4284
J. Am. Chem. Soc. 1999, 121, 4284-4285
isopropoxides were found to promote cyanide addition (TMSCN,
20 mol % ligand and metal salt, toluene, 4 °C): Ti(OiPr)4 (28%
ee), Zr(OiPr)4 (0% ee), Al(OiPr)3 (-4% ee), Sm(OiPr)3 (0% ee),
Hf(OiPr)4 (3% ee), Ba(OiPr)2 (-6% ee), Y5(OiPr)13O (-6% ee),
Sr(OiPr)2 (-7% ee), Yb(OiPr)3 (-5% ee), Nd(OiPr)3 (9% ee).7
With Ti(OiPr)4 as the metal salt of choice, a range of solvents
were examined. Both trichloroethane and toluene emerged as
superior media (better enantioselection); we opted for toluene due
to environmental concerns and the convenience of using a less
volatile solvent in library screening. Similar trends were observed
for N-diphenyl methyl imines (cf. Table 1). We selected the latter
class of substrates because, as mentioned above, CN hydrolysis
and amine deprotection may be performed in a single operation.
A survey of potential cyanide donors was carried out. From among
seven candidates,8 TMSCN delivered the most selective, efficient,
and reproducible results.
Next, we utilized our ligand optimization protocol5 to determine
the most selective peptide-Schiff base ligand (cf. 2 in eq 1). Thus,
according to the procedure reported previously, for the Ti-
catalyzed addition of TMSCN to meso epoxides, the three
structural modules of the chiral peptide ligand were modified
systematically. In all the cases examined, the most effective
ligands bear a t-Leu in the AA1 site (adjacent to the Schiff base)
and Thr(t-Bu) in the AA2 position (cf. Table 1). However,
depending on the imine substrate, the optimum Schiff base moiety
within the ligand varies.
Ti-Catalyzed Enantioselective Addition of Cyanide to
Imines. A Practical Synthesis of Optically Pure
r-Amino Acids
Clinton A. Krueger, Kevin W. Kuntz, Carolyn D. Dzierba,
Wolfgang G. Wirschun, John D. Gleason,
Marc L. Snapper,* and Amir H. Hoveyda*
Department of Chemistry, Merkert Chemistry Center
Boston College, Chestnut Hill, Massachusetts 02467
ReceiVed NoVember 24, 1998
The development of asymmetric methods for the synthesis of
nonproteinogenic R-amino acids has been the subject of extensive
research.1 Nonetheless, there is a scarcity of processes that are
catalytic, highly enantioselective, practical and cost-effective, and
deliver R-amino acids not accessible by asymmetric hydrogena-
tion.2 Various recent disclosures indicate that the addition of HCN
to protected imines can be promoted by peptide-derived catalysts.
In a study reported by Lipton,3 selectivities range from <10% ee
(fVal) to >99% ee (fPhGly) (71-97% yield); CN hydrolysis
and amine deprotection were carried out simultaneously by
exposure of CN addition products to HCl. A procedure by
Jacobsen4 provides selectivities from 70% ee (f4-MeOPhGly)
to 91% ee (fPhGly) (65-92% yield), but requires the removal
of the N-allyl protecting group with a Pd catalyst in addition to
a CN hydrolysis step. The latter study utilizes the high-throughput
screening of parallel libraries of Schiff base-peptides to identify
an optimum catalyst.5 A (salen)Mn-catalyzed process has also
been reported (N-allyl imines as substrates) that afford CN
addition products (69-99% yield) in 37% ee (ft-Leu) to 95%
ee (fPhGly).6
Herein, we report that CN addition to a variety of imines is
catalyzed by Ti-tripeptide Schiff base complexes. The reaction
is efficient (g93% conversion) and proceeds with excellent
enantioselectivity (85-97% ee). In most cases, optically pure
(>99% ee) products can be isolated in >80% yields. Moreover,
conversion to the derived R-amino acids proceeds efficiently, with
inexpensive hydrolytic reagents, and without loss of enantiopurity.
Imine 1 was used as the substrate and the modular tripeptide-
Schiff base 2 as the chiral ligand prototype (eq 1) in brief surveys
As illustrated in the left-hand column of Table 1, the
aforementioned search allows the identification of ligands that
afford the derived amino nitriles in 84-97% ee. However, these
transformations proved inefficient and sluggish: conversions of
15-39% were typically obtained after 48 h (Table 1).
To address the efficiency problem, based on various experi-
mental observations and mechanistic hypotheses, we examined
the influence of several protic additives on the rate of CN addition.
On a number of occasions, when transformations were carried
out on larger scale, there was a notable reduction in reaction
efficiency. We argued that adventitious moisture may facilitate
processes performed in smaller quantities, where exclusion of
undesired components is usually more difficult. Furthermore, we
conjectured that, as illustrated in Scheme 1, cleavage of the Ti-N
bond and removal of the Me3Si unit within the purported Ti
complex III (R ) TMS) might lead to the more facile regeneration
of the active catalyst I (enhanced turnover rate). As illustrated in
Table 1 (right column), in the presence of 1.5 equiv of i-PrOH,9
notable enhancements in reactivity are indeed observed.10 Catalyst
turnoVer is facilitated significantly in the presence of i-PrOH; in
several instances enantioselectivities are improved as well (entries
3-5).11
The exact reason for the influence of i-PrOH on CN addition
reactivity and selectivity requires additional investigation. It
nonetheless merits mention that slow addition of a solution of
HCN (in toluene) to a mixture of 13, chiral ligand 9 (10 mol %),
to establish the most appropriate metal center. Ten metal
(1) Williams, R. M. Synthesis of Optically ActiVe R-Amino Acids; Perga-
mon: Oxford, 1989.
(2) Burk, M. J.; Allen, G. J.; Kiesman, W. F. J. Am. Chem. Soc. 1998,
120, 657-663 and references therein.
(7) In the absence of a metal salt, <5% product is observed.
(8) Other cyanating agents examined were as follows: TBSCN, Bu4NCN,
Et2AlCN, acetone cyanohydrin, tert-butyl isocyanide, and HCN.
(9) Other alcohols also provided enhanced catalytst turnover; as an example,
with 15 as the substrate, catalytic addition proceeds to 98% conversion in the
presence of 1 equiv of p-MeO-phenol. See entry 7 of Table 1.
(3) Iyer, M. S.; Gigstad, K. M.; Namdev, N. D.; Lipton, M. J. Am. Chem.
Soc. 1996, 118, 4910-4911.
(4) Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 4901-
4902.
(5) For initial reports on screening of parallel peptide-based libraries, see:
(a) Cole, B. M.; Shimizu, K. D.; Krueger, C. A.; Harrity, J. P.; Snapper, M.
L.; Hoveyda, A. H. Angew. Chem., Int. Ed. Engl. 1996, 35, 1668-1671. (b)
Shimizu, K. D.; Cole, B. M.; Krueger, C. A.; Kuntz, K. W.; Snapper, M. L.;
Hoveyda, A. H. Angew. Chem., Int. Ed. Engl. 1997, 36, 1704-1707. (c)
Shimizu, K. D.; Snapper, M. L.; Hoveyda, A. H. Chem. Eur. J. 1998, 4, 1885-
1889. (d) Hoveyda, A. H. Chem. Biol. 1998, 5, R187-R191.
(6) Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 5315-
5316. While this paper was under review, a Zr-catalyzed addition of Bu3-
SnCN to imines was reported (highest ee ) 92%): Ishitani, H.; Komiyama,
S.; Kobayashi, S. Angew. Chem., Int. Ed. Engl. 1998, 37, 3186-3188.
(10) Initial studies indicate that the CdN bond of the chiral ligand remains
largely unreacted. As an example, 85% of 9 is recovered from the reaction
with 7; the identity of the products arising from the remainder of 9 and whether
they are involved in any catalytic CN addition remains to be determined.
(11) i-PrOH must be added over 20 h to achieve high selectivity and
efficiency. As an example, initial treatment of 10 with 1 equiv of i-PrOH
affords the desired adduct in 70% conversion after 20 h but only in 20% ee
(vs 93% ee with slow addition). It is likely that rapid addition leads to (i)
formation of large amounts of HCN which at 4 °C add to imines in the absence
of a catalyst and (ii) displacement of the chiral ligand from the transition
metal center.
10.1021/ja9840605 CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/05/1999