IMAGE EVALUATION
TEST TARGET (MT-3)

>"a

%°

f ^

^ W/J.

1.0

I.I

1.25

f llitt
t m

i.4

11^

1 2.0

1.6

■/y >

-r^

^i ^i \^

Photographic

Sciences
Corpordtion

23 WEST MAIN STREET

WEBSTER, NY 14580

(716) 872-4503

L17

■ *

%^^

f^?

-?<.

^\%<-'^

<?>

-^^N-

CIHM/ICMH

Microfiche

Series.

CIHM/ICMH
Collection de
microfiches.

Canadian Institute for Historical Microreproductions Institut canadien de microreproductions historiques

1980

Technical and Bibliographic Notes/Notes techniques et bibliographiques

The Institute has attempted to obtain the best
original copy available for filming. Features of this
copy which may be bibliographically unique,
which may alter any of the images in the
reproduction, or which may significantly change
the usual method of filming, are checked below.

D
D

□

Coloured covers/
Couverture de couleur

Covers damaged/
Couverture endommag^e

Covers restored and/or laminated/
Couverture restaurde et/ou pellicuide

□ Cover title missing/
Le titre de couverture manque

□ Coloured maps/
Cartes g^ographiques en couleur

n Coloured ink (i.e. other than blue or black)/
Encre de couleur (i.e. autre que bleue ou noire)

□ Coloured plates and/or illustration^/
Planches et/ou illustrations en couleur

Bound with other material/
Reli6 avec d'aut. 's documents

Tight binding may cause shadows or distortion
along interior margin/

La reliure serree peut causer de I'ombre ou de la
distortion le long de la marge int^rieure

Blank leaves added during restoration may
appear within the text. Whenever possible, thesa
have been omitted from filming/
II f.e peut que certaines pages blanches ajout^es
lors d'une restauration apparaissent dans le texte,
mais, lorsque cela 6tait possible, ces pages n'ont
pas 6t6 film^es.

Additional comments./
Commentaires suppl6mentaires;

L'Institut a microfilm^ le meilleur exemplaire
qu'il lui a 6t6 possible de se procurer. Les details
de cet exemplaire qui sont peut-§tre uniques du
point de vue bibliographique, qui peuvent modifier
une image reproduite, ou qui peuvent exiger une
modification dans la mdthode normale de filmage
sont indiqu6s ci-dessous.

□ Coloured pages/
Pages de couleur

1 I Pages damaged/

Pages endommag^es

Pages restored and/or laminated/
Pages restaur6es et/ou pellicul6es

Pages discoloured, stained or foxed/
Pages d^colordes, tachetdes ou piqu^es

I j Pages detached/

Pages ddtach^es

Showthrough/
Transparence

□ Quality of print varies/
Qualit^ in^gale de I'impression

I I Includes supplementary material/

D
D

Comprend du materiel supplementaire

Only edition available/
Seule Edition disponible

Pages wholly or partially obscured by errata
slips, tissues, etc., have been refilmed to
ensure the best possible image/
Les pages totalement ou partiei'ement
obscurcies par un feuillet d'errata, une pelure,
etc., ont 6t6 film^es d nouveau de fapon d
obtenir la meilleure image possible.

This item is filmed at the reduction ratio checked below/

Ce document est filmd au taux de reduction indiqu6 ci-dessous.
10X 14X 18X 22X

26X

30X

::^

12X

16X

20X

24X

28X

32X

itails
5 du
lodifier
r une
Image

The copy filmed here has been reproduced thanks
to the generosity of:

Library,

Geological Survey of Canada

The images appearing here are the best quality
possible considering the condition and legibility
of the original copy and in keeping with the
filming contract specifications.

L'exemplaire filmd fut reproduit grSce d la
gdn^rositd de:

Blbliothdque,

Commission G6ologique du Canada

Les images suivantes ont 6t6 reproduites avec le
plus grand soin, compte tenu de la condition et
de la nettetd de l'exemplaire film6, et en
conformity avec les conditions du cuntrat de
filmage.

Original copies in printed paper covers are filmed
beginning with the front cover and en 'ng on
the last page with a printed or illustrated impres-
sion, or the back cover when appropriate. All
other original copies are filmed beginning on the
first page with a printed or illustrated impres-
sion, and ending on the last page with a printed
or illustrated impression.

Les exemplaires originaux dont la couverture en
papier est imprimde sont film6s en commenpant
par le premier plat et en terminant soit par la
dernidre page qui comporte une empreinte
d'impression ou d'illustration, soit par le second
plat, selon le cas. Tous les autres exemplaires
originaux sont film6s en commenpant par la
premidre page qui comporte une empreinte
d'impression ou d'illustration et en terminant par
la dernidre page qui comports une telle
empreinte.

The last recorded frame on each microfiche
shall contain the symbol — ♦- (meaning "CON-
TINUED"), or the svmbol V (meaning "END"),
whichever applies.

Un des symboles suivants apparaitra sur la
dernidre image de cheque microfiche, selon le
cas: le symbole — ^ signifie "A SUIVRE ", le
symbole V signifio "FIN".

Maps, plates, charts, etc., may be filmed at
different reduction ratios. Those too large to be
entirely included in one exposure are filmed
beginning ir the upper left hand corner, left to
right and top to bottom, as many frames as
required. The following diagrams illustrate the
method:

Les cartes, planches, tableaux, etc., peuvent dtre
film^s d des taux de reduction diff^rents.
Lorsque le document est trop grand pour dtre
reproduit en un seul clich6, il est film^ d partir
de Tangle sup6rieur gauche, de gauche ^ droite,
et de haut en bas, en prenant le nombre
d'images n6cessaire. Les diagrammes suivants
illustrent la mdthode.

errata

I to

i pelure,

on d

32X

-^-^^

*=r

/ / / O

H^-^ —

r A rvi .
THE

M J

HIM \m i)()L(;las puo( e

•<^.

.1'

FOR

EXTRACrnXG COPPER FROM ITS ORES.

AVITII A\ APPK.NDIX

iNci.rniNG

NOTES OX THE TIIKATMKNT OF SILVER
AND GOLD OlIES, .

AND A rr.ATi:.

BOSTON:

ri;i;ss of a. a. kixcim ax.
1S7().

THE

HUNT AxND DOUGLAS PROCESS

FOR

EXTRACTING COPPER FROM ITS ORES.

WITH AN APPENDIX

INCLUDING

KOTES ON THE TREATMENT OF SILVER
AND GOLD ORES,

AND A PLATE.

BOSTON:
PRKSS OF A. A. KINGMAN.

1876.

• •

t <
t I

• • • •

• I •

I • I

I f

• • •• (

• • •

.•

• •• • . .•

:•■• •■

• • • «

• I •

TABLE OF CONTENTS.

Intuoduct.on: Wet and Dry rroccsses of Copper Extraction

Tlie Hunt and Donglas Process ; Working Directions
I. Grinding the Ore
II. Calcining tlie Ore

III. Dissolving the Copper

IV. Precipitating tJie Coj)per
V. Melting and retining the Coi)per

VI. Arrangement of tlie Plant .

8
12
17
18
19

Appendix.

II.

III.

IV.

VI.

VII.

VIII.

Assaying for Copper and for Iron
Preparation and use of Iron Sponge .
Use of Tin Plate Scrap .
Chemistry of the Hunt and Douglas Process

Patent Specification

What Ores may be treated by the Process

Practical Working of the Process

The Treatment of Silver and Gold Ores

THE

HUNT AND DOUGLAS COPPER PROCESS,

This is wliat is technically called a wet method, because the
copper is removed from its ores in a dissolved state, the sol-
vent employed in the present process bein<^ a watery solution of
neutral protochlorid of iron and common salt. Most oxydized
compounds of copper— whether obtained artificially by roasting
sul|)huretted ores, or found in nature in the forms of carbonates
and oxyds, — when digested with such a solution are converted
into a mixture of protochlorid and dichlorid of copper, which are
dissolved, wdu'Ie the iron of the solvent separates in the form of
insoluble hydrous peroxyd of iron. When the solution of the
chlorids of copper thus obtained is brought in contact with me-
tallic iron the copper is separated in a metallic crystalline state,
while the iron passes into r.ohition, rej)roducing the i)rotochlorid
of iron; thus restoring its solvent powers to the liquid, which
we shall call " the bath," and fitting it for the treatment of a
fresh portion of copjier ore. This j)rocess of solution and pre-
cipitation can, under proper conditions, be repeated in efinitely
with the same bath, the oidy reagent consumed being the
metallic iron.

The chief advantage wliich wet j)rocesses jwssess over smelt-
ing lies in the economy of fuel. To extract cop|)er from a low
grade ore by smelting, five or six furnace-operations are neces-
sary, and about one ton of coal is consumed for each ton of ore
treated ; while for the various wet processes a single calcination,

(5)

in which not more than three hundred weiglit of coal is con-
sumed for each ton of ore, is the only furnace-operation required
to obtain the metallic copper in a precipitated form known as
cement copper. An important item of cost in wet processes is
the metallic iron employed to separate the metallic copper from
its solutions. The same amount of iron is required to pre-
cipitate a ton of copper whether extracted from a poor or a rich
ore, but as for the smeltino; of the latter much less fuel is
required, it follows that rich ores are generally treated by smelt-
ing rather than in the wet way, any saving of fuel in the latter
being more than compensated for by the cost of iron. No gen-
eral rule however can be laid down to determine what grade of
ore can be more profitably treated by one method or the other,
inasmuch as circumstances of locality, affecting the cost of fuel
and the price of iron, must in each case be taken into account.
The various other wet methods of copper-extraction may be
divided into two classes : those in which the previously oxydized
ore is treated with hydrochloric or sulphuric acid to dissolve the
oxyd of copper, and those in which sulphuretted ore, generally
after a preliminary roasting, is calcined with an admixture of
sea-salt or of sulphate of soda, by which the copper is converted
into chlorid or into sulphate. All of these methods, when
properly applied, effect a pretty thorough extraction of the
copper, but the cost of the reagents which have to be added to
every charge of ore, preclude altogether the use of some of
these methods, except in certain favored localities, and render
them in almost all cases, it is believed, less economical than the
present one with the Hunt and Douglas bath, for wiiich the
followiuff advantajres are claimed :

I. It is a general method adapted to all compounds of copper,
while that by calcination with salt is only aj)plicable to sulphur-
etted ores.

II. It does not require the addition of reagents sucii as acids,
salt or sulphate of soda to each charge of ore, since in the regidar
course of the operation the solvent required for the treatment
of the ore is constantly reproduced.

III. The bath employed being neutral, certain impurities of
the ore, such as arsenic, which pass into solution and con-
taminate the product in the wet processes, remain undissolved,
so that a purer copper is obtained.

IV. As the solution obtained is neutral and free from per-
salts of iron, there is no uiniecessary waste or consumption of
metallic iron in the process of precipitation. Moreover, as the
result of the action of the protochlorid of iron of the bath on
protoxyd of copper, one-third of the copper is obtained as pro-
tochlorid, and two-thirds as dichlorid. Now since the latter
requires for each one hundred parts of copper precipitated only
forty-five parts of iron, it is found in practice that not more
than three-quarters of a ton of iron are consumed to precipitate
one ton of metallic copper, while in the other methods, in which
the copper is obtained as protochlorid, the consumption of iron
amounts to a ton, and in many cases greatly exceeds it.

WORKING DIRECTIONS.

The application of the Hunt and Douglas process to the treat-
ment of copper ore may be considered under the following
heads :

I. Grinding the ore.

II. Calcininji the ore.

III. Dissolving the copper.

IV. Precipitating the co|»per.

V. Melting and refining the copper.
VI. Arrangement of the plant.

I. Grinding the Ore. The degree of fineness to which the
ore must be ground will depend entirely upon the character of
the gangue. If the metal be scattered in fine particles through
an impermeable rock, it will be necessary to grind it to the size
of sand, so that the copper, if a sulphuret, may be exposed to
the oxydizing action of the air during calcinati(m, and to the
solvent action of the protochlorid of iron bath during lixiviation.

If, on tlie contrary, tlie copper-suli)liurot be mixed, as is often the
case, vvitli iron i)yrites, which by calcination becomes porous, tlie
ore need not be r^round so fine. Experiment in each case must
deternnne the point, and upon the decision must depend the
machinery which should be chosen to effect the grinding; —
Cornish rolls being preferable for coarse crushing and stamps
for finer work. Two pairs of rolls, — one ])air of 24 or 30
inches diameter, and one pair of 12 or 15 inches, with a screen
between them to sift out what is not broken sufficiently fine by
the nj)per pair, will crush about twenty tons of stuff in twenty-
four hours so that it will pass through a sieye of fifteen holes to
the linear inch, a dejiree of fineness sufficient for most ores. A
rock-breaker with jaws set close may be substituted for the
upper pail of rolls.

II. Calcining the Ore. It is not necessary to calcine car-
bonates or protoxyds, but mixtures in whicli there is a large
proportion of red or dinoxyd need a slight roasting to convert
at least a ])art of this into protoxyd ; wliile all sulphuretted ores
require much more calcination. The mode of effecting this will
vary with the character of the ore. When it contains 20 p. c.
or n|)ward of sulphur, it may be broken into lumps of an inch
or more in diameter, and exposed to a preliminary roasting in
heaps or kilns, whereby, without the aid of fuel, the greater
part of the sul[)hur will be driven off, and the metallic ingre-
dients more or less completely oxydized. The lumps thus
partially roasted should then be crushed and calcined in a muffie
or reverberatory fui'nace. The calcination of all ores in an
earthy gangue nuist be effected wholly in such furnaces.

The first rule in roasting is to expose the ore at the beginning
to a low heat, which is to be gradually increased as the sulphur
is driven off. If the temperature be too high at the commence-
ment of the operation, the ore, if highly sulj)huretted, may
l)ecome softened and agglutinated or fritted, after which it is
imj)ossible to effect a i)roper roast. But even if this should not
happen, too high a heat at first, or indeed at any stage of the
process, brings the copper into a condition in which it is diffi-

I
i

3 often the
)orous, tlie
case must
epend tlie
ncHng ; —
id stamps
24 or 30
a screen
y fine by
1 twenty-
1 lioles to
ores. A
for the

Icine car-
5 a larofe
) convert
ttod ores

this will

20 p. c.

an inch
isting in

<ireater
: ino;re-
|)s thus
muffle
in nn

nnnino;
ulpliiu"
nence-
may
1 it is
lid not
of the
s diffi-

cultly soluble in the bath. A lono; furnace is more easily
manaired than a short one, since In the former the fire can
always be kept strong and the ore moved forward from a cooler
to a hotter j)ortion, while in a short furnace the gradation of
heat can only be attained by close attention to the firini;.

A lon<: muffle furnace always gives a good roast, as the tile
floor protects the ore from excessive heat, and there is sure to
be an oxydizing atmosphere in the furnace, which is not always
the case in a reverberatory, where the flame comes in contact
with the ore. But the construction of the muffle furnace is
expensive, and a cheap and efficient furnace is a three-hearth
reverberatory. When a number of such furnaces are needed,
they may be built side by sidi', in a row, the rabbling-doors
oi)ening before and behind, and the arches of the whole row
being supported by a stone buttress at each end, — the only
binding necessary. The fire-boxes of adjacent furnaces are
placed side by side. The dimensions which have been found
advantan-eous for these furnaces are as follows : lower hearth
ten feet wide by sixteen feet long; u])per hearths twelve feet
wide by fifteen feet lonjj. The lower hearth is contracted in
width by the fire-i)lace, and the upper hearths in length by the
flues which lead from hearth to hearth. The details of con-
struction are shown in the accompanying plan.

The advantages of such a form of furnace are cheapness of
construction and economy of heat, on accouut of the exposure of
a less amount of cooling surface than in the long reverberatory
with rabbling-doors on the side. On the other hand the upper
hearths are not verv accessible to the rabble rs. If such a
furnace be used, the heat should only be sufficient to thoroughly
dry and warm the ore on the up[)ermost hearth. Oxydation
should take place, with the elimination of the greater part of the
sulphur, on the second hearth, so that when the ore is exposed
to the higher temperature of the lower hearth there may be no
danger of fritting. The quantity of ore which may be roasted
in such a furnace will depend on the character of the ore and
the proportion of 8ulj)hate of copper which it may be desirable

to obtain. If tlic ore is liin;lily sulphuretted and has not re-
ceived a preliminary roast before grinding, only two or three
tons can be calcined in twenty-four hours, whereas double that
quantity may be treated if the ore be poor in sulphur. An ore
with from 15 ]). c. to 20 p. c. of sulj)hur maybe added in charges
of 2500 lbs. and shifted from hearth to hearth every eight hours,
while one containing from 5 p. c. to 7 p. c. .of sulphur may be
shifted every five hours.

If the ore contains no carbonate of lime or magnesia (which
will deprive the bath of the chlorid of iron in the subsequent
operation of solution), the roast need not contain over one-
fourth of its coj)j)er in the state of sulphate. This will be more
than sufficient to repair unavoidable losses in the iron-chlorid of
the bath. The presence of portions of these obnoxious ele-
ments may, however, make it desirable to obtain in the roast a
larger ])roportion of sulphate of copper (which is soluble in
water and by its precipitation by metallic iron yields an iron-
salt). To obtain this the ore should be roasted more slowly
and in larger charges, say of 5000 lbs. each, in which case the
yield of ore from the furnace will be somewhat diminished.

The quantity of fuel consumed will vary with the different
ores, but as a rule one cord of wood will suffice for three tons,
and one ton of coal for eight tons of ore.

When a sulj)huretted ore has been properly roasted it loses,
when being rabbled, that aj)j)arent fluidity which ore still giving
off sulphurous acid exhibits, and when withdrawn and cooled
should have a bright red color. If the heat has b'^on too great
the color of the cooled ore will varv tliroujxh dull red to black.
There is more danger of havinn; too much than too little heat
in the furnace. The ore on the upper hearth should never be
in a glow, and that on the lower hearth should never attain a
higher neat than dull redness. Resides reculatinff the heat, it
is important to attend to tlie admission of air. As the roasting
of the ore is an oxydizing process an abundance of air is es-
sential to the oj)eration, and that this may be sup])lied, the
furnace must possess a good draft and be provided with openings

las not re-
o or three
louble tliat
An ore
in cliarges
gilt lionrs,
xr may be

da (which
ubseqiient
over one-
II be more
clilorid of
xious ole-
lie roast a
soluble in
; an iron-
re slowly
case the
hod.

lirt'erent
iroe tons,

it loses,

ill Ji'ivinfr

cooled

o() great

tt) black.

tie heat

ever be

It tain a

heat, it

•oastinor

r is cs-

ied, the

)ennii;s

sufficiently large and numerous. If the furnace be defective in
these ])oints the ore will be scorched and its coj)per rendered
insoluble by a reducing action on the lower hearth, while the
uj)per hearth will be liable, at the same time, to become too hot.

The more completely the sulj)huret of copper is oxydized in
the roasting, the more thorough will be the subsequent ex-
traction of the copper, but to oxydize the last traces of sulphuret
requires a disproportionate expenditure of time, labor and fuel.
Ujion the relative value of the raw ore, and of labor and fuel,
will therefore depend the degree of thoroughness to which it
may be profitable to carry the extraction of the copper at any
given reduction-works. While it is desirable to oxydize as
completely as consistent with economy the sulphurets of the
ore, it should be borne in mind that a dead roast, as it is called,
or the elimination of that portion of sulphur which, after oxy-
dation, remains combined as sulphate of copper, is to be avoided,
since, as already pointed out, to provide for unavoidable loss of
chlorid of iron it is desirable to leave a portion of sulphate of
copper in tlie roasted ore. The composition of the roast may
be seen from the following examples.

At the Ore Knob Mine in North Carolina, the averajje of the
ore roasted by this process was, according to Mr. Olcott, (Trans.
Amer. Inst. Min. Engineers, vol. III. p. 395.) ;

Copper as sulphate , 3.76

Cop|)er as oxyd 7.75

Cupper iis giilphid .39

11.90
At Phoenixville, Pennsylvania, where the ore contains a con-
siderable quantity of carbonate of magnesia, the eftc'ct of
which has to be neutralized by a large proportion of sulphate
of copper, and where charges of oOOO lbs. of ore are calcined
for twenty-four hours on each hearth of the dimensions above
given, the roasted ore has the following average composition :

Copper as sulphate ....... 1.25

Copper iis oxyd . , . . . . . .1.10

Copper as sulphid .40

2.75

For tlie mctliod of determining by assay the composition of
the roasted ores, see Appendix, p. 21.

III. Diamlvlng the Copper. The solvent or bath employed
for the extraction of the copper is, as has been stated, a neutral
solution of protochlorid of iron with common salt. This proto-
chlorid may be obtained in various ways. In localities where
acids are cheap it is easily made by dissolvino; scrap iron in
diluted muriatic or sulphuric acid ; the first yields directly pro-
tochlorid, the second protosulphate of iron, which when mixed
with a solution of salt gives rise to the protochlorid, together
with a j)ortion of sulphate of soda. In places where acids
are not so easily had, the commercial protosulj)hate of iron
(green copperas) is the most convenient source of the proto-
chlorid, as exj)lained in the specification. 100 lbs. of the
commercial acid and bQ lbs. of scrap iron will make 280 lbs.
of copperas. Knowing the relative cost of these substances at
any locality, it will be easy to calculate whether it is cheaper to
make the copperas or to purchase it. Where highly sulj)hur-
etted coj)per ores or copper pyrites are to be had these, by
calcining at a low red heat (as already stated) yield large
pro})ortions of sul[)hate of copper and sulphate of iron, both of
which are soluble. By leachino; these roasted ores with water
and difjestinrj the solution thus obtained with scrai) iron the
dissolved copper is thrown down as metal, and a solution of
protosulphate of iron obtained, which may be mixed with salt
to form the bath.

In the original specification of the process it was directed in
making the bath by the use of protosulphate of iron to take
280 lbs. of this, (equal to 56 lbs. of metallic iron) and 120 lbs.
of salt, suflicient to convert it into protochlorid. These dis-
solved in 1000 lbs. of water (100 imperial gallons) with a
a farther addition of 200 lbs. of salt made the strongest bath,
but a weaker one was also recommended in which these same
incjredients were to be dissolved in 2000 lbs. of water. Ex-
])erience has shown that the latter is strong enough for the
treatment of all ordinary ores.

osition of

employed
a neutral
his proto-
ies where
p iron in
.'ctly pro-
en mixed
, together
ere acids
e of iron
lie j)roto-
. of the
280 lbs.
tances at
leaper to
sulj)luir-
hese, by
1(1 laro-e
I, both of
h water
ron the
iti(m of
itli salt

Icted in

to take

20 lbs.

|se dis-

Iwith a

bath,

same

Ex-

br the

The bath mav be broiifj-ht in contact with the ore either bv
percolation in leachino; tanks, or by an;itation in vats arranged
with stirrers. If the ore be finely ground and .•slimy, the latter
must be used, but if it is coarse, and contains nothing which
when wetted will form mud, it is best treated by leaching.
When agitation is required the tanks should be rouiul, ten or
twelve feet in diameter, and five or six feet high, and made of
three-inch staves. A convenient stirring apparatus consists of
two oblique blades fixed to the base of a vertical shaft, which
rests on the vertex of a conical bottom. The tii)s of the blades
should reach to within an inch of the sides of the tank, and be
raised about fifteen inches above the level of the botttnn of the
tank at the periphery. The object of thus elevating the stirrer
on a cone above the bottom is to permit the ore to settle below
the blades, so that the stirrer, after having been stopped, can be
started at will ; whereas were the bottom flat and tlie distance
between it and the blades the same at all points, the ore would
accumulate around the shaft and thus escape agitation. The
stirrer should make about twenty revolutions a minute. A vat
of the above dimensions, having a capacity of about f;JOOO gal-
lons, and two-thirds filled with bath, will serve to agitate and
dissolve the copper from 8000 lbs. of roasted ore conta.ning five
or six p. c. of cop})cr oxyd in six to eight hours, the temperature
being from 120° to 150° F. The stirrers are then stopped, the
whole allowed to settle, the clear liquor drawn off into the pre-
cipitating tanks, and the muddy portions into settling tanks,
after which the residue may be washed, first with bath, and then
with water, to remove the adherent coj)per solution.

When ])ercolation can be adopted it is preferable to stirring,
since, though the operation is slower, we are enabled to dispense
with the settling tanks which the latter plan re(|uires, and the
handling of the slimes which accumulate in these. Moreover,
as the solution of the copper takes place in the mass of ore out
of contact of air, a larger proportion of dichlorid of copper is
found and less iron is lost by oxydation than when the solution

holding tlie dissolved iron and copper salts is exposed to tiie air
by constant agitation.

The vats for filtration are made of wood or of brick. For
the latter the bricks are laid in Roman cement and coated
within by a layer of the same cement mixed with sihcate of
soda. Tliis, when afterwards washed with a solution of chlorid
of calcium, forms a coating which resists the action of the metallic
salts of the bath. If wood be used the vats may either be
square or round, but in any case they should be somewhat
wider at the toj) than at the bottom, otherwise the settling and
contraction of the moistened ore will leave a space along the
walls through which the bath may descend without percolating
the mass. Filteriuii vats need not be more than three feet hiy-h.
The filter may be made by laying on the bottom of the vat
three inches of small stones, broken cinder or coke, and cover-
ing this by a layer of coarse sand upon which the ore may be
laid to the depth of one or two feet according to its coarseness
or fineness. Instead of this arrangement a false bottom, con-
sisting of perforated planks or of narrow boards loosely laid
ogether, may be covered over with coarse s.acking upon which
the ore is sj)read. A hole in the side near the bottom of the
tank, into which is fitted an india-rubber tube provided with
a squeezer or pincli-cock, gives vent to the liquor after
its j)assage through the ore. The vats should be fitted with
close covers so as to exclude the air and retain the heat ; these
are providetl with a small hole through which enters a tube to
suj)ply the bath. It is well to spread the ore in the vats already
partially filled with heated bath, as when thus wet down it will
not cake but will permit the bath to percolate uniformly through
it. When the desired quantity of ore has been added a wooden
float should be secured beneath the opening in the cover so that
the bath, as it flows in, may fall thereon, otherwise it would
make a depression in the mass and thus the percolation would
be unequal. About two or three inches of bath should be kept
on the top of the ore, and it should be supplied as rapidly as it
escapes from the tube below. When the outflowing liquid is

to tl

le air

•ick. For
id coated
silicate of
of chlorid
\c metallic
either be
somewhat
ttling and
along the
LTcolatinfj
feet hiixh.
)f the vat
nd cover-
•e may be
oarseness
toni, con-
)sely laid
on which
of the
d with
after
ted with
; these
tube to
already
n it will
through
wooden
so that
t would
1 would
be kept
lly as it
quid is

found, bv testinnr with a bit of iron wire, to contain no more
copper than tlie liquid entering above, we know that tin' soluble
copper has been removed and it only remains to stoj) the suj)ply
of bath, allow the layer above to filter through and then dis-
place that which remains in the pores of the exhausted mass by
the addition of a little water. The extraction of the copper by
filtration may not be completed in less than three or four days,
the time of course depending on the richness of the ore and the
strength and the temperature of the bath.

The solution of the cop[)er is much accelerated by heating
the bath, which may be done by the injection of steam. It* the
liquid be heated to from 120° to 180° F. it will flow through the
ore in the leacliing tanks with a very little re(hiction of tempera-
ture, and the heat generated in the process of i)recipitation will, if
the tanks for this operation are well covered, maintain the batli
in these at a sufficient temperature to ensure a quick separation
of the co[)per, so that it is only the liquid in the store tanks that
will require heating by steam.

The bath after it is withdrawn from the precipitating tanks
generally contains a little copper. If, however, care be taken
to leave it there till the whole of the copper is separated, the
liquid will then be without action on metallic iron, and steam
coils may be used to heat it in the store tank, or in passiui: from
this to the leach vats it may be made to pass through a coil of
iron pipe heated by a stove.

Where kilns are used for roasting, the heating of the liquors,
as well as the eva[)oration of the excess of liquid derived from
the wash-waters, may be effected in Gay-Lussac towers, which
are small brick or stone chambers, tall and narrow, filled with
fraixments of coke or broken bricks, in which an ascending
current of the hot air and sulphurous va|)ors meets a descend-
ing current of the liquid. The hot gases from the kiln or from
a muffle may also be utilized by drawing them through a pipe
from four to six inches in diameter by means of a small steam
jet introduced at the bend in an injection pipe, which at that
point should be contracted to two inches, and may dip two feet

or more into the liquid in tlie store tank. Such an arranj:feinont
saves steam and serves to impregnate the bath witii sulj)hur()us
acid, which in its passage tlirou<^h the ore in tlie filtering vats
serves to attack the separated peroxyd of iron, converting it into
soluble protosalt.

The use of the sulphurous acid fumes, which thus serve to
supply the losses of protochlorid of iron, need not be resorted
to except in treating native carbonates or oxyds of copper, or
such ores as contain carbonates of lime or majjuesia or oxyd of
lead or of zinc, all of which cause a loss of the j)rot()cliIorid of
iron. In such cases the best mode of applying the suli)hurous
acid is by using stirring tanks and j)assing the gas over the sur-
face of the licjuid, which is agitated during the solution of the
copper. The gas should be as little diluted with air as possible.
If the roastinij!; kiln or mufHe furnace be connected with the
stirring tunk by an earthenware tube which enters eitlier the
cover or the side of the tank at a point opj)osite to that by
which a wooden tube (best connected with a flue) gives exit
to the unconsunied gas, a sufficiently rapid current of the gas
^vill be kejit up, and will be readily absorbed by the Ii(juid in
the tank.

It is seldom, however, that the process is thus complicated by
the necessity of using sulphurous acid gas, for unless the ob-
jectionable matters mentioned above are present in considerable
quantities in the ore, this, if a sulphuret of copper, will yield
by careful calcination, as already explained, enough sulphate of
copper to compensate for the loss of chlorid of iron wiiich these
would occasion.

By the introduction of steam for heating and of water for
washing the residue, the volume of the bath becomes slowly aug-
mented. To reduce this it is therefore necessary to resort to
evaj)oration, and for this purpose it has been found that in the
case of the three-heated reverberatory furnaces already de-
scribed, the u})per hearth may with advantage be used for this
purpose. A shallow basin six inches deep, of which the upper
hearth is the bottom, is built of brick lined with Roman cement

ii !

rrctnent
•liurous
nil vata
t it into

3rve to
esorted
jjKM', or
oxyd of
lorid of
)hurous
tlic sur-
i of the
)ossible.
•ith the
her the
that by
a«s exit
the gas

brick leachinjx tank

s, wi

d in

ited by
lie ob-
k'rable
yield
hate of
11 these

ater for
•ly aug-
esort to
in the

ly de-
fer this
e upper
cement

prepared as ah'eady described 1
silicate of soda and chlorid of calcium. Through the middle of
it passes a funnel or hopper connecting with the second hearth
and opening on the roof of the furnace for the purjiose of
charging the ore. The roof over the pan should be higher than
if it were over a calcining liearth, in order to give ample room
to use a scraper with which to remove the crystals of salt which
separate during evaporation, and two doors instead of three
should open at each end so as to give free access to the whole
surface of the pan. Care should be taken to keep the floor of
the pan free from accumulations of salt, else a crust will bo
formed which is difficult to remove. When this arranij-ement is
used, the li(piors from the store tank are run into the pan where
they are ex})osed to the escaping current of hot air and gases by
which they are rajiidly evajiorated, while at the same time they
absorb a considerable amount of sulphuric acid (which, together
with sulphurous acid, is formed in the slow roasting of ])yritou3
ores), and thus become strongly acid. An evaporator of this
kind, with a surface of 100 s(}uare feet, is easily and cheaply
constructed, is tight and durable, and will evaporate a layer of
four inches of liquid in twenty-four hours, by the waste heat.
If the three hearths are required for calcination, large shallow
tanks of the kind described may be constructed between the
furnaces and the stack, so that the whole current of the hot
gases shall pass over the surface of the liquid, thus dispensing
with the use of steam for heating the liquors and evaporating
the bath at the same time.

IV. Precipitating the Copper. The copper liquors, whether
taken from the stirring or settling tanks, or flowing from the
leaching-vats, are received in tanks of any convenient size,
where in contact with metallic iron the chlorids of copper are
decomposed, and the copper is precipitated in crystalline grains,
plates or crusts, the texture of which will vary according to the
strength of liquors. AVrought-iron precipitates the copper more
rapidly than cast iron, but where this latter is the cheaper it
should be used. If small scrap is employed it must be spread

]|8

on trays arranged in the vats. In the Appendix will be found
described a method of cleaning tinned iron scrap by -which the tin
may be saved and the iron fitted for precipitating cop])er. If
the residue after the extraction of the copper is a nearly pure
oxyd of iron, this may be reduced to a spongy metallic iron by
heating it for some hours at a red heat with pulverized coal in
a closed vessel. This spongy iron, which may also be easily
made from ordinary iron ores, precipitates copper very rapidly
from its solutions, and is used for that purpose in England, where
it is prepared in a reverberatory furnace. Details of the
pre])aration and use of the iron sponge will be found in the
Appendix.

From the precipitating vats the liquor which has been de-
prived of its copper, and in this process has been recharged with
protochlorid of iron, is drawn off after twenty-four hours or
more, and pumped up into the store tank, which should be at a
higher level than the leach vats. It is then ready for the
treatment of fresh j)orti()ns of ore. A working drawing of a
cheap ])unip nuide entirely of wood, is included in the plan
annexed.

V. Melting and Refining the Copper. The precipitating
tanks are emptied from time to time, the cement copper is
washed with water and if small iron scrap has been used, is
passed through a screen or sieve to separate any fragments of
the latter. It is then dried at a gentle heat, when it is ready
for refining. In the treatment of coj)per obtained in wet pro-
cesses it is customary to melt the cement with a portion of
matte or sulphuretted ore, and thus obtain a crude copper which
is refined by a second fusion. Experience on a large scale has,
however, shown that the purity of the cement obtained by the
l)resent j)rocess is such that but a single fusion is recpiired to
convert it into fine copper. The dried, or even the moist ce-
ment, is melted down in a furnace such as is used for refinin<'
blister coj)per, poled in the usual manner and then cast into
injiots. It is found advantatfeous to mix the cement with one
or two hundredths of coul dust, and if compressed into blocks

bo found
h the tin
•per. If
rly pure
: iron by
I coal in
)e easily
J rapidly
d, where
; of the
id in the

been de-
cked with
hours or
d be at a
' for the
infi of a
the plan

'ipi fating
;oppor is
used, is
in cuts of
is ready
[wet pro-
>rtion of
'r which
;ale has,
by the
luired to
iiuist ce-
I refining
:ast into
ivith one
lo blocks

it can be handled with greater advantage than if in a loose
powder.

VI. Arrangement of Plant. It is well, when it can be
done, to choose a hill-side as the site for works for carrying out
this process, so as to place the leaciiing vats below the level of
the calcining furnaces. Above these vats should be a water
tank and also a store tank for bath, from which it can be made
to flow into the vats j)laced in rows on the lower level. Below
these should be placed tlie precipitating tanks and still lower a
large tank into which the bath when de})rived of copper can be
allowed to flow, and from which by means of the wooden pump
it is to be pumped up into the store tank for redistribution, thus
establishing a continuous circulation. Wooden tubes securely
coupled together are the best conductors for the bath. A hor-
izontal line of such tubino; should run above the leachino; vats
and be connected with the store tank by a piece of india-rubber
tube or hose, which can be closed at will by a wooden squeezer.
From this line of wooden tubing the bath is to be conducted to
each leach tank by an india-rubber tube, the flow through
which is to be regulated by squeezers. From the leach tanks the
coj)per liquors should be conducted through similar india-rul)ber
tubes into a covered trough or launder, running the whole length
of the row of precipitating vats. Such a trough is better for
this purj)ose than a closed tube, for the reason that when the
bath is too cool or does not hold a sufficient amount of salt to
retain the whole of the dichlorid of copj)er in solution, a por-
tion of this may be dejjosited and fill up the tube, while the
launder can be watclied and this state of things guarded against.

In localities where a hill-side cannot l)e chosen for the site it
will be better to place both the leaching and the precipitating
vats on the same level with the calciners, for it is easier and
cheaper to jmmp the copper li(pu>rs into the precipitating tanks
than to elevate tlie ore. In the plan, however, for clearness of
iUustration the tanks are shown on successive levels in a build-
ing of three stories.

APPENDIX.

I. Assaying fou Copper and for Iron.

The following diroctions will enable any one, even without a knowlerlgc
of chemistry, to make the tests necessary for the successful working of this
process; namely, the determination of the quantity of copper in the ore,
the character of the roast, and the condition of the bath.

Copper. The most expeditious and convenient method for the determi-
nation of copper depends upon the property of cyanid of potassium to de-
colorize the deep-blue solution which is got by adding ammonia in excess
to the protosalts of copper. Begin by dissolving in clear water ordinary
comnjercial cyanid of potassium, in the proportion of about a half a
pound to a gallon. Next weigh out carefully five grains of pure metallic
copper, such as copper foil ; dissolve it in a little nitric acid ; add water to
the bulk of about four ounces, and then caustic ammonia (J'ninnr nmmoni(v)j
until a (U'cp-blue liiiuid is obtained. Next fill a graduated tube, known as
a burette and generally divided into cubic centimeters and fractions
thereof, with the solution of cyanid of potassium, and allow this to drop
from the burette into the copper solution, till the blue color disappears, first
giving place to a pinkish hue. Great care must be taken to add slowly
towards the end of the operation, so as to avoid the addition of an excess
of the cyanid. Now read olf on the graduated tube the quantity of the
solution of cyanid which has been reipiired to produce the decoloration of
the solution of five grains of copper. Suppose 2.J.0 cubic centimeters
fee.) have been consumed; then it is clear thsit one grain of cojjper cor-
responds to 5 c.c. of the cyanid solution.

To make an assay of i copper ore, reduce it to fine powder, and weigh
out carefully 25, fiO, or 100 grains, according as it is rich or iu)or in copper.
Add to this, in a small (lask or beaker-glass, common nitric acid, sufllcient
in (piantity to cover the ore. Apply heat to dissolve the copper, and if the
ore is a sulphin-et boil it, until, on adding a fresh portion of acid, no more
red vapors are given olf. Now add from four to eight ounces of water, and
pour in ammonia till the blue color is obtained, and the liquid shows by its

(20)

smell that a slight excess of ammonia has been added. By this means the
dissolved iron (which is almost always present in the ore) will be separated
as a bulky, reddish-hrov»n peroxyd, which rapidly settles, leaving the clear
blue liquid above. Add to the mixture at once the cyanid of potassium
from the burette, stirring the while, and allow the suspended oxyd of iron
to settle from time to time, so as to judge of the progress of the operation
from the color of the clear licpiid, which soon appears above the subsiding
precipitate of the brown peroxyd. The operation may require ten or
fifteen minutes. Wiien the color has faded out, as in the previous example,
note on the burette the quantity of the cyanid solution consumed. Sup-
pose this to be 48 e.c. Now as 5 c.c. are equal to a grain of copper, we have
the proportion 5:48:: 1 : 9.G; so that, if the quantity of ore was 100
grains, the ore contains 9.6 p. c. of copper, or, if 50 grains were employed}
19.2 p. c.

In testing the roasted sulphuret ore, there shouM be determined in each
sample three things : (I) the quantity of sulphate of copper, a portion of
which, as already explained, should always be present; (2) the amount of
oxyd of copper ; and (3) the quantity of unoxydized copper, existing ia
the form of sulphuret. As sulphate of copper is readily soluble in water,
it is only necessary for the first deteruiination to boil a weighed portion of
the ore with a little water, pour olf the clear soluti(jn, wash the residue
with cold water, add ammonia to the liquid, and proceed as butbre. Muri-
atic acid diluted with twenty-five times its bulk of water will, at a boiling
heat, readily dissob'c the oxyd of copper from the roasted ore, without at-
tacking the sulphuret. If then we boil the roast for two or three minutes
with a sulHcient (juantity of such dilute acid, allow the undissolved portion
to settle, and wash it thoroughly with several waters, the only copper left;
in the residuewill be that of the sulphuret which has escaped roasting
and which may be dissolved by boiling with nitric acid.

Suppose 100 grains of a ro;isted ore to be boiled with water: the clear
solution is poured of!", mixed with ammonia, and treated with the standard
solution of cyanid of potassium, of which 15 c.c. are consiuued. Then
5 : 15 :: 1 : 3; so that the sample holds 3.0 p. c. of copper Jis soluble sulphate.
The residue is now boiled for two or three ujinutes with two ounces or more
of dilute nuu'iatic acid, as already described, and the solution then oljtained
decanted and boiled for a minute with a ihw drops of nitric acid, to convert
any dichlorid into protochlorid of copper (this is necessary to get the full
color, because the ammoniacal solution of the dichlorid is colorless). Am-
monia is now added, and then the standard cyanid solution. K of this 32
c.c. are recpiired to decolorize, we have 5 : 32 :: 1 : G.4, or G.4 p. c. of cop-
per in the state of soluble oxyd. The insoluble residue is then boiled with
nitric acid till red fumes are no longer given oil', which may take as much

as five or ten minutes, water is added, and excess of ammonia, when it is
found that 2 c.c. of the standard solution of cyanid are required for decol-
orizing. We have thus 5 : 2 :: 1 : 0.4; so that there remained 0.4 p.c. of
copper as insoluble sulphuret. The result will then stand as follows : —

Copper as sulphate 3.0

*' " oxyd 6.4

" " BUlphid 0.4

9.8
In thus assaying a sample of ore, the percentage of copper in which is
known, we may dispense with directly determing the oxyd, and by
deducting from the total the amount of soluble sulphate and insoluble sul-
phuret, we find the quantity of oxyd by loss. In the regular working of the
process, in fact, it is only necessary to determine the sulphate and the
sulj)huret of the roast. The former is important, because as already ex-
plaim'd, it is necessary to have a certain portion of sulphate to make up for
the loss of protochlorid of iron in the bath, and the second because the
copper which remains unroasted in the form of insoluble sulphuret is not
extracted by the bath, and therefore is lost.

The above process of copper assaying, though the most simple and ex-
peditious, is not directly applicable to ores containing silver, zinc, cobalt,
nickel or manganese. The absence of the first may always be ensured by
adding a few drops of muriatic acid to the solution, by which silver is
thrown down as a white insoluble chlorid. When, as is not infrequently
the case, one or more of the other metals mentioned are present, the fol-
lowing modification of the process may be adopted. The solution from th^
roasted ore by water, mixed with a little muriatic acid, or the solution of
the oxyd of copper in this acid, is digested with metallic iron or zinc, by
which the whole of copper is thrown down in a spongy metallic form, while
the metals zinc, cobalt, nickel and manganese, which would vitiate the
operation of decoloration, remain dissolved. Iron wire or a slip of sheet
zinc may be used, and if heat bo applied the precipitation of the copper
will generally be finished in half an hour. It may be known to be complete
when a bit of clean bright' iron dipped for a minute in the liquid gets no
color of copper upon its surface. The metallic copper precipitated is care-
fully separated from the iron or zinc, washed with water, dissolved in a
little nitric acitl, mixed with an excess of ammonia and determined by the
use of a cyanid solution as before.

When a copper ore htus been dissolved by nitric acid it is necessary to get
rid of this acid before precipitating the copper. To this end the nitric
solution is to be evaporated nearly to dryness, then mixed with about as
much muriatic acid as had been used of nitric, and again evaporated nearly
to dryness. Water ia now added, and from the solution the metallic cop

por precipitated by iron or by zinc, and determined in the manner already
described. The amount of copper held in solution at any time by the bath,
is also readily determined by precipitation with iron or zinc. It is well to
add thereto a few drops of muriatic acid, and heat 'juickena the process. If
the quantity of copper precipitated is considerable, we may conveniently
wash it with water several times, then with a little alcohol, dry it at a j^cntle
heat and weigh it directly. Sm.aller quantities are liowever best <letermined
by dissolving the precipitated copper in nitric acid, adding ammonia and
the standard solution of cyanid.

Tlie cyanid solution is liable to slow decomposition, by which its strength
is diminished, so that from time to time the stock of the solution (which
should be kept in well-stoppered bottles and out of the direct liglit,), must
be standardized afresh by a solution of a known quantity of copper as
already explained. If the solution becomes turbid it may be filtered
through paper.

Iron. Tlie following simple and rapid mode of assaying for iron in the
bath, is based upon the fact that the deep crimson color of a solution of
permanganate of potash (mineral chameleon) is at once destroyed by a
protosalt of iron, such as protosulpliate or protochloritl. It is therefore
only necessary to add to the acid and dilute solution of a protosalt of iron,
wliich is colorless, a solution of the crimson permanganate of known
strength until the tint of the latter remains in the liquid. We begin by dis-
solving five grains of pure clean iron wire (like that used for piano strings) in
dilute sulphuric or muriatic acid, and then add half a pint or more of pure
cold water, and a few drops of sulpliuric acid. Now pour from a tube or
burette graduated like that used for tiie cyanid solution in copper testing, a
dilute solution of the permanganate, rapidly at first, and then more slowly
constantly stirring, till the iron solution acquires a faint crimson tint. The
quantity of chameleon solution employed, divided by the number of grains
of iron, will give the number of cubic centimeters wliich correspond tj a
grain of this metal. As the bath used in the Hunt and Douglas process
contains the iron in the state of protochlorid it is only necessary to take a
known volume of it and proceed as above to determine its content of iron.
If, for example, the permanganate solution is of such a strength that a
solution of five grains of iron reijuires 42 cubic centimeters of chameleon
solution to color it, equal to 8.4 c.c. for a grain of iron, we take two fluid
draclnns (a (juarter of a fluid ounce) of the bath, add thereto twenty drops of
sulphuric acid and half a pint of pure cold water and find that it recjuires
12.6 c.c. of chameleon to give it a crimson tint, then 8.4 : 12.6 :: 1.0 : L.O
80 that this quantity of the bath contains 1.5 grains of metallic iron, eiiual
to 6 grains to the fluid ounce. By testing in this manner from day to day
a measured (juantity of the bath, its loss or gain in iron can be very easily
determined.

The solution of the permanganate should be kept in a glass-stoppered
bottle and should be tested from time to time with a solution containing a
known weight of iron. For this purp(jse, instead of iron wire, which re-
quires some time to dissolve in acid, we may conveniently use the double
sulphate of protoxyd of iron and annnonia (ammonio-ferrous sulphate).
The green readily soluble crystals of this salt contain exactly one-seventh
their weight of iron, so that 35 grains of this salt, with 20 drops of sulphuric
acid dissolved in half a pint of water, corres2)ond to a solution of 5 grains of
metallic iron.

n. PUEPARATIOX AND UsE OF luON SpOXGE.

Tlie use of spongy metallic iron for precipitating copper in a metallic
condition from its solutions was proposed in 1851) by Mr. William
Gossage, of Widnes, England, to whom letters patent for improve-
ments in extracting copper were then granted in Great Britain, the
dates of the specification being March 7 and September 7 of that year
(patent No. 194). He claimed the extraction of copper from the residues
of calcined pyrites by the use of solutions of pcrsalts of iron. The copper
thus dissolved being thrown down by metallic iron, he obtained a protiisalt
of iron (protosulphate or protochlorid), which he converted by evaporation
and exposure to the air at a heat below redness into a mixture of insoluble
peroxyd and of a persalt soluble in water, which could be used to dissolve
another portion of copper. In the case of silver-bearing ores he combined
the persalt of iron with common salt, to dissolve the chlorid of silver formed.

In the same patent he claimed for the precipitation of the dissolved cop-
per linely divided or spongy iron prepared from calcined pyrites residues
containing a portion of copper, '' by mixing them with about one-fourth
their weight of coal, coke, or charcoal, in coarse powder, and introducing the
mixture into an oven or closed furnace kept at a red heat for about twelve
hours, or until the greater part of the oxyd of iron contained therein has
become converted into metallic iron. I then withdraw the product as quickly
as possible, and I receive it either in vessels containing water, or in vessels
from which the air can be excluded till the contents have become sufficiently
cooled." " I employ metallic iron obtained by the means herein described
for the precipitation of ci>pper from its solutions, in the same manner that
other metallic iron is employed for etfecting such precipitation." In 1862
Gustav liischof, Jr., obtained letters patent in Great Britain for the man-
ufacture of spongy iron for precipitating copper, the materials used and
the mode of working being essentially the same as those of the earlier
patent of Gossage, and in 1808 a United States patent for the same inven-
tion was granted Mr. Bischof and his assignee Mr. John S. Kidwell.

The manufacture of iron sponge for this purpose as carried on at New-
ca'itle, in England, is described in a paper read before the Newcastle
Chemical Society by Mr. (iibbs, and publislied in Engineering and in Van
Nostrand's Engineering Magazine for October, 1875, from whicli the fol-
lowing notes and extracts arc made :

The ore now treated for copper by ■vvet process in England is the residue
of the Spanish pyrites (of which nearly 400,000 tons are there annually
consumed), which has been calcined to extract its sulphur, and is then known
as burnt ore. It contains 3 or 4 p. c. of copper, as much sulplun-, a little
arsenic, lead, and zinc, and about 2 p. c. of silicious matter. By calcination
with about 1.3 p. c. of salt, in what is known as the Longmaid or Henderson
process and washing with water and with dilute uuiriatic acid, the copper,
sulphur and arsenic are removed, and the residue, known as purple ore, is
nearly pure peroxyd of iron. Both Gossage and Bischof proposed the use
of the burnt ore ibr the manufacture of the sponge on the ground that the
copper present would be obtained with the precipitate; but the arsenic
which remains in the burnt ore is such an objectionable impurity that, ac-
Qprding to ^Ir. Gibbs, only the purified or purple ore is now employed for
the manufacture of sponge for copper precipitation. He thus describes the
furnace and its working:

"This is essentially a reverberatory furnace, 30 fl. long, with a provision
for conveying the flame under the hearth, after it has passed over the
charge. The hearth of the furnace is 23 ft. loni"; and 8 ft. wide, and is
divided into three workinj^-beds bv bridy;es. Each bed has two working;-
doors on one side. The doors slide in grooves, and close air-tight. The
fire-box is 4 ft. by 3 ft., with bars 4 ft. 8 in. below the bridge, thus allow-
ing for a considerable depth of burning fuel. The fire-door slides in
grooves like the working-doors. The hearth is formed of tiles sustained
on brickwork partitions, forming flues through which the flame returns after
passing over the hearth. From tiiese flues the flame drops by a vertical
flue alonj'.side the nre-brid^e, to an underu;round flue conununicating with a
chimney. The entrance to the latter flue is provided with a fire-tile dam-
per, which is closed whenever the working or fire-doors of the furnace have
to be opened. A cast-iron pan, 20 ft. by 10 ft., is carried by short columns
and girders over the furnace-roof. In this pan the ore is dried and nuxed
■with coal, and from it is charged into the hearth through cast-iron pipes »
built into tlie furnace-arch. The furnace is elevated on brick jtillars, to
allow of iron cases running under it, and it is worked from a platform of
cast-iron plates. A vertical pipe, 6 in. diameter, passes through the hearth
of the furnace, inside each working-iloor; and through these pipes the re-
duced iron is discharged into iron c;ises placed beneath. These cases are
horizontally rectangular, and taper upwards on all sides. The cover is
3

fixed, and in its centre is a hole 6 in. diameter, with a flange upwards
which serves to connect the case with the discharging-pipe. The bottom of
the case is closed by a folding-door, hinged on one side, and secured by-
bolts and cutters on the other. The case is fitted with four wheels, clear
of the door, and is covered with a cast-iron plate, fitting loosely into the
opening on the upper side. It stands 4 ft. 8. in. high, and has a capacity
of 12 cubic feet."

" The furnace-hearth being at a bright red heat, each of the three work-
ing-beds is charged with 20 cwt. of dry purple ore, and 6 cwt. ground coal
from the cast-iron pan undt^r the roof. The fire and working doors are
closed, and the only air entering is that through tlie fire, in working which
care is taken to prevent the mass of burning fuel getting hollow. The
charge in the first bed from the fire-bridge is reduced in from nine to twelve
hours; in the second, in eighteen hours; and in the third, in about twenty-
four hours. Each charge is stirred over two or three times during the
period of reduction. Before opening any door the flue-damper is closed, to
prevent a current of air entering over the charge. On the complete reduc-
tion of the charge on any working-bed, two cases are run under the bottom
pipe, to which their mouths are luted by clay, and the charge is quickly
drawn into them, by rakes worked through the doors. The cases are then
closed with cast-iron plates. In about forty-eight hours the iron is cooled
suflicicntly to be discharged ; and this is simply done by rilsing the case by
a crane, and knocking out the cutters fastening the hinged door on the
bottom, when, from the tapering form of the case, the mass of reduced iron
falls out readily. The sponge is ground to powder under a pair of heavy
edge-stones, G fl. in diameter, and is passed through a sieve of fifty holes
per linear inch."

The heat beinfj a brijiht red, the reduction of the charges is sure to take
place on the hearths in the times specified; but it may be completely though
more slowly etfected at a very dull red. The material used is said to con-
tain 95 p. c. of peroxyd of iron ; and the product holds metallic iron 70.40,
peroxyd 8.1.5, protoxyd 2.40, sulphur 1.07, copper 0.24, with small portions
of lead and zinc, 7.60 of carl)on, and 9.80 of silicious residue; so that
about 90 p. c. of the iron of the charge is reduced to the metallic state.
If we sujjpose that four charges of 20 cwt. each of such ore are reduced in
twenty-four hours, the yield would be nearly 48 cwt. of metallic iron.

" In using spongy iron in precipitating copper, the liquids are agitated by
an air-blast while the iron is gradually added. By this means a very per-
fect mixture is obtained, and a copper precipitate can be readily produced,
containing not more than 1 p. c. of metallic iron. As compared with pre-
cipitation by scrap iron, the economy of space required and facility of ma-
nipulation are very great. On the side of spongy-iron precipitation are

cheapness of material and economy of application; while against it is the
presence with the precipit.ited copper of the unre(hiced iron oxides and
excess of carbon from the reduction. In employ in<j spongy iron, the
copper-extractor has the production of the precipitant in his own hands,
and avoids the troublesome handling of a material so cumbrous as scrap
iron."

It is comparatively rare that the residue from copper ores treated in the
wet way, will yield an oxyd of iron approaching in purity to that obtained
from Spanish pyrites; and it is clear that an oxyd containing a large pro-
portion of ejirthy matter is not fitted for the production of iron sponge for
copper precipitation, inasmuch as the whole of this would remain as an
impurity in the cement copper. The purer native oxyds of iron, however,
such as the magnetic and specular ores, may be advant<ageously employed
for the production of sponge, by grinding them to powder and heating
with an admixture of coal, as already described.

The production of pure spongy iron, on a commercial scale, has lately
been perfected by Mr. T. S. Blair, of Pittsburg, by means of an ingenious
gas-furnace in which the ore in small masses is heated to redness with
coarsely powdered charcoal. The iron thus prepared, being free from oxyd
and from the earthy matters of the fuel, is fitted for the manufacture of fine
steel; but this perfected process is not adapted to powdered ores and
residues like those from pyrites. As a matter of economy it remains to be
seen whether the somewhat impure iron sponge, which m.ay be very econon-
ically prepared from these, is to be preferred to the purer sponge, which is
made on a large scale and cheaply in Mr. Blair's furnaces.

III. Use of Tin Plate Scrap.

Tin plate, which is iron coated with tin, may be advantageously used for
precipitating copper, when, as in the Hunt and Douglas process, the so-
lutions contain protochlorid of copper, together with a soluble sulphate,
as sulphate of soda. A heated solution of this kind very readily removes
the tin from the iron, and causes its separation in white flakes of insoluble
hydrated peroxyd of tin, with the liberation of a portion of free hydrochloric
acid,^ leaving the iron in a state fit for the preci[)itation of the copper. In
practice the tin plate scrap may be ised directly in the vats in place of

1 In this reaction protochlorid of copper and metallic tin yield dlchlorid of copper and
perchlorid of tin, which last, In solution and in the presence of a sulphate, is broken up
into hydrated peroxyd of tin and trae hydrochloric acid. The final result of the reaction
may be thus represented :

4CuCl + Sn + Hj,0, = 2CujCl + SnO, + 2HCL

8or.')p iron, in which case the tin oxyd remains mixed with the cement
copper, and will be more or less completely washed away in the sul>so(juent
stages of the process. Otherwise the tin [date scrap may, by a simple ar-
rangement, be immersed ibr a few minutes in the hot copper solution till
the tin is taken oil", and may then be removed to the copper-j)recipitation
tanks. The separated tin oxyd may be collected by subsidence, freed from
any ailhering metallic copi)er by washinj^ with a portion of the hot solu-
tion containing protochlorid of copper, and when tJnis purified reduced to
the metallic slate or used for the manufacture of stanuate of soda. Ordi-
nary tin plate carries 3 or 4 p. c. of tin.

As tin plate scrap has, in most places, little or no commercial value, it
may often be advantageously employed for the precipitation of coj)per from
its solutions. The process above described was made the subject of letters
patent granted to Thomas Sterry Hunt, May 19, 1874, (No. 150,957) for
" An improvement in precipitating copper by means of tin scrap," the
claim beiiijr as follows:

1. " The use and anplicatiou of tin plate scrap or waste for precipita-
ting copper from its solutions, substantially as above described."

2. '' The recovery and utilization of the tin from the tin plate scrap by
means of its solution and subsecpient precipitation as o.xyd of tin in solu-
tions containing protochlorid of c(jpper and a sulphate, substantially as
alxtvc described."

IV. ClIKMISTUY OF TllH IICNT AND DofCil.AS PuOCESS.

The peculiarity of this method is the use of a solution oi protochlorid
of iron and clilorid of sodium to render soluble the oxy<lized com-
pounds of eoj)per. In the wet process now generally adopted in Great
Britain and mentioned on page 25, where the ores are calcined with com-
mon salt, this is in great part decomposed with the formation of sulphate of
soda and chlorids of copper, which are, in their turn, decomposed when in
solution, by contact with iron, with separation of metallic co])per and
proiluction of j)rotochlorid of iron.^

' Till' eimcentrutod liquid obtained bj- louobing tho oiu's in tliis process at AVidnes in Eng-
liHi'l .Uiivc, iU'ooi'(liii,i; to Cliiiiilt'l, lor a litrt! of sijcimHc ;j;ravity 1.21; siiliiliiito of soda 14.11
gnininics; olilorid ot 'i^odiuni li.aU; chloride* of coiipfi- and oilier nietal.s 12.75, containing
chloriue 6 CI, copiK'r 5.28, zinc 0.G8, load 0.067, iron 0.045, silver 0.004, besides a little gold
luul .small but undetermined quantities of ur.^enie, antimony and bismuth. Of the cojiper
O.oSO was in the state of diclilorid. The silver e.xtracted from this solution by t'laudct's
method, with iodine, contains about l.o p. c. of gold. Chemical yews, Vol. XXii, p. 184.

The llqui.l thns obtalnod, hol,,i„. „„ ,,„„,„„, ^f prot.n.hl.n,! of i,..,„
with a httio ohlori.l of sodi„n.. is f.u.n.l tn l,:,v. l„,t a feeble solvonf notion
"pon tl.o oxyd of ooppor an.l is ac-.-nr.iinjrly tl.rown awav, poll„ti„. tl.o
nvdrs, an.l tl.ns rrjvln^ rise to sorions (linic.,lti,.s in 7':n.rlan.l. Varions .t-
tempt, have heon tna.le to utilize the chlori.l of i.-on in these waste iionors
for chlondizin- copper. Gossa^^a. patented a plan whi.-h eonsisf.,! in ev.p-
oratms them to dryness and heatin,. the residue in contaet with air to hnv
redness, by whieh means there is obtained a mixtun. of insoluble peroxv.l
and s„l,d,le;„rc-7./o,vV/ ,>//.«„ (see pac;e 24). Henderson effeets ,be same
result by the aotion of air on the liquors at ordinary temperatures. He
also by decomposing the evaporated waste liquors at a stron.r re.l heit in
contact with silicious n.atters, jrots porchlori.l of iron in va,ror with some
hydrochloric acid and free chlorine, and .lissolves these in a soluti<,n of
protochlor.d of iron, thus prottinjr a solution o? perchlorid of iron, the sol-
vent action of which on oxy.l of copper is well known. (British patent of
JNIay, 1865, No. 1255, and United States patent, Dec, IHCG, No. 60 514 )

To dispense with these te.lious and costly processes and enable' li.n.or
conta,n.n.2;)ro/oc/,7onV/ of iron to be directly used for the solution of copper
was much to be desired. It was found that when a solution of proloehlorid
of iron IS brought in contact with either protoxvd or dinoxyd of copper
d.chlorul of copper is for.ned, which, being insoluble in wa'ter, soon .-oats
over the oxyd and arrests the chloridizinjr process. To overcome thi< .lidi-
culty, however, it was only necessary to add a hot and strong solution of
common salt in which (as in all other solutions of chlorids) the di.-hlori.l of
copper has a considerable .lei-ree of soh.bility. The reactions of the two
oxyds of copper with protochlorid of iron are unlike. Three equivalc.ts ot
the protoxyd, containing 95.25 of copper, when brought in contact with an
excess of solution of the protochlori.l under the con.litions just explained
react with two equivalents of it, containing 56.00 of iron, "and vield onJ
equivalent of the protochlorid of copper, which is rea.lilv soluble 'in water
and contains .31.75 of copper, and ;55.50 of chlorine, and^ one equivalent of
the insoluble <lichlorid. in which the same amount of chlorine is unite<l wiih
twice as much, or 63.50 of copper. When the copper is in the state of the
dinoxyd, only one-half as much protochlorid of iron is consumed, and there
IS formed tor the same amount of dichlorid as before, one equivalent, or
31./5 of metallic copper. This woul.l remain undissolved if the dinoxyd
a one were treated, but metallic copper in presenc,. of an excess of prot'o-
chh.rid of copper is at once converte.l into the .lichlorid, so that if one-half
of the oxydized copper in a mixture treate.l with an excess of profoHiIorid
of iron, IS protoxyd and one-half dinoxyd, the whole of the copper passes

into the state of dichlorid.* For this reason it is necessary, in submitting
(linoxyd ores to this process, either to mix them with a snflieient amount of
ores contaiiiin}^ protoxyd, or to eah'ine them slij^litly in the air, so as to con-
vert one-half or more of tlie (linoxyd into j)rotoxyd of copper.

In the reaction between the oxyd of copper and protochlorid of iron, tho
iron of the latter separates from the solution as a reddish brown insoluble
precipitate of hydrous peroxyd, which carries with it a small jjortion of
chlorine in the form of an oxychlorid of iron, duo to secondary reactions
and in part to the action of the air upon the solution of protochlorid of
iron. The amount of chlorine thus removed, and consecpiently lost to the
bath, was found, in carefully conducted experiments, to vary from .') to 10
p. c. of that orij^inally imited with the iron, that is to say for 100 parts of
protochlorid of iron consiuned in chloridizin;^ copper, the re<»enerated bath
will contain from 90 to 9!) parts. This loss of chlorine must in all cases be
supplied if the strength of the bath is to be kept up, an end which is readily
obtained in one or two ways. When suljihuretted ores are oxydized, there
is always formed a ))ortion of sulphate of copper, which, with careful roasting
(page 10), may e(|ual one-fourth or even one-half of the copper present.
This sulphate when decomposetl by metallic iron gives protosulphate of
iron, which, by its reaction with salt yields, as we have seen in the prepar-
ation of the bath, sulpliate of soda and protochlorid of iron, Avhich in
ordinary cases more than suffices to supply any loss of chlorine. If, as
sometimes happens, there is found too large a portion of these compounds,
this niny be corrected l)y Hiding to the ha.i\\, previoushj freed from copper,
a small (piantity of slaked lime, by which means the excess of sulphate and
of iron are precipitated in the form of sulphate of lime and proto.xyd of
iron, from which the clear liquid may be drawn or filtered off.

The bath made, as already described, with 280 pounds of copperas (equal
to ,5t) pounds of iron) and .S20 pounds of salt in 2000 pounds of water, has
a specific gravity of about 22° Beaunie or 1.150 at ordinary temperatures,
water being 1.000. A cubic foot of it weighs 1150 ounces avoirdupois and
contains 3.52 lbs. of protochlorid of iron (besides an equal quantity of
sulphate of soda), and about 5^ lbs. of salt. This amount of protochlorid
contains 1.54 lbs., or 10,780 grains of metallic iron, and as a cubic foot is

' The reaction!* between protochlorid of iron and the oxyds of copper are thus expressed
in chemical symbols, using, as has been done in the note on page 27, the older notation, iu
wliicli Cu = 31.75, l-e = 28, CI = 36.5, and = 8.
For the protoxyd of copper.

3CujOj + 4FeCl = 2Fo.fi^ + 2CUjCl + 2CuCl.
For the diuoxyd of copper,

SCUjO + 2FoCl = FCjO., + 2CUjCl + 2Cu.
In the reaction between the protochlorid of copper and the metallic copper,

2CuCl + 2Cu = 2CujCl.

equal to almost ex.actly 1000 fluid-ounces, oach fluid-ouiipo holds in solution
10.78 grains of iron as protochlorid. By the method of assay described
above, the amoimt of iron held in solution by a fluid-ounce of the li(|uid
is very easily determined, and in this way the elliciency of a bath is most
conveniently desigrated. Solutions containing 5.0 grains, an<l even 3.0
grains of dissolved iron to the fluid-ounce, may bo used, but the strongest
are most eflicient.

The protochlorid of iron serves to chloridize the oxyd of copper in the
ore. A cubic foot of bath containing 10,780 grains of dissolved iron will
chloridize 18,287 grains (or 2.61 lbs.). of copper in the state of protoxyd,
converting one-third of it into protochlorid, and two-thirds of it into di-
chlorid of copper, of which latter compound (consisting of copper 63.."),
chlorine 35.5) there will be formed 29,057 grains, or about 4.15 lbs. The
dichlorid is insoluble in water though readily soluble in strong brine, espe-
cially if this be heated; hence the necessity of a large excess of salt in the
bath. A cubic foot of saturated brine at a temperature of 194° F. will
dissolve about 10.0 llis., and at 104° F. about 5.0 Ib.s. • he dichlorid,
•while a cubic foot of brine holding 15 p. c. of salt, will dis>- ;it 194° F.
6.25 lbs., at 184° F. 3.75 lbs., and at 57° F. 2.18 lbs. of dichl>r 1 of copper,
and the same amount of brine holding only 5 p. c. of salt wm dissolve at
198° 1.65 lbs., and at 104° 0.70 lbs. of dichlorid. The bath above described,
with 5^ pounds of salt to the cubic foot, contains not quite 8 p. c. of salt.
It will thus be understood why in some cases it may become necessary to
increase the amount of salt in the bath in order to aujiment its solvent
power for the dichlorid. Both by cooling and by dilution with water the
dichlorid separates in the form of a white heavy crystalline powder, which
is readily converted by simple contact with metallic iron into pure crys-
talline copper.

In treating copper ores which contain no sulphur and consequently form
no soluble sulphate in roasting, the loss of the bath in chlorine may be sup-
plied by adding, from time to time, small portions of protosulphate of iron,
or still better by passing over or through the liquid in the stirring or leach-
ing vats, as already described (p<age 16), a current of sulphurous jicid gas.
This, being absorbed, converts the separated hydrous pcroxyd of iron
into a mixture of protosulphate and protosulpliite of iron, at the same time
liberating the combined chlorine of the oxychlorid in the form of soluble
protochlorid.

As the protoxyd of copper is a comparatively feeble base, the solutions of
the protochlorid are readily decomposed by the oxyds of zinc and lead,
which are often present in roasted ores. These cause the separation from
the solutions of a green insoluble oxychlorid composed of oxyd and proto-
chlorid of copper, chlorid of zinc or of lead being formed at the same time.

In llic jjroscnce of an excess of protochlorid of iron this oxydilorid of
coppiT is iiinnodiately dissolved, as oxyd of copper woidd be, l)iit in tlie re-
action a certain anuMint of clilorine is consumed in forinin;; tlie cldovids of
zinc and l(fail. An cx(;ess lA' oxyd of copper also unites with jtrotochlorid
of c^pjjcr U) form this oxychloriil, so that in leachin<^ ores char^eil with oxyd,
the protochlorid of copper formed is at first retained in a form insoluble in
water ami in brine, but as it is completely dissolved in an excess of proto-
(ddorid of iron, this reaction gives rise to no dillicnlty in working.

Jn like manner carbonate of lime, though without action on solutions of
protochlorid of iron below 212° F. readily decomposes protochlorid of
copper at 140° F., with separation of a similar oxychlorid, which recpiires
protochlorid of iron to redissolve it. In this way the presence of carbonate
of lime in copptT ore imlirccthi causes a loss of jtrotochlorid of iron, wdiich
must be supplied in one of the ways already set forth. The action of car-
bonate of magnesia is similar to that of the carbonate of lime. Neitlier
these substances nor the oxyds of lead or zinc separate the copjier from the
dichlorid.

In the preci|)itation of the copper by metallic iron 28 parts of this metal
unite with .'15. .5 parts of chlorine, and in so doing separate fi-om a solution
of th(^ protochlorid 31.7.J of copper, and from (he dichlorid twice that
amount. Hence to obtain 100 parts of copper from the first recpiires 88.2
parts, and from the second 4 1.1 parts of iron, while! from sohuious in which
one-half (he copper exists as protochlorid and one-half as dichlorid the
amount of iron reepiired will be the mean of these two, or about GG parts
for 100 of copper.

In the roasting of sulphuretted copper ores the greater part of the coj)per
(apart from the sulphate) is obtained as protoxyd, besitles a variable amount
of dinoxyd, sometimes, according to PlaKuer. as much as 20 or 30 p. c. of
the copper.^ Such a i lixture when treated in the bath gives rise, of course,
to a corresponding); hirge amount of dichlorid, which is, however, gen-
erally nearly counterbalanced by (lie protochlorid resulting from the
reaction between the sulphate of copper and the salt of the bath, so that
the proportion of iron recpiired to separate 100 j>arts of co|)per from the
sohition of such roasted ores varies fiom GO to 70 parts. Ilenci' the present
process presents, in this respect, a great economy over the ordinary wet
methods in which the ])recipitati«m of 100 parts of cojjper reipiires 100 and
often 120 or more parts of metallic iron.

' 111 sonit" cases we Imvc fdiiiul in siu'ti rousted oros a portion of sn1|ili:itc of illTioxyd of
copper. Tliis reiuuins wlieii tlie ordinary siilpliiite (of jirotoxyd) lias Inien removed by
water, and may be dissolvd from the roidiie by a hot solntion of oominon salt, by which
this insoluble sulphate is converted into dichlorid. Some cup|)er ores of 15 or '-(• j). c. have
yielile;! as much us one jier cent, of cojiper in this form, which is of course roudily soluble
in the protochlorid uf iron bath.

A solution of protoeLlorld of copper wl.cn mixo.l witl, salt not onlv has
the power of ehlori.lizin. and dissolvin,, metalHe eopper, as already cle-
scribcl, l.„t readily take, up the copper from sulphuretted ores, such as the
v.treous and variegated species, and from copper matte or re<r„lus, with
separation of sulphur and fonnation of dichlorid. This reactUm mav in
some cases, be taken advantage of by causing a hot solution nearlv satu-
rated with salt and holding protochlorid of copper, to f.Iter through a laver
of suchore or reo-ulusin coarse powder. The metal is rapidlv taken ur^, and
solu K.ns obtained in which the whole of the copper is present as dichlorid.
In t^h.s way an additional amount of copper is .Ussolved and mav be separated
.n the metalhc state with very little cost. The fi6 parts of iron required
toprec.p.tate 100 of copper from the ordinary solutions of n.ixed proto-
chlond and dichlorid will separate from solutions of pure <lichlori!l 150 •
parts of coi)per.

V. Patent Spkcificatiox.

Letters patent for the process above described were -ranted to T Sterrv
Hunt and James Dou.las, Jr., in 18Gf). in the United States. C.reat Britain
and Canada, the date of the United States patent beln. F.-b. 9, 1809.
The „a ,,re of the process and the mode of apj,lyin^ it havin, been fully
so forth ,n the precedino: pa-es, it will be suflicient to .ive the following
extracts from the specification: "

"We do not claim the use of any particular form of furnace, nor of any
spccuvl arrangement for calcining, lixiviating or precipitating, reserving to
ourselves the choice of the best forms of apparatus for these purposes,
either do we claun the use of protosalts of iron otherwise than in solution
no. the use of perchlorul or other persalts of iron, nor vet the use of sul!
!:;:rs:::tr ^"' ''''''' '- <-onnection with p^tosaUs of iron, as
" What we claim as our invention is:

"I. The use and application ofa solution of neutral protochlorid of iron
or of mixtures containing it, for the purpose of converting the oxvd or
s.iboxyd of copper, or their compounds, into chlorids of copper

" n. riie u.e of sulphurous acid for the purpose of decomposing, the
o.xychlorid of n-on forme.l in the preceding re-action. °

" III The use of a process tor the purpose of extracting copper from its
naturally or artihcially oxydlzed compounds by the aid of the first, or the
let fril.'' ''" '^ ''" '^'''' ''''"°"'' ^"l^^tautially in the manner already

VI. What Ouks of Copper may be Treated by this Process.

The forms in which copper occurs in nature may be conveniently grouped
in tliree classes to each, of which, under certain conditions, tlie Hunt and
Douglas process may be advantageously applied.

In the first class may be included the various sulphuretted ores, such as
copper pyrites (often mixed with iron pyrites) and the variegated and
vitreous sjlphurets, all of which are readily oxydized by calcination. In
addition to these are the fahl-ores, which contain besides sulphur, arsenic
and antimony. These objectionable elements by calcination are cither ex-
pelled or rendered insoluble. All the above named ores yield their copper
after oxydation to the Hunt and Douglas bath. The (juestion of the com-
parative fitness of this method for rich and poor ores has already been
discussed on page 6.

In the second class are included the oxydized compounds of copper, such
as the red and black oxyds, the green and blue carbonates, salts like the
oxyohlorid and also silicates of copper like chrysocolla. All of these are
readily attacked l)y the balh without previous calcination; but in the case
of the red or dinoxyd, as already explained above, it should either be
mixed with jirotoxyd ores or in part converted into protoxyd by a slight
calcination in order to render tfie copper wholly soluble. The carbonates
of copper, which are readily dissolved in the bath, give off their carbonic
acid so as to cause frothing, to prevent which it may be well to give tliom a
slight calcination or roasting. Heating them to low redness in a kiln or
furnace for a few minutes will be sullicient to convert the carbonates into
protoxyd.

The common silicate of copper called chrysocolla, readily gives up its
copjier to the bath of protochlorid of iron and salt, so that its treatment,
whether alone or mixed with other ores, presents no diflictilty. A peculiar
ore which is now treated successfully by this pr cess at I'lin'iiixvillc, Penn.,
is a hydrated silicate of oxyd of copper with magnesia, alumina and per-
oxyd of iron, containing when pure about 13 p. c. of copp<>r. This mineral,
which may be described as a copper-chlorite, is readily and completely
decomposed by acids but is not attacked by the bath of protochlorid of iron.
To extract the copper it is treated as follows: The crude ore, which is
mixed with clay and sand and carries from 3.0 to (1.0 p. c. of copper, is
heated to low redness for some hours in large vertical muflles each holding
15,000 lbs., having been previously uiixed with one-tenth its weight of coal
in coarse powder, by which the combined oxyd of copper is reduced to the
metallic state. This, on withdrawing the heated charge, is at once oxydized
the air, vielding a mixture of protoxyd and dinoxv<l of copper, which

are readily and comjjletely removed by the subsequent operation of leach-
ing with the Hunt and Douglas bath. The copper is chielly dissolved in

the form of diclilorid, as is shown by the fact that not more than 50 parts of
metallic iron are re»jiiired to precipitate 100 parts of copper from the
solution.

The third class includes the deposits of native or metallic copper, which
in almost all instances are most advantageously treated by mechanical
means. In those rare cases, in which the copper is too finely divided to
be thus profitably extracted, it will be found that by careful calcination
at a low red heat it may be oxydized so as to become soluble in the pro-
tochlorid of iron bath. In this, as in all other cases of non-sulphuretteil
ores, it is as already explained (page 16) indispensable to supply the loss
of chlorine by the use of sulphurous acid funics, or by the a<ldition from
time to time of a protosalt of iron.

The presence of carbonate of lime or carbonate of magnesia in any ore
is objectionable, since as already explained (page 32), it decomposes the
protochlorid of copper and thus indirectly precipitates the iron from the
bath. The action of oxyds of lead and zinc, which come from the roasting
of blende and galena when these are present in the ore, proiluces a similar
elfect. When not present in too large cpiantitics, tlie effect of all these
substances may be corrected by careful roasting, which forms a large pro-
portion of sulphates, or by the use of sulphurous fumes, but ores containing
much carbonate of lime or carbonate of magnesia are not adapted to treat-
ment by this or any other wet process.

YII. Practical Woukixg of thk Process.

The Hunt and Douglas process, after some experimental trials, was first
worked c:()ntinuously for a year in 1872-73 at the Davidson Mine in North
Carolina, under the direction of the ^lessrs. Clayton. The ore, a pyritous
copper in a slaty gangue, was dressed up to five or six per cent., crushed to
pass through a sieve of forty meshes to the linear inch, roasted in three-
hearth revcrl)eratory furnaces so as to contain about one-fourth its copper
as sulphate, and treated in stirring vats in charges of 3o00 lbs. The loss
of copper in the residue w;us found to be from 0.3 to 0.5 p. c, and the bath
maintained its strength in chlorid of iron without the use of copperas or
sulphurous acid. The amount of inm consumed was ecjual to 70 p. c,
and the salt, to supply unavoidable losses, to 25 p. e. of the copper pro-
duced. These details are from a letter from the manager of the works,
Mr. James E. Clayton, published in the V.ngiiifering atui Mlniiuj Journal
for.Iidy, 1H73, Injm which it apjjcars that the entire cost of producing cement
copper from the dressed ore of 5-^ p. c. was estimated to be three aud two-
thirds cents a pound.

This niino. was suhsequently .abandnnod, and tlio same proprietors in 1874
erected works with six calcining furnaces fur the treatment of twelve tons
of pyritous ore daily by this process at the Ore Knob mine in Ashe Comity,
North Carolina. Up to the first of .lannary, 1875, over 200 tons of copi)er
had tlu-re been made by this process. In the report bearing that date of
the directors of the Ore Knob Co., James E. Tyson of Baltimore, ))resident,
it is said, " From the data furnished by the superintendent in his Report
from the mine, and a careful estimate made here, Ave find the cost of making
copper, mining, and all expenses included, to be loss than eight cents a
pound."

These works were soon after enlarged to nearly three times their former
capacity, but in sinking below the water-line in the mine the ore. hitherto
free from lime, was found to contain 30 p. c. or more of carbimate of lime
with some magnesia. The direct treatirient of such an ore by any moist
process was impracticable, and the reduction works were accordingly sus-
pended pending the erection of dressing-works in which it is proposed to
concentrate the ore by crushing and washing, removing thereby the car-
bonate of lime of the gangue. The concentrating machinery, as we are
inlbrnied by the managing director of th(^ Ore Knob Copper Co., Mr.
James E. Clayton, will be in operation in June, 187G, when it is proposed
to recommence at once the treatment of the purified ores Ijy the Hunt and
Donglas process.

lleiluction-works arc now in successful operation at PhaMiixville, Penn-
sylvania, where copper ores of two kinds are treated by the limit and
Douglas process, the first of wliich is a magnetic iron ore from Herks Co.,
Penn., containing about .3 p. c. of cop{)er, chielly as cojiper pyrites, mixed,
however, with a little carbonate and silicate of copper. This ore, of
which iOjOOO lbs. are treated daily, is crushed so as to pass through a sieve
of seven meshes to the linear inch, roasted as already explained on page
11, and subsccpiently treated by leaching. The residue, which contains
about 0.5 p. c. of copper, is a rich iron ore which is use<l lor lining puil-
dling furnaces. The second ore is the jieculiar hydrated silicate described
on page 34, of which 15,000 lbs. are treated daily. The leached residues
of tliis do not retain over 0.3 p. c. of copper.

The works of the Stewart Reduction Co., at Georgetown, Colorado, in
which this jirncess is ap])lied to mixtures of silver a'ul copper ores, will lie
again referreil to.

VIII. Treatment of Silver and Gold Ores.

The use of soluble compounds of copper as an agent in treating silver
ores and rendering them fit for amalgamation, has long been known, and is
the basis of the ^lexican j)atio jjnx-ess and its modifications, as well as of
the AVashoe process now largely emi)loyed in the west. The theory of the
action of the copper salts in the first of these methods, where tlie materials
are exposed for a long time to the action of the air, is still somewhat ob-
scure. In the Washoe method sulphate of copper and connnon salt are
added together to the ground ore mixed with Avater, and from tlii'se by the
reactions Avliich take ])lace in the i)ans, dichloriil of copper is soon formed.
This substance dissolved in brine is used directlv with advantage in the
treatment of silver ores by Janin and by Kriincke. From the results of
various experimenters, it is clear that solutions, both of ])rotochlorid and
dichlorid of copper, mixed with couunon salt, when at an elevated temper-
ature, effect a complete chloriuaticju of sulphuretted and arsenical silver
ores, or at least render them susce2)tible of ready and complete amalga-
mation.

The use of the chlorids of copper as hitherto applied, presents, however,
several dilliculties : 1st. The sulphate of copper from which they are gen-
erally prepared is costly, and in some places diflicult to jirocure; 2d. Tro-
tochlorid of copper i.-^ readily decomposed and separated from hot solutions
as an insoluble oxychlorid l)y the carbonate of lime ofteu found with the
ores; 3d. Solutions of dichlorid of copper in brine very readily absorb
oxygen from the air, forming, besides j)rotochlorid of copjjer, also an insolu-
ble oxvchlorid. These oxvchlorids are without action on silver ores,
though they attack the mercury when amalgamation is attempted simul-
taneously with the treatment by copi)er salts, forming an insoluble chlorid
of this metal, and thereby causing a considerable loss.

To meet these objections there is needed a cheap and ready method of
preparing the chlorids of copper, and a simple means of preventing their
precipitation in inert or noxious forms by the action of the air or carbonate
of lime. It will be apparent from the preceding account of tjie chendstry
of the Hunt and Douglas copper proci-ss, I'l -n the use of a heated solu-
tion of i)rotochlorid of" iron and salt, aidcil by si.lphnrous acid, for the solu-
tion of the oxydized compounds of copper, meets the conditions of the
problem in the following manner:

1st. The Hunt and Douglas bath gives readily <ind cheaply strong
solutions of the mixed protochlorid and ilichlorid of copper wherever car-
bonates, oxyds, or calcineil sulphurettetl ore of this metal can be had.

2cl. It dissolves the oxychlorids of copper, by whatever means produced,
chanjiing tlicm into a mixture of protochlorid and dichlorid of coj)per, and
thus prevents any deterioration of tlie copper solution by the action of the
air or of carbonate of lime.

The Hunt and Douglas bath may be advantageously applied: ^
I. To effect more cheaply and more completely the chlorination and
the amalgamation of such silver ores as are now treated in the raw state
with chemicals, as they are called, — that is to say, sulphate or chlorid of
copper with common salt.

II. To chlorinate such silver ores as have been calcined without the
addition of salt.

III. To complete the chlorination of silver ores which have been par-
tially chlorinated by calcining with salt, thus securing a much more com-
plete extraction of the silver than has liitherto been attained.

Jn all of these cases it will be understood that some oxydized form of
copper, such as carbonate, native oxyd or calcined sulphuretted ore, is to
be added, unless it is already present in the silver ore to be treated. It may
be added even in large quantities with advantage, and from the solutions
charged with copper a portion, or the whole of this metal may be precipi-
tated from time to time by metallic iron as cement copper.

In localities where salts of iron are not readily obtained, and where sul-
phur ores are abundant. It will be found that by passing sulphurous acid
gas into or over a solution of salt holding pulverized oxyd or carbonate of
copper in suspension, a solution of dichlorid of copper will be readily
formed, and this reaction may be rendered available for the treatment
of silver ores. By precipitating the copper solution thus obtained with
metallic iron, protochlorid of iron is at once readily and cheaply obtained.

Silver ores chlorinated by the Hunt and Douglas bath, may be subse-
quently treated, either by dissolving the silver from the washed residues by
a solution of hyposulphite or of chlorid of sodium, or by amalgamation.
The use of mercury is to be preferred for ores holding, besides silver, a
portion of gold. Such ores should be treated with the bath in the raw state,
or after i-imple calcination, roasting with salt being for them objectionable.

United States letters patent (No. 151,763) for the use of the Hunt and
Douglas bath of protochlorid of iron and common salt, conjointly with
sulphurous acid, for the treatment of silver ores, or silver and gold ores,
mixed with oxydized ores of copper, were granted June 9, 1874, to James
Douglas, Jr., Thomas Sterry Hunt and James Oscar Stewart. This pro-
cess has now l)een most successfully applied for more than a year on a large
scale in the working of silver ores by Mr. Stewart, who will publish in the

1 Later obsorvations lihow that this process may be advantageously applied to the treat-
ment of the tellurids of silver and gold.

course of tlie summer of 1876 a detailed description of the method as
adapted by hun to various kinds of silver ores. Copies of tins (an,l also
of the present pamphlet) may be had by addressing J. Oscar Stewart,
Georgetown, Colorado.

For further information concerning the Hunt and Douglas process as
applied to copper extraction, address

JAMES DOUGLAS, Jr.,

Phcenixville, Penn.,
or

May, 1876.

T. STERRY HUNT,

Boston, Mass.