Enantioselective Strecker-type reaction promoted by polymer-supported bifunctional catalyst

Article abstract of DOI:10.1016/S0040-4039(00)01923-7

The JandaJEL-supported bifunctional catalyst 3 (10 mol%) promoted the Strecker-type reaction of aromatic and α,β-unsaturated imines in excellent yields with 83-87% ee in the presence of 'BuOH (110 mol%). The reactivity of 3 was comparable to the homogeneous analogue 1, and 3 could be recycled at least four times.

Full text of DOI:10.1016/S0040-4039(00)01923-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 279–283
Enantioselective Strecker-type reaction promoted by
polymer-supported bifunctional catalyst
Hiroyuki Nogami, Shigeki Matsunaga, Motomu Kanai and Masakatsu Shibasaki*
Graduate School of Pharmaceutical Sciences, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
Received 10 October 2000; accepted 27 October 2000
Abstract—The JandaJEL™-supported bifunctional catalyst 3 (10 mol%) promoted the Strecker-type reaction of aromatic and
t
a,b-unsaturated imines in excellent yields with 83–87% ee in the presence of BuOH (110 mol%). The reactivity of 3 was
comparable to the homogeneous analogue 1, and 3 could be recycled at least four times. © 2000 Elsevier Science Ltd. All rights
reserved.
The importance of immobilized asymmetric catalysts is
rapidly growing in view of easy separation of the
product and reusability of the catalyst.1 Solid-sup-
ported catalysts show considerable advantages over ho-
mogeneous ones, especially in large-scale syntheses and
high throughput organic chemistry. We have recently
reported an enantioselective Strecker-type reaction pro-
moted by the homogeneous bifunctional catalyst 1.2,3
The products were obtained with generally high ee
(70–95% ee) from a relatively wide range of substrates,
such as aromatic and aliphatic imines including a,b-un-
saturated imines. The origin of the high enantioselectiv-
ity and substrate generality is considered to stem from
the dual activation of the imine and TMSCN at the
defined positions by the Lewis acid (Al) and the Lewis
base (phosphine oxide) of the catalyst.4 Considering the
importance of catalytic enantioselective Strecker-type
reaction for providing various chiral amino acid precur-
sors in both laboratory and industrial scales, it is highly
desirable that the catalyst could be immobilized on a
solid phase and recycled many times.5 In this paper, we
disclose our initial success in this direction.
Since the high enantio-differentiation by the bifunc-
tional catalyst depends on the simultaneous activation
of substrate and reagent, it seemed important that the
polymer support should not affect the balance of the
activation ability of the Lewis acid and the Lewis
base. On the basis of preceding reports concerning
polymer-supported binaphthyl ligands,6 we designed
the polymer-supported catalysts 2 and 3 possessing a
sufficiently long spacer at the 6-position to avoid an
adverse effect of the spacer on the asymmetric envi-
ronment. Since the catalyst contains a Lewis acid
metal, a non-coordinating alkenyl linker was selected.
The syntheses of 2 and 3 are shown in Scheme 1.
After regioselective Friedel–Crafts acylation,6a the at-
tachment of the chiral ligand to the polymer was
achieved by Wittig reaction.7 The purity of the final
polymer-supported ligand was checked by 31P swol-
len-resin magic angle spinning (SR-MAS) NMR.8 The
loading of the ligand on polymers was determined
based on mass balance (1.02 mmol/g for 2 and 0.52
mmol/g for 3).
Keywords: polymer-supported; bifunctional asymmetric catalyst; Lewis acid; Lewis base; Strecker-type reaction.
* Corresponding author.
0040-4039/01/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)01923-7

280
H. Nogami et al. / Tetrahedron Letters 42 (2001) 279–283
Scheme 1. Reagents and conditions: (a) MeI, K2CO3, acetone, 92%; (b) ethyl 4-chloro-4-oxobutyrate, AlCl3, CH2Cl2, 80%; (c)
Pd/C, H2, CH3SO3H, AcOH/AcOEt/EtOH, 68%; (d) BBr3, CH2Cl2, 64%; (e) MOMCl, iPr2NEt, CH2Cl2; (f) LAH, THF, 76% (two
steps); (g) TBSCl, imidazole, DMF, 91%; (h) 1) BuLi (5 equiv.), Et2O, 2) DMF (6 equiv.), 61%; (i) NaBH4, MeOH, 98%; (j) 1)
MsCl, Et3N, CH2Cl2, 2) LiCl, DMF, 86%; (k) Ph2P(O)H (3 equiv.), NaOtBu (3.3 equiv.), THF, 100%; (l) TBAF, THF, 89%; (m)
(COCl)2, DMSO, Et3N, CH2Cl2, 94%; (n) 1) 12 (1 equiv.), KHMDS (5 equiv.), THF–toluene, rt, 4 h, 2) wash with THF, 3) 11
(1.5 equiv.), THF, −78°C to rt, 15 h, 4) acetaldehyde (capping); (o) TsOH·H2O, CH2Cl2–MeOH, 40°C.
polymer-supported 2 might be explained as follows. The
protic additive should partly react with TMSCN to
generate HCN.10 In the case of the reaction promoted
by the homogeneous catalyst 1, the reactivity of HCN
toward the imines was considerably lower than that of
TMSCN, since only TMSCN could be activated by the
Lewis basic phosphine oxide of 1.2 However, in the case
of the polymer-supported catalyst 2, the reactivity dif-
ference between HCN and TMSCN is not so significant
any more, possibly due to the hindered accessibility of
TMSCN to the phosphine oxide by the bulky polymer
core. We have confirmed that the reaction using HCN
proceeded with the similar rate as TMSCN in the
presence of 2. Compound 5a was obtained in 41% yield
in 30 h with 0% ee by HCN (versus 50% yield with
TMSCN-ROH, entries 4, 5). Therefore, the difference
of the ee of the product, depending on the proton
source, could possibly stem from the amount of HCN
in the reaction mixture. The more acidic PhOH should
Using the homogeneous catalyst 1, we have found that
addition of a proton source was necessary to enhance
the reaction rate.2 The best proton source was PhOH,
although other alcohols gave only slightly lower enan-
tioselectivity. Our rationale for the additive effect is that
it should protonate the anionic nitrogen generated by
the addition of the cyanide to the imine, thus facilitating
the ligand exchange on the catalyst. Since PhOH should
be partially regenerated from TMSOPh and the product
amine, only a catalytic amount (20 mol%) of PhOH was
sufficient to give the products with the same ee as with
110 mol% of PhOH. So, we first investigated the effect
of the protic additives, using the Merrifield resin-sup-
ported catalyst 2. As shown in Table 1, applying the
best reaction conditions to benzaldehyde imine 4a, us-
ing 2 (10 mol%), the product 5a was obtained in 83%
yield with 65% ee (entry 2). However, when a stoichio-
metric amount of PhOH (110 mol%) was used, the ee
became significantly lower (43%, entry 3). Another
different tendency from the homogeneous catalyst 1 was
observed when an aliphatic alcohol was used (entries
4–6). The ee was improved up to 78% using 110 mol%
t
generate a higher amount of HCN than BuOH. As a
result, the competitive racemic pathway by HCN should
become more problematic, especially in the case when a
stoichiometric amount of PhOH was used. From these
results, we expected that, if the phosphine oxide of the
catalyst becomes more accessible to TMSCN, the
t
of BuOH as a proton source (entry 6).9 These different
tendencies between the homogeneous catalyst 1 and
Table 1. Optimization of the reaction conditions using benzaldehyde imine 4a
Entry
Catalyst
Catalyst synthesis
Proton source (mol%)
Reaction
Temp. (°C)
Time (h)
Yield (%)
Ee (%)
1
2
3
4
5
6
7
8
9
1
2
2
2
2
2
2
2
2
3
–
PhOH (20)
PhOH (20)
–
−40
−40
−40
−40
−40
−40
−40
−40
−40
−50
44
43
43
30
30
64
16
18
86
60
92
83
91
50
50
86
16
18
86
98
95
65
43
71
71
78
55
16
15
87
Stir
Stir
Stir
Stir
Stir
Stir
Shake
Shake
Stir
Stir
Stir
Stir
Stir
Stir
Shake
Stir
Shake
Stir
PhOH (110)
MeOH (110)
iPrOH (110)
tBuOH (110)
tBuOH (110)
tBuOH (110)
tBuOH (110)
tBuOH (110)
10

Table 2. Catalytic enantioselective Strecker-type reaction of various iminesa
/
3
2
1b
Entry
Imine (R)
Product
Time (h)
Yield (%)
Ee (%)
Time (h)
Yield (%)
Ee (%)
Time (h)
Yield (%)
Ee (%)
1
2
3
4
5
6
4a (Ph)
4b (p-MePh)
4c (p-ClPh)
4d (p-MeOPh)
4e (3-Furyl)
4f ((E)-PhCHꢀCH)
5a
5b
5c
5d
5e
5f
60
64
59
41
66
66
98
100
98
98
97
96
87
83
85
83
86
83
64
–
85
85
64
86
86
–
74
85
81
55
78
–
80
62
76
53
44
–
44
44
44
41
92
–
92
93
92
80
95
–
95
93
90
96
4
(
a The reaction was performed at −50°C using 3, and −40°C using 2 or 1.
279
b See Ref. 2 for details.
283

282
H. Nogami et al. / Tetrahedron Letters 42 (2001) 279–283
Table 3. Recycle of 3a
Cycle
Time (h)
Yield (%)
Ee (%)
1
2
3
4
5
60
44
44
110
204
98
95
78
97
83
87
81
83
80
77
a 4a was used as substrate.
promoted by the JandaJEL™-supported bifunctional
catalyst 3 are summarized in Table 2.15 In all cases, 3
was more reactive and gave higher enantioselectivities
than 2. Although 3 was slightly less enantioselective
than homogeneous 1, the reactivity was almost com-
parable. The sense of enantioselection was the same as
for the homogeneous catalyst 1, indicating that the dual
activation of the imine and TMSCN by the Lewis acid
metal (Al) and the phosphine oxide should take place,
also in the case of the polymer-supported catalyst 3.
However, using aliphatic pivaladehyde imine, the ee of
the product was significantly lower (20% ee) compared
to 1 (78% ee), which should be the target of the future
investigation.
Figure 1. SEM pictures of the resin. A. Resin prepared by
stirring. B. Resin prepared by shaking.
Finally, we found that 3 could be recovered and recy-
cled at least four times (Table 3), even though the
macroscopic structure of the polymer was broken.16,17
highly enantioselective dual activation pathway by TM-
SCN should be facilitated, thus giving the improved
enantioselectivity.
In conclusion, we have succeeded in immobilizing the
Lewis acid–Lewis base bifunctional catalyst 1 on poly-
mer support. Both of the macroscopic and molecular
structures of the polymer had a profound effect on the
reactivity and enantioselectivity. Using JandaJEL™
and a catalyst preparation method under stirred condi-
tions, the catalyst center became more accessible and
the highly enantioselective dual activation pathway in
which TMSCN acted as the nucleophile became pre-
dominant. Moreover, the catalyst was recyclable. Fur-
ther studies to extend this methodology to a practical
large-scale synthesis are underway.
Next, we investigated the effect of the macroscopic
structure of the polymer beads. The macroscopic poly-
mer structure could affect the accessibility of the sub-
strate to the catalytic center. Since we synthesized the
catalyst with stirring, the polymer beads were physically
broken (Fig. 1, A). Therefore, we synthesized the cata-
lyst by just gently shaking the beads. Thereby, we could
obtain the catalyst containing the spherical polymer
structure (Fig. 1, B). However, this spherical polymer-
supported catalyst turned out to be less active as well as
less enantioselective (Table 1, entries 8, 9). These results
again indicated that the more porous, thus more acces-
sible polymer-supported catalyst (A), should contain
the higher activity as well as enantioselectivity.11
Acknowledgements
This work was supported by CREST and RFTF. We
thank Professor Kobayashi and Mr. Akiyama at Tokyo
University for their kind help measuring SR-MAS
NMR.
Therefore, we tried JandaJEL™ as polymer support
(catalyst 3).12 Recently, Janda developed a polystyrene
resin containing flexible tetrahydrofuran derived cross-
linkers.13 This resin is reported to swell better than
Merrifield resin, and thus the reagents should be more
accessible to the catalytic center. As expected, the
product 5a was obtained in 98% yield with 87% ee,
using 3 as catalyst (Table 1, entry 10). These results
clearly showed that well-swollen JandaJEL™ should
make the catalyst more accessible, thus making the
highly enantioselective dual activation pathway with
TMSCN as the nucleophile more predominant.14 The
results of the enantioselective Strecker-type reaction
References
1. Review: Shuttleworth, S. J.; Allin, S. M.; Wilson, R. D.;
Nasturica, D. Synthesis 2000, 1035–1074.
2. (a) Takamura, M.; Hamashima, Y.; Usuda, H.; Kanai,
M.; Shibasaki, M. Angew. Chem., Int. Ed. 2000, 39,
1650–1652. (b) Takamura, M.; Hamashima, Y.; Usuda,
H.; Kanai, M.; Shibasaki, M. Chem. Pharm. Bull. 2000,
48, 1586–1592.

H. Nogami et al. / Tetrahedron Letters 42 (2001) 279–283
283
3. The development of catalytic asymmetric Strecker-type
reactions is one of the most intensively investigated top-
ics, recently: (a) Iyer, M. S.; Gigstad, K. M.; Namdev, N.
D.; Lipton, M. J. Am. Chem. Soc. 1996, 118, 4910–4911.
(b) Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc.
1998, 120, 4901–4902. (c) Sigman, M. S.; Jacobsen, E. N.
J. Am. Chem. Soc. 1998, 120, 5315–5316. (d) Sigman, M.
S.; Vachal, P.; Jacobsen, E. N. Angew. Chem., Int. Ed.
2000, 39, 1279–1281. (e) Ishitani, H.; Komiyama, S.;
Kobayashi, S. Angew. Chem., Int. Ed. Engl. 1998, 37,
3186–3188. (f) Ishitani, H.; Komiyama, S.; Hasegawa,
Y.; Kobayashi, S. J. Am. Chem. Soc. 2000, 122, 762–766.
(g) Krueger, C. A.; Kuntz, K. W.; Dzierba, C. D.;
Wirschun, W. G.; Gleason, J. D.; Snapper, M. L.; Hov-
eyda, A. H. J. Am. Chem. Soc. 1999, 121, 4284–4285. (h)
Porter, J. R.; Wirschun, W. G.; Kuntz, K. W.; Snapper,
M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 2000, 122,
2657–2658. (i) Corey, E. J.; Grogan, M. J. Org. Lett.
1999, 1, 157–160.
10. Mai, K.; Patil, G. J. Org. Chem. 1986, 51, 3545–3548.
11. The enantioselectivity was not improved using catalysts
containing longer linkers. Merrifield resin-supported cata-
lysts synthesized under shaking with n=5 and 7 gave 5a
with 16 and 0% ee, respectively (versus 16% ee by 2
synthesized by shaking).
12. Catalyst 3 was synthesized by stirring.
13. (a) Toy, P. H.; Janda, K. D. Tetrahedron Lett. 1999, 40,
6329–6332. (b) Reger, T. S.; Janda, K. D. J. Am. Chem.
Soc. 2000, 122, 6929–6934.
14. The difference of loading density of the catalyst on resins
may also affect the activity and enantioselectivity.
15. Representative procedure: To the swollen polymer-sup-
ported ligand (19.1 mg, loading level=0.52 mmol/g, 10
mmol) in CH2Cl2 (0.4 mL), Et2AlCl (0.95 M in hexane, 10
mL, 9.5 mmol) was added at ambient temperature and the
whole was stirred for 1 h. Cooling down to −50°C, imine
4a (27 mg, 0.1 mmol) in CH2Cl2 (0.25 mL), TMSCN (54
t
mL, 0.4 mmol) and BuOH (0.11 mmol) in CH2Cl2 (0.25
4. For the reaction mechanism of the bifunctional catalyst,
see: Hamashima, Y.; Sawada, D.; Kanai, M.; Shibasaki,
M. J. Am. Chem. Soc. 1999, 121, 2641–2642.
mL) were added successively with 10 min interval. After
60 h, satd NaHCO3 (1 mL) was added. Usual workup
and purification by silica gel column chromatography
afforded pure 5a in 98% yield. Ee of the product was
determined by chiral HPLC (see Ref. 2 for details).
16. The procedure for recycling the catalyst 3 is as follows.
After the reaction was completed, dry Et2O (five times
volume to CH2Cl2) was added and the reaction mixture
was stayed for 3 h. The supernatant containing the
product was then taken by a syringe and the catalyst was
washed with dry Et2O five times under an inert atmo-
sphere. Drying the catalyst under reduced pressure for 30
5. For a catalytic asymmetric Strecker reaction by resin-
bound catalyst, see Ref. 3d.
6. (a) Bayston, D. J.; Fraser, J. L.; Ashton, M. R.; Baxter,
A. D.; Polywka, M. E. C.; Moses, E. J. Org. Chem. 1998,
63, 3137. (b) Nozaki, K.; Itoi, Y.; Shibahara, F.; Shi-
rakawa, E.; Ohta, T.; Takaya, H.; Hiyama, T. J. Am.
Chem. Soc. 1998, 120, 4051. (c) Fujii, A.; Sodeoka, M.
Tetrahedron Lett. 1999, 40, 8011. (d) Kobayashi, S.;
Kusakabe, K.; Ishitani, H. Org. Lett. 2000, 2, 1225. (e)
Matsunaga, S.; Ohshima, T.; Shibasaki, M. Tetrahedron
Lett. 2000, 41, 8473.
t
min, CH2Cl2 solvent, the imine, TMSCN, and BuOH
were added at −50°C to start the new cycle.
7. Nicolaou, K. C.; Winssinger, N.; Vourloumis, D.;
Ohshima, T.; Kim, S.; Pfefferkorn, J.; Xu, J.-Y.; Li, T. J.
Am. Chem. Soc. 1998, 120, 10814.
17. The slightly lower yield in cycle 3 might be due to
inevitable loss of the catalyst when the supernatant was
taken by a syringe. The longer reaction time in cycles 4
and 5 might be caused by the above-mentioned effect
and/or by the partial decomposition of the catalyst.
8. Kobayashi, S.; Akiyama, R.; Furuta, T.; Moriwaki, M.
Molecules 1998, 2, 35.
t
9. Using 20 mol% of BuOH did not improve the ee.
.