organocadmiums2ketones

organocadmiums2ketones, biotransformation

[ Pobierz całość w formacie PDF ]
THE USE OF ORGANOCADMIUM REAGENTS FOR THE
PREPARATION OF KETONES
JAMES CASOS
Department
of
Chemistry, University
of
California, Berkeley, California
Received
August
86,
1946
CONTENTS
Introduction.,
....
..................................................
000
............
000
I.
11.
111.
2.
Ester
groups..
...
............
............
000
............
000
............
000
............
000
1.
Ester formation..
..............
IV
.
V.
VI.
.............................
000
I. IKTRODUCTION
The preparation of ketones by the reaction betxeen an organocadmium com-
pound and an acid chloride was first recommended by Gilman and Nelson (19)
in 1936.
The following reactions are involved in the preparation:
RBr
+
Mg
RMgBr
RCdR
+
MgBrz
+
MgClz
2RRIIgBr
+
CdClz
RCdR
+
BR’COCL
+
BRCR’
+
CdClz
The cadmium reagent may be prepared from a lithium derivative instead of a
Grignard reagent, if desired (19). This sequence may be carried through in
a
few hours without isolation of intermediates; however, the general utility of the
method mas not recognized for several years, possibly on account of experimental
difficulties encountered in carrying out these reactions
(cf.
Section IV). At
least one report (38) of a low yield in such a preparation has appeared. Since
1941, however, considerable information concerning this reaction has been
gained, and the method may nom be regarded as one
of
the best and most widely
applicable procedures
for
laboratory-scale preparation of either simple ketones
or
various polyfunctional compounds containing a keto group.
The required
15
16
JAMES CASON
starting materials are relatively easily available, the experimental procedure is
equivalent to a one-step process, and the yields are usually rather good.
WITH
OTHER METHODS
Although simple symmetrical ketones may be conveniently prepared from
the appropriate acid or its salt, the most useful methods for preparing unsym-
metrical ketones involve a Friedel-Crafts reaction, an enolate condensation, or
some type of Grignard reaction. It is of interest, perhaps, to consider the rela-
tive merits
of
these methods
as
compared with the use of organocadmium re-
agents.
The Friedel-Crafts reaction is quite useful in cases where
it
is applicable.
However, it is useful only in the aromatic series and with compounds containing
no meta-directing groups; also, only certain orientations may be obtained. In
the latter two categories mentioned above, the following general reactions may
be considered
:
11.
COMPARISON
(1)
hydrolysis
(2)
decarboxylation
+
CH3CCH2R
I/
0
C
02
C2Hs
e
CH3C-CCO2C2H~
+
R’COC1
+
C1
8
(2)
C€LC-(!F-CR’
-+
II
I
II
OR0
(1)
cleavage of acetyl
(2)
II
I
OR
-+
RCHzCR’
II
hydrolysis
0
(3)
decarboxylation
COzCzHs
COzCzHs
+
RCOCl
le
(3)
CH
I
C02CzHs
+c’l
RC-CH
II I
0
COzCzHs
(2)
decarboxylation
’
hydrolysis
RCOCHI
hydrolysis
(1)
(4)
RMgX
+
R’C=N
+
imine
-+
RCOR’
hydrolysis
(5)
RMgX
+
R’CONHz
imine
+
RCOR’
---f
-
70°C.
(6)
RMgX
+
(CH3CO)zO
-+
RCOCH3
+
CH3COzMgX
RZnCl
+
R’COC1
+
RCOR’
+
ZnCl2
(7)
or
RzZn
17
ORGANOCADMIUM REAGENTS IN PREPARATION
OF
KETONES
Method
1
has been widely used for the preparation of ketones; however, it
involves two separate operations and the over-all yield of ketone rarely equals
the yield obtained by use of the organocadmium reagent. This method may be
extended to the preparation of keto acids
(7, 32, 36)
if a bromo ester is used
instead of an alkyl halide; however, the difficult availability of most types of
bromo esters is often a handicap. On the other hand, keto esters may be pre-
pared in one step by the reaction of an organocadmium reagent with the ester
acid chloride
of a
dibasic acid, C2H602C(CH2),COC1,
a type of starting material
which is relatively easily obtained
(1
1, 39).
Method
1
has the advantage. how-
ever, that it may be used for the preparation of keto acids having a branching
group between the keto and carboxyl groups.
The chief usefulness of method
1
is in the preparation of methyl ketones, for
higher @ketoesters are obtainable
(5,30,36)
only by fairly laborious procedures.
For this reason, method
2
has found considerable application in the preparation
of other types of ketones, especially keto acids
(32, 36).
The chief difficulties
with the method are that the over-all process is laborious and the yields are not
especially good.
Side reactions are oxygen-acylation and cleavage of the higher
acyl group.
Method
3
is limited
to
the preparation of methyl ketones, for acylation of
alkylmalonic esters
(23,40)
gives
a
very poor yield, presumably owing to oxygen-
acylation. For methyl ketones this method gives good yields, and it is about
as useful as the cadmium method.
For such ketones, there seems little basis for
choice between the two methods.
Methods
4
(2, 17, 18)
and
5
(17, 22, 38)
give good yields of ketones, provided
there is present in the molecule no other functional group which will react with
the Grignard reagent. The nitrile and amide groups are among the least reac-
tive toward this reagent. If this condition is met these methods give yields as
good
as
or better than the cadmium method, and the experimental manipula-
tions are similar. The scope of the cadmium method is much greater however,
for the organocadmium reagents fail to react with most functional groups (cf.
Section 111,D).
Method
6
has been introduced only recently
(27,28)
but seems quite promising
as a general preparative method for methyl ketones. The yields are in the same
range
as
those obtained from cadmium reagents. It is also reported that this
method is satisfactory with secondary and tertiary alkyl halides, types which are
not suitable for conversion to dialkylcadmium reagents.
Method
7
is identical with the preparation involving cadmium, except that
zinc is used. Although Blaise
(3)
originally claimed excellent yields in this type
of preparation,
a
succession
of
investigators
(25, 26, 33, 34)
has failed to obtain
better than moderate yields. The superiority of the cadmium reagent arises
from its much easier preparation and its lower reactivity toward the carbonyl
group. Diethylzinc has been shown
(19)
to react with the carbonyl group
about
3.5
times
as
fast as does diethylcadmium.
A
few ketones have been made
from both cadmium and zinc reagents
(15,
19),
and in each case the yield was
significantly larger when the cadmium reagent was used.
Thus, it seems safe to say that no other method of ketone preparation has the
18
JAMES CASON
scope of the method utilizing the organocadmium reagents, and in relatively few
specific cases are other methods superior.
The actual scope and limitations of
this method will be considered next.
111.
SCOPE
AND
LIMITATIONS
OF THE
METHOD
A. THE ALKYL
OR
ARYL HALIDE
1.
The
halogen
Several investigators (12, 16, 19) have reported that if the Grignard reagent
from which the cadmium derivative is prepared is made from an iodide, the
yield of ketone is much poorer than if the bromide is used. It has also been
reported (12) that in the case of n-butyl halides the chloride is somewhat inferior
to the bromide, although much better than the iodide. In the preparation of
methyl 4-ketooctoate from di-n-butylcadmium and P-carbomethoxypropionyl
chloride, the yields mere respectively 79.5,
63,
and 45 per cent when the halides
(C4Hs)zCd
+
2ClCO(CH,)zCO&H3
+
C4HsCO(CHz)zCOzCH3
+
CdClz
mere butyl bromide, chloride, and iodide. Cole and Julian (15) reported the
preparation of a methyl ketone using dimethylcadmiurn, and obtained the same
yield when either methyl iodide or methyl bromide was used; however, a fivefold
excess of cadmium reagent was used in this preparation,
so
the relative value
of the halides may have been obscured. This seems probable, since De Benne-
ville (16) reported methyl iodide inferior to the bromide.
2.
The organic
radical
Aromatic Grignard reagents form cadmium derivatives readily (12, 15, 16, 19)
and give good yields of aromatic ketones. If the organic radical is alkyl, then
it must be primary in order for the preparation to be useful. This seems to be
the chief limitation of the organocadmium reagents. Gilman and Nelson (19)
first reported that secondary and tertiary cadmium derivatives are stable only
at low temperature. Other investigators (12, 14) have found that in the prep-
aration of methyl 4-keto-5-methyloctoate from di-2-amylcadmium and P-carbo-
methoxypropionyl chloride, the yield of keto ester is nearly zero if the reaction
(
Cd
+
2CICO(CHz)zCOzCHs
--+
C~H~CHCO(CHZ)ZCO~CH~
C3H7 CH-
I
CH3
AH3
z
+
CdClz
temperature is allowed to rise somewhat above 0°C. (12a), and the yield is only
21.5 per cent when the reaction is conducted at
-5"
to -10°C. Diisopropyl-
cadmium seems to be somewhat more stable, for Gilman and Nelson, operating
at O"C., obtained n-propyl isopropyl ketone in 60 per cent yield:
CH3CH-
Cd
+
2CaH7COCl
+
CdClz
+
CHSCHCOCaH7
(
I
AH3
Z
CHI
19
ORQANOCADMIUM REAGENTS IN PREPARATION OF KETONES
Di-tert-butylcadmium has been used in the reaction with acetyl chloride (19)
at -70°C. to give a 17 per cent yield of ketone. De Benneville (16) reported a
40
per cent yield (based on anhydride) in the reaction, at -7O"C., between
di-
tert-butylcadmium and benzoic anhydride
;
however, in view of the low reactivity
of cadmium reagents toward anhydrides (cf. Section II1,C) and the preparation
of ketones from Grignard reagents and anhydrides at -70°C. (27), it seems
quite possible that the ketone obtained in this case resulted from Grignard
reagent rather than cadmium reagent.
It may be mentioned at this point that Nightingale and Wadsworth (29) at-
tempted the conversion of phenylethynylmagnesium bromide (I) to the corre-
sponding cadmium derivative.
After stirring for
1
hr. in ether the test for
OH
I
6H.5
CaH6C=CMgBr
C
6H5
CECCC~CC
I
CH3
I1
Grignard reagent was still positive, and when the resultant mixture was treated
with acetic anhydride or acetyl chloride the only product isolable (in 14 per cent
yield) was that expected from the Grignard reagent,
bis(phenylethyny1)methyl-
carbinol (11).
I
B.
THE CADMIUM HALIDE
It has been shown (19) that cadmium chloride is at least as effective as is cad-
mium bromide for preparing the cadmium reagents. Since the bromide is more
expensive and much more hygroscopic, the chloride is always used. It has also
been reported (19) that the yields are approximately the same from a dialkyl-
cadmium compound and an alkylcadmium halide;
so
the former is customarily
used, since it requires only half as much cadmium chloride.
C. THE ACID CHLORIDE OR &'HYDRIDE
Organocadmium reagents will react with acid anhydrides
(10,
12, 14, 16)
;
however, the reaction seems to be uniformly inferior to that obtained with acid
chlorides. The most interesting reaction with anhydrides is that with cyclic
anhydrides.
Unfortunately, the yields of keto acids (on the basis of alkyl or
-
-
CHzC\
//O
R2Cd
+
2
1
0
2RCCHz CHz COzH
H20+
II
0
CH2 C\
/
0
aryl halide) in this reaction have been only
25-30
per cent;
so
unless the halide
is quite cheap, it seems profitable to convert the anhydride to the half-ester and
this to the ester acid chloride
(11,
39).
Reaction of the latter compound with
the cadmium reagent gives the keto ester corresponding to the keto acid obtained
[ Pobierz całość w formacie PDF ]

zanotowane.pl

doc.pisz.pl

pdf.pisz.pl

anio102.xlx.pl