A green and selective reduction of aldehydes

Article abstract of DOI:10.1016/S0040-4039(00)00435-4

Aldehydes are reduced to alcohols with sodium formate in subcritical water at temperatures and pressures below those required to reduce cyclic ketones. Acyclic ketones afford only minor amounts of alcohol with sodium formate even under more forcing conditions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00435-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 3523–3526
A green and selective reduction of aldehydes
T. A. Bryson,a,∗ J. M. Jennings a and J. M. Gibson b
aDepartment of Chemistry and Biochemistry, University of South Carolina, Columbia, SC 29208, USA
bDepartment of Chemistry and Physics, Wingate University, Wingate, NC 28174, USA
Received 17 December 1999; revised 13 March 2000; accepted 14 March 2000
Abstract
Aldehydes are reduced to alcohols with sodium formate in subcritical water at temperatures and pressures below
those required to reduce cyclic ketones. Acyclic ketones afford only minor amounts of alcohol with sodium formate
even under more forcing conditions. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: selective aldehyde reduction; subcritical water.
Developments in highly specialized (chemo-, regio- and stereoselective) reagents can be at odds with
the goals of green chemistry.1 For example, tetrabutylammonium cyanoborohydride in acidic HMPA or
NaBH(StBu)3 elegantly provides selectively for the reduction of aldehydes in the presence of ketones,2
but reaction conditions and by-products are not environmentally benign.3 Here we describe a selective
reduction of aldehydes that is profoundly green4 and achievable with very simple equipment.5 We needed
an aqueous reducing agent that had some thermal and aqueous stability. Sodium formate has an extensive
history6 and was chosen based on its green by-products. Other possible hydrogen sources were judged
to be either environmentally unfriendly or were achieved by a complicated work-up. For example,
reductions with cyclohexene or cyclohexadiene7 would form benzene or reductions with alcohol8 in
water would change the critical point properties of water and yield a product(s) in alcohol interfering
with a green work-up and separation.
Indeed, aqueous HCO2Na in water at 250–350oC (∼1300 psi)9 without a co-solvent or added catalyst
will reduce aldehydes and a number of ketones. These conditions are below the critical point of
water but were at temperatures and pressures selected such that the organic substrate was believed to
have reasonable solubility.10 Yields of alcohol have not been optimized and are obtained generally by
distillation of the organic phase that has been separated from the water layer.11
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00435-4
tetl 16715

3524
Experiments comparing the amount of sodium formate to the efficiency of the carbonyl reduction have
leveled at 3 equivalents of sodium formate being generally favored. Even using 3 equivalents there is a
noticeable positive pressure when opening the cooled pressure reactor due to formed CO2. Running at
different temperatures reveals selectivity for aldehyde reduction. Aldehyde reduction is observed at 250
to 300°C and ∼1200 psi with reaction times up to 3 h; at 300°C or greater cyclic ketones show substantial
reduction, and acyclic ketones afford only poor conversion to alcohols at 340°C even with extended
reaction times. If selectivity is not an issue, higher temperatures (Table 1) and/or longer reaction times
generally improve yields. To verify the tunability of this reduction, a mixture of 1 and 2 (1:1, 310°C)
are treated with 3 equivalents of sodium formate effecting reduction of only benzaldehyde (to 3) (Table
2). Higher temperatures generally increase yields of primary alcohols but other processes (presumably
aldol, Cannizzaro reaction, and dehydration) become significant making separations difficult. Similarly,
reduction (350°C) of a 1 to 1 mixture of heptanal (5) and 2-heptanone (2) yields a mixture of ketone 2
and 1-heptanol (60% based on heptanal) with no 4.
Table 1
Aldehyde reduction with 3 equivalents of HCO2Na
Table 2
Representative ketones
This selectivity was further tested with 8-oxononal (7) and 2-(2-oxocyclohexyl)ethanal (8). Sodium
formate and 7 (290°C, 0.5 h) afforded a 7 to 1 mix of ketone 9 and diol 10 in 56%, while 8 (240°C, 0.5
h) formed 2-(2-hydroxyethyl)cyclohexanone (11), 78%, with less than 2% of diol 12 formed.13

3525
Typically, a 10 cc pressure reactor equipped with a thermocouple5 is charged with a 4 ml mixture of
degassed water, carbonyl compound (0.05 to 0.4 g, 1 equiv.) and sodium formate (3 equiv.) and flushed
with nitrogen. The reactor is placed in a preheated muffle oven, brought to and maintained at the targeted
temperature (250/310/340°C) for 0.5 h (or 3 h) and quickly cooled. The reactor is cautiously opened, the
organics separated by heavily salting the aqueous phase and the product distilled.14
Aldehydes and some ketones can be reduced with sodium formate under pressure and at elevated
temperatures in water without co-solvents or additional catalysts or reagents.15 Aspects of this process
and other green transformations are currently under study.
Acknowledgements
J. M. Jennings received a 1998 Ken Hancock Memorial Green Chemistry Scholarship generously pro-
vided by the ACS Division of Environmental Chemistry and supported by EPA and NSF. The University
of South Carolina, The SC EPA/EPSCoR Program (R826399-01-0) NSF (Cooperative Agreement EPS-
9630167) DoD (ONR-N00014-97-1-0806), Unilever Corporation and Wingate University are gratefully
acknowledged for their support of this work.
References
1. For Green Chemistry, see: http://www.epa.gov/opptintr/greenchemistry/whatis.htm.
2. (a) Hutchins, R. O. J. Amer. Chem. Soc. 1973, 95, 6131. (b) Maki, Y.; Kikuchi, K.; Sugiyama, H.; Seto, S. Tetrahedron
Lett. 1977, 263.
3. Trost, B. M. The atom economy — a search for synthetic efficiency. Science 1991, 254, 1471.
4. By-products are CO2 and H2O or NaOH.
5. Equipment: Two types of reactors have been employed: A Parr pressure reactor heated in a sand bath or oven and a capped
316 stainless steel tube heated in a muffle oven. The latter can be equipped with a thermocouple and fiber optic probe.
6. Johnstone, R. A. W.; Wilby, A. H.; Entwistle, I. D. Chem. Rev. 1985, 85, 129.
7. Reductions in Organic Chemistry, 2nd edn.; Hudlicky, M.; ACS Monograph 188, ACS: Washington, DC, 1996; p. 15, and
references cited therein.
8. For reduction of carbonyls with alcohols, see: (a) Bagnell, L.; Strauss, C. Chem. Commun. 1999, 287. (b) Malwitz, D.;
Metzger, J. O. Angew. Chem., Int. Ed. Engl. 1986, 25, 762. (c) Malwitz, D.; Metzger, J. O. Chem. Ber. 1986, 119, 3558.
9. Calculation of pressure: Introduction to Thermodynamics and Heat Transfer; Çenge, Y. A.; The McGraw-Hill Companies:
New York, 1997; pp. 849–855.
10. Connelly, J. F. J. Chem. Eng. Data 1966, 11, 13.
11. GC analysis of products indicates yields are much better than reported above.
12. Based on data from GC–mass spectra of the products, it appears aldol/dehydration (2,4-diphenylbut-2-enal) and subsequent
reduction (2,4-diphenylbut-2-en-1-ol) are the major reactions with this highly enolizable aldehyde.

3526
13. Product analysis shows 60% of 11 plus a mixture of enol ethers (18%) presumably formed from dehydration of the
hemiacetal on heating (GC or distillation). These same products formed on hydrolysis of the ketal of 11. Amounts of
12 were based on GC comparison to an authentic, epimeric sample.
1
14. Products were characterized by IR, 13C and H NMR, mass spectroscopy and chromatographic comparison to authentic
compounds. Some work-up procedures required pH adjustment and chromatography.
15. Preliminary results indicate that ethyl formate affords a similar selectivity for aldehyde reduction. The primary difference
observed, so far, between sodium formate (basic) and ethyl formate is that the latter tends to promote dehydration of certain
alcohol reduction products.