Chemical Library
chemical Libaº
Q,\\
2
ºf
ºn 3
*To PREPARE AN ALIPHATIC
B-DIAZO CCMPOUND
BY
VIRGIN IA BARTOW
-
A. B. V.Assar College, 1918
A. M. UNIVERSITY of ILLINois, 1921
T H E S IS
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, 1923
URBANA, ILLINOIS
1923


AN ATTEMPT
TO PREPARE AN ALIPHATIC
B-DIAZO CCMPOUND
- BY
VIRGINIA BARTOW
A. B. VASSAR College, 1918
A. M. UNIVERSITY of ILLINois, 1921
T H E SIS
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, 1923
URBANA, ILLINOIS
19 2 3
42%
3-22 - 22%
1. INTRODUCTION.
Twenty-five years after the aromatic diazo compounds were discov-
ered by Johann Peter Griess, the first aliphatic diazo compound was
obtained by Schiff and Meissen who prepared diazo camphor from
camphor imide by means of a nitrite. Judging by the literature on the
subject, it is perhaps more usual to designate the preparation by Cur-
tius” in 1883 of diazo acetic ethyl ester by the action of nitrous acid on
glycocol ester hydrochloride as the real impetus to definite investigation
in this field. Since the original production of the aliphatic diazo com-
pound, there have been many efforts to expand the knowledge of this
subject for various reasons. However, the very broad application of
the diazo reaction in synthetic work in the aromatic series has never been
extended to the aliphatic compounds owing to the unexplained peculiar-
ities which make the preparation of the latter difficult.
Synthetic interests have been displaced in the last twenty years by the
controversy concerning the fundamental structure of the aliphatic diazo
type. Although much of the investigation has been developed to explain
the mechanism of the reactions of these compounds, a second interest
is probably more fundamental since it involves a deeper insight into the
methods by which atoms combine to form molecules. This is due to the
conception and gradual evolution of the electron theory. Obviously,
the preparation of any electromers which the theory indicates should
exist, would be an important point in its favor. The wide application
of this theory which is donsidered unwarranted by some would then
. have concrete evidence for its support. -
* Professor Noyes has been interested for some time in the possibility
of preparing such electromers. The aliphatic diazo compounds because
of the accepted view of their structure should be an excellent type of com-
pound to produce these electromers as has already been stated by Mar-
vel and Noyes 4 and Chiles and Noyes". This is because their method
• 3
of formation is such that the two nitrogen atoms are derived from am-
monia and nitrous acid which would have the following electronic form-
ulas:– -
+
H
/ +
– + - - -H ++ - -
N — H - H — O — N = O
N +
H
If then, one nitrogen atom is positive while the other is negative, the
electronic formula of an aliphatic diazo compound which is
++
R — C – R^
—/N+
+ *=º:
N = N
according to Curtius," or given as a straight chain
R -
C — —H N = N
/
• R’
by Angeli" and Thiele * indicates the possibility of electromers.
Accordingly, an attempt was made to obtain an aliphatic beta diazo
compound which might give evidence of the existence of electromers.
Unfortunately, the apparent non existence of the beta diazo compounds,
at least in a stable form, has rendered their preparation extremely diffi-
cult. This investigation has been concerned with preliminary materials
which have failed to give the desired results. .
II. THEORETICAL BASIS.
1. The Development of the Electron Theory of Valence. .
The original conception of the possibility of the existence of electrons
as a part of the structure of atoms and molecules upon which a theory
of valence can be built, can not be definitely stated. It seems rather to
be the outgrowth from suggestions that various observers have made to
explain experimental facts. Previous to the nineteenth century, the
idea of an electron was undeveloped but the explanation of chemical phe-
nomena by electrical means was by no means untried. Lavoisier" had
a true dualistic conception of chemical combination. Davy 19 decomposed
the oxides of potassium and sodium by the electric current which indica-
4.
ted some definite relationship between chemical and electrical phenomena.
Berzelius,” the strong advocate of the dualistic theory which was the
result of such experiments as Davy's, defined chemical combination as
electrical in nature. When Dumas 14 found that hydrógen, heretofore
considered positive, could be replaced in an organic compound by the
negative element chlorine, the electrical theory was abandoned for the
well known theory of types. In spite of the obscurity into which the
older dualistic explanation of chemical combination passed, similar ideas
were destined to appear again. 1887 brought Arrhenius's theory of
dissociation. In less than ten years, what might be considered the be-
ginnings of the electron theory came.
In 1895 van't Hoff,” who had noticed that in an oxidation the rate
was proportional to a part of the oxygen pressure, suggested that the
molecules of oxygen might be separated into ions not atoms. He made
the statement without a definite expression of the existence here of an
electrically charged state. It was Noyes and Lyons ** who first made
the suggestion that the chlorine molecule might very likely dissociate
into positive and negative chlorine. This was to explain the formation
of nitrogen tri-chloride by the action of chlorine on a dilute solution of
ammonia. Stieglitz" published a communication in which he said that
for some time he had presented similar ideas to his classes.
An observation that electrolytic chlorine had a greater re-activity than
ordinary chlorine was explained by Larsen” by the fact that there
might be a change in the structure during the electrolytic process. This
was referred to by Walden " who offered a comparable explanation to
account for the fact that the electric current is carried into a solution
of bromine or iodine in liquid sulfur dioxide. His experiments showed
that these halogens may have different forms. Wislicenus” in a brief
note is quite ready to agree with Walden that there are “verschiedene
Halogene”. • ‘
At practically the same time, J. J. Thomson” published his con-
ception of the electron theory of valence in which he stated that there is
an actual transfer of electrons from one atom to another when they
combine to form molecules. This was substantially the holding force or
valence power. A few years later 49 he referred to the possibility of
isomers in simple substances and to the electric conductivity of heated
gases which would seem to indicate a dissociation into charged par-
ticles. Abegg's* correlation of the corpuscular theory of valence
with the electron theory accompanied by the ideas of normal and
5
{~,
contra valence expressed the same hypothesis. It remained for Ram-
say” to propose that the electron was the elementary unit of electricity
or atom of the element electricity. With the development of these fun-
damental conceptions of the electron theory, the study of the electrons as
a real part of atomic and molecular structure became established. Those
investigators who have enlarged the field particularly are Falk and
Nelson,” Fry,” L. W. Jones,” and Stieglitz.”
By recent work in this laboratory, Noyes” with his assistants has
shown that nitrogen trichloride could be prepared by the action of an-
hydrous ammonia and anhydrous chlorine. The nitrogen trichloride
thus formed was converted quantitatively by anhydrous hydrochloric
acid gas into ammonium chloride. This reaction seemed to give evidence
that non-electrolytes do separate into positive and negative atoms during
the reaction. Further work by Noyes and Wilson” in the dehydration
of hypochlorous acid indicated the presence of positive chlorine. Investi-
gations” of hydrated amine oxides showed a significant difference in the
attachment of the hydroxyl group in these compounds.
2. THE DIAzo CoMPOUND: STRUCTURE AND APPLICABILITY
• ‘ TO THE PROBLEM
The study of diazo compounds in the aliphatic series has led to the
proposal of two distinctly different yet possible structures for them.
Curtius" offered the following ring structure to explain the facts known -
at the time when he studied them. This was
R — C — R^
/ N
N = N
and was established without question until about fifteen years ago when
Angeli’ proposed the straight chain instead of the nitrogen-carbon ring.
Jº . R
R/ - -
After Staudinger” prepared this compound by the action of mercuric
oxide on hydrazones, Thiele” gave more publicity to the straight chain
structure which bears his name with Angeli. Evidence as to which is
correct is conflicting. Hantzsch and Lifschitz” agreed with Curtius
after investigation of the absorption spectra of diazo methane. Forster
- and Cardwell” felt that the production of a hydrazone by the action of
6
the Gringard reagent on a diazo compound indicated the existence of the
nitrogen atoms in the straight chain. Darapsky and Prabhakar” by
reducing diazo acetic ethyl ester obtained a hydrazone which seemed to
show the Thiele formula to be more logical. The preparation of diazo
methane from hydrazone, potassium hydroxide and chloroform * can be
explained either way. Staudinger and his co-workers* conducted a
very thorough investigation of the diazo compounds in which they did
not settle definitely which structure was the most suitable. They did
try to prepare isomers which they thought might have the two formulas
but without success. Just recently Staudinger* has published a further
paper which supports the Curtius formula which he has felt before to be
rather better. Finally Langmuir.” supported Thiele if one bond be-
tween the two nitrogen atoms were dropped. The principal reactions
upon which Curtius originally based his claim to a ring structure were
some eleven in which the substances replacing the nitrogen were attached
directly to the carbon atom in two parts. No newer evidence can really
displace the Curtius formula so that it is perhaps more generally accept-
ed although both structures explain the reaction which the diazo com-
pounds undergo.
Recently a series of papers by Levene have given evidence that elec-
tromers in the diazo series do exist. Fischer” and his assistants found
that the ester and acid of certain amino acids did not give the same
hydroxy acid on treatment with nitrous acid. To explain this fact sev-
eral investigators have tried to present a theoretical foundation. The
theory that Levene has been trying to develop for this purpose is that the
diazo form is an intermediate in the reaction by which alpha amino acids
are converted into alpha hydroxy acids. His basis for believing this was
the isolation of a diazo compound after the reaction of nitrous acid on
the benzal of chitosamine ethyl ester.” He then reasoned in this way:
if this is true, then the diazo compounds formed from epimeric amino
acid esters should be the same if the accepted structure for the diazo
type is correct. If, however, the two epimeric diaminized acid esters
formed finally, differ, then it is reasonable to suppose that the diazo
cómpounds in the two cases differ in some way. They would very likely
exist in two forms which would be of the nature of electromers.
Levene prepared three pairs of epimeric amino acids which gave in
each case a distinct, active anhydro acid by the action of nitrous acid.
This was followed by definite evidence of the existence of asymmetry in
diazo succinic acid ester which was prepared from d-aspartic acid ester.”
7
Proof of this activity which showed but a small degree of optical rota-
tion was further established by conversion of the diazo compound to
diethyl malate,” the activity of which was known, by formation of
active halogen compounds from the diazo and finally by preparing ben-
zoyl diethyl malate.* Levene's latest papers have been concerned with
benzilidene 1 ethyl 2 diazo gluconate” which gave no epimers with
anhydrous hydrochloric acid and hydrobromic acid gas. Reduction of
this substance by aluminum amalgam * did give two products. This is
slightly in contrast to Staudinger* who obtained two similar products
from diazo acetic ethyl ester by using aluminum amalgam in one case
and hydrogen in the other. These last papers seem to indicate that
there may be two distinct diazo compounds as well as the previous pre-
paration of such a substance which was optically active. -
In this laboratory, there have been two attempts to prepare active
diazo compounds. Marvel and Noyes" could get no isolated pro-
duct, using aliphatic alapha amines for diazotization. Chiles and Noyes"
were unsuccessful in attempts to prepare a gamma diazo compound
from gamma amino valeric acid ester but did succeed in satisfactorily
isolating six active diazo compounds from alpha amino esters.
(3). PREPARATION
The actual preparation of the aliphatic beta diazo compound was
not accomplished. For years, it has been well known that the corre-
sponding alpha compounds have been made with great difficulty. Leu-
cine, tyrosine, alanine, aspartu and glutamic acid esters give only diazo-
succinic ester. Any product was purified only with special care.”
A thorough investigation by Curtius” after preparation of new amino
acid esters by E. Fishcher” and E. Erlenmeyer Jr.” using nitrous acid
on amino acids and amino acid esters, resulted in the statement that
alpha-diazo esters in the aliphatic series could be prepared only from
alpha amino acid esters. Beta and gamma diazo compounds could not be
isolated if they existed at all. The difficulties of the case deserve to be
considered and avoided if possible. -
Neither of the two other general methods for the preparation of
an aliphatic diazo compound had produced this group in the beta posi-
tion. These methods are von Pechmanns's" involving the decom-
position of nitrosamines and Staudinger’s” oxidation of hydrazones
of aldehydes and ketones by mercuric oxide. Since the latter two
methods were found impractical by previous work,” the preparation
8
by the Curtius method, which was the other alternative, necessitated
the choice of an amine. -
To decide what amine to use, not only was the ultimate formation
of the diazo structure of first importance but also it was imperative that
the nitrogen group be contained in a compound in which no other asym-
metry except that due to the nitrogen could occur. An amine had to
be selected that should diazotize, as difficulty has often been encoun-
tered at that point. To obtain such an amine, for two reasons, an alpha
dialkyl substitution product of aceto acetic ester was chosen. The first
was the fact that this keto ester had two like groups on every carbon
atom so that no evidence of asymmetry could possibly be due to an asym-
metric carbon atom. In the second place, there are the alkyl radicals
on the alpha carbon atom in place of the mobile hydrogen atoms on the
corresponding carbon atom in aceto acetic ester. It was thought that
single hydrogens in this position might seriously hinder or even prevent
the formation of the diazo compound in the beta position.
Two methods for production of the amine from this type of beta
ketonic ester were considered. The first involved the preparation of an
Oxime which should reduce to the desired product. The second method
was to prepare the hydroxy compound by the reduction of the same
beta ketonic ester. Following this reduction of the ketone group to
that of a secondary alcohol, a halogen could be substituted for the hy-
droxyl formed and this halogen in turn might be replaced by the amino
group by means of ammonia under proper conditions. Resolution and
diazotization were the final steps.
Dimethyl aceto-acetic ester was prepared by using dimethyl sulfate
as the methylating agent, since methyl iodide was too expensive to use
in experimentation if a substitute could be found. Although the for-
mer reagent has been used but very little for similar alkylations, Grand-
mougin” used it with success in preparing the desired compound. Di-
ethyl aceto acetic ester, the next homologue in the series, was also pre-
pared and much more satisfactorily than the dimethyl aceto acetic ester
since ethyl bromide was readily available. In the course of the investi-
gation these two beta ketonic esters were used somewhat interchangeably.
It was expected that the results that they would give would either check
each other or one give promise of better results due to slightly different
properties. - -
Attempts were made to convert the dimethyl aceto acetic ester into
the corresponding oxime by the action of both the sulfate and the chloride
9
of hydroxylamine in absolute alcohol. Efforts to isolate the oxime failed.
Due to the very evident loss of this oxime during its unsuccessful isola-
tion, it was considered possible to reduce it at once by metallic sodium
in the original alcohol solution. As the very slight amount of product
obtained from the ether extraction of this solution and the remaining
mother liquors indicated that the main reaction did not form oxime and
amine, this means to obtain the beta amine easily was abandoned, as some
variations in the method proved to be no better. -
Since the oxime of dimethyl aceto acetic ester could not be obtained in
pure form, the oxime of diethyl aceto acetic ester which is crystalline,
was prepared according to Betti.” In spite of low yields, a known
weight of the oxime was a far more satisfactory material to use for re-
duction. As in the case of dimethyl aceto acetic ester, the sodium and
absolute alcohol method was tried. The ether extract of the alcoholic
solution after the unchanged oxime was removed gave only very faint
evidence of the formation of some possible amine hydrochloride when dry
hydrochloric acid gas was passed into it. It could not be found that the
mother liquors had retained the amine. .
As Goldschmidt's* very useful method for reducing an oxime em-
ployed two and a half per cent sodium amalgam, which is a far less ener-
getic reagent than metallic sodium and alcohol, unproductive attempts
were made to procure the amine by its use. -
The production of the amine by first reducing the beta ketone was
considered a greater share of the time as more promising. To avoid
interfering side reactions, the catalytic method was investigated. The
apparatus used in practically all the experiments was that of Skita.”
modified by Lochte, N oyes and Bailey.” ". Although the literature is
filled with many papers in which catalytic reduction is used, there seems
to be no real way of determining just what conditions a new compound
would require. - .
Platinum black was used as a catalyst in the preliminary experiments.
Since the dimethyl aceto acetic ester prepared from dimethyl sulfate
failed to be reduced by hydrogen in the presence of this catalyst, the
conclusion was drawn that some sulfur which would have a poisonous
effect might have been present. For this reason, diethyl aceto acetic
ester was used in all further catalytic experimentation. With the plati-
num black this compound was not reduced in ether or alcohol solution.
A method of preparing very active platinum oxide was developed in
this laboratory at this time by Adams and Vorhees” with considerable
10
success. With this there appeared to be enough reduction to warrant
attempts to substitute bromine for the hydroxyl group. Finally, since
this method was not completely successful and it has been found that
excellent reduction in certain cases took place when the platinum oxide
was not absolutely pure but contained a small amount of impurity in
the form of ferric chloride,” which seemed to catalyze the catalyst, a
trial was made using this means with no better result. -
Platinum in a colloidal form was employed as the last possibility. No
evidence of reduction which was at all satisfactory was obtained. For
this experiment a trial apparatus was used in which the absorption of
the hydrogen was carried out at atmospheric pressure. As this apparatus
had been successfully used by others, there was no reason to believe that
it was faulty. -
When catalytic reduction seemed to be unpromising, nascent hydrogen
formed under various conditions was the alternative. The action of
three and a half to four per cent sodium amalgam on diethyl aceto
acetic ester in an investigation by Schnapp" had resulted in the prepar-
ation of the sodium salt of the beta hydroxy acid. This salt was neu-
tralized by dilute sulfuric acid to give the free acid. No yields or anal-
yses of the free acid before polymerization were given, although later
workers in the same field used, this method to prepare the sodium salt.
In this present investigation, the product of acid ketonic hydrolysis, die-
thyl acetic acid was obtained always, usually in such large amounts that
it was useless to proceed. - - -
Attempts were made to use sodium and alcohol at low temperatures
in the hope that the more rapid reaction would prove to be more success-
ful. Hydrolysis still occurred with both beta ketonic esters. Qualita-
tive results were positive enough here to warrant introducing bromine
into the molecule in place of the beta hydroxyl group in both the acid
and the ester. There was a possibility that a beta bromo compound
might be obtained in pure form. An effort to prepare the beta bromo
ester by Burton” was not successful, due to apparent inability to pre-
serve the molecule intact upon purification by distillation. Using both
halides of phosphorus and the halogen acids, his distilled product was
the diethyl acetic acid derivative. He gave, however, no evidence of
using precautions during the heating of the reaction and distilled his
material under ordinary atmospheric pressure. With this in mind, the
preparation of the bromo compound was carried out more carefully.
A half hour's treatment on the steam bath with fuming hydrobromic
gº 11
acid gave the best results, though distillation of the red oily substance
formed under 12 mm. pressure brought about decomposition.
Therefore, the impure bromo compound, obtained in this manner,
was treated with ammonia with the hope that perhaps the amine could
be isolated. Of the methods tried, anhydrous ammonia passed into
an ethereal solution of the beta bromo ester was probably the best to
obtain an easily isolated product. No definite indication that the amine
was the material precipitated could be gained by analysis.
The latest efforts have been concerned with the dimethyl aceto acetic
ester once more. The beta hydroxy acid was prepared from this accord-
ing to a modified method of Wogrintz” whereby considerable time was
saved. The acid was readily esterfied to an ester which could be dis-
tilled without decomposition.
Finally, experiments have been directed towards the substitution of
a halogen for the beta hydroxy group in the dimethyl aceto acetic ester
derivative. Methods using the free acid, sodium salt of the acid and
ethyl ester that have been carried out with success in other cases, have
been tried. Attempts to isolate the amine salt from the impure ester
containing the halide have not as yet given the desired result.
An explanation of the lack of definite result is of course problematical.
Conditions mean so much in all cases that right ones may not have been
found. Further experimentation might be rewarded by finding out
what they are. Admittedly small amounts of materials were used.
However, it could not be considered really successful to have obtained
a product where the ratio of that product to the initial substance was
that of a very small amount to a large one. Such results would not be
indicative of a main reaction but rather of a chance product.
The fact that the reduction of the dimethyl aceto acetic ester by so-
dium amalgam did take place easily under conditions where the reduction
of diethyl aceto acetic ester was peculiarly impossible to carry out with-
out obtaining the hydrolysis product, may be explained, perhaps rather
tritely, by saying that the heavier radicals on the alpha carbon atom
promote the break between it and the beta carbon. The conditions were
doubtless favorable for the so called acid ketonic hydrolysis of a beta
keto ester in the presence of alkali. Such beta keto esters are gener-
ally considered markedly unstable. Steoric hindrance is another theory
that chemists have used to explain the unexplainable. Here, of Course,
the larger, longer ethyl group might have inhibited the action on the
beta carbon atom. -
12
Finally, it may be repeated in behalf of the continued failure to re-
duce diethyl aceto acetic ester, that the original method followed gave
physical characteristics of the product, but no direct analysis of the free
acid before so called polymerization. The second preparation followed
the first without recording analyses. It is to be wondered if there was
not a considerable impurity in the reduction product as it was first pre-
pared.
III. EXPERIMENTAL WORK.
PREPARATION OF DIMETHYL ACETO ACETIC ESTER :
The method used was essentially that of Grandmougin” which in-
troduces the two alkyl groups separately into the alpha position of aceto
acetic ester. Twenty-three grams of metallic sodium was gradually
added to 150 cc. of absolute alcohol in a flask previously fitted with a
reflux condenser, thermometer and dropping funnel. Occasional shak-
ing took the place of a mechanical stirrer. To the sodium ethylate
formed in this way, 130 grams of ethyl aceto acetic ester and 132 grams
of dimethyl sulfate were added drop by drop. Meanwhile the tempera-
ture was held between 60° and 70°. When all the reagents had been
added, the mixture was stirred or shaken until it was cold and was no
longer alkaline in water, which indicated the completion of the reaction.
After the alcohol was distilled off under diminished pressure at the tem-
perature of the steam bath, the oily product was separated by adding
water in sufficient quantity to dissolve the solid sodium sulfate that had
been formed. As a negligible amount of organic material was found
to remain in the water layer, the oily upper portion was at once removed
and dried over anhydrous sodium sulfate. Vacuum distillation of the
dried material gave a 48% yield of the mono methyl aceto acetic ester,
a product boiling at 68°-72° at 12 mm. By using 72 grams of the mono
methyl aceto acetic ester in exactly the same way, a 48% yield of the
dimethyl aceto acétic ester was obtained. Both yields are 35% lower
than those claimed in the literature. - -
PURIFICATION OF DIMETHYL ACFTO ACETIC ESTER
The purification was necessary after distillation since the boiling
points of the related esters and dimethyl sulfate are so similar.
(CH,),SO, b. p. 188.”
CH,COCH (CHA) COOEt. b. p. 186.8°
CH,COC(CH,),COOEt. b. p. 184.
CH,COCH,COOEt. b. p. 180-181°.
13
Ferric chloride is a delicate indicator for the purity of the product as
it gives a violet color with aceto acetic ester, blue violet with mono-
methyl aceto acetic ester and no color with the desired dialkyl compound.
This is due to the lack of a mobile hydrogen atom in the alpha position
which gives the reactive enolic form in the first two esters. A single
washing with a five per cent solution of sodium hydroxide removed the
unchanged aceto acetic ester and monomethyl ester, using this ferric
chloride test as the proof of its efficiency. Sodium chloride was used
to break up the emulsion at this point which otherwise would have
lengthened the time the ester was in contact with the water. Prolonged
contact might have caused a splitting of the molecule by hydrolysis. The
dimethyl aceto acetic ester, after drying and redistilling, kept without
change.
PREPARATION OF DIETHYL ACETo ACTIC ESTER
The most satisfactory method to follow was that described by Bar-
nett.” Twenty-three grams of metallic sodium were carefully added to
300 cc. of very absolute alcohol contained in a flask which was fitted in
the same way as in the preparation of dimethyl aceto acetic ester. The
sodium ethylate formed was cooled to 40° by surrounding the flask with
ice. One-hundred and thirty grams of aceto acetic ester, and 150 grams
of ethyl bromide were added drop by drop to this cold sodium ethylate.
The whole was heated for three hours on the steam bath which kept the
mixture just below the boiling point. The flask and its contents were
then cooled to 30°. Without change, the product was used in the
place of the aceto acetic ester in a repetition of the above procedure.
When the second three hour period on the steam bath was completed, the
alcohol and any excess bromide were removed by distillation under di-
minished pressure from the steam bath. It should be stated that excess
heating should be avoided throughout, due first to the low boiling point
of the ethyl bromide and second to the tendency of the aceto acetic ester
derivitives to decompose at even moderately high temperatures. After
the distillation, water was added to dissolve the solid matter. The oily
layer was separated, dried over anhydrous sodium sulfate, and if neces-
sary, purified. It was found ordinarily that any fraction boiling above
90° at pressures between 14 and 16 mm. gave no-color with ferric chlor-
ide and was insoluble in base. Using an excellent grade of absolute al-
cohol that was made available by experimental work by Noyes “a 64-
68.9% yield was regularly obtained. Barnett gives 50 to 60%.
14
ATTEMPT to PREPARE THE oxiME FROM DIMETHYL AcETo AcETIC ESTER
The oxime of dimethyl aceto acetic ester was a non isolated inter-
mediate in an investigation carried out by Wallach,” who was interested
not in the oxime but rather in some end products. His method which
in principle was the usual one for the formation of ketoximes, was fol-
lowed closely. A mixture of metallic sodium and absolute alcohol, which
formed sodium ethylate, and hydroxylamine sulfate or chloride in a
stoppered tube was the solution prepared into which the ester was intro-
duced. The sodium ethylate neutralized the acid forming the amine
salt, leaving the hydroxylamine free to react with the ketonic ester. Two
tenths of a mol of ester was used with a slight excess of the hydroxyl-
amine and enough sodium ethylate to neutralize the acid. After shaking
twenty minutes, the tube was allowed to stand for a time varying from
half an hour to twelve hours, while the solid sodium chloride settled out.
The liquid was decanted and distilled under a pressure of 17 mm. at
40° to remove the alcohol. The remaining yellowish oil solidified when
treated with ether, contrary to expectations, considering the properties
of an oxime. This substance contained nitrogen but attempts to separate
the oxime from it by ether and alcohol as solvents failed.
ATTEMPT TO PREPARE BETA AMINo, ALPHA DIMETHYL BUTYRIC ESTER
This compound was unknown, so reduction methods had to be analo-
gous to those already proven successful in preparing known amines from
oximes. The alcoholic solution of the unisolated beta oxime was
treated with six atoms of metallic sodium for every atom of oxime sup-
posed to be present. The product, when poured upon ice, gave a thick
yellow, oily liquid with a spicy odor. No satisfactory test for nitrogen by
the sodium fusion method could be obtained; using a small portion of the
very insignificant amount of material obtained by ether extraction from
the solution on ice. Investigation of the mother liquors which might
have retained the oxime due to the presence of so much alcohol, accom-
plished nothing. A modification of this method of preparing the amine,
in which the alcoholic solution was acidified and dried proved no better.
This should have given a salt of the amine but no portion of the dry
residue could be extracted with alcohol. -
| PREPARATION of THE oxiME FROM DIETHYL ACETo ACETIC ESTER
This was prepared in a pure state by repeating the work of Betti.”
Equivalent weights of dimethyl aceto acetic ester, anhydrous sodium
carbonate and hydroxyl amine hydrochloride were added to 4 or 5 vol-
umes of ordinary alcohol and the mixture was placed on the steam bath
15
to reflux until the original pink color turned to a yellow. This required
about three hours when 3.6 grams of the ester were used. After the
color change, a few cubic centimeters of water were added. This solu-
tion was distilled under ordinary pressure to remove the alcohol until the
original volume was regained. A few cubic centimeters of water were
again added which dissolved the sodium chloride and sodium carbonate.
A thick, oily layer appeared which was dissolved in the least amount of
alcohol that would suffice. To this solution, kept in a beaker surround-
ed by ice, water was added drop by drop until it was very faintly tur-
bid. Standing was said to result in crystal formation. Nevertheless,
energetic stirring of the solution in this state appeared to be more effi-
cient. Time was sometimes saved by heating the solution on the steam
bath before the beaker containing it was set in ice. The sudden fall in
temperature promoted crystallization. Recrystallization from alcohol
and water in the same way gave a white crystalline product. Melting
point 56°. Yield 26%. Betti published no yield. *
A number of slight variations were made in order to increase the
amount of product. Neither excess of hydroxylamine salt and sodium
carbonate, prolonged heating of the original mixture on the steam bath or
change in the order of the addition of reagents had any effect.
ATTEMPTs To REDUCE THE BETA oxiME of DIETHYL AcETo
< . ACETIC ESTER .
A. One gram of the beta oxime was taken for reduction by the the-
oretical amount of metallic sodium in absolute alcohol. Upon comple-
tion of the evolution of hydrogen, two-thirds the amount of glacial acetic
acid required to combine with the sodium ethylate was added at once
to combine with that base. This mixture was poured upon ice while still
alkaline. The attempt was then made to extract the free amine with
cold ether. Extreme care was exercised throughout to keep apparatus
and solution cold, for fear the amine might be volatile.
The ether extract was washed with water, dried over anhydrous so-
dium sulfate and anhydrous hydrochloric acid gas was passed into it.
This should form amine salt which would be insoluble in ether, while
any unchanged oxime which might have been extracted by the ether from
the ice-water solution, would not precipitate. Nothing more than white
fumes appeared. The quantity of sodium was varied up to ten atoms for
one molecule of oxime. Ice was kept around-the tube during reduction
so that the temperature was 10° the entire time. Ice was poured into the
reduction tube but no precautions improved the preparation. The moth-
16
er liquors were acidified, evaporated to dryness, tested for nitrogen and
extracted with alcohol. Nothing significant appeared.
B. In a variation, one gram of oxime in 30 cc. of alcohol was treated
for reduction of the oxime with 160 grams of two and a half per cent
sodium amalgam, according to Goldschmidt's method.” Acetic acid, ap-
proximately 1-normal, was added from time to time to keep the solution
acid. 40° being a temperature recommended, the temperature was main-
tained at that point until the amalgam was liquified. To remove any
unchanged oxime, the acid solution was extracted with ether. 50% was
removed in one case. The average was 15%. Room temperature, and
a temperature lowered to 5° by an ice and salt mixture gave no improve-
ment. When a 90% recovery of the oxime was obtained using an at-
mosphere of carbon dioxide, the method was considered impractical. In
each experiment, the ether extraction from the alkaline solution and the
mother liquors were investigated as in the sodium and alcohol reduction.
This oxime is very evidently difficult to convert to the amino compound.
EXPERIMENTS WITH CATALYTIC REDUCTION
Attempt to prepare beta hydroxy, alpha dimethyl butyric acid ester.
The apparatus used " " " for the reduction of dimethyl aceto acetic
ester by hydrogen under pressure consisted of a prestolite tank, without
filling, which was used as a container for the hydrogen. This was con-
nected by a tube fitted with a valve and pressure guage to a large cylin-
der containing hydrogen. A second tube into the tank was connected to
two valves, one of which was connected to a vacuum pump and the other
to the pressure bottle into which the material to be reduced was poured.
This bottle in which the reaction was to take place was arranged so that
it could be constantly shaken by motor power. By adjustment of the
valves, the whole apparatus could be evacuated and then filled with hy-
drogen. The amount of hydrogen absorbed was read by readings of the
pressure guage. Obviously, variations in temperature and absorption of
hydrogen by the catalyst itself had to be taken into consideration.
Platinum black, a well-known catalyst was tried first. It was pre-
pared as follows:–two grams of chloroplatinic acid, 2 cc. of water, 4
cc. of 40% formaldehyde were combined. To this 6 cc. of 50% sodium
hydroxide were added. The chloroplatinic acid was quickly reduced.
After three hours heating on the steam bath with 50 cc. of water the
liquid around the platinum black had been colorless for some time. After
decanting the liquid, the catalyst was washed until free from chlorides
and dried for use in a vacuum desiccator.
17
*
For reduction, ten grams of dimethyl aceto acetic ester in absolute al-
cohol were added to 0.25 gram of this prepared platinum black which
had been previously shaken with air in the pressure bottle. The appara-
tus was evacuated and hydrogen introduced under pressure. Neither
twenty-nine or forty pounds of pressure for nineteen hours showed ab-
sorption of hydrogen. As Vavon “” had reduced aceto acetic ester with
platinum black in ether solution, the solvent was changed with no better
result. Either the conditions were not suitable or sulfur from the di-
methyl sulfate used in preparing the ketonic ester might have poisoned
the catalyst. The same quantities and conditions for diethyl aceto acetic
ester proved to be no better. Subsequent reductions by any catalytic
method were undertaken with diethyl aceto acetic ester alone.
Using the same apparatus, the platinum black was replaced by plati-
num oxide as the catalyst. This substance was prepared according to
Vorhees and Adams” as follows:—chloroplatinic acid solution was evap-
orated to dryness on the steam bath, One gram of this with 10 grams
of moist sodium nitrate were mixed in a small casserole. When the
mixture was homogeneous, it was heated over a free flame with constant
stirring to prevent spattering. Brown oxides of nitrogen come off at
320° and, when the mixture was fused a brownish powder appeared.
After about ten minutes heating the brown oxides were no longer evolved
and the mixture was cooled. Water was added to dissolve the nitrate.
The platinum oxides were allowed to settle in order to wash by decanta-
tion. After decantation through a filter a few times, the entire pre-
cipitate was transferred to the filter to be washed free of nitrates. The
catalyst was dried in a desiccator and the washings saved for platinum.
The same amounts of catalyst and keto-ester were used for these re-
duction experiments, as before. After reduction the product was frac-
tionated and that portion distilling from 102° to 105° under a pressure
of 14 mm. was used for tests. The Beckmann” test for secondary al-
cohols was positive and taken as evidence that reduction had taken place.
As alcohol itself gives the Beckmann test, ether was used as a solvent
with the same results. The fact that diethyl aceto acetic ester distills at
94°-95° under 16 mm. pressure is favorable evidence for the presence of
some reduction product. . . - -
To try the use of an impure catalyst” one-fifth mol or 30 cc. of die-
thyl aceto ester in 60 cc. of absolute alcohol-with one-thousandth of a
gram mol of platinum as the platinum oxide and one-ten-thousandth of
a gram mol of ferric chloride was introduced into the reaction bottle of
18
the reducing apparatus. The pressure was thirty-five pounds. Experi-
ence had shown that if reduction was to take place, it would start very
definitely during the first hour. No diminution in pressure of the hy-
drogen was observed. Addition of sodium acetate did not further the
reaction. - -
The preparation of colloidal platinum as the last alternative for a
catalyst was modeled after Skita's" method. One gram of gum arabic
was boiled in 20 cc. of water to which hydroxylamine hydrochloride in al-
kaline solution was added. To this, 1 gram of chloroplatinic acid, with
water to dilute to 50 cc. was added last of all. The resulting solution
was filtered but not dialized.
The reduction of the diethyl aceto acetic ester by this colloidal plati-
num as catalyst was carried out in a simple experimental apparatus which
had given instantaneous results with compounds readily reduced cataly-
tically under pressure. The principle was the same as in the Skita appar-
atus with the advantage that smaller amounts could be given a prelim-
inary trial under atmospheric pressure. The apparatus was first evacu-
ated and then filled with hydrogen. The gas was confined at one end
in a gasometer, the base of which was partly filled with water from a
small reservoir. The hydrogen could pass up out of the gasometer and
into the tube where the sample was placed. As the gas was absorbed,
the lessened pressure caused the water to rise in the gasometer. The
number of cubic centimeters of gas disappearing could be read in this
way. The results were most unsatisfactory when 2 cc. samples of ester
were used. No positive Beckmann test could be obtained, so it was not
thought feasible to try colloidal platinum on a large scale. One-tenth to
0.2 gram of platinum was used as a catalyst and the time the ester was
left in contact with the hydrogen varied from one and three quarters to
six hours. Temperature and atmospheric changes were negligible. A
small amount of air in the apparatus gave no better results. * -
REDUCTION OF DIETHYL ACETO. ACETIC ESTER BY SODIUM AMALGAM.
Schnapp's° method of reduction of diethyl aceto acetic ester which
was very indefinite but given in more detail by Jones” was followed.
Ten grams of diethyl aceto acetic ester and two and a half per cent
sodium amalgam containing twice the theoretical amount of sodium re-
quired to reduce this amount of keto ester was added to 50 cc. of alcohol
to which enough water had been added to make it slightly turbid. The
rapid evolution of hydrogen at first produced an appreciable amount of
heat. To keep this heat from promoting hydrolysis, the container was
19
surrounded by a freezing mixture. 1-normal sulfuric acid was added
from time to time to keep the solution from becoming strongly alkaline.
The reaction was apparently complete when no water in soluble oil sepa-
rated out upon further dilution with water. This took about five days.
Before the solution was placed on the steam bath to evaporate, there was
enough water added to dissolve the solid sodium sulfate formed after
complete neutralization and the aqueous layer was decanted from the
mercury. Any oil was separated but it was rarely present. After evapo-
ration, the residue was extracted with 95% alcohol. The sodium salt of
the organic acid was very soluble in this solvent while the inorganic
sodium sulfate scarcely dissolved. Evaporation of the alcoholic extract
and redissolving the residue from this in absolute alcohol gave a 10%
yield of the sodium salt which answered Schnapp's description. Varia-
tions did not increase the yield. By treating the sodium salt with dilute
sulfuric acid, a yellow oil separated out. This was extracted with ether,
dried over anhydrous sodium sulfate and the ether evaporated. The acid
resulting answers Schnapp's description, with the odor of diethylacetic
acid. Since the yield was only 2–4%, a better method for reducing the
diethyl aceto acetic ester was sought.
REDUCTION OF DIETHYL ACETO ACETIC ESTER BY SODIUM AND
ETHYL ALCOHOL.
Twenty grams of diethyl aceto acetic ester were poured into a bottle
fitted with a reflux tube, thermometer and dropping funnel. To this,
500 cc. of absolute alcohol was added. The temperature was kept be-
tween 5° and 15° during the reaction by surrounding the bottle with
ice. After lowering the temperature of the alcohol solution to about
0°, 30 grams of metallic sodium were added in small pieces. Intermit-
tent shaking of the container throughout the four or five hours the
sodium, required for its reaction at that low temperature, kept the solu-
tion from becoming too warm in the neighborhood of the sodium. When
the reaction was complete, the solution was still kept at 20° during
neutralization by 6N-sulfuric acid. The solid sodium sulfate was filtered
off and the alcohol distilled off under diminished pressure. The pro-
cedure followed to obtain the reduction product follows. After the
alcohol had been removed, any unchanged ester or reduced ester appeared
as an oil on top of the water solution. This oil was extracted with ether,
dried and distilled. The larger portion distilled from 105° to 110°
under a pressure of 14 mm. which is higher than the temperature of
20
94° at 14 mm. at which diethyl aceto acetic ester distills. The product
was a yellow oil, insoluble in base and gave a positive Beckmann test.
Yield 10 to 20%. -
This distilled product was used for further work, assuming that it
contained a considerable portion of the desired beta hydroxy alpha diethyl
butyric ester. A sample of this ester with a constant boiling point of
105° at 12 mm. was analyzed. The results indicate that reduction had
not occurred.
Analysis subs. 0.1722 gm. : 0.2208 gm. CO, 0.4057, 0.5155;
H.O, O.1506, 0.1875. -
Calc. for Clo HisOs : C, 64.46 H, 9.74.
Found C, 64.25 H, 9.78
63.66 9.52
After extraction of the ester the sodium salt of the acid was recovered
from the aqueous solution by acidification with sulfuric acid and extrac-
tion four to five times with ether. After drying the ether extract over
anhydrous sodium sulfate and evaporation of the ether, the yellow liquid
residue was dissolved in sodium bicarbonate, to remove any ester which
might be present. A small amount of ester was removed by ether, the
sodium bicarbonate solution acidified by dilute sulfuric acid and ex-
tracted with ether. The pure acid was obtained from the ether as before.
It was a clear yellow liquid, strongly acidic, with a sharp, penetrating
odor, apparently like the acid of Schnapp. It gave the Beckmann test
for a secondary alcohol group. To ascertain whether it was the product
desired, neutral equivalents were taken of the samples of acids as they
were prepared. This was convincing evidence that the acid if formed
was very readily hydrolyzed to diethyl acetic acid. Yield 20 to 40%.
Example subs., (hydroxy acid) 0.2222: 16.8 cc. of 0.97605N
NaOH - - -
Neutral equivalent calc. for Cs H.O., 160.12:
for Cs H.O. 116.09:
Found 127.8
No variations improved the method. The principal ones tried were,
use of a 50% solution of ether in alcohol instead of pure alcohol which
kept the metallic sodium at the bottom of the bottle and a freezing mix-
ture which lowered the temperature to -5° during the entire period.
21
ATTEMPT TO SUBSTITUTE BROMINE FOR THE HYDROXYL GROUP OF A.
REDUCED DERIVITIVE OF DIETHYL ACETO ACETIC ESTER.
Preliminary experiments to determine qualitatively whether bromina-
tion by 34% or 48% hydrobromic acid of the sodium salt or free hy-
droxyacid obtained by reduction, in the cold for times varying from one
day to seven, were not significant. One gram samples with 5 cc. of the
acid were used. The acid solution was diluted and the product was ex-
tracted with ether. The tests for the presence of a halide were first
made by a sodium fusion. The best result obtained was from 1 gram
of the hydroxy acid with 5 cc. of fuming hydrobromic acid heated on
the steam bath for a half hour. The solution was diluted, extracted with
ether; the ether solution was washed free of hydrobromic acid and dried
over anhydrous sodium sulfate. After distilling off the ether in vacuum,
the red liquid obtained was analyzed quantitatively for bromine by
means of the method using metallic sodium and absolute alcohol.” The
analysis had to be made from the substance without purification by dis-
tillation as even distillation under diminished pressure căused immediate
decomposition. -
Subst. 1.895 gms. gives .5368gm Ag Br. Per cent possible bromo COIIl-
pound present C, H, O, Br, 100.00. Found 26.4.
To make the silver salt of the beta bromo acid, one gram of it was
treated with excess solid calcium carbonate. When the evolution of car-
bon dioxide ceased, the thick mixture was not wholly soluble in 10 cc. of
water which was added. After filtering the filtrate gave a white, pasty
silver salt by the addition of silver nitrate, that could be dried only in the
air on a porous plate. A very slight amount of heat and even long con-
tinued standing in air darkened it. The salt was completely soluble in
dilute nitric acid. Analysis of samples of this salt for both silver and
bromise indicated that the bromo acid had been formed. To do this,
the amount of silver in a weighed sample was determined in the usual
manner by dissolving the salt in very dilute nitric acid and precipitating
the metal by the exact amount of dilute hydrochloric acid. The precipi-
tate of silver chloride was filtered through a weighed Gooch crucible,
dried and weighed. The filtrate containing the free beta bromo acid
was heated on the steam bath for at least twelve-hours with concentrated
aqueous ammonia. The solution upon acidification with nitric acid and
addition of silver nitrate solution gave a precipitate of silver bromide
which was treated as in the case of silver chloride. -
22
1. Subst. (Ag salt of beta bromo acid) 0.1309 : 0.0693 AgC1,
0.0576 0.0309
Ag in Cs H.O.BrAg 32.7 Found 33.0, 33.2
II. Subst. (Ag salt of beta bromo acid) 0.0586gm Ag C1, 0.0274
% Ag calc. 32.7 Found 35.1
wgt.AgBr 0.01459 Br calc. 24.08 Found 10.4
wgt.Br. 100 0.0254gm Br acid -
24.8
0.0586 gm–0.0254gm = 0.0332gm OH acid -, .
Calc. wet Ag that should be precipitated from both 0.0289gm.
Found 0.0274gm. . -
Five grams of the hydroxyester instead of the acid were heated with
occasional shaking with 20 cc. of fuming hydrobromic acid gas for one
hour on the steam bath. The substance was diluted, extracted with
ether and the ether extract washed with water and dried as usual. The
residue was the red oil that was insoluble in water. The yield was 4
grams of impure product. This could not be successfully distilled. The
percentage bromination indicated by 0.52 gram of the unpurified ma-
terial was 32.4 as determined by heating with ammonia for twenty-four
hours to release the halogen which was precipitated as silver bromide.
The residue from the filtrate was examined for the amine. Nitrogen
was present. Sodium hydroxide added to that portion soluble in alcohol
gave the distinct odor of ammonia and no oily amine.
ATTEMPT TO PREPARE THE BETA AMINE
A. The first method used to prepare the amine was that of treating
the impure bromo acid in slightly larger amounts than the samples used
to determine the percentage of bromination with aqueous ammonia. One
and nine tenths gram of the beta bromo acid was heated with 20 cc. of
concentrated aqueous ammonia on the steambath for four hours. The
excess ammonia was then driven off by evaporation until there was
neither an odor of ammonia nor was moistened red litmus changed to
blue in the vapors from the flask. The product formed should be the
ammonium salt of the amine hydrobromide. To remove the ammonium
group, lead oxide was added and the whole boiled with about thirty
cubic centimeters of water, until no more ammonia could be detected as -
23.
before. The lead was precipitated from the solution by saturating it
with hydrogen sulfide. After filtering off the solid lead sulfide the
filtrate was evaporated to dryness and extracted with ether. Since the
residue from the ether extract was scarcely visible, very little unchanged
substance could have been present at this point. The remainder of the
material, insoluble in ether, was soluble in 95% ethyl alcohol and water.
Fusion with soda lime gave a qualitative test for nitrogen. The sub-
stance gave a yellow precipitate with chloroplatinic acid which was solu-
ble in very slight excess of water. Since only 0.0100 grams of the
chloroplatinate could be obtained for the sample and the analysis for
platinum was half way between the percentage of platinum in the amine
and ammonium chloroplatinate, the evidence was not conclusive. Sub-
sequent experimentation gave no larger yields. -
B. The second general method tried was to treat the beta bromo
ester with dry ammonia. The amount of impure bromo ester obtained
from 3 grams of beta hydroxy diethyl butyric ethyl ester gave a white
precipitate of 0.3582 grams. This white crystalline substance retained
moisture, yet could not be dried in an oven at 50°. The material was,
therefore, allowed to dry on a porous plate in a desiccator. This sub-
stance which should be the amine hydrobromide of the ester gave a per-
centage of bromine much too low and very variable. It gave the odor
of ammonia with no separation of oil upon adding sodium hydroxide.
Substance 0.0199 : 0.0445 Ag Br 0.01.16, 0.0224
Calc, Br for C, H, O, NBr 29.9 Found 24.8, 21.3
PREPARATION OF BETA HYDROxY ALPHA DIMETHYL BUTYRIC ACID.
The method followed was essentially that of Wogrintz.” Twenty
grams of dimethyl aceto acetic ester were added to a solution of 50 cc.
of 95% alcohol and 70 cc. of distilled water. This gave a liquid with
slight turbidity. Four hundred grams of 4% sodium amalgam were
added from time to time for reduction. In order to avoid any decom-
position that a rise of temperature might favor, the container was sur-
rounded with ice until the hydrogen was evolving evenly. After the
initial increased evolution of hydrogen ceased, the reaction could be con-
tinuously run quite satisfactorily at room temperature. During the en-
tire operation the mixture was stirred mechanically and carbon dioxide
was passed through the solution. With the consequent formation of
sodium carbonate, this salt which was not readily soluble in this aqueous
24
alcoholic mixture, gradually formed a thick layer over the amalgam.
Water was added in 25 cc. amounts to partially dissolve the salt which
trapped the hydrogen. This caused no separation of liquids. During
the latter part of the reduction the amalgam itself could be stirred.
This mechanical stirring lessens the time from two to two and a half
days in which the reduction takes place.
When the amalgam was completely liquified, the solid salt formed
was filtered off from the liquid. Excess water was added to separate
any unchanged keto ester. The oil was separated and the aqueous solu-
tion containing the sodium salt of the reduced compound was evaporated
on the steam bath. Since the sodium salt is crystallized with difficulty,
it appeared with some solid sodium carbonate as a viscous liquid when
the water had evaporated. Extraction of this residue with 95% alcohol,
evaporation of the alcoholic filtrate, followed by reëxtraction with abso-
lute alcohol gave the characteristic, colorless, difficultly crystallizable so-
dium salt. The salt was dissolved in the least amount of water neces-
sary. Dilute sulfuric acid was added in excess to acidify. As the acid
was extremely soluble in water, extraction of the sulfuric acid solution
with ether ten times is profitable. The ether extract was dried and dis-
tilled in the usual manner. The free acid distilled at 144° under a
pressure of 14 mm. It was a colorless, viscous liquid, soluble in ether,
alcohol and water. Yield 60%.
Subs., (beta hydroxyacid) 0.3785; 29.9 cc. Of 0.10331 N-NaOH
Neutral equivalent calc. for C.H.O., 132. Found 130.1
A further experiment to see if the time for reduction could be lessened
resulted in a loss. Twenty grams of the dimethyl aceto acetic ester were
taken as before but the reduction was continued for just twenty-four
hours. The unchanged oil was separated at the end of that time and
the sodium salt of the hydroxy acid was obtained from the aqueous
solution as in the method previously given. To the oil, enough more
dimethyl aceto acetic ester was added to make 20 grams. This was
treated for a second twenty-four hour period. The combined sodium
salts of three such periods were converted to the acid, giving a yield of
36% when the exact amount of original ester which had disappeared
was taken into consideration.
25
PREPARATION OF THE ETHYL ESTER OF BETA HYDROXY ALPHA
DIMETHYL BUTYRIC ACID
This ester has been prepared by Courtot.” To 5 grams of the free
acid in 40 cc. of absolute alcohol, 1.5 grams of dry hydrogen were
added. The solution was refluxed on the steam bath for four hours.
After the alcohol was distilled off under diminished pressure, the organic
residue was dissolved in a concentrated solution of sodium bicarbonate
to remove any unchanged free acid as the sodium salt. The bicarbon-
ate solution was extracted with ether. The extract was dried as usual
and distilled. The product is a clear colorless liquid without a charac-
teristic odor, distilling at 87° under a pressure of 14 mm. Yield 66.6%.
No yield was given for a preparation by esterification. -
AN ATTEMPT TO PREPARE A BETA BROMO COMPOUND FROM A
DIMETHYL ACETo AcETIC ESTER REDUCTION PRODUCT.
One and five tenths gram of the hydroxy acid in 10 cc. of ethyl alcohol
concentrated aqueous ammonia on the steambath for four hours. The
in the hope that the acid would not only be esterified but also that
bromine would be introduced in the beta position in place of the hydroxyl
group. The resulting product obtained after evaporation of the alcohol
was insoluble in water and concentrated sodium bicarbonate solution.
The impure product was 0.6 gram. Analysis indicated that not more
than 12% of the hydroxy acid had been converted into the bromo com-
pound. To improve the yields other substances were tried.
Saturation of an absolute alcohol solution of one gram of the sodium
salt of the hydroxy acid with anhydrous hydrogen bromide gave no
better result. - w . . . - -
One gram of acid was treated with 10 cc. of fuming hydrobromic acid
in the cold. A perfectly homogeneous liquid resulted at once without
heat from a reaction. After twenty-eight hours the solution was diluted
and extracted with ether. From the ether extract, treated as usual, 0.2
gram of substance was obtained. This was readily explained when
evaporation of the washings and aqueous solution gave 0.9 gram of a
viscous product resembling the original hydroxy acid. This residue could
not be recovered, as the hydroxy acid, due to decomposition upon dis-
tillation. Heating a 1-gram sample of the hydroxy acid with fuming
hydrobromic acid for a half hour on the steam bath gave no better re-
sult. Bromine could not be introduced by hydrobromic acid in a practi-
cal manner. - . .
26
To use phosphorus and bromine, 2.6 grams of the hydroxy acid and
0.7 gram of red phosphorus were added to 30 cc. of anhydrous ether in
a bottle fitted with a dropping funnel, thermometer and outlet tube to
which a drying tube was attached. Five and two tenths grams of bro-
mine were slowly added from the dropping funnel. The temperature
was maintained at 0° as long as the bromine was being added. The red
phosphorus gradually disappeared due to the formation of phosphorus
tribromide. After the last portion of bromine had been added, the solu-
tion was allowed to stand at 20° for a half hour. The ether was dis-
tilled off under diminished pressure without heat, leaving a deep red,
heavy liquid which should have been the bromo acid bromide of the beta
bromo acid. Hydrolysis of this liquid by the use of ice to remove the
bromine in the molecule which formed the acid bromide was continued
for a half hour. The red color of the water insoluble oil gradually
faded to yellow. This oil was extracted with ether, the ether extract
being washed and dried in the usual manner. The residue, weighing
0.8 gram could not be crystallized from varying amounts of water and
alcohol. The evaporation of the ether washings gave 2.1 grams of the
viscous residue. Lengthening the time in which the bromination was to
take place, to six, twelve and twenty-four hours, failed to increase the
yield. In all but the first case, a qualitative test to roughly estimate the
completeness of the bromine substitution gave evidence of incompleteness
of the reaction. - -
Following the trials with the phosphorus and bromine in anhydrous
ether, 1.3 grams of the hydroxy acid was treated with 4 grams of liquid
phosphorus tribromide. The mixture was allowed to stand two hours
and then treated as in the previous cases. The result was even less
satisfactory. - - -
AN ATTEMPT TO PREPARE A BETA CHLORO COMPOUND FROM A
DIMETHYL ACETo ACETIC ESTER REDUCTION PRODUCT
One gram of the hydroxy acid plus 3.5 grams of phosphorus penta-
chloride in powdered form were allowed to react by first dissolving the
acid in 10 cc. of anhydrous ether and carefully adding the pentachloride.
There was an instantaneous reaction as indicated by a production of heat
which caused the ether to boil. In the course of fifteen minutes all the
phosphorus pentachloride had disappeared. The phosphorus oxychloride
formed by this reaction and the ether were distilled off under diminished
pressure at the lowest temperature possible by using an oil bath. Esteri-
27
fication without isolation was attempted by adding to the residue 10 cc.
of absolute alcohol and refluxing on the steam bath for two and a half
hours. The product recovered after removal of the alcohol and extrac-
tion with ether contained very little halogen. To determine the amount
of esterification this residue was treated with sodium bicarbonate which
would dissolve the unchanged acid, leaving the ester. Seventy-five per
cent of the acid was recovered as the sodium salt.
One gram of the hydroxy ester instead of the acid was treated in the
same way with 1.5 grams of phosphorus pentachloride, with the excep-
tion that the resulting product was hydrolyzed with ice. The odors of
phosphorus pentachloride and phosphorus oxychloride were dispelled
when the hydrolysis was complete. The insoluble product was extracted
with ether and the ether extract washed free of chlorides, dried and evap-
orated. The product which was 0.8 gram was a liquid with an odor
resembling peppermint. It contained some unsaturated compound since
it decolorized potassium permanganate. It also contained considerable
chlorine as demonstrated by addition of silver nitrate, boiling with am-
monia and acidification with nitric acid. This behavior corresponded to
the results of Bouveault.” . - -
Preliminary experiments with phosphorus and iodine to introduce the
halide were not encouraging. Replacement of the chlorine by amino
group by the passage of anhydrous ammonia into anhydrous ether solu-
tion of the beta chloro compound or moist ammonia introduced the same
way gave unsatisfactory results. The ether solution from this procedure
gave a residue from which halogen could still be obtained free by boiling
with aqueous ammonia. Some definite method which would replace
the halogen quantitatively remains to be established.
IV. SUMMARY.
1. Diethyl aceto acetic ester was prepared by using ethyl bromide
as alkylating agent. Yield 68.9%. Reported yield 50-60%.
2. The oxime from diethyl aceto acetic ester was prepared with a
yield of 20%. No yield was recorded.
3. The reduction of diethyl aceto acetic ester by hydrogen under
pressure in the presence of various catalysts was unsuccessful.
4. The attempts to reduce diethyl aceto acetic ester by known meth-
ods, using sodium amalgam, seem to indicate that the original investiga-
tors had a large percentage of impurity in their product. -
28
5. The reduction of dimethyl aceto acetic ester by sodium amalgam
was repeated. Yield 60%. Esterification of the beta hydroxy acid gave
a yield of 66.6%. No yield was recorded for the esterification.
6. The introduction of a halogen in place of the beta hydroxyl
group of either the dimethyl or diethyl aceto acetic ester derivative gave
no product that could be distilled. Qualitative and quantitative evi-
dence of the amount of halogen introduced indicated that a reaction had
taken place. -
7. The preparation and isolation of a beta amine from the impure
beta halogen derivative was unsuccessful.
29
2.
i
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
BIBLIOGRAPHY
Griess, Ann; 106, 123 (1858).
Schiff and Meissen, Gazz. chim. ital; 11 171 (1881).
Schiff, Ber; 14, 1375 (1881). -
Curtius, Ber; 16, 2230 (1883).
Marvel and Noyes, J. Am. Chem. Soc; 42, 2259 (1920).
Chiles and Noyes, J. Am. Chem. Soc; 44, 1798 (1922).
Curtius, J. Prakt. Chem; [2] 38, 394 (1888).
Angeli, Atti. accad. Lincei; 16, II, 790 (1907); 20, I, 626 footnote (1911).
Thiele, Ber; 44, 2522 (1911).
Lavoisier, Traite elementaire de Chimie Vol. I, p. 123 (1789).
Davy, Phil. Trans. 97, 1 (1807). .
Berzelius, Schweigger's Jour. 8, 119 (1812); Ann; 24, 158 (1837).
Dumas, Ann; 33, 259 (1840).
van't Hoff, Z. Physik. Chem; 16, 411. (1895).
Noyes and Lyons, J. Am. Chem. Soc; 23, 460 (1901).
Stieglitz, J. Am. Chem. Soc; 23, 797 (1901). - -
Alf Sinding Larsen, Z. Elektrochem; 6, 112 (1900); 8, 500 (1902).
Walden, Z. Physik, Chem; 43, 413 (1903). - ~
Wislicinus, Z. Elektrochem; 8, 504 (1904).
Thomson, J. J., Phil. Mag; (6) 7, 237 (1904).
Thomson, J. J., Corpuscular Theory of Matter, p. 130 (1907).
Abegg, Z. Anorg, Chem; 39, 342 (1904).
Ramsay, J. Chem. Soc; 93, 774 (1908).
Falk and Nelson, School of Mines Quart; 30, 179 (1909); J. Am. Chem. Soc;
32, 1637 (1910); 33, 1140 (1911); 35, 1810 (1913); 36, 209 (1914); 37, 274,
1732 (1915); J. Prakt Chem. N. F.; (2) 88, 97 (1913); 8th Internal. Cong.
Applied Science; 6, 212 (1912); Pro. Am. Phil. Soc; 53, 25 (1914).
Fry, E. Physik, Chem.; 76, 385, 398, 591 (1910); 80, 29 (1912); 82, 665
(1913); 90, 458 (1915). J. Am. Chem. Soc; 34, 664 (1912); 36, 248,
262, 1035 (1914); 37, 855, 2386 (1915); 38, 1323 (1916); 39, 1688 (1917).
Jones, L. W., J. Am. Chem. Soc; 36, 1288 (1914); 39, 674 (1917); Science,
46, 493 (1917); Am. Chem. Jour; 48, 25 (1912); 50, 414 (1913).
Stieglitz, J. Am. Chem. Soc; 44, 1293 (1922); Am. Chem. Jour; 29, 49
(1903); Ber; 43,782 (1910).
Noyes, J. Am. Chem. Soc; 35, 767 (1913); -
Noyes and Haw, J. Am. Chem. Soc; 42, 2167 (1920).
Noyes, J. Am. Chem. Soc; 42, 2173 (1920).
Noyes and Coleman, J. Am. Chem. Soc; 43,2211 (1921).
Noyes and Wilson, J. Am. Chem. Soc; 44, 1630 (1922).
Noyes and Hibben, J. Am. Chem. Soc; 45, 1356 (1923).
30
29.
30.
31.
32.
33.
• 34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65. .
66.
67.
68.
69.
70.
71.
72.
Staudinger and Kupfer, Ber; 44, 2197 (1911).
Hantzsch and Lifschitz, Ber; 45, 3022 (1912).
Forster and Cardwell, J. Chem. Soc; 103, 861 (1913).
Darapsky and Prabhakar, Ber; 45, 1654, 2617 (1912).
Staudinger and Kupfer, Ber; 45, 501 (1912). -
Staudinger and others, Ber; 49, 1884-1894 (1916).
Staudinger, Helvetica Chim. Acta; 5, 87 (1922).
Langmuir, J. Am. Chem. Soc; 41, 1546 (1919); 42, 274 (1920).
Fischer, Ann; 381, 123 (1911).
Levene and LaForge, J. Biol. Chem; 21, 345 (1915).
Levene, J. Biol. Chem; 36, 89 (1918).
Levene and Mikeska, J. Biol. Chem; 45, 593 (1921).
Levene and Mikeska, J. Biol. Chem; 52, 485 (1922).
Levene, J. Biol. Chem; 54, 101 (1922).
Levene, J. Biol. Chem; 53, 449 (1922).
Levene, J. Biol. Chem; 54, 809 (1922). .
Staudinger, Gaule, and Siegwart, Helvetica Chim. Acta; 4, 212 (1921).
Curtius, Ber; 17, 959 (1884). - -
Curtius and Muller, Ber; 37, 1261 (1904); Angeli Ber; 37, 2080 (1904).
Fischer, Ber; 34, 433 (1901). -
Erlenmeyer Jr., Ann; 275, 18 (1893); 307, 138 (1899).
von Peckmann, Ber; 27, 1888 (1894). -
Staudinger, Ber; 44, 2198 (1911). -
Grandmougin, Havas and Guyot, Chem. Zeit; 87, 812 (1913).
Betti, Gazz. Chim. ital; 28, I, 274 (1898).
Goldschmidt, Ber; 19, 3232 (1886); 20, 492 (1887).
Skita and Meyer, Ber; 45, 3589 (1912). * -
Lochte, Noyes and Bailey, J. Am. Chem. Soc; 43, 2601 (1921).
Diagram Univ. of Illinois Bull. Vol. 20, No. 8, p. 62.
Vorhees and Adams, J. Am. Chem. Soc; 44, 1397 (1922).
Carothers and Adams, J. Am. Chem. Soc; 45, 1071 (1923).
Schnapp, Ann; 201, 62 (1880).
Burton, Am. Chem. Jour; 3, 395 (1881-82).
Wogrintz, Monatsh; 24, 248 (1903).
Barnett, The Preparation of Organic Compounds p. 121, (1920).
Noyes, J. Am. Chem. Soc; 45, 857 (1923). -
Wallach, Ann; 248, 170 (1888). . . .
Vavon, Compt. Rend; 155, 286 (1912). • -
Beckmann, Ann; 250, 325 (1889); Noyes, Organic Chemistry for the Labora-
tory, p. 86 (1916). . -
Skita, Ber; 42, 1630 (1909). . - -
Jones, Ann; 226, 287 (1884). -
Stepanow, Ber; 39, 4056 (1906); Noyes, Organic Chemistry for the Labora-
tory, p. 23 (1916). « - -
Courtot, Bull. Soc. Chim; Sec. 3, 35, 111 (1906).
Bouveault, Bull. Soc. Chim; Sec. 3, 21, 1062 (1899).
31
VITA.
The writer was born in Rochester, New Hampshire, December 20,
1896. She entered Vassar College in the fall of 1914 and graduated
with the degree of Bachelor of Arts in 1918.
She was appointed assistant in the department of chemistry of Goucher
College for the school year 1918-1919. This appointment was renewed
and the position held for the following year 1919-1920.
She began her graduate work at the University of Illinois in the fall
of 1920. She received the degree of Master of Arts in 1921. While at
the University of Illinois, she held the following appointment:-Scholar
in Chemistry 1920-1921. -
ACKNOWLEDGEMENT.
The writer is glad to have the opportunity to express her sincere
appreciation for the guidance of Dr. William Albert Noyes who sug-
gested this problem and under whose direction the work was carried out.
32
ºffilſ of McHGAN
|||||||||||
3901s05068.36%
|
|
|
/