March ±, 1871.] 
THE PHARMACEUTICAL JOURNAL AND TRANSACTIONS. 
701 
CHLORAL HYDRATE AND CHLORAL 
ALCOHOLATE. 
C 2 CI 3 H 0, H 2 O.—C 2 Cl 3 H O, C 2 H c O. 
BY DR. F. VERS MANN. 
The sale and use of chloral hydrate has of late so 
largely increased, that it becomes necessary to be 
quite certain of the quality of the articles, and of the 
exactness of the method of testing it. In analysing 
a great many samples, from, I believe, almost every 
maker, I was naturally led to inquire into the phy¬ 
sical properties of the hydrate, and also of the 
chloral alcoholate which may possibly be met with 
in the market as a substitute for the former. 
Dr. Paul, in his article in this Journal of 4th 
ult., has drawn attention to the difference in the 
crystalline forms in which the hydrate is sold; this 
is, no doubt, owing to the different solvents employed 
for recrystallization. Thus, a concentrated aqueous 
solution, placed under the air-pump, gives rhomboid 
crystals, ether gives small hard crystals, acetone 
fine needles, warm benzole supersaturated deposits 
on cooling also fine needles; whereas a solution in 
benzole, allowed gradually to evaporate, deposits 
large crystals sometimes £ in. long. Bisulphide of 
carbon in same manner yields either fine needles or 
large crystals. A saturated alcoholic solution gives 
beautiful long feathery crystals. I obtained some 
14 in. long, which have all the appearance of the 
alcoholate, and which were found to be so. This is 
remarkable, and may possibly account for the fact of 
one sample, obtained as hydrate, being pure alcohol¬ 
ate, as the manufacturer may have recrystallized 
the impure hydrate from alcohol without being 
aware of the change produced. 
The hydrate is extremely hygroscopic, the more 
so the smaller the crystals are. 10 grains of fine 
needles left in an open vessel became quite fluid in 
a day, whereas the same quantity of hard crystals 
became only opaque, and again transparent the fol¬ 
lowing day. But in both forms the compound is so 
volatile at ordinary temperature, that the first had 
completely evaporated after five days, carrying away 
with it the moisture absorbed, and the last in eight 
days. 
The hydrate is extremely soluble in water, 100 
parts of water dissolve as much as 360 parts of dry 
crystals; the alcoholate is also soluble in water, but 
to a much smaller extent, and much slower. 
In fact, the two may be readily distinguished in 
the following manner:—Take a pretty wide beaker- 
glass, 6 or 8 in. high, full of water, drop a few crys¬ 
tals into it: the hydrate sinks down at once, and is 
almost dissolved before it reaches the bottom. With 
the alcoholate the larger crystals only will sink to 
the bottom, and lie there for several minutes before 
they gradually and very slowly disappear; but small 
crystals or fragments of crystals will float on the 
surface of the water, and as soon as they are at¬ 
tacked by the water, the slight current of the saline 
solution sinking down occasions sufficient disturb¬ 
ance to apparently impart life to the solid particles, 
—they begin to spin round and round, and dart from 
one side of the beaker to the other, until the very 
last solid particle has disappeared. This is not only 
a very pretty and amusing sight, but it is really a 
distinctive mark between the hydrate and the alco¬ 
holate. Slightly tepid water makes the action even 
more violent. 
The specific gravities of solutions in water of the 
Third Series, No. 36. 
two compounds also show a great difference, as will 
be seen by the following figures :— 
Temp. 15'5° C. (G0° F.) Hydrate. Alcoholate. 
20 per cent, solution. . . 1085 1072 
15 „ „ ... 1062 1050 
10 „ „ ... 1040 1028 
5 „ „ ... 1019 1007 
The specific gravity of the two substances in the 
liquid state is another criterion, that of the hydrate 
being 1610 at 49° C. (120° F.), and of the alcoholate 
1143 at 40° C. (104° F.) 
I do not attach any value to the boiling-point as a 
test for purity, and for this reason. Both hydrate 
and alcoholate at that temperature begin to decom¬ 
pose into chloral and water or alcohol respectively, 
and it is sometimes extremely difficult to take cor¬ 
rect observations. I have had undoubtedly good 
samples of hydrate commencing to boil only above 
100° C., and of alcoholate commencing to boil at 
80° C. 
So far I have treated the two substances sepa¬ 
rately, but if the alcoholate really should be intro¬ 
duced as an adulterant, the direct proof of alcohol 
will become necessary. For this purpose Lieben’s 
method is the best, who converts the alcohol into 
iodoform, and detects minute quantities. 
All the samples I have hitherto had occasion to 
test were true chloral hydrate, with the exception of 
one ; but they varied very much in the percentage of 
moisture, and for this reason it is necessary to have 
a ready and accurate method of determining the 
percentage of true hydrate in a sample. 
We now know only the ammonia test. Tliis me¬ 
thod, if properly carried out, is sufficiently exact to 
decide between hydrate and alcoholate, where the 
percentage of the resulting chloroform differs as 
much as 72*2 and 61*76; but it cannot claim analv- 
tical accuracy. 
The column of ammonium formiate solution al¬ 
ways takes up some chloroform, and the chloroform 
layer is never free from water; chloroform is dis¬ 
solved while it separates from the ammonia, and 
its true percentage invariably decreases. 
It has also been suggested that during the twelve 
hours contact of the ammonia and chloroform, the 
last might be further decomposed into hydrochloric 
and formic acid; but this I find not to be the case. 
Experiments made with chloroform and ammonia 
and with chloroform and pure water gave in both 
cases a loss of chloroform amounting to 0*2 c. c., the 
same well-stoppered tubes being used as in the analy¬ 
sis of hydrate, of which always 10 grammes were 
taken. It is evident, therefore, the decrease in 
chloroform is not owing to the action of ammonia, 
but to its solubility in water. 
Tliis and the length of time required for the am¬ 
monia test are certain drawbacks; and I have tried 
another plan, which is both more accurate and more 
expeditious. 
I take advantage of the facility with which the 
chloral hydrate and the chloral alcoholate are de¬ 
composed by strong sulphuric acid with separation 
of chloral, which, in a graduated tube, may be read 
off and the percentage of hydrate calculated. 
I take about equal parts by weight, i. e. 10 
grammes of the hydrate and from 5 to 6 c. c. of sul¬ 
phuric acid; the quantity of the acid is of no conse¬ 
quence within certain limits; 5 parts of hydrate and 
one part of acid do not separate chloral, even when 
