A Novel B(C6F5)3-Catalyzed Reduction of Alcohols and Cleavage of Aryl and Alkyl Ethers with Hydrosilanes

Article abstract of DOI:10.1021/jo000726d

The primary alcohols 1a-e and ethers 4a-d were effectively reduced to the corresponding hydrocarbons 2 by HSiEt₃ in the presence of catalytic amounts of B(C₆F₅)₃. To the best of our knowledge, this is the first example of catalytic use of Lewis acid in the reduction of alcohols and ethers with hydrosilanes. The secondary alkyl ethers 4j,k enabled cleavage and/or reduction under similar reaction conditions to produce either the silyl ethers 3m-n or the corresponding alcohol 5a upon subsequent deprotection with TBAF. It was found that the secondary alcohols 1g-i and tertiary alcohol 1j, as well as the tertiary alkyl ether 4l, did not react with HSiEt₃/(B(C₆F₅)₃ reducing reagent at all. The following relative reactivity order of substrates was found: primary ? secondary > tertiary. A plausible mechanism for this nontraditional Lewis acid catalyzed reaction is proposed.

Full text of DOI:10.1021/jo000726d

J . Org. Chem. 2000, 65, 6179-6186
6179
A Novel B(C6F 5)3-Ca ta lyzed Red u ction of Alcoh ols a n d Clea va ge of
Ar yl a n d Alk yl Eth er s w ith Hyd r osila n es†
Vladimir Gevorgyan,*,‡ Michael Rubin,‡ Sharonda Benson,‡ J ian-Xiu Liu,§ and
Yoshinori Yamamoto*,§
Department of Chemistry, University of Illinois at Chicago, 845 Taylor Street,
Chicago, Illinois 60607-7061, and Department of Chemistry, Graduate School of Science,
Tohoku University, Sendai 980-8578, J apan
vlad@uic.edu
Received May 13, 2000
The primary alcohols 1a -e and ethers 4a -d were effectively reduced to the corresponding
hydrocarbons 2 by HSiEt3 in the presence of catalytic amounts of B(C6F5)3. To the best of our
knowledge, this is the first example of catalytic use of Lewis acid in the reduction of alcohols and
ethers with hydrosilanes. The secondary alkyl ethers 4j,k enabled cleavage and/or reduction under
similar reaction conditions to produce either the silyl ethers 3m -n or the corresponding alcohol
5a upon subsequent deprotection with TBAF. It was found that the secondary alcohols 1g-i and
tertiary alcohol 1j, as well as the tertiary alkyl ether 4l, did not react with HSiEt3/(B(C6F5)3 reducing
reagent at all. The following relative reactivity order of substrates was found: primary . secondary
> tertiary. A plausible mechanism for this nontraditional Lewis acid catalyzed reaction is proposed.
Reactions of hydrosilanes in the presence of Lewis
acids are very important tools in modern synthetic
organic chemistry. Thus, the Lewis acid-catalyzed hy-
drosilylation of carbon-carbon unsaturated systems is
a powerful approach for the synthesis of various types of
organylsilanes,1 whereas the Lewis acid-catalyzed reduc-
tion of carbonyl function equivalents with hydrosilanes
serves as a useful synthetic tool for the preparation of
alcohols.2 Another area of application of Lewis acid-
hydrosilane combination is the reduction of alcohols and
ethers. The known reducing methods of this type require
at least stoichiometric amounts of Lewis acid.3 Further-
more, the previous methods are most effective for the
reduction of C-O bond at tertiary carbon,3 much less
effective for the reduction of secondary substrates,4 and
absolutely noneffective for that of primary alcohols5 and
ethers (Scheme 1).3 Such reactivity order of tertiary,
secondary, and primary substrates is well understood in
terms of the classical SN1 mechanistic pathway (Scheme
1).3
We have recently communicated6 the following: (1)
even catalytic amount of B(C6F5)3 is enough to effectively
reduce certain alcohols and ethers with HSiEt3; (2) the
reactivity order for the reduction of tertiary, secondary,
and primary substrates with HSiEt3/cat.-B(C6F5)3 is
completely reverse from that of the traditional HSiR3/
Lewis acid reducing systems (Scheme 2).3 In this paper,
we report a full account on this B(C6F5)3-catalyzed
reaction, involving reduction of alcohols and reductive
cleavage of alkyl and aryl ethers, as well as mechanistic
studies of these novel transformations.
Resu lts a n d Discu ssion
† Presented at the Organic Reactions and Processes section of the
Gordon Research Conferences J uly 1999, Hanniker, NH, and at the
218 ACS National Meeting, August 1999, New Orleans, LA.
‡ University of Illinois at Chicago.
B(C6F 5)3-Ca ta lyzed Red u ction of Alcoh ols w ith
Hyd r osila n es. During our studies on the B(C6F5)3-
catalyzed hydrostannation of carbon-carbon multiple
bonds,7 we noticed remarkably strong affinity of B(C6F5)3
toward the hydride of hydrostannanes.8 This fact, to-
gether with the exceptionally high stability of B(C6F5)3,8
encouraged us to investigate the possibility of reduction
of C-O bonds with hydrosilanes in the presence of
catalytic amounts of this unique Lewis acid (eq 1). In a
test experiment, we found that 1-hexadecanol (1a ) un-
derwent complete dehydrocondensation with 1.1 equiv of
HSiEt3 in the presence of 5 mol % of B(C6F5)3 to give the
corresponding silyl ether 3a (Table 1, entry 1).9 Surpris-
§ Tohoku University.
(1) (a) Yamamoto, K.; Takemae, M. Synlett 1990, 259. (b) Asao, N.;
Sudo, T.; Yamamoto, Y. J . Org. Chem. 1996, 61, 7654. (c) Sudo, T.;
Asao, N.; Gevorgyan, V.; Yamamoto, Y. J . Org. Chem. 1999, 64, 2494.
(2) (a) Doyle, M. P.; West, C. T.; Donnelly, S. J .; McOsker, C. C. J .
Organomet. Chem. 1976, 117, 129. (b) Kitazume, T.; Kobayashi, T.;
Yamamoto, T.; Yamazaki, T. J . Org. Chem. 1987, 52, 3218. (c) Kano,
S.; Yokomatsu, T.; Iwasawa, H.; Shibuya, S. Tetrahedron Lett. 1987,
28, 6331. For B(C6F5)3-catalyzed reduction of aldehydes, ketones and
esters, see: (d) Parks, D. J .; Piers, W. E. J . Am. Chem. Soc. 1996, 118,
9440.
(3) (a) Adlington, M. G.; Orfanopoulos, M.; Fry, J . L. Tetrahedron
Lett. 1976, 2955. (b) Fry, J . L.; Orfanopoulos, M.; Adlington, M. G.;
Dittman, W. R.; Silverman, S. B. J . Org. Chem. 1978, 43, 374. (c)
Orfanopoulos, M.; Smonou, I. Synth. Commun. 1988, 18, 833. (d)
Larsen, J . W.; Chang, L. W. J . Org. Chem. 1979, 44, 1168. (e) Yato,
M.; Ishida, A. Heterocycles 1995, 41, 17. (e) Smonou, I. Synth. Commun.
1994, 24, 1999.
(4) For a report on the reduction of secondary benzyl alcohols in
the presence of primary alkyl alcohols with HSiEt3/BF3 system, see
ref 3c.
(5) Recently an effecient one-pot radical-initiated reduction of
primary and secondary alcohols with Et3SiH was reported, see:
Ferreri, C.; Costantino, C.; Chatgilialoglu, C.; Boukherroub, R.; Man-
uel, G. J . Organomet. Chem. 1998, 554, 135.
(6) Gevorgyan V.; Liu J .-X.; Rubin M.; Benson S.; Yamamoto Y.
Tetrahedron Lett. 1999, 40, 8919.
(7) (a) Gevorgyan, V.; Liu, J .-F.; Yamamoto, Y. J . Org. Chem. 1997,
62, 2963. (b) Gevorgyan, V.; Liu, J .-F.; Yamamoto, Y. J . Chem. Soc.,
Chem. Commun. 1997, 37.
(8) Strong affinity of B(C6F5)3 toward hydride of hydrosilanes is well-
known; for a review, see: Piers, W. E.; Chivers, T. Chem. Soc. Rev.
1997, 26, 345. See also ref 2d.
10.1021/jo000726d CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/22/2000

6180 J . Org. Chem., Vol. 65, No. 19, 2000
Gevorgyan et al.
Sch em e 1. Tr a d ition a l Red u ction of Alcoh ols a n d Eth er s w ith Stoich iom etr ic LA-HSiEt3 System
Sch em e 2. Novel Red u ction of Alcoh ols a n d
Eth er s w ith Ca ta lytic B(C6F 5)3-HSiEt3 System
was found that stoichiometric amounts of HSiEt3, in the
ingly, we found that in the presence of 3 equiv of HSiEt3
the primary alcohol 1a was quantitatively reduced into
n-hexadecane (2a , entry 2).10 All other primary alcohols
tested (1b-e) under the same reaction conditions (see
the Experimental Section for details) were also smoothly
reduced into the corresponding hydrocarbons 2b-e in
high to quantitative yields (entries 3-6). Other hydrosi-
lanes tested, such as HSiPh3, HSiMePh2, and HSiMeEt2,
were similarly effective. As expected, phenol (1f) did not
undergo reduction even in the presence of 6 equiv of
hydrosilane, the phenyl triethylsilyl ether 3b was ob-
tained quantitatively, instead (entry 7). Surprisingly
again, the secondary alcohols 1g-i, in contrast to pri-
mary ones, produced the silyl ethers 3c-e in essentially
quantitative yields (entries 8-10), thus exhibiting a
complete resistance toward the reduction.11 The tertiary
alkyl alcohol (1j) did not undergo the reduction at all but
gave 96% of the dehydrocondensation product, silyl ether,
3f (entry 11). In contrast to the alkyl analogues, the
secondary alcohol 1k , possessing two phenyl groups and
tertiary trityl alcohol (1l), smoothly underwent the
reduction even upon treatment with 1.1 equivalents of
hydrosilane to give diphenylmethane 2f and triphenyl-
methane (2g) almost quantitatively (entries 12 and 13,
Table 1).
presence of 5 mol % B(C6F5)3, enabled the cleavage of
linear primary alkyl ethers 4a ,b into the corresponding
hydrocarbons 2a ,e and silyl ethers 3a ,h , respectively
(Table 2, entries 1 and 3), whereas in the presence of
excess amounts of HSiEt3, 4a ,b underwent smooth
exhaustive reduction into the hydrocarbons 2a ,e in
quantitative to high yields, respectively (entries 2 and
4). Very similarly, reduction of a cyclic primary ether 4c
in the presence of 1.1 equiv of HSiEt3/10 mol % B(C6F5)3
gave the corresponding ring-cleaved silyl ether 3i in 96%
yield (entry 5). Here again, the use of 3.0 equiv of HSiEt3/
10 mol % B(C6F5)3 afforded the corresponding hydrocar-
bon 2h (entry 6). Phthalan (4d ) was easily reduced into
the o-xylene in 78% yield (2i, entry 7). As expected, the
aryl C-O bonds in 2,3-dihydrobenzofuran (4e) and
methylenedioxybenzene (4i) were tolerant toward the
reduction, and consequently, the cleavage products tri-
ethylsilyl ether of o-ethylphenol (3j) and catechol (5a )
were produced in quantitative and 79% yields, respec-
tively (entries 8, 12). Similarly, anisole derivatives 4f,g,h
were readily cleaved to form the corresponding phenyl
silyl ethers 3b,k ,l in virtually quantitative isolated yields
(entries 9-11). This method could serve as a useful tool
for the deprotection of alkyl aryl ethers, because it allows
one to perform a quantitative demethylation of anisoles
under very mild conditions,12 unlike the known meth-
ods.13 Obviously, TES-ethers of phenols can be easily
desilylated in situ by a variety of known procedures.13 It
was interesting to find that the secondary alkyl ether 4j
in the presence of 3 equiv of HSiEt3 was quantitatively
cleaved to give the silyl ether 3m (entry 13). Thus, the
secondary alkyl silyl ether 3m (as well as the secondary
B(C6F 5)3-Ca ta lyzed Clea va ge a n d /or Red u ction of
Alk yl Eth er s w ith Hyd r osila n es. Inspired by the
successful reduction of alcohols with HSiEt3/cat-B(C6F5)3
(eq 1, Table 1), we attempted to apply this new reducing
system for the reduction of ethers 4 (eq 2, Table 2). It
(9) While our project was underway, a paper describing silylation
of alcohols in the presence of HSiEt3/B(C6F5)3 system was published;
see: Blackwell, J . M.; Foster, K. L.; Beck, V. H.; Piers, W. E. J . Org.
Chem. 1999, 64, 4887.
(10) It is well-known that primary alcohols and ethers do not
undergo reduction with traditional hydrosilane/Lewis acid reducing
system; see refs 3 and 4.
(11) The observed reduction of primary alcohols and resistance of
secondary alcohols toward reduction by HSiEt3/B(C6F5)3 system is
absolutely opposite to the reactivity order which was observed in the
reduction of alcohols by the classical methods; see refs 3 and 4 (eq 1).
(12) Although reproducible results were obtained with 5 mol % of
commercially available boron catalyst, it was found that the freshly
prepared catalyst was notably more efficient. Thus, only 1 mol % of
B(C6F5)3 was enough for the complete cleavage of anisoles 4g,h . For
the routine synthesis of B(C6F5)3, see: Massey, A. G.; Park, A. J . J .
Organomet. Chem. 1964, 2, 245.
(13) Greene, T. W.; Wuts, P. G. Protective Groups in Organic
Synthesis, 3rd ed.; Wiley: New York, 1999.

B(C6F5)3-Catalyzed Reduction of Alcohols
J . Org. Chem., Vol. 65, No. 19, 2000 6181
Ta ble 1. Red u ction of Alcoh ols 1 w ith HSiR3/Ca t.-B(C6F 5)3 System
a
b
NMR yield. GC yield.
alkyl alcohols, see also Table 1, entries 8-10) exhibited
striking resistance toward reduction. Cyclic secondary
ether 4k behaved similarly, producing the cleavage
product 3n in very high yield (entry 14). Ether 4l,
possessing both tertiary and primary alkyl units, did not
undergo reduction at all (entry 13).14
Mech a n istic Stu d ies a n d Discu ssion . The observed
unusual high reactivity of primary alcohols 1a -e (eq 1,
Table 1) and ethers 4a -h (eq 2, Table 2) toward reduc-
tion with the HSiEt3/cat.-B(C6F5)3 system seemed to be
easily understood in terms of the SN2 mechanistic
pathway rather than SN1 protocol.15 This proposal can
be examined by investigating the stereochemistry of the
reduction of the chiral alcohol (e.g., (S)-(-)-1i) with
deuterated silane (Scheme 3). Indeed, if the reduction
proceeds through the carbenium intermediate 6 (SN1
pathway, Scheme 3), the formation of racemic hydrocar-
bon 7 is unavoidable.3 However, if the concerted mech-
anism is involved (transition state 8), the formation of
product 9 with complete or notable inversion of config-
uration at the secondary carbon center16 should be
observed (SN2 pathway, Scheme 3). Since the above-
mentioned study cannot be applied to primary alcohols,
we searched for a suitable secondary substrate. It was
found that triethylsilyl ether of (S)-(-)1-phenylethanol
(S)-(-)-1i could serve for this purpose. Although the
secondary silyl ether 3e did not undergo reduction with
triethylsilane (Table 1, entry 10), the test experiments
indicated that it could be reduced into 2d by treatment
with less bulky dimethylethylsilane. Therefore, to study
the stereochemistry of the reduction, (S)-(-)-3e was
prepared. The experiment on the reduction of (S)-(-)-3e
(14) Similarly, rather bulky silyl ethers of benzyl alcohol, such as
TIPSOBn (1m ), TBDPSOBn (1n ), and TBDMSOBn (1o), were resistant
toward reduction with HSiEt3/B(C6F5)3 system.
(15) Generally, the nucleophilic substitution at primary sp3 carbon
should proceed via SN2 rather than through an SN1 pathway. For a
review, see, for example: Hartshorn, S. R. Aliphatic Nucleophilic
Substitution; Cambridge University Press: Cambridge, 1973.
(16) For trapping of carbenium cation with chiral hydrosilane, see:
Fry, J . J . Am. Chem. Soc. 1971, 93, 3558.

6182 J . Org. Chem., Vol. 65, No. 19, 2000
Ta ble 2. Clea va ge a n d /or Red u ction of Eth er s 4 w ith HSiEt3/Ca t.-B(C6F 5)3 System a
Gevorgyan et al.
a
b
All reactions were carried in CH2Cl2. For more detailed reaction conditions, see the Experimental Section. NMR yield. c Yields of
alcohols 5 after TBAF deprotection of the corresponding alkoxysilanes 3.
Sch em e 3. Cla ssica l SN1 a n d SN2 P a th w a ys for th e Red u ction of Alcoh ols w ith LA-HSiEt3 System
with DSiEtMe2/B(C6F5)317 in pentane revealed substantial
inversion of its configuration (41% ee of (R)-(-)-phenyl-
ethane-1-d (9) was obtained,18 Scheme 3). However, in
slightly more polar solvent, dichloromethane, the reduc-
tion of (S)-(-)-3e produced nearly racemic 7 (2% ee,
inversion). The last result is in agreement with known
data, where 2% ee (retention) was observed in the
reduction of the same secondary alcohol (S)-(-)-1i under
the traditional method: HSiEt3-stoichiometric amounts
(17) The isotopically homogeneous DSiEtMe2 was prepared under
the phase-transfer conditions, see: (a) Gevorgyan, V.; Ignatovich, L.;
Lukevics, E. J . Organomet. Chem. 1985, 284, C31. (b) Liepins, E.;
Gevorgyan, V.; Lukevics, E. J . Magn. Reson. 1989, 85, 170.
(18) On absolute configuration of enantiomerically pure deuterated
phenylethanes, see: Elsenbaumer, R. L.; Mosher, H. S. J . Org. Chem.
1979, 44, 600.

B(C6F5)3-Catalyzed Reduction of Alcohols
J . Org. Chem., Vol. 65, No. 19, 2000 6183
of BF3(gaseous).19 Intrigued by the contradictory results
of stereochemistry on the reduction of secondary sub-
strate (S)-(-)-3e, we were eager to learn about the
mechanism for the reduction of primary substrates. The
study of kinetic hydrogen/deuterium isotope effect in the
reduction of primary alcohols could give an important
missing information on this matter.20 Provided that the
concerted mechanism with certain degrees of symmetry
(traditional SN2 pathway, Scheme 2) is responsible for
the reduction of primary substrates, then the substantial
primary isotope effect should be observed.20 This effect
will occur because the transition of deuteride from the
reagent system to the electrophilic carbon center will take
place at the rate-determining step.20 Otherwise, if the
reduction of primary substrates proceeds via the carbe-
nium intermediate 6, the deuteride transfer would not
be a rate determining step,21 consequently no notable
isotope effect should be detected (Scheme 3). For the
measurement of the primary kinetic isotope effect, we
chose the reduction of triethylsilyl ether of phenylethanol
3o22 with 1:1 mixture of HSiEt3 and DSiEt3 in the
presence of 10 mol % of B(C6F5)3. These studies revealed
negligible primary hydrogen/deuterium isotope effect
(1.11 ( 0.03, and 0.97 ( 0.03 for two series of experi-
ments), thus confirming the classical SN1 type reaction
pathway for the reduction of primary alcohols and ethers
(Scheme 3). It is reasonable to propose that the reduction
of 3o proceeds via the well-known23 phenonium cation i
which leads to a 1:1 mixture of 2d and 2j (eq 3). This
pathway can be easily justified by reduction of 1,1-
dideuteriophenethyl alcohol 1m (eq 4). Indeed, if the
reaction conditions. In contrast to expectations, minor
deuterium scrambling was observed. A 90:10 mixture of
2k and 2l was obtained, thus ruling out an involvment
of the ii as a major intermediate in the mentioned
transformation. Puzzled by the confrontational results
obtained from the stereochemical and isotope effect
studies, we conducted kinetic investigations on the
reduction of phenylethanol 1d with 3 equivalents of
HSiEt3 in the presence of 10 mol % of B(C6F5)3 catalyst.
The plot of the reaction coordinates vs time is presented
in Figure 1. It became clear that the reduction of the
alcohol 1d proceeded in two steps. At the first step, 1d
underwent fast dehydrocondensation27 with hydrosilane
to produce the silyl ether 3o. Thus, after only 1 min, the
reaction mixture consisted of 78% of 3o and 22% of 1d ,
whereas in two minutes the transformation 1d f 3o was
almost quantitative (Figure 1). The second step, reduction
of the silyl ether 3o, appeared to be dramatically slower
than the first step and it took more than 8 h to complete
the formation of the hydrocarbon 2d (Figure 1).
Based on the above-mentioned stereochemical, kinetic,
and isotope effect studies, we propose the following
mechanistic rationale for the observed unusual reduction
of alcohols and ethers in the presence of HSiEt3/cat.-
B(C6F5)3 system (Schemes 4 and 5). The plausible mech-
anisms for the silylation of alcohols and for the reduction
of particular alcohols are depicted in the Scheme 4. The
reduction of diphenylmethanol (1k ) and triphenylmeth-
anol (1l), the alcohols possessing strong cation-stabilizing
groups, could be explained in terms of the classical (path
A) or the modified (path B) SN1 pathways (Scheme 4).
Reversible interaction of 10 with 1k or 1l would form an
oxonium complex 11, which would be transformed into
the carbenium intermediate 12. The latter would react
with hydrosilane 13 to produce diphenylmethane (2f) or
triphenylmethane (2g) and regenerate the catalyst 10
(path A, Scheme 4). According to an alternative mecha-
nism (path B, Scheme 4), the reversible interaction of 10
with 13 would produce an ate complex 15. The silicenium
cation of 1528 could serve as a new Lewis acid that would
coordinate to 1 to form the oxonium complex 16,29,30
which via the carbenium intermediate 17 would produce
hydrocarbons 2f or 2g and would regenerate 10. It is hard
to distinguish between these two possible SN1 pathways
(paths A and B), since both of them reasonably explain
the formation of the same reaction products, the hydro-
carbons 2f or 2g and the byproduct silanol 1431 (Scheme
4). In the case of other alcohols, which do not possess
any strong cation-stabilizing groups, the oxonium com-
plex 16 does not transform into the carbenium interme-
diate 17, instead it collapses via the dehydrocondensation
process into the silyl ether 3,9 boron catalyst 10, and
dihydrogen27 (path C, Scheme 4). Accordingly, in the
phenonium ion ii is the true intermediate in the reduc-
tion of 1m , then the formation of nearly equimolar
mixture of 2k and 2l24 is unavoidable25 (eq 4). Accord-
ingly, the deuterated alcohol 1m was prepared26 and
subjected to the reduction with HSiEt3 under the typical
(19) Smonou, I.; Orfanopoulos, M. Tetrahedron Lett. 1988, 29, 5793.
(20) For a discussion on interpretation of kinetic and product isotope
effects at the rate-determining step in the ene reaction, see, for
example: (a) Douglas, Z. S.; Beak, P. J . Org. Chem. 1987, 52, 3938.
(b) For investigation of kinetic hydrogen/deuterium primary isotope
effect in the reduction of alcohols with hydro- and deuteriosilanes in
the presence of stoichiometric amount of BF3, see ref 19.
(21) For kinetic studies on hydride transfer from hydrosilanes to
carbenium ions, see: Mayr, H.; Basso, N.; Hagen, G. J . Am. Chem.
Soc. 1992, 114, 3060.
(22) We used silyl ether 3k for investigation of isotope effect since
the reduction of silyl ethers seems to be the slowest step in the overall
transformation: alcohol f hydrocarbon (see Figure).
(23) For earlier reports on phenonium ions, see: (a) Cram D. J . J .
Am. Chem. Soc. 1949, 71, 3863. (b) Cram D. J .; Davis R. J . Am. Chem.
Soc. 1949, 71, 3875. (c) Cram D. J . J . Am. Chem. Soc. 1964, 86, 3764.
(24) We assume that k∆ . kS and that the secondary isotopic effect
is negligible.
(25) Saunders W. H.; Asperger S.; Edison D. H. J . Am. Chem. Soc.
1958, 80, 2421.
(26) Wong, C. W.; Hamilton J . T. G.; O’Hagan, D.; Robins, R. J Chem.
Commun. 1998, 1045.
(27) Vigorous hydrogen emission was observed.
(28) We do not postulate an involvement of free silicenium cation
in the proposed mechanism.
(29) Involvement of such ate-complexes in the dehydrocondensation
of alcohols,9 as well as in the reduction of carbonyl group equivalents,30
have been recently unambiguously demonstrated.
(30) Parks, D. J .; Blackwell J . M.; Piers W. E. J . Org. Chem. 2000,
65, 3090.
(31) Formation of 14 was confirmed by GC-MS analyses of the
crude reaction mixtures.

6184 J . Org. Chem., Vol. 65, No. 19, 2000
Gevorgyan et al.
F igu r e 1. Reaction coordinates on reduction of 1d with HSiEt3/cat.-B(C6F5)3 system.
Sch em e 4. P r op osed Mech a n ism for th e Red u ction of 1k ,l a n d Silyla tion of 1a -j w ith 1 Equ iv of Et3SiH
reduction of the “regular” alcohols, the first equivalent
of hydrosilane is completely consumed for the formation
of silyl ethers (entry 1, Table 1; Figure 1; path C, Scheme
4).
unusual reactivity order of primary, secondary and
tertiary alcohols and ethers toward reduction. Thus, only
less hindered triethylsilyl ethers 3 derived from the
primary alcohols 1a -e (Table 1, entries 2-6; Figure 1)32
and primary ethers 4a -e (Table 2, entries 2-8) enabled
to undergo the reduction. In contrast, the triethylsilyl
ethers, obtained from more bulky secondary33 or tertiary
alcohols and ethers,33 exhibited no detectable reduction
under similar reduction conditions (Table 1, entries 8-11;
Table 2, entries 13-14). Reasons for the observed notable
inversion of configuration of the secondary silyl ether (S)-
(-)-3e (Scheme 3 and text above) are not clearly under-
stood. It is hypothesized that (S)-(-)-3e being a border-
We propose that the ate-complex 15 is responsible for
the reduction of silyl and alkyl ethers (Scheme 5). The
silicenium cation of the ate-complex 15 would reversibly
coordinate to the oxygen of silyl ether 3 (or alkyl ether
4) to form an oxonium complex 18, which would produce
the reaction product, hydrocarbon 2, siloxane (or silyl
ether 3) and would regenerate 10 (Scheme 5). The
hydride transfer step 18 f 2 should be a fast step, as it
was supported by the negligible kinetic hydrogen/
deuterium isotope effect in the reduction of 3o (see the
text above). In contrast, the reversible coordination of 15
to 3 to form 18 seems to be the slowest step among the
overall reduction process and completely sterically con-
trolled. This is the key step for understanding the
(32) Test experiments revealed that primary triethylsilyl ethers
underwent reduction into the corresponding hydrocarbons 2.
(33) Control experiments showed that secondary and tertiary tri-
ethylsilyl ethers in contrast to primary ones32 did not undergo reduction
with excess amounts of HSiEt3.

B(C6F5)3-Catalyzed Reduction of Alcohols
J . Org. Chem., Vol. 65, No. 19, 2000 6185
Sch em e 5. P r op osed Mech a n ism for th e Red u ctive Clea va ge of Silyl-Alk yl, Ar yl-Alk yl, a n d
Alk yl-Alk yl Eth er s
an argon atmosphere to a mixture of B(C6F5)3 (5 mol %) and
alcohol 1 or ether 4 (1 mmol) in hexane or CH2Cl2 (1 mL). After
being stirred for 20 h at room temperature, the reaction
mixture was quenched by addition of triethylamine (0.05 mL),
filtered through Celite, and concentrated. After the addition
of appropriate internal standard (CH2Br2 or CH3CCl3 for NMR
or n-pentadecane for GC), the mixture was analyzed by
capillary GC or NMR. In the case of low bp products, the
reaction mixture was analyzed by 1H NMR without concentra-
tion. Isolated yields were determined after preparative column
chromatography on silica gel.
Kin etic Isotop e Effect Stu d ies (Red u ction of 3o). A
mixture of HSiEt3 and DSiEt3 was added at once to a stirred
solution of B(C6F5)3 (10 mol %) and 3o (1 mmol) in anhydrous
hexane (1 mL) under an argon atmosphere. After stirring for
20 h at room temperature, the reaction mixture was quenched
with triethylamine (0.05 mL), filtered through Celite, and
concentrated. The ratios 2d /2j were determined by 1H NMR
analyses. The isotope effects were found as 1.11 ( 0.03 and
0.97 ( 0.03 for two series of experiments with the following
ratios 3o/HSiEt3/DSiEt3 ) 1.0:1.5:1.5 and 1.0:0.5:0.5, respec-
tively.43
line substrate between simple alkyl alcohols and 1k ,l
undergoes reduction via both intermediate 18 (Scheme
5) and some free carbenium intermediate, similar to 12
or 17 (Scheme 4).
In conclusion, we have developed a novel, effective,
nontraditional method for reduction of primary alcohols
and ethers and for deprotection of aryl alkyl ethers with
hydrosilanes in the presence of catalytic amounts of
B(C6F5)3. The reactivity order for the reduction of pri-
mary, secondary and tertiary susbstrates (Scheme 2) is
reverse to that observed in the classical reduction pro-
tocols (Scheme 1).3,4 We believe that the present novel
methodology will serve as a useful tool in synthetic
organic chemistry, complementary to existing methods.
Exp er im en ta l Section
Gen er a l In for m a tion . All manipulations were conducted
under an argon atmosphere using standard Schlenk tech-
niques. Anhydrous solvents were purchased from Aldrich.
Tertiary alcohol 1k ,34 primary ether 4a ,35 and tertiary ether
4l36 were prepared according to the standard procedures. All
other compounds used were commercially available and pur-
chased from Aldrich. Products 3b,37 3d ,38 3e,39 3h ,40 3i,41 and
3m 42 as well as 4l are known compounds and their analytical
data were in agreement with literature data. The spectral data
of new compounds 3a ,c,j-l,n are provided below. All other
reaction products are commercially available compounds and
their analytical data were in perfect agreement with the
authentic samples.
Ster eoch em ica l Stu d ies on th e Red u ction of (S)-(-)-
3e. EtMe2SiD (15 mmol) was added dropwise to the stirred
mixture of (S)-(-)-3e (10 mmol), B(C6F5)3 (5 mol %), and
anhydrous solvent (0.5 M). The reaction mixture was stirred
for 20 h at room temperature, quenched with isopropylamine
(0.5 mL), and filtered through a short column of Silica gel
(eluent: dichloromethane). Eluate was concentrated under
ambient pressure, and the residue was purified by column
chromatography on silica gel (eluent: pentane). Combined
fractions containing 1-deuterioethylbenzene (7 or 9) were
concentrated under an ambient pressure. The resulting con-
B(C6F 5)3-Ca ta lyzed Red u ction of Alcoh ols a n d Eth er s
w ith HSiEt3 (Gen er a l P r oced u r e). HSiEt3 was added under
centrate of 7 or 9 was used for [R]20 determination. Exact
D
concentration of ethylbenzene-d (c ) 30-50) was determined
by 1H NMR using CH2Br2 as an internal standard. The
following stereochemical results were obtained, for the reduc-
tion in pentane: [R]20D ) - 0.356° (c ) 50, pentane), ee ) 44%
and for dichloromethane: [R]20D ) - 0.016° (c ) 30, pentane),
ee ) 2%.
(34) For a review on preparation of alcohols via alkylation of ketones,
see for example: Ashby, E. C.; Laemmle, J .; Neumann, H. M. Acc.
Chem. Res. 1974, 7, 272.
(35) Lenne, H.-U.; Mez, H. C.; Schlenk, W., J r. J ustus Liebigs Ann.
Chem. 1970, 732, 70.
(36) Reuchardt, C.; Grundmeier, M. Chem. Ber. 1975, 108, 2448.
For preparation of benzyl ethers, see, for example: Fukuzawa, A.; Sato,
H.; Masamune, T. Tetrahedron Lett. 1987, 28, 4303.
(37) Hudrlik, P. F.; Minus, D. K. J . Organomet. Chem. 1996, 521,
157.
(38) Onaka, M.; Higuchi, K.; Nanami, H.; Izumi, Y. Bull. Chem. Soc.
J pn. 1993, 66, 2638.
(39) Wright, A.; West, R. J . Am. Chem. Soc. 1974, 96, 3214.
(40) (a) Liepins E.; Zicmane I.; Luckevics E. J . Organomet. Chem.
1986, 306, 167. (b) Fujita M.; Hiyama, T. J . Org. Chem. 1988, 53, 5405.
(41) (a) Bourhis, R.; Frainnet, E. J . Org. Chem. 1975, 40, 205. (b)
Diekman et al. J . Org. Chem. 1967, 32, 3904.
(42) Lorenz C.; Schubert U. Chem. Ber. 1995, 128, 1267. See also
ref 40a.
3a : 1H NMR (CDCl3, 200.13 MHz) δ 3.59 (t, J ) 6.6 Hz,
2H), 1.52 (m, 2H), 1.26 (m, 26H), 0.96 (t, J ) 8.0 Hz, 9H), 0.88
(t, J ) 6.8 Hz, 3H), 0.59 (q, J ) 8.0 Hz, 6H); 13C NMR (50.32
MHz, CDCl3) δ 62.9, 32.8, 31.8, 29.6, 29.4, 29.3, 25.7, 22.6,
14.0, 6.6, 4.3; GC/MS m/z 355 (M+ - 1, <1), 327 (100).
3c: 1H NMR (CDCl3, 400.13 MHz) δ 3.76 (m 1H), 1.26 (m,
26H), 1.12 (d, J ) 6.3 Hz, 3H), 0.96 (t, J ) 7.9 Hz, 9H), 0.88
(43) For the measurement and calculation of kH/kD, see, for ex-
ample: Beak, P.; Berger, K. R. J . Am. Chem. Soc. 1980, 102, 3848.
See also ref 19.

6186 J . Org. Chem., Vol. 65, No. 19, 2000
Gevorgyan et al.
(t, J ) 6.7 Hz, 3H), 0.59 (q, J ) 7.9 Hz, 6H); 13C NMR (100.61
MHz, CDCl3) δ 68.4, 39.8, 31.9, 30.0, 29.3, 25.8, 23.7, 22.6,
14.0, 6.8, 4.9; GC/MS m/z 355 (M+ - 1, <1), 327 (100).
9H), 0.81 (q, J ) 7.9 Hz, 6H); 13C NMR (100.61 MHz, CDCl3)
δ 152.8, 128.4, 128.2, 121.1, 17.5, 6.8, 5.8; GC/MS 236 (M+,
45), 207 (100).
3l: 1H NMR (CDCl3, 500.13 MHz) δ 7.08 (d, J ) 8.0 Hz, 1H),
6.77 (d, J ) 2.2 Hz, 1H), 6.67 (dd, J ) 8.0, 2.2 Hz, 1H), 2.91 (t,
J ) 7.4 Hz, 2H), 2.87 (t, J ) 7.4 Hz, 2H), 2.10 (ps-quintet,
J ) 7.4 Hz, 2H), 1.05 (t, J ) 7.9 Hz, 9H), 0.78 (t, J ) 7.9 Hz,
6H); 13C NMR (125.77 MHz, CDCl3) δ 154.5, 146.0, 137.1,
125.0, 118.0, 116.3, 33.5, 32.5, 26.2, 7.1, 5.4; GC/MS m/z 248
(M+, 56), 219 (100).
3n : 1H NMR (CDCl3, 400.13 MHz) δ 3.77 (m, 1H), 1.29 (m,
6H), 1.12 (d, J ) 6.2 Hz, 3H), 0.96 (t, J ) 7.9 Hz, 9H), 0.89 (t,
J ) 6.9 Hz, 3H), 0.59 (q, J ) 7.9 Hz, 6H); 13C NMR (100.61
MHz, CDCl3) δ 68.3, 39.4, 27.9, 23.7, 22.7, 14.0, 6.8, 4.8; GC/
MS m/z 215 (M+ - 1, <1), 103 (100).
1
3f: H NMR (CDCl3, 500.13 MHz) δ 7.29 (m 2H), 7.24 (m
3H), 2.76 (s 2H), 1.24 (s 6H), 0.95 (t, J ) 7.9 Hz, 9H), 0.59 (q,
J ) 7.9 Hz, 6H); 13C NMR (125.75 MHz, CDCl3) δ 139.3, 131.2,
127.9, 126.3, 74.0, 51.7, 30.1, 7.5, 7.2; GC/MS m/z 249 (M+
15, 3), 173 (100).
-
3j: 1H NMR (CDCl3, 400.13 MHz) δ 7.19 (dd, J ) 7.3, 1.6
Hz, 1H), 7.10 (td, J ) 7.7, 1.6 Hz, 1H), 6.93 (td, J ) 7.3, 0.9
Hz, 1H), 6.82 (dd, J ) 7.7, 0.9 Hz, 1H), 2.86 (q, J ) 7.5 Hz,
2H), 1.24 (t, J ) 7.5 Hz, 3H), 1.06 (t, J ) 7.9 Hz, 9H), 0.82 (q,
J ) 7.8 Hz, 6H); 13C NMR (100.61 MHz, CDCl3) δ 153.4, 134.4,
129.0, 126.4, 120.8, 118.0, 23.5, 14.0, 6.5, 5.2; GC/MS 236 (M+,
48), 207 (100).
3k : 1H NMR (CDCl3, 400.13 MHz) δ 7.00 (d, J ) 7.4 Hz,
2H), 6.82 (t, J ) 7.5 Hz, 1H), 2.27 (s, 6H), 1.03 (t, J ) 7.8 Hz,
J O000726D