%

100.0
93.6%

Per cent of the sand passing-
100-mesh sieve

88.2%

91.9%

EXTRACTED BITUMEN.

Softens 210F.

Flows. . ... 220 F.

Fixed carbon 19.5%

A8PHALTIC SANDS AND LIMESTONES.

243

Bituminous Sands of California. The location of the bitu-
minous sands of California has been described in detail by Eld-
ridge. It will suffice to remark here upon the character of some
of the most important.

Santa Cruz Bituminous Sands. The quarries of bituminous
sand near the summit of the Empire Ridge, facing the Bay of
Monterey and the Pacific Ocean, are of very large extent. The
individual strata are very variable in composition, as can be
seen from the results of an examination of the various types
found there, collected by the author in 1898:

BITUMINOUS SANDS, SANTA CRUZ, CALIFORNIA,

Test No. 13578. Soft material from the foot of Point Quarry.

13579. Top of stratum, Side Hill Quarry.
" 13580. Richest rock, Side Hill Quarry.

" 13581. Lowest stratum, Rattlesnake Quarry.

" 13582. 6 to 9-foot vein, Hole Quarry.

" 13583. Poorer rock, Hole Quarry.

" 13584. Gray rock, Hole Quarry.

13585. 6 to 9-foot vein, Last Chance Quarry.

Test number

13578

13579

13580

13581

14 4%

15 4%

13 2%

15 1%

Passing 200-mesh sieve

6.4

5.2

8.6

1 5

100-

2.2

3.4

5.2

7 4

80-

10.0

12.0

50-

40-

30-

" 20-

9.0

6.0

" 10-

4.0

3.0

100.0

Test number . .

13582

13583

13584

13585

Bitumen soluble in CS 2

17 3%

11 4%

11 7%

14 2%

Passing 200-mesh sieve

5 6

1 5

2 4

100- "

24.1

4.1

26 6

2 4

" 80- "

39.0

12.0

" 50- "

11.0

35.0

20.0

40- " .
" 30- "

3.0

20.0
11

3.0
1

18.0
14

" 20- "

0.0

" 10- "

0.0

100.0

244

THE MODERN ASPHALT PAVEMENT.

The bitumen which these sands contain is in the form of
maltha, much of it readily staining the hands when the sands
are handled. It hardens on heating with a loss oi the lighter
oils and a reduction in the percentage of bitumen to a point which
makes it possible to produce a surface mixture which will with-
stand traffic.

It will be noted that the grading of these sands is sufficiently
fine and that they contain a certain amount of 200-mesh material.

The streets which have been paved with the Santa Cruz bitu-
minous sands in San Francisco have been only fairly satisfactory.
They have required large repairs which, however, are readily made
by reheating the material, but there is now a tendency to abandon
this form of asphalt pavement and to construct surfaces from
properly graded sand combined with filler and a suitable pure
bitumen.

San Luis Obispo Bituminous Sands. Deposits of bituminous
sands near San Luis Obispo, in the county of the same name,
were formerly worked to a very considerable extent, but these
sands were much more variable in character than those found at
Santa Cruz. Different strata contain from 8 to 16 per cent of
bitumen and generally below 10 per cent. The following analyses
show the characteristics of those which were available in 1898:

BITUMINOUS SANDS, SAN LUIS OBISPO, CALIFORNIA,

Test number

13576

13577

Bitumen soluble in CS 2 . . .
Passing 200-mesh sieve. . .

8.8%
11.9

11.4%
4.4

100-

6.1

80-

10.2

16.1

50-

50.0

44.0

40-

8.0

9.6

30-

1.0

5.0

20-

1.0

3.0

10-

3.0

1.0

100.0

The supply of the sands, which is readily available, is now
nearly exhausted and they are no longer a commercial factor.

ASPHALTIC SANDS AND LIMESTONES.

245

Bituminous Sands in Santa Barbara County. Large deposits
of bituminous sands occur in Santa Barbara County in the Sisquoc
Hills, the location and geological relations of which are described
by Eldridge. 1

The deposit worked by the Alcatraz Company had the following
composition:

SANTA BARBARA COUNTY, CALIFORNIA.

Test number

6484

6485

Bitumen soluble in CS 2
Passing 200-mesh sieve

18.5%
12.5

16.5%
7.5

100-

8.0

7.0

80-

3.0

6.0

50-

39.0

20.0

40-

10.0

20.0

30-

8.0

14.0

20-

1.0

8.0

10-

0.0

1.0

100.0

This is , sand of medium grade, largely 50- and 40-mesh grains,
but carries a very considerable amount of 200-mesh material.
The bitumen is in the nature of a maltha and was extracted from
the sand with naphtha, sent down to the seacoast by pipe-line
and there recovered by distillation. On heating, the original soft
bitumen was hardened to a proper consistency for use for paving
purposes. The process proved to be an expensive one and the
material when extracted was of no better quality than that obtained
by the distillation of ordinary California petroleum. After the
expenditure of a vast amount of money the process was abandoned.
Some of the bitumen prepared in this manner had the following
characteristics:

BITUMEN EXTRACTED FROM SANTA BARBARA COUNTY,

CALIFORNIA, BITUMINOUS SANDS.

TEST No. 35202.

Penetration at 78 F 48

Bitumen soluble in CS 2 , air temperature 89.4%

Difference .3

Inorganic or mineral matter 10 . 3

100.0

The Asphalt and Bituminous Rock Deposits, 1901, 429.

246

THE MODERN ASPHALT PAVEMENT.

Carpinteria Sands. One of the first bituminous sands to be
worked in California for the purpose of obtaining a pure bitumen
was that known as the Las Conchas deposit, occurring near the
beach at Carpinteria, Santa Barbara County. The sand at this
point was worked from the surface. It had the following com-
position :

LAS CONCHAS DEPOSIT AT CARPINTERIA, SANTA BARBARA,

CALIFORNIA.
TEST No. 6475.

Bitumen soluble in CS 2

18.9%

18.4%

Passing 200-mesh sieve

1.1

4.5

100- ' '

3.0

80- '

28.0

25.0

50- ' '

45.2 '

48.0

40- ' '

1 1

30- ' '

100.0

Per cent of total bitumen

soluble in 88 naphtha ....

83.0%

Attempts were made, which were never very successful prac-
tically or commercially, to extract the bitumen by boiling the
sand with water. The material is of interest to-day only his-
torically and as being typical of a certain class of soft bitumens
the nature of which has been already referred to on page 128.
They harden so on heating that a soft maltha will become con-
verted into a brittle pitch most readily and on this account were
the cause of the failures in the early attempts to lay asphalt sur-
faces with California material.

The deposits of solid bitumens in California have been con-
sidered under the heading " Asphalt."

Colorado. The bitumens of Colorado consist only of a paraffme
petroleum, in the Florence oil field, of some veins of gilsonite
in the western portion, and of a grahamite found in Middle Park,
the location and manner of occurrence of the latter being accu-
rately described by Eldridge. He speaks of it as an asphalt closely
resembling gilsonite which is, of course, quite an erroneous descrip-
tion as it does not melt and yields 47 per cent of fixed carbon.
It has already been described under grahamite. 1

1 See page 2147~

ASPHALTIC SANDS AND LIMESTONES.

247

The paraffine petroleum furnishes a flux which, when carefully
prepared, is entirely satisfactory for use in the asphalt paving
industry.

As far as the author is aware no asphaltic sands or limestones
occur in Colorado which are of commercial importance.

The Gilsonites and Other Solid Native Bitumens of Utah.
Utah has deposits of bitumen of very varied character. Gil-
sonite veins are characteristic of this state and the material which
they furnish has already been described. Wurtzilite and Ozo-
cerite are found in small amounts but are of no importance to
the paving industry, nor is the albertite which is found about eight
miles from Helper Station on the Rio Grande & Western R.R.,
which Eldridge has unfortunately described under the new specific
name of Nigrite, which is quite unnecessary and illustrates the dupli-
cation of names which is common among investigators who are not
widely acquainted with the materials which they examine. It is
plainly an albertite as can be seen from the following determinations
in comparison with some for the type albertite found hi Nova Scotia.

ALBERTITE.

Test number

19187

7834

Utah

Nova Scotia.

Color of powder .

Black

Fracture

Irregular

Smooth

Fusibility

Does not

Intumesces

Specific gravity, 78 F./78 F

intumesce
1 092

1 076

Bitumen soluble in CS air temperature

5 6%

*t Q%

94.2

<j.y /O

94 1

Inorsanic or mineral matter

Trace

Bitumen yields on ignition :

37 0%

29 8%

1.06%

1 2%

Wurtzilite might be a valuable material for industrial purposes
were it available in commercial quantities, but this is not the case.
Ozocerite could never be of any value to the paving industry

248

THE MODERN ASPHALT PAVEMENT

as it is a hard paraffine. The location of all these deposits are
closely fixed by Eldridge.

Bituminous Sands and Limestones. Asphaltic limestone is
found in the same geological horizons as those in which the alber-
tite and wurtzilite of Utah occur and its bitumen is probably of
the same origin. That located by Eldridge between Strawberry
and Soldier creeks, 7 miles northwest of Clear Creek Station,
on the Rio G. & W. R.R., is far from uniform in composition,
which has been found in the author's laboratory to be as follows:

ASPHALTIC LIMESTONE FROM NEAR CLEAR CREEK
STATION, UTAH.

21633

21634

21635

21636

Bitumen soluble in CS 8

13.7%

13.3%

7.3%

5.2%

Penetration of extracted bitumen at
78 F

Part soluble in HC1

62 3%

58 1%

52 9%

64 2%

The ignited residue effervesces with acid.

A limited supply of fairly pure bitumen has been obtained from
this rock, which has the characteristics given in the table on page 249.

This is a most remarkable bitumen since there is such a great
variation in the solubility in 88 and 62 naphthas and since it
yields no fixed carbon on ignition. From a scientific point of
view it is worthy of careful study.

Eldridge mentions deposits of asphaltic limestones in the same
locality as that in which the wurtzilite veins occur along por-
tions of the outer face of the Roan Plateau, on its westward exten-
sion, across Soldier Summit. These deposits have not been iden-
tified as any that have come into the author's hands.

In Grand County, near the western border of Colorado, at
the head of the West Water Canon, 20 miles north of West Water,
free bitumen has been obtained to a certain extent, both in soft
and hard form. This material when examined in the author's
laboratory was found to have the characteristics given in the table
on page 250, Test No. 60532.

ASPHALTIC SANDS AND LIMESTONES. 249

BITUMEN EXTRACTED FROM LIMESTONE ROCK FOUND

NEAR CLEAR CREEK STATION, UTAH.

TEST No. 21632.

Specific gravity, 78 F./78 F 1 . 20

Color Light brown

Lustre Dull shining

Structure Compact

Fracture Conchoidal

Hardness, original substance 1

Fuses Readily

Softens 210 F.

Flows 220 F.

Loss, 212 F., 1 hour .6%

Bitumen soluble in CS 2 , air temperature 75 . 3%

Difference 3.4

Inorganic or mineral matter. 21.3

100.0
Mineral matter soluble in HC1. 48 .4% ,

Bitumen soluble in 88 naphtha, air temperature 48 . 3%

This is per cent of total bitumen 64 . 3

Bitumen soluble in 62 naphtha, air temperature 72 . 8%

This is per cent of total bitumen 96 . 7

Bitumen yields on ignition:

Fixed carbon 0.0%

Penetration of extracted bitumen at 78 F 45

From the small percentage of fixed carbon which the Grand
County bitumen yields it is evident that it is not a true asphalt,
that it approaches in composition more nearly that of the paraffine
series, and resembles to some degree the material described from
the locality near Clear Creek station.

The soft bitumen found at this point is a maltha which is very
pure, 98.6 per cent of bitumen, which consists almost entirely
of maHhenes soluble in 88 naphtha, 94.5 per cent.

It has a specific gravity of .9874 and after heating for 7 hours
at 325 F. hardens to a consistency of 53 and to 23 after the
same length of time at 400 F.

These bitumens are of no commercial, but of great scientific
interest as they differ so markedly in their characteristics from.

250 THE MODERN ASPHALT PAVEMENT.

other asphalts. Gilsonite may have been derived from such a
material.

BITUMEN FROM GRAND COUNTY, UTAH.
TEST No. 60532.

DRIED CRUDE.

Bitumen soluble in CS 2 , air temperature 43 .2%

Difference 7.5

Inorganic or mineral matter 49 . 3

100.0

EXTRACTED BITUMEN.

Specific gravity, 78 F./78 F 1 .037

Color Black

Hardness Variable

Odor. Asphaltic

Softens 203 F.

Flows 221 F.

Penetration at 78 F 22

Loss, 212 F.,1 hour 2.8 %

Bitumen soluble in CS 2 , air temperature 94 . 8%

Difference 1.6

Inorganic or mineral matter 3.6

100.0

Bitumen soluble in 88 naphtha, air temperature 68 . 7%
This is per cent of total bitumen 71.0

Bitumen soluble in 62 naphtha, air temperature 90 . 3%
This is per cent of total bitumen . . 93 . 3

Bitumen yields on ignition :

Fixed carbon. . .< 8.0%

Bituminous Sands. Bituminous sandstones occur in various
parts of Utah. The A. L. Hobson mine, H- miles from Thistle
Junction, is a material of the following composition:

BITUMINOUS SAND FROM A. L. HOBSON MINE, THISTLE
JUNCTION, UTAH.

TEST No. 21730.
Loss, 212 F., until dry 0.1%

Bitumen soluble in CS 2 11 . 6%

Part soluble in HC1. . .20.0

ASPHALTIC SANDS AND LIMESTONES. 251

It appears that this is a mixture of sand and silicates.

About 8 miles from Sunnyside, in Carbon County, on the Rio
G. &. W. R.R., a bituminous sand is found in large quantities
which has the following composition:

BITUMINOUS SAND, SUNNYSIDE, CARBON COUNTY, UTAH.
TEST No. 37048.

Bitumen soluble in CS 2 11 .2%

Passing 200-mesh sieve 16 .8

" 100- " "

" 80- " "

50- " "

" 40- " "

" 30- " "

" 20- " " ,

" 10- " "

100.0

Mineral matter Quartz sand

Extracted bitumen Pulls to a thread

The mineral matter consists of quartz sand and the extracted
bitumen possesses the characteristics of a maltha.

In Whitmore Canon bituminous sandstone occurs nearly free
from carbonates, the bitumen having a penetration of 35. It
has the following characteristics:

BITUMINOUS SAND, WHITMORE CANON, UTAH.
TEST No. 21729.

Bitumen soluble in CS 2 . 10.9%

Passing 200-mesh sieve 17.9

" 100- " " 16.1

" 80- " " 16.1

" 50- " " 21.4

" 40- " " 14.2

" 30- " " 3.4

100.0

Per cent soluble in HC1 2.6%

Extracted bitumen, penetration at 78 F. = 35

The bitumen obtained from this sand is a maltha which has
been examined by the author with the following results:

252 THE MODERN ASPHALT PAVEMENT.

BITUMEN EXTRACTED FROM SAND FROM WHITMORE
CANON, UTAH.

TEST No. 21731.

Penetration at 78 F Too soft

for test

Loss, 212 F., until dry 18.6%

Loss, 325 F., 7 hours 6.6%

Residue after 325 F. penetrates 145

Bitumen soluble in CS 2 , air temperature 97 . 8%

Difference 0.6

Inorganic or mineral matter 1.6

100.0

Bitumen soluble in 88 naphtha, air temperature 89.8%
This is per cent of total bitumen 91 . 8

Bitumen soluble in 62 naphtha, air temperature 97.0%
This is per cent of total bitumen 98 . 7

Bitumen yields on ignition :
Fixed carbon. 5.0%

It is evident from the small amount of fixed carbon which it
yields that it is not asphaltic and it, therefore, corresponds in
this respect with the bitumen found in similar Utah bituminous
sands and limestones previously described. It would seem, there-
fore, that the bitumens of this nature found in Utah are more closely
allied to ozocerite or to gilsonite than they are to the asphalts.

Deposits in Other States. Seepages of maltha and sand and
limestone impregnated therewith are found in many other States,
the distribution of bitumen being much more general than would
be supposed. None of these deposits are of any commercial inter-
est and must, therefore, be passed over.

Continental Rock Asphalts. The asphaltic limestones from
the Continent of Europe, which have been the main source of
the material for the asphalt paving industry in that country,
are scattered through France, Switzerland, Germany, Sicily, and
Italy. As these rocks reach the United States they have the
composition given on pages 253 and 254.

ASPHALTIC SANDS AND LIMESTONES.

253

CONTINENTAL ROCK ASPHALTS.

Test No. 47137. Ragusa, Sicily.

" " 47147. Seyssel, France.

" " 47153. Vorwohle.

" " 47156. Sicula, Sicily.

" " 47159. Neuchatel, Val de Travere.

" " 47162. Mons.

Test number

47137

47147

47153

47156

47159

47162

Bitumen soluble in CS 2 . .

9.9%

5.9%

7.5%

10.2%

9.1%

8.9%

Passing 200-mesh sieve
100-

37.1
17.0

44.1
10.0

18.5
14.0

33.8
16.0

36.9
14.0

53.1
9.0

80-

6.0

5.0

21.0

9.0

15.0

4.0

50-

14.0

9.0

25.0

18.0

14.0

7.0

40-

4.0

7.0

8.0

4.0

5.0

30- <

2.0

7.0

2.0

3.0

4.0

3.0

20- " .

5.0

6.0

3.0

1.0

2.0

5.0

a 10 _ a

5.0

6.0

2.0

1.0

5.0

100.0

For some of the rocks which have not been examined by the
author reference must be made to the analyses of others. 1 See
the table on page 254.

These asphaltic limestones are characterized more by differences
in the grain of the limestone than of their bitumen contents. As
seen in thin sections it appears that the Continental asphaltic lime-
stones consist of the remains of marine animal life, and it is
undoubtedly this fact which gives them their uniform impregna-
tion and their faculty of being readily compacted, as distinguished
from American asphaltic limestones which contain very con-
siderable proportions of hard crystalline calcite not impregnated
with bitumen.

The Sicilian rock may vary in bitumen from 6.6 to 11.4 per
cent. The rock exported by the Sicula Company is about as rich
3000 tons examined by the author, in three samples, containing
9.5, 9.3, and 9.9 per cent of bitumen, though some of it reaches
12 per cent. The Mons rock is not evenly impregnated; veins
which are pure white being scattered through the material. This
rock is used more on account of the character of the grain than

1 Dietrich, Die Asphaltenstrassen.

254

THE MODERN ASPHALT PAVEMENT.

for its bitumen contents, which will average 6.5 per cent to 9.0*per
cent. The rock obtained from the Seyssel mine at present is very
poor in bitumen, not exceeding 6 per cent and in some cases drop-
ping to 1 per cent. It is used on account of the character of the
grain of the stone.

CONTINENTAL ROCK ASPHALTS.

1. Val de Travers. 5. Cesi.

2. Seyssel, Pyrimont. 6. Roccamorice.

3. Lobsann. 7. Limmer.

4. Ragusa. 8. Vorwohle.

1
10.15%

8 15%

12 32%

8 92%

88.40

91.30

71 43

88*21

0.25

0.15

5 91

5.18

Carbonate of magnesia

6.30

0.10

0.31

Sand

3 15

Insoluble in acid

6.45

0.10

0.45

0.20

1 70

7.15%

12.46%

7
14 . 30%

8
8.50%

Carbonate of lime

73.76

77 53

67 00

80 04

1.72

2 63

Alumina and iron oxides

3.02

2.17

1 A

j 4. 03

Carbonate of magnesia

14.24

4.71

17 52

0.55

Sand

1 A m

Insoluble in acid

J4.77

Difference

1 18

2.11

The richer Sicilian rock by itself does not form a stable pave-
ment but when some of the Seyssel or Mons rock is added to it
stability is obtained.

Continental rock asphalts are now used in this country almost
solely in mastics, the extreme slipperiness of the pavement made
with them having proved so objectionable in comparison with
the asphaltic sand pavements that the former are no longer toler-
ated.

ASPHALTIC SANDS AND LIMESTONEa 255

SUMMARY.

The asphaltic sands and limestones of the United States have
not been shown to be attractive to those interested in the con-
struction of asphalt pavements. The asphaltic sands of Kentucky
are too deficient in bitumen to make a satisfactory surface mix-
ture and at the same time the character of the bitumen which
they contain is altogether too oily. Successful surfaces have never
been made with these materials unless they have been largely
amended by the addition of a considerable amount of a harder
bitumen and a proper proportion of filler.

The bituminous sands of California, although they have been
used to a very considerable extent, are now known to give results
which cannot compare favorably in any way with the artificial
mixtures which have been laid along parallel streets. Their use
for heavy traffic work will no doubt be soon abandoned.

The bituminous limestones and sands of Oklahoma
occur in such small masses and pockets that their uniformity
can never be guaranteed. In a few instances excellent street
surfaces have been constructed from the mixed sands and lime-
stones obtained near Dougherty, but the character of the asphaltic
limestones is such that they can never be used in the same way
as the Continental asphaltic limestones, owing to the structure
of their mineral aggregate.

As a whole it is probable that more money has been lost hi
attempting to develop the isphaltic deposits described in this
chapter than will ever be recovered by working them.

CHAPTER XIII.

RESIDUAL PITCHES, OR SOLID BITUMENS DERIVED FROM
ASPHALTIC AND OTHER PETROLEUMS.

IF the distillation of the asphaltic petroleums of California,
of the semi-asphaltic petroleums of Texas, or even of Russian oil
and some paraffine petroleums, is carried sufficiently far the residue
on cooling will be found to be a solid bitumen, and from asphaltic
oils of a more or less asphaltic nature. The properties of these
solid bitumens and their availability for industrial purposes depend
largely, of course on the nature of the petroleum from which
they are derived, the care with which the distillation is conducted
and the amount of cracking which has taken place in the process.

Residual Pitches from California Petroleum. The residues
from California petroleum have been used to a very consider-
able extent in the paving industry and are generally known as
" D " grade asphalt or under some trade designation or brand,
such as Diamond, Obispo, Acme, or Hercules.

They have been, usually, all moreor less carelessly manufactured
without laboratory control and consequently vary in character and
consistency. As a rule they are by-products resulting from the
recovery of distillates of different gravities from crude petroleum
and are not prepared especially for paving purposes. That
they are badly cracked in the process of manufacture, the oil
often being heated as high as 900 F., appears from the fact that
if the petroleum from which they are obtained is distilled or evap-
orated under such conditions that cracking will not occur, as
much as 60 per cent of a hard residue will remain, as shown by
the following figures obtained in the author's laboratory, as com-
pared with 30 per cent by industrial methods.

256

RESIDUAL PITCHES. 257

RESIDUAL PITCH FROM CALIFORNIA PETROLEUM PREPARED
IX THE LABORATORY.

TEST No. 69209.

Loss, 212 F , to constant weight 2.8%

Loss, heating until 59 penetration is obtained. ... 38 . 9%

Residual solid bitumen penetrating 59. 61 . 1

100.0

ANALYSIS OF BITUMEN PENETRATING 59.

Loss, 400 F., 4 hours 4.5%

Penetration of residue after heating 29

Bitumen soluble in CS 2 , air temperature 99.8%

Difference 1

Inorganic or mineral matter. 1

100.0
Malthenes:

Bitumen soluble in 88 naphtha, air temperature 77 . 6%
This is per cent of total bitumen 77.8%

Carbenes :

Bitumen insoluble in carbon tetrachloride, air

temperature. , 0.5%

Bitumen yields on ignition:

Fixed carbon 10.5%

As a matter of fact, under the conditions which obtain indus-
trially, only 30 to 40 per cent of solid residue is recovered, the
remainder being cracked and volatilized, and such residues con-
tain a much larger amount of fixed carbon, 15 to 20 per cent, than
is found on careful evaporation. With the form of still at present
in use and with the most careful handling the temperature rises
to 720 F. and tho residual pitch is much smaller in amount, in
the author's experience, than it should be. As an illustration
of this, at an oil works under the author's observation, where
an endeavor was made to produce the best " D " grade material
for paving purposes, the petroleum in use, on careful evaporation
at 400 F., left a residuum of solid bitumen amounting to 61.1 per
cent, and penetrating 59, as appears in the preceding table, whereas
industrially only 33 per cent was recovered having about the same
penetration. The physical characteristics and proximate com-
position of the industrial product obtained in this way are given
in the accompanying tables. See pages 258, 259, 260, 261 , and 262.

258

THE MODERN ASPHALT PAVEMENT.

"DEGRADE CALIFORNIA ASPHALT PHYSICAT

Test number 18250

Year received 1898

PHYSICAL PROPERTIES.

Specific gravity 78 F./78 F., original substance, dry 1 .089

Color of powder or streak Black

Lustre Lustrous

Structure Uniform

Fracture '. Conchoidal

Hardness, original substance 1

Odor Asphaltic

Softens 150 F.

Flows 162 F.

Penetration at 78 F 25

CHEMICAL CHARACTERISTICS.

Loss, 325 F., 7 hours .83%

Residue penetrates at 78 F 17

Loss, 400 F., 7 hours (fresh sample) 4.9%

Residue penetrates at 78 F ., 9

Bitumen soluble in CS 2 , air temperature 98 . 3%

Difference . 0.5

Inorganic or mineral matter 1.2

100.0

Malthenes :

Bitumen soluble in 88 naphtha, air temperature 65 .0%

This is per cent of total bitumen. 68 . 6

Per cent of soluble bitumen removed by H,jSO 4 50.0

Per cent of total bitumen as saturated hydrocarbons 33 . 1

Carbenes :

Bitumen insoluble in carbon tetrachloride, air tempera-
ture 7.0

Bitumen yields on ignition:

Fixed carbon 19.0%

It will be noted by a comparison of the above data with
those given on page 257 that while the solubility of the bitumen
in carbon disulphide is unaltered in the process of distillation a
very considerable portion of it has often been rendered insoluble
in cold carbon tetrachloride and in 88 naphtha. At the same
time the amount of fixed carbon in the industrial product is largely

RESIDUAL PITCHES.
CHARACTERISTICS AND PROXIMATE COMPOSITION.

259

68488

69549

69550

69605

69606

1903

March 1904

April 1904

1.062
Black
Lustrous
Uniform
Sticky
Tacky
Asphaltic
142 F
156 F.
52

1.052
Black
Lustrous
Uniform
Sticky
Tacky
Asphaltic
178 F.
190 F.
45

1.046
Black
Lustrous
Uniform
Sticky
Tacky
Asphaltic
106 F.
120 F.
65

1.055
Black
Lustrous
Uniform
Sticky
Tacky
Asphaltic
128 F.
141 F.
50

1.071
Black
Lustrous
Uniform
Conchoidal
-1
Asphaltic
120 F.
135 F.
52

83%
Hard

H'&S 6

r 3 ^ %

2.1%
23

2.7%
29

Hard

9.6%
Hard

to 4 %

6-7%
10

7.1%
16

99.3%
!3

99.2%
.8
Trace

99.6%

99.6%
.3
.1

99.7%
.3
Trace

100.0

77.0%
77.5
47.8
40.5

66.6%
67.0
56.9
28.9

70.5%
70.8
62.7
26.4

68.5%
68.8
57.7
29.2

57.3

42.8

0.5%

7.3%

2.8%

2.2%

6.0%

15.0%

18.0%

16.7%

18.0%

18.8%

increased and this increase corresponds to the degree of severity
of the heat to which the oil has been subjected. These differ-
ences characterize the California pitches as being, to a certain
extent, products of decomposition and on this account undesirable
material.

Considered as a class they are also undesirable because they

260

THE MODERN ASPHALT PAVEMENT.
"D" GRADE CALIFORNIA ASPHALT.

18250

Carelessly
Prepared

1.089
Black
Lustrous
Uniform
Conchoidal
-1
Asphaltic
150 F.
162 F.
25

83%
17

4.9%
9

98.3%
0.5
1.2

100.0

65.0%
68.6

50.0
33.1

68488
More
Carefully
Prepared

1.062
Black
Lustrous
Uniform
Tacky
Sticky
Asphaltic
142 F.
156 F.
52

Hf?d %

99.3%
!3
100.0

77.0%
77.5

47.8
40.5

80.2%
80.8

0.5%
15.0%

PHYSICAL PROPERTIES.

Specific Gravity, 78 F./78 F., original sub-
stance dry

Odor

Softens. . .

Flows .,

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance:
Loss 325 F 7 hours

Loss 400 F., 7 hours (fresh sample)

Residue penetrates at 78 F

Bitumen soluble in CS 2 air temperature

Malthenes:
Bitumen soluble in 88 naphtha, air tern-
perature

This is per cent of total bitumen

Per cent of soluble bitumen removed by
H SO

Per cent of total bitumen as saturated hy-

Bitumen soluble in 62 naphtha

Carbenes:
Per cent of bitumen insoluble in carbon
tetrachloride air temperature

7.0%
19.0%

Bitumen yields on ignition:

RESIDUAL PITCHES.

261

are not uniform in character, as shown by the different degree
of solubility of the bitumen in cold carbon tetrachloride and by
the very considerable variation in the amount of fixed carbon
which they yield.

"D" GRADE ASPHALT FROM REFINERY AT LOS ANGELES, CAL.
AVERAGE AND EXTREMES OF COMPOSITION IN 1904.

Average.

Highest.

Lowest.

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original sub-

1 060

1 066

1 054

Flashes, F , N Y. State oil-tester

406 F.

420 F.

385 F.

Softens

137 F.

190 F.

124 F.

Flows

150 F.

180 F.

140 F

Penetration at 78 F . . .

118

CHEMICAL CHARACTERISTICS.

Loss 400 F 4 hours.

7.12%

9.40%

5 52%

Residue after heating penetrates at 78 F.. . .

Bitumen soluble in CS a , air temperature ....
1 hfterence ."

99.4%
.4

99.9%
1.59

98.1%
.0

Inorganic or mineral matter

.54

Malthenes:
Per cent of total bitumen soluble in 88
naphtha air temperature ...

100.0
71 61%

83 81%

66 01%

Carbenes :
Insoluble in carbon tetrachloride, air tern-

4.37%

6.91%

32%

565

From an average of a very large number of analyses of " D "
grade asphalts it has been found that the amount of fixed carbon
which they yield, when prepared as carefully as possible by the
present industrial process, does not vary far from 15 per cent,
although at times it reaches 19 per cent where the product is care-
lessly handled, and should not exceed 10 per cent as shown by
our laboratory results. This characteristic of the California
pitches is important in differentiating them from those made

262

THE MODERN ASPHALT PAVEMENT.

from Texas oil, which yield a much higher percentage of fixed
carbon.

"D" GRADE CALIFORNIA ASPHALT. BITUMEN INSOLUBLE IN
CARBON TETRACHLORIDE.

Test Number.

Bitumen Insolu-
ble in Carbon
Tetrachloride,
Air Temperature.

18250

7.0%

63847

69549

7.3

73798

73799

73800

73801

73959

73960

73961

73962

74087

74088

2.8

74089

74090

74091

1.3

For the purpose of preparing a pitch suitable for paving pur-
poses it is, of course, desirable that some of the malthenes should
be converted to asphaltenes, although not to carbenes. The bitu-
men, soluble in 88 naphtha, should be reduced to about 70-75
per cent and the fixed carbon should reach 15 per cent.

Harder Residual Pitches. Where the consistency of the
asphaltic residue is harder, its character has been denominated
by other letters than " D." For example, A, B, and C grades
are found, and much of the " D " grade put upon the market
corresponds to these materials rather than to a true " D " grade.
Where an attempt is made to manufacture the different grades
they are expected to be of a consistency corresponding to the
following penetrations on the Bo wen scale:

A Grade 9

B " 15

C " 25

D " 46 and above.

RESIDUAL PITCHES. 263

Residual bitumens having a penetration of less than 46 are
deficient in the less viscous malthenes and require a very large
amount of flux, to bring them to a proper consistency for paving
cement. This results in the presence of too large a percentage
of both brittle asphaltenes and the lighter forms of malthenes.
Where these very hard residual pitches are in use in the produc-
tion of a paving cement the results have been disastrous in climates
where severe conditions are met, although they may be passable
in the climate of Southern California.

Of late years, the use of such very hard residual pitches has
been done away with, and those which are used for paving purposes
are now very skilfully prepared at low temperatures, of very nearly
the proper consistency for use as a paving cement, so that it is
not necessary to add more than two or three pounds of flux, and
often none at all, before using the bitumen.

" Specifications f or ' D ' Grade Asphalt. ' D ' grade asphalt
should be the residue from the careful distillation, with steam
agitation, of some suitable California petroleum at as low a tem-
perature as possible and certainly not exceeding 700 F. It shall
be free from free carbon or suspended insoluble matter, which
are evidences of excessive cracking.

" It shall be soluble to the extent of at least 98 per cent in
carbon disulphide, 95 per cent in cold carbon tetrachloride and
not less than 65 nor more than 80 per cent of it shall be solu-
ble in 88 Pennsylvania naphtha, preferably nearer the former
figure.

" It shall not flash below 450 F. and shall have a density be-
tween 1.04 and 1.06. It shall not volatilize more than 8.0 per cent
at 400 F. in 4 hours, and shall have a penetration between 40 q
and 70. It shall melt at not less than 140 nor over 180 F. on
mercury, according to the method in use in the New York Test-
ing Laboratory, and shall yield not more than 15 per cent of
fixed carbon on ignition.

" It shall have a consistency of not less than four (4) milli-
meters penetration at 78 F. when tested for five (5) seconds with
a No. 2 needle weighted with 100 grams."

The lower the temperature at which the asphalt is produced
the smaller the percentage of cracked products it will contain
and the smaller the loss will be on heating for 4 hours at 400 F.

264

THE MODERN ASPHALT PAVEMENT.

The difference in its character when run down in 20 and 65 hours
can be seen from the following figures:

COMPARISON BETWEEN "D" GRADE TAKING 65 AND 20
HOURS TO COME TO GRADE.

Still

Small

Large

99.65%

99.90%

.25

.10

Inorganic or mineral matter. ..

.10

Malthenes:
Bitumen solution in 88 naphtha, air temp
This is per cent of total bitumen

100.00

71.77%
72 02

100.00

72.50%
72 57

Penetration of still sample at 78 F

'* " barrelling sample at 78 F

Specific gravity 78 F./78 F

1.057

1 054

Softens

120 F.

124 F.

Flows

138 F.

140 F.

Loss, 400 F., 4 hours.
Penetration, at 78 F., of residue after heating . .
Yield

9.4%
12
33 0%

6.8%
15
43 5%

For comparison with the preceding " D " grade product a
bitumen procured on the market in Los Angeles, Cal., in 1904,
will serve. See table on page 265.

Here the percentage of fixed carbon is very high and that of
the malthenes is low, the total bitumen at the same time amount-
ing to only 93 per cent, while a very large proportion of bitumen
soluble in carbon disulphide but insoluble in cold carbon tetra-
chloride and of free carbon are present. This material has been
plainly overheated and it will require from 30 to 40 pounds of flux
instead of the much smaller quantity necessary with a properly
prepared asphalt. In this connection it may be of interest to
remark that the hardness of a " D " grade asphalt is proportional,
as in the native bitumens, to the percentage of naphtha soluble
bitumen which it contains as appears from the following deter-

RESIDUAL PITCHEa

265

"D" GRADE ASPHALT FROM AN ASPHALTUM OIL AND
REFINING CO.

PHYSICAL PROPERTIES.

Test number 69014

Specific gravity, 78 F./78 F., original substance, dry 1 .077

Softens 195 F.

Flows 205 F.

Penetration at 78 F 27

CHEMICAL CHARACTERISTICS.

Original substance:

Loss, 212 F. , 1 hour * 0.0%

Dry substance:

Loss, 325 F. , 7 hours 1.3%

Character of residue Surface

smooth.

Loss, 400 F., 7 hours, additional loas. 5.3%

Character of residue Shrivelled

surface,
penetration
5.

Bitumen soluble in CS,, air temperature 92 .6%

Difference (largely carbon) 7.3

Inorganic or mineral matter .1

100.0
Malthenes:

Bitumen soluble in 88 naphtha, air temperature 64 . 4%

This is per cent of total bitumen 69 . 5

Bitumen soluble in 62 naphtha 65 . 6%

This is per cent of total bitumen 70 . 8

Carbenes :

Bitumen insoluble in carbon tetrachloride, air temperature 13.1%

Bitumen yields on ignition:

Fixed carbon 19 .0%

REMARKS: A small amount of suspended matter is noted under the
microscope.

266

THE MODERN ASPHALT PAVEMENT.

minations on the products produced at one plant under the same
conditions :

Number

103

104

105

106

Penetration at 78 F

Naphtha soluble bitumen. . .

66.0%

70.6%

72.4%

Asphaltic Residues from Texas Oil. The semi-asphaltic oil
from the Beaumont field in Texas leaves a residue of solid bitu-
men on distillation which, however, as in the case of California
oil, varies in character according to the method of distillation
employed. In the case of California oils, with careful distillation
a larger percentage of residue was obtained than was the case indus-
trially. With the Beaumont oil the reverse is the case; on dis-
tillation in vacuo but 9.0 per cent of solid bitumen was recovered
while industrially as much as 30 per cent is obtained. This is,
probably, due to the fact that condensation goes on in the case
of the Beaumont oil instead of cracking as in the case of California
petroleum. Analyses of residual pitches from Texas petroleum
representing the product as turned out in 1903 and again in
1907, are given on the opposite page, the two materials originat-
ing from different manufacturers.

An examination of the preceding results shows that the dis-
tillation has been carried much further than is the case in the
production of the California asphalts. This is evidenced by the
greater density of the product and the very much higher per-
centage of fixed carbon which it yields. It should also be noted
that the two forms of residual pitch are differentiated by the fact
that that from the Texas oil contains a larger percentage of satu-
rated hydrocarbons than that from the California oil, a fact which
might be expected as the stability of the former is much greater
than that of the latter, owing to the amount of paraffine hydro-
carbons which it contains. The asphaltic residue from the Texas
oil is marked by the presence of a little over 1 per cent of paraffine
scale, but the amount is insufficient to give it the character of a
paraffine material.

The residual pitches from Texas oil are no more uniform in

RESIDUAL PITCHES.

267

RESIDUAL PITCH FROM TEXAS PETROLEUM.

1903
68943

1907
102529

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original sub-
stance dry

1.0803

1.0703

Black

Lustrous

Lustrous-

Uniform

Semi-conchoidal

Conchoidal

-1

Odor

Asphaltic

Petroleum

Softens

230 F.

143 F.

Flows

247 F.

164 F.

Penetration at 78 F , Bo wen

Penetrometer at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance:
Loss 325 F , 7 hours

.13%

95%

Loss 400 F., 7 hours

.19

iSr

Bitumen soluble in CSj, air temperature . . .
Inorganic or mineral matter

99.0%

98.2%
Trace

Difference

1.8

Malthenes:
Bitumen soluble hi 88 naphtha, ah- temp.
This is per cent of total bitumen

100.0

65.4%
66.1

100.0

69.6%
70 9

Per cent of soluble bitumen removed by
H 2 SO 4 .

32.1

51.1

Per cent of total bitumen as saturated
Hydrocarbons . ....

44 8

65 6

Bitumen soluble in 62 naphtha

71.5%

70 1%

72.2

7i:4

Carbenes:
Bitumen insoluble in carbon tetrachloride,

5.1%

14.6%

Bitumen yields on ignition:

24.0%

19.5%

1.2%

.8%

268

THE MODERN ASPHALT PAVEMENT.

character than those prepared from California petroleum and, no
doubt, for the same reason. That they are very variable in char-
acter can be seen from the results of an examination of five
samples, taken from one delivery, which were submitted to the
author for examination.

RESIDUAL PITCH FROM BEAUMONT, TEXAS, PETROLEUM
TAKEN FROM ONE DELIVERY.

72550

72589

72590

72591

72592

Penetration at 78 F./78 F

Flow

None

100 0%

76 0%

86 0%

98.1%

97 7%

99 3

99 1

Difference

1.8

2.2

Inorganic or mineral matter.

Carbenes:
Bitumen insoluble in carbon
tetrachloride,air temperature.

Bitumen yields on ignition:
Fixed carbon

100.0
10.5%

100.0

12.7%
23.0%

100.0
6.7%

100.0
7.0%

100.0
6-7%

In this delivery material was found which was so hard as to
hardly flow at 212 F. (No. 72550) and so soft as to be readily
melted (No. 72590).

Other lots which have been examined by the author have
shown an equal lack of uniformity, as can be seen from the follow-
ing figures:

RESIDUAL PITCH FROM BEAUMONT, TEXAS, PETROLEUM.

Test number

63526

63527

63528

Penetration at 78 F./78 F

Too soft

110

98.3%

96.6%.

95.7%

Malthenes :
Bitumen soluble in 88 naphtha, air tempera-
ture

78.2%

72.2%

67.9%

This is per cent of total bitumen

79.6

74 7

Carbenes :
Bitumen insoluble in carbon tetrachloride,

8 6%

12.8%

12 5%

Bitumen yields on ignition:
Fixed carbon. .

14.5%

17.6%

21.1%

RESIDUAL PITCHES.

269

The most important characteristic of the residual pitches from
Texas oil is that they yield, as prepared, and as found on the
market, more than 20 per cent of fixed carbon as compared with
15 per cent for the California pitches. This characteristic while
it may be due somewhat to the fact that the Texas pitch is a denser
material, because the distillation has been carried to a point beyond
that to which the California oil is submitted, is an important one
industrially as it makes it possible to differentiate and determine
the origin of any of these forms of bitumen. The two can also be
differentiated by determining whether paraffine is present, none
being found in the California products and about 1 per cent in
those from Texas.

Examples of the variation in the character of Texas residual
pitches as revealed by the percentage of carbenes which they con-
tain is shown by the folio whig analyses:

RESIDUAL PITCHES FROM BEAUMONT, TEXAS, PETROLEUM.

Bitumen Insolu-

Test Number.

ble in Carbon
Tetrachloride,

Air Temperature.

63526

8.6%

63527

12.8

63528

12.5

68943

5.1

69015

5.1

72550

10.5

72589

12.7

72590

6.7

72591

7.0

72592

6.7

The amount is generally much larger than is found in the more
carefully prepared California " D " grade asphalt and points to
overheating in the preparation of these particular specimens.
This is much greater in the pitch which has been produced in 1907
than in the supply of 1903. The material has recently been of
very poor quality, and, in the opinion of the author, it is entirely
unsuitable for use in the paving industry.

270

THE MODERN ASPHALT PAVEMENT.

Ebano Asphalt. A residual pitch made from the asphaltic
oil or maltha occurring at Ebano, Mexico, about thirty miles
from Tampico, on the line of the Mexican Central Railroad, has
been on the market in the United States and abroad, for a few
years, under the designation of "Ebano Asphalt, "a well sunk
at that point to a depth of 1600 feet, producing from 1200 to
1500 barrels of petroleum a day. It corresponds in its preparation
and characteristics to the California residual pitches. Following
are the results of examinations of various grades of the material:

T 1 6st number .

102261

102262

102263

102264

"E"

"DX"

"D"

"B"

PHYSICAL PROPERTIES.

Penetration at 78 F. Bowen

Softens

125 F.

185 F.

190 F.

260 F.

163 F.

200 F.

206 F.

278 F.

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 , air temp

99.4%

97.9%
2

97.8%
none

95.8%
3

1.9

2 2

3 9

Carbenes:
Bitumen insoluble in carbon tetra-
chloride, air temperature

100.0%

.4%

100.0%
14.4%

100.0%
15.5%

100.0%
19.7%

Malthenes:
Bitumen soluble in 88 naphtha, air
temperature

69.0%

66.0%

55 5%

48 1%

This is per cent of total bitumen ....
Per cent of soluble bitumen removed
byH 2 SO 4

69.4
89.0

66.5
91.0

56.3
92.0

50.2
93.0

Bitumen yields on ignition:
Fixed carbon

19.2%

23.9%

24 9%

30 5%

Paraffine scale

1.0%

1.3%

1.9%

Where the Ebano asphalt is made of " E " grade consistency,
it seems to be of a character as good as that of the same material
made in California, but where the distillation has been carried to
such a point that the consistency is much harder, the percentage
of carbenes present shows that the bitumen has been, at least
in some cases, very seriously decomposed and such a product
is, on this account, very unsatisfactory.

Baku Pitch. On the Continent a residual pitch from the
distillation of Russian petroleum is an industrial product. This
hae the following composition:

RESIDUAL PITCHES. 271

BAKU PITCH.
Test number 63200

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 1 . 1098

Color of powder or streak Black

Lustre Lustrous

Structure Uniform

Fracture Semi-

conchoidal

Hardness, original substance 1

Odor Petroleum

Softens 140 F.

Flows 150 F.

Penetration at 78 F 10

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 , air temperature 91 . 6%

Difference 8.4

Inorganic or mineral matter Trace

100.0
Malthenes:

Bitumen soluble in 88 naptha, air temperature 54.6%

This is per cent of total bitumen 59 . 6

Per cent of soluble bitumen removed by H^C^ 44 . 1

Per cent of total bitumen as saturated hydrocarbons 33 .3

Bitumen soluble in 62 naphtha 61 . 3%

This is per cent of total bitumen 66 . 9

Carbenes:

Bitumen insoluble in carbon tetrachloride, air temp 10 . 4%

Bitumen yields on ignition:

Fixed carbon 26.8%

Paraffine scale 1 . 7%

This pitch has a comparatively high density, yields a large per-
centage of fixed carbon, bitumen insoluble in cold carbon tetra-
chloride and much organic matter not bitumen; showing that the
distillation has been pushed to an extreme. It contains 1.7 percent
of paraffine scale. It is a remarkable fact that the softening-point
of this material is much nearer that of the California residue than of

272 THE MODERN ASPHALT PAVEMENT.

that from Beaumont, Texas, oil. It might, perhaps, be possible
to use a small amount of this bitumen in the paving industry as
an amendment to some asphalts.

Solid Bitumens the Product of the Condensation of Heavy
Oils. Another class of bitumens are the artificial ones obtained
by the treatment of any of the fluxes which have been described
with sulphur or oxygen at high temperatures. In then- uses and
consistency they may be ranked between the fluxes and the solid
bitumens.

Pittsburg Flux. The first bitumen of this description to be
put upon the market was known as Pittsburg Flux. It was made
by adding to an ordinary Pennsylvania petroleum residuum about
one pound of sulphur to every gallon of oil and heating the
material to a point a little below that of distillation and main-
taming it at that temperature until the evolution of hydrogen
sulphide ceases. The residuum is, in this way converted into a
semi-solid cheesy bitumen which is very short, that is to say,
has little ductility, and is very slightly susceptible to changes of
temperature. The reaction which takes place is represented by
the following equation:

The reaction, in reality, is not as simple as this but the result is
explained as well. Two molecules are condensed to one with
the accompanying evolution of hydrogen sulphide gas and with
the resulting changes in the properties of the bitumen. The great
expense incurred for sulphur in this process made it necessary to
utilize the by-product of hydrogen sulphide. This was done by
converting it into sulphuric acid. The business was not profitable
even under these conditions and was soon abandoned. The mate-
rial could not be used as the principal source of bitumen in making
an asphalt cement, being too short, and only as an addition, in
small amounts, to the ordinary asphalts. Used hi this way it
has been successful in one or two instances.

An analysis of this material is presented in the table on page 275.

Ventura Flux. Later on an attempt was made to make a
similar substance from the asphaltic petroleum of California. The

RESIDUAL PITCHES. 273

product was a slight improvement on the Pittsburg Flux but pave-
ments made with it without the addition of native solid bitumen
were failures in Allegheny, Pa. Its manufacture was abandoned
after a few years.

Byerlyte. In the meantime Byerly, of Cleveland, had found
that the oxygen of the air was as satisfactory a condensing agent
as sulphur, imitating the practice of blowing certain vegetable
and fish oils in order to thicken them and give them greater viscosity.
He produced a substance similar to Pittsburg Flux by drawing
air through residuum while the latter was maintained at a high
temperature. Depending on the length of time during which
the air was allowed to act the product was soft or as hard as pitch.
This material has been used to some extent in making asphalt
blocks in Washington, but even this use has now been abandoned.
In mixture in small proportion with the native solid bitumens
it can be used but there is no advantage in doing so commen-
surate with the expense involved in the treatment of the original
residuum.

Hydroline " B." Still more recently the asphaltic residuum
from the asphaltic petroleums of Texas has been put upon the mar-
ket, after having been blown, under the name Hydroline- " B."
It possesses in this form no qualities which could recommend it
very strongly for paving purposes except, perhaps, as an amend-
ment to certain inferior asphalts and it need hardly be considered
here.

Sarco Asphalt. This material, which has recently been on
the market, is a blown product like Hydroline ' *B," prepared
from Kansas residuum, and it possesses much the same properties,
having little ductility and being very short.

A specimen recently examined in the author's laboratory,
showed the following characteristics:

Test number 96891

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F., original substance, dry 981

Softens 190 F.

Flows 210 F.

Penetration at 78 F. Bowen 120

Condition at 140 F. after 1 hour {

274 THE MODERN ASPHALT PAVEMENT.

CHEMICAL CHARACTERISTICS.

Bitumen soluble in CS 2 , air temperature 99 . 1%

Inorganic or mineral matter .7

Difference .2

100.0%
Malthenes:

Bitumen soluble in 88 naphtha, air temperature 69 .8%

This is per cent of total bitumen 70 . 4

Carbenes:

Bitumen insoluble in carbon tetrachloride, air temperature . 2%

Bitumen yields on ignition:

Fixed carbon , 10 .3%

Paraffine scale 12 .3%

It is notable for the very high percentage of paraffine scale
which it contains, which, in addition to the properties conferred
by blowing the oil with air, adds to its shortness. It is made
of several grades, having a penetration of from 50 to that of the
sample which was examined.

According to analyses made by Mr. Francis P. Smith, three
specimens which he examined had the following properties:

Test number 1175 1176 1210

Melting-point, degrees F 171 188 198

Penetration at 32 F.(Dow, 1 rain., 200 gms.) ... 42 49 31

Penetration at 77 F. (Dow, 5 sec., 100 gms.) ... 60 70 45

Penetration at 115 F. (Dow, 5 sec., 50 gms.) . . 97 85 71

Bitumen soluble in CS, . 98.0% 97.7% 98.7%

Organic insoluble 1.0 1.8 1.3

Ash 1.0 .5 Trace

Naphtha soluble bitumen 79 .9% 79 .0% 79 . 2%

Naphtha soluble bitumen (per cent total bit.) ..81.5 80 . 9 80 . 3

Loss on heating 5 hours at 325 F 08% .4% . 15%

Penetration after heating 60 61 33

Ductility 3 cm. 3 cm. 2 cm.

Fixed carbon 10.6% 10.8% 10.8%

It is hardly possible that a satisfactory pavement could be
made with such material, although it might be combined to a
certain extent with the ordinary asphalts.

The character of Pittsburg Flux ; Byerlyte and Hydroline " B "
is shown in the table on page 275.

From these figures it appears that the materials are nearly
pure bitumens and that, not having been subjected to suffi-
ciently high temperatures to produce cracking, the amount of
bitumen insoluble in carbon tetrachloride is practically nothing.
According to their derivation the materials carry more or less
paraffine but the Hydroline " B " being derived from a Texas oil

RESIDUAL PITCHES.

275

6123

Byerlyte

(paving)

1.023
Black
Dull
Uniform
Cheesy
Petroleum
245 F.
294 F.
63

.93%
Smooth

5-9%
Smooth

99.7%
.3
.0

100.0

62.0%
62.2

25.5
46.3

67.2%
67.4

18.0%
4.6%

6124

Byerlyte

(roofing)

.9070
Black
Dull
Uniform
Cheesy
Petroleum
230 F.
254 F.
107

1-8%
Smooth

6.5%
Smooth

99.>%
.5
.0

71436

Hydro-
line "B".

1.0043
Black
Dull
Uniform
Cheesy
Petroleum
206 F.
220 JF.
55

1.0%

Smooth
penetra-
tion, 45

5.8%
Smooth
penetra-
tion, 40

99.9%
.1
.0

Pittsburg
Flux

.9879
Black
Dull
Uniform
Cheesy
Petroleum
295 F.
353 F.
74

1.7%
Smooth

4.4%
Blistered

99.7%
.3
.0

PHYSICAL PROPERTIES.

Specific gravity, 78 F./78 F.,
original substance, dry

Color of powder or streak

Lustre

Structure . .

Fracture

Odor

Softens

Flows

Penetration at 78 F

CHEMICAL CHARACTERISTICS.

Dry substance :
Loss 325 F 7 hours

Character of residue .

Loss, 400 F., 7 hours (fresh
sample)

Character of residue

Bitumen soluble hi CS* air temp.
Difference . .

Inorganic or mineral matter

Malthenes :
Bitumen soluble in 88 naphtha,
air temperature

100.0

67.1%
67.3

14.8
57.3

71.5%
71.7

.3%

14.7%
10.3%

100.0

66.8%
67.1

17.4
55.5

72.0%
72.3

14.3%

5.7%

100.0

69.3%
69.4

12.7
60.6

.5%

12.2%
1-0%

This is per cent of total bitumen.
Per cent of soluble bitumen re-
moved by H 2 SO 4
Per cent of total bitumen as sat-
urated hydrocarbons

Bitumen soluble in 62 naphtha
This is per cent of total bitumen.

Carbenes :
Bitumen insoluble in carbon te-
trachloride, air temperature. .

Bitumen yields on ignition :
Fixed carbon

Paraffine scale

276 THE MODERN ASPHALT PAVEMENT.

contains no more than is found in the residual pitch from
the same oil. It is worthy of remark as to Byerlyte that,
although made from paraffine oil, it contains much less paraf-
fine scale than would be expected, and would point to the fact
that this material has become altered in the process of manu-
facture.

With none of these materials, as the principal constituent of
a paving cement, is it possible to produce a satisfactory surface
mixture. They are all too short, but they may be used as an
amendment in an amount not exceeding 10 per cent. Owing to
the fact of their great lack of susceptibility to change in con-
sistency within wide ranges of temperature they present some
advantages.

Differentiation of the Residual Pitches from the Natural
Asphalts. The residual pitches, it appears from the preceding
data, contain practically no mineral matter. With only one excep-
tion there is no native bitumen in use in the asphalt paving indus-
try which has the same characteristic. It is, therefore, possible
to differentiate, except in the case of gilsonite, an oil asphalt,
so-called, from a native bitumen by determining the amount of
mineral matter present. The mineral matter in the latter is
generally of a ferruginous nature while that derived from the
native bitumens generally contains silica. A microscopic exam-
ination of the residue left on ignition will, therefore, aid in the
determination. Even in the case of gilsonite the color of the
ash is quite different from that obtained from the residual pitches.
Unfortunately the amount of fixed carbon which the California
" D " grade asphalt yields and that from the native bitumens
is so nearly the same that this characteristic cannot be success-
fully used, although the amount obtained may be of value as
indicating the presence of grahamite which in itself has a high
fixed carbon. The native bitumens carry, however, less bitu-
men insoluble in cold carbon tetrachloride but soluble in carbon
disulphide than the residual pitches, unless the latter are very
carefully made, and in case of doubt the differentiations of the
two classes of materials may be assisted by comparative deter-
minations of this form of bitumen.

SUMMARY.

All petroleums on evaporation under suitable conditions leave
a pitchy residue. The residue from the asphaltic or semi-asphaltic

RESIDUAL PITCHES. 277

petroleums resembles the native asphalts. The principal supplies
available for use in the paving industry are residual pitches from
California and from Texas oil. These are each made in such a
careless way that they consist largely of alteration products of
the original hydrocarbons as shown by the lack of solubility of
some of their constituents as compared with those found in the
original oil. On this account the material is not always satisfac-
tory and, moreover, requires great skill to use it.

The residual pitches from Texas and from California oils can
be readily differentiated by certain characteristics, that from the
Texas oil generally yielding a higher percentage of fixed carbon
than the pitch obtained from the California oil.

The blown or oxidized petroleum residues are characterized
by their lack of susceptibility to temperature changes, but are
extremely short, which prevents their use as the main source of
bitumen in the paving mixture. Pavements laid with them have
generally eventually proved failures. They may possess some
desirable qualities as an amendment to the native asphalt to
an extent not exceeding 10 per cent, and, used in this way, should
be considered as fluxes.

CHAPTER XIV.

COMPARISON OF VARIOUS NATIVE ASPHALTS AND THEIR
RELATIVE MERITS FOR PAVING PURPOSES.

In attempting to form an opinion on the availability of any
native bitumen for paving purposes a number of things must be
taken into consideration, which may be tabulated as follows:

1. The quantity available.

2. Its uniformity in character.

3. Its stability in a melted condition at high temperatures.

4. Its stability in consistency at the extremes of temperature
which it meets in an asphalt pavement.

5. The proportion of malthenes to asphaltenes which it con-
tains.

6. The proportion of flux which is required to make an asphalt
cement.

7. Mineral matter present and its character.

i. The Quantity Available. No native bitumen can be of any
great importance in the paving industry without a large supply
of ^t is available. Pavements can no doubt be constructed of a
bitumen of which not more than 500 to 1000 tons can be gathered
together with difficulty in any one year, but such supplies are too
unreliable to permit of their being of permanent interest. There
are hundreds of such deposits in which many thousands of dollars
have been sunk without any adequate return for the investment.
A deposit to be of any great value should afford a supply of at
least 50,000 tons annually without difficulty. The first thing
to be done, therefore, in considering the availability of native bitu-
men is to learn whether the deposit is such that the amount which

278

COMPARISON OF VARIOUS NATIVE ASPHALTS. 279

can be obtained from it will prove large enough to be of industrial
importance.

2. Its Uniformity in Character. A great consideration in the
turning out of a regular asphalt surface mixture is that the bitu-
men from which it is made shall be of such a nature that every
cargo or shipment of it may be exactly like all others. If this
is not the case each lot will, of necessity, require a different method
of handling, which necessitates great experience and skill which
are not always to be found among those who are engaged in the
laying of asphalt pavements.

3. Its Stability in a Melted Condition at High Temperatures.
As the asphalt cement made from the native bitumens as they
occur in the refined condition in the trade is necessarily main-
tained in a melted condition at high temperature for considerable
periods of time it is equally important that it should consist of a
bitumen which does not become changed in consistency, under
these conditions, owing to the rapid volatilization of certain of
its constituents.

There is a very decided difference in the behavior of different
bitumens in this respect and this should be borne in mind in deter-
mining whether one has a preference over another for use in the
construction of asphalt pavements.

4. Its Stability in Consistency at the Extremes of Tempera-
ture which it Meets hi an Asphalt Pavement. There is a difference
in the behavior of different bitumens, as far as their consistency
is concerned, at the extremes of temperature which are met with
in summer and winter, that is to say, some of them are much
more susceptible to changes in consistency between very low and
very high temperatures than others. This is an important con-
sideration since, although a given bitumen may enable one to con-
struct an asphalt surface which is of proper consistency at medium
temperature, say 78 F., it may become extremely hard and brittle
at zero or extremely soft and oily at 120 F., a temperature which
asphalt surfaces frequently reach under our hot summer sun.

5. The Proportion of Malthenes to Asphaltenes which it Con-
tains. The relative proportion of malthenes, those constituents
which are soluble in light petroleum naphtha, and of asphaltenes,

280 THE MODERN ASPHALT PAVEMENT.

the other constituents not soluble in this medium, has a bearing
upon the availability of any native bitumen for paving purposes.
Although to-day the deficiencies in this respect may be modified
by the use of certain fluxes, which supply the missing constituents,
this is not always the case and, when it is so, it requires very
considerable skill to accomplish it. This has been illustrated more
fully in another place. 1

6. The Proportion of Flux which is Required to Make an
Asphalt Cement. The question of the amount of flux which is
necessary to use with any bitumen is one of importance. If the
native bitumen is so hard as to require a very large percentage
of flux there is a very great probability, although this is not uni-
versally the case, that the resulting cement will be too oily and
too susceptible to high temperature.

Gilsonite is a bitumen which is an exception to this rule. To
flux this material in making an asphalt cement, about equal parts
of it and of heavy asphaltic oil are necessary, but the cement
contains normal proportions of malthenes soluble in 88 naphtha,
and of asphaltenes insoluble. Such a cement has an extremely
rubbery consistency and a considerable amount of elasticity,
a cylinder of the cement springing back to its original formation
when bent nearly double. Such a cement, in addition, is not as
susceptible to temperature changes as those made from the ordi-
nary asphalts. A statement in regard to the relative proportions
of various fluxes which must be combined with the various asphalts
to produce cements of different penetrations is given on page 308.

7. Mineral Matter Present and its Character. The mineral
matter present in any native bitumen may be desirable or unde-
sirable. If it contains so much or if it is so coarse as to render
it impossible to maintain it uniformly in suspension in the melted
asphalt cement which is prepared from the bitumen it is undesir-
able. If, on the other hand, it is extremely fine and acts as a
filler, as in the case of Trinidad asphalt, it is very desirable.

The fact that it may reduce the percentage of bitumen present
is of no importance, since in the case of an asphalt consisting of

1 See pages 139 and 178.

COMPARISON OF VARIOUS NATIVE ASPHALTS. 281

99 per cent of bitumen it will be necessary in building up a satis-
factory surface mixture to add a certain amount of filler which
cannot be done as successfully by any artificial means as is done
by nature.

In this connection the following correspondence between the
President of the Board of Public Works of a western city and a
local chemist, many years ago, may prove of interest as well as
the latter's answers to several other questions which are frequently
asked.

" May 22, 1893.

" DEAR SIRS: A discussion has been going on in this city,
which the citizens are largely interested in, in regard to asphalt
paving.

" It is claimed on one hand that asphalt is asphalt, no matter
where it is found, and the only difference in asphalts is in the
amount of bitumen which they contain. As a well-known and
practical chemist in this city, I would thank you to answer the
following questions, and send me a bill for your expert opinion.

" 1. Does the percentage of bitumen determine the value
of an asphalt for paving purposes?

"2. May or may not an asphalt contain a very large percent-
age of bitumen and still be worthless for paving purposes?

" 3. Might or might not an asphalt, which in its natural state
is good for paving purposes, be so destroyed by heat that it is
practically worthless for paving purposes, and still the material
after subjection to heat, be asphalt?

" 4. Might or might not two asphalts contain the same amount
of bitumen, and one be so unstable that it will not stand expo-
sure to the sun, and the other be comparatively permanent?

" 5. Is it possible that one asphalt might contain twice as
much bitumen as another, and still be far inferior for street con-
struction to the one containing a less quantity?

" 6. Is there any real system by which a chemist can tell to a
certainty, by analysis, whether a given asphalt which has never
been tried will make as good, permanent, and durable a pave-
ment as another which has proved a success?

" 7. Are there or are there not qualities required of an asphalt

282 THE MODERN ASPHALT PAVEMENT.

for paving purposes which makes it impossible for a chemist who
has not made the subject a special study, to state for a certainty
whether a given untried asphalt will make as good a pavement
as another asphalt which has proved a success?

" 8. What is the real test of standard or quality which will
give the value of an asphalt for paving purposes?

"9. Might an asphalt pavement stand for one or two years,
and fail from effect of elements in succeeding years, and might
two asphalts stand equally for two years and show marked differ-
ences in wear in succeeding years?

" An early reply will oblige,

"Yours respectfully,

(Signed) "PRESIDENT BOARD OF PUBLIC WORKS.'!

In reply to the above questions the following opinion was
rendered :

"1st Answer. The percentage of bitumen does not determine
the value of an asphalt for paving purposes.

" 2d Answer. An asphalt might contain a very large percentage
of bitumen, and still be comparatively worthless for paving pur-
poses.

" 3d Answer. An originally good asphalt for paving purposes
might be so altered by heat as to be practically worthless, and yet
the altered material would still be asphalt in the sense that it could
not be distinguished from asphalt, notwithstanding its marked
inferiority to the particular asphalt from which it was produced.

" 4th Answer. Two asphalts might contain the same amounts
of bitumen and yet possess entirely different powers of resistance
to the destructive action of the elements. One might thus
be comparatively permanent and stable, and the other greatly in-
ferior.

" 5th Answer. As the percentage of bitumen in an asphalt
does not determine its value for paving purposes, it is quite pos-
sible for one asphalt to contain a much higher percentage than
another and yet be decidedly inferior for making a durable pave-
ment.

" 6th Answer. There is no system of chemical analysis that will

COMPARISON OF VARIOUS NATIVE ASPHALTS. 283

determine for a certainty that a given untried sample of asphalt
will make, in every way, as good a pavement as another asphalt
which has proved a success.

"7th Answer. The requirements of an asphalt for paving
purposes are of such a peculiar nature that it would be impossible
for a chemist who had not made the subject a special study to
state with certainty, from the results of analysis, whether or not a
given sample would make as good a pavement as an asphalt which
has proved a success.

" 8th Answer. The real and final test of the quality of an
asphalt for paving purposes is actual trial for a proper length of
time. Proper chemical and physical tests of a new variety of
asphalt may strongly indicate its probable value as a paving mate-
rial, but these tests, though of great assistance in forming an
opinion, really only show the advisability of submitting the asphalt
to the final and infallible test of actual trial.

" 9th Answer. A test of one or two years under any condition
demonstrates only that that particular asphalt pavement is good
for that length of time under those conditions, and does not demon-
strate how much longer it will last under the same conditions or
whether it will last as long under other or more unfavorable con-
ditions. Two asphalt pavements might endure equally well for
a given short time, and yet show decided difference under a long
trial.

"Having thus briefly answered the questions asked it may,
perhaps, be well to give some explanation of the subject, in order
to indicate the reasons for the opinions expressed. First of all,
it may be stated that asphalt is not a chemical compound or mineral
of fixed and invariable composition. According to Dana it is a
mixture of hydrocarbons, and the asphalts of different localities
have various compositions. Mineralogically, bitumen is simply
another name for asphalt or asphaltum. In paving parlance,
however, bitumen has come to mean only the pure portion, so to
speak, of the asphalt, the latter term being applied to the entire
mixture of earthy and other impurities with the true bitumen.
This view of bitumen having evidently been taken in the ques-
tions asked, it was similarly considered in the replies. It is to

284 THE MODERN ASPHALT PAVEMENt.

be understood, then, that asphalt is an impure bitumen, and that
bitumen is the pure article considered by Dana in his Mineralogy.
But, as before stated, bitumen has no fixed composition or com-
bination of qualities. Its nature and physical properties are as
various as the localities where it is found. It can be no more
strictly defined than coal. It is simply a mixture of various hydro-
carbons, and may be either a solid or a liquid. Two bitumens of
precisely similar percentage composition may have widely different
properties, so that while one would furnish a most excellent paving
material the other would be practically worthless. Such instances
of substances of entirely dissimilar nature having the same per-
centage composition abound in chemistry. Charcoal, the diamond,
and plumbago, or black lead, may be mentioned as a familiar
example. Light naphtha or gasoline and solid paraffine is another.
It takes more than an ordinary chemical analysis to distinguish
between such substances. Evidently, then, the mere percentage
of bitumen in an asphalt would not determine its value for paving
purposes, for this bitumen might have a consistency varying any-
where from a non-cohesive liquid to a brittle worthless solid. By
the action of the elements all asphalts undergo change. This
change is due to oxidation, volatilization and other molecular
disruption, and tends to produce greater solidification or apparent
drying, and the asphalt may pass through all the stages of brittle-
ness to final crumbling or disintegration. In all these stages the
substance is still asphalt, although at many points it is evidently
worthless as a paving material. While these changes are slow
in nature, some of them may be greatly hastened by the applica-
tion of heat, as in incautious or unskilful refining so as to greatly
injure an originally good asphalt. It is evident, also, that an
asphalt may be so far gone in the process of natural decay that,
while it may serve to make what appears to be an excellent pave-
ment, the life of such a pavement must be comparatively short.
" Having thus shown how much depends upon the quality or
nature of the bitumen in an asphalt rather than upon its mere
percentage, it becomes important to know to what extent the
chemist can distinguish this valuable quality, and so prevent dis-
astrous mistakes in pavement work. It may be answered that a

COMPARISON OF VARIOUS NATIVE ASPHALTS. 285

chemist who has made a special study of the subject can, by proper
chemical analysis, aided by certain physical tests, point out what is
probably good or worth trying in the case of new varieties, but it
is impossible for him to state for a .certainty that a particular new
variety will be fully equal in every essential respect to some stand-
ard asphalt that has proved a success. Having learned by experi-
ence the chemical and physical differences between good and bad
samples of any particular asphalt, the chemist may thereafter
afford valuable assistance in the use of that asphalt.

" In view of the foregoing facts it would seem that the extensive
use, for paving purposes, of any variety of asphalt that has not
previously been proven a success by the test of actual trial for a
sufficient length of time, under sufficiently adverse conditions, is
in the nature of a rather hazardous experiment.

" Trusting that the above answers and explanations will prove
clear and satisfactory, we will add that they are given without
prejudice and according to our best knowledge of the subject,

" Yours respectfully,
(Signed) " CHEMIST."

Action of Water on Asphalt in the Laboratory. It has fre-
quently been claimed that there is a preference for one asphalt
over another based upon the manner in which it behaves towards
water when it is placed in contact with it in the laboratory. From
data which will be given elsewhere l it appears that this method
of examining them is not one the results of which are confirmed
by practice. All asphalts are attacked by water under certain
environments and some more than others under certain laboratory
conditions. In practice, however, the results obtained in the
laboratory are not confirmed if the asphalts are employed so as
to bring out their most desirable qualities.

Bearing in mind all these considerations it is not difficult to
form a decided opinion as to the desirability of any native bitu-
men for the uses to which it is put in the paving industry.

1 See page 460.

286 THE MODERN ASPHALT PAVEMENT.

CONCLUSIONS.

From the preceding data and discussion it is very evident
that while many native bitumens may be denominated asphalt,
from an industrial point of view, they possess no great uniformity
in their physical and chemical properties and that some of
them are far preferable to others for paving purposes. Some
of them are extremely stable bitumens while others are more
or less changeable on the application of heat. Some of them
are hard, others are comparatively soft. Some evolve gas on
heating, showing that they are unstable. Some lose on heating
a considerable amount of light hydrocarbons, petrolenes, with
corresponding hardening in the consistency of the material. Some
asphalts are obtainable in unlimited amounts and of great uni-
formity in composition. Others, while obtainable in large amounts,
are very variable in their consistency, the character of no two
shipments corresponding in this respect. Some asphalts, such as
those which are obtained by collecting the exudation from maltha
springs, are not only very variable in their character but, being
still in a state of transformation from maltha to asphalt and,
therefore, not in equilibrium, are unsatisfactory materials for use
in the paving industry or require such great skill or judgment in
their treatment as to make it difficult to construct good work
with them.

From the results of the author's experience with all the bitu-
mens which have been used in the construction of asphalt pave-
ments during the last eighteen years the conviction has been
forced upon him that none of them, with the exception of gilsonite,
which has recently proved itself under service tests to be a very
desirable material, is as uniformly satisfactory as that obtained
from the Trinidad pitch lake, and for the following reasons:

1. The available supply is unlimited.

2. The supply is of great uniformity, as appears from data
given in the preceding pages.

3. Asphalt cements prepared from Trinidad lake asphalt and
stable flux are less liable to change in consistency when main-
tained in a melted condition at high temperatures for any con-
siderable length of time or on being tossed about in a mixer with

COMPARISON OF VARIOUS NATIVE ASPHALTS.

287

excessively hot sand, something that unfortunately happens too
frequently, than one derived from any other form of bitumen.

4. It is less susceptible to changes in consistency at extremes
of temperature than any other native bitumen which is now used
extensively in the construction of asphalt pavement.

5. The relation of malthenes to asphaltenes is such that the
proportion of flux which is necessary to produce an asphalt cement
of normal consistency is not excessive.

6. The mineral matter which it contains is of a nature most
suitable to play the role of a filler and it is mixed by nature with
the bitumen in a way that it is impossible to imitate by adding
finely powdered mineral matter to a purer form of bitumen.

Trinidad mixtures, when the mineral aggregate is properly
graded and regulated, are not attacked by water to any greater
extent on the street than mixtures made with other asphalts. In
fact surface mixtures of Trinidad asphalt resist impact more sat"
isfactorily after three months exposure to running water than
those made with Bermudez asphalt, as shown by the following
figures:

IMPACT TESTS OF ASPHALT SURFACE MIXTURES.

New

York.

Trinidad.

Bermudez.

2.24

Number of blows :

21-20

16-14

After 3 months' exposure to running water
Water absorbed:

20
.129

13
.157

Bermudez asphalt possesses the disadvantage that it is far from
uniform in character, that the bitumens of which it consists are
susceptible to volatilization at high temperature with a resulting
hardening of the material, as for example when it is mixed with

288 THE MODERN ASPHALT PAVEMENT.

very hot sand ; that is to say, it does not form an asphalt cement
which can be maintained at high temperatures or mixed with
sand at high temperatures satisfactorily and for this reason cannot
be used in- cold weather, and because it is deficient, in com-
parison with Trinidad asphalt, in mineral matter forming a natural
filler. As has already been shown, surface mixtures made with
Bermudez asphalt are more deteriorated by the continued action
of water, as far as their resistance to impact is concerned, than
those made with Trinidad asphalt.

Maracaibo asphalt is not a normal asphaltic bitumen and
possesses characteristics which throw a doubt upon its suitability
for the preparation of a paving cement, which can only be removed
by a study of its behavior after a long period of years in actual
practice.

Mexican asphalts are far from uniform and possess the same
disadvantages that pertain to Bermudez asphalt. Their use would
involve greater care and skill than any of the materials that have
been mentioned.

Cuban asphalts are very hard materials, approaching grahamite
in composition, and must be fluxed with very large proportions
of asphaltic oil. Their value as paving materials has never been
satisfactorily demonstrated.

The solid residuals from asphaltic or semi-asphaltic oils are
far from uniform and are generally somewhat damaged or cracked
in the course of their preparation. The closest scrutiny of these
materials in the laboratory and the greatest skill in handling them
is necessary to enable them to be used satisfactorily in the con-
struction of asphalt surface mixture.

Gilsonite has now been submitted to sufficient service test
to show that as a paving cement it is equal in character to any
asphalt in use for this purpose. A pavement was laid with it
in 1903 on Missouri Avenue, Main to Delaware, Kansas City,
Mo., under the author's direction, and it has proved entirely
satisfactory. Since that time it has been used equally successfully
to a large extent in other cities. Gilsonite may therefore be
added to the list of those bitumens which are entirely suitable
for paving purposes, if properly handled. This, of course, means

COMPARISON OF VARIOUS NATIVE ASPHALTS. 289

that the material shall be fluxed with an asphaltic oil, as a paraffine
flux is entirely incompatible with the material.

The preceding criticisms of the various asphalts which have been
used in the construction of asphalt pavements lead at once to the
conclusion that Trinidad lake asphalt or gilsonite is the best
for this purpose. In the author's mind there is no reasonable
doubt that this conclusion is correct. It is not intended, how-
ever, to assert that satisfactory pavements cannot be constructed
from the other asphalts, especially where the latter are not sub-
jected to trying environment or a heavy traffic and where con-
siderable skill is exercised in their use. It is asserted, however,
that with Trinidad asphalt there is greater probability that a
pavement constructed with it will be satisfactory, taking into
account the fact that a greater or less lack of care is inevitable
in preparing an asphalt surface mixture from any bitumen. Trini-
dad asphalt will stand more abuse than any other material with
which we are acquainted, and on this account is to be strongly
recommended, as well as because less skill is required in handling it.

SUMMARY.

In this Chapter there is outlined the characteristics which make
any solid native bitumen available and desirable for paving pur-
poses. These characteristics are as follows:

1. The quantity available.

2. Its uniformity in character.

3. Its stability in a melted condition at high temperatures.

4. Its stability in consistency at the extremes of temperature
which it meets in an asphalt pavement.

5. The proportion of malthenes to asphaltenes which it con-
tains.

6. The proportion of flux which is required to make an asphalt
cement.

7. Mineral matter present and its character.

It appears that Trinidad lake asphalt and gilsonite fulfil more
of the necessary requirements than any other commercial supplies
of native bitumen for the purpose of constructing asphalt pave-
ments.

290 THE MODERN ASPHALT PAVEMENT.

It also appears that properly constructed surface mixtures
made with Trinidad lake asphalt are no more acted upon by water
than those made with other asphalts, and that the charge that
they are acted upon to a greater extent is dependent purely upon
laboratory experiments without regard to making the conditions
under which they are carried on conform to tnose which are met
with on the street.

PART IV.
TECHNOLOGY OF THE PAVING INDUSTRY.

CHAPTER XV.
REFINING OF SOLID BITUMENS.

ASPHALTS which contain water as they occur in nature must
be freed from it before they are in a condition to be used in the
paving industry. The process of bringing this about is called
refining. It really is nothing more than some method of drying the
asphalt, in some cases removing the more volatile hydrocarbons,
the loss of which, at a later period, from the asphalt cement
would make the latter of unstable consistency, and skimming off
any vegetable matter which may rise to the surface of the
melted material. The process was originally called refining
because, before the value of the fine mineral matter was un-
derstood, much of this was separated out by subsidence
from the melted bitumen and the resulting asphalt was actually
refined, having been made purer or richer in bitumen. To-day
the refining goes no further than the removal of such organic
contamination as may rise to the surface of the melted asphalt
after the water has been evaporated, and the volatile hydrocar-
bons have gone off with the steam, and the thorough mixing of
the residual mass to a condition of uniformity in composition,
the mineral matter being maintained for this purpose in suspen-

291

292 THE MODERN ASPHALT PAVEMENT .

sion, in the meanwhile, by agitation of the melted material in
any convenient way.

The drying process is conducted in two different ways. The
material is filled into an iron tank or melting-kettle which is heated
by a free fire, the bottom of the kettle being protected by an arch
of brick; or a large rectangular tank is used, the interior of which
is filled with gangs of pipe through which steam is conducted at
such a pressure as to raise the asphalt to the same temperature
that is produced over the free flame but without any danger of
exceeding the highest temperature which the pressure of the
steam will yield. Fig. 18. 1 In either case, since convection in
such a viscous mass is very slow, agitation is carried on with either
a current of air or steam, in the latter case the current not being
admitted until the asphalt is melted and exceeds the boiling-point
of water, this being necessary to prevent condensation and sub-
sequent foaming. The temperature must, of course, be raised
slowly at first to avoid foaming when the bitumen melts easily
and the asphalt contains much water. When the temperature
has been raised to a point where the material is thoroughly melted
and steam is no longer given off the process is finished and the
refined asphalt is ready to be drawn off. The details of this process
are ones purely of economy, the object being to dry and get the
asphalt into packages suitable for handling. Where the material
is to be made into cement and used on the spot, the latter
is unnecessary. The packages are usually old hydraulic cement
barrels which before use are clayed on the inside by being re-
volved in a bath of clay and water. The claying is done to make
it possible to strip the staves from the asphalt more easily when
preparing it at its destination to be made into cement and to do
this with the loss of the least possible amount of asphalt adherent
to the staves.

In the fire refining method four or five days are required to
complete the operation. In refining solid bitumens in this way
danger is always incurred of overheating them, with the formation
of coke and the cracking of the hydrocarbons. There is a certain
formation of coke in all cases where a direct flame is in use and

1 Page 405.

REFINING OF SOLID BITUMENS. 293

that some asphalts are injured during the process is shown by the
fact that the resulting bitumen is not entirely soluble in cold car-
bon tetrachloride. To avoid such difficulties a very thorough
mechanical or other form of agitation is absolutely essential.

By the steam process the refining is completed in 24 hours
or less without any danger of injury to the bitumen from over-
heating. The agitation in this process is generally by means of
dry steam. The use of steam results in the volatilization of a
rather larger amount of lighter oils than occurs with air agitation
and it may be possible, for this reason, that it could be replaced by
air beneficially, although hot air has a decidedly strong effect upon
native bitumens as has been shown in connection with the con-
densed oils. 1

From ten to twenty-five or more tons are refined at once, the
larger amounts by the steam process, and the temperature reached
is about 325 F.

Almost all asphalts require refining but some other native
bitumens which can be, and are, used to a small extent in the
paving industry, are anhydrous and need no drying. Gilsonite and
grahamite need no refining, being practically dry and pure bitu-
mens.

The Preparation of the Asphalt Cement. Whatever solid
bitumen and flux are selected for the purpose, their careful com-
bination is necessary for the preparation of a satisfactory asphalt
cement. The carefully weighed asphalt is melted and raised to
a temperature of about 300 or 325 F., or if the material is taken
on the immediate completion of refining, as happens where a
refinery and paving plant are associated, it is carefully gauged.
The flux, having preferably been heated with steam coils in the
receptacle containing it to 150 to 200 F., is then slowly run into
the melting-tank holding the asphalt, agitation with air or steam
having been established, the air or steam being admitted through
lines of pipe, perforated with frequent openings and which lie along
the bottom of the tank. A satisfactory and sufficient agitation
is most essential and steam has been found more suitable than
air where its use is possible. It should, of course, be high pressure

1 See page 273.

294 THE MODERN ASPHALT PAVEMENT.

steam and it should not be admitted to the melted asphalt until
the latter is at such a temperature as to prevent condensation.
Every provision should also be made that the steam be quite dry
by blowing all condensed water out of the pipes carrying it. Neg-
lect to do this will, otherwise, cause dangerous foaming. A check-
valve should also be provided at a point above the surface of the
melted asphalt to provide for the admission of air when the steam
is shut off and prevent condensation and the production of a
vacuum which will draw the melted asphalt cement back into
the agitation pipes and clog them. Air agitation is simple and
fairly satisfactory but the effect of blowing hydrocarbon oils
with air results in hardening them and changing their consistency
in a marked degree and on that account is undesirable.

The agitation, of whatever kind, should be kept up until the
solid bitumen and liquid flux are thoroughly mixed and in homo-
geneous solution. The length of time required will depend on the
force of the current of steam or air and the character and tem-
perature of the melted materials. Under the most favorable
circumstances three hours are requisite and with inferior agita-
tion eight or more may be necessary.

To the eye of the experienced yard foreman the point at which
the combination is complete and the mixture homogeneous will
be evident, but the material can be tested by pouring some of it
into a pail of cold water and examining it on cooling Any oili-
ness is a sign that more agitation is necessary.

The asphalt cement having been found to be homogeneous
the next step is the determination of the fact that the consistency
is that which is desired. This can be arrived at in various ways
of greater or less refinement. The ordinary, and always the pre-
liminary test, is that of chewing a small piece of the cement
cooled by pouring it into cold water. On putting the cement
in the mouth and working it between the teeth it rapidly assumes
the temperature of the mouth which is a very uniform one, that
of the normal temperature of the body, 98.4 F. The amount
of work that is done by the jaws upon the cement will readily
show whether it is harder or softer than what experience has
taught to be a proper consistency and it is not difficult for one

REFINING OF SOLID BITUMENS. 295

who makes this test daily to decide whether the asphalt in ques-
tion is within four or five points of the consistency desired and
registered by the more accurate penetration machine. In experi-
enced hands it is questionable whether a more accurate test is
absolutely necessary, except as a matter of record.

A more refined test which is available for use by the yard fore-
man at the plant is that known as a flow test which permits, ac-
cording to a method described in Chapter XX VIII of comparing the
relative flow, at temperatures above the softening point of the
cement, of the material to be tested with that of a standard of
the desired consistency prepared in the laboratory.

Where a definite determination is required for purposes of
record one of the several penetration machines may be used, but
these require careful manipulation and their use sometimes neces-
sitates greater refinement than a yard foreman is capable of.

Under any circumstance it is absolutely necessary that the
consistency of the asphalt cement shall be so regulated that it
will be entirely uniform for any one piece of work. What this con-
sistency shall be will depend upon the character of the work which
is being done and upon its environment, both as to traffic and
climate. The variations in this respect will be discussed later.

If the cement is to be held in a melted condition for any length
of time agitation must be maintained, especially if it contains
mineral matter. The purer native bitumens and residues from
asphaltic petroleums require very little beyond- that necessary to
prevent any one portion remaining for any great length of time in
contact with the source of heat, whether the walls of a tank heated
by direct flame or the steam coils. All cements can be allowed to
become solid and cold if they are thoroughly agitated again on
remelting. On the other hand too powerful agitation is injurious
as it volatilizes the lighter portions of the cement and hardens it.
Continued agitation with air has a marked effect upon the charac-
ter of all oils by the extraction of hydrogen and condensation of
the hydrocarbons to a short rubbery solid such as the blown petro-
leum now to be found on the market as an article of commerce,
and which has been already described. The result of continued
air agitation, therefore is to harden an asphalt in two ways, by

296 THE MODERN ASPHALT PAVEMENT.

the volatilization of the lighter oils and also by increasing their
density by condensation of two molecules into one. Steam hardens
a cement only by the volatilization of the lighter constituents.
Steam is, therefore, probably preferable for the agitation of a fin-
ished asphalt cement, although air may be more desirable as a
means of agitation during refining.

The actual changes which take place with different fluxes and
different asphalts will be shown further on.

Character of Various Asphalt Cements. The character of an
asphalt cement depends upon that of the solid bitumen and of
the flux from which it is prepared.

Asphalt cements may be divided into several classes.

1. Those composed of the standard solid native bitumens, such
as Trinidad and Bermudez asphalts, and paraffine petroleum
residuum.

2. Those composed of the same asphalts and fluxes or residuums
from asphaltic petroleums.

3. Those made from the same asphalts and fluxes which are
mixtures of asphaltic and paraffine oils.

4. Those made from other solid native bitumens and asphaltic
fluxes.

5. Those composed of solid residual bitumens from asphaltic
petroleum brought to a proper consistency with residuum of the
same origin.

6. Any of the first four classes with the addition of small
amounts of the condensed or blown oil, or other forms of bitumen
not constituting one of the main constituents of the cement.

Asphalt Cements Composed of Trinidad or Bermudez and
Similar Asphalts and Paraffine Petroleum Residuum. Some years
ago there was a popular prejudice against the use of paraffine
petroleum residuum as a fluxing agent for asphalt. This was not
founded on the results of any careful investigation or tangible
evidence. It arose at first from a desire to find some excuse for
the poor wearing quality of some carelessly constructed asphalt
pavements and from the fact that the earlier surfaces were readily
attacked by water where subjected to its action for any length of
time. It was claimed:

That a part of the residuum of paraffine petroleum is soluble in

REFINING OF SOLID BITUMENS. 297

water and that by the continued action of the latter on the oil in
the asphalt cement, the cement is deteriorated.

That on standing in a melted condition the petroleum oil rises
to the top of the cement and can be "skimmed off like cream."

That the bitumen of Trinidad and other asphalts are not com-
pletely soluble in paraffine residuum but are only mechanically
mixed.

The fallacies in two of these claims are readily shown.

That the first is false is shown by the fact that distilled water
which has been allowed to stand in a glass-stoppered bottle in
contact with standard paraffine residuum for four years is as
bright and clean as when first put there and contains nothing in
solution. 1

The second is equally wrong since a tank of asphalt cement
maintained one week at a temperature of 300 F., without agita-
tion, on cooling, was not to the slightest degree oily or greasy on
the surface, which would be the case if any oil had separated like
cream.

The proposition that the bitumen of Trinidad and other asphalts
is not completely soluble in standard paraffine petroleum residuum
can be equally well disproved and it can be shown that asphaltic
bitumens are as soluble in paraffine residuum as in the asphaltic
oils of California. The results of some experiments in this direc-
tion by the author were presented in articles in "Municipal
Engineering " for June, July and August, 1897, and for June, 1899.
The experiments and conclusions arrived at and presented in the
latter article were, in the main, as follows:

"Three asphalt cements, prepared with great care and uni-
formity, as appears from the results of duplicate analyses of the
original material, were placed several inches deep in glass tubes
8 inches long and } of an inch in diameter and maintained at a
temperature of 325 F. for three days, being centrifugaled at that
temperature several times to assist any separation that might take

1 Messrs. Whipple and Jackson in a paper read before the Brooklyn En-
gineers Club and published in the Engineering News for March 22, 1900, have
shown that petroleum residuum is affected less by water than any bitumin-
ous substance that they examined.

298

THE MODERN ASPHALT PAVEMENT.

"RESULTS OF CENTRIFUGAL ACTION ON VARIOUS ASPHALT

CEMENTS.

"100 Ibs. Trinidad + 20 Ibs. of paraffine residuum.

Original
Cement.
Duplicates.

Top
45 Per Cent.
Duplicates.

Bottom
45 Per Cent.
Duplicates.

Sedi-
ment,
10 Per
Cent.

' ' Bitumen soluble in chloroform

27.6
26.8
22.3

83.2
16.8

65.5
5.4

Bitumen soluble in CS 2

63.4 63.5
48.6 48.7

76.6 76.6
23.4 23.4

30.3
6.3
55

69.9 70. 1
53.4 53.4

76.3 76.2
23.7 23.8

23.9
6.2
49

68 . 5 68 . 7

52.7 52.9

76.9 77.0
23.1 23.0

25.1
6.4
51

Bitumen soluble in 88 naphtha. .
Per cent of total bitumen thus sol-
uble

Per cent of total bitumen insoluble

"100 Ibs. Trinidad +27
"Bitumen soluble in chloroform. .

' Ibs. Califoi

*nia 'G' gra

de flux.

27.3
26.8
22.5

84.0
16.0

65.8
7.4

07. 5
66.8
53.9
80 7
19.3

15.0

8.2

Bitumen soluble in CS 2 . . .

64.7 64.8
51.1 51.3

79.0 79.2
21.0 20.8

28.9
6.4
46

-18 Ibs. of p

71.6 71.9
55.7 56.0

77.8 77.9
22.2 22.1

22.9
5.5
47

araffine resi<

70.2 70.4
55.1 55.4

78.5 78.7
21.5 21.3

23.0
6.8
50

iuum.

Bitumen soluble in 88 naphtha. . .
Per cent of total bitumen thus sol-
uble

Per cent of total bitumen insoluble
Mineral matter.

Penetration

"100 Ibs. Bermudez-f
"Bitumen soluble in chloroform

Bitumen soluble in CS 2
Bitumen soluble in 88 naphtha. .
Per cent of bitumen thus soluble . .
Per cent of total bitumen insoluble

Mineral matter.

92.6 92.8
73.1 73.2
78.9 78.9
21.1 21.1

3.2 3.3
4.2 3.9
60

96.2 96.4
74.5 74.7
77.4 77.5
22.6 22.5

1.9
1.9

95.2 95.4
73.6 73.8
77.3 77.4
22.7 22.6

2.2
2.6
55

Penetration at 78 F.

REFINING OF SOLID BITUMENS. 299

place. On cooling the asphalt was divided into three parts; an
upper, 45 per cent; a lower, 45 per cent; and the bottom, 10 per
cent, of sediment. The consistency and composition of these por-
tions were then determined by the most careful methods, extract-
ing with naphtha and carbon disulphide, filtering on a Gooch cru-
cible and burning the bitumen solution for any inorganic correction.
The results were obtained in duplicate, except in the case of the
sediment. Their agreement is confirmatory of their accuracy. See
results tabulated on page 298.

"These results show that there is no difference in the char-
acter of the bitumen in the cement made from Trinidad asphalt
and residuum in the two portions of cement above the residue,
after heating and subsidation, from that of the original material.
With California oil and Bermudez asphalt there is a slight loss of
oil in the upper portions and consequent small reduction in the
percentage of naphtha soluble bitumen. In the sediments the pro-
portion of naphtha soluble to total bitumen has increased in all
three cements. The fact that something in the cement more solu-
ble in naphtha and heavier than the ordinary constituents has been
thrown down, is peculiar."

What this is it is impossible to say at present, but it appears
from a paper by R. P. Van Calcar, which has recently appeared in
the Recueil des Travaux Chimiqices des Pays-Bas, 1 that where solu-
tions of various salts in water were subjected to a centrifugal force
400 times that of gravity they became more concentrated at the
periphery after a few hours, contrary to the preconceived ideas that
the molecules in a true solution were unaffected by gravity, and
hence were in a different state from those in colloidal solutions.
As a consequence of Van Calcar s conclusions the results obtained
with the asphalt cement are not unexpected.

"As a whole the results seem to the writer to refute the state-
ments which have been made in regard to residuum and to show
that 90 per cent of the asphalt cement made of Trinidad lake
asphalt and residuum was unchanged to any perceptible degree
after the severe treatment it had been subjected to by standing
and centrifugaling at a high temperature, while any changes that
1 Science, 1904, August 19, 20, 250.

300

THE MODERN ASPHALT PAVEMENT.

took place in the sediment were found as well in cements made
with the California residuum or so-called asphaltic flux."

As a matter of fact there is no evidence to show that there is
any essential difference between the California fluxes and paraffine
residuums in their power of dissolving the bitumen of Trinidad and
Bermudez asphalts, or that the latter is not a satisfactory flux,
on this account, for making asphalt cements with these asphalts.
The successful use of it in many pavements laid twenty years ago,
which are now in perfect repair, is the best evidence that it is
satisfactory.

Amount of Residuum Necessary in Making An Asphalt Ce-
ment. The amount of paraffine residuum oil which it is necessary
to use per 100 pounds of Trinidad or Bermudez asphalt to make a
cement of satisfactory consistency depends on the character of
this flux. It may be very variable, but with a standard material
should not vary more than 4 pounds per hundred of the asphalt.
With some less carefully prepared residuums the difference may be
6 pounds. For example the oil in use by one company in 1899
and by another in 1898 had the following characteristics: .

Residuum

Light

Heavy

Specific gravity, 78 F./78 F., orig. mat., dry
Beaum6

.9197

22.7

.9331
20.5

Flashes, F

330 F.

442 F.

Loss, 400 F., 7 hours

17.3

3.8%

Pounds per 100 of asphalt to make asphalt
cement of 60 penetration:
Trinidad lake

Bermudez 1899

Much less of the lighter oil would produce the same softening
effect as the larger quantity of the heavier residuum. It becomes
a question then to determine as far as possible which is the most
desirable cement and the only evidence that is available are the
results of an examination of the two cements as to the change in
their consistency at such extremes of temperature as are common
in pavements and as to their change in penetration on being main-

REFINING OF SOLID BITUMENS.

301

tained in a melted condition for some time. Experiments in these
directions furnish the following information:

COMPARISON OF CONSISTENCY OF ASPHALT CEMENTS AT DIF-
FERENT TEMPERATURES WHEN MADE WITH DIFFERENT
FLUXES.

Asphalt.

Residuum.

Pounds
per 100

Penetration al

of Asphalt.

45 F.

78 F.

100* F.

Bermudez

Light
Heavy

14
23

30
32

60
60

105
125

Trinidad.

Light

120

Heavy

115

PENETRATION AND LOSS AFTER HEATING TO 300 FAHR.

Penetration at 78 F.

Loss.

23:

After
Heating
4 Hours

After
Heating
6 Hours

1st
2 Hours

2d
2 Hours

2 Hours

TotaL

Bermudez. . .

Light

1.36%

.79%

.71%

2.86%

Heavy

.50

.40

.30

1.20

Trinidad. . . .

Light

1.56

.75

.58

2.89

Heavy

.98

.34

.33

1.65

At temperatures between 78 and 45 F. there is no great
difference in the penetration of cements made with heavy and
light residuum. At higher temperatures there is a considerable
but not constant difference. In the case of Bermudez cements,
that made with the heavy oil is softer at 100 because of the
greater softening effect of the larger amount of flux, 23 pounds,
as compared to 14 of the lighter oil, while with the harder Trinidad
the reverse is the case. As will be seen later, a flux which is so
dense that an excess of it is required to produce a cement of normal
consistency at ordinary temperatures may make a cement more
susceptible to high temperatures than a lighter or less dense one.

As to the permanency of the two classes of cement, however
the figures show that on maintaining it in a melted condition,

502

THE MODERN ASPHALT PAVEMENT.

and of course on mixing with hot sand, there is a much larger loss
of oil and a greater hardening of the cement fluxed with light than
with heavy residuum. For this reason alone cements made with
the heavier oil seem, up to a certain point, decidedly preferable to
those made with the lighter forms in use. Determinations of the
consistency of the bitumen in old surfaces laid with cements made
with light residuum as compared with others holding heavy oil
confirm this. Surfaces in Omaha were laid in 1890 in part with a
light, so-called summer oil, and in part with a heavy one. The
consistency of the bitumen in these different surfaces when laid
and again on extraction was as follows:

Flux in Cement.

Original Pen.

Pen. 1899.

Loss.

Light
Heavy

67
50

35
30

32
20

It seems that the cement made with the very light oil has
hardened, either in the mixer or by age, to a much greater extent
than the other.

Many good pavements have been made with the lighter fluxes,
however, and it would be unfair to condemn them entirely, or to
say that they are necessarily the cause of defects in asphalt sur-
faces, but it seems plain that the heavier oil is in general the more
satisfactory although more of it must be used.

In the light of the previous results no valid objection can be
raised and maintained against the use of a suitable paraffine petro-
leum residue as a flux for Trinided lake and Bermudez bitumens
in the preparation of an asphalt cement, and this is not surprising
when it is considered that many million yards of satisfactory
pavement have been laid with such a cement.

Paraffine residuums are to be found on the market, and this
was the case very frequently in the early days of the industry, which
are, owing to the manner in which they have been prepared, quite
unsuitable for use, but this has no bearing on the question of the
availability of standard material.

For fluxing Trinidad land asphalt and others of a hard nature
paraffine residuum is not suitable because of the deficiency of

KEFINING OF SOLID BITUMENS. 303

lighter malthenes in these asphalts, the lack of which is not made
up by the hydrocarbons of such a residuum.

Asphalt Cements Composed of Trinidad or Bermudez and Simi-
lar Asphalts and Flux or Residuum from Asphaltic Petroleums.
Trinidad, Bermudez, and other similar asphalts can be satisfactorily
fluxed with the asphaltic residuums which are prepared in the
East from Texas oil and are now on the market. The char-
acter of this residuum has already been described. It is a
most desirable material and can be used in about the same pro-
portions and in exactly the same way as the paraffine petroleum
residuum. It should not, however, be so dense as to necessitate
the use of excessive amounts of it, since under these conditions,
as was shown to be the case with paraffine residuum, the resulting
asphalt cement will be too susceptible to high temperatures. The
density should be such that not more than 22 pounds of oil per
100 of refined Trinidad or of Bermudez asphalt shall be required
to produce a cement of 65 penetration on the Bo wen machine.
Such an oil will have a density of .95. The heavier residuum of
a density of .97, also found on the market, is not satisfactory,
although this flux has its use with certain other native bitumens.

Comparing the general characteristics and stability of the
two forms of residuum it has been found and confirmed by prac-
tical experience that there is probably a slight preference in favor
of a not too dense asphaltic flux, but this difference is not sufficiently
great to make it obligatory to use the latter except in work of the
very highest character on streets which carry very heavy traffic
and where it is certain that the character of the asphaltic will
be as uniform and satisfactory as that of the paraffine flux, and
unfortunately this is not always the case. The greatest care is
necessary in its preparation, as any overheating or cracking in the
latter will result in the presence of light oils which volatilize readily
and cause a rapid change in the consistency of the cement, while
maintaining it in a melted condition or during the time that the
cement is being tossed about in the mixer in contact with hot
sand during the preparation of surface mixture. It is possible,
therefore, that in comparison with an asphaltic flux of inferior
grade a paraffine residuum may be preferable.

304 THE MODERN ASPHALT PAVEMENT.

Combinations of Trinidad, Bermudez, and similar asphalts with
the heavy California flux known as No. 2 or G grade are not satis-
factory, since the proportion of such a flux to the asphalt in order
to produce a cement of proper consistency is so large, being in the
neighborhood of 60 pounds of flux to 100 of Trinidad asphalt, that
the resulting material is excessively susceptible to high tempera-
tures. Such combinations are, therefore, rarely used. Where
the solid asphalt is one that has been much hardened by age or
exposure, as in the case of that from La Patera in California, a
supply of which is no longer on the market, the mine being ex-
hausted, the use of a heavy California flux or a very dense Texas
residuum is imperative, at least as a preliminary fluxing material,
to supply the lack of denser malthenes in the asphalt. If a certain
amount of this flux is used, however, the remainder can be of a
lighter form, and probably preferably so. Asphalts of this descrip-
tion are not at present of commercial interest, with the exception;
perhaps, of that obtained in Cuba from the Bejucal mine.

Asphalt cements have sometimes been made from the solid
native bitumens, including the asphalts, and the natural malthas.
Pavements constructed with asphalt cements made in this way
have proved, however, to be unsatisfactory. Some experiments
were conducted some years ago by the writer to determine why
such asphalt cements were not satisfactory.

In the laboratory it was found that the solubility of the bitu-
mens of Trinidad and Bermudez asphalts was as great in the
ordinary malthas as in the residuums from paraffine and asphaltic
petroleums. There was no preference in this respect. When,
however, the permanence of consistency of malthas when exposed
to heat was compared with that of residuums, there was found
originally to be a great deficiency in that of the malthas.

Residuum such as is at present in use has already been shown
to volatilize but a small amount when heated in an open dish
in a bath kept at 400 F. for 7 hours, and to remain of the same,
or very nearly the same, consistency after as it was before heating.
The desirable features of a carefully prepared residuum as a soften-
ing agent are not lost on continued heating, nor is there sufficient

REFINING OF SOLID BITUMENS. 305

oil volatilized at the high temperatures at which asphalt cement
is maintained in a melted condition, with agitation for consider-
able periods of time in large masses, to change the consistency
to any marked degree. As an example, a Trinidad lake asphalt
cement made on February 29, 1896, by mixing 100,000 pounds
of refined asphalt with 20,000 pounds of residuum had a pene-
tration of 55. It was held over a very low fire in a melted con-
dition for 48 hours and then had changed in consistency so little
as to penetrate 49. After 73 hours melting the penetration was
46. This is an extremely small change for such a considerable
length of time.

Asphalt cements made with the native malthas behave quite
differently. On heating for any considerable tune they are con-
verted into hard and glassy pitches by volatilization of oil, and,
perhaps, by- condensation of its hydrocarbon constituents. Such
material is unstable and cannot form a cement which can be
maintained at a uniform consistency.

On Saturday, February 29, 1896, in the early morning, 500
pounds of Bakersfield maltha was added to 2000 pounds of refined
Trinidad asphalt, or at the rate of 25 pounds to the 100. After
agitation the resulting asphalt cement penetrated 55. It was
allowed to stand with a low fire until the following Monday morn-
ing, March 2. The penetration had then fallen to 25. On stand-
ing another 24 hours the penetration was found to be 22. 270
pounds of additional maltha was then added, corresponding to
13.5 pounds per 100, whi^h, after agitation, raised the penetration
to 54. After 4 hours of heating and agitation a sample was
taken and found to penetrate but 35.

It appears from this experiment that the light native Cali-
fornia malthas are not suitable for the preparation of an unchange-
able cement. Why this is so can be seen in the results of an exam-
ination of the maltha in the laboratory. While it is thick enough
to require 5 pounds more of it to every 100 pounds of Trinidad
refined asphalt to make a cement of the same consistency that
is obtained with residuum, it loses on heating for 7 hours to 400 F.
20.3 per cent. This light oil is, of course, volatilized, in the same

306 THE MODERN ASPHALT PAVEMENT.

way, though more slowly, from the asphalt cement made with
the maltha, and the loss causes the rapid fall in penetration and
hardening of the cement.

An experiment with one of the asphaltic oils extracted from the
abundant supply of asphaltic sandstone rock in Texas resulted
similarly. The asphaltic oil, or maltha, as received was heated for
some time at a low temperature to drive off any water and very
volatile oil. A cement was then made of Trinidad asphalt and the
maltha in the proportion of 100 to 80, which had a penetration of
65. This cement was then maintained for 9 hours at a tempera-
ture of 325 F., when it was found to have hardened so much as to
penetrate but 24.

Of course satisfactory surface mixtures for paving cannot be
made with such changeable material, and this is the reason that
much of the earlier work done with California asphalt was a fail-
ure.

In fact, slight reflection shows that for a fluxing agent for
softening hard asphalt a substance is needed which does not
change its consistency after prolonged heating, and not
another, though perhaps softer asphalt, which gradually
becomes converted into a hard asphalt under the influence
of heat.

Asphalt Cements Composed of other Solid Native Bitumens
than Ordinary Asphalts and Asphaltic Fluxes. Entirely satis-
factory cements for use in the paving industry have been made
from some of the other native bitumens, especially Gilsonite,
by fluxing them with an asphaltic oil. For this purpose about
equal parts of the Gilsonite and flux are necessary. The resulting
cement contains bitumen in the form of the classes known as
malthenes and asphaltenes in normal proportions, about the
same as in Trinidad and Bermudez asphalt cements, as will appear
from data on page 308. On this account, such cements may be
regarded as entirely normal in character, and the fact that they
contain a large proportion of flux is not open to comment, espe-
cially as the hard bitumen and the oils combine homogeneously
and no free flux is found in the cement.

REFINING OF SOLID BITUMENS. 307

Asphalt Cements Composed of Solid Residual Bitumens from
Asphaltic Petroleum Brought to a Proper Consistency with
Residuum of the Same Origin. From the asphaltic petroleums,
such as those found in California and Texas, residual pitches or
solid bitumens are prepared by distillation, and the charac-
teristics of these solid bitumens have been already described.
Asphalt cements can be prepared for paving purposes from these
residual pitches by bringing them to a proper consistency with
a residuum or flux made from the same petroleum. These cements
have been used with some success and also with many resulting
failures. They are very susceptible to temperature changes,
which necessitates the use of a very carefully graded mineral
matter with plenty of filler and the greatest skill hi handling
them in order that they may not harden while being mixed with
hot sand or reach the street at such a degree of softness that they
mark up very rapidly under hot summer suns.

Cements of this description have been used to a very consider-
able extent on the Pacific Coast, and they are quite suitable for
the climate of Southern California. In Washington and Oregon
some difficulties have been met with where they have been employed.

Asphalt Cements of Any of the Previous Classes to which*
Amendments of Residual Pitches or Blown Oils Have Been Added.
Excellent asphalt cements have been prepared for paving purposes
to which additions, not exceeding 10 per cent in amount, of con-
densed or blown oils, such as Pittsburg flux, Ventura flux, or blown
Beaumont oil have been made. While tiie materials constituting
these additions are in themselves unsuitable for paving purposes,
they seem in some instances to modify the properties of native
bitumen in such a way as to improve them, although in a manner
that cannot be described. Such cements are not to be objected
to, since they have been shown by experience to give satisfactory
results.

Characteristics of Asphalt Cement. It is of interest to note
the characteristics of asphalt cement prepared from various bitu-
mens with various proportions of fluxes, especially as to the rela-
tive proportions of the malthenes and asphaltenes which they

308

THE MODERN ASPHALT PAVEMENT.

contain. Some data in regard to this are given in the following
table:

Asphalt.

Flux.

Proportions.

1
1

Bitumen by CSj.

Mineral Matter.

Difference.

Per Cent Total.

Fixed Carbon.

Trin. Lake .

Texas

100-19.0

64.2

31.0

4.8

46.7

72.7

8.3

asphal-

100-25.0

67.6

28.3

4.1

50.2

74.3

8.2

tic

100-33.3

70.0

25.9

4.1

52.2

74.6

7.6

Trin. Lake .

Paraf-

100-19.5

64.7

30.5

4.8

46.0

71.1

8.9

fine

100-23.5

66.1

29.2

4.7

47.2

72.6

8.8

100-26.6

67.5

27.7

4.8

48.6

72.0

8.1

Trin. Land

Paraf-

100-17.6

63.1

30.9

6.0

44.3

72.0

9.2

fine

100-25.0

64.3

29.6

6.1

47.3

73.5

9.8

100-28.2

65.5

28.8

5.7

49.0

74.8

9.0

Berm. Lake.

Texas

100- 5.3

94.8

2.4

2.8

69.0

72.8

13.0

asphal-

100-14.9

95.2

2.4

71.6

75.1

12.1

tic

100-22.0

95.0

2.1

2.9

72.2

76.0

11.6

Berm. Lake.

Paraf-

100- 5.3

94.2

2.2

3.6

68.0

72.2

12.7

fine

100-14.9

95.5

2.0

2.5

71.0

74.3

12.4

100-22.0

95.4

2.2

2.4

71.3

74.7

12.2

Gilsonite . . .

Texas

100-85.2

99.7

76.3

77.1

8.1

asphal-

tic

100-96.1

99.8

76.8

77.6

7.7

D grade, Cal.

G grade

100-12.4

99.7

70.8

71.5

16.0

Texas resid-

Texas

100-20.4

98.6

1.2

70.5

71.5

19.1

ual pitch

asphal-

100-25.0

98.9

71.9

72.7

18.3

tic

100-33.3

98.8

1.1

73.8

74.7

17.2

Cuban Beju-

Texas

100-49.0

68.3

26.1

5.6

48.8

71.4

11.9

cal

asphal-

100-50.0

70.0

23.5

6.5

50.7

72.4

11.1

tic

100-55.0

76.0

18.1

5.9

56.0

73.7

11.0

Grahamite. .

Texas

100-150.0

99.5

63.3

63.6

9.8

asphal-

100-203.0

99.5

70.4

70.7

8.4

tic

100-233.3

99.5

71.3

71.6

8.1

Physical Properties of Asphalt Cement. The character of
a sheet asphalt surface of ordinary type will depend very largely
on the properties of the cementing material which binds the mineral
aggregate together, even if the latter is of the most approved

REFINING OF SOLID BITUMENS. 309

grading and consequent stability. When the latter is not careiolly
arranged the physical properties of the asphalt cement will have
an even greater influence on the behavior of the asphalt surface,
more particularly under great extremes of temperature. It is
important, therefore, to examine the properties of asphalt cements
prepared from different solid native bitumens and softened with
various fluxes. The properties which are of the greatest impor-
tance have been generally accepted to be their greater or smaller
susceptibility to changes in consistency at the extreme temperatures
which they meet under different climatic conditions and to their
variable ductility.

The fact that asphalt cements vary in consistency, with change
of temperature, means that at certain temperatures they are very
viscous liquids and at low temperatures slightly viscous solids, the
transition from one state to another being very gradual, although
under modern theories of physical chemistry substances which are
not crystalline can hardly be regarded as being solids. The slow
flow of crude Trinidad asphalt, where large heaps of it are stored,
is a well-known occurrence arfd corresponds very closely to that
of the glacial flow of ice. The flow of an asphalt cement con-
taining a very considerable proportion of flux is, of course, much
more rapid. Mr. A. W. Dow 1 has shown that when cubes of asphalt
cement are placed over a hole in a board at temperatures of 26 F.,
75 F. and 140 F., the movement of the material into the hole
was visible in 1 hour at 140 F., in a longer time at 75 F., and in
I week at the lowest temperature. The fact that an asphalt cement
will flow at this low temperature is of great importance in connec-
tion with the behavior of asphalt surface on the street in the winter
months. Unless there is some ductility to allow for the contraction
in the mass of the mineral aggregate all asphalt surfaces would
crack at such a season. That they do not do so in all cases is to
be attributed to this and to the fact that a suitable asphalt cement
possesses such a consistency and lack of susceptibility to change
*n this respect between the lowest and the highest temperature
to which it is exposed as to prevent it. Cracking frequently does

1 Municipal Engineering, 1898, 15, 364.

310 THE MODERN ASPHALT PAVEMENT.

take place in asphalt pavements from the lack of such qualities
in the asphalt cement of which they are composed or from the
absence of a sufficient amount of it as will appear in the discussion
in later pages on the defects in asphalt pavements, 1 but is oftener
due to the hardness of a cement rather than to its lack of ductility
as experiments have shown that some asphalt cements, if suffi-
ciently soft, although very short, may be sufficiently ductile to
meet the demands made upon them at low temperature.

Experiments have shown that the ductility of an asphalt
cement is proportionate to the amount of flux which it contains
rather than to the character of the same and as the result of a
very extended investigation, the results of which are too lengthy
and numerous to introduce here, it has been made evident that
too much dependence cannot be placed upon this characteristic
in forming an opinion as to the availability of an asphalt for paving
purposes.

The susceptibility of asphalt cements to changes in consistency
with change in the temperature of its environment can be shown
in several ways, most conveniently by determining the consistency
at different temperatures with one of the various penetration
machines in use for this purpose, by the relative elongation of
cylinders of different cements under tension at different tempera-
tures or, in the case of high temperatures, by the length of flow
of small cylinders of cement on a corrugated brass plate in the
manner described in Chapter XXVIII. If asphalt cements are
prepared from different asphalts and fluxes of such a consistency
that they all have the same penetration at the normal tempera-
ture, 78 F., and are then again penetrated at extremely low
and high temperatures the relative changes in the consistency
can then be determined, as has been already shown. 2

In the following table are presented the results of the deter-
mination of the consistency of the asphalt cement made from
various asphalts with fluxes of different character at 41 F. and
100 F. all the cements having the same consistency at 78 F.

Although the results speak for themselves it may be well to

1 See page 480-2. 2 See page 301.

REFINING OF SOLID BITUMENS.
THE MODERN ASPHALT PAVEMENT.

311

Asphalt.

Flux.

Parts
Flux to
100 of

Asphalt.

Pen.
Bowen,

78 F.

Penetrometer.

41 F.i

78 F.i

100 F.i

Trinidad Lake

1 1 1 1

it ft
Bermudez

Heavy

25
34
22

17.5
24
15

65
65
65

65
65

5
3
3

5
5

6
5

41
42
32

44
40
39

34
37
30

28
22

97
113
91

100
109
108

82
48

35
37

Extra Heavy . . .
Paraffine

Heavy

1 1

Extra Heavy . . .
Paraffine

None

Gilsonite . . .

He aw

127
170
122

270

Extra Heavy . . .
Paraffine

Grahamite .

Extra Heavy . . .
Heavy . .

Cuban Bejucal

1 41 F. = 200 grams, 5 seconds; 78 F. = 100 grams, 5 seconds; 100 F.=
50 grams, 5 seconds.

call attention to some of the facts that are brought out by them.
Among the Trinidad cements that made with extra heavy oil y
and consequently requiring the largest proportion of flux, is
much more susceptible to high temperatures, having a penetra-
tion at 100 F. of 113, while that made with the paraffine oil,
of which only 22 pounds per 100 was employed, has a penetra-
tion of only 91, while all the Trinidad cements had practically
the same penetration at 41 F.

Asphalt cements made with gilsonite and grahamite are much
less susceptible to changes in consistency at extreme tempera-
tures. At 41 F. cements made from these materials, although
having the same penetration as the Trinidad, Bermudez, and
California cements at 78 F., are much less hard and in the same
way are softer to a less degree at higher temperatures. This
would show that such cements would be more satisfactory for use
in the paving industry where extremes of temperature are to be
met than those composed of true asphalts.

312 THE MODERN ASPHALT PAVEMENT.

SUMMARY.

In the preceding chapter the technology of the paving indus-
try has been discussed in detail from the refining of the native
bitumen to the preparation of the asphalt cement, together with
a study of the character of the various asphalt cements made
from different solid bitumens and with different fluxes. This
chapter in its detail will interest principally the asphalt expert,
the engineer, and the specialist.

CHAPTER XVI.
SURFACE MIXTURES.

THE surface mixtures of the early days of the asphalt paving
industry consisted, as they do to-day, of asphalt cement, ground
limestone, and sand; but even in 1893 very little attempt was made
to specify fhe character of these consituents or to determine what
rdles they play in the finished pavement. The asphalt cement was
at one time soft, at another hard, at one time too small in amount
and again too large, but oftener too small; the ground limestone
was expected in 1884 to be only so fine that 16 per cent should be
an impalpable fine powder and all should pass a No. 26 mesh,
hardly what would be considered a dust to-day, and at times it
was held to be doubtful if there were any necessity for the use
of dust at all. The sand was sometimes coarse and sometimes
fine, depending on the most available local supplies, and only in
the later years were two kinds mixed and that without much
reasoning.

In 1884 it was specified that the sand in use in Washington should
all pass a 20-mesh sieve and none of it an 80. 1

Surfaces with coarse sand and much cement marked or pushed
and then the bitumen was reduced; with fine sand and low bitumen
they cracked and the other extreme was again sought. Every-
thing was done by rule of thumb and without reason. To this
state of affairs much, but not all, of the cracking, displacement,
and defects in pavements laid in the early nineties was due. The
average consistency of the cement was the same for years and was

1 Annual Report of the Operations of the Engineer Dept., District of
Columbia, for the year ending June 30, 1884, 101.

313

314

THE MODERN ASPHALT PAVEMENT.

too hard in most cases because the defective mineral aggregate
would not permit the use of a softer one. The limestone dust was
most of it sand and in consequence the amount of filler in the mix-
ture was very deficient. But, worst of all, the sand grading was
arranged by chance. Specimens of old surfaces were collected
in 1894 and studied by the author at the request of the President
of the Barber Asphalt Paving Company as being representative
of the best work of the company up to that time, although they,
on this account, hardly illustrate the average pavement of that day.
They were analyzed in Washington and showed the following
variations in their mineral aggregate, filler, and bitumen. Among
these variations those in the sand grading are most striking.

AVERAGE COMPOSITION OF SURFACES FROM VARIOUS CITIES,
LAID BEFORE 1894, ARRANGED ACCORDING TO THE PER-
CENTAGES OF 100- AND 80-MESH SAND THEY CONTAIN.

aty.

Bitumen.

Mineral Aggregate Passing Mesh

Total.

200.

100 and 80

50 and 40.

30,20, and
10.

Washington

10.29
8.91
8.81
9.61
9.06
9.87
10.97
10.64
11.75
9.85
10.32
9.65
9.24
9.44

9.89
10.5

9.72
14.50
8.38
10.87
10.93
11.27
12.13
12.15
14.46
13.10
11.91
11.32
9.33
12.80

11.63
13.0

6.45
9.06
9.48
12.12
12.77
15.34
16.39
22.01
35.32
25.92
29.19
30.53
35.95
41.98

21.61
26.0

42.06
38.30
41.26
60.10
49.06
52.59
34.14
37.58
26.19
31.31
39.38
44.68
38.83
24.93

40.03
34.5

31.48
29.23
32.07
7.30
18.18
10.93
26.37
17.62
12.28
19.82
9.20
3.82
6.65
10.85

16.84
16.0

= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%
= 100%

= 100%
= 100%

Louisville

Newark

St Louis

Youngstown
New Orleans
New York
Scranton

Boston

Kansas City

Schenectady

Buffalo

Chicago

Omaha

Average

FOR COMPARISON.

Standard mixture. . .

These surfaces present every variety of grading in their com-
position and it is apparent that they could not all be satisfactory.
Evidently no system was carried out in their production. The
percentage of bitumen varies from 8.8 to 11.7, probably not from

SURFACE MIXTURES. 315

carelessness entirely but because the mineral aggregate would
permit of the use of a large amount in certain cases and less in others.
The amount of filler or 200-mesh dust is deficient in many cases,
although in some it seems high enough, owing to the presence of
sand of 200-mesh size which, as will appear, does not act as filler.

As will be shown later, the amount of 100- and 80-mesh material
is deficient in the first seven cities and unbalanced in all. The
grains of 10-, 20-, and 30-mesh sizes are present in far too large a
degree in Washington, Louisvile, Newark, and New York, and are
not well regulated in most of the cities. In fact the mineral aggre-
gate in none of these towns was, at that time, well graded.

If the records of the Barber Asphalt Paving Company are
studied for the 10 years before 1899, as summarized in the following
tables, the reason for the varied composition of the preceding
mixtures is explained. All sorts of sands were used, and probably,
although there are no records, all kinds of filler. See results tabu-
lated on pages 316, 317, 318, and 319.

If the average grading of the sand and of the mineral aggregates
of these same cities be examined as far as the incomplete data will
admit, the peculiarities of this part of the surface mixture are
apparent. As fine sieves were not in use in the early days of the
industry our knowledge of the grading of the finer part of the
aggregate is limited, but in the sands of 1889 it is readily seen that
in that used in Chicago there were not enough coarse particles,
while the Newark and New York sands were only fit for concrete.
Buffalo was deficient in 80- and 100-mesh particles, as were Kansas
City, Louisville, and Washington. Other defects were apparent
in this and the following years, to which it is unnecessary to call
attention here, as the data are open to examination in the tables
and the most striking points have been marked with asterisks.
In 1893, for instance, all the sands were much too coarse except
in Buffalo and Chicago. In 1894 they were much better. It is
sufficient to say that up to 1894 no effort based on any well-defined
reasons had been made to regulate the grading of the sand in
surface mixtures or to accommodate the dust and asphalt cement
to the demands of the latter, although it had been determined in
1892 that in the better class of pavements the sand was fine.

316

THE MODERN ASPHAL1 PAVEMENT .

AVERAGE SANDS.

1889.

City.

Passing

Ret. 10

Boston

8*
29
20*
12*
6*

10
41

28
9*

50
49
22
43
10

15
33

31
10*

26
16
18
22
11

15
10

17
24

11
5
20
12
17

sot

3*
1*

tit

5
23

18
3

7
19f

1*
0*
7
3
27

18
2

6
6

1
1
6

4
1

1
3

Buffalo 1-B

Chicago

Kansas City

Louisville

Newark

New Orleans

New York

Omaha

St Louis

Scranton

Washington

1891.

Passing

200

Boston

Buffalo 1-B

Chicago

Kansas City
Louisville

14f

12f

Newark

23t

New Orleans
New York

Omaha
St Louis

16*

Scranton

16t

lit

Washington

13*

29t

16t

* Too low.

t Too high.

SURFACE MIXTURES.

317

MINERAL AGGREGATE.
1892

rs*

Passing

Ret 10

City.

200

Boston

16f

Buffalo 1-B

Chicago

Kansas City
Louisville
Newark
New Orleans
New York

7
12
7
6
12

6*
21
11*
0*
21

7
20
17
11
13

13
25
22
39
14

20
13
20
32
15

28f
6
18
11
15

18t
3
6
2
10

1
1
3

Omaha

21 1

17f

St. Louis
Scranton

13*

29f

12f

Washington

1893.

Boston

22f

20f

Buffalo

51 f

Chicago

Kansas City
Louisville
Newark

41
16*

16
12

13
17

10
23

lot

New Orleans
New York

19t

14f

Omaha

14f

lit

St Louis

17f

Scranton

Washington

20*

13f

1894.

Pitv

Passing

Ret 10

City.

100

Boston

Buffalo

Chicago
Kansas City
Louisville

15
29

19
21

33
24

14
9

7
7

5
5

Newark
New Orleans
New York

14t

Omaha

St Louis

Scranton

Washington

31t

17t

* Too low.

t Too high.

318

THE MODERN ASPHALT PAVEMENT.

MINERAL AGGREGATE.

1895.

Passing

City.

100

Ret. 10

Boston

Buffalo

Chicago
Kansas City

22
27

19
19

9
5

Louisville

Newark

12*

14f

14t

New Orleans
New York

18
21

14*
15*

25
15

19
12

13
13

Omaha

St Louis

Scranton

12*

Washington

1896.

City.

Passing

200

100

Boston

12.1
7.4

9.9
13.7
12.2
11.2
11.6
14.4
10.6
15.9
11.0
8.7

14.4
6.2
13.7
8.7
5.4*
8.9*
9.4*
11.6
11.9
10.4
9.5*
6.7*

14.3
20.1
18.8
17.5
12.6
10.2
11.3
11.9
20.3
14.6
12.3
10.4

27.0
49.3
31.1
40.6
53.8
25.0
31.6
25.5
33.8
36.6
29.1
31.2

11.0
8.1
9.2
6.7
8.1
14.8
14.7
13.3
9.2
11.4
15.0
20.1

9.2
3.8
6.5
5.3
3.7
12.9
12.5
10.6
6.5
6.8
12.3
14.7

6.0
2.1
5.2
3.6
2.0
8.4
6.0
6.5
4.1
4.9
6.6
5.4

6.0
3.0
5.6
3.9
2.1
8.6
3.0
6.2
3.5
1.2
4.2
2.8

Buffalo 1-B. .....
Chicago Pit. 1. ...
Kansas City
Louisville

Newark

New Orleans
New York

Omaha

St. Louis
Scranton

Washington

1897.

Boston

16.5

14.9

12.5

26.6

10.1

8.5

5 2

5 7

Buffalo 1-B

15 3

12 6

16 3

46 3

5 7

2 5*

Chicago

15 7

16 4

19 5

36 3

2 7*

2*7

1 7

Kansas City

21 1

14.6

15 3

34.5

5.1

3 6

2 9

2 7

Louisville
Newark

14.9
17.3

6.4*
12.2

10.5
9.8*

54.4
27.3

7.5
10.5

3.2
10.9

1.8
7 2

1.2

4 8

New Orleans
New York

14.1
18.4

11.1
14.5

9.9*
13.9

28.9
27.7

13.3
9.4

11.6
8.0

6.6
4.6

4.5
3 5

Omaha

14 6

17 5

35 5

7 4

5 6

2 9

2 5

St Louis . .

23 2

12 7

18 2

32.1

6.7

3 8

1 2

Scranton

13.7

9.0

11.5

31.1

12.3

11.8

5 9

4 6

Washington

11.4

8.4

7.5

27.1

17.2

13.4

9.5f

5.4

* Too low.

t Too high.

SURFACE MIXTURES.

319

MINERAL AGGREGATE.
1898.

City.

Passing

200

100

16.4
15.8
14.6
16.2
16.7
12.9
12.3
15.3
13.0
14.9
13.8
13.8

15.2
14.3
17.3
13.6
11.8
12.5
8.7*
14.6
15.6
10.9
12.0
9.3

14.3
19.2
14.8
15.6
11.3
8.4*
11.1
15.0
19.2
16.6
11.7
8.6

28.3
40.1
37.6
31.9
34.3
20.6
34.6
25.3
32.9
41.9
29.7
29.1

11.0
6.1
9.4
8.3
10.9
14.4
15.8
11.2
8.0
6.4
13.2
20.0

7.5
2.8*
3.4*
6.5
7.3
12.0
11.1
8.5
5.3
4.3
9.9
10.7

4.4
1.6
2.1
4.5
3.8
11.6
4.1
6.3
3.6
3.0
5.6
5.2

2.9
0.1
0.7
3.1
4.0
7.6
2.3
3.9
2.4
1.9
3.9
3.2

Buffalo 1-B

Chicago

Kansas City

Louisville

Newark

New Orleans

New York

Omaha ....

St Louis

Scranton

Washington

1899.

Boston

16.2

13.6

10.4

24.9

15.7

8.4

6.4

4.4

Buffalo 1-B
Chicago
Kansas City
Louisville

14.7
12.9
13 5
17.4

11.8
16.0
10.3
8.0*

18.7
17.3
15.5
5.1*

41.6
38.5
43.9
41.3

7.0
8.5
7.9
21.6

2.8*
3.1*
4.0*
3.4*

1.9
2.0
3.0
2.0

1.5
1.6
1.9
1.2

Newark
New Orleans

16.2
14

15.2
12 4

10.7
10 4

12.9
28 4

12.5
19

10.8
7 7

11.7
5 7

9.9
2 4

New York

14 5

14 2

14 1

26.7

13 5

7 4

5 9

3 7

Omaha

14 8

14.7

13.9

28.6

12.9

6.3

5.2

3.6

St. Louis

13.8

13.7

12.3

33.2

12.7

6.0

4.8

3.5

Scranton

14.5

12.9

12.2

25.6

16.7

8.6

5.8

3.7

Washington

14.2

8.5*

5.8*

19.5

22.4

11.2

9.8

8.5f

'Too low.

t Too high.

Bitumen in the Surface Mixtures of the Earlier Days of the
Industry. Up to 1896 the bitumen in surface mixtures was very
variable in amount, and, as a rule, too low, owing probably to
the necessity of keeping it at such a point because of the poor
sand grading and the absence of binder, to avoid displacement
of the street surfaces. The following tables show the average
and extreme per cents of bitumen in the surfaces laid by the Bar-
ber Asphalt Paving Company in a number of cities during the
earlier years of which records are available.

320

THE MODERN ASPHALT PAVEMENT.

AVERAGE AND EXTREME PERCENTAGES OF BITUMEN IN
SURFACES OF THE BARBER ASPHALT PAVING COMPANY
FOR ELEVEN YEARS, 1889-1899.

City.

1889

1890

1891

1892

1893

Boston

9.3-10.9

8.9-14.5

8.6-12.0

9 3-10 3

Buffalo

10.0
8.4-12.2

7.6^i2.2

11.1
8.1-12.5

10.2
7.2-15.9

9.8
8.1-13.7

Chicago

9.8
8 6-10 5

10.2
9 0-11 5

10.0
7 9-11 7

10.1
8 8-11.5

10.3
9 2-11.7

Kansas City

9.8
8.0-12.3

10.1
8.0-11.1

9.5

7.7-11.7

10.1
8.9-12.3

10.3
9.0-11.3

Louisville

9.9
10.4-11.5

9.3
7.8-10.7

10.2
10.1-12.1

10.3
8.9-10.8

10.2

Newark

10.7
6.7-10.7

9.3

10.9
6.9-13.0

9.6
7.8-10.9

9.2-10.2

New Orleans.

9.0
8.1-10.6

9.6

9.8
8.8-10.4

9.7

9.2

9.5

New York

8.4-12.9

8 6-12.7

7 9-13.1

8.3-11.7

8.9-13.4

10.8
8.5-11.0

10.8
8.6-12.4

10.6
8.0-11.6

10.1
7.7-11.5

10.5

8.8-9.8

St Louis

9.8
9.6-11.2

10.2

9.8

9.4
9 6-12.8

9.4
7 3-11.7

10.1

9.7

8.5-11.2

8.8-11.6

8.2-14.2

9.2-12.2

Washington

10.1

8.8-15.5

10.6
9.1-11.4

10.5
8.7-11.5

10.6

8.8-12.8

9.6-10.9

Averaee. .

9.8
9.9

10.2
10.1

10.3
10.2

10.7
10.0

10.2
10.0

City.

1894

1895

1896

1897

1898

1899

Boston

8.4-10.0
9.4
6.9-11.9
10.0
8.2-11.7
9.8
8.0-12.1
9.9

7.5-11.1
9.9
8.3-10.6
9.4
8.4-11.0
9.9
8.3-10.7
9.3
8.0-11.6
10.0
7.8-10.6
9.3
9.1-10.9
10.0
7.9-11.3
9.9
7.4-10.9
9.1

8.7-11.1
9.9
8.7-10.6
9.7
8.5-11.6
10.2
8.4-10.8
9.4
9.1-11.4
10.3
9.3-11.6
10.3
9.0-11.7
10.2
9.2-11.6
10.5
7.3-10.4
9.4
9.0-11.3
9.9
9.3-11.2
10.2
9.2-12.0
10.8

10.1

9.0-11.0
10.1
9.3-11.3
10.4
9.4-11.8
10.8
9.5-11.3
10.4
9.2-10.5
9.8
8.0-11.3
10.2
9.0-11.9
10.1
9.6-12.3
10.7
8.1-11.6
9.1
9.4-11.6
10.5
10.0-11.2
10.4
9.1-12.0
10.8

10.3

9.3-11.3
10.6
9.6-11.2
10.4
9.8-11.3
10.5
9.3-11.7
10.4
9.2-10.7
9.9
9.1-11.1
10.2
9.3-10.7
10.0
9.3-11.4
10.5
7.9-11.1
9.0
10.4-12.5
11.3
10.0-11.2
10.2
11.5-13.2
12.1

10.4

9.4-11.5
10.7
8.6-11.6
10.4
9.2-11.7
10. G
9.3-11.3
10.4
9.5-11.7
10.7
9.3-10.6
9.9
9.0-11.6
10.3
9.0-12.1
10.5
8.2-11.1
9.5
9.9-11.3
10.7
9.4-11.6
10.4
9.7-13.0
10.8

10.4

Buffalo
Chicago

Kansas City. . .
Louisville

Newark

New Orleans . . .
New York . . .

's'.s^ioii

9.4
8.4-11.7
10.1
8.1-9.9
9.0

Omaha

St. Louis

Scran ton. .

9.9-11.9
10.6
10.1-11.0
10.5

9.9

8.5-10.3
9.4

Washington.. . .
Averaee. .

9.6

SURFACE MIXTURES.

321

Since 1896 these irregularities in bitumen have grown smaller
and the average percentage of bitumen higher, as can be seen
from the average percentage of bitumen in the mixtures laid under
the author's supervision as long ago as 1897:

AVERAGE COMPOSITION, SURFACE MIXTURES, 1897.

City.

Bitu-
men.

Passing Mesh.

2*4
<

200

100

St. Louis, Mo

10.5
10.4
10.9
10.8
10.7
10.1

10. e

10.2
10.1
10.1
10.5
10.2
11.4
10.5
10.0
10.4
10.6
10.8
10.6
10.1
10.5
10.1
10.4
10.3
10.8
11.1

10 R

20.8
18.9
8.9
14.0
16.4
14.0
14.8
14.5
14.8
14.2
14.6
15.5
14.6
16.0
13.1
13.7
14.5
12.4
11.3
12.7
12.8
13.2
12.3
13.1
10.2
10.9

11.4
13.1
23.0
16.4
12.9
15.2
14.2
13.9
13.4
12.5
12.1
10.9
11.4
9.7
12.1
11.5
9.9
11.3
11.9
10.0
8.2
7.6
8.1
6.9
7.5
6.4

16.3
13.7
25.6
16.5
12.4
14.4
11.1
13.2
11.2
9.2
10.7
8.8
13.9
9.8
7.8
14.6
10.8
13.7
12.8
8.9
8.5
7.3
10.3
5.0
6.7
12.7

28.7
30.9
27.8
32.4
24.7
38.7
38.2
37.5
23.9
24.6
25.3
24.5
26.7
28.7
29.7
41.5
21.7
41.2
32.8
25.9
26.6
30.5
27.9
34.1
24.2
41.0

6.0
4.6
2.4
4.5
8.4
4.9
7.1
6.9
9.1
10.8
9.7
9.4
8.8
9.2
15.6
5.1
11.2
4.3
7.6
11.9
13.7
15.6
11.0
19.9
15.4
8.6

3.4
3.4

1.0
2.4
7.1
1.9
2.5
2.9
7.6
10.2
7.5
9.8
6.3
8.1
8.7
2.1
10.2
24
7.1
10.4
9.7
12.2
10.6
6.8
12.0
5.9

1.8
2.6
0.3
1.5
4.1
0.6
0.8
0.6
4.7
5.6
4.3
6.5
4.0
4.2
2.1
0.7
6.5
1.7
3.2
5.9
6.3
2.7
5.3
2.8
8.5
1.9

1.1

2.5
0.1
1.5
3.2
0.3
0.7
0.2
5.2
2.8
4.7
4.3
2.9
3.7
1.0
0.4
4.6
2.2
2.6
4.1
3.7
0.8
4.1
1.1
4.8
1.4

63
61
75
56
58
61
67
63
59
62
58
58
55
61
56
61
51
57
51
45
51
54
57
63
60
57

Kansas City Mo. . . .

Elmira, N. Y

Chicago, III

New York N Y

Buffalo, N.Y..4-B Pit...
" " 3-B " .
Niagara Falls, N. Y. .

Boston, Mass

New York, N. Y., Bronx.
Yonkers, N. Y.*

Newark N J

Jersey City N J

Sioux City, Iowa
Saginaw, Mich
Buffalo, N. Y., 1-B Pit...
New Orleans, R.R. Pit. . .
Detroit, Mich

Pittston Pa

New Orleans La

Wilkesbarre Pa.

Rockford 111 . ...

Scranton, Pa

Harrisburg, Pa

Washington D C

Akron, Ohio. .

Average

* Retained on 10-mesh, .5%.

Experience has shown, however, that in these surfaces, although
the average percentage of bitumen reached 10.5, it was in most
cases too hard, averaging 58 ; which has resulted in some cracking.
In subsequent years, therefore, it has been the practice to use
softer asphalt cement. The results have been very satisfactory.

Too small an amount of bitumen in a mixture permits the
easy entrance or absorption of water, which eventually disin-
tegrates and rots the surface. It reduces the tensile strength

322 THE MODERN ASPHALT PAVEMENT.

and prevents the accommodation of the surface to the contrac-
tion of the mineral aggregate, which follows a rapid fall of tem-
perature, as this can only be met by the elongation of the bitu-
men. In both of these ways lack of bitumen is a direct cause of
cracking and deterioration of pavements.

Analyses of specimens of old surface from Omaha, grouped
and arranged according to their condition, show that the badly
cracked pavements in that city contain the least bitumen and
the better pavements the most, as appears from the following
figures :

AVERAGE BITUMEN IN OMAHA ASPHALT SURFACES.

Good 10.0%

Medium 9.4

Badly cracked 8.6

The results of the examination of the surfaces collected in
1894 and of the data available at that time having shown nothing
more than the fact that there was no uniformity in the way the
mixture was made before then, and that it would be necessarv
to extend the investigation still further to find which was the
most desirable composition, this work was continued as oppor-
tunity offered during a period extending over two years and with
extremely interesting results, which were published in Bulletin
No. 1 of the Office of the Superintendent of Tests of the Barber
Asphalt Paving Company, in March 1896, the substance of which
was as follows:

" The attention of the author was attracted, as long ago as
1889, to a particularly good asphalt surface on Vermont Avenue,
in Washington, D. C., which, although subjected to light traffic,
had had scarcely a repair after, at that time, ten years' service.
An analysis of this surface gave the following results :

"Bitumen 11.3%

Passing 200-mesh sieve 16.0

100-
80-
50-
40-
30-
20-
10-

8.7

5.2

32.0

16.4

6.0

2.7

1.7

100.0
Density 2.18

SURFACE MIXTURES. 323

" The high percentage of bitumen and of dust, both unusual
at the time the surface was examined, led to the conclusion that
the desirable properties of this surface were due to the presence
of plenty of bitumen and dust. In order to confirm this, several
other surfaces were selected in Washington which were typically
good or bad, and it was found that the best were characterized
by a similar composition to that of the Vermont Avenue surface,
while the inferior were deficient in both asphalt cement and dust.
An inquiry as to the conditions under which the Vermont Avenue
surface was laid showed that the sand in use was from a pit and
contained much fine material, on which account it was eventually
abandoned.

" In 1893 attention was called to the excellent character of
the asphalt surface on Court Street in Boston, which had sustained
successfully a very heavy traffic. The surface mixture was
examined by the author and found to have the following compo-
sition :

" Bitumen 11 . 7%

Passing 200-mesh sieve 14.5

" 100- " " 11.2

" 80- " " 24.1

" 50- " " 20.5

" 40- " " 5.8

" 30- " " 4.6

" 20- " " 4.0

" 10- " " 3.6

100.0

" In this mixture high percentages of bitumen, of dust and of
fine sand were found, which was in confirmation of the original
conclusion that the Vermont Avenue surface in Washington wore
well because it contained high percentages of these materials.
These results led to the suggestion that the inquiry should be
extended to a collection of representative surfaces from various
parts of the country. This was undertaken and led to the same
general conclusion, namely, that surfaces carrying the most bitu-
men and dust or filler are the most satisfactory.

" In 1895 the inquiry was extended still further, and an exam-
ination of the street surfaces laid in a western city during the

324

THE MODERN ASPHALT PAVEMENT.

period extending from 1888 to that year, some of which were much
more satisfactory than others, was made. The results of the
analyses of surfaces representing different years' work were as
follows :

1888

1889

1890

1891

1892

1893

1894

1895

" Bitumen

9 85

10 35

9 35

9 05

10 55

9 85

9 35

9 50

Passing 200-mesh sieve. . . .

9.00

9.60

7.50

10.60

9.25

9.00

11.40

10.95

100-

8.80

25.45

10.10

9.60

5.80

6.30

16.60

13.95

80-

10.20

23.05

10.50

14.50

6.65

5.40

16.30

17.50

50-

26.00

20.05

20.00

30.20

26.50

36.30

21.50

31.00

40-

12.30

4.55

12.70

8.20

14.40

10.80

6.10

4.95

30-

11.90

4.05

13.90

6.90

12.55

8.20

7.80

6.40

20-

6.60

2.80

7.60

5.00

8.25

6.50

6.00

2.70

" 10-

3.35

2.10

8.35

5.95

6.05

7.65

4.95

3.08

"If these results are grouped more closely, calling 200-mesh
material dust, 100- and 80-mesh material fine sand, 50- and 40-mesh
medium, and 30-, 20-, and 10-mesh coarse sand, the analyses catch
the eye more quickly.

1888

1889

1890

1891

1892

1893

1894

1895

" Bitumen

9.85
9.00
19.00
40.30
21.85

10.35
9.60
46.50
24.60
8.95

9.35

7.50
20.60
32.70
29.85

9.05
10.60
24.10
38.40
17.85

10.55
9.25
12.45
40.90
26.85

9.85
9.00
11.70
47.10
22.35

9.35
11.40
32.90
27.60
18.75

9.50
10.95
31.45
35.95
12.18

Dust

Fine sand

Medium sand . . .

Coarse sand . .

"The characteristics of these surfaces as noted on the streets
were:

" 1888 Soft; pushes and calks.

1889 Considered one of the best mixtures.

1890 Calks badly.

1891 Calks badly.

1892 Calks badly.

1893 Calks worst of all; very mushy.

1894 Hardly marked; very stable.

"The 1894 mixture is the only one which has produced a surface
which is reasonably free from calking in hot weather. Of the

SURFACE MIXTURES. 325

1895 surfaces we cannot judge until another year, although they
at present are very promising and probably quite as good as those
of 1894. 1 The 1889 surfaces are in better form than the surfaces
of years prior to 1894. Those of 1891, 1892, and 1893 are so yield-
ing as to be a mass of calk marks hi summer. How this occurs is
seen from the differences which are brought out by analyses. The
1888, 1890, 1891, 1892, and 1893 surfaces are deficient in fine sand
as compared to those of 1894, and this is especially the case with
those of 1892 and 1893, where there is but 12.45 and 11.70 per cent,
respectively, of fine material. They do not carry enough sand
grains of this size to make the surface dense, and of course con-
versely they have too much coarse material. It is apparent, there-
fore, that, with the available sand, the grading must be so arranged
that the coarse part shall not run as high as 20 per cent, preferably
about 15 per cent, and the fine shall reach about 30 per cent, the
dust being about 11 per cent, to give a stable surface. The bitumen
in these mixtures is too low when compared with the amount found
necessary for good surfaces in most cities, but here it was due to
peculiarities in the sand, owing to which it will not carry more,
and is therefore not as serious a defect as it would be in some
other places.

"As a whole the experience in this city was very encouraging
in its confirmation of previous conclusions, and sufficiently so to
render an attempt to follow them out in practice in other places
desirable."

The bulletin then goes on to present two illustrative cases
where an explanation had been sought for the good or bad wearing
properties of asphalt surfaces:

" In New York there has been during the present winter (1896)
some scaling of surfaces laid in the autumn of 1895, whereas others
have shown nothing but the best results. Typical of these were
surfaces on Fifth Avenue at Fifty-ninth Street and on Eighth
Avenue at Twenty-eighth Street. Analyses were made of speci-
mens of these surfaces, which quickly explained the differences in
behavior. The results were as follows:

1 At the present time it can be seen that the mixtures of 1895 have proved
as desirable as it was expected they would.

326

THE MODERN ASPHALT PAVEMENT.
NEW YORK MIXTURE LAID IN 1895.

Per Cent.

" Bitumen

9.8
7.3
5.9 \ u 7
8.8/ 14 ' 7
24.4
12.4
9.1)
10.7^30.6
10.8)
.8

11.1
9.8

ll.llio

7.5/ 18
28.6
6.6
10.3 )
8.0^25
7.0J

.6
.3

Passing 200-me
100-
80-
50-
40-
30-
20-
10-
Retained on 10

*sh sieve

< i

1 1

< t

100.0

" The same striking contrast between a good surface and a
poor one is here again well illustrated in the difference in bitumen
and fine material in the two specimens.

" Again the unsatisfactory surfaces from St. Louis, samples of
which were sent in recently for examination, show that a coarse
mixture is an inferior one and likely to scale. The results of an
examination of the St. Louis surfaces were as follows. See results
given in first two tables on page 327.

"In this way the original conclusions of earlier years have
been confirmed, and it has become the present policy to work upon
the lines above indicated in laying surface. While nothing abso-
lutely fixed can be suggested as a universal mixture, perhaps for
the present the following may be considered as an ideal towards
which to work.

"Bitumen

10.0%

or above

Passing

200-mesh

sieve

10.0%

1 1

100-

1 1

10.0%

80-

1 1

20.0%

50-

24.0%

1 1

40-

10.0%

30-

1 1

8.0%

20-

1 1

5.0%

10-

3.0%

1 1

" The grading should not, apparently, be stretched too far as
in such a case but little asphalt cement can be gotten in, and the
surface will lack elasticity.

SURFACE MIXTURES.

327

ST. LOUIS SURFACES OF 1892.

Per Cent.

"Bitu
Passin

it
tt
it
tt
tt

9.7

6.*9

9.7
17.7)
20. 2V 56. 5
18.6)

100.0

10.0
7.1
7.31 1?8
10. 5/ 17 -*
19.0
8.4
13.9)
12.4V37.7
11.4)

g 200-mesh sieve

100- "

80- " .

50- "

40- "

30- "

20- M ...

10- " "

100.0

ST. LOUIS SURFACES OF 1893.

Per Cent.

Per Ont.

Per Cent. N

Per Cent.

" Bitumen.

9.2

10.6

9.4

10.2

9.3

Passing 200

2.8

5.5

7.0

6.3

6.6

" 100
80

9.4 r 19 -

si} 13 -

S.'SJ 13 ' 8

5.61 9 g
4.3J y ' 9

5.61 Q3
3.7/ 9 ' 3

37.0

27.3

17.8

18.4

10.2

" 40

10.7

11.1

12.7

10.0

30
" 20
" 10

10.8)
5.7 f-21.3

4.8)

15.3)
9.6 [32.5
7.6)

18.1 )
15.5 V40.9
7.3)

15.5)
13.2V42.5
13.8)

21.1 )
20.4 V54. 6
13.1 )

100.0

" Finally, it must be remembered that many mixtures which
are quite different from the fine ones which have been mentioned
have furnished good surfaces for light traffic. In Washington, D. C.,
for instance, a recent mixture (1896) analyzed as follows, and
will, no doubt, serve entirely well there:

"Bitumen 11.4%

Passing 200-mesh sieve 7.2

100-
80-
50-
40-
30-
20-
10-

2.9
3.1

14.4
16.9
19.1
16.4
8.6

100.0

" In the same way many coarse streets in Buffalo have served
as well as could be desired.

328 THE MODERN ASPHALT PAVEMENT.

" In conclusion, attention must be called to the fact that a
high percentage of bitumen is not safe in a mixture which is defi-
cient in dust and fine sand, especially when it is to meet heavy
traffic, because such a surface is unstable without the material
which gives it stiffness and capacity to resist pushing and mark-
ing. This is the reason the use of large percentages of asphalt
cement in some cases has been the cause of trouble and has led
to the use of mixtures deficient in asphalt cement for streets of
heavy traffic. In most cases, with plenty of dust and fine sand,
the per cent of asphalt cement in the mixture with steam re-
fined Trinidad asphalt can be carried well above 15 per cent.

" CLIFFORD RICHARDSON,

"Superintendent of Tests.

" Long Island City, N. Y., March 10, 1896."

Soon after the appearance of this bulletin the author carried
out the ideas contained in it in laying a Trinidad lake asphalt
pavement, on the King's Road in Chelsea, London, England.
The composition of this mixture was as follows:

Bitumen 10.8%

Passing 200-mesh sieve 13 . 6

" 100- " " 7.3

" 80- " " 22.5

" 50- " " 25.5

" 40- " " 8.9

" 30- " " 6.6

" 20- " " 3.0

" 10- " " . 1.8

100.0

As this surface resisted entirely successfully the heavy traffic
and fogs of London, where previous attempts with coarser sand
and less filler had failed, it seemed to settle the fact that the con-
clusions drawn in the bulletin were correct, and from that time
to the present all the work under the supervision of the author
on streets of much travel has been done with surface mixtures
made on these lines.

After the London work the next important surface which was
laid was that on Fifth Avenue, in New York. This has proved
successful. Its average composition is as follows:

SURFACE MIXTURES.

329

Bitu-
men.

Passing Mesh

200

100

1896 . .

10.8
10.6

15.4
17.4

10.5
12.3

10.7
11.1

22.3
23.3

10.9
8.8

8.9
8.0

4.9
4.7

5.6
3.7

1897

FOR COMPARISON.

London. .

10.8

13.6

7.3

22.5

25.5

8.9

6.6

3.0

1.8

After from twelve to thirteen years' use the surfaces laid in Lon-
don and that on Fifth Avenue, New York, have proved them-
selves most satisfactory.

Since the year 1896 the asphalt pavements laid by the Barber
Asphalt Paving Company have been constructed, as a rule, in a
similar way to that which has been described; but there has been a
decided improvement in the character of the work with each suc-
ceeding year, as can be seen by comparing the average com-
position of mixtures laid in 1896 in several important cities with
those laid in the same places in 1899 and in 1904.

BITUMEN AND MINERAL AGGREGATE.
1896.

City.

Bitu-
men.

Passing Mesh

200

100

Boston

9.9
9.7
10.2
9.4
10.3
10.3
10.2
10.5
9.4
9.9
10.2
10.8

10.1

10.9
6.7
8.9
12.4
10.9
10.0
10.4
12.9
9.6
14 3
9.9
7.8

13.0
5.6
12.3
7.9
4.9
8.0
8.4
10.4
10.8
9.4
8.5
6.0

12.9
18.1
16.9
15.8
11.3
9.1
10.1
10.6
18.4
13.3
11.0
9.3

24.4
44.4
27.9
36.7
48.3
22.4
28.3
22.8
30.6
33.0
26.1
27.8

9.9
7.3
8.3
6.1
7.3
13.3
13.2
11.9
8.3
10.3
13.5
17.9

8.3
3.4
5.8
4.8
3.3
11.6
11.2
9.5
5.9
6.1
11.0
13.1

5.4
1.9
4.7
3.3
1.8
7.5
5.4
5.8
3.7
2.6
5.9
4.8

5.4
2.7
5.0
3.5
1.9
7.7
2.7
55
3.2
1.1
3.8
2.5

Buffalo 1-B. . .

Chicago Pit. 1

Louisville

Newark .

New Orleans. . . .

New York

Omaha

St Louis

Scran ton .

Washington
Average per cent . . .

330

THE MODERN ASPHALT PAVEMENT.

BITUMEN AND MINERAL AGGREGATE.

1897.

City.

Bitu-
men.

Passing Mesh

200

100

Boston. . . .

10.1
10.4
10.8
10.4
9.8
10.2
10.1
10.7
9.1
10.5
10.4
10.8

10.3

14.8
13.7
14.0
18.9
13.4
15.5
12.7
16.4
13.3
20.8
12.3
10.2

13.4
11.5
16.4
13.1
5.8
10.9
10.0
12.9
12.7
11.4
8.1
7.5

11.2
14.6
16.5
13.7
9.5
8.8
8.9

23.9
41.5
32.4
30.9
49.1
24.5
25.9

9.1
5.1
4.5
4.6
6.8
9.4
11.9

7.6
2.1
2.4
3.4
2.9
9.8
10.4

4.7
0.7
1.5
2.6
1.6
6.5
5.9

5.2
0.4
1.5
2.5
1.1
4.3
4.1
3.2
2.3
1.1
4.1
4.8

Buffalo 1-B. . .

Chicago

New Orleans

NT WvWlr

IN ew i OFK
Omaha. ...

15.9
16.3
10.3
6.7

32.3

28.7
27.9
24.2

6.7
6.0
11.0
15.4

5.1
3.4
10.6
12.0

2.6
1.8
5.3

8.5

St. Louis . ...

Scranton

Washington

Average per cent . . .

1898.

10.6

14 7

13.6

12.8

25 3

9 8

6.7

3 9

2 6

Buffalo 1-B

10 4

14 1

12 6

17 2

35 9

5 5

2 5

1 4

Chicago

10 5

13 2

15 5

13 2

33 6

8 4

1 9

Kansas City

10 4

14 6

12 4

28 6

7 4

5 8

2 g

Louisville

9 9

10 6

10 2

30 9

9 8

6 6

3 4

3 6

Newark

10 2

11 6

11 2

7 5

18 5

12 9

10 6

10 4

6 8

New Orleans

10.0

11 1

7 8

31 1

14 2

3 7

2 1

10 5

13.7

13.1

13 4

22 6

10.0

7.6

5 6

3 5

Omaha

11 8

14 2

17 5

29 9

7 3

4 8

3 3

2 2

St Louis

11 2

13 2

9 7

14 7

37 2

3 8

2 7

1 7

Scranton .

10 2

12 4

10 8

10 5

26 6

11 8

8 9

5 2

3 5

Washington

12.1

12 1

8 2

7 6

25 6

17 6

9 4

4 6

2 8

Average per cent. . .

10.4

1899.

Boston . . ...

10 7

14 5

12 1

9 3

22 2

7 5

5 7

Buffalo 1-B

10 4

13 2

10 6

16 7

37 3

6 3

2 4

1 7

1 3

Chicago

10 6

11 5

14 3

15 5

34 5

7.6

2 8

1 8

1 4

Kansas City
Louisville

10.4
10 7

12.1
15 5

9.2

7 1

13.9
4 6

39.3
36 9

7.1
19 3

3.6
3

2.7
1 6

1.7
1 i

Newark

9 9

14 8

13 7

9 6

11 6

11 3

9 7

10 5

8 9

New Orleans ....

10 3

12 6

11 1

9 3

25 5

6 9

5 1

2 2

New York

10 5

12 7

12.6

23 9

12 1

6 6

5 3

3 3

Omaha

9.5

13.4

13.3

12.6

25.9

11.7

5.6

4.7

3 3

10.7

12.3

12.2

11.0

29.7

11.3

5.4

4.3

3.1

Scranton .

10 4

11 6

10 9

22 9

7 7

5 2

3 3

Washington

10 8

12.7

7.6

5.2

17.4

8.7

7.6

Average per cent. . .

10.4

SURFACE MIXTURES. 331

AVERAGE COMPOSITION SURFACE MIXTURE. 1904.

at y .

Bitu-
men.

Pa*

ing U

e.-h

2
1

age
net ration
A. C. 1

200

100

Alexandria, La. . . .
Allegheny, Pa
Auburn Ind

10.0%
11.2
10 5

11.0%
12.8
12.5

18%
5
13

7%
9
1?

19%
42

15%
11
9

13%
4
6

5%
4

2%
5

62
68

Boston, Mass

11.2

12.8

3?,

Buffalo, N. Y

10.4

17.6

Chicago 111

10 5

11 5

Cincinnati, Ohio. . .
Decatur 111

10.2
11 4

12.8
15.6

13
11

3
5

2
5

54
63

Des Moines, Iowa. .
Detroit Mich

11.0

10 7

12.0
10 3

9
12

30
33

9
10

8
7

3
3

2
?

Ft. Dodge, Iowa. . .
Ft. Wayne, Ind
Grand Rapids, Mich.
Harrisburg, Pa
Kansas City, Mo. . .
New York, N. Y. . .
Los Angeles, Cal. . .
Louisville, Ky
New Albany, Ind. . .
New Orleans, La. . .
Niagara Falls, N. Y.
Omaha Neb

10.9
10.6
10.2
10.6
10.3
10.9
11.0
11.2
10.8
10.1
10.3
10 9

12.1
11.4
11.8
13.4
21.7
14.1
12.0
15.8
17.2
10.9
12.7
13.1

12
11
11
6
18
11
14
12
7
15
11
7

13
12
10
8
13
10
12
8
5
13
13
13

27
27
23
32
16
28
20
26
23
23
26
39

13
14
12
15
6
13
11
13
17
14
10
6

6
7
11
7
6
7
8
7
12
8
9
4

5
6
5
6
4
8
4
6
4
3
4

2
2
4
3
3
2
4
3
2
2
4
3

i
i

64
66
62
66

53
65
54
67
69

Ottawa Ont

10 6

15.3

Pittsburg Pa. . . .

10 4

12.6

Sandusky, Ohio. . . .
Seattle, Wash
Spokane, Wash. . . .
St. Louis, Mo

10.4
12.3
12.9
11.3

8.6
12.7
11.1
15.7

13
14
14
15

12
11
11
14

28
25
22
27

12
11
8
6

7
7
9
5

7
4

4
3
5
2

67
80
77
75

St. Paul Minn . .

10 9

14 1

Tacoma, Wash
Toronto, Ont
Trenton, N. J.

11.9
10.7
10 5

12.1
16.3
10.5

13
21
9

10
14
14

24
27
29

6
14

8
3

I
3

1
2

...

76
61
73

Walla Walla, Wash.
Wichita Kan

13.4
10 3

7.6
11.7

14
10

12
16

27
3?

8
10

8
5

9
3

...

Average

10 9

The general improvement and greater uniformity reached by
experience between 1896 and 1899 and 1899 and 1904 is marked
in several particulars. The average per cent of bitumen in the
more recent mixtures is at a far better figure, 10.9, because the
grading in the late* <nineral aggregate is more satisfactory, since

332 THE MODERN ASPHALT PAVEMENT.

it holds more 200-mesh dust and, as a rule, a better percentage
of 100- and 80-mesh sand. In some cases, of course, the sand
grading could still be improved upon in this direction, but this is
the case only where no suitable fine sand was available within
reasonable distances, as in Washington and Louisville, while a
falling off of the 100- and 80-mesh grains in New York in 1904
is due to inability to find such sand, this material being derived
in 1899 from ballast coming to the port, none of which is now to
be had. The New York surface mixture of 1904 is, therefore,
not as satisfactory as it was in the former years.

The New York mixture of 1899 was regarded at that time as
an unexceptional one, and it was decided to consider it a standard
for mixtures on streets of any traffic for the remainder of the country.
In round numbers the composition of this mixture and of the sand
of which it was composed, regardless of the small amount of 200-
mesh material which it contained, was as follows:

Sand.

Bitumen 10 . 5%

Passing 200-mesh sieve 13.0

100-
80-
50-
40-
30-
20-

13.0 17.0%

13.0 17.0

23.5 30.0

11.0 13.0

8.0 10.0

5.0 8.0

10- " " 3.0 5.0

100.0 100.0

With the object of explaining to the practical man, the super-
intendent or yard foreman, the features of such a standard mix-
ture it was considered from the point of view of consisting of a
mineral aggregate composed of sand and dust and a proper per-
centage of bitumen. The mineral aggregate must be regarded
as being made up of three elements the fine sand, which is the
most important, the coarse sand, which is desirable, and the dust
or filler, which is absolutely necessary. The mineral aggregate of
a standard mixture may, therefore, be considered from the following
points of view:

1st point 100- and 80-mesh sand 17-1- 17 =34%

2d " 10-, 20-, and 30-mesh sand 10 + 8 + 5 = 23

3d t Filler + 200 sand. Dust + fine sand . . =17

SURFACE MIXTURES.

333

Or for the complete surface mixtures:

1st point 100 and 80 sand 13 + 13=26%

2d " 10, 20, and 30 sand 3 + 5 + 8 = 16

3d " Filler + 200 sand =13

4th " Bitumen =10.5

Or these points may be expressed in one of the. following ways.
ASPHALT SURFACE MIXTURE.

Bitumen
10.5%
(4th point)

FUler, 13.0%

Correct surface

(3d point)

mixture, 100%

Mineral aggre-

gate, 89.5%

(1st point)

(2d " )

, (3d " )

Sand, 76.5%

(1st point)

(2d " )

26.0%

Mesh.
100.. 13.0

80.. 13.0
(1st point)

50 23.5%

40 11.0%

30. ..8.0

20... 5. Oj-16.0%

ASPHALT SURFACE MIXTURE.

30. ..8.0)
20...5.0V16.
10... 3.0)
(2d point)

Composition,

i 4th point

Mineral
aggre-
gate,
89.5%

Correct
asphalt
mixture,
100%

Filler+ 200 sand 13

.0 3d point '

Sand,

76.5%

^'^oKo-Ut point
50 23.5

40 11.0

30 8.0)

20 . 5 0>16.0 2d point

10.. . 3.0>

334 THE MODERN ASPHALT PAVEMENT.

The surface mixture, therefore, may be regarded:

1st. As a whole.

2d. As a mixture of bitumen and a mineral aggregate.

3d. As a mixture of bitumen, filler, and sand.

4th. As a mixture of bitumen, filler, 100- and 80-mesh sand
and 10-, 20-, and 30-mesh sand in suitable proportions.

For example take a New York mixture:

1st. New York mixture.

2d. 10.5 per cent bitumen, 89.5 per cent mineral aggregate.

3d. 10.5 per cent bitumen, 13.0 per cent dust, 76.5 per cent sand.

4th. Bitu- 200 100 80 50 40 30 20 10

men, Dust, , . ,

10.5 13.0 26.0 23.5 11.0 16.0

10.5 13.0 13.0 13.0 23.5 11.0 8.0 5.0 3.0

In forming an opinion, therefore, of an old or new surface mix-
ture it becomes evident that the four points which have been de-
scribed must be considered. These points may be differentiated
from the composition of an old mixture or combined to form a
new one.

The primary consideration is the sand and the first point that it
shall contain a normal and sufficient amount of 100- and 80-mesh
material. This was, and undoubtedly is to-day, the most essential
feature in making a satisfactory mixture. It is essential because
without this fine sand the mixture is porous and open, and more
particularly because, unless it is present, a sufficient amount of dust
or filler cannot be used. The fine sand prevents the dust from
balling up and making a lumpy mixture and, as will eventually
appear, the larger the amount of fine sand the more dust can be
introduced without difficulty. In the earlier mixtures, 1880 to
1896, a large percentage of dust could seldom be used, although
the attempt was often made, as the resulting mixture was difficult
to handle and rake.

The second point or consideration lies also in the sand grading
and is the regulation of the amount of the 10-, 20-, and 30-mesh
sand grains. In the Fifth Avenue mixture this material amounted

SURFACE MIXTURES. 335

to 16 per cent. It was unavoidable there, owing to the character
of the sand available, but was believed to be desirable in several
ways. In the first place, it seemed to fill the place taken by broken
stone in hydraulic concrete, and to carry the traffic, so to speak.
In the second place, it gave a less slippery surface than a finer
mineral aggregate. In both these ways the coarse material is
desirable, but closer study and experience has shown that at times
it may be reduced or largely omitted to advantage, especially in
damp climates.

To bring about a satisfactory arrangement of the first two points,
or sand grading, one or more kinds of sand are necessary, usually
more than one. For example, in the Fifth Avenue mixture the
main sand supply was deficient in 100- and 80-mesh grains. It
was, therefore, necessary to add a certain amount of fine sand
consisting predominantly of grains of this size.

The third point, and one also of great importance, is that the
amount of filler or dust shall be sufficient. In the standard mix-
ture of 1899 this was intended to reach, together with the small
amount of 200-mesh sand and the natural filler present in Trinidad
asphalt, 13 per cent. In the older, coarse Washington and St.
Louis mixtures of the early nineties the filler and 200 sand rarely
reached 7 per cent, and hi St. Louis fell, at tunes, below 3 per
cent. This was attributable to two causes: one, the fact that
such coarse mixtures would not carry much dust without balling,
and the other, because it was considered at that tune uncertain
if there were any merit in using a filler. We now know that dust
gives stability to the mixture, aids in excluding water, and that
the best surfaces are those which, up to a certain limit, contain
the most filler. In the standard mixture of 1899 the largest amount
of dust which such a sand grading could carry was about 13 per
cent, owing to the relatively small amount of 100 and 80 sand
grains. Beyond this percentage the mixtures would become
greasy or would ball.

With the first three points arranged in a satisfactory way, the
fourth or last point was to decide on how much asphalt cement
the mineral aggregate would carry. This has been determined
in recent years by the pat test, described on pages 351 and 514,

336 THE MODERN ASPHALT PAVEMENT.

which readily shows whether an excess or deficiency in asphalt cement
has been used. This test cannot, in all probability, be improved
upon. If each grain of material in the mineral aggregate is coated
with asphalt cement and the voids more than filled the excess
will be squeezed out hi making a pat and stain the paper exces-
sively. If the voids are not filled the only stain on the paper
will be a light one from the cement coating the grains of sand. A
perfect mixture will contain just enough cement to fill the voids
in the aggregate, stain the paper well but not excessively (Figs.
6, 7, 8, and 9). The hotter the mixture the more liquid the
asphalt cement and the freer the stain. Cold mixtures will give
no indication, while the difference in the markings of a fine and
coarse sand will be readily learned by experience.

The preceding instructions are satisfactory for turning out
a mixture for the conditions ordinarily met with if the available
materials admit of following them, or for judging the character
of old surfaces when they have been resolved into their constitu-
ents by analytical methods.

In cases where there may be an excess of fine sand, particu-
larly of 200-mesh material, some modification of the method of
procedure which has been described will be necessary. This
will be taken up later. 1

Work on the Old Rule-of -thumb Basis. In comparison with a
standard mixture made according to the previous instructions
work done without any rational method of control is instructive.
Several such mixtures have been examined which were laid in
Chicago in 1898 and 1899 by contractors exercising no technical
supervision over their work. See first table on page 337.

There is hardly a mixture among these that is not open to
criticism in one respect or another, while that laid under the
author's supervision could in itself be slightly improved. The
mixtures are more or less deficient in coarse sand as compared
with the standard adopted. This is general, if it is a defect, and
is due to the character of the local sand. The Bermudez mixture
is very deficient in bitumen and for no other reason except that

1 See page 345.

SURFACE MIXTURES.

337

enough asphalt cement has not been put in. It would easily
carry more, as the sand is very fine, quite too much so.
CHICAGO, ILL., MIXTURES OF 1898 AND 1899.

Bitu-
men.

Passing Mesh

Retained
on 10.

200

100

8.9
11.2
10.3
8.5
10.8

11.1
12.8
13.7
8.5
6.2

20.0
8.0
5.0
5.0
18.0

35.0
11.0

28.0
28.0
26.0

20.0
43.0
37.0
41.0
31.0

1.0
7.0
4.0
7.0
5.0

1.0
4.0
2.0
2.0
1.0

1.0
2.0
0.0
0.0
1.0

1.0
1.0
0.0
0.0
1.0

1.0

Trinidad lake asphalt. . . .
Alcatraz asphalt

Standard asphalt

Trinidad land asphalt.. . .

FOR COMPARISON.

Author's supervision

10.611.414.0

15.o|35.o| 8.o| 3.o| 2.0 l.OJ

The second mixture is one which is hardly open to serious
comment. The mineral aggregate should have, however, rather
more 80- and 100-mesh grains, so that they should together reach
25 to 27 per cent.

The Alcatraz has only 6 per cent of sand coarser than a 50-mesh
sieve, and it is unbalanced in its 80- and 100-mesh sizes.

The Standard mixture is very inferior and cannot prove sat-
isfactory. It is very deficient in bitumen, dust, and 100-mesh
sand, three of the important factors hi a good wearing surface.

The Trinidad land asphalt should have more filler, but it is
not otherwise defective, in so far as the mineral aggregate is con-
cerned, except in the usual absence of coarse sand.

As a whole these mixtures are excellent examples of ordinary
work done empirically and without proper control.

In other cities even more glaring defects are often met, as can
be seen from a few examples:

Cities.

Bitu-
men.

Passing Mesh

200

100

Toronto, Canada. . .
Utica N. Y

8.4*
8.0*
9.9

24.6*
26.5*
13.1

22.0
31.7
3.0*

12.0 22.0
17.9 ! 7.7
3.0*24.0

5.0
5.6
22.0

3.0
1.5
12.0

2.0
0.8
7.0

1.0
0.3
6.0

Indianapolis, Ind. . .

* The shortcomings are here marked with an asterisk-

338 THE MODERN ASPHALT PAVEMENT.

That the cardinal points in a mixture were often neglected
in the earlier days can also be seen from an examination of the
materials and their proportions in a mixture sent out on August 8,
1895, for use on Eighth Avenue in New York:

SAND COW BAY. (GOODWIN.)

Passing 200-mesh sieve Trace

" 100- " " "

" 80- " " "

" 50- " " 5.0%

" 40- " " 13.0

" 30- " " 42.0

" 20- " " 37.0

" 10- " " 2.0

DUST TUBE-MILL.

Passing 200-mesh sieve 66 . 0%

" 100- " " 20.0

80- " " 14.0

ASPHALT CEMENT.
Bitumen 65.0%

PROPORTIONS.

Sand 801 Ibs 79.5%

Dust 60 " 5.9

A. C. . ..147 " . 14.6

1008 100.0

A mixture made from the above materials in the proportions
given would have .about the following composition:

Bitumen 9.5% 4th point

Passing 200-mesh sieve 8.9 3d "

100-

" 131

80- " . l.O/ 2 ' 3lst

50- ' ' 4.1

" 40- " " 10.4

" 30- " " 33.

20- " 29.6^64.8 2d

* 10- " " .

13. 5-|
!9.6[64.!
1.7J

100.0

SURFACE MIXTURES. 339

It is evident that none of the four points in a good mixture
is approached in this one. It is deficient in fine sand, far too
coarse, contains too little dust, and would not hold enough bitu-
men.

This is, of course, an exaggerated case, but much mixture of
a similar description has been sent out and is being made to-day
by ignorant contractors.

Problems Arising from Lack of Sand Suitable for Obtaining
the Standard Grade. Our illustrations and experience have shown
that at tunes the sands to be found in any locality do not permit
of attaining the standard grade which has been proposed. For
example, in Washington, D. C., there is no sand available which
will supply the proper amount of 100- and 80-mesh material
in sufficient amount. In other cities there may be an excess of
200-mesh sand. Again, in some localities, coarse sand is an expen-
sive article, and it is impossible to introduce into the mixture
the normal amount of 10-, 20-, and 30-mesh grains at any reason-
able cost. Finally, questions arise as to whether under some
trying conditions a mixture cannot be made which is more resistant
to unfavorable environment than the standard and as to whether
sands of the same grading in different localities the grains of
which may have a different surface and a different shape, and in
consequence of the last fact may have different voids, can be
handled in the same way as New York sand. The proper amount
and the consistency of the asphalt cement to be used under vari-
ous conditions must also be determined. These points have
been so far settled by the results of investigations carried out
during the last few years that the problems can now be discussed
fairly intelligently.

Mixtures Necessarily Coarser than Standard. The City of
Washington was once in a situation of this kind. Until recent
years, there has been no available supply there of what is known
as "tempering sand," and the surface mixture in use was on this
account deficient, more or less, in 80 and 100 mesh grains. In
the early days of the industry this deficiency was a serious one.
The average composition of the surface mixture laid in 1889 was
as follows:

340 THE MODERN ASPHALT PAVEMENT.

Density 2.10

Bitumen 9.7%

200-mesh-sieve 9.3

100 " " 3.0

80 " " 5.0

50 " " 20.0

40 " " 20.0

30 " " 18.0

20 " " 8.0

10 " " 7.0

100.0

This mixture is plainly deficient in 80- and 100-mesh sand
grains, in the percentage of bitumen, and probably in filler or
actual dust, since less than 4 per cent of ground limestone, of
which not more than 60 per cent passes a 200-mesh sieve, was
added to the mixture, although the 200-mesh material reaches
9.3 per cent, more of this being in the form of sand grains than of
dust, a condition which investigations to be described later will
show has a decided effect upon the character of the mixture.

The streets on which the surfaces of 1889 were laid were sub-
jected to very light travel and were fairly satisfactory for that
period, but when they were examined in 1894 it was evident
that they could have been improved upon by the selection of
a better mineral aggregate, containing more filler, and which,
consequently, could carry more bitumen.

Of recent years there has been a distinct iimorovement
in the grading of the mixture laid in Washington, as shown
by the following analysis of the pavement laid by the Brennan
Construction Company on Pennsylvania Avenue in 1907.

PROPORTIONS.

Asphalt cement (Bermudez) 102 lbs.= 11.5%

Limestone dust 55 " 6.2

Sand. . 731 " = 82.3

888 " =100.0%

SURFACE MIXTURES. 341

ANALYSIS.

Bitumen 10.5%

Sand, passing 100-mesh sieve 25 .0

11 retained on 100-mesh sieve 10.3

" " " 80 " " 12.4

" " 60 " " 26.6

it a "40" " 23 1

n n n 20 " tf ?.6

In some other cities similar conditions are met with in regard
to the available materials for making the mineral aggregate, and
experience has shown that where the streets to be paved are care-
fully drained and the traffic is not heavy such a mixture will prove
satisfactory, although it is probable that if the standard grading
had been employed with a cement which is somewhat softer than
that which would be used on a heavy-traffic street the life of the
pavement would be somewhat extended.

On the other hand, it is the opinion of certain experts that a
coarser mixture is more desirable for streets of light traffic, and
that where the surface is not thoroughly rolled out and closed up
thereby it is more satisfactory than a finer one. There is a possi-
bility that this may be so if the finer standard mixture is not made
with a softer asphalt cement, the experience of the author in 1896,
in several western cities where streets were paved having no traffic
at all, having shown that standard surface mixture laid with a rather
hard cement cracked to a very considerable extent after two years.
Where, however, a standard mixture was laid on such streets with
a very soft cement, cracking has not taken place under the same
conditions. For the reason that finer sands are not available in
Washington, or in the belief that a coarser mixture is more desirable,
the more recent specifications for asphalt pavements in that city
call for a sand of the following grading:

100-mesh At least 10%

80- and 100-mesh ...." " 25%

" 10-, 20-, and 30-mesh " " 15%

342

THE MODERN ASPHALT PAVEMENT.

In view of the above facts, where fine sand is not to be pro-
cured readily, a modified standard for sand grading and finished
mixture has been adopted.

STANDARD GRADING FOR LIGHT TRAFFIC.

Sand.

Mixture.

Bitumen

10%

Passing 200-mesh. . .

100- " ...

11 | __

91 -.0

80- "

ll)22%

9/ 18

60-

40-

30-

13]

101

20-

10 V30

8 (-24

" 10-

7 1

6 1

100

A very considerable amount of work which has proved entirely
satisfactory in small cities and towns has been done on this basis
under the author's supervision. Such mixtures would not, how-
ever, be satisfactory in all large cities, except on residence streets,
and it is because most of the mixtures of the careless contractor
are never more satisfactory in their grading than this that they
are not entirely successful in their work where it is subjected to
heavy traffic.

Excess of Fine Sand of 100- and 8o-Mesh Size. Where the
regular sand supplies contain an excess of 100- and 80-mesh material,
and where it is impossible to introduce into the mixture the normal
amount of 10-, 20-, and 30-mesh grains at any reasonable cost, a
new problem is brought to our attention. Such a situation is
complicated by the fact that an excess of 100- and 80-mesh grains
may or may not be accompanied by the presence of a large
amount of 200-mesh material.

If the 200-mesh material is not present the mineral aggregate
can, generally, be treated in much the same way as the standard

SURFACE MIXTURES. 343

grading, merely allowing for the fact that the greater surface exposed
by the grains of the fine material necessitates the use of a larger
percentage of asphalt cement. The resulting mixture may be
quite satisfactory and, on the other hand, may possess less stability
than it should and be more liable to cracking at low temperatures
and to displacement. In other respects it may be preferable to
the standard mixture, if sufficient filler is used, owing to the fact
that the surface is a closer one than when the coarser particles are
present.

It may also be necessary to use sand in which, while the coarser
particles are present in nearly normal amount, the distribution of
the finer sand, the 80- and 100-mesh grains, may be quite different
from that found in the standard grading. Such a condition will
necessitate changes in the handling of such a sand, as will appear
when the consideration of the amount of bitumen which a mineral
aggregate will carry is arrived at, and this may be conveniently
taken up at this point.

The standard New York sand without 200-mesh material or
filler should contain 17 per cent of grains passing the 100-mesh
and 17 per cent passing the 80-mesh screen, resulting in the pres-
ence of only 13 per cent of each of these grades in the finished
mixture. For the purpose of studying the effect of an alteration of
the proportions of these two sands some sands have been made up
on an experimental basis and the voids, weight per cubic foot, with,
and without filler, determined. See results tabulated on page 344,

In the sand, both with and without filler, No. 1, the lack of 100-
mesh grains and increase of 80 above the usual proportion causes
an increase in the voids over those found in the standard New
York grading. An increase in both 80- and 100-mesh grains
to 5 and 6 per cent each above the normal, No. 5, reduces the
voids decidedly with the plain sand and slightly when filler is present,
but with all the other arrangements, while the sands alone may
be improved, there are larger voids when the filler is present than
in the normal mixture. It seems, therefore, that unequal amounts
of 80- and 100-mesh are not desirable, but that perhaps larger
amounts of both might be, since the voids, when 45 per cent of
the two sands are present instead of 34 per cent, are reduced

344

THE MODERN ASPHALT PAVEMENT.

WEIGHT PER CUBIC FOOT AND VOIDS IN NEW YORK SAND r
WITH VARYING PERCENTAGES OF 100- AND 80-MESH
MATERIAL.

N. Y.

Regular
Grading
with same
Sand.

Passing 100-mesh sieve

30%

28%

17%

22%

17%

< < SO- ' ' ' *

" 50- " "

a 40 _ u

" 30- " "

<i 20- " "

tt 1Q ,

100

Per cent 100-mesh grains.
on '( ft

Low
High

High
Low

High
Normal

Normal
High

High
High

Normal
Normal

Weight per cubic foot
with no 200

108.3

110.0

111.5

112.5

109.9

Voids.

35.1

34.1

33.2

32.8

34.2

Weight per cubic foot
with 13 per cent dust . .

119.4

119.6

118.9

119.0

121.7

120.4

Voids.

28.5

28.3

28.8

28.7

27.1

27.8

slightly. The presence of so much fine material, however, it is
feared, would make the mixture mushy, and in addition it is generally
very difficult and expensive to accomplish this, since such material
is not always available. More asphalt cement is also necessary to
cover the fine grains, which makes the mixture more expensive,
without an adequate return.

The effect of such changes in the standard grading upon the
percentage of bitumen which the mixture will carry is well illustrated
by the analyses of the following mixtures which were turned out
in New York under the author's supervisions in 1899. See table on
page 345.

It must be added, however, that some of the difference in
the percentage of bitumen in these cases may be due to a variation
in the shape or surface of the sand grains as well as to the grading,
and that similar results might not be obtained with the same
grading for sands from other localities.

SURFACE MIXTURES.

345

NEW YORK MIXTURE PAT PAPERS ALL WELL STAINED.

Standard
Average.

D
Av. N. Y.

Week
Ending

Date . ...

\.Y '99

2_7-'00

9-20-' 99

2-28-'00

8-26- ? 99

Proportions :

Sand

790

775

765

Dust

100

110

Asphalt cement

149

95 (Ber.)

167

Remarks:

Per cent of 100-mesh grains

ft Cf)_

Normal
Normal

Low
High

Normal
Low

High
Normal

Normal
High

Passing 100-mesh sieve

17%

12%

19%

28%

16%

" 80-

50-

" 40-

" 30-

" 20-

10- "

100

Bitumen

10 5^

9 5<7r

9 5^

11 3v

10 4n

Dust and sand (passing 200

sieve)

15 5

14 7

1 1

Sand. .

76 5

77 5

100.0

Effect of 2oo-Mesh Material. In the case of sands which con-
tain a very considerable amount of material passing the 200-mesh
sieve the conditions will be found to be different from any of those
which have been previously discussed. That portion of the sand
which will pass a 200-mesh sieve may consist, as has been previously
shown in considering pulverized mineral matter for use as a filler,
of particles resembling sand and of more impalpable material,
which may be considered as true dust or filler. The effect of a
large proportion of 200-mesh grains in the sand on the surface
mixture will depend largely, therefore, on whether they are sandy
or fine enough to act as a filler, and also largely on the character
of the sandy grains themselves, that is to say, their shape and
surface. If the coarser 200-mesh grains are round a mixture con-

346 THE MODERN ASPHALT PAVEMENT.

taining any considerable amount of them, especially if the remainder
of the sand is largely of 100- and 80-mesh size, will be very mushy
and readily displaced. In 1901, in Kansas City, Mo., a mixture
was turned out the sand of which contained as much as 14 per cent
of sandy grains of 200-mesh size. With 10 per cent of filler the
resulting mixture was very mushy and marked badly on the street,
although it showed by analysis over 19 per cent of 200-mesh material
and consequently might be supposed to be a stable mixture. An
increase of the filler to 12 per cent improved the general character
of the mixture very much, but it was never satisfactory and the
use of this sand was abandoned, although it was at first hoped that
such a fine mixture, giving an extremely close surface, might be
more satisfactory than the coarser standard mixture.

On the other hand, in Toronto, Ont., and in Rochester, N. Y.,
where the sands at the same time contained 23 and 20 per cent,
respectively, of 200-mesh material, the amount of filler could not
be carried beyond 4 per cent, as a larger quantity made both mix-
tures very bally and impossible to roll and rake on the street. In
these cases the 200-mesh material apparently acted in itself largely
as a filler. At other points loamy sands have been found the
loam in which, when it does not bake into balls on being heated
in the sand-drums, proves to be a satisfactory filler.

The character of a 200-mesh .material in any sand cannot be
determined by the use of sieves, as nothing finer than the 200-mesh
sieve is available and this will not differentiate between sand
grains of 200-mesh size and the impalpable powder which acts
as a filler, but this may be done by elutriating the material by
the method described elsewhere. In the sands in use in New
York, especially in that from Cow Bay on Long Island, consider-
able extremely fine material is found, this amounting at times
to 10 per cent or more passing a 200-mesh sieve. On separation
of this material and elutriation it was found that between 50 and
60 per cent of it would at times be in the nature of a filler and
at others not more than 30 per cent. In a case where the pro-
portion of sand and filler were about the same it was found that the
mineral aggregate would still carry a very considerable further
proportion of filler and that the grading must be regarded as

SURFACE MIXTURES.

347

being extended in the fine direction as if there were a possibility
of differentiating the material with finer sieves than are available.
Such a mineral aggregate would carry between 11 and 12 per
cent of bitumen, frequently approaching the latter. As will be
seen when considering the grading of a coarse asphaltic concrete,
in such a material the percentage of bitumen is much reduced by
the addition of the larger particles, and it may, therefore, be
assumed that on either side of our standard grading we may place
other gradings according to the following scheme. It is, of course,
to be understood that with very fine grains a certain amount of
fine filler would be present, although theoretically absent, while
the same would hold in regard to fine material with coarser
mixture.

Sieves.

Stand.
Mix.

Bitumen

10 5

600 ....

500

400

300

200

100
80
50 . .

8
5
3

11
8
5

24
11

13
24
11

13
13
24

13
13
13

13
13

40
30

5
3

8
5

24
11

13
24

13
13

The above diagram shows that, theoretically, the standard
can be pushed up and down, according to the amount of fine and
coarse material which is present, and that at the same time it
will be found that the amount of bitumen which the mineral
aggregate will carry will change to a marked degree. This may
be illustrated by the following mixtures:

348

THE MODERN ASPHALT PAVEMENT.

New York,
1904.

Chicago,
1901.

Boston,
1901.

Newark,
1901.

Biti
Pas

imen

12.0%
19.0

io.o\ 1Q

9.0/ 1
26.0
12.0
6.0]
4.0^ 12
2.0J

11.4%

18.6
33.01 4?
14. 0/ 47
18.0
2.0
1.0]
1.0* 3
l.Oj

10.7%
14.3
12.01
12.0 J 24
26.0
10.0
7.0]
5.0 [ 15
3.0J

9.6%
9.4

ii.o\ 17

6.0/ 17
18.0
16.0
12.0]
9.0 1-30
9.0J

sing 200-mt
100-
80-
50-
40-
30-
2C-
10-

;sh sie

ve. . .

100.0

ASPHALTIC CONCRETES.

Barber Asphalt
Paving Co.,
New York.

Warren Bros.,
St. Louis, Mo.

Bitumen

6 2%

3 4% 1

Filler .

7 8

2 9

Sand. .

Stone passing \" screen. . . .

18.0]

4.2]

21.0^57
18. Oj

6.6l gl -
55.6 81 ' 7

' ' retained on 1" screen .

15. 3J

100.0

1 Coal-tar soluble in CSz.

It is very evident from the preceding that the grading of a
mineral aggregate has a very large bearing on the amount of bitu-
men that it can carry, and it may be stated as a general rule that :

1. If the sand is finer than the standard, increase the filler
and the asphalt cement.

2. If the sand becomes coarse, reduce the filler and the asphalt
cement.

3. If the 200-mesh material in the sand is high, determine
whether it is sand or a filler. If it is a sand, and cannot be avoided
by the use of sand from some other source, and if it acts badly
in the mixture, get rid of it if possible by means of blowing it
out with a forced draft or suction, or add more filler and more
asphalt cement if this is impossible. If a portion of it is of the
nature of a filler allow for this in the amount of filler that is

SURFACE MIXTURES.

349

added. If the only available sand contains 200-mesh grains,
which make a mushy mixture, remedy this by the addition of
more filler if possible. In some cases the 200-mesh sand will give
a mushy mixture under all circumstances and in this case every
effort should be made to do away with the use of such material,
or to remove the defect by mixing it with some other sand supply.
Examples of the percentage of 200-mesh material which is
so fine as to act as a filler, since it does not settle in water in 15
seconds, is shown for the sands of various cities in the following
table:

ELUTRIATION OF 200-MESH MATERIAL FROM PLATFORM SAND

Test number

71714

Kansas City,
Mo.

3%
17.2%

55560

Chicago,
111.
6%
10.9%

56372

Toronto,
Ont.

23%
8.6%

City . ...

Material passing 200-mesh. . .
Acting as filler

74814

Ottawa,
Ont.

13%

27%

73176

Seattle,
Wash.

10%
46.4%

72355

Buffalo,
N.Y.

12%
43.2%

73420

Long Isl-
and City,
N.Y.

18%
54.3%

City. .

Material passing 200-mesh . . .
Acting as filler

Examples of very mushy mixtures have not been infrequent
in the West. Some years ago a mixture was turned out in Louis-
ville, Ky., having the following composition:

Bitumen 11-7%

Passing 200-mesh 17.3

" 100-
tt

tt
tt

80-
50-
40-
30-
20-
10-

to}"

9.0
9.0
1.0i
l.Oj 3

l.OJ

39.0
19.0

100.0

550

THE MODERN ASPHALT PAVEMENT.

In this mixture about 10 per cent of dust was being used before
it was brought to the attention of the author. On examination
it was found that the sand was deficient in 100- and 80-mesh
grains and contained from 5 to 7 per cent of loam acting as a filler.
The reduction of the dust to 4 per cent changed its character so
much that it was no longer mushy. The modified surface has
proved entirely satisfactory.

In the early days of the industry much difficulty was met
with, as has been mentioned in discussing the nature of sands,
in turning out a satisfactory mixture in two western cities where
river sands were in use. In both cases this was only overcome
by the entire abandonment of the supplies in use and the selection
of others. With the old supplies the mineral aggregate would
not carry ten per cent of bitumen and the amount varied with
different deliveries of sand. In another city the mineral aggre-
gate, while well graded, carried too little bitumen to permit the
finished surface from responding to the great contraction, due
to sudden drops of temperature, with the result that all the pave-
ments in this city were a mass of cracks. With the selection of
other sand supplies this has been overcome, and mixtures having
the following composition have been produced:

City

Mo. 1.

City

No. 2,

1896.

1904.

1896.

1901.

9 9%

11 3%

9 4%

10 6%

Passing 200-mesh . .

14 1

15 7

9 5

114

" 100- "

1 1

" 80- "

170

" 50- "

" 40- "

11.0

" 30- "

" 20- "

" 10- "

100.0

Here the difficulty lay in the fact that the sand first in use
was composed of grains which had a surface of such a nature that a
thick coating of asphalt cement would not adhere to them.

SURFACE MIXTURES. 351

The Amount of Asphalt Cement or Bitumen which a Mineral
Aggregate Will Carry. It has become very evident from what has
been said in the preceding pages that the amount of bitumen or
asphalt cement in any mixture is very variable, depending upon the
grading of the mineral aggregate and upon the peculiar surface of
the sand grains. It is a self-evident fact that this amount in any
mixture should be sufficient to thickly coat every particle of mineral
matter and fill the voids in the sand, if the latter are sufficiently
small, in the actual size of the spaces between the grains but not
in volume, to permit of doing so without making the resulting
asphalt surfaces too susceptible to temperature changes. With
too much bitumen the sand grains composing the mineral aggregate
are readily displaced among themselves, especially in the absence
of a sufficient amount of filler, and the surface is not stable and
will mark badly and push out of shape. With too little bitumen
the surface cracks, owing to its inability to withstand sudden
changes in temperature, and also becomes displaced because it is
not a solid mass. Mr. Dow's illustration, which compares an asphalt
pavement to a seabeach at different states of the tide, is an excellent
one. Beach sand with the voids just rilled with water, as the
tide goes out, is firm and stable. A horse hardly marks it. When
it begins to dry out it is loose and is readily displaced. When it is
supersaturated with water it is a quicksand.

The proper amount of bitumen for various mineral aggregates
in' common use may vary from 9 to over 14 per cent. As examples
sands found and in use in Moline, HI., in 1902, would carry but 8.5
per cent of bitumen, while in Paris, France, and London, England,
11.5 per cent could be used, and in Glasgow, Scotland, and Seattle,
Wash., over 12.5 per cent, as shown by the following analyses. See
table on page 352.

If a strict interpretation of the instructions is followed and only
10.5 per cent of bitumen is introduced the mixture with such sand
will, of course, be unsatisfactory, and such difficulties have been
frequently met with owing to lack of judgment on the part of
yard foremen and superintendents.

The actual amount to be used in any case must be determined
by the pat-paper test described on page 514. This test, however,

352

THE MODERN ASPHALT PAVEMENT.

is deceptive unless the mixture is at a temperature where the asphalt
cement is quite liquid. With cold mixtures the test is of no value,
while excessively hot ones may stain the paper too freely.

Citv .

Moline

Paris

London

Glasgow

Seattle

Bitumen soluble in CS 2
Passing 200-mesh sieve . .

8.4%
15 6

11.2%
14 7

11.1%
15 3

12.0%
18

12.3%
12 7

" 100- "

18 7

12 7

11.0

" 80- "...

23 1

20 5

9.0

" 50- " ...

16.0

26 3

33 7

23.0

40- "
" 30- "
" 20- " "
" 10- " "

17.0
13.0
9.0
3.0

3.9
1.6
.4
.1

3.6
1.5
1.1
.5

4.0
2.0
0.0
0.0

15.0
10.0
5.0
2.0

iOO.O

100.0

Characteristic pat papers are illustrated on the following sheets.

The paper, Fig. 6, illustrates a light stain made by a mixture
which, although of a proper temperature, is deficient in bitumen.
The paper reproduced in Fig. 7 illustrates a medium stain,
showing a slight deficiency in bitumen. The paper reproduced
in Fig. 8 shows a strong stain produced by a standard mixture
carrying a suitable amount of bitumen. The paper reproduced
in Fig. 9 represents a heavy stain, pointing to the presence of
an excess of bitumen if the temperature of the latter is not abnor*
mally high.

Coarse sands such as are found in abnormal mineral aggregates
give a rather different stain. Experience will prove the best
means of interpreting the test. It is a very valuable one with
Trinidad lake asphalt mixture, but less so with others, as other
bitumens are more susceptible to temperature changes.

In making a pat test the appearance of the surface of the ho<*
pat is quite as instructive as that of the stain upon the paper,
since if the mixture is unbalanced in any way greasiness is often
visible, which should be removed by the adjustment of sand, filler,
and bitumen, which can only be accomplished by experiment,
the reduction of the amount of filler accomplishing this at one
time and increasing it at another.

FIG. 6. Light Stain.

353

FIG. 7. Medium Stain.

354

FIG. S. Strong Stain.

355

FIG. 9. Heavy Stain.

356

SURFACE MIXTURES.

35"

Reasons for the Necessity for a Larger Percentage of Bitumen
in a Fine than in a Coarse Mixture. It has already been mentioned
that one reason why the finer mixture requires a larger percentage
of bitumen than the coarser one is that the extent of surface of
the sand grains to be covered with bitumen is much larger in the
former than in the latter case.

The number of particles in a gram of grains of uniform diameter
and of different sizes and the square centimeters of surface exposed
by one gram of such grains are presented in the following tables.

These figures are obtained by means of the following formulas:

and

surf ace =x(d) 2 n,

where a is the weight of the particles, in this case one gram, d the
diameter of the particles, aj the specific gravity of them, and where
n is the number of particles.

Where the diameter, d, is given in centimeters and a in grams,
the following constants are useful:

LOGARITHMS OF CONSTANTS IN CENTIMETERS.

Vlesb, Sieve

*d)*

log ic(d)*n.

.150

6.6705180

8.8493321

.084

6.9150320

8.3457081

.058

6.4325281

8.0240055

.040

5.9484241

7.7012695

.026

5.3871640

7.3270961

.020

5.0453441

7.0992095

100

.013

4.4840743

6 . 7250363

200

.008

3.6515141

6.3033295

1 minute

.005

3.2491541

5.8949895

30 minutes

.0025

2.3360641

5.2930295

2 hours

.00075

0.7674280

4.2472721

16 "

.00025

0.3360641

3.2930295

log ^- = . 1422441.

358

THE MODERN ASPHALT PAVEMENT.

ONE GRAM OF SAND OF UNIFORM SIZE CONTAINS.

Mesh Sieve.

Millimeter.

Particles.

Square
Centimeters.

1.50

212.8

15.0

.84

1,215.9

27.0

.58

3.693.6

39.4

.40

11,261.0

56.6

.26

41,005.0

87.1

.20

90,066.0

113.2

100

.13

328,032.0

174.2

200

.08

1,407,620.0

283.0

1 minute

.05

5,643,700.0

442.4

30 minutes

.025

46,124,900.0

905.7

2 hours

.0075

6,800,990,000.0

1,201.6

16 "

.0025

46,124,900,000.0

9,056.6

1 gram to 1 Ib. X 453.59, log. 2.6566654 for particles.
Sq. cm. to sq. ft., divide by 2.9680569 for surface.

Where it is desired to determine the number of particles and
the surface exposed by grains of different sizes which go to make
up an aggregate of definite weight, the preceding formulas become:

where a is the weight of each group of particles and A the total
weight of the material, in the following case one pound. Using
this, the number of particles and the square feet of surface in one
pound of the mineral aggregates in New York mixtures of 1895
and 1898 are found to be as follows. See tables on page 359.

It appears that the finer aggregate presents a surface of
60.5 square feet to the pound, the coarser only 44.4, or 39,407 and
52,093 square feet per 9 cubic foot box respectively; that is to
say, the finer has one- third more surface, and as this must be covered,
more asphalt will be required for the finer than the coarse aggre-
gate. This explains why the addition of dust will always increase
the amount of asphalt cement which a mixture will hold, although
the voids may be reduced, as a pound of the best Long Island

SURFACE MIXTURES.

359

NUMBER OF PARTICLES AND THEIR SQUARE FEET OP

SURFACE IN ONE POUND OF SAND AND DUST.

NEW YORK MIXTURE, 1895.

Mesh Sieve.

Per Cent.

Particles.

Square Feet
of Surface.

12,592

.958

66,186

1.579

167,547

1.901

664,250

3.593

5,021,870

11.479

4,086,220

5.527

100

10,415,700

5.952

200

31,924,300

6.909

.005 mm.

76,671,400

6.480

100

129,030,065

44.378

One box of mixture (average 888 Ibs.), 39,407.664
NEW YORK MIXTURE, 1898.

3,674

.295

38,808

.921

150,796

1.715

561,866

3.033

4,835,870

11.054

6,127,910

8.099

100

22,319,400

12.754

200

44,694,000

9.672

.005mm.

153,343,000

12.960

100

232,075,324

60.503

One box of mixture (average 861 Ibs.), 52,093.083

City dust with the following siftings will contain the number of
particles and the surface in square feet given below :

Size Particles,
Centimeters.

Per Cent.

Number of Particles.

Square Feet
of
Surface.

.008
.005

.0025
.00075
.00025

18.8
17.7
51.3
5.0
7.2

120,035,300
452,360,200
10,732,9-50,000
30,775,730,000
150,634,400,000

25.976
38.231
226.825
58.590
178.199

100.0

192,715,475,500

527.821

or if all of t
.0025 cm.

ie dust is of
n diameter:

20,922,050,000

442.157

360 THE MODERN ASPHALT PAVEMENT.

This enormous area of surface allows the presence of a larger
quantity of bitumen in a fine mixture than in a coarse one. The
question of the thickness of the film of the melted asphalt cement
on the extended surface of the sand grains is one which, from
the elaborate studies of soil physics, it is plain must be of great
importance in regulating the amount of bitumen which different
sands will require to prevent porosity and instabilty in the mix-
ture. Much pertinent information on this point will be found
in Whitney's Bulletins, Weather Bureau, Division of Soils, U. S.
Department of Agriculture, and Wiley's Principles of Agricultural
Analysis, but our understanding of the question at present is
insufficient to permit of going into the matter at this time. It
will be investigated in the future.

In this connection some recent determinations by Messrs.
Briggs and McCall, of the Bureau of Soils, U. S. Department of Agri-
culture, " On the Thickness of Adsorbed Aqueous Films," are of
interest. They found the following values for several materials:

Silica 167.00X10-' cm.

Glass 18.00X10- 6 "

Quartz 45X10~ 8 "

The application of these data to an asphalt surface lies in
the fact that sand may consist of particles which may vary as
largely in the thickness of the film of asphalt which will adhere
to them as the materials experimented with above, and this may
explain why one sand with the same grading and voids as another
may hold different percentages of bitumen.

The New York mixture is desirable in so far as it will usually
carry 10.5 to 11.5 per cent of bitumen, when the grains are all
coated and the voids filled, as shown by the paper pat test, and
this affords a sufficient amount to keep out water and provide
for the contraction due to a rapid fall in temperature.

With the low voids it might at first be assumed that such $ mix-
ture would hold less asphalt than one with a greater volume, but
it has already been shown that the low voids, when accompanied
by plenty of fine sand, do not have this effect, as the adsorbed
bitumen, or that necessary as a paint coat to cover the more numer-

SURFACE MIXTURES.

361

ous small grains, is something to be considered beyond that necessary
to fill the voids. On the contrary, a well-graded fine mixture
with small voids will often carry more bitumen than a coarse
one with larger voids.

Comparison of the Characteristics of Different Sands Having
the Same Sand Grading. If the sand used in New York, arranged
according to the grading in the mixture laid in that city in 1898
and 1899, is to be regarded as most satisfactory, as shown in the
following figures, it must be by no means assumed that on
that account the New York sand is the best sand; that is to
say, consists of the best shaped grains or of those having the
best surface to afford a proper adhesion of the asphalt cement
and allow of a sufficiently thick coating. As a matter of fact
the contrary is the case. It is possible that with other sands
accommodated to the New York grading even better results could
be obtained than with the New York sands themselves.

1898.

1899.

Bitum
Passin

tt
tt
tt
tt
tt
tt

10.5%
13.7
13.1
13.4
22.6
10.0
7.6
5.6
3.5

10.5%
13.0
12.7
12.6
23.9
12.1
6.6
5.3
3.3

T ^00-mesli sieve

3 100- " "

80- " " .

50- " " . .

40- " "

30- " "

20- " "
10- " "

100.0

Experiments have been undertaken and completed for determin-
ing what the differences are in this respect in the available sands
in different cities of the country.

It has been shown that the grading of the New York mineral
aggregate is such that the New York mixture is the densest of any
satisfactory one in the country, and it appears that in the case of
the sand of every other city hi the country, when the grading
according to which it was used some years ago is changed to that
of the New York mineral aggregate of to-day (1899), the density

362

/HE MODERN ASPHALT PAVEMENT.

of the resulting mixture is increased with one or two exceptions,
and in the same way if the New York aggregate is changed from
its own grading to those of other cities its density is decreased.

The grading of the local sands with and without dust, the voids
and the weight per cubic foot of each on its own grading, on the
grading of the New York sand and aggregate and of the New York
sand on the local grading as determined in 1900 are given in the
following tables:
GRADING OF SANDS IN AVERAGE MIXTURES IN DIFFERENT

CITIES WITH 200-MESH MATERIAL REMOVED 1898 AND

1899.

Passing Mesh.

Year.

100

Philadelphia, Pa. .
Chicago, 111
Kansas City, Mo. .
New York, N. Y..
Omaha, Neb. .

22
21
17
17
17
16
16
15
11
10
10
7

18
17
19
17
16
14
12
13
14
7
6
7

30
44
37
30
34
38
30
35
25
23
50
23

14
10
10
13
15
15
19
17
14
26
26
26

7
4
8
10
8
7
10
10
12
13
4
13

5
3
5

8
6
6
8
6
13
11
2
11

4
1
4
5
4
4
5
4
11
10
2
10

1899
1898
1898
1898
1899
1899
1899

1898

1899
1898

St. Louis, Mo ....
Boston, Mass
Paterson, N. J. . .
Trenton, N. J. ...
Washington, D. C.
Louisville, Ky. . . .
Youngstown, O. .

GRADING OF SANDS IN AVERAGE MIXTURES IN DIFFERENT
CITIES WITH 13 PER CENT OF 200-MESH DUST.

Passing Mesh.

Year.

200

100

Philadelphia, Pa. . .
Chicago 111

13
13
13
13
13
13
13
13
13
13
13
13

15
18
15
15
15
14
14
13
10
9
9
6

15
15
16
15
14
12
10
11
12
6
5
16

26
38
32
26
30
33
26
31
22
20
44
31

11
9
9
11
13
13
17
15
12
23
23
14

9
3
7
9
7
6
9
9
10
11
3
13

7
3
4
7
5
5
7
5
11
9
2
6

4
1
4
4
3
4
4
3
10
9
1
1

1898
1898
1898
1898
1899
1899
1899
1898
1898
1899
1899
1898

Kansas City, Mo. .
New York, N. Y . .
Omaha, Neb

St. Louis, Mo
Boston Mass

Paterson, N. J. . . .
Trenton, N. J
Washington, D. C. .
Louisville, Ky
Youngstown, O. . .

SURFACE MIXTURES

363

WEIGHT PER CUBIC FOOT AND VOIDS IN NEW YORK SAND,
WITH NO 200-MESH SAND AND WITH 13 PER CENT DUST
MADE UP ON THE GRADING OF VARIOUS CITIES, COM-
PARED WITH THE SAND FROM THESE CITIES OF THE
SAME GRADING AND WITH THAT OF THE CITIES MADE
UP ON THE NEW YORK GRADING.

Specific
Grav-
ity.

With no 200.

With 13 Per Cent
Dust.

Weight
per Cubic
Foot.

Voids.

Weight
per Cubic
Foot.

Voids.

New York . . .

2.67
2.63

2.61
2.63
2.63
2.63
2.66
2.68
2.67
2.66
2.62

109.6

113.3
114.1
110.0

111.1

109.0
113.7

110.4
111.8
110.3

111.9
113.8
109.7

110.6
110.0
110.6

109.6
111.8
106.8

109.1
112.5
107.8

107.8
109.6
110.3

106.1
107.2
112.4

104.5
107.2
108.6

34.1

30.9
30.3
33.3

31.6
32.6
31.7

32.6
31.8
33.7

31.7
31.9
34.1

32.6
32.9
33.5

33.9
32.4
35.8

34.6
32.6
35.1

35.2
34.1
33.7

36.5
34.3
32.4

36.0
34.3
34.7

118.9

124.5
126.0
121.4

123.5
118.6
125.6

122.4
124.0
121.9

122.2
124.0
121.1

121.0
122.2
121.4

120.5
124.2

118.8

120.4
123.2
119.1

119.4
121.7
121.5

119.0
117.3
124.0

115.7
117.3
120.4

28.5

24.0
23.1
27.2

24.1

26.9
24.5

25.3
24.3
26.7

25.4
25.7
27.1

26.2
25.4
27.1

27.3
25.0
28.6

27.9
26.2
28.1

28.2
26.8
27.0

26 8
26.2
25.5

29.1
28.2
27.6

Omaha, N. Y. "

N. Y., Omaha "

Trenton, local grading

Trenton, NY "

N. Y., Trenton ' '

Kansas City, local grading. . . .
Kansas Citv, N. Y. "
N. Y., Kansas City "

St. Louis, local grading

St Louis NY "

N Y , St Louis ' '

Paterson , local grading

Paterson, N. Y. "

N Y , Paterson ' '

Buffalo, local grading

Buffalo, NY "

N. Y., Buffalo "

Chicago, local grading

Chicago, NY "

N. Y. , Chicago ' '

Philadelphia local grading. .
Philadelphia, N. Y. "
N. Y., Philadelphia " ....

Washington, local grading. . . .
Washington, N. Y. "
N. Y., Washington "

Louisville, local grading

Louisville, N. Y "

N. Y., Louisville " ....

364

THE MODERN ASPHALT PAVEMENT.

In the preceding determinations of voids in different local
sands the sands have been freed from all 200-mesh grains before
adding the filler. In practice these sands always contain from
1 per cent of this material in Chicago to 13 per cent in Buffalo.
Where so much 200-mesh material not dust is present the full
amount of filler cannot always be used. The actual average
amount of 200-mesh sand in the supplies of the special cities ,
which have been examined, and that of the filler used in 1899,
with the percentage of the latter passing a 200-mesh sieve, is
given in the following table :

AVERAGE PER CENT OF SAND PASSING 200-MESH TAKEN
FROM WEEKLY REPORTS; ALSO PER CENT USED IN
MIXTURE.

City.

Year.

Average
Per Cent
Passing
200-Mesh.

Average
Per Cent
Dust in
Mixture.

Per Cent
Dust
Passing
200-Mesh.

Per Cent
200 Dust
Added.

Total
200 Sand
and Dust.

12.6
9.5

10.5
15.6
13.0

8.8

9.6
8.7
16.2

16.0

New York
Chicago

1898
1898
1899
1899
1898
1899
1898
1899
1898
1899
1899
1899

5.0
1.0
7.0
10.0
6.0
4.0
2.0
6.0
1.0
11.0
8.0
13.0

8.0
10.0
5.0
8.0
10.0
8.0
7.0
6.0
11.0
8.0
7.0
5.0

95.0
85.0
70.0
70.0
70.0
60.0

60.0
70.0
65.0

60.0

7.6
8.5
3.5
5.6
7.0
4.8

3.6

7.7
5.2

3.0

St Louis

Louisville

Kansas City

Omaha

Trenton

Paterson

Youngstown

Washington
Boston . .

Buffalo

In Buffalo, with 13 per cent of 200 sand, only 3 per cent of
filler can be used, while in Chicago, with only 1 per cent, 8.5 per
cent or more is used. The total per cent of 200-mesh sand and
dust in many cases is below the amount which should be found
in a good mineral aggregate, but it must be remembered that
nearly 3 per cent of 200-mesh filler is contributed to the mixture
by the fine mineral matter where a Trinidad asphalt cement is
in use. Where Bermudez asphalt is the cementing material an
additional amount of filler is of course required.

SURFACE MIXTURES.

365

If all these sands are taken and enough dust added to make
the total 200-mesh material in the aggregate up to 15 per cent
the voids in these aggregates can be determined and the influence
of the presence of the 200-mesh sand investigated. This has
been done and the results follow:

WEIGHT PER CUBIC FOOT AND VOIDS IN THE AVERAGE
SAND OF 1899 FROM VARIOUS CITIES, WITH THE AVERAGE
AMOUNT OF 200-MESH SAND AND ENOUGH FILLER ADDED
TO BRING IT UP TO 15 PER CENT, PASSING 200-MESH, AND
WITH 200-MESH SAND REMOVED AND REPLACED BY FILLER.

City.

Average
200 Sand.

Amount
Dust
Added.

Weight per Cu Ft.

Voids.

Without
200-Mesh
Sand.

With
200-Mesh
Sand.

Without
200-Mesh
Sand.

With
200-Mesh
Sand.

New York

5.0
1.0

7.0
10.0*
6.0
4.0
2.0
6.0
11. Of
13.0
4.0

10.0
14.0
8.0
5.0
9.0
11.0
13.0
9.0
4.0
2.0
11.0

118.9
120.4
122.2
115.7
122.4
124.5
123.5
121.0
119.0
120.5
119.5

120.4
120.7
121.1
114.5
121.4
125.3
125.2
121.2
117.4
117.0
120.9

28.5

27.9
25.4
29.1
25.3
24.0
24.1
26.2
28.8
27.3
28.2

27.6
27.7
25.8
30.0
26.0
23.6
23.0
26.0
28.4
29.4
27.3

Chicago . . .*

St Louis

Louisville

Kansas City

Omaha
Trenton
Paterson

Washington ....

Buffalo

Philadelphia

* Largely fine loam acting as a filler.

t Largely crushed-stone dust acting as a filler.

With only 1 per cent of 200 sand, as in Chicago, little difference
is occasioned, but in Buffalo, with 13 per cent of 200 sand, the voids
are greater with the sand than with this taken out and substi-
tuted by filler. Where the 200-mesh material in the sand is more
of the nature of filler than sand there is little difference, but if
the 200-mesh material is really sand of the largest size which will
pass a 200 sieve the difference is striking. The substitution of
such a sand for filler has been made with the sands from the several
cities and the results show the effect when compared with those
obtained with filler on a previous page :

366 THE MODERN ASPHALT PAVEMENT

EFFECT OF SUBSTITUTION OF SAND FOR FILLER.

13 Per Cent 200-Mesh Sand.

Weight per
Cubic Foot.

Voids.

New York 115.6

Omaha, local grading 118 . 6

Omaha, N. Y. " 118.9

N.Y., Omaha " 118.0

Trenton, local grading 117.1

Trenton, N. Y. " 114.6

N. Y., Trenton " 120.0

Kansas City, local grading 115.0

Kansas City, N. Y. " 116.8

N. Y., Kansas City " 117.0

St. Louis, local grading 116.1

St. Louis, N. Y. " 117.1

N. Y., St. Louis " 118.2

Paterson, local grading 115.6

Paterson, N. Y. " 116.9

N. Y., Paterson " 118.2

Buffalo, local grading 115.9

Buffalo, N. Y. " 118.4

N.Y., Buffalo " 114.8

Chicago, local grading 113

Chicago, N. Y. " 117.6

N. Y., Chicago " 115.1

Philadelphia, local grading 113 . 3

Philadelphia, N. Y. " 112.7

N. Y., Philadelphia " 117.0

Washington, local grading 114.8

Washington, N. Y. " 112.3

N. Y., Washington 120.2

Louisville, local grading 110 . 9

Louisville, N. Y. " 112.8

N. Y., Louisville " 117.3

30.5

27.6
27.4
29.1

28.2
29.5

27.8

29.9

28.8
29.7

29.1
29.9
28.9

29.6
28.6
29.0

30.1
28.4
31.0

32.3
29.6
30.8

33.1
32.3
29.7

31.4
32.8

27.8

32.1
30.9
29.5

200-mesh sand is generally undesirable because it tends to
make the mixture less stable and liable to move, as has already
been shown Our ideal mixture should, therefore, as a rule con-

SURFACE MIXTURES. 367

tain none of this material, and in this respect the New York mix-
ture is at times capable of some improvement, although at others,
with quite large amounts, an extremely satisfactory result is
obtained.

It is evident from the preceding facts that something besides
the mere grading of the sand has large influence on the character
of the mineral aggregate and the asphalt surface mixture pre-
pared from it, and this can probably be explained by consider-
ation of the fact that the shape of the sand grains which are of
size to pass any given sieve may be so entirely different that they
fit together with different degrees of compactness, while the power
of adsorption l of the surface of the sand grams will have an equal
influence. This is not astonishing from what has been observed
in regard to the character of various sand supplies when the
subject of sand was under consideration.

Further Characteristics Indicative of the Properties of Old
and New Asphalt Surfaces. In addition to the consideration of
the preceding characteristics in judging a surface mixture, cer-
tain of its physical properties or those of old surfaces, if one of
these is under examination, must be determined, such as its density
and capacity for absorbing moisture, while others may throw
some light on the nature of old surfaces, more especially such as
their tensile, crushing, and shearing strength. Old surfaces can
also be reheated and the general appearance of the mixture in
this condition noted, including the surface of a pat and the stain
on a pat paper 2 made, at carefully regulated temperatures, as
with a new mixture.

Density. The density of the best mixtures when thoroughly
compacted either by traffic or in the laboratory, as illustrated
by that turned out in New York at the present time, should be
about 2.22 to 2.25 when made with ordinary limestone and 2.27
when made with Portland cement. The density of such a mix-
ture calculated from that of their constituents is about 2.27 and
2.29, so that hi this mixture but a small volume of voids is found.
A comparison with these figures of the actual density of old
street surfaces which have been examined is therefore of value.
1 See page 360. 2 See pages 353 to 356.

368

THE MODERN ASPHALT PAVEMENT.

In the old surfaces as they exist in the streets of many cities
of the country densities of from 1.89 to 2.26 were found. That
on Dodge Street in Omaha had the latter density, and two good
surfaces, one from Warwick Boulevard in Kansas City, laid in

1892, and one from Cumming Street in Omaha, laid in 1893,
had a density of 2.24. These densities are nearly theoretical, and
such surfaces should be able to keep out the water. Howard
Street in Omaha, laid in 1895, has a density of only 1.89, and Ohio
Street in Chicago, laid in 1894 by the Standard Company, has a
density of 2.04. Both of these pavements are cracked, the former
very badly. Attempts to compress the latter mixture in the
laboratory resulted in obtaining a no greater density.

It is not always the case, however, that a surface of high density
does not crack or the reverse. In Omaha, Dodge Street, laid in

1893, has a density of 2.26, but it has cracked probably because
the bitumen was too hard. In Chicago, Tripp Avenue, with a
density of 2.21, has cracked, as has Walrond Avenue in Kansas
City, with 2.24, for the same reason. Baltimore Avenue in Kansas
City, of a density of only 2.11, has not cracked, nor has Thirty-ninth
Street in Omaha, with a density of 2.10. The extreme densities
of the surfaces examined in Chicago, Omaha, and Kansas City
were:

Chicago.

Omaha.

Kansas City.

Good surfaces:
High density
Low "

2.20
2 16

2.24
2 10

2.24
2 11

Medium surfaces
High density. . . .
Low " ....
Cracked surfaces
High density. . . .

LOW "!!.!!

2.15
2.04

2.26
2.09

2.21
1.89

2.15
2.13

It seemed possible that a low density might be due to lack of
compression in surfaces laid in winter. The average density of
the summer surfaces as compared with those laid after Novem-
ber first in Kansas City and Chicago seems to confirm this idea,

SURFACE MIXTURES.

369

but in Omaha the density is slightly in favor of the one winter
surface examined.

Summer
Pavements,
Density.

Winter
Pavements,
Density.

Cracked
Good

Badly ci
Medium
Good

KANSAS
pavements

CITY, Mo.

2.235(1) l
2.201 (3)

jo, ILL.

2.155(5)
2.183(2)

L, NEB.

2.180(2)
2.170(9)

2.145(3)
2.136(2)

2.080(1)

2.182(1)
2.184(2)

CHICAC
pavements

OMAHJ

"acked pavements,
good

1 Number of surfaces examined.

It must be remembered, however, that variations in the rela-
tions of bitumen to sand may make a marked difference in the
densities, since the greater the percentage of bitumen in a mixture
the lower will be its volume weight; that is to say, an excess of
bitumen added to an aggregate will lower the density as much
as a deficiency. The density of the densest mineral aggregate
before the addition of bitumen has been found to be 2.00 in Omaha,
the lowest 1.86 in Louisville. It would not be expected, there-
fore, that a mixture having low voids would have the same gravity
in Louisville as in Omaha.

Capacity for Absorbing Water. Surfaces will absorb water
in amount varying with the density and the percentage of bitu-
men which they contain. With the New York mixture the amount
of water absorbed by it in milligrams per square inch and in pounds
per square yard when a thoroughly compacted cylinder of the
above density is immersed in it for various lengths of time is as
follows. See table on page 370.

It appears that more water is absorbed in the first day's immer-
sion than in any subsequent day and that it diminishes in a good
mixture as time goes on. When the New York mixture is made
with bitumens of different origin the amount of water absorbed

370

THE MODERN ASPHALT PAVEMENT

WATER ABSORBED BY CYLINDER OF NEW YORK TRINIDAD
LAKE ASPHALT MIXTURE, DENSITY 2.24, WHEN IM-
MERSED FOR DIFFERENT PERIODS.

Time.

Milligrams per
Square Inch.

Total.

Pounds per
Square Yard.

Total.

1 day. .

.0169

.0480

2 days

.0021

.0190

.0060

0540

7 "

.0092

.0282

.0263

.0803

15 " .

0045

0327

0127

0930

28 ' ' . .

0035

0362

0101

1031

will vary. This is illustrated by some data, 1 wherein it is seen
that Trinidad asphalt-surface mixtures are quite as impervious
as those made with other asphalts and after a lengthy exposure
in running water are able to resist impact better than any others.

Comparison of Street Surfaces with New York Mixture. A
comparison of the absorption of water by some of the typical
surfaces from old streets in the western cities with that absorbed
by the New York mixture will be of interest. See results tabulated
on pages 371 and 372.

These results show that the absorption in the old-time, poorly
graded surfaces is in inverse proportion to the amount of bitumen
they contain and that those of high density, unless they contain
enough bitumen to fill the voids, as shown by a paper test, gain
more than less dense mixtures with sufficient bitumen,

Such a surface as that on Thirty-ninth Street, Omaha, which
the pat paper shows is excessively rich in asphalt cement, excludes
water better than the New York mixture, as does the rich War-
wick Boulevard surface from Kansas City. The old Howard
Street surface with less than 8 per cent of bitumen, of course,
absorbs more water than any of the others which were examined.
The peculiarities of the other surfaces appear from an inspection
of the results in the table. Twenty-sixth Street in Omaha, although
it has cracked some,* absorbs a comparatively small amount of
water, but it must be remembered that water absorption results
more in disintegration, scaling, and rotting than in cracking.

1 See page 468.

SURFACE MIXTURES.

371

WATER ABSORBED BY CYLINDERS OF OLD SURFACE.

Twt No.

Street.

Density of

Compacted

Cylinder.

Stain on Pat Paper.

NEW YORK MIXTURE.

Fifth Ave. mixture 2.24 Heavy

KANSAS CITY, Mo.

21440 Baltimore Ave. good 2.240 Heavy

21442 Garfield Ave. cracked 2 . 158 Very light

21445 Walrond Ave. cracked 2. 195 Heavy

21446 Warwick Blvd. good 2 . 274 Very heavy, coarse

21447 Seventh good 2.248 Medium

OMAHA, NEB.

21448 23d cracked 2. 142 Strong

21456 20th medium good 2 . 205 Medium

21461 39th good 2.209 Very heavy

23253 Gumming good 2 . 235 Medium

23254 26th cracked 2.217

23256 Capitol medium good 2 . 210 Heavy

23257 Howard cracked 1 .904 None

CHICAGO, ILL.

21431 Prairie good 2.231 Heavy

21433 Tripp cracked 2 . 193 Strong

21435 So. Park Ave. cracked 2.156 Strong

21438 Washington Blvd good 2.201 Medium

Why the Standard Mixture is Satisfactory. The standard
mixture which has been suggested by the author and which is
now universally laid under his supervision where this is possible,
on streets of heavy traffic and elsewhere, has been arrived at by
the examination of surfaces which have proved successful and
not by any theoretical reasoning or experimenting. Practice
during the last eleven years has shown that such a mixture is
successful. The results of laboratory investigations on the sub-
ject have, however, made it possible to explain theoretically and
with a good deal of satisfaction why the standard mixture has
been a satisfactory one. The greatest factors in the construction

372

THE MODERN ASPHALT PAVEMENT

WATER ABSORBED BY CYLINDERS OF OLD SURFACE.

ABSORPTION, POUNDS PER SQUARE YARD.

Test No.

IDay.

2 Days.

7 Days.

15 Days.

28 Days.

Add.

Total.

Add. i Total.

Add. Total.

Add.

Total

NEW YORK MIXTURES.
.0480J .0060| .0540J .0263| .0803| .0127| .0930| .

KANSAS CITY, Mo.

.1031

21440

.089

.029

.125

.113

.238

.096

.334

.146

.480

21442

.141

.063

.204

.342

.546

.354

.900

.379

1.279

21445

.095

.080

.175

.231

.406

.357

.763

.580

1.343

21446

.050

.017

.067

.068

.134

.074

.208

.119

.327

21447

.115

.065

.181

.319

.499

.286

.785

.434

1.219

OMAHA, NEB.

21448

.128

.051

.178

.292

.469

.426

.896

.693

1.589

21456

.106

.060

.166

.234

.399

.185

.585

.267

.852

21461

.032

.012

.045

.052

.097

.046

.142

.091

.234

23253

.062

.027

.089

.125

.215

.146

.361

.210

.571

23254

.068

.025

.093

.118

.211

.133

.344

.226

.570

23256

.066

.043

.109

.415

.524

23257

.273

.103

.376

.531

.907

.767

1.674

.505

2.259

CHICAGO, ILL.

21431

.058

.017

.075

.069

.144

.073

.217

.103

.319

21433

.096

.043

.139

.221

.360

.282

.642

.386

1.028

21435

.106

.053

.159

.291

.450

.338

.787

.473

1.253

21438

.089

.037

.126

.208

.334

.241

.575

.309

.884

of a successful asphalt surface is that the mixture shall be so
dense as to resist the action of water and impact and at the same
time contain sufficient bitumen to permit its responding, without
cracking, to a sudden fall in temperature. The standard mixture
seems to offer these advantages in a way not supplied by a coarser
and more carelessly prepared mixture.

The only way to keep water out of an asphalt surface is to
have the voids in the surface mixture as small as possible in size,
but not necessarily so in volume, to fill them with bitumen of a
consistency which will permit of contraction and to stiffen the
latter with a proper amount of filler which will alone permit of
the use of a sufficiently soft cement. If the interstitial spaces
are few in number but large in size, the asphalt occupying them
will be in such large masses, if the voids are entirely filled, that

.SURFACE MIXTURES. 373

they will easily yield to stress and cause the surface to mark and
push and the pavement to appear soft. If the voids are not filled
water quickly enters and destroys the pavement. If fine sand
is introduced in proper proportions the size of the interstitial
spaces is much reduced, the volume of the masses of asphalt filling
them is reduced in the same way, and the voids can be thoroughly
filled without danger of movement. This is made more certain
by the introduction of a filler into the cement, thus stiffening it as
it exists between the voids. The function of a filler can be seen
by rolling out two cylinders of a cement of the same consistency,
one containing 25 per cent of filler and the other none. Their
ductility or elongation under stress is then found to be as follows:

AT 78 F.

Without filler elongation 20 . 6%

With 25 per cent filler elongation 34 . 5

The part played by the filler in an asphalt surface mixture
is thus made apparent.

Fine sand of 100- and 80-mesh size is desirable, since it is evi-
dent that grains of this size if introduced in the proper propor-
tions among coarser sand grains must reduce the size of the inter-
stitial spaces between the grains even if they do not reduce the
volume of the latter. In this way they play an important part
in the stability of the pavement, but they play a still more important
part in making it possible to use a desirable amount of filler in
the mixture. In the early days of the industry, as it was carried
on in the city of Washington, it was possible to use only a very
small amount of filler in the surface mixture and this never
exceeded 3 or 4 per cent. If a larger amount was added, either
there or elsewhere, where coarse sands were employed, it was
found that when an attempt was made to lay the mixture upon
the street it would not rake or spread with ease and was in a con-
dition which was known as " bally." It was impossible, there-
fore, under such conditions to attempt to close up the surface
of the finished pavement by the use of large percentages of filler,
although attempts were made to do so. When it was found that
the most desirable surfaces contained a considerable percentage
of filler not intentionally introduced into them, and that this

374 THE MODERN ASPHALT PAVEMENT

was accompanied by a similar amount of 100- and 80-mesh
sand grains, attempts to duplicate these mixtures with the fine
sand present showed that in the presence of the latter much higher
percentages of filler could be added without resulting in a " bally "
condition of the hot mixture on the street. A consideration of this
state of affairs will quickly show that this is due to the fact that
in the coarse sand where the size of the spaces between the indi-
vidual grains were large there was an opportunity for the filler to
become balled up with the comparatively large masses of asphalt
cement present there, but when the fine sand was introduced this
material as it was tossed around in the mixer in a hot condition it
broke up these balls and made a smooth and homogeneous mix-
ture which could be raked out on the street with ease. The value
of sand of 100- and 80-mesh sizes is, therefore, to be attributed
to the two causes mentioned above: one, its reduction of the size
of the spaces between the individual sand grains, and, secondly,
to the fact that it permits the use of a proper amount of filler
in the mixture by preventing the collection of the filler into bally
masses.

SUMMARY.

The preceding chapter consists of an elaborate discussion of
the theory of asphalt-surface mixtures which does not admit of
summarization beyond the statement that the construction of a
standard mixture is dependent upon an intimate knowledge of
the behavior of sand and the complete mineral aggregate towards
the bitumen and of the finished surface mixture towards its
environment.

It shows that an asphalt surface to be successful must be so
constructed as to resist weathering and impact, which are the two
most serious enemies of such a surface, and it also shows how
this can be done.

As the surface mixture is one of the most important elements
of the pavement the data collected here will be of great interest
to the asphalt expert or the person desiring to make himself one,
and also to a very considerable extent to the general reader in
revealing the amount of skill which is necessary in handling the
material which enters into the composition of an asphalt surface.

CHAPTER XVII.
ASPHALTIC OR BITUMINOUS CONCRETE.

The term asphaltic or bituminous concrete has been in use
for many years to denote a concrete in which asphalt or some other
bituminous substance has been used as a cementing material
for a mineral aggregate instead of hydraulic cement. Such a
concrete has been employed particularly as a foundation for
heavy machinery, the vibration of which it has been desired to
do away with where this is a nuisance. Owing to its elasticity
and lack of rigidity, it absorbs and does not transmit shock. 1

As a pavement or roadway bituminous concrete in which coal
tar was the cementing material was exploited to a very large extent
nearly half a century ago in the United States. Roadways of
this description were laid in Washington, D. C. in the early J 70's,
among them one on Connecticut Avenue by C. E. Evans in 1873,
under a contract dated August 7, 1872. This pavement was a
failure owing to the character of the cementing material and was
resurfaced and covered up with a new one after a short time. It
remained in place, however, as a foundation of the roadway,
until 1906, when it was removed, on repaving the street, and the
construction of a hydraulic concrete foundation. The author
was present when the old surface was torn up, and collected
specimens of the original Evans concrete as being of interest and

1 See Delano, "Twenty Years' Practical Experience of Natural Asphalt
and Natural Bitumen," London, E. & F. N. Spon, 1893. Malo, "L'As-
phalte. Son Origine, Sa Preparation, Ses Applications," Paris, 1866 and
1898. Letouze et Loyeau, "Traite pratique des Travaux en Asphalte/'
E. Bernhard et Cie, Paris, 1897.

375

376

THE MODERN ASPHALT PAVEMENT.

ASPHALTIC OR BITUMINOUS CONCRETE.

377

as illustrating the earliest bituminous concrete in use as a road-
way, which has been preserved. A sawn section of this pavement

is shown in Fig. 10. It will be seen that the mineral aggre-
gate in this bituminous concrete consists of broken stone of
various sizes, sand and coal tar as a cementing material. This

378 THE MODERN ASPHALT PAVEMENT.

is more strikingly shown by an analysis of the material which
resulted as follows.

EVANS' CONCRETE PAVEMENT, 1873.

Entire Mineral

Pavement. Aggregate.

Bituminous matter soluble in CS 2 3 . 0%

Mineral matter passing 200-mesh screen 3.2 3.4% 3.4

10 " " 37.1 37.1 38.2 38.2

" " 8 " " 2.7 2.8

" " i inch " 10.5 13.2 10.8 13.6

" " " % " " 14.3 14.7

" " " f " " 14.7 15.2

" " " 1 " " . 14.5 43.5 14.9 44.8

100.0 100.0

Density of concrete 2 . 73

" " stone 2.86

" sand 2.65

Voids in mineral aggregate 23 .7%

The concrete does not differ in a great degree from similar
material which has been used as street roadways within the last
few years, at least as far as the grading of the mineral aggregate
is concerned, as will appear from analyses which follow. Much
of it was laid in Washington at the time mentioned, but all of
it was unsatisfactory and required resurfacing within a few years.
It is of interest only as showing that a bituminous concrete was
recognized as long ago as 1873 as a possible material for paving
purposes.

Asphaltic concrete, as far as the author is informed, was first
proposed for use as a pavement in 1896. A sidewalk of this
material was constructed on West Avenue at Sixth Street, in
Long Island City, N. Y., at the plant of the Barber Asphalt
Paving Company at that point. The development of the concrete
used was the result of some investigations conducted to determine
whether a satisfactory material of this description could be assem-
bled from a well graded mineral aggregate and an asphaltic cement
which would be suitable for lining a canal. In order to obtain

ASPHALTIC OR BITUMINOUS CONCRETE.
COAL-TAR CONCRETES, 1902-1906.

379

Cleveland,
Ohio.

1902

St. Louis,
Mo.

1903

St. Louis,
Mo.

1904

Toronto,
Ont.

1904

Water-
bury,
Conn.

1905

Soluble in CS 2 . .

3 9%

3 4%

5 0%

5 5%

4 30?

Passing 200 screen

7.1

2.9

5.6

100 "

18.8

12.0

23.8

21 9

1-inch screen . . .
\ " "

13.9

4.2
6 6

8.2
19

14.7
17 5

6.3
3 6

3 11 (f

16 8

10 8

i " :::

Retained 1 " " ...

56.3

55.6
15.3

38.4
.'0

8.4
6 1

23.3
24 3

Total

100

S &-
Mass'.

1905

Cam-
bridge,
Mass.

1905

Boston,
Mass.

1905

Lynn,
Mass.

1905

Port
Huron,
Mich.

1906

Soluble in CS 2 .

5 3%

4 8%

6 2%*

6 Q%

Passing 200 screen
" 100 "

3.3
29 4

4.4
30 9

4.7
26 5

5.5

1.8

13 7

' l -inch screen . . .

tt i << (f

2
tt 3 <( <(

(( J tl ll

Retained 1 " " '.'.'.

11.4
3.3
12.8
12.5
22.0

15.9
21 A
15.6
6.5
.0

14.4
24.0
23.0
2.6
.0

10.5
23.1
23.1
10.1
2.5

13.2
10.9
5.5
4.6
43.4

Total

100

* Percentages of components used.

a mineral aggregate with a small percentage of voids, it was
recognized that it would be necessary to combine the coarsest
stone which was to be used with finer stone to fill the voids in
the coarser, and this again with sand and fine mineral matter to
further reduce them. The relative proportions were determined
by experiments.

In the actual production of this concrete mixture, crushed
stone, the largest particles of which would pass a screen with
openings three-quarters of an inch in diameter, but containing
a considerable amount of quarter inch material, was heated and
separated into two sizes, coarse particles and those passing a

380 THE MODERN ASPHALT PAVEMENT.

screen with openings three-eighths inch in size. Sand, such as
was ordinarily used in the paving business, was heated separately.
Finely ground limestone was used cold. These four components
were then mixed while the first three were hot, and to them was
added an asphalt cement of proper consistency in the predeter-
mined proportions. The resulting concrete mixture was tamped
into a wooden form from which the resulting block was removed
on cooling. Its dimensions were 3X4X1 in feet. It was a very
successful piece of bituminous concrete, and after exposure to the
sun and weather for nine years had preserved its form and showed
no signs of deterioration. It was very carefully analyzed in
1905, with the following results :

Test number 81662

Bitumen 7.9%

Passing 200-mesh screen 9.3 9.3

" 10 " " 34.7

" 8 " " 6

' ' |-inch screen 6.1 6.7

\ " " 31.6

" 1 " " 9.8

Retained 1 " " .0 41.4

100.0
Voids in mineral aggregate 15.0%

This concrete, as appears from the above data, consists of
a well graded mineral aggregate, the voids in which amount to
only 15% of its volume, cemented together into a resistant mass
by an asphalt cement. It was so satisfactory that its application
as a surface for pavements at once suggested itself. In order to
determine whether it could be used in this way practically, a
driveway in a store-house at the plant of the Barber Asphalt
Paving Company, Long Island City, covering an area of 36 square
yards, was laid with such a concrete, the character of which is
shown by the following determinations made upon a sample of
the surface in 1905, after the roadway had been subjected to the
traffic of loaded vans for nine years, with no repairs.

ASPHALTIC OR BITUMINOUS CONCRETE 381

Test number 80951

Bitumen 6.2%

Passing 200-mesh screen 5.1 5.1

10 " rt 33.6

" 8 " " 3.1

" i-inch screen 9.3 12.4

" \ " ' l 27.0

1 " " 15.7

Retained 1 " " 42.7

100.0
Voids in mineral aggregate 14 . 1%

The general similarity between the concrete of the driveway
and the block is evident. The latter contains more coarse stone,
but the voids in the mineral aggregate are practically the same,
14.1 as compared to 15.0 per cent in the samples examined. The
fact that an asphaltic concrete, the basis of which is a mineral
aggregate of such compact and dense nature as to contain a very
low percentage of voids, could be turned out, in a rational and
not an empiric way, on different occasions, and successfully
placed in forms and as a roadway was thus demonstrated.

At that time there was no sidewalk in front of the office and
laboratory of the Barber Asphalt Paving Company on West
Avenue, the building having "been only recently completed. In
view of the success met in laying the driveway, it was decided
to construct this walk of the same material, so that its behavior

Test number 82157

Bitumen 7.3%

Passing 200-mesh screen* 8.2 8.2

" 10 " " 24.2

ft g ce it -j 2

\ -inch screen 7.9 9.1

\ " " 30.9

1 " " 20.3

Retained 1 " " 51.2

100.0
Voids in mineral aggregate 15 . 5%

382 THE MODERN ASPHALT PAVEMENT.

might be observed in the open air. The area covered was about
90 square yards, and the surface was supported upon a broken
stone and cinder foundation. The character of the mixture
laid is shown on the bottom of preceding page, as determined in
1905 from several samples taken from the pavement at that time.
In producing the preceding mixture, the records show that
the components were combined in the following proportions :

Stone, passing f-inch screen 450 Ibs. = 46 .2%

" " " " 125 " = 12.8

Sand, " 10 mesh screen 250 " = 25.6

Filler, dust, 60% passing 200-mesh-

screen 50 " = 5.1

Asphalt cement 100 " = 10.3

975 100.0

The mineral aggregate, it will be seen, as has already been
mentioned, was made up of two sizes of stone, of sand, and a
filler in the shape of a fine mineral dust. They were combined
in such proportions as to make a very dense aggregate having
voids, varying on account of some segregation in laying the mate-
rial, of from 13 to 17 per cent, and averaging 15.5 per cent. This
aggregate was assembled on the previously well established prin-
ciple that to accomplish the desired result a dense concrete
the voids in the coarsest stone must be filled with particles of
suitably smaller size, and the voids in this combination again
with other still smaller ones in this case sand and still further
with material finer than sand in this case mineral dust. The
actual proportions were worked out by packing the materials
solidly in a box holding a cubic foot and determining the percent-
ages of each necessary to obtain the densest aggregate.

The bituminous concrete made in this way has given great
satisfaction for eleven years. When a demand arose for such a
mixture for roadways, as a novelty, it was very evident that it
was only necessary to duplicate the Long Island City mixture
of 1896. This was done on a considerable scale in Muskegon,
Mich., in 1902, on Muskegon Avenue, under the author's direction.
The area covered was 10,736.16 square yards. The mixture had

ASPHALTIC OR BITUMINOUS CONCRETE. 383

the following composition, as shown by a very careful analysis
made of a specimen taken from the street in 1905:

Test number 81117

Bitumen 7.4%

Passing 200 mesh screen 7.4 7.4

10 " " 34.0

" 8 " " 2.8

" Hnch " 11.2 14.0

" \ " " 18.0

l" " " 19.2

Retained 1 " " .0 37.2

100.0
Voids in mineral aggregate 16 .8%

A comparison of these data with those for the mixture of
1896 shows the striking resemblance of the two materials, al-
though the Muskegon mixture contained less stone. The pave-
ment was, in fact, a duplication of the Long Island City work. It
has proved entirely satisfactory after six years use.

In the following year, 1903, an asphaltic concrete pavement
was laid in Owosso, Mich., with a mixture having a similar com-
position, but containing more coarse stone.

Test number 80904

Bitumen 7.4%

Passing 200 mesh-screen 5.2 5.2

" 10 " " 29.4

" 8 " " 2.2

" i-inch screen 3.5 5.7

" \ " " 9.8

" 1 " " 33.5

Retained 1 " " 9.0 52.3

100.0
Voids in mineral aggregate 13 . 2%

Similar surfaces have been laid in Scranton, Pa., Paterson,
N. J., Newark, N. J., in the Borough of Richmond, New York City,

384

THE MODERN ASPHALT PAVEMENT.

and elsewhere in subsequent years. They have proved satis-
factory on streets of light traffic where street car tracks are absent,

FIG. 12. Asphaltic Concrete Sidewalk, Long Island City, New York, 1896.

but, like all pavements, cannot be maintained against rails that
are not rigid. They will not withstand heavy traffic, as the

FIG. 13. Asphaltic Concrete Surface, Broadway, Paterson, N. J., 1906.

coarse particles of stone ravel out under the continuous impact
of the horses' hoof and shoe, the commonest cause of deterioration
of all forms of street pavement. If, however, the asphaltic con-

ASPHALTIC OR BITUMINOUS CONCRETE.

385

Crete is covered with an inch of standard surface mixture the
resulting pavement has been found to exceed for durability any-
thing that has been hitherto constructed, especially where there

is a possiblity of vibration, as along street railway tracks and on
a base which is not perfectly rigid. There is every evidence
that this form of construction will be the one to be adopted on

386

THE MODERN ASPHALT PAVEMENT.

streets of heavy traffic in the future, and its use is recommended
by the author where trying conditions are to be met. This form
of construction is shown in section in Fig. 16.

Production of Asphaltic Concrete. Asphaltic concrete can be
turned out in any plant which is fitted for the production of the

ASPHALTIC OR BITUMINOUS CONCRETE.

387

binder and surface mixture for the ordinary form of sheet asphalt
pavement, with certain modifications, permitting of the separation
of the stone into two sizes after it has been heated in the ordinary
heaters provided for binder stone and storage in separate bins.
For this purpose the heated stone is passed over a revolving

FIG 16.

apparatus provided for this purpose. The three materials are
then weighed or measured out in the predetermined proportions,
screen of sheet metal having perforations f inch in diameter. The
finer stone which passes the screen is collected in one bin and
the coarser into another. There is no necessity for separating them
into a greater number of sizes. The sand is heated in the ordinary

388 THE MODERN ASPHALT PAVEMENT.

the dust or filler added, and the combined aggregate thoroughly
mixed in the usual pug mill, after which the asphalt cement is
added and the mixing continued until the concrete is homo-
geneous. Portions of the mixture are then compacted under a hot
tamper upon a firm surface to determine how satisfactory it is.
The hot tamper should bring to the surface of the specimen a
slight excess of bitumen, if this is present in correct amount.
If the surface remains dry, more asphalt cement must be added,
but if it is too greasy, it must be reduced. On cooling the specimen
may be broken.

If the concrete is to be used as a binder course, there must be
no excess of bitumen and a slight deficiency of sand over that
required to fill the voids in the stone, as otherwise the surface
will be too readily displaced on such a course under traffic.

The character of the concrete, in either case, can be determined
also by its appearance in the truck after the haul from the plant
to the street. Where it is to be used as the surface to carry
traffic, it should have settled to a compact mass, the top of which
should show a slight excess of bitumen as a rich coating, whereas
if it is intended for a binder course no excess of bitumen should
appear and there should be some evidence of the coarser particles
of stone at the top of the load.

In either case, the asphalt cement should be much softer than
that in use in the ordinary surface mixture containing sand alone.
The penetration should be about 90, as determined by the Bowen
penetration machine.

SUMMARY.

An asphaltic concrete pavement is not a novelty. Such pave-
ments have been laid for over twelve years, and have proved
satisfactory where not exposed to heavy traffic. On streets like
Broadway, in New York, it is unsatisfactory.

CHAPTER XVIII.
ASPHALT BLOCKS.

Pavements composed of blocks of asphaltic concrete in which
the coarsest particles of the aggregate are no larger than \ or f
inch in size, have been laid for a great many years with considerable
success on streets of light or very moderate traffic. The idea
of making such blocks originated in a crude way in San Francisco
in 1869. The first Board of Commissioners of the District of
Columbia stated in its report of 1878 that an experimental piece
of pavement of this description had been laid in Washington on
E Street, the Board having been influenced to do so by favorable
reports of its durability in Providence and Philadelphia. It was
not, however, until the application of powerful mechanical presses
to the compression of the blocks in 1880 that they proved at all
successful. 60,774 square yards of this form of pavement had
been laid in Baltimore up to June 30, 1885, and 45,000 had been
laid in Washington on a foundation of five inches of bank gravel
and a cushion of two inches of sand. The Engineer Commissioner
of the latter city reported that the character of the block had much
improved in the two preceding years. The blocks were four inches
deep, five inches wide and twelve inches long. Many of these
pavements are in existence to-day in satisfactory condition, having
been subjected only to the most moderate travel on residence
streets, and prove the adaptability of uch surfaces to similar
conditions. The blocks were of sufficient depth to support each
other laterally. By June 30, 1889, the area of block pavements
in Washington had increased to 117,164 yards, and it had spread

389

390 THE MODERN ASPHALT PAVEMENT.

to other towns, notably, Baltimore, Md., and Chester, Pa., where
40,000 square yards of block had been laid. This form of pavement
made no great further progress until 1896, when the amount reached
200,000 yards, 33,874 of this being in the Borough of Man-
hattan, and the blocks being 4X4X12 inches in dimension. In
1901 the block as at present laid, 3X5X12, was brought out
and laid on East 27th Street from Third to Madison Avenues,
since which time large areas of blocks of this description have
been put down with varying success, the pavement failing uni-
versally under heavy travel and requiring renewal or extensive
repairs in from one to two years, and revealing the fact that an
asphalt block pavement at least, as at present constructed, is
suitable only for residence streets of the lightest travel. The
causes of this are not far to seek after a study of the nature of
the blocks and of the technology of the industry.

Asphalt blocks consist of a mineral aggregate and a bituminous
cementing material, as is the case in the sheet or monolithic
asphalt pavement. They differ essentially from the latter form
of construction in that the aggregate is much coarser, containing
particles as large as J or f inch in size, while the cementing mate-
rial is of a much harder consistency, to enable the blocks to be
handled in the course of their transfer from the place where they
are made to the street and to be stored for some time without
losing their shape. Blocks of the ordinary surface mixture which
is used in the construction of a sheet asphalt pavement which
would withstand the heaviest travel could not be transferred from
the plant to the street without doing so, and under a hot summer
sun would soon become a monolithic mass in the street, even if
it were possible to place them there before they were so much
out of shape as to make it impossible to lay them. One of the in-
herent defects in asphalt blocks, at least as they have been laid
up to the present time, is that the cementing material which binds
them together is too hard to enable the block to resist heavy travel,
and they have gone to pieces, especially in cold and wet weather,
after being exposed to it for but short periods of time. In ad-
dition, the aggregate of an asphalt block is not a satisfactory one
as far as grading is concerned. It consists to-day of crushed

ASPHALT BLOCKS.

391

trap-rock or similar hard stone, which will pass a screen having
openings f inch in diameter. Whether these screenings are made
particularly for the purpose with steel rolls or are the finer
portions of stone which has been crushed for concrete, they al-
ways contain an excess of particles of a size which will pass a
screen of ten meshes to the linear inch, and be retained on one of
twenty meshes. The screening of such material does not vary
largely from year to year, as can be seen from the following figures:

Year.

1906

1907

Passing 200-mesh screen

" 100 " "

' 80 " "

" 50 " "

u 40 " "

" 30 " "

" 20 " "

" 10 " "

" t ^-inch mesh screen

100

In 1906 there was a considerably larger percentage of grit
than in 1907, but otherwise the grading did not differ essentially.
If asphalt blocks are, however, regarded as being composed of
small particles of stone or grit, the voids in which are filled with
a standard mixture such as had been found most desirable for
sheet asphalt surfaces, the portion of the crushed trap-rock which
plays the role of sand, it will be seen on calculation, has a grading
which is entirely "unsuited for the purpose for which it is used,
and one which would never be employed in the sheet asphalt
industry. There is no way of modifying this grading except
by rejecting the excess of 10-mesh material, and even then there
remains a deficiency in the grains passing sieves of 80- and 100-
meshes, while the expense involved makes it quite impossible
to attempt such a modification. The crushed rock must be used
just as it comes from the rolls or from the crusher, and the grading
is uniformly bad, as can be seen from the analyses of blocks made

392 THE MODERN ASPHALT PAVEMENT.

by various firms and corporations. There would be no possi-
bility of obtaining good service from a sheet asphalt surface with
a mineral aggregate of such grading, and it is not surprising,
therefore, that asphalt blocks do not wear when subjected to
heavy travel, especially as most blocks are made, as has been
said, with an asphalt cement, which is far too hard and which, if
used in a sheet asphalt pavement, would result in scaling of the
surface during damp, cold weather.

The question at once arises as to what opportunities are possible
for modifying and improving the character of an asphalt block.
As a modification of the grading of the crushed trap-rock is im-
practicable for economic reasons, if a more satisfactory mineral
aggregate is to be obtained, recourse must be had to materials
found in nature which can be combined to give the desired grad-
ing, such as a well graded sand or clean trap-rock grit of proper
size or gravel. There would be no difficulty in obtaining proper
sands, such as are used in the sheet asphalt industry, but suitable
gravel would be more difficult to find. An opportunity, however,
is afforded for improving somewhat the character of the block
in another direction, by the use of a more suitable asphalt cement
than that which has been in use hitherto, one which is not so
susceptible to extremes of temperature. With such a material,
a consistency which would be suitable for making a block which
would not be so brittle at low temperatures could be used, and,
at the same time, the block would not be so soft under a summer
sun as to lose its shape on storage without lateral support or
under pressure before being placed in the street. Such a cement
is found in one prepared from Gilsonite and an asphaltic flux. It
can be used of a much softer consistency than one made with
Trinidad or Bermudez asphalt without danger of producing a
block which will not retain its shape. This can be seen by a com-
parison of the consistency of the three cements, as they are used in
making blocks, at different tejnperatures.

Penetration-millimeters.
32 F. 77 F. 110F.

Trinidad asphalt 2 1.4 6.8

Bermudez asphalt 2 1.3 7.3

Gilsonite 5 2.6 7.5

ASPHALT EfLOCKS.

393

r> Q X

cSS 10

* co

<n <N

<J -<

3 * 2

8 -S x

f, J

.s x

*^5

8 - iM S

ao (N

.. bC nH

^ .s x

S -s

QO X

HH CO

Ttoo Oi

| x
8 fi x

^ ^

^5 CO

^ ;
w ^!

-g X

I x

'HCOO*OO5

O5O> s -
^J* 4 s * OJ

<N(N 00

t t/.

If.

53 -

I s

to
o

1 :

fa J

00 oT

l>- C

* 3

1 >
f 2
g oo

5 3

1 1

I I S

394 THE MODERN ASPHALT PAVEMENT.

Under any circumstances, asphalt blocks which are suitable
to resist heavy traffic will be sufficiently soft to become, in summer,
almost a monolothic surface under travel.

The character of asphalt blocks is usually determined both
by chemical and physical tests, the latter being the more important
and including determinations of density, modulus of rupture,
and loss from attrition in a rattler of the type used in testing
paving brick. The results of such tests on various blocks produced
by the most prominent manufacturers are presented in the ac-
companying table,

SUMMARY.

Asphalt blocks are very satisfactory under certain circum-
stances as a form of pavement for residence or light traffic streets,
but experience has shown that such a pavement is unable to
wear for any length of time under heavy traffic.

CHAPTER XIX.

THE PROCESS OF COMBINING THE CONSTITUENTS INTO A
SURFACE MIXTURE.

ASPHALT cement of a desirable nature, a sand or sands which
will afford a satisfactory grading, and a sufficiently finely divided
mineral matter for a filler being available, it is necessary that
these materials should be combined with great care, skill, and
uniformity in order to produce a surface mixture which shall be
free from criticism.

To bring about this combination some type of plant is necessary
which shall make it possible to meet the following conditions:

1. To feed a sand or mixture of sands into the sand heater
with great regularity and to have it pass through the drum in
such a way that it is uniformly heated and the particles not segre-
gated according to size.

2. To raise the heated sand to a temperature of from 330
to 380 F. as it emerges from the heater, without reducing its tem-
perature essentially, pass it through a sieve which shall remove
all particles larger than those passing a laboratory screen of
10 meshes to the inch, and collect it hi some form of bin where
it can be held for some time without too much radiation and
from which it can be drawn without delivering at one time a finer
and at another tune a coarser material.

3. To have a melting-tank where asphalt cement can be main-
tained in a melted condition and at a uniform temperature and
provided with suitable means of agitation, either air or steam.

4. To have suitable provisions for determining accurately the
weight or volume of the constituents entering into the compo-
sition of the surface mixture.

395

396 THE MODERN ASPHALT PAVEMENT.

5. To have a mixer which shall make an entirely uniform and
homogeneous combination of all the constituents which go into it.

6. With a satisfactory plant it is equally necessary to have
a foreman to run it who not only understands how to make every-
thing move uniformly but who has had experience in and under-
stands the technology of the industry and the reasons for each
step that is taken.

These necessities may now be considered in greater detail.

Sand. To obtain a sand of satisfactory grading it has already
been shown that it is usually necessary to mix two or more sands.
To obtain a uniform mixture from sands which will give a proper
grading requires care and attention and proper facilities for storing
the sand conveniently in the rear of the sand-drums. The two
or more sands are then wheeled up or shovelled in separate piles
in the neighborhood of the bucket elevators, which are to elevate
it to the drums. The sands are then fed into the buckets with a
shovel or hoe by laborers in such proportions as may be found
by experiment to be necessary. This feeding must be carefully
watched, as, owing to the class of labor employed, little depend-
ence can be placed upon the laborer himself. If the feeding is
irregular the surface mixture will also be irregular. The regu-
larity of the feeding is determined on the platform by sifting.

To obtain uniformity in the temperature of the sand the type
of sand-heater must be a satisfactory one and the firing must
be as carefully done as with a steam-boiler. Only experienced
firemen should be employed and they should be instructed to
watch the temperature of the effluent sand closely.

The hot sand falls from the drums into a boot, from which it
is raised to the sand -screen over the sand-bin. This elevator
should be well closed in to prevent radiation and the screen and
bin should also be enclosed.

The screen should be cylindrical or conical in shape; in the
latter case, 3 feet in diameter at one end and 20 inches at the
other. It should revolve about 12 revolutions per minute. At
the end of the larger diameter it is covered for half of its length
with cloth of 10 meshes to the inch, No. 22 wire. The remainder
should be 8 meshes of No. 18 wire, the entire length being 5 to 6 feet.

THE PROCESS OF COMBINATION. 397

Angle-irons placed lengthwise of the screen at each quadrant
will strengthen it and increase its capacity by throwing the sand
about.

The sand should not be allowed to fall into the bin irregu-
larly and from whatever point it passes the screen. It should
all fall into a hopper which opens over the centre of the bin. This
is quite necessary to prevent segregation, as otherwise the fine
sand would pass the screen first and go to one side of the bin,
the coarser particles collecting at the other.

' At best a certain segregation results on drawing sand from
the bin. Unless the bin is kept more than half full there is a
tendency to form a hollow cone in the centre of the mass of sand,
down the sides of which the coarse particles run and accumulate,
so that every now and then there is a delivery of coarse and again
a delivery of fine sand. This can only be avoided by not draw-
ing the bin down too low.

Various shapes of bin have been suggested, but a cylinder
and cone or a half cylinder and half cone are probably the best.
The gate should be in the bottom of the cone and not in the side.

Segregation is, however, apt to take place in -any form of bin,
and that form which prevents this to the greatest extent is the
most desirable.

The temperature at which it is necessary to maintain the sand
in order to produce a satisfactory mixture will depend on the
character of the mixture that is being turned out, the nature
of the asphalt in use, the weather, and the distance to the street
from the plant. If the mixture is a fine one, carrying much filler
and bitumen, if Trinidad asphalt is the cementing material, and
if the weather is cool, 385 F. is not too high a temperature for
the sand in the bin. If a smaller amount of filler is employed
the temperature need not exceed 335. In any case the mixture
should reach the street at such a temperature that it can be raked
freely. In the best New York mixtures the temperatures aver-
age 330 F. The average mixture of the country will reach the
street at 310. The above applies to a standard Trinidad mix-
ture. If other asphalts are used the temperatures must be con-
siderably reduced, as they will suffer from such heat. The harden-

398 THE MODERN ASPHALT PAVEMENT.

ing effect of hot sand on asphalt cement has already been noted,
and should always be allowed for in those mixtures made with
susceptible bitumens such as Bermudez and the hard residues
from petroleums or where the flux in use is one carrying much
volatile oil.

Melting-tanks. The melting-tanks in which the asphalt cement
is made at the smaller plants or into which it is drawn from the
refining-tanks at the larger ones should be so constructed that
no portion of their contents shall become readily overheated.
The bottoms should be protected from direct flame by a fire-
brick arch. Agitation is necessary, not only with such an asphalt
as Trinidad, for the purpose of keeping the mineral matter in sus-
pension, but with others as well, to keep the material from too
long contact with the sides of the tank, and of even temperature,
since convection in melted asphalt results in but a very slow
motion of the mass.

Agitation with steam is undoubtedly the best method, as the
action of air on all oils at a high temperature is very strong, the
result of blowing an asphaltic residuum at a temperature of 350
with air for twenty-four hours being to convert it into a semi-
solid buttery mass.

Agitation is also a matter of economy as far as the life of the
tanks themselves are concerned, as they will burn out very rapidly
if sediment or coke is allowed to collect in them.

In plants where a large amount of work is done some pro-
vision in the form of a pneumatic lift should be made for raising
the cement to the asphalt bucket, where it is gauged or weighed
without the aid of manual labor.

The requisite amount of asphalt should be weighed and not
measured and the same may be said of the sand. The same volume
of sand may vary very much in weight according to the way it
runs into the receptacle.

The type of mixer in use in combining the constituents is,
of course, of importance and still more so the way in which it is
kept in repair and good order. It is usually constructed to mix
a volume of 9 cubic feet at one operation, but as large a volume as
18 can be thoroughly mixed in a properly constructed mill and

THE PROCESS OF COMBINATION. 399

corresponding economy attained. The mixer should be pro-
vided with a liner which can be renewed when worn. It should
be provided with a set of teeth made of chilled iron or having
steel tips which reach within a quarter of an inch of the lining.
The teeth should be set on a shaft the bearings of which can be
raised or lowered by the introduction or removal of shims so as
to bring the teeth nearer or farther away from the liner. It should
have a gate which is tight and will prevent the leaking of either
asphalt or sand. In the larger types of mixer the gate is controlled
by some power appliance.

The mixer rests on what is technically known as the platform,
which is sufficiently elevated to admit a truck beneath. Behind
the mixer is the sand-box in which the sand and filler are weighed
or measured out. Over it is the bucket for the asphalt cement
suspended from a scale. The sand and filler having been weighed
in the box and the A. C., the technical designation of the asphalt
cement, in the bucket, the two former are allowed to run out
through a gate into the covered mixer, which is, of course, in motion,
or if the sand-box is one that has no gate, it is dumped into the
mixer. It is allowed to remain there for from 15 to 20 seconds to
mix the dry materials thoroughly. The asphalt cement is then
poured or run directly into the middle of the dry mix and not spread
about over different parts of it, as the mixer teeth will bring all the
sand to the centre to meet the bitumen, but will not be able to do so
as readily with the latter. After the introduction of the asphalt
cement the mixing is continued for about one hundred revolutions.
The gate to the mixer is then opened and the mixture dropped
into the truck.

Where the platform is large enough a testing-room should be
provided there for the use of the yard foreman; otherwise it
should be upon the ground in front of the mixer, as in the case of
a railroad plant. He should have there a set of sand-screens,
a sand-balance, a flow outfit for controlling the consistency of his
A. C. and manilla paper for making pat tests of his mixture. He
should screen samples of sand taken from the boot of the sand-
drums and from the bin in order to be sure that the sand mixture
is being fed in the proper proportions and evenly. He should

400 THE MODERN ASPHALT PAVEMENT.

make comparative flow tests of the asphalt cement he has in use
with that of the standard furnished him for the purpose. Finally,
he should make frequent pats of his mixture to determine whether
it is carrying a proper amount of A. C. and whether it is properly
balanced. These points are recognized by the stain made by
the bitumen on the paper and by the appearance of the surface
of the hot pat when it is held on a level with the eye. Experi-
ence is, of course, required to interpret these latter tests and to
understand the indications which are afforded. In addition
such samples should be taken on the platform as are needed for
examination in the laboratory by more accurate methods.

The Production of Binder. Binder is turned out in exactly
the same way as the surface mixture except that the tempera-
ture and not the grading of the stone is to be watched and the
mixing is done in a mixer having fewer and shorter teeth. The
temperature of the binder can be appreciably lower than that
of the surface. It should certainly not be so hot as to cause the
asphalt to run off the stone, as in that case it will reach the street
without sufficient cementing material to hold it together, a result
often noticed in careless work.

Types of Plants and Machinery. In the preceding pages no
mention has been made of any particular type of plant or machinery,
as this seemed unnecessary. It is the results obtained and the
way in which they are attained which are of the greatest interest
to the engineer and to the private individual to whom this book
is addressed. One type may do slightly more economical work
than another or slightly better. If the latter is true it can be
better judged from the character of the material turned out or
from the celerity with which the work is accomplished.

It will be of interest here, however, to describe the type of
the machinery which the author has found most satisfactory.

Permanent Plants. In large cities where a very considerable
amount of work is done every year a permanent plant will, of course,
be established, consisting of proper storage, for two or more
grades of sand, in the shape of bins, which can be readily and
economically filled from the boats, cars, or other means of trans-
portation which supply the sand. This material should natu-

THE PROCESS OF COMBINATION. 401

rally be handled, for economy's sake, as far as possible by power
and the bins should be so placed in relation to the sand-heaters
that as little labor as possible will be necessary to feed the sand
to the heaters.

Sand-heaters. The sand used in different localities varies
from dry bank to dripping-wet river sand, and the heater is required
to drive off the moisture and heat to a temperature of 330 F.
or over the maximum amount of sand per unit of time with the
minimum amount of fuel. As the result of actual trial of many
different designs the Iroquois Iron Works has adopted a setting
of two horizontal revolving drums, fired with induced draft, as
giving the greatest efficiency. By the proportioning of the size
of the drum, furnace, and induced draft, all the available heating
power is obtained, and by heavy construction and perfection of
details the durability of the machine is assured.

The drum-shells are made of j" steel plate 20 feet long rolled
to a 40" diameter circle, and are riveted together with two hori-
zontal butt-strap joints. These shells are carried at both ends
by heavy cast-iron spiders. Shafts from these spiders extend
out beyond the shells and form the journals. The bearing at the
hot sand end takes the thrust, being grooved similar to a propeller
shaft-bearing. The bearing boxes are fitted with trunnions allow-
ing swinging in a vertical plane. The trunnions rest on swivel
brackets, permitting swinging in a horizontal plane; consequently
there can be no binding in the boxes, they being able to align
themselves at all tunes, a necessary qualification for a drum revolv-
ing in a furnace. Sheet-steel shelves are riveted the entire length
of the ulterior, which give additional heating surface and at the
same time are continually lifting the sand and allowing it to fall
through the diameter of the drum. The grates are located directly
under the cold sand end of the drum-shells.

By means of a fan the combustion gases are drawn along under
and back through the drums, coming in contact with the sand,
which, by the shelves on the interior of the drums, is distributed
through them. By this method the surfaces of the drum are
heated by direct radiation from the gases of combustion, and the
gases, being drawn through the drum and coming in intimate

402 THE MODERN ASPHALT PAVEMENT.

contact with the particles of sand, not only draw off the released
moisture and discharge it outside the setting but also assist in
heating the sand to the required temperature. This induced
draft is a valuable feature and, as it is easily regulated, increases
the capacity and ensures obtaining the maximum calorific value
from the fuel. Such a sand-heater is illustrated in Fig. 17.

These drying cylinders are mounted in a brick or steel-plate
setting, as desired. For permanent installation the brick setting
is preferred. The steel-plate setting is well shown in the illustra-
tion, Fig. 17, from which it will be seen that the bearing boxes
supporting the cylinders are mounted on a built-up framework
at both hot and cold sand ends. This framework consists of a
base of riveted channels and steel plate, upon which is mounted
cast-iron brackets for carrying the bearing boxes. The sides are
made of J" steel plate reinforced with angles. The roof consists of
two steel plates. The inner, which is \" thick, is curved to con-
form to the arc of the drums, thus holding the heat against the
latter. The outer covering of medium gauge sheet steel is carried
straight across to form a rain-shed. The inner and outer covers
are riveted to a trussed angle-iron frame which is absolutely self-
supporting. The sides, ends, and roof are made in sections and
bolted together, permitting the entire setting to be dismantled
into small units which are easily handled and packed for ship-
ment. For the protection of the steel plate it is customary to
lay up a lining of one thickness of fire-brick at the sides extending
two-thirds the length of the setting. The exhaust-fan is mounted
on a timber frame to one side and can discharge into the air or
be connected with a dust-collector, as may be preferred.

The drums discharge into a boot, from whence the hot sand
is elevated by means of steel buckets on a steel pin chain to a
rotary screen covered with cloth of the dimensions which have
been described. 1 The screen discharges into a steel storage-bin
of from 6 to 9 cubic yards capacity. From the storage-bin it
flows by gravity to a triangular-shaped weighing-box mounted
on a beam-scale. Here the required amount of stone dust is

1 See page 396,

=fc "3

404 THE MODERN ASPHALT PAVEMENT.

added and the charge brought up to accurate weight ready for
the mixture.

Melting-tanks. For melting the asphalt two types of kettle
are used, fire and steam. The fire melting-tanks are either cylin-
drical or rectangular, with semicircular bottom, and are set in
the furnace with a fire-brick arch between the grates and the bot-
tom of the tank to prevent too rapid heating, which would tend
to coke the material on the bottom. The fire melting-tanks are
now more generally used for small portable plants in small units
of 4 tons capacity. For larger and more permanent plants the
steam melting-tanks are preferable.

The steam melting-tanks are rectangular in shape, contain-
ing horizontal oval-shaped coils of 1J" pipe. Fig. 18. One
hundred and twenty-five pounds steam pressure is generally
used, which gives a temperature in the coil of 345 F. Agitation
is accomplished in both fire and steam melting-tanks by hori-
zontal pipes laid at the bottom, with small perforations. To
these pipes are deliverd either steam at boiler pressure or air
at about 20 pounds pressure. When the asphalt is melted in
these kettles it is reduced with the required amount of warm
flux, which is measured in a special measuring-tank and flows
by gravity to the melting-kettles. The asphalt cement result-
ing is now ready for the mixer and is delivered to it in one of three
ways : In small semi-portable plants a bucket carried by a traveller
mounted on wheels on a track is run out from the mixer over
the melting-kettles and the cement dipped into the bucket. In
many permanent plants the melting-kettles are mounted on a
structure sufficiently high for the asphalt cement to flow by gravity
directly into the bucket. The third and very largely used
method is setting a pneumatic lift just below the bottom of the
kettles. This pneumatic lift consists of a steam-jacketed steel
cylinder fitted with inlet- and discharge-pipe, air-pipe, and a system
of valves whereby the cement flows into the lift from the kettles
by gravity, and by the operation of suitable air- valves, air-pres-
sure, which need not be over 5 pounds and is generally the same
as the agitation pressure, is admitted on top of the cement, thereby
automatically closing the valve in the intake and forcing the

FIG, 18. Steam Melting Tank.

405

406 THE MODERN ASPHALT PAVEMENT.

cement up through the discharge-pipe to the weighing-bucket
at the mixer.

Mixer. The mixer consists of a rectangular-shaped steel shell
with semicircular bottom, containing two horizontal square
shafts, upon which shafts are bolted blades, or teeth, as they are
commonly called. Fig. 20. These shafts are made to revolve
by gearing at a speed of from 60 to 75 revolutions. The teeth
are made in two different shapes, called right and left hand, and
are so set upon the shafts that they work the material horizontally
towards the centre, and at the same time are continually tossing
it vertically. The result is that an absolutely homogeneous
mixture of the sand and asphalt cement is obtained within a
minute and a half, when the mixer-man pulls a lever, opening
the slide in the bottom, and the finished topping is discharged
into the wagon ready for hauling to the street.

Where a plant has but one mixer for turning out both sur-
face mixture and binder it is provided with another shaft for
carrying shorter teeth at much wider intervals. This shaft re-
places the one used for surface mixture when binder is to be pro-
duced.

Portable or Semi-portable Plants. In cities where work is
only done at intervals and where the amount is not sufficiently
great to justify the construction of a permanent plant, portable
or semi-portable plants are used.

The first type of portable plant consisted of two railroad flat
cars, one to carry the boiler and melting tanks and the other for the
sand heaters, with the mixing platform between the cars when
the plant was set up for use. Such a plant is illustrated in Fig.
21. The cost of setting up and taking down this form of
plant when moving from place to place was rather large. The
plant has therefore been modified to occupy a single car, render-
ing the cost of putting it in operation extremely small. A plant
of this type, as manufactured by the Iroquois Iron Works, is
shown in Fig. 19.

The semi-portable plant is one which is readily taken down
and erected, but is not fixed upon a car. It consists of a steel
tower with mixer platform, as illustrated in Fig. 22, which is
accompanied by the necessary melting-tanks, usually heated

THE PROCESS OF COMBINATION.

407

by fire. This type of plant has been found very successful in

late years and has a large future before it for work in small towns.

Plants of the two previous types require skilled handling and

close attention, but with a good foreman and engineer equally

FIG. 19.

good work can be turned out from them as from a permanent plant,
and they are highly recommended by the author.

Bituminous Concrete Plant. For the production of bitu-
minous concrete any of the previous types of plant will serve, if
the screen over which the heated material passes is arranged so
as to separate the heated mineral matter into three sizes, sand,
fine stone passing openings f inch in diameter, and coarse stone
up to f inch in size, and if the bin is divided into compartments
to hold these separate grades. A special form of plant is not
required for this purpose, nor is it necessary to separate the stone
into more than three sizes, as has been shown by the successful
work done in this way as long ago as 1896.

FIG. 22.- Semi-portable Mixing Platform

410

THE PROCESS OF COMBINATION. 411

SUMMARY.

This chapter describes the process of combining the con-
stituents into a surface mixture, including the machinery and
plant necessary for heating the sand and mixing the hot min-
eral aggregate with asphalt and the type of tanks necessary for
melting the latter.

PART V.

HANDLING OF BINDER AND SURFACE MIXTURE
ON THE STREET.

CHAPTER XX.
THE STREET.

Transportation of the Materials to the Street. The transpor-
tation of the binder and surface mixture from the plant to the
point where the pavement is being constructed is something that
cannot be undertaken carelessly and with no other thought than
merely getting it there. In the case of the binder the only con-
sideration is that it be so protected that it will not become cold.
The condition of the surface mixture when it reaches the street
is much more influenced by the conditions to which it has been
subjected en route. In the early days of the industry, in Washing-
ton, D. C., for instance, the old-fashioned dump-cart, holding
from 18 to 27 cubic feet, was in use. Aside from a matter of
economy, this is the ideal way to haul the material. Later on,
as the size of the mixer was increased in the larger cities, trucks
were employed which would hold six batches of 18 cubic feet,
or six tons. It was soon found that this method of transporta-
tion was unsatisfactory, as the larger mass of surface mixture,
during the long hauls which are unavoidable in cities of the size
of New York and the constant jarring over rough pavements,
became so compacted that it was difficult to break it up after
it was dumped on the street, especially if the mixture was a dense
one carrying a large percentage of filler and asphalt cement. To

412

THE STREET. 413

offset this disadvantage, however, the larger mass loses heat much
less rapidly than is the case with the smaller load, and this is a
distinct gain in cold weather and for repair work. A medium
course is, therefore, now pursued and loads of about four tons
are hauled.

In the smaller cities and towns the question is often a serious
one, as the trucks available locallyare often not suitable for hauling
surface mixture. The worst type is the ordinary dirt truck which
dumps by turning over slats which form the bottom of the truck.
This truck does not protect the mixture from cooling rapidly, and
in dumping the entire mass is so loosened up as to be still further
cooled, while much of the material is lost by being carried away
on the running-gear.

Whatever the form which the load may take, the surface should
be protected from the air, at all seasons of the year, by tarpaulins.
The loss hi temperature, if the protection is suitable, will not
exceed 10 from plant to street hi two hours, or often after longer
intervals, in warm weather.

It has been found possible to transport large masses of hot
surface mixture by rail or by scow for long distances. As an example
of this may be cited work done in New Rochelle, N. Y., in 1899.
Three hundred tons of mixture w ere placed on a scow at Long
Island City, N. Y., and taken by a tug to the point where the sur-
face was to be laid. Owing to the inclement weather it was impos-
sible to place the material for thirty hours, but the majority of
it was in good condition to be laid at that time after the outer
cooled portions had been removed. The asphalt pavement laid
in this way has been entirely satisfactory.

Construction Work on the Street. Of the work of construction
of an asphalt pavement on the street consideration need be given
only to that portion above the base, that of the latter involving
no principles which have not already been exploited. In earlier
pages the desirable characteristics of a base have been shown,
and it is here assumed that the bituminous surface is to be applied
to such a base.

The Binder Course. In general a binder course is the first
applied. In the description of the preparation of the binder

414 THE MODERN ASPHALT PAVEMENT.

it appeared that it was sent to the street at a temperature some-
what lower than that of the surface mixture. Arriving there it
should be dumped sufficiently distant from the point where the
spreading is to be begun or from the point where the previous
load ended to permit of turning all the material over without
finding it necessary to finally distribute any of the binder over
the base at the point where it has been dumped. This is quite
necessary, although not as much so as in the case of surface mixture,
to permit of spreading the binder course evenly, that portion
lying at the point where the load was dumped being consider-
ably compressed by the fall and the weight of the incumbent
mass, so that were this not shovelled over the thickness at this
point would be greater than elsewhere in the street.

The surface of the load of binder should be bright and glossy,
as should the whole mass after it has been dumped. On the
other hand, there should be no excess of bitumen, as evidenced
by asphalt running from the bottom of the truck or by too great
richness of the bottom of the load. Too hot stone may cause
the bitumen to run off the binder. One should not be misled
by such an occurrence into the belief that the load is too rich.
In such a case, however, the surface of the load will generally
be dead. Unless the stone is covered with a bright coat of
bitumen the binder will have no coherence and should be re-
jected.

The binder is spread with rakes with long tines. It may be
allowed to cool to a very considerable degree before rolling. If
rolled too hot it will be much more liable to displacement and
to being picked up by the roller. It should be rolled directly
with a steam-roller weighing from five to seven tons.

Immediately after rolling it is ready for the application of
the surface. If the surface is not applied at once the binder
should be protected from becoming soiled by traffic or otherwise.
A slight coating of dust will do no harm. The hauling of a
sufficient amount of surface mixture over it to cover it should not
break it up. If this happens, except on a weak base, it is a sign
that it is not of the best quality.

Too often the thickness of binder specified is too small, and

THE STREET. 415

in this case it is impossible to lay it so that it will not break up
to a certain degree in putting on the surface. An inch of binder
is never satisfactory. Binder is composed of stone the larger
particles of which are at least an inch in diameter. It is readily
seen that no satisfactory bond of such materials can be obtained
in such a thickness.

Where an asphaltic concrete course is substituted for an open
binder this must be spread with shovels and the back of the rake.
The tines should not be used at all, since they have a tendency
to pull the larger stones to the surface and cause a segregation of
the material.

The Surface Course. The mixture which is to form the sur-
face should arrive upon the street at a temperature which cannot
be denned in degrees of the thermometer. It should be hot enough
to work freely under the rake if it has a properly balanced mineral
aggregate, but in no case should exceed in temperature one which
the particular asphalt cement can resist without being too much
hardened, especially in the mixer when being violently agitated
with hot sand in contact with ah*. As different asphalts are very
variable in respect to their volatility and stability, the extreme
temperature to which mixtures made with them may be heated
is quite different. This has already been shown on previous
pages. The temperature will also vary with the character of
the mineral aggregate. A dense mixture may be heated much
hotter without injury than an open one. As a general rule, it
may be laid down that some dense Trinidad mixtures, such as
that made with a Portland-cement filler, may with safety be raised
to a temperature of 340 to 350, if it is necessary, in cold weather.
By this it is not meant that such a temperature is desirable if
the mixture can be worked at a lower one, but that no danger
will be incurred by its use which is commensurate with the dis-
advantages arising from inability to handle a cold mixture on
the street and consequent poor workmanship.

A Bermudez mixture hardens rapidly at temperatures above
300 and should not be heated above that point unless provision
is made for the resulting hardening by making the asphalt cement
about ten points too soft. The same may be said of those asphalts

416 THE MODERN ASPHALT PAVEMENT.

which resemble Bermudez, Mexican, western Venezuelan, and
the like. The best oil asphalts will resist high temperatures well,
but mixtures made with them do not require to be very hot, as
bitumen of this character is so liquid at a comparatively low heat
that no difficulty is experienced in working them, even the densest,
at 280.

As a general rule, it may be laid down that the following tem-
peratures may be considered normal on the street:

Trinidad asphalt:

Dense mixture 325 F. to 340 F.

Average " 300 F. " 325 F.

Open " 280 F. " 300 F.

Bermudez asphalt:

All mixtures 280 F. " 300 F.

The lowest temperature at which a mixture may reach the
street and still be considered satisfactory is that at which it may
be raked to a proper grade without too much difficulty.

The character of a mixture can be judged, to a very considera-
ble degree, by the appearance of its surface in the truck as it
comes upon the street, if the haul has extended for any distance,
and by its cohesion when it is dumped. The best mixture, carry-
ing plenty of filler, should have, before dumping, a nearly level
and rather bright surface. If the material stands up in a heap
as it was dropped from the mixer it is not rich enough. If it
tumbles to pieces on dumping, it does not contain enough filler.
The best mixtures, which are the only ones suitable for heavy-
traffic streets, should stand up and show in part the shape of
the truck from which they have been dumped.

This applies, however, only to the natural asphalts. The
residual asphalts from asphaltic petroleums become so liquid
at temperatures at which surface mixtures are handled that the
latter are quite sloppy.

A load of surface mixture, for the same reasons as in the case
of binder, only more emphatically so, should be dumped upon
the binder so far from the material previously raked out that
it will be possible and necessary to shovel it all over in order to
get it in place. This is most important, and care in this direction
is often lacking. If the mixture at the point where the load is

THE STREET. 417

dumped is not shovelled over, but merely brought to grade before
rolling, there will be an excess of mixture at that point which will
not compress as much under the steam-roller as the neighboring
surface, with the result that after the street has been subjected
to traffic for some time that part is higher than the rest.

The surface mixture is distributed with hot shovels from the
point where it has been dumped to the place where it is to be
raked out to the proper thickness for compression. This opera-
tion should not be conducted too rapidly. No more should be
spread than the rakers can handle. If it is spread too rapidly
the rakers will find it necessary to step in it in correcting inequali-
ties of grade, thus compressing the part where their feet fall. This
depression they afterward fill with more mixture and therefore
leave at that point more than there should be. After the street
is opened for traffic this part of the surface does not yield as much
to final compression as the remainder and the result is an uneven
and bumpy grade which is accentuated with the lapse of time
and aids in the disintegration of the pavement from the blows
of wheels bounding from the elevated spot to the adjoining sur-
face. Rakers should on no account be allowed to place their
feet on the uncompressed surface mixture. If absolute necessity
arise the depression should not be refilled.

The mixture after it is spread should be thoroughly raked
out with rakes having long and strong tines which will penetrate
through its entire depth. It is necessary, in order to obtain a
regular surface to the finished street, that all the hot mixture
shall be broken up to a uniform state of looseness. None of the
material in the state of compression which it has acquired in the
truck during the haul to the street should be allowed to remain
in lumpy form. If lumps remain in the loose hot material the
effect will be the same as that occurring at the points where the
rakers place their feet. It is insufficient that the actual surface
of the loose hot mixture should represent a uniform thickness; it
must also be of uniform consistency all the way through.

The lack of perfect form in asphalt surfaces is oftener due
to this cause than any other, but it is, of course, much empha-
sized on streets of heavy travel where traffic depresses that part

418 THE MODERN ASPHALT PAVEMENT.

containing the least material. With coarse mixtures, and those
deficient in filler, which do not become so much compressed in
the trucks, and with mixtures poor in bitumen, results such as
have been described are not so apt to occur or are brought out
less under lighter traffic, and, as a matter of fact, with such mix-
tures it is possible to give a street surface a much prettier original
finish than can often be obtained with standard mixture which
carries a high percentage of fine sand, filler and bitumen.

The raking of the material to a proper grade requires a good
eye on the part of the workman and proper supervision and a bet-
ter eye for such work on the part of the foreman. Constant atten-
tion and great care are, however, the great desiderata. It is
not difficult to make a good raker out of an inexperienced man
if he is under good supervision. The great difficulty with all
rakers is to make them pull out all their material to a loose con-
dition, especially with a dense mixture such as it is necessary to
lay on heavily travelled streets.

The hot mixture having been raked to grade, it was the cus-
tom, in the early days of the industry when the mixture was more
loose and open and carried less filler and asphalt cement and
consequently had less density, to give it its first compression with
a hand-roller of comparatively light weight. This may be advis-
able even to-day with similar mixtures. As a rule, however,
the modern mixture has sufficient density to permit the use of a
steam-roller at once, and this is the general custom in good prac-
tice. The hot mixture is, however, allowed to cool to a point
where it will not be displaced or picked up. To avoid the latter
difficulty it may be necessary with some mixtures to oil the roller
with a mixture of kerosene and water, and it is generally found
to be preferable to run the lighter or guide rolls of the roller on
the surface first. After the preliminary compression the surface
is sprinkled with any fine mineral matter which will give it a
color pleasing to the eye. It is not necessary that this should
be a hydraulic cement. The excess of dust having been swept off,
the surface is allowed to cool still further until the roller can go
on and shape it up without displacing it. Experience can alone
determine what length of time to allow for cooling. In winter

THE STREET. 419

it cannot be long, since if a hardened crust is allowed to form by
the chilling of that portion of the mixture exposed to the air, this
will have a tendency to break up on further rolling, and fail to
make a bond with the main mass, resulting in subsequent scaling.

The aspect of the finished pavement, especially after it has
been subjected to traffic for a year or two, will depend as much
on the way it was rolled as on the way the mixture has been raked
out. The management of a steam-roller requires experience,
skill, and judgment. The roller if not run with great regularity
and stopped with care at the end of a run will readily displace
the surface so that it cannot be easily brought back into form
again. The first rolling should be with the length of the street.
It should then be rolled diagonally where this is possible as soon
as it is evident to the roller engineer that nothing is being accom-
plished in the original direction. No rule can be laid down as
to the length of time necessary for rolling a given area of surface.
The time will depend very much on the season of the year and
more on the character of the mixture. The hot surface mixture
will cool more rapidly in the autumn, and cannot be rolled for
the same length of time as in summer, or at least with any effect.
Mushy mixtures, where the local sands make mixtures of this
description, should not be rolled too much. This would injure
them by breaking the bond in the cool mixture. Mixtures on a
weak base cannot be rolled to the same extent as those on one
that is firm. Certain mixtures which are readily displaced may
require a final shaping up with a roller of wider tread than that
used for the original compression, one of ten or eleven tons weight
and tread. This is by no means always necessary if the roller
engineer is skilful and the mixture a good one; although the weight
per inch tread is greater hi one case than in the other. Follow-
ing are some determinations of the pressure per inch run for
various rollers. See table on page 420.

All rollers are not equally suited for the work. Some are
strikingly defective in that they are not well balanced. The
side carrying the motive power is much heavier than the other,
and the result is that the roller sways, especially when it meets
an elevation or depression, thus producing a wavy surface. The

420

THE MODERN ASPHALT PAVEMENT.
ROLLERS PRESSURE PER INCH RUN.

IROQUOIS IRON WORKS.

Size of roller

2\ tons

5 tons

8 tons

13 tons

Duplex engine .

4X5

6X6

7X7

8X10

Main roll, width of tire. ....
Pressure per linear inch

30 ins.
125 Ibs.

38 ins.
210 Ibs.

48 ins.
250 Ibs.

60 ins.
300 Ibs.

best type of modern roller has a compensating weight attached
to the channel iron on the side opposite to that carrying the motive
power. Whether a roller is properly balanced or not can be deter-
mined by running it on a scantling so that the latter is exactly
in the middle of the rolls and then noting whether it has a ten-
dency to tip toward the side carrying the motive power. If
it does it should be balanced by bolting a weight to the channel
iron on the side which is too light.

Rollers should be provided with a steering-gear which can
be controlled by power, thus enabling the engineer to give his
undivided attention to the street. Such rollers are available.
Throttle-valves should be of a kind which permit the gradual
reduction of speed.

The importance of having a perfect roller, if good work is to
be done, should not be overlooked. That offered by the Iroquois
Iron Works, Buffalo, N. Y., is the best balanced and most care-
fully constructed roller with which the author is acquainted. It
is illustrated in Fig. 23.

These rollers are made of various sizes, 2J, 5, 8, and 13 tons,
the latter being used for rolling the base and for finally shaping
the asphalt surface where mixture requires it after previous com-
pression with a lighter roller. Such shaping is only necessary
when the mixture is of a mushy nature and consequently some-
what displaced by the roller of narrower tread.

Work Under Particular Condition. In the preceding para-
graphs consideration has been given only to what may be called
straight work; that is to say, the laying of an extended area where
everything connected with the work goes on in a perfectly nor-
mal way. This, however, is not the only kind that is met. There

422 THE MODERN ASPHALT PAVEMENT.

are joints to be made, between the work of different days, with
the curb and with headers, around boxes, manholes, and similar
protuberances, and with railway tracks or the brick or stone
stretchers along them. It is also quite possible, owing to unavoid-
able circumstances, that the surface may be found to be, after
its preliminary compression, too high at one point in the gutter
for example or too low at another. Owing to chilling of the
surface it may not close up properly or from inaccessibility to
the roller fail to be sufficiently compressed. These conditions
must be met and the defects remedied, all in their own way, and
in many cases tools especially provided for the purpose, known
as tampers and smoothers, must be used. These tools are shown
with some others in the accompanying illustration, from the cata-
logue of the Iroquois Iron Works of Buffalo, N. Y. Fig. 24.

The tampers and smoothers must be used with great care and
should not be too hot. If the smoothers are hot and it is difficult
to tell their temperature in bright sunshine, they may do much
damage by hardening the bitumen in the surface, and their use
is only advisable in very skilful hands. They are a relic of the
days when mixtures were used which would not close up readily
on account of poor grading and deficiency in bitumen. With
standard mixture they are seldom necessary except on joints.
The tampers are not as dangerous, since they are not left in
contact with the surface as long as the smoothers. They are
used on joints, around manholes, and along rails at points
which the roller cannot reach and in reducing inequalities in the
gutter.

Attempts to reduce projections above the proper grade with
tampers are rarely successful ; the material at the point is merely
more strongly compressed than that in the immediate neighbor-
hood, with the result that the latter goes down under traffic and
the original elevation is brought out again. Especially in gutters,
where the surface is too high, the excess should be removed with
a hot shovel, if necessary, after heating it with a smoothing iron.
Except on joints between days' work, the use of tampers and
smoothers is a makeshift to cover up poor raking, and the best street
foreman will be the man who finds the least necessity for their

424 THE MODERN ASPHALT PAVEMENT.

use, especially that of the smoothers, for reasons which have been
already mentioned.

Joints between different days' work, according to the prefer-
ence of the street foreman, are made in different ways. Usually
the mixture is compressed to a feather-edge under the steam-
roller and left in this condition. On the following morning the
feather-edge is cut back to a point where the full thickness of
surface is shown. This edge is painted with melted asphalt and
the joint between the two days' work is made in this way.

In the middle West an excellent joint is made by imbedding
in the soft material, while still hot and after it has been raked
off to a feather-edge, a rope of about f of an inch in diameter
to which a flap of canvas is attached. The steam-roller is run
over this until final compression is obtained, and on the following
day the rope and canvas are easily detached, leaving an excellent
section to work to without the necessity of cutting back and
losing good material. This form of joint is to be recommended.

SUMMARY.

The preceding chapter describes the method of transporting
and handling on the street the surface mixture, together with
the use of the tools employed in laying it.

PART VI.

THE PHYSICAL PROPERTIES OF ASPHALT SUR-
FACES.

CHAPTER XXI.

RADIATION, EXPANSION, CONTRACTION, AND RESISTANCE TO

IMPACT.

THE physical properties of asphalt pavements are of interest
in two ways: first, from a general point of view as pertaining to
asphalt surfaces as a whole, and, second, the peculiar character-
istics which are dependent on particular mixtures according as
they differ in sand grading, the amount and character of filler
they contain, and the consistency and character of the cementing
material with which they are bonded.

Radiation from Asphalt Pavements. Asphalt pavements have
been frequently criticised because of their great absorption of heat
when exposed to summer suns in an atmosphere of high tem-
perature and its radiation again during the ensuing night. With a
view of determining the number of thermal units of heat thus
absorbed and radiated numerous inquiries have been addressed
to the author as to the specific heat of asphalt. It has been
assumed that the specific heat of asphalt could not be very differ-
ent from that of other native bitumens the records of which are
available. For example, the specific heat of petroleum is given
by Pagliani 1 as .498 to .511. It must be remembered, how-

1 Atti di Torino, 1881, 17, 97.

425

426 THE MODERN ASPHALT PAVEMENT.

ever, that but 10 to 11 per cent of an asphalt surface consists of
bitumen; the remainder is quartz and mineral matter which has a
specific heat no greater than .201. The specific heat of an asphalt-
surface mixture cannot, therefore, be much greater than that of a
granite pavement, or be the cause of any great difference in its
temperature. That the asphalt pavement seems hotter must be
due to other causes, and this is to be attributed to the fact that
having a blacker surface it absorbs heat rather more rapidly than
the granite and radiates it much more rapidly after sunset. There
is, therefore, not much more heat given out by asphalt than by
granite pavement, but since it may be given out more rapidly
it may be more noticeable. Each form will give up about the
same quantity during the entire night.

That the figures assumed for the specific heat of asphalt are
not far out of the way may be seen from the following informa-
tion furnished in " Municipal Engineering " 1 by Mr. A. W. Dow.

SPECIFIC HEAT.

Refined Trinidad asphalt 350

Cuban asphalt cement 401

Trinidad " " 381

Bermudez " " 413

Maracaibo " " 447

Quartz sand 201

Maracaibo and Bermudez asphalts, being nearly pure bitu-
men, afford the best idea of the true specific heat of asphalt.
This factor is somewhat smaller than that for petroleum oil, and
this is evidently due to the presence of more condensed molecules
in the asphalt than in the oil. 2

Some experiments conducted in the summer of 1907 with
cylinders of granite, brick and asphalt surfaces, confirmed the
previous conclusions, as will be seen from the following data :

1 1904, July, 27, 22.

'Additional data in regard to the specific heat of mineral oils will
also be found in "Petroleum," Berlin, 2, 521, April 3, 1907.

RADIATION, EXPANSION, ETC.

427

ABSORPTION AND RADIATION OF HEAT BY VARIOUS
PAVEMENTS.

MAY, 1907.

Time.

Granite.

Brick.

Asphalt.

Thermom.
Temp.

Gain or
Lose.

Thermom.
Temp.

Gain or
Loss.

Thermom.
Temp.

Gain or
Loss.

10.30
11.30
12.30
3.00
4.00

72.0
93.2
102.2
98.6
84.2

+ 21.2
-f 30.2
- 4.6
- 18.0

73.4
93.2
101.3
101.3
86.0

72.0
91.4
100.4
100.4
86.0

+ 19.8
+ 27.9
+ 27.9
- 15.3

+ 19.4

+ 28.4
+ 28.4
- 14.4

JUNE, 1907.

11.05

80.6

78.8

11.35
12.05
2.35
3.05
4.05

88.0
91.5
92.8
87.8
84.2

-f 7.4
+ 10.9
-f 12.2
5.0
- 8.6

90.6
93.2
93.6
89.4
84.2

-1- 10.0
-f 12.6
-1- 13.0
- 4.2
- 9.4

89.6
92.0
91.4

87.8
84.2

+ 10.8
+ 13.2
+ 12.6
- 3.6
- 7.2

JUNE, 1907.

11.15

81 5

83.4

82.4

11.45

89.6

+ 8.1

93.1

+ 9.7

93.1

+ 10.7

12.05

93.0

+ 11.5

96.8

+ 13.4

96.8

+ 14.4

2.05

103.8

+ 22.3

105.8

+ 22.4

105.8

+ 23.4

3.15

93.0

- 10.8

93.2

- 12.6

91.4

-14.4

4.30

86.0

- 17.8

85.2

- 20.6

84.2

- 21.6

Black bulb thermometer in each case 90 F. to 98 F. in sun.
Air temperature, May, 73 F. to 84 F.

June, 81 F. to 92 F. and 81 F. to 93 F.
Clearer atmosphere on second day.

Expansion and Contraction of an Asphalt Surface. Asphalt
surfaces necessarily expand and contract with changes of tem-
perature. As they consist very largely, to the extent of about
89 per cent, of mineral matter, this expansion must be closely
that of quartz, of which the mineral aggregate is principally com-
posed and must be fairly constant for all surfaces. Whether the
cementing material is sufficiently strong to yield to such con-

428 THE MODERN ASPHALT PAVEMENT.

traction without the fracture will determine whether the pave-
ment cracks or not. It is, of course, a feature which will vary
with the character of every mixture, depending upon its com-
position. This subject will be taken up in detail in a succeeding
chapter, where the cause of cracks in pavements is considered. 1

Impact Tests of Asphalt-surface Mixture. The force which
has the greatest tendency to injure an asphalt surface is that of
impact. The blows from horses' hoofs or from the wheels of
vehicles where the surface is irregular deteriorates it to a much
larger extent than attrition or ordinary travel. A well-con-
structed asphalt surface has been known to carry without injury
a load of 51 tons on a truck with broad tires, whereas the same
pavement under constant impact would deteriorate perceptibly
in the course of years.

The capacity of any asphalt surface to resist impact can be
readily tested in the laboratory with appropriate testing
machines, such as that devised by Mr. Logan Waller Page,
of the Office of Public Roads, and described on page 34
of Bulletin No. 79 of the Bureau of Chemistry, and which
is illustrated in Fig. 25. Cylindrical test-pieces are made
from the surface mixture to be examined in a mold which permits
of its being filled with the hot mixture at an appropriate tem-
perature and of being compressed by means of blows with a ham-
mer of four pounds weight, ten of which are given to each end
of the cylinder. These cylinders are usually made 1.25 inches
in diameter and 1 inch high, and weigh about 50 grams. When
brought under the impact machine they are held firmly with a
plunger resting on the surface with a spherical bearing having a
radius of 4/10 of an inch. The hammer is then allowed to drop
from a distance of 1 cm., and this distance is increased by this
amount at each blow until the test-piece yields. Under these
conditions tests have shown that the resistance of any asphalt
surface to impact will depend upon:

1. The sand grading.

2. The character of the sand.

1 See page 480.

RADIATION, EXPANSION, ETC

429

FIG. 25. Instrument for Impact Test.

430

THE MODERN ASPHALT PAVEMENT.

CM CM W <N CMC*

2SS

fafa^'

JrS'S OJ3
J"S 3
* J3 g^^

O OOCOIM
<N (N rH -^

rH O rH rH

rH rH CM rH

' OS -^ OO CO O C^l CO <*!

S= - 2

RADIATION, EXPANSION, ETC.

431

~ --0

O ^

t^. O<NO

06 osos oi

I'l's 1

&, h a

23 2282 2oS o&S&Sog 3o

lO t i

<N O'

O O'

t-OO> i COOOOOO CO -^ I 00 O t>-

-co

i i

o o

ffl PQ

(

s s

432

THE MODERN ASPHALT PAVEMENT.

Og.2g

fill

SCO C100
CO *O W5

OCO
00 t^

CO CO C^l CO CO CO C*

00 CO

rH IO rH (N

00 00 O5 O

t CO CO

0005
00 t^-

M* co co co co co co

COOS OOS rH

00 O5

O5 CO COCO rH
*H ^H CO CO CO

oot>- oor>-

l>00 CO CO CO<N OOO <N

OOO 00 OS rH'

O O rH rH CO rH rH

rH CO CO Tf CO
<N rH rH rH rH

oo co oq5 q
^ co co

ooo

qq osos
do do

os q co q q c<j rf c<i 10 o

OO rH rH rH rH OO OO O*

(NO

CO CO

SrH TjH
rH rH

iO rHrH rH rH rHrH OO O 1

fcrh'

c-2^S
CQo

^ &H

c^ c5

000

co co

00 Oi

I>00

88 S:

CO
(N

(N (N (N <N

^ OJ -

ly
co

B
V

b |b

(U J3.0>

Harrisburg, Pa
Deficient in fine sa

t mixt

City, N
fine s

SI n.S

SII 1

O'cu ^-^>
S 4Sg

|| || ll.i B

Pa
fine sand

Pittsburg
Deficient

oT bp

rH^ -CO

N CO

rH CO O (N

CO ^^ CO CO

Tt^ O5
1C CO
rH CO

t>.

RADIATION, EXPANSION, ETC.

433

3. The amount of filler present.

4. The character of the asphalt in use.

5. The consistency of the asphalt.

6. The density and degree of compaction of the test-piece.

7. The percentage of bitumen in the mixture.

The data in the preceding tables will illustrate the application
of the test.

From these results it appears that the old-time mixtures which
are low in bitumen, those from Minneapolis and Rochester, do not
withstand impact to nearly the extent that the more modern
mixtures do, and reveal the cause of the inferiority of the pave-
ments constructed with such mixtures.

The results of tests by impact at different temperatures of
the standard New York mixture made with various asphalts
show that those in which Trinidad lake asphalt is the cementing
material give a much greater resistance to impact at low and
medium temperatures, and are much less affected by tempera-
ture changes than those made with Bermudez asphalt, while the
variation in the character and grading of the sand, and the per-
centage of filler and of bitumen, are made evident by other data
in the table.

It should also be observed that the work done in fracturing
the test-pieces is relatively as the square of the number of blows.

The impact test is also valuable in revealing the difference in
susceptibility to water action of different mixtures. Cylinders
of standard Trinidad and Bermudez asphalt mixtures were pre-
pared and some of them tested by impact at 78 Fahr. as soon as
made. Others were immersed hi water for three months and the

Trinidad.

Bermudez.

Density

2.24

Number of blows:
Original material

After three months' exposure
to running water

Water absorbed:
Pounds per square yard

129

157

434 THE MODERN ASPHALT PAVEMENT.

amount absorbed determined, after which they were subjected to
the impact test. The results are shown in the table at bottom of
page 433.

It is readily seen that the Bermudez mixture is much more
weakened by immersion in water than that made of Trinidad
asphalt. The value of the impact test is, therefore, assured from
the results thus far obtained and the investigation of the subject
will be carried out in greater detail in the future.

SUMMARY.

In the preceding pages the question of the radiation of heat
from asphalt pavements, their expansion and contraction, and
resistance of various asphalt surface mixtures to impact are
considered.

PART VII.

SPECIFICATIONS FOR AND MERITS OF ASPHALT
PAVEMENTS.

CHAPTER XXII.
SPECIFICATIONS.

As has already been mentioned, the specifications for the con-
struction of asphalt pavements which are prepared by engineers
who are not thoroughly acquainted with the subject are often
wanting in some respects or make certain requirements which
are undesirable, unessential, or unnecessarily increase the cost of
the pavement. For the construction of an asphalt pavement
which is to meet the requirements of ordinary traffic in a majority
of our cities the following, in the author's opinion, will be found
to be a thorough protection of the city's interests, and not open
to objection by the bidders that the provisions are unreasonable.

SPECIFICATIONS FOR ASPHALT PAVEMENT ON PORTLAND-CEMENT
CONCRETE FOUNDATION. 1

Extent of Work. The work shall consist of regulating and
grading the entire street, constructing combined curb and gutter,
laying asphalt pavement, and all work incidental thereto, all in
accordance with the following specifications:

Removal of Old Materials. All old material which is not to
be used in the work, shall become the property of the contractor
and be removed by him.

1 As an example of the type of specifications in actual use in cities, those
portions pf the specifications of Kansas City, Mo., which apply to the tech-
nical portion of the work are presented in an appendix.

435

436 THE MODERN ASPHALT PAVEMENT.

GRADING.

Excavation, Grading and Preparation of Foundation. Any

old material to be used again shall be compactly piled on the side
of the roadway. The excavation shall be carried to the grade
established by the Engineer. When completed, the sub-grade
shall present a line and contour parallel with and approximately
. . .inches below the surface of the finished pavement to be con-
structed. Should any soft, spongy, vegetable or other objection-
able matter be disclosed by the excavation thus made, or be
located where filling is to be done, such material shall be removed
and replaced with suitable material, which shall be thoroughly
compacted. Whenever deemed necessary by the Engineer, the
sub-grade shall be rolled with a suitable steam roller.

Inspection and Piling of Material. The materials for con-
struction, when brought upon the street, shall be neatly piled
so as to present as little obstruction as possible to travel. No
material shall be used without the approval of the Engineer, the
contractor furnishing all labor necessary for inspection, without
&ny charge.

Filling .and Embankments. Embankments shall be brought
up to the designated grades, and the top shaped off and compacted
as defined for earth excavation. Such excavated material as
may be fit for the purpose and as may be necessary, shall be used
U> fill in those parts of the streets which are below the aforesaid
grades.

CONCRETE.

Upon the sub-grade, prepared as above described, Portland-
cement concrete composed of Portland cement, clean sharp sand,
and broken stone or slag, will be laid to an average thickness of

inches. Suitable gravel may be used in combination with

the stone or slag, as hereinafter provided. The cement shall be
of the best quality of American manufacture and shall be sub-
mitted to the Engineer for "inspection at least ten (10) days before
it is used. It shall conform to the following tests, conducted

SPECIFICATIONS. 437

according to the methods recommended by the Committee on
Uniform Tests of Cement of the Am. Soc. of C. E. It shall not
set in less than one (1) hour. When mixed in the proportion of
one (1) part of cement, by weight, and three (3) parts of standard
sand, it shall have a tensile strength after exposure of one (1)
day in air and six (6) days in water of at least one hundred and
fifty (150) pounds.

The sand shall be clean and sharp, not more than 20 per cent
of which shall pass a 50-mesh screen. It shall be free from loam
adherent to the sand grains. The gravel shall be of such size
that it will satisfactorily fill the voids in the broken stone, and
it shall not contain dirt or foreign matter. The broken stone
shall consist of granite, trap, or other hard rock or slag which shall
be satisfactory to the Engineer. It shall be of such a size that
all will pass through a revolving screen, having holes two and one-
half (2^) inches in diameter, and be retained by a screen having
holes one-half () inch in diameter. Stone which is the run of the
crusher may be used when provision is made for the consideration
of finer particles than one-half (^) inch which it contains as sand.
The unit of measure in mixing these materials will be the barrel
of cement, weighing 380 pounds, and four (4) cubic feet for sand,
gravel, and stone. They shall be mixed in the following propor-
tions and in the following manner:

The sand and cement shall be mixed dry in the proportion by
volume of one (1) of cement to three (3) of sand, and then made
into a mortar by the addition of water. To this mortar will
be added six (6) measures of wet broken stone, and the whole
thoroughly mixed by hand or machinery until it is entirely uni-
form.

Where gravel is available this may be used in such proportion
that the gravel will fill the voids in the broken stone, with a con-
sequent decrease in the amount of mortar necessary to make a
compact concrete. For example: A one (1) to three (3) mortar
which could be mixed with only six (6) parts of broken stone
may be mixed with a combination of two (2) to three (3) parts
of gravel and four (4) to six (6) parts of broken stone. The con-
crete thus mixed will be of such a consistency, owing to the per-

438 THE MODERN ASPHALT PAVEMENT.

centage of water which it contains, that it shall quake very slightly
when thoroughly rammed.

The concrete as thus prepared shall then be spread on the
sub-grade and rammed, the surface being so graded that in its
finished condition it shall average .... inches below that of the
finished pavement. No concrete shall be used that has been
mixed more than one hour.

The concrete, after laying, shall be properly protected and
the surface shall be kept moist in warm weather by sprinkling
at proper intervals.

At the expiration of such a period as is found to be necessary
in order that the concrete shall have attained a sufficient set to
sustain a steam roller, the binder course shall be laid.

ASPHALT PAVEMENT.

Definition. The pavement proper shall consist of a binder
'course .... inches in thickness and a wearing surface. .. .inches
in thickness, equal in composition to the pavement mixture here-
inafter described.

BINDER COURSE.

The alternative of an open or close binder course is given.
In any specification but one should be called for. The close
binder course is much more desirable.

Open Binder. Stone. The binder shall be composed of
suitable clean broken stone passing a one and a quarter (1J)
inch screen.

Asphaltic Cement. The stone shall be heated in suitable
appliances, not higher than 300 Fahrenheit, and then thoroughly
mixed by machinery with asphaltic cement equivalent in com-
position to that hereinafter set forth, in such proportion as will
cover the stone with a glossy coat and without any excess of
asphaltic cement.

Laying. The binder must be hauled to the work and spread
while hot upon the foundation to such thickness that, after
being immediately compacted by rolling, its average depth
shall be .... inches, and its upper surface shall be approximately

SPECIFICATIONS. 439

parallel to the surface of the pavement to be laid. Upon this
binder course shall be laid the wearing surface.

No traffic, except such as may be required in depositing the
surface mixture or in otherwise prosecuting the work, shall be
allowed on the binder course.

Compact or Close Binder. Compact or close binder shall
be composed of hard, clean broken stone which shall pass an
opening one (1) inch in diameter, the voids in which are filled
with finer stone passing an opening three-eights (|) inch in diameter,
while the voids in the mixed stone shall be filled with a well
graded sand or old asphalt surface mixture which has previously
been in use, and which has been softened by heating so as to allow
it to be incorporated properly with the stone. If sand is used,
sufficient asphalt cement shall be added to thoroughly coat the
mineral aggregate with bitumen without showing any excess on
compression with a hot tamper, while if old surface material is
used, sufficient cement shall be used to coat the stone and produce
the same result.

Care should be exercised with close binder that it does not
carry an excess of fine material or of asphalt cement, as in this
case the wearing surface of the pavement may have a tendency
to displacement.

Asphaltic Cement. The asphaltic cement for the binder
course should have a consistency of at least twenty (20) points,
as indicated by the Bowen machine, higher than that in use
in the surface.

Pavement Mixture. The pavement mixture for the wearing
surface shall be composed of:

(a) Asphaltic cement (Refined asphalt and flux).

(6) Sand of satisfactory grading and grain.

Asphaltic Cement. The asphaltic cement shall be composed
of refined asphalt and flux of such character that the bitumen,
without regard to the mineral matter present, shall be a homo-
geneous solution, and present the following characteristics:

1. The proportion of bitumen soluble in 88-degree naphtha

440 THE MODERN ASPHALT PAVEMENT.

shall not exceed seventy-eight (78) per cent, nor fall below sixty-
four (64) per cent.

2. A No. 2 needle, weighted with one hundred (100) grams,
shall penetrate, in five (5) seconds, at 78 F., a distance of from
three (3) to nine (9) millimeters.

3. When twenty (20) grams are heated in a receptacle about
two and a half (2) inches in diameter and about two (2) inches
high, for seven (7) hours, in an oven maintained at a uniform
temperature of about 325 F., as determined by a thermometer,
the bulb of which is immersed in a similar receptacle filled with
oil, it shall not lose more than four (4) per cent. Its flash point
as determined in a New York State closed oil tester, shall not
be less than 350 F.

4. Ninety-five (95) per cent shall be soluble in carbon disulphide
at air temperature, and not more than one and a half (1) per cent
of the bitumen shall be less soluble in carbon tetrachloride than
in carbon disulphide, the test for the former being conducted
by submitting the bitumen to the action of the tetrachloride
for twenty-four (24) hours before filtration.

Sand. The sand shall consist of hard grains, not necessarily
sharp, but not containing more than one (1) per cent of clay
or loam. On sifting, the entire amount shall pass a ten (10)
mesh screen, at least fifteen (15) per cent shall pass an eighty
(80) mesh screen, and seven (7) per cent a one-hundred (100)
mesh screen.

Filler. 1. Powdered Mineral Matter or Dust. The filler
shall consist of ground limestone or any other mineral matter
of sufficient density to produce a powder having a volume weight
of at least ninety (90) pounds to the cubic foot. It shall be so
fine that at least sixty-six (66) per cent shall pass a two-hundred
(200) mesh screen, and all of it shall pass a fifty (50) mesh screen,
while, when thoroughly agitated with distilled water at a tem-
perature of 68 F., by means of an air blast, avoiding cyclonic
effects, not more than forty (40) per cent shall subside on standing
for fifteen (15) seconds.

2. Portland Cement. The filler shall consist of Portland

SPECIFICATIONS. 441

cement of such a degree of fineness that seventy-five (75) per cent
will pass a two-hundred (200) mesh sieve.

Only one or the other of the materials described as filler should
be called for in any specification, as the difference in cost between
the two is sufficient to require a separate estimate on each.

Combining Materials. The materials complying with the
above specifications shall be mixed in proportions by weight,
depending upon their character. The percentage of matter soluble
in carbon disulphide in any pavement mixture shall be not less
than 9.5 nor more than 12.0 per cent.

The sand and the asphaltic cement will be heated separately
to such temperatures that the finished mixture shall, depending
on the asphalt in use, have a temperature of from 290 to
330 Fahr. The filler shall be mixed, while cold, with the hot
sand. The asphaltic cement will then be mixed with the sand
and stone dust, at the required temperature and in the proper
proportion in a suitable apparatus, so as to effect a thoroughly
homogeneous mixture.

Laying the Pavement. The above mixture shall be hauled
to the street in trucks properly protected from radiation by
tarpaulins at a temperature of not less than 250 Fahr., and spread
upon the binder to such a depth as will insure an average thick-
ness of inches after ultimate compression. This compression

will be attained by first smoothing the surface with a hand-roller,
or light steam-roller, after which hydraulic cement or stone dust
shall be swept over it, when the rolling will be continued with a
steam-roller until the surface is properly compacted.

BITUMINOUS CONCRETE PAVEMENT.

For a bituminous concrete surface, the character of the asphalt
cement, sand, filler and stone will be the same as those provided
for in the previous specifications.

Determination of Proportions. The wearing surface mixture
shall be composed of the materials previously specified combined
in proportions to be determined as follows:

The stone shall be separated into fragments which are retained

442 THE MODERN ASPHALT PAVEMENT.

and those which pass a screen with circular openings of three-
eighths (|) inch size. While hot, a layer of the coarser fragments
are to be spread on the bottom of a strong box exactly one foot
square and shaken down. Hot fragments of the smaller size
are then shaken into the voids in the larger stone until the latter
are filled. Into this mixture hot sand and filler, in the proportion
of ninety (90) per cent sand and ten (10) per cent filler, are sifted,
with continued jolting and jarring of the box, until the stone has
taken up all of the fine material possible. This operation is
repeated with subsequent layers until the box is full. It is then
struck off evenly and weighed. The total weight of the mineral
matter in the box obtained in this way is divided by the weight
of a solid cubic foot of mineral matter of the same density, and
the voids in the mixture of rock, sand and filler determined. The
voids should not exceed 25 per cent.

The proportions by weight in which the different components
are present is then determined either by screening the contents
of the box or by originally taking a definite amount of each and
determining by difference the amount actually placed in the box.
These proportions shall be those which shall be subsequently used
in making up the mineral aggregate for the surface mixture
unless observation on the street in laying the mixture should
indicate the necessity for increasing the amount of fine material
present.

The proportions will ordinarily vary between the following
limits:

Coarse stone 56 to 30%

Fine stone 16" 26

Sand 25" 33

FUler 3" 5

The variations will depend upon the shape of the fragments
of the coarse stone and on the uniformity in size of the fragments
of the two dimensions.

The amount of bituminous cement necessary to coat the
particles of the above mineral aggregate and fill the voids therein
as nearly as possible, shall be determined as follows:

SPECIFICATIONS. 443

To the hot mineral aggregate assembled in the proportions
arrived at as above, shall be added the bituminous cement in
an amount which previous experience points out as within the
limits for such a mixture. After thorough mixing the material
prepared in this way is placed upon a firm surface (such as one
of hydraulic concrete) and compacted with a hot tamper until
it is thoroughly compressed. If bitumen comes to the surface
to a slight extent, the amount used is satisfactory, but if it appears
in excess, the amount of bituminous cement must be diminished,
or, if it appears dry, it must be increased until the proper appear-
ance is obtained. The percentage thus determined will be followed
in making the surface for the street, but it may be modified to
a certain extent to correspond to unavoidable variations in the
mineral aggregate, but in no case shall it exceed 8.0 per cent,
nor fall below 6.5 per cent.

Preparation and Mixing of Bituminous Wearing Surface.
The sand and stone shall be heated in suitable heaters to the re-
quired temperature, which temperature shall be dependent upon
that of the atmosphere at the time the mixing is in progress, and
upon the character of the bituminous cement in use. These
materials, while in this heated condition, shall be separated into
not less than three sizes by passing over a screen having perfora-
tions or openings of at least two diameters, so arranged as to
separate the stone and sand into particles larger than three-eighths
(I) inch in their greatest diameter, particles passing a screen
with openings three-eighths (f ) in size and retained on a 10-mesh
screen, and particles passing the latter screen; the material of
each size being collected separately in suitable bins. The material
of each size shall then be weighed or measured out in the proportions
determined upon, together with a proper proportion of filler
(finely ground mineral matter) not previously heated. The
aggregate shall then be mixed with the bituminous cement, pre-
pared as described in the paragraph on this material above given,
in a suitable mixing appliance, either pans or a double-bladed
revolving mixer of the type generally used in the preparation of
bituminous paving mixtures, the operation of mixing to be con-
tinued until uniformity is attained.

444 THE MODERN ASPHALT PAVEMENT.

Laying Bituminous Wearing Surface. In this condition the
bituminous wearing surface mixture shall be hauled to the street
in suitable carts or wagons, properly protected from the weather,
so as to prevent an undue loss in temperature. It shall be spread
upon the foundation with hot rakes to such a depth that after
receiving its ultimate compression by rolling with a steam roller,
it will have an average thickness of two (2) inches.

Surface Finish. After the rolling of the bituminous wearing
surface has been completed, there shall be spread over it a thin
coating of quick-drying bituminous flush-coat composition,
specially prepared, the purpose of this coating being to thoroughly
fill and smooth out any unevenness which may be on the surface
of the coarser mixture.

There shall then be spread over and rolled into the wearing
surface a thin layer of coarse sand or stone chips for the purpose
of presenting a gritty surface, which shall not be slippery. The
character and size of the sand or stone chips to be spread upon the
surface, and the consequent degree of roughness of surface, shall
be determined by the Engineer, who may, at his discretion,
omit the use of stone chips and substitute therefor a light sweep-
ing of hydraulic cement, stone dust, or similar material, in order
to secure a smoother surface.

MATERIAL FOR REPAIRS.

Repairs. In case of repairs, it shall be required that such
repairs be made with a pavement mixture equal to the above
described.

CLEARING UP.

All surplus materials, earth, sand, rubbish, and stones are to
be removed from the line of the work. All material covering the*
pavement and sidewalks shall be swept into heaps and imrne
diately removed from the line of the work.

SPECIFICATIONS. 445

MAINTENANCE.

Contractor to Make Repairs. The contractor shall make all
repairs which may be necessary during a period of .... years l
from the date of acceptance of the pavement, which are caused
by improper construction due to defective materials or workman-
ship, and he guarantees that he will repair during this period,
any and all defects arising through its usual use as a roadway.
In case of the contractor's failure or neglect to do so within thirty
days after notification by the duly authorized official, such notice
to be served upon him in writing, either personally or by leaving
said notice at his place of business or with his agent in charge
of the work, the said duly authorized official may repair, or cause
such defects to be repaired, and may recover cost thereof from
the contractor or his sureties.

Temporary Repairs in Winter. The contractor shall have
the right, in case of trenches, to provide against settlement by
covering the surface of the cut with broken stones and maintain-
ing the surface for a sufficient period, and during winter weather
any hole in the pavement may be filled and maintained with
binder, asphalt mastic, or other suitable material.

Repairs to Openings. During the period of maintenance,
the contractor shall, within thirty (30) days after the receipt
of notice so to do, restore the pavement over all openings made
under permits legally issued by the duly authorized official for
new service connections, or repairing, renewing, or removing
the same, and over all trenches made for carrying sewers,
water or gas pipes or any other sub-surface pipes or conduits,
for the sum of $3.50 per square yard for any opening less than

1 The author has suggested no definite period of guaranty as there is a
tendency on the part of the larger cities to shorten this to a very "consider-
able extent, New York, Chicago, Philadelphia, and St. Louis now calling
for five years, while in the State of California no guaranty whatever is
required. Shorter periods are called for in view of the fact that municipal
officials are now in a position to judge of the character of the work that is
being done on any contract before acceptance, which was not the case
years ago.

446 THE MODERN ASPHALT PAVEMENT.

ten (10) square yards in area, and $3.00 per square yard over any
trench measuring more than ten (10) square yards in area; $3.25
per square yard for restoring the pavement over any opening
between or alongside of surface railroad tracks which shall exceed
ten (10) square yards in area, and in case of any injury to the
surface of the pavement caused by fire or accident, same shall be
replaced for the sum of $2.00 per square yard. The concrete
foundation, if relaid, shall be of the same thickness as that origi-
nally laid. The contractor will not be held responsible, during
the period of guaranty, for any defects in the pavement or mate-
rials, which may have been caused by the opening of trenches
or by the improper backfilling of the same, and will not be com-
pelled to do the repaving over said trenches, except as herein-
before provided, nor shall he be held responsible for deterioration
of the pavement along the rails of street railway tracks, if the
form of construction is such that the rail vibrates under the traffic
it carries, unless the track construction has been in charge of the
contractor himself, nor shall he be obliged to replace cracks in
the surface, unless they have resulted in the disintegration of
the surface.

CEMENT CURB AND GUTTER.

Cement curb and gutter shall be composed of concrete formed
as follows:

One (1) part of Portland cement.

Three (3) parts of clean sharp sand, or other suitable material.

Five (5) parts of crushed stone.

Cement and Sand. Cement and sand shall be equal to the
materials hereinbefore described for use in concrete foundation.

Stone. The crushed stone shall be clean, free from dirt, and
crushed to such size as to measure not more than one (1) inch in
any dimension. It shall be deposited at the site of the work in
such manner as to insure its cleanliness.

Foundation. The curb and gutter composed of the above
materials shall rest on a foundation of cinders six (6) inches in
thickness after being thoroughly compacted by ramming.

SPECIFICATIONS. 447

Dimensions. The gutter flag shall be eighteen (18) inches
wide and five (5) inches thick; the curb shall be six (6) inches
thick throughout, except at the upper face corner, which is to
be rounded to a radius of one and one-half (1) inches. The

height of the curb above the gutter flag shall be inches, all

as shown on plans.

Finish. All exposed surfaces shall be covered with a finish-
ing coat of mortar three-eighths (f) inches in thickness, composed
of one (1) part of cement thoroughly mixed with one and one-
half (1J) parts of sand.

Before the concrete sets, the curb and gutter shall be cut into
sections not exceeding six (6) feet hi length.

Construction. The curb and gutter as hereinbefore described
shall be constructed at the grade and to lines established by the
Engineer, . . . .feet from and parallel with the centre line of the
street, except at intersections of streets and alleys, at which points
it shall be returned to the street line, the necessary circular sec-
tions being built to radii established by the Engineer. The curb
shall be properly back-filled to the top thereof.

PROPOSALS.

Bidders will be required to make proposals on blank forma
furnished by the Engineer, w r hich proposals shall state:

A price per cubic yard for excavation;

A price per cubic yard for embankment or filling;

A price per cubic yard for hauling excavated material or
material for use in embankment, for each 1000 feet hi excess of
one-half mile;

A price per lineal foot for straight combined curb and gutter;

A price per lineal foot for circular combined curb and gutter.

An inclusive price for asphalt pavement, including foundation,
open or close binder, and wearing surface, together with main-
tenance for a period of ... .years. 1

1 Attention must be called to the fact that propositions for the use of
one kind of binder alone and one type of filler alone should be preferably
specified for any one street. If th^y are both specified, there should be some
provision for alternative bidding.

448 THE MODERN ASPHALT PAVEMENT.

An inclusive price for bituminous concrete pavement, in-
cluding foundation and a bituminous concrete surface, \yith
maintenance for a period of.. . .years. 1

Other Specifications. Specifications of another type will be
found in the appendix, those issued by Kansas City being char-
acterized by demanding an asphalt cement of a certain character
without regard to the fine material from which it is made, and are, on
this account, much to be recommended from many points of view.

Clay Soils in Cold Climates. Although the preceding speci-
fications are, as has been said, satisfactory in the majority of
instances there are cases where, owing to the character of the
climate, sub-soil, or heavy traffic to be carried by the pavement,
special provisions must be made. On clay sub-soil in cold climates
some special provisions, such as are made in Manitoba, may be
desirable in the treatment of the sub-soil base. Such a provision
may be outlined as follows:

In clay soils trenches shall be dug across the line of the street
from the centre to the trenches in which the curb is laid on each
side of the street, to a depth of six (6) inches and filled with broken
stone or large gravel. The entire roadbed will then be thor-
oughly rolled with a steam roller, having a tread of at least (60)
inches and a pressure per linear inch of tread of at least 310 pounds,
along the large roll, until it is compacted. All soft spots which
are developed should be refilled and rerolled. The surface thus
prepared must conform closely to the prescribed cross-section of
the street.

Upon this foundation, broken stone, preferably the run of
crusher, gravel or clean sand, will be laid to the depth of three
(3) inches and thoroughly consolidated by rolling, to be followed
by the hydraulic concrete, the latter in this way not being brought
in contact with the soil and drainage being provided by the broken
stone.

This form of construction is most necessary in very cold climates
and especially on clay soils where a thaw is apt to occur from
the action of frost, and where cracks have been observed to open
in the ground and extend through the concrete and the asphalt
surface.

1 See footnote on page 447.

SPECIFICATIONS. 449

As an additional precaution under such conditions the follow-
ing provisions are made as to setting the curb:

Curbing. Curb of the character described shall be set hi con-
crete hi a trench. .. .niches deep, the bottom of which is filled

with broken stone to a depth of inches, connected with the

broken stone cross trenches of the base, upon which shall rest
the concrete foundation for the curbstone, not less than six (6)
inches thick and seventeen (17) inches wide, made of the mate-
rials in the proportions previously described for the concrete base,
except that the stone shall not exceed one and one-quarter (1J)
niches hi maximum dimension. The curb shall be imbedded
immediately hi the centre of this concrete and backed up with
additional concrete for a width of six (6) inches, extending from
the concrete base to within four (4) inches of the top of the curb.
The broken stone underlying the concrete shall be graded to catch-
basins for the removal of ground-water.

Sandy Soils. On sandy soils at the seashore, where it is diffi-
cult to compact the sand under the roller, it may be provided
that a course of one (1) or two (2) inches of gravel or other suit-
able material may be spread and rolled over the sand before the
concrete foundation is constructed

Asphalt. While, in the author's opinion, there is no question
that the best asphalt surface, in the present state of the industry, can
be constructed with Trinidad lake asphalt orgilsonite, he would not
be understood to affirm that satisfactory work cannot be done with
other bitumens, and for streets of light traffic the Engineer may
exercise such choice as he may believe to be desirable from the
point of view of competition. He should, however, bear in mind
that the skill which the contractor may possess and his knowl-
edge of the art of constructing an asphalt pavement is of as great
importance as the materials in use or the price which may be bid.
At an equal cost the pavement constructed with skill and intelli-
gence may be worth a very much larger sum when completed
than another surface carelessly constructed. These points should
weigh largely hi the awarding of contracts if the cost of main-
tenance of the surface to the city after the expiration of the guar-
antee period is to be considered.

450

THE MODERN ASPHALT PAVEMENT.

Grades of Streets on which Asphalt Pavements may be Con-
structed. The general impression has gained ground, very natu-
rally, that asphalt pavements are unsuited to grades of more than
4 to 5 per cent. That this is an erroneous conclusion may be seen
from the fact that in 1890 an asphalt surface was laid in Wash-
ington, D. C., on Thirty-fourth Street, N. W., from M Street to
Prospect Street, 275 feet long, the grade of which is 9.74 per cent,
and that in Kansas City, Mo., the following streets have been con-
structed with the grades given. 1

GRADES IN KANSAS CITY, MO., FOR ASPHALT PAVEMENTS.

Year Laid.

Street.

Grade.

1898

Jefferson Street 18 to 20.

12 5%

1895
1897
1895

11 Street, Maine to Wyandotte. .
Troost Ave., 19 to Belt Line. . . .
Central Street, 16 to 17

7.5

8.0
10

1894

Forest Ave., Independence to 8.

8.0

All of these streets are in constant use and are satisfactory
except on occasions where a thin coating of moisture has
become congealed on the surface. Several of the streets in Kan-
sas City are only paved with asphalt in the centre, the sides
having a stone or brick surface. Nevertheless, the asphalt sur-
face is universally used in preference to the brick or stone, and
appears to be no more slippery even under the most trying con-
ditions. Where a film of ice causes asphalt to be slippery traffic
is diverted to other streets with lighter grades, rather than to the
brick or stone. As a matter of fact the limiting conditions in
determining the extent to which the steepness of a grade will
prevent the use of an asphalt surface mixture will depend entirely
upon the climate and the nature of the traffic which uses the street.
Eight per cent is not an excessive grade under ordinary eastern
conditions, while in a climate like Seattle, Wash., a 10 per cent
or 12 per cent grade is quite possible.

1 Tillson cites a grade of 17 per cent on a portion of Bates Street, in
Pittsburg, Pa., and 12 per cent in Scranton, while Baker mentions one of
16 per cent in San Francisco, Cal.

SPECIFICATIONS.

451

Crown or Camber. That an asphalt pavement should show
in transverse section a proper profile for the surface is as important
as that the grade should be sufficient to provide for proper drain-
age. There is little agreement among engineers in regard to
what this proper form should be, but it is quite certain that the
tendency in America is to make all the asphalt pavements much
too flat. Theoretically, no doubt, an asphalt pavement should
demand but a low crown or camber. In practice, however, the
pavement will prove much more satisfactory and pleasing to
the eye if this is maintained at a comparatively high figure, since
in wet weather the slight depressions which it is impossible to
avoid in laying such a surface, or which are formed by unequal
compression and traffic, will not then be revealed as small pools
of water. This is especially the case if the profile of the surface
shows a plane surface from the gutter to the crown instead of a
curve.

It is generally assumed that for a roadway 30 feet wide a
crown of 4 inches should be adopted, with a curve towards the
gutter having a somewhat greater fall near the latter and decreas-
ing towards the crown. The objection to this profile is that the
street is too flat on the crown, with the result that depressions form
there which retain water. It is, therefore, much better to keep
the crown raised sufficiently to avoid this. On the other hand,
with a nearly flat crown, the centre of the street which is used
principally for traffic is of a more acceptable form. Mr. G. W.
Tillson l gives the following table showing the necessary crown
for streets of a width from 24 to 60 feet.

Width of roadway . ...

24 ft.

30 ft.

36 ft.

48 ft.

60 ft.

Crown. ...

3 ins.

4 ins.

6 ins.

5 ins.

6 ins.

8 ins.

Fall towards gutter in cen-
tral J of roadway
Rate per 100
Fall towards gutter in sec-
ond J of roadway. .

Jin.

8 ins.

1 in.

4 in.

9 ms.

1^ ins.

in.
13 ins.

2 ins.

f in.
9i ins.

1 ins.

f in.
8 ins.

2 ins.

f in.
8| ins.

2| ins.

Rate per 100

9' 1"

2' 3"

3' 4"

2 / 4//

2' 1"

2' 3"

Fall to gutter in of road-
way adjacent to curb. .
Rate'per 100. .

1 ins.
3' 6"

2f ins.
3' 8"

3 ins.
5' 6"

21 ins.
3' 3"

3J ins.
3' 6"

4f ins.
3' 9"

1 Street Pavements and Paving Materials, 202.

452

THE MODERN ASPHALT PAVEMENT.

The author would regard an 8-inch crown as none too high
for a 60-foot roadway, while on many flat streets a 6-inch crown
is not too high for a 30-foot roadway. Of course the steeper the
grade of the street the smaller the height of crown which is
necessary, and this fact does not seem to have been taken into
consideration in the table which Tillson offers.

In Paris, France, the crown for asphalt streets is determined
by the formula, Fig. 26.

Provision is made for a drop of 10 per cent from the point A
towards the curb for a space of 50 cm. (19.7 inches). This latter
provision seems to the author to be an excellent one and, from
his experience in Paris, the form of street profile to be a very suc-
cessful one.

Baker l gives a resume of the specifications of various cities

FIG. 26.

ATL 2
'NL-l'
AB= 0.05 meters.

for crowns of asphalt pavements to which the reader may refer.
The provisions of the City of Omaha are also very excellent.

1 Roads and Pavements, 348.

SPECIFICATIONS.

453

TABLE OF STANDARD CROWNS.
(City Engineers Office, Omaha, Neb., 1902.)

Crowns for American Sheet Asphalt Pavement in Feet.

Distance

Grade of Street.

Between

Curbs.

10%

11%

12%

20 feet . . .

.40

.38

.37

.35

.34

.32

.30

.29

.27

.26

.24

.22

.21

.50

.48

.46

.44

.42

.40

.38

.36

.34

.32

.30

.28

.26

.60

.58

.55

.53

.50

.48

.46

.43

.41

.38

.36

.34

.31

.70

.67

.64

.62

.59

.56

.53

.50

.48

.45

.42

.39

.36

.80

.77

.74

.70

.67

.64

.61

.58

.54

.51

.48

.45

.42

.90

.86

.83

.79

.76

.72

.68

.65

.61

.58

.54

.50

.47

1.00

.96

.92

.88

.84

.80

.76

.72

.68

.64

.60

.56

.52

1.10

1.06

1.01

.97

.92

.88

.84

.79

.75

.70

.66

.62

.57

1.20

1.151.10

1.06

1.01

.96

.91

.86

.82

.77

.72

.67

.62

1.30

1.2511.20

1.14

1.09

1.04

.99

.94

.88

.83

.78

.73

.68

1.40

1.341.29

1.23

1.18

1.12

1.06

1.01

.95

.90

.84

.78

.73

1.50

1. 4411. 38

1.32

1.26

1.20

1.14

1.08

1.02

.96

.90

.84

.78

1.60

1.54

1.47

1.41

1.34

1.28

1.22

1.15

1.09

1.02

.96

.90

.83

NOTE. The formula used for the construction of the table is as follows :

TF(100-4/).

5000

C crown of pavement in feet;
W = distance between curbs in feet;
/= number of feet fall per 100 feet of street.

Note. Where the crown is less than 0.5 foot make the gutter 0.5 foot,
and where it is 0.7 foot make the gutter 0.7 foot, and for intermediate crowns
make the gutter equal the crowns. Andrew Rosewater, M. Am. Soc. C. E.,
City Engineer.

None of these methods of arriving at a proper figure for crown
is applicable if opposite sides of the street are not of the same
elevation.

The subject of the form to be given to streets is discussed at
much length and with reference to various cases in ' ' Traite
Pratique des Travaux en Asphalte," Letouze etLoyeau, Paris, 1897,
and reference must be made to this work for more detailed formula
and special applications.

Gutters. In many cities where concrete curb and gutter are
not in use, there has been a tendency to use either stone or brick

454 THE MODERN ASPHALT PAVEMENT.

as a substitute for an asphalt surface in gutters of asphalt streets.
From what has been shown in the previous pages it is evident
that where the asphalt-surface mixture is made on the lines laid
down by the author, and where the form of construction employed
in the street is such as to provide satisfactory drainage, there is no
reason why the asphalt surface should not be carried from curb to
curb, nor is there any necessity for painting such a surface with
asphalt or bitumen, if the mixture of which it is composed is
constructed on modern lines. In the early days of the industry
this painting became the custom, owing to the fact that the
mixtures were made with sand so badly graded and so porous
that they could not resist water action.

CHAPTER XXIIL
THE MERITS OF THE MODERN SHEET-ASPHALT PAVEMENT.

WHETHER a sheet-asphalt pavement possesses any lasting degree
of merit will depend entirely upon the manner in which it is con-
structed from the base up, including proper drainage and the charac-
ter of the asphalt mixture which forms the surface. It has already
been made evident that the greatest care is necessary in all these
respects. It will be only worth while, therefore, to consider
what the merits are of a sheet-asphalt pavement of standard
construction. Such a pavement is desirable for the following
reasons :

1. It does not disintegrate under impact or attrition, and con-
sequently produces neither mud nor dust.

2. It can be kept perfectly clean if the proper efforts are made
to do so.

3. It has an impervious surface and does not absorb filthy
liquids, as is the case with wood blocks.

4. It affords the best foothold for horses except under occa-
sional conditions.

5. Traction on such a surface can be carried on with a smaller
expenditure of force than on any other form of pavement.

6. Its wearing properties compare more than favorably with
granite and exceed that of any other form of pavement under
heavy traffic.

7. Deterioration hi a standard asphalt pavement is of a kind
that can be readily and economically met owing to the simplicity
of making repairs, something that cannot be done satisfactorily
with any other form of pavement.

455

456 THE MODERN ASPHALT PAVEMENT.

8. Cuts in the pavement for underground work can be replaced
in a manner which makes the repairs undistinguishable from
the original surface, whereas they are quite evident in the case
of other pavements.

9. It increases the actual and rental value of all real estate
abutting on streets where it is laid to a larger extent than any
other form of pavement.

10. The wear and tear upon horses and carriages is largely
reduced by asphalt pavements, and it has been estimated for
Philadelphia l that the repairs to vehicles in that city due to
rough pavements existing there in 1885, which could be saved
by sheet-asphalt pavements, would amount to $1,000,000 annually.
The universal testimony of fire-department chiefs is that there is
far less wear and tear to the running gear of the engines, hose
carriages and trucks on asphalt pavements than on stone blocks,
and consequently less liability to break down or to have acci-
dents, while much better tune is made in going to fires.

That asphalt pavement will sustain the heaviest traffic that
is carried by any street in the world can be seen from the follow-
ing determination of the number of vehicles and the tonnage on
Fifth Avenue and some other streets in New York City during the
months of November and December, 1904. See table on page 457.

The heaviest traffic in London, as determined in 1879, was
422 tons per foot of width per day. The traffic on Fifth Avenue,
which has been an extremely successful asphalt pavement, is,
therefore, equal to, if not greater than, that sustained by the pave-
ment on many of the most heavily travelled streets of Europe.

The defects which have generally been assigned to an asphalt
pavement are its comparatively great first cost and cost of main-
tenance. Its first cost may be larger than that of some other
inferior forms of pavement, but considering the length of time
that an asphalt surface will wear, if of standard construction, this
cost is smaller per annum and per ton of traffic carried than
that of any other form. The cost of maintenance has no doubt
been large for many pavements constructed in the past and will
be large for many constructed in the future which are not of stand-

1 The Philadelphia North American, Oct. 12, 1885.

MERITS OF THE MODERN SHEET-ASPHALT PAVEMENT. 457

ard composition. With the best form of construction the cost
per yard will not be excessive and the public will have the advan-
tage, if the city maintains its streets, which unfortunately is not
always the case, of having a perfect pavement at all periods of
its existence, instead of one which becomes worse and worse with
each year of its age.

Another defect has been said to be the fact that it is unsuited
for steep grades. From the figures given on pages 450 and 451
it is evident that this is not so.

There can be no question that a standard sheet-asphalt pave-
ment possesses more merits than any other, and fewer defects.
It is undoubtedly the pavement of the present and of the future.

TRAFFIC RECORD TAKEN ON STREETS PAVED WITH ASPHALT
IN NEW YORK, N. Y., NOVEMBER AND DECEMBER, 1904.

Tonnage
11 Hours.

Average
Tonnage
per Linear
Foot of
Width per
11 Hours.

Average
Tonnage
per Hour.

Average
Number of
Vehicles

11 Hours.

Average
Tonnage

Vehicle.

Fourth Street, from Wooster
to West Broadway

9254 22

289 18

841 35

3394

2 73

Eighth Avenue, from 35th
to 36th Streets
Thirty-fourth Street, from
Broadway to Seventh Av .
First Avenue, from 26th to
27th Streets . . .

13024.52
2176.60
19253 76

296.02
89.22
435 58

1184.05
197.86
1750 34

5720
1072
6034

2.28
2.03
3 18

Fifth Avenue, from 33d to
34th Streets. . . .

19274 47

481 85

1752 20

11787

1 64

Broadway, from 18th to 19th
Streets

7491 70

299 63

681 06

3817

1 97

The Cost of Asphalt Pavements. No general statement can
be made in regard to the cost of an asphalt pavement, as it is a
function of too many variables ; these variables can, however, be
considered individually. They are:

1. Freight rates for the transportation of the plant and mate-
rials of construction to the locality where the pavement is to
be laid.

2. Local cost of materials of construction, such as sand, cement,
gravel, broken stone, and filler.

458 THE MODERN ASPHALT PAVEMENT.

3. The cost of local labor.

4. The form of construction which is specified.

5. The character of the traffic which the street is to carry,
its grade, and the character of the pavement on adjoining streets.

6. The period of guarantee demanded.

7. The terms of payment.

With so many changeable conditions it would, of course, be
impossible to give any general data as to the cost of an asphalt
pavement. It may vary from $4 or $5 per yard on a street of
extreme traffic in a large city which is guaranteed for 15 years,
and $1.25 per yard on old brick pavement for the base where
the traffic is very light, as in a residence street or where no guar-
antee is demanded, as is the case in the State of California.

Cost of Maintenance. The cost of maintenance is quite as
uncertain an element as the cost of construction; and even more
so, since it will depend not only on the character of the original
work but upon the amount of attention which is given by the
company constructing the pavement, or by the authorities after
the former's guarantee has expired, to keeping the surface in first-
class condition. If it is neglected the cost of maintenance may
become large, whereas if carefully kept up this may well be small.

The character of the base which supports the pavement will
have more to do with the cost of its maintenance, if the surface
is a standard one, than any other controlling condition. In the
observation of the writer at least 90 per cent of all maintenance
work on asphalt pavements with well-constructed surfaces is due
to weakness and deficiency in the base.

It is of interest in this connection, however, to note the data
contained in a paper by Capt. H. C. Newcomer, Corps of Engineers,
United States Army, in regard to the cost of maintenance of the
asphalt pavement in Washington, D. C., especially as the sur-
face mixtures laid in that city have been, unfortunately, not of
standard quality, owing to deficiencies in the character of the
local sand supply. Capt. Newcomer has found l that " the aver-
age cost per square yard per annum for the second five-year period
of the life of the pavements considered was 1.65 cents; for the

1 Engineering News, 1904, Feb. 18, 51, 165.

MERITS OF THE MODERN SHEET-ASPHALT PAVEMENT. 459

third five-year period, 3.37 cents; for the fourth five-year period,
3.78 cents, and for the fifth five-year period, 2.56 cents. The
average cost for all ages tabulated was 2.8 cents."

It is worthy of note in this connection that of the 2,425,732
square yards of bituminous pavements of all kinds, including coal-
tar, in the preceding estimate, many of which were very inferior,
not less than 2,161,181 square yards were constructed of Trinidad
asphalt.

CHAPTER XXIV.
ACTION OF WATER ON ASPHALT PAVEMENTS.

THE action of water on asphalt and on asphalt pavements
has been a prominent topic of discussion from the early days of
the industry, and the subject, for many reasons, remains one of
peculiar interest to-day, since many mistaken ideas in regard
to it are still in vogue.

Asphalt surface mixtures, the mineral aggregate of which is
not properly graded and balanced and which, in consequence,
lack density and an impervious surface, are attacked by water
when subjected to its continued action, from lack of proper drain-
age or other reasons, without regard to the nature of the asphalt
of which the mixture is made, although under these conditions
one asphalt may be attacked more than another. With the stand-
ard surface mixture constructed on the ideas laid down by the
author in the previous pages surface mixtures may be constructed
of any asphalt which are equally resistant to water action, but
all of which are attacked more or less by the water unless allowed
to dry out at intervals. In a properly constructed pavement no
important deterioration from water action should ensue within
the life of the pavement and, as a matter of fact, in the author's
experience, the deterioration in the asphalt surfaces laid under
his supervision has in the last ten years become an item which
is hardly worth consideration, where the form of construction
has provided satisfactory drainage.

As it must be admitted that asphalts are attacked by water
to different degrees when the mixtures are not dense, and espe-
cially in laboratory tests, it is of interest to examine into the rea-

460

ACTION OF WATER OX ASPHALT PAVEMENTS. 461

son for this, in order that we may be able to compare practical
results with those obtained by experiment and determine the
means for preventing such action on pavements actually in use.

Numerous observers have detected and noted the fact that
there is a difference in the degree to which water acts upon vari-
ous asphalts and fluxes. Messrs. Whipple and Jackson have made
an elaborate investigation of the subject, the results of which
were presented in a paper read before the Brooklyn Engineers'
Club in March, 1900, which was published in the Engineering
News for March 22, 1900. These results, although of little inter-
est as showing the effect of water on a well-constructed asphalt
paving mixture, since the pure bitumens were themselves exposed
by these investigators directly to the continued action of water
in order to determine their relative value for the construction of
concrete lining for reservoirs and not for pavements, are of inter-
est as showing that experiments conducted under conditions
employed by these investigators may lead to conclusions which
are utterly erroneous as applied to the paving industry, except
when the asphalts in question are used in an unskillful way.

Actual Results on the Streets. It is of interest to consider
what the practical experience has been on the street during
the last fifteen years as regards the action of water on asphalt
surface mixtures. In the early days of the industry, as has already
appeared, the asphalt surface mixtures were very open. At the
time that the author was connected with the Engineer Department
of the District of Columbia the Trinidad asphalt surface mixtures
were constructed with coarse sand and very little filler. The
gutters of streets which were paved with mixtures of this descrip-
tion were much given to deterioration from the action of water,
and the same conditions were met in other cities, so that at that
time every one believed that it would be impossible to construct
a Trinidad surface mixture which would not be attacked by water.
This idea has persisted in the minds of many who have not followed
the industry carefully down to the present day, and it was only
dissipated in the author's mind by an experience extending from
1894 to 1896 during an attempt to introduce the American form
of asphalt pavement in London, England. In 1894 a Trinidad

432 THE MODERN ASPHALT PAVEMENT.

asphalt surface was constructed on Pelham Street, Kensington,
and on King's Highway, Chelsea, in London, using Trinidad asphalt
in much the same way that he had employed it in previous years
in Washington, D. C., the modern methods of constructing a
mixture to withstand heavy traffic and wet climate not having been
developed at that time. The results were that the pavements
were not an entire success and scaled. It was suggested that
this was due to the fact that the asphalt in use was Trinidad and
that this was constantly attacked by the continuous fogs of London.
The pavements were, therefore, replaced in the following year
with a mixture made with Bermudez asphalt. These surfaces
went to pieces much more rapidly than the previous Trinidad
surfaces. By this time the principles which have been elucidated
in the preceding pages had been largely worked out. In the third
year Trinidad asphalt surfaces were laid in London on these lines
which not only were not attacked by the continued wet weather
and fogs of that climate, but which have remained there to the
present time, having shown no deterioration due to water action.
If the Bermudez surfaces had been constructed with the same
regard to the mineral aggregate and to the character of the asphalt
cement prepared from it they would undoubtedly have shown
an equal freedom from the action of water, but it is not probable
that they would have shown an equal resistance to the deteriorating
influence of the heavy traffic on the streets on which the pavements
were laid. Practical experience rather than theory, therefore, leads
the author to conclude that an asphalt which may not appear to be
as satisfactory in laboratory tests may prove more so in actual
construction.

That asphalts which are not attacked by water in the laboratory
may be seriously affected by it in asphalt surface mixture has
frequently been revealed, but never in a more striking way than
in Reading, Pa., where house drainage is conducted along the
gutters of the Bermudez asphalt pavements of that town, in con-
sequence of which they have entirely disintegrated, as shown in
the accompanying illustration, Fig. 27.

It appears then that it is the manner in which the asphalt is
used and the practical results obtained with it rather than its

~
_z

464

THE MODERN ASPHALT PAVEMENT.

properties as revealed by laboratory tests which should control
our judgment in forming an opinion of its behavior towards water
in an asphalt-surface mixture on the street. 1

As a practical example of the difference between surface mix-
tures actually in use and made with different asphalts in their rela-
tion to water absorption in the laboratory, the results of an examina-
tion of the Trinidad and Bermudez surface mixtures which were
being laid in the city of New York in the year 1904 may be of
interest. These mixtures consisted of the following material.* ^
the proportions given and had the following composition '

Proportions.

Trinidad.

Bermudez.

Sand (1 Cow Bay and 1 Grossman)
Filler (P C dust) .

74.7%
9 6

73.7%
14 8

Asphalt cement. . .

15 7

11 5

Analyses.
Bitumen. ...

100.0
11 0%

100.0

11 2%

17 8

100- ' '

11.0

80-

11.0

10.0

50-

27.0

40-

12.0

30- . .

20-

10-

100.0

Of the above mixtures cylinders 1 inch in height were made
having the greatest density possible, by compressing them under
impact in a diamond mortar of a diameter of 1.25 inches. The
cylinders of mixtures had the following densities and weight:

1 This subject has been discussed at length in the Engineering News,
1904, June 2, 51, 520.

ACTION OF WATER ON ASPHALT PAVEMENTS.

465

Cylinder
Number.

Trinidad.

Bermudez.

Density.

Weight
(Grams).

Density.

Weight
(Grams).

2.247
2.200
2.204
2.217
2.214
2.223

2.217

47.522
47.227
46.963

46.885
48.536
49.548

47.780

2.262
2.225
2.232

2.277
2.222
2.260

2.246

46.728
47.410
47.551
49.631
49.410
52.257

48.831

Average.

These cylinders were exposed to the action of running water
for a length of time. The gain in weight of the cylinders at various
intervals is shown in the following table in fractions of a pound per
square yard :

ABSORPTION OF WATER. POUNDS PER SQUARE YARD.

Cylinder
Number.

Trinidad.

Bermudez.

1
Week

4
Weeks.

Months.

Week.

4
Weeks.

Months.

3
Months

1..
2 .... .

.0849
0789
.0763
.0869
.0839
.0706

.0803

.1111
.1009
.0991
.1069
.1066
.0943

.1031

.1262

.1190
.1149
.1199
.1237
.1137

.1196

.1291
.1194
.1171
.1240
.1254
.0914

.1177

.0961
.1485
.0868
.0894
.0909
.0526

.0940

.1020
.1428
.0914
.0883
.0966
.0566

.0963

.1183
.1640
.1134
.1077
.1198
.0746

.1163

.1174
.1694
.1103
.1089
.1194
.0729

.1164

3
4.. ....

5.
6.

Average.

After an exposure of three months none of the cylinders
showed any signs of softening. Those containing Bermudez
asphalt could, however, be distinguished from those made with
Trinidad asphalt by slight excrescences the size of pin-heads,
which had appeared upon the surface. Fig. 28. It must be
remembered, too, that the cylinders were not prepared from mix-
tures made in the laboratory, but were made from material which
was actually being used on the street.

466

THE MODERN ASPHALT PAVEMENT.

When pieces of glass were coated with Trinidad and Bermudez
asphalt cement and with one made from a California residual pitch
and immersed in running water for a week they were all more or
less attacked thereby, as can be seen from the accompanying
illustration, Fig. 29. From the preceding results it is apparent

Trinidad Lake Asphalt, Bermudez Asphalt.

Surface Mixture.

In running water five months.

FIG. 28.

at once that although all asphalts under certain circumstances
are attacked by water, Trinidad asphalt when properly used in an
asphalt surface mixture is the equal of any other in resisting
power, and this fact being proved to a contractor's satisfaction, he

California Oil
Asphalt Cement.

Bermudez

Asphalt Cement.

In running water one week.

FIG. 29.

Trinidad Lake
Asphalt Cement.

ACTION OF WATER ON ASPHALT PAVEMENTS. 467

prefers to employ it for the many reasons which have been given
in another place, namely, because no other material offers such a
uniform supply as that taken from the Trinidad pitch lake, every
cargo being handled in the same manner as those preceding it,
because the bitumen which it contains is free from hydrocarbons,
which are volatile at the temperature at which it is necessary to
maintain a surface mixture, in consequence of which asphalt cement
made with it from proper flux is peculiarly non-volatile and non-
changeable at this temperature, and because it can be maintained
for a considerable length of time at high temperatures without
hardening excessively, even when tossed about loosely with exces-
sively hot sand in any process of turning out the surface mixture.

Cause of the Action of Water on Asphalt Under Certain Cir-
cumstances. All bitumens, as has been seen, are more or less
acted upon by water under certain conditions. It is a matter of
great interest to determine what conditions are most favorable
for the destructive action of water, how far this action is inherent
in certain properties of the bitumen, and how far to the presence
of gases or salts soluble in water, or of the latter mixed with
the asphalt.

There has been a great cry that the soluble salts in Trinidad
asphalt were the cause of the deterioration of this material in the
presence of water. The idea unfortunately originated with the
author many years ago on insufficient evidence. It was soon shown
that the addition of 5 per cent of common salt to a Trinidad asphalt
surface mixture or immersion of the latter in salt water com-
pletely prevented any disintegration, even of the old-time open
surface mixture. This, of course, quite does away with the idea
that the presence of soluble salts in Trinidad asphalt has any-
thing to do with its disintegration when exposed in the refined
condition to the continued action of water. As a matter of fact,
the bitumen of Trinidad asphalt is not in itself attacked by sea-
water under the conditions imposed by Messrs. Whipple and
Jackson.

When the material which has become disintegrated and brown
under these tests is remelted the original bitumen is recovered in

468

THE MODERN ASPHALT PAVEMENT.

an unchanged condition, both as to consistency and softening point.
The action of the water seems, therefore, to be in this case caused
by its absorption by some of the organic matter or non-bituminous
matter which the asphalt contains. If the vegetable matter is so
sealed up in the asphalt or in the surface mixture as to prevent
diffusion no disintegration occurs. As a preventive against the
slightest diffusion the presence of a material, such as Portland
cement, which will combine with the water is desirable. 1

That asphalt surfaces can be constructed from any asphalt
so that they will not be attacked by water has been conclusively
proved within the last few years. In the same way it has been
equally conclusively proved that all asphalts, under certain con-
ditions, are more or less attacked by water.

That a distinct advance has been made along this line can be
seen by comparing the amount of water absorbed by the surfaces
of 1894 as compared with those of ten years later, as shown in the
following table:

ABSORPTION OF WATER BY CYLINDERS OF ASPHALT SUR-
FACE. IN POUNDS PER SQUARE YARD.

Washington, 1893.

Standard Mixture, 1904.

Trinidad.

Bermudez.

Trinidad.

Bermudez.

7 days. .
28 " !!

.314
.434
.502

.063
.194
.306

.080
.093
.107

.094
.093
.104

The conditions to which asphalt surface mixtures are sub-
jected in the street and in the laboratory bear no relation to one
another. In the ordinary laboratory tests surface mixtures are
submitted to the continued action of water, except when bur-
nished from time to time with a burnishing-tool for experimental
purposes, and receive no compaction as does the street surface
from the traffic which it receives. In the street an asphalt sur-

1 See page 27: P. C. as a Filler.

ACTION OF WATER ON ASPHALT PAVEMENTS. 469

face is never subjected, at least in well constructed pavements,
to the continued action of water. The conditions are, there-
fore, in this respect very different from any which are found in
laboratory experiments. The conclusion must, therefore, be
drawn that we must be guided in forming an opinion in regard
to the availability of any material by the results obtained in prac-
tice and not by theoretical deductions from laboratory experi-
ments. The asphalt surface laid on Fifth Avenue, New York,
a thoroughly well-constructed surface, if the presence of an open
binder is barred, has been practically unacted upon by water,
although made of Trinidad asphalt, in the ten years of its exist-
ence, since no repairs of any amount have been made to the pave-
ment due to deterioration of the mixture. No doubt a very
Email amount of deterioration may be detected in the gutters
along the curb, but this would have been the same in the case of
other asphalts as in the Trinidad mixture and does not reach an
extent to demand consideration.

In this connection it is of interest to call attention to the fact
that a process has been patented for washing crude Trinidad
asphalt for the removal of soluble salts before it is refined, and
that it is true that material thus treated withstands laboratory
tests to a somewhat better degree than the untreated material
when exposed to the action of water in the refined condition, but
the behavior of the surface mixture is not improved by it to any
appreciable extent, and the process must, therefore, be regarded
as involving an additional expense with no compensating return.

In conclusion the author may state with the utmost conviction
that no Trinidad asphalt pavements which have been laid under
his direction in the last ten years have suffered from the attack
of water when a proper form of construction has been employed.
All attempts which have been and are now being made to prove
the contrary are based purely upon personal and political attempts
to disparage the nature of the material.

SUMMARY.

An endeavor is made in the preceding chapter to show that
the conclusions derived from laboratory experiments and from

470 THE MODERN ASPHALT PAVEMENT.

the results of poor workmanship, as regards the action of water
on asphalt surfaces, are not practical, but merely theoretical. It
is shown that with requisite skill, surface mixture can be made
from those asphalts which are themselves attacked by water in
the refined state in the laboratory which will not be at all
attacked by water either in the laboratory or on the street.
Statements to the contrary generally originate in a desire to damage
the reputation of a material for reasons arising in business rivalry.

PART VIII.

CAUSES OF THE DEFECTS IN AND THE DETERIO-
RATION OF ASPHALT SURFACES.

CHAPTER XXV.
DEFECTS IN AND DETERIORATION OF ASPHALT PAVEMENTS.

ASPHALT surfaces, like all pavements, necessarily deteriorate
with age even when they are originally of the most acceptable
form of construction. When they are not well constructed they
deteriorate very rapidly.

Defects in asphalt surfaces are more apparent than in any
other form of pavement, since it is a continuous, smooth surface
without joints. The eye, as well as the effect of any irregularity
upon the vehicle passing over it, reveals them at once, where the
difference between a perfect and worn stone or brick surface is
not as noticeable.

The proper method of construction of an asphalt pavement
and the characteristics of a desirable asphalt surface mixture
have already been elaborated. At this point it seems appropriate
to sum up the causes of the deterioration in such surfaces which
are due to defects in construction or environment and to follow
this with an examination of the causes of legitimate wear.

Deterioration of or defects in asphalt pavements are attributed
to three principal causes and many minor ones:

471

472 THE MODERN ASPHALT PAVEMENT.

1. Defects in construction due to

A. Improper specifications or form of construction.

B. Lack of lateral support.

C. Inferiority of sand, in the character of the filler or lack

of a sufficient amount of it.

D. Inferiority in the asphalt or lack of intelligence in its use.

E. Careless workmanship and ignorance.

2. Unfavorable environment.

A. Climate.

B. Lack of cleanliness and general neglect.

C. Action of water, of illuminating-gas, or of gas and water

combined.

D. Flushing with water under pressure.

E. Constant opening of the surface for underground work.

3. Age.

A. Natural wear.

B. Neglect of maintenance.

Improper Specifications. It often happens that from motives
of economy specifications provide for a form of construction of
asphalt pavements which is deficient in one or more respects from
what is necessary to enable them to meet the conditions to which
they are to be exposed.

Particular attention has already been drawn to faulty pro-
visions for a suitable foundation and for proper drainage. Itishardly
necessary to recur again to this matter here except to emphasize
the fact that without a rigid foundation and pro per protection of the
surface mixture from water reaching it from the bottom, or standing
on or flowing constantly over the top, an asphalt pavement in
every other way of the highest type of construction cannot have
a long life, at least without extensive maintenance.

Specifications are also at fault in regard to the depth of the
binder course required. An inch of binder made of inch stone
cannot, in the writer's opinion, form a sufficient bond to keep it
from going to pieces under constant traffic, especially if it is sup-
ported by only a weak base. It is probable that on the heaviest
travelled streets in our large cities an open binder course is an
unsatisfactory form of construction. In summer when the surface
is soft the binder is crushed under the weight of trucks with too

DEFECTS ^AND DETERIORATION. 473

narrow tires, carrying loads of as much as seven tons, particularly
when the binder stone is not hard. The binder in such cases should
be replaced by a denser mixture, one in which the voids in the
stone are filled by a bituminous mortar, in fact, the regular asphaltic
surface mixture. Such an asphaltic concrete supports the surface,
most satisfactorily distributes the load over the base, and is of
great advantage when placed over a base subject to vibration,
such as stone blocks which have been reset, or laterally against a
vibrating rail. Specifications for such a course have already
been given.

The thickness of surface specified is less often at fault. If
properly supported, an inch and a half of surface made of desirable
constituents has satisfactorily carried heavy traffic. A greater
thickness may often be preferable for business streets, but the
greater the thickness the greater the difficulty hi raking out the
hot mixture evenly and obtaining uniform compaction and the
greater its liability to displacement with the formation of waves or
inequalities hi the surface.

A very frequent fault in specifications for asphalt pavements
is that it is provided that the street should be constructed without
sufficient crown. The only objection that can be raised against a
high crown is that the pavement is slippery on the quarters, but
this is a very small objection compared to the fact that flat streets
are unsightly because it is impossible to so grade them as to throw
off all the water and because where water stands in this way it
cannot but have an undesirable effect upon the surface. The
height of a crown which a pavement should have has already been
considered. 1 It may be added that defects due to lack of crown
are more emphasized in careless work than when the pavement is
laid with skilled labor and supervision.

Lack of Lateral Support. Attention has been called to the fact
that a sufficient lateral support, free from vibration, is as essential
as a rigid foundation. An aaphaltic surface cannot be expected not
to deteriorate against a rail which vibrates or against a header
which is not rigid, where the asphalt joins some other form of
roadway.

1 See page 451.

474 THE MODERN ASPHALT PAVEMENT.

Proper provision for avoiding deterioration from these causes
is rarely made and should receive more attention. Along a rail
which shows the least tendency to vibration, paving brick in three
or four rows, all laid as stretchers with broken joints in Portland-
cement mortar, experience has shown is by far the most advanta-
geous form of construction, or, if the asphalt surface must be car-
ried to the rail, it should be supported on an asphalt concrete and
not a binder.

Inferiority of Available Sand. The important role which sand
plays in the construction of an asphalt surface and the great varia-
tions which are met with in the character of this material have been
made plain in preceding pages. It is evident that the sands
available in one city may be far inferior to those found in another,
but this demands only the more care in selecting the best and
using them with the greatest skill. In two western cities it is only
after seventeen years' experience and search for sand that the proper
supply has been found. The great improvement brought about
in the character of the surface mixtures now laid in these cities by
the use of the sand finally selected is most evident and satisfactory.
The result points out the great advantage derived from a thorough
knowledge of the characteristics of various sands and by having in
charge of securing supplies superintendents who are thoroughly
acquainted with the subject. Where the superintendents are
incompetent and do not pay sufficient attention to their sand, the
surface mixtures which they produce are inferior. This is illus-
trated by the mixtures laid by six companies in the city of New
York in 1904, the grading of which is given on the following page
in comparison with that produced under the author's supervision.

It will be noted in the table that the mixture turned out
under the author's supervision in 1904 is not up to the stand-
ard. This is due to the fact that -the available sand supply in
that year was unsatisfactory. The other mixtures are, however,
much more unsatisfactory, and, although they contain in all
cases a sufficient amount of bitumen, they are very deficient in
sand grains passing the 100- and 80-mesh sieves and generally
contain far too much coarse material of 10-, 20-, and 30-mesh
size. Such mixtures, on this account, cannot result in a sur-

DEFECTS AND DETERIORATION.

475

AVERAGE COMPOSITION OF MIXTURES PRODUCED IN NEW
YORK CITY IN 1904 WITHOUT PROPER SUPERVISION OF
THE GRADING.

Com-
pany

Bitu-

Passing!

flesh.

Re-
tained

P No y

men.

200

100

on 10.

11 o%

2
3
4

11.3
10.4
11.8

10.7
9.6
12.2

5
5

4
7
6

18
18
20

11
14
12

11
13
14

14
12
11

13
10
5

10.7

6.3

10 8

8 2

7 l ....

10.9

14.1

1 Author's mixture, 1904.

face which will be impervious to water. It will also be noted
that the percentage of 200-mesh material is lower than in
that which the author supervises, although in two instances it
is above 10 per cent, in two others over 9 per cent. It must
be borne in mind in this connection that the sand in use contains
a very considerable percentage of 200-mesh material, often 6 to
9 per cent. This material is largely sand and does not act as a
filler, so that the deficiency in the above mixtures does not seem
as large as it really is, but they are all of them actually deficient
in filler. In the case of companies 5 and 6 the deficiency is very
large, and these mixtures may be pronounced very inferior on
this account and because this deficiency is accompanied by a
similar one in fine sand and by the presence of a very large amount
of coarse material.

In other cities mixtures have been laid which show even greater
deficiencies, and illustrate very well the inferior character of the
material which is turned out without a thorough understanding
of the principles underlying the production of a standard surface
mixture, and without proper laboratory control. Had the latter
been exercised, the defects in these mixtures would have become
apparent before the material was laid. See table on page 476.

More gross defects in pavements are due to the improper use
of sand than to any other causes except too hard bitumen or weak
foundation.

476

THE MODERN ASPHALT PAVEMENT.

City.

Bit-
umen.

Passing Mesh.

16
1

200

100

Buffalo N. Y

9.2%
10.6
8.8
11.3
9.9
9.9
9.3
10.6
9.9
11.1
8.7
9.5
9.0
12.7
9.8

4.8
8.4
16.2
6.7
4.1
9.1
7.7
6.4
12.1
4.9
8.3
11.5
5.0
5.3
5.2

12
4
16
4
4
5
3
6
18
10
3
4
11
8
10

19
8
30
6
27
12
8
7
30
14
4
5
12
5
14

53
31
22
45
42
44
45
35
28
33
32
32
27
53
35

2
5
2
15
5
10
11
15
2
10
26
23
15
8
14

3
2
9
4
5
6
9

11
12
11
12
4
6

5
1
1
2
3
5
5

2
4
3
5
2
4

9
2
2
1
2
5
6

2
2
1
4
2
2

it tt

Chicago, 111

it ft

tt (t

Cedar Rapids, Iowa.. .
Erie Pa

Long Island City, N.Y.
Louisville Ky

Newark N J.

New Orleans La.. . .

tt a tt

Omaha Neb

Pittsburg, Pa

Toronto, Ont

It may happen, of course, that in some places the highest grade
of asphalt-surface cannot be laid with the available sand supplies
and that their lasting or wearing properties in such cities must,
therefore, be inferior to those which can be laid in others with
more suitable sand.

Character of the Filler. As has been shown 1 the character
of the filler in use in asphalt-surface mixtures is very variable. If
it is coarse and used in insufficient amount the result will be a
decidedly inferior mixture. As an example of this, certain streets
are known to the author, which were laid in the downtown sec-
tion of New York in 1895, with an asphalt surface mixture con-
taining no filler. These streets rapidly lost their shape through
displacement of the surface or lack of stability in the mixture,
and they have long since been resurfaced.

It can be seen, therefore, that deterioration of asphalt pave-
ments may at times be attributed to the lack of filler, to its poor
character or to its unintelligent use.

On street surfaces which are to be subjected to the heaviest

1 See page 89.

DEFECTS AND DETERIORATION. 477

travel the use of Portland cement as a filler has been found to
well repay the extra expense incurred.

Inferiority in the Asphalt or Lack of Intelligence in its
Use. Defects due to the character of the asphalt hi use and
lack of intelligence in handling it are and have been the most
frequent in pavements laid by inexperienced or unintelligent
persons. By a proper combination of different najtive bitumens
of different properties an asphalt cement can be made hi which
more or less of any available kind may form a part, but certain
bitumens require much more skill in handling, while others will
stand much greater abuse. Trinidad lake asphalt has been shown
to be of the latter class, while others, either deficient in hydro-
carbons of the malthene group, or containing light oils volatile
at high temperatures or unsaturated hydrocarbons which readily
become altered in their state of molecular aggregation and con-
sequently in their consistency, are of the class which require skill
and care in their manipulation. Others again necessitate the
use of particular fluxing agents and result in comparative failures
when improper ones are used. These differences have been taken
up in the description of the properties of the several native bitu-
mens.

Asphalt cements made in this way with a flux which is unsuit-
able for the purpose may thus be the cause of failure or deteriora-
tion. Such a cement may contain an excess of paraffine scale,
of light oils, of cracked products, or of unsaturated hydrocar-
bons, which are rapidly converted to pitch on heating. Defects
in asphalt surfaces have been due frequently to such reasons in
the past. They are not as frequent to-day, but public officials
cannot be too careful in determining the quality of the flux hi
use in preparing the asphalt cement with which a surface for
which they are responsible is laid. Large numbers of Bermudez
asphalt pavements laid between 1894 and 1900 were failures
because the asphalt cement of which they were made was not
handled with skill.

Careless Workmanship. Poor workmanship may be due to
either ignorance or intention, and unfortunately it is too often
due to both combined. The careless and irresponsible contractor

478 THE MODERN ASPHALT PAVEMENT.

who looks to immediate profits, who has little experience in the
cost of maintenance of pavements, who does not set aside a cer-
tain amount of money for this purpose or consider it in his bid
for construction, is doing more to discredit asphalt pavements
to-day than any inherent defects in the pavement^ except per-
haps the public officials who will not maintain their asphalt pave-
ments after the expiration of the guarantee period.

Aside from the defects due to the nature ol the asphalt, to
the use of improper sand and the careless regulation of the mineral
aggregate, others are attributable to asphalt cement made, as
has been shown, with an unsatisfactory flux, or to the fact that
it is too hard or too soft, irregular in amount or hardened, burned
as the saying is, by too hot sand. All lack of attention to pre-
cautions for avoiding such defects, which are known to be fatal
to the production of the best surface mixture, may be set down,
largely, to carelessness as well as ignorance.

Ignorance or lack of technical knowledge on the part of the
contractor can be readily learned by inquiry as to whether the
requisite technical supervision is exercised over his work by labora-
tory methods. A high-grade surface, it has been shown, cannot
be laid without such a supervision of all the elements entering
into its construction.

Intentional neglect of the proper construction from motives
of economy can be detected by public officials if they are suffi-
ciently acquainted with the technology of the industry. Unfor-
tunately City Engineers are usually themselves insufficiently
informed to do so, and it is for the purpose of instructing them
that this book has been written. They must, as a rule, depute
any inspection to subordinates, who are even less well informed,
who quibble over small details and miss the important points,
or to experts, men of no wide practical experience but rather
theorists, with one theory one year, another the next, abandoning
an old one for the novelty of the new, but not founding any of
them on more than closet work and experiment, and failing to
look back and draw conclusions of weight from practical results.

The asphalt surfaces which are laid to-day on a rational basis,
under the writer's supervision, are built on no theory but by deter-

DEFECTS AND DETERIORATION. 479

mining from a study of the composition of actual surfaces which
have given the greatest satisfaction what a desirable form of con-
struction is. The manner of working out this problem has been
elaborated in previous pages.

Public officials are advised in determining the character of
the work which is being done by any contractor who employs
no scientific supervision of his process to note:

The number of barrels of cement and the amount of sand and
stone used in a definite area of base.

The consistency of the asphaltic cement and its regularity,
together with the character of the flux used in its preparation.

The character of the sand and its capacity for carrying
bitumen and a proper amount of filler.

The grading of the mineral aggregate.

The regulation of the amount of bitumen in the surface
mixture by means of the pat paper test.

The temperature of the materials.

The skill in handling the materials at the plant and on the
street.

The Manner in Which Defects in Asphalt Surfaces Due to
Faulty Construction are Manifested. Defects in asphalt pave-
ments due to the faulty methods of construction which have
been described, are manifested in several ways.

The surface cracks, but does not disintegrate.

The surface cracks when the lateral support is weak and then
goes to pieces under traffic.

The surface disintegrates in various parts of the roadway,
forming depressions or holes extending to the base.

The surface, when wet, scales off in large thin patches.

The surface is displaced upon the base becoming wavy, high
at one spot and below grade at another.

The surface is raised into waves by expansion of the hydraulic
cement in the concrete base.

Cracking in asphalt surfaces have been found to be due to
many different causes:

Induced by cracks in the hydraulic concrete forming the

480 THE MODERN ASPHALT PAVEMENT.

Produced by too hard a bitumen in the surface mixture, or
by one which is not sufficiently ductile at low temperatures.

Produced by too small a percentage of bitumen in the surface.

Produced by the use of an unsuitable bitumen.

Produced by an unsuitable mineral aggregate.

Produced by lack of compression.

Produced by lack of traffic.

Produced by sudden changes in temperature.

Produced by vibration of rails, manholes and valve-boxes.

Cracking of Asphalt Surfaces. Cracks in the hydraulic con-
crete base are at times reproduced in asphalt surfaces, even when
the latter are of the best quality. The causes of cracks in base of
this description must be referred to defects in the cement of which
it is made, some of them expanding or contracting for some years
after their use. Cracks due to this cause may be directly across
the street or run in zig-zag directions along the crown and else-
where, as shown in the illustration, Fig. 1, where cracks in the
surface have been cut out to show those in the base. This form of
cracking occurs both with natural and Portland cement, and with
the very best surface mixtures under traffic, as well as with inferior
ones under no traffic.

If the cracked portions are renewed after the cement has attained
volume constancy with age and the surface repaved, the cracks
do not return. They are not a common form of defect in an
asphalt surface.

Cracks of the second description, due to the use of asphalt
cement which is too hard or which has become hardened by being
mixed with too hot sand, or to this cause combined with others,
are the form which is most commonly met with. They are fre-
quent in the hard Bermudez pavements laid in the Central States
in 1898 and 1899, where the work was done according to a formula
suitable for the materials available in 1893, but which with changed
conditions resulted in later years in an asphalt cement of great
hardness. Intelligence or proper supervision would have detected
the unsuitable consistency of the cement. The results indicate
the danger of following a blind formula.

Such cracks are of course due to the fact that the hard asphalt

DEFECTS AND DETERIORATION.

481

is too brittle at low temperatures to yield to the contraction of
the surface It fractures under the tensile stress imposed upon it.

An actual measurement of the contraction of an asphalt
surface made by Mr. E. C. Wallace, formerly Chemist of the
Warren-Scharf Asphalt Paving Company, outside the window
of his laboratory during cold winter weather, has shown that above
32 F. it is less than the average contraction of steel, and below
freezing greater. This contraction is about that of quartz, and as
quartz or similar mineral matter forms nearly 90 per cent of the
mixture such a contraction would be expected.

Determinations of the coefficient of expansion of various
materials have been collected in the following tables from the
literature of the subject, and a few determinations made by the
writer are given for that of residuum and asphalt cements.

COEFFICIENTS OF LINEAR EXPANSION FOR 1 C.

Substance.

Temperature.

Coefficient.

Authority.

Quartz, mean

0-100 C.

1,000,010 67

Benoit

~~ < <

0-100 C.

,000,011.80

Pulfrich

Steel

0-100 C

000010 9

Benoit

Petroleum 26 B

0-100 C.

000095

Sharpless

100-101 C.

,000,147

Paraffine, hard

0- 16 C.

,000,106.6

Rodwell

16- 38 C

000 130 3

< <

Beeswax

10- 26 C

000 230

KODD

< <

26- 31 C.

,000312

ixupp

Eastern petroleum:
Residuum, 21 B. . .

14- 27 C.

,000,989

Richardson

n a

23- 38 C

000 838 9

Bermudez asphalt cement

100 asphalt, 20 residuum. . .
< < it

6- 20 C.
20- 45 C.

,000,544
,000,302

tt
tt

The coefficient of expansion of petroleum residuum does
not, like that of most oils, increase with rise in temperature, prob-
ably due to the presence of paraffine, which solidifies at low tempera-
tures and contracts rapidly. The bitumen of asphalt and asphalt
cements contracts or expands in the same way. These results at
first seemed rather startling, but reference to the literature of
the subject confirms them. A paper by Holde in Mittheilungen
der Konig, Technische Versuchsstation, 1893, 45-68, shows that:

482 THE MODERN ASPHALT PAVEMENT.

" The heavy viscous products of distillation or residues from
crude petroleum of different origin, possessing a specific gravity
of at least 0.908, do not sho.w any marked difference in their expan-
sions between +20 C. and 78 C. Their coefficient of expansion
varies from 0.00070 to 0.00072. Those oils holding solid paraffine
suspended at temperatures below + 20 C. (as German oils) have
a higher coefficient of expansion between 18 C. and 20 C., viz.
0.00075 and 0.00081, owing to the melting of the solid particles.

" The heavy liquid products of distillation, of specific gravities
below 0.905, at + 15 C. possess a higher coefficient of expansion
between 20 C. and 78 C., viz. 0.00072 to 0.00076. The American
and Scotch oils belong to this class.

"As to the completely fluid lubricating oils, their coefficients
of expansion rise gradually in proportion to the increase of tempera-
ture."

In an asphalt surface one thousand feet long between 20 F.
and 130 F., extremes of temperature that are met with by Omaha
surfaces, the contraction of the sand alone, forming 90 per cent
of the pavement, would amount to from .902 to .920 feet, or from
10 to 11 inches. The contraction of the bitumen need not be
considered, as this either elongates under the stress, or fractures.
As low as 26 F. the elongation of a bitumen of proper consistency
has been shown by experiments, to be described later, to take
place quite readily. The contraction need therefore be considered
only for the temperature between 26 and -20, 46 F. or 25 C.
For such an interval it would amount to about .29 of a foot per
1000 feet, or about 1 inch in a Fifth Avenue, New York, block. It
is not surprising, therefore, that with a hard cement rupture of the
surface takes place, but rather that it does not always take place.

The above conclusions are based entirely on theoretical consider-
ations. Mr. S. Whinery and Mr. E. C. Wallace, of the Warren-
Scharf Asphalt Paving Company, made some actual determinations
some years ago of the expansion of an asphalt surface mixture on a
laboratory scale. The means of measurement were not more
accurate than 1/1000 inch. Using only those observations where
the difference of temperature was 20 F. or more, and the expan-

DEFECTS AND DETERIORATION.

483

sion or contraction, therefore, so considerable that it could be
measured in this way with a fair degree of accuracy, they arrived
at results which seem to be worthy of some confidence. There
were four series of these observations, and the results are given
in the following table.

COEFFICIENT OF EXPANSION BY HEAT OF SHEET ASPHALT

PAVEMENT. FROM OBSERVATIONS MADE BY LABORATORY

OF THE WARREN-SCHARF ASPHALT PAVING COMPANY.

Series.

No. of Observa-
tions Used.

Mean Coefficient
of Expansion.

A .

0000135

.0000140

C .

0000139

0000131

As the coefficient of expansion of structural steel is about
.0000065, it is nearly correct to say that the coefficient of
expansion of asphalt surface mixture as determined in this way is
double that of steel.

These results are about half of that which is arrived at theo-
retically. What weight can be given to the results is some-
what doubtful, but they are quoted here for what they are
worth.

Cracks which are due to the fact that the mixture is deficient
in bitumen, in consequence of which the surface does not possess
sufficient tensile strength, regardless of ductility, at low winter
temperatures, are not as frequent as those due to a hard bitumen,
since in such a mixture, disintegration with the formation of
holes takes place, as a rule, before cracking.

Cracks may be caused by the use of an asphalt cement which
is unsuitable for the purpose to which it is applied. It may be
too susceptible to temperature changes, so that, even if made so
soft that the surface marks badly under a summer sun, it may
be brittle at zero.

484 THE MODERN ASPHALT PAVEMENT.

Finally, an asphalt cement may so harden with age that it
becomes brittle in the course of a few years. Coal-tar is an
example of such a material.

Cracking may be caused even with the most satisfactory asphalt
cement by an unsuitable sand or mineral aggregate. Sands are
known and have been used, the surface of the grains of which
are of such a nature that melted asphalt cement will not adhere
to them in sufficient thickness, and the voids in which are so small
as to prevent the mixture from holding enough bitumen to give
the pavement ductility. The sands available in other cities,
without appreciable difference from those in use elsewhere, pro-
duce a surface which never cracks, even under unfavorable con-
ditions and inferior workmanship.

Too fine a mineral aggregate may be a disadvantage on streets
of little or no traffic.

Lack of density in the surface also favors cracking, whether
brought about by insufficient compaction when the surface is
laid or subsequent lack of traffic.

Traffic and the lack of it play a large part in preventing or
causing the cracking of pavements.

Traffic releases tension to a large degree in a cold asphalt sur-
face and assists elongation of the bitumen, so that heavy-traffic
streets do not crack as readily as those of light or no traffic, and
oftener crack only in the gutters, if at all, while light-traffic streets
crack entirely across the roadway.

Suburban streets, not benefited by traction, at least in certain
cities, are much more liable to crack than those which have a
medium traffic. This is particularly well illustrated in a Canadian
city where surfaces with only 8 to 9 per cent of bitumen are free
from cracks on the downtown streets but are a mass of cracks
in the suburbs, the bitumen present not being sufficient to give
any elasticity unless the tension produced by contraction is
released by traffic.

Climate, of course, plays a large part in deteraiining the fre-
quency of cracks in asphalt surfaces. Mixtures of almost identical
composition will fracture under the conditions met with in one

DEFECTS AND DETERIORATION.

485

city and not in another. In those cities in the Missouri valley
where sudden changes in temperature reaching 60 in a few hours,
from 40 above zero to 20 below, cracking frequently results,
where the same changes occurring more slowly in more protected
locations do no damage.

Cracking along rails and around boxes and manholes is due
to lack of support and may occur with the best mixtures. The
causes have been considered elsewhere.

In all of the causes of cracking which have been cited, except
the last, laboratory investigations have thrown some light on the
subject and the results obtained are of interest.

Strength of Asphalt Surfaces. An asphalt surface having the
least ductility and tensile strength will, of course, rupture most
readily under the tensile stress produced by contraction due to
a fall of temperature. The tensile or crushing strength of an
asphalt surface is, of course, a function of the temperature being
greater at low than high temperature. Following are illustrations:

FIFTH AVENUE, NEW YORK. LAID IN 1897.

Temperature.

6 F.

38 F.

76 F.

Pounds per square inch:
Tensile strength

880

568

300

Crushing ' *

4862

2836

1820

In considering the subject of cracked pavements our interest
is entirely in the strength and ductility of the surfaces at low
temperatures.

From a great many old surfaces, some of which had cracked
and some of which were free from them, briquettes were made
and broken at 6 F. Averages of these determinations for the
cracked and good surfaces in two cities are as follows:

486 THE MODERN ASPHALT PAVEMENT.

TENSILE STRENGTH IN POUNDS PER SQUARE INCH AT 6 F.

No. 1.

Cracked pavements 664 (6)

Good " 722 (2)

No. 2.

Cracked pavements 497 (2)

Good " 614 (6)

There is a striking difference in the strength of the cracked
and the good pavements in both cities in favor of the latter.

It is of interest in this connection to know what peculiarities
contribute to the strength of asphalt surfaces. Experiments have
shown that the principal conditioning elements are:

Asphalt Cement.

Character.

Consistency.

Amount.
Filler.

Amount.
Sand.

Grading.
Density of the surface.

This subject was looked into to a considerable extent by the
writer in 1894. Unfortunately this work was done with the
old coarse Washington surface mixture. With the modern well
graded New York material more satisfactory results would now be
obtained. The experiments suffice, however, to bring out several
points. Following are the available data. See table on page 487.

These results show that the character of the cementing material
has a very decided influence on the crushing, and it would also be
found to be the same on the tensile strength of the surfaces. This
subject was thoroughly discussed by the writer in a letter in the
Engineering News for June, 1894, and it need only be said here
that the strongest mixture is not the best, without regard to the
nature of the bitumen, but that, with a proper cement, weakness

DEFECTS AND DETERIORATION.

487

EFFECT OF THE CHARACTER OF THE BITUMEN ON THE
CRUSHING STRENGTH.

CRUSHING STRENGTH OF MIXTURES OF COAL-TAR, TRINIDAD LAKE AND

LAND, BERMUDEZ, AND PEDERNALES ASPHALT.

POUNDS PER SQUARE INCH.

Mixture.

Density.

At 38 F.

At 77 F.

Coal-tar 15%. . .

2.16

3880

1254

" 10%

2.07

3845

2655

Land pitch cement 15% or 10% bitumen

2 13

1813

761

Bermudez cement, 10% or 10% bitumen

2.10

1955

635

Pedernales asphalt, 10% or 10% bitumen

2 06

2125

550

Lake pitch cement, 15% or 10% bitumen

2.14

1375

548

Ten per cent of dust and Washington sand in all mixtures.

at low temperatures due to ductility or elongation is preferable
to high strength; weakness due to lack of bitumen, on the contrary,
is not.

EFFECT OF THE CONSISTENCY OF ASPHALT CEMENT ON
STRENGTH OF SURFACES.

WASHINGTON MIXTURE, 1894, POUNDS PER SQUARE INCH.

Softness.

Crushing Strength.

Shearing Strength.

36 F.

77 F.

36 F.

77 F.

Trinidad cement:
Normal consistency. . . .

1703
1610

2416
3028

680
682

741
602

2865
2005

3193
2569

1425

1873

1528
1876

Softer cement 3 Ibs. more oil. .

Bermudez cement:
Normal consistency

Softer "

These results show that a softer Trinidad cement makes a
mixture which has, owing to its great ductility, a smaller crushing
strength at 36 than the one made with a hard cement; but with
Bermudez cement this is not the case, as this cementing material
is more easily affected by a fall of temperature.

With a better sand grading, however, the above results might
be somewhat modified.

488

THE MODERN ASPHALT PAVEMENT.

EFFECT OF THE QUANTITY OF ASPHALT CEMENT ON

STRENGTH OF SURFACES.
WASHINGTON MIXTURE, 1894, POUNDS PER SQUARE INCH.

Crushing

Strength.

Shearing

Strength.

36 F.

77 F.

36 F.

77 F.

Trinidad cement :
Normal 15%. . .

1703

880

2865

1425

More cement 16 5%.

2425

768

2882

2378

Bermudez cement:
Normal 10%

2416

741

3193

1528

More cement 11%

2544

961

2511

1636

These results show that an increase in the amount of asphalt
in a mixture, up to a certain point, increases the strength of a
Trinidad mixture in all cases; but, as before, not always, with
Bermudez asphalt due to differences in the physical properties of
the two bitumens at low temperatures.

EFFECT OF INCREASE OF AMOUNT OF DUST IN SURFACE

MIXTURES ON THEIR TENSILE STRENGTH.

POUNDS PER SQUARE INCH.

Dust
per Cent.

Trinidad
at 36 F.

Bermudez
at 36 F.

Trinidad
at 77 F.

Bermudez
at 77 F.

7.5
10.0
15.0
20.0

501
604
646
701

449
611
662

857

188
205
273
270

Ill

171
186
192

The addition of increased amounts of dust gives decided evi-
dence of improvement of the mixture in all cases, and shows the
necessity of using plenty of filler.

EFFECT OF DENSITY ON STRENGTH OF SURFACES.

Tensile Strength at

Compaction.

40 F. | 77 F. | 90 F.

Density.

Pounds Per Square Inch.

Least dense

463
646

152
273

101
166

2.08
2.23

Densest

DEFJECTS AND DETERIORATION.

489

The above results show that there can be no doubt that the
densest mixtures are the strongest, at least with the same mineral
aggregate and a sufficient amount of bitumen.

EFFECT OF SAND GRADING ON STRENGTH OF SURFACES.
WASHINGTON AND NEW YORK.

Composition.

Bitu-
men.

Passing.

200

100

New York..
Washington.

10.6
10.5

14.4
9.7

11.0
3.2

12.0
5.4

27.0
22.3

11.0
20.5

7.0
13.2

4.0
7.8

3.0
7.4

TENSILE STRENGTH, POUNDS PER SQUARE INCH.

At 38 F.

At 78 F.

New York

568 Ibs.

300 Ibs.

^ r Lshinfirton

604 "

205 "

The difference between the fine and coarse mixture at 36 F.
is slightly in favor of the coarse; at 78 F. in favor of the fine.

It must be remembered, however, that in these tests there
are a number of conditions beside the grading of the sand that
enter into the problem, so that final conclusions can hardly be
drawn from so few experiments.

As a whole the results of these physical tests throw considerable
light on the peculiarities of mixtures of varying composition and
give us some information as to why some crack and others do not.

The possibility of cracking in asphalt surfaces are seen to
be large, and it is remarkable how well they have been overcome
by intelligent study of the conditions that are to be met. There
is still much to be learned in this direction. It is impossible, as
yet, to say why cracking has never occurred in pavements, even
when not laid with the greatest care, in one city, while they are
of general occurrence in another where the greatest care is exer-
cised. It must, of course, be due to peculiarities in the surface
of the grains of the sands in use, and the relative degree of
adhesion of asphalt to them.

490 THE MODERN ASPHALT PAVEMENT.

Cracks are never known to heal. Experiments have shown
that an asphalt contracts longitudinally ; but expands vertically,
BO that cracks once formed increase in width every winter, not
only for this reason but because they become filled with dirt.

Cracks may be merely unsightly or they may be the essential
cause of subsequent disintegration.

Asphalt surfaces on suburban streets in some climates crack
to a marked degree after about three years service ; but if the sur-
face mixture is a good one, no disintegration follows and the pave-
ment continues to be satisfactory in every other respect for as
long a period as if no cracks existed. If disintegration takes
place in such cases it is due to inferiority hi the character of the
mixture. There need be no alarm if no disintegration sets in.
If local prejudice against a very soft surface which marks excessively
when first laid does not exist, cracks may be largely avoided by
using a very soft asphalt cement in the surface. In a north-
western city, where no asphalt surface had ever been laid on a
residence street without cracks appearing hi a few years, this was
avoided in some laid by the writer by using a cement of 90 to
100 penetration insead of one of 65, as had been previously the
case. The surface marked up under traffic excessively, however,
during the first summer and aroused much comment. Com-
munities soon become accustomed to this and the marking in
new surfaces is objected to no longer, as it is understood that the
pavement will eventually be a superior one. In consequence,
much of the cracking of asphalt surfaces can now be avoided
if they are originally laid with sufficiently soft bitumen, and the
reason for the ensuing marking is properly explained to the public.

Disintegration. Disintegration of the surface in various parts
with the formation of depressions or holes extending to the base
is the commonest defect in asphalt surfaces. Many defects of
this kind are attributable to faults of construction, but they may
also be due to unfavorable environment with the best of surface
mixtures. Altogether they may be summed up as:

Deteriorations or defects due to:

Weak foundation an extremely common cause.

Inferior mixture.

DEFECTS AND DETERIORATION. 491

Action of illuminating-gas.

Action of water.

Uneven thickness of surface and constant pounding on depres-
sions in an unbalanced mixture.

Weak Foundation. An asphalt surface cannot resist the impact of
traffic if the foundation does not furnish adequate support. This, as
has been reiterated many times in these pages, is one of the most seri-
ous causes of the deterioration of asphalt pavements in large cities
and where they are subject to heavy traffic and moisture. Any
vibration in the surface, especially under unfavorable surface con-
ditions such as dirt and continued moisture, is extremely liable to
result in deterioration of a poor mixture and will aid in destroying
the best surface. A defect due to weak foundation may be manifest in
many ways. The surface may merely break up and go to pieces,
or it may at first merely separate into small individual masses
which become rounded at their edges and form a collective group
of almond-shaped patches, which eventually go to pieces with a
resulting hole.

Inferior Mixture. Disintegration due to inferior mixture has
been too thoroughly discussed in previous pages to necessitate
a recurrence to the reasons therefor. It is, of course, in careless
work the chief cause of defects in asphalt surfaces, but too often
disintegration is attributed to a poor mixture which is due entirely
to other causes.

Action of Illuminating-gas. The disintegrating effect of the
action of illuminating-gas is a subject which has not been consid-
ered hitherto in these pages. The writer cannot do better than
by allowing Mr. A. W. Dow, his successor in the Office of Inspector
of Asphalt and Cements, in the District of Columbia, to speak of
his experiences in Washington with this cause of deterioration of
asphalt surfaces, as all that he says applies as well to other cities.
He writes as follows in his report to the Engineer Commissioner
of the District of Columbia, for the fiscal year ending June 30,
1899:

"Disintegration of pavements from the absorption of illuminat-
ing-gas, escaping from leaky gas-pipes or mains under the pave-
ment: There are several streets in the city being ruined by this

492 THE MODERN ASPHALT PAVEMENT.

means, and it appears to be a common thing in all cities having
gas. The pavements are affected in very much the same way as
when disintegrated by coal-tar binder, except the fine cracks,
running parallel with the street, make their appearance sometime
before the pavement begins to crowd. Pieces of the surface
mixture taken up smell very strongly of illuminating-gas, and in
some cases the gas can be ignited by applying a match to the under
surface when it has just been taken up. In nearly every case
enough gas will be given off by heating a small piece of the affected
pavement in a tube to have it flash by igniting.

" As it has been doubted by some that this disintegration is
really due to illuminating-gas, I have made a most thorough investi-
gation of the subject and believe have positively proven that
gas is the cause. Samples of pavements were obtained from
several affected spots, and in all cases I have been able to obtain
from them a gas that exploded by passing an electric spark after
mixing with ah*. The method employed to obtain the gas from
samples of the surface mixture was by heating them under boiling
water and collecting the gas given off in an inverted funnel. Those
not acquainted with the properties of asphalts might suggest that
heating any asphalt to this temperature might make it give off a
gas. This is impossible, as an asphalt cement such as is used in
paving will lose only 3 or 4 per cent at the most on being kept at
a temperature of 400 F. for 30 hours, and only an infinitesimal
part of this loss is a gas at ordinary temperatures. To make a
more practical demonstration of this, two samples of a pavement
were taken, one from an affected spot and the other from a good
portion of the pavement about 10 feet away. These samples
were treated under boiling water until they ceased to evolve gas.
The affected sample gave several times more gas than did the other.
On testing, the gas from the good sample was found to consist of
oxygen and nitrogen, which was evidently just the air from the
voids of the pavement. The gas from the affected piece gave on
analysis:

DEFECTS AND DETERIORATION. 493

Carbon dioxide 8.4%

Oxygen 10.8

Heavy hydrocarbons 13 . 4

Carbon monoxide 0.7

Hydrogen 6.6

Methane .2.0

Nitrogen 58 . 1

" Having now found that a gas is present in the pavement so
affected, let us proceed to examine as to its source. It cannot be
a natural gas or marsh-gas, for there is no analysis of such gases
on record that contains appreciable amounts of heavy hydrocar-
bons, while the gas from the pavement is rich in these compounds.
The same would also apply to sewer air or gas. The only
remaining source is illuminating-gas, the analysis of which is
here given:

Carbon dioxide 0.2%

Oxygen 0.0

Heavy hydrocarbons 12.1

Carbon monoxide 25 . 5

Hydrogen ! 39.2

Methane 23 .0

Nitrogen 0.0

" On comparing the composition of the gas given off from the
disintegrating pavement with the illuminating-gas it is seen that
they are not at all similar in composition. At first glance it would
not seem possible that the former gas could originate from the
latter, but when the properties of asphalt are considered it is easily-
explained.

" Heavy hydrocarbons, to which class asphalts belong, are
known to absorb other gaseous hydrocarbons; the heavier the gas
the more affinity between it and the heavy hydrocarbons. Know-
ing this, the ingredients of the illuminating-gas that asphalt would
have the greatest affinity for would be the heavy hydrocarbon
gases, a slight affinity for the marsh-gas or methane, and no affinity
for any of the other ingredients. If we examine the ingredients
of the gas from the affected pavement, it will be found to consist
of some carbon dioxide, ah- that was in the voids and cracks of the
pavement, and the constituents of illuminating-gas with the heavy

494

THE MODERN ASPHALT PAVEMENT.

hydrocarbon gases very much in excess, which is what we would
expect. To practically demonstrate that the above takes place
when asphalt is in contact with illuminating-gas, I took two samples
of gas from a tap in the laboratory. One was analyzed, while the
other was kept for several weeks in a tube the interior of which
was coated with asphalt cement such as is used in pavements,
after which it was analyzed. The results of the two analyses are
here given :

Original
Gas.

Gas after
Asphalt
Absorption.

Carbon, dioxide ...

Oxvsren .

12.1

7.2

25.5

27.3

39.2

42.2

Methane

23 2

Nitrogen . .

" It is evident from this that the asphalt cement has absorbed
over 5 per cent of the heavy hydrocarbon gases, a little methane,
and practically nothing else.

" I have ascertained by experiment that one part by volume
of asphalt cement will absorb forty-two parts of illuminating-
gas in somewhat over a month. I have also practically shown
that asphalt is much softened by asborbing gas, the ordinary
asphalt cement becoming as soft as a thick maltha after being in
an atmosphere of illuminating-gas for several months. As to the
quantity of gas contained in the affected pavements this of course
varies, but in one instance 1000 c.c. of pavement gave off 500 c.c.
of gas.

" There is but one way to stop the disintegration of a pave-
ment from this cause, and that is to stop the leak of gas; for it
is useless to patch the pavement, as it will not be long before
the patch disintegrates. I have known of cases where a pave-
ment so affected was repaired and in fourteen months the patches
were showing signs of disintegration."

The writer's investigations have in every respect confirmed the
conclusions of Mr. Dow.

DEFECTS AND DETERIORATION. 495

Water Action. The action of water on poor and unsatisfactory
asphalt surfaces has also been considered at length. Continued
standing or running water will destroy the best asphalt surface, but
such a defect is one which can be avoided by proper provision for
the prevention of such conditions. The best surfaces will resist for
years any reasonable water action. Unfortunately provisions for
preventing such action are not always adequate either from the
presence of a porous base, seepage from soil in terraces above the
level of the base, or poorly arranged grades to the actual surface
of the pavement, which permit water to stand in the gutters. These
defects, not inherent in the pavement but merely in the mode of
construction, can be readily provided against.

Of the results of the action of water reaching the asphalt through
a porous base, Mr. A. W. Dow writes as follows:

" Disintegration by Water Entering a Pavement by Oozing up
Through the Base. I believe if more thorough investigation were
made into the cause for the disintegrating of pavement, this would
be found to be one of the most common, especially in small towns
and cities where there are terraces or considerable lawns in front
of houses. There have been a number of cases in this city where
the water has entered a terrace or parking where they were above
the grade of the street and worked its way up through the con-
crete base to the asphalt surface.

" This disintegration manifests itself differently, depending
on the character of the pavement. If the asphalt surface is soft
or the concrete smooth, the first defect noticed will be the ten-
dency of the pavement to crowd in warm weather. This is due
to the under portion of the surface mixture rotting, so to speak,
thus destroying the cementing properties of the asphalt. The
upper portion, although good, being deprived of the support of
the affected mixture under it, will be crowded out by traffic. This
crowding is assisted by the concrete base being smooth, and also
the bond between the base and binder are destroyed by the moisture.

" In cases where the concrete base is rough and the surface
mixture hard, the principal disintegration will take place in cold
weather, nothing abnormal being noticed until the pavement
begins a rapid crumbling away in the affected spots under traffic.

496 THE MODERN ASPHALT PAVEMENT.

" On examining a section of asphalt surface disintegrating
from this cause, especially where it has not been going on for
too long a time, there will be found a layer of perfectly sound
and good material at the surface of the pavement, while under-
neath the mixture will show evidence of being disintegrated by
water that is, the sand will appear clean and white in spots, as
though there had been an insufficiency of asphalt cement to cover it.
The concrete base under the affected pavement will generally
be found damp or even wet. We have prevented the destruction
of several pavements from this cause by the use of blind drains
put in under the gutter next to the lawn or terrace, and even
run herringbone under the pavement.

" This last cause for disintegration would, of course, not occur
in a pavement constructed with an asphalt that was unacted
on by water, but water soaking up through a concrete base might
injure any pavement by freezing.

" It is always advisable where a pavement shows signs of dis-
integrating to examine into the cause in a most careful manner
and not pass snap judgment. It seems only too easy for the
majority of people, whether experienced or not, to place the blame
for the failure of a pavement on the manufacturers. I have heard
men with considerable experience, commenting on a bad place
in a pavement that they had not carefully examined, remark,
' They used bad oil or asphalt in that piece of work.' They have
not taken into consideration that all but possibly a square yard
of the pavement is in good condition and that it would be no
economy to a contractor to use bad material in one small place
of the pavement. A careful examination into the disintegrating
of a pavement may, in many cases, show a cause that is entirely
foreign to the composition of the materials and a cause that could
be easily remedied with a little common sense."

The conclusions of Mr. Dow in regard to such causes of dis-
integration are most reasonable.

Poor Workmanship. The raking of the hot asphalt mixture
to an uneven thickness before compression, resulting in the forma-
tion of depressions in the surface under the final compression
obtained from traffic results in disintegration of the less satisfactory

DEFECTS AND DETERIORATION. 497

mixtures owing to the constant impact of the wheels of vehicles
suddenly dropping into such depressions. This is a fault often due
to carelessness in construction and is not a common one.

Scaling of Asphalt Surfaces. Scaling of asphalt surfaces has
been in individual cases a serious cause of the deterioration. It is
something which happens only in moist climates, particularly in
those near the seacoast, or where fogs are prevalent, and where it
is the custom to water streets continually without the removal of
the accumulated dirt. It is particularly frequent with coarse
mixtures, and hi order to avoid it the grading of the sand, the
character of the filler, the character of the asphalt and flux in use,
their proper combination, together with the support of the resulting
surface on a base free from vibration, must receive the most care-
ful attention.

That the problem can be successfully met has been proved by
the fact that pavements which have not suffered from scaling
have been laid on Broadway and Fifth Avenue in New York, and
in the foggy and damp climates of London, Glasgow, and Paris,
in the first of which cities the successful application of the modern
asphalt surface mixture was worked out in 1896 after two failures,
to be equally successfully followed by similar results later in Glas-
gow and Paris.

Scaling is characterized by the separation under traffic, when
the streets are wet and the ah* so humid as to prevent their drying,
of a thin film of the asphalt mixture from the surface of the pave-
ment. No satisfactory explanation of this phenomena has been
advanced. We must content ourselves with noting its occurrence.

After drying out the asphalt surface resumes its normal appear-
ance, rolling out smoothly under traffic, but the thickness of the
pavement is decreased. The same thing will happen again under
like conditions until a depression or hole is worn and repairs are
necessary.

Enough is known, as has been said, to show that the weaker
poorly graded mixtures lacking in bitumen and made with unsatis-
factory asphalt cement suffer most from scaling. It is also much
more in evidence where the base is weak. Mixtures having a filler
of Portland cement are the most resistant, and finally it never

498 THE MODERN ASPHALT PAVEMENT.

occurs in a pavement thoroughly well constructed, such as that
on Fifth Avenue in New York, or those properly laid in London
and Paris; that is to say, it can be avoided entirely if the proper
precautions are used. The watering-cart and lack of cleanliness
are great aids to the production of scaling and, in fact, to the
general diminution of the life of an asphalt pavement, but this is
a condition the consideration of which must properly be taken up
under the head of the effect of environment upon such surfaces.

Displacement of Asphalt Surfaces. The displacement of
asphalt pavements under traffic resulting in a rolling or wavy
surface was a serious defect in the early days of the industry.
It was due to the fact that the asphalt mixtures were unbalanced,
the mineral aggregate was not properly graded, and the bitumen
was present either in too small or too great an amount, with the
result that the surface, not having sufficient internal stability,
moved upon the more or less smooth hydraulic base, to which it
was not tied by any intermediate course, under the pressure and
impact of traffic. It has even moved in cases where such a course
exists where the stability is unusually small. Defects of this
description, which were at one time common, have been avoided
not only by the introduction of a binder or paint course, but also,
in the few surfaces constructed in recent years without such a
course, by the greater stability of the surface mixture.

Displacement of the surface and formation of waves in asphalt
surfaces do, however, occur occasionally at present, and this
has been found to be due to the percolation of water through an
open binder course, and its attacking the under surface of the
wearing surface of the pavement, thus reducing its stability and
permitting of the movement of the upper portion of the surface
upon the loosened lower part. This occurrence is particularly
noticeable where a pavement of this form of construction adjoins
a street railway track which is not stable, vibrating and admitting
water between the rail and the pavement. Such occurrences
bring out very strongly the injury to all forms of pavement, of
?.n unsatisfactory track construction, and of the difficulty of
*naintaining any pavement under such conditions.

DEFECTS AND DETERIORATION. 499

Expansion of Cement in the Foundation. The surface of an
asphalt pavement has been at times raised transversely into waves
by the expansion of the cement used in the construction of the base
with the direct result of raising the latter at points of least resist-
ance generally at joints between different days' work and the
immediate elevation of the asphalt surface above these points.
This has occurred with both natural and Portland cement, but
usually with magnesian cements of the former type. When the
expansion has ceased after the lapse of several years, removal of
the excess of base and replacement of the surface at a normal
grade obviates any further trouble.

Deterioration of Asphalt Pavements Due to Environment.
Deterioration in asphalt surfaces is brought about, even in those
of the best form of construction, or to a much greater degree, of
course, in those which are poorly constructed, by the nature of
the environment to which they are subjected.

Difficult climatic environment is something that cannot be
escaped and must be met as well as possible by the form of con-
struction of the pavement and by the character of the surface
mixture with which the pavement is constructed. That this
must be accommodated to the climatic conditions which it is
to meet is apparent and is generally understood, at least as far
as temperature in its relations to latitude is concerned and with
reference to sudden falls in temperature. The unfavorable envi-
ronment due to prolonged humidity is, however, the most serious
condition to be met, especially where the traffic is heavy, the sur-
face not kept clean, and the air temperature for long periods lying
between freezing and 45 F. As examples of the former con-
dition might be cited the climate of St. Paul, Omaha, and New
Orleans.

In the latter place it is very necessary to make the surface
with a bitumen sufficiently hard in consistency not to prove unde-
sirable in the summer months, as the temperatures in winter are
not low enough to produce cracking. For this purpose a cement
of 50 penetration on the Bowen machine is usually employed.
In St. Paul, on the other hand, where the winter temperatures
are very low a penetration of 90 is used. A cement of this degree

500 THE MODERN ASPHALT PAVEMENT.

of softness will naturally result in a pavement which marks up
to a notable extent in the summer when first laid, but this dis-
agreeable feature disappears after the second winter, and such a
surface does not crack as would those laid with a harder cement.

A still more difficult climatic feature to meet is that of sudden
drops in temperature, often as much as 50 in a few hours, which
are met with in cities like Omaha. The immediate contraction
caused by such a drop is so great as to overcome the elasticity
or ductility of the bitumen, and cracking can only be prevented
by using in the mixture as it is originally laid a very soft asphalt
cement. An example of such a surface, that laid on Thirty-ninth
Street in Omaha, Neb., will serve. It was so soft when it was
completed and marked so freely that it was not at once accepted
by the city. To-day it is the only pavement of its age in that
city which has not cracked and now does not mark exceptionally
under the hottest summer suns.

Experience has taught how these difficulties may be met with
in the manner described, but pavements constructed by an inex-
perienced contractor, with an unbalanced mineral aggregate which
will not permit the use of cement of sufficiently soft consistency,
will inevitably show cracks in a colder climate in the course of
two or three years.

A still more serious climatic condition to contend with is that
met with in climates where there is excessive humidity in the winter
months. Where such a condition exists only the most carefully
prepared surface mixture will resist the combined action of moisture
and heavy traffic. This was well illustrated in the earlier attempts
to lay asphalt pavements in London, Glasgow, and on the north-
western Pacific Coast. It is even met with in some of our cities
on the Atlantic Coast where asphalt pavements are placed on
very heavy-traffic streets. Much of the earlier scaling in New
York City was due to humidity combined with temperatures
between 45 and the freezing-point. Below a freezing tempera-
ture scaling and disintegration due to this cause does not take
place.

The deterioration of an asphalt pavement caused by the unfavor-
able environment produced by the leakage of coal-gas from gas-

DEFECTS AND DETERIORATION. 501

mains is an important one and has already been discussed under
the heading " Disintegration."

Attention has already been called to the fact that asphaltic
paving mixtures will not withstand the constant action of ground
or running water, and where they are subjected to such an environ-
ment they will inevitably deteriorate more rapidly than is necessary.
The remedy for defects due to the constant action of water is
the removal of the cause by the introduction of proper provisions
for drainage.

Asphalt pavements also suffer in one or two of our cities from
flushing with water under a very considerable head with a hose
and nozzle. No surface of any description can be expected to
withstand such hydraulic mining. If it is carried on the city
must expect to have the cost of maintenance of its asphalt streets
much increased at the end of the guarantee period, and this will
be the greater the more inferior the asphalt surface mixture is
in the beginning.

A still greater cause of deterioration of asphalt pavements
is found in the lack of cleanliness and general neglect. If the
pavements are not carefully cleaned and filth is allowed to lie
upon the surface for a great length of time, becoming mud as
soon as they are sprinkled or rained upon, the deterioration is
very rapid and even worse than when they are subjected to the
action of clean water. Permitting mud and slime to remain upou
an asphalt surface displays great ignorance, upon the part of pub-
lic officials of the nature and behavior of asphalt pavements and
should never be allowed to take place. For the same reason
asphalt pavements should never be sprinkled if possible/ The
dirt should be removed and the situation not temporized with it
by converting it into a slimy mud. This is doubly the case since
such a slimy coating results in making the pavement extremely
slippery, a feature not inherent in the asphalt surface itself, but
attributable only to the film of mud.

Finally, asphalt surfaces in cities like New York, suffer enor-
mously from the constant disturbance to which they are subjected

502 THE MODERN ASPHALT PAVEMENT.

by being taken up in connection with underground work. Though
an asphalt pavement can be repaired more readily than any other,
its constant removal and renewal cannot but injure it, as the
foundation in such cases is never as rigid as the original. It can
be asserted without doubt that not only asphalt pavements, but
all others in a large city, can never be kept in a state of perfect
maintenance until pipe galleries are provided under the streets
which shall do away with the constant opening of the surface.
In some of the streets of New York, from one-third to more than
the entire area of the street has been often renewed in the course
of the guarantee period for this reason.

Deterioration Due to Natural Wear and Neglect of Mainte-
nance. An asphalt surface is naturally more or less deteriorated
by usage, like all materials of construction, and the amount which
it suffers in this respect depends entirely upon the character of
the original workmanship and the traffic and other conditions to
which it is to be exposed. Some asphalt pavements under
light traffic, such as that opposite the Arlington Hotel in
the city of Washington, have given good service for 30 years
and may be expected to last much longer. On the heaviest
traffic streets constructed with the greatest skill some minor
repairs may be expected at the end of from 3 to 5 years, depend-
ing upon the rigidity of the base which supports the surface
and upon the manner in which these repairs are made, the
extent of its deterioration will largely depend. The question of
maintenance will be discussed in the following chapter.

SUMMARY.

To the general reader the preceding chapter will probably be
one of the most interesting and instructive in the book, and it should
be read in detail, as it explains the reasons for defects in and the
causes of the deterioration in asphalt surfaces. The chapter may be
summarized briefly as follows :

Defects in asphalt pavements are, to the greatest extent, to be
attributed to faults of construction.

DEFECTS AND DETERIORATION. 503

1. Due to improper specifications for the form of construction,
the fault of the city officials.

2. Due to careless construction on the part of the contractor,
and also

3. To improper maintenance when the age of the pavement is
such that it should be given careful attention, as unfortunately
the American public and many city officials seem to believe that
when a street is once paved it should be expected to last forever
without maintenance.

4. The action of illuminating-gas escaping from the mains.

5. Constant opening for underground work.

CHAPTER XXVI
MAINTENANCE OF ASPHALT PAVEMENTS.

THE question of the proper maintenance of asphalt pavements
demands, but has not received, very careful consideration on
the part of most of our municipalities and municipal officials,
but it fortunately is receiving more to-day than in the past. Owing
to the fact that asphalt pavements have generally been laid under
guarantees by the contractor for their maintenance for a con-
siderable period, a custom which, as has been said, 1 is rapidly
disappearing, too little consideration has 'been given to their
maintenance after its expiration. As a result of this, they have
deteriorated very rapidly when thrown upon the hands of the
municipality, and their condition has become such as to cause
very unfavorable comment and great dissatisfaction. In a few
cities, such as Washington, D. C., it has been demonstrated that
there is no difficulty in maintaining an ordinary asphalt street
in good condition for from 15 to 20 years, at a moderate cost,
as has been shown by Capt. H. C. Newcomer in a report published
in the Engineering News for February 18, 1904, where he shows
that of the 2,425,732 square yards of bituminous pavements
maintained by that city, of which not less than 2,161,181 square
yards are laid with Trinidad Lake asphalt, the cost of main-
tenance was as follows, the average age of the surfaces being
about 14.8 years, while there are over 700,000 square yards that
are over 18 years of age. He also shows that the average age

1 See page 445.

504

MAINTENANCE OF ASPHALT PAVEMENTS.

505

of the areas resurfaced during the fiscal year ending July 1, 1903,
was 21 years, and this may be regarded as well within the limits
of the duration of a standard asphalt pavement, if it is properly
maintained during the period, especially as the older mixtures
laid in Washington were by no means up to the standard of ex-
cellence of those which are now being put down.

tlOST OF MAINTAINING ASPHALT PAVEMENTS OF VARIOUS
AGES AT WASHINGTON, D. C.

Age in Years.

Area,
Square Yards.

Cost of Repairs
for the Year.

Average Cost per
Square Yard
per Year.

1,841,435

$11,897

$0 0065

1,809,869

13,965

0077

1,747,461

31,385

0180

1,653,811

38,531

0233

1,597,313

42,871

.0269

1,476,575

38500

0260

1,1|2,200

43,003

0333

1,068,848

42,270

.0396

913,795

31,546

.0345

804,420

28,435

.0354

698,826

21 576

0309

608,117

23,479

0386

560,823

18,913

0338

504,995

23,012

0456

374,800

11 951

0319

272,040

7,182

0264

192,643

3,879

0201

104,001

2,887

0280

36,332

678

0187

35,647

1,268

0356

Neglect of maintenance, however, has resulted in quite a
different condition in many other cities. There are one or two
cities in the United States where, at the expiration of the guarantee
period, no attempt is made at further maintenance, and, as a
result, the asphalt pavements in these cities in a few years are
in a wretched condition, arousing comment and adverse criticism
of this form of pavement on the part of all the citizens. This
can in nowise be attributed to the character of the pavement
itself, but to the narrow policy pursued by the public officials in
charge of the streets. Nothing can be so far from economical

506 THE MODERN ASPHALT PAVEMENT.

as to allow an asphalt pavement to go without repairs when they
are needed, as, in this case as in all others, a ' 'stitch in time saves
nine."

On the other hand, there are numerous cities, such as Phila-
delphia and Washington, which have met with no difficulty
whatever in maintaining their asphalt pavements in good condition
under the contract system, a proper amount of funds being made
available in the spring of the year. In at least one other city
where the money is not available until mid-summer, July 1st,
the contract system has not proved as desirable. In several cities,
the failure of the contract system has been due more to the pro-
miscuous way in which cutting up of the pavement is permitted
by the city authorities, than to the manner in which repairs are
made.

This situation has been discussed, as far as maintenance is
concerned, in an article by Mr. S. Whinery, in Engineering News
for May 9, 1907. He says in psfk:

" It cannot be denied that the condition of asphalt pavements
in most cities is unsatisfactory, while in many it is deplorable and
is rapidly becoming intolerable. To attribute this condition to
inherent inferiority of this kind of pavement, as many people do,
is wholly unwarranted. Everything considered, a well constructed
and properly maintained asphalt pavement is not inferior to any
other kind of pavement now in use. At the same time it must be
conceded that an asphalt pavement usually requires attention and
repairs at an earlier period in its life than most of the other pave-
ments in common use, and it is probable, though not yet certain,
that the total cost of maintenance during its useful life is also
somewhat greater. It is certain that it requires more frequent
and careful attention during its life to keep it in a proper state of
repair than do stone or brick pavements. It is often the case
that high grade articles or structures require more careful and
skillful attention than the cruder varieties of the same class.
It will certainly not be claimed that the value of a pavement
may be measured by the amount of neglect and abuse it will
endure."

"As the first important step in this direction the city must

MAINTENANCE OF ASPHALT PAVEMEN*S. 507

avoid, in the contract and specifications, requirements that are
indefinite and ambiguous, and under which the contractor's
duties cannot be clearly determined and enforced. We have
seen that the long guaranty involves inherent conditions which,
under the above stipulation, it is necessary to avoid. It should
therefore be omitted from all future contracts. The art of con-
structing asphalt pavements is now so well understood that it
is entirely possible to frame and administer specifications that
will insure the production of a first class pavement, and the con-
ditions that originally dictated these guarantees therefore no
longer exist. The attempt thus to make the contractor responsible
for the character of the work he does has not, in practice, been
found effectual; moreover, our present knowledge of how the
work should be done and the practicability of thorough inspection
and supervision, renders the resort to such an expedient largely
unnecessary. But if a time guaranty is still thought advisable,
the period may be made so comparatively short that questions of
faulty construction shall not be confused with those of mainte-
nance. Barring accidental and unusual injuries, maintenance
questions ought not to arise within a period of two years after the
completion of the construction work, while palpable construction
faults should be revealed within that time. The guaranty
period should not therefore, exceed two, or at the most, three
years. Maintenance after that period ^should be provided by
the city."

Surface Heaters. The question of the best way to maintain
an asphalt pavement lies between cutting out the entire surface
and binder, and ihe use of what are known as ' 'surface heaters,"
which consist of a machine which directs a blast of flame from a
series of naphtha burners upon the asphalt surface and softens
it so that it can be removed to such depth as may be necessary
to reach undisintegrated material, and then applying sufficient
fresh material upon the area thus laid bare to renew the pavement
to proper thickness. Such work has been extremely satisfactory
in the past. As an example, the surface of Madison Avenue,
between 23d and 26th Streets, in New York City, was entirely
replaced in this manner in 1895, and has remained in excellent

508 THE MODERN ASPHALT PAVEMENT.

condition up to the present time. Of course the work cannot
be done in wet or cold weather. At such periods the entire sur-
face where disintegration occurs must be cut out to the foundation.
The surface heater work is of course much more economical.

Under any circumstances, all maintenance and repair work
will require the greatest skill in their execution, and too much
care cannot be devoted to the details of it.

PART IX.
CONTROL OF WORK.

CHAPTER XVII.

INSTRUCTIONS FOR COLLECTING AND FORWARDING TO THE
LABORATORY SAMPLES OF MATERIALS IN USE IN CON-
STRUCTING ASPHALT PAVEMENTS.

IN order that a laboratory examination, which has already been
shown to be necessary, may be satisfactorily carried out the samples
which are collected for this purpose should be carefully taken and
according to some system.

The following directions have been prepared by the author for
the use of superintendents and yard foremen.

The materials for the construction of asphalt pavements
which require inspection in the laboratory may be classified as
follows:
IN USE IN FOUNDATION.

Broken stone, gravel, sand, hydraulic cement.
IN USE IN BINDER.

Broken stone, bituminous cement.
IN USE IN SURFACE.

Stone for asphaltic concrete, sand, dust, or filler, refined asphalt,
fluxing agents, either eastern residuum, California asphaltic
oil, or other similar substances, and prepared from these
materials, asphalt cement and the surface mixture itself.

509

510 THE MODERN ASPHALT PAVEMENT.

These materials should be of satisfactory quality, and in order
to determine this, samples should be sent for examination and
report to the New York Testing Laboratory, Maurer, N. J. Fol-
lowing are directions, which must be closely observed, for collect-
ing and forwarding these samples from every city where contracts
are made and under way.

Samples and Specimens. 1. To begin with, it must be
explained that there is a decided difference between a sample
and a specimen of any material. A specimen is some of the mate-
rial selected to show its prominent characteristics, either of an
inferior or desirable nature. A sample, if properly taken, represents
the average composition and character of the material it represents.

Specimens are preferable to samples in certain instances and
the reverse. When it is desired to emphasize the peculiarities of
some material, a specimen is needed; but when a quantitative
determination of its characteristics is to be made, a sample is
necessary.

This distinction must be borne in mind in sending materials
to the laboratory for examination, and good judgment must be
used in regard to the most satisfactory means of arriving at the
desired end.

Materials for Foundation. 2. No samples of broken stone,
sand, or cement need be forwarded, unless there is some question as
to their suitability or quality, or unless they fail to meet the approval
of the local engineers. Under the latter circumstance, two or
three fragments of broken stone, a small sample box of sand, or
four pounds of hydraulic cement should be sent to the laboratory,
carefully identified as to the source from which the material comes
and as to the parties furnishing the same.

Materials for Binder. 3. No samples of broken stone or gravel
for binder need be forwarded unless their quality be in question.
Samples of asphalt cement for binder should be sent in the same
way as those for surface mixture, if especially made for binder.

Materials for Surface Mixture. 4. Sand. In the case of
sand supplies which have not been previously in use, and in every
case at the beginning of a new season's work, samples of the one

SAMPLES OF MATERIALS FOR THE LABORATORY. 511

or more sands the use of which is proposed, or information in
regard to the nature of which is desired, should be sent to the
laboratory with definite statements as to source, whether river,
lake, bank, etc., with the name of the party furnishing it and the
locality in which the sand is found. These samples should weigh
from two to three pounds, as much as will fill a cigar-box holding
fifty cigars. The sample should be moistened and tightly packed
so that the finer sand particles cannot sift out and be lost.

Samples of sand supplies which have been approved need
not be sent again during the same season, unless the character
of the deliveries appears to have changed decidedly or is sus-
pected to have done so.

Samples of the sand in use on the platform, after it has been
heated and screened should be sent to the laboratory once a week,
when the plant is running. The quantity contained in the ordinary
screw-top tin sample box is sufficient in this case unless other-
wise directed.

5. Dust and Filler. The ground mineral matter, or dust,
proposed for use as a filler should be sent in whenever a new
source of supply is contemplated, and its use not begun until its
quality has been approved.

From each delivery of such material a sample should be for-
warded for examination.

The amount contained in an ordinary tin sample box is, in
either case, sufficient for this purpose.

6. Refined Asphalt. Of refined Trinidad or Bermudez
asphalt it is unnecessary to send samples from the paving plants,
as this material has usually been inspected at the refineries. If,
however, a shipment or any part of it appears to be of inferior
quality or dirty, a convenient sized specimen, showing the defects
noticed, should be provided for examination.

Asphalt from any other source which may happen to be in
use, either experimentally or otherwise, should be sent in for
examination in the form of a convenient sized representative
specimen.

7. Fluxing Agents. A sample of each shipment or tank
car of residuum, California soft asphalt or similar material in use

512 THE MODERN ASPHALT PAVEMENT.

for softening the hardei asphalts in making asphalt cement r
should be sent in a tin can by express, not less than a pint in
amount.

Materials similar to blown oils can be sent in a box.

8. Asphalt Cements. Samples of asphalt cements from
every tank that is put in use should be taken at that time and for-
warded to the laboratory. If the consistency of that tank of
cement becomes altered at any time during the day by the addi-
tion of oil or flux, a new sample should be taken.

If the asphalt cement in any tank is not exhausted in one
day's run, and is in use again one or more days afterwards, samples
for each of these days should be taken and sent to the laboratory,
stating at the same time on a postal card, giving the number of
the sample, the amount of oil or flux that has been added per every
hundred pounds of cement estimated to be in the tank or dipping-
tank at the time the oil was added. The screw-top tin sample
boxes are to be used for this purpose.

9. Surface Mixture. Samples of surface mixture should
be forwarded daily. For ordinary work one is sufficient, but
where an important piece of work, subject to trying conditions,
is completed in one day, two or three samples, taken at intervals
while the mixture is being sent out, should be sent, to better illustrate
the average composition of the surface.

Sampling Methods to be Employed. 10. Unless the sampling
of any material that is to be examined is carefully done, the sample
will not be a representative one, and all work done upon it will be
wasted. The results will be worse than useless that is to say,
deceiving. Too much emphasis cannot be placed, therefore, on
the necessity for great care in this direction. In addition to
what has already been said in regard to sending samples to the
laboratory, the following suggestions for taking them should be
followed closely.

11. SAMPLING SAND:

1st. From Pit or Bank. It must be borne in mind that in a
pit or bank the sand lies in layers of different grading, which can
almost never be taken out separately. Experience has shown
that the best that can be done is to obtain a supply representing

SAMPLES OF MATERIALS FOR THE LABORATORY. 513

the average composition of the face of the bank. It is useless,
therefore, to send specimens of sand from strata that cannot be
isolated; or, if they are sent, specimens of the other layers in the
bank should accompany them, with a statement of their relative
thickness. A proper sample can be obtained by cutting a groove
down the face of the bank and collecting the material in a pile
and sampling as described below.

2d. From Rivers or Lakeshores. In case it is desired to sample
sands from river bottoms or lakeshores, it is impossible hi ordinary
cases to send in more than what is considered to be a representative
specimen of the material, and final sampling must await deliveries
on scow or car.

3d. Deliveries of Sand should be sampled as follows: Small
scoopfuls or shovelfuls are taken from different parts of the pile,
car, or boat load, and at different depths, in such number as will
fairly represent the lot, three to six, from a canal-boat or barge
and at depths of a foot or more, two from a car, and more or less
from a pile, depending on its size. When the sand is in a pile
the coarser grains will have rolled to the bottom, so care must be
exercised not to take the sand from that point or the top alone. It
is also well to dig some distance into the heap for some scoop-
fuls.

All the sand thus collected is dried, and, if large in amount,
is made into a heap, cut back and forth with shovels like a batch
of concrete and quartered, all but one quarter being rejected.
This is continued until the heap is reduced to such a size that it
can be passed through a Clarkson sampler, found at some of the
works, or sampled by rolling first in one direction and then at
right angles on brown paper and halving the mass, this being done
several times until it is reduced to the required size for shipping.

4th. Sand from Platform. Samples of the hot screened sand
in use in the mixer should be taken from the spout of the sand-
bin while the sand is running out freely into the box in the
process of filling it. It should be collected by running a shovel or
scoop back and forth several times along the edge of the distribu-
tor and then sampling the lot so gathered, either in a Clarkson
sampler or by rolling on paper in the usual way. ,

514 THE MODERN ASPHALT PAVEMENT.

Where there is no sand-bin a sample may be taken from the
floor pile, as already described from a delivery pile.

12. For sampling material of larger size, such as a barrel
of refined Trinidad asphalt, it should be broken up on a tarpaulin
and reduced to a size so fine that it can be treated in the same
way as sand. Usually, as has been said, specimens only of rock,
gravel, refined asphalt, and such materials are sufficient.

13. Sampling Asphalt Cement. It is difficult to always
obtain uniform samples of asphalt cement. From the same
bucket samples have been taken which varied as much as six
points in penetration, owing to imperfections in the way it was
dipped, a dirty bucket, or lack of uniformity in the cement. Great
care should be used, therefore, to see that none of the conditions
surrounding the dipping is abnormal. The dipping-bucket should
be immersed in the cement until it is of the same temperature,
and should then be moved about rapidly and submerged upside
down and full of air to the middle depth of the still and then
turned over, filled, withdrawn, and the tin sample boxes filled
where dust cannot reach them.

14. Sampling Surface Mixture. A small wooden paddle
with a blade 3 to 4 inches wide, 5 or 6 inches long, and \ an inch
thick, tapered to an edge at one end and with a convenient handle
at the other, is used to take as much of the hot mixture from the
wagon as it will hold, being careful to avoid any of the last drop-
pings from the mixer which may not be entirely representative
of the average mixture. Samples of mixture should never be
taken from the mixer itself, but only from the wagon after mixing
is completed.

In the meantime a piece of brown manilla paper with a fairly
smooth surface, 10 or 12 inches wide, and torn off at the same
length from a roll of this paper, which can be had at any paper
warehouse, is creased down the middle and opened out on some
very firm and smooth surface of wood, not stone or metal, which
would conduct heat too rapidly. The hot mixture is dropped
into the paper sideways from the paddle and half of the paper
doubled over on it. The mixture is then pressed down flat with
a block of wood of convenient size until the surface is flat. It

SAMPLES OF MATERIALS FOR THE LABORATORY. 515

is then struck five or six sharp blows with the block, until the
pat is about a inch thick. The paper should then be opened
and the pat trimmed with an ordinary table knife or spatula to a
size of about 2 by 4 inches, and a crease made along the narrower
edge at a distance of J an inch to facilitate breaking off a piece
for analysis when the pat is cold. Before the mixture is entirely
cold the proportions of sand, dust, and asphalt cement, together
with the sample number, date, and abbreviation of the name
of the city where the sample is taken, is impressed upon it with
steel stamps in letters and figures } of an inch high. The paper
is also marked with a rubber stamp, identifying it with the pat.
Additional information as to street, kind of dust, asphalt,
etc., can also be provided for in blank spaces opposite headings
printed by the rubber stamp. Such a stamp may be arranged
as follows:

Name of city. . . ,
Sample number. .
Date and hour. . .

Street

Sand, coarse

Sand, medium. . .

Sand, fine

Filler, kind

A. C

Asphalt, source. .

Flux, kind

Penetration A. C.
Temperature. . . .

The pat papers should be wrapped about the pat when cold
and both placed in a heavy clasp envelope for mailing at fourth-
class rates.

The pat paper is sent because the stain made upon it by the
asphalt of the hot mixture, when considered in connection with
the temperature of the mixture as it goes on the street, is of great
value in determining whether a suitable amount of bitumen is
present. Nothing should be written on the pat paper, as this
renders the entire pat liable to letter rates in mailing, but the
information required may be sent by filling in the blanks furnished

516 THE MODERN ASPHALT PAVEMENT.

by the rubber stamp on a postal card and mailing this at the
same time.

15. Samples of Old Asphalt Surfaces. Where the determina-
tion of the characteristics of an old asphalt surface is desired a
piece of the surface, together with the adhering binder course,
if one has been used, is selected which will represent the average
condition of the street. This should weigh at least 1 pound, and
it is generally desirable that two or more samples from each street
should be taken.

Collecting Samples. 16. Samples of stone, cement, sand,
refined asphalt, flux, etc., can be taken by any one about the
plant who is competent to follow the directions which have been
given but samples of mixture and asphalt cement should be taken
by the plant foreman himself and by no other person. When
there is a sub-laboratory at a paving plant the chemist in charge
will have general supervision, and may, if requested by the plant
foreman, attend to the collection of samples. The plant foreman
will be held responsible in all cases, through the local superin-
tendent, for the representative nature of the samples or speci-
mens which are forwarded to the laboratory and for any deviation
from the preceding instructions. It is especially urged upon the
superintendents that they shall see that these instructions are
carried out, and it is suggested that, to fully benefit by the results
of the laboratory examinations, samples should be sent not only
of the best but of the poorest work at each plant, in the latter
case calling attention to that fact and giving the cause of the
defect.

It is recommended that superintendents require their plant
foremen to initial all reports from the laboratory in order that it
may appear that the information given there has been brought to
their attention.

Samples of asphalt cement and surface mixture should usually
be taken as soon, after starting up the plant, as the work is going
on regularly. The dipping-tank may be sampled at once so as
to be able to mail it, as soon as cool, at an early hour. Any change
in the character of the cement or addition to it demands a new
sample, coupled with details of the change. If a second lot of

SAMPLES OF MATERIALS FOR THE LABORATORY. 517

cement goes into use later in the same day this should also be
sampled immediately and sent to the laboratory as soon as possible.

A second sample of mixture should be taken if any decided
change in it is made, such as increasing or decreasing the amount
of asphalt cement, dust, or proportions of different sands.

The capacity of the laboratory for work is large, and if the
rejection of any sample is necessary it is better done according
to judgment exercised there than at the works.

Numbering and Mailing Samples. 17. The samples of
residuum, sand, and dust should be numbered consecutively,
regardless of each other and of all other samples. For instance,
the first sample of residuum sent for analysis would be No. 1, the
second sample No. 2, and so on. The same would be true of the
sand and dust, the first sample of each of these materials being
No. 1 and the second No. 2, etc.

Samples of surface mixture will be numbered consecutively,
but in case two samples of asphalt cement and only one of sur-
face mixture are sent on the same day, the number of the second
sample of asphalt cement should be the same as the first, but a
figure " 2 " should be placed slightly above the right-hand upper
corner and a " 3 " for a third corresponding to .the same sample
of surface. For instance, supposing that only one sample of
surface mixture, No. 10, is sent on one day, but two of asphalt
cement, the latter would be numbered 10 and 10 2 . In this way
the A. C. sample number will be made to agree with that of the
surface mixture in which it was used. If two samples of surface
mixture and two of asphalt cement are sent on the same day
the numbers on each should correspond.

It must be insisted upon that too much care cannot be taken
to so thoroughly identify samples that there may be no difficulty
in recognizing their origin and source, even after the lapse of
years. The habit of sending materials with a designation by
number, or otherwise known only to the local superintendent,
must be discontinued.

Samples from the plant should be mailed at once, or, if hot,
as soon as cool, and usually at or about noon of the day on which
the material represented by the samples is used. It may, of

518 THE MODERN ASPHALT PAVEMENT.

course, be necessary to mail others in the afternoon, but morn-
ing samples should in all cases be sent promptly and from the
nearest letter-box to the works. The practice of sending samples
to the superintendent's office should be discontinued. Duplicate
samples can be sent to him for his inspection if he desires.

In order to check delays in the mails and in other ways it
is well to mark the date and time of mailing on each package, and
it should be made the special duty of some one person to attend
to this matter, and he should be held responsible for mailing.

A new series of sample numbers for each material or class of
materials should be started on January 1 of each year.

CHAPTER XXVIII.

METHODS EMPLOYED IN THE ASPHALT-PAVING INDUSTRY
FOR THE CHEMICAL AND PHYSICAL EXAMINATION OP
THE MATERIALS OF CONSTRUCTION.

THE results of the examination of the materials in use in the
construction of asphalt pavements will be of no value unless the
samples or specimens representing these materials are collected
with great care, so that they shall be truly representative of
whatever is to be examined The directions given hi the preceding
chapter should, therefore, be followed in taking them.

The methods practised in the asphalt-paving industry in judging
the different materials are as follows. It may be said in the begin-
ning that these methods make no claim to great analytical accuracy,
but afford that information needed in the industry with a maximum
amount of accuracy and rapidity at a minimum expenditure of
time. Many of them are only of relative value; that is to say, they
do not furnish absolute data and can only be used when some well-
known specimen of the same material is treated in a parallel
manner and used as a standard of comparison, as will appear later.

Stone, Gravel, Slag, etc. These materials, which form the coarser
part of the aggregate in the foundation and the entire aggregate
of the binder, can generally be examined by the hand-and-eye
method without submitting them to laboratory tests. They should
be free from adventitious matter, soil, vegetable debris, etc. They
should be hard and when shaken together or passed through heating-
drums should produce but a small amount of detritus. If necessary
the percentage of this, formed in a given length of time when a
definite weight of the material is revolved in a rattler, such as is
used in testing paving bricks, may be determined.

519

520 THE MODERN ASPHALT PAVEMENT.

For binder, stone should not be too porous and tests for the
amount of water absorbed in twenty-four hours should be made.

As a rule the examination of these materials for use in the
asphalt-paving industry is in no respects different from that which
would be made were they to be used for other purposes. A more
elaborate examination of any stone available for road construction
will be made by the Office of Public Roads, U. S. Department
of Agriculture, Washington, D. C., on application on forms sup-
plied by the Department.

Where it is desirable to determine the proportions of stone of
different sizes which go to make up the aggregate of crushed rock,
as in arranging the grading of an asphaltic concrete, this can be
determined by means of riddles. These consist of perforated
metal set in circular wooden frames sixteen inches in diameter,
with circular openings of the following diameter:

1 inch, J inch, J inch, f inch and J inch.

Binder Cement. This is generally examined for its consistency
alone by the same method as asphalt cement for surface mixture.

Sand. This is examined as to the material of which the grains
are composed, their shape, the character of their surface, the
amount of silt, clay, loam, coal, or vegetable debris it contains,
the size of the grains, or the sand grading as it is called, and the
voids in the sand when compacted. The grains of sand are more
often quartz than other minerals. At times some limestone
grains or shells are present and rarely a considerable portion of the
grains are silicates of igneous and volcanic origin, feldspar, horn-
blende, shales, magnetite, etc., as at Siboney beach, in Cuba, and
in Mexico. At Santiago sands derived from coral reefs occur.
These different minerals are determined in the usual way by exam-

METHODS OF ANALYSIS. 521

ination under the glass and with reagents. The shape of the
sand grains is important and is noted with a glass. It may be
classed as sharp, round, medium, irregular, and with greater detail
as to special peculiarities. The surface of the grains should be
examined and note made as to whether it is smooth and polished,
rough like ground glass, or between these grades, and if it is covered
with any cementing material such as the ferruginous matter adhering
to the New Jersey sands. The capacity of the surface of the grains
for adsorbing aqueous vapor may also be determined with the
object of learning the thickness of the film of asphalt cement
which it will probably retain. The silt or clay and vegetable
debris are detected by shaking a volume of 30 cubic centimeters
of sand with 100 cubic centimeters of water in a graduated
cylinder until thoroughly wetted and allowing the coarse particles
to subside. A rough estimate of the amount of silt, etc., may then
be made by the eye. The separation of sands into grains of various
sizes by the use of sieves is the most important means of determin-
ing their availability for paving purposes. Another estimate may
be reached from the character and amount of material passing the
finest sieve in use and determining the size of the grains.

Determination of the Grading of Sands. This is done with
a series of sieves consisting of carefully woven brass wire cloth
stretched upon a tin frame. These cloths were originally so
selected that the average diameter of the particles which each sieve
passed bore some definite relation to those passed by the next
finer sieve. The average diameter of these particles as passed by
the sieves in 1908 and the names given in the trade to the cloths
are:

200-mesh. . . .085 millimeter

100-

173

80-

231

50-

377

40-

469

30-

595

20-

. . 1 002

10- '

* . . 2.133

The 200-mesh cloth was selected as a basis of measurement,
being the finest available wire cloth, and it was found that the
average diameter of the largest particles it will pass is .085 mm.

522

THE MODERN ASPHALT PAVEMENT.

It seemed unnecessary to use any cloth, such as the 150-mesh,
between this and the 100-mesh sieve, as the largest particles passed
by the latter were only twice the size of those passed by the former.
For the same reason an 80-mesh sieve was selected for the third
sieve, as its largest particles were more nearly three times the
size of those passed by the 200-mesh sieve than any other. A
50-mesh sieve passing particles about four times as great in diam-
eter was selected, rejecting the use of the 60. From this point
the increase in size is greater at each step, as no intermediate sieves
are available or necessary. The particles passing the 40-, 30-,
20-, and 10-mesh sieves are approximately six, eight, twelve, and
twenty-four times the diameter of the finest particles.

Obtaining satisfactory sieves of this description is not readily
accomplished. Most of those found in the trade are made of
poorly woven cloth or the cloth is so stretched in putting it on
the frames that the interstices between the wires are much altered
in size and no two sieves of the same number will agree. It is
only satisfactory, therefore, to use sieves which have been care-
fully tested and compared among themselves and with a standard
set, or to determine the factor for correction as recommended by
Hazen. 1

Sieves can now be had in such perfection from Howard & Morse,
1197 DeKalb Avenue, Brooklyn, N. Y., that a sand sifted on
one set of sieves in Mexico and on another in the New York Test-
ing Laboratory by different operators agreed remarkably well.

Mexico.

New York
Testing
Laboratory.

Passin

g 200-mesh

100-

1 1

80-

1 1

50-

40- ...

30-

20-

10-

3
100

4
100

Report of Mass. State Board of Health, 1892, 541.

METHODS OF ANALYSIS.

523

The finest cloth, in the 200-mesh sieve, is so delicate that it
must be used with care and continually watched to detect any
deterioration. Tearing away from the frame is a frequent occur-
rence, but such a defect or a small hole in the cloth can be stopped
readily with soft solder, making the sieve as good as new. New
sieves often have spots of solder in them where defects due to imper-
fect weaving or strains in mounting the cloth have been stopped
out.

The sieves are made in nests, the finest being of the largest
diameter, about 8 inches, as the greatest area of sifting surface

FIG. 30. Sand Scale.

is needed with this cloth. The diameter decreases until that
of the 10-mesh sieve is only 5 inches. For storage and shipment
the sieves thus occupy a small space.

With such a set of sieves the size and grading of the particles
of a sand can be satisfactorily determined and intelligently expressed.
The operation of sifting and weighing is conducted as follows: A
definite weight of sand, 50 grams, is taken. This is weighed out
on a scoop and beam-scale especially constructed for use in making
sand siftings and sensible to half a gram, Fig. 30. It is the ordi-
nary No. 485 Fail-bank's scale, 1 supplied to seedmen for deter-

1 Fairbanks Co., New York.

524 THE MODERN ASPHALT PAVEMENT.

mining the dust and dirt in flaxseed, modified to weigh a normal
weight of 50 grams instead of 1 pound. The 50 grams are
divided by graduations on the beam of this balance into 100 parts
representing per cents, thus doing away with any calculations.

The sand thus weighed out is thrown upon the 200-mesh sieve.
It is important, however, before sifting is begun that the cloth
of all the screens, and especially the three finest, be thoroughly
cleaned with a stiff bristle brush from all particles which may
have become fixed in the meshes. For this purpose a small stiff
stencil brush, or one that is found in house-furnishing stores
for scrubbing porous filters, may be used. After rubbing with a
brush the screen is struck several sharp blows up and down to
free it from loose particles.

The knack of using the sieves satisfactorily and quickly can
hardly be described in print. After some shaking from side to side,
the sieve is hit sharply on the table orchard surface to dislodge
particles which have filled the meshes but will not pass through.
Many attempts have been made to do the sifting by mechani-
cal shaking devices and with sieves of all sizes at one time, but
with little success. Where a large number of siftings is required,
hand work is more reliable and quicker.

The most satisfactory mechanical device we have found is
the "Per Se" Tier shaker,* which may be operated by power or
hand, one of which is in use in the New York Testing Laboratory
for screening concrete aggregates or when large quantities of
sand must be considered.

The sifting is done over a clean piece of paper, and when nothing
passes the 200-mesh sieve, all lumps of loam or clay which readily
break up under the fingers having been rubbed to a powder on
the sieve and the coarser grains cleaned from dust by attrition, the
residue is returned to the scoop of the balance.

It will be noticed that some material at this point usually
remains in the meshes of the wire cloth. This is allowed to remain
there when the residue is poured from the sieve, and being more
nearly the size of the grain passed by this sieve than the next
larger is included with the 200-mesh or the coarser mesh sieves
with their particular size grains and counted as passing that sieve.
When the cloth of the sieve is brushed these grains are rejected,
and, as all the determinations are made by loss and not by direct
weighing, this is satisfactory.

* Howard & Morse, Brooklyn, N. Y.

METHODS ol ANALYSIS. 525

The per cent of material passing the 200-mesh sieve is arrived
at as follows: The beam at the point where the poise is put to
weigh out the original 50 grams of sand is graduated zero, and
between this point and where the poise balances the empty scoop
is graduated into 100 parts which may be read as per cents. The
amount of material which has been passed by the sieve and rejected
may then be seen at once in per cents on weighing the residue in
the scoop, and so for subsequent sieves, subtracting of course each
time the previous reading from the last.

It may be asked why the 200-mesh sieve is used first and why
the results are not stated in per cents of materials retained on
the different sieves, as is commonly the case. One of the reasons
for using the finest sieve first is that if the dust or fine material
is not removed at once much of it is blown away and lost in the
process of sifting. As in this method the percentages are deter-
mined by loss, the amount disappearing during the sifting of the
200-mesh sieve first is of no consequence. Another is that the
fine material adhering to the coarser grains is more readily separated
from them by attrition in the finer sieves, and that the presence of
the coarser sand aids in breaking up lumps and expedites sifting.
In fact when sifting very fine material like dust or filler it is usual
to place some coarse gravel not passing a 10-mesh sieve,
or a few pennies, on the 200-mesh to aid in keeping the fine
cloth clear and to break up lumps, thus doing the work of the coarser
particles in sand or mineral aggregates which are entirely absent
from the filler. The results are stated in per cents passing a
given sieve rather than that retained, because this results in making
a uniform statement and not some figures passed and some retained,
and because it is easier to associate the percentages with definite
sized particles rather than with sieves which will not pass them.

After the 200-mesh sieve, the others are used in order, and of
course more rapidly as they become coarser. The greatest care is,
of course, necessary with the finest sieve to clean the coarse grains
and to break up lumps of clay, etc., for which the finger ends are
most suited, as their pressure can be graduated and no undue force
exerted upon the cloth or upon the particles which do not readily
disintegrate.

526

THE MODERN ASPHALT PAVEMENT.

The actual subtractions made in a sifting appear in the follow-
ing facsimile page of a laboratory record, a rubber stamp being
a decided convenience for reporting purposes:

No. 30402

100.0

13
4

27
13

100
95

No. 30402

Mesh
No.

Per Cent.
Passing.

Bit.

200

4.0

100

9.0

14.0

30.0

20.0

11.0

7.0

5.0

R. 10

Total. . .

100.0

Remarks :

Voids in Sand and Mineral Aggregates. An important con-
sideration in connection with a sand or mixture of sands for use in
asphalt pavements is the percentage of voids which it contains
on compaction.

METHODS OF ANALYSIS. 527

The ordinary methods of determining voids by measuring the
volume of water that must be added to the compacted mineral
aggregate to fill them, or pouring the compacted aggregate into a
measured volume of water in a graduate and noting the increase of
volume, are not satisfactory, because in the first way it is difficult
to displace all the air in the voids or to know when it is displaced ;
in the second, because in light sand and sand mixed with filler a cer-
tain amount of the fine material is with difficulty persuaded not to
float or leaves at best a meniscus which cannot be read.

The ordinary means of attaining ultimate compaction is also
deficient in accuracy. At air temperatures the grains of any
sand or dust are surrounded by a film of adsorbed aqueous vapor
which prevents their packing as closely as possible. To attain
satisfactory compaction the aggregate must be above the tempera-
ture of boiling water. It is necessary, therefore, to use hot sand
in determining the voids in fine materials and the finer the ma-
terial the more necessary it is.

The sand aggregate should be heated to about 250 F. in a small
deep-form iron sand-bath 1 and then compacted in one of the fol-
lowing ways.

First Method. A narrow-necked flask, graduated to 100 c.c.,
is filled with hot sand, the neck taken in one hand and with the
other hand the body of the flask is struck back and forth from the
neck to the bottom with a peculiar jarring motion with a wooden
rod of about of an inch diameter and 12 inches long, covered
for 3 inches with a piece of rubber tubing. The jarring settles
the sand together rapidly in the flask and it is necessary to add
more from tune to time. When jarring ceases to compact the sand
further it is, after having been brought to a definite volume, emp-
tied on a balance and weighed. For ordinary purposes this
weight divided by the weight of an equal volume of quartz,
will give the actual volume the particles of sand occupy,
and from the difference between this and the volume of the
flask the voids are learned. When greater accuracy is re-
quired the flask with the hot sand must be allowed to cool to
ordinary temperatures, to allow for contraction of the quartz in

1 Eimer & Amend, No. 4555, 6-inch.

528 THE MODERN ASPHALT PAVEMENT.

volume, and again filled to the mark. The specific gravity of the
sand grains is ordinarily assumed to be 2.65, but it may vary, and
the possibility of this can be determined at a glance. When this is
the case the density must be determined, often in petroleum or
alcohol when fine sand is present. The apparatus designed by
D. D. Jackson, which has been described by him in a paper before
the Society of Chemical Industry in 1904,* and which can be
obtained from Emil Greiner Co., 45 Cliff Street, New York City,
has been found to be very convenient for this purpose.

The advantage of this process is that the flask is filled with sand
at once and there is no segregation of particles of different sizes,
especially dust, which sometimes takes place in the next method.

Second Method. The hot sand is taken as before, but it is
put into a graduated 100 c.c. cylinder, 10 c.c. at a time, and com-
pacted by tamping on a block of wood after each addition. When
the compaction has reached its ultimate limit, this is known by
a disappearance, especially with fine material, of a noise due to
the presence of an air-cushion between the particles before this
point is reached. Segregation takes place to a limited extent,
especially in mixtures of sand and dust, but even with this error
a greater compaction and density is obtained and less voids are
found than with the flask method for the same aggregate, the
difference being about 1 per cent in favor of the cylinder.

Volume Weight of Sand. From the weight in grams of 100
c.c. of a sand or aggregate, obtained in the determination of the
voids as just described, the weight per cubic foot in pounds can be
found by multiplying by the factor .624. This is of value in deter-
mining what difference in the weight per cent of bitumen to expect
from the addition of the same volume of asphalt cement to sands
of different volume weight.

For example, an aggregate from a western city in 1899 weighed
124.5 pounds per cubic foot, from another 115.7. It is very easily
seen that, with the same volume or weight of asphalt cement,
added to each, the percentage by weight in an ordinary mixture
will be much lower in the first than in the second city, and in

* J. Soc. Chem. Ind. 24, 593.

METHODS OF ANALYSIS. 529

practice it is found that in one case it was 10 per cent, in the other
11.3 per cent. This determination of volume weight therefore
serves as an aid to our interpretation of our gravimetric analysis.

Dust or Filler. A dust or filler of ideal quality should consist
of particles, all of which should be so fine that they will pass a
200-mesh screen. Everything coarser merely acts as sand. It
is important, too, that the particles should be much finer than
a size that will merely pass this sieve. They should be impal-
pably fine. 200-mesh sand is not the same as dust and is, in fact,
often undesirable in a surface mixture.

In examining a dust or filler, therefore, it is necessary to deter-
mine with the 200-mesh sieve the percentage passing it and to
study the character of that which passes. The latter examination
can be made in two ways. As we have no sieve available for the
purpose it must be done with a microscope or powerful lens which
will show the character of the grains, or by elutriation.

Elutriation Method. Until recently the only means of deter-
mining the fineness of a dust or filler has been by means of a
200-mesh sieve, but as the material passing this sieve might con-
sist in whole or in part of grains as large as .10 mm. in diameter
which can hardly be considered as dust, but are, on the contrary,
only fine sand, something more satisfactory is demanded. This
has been found in the elutriation process in use in soil analysis.
Five grams of the dust to be examined are placed in a beaker
about 120 mm. high, holding about 600 c.c. The beaker is nearly
filled with distilled water, at a temperature of 68 F., and agitated
with an air-blast until the dust and water are thoroughly mixed.
On stopping the blast the liquid is allowed to stand exactly 15
seconds and the water above the sediment immediately decanted
without pouring off any of the latter. This washing is repeated
twice. The sediment is washed out into a dish, dried, and weighed.
The loss in weight represents what may be considered as dust free
from sand. The washing must be done with distilled water, since
water containing salts in solution, as is well known, induces floc-
culation. This method can also be used with hydraulic cements,
since the material acted upon by water is retained in suspension
and removed, while that which subsides is practically unacted

530 THE MODERN ASPHALT PAVEMENT.

upon and can be dried and weighed without difficulty. The
differentiation in this case can, however, not be carried beyond
that resulting in 15 seconds. With other materials the differ-
entiation of the particles not subsiding in 15 seconds can be carried
further, if desired, by reagitating the decanted material and allow-
ing the sedimentation to go on for 1 minute, 30 minutes, 1 hour,
and so on. The preceding method is an adaptation of that pro-
posed by Osborne for the separation of the particles of soil of
various sizes, for further details of which reference must be made
to the Connecticut Agricultural Station Annual Report, 1886,
page 141, " Principles and Practice of Agricultural Analysis,
Wiley, Vol. 1, page 196, and Hazen, 24th Annual Report Mass.
State Board of Health, 1892, page 543.

Crude Hard Asphalts. The analysis of crude asphalt is con-
ducted in much the same way as that of the refined product except
that it is necessary to determine, in the former, the loss of water
and light oil, which are not or should not be found in the refined
material. If there is any question as to the dryness of the refined
material this should, of course, be first determined in the same
manner as with the crude. The determination of water can be
made in two ways.

Ordinary Method. Ordinarily it is sufficiently accurate to
weigh out 2 to 5 grams of the material in a crucible, or preferably
on a watch glass to expose more surface, and to subject it to a
temperature of 100 C., in a well regulated air-bath with the pre-
cautions described on pp. 534, 535, until it ceases to lose in weight to
an extent of more than .2 to .3 per cent on successive heating.
A greater concordance is not sought, as many asphalts continue to
lose light oils gradually at this temperature. The oven which
is used for this purpose in the author's laboratory is one of the
Lothar-Meyer form, or a modification of this, which is fully described
on page 536, Figs. 31 and 32. The degree of fineness to which the
crude asphalt should be reduced before weighing out is dependent
upon the amount of water it contains. In powdering some asphalt,
such as crude Trinidad, the material, since it contains 29 per cent
of water in emulsion with bitumen, begins to lose water at once.
It can, therefore, only be broken into coarse lumps and not reduced

METHODS OF ANALYSIS. 531

to a powder until after a preliminary determination of the water
thus lost by the coarse material. Other asphalts, containing only
a small amount of hygroscopic or adventitious water, may be
ground up at once, while some which are not readily powdered
may be cut into small pieces. If it is necessary to determine the
water absolutely it may be absorbed and weighed and the difference
stated as gas or light hydrocarbons. This is hardly necessary from
a technical point of view.

Alternate Method. For asphalts such as crude Trinidad, in
which the difficulties described above are met, a different method
of procedure is advisable. The substance is very quickly reduced
to a coarse powder only, in a mortar provided with a cover, through
which the pestle passes. Five grams of it are spread out on a
4-inch watch-glass, and this is placed in vacuo over sulphuric
acid for twelve hours and the loss determined. It should then
be reground to a fine powder and exposed again in vacuo until
it ceases to lose weight. The loss may be stated as water.

In the examination of a cargo of crude Trinidad for technical
purposes, about 50 grams in small lumps are placed in a flat 8-oz.
oblong tin box and dried to practically constant weight at 325 F.
The results thus obtained are comparable with those obtained
in the refining by steam at this temperature.

In whichever way the asphalt is dried a sufficient quantity is
prepared and preserved in this condition in a tightly stoppered
bottle, for analysis. Asphalts which cannot be reduced to powder
are used in mass. The powdered asphalts have a slight tendency
to absorb hygroscopic moisture and must be protected from the
air.

In the dried condition crude asphalts can be considered, as
far as analysis is concerned, simply as refined material, and all
determinations should be done with and percentages calculated to
this material, including the water or loss, by calculation, in the
final results if desired.

Refined Asphalts. Examination of refined asphalts in their
most extended form include determinations given on the accompany-
ing form, used as a convenience in reporting. With well-known
asphalts but a limited number of determinations are necessary for

532 THE MODERN ASPHALT PAVEMENT.

the purpose of detecting the lack of uniformity or peculiarities in
the material.

NEW YORK TESTING LABORATORY.

Test number MAURER, N J

Source of supply

PHYSICAL PROPERTIES.
Specific gravity, 78 F./78 F. original substance, dry.

" " " pure bitumen

Color of powder or streak

Lustre

Structure

Fracture

Hardness, original substance "

Odor

Softens

Flows

Penetration at 78 F

CHEMICAL CHARACTERISTICS.
Original substance:

Loss, 212 F., 1 hour

Dry substance:

Loss, 325 F., 7 hours

Character of residue

Penetration of residue at 78 F

Loss, 400 F., 7 hours (fresh sample)

Character of residue

Penetration of residue at 78 F. . .

Bitumen soluble in CS 2 , air temperature.

Inorganic or mineral matter

Difference. .

Malthenes :

Bitumen soluble in 88 naphtha, air temperature.. . .

This is per cent of total bitumen

Per cent of soluble bitumen removed by H 2 SO 4

Per cent of total bitumen as saturated hydrocarbons.

Bitumen soluble in 62 naphtha.
This is per cent of total bitumen.

METHODS OF ANALYSIS. 533

Carbenes:

Bitumen insoluble in carbon tetrachloride, air temperature.
Bitumen more soluble in carbon tetrachloride, air tem-
perature

Bitumen yields on ignition:

Fixed carbon. .

Sulphur

Ultimate composition B

Remarks:

Physical Properties. Specific Gravity. The specific gravity
of the dried asphalt is taken in a picnometer at 25 C. and referred
to water at the same temperature. This temperature has been
selected as the most convenient mean between the room tem-
peratures of winter and summer, and is much more suitable in
our surroundings than the lower temperature generally in use
abroad, 15 C. Determinations at the latter temperature are
much hampered by the great difference between it and OUT labora-
tory temperatures, the rapid expansion after cooling to 15 C.
being difficult to provide for, as well as the condensation of mois-
ture on the surface of the picnometer when the dew point is high,
a room with the temperature in use for penetrations is always
available for density and other temperature work. This tem-
perature of 78 F. has therefore been taken as a normal one for
all physical work on asphalts. The specific gravity of the pure
bitumen extracted from those asphalts carrying a considerable
amount of mineral matter, in a way to be subsequently described,
is also determined in the same way.

The usual determination of the outward physical features of
any mineral substance, Color of Streak, Structure, Fracture, Hard-
ness, and Odor if any, are noted.

The color of the streak or of the powder of a hard asphalt is
in certain cases characteristic. For example, in the case of refined
Trinidad lake asphalt the powder is a bluish-black color, while
that of the refined Trinidad land asphalt is much browner. Pow-

534 THE MODERN ASPHALT PAVEMENT.

dered gilsonite is of a very light-brown color. Powdered gra-
hamite is quite black.

The structure of a solid native bitumen may be either homo-
geneous or it may show the presence of cavities containing gas,
particles of adventitious mineral matter, shale or clay, or other
peculiarities.

The fracture may be, in the case of very pure bitumens, con-
choidal or semi-conchoidal, pencillated or hackley in the case of
grahamite, or irregular.

The hardness of the original material, if it contains much
mineral matter, may be stated in degrees of Mohrs scale, that
of the pure bitumen in several ways. It is either brittle, like
glance pitch, or soft enough to be penetrated with the needle of
a penetration machine in which case the hardness is expressed
in degrees of this machine.

The odor in the case of many bitumens is characteristic. That
from Venezuelan asphalt, found near the Gulf of Maracaibo, is
extremely strong and rank, while others are more purely asphaltic,
especially on warming. The heat in any case brings out the
odor to a degree not observed in the cold material.

Loss on Heating. It is sometimes necessary to determine
the loss which an asphalt suffers on heating for a time to definite
temperatures. The length of time has been arbitrarily taken
as seven hours and the temperature 325 F. and 400 F.

The determination is made as follows: In a No. 1 crystallizing-
dish, 2J inches in diameter and 1 T \ inches high, or 3 oz. deep seam-
less tin box of this size, 1 are placed 20 grams of the material
under examination. The exact dimensions of the dish are of no
great importance, as can be seen from the following determina-
tions:

1 American Can Co., New York.

METHODS OF ANALYSIS

535

TWENTY GRAMS OF RESIDUUM AT 325 F. FOR SEVEN HOURS.

Weight of dish

23.3465

18.1825

19.1855

1.51"

1.47"

1.28"

2.15"

2.20"

.06

.045

.05

.925

.920

.950

Position in b&th.

Left

Right

Middle

Should it be necessary to use a very much larger dish the
weight of the material to be taken should be calculated so that
the volume which it holds shall bear the same relation to the
surface exposed as in the case of the smaller dish. It is necessary
to take separate portions of the substance for each determination,
and not to attempt to determine the loss at 400 F. from the
sample which has been previously heated at 325 F.

The dish is heated to the requisite temperature for the given
length of time in an oven the temperature of which is uniform in
all parts, something that is not as easily accomplished as might
be supposed, and with the assurance that the materials are main-
tained at the proper temperature, a temperature which it has
been found is not indicated by that recorded by a thermometer,
that registers merely the temperature of the air in the bath. Such
an oven is not only difficult to obtain, but the manner in which
the best form is used is of great importance.

Farm of Oven Employed in the Author's Laboratory. Extended
experience with various ovens thoroughly convinced the author
that none of the forms ordinarily furnished by the supply dealers
were satisfactory, especially if they are heated by the direct appli-
cation of the flame to the bottom of the oven. The air-bath of
Lothar-Meyer l was found to be by far the best, but inconvenient,

1 Berichte, 1889, 22, 1, 879.

536

THE MODERN ASPHALT PAVEMENT.

owing to the fact that the interior is not readily reached. On this
account an oven has been designed, possessing all the advantages
of Meyer's form, but much more convenient for use in an asphalt
laboratory. The accompanying illustrations show its construc-
tion, Figs. 31 and 32.

FIG. 31. New York Testing Laboratory Oven.

It will be seen that the bath instead of being heated from the
bottom is heated by a 10-inch ring burner immediately under-
neath the space between the oven itself and the outside wall, as
in the Lothar-Meyer form. The perforations in this ring are neces-
sarily systematically spaced to allow for the greater gas pressure at

METHODS OF ANALYSIS. 537

the point where the latter enters, in order that a uniform amount of

ii n \ i rm

FIG. 32. New York Testing Laboratory Oven.

heat may be furnished at all -points in the circumference. If this
is properly arranged, and the inner chamber of the bath is well
made, no further precautions are necessary to avoid inequalities of
temperature caused by the direct entrance of hot air from the burner
into the oven. In standardizing the bath it is necessary that a
number of thermometers should be inserted at different points in
order to determine that the temperature at all points is uniform.
If this is not the case the openings in the ring burner must be
rearranged until this is accomplished. The interior of the bath, it
will be noticed, is provided with a fan for causing a circulation of
air to bring about still greater uniformity, this fan being moved by
any convenient source of power.

The inner chamber is provided with a perforated shelf of metal.
The dishes containing the material to be subjected to the desired
temperature, it has been found, cannot be placed directly on this

538 THE MODERN ASPHALT PAVEMENT.

shelf with the assumption that they will not exceed the tempera-
ture recorded by the thermometers in the air circulating in the
bath. The conductivity of the metal shelf is' so much greater than
that of the air that the dishes will attain a much higher temperature
than the air in the bath. This difficulty can be avoided to a very
considerable extent by placing a sheet of asbestos over the shelf;
but even then the temperature of the material in the dish will be
somewhat different from that of the air in the bath. In order to de-
termine what the temperature of the former is it has been found nec-
essary to use a thermometer, which is immersed in heavy residuum
oil placed alongside the material under examination. The reading
of this thermometer will give the temperature to which the material
under examination is being subjected. The thermometer exposed
only to the air of the bath is then observed merely for the purpose
of detecting any sudden changes.

It will be noted that the cover to this bath is hinged so that it
may be opened conveniently for inserting and removing the dishes
containing the material under examination. It is provided with
numerous openings for the insertion of thermometers and a gas
regulator, and for the escape of the vapor of hydrocarbons which
have been volatilized. The outer shell of the bath is covered
with asbestos for insulating purposes. 1

Melting or Flowing Point. The solid native bitumens can
have no definite melting point, for the reason that they are com-
posed of mixtures of hydrocarbons. It is only possible, therefore,
to determine rather arbitrarily the point at which the material
softens or flows, and with special reference to the relation of this
point toward some standard bitumen. This is determined as
follows:

A crystallizing dish, about 2^ inches in diameter and with 1^
inch sides, filled with clean mercury to a distance of J inch from
the top, is placed over a 20-mesh wire gauze and heated by a
small flame protected from draughts by a chimney. On the sur-
face of the mercury is placed a thin microscopic cover-glass,
No. 2-0, carrying the specimen of asphalt under examination.

1 The bath is constructed for the author by the Hauck-Seebach Co., 291
Essex Street, Brooklyn, N. Y.

METHODS OF ANALYSIS.

539

When dealing with hard asphalts that can be ground rather
coarsely, several fragments which will pass a 40-mesh sieve and
be retained on a 50-mesh sieve (about .50 mm. diameter) are
spread on the cover glass and placed upon the surface of the

FIG. 33

mercury, covered with a funnel, from which the stem has been
cut and the thermometer passed through the orifice until the
bulb is immersed in the mercury. It is held in position by a
clamp attached to the ring-stand holding the dish. Under the
dish a burner is placed that can be regulated to a small flame and

540 THE MODERN ASPHALT PAVEMENT.

heat, so that the rise of temperature will be from three to five
degrees per minute. In a short time it will be noticed that the
specimens will have changed from the brown or brownish-black
color of the powder to that more nearly approaching tlie original,
with a slight rounding of the individual grains. On further heat-
ing these globules flow together and form a thin sheet on the glass.
The point at which the specimen begins to flow, as indicated by
the thermometer, is noted as the melting or flowing point.

Asphalts that cannot be ground are softened and pulled out
to a thread and cut into small pieces, about 1 cubic mm. Sev-
eral pieces should be placed on the glass together, as one will
serve as a check on the other, and thereby lessen the chance of
error. The softening point may be noted by the rounding of
the particles and the beginning of the flow, or when the specimen
begins to spread out, which is always at the point of contact with
the cover-glass, is set down as the flowing point or the tempera-
ture at which the specimen will melt.

Determination of Total Bitumen. One gram of the dried
or refined material, in a state of very fine powder, if possible, is
weighed out and introduced into a 200 c.c. Erlenmeyer flask of
Jena glass and covered with about 100 c.c. of bisulphide of carbon.
It is then set aside for at least five hours, or overnight, at the tem-
perature of the laboratory. In the meantime a Gooch crucible
is prepared with an asbestos felt and weighed. This Gooch
crucible is of special form with a large filtering surface. It holds
30 c.c., is 4.4 cm. wide at the top, tapering to 3.6 cm. at the
bottom and 2.6 cm. deep. This is much better for percolation work
than the usual narrow form of Gooch. The felt is made by beating 1
up long-fibre Italian asbestos in a mortar, and suspending the finei
particles in water and quickly pouring off from the coarse particles.
Too much of the latter should not be removed, or the felt will be
too dense. The decanted asbestos and water can be kept in a
bottle for use. To prepare the felt the asbestos and water are
shaken up and what is found to be a proper amount poured into
the crucible, which has in the meantime been attached to a vacuum
filtering-flask by the proper glass and rubber connections. As
soon as the asbestos has somewhat settled the vacuum-pump is

METHODS OF ANALYSIS. 541

started and the felt firmly drawn on the bottom of the crucible.
It is then dried, ignited, and weighed.

After standing a proper time the bisulphide is decanted very
carefully upon the filter which is supported in the neck of a wide-
mouth flask and allowed to run through without pressure. The
flask after being tipped to pour the first portion is not again placed
erect in order to avoid stirring up the insoluble material, but is
held at an angle on any suitable base, such as a clay chimney.
After all the bisulphide has been decanted more is added and
the insoluble matter shaken up with it. This is allowed to settle
and decanted as before, the insoluble matter being finally brought
on the filter and washed with the solvent until clean. The excess
of bisulphide is allowed to evaporate from the Gooch crucible
at the temperature of the room. It is then dried for a short time
at 100 C. and weighed. The loss of weight is the percentage of
bitumen soluble in 82-

In the meantime, however, the bisulphide which has passed
the filter is allowed to subside for twenty-four hours, if possible,
and is then decanted carefully from the flask in which it has been
received into a weighed platinum or unweighed porcelain dish.
If there is any sediment in this flask it must be rinsed back into
the Gooch crucible with bisulphide and the crucible again washed
clean. The solvent in the dish is placed in a good draught and
lighted. When all the bisulphide has burned, the bitumen
remaining in the dish is burned off over a lamp and the mineral
residue, which was too fine to subside, is weighed, if the burning
was done in a platinum dish, or dusted out and added to the cru-
cible if in a porcelain one. In the former case the weight is added
to that of the Gooch crucible or subtracted from the per cent
of bitumen, found without its consideration, as a correction.
Care must be used in this method of procedure that the solvent
does not creep over the sides of the crucible and that the outside
is free from bitumen before weighing. In order to avoid this the
crucible is supported in the neck of a flask with three constrictions,
the neck extending above the top of the crucible and the latter
being covered with a watch-glass. These flasks are made for

542 THE MODERN ASPHALT PAVEMENT.

the author by E. Machlett & Son, 143 East Twenty-third Street,
New York.

Mineral Matter or Ash. One gram of the same sample of
material used for the determination of bitumen is weighed out in a
No. Royal Berlin porcelain crucible and burned in a muffle or
over a flame until free from carbon. This must be determined
by breaking up the cake of ash, moistening with water or alcohol,
and observing if any black particles of coke are present. The
weight of the residue is stated as inorganic or mineral matter.

The determination is of course not exact, sulphuric acid and
the alkalies being volatilized in many cases, but it is satisfactory
for technical purposes.

Naphtha Soluble Bitumen. For the purpose of determining the
percentage of bitumen soluble in naphtha distillates, 88 and 62 B.
are used. It is extremely important that these naphthas should be
of the exact degree specified, since differences in density will make
an appreciable difference in the amount of bitumen extracted.
The distillate should be that obtained from a paraffine petroleum.
The density of each lot should be carefully determined with a
Westphal balance at 60 F. and if it is too dense or too light, it
can be brought to the proper density by diluting with a heavier
or lighter naphtha as required. Redistillation of these naphthas is
unnecessary as the products of distillation are no more uniform
than the original naphtha.

It will be found very necessary that hard bitumens should be
reduced to an impalpable powder before attempting to extract
them, as otherwise the extraction will not be complete. The softer
bitumens should be divided as much as possible.

The bitumen is usually extracted with naphthas of both densi-
ties hi order to determine the difference in their action. If the
amount extracted by each is the same or nearly the same it will point
to the fact that the bitumen consists of hard asphaltenes mixed with
light malthenes, the latter equally soluble in naphtha of both
degrees of density, and but little intermediate hydrocarbons, or of
the very hard asphalts fluxed artificially with some light oil.
If, on the other hand, there is a very considerable increase in the
percentage dissolved by the 62 over the 88 naphtha it may be

METHODS OF ANALYSIS.

543

assumed that the malthenes are well graded and natural constitu-
ents of the bitumen which is being examined. In certain cases,
however, the use of the two naphthas is unnecessary. It would be
useless to extract a maltha with a dense naphtha or glance pitch
or albertite with a lighter one.

In determining the naphtha soluble bitumen in asphalts and
other hydrocarbons it was the custom from 1887 to 1899 to make
the extractions in small beakers, No. 0. One gram of the substance
was weighed out and covered with a sufficient amount of naphtha,
about 75 c.c., and placed on the steam bath and allowed to boil
until the solvent became thoroughly saturated. It was then
decanted through a weighed Gooch crucible and the residue succes-
sively treated until free from bitumen soluble in naphtha. As it
was almost impossible to get concordant results in this way, on
account of the loss of the lighter constituents of the naphtha and
the consequent increase of density of the solvent, resort was had
to the use of Erlenmeyer flasks, about 12 cm. high and 200 c.c.
capacity. One gram of the substance was weighed out and boiled
with the naphtha in a loosely stoppered flask for from one-half
to one hour, according to the character of the material to be ex-
tracted. The solution was decanted as with the beaker method
and the treatment repeated. The results were a slight improve-
ment over the open beaker, but not entirely satisfactory. The use
of a return cooler was then tried and gave good results with 62
naphtha, but as the loss of light hydrocarbons from the 88 naphtha
could not be controlled, even in this way, any heating with this
very volatile solvent was abandoned. The change in the two grades
of naphtha on heating are shown from the following experiments:

EFFECT OF HEATING NAPHTHA AS IF USED AS A SOLVENT.

Degrees B.

Gravity
15 C./15 C.

Treatment.

Loss by
Weight,
Per Cent.

Loss by
Volume,
Per Cent.

Residue,
Specific
Gravity.

88
<

0.6379
0.6379
0.6379

Return cooler

i ( ( t

Open flask

51.2
65.6

53.5

40.0

37.0
48.0

0.6523
0.6523
0.6585

0.7321
0.7321

Return cooler
Open flask

3.0
10.0

1.0
3.0

0.7352
0.7393

544 THE MODERN ASPHALT PAVEMENT.

It appears, therefore, that heating increases the density of both
naphthas, and consequently their solvent powers, from inability
to condense the more volatile parts, but that the change in the 62
naphtha is small, so that it can be safely heated to a slight extent.

As a result of these experiments all determinations are now
made with cold naphtha by the following method:

One gram of the substance is weighed into a 200 c.c. Erlenmeyer
flask, covered with naphtha and allowed to stand, as in estimating
total bitumen, in fact the entire process is the same with the ex-
ception that one or two precautions must be observed. It is well
not to attempt to break up any lumps with a stirring rod, as the sub-
stance, especially the softer asphalts, may then adhere to the rod
or flask and be difficult to detach. It may also be necessary to
treat the substance with several portions of the solvent instead
of with two or three, as in the case of carbon disulphide. No heat
is applied at any time in the process.

The naphtha soluble bitumens are frequently denominated
petrolenes. The writer has recently suggested the name malthenes
as bitumen of this nature closely resembles maltha in its consis-
tency. Objection has been raised by partisans to the use of the
name petrolene as leading to the conclusion that petrolene is a defi-
nite compound. Of course it is no more a definite compound than
kerosene, but a mixture of various hydrocarbons like the latter.
The objection to this designation must, therefore, fall to the ground,
although petrolenes or malthenes may be more satisfactory as being
less misleading.

Determination of the Character of the Malthenes or Naphtha
Soluble Bitumens. The determination of the relative proportion
of saturated and unsaturated hydrocarbons which constitute the
malthenes is very important in differentiating the solid bitumens.
It is made as follows:

The 88 naphtha solution of the bitumen under examination
is made up to a volume of 100 c.c. or reduced to that volume by
evaporation. It is then placed in a 500-c.c. separatory funnel.
An equal volume of the solvent naphtha is placed in another sepa-
ratory funnel. The naphtha solution and the naphtha are then
subjected to the action of 30 c.c. of sulphuric acid of specific

METHODS OF ANALYSIS. 545

gravity 1.84, the acid and the naphtha V>eing shaken together for
exactly three minutes. This is most important, since the action
of the acid on the hydrocarbons in the bitumen under examina-
tion is not a fixed one, but will continue more or less indefinitely.
After the shaking, the acid and the naphtha solution are allowed
to stand overnight. The acid is then carefully drawn off and the
shaking again repeated with another volume of acid of the same
amount. This will require a shorter time for the separation of the
acid and it can be drawn off within a few hours. If the second acid
is very strongly discolored the acid treatment should be continued
a third tune. In the case of the blank determination with the
plain solvent one treatment will be sufficient. The naphtha
solution and the naphtha are then washed twice with water and
afterwards once with a 5 per cent carbonate of soda solution,
after which one further washing with water takes place. The
naphtha solution of the bitumen which is being treated and the
blank naphtha are then poured into crystallizing dishes 3 inches
in diameter and 2 inches deep. In the plain naphtha is dissolved
from .50 to .75 gram of some extremely stable petroleum residuum.
The two dishes are then placed upon the steam-bath to evaporate
the naphtha. In order to avoid creeping, the sides of the dishes
are imbedded in a mass of cotton waste reaching to the top, as
creeping is much diminished by having the sides of the dish warm.
The evaporation is carried on on the steam-bath until the naphtha
is volatilized and until the blank shows on weighing that the
residue has returned to its original weight. It is then as-
sumed that the other dish is free from naphtha, and from
the water which the latter has dissolved in the process of washing.
This, under the conditions observed in the author's laboratory,
will require about six hours, but the exposure on the water-bath
is generally continued one hour after the blank has reached a
constant weight and further for fifteen minutes in an air-bath at
100 C. as control. The results obtained in this way are of no
absolute value, but are of relative importance in comparing differ-
ent fluxes and solid bitumens. It cannot, of course, be applied
where the bitumen contains an appreciable amount of hydro-
carbons volatile at 100 C.

546 THE -MODERN ASPHALT PAVEMENT.

Where 62 naphtha is the solvent its volatilization from the
residue of bitumen which has been treated is extremely difficult,
and such a determination is, therefore, not recommended.

Determination of Bitumen Soluble in Carbon Tetrachloride.
While in the large majority of cases the same, or nearly the same,
amount of bitumen is dissolved by carbon tetrachloride as by bisul-
phide of carbon, bitumens are known in which hydrocarbons
exist which are not as soluble in the former solvent for example,
one of the Venezuelan asphalts when overheated in refining,
grahamite, and some of the residual pitches. The use of this
solvent may, therefore, be desirable at times for the purpose of
differentiating the native bitumens. The insoluble matter is
extremely fine and is with difficulty retained on the closest filters.
It appears to coagulate after several hours and in conducting the
operation should be allowed to stand over night before filtering.
A mild current of air passed for an hour through the solution
accomplishes the same purpose, and may be used to hasten
the analysis when desired.

One gram of the sample is treated with 100 cc. of the cold
solvent, and with the exception of the precautions above noted,
filtered in exactly the same way as with carbon disulphide.

The following data are given in illustration of the effect of
different methods of handling.

Filtered as soon as dissolved 89 .8%

' ' after standing 15 hours 83 .6

" " " 48 " 83.6

" " blowing with air 1 hour as soon as dissolved . . 83 . 5

The commercial supply of carbon tetrachloride contains more
or less carbon disulphide, and this naturally affects its solvent
power, so that different lots may vary in this respect. As the carbon
disulphide is much more volatile than the carbon tetrachloride,
the majority of the latter can be removed by redistillation and
rejecting all that which goes over below the boiling-point of
the carbon tetrachloride, 76 C. It is also possible that it may
be removed by blowing a current of air through the carbon tetra-
chloride.

METHODS OF ANALYSIS. 547

Preparation of Pure Bitumen. The preparation of the pure
bitumen is a necessity where the percentage in the crude or refined
material does not exceed 50 per cent, as under these circumstances
its properties are so much concealed by the materials which are
mixed with it that it is impossible to determine them, especially
the hardness, softening point, and other physical data. The
process which has been worked out for this purpose applies equally
well to native bitumens and to artificial mixtures, such as old
surfaces where it is desired to determine the consistency of the
bitumen in the pavement.

Such an amount of crude, refined material or old surface is taken
as analysis shows will afford about 20 grams of pure bitumen.
At the same time 20 grams of a bitumen or asphalt cement of
corresponding character and of known consistency is taken and
treated in the same way as the material under examination. JThis
is done for a control, as will appear. The original material and
that for the control determination are placed, in small pieces,
in a 600 c.c. Erlenmeyer flask and covered with 300 c.c. of redis-
tilled bisulphide of carbon. This with shaking is allowed to stand
overnight or until all lumps are broken down and the bitumen
is dissolved. After thorough sedimentation the solvent is decanted
as carefully as possible into a litre flask and 200 c.c. of fresh bisul-
phide poured upon the residue. This should be shaken and allowed
to stand again until the insoluble matter has subsided, when the
solution of bitumen is decanted as before and added to the first
300 c.c. This process is renewed with several portions of 100 c.c.
of bisulphide until the residue is clean. The entire solution is
allowed to stand overnight, again decanted from the finer sedi-
ment of mineral matter, and then swung hi a centrifugal machine
to remove as much of the still finer mineral matter as possible.
If organic debris is present the solution must also be filtered,
In case a more rapid method is desired for old surface mixtures,
it is probably quite as satisfactory to swing the solution obtained
in the first 300 c.c., as this is, of course, representative of the
total bitumen, although only a portion of it.

If no centrifugal is available the different bisulphide solutions
are well mixed, allowed to stand for some days and decanted.
The solutions of bitumen, the one holding that under exainina-

548 THE MODERN ASPHALT PAVEMENT.

tion and the control, are, one after the other, placed in the same
flask and the solvent distilled off as far as possible at the heat
of a steam-bath. The hot and thick residue is poured into an
iron dish, the 6-inch-deep-form sand-bath already described.
This is placed on a suitable sized hole on the steam-bath and
heated. The remaining bisulphide is largely driven off in this
way. To prevent the vapor from the hot bisulphide from taking
fire, it will do so without the presence of flame in contact with
a hot steam-pipe, or from foaming over, a current of dry steam
is blown over the surface of the liquid as long as vapor is evolved.
Finally, the presence of the last traces of vapor are tested for
with a small flame such as is used for determining the flashing-
point of oils. If all vapor of bisulphide which can be distilled
in this way has disappeared, the bitumen is in a condition to be
brought over a flame or sand-bath and heated, with constant
stirring, to a temperature depending on its softness, and until it is
sufficiently fluid to be poured into a tin box for further treatment.
This temperature should in no case exceed 325 F. These tin
boxes are of the kind used in taking samples of asphalt cement
for penetration and shipping them to the laboratory. A con-
venient form and size is the flat, 2-oz., screw top, Gill style. 1

The bitumen or bitumens under examination and the control
bitumen, after having been well identified in the tin boxes, are
brought to the standard temperature and their consistency
determined with the penetrometer. The control will usually
be found to be softer by twenty or more points than in the original
condition. If this is the case both or all of the extracted bitu-
mens are put in an oven and heated for a length of time to 300 F.,
depending upon their excess of softness. It is important, of
course, that the conditions in the air-bath are uniform, and that
the same precautions should be used as in the determination of
loss at 325 F. and 400 F., as previously described.

When the control bitumen has reached its original known
consistency it is assumed that the bitumen or bitumens under exam-
ination have done the same thing, and the product is taken as the
pure bitumen as it occurs in its original consistency in the crude or
refined material or of the cement as it exists in the surface mixture.
1 American Can Co., New York.

METHODS OF ANALYSIS.

549

Experience shows that this determination is reliable within
five points on duplicate determinations.

Fixed Carbon. The fixed carbon is determined usually on
the pure bitumen according to the method recommended by the
Committee on Coal Analysis of the American Chemical Society and
published in the Journal of the Society for 1899, 21, 1116. It is
as follows :

Place 1 gram of pure bitumen in a " platinum crucible weigh-
ing 20 or 30 grams and having a tightly fitting cover. Heat
over the full flame of a Bunsen burner for seven minutes. The
crucible should be supported on a platinum triangle with the bottom
6 to 8 cm. above the top of the burner. The flame should be
fully 20 cm. high when burning free, and the determination
should be made in a place free from draughts. The upper surface
of the cover should burn clear, but the under surface should remain
covered with carbon."

FIXED CARBON IN BITUMENS.

Extn

;mes.

High

High.

Low.

Grade.

53.3

35.3

53 3

Albertite.

54.2

29.8

29 8

Gilsonite

26.2

3 3

14 5

Asphaltenes from Trinidad bitumen

25 8

Glance pitch . . .

15 0*

Asphalts .

17 9

10 8

14 2

Bverlyte (artificial asphalt)

14 3

Standard Asphalt Co.'s mine soft gilsonite. . . .

7 3

Malthenes from Trinidad bitumen

6 3

Wurtzilite Utah ...

8 8

5 3

8 2

Residuum, Pennsylvania field

3.4

2 7

1 Egyptian.

The residue minus the small impurity of ash in the pure bitumen
is the fixed carbon, which should be calculated to 100 per cent with
the volatile hydrocarbons, excluding the inorganic matter. As
the committee states, this determination, like most industrial
ones, is arbitrary, but it is of the greatest value in determining the

550 THE MODERN ASPHALT PAVEMENT.

nature of a bitumen quickly. Experience has shown that true
hard asphalts have never been found which yielded more than
17.9 per cent or less than 10.0 per cent of fixed carbon, while
grahamite yields over 53 per cent, albertite over 29 to 54 per cent,
and some other bituminous materials characteristic amounts of
fixed carbon.

EXAMINATION OF HEAVY PETROLEUM OIL.

Fluxing Agents and Oils. The examination of materials
under the above heading includes the determinations given in the
accompanying forms:

NEW YORK TESTING LABORATORY.

MAURER ; N. J.,

Report on sample of FLUXING AGENT received from. . . .

Character of flux

Date when sample was gathered

Name of manufacturer

Tank-car number

Sample number Test number

Specific gravity, Beaume Actual At 78 F.

Flash-point F

Loss, 212 F., . '.hours

Loss, 325 F., 7 hours

Loss, 400 F., 7 hours

Character of residue at 78 F

Bitumen insoluble in 88 naphtha, air temperature. Pitch. .
Per cent of soluble bitumen removed by H 2 SO 4

Paraffine scale

This material is quality

Remarks:

METHODS OF ANALYSIS. 551

NEW YORK TESTING LABORATORY.

MAURER, N. J.,

Test number:

Source of supply

PHYSICAL PROPERTIES.
Specific gravity, dried at 212 F., 78 F./78 F. .. .

Flows, cold test

Color

Odor

Under microscope

Flashes, F., N. Y. State oil-tester
Viscosity

CHEMICAL CHARACTERISTICS.
Original substance:

Loss, 212 F., 1 hour or until dry

Dry substance :

Loss, 325 F., 7 hours

Character of residue

Penetration of residue at 78 F

Loss, 400 F., 7 hours (fresh sample)

Character of residue

Penetration of residue at 78 F

Bitumen soluble in CS 2 , air temperature.

Difference

Inorganic or mineral matter

Bitumen insoluble hi 88 naphtha, air temperature. Pitch. .

Per cent of soluble bitumen removed by H.jSO 4

Per cent of bitumen as saturated hydrocarbons

Per cent of solid paraffines.

Bitumen yields on ignition: ,

Fixed carbon . ....

Ultimate composition:

Remarks:

552 THE MODERN ASPHALT PAVEMENT.

The methods used in making these determinations are, as a
whole, the same as those described for hard asphalts with the
following modifications.

Specific Gravity. The specific gravity of oils or fluxes is
taken on the material either dried at 212 F., or, if there are light
oils present, volatile at this temperature, on some of the oil freed
from water by being swung in the centrifugal.

Heavy fluxes too dense to employ a picnometer with are filled
into an open specimen tube, 10 cm. long, 2 cm. in diameter,
and holding about 27 grams of water, even with the top, which
is ground flat and parallel to the base. The weight of this volume
of oil at 78 F. is compared with that of water at the same tempera-
ture. Lighter oils are examined with the picnometer or Westphal
balance. The industrial methods for the determination of the
specific gravity of dense oils admit of much improvement and are
now probably not accurate beyond the second place of decimals.

Mr. Lester Kirschbraun, chemist of the Chicago City Labor-
atory, Bureau of Engineering, describes the following method
which is employed by him satisfactorily.

Take a test tube about a half inch in diameter and cut
it off to give a 1J inch by i inch tube. Flare it out to carry a
fine wire. Put about 10 grams of the oil or asphalt into it,
and suspend it in an oven to remove air bubbles and drive off the
water. Cool and weigh accurately in air and immerse in distilled
water at 25 C. to a fixed mark on the wire and weigh. Previous
to filling the tube with the sample, determine its weight carefully
in air and in water at 25 C., immersed to the fixed mark. These
weighings give the weights of the tube alone, in air and
in water, and the combined weights of the tube and the sample,
in air and water. The gravity is calculated in this way, represent-
ing.

Weight of tube in air a

Weight of tube and sample in air 6

Weight of tube in water c

Weight of tube and sample in water d

Wt. in air of sample

bD. gT. = 9 f, ;

Loss of wt. in water

METHODS OF ANALYSIS. 553

in this case becomes

7H !T\ TA IT* V-*'/

When the gravity of the sample is less than unity, the second
expression in the denominator is a negative (d c)quantity and
is added to the original weight of the sample (b a), inasmuch as
the buoyancy of the sample will overcome its own weight and
to a certain extent will also reduce the weight of the tube in water,
making c greater than d.

The formula in this case may be also expressed so:

(2)

When the gravity of the sample is greater than unity, d
will be greater than c and the first formula applies without
confusion. As an example, take a blown oil, the gravity of which
was determined in our laboratory according to this method.

Weight of tube in air " a = 4 .7870

Weight of tube and sample in air 6 = 16 .7900

Weight of tube in water, 25 C c = 2 .8565

Weight of tube and sample in water d = 2 .6425

by formula (2)

16.7900-4.7870 = 12.003p =

~ (16.7900-4.7870) +(2.8565-2.6425) " 12.2170"

This may seem a bit complicated at first, but it will be noted
that the factors a and c are constants which can be used
for the same tube without change, and the calculation becomes
very simple after a few trials.

The advantages of the method are in doing away with the
inconveniences of handling a large amount of oil, and the ease
with which air bubbles can be removed, particularly in handling
refined asphalt. Of course it can be used only with oils which
are viscous enough to be retained in the tube when under water.

554 THE MODERN ASPHALT PAVEMENT.

Flow Test. Some of the oil is chilled in a large test-tube and
gradually allowed to attain the temperature of the room. The
point at which it will flow in the inclined tube is the flow-point.

Color. This is found by examining the reflection from the
surface of the cold oil. It is intended to be that revealed by
reflected and not by transmitted light through a thin film.

Odor. The odor can be described as that corresponding to
different kinds of known petroleum in the cold or on heating.

Microscopic Examination. The appearance of an oil that has
been heated is noted under the microscope to determine the presence
of material insoluble in the main mass of the oil.

Flash-point. The flash-point is determined in a New York
State oil-tester. 1 The water-bath is of course removed and the
oil heated directly with a flame of a size to raise the temperature at
the rate of 20 per minute and a small flame from a capillary glass
or metal tube is used for flashing. The flame should be applied
at 5 intervals. The determination should be repeated on such oils
as flash at unexpected temperatures. The water must be removed
from the oil or flux before putting it in the tester cup, either by heat
or by the centrifugal.

Open tests of high flashing oils are not reliable and at the
best with the closed tester a reading of 5 intervals only need be
sought.

Viscosity. Thf> Engler viscosimeter is used for this purpose,
it being available for temperatures as high as 500 C.

The rate of flow in seconds of time is compared with that of
an equal volume^ of water at the same or some standard tem-
perature.

Loss at 212 F. The water or loss of light oils at 212 F. is
determined by weighing out 20 grams in a glass or tin dish, such
as is described for use in the determination of loss at 325 F. in
hard asphalts, and heating in the oven described, at the temper-
ature named, until the oil has ceased foaming, or to practically
constant weight. The precautions previously noted should be
observed. When oils contain a large percentage of water this is

1 E. & A., No. 4160.

METHODS OF ANALYSIS. 555

better determined by the centrifugal method or by dilution with
naphtha.

Drying an oil or flux for subsequent examination is done by
heating a large volume in an iron dish over a flame, with constant
stirring, unless it contains much light oil, when the centrifugal
method alone can be used.

Loss at 325 F. and 400 F. in Seven Hours. Separate portions
of 20 grams of the dried material are taken for each determina-
tion and are heated to these temperatures in the manner described
for solid bitumens. The residues from the oils and fluxes are
examined after heating to these temperatures more in detail than
those from hard asphalts which have been treated similarly.

After cooling and weighing the appearance of the residue is
noted, especially as to whether it is smooth or granular owing to the
presence of paraffine, the temperature at which it flows, whether
it pulls out to a long string or is short. If it is so hard that it does
not flow except on raising the temperature above 100 F., its con-
sistency is determined with the penetrometer either at 100 F. or
at 78 F. or at lower temperatures.

The residue should also be examined under the microscope to
determine whether, owing to the nature of the fluxes, they have
been at all decomposed at these temperatures with a separation
of insoluble pitch, which is an evidence that the original flux must
have been more or less cracked in the process of manufacture.

Total Bitumen ; Inorganic Matter and Naphtha Soluble Bitu-
men. These determinations are arrived at by the methods
already described for hard asphalts.

As the oils and fluxes are more easily soluble it is unnecessary
to let the solvents act on them for so long a time as in the case
of hard asphalts. There is little object in using 62 naphtha with
oils or fluxes, as there is too little difference between its solvent
power and that of bisulphide of carbon with such materials to make
it worth while. The residue insoluble in 88 naphtha, however,
shows how much decomposition there has been in fluxes which
have been subjected to excessive heat.

Determination of the Character of the Hydrocarbons in Fluxes.
The character of the hydrocarbons in any of the heavy oils used

556 THE MODERN ASPHALT PAVEMENT.

for fluxing purposes is determined by treatment with sulphuric
acid after the method described for use with malthenes from hard
asphalts.

Determination of the Amount of Hard Paraffine Scale. The
amount of hard paraffine scale contained in any flux or heavy
oil can be readily determined by the author's modification 1 of
the method of Holde. 2

The method in detail is as follows:

The Determination of Paraffine in Petroleum Residues,
Asphaltic Oils, and Asphalts Fluxed with Paraffine Oils. For
this purpose, one, two, or more grams, of the substance to be ex-
amined is taken and covered in an Erlenmeyer flask with 100 c.c.
of 88 naphtha. The amount will depend on the paraffine present
and on the percentage of oil which remains after the preliminary
treatment with naphtha and acid. Of a residuum from east-
ern pipe-line oils one gram is sufficient, as the substance con-
sists of a nearly pure bitumen containing from 4 to 12 per cent of
paraffine. Ten grams of a residual pitch from asphaltic oil should
be used, as this, in some cases, contains only 65.0 per cent of its
bitumen soluble in naphtha, less than 50 per cent unacted on by acids,
and only about 1.0 per cent paraffine. Several grams can be taken
of a Trinidad asphalt cement, made of asphaltum and paraf-
fine residuum, which contains 26.0 per cent of mineral matter
and only 70.0 per cent of its bitumen is in a form soluble in 88
naphtha.

The object of the naphtha treatment is to separate the paraffine
from substances of a non-bituminous nature and from some of
the asphaltic hydrocarbons insoluble in naphtha which would
be precipitated in the ether alcohol solvent and contaminate the
paraffine.

By this means all the unsaturated hydrocarbons and those
of an asphaltic nature, readily precipitated by alcohol from

1 J. Soc. Chem. Ind., 1902, 21, 690.

2 Mitt. a. d. Konig, tech. Vers-anst, Berlin, 1896, 14, 211. Abs. J. Soc.
Chem. Ind., 1897, 16, 362. Lunge, Chem. tech., Untersuchungs, Methoden.
3.9.

METHODS OF ANALYSIS.

557

their ether solution, are removed and the possibility brought
about of recovering the paraffine in a pure condition.

Some determinations made in the manner described resulted
as follows:

PETROLEUM RESIDUUM FROM PIPE-LINE OIL.
Specific gravity, 0.93.

Number.

Weight
taken.

Soluble in Naphtha.

Not acted on by
H(V

Paraffine.

Grams.

1.0

Per Cent.
96.0

Per Cent.
No treatment

Per Cent.

7.95

1.0

96.0

89.5

5.55

1.0

Distilled in vacua

No treatment

5.95

TRINIDAD ASPHALT CEMENT.

Number.

Weight Taken,
Grams.

Soluble in Naphtha.

Not Acted on by

H2SC-4.

Paraffine.

No treatment

2 95

Treated

In each case the paraffine recovered after treatment was white
and pure, while that obtained in the other way, even by distilla-
tion in vacuo, was colored. The results after treatment were, of
course, lower and more correct.

The Trinidad asphalt cement was made from 100 parts of
Trinidad asphalt and 20 parts of a residuum similar to the one
analyzed. The asphalt contained, of course, no paraffine; the
residuum, 5.55 per cent. The calculated amount in the cement
is therefore 0.925 per cent, and 0.95 per cent was found.

In this way it can be determined whether the flux which has
been used in the preparation of an asphalt cement has been derived
from paraffine petroleum or from one having an asphaltic base,
since if paraffine is found to such an extent as shown above it will
necessarily point to the use of a paraffine flux, as no native solid
bitumen hi use in the paving industry contains paraffine.

558 THE MODERN ASPHALT PAVEMENT.

Rapid Method for Paraffine in Crude Petroleum Fluxes and
Residues. 1 Standard Oil Company. The preceding method, while
not lacking in accuracy, cannot be completed in a single day, and
the following method, while yielding lower results, has the ad-
vantage of rapidity.

One hundred grams of the oil is distilled rapidly in a 6-oz.
retort to dry coke.

Five grams of the well mixed distillate is treated in a 2-oz. flask
with 25 cc. Squibb's ether; after mixing together thoroughly,
25 cc. Squibb's absolute alcohol is added, and the flask packed
closely in a freezing mixture of finely crushed ice and: salt for at
least thirty minutes. Filter off the precipitate quickly by means of
a suction pump, using a No. 575 C. S. & S. 9-cm. hardened filter,
cooled by the above freezing mixture in a suitable apparatus.

Rinse and wash the precipitate with 1 to 1 Squibb's alcohol
and ether mixture cooled to F. until free from oil. Fifty cc.
of the washed solution is usually sufficient. When sucked dry,
remove the paper, transfer the waxy precipitate to a small glass
crystallizing dish. Dry on a steam bath and determine the
weight of paraffine scale remaining in the dish.

Calculation. Weight of paraffine scale divided by weight of
distillate taken and multiplied by per cent of total distillate ob-
tained from original sample, equals per cent of paraffine scale.

Fixed Carbon. This determination is sometimes desirable
with fluxes, especially with harder ones, to show the same facts
revealed by the 88 naphtha extract. It is carried out in the
same way as with hard bitumen.

Asphalt Cement. Asphalt cement is examined to determine
its consistency, the amount of bitumen, inorganic or mineral
matter and organic matter, not soluble, it contains. Rarely the
naphtha soluble bitumen is extracted and examined to determine
the nature and quantity of the flux of which it has been made.
The permanency of its consistency, when it is maintained in a
melted condition at 325 F. for some time, may be noted.

The consistency of asphalt cement is determined in several
ways, the most desirable of which in the laboratory is the pene-

1 Kindly furnished by Mr. George M. Saybolt.

METHODS OF ANALYSIS. 559

tration machine, for the reason that it admits of an absolute record
in figures. Penetration machines have been designed by Bowen,
Kenyon, Dow, and tho New York Testing Laboratory. The two
latter, however, have superseded the earlier and less desirable types.

The flow test originated by the Warren-Scharf Asphalt Paving
Company, as well as the test by chewing, which is a rough one
but always available, are of decided value and convenience for
use at plants.

The construction of the various penetration machines and the
method of using them is described as follows:

Bowen's Penetration Machine or Viscosimeter for Bitumi-
nous Solids. Principle of the Machine. The machine, Fig. 34, is
designed to register on a dial in degrees, arbitrarily selected, the
depth to which, a No. 2 cambric needle, attached to a weighted
arm or lever, penetrates into the surface of the material to be
tested when allowed to act upon it for one second at a standard
temperature.

The machine has been described by H. C. Bowen, in the
School of Mines Quarterly, 10, 297.

The Dow Penetration Machine. A penetration machine has
been designed by Mr. A. \V. Dow, formerly Inspector of Asphalt and
Cements, of the District of Columbia, Washington, D. C , which in so
far as it is based on the measurement of the millimeters to which a
definite needle penetrates into the asphalt cement under a definite
weight at a definite temperature is concerned, is a more truly
scientific instrument than the one previously described. The read-
ings by this machine are about 20 points lower than those obtained
with the Bowen instrument, but it possesses the disadvantage
that when large numbers of asphalt cements are to be examined
at any one time it requires greater delicacy of manipulation and
much more time than is the case when the Bowen machine is
used. Fig. 35. Mr. Dow describes its use as follows:

" Description and Directions for Using the Dow Penetration
Machine. The object of the penetration test is to ascertain the
softness of asphalt, etc., and is accomplished by determining the
distance a weighted needle will penetrate into the specimens under
examination.

560 THE MODERN ASPHALT PAVEMENT.

FIG. 34. Bowen Penetration Machine.

METHODS OF ANALYSIS.

561

"So that all tests may be comparable, a standard needle
should be used, weighted with a constant weight. The tests should
be made on samples at a standard temperature and be made for

A H

FIG. 35. Dow Penetration Machine.

the same length of time in every case. The standards used in
this machine for testing cements to see that they are of uniform
consistency are a No. 2 needle, weighted with 100 grams, pene-
trating for five seconds into the sample at a temperature of of 77 F.
(25 C.)

" The apparatus consists of a No. 2 needle A, inserted in a

562 THE MODERN ASPHALT PAVEMENT.

short brass rod which is held in the aluminum rod C by the
binding screw B. The aluminum rod is secured in a frame-
work so weighted and balanced that when it is supported on the
point of the needle A the framework and rod will stand in an
upright position, allowing the needle to penetrate perpendicu-
larly without the aid of a support, thus doing away with any
friction.

" The frame, aluminium rod, and needle weigh 100 grams with
the weight on bottom of frame ; without weight 50 grams. Thus
when the point of the needle rests on the surface of the sample
of material to be tested as to the penetration, it will penetrate
into the sample under a weight of 100 grams or 50 grams as desired.

" The needle and weighted frame are shown in Fig. 35, side
and front views of the entire apparatus put together and ready
for making a penetration. D is the shelf for the sample, E,
is the clamp to hold the aluminum rod C until it is desired
to make a test, F is a button which when pressed opens clamp E.
By turning this button while the clamp is being held open it will
lock and keep the clamp from closing until unlocked. The device
to measure the distance penetrated by the needle consists of a
rack, the foot of which is G. The movement of this rack up or
down turns a pinion to which is attached the hand which indicates
on dial K the distance moved by the rack. One division of the
dial corresponds to a movement of the rack of 1/100 cm. The rack
can be raised or lowered by moving counterweight H up or down.
L is a tin box containing sample to be tested which is covered
with water in the glass cup, thus keeping its temperature constant.
MM' are leveling screws. A clock movement having a 10-inch
pendulum is attached to the wall to one side of the machine. Make
a mark P on the wall just at the extremity of the swing of the
pendulum; a double swing of this pendulum, that is from the time
it leaves P until it returns, is one second.

" The only other things necessary to complete the outfit are a
large dish-pan, a pitcher to hold ice-water and a tin for hot water;
a coffee-pot is a good thing.

" To make penetration tests place the materials contained in
circular tins, along with the glass dish, under five or six inches of

METHODS OF ANALYSIS. 563

water in the dish-pan, which should have been previously brought
to a temperature of 77 F. by the addition of hot water or cold
water.

" While the sample are under the water it should be stirred
every few minutes, with the thermometer and the temperature
kept constant at 77 F. by the addition of hot or cold water as the
case may require. The samples should remain under the water for
at least fifteen minutes and in cases where they are very cold or
hot, at least one-half hour. The most expeditious way to proceed in
testing a sample just taken from a still or tank is to immerse it in
ice-water as soon as it has hardened sufficiently and keep it there for
ten minutes, then in the water at 77 F. and keep it there for fifteen
minutes. When the sample has remained in the water for the speci-
fied time it is ready to penetrate.

" The aluminum rod C should be pressed up through the clamp
E so that it will be at such a height that the glass cup will easily
pass under it when placed on shelf D.

" A sample in tin box should now be placed in the glass cup and
removed in it covered with as much water as convenient without
spilling.

" The glass cup containing sample is placed on shelf D under
C. Insert brass rod with needle into C and secure by tightening
binding screw B, lower C until the point of the needle very nearly
touches surface of sample; then, by grasping the frame with two
hands at S and S', cautiously pull down until needle is just in con-
tact with surface of sample.

" This can best be seen by having a light so situated that,
looking through the sides of the glass cup, the needle will be reflected
in the surface of the sample. After thus setting the needle, raise
counterweight H slowly until the foot of the rack G rests on
the head of rod C; note reading of dial, place thumb of right
hand on R and press button F with forefinger, thus opening the
clamp.

" Hold open for five seconds and then allow it to close. The
difference between the former reading of the dial and the present
is the distance penetrated by the needle, or the penetration of the
sample. Raise rack, loosen binding screw B raise rod through

564 THE MODERN ASPHALT PAVEMENT.

clamp, leaving the needle sticking in sample. Remove needle
from sample, clean well by passing through a dry cloth, replace
needle in C and the machine is ready for another test.

" Do not clean needle on oily cloth, or waste.

" Do not allow rack to descend too rapidly on rod C as it
may force C through the clamp, thus spoiling the reading.

" After using the machine, leave it so the top of the rack is
just level with its base. You will thus prevent dust from entering
and getting into pinion. When not in use keep machine covered
with a cloth to protect from dust.

" Examine point of needle from time to time with magnifying
glass to see that it is not injured in any way.

"If the needle is found defective remove by heating the brass
rod, when the needle can be withdrawn with pincers. Break eye
from one of the extra needles and press into brass rod previously
heated.

" If needle does not stay in well, insert it with a small lump of
asphalt.

"If when this framework is supported on the point of the
needle it does not balance so that the aluminum rod C stands
perfectly perpendicular, the frame is bent and should be straight-
ened until the rod stands perpendicular. This can easily be done
by hand.

"If rack G does not descend readily of its own weight when
counter weight H is raised, it is likely that dust has gotten into
the pinion. To get at pinion to clean, remove dial K and bear-
ing T, when pinion can be pulled out sufficiently far to clean.

"Never oil rack and pinion, as it prevents a free movement
of rack.

"If machine is unsatisfactory write and explain trouble.

"Test for Susceptibility to Changes in Temperature. The
standards that I have adopted for this test are:

"The distance penetrated by the No. 2 needle into the sample
at 32 F. in one minute with 200 grams on frame.

"The penetration at 77 F., as described before, and the pene-
tration into the sample of the No. 2 needle in five seconds at 100 F.
with 50-gram frame. In some cases I use 100 grams, which

METHODS OF ANALYSIS.

565

is preferable if the depth of the sample will permit. In all cases
when you give a penetration of cement state in parentheses how
it was made, as for example (No. 2N., 5 sec., 50 grams, 100)
means that the penetration was made with a No. 2 needle pene-
trating 5 seconds with 50-gram frame at 100 F.

"If a statement is made like this there can never be any doubt
about the figures and they will be understood by all familiar with
the machine."

New York Testing Laboratory Penetrometer, In the Dow in-

FIG. 30.

strument, all the parts are of very light construction and require
a certain delicacy of touch to use it satisfactorily, and consequent-

566

THE MODERN ASPHALT PAVEMENT.

ly ; it is not easily manipulated by such men as are found at paving
plants, although these disadvantages are of slight consideration
in the laboratory. The shelf upon which the sample to be tested
is placed is fixed and the needle must, therefore, be brought

FIG. 37.

down toward the surface of the bituminous cement until it is in
contact Avith it. On each side of the shelf which holds the sample
are the tw r o rods, of the frame from which the weight which acts
upon the needle is suspended. These are much in the way in
adjusting the needle in contact with the surface of the bituminous
cement, and restrict the size of the sample of the latter.

METHODS OF ANALYSIS. 567

In the Penetrometer (Figs. 36 arid 37), an attempt has been
made to overcome these features, without seriously imparing the
accuracy of the instrument. The sample to be tested, which may,
if desired, be of large or small size, is placed upon a table which is
supported upon a screw, enabling the needle to be set at zero,
and the sample to be brought up to contact with the point by
elevating it with the screw. The entire weight acting upon the
needle is contained in the tube which holds it, and. being placed
at the lowest possible point in the tube, does not tend to seriously
divert it from the vertical on the release of the clamp and makes
it possible to do away with the frame which is so much in the
way in the Dow apparatus. The Penetrometer is given greater
stability than the Dow apparatus by making a much larger and
firmer clamp to hold the tube, while all the other parts of the
instrument are rigidly constructed. The increased friction of
the clamp on the rod reduces the extent of the penetration of
the needle by one- or two-tenths of a millimeter, but this is prac-
tically of no consequence, as it is a variation not greater than that
due to different needles or to the personal equation of various
operators.

The determination of the depth to which the needle has pen-
etrated is arrived at by a similar means to that employed in the
Dow machine, a rack and pinion, but the rod to which the rack
is attached is prevented from moving, not by a balance weight
attached to a cord, as in the Dow machine, but by a spring ex-
erting a slight pressure against its side. This does away with the
opportunity for the cord and balance weight to be knocked out
of place or lost, and makes the apparatus much more convenient
to move from place to place. The penetration of the needle is
registered in tenths of millimeters, on the same scale as in the
Dow machine, and the time during which the weight acts is made
five seconds instead of one, as is the case of the Bowen apparatus.
The detail of the manipulation of the tests is the same with the
Penetrometer as with the Dow apparatus, previously described.

The Penetrometer is manufactured by Messrs. Howard &
Morse, 1197 DeKalb Avenue, Brooklyn, X. Y.

Flow Test. The consistency of asphalt cements can also be
controlled by means of a flow test. This is a comparative one

568

THE MODERN ASPHALT PAVEMENT.

and gives nothing but ocular evidence as to the relative softness
of two cements at or near their flowing point. The test consists
in making, in a suitable mould, cylinders f inch long and f inch
in diameter, of a standard cement and of the one to be examined,
placing them on a brass plate with corrugations corresponding
in size to that of the cylinders and exposing them at an angle of
45 to a temperature at which the cements will soften and flow.

FIG. 38. Flow Plate.

Cements made of the same asphalt and flux are of the same con-
sistency if they flow to the same length, Fig. 38.

As a quick, rough test this is very satisfactory; but care must
be taken that the cylinders are exposed to a uniform temperature
and that one part of the brass plate is not hotter than another,
which may readily happen if it touches hot metal or any good

METHODS OF ANALYSIS. 569

conductor. The plate, for safety, should only be warmed by air
and should be isolated from contact with metals by asbestos, paper
or wood.

Cylinders are made by softening the cement to be examined
until it can be rolled out on a board to about the proper size. It
is then pressed in a brass mould of the exact size, which is made in
halves, and cut off to the right length with a hot knife. With any
particular cement a weighed amount known to make a cylinder of
proper size may be taken and rolled to the right length instead of
using a mould.

The cylinders, while still warm, are pressed upon the brass
plate until they adhere and allowed to come to a constant tem-
perature by immersion in water before warming for the flow test.
The cylinders must stick to the flow plate in the beginning or they
may slide instead of flow. The flow plates are 8x2| inches hi
size and have corrugations for four cylinders. As has been said
the selection of some means of affording a uniform temperature is
the most difficult one. In the New York Testing Laboratory,
a carefully regulated rectangular air bath illuminated by a
small incandescent bulb and containing a mica covered window
in the door is employed. At the plants a box heated by a
<;oil of pipe through which steam is conducted can be arranged;
or, for rough work, the plate is placed over a stove with-
out being in contact with the metal. In trying any new
method of heating it is well to put duplicates of the same material
on different parts of the plate and see if they flow alike. This will
determine whether the oven is sufficiently uniformly heated for
practical purposes.

Flows of Trinidad asphalt cements are made at about 160 F.,
others at somewhat lower or higher temperatures, as the case may
demand. One kind of cement cannot, of course, be compared with
that made from another bitumen, even if they have the same pene-
tration at 78 F.

Composition of Asphalt Cement. The percentages of bitumen,
organic insoluble matter and inorganic or mineral matter in all
asphalt cements can be determined exactly as in refined asphalts.

The naphtha soluble bitumen is sometimes sought with a view to

570 THE MODERN ASPHALT PAVEMENT.

its examination and the determination of the nature and amount of
flux which has been used in making the cement. .This is done in the
same way as with refined asphalts or fluxes, but the naphtha
solution is evaporated and the residual bitumen examined. Al-
though it will contain the malthenes of the asphalt as well as those
of the flux used in making the cement, the percentage of the former
being known for any given asphalt, it is possible to calculate the
latter if the cement has not been maintained at a high temperature
in a melted state for too long a time with volatilization and loss of
oil. From the physical characteristics and distillation the nature
of the flux can generally be determined as between an eastern
or California residuum, and the presence of coal-tar or dead-oil
is easily detected.

The bitumen in asphalt cements holding much organic matter
can be estimated only by percolation with the Gooch crucible,
but in Trinidad and other cements carrying much mineral matter,
when examined in large number, the bitumen can be more expedi-
tiously determined by the centrifugal machine. The centrifugal
machine in use for this purpose in the New York Testing Labora-
tory, at Maurer, N. J., is a large Troy laundry extractor with a bas-
ket 30 inches in diameter, which has been filled about the circum-
ference with solid boxwood, leaving an opening 11 inches in
diameter in the centre. In this boxwood are bored about three
dozen one-inch holes to a depth of 6^ inches, sloping downward
at an angle of 15, provided with metal liners, in the bottom of
which a piece of sponge is placed to form a cushion with water, and
which in turn hold the glass tubes, about 1 inch in exterior diameter
and 8 inches long, weighing 50 to 60 grams, in which, after being
accurately weighed, is placed 1 gram of the asphalt cement stirred
up with bisulphide of carbon to reach to a height not greater than
4J to 4J inches in the tube and amounting to 30-35 c.c. of solvent.
The tubes and substance thus prepared are placed in the centrifugal
in such a way as to balance the basket and the power is applied
to give a revolution of 1500 per minute. This is kept up for fifteen
minutes, when the tubes are taken out and decanted carefully, with-
out pouring off any sediment into flint 8-ounce wide-mouth bottles
labelled with the same number as the tubes. More bisulphide of car-

METHODS OF ANALYSIS. 571

bon is then added, the sediment thoroughly mixed with it by means
of an iron rod, which is afterwards washed off with the solvent,
and the tubes again placed in Ihe centrifugal and run ten minutes. .
The decanting is repeated into the correction bottle, more solvent
added as before, and the tubes swung a third time. The third
decantation usually leaves the residue free from any amount of
bitumen which would influence the results. The tubes are placed
in a warm spot to volatilize the remaining solvent and when dry
are weighed. In the meantime the bisulphide of carbon solution
is burned for a correction, as in the analysis of refined asphalts,
and the weight added to that of the tube. The loss of weight of
the tube gives the percentage of bitumen in the cement.

An excellent power centrifuge holding six tubes which is
driven by electricity is furnished by the American Name Plate
Company, 62 Sudbury Street, Boston, Mass.

Change in Consistency of Asphalt Cements on Maintaining
in a Melted Condition. This change can be found by heating some
of the cement to any desired temperature, as hi the determination
of loss at 325 F. in fluxes and making penetrations before and
after heating.

This treatment, however, is much more severe than any that
a cement would ever receive at a plant, as the surface, as com-
pared to the volume under treatment, is very much larger than
is the case in a melting-kettle or dipping-tank. Such determina-
tions are, in consequence, of relative value only in comparing
cements made with different fluxes.

Mineral Aggregate. The mineral aggregate is determined in
the same manner as in solid bitumens.

Examination of the Finished Surface Mixture. Samples of
surface mixture are examined as to the per cent of bitumen they
contain and the grading of the mineral aggregate.

Bitumen. The amount of bitumen is determined in one of two
ways:

1. A funnel, 2J inches in diameter, with a short stem, is placed
in a conical flat-bottom assay flask, holding about 250 Q.C.
A Schleicher & Schiill 9 cm. 597 filter-paper is folded and placed hi
the funnel. Ten grams of the surface mixture are weighed out

572 THE MODERN ASPHALT PAVEMENT.

in fair-sized pieces on the balance to be described later and placed
upon the filter. With a washing-bottle provided with two tubes
through its cork, one reaching to the bottom of the bottle and the
other only just passing the cork, but with a capillary orifice, a
small stream of bisulphide of carbon can be delivered on inverting
the flask without the necessity of using pressure from the mouth
and inhaling the noxious vapor of the solvent. With this bottle
a fine stream is directed on the surface mixture, but no more than
it can absorb. It is allowed to stand until it has softened and
settled upon the filter. The latter is then filled up to an eighth
of an inch below the rim and the funnel covered with a 2J-inch
watch-glass. It is not filled up at first, as before the mixture has
been softened and settled upon the paper the solvent would have
run through the filter-paper and would not have been used econom-
ically. As the percolation goes on the solvent is renewed, and
if it goes too slowly the rate may be hastened by washing between
the paper and the funnel with bisulphide, which will Dissolve the
bitumen, which may have hardened and closed the pores by evapo-
ration, or by lifting the filter a little and letting it drop back.
On the day the analysis is started the sand is washed as clean as
possible, but nothing more is done. The filter with the sand and
the percolate is allowed to stand overnight to permit anything
that has run through to settle out.

In the morning the funnel is placed in a clean assay flask and
the percolate is carefully decanted into a correction bottle, being
careful not to disturb the sediment.

Some bisulphide of carbon is poured on this, it is shaken up
and poured back on the filter, the first assay flask being thoroughly
cleaned with a feather and everything brought upon the original
filter-paper. The mineral aggregate is washed clean with the solvent.

The percolate, or solution of bitumen, in bisulphide of carbon
is poured from the correction bottle into a dish, burned, ignited,
and the correction obtained.

In the meantime the mineral aggregate after drying is separated
from the filter over a piece of glazed paper by scraping with a
blunt spatula or rubbing between the fingers in an appropriate way
until all the mineral matter that can be removed is separated,

METHODS OF ANALYSIS. 573

taking care, of course, not to detach any fibres of the paper. It
is then dusted into a weighed No. 2 Royal Berlin porcelain cnicible
and set aside. The filter-paper, containing much fine mineral
matter in its pores, is burned either with the correction in its dish
or in any satisfactory way, its ash and the correction added to the
mineral aggregate and the crucible's entire contents, after one
is assured that no trace of solvent remains, is weighed. The
difference in the weight of the aggregate and the ten grams of
surface taken is that of the bitumen and gives the per cent of
bitumen in the mixture, which should be calculated to the nearest
tenth of one per cent. The expression of the percentage in hun-
dredths is beyond the limit of accuracy of the method and is
cumbersome and unnecessary.

Centrifugal Method for the Examination of Surface Mix-
ture. In laboratories where large numbers of surface mixtures
are examined daily, as in that of the author, where the number
has reached 70, the centrifugal method as described on page 570
is commonly used. Ten grams of the surface mixture are weighed
out in a glass tube and submitted to the same treatment as is
applied to asphalt cements. The loss of weight of the tube minus
that of the correction obtained on burning the extracted bitumen
gives the percentage of bitumen in the mixture. Should the
mineral aggregate contain coal or other material too light to
be thrown out by centrifugal action, the decanted bisulphide
must be filtered before burning or the percolation method alone
can be used.

Grading of the Mineral Aggregate in a Surface Mixture. The
mineral aggregate from the porcelain crucible after having been
weighed for the determination of bitumen or the mineral matter
from the glass tube which has been submitted to the centrifugal
process is emptied upon the 200-mesh sieve. The particles of
dust, which are caked together, are broken up by gentle pressure
with the finger tips and the coarser sand grains thoroughly cleaned
by attrition. When nothing further passes the sieve the residue
is transferred hi any convenient way to the pan of a balance,
preferably one which, while weighing accurately to a hundredth
of a gram or one-tenth per cent of the amount of surface mixture

574

THE MODERN ASPHALT PAVEMENT.

taken, does not require the use of weights but can be rapidly
manipulated.

Such a balance is supplied by the C. H. Stoelting Co., 31 W.
Randolph St., Chicago, 111., or Eimer & Amend, New York, in the
Chaslyn balance, Fig. 39. This is a beam balance, very much of
the Westphal specific gravity type, which weighs readily to 10 m^s.
by moving ring.? of different weight along the beam. It is exactly
suited for rapid work with surface mixtures where results no closer
than iV of 1 per cent are sought. With this balance the weight

FIG. 39. Chaslyn Balance.

of the residue on the 200-mesh sieve is obtained and the differ-
ence between this and the weight of the mineral aggregate gives the
percentage of 200-mesh material and filler in surface mixture. It is
a determination by loss, and so no effort is necessary to save the
dust which passes the sieve, unless it is desired tc determine its
chemical nature or subdivide further by elutriation when it
should be caught in a closely fitting pan.

METHODS OF ANALYSIS.

575

The other sieves are used in succession after the 200-mesh
with the precautions mentioned under sands and the percentages
passed by each determined.

The percentage of bitumen, dust, or filler, and various sized
sand should amount to 100 per cent.

The results are reported on the following form:

NEW YORK TESTING LABORATORY,

MAURER, N. J.,

Mesh composition and quality of

Received from . .

Test No.
Test No.
Test No.
Test No .

Sample No.
Sample No.
Sample No.
Sample No.

Test No.

Standard Mixture.

Per Cent.

PerCent.

Heavy
Traffic

Light
Traffic

No.

Pass-
ing

Total

Pass-
ing

Total

Pass-
ing

Total

Pass-
ing

Total

10.5

10.0

Bit.

13.0

10.0

200

13.0\ 9ft n
13. 0/ 260

J18.0

100
80

24.0

11.0

8.0]

]

5.0[16

^24.0

3.0J

Penetration of A. C.

Pat paper stain. . . .

Remarks:

576 THE MODERN ASPHALT PAVEMENT.

Method for the Examination of Asphaltic Concrete and As-
phalt Blocks. The quantity of concrete which should be taken
for the analysis will depend upon the size of the largest particles
of the mineral aggregate.

Three hundred grams of an asphalt block, 500 grams when
the largest particles range from ^ to J inch, or 1000 grams when
1 inch particles predominate, will be sufficient. Several com-
parative analyses of the latter class, employing respectively, 1,
2, and 10 kilos of the sample, were in very close agreement.

In order to retain the original relation of the various particles
of the mineral aggregate, it is obvious that none of the stones
should be fractured in taking the portion for analysis, and the
sample should accordingly be softened by heating so that it may
readily be broken apart by hand.

Analysis by Decantation. An appropriate amount of the
sample is placed in a metal beaker or measure, covered with
carbon disulphide and allowed to settle for two hours; decant
and cover with fresh carbon disulphide; repeat this treatment
until the solution becomes clear.

Burn off the carbon disulphide to recover the mineral matter
carried over with the extract, in a platinum or porcelain dish.
Dry and weigh residue together with the correction. The loss
from weight of concrete taken is bitumen.

The mineral aggregate may now be further examined by care-
fully compacting in a suitable vessel to determine the amount
of voids or sifted through a series of perforated metal and wire
cloth screens to ascertain the size of the various particles or its
grading.

Centrifugal Method. While the preceding method is not lack-
ing in accuracy, when intelligently manipulated, it is far from
rapid, especially when a large mass of material is under considera-
tion, and more or less extravagant in the use of solvent. The
following method has therefore been developed, applying the
essentials of the special centrifugal extractor as closely as they
could be gathered from a brief verbal description, very kindly
supplied by its ingenious inventor, Mr. August E. Shutte.

The body of the apparatus consists of a double walled iron

METHODS OF ANALYSIS.

577

casting, the inside dimensions of which are 9^ inches diameter by
7 inches deep, set upon three legs and provided with a tightly
fitting cover carrying a reservoir, also consisting of a single cast-
ing. The combined portions form in effect a large Soxhlet
extractor, as will be observed upon inspection of the illustrations,
Figs. 40 and 41.

FIG. 4.

A slotted shaft extends through the bottom of the extractor
and a gland is provided for making a tight joint. This, as well as
the gasket under the cover, may be packed with ball lamp wicking
and soap.

The sample is broken down as finely as practicable and placed on

578

THE MODERN ASPHALT PAVEMENT.

the cast iron plate which fits upon the end of the slotted shaft. A
filter ring cut from No. 80 soft roofing felt about .090 inches thick

FIG. 41.

is placed under the rim of the dome, which is now bolted down
hard upon the plate over the sample. Fig. 42.

A dome 8 inches diameter by 3^ inches high, spun out of No. 16
brass, will accommodate one kilogram of concrete. The plate is

METHODS OF ANALYSIS.

579

Si inches diameter. The bolt should be tool steel, -^ inch
diameter. A shallow brass pan is placed in the device to receive
the extract as it is thrown out of the revolving dome.

The dome having been placed in position upon the shaft, about
300 cc. carbon disulphide is poured upon the sample through the
perforations in the top and the apparatus closed.

Steam is turned into the lower jacketed bod}' and cold water
through the condenser. After allowing a few minutes to become
warm, the dome is revolved at a rate of from 1500 to 1800 R. P. M.
by any suitable power, and as the solvent carrying the bitumen
is thrown out into the receiving pan, it is volatilized and passes into

FIG. 42.

the reflux condenser, falling back into the reservoir on the cover.
AVhen about 250 cc. has accumulated, it is automatically syphoned
back into the dome upon the sample as fresh solvent.

As but three minutes are required to swing the sample dry,
it is not necessary to revolve the dome continuously; stop it after
this period of time and do not start again until the gauge glass on
the reservoir indicates that a fresh portion of solvent has passed
over upon the sample.

About four treatments are sufficient for a kilogram sample,
after which the flow of steam is discontinued and cold water re-
versed through the jacket to hasten cooling.

580 THE MODERN ASPHALT PAVEMENT^

The bitumen free from all but minute traces of the finest portion
of the mineral aggregate, which may be disregarded, will be found
in the pan, while the clean aggregate is allowed to dry spontane-
ously and then weighed, the bitumen being determined by dif-
ference. It is examined further as desired, by sifting, etc., as
previously described.

The entire extraction may thus readily be completed in less
than an hour, with little loss of solvent.

In handling asphalt block mixtures employing 300 grams of
sample and but three portions of cold carbon disulphide, an ex-
traction may be made in about twenty minutes. In the latter
case, the solvent is recovered in a separate distilling apparatus
arranged to receive the pan holding the extract directly, or an
accumulation is worked over at convenient times.

Density and Voids in Surface Mixtures. The density which
a surface mixture can attain on compaction is often a source of
information as to its quality, and from this the voids in the com-
pacted material can be calculated. The mixture is compressed
in a mould made for the purpose. This should consist of a base
8 inches long, 5 inches wide and 3| inches high. On top of this
base are found a cylindrical boss or post 1J inches in diameter
and 1 inch high, and a hole of the same or a little larger diameter,
opening into a hollow in the base. A hollow cylindrical mould
or sleeve of steel of the same internal diameter as the boss and
3J inches high is provided and a solid plunger of steel to fit this

accurately. About grams of the surface mixture is heated

in a deep iron dish to 325 F., the cylinder and plunger being also
heated. The cylinder is placed over the post and filled with the
hot mixture and compressed with the plunger and sharp blows of
a heavy hammer, or better in any suitable press such as the
Riehle or Olsen tensile and crushing power machines. When
ultimate compression has been attained in this way the cylindrical
mould is removed from the boss and turned over or reversed and
again placed on the boss. The space formerly occupied by the boss
now gives an opportunity for the insertion of the plunger and
compression of the cylinder of asphalt mixture from the other

METHODS OF ANALYSIS. 581

end. Finally, the mould is placed over the opening in the base
and the cylinder of surface knocked out with a few blows of the
plunger. It should be between 1 and 2 inches long.

Its density can be determined by weighing it in air and water,
but the quickest way is to measure its length with calipers to .01
inch and find its volume in cubic centimeters by reference to the
table on page 582.

The weight of the cylinder divided by the volume gives the
density. This should not fall below 2.20 for good mixtures, made
with quartz sand. The voids in such a cylinder can be calculated
from the known proportions and density of the materials of which
it is composed, as can be seen from the following example:

The customary surface mixtures consist, in parts by weight, of:

Sand 75%

Dust 10

Trinidad asphalt cement 15

100%

By volume this would be, the density of the sand being 2.65, that
of dust 2.60, and that of the asphalt cement 1.25:

Sand 75/2 65... 28.30 or 64.10

Dust 10/2.60... 3.85 " 8.72

Asphalt cement.. 15/1. 25... 12.00 " 27.18

44.15 100.00

The cement in use, being 27. 18% of the entire volume of the mix-
ture, will fill the voids which ordinarily exist in the mineral aggregate
if the mixture receives its ultimate compression and density. If
the voids are larger it will be too little and some voids will remain
unfilled; if they are smaller it will be too much and will make the
surface too yielding.

Considering the proportions given as being theoretically cor-
rect, the density of the resulting mixture, when it receives its
ultimate compression, should be:

64 . 1 vols. at 2 . 65 density 1 . 699

8.7 " " 2.60 " 226

27.2 " " 1.25 " .340

100.0 2.265

582

THE MODERN ASPHALT PAVEMENT.

TABLE FOR DETERMINING CONTENTS IN CUBIC CENTIMETERS
OF CYLINDERS 1.25 INCHES IN DIAMETER AND VARIOUS
HEIGHTS IN INCHES.

Height,
Inches.

Cubic
Inches.

Cubic
Centimeters.

! Height,
Inches.

Cubic
Inches.

Cubic
Centimeters.

.95

1.17

18.17

.48

1.81

29.66

.96

1.18

19.34

.49

1.83

29.99

.97

1.19

19.50

.50

1.84

30.15

.98

1.20

19.66

.51

1.85

30.32

.99

1.22

19.99

.52

1.86

30.48

.00

1.23

20.16

.53

1.87

30.64

.01

.24

20.32

.54

1.89

30.97

.02

.25

20.48

.55

1.90

31.14

.03

.26

20.65

.56

1.91

31.30

.04

.28

20.98

.57

1.93

31.63

.05

.29

21.14

.58

1.94

31.79

.06

.30

21.30

.59

1.95

31.95

.07

.31

21.47

.60

1.96

32.12

.08

.33

21.79

.61

1.98

32.45

.09

.34

21.96

.62

1.99

32.61

1.10

.35

22.12

.63

2.00

32.77

1.11

.36

22.28

.64

2.01

32.94

1.12

.37

22.45

.65

2.02

33.09

.13

.39

22.78

.66

2.04.

33.42

.14

.40

22.94

.67

2.05

33.59

.15

.41

23.11

.68

2.06

33.76

.16

.42

23.27

.69

2.07

33.92

.17

.44

23.60

.70

2.09

34.25

.18

.45

23.76

.71

2.10

34.41

.19

.46

23.92

.72

2.11

34.58

.20

.47

24.08

.73

2 12

34.74

.21

1.48

24.25

1.74

2'l4

35.07

.22

1.50

24.58

1.75

2.15

35.23

.23

1.51

24.74

1.76

2.16

35.40

.24

1.52

24.91

1.77

2.17

35.56

.25

1.53

25.07

1.78

2.19

35.89

.26

1.55

25.40

1.79

2.20

36.05

.27

1.56

25.56

1.80

2.21

36.21

.28

1.57

25.73

.81

2.22

36.38

.29

1.58

25.89

.82

2.23

36.54

.30

1.60

26.22

.83

2.25

36.87

.31

1.61

26.38

.84

2.26

37.03

1.32

1.62

26.55

.85

2.27

37 20

1.33

1.63

26.71

.86

2.28

37.36

1.34

1.65

27.04

1.87

2.29

37.53

.35

1.66

27.20

1.88

2.31

37.85

.36

1.67

27.35

1.89

2.32

38.02

.37

1.68

27.53

1.90

2.33

38.18

.38

1.69

27.69

1.91

2.34

38.35

.39

1.70

27.86

1.92

2.36

38.67

.40

1.72

28.18

1.93

2.37

38.84

.41

1.73

28.35

1.94

2.38

39.00

.42

1.74

28.51

1.95

2.39

39.16

1.43

1.75

28.68

1.96

2.41

39.49

1.44

1.77

29.00

1.97

2.42

39.66

1.45

1.78

29.17

1.98

2.43

39.82

1.46

1.79

29.33

1.99

2.44

39.98

1.47

1.80

29.50

I 2.00

2.45

40.15

METHODS OF ANALYSIS. 583

The density of the compacted mixture is usually found to be not
over 2.22, and at this figure there would be about 2 per cent of
voids. At a density of 2.18 the voids would reach 3.7 per cent.

The density of coarse mixtures, such as asphalt blocks or as-
phaltic concrete, may be determined by compressing the hot
mixture in a mold of appropriate size under suitable pressure in a
power press.

Asphalt blocks are weighed as received, first in air and then
in water, in the most convenient manner, and the density cal-
culated from the data thus obtained.

The specific gravity of the mineral aggregate in such mixtures
is most conveniently determined in a Jackson's apparatus, pre-
viously referred to.

Water Absorption of Surface Mixtures. Cylinders prepared as
previously described, or, if such a mould is not at hand, compressed
to the best possible extent in an ordinary diamond mortar, are
weighed in air and then suspended by a horsehair and immersed
in distilled water at ordinary temperature and again weighed, while
still immersed in water and suspended by the hair in the same
way, at intervals of 1, 2, 7, 15 days and one month. The gain in
weight shows the water absorbed, which is calculated to milligrams
per square centimeter, or inch, or pounds per square yard, as may
be desired, by determining from its dimensions the number of
square inches of surface the cylinder has.

Where cylinders are of such density that the surface is but slightly
acted upon by water and there is no disintegration, they may be
carefulty wiped off and weighed directly.

An example of the amount of water absorbed by a good mix-
ture is seen in the following determinations:

584

THE MODERN ASPHALT PAVEMENT.

GAIN IN GRAMS PER SQUARE INCH AND POUNDS PER SQUARE
YARD OF TRINIDAD SURFACE MIXTURE FROM THE LONG
ISLAND CITY PLANT OF THE BARBER ASPHALT PAVING
COMPANY.

Height of cylinder 1.15 inches

Diameter 1 . 25 "

Surface 6 . 97 square inches

Interval

Mgs. Per Square Inch.

Pounds Per
Square Yard.

After

Immersion.

Gain in

Total

Interval.

Gain.

24 hours

.0169

48 "

.0021

.0190

.0540

7 days

.0092

.0282

.0803

15 "

.0045

.0327

.0930

28 "

.0935

.0362

. 1031