These belong to a new thread as they have little to do with Sommelet reaction.
Here is Tetrahedron Letters, 21, 2869-2872 (1980) scanned and OCR'ed:
REINVESTIGATION OF THE GRIGNARD REACTIONS WITH FORMIC ACID.
A CONVENIENT METHOD FOR PREPARATION OF ALDEHYDES
Fumie Sato, Kaoru Oguro, Hiroshi Watanabe, and Masao Sato
Department of Synthetic Organic Chemistry, Faculty of Engineering
Kyoto University, Kyoto, Japan
Summary: Grignard reagents react with formic acid in tetrahydrofuran
to produce aldehydes in relatively good yields. Various aldehydes
such as alkyl, aryl, allyl, benzyl and vinyl aldehydes were prepared
from the corresponding Grignard reagents. The reaction with vinyl
Grignard reagents proceeded with retention of configuration.
Many methods have been reported for the preparation of aldehydes by
the reactions of Grignard reagents with masked formic acid or formal
derivatives1. However, the reaction with formic acid itself,
which is obviously the most direct approach, was reported to afford
aldehydes only in very low yields2,3, and, therefore, has not
been used for aldehyde synthesis.
Recently, a striking solvent effect of THF on Grignard reactions
with acid chlorides was observed. In marked contrast to the reaction in
ether, the reaction in THF gave ketones almost quantitatively. This finding
prompted us to reinvestigate the reaction of Grignard reagents with formic
acid in THF in expectation of production of aldehydes in better yields than
In the preliminary investigation, the reaction of two moles of
hexylmagnesium bromide with a mole of formic acid in THF was carried out,
and a satisfactory yield (72%) of heptanal was obtained. However, rather
suprisingly, heptanal was found to be produced in a 55% yield even when
ether was used as the solvent. The yield was lower than for the reaction in
THF but not so low as previously reported2,3. Though the precise
reason is not clear, we believe that insufficient drying of formic acid was
responsible for the low yield of aldehydes in the previous investigations.
In view of the potential economy of this aldehyde synthesis in THF,
we decided to carry out the reaction which gives the magnesium salt of
formic acid by treatment with an equimolar amount of a readily available
Grignard reagent such as ethylmagnesium bromide, and then to treat the
magnesium salt with one mole of the desired Grignard reagent as shown in eqs
1 and 2.
(1) HCOOH + C2H5MgBr ---> HCOOMgBr + C2H6
(2) HCOOMgBr + RMgBr ---> HCR(OMgBr)2 ---H2O--> RCHO
In a representative procedure, ethylmagnesium bromide in THF (0.84 M; 16 ml,
13.4 mmol) was added dropwise (20 min) to a solution of formic acid (0.45
ml, 11.9 mmol) in dry THF (15 ml) while stirring under argon at 0°C.
Hexylmagnesium bromide (0.79 M; 10.5 ml, 8.3 mmol) was then added (10 min)
to this solution, and the reaction mixture was stirred for 30 min at room
temperature. The reaction mixture was decomposed with 2N HCl, extracted
with ether, dried over Na2SO4, and then distilled
under reduced pressure (62°C, 35 Torr) to afford heptanal (0.72 g, 75%
The presumed intermediate (1) seemed stable under the reaction
conditions and was converted to aldehyde only by hydrolysis, as the
1H NMR analysis of the reaction mixture in THF showed no peak due
to formyl protons before hydrolysis, whereas the peak appeared after the
addition of 2N HCl to the sample in the NMR tube.
As illustrated by the entries in Table 1, the method can be readily
extended to various Grignard reagents such as alkyl, aryl, benzyl, allyl and
vinyl Grignard reagents. The listed yields were obtained using the above
procedure without optimization for each substrate. It can be seen that the
yields were satisfactory in all cases except for secondary aliphatic
Grignard reagents. The reaction with vinylic Grignard reagents followed by
hydrolysis under neutral conditions (by addition of H2O)8
afforded an alpha,beta-unsaturated aldehyde with the same
stereoconfiguration as the starting Grignard reagents, indicating that the
reaction of vinylic Grignard reagents with formic acid proceeds with
retention of configuration. Unfortunately, even if a Grignard reagent is
prepared from a pure (E) or (Z)-1-alkenyl halide, it becomes a mixture of
(E) and (Z)-1-alkenyl- magnesium halides9,10,11. Therefore, pure
(E) and (Z)-2-alkenals could not be obtained directly. However, as
(Z)-2-alkenal is easily isomerized to the stable (E)-isomer by acid, pure
(E)-2-alkenal may be obtained easily by the present method. Thus, pure
(E)-cinnamaldehyde was prepared from the Grignard reagent from either (E) or
(Z)-beta-bromostyrene by hydrolysis with 2N HCl after the reaction. Most
aldehyde syntheses by the reactions of Grignard reagents with formic acid
derivatives in the literatures need the acidic hydrolysis step. Therefore,
it seems difficult to prepare (Z)-2-alkenals by these methods. Therefore, it
is important to note that the present aldehyde synthesis makes it possible
to prepare (Z)-2-alkenals from (Z)-l-alkenylmagnesium halides from either
(E) or (Z)-beta-bromostyrene by hydrolysis with 2N HCl after the reaction.
Most aldehyde syntheses by the reactions of Grignard reagents with formic
acid derivatives in the literatures need the acidic hydrolysis step.
Therefore, it seems difficult to prepare (Z)-2-alkenals by these methods.
Therefore, it is important to note that the present aldehyde synthesis makes
it possible to prepare (Z)-2-alkenals from (Z)-l-alkenylmagnesium halides.
RMgBr RCHO Yield %
C6H13MgBr C6H13CHO 75
C8H17MgBr C8H17MgBrCHO 70
C3H7CH(CH3)MgBr C3H7CH(CH3)CHO 38
BrMg-(CH2)4-MgBr OHC-(CH2)4-CHO 55
C6H5MgBr C6H5CHO 81
C6H5CH2MgBr C6H5CH2CHO 61
C3H7-CH=CH-CH2MgBr C3H7-CH(CHO)-CH=CH2 50
E/Z = 15/85 E/Z = 16/84 60
E/Z = 65/35 E/Z == 65/35 58
E/Z = 89/ll E/Z = 87/13 67
References and Notes
1. For a recent review, see C.A. Buehler and D.E. Pearson, "Survey of Organic Syntheses" vol
2. N.D. Zeiinsky, Chem. Ztg., 28, 303 (1904).
3. J. Houben, Chem. Ztg., 29, 667 (1905).
4. L.I. Smith and J. Nichols, J. Org. Chem., 6, 489 (1941).
5. F. Sato, M. Inoue, K. Oguro. and M. Sato, Tetrahedron Lett., 4303 (1979).
6. Formic acid dried over boric acid anhydride by the method of Schlesinger and Martin was
7. H.I. Schlesinger and A.W. Martin, J. Amer. Chem. Soc., 36, 1590 (1914).
8. Instead of 2N HC1, H2O was used for hydrolysis in the case of vinylic Grignard reagents;
9. B. Mechin and N. Naulet, J. Organometal. Chem., 39, 229 (1972).
10. T. Yoshino and Y. Manabe. J. Amer. Chem. Soc., 85, 2860 (1963).
T. Yoshino, Y. Manabe, and Y. Kikuchi, J. Amer. Chem. Soc., 86, 4670 (1964).
11. G.J. Martin and M.L. Martin, Bull. Soc. Chem., 1636 (1966).
12. R.A. Raphael and F. Sondheimer, J. Chem. Soc., 2693 (1951).
I also found this. Tetrahedron Letters, 25, 1843-1844 (1984):
FORMYLATION OF ORGANOMETALLIC COMPOUNDS WITH LITHIUM (OR SODIUM) FORMATE.
PART I. A FACILE SYNTHESIS OF ALDEHYDES FROM GRIGNARD REAGENT
M. Bogavac, L. Arsenijevic, S. Pavlov and V. Arsenijevic
Department of Organic Chemistry, Faculty of Pharmacy
Summary: Grignard reagent reacts with lithium (or sodium) formate in boiling
THF giving the corresponding aldehydes in good yields. This reaction can be
carried out at room temperature as well, but stirring of the reaction
mixture for two or three days is required.
Aldehydes can be obtained by the reaction of Grignard reagent with a
variety of compounds1 and, recently, with several N-formyl
amines2-4. Some of this formamides give the aldehydes free of
secondary alcohol by-product. It has been also shown that aldehydes can, be
prepared in good yields by the reaction of scrupulously anhydrous formic
acid with two moles of Grignard reagent, provided the reaction is carried
out in THF solution5. The earlier described synthesis of
aldehydes from formic acid [or copper(II) formate] and Grignard reagents, on
account of low yields, was of no preparative value6-7.
In this paper we have investigated the possibility of avoiding the
use of formic acid (the drying of this acid is a rather tedious process) and
have found that the formylation of Grignard reagent with lithium or sodium
formate is very convenient, one pot procedure, and gives aldehydes without
secondary alcohol as by-product ; also, in this reaction only one mole of
Grignard reagent is used and the transfer of the reagent to the dropping
funnel is unnecessary. These salts are commercially available or may be
easily prepared in the laboratory.
Alkaline salts of formic acid are very slightly soluble in THF, but
on heating, they react with Grignard reagent whereby the latter is added to
the formate carbonyl group. If the reaction is carried out in ether, the
yields are considerably lower, even after refluxing the reaction mixture for
three days. If a mixture of ether and THF (1:2) is applied, the yields are
approximately the same as those obtained in THF, but a longer heating is
required ; this is significant in cases when it is more convenient to
prepare Grignard reagent in ethereal than in THF solution.
Representative procedure: to 0.192 mole of Grignard reagent (prepared
from 4.8 g of magnesium, 37.4 g of 2-bromoanisol and 200 ml of THF), 11.5 g
(0.22 mole) of lithium formate is added and the reaction mixture is heated
with boiling in a nitrogen atmosphere until an almost clear solution is
obtained (about 2 h), the major part of THF is removed by distillation, 100
ml of ether and about 0.1 g of hydroquinone are added to the residue, and
the reaction mixture is decomposed with dilute HCl (cooled to 0°).
2-Methoxybenzaldehyde is isolated in usual way and distilled under nitrogen:
bp 114°C/17 mmHg, mp 36-7°C. The yield is 23 g (85%).
Entry RMgBr(Cl)8 R-CHO Yield %
1. C6H5MgBr C6H5CHO 85 79
2. C6H5MgCl*) C6H5CHO 80 72
3. o-CH3OC6H4MgBr o-CH3OC6H4CHO 85 76
4. (CH3)2CHMgCl (CH3)2CHCHO 80 75
5. p-BrC6H4MgBr p-BrC6H4CHO 83 69
6. C6H5CH2MgCl C6H5CH2CHO 79
7. 1-Naphthyl-MgBr 1-Naphthaldehyd 78
8. C6H5CH=CHMgBr C6H5CH=CHCHO 72
*)Prepared according to H. Normant8.
1. R.S. Brinkmeyer, E.W. Collington, A.I. Meyers, Org. Synth., 54, 42 (1974) and references therein.
2. D. Comins, A.I. Meyers, Synthesis, 403 (1978).
5. G.A. Olah, M. Arvanaghi, Angew. Chem., 93, (1981), 925.
4. W. Amaratunga, M.I. Erechet, Tetrahedron Lett., 24, 1143 (1983).
5. F. Sato, M. Inoue, K. Oguro, M. Sato, Tetrahedron Lett., 21, 2869 (1980).
6. N.D. Zelinsky, Chem. Ztg., 28, 303 (1904).
7. J. Houben, Chem. Ztg., 29, 667 (1905).
8. H. Normant, Bull. Soc. Chim. Fr., 728 (1957).
Moo, you've made the sun shine brighter today
What a find, especially the second article where they skip the anhydrous formic acid and work with plain formates of the alkali metals. I guess they choose lithium formate over sodium formate because the solubility of the lithium derivative in THF should be higher and, of course, Li is cross-related to Mg in the periodic table. Wonder about how potassium formate would behave in this reaction, as its solubility (in water, of course) is the highest of all three formates, 331g of potassium formate dissolve in 100g of cold (18°C) water, 657g dissolve in 100g hot (80°C) water, according to the "Handbook of Chemistry and Physics". The solubility of potassium formate in THF should also be higher (theoretically) than that of the Li or Na salt. Maybe the use of potassium formate could boost up the yield a little bit...
The reaction time (2h) of this "improved Sato synthesis" is short, the Grignard can be handled in the usual manner (very convenient) and most of the THF is recycled, the THF is anhydrous after recycling and can be directly used in a Grignard again. Yields of products are good to very good and this procedure should work where the Sommelet doesn't because of steric hinderance or because one wants to synthesize non-aromatic aldehydes. The greatest difference to the Sommelet is that the Sato (or the improved Sato) reaction is an addition, you will end up with an CHO group added to the molecule, not with an -CH2X group changed (oxidized) to CHO.
I really begin to like this reaction
Quidquid agis, prudenter agas et respice finem!