Influences of electronic effects and anions on the enantioselectivity in the oxazaborolidine-catalyzed asymmetric borane reduction of ketones

Article abstract of DOI:10.1021/jo048959i

The influence of electronic effects on the enantioselectivity of the oxazaborolidine-catalyzed asymmetric borane reduction of ketones has been observed and investigated with use of parasubstituted acetophenones and propiophenones with a variety of functional groups and B-unsubstituted and B-methoxyoxazaborolidines derived from (S)-2-(diphenylhydroxymethyl)-pyrrolidine with borane and trimethyl borate as catalysts in toluene and THF. The results indicate that Hammett linear free energy electronic effects on the enantioselectivity in the asymmetric reduction were observed and rationalized. Tuning electronic effects of the catalyst can improve the enantioselectivity in the reduction. Another interesting finding to be noted is that anions heavily affect the enantioselectivity, especially for the B-methoxy catalyst, because of their coordination with the boron atom in the catalysts.

Full text of DOI:10.1021/jo048959i

In flu en ces of Electr on ic Effects a n d An ion s on th e
En a n tioselectivity in th e Oxa za bor olid in e-Ca ta lyzed Asym m etr ic
Bor a n e Red u ction of Keton es
J iaxi Xu,* Tiezheng Wei, and Qihan Zhang
Key Laboratory of Bioorganic Chemistry and Molecular Engineering of Ministry of Education,
Department of Chemical Biology, College of Chemistry and Molecular Engineering,
Peking University, Beijing 100871, People’s Republic of China
jxxu@pku.edu.cn
Received J une 21, 2004
The influence of electronic effects on the enantioselectivity of the oxazaborolidine-catalyzed
asymmetric borane reduction of ketones has been observed and investigated with use of para-
substituted acetophenones and propiophenones with a variety of functional groups and B-
unsubstituted and B-methoxyoxazaborolidines derived from (S)-2-(diphenylhydroxymethyl)-
pyrrolidine with borane and trimethyl borate as catalysts in toluene and THF. The results indicate
that Hammett linear free energy electronic effects on the enantioselectivity in the asymmetric
reduction were observed and rationalized. Tuning electronic effects of the catalyst can improve the
enantioselectivity in the reduction. Another interesting finding to be noted is that anions heavily
affect the enantioselectivity, especially for the B-methoxy catalyst, because of their coordination
with the boron atom in the catalysts.
In tr od u ction
a ketone or borane into the reductive system,2d,7c the
reduction temperature,7c,9 solvent,6a,7c,8c additive,9g,10 sec-
ondary reduction,10a,11 and stabilizer in borane,12 etc.
Although a few papers have considered the influence of
Since Itsuno et al. discovered that the reduction of ke-
tones by borane is strongly catalyzed by a chiral ox-
azaborolidine,1 the enantioselective 1,3,2-oxazaborolidine-
catalyzed borane reduction of prochiral ketones to chiral
secondary alcohols has become an important reaction in
asymmetric syntheses, which has been widely used in the
preparation of various secondary alcohols during the past
decade.2 Numerous new efficient oxazaborolidines as
catalysts have been reported and a plethora of applica-
tions have appeared until now. In comparison with the
numerous attempts to search for new catalysts and to
improve the enantioselectivity, several papers have
concentrated on the mechanistic investigation of the cat-
alytic asymmetric reduction reaction,3,4 and some papers
have paid attention to the factors which affect the enan-
tioselectivity in the asymmetric reduction, such as the
structure,2,3,5 stability 3a,6 (including dimerization) and the
amount of the catalyst,3a,6a,7 the borane source8 and the
borane amount,3a,7c the order and rate of the addition of
(5) (a) J ones, T. K.; Mohan, J . J .; Xavier, L. C.; Blacklock, T. J .;
Mathre, D. J .; Sohar, P.; J ones, E. T. T.; Reamer, R. A.; Roberts, F. E.;
Grabowski, E. J . J . J . Org. Chem. 1991, 56, 763-769. (b) Puigjaner,
C.; Vidal-Ferran, A.; Moyano, A.; Pericas, M. A.; Riera, A. J . Org. Chem.
1999, 64, 7902-7911.
(6) (a) Mathre, D. J .; Thompson, A. S.; Douglas, A. W.; Hoogsteen,
K.; Carroll, J . D.; Corley, E. G.; Grabowski, E. J . J . J . Org. Chem. 1993,
58, 2880-2888. (b) Corey, E. J .; Azimioara, M.; Sarshar, S. Tetraherdon
Lett. 1992, 33, 3429-3430. (c) Zhao, J . K.; Bao, X. H.; Liu, X. M.; Wan,
B. S.; Han, X. W.; Yang, C. G.; Hang, J . F.; Feng, Y.; J iang, B.
Tetrahedron: Asymmetry 2000, 11, 3351-3359.
(7) (a) Masui, M.; Shioiri, T. Synlett 1997, 273-274. (b) Xu, J . X.;
Su, X. B.; Zhang, Q. H. Tetrahedron: Asymmetry 2003, 14, 1781-1786.
(c) Xu, J . X.; Wei, T. Z.; Zhang, Q. H. J . Org. Chem. 2003, 68, 10146-
10151. (d) Xu, J . X.; Wei, T. Z.; Xia, J . K.; Zhang, Q. H.; Wu, H. S.
Chirality, 2004, 16, 341-346.
(8) (a) Cho, B. T.; Chun, Y. S. J . Chem. Soc., Perkins. Trans. 1 1999,
2095-2100. (b) Cho, B. T.; Chun, Y. S. Tetrahedron: Asymmetry 1999,
10, 1843-1846. (c) Gilmore, N. J .; J ones, S. Tetrahedron: Asymmetry
2003, 14, 2115-2118.
(9) (a) Brunel, J . M.; Maffei, M.; Buono, G. Tetrahedron: Asymmetry
1993, 4, 2255-2260. (b) Stone, G. B. Tetrahedron: Asymmetry 1994,
5, 465-472. (c) J iang, Y. Z.; Qin, Y.; Mi, A. Q. Chin. Chem. Lett. 1995,
6, 9-12. (d) Prasad, K. R. K.; J oshi, N. N. Tetrahedron: Asymmetry
1996, 7, 3147-3152. (e) Santhi, V.; Rao, J . M. Tetrahedron: Asymmetry
2000, 11, 3553-3560. (f) Garrett, C. E.; Prasad, K.; Repic, O.;
Blacklock, T. J . Tetrahedron: Asymmetry 2002, 13, 1347-1349. (g) Fu,
X. Y.; McAllister, T. L.; Thiruvengadam, T. K.; Tann, C.-H.; Su, D.
Tetrahedron Lett. 2003, 44, 801-804. (h) Huertas, R. E.; Corella, J .
A.; Soderquist, J . A. Tetrahedron Lett. 2003, 44, 4435-4437.
(10) (a) Cai, D. W.; Tschaen, D. M.; Shi, Y. J .; Verhoeven, T. R.;
Reamer, R. A.; Douglas, A. W. Tetrahedron Lett. 1993, 34, 3243-3246.
(b) Shi, Y. J .; Cai, D. W.; Dolling, U. H.; Douglas, A. W.; Tschaen, D.
M.; Verhoeven, T. R. Tetrahedron Lett. 1994, 35, 6409-6412. (c)
Tschaen, D. M.; Abramson, L.; Cai, D. W.; Desmond, R.; Dolling, U.
H.; Frey, L.; Karady, S.; Shi, Y. J .; Verhoeven, T. R. J . Org. Chem.
1995, 60, 4324-4330. (d) Ponzo, V. L.; Kaufman, T. S. Synlett 2002,
1128-1130.
* Address correspondence to this author. Phone: +86-10-6275-1497.
Fax: +86-10-6275-1708.
(1) Itsuno, S.; Sakurai, Y.; Ito, K.; Hirao, A.; Nakahama, S. Bull.
Chem. Soc. J pn. 1987, 60, 395-396.
(2) For recent reviews, see: (a) Wallbaum, S.; Martens, J . Tetrahe-
dron: Asymmetry 1992, 3, 1475-1504. (b) Singh, V. K. Synthesis 1992,
605-617. (c) Deloux, L.; Srebnik, M. Chem. Rev. 1993, 93, 763-784.
(d) Corey, E. J .; Helal, C. J . Angew. Chem., Int. Ed. 1998, 37, 1986-
2012.
(3) (a) Corey, E. J .; Bakshi, R. K.; Shibata, S. J . Am. Chem. Soc.
1987, 109, 5551-5553. (b) Corey, E. J .; Bakshi, R. K.; Shibata, S.;
Chen, C.; Singh, V. K. J . Am. Chem. Soc. 1987, 109, 7925-7926. (c)
Corey, E. J .; Link, J . O. Tetrahedron Lett. 1989, 30, 6275-6278.
(4) Nevalainen, V. Tetrahedron: Asymmetry 1991, 2, 63-73, 429-
435, 827-842, and 1133-1155.
10.1021/jo048959i CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/03/2004
6860
J . Org. Chem. 2004, 69, 6860-6866

Oxazaborolidine-Catalyzed Asymmetric Reduction
SCHEME 1
have been proven to be effective catalysts in the asym-
metric reduction with excellent yields and enantioselec-
tivities for a wide variety of ketones.2,3,7 However, 1a and
1b are not entirely satisfactory, particularly for large-
scale productions, owing to the necessity of relatively
expensive reagents, such as trimethyl boroxin or meth-
ylboronic acid, for their preparation3 and the necessity
of the preparation before catalytic reactions in order to
get excellent and stable enantioselectivity. Furthermore,
the catalyst 1a is air and moisture sensitive.3 The
catalyst 1d , modified by the catalyst 1b, was developed
and applied successfully in the enantioselective reduc-
tion.7 The advantage of the catalyst 1d is that it can be
prepared from (S)-2-(diphenylhydroxymethyl)pyrrolidine
and inexpensive trimethyl borate in situ, and used
directly without any further separation and purification.7
After the investigation on the effect of the temperature
on the enantioselectivity, it seems that the catalyst 1d
is an effective, convenient, and practical catalyst for the
asymmetric borane reduction of ketones. We wish to
extend the utility of the robust catalyst 1d for the prep-
aration of secondary alcohols with a variety of functional
groups. However, when we reduced the acetophenones
with electron-withdrawing groups in the para position
under our optimal reductive conditions, lower enantiose-
lectivities were obtained. The results prompt us to
consider the influence of the electronic effects of the
substituents in the ketones on the enantioselectivity in
the asymmetric reaction. To investigate the influence,
two series of ketones, para-substituted acetophenones
and propiophenones, were purchased or prepared. The
steric surroundings of the carbonyl groups of the ke-
tones had to be kept as constant as possible to obtain
reliable data concerning the electronic effect. Acetophe-
nones and propiophenones have been electronically modi-
fied simply by variation of the substituent attached to
the phenyl moiety on their para positions. So that
changes in the steric situation at the carbonyl group are
negligible, they were reduced by using the same re-
ductive conditions, our optimal standard conditions, with
the oxazaborolidine 1d as a catalyst first. In our asym-
metric reduction, the catalyst 1d was prepared in situ
through addition of trimethyl borate into a solution of
(S)-2-(diphenylhydroxymethyl)pyrrolidine in toluene. Af-
ter the resulting mixture was stirred for 2 h and bor-
ane was added, a ketone was added dropwise at 25 °C.
After the reaction mixture was quenched with meth-
anol and the usual workup, the enantiomeric excess
value was determined by using a chiral column on HPLC.
The results are presented in Table 1, columns 5 and 6.
Obvious electronic effects on the enantioselectivity were
observed.
the electronic effects of ketones and catalysts on the
enantioselectivity, all of the results indicated that no
obvious influence has been observed in the asymmetric
reduction5a,6a,13 except for the observation of Corey et al.13c
Very recently we investigated the temperature-dependent
enantioselectivity of the asymmetric reduction and found
that the noncatalytic reduction is an important factor for
the enantioselectivity in asymmetric reductions.7c We also
observed obvious influences of electronic effects and
anions on the enantioselectivity in the asymmetric reduc-
tion. Herein, we wish to report our observations and
rationales on these influences.
The factors governing enantioselectivity in the catalytic
asymmetric reaction are usually interpreted in steric
terms,14 affected by the temperature and solvent, etc.6a,7c,8c,9
Electronic effects have been reported to control the
enantioselectivity recently. In reported studies on the
electronic effects on the enantioselectivity,15,16 the un-
derlying reasons are poorly understood in most of the
cases. It should be very useful to understand all factors
that affect the enantioselectivity in this asymmetric
reduction in detail. And it will be helpful to apply the
reduction effectively in the preparation of secondary
alcohols with a variety of functional groups.
Resu lts a n d Discu ssion
Many chiral 1,3,2-oxazaborolidines derived from chiral
vicinal amino alcohols have been prepared and evaluated
in the asymmetric borane reduction of ketones, and some
of the best enantioselectivities have been achieved with
them.2 Among them, (S)-2-substituted 4,4-diphenyl-3,1,2-
oxazaboro[3.3.0]octanes 1, derived from (S)-2-(diphenyl-
hydroxymethyl)pyrrolidine, with several different sub-
stituents (such as H for 1a , Me for 1b, Ph for 1c, and
MeO for 1d , as representatives, Scheme 1) on the B atom,
(11) Douglas, A. W.; Tschaen, D. M.; Reamer, R. A.; Shi, Y. J .
Tetrahedron: Asymmetry 1996, 7, 1303-1308.
(12) Nettles, S. M.; Matos, K.; Burkhardt, E. R.; Rouda, D. R.;
Corella, J . A. J . Org. Chem. 2002, 67, 2970-2976.
(13) (a) Cho, B. T.; Kim, D. J . Tetrahedron: Asymmetry 2001, 12,
2043-2047. (b) Cho, B. T.; Choi, O. K.; Kim, D. J . Tetrahedron:
Asymmetry 2002, 13, 697-703. (c) Corey, E. J .; Helal, C. J . Tetrahedron
Lett. 1995, 36, 9153-9156.
(14) (a) Corey, E. J .; Noe, M. C. J . Am. Chem. Soc. 1993, 115, 12579-
12580. (b) Helal, C. J .; Magriotis, P. A.; Corey, E. J . J . Am. Chem.
Soc. 1996, 118, 10938-10939.
(15) (a) J acobsen, E. N.; Zhang, W.; Guler, M. L. J . Am. Chem. Soc.
1991, 113, 6703-6704. (b) Nishiyama, H.; Yamaguchi, S.; Kondo, M.;
Itoh, K. J . Org. Chem. 1992, 57, 4306-4309. (c) RajanBabu, T. V.;
Ayers, T. A.; Casalnuovo, A. L. J . Am. Chem. Soc. 1994, 116, 4101-
4102. (d) Park, S. B.; Murata, K.; Matsumoto, H.; Nishiyama, H.
Tetrahedron: Asymmetry 1995, 6, 2487-2494. (e) Zhang, H. C.; Xue,
F.; Mak, T. C. W.; Chan, K. S. J . Org. Chem. 1996, 61, 8002-8003. (f)
Wong, H. L.; Tian, Y.; Chan, K. S. Tetrahedron Lett. 2000, 41, 7723-
7726. (g) J ew, S. S.; Yoo, M. S.; J eong, B. S.; Park, I. Y.; Park, H. G.
Org. Lett. 2002, 4, 4245-4248.
(16) (a) Schnyder, A.; Hintermann, L.; Togni, A. Angew. Chem., Int.
Ed. 1995, 34, 931-933. (b) Wiegers, A.; Scharf, H. D. Tetrahedron:
Asymmetry 1996, 7, 2303-2312. (c) Garner, C. M.; Chiang, S.; Nething,
M.; Monestel, R. Tetrahedron Lett. 2002, 43, 8339-8342.
To further observe the generality of the influence of
the electronic effects on the enantioselectivity in the
asymmetric borane reduction, both of these series of
ketones were further reduced with catalyst 1a under the
same conditions. In our asymmetric reduction, catalyst
1a (0.1 equiv to ketone) was prepared first through
addition of borane-dimethyl sulfide complexes (1.5 equiv)
into a solution of (S)-2-(diphenylhydroxymethyl)pyrroli-
dine in toluene or THF. After the resulting mixture was
stirred for 14 h at 45 °C, the mixture was cooled to 25 °C
and to the borane-dimethyl sulfide complexes (1.0 equiv
to ketone) was added a ketone dropwise over 1 h at 25
J . Org. Chem, Vol. 69, No. 20, 2004 6861

Xu et al.
TABLE 1. Oxa za bor olid in e-Ca ta lyzed Asym m etr ic
Bor a n e Red u ction of Keton es
F IGURE 1. Hammett plot of the catalyst 1a -catalyzed borane
reduction of acetophenones in toluene (log([R]/[S]) vs Hammett
constant σp: MeO, -0.27; Me, -0.14; H, 0; F, 0.15; Cl, 0.24;
Br, 0.26; O2N, 0.78, the same as in Figures 2-4.
cat. 1d
cat. 1a
cat. 1a
in PhMe
in PhMe
in THF
yield
R2 (%)a
ee
yield
ee
yield
ee
entry ketone
R1
(%)b
(%)a (%)b (%)a (%)b
1
2
3
4
5
6
2a
2b
2c
2d
2e
2f
Me2N Me 95
MeS Me 98
MeO Me 99
86.5c
99.2d
99.1e
98.7e
98.9f
0.5g
96 99.9
98 99.3
98 98.5
97 99.1
91 90.1
93 89.9h
95 90.4
94 90.1h
93 87.5
95 88.1h
94 73.8
Me
H
Br
Me 97
Me 99
Me 92
93
90.9h
1.2g
7
8
2g
2h
Cl
Me 91
95
95.2h
F
Me 92 -0.8i
94
92.1h
66.7j
64.7c
97.5d
94.2e
93.4e
93.0f
F IGURE 2. Hammett plot of the catalyst 1a -catalyzed borane
reduction of propiophenones in toluene.
9
10
11
12
13
14
15
2i
2j
O2N
Me 96
Me2N Et
MeS Et
MeO Et
95
96
97
99
96
2k
2l
2m
2n
2o
95 94.9
97 93.6
98 91.5
98 88.9
94 81.3
96 80.7h
92 79.5
95 79.4h
95 79.7
96 79.2h
96
98
98
97
95
96.3
91.4
87.7
87.5
65.8
remote electronic effect to the substrate ketones. Thus,
ketones with electron-donating groups in the para-
position of acetophenones and propiophenones afforded
higher enantioselectivity than those bearing electron-
withdrawing groups. To investigate the influence of the
solvent, we chose two widely used solvents, toluene and
THF, to run the asymmetric reduction of propiophenones,
which show a more obvious influence than acetophe-
nones, under the same conditions. The results indicate
that the same tendency of the influence of the electronic
effects on the enantioselectivity was observed. From the
Hammett plots of log ([R]/[S]) versus Hammett constants
σp,19 a straight line for each of the asymmetric reductions
catalyzed by catalysts 1a and 1d was obtained,20 with
R2 ) 0.87 for acetophenones in toluene catalyzed by 1a
(Figure 1), R2 ) 0.93 for propiophenones in toluene
catalyzed by 1a (Figure 2), R2 ) 0.90 for acetophenones
in toluene catalyzed by 1d (Figure 3); and R2 ) 0.93 for
propiophenones in toluene catalyzed by 1d (Figure 4).
The reduction of propiophenones in THF catalyzed by 1a
did not show a good linearity because THF as a Lewis
basic solvent can coordinate with the boron atom in the
catalyst as a competing coordinating reaction in the
reduction system. Its Hammett plot is not shown. On the
basis of our observations (Figures 1-4), we conclude that
the electron-donating substituents in ketones increase
the enantioselectivity and the electron-withdrawing sub-
stituents in ketones decrease the enantioselectivity in the
Me
H
Br
Et
Et
Et
94 -1.7g
95
89.0h
16
17
2p
2q
2r
Cl
Et
Et
Et
94 -0.4g
93
93
94
59.2
47.2
56.7
96
91
93
96
87.8h
1.4i
F
90.5h
18
O2N
52.4j,k 95 63.5
a
b
Isolated yields after the column chromatography. Ee values
were the average values of twice parallel experiments with less
than 1% ee difference and each of the ee values was determined
by HPLC analysis, using AS, OD, or OJ chiral columns (4.6 × 250
mm, Chiralcel) and a mixture of n-hexane and 2-propanol as an
eluent at an eluent rate of 0.5 to 0.8 mL/min at monitoring wave
228 nm. Configuration was assigned according to the rotation
value. In each case a positive rotation was obtained, indicating
that the selectivity was for the (R)-enantiomer in agreement with
reported work (refs 6a, 7d, 21, and 22). c OD column, n-hexane:
d
2-propanol (96:4, v/v). AS column, n-hexane:2-propanol (95:5,
v/v). e OD column, n-hexane:2-propanol (95:5, v/v). f OJ column,
g
n-hexane:2-propanol (90:10, v/v). OD column, n-hexane:2-pro-
panol (97:3, v/v). Minus ee values indicate that the (S)-enantiomer
h
i
predominated. Ee value for AgNO3-washed ketone. OD column,
n-hexane:2-propanol (99:1, v/v). j OD column, n-hexane:2-propanol
k
(90:10, v/v). Configuration was tentatively assumed according to
the mechanism and its rotation sign.
°C. After the reaction mixture was quenched with
methanol and the usual workup, the enantiomeric excess
value was determined by using a chiral column on HPLC.
The results are summarized in Table 1, columns 7 and 8
in toluene, and columns 9 and 10 in THF. From Table 1,
an obvious influence of the electronic effects on the
enantioselectivity was observed for both of these series
of ketones in toluene and in THF. Now a conclusion may
be drawn that the enantioselective 1,3,2-oxazaborolidine-
catalyzed borane reduction of prochiral ketones to chiral
secondary alcohols shows an obvious influence of elec-
tronic effects on the enantioselectivity.
(17) Gregory, W. A.; Iler, R. K.; Schlatter, R. U.S. patent 2802033,
1957; Chem. Abstr. 1958, 52, 1238d.
(18) Firouzabadi, H.; Abbasi, M. Synth. Commun. 1999, 29, 1485.
(19) Isaacs, N. Physical Organic Chemistry, 2nd ed.; Addison-Wesley
Longman: Essex, England, 1995; p 152.
(20) The data for dimethylamino- and methylthio-substituted ke-
tones were not used in Hammett plots because dimethylamino ketones
show strong amine-boron coordination, and methylthio is a seldom
used substituent in the Hammett linear free energy investigation. If
the data of methylthio-substituted ketones were used, R2 will become
lower. Reference 19 gives 0.06 of the σ value for the MeS group.
However, on the basis of our experimental results, the σ value of the
MeS group should be suggested as -0.51, which is an average of the
σ values calculated from Figures 2 and 4 because substituted pro-
piophenones show good linearities in the Hammett plots.
On the basis of our results, the influence of the
electronic effects on the enantioselectivity is obvious. The
enantioselectivity of the reduction is subjected to a
6862 J . Org. Chem., Vol. 69, No. 20, 2004

Oxazaborolidine-Catalyzed Asymmetric Reduction
shows weaker coordination with the boron atom in the
catalysts than that of the ketones with electron-donating
groups. This suggests that larger amounts of the ketones
with electron-withdrawing groups were reduced by non-
catalytic borane to produce racemic alcohols. Thus, their
ee values are lower than those of the ketones with
electron-donating groups. The results should also indicate
that the coordination step in the catalytic cycle is a key
step for the enantioselectivity in the reductions (Scheme
2). This should be an ee-determining step in the reduction
cycle.
F IGURE 3. Hammett plot of the catalyst 1d -catalyzed borane
reduction of acetophenones in toluene.
Although the mechanism of the oxazaborolidine-cat-
alyzed asymmetric borane reduction of ketones is prob-
ably very complicated as shown in Scheme 2 suggested
by Corey et al.,2d most of the experimental results in-
dicated that the catalytic cycle involving steps i-iv is a
major pathway in the reduction3a because high enantio-
selectivity was obtained in most cases, especially for
catalyst oxazaborolidines derived from (S)-2-(diphenyl-
hydroxymethyl)pyrrolidine. And the oxazaborolidine-
catalyzed reduction rate is very fast in most cases.1,2,3a
Thus, the intermediates C and D are very short-life and
low-concentration intermediates in the reduction system.
Intermediate E seems to be a long-life species.25 However,
as an alkoxyborane, its reductive activity is weaker than
that of intermediate A and borane in the reduction
conditions because no dialkoxyborane was observed in
the borane reduction of acetone.25
F IGURE 4. Hammett plot of the catalyst 1d -catalyzed borane
reduction of propiophenones in toluene.
SCHEME 2. P r op osed Mech a n ism for th e
Oxa za bor olid in e-Ca ta lyzed Asym m etr ic Bor a n e
Red u ction of Keton es by Cor ey et a l.2d
The influence of electronic effects on the enantioselec-
tivity in the asymmetric borane reduction of ketones has
been considered several times in the literature.5a,6a,13
However, no obvious influence has been observed when
the B-alkyl- and aryl-substituted oxazaborolidines 1b and
1c were used as catalysts, and different alkyl para-
substituted aryl ketones were used as substrates except
for the report of Corey et al.13c Mathre et al. investigated
the influence of para-substituted acetophenones using
oxazaborolidine 1b as the catalyst under stoichiometric
reduction conditions in most cases.6a J ones, Blacklock,
and their co-workers paid much attention to the influence
using the oxazaborolidine 1b itself and its derivatives
with different para substituents on the phenyl ring of
the diphenylprolinol, the oxazaborolidine 1c itself, and
its derivatives with different para substituents on the
phenyl ring attached to the boron atom in the asymmetric
reduction.5a Unfortunately, no significant electronic effect
was observed with acetophenones with a variety of
substituents and with both the substitution on the phenyl
groups in the diphenylprolinol and that on the boron-
bearing phenyl group. Here, we investigated the influence
using para-substituted acetophenones and propiophe-
nones as substrates and the other two widely used chiral
oxazaborolidines 1a and 1d as catalysts. An obvious
influence was observed for both of these catalysts. Both
of them show a Hammett linear free energy electronic
effect on the enantioselectivity in asymmetric reductions.
p-Dimethylaminoacetophenone 2a and p-dimethylami-
nopropiophenone 2j with an electron-donating group gave
relative lower ee values (86.2% and 64.7%, respectively,
Table 1, entries 1 and 10). The reason is that noncatalytic
reduction is increased because the nitrogen atom in the
amino group shows a strong coordination with borane
asymmetric borane reduction. The conclusion is in ac-
cordance with the observation of Corey et al.13c A similar
electronic effect was also observed in the chiral bisox-
azoline-Cu-catalyzed asymmetric aziridination of chal-
cones, R,â-unsaturated ketones, recently.23 Another im-
portant evidence is that diborane reduces trimethylacetal-
dehyde readily, but fails to react with chloral under the
usual mild conditions.24 The inertness of chloral toward
diborane is attributed to the decreased basic property of
the oxygen atom of the carbonyl group resulting from the
powerful inductive effects of the halogen substituents.
The enantioselectivity is an electronic effect dependent
in the reductions. It was rationalized that the oxyatom
of the carbonyl group in the ketones with electron-
withdrawing groups show weaker basicity than that of
the ketones with electron-donating groups. The oxyatom
(21) Dai, W. M.; Zhu, H. J .; Hao, X. J . Tetrahedron: Asymmetry 2000,
11, 2315-2337.
(22) Vyskocil, S.; J aracz, S.; Smrcina, M.; Sticha, M.; Hanus, V.;
Polasek, M.; Kocovsky, P. J . Org. Chem. 1998, 63, 7727-7737.
(23) Xu, J . X.; Ma, L. G.; J iao, P. Chem. Commun. 2004, 1616-1617.
(24) Brown, H. C.; Schlesinger, H. I.; Burg, A. B. J . Am. Chem. Soc.
1939, 61, 673-680.
(25) Fehlner, T. P. Inorg. Chem. 1972, 11, 252-256.
J . Org. Chem, Vol. 69, No. 20, 2004 6863

Xu et al.
TABLE 2. Oxa za bor olid in e-Ca ta lyzed Asym m etr ic
Bor a n e Red u ction of Acetop h en on es (w ith Differ en t
Ad d itives)
SCHEME 3. Effects of An ion s on th e
En a n tioselctivity in th e
Oxa za bor olid in e-Ca ta lyzed Asym m etr ic Bor a n e
Red u ction
catalyst
ketone
(R)
loading
additive
(equivb)
yield eec
(%) (%)
entry
catalyst (equiva)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2d (H)
2i (ON2)
2i (ON2)
2i (ON2)
2i (ON2)
2d (H)
2d (H)
2d (H)
2d (H)
1d
1d
1d
1d
1d
1a
1a
1a
1a
1a
1a
1d
1d
1a
1ee
1ff
1d
1d
1d
1d
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
LiCl (1.0)d
95 37.2
TBAF (1.0)
TEBA (1.0)
HTMAB (1.0)
TBAI (1.0)
TEBA (0.5)
TEBA (1.0)
TEBA (1.5)
TEBA (2.0)
TEBA (2.5)
96
94
96
94
0.7
0.2
0.8
0.5
and the coordinated borane can reduce the ketone to give
a racemic alcohol. Thus, these two ketones were not used
in further experiments to avoid other effects. p-Methyl-
sulfonyl- and p-methylsulfinyl-substituted acetophenones
and propiophenones were also prepared by oxidation of
methylthioacetophenone 2b and methylthiopropiophe-
none 2k with peroxy acetic acid and freshly made man-
ganese dioxide, respectively.17,18 However, unfortunately,
all of them yielded the corresponding 1-(4-methylthio-
phenyl)ethanol 3b and 1-(4-methylthiophenyl)-1-pro-
panol 3k , respectively, under our reduction conditions.
Thus, both sulfinyl and sulfonyl groups were reduced to
sulfide (data were not shown in Table 1). Although
sulfone-containing ketones were reduced successfully
to chiral sulfone-containing alcohols previously,13a our
ketones which contain a sulfone or sulfoxide group
adjacent to phenyl groups were reduced to chiral sul-
fide-containing alcohols under the current reduction
conditions.
Another interesting observation is that it is obviously
different to reduce halogenated ketones in the reductions.
The catalyst 1a gave enantioselective products, while the
catalyst 1d produced racemic products with commercial
available and self-prepared halogenated ketones. Why did
they generate different results? Considering the struc-
tures of the catalysts, the reason may be caused by the
halide anion coordination to the boron atom in the
catalysts (as shown in Scheme 3). The halogenated
ketones contain some halide anions which can coordinate
with the boron atom in the catalysts. When they coordi-
nate with the boron atom in catalyst 1d , the halide-
coordinated catalyst 1d cannot further coordinate with
the carbonyl group of ketones. Thus, it is an inactive
catalyst. The halogenated ketones were reduced only by
the noncatalytic borane to yield racemic alcohols. How-
ever, for the catalyst 1a , the halide-coordinated catalyst
1a is a chiral reductant, which reduces halogenated
ketones to produce chiral alcohols. Moreover, the halide-
coordinated catalyst 1a can lose a hydrogen anion to form
B-halogenated catalysts, which are also chiral catalysts
in the borane reduction of ketones. This is the reason
these two catalysts show different effects in the presence
of halide anions in the reduction system.
95 98.6
95 98.3
96 95.2
97 51.7
95 23.1
PhCO2Na (1.0)d 96 96.2
PhCO2Na (1.0)d 95 96.9
96 66.7
94 73.8
97 76.6
96 76.9
97 98.0
98 94.4
95 93.5
98 88.9
0.05
0.01
0.005
0.001
a
b
Equiv to ketone. Equiv to the catalyst. c Ee values were
determined by HPLC analysis. Analytic conditions are provided
d
in the footnotes of Table 1. Actually only saturated solution due
to poor solubilities of LiCl and PhCO2Na in toluene. e 1e was
prepared in situ via the exchange reaction of 1a and TEBA as
shown in Scheme 3. f 1f was prepared in situ via the exchange
reaction of 1a and TBAF as shown in Scheme 3.
anion, hexadecyltrimethylammonium bromide (HTMAB)
for bromide anion, and tetrabutylammonium iodide
(TBAI) for iodide anion. The results indicate that halide
anions can decrease the enantioselectivity of the borane
reduction of kenones. For catalyst 1d , addition of an
equivalent of halide anion to the catalyst can completely
inactive catalyst 1d . However, for catalyst 1a , it is
obviously inactive with more than 2 equiv of chloride
anion. Its activity is not affected when the amount of
chloride is less than equivalent to the catalyst 1a . These
results support our rationale. The benzoate ion, a biden-
tate anion, was also evaluated in the asymmetric reduc-
tion of acetophenone, and the results indicate that it just
shows a slight effect due to its very poor solubility in
toluene (Table 2, entries 11 and 12). This implies that
soluble and coordinate-able anions can affect the enan-
tioselectivity in the reduction.
Because the halides can affect the enantioselectivity
in the asymmetric reduction, all of halogenated ketones
were washed twice with aqueous silver nitrate, dried over
anhydrous sodium sulfate, and asymmetrically reduced
again to give chiral alcohols. The results are listed in the
Table 1. It was found that no obviously difference was
observed for the results catalyzed by catalyst 1a before
and after washing with aqueous silver nitrate.
To further verify this rationale, acetophenone was
reduced by using catalysts 1a and 1d in the presence of
different halide anions and different amounts of chloride
anion (Table 2, entries 1-10) with tetrabutylammonium
fluoride (TBAF) for fluoride anion, lithium chloride and
benzyltriethylammonium chloride (TEBA) for chloride
6864 J . Org. Chem., Vol. 69, No. 20, 2004

Oxazaborolidine-Catalyzed Asymmetric Reduction
It was reported previously that the electronic tuning
of asymmetric catalysts can change the stereoselectivity
in catalytic asymmetric reactions.15a,b,d Regulation of
electronic effects of catalysts is another important path-
way to find effective catalysts. Regulation of the electronic
effects of catalysts to match with substrates should be
important especially for asymmetric transformations
where stereoselectivity relies purely on nonbonded in-
teractions. The interaction performs stereochemical com-
munication between a chiral catalyst and a substrate.
The regulation should be a very useful tool to determine
reaction mechanisms, especially to identify the ee-
determining step in the catalytic cycle.
For the p-nitro ketones 2i and 2r , it was found that
they showed higher enantioselectivities with catalysis of
1a than those with catalysis of 1d . It was rationalized
that the difference was caused by the different substit-
uents on the boron atom of catalysts 1a and 1d . The
electron-donating group methoxy attached to the boron
atom in the catalyst of 1d decreases the Lewis acidity of
the boron atom in catalyst 1d so that the boron atom
shows a weaker coordination with the oxyatom of the
carbonyl group in the ketones with an electron-with-
drawing nitro group because the oxyatom of the carbonyl
group is a weaker Lewis base than that in the ketones
with electron-donating groups owing to the existence of
the electron-withdrawing nitro substituent. Larger
amounts of the nitro ketones were reduced by the
noncatalytic borane to produce racemic alcohols under
the catalysis of 1d . Thus, the ee values are lower than
those under the catalysis of 1a .
On the basis of the above analysis, we rationalized that
a catalyst with an electron-withdrawing group, such as
fluoro or chloro substituents, attached to the boron atom
should be a suitable catalyst for the asymmetric reduction
of the ketones with electron-withdrawing groups. The
boron atom in the catalysts with an electron-withdrawing
group attached to the boron atom should show a stronger
Lewis acidity than that in the B-unsubstituted cat-
alyst 1a and that in the catalysts 1b-d with electron-
donating groups attached to the boron atom. It should
show a stronger coordination with the oxyatom of the
carbonyl group in the ketones with electron-withdrawing
groups, such as nitro and halo groups. Thus, if the
ketones with electron-withdrawing groups were reduced
by using the catalyst with an electron-withdrawing
group, their enantioselectivity may be improved. The
B-fluoro- and B-chloro-substituted catalysts 1e and 1f
may be prepared by the exchange reaction with halide
anions in the reduction system as shown in Scheme 3.
First, the B-unsubstituted catalyst 1a was prepared and
then was converted into the B-fluoro- and B-chloro-
substituted catalysts 1e and 1f by in situ exchange
reactions with TBAF and TEBA, respectively, as shown
in Scheme 3. The freshly prepared catalysts 1e and 1f
were tested in the asymmetric reduction of p-nitro-
acetophenone, which gave lower enantioselectivity under
the reductions catalyzed by catalysts 1a and 1d . The
results indicated that the enantioselectivity of p-nitro-
acetophenone was slightly improved (Table 2, entries 15
and 16). The current studies indicate that tuning the
electron effects of a catalyst can improve its enantiose-
lectivity in the asymmetric borane reduction (Table 2,
entries 13-16).
One can find that catalyst 1d always shows excellent
enantioselectivities to the ketones with electron-donating
substituents or without substituent (see Table 1). It was
assumed that the Lewis acidity of the boron atom in
catalyst 1d is decreased because of its bearing an
electron-donating methoxy group. The coordination be-
tween the oxyatom of the carbonyl group of ketones and
the boron atom of catalyst 1d is diminished. The interac-
tion between the oxyatom of alkoxy and the boron atom
of catalyst 1d (Scheme 2, C and D) is also weaker. In
the catalytic reduction cycle, reduced product, alkoxy-
borane E, could be released more easily (Scheme 2, steps
iv and vii). Thus, catalyst 1d undergoes the catalytic
reduction cycle at a faster speed than catalysts 1a -c,
and shows a higher catalytic efficiency for the ketones
with electron-donating groups. To further confirm the
assumption, acetophenone was reduced with the decreas-
ing amount of catalyst loading (Table 2, entries 17-20).
It was found that the catalytic efficiency of catalyst 1d
is obviously higher than that of catalyst 1a .3a Catalyst
1a gave an enantioselectivity of 80% when it was used
in the reduction of acetophenone with 0.5% equiv of
catalyst,3a while catalyst 1d yielded an enantioselectivity
of 88.9% with 0.1% equiv of the catalyst loading in the
same reduction reaction. This supports our assumption
and seems to indicate that the releasing steps (Scheme
2, steps iv and vii) are rate-determining steps in the
catalytic cycle.
The present investigation indicates that the electronic
tuning of catalysts 1 can improve their enantioselectivity
and efficiency in the asymmetric borane reduction and
the electronic tuning of catalysts and ketones can identify
that the ketone and catalyst coordinating step (Scheme
2, step ii) is a key step for the enantioselectivity in the
catalytic cycle.
On the basis of investigation, one can select a suitable
catalyst for the borane asymmetric reduction of ketones.
We hope that the present study has addressed an
important issue: the regulation of electronic effects of
catalysts to match with substrates should be important
in asymmetric catalytic reactions. It should be an im-
portant way to improve the enantioselectivity for some
substrates. We hope that the present study has also
addressed an important issue regarding the catalyst
selection in the oxazaborolidine-catalyzed asymmetric
reduction of ketones. We recommend oxazaborolidine 1d
as the most suitable catalyst for ketones with electron-
donating groups because of its convenient preparation,
economic starting materials, high efficiency, and excellent
yields and enantioselectivity, and the oxazaborolidines
1b and 1c as good choices for all ketones with either
electron-donating groups or electron-withdrawing groups,
but especially for ketones with electron-withdrawing
groups. Although oxazaborolidine 1d as the most suitable
catalyst is limited to ketones with electron-donating
groups, it is convenient and practical in the preparation
and application, especially for large-scale preparation in
industry due to low catalyst loading in the asymmetric
reduction.
Con clu sion
The influence of the electronic effects on the enantio-
selectivity of the oxazaborolidine-catalyzed asymmetric
J . Org. Chem, Vol. 69, No. 20, 2004 6865

Xu et al.
hydroxymethyl)pyrrolidine (12.5 mg, 0.05 mmol) in dry toluene
(2.5 mL) was added 2 M borane-dimethyl sulfide complexes
in THF (0.038 mL, 0.075 mmol), and the mixture was stirred
under nitrogen at 45 °C for 14 h. After the mixture was cooled
to 25 °C, 2 M borane-dimethyl sulfide complex in THF (0.25
mL, 0.5 mmol) was added, then a solution of ketone (0.5 mmol)
in dry toluene (2.5 mL) was added dropwise over 1 h. The
mixture was stirred at 25 °C until the ketone disappeared on
gas chromatographic monitoring. The resulting mixture was
quenched with methanol in an ice bath and concentrated under
reduced pressure. The residue was purified on a silica gel
column with a mixture of petroleum ether (60-90 °C) and
ethyl acetate (5:1, v/v) as an eluent to give the chiral secondary
alcohol as a colorless oil.
borane reduction of ketones has been observed and
investigated with use of para-substituted acetophenones
and propiophenones with various functional groups and
B-unsubstituted and B-methoxyoxazaborolidines derived
from (S)-diphenylprolinol with borane and trimethyl
borate as catalysts in toluene and THF. The results
indicate that they show Hammett linear free energy
electronic effects on the enantioselectivity in the asym-
metric reduction. Tuning the electronic effect of the
catalyst can extend its utility effectively and improve the
enantioselectivity for some substrates in the reduction.
Another interesting finding is that soluble and coordinate-
able anions in the reductive system heavily affect the
enantioselectivity, especially for the B-methoxy catalyst,
because of their coordination with the boron atom in the
catalysts.
(R)-1-(4-Nitr op h en yl)-1-p r op a n ol (3r ). Colorless liquid;
[R]25 +20.4 (c 1.20, CHCl3), ee 63.3%; 1H NMR (300 MHz,
D
CDCl3) δ 0.93 (t, J ) 7.2 Hz, 3H, CH3), 1.78 (dq, J ) 6.4, 7.2
Hz, 2H, CH2), 2.80 (s, br, 1H, OH), 4.74 (t, J ) 6.4 Hz, 1H,
CH), 7.51 (d, J ) 9.0 Hz, 2H, ArH), 8.16 (d, J ) 9.0 Hz, 2H,
ArH); 13C NMR (75.5 MHz, CDCl3) δ 9.6, 31.9, 74.6, 123.4,
126.6, 146.9, 152.1; MS (EI) m/z:181 (M+, 1.2), 163 (M+ - H2O,
1.4), 152 (M+ - Et, 100), 122 (O2NPh+, 28); IR v (cm-1) 3220
(OH), 2966, 2928, 1604, 1519, 1346. Anal. Calcd for C9H11NO3
(181.19): C, 59.66; H, 6.12; N, 7.73. Found: C, 59.53; H, 6.32;
N, 7.57.
Exp er im en ta l Section
Gen er a l P r oced u r e for th e Asym m etr ic Red u ction of
Keton es w ith Ca ta lyst 1d . To a solution of (S)-2-(diphenyl-
hydroxymethyl)pyrrolidine (12.5 mg, 0.05 mmol) in dry toluene
(2.5 mL) was added trimethyl borate (6.0 mg, 0.06 mmol), and
the mixture was stirred under a nitrogen atmosphere at room
temperature for 2 h. After 2 M borane-dimethyl sulfide
complex in THF (0.25 mL, 0.5 mmol) was added, a solution of
ketone (0.5 mmol) in dry toluene (2.5 mL) was added dropwise
over 1 h. The mixture was stirred at 25 °C until the ketone
disappeared by gas chromatographic monitoring. The resulting
mixture was quenched with methanol in an ice bath and
concentrated under reduced pressure. The residue was puri-
fied on a silica gel column with a mixture of petroleum ether
(60-90 °C) and ethyl acetate (5:1, v/v) as an eluent to give a
colorless oil chiral secondary alcohol. The spectral and analyti-
cal data of all obtained alcohols are in agreement with those
reported in the literature.6a,7d,19,20
Ack n ow led gm en t. This work was supported in part
by the National Natural Science Foundation of China
(Project No. 20272002), Ministry of Education of China
(SRF for ROCS and EYTP), and Peking University
(present grant).
Su p p or tin g In for m a tion Ava ila ble: The chromatogram
for the determination of enantiomeric excess of the unknown
1
chiral alcohol 3r and its H NMR and 13C NMR spectra. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Gen er a l P r oced u r e for th e Asym m etr ic Red u ction of
Keton es w ith Ca ta lyst 1a . To a solution of (S)-2-(diphenyl-
J O048959I
6866 J . Org. Chem., Vol. 69, No. 20, 2004