close

Вход

Забыли?

вход по аккаунту

?

I. THE REDUCING ACTION OF GRIGNARD REAGENTS. II. STUDIES ON NEOPENTYL ALCOHOL

код для вставкиСкачать
The Pennsylvania State College
The Graduate School
Department of Chemistry
I. The Reducing Action of Grignard Reagents
II. Studies on Neopentyl Alcohol
A Thesis
hy
John D. Zech
Submitted in partial fulfillment of
the requirements for the degree of
Doctor of Philosophy in Chemistry
June, 1940
Dept, of Chemistry
Head of the Dept.
ACKNOWLEDGMENT
The author wishes to express his most
sincere thanks and appreciation to Dr. Frank C.
Whitmore who directed this research for his
kindly interest and suggestions.
Sincere thanks
are also extended to Dr. George H. Fleming for
his helpful advice, and to Dr. R. V. MeGrew who
kindly furnished some of the chemicals required
in the course of these investigations.
TABLE OF CONTENTS
Part I
Page
I. introduction . .
...........................
]_
II. Historical...................................
3
III. D i s c u s s i o n .............................
7 '
A. Reaction of n-butylmagnesium bromide with
2 -hexanone.......................
7
B. Reaction of n-butylmagnesium bromide with
ethyl chloroacetate .....................
7
C. Reaction of t-butylmagnesium chloride with
ethyl chloroacetate .....................
9
D. Reaction of n-butylmagnesium bromide with
ethyl trichloroacetate...................
9
E. Reaction of t-butylmagnesium chloride with
ethyl trichloroacetate...................
10
F. Reaction of t-butylmagnesium chloride with
benzaldehyde.............................
11
G-. Reaction of t-butylmagnesium chloride with
benzil....................................
12
Iv . Experimental..................................
14
A. Addition of hexanone-2 to n-butyl magnesium
b r o m i d e ..................................
14
B. Addition of ethyl chloroacetate to n-butyl
magnesium b r o m i d e ............
16
C. Addition of ethyl chloroacetate to t-butyl
magnesium chloride.......................
20
Page
D. Addition of ethyl trichloroacetate to
n-butyl magnesium bromide................
25
E. Addition of ethyl trichloroacetate to
t-butyl magnesium chloride ..............
29
p. Addition of benzaldehyde to t-butyl
magnesium c h l o r i d e .......................
31
G. Addition of benzil to t-butyl magnesium
c h l o r i d e ..................................
H. Description of the fractionating columns
34
36
V. Summary..........................................
37
VI. Bibliography....................................
39
Part II
I. Introduction . . . . . ........................
40
II. Historical......................................
41
III. D i s c u s s i o n ....................................
51
A. Preparation of neopentyl alcohol . . . .
51
B. Attempted preparation of neopentyl sulfate
52
C. Reaction of neopentyl alcohol with thionyl
chloride in the presence of anhydrous
potassium carbonate and chloroform . . .
53
D. Dehydration of neopentyl alcohol over
aluminum o x i d e ...........................
53
E. Dehydration of t-amyl. alcohol over
aluminum o x i d e ...........................
58
F. Dehydration .of isoamyl alcohol over
aluminum o x i d e ...........................
59
Page
G. Isomerization of isopropylethylene over
60
aluminum oxide.........................
H. Isomerization of trimethylethylene over
60
aluminum oxide.........................
I. The action of gaseous hydrogen bromide
on boiling neopentyl alcohol..........
61
J. The action of phosphorus tribromide on
neopentyl a l c o h o l ..................
.
63
K. The action of constant boiling hydrobromic acid on neopentyl alcohol.
...
64
L. The action of a solution containing
hydrobromic acid,
sulfuric acid, and
sodium bromide on neopentyl alcohol . .
65
M. The stabilitj' of neopentyl bromide to a
solution containing hydrobromic and
sulfuric acids....................
65'
N. The stability of neopentyl stearate to
lieat...............................
66
0. Treatment of neopentyl alcohol with hot
concentrated sulfuric acid........
66
P. The effect of heat on a mixture of neo­
pentyl alcohol and sodium neopentylate.
17. E x p e r i m e n t a l ...........................
A. Preparation of neopentyl alcohol.
67
69
...
(1 ) From t-butyl magnesium chloride and
methjdL f o r m a t e ..............
69
(2) From t-butyl magnesium chloride and
69
Page
paraformaldehyde...................... 70
B. Attempted preparation of neopentyl sulfate 71
(1 ) preparation of neopentyl sulfite. .
71
(2) Treatment of neopentyl sulfite
with sulfuryl chloride............... 73
C. Treatment of neopentyl alcohol with
thionyl chloride in the presence of
anhydrous potassium carbonate and
chloroform.................................. 75
D. Dehydration of neopentyl alcohol over
aluminum oxide.............................. 77
(1) Description ofthe apparatus.
.. .
77
\2 ) Preparation of the aluminumoxide
catalyst.............................. 79
(3) Trial dehydration of n-butyl alcohol
over aluminum o x i d e ................. 79
(4) Dehydration of
neopentylalcohol.
.
8u
S. Dehydration of t-emyl alcohol over
aluminum oxide.............................. 8 8
P. Dehydration of isoamyl alcohol over
aluminum oxide.............................. 92
G. Isomerization of isopropylethylene over
aluminum oxide.
..........................96
H. Isomerization of trimethylethylene over
aluminum oxide.............................. 96
I. Action of gaseous hydrogen bromide on
boiling neopentyl alcohol ...............
98
Page
J. Action of phosphorus tribromide on
neopentyl alcohol.....................
104
K. Action of constant boiling hydrobromic
acid on neopentyl alcohol.............
107
L. A.ction of a solution containing hydro­
bromic acid, sulfuric acid, and sodium
bromide on neopentyl alcohol ........
M. Stability of neopentyl bromide
109
to a
mixture of hydrobromic and sulfuric acidslll
N. Stability of neopentyl stearate to heat
il) Preparation of neopentyl stearate
112
112
(2) Stability of neopentyl stearate
to h e a t .
113
0. Treatment of neopentyl alcohol with hot
.
concentrated sulfuric a c i d
114
P. The effect of heat on a mixture of neo­
pentyl alcohol and sodium neopentylate
115
Q,. Description of the fractionating columns 116
V. Summary.............
117
71. Bibliography......
119
PART I
The Reducing Action of Grignard Reagents
INTRODUCTION
The Grignard reagent has been one of the most useful
reagents available to the organic chemist for the synthesis
of organic compounds by virtue of its ability to add to the
carbonyl group.
This reaction of the Grignard reagent, while
it is the most important one, is sometimes accompanied by one
or more other reactions such as condensation, enolization
and reduction.
In some cases one or more of these reactions
may take place to a much greater extent than addition to the
carbonyl group.
Of these other reactions, reduction has long been under
investigation.
Cases are known where reduction of the car­
bonyl group is practically quantitative, no addition taking
place at all.
There are also many cases which fall between
the two extremes of complete addition to and complete reduc­
tion of the carbonyl group.
Much work has been done on the
reducing action of Grignard reagents in this and other labo­
ratories.
Theories have been proposed to explain this re­
ducing property of the Grignard reagent.
Most of these
i
theories have been attempts to correlate chemical structure,
both of the Grignard reagent and of the carbonyl compound,
with the amount of reduction obtained.
While many of these theories have been useful and have
many facts to support them, none is completely satisfactory.
They have been unsatisfactory because they were based on in­
sufficient evidence, evidence which did not include all of
2
the factors that determine the amount of addition and re­
duction that will take place in any reaction.
Only by the discovery of all of these factors can one
hope to find the answer to the question "What is the mechan­
ism of reduction by Grignard reagents?"
This investigation
was undertaken, not in the expectation of answering this ques­
tion, but in the hope of finding new facts which are needed
for a satisfactory solution of the problem.
HISTORICAL
The literature on reduction by Grignard reagents is
voluminous and spreads over a period of about 40 years.
Due to the extreme interest taken in this field of organic
chemistry in the last ten years, this literature has been
reviewed very adequately by ICharasch
by workers in this laboratory.
etc., as well as
Any such review here vould
only be needless repetition and will therefore not be
attempted.
The literature reviewed here will be confined
to reactions which are closely related to the reactions in­
vestigated in this research.
Since a large part of the Grignard reduction work
studied in this thesis dealt with esters of the chloroacetic
acids, it is in order to review the literature on subjects
closely related to it, namely the action of Grignard reagents
on halogen acids, esters, aldehydes, etc.
The reaction of chloral with Grignard reagents has been
the subject of numerous investigations reported in the litera­
ture.
Herbert (2 ) found that the chlorine atoms of chloral
are not removed with Grignard reagents.
He reported the re­
action as taking place in the following manner
CC1 3 CH0
+
RMgX
— ---
>
CCI 3 CHOHR
This is in agreement with the findings of Henry
(3 )
who
found methyl Grignard to add to chloral according to the
equation
GCI 3 GHO
+
MeMgX
■>
CClgCHOHMe
Savariau ^4 ) reported a similar addition to chloral
as represented by the equation
Me
Me
chloral
CC1,CH0HMe
Me
Howard (5 ,0 ) prepared a series of trichloromethyl sec.
alcohols by the reaction of chloral with ethyl, propyl,
isopropyl, butyl, and benzyl Grignard reagents in respective
yields of 32,22,33,26, and 18$.
No other products were re­
ported.
Dean and Wolf l7 ) reported rather unusual products from
the reaction of beta-phenylethyl, gamma-phenylpropyl, and
delta-phenylbutyl magnesium bromides with equimolecular quan­
tities of chloral.
With beta-phenylethylmagnesium bromide
they obtained 32-40$ of trichloroethanol, 2-4$ of 1,4diphenyl butane and 2.2-4.3$ of beta-phenylethyl bromide to­
gether with a large amount of styrene..
With gamma-phenyl-
propylmagnesiurn bromide they obtained 32-48$ of trichloro­
ethanol, 5.2-12$ of 1,6-diphenyl hexane and large quantities
of 3-phenyl propene-1, while delta-phenylbutylmagnesium bro­
mide gave 43-48$ of trichloroethanol, 4.4$ of 1,8-diphenyl
octane and large amounts of 4-phenyl butene-1.
can be represented by the equation
This reaction
5
CC13 CH0
+
RCH 2 CH 2 M g X — ► GC1 3 CH 2 0H
+
RCH=CH 2 + (RCH2 CH 2 )2
Halogen acids and derivatives of halogen acids have also
been treated with Grignard reagents to give a variety of prod­
ucts.
Stolle (g) treated bromoacetic ester with magnesium in
the presence of dry ether to obtain acetoacetic ester, gammabromoacetoacetic ester and other products which he believed
were condensation products of these compounds.
Rottinger and
Wenzel (9 ) reported similar results in the same reaction.
Sommelet and Hamel (1 0 ) treated chloroacetic ester with
magnesium in dry ether and mercuric chloride as a catalyst
and reported gamma-chloroacetoacetic ester as their only
product.
If this reaction was carried out in the presence of
ethyl acetate they obtained acetoacetic ester as the main
product.
McKenzie, Drew, and Martin (1 X ) treated dl-alpha-chlorophenylacetic acid with 2-7 mols of methylmagnesium iodide
and obtained alpha, beta-diphenylsuccinic acid in small
yields together with small amounts of alpha-hydroxy phenylacetic acid and glycols.
dl-Bromo-phenylacetic acid when
treated with 4 mols of phenyl Grignard yielded a mixture of
products among which were diphenyl, phenol, diphenylacetic
acid, and triphenylethylene glycol.
Boyle, McKenzie, and Mitchell (1 2 ) obtained 1,1,2,2tetraphenylethyl alcohol as the main product from the addi­
tion of chloro-phenylacetyl chloride to phenyl Grignard.
The reverse addition gave mainly tar with a small amount of
phenyl diphenylmethyl ketone.
The addition of p-totyl and
alpha-naphthyl Grignards to chloro-phenylacetyl chloride
likewise gave tars.
The addition of chloroacetyl chloride to phenyl Grignard
gave 1 ,2 ,2 -triphenylethanol and some chloromethyl diphenyl
carbinol, while the addition of chloroacetic acid to phenyl
Grignard gave triphenylethylene glycol as the main product.
Treatment of 3-hydroxy-4,4,4-trichloro-butyric acid and
its ethyl ester with phenyl Grignard gave dl-l,l-diphenyl-3trichloromethyl-1 ,3-dihydroxy propane.
Examination of these reactions reveals that in some
cases the halogen atoms underwent a Wurtz reaction with the
Grignard reagent and in other cases the halogen atoms re­
mained unreacted.
7
DISCUSSION
A. Reaction of n-butylmagnesium bromide with 2-hexanone ^1 3 ).
The reaction of n-butyl magnesium bromide with ethyl
acetate 11 3 ) lead to the formation of a 3% yield of hexanol-2
in addition to a 64$ yield of the tertiary carbinol.
The for­
mation of hexanol-2 in this reaction probably resulted from
the reduction of the intermediate hexanone-2.
It therefore
seemed of interest to determine the extent of reduction which
would result from the addition of hexanone-2 to n-butyl
Magnesium bromide.
Hexanone-2 was added as rapidly as possible to an excess
of n-butyl Magnesium bromide at reflux temperature ^about
ox
40 ) to insure the maximum possible amount of reduction.
Fractionation of the products revealed the formation of hex­
anol-2
(9fo
yield) and methyl-di-n-butyl carbinol
[60fo
yield),
10% of the ketone remaining unreacted, perhaps as a result
of enolization.
B. Reaction of n-butyl Magnesium bromide with ethyl
chloroacetate.
The amount of reduction of carbonyl compounds by
Grignard reagents has been explained as being a function
both of the negativity of the groups adjacent to the car­
bonyl group and of the alkyl group of the Grignard reagent.
Such an explanation seems reasonable in view of such reac­
tions as t-butyl Magnesium chloride with pivalyl chloride
^1 4 )
give neoperityl alcohol in 95% yields, n-butyl
8
Magnesium bromide with pivalyl chloride (1 3 ) to give neo­
pentyl alcohol in 27% yield plus a 69% yield of n-butyl-tbutyl- carbinol, and n-butyl Magnesium bromide with acetyl
chloride (1 3 ) to give ethyl alcohol in an 8 % yield.
Many
other supporting reactions could be cited from the literature.
However there are also many apparent refutations of this
theory.
For example, why is there no neopentyl alcohol
formed from the action of n-butyl Magnesium bromide on ethyl
trimethylacetate (1 3 ) while a 27% yield of neopentyl alcohol
is obtained from n-butyl Grignard and pivalyl chloride?
In this connection the effect of negative groups, such
as chlorine atoms, on the amount of reduction by Grignard
reagents appeared to be an interesting and perhaps valuable
study.
For this study, ethyl chloroacetate and ethyl tri­
chloroacetate were chosen as the halogen compounds.
Ethyl chloroacetate was added rapidly to a large ex­
cess of n-butyl Magnesium bromide at reflux temperature so
as to present the most favorable conditions for reduction.
Examination of the products revealed that no ethylene
chlorohydrin (the possible primary reduction product)
obtained.
m. s
Among the products obtained were di-n-butyl-n-
amyl carbinol (47.2% yield) and a small amount of a very
active halogen compound which was not identified but may
have been chloroacetaldehyde.
A rather large residue of
very high boiling material was also obtained.
The reaction
was not investigated thoroughly since the primary aim was
simply the determination of the amount of primary reduction
product formed, namely ethylene chlorohydrin.
C. Reaction of t-butyl Magnesium chloride with ethyl
chloroacetate.
Since t-butyl Magnesium chloride is a much better re­
ducing agent than n-butyl Magnesium bromide it appeared
that the formation of ethylene chlorohydrin was much more
likely to result in this reaction than in the previous one.
In view of the large amount of coupling product (47$) ob­
tained in the previous reaction, the possibility of the for­
mation of neopentyl carbinol also presented itself.
This reaction was carried out in the same manner as
the previous one.
No ethylene chlorohydrin, ethylene oxide
or neopentyl carbinol were found among the products.
As in
the previous reaction, a small amount of unidentified active
halogen compound was formed which may have been chloroacetaldehyde.
The majority of the products consisted of a
black tarry residue.
This reaction was not investigated
thoroughly since the primary purpose was simply the deter­
mination of the amount of primary reduction products resul­
ting from it.
D. Reaction of n-butyl Magnesium bromide with ethyl
trichloroacetate.
This reaction was carried out in the same manner as
the two previous reactions.
reduction product) was found.
No trichloroethanol (primary
Among the products formed
were n-butyl chloride and n-octane.
The formation of these
10
two compounds is not surprising in light of the work of
Dean and Wolf (7 ) who reported similar formations of alkyl
halide and hydrocarbon in the reactions of Grignard reagents
with chloral.
A very small amount of a third product, boiling be­
tween n-butyl chloride, and n-octane, was also formed but
this substance could not be obtained pure and identified.
A very large tarry residue was also obtained.
The presence
of the active halogen atoms apparently causes complex re­
actions to take place which lead to the formation of these
tars.
While Howard (5 ,5 ) did not report the formation of
tars from the reaction of various Grignard reagents with
chloral, the fact that his shields
sec. alcohols (less
than 35%) account for only a small part of the chloral
used, leads one to suspect that he too obtained comparable
products.
I. Reaction of t-butyl Magnesium chloride with ethyl
trichloroacetate.
The formation of n-butyl chloride and n-octane in
the preceeding reaction leads one to suspect a similar re­
sult in this reaction.
Examination of the products failed
to show the presence of any t-butyl chloride (or t-butyl
alcohol which would result from the hydrolysis of the
chloride).
Hexamethylethane was obtained but the amount
did not seem to be any larger than that usually obtained
in any reaction involving the use of t-butyl Magnesium
chloride.
No trichloroethanol was formed.
A very large
yield of tar was obtained.
From the large yields of tars obtained in these re­
actions of ethyl chloroacetates with Grignard reagents one
is lead to conclude that the activity of the halogen atoms
is too great to be of any value in a study on the effect of
negativity on the reduction of an adjacent carbonyl group
by Grignard reagents.
The halogen atoms, rather than re­
maining attached to the carbon atom of the esters, appar­
ently enter into various types of reactions, such as coup­
ling IWurtz reaction), etc. with the resulting production
of tars.
In closing the discussion on the reaction of the
chloroacetic esters with Grignard reagents it must again
be emphasized that no attempt was made to study these re­
actions completely.
Many products were found which were
not identified because the purpose of this work was not a
complete reaction study but, as stated before, was a search
for reduction products.
F. Reaction of t-butyl Magnesium chloride with benzaldehyde.
This reaction as well as the following one was studied
as a result of the work of Clapper(1 5 ) of this laboratory
on the action of t-butyl Magnesium chloride on benzoyl
chloride.
Among the products which he obtained were ben­
zoin and benzil.
In order to throw more light upon the
subject of a possible mechanism for the formation of the®
two products, it appeared that the reaction of benzaldehyde and benzil with t-butyl Magnesium chloride might he]p
solve this problem.
Did two benzoyl groups couple to form benzil which
was then reduced to benzoin, or was the benzoin formed
as a result of a benzoin condensation of benzaldehyde
which might have been formed as an intermediate by the re­
duction of benzoyl chloride?
Pure benzaldehyde was added as rapidly as possible to
an excess of t-butyl Magnesium chloride at reflux tempera­
ture.
Benzyl alcohol (22% yield), t-butyl phenyl carbine!
(51°jo yield) and a large viscous oily residue were obtained.
By crystallization from petroleum ether it was possible to
obtain a small amount of a crystalline compound, which
formed an acetyl derivative with acetyl chloride, from this
residue.
This compound was probably dl-hydrobenzoin on the
basis of its physical constants and those of its acetyl
derivative.
This large oily residue was probably a complex mixture
of compounds, petroleum ether being the only solvent from
which a crystalline product could be obtained.
G. Reaction of t-butyl Magnesium chloride with benzil.
Benzil was dissolved in benzene and added to an ex­
cess of t-butyl Magnesium chloride at reflux temperature.
After decomposition of the reaction mixture and removal of
solvents a viscous liquid resembling the residue in the preceeding reaction was obtained.
By means of Girard’s reagent this material was separ­
ated into two fractions.
The ketone fraction consisted of
benzoin (46% yield).
The non-ketone fraction was a vis­
cous oil which yielded a white solid m.p. 133-135° amount­
ing to 1 0 .8 % of the total products on crystallization from
petroleum ether.
This compound was not identified but
appeared to be one of the isomeric hydrobenzoins.
This
fraction probably also contained compounds resulting from
the addition of the Grignard reagent to one or both of tie
carbonyl groups since the weight of the products was great­
er than the weight of the benzil used.
EXPERIMENTAL
A. Addition of hexanone-2 to n-butyl Magnesium bromide.
The hexanone-2 used in this reaction was prepared
as follows.
Three kg. of Eastman "Technical" hexanol-2
were dehydrogenated over a copper catalyst in the labora­
tory dehydrogenator and then fractionated with column //I
into 67 fractions of about 40 g. each.
The best ketone
fractions were combined and refractionated with column B
into 40 fractions of about 13 g. each.
The pure ketone
fractions b.p 125° (732 mm), n^° 1.4010 were used for
addition to the Grignard reagent.
n-Butyl Magnesium bromide was prepared in the usual
way from 5 moles of magnesium and 5 moles of n-butyl brom­
ide.
The clear Grignard solution obtained was siphoned
from the unreacted magnesium.
The addition of 200 g. (2
moles) of the pure hexanone-2 to this Grignard solution
(2050 cc., 2.18 normal, 4.47 moles) was completed in one
hour.
The reaction mixture was heated on a waterbath with
stirring for another hour after the addition of the ketone
was completed.
The reaction mixture was poured upon about 2500 g. of
cracked ice.
The ether layer was decanted and the aqueous
layer steam distilled till no additional oil steam dis­
tilled.
The layers were separated and the aqueous portion
of the steam distillate was extracted twice with 300 cc.
portions of ether.
The ethereal solutions were combined
and dried over anhyd. K 2 C03 .
The ether was removed on
the steam bath through a 110 cm. indented column.
residue was then fractionated with column
tfZ.
Bath
Col.
12; 40
1; 28
1; 51
2; 18
2; 43
3; 11
3; 44
4; 00
112
1 24
125
126
126
127
132
13 3
59
62
62
65
76
68
68
68
45
55
60
6 1.5
62 .5
67
69
7 0.5
1
2
3
4
5
6
7
8
10; 55
11; 40
12; 20
12;36
1;05
1;42
2; 28
2; 53
3; 34
3; 53
4; 37
139
146
1 48
149
148
147
146
124
133
124
126
67
70
76
76
76
74
108
1 00
1 11
1 09
111
69
70
7 5.5
80
80
79
77
82
95
98
97
9
10
11
12
13
14
15
16
17
18
19
3 .1
6.7
4 .4
2.0
4 .4
6 .8
5 .3
1 .6
4 .9
7.7
12.6
1.4020
1.4011
1.4030
1.4079
1.4138
1.4152
1.4 1 6 0
1.4186
1.4331
1.4 3 6 0
1.4359
10 ;02
10; 17
10; 33
10; 45
11; 00
11; 15
11; 26
11; 40
12;00
12; 15
12; 33
12; 50
1;01
1; 57
101
103
104
105
105
105
105
105
1 05
106
106
106
1 09
1 28
93
99
95
94
93
93
92
92
93
95
96
96
95
127
65
67
65
66
65
66
65
66
67
67
67
70
70
78
20
21
22
23
24
25
26
27
28
29
30
31
32
33
1 0.9
17.2
1 8.0
18.5
18 .1
17.8
1 7.1
19.7
18.4
1 7 .8
18.2
18.3
7 .9
9.9
1.4356
1.4358
it
ii
ii
it
ti
it
•
•
0 .7
2.0
1 .9
3 .3
2 .1
2 .4
3 .0
2 .5
Index Pressure
1 . 4 0 5 1 1 2 5 mm
tt
1 .4 1 1 0
it
1 .4122
it
1.4000
1 .3982 n
it
1.4020
it
1 .4045
it
1.4027
ii
ii
1.4360
ii
1.4358
it
o
i—i
A residue of about 10
-p
Time
Vapor Frac
The
tt
it
98
87
74
65
24
20
20
20
3
tt
it
it
it
it
ii
iiit
ii
it
ii
it
remained.
g .
Fractions 1 to 12 were combined and refractionated
with column C.
Time
4;48
5;09
Bath
145
146
Col.
115
119
Vapor Frac.
105
109
1
2
Wt.
Index Pressure
2.3
4.1
1.4060 739 mm
1.4043 n
5; 23
148
122
111
3
3.8
1.4008 739 nun
7; 53
152
154
161
187
195
124
124
125
133
141
116
116
116
118
118
4
5
7
3.4
5.1
5.2
3.6
8
2.0
1.4017
1.4022
1.4021
1.4021
1.4030
8 ;29
9; 05
9; 35
9 ;48
6
A residue of 5 g. remained.
Fractions 4-8 were hexanone-2, 2,4-dinitrophenyl-
o
hydrazone m.p. and mixed m.p., 106-107 .
Fractions 13-16 were refractionated with column D.
Time
Col.
Vapor
Frac.
Wt.
Index
Pressure
3; 55
4; 22
4; 50
5;05
5; 26
140
141
143
143
143
131
133.5
it
135
135
1
2
1.9
1.5
0.7
1.5
1.7
1.4069
1.4129
1.4150
1.4160
1.4162
739 ram
n
7; 23
7; 33
7; 41
7 ;46
143
143
143
143
135
136
136
136
6
1.5
7
8
1.0
1.2
9
0.9
1.4162
1.4163
1.4162
1.4162
Residue 0.9 g.
3
4
5
ii
it
it
n
it
it
ii
These fractions were identified as hexanol
2 by oxidizing fractions 6,7, and 8 with acid dichromate
at room temperature to hexanone-2 which gave a 2,4-dinitro
o
phenylhydrazone m.p. and mixed m.p. 106-107 .
A test on
these fractions for hexanone-2 with 2 ,4-dinitrophenylhydrazine before oxidation showed that no hexanone-2 was
present.
B. Addition of ethyl chloroacetate to n-butyl Magnesium
bromide.
n-Butyl Magnesium bromide was prepared in the usual
way from 8 moles of magnesium and 8 moles of n-butyl
bromide to give 3025 cc. of 2.24 normal Grignard solution.
To this Grignard solution were added 245 g. (2 moles)
of ethyl chloroacetate, b.p. 140° (718 mm) ,• ti2° 1.4210,
during 2 hours with vigorous stirring.
There was a vig­
orous evolution of gas throughout the addition of the ester.
The reaction flask was surrounded with a salt ice
mixture and the Grignard addition complex decomposed by
the addition of 1500 cc. of water with stirring during 1
hour and 40 minutes.
The ether layer was decanted and
the residue extracted twice with 500 cc. portions of ether.
The residue was then neutralized by the addition of about
500 cc. of conc. HC1 diluted with 500 cc. of water.
The
ether layer was separated and the aqueous layer extracted
with two 700 cc. portions of ether.
The ethereal solutions
were combined and ether distilled off through a 1 1 0 cm. in­
dented column till a residue of about 1500 cc. remained.
This residue was dried over anhyd. K gC0 3 and the distilla­
tion of ether continued till a residue of about 700 cc. re­
mained.
Time
This residue was then fractionated with column $1.
Bath
Col. Vapor Frac.
60
61
61
61
67
70
1+
2+
9; 40
142
143
144
3; 00
5; 15
6.03
7; 15
7; 40
8 ;23
147
152
156
160
179
191
62
63
63
80
97
4+
5+
6+
7+
100
71
73
74
75
79
80
9; 14
9; 20
9; 58
145
147
151
72
73
73
53
71
71
8 ;20
8 ;45
3+
Wt.
Index
Pre;
5.6
3.2
5.0
1.3663
1.3832
1.3917
735
9.0
1.3958
1.3949
1.3744
1.3744
1.3735
1.4021
U
6.6
6.1
90
4.8
3.0
3.3
10
110
120
3.0
4.0
8
2.0
1.3729
1.4067
1.4063.
IT
!1
II
TT
IT
IT
I!
128
ii
120
43
55
40
155
157
157
72
74
81
65
61
58
130
14o
15
3.6
4.0
1.9
1.4065
1.4074
1.4093
10 0
86
11
1
53
15
3 55
161
159
160
85
91
115
63
77
92
16
17
18
2.2
3.3
5.3
1.4149
1.4261
1.4398
60
45
35
4 00
5 15
8 30
9 50
11 30
157
158
163
162
144
122
122
87
89
19
6 .6
127
142
125
110
20
21
22
9.2
3.0
7.0
3.8
1.4450
1.4435
1.4439
1.4421
1.4430
20
20
20
20
11 20
12 00
12 30
2 30
2 45
3 00
118
130
128
132
136
136
136
136
135
135
136
133
140
1.4438
1.4461
1.4463
1.4461
3
3
12
155
156
157
158
158
158
158
158
158
158
158
160
161
10 00
10 50
175
188
138
160
10
11
12
3
3
3
3
4
4
5
15
28
42
57
15
42
119
80
23
82
88
106
n
ii
n
ii
ii
ii
ii
ii
ti
ii
ti
ii
24
25
26
27
28
29
30
31
32
33
34
35
36
5.5
15.6
17.1
15.5
19.2
15.9
16.0
16.1
17.3
17.0
15.4
15.3
37
38
15.6
16.1
11.1
it
73
3
ii
ii
ii
ii
ii
H
ii
II
ii
II
it
It
it
It
it
11
ii
II
2
It
IT
1.4477
II
Residue 55 g. of black liquid.
1+) These fractions consisted of two layers.
l0 ) These fractions gave off halogen acid slowly and
gradually assumed a very dark color.
The density of fraction
11.4657
7.4838
3.9823
4.7364
density =
3.9823
4.7364
wt.
wt.
wt.
wt.
j f 31
of
of
of
of
was determined.
picnometer'+ sample
picnometer
sample
an equal vol. of water at
= 0.8408
Determination of molecular refraction.
m. r. =
,r. 1,)
a (n2+ 2)
M. R. =■ . .2.1.4.,2 (1.4461
.8408 (T75¥5I
14 C
30 H
1 0
M. R.
- 1
+ 2
_ 57^5
= 33.80
= 33.00
= 1.525
= 68.325
These calculations were made on the basis that the com­
pound was di-n-butyl-n-amyl carbinol.
The index of re­
fraction and boiling point compare favorably with those
given by Whitmore and Williams t1 6 ).
The molecular weight of this compound Ifractions 2537) was determined by means of the benzene freezing point
depression method,
M
M
K
g
G
T
=
=
=
=
e
=
using thiophene free benzene.
1000 Kg
GT
molecular wt. of solute
5.1 (freezing point constant of benzene)
weight of solute
8.73 (weight in grams of solvent, benzene)
freezing point depression
Run #1.
23.0395
22.8097
.2298
wt. of container with liquid
wt. of container after removal of sample
wt. of sample
Freezing point of pure benzene = 5.16
Freezing point of the solution = 4.55
Freezing point depression = 5.16 - 4.55 = 0.61
M =
1000 x 5.1 x .2298
8.73 x 0.61
=
pp.n
The molecular weight calculated for C 14 H 2 9 0H = 214.2
22.8096
22.5705
.2393
wt. of container with liquid
wt. of container after removal of
wt. of sample
sample
Freezing point of the solution = 4.53
Freezing point depression = 0.63
M =
1000 x 5.1 x .2393
6.73 x 0.63
=
p.p.p.
Thus this material is apparently di-n-butyl-n-amyl
carbinol.
An attempt was made to convert some of this material
to the chloride.
Fraction #31 was placed in a small sep­
aratory funnel and shaken with three portions of conc. HC1,
each portion being 2.5 times the volume of tert. carbinol.
The organic layer was then washed twice with 1.5 times its
volume of conc.
H 2 S04 . It was then dried over anhyd. K 2 C03.
Itsindex of refraction was found to be 1.4556
as
compared
with 1.4490 given by Williams li6 ).
Fractions 9 - 1 4
contained halogen, gave off halogen
acid slowly, and became very dark and gummy.
It had a
rather pungent odor.
C. Addition of ethyl chloroacetate to t-butyl Magnesium
chloride.
R u n 7fl.
Ethyl chloroacetate, 122,5 g. (1 mole), was
added with stirring to 1800 cc. (3.02 moles) of clear fil­
tered t-butyl Magnesium chloride during 2.5 hours and then
refluxed on a water bath for 25 minutes.
During the addi­
tion of the ester there was a steady evolution of gas.
The
reaction mixture was poured on 2 000 g. of cracked ice
which caused the formation of a yellow precipitate.
The
mixture was subjected to steam distillation after dissol­
ving most of the precipitate by the addition of dilute HC1.
The steam distillation residue was extracted with ether and
the ether distilled leaving a residue of 27 g. of black tar.
The layers of the steam distillate were separated, and
the aqueous portion extracted several times with 250 cc.
portions of ether.
The ethereal solutions were combined,
dried over anhyd. K 2 C03 , and the ether removed through a
110 cm., indented column.
column
The residue was fractionated with
3.
Time
Bath Col.
Vapor Fract.
Wt.
Index
Pressure
3;00
3; 25
3; 56
4; 22
4; 50
5;02
149
148
172
176
179
180
85
85
87
90
103
73
74
76.5
79
90
2.6
1.3780
1.3837
1.3846
1.3880
1.3908
732 mm
112
100
4;15
4; 52
5; 18
5; 32
8 ;35
8 ;50
9; 25
126
126
138
143
146
140
157
108
110
1.4221
1.4311
1.4357
1.4369
1.4358
1.4367
1.4421
' 97
94
76
60
30
15
114
113
123
123
158
1
2
3.6
4.8
3.6
3.9
3
4
5+
6+
12.1
88
7+
1.7
90
92
93
84
76
82
8
1.8
9
3.1
3.0
10
11
12
2.0
1.2
13
3.0
(+) These fractions were mainly hexamethylethane.
ii
n
ii
it
it
2
A resi­
due of 5 g. remained.
Fraction
3 gave an"alpha-naphthyl urethan, m.p. and
mixed m.p. 78-79° with the urethan of ethyl alcohol.
Fraction £11 did not form a 2 ,4 -dinitrophenylhydrazone
and was thus not a carbonyl compound.
It gave a slight
reaction with sodium, but gave no urethan with phenyl
isocyanate.
Since the amount of products was so small and no evi­
dence of the formation of ethylene chlorohydrin was found,
a second run was made.
Run
jf2.
The addition of 245 g. (2 moles) of ethyl
chloroacetate to 4150 cc. of clear filtered t-butyl Mag­
nesium chloride which was 1.57 normal, was completed in 5
hours.
The complex was decomposed by the addition of 380 cc.
of water, the resulting solid mass extracted with ether and
then dissolved in dilute HC1.
The layers were separated
and the aqueous portion extracted with ether.
All of the
ether solutions were combined and the ether distilled
through a 1 1 0 cm. indented column till a residue of about
1 liter remained.
This residue was dried with 30 g. of
anhyd. Na 2 S0 4 and then fractionated with column
Col. Vapor Fract. Wt.
Time
Bath
4; 15
5; 10
81
87
32
36
3 4.5
it
1
2
3
1;00
1; 37
2; 10
2; 55
3; 20
3; 55
4; 27
5;03
6; 20
6; 39
6; 58
7; 27
8; 15
98
98
108
1 13
1 16
1 26
130133
140
140
1 41
142
144
53
54
54
54
54
54
70
82
86
86
88
92
98
41 .5
60
65
68
69
68
65
73
79
87
95
99
1 00
4
5
6
7
8
9
10
11
12
13
14
15
16
1 2 .1
2.1
2 .2
8 .5
7 .4
1 1 .2
5 .5
4 .1
4 .3
2.0
2 .4
3.9
4 .7
3; 15
144
102
1 01
17
4 .8
Index
4 2 2 . 0 g. ether
203.5
1.3524
—
—
1.3547
1.3914
1.3 9 3 1
1.3932
1.3978
1.3959
1.3959
1+ )
it
u
f f 2.
Pressi
730
ti
7 33
ii
ii
ii
it
ti
it
ii
it
tt
ii
it
n
n
ii
HME
7 25
3; 45
5;15
143
158
102
101
108
-
18
19
3.1
11.5
1;25
1;55
2 ;38
3; 40
4; 40
139
139
135
145
152
81
81
83
77
83
73
76
70
20
21
22
1.6
68
23
24
76
2.5
4.7
4.3
5.9
725
734
HME
ii
1.4224
1.4300
1.4305
1.4343
1.4341
46
32
36
25
23
A large residue of black tar remained.
{+)
These fractions had a dark color and gave off halogen
acid.
No line could be seen in the refractometer, thus on
refractive index is given.
Fractions 5 to 10 consisted of
two immiscible liquids, the index readings being for the
top organic layers.
HME refers to hexamethylethane.
Fraction yl2 when tested with alcoholic AgN0 3 immed­
iately gave a large amount of white precipitate which was
soluble in excess conc. NH 4 0H.
This material gave a test
for carbonyl compounds with 2 ,4-dinitrophenylhydrazine,
qualitative reagent (page 38, Fuson and Shriner).
However
it was not possible to obtain a 2,4-dinitrophenylhydrazone
from this material.
It was thought that this material might be chloroacetaldehyde.
An attempt was made to convert it to phenoxy-
acetic acid as follows.
One-half g. of KMn0 4 was dissolved
in 10 cc.'of water and 0.5 cc. of fraction 12 added.
solution became warm and finally rather hot.
The
More perman­
ganate was added till a permanent permanganate color remained.
The excess permanganate was then removed by the addition of
a little ferrous sulfate.
After filtering the solution and
evaporating to dryness on the steam hotplate, a solution of
about 0.5 g. of sodium phenolate was added and the mixture
allowed to stand for a few days.
The mixture was filtered
and the filtrate acidified with HC1, then extracted with 10
cc. of ether and the ether evaporated leaving a small liquid
residue which had the odor of phenol but showed no evidence
of phenoxyacetic acid.
This same procedure when carried out
on a small amount of chloroaeetic acid gave phenoxyacetic
acid as a derivative.
Fraction 21 gave a faint halogen test when treated
with alcoholic AgN03 .
evolution of gas.
It reacted slowly with sodium with
There was no visible reaction with
acetyl chloride however.
It gave a cloudy solution with
2,4-dinitrophenylhydrazine qualitative reagent.
The aqueous solution from the Grignard decomposition
was examined for ethylene chlorohydrin as follows.
The
solution was distilled, giving a distillate of about 1500
cc. and a residue which solidified to a hard mass on cool­
ing.
The distillate was fractionated with column f f 4 but
o
no material boiling above 99 at 738 mm was obtained this
indicating the absence of ethylene chlorohydrin.
A third run was made, the results being essentially
the same as those of the first two runs.
A fourth run was
made for the purpose of examining the evolved gases for
ethylene oxide.
The evolved gases were passed into a trap
of glacial acetic acid surrounded by an ice bath.
The re­
sulting solution was refluxed to convert any dissolved
ethylene oxide to the monoacetate.
After three hours of
refluxing, the acetic acid was distilled off.
During the
distillation the temperature of the vapors never went above
the boiling point of acetic acid.
No residue remained,
indicating that no glycol monoacetate was formed.
D. Addition of ethyl trichloroacetate to n-butyl Magnesium
bromide.
Run #1.
SO
Ethyl trichloroacetate, b.p. 103.5° (100 mm),
1.4495, 383 g. (2 moles ) was added to the n-butyl Mag­
nesium bromide prepared from 5.5 moles of n-butyl bromide.
The ester reacted violently with the Grignard with the evo­
lution of gas.
Three hours were required for the addition
of the ester to the Grignard.
The reaction mixture was poured on 2.5 kg. of cracked
ice, the ether layer decanted from the thick sirupy mater­
ial which was dissolved by the addition of HC1 and then ex­
tracted with four 500 cc. portions of ether.
The ethereal
solutions were combined and the ether distilled through a
110
cm indented column leaving a thick viscous residue.
About 1 liter of water was added to this residue and
the mixture steam distilled.
The steam distillation resi­
due was extracted with ether and the ether distilled leav­
ing a residue of black tar.
The layers of the steam dis­
tillate were separated, the aqueous portion extracted with
ether and the ethereal extracts added to the organic layer.
The ethereal solution was dried with anhyd. K 2 C0 3 and the
ether distilled through a 1 1 0 cm. indented, column.
The residue was fractionated with column fZ.
Bath
Time
105
109
1;00
1; 30
2 ;20
3;15
4;00
4;45
11 0
115
120
124
123
130
138
145
147
146
146
146
148
148
148
10; 46
11;15
1 1 ;55
12;00
12; 40
1; 29
2 ;14
2; 58
5; 43
4; 35
5; 22
1 1 ;00
11; 45
12; 35
1 ;20
5; 35
3; 55
4;07
4; 25
4; 25
5; 18
8;00
8 ;13
151
152
155
158
118
129
137
141
141
141
141
146
Col.
Vapor
52
56
58
70
70
70
40
64
77
77.5
78
78
72
73
80
92
98
108
78
78
78
81
120
121
121
121
128
125
126
127
140
68
74
80
84
88
95
136
158
101
117
120
121
123
124
124
124.5
125
123
10 0
67
70
66
75
80
81
10 1
108
Fract . Wt.
1
2
2.4
2.7
5.6
5.7
3
4
5
10.2
6
12.7
7
7.9
9.3
6.3
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
2.6
3.9
3.0
3.8
5.6
5.5
6. 2
5.8
5.4
7.7
7.9
7.0
2.6
3.1;
2.7
1.9
2.1
4.1
7.0
1.9
Index
Pressure
1.3547
1.3770
1.4036
1.4030
1.4013
1.4013
731
1.4013
1.4013
1.4013
1.4019
1.4110
1.4150
1.4015
1.3997
1.3989
1.3980
1.3976
734
1.3975
1.3973
1.3973
1.3970
1.3970
1.3968
1.3970
1.4000
1.4233
1.4350
1.4262
1.4368
746
IT
T!
It
1!
It
ti
ti
it
it
it
ii
it
it
ii
it
ii
ii
ii
100
ii
80 .
62
50
24
20
20
A residue of 5 g. remained.
Fractions 5-9 appeared to be n-butyl chloride.
A
Grignard reagent was prepared from 3.5 g. of this material
in the usual way.
The grignard reagent was treated with
phenyl isocyanate to give an anilide m.p. and mixed m.p.
60-61° with valeranilide.
Thus this material was identi­
fied as n-butyl chloride.
Fractions 13 - 25 which appeared to be n-octane were
combined and fractionated with column
j f 3.
Time
Bath
Col.
1; 45
2 ;52
3; 42
5; 25
143
140
142
144
120
125
125
126
3 ;14
4 ;12
4; 55
5 ;24
143
142
141
143
8;16
8 ;42
9;04
9; 30
144
145
151
175
Residue 2 g.
Vapor
• 105-118
Fract. Wt.
Index
Pressure
1.3
3.5
7.1
3.9
1.4003
1.4000
1.3988
1.3990
734
121.5
30
31
32
53
128
128
129
130
123
123
123
123.5
34
35
36
37
2.0
1.3980
1.3976
1.3970
1.3972
739
3.8
5.2
5.1
130
130
131
143
123
123
125
123
38
39
40
41
7.2
5.7
5.2
2.4
1.3970
1.3970
1.3970
1.3970
ti
120
120
tt
n
ii
ii
ii
n
ii
it
ti
The density of fraction fr37 was determined.
10.8799
7.4854
3.3965
4.7364
wt. of sample +
picnometer
wt. of picnometer
wt. of sample
o
wt. of an equal vol. of water at 4 .
20
density — ^
=
5.5965 =
4.7364
0.7171
Fractions 26 -29 yielded no urethans when treated with
phenyl isocyanate in the usual manner.
Run #2.
n-Butyl bromide, 6 moles, yielded 5.7 moles
of filtered n-butyl Magnesium bromide to which 342 g.
11.79 moles) of ethyl trichloroacetate were added with
stirring during 3 hours.
The following day the reaction
mixture was poured on 2 kg. of cracked ice and the mixture
steam distilled.
The layers of the steam distillate were
separated and the aqueous portion extracted with ether.
The ethereal solutions were combined and dried over anhyd.
K 2 C03 .
The ether was distilled off during about 25 hours
through column $4.
The residue was then fractionated
with the same column.
Time
Bath
Col.
Vapor
00
10
03
10
18
32
40
97
1 04
106
108
109
111
1 12
45
62
63
63
63
63
64
35
56
66
67
67
67
67
9; 20
10; 00
10; 30
10; 48
ii; 27
11; 59
12; 34
1 15
117
118
118
119
119
126
64
67
74
74
74
74
74
3; 43
4; 42
140
148
74
75
1; 50
3; 50
4; 25
6; 10
7;05
7; 33
8; 20
9; 04
9; 25
155
155
1 56
1 56
158
158
1 63
171
180
87
95
108
119
1 22
1 22
122
1 23
12 4
2; 05
2; 34
3;13
3; 25
1 44
1 66
1 82
186
70
72
94
94
1
2
3
4
5
5
7
Fract . Wt.
Index
Pressure
1
2
3
4
5
6
7
8.1
4.0
6.0
7.7
7.6
8 .8
9 .0
1.3579
1.3600
1.3861
1.3905
1.3910
1.3910
1 .3910
727 mm
fl
IT
tt
It
tt
tt
67
67.5
75
77
78
78
78
8
9
10
11
12
13
14
16.8
15.7
5.2
3.5
1 0.5
1 8.6
24.9
1.3913
1.3919
1.3963
1.4024
1 .4053
1.4083
1.4095
732
it
it
tt
it
tt
it
78
78
15
16
2 4.8
22.6
1.4101
1 .4132
it
it
17
18
19
20
21
22
23
24
25
7.9
6.4
12.3
6.3
4 .6
6.8
12 .0
12.6
7.2
1.4 1 8 0
1.4253
1.4322
1.4080
1.3978
1.3976
1.3971
1.3971
1.3968
728
ti
it
ii
ti
ti
it
it
ti
4 .4
5 .3
6.1
4 .1
1.3971
1.3965
1.3981
1.4207-
100
80
96
1 03
118
121.5
122
1 23
123
1 23
70
—
70
85
26
27 '
28
29
ii
43
26
Residue 7.0 g.
Qualitative tests on the material boiling at 67
o
showed that it was insoluble in water and insoluble for
the most part in cold conc. H 2 S04 , some heat being evol­
ved.
ride.
Heat was also evolved on treatment with acetyl chlo­
It reacted with sodium with evolution of gas, and
gave a halogen test with alcoholic AgN03 .
It would not
form a Grignard reagent.
It appeared that it was a mix­
ture of an alcohol and an alkyl halide.
Treatment with
phenyl isocyanate yielded a urethan, m.p. and mixed m.p.
o
49 - 50 with phenyl urethan of ethyl alcohol.
Ten cc. of fraction 7 were placed in a separatory
funnel and washed with 15 cc. of conc. H2S04 to remove
the ethyl alcohol,
a
colorless organic layer of 6.5 cc.
remained.
Thus fraction 7 was approximately 50°jo ethyl
alcohol.
After treating ■with a little sodium carbonate
it had a ref. index of 1.4054.
This material formed a
Grignard reagent which was treated with oxygen to give an
alcohol which formed a urethan, m.p. and mixed m.p. 59.5o
60 , with phenyl isocyanate. Thus the n-butyl chloride
was again identified.
Fractions 4 to 14 are ethyl alco­
hol and n-butyl chloride.
Fractions 21 to 27 are n-
octane, this material being inert to nearly all reagents.
Fractions 15 to 20 are intermediate fractions which appar­
ently contain a third compound which was not identified.
E. Addition of ethyl tri-chloroacetate to t-butyl Mag­
nesium chloride.
Ethyl trichloroacetate, 288 g. (1.5 moles J, wras add­
ed with stirring to 4500 cc. of 1.38 normal t-butyl Mag­
nesium chloride (clear filtered) during 2 hours and 45
minutes.
The ester reacted vigorously with evolution of
gas, the reaction mixture changing to a red color, then
to orange and finally back to a red color again.
The
reaction mixture was stirred for 4 hours and 20 minutes
after the addition of the ester was completed.
After
standing for a day the ether solution was decanted from
the large amount of solid which had separated.
The ether
solution and solid portion were decomposed separately by
the addition of cracked ice.
The ether solution gave very
little reaction with the water but the solid material re­
acted vigorously.
The two portions were combined and the
ether distilled off by heating on the steam bath.
The
ether distillate was fractionated carefully with column
ir 4
into fractions 1-9, b.p. 33-36.5° (736 mm) n|° 1.3528,
and fraction 10, 1 g . , b.p.38° (736 mm).
It was thus
evident that there was no t-butyl chloride in the ether
distillate.
The aqueous residue above was steam distilled, 100 cc.
of conc. HOI diluted with 100 cc. of water having been added
to prevent foaming.
The steam distillation residue was ex­
tracted with ether, and the ether distilled off leaving
87 g. of black tar which could not be distilled under re­
duced pressure.
The layers of the steam distillate were separated,
the aqueous layer extracted twice with 200 cc. portions of
ether, and the ethereal solutions combined and dried over
anhyd. NasS04 .
The dried solution was fractionated v/ith
column y 2 to give;-
fractions 1-4, b.p. 34-35
(737 mm);
fractions 5-6, 2.7 g. b.p. 62-71° (737 mm) n|° 1.3562-
1.3765; fractions 7-9, 29.5 g. b.p. 101° (729 mm);
fractions 10-13, 15.1 g. b.p. 67° (18 mm) - 101° (16
.
20
mm) n^
1.4288 - 1.4465; residue 2 g. of tar.
Fractions
1-5 were mostly ether, and fractions 6-9 were mostly hexamethylethane.
Fractions 10-13 consisted of unidentified
material which formed no derivatives with phenylisocyanate
or 2,4-dinitrophenylhydrazine.
F. Addition of benzaldehyde to t-butyl Magnesium chloride.
Benzaldehyde, 212 g. (2 moles) b.p. 73° (20 mm) n^°
1.5443, was added to a clear magnesium free solution of
5.46 moles of t-butyl Magnesium chloride with stirring dur­
ing 1 hour and 15 minutes.
The addition complex was decom­
posed by pouring upon 2 kg. of cracked ice.
The ether so­
lution was decanted and the gelatinous residue extracted
twice with 600 cc. portions of ether.
The gelatinous resi­
due was then dissolved for the most part by the addition
of 325 cc. of conc. HC1 diluted with 600 cc. of water.
The resulting aqueous solution was extracted twice with
300 cc. portions of ether.
The ethereal solutions were
combined and dried over anhyd K 2C03 .
The ether was re­
moved by distillation through a 110 cm. indented column
leaving a residue which was fractionated with column y2.
Time
Bath
Col. Vapor Fract. Wt.
11; 40
12; 50
1;05
2; 50
3; 50
147
135
137
145
141
75
66
98
100
103
39
35
73
91
68
1
2
3
4
5
2.4
1.5
2.8
5.3
7.2
Index
1.5857
1.3900
1.5037
1.5365
1.5403
Pressure
145 mm
90
20
19
3
42
45
15
24
12
50
00
14
35
48
13
143
146
149
155
156
156
156
156
156
157
160
95
98
100
108
108
109
108
108
108
108
111
66
66
68
75
76
77
77
81
78
78
78
6
7
8
9
10
11
12
13
14
15
16
12.5
12.2
11.6
8.6
12.7
21.0
20.8
21.9
22.0
21.5
19.7
2 58
3 35
4 20
172
182
208
112
114
157
82
83
106
17
18
19
13.0
2.5
2.6
4
5
7
8
9
9
10
10
10
10
11
1.5328
1.5362
1.5364
1.5217
solid
it
ii
ti
it
ti
ii
ii
it
2
2
2
2
2
3
3
3
3
3
3
4
4
4
Residue 75 g. of light yellow viscous oil.
Fractions 3 to 8 were combined and refractionated with
column //3.
Time
Bath
Col. Vapor Frac.
Wt.
Index
Pressure
13 mm
8
6
3; 50
4; 25
4; 58
150
142
144
103
103
108
86
95
80
20
21
22
1.9
3.4
3.3
1.5300
1.5347
1.5370
8; 36
9;03
10; 00
146
146
164
109
109
122
82
82
80
23
24
25
8.1
8.6
7.3
1.5395
1.5395
1.5395
Residue 9 g.
6
6
5
The refractometer was not working properly
at the time these readings were taken and consequently
they are not accurate.
Fractions 20 to 26 were mostly benzyl alcohol, phenyl
o
urethan m.p. and mixed m.p. 77 - 78 .
Fraction 9 was an intermediate fraction between benzyl
alcohol and t-butyl phenyl carbinol.
Fractions 10 - 19 were t-butyl phenyl carbinol m.p.
42.5-43.5° (51$ yield), phenyl urethan m.p. 105 -106°.
The 75 g. residue was subjected to vacuum distilla­
tion from a 500 cc. r.b. flask under the lowest possible
pressure that could be obtained with the vacuum pump.
Using an oil bath temperature up to 210°, 19 g. of dis­
tillate were obtained at a pressure of 3 mm.
Distillation
was extremely slow and there was evidence of decomposition
so distillation was stopped.
This distillate was dissol­
ved in an equal volume of hot petroleum ether.
Upon cool­
ing a white solid separated which was extremely insoluble
in hot pet. ether.
It was filtered off and after extract­
ing it several times with hot pet. ether, it was recry­
stallized from a mixture of 95$ ethyl alcohol and pet.
o
ether.
It melted at 110-115 ,
o
zation it melted 119-121 , and
o
zation it melted at 120-121 .
o
acid it melted below 100 thus
after a second recrystalliafter a third recrystalliWhen mixed with benzoic
proving it was not benzoic
acid.
The undistilled portion of the 75 g. residue was
diluted with hot pet. ether.
days a white solid separated.
for recrystallization.
After standing for several
Various solvents were tried
It was fairly soluble in chloro­
form, the solubility being decreased by the addition of
pet. ether.
It was recrystallized from a 3;1 mixture of
pet. ether and chloroform after which it melted at 105120°.
o
120 .
After a second recrystallization it melted at 118It was then extracted with a little pet. ether con­
taining a small amount of chloroform leaving an insoluble
potion m.p. 119-120°.
This compound, judging from its
solubility behavior and m.p., appeared to be dl-hydro­
benzoin which is reported in the literature as melting
variously from 119 to 122°.
It seems that the solid be­
fore recrystallization was a mixture of two or more com­
pounds, one of which was very insoluble in pet. ether.
The compound m.p. 119 - 120° was heated with acetyl
chloride to yield a compound which after recrystallization
from
95%
ethyl alcohol melted at 117-117.5°.
A mixture
of it with the starting material melted at 93 - 97° thus
proving it to be a derivative rather than unreacted mater­
ial.
The diacetate of dl-hydrobenzoin melts at 117-118°.
It thus seems fairly certain that the compound was dlhydrobenzoin.
Concentration of filtrates from the como
pound m.p. 119 - 120 yielded a viscous liquid from which
no crystalline compound could be obtained with any common
organic solvents.
G. Addition of benzil to t-butyl Magnesium chloride.
o
Benzil, 50 g . , 0.24 moles, m.p. 94-95 , dissolved
in 120 cc. of dry benzene was added to 580 cc. of 1.65
normal filtered t-butyl Magnesium chloride with stirring
during 40 minutes.
The reaction mixture was refluxed for
2 hours and 20 minutes till no more gas was evolved.
The
addition complex was decomposed by pouring upon 500 g. of
cracked ice.
Most of the Mg(0H)2 was dissolved by the
addition of dil. HCl, the remaining Mg(0H)2 being dis­
solved by the addition of NH4C1.
The ether layer was
separated, and the aqueous layer extracted three times
with 150 cc. portions of ether.
The ether solutions
were combined and the solvents distilled off under
vacuum leaving a residue of 71.5 g. of viscous yellow
liquid.
A little ether was added making 76.5 g. of
solution.
The following day some solid material had
separated.
The mixture was heated to make a homogeneous solu?
tion.
Ten g. of this solution were placed in a 250 cc.
flask with 70 cc. of 95$ ethyl alcohol and 10 g. of
Girard's reagent.
This mixture was refluxed on the steam
hath for 20 minutes, and then poured upon 100 g. of
cracked ice.
The mixture was then extracted three times
with 50 cc. portions of ether.
The combined ether ex­
tracts were then extracted once with water.
The ether
was then distilled leaving a residue of 5.6 g. of vis­
cous liquid which did not crystallize.
The aqueous solutions containing the dissolved
ketones were combined, 31 cc. of conc. HC1 added, and
the solution refluxed on the steam bath for one hour.
The solution was cooled and the white solid filtered
off, wt. 3 g.
This solid (white) was recrystallized
from 95$ ethyl alcohol and then from pet. ether after
o
which it melted at 131-133 . A mixture v/ith benzoin
melted at 131-133°, proving it to be benzoin, yield 23
g . , 46$ based on 3 g. per 10 g. of the 76.5 g. solution.
The non-ketone fraction above was dissolved in a
little hot pet. ether.
1
After standing for several days
g of white solid had crystallized.
After recrystalli­
zation from pet. ether containing a little chloroform it
o
melted at 130-132 . A mixture with benzoin melted below
115° proving it was not benzoin.
This compound was re­
crystallized again, separating in shiny silvery plates,
o
m.p. 133-135 . This compound was not identified but it
seems likely that it is one of the isomeric hydrobenzoins.
H. Description of the fractionating columns.
The fractionating columns used were of the total
condensation, variable take-off type, Whitmore
i 1 7 ),
containing single turn, 4mm glass helices t1 6 ) as pack­
ing.
The dimensions are reported as length of the packed
portion and internal diameter of the column;- column £l,
89 x 1.5 cm.; column y 2, 62 x 1.1 cm.; column y3, 40 x
0.9 cm.; column ;/-4, 110 x 1.5 cm.; column B, 58 x 1.1
cm.; column G, 46 x 0.9 cm.; and column D, 27 x 0.5 cm.
SUMMARY
The addition of hexanone-2 to an excess of n-butyl
Magnesium bromide lead to the formation of 9% of reduc­
tion product, hexanol-2.
The addition of ethyl chloroacetate to an excess of
n-butyl Magnesium bromide gave a 47% yield of di-n-butyln-amyl carbinol as the main product.
No ethylene chloro­
hydrin was found among the products.
The addition of ethyl chloroacetate to an excess of
t-butyl Magnesium chloride gave a tarry material as the
main product.
No ethylene chlorohydrin, ethylene oxide,
or neopentyl carbinol were formed.
The addition of ethyl trichloroacetate to an excess
of n-butyl Magnesium bromide lead to the formation of n-bu­
tyl chloride, and n-octane as well as a large amount of
tar.
No trichloroethanol was formed.
The addition of ethyl trichloroacetate to an excess
of t-butyl Magnesium chloride gave a tarry material as
the product, no trichloroethanol being formed.
The addition of benzaldehyde to an excess of t-butyl
Magnesium chloride gave a 22% yield of benzyl alcohol,
a 51% yield of t-butyl phenyl carbinol and a residue of
high boiling material from which a compound believed to be
dl-hydrobenzoin was crystallized.
The addition of benzil to an excess of t-butyl Mag­
nesium chloride gave a 46% yield of benzoin as well as a
non-ketone fraction which probably consisted, of a mix­
ture of hydrobengoins and glycols resulting from the addi­
tion of the Grignard to one or both carbonyl groups of
benzil.
0 5?
BIBLIOGRAPHY
1. Kharasch and Weinhouse - J. Org. Chem. 1
2.
Herbert - Bull Soc Chim 27
3.
Henry - Compt rend
45 (1920)
158 205 (1904)
4. Savariau - ibid 146
297 (1908)
5. Howard - J. Am. Chem. Soc. 48
6. Howard - ibid 49
774 (1926)
1068 (1927)
7. Dean and Wolf - ibid_58
8. Stolle - Ber. 41
209 (1936)
332 (1936)
945
9. Rottinger and Wenzel - Monatsh 34
1867
10.
Sommelet and Hamel - Bull Soc Chiin 29545 (1921)
11.
McKenzie, Drew and Martin -
J,
Chem. Soc. 107 26 (1915)
12..Boyle, McKenzie and Mitchell - Ber. 70B
2153 (1937)
13. Whitmore, Popkin, Whitaker, Mattil and Zech I. Am. Chem. Soc. 60
2458 (1938)
14. Greenwood, Whitmore and Crooks - ibid J30
2028 (1938)
15. Clapper - Penn State Thesis, M.S.
16. Whitmore and Williams - J. Am. Chem. Soc. J55
17. Whitmore and Lux - ibid
54
406 (1933)
4351 (1932)
18. Wilson, Parker, and Laughlin - ibid 55
2795 (1953)
PART II
Studies on Neopentyl Alcohol
•KV
I'
•
*
<
INTRODUCTION
Neopentyl alcohol has been under investigation
from time to time in this laboratory during the past
ten years.
The greatest amount of the work was done in
connection with the neopentyl halides and its relation
to the Whitmore theory of rearrangements.
The results
of these investigations might be summed up briefly by
saying that reactions of neopentyl alcohol' and neopentyl
halides which involve the removal of the hydroxyl group
or the halogen atom, respectively, are always accompan­
ied by rearrangements of the carbon skeleton, the yields
of rearranged products being much greater than those of
the normal or expected products.
Since many of the investigations on neopentyl alco­
hol in this laboratory were incomplete and inasmuch as
the dehydration of neopentyl alcohol has not been des­
cribed in the literature, the further study of neopentyl
alcohol seemed to be in order.
HISTORICAL
The history of neopentyl alcohol is an interesting
one and deals mainly with its use in the attempted prep­
aration of the neopentyl halides by methods which are gen­
erally used for the conversion of alcohols to their corres­
ponding halides.
Much of the early work is extremely
doubtful in light of the work of the last few years, par­
ticularly in light of the findings of the workers in this
laboratory.
Freund and Lenze
{± )
were the first to report the
preparation of neopentyl alcohol.
By the treatment of
r
neopentylamine hydrochloride with AgN02 they obtained an
alcohol boiling at 102 - 103°.
Their alcohol was prob­
ably t-amyl alcohol as Tissier (2 ) soon pointed out.
A
little later Freund i3 ) himself reported that t-amyl al­
cohol was formed when neopentylamine is treated with
nitrous acid.
Tissier (4 ) was probably the first to prepare neo­
pentyl alcohol.
By dropping a mixture of trimethylacetyl
chloride and trimethylacetic acid on
sodium amalgam
he obtained a fraction boiling at 105 - 120° from which
he isolated an alcohol^jn.p. 48 - 50°, b.p. 112 - 113°.
Another fraction b.p. 150 - 180° gave an ester b.p. 164 o
166 which on boiling with dry K0H gave neopentyl alcohol
and trimethylacetic acid.
He reported that the alcohol
was oxidized with chromic acid to trimethylacetic acid.
He also reported that the alcohol reacted with acetyl
chloride or acetic acid to yield an acetate b.p. 186°.
A little later Tissier (5 ) attempted the preparation
of neopentyl chloride.
He chlorinated neopentane, treated
neopentyl alcohol with dry HG1 below 0° and also with
phosphorus pentachloride.
He reported that the halide
obtained was the same in these three reactions.
He be­
lieved it to be neopentyl chloride and reported that it
was unstable, decomposing partially when distilled.
He
reported a specific gravity of 0.8792 at 0° for the chlo­
ride.
He also stated that the chloride could be converted
to an acetate which could be saponified to the original
alcohol.
He also reported that neopentyl alcohol on treat­
ment with sulfuric acid gave neopentyl acid sulfate.
Continuing his work on neopentyl halides, Tissier (6 )
reported that he prepared a bromide from neopentyl alcohol
by saturating it with HBr in a salt-ice bath and then heato
ing in a sealed tube at 35 for many hours. He demon­
strated the primary nature of the bromide only by the pro­
duction of the nitrolic acid color test.
The bromide gave
trimethylethylene on treatment with alcoholic pottasium
hydroxide and on distillation gave t-amyl bromide. He preo
pared an iodide b.p. 127 - 129 , dQ 1.0502, in a similar
manner.
His iodide on treatment with silver acetate gave
an acetate b.p. 124° which he claimed was saponified to
neopentyl alcohol.
He stated however that heating neopentyl
alcohol, saturated with HI at 0°, for 3 - 4
o
^
„ .
100 gave t-amyl iodide.
hours at
Schueble and Loebl (7 ) reported the preparation of
neopentyl alcohol by treating trimethylacetamide with so­
dium amalgam in amyl and octyl alcohols.
They also re­
ported the formation of some neopentylamine in this
reaction.
Bouveault (e ) reported that neopentyl alcohol is
formed from t-butyl chloride through the Grignard reagent
which reacts with methyl formate to give neopentyl alcohol
and some trimethylacetaldehyde.
Meyersberg {9 ) treated 2,2-dimethyl propanediol-1,3
with excess fuming HI at 100 - 110° for 30 hours and
stated that he obtained the corresponding iodohydrin and
neopentyl iodide b.p. 42 - 44° at 20 mm, sp.gr. 1.5317 at
o
13 . Treatment of the iodide with silver acetate in glacial
acetic acid yielded the acetate which he stated was hydro­
lyzed by dilute aqueous potassium hydroxide to yield neo­
pentyl alcohol, b.n. 110 - 120°.
Courtot (1 0 ) prepared neopentyl alcohol from t-butyl
magnesium chloride and paraformaldehyde.
given.
No yield was
A little later Samec i n ) prepared neopentyl al­
cohol from t-butyl bromide, magnesium and paraformaldehyde
below 15°.
His yield of alcohol was 4/6.
He reported that
oxidation of the alcohol with sodium dichromate and H 2S04
gave a 45$ yield of trimethylacetaldehyde b.p. 174°,
methyl isopropyl ketone and neopentyl trimethylacetate.
Menschutkin {l s ) working on the rates of esterification of alcohols in 1909 reported that neopentyl alco­
hol has a smaller esterification velocity than sec. amyl
alcohol, (pentanol-2 ).
Richard l1 3 ). reduced methyl trimethylacetate with
sodium in absolute ethyl alcohol to get a 60% yield of
neopentyl alcohol m.p. 50°, b.p. 113 - 115° at 760 mm,
o
phenylurethan, m.p. 114 . By treating the alcohol with
dry HG1, he obtained a chloride b.p. 87 - 90° which was
unstable, decomposing under the influence of.heat or on
vigorous drying to yield trimethylethylene and HC1.
His
chloride formed a Grignard reagent which on treatment
with oxygen and carbon dioxide gave dimethylethyl carbinol and dimethylethylacetic acid respectively.
He believed
that the isomerization occurred either in the formation of
the Grignard reagent or in the reaction of the Grignard
reagent with oxygen and carbon dioxide.
Franke ^ 4 ) reported the preparation of neopentyl
alcohol in a 12.5% yield by treating 3-bromo-2,2-dimethyl
propanol- 1 in aqueous methyl alcohol solution with sodium
amalgam.
The alcohol was oxidized to trimethylacetic acid
in good yields with chromic acid.
Franke and Hintreberger i1 5 ) passed neopentyl alcohol
vapors over brass turnings at dark red heat to obtain a
52% yield of trimethylacetaldehyde.
The evolved gases from
this reaction contained 0.5$ of C02 , 11.5$ of unsaturated
hydrocarbons, 13.4$ of CO and 70$ of hydrogen.
Hinterberger l1 6 ) could find no evidence of a Canniz­
zaro reaction with trimetnylacetaldehyde in the presence of
concentrated potassium hydroxide in water solution contain­
ing a little alcohol.
Ingold (1 7 ) stated that Robinson in unpublished work
dehydrated neopentyl alcohol to obtain 2 -methyl butene -1
and trimethylethylene, the former being converted to the
latter under the conditions of the experiment which were
not disclosed.
This is the only reference in the literature
concerning the dehydration of neopentyl alcohol.
Conant ^1 8 ) modified the procedure of Franke ^ 5 )
for the preparation of trimethylacetaldehyde from neopentyl
alcohol.
He passed the alcohol vapors over a copper cata-
lyst at 250 - 300
0
to get 60 - 66 ^ yields of trimethylacet0
aldehyde b.p. 74 - 76 .
The neopentyl alcohol was prepared
according to the following scheme.
HCl
Mg
HCHO
Me 3 C-0H *---- =» M e 3 C-Cl ----»> M e 3 C-MgCl. ---- =► Me 3 CCH 2 0H
He reported that if considerable excess of formalde­
hyde is passed into the Grignard solution as much as twothirds of the product may be in the form of the formal,
H 2 C(0CHsCMe3 )g , b.p. 182 -185°.
The formal can be hydro­
lyzed by refluxing v/ith alcoholic HCl, while it is not
appreciably changed by steam distillation from 30$ H 2 SO4 .
Contrary to the findings of Hinterberger (1 6 ), Conant,
et al obtained a 59'Jjo yield of neopentyl alcohol and a
55$ yield of trimethylacetic acid by treating trimethyl­
acetaldehyde with 50$ alcoholic KOH containing a little
water after letting the homogeneous solution stand for
one day at room temperature.
Using aqueous alkali they
obtained chiefly unchanged aldehyde and some indefinite
higher boiling condensation products.
Treatment of trimethylacetaldehyde with n-propyl
Grignard gave only a trace of neopentyl alcohol.
Iso­
propyl magnesium bromide gave a 10$ yield of neopentyl
alcohol while t-butyl magnesium chloride gave a quanti­
tative reduction of the aldehyde to neopentyl alcohol.
Adkins and Folkers l1 9 ) reduced ethyl trimethyla­
cetate with hydrogen under a pressure of 220 atmospheres
o
at 250 in the presence of a copper chromite catalyst to
neopentyl alcohol in 88$ yields.
Whitmore and Rothrock (2 0 ) demonstrated the stabil­
ity of neopentyl alcohol to various reagents.
Heating
o
the alcohol with a crystal of iodine at.230 - 240 for
48 days produced no change.
The alcohol was stable to
o
anhydrous K2C03 for seven days at 225 .
Heating the al­
cohol with a trace of dry HCl at 200° for two weeks pro­
duced no change.
With cold conc. sulfuric acid a stable
acid ester was formed.
No trace of chloride was detected
when neopentyl alcohol, which had been saturated with dry
HCl at 10°, was heated in a sealed tube at 62 - 65° for
206
hours.
chloride.
Thionyl chloride and pyridine at 0° gave no
Treatment of neopentyl alcohol with 48% HBr
gave no bromides.
at 40
o
Heating with dry HBr in a sealed tube
for weeks gave no bromides while treatment at 65
o
for forty days produced bromides b.p. 22 - 32° at 23 - 35
mm, which were not identified.
Ingold (2 1 ) reported that neopentyl iodide is obtained
by the treatment of neopentyl alcohol with red phosphorus
and iodine.
He also prepared trimethylneopentyl ammonium
hydroxide and found that upon the distillation of it, no
neopentyl alcohol was formed.
Only methyl alcohol was
formed, no gaseous products being obtained.
The Raman spectrum of neopentyl alcohol has been de­
termined and reported by Kohlrausch and Koppl (2 4 ).
The reducing action of Grignard reagents 011 pivalyl
chloride to form neopentyl alcohol has been investigated
rather thoroughly in this laboratory.
Greenwood, Whitmore
and Crooks l2 2 ) found that when t-butyl magnesium chloride
was added to a large excess
of pivalyl chloride at 10
0
an 8 % yield of the neopentyl ester of trimethylacetic acid
was obtained.
When pivalyl chloride was added to a large
0
excess of t-butyl magnesium chloride at 40 , a 94% yield
of neopentyl alcohol was obtained.
Whitmore (2 3 ) reported that the addition of pivalyl
chloride to excess n-butyl magnesium bromide gave a 27%
yield of neopentyl alcohol.
This is what could be ex­
pected in view of the results of Conant
011
action
of various Grignard reagents on trimethylacetaldehyde
to yield neopentyl alcohol.
Rice i8 5 ) prepared derivatives of neopentyl alcohol.
The a-naphthyl urethan melts at 99 - 100°, the acid
o
phthalate at 70 - 71 and the acid tetrachloroplatinate
o
at 140 - 141 . The latter was prepared by refluxing
tetrachloroplatinic acid with a slight excess of neo­
pentyl alcohol in benzene, removing the water as formed.
Ginnings and Baum (2 6 ) determined the aqueous sol­
ubility of the eight amyl alcohols at 20, 25, and 30° C.
They found the tertiary alcohol more soluble than the
three secondary alcohols which were more soluble than
the four primary alcohols.
Also the solubility of the
alcohols increases as the OH group approaches the center
of the molecule.
Limitation of the comparisons to either
the primary or secondary group reveals that the more com­
pact the molecular structure, the greater the aqueous
solubility of all eight isomers was found to decrease as
o
the temperature increases from 20 to 30 .
Magnani and McElvain l2 7 ) heated one mole of sodium
neopentylate with four moles of neopentyl benzoate at
175 - 180° for two to three hours.
From the reaction mix­
ture they obtained 1.9 moles of unreacted ester, 0.86
moles of benzoic acid, 0.83 moles of neopentyl alcohol,
0.14 moles of benzyl benzoate, 0.12 moles of benzyl tri­
methylacetate, 0.8 moles of neopentyl trimethylacetate,
0,12 moles of trimethylacetaldehyde, and 0.14 moles of
trimethylacetic acid, leaving a residue amounting to
12.3# of the weight of the reacting ester.
Wittle (2 8 ) obtained neopentyl iodide in 6 - 14#
yields from the treatment of neopentyl alcohol with red
phosphorus and iodine at room temperature.
Neopentyl al­
cohol which was saturated with dry HI in a salt-ice bath
was unchanged at the end of three months of standing at
room temperature.
Later Whitmore (2 9 ) and Wittle (3 0 ) reported the
preparation of neopentyl iodide in 4 - 9# yields from
the treatment of neopentyl alcohol with red phosphorus
and iodine.
The iodine was added to a mixture of the al­
cohol and red phosphorus, and after standing for one week
at room temperature, the mixture was refluxed for six
hours and then the products worked up.
Karnatz (3 1 ) of this laboratory treated neopentyl
alcohol with thionyl chloride in the presence of pyridine.
He reported a 25# yield of chlorides from which he iden­
tified 4.6# of neopentyl chloride.
Wittle (3 0 ) treated neopentyl alcohol with phos­
phorus pentachloride and obtained only products boiling
above 100°, thus no neopentyl chloride was formed.
The
distillate from this reaction reacted vigorously with
water, then settled to the bottom as a viscous fluid
heavier than water.
It was not investigated further.
His
neopentyl alcohol was prepared in a 40# yield (based on
the t-butyl chloride) by passing formaldehyde gas into
trimethylacetic acid, leaving a residue amounting to
12.3# of the weight of the reacting ester.
Wittle (2 8 ) obtained neopentyl iodide in 6 - 14#
yields from the treatment of neopentyl alcohol with red
phosphorus and iodine at room temperature.
Neopentyl al­
cohol which was saturated with dry HI in a salt-ice bath
was unchanged at the end of three months of standing at
room temperature.
Later Whitmore i3 9 ) and Wittle (3 0 ) reported the
preparation of neopentyl iodide in 4 - 9# yields from
the treatment of neopentyl alcohol with red phosphorus
and iodine.
The iodine was added to a mixture of the al­
cohol and red phosphorus, and after standing for one week
at room temperature, the mixture was refluxed for six
hours and then the products worked up.
Karnatz (3 1 ) of this laboratory treated neopentyl
alcohol with thionyl chloride in the presence of pyridine.
He reported a 25# yield of chlorides from which he iden­
tified 4.6# of neopentyl chloride.
Wittle (3 0 ) treated neopentyl alcohol with phos­
phorus pentachloride and obtained only products boiling
o
above 100 , thus no neopentyl chloride was formed. The
distillate from this reaction reacted vigorously with
water, then settled to the bottom as a viscous fluid
heavier than water.
It was not investigated further.
His
neopentyl alcohol was prepared in a 40# yield (based on
the t-butyl chloride) by passing formaldehyde gas into
t-butyl magnesium chloride.
formed.
Some of the formal was also
Treatment of neopentyl alcohol with thionyl
chloride gave a 65$ yield of neopentyl sulfite which was
hydrolyzed with aqueous KOH to neopentyl alcohol.
Whitmore ^3 2 ) et al, found that neopentyl iodide with
alcoholic KOH at 180 - 190° for 20 hours gave 3 - 5 $ yields
of neopentyl alcohol among other products.
Ethyl neo­
pentyl ether was prepared in a 30$ yield from sodium neopentylate and ethyl iodide.
Refluxing of the ether with
sodium raised the index of refraction apparently with the .
formation of sodium neopentylate.
They also found that
neopentyl iodide reacts with potassium acetate in ethyl
alcohol or acetic acid to give neopentyl acetate in 40 70$ yields.
The acetate is hydrolyzed with aqueous KOH
to give theoretical yields of neopentyl alcohol.
Neo­
pentyl alcohol reacts with acetyl chloride to-give 87$
yields of neopentyl acetate.
DISCUSSION
A. Preparation of neopentyl alcohol.
Of the various reactions for the preparation of neo­
pentyl alcohol reported in the literature, only three ap­
pear to he of much value.
They are (1) addition of methyl
formate to an excess of t-butyl Magnesium chloride, (2)
reduction of pivalyl chloride with t-butyl Magnesium
chloride, and (3) addition of formaldehyde or its poly­
mers to t-butyl Magnesium chloride.
Of these the latter
promised to be best from the standpoint of ease of manipu­
lation, time and apparatus required, and also the cheapest
on the basis of the chemicals required.
Reaction ^1) is
probably only slightly more expensive than reaction 13)
but is more difficult to carry out due to the fact that
soon after the reaction is begun the mixture becomes so
thick that stirring is almost impossible.
Reaction (2)
is expensive from the standpoint of time and chemicals
since the pivalyl chloride must be made from t-butyl
chloride through trimethylacetic acid.
Since gaseous formaldehyde tends to change to the
solid polymer, paraformaldehyde, and is therefore diffi­
cult to handle, the addition of paraformaldehyde to tbutyl Grignard is the easiest method of preparation.
Powdered paraformaldehyde is simply added to the Grignard
reagent at room temperature and the mixture stirred for
about 24 hours.
The yields of neopentyl alcohol are about
40$, with varying amounts of the formal of neopentyl
alcohol as a by-product.
erated from the
Neopentyl
formalby
alcohol can be
lib­
refluxingwith alcoholicHCl.
B. Attempted preparation of neopentyl sulfate.
A method of preparing alkyl sulfates t,3 3 ) has re­
cently been published.
It consists of the reaction of
an alkyl sulfite with sulfuryl chloride to form the sul­
fate according to the equations
2 R-OH
R sS 0 3
ROSOgCl
SOClg -- * R 2 S0 3
+
+
+
+
2 HCl
S0 2 C1 2--- > ROSOgCl
+
RgS0 3
RCl
> RgSO 4
+
RC1
+
+
S0 2
S0 2
Since an alkyl chloride is a by-product in this
preparation, the question arose as to whether or not
any neopentyl chloride would be formed if the reaction
were applied to neopentyl alcohol.
Neopentyl sulfite was prepared in a 64$ yield
according to the first equation, there being no evidence
of the formation of chlorides.
The reaction of neopentyl sulfite with sulfuryl
chloride took place with the formation of much gas (S02 ),
a small amount of alkyl halide which was not identified
but appeared to be t-amyl chloride las evidenced by the
extreme reactivity of the halogen toward aqueous AgN03 ),
some neopentyl alcohol, and a black liquid which dis­
tilled at 5 mm with evidence of decomposition giving a
distillate which contained sulfur.
was not investigated,
The sulfur compound
since there was no evidence for
the formation of any neopentyl chloride, the reaction
was not investigated further.
C. Reaction of neopentyl alcohol with thionyl chloride
in the presence of anhj^drous potassium carbonate and
chloroform.
It has been the experience of this laboratory that
in some cases better yields of alkyl chlorides can be ob­
tained from the alcohol and thionyl chloride in the pres­
ence of anhyd. K 2 C0 3 than can be obtained in the presence
of a base such as pyridine.
In view of this, the possi­
bility that neopentyl chloride could be prepared from neo­
pentyl alcohol and S0C12 in better yields than those of
Karnatz (3 !), presented itself.
The reaction was carried out by adding the S0C1 2 to
the mixture of neopentyl alcohol and K 2 G0 3 in chloroform
while cooling the reaction flask with an ice bath.
The
temperature of the reaction mixture was then gradually
o
raised to 85 . Thirty percent of the neopentyl alcohol
was recovered unreacted and a 54$> yield of neopentyl sul­
fite was obtained.
The remainder of the neopentyl alcohol
was apparently converted to olefins as no chlorides were
found.
D. Dehydration of neopentyl alcohol over A1 2 0 3.
The dehydration of neopentyl alcohol has never been
54
described in the literature.
Ingold (1 7 ) in 1923 re­
ported that Robinson in unpublished work had dehydrated
neopentyl alcohol to yield two amylenes but the method
of dehydration was not given.
Due to the great stability of neopentyl alcohol,
Rothrock (2 0 )» it seemed likely that many of the common
methods used for the dehydration of alcohols were doomed
to failure in the case of neopentyl alcohol.
Catalytic
dehydration over A1203 appeared to be the most promising
method and was accordingly chosen.
'Since neopentyl alcohol has no hydrogen alpha to
the OH group, the dehydration of neopentyl alcohol must
result in the formation of the cyclic hydrocarbon, 1,1dimethyl cyclopropane, di-neopentyl ether, or amylenes
resulting from the rearrangement of the fragment
I
CHS
(#) indicates carbon atom with a sextett of electrons.
*
G-C-C-C
■
> C=C-C-C
+
H
+
H
c
c
(II)
C-C-Cxf
C
(III)
From this scheme it appears that three amylenes are
likely to result from the rearrangement of the neopentyl
fragment, it toeing very unlikely that fragments II or III
would further rearrange to form the straight chain amylenes.
One would also predict that fragment II would lose a
secondary hydrogen predominately to form mostly trimethylethylene and some 2-methyl butene-1.
Fragment III would
be expected to lose a tertiary hydrogen almost exclusively
to form trimethylethylene with a little isopropylethylene
perhaps.
Some present day workers, Taylor (3 4 ) feel that de­
hydration of alcohols over A120 3 involves the intermediate
formation of a cyclopropane hydrocarbon.
While it is pos­
sible that the dehydration of neopentyl alcohol would lead
to the formation of 1,1-dimethyl cyclopropane, it is im­
probable in view of the findings of Ipatiev (35) who re­
ports that 1,1-dimethyl cyclopropane when passed over
o
A1203 at 340 - 345 is changed almost completely to tri­
methylethylene.
Of course it might be argued that 1,1-
dimethyl cyclopropane is formed momentarily and is almost
immediately isomerized to amylenes.
While this possibili­
ty exists, it would be difficult to prove definitely.
Alvarado 1 3 5 ) reports that the dehydration of ethyl
alcohol over Al203 is a two step process,
EtOH
----
>
Et20
----
>
CH2=CH2
and that the time of contact and the temperature determine
the amount of ethyl ether formed.
Senderens (3 7 ) re­
ported that methyl alcohol passed over A1 2 0 3 at 350 370 , gave dimethyl ether, while n-propyl alcohol gave
propylene and a 30# yield of dipropyl ether.
Due to
the absence of alpha hydrogen in neopentyl alcohol, the
dehydration of neopentyl alcohol over A1 2 0 3 might be ex­
pected to lead to dineopentyl ether in light of these two
papers.
However the formation of dineopentyl ether would
probably involve the addition of a neopentylate ion to a
neopentyl fragment.
9
C-C-C* 0 ’
•
»
c
+
9
O’
-O-C
— --- >
•
c
9 . 9
C-C:C:6:C:C-C
1
c
..........
1
c
This is improbable, since the evidence obtained in
the attempted preparation of neopentyl halides from neo­
pentyl alcohol indicates that the neopentyl fragment re­
arranges almost instantly.
Thus it seems probable that
the dehydration of neopentyl alcohol over A1 2 0 3 would
lead to the formation of one of three or perhaps all
three of the amylenes having the isopentane carbon
skeleton.
The A1 2 0 3 catalyst used for this work was prepared
by precipitating A1(0H )3 from Al(N0 3 )3 solution with
ammonium hydroxide, filtering, washing and drying at
360 - 400°.
As no special preoftutions were taken, it
was probably not acid free which is of significance.
Matignon l3 8 ) reports that pure Al 2 0 3 causes the formation
of 1 -butene only from n-butyl
oohol but if traces of
acid are present, it is partly transformed to 2 -butene.
Ipatiev (3 9 ) and Pines (4 0 ) working with n-butyl and
isobutyl alcohols report similar results.
This point
is further illustrated by Komarewsky I4 1 ) who found
that A1 2 0 3 containing H 3 P0 4 dehydrates n-butyl alcohol
to form 1-butene and 2-butene.
Cramer and Glasebrook
(4 P ) found that t-butyl-ethylene remains unchanged when
passed over pure Al 2 0 3 but is isomerized over an acid
catalyst such as A1 2 (S04 )3 .
In view of these facts, it appeared probable that
a mixture (possibly an equilibrium mixture) of olefins
would result from the dehydration of neopentyl alcohol
rather than a single pure olefin.
what happened.
An 81.5
c/o
This is precisely
yield of crude olefin was ob­
tained from the dehydration of neopentyl alcohol over
A1 2 0 3 at 355 - 390°.
This olefin mixture was fraction­
ated, with a column having over 25 theoretical plates
and packed with stainless steel helices, into three
olefins.
The low-boiling olefin which amounted to 4 cjo
of the olefins was ozonized to yield formaldehyde and
isobutyraldehyde, proving it to be isopropylethylene
rather than 1,1-dimethyl cyclopropane.
The intermediate
olefin, amounting to 34 '■jo of the olefins, had physical
constants identical with 2 -methyl butene -1 and the high
boiling olefin, amounting to 62 % of the total olefins,
had physical constants identical with trimethylethylene.
The question then arose as to whether or not this
mixture of olefins was an equilibrium mixture which would
result if any one of these three olefins was passed over
the catalyst, or was it determined solely by the condi­
tions of temperature and time of contact with the c a t a l y s t
In order to gain some light on this subject, it was
proposed to dehydrate t-amyl and isoamyl alcohols over tiies
same catalyst and also to pass two of the pure olefins,
isopropylethylene and trimethylethylene, over the c a t a l y s t
under approximately the same conditions which prevailed
during the dehydration of neopentyl alcohol.
Whether or not the olefin mixture obtained was an
equilibrium mixture cannot be stated definitely in light
of the results which follow on the dehydration of t-amyl
and isoarnjd alcohols and on the isomerization of
isopropyl ethylene and trimethylethylene over Al203 .
To
prove this point it would be necessary to carry out these
reactions at various velocities at the same temperature
and note the effect on the relative amounts of the
three olefins in the resulting mixtures.
E.
Dehydration of t-amyl alcohol over Ala0 3 .
Since the dehydration of t-amyl alcohol and isoamyl
alcohol over Al30 3 as well as the isomerization of isopropylethylene and trimethylethylene over Alg03 were to b e
compared with the dehydration of neopentyl alcohol over
Algo.-*, it was desirable to carry out these reactions at
the same temperature and velocity which was used on
neopentyl alcohol.
However temperature variation was so
difficult to control that it was possible to keep the
temperature only approximately the same as that used
during the dehydration of neopentyl alcohol.
The same
trouble was encountered in regulating the rate of flow
of alcohol and of olefin through the dehydration tube.
T-amyl alcohol was dehydrated over Al203 at 350 400° to yield 80.7
%
of crude olefin which on fraction­
ation was separated into 60.5
30
fo
%
of trimethylethylene,
of 2-methyl butene-1, and 9.5
upwards from 12°.
of material boiling
fo
No satisfactory explanation for this
low-boiling material, which is below that of the lowest
boiling amylene (20°), has been found.
t-amyl alcohol was dehydrated with 15
Some of the same
c/o
H 2S04 .
The
Olefins from this dehydration distilled from 30 - 38°.
Perhaps the low boiling point of the olefins from the
dehydration of t-amyl alcohol over Aly>03 was due to
cracking.
F.
Dehydration of isoamyl alcohol over Al203.
Isoamyl alcohol was dehydrated over A1203 at 350 390° to yield 91.7 $ of crude olefins which were frac­
tionated to give approximately 36
°/o
of isopropylethylene,
25 $ of 2-methyl butene-1, and 39 # of trimethylethy­
lene..
These olefins were not identified by ozonolysis
but simply by their physical constants.
G.
Isomerization of isopropylethylene over A1203.
The constant boiling, constant refractive index
fractions of isopropylethylene obtained from the fraction­
ation of the dehydration products of isoamyl alcohol over
A1203 were used for this experiment.
This olefin was
passed over the catalyst at 360 - 410n and the resulting
olefin mixture fractionated into approximately 12
%
of
.
A
isopropylethylene, 27 $ of 2-methyl butene-1 and 61 fo of
trimethylethylene.
Again the olefins were identified
simply by their physical constants.
H.
Isomerization of trimethylethylene over A1203.
The constant boiling, constant refractive index
fractions of trimethylethylene obtained from the various
fractionations of amylenes from the dehydration of neo­
pentyl, t-amyl, and isoamyl alcohols were used for this
experiment.
This olefin was passed over the catalyst at
375 - 393° to yield olefins which were fractionated into
3.5
fo
of isopropylethylene, 29.5
and 63.5 fo of trimethylethylene.
fo
of 2-methyl butene-1,
Again the olefins were
identified by physical constants.
This fractionation curve resembles very closely the
curve for the olefins from the dehydration of neopentyl
alcohol.
If it is assumed that the slight difference
in the conditions of the two experiments are negligible,
it would appeaf that the dehydration of neopentyl
alcohol over the A120 3 catalyst leads to the formation
of trimethylethylene exclusively, which in turn is then
isomerized by the acid present on the catalyst to the
mixture of the three olefins.
If this is the case, the
dehydration of neopentyl alcohol over acid free A120 3
should yield pure trimethylethylene.
I.
The action of gaseous HBr on boiling neopentyl
alcohol.
The action of HBr in various forms on neopentyl
alcohol as reported in the literature by Tissier (6 ) and
Whitmore (2 0 ) was given in the historical part above.
While Whitmore and Rothrock (2 0 ) believed that approxi­
mately 20
%
of one or more primary bromides was formed
when neopentyl alcohol was treated in a sealed tube at
65° for 40 days with HBr, no primary bromides were
actually identified.
In this experiment it was proposed to pass gaseous
HBr through boiling neopentyl alcohol continuously till
the reaction with neopentyl alcohol was complete, remov­
ing the products as formed.
To accomplish this gaseous
HBr was passed into boiling neopentyl alcohol in a
flask which was attached to a fractionating column.
The
exit HBr was collected in a dry-ice trap in the form of
liquid HBr.
The products were to be drawn off at the
head of the column as formed but this was found to be
impossible due to the presence of two immiscible liquids
in the take-off reservoir at the head of the column, the
bottom layer being concentrated aqueous HBr.
In practice
the aqueous HBr was drawn off from time to time during
the day.
Overnight all of the products were left to re­
flux back into the reaction flask.
At the end of six days
the reaction products were distilled out of the flask
through the column.
When about one half of the charge had
distilled solid neopentyl alcohol collected in the head
of the column indicating that the reaction was not com­
plete.
The distillate was returned to the reaction
flask and gaseous HBr passed through for 15 days more.
The products were hydrolyzed with aqueous alkali to
convert any t-amyl bromide to the alcohol and thus make
separation from neopentyl bromide easier.
The products
were then fractionated yielding 26.7 fo of neopentyl bro­
mide, 19.5 fo of t-amyl alcohol, 13 fo of unreacted neo­
pentyl alcohol, and a residue amounting to 3 fo by weight
of the neopentyl alcohol used up in the reaction.
Since this adds up to only 61
%
of the neopentyl alcohol
used, the question arises as to where the other 39
went.
fo
Unfortunately part of this loss can be accounted
for as being absorbed by the rubber connections used in
the apparatus.
Some of the rubber connections (stoppers,
etc.) were badly attacked by the HBr and organic
liquids, and had to be replaced several times during
the reactions.
Undoubtedly quite a bit of material was
lost in this manner.
In order to get a better material
balance in this reaction it would undoubtedly be
necessary to avoid the use of rubber connections and use
glass connections throughout.
Since t-amyl bromide is not too stable to heat,
refluxing for 21 days would be expected to decompose some
of it into amylenes which on account of their low boil­
ing point would be carried along with
the gaseous HBr.
The neopentyl bromide was identified by preparing
the Grignard reagent which was treated with oxygen to
yield neopentyl alcohol which was identified by its
phenyl urethan.
The Grignard was also treated'with phen­
yl isocyanate to yield t-butyl acetanilide.
The t-amyl alcohol was identified by converting it
to the chloride which formed a Grignard reagent which
reacted with phenylisocyanate to give dimethylethyl
acetanilide.
I.
The action of PBr? on neopentyl alcohol.
Neopentyl alcohol was treated with PBr3 at ice tem­
perature, the temperature being raised gradually to 76°
following the addition of the PBr3 to the alcohol.
This
reaction yielded bromides which were treated with aqueous
alkali to hydrolyze any t-bromide.
Fractionation of the
products yielded 13.1
%
of neopentyl bromide, 18.6
t-amyl alcohol from the bromide, 17.8
%
°/o
of
of unreacted
neopentyl alcohol and quite a large amount of high boil­
ing residue’which could not be distilled without decom­
position iinder reduced pressure.
This material probably
consisted of esters of phosphorus and was not investi­
gated further.
K.
The action of constant boiling HBr on neopentyl
alcohol.
One mole of neopentyl alcohol was refluxed with an
excess of constant boiling HBr for 7 days.
The neopentyl
alcohol was destroyed completely, yielding only a
relatively small amount of liquid products which could
not be separated into any pure fractions by fractiona­
tion.
The greater part of these products distilled above
118°, the boiling point of neopentyl alcohol, and con­
tained a halogen compound which would not form a Grignard
reagent.
There was no evidence of any neopentyl bromide
being formed.
The reaction was not investigated further
to determine what the products of the reaction were.
The treatment was perhaps too drastic for any neo­
pentyl bromide which might have formed in light of the
results to be presented later on the stability of
neopentyl bromide to refluxing with an HBr - H2S04 solu­
tion.
L.
The action of a solution containing HBr, H 2S04 , and
NaBr on neopentyl alcohol.
One mole of neopentyl alcohol was refluxed for
18.5 hours with a solution of HBr and H2S04 (prepared
from Brg S02 and ice) to which sodium bromide was added.
The organic products were separated and hydrolyzed with
aqueous alkali to hydrolyze any tertiary bromide.
Frac­
tionation of the products indicated a yield of 4-1 C
/S of
t-amyl bromide.
fied.
No neopentyl bromide could be identi­
No unreacted neopentyl alcohol was recovered.
A
small amount of halogen compound boiling higher than
t-amyl bromide was obtained which did not form a
Grignard reagent.
The reaction was not investigated fur­
ther.
M.
The stability of neopentyl bromide to a solution of
HBr and H2S04 .
Neopentyl bromide (10 grams) was refluxed with a
solution of HBr and H 2S04 (prepared from Br2 , S02 , and
ice) for two days.
At the end of this time the reaction
mixture had assumed a black color indicating that some
reaction had taken place.
From the reaction mixture 8.7
g. of crude bromide were recovered.
Distillation of this
material gave 4.E g. of distillate which gave a Grignard
reagent which when treated with phenyl isocyanate gave
t-butyl acetanilide.
tyl bromide.
It was therefore unchanged neopen­
It is apparent however that neopentyl
bromide is not entirely stable under the conditions of
the experiment.
N.
The stability of neopentyl stearate to heat.
The pyrolysis of esters is of value in certain in­
stances for the dehydration of alcohols.
Thus the dehy­
dration of pinaoolyl alcohol yields rearranged olefins,
but the pyrolysis of pinacloyl acetate is perhaps the
best method for the preparation of t-butylethylene.
In
view of the stability of neopentyl alcohol, it seemed
very likely that a high molecular weight ester of neopen­
tyl alcohol such as neopentyl stearate would distill at
atmospheric pressure without decomposition.
Neopentyl stearate was prepared by treating stearoyl
chloride with neopentyl alcohol.
It was found to boil at
359° at 730 mm with little or no evidence of decomposi­
tion, thus giving further evidence of the extreme stabil­
ity of neopentyl alcohol.
0.
Treatment of neopentyl alcohol with hot conc.
H r , S O
4
.
Tissier (5 ) and Whitmore and Rothrock (20) reported
that neopentyl alcohol on treatment with cold conc. H 2S04
gave a stable acid sulfate.
Since sulfuric acid is often
used as a dehydrating agent for alcohols the question
arose as to what would happen if neopentyl alcohol were
heated with conc.
H
r , S 0 4
.
Neopentyl alcohol (0.25 moles) was heated by means of
0
an oil bath at 183
9 hours.
with 0.125 moles of conc. H2S04 for
A liquid distillate which smelled of sulfur
dioxide and which consisted of two layers was obtained,
together with a small residue of black tar.
A small
part of the organic portion of the distillate distilled
o
at 32 - 38 and was probably a mixture of amylenes.
Dineopentyl sulfate probably formed which was un­
stable at the high temperature to which it was subjected
and decomposed into sulfur dioxide and amylenes among
•other products.
P.
The reaction was not investigated further.
The effect of heat on a. mixture of neopentyl alcohol
and sodium neopentylate.
In 1898 Guerbet reported that n-propyl alcohol when
heated with an equimolecular amount of sodium n-propylate
o
under pressure at 250 undergoes a change, which he said
was characteristic of primary and secondary alcohols, in
which the net result is the removal of ONa with an alpha
hydrogen from another molecule to form 2-methyl pentanol1 and NaOH.
alcohol.
More vigorous treatment will produce tripropyl
Many alcohols when heated with soda lime yield
salts of the corresponding acids.
The question arises as
to whether or not neopentyl alcohol will give reactions
similar to these or is ifc too stable to give such reac­
tions?
One gram of sodium was heated with ten grams of neo­
pentyl alcohol in a bomb tube till nearly all of the
sodium had dissolved.
The tube was then sealed and
heated for 34 hours at 220 - 240°.
There was no evi­
dence of any reaction, the neopentyl alcohol being re­
covered unchanged.
EXPERIMENTAL
A. Preparation of neopentyl alcohol.
Two methods were used,
(1) from t-butyl Magnesium
chloride and methyl formate, and (2) from t-butyl Mag­
nesium chloride and paraformaldehyde.
The latter method
was found to be the most satisfactory and most of the
neopentyl alcohol used in these experiments was prepared
by the latter method.
However both methods will be des­
cribed.
(1). From t-butyl Magnesium chloride and methyl
formate.
Tert. butyl Magnesium chloride was prepared in the
usual manner in a 3-L 3-neck flask from 120 g. (5 moles)
of magnesium turnings and 463 g. of t-butyl chloride in
anhydrous ethyl ether.
The resulting solution of Grig-
nard reagent (2600 cc. ) contained 3.33 moles of the Grignard reagent as determined by titration.
Methyl formate, Eastman ^ 1227, (118 g . , 1.97 moles)
was added to the Grignard reagent from a dropping funnel
with vigorous stirring during six hours.
Soon after the
addition of the ester was begun the reaction mixture be­
came very thick due to the formation of a grey precipi­
tate, stirring becoming very inefficient as a result of
it.
Ether (600 cc.) was added to the reaction mixture
during the addition of the ester in order to aid stirring.
Throughout the addition of the ester there was a
vigorous evolution of gas.
11
Stirring was continued for
hours after the addition of the ester was complete.
The reaction mixture was divided into two parts and
decomposed hy the addition of 150 g. of cracked ice to
each part.
The clear ether layers were decanted and the
solid residues were each extracted twice with 400 cc. por­
tions of ether.
The Mg(0H )2 residues were then dissolved
by the addition of HC1, the small ether layers separated,
and the aqueous solution extracted with 500 cc. of ether.
The ether solutions v/ere combined in a 5-L flask and the
ether removed by distillation on the steam bath through a
110 cm indented column.
When about one-half of the ether
had been distilled off the residue was combined with a sim­
ilar neopentyl alcohol solution residue obtained in a simi­
lar manner from 3350 cc. of 2.16 normal t-butyl Grignard
and 220 g. (3.67 moles) of methyl formate.
The distilla­
tion of ether was then continued till a residue of about
750 cc. remained.
This material was dried with 20 g. of
anhyd. MgS0 4 and fractionated v/ith column ^2.
Yield, 377
o
g. (4.28 moles), of neopentyl alcohol b.p. 110.5 - 112 at
737 mm, 40 .570 based on the Grignard used.
(2).
From t-butyl Magnesium chloride and paraform­
aldehyde ♦
Several runs v/ere made, but only a single run will
be described.
To 3.7 moles of t-butyl Magnesium chloride
^prepared from 5 moles of magnesium and 5 moles of t-butyl
chloride which gave 2500 cc. of 1.48 normal Grignard
reagent) filtered through glass wool to remove the ex­
cess magnesium, 111 g. (3.7 moles of HCHO) of powdered
Eastman paraformaldehyde were added during 2 hours and
40 minutes with vigorous stirring.
After stirring con­
tinuously for 36 hours the reaction was considered com­
plete.
The reaction mixture was decomposed in the same man­
ner as described under the methyl formate method above.
The ethereal solution of neopentyl alcohol obtained was
combined with the ethereal solution of neopentyl alcohol
obtained from 2150 cc. of 2.7 normal t-butyl Grignard
(5.8 moles) and 176 g. (5.9 moles of HCHO) of paraform­
aldehyde in the same manner as just described.
The ether was distilled off through a 110 cm in­
dented column till a residue of about 1700 cc. remained.
This solution was dried with 30 g. of anhyd. MgS0 4 and
fractionated with column y2 to yield 397 g. (4.51 moles)
o
.
of neopentyl alcohol b.p. 110 - 111 at 731 mm (47.5%
based on the Grignard used) and a residue of 63 g. which
consisted of the formal of neopentyl alcohol.
This resi­
due was combined with other similar residues and fractionated to obtain the pure formal b.p. Ill
o
20
at 90 mm, n^
1.4080.
B. Attempted preparation of neopentyl sulfate.
(1). Preparation of neopentyl sulfite.
The procedure followed was taken from Organic
Syntheses, Vol. XIX, page 29.
0. 6
Neopentyl alcohol (53 g. ,
moles) was placed in a 200 cc. 3 -neck flask fitted
with a mercury sealed stirrer, reflux condenser and a
dropping funnel.
Eastman Practical thionyl chloride
(33 g., 0.28 moles) was added with stirring from the
dropping funnel during 35 minutes.
The flask was cooled
o
with a water-bath at 25 during the addition of the first
half of the S0C12 .
began.
At this point the evolution of gas
The water-bath was removed and the flask was
heated just enough so that it remained warm to the hand
(about 40° C) during the addition of the remainder of
the S0C12 .
The dropping funnel was then replaced with a
thermometer which extended down into the reaction mixture.
While stirring the temperature was raised during 45 mino
utes to 148 and held there for 3 minutes.
There was
gentle refluxing at this point and the reaction mixture
darkened slightly.
The reaction mixture was transferred to a 125 cc.
Glaisen flask and vacuum distilled.
Vapor Frac •Wt.
Time
Bath
11; 30
11; 47
11; 55
12; 30
11 0
72
137
141
190
112
11 0
137
1
2
4.2
3
4
9.0
28.7
2.6
Index
solid
1.4219
1.4262
1.4242
Pressure
12
mm
24
16
12-89
Neopentyl sulfite is apparently unstable to heat,
as there was evidence of decomposition.
The pressure in
the system rose slowly as the temperature of the bath was
raised.
Residue about 1 g. of black tar.
neopentyl alcohol.
Fraction 1 was
12). Treatment of neopentyl sulfite with sulfuryl
chloride.
The procedure used was taken from Organic Syntheses,
Vol. XIX, page 27.
Neopentyl sulfite 140 g., fractions 2,
3, and 4 above) was placed in a 200 cc. 3-neck flask fitted
with a mercury sealed stirrer, reflux condenser and a
dropping funnel.
Provision was made to pass the evolved
gases through NaOH solution to remove S0 2 and HC1 and the
remainder of the gas, if any, through a trap cooled with
an ice-conc.HCl mixture in order to catch any amylenes
which might be formed in the reaction.
Sulfuryl chloride (12.2 g. "Eastman" P 322) was
placed in the dropping funnel and added dropwise with
stirring and cooling to the neopentyl sulfite during 23
minutes.
The flask was then heated with an oil bath,
raising the temp, very slowly. When the temperature
0
reached 1 2 2
the solution began to reflux slowly and a
rapid stream of gas v/as evolved.
When the temperature
o
reached 130 the condenser was set downward for distillao
tion.
The temperature was raised to 142 and held there
till no more distillate or gas v/as obtained.
late amounted to 9.5 g.
The distil­
No liquid v/as caught in the trap.
A black residue remained in the reaction flask.
The distillate was washed with Na 2 C0 3 solution and
20
then dried over anhyd. K 2 C0 3 leaving 8 g. of liquid, n^
1.4053.
This material when tested v/ith aqueous 5c/o AgN0 3
gave a precipitate thus resembling t-amyl chloride rather
than neopentyl chloride.
.
20
pentyl chloride are n d
The physical constants of neo1.4040, b.p. 84
o
at 740 mm.
This
material was placed in an 18 cc. distilling flask and
distilled.
iction
Wt.
1
2
3
4
5
6
7
8
0 . 5 g.
0 .5
1.0
0 .9
0 .9
1.0
1.0
0 .2
B. P.
0
75
85
95
101
1 06
111
116
---
Pressure
7 19 mm
it
it
tt
tt
tt
it
tt
Residue, 0.2 g. of dark liquid.
A G-rignard reagent could not be prepared from
fraction 4 which gave a test for halogen with AgN0 3
(5°/o aqueous).
ride.
It probably contained some t-amyl chlo­
Fraction $7 solidified on cooling and was mostly
neopentyl alcohol.
It appears that very little if any
neopentyl chloride was formed.
The black reaction residue
was diluted with ether
and washed with Na 2 C0 3 solution and water.
The black
ethereal solution was dried with 6 g. of anhyd. CaCl 2
for one day and then with anhyd. K 2 C0 3 for one week.
solution was filtered .into a 50 cc. Claisen flask.
The
After
distilling off the ether the black residue was vacuum
distilled.
While heating with a water bath, 4.7 g. of
solid neopentyl alcohol distilled over at a pressure of
106 mm.
The residue was then distilled at 5 mm pressure.
When the oil bath temperature reached 200° the pressure
rose to 8 mm indicating a slight amount of decomposition.
The distillation was stopped at this point, the vapor
o
temperature being 98 . 1.8 g. of yellow distillate were
obtained leaving a tarry residue of 3.5 g.
The presence
of sulfur in the distillate was shown by means of a sodium
fusion.
Addition of lead acetate solution to the solution
from the sodium fusion gave a black precipitate of lead
sulfide.
Since the amount of this material obtained was
so small it was not investigated further.
G. Treatment of neopentyl alcohol with thionyl chloride
in the presence of anhyd. KgCO^ and CHGl^.
A 500 cc. 3-neck flask containing 69 g. (0.5 moles)
of anhyd. K 2 C03 , and fitted with a mercury sealed stirrer
was heated to drive off all the water which may have been
taken up by the carbonate.
After the flask had cooled,
44 g. (0.5 moles) of neopentyl alcohol and 60 g. (0.5
moles) of dry CHC1 3 were added and the flask fitted with
a dropping funnel and reflux condenser.
72 g. (0.6 moles)
o
of pure S0C12 , b.p. 75 - 75.5 at 732 mm, were added to
the reaction mixture with stirring during 1 hour and 50
minutes, the evolved gases being removed through a water
trap.
During the addition of the S0C1 2 the reaction flask
was cooled with an ice bath.
The temperature was allowed
to rise to room temp, during the next two hours and was
o
then heated on a water bath to 85 during another two hour
o
period. After holding the temperature at 85 for another
hour the reaction mixture was allowed to cool.
When cold,
the excess S0C1 2 was destroyed by adding 25 cc. of water
slowly with stirring, after which 175 cc. of water was
added rapidly.
Stirring was continued for 24 hours to
hydrolyze any possible t-amyl chloride.
layer remained alkaline.
The aqueous
The layers were separated and
the aqueous layer extracted with two 25 cc. portions of
ether which were added to the CHC1 3 solution.
The solu­
tion was dried with 10 g. anhyd. K 2 C0 3 and fractionated
with column #3.
Time
Col. Vapor
1; 30
3; 05
5; 30
54
54
63
10; 15
10; 48
11; 45
89
92
105
5; 00
2; 45
Fract.
34
59. 5
61
1
2
61
91
4
5
110
132
112
3
Wt.
4.3 g
24.2
Index
Pressure
ether
1.3936
1.4400
730 mm
it
ii
1.0
1.4378
6
0.5
1.5
1.4061
IT
It
II
112
7
12.0
solid
735
90
8
30.0
1.4273
There was no residue.
6
Fraction $ 8 gave a test for
sulfur and it seemed probable that it was neopentyl sul­
fite.
A portion of fraction f/8 (5 cc. ) was placed in a
50 cc. r. b. flask and hydrolyzed by stirring for three
days with a solution of 5 g. of KOH in 25 cc. of water
while heating on a steam bath.
When the reaction mixture
had cooled the top organic layer solidified.
It was
separated and weighed, wt. 3.2 g.
A phenyl urethan was
o
prepared in the usual way, m.p. 112 - 113 , mixed m.p.
112 - 113° using phenyl urethan of neopentyl alcohol.
A test on the aqueous layer from the hydrolysis revealed
the presence of sulfite ion, thus proving fraction y 8
to he neopentyl sulfite.
Fractions # 6 and 7 were neoo
pentyl alcohol, phenyl urethan m.p. 112 - 113 . From
the fractionation table it can be seem that very little
if any alkyl halides were formed.
D. Dehydration of neopentyl alcohol over A 1 P0 ^.
(1) Description of the apparatus.
A diagram of the dehydration tube is shown in Fig.
1.
It consisted of a pyrex glass tube, 44 x 1.8 cm.,
with a side arm large enough to admit a thermometer at
the center of the tube.
One end of this tube was sealed
to a dropping funnel provided with a drop counter.
The
other end of the tube was open and connected to a conden­
ser by means of rubber stoppers and glass tubing.
The
top of the dropping funnel was connected to the exit end
of the tube by means of rubber tubing so as to equalize
the pressure at both ends.
The dehydration tube was
filled with Al 2 0 3 (2 to 8 mesh) prepared as described
below.
The tube v/as v/ound with ribbon chromel resistance
wire ^about one-eighth inch betv/een windings) for heating.
The tube was insulated by means of an asbestos covering
about one-half inch thick.
Since neopentyl alcohol is a
solid, provision had to be made to keep it in the liquid
state in the dropping funnel.
This was accomplished by
winding the dropping funnel and dropping funnel stem v/ith
ribbon chromel resistance wire for heating electrically.
(2) Preparation of the Al 2 0 3 catalyst.
Aluminum nitrate (750 g . , 2 moles, Al(N0 3 )3 .9H 2 0 )
was dissolved in 4 liters of water and 500 cc. of conc.
NH4 0H added to precipitate the A1(0H)3 .
The mixture was
shaken vigorously and then filtered with suction on a
large Buchner funnel.
The A1(0H )3 was washed on the
Buchner funnel with about 2 liters of distilled water.
The A 1 ( 0 H )3 gel was placed in an oven at 90° on a por­
celain plate.
After drying for 6 days at 90 - 115° in
the oven, the glassy Al 2 0 3 was broken up and screened
over 8 mesh wire cloth, discarding that which went through.
The dehydration tube described above was filled with
o
A1 2 0 3 particles larger than 8 mesh and heated to 360
while passing a slow stream of nitrogen through it.
White fumes as well as oxides of nitrogen came out of
the tube indicating decomposition of ammonium salts still
present in the A1 2 0 3.
When no more fumes were evolved,
heating was stopped and the catalyst removed from the
tube.
It had shrunk to about one-third of its original
volume.
This process was repeated until enough catalyst,
which had been freed from ammonium salts, was obtained to
fill the tube with 3 - 8
mesh alumina.
(3) Trial dehydration of n-butyl alcohol over A1 2 0 3 .
n-Butyl alcohol was dehydrated by means of the appa­
ratus just described, as a test to see if it worked as it
was supposed to.
After having swept out the tube with
nitrogen gas at 360° for 50 minutes, 75 cc. (0.82 moles)
of stock n-"butyl alcohol,
20
1.3998, which had been
dried over anhyd. MgS04 , was placed in the dropping fun­
nel of the dehydrator and passed over the catalyst during
one hour and 13 minutes at 353 - 377°.
were collected in a dry ice trap.
was 44 g. , 0.78 moles,
95fo
water were also collected.
The olefins formed
The yield of olefins
of the theoretical.
15 cc. of
It thus appeared that the
apparatus was working satisfactorily.
(4) Dehydration of neopentyl alcohol.
Run
ffl
.
This was a trial run made on a small amount
of neopentyl alcohol to find out if neopentyl alcohol
would be dehydrated under the conditions used on n-butyl
alcohol above, before making a run on a large amount of
the alcohol.
. A 1E5 cc. distilling flask surrounded by an ice bath
served as a receiver for the products of dehydration.
The
side arm of the distilling flask was connected to a dry
ice trap to catch any olefins not condensed by the condeno
ser. Water at 0 was circulated through the condenser by
means of a small motor-driven centrifugal pump.
After sweeping out the dehydrator with nitrogen at
360° for 1 hour, 16 g. (0.182 moles) of neopentyl alcohol
were placed in the dropping funnel and passed over the
catalyst during 20 minutes.
When the passage of neopentyl
alcohol was begun the temperature of the dehydration tube
rose rapidly from 355 to 382° and varied from 375 to 385°
during the period of dehydration.
However the tempero
ature stood at 582 for the greater part of the time.
A slow stream of nitrogen was passed through the tube
for 15 minutes after the dehydration was complete in
order to drive out all of the gaseous products.
The
products consisted of 13.5 cc. of upper olefin layer and
3.5 cc. of water.
Very little liquid was caught in the
dry ice trap.
Run #2.
Neopentyl alcohol (87.5 g . , 0.994 moles)
was dehydrated during 1 hour and 26 minutes in the man­
ner described above.
The temperature of the dehydration
tube was 338° when the dehydration was begun.
It rose
rapidly to 380°and varied from 370 to 390° during the
dehydration.
The temperature stood at 385° for the
greater part of the time.
At the end of the dehydration
the system was swept out with nitrogen for 5 minutes.
The products consisted of 18 cc of water and an olefin
layer (59.5 g. ) which was dried with a little anhyd.
K 2 C03 , The dried olefins (64.0 g) from runs //;1 and 2
were combined in a 250 cc. flask containing a few boil­
ing chips and 1 g. of anhyd. K 2 C0 3 and fractionated with
column
f t 3.
Ice water was circulated through the conden­
sers and fraction cutter by means of a small centrifugal
pump.
A dry-ice trap was attached to the fraction cut­
ter to prevent loss of olefins.
Bath
Col.
Yapor Fract.
1; 15
3; 40
6; 30
9; 50
12;15
40
40
43
43
43
30
31
3 1.5
31
31
22-24
2 8 .5
3 1 .5
3 2 .5
33
1
2
3
4
5
1 .6
1.9
2.3
2.7
4.0
5; 15
8;00
11; 30
47
48
48
3 2 .5
33
33
3 3 .5
■ 34
3 4 .5
6
7
8
2 .4
3 .5
4 .2
1.3816
1 .3824
3;00
5; 30
9; 45
49
50
50
34
35
35
35
3 5.5
36
9
10+
11
3 .5
1.0
3.0
1 .3 8 3 4 727
it
it
1.3845
10; 00
12; 15
49
49
35
35
3 6.7
37
12
13
2 .4
3 .2
1 .3852 728
it
1 .3859
12;00
51
35
38
14
4 .2
1 .3 8 6 3 730
11; 00
51
35
38
15
5 .1
1 .3868
11; 05
12; 10
51
86
35
37
3 8 .3
33
16
17
8 .4
3.1
1 .3 8 7 1 733
it
1.3871
Time
Weight Index
Pressure
__
7 3 2 mm
it
it
1.3761
u
1 .3783
n
1.3800
__
tt
n
it
it
20
Residue 2.6 g . ,
1.4037.
A liquid fraction (0.6 g)
was caught in the dry ice trap.
(+) Due to a stopcock
leak, some of this fraction was lost, the loss being es­
timated as 2.0 g.
The results of this fractionation are
shown graphically on Fig. 2.
From the curve it can be
seen that a more efficient column is necessary for the
separation of the olefins in the mixture.
Run $3.
Neopentyl alcohol (267 g . , 3.04 moles) was
dehydrated during 4 hours and 20 minutes in the manner
described under run ^1 above.
The dehydration was begun
o
with the dehydration tube temperature at 358 . The temper­
ature rose quickly
to 387°, averaged around 375° during
the dehydration but varied from 355 to 390°.
The products
41
8
Hat}
w
cl "0
Forcont of diotillnto
consisted of 53 g. of water and an olefin layer of 173 g.
which was dried over 20 g. of anhyd. K 2 C0 3 in the ice-box.
The dried olefins were placed in a 500 cc. flask with a
little anhyd. K 3 C0 3 and fractionated with column E.M. I.
Ice-water was circulated through the condensers and frac­
tion cutter by means of a centrifugal pump and the fraction
cutter was attached to a dry-ice trap to prevent loss of
olefins.
Time
Bath Col.
Vapor
Fract . Wt.
Index
5 15
7 00
12 00
38
40
40
30
31
31
19-20
23
28
1
2
3
2.2
2 .3
2.0
1.3657
1.3 6 9 4
1
2
3
4
6
9
11
05
30
50
25
00
10
30
40
40
41
44
42
44
45
32
32
32
33
33
33
34
3 0 .5
31
31.5
32
32
32
32
4
5
6
7
8
9
10
3 .8
4 .9
6.0
6.4
5.7
8.0
5 .4
1.3766
1 .3 7 76
1.3780
1.3790
1 .3790
1 .3786
1.3790
11
2
4
7
10
11
50
10
25
50
00
40
47
47
47
48
49
49
35
35
36
36
36
36
32
32
35
37
38
38
11
12
13
14
15
16
4 .3
2.7
3 .3
3 .8
3 .4
4 .1
1.3 7 9 3
1.3797
1.3813
1.3848
1 .3 8 6 8
1 .3872
9
10
10
10
11
11
12
12
2
45
10
30
55
20
45
07
55
20
49
49
49
49
49
50
56
60
92
36
36
36
36
36
36
36
36
36
38
38
38
38
38
38
38
38
3 7 .5
17
18
19
20
21
22
23
24
25
4 .9
6 .4
8 .0
1 1 .5
1 1 .2
1 4 .9
13.9
5 .1
4 .0
1.3870
1.3 8 7 0
740 mm
735
IT
1!
n
it
IT
ii
tr
One gram of liquid /was caught in the dry-ice trap,
Residue 7.6 g. of yellow liquid ngo 1.3989.
A phenyl
*irethan m.p. 112 - 113°, mixed m.p. 112 - 113° using
phenyl urethan of neopentyl alcohol, was prepared
in the
usual manner from the residue.
The residue therefore con­
tained some neopentyl alcohol.
The results of this frac­
tionation are shown graphically on Fig. 3.
Fraction ^1 was ozonized as follows;-
Fraction
ff1
was placed in the ozonolysis tube with 35 cc. of solvent.
The tube was placed in a Dewar flask and packed in dryice.
Ozone was passed into the solution at a slow rate
during 9 hours and 4=5 minutes.
The end-point of the
ozonolysis was determined by passing the exit gases into
a 5°/o K 1 solution and watching for the sudden color change.
The resulting ozonide was decomposed in the following manner.
A 3-neck 200 cc. flask containing 5 g. of zinc dust,
100 cc. of water and traces of AgW0 3 and hydroquinone.,was
fitted with a Liebig condenser and a dropping funnel.
The
ozonide was placed in the dropping funnel and added dropwise to the boiling mixture in the flask during 20 minutes.
The solvent was collected in a dry-ice trap which was con­
nected to the top of the condenser.
The solvent collected
in this manner was placed in the dropping funnel and run
through again in the same manner.
This was repeated a
third time to make sure the ozonide was completely decom­
posed.
The condenser was then set downward for distillation
and about 60 cc. of water distilled.
Five cc. of this
distillate were added to 75 cc. of a saturated solution
40
30
»*)
H*
013
It
I'eroent cl' d i s t i l l a t e
a 0
o
of dimethyldihydroresoreinol and allowed to stand for SO
minutes.
The precipitate was then filtered off and the
filtrate set aside for further precipitation.
The first
precipitate was recrystallized from methanol, filtered
o
o
and dried, m.p. 189 - 190 , mixed m.p. 189 - 190 with
dimetol of formaldehyde.
Thus formaldehyde was identi­
fied as one of the ozonolysis products.
At the end of an
hour, the second precipitate was filtered off, recrystal­
lized from methanol
\& 0$)
and dried, m.p. 188 - 190°.
The filtrate was set aside and the following day the
third precipitate was filtered off, recrystallized from
dilute methanol and dried, m.p. 131 - 153°.
The amount
of this material was very small so more of it was prepared
by treating 1 0 cc. of the ozonolysis distillate with 200
cc. of saturated dimethyldihydroresorcinol solution in
the manner just described.
The third precipitate obo
tained from this material melted at 130-146 after re­
crystallization from 50fo methanol.
The mixed m.p. with
o
(m.p. 151-152 ) was 125-
dimetol of isobutyraldehyde,
o
'
136 . After another recrystallization from 1 cc. of
o
o
methanol it melted at 185-187 , mixed m.p. 186-190 with
dimetol of formaldehyde.
It thus appeared that formalde­
hyde was the only aldehyde present in the aqueous distil­
late.
The filtrate from the third precipitate was heated
on the steam bath overnight and then cooled with ice.
After standing for another week no precipitate was obtained.
The remainder of the aqueous distillate was extracted
with two 30 cc. portions of ether and the ether distilled
through a small column leaving a residue of 0.5 cc.
This
residue failed to give a 2,4-dinitrophenylhydrazone.
Sinoe only formaldehyde was found in the aqueous
distillate the ozonolysis solvent collected in the dry
ice trap was examined for carbonyl compounds.
The solvent
was distilled through a small packed column leaving a resi­
due of about 1 cc. which was partly water.
A 2,4-dinitro-
phenylhydrazone was prepared in the usual way. from a few
drops of this residue which had an odor resembling an
o
aldehyde.
The crude hydrazone melted at 168-174 . After
recrystallization from 95% ethyl alcohol it melted at 176o
.
o
177 , mixed m.p. 178-161 with 2,4-dinitrophenylhydrazone
of isobutyraldehyde (m.p. 181.5-182.5°).
Isobutyralde-
hyde was thus identified as an ozonolysis product of frac­
tion
fi-1
which therefore must have been isopropylethylene.
The other two olefins, 2-methyl butene-1 and trimethylethy­
lene, obtained from the fractionation of the dehydration
products of neopentyl alcohol were identified by their
physical constants (ICilmer, Penn State Thesis, M.S.).
Dehydration of t-amyl alcohol over A1P0 B ;T-amyl alcohol was purified as follows.
One liter
of stock t-amyl alcohol was refluxed in a 2-L r.b. flask
with 350 g. of CaO for 5 hours.
The alcohol was then dis­
tilled off and refluxed with 21 g. of metallic sodium for
1.5 hours.
The reaction with sodium was very slow.
The
following day the t-amyl alcohol was distilled from the
mixture.
The distillate was fractionated with column ^-4
over a small piece of sodium.
Time
12;45
1; 30
2;00
2; 30
2; 55
3; 15
3; 40
3; 55
Col. Vapor Fract.
Wt.
Index
Pressure
99
14.3
18.7
14.0
106.0
82.0
92.0
106.0
85.5
1.3910
1.4039
1.4038
1.4039
1.4039
1.4039
1.4039
1.4045
731 mm
93
93
93
93
95
94
94
96
100
100
101
101
101
101
102
1
2
3
4
5
6
7
8
u
ii
n
ii
it
ii
T- amyl alcohol (350 cc., 3. 2 2 moles from fractions
4,5,6, and 7 above) were dehydrated during 3 hours and
5 minutes in the same manner that the neopentyl alcohol
was above.
During the dehydration the temperature of
the dehydration tube varied from 350 - 400°, and averaged
o
about 370 . The products consisted of 57 cc. of water
and an olefin layer (182 g . ) which was dried over about
30 g. of anhyd. K 2 C0 3 in the ice-box.
The yield was thus
98.2$ based on th^water formed and 80.7$ based on the
crude olefins.
The dried olefins were fractionated over
a little anhyd. K 2 C0 3 through column E.M.J. using icewater in the condensers and fraction cutter.
A dry-ice
trap was used to prevent the loss of olefins.
Time
Bath Col. Vapor Fract.
Wt.
Index
Pressure
-8 ;50
11; 40
1; 05
39
41
42
29
30
31
12
20
1
2
7.0
4.2
30
3
6.2
1.3721
ii
1 1 ;00
11; 55
47
41
32
32
31
31
4
5
5.5
5.2
1.3771
1.3774
741
734 mm
--
it
ii
1 ;00
2; 40
4; 20
5; 20
6; 25
10; 12
10; 50
42
42
45
45
46
47
47
32
34
34
35
35
35
35
31
3 1 .5
.. 3 2
33
35
37
3 7 .3
6
7
8
9
10
11
12
6 .3
8 .5
9 .5
5 .5
5 .8
5 .9
5 .4
1.3 77 7
1.3778
1 .3 7 8 2
1 .3 7 9 5
1.3 8 1 3
1 .3830
1 .3849
7 41
it
ir
ii
I!
II
11
10; 30
11; 12
12; 10
1; 20
2; 13
2; 27
2; 45
3;05
50
49
49
49
49
49
50
89
36
36
36
36
36
36
36
36
38
38
38
38
38
38
38
58
13
14
15
16
17
18
19
20
6.9
5 .2
8 .6
9.0
1 2 .9
1 4 .3
1 2.6
1 3.0
1.3862
1.3860
1 .3 8 6 4
1 .3869
1.3870
1 .3870
1.3870
1.3870
7 40
,
r-
Residue 4.5 g. ,
in dry-ice trap.
2 0
1.5980.
ii
it
ii
it
ii
it
it
3.2 g. of liquid caught
The results of this fractionation are
shown graphically on Fig. 4.
In view of the fact that the lowest boiling amylene
o
boils at 20 , the formation of the low boiling material
of fraction yl was rather unusual.
Perhaps it was due to
impurities in the alcohol in which case dehydration by
another method such as 15% H 2 S04 might also be expected
to yield this material, or perhaps it was the result of
a slight amount of cracking.
At any rate, it appeared advisable to dehydrate
some of the same t-amyl alcohol by another method.
this purpose
15%
H 2 S0 4 (Kilmer (4 3 ) ) was chosen.
For
T-amyl
alcohol (177.5 g . , 2.02 moles, from fractions 7 and 8
above) was placed in a 1-L r.b. flask with an equal vol­
ume (220 cc. ) of 15% I-I2 S04 .
The flask was fitted with a
12 mm. pyrex tube 2 ft. long and having six 25 mm. bulbs
on it and a side arm at the top.
A thermometer was fitted
40
r p r * '
’
30
H*
(ft
rr
%
::::
t :t
20
10
50
Percent of distillate
75
into the top of the tube and the side arm was fitted with
a condenser through which ice-water was circulated.
A
250 cc. distilling flask surrounded with ice served as a
receiver.
The side arm of the distilling flask was con­
nected to a dry-ice trap.
The reaction flask was heated
with a small flame at such a rate that the olefins pass­
ing over would not distill over above 40°.
At the end
of 3.5 hours a small oil layer remained on the surface
of the H 2 S04 but no more olefin distilled over.
The
crude olefin (123.5 g . ) was dried over 30 g. of anhyd.
K 2 C0 3 giving 111 g. of dried olefin which was fraction­
ated with column IS.M.J.
1; 55
3; 10
4; 10
5; 50
7; 40
9; 56
1 1 ;10
12; 25
Bath
•
l—1
o
o
Time
45
45
45
49
49
51
51
99
34
34
35
35
36
36
36
36
Vapor
Fract.
Wt.
Index
Pressure
1.3783
734 mm
30-31
1
1.0
2.8
1.3798
32
2
1.3798
2. 6
33
3
1.3811
5.1
34
4
36
1.3827
5
6. 0
7.8
37.7
6
1.3851
1.3865
3.5
38
7
38
8
1.3870
70.0
so
A residue of 8.4 g. n^ 1.3960 remained. 0.5 g. of
liquid were caught in the trap.
The results of this
fractionation are shown graphically on Fig. 5.
Dehydration of isoamyl alcohol over A1P0 3 .
Isoamyl alcohol (370 cc. , 3.41 moles b.p. 129° at
733 mm., n|° 1.4063) was dehydrated during 3 hours and 24
minutes in the manner described for neopentyl alcohol.
The temperature varied from 350 - 390° during the
Point
Boiling
Percent of distillate
1
0
dehydration hut averaged about 375 .
The products con­
sisted of 61 cc. of water and 218.5 g. of olefin layer
which was dried over 20 g. of anhyd. K 2 003.
The yield
was 99.5°/o based on the water and 91.7% based on the
crude olefin obtained.
The dried olefin was fraction­
Vapor
25
25
42
55
13
34
34
35
35
39
22
22
23
26
30
20.3
IT
tt
IT
21
1
2
3
4
5
10; 10
11; 45
1; 40
3; 25
4; 45
6; 07
7; 4 5
40
43
43
45
45
46
46
31
32
33
34
35
35
36
22
27
2 9.5
31
32
33
35
11; 30
12; 40
2; 13
3;05
3; 40
4; 00
4; 25
4; 35
46
46
48
49
50
51
76
1 00
36
36
36
36
36
36
■36
38
36
37
38
38
38
38
38
38
12;
2;
4;
6;
8;
Fract.
Index
Pressure
11.0
1 1 .3
12.0
1 0 .8
6.0
1 .3 6 4 5
IT
II
II
1.3650
7 3 5 mm
6
7
8
9
10
11
12
5.0
7 .4
9 .3
8 .3
6.9
6.7
8 .2
1.3647
1.3700
1.3738
1.3 7 6 7
1 .3 781
1.3 7 9 7
1 .3 819
736
13
14
15
16
17
18
19
20
9 .5
2 .9
8 .3
13.0
1 3 .5
10.7
8 .6
2.0
1.3842
1.3850
1.3 8 6 0
1.3870
it
•
Time
■p
Bath
o
o
•
ated with column E.M.J.
11
it
it
it
11
11
11
it
11
n
11
11
11
11
it
IT
It
II
11
it
11
A liquid fraction of 0.5 g. was caught in the dry-ice
20
trap.
Residue 9.0 g. of yellow liquid n^ 1.3991. A
o
phenyl urethan m.p. 53-55 was prepared in the usual
manner from a portion of the residue. A mixed m.p.
o
53-55 was obtained using phenyl urethan of isoamyl
alcohol im.p. 52-54°).
The residue therefore contained
some undehydrated isoamyl alcohol.
The results of this
fractionation are shown graphically on Fig. 6
Bolling
Point
30
20 '
10
50
Percent of distillate
75
G. Isomerization of isopropylethylene over AlpCU.
Isopropylethylene (45 g. , 0.64 moles, fractions 1
to 5 from the dehydration of isoamyl alcohol above) was
passed over the A1 2 0 3 catalyst at 360 - 410° in the same
manner as described for the dehydration of neopentyl alco­
hol except that it was not dropped into the dehydration
tube from the dropping funnel because of its low boiling
point.
Instead, it was distilled from a bottle and passed
into the tube as a gas.
The time required for the passage
of the olefin was 1 hour and 5 minutes.
The recovered
olefin (45 g . ) was dried over 5 g. of anhyd. K sC03 and
fractionated with column E.M. I.
2
35
3 00
4 30
6 00
7 25
8 40
9 55
11 05
11 25
11 27
12 20
12 37
Bath
•
l
—1
o
o
Time
41
42
43
45
49
46
48
53
52
52
54
95
31
32
32
34
34
35
35
35
35
35
35
35
Vapor Fract.
Wt.
17-21
1.8
22
30
31
32.5
34.5
37
38
38
38
38
38
20
Residue 3.2 g. n d
1.3967 .
in the dry-ice trap.
1
2
3
4
5
6
7
8
9
10
11
12
1.7
1.3
2.2
2.5
2.3
2.9
3.2
0.9
1.4
7.6
4.9
Index
™
1.3650
1.3734
1.3767
1.3784
1.3798
1.3832
1.3850
1.3857
1.3858
1.3863
1.5868
Pressure
734 mm
ir
I!
ii
ii
it
it
ti
it
ii
ii
Liquid (1. 4 g. ) was caught
The results of this fractionation
are shown graphically on Fig. 7.
H. Isomerization of trimethylethylene over A1P0 3.
Trimethylethylene (171 g . , 2.44 moles, b.p. 38°,
n|° 1.3870 obtained in the fractionations of the
40-
■-! •
3olline
Point
30 -
?'
10
25
50
Pcrcer.t of distillnto
75
deh3rdration products of neopentyl, t-amyl,
and isoamyl
alcohols over A1 2 0 3 described above, was passed over the
AlgOg catalyst at 375 - 395° during 3 hours and 40 min­
utes in the manner described for neopentyl alcohol.
The
recovered olefin was dried over 20 g. anhyd. K 2 C03.
The
dried olefin ( 1 5 3 . 5 g . ) was fractionated with column E.M.J.
Bath Col.
Time
1;
2;
3;
6;
Vapor
Fract.
wt.
Index
Pressure
7 3 6 mm
tt
IT
II
42
27
55
20
43
41
42
42
32
32
33
32
21-24
28
30
30
1
2
3
4
3 .8
2 .8
4.7
4.7
1.3681
1.3700
1.3749
1.3760
1;40
4; 55
7; 25
1 0 ; 25
1; 35
4 ; 45
7; 47
1 0 ; 50
1;05
2; 25
3; 55
47
45
46
49
48
52
51
48
48
48
88
33
33
34
34
35
35
36
36
36
36
36
31
31
31
31.5
33
35
37
38
38
58
38
5
6
7
8
9
10
11
12
13
14
15
5.0
4 .9
5.0
5 .4
5.7
5 .5
5 .5
5 .5
3.9
2 .4
7 8.0
1 .3 771
1.3774
1.3 7 7 8
1.3780
1.3 7 8 8
1.3809
1 .3 8 5 3
1.3850
1.3860
1.3860
1.3869
20
Residue 4.6 g. n^ 1.3975.
in the dry-ice trap.
It
It
II
It
II
II
11
11
n
it
ii
1.0 g. of liquid was caught
The results of this fractionation
are shown graphically on Fig. 8 .
I. Action of gaseous HBr on boiling neopentyl alcohol.
Neopentyl alcohol (146.5 g . , 1.66 moles) was placed
in a 500 cc. r.b. 3-neck flask which was attached to frac­
tionating column
i f 2.
The flask was fitted with a thermo­
meter and a glass bubbling tube which extended to the bot­
tom of the flask.
The flask was heated on an oil bath
which was heated with an electric hot-plate.
When the
bath temperature had risen to 90° the passage of HBr
40
r -
30
H-
CJtj
•
'eraent of d i s t i l l a t e
rc
00
into the neopentyl alcohol was begun.
The HBr was gen­
erated by dropping bromine upon naphthalene flakes and
passing the HBr through a tube filled with red phosphorus
to remove bromine vapors.
Gaseous HBr was passed into the alcohol continuously
for six days at a slow rate while the bath temperature was
o
held between 120 and 130 . Aqueous HBr was drawn off from
the head of the column from time to time.
At the end of
six days the contents of the reaction flask were distilled
out through the column.
had distilled,
When about one-half of the contents
it was noticed that unreacted neopentyl al­
cohol was still present.
The distillation was stopped and
the distillate returned to the reaction flask.
After stand­
ing at room temp, for 13 days, the flask was again heated
on the oil bath as before while passing a slow stream of
gaseous HBr through the boiling alcohol for 15 days.
aqueous HBr was drawn off from time to time.
HBr was liquefied in a dry-ice trap.
Again
The exit
The liquid HBr ob­
tained in this way was distilled into water leaving a resi­
due of organic material which was later combined with the
rest of the products of the reaction.
The aqueous HBr
which was drawn off during the reaction.was diluted with
water which caused the separation of organic material which
was separated and later added to the other products of the
reaction.
At the end of 21 days of reaction with gaseous HBr
the contents of the reaction flask were transferred to a
500 cc. 3-neck flask containing a solution of 70 g. of
KOH in 200 cc. of water.
The flask was fitted with a re­
flux condenser and a mercury sealed stirrer and the con­
tents stirred vigorously at room temperature for two days
to hydrolyze any tertiary bromide which might have been
present.
The organic material was then separated from the
alkaline aqueous layer and washed with a little water.
aqueous layer was extracted with 100 cc. of ether.
The
After
distilling the ether through column $ 2 , the residue was
added to the main organic layer and dried over 20 g. of
anhyd. K 2 C03 .
The liquid was decanted from the carbonate
into a 200 cc. flask and the carbonate washed with 5 cc.
of dry acetone which was added to the rest of the products
and fractionated with column 7/2 .
Time
Bath
Col.
Vapor
Fract.
Wt.
Index
1
2
3
4
5
4 .0
2 .8
1 .2
2.6
7 .6
1 .3 7 1 3
1.3 69 5
1.3593
1.4129
1 .4 2 0 3
730 mm
1 1.6
16.1
1 5 .2
14.7
1 3.0
1 2 .5
3 .9
4 .8
1 .4 2 1 8
1.4230
1.4237
1.4240
1 .4 2 4 4
1 .4240
1.4214
1 .4094
it
tt
it
tt
tt
it
tt
732
1 0 .5
3 .0
solid
tt
tt
3; 40
4; 55
8;05
9; 55
1 1; 40
130
13 2
140
134
134
65
85
85
92
93
11; 50
2; 25
4; 50
10; 27
12; 00
1; 55
4;00
4; 35
157
133
139
141
146
147
151
1 57
93
93
93
95
96
99
1 00
103
100
it
101
1 01
1 03
1 06
1 10
6
7
8
9
10
11
12
13
2; 10
2; 37
194
2 22
105
113
1 11
105
14
15
31-51v
69
90
98
99
It
tt
Pressure
11
tt
it
tt
Residue 4.5 g.
Fraction ^14 was identified as neopentyl alcohol by
means of its phenyl urethan m.p. 112-1130.
Fraction #7 gave a very faint halogen test with
aqueous AgN0 3 and a somewhat better test with alcoholic
AgN03 .
of gas.
It reacted slightly with sodium with liberation
After standing for several hours the sodium had
a blue coating similar to that obtained in Wurtz reac­
tions.
These tests seemed to indicate the presence of a
neopentyl halide in this fraction.
Fraction
#6
was extracted with 75 cc. of water leav­
ing a heavy bromide layer which after washing with 10 cc.
of conc. H 2 S04 weighed 6 . 8 g.
A Grignard reagent was pre­
pared from this material in the usual manner.
Treatment
of the Grignard reagent with phenyl isocyanate yielded
o
an anilide m.p. 129.5 - 130.5 after recrystallization
from dilute ethyl alcohol.
It gave a mixed m.p. of 130o
o
131 with t-butyl acetanilide (m.p. 131-131.5 ). Neo­
pentyl bromide had thus been identified.
The aqueous extract from fraction 6 was saturated
with K 2 C0g which caused the separation of an organic layer.
This organic material was shaken with conc. HC1 to yield
1.8 g. of a chloride which formed a Grignard reagent.
Treatment of the Grignard reagent with phenyl isocyanate
o
produced an anilide m.p. 89-90 . It gave a mixed m.p. of
o
o
89-90 with dimethylethyl acetanilide (m.p. 91-91.5 ).
Thus t-amyl alcohol was proven to be a constituent of
fraction 6 .
Four cc. of fraction 7 were shaken with 15 cc. of
conc. HC1 producing a material which after drying with
anhyd. K 2 C0 3 and CaCl 2 had a refractive index n^° of
1.4276.
This material formed a Grignard reagent which
was treated with oxygen to produce a material which had
an odor resembling neopentyl alcohol.
This material on
treatment with phenyl isocyanate in the usual manner
o
gave a urethan m.p. 112-112.5 . This urethan gave a
o
mixed m.p. of 112-113 with phenyl urethan of neopentyl
alcohol.
Thus neopentyl bromide was identified as a
constituent of fraction 7.
All remaining portions of fractions 4 to 11 inclu­
sive were combined in a 500 cc. separatory funnel and ex­
tracted with 200 cc. of water.
The crude bromide was then
washed m t h 70 cc. of conc. H 2 SO4. and then twice with 30
cc. portions of conc. H 2 S04 .
The bromide was then treated
with a little anhyd. K 2 C0 3 giving an over-all yield of
20
64.8 g. of neopentyl bromide n^
1.4371 from fractions
4 to 11 inclusive.
The refractive index of fraction 6 lies 0.56 of the
way between that of pure t-amyl alcohol and pure neopentyl
bromide.
0.56 x 11.6
frac. 6 ).
=
6.4 g. (6 . 8 g. were obtained from
On this basis fraction 12 must contain 2.1 g.
of neopentyl bromide bringing the total yield of neopentyl
bromide to 66.9 g. (0.444 moles).
19 g. of neopentyl al­
cohol (fractions 12 to 15) were recovered.
146.5
cohol used.
-
19
=
127.5 g. (1.45 moles) neopentyl al­
The yield of neopentyl bromide was thus 30.6$
based on the 1.45 moles of neopentyl alcohol consumed
in the reaction.
The total weight of fractions 4 to 11
inclusive is 93.3 g.
93.3
cohol.
-
64.8
=
28.5 g. (0.324 moles) t-amyl al­
This corresponds to a yield of 22.3$ of t-amyl
bromide.
Twenty cc. of the neopentyl bromide were placed in
the Cottrell boiling point apparatus at 735 mm. and diso
o
tilled.
Initial b.p. 105 , b.p. 106 when one-half had
distilled.
J. Action of PBr 3 on neopentyl alcohol.
A 500 cc. 3-neck flask containing 93 g. (1.06 moles)
of neopentyl alcohol was fitted with a mercury sealed
stirrer, and a dropping funnel.
The flask was cooled
with a salt-ice bath and 95 g. (0.35 moles) of PBr 3 were
added dropwise to the neopentyl alcohol during 1 hour and
47 minutes.
A few minutes after the addition of PBr 3 was
begun enough of the neopentyl alcohol had melted to permit
stirring.
After the addition was completed the cooling
bath was removed and the reaction mixture allowed to rise
to room temperature during the next two hours.
The flask
was then warmed on a water bath for 5 hours during which
o
time the temperature was gradually raised to 76 . After
standing at room temperature for two days, the flask was
fitted with a condenser for distillation and the reaction
mixture distilled on a water bath (temp, up to 83°) under
a vacuum down to 25 mm.
late were obtained.
In this manner 64 g. of distil­
A residue of 74 g. of yellow liquid
remained.
The distillate was hydrolyzed by stirring at room
temperature with a solution of 20 g. of KOH in 200 cc. of
water.
The aqueous layer was extracted with two 50 cc.
portions of ether which were added to the organic layer and
dried over anhyd. K 2 C0 3 to give 114 g. of dried solution
which was fractionated with column ?/-2 .
Time
Bath
1 2 ;00
1;00
11 0
4; 20
5; 30
10; 45
122
125
129
11; 36
12; 30
1;25
134
150
175
115
Col. Vapor
Fract.
Wt.
35
40
97
98
98
34
37
93-97
97
97.5
1
2
64.5
3
4
5
3.0
2.7
97
tt
6
101
110
IT
7
8
Index
10.8
1.3530
1.3535
1.4102
1.4162
1.4173
10.8
8.2
2.8
1.4181
1.4188
1.4206
1.0
Pressure
731 mm
TT
TT
11
11
IT
TT
1!
Residue 1.5 g.
Fraction $5 had a density greater than that of water
and gave a faint halogen test with alcoholic AgN0 3 thus
indicating the presence of a bromide, possibly neopentyl
bromide.
Fraction 6 was extracted twice with 50 cc. por­
tions of water leaving a bromide which was washed with 15
20
cc. of conc. H 2 S04 to yield 6.4 g. n^
1.4570.
After
treating with a little anhyd. IC2 C0 3 a Grignard reagent
was prepared which on treatment with phenyl isocyanate
gave an anilide m.p. 130-131°.
The anilide gave a mixed
m.p. of 129-1310 with t-butyl acetanilide (m.p. 129-1300).
Neopentyl bromide was tbus identified as a constituent
of fraction 6.
Fractions 3 to 8 inclusive were combined in a separa­
tory funnel and extracted with 100 cc. of water.
The
bromide layer was washed with 43 and 15 cc. portions of
conc. H 2 S04 , then with water and then dried with anhyd.
20
K 2 C0 3 leaving 14.6 g. of neopentyl bromide n^
1.4369.
The total yield of neopentyl bromide was thus 21 g. (0.139
moles) 13.1$ of the theoretical.
The aqueous extract was saturated with K 2 C0 3 to give
6.2
g. of an organic layer which was separated and shaken
with 4 volumes of conc. HC1 to give 6.2 g. of a chloride
20
1.4048
which was dried over anhyd. K 2 C03 .
A Grignard
reagent was prepared which on treatment with phenyl isoo
cyanate gave an anilide m.p. 90-91 . It gave a mixed m.p..
o
o
of 89-91 with dimethylethyl acetanilide (m.p. 89-90 ).
T-amyl alcohol was thus identified.
Yield 17.3 g. (0.197
moles) 18.6$ of the theoretical.
The 74 g. reaction residue above was poured upon 200
g. of cracked ice.
After the ice had melted, 50 cc. of
ether were added and the lower water layer drawn off.
The
ether solution was transferred to a 200 cc. Claisen flask
and vacuum distilled after distilling off the ether.
A
distillate amounting to 20 g. was obtained under vacuum
down to 35 mm.
The last part of the distillate obtained
at this pressure was a solid, probably neopentyl alcohol.
When no more solid distilled, receivers were changed and
distillation was resumed at 7 mm using an oil bath tempero
ature of 166 . A sudden rise in pressure indicated that
decomposition was taking place so the distillation was
discontinued.
Residue 29.0 g.
Distillate 5 g.
The dis­
tillate and residue were not investigated but probably con­
sisted of phosphorus compounds.
The distillate obtained at 35 mm was fractionated
with column $3.
Time
Bath
Col. Vapor
10; 45
11; 10
144
156
112
112
107
109.5
11; 30
1; 30
164
196
112
115
110
112
Residue 1.0 g.
Fract.
Wt.
Index
1
2
2.5
1.2
1.3360
1.3992
3
4
1.7
12.0
solid
n
Pressure
.736 ram
n
ir
ir
Fraction 1 was nearly all water with a
small organic layer on top.
The remainder of the fractions
consisted of nearly pure neopentyl alcohol, phenyl urethan
o
m.p. 112-113 . Thus approximately 16 g. of neopentyl al­
cohol were recovered (0.182 moles) 1 7 . 2 of the original.
K. Action of constant boiling HBr on- neopentyl alcohol.
Neopentyl alcohol (89 g . , 1.01 moles) was placed in
a 1 liter r.b. flask with 500 cc. of constant boiling HBr
(b.p. 124° at 732 mm., 9 normal) and refluxed on an oil
o
bath at 130 - 165 for 7 days. At the end of this time
the condenser was set for distillation and the products
distilled out.
The following morning the layers of the
distillate were separated and the upper organic layer
placed in a 500 cc. flask with a solution of 20 g. of
KOH in 200 cc. of water.
The mixture was stirred vigor­
ously at room temperature for 24 hours to hydrolyze any
tertiary bromide.
The layers were then separated, and
the aqueous layer extracted with two 50 cc. portions of
ether which were added to the organic layer.
The ethereal
solution was dried over 20 g. of anhyd. K 2 C0 3 and frac­
tionated with column #3.
Bath
Time
Col.
Vapor
Fract.
Wt.
Index
Pressure
2; 40
5; 15
99
99
58
60
34
44
1
2
ether
0.9
1.3556
1.3797
731 mm
tt
2; 30
5; 00
9; 15
12;05
114
64
106
3
4
5
1.2
122
120
1.3823
1.3957
1.4173
1.4368
730
ir
n
I!
1.4397
1.4410
1.4380
1.4363
IT
n
IT
TT
112
57
69
106
126
116
112
6
4.0
1.5
3.8
11; 35
137
3; 05
143
5; 55 . 2 1 2
235
6 ;30
127
136
154
117.5
130
150
7
3.3
210
—
8
6.1
9
9.0
10
6.2
Residue 5.1 g.
Fraction 9 decolorized bromine in carbon tetrachlor­
ide with the evolution of HBr, produced a slow decolorization of dilute aqueous permanganate, was insoluble in
cold conc. H 2 S0tf, and gave a very generous precipitate
with alcoholic AgN03 .
Five cc. of fraction 9 were washed in a snail sep­
aratory funnel with 10 cc. of conc. H 2 S04 which assumed
a yellow color.
After separating the H 2 S04 , the organic
layer (3.3 g . ) was treated with a little anhyd. K 2 C0 3
after which it had a ref. index of 1.4479.
This material
gave a very good instantaneous precipitate with alcoholic
AgN0 3 and reacted slightly with sodium with evolution
of gas.
An attempt was made to prepare a Grignard re­
agent from this material.
It appeared to react with the
magnesium as the solution began to reflux spontaneously
without the application of heat.
However the magnesium
showed little or no evidence of Grignard formation.
Treatment of this mixture with phenyl isocyanate produced
no petroleum ether soluble compound.
The fractionation table shows no evidence for the
formation of neopentyl bromide in this reaction and there­
fore it was not investigated further.
L. Action of HBr
*
H PSOA
+
NaBr on neopentyl alcohol.
Bromide, 35 cc. (0.7 moles), and 110 g. of cracked
ice were placed in a 1-L r.b. flask and S0 2 passed into
the bromine till the color had disappeared.
Neopentyl
alcohol, 8 8 g. (1 mole), was added to this solution of
HBr and H 2 S04 and then 135 g. (1.13 moles) of sodium
bromide were added.
The flask was fitted with a large
reflux condenser and the mixture v/as refluxed for 18.5
hours.
After cooling, the layers were separated and the
dark organic layer was stirred for one day v/ith a solu­
tion of 50 g. of KOH in 200 cc. of water.
After separ­
ating the layers the aqueous layer was extracted with 90
cc. of ether.
The ether extract v/as combined v/ith the
organic layer, dried over 30 g. of anhyd. K 2 C0 3 and
fractionated with column 2 .
Time
Bath
2; 30
95
36
34
1
5; 00
110
89
38
4; 30
109
87
2; 30
110
4; 30
Col. Vapor Frac.
wt.
Index
ether
1.3555
732 mm
2+
5.3
1.3714
733
38
3+
2.8
1.3803
732
89
38
4+
27.0
1.3860
729
11 0
89
38
5+
4.8
1.3850
732
5; 15
132
97
85
6
0.8
1.3612
731
2; 30
3; 10
4; 25
148
150
164
99
94
96
--
7
100
8
3.5
5.8 .
4.4 '
1.4089
1.4325
1.4368
730
if
it
123
9
Pressure
Residue 9.8 of black liquid.
(+) These fractions contained small water layers (6.7 g.
20
nd
1.3390).
This water apparently came from t-amyl al­
cohol which was slowly dehydrated during the fractiona­
tion.
The organic portions of these fractions were def­
initely olefinic, decolorizing bromine in carbon tetrachlo­
ride, and dilute aqueous permanganate as well as having
the odor of amylenes.
The amount of this olefinic ma­
terial calculated as amylenes amounted to 0.47 moles
which would correspond to a 4 7°/o yield of t-amyl bromide
from this reaction.
Fraction 9 after drying over anhyd. K 2 C0 3 gave an
instantaneous precipitation with alcoholic AgN03.
Frac­
tion 8 reacted with sodium which took on a blue coating
such as obtained in Wurtz reactions.
Fraction 8 was
washed with 10 cc. of conc. H 2 S04 in a small separatory
funnel.
After separating the acid, the material was
treated with a little anhyd. I\2 C03.
This material slowly
gave off halogen acid and had a ref. index of 1.4453.
While these fractions might have contained some neopentyl
bromide, these tests indicated t-amyl bromide rather
than neopentyl bromide.
An attempt to convert this
material to an anilide through the Grignard reagent failed.
M. Stability of neopentyl bromide to HBr + HpSOA .
A solution of HBr + H 2 S04 was prepared by passing
S0 2 into a mixture of 4 cc. of bromine and 20 g. of
cracked ice till the bromine color had disappeared.
This
acid solution was placed in a 10 0 cc. flask having a
ground glass joint fitted v/ith a reflux condenser.
10
g. of neopentyl bromide were added to the acid solution
o
and the mixture refluxed on an oil bath at 125 - 130
for two days.
At the end of this treatment the mixture
v/as so black that no layers could be seen.
The mixture
v/as placed in a small separatory funnel and 25 cc. of
water added.
The lower aqueous layer was drawn off and
the organic layer washed with 40 cc. of water leaving
8.7 g. of organic liquid which v/as dried over anhyd.
K 2 C03 .
The dried liquid was placed in an 18 cc. distil20
ling flask and distilled, giving 4.2 g. nd
103-107° at 730 mm.
1.4340, b.p.
This distillate gave a faint halo­
gen test with alcoholic AgN0 3 thus resembling neopentyl
bromide.
The first attempt to prepare a Grignard reagent
from this distillate failed even though a little ethyl
bromide was added to start the reaction.
On the second
attempt a Grignard reagent was obtained after refluxing
for 1 2 hours with magnesium and dry ether to which a few
drops of ethyl bromide had been added.
Treatment of the
Grignard reagent with phenyl isocyanate gave an anilide
o
which melted at 129.5-131 after three recrystallizations.
o
It gave a mixed m.p. of 129.5-131 with t-butyl acetani­
lide.
Thus neopentyl bromide was identified.
It was
apparent however that the neopentyl bromide v/as not en­
tirely stable to the acid mixture under the conditions of
the experiment.
N. Stability of neopentyl stearate to heat.
(1) Preparation of neopentyl stearate.
Stearic acid, 35 g . , 0.123 moles, was placed in a
200 cc. 3-neck flask fitted with a reflux condenser and
a dropping funnel.
The flask was heated on the steam
bath and 17 g. ^0.145 moles) of thionyl chloride were
added during one-half hour.
The mixture was heated for
an additional 3 hours to complete the reaction.
The crude
chloride was transferred to a 100 cc. Claisen flask and
subjected to vacuum distillation at 5 mm., heating the
20
flask with an air bath at 220°.
Yield 25.0 g. n^ 1.4526,
b.p. 169-178°.
Residue 10 g. of solid.
Stearoyl chloride, 25 g.
{ 0.083
moles), was placed
in a 100 cc. flask and 7.3 g. ^0.085 moles) of neopentyl
alcohol were added.
evolved rapidly.
The mixture became hot and HC1 was
After the reaction had subsided the
flask was fitted with a reflux condenser and heated on
the steam bath for 5 hours to complete the reaction.
Vacuum fractionation of the product was impossible due to
the high viscosity and boiling point.
at 8 mm into two fractions.
It was distilled
20
Fraction 1, 21.9 g. n^
1,4404, fraction 2, 2.9 g. which partly solidified at
room temperature.
Two grams of solid residue remained.
Fraction 1 gave an acid reaction to litmus paper.
It
was dissolved in 1 0 0 cc. of ether and washed with 100 cc.
of 5io Na 2 C0 3 solution to remove free acid.
After distil­
ling off the ether a residue of 2 0 g., n^° 1.4405 remained.
{2)
Stability of neopentyl stearate to heat.
Neopentyl stearate, 20 g . , was placed in a 50 cc.
distilling flask fitted with a thermometer and a conden­
ser.
The system was connected to a dry-ice trap to col­
lect any olefins which might result from pyrolysis.
The
flask v/as heated on a sand bath to distill the ester at
atmospheric pressure if possible.
The ester boiled at
o
359 at 730 mm with the thermometer bulb about 0.5 cm.
above the surface of the boiling liquid.
The temperature
o
of the sand bath was held well over 400 for nearly an
hour.
Only ab.out 3 drops of distillate were obtained
in this time because the neck of the distilling flask
served as a reflux condenser in spite of insulation with-.
asbestos.
While the liquid was refluxed for an hour in
this manner there was little if any evidence of decomposi­
tion.
No olefins were collected in the dry ice trap.
The refractive index of the ester at the end of the one
hour of refluxing was 1.4410.
Neopentyl stearate is ap­
parently very stable to heat.
0. Treatment of neopentyl alcohol v/ith hot conc. H p S0a »
Neopentyl alcohol, 22 g. (0.25 moles), was placed in
a 200 cc. 3-neck flask fitted with a mercury sealed stir­
rer and a tube leading to a dry-ice trap.
Seven cc. of
conc. H 2S04 were added, the solution becoming very hot.
The mixture v/as heated with stirring on an oil bath at
o
135-145 for 7.5 hours.
The temperature was then grado
ually raised to 183 during the next 1.5 hours. Heating
v/as then discontinued.
Twenty g. of products (consisting
of two layers) were collected in the dry-ice trap.
material smelled of S02 .
The layers were separated and
the oil layer (12 g . ) was dried with anhyd. K 2C03.
dried products were fractionated v/ith column
Fraction
1
2
This
B.P.
Weight
Index
32-38
-90
1.6 g.
1.7
1.3857
1.3979
The
3.
20
Residue 3.6 g . , n^ 1.4353. An oil bath temperature
o
up to 180 was used to obtain fraction jj-2. Fraction 1 de­
colorized bromine in carbon tetrachloride and dilute
aqueous permanganate indicating it to be olefinic in nature.
The reaction residue was black and solidified partially
on cooling.
It was treated with ice and then extracted
with 75 cc. of ether.
g. of black tar.
The ether v/as evaporated leaving 4
The reaction v/as not investigated further.
Apparently what happened in this reaction was the initial
formation of esters of sulfuric acid which were unstable
to heat and decomposed into S0 2 and amylenes among other
products.
P. The effect of heat on a mixture of neopentyl alcohol
and sodium neopentylate.
Neopentyl alcohol, 10 g., and 1 g. of sodium were
placed in a Dornb tube and heated on the steam bath till
the mixture became solid.
To promote the solution of
the remaining sodium the tube was heated on an oil bath
o
at 180 for 1.5 hours. At the end of this time only a
little sodium remained undissolved.
The tube was sealed
and placed in the bomb furnace at 23u°. After heating in
o
the furnace at 220 - 240 for 34 hours and 15 minutes
the tube was cooled and opened.
The white solid in the
tube looked the same as the starting mixture.
Water (25
c c.) was added and the tube heated on the steam bath till
the white solid had changed to a liquid.
were poured out of the tube and cooled.
The contents
The upper organic
layer solidified and was separated from the water layer.
o
This solid (10 g . ) gave a phenyl urethan m.p. 112-113 ,
indicating it to be recovered neopentyl alcohol.
The
aqueous layer v/as extracted with ether and then acidified
v/ith HC1.
No organic acids separated.
The acidified so­
lution was extracted v/ith ether and the ether evaporated
leaving no residue.
It was thus apparent that very little
if any reaction had taken place.
The reaction v/as not in-
vestigated at a higher temperature.
Q. Description of the fractionating columns.
All of the columns used were of the total condensa­
tion, adjustable take-off type described by Whitmore and
Lux (4 4 ).
Column
of 1.3 cm.
([■2
was 62 cm long with an external diameter
It was filled with glass helices and had an
H.E.T.P. of 4.3 cm.
Column y3 was 40 cm long v/ith an internal diameter
of 0.9 cm.
It was filled with glass helices and had an
H.E.T.P. of 3.9 cm.
Ground glass joints were used through­
out on this column.
Column #4 was 110 cm long with an external diameter
of 1.8 cm.
It was filled with glass helices and had an
efficiency of approximately 19 theoretical plates.
Column E.M.J. was 100 cm long with an internal diame­
ter of 1.0 cm.
It was packed with single turn stainless
steel helices and had an efficiency of at least 35 theoret­
ical plates.
SUMMARY
Neopentyl alcohol reacted with thionyl chloride to
give neopentyl sulfite.
Neopentyl sulfite did not react
with sulfuryl chloride in the same manner that other pri­
mary alkyl sulfites, such as n-hutyl sulfite, do to yield
the corresponding alkyl halide and alkyl sulfate.
Neopentyl alcohol reacted with thionyl chloride in
chloroform solution in the presence of anhydrous potassium
carbonate to give a 54% yield of neopentyl sulfite.
No
chlorides were obtained.
Neopentyl alcohol was dehydrated when passed over
aluminum oxide at 355 - 390° to give an 81.5% yield of
amylenes consisting of isopropylethylene, 2 -methyl 1 butene and trimethylethylene in the ratio of 1:9:17
Isopropylethylene was converted to a mixture of
isopropylethylene, 2 -methyl 1 butene and trimethylethylene
in the ratio of 1:2:5 when passed over aluminum oxide at
360-410°.
Trimethylethylene was converted to a mixture of
isopropyl ethylene, 2 -methyl 1 -butene and trimethylethylene
in the ratio of i;8.'19 when passed over aluminum oxide
at 375 - 395°.
Neopentyl bromide was obtained in a 26.7% yield by
passing gaseous hydrogen bromide into boiling neonentyl
alcohol for a period of 21 days, 15% of the neopentyl
alcohol being recovered unreacted.
Neopentyl bromide was obtained in a 13.1% yield by
the action of phosphorus trihromide on neopentyl alcohol
together with 18.6% of t-amyl bromide and higher boiling
products, probably esters of phosphorus.
Weopentyl alcohol was destroyed completely by refluxing for 7 days with constant boiling hydrobromic
acid.
No neopentyl bromide was found but a high boiling
bromide which was not identified was obtained.
t-Amyl bromide was obtained in a 41% yield by refluxing neopentyl alcohol with a solution containing
HBr, H 2 S04 , and NaBr.
No neopentyl bromide was obtained
but a small amount of an unidentified high boiling bromide
was found.
Neopentyl bromide was slowly destroyed by refluxing
with a solution of hydrobromic and sulfuric acids prepared
from bromine and sulfur dioxide.
Neopentyl stearate was found to be very stable, dis­
tilling without apparent decomposition at atmospheric
pressure.
Neopentyl alcohol when heated with concentrated sul­
furic acid gave olefins and sulfur dioxide among other
products.
Neopentjrl alcohol did not react with sodium neopentylate when heated at 220 - 240° for 34 hours.
ii9
BIBLIOGRAPHY
1. Freund and Lenze
- Ber.
2. Tissier - Ber. 24 (2)
25
2886 (1890)
557
(1891)
3. Freund and Lenze
- Ber. 24
2150 (1891)
4. Tissier - Compt.
rend. - 112
1065 (1891)
5. Tissier - Ann Claim Phys - (6) 29
6 . Tissier
ibid
^6)
29
357 (1893)
7. Scheuble and Loebl - Monatsh.
_25
8 . Bouveau'lt - uompt. rend. 158
9. Meyersberg - Monatsh.
26
551
12. Menschutkin - Ber.
1081 (1904)
1108 (1904)
41
10. Courtot - Bull. Soc. Chim.
11. Samec - Ann.
321 (1893)
(1905)
(3)
35
985 (1906)
255 (1907)
42
4020 (1909)
13. Richard - /inn Chim Phys (8)
323 (1910); C Zent.
21
1911 (1) 60
14. Franke - C. Zent. 1914 (1) 861
15. Franke and Hinterberger - Monatsh.
43
655 (1922)
16. Hinterberger - Dissertation, Vienna (1923)
17. Ingold - I. Chem. Soc.
125
1706
(1923)
18. Conant, Webb, and Mendum - J.Am. Chem. Soc. 51
19. Adkins and Folkers - ibid
_53
20. Whitmore and Rothroclc - ibid
1095
54
21. Ingold and Patel - J. Chem. Soc.
1246 (1929)
(1931)
3431 (1932)
67 (1933)-
22. Greenwood, Whitmore, and Crooks - J.Am.Chem.Soc.
2028
(1938)
23. Whitmore - Rec.Trav.Chim.
57
562
(1938)
_60
24.
Kohlrausch and Ivoppl - Monatsh.
_63
255
(1933)
25.
Rice, Jenkins, and Harden - J.Am.Chem.Soc.
59
2000
(1937)
26.
Ginnings and Baum - ibid _59
1111
^1937)
27. Magnani and McElvain - J.Am.Chem.Soc.
_60 813 (1938)
28. Wittle - Penn State Thesis, M. S.
29. Whitmore, Wittle, and Harriman - J.Am.Chem.Soc.
1585
61
(1939)
30. Wittle - Penn State Thesis, Ph. D.
31. ICarnatz - Penn state Thesis, Ph. D.
32. Whitmore, Wittle, and Poplcin - J.Am.Chem.Soc.
1586
_61
(1939)
35. Organic Syntheses, Vol. XIX, pages 27 - 29
34.
Goldwasser and Taylor- J.Am.Chem.Soc.
35.
Ipatiev - Ber. 36
36.
Alvarado - J.Am.Chem.Soc. 56
790
(1928)
37.
Senderens - Compt. rend. 148
927
(1909)
2015
J31
1751 (1939)
(1903)
38. Matignon, Moureu, and Dode - ibid
196
973 (1933)
39.
Ipatiev - Ber. 36
1990 (1903)
40.
Pines - J.Am.Chem.Soc.
41.
Komarewslcy, Johnston and Yoder - ibid _56
42.
Cramer and Glasebrook - ibid
55
3892 (1933)
_61 230
2705 (1934)
(1939)
43. Kilmer - Penn State Thesis, M. S.
44. Whitmore and Lux - J.Am.Chem.Soc.
54
3451 (1952)
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 655 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа