NRLF

EXCHANGE

A Study of the Hoesch Reaction

WILSON DAVIS LANGLEY,
B. S. Wesleyan University, 1918
M. S. Wesleyan University, 1919

THESIS

bmitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy in Chemistry in the Graduate School
of the University of/Illinois, 1922

Reprinted from the Journal of the American Chemical Society
Vol. XLVI, No. 10. October, 1922

A Study of the Hoesch Reaction

WILSON DAVIS LANGLEY,
B. S. Wesley an University, 1918
M. S. Wesleyan University, 1919

THESIS

* . ;

Submitted in partial fulfillment of the requirements for the degree of

Doctor of Philosophy in Chemistry in the Graduate School

of the University of Illinois, 1922

Reprinted from the Journal of the American Chemical Society
Vol. XLVI, No. 10. October, 1922

XC MANGA

[Reprinted from the Journal of the American Chemical Society,
Vol. XLVI, No. 10. October, 1922.]

[CONTRIBUTION FROM THE CHEMICAL LABORATORY OP THE UNIVERSITY OF ILLINOIS]

CONDENSATION OF CERTAIN NITRILES AND VARIOUS POLY-
HYDROXYPHENOLS TO FORM PHENOLIC ACIDS

BY WILSON D. LANGLEY WITH ROGER ADAMS'

Received July 10, 1922

From the time Hoesch 2 first discovered that nitriles reacted with certain
phenols in the presence of anhydrous zinc chloride and dry hydrogen chlor-
ide to give ketones according to the following equations

RCN

ZnCl 2 -fHCl 1^ / JC(=NH-HC1)R

this reaction has been applied extensively in organic chemistry. It has
proved of particular value in the synthesis of various natural products,
in the preparation of certain synthetic drugs, in the formation of coumar-
anones, and various other types of ketones and aldehydes.*

Another application of this reaction, namely, the condensation of 0-
chloro-propionitrile with resorcinol to prepare 7-hydroxychromanone-4
was tried as follows:

1 This communication is an abstract of a thesis submitted by W. D. Langley in
partial fulfilment of the requirements for the degree of Doctor of Philosophy in Chemistry
at the University of Illinois.

2 Hoesch, Ber., 48, 1122 (1915).

8 (a) Hoesch, ibid., 50, 462 (1917). (b) Sonn, ibid., 50, 1262 (1917); (c) 51, 821.
1829 (1918); (d) 52, 923 (1919). (e) Stephen, J. Chem. Soc., 117, 309, (f) 1529 (1920),
(g) Karrer, Helvetica Chim. Ada, 2, 89, 462, 486 (1919); 3, 261, 392, 541 (1920); 4,
203, 707 (1921). (h) Fischer, Ber., 50, 611, 693 (1917). (i) Bauer, Arch. Pharm., 259,
53 (1921).

CONDENSATION OF NITRIDES AND PHENOLS
HOr" "^OH C1CH 2 CH 2 CN

ZnCl 2 + HC1 ^ ^C( = NH-HC1) H 2 O
CH 2 CH 2 C1

HO/\/\

L.H 2

k ;COCH 2 CH 2 C1

from which by treatment with butyl nitrite and acid the compound 3,7-
dihydroxychromone-4 would be produced. This compound with alcoholic
sodium hydroxide should give fisetol.

/\/\
H0 Y C

H HO/NOR

rj, COH >

The initial reaction, however, did not take place as expected and the
results of the abnormal reaction are reported in this communication.

The condensation of j3-chloro-propionitrile with resorcinol takes place
smoothly. The final product is not a ketone but an acid, /3-(2,4-dihydroxy-
phenyl)propionic acid (III). The formation of analogous compounds
takes place when resorcinol monomethyl ether, orcinol or phloroglucinol
is used; in the cases of orcinol or phloroglucinol unstable acids are pro-
duced which immediately decompose into water and the corresponding
lactones. In no instance during these experiments, however, was the
formation of any ketone observed. This is a surprising type of reaction
in view of the fact that chloro-acetonitrile condenses normally with re-
sorcino! 3b as follows.

H C1CH 2 CN HOf V>H

ZnCl 2 + HC1 \y C( = NH-HC1)CH*C1

HO/NOH HO/V- 0\

> >CH 2

L/COCH,C1 \/~ C

There have already been described in the literature other types of ni-
triles which condense abnormally with resorcinol in the presence of hydro-
gen chloride and zinc chloride. E. Fischer discovered that cinnamo-

2322

WESSON D. BANGLE Y WITH ROGER ADAMS

nitrile and certain substituted cinnamonitriles condense with phloro-
glucinol, 3h not as would be expected, but instead to give the corresponding
substituted chromanones according to the following equation,

C 6 H 6 CH = CHCN

= NH-HCl
H 2

H 2

CH 2

a reaction which is very similar to that observed in this research. Fischer
assumed as a mechanism for this reaction that the phloroglucinol first
adds to the unsaturated nitrile and that the hydrogen chloride then con-
verts the nitrile produced into the imide lac tone which in turn is readily
hydrolyzed to the lactone. This mechanism seems unlikely in view of the
observations made with /3-chloro-propionitrile. It is well known that un-
saturated nitriles add halogen acid 4 with great ease so that the primary
reaction with cinnamonitrile is undoubtedly the addition of hydrogen
chloride. There will thus be produced a 0-chloro-propionitrile derivative
and the subsequent mechanism will undoubtedly be similar to that observed
with' /3-chloro-propionitrile itself.

When the /3-chloro-propionitrile is condensed with resorcinol in an-
hydrous ether in the presence 'of hydrogen chloride, a white solid slowly
separates. This is difficult to purify, contains both chlorine and nitrogen,
and probably has Structure I.

NH-HC1

H 2 O

H(y\OH

H 2 CH 2 COH 2

III

When I is treated with cold water, it dissolves completely, and within
about a minute an oil separates which on cooling solidifies. This is the
lactone, II, and is formed in practically quantitative yields. The structure
of this lactone is certain, as shown by a direct synthesis from umbelli-
ferone, the corresponding unsaturated lactone. Upon heating either the
lactone or the imide hydrochloride with water, the corresponding /3-(2,4-
dihydroxyphenyl)propionic acid (III) is obtained.

There are two possible mechanisms by which these substances might
be formed.

4 Moureu, Bull. soc. chim.. 27, 901 (1920).

CONDENSATION OF NITRII^S AND PHENOLS 2323

HO/NoH
A * 1 V JCH 2 CH 2 CN

H YY^ = NH.HC1
HO/No C(=NH-HC1) f ^ H2

I J. CH 2 CH 2 C1 X ^\C^ /

V H 2

iv n

At first Mechanism B might seem the more likely since Stephen 35 has
presented evidence that the mechanism of the reaction of nitriles with
resorcinol to give ketones is first a condensation of the cyano group with
the phenolic hydroxyl group yielding a structure similar to that in (IV),
then rearrangement of the group into the ring, and finally hydrolysis.
If such a primary mechanism is assumed in the case at hand, it is necessary
merely to suppose that in Compound IV the reaction of the chlorine
attached to the carbon atom with the hydrogen in the ortho position in the
benzene ring to eliminate the halogen acid and to form a ring structure,
has a greater tendency to take place than the rearrangement of the C
( = NH -HC1)CH 2 CH2C1 group from the oxygen to the ortho carbon atom,
and then the formation of a ring. This mechanism B, is shown to be in-
correct, however, since it is possible in the case of the condensation of
/3-chloro-propionitrile with resorcinol monomethyl ether to isolate a nitrile V.
The formation of such an intermediate excludes the possibility of the pri-
mary formation of Compound IV, and is strong evidence for Mechanism A.

The condensation of j8-chloro-propionitrile with resorcinol does not take
place in the presence of anhydrous zinc chloride alone. From this it might
be concluded that the hydrogen chloride which must be present for the
condensation to take place first reacts with the 0-chloro-propionitrile to
give a chloro-imide and that this resulting compound condenses with the
resorcinol. The fact that the condensation takes place slowly in the pres-
ence of hydrogen chloride alone is further evidence for this mechanism.
However, conclusive proof that the correct mechanism is presented by
Mechanism A is afforded in the condensation with resorcinol monomethyl
ether, by the isolation of the nitrile, V.

The condensation of resorcinol, orcinol and phloroglucinol with acrylic
nitrile also runs smoothly and yields the same phenolic lactones or acids
as are produced by using /3-chloro-propionitrile.

It seemed possible, since the chlorine in the position to a cyano group
reacts so readily with resorcinol, that the chlorine in #-chloro-propionic
ester, or even that the halogen in such compounds as allyl bromide, or
trimethylene bromide might also condense under the same conditions.

2324 WILSON D. I,ANGl,8Y WITH ROGER ADAMS

However, it is impossible to cause any condensation with these substances,
thus indicating that the cyano group undoubtedly plays an important
part in the reaction.

An attempt was also made to condense /3-chloro-propionitrile with
phenol in presence of zinc chloride and hydrogen chloride, but no reaction
occurred.

More surprising is the fact that 7-chloro-butyronitrile also reacts ab-
normally with resorcinol to give 2 unstable intermediate compounds which
hydrolyze to 7- (2,4-dihydroxyphenyl) butyric acid, VI.

TT/*k/\/

CHgO/NoH HO/NOH

HO | I C = NH-HC1

CH a

njw njn nui JWXT

IJcHaCHaCN lJcH,CH 2 CH,CO,H V/V/

CHCH,
V VI VII

The compound which is obtained directly from the original reaction mix-
ture, is probably the chloro-imide hydrochloride. Upon washing with
water, it is converted into a compound containing less chlorine, perhaps
a chloro-imide, which by contact with water for a longer time is hydro-
lyzed to the acid. The intermediate compounds were difficult to obtain
pure so that their structure was not definitely established. The yields
are lower, to be sure, than in the case of the -jchloro-propionitrile-resorcinol
condensation; nevertheless, the reaction takes place with comparative
ease and apparently gives no product in which the cyano group condenses
directly with the ring to give a ketone. The reaction of 7-chloro-butyro-
nitrile with orcinol or phloroglucinol does not go readily and no well de-
fined products were isolated.

The constitution of Compound VI from 7-chloro-butyronitrile and re-
sorcinol was proved by analysis, and by the fact that it does not form the
corresponding lactone under the same conditions as does j8- (2,4-dihydroxy-
phenyl) propionic acid. This fact excludes the possibility that 7-chloro-
butyronitrile undergoes a transformation before reacting with the resorci-
nol, namely the elimination and re-addition of hydrogen chloride to give
/3-chloro-butyronitrile, which then would condense to give /3-methyl
/3- (2,4-dihydroxyphenyl) propionic acid lactone imide hydrochloride, VII.

An attempt was made to condense ethylene cyanohydrin with resorcinol,
but only a small amount of 0- (2,4-dihydroxyphenyl) propionic acid was ob-
tained from the reaction mixture. If any ketone was formed, it could not
be isolated. The reaction does not run smoothly and this is probably due
to the molecule of water which must necessarily be eliminated in the re-
action. It is interesting, however, that the cyanohydrin of acetaldehyde
condenses normally with resorcinol. 36

CONDENSATION OP NITRIDES AND PHENOLS 2325

Experimental

jS-Chloro-propionitrile, C1CH 2 CH 2 CN. This substance was prepared according
to the method of Henry 5 , by the action of phosphorus pentachloride on ethylene cyano-
hydrin. The reaction was carried out in toluene suspension in preference to no solvent.
For successful results in the following experiments, the product must be pure. After
being freed as thoroughly as possible from phosphorus oxychloride by fractionation in a
vacuum (20-25 mm.), the /3-chloro-propionitrile fraction boiling at 65-72 at 20 mm. was
poured carefully into cold water and shaken, the temperature being kept low. The
oily nitrile was separated, washed with sodium carbonate solution, then with water,
dried over anhydrous sodium sulfate, and finally purified by distillation in a vacuum.

Acrylic Nitrile, CH 2 = CHCN. This substance was prepared by the method of
Moureu. 4

7-Chloro-buryronitrile, C1CH 2 CH 2 CH 2 CN. This was prepared from trimethylene
chlorobromide and sodium cyanide according to the method of Gabriel. 6

OHf
/3-(2,4-Dihydroxyphenyl)propionic Acid . In a 2-liter round-

bottom flask protected with a calcium chloride tube were placed 130 g. of c. P.
resorcinol, 90 g. of pure /3-chloro-propionitrile and 700 cc. of dry ether. To this solution
was added 40 g. of zinc chloride which had been freshly fused and then powdered, and
dry hydrogen chloride was passed in as long as there was any absorption. The flask was
then tightly stoppered and alloVed to stand overnight. Dry hydrogen chloride was
again slowly passed in for 5 hours and the flask was allowed to stand for 36 hours longer.
The mass of crystals which separated was sticky and hard to handle due to the presence
of zinc chloride but was filtered from the red solution and washed with dry ether. The
original filtrate was again stoppered and allowed to stand for 48 hours, during which 38 g.
more of solid formed. After filtering and allowing the filtrate to stand for a week longer,
25 g. more of crystals was produced.

The total quantity of crystals was dissolved in 450 cc. of water and heated on a
steam-bath for 4 hours. An oily layer of /3-(2,4-dihydroxyphenyl)propionic acid lactone
first separated which, if cooled, solidified. The layer however, was not removed, but
the reaction mixture was heated further, thus causing the lactone to go gradually into
solution. After this had cooled and stood for some hours, 86.5 g. of /3-(2,4-dihydroxy-
phenyl)propionic acid crystallized and was filtered. The aqueous filtrate, upon evapora-
tion in a vacuum to 175 cc. and cooling, yielded a second crop of crystals which weighed
22.5 g. Further concentration and cooling of the filtrate yielded only inorganic salts.
The total yield of product was thus 109 g. (56%). The substance was almost always
light brown and this color was difficult to remove even though several crystallizations
from water and bone black were made. The substance always separated from the aque-
ous solution very slowly. In spite of the color, the product melted sharply at 165 with
decomposition, the same point as pure white material obtained by hydrolysis of the pure
lactone.

Analyses. Subs., 0.3679: CO 2 , 0.7967; H 2 O, 0.1799. Subs., 0.2973: 24.24 cc. of
0.0717 N NaOH. Calc. for C 9 Hi O 4 : C, 59.34; H, 5.49; neut. equivi, 182. Found:
C, 59.05; H, 5.43; neut. equiv., 171.

This substance was prepared by Hlasiwetz 7 by the reduction of umbelliferone, but
he stated that the acid decomposed when heated above 110. In order to make certain

6 Henry, Bull. acad. roy. med. Belg., [3] 35, 360 (1898).
Gabriel, Ber., 23, 1771 (1890).

7 Hlasiwetz, Ann., 139. 102 (1866).

2326 WIIvSON D. LANGLEY WITH ROGER ADAMS

that the compound which was obtained in this investigation had the structure assigned
to it, umbelliferone was made by the action of malic acid upon resorcinol by means of
cone, sulfuric acid. The product was carefully purified and then reduced with sodium
amalgam as described by Hlasiwetz. Upon acidification with hydrochloric acid, red
crystals formed, which were boiled with bone black in water solution, and filtered.
Nearly colorless crystals melting at 165 were obtained which, when mixed with /3-(2,4-
dihydroxyphenyl)propionic acid, melted at the same temperature as when alone. For
further confirmation, another sample of the acid was treated with excess of acetic anhy-
dride and boiled for 20 minutes. The reaction mixture was poured into water to de-
compose the excess of anhydride, and the brown solid which separated was filtered,
washed and crystallized from 20% methyl alcohol. A white powder, the acetate of
dihydro-umbelliferone, [/?-(2-hydroxy-4-acetoxyphenyl) propionic acid lactone], was thus
produced which melted at 112 and proved to be identical with the compound prepared
by the action of acetic anhydride on ]3-(2,4-dihy droxyphenyl) propionic acid made from
resorcinol and 8-chloro-propionitrile.

/3-(2,4-Dihydroxyphenyl)propionic acid may also be made by the action of acrylic
nitrile upon resorcinol. A solution of 12 g. of resorcinol and 6 g. of acrylic nitrile in 70
cc. of dry ether was treated with 4 g. of freshly fused and powdered zinc chloride, after
which dry hydrogen chloride was passed in until the solution was saturated. At the
end of 12 hours, a heavy white crystalline product had formed. The ether was decanted
and a small portion of the solid treated directly with water. There was thus produced
/3-(2,4-dihy droxyphenyl) propionic acid lactone, which melted at 133-134, and proved
to be identical with the substance prepared from resorcmol and /3-chloro-propionitrile.

The remainder of the solid was boiled with 50 cc. of water for a half hour, and then
allowed to stand. Crystals separated which proved to be 0-(2,4-dihydroxyphenyl)-
propionic acid.

/3-(2,4-Dihydroxyphenyl) propionic Acid Lactone, C H . /3-(2,4-Dihy-

c
H 2

droxyphenyl) propionic acid was heated in an oven at 130-135 for 2 hours. The
melt which formed was allowed to cool and the product then crystallized from
toluene. White cubical crystals resulted which, upon complete purification melted at
132-133. This substance is identical in all respects with that obtained by the treat-
ment of the original reaction product of resorcinol and 0-chloro-propionitrile with cold
water.

Analyses. Subs., 0.1560: CO 2 , 0.3795; H 2 O, 0.0072. .Calc. for C 9 H 8 O 8 : C, 65.85;
H, 4.88. Found: C, 66.33; H, 4.96.

The substance was insoluble in sodium carbonate until the solution was heated,
when it dissolved. Upon acidification with acid ^-(2,4-dihydroxyphenyl) propionic acid
gradually deposited in the form of hard crystals.

/3-(2-Hydroxy-4-acetoxyphenyl) propionic Acid Lactone,

c /
H 2

A solution of 7 g. of /3-(2,4-dihy droxyphenyl) propionic acid in 10 g. of acetic anhydride
was heated on a water-bath for an hour. A red solution was produced ; this was cooled
and then poured into 70 cc. of cold water when a yellow crystalline solid separated

CONDENSATION OF NITRIDES AND PHENOLS 2327

which weighed 9 g. Upon crystallization from 20% methyl alcohol the substance was
readily obtained pure and formed white crystals melting at 111-112.

Analyses. Subs., 0.3116, 0.3278: CO 2 , 0.7382, 0.7742; H 2 O, 0.1357, 0.1387.
Calc. for C u Hio0 4 : C, 64.1; H, 4.89. F^und: C, 64.63, 64.40; H, 4.83, 4.67.

It was insoluble in sodium carbonate until heated. Upon hydrolysis either with
sodium carbonate or water, /3-(2,4-dihydroxyphenyl)propionic acid resulted.

-(2,4-Dimethoxyphenyl)propionic Acid, . A solution of

35 g. of /3-(2,4-dihydroxyphenyl)propionic acid in 100 cc. of 10% sodium hydroxide
was treated with 36 g. of dimethyl sulfate. The mixture was thoroughly shaken for
some minutes and then heated on a water-bath under a reflux condenser. After the
initial reaction had taken place, 100 cc. more of 10% sodium hydroxide solution and 35 g.
of dimethyl sulfate were added and the mixture was again heated until the dimethyl
sulfate was completely decomposed. Upon cooling the reaction mixture and acidifying
with hydrochloric acid, needle-like crystals separated. The yield was 28.5 g. (68%).
It was readily purified by crystallization from water and then melted at 102.5-103.5.

Analyses. Subs., 0.2991: CO 2 , 0.6896; H 2 O, 0.1739. Subs., 0.1260: 8.7 cc. of
0.0717 N NaOH. Calc. for CuH 14 O 4 : C, 62.8; H, 6.6; neut. equiv., 210. Found:
C, 62.87; H, 6.5; neut. equiv., 201.

This substance is without doubt the same as that prepared by W. Will 8 by the re-
duction of dimethoxy-umbellic acid. He reported the melting point as 105.

0-(5-Nitroso-2,4-dihydroxyphenyl)propionic AcidLactone,

N \ C /

H 2

A solution of 3 g. of /3-(2,4-dihydroxyphenyl)propiomc acid in 50 cc. of alcohol was treated
with 4 g. of freshly distilled butyl nitrite. The flask was then cooled and 10 cc. of cone.
hydrochloric acid added. Fumes of nitrogen dioxide were evolved, heat was generated,
and a red precipitate was formed. An additional 40 cc. of cone, hydrochloric acid was
then added and the flask allowed to stand for about 30 minutes. The solid was filtered,
washed twice with cold water, and dried. The yield was 4 g. The crude material
varied in color in different experiments from a cream color to red, and upon exposure to
air it generally turned green. The crude substance was purified by dissolving 4 g. in
70 cc. of boiling water, filtering hot and allowing to cool. In spite of the fact that the
aqueous solution was green, a cream-colored solid separated. Upon filtering, washing
and drying over sulfuric acid in a vacuum desiccator, the product did not change color
and melted sharply at 147.5-148, with decomposition.

Analysis. Subs., 0.1873: N 2 , 13.1 cc. (26 and 737 mm.). Calc. for C 9 H 7 O4N:
N, 7.25. Found: 7.61.

This substance could not be formed by treatment of 0-(2,4-dihydroxyphenyl)-
propionic acid lactone with butyl nitrite, but had to be prepared from the acid.

CH 3 O/NOH
j8-(2-Hydroxy-4-methoxyphenyl)propionic Acid,

A mixture of 30 g. of resorcinol monomethyl ether and 22 g. of /S-chloro-propionitrile
8 Will, Ber., 16, 2116 (1883).

2328 WILSON D. LANGLEY WITH ROGER ADAMS

in 300 cc. of dry ether was treated with 20 g. of freshly fused and powdered zinc chloride.
The mixture was cooled in an ice-bath and dry hydrogen chloride passed in until the
ether was saturated. Upon standing overnight, a sirupy mass deposited. The ether
was decanted and the residue treated with 30 cc. of water. After standing for 15 hours,
a small amount of solid and some liquid separated. The oil (10 g.) was filtered from the
solid (3.5 g.) by the use of suction. The solid, upon crystallization from water, was
obtained pure and melted at 138-139.5. It proved to be /9-(2-hydroxy-4-methoxy-
phenyl)propionic acid.

Analyses. Subs., 0.3819: CO 2 , 0.8604; H 2 O, 0.1982. Calc. for Ci 2 Hi O 4 : C, 61.2;
H, 6.1. Found: C, 61.4; H, 5.8.

/3-(2-Hydroxy-4-methoxyph'enyl)propionicAcidLactone,

\/\C/

H 2

The oil obtained by filtration of the 0-(2-hydroxy-4-methoxyphenyl)propionic acid was
insoluble in cold sodium hydroxide solution, but dissolved on heating. It could not be
distilled under a pressure of 6 mm. It was found, however, that by boiling the oil with
sodium carbonate solution for some time, and acidifying the resulting solution, it was
converted completely into /3-(2-hydroxy-4-methoxyphenyl)propionic acid, which melted
at 138-139. The same oil was produced by heating /3-(2-hydroxy-4-methoxyphenyl)-
propionic acid in an oven at 132 and was, therefore, without doubt /3-(2-hydroxy-4-
methoxyphenyl)propionic acid lactone.

CH 8 C/\OH
/3-(2-Hydroxy-4-methoxyphenyl)propionitrile, .The ether

solution which was decanted from the original reaction mixture between /3-chloro-
propionitrile and resorcinol monomethyl ether was evaporated in a vacuum and about
12 g. of unchanged /3-chloro-propionitrile recovered. A higher-boiling fraction of re-
sorcinol monomethyl ether (21 g.) was also obtained. The residue in the flask solidified
on cooling. It was purified by crystallization from dil. alcohol and then melted at 126.5-
127.5. Analysis showed it to be /3-(2-hydroxy-4-methoxyphenyl)propionitrile.

Analyses.' Subs., 0.3718: N 2 , 27.2 cc. (28 and 749 mm.). Subs., 0.2968: CO 2 ,
0.7353. Calc. for Ci H u 2 N: C, 67.6; H, 6.2; N, 7.9. Found: C, 67.5; H, 6.5; N, 8.0.

It was insoluble in sodium carbonate but soluble in sodium hydroxide solution and
was recovered unchanged upon acidification. When it was boiled with alkali ammonia
was evolved and, after acidification, /3-(2-hydroxy-4-methoxyphenyl)propionic acid was
obtained.

HO/NoH

-y-(2,4-Dihydroxyphenyl)butvric Acid, . A mixture of

1 X/ JCH 2 CH 2 CH 2 C0 2 H

64 g. of resorcinol and 60 g. of 7-chloro-butyronitrile was dissolved in 400 cc. of dry
ether; 30 g. of fused and powdered zinc chloride was added, and dry hydrogen chloride
was passed in rapidly for 2 hours. The flask was then sealed, and allowed to stand.
Hydrogen chloride was passed in for 1 hour on each of 2 successive days, and the
flask was allowed to remain sealed for 4 more days. By that time the oil which had
separated had entirely crystallized. The ether was filtered and the solid dissolved in
250 cc. of water, heated for 3 hours on a steam-bath and then cooled. The crystals
which slowly formed were filtered and dried, and weighed 24.5 g. (20.8%).

CONDENSATION OF NITRILES AND PHENOLS 2329

For purification, 10 g. of the product was dissolved in 57 cc. of boiling water, ancj
filtered hot. An oil separated, which when cooled and stirred, solidified. The weight
recovered was 9.5 g., and the substance melted at 89-99. Repeated crystallization did
not make this melting point sharper, but after the melt had been allowed to solidify, it
reinelted at 118-119. This indicated that the low-melting product contained water
of crystallization, and a moisture determination confirmed this.

Analyses. Subs., 0.7322, 0.5297: H 2 O, 0.0655, 0.0471. Calc. for CnH 12 O4.H 2 O:
H 2 O, 8.4. Found: H 2 O, 8.94, 8.89.

The anhydrous product was readily crystallized from benzene hi the form of color-
less plates, and melted at 118.5-119.

Analyses. Subs., 0.1628: CO 2 , 0.3640; H 2 O, 0.0865. Subs., 0.1491: 9.98 cc. of
0.0717 N NaOH. Calc. for Ci H 12 O 4 : C, 61.22; H, 6.12; neut. equiv., 198. Found:
C, 60.96; H, 5.95; neut. equiv., 208.

A portion of the anhydrous y- (2,4- dihydroxyphenyl) butyric acid was heated in an
oven at 133-134 for 0.5 hour. The solid melted to a red liquid which solidified on cool-
ing and, upon recrystallization from benzene, proved to be unchanged starting material.
Therefore, no lactone of this acid formed under the same conditions as were used with
/3- (2,4-dihydroxyphenyl)propionic acid.

In order to study the intermediate products formed in the condensation of resorcinol
and 7-chloro-butyronitrile, another condensation was made under the same conditions
as are given above. The initial crystalline .reaction product was treated with water,
warmed to 50, cooled, and the solution partly neutralized with dil. sodium carbonate
solution. The oil which formed solidified, and the resulting substance was filtered and
washed with dil. hydrochloric acid. It was a white powder, melting at 214216, but
when it was washed with water the color changed to a canary-yellow, and the melting
point dropped to 190-192. When the yellow compound was washed with methyl al-
cohol saturated with hydrogen chloride, it became white, and the original higher-melting
product resulted.

The wash waters, on standing, gave a small amount of white solid which proved to
be /3-(2,4-dihydroxyphenyl) butyric acid.

/3-(2,4-Dihydroxy-6-methylphenyl)propionic Acid Lactone,

Five g. of orcinol and 5 g. of /3-chloro-propionitrile were dissolved in 100 cc. of dry ether,
and 5 g. of freshly fused and powdered zinc chloride added. The ether was cooled in an
ice-bath, and dry hydrogen chloride passed in until the ether was saturated. A white
solid formed overnight, and after 10 hours there appeared to be no more deposit forming.
The ether was decanted, the solid washed with dry ether, treated with 20 cc. of water,
warmed to 50 and cooled in an ice-bath. The oil which formed could not be made to
crystallize, even when dry toluene was used as the solvent. The heating was, therefore,
continued and ammonia was added until the solution was just neutral. After 1 hour the
solution was boiled with bone black, filtered and acidified with hydrochloric acid. On
cooling, an oil layer separated which slowly solidified on standing. This was filtered and
dried, when it weighed 4 g. and melted at 125-133. It was recrystallized from hot
dil. methyl alcohol, 4 g. dissolving in 90 cc. of 10% alcohol; 1.7 g. of solid was obtained
melting at 140-141 .5. The remainder of the product came out as an oil which solidified
after 2 weeks' standing. This was again boiled with bone black and filtered, when 0.2 g.
more of solid was obtained.

2330 WILSON D. LANGLEY WITH ROGER ADAMS

Analyses. Subs., 0.3923: CO 2 , 0.9641; H 2 O, 0.1934. Calc. for C 10 H 10 O,: C, 67.4;
H,5.6. Found: C, 67.01 ;H, 5.5.

This product was difficultly soluble in hot water, but readily soluble in hot alkali.
Since the lactone was obtained on acidification of the alkaline solution, the acid is not
stable under ordinary conditions. The same product was obtained in a purer form from
acrylic nitrile, 3.6 g. being obtained from 5 g. of orcinol and 2.3 g. of acrylic nitrile.

j8-(2,4,6-TrihydroxyphenyI)propionic Acid Lactone, Q H(J! . Five

<=H,

g. of pure phloroglucinol, dried at 120 to free it from water of crystallization, and
5 g. of /3-chloro-propionitrile were dissolved in 100 cc. of dry ether, and 4 g. of powdered
zinc chloride was added. The flask was cooled and dry hydrogen chloride was passed in
until the solution was saturated. The mixture was allowed to stand for 20 hours, the
ether was decanted from the solid which had separated and the residue washed with dry
ether. It was then treated with 20 cc. of water, warmed to about 40 for 5 minutes, and
cooled with ice. No oil layer separated. After standing for 24 hours, the water solution
was extracted twice with 50 cc. portions of butyl alcohol, the butyl alcohol in turn
washed with water, and the wash water extracted with fresh butyl alcohol. The
butyl alcohol was distilled in a vacuum and 10-12 g. of dark brown liquid was obtained.
This was taken up in 150 cc. of boiling water, heated until all of the butyl alcohol had
been removed, treated with bone black, and filtered. The filtrate was light brown, and
from it there separated about 5 g. of a viscous oil. This could not be obtained crystal-
line. It changed to a glassy mass on standing for a long time in a desiccator over sul-
furic acid, and could not be crystallized from anhydrous solvents.

This -oil decomposed carbonate solutions when heated, and went into solution.
On acidification, the oil formed again. Attempts were made to prepare derivatives of
the oil, such as those with phenyl-isocyanate, diphenyl-carbamine chloride, and the
acetate and benzoate. None of these could be obtained in crystalline form. Acrylic
nitrile gave the same type of oil, which possessed all the properties of that obtained
above.

Summary

1. /3-Chloro-propionitrile condenses with resorcinol in the presence
of anhydrous zinc chloride, hydrogen chloride and dry ether to give a
crystalline intermediate product which hydrolyzesto form /3-(2,4-dihydroxy-
phenyl)propionic acid.

2. By a similar procedure /3-chloro-propionitrile and resorcinol-mono-
methyl ether give a mixture of j3-(2,4-dihydroxyphenyl)propionitrile and
the corresponding acid and lactone. From orcinol and phloroglucinol
intermediate products are formed which give on hydrolysis the analogous
lactones.

3. 7-Chloro-butyronitrile and resorcinol condense under similar
conditions to give a product which hydrolyzes to form 7-(2,4-dihydroxy-
phenyl)butyrfr acid.

4. The mechanism by which the above products are formed is discussed.
URBANA, ILLINOIS

VITA

The writer was born in Charleston, South Carolina, on January
7, 1895. At the age of nine years, he moved to Schenectady, New York,
where he completed his grammar and high school education. He
entered Wesleyan University, Middletown, Connecticut, in the class of
1918, and secured the degree of Bachelor of Science in that year. He
completed one year of graduate study at the same institution, and re-
ceived the degree of Master of Science in 1919. He has since been
a graduate student at the University of Illinois.

His teaching appointments have been as follows :

1919-20 Graduate Assistant in Chemistry

1920-21 Assistant in Chemistry

1921-.22 (first semester) Graduate Assistant in Chemistry.

UNIVERSITY OF CALIFORNIA LIBEARY,
BERKELEY

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW

Books not returned on time are subject to a fine of
50c per volume after the third day overdue, increasing
to $1.00 per volume after the sixth day. Books not in
demand may be renewed if application is made before
expiration of loan period.

SEP

d 1944

,, Pft

2U1954

15m-12,'24

530283

UNIVERSITY OF CALIFORNIA LIBRARY