Highly chemoselective catalytic hydrogenation of unsaturated ketones and aldehydes to unsaturated alcohols using phosphine-stabilized copper(I) hydride complexes

Article abstract of DOI:10.1016/S0040-4020(99)01098-4

A base metal hydrogenation catalyst composed of the phenyldimethylphosphine-stabilized copper(I) hydride complex provides for the highly chemoselective hydrogenation of unsaturated ketones and aldehydes to unsaturated alcohols, including the regioselective 1,2-reduction of α,β- unsaturated ketones and aldehydes to allylic alcohols. The active catalyst can be derived in situ by phosphine exchange using commercial [(Ph₃P)CuH]₆ or from the reaction of copper(l) chloride, sodium tert-butoxide, and dimethylphenylphosphine under hydrogen. The catalyst derived from 1,1,1- tris(diphenylphosphinomethyl)ethane is mechanistically interesting but less synthetically useful. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(99)01098-4

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 2153±2166
Highly Chemoselective Catalytic Hydrogenation of Unsaturated
Ketones and Aldehydes to Unsaturated Alcohols Using
Phosphine-Stabilized Copper(I) Hydride Complexes
*
Jian-Xin Chen, John F. Daeuble, Donna M. Brestensky and Jeffrey M. Stryker
Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada
Accepted 7 December 1999
AbstractÐA base metal hydrogenation catalyst composed of the phenyldimethylphosphine-stabilized copper(I) hydride complex provides
for the highly chemoselective hydrogenation of unsaturated ketones and aldehydes to unsaturated alcohols, including the regioselective
1,2-reduction of a,b-unsaturated ketones and aldehydes to allylic alcohols. The active catalyst can be derived in situ by phosphine exchange
using commercial [(Ph3P)CuH]6 or from the reaction of copper(I) chloride, sodium tert-butoxide, and dimethylphenylphosphine under
hydrogen. The catalyst derived from 1,1,1-tris(diphenylphosphinomethyl)ethane is mechanistically interesting but less synthetically useful.
q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
a,b-unsaturated ketones and aldehydes.
Despite the considerable potential for applications to large-
scale reduction of polyfunctional carbonyl-containing
compounds, hydrogenation catalysts rigorously chemo-
selective for carbonyl reduction over alkene hydrogenation
remain very rare.1 With the exception of the ruthenium/
diamine/base catalyst system recently disclosed by Noyori
et al.,2 the reported catalysts3±5 reveal only modest selec-
tivity or limited substrate range for conjugated or non-
conjugated substrates, with the highest and most general
selectivity obtained for the hydrogenation of a,b-unsatu-
rated aldehydes.4 Greater selectivity for carbonyl reduction
can be obtained by using catalytic hydrosilylation method-
ology,6 but this approach is accompanied by the additional
costs associated with using a silane reductant and the subse-
quent hydrolysis of the resultant silyl ethers.
ꢀ1†
We previously reported that the hydrogenation catalyst
derived from the thermally stable copper(I) hydride
complex [(Ph3P)CuH]6 (1) and excess triphenylphosphine
7
mediates the chemoselective reduction of a,b-unsaturated
ketones to saturated ketones or to saturated alcohols,
depending on reaction conditions, yet is completely unreac-
tive toward the hydrogenation of isolated alkenes, even
under high hydrogen pressure and prolonged reaction
time.8 The hexameric complex 1, however, is itself unreac-
tive toward isolated ketones, leading to the conclusion that
the hexamer must be transformed into more reactive
copper(I) hydride species during the hydrogenolysis of the
copper(I) enolate intermediate formed by initial conjugate
hydride addition.8a Mechanistic investigations8b,c revealed
that although the conjugate reduction is inhibited by the
presence of excess triphenylphosphine, the ketone reduction
is accelerated, suggesting that the catalyst responsible for
ketone reduction is probably a lower aggregate (possibly
monomeric) with a phosphine/copper stoichiometry of 2:1.
Such complexes are expected to be both kinetically more
reactive and more hydridic than the hexamer, leading to the
subsequent reduction of the isolated ketone.
In this report, we describe phosphine-stabilized copper(I)
hydride hydrogenation catalysts chemoselective for the
reduction of ketones in the presence of alkenes and a
broad range of other functionality typically reactive toward
conventional hydrogenation catalysts (Eq. (1)). The
copper(I) hydride catalyst stabilized by dimethylphenyl-
phosphine is particularly selective, providing complete
chemoselectivity for carbonyl reduction in nonconjugated
unsaturated ketones and aldehydes and moderate to high
levels of 1,2-reduction selectivity in the hydrogenation of
Keywords: copper and compounds; hydrogenation; catalysis; carbonyl
compounds.
* Corresponding author. Fax: 11-780-492-8231;
e-mail: jeff.stryker@ualberta.ca
We therefore initiated an investigation of sterically smaller
and more basic phosphine ligands, including multidentate
phosphines, to develop catalysts capable of ketone
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(99)01098-4

2154
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
hydrogenation under mild conditions. Since the inhibition of
conjugate reduction was a function of phosphine concen-
tration, it also appeared reasonable to propose that catalysts
of the composition [(R3P)2CuH]n might show selectivity for
direct 1,2-reduction of a,b-unsaturated ketones and
aldehydes.
equilibrium dissociation of the catalytically inactive dimer
2 into reactive monomeric fragments.14
The pressure dependence of the hydrogenation reaction is
most unusual. Slow catalyst decomposition is observed at
hydrogen pressures below 50 psi, but sustained turnover and
optimum reaction rate are obtained at pressures from 50±
70 psi. At higher pressure, however, turnover is strongly
inhibited. To con®rm this unexpected observation, the
pressure dependence was evaluated at two different concen-
trations; the results are presented in Fig. 1. While the obser-
vation of saturation kinetics in homogeneous hydrogenation
at higher pressures is not unusual, turnover inhibition by
hydrogen is, to our knowledge, unique. The origin of this
phenomenon was probed in a series of control experiments
in which the catalyst alone and catalytic reduction reaction
mixtures were `aged' under hydrogen pressure (both high
and low) for an extended period. No decomposition of
[(tripod)CuH]2 was observed under these conditions and
the rates of catalytic hydrogenation were fully restored
upon returning to typical operating pressures. The catalyst
is thus reversibly inhibited by hydrogen at high pressure.
Although no supporting evidence can be provided, we
speculate that the reversible formation of a catalytically
inactive Cu(I) dihydrogen complex, e.g. [(h2-tripod)-
CuH(H2)], may be responsible for the observed pressure
dependence. A simple, but related, copper(I) dihydrogen
complex [Cu(h2±H2)Cl] has been prepared and character-
ized by infrared spectroscopy at low temperature in an argon
matrix; the structure assignment is also supported by ab
initio calculations.15 The rate inhibition is thus analogous
to that observed in enzyme-catalyzed two-substrate compul-
sory order reactions in the presence of a high concentration
of the second substrate.16 Consistent with this proposal, a
normal concentration dependence for the rate of hydro-
genation is restored at very high hydrogen pressure (see
Fig. 1): the coordination of dihydrogen by the monomeric
copper fragment displaces the dimer/monomer equilibrium
toward a monomeric precatalyst complex. The reduction of
tert-butylcyclohexanone is markedly inhibited by the
presence of cyclohexene or stilbene, neither of which is
consumed in the reaction. Nonpolar double bonds thus
coordinate to, but are not reduced by, this catalyst.
Results and Discussion
Catalytic ketone hydrogenation using [h2-(1,1,1-tris(di-
phenylphosphinomethyl)ethane)CuH]2
Copper(I) hydride complexes incorporating nominally
bidentate phosphine ligands do not function as robust hydro-
genation catalysts. The known complex [(dppp)4Cu8H8]9
(dpppˆ1,3-bis(diphenylphosphino)propane), for example,
is a poor catalyst for ketone reduction, even in the presence
of excess dppp ligand.10 Catalysts derived in situ by
phosphine exchange using 1,2-bis(dimethylphosphino)ethane
or 1,2-bis(methylphenylphosphino)ethane11 (stoichiomet-
rically or in excess) and [(Ph3P)CuH]6 are also ineffective.
Surprisingly, however, the known binuclear bidentate
12
complex [(h2-tripod)CuH]2 (2, tripodˆ1,1,1-tris(diphenyl-
phosphinomethyl)ethane) indeed provides a synthetically
modest but mechanistically interesting catalyst for chemo-
selective ketone reduction.
An initial investigation of ketone hydrogenation using [(h2-
tripod)CuH]2 as the catalyst revealed several highly unusual
features. In contrast to [(Ph3P)CuH]6, the tripod catalyst
decomposes slowly in aromatic solvents, requiring the use
of tetrahydrofuran as solvent. The catalyst essentially
ignores the anticipated bene®cial effects of chelation: to
sustain turnover and maintain catalyst homogeneity, the
presence of excess tripodal phosphine ($2 equiv./Cu) is
required. This suggests that in the absence of free tripod,
catalyst decomposition is triggered by total dissociation of
the chelating tris(phosphine) ligand from the metal. This
conclusion is supported by an independent assessment:
upon addition of triphenylphosphine (1 equiv./Cu) to a
solution of [(h2-tripod)CuH]2 in benzene-d6, a dynamic
equilibrium with [(Ph3P)CuH]6 and free tripod is established
in which a surprising 17% of the copper content is present as
the hexamer, as established by NMR spectroscopic analysis.
Reaction conditions were optimized for the reduction of
tert-butylcyclohexanone, a substrate that also allows for a
sterically unbiased evaluation of the stereoselectivity of the
reaction. The rate of catalytic hydrogenation is concen-
tration dependent, with faster turnover observed upon
decreasing the catalyst concentration. Thus, turnover
increases from 1.2 h21 at 0.2 M (in substrate) to 3.5 h21 at
0.05 M (in substrate) for reactions run at 50 psi (gauge
pressure) of hydrogen pressure and 2.5 mol% [(h2-tripod)-
CuH]2, but then decreases upon further dilution.13 This
behavior is consistent with a mechanism involving
Figure 1. Pressure dependence in the reduction of tert-butylcyclohexanone.
Conditions: 2.5 mol% 2, 10 mol% tripod, THF, RT, H2 as noted. Concen-
tration, in substrate: Oˆ0.2 M; Aˆ0.1 M.

J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
Table 1. Catalytic reduction of ketones and aldehydes using [(h2-tripod)CuH]2 and H2
2155
aReduction conditions: A, 2.5 mol% [(tripod)CuH]2, 2 equiv. tripod/Cu, THF, 0.1 M in substrate, 50±60 psi H2, RT; B, as A, except 0.8 mol% [(Ph3P)CuH]6,
3 equiv. tripod/Cu, 0.2 M; C, as A, except 0.2 M; D, as A, except 0.05 M; E, as C, except 5 mol% catalyst; F, as B, except 0.1 M; G, as B, except 500 psi H2.
bAll yields refer to isolated puri®ed products, assignments by comparison to known compounds or authentic materials. All reactions were set up under N2;
quenched by exposure to air and addition of sat. aq NH4Cl. Products were puri®ed by ¯ash chromatography.
cMinor isomer is epimeric at the hydroxy center.
dStereoisomer unassigned.
eConversions and product ratios determined by 1H NMR spectroscopy.
Under optimized conditions, the synthetic scope of catalytic
hydrogenation using [(h2-tripod)CuH]2 is somewhat
limited, principally by the slow rates observed for substrates
with even modest steric hindrance. Nonetheless, reductions
that do proceed are clean and nearly quantitative (Table 1).
The reduction of cyclic ketones proceeds with reasonable
stereoselectivity (entries 1, 7).17 The catalyst can be
prepared conveniently by in situ phosphine exchange
starting with commercially available hexamer 1, although
catalyst turnover is attenuated and the stereoselectivity
diminished (entries 2, 6).18 While turnover is again inhibi-
ted, no reduction of isolated ole®ns is observed (entry 4), but
the catalyst maintains a preference for conjugate reduction
of a,b-unsaturated ketones. The catalyst neither opens
epoxides (entry 7) nor reduces carboxylic esters. Nitrile
functionality, such as is present in 5-cyano-2-pentanone,
inhibits all catalytic activity. The catalyst is very sensitive
to steric effects neighboring the carbonyl group, requiring
higher catalyst loading to achieve complete conversion in a
reasonable reaction time (entry 5). Cyclopentanone
substrates (e.g. indanone, 2-allyl-2-carboethoxycyclopenta-
none) are reduced very slowly by this catalyst. Catalytic
reduction of aldehydes (entries 8, 9) requires elevated
pressure to suppress the competitive Tishchenko
condensation.19
genation reactivity of copper(I) hydride complexes was
initiated. Our preliminary evaluation of simple, strongly
basic monophosphine ligands revealed that very subtle
alterations of the phosphine structure lead to incommensu-
rate changes in the structure and function of the resulting
copper complexes.20 Thus, although discrete, isolable
hexameric copper(I) hydride complexes are formed from
the hydrogenolysis of copper(I) tert-butoxide in the
presence of either ethyldiphenylphosphine or diethylphenyl-
phosphine,21 neither complex is an effective hydrogenation
catalyst. The corresponding methyldiphenylphosphine and
dimethylphenylphosphine derivatives, however, manifest
dramatically different structure and reactivity properties,
despite the closely analogous steric and electronic character
of the ligands themselves. While methyldiphenylphosphine
has a slightly smaller cone angle22 and marginally greater
nucleophilicity23 than does ethyldiphenylphosphine, no
change in the stretching frequency of coordinated carbon
monoxide is observed when these phosphines are inter-
changed at various metal centers.24 In other catalytic hydro-
genation reactions, the function of these two ligands is
virtually indistinguishable.5b
Nonetheless, no well-de®ned copper(I) hydride complex
can be isolated from the hydrogenolysis of [Cu(OtBu)]4 in
the presence of excess methyldiphenylphosphine. Instead, a
thermochromic oil is obtained, which varies in color
from deep red at ambient temperature to yellow at
2408C.20 The product is thus variable in composition,
suggesting an equilibration between a red hexameric
aggregate of 1:1 phosphine/copper stoichiometry and a
yellow dimeric complex with a 2:1 phosphine/copper ratio
Catalytic ketone hydrogenation using dimethylphenyl-
phosphine-derived copper(I) hydride catalysts
Given the apparent failure of chelation to limit phosphine
lability in copper(I) hydride complexes, an empirical
investigation of ligand effects on the structure and hydro-

2156
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
over this temperature range, consistent with analysis of the
product by variable temperature 1H and 31P NMR spectros-
copy. Under otherwise identical conditions, the hydro-
genolysis of [CuOtBu]4 in the presence of excess
dimethylphenylphosphine leads initially to the formation
of a yellow solution, which darkens rapidly and deposits
an intractable precipitate. Similarly, addition of excess
dimethylphenylphosphine to a solution of [(Ph3P)CuH]6
initially gives an orange±yellow solution, which
deteriorates visibly within
temperature.
a few hours at room
DespiteÐor perhaps because ofÐthis instability, the
dimethylphenylphosphine-derived copper(I) hydride
complex is an excellent catalyst for the chemoselective
hydrogenation of ketones under very mild conditions
(Table 2). The catalyst can be conveniently generated in
situ by one of two procedures: (i) ligand exchange using
Table 2. Catalytic reduction of ketones using Me2PPh-stabilized copper(I) hydride and H2
aConditions. A, 0.33 mol% [(Ph3P)CuH]6 (2 mol% Cu), Me2PPh (6 equiv./Cu), 1 atm H2, C6H6 (0.4±0.8 M in substrate), tert-butanol (10±20 equiv./Cu), RT;
B, as A, except 5 mol% CuCl and NaOtBu, Me2PPh (6 equiv./Cu); C, as A, except 0.83 mol% [(Ph3P)CuH]6 (5 mol% Cu); D, as A, except 1.67 mol%
[(Ph3P)CuH]6 (10 mol% Cu).
bMany reaction times not optimized.
cMajor stereoisomer; minor epimeric at alcohol center. Stereochemical ratio in parentheses; determined by isolation and separation or by 1H NMR integration
of crude product.
dIsolated yields after puri®cation by chromatography. Products identi®ed by comparison with authentic materials prepared by unambiguous synthesis.
e16% starting material recovered.
fSecond isomer epimeric at methyl and alcohol centers; third isomer epimeric at alcohol position.

J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
2157
commercially available [(Ph3P)CuH]6 (1) and excess
dimethylphenylphosphine, or (ii) direct synthesis from
cuprous chloride, sodium tert-butoxide, and dimethyl-
phenylphosphine (1:1:6) in anhydrous benzene under a
hydrogen atmosphere at room temperature. The exchange
procedure is operationally simpler, but for some substrates
(vide infra) leads to a deterioration in stereoselectivity,
arising from the presence of triphenylphosphine in the
reaction medium. The latter procedure is most attractive,
using only inexpensive, commercially available reagents,
but proceeds via in situ formation of the oxygen-sensitive
copper(I) tert-butoxide complex and demands greater
attention to reagent purity and experimental technique.
reduced only very slowly by this catalyst; such reactions
generally do not proceed to completion at 10 mol%
catalyst.25
Catalytic 1,2-reduction of a,b-unsaturated aldehydes
and ketones
In contrast to the reactivity of catalysts derived from
[(Ph3P)CuH]6 and [(h2-tripod)CuH]2, the dimethylphenyl-
8
phosphine-stabilized copper(I) catalyst retains its selectivity
for carbonyl reduction in the hydrogenation of a,b-unsatu-
rated ketones and aldehydes. The hydrogenation reaction
tolerates the use of either benzene or tetrahydrofuran as
solvent, but requires a somewhat higher concentration of
tert-butanol as a co-solvent to ensure sustained turnover
and reaction homogeneity. In contrast to the reduction of
isolated ketones, hydrogen pressures above one atmosphere
are generally required to obtain complete conversion of
conjugated substrates, perhaps due to greater competitive
inhibition arising from coordination of the electron-de®cient
ole®n of the substrate or product to an unsaturated copper(I)
intermediate.26 In the reduction of conjugated aldehydes, the
higher hydrogen pressure also inhibits the formation of
pseudo-dimeric esters from competitive Tishchenko
reaction.
The dimethylphenylphosphine-derived catalyst functions
slowly, normally requiring between 2 and 10 mol% copper
to obtain reasonable reaction time. Excess phosphine
(6 equiv. per copper) is required for sustained turnover;
the use of lower phosphine concentration leads to slow
decomposition of the catalyst during the reduction of
some substrates. The catalyst does not tolerate high
temperatures, but appears to be reasonably robust at room
temperature: the reduction of tert-butylcyclohexanone
proceeds to completion at atmospheric hydrogen pressure
using as little as 0.5 mol% copper. The infusion of
additional substrate to a completed hydrogenation reaction
results in resumption of conversion with a negligible impact
on turnover, provided the catalyst mixture is not allowed to
age for a prolonged period in the absence of substrate. The
catalyst exhibits normal pressure dependence, but for some
substrates, the use of higher pressure leads to slow deterio-
ration of the catalyst, a phenomenon that is, as yet, poorly
understood. Finally, the presence of tert-butanol
(10±20 equiv./copper) in the reaction medium prolongs
the lifetime of the catalyst, presumably by inducing alco-
holysis of less stable copper(I) alkoxide intermediates to
give the thermally more robust copper(I) tert-butoxide
complex,8 which then undergoes hydrogenolysis without
competitive decomposition.
Under optimized conditions, the catalytic hydrogenation of
a,b-unsaturated aldehydes is general for the formation of
allylic alcohols in high yield (Table 3, entries 1±4). The
reaction is highly regioselective regardless of substitution
pattern, but returns a minor amount of the saturated alcohol
from complete reduction of both the conjugated double
bond and the carbonyl group. Control experiments establish
that the minor product arises from the competitive conjugate
reduction followed by subsequent reduction of the carbonyl;
allylic alcohols are completely unreactive toward this
catalyst.27
The selectivity obtained in the hydrogenation of a,b-unsatu-
rated ketones is more variable (Table 3, entries 5±9).
Acyclic substrates are reduced with high selectivity for
the formation of the corresponding allylic alcohol, but the
reduction of cyclohexenones (e.g. entry 8) is more proble-
matic, delivering a modest 2.7:1 selectivity using standard
exchange conditions. The selectivity is improved by
removing the triphenylphosphine from the reaction medium
(entry 9): by generating the catalyst from copper(I) chloride,
sodium tert-buoxide, and dimethylphenylphoshine, the
product ratio is raised to approximately 4.4:1 in favor of
the allylic alcohol. Although we do not understand the
origin of the lower regioselectivity observed for the
reduction of cyclohexenones, it is interesting to note that
these derivatives also return the poorest selectivity in hydro-
genations catalyzed by Noyori's ruthenium system.2 The
reduction of cyclohexenone substrates using catalysts
derived from methyldiphenylphosphine, triethylphosphine,
and tri-n-butylphosphine all proceed with high conversion,
but deliver markedly lower selectivity for allylic alcohol
formation. Structurally related ligands such as trimethyl-
phosphine and tricyclohexylphosphine do not provide
productive hydrogenation catalysts.
Synthetically, the system functions for a broad range of
organic substrates and shows unparalleled chemoselec-
tivity, reducing ketones in the presence of isolated alkenes
(entries 4±6, 9), alkynes (entry 7), dienes (entry 12), vinylic
halides (entry 12) and allylic and benzylic oxygenates (entry
12), all functionality typically reduced by standard hydro-
genation catalysts. The hydridic catalyst does not open
epoxides (entry 8) nor reduce esters (9, 11). Primary
p-toluenesulfonate functionality survives this reduction
reaction, but the catalyst slowly deactivates during the
course of the reduction, which does not proceed to com-
pletion even at a catalyst loading of 10 mol% (entry 10).
The hydrogenation rate is inhibited by the presence of
sterically accessible alkene functionality (e.g. entries 9,
12), suggesting that the catalyst binds competitively to
alkenes. The catalyst generally provides the same sense of
stereoinduction as that delivered by standard hydride
reagents, but the stereoselectivity of the catalytic reduction
is generally superior. Stereogenic centers adjacent to the
ketone functionality are unfortunately epimerized during
the reduction, presumably as a consequence of the strongly
basic copper(I) alkoxide intermediates formed during the
reaction (entries 13, 14). Cyclopentanone substrates are
The reactivity of copper(I) hydride hydrogenation catalysts

2158
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
is thus exceptionally sensitive to the structure of the phos-
phine ligand. In the hydrogenation of 4-phenyl-3-buten-2-
one, for example, it is possible to obtain each of the three
reduction products with high selectivity, simply by varying
the ancillary ligand added to the reaction mixture (Eq. (2)).
exhibits an unprecedented level of chemoselectivity in the
hydrogenation of highly functionalized ketone substrates,
with a reactivity pro®le more consistent with classical
hydride reagents than with typical hydrogenation
catalysts. The dramatic differences observed in the
Table 3. Catalytic reduction of a,b-unsaturated ketones and aldehydes using Me2PPh-stabilized copper(I) hydride and H2
aReaction conditions. A, 0.83 mol% [(Ph3P)CuH]6 (5 mol% Cu), PhPMe2 (6 equiv./Cu), 70 psi H2, C6H6 (0.4±0.8 M in substrate), tert-butanol (40 equiv/Cu),
RT; B, as A, 400 psi H2; C, as A, except 500 psi H2; D, as A, except copper introduced as CuCl (5 mol%) with NaOtBu (5 mol%), 1000 psi H2.
bMajor product indicated; minor product is saturated alcohol (1,411,2 reduction). Product ratio is given in parentheses.
cIsolated yields after puri®cation by chromatography. Products identi®ed by comparison with authentic materials prepared by unambiguous synthesis (see
Experimental).
dRemaining material identi®ed (NMR spectroscopy) as the Tishchenko reaction product PhCHyCHCH2OC(O)CHyCHPh.
eE/Zˆ10:1 (substrate), 10:1 (product).
fE/Zˆ2:1 (substrate and product).
gIsolated yield after acetylation (Ac2O/pyr) and puri®cation.
hMajor allylic alcohol stereoisomer indicated (12:1); minor epimeric at hydroxyl center.
ꢀ2†
Compared to Noyori's ruthenium catalyst,2 the copper(I)-
based catalyst functions more slowly and, for the
hydrogenation of conjugated ketones and aldehydes, is
less regioselective for 1,2-reduction. Nonetheless, the
selectivity is impressive and examples of inexpensive base
metal hydrogenation catalysts remain very rare. The
dimethylphenylphosphine-stabilized copper(I) system
structure, dynamics, and function of copper(I) hydride
complexes derived from closely related phosphine
derivatives are dif®cult to rationalize, although an
understanding of this subtle interplay of steric and elec-
tronic factors is critical to the development of an asym-
metric version of this highly chemoselective catalytic
reduction.

J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
2159
Experimental
General experimental
(oxalyl chloride, benzene, RT) and treatment with LiMe2Cu
(Et2O, 2788C). 5(S),6-(Cyclohexylidenyl-1,1-dioxy)-4(S)-
methoxy-2(R)-methylhexanoic acid41 was furnished by
Prof. D. R. Williams (Indiana University) and converted
to 6(S),7-(cyclohexylidenyl-1,1-dioxy)-5(S)-methoxy-3(R)-
methylheptan-2-one by treatment with oxalyl chloride
followed by LiMe2Cu, as described above.
Unless otherwise noted, all manipulations were performed
under an inert atmosphere using standard Schlenk tech-
niques or in a glove box (#1 ppm O2). Reactions run
under hydrogen pressure were performed in a Fischer &
Porter (Andrews Glass) medium pressure bottle28 (20±
75 psi) or in a magnetically stirred, glass-lined, Parr stain-
less steel autoclave (75±1500 psi), each equipped with
Swagelok Quick-connects and pressure gauges. Vacuum
transfers of known amounts of air sensitive volatile
compounds were performed at high vacuum (1025 mmHg)
using MKS Baratron digital pressure transducers for
measurement of vapor pressures in known volume ¯asks.
All solvents, reagents, and commercial substrates were used
as received or puri®ed by standard methods. Toluene,
benzene, benzene-d6, diethyl ether, hexanes, pentane, and
tetrahydrofuran were dried and deoxygenated by distillation
from sodium or potassium benzophenone ketyl. Aceto-
nitrile, dichloromethane, and pyridine were distilled from
calcium hydride and deaerated by repeated freeze±pump±
thaw cycles or by purging with nitrogen. Chromatography
solvents were used as received. Product puri®cation was
performed by preparative-layer chromatography on pre-
Product identi®cation
Reaction products were identi®ed spectroscopically and
compared to authentic materials prepared by unambiguous
synthesis. Isolated alcohols were prepared by standard
ketone reduction using NaBH4 in methanol or LiAlH4 in
diethyl ether. Allylic alcohols were prepared from
a,b-unsaturated aldehydes and ketones by treatment with
NaBH4/CeCl3 in methanol.42 Fully reduced alcohols were
prepared by catalytic hydrogenation of the allylic alcohols
using Pd/C and H2 (1 atm)43 or, in the case of a,b-unsatu-
rated aldehydes, by reduction using stoichiometric
[(Ph3P)CuH]6 (0.4±0.5 equiv.).44 High resolution spectro-
scopic data are included for selected literature compounds
for which the data are unreported or of poor quality.
General procedure for catalytic reductions using [(h2-
tripod)CuH]2
coated glass-backed silica gel plates (E. Merck 60 F254
,
0.5±1.0 mm) or by ¯ash chromatography31 using E.
Merck Kieselgel 60, 230±400 mesh.
In the glove box (or on the bench with rigorous oxygen
exclusion), the Fischer & Porter bottle was loaded with a
solution with [(h2-tripod)CuH]2 (5±10 mol% in copper) and
tripod (2±3 equiv. per Cu) in THF (1±3 ml). To this solu-
tion was added the ketone in anhydrous THF (1 ml, ®nal
concentration: 0.2±0.05 M in substrate) and a stirbar. The
vessel was sealed and connected via hydrogen-¯ushed
pressure tubing to a hydrogen tank, ¯ushed with hydrogen
(pressurized to .50 psig and vented, ®ve times), and
pressurized (50±70 psig). The yellow solution was stirred
vigorously for the speci®ed length of time (Table 1) or until
TLC analysis indicated complete reduction. The vessel was
vented to air, quenched with a few drops of saturated
aqueous NH4Cl, diluted with ether, and dried over
MgSO4. After ®ltration through Celite and concentration,
the residue was puri®ed by chromatography.
Materials
tert-Butanol was distilled from sodium, deaerated and
stored under nitrogen. [(Ph3P)CuH]6 and [(h2-tripod)-
CuH]2 were prepared and puri®ed according to literature
29
12
procedures. All tertiary phosphines except commercial
triphenylphosphine and triethylphosphine were prepared
by literature methods11,30 and deoxygenated before use.
2-Undecanone, 6-undecanone, 4-tert-butylcyclohexanone,
4-phenyl-2-butanone,
2-methyl-1-phenyl-1-propanone,
dihydrocarvone (2 isomers), benzaldehyde, cinnamalde-
hyde, 2,4-dimethyl-2,6-heptadienal, citral, perillaldehyde,
b-ionone, 1-acetyl-1-cyclohexene, and 3,5-dimethylcyclo-
hexenone were purchased from commercial suppliers.
4-Phenyl-5-hexen-2-one,32 isophorone oxide,33 dihydro-b-
ionone,34 6,10-dimethyl-9-undecen-2-one,35 2-allyl-2-
Reduction of 4-tert-butylcyclohexanone using a catalyst
derived from [(Ph3P)CuH]6 and tripod
carbethoxycyclohexan-1-one,36and
1(b)-acetoxy-8a(b)-
methyl-1,2,3,4,4a(a),5,8,8a-octahydronaphthalene-6(7H)-
one37 were prepared according to literature procedures.
Cyclooct-4-en-1-one was prepared by the oxidation of
cyclooct-4-en-1-ol38 (aqueous chromic acid, acetone, 08C).
Trideca-2-one-6-yne was prepared by alkylation of 1-lithio-
1-octyne (1-octyne, n-BuLi, NH3, 2788C) with the ethylene-
dioxy acetal of 5-chloro-2-pentanone (NH3, 2338C),
followed by hydrolysis (0.5N HCl, THF). 3-[4-(p-Toluene-
sulfonyloxy)butyl]-2-cyclohexanone39a was prepared by
tosylation of 3-(4-hydroxybutyl)-2-cyclohexanone39b under
standard conditions. 4-Benzyloxy-6-bromo-(5E,7E)-1,5,7-
undecatrien-11-ol40 was furnished by Prof. W. R. Roush
(University of Michigan) and converted to 4-benzyloxy-6-
bromo-(5E, 7E)-1,5,7-dodecatrien-11-one by oxidation
(aqueous chromic acid, acetone, 08C), followed by conver-
sion of the resulting carboxylic acid to the acid chloride
In the glove box, a solution of 4-tert-butylcyclohexanone
(62.6 mg, 0.406 mmol) in anhydrous THF (1 ml) was added
to a Fischer & Porter bottle charged with a solution of
[(Ph3P)CuH]6 (6.6 mg, 0.0034 mmol) and tripod (38.2 mg,
0.0612 mmol) in THF (1 ml). The vessel was sealed,
removed from the glovebox, ¯ushed with hydrogen and
pressurized to 50 psig. The yellow solution was stirred
vigorously for 24 h. The vessel was vented to air, quenched
with a few drops of saturated aqueous NH4Cl and diluted
with ether. After drying with MgSO4 and ®ltration through
Celite, concentration and chromatography (15% EtOAc/
hexanes) gave trans and cis 4-tert-butylcyclohexanol
1
(62.4 mg, 98%, 4:1 trans/cis). Partial H NMR spectrum
(500 MHz, CDCl3): trans-4-tert-butylcyclohexanol d 3.51
(1H, tt, Jˆ4.3, 10.9 Hz, CHOH), 0.84 (9H, s); cis-4-tert-
butylcyclohexanol d 4.03 (1H, quintet, Jˆ2.7 Hz), 0.85

2160
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
(9H, s). The products were identi®ed by comparison to a
commercial sample.
C9H16O2ˆ156.1146 (M1), foundˆ156.1150. The data are
consistent with the literature values.46
Reduction of 4-tert-butylcyclohexanone using [(h2-
tripod)CuH]2
Reduction of benzaldehyde using a catalyst derived from
[(Ph3P)CuH]6 and tripod
Using the general procedure, 4-tert-butylcyclohexanone
(31.2 mg, 0.203 mmol), [(h2-tripod)CuH]2´2THF (7.7 mg,
0.005 mmol), and tripod (12.6 mg, 0.020 mmol) were
hydrogenated for 7 h, giving 4-tert-butylcyclohexanol
(29.0 mg, 92%, 8:1 trans/cis). The products were identi®ed
as described above.
Using the procedure described above for 4-tert-butylcyclo-
hexanone,
benzaldehyde
(21.5 mg,
0.203 mmol),
[(Ph3P)CuH]6 (6.6 mg, 0.0034 mmol), and tripod
(50.4 mg, 0.081 mmol) were hydrogenated at 500 psig
hydrogen in the steel autoclave for 25 h, giving a mixture
of benzyl alcohol and benzyl benzoate (19:1) in quantitative
yield. The products were identi®ed by comparison to
authentic samples.
Reduction of 2-undecanone
Using the general procedure, 2-undecanone (68.8 mg,
0.404 mmol), [(h2-tripod)CuH]2´2THF (15.4 mg, 0.010
mmol), and tripod (38.0 mg, 0.061 mmol) were hydroge-
nated for 36 h, giving 2-undecanol (62.5 mg, 90%). The
product was identi®ed by comparison to a commercial
sample.
General procedure for reduction of saturated ketones
using [(Ph3P)CuH]6 and Me2PPh
In the glove box, [(Ph3P)CuH]6 (1±10 mol% Cu), Me2PPh
(6 equiv./Cu) and tert-butanol (10±20 equiv./Cu) were
combined in a Schlenk ¯ask and dissolved in benzene. To
this solution was added a solution of the substrate (10±
100 equiv./Cu) in benzene (0.4±0.8 M in substrate). The
¯ask was sealed, transported out of the drybox and placed
under a slight positive pressure of hydrogen after one
freeze±pump±thaw degassing cycle. The resulting
yellow±orange homogeneous solution was allowed to stir
until complete by TLC analysis. The reaction mixture was
exposed to air, diluted with ether, and treated with a small
amount of silica gel. This mixture was stirred in the air for
$0.5 h prior to ®ltration, concentration in vacuo, and puri-
®cation by ¯ash chromatography. When the polarity of the
product was similar to that of the residual phosphine, the
crude mixture was treated with sodium hypochlorite (5%
aqueous solution) and ®ltered through silica gel/MgSO4
prior to chromatography.
Reduction of 4-phenyl-5-hexen-2-one
Using the general procedure, 4-phenyl-5-hexen-2-one
(35.5 mg, 0.204 mmol), [(h2-tripod)CuH]2´2THF (15.4 mg,
0.010 mmol), and tripod (38.0 mg, 0.061 mmol) were
hydrogenated for 60 h, giving 4-phenyl-5-hexen-2-ol45
(33.0 mg, 92%) as an unassigned 3:2 mixture of diastereomers.
The products were identi®ed by comparison to an authentic
sample prepared by unambiguous synthesis.
Reduction of 2-methyl-1-phenyl-1-propanone
Using the general procedure, 2-methyl-1-phenyl-1-propan-
one (30.0 mg, 0.203 mmol), [(h2-tripod)CuH]2´2THF
(15.4 mg, 0.010 mmol), and tripod (38.0 mg, 0.061 mmol)
were hydrogenated for 24 h, giving 2-methyl-1-phenyl-1-
propanol (28.0 mg, 91%). The product was identi®ed by
comparison to an authentic sample prepared by unam-
biguous synthesis.
General procedure for reduction of saturated ketones
using CuCl, NaOtBu, and Me2PPh
Under an inert atmosphere, a solution of the substrate in
benzene was added to a slurry of freshly puri®ed CuCl
(5 mol%), NaOtBu (5 mol%), Me2PPh (6 equiv./Cu) and
tert-butanol (10 equiv./Cu) in benzene (®nal concentration:
0.4±0.8 M in substrate). After degassing with one freeze±
pump±thaw cycle, the suspension was placed under a slight
positive pressure of hydrogen and allowed to stir until
complete by TLC analysis. The product was isolated and
puri®ed as described above.
Reduction of isophorone oxide using [(h2-tripod)CuH]2
Using the general procedure, isophorone oxide (31.4 mg,
0.204 mmol), [(h2-tripod)CuH]2´2THF (7.0 mg, 0.005
mmol), and tripod (12.6 mg, 0.020 mmol) were hydro-
genated for 30 h, giving a mixture of isomeric 2,3-epoxy-
3,5,5-trimethylcyclohexan-1-ols,46 which were separated by
chromatography (4:1 hexanes/ethylacetate) to give pure
trans-epoxy alcohol (27.7 mg, 87%) and cis-epoxy alcohol
(2.5 mg, 8%). Major isomer: TLC Rfˆ0.31, 2:1 hexanes/
ethyl acetate; IR (neat, cm21) 3420, 2980±2880, 1450,
1410, 1375, 1360, 1270, 1250, 1180, 1080, 1040, 990,
Reduction of 4-tert-butylcyclohexanone using
[(Ph3P)CuH]6 and Me2PPh
1
940, 810; H NMR (400 MHz, CDCl3) d 4.15 (1H, m),
2.95 (1H, s), 1.9 (1H, d, Jˆ4.8 Hz), 1.70 (1H, d,
Jˆ15.2 Hz), 1.63 (1H, dd, Jˆ5.6, 13.6 Hz), 1.51 (1H, d,
Jˆ15.2 Hz), 1.33 (3H, s), 1.16 (1H, dd, Jˆ7.6, 13.6 Hz),
0.96 (3H, s), 0.90 (3H, s); Minor isomer: TLC Rfˆ0.22, 2:1
Using the general procedure, 4-tert-butylcyclohexanone
(0.10 g, 0.65 mmol), [(Ph3P)CuH]6 (0.0041 g, 0.0021
mmol, 1.85 mol% Cu), tert-butanol (0.019 g, 0.26 mmol),
and dimethylphenylphosphine (0.011 g, 0.078 mmol)
were combined and hydrogenated for 50 h, giving 4-tert-
butylcyclohexanone (0.098 g, 97%, 3:1 trans/cis). The
products were identi®ed by comparison to commercial
samples.
1
hexanes/ethyl acetate; H NMR (400 MHz, CDCl3) d 4.07
(1H, m), 3.14 (1H, d, Jˆ2.0 Hz), 1.64 (1H, d, Jˆ14.8 Hz),
1.52±1.40 (2H, m), 1.35 (3H, s), 1.66 (1H, dd, Jˆ11.2,
12.4 Hz), 0.89 (3H, s), 0.86 (3H, s); HRMS calcd for

J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
2161
Reduction of 6-undecanone
0.26 mmol), [(Ph3P)CuH]6 (4.2 mg, 0.0021 mmol, 5 mol%
Cu), tert-butanol (0.0095 g, 0.130 mmol), and dimethyl-
phenylphosphine (0.011 g, 0.077 mmol) were combined
and hydrogenated for 24 h, giving trideca-2-ol-6-yne
(0.050 g, 99%). TLC Rf ˆ0.25, 5:1 hexanes/ethyl acetate;
IR (neat, cm21) 3350, 2960±2840, 1430, 1370, 1325, 1120,
Using the general procedure, 6-undecanone (0.050 g,
0.29 mmol), [(Ph3P)CuH]6 (0.0096 g, 0.0049 mmol,
10 mol% Cu), tert-butanol (0.022 g, 0.29 mmol), and
dimethylphenylphosphine (0.024 g, 0.176 mmol) were
combined and hydrogenated for 12 h, giving 6-undecanol
(0.048 g, 95%). The products were identi®ed by comparison
to an authentic sample.
1
1080, 985, 935; H NMR (400 MHz, CDCl3) d 3.84 (1H,
sextet, Jˆ6.4 Hz), 2.17 (2H, tt, Jˆ2.4, 6.8 Hz), 2.13 (2H, tt,
Jˆ2.4, 6.8 Hz), 1.65±1.41 (7H, m), 1.40±1.22 (6H, m), 1.19
(3H, d, Jˆ6.4 Hz), 0.88 (3H, t, Jˆ6.8 Hz); 13C{1H} NMR
(100 MHz, CDCl3) d 80.7, 79.8, 67.7, 38.4, 31.4, 29.1, 28.5,
25.3, 23.5, 22.6, 18.7, 14.0. The product was identical to the
material prepared by unambiguous synthesis.
Reduction of 4-phenyl-2-butanone
Using the general procedure, 4-phenyl-2-butanone (0.025 g,
0.169 mmol), [(Ph3P)CuH]6 (2.8 mg, 0.0014 mmol,
,5 mol% Cu), tert-butanol (0.0125 g, 0.169 mmol), and
dimethylphenylphosphine (7.0 mg, 0.050 mmol) were
combined and hydrogenated for 30 h, giving 4-phenyl-
butan-2-ol (0.024 g, 95%). The products were identi®ed
by comparison to an authentic sample.
Reduction of isophorone oxide using [(Ph3P)CuH]6 and
Me2PPh
Using the general procedure, isophorone oxide (0.050 g,
0.324 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027 mmol,
5 mol% Cu), tert-butanol (0.012 g, 0.162 mmol), and
dimethylphenylphosphine (0.0134 g, 0.097 mmol) were
combined and hydrogenated for 48 h (unoptimized), giving
a mixture of isomeric of 2,3-epoxy-3,5,5-trimethylcyclo-
hexan-1-ols46 (25:1, 0.042 g, 83%), which were separated
by ¯ash chromatography. Major, trans epoxy-alcohol:
0.0403 g, 80%; minor, cis epoxy-alcohol: 0.0017 g, 3%.
The products were spectroscopically identical to those
described above.
Reduction of dihydro-b-ionone
Using the general procedure, dihydro-b-ionone (0.050 g,
0.257 mmol), [(Ph3P)CuH]6 (4.2 mg, 0.0021 mmol,
5 mol% Cu), tert-butanol (0.0095 g, 0.13 mmol), and
dimethylphenylphosphine (0.011 g, 0.077 mmol) were
combined and hydrogenated for 30 h, giving 4-phenyl-
butan-2-ol (0.024 g, 95%). The products were identi®ed
by comparison to an authentic sample.47
Reduction of 2-allyl-2-carbethoxycyclohexan-1-one
Reduction of 6,10-dimethyl-9-undecen-2-one
Using the general procedure, 2-allyl-2-carbethoxycyclo-
hexan-1-one (0.080 g, 0.38 mmol), [(Ph3P)CuH]6 (12.7
mg, 0.0065 mmol, 10 mol% Cu), tert-butanol (0.029 g,
0.39 mmol), and dimethylphenylphosphine (0.032 g,
0.23 mmol) were combined and hydrogenated for 30 h.
Puri®cation by ¯ash chromatography gave the two
diastereomers of the 2-allyl-2-carbethoxycyclohexan-1-ol,
identi®ed by spectroscopic comparison to literature data36
(10:1, 0.063 g, 79%). Major isomer (alcohol cis to ester):
0.057 g. TLC Rfˆ0.34, 5:1 hexanes/ethyl acetate; IR (neat,
cm21) 3500, 3080, 2980±2860, 1720, 1635, 1450, 1220,
Using the general procedure, 6,10-dimethyl-9-undecen-2-
one (0.045 g, 0.229 mmol), [(Ph3P)CuH]6 (3.8 mg,
0.0019 mmol, 5 mol% Cu), tert-butanol (8.5 mg,
0.12 mmol), and dimethylphenylphosphine (9.4 mg,
0.068 mmol) were combined and hydrogenated for 36 h,
giving 6,10-dimethyl-9-undecen-2-ol (0.039 g, 87%). TLC
Rfˆ0.07, 10:1 hexanes/ethyl acetate; IR (neat, cm21) 3350,
1
3040, 3020, 2960±2850, 1450, 1375, 1115, 1065, 820; H
NMR (500 MHz, CDCl3) d 5.09 (1H, tm, Jˆ7.2 Hz), 3.79
(1H, sextet, Jˆ5.8 Hz), 1.95 (2H, m), 1.67 (3H, d,
Jˆ0.8 Hz), 1.59 (3H, s), 1.49±1.22 (8H, m), 1.18 (3H, d,
Jˆ6.2 Hz), 1.17±1.07 (2H, m), 0.86 (3H, d, Jˆ6.6 Hz);
13C{1H} (125 MHz, CDCl3) d 131.0, 124.9, 68.2, 39.7,
37.1, 36.9, 32.3, 25.7, 25.5, 23.5, 23.2, 19.5, 17.6; HRMS
calcd for C13H26Oˆ198.1985 (M1), foundˆ198.1999. The
product was identical to the material prepared by unam-
biguous synthesis.
1
1200, 1130, 980, 915; H NMR (400 MHz, CDCl3) d 5.79
(1H, m), 5.09±5.04 (2H, m), 4.18 (2H, m, ABX3), 3.54 (1H,
d, Jˆ10.0 Hz), 3.43 (1H, apparent dt, Jˆ3.6, 10 Hz), 2.54
(1H, dd, Jˆ7.6, 13.6 Hz), 2.37 (1H, dd, Jˆ7.6, 13.6 Hz),
2.14±2.08 (1H, m), 1.92±1.84 (1H, m), 1.72±1.64 (1H, m),
1.56±1.4 (2H, m), 1.34±1.14 (3H, m), 1.27 (3H, t,
Jˆ7.2 Hz); Minor isomer (alcohol trans to ester): 0.006 g.
TLC Rfˆ0.29, 5:1 hexanes/ethyl acetate; 1H NMR
(400 MHz, CDCl3) d 5.79 (1H, m), 5.10±5.03 (2H, m),
4.18 (2H, m, ABX3), 3.96 (1H, apparent dt, Jˆ3.6, 8.8 Hz),
2.8 (1H, bs), 2.59 (1H, dd, Jˆ7.6, 13.6 Hz), 2.40 (1H, dd,
Jˆ7.6, 13.6 Hz), 1.84±1.18 (8H, m), 1.28 (3H, t, Jˆ7.2 Hz).
Reduction of 4-cyclooctenone
Using the general procedure, 4-cyclooctenone (0.050 g,
0.403 mmol), [(Ph3P)CuH]6 (6.6 mg, 0.0034 mmol,
5 mol% Cu), tert-butanol (0.015 g, 0.201 mmol), and
dimethylphenylphosphine (0.017 g, 0.121 mmol) were
combined and hydrogenated for 30 h, giving 4-cyclo-
octen-1-ol38 (0.024 g, 95%). The products were identi®ed
by comparison to literature data and an authentic sample.
Reduction of 3-[4-(p-toluenesulfonyloxy)butyl]-2-cyclo-
hexanone
Using the general procedure, 3-[4-(p-toluenesulfonyloxy)-
butyl]-2-cyclohexenone (0.050 g, 0.15 mmol), [(Ph3P)CuH]6
(5.0 mg, 0.0025 mmol, 10 mol% Cu), tert-butanol
(0.0011 g, 0.15 mmol), and dimethylphenylphosphine
(0.013 g, 0.093 mmol) were combined and hydrogenated
Reduction of trideca-2-one-6-yne
Using the general procedure, trideca-2-one-6-yne (0.050 g,

2162
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
for 60 h. Puri®cation by ¯ash chromatography (8:1 hexanes/
ethyl acetate) gave the starting ketone (0.008 g, 16%) and a
separable mixture of 3-[4-(p-toluenesulfonyloxy)butyl]cyclo-
hexan-1-ol diastereomers (1.6:1, 0.38 g, 76%). Minor trans-
isomer (0.015 g, 30%): TLC Rfˆ0.1, 2:1 hexanes/ethyl ace-
tate; IR (CDCl3, cm21) 3600, 2925, 2850, 1595, 1350, 1180,
1170, 1060, 1020, 960; 1H NMR (400 MHz, CDCl3) d 7.78
(2H, d, Jˆ8.3 Hz), 7.34 (2H, d, Jˆ8.3 Hz), 4.02 (3H, m,
containing a 2H, t, Jˆ6.4 Hz), 2.44 (3H, s), 1.62 (7H, m),
1.51±1.35 (3H, m), 1.30(2H, m), 1.25±1.08 (3H, m), 0.90
(1H, m); 13C{1H} NMR (100 MHz, CDCl3) d 144.6, 133.3,
129.8, 127.9, 70.6, 66.7, 39.5, 35.7, 33.3, 32.0, 31.3, 29.0,
22.7, 21.6, 19.9. Major cis-isomer (0.0235 g, 47%): TLC
in the drybox. Of this solution, 0.20 ml (10 mol% Cu) was
added to a small volume glass reaction bomb containing
4-benzyloxy-6-bromo-(5E,7E)-1,5,7-dodecatrien-11-one
(0.010 g, 0.028 mmol) in benzene (0.50 ml) and a stir bar.
The reaction vessel placed under a slight positive pressure
of hydrogen and stirred for 30 min. Standard work-up and
puri®cation by ¯ash chromatography (20:1 hexanes/ethyl
acetate) gave recovered starting material (0.0008 g, 8.0%)
and 4-benzyloxy-6-bromo-(5E,7E)-1,5,7-dodecatrien-11-ol
(0.0090 g, 89%) as an inseparable 1:1 mixture of diastereo-
mers. TLC Rfˆ0.22, 4:1 hexanes/ethyl acetate; IR (CCl4,
cm21) 3620, 3350, 3060, 3030, 2970, 2930, 2860, 1940,
1870, 1830, 1800, 1640, 1605, 1490, 1450, 1370, 1330,
1260, 1240, 1200, 1025, 985, 950, 915, 850, 840, 730,
Rfˆ0.07, 2:1 hexanes/ethyl acetate; IR (CDCl3, cm21
)
1
3620, 2950, 2875, 1595, 1360, 1190, 1180, 1080, 1020;
1H NMR (400 MHz, CDCl3) d 7.78 (2H, d, Jˆ8.0 Hz),
7.34 (2H, d, Jˆ8.0 Hz), 4.0 (2H, t, Jˆ6.5 Hz), 3.52 (1H,
tt, Jˆ4.3, 10.9 Hz), 2.45 (3H, s), 1.93 (2H, bm), 1.74 (1H,
dp, Jˆ13.4, 3.4 Hz), 1.65±1.55 (3H, m), 1.47 (1H, bs),
1.35±1.05 (7H, m), 0.81 (1H, apparent q, Jˆ11.6 Hz),
0.71 (1H, apparent qd, Jˆ11.6, 3.7); 13C{1H} NMR
(100 MHz, CDCl3) d 144.6, 133.3, 129.8, 127.9, 70.7,
70.5, 42.5, 36.2, 35.8, 31.9, 29.0, 24.0, 22.7, 21.6; HRMS
cacld for C17H26O4Sˆ326.1553 (M1), foundˆ326.1544.
695; H NMR (400 MHz, CDCl3) d 7.37±7.25 (5H, m),
6.22±6.06 (2H, m), 5.92±5.78 (2H, m), 5.14±5.02 (2H,
m), 4.50 (2H, AB quartet), 4.47 (1H, m), 3.84 (1H,
apparent sextet, Jˆ6.0 Hz), 2.55±2.21 (4H, m), 1.59
(2H, m), 1.31 (1H, bs), 1.22 (3H, d, Jˆ6.8 Hz);
HRMS calcd for C16H20O2Br 323.0641 (M1-allyl),
foundˆ323.0635.
Reduction of trans-dihydrocarvone [2(R)-methyl-5(R)-
(2-propenyl)cyclohexan-1-one]
Reduction of 1(b)-acetoxy-8a(b)-methyl-
1,2,3,4,4a(a),5,8,8a-octahydronaphthalene-6(7H)-one
Using the general procedure, dihyrocarvone (0.070 g,
0.459 mmol), [(Ph3P)CuH]6 (7.5 mg, 0.0038 mmol,
5 mol% Cu), tert-butanol (0.017 g, 0.23 mmol), and
dimethylphenylphosphine (0.025 g, 0.184 mmol) were
combined and hydrogenated for 24 h. Puri®cation by ¯ash
chromatography (10:1 hexanes/ethyl acetate) gave 0.064 g
(90%) of the following mixture dihydrocarveol48 isomers:
2(R)-methyl-5(R)-(2-propenyl)cyclohexan-1(S)-ol (60%),
2(R)-methyl-5(R)-(2-propenyl)cyclohexan-1(R)-ol (30%),
Using the general procedure, 1(b)-acetoxy-8a(b)-methyl-
1,2,3,4,4a(a),5,8,8a-octahydronaphthalene-6(7H)-one (0.075
g, 0.33 mmol), [(Ph3P)CuH]6 (5.5 mg, 0.0028 mmol,
5 mol% Cu), tert-butanol (0.015 g, 0.21 mmol), and
dimethylphenylphosphine (0.018 g, 0.13 mmol) were
combined and hydrogenated for 27 h. Spectroscopic
analysis of the crude reaction mixture (1H NMR,
400 MHz, CDCl3) showed 1(b)-acetoxy-6(b)-hydroxy-
trans-decalin37 and 1(b)-acetoxy-6(a)-hydroxy-trans-deca-
lin37 in a ratio of 16:1 along with two other very minor
components. Puri®cation by ¯ash chromatography (8:1
hexanes/ethyl acetate) gave 1(b)-acetoxy-6(b)-hydroxy-
trans-decalin and 1(b)-acetoxy-6(a)-hydroxy-trans-decalin
(12:1, 0.070 g, 93%), spectroscopically consistent with data
from the literature.37 Major isomer (0.065 g, 86%): TLC
and
2(S)-methyl-5(R)-(2-propenyl)cyclohexan-1(R)-ol
(10%), as determined by analysis of the 500 MHz 1H
NMR spectrum and comparison to literature values. The
fourth isomer, 2(S)-methyl-5(R)-(2-propenyl)cyclohexan-
1(S)-ol was not detected. A similar reaction run to low
conversion gave recovered dihydrocarvone as a mixture of
cis and trans isomers (ca. 1:1).
Rfˆ0.14, 2:1 hexanes/ethyl acetate; IR (CDCl3, cm21
)
3610, 3450, 2980, 2940, 2860, 1725, 1470, 1450, 1380,
Reduction of 6(S),7-(cyclohexylidenyl-1,1-dioxy)-5(S)-
methoxy-3(R)-methylheptan-2-one
1
1260, 1160, 1110, 1090, 1040, 1030; H NMR (500 MHz,
CDCl3) d 4.47 (1H, dd, Jˆ4.5, 11.5 Hz), 3.59 (1H, tt-appar-
ent septet, Japparentˆ6.1 Hz), 2.02 (3H, s), 1.80 (1H, m), 1.72
(2H, m), 1.63 (3H, m), 1.53 (1H, m), 1.41 (2H, m), 1.33±
1.18 (4H, m), 1.09 (1H, dt, Jˆ3.8, 13.5 Hz), 0.91 (3H, s).
Minor isomer (0.0055 g, 7%): TLC Rfˆ0.19, 2:1 hexanes/
ethyl acetate; IR (CDCl3, cm21) 3610, 3450, 2930, 2860,
1725, 1470, 1450, 1380, 1260, 1070, 1095, 1030, 1010, 995,
Using the general procedure, 6(S),7-(cyclohexylidenyl-1,1-
dioxy)-5(S)-methoxy-3(R)-methylheptan-2-one (0.050 g,
0.185 mmol), [(Ph3P)CuH]6 (3.0 mg, 0.0015 mmol,
5 mol% Cu), tert-butanol (7.0 mg, 0.091 mmol), and
dimethylphenylphosphine (0.010 g, 0.074 mmol) were
combined and hydrogenated for 30 h. Puri®cation by ¯ash
chromatography (3:1 hexanes/ethyl acetate) gave 6(S),7-
(cyclohexylidenyl-1,1-dioxy)-2-hydroxy-5(S)-methoxy-3-
methylheptane (0.047 g, 93%) as an approximately 1:1:1:1
mixture of four diastereomers. IR (neat, cm21) 3440, 2980±
2850, 1460±1430, 1360, 1320, 1280, 1260, 1220, 1160,
1140±1060, 1030, 920; partial 1H NMR (400 MHz,
CDCl3) d 3.48 (3H, s), 3.47 (3H, s), 3.46 (3H, s), 3.42
(3H, s); 2.85 (1H, d, Jˆ3.6 Hz), 2.25 (1H, d, Jˆ5.2 Hz),
2.16 (1H, d, Jˆ4.4 Hz), 2.01 (1H, d, Jˆ4.4 Hz); 1.17±1.13
(overlapping d's, 4£3H); 0.94±0.90 (overlapping d's,
4£3H); HRMS calcd. for C15H28O4ˆ272.1988 (M1),
1
975; H NMR (500 MHz, CDCl3) d 4.57 (1H, dd, Jˆ4.2,
11.4 Hz), 4.05 (1H, m), 2.03 (3H, s), 1.77±1.64 (4H, m),
1.63±1.36 (8H, m), 1.30±1.08 (2H, m), 0.89 (3H, s).
Reduction of 4-benzyloxy-6-bromo-(5E,7E)-1,5,7-
dodecatrien-11-one
[(Ph3P)CuH]6 (0.0045 g, 0.0023 mmol), tert-butanol
(0.010 g, 0.14 mmol), and dimethylphenylphosphine
(0.012 g, 0.083 mmol) were combined in benzene (1 ml)

J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
2163
foundˆ272.1989. Starting material recovered from a hydro-
[(Ph3P)CuH]6 (5.3 mg, 0.0027 mmol, 5 mol% Cu), tert-
butanol (0.048 g, 0.65 mmol), and dimethylphenyl-
phosphine (0.013 g, 0.097 mmol) were combined and
hydrogenated under 500 psig of hydrogen for 15 h. 1H
NMR spectroscopic analysis of the crude mixture indicated
the formation of geraniol (mixture of E and Z isomers) and
citronellol in a ratio of 11:1, identi®ed by comparison to
authentic (commercial) samples. The inseparable products
(0.045 g, 90%) were isolated by ¯ash chromatography (7:1
hexane/ethyl acetate).
genation run to low conversion was completely equilibrated.
General procedure for reduction of a,b-unsaturated
ketones and aldehydes using [(Ph3P)CuH]6 and Me2PPh
In the glove box, [(Ph3P)CuH]6 (5 mol% Cu), Me2PPh
(6 equiv./Cu) and tert-butanol (40 equiv./Cu) were
dissolved in benzene. The substrate (20 equiv./Cu) was
added and the resulting solution was rapidly transferred to
a Fischer & Porter bottle equipped with a magnetic stirbar or
to a similarly equipped steel autoclave (total volume: 0.4±
0.8 M in substrate). The apparatus was sealed, transported
out of the glove box, and connected to a tank of prepuri®ed
hydrogen. After pressurizing and releasing several times to
¯ush the apparatus, the vessel was charged to the desired
pressure and stirred for the indicated reaction time. After
releasing the pressure, the contents were exposed to air,
stirred several minutes, and ®ltered through celite. The
solvent was removed in vacuo and the crude mixture
Reduction of perillaldehyde
Using the general procedure, (S)-(2)-perillaldehyde
(0.049 g, 0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027
mmol, 5 mol% Cu), tert-butanol (0.048 g, 0.65 mmol),
and dimethylphenylphosphine (0.013 g, 0.097 mmol) were
combined and hydrogenated under 500 psig of hydrogen for
1
18 h. H NMR spectroscopic analysis of the crude mixture
indicated the formation of (S)-(2)-perillyl alcohol and
1-hydroxymethyl-4-isopropenyl-1-cyclohexane49 (stereo-
chemistry unassigned) in a ratio of 32:1. The major product
was identi®ed by comparison to an authentic (commercial)
sample; the minor was identi®ed by comparison to the
literature data. The inseparable products (0.047 g, 95%)
were isolated by ¯ash chromatography (7:1 hexane/ethyl
acetate).
1
analyzed by H NMR spectroscopy (integration at long
pulse delay) to determine product ratios. Puri®cation and
isolation of the product(s) was accomplished by ¯ash
chromatography on silica gel.
Reduction of trans-cinnamaldehyde
Using the general procedure, cinnamaldehyde (0.043 g,
0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027 mmol, 5 mol%
Cu), tert-butanol (0.048 g, 0.65 mmol), and dimethyl-
phenylphosphine (0.013 g, 0.097 mmol) were combined
Reduction of trans-4-phenyl-3-buten-2-one
Using the general procedure, trans-4-phenyl-3-buten-2-one
(0.047 g, 0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027 mmol,
5 mol% Cu), tert-butanol (0.048 g, 0.65 mmol), and
dimethylphenylphosphine (0.013 g, 0.097 mmol) were
combined and hydrogenated under 500 psig of hydrogen
for 18 h. After releasing the pressure, the solution was
cooled to 08C and acetic anhydride (0.5 ml) and pyridine
(1 ml) were added. The solution was stirred at 08C for 1 h
and warmed to room temperature until the reaction was
complete by TLC analysis. The volatiles were evaporated
in vacuo and the residue was puri®ed by ¯ash chromato-
graphy to give the inseparable trans-2-acetoxy-4-phenyl-3-
butene and 2-acetoxy-4-phenylbutane as a 12:1 mixture
(0.051 g, 84%). The products were identi®ed by comparison
1
and hydrogenated under 70 psig of hydrogen for 4 h. H
NMR spectroscopic analysis of the crude mixture indicated
the formation of cinnamyl alcohol and 3-phenyl-1-propanol
in a ratio of 32:1, identi®ed by comparison to authentic
samples. The inseparable products (0.041 g 94%) were
isolated via ¯ash chromatography on silica gel (10:1
hexane/ethyl acetate), along with an impure by-product
(ca. 0.002 g) tentatively identi®ed as PhCHvCHCH2O-
1
C(O)CHvCHPh: H NMR (200 MHz, C6D6) d 4.75 (dd,
Jˆ7.5, 1.5 Hz, 2H), 6.20 (dt, Jˆ15.5, 7.5 Hz, 1H), 6.49 (d,
Jˆ15.5 Hz, 1H), 6.68 (d, Jˆ15.5 Hz, 1H), 6.9±7.3 (m,
aromatic-H), 7.85 (d, Jˆ15.5 Hz, 1H).
1
Reduction of 2,4-dimethyl-2,6-heptadienal
to samples prepared by unambiguous synthesis. H NMR
(300 MHz, C6D6), trans-2-acetoxy-4-phenyl-3-butene: d
1.21 (d, Jˆ7.5 Hz, 3H), 1.72 (s, 3H), 5.58 (m, 1H), 6.08
(dd, Jˆ15.5, 7.5 Hz, 1H), 6.51 (d, Jˆ15.5 Hz, 1H), 6.95±
7.25 (m); 2-acetoxy-4-phenylbutane: d 1.11 (d, Jˆ6.1 Hz,
3H), 1.79 (s, 3H), 1.83 (s, 2H), 2.45±2.70 (m, 2H), 4.90±
5.08 (m, 1H), 7.02±7.38 (m).
Using the general procedure, 2,4-dimethyl-2,6-heptadienal
(0.045 g, 0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027 mmol,
5 mol% Cu), tert-butanol (0.048 g, 0.65 mmol), and
dimethylphenylphosphine (0.013 g, 0.097 mmol) were
combined and hydrogenated under 400 psig of hydrogen
for 30 h. 1H NMR spectroscopic analysis of the crude
mixture indicated the formation of 2,4-dimethyl-2,6-
heptadien-1-ol and 2,4-dimethyl-6-hepten-1-ol44 in a ratio
of 16:1, identi®ed by comparison to authentic samples
(2,4-dimethyl-2,6-heptadien-1-ol is commercially avail-
able). The inseparable products (0.042 g, 91%) were
isolated by ¯ash chromatography (3:1 hexane/ethyl
acetate).
Reduction of b-ionone
Using the general procedure, b-ionone (0.062 g,
0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027 mmol, 5 mol%
Cu), tert-butanol (0.048 g, 0.65 mmol), and dimethyl-
phenylphosphine (0.013 g, 0.097 mmol) were combined
and hydrogenated under 500 psig of hydrogen for 26 h.
NMR spectroscopic analysis of the crude product indicated
the formation of the allylic alcohol and saturated alcohol in
a ratio of 49:1. The inseparable products (0.056 g, 89%)
were puri®ed by ¯ash chromatography (10:1 hexane/ethyl
Reduction of citral
Using the general procedure, citral (0.049 g, 0.32 mmol),

2164
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
acetate). The allylic alcohol was identi®ed by comparison to
an authentic sample: H NMR (300 MHz, C6D6) d 1.02 (s,
the glove box, pressurized to 1000 psi of hydrogen, and
allowed to stir for 24 h. After work up as described in the
general procedure, analysis of the crude product mixture by
1H NMR spectroscopy indicated the formation of the
3,5-dimethyl-2-cyclohexen-1-ol (12:1 mixture of stereo-
isomers) and 3,5-dimethyl-1-cyclohexanol (,3:1 mixture
of stereoisomers) in a ratio of 4.4:1. The inseparable
products (0.037 g, 92%) were isolated via ¯ash chroma-
tography (3:1 hexane/ethyl acetate) and identi®ed as
described above.
1
6H), 1.18 (d, Jˆ6.1 Hz, 3H), 1.41 (m, 2H), 1.52 (m, 2H),
1.67 (s, 3H), 1.90 (m, 2H), 4.13 (m, 1H), 5.47 (dd, Jˆ15.5,
6.2 Hz, 1H), 6.03 (d, Jˆ15.5 Hz, 1H). The minor saturated
alcohol was identi®ed by comparison to an authentic sample
and literature data.47
Reduction of 1-acetyl-1-cyclohexene
Using the general procedure, 1-acetyl-1-cyclohexene
(0.040 g, 0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027
mmol, 5 mol% Cu), tert-butanol (0.048 g, 0.65 mmol),
and dimethylphenylphosphine (0.013 g, 0.097 mmol) were
combined and hydrogenated under 500 psig of hydrogen for
Acknowledgements
Financial support from NSERC and the University of
Alberta is gratefully acknowledged.
1
20 h. H NMR spectroscopic analysis of the crude product
indicated the formation of the 1-cyclohexenylethanol and
the commercially available 1-cyclohexylethanol in a ratio
of 17:1. The inseparable products (0.038 g, 94%) were puri-
®ed by ¯ash chromatography (4:1 hexane/ethyl acetate) and
identi®ed by comparison to authentic materials. 1-Cyclo-
hexenylethanol: 1H NMR (400 MHz, C6D6) d 1.16 (d,
Jˆ6.5 Hz, 3H), 1.35±1.60 (m, 4H), 1.75±2.05 (m, 4H),
3.94 (q, Jˆ6.3 Hz, 1H), 5.53 (br s, 1H).
References
1. General hydrogenation reviews: (a) James, B. R. Homogeneous
Hydrogenation; Wiley: New York, 1973. (b) James, B. R. Adv.
Organomet. Chem. 1979, 17, 319. (c) Rylander, P. N. Catalytic
Hydrogenation in Organic Synthesis; Academic Press: New York,
1979. (d) Mestroni, G.; Camus, A.; Zassinovich, G. Aspects of
Homogeneous Catalysis 1981, 4, 71.
Reduction of 3,5-dimethylcyclohexenone using
[(Ph3P)CuH]6 and Me2PPh
2. Ohkuma, T.; Ooka, H.; Ikariya, T.; Noyori, R. J. Am. Chem.
Soc. 1995, 117, 10417.
Using the general procedure, 3,5-dimethylcyclohexenone
(0.040 g, 0.32 mmol), [(Ph3P)CuH]6 (5.3 mg, 0.0027
mmol, 5 mol% Cu), tert-butanol (0.048 g, 0.65 mmol),
and dimethylphenylphosphine (0.013 g, 0.097 mmol) were
combined and hydrogenated under 500 psig of hydrogen for
3. Hydrogenation of nonconjugated unsaturated ketones: (a) Kaye,
I. A.; Matthews, R. S. J. Org. Chem. 1964, 29, 1341. (b) Mestroni,
G.; Zassinovich, G.; Camus, A. J. Organomet. Chem. 1977, 140,
63. (c) Mestroni, G.; Spogliarich, R.; Camus, A.; Martinelli, F.;
Zassinovich, G. J. Organomet. Chem. 1978, 157, 345. (d) Jenck, J.;
Germain, J.-E. J. Catalysis 1980, 65, 141. (e) Spogliarich, R.;
Tencich, A.; Kaspar, J.; Graziani M. J. Organomet. Chem. 1982,
240, 453. (f) Ishiyama, J.; Maeda, S.; Takahashi, S.; Senda, Y.;
Imaizumi, S. Bull. Soc. Chem. Jpn. 1987, 60, 1721.
1
30 h. H NMR spectroscopic analysis of the crude product
indicated the formation of the 3,5-dimethyl-2-cyclohexen-
1-ol (12:1 mixture of stereoisomers) and 3,5-dimethyl-1-
cyclohexanol (,3:1 mixture of stereoisomers) in a ratio of
2.7:1. The inseparable products (0.036 g, 90%) were puri-
®ed by ¯ash chromatography (3:1 hexane/ethyl acetate) and
identi®ed by comparison to authentic materials. The
isomeric 3,5-dimethyl-2-cyclohexen-1-ols were identi®ed
by comparison to authentic materials prepared by unam-
4. Hydrogenation of conjugated unsaturated aldehydes: (a)
Adams, R.; Carvey, B. J. Am. Chem. Soc. 1926, 48, 477. (b)
Pummerer, R.; Aldebert, F.; Graser, F. Sperber, H. Ann. 1953,
583, 225. (c) Rylander, P. N. Tetrahedron Lett. 1969, 1579. (d)
Hotta, K.; Kubomatsu, T. Bull. Soc. Chem. Jpn. 1973, 46, 3566. (e)
Gradeff, P. S.; Formica, G. Tetrahedron Lett. 1976, 4681. (f)
Mizoroki, T.; Seki, K.; Meguro, S.; Ozaki, A. Bull. Soc. Chem.
Jpn. 1977, 50, 2148. (g) James, B. R.; Morris, R. J. Chem. Soc.,
Chem. Commun. 1978, 929. (h) Strohmeier, W.; Holke, K.
J. Organomet. Chem. 1980, 193, C63. (i) Hotta, K. J. Mol.
Catal. 1985, 29, 105. (j) Sanchez, R. A.; Andriollo, A.; Valencia,
A. J. Chem. Soc., Dalton Trans. 1985, 1859. (k) Farnetti, E.;
Kaspar, J.; Spogliarich, R.; Graziani, M.; Pesce, M. J. Mol.
Catal. 1987, 43, 35. (l) Suarez, T.; Fontal, B. J. Mol. Catal.
1988, 45, 345. (m) Joo, F.; Benyei, A. J. Organomet. Chem.
1990, 58, 151. (n) Nitta, Y.; Hiramatsu, Y.; Imanaka, T. J. Catal.
1990, 126, 235 and references therein. (o) Grosselin, J. M.;
Mercier, C.; Allmang, G.; Grass, F. Organometallics 1991, 10,
2126. (p) Wieland, S.; Panster, P. In Catalysis of Organic Reac-
tions, Scaros, M. G.; Prunier, M. L., Eds. Chem. Ind. 1995, 62;
Marcel Dekker: New York; 1995; pp 383±392. (q) Kaneda, K.;
Mizugaki, T. Organometallics 1996, 15, 3247 and references
therein.
1
biguous synthesis; H NMR (400 MHz, C6D6, resonances
used in product identi®cation only), major cis-isomer: d
5.45 (br s, 1H), 4.19 (complex m, width at half-height
,21 Hz, 1H), 1.53 (s, 3H), 0.81 (d, Jˆ6.4 Hz, 3H); minor
trans-isomer: d 5.5 (m, 1H), 4.10 (br s, width at half-height
,7 Hz, 1H). The 3,5-dimethyl-1-cyclohexanols were
1
identi®ed by comparison to the commercial material; H
NMR (400 MHz, C6D6, resonances used in product identi-
®cation only): major isomer, 3,5-cis-dimethyl-1-trans-
cyclohexanol: d 4.05 (tt, Jˆ2.9, 2.7 Hz, 1H), 0.86 (d,
Jˆ6.7 Hz, 3H); minor isomer, 3,5-cis-dimethyl-1-cis-cyclo-
hexanol: d 3.57 (tt, Jˆ10.9, 4.2 Hz, 1H), 0.74 (d, Jˆ6.6 Hz,
6H).
Reduction of 3,5-dimethylcyclohexenone using CuCl
Cuprous chloride (0.0016 g, 0.016 mmol), NaOtBu
(0.0015 g, 0.016 mmol), Me2PPh (0.0134 g, 0.096 mmol),
tert-butanol (0.048 g, 0.65 mmol) and 3,5-dimethylcyclo-
hexenone (0.040 g, 0.32 mmol) were combined in C6H6
(0.6 ml), transferred to a steel autoclave, removed from
5. Hydrogenation of conjugated unsaturated ketones: (a) Csuros,
Â
Z.; Zech, K.; Geczy, I. Hung. Acta Chim. 1946, 1, 1. (b) Henbest,
H. B.; Trocha-Grimshaw, J. J. Chem. Soc., Perkin 1 1974, 601,

J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
2165
and references therein. (c) Farnetti, E.; Kaspar, J.; Spogliarich, A.;
Graziani, M. J. Chem. Soc., Dalton Trans. 1988, 947. (d) Nakano, T.;
Umano, S.; Kino, Y.; Ishii, Y., Ogawa, M. J. Org. Chem. 1988, 53,
3752. (e) Mashima, K.; Akutagawa, T.; Zhang, X.; Takaya, H.;
Taketomi, T.; Kumobayashi, H.; Akutagawa, S. J. Organometallic
Chem. 1992, 428, 213. (f) Bianchini, C.; Farnetti, E.;
Glendenning, L.; Graziani, M.; Nardin, G.; Peruzzini, M.;
Rocchini, E.; Zanobini, F. Organometallics 1995, 14, 1489. (g)
Bianchini, C.; Peruzzini, M.; Farnetti, E.; Kaspar, J.; Graziani, M.
J. Organomet. Chem. 1995, 488, 91 and references therein.
6. Catalytic hydrosilylation of unsaturated carbonyl compounds:
(a) Lappert, M. F.; Nile, T. A. J. Organomet. Chem. 1975, 102,
543. (b) Ojima, I.; Kogure, T.; Organometallics 1982, 1, 1390 and
references therein. (c) Nishiyama, H.; Kondo, M.; Nakamura, T.;
Itoh, K. Organometallics 1991, 10, 500. (d) Berk, S. C.; Kreutzer,
K. A.; Buchwald, S. L. J. Am. Chem. Soc. 1991, 113, 5093. (e)
Zheng, G. Z.; Chan, T. H. Organometallics 1995, 14, 70. (f)
Lipshutz, B. H.; Keith, J.; Papa, P.; Vivian, R. Tetrahedron Lett.
1998, 39, 4627.
stereomutation to the thermodynamically favored trans-isomer in
solution at room temperature.
18. The attenuation of stereoselectivity is attributed to the
incomplete phosphine exchange, resulting in the presence of the
triphenylphosphine-derived catalyst in low concentration. This
catalyst is also stereoselective for cyclohexanone reduction, but
favors the formation of the epimeric alcohol.8a
19. Goeden, G. V.; Caulton, K. G. Copper(I) catalyzed
Tishchenko reaction. J. Am. Chem. Soc. 1981, 103, 7354.
20. Preliminary account: Daeuble, J. F.; Stryker, J. M. In
Catalysis of Organic Reactions; Scaros, M. G., Prunier,
M. L., Eds. Chem. Ind. 1995, 62; Marcel Dekker: New
York, 1995; pp 235±247.
21. Goeden, G. V., Ph.D. Dissertation, Indiana University, 1982.
Both Ph2PEt and PhPEt2 complexes are poorly characterized
red aggregates, but show spectroscopic characteristics
analogous to the deep red triphenylphosphine hexamer 1. In
contrast, the dimeric complex [(h2-tripod)CuH]2 is yellow,12 as
is the working catalyst derived from complex 1, triphenyl-
phosphine, and substrate.8
7. (a) Churchill, M. R.; Bezman, S. A.; Osborn, J. A.; Wormald, J.
Inorg. Chem. 1972, 11, 1818. (b) Bezman, S. A.; Churchill, M. R.;
Osborn, J. A.; Wormald, J. J. Am. Chem. Soc. 1971, 93, 2063.
(c) Goeden, G. V.; Caulton, K. G. J. Am. Chem. Soc. 1981, 103,
7354.
22. Tolman, A. C. Chem. Rev. 1977, 77, 313.
23. Henderson Jr., W. A.; Buckler, S. A. J. Am. Chem. Soc. 1960,
82, 5794.
24. Vastag, S.; Heil, B.; Marko, L. J. Mol. Catal. 1979, 5, 189.
25. The rate attenuation observed for the reduction of cyclopen-
tanone substrates is consistent with that observed for sodium boro-
hydride reductions of cyclopentanones versus cyclohexanones.
See, for example, Ref. 36.
8. (a) Mahoney, W. S.; Stryker, J. M. J. Am. Chem. Soc. 1989, 111,
8818. (b) Stryker, J. M.; Mahoney, W. S.; Daeuble, J. F.;
Brestensky, D. M. In Catalysis of Organic Reactions; Pascoe,
W. E., Ed. (Chem. Ind. 1992, 47); Marcel Dekker: New York;
1992; pp 29±44. (c) Mahoney, W. S. Ph.D. Dissertation, Indiana
University, 1989. (d) See also Ref. 6f for the use of [(Ph3PCuH]6 in
catalytic hydrosilylation reactions.
26. Zavalii, P. Y.; Fandamenski, V. S.; Myskiv, M. G.;
Gladyshevskii, E. I. Sov. Phys. Crystallogr. 1983, 28, 390.
27. Control experiments were performed by the addition of an
allyl alcohol (e.g. geraniol) to a reaction mixture of a working
hydrogenation reaction, subjecting the allylic alcohol to all
catalysts and catalyst-derived intermediates present during an
active reduction.
9. Lemmen, T. H.; Folting, K.; Huffman, J. C.; Caulton, K. G.
J. Am. Chem. Soc. 1985, 107, 7774.
10. It is interesting that despite the expectation of chelation, the
isolated dppp complex retains a thermodynamic preference for 1:1
phosphine/copper complexation, presumably a function of the
highly electron-rich d10-copper(I) center. Neither 1,2-bis(diphenyl-
phosphino)ethane nor bis(diphenylphosphino)methane survive
exposure to strongly basic copper(I) alkoxides.9
28. Messerle, L. Experimental Organometallic Chemistry; In ACS
Symposium Series 357, Wayda, A. L., Darensbourg, M. Y., Eds.;
American Chemical Society: Washington, DC, 1987.
29. Brestensky, D. M.; Huseland, D. E.; McGettigan, C.; Stryker,
J. M. Tetrahedron Lett. 1988, 29, 3749.
11. Chou, T.; Tsao, C. H.; Hung, S. C. J. Org. Chem. 1985, 50,
4329.
30. (a) Chou, T.; Yuan, J.; Tsao, C. J. Chem. Res. 1985, 18. (b)
Hewertson, W.; Watson, H. R. J. Chem. Soc. 1962, 1490.
31. Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43,
2923.
12. Goeden, G. V.; Huffman, J. C.; Caulton, K. G. Inorg. Chem.
1986, 25, 2484.
13. In these experiments, both the substrate and catalyst concen-
tration were lowered. Despite the net decrease in Cu(I) concen-
tration, equilibrium constant calculations show that the ratio of
catalyst monomer to substrate concentration increases on dilution.
At suf®ciently high dilution, however, the second order nature of
the reduction reaction imposes a decrease in rate.
32. Accary, A.; Infarnet, Y.; Huet, J. Bull. Soc. Chim. Fr. 1973,
2424.
33. Wasson, R. I.; House, H. O. Org. Synth. Coll. 1963, IV, 552.
34. Keinan, E.; Greenspoon, N. J. Am. Chem. Soc. 1986, 108,
7314.
14. This phenomenon is often observed in reactions catalyzed by
metal cluster compounds. See, for example, Laine, R. M. J. Mol.
Catal., 1982, 14, 137.
35. (a) Teisseire, P.; Corbier, B. Recherches 1962, 12, 74. (b)
Sasaki, K. Bull. Chem. Soc. Jpn. 1968, 41, 1252.
36. von Frater, G. Helv. Chim. Acta 1980, 63, 1383.
37. Lin, Y. Y.; Jones, J. B. J. Org. Chem. 1973, 38, 3575.
38. Aranda, V. G.; Barluenga, J.; Yus, M.; Asensio, G. Synthesis
1974, 806.
È
È
15. Plitt, H. S.; Bar, M. R.; Ahlrichs, R.; Schnockel, H. Angew.
Chem., Int. Ed. Engl. 1991, 30, 832.
16. (a) Cornish-Bowden, A. Fundamentals of Enzyme Kinetics;
Butterworths: London, 1979; Chapter 6. (b) Engel, P. C. Enzyme
Kinetics. The Steady-State Approach; Chapman & Hall: New
York, 1981; Chapter 5.
39. (a) Koenig, T. M.; Daeuble, J. F.; Brestensky, D. M.; Stryker,
J. M. Tetrahedron Lett. 1990, 31, 3237. (b) Conia, J.-M.; Rouessac,
F. Tetrahedron 1961, 16, 45. (c) Cahiez, G.; Alexakis, A.;
Normant, J. F. Tetrahedron Lett. 1978, 19, 3013.
17. The stereochemical preference for axial delivery of hydride
represents the kinetic selectivity of the catalytic reduction: the
stereoisomeric tripod-stabilized copper(I)-cis-4-tert-butylcyclo-
hexoxide complex, generated independently by treatment of the
corresponding potassium alkoxide with (tripod)CuCl, shows no
40. Roush, W. R.; Moriarty, K. J.; Brown, B. B. Tetrahedron Lett.
1990, 31, 6509.
41. Williams, D. R.; Benbow, J. W. J. Org. Chem. 1988, 53, 4643.
42. Luche, J.-L. J. Am. Chem. Soc. 1978, 100, 2226.

2166
J.-X. Chen et al. / Tetrahedron 56 (2000) 2153±2166
43. Siegel, S.; Smith, G. V. J. Am. Chem. Soc. 1960, 82, 6087.
44. Brestensky, D. M.; Stryker, J. M. Tetrahedron Lett. 1989, 30,
5677.
47. Schmidt, C.; Adams, K. L.; Fechner, U. Can. J. Chem. 1974,
52, 1732.
48. An isomeric mixture of dihydrocarveols is commercially
available. Spectroscopic data: Fairlie, J. C.; Hodgson, G. L.;
Money, T. J. Chem. Soc., Perkin Trans. 1 1973, 2109.
49. Murphy, R.; Prager, R. Aust. J. Chem. 1978, 31, 1629.
45. Takai, K.; Mori, I.; Ishima, K.; Nozaki, H. Bull. Chem. Soc.
Jpn. 1984, 57, 446.
Â
46. Magnussen, G.; Thoren, S. J. Org. Chem. 1973, 38, 1380.