Highly enantioselective deracemization of linear and vaulted biaryl ligands.

Article abstract of DOI:10.1021/ol0275769

[reaction: see text] A copper-mediated deracemization of the vaulted biaryl ligands VANOL and VAPOL can be readily achieved in the presence of (-)-spartiene. The optimal procedure involves the in situ generation of copper(II) and leads to the reproducible formation of (S)-VANOL and (S)-VAPOL in greater than 99% ee from the racemates. This method is superior to existing procedures for BINOL (92% ee).

Full text of DOI:10.1021/ol0275769

ORGANIC
LETTERS
2
003
Vol. 5, No. 11
813-1816
Highly Enantioselective Deracemization
of Linear and Vaulted Biaryl Ligands
1
Yu Zhang, Siu-Man Yeung, Hongqiao Wu, Douglas P. Heller, Chunrui Wu, and
William D. Wulff*
Department of Chemistry, Michigan State UniVersity, East Lansing, Michigan 48824
wulff@cem.msu.edu
Received December 31, 2002
ABSTRACT
A copper-mediated deracemization of the vaulted biaryl ligands VANOL and VAPOL can be readily achieved in the presence of (−)-spartiene.
The optimal procedure involves the in situ generation of copper(II) and leads to the reproducible formation of (S)-VANOL and (S)-VAPOL in
greater than 99% ee from the racemates. This method is superior to existing procedures for BINOL (92% ee).
The development of highly efficient catalytic asymmetric
reactions has been a major focus in modern synthetic organic
chemistry and the design and synthesis of effective ligands
by crystallization. Adaptation of this protocol to the resolu-
tion of VANOL and VAPOL was successful provided that
(-)-brucine was used in salt formation with VANOL and
1
2
necessarily plays an important role in this endeavor. The
(-)-cinchonidine was used with VAPOL. The major
C
3
2
-symmetric vaulted biaryl ligands VANOL 2 and VAPOL
developed in our laboratories have proved to be excellent
drawback of this resolution protocol lies with the fact that it
is a four-step sequence.
2
ligands for several important catalytic asymmetric reactions
Subsequently, alternative methods for obtaining optically
pure BINOL have been developed, including the optical
resolution of diastereomeric salts prepared directly from
3
4
such as Diels-Alder reactions, imino aldol reactions, and
5
aziridination reactions. The resolution procedures that are
7
8
published for these two ligands follow a four-step procedure
developed for BINOL that involves the bis-ester of phos-
BINOL, asymmetric oxidative coupling of 2-naphthol, and
the copper-mediated deracemization of BINOL in the pres-
6
9
phoric acid. Salt formation of the BINOL bis-ester with (+)-
ence of a chiral amine or diamine. As an example of the
cinchonine leads to diastereomers which can be separated
latter, the deracemization of BINOL with cupric chloride in
(
1) (a) Noyori, R. Asymmetric Catalysis in Organic Synthesis; Wiley:
(6) (a) Jaques, J.; Fouquey, C. Org. Synth. 1988, 67, 1. (b) Truesdale,
L. Org. Synth. 1988, 67, 13.
New York, 1994. (b) Ojima, I., Ed. Catalytic Asymmetric Synthesis, 2nd
ed.; VCH: New York, 2000.
(7) (a) Kawashima, M.; Hirayama, A. Chem. Lett. 1990, 2299. (b)
Tannaka, K.; Okada, T.; Toda, F. Angew. Chem., Int. Ed. Engl. 1993, 32,
1147. (c) Cai, D.; Hughes, D. L.; Verhoeven, T. R.; Reider, P. J. Tetrahedron
Lett. 1995, 7991. (d) Toda, F.; Tanaka, K. J. Chem. Soc., Chem. Commun.
1997, 1087.
(8) (a) Hamada, T.; Ishida, H.; Usui, S.; Watanabe, Y.; Tsumura, K.;
Ohkubo, K. J. Chem. Soc., Chem. Commun. 1993, 909. (b) Smrcina, M.;
Polakova, J.; Vyskocil, S.; Kocovsky, P. J. Org. Chem. 1993, 58, 4534. (c)
Nakajima, M.; Miyoshi, I.; Kanayama, K.; Hashimoto, S. J. Org. Chem.
1999, 64, 2264. (d) Li, X.; Yang, J.; Kozlowski, M. C. Org. Lett. 2001, 3,
1137. (e) Hon, S.-W.; Li, C.-H.; Kuo, J.-H.; Barhate, N. B.; Liu, Y.-H.;
Wang, Y.; Chen, C.-T. Org. Lett. 2001, 3, 869. (f) Luo, Z.; Liu, Q.; Gong,
L.; Cui, X.; Mi, A.; Jiang, Y. Angew. Chem., Int. Ed. 2002, 41, 4532.
(2) Bao, J.; Wulff, W. D.; Dominy, J. D.; Fumo, M. J.; Grant, E. B.;
Rob, A. C.; Whitcomb, M. C.; Yeung, S.; Ostrander, R. L.; Rheingold, A.
J. Am. Chem. Soc. 1996, 118, 3392.
(
3) (a) Bao, J.; Wulff, W. D.; Rheingold, A. L. J. Am. Chem. Soc. 1993,
15, 3814. (b) Bao, J.; Wulff, W. D. Tetrahedron Lett. 1995, 36, 3321. (c)
Heller, D. P.; Goldberg, D. R.; Wulff, W. D. J. Am. Chem. Soc. 1997, 119,
1
1
0551.
(4) Xue, S.; Yu, S.; Deng, Y.-H.; Wulff, W. D. Angew. Chem., Int. Ed.
2
001, 40, 2271.
(
5) (a) Antilla, J. C.; Wulff, W. D. J. Am. Chem. Soc. 1999, 121, 5099.
b) Antilla, J.; Wulff, W. D. Angew. Chem., Int. Ed. 2000, 39, 4518. (c)
Loncaric, C.; Wulff, W. D. Org. Lett. 2001, 3, 3675.
(
1
0.1021/ol0275769 CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/03/2003

a
Scheme 1
Table 1. Decacemization with CuCl ‚2H O
2
2
scale
(g)
solvent
time yieldb
eec
entry biaryl
MeOH/CH2Cl2
(h)
(%)
(%)
1d
2
3
4e
5
6
7
8
9
0
11
12f
3g
4h
5
3
3
3
3
3
3
3
4
2
2
2
2
2
2
2
0.1
0.1
0.1
0.1
0.1
0.1
1.0
0.2
0.6
0.6
0.6
0.6
0.6
0.6
1.0
MeOH only
1:2
1:2
1:2
1:2
1:3.4
1:3.4
2:1
2:1
2:1
24
20
2
2
1
24
24
24
20
46
50
50
50
50
50
62
59
54
73
82
64
<10
50
71
42
77
66
55
61
59
71
75
49
25
34
99
>98
90
98
99
90
93
93
93
1
the presence of (-)-sparteine has been reported to give the
2:1
2:1
2:1
2:1
9
b
S-enantiomer of BINOL in 94% yield and in 80% ee. The
subject of the present work is our studies on the adapation
of this deracemization process to the VANOL and VAPOL
ligands. It was found that this process works exceptionally
well for both ligands with proper modifications of the
experimental procedure and, furthermore, that this new
procedure is also superior for the deracemization of BINOL.
The results of the direct application of the deracemization
procedure reported for BINOL to the deracemization of
VAPOL are shown in entry 1 of Table 1. Neither the yield
nor the asymmetric induction is as high as that reported for
BINOL. Given the lower solubility of racemic VAPOL in
methanol, the reaction was repeated with a 1:2 mixture of
methanol and methylene chloride as solvent, which provides
a homogeneous solution throughout the reaction. Nonethe-
less, the outcome is essentially unchanged (entry 2). All other
reactions in this work are also homogeneous. Decreasing the
temperature of the reaction (entry 4) or decreasing the
reaction time (entry 5) led to increased recovery, and the
asymmetric induction in both cases dramatically decreased.
In many cases the low yields are accompanied by the
formation of yellow side-products that have not yet been
characterized. It was shown that the quinone 5 is not
produced in these reactions by oxidation with copper(II). This
quinone can be obtained in 30-40% yield by the oxidation
of VAPOL with cerric ammonium nitrate. Optimal results
were found when the reaction was performed with 1.4 equiv
of copper chloride and 2.8 equiv of (-)-sparteine, which gave
a 64% yield of (S)-VAPOL with 99% ee (entry 7).
Unfortunately, this success was not always reproducible as
indicated in entry 7 when an attempt was made to scale this
reaction to 1.0 g, which resulted in less than 10% recovery
1
1
1
2:1
a
Unless otherwise specified, the reaction was performed with 1.4 equiv
of CuCl2‚2H2O and 2.8 equiv of (-)-sparteine according to the general
procedure (see Supporting Information). Entries 1-5 used 1.0 equiv of
CuCl2‚2H2O and 2.0 equiv of (-)-sparteine. The concentration was 15 mM
for 3 and 7.6 mM for 2. Entries 1-10 were deoxygenated by Argon purge,
b
entries 11-15 were deoxygenated by the freeze-thaw method. Isolated
yield by chromatography. c Determined by HPLC on a Pirkle D-phenyl-
d
glycine column. The S-enantiomer was obtained in all cases. The VAPOL
was recovered in 35% yield from a precipiate in 48% ee and in 27% yield
e
f
from solution in 99% ee. Reaction at 0 °C. Yield is average of three
runs. g Reaction at 40 °C. h Concentration of 2 was 15 mM.
of the VAPOL. This procedure could be successfully
extended to the VAPOL derivative 4.
The deracemization of VANOL was performed under the
conditions found optimal for VAPOL and the results are
shown in Table 1, entries 9-15. After 20 h this produced a
1% recovery of VANOL that was 90% ee (entry 9).
Monitoring of the time course of the reaction revealed that
the deracemization is fast, giving 77% ee after the first hour
1
0
7
(
conditions as in entry 9). The optical purity could be
increased to 98% ee if the reaction was allowed to continue
for 46 h; however, the yield dropped to 42%. To probe the
sensitivity of the reaction to traces of oxygen, these condi-
tions were reinvestigated with reaction mixtures that were
deoxygenated by the freeze-thaw method (entries 11-15).11
We were delighted to find that the recovery was greatly
improved (entries 10 vs 11) and the induction was still high
99% ee). However, the inductions were still variable as
indicated by the fact that the reaction in entry 11 was repeated
three times and the average of those three runs (entry 12)
(
(
9) (a) Brussee, J.; Groenendijk, J. L. G.; te Koppele, J. M.; Jansen, A.
C. A. Tetrahedron 1985, 41, 3313. (b) Smrcina, M.; Lorenc, M.; Hanus,
V.; Sedmera, P.; Kocovsky, P. J. Org. Chem. 1992, 57, 1917. (c) Smrcina,
M.; Polakova, J.; Vyskocil, S.; Kocovsky, P. J. Org. Chem. 1993, 58, 4534.
(10) The preparation of 4 will be reported separately.
(11) If the deracemization of VANOL was performed in air under the
conditions in Table 1, no VANOL could be recovered.
1814
Org. Lett., Vol. 5, No. 11, 2003

was 90% ee. Thus, although the use of the freeze-thaw
deoxygention technique generally gave better results, the
procedure is time consuming, the results are variable, and it
would be impractical on a larger scale.
solution, presumably a copper(II) species (Vide infra). If this
solution is added to a solution of racemic VAPOL and
allowed to react for 5 h the VAPOL is recovered in 75%
yield and 95% ee. This is a much faster reaction than was
It has been suggested, but not demonstrated, that (R)-
BINOL could be produced from a copper(I) complex with
2
seen for the deracemization with CuCl . In addition to the
increased rate, this improved procedure is more reproducible
and can be scaled up (entries 13 and 14).
9
c
(-)-sparteine. Since (+)-sparteine is not available, this
suggestion takes on great magnitude since it could provide
for a direct route from racemic bis-phenols to either desired
enantiotropic atropisomer. This possibility was tested on
VAPOL and conducted with the aid of ultrasound due to
the low solubility of cuprous chloride. The reaction was
performed by deoxygenating a slurry of 1.67 equiv of CuCl,
The deracemization of VANOL with in situ generated
copper(II) was found to be approximately twenty times faster
than with commercial copper(II) chloride. Beginning with
copper(II) chloride, the deracemization of VANOL occurs
in 50 h to give a 59% recovery of material that is 93% ee
(Table 1, entry 15). In contrast, beginning with copper(I)
chloride, the deracemization of VANOL under the same
conditions gives a 61% recovery of 99% ee material in only
2.5 h. It was also found that the deracemization of VANOL
with CuCl ‚2H O under the aegis of ultrasound was also
slow, requiring 24 h to go to completion (52% yield, 99%
ee). A beneficial consequence of the shorter reaction time is
that deoxygenation by the freeze-thaw method is not
necessary to ensure good mass recoveries. All of the reactions
in Table 2 are performed by a simple purging of the solution
with argon prior to reaction. Importantly, it was found that
the scale could be increased from 1 g to 5 g with no loss in
recovery or asymmetric induction.
8d
3.3 equiv of (-)-sparteine, and 1.0 equiv of racemic VAPOL
in a 1:4 mixture of methanol and methylene chloride by the
freeze-thaw technique and then placing the mixture in an
ultrasound bath for 10 min and allowing the homogeneous
faint yellow reaction mixture to stir for 6 h in the absence
of ultrasound. The VAPOL was recovered in 93% and was
found still to be racemic (entry 9). The same reaction with
CuCN or CuI also led to racemic VAPOL.
When the sonification of CuCl and (-)-sparteine is
conducted in the presence of air, there is a rapid dissolution
of CuCl along with concomitant formation of a green
2
2
The deracemization of BINOL was also examined with
the in situ generated copper(II) procedure. Employing
conditions previously reported with copper(II) chloride, the
in situ procedure gave a 77% recovery of BINOL with 93.5%
ee. Other changes in the conditions were examined as
outlined in Table 3 which lead to higher yields, but the
Table 2. Deracemization of Vaulted Biaryls with CuCla
9b
Table 3. Deracemization of BINOL with CuCla
scale
(g)
solvent
time
(h)
yieldb
(%)
eec
entry biaryl
MeOH/CH2Cl2
(%)
1
2
3
4
5
6
7
8d,e
9f
0e
1g
2
3h,i
4h,j
2
2
2
2
2
2
2
2
3
3
3
3
3
3
0.3
1.0
1.0
1.0
1.0
2.0
5.0
4.0
1.0
1.0
1.0
2.0
5.0
5.0
1:4
2:1
1:4
1:9
1.75
2.5
2.75
2.5
48
2
2
2
6
5
79
61
72
48
49
70
69
54
93
75
72
74
60
84
99
99
99
98
84
99
99
99
0
95
72
98
95
99
concn
scale
(g)
solvent
time yieldb
eec
entry (mM)
MeOH/CH2Cl2
(h)
20
8.5
24
(%)
(%)
CH2Cl2 only
1:4
1:4
1:4
1:4
1:4
1:4
1:4
1:4
1:4
1
2
3
4
33
7.6
7.6
15
15
30
30
0.286
0.2
0.2
0.2
0.2
MeOH only
1:4
1:2.5
1:2.5
1:2.5
77
94
83
93
96
97
98
93.5
92
92
92
92
24
24
24
24
1
1
1
1
1
5d
6
4
48
6
6d
0.2
5.0
1:2.5
1:2.5
92
87
7d
a
Unless otherwise specified, the reaction was performed with 1.4 equiv
of CuCl and 2.8 equiv of (-)-sparteine according to the general procedure
b
a
(see Supporting Information). Isolated yield by flash chromatography.
Unless otherwise specified, the reaction was performed with 1.4 equiv
c
Determined by HPLC on a Chirapak AS column. d Reaction run at 25 °C
of CuCl and 2.8 equiv of (-)-sparteine according to the general procedure
see Supporting Information). The concentration was 6.2 mM for 3 and 7.6
mM for 2. Isolated yield by flash chromatography. Determined by HPLC
on a Pirkle D-phenylglycine column. Concentration of 2 is 15 mM. Yield
is average of two runs. See text for procedure. Concentration of 3 is 24
mM. 1.67 equiv of CuCl and 3.3 equiv of (-)-sparteine was used. The
product was purified by recrystallization with CH2Cl2/hexane. The starting
for 8 h and then -20 °C for 16 h and then quenched at -20 °C.
(
b
c
d
e
f
g
h
i
asymmetric inductions appear to be limited to 92-93% ee.
The effect of temperature on the induction was also probed
with no change in the equilibrium observed for a reaction
carried out at -20 °C for 16 h (entry 5). A higher recovery
j
material used in entries 11-13 was purified by crystallization from ethyl
acetate. In entry 14 the starting material was crystallized twice.
Org. Lett., Vol. 5, No. 11, 2003
1815

is observed for those reactions that are quenched at -20 °C
entries 5-7). Nonetheless, the induction by this new
(
procedure is significantly improved over that previously
reported for BINOL (Scheme 1).
The reaction of copper(I) salts with diamines in the
presence of oxygen is a well-known process that gives υ-oxo
complexes that in most cases are dimers or trimers of Cu-
(
(
diamine)X(OH).12 The reaction of CuCl with 1 equiv of
-)-spartiene in methanol in the presence of air gave a 63%
yield of a green solid that were assumed to be Cu(spartiene)-
1
2c
Cl(OH). This material was not as effective in the derace-
mization of VAPOL as the in situ generated complex. It gave
only a 47% recovery of VAPOL with 70% ee after 24 h in
a 1:4 mixture of CH
2 2
Cl /MeOH. This could be improved to
a 57% recovery with 97% ee in the presence of 1.2 equiv of
(-)-spartiene.
The preferential formation of the S-enantiomer is observed
in the deracemization of BINOL, VANOL, and VAPOL. To
probe the source of this selectivity we have performed PM3-
Figure 1.
The deracemization procedure described above involving
(
(
(
(
tm) calculations with the Spartan program on the copper-
II) complexes of (-)-spartiene and R- and S-VANOL
Figure 1). As has been observed experimentally for copper-
II) complexes of (-)-spartiene,12c both complexes are
seriously distorted from the normal square-planar geometry.
The angle between the N-Cu-N plane and the O-Cu-O
planes is 85.5° for the R-complex and only 40.6° for the
S-complex, revealing that the R-complex is much more
distorted from planarity. The S-complex was found to be
an in situ generated copper(II) species is superior to previous
methods for the deracemization of BINOL and for the first
time provides for a method for the deracemization of the
VAPOL and VANOL ligands. Since (+)-spariene is not
available this method is limited to preparation of the
S-enantiomers. Recently a surrogate for (+)-spatiene has been
13
introduced and we plan to try this for the deracemization
of these ligands to their R-enantiomers.
3.91 kcal/mol more stable than the R-complex, which is more
Acknowledgment. This work was supported by the
than enough to account for the selectivities seen (KS/R ) 730).
We imagine that the deracemization could involve keto forms
of the naphthol and phenanthrol units as has been suggested
previously for BINOL-type ligands.1
National Institutes of Health Grant GM 63019.
Supporting Information Available: Experimental pro-
cedures and data for new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
2d
(
12) (a) Noji, M.; Nakajima, M.; Koga, K. Tetrahedron Lett. 1994, 35,
OL0275769
7
983. (b) Cole, A. P.; Root, D. E.; Mukherjee, P.; Solomon, E. I.; Stack, T.
D. P. Science 1996, 273, 1848. (c) Funahashi, Y.; Nakaya, K.; Hirota, S.;
Yamauchi, O. Chem. Lett. 2000, 1172. (d) Kozlowski, M. C.; Li, X.; Carroll,
P. J.; Xu, Z. Organometallics 2002, 21, 4513.
(13) Dearden, M. J.; Firkin, C. R.; Hermet, J.-P.;, R.; O’Brien, P. J. Am.
Chem. Soc. 2002, 124, 11870.
1816
Org. Lett., Vol. 5, No. 11, 2003