Direct Zn-diamine promoted reduction of C=O and C=N bonds by polymethylhydrosiloxane in methanol

Article abstract of DOI:10.1039/b210144k

Ketones and imines are chemoselectively reduced at room temperature in methanol to the corresponding alcohols and amines in high yields in a one-step procedure using polymethylhydrosiloxane (PMHS) and a simple zinc-diamine catalyst.

Full text of DOI:10.1039/b210144k

Direct Zn–diamine promoted reduction of CNO and CNN bonds by
polymethylhydrosiloxane in methanol
Virginie Bette,a André Mortreux,a Christian W. Lehmannb and Jean-François Carpentier*ac
a Chimie Organique Appliquée, UPRESA 8010 CNRS-Université de Lille 1, ENSCL, BP 108, 59652
Villeneuve d’Ascq Cedex, France
b Max-Planck-Institut für Kohlenforschung, Strukturchemie, Postfach 101353, 45466 Mülheim/Ruhr,
Germany
c
Organométalliques et Catalyse, UMR 6509 CNRS-Université de Rennes 1, Institut de Chimie, Campus de
Beaulieu, 35042 Rennes Cedex, France. E-mail: jean-francois.carpentier@univ-rennes1.fr
Received (in Cambridge, UK) 16th October 2002, Accepted 12th December 2002
First published as an Advance Article on the web 20th December 2002
Ketones and imines are chemoselectively reduced at room
temperature in methanol to the corresponding alcohols and
amines in high yields in a one-step procedure using
polymethylhydrosiloxane (PMHS) and a simple zinc–dia-
mine catalyst.
significantly faster than using the two-step procedure in toluene.
For instance, the reduction of acetophenone (1a) is completed
within 1 h in methanol while only ca. 8% conversion is
observed in toluene over the same reaction time. Also, the
conversion of a-chloroacetophenone (1b) in toluene reaches a
maximum of 43% after 24 h, but is quantitative within 1 h in
methanol. The reaction tolerates a variety of functional groups.
In particular, the [Zn–diamine]–PMHS–MeOH system is not
active under mild conditions (RT) in the reduction of esters and
is thus totally chemoselective for the reduction of CNO towards
COOR functions, enabling the valuable conversion of a- and b-
keto esters (1d–g) into a- and b-hydroxy esters (2d–g). When
carried out in toluene, the reduction using the Zn–diamine
catalyst is not selective in the case of methyl pyruvate (1e) and
does not proceed at all in the case of methyl benzoylacetate (1f);
those substrates are yet fast and chemoselectively reduced with
the present catalyst system. Similarly, a- and b-ketoamides (1h-
j) are reduced into a- and b-hydroxyamides (2h-j), which are
readily recovered in > 96% yield and high purity from the final
reaction mixture, due to their very low solubility in hydro-
carbons contrary to the silicon residues.†
Polymethylhydrosiloxane (PMHS), a safe and inexpensive
polymer co-product of the silicon industry, has proved to be an
efficient alternative reducing agent of CNO and CNN bonds
when associated to different types of catalysts.1 In this context,
Mimoun recently reported a new system based on zinc–hydride
catalysts that enables to reduce chemoselectively non-function-
alised (or a,b-unsaturated) aldehydes, ketones and even esters.2
However, those systems can not operate in protic solvents due
to dehydrogenative silylation of the latter by PMHS; therefore,
the recovered product is a silyl ether which must be subjected to
a separate and somewhat delicate hydrolysis step. Mimoun et al.
also developed a chiral version employing catalyst combina-
tions of ZnEt2 with optically active diamines to reduce aryl
alkyl ketones in valuable enantioselectivities via the same two-
step procedure ((i) hydrosilylation; (ii) hydrolysis).3 We report
here a simple diamine–zinc catalyst system that operates in an
alcoholic solvent, and that overcomes these limitations and
broadens the scope of this process.
The typical catalyst system is formed from a 1+1 combination
of diethylzinc and commercially available N,N’-dibenzylethy-
lenediamine (dbea). With this system, most conveniently
generated in situ (vide infra), a large range of ketones and
imines are chemoselectively reduced at room temperature in
methanol to the corresponding alcohols and amines in high
yields in a one-step procedure (Scheme 1, Table 1). For all the
ketones investigated, the reduction in methanol proceeds
Only 1.0 equivalent of PMHS can be used to complete the
reduction of some ketones, e.g. a,a,a-trifluoroacetophenone
(1c) and methyl phenylglyoxylate (1d). On the other hand,
excess PMHS (2–5 equiv.) proved necessary in most cases, in
particular for readily enolisable ketones such as 1f, 1g and 1i. In
Table 1 One-step reduction of ketones 1a–j and imines 3a,b with the [Zn–
dbea]–PMHS–MeOH system.a
Equivalents
of PMHS
Yield 2/4c
(mol %)
Entry
Substrate
Timeb/h
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1a
1a
1b
1c
1d
1e
1f
1
2
2
1
1
2
2
5
2
5
2
2
5
1
2
2
2
4
1
1
1
1
1
55
99 (86)
99
99
99 (67)
99
50
76
68
99
99 (98)
70
99
75
99 (96)
99
1
24
24
6
1
1
1
1
1
1
1f
1g
1g
1h
1i
1i
1j
1j
3a
3b
3b
18
24
24
65
99 (91)
a General conditions: 1 or 3–ZnEt2–dbea = 2.75+0.055+0.055 mmol in
methanol–toluene (80:20 v/v, 2.5 mL), T = 20 °C. b Reaction time was not
necessarily optimised. c Yield of 2 and 4 as determined by quantitative GLC
(BPX 5 column) and 1H NMR; no side product was formed; data into
parentheses refer to isolated yields of spectroscopically pure products.
Scheme 1 One-step reduction of ketones and imines with the [Zn–dbea]–
PMHS–MeOH system.
332
CHEM. COMMUN., 2003, 332–333
This journal is © The Royal Society of Chemistry 2003

the same vein, some imines, such as 3a, are readily reducible
using a slight excess of PMHS while others, e.g. 3b, require a
larger amount to go to completion. Two explanations account
for the consumption of extra equivalents of PMHS: (i) the [Zn–
dbea]–PMHS system is still moderately active for the dehy-
drogenative silylation of MeOH and the overall efficiency of the
process depends on the relative kinetics for CNO (CNN)
reduction vs. MeOH silylation; (ii) possibly, the catalyst system
is also operative for the dehydrogenative silylation of the enol
tautomer of ketones to give an enol-silyl ether (not observed by
NMR) that would hydrolyse back in methanol to the free enol/
ketone. Molecular hydrosilanes such as PhSiH3, Ph2SiH2,
Et2SiH2, can be equally used but are not as convenient as PMHS
due to their higher sensitivity and price.
mixtures of complexes that have not been formally identified so
far.
In conclusion, we have developed an alternative, simple and
cost-effective procedure that allows the chemoselective reduc-
tion of a variety of carbonyl and imine compounds. Current
efforts are directed towards mechanistic issues to rationalize the
effectiveness of the diamine-modified system in protic solvents
and the development of an efficient enantioselective version of
this system with chiral diamines for the reduction of function-
alised ketones and imines.
We thank the CNRS and PPG-SIPSY for financial support of
this research (Ph.D. grant to V. B.).
So far, methanol has been found the most suitable protic
solvent for this system. Generally, reduction of the ethyl
analogues of 1d–g performed in ethanol was found to be
somewhat less selective and more sluggish compared to
experiments conducted in MeOH. An exception to this trend
concerns aromatic a-ketoesters; i.e., reduction of ethyl phenyl-
glyoxylate in ethanol to ethyl mandelate proceeds quantitatively
within 1 h under typical conditions (Table 1).6 Though
experiments were routinely carried out using 2.0 mol% of
catalyst precursors, reduction of ketoester 1d using as low as 0.5
mol% of [Zn–dbea] is completed within 75 min. ZnEt2, which
alone has no catalytic activity, may also act as a scavenger;3
then, a substoichiometric amount of dbea, as low as 0.2 mol%,
suffices to activate the 1 mol% of ZnEt2 used (total conversion
of 1d within 2 h).
Notes and references
† In a typical experiment (entry 11), to a solution of dbea (13.2 mg, 0.055
mmol) in freshly distilled toluene (0.5 mL) under nitrogen, were
successively added ZnEt2 (50 mL of a 1.1 M solution in toluene, 0.055
mmol), a solution of 1h (657 mg, 2.75 mmol) in MeOH (2.0 mL), and
finally PMHS (0.32 mL, 5.0 mmol). The resulting solution was stirred with
a magnetic stir bar and the reaction was monitored by GLC. After
completion of the reaction, volatiles were removed under vacuum to give a
white oil which was triturated with pentane (2.0 mL). The resulting
precipitate was separated off from the liquid phase, washed with a minimal
amount of pentane, and dried under vacuum to give expected hydroxyamide
2h as a spectroscopically pure white powder (650 mg, 98%).
‡ Crystal data for zinc complex I: C36H48N4Zn2, M
= 667.52, or-
thorhombic, Pbca (no. 61), a = 8.6479(4), b = 19.4712(9), c = 19.5345(9)
Å, V = 3289.3(3) Å3, T = 100 K, Z = 4, m(Mo-Ka) = 1.489 mm21, Dc
= 1.348 g cm23, 34336 reflections measured, 6259 independent (Rint
=
Preliminary mechanistic investigations have shown that
ZnEt2 and dbea react rapidly ( < 15 min) at 20 °C in toluene
solution to give a single complex, [EtZnNBn(C2H4)NHBn]2 (I),
which has been formally characterized by elemental analysis,
NMR4 and a single crystal X-ray diffraction study‡ (Fig. 1).
Such ready formation of a dimeric alkyl(amino-m-amido) Zn
complex, with concomitant release of ethane, was unexpected
considering the relatively harsh conditions (60–70 °C, 3–5 h)
required for preparing analogous complexes, e.g.
[RZnNMe(CH2)nNMe2]2 (R = H, Me, Et; n = 2, 3),5 and that
N,NA-ethylenebis(1-phenylethylamine) (ebpe) reacts with ZnR2
(R = Me, Et) to form stable adducts ZnR2(ebpe).3 Dimer I is
only a catalyst precursor, which transforms rapidly in the
presence of MeOH, ketone substrate and hydrosilane into
0.0901), F2 refinement, R1 = 0.0535, wR2 = 0.0889, 3713 independent
observed reflections [I > 2s(I)], 195 parameters. CCDC reference number
195656. See http://www.rsc.org/suppdata/cc/b2/b210144k/ for crystallo-
graphic data in CIF or other electronic format.
1 For a review on PMHS, see: (a) N. J. Lawrence, M. D. Drew and S. M.
Bushell, J. Chem. Soc., Perkin Trans. 1, 1999, 3381–3391; for a recent
review on metal-catalyzed hydrosilylation of ketones and imines, see (b)
J.-F. Carpentier and V. Bette, Curr. Org. Chem., 2002, 6, 913–936; (c) M.
T. Reding and S. L. Buchwald, J. Org. Chem., 1995, 60, 7884–7890; (d)
Y. Kobayshi, E. Takahisa, M. Nakano and K. Watatani, Tetrahedron,
1997, 53, 1627–1634; (e) M. D. Drew, N. J. Lawrence, D. Fontaine and
L. Sehkri, Synlett, 1997, 989–991; (f) M. D. Drew, N. J. Lawrence, W.
Watson and S. A. Bowles, Tetrahedron Lett., 1997, 38, 5857–5860; (g)
X. Verdaguer, M. C. Hansen, S. C. Berk and S. L. Buchwald, J. Org.
Chem., 1997, 62, 8522–8528; (h) J. Yun and S. L. Buchwald, J. Am.
Chem. Soc., 1999, 121, 5640–5644; (i) N. J. Lawrence and S. M. Bushell,
Tetrahedron Lett., 2000, 41, 4507–4512; M. C. Hansen and S. L.
Buchwald, Org. Lett., 2000, 2, 713–715.
2 H. Mimoun, J. Org. Chem., 1999, 64, 2582–2589.
3 H. Mimoun, J. Y. de Saint Laumer, L. Giannini, R. Scopelliti and C.
Floriani, J. Am. Chem. Soc., 1999, 121, 6158–6166.
4 Variable temperature 1H NMR spectroscopy of I in toluene-d8 showed
dynamic phenomena with a single species observed on the NMR time
scale at 40 °C and two species (6+1 ratio) observed at 240 °C. The
relative intensity of these two sets of signals is not affected by a 50%
dilution, indicating that this is not a mixture of dimer and monomer but
rather of conformers5c
.
5 (a) N. A. Bell, P. T. Moseley, H. M. M. Shearer and C. B. Spencer, J.
Chem. Soc., Chem. Commun., 1980, 359–360; (b) N. A. Bell and A. L.
Kassyk, Inorg. Chim. Acta, 1996, 250, 345–349; (c) M. A. Malik, P.
O’Brien, M. Motevalli and A. C. Jones, Inorg. Chem., 1997, 36,
5076–5081.
6 Reduction of methyl ketoesters in ethanol and vice versa afforded
mixtures of methyl and ethyl hydroxyesters due to relatively slow
transesterification. No transesterification was observed for the reduction
of tert-butyl ketoesters conducted in methanol.
Fig. 1 Structure of catalyst precursor I. All hydrogen atoms have been
omitted for clarity.
CHEM. COMMUN., 2003, 332–333
333