вход по аккаунту



код для вставкиСкачать
The Graduate School
Department of Chemistry
Metathetical R e a c t i o n s of Some Alpha Bromo Ketones,
II The Synthesis
of n-Butane.
A Thesis
R i c h a r d B e r n a r d Greenburg
Submitted in partial fulfillment
of the requirements
for the degree of
Doctor of Philosophy
June 1940
of Chemistry
Head of Department
The a u t h o r ’s grateful thanks are tended to
Professor J. G. A s t o n w h o originated these problems
and offered a multitude
during the progress
of helpful suggestions
of the research.
Part I. Meta t h e t i c a l Reactions of Some Alpha Bromo Ketones
Discussion and correlation
1. The r eaction of 3-bromo-3-methyl-2-butanone
w i t h methylmagnesium iodide
2. The r eaction of tertiary bromo-ketones
wi t h alcoholic sodium methylate
3. The action of sodium methylate in anhydrous
ether on 3-bromo-3-methyl-2-butanone
4. The hydrolysis
of tertiary bromo-ketones
5. The etherification of the hydroxyl group in
the treatment of tertiary hydroxy ketones
w i t h alcoholic HC1
Pentamethylethanol from 3-bromo-3-methyl-2butanone
3 . 3-Dimethoxy-2-methyl-2-butanol
2.4-Dinitrophenylhydrazine derivatives
Hydrolysis of 3, 3 - dimethoxy-2-methyl-2-butanol
3 , 3~Diethoxy-2-methyl-2-butanol
2.4-Dinitrophenylhydrazine derivatives
of 3 ,3-diethoxy-2-raethyl-2-butanol
3 ,3-Diisopropoxy-2-methyl-2-butanol
The a c tion of 2 ,4 -dinitrophenylhydrazine on
The action of isopropylmagnesium "bromide on
2.4-Dinitrophenylhydrazine derivatives
Methyl pivalate
from 3-bromo-3-methyl-2-butanone
Ethyl pivalate from 3-bromo-3-methyl-2-butanone
Isopropyl pivalate f r o m 3-bromo-3-methyl-2"butanone
2.4-Dinitrophenylhydrazine derivatives
2.4-Dinitrophenylhydrazine derivatives
3-Hydroxy- 3-methyl-2-pentanone
Methanolic HC1 on 3-hydroxy-3-methyl-2-butanone
Dimethyl sulfate
Appendix A
on 3-hydroxy-3-methyl-2-butanone
P reparation of starting materials
Methyl isopropyl ketone
Ethyl isopropyl
Ethyl isopropyl ketone
Methyl sec-butyl ketone
Appendix B
Miscellaneous reactions
Reaction of diacetyl with m e t h ylmagnesium iodide
Dimethyl sulfate
on 3-hydroxy-3-methyl-2-
in dime thylaniline
R eduction of 3-bromo-3-methyl-2-butanone w i t h
zinc and acetic acid
Summary of Pa r t I
Bibliography, Part I
Part II The synthesis ofn-Butane
Summary of Part II
Bibliography, Part II
Metathetical Reac t i o n s of Some Alpha Bromo Ketones
W h e n a Grignard reagent reacts w i t h a compound
of the general formula R ’-C-Y-R" there are several
X x
courses w h i c h the reactions ma y take.
In this formula
R and R 1 may he aryl, alkyl or hydrogen, X may he Cl
or Br, Y may he carhon or nitrogen, and R IT ma y he aryl
or alkyl
(or a free pair of electrons if Y is nitro­
The possible reactions ma y he any one of the
following, u s i n g R " ’Mg B r as the general Grignard re ­
(1) Enolization, where R and/or R 1 is hydrogen:
R \
Ri-C-f-R" +- R ,T,MgBr —>
-+- R ,n H
X /
(2) Elimination of HX, where R and/or R ’ have alpha
hydr o g e n :
- CH2 n
R « - C - t - R n ■+■ R ,T,Mg B r
(3) Metathesis,
-CH,n 0
C-t-Rn +
R 1'
R M,H-f-MgXBr
replacing X hy H:
R \
R' -C-Y -R” + R n ’MgBr
X /
— >
R \
R 1'
(4) Metathesis, replacing X "by R U | :
R v
r i _C-Y-R"
R v
R ’-C-Y-R"
R tM
+ R n ’M g B r
+ MgXBr
(5) Normal addition to the double hond, Y:0:
R*-C-Y-R" -t R n,M g B r - >
R \ ^OMgX
R ’-C-Y-R"
X ^
(6) Replacement
n R,m
of X w i t h rearrangement:
R v
R .0
(a) R 1-C-Y-R" f R ,T’M g B r — > R ’- C - Y - R ,T1+ MgXBr.
R ,f/
R s
(h) CH*-C-Y-R" -+- R"»MgBr - * R » ' - C H 2-C-Y-R" + MgXBr.
It w a s the purpose
of this research to determine
whi c h of these courses the reaction of an aliphatic ter-
tiary-hromo ketone w o uld take.
had been done
Very little
early w o r k
on the aliphatic series where Y is carbon,
but if Y is n itrogen both normal Reaction 4 ard .re ­
arrangement Reaction 6b occur
(12), together with
addition R e a c t i o n 5.
For the system R - b - C - R T we ma y expect any one of
X /
three possible
courses in reaction w i t h a solution
of R ,r0Na in anhydrous R ”0H.
(7) Metathesis,
replacing X b y 0 R ,T:
CHcu 0
R-C-ft-R’ + H a O R ” -»
CH 3. 0
R-C-ft-R* + Ha X
R ”0"
(8) E l i m i n a t i o n of HX:
CHa, 0
R-C-ft-R1 t- H a O R ”
CH 2„ 0
R-C-ft-R* -r
N a X + R ”0H
(9) Formation of a hydroxy acetal:
CH3, 0
R-C-fi-R* +
X 7
C %
HO" R ’
This r eaction was also investigated w i t h a terti­
ary b r o m o - k e t o n e , since it would appear from analogy
w i t h the Grignard reaction that some reaction corres­
ponding to Rea c t i o n 6a should be possible.
This was
found to be the case, Reaction 10 occurs u n d e r certain
conditions w i t h alcoholic solutions of alcoholates.
(10) Replacement
of X wi t h rearrangement:
CH3 v
R - C — ft-R* + NaOR" — *
R -C-ft-OR”
F r o m th6 abovfe it w o u l d sosip ths."fc sodium alooRoi"
ates in a neutral solvent such as ether should give
Reactions 7,8,
and 10.
This reaction was also inves­
tigated w i t h the same tertiary hromo-ketone us e d in
the ahove work.
The reaction of sodium hydroxide i n water would
he expected to follow one of two courses, provided no
condensation reactions occurred.
Either Reaction 8,
splitting out HX would he expected,
or a direct hydrol­
CH3 v 0
(11) R-C-6‘- R » N a O H
X '
CH3 n 0
— > R-C-fi-R' + N a X
HO /
This reaction was also investigated, and while Re ­
action 11 occurred easily, no evidence of splitting out
of H X was observed.
Certain peculiar reactions of alpha-
hydroxy-ketones were also investigated in detail because
of their hearing on the structure of the alpha-hydroxy-
The ahove outlined investigations constitute
the first part of this dissertation.
The reaction of 3-bromo-3-methyl-2-butanone wit h
methyl-magnesium i o d i d e .
In the r e action of primary halo-lcetones of the type
X-CHg-CO-R w i t h the Grignar d reagent, Reactions 1, 5,
and 6a (see Introduction) have been observed.
and Tishler
(l) reported that enolization was to be ex­
pected whenever a hydrogen a t o m was situated in the
proper position.
This enolization reaction was always
accompanied by some other reaction.
T iffeneau (2)
treated ehloracetone w i t h phenylmagnesium bromide, and
isolated methyl benzyl ketone, a seemingly good example
of Reaction 4.
However wh e n the same author
treated omega-chioracetophenone w i t h ethylmagnesium
bromide he found that the product was ethyl benzyl ke­
tone, an example of Reactio n 6A. This reaction was ex­
plained by a free di-radical mechanism at the time of
M a n y years later Tiffeneau and Tchoubar
(3), realizing the improbability of their mechanism,
published another, utilizin g the "small ring" mechanism
of rearrangement.
and Fuson (4) state
that the a d dition of Grignard reagents to omega-bromoacetophenone is rapid,
furnishing an example
of Reaction
Of the secondary halo-ketones,
the ones w h i c h have
attracted the most interest are the cyclics,
Bouveault and Chereau (5)
reported that this ketone gave two products whe n
treated w i t h Grignard reagents.
The first was the
alpha-alkyl ketone formed by normal metathesis,
tion 4, while the second was an alkyl cyclopentyl
ketone, w h i c h according to the findings here i n reported,
is probably formed by React i o n 6a.
Quite recently
Bartlett and R osenwald (6) have repeated this w o r k and
confirmed the products obtained.
found that alpha-bromo-beta,
phenone was
Kohler and Tishler
converted to an enolate of the type shown
in R e a c t i o n 3, replacing the halogen w i t h hydrogen.
They offered definite proof that the compound was a
true enolate,
and not a "metal on carbon" type of com­
by isolation of a known enolate peroxide.
Their w o r k was
in confirmation of the earlier wor k of
Kohler and Johnstin (8).
(32) has observed
a somewhat novel reaction of the cyclic secondary
He found that alpha-chloro-cyclo-
hexanone was easily reduced by Grignard reagents to
give the m a g n e s i u m halide salt of the chlorohydrin.
This reaction depends on the Grignard reagent being
one w h i c h has the property of reducing action (33).
The tertiary bromo-ketones have been found to
give a rea c t i o n of the type of Reaction 3 in almost
every case studied until now.
Kohler and Tishler (7)
found that this was the case w i t h alpha, beta, betatriphenyl-alpha-bromo-propiophenone, and Lowenbein and
(9) found the same results w i t h alpha,
is the wo r k of U mno w a
Of especial interest
(10), who reported that alpha-
bromo-pentamethylacetone was converted to pentamethylacetone by methylmagnesium iodide, w i t h evolution of
methyl bromide.
She also reported that alpha, a l p h a 1-
dibromo-isobutyrone was converted to alpha-phenyl-isobutyrone b y phenylmagnesium bromide,
evidently by R e ­
action 3 on one bromine and Reaction 4 on the other.
The above is not intended to be a complete litera­
ture survey of the subject, since an adequate survey
has be e n made by Harriman
(11), but is merely offered
to illustrate the abnormal reactions which alpha-halo
ketones may undergo.
case where Y
(IE) has investigated this reaction for the
is nitrogen.
He found both Reaction 4 and
Reaction 5 w o u l d occur as might be expected, but that
in addition a rearrangement of the alkyl group attached
to Y would sometimes occur, possibly by a reaction simi­
lar to Reaction 6b.
In the course of this research the action of methylmagnesium iodide u p o n 3-bromo-3-methyl-2-hutanone was
It was f o und that the reactions whi c h
occurred were 4 a n d 5,
the completely normal ones.
Uo evidence was found for any of the products of re ­
actions 3 or 6.
The product obtained, pentamethyl-
ethanol, w a s reco v e r e d in 62$ yield.
This yield is
not q u a n t i t a t i v e , but some loss is unavoidable in the
handling of this compound, w h i c h has a high vapor pres­
sure at room temperature.
This was not the anticipated product,
since the
w o r k of Umnowa w o u l d seem to at least partially dis­
credit steric hindrance as the m a i n cause of Reaction
Also, no hindrance seems apparent in the brom-
nitroso-propane w h i c h Mena r d
(12) found to give both
normal and r earranged products.
In the face of the variable nature of this r e ­
action it does not seem prudent to propose a mechanism
at this time,
and it is doubtful if any common basis
can be found for all of the reactions.
2. The reaction of tertiary bromo-ketones w i t h alcoholic?
sodium m e t h y l a t e ,
The primary halo-ketones m ay give either Reaction
8 or Reaction 9 wi t h alcoholic sodium methylate.
action 8 does not occur u nl e s s a tertiary hydrogen atom
is available.
(13) has reported that the reaction
of sodium phenoxide w i t h chloracetone gives only tars
unless conducted in a large excess of neutral solvent.
Their use of the action of Grignard reagents on alkoxynitriles to prepare alkoxy-ketones is representative of
all of the available literature.
Ho mention was found
of any alkoxy-ketone prepar e d by the "Williamson syn­
The only analogous reaction is found in the
w o r k of Rotbort
(14), who prepared alkoxyacetals by
the reaction of alcoholates on bromacetal.
and Miclceley (15) found that bromacetone gave the methyl
acetal of acetol w h e n treated with alcoholic sodium
This takes place according to Reaction 9.
T h e y reported the acetal as an unstable oil which de­
composed almost instantly on heating to give 2,5-di­
me thyl-2,5-dimethoxy-1,4-dioxane by loss of two mol e ­
cules of methanol from two molecules of the acetal.
Similar results m a y be expected, and were found,
f r o m the secondary halo ketones.
War d (16) showed that
desyl chloride gave the ethyl acetal of benzoin w h e n
treated w i t h sodium ethylate in absolute alcohol,
other example of Re a c t i o n 9,
In his paper he takes
exception to the earlier w o r k of Fisher (17) and
Irvine and Nicoll
(18) who said that desyl chloride
gave the expected ether of b e n z o i n on treatment w i t h
sodium ethylate,
according to Reaction 7.
No literature is available on the reaction of t er­
tiary bromo-ketones w i t h alcoholic sodium methylate.
When this rea c t i o n was carried out using 3-bromo-3
methyl-2-butanone, the product was found to be the
other example of R eaction 9.
It was definitely proved
that this was the structure of the compound, and that
no carbon skeleton rearrangement had occurred.
Re ­
cently Froning and Hennion (19) have synthesized this
compound by an independent method, and their reported
physical constants check w i t h those found in this work.
It is noteworthy that neither the unsaturated ke ­
tone nor the expected keto ether was a product of this
No un s a t u r a t e d product occurred at a n y t i m e ,
and none of the lower fractions contained any ketonic
Since none of the previous workers in this
field have r e ported any normal product,
primary halo-ketones,
except with the
it seems that the substitution of
an acyl for an alkyl group has a profound effect on the
W illiamson reaction.
Since the shift of a methyl group in this reaction
w ou l d not necessitate the formation of a new compound,
it was determined to investigate the next higher member
of the series,
ethyl isopropyl ketone.
this ketone w a s prepared,
sodium methylate in absolute methanol.
and treated w i t h
The hydroxy
acetal was isolated and a derivative prepared which
was compared w i t h one from the hydroxy ketone and found
The hydroxy ketone had been prepared by
hydrolysis of the b r o m o - k e t o n e , w h i c h Favorskii
had shown to be a satisfactory method of preparation
without rearrangement.
As further evidence that re­
arrangement did not occur during metathesis, methyl
hydroxy-sec-butyl ketone was prepared and its deriva­
tive showh to be non-identical w i t h that from the
M e t h y l hydroxy-sec-butyl ketone would be the
expected product if rearrangement occurred.
This reaction was also carried out using a solu­
tion of sodium ethylate in absolute ethanol.
The ethyl
acetal of the hydroxy ketone was obtained as well as a
amall amount of ethyl trimethyl acetate according to
R eaction 10.
(This unusual product is discussed below.)
When isopropyl alcohol was used,
some of the acetal was
formed, but the greater part of the bromo-ketone was
converted to iso-propyl trimethylacetate,
(see below)
In all of the previous w o r k on the hydroxy acetals formed "by this reaction, no evidence has been
offered that the compound could not be the alkoxy
hemiacetal, w h i c h would have the same empirical for­
This compound w ou l d also be susceptible to hy­
and w o u l d give the same derivatives.
these tertiary alpha-alkoxy ketones, the alkoxyl
group is apparently easily split off, as is shown by
the w o r k of Schmidt and A u s t i n (21) who prepared the
oxirae of 3-methoxy-3-methyl-2-butanone b y the action
of sodium methylate on the nitrosate of trimethylethylene.
W h e n they attempted to hydrolyze the oxime
to the m e t h o x y ketone,
they found the methoxyl group
completely split off.
That the product was the true acetal was shown by
two reactions.
Wh e n the acetal was heated for one hour
w i t h 2,4-dinitrophenylhydrazine, no reaction took place,
although u n d e r the same conditions a hemiacetal or a
ketone will react quite rapidly.
the acetal
was shown to be unaffected by isopropylmagnesium br o ­
mide, while the hemiacetal form wou l d have yielded the
monomethyl ether of a pinacol.
After decomposition of
the Grignard reaction only the original acetal was
At first glance it would seem that a Zerewitnoff
analysis w o u l d differentiate b e t w e e n the two forms,
but reference to the w o r k of Chichibabin (22) and
ICranzf elder and Vogt
shows that under the condi­
tions of this analysis an acetal will give one mole
of addition,
so that the net reaction from either form
w o u l d be one mole of active hydrogen and one mole of
CH3-C(0CH3 )2-C(0H)(CH3 )2 -f- 2CHgMgI -*■
CH4 + CHgOMgl +CHg-C(0CH3 )CHg-C(OMgl) (CH3 )2
CH3-C(0H) (0CH3 )-C(0CH3 ) (CH3 )2 ■+- 2CH3MgI
CH4 -t-CHgOMgl
CH3-C (OMgl) (CHg)-C (0CH3 )(CHg)2
This was found to be the case w h e n b o t h the methyl
and ethyl acetals of 3-hydroxy-3-methyl-2-butanone were
analyzed in the"Grignard machine" of Kohler,
and Fuson (24).
3. The action of sodium methylate in anhydrous ether
on 3-bromo-3-methyl-2-butan o n e .
The use of socLium methylate in ether as a react­
ant for the W i l l i a m s o n synthesis seems to have been
overlooked by the earlier workers in this field.
reference has been found discussing the use of this
reagent except u p o n p r imary halo-ketones.
Henze and
Whitney (13) and Henze a n d Calaway (25) have u s e d sol­
utions of phenoxides in toluene to obtain aryloxy
acetones according to R e act i o n 7.
They have prepared
many other alkoxy ketones in their work, both secondary
and primary, but except for the aryloxy derivatives
have always u t ilized the action of a Grignard reagent
u p o n an alkoxy-nitrile for their syntheses.
W h e n 3-bromo-3-methyl-2-butanone was added to an
equivalent quantity of sodium methylate in anhydrous
ether the product was found to be the methyl ester of
trimethylacetic acid.
This product is possible by. a
rearrangement of the carbon skeleton and a'hormal”
of the halogen, according to Reaction 10.
It was found that the reaction could be carried out
with other alcoholates,
giving the corresponding
If any alcohol was present,
a higher boiling
component also appeared, which was identified as the
acetal formed from the bromo-ketone by the action of
sodium methylate according to Reaction 9.
(see above).
This reaction (Reaction 10) has be e n shown to
take place with other bromo-ketones where the bromine
is t e rtiary (26),
and seems to offer a general meth o d
of synthesis of tri-substituted acetic acids.
4. The hydrolysis of tertiary bromo-ketones.
The hydrolysis of bromo-ketones to the correspond­
ing ketols according to Reaction 1 1 has been investi­
gated at great length,
ketol in most cases.
and has b e e n shown to yield the
(£0) prepared a large
number of ketols by hydrolysis of the corresponding
That this reaction is n ot without its anomalies
is aptly demonstrated b y the w o r k of Favorski and
(27) who converted 2-chloro-cyclohexanone
to cyclopentane
By use
carboxyllic acid by the action of KOH
This is somewhat similar to Reaction 6 a.
of the 3-methyl compound they showed that the
carboxyl group occupied a place on the atom which car­
ried the halogen,
later these same workers
(£8 ) car­
ried out the same reaction using chloro-cyclopentanone
and reported that no cyclobutane carboxyllic acid was
(29) has shown that the ketols sometimes
undergo a rearrangement whe n heated in sealed tubes
w i t h dilute acid.
This rearrangement does not seem to
involve a carbon skeleton change.
In this w o r k it was found that a slight excess of
NaOH in water w o u l d give a nearly quantitative hydrolysis
in a very short time.
Ro product other than the
3 -hydroxy-3-methyl-2-butanone was
results were obtained w i t h 3-bromo-3-methyl-2-pentanone and 2-bromo-2-methyl-3-pentahone.
5. The etherification of the hydroxyl group in the
of tertiary hydroxy ketones w i t h alcoholic HC1
That ethers in which one radical is tertiary are
easily sp^-it h y aqueous HC1 was
shown "by Norris and
Rigby (30) who determined the rate of hydrolysis of
methyl etad tertiary-hutyl ether w i t h concentrated HC1.
The synthesis of methyl glucosides hy the action of
alcoholic HC1 is the standard method of synthesis of
these compounds
Fisher (17) and Irvine and Nicoll
(18) state that
the ethyl ether of benzoin is split hy aqueous HC1, hut
their work is disputed h y W a r d (16), who synthesized
the compound and reported that it was stable to dilute
Schmidt and Austin (21) found that hydrolysis of
the oxime of 3-methoxy-3-methyl-2-butanone w i t h HC1
gave the h y droxy ketone.
In th i s w o r k attempts were made to prepare the
2 ,4 -dinitrophenylhydrazones
from the hydroxy ketones
obtained hy hydrolysis of their acetals.
It was found
that wh e n an absolute methanol solution was used, with
the HC1 incident to this preparation,
the product was
not the 2 ,4 -dinitrophenylhydrazone of the hydr o x y ketone
hut the derivative
of.the corresponding methoxy ketone.
If a saturated solution of 2,4-dinitrophenyl­
hydrazine in 2N HC1 was u s e d to prepare a derivative
from the
the expected hydroxy derivative was
It w a s found possible to convert the hydroxy
2,4-dinitrophenylhydrazo ne to that of the
methoxy compound by several recrystallizations from
absolute m ethanol
containing a trace of HC1.
heating w i t h dilute HC1 converted the methoxy derivate
to the corresponding h y droxy compound.
were obtained w i t h ethanol.
Similar results
Boiling water v/as without
effect u p o n the alkoxy derivatives.
F r o m this
it would seem that the ethers of the
tertiary h y d r o x y ketones are b o t h easy to f o r m and easy
to split.
it was
found that w h e n attempts
were made to prepare these ethers, no methylation was
either w i t h dimethyl sulfate or methanol and
The p o s s i b i l i t y that some product might have been
formed in the above mentione d reactions v/as discarded
on the following grounds.
The m aterial r ecovered had the same boiling point
and refractive
index as the starting material.
fication might be expected to lower both of these,
pecially the former.
It is n ot inconceivable, however,
that these properties could coincide.
The ethers of
the ketone derivatives are knov/n to be stable in bo i l ­
ing water.
Accordingly the semicarbazones of the isola­
ted products were prepared in water solution, buffered
with sodium acetate.
These preparations were heated
only to approximately 80° for a short time.
derivative formed was
that of the starting material.
This evidence leaves only the possibility that
the ether ketone is instantly decomposed b y water,
while its derivatives are not.
The remoteness of this
together w i t h the amount of w o r k w h i c h
would be n e e d e d to confirm or reject it, made it illog­
ical to pursue it further.
In the following work when e v e r fractionation is
referred to the w o r k was done usi n g either a 65x1.5 cm.
total condensation column pack e d wit h single turn
or a 40x0.9 cm.
the large
column of the same type.
column was u s e d whenever 35 cc. or
more of material w a s available.
Pentamethylethanol from 3-bromo-3-methyl-2-hutanone.
Grignard reagent was prepared in the usual manner from
2 moles of M e l in one liter of commercial anhydrous
Afte r reaction had ceased the solution was
stirred over night to ensure completion.
To this solution
was added 0.75 mole of the bromo-ketone over a period
of five hours,
the rate being such that only a slight
refluxing of the ether occurred.
After the addition
was complete the reaction mixture was
hours at ro o m temperature.
stirred for ten
The complex was decomposed
by pouring on ice and the ether layer decanted.
residue was dissolved in HC1 and the solution well ex­
tracted w i t h ether.
After washing w i t h dilute sodium
solution this ether was added to the main
The combined extracts were dried over anhydrous
potassium carbonate and concentrated to a volume of
100 cc. by fractionation.
The residue was distilled
from a small flask to yie l d a mixture of pentamethyl-
ethanol and pentamethylethanol hydrate which hoiled
at 120-125°/740 ram.
This material was all converted
to the hydrate b y treatment with a small amount of cold
water and filtration.
Steam distillation of the water
layers did not yie l d any additional product.
The yield
was 0.47 mole or 62$ of material melting at 82-83°.
A mixed melting point w i t h a sample,
82-83°, pre­
pared from the product of the reaction of trimethylacetyl
chloride and m e t hylmagnesium chloride
(W. S. Forster)
was not depressed.
3,3-Dimethoxy-2-methyl-2-b utanol.
A solution of
sodium methylate was prepared by addition of 2 0 . 1 gm.
(0.875 mole)
of sodium to 300 cc. of absolute methanol.
The solution was cooled in ice and 145 gms.
of 3-bromo-3-methyl-2-butanone was added over a
period of two hours.
The solution was then allowed to
come to room temperature with continued stirring and
The precipitated HaBr was washed wit h methanol
and the washings added to the filtrate.
The methanol
was taken off through a column to one-third volume, and
the residue again filtered.
Fractionation of the
residual liquid gave 99.3 gms.
oil, b.p. 159-161° at 730 mm.; n|°,
MR (c a l c d . ), 39.34; M R (obs.),
of colorless
1.4238; d|°,
0CH3 , 41.8.
for C 5 H 1 0 0 (0 C H 3 )2 : C,56.8; H, 10.8;
Found. C, 56.8,
56.8; H, 11.2,
0CH3 , 40.8.
The 2,4-dinitrophenylhydrazine derivatives from
the h y d r o x y - a c e t a l .
Treatment w i t h a solution of 2,4-dinitrophenylhydrazine in 2N HC1 gave a derivative w h i c h w he n re­
crystallized from petroleum ether melted at 139-140°
and did not depress an authentic sample of the deriva­
tive of the h y droxy ketone prepared below.
46.8; H, 5.0.
Found: 0, 46.5; H, 5.3.
Treatment w i t h 2 ,4-dinitrophenylhydrazine in
methanol containing at the most only a trace of water
and addition of methanolic HOI gave a derivative which
when recrystallized from petroleum ether melted at
138-139° and d i d not depress the melting point of the
derivative of 3 -methoxy- 3 -methyl- 2 -butanone prepared
below from the hydroxy ketone.
0CH3 , 10.0.
A mix e d melting point of the two derivatives pr e ­
pared above from the acetal was depressed to 112-114°.
Hydrolysis of
,3 -dimethoxy- 2 -methyl.-2 -butanol.
Seventy seven grams (0.52 mole) of the hydroxy acetal
was p l a c e d u nder a column w i t h 50 cc.
of 2$ HC1.
mixture was refluxed for ninety minutes before take-off
was b e g u n •
n ^
Fractionation was continued up to a head
of 65°,
giving 31.1 grams
(0.97 mole)
or 94 Jo of the theoretical amount.
column was then allowed to drain and the pot residue
Saturation of this residue w i t h potassium
caused the separation of an oil layer, which
was taken up in ether.
The v/ater was extracted wit h
ether and the combined ether layers dried over anhydrous
potassium carbonate and fractionated.
gave 48.6 grams
141°, n ^
(0.48 mole)
This fractionation
of hydroxy ketone, b.p. 139-
This amounts to 97$> of the expected
The hydroxy ketone was identified as the semi-
carbazone, melting point 164-165°
3 , 3-Diethoxy-2-methyl-2-bu tanol.
This compound
was pre p a r e d b y a similar reaction to that u s e d for the
methyl acetal, w i t h the exception that after removal of
the alcohol by distillation a small amount of water was
The acetal layer was separated from the aqueous
NaBr layer,
54.4 gms.
dried and fractionated.
of acetal, b.p. 110-112 at 98 mm.;
n 20 1.4189;
A one mole run gave
0.919; M R (calc. ) 48.57;
LIE (obs), 48.40.
for C 9 H 2 o03 : C, 61.4; H, 11.4.
C, 61.5; H, 11.3.
From this r eaction there was also isolated 17.6
of ethyl pivalate,
59-62°at 100 mm.
(see b e l o w ) .
2 14-Dinltrophenylhydraz ine de rivat ive s .
This acetal gave derivatives in the same manner and
with the same m e l t i n g points as those prepared from the
methyl acetal.
Fr o m it a 2,4~dinitrophenylhydrazone
was also pr e p a r e d
in absolute ethanol.
Calcd. for C 1 1 H 1 3 N 4 0 4 0 E t : C, 50.3;
H, 5.8; OEt,
It melted at
C, 50.3; H, 5.5; OEt, 14.3.
Hydrolysis of 3,3-diethoxy-2-methyl-2-butanol.
A hydrolysis
similar to that described above was carried
out on 12.5 grams
(0.071 mole)
of the ethyl acetal.
recovery amounted
to 86 $ of the theoretical
80$ of the theoretical hydroxy ketone.
3,3-Diisopropoxy-2-methy l - 2 - b u t a n o l .
A preparation
was carried out similar to those above using a solution
of sodium isopropylate in isopropyl alcohol.
The prin­
cipal product of this reaction was isopropyl pivalate,
b.p. 49-51° at 46 mm.
Since this product was unexpected
the distillation was not carried out under conditions
which w o u l d give a satisfactory recovery of a low-hoiling
However, the isopropyl pivalate isolated
amounted to 20$.
The reaction also gave 8 $ of higher
boiling material w h i c h was poss i b l y the acetal,
it could be decomposed to give ketonic derivatives.
The amount was too small to purify satisfactorily for
It b o i l e d f r o m 67
a refractive
to 95
at 46 mm.
and had
index of 1.4305-1.4382.
The action of 2,4-dinitrophenylhydrazine
d i m ethoxy-2-methyl-2-butanol.
on 3,3-
One gram of the acetal
was placed in a 30 cc. glass-stoppered Erlenmeyer
flask w i t h one gram of 2,4-dinitrophenylhydrazine and
the mixture heated for two hours at 100°.
The mixture
was then extracted w i t h hot p etroleum ether,
and the filtrate
solution was
(25 c c . ) concentrated to 5 cc.
cooled and seeded w i t h a sample of the
hydroxy ketone dinitrophenylh y d r a z o n e , but no crystals
appeared after five days in the ice chest.
tion to 3 cc. and repetition of the above produced no
U n d e r the same conditions bot h methyl iso­
propyl ketone and glucose gave a good yield of crystals
after t e n minutes heating.
The action o f isoprop.ylmagnesium bromide up o n
3 13-d i m e t h o xy-2-methyl-2-b u t a n o l .
A Grignard reagent
was p r e p a r e d in the usual manner from 4 9 . 2 gm.
of isopropyl bromide.
added 14.8 gms.
To this solution was
of the acetal.
tion took one hour, during w h i c h
(0 . 4
The addi­
time gas was evolved
Aft e r completion of the addition the
solution was refluxed and stirred for one hour.
It was
then decomposed b y cautious addition of water and the
ether layer decanted.
The residue was steam distilled,
the distillate extracted with ether an d the ether layers
The ether was dried over anhydrous potassium
carbonate a n d distilled.
oil was
Eleven and one-half grams of
obtained boiling at 37-39° at 1 mm.
This oil
was analysed for methoxyl.
Calcd.for 2,3,4-tri»ethyl-2-methoxy-3-pentanol,
OMe, 18$.
OMe, 41.8$.
for 3,3-dimethoxy-2-methyl-2-butanol,
OMe, 40.0$.
A p p a r e n t l y the acetal has come thro u g h the reaction u n ­
as the analysis shows.
2, 3,4-Trimethyl-2-meth-
oxy- 3 -pentanol is the product w h i c h would have been ex­
pected from the hemi-acetal.
A 2, 4 - dinitrophenylhydrazone
of this product
was prepared b y the a c t i o n of a saturated solution of
the reagent in 2N HC1.
M e l t i n g point and mixed melting
point wi t h a n authentic sample of the derivative from
3 -hydroxy-3-methyl-2-hutanone was 138-139°.
3,3-Dimethoxy-2-methyl- 2 - p e n t a n o l .
This compound
was prepared from 2 - b r o m o - 2 -methyl-3-pentanone
isopropyl ethyl ketone) b y the same procedure us e d
above in the preparation of 3,3-dimethoxy-2-methyl-2butanol.
A one-half mole run gave 65.5$ yield of an
oil, b.p.
82-g-0 at 100 mm.; n ^
(calc.) 43.96, MR (found)
C, 59.0,
li4088; df° 0.8996;
for CgH 1 8 03 : C, 59.3; H, 11.1.
59.1; H, 11.0,
2 - 4 -Dinitrophenylhydrazine
derivatives from the
hy droxy-acetal.
A 2 ,4 - dinitrophenylhydrazone was prepared by the
action of a saturated solution o f the reagent in 2N HC1.
Up o n recrystallization from petroleum ether the deriva­
tive melted at 125-126°.
This was the derivative of
the hydroxy ketone as expected.
C, 48.2,
for C-^gH^QlT^Os: C, 48.6; H, 5.4.
48.4; H, 5.7,
A 2,4-dinitrophenylhydrazone was prepared in meth­
anol containing at the most only a trace of water.
Up on two recrystallizations from petroleum ether it
melted at 139-139-|-0 .
This was evidently the derivative
of 2 -methoxy- 2 -methyl-3-pent a n o n e ,
for C ^ H ^ g l ^ O g : C, 50.3; H, 5.8.
Found; C, 50.1; H,
A 2 ,4-dinitrophenylhydrazone was prepared in eth­
anol containing at the most only a trace of water.
recrystallization f r o m petroleum ether it melted at
This was evidently the derivative of 2-ethoxy-
2 -methyl-3-pentanone•
Calcd. for CJ 4 H 2 0 N 4 O 5 : C, 51.9; H, 6.2.
C, 51.6; H,
A mixed melting point of the hydroxy and methoxy
derivatives was depressed to 117-122 , a m i x e d melting
point of the hydroxy and ethoxy derivatives was depressed
to 108-114°.
In order to demonstrate that no rearrangement has
the 2 ,4 -dinitrophenylhydrazone of 3 -methyl-
3-hydroxy-2-pentanone was prepared.
It melted at 86-87°,
and a mixed melting point wit h a sample of the hydroxy
derivative above
(m.p. 125-126 ) was depressed to below
(For the preparation of 3-methyl-3-hydroxy-2-
see b e l o w . )
M e t h y l pivalate from 3-bromo-3-methyl-2-butanone.
A suspension of sodium methylate in anhydrous ether was
prepared b y the addition of 11.5 gm.
of finely cut
sodium (0.5 mole) to a solution of 30 cc. of methanol
(0.75 mole)
in 500 cc.
of absolute ether.
The solution
was refluxed until solution of the sodium appeared to
be complete, t h e n addition of 82.5 gm. (0.5 m o l e )A the
bromo-ketone was begun at such a rate as to cause
continuous refluxing.
After addition was complete the
mixture was refluxed for one hour, then wa t e r was added
in a quantity sufficient to dissolve all of the sus­
pended solid.
The ether layer was ss/»arated, dried over
anhydrous sodium sulfate at 0 ° and fractionated to
yield two fractions:
Fraction A; 22 gms., b.p.
31-45° at 100 mm., nj) 1.3870-
Fraction B; 12 gms. b.p.
70-80° at 100 mm., n ^ 1.4208-
Fraction A was identified as methyl pivalate,
yield 39c
for C^H-LgOg: C, 62.0; H, 10.4; M.W.,
114; N.E., 114.
Found.: C, 61.7; H, 10.3; M . W . , (cryo-
soopic in "benzene) 112,
113; H.E.
Saponification gave MeOH,
identified as the 3 ,5 -
dinitrobenzoate, melting point and mix e d melting point
105i-107°; and pivalic acid, melting point and mixed
melting point 34-36°,
amide, melting point and mixed
melting point 153-154°.
Fraction B w a s apparently the hydroxy acetal,
since u p o n treatment
with acid it was decomposed to a compound which gave
the 2,4-dinitrophenylhydrazone
of 3-hydroxy-3-methyl-
2-hutanone, melting point and mixed melting point 138139°.
The formation of this product is probably due to
the excess of alcohol used,
since it does not appear in
the p r e p a ration below where no excess of alcohol was
The low yield of ester is no doubt due to the
low temperature of fractionation,
since vacuum was us e d
to prevent decomposition of the product, whose nature
was then unknown.
Ethyl pivalate from 3 -bromo- 3 -methyl- 2 -butanone.
A suspension of sodium ethylate in absolute ether was
prepared b y adding 11.5 gm.
29.2 cc.
(0.5 mole)
(0.5 mole)
of absolute alcohol to 500 cc. of
ether in a vessel w i t h condenser,
of sodium and
and dropping
The mixture was refluxed for 48 hours to ensure
82.5 gm.
The solution was then cooled in ice and
(0.5 mole)
of 3-Bromo-3-methyl-2-Butanone
was added over a p e r i o d of two hours.
was refluxed for three hours,
The solution
then water was added to
dissolve the precipitated NaBr.
The layers were sep­
arated and the ether dried and fractionated.
was 39.8 gms.
mm. n 20 1.3912;
The yield
of ester Boiling a i? 116° at 725
d 2 0 0.856; M R
36.18; M R
Wt. 130; N.E.
for C 7Hi^Og: C, 64.6; H, 10.8; Mol.
Found: C, 64.4; H, 10.8; Mol. Wt.
cryoscopic in Benzene)
129, 130. K.E.
Sa p o nification of a sample gave ethanol,
as the 3 , 5 - d i n i t r o B e n z o a t e , melting point and mixed
melting point 91-92°,
and pivalic acid, m.p.
amide, melting point and mixed melting point 153-154 .
Isopropyl pivalate from 3-Bromo-3-methyl-2-Butanone.
This p r e p a ration was carried out exactly as that for
ethyl pivalate above, yielding 64/b of ester in a one-half
mole run.
No acetal was found.
for CqH-^Os:
C, 6 6 , 6 ; H, 11.2, M.W.
C, 66.4; H, 11.2; M. W. (cryoscopic in Benzene)
141, 141.
S a p o nification of a sample gave pivalic acid; amide,
melting point 152-1540.
3-Hy dr oxy- 5-met hyl- 2-but anone from hydrolysis of
Ninety four grams
of the bromo-ketone was refluxed for one hour wit h
a solution of 35.1 grams of K OH (0.63 mole)
of water.
in 250 cc.
Solution of the bromo-ketone took place al­
most immediately,
and at the end of the reaction only a
very small amount of undisso l v e d oil remained.
The re ­
action mixture was saturated w i t h NaCl and extracted with
ether in a continuous extractor.
The extraction was con­
tinued for 1 0 0 hours,
although less time wou l d probably
have b e e n sufficient.
The ether extract was dried w i t h
anhydrous magn e s i u m sulfate,
the ether stripped off and
the residue fractionated to yield 46 grams
(76$) of the
hydroxy ketone,
b.p. 140° at 727 mm.,
A c i d ification of a sample of the hydrolysis mixture
gave no evidence of trimethylacetic acid,
although this
product might have been expected from the wor k of Favorskii and Bozhovskii
A semicarbazone was prepared in aqueous solution.
After recrystallization from water it melted at 164-165°
as reported by F r e o n (34).
The 2 ,4 -dinitrophenylhydrazine derivatives from the
hydr oxy-ketone.
A 2 ,4 -dinitrophenylhydrazone was prepared by the
action of a saturated solution of the reagent in 2N HC1.
After recrystallization from petroleum ether it melt e d at
138-139° and gave no depression w h e n mixed w i t h a
sample of the 2 ,4-dinitrophenylhydrazone prepared
above hy the a c t i o n of the aqueous reagent on 3 ,3 dime th,oxy- 2 -me thyl- 2 - but a n o l •
for C-jjH^IT^Og; C, 46.8; H, 5.0.
C, 46.6; H, 5.1.
A 2,4-dinitrophenylhydrazone was prepared in ab­
solute methanol, u s ing methanolic HC1.
The derivative
was reorystallized from petro l e u m ether to a melting
point of 138-139°.
It did not depress the melting
point of the 2,4-dinitrophenylhydrazone prepared above
from 3,3-dimethoxy-2-methyl-8-butanol in a similar ma n ­
It depressed the melting points of the derivatives
of 3-hydroxy-3-methyl-2-butanone prepared from both
the ketone and its acetal.
for C ^ H ^ g ^ C ^ O C i ^ ; OCH 3 , 10.5.
OCH 3 , 10.0.
2-Hydroxy-2-methyl-3-pentanone was prepared in a
manner similar to the above preparation of 3-hydroxy3 - m e t h y l - 2 - b u t a n o n e , in 70 fo yield.
It boiled at 95-97
at 100 mm.
as reported by Favorskii
,4 -dinitrophenylhydrazine derivatives from
the hydroxy-ketone.
A 2, 4 -dinitr ophenylhydrazone was prepared b y the
action of a saturated solution of the reagent in 2N HC1.
U p o n recrystallization it melted at 125-126
and did not
depress the melting point of the derivative prepared
above from 3,3-dimethoxy-2-methyl-2-pentanol by the
action of the same aqueous reagent,
A 2,4-dinitrophenylhydrazone was prepared in ab­
solute methanol, with
addition of methanolic HC1,
It melted at 138-139°
did not depress the melting
point of the derivative prep a r e d from 3,3-dimethoxy-2raethyl-2 -pentanol u n d e r similar conditions.
3-Hydroxy-3-methyl-2-pentanone was prepared by the
hydrolysis of 19.8 gms.
(0.I I 7mole)
of 3-bromo-3-methyl-
2 -pentanone w i t h a solution of 8 grams
NaOH in 100 cc.
ten hours,
of water.
(0 . 2 mole)
The reaction was refluxed for
then well extracted wi t h ether.
The ether
layer was dried over anhydrous potassium carbonate and
distilled to give 10.1 grams of product.
80$, b.p. 148-150° at
31.19; M R
The yield was
mm.; nj u 1.4211; d^ 0.9432;
for C 6H 1 2 02 : C, 62.4; H, 10.5.
C, 62.1; H, 10.4.
A 2 ,4 -dinitrophenylhydrazone was prepared by the
action of a solution of 2 ,4 -dinitrophenylhydrazine in
acidified 50$ aqueous ethanol.
After recrystallization
from p etroleum ether it melt e d at 86-87° and depressed
the melting points of the derivatives of 2 -hydroxy- 2 m e thyl- 3 -pentanone prepared from the ketone and from the
methyl acetal.
Calcd. for C 1 2 H 1 6 H 4 0 5 : C, 48.6; H, 5.4.
C, 48.3,
H, 5.7.
The a c t i o n of methanolic HC1 on 3-hydroxy-3-methyl2-butanone.
T e n grams
(0.1 mole)
of 3-hydroxy-3-methyl-
2-butanone was placed in a 50 cc. flask w i t h 10 cc.
of absolute methanol.
To this mixture was added
5 drops of a saturated solution of HC1 in methanol.
condenser w a s t h e n attached and the mixture refluxed
for one hour.
During this time the reaction was pro-
tected from external moisture.
DistillationAgave 9.2
grams of methanol boiling betw e e n 63.5 and 75° and two
fractions of product:
Fraction 1. 2.0 grams b.p. 135-138° n ^
Fraction 2. 7.0 grams b.p. 138-139° n ^
The intermediate cut,
1 was u s e d
75-135° was negligible.
in the preparation of a semicarbazone, using
a solution of semicarbazide hydrochloride buffered wi t h
sodium acetate.
The derivative melted at 164-165°,
same as that of the starting material.
point was not depressed.
A mixed melting
Fraction 2 has exactly the
physical constants of the starting material.
It also
gave a semicarbazone w h i c h was identical w i t h that of
the starting material.
The action of dimethyl sulfate upo n 3-h.ydroxy-3-
m e thyl - 2 - b u t a n o n e .
A solution of 37.1 grams
of the hydroxy ketone
in 150 cc. of water was
To this was added 16.0 grams
(0.4 mole)
sodium hydroxide and the resulting solution cooled in
an ice hath.
The. solution-was stirred mechanically
and 22.9 grams
(0.2 mole)
of dimethyl sulfate added
dropwise over a period of one hour.
The mixture was
stirred for three additional hours and then steam dis­
Seventy-five cc. of distillate were collected,
including 5 cc.
of an oil layer,
The distillate was
lighter than water.
saturated w i t h potassium carbonate
and w e l l extracted with ether.
The ether layer was
dried over anhydrous potassium carbonate and fraction­
ated to give 27 grams of a product boiling at 138° and
having a refractive index of 1.4152-1.4155.
stants are those of the starting material.
These con­
A semiear-
bazone was prepared in the same manner as in the experi­
ment immediately above.
The melting point,
and mixed
melting point w i t h a sample prepared from the starting
material, was 164-165°.
Preparation of starting materials u s e d in the
foregoing work.
Meth y l isopropyl ketone was prepared b y the hydrol­
ysis of trimethylethylene dibromide according to the
directions of Whitmore, Evers,
a nd Rothrock (35).
g-Bromo-3-methyl-2-butanone was prepared by a m o d ­
ification of the method of Favorskii
(86 grams)
One mole
of methyl isopropyl ketone was placed in a
three-necked flask fitted wi t h a gas inlet tube reaching
to the bottom,
a dropping funnel,
and a reflux condenser
whose outlet v/as fitted wi t h a water trap.
The ketone
was cooled in an ice ba t h and one mole of bromine was
added at such a rate that the solution was never colored
by the bromine.
During the addition carbon dioxide was
bubbled through at a slow rate.
After addition was
complete the ice bath was removed and the mixture al­
lowed to w a r m to room temperature.
Passage of carbon
dioxide v/as continued until the exit gases contained only
a small amount of HBr.
Fractionation of the mixture
gave from 105 to 125 grams
at 44-45° at 20 mm.
(65 to 75$)
of product boiling
The product had n^ ^ 1.4585.
was stable if protected from air.
A n alternative meth o d was improvised when no carbon
dioxide was available.
as above,
The bromination was carried out
omitting the passage of gas, and the product
poured into w a t e r as soon as addition was complete.
washing twice with water the oil was dried over CaClg
and fractionated as above.
The yields were substantially
the same, w i t h the disadvantages that troublesome
emulsions sometimes formed, an d that the crude material
is strongly lachrymatory*
Ethyl isopropyl carhinol was prepared by the method
of Favorskii
(20), u s i n g ethylmagnesium bromide and iso-
Decompositio n of the complex and frac­
tionation of the products gave 80$ yield of carbinol,
b.p. 125-128 at 730 mm.
Ethyl isopropyl ketone was prepared by oxidation
of three and one-half moles
(356 gms.)
of the corres­
ponding carbinol v/ith one and one-sixth moles
sodium dichromate and five moles
2 liters of water.
(490 gms)
(350 gms.)
of HgSO^ in
Steam distillation of the reaction
mixture and fractionation of the oil layer
(325 gms.)
from the distillate gave 75$ yield of the ketone, b.p.
2-Bromo-2-methyl-5-pentanone was prepared by bromination of ethyl isopropyl ketone,
following the first
method given above for 3-bromo~3-methyl-2-butanone.
Fractionation gave:
Fractions Weight
21 mm.
75 gm.
Boiling point
No residue.
Fractions 1-7 were m o stly unbrominated ketone.
Fractions 8-13 were the
desired product,
toiling point r eported by Favorskii
having the
F raction 14 was intermediate.
Fractions 15-20 were apparently 2,4“ di'bror 2-methyl-pentanone.
for CgH-LQOBrg: Br,
62.0. Found: Br,
It is peculiar that apparently no secondary hromo-ketone
was formed, as this w o u l d have "been expected to boil
between the tertiary bromo-ketone and the d i b r omo-ketone,
where only a small intermediate fraction was found.
Methyl sec-butyl-carbinol was prepared by the ad­
dition of four moles of acetaldehyde to four moles of
sec-butylmagnesium chloride.
The complex was decomposed
and worked up in the usual manner.
The product was not
distilled, but instead the greater part of the ether
was distilled off on a steam bat h and the residue
stripped through a column until the head temperature
r eached 105°.
The column was then allowed to drain and
the pot residue u s e d as methyl sec-butyl carbinol in the
preparation of rnethyl sec-butyl ketone below.
Methyl sec-butyl ketone was prepared by the oxidation
of the crude methyl sec-butyl carbinol prepared above
w i t h one and one-third moles
(400 gms)
and five a n d one-third moles
(525 gms.)
sodium dichromate
of sulfuric
acid in 2-|- liters of water.
The theoretical amount of
oxidant was u s e d assuming that the entire weight
the residue was carbinol.
This oxidation was carefully
kept at room temperature or below at all times,
excess Me OH was added to d e s t r o y any unused oxidant
before steam distillation of the product.
of the oil layer from the steam distillation g a v e .50$
yield b a sed on the acetaldehyde used.
The product
boiled at 116° at 740 mm. and ha d n^® 1.4002.
3-Bromo-3-methyl-2-p entanone was prepared b y bromination of methyl sec-butyl ketone according to the
first method given above for 3-bromo-3-methyl-2-butanone
Fractionation of the product gave 40 grams of fore-run
and 100 grams
n^A 1.4637;
(56fo) of product b.p.
df° 1.3142; M R
57° at 19 mm;
37.56; M R (found)
for CgH-^OBr: Br, 44.7.
Found: Br, 44.4.
Miscellaneous reactions
The reaction of diaoetyl wit h methylmagnesium
i odide.
It was thought that the reaction of one mole of
m ethylmagnesium iodide w i t h one mole of diacetyl would
provide a good m e thod of preparation of 3-hydroxy-3methyl-E-butanone.
However, w h e n the reaction was tried
no recognizable product was obtained,
the alkaline Grig-
nard reagent apparently polymerizing the diacetyl.
details of the attempt are given here:
A solution of one mole of methylmagnesium iodide in
500 cc. anhydrous ether was prepared and transferred to
a separatory funnel under nitrogen pressure.
This solu­
tion was then slowly added to a solution of one mole of
diacetyl in 500 cc.
of anhydrous ether.
The sortition
turned bright red and a red solid separated.
By the
time the addition was complete the solution was nearly
U p o n decomposition w i t h aqueous HC1, a dark solid
formed in the water layer, while the ether layer be ­
came bright red.
U p o n distillation the ether layer
turned to a viscous gum as the solvent was removed.
Neither the gum nor the solid could be crystallized from
common solvents.
The action of dimethyl sulfate and dimethylaniline
on 3-hydroxy-3-methyl-2-but anone.
Since attempts to etherify this hydroxy-ketone
w it h dimethyl sulfate in water and w i t h methanol and
HC1 failed,
it was determined to attempt a reaction
w i t h dimethyl sulfate in w h i c h no water was present.
Dimethylaniline was added to this reaction in order to
prevent the sulfuric acid, w h i c h is a product of dimethyl
sulfate m e t h y l a t i o n s , from attacking the ketone.
reaction was apparently a total failure.
The details
of the experiment were:
Twenty-five cc.
of dimethyl sulfate
was added to a cooled mixture of 30 gm.
hydroxy ketone and 50 cc.
(0.22 mole)
(0.3 mole)
of dimethylaniline.
The mi x ­
ture was placed und e r a column and allowed to w a r m slow­
Heating was begun after one hour.
ature rose to 67° and a small amount
alcohol came off.
The head temper­
(0.5 cc) of methyl
Ho further distillate came over at
a pot temperature of 250°, w i t h the column at 200°.
The reduction of 5~bromo-3-methyl-2-butanone with
zinc and acetic a c i d .
Because of the tendency of this bromo-ketone to.
rearrange during reactions w h i c h involve removal of the
bromine atom,
it seemed advisable to investigate the
of reduction of this ketone.
Should rearrange­
ment take place in the manner noted before,
either tri­
met hylacet aldehyde or neopentyl alcohol should be the
Because of the extreme ease of hydrolysis of
this bromo-ketone in alkaline solutions the reduction was
carried out b y means
of zinc in acetic acid.
arrangement was found,
the product being exclusively
methyl isopropyl ketone,
No r e ­
formed by reductive dehalogen-
The details of the experiment follow*
One h u ndred grams of granulated zinc was placed
in a flask w i t h 200 cc.
solution 82.5 grams
slowly added.
of glacial acetic acid.
(0.5 mole)
To this
of the bromo-ketone was
The reaction was allowed to run for three
hours, then filtered and the liquid layer distilled up
to a temperature of 118°.
Sufficient distillate was
taken off at this temperature to ensure removal of all
(b.p. 113°).
or neopentyl alcohol
The distillate was neutralized with
sodium hydroxide and the non-aqueous layer dried and
The product amounted to 37 grams
of methyl isopropyl ketone, b.p. 90° to 93°, n ^
No product was observed outside this temperature
1) The reaction of 3-bromo-3-methyl-2-butanone with
excess methylmagnesium iodide has b e e n shown to
give only pentamethylethanol.
2) The reaction of tertiary alpha-bromo-ketones with
sodium alcoholates in absolute alcohols has been
shown to yield the corresponding hydroxy acetals.
3) A new m e thod of synthesis of tri-alkyl acetic acids
has been found, by rearrangement of the tertiary
alpha-bromo-ketones with sodium methylate in ether.
4). The etherification of the hydroxyl group in tertiary
alpha-hydroxy ketones has been investigated.
Kohler and Tishler,
Tiffeneau, Ann.
J. Am. Chem.
(8), 10,
3. T iffeneau and Tchoubar,
4. Fisher,
Soc. 57,
J. Am.
1 9 8 , 941,
Chem. Soc.
5. Bouveault and Chereau,
Compt. rend.
and Fuson,
Compt. rend. 142, 1986,(1906)
6 . Bartlett and Rosenwald,
J. Am.
Chera. Soc. j56, 1990,
Kohler and Tishler,
i h i d . , 54,' 1594,
Kohler and Johnstin, Am.
Chem. J. 33, 45,
9. L owenhein and Schuster, Ann. 4 8 1 , 106,
10. Umnowa,
J. Russ. Phys.
Chem. Zentr.
11. Harriman, Ph.D.
Menard, Ph.D.
The Penna.
J. Chem.
17. Fisher, Ber.
State College,1938
State College,
1 9 6 , 2013,
60, 1148,
(1933); idem,
15. Bergmann and Mickely,
16. Ward,
The Penna,
J. Am.
1 9 7 , 1225,
Soc. 45,
84, I, 1402,
13. Henze and Whitney,
14. Rotbort,
64, 802,
Soc. 1 9 2 9 , 541.
18. Irvine and McNieoll,
J. Chem.
19. Froning and Hennion,
J. Am.
20. Favorskii,
J. prakt.
1 9 P 8 , 1604
62, 653,
88, 641,
Schmidt and Austin, Ber. 35, 37£1,
2£. Chichihabin,
Chem. Abstr.
£3. Kranzfelder and Vogt,
£4. Kohler, Richtmeyer,
8, 1773,
J. Am.
and Fuson,
60, 1714,
J. Am. Chem.
3 1 8 1 , (19£7); Kohler and Richtmeyer,
Soc. 49,
ibid., 52,
Henze and Caloway,
Jr. Am. Chem.
Aston and Burgess, Unpubl i s h e d work, this laboratory.
£7. Favorskii and Bozhtbvskii,
46, 1097,
50, 58£,
9, 1900,
(1920); ibid., 18, 1476,
Chim. 39, 216, (1926)
J. Am.
31. Helferich and Schafer,
32. Bartlett,
J. Russ. Phys.
(1914); Chem. Abstr.
30. Norris and Rigby,
Soc. 61, 1355,
54, 2088,
Org. Syntheses,
Coll. Vol.
J. Am.
33. Whitmore et al, ibid.,
34. Freon, Ann.
35. Whitmore, Evers,
Soc. 57_, 224,
60, 2462, 2788,
and Rothrock,
Syntheses, 13,
6 8 , (1933)
Part II
The Synthesis of n-Butane
This w o r k was u n d e r t a k e n in order to prepare a
pure sample of n-butane for the determination of its
thermodynamic properties by the workers in the Cry­
ogenic laboratory.
Work of that nature requires a
sample of utmost purity,
a property seldom found in
commercially available hydrocarbons.
This w o r k was
directed toward the preparation of such a sample.
M a n y methods are available for the preparation of
as may be seen from the list of workers who
have prepared this compound (1).
From the variation
of physical constants reported (1), it is apparent that
ease of preparation has been preferred to purity in
m any cases.
The best samples previously prepared are
those of Huffman, Parks,
and Barmore
(3) who fraction­
ated a commercial product, Burrell and Robertson (4)
who reduced n-butyl iodide w i t h a zinc-copper couple,
Jenkins, Burdick,
and Timm (5) who isolated
their sample from natural gas, Kohlrausch and Koppel
(6) w h o decomposed n-butylmagnesium bromide with water,
and Coffin and Maass
(7) who hydrogenated the olefin
from the dehydration of n-butyl alcohol over alumina.
in this work it was of prime importance to
obtain a sample
free from impurities, wit h yield and
cost of starting materials occupying a place of secondary
especially since a large sample was not
The method finally decided up o n was the Wurtz re­
action, using ethyl iodide and metallic sodium.
reaction should give as by products ethane,
and possibly hydrogen,
ethyl alcohol,
and from possible
side reactions
and hydrocarbons of the Cg and
higher series*
These possible impurities are those
to be expected from , a)' free •radical mechanism of
reaction, b) possible introduction of moisture.
all of these impurities are more or less
easily removed.
The olefins may be removed by either
sulfuric acid or alkaline permanganate, reagents which
will also remove any ethanol which might appear.
ethyl iodide carried over would easily be removed by
alcoholic silver nitrate.
This would leave only ethane,
and hydrocarbons of Cg or higher.
boil sufficiently far from n-butane to make their com­
plete removal by fractionation possible.
H o reactions have been noted, and no mechanism seems
plausible which w o u l d allow a rearrangement to occur
during the Wurtz reaction.
This is especially desirable,
since the removal of any iso-butane by fractionation
would probably require the use of better low temperature
columns than are at present available, and in any case
removal to the desired degree would be impossible.
It is remarkable,
considering the results obtained
that so far as we can find, nowhere in the liter­
ature has this method of preparation been us e d by workers '
who desired a pure sample for physical measurements,
though Lowig (8) in 1860 reported the preparation of
n-butane by the action of sodium amalgam on ethyl iodide.
Contrary to popular belief,
the reaction of ethyl
iodide w i t h sodium is not spontaneous.
In fact a mi x ­
ture of the two substances may be refluxed without any
apparent reaction.
Addition of dibutyl ether to raise
the boiling point is without
Metallic potassium
reacts very slowly wi t h ethyl iodide.
Reference to a paper b y Michael
also observed this unreactivity.
(2) shows that he
He found that a few
drops of acetonitrile would catalyze the reaction, and
that once started,
it w o u l d proceed smoothly.
was found to be the case, the characteristic blue color
of the Wurtz reaction appearing where the acetonitrile
touched the sodium,
and rapidly passing through the
entire reaction.
B y this method we were able to obtain a sample with
less than 0.001 mole percent impurity.
Beside the ther­
modynamic w o r k the sample is to have another very inter­
esting purpose.
The density of the gas, together wi t h
the data of state, will yield a highly accurate value
for the atomic weight of carbon (accurate to 0.002 units).
The origin of the carbon is therefore
of interest.
The ethyl iodide was prepared from absolute ethyl alco­
hol obtained from the Unit e d States Industrial Alcohol
C o . , and was prepared by the hydration of ethylene from
cracking petroleum.
The reaction was carried out in a two-liter three­
necked round-bottomed flask equipped w i t h a stirrer wit h
a mercury seal, a dropping funnel and a reflux condenser.
The outlet of the condenser carried a Y tube,
one outlet
of w h i c h led back to the top of the dropping funnel to
serve as a pressure balance.
The other outlet of the Y tube led to a mercury
"safety” trap, w h i c h also served as a manometer,
thence to a purifying train w h i c h consisted of five
double gas washing bottles of the
"Gilman" type.
contained in order:
1) Concentrated sulfuric acid.
2) Saturated solution of potassium permanganate in
AzQffo potassium hydroxide.
3) Saturated solution of silver nitrate in alcohol.
4 )
5) Concentrated sulfuric acid.
(This trap was changed
frequently during the run).
From the last trap the gas was passed through a phosphor­
us pentoxide tube and then condensed in large traps
cooled in a mixture of "dry ice" and acetone.
The trap
outlet led to an open tube manometer and through another
sulfuric acid trap to a water pump fitted wit h a bleed
A bunsen burner served admirably for this last
The bleed v/as so adjusted that the reaction
end of the system was at atmospheric pressure.
required a vacuum of about ten millimeters at the out­
let end.
The reaction was carried out as follows:
(2.2 moles)
of cut sodium were placed in the flask
together wi t h 2 5 -grams of ethyl iodide.
of the ethyl iodide,
(total 312 grams,
placed in the dropping funnel.
was then added down the
The remainder
2.0 moles) was
One cc.
of acetonitrile
condenser and the system im­
mediately attached for washing and condensing the gas.
The reaction began immediately,
and was maintained by
slow addition of ethyl iodide from the dropping funnel.
It was found advisable to keep a water trough handy for
cooling the reaction in case refluxing of the ethyl
iodide became too violent.
The reaction takes about
five hours a n d gives 50 to 60 cc.
measured at ^80°.
sodium m a y be
The prodi
of condensed gas,
After the reaction is over the excess
;royed by addition of methanol.
fas purified b y passage through a
v/ash bottle system where it was scrubbed with:
1) Saturated silver nitrate in alcohol
2) Saturated potassium permanganate in 40$ potassium
3) Saturated silver nitrate in alcohol
4) Concentrated sulfuric acid
Concentrated sulfuric acid
The gas was th e n pass e d through a phosphorus pentoxide
drier and condensed in the "dry ice" trap.
Of this
p urification the first 10 cc. of liquid was distilled
into and through the train before the condensing traps
were attached and the last 10 cc. of liquid allowed to
remain behind.
Fractionation of the sample was carried out in the
low temperature column of the Cryogenic Laboratory.
The products of three such runs as described above v/ere
combined for fractionation and the middle cut taken for
This sample had the following physical constants:
272.66 ±
0 . 05°K
134.87 ±
Its p u rity was 99.999 £
0.001 mole jo as calculated from
observation of the melting point w i t h various fractions
over a period of twelve hours.
no solid solution.
This assumes
The weight of the cut from which
the sample was taken was 40 g m s . , or 33$ of the total
This is 22$ based on ethyl iodide.
Sincere thanks are due to Dr. G. H. Messerly of
the Cryogenic Laboratory who fractionated the n-butane
and determined its physical constants and purity,
to Mr.
James Fritz, who prepared the ethyl iodide used
in the synthesis.
A sample of n-lmtane of 99.999 mole $ purity was
p r epared and its physical constants determined.
1 . Eglof,
"Physical Constants of Hydrocarhons",
Reinhold Publishing Co., Hew York,
2. Michael,
Jr. 25, 419,
3. Huffman, Parks, and Barmore,
J. Am. Chem. Soc. 5 3 ,
4. Burrell and Robertson,
5. Dana,
ibid., 37, 2188,
Jenkins, Burdick, an d Timm, Ref. Eng.
6 . Kohlr a u s c h and'Koppel,
Z. Physik.
2 6 B ’, 209,
7. Coffin and Maass,
8 . Lowig,
J.Am. Chem.
Jahr, u. fort.
50, 1427,
Chem., 1 8 6 0 , 397.
Без категории
Размер файла
2 348 Кб
Пожаловаться на содержимое документа