Synthesis of α-Amino Acids via Asymmetric Phase Transfer-Catalyzed Alkylation of Achiral Nickel(II) Complexes of Glycine-Derived Schiff Bases

Article abstract of DOI:10.1021/ja035465e

Achiral, diamagnetic Ni(II) complexes 1 and 3 have been synthesized from Ni(II) salts and the Schiff bases, generated from glycine and PBP (7) and PBA (11), respectively, in MeONa/MeOH solutions. The requisite carbonyl-derivatizing agents pyridine-2-carboxylic acid(2-benzoyl-phenyl)-amide 7 (PBP) and pyridine-2-carboxylic acid(2-formyl-phenyl)-amide 11 (PBA) were readily prepared from picolinic acid and o-aminobenzophenone or picolinic acid and methyl o-anthranilate, respectively. The structure of 1 was established by X-ray crystallography. Complexes 1 and 3 were found to undergo C-alkylation with alkyl halides under PTC conditions in the presence of β-naphthol or benzyltriethylammonium bromide as catalysts to give mono- and bis-alkylated products, respectively. Decomposition of the complexes with aqueous HCI under mild conditions gave the required amino acids, and PBP and PBA were recovered. Alkylation of 1 with highly reactive alkyl halides, carried out under the PTC conditions in the presence of 10% mol of (S)- or (R)-2-hydroxy-2′ -amino-1,1′-binaphthyl 31a (NOBIN) and/or its N-acyl derivatives and by (S)- or (R)-2-hydroxy-8′-amino-1,1′-binaphthyl 32a (iso-NOBIN) and its N-acyl derivatives, respectively, gave rise to α-amino acids with high enantioselectivities (90-98.5% ee) in good-to-excellent chemical yields at room temperature within several minutes. An unusually large positive nonlinear effect was observed in these reactions. The Michael addition of acrylic derivatives 37 to 1 was conducted under similar conditions with up to 96% ee. The ¹H NMR and IR spectra of a mixture of the sodium salt of NOBIN and 1 indicated formation of a complex between the two components. Implications of the association and self-association of NOBIN for the observed sense of asymmetric induction and nonlinear effects are discussed.

Full text of DOI:10.1021/ja035465e

Published on Web 09/30/2003
Synthesis of r-Amino Acids via Asymmetric Phase
Transfer-Catalyzed Alkylation of Achiral Nickel(II) Complexes
of Glycine-Derived Schiff Bases
Yuri N. Belokon,*,† Natalia B. Bespalova,† Tatiana D. Churkina,† Ivana C´ısarˇova´,‡
Marina G. Ezernitskaya,† Syuzanna R. Harutyunyan,† Radim Hrdina,‡,¶
Henri B. Kagan,§ Pavel Kocˇovsky´,*,¶ Konstantin A. Kochetkov,† Oleg V. Larionov,†
Konstantin A. Lyssenko,† Michael North, Miroslav Pola´sˇek,∇
Alexander S. Peregudov,† Vladimir V. Prisyazhnyuk,† and Sˇ teˇpa´n Vyskocˇil*,‡,¶,∆
Contribution from A. N. NesmeyanoV Institute of Organo-Element Compounds,
Russian Academy of Sciences, 117813, Moscow, VaViloV 28, Russian Federation,
Department of Organic Chemistry, Faculty of Science, Charles UniVersity, HlaVoVa 2030,
12840 Prague 2, Czech Republic, Department of Chemistry, Joseph Black Building, UniVersity
of Glasgow G12 8QQ, United Kingdom, Institut de Chimie Moleculaire d’Orsay, Laboratoire de
Synthe`se Asyme´trique (UPRESA 8075), UniVersite´ de Paris Sud, 91405-Orsay Cedex, France,
Department of Chemistry, King’s College, Strand, London, WC2 R2LS, United Kingdom, and
J. HeyroVsky´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic,
18223 Prague 8, Czech Republic
Received April 4, 2003; E-mail: yubel@ineos.ac.ru; P.Kocovsky@chem.gla.ac.uk; stepanv@natur.cuni.cz
Abstract: Achiral, diamagnetic Ni(II) complexes 1 and 3 have been synthesized from Ni(II) salts and the
Schiff bases, generated from glycine and PBP (7) and PBA (11), respectively, in MeONa/MeOH solutions.
The requisite carbonyl-derivatizing agents pyridine-2-carboxylic acid(2-benzoyl-phenyl)-amide 7 (PBP) and
pyridine-2-carboxylic acid(2-formyl-phenyl)-amide 11 (PBA) were readily prepared from picolinic acid and
o-aminobenzophenone or picolinic acid and methyl o-anthranilate, respectively. The structure of 1 was
established by X-ray crystallography. Complexes 1 and 3 were found to undergo C-alkylation with alkyl
halides under PTC conditions in the presence of â-naphthol or benzyltriethylammonium bromide as catalysts
to give mono- and bis-alkylated products, respectively. Decomposition of the complexes with aqueous HCl
under mild conditions gave the required amino acids, and PBP and PBA were recovered. Alkylation of 1
with highly reactive alkyl halides, carried out under the PTC conditions in the presence of 10% mol of (S)-
or (R)-2-hydroxy-2′-amino-1,1′-binaphthyl 31a (NOBIN) and/or its N-acyl derivatives and by (S)- or (R)-2-
hydroxy-8′-amino-1,1′-binaphthyl 32a (iso-NOBIN) and its N-acyl derivatives, respectively, gave rise to
R-amino acids with high enantioselectivities (90-98.5% ee) in good-to-excellent chemical yields at room
temperature within several minutes. An unusually large positive nonlinear effect was observed in these
reactions. The Michael addition of acrylic derivatives 37 to 1 was conducted under similar conditions with
up to 96% ee. The 1H NMR and IR spectra of a mixture of the sodium salt of NOBIN and 1 indicated
formation of a complex between the two components. Implications of the association and self-association
of NOBIN for the observed sense of asymmetric induction and nonlinear effects are discussed.
Introduction
tert-butyl ester with alkyl halides, developed by O’Donnell et
al.2 Catalytic asymmetric versions of this reaction are being
The synthesis of nonproteinogenic R-amino acids remains a
subject of considerable interest because of their great importance
in biology, medicine, and synthetic chemistry.1 An increasingly
popular approach to chiral R-amino acids with a tertiary
R-carbon atom relies on the C-C bond formation via alkylation
of glycine derivatives, such as N-(diphenylmethylene)glycine
sought, and following the seminal work by O’Donnell et al.2c
on asymmetric alkylations with cinchona alkaloid derivatives
as chiral phase-transfer catalysts (PTC), dramatic improvements
have been achieved.3 Purely synthetic, chiral C2 symmetrical
ammonium salts have recently been prepared and shown to be
highly efficient in the same set of reactions.4 Nevertheless,
† A. N. Nesmeyanov Institute.
‡ Charles University.
(1) (a) Heimgartner, H. Angew. Chem., Int. Ed. Engl. 1991, 30, 238-264. (b)
Williams, R. M.; Hendrix, J. A. Chem. ReV. 1992, 92, 889. (c) Duthaler,
R. O. Tetrahedron 1994, 50, 1539. (d) Wirth, T. Angew. Chem., Int. Ed.
Engl. 1997, 36, 225. (e) Gibson, S. E.; Guillo, N.; Tozer, M. J. Tetrahedron
1999, 55, 585. (f) Diaz-de-Villegas, M. D.; Cativiela, C. Tetrahedron:
Asymmetry 1998, 9, 3517. (g) Cativiela, C.; Diaz-de-Villegas, M. D.
Tetrahedron: Asymmetry 2000, 11, 645.
¶ University of Glasgow.
§ Universite´ de Paris Sud.
King’s College.
∇ J. Heyrovsky´ Institute.
∆ Current address: Department of Chemistry, The Scripps Research
Institute, La Jolla, CA 92037.
9
12860
J. AM. CHEM. SOC. 2003, 125, 12860-12871
10.1021/ja035465e CCC: $25.00 © 2003 American Chemical Society

Synthesis of R-Amino Acids
A R T I C L E S
Chart 1. Ketimine and Aldimine Ni(II) Complexes
Scheme 1. Synthesis of Ni(II) Complexes 1 and 2
despite the recent progress in the catalytic asymmetric synthesis
of R-amino acids,5 asymmetric PTC alkylation of glycine or
alanine derivatives still represents the simplest and most
straightforward route to a variety of enantiomerically enriched
R-amino acids.
Previously, we have reported on the synthesis and application
of the square-planar nickel(II) complex 1 (Chart 1) in asym-
metric Michael reaction, catalyzed by (R,R)-TADDOL [(4R,5R)-
2,2-dimethyl-1,3-dioxolane-4,5-bis(diphenylmethanol)], which
led to 4-methylglutamic acid with low enantioselectivity (28%
ee).6a More recently, we employed NOBIN (2-amino-2′-hy-
droxy-1,1′-binaphthyl) as a novel type of PTC catalyst in the
alkylation reaction of N-(phenylmethylene)alanine esters pro-
ducing R-methyl-R-amino acids with modest enantioselectivity
(68% ee).6b
Scheme 2. Synthesis of Ni(II) Complexes 3 and 4
Herein, we report on the introduction of one or two alkyl
groups (identical or different) into the achiral Ni(II) complex
1, derived from the Schiff base of glycine and pyridine-2-
carboxylic acid(2-benzoyl-phenyl)-amide 7 (PBP), in a selective,
stepwise manner. This approach represents a viable route to the
preparation of either racemic or enantiomerically enriched
R-monosubstituted R-amino acids or R,R-disubstituted R-amino
acids. Also reported is the achiral Ni(II) complex 3, obtained
(2) (a) O’Donnell, M. J.; Boniece, I. M. Tetrahedron Lett. 1978, 19, 2641. (b)
O’Donnell, M. J.; Eckrich, T. M. Tetrahedron Lett. 1978, 19, 4625. (c)
Scott, W. L.; Zhou, Ch.; Fang, Z.; O’Donnell, M. J. Tetrahedron Lett. 1997,
38, 3695. (d) Griffith, D. L.; O’Donnell, M. J.; Pottorf, R. S.; Scott, W. L.;
Porco, J. A. Tetrahedron Lett. 1997, 38, 8821. (e) O’Donnell, M. J.; Bennett,
W.; Bruder, W.; Jacobsen, W.; Knuth, K.; LeClef, B.; Polt, R.; Bordwell,
F.; Mrozack, S.; Cripe, T. J. Am. Chem. Soc. 1988, 110, 8520. (f)
O’Donnell, M. J.; Delgano, F.; Pottorf, R. S. Tetrahedron 1999, 55, 6347.
(g) O’Donnell, M. J.; Bennett, W. D.; Wu, S. J. Am. Chem. Soc. 1989,
111, 2353. (h) O’Donnell, M. J.; Dominguez, E.; Scott, W. L.; Delgado,
F.; de Blas, J. Tetrahedron: Asymmetry 2001, 12, 821. (i) O’Donnell, M.
J.; Delgano, F.; Hostettler, C.; Schwesinger, R. Tetrahedron Lett. 1998,
39, 8775.
(3) (a) Corey, E. J.; Xu, F.; Noe, M. J. Am. Chem. Soc. 1997, 119, 12414. (b)
Corey, E. J.; Noe, M.; Xu, F. Tetrahedron Lett. 1998, 39, 5347. (c) Corey,
E. J.; Bo, Y.; Busch-Petersen, J. J. Am. Chem. Soc. 1998, 120, 13000. (d)
Lygo, B.; Crosby, J.; Lowdon, T. R.; Wainwright, P. G. Tetrahedron Lett.
1997, 38, 2343. (e) Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1997,
38, 8595. (f) Lygo, B.; Wainwright, P. G. Tetrahedron Lett. 1998, 39, 1599.
(g) Lygo, B.; Crosby, J.; Peterson, J. A. Tetrahedron Lett. 1999, 40, 8671.
(h) Chinchilla, R.; Mazon, P.; Najera, C. Tetrahedron: Asymmetry 2000,
11, 3277.
from the Schiff base of glycine and pyridine-2-carboxylic acid-
(2-formyl-phenyl)-amide 11 (PBA) as a convenient substrate
for the preparation of achiral, highly constrained R,R-disubsti-
tuted R-amino acids. Finally, we explore asymmetric catalytic
C-alkylation of 1 with alkyl halides and Michael acceptors under
PTC conditions, using NOBIN, iso-NOBIN, and their deriva-
tives as catalysts.7
Results
(4) (a) Ooi, T.; Kameda, M.; Maruoka, K. J. Am. Chem. Soc. 1999, 121, 6519.
(b) Ooi, T.; Takeuchi, M.; Kameda, M.; Maruoka, K. J. Am. Chem. Soc.
2000, 122, 5228. (c) Ooi, T.; Uematsu, Y.; Kameda, M.; Maruoka, K.
Angew. Chem., Int. Ed. 2002, 41, 1551.
(5) (a) Bogevig, A.; Juhl, K.; Kumaragurubaran, N.; Zhuang, W.; Jørgensen,
K. A. Angew. Chem., Int. Ed. 2002, 41, 1790 and references therein. (b)
Sigman, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 1998, 120, 5315. (c)
Krueger, C. A.; Kuntz, K. W.; Dzierba, C. D.; Wirschun, W. G.; Gleason,
J. D.; Snapper, M. L.; Hoveyda, A. H. J. Am. Chem. Soc. 1999, 121, 4284.
(d) Ishitani, H.; Komiyama, S.; Hasegawa, Y.; Kobayashi, S. J. Am. Chem.
Soc. 2000, 122, 762 and references therein.
(6) (a) Belokon, Y. N.; Kochetkov, K. A.; Churkina, T. D.; Ikonnikov, N. S.;
Orlova, S. A.; Smirnov, V. V.; Chesnokov, A. A. MendeleeV Commun.
1997, 137. (b) Belokon, Y. N.; Kochetkov, K. A.; Churkina, T. D.;
Ikonnikov, N. S.; Chesnokov, A. A.; Larionov, O. V.; Singh, I.; Parmar,
V. S.; Vyskocˇil, Sˇ.; Kagan, H. B. J. Org. Chem. 2000, 65, 7041.
Synthesis and Structure of Ni(II) Complexes 1 and 3.
Complexes 1 and 3 were prepared from glycine, Ni(NO3)2, and
the respective ligand precursors 7 (PBP) and 11 (PBA) in the
presence of KOH or MeONa in methanol (Schemes 1 and 2).
The red-colored, crystalline, diamagnetic complexes 1 and 3
can be purified by chromatography or crystallization from
CHCl3. Racemic complexes 2 and 4 were obtained in the same
way, using (()-alanine instead of glycine. The ketone precursor
7 (PBP) was obtained by condensation of the in situ-generated
chloride of R-picolinic acid (5) with o-aminobenzophenone (6),
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12861

A R T I C L E S
Belokon et al.
Table 1. Selected Bond Lengths (Å) and Angles in Complexes 1
and (()-2
(bond/angles)
atoms
1
2
Ni(1)-O(1)
Ni(1)-N(1)
Ni(1)-N(2)
Ni(1)-N(3)
O(1)-C(21)
O(2)-C(21)
O(3)-C(6)
N(1)-C(1)
N(1)-C(5)
N(2)-C(6)
N(2)-C(7)
N(3)-C(13)
N(3)-C(20)
1.851(2)
1.876(3)
1.861(3)
1.843(2)
1.289(4)
1.231(4)
1.220(4)
1.333(4)
1.352(4)
1.381(4)
1.395(4)
1.292(4)
1.486(4)
1.857(2)
1.881(3)
1.874(3)
1.846(3)
1.299(4)
1.217(4)
1.217(4)
1.353(8)
1.323(4)
1.370(4)
1.402(5)
1.297(4)
1.486(4)
Figure 1. Scheme illustrating the formation of the heterochiral dimers in
the crystal of 1.
O(1)-Ni(1)-N(1)
O(1)-Ni(1)-N(2)
O(1)-Ni(1)-N(3)
N(1)-Ni(1)-N(2)
N(1)-Ni(1)-N(3)
N(2)-Ni(1)-N(3)
90.8(1)
176.7(1)
87.6(1)
86.0(1)
176.0(1)
95.6(1)
90.7(1)
173.3(1)
87.1(1)
86.1(1)
175.4(2)
96.5(1)
complexes. Although both complexes 1 and (()-2 are assembled
into centrosymmetric dimers, the nature of the interaction
between the stacks is different. The heterochiral dimer of 1 is
interconnected by the weak Ni(1)‚‚‚N(1) contacts [Ni(1A)‚‚‚
N(1A) 3.287(2)Å] and by the interaction of the N(3) atom with
the π-system of the pyridine ring [N(3)‚‚‚C(2A) 3.319(3)Å]
(Figure 1). By contrast, the presence of the methyl group in
(()-2 makes this stacking-type interaction impossible, and as a
result, the interdimer interactions are limited to only a weak
contact of the nickel atom with the carbonyl group (Ni(1)‚‚‚
O(3A) 3.287(2)Å) (Figure 2).
Synthesis of Racemic Amino Acids by Alkylation of Ni-
(II) Complexes 1 and 3. The alkylation of both 1 and 3 with
alkyl halides was carried out in the presence of Bu4NBr, Bu4-
NCl, or â-naphthol as PTC catalysts in CH2Cl2 with solid NaOH
as a base (Scheme 3), and the reaction was monitored by TLC.
After completion, the reaction mixture was neutralized and the
red-colored solid residue was purified either by chromatography
or crystallization. In most cases the yields exceeded 95%, and
the purification was not necessary. Decomposition of the
resulting complexes 12, 13-16, and 22-24 was effected by
diluted methanolic HCl within 5 min at 50 °C to produce the
corresponding mono- and bis-alkylated amino acids 17, 18-
21, 25, and 26, respectively. The process was easily followed
by the change of the solution color from red to blue. The
hydrochlorides of 7 (PBP) and 11 (PBA) were removed by
filtration in almost quantitative yields, and NiCl2 and the amino
acid were easily isolated by ion exchange chromatography. The
results of the alkylations are summarized in Table 2.
Figure 2. Scheme illustrating the formation of the heterochiral dimers in
the crystal of (()-2. The other position of the disordered ligand is omitted
for clarity.
as reported in our preliminary communication;7a an improved
procedure is described in the Experimental Section of the
Supporting Information. The aldehyde precursor 11 was syn-
thesized as follows: condensation of the in situ-generated
chloride of R-picolinic acid (5) with methyl o-anthranilate (8)
afforded amide 9, which was reduced with LiBH4 to give alcohol
10. Swern oxidation of the latter alcohol provided the required
aldehyde 11.
The X-ray structure of 1 and (()-2 (Figures 1 and 2, Table
1) suggests that the complexes are neutral, with the two positive
charges at the central Ni ion neutralized by two negative charges
(CON- and COO-) of the tetradentate ligand. The ligand is
slightly puckered with two enantiomeric conformations of two
molecules of 1 and (()-2 in the crystal cell, restoring the overall
racemic crystal arrangement. The principal bond lengths and
angles in complexes 1 and (()-2 are close to each other and
are typical for this type of compound (see Table 1). The slight
variation of the bond lengths in the ligand in 1 and (()-2 cannot
be rationalized as the consequence of the presence of the methyl
group at C(20) or the influence of crystal packing and seems to
originate from a systematic bias introduced by the disorder in
(()-2.
The dihedral angles between the Ni(1)O(1)N(1)N(2)N(3)
plane and the phenyl ring in 1 and (()-2 differ slightly (90.8°
and 108.8°, respectively). The difference most likely originates
in the steric interaction of the methyl substituent and the Ph
group in (()-2 that is absent in 1. The effect of the methyl
group is also reflected in the supramolecular assembly of the
At a 1:1 molar ratio of the alkylating agent to the substrate,
the monoalkylation of the ketimine complex 1 proceeded
quantitatively both in CH2Cl2 under PTC conditions and in DMF
in the presence of NaOH or NaH (Table 2, entries 1-4). The
use of sterically hindered alkyl halides such as i-PrI gave rise
to mono-alkylated products with 1, even at a 3:1 ratio to the
substrate in DMF (Table 2, entry 5). On the other hand, bis-
alkylation of 1 can be performed by employing 2-3 equiv of
the more reactive alkylating agents, such as benzyl and allyl
bromide (Table 2, entries 6 and 7) in DMF. R,R′-Dibromo-o-
xylene can be employed to give cleanly the corresponding
(7) Preliminary data on the asymmetric alkylation of 1 under PTC conditions,
catalyzed by NOBIN, have been reported by us earlier: (a) Belokon, Y.
N.; Kochetkov, K. A.; Churkina, T. D.; Ikonnikov, N. S.; Larionov, O. V.;
Harutyunyan, S.; North, M.; Vyskocˇil, Sˇ.; Kagan, H. B. Angew. Chem.,
Int. Ed. 2001, 40, 1948. (b) Vyskocˇil, Sˇ.; Meca, L.; Tisˇlerova´, I.; C´ısaˇrova´,
I.; Pola´sˇek, M.; Harutyunyan, S.; Belokon, Y. N.; Stead, R. M. J.; Farrugia,
L.; Kocˇovsky´, P. Chem.-Eur. J. 2002, 8, 4633.
9
12862 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Synthesis of R-Amino Acids
A R T I C L E S
Scheme 3. Alkylation of Ni(II) Complexes 1 and 3a
that various reactions could be easily performed on the groups
of the side chains. As an illustration, a ruthenium-catalyzed ring-
closing metathesis was carried out with the diallylglycine
complex 14, which resulted in a ready formation of 1-amino-
1-carboxycyclopent-3-ene 28 after decomplexation of the in-
termediate 27 (Scheme 4).
Synthesis of Enantiomerically Enriched r-Amino Acids
by Asymmetric Alkylation of the Ni(II) Complex 1 with
Alkyl Halides, Catalyzed by NOBIN, iso-NOBIN and Their
Congeners 31-32. Asymmetric alkylation of 1 in CH2Cl2
(Scheme 5) was carried out in the presence of cinchonine
derivative 29, (R,R)-TADDOL 30, NOBIN 31a (and its deriva-
tives 31d-h), and iso-NOBIN 32a (and its derivatives 32b-g)
as catalysts (Chart 2). Catalysts 29 and 30 gave low chemical
yields (less than 50%) even after prolonged treatment (1 h),
and the ee of the resulting phenylalanine (17a) was in the range
of 5-16% (Table 3, entries 1-3). By contrast, NOBIN-type
binaphthyls 31 and 32 proved much more efficient.
Thus, benzylation of 1, catalyzed by (R)-NOBIN 31a (or its
enantiomer) in toluene (Scheme 5), gave the mono-alkylated
complex 12a in a 50% chemical yield, and the released
phenylalanine (17a) was of 89% ee (Table 3, entry 4). The
reaction carried out in CH2Cl2 gave (R)-Phe [or (S)-Phe] in 88-
90% chemical yield with 96-97% ee within 8 min (Table 3,
entries 5 and 6). As expected, the increase in solvent polarity
(MeCN) diminished the ee of the alkylation (Table 3, entry 7),
whereas (CH2)2Cl2 served as a good substitute for CH2Cl2 (Table
3, entries 8 and 9), allowing the reaction to be carried out at
higher temperatures (up to 70 °C) without a significant loss in
the product ee (Table 3, entry 9).
The nature of the base was important in these reactions as
the transition from solid NaOH to KOH and then to CsOH‚
H2O brought the ee of the reaction progressively from 96% to
16% and finally to 10% (Table 3, compare entries 6, 10, and
11). Switching from solid NaOH to 50% aqueous NaOH was
detrimental to the enantioselectivity, which fell from 96% ee
to 55% ee; simultaneously, the chemical yield dropped to a
meager 5% after 1 h (Table 3, entry 12). Significantly, solid
NaH proved to be almost as efficient as NaOH (Table 3,
compare entries 6 and 13).
An attempt at using BINOL (31b) or 2,2′-diamino-1,1′-
binaphthyl (31c) as catalysts resulted in both low ee and
chemical yields of the product (Table 3, entries 14 and 15).
The modifications of 31a by replacing the NH2 group with
NMe2 (31d) or NHPh (31e) invariably decreased the efficiency
of the reaction by slowing the rate and decreasing the enantio-
selectivity to 3-5% ee (Table 3, entries 16 and 17). N-Formyl
NOBIN 31f was also inactive (Table 3, entry 18), whereas
modest restoration of reactivity was observed for the (R)-N-
acetyl derivative 31g, which gave the final (S)-Phe of 28% ee
(Table 3, entry 19). Interestingly, the latter instance constitutes
the reversal of the sense of chirality as compared to the catalysis
by (R)-NOBIN (Table 3, compare entries 6 and 19). The
introduction of three fluorine atoms into the N-acyl-moiety (31h)
resulted in a total loss of catalytic activity (Table 3, entry 20).
(S)-iso-NOBIN (32a) proved to be a fairly efficient catalyst
for the production of (S)-Phe with 87.5% ee and 36% chemical
a
For catalysts, conditions, and results, see Table 2.
complex of 2-amino-2′-carboxy-indane 15 (Table 2, entry 8)
from which the amino acid 21 can be released. Notably, the
alkylation of a chirally modified O’Donnell substrate with this
alkyl halide was reported as being accompanied by simultaneous
N-alkylation of the nitrogen atom of the glycine moiety, giving
rise to the corresponding heterocyclic derivative of amino acid.8
A procedure for the stepwise bis-alkylation of 1 was
elaborated, starting with mono-alkylation of 1 with an alkyl
bromide under PTC conditions in CH2Cl2, followed by a second
alkylation with another activated alkyl halide in DMF. In this
way, alkylation of 1 with EtBr, followed by the alkylation of
the resulting mono-alkylated complex with benzyl bromide, gave
R-ethylphenylalanine 20 after the decomposition of the bis-
alkylated complex (Table 2, entry 9).
The glycine moiety of the aldimine complex 3 is much less
sterically hindered, and only bis-alkylated complexes were
formed selectively under PTC in CH2Cl2 or DMF even at a 1:1
ratio of the alkylating agent to the substrate. The increase of
the latter ratio to 2.5 led to the formation of the bis-alkylated
complexes in quantitative yields (Table 2, entries 10 and 11).
Even the sterically hindered iso-propyl iodide reacted readily
to give the corresponding bis-alkylated complex (Table 2, entry
11), from which R,R-diisopropylglycine 26 was released in a
very good chemical yield.9
The Ni chelation serves as a means of protection for both
the amino and carboxyl groups of the amino acid moiety so
(9) The Schiff base derived from benzaldehyde and i-propyl glycinate gave
no products of bis-C-alkylation under the same conditions; only a mixture
of the mono-alkylated product, i.e., ValO-i-Pr Schiff base, and unidentified
material was detected by 1H NMR in the reaction mixture.
(8) Guillena, G.; Najera, C. J. Org. Chem. 2000, 65, 7310.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12863

A R T I C L E S
Belokon et al.
Table 2. Mono- and Bis-alkylation of Ni(II) Complexes 1 and 3 with Alkyl Halides (Scheme 3)a
entry
substrate
alkyl halideb
solvent/base
catalystc (10−15%)
time (h)
mono-alkylationd
bis-alkylationd
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
PhCH2Br
CH2Cl2/NaOH
CH2Cl2/NaOH
CH2Cl2/NaOH
DMF/NaOHa
DMF/NaH
DMF/NaOH
DMF/NaH
DMF/NaH
TEBA-Br
TBAB
TEBA-Br
1
1
>95
>95
>95
>83
>95
AllylBr
EtI
Ph2CHCl
15
0.5
1
i-PrI (3 equiv)
PhCH2Br (2 equiv)
AllylBr (3 equiv)
o-C6H4(CH2Br)2 (1equiv)
(1) EtI
(2) PhCH2Br
EtI (2.5 equiv)
i-PrI (3 equiv)
o-C6H4(CH2Br)2 (1equiv)
1
81
59
1
1
32e
CH2Cl2/NaOH
DMF/NaOH
DMF/NaOHf
DMF/NaH
TEBA-Br
15
0.5
0.5
1
>95
>95
>95
70
10
11
12
3
3
3
DMF/NaOH
1
48e
a
b
The substrate 0.1-0.4 M. Unless indicated otherwise, 1.1 equiv. Methyl iodide and other substituted benzyl bromides could also be used successfully.
c
d
â-Naphthol (entries 1-5) could also be used as catalysts. The chemical yields of the products were calculated on the initial 1 or 3 and determined by
e
f
weighing the separated complexes before their decomposition. The initial substrates were recovered from the reaction mixture in 30-40% yields. NaH
can also be used as a base in the reaction.
Scheme 4. Grubbs Cyclization of 14
Chart 2. Chiral Phase Transfer Catalystsa
Scheme 5. Asymmetric Alkylation of Ni(II) Complexes 1, 2, and 4a
a
Both enantiomers were available for 31a, 31b, and 32a; Ad )
1-adamantyl.
Table 4 summarizes the alkylation of substrate 1 with different
alkyl halides catalyzed by (S)- or (R)-NOBIN 31a under optimal
conditions (as in Table 3, entry 6). As can be seen from the
data, all the activated alkyl halides participated in the reaction,
affording the alkylated products in good chemical yields and
high ee (92-98.5%) at room temperature within 4-30 min
(Table 4, entries 1-7). As expected, unactivated alkyl halides
were much less efficient so that relatively low yields of the
alkylated product were obtained within short time intervals
(Table 4, entries 8, 9, and 11), although the ee of the resulting
amino acids was still good. The yields could be improved by
using larger amounts of NOBIN (Table 4, entry 9). Longer
reaction times resulted in a gradual decrease of the ee of the
final amino acid because of a partial racemization of the final
complex (Table 4, entry 10). The introduction of electron-
withdrawing groups into the side chain of the amino acid moiety
a
For catalysts, conditions, and results, see Tables 3 and 4.
yield after 13 min (Table 3, entry 21). (S)-N-Acyl derivatives
32b-d also turned out to be catalytically active, furnishing
phenylalanine of 71-92% ee with reversed sense of chirality,
as compared to the parent catalyst 32a of the same configuration
(Table 3, entries 22-24). The size of the acyl groups was
identified as a decisive factor: the larger that group, the higher
the ee. However, introduction of an N-tosyl group (32e) ruined
the catalytic activity (Table 3, entry 25). By contrast, the
NOBIN-catalyzed benzylation of monoalkylated complexes 2
and 4 (derived from alanine) proved to be very slow and
exhibited low asymmetric inductions (Table 3, entries 26 and
27).
9
12864 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Synthesis of R-Amino Acids
A R T I C L E S
Table 3. Asymmetric PTC Alkylation of Ni(II) Complexes 1, 2, and 4 with Benzyl Bromide Promoted by Different Catalysts and Types of
Bases at Room Temperature (Scheme 5)a
Ni(II)
complex
time
(min)
yieldb
(%)
ee of Phe (%)
(configuration)c
entry
solvent
catalyst
29
30
30
base
NaOH
1
2
3
4
5
6
7
8
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
4
CH2Cl2
CH2Cl2
C6H5CH3
C6H5CH3
CH2Cl2
CH2Cl2
MeCN
60
10
45
45
8
8
20
5
50
10
10
50
90
88
80
70
86
80
25
5
50
14
16
15
10
traces
30
traces
36
90
60
70
traces
90
92
16 (R)
12 (R)
5 (S)
89 (R)
97 (R)
96 (S)
17 (R)
97 (R)
93 (R)
16 (R)
10 (R)
55 (R)
97 (R)
17 (S)
18 (S)
5 (R)
3 (R)
not determined
28 (S)
not determined
87.5 (S)
71 (S)
90 (R)
92 (R)
not determined
44 (R)
20 (R)
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
KOH
(R)-31a
(R)-31a
(S)-31a
(R)-31a
(R)-31a
(R)-31a
(R)-31a
(R)-31a
(R)-31a
(R)-31a
(S)-31b
(R)-31c
(R)-31d
(R)-31e
(R)-31f
(R)-31g
(R)-31h
(S)-32a
(R)-32b
(S)-32c
(S)-32d
(S)-32e
(R)-31a
(R)-31a
(CH2)2Cl2
(CH2)2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
9d
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
5
7
CsOH × H2O
50% aq NaOH
NaH
15
60
11
60
30
15
60
35
30
20
13
6
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
NaOH
7
9
30
1200
40
a
Reaction conditions: concentration of the substrate (1, 2, or 4): 0.17-2.0 M; ratio of the PhCH2Br/substrate, 1.2:1; catalyst/substrate, 1:10; MOH/
b
substrate, 10:1; room temperature unless indicated otherwise; under Ar, all solvents were carefully dried. Chemical yield of the alkylated complex determined
c
after its separation from unreacted substrate. Determined for phenylalanine recovered from the complex by chiral GLC analysis on chiral Chirasil-Val
columns. d The reaction was carried out at 70 °C.
made racemization faster, as illustrated by the decline of ee in
the alkylation with tert-butyl bromoacetate (Table 4, entries 12-
14).
Another interesting feature of the NOBIN catalysis was the
observation of a significant positive nonlinear effect (NLE)10
for the alkylation of 1 with BnBr (Figure 4). In practical terms,
this means that even the catalyst of 30% ee is sufficient to bring
about the same level of asymmetric induction as the enantio-
merically pure catalyst.
The intermediate alkylated complexes can be purified either
by chromatography or by crystallization before releasing the
amino acid. With the crystallization procedure, the ee of the
complex or recovered amino acid was further improved (Table
4, entry 1). Enantiomeric purity of the amino acid moiety can
be roughly assessed by measuring the optical rotation of the
complexes before crystallization, as the specific rotation of
enantiomerically pure samples were only marginally influenced
by the structure of its side chain. Furthermore, the same
principles apply to the CD spectra of the complexes: Figure 3
illustrates the positive Cotton effects for the complexes of amino
acids of (S)-configuration and negative effects for the (R)-amino
acid complexes. Despite different structures of the side chains
of the amino acid moieties (phenylalanine and glutamic acid),
the CD curves were almost mirror images. Thus, the absolute
configuration of any amino acid moiety can be easily assigned
by checking the sign of the Cotton effect of the corresponding
Ni complex.
The solubility of 1 in CH2Cl2 [or (CH2)2Cl2] was found to
greatly increase in the presence of sodium NOBIN-ate (Figure
5). Since the solubility experiments were always conducted
under excess of insoluble 1, the observed straight line depen-
dence of the concentration of 1 in solution versus the concentra-
tion of sodium (S)- or (R)-NOBIN-ate reflects the formation of
a well soluble complex formed between 1 and the enantiomeri-
cally pure NOBIN-ate in a 1:1 ratio with an equilibrium constant
of 165 M-1. Self-association of the racemic mixture of sodium
NOBIN-ates, which effectively decreases the concentration of
active monomeric sodium NOBIN-ate, appears to be the most
plausible explanation for the ineffectiveness of the racemate.
Synthesis of Enantiomerically Enriched r-Amino Acids
by Asymmetric Michael Addition of the Ni(II) Complex 1
to Michael Acceptors 37, Catalyzed by NOBIN, iso-NOBIN,
and Their Congeners 31-32. Since enolates react with alkyl
halides via SN2 mechanism, the above asymmetric alkylation
is limited to the use of good electrophiles, such as benzyl, allyl,
and primary alkyl halides. Further extension of this methodology
should, therefore, be sought in the area of nucleophilic addition
to suitable electrophiles, such as Michael acceptors or aldehydes.
As in the case of racemic synthesis, the amino acids and PBP
(7) could be easily recovered in a quantitative yield and PBP
(7) reused to prepare complex 1. NOBIN (31a) could also be
partly recovered from the reaction mixture by chromatography
(usually in less than 20% yield), though significant amounts of
alkylated NOBIN derivatives of unidentified structure were also
detected in the reaction mixture. Fortunately, the latter byprod-
ucts were not catalytically active in the reaction, as shown by
control experiments.
(10) (a) Girard, C.; Kagan, H. B. Angew. Chem., Int., Ed. 1998, 37, 2922. (b)
Mikami, K.; Terada, M.; Korenaga, T.; Matsumoto, Y.; Ueki, M.; Angelaud,
R. Angew. Chem., Int. Ed. 2000, 39, 3532.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12865

A R T I C L E S
Belokon et al.
Table 4. Asymmetric PTC Alkylation of Ni(II) Complex 1 with Alkyl
Halides Promoted by (R)-NOBIN (31a) at Room Temperature
(Scheme 5)a
Figure 4. Positive nonlinear effect in the alkylation of 1 with BnBr
catalyzed by (R)-NOBIN.
a Reaction conditions: the reactions were carried out in CH2Cl2;
concentration of 1 was 0.17-2.0 M; ratio of the RX/substrate, 1.2:1; (R)-
NOBIN/substrate, 1:10; NaOH/substrate, 10-20:1; room temperature; under
b
Ar. Chemical yields of the alkylated complex were estimated after the
c
product separation from unreacted 1. Determined by chiral GLC analysis
on chiral Chirasil-Val columns for the amino acid released from the crude
complex. In several instances, the amino acids thus obtained were
subsequently recrystallized from an i-PrOH-water mixture to give an
d
enantiopure product. The amino acid was released from a recrystallized
e
complex. The bis-alkylated complex (6%) was detected in the reaction
Figure 5. Increase in the solubility of 1 in (CH2)2Cl2 affected by sodium
(R)-NOBIN-ate and racemic NOBIN-ate.
f
mixture. Fifty percent of 31a was used as catalyst.
acids obtained by decomposition of the latter complexes was
determined by chiral GLC. While PBP (7) was easily recovered
in most cases, the recovery of the catalysts was attempted only
in the case of 32b (70% yield); its reuse did not lead to any
noticeable loss of the ability to effect asymmetric induction.
TADDOL 30 did catalyze the reaction but the asymmetric
induction was relatively low at the beginning of the reaction
(40% ee by extrapolation to 0% conversion) and progressively
decreased as the reaction proceeded. As can be seen from the
experimental data summarized in Table 5, unsubstituted NOBIN
31a still catalyzed the addition of 1 to methyl acrylate (37a),
but gave poor asymmetric induction even at low temperatures
(Table 5, entries 1-3). In addition, the sense of chirality of the
product was reversed, as compared with the alkyl halide
alkylations catalyzed by 31a (Table 5, entries 1-3; compare
with Tables 3 and 4).
N-Formyl NOBIN 31f was still a poor asymmetric catalyst
(Table 5, entry 4), but the N-acetyl derivative 31g exhibited an
improved asymmetric induction with the same sense of chirality
of the products for both alkyl halide alkylation and Michael
addition reactions (Table 3, entry 19; Table 5, entry 5). N-BOC-
NOBIN 31i was less efficient than 31g but still retained some
asymmetric catalytic efficiency (Table 5, entry 6).
The behavior of iso-NOBIN 32a turned out to be similar to
that of NOBIN 32a, with low asymmetric efficiency and the
sense of chirality of the Michael adduct opposite to that of the
alkyl halide alkylation product (compare Table 5, entry 7, with
Table 3, entry 21). By contrast, N-acyl derivatives of iso-NOBIN
Figure 3. CD curves of Ni-PBP-(R)-Phe (99% ee), negative Cotton effect
at 538 nm, and that of Ni-PBP-(S)-Glu (91% ee), positive Cotton effect
at 536 nm.
Asymmetric Michael addition of substrates 1, 2, and 4 to
Michael acceptors, catalyzed by TADDOL, NOBIN, iso-NOBIN
(30-32), and their derivatives was conducted in CH2Cl2 in the
presence of a base (NaH or NaOH; Scheme 6). The resulting
alkylated complexes could be isolated from the reaction mixture
by chromatography and analyzed as such or directly decomposed
without prior isolation. The enantiomeric purity of the amino
9
12866 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Synthesis of R-Amino Acids
A R T I C L E S
Scheme 6. Asymmetric Michael Additiona
reacted very slowly, affording the Michael adduct in 9% yield
with inferior enantioselectivity (31% ee) after 240 min (Table
5, entry 12). The enantiomeric purity of the product can be
increased up to 88% ee by carrying the reaction of 4 in toluene
(Table 5, entry 11).
Other Michael acceptors included methyl metacrylate (37b),
which reacted with 1 to produce the (S,R)/(S,S) isomers in ca.
7:1 ratio when catalyzed by 32b (Table 5, entry 19). Under the
same conditions, acrolein and acrylamide were found to be
unsuitable substrates, with low ee (Table 5, entries 20 and 25).
Another substrate, 37d, having the NH protons of acrylamide
replaced by alkyl groups, reacted with 1 to give the Michael
adduct with good enantioselectivity (80, 86, and 75% ee), when
32b, 32d, and 32f, respectively, were employed as catalysts
(Table 5, entries 21-23).
Attempts to carry out a C-C bond forming cascade by
trapping the intermediate enolate with benzyl bromide or
aldehydes failed, as only the products of the initial Michael
addition were found in the solutions. A positive nonlinear effect
was also observed in the Michael addition of 1 to 37a promoted
by 32b (Figure 6).
1H NMR and IR Analysis of the Mixtures of 1 and Sodium
Salt of NOBIN (31a) and N-Acetyl-NOBIN (31g). Figure 7
illustrates the changes that occur in the 1H NMR spectra of both
1 and sodium NOBIN-ate in CD2Cl2 when mixed at a 1:1 and
1:3 ratio, respectively. The most salient features of the spectra
are the significant shifts of almost all the protons of sodium
NOBIN-ate, with one of the protons shifted by 0.4 ppm to lower
fields (from 6.2 to 6.6 ppm). The chemical shifts of the protons
of 1 were influenced to a lesser extent, but they all became
broadened. At a 1:3 ratio of 1 to sodium NOBIN-ate, the
broadening became even more evident. By contrast, very few
changes were observed in the spectra of mixtures of 1 and the
sodium salt of 31g even at a 1:3 ratio.
a
For conditions and results, see Table 5.
An IR study of the solutions of 31a, the sodium salt of 31a,
and their mixtures with 1 was conducted to assess the mutual
interactions. The IR spectrum of NOBIN 31a in CCl4 exhibits
one band for ν(OH) at 3527 cm-1 and two bands (νas and νs) at
3482 cm-1 and at 3394 cm-1 for the NH2 group. The NH
stretches have the same values as those for â-naphthylamine
whereas the ν(OH) is 81 cm-1 lower compared to that for
â-naphthol. On decreasing the concentration by 3 orders of
magnitude, no band at higher frequency could be attributed to
the free OH group. The low-frequency shift of ν(OH) and the
lack of concentration dependence indicate that the OH group
in NOBIN is involved in an intramolecular H-bonding. Note
that if this were an OH‚‚‚N or NH‚‚‚O hydrogen bond, the NH
stretches would also differ from those for â-naphthylamine. In
addition, this type of H-bonding could hardly take place because
the naphthyl moieties in NOBIN are not coplanar. Thus, it seems
reasonable to assume that the OH group of NOBIN is H-bonded
to one of the neighboring naphthalene π-systems.11 In contrast
to free NOBIN, sodium NOBIN-ate in the solution is associated
via intramolecular coordination of the NH groups with the
sodium cations as demonstrated by the IR spectrum of sodium
NOBIN-ate in dichloromethane, in which the two bands for NH
stretches appeared at 3406 and 3324 cm-1, i.e., by 76 and 70
cm-1 lower, respectively, than those for pure NOBIN (vide
supra).
32b-d and 32f proved to be highly efficient catalyst in the
addition of 1 to methyl acrylate (37a) (Table 5, entries 8, 13,
14, and 16), with the ee of the product as high as 96%. There
is a tendency toward increased enantioselectivity of the catalyst
as the size of the side chain of the acyl moiety becomes larger
with the ee increased from 90-94% ee for N-acetyl derivative
32b to 96% for the catalysis by N-pivalyl derivative 32d.
However, further increase in the steric bulk of the acyl moiety,
as in the adamantyl derivative 32f, reduced the ee to 84% (Table
5, entry 16). Unfortunately, partial racemization accompanied
the reaction catalyzed by 32b (and most likely those catalyzed
by all the other derivatives of NOBIN and iso-NOBIN), with
the ee of the product falling from 90-94% to 83% and 68% as
the reaction time was increased from 2 min to 20 and 60 min
(Table 5, entries 8, 9, and 10). Replacement of the N-acyl group
by the N-p-toluenesulfonyl moiety (32e) resulted in a dramatic
decrease in the effectiveness of the catalyst, with only 8% ee
and the chemical yield of 75% after 40 min (Table 5, entry
15). Finally, 32g, a modification of 32b in which the NHAc
group was replaced by Br, was still an efficient chemical catalyst
(as was 2-naphthol) although the ee of the reaction was only
13% (Table 5, entry 18).
The addition of 4 to methyl acrylate was also catalyzed by
32b and 32f to give the corresponding complexes of R-meth-
ylglutamic acid with 30-39% ee (Table 5, entries 11 and 17).
Under similar conditions, with 32b as catalyst, substrate 2
(11) Iogansen, A. V. Spectrochim. Acta, Part A 1999, 55, 1585.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12867

A R T I C L E S
Belokon et al.
Table 5. Asymmetric Michael Addition of Ni(II) Complexes 1, 2, and 4 (Scheme 6)a
Ni(II)
complex
base
(mol %)
acrylic acid
derivative
time
(min)
yield
(%)
ee of amino acids
(configuration)
entry
catalyst
1
2b
3c
4
5
6
7
8
9
10
11d
12
13
14
15
16
17
18
19
20
21
22
23
24
25
1
1
1
1
1
1
1
1
1
1
4
2
1
1
1
1
4
1
1
1
1
1
1
1
1
(R)-31a
(R)-31a
(R)-31a
(R)-31f
(R)-31g
(R)-31i
(R)-32a
(R)-32b
(R)-32b
(R)-32b
(R)-32b
(R)-32b
(S)-32c
(S)-32d
(R)-32e
(R)-32f
(R)-32f
(S)-32g
(R)-32b
(R)-32b
(R)-32b
(S)-32d
(R)-32f
(R)-31a
(R)-32a
NaH (10)
NaH (10)
NaH (10)
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37a
37b
37c
37d
37d
37d
37d
37e
3
30
600
6
3
2
40
15
5
26 (S)
33 (S)
45 (S)
2 (R)
55 (S)
30 (S)
13 (S)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaH (100)
NaOH (100)
NaOH (100)
NaOH (100)
NaOH (100)
NaOH (100)
65
50
70
50
70
96
98
52
9
80
80
75
88
62
80
60
50
70
90
85
80
85
3
2
90-94 (S)
83 (S)
20
60
2
240
4
4
40
4
15
2.5
9
300
3
68 (S)
39 (S) [88 (S)]d
13 (S)
90 (R)
96 (R)
8 (S)
84 (S)
32 (S)
13 (S)
61 (S,R), 54 (S,S)e
0
80 (S)
86 (R)
75 (S)
3
3
3
60
∼10 (S)
∼10 (R)
a
The substrate 0.07-0.1 M, at ambient temperature (unless indicated otherwise); catalyzed by 15 mol % of NOBIN, iso-NOBIN, and their derivatives
b
c
d
e
in CH2Cl2. Run at -5 °C. Run at -78 °C. In brackets are the results obtained in toluene as solvent within 5 min; chemical yield 20%. The ratio of
(S,R)/(S,S)-isomers was 7:1. The enantioselectivity shown here is lower than that reported by us in a preliminary communication.7b The present figure is the
lowest one observed in several experiments and is likely to originate from partial racemization.
to the spectra of the initial substances. In the range of CO
stretches, the intensity of νasym of the carboxyl group gradually
decreased with the increase in the NOBIN-ate excess. All
changes in the IR spectra were reversible. The IR spectra were
also measured in CD2Cl2 with a 3:1 sodium NOBIN-ate to
complex 1 ratio in parallel with the 1H NMR experiment.
Analysis of the IR spectra can be summarized as follows: (1)
There is no interaction between the NH2 group of NOBIN-ate
and any fragment of complex 1. (2) The frequency of the νasym
mode of the carboxyl group decreases in the presence of
NOBIN-ate. Unfortunately, the solvent and NOBIN-ate absorp-
tion mask a large part of the spectrum, and therefore, it is
difficult to determine the position of the new band. (3) The bands
of the CH2 group at 2927 and 2855 cm-1 were still observed in
the spectrum of the mixture, indicating that the CH2 group of
Figure 6. Positive nonlinear effect in the Michael addition of acrylic ester
37a to 1 catalyzed by (R)-8-acetamido-2′-hydroxy-1,1′-binaphthyl (32b).
complex 1 was not involved in the interaction with NOBIN-
ate.
The IR spectrum of the Ni substrate 1 in the region of CO
Discussion
stretches exhibits two bands: the band at 1645 cm-1 is assigned
Evidently, the glycine-derived complex 1 is a very convenient
to amide I mode, while the other band at 1675 cm-1 is assigned
substrate for the synthesis of racemic R-amino acids, retaining
to νasym of the carboxyl group (νsym of the carboxyl group is
one R-proton of the original glycine moiety. Simple alkylation
observed at 1329 cm-1). The difference between νasym and νsym
of 1 under PTC conditions with alkyl halides of varying
modes is 346 cm-1, which is typical for monodentate coordina-
activities resulted in the selective formation of mono-alkylated
tion of the ionized carboxylic group with metal cations.12
products, from which a set of racemic R-amino acids could be
released by decomplexation (Table 2, entries 1-5). As expected,
steric hindrance imposed by the phenyl substituent at the CdN
moiety of 1 provided for much more efficient mono-alkylation
of the glycine moiety, as compared to 3 (Table 2, compare
entries 3 and 5 with 10 and 11).
O’Donnell discussed in detail similar differences in the
The IR spectra of the mixtures of free NOBIN and 1 in
dichloroethane or dichloromethane in a wide range of ratios of
complex 1 to NOBIN did not provide evidence for any
interaction between the two compounds.
The spectra of mixtures of sodium NOBIN-ate and 1 at
various ratios were recorded in dichloroethane and compared
behavior of the Schiff bases of glycine esters of benzophenone
and benzaldehyde and attributed the predominant mono-alkyl-
ation of benzophenone substrate to the diminished CH acidity
(12) (a) Deacon, G. B.; Phillips, R. J. Coord. Chem. ReV. 1980, 33, 227. (b)
Maslowsky, E. Vibrational Spectra of Organometallic Compounds;
Wiley: New York, 1977. (b) Dance, N.; McWhinnie, W. R. J. Organomet.
Chem. 1976, 104, 317.
9
12868 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Synthesis of R-Amino Acids
A R T I C L E S
Figure 7. 1H NMR spectra of 1, sodium salt of NOBIN, and mixtures of both at ratios 1:1 and 1:3 (from the bottom to the top).
of its mono-alkylated products.2b The comparison of the X-ray
structures of 1 and 2 (Figures 1 and 2) clearly indicated an
interaction of the methyl substituent of the alanine moiety with
the adjacent phenyl group of the benzophenone fragment,
causing a significant additional puckering of the chelate rings
and a certain degree of rotation of the Ph relative to the
coordination plane of the complex. Here, the methyl group and
R-proton occupy a pseudoaxial and pseudoequatorial position,
respectively. Clearly, in addition to the probable decrease in
the R-CH acidity of the amino acid moiety in the mono-alkylated
complexes derived from 1, further steric hindrance to the
approach of the second alkylating agent also contributes to the
predominant mono-alkylation of 1.
appears to be a plausible explanation. Most likely, the
mechanism of the reaction includes the formation of the
phenolate of â-naphthol on the surface of the solid NaOH.16
The sodium phenolate could penetrate the layers of the
molecules of 1 in the crystals, thereby compensating the crystal
lattice-packing forces by the lipophilic arene-arene interactions;
additional interactions with the central nickel may also be
involved. These effects can be assumed to work in synergy,
which would result in the extraction of the otherwise insoluble
1 into CH2Cl2. In the next step, 1 is deprotonated to form an
ion pair containing a molecule (or molecules) of naphthol, which
occurs in the solution, where the subsequent alkylation takes
place.
As expected, complex 3, lacking such severe intramolecular
interactions, was a convenient substrate for double alkylation,
allowing the synthesis of achiral R-amino acids with a quater-
nary carbon atom carrying two identical R-substituents (Table
2, entries 10 and 11).
Rather unexpectedly, â-naphthol was found to be a very
efficient phase-transfer catalyst for the alkylation of 1, in fact
even more efficient than Bu4NBr(Cl).13 As the pKa of phenol
equals 18 in DMSO14 and those of Ni-BPB-Gly complexes of
similar structure to 1 lie in the range of 17-19,15 the formation
of a lipophilic phenolate base functioning as a PTC catalyst
Presumably, NOBIN, iso-NOBIN, and their derivatives (31
and 32) function in the same way as â-naphthol, with the
additional advantage of being chiral and capable of chelation
(rather than simple coordination) of the sodium cation in the
transition state of alkylation. This concept is supported by the
observed difference in the enantioselectivity of the reaction as
the cation of the base is changed (Table 3, entries 10-12). The
chiral ionic complex of the chiral ligand, 1, and Na+ is then
alkylated by the alkyl halide with a significant enantiofacial
selection. The role of the primary amino group of the ligand is
crucial, since its alkylation led to a dramatic decrease in the
asymmetric induction (Table 3, entries 16 and 17). These effects
strongly suggest that the NH2 group interacts with the Na+ ion.
However, direct interaction of the NH2 group with the central
(13) Similar behavior of phenolates was reported for PTC dehydrohalogenation
of alkyl halides: Ma¸kosza, M.; Chesnokov, A. Tetrahedron 2000, 56, 3553.
(14) Bordwell, F. G.; Liu, W.-Z. J. Am. Chem. Soc. 1996, 118, 10819.
(15) Terekhova, M. I.; Belokon, Y. N.; Maleev, V. I.; Chernoglazova, N. I.;
Kochetkov, K. A.; Belikov, V. M.; Petrov, E. S. Bull Acad. Sci. USSR,
DiV. Chem. Sci. (Engl. Transl.) 1986, 35, 824; IzV. Akad. Nauk, Ser. Khim.
1986, 905.
(16) We cannot exclude the presence of a molecular layer of water on the surface
of the finely ground NaOH, which may have initiated the phase-transfer
reaction.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12869

A R T I C L E S
Belokon et al.
Scheme 7. Proposed Mechanism for the Asymmetric Alkylation of
1
Ni ion of 1 or its enolate via H-bonding cannot be ruled out.
Apparently, the mechanism in the case of 1 is more complex
than that in the usual PTC alkylation. Note that alkylation of
tert-butyl N-(diphenylmethylene)glycinate with BnBr in CH2-
Cl2, catalyzed by (R)-NOBIN under the experimental conditions
defined in Table 3, led to racemic phenylalanine in 50%
chemical yield after 1 h. Similarly, alkylation of the Schiff bases
of alanine esters with BnBr under similar conditions always
led to racemic products. Less polar toluene or hexane had to
be employed in the case of these substrates to obtain good
enantioselectivities in the alkylations.6
1
The H NMR (Figure 7) and IR spectra of sodium NOBIN-
ate, 1, and their 1:1 and 1:3 mixtures in CD2Cl2 lend further
credence to the existence of special interactions of sodium
NOBIN-ate with complex 1. Apparently, both compounds
establish strong and rapidly reversible interactions with each
other, with the aromatic protons of NOBIN situated over the
Ni(II) ion, which results in a low-field shift of the signals.17
The broadening of the resonances of 1 indicates fast exchange
between several species in the solution and may originate from
the formation of a paramagnetic octahedral mixed complex of
1 and sodium NOBIN-ate, differing in geometry from the
square-planar arrangement of ligands around the Ni ion. No
significant enolate formation was observed in the mixture of
sodium NOBIN-ate and 1, as indicated by the IR data and by
almost no change in the UV-vis spectra of the mixture as
compared to 1. As the NH2 group of free NOBIN (31a) was
shown by IR not to be involved in any coordination prior to
the reaction, it seems that the phenolate oxygen atom replaced
the carboxyl group in the coordination sphere of Ni(II), and
thus, the group became partially or fully liberated, forming a
strong ionic bond with Na+. The NH2 group of NOBIN may
stabilize the enolate by formation of a hydrogen bond during
the next stage of the reaction. The importance of the latter
bonding is evidenced by the inhibitory effect of water on the
stereoselectivity of the reaction (Table 3, entry 12).
A plausible structure of the intermediate is shown in Scheme
7 for the case of sodium salt of (S)-NOBIN and 1. Although it
remains to be verified whether this species does lie on the
reaction coordinate, its reversible formation would help rational-
ize the stereochemistry of the alkyl halide alkylation catalyzed
by NOBIN. As shown in Scheme 7, the re-face of the
intermediate enolate 42 is shielded by the molecule of (S)-
NOBIN. The electrophilic attack should therefore preferentially
occur from the si-face, giving rise to the formation of (S)-amino
acids, which is consistent with the experimental observation
(Tables 3 and 4).
homochiral NOBIN molecules take part in the transition state
of the reaction, for example, one forming a mixed complex with
1 and solubilizing it, and another one removing the R-proton
from the glycine moiety of the adduct. However, in light of the
experimental results discussed here, the latter alternative appears
unlikely.
The mechanism of the Michael addition of 1 to acrylic esters
must differ from that of alkyl halide alkylation, as reversal in
the sense of chirality of the products was observed. As the pKa
of 1 in DMSO is close to 18-1915 and that of acetic esters is
close to 30,18 the Michael addition is unlikely to generate the
γ-enolate as a free species. Probably one of the functions of
NOBIN is to protonate the incipient enolate simultaneously with
the formation of the C-C bond. Failure to capture the
intermediate γ-enolate by the alkylation with added alkyl halides
or aldehydes is indicative of the absence of any free, long-lived
γ-enolates on the reaction coordinate. In stereochemical terms,
the concerted C-C and C-H bond formation mechanism
implies that the acrylate attack occurs from the same side of
the enolate where NOBIN and its acidic OH group are located,
with the inevitable reversal of the direction of attack compared
to the alkyl halide alkylations, as observed experimentally (Table
5).
N-Acyl derivatives of NOBIN 31f-h are much less efficient
catalysts in the alkylation of 1 by alkyl halides. In fact, 31f and
31h failed to catalyze the reaction entirely. On the other hand,
N-acyl derivatives of iso-NOBIN 32b-d and 32f proved to be
highly efficient catalysts for both SN2 alkylation and Michael
addition with identical absolute configurations of the products
obtained (when the same catalyst was used). As the electron-
withdrawing groups at nitrogen (31h, 32e) greatly diminished
The positive NLE (Figures 4 and 6) indicates that, most likely,
the ionized NOBIN phenolate generated heterochiral aggregates
with lower reactivity than either the homochiral aggregates or
the monomeric species. As a result, the remaining monomer
(or the homochiral aggregates) with higher ee than that of the
starting ligand becomes the active species. This amplification
is related to the reservoir effect (see the discussion in ref 10a).
Greater self-association of racemic NOBIN-ate as compared to
enantiomerically pure NOBIN-ate was clearly demonstrated in
the solubility studies of 1 in the presence of the NOBIN-ates
(Figure 5). An alternative mechanism would require that several
(17) (a) Warner, L. G.; Rose, N. J.; Busch, D. H. J. Am. Chem. Soc. 1968, 90,
6938. (b) Ito, T.; Busch, D. H. J. Am. Chem. Soc. 1973, 95, 7528.
(18) Bordwell, F.; Zhang, S.; Zhang, X.-M.; Liu, W.-Z. J. Am. Chem. Soc. 1995,
117, 7092.
9
12870 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Synthesis of R-Amino Acids
A R T I C L E S
are not required for the effective asymmetric induction, as the
observation of a large positive nonlinear effect indicated. The
alkylation procedures are very fast and, therefore, particularly
suited to the synthesis of amino acids labeled with short-lived
isotopes for PET diagnostics, as the first applications of the
protocol for the syntheses of 18F-labeled (S)-tyrosine and (S)-
DOPA has indicated.20A mechanism has been proposed to
rationalize the observed effects and the stereochemical outcome.
However, further experiments will be needed to shed more light
on this complex problem; work toward this direction is
underway in these laboratories.
Acknowledgment. The authors are grateful to Prof. D.
Seebach for kindly supplying a set of TADDOLs used in the
present project and to Dr. N. Ikonnikov for carrying out chiral
GLC analyses of the amino acids. Financial support from a
Russian Grant for Fundamental Research No. 02-03-3209, ISTC
Grant No. A-356, Swiss Grant Scopes No. 7SUPJ062125,
Descartes Prize 2001, a scholarship from DuPont during July
7, 2001-July 7, 2002 to S.R.H., a NATO Fellowship (admin-
istered by the Royal Society) to Sˇ.V., GACˇ R (Grant No. 203/
01/D051), GAUK (Grant No. 243/20B CH), and Ministry of
Education of the Czech Republic (Project MSM 113100001)
are gratefully acknowledged.
Supporting Information Available: Experimental procedures
and analytical and spectral characterization data for all com-
pounds, all NMR spectra of alkylated substrates, NMR experi-
mental procedure for mixture of complex 1 and N-Ac-iso-
NOBIN 32b, IR spectra, IR experimental procedure (all PDF),
and crystallografic information files (CIF). This material is
available free of charge via the Internet at http://pubs.acs.org.
Figure 8. ORTEP diagrams for N-acyl iso-NOBIN derivatives (R)-32b,
(S)-32d, and (R)-32f.
JA035465E
(19) Synthesis of NOBIN: (a) Smrcˇina, M.; Lorenc, M.; Hanusˇ V.; Kocˇovsky´,
P. Synlett 1991, 231. (b) Smrcˇina, M.; Lorenc, M.; Hanusˇ, V.; Sedmera,
P.; Kocˇovsky´, P. J. Org. Chem. 1992, 57, 1917. (c) Smrcˇina, M.; Pola´kova´,
J.; Vyskocˇil, Sˇ.; Kocˇovsky´, P. J. Org. Chem. 1993, 58, 8, 4534. (d) Smrcˇina,
M.; Vyskocˇil, Sˇ.; Ma´ca, B.; Pola´sˇek, M.; Claxton, T. A.; Abbott, A. P.;
Kocˇovsky´, P. J. Org. Chem. 1994, 59, 9, 2156. (e) Smrcˇina, M.; Vyskocˇil,
Sˇ.; Hanusˇ, V.; Pola´sˇek, M.; Langer, V.; Chew, B. G. M.; Zax, D. B.; Verrier,
H.; Harper, K.; Claxton, T. A.; Kocˇovsky´, P. J. Am. Chem. Soc. 1996,
118, 487. (f) Smrcˇina, M.; Vyskocˇil, Sˇ.; Pol´ıvkova´, J.; Pola´kova´, J.;
Kocˇovsky´, P. Collect. Czech. Chem. Commun. 1996, 61, 1520. (g) Smrcˇina,
M.; Vyskocˇil, Sˇ.; Pol´ıvkova´, J.; Pola´kova´, J.; Sejbal, J.; Hanusˇ, V.; Pola´sˇek,
M.; Verrier, H.; Kocˇovsky´, P. Tetrahedron: Asymmetry 1997, 8, 537. (h)
Vyskocˇil, Sˇ.; Smrcˇina, M.; Lorenc, M.; Hanusˇ, V.; Pola´sˇek, M.; Kocˇovsky´,
P. Chem. Commun. 1998, 585. (i) Vyskocˇil, Sˇ.; Jaracz, S.; Smrcˇina, M.;
Sˇt´ıcha, M.; Hanusˇ, V.; Pola´sˇek, M.; Kocˇovsky´, P. J. Org. Chem. 1998, 63,
7727. (j) Ding, K.; Xu, Q.; Wang, Y.; Liu, J.; Yu, Z.; Du, B.; Wu, Y.;
Koshima, H.; Matsuura, T. Chem. Commun. 1997, 693. (k) Mahmoud, H.;
Han, Y.; Segal, B. M.; Cai, L. Tetrahedron: Asymmetry 1998, 9, 2035. (l)
Singer, R. A.; Buchwald, S. L. Tetrahedron Lett. 1999, 40, 1095. (m) Ding,
K.; Wang, Y.; Yun, H.; Liu, J.; Wu, Y.; Terada, M.; Okubo, Y.; Mikami,
K. Chem.-Eur. J. 1999, 5, 1734. (n) Ko¨rber, K.; Tang, W.; Hu, X.; Zhang,
X. Tetrahedron Lett. 2002, 43, 7163. (o) Singer, R. A.; Brock, J. R.;
Carreira, E. M. HelV. Chim. Acta 2003, 86, 1040. For applications of
NOBIN in asymmetric catalysis, see ref 19i and the following: (p) Carreira,
E. M.; Singer, R. A.; Lee, W. J. Am. Chem. Soc. 1994, 116, 8837. (q)
Carreira, E. M.; Lee, W.; Singer, R. A. J. Am. Chem. Soc. 1995, 117, 3649.
(r) Singer, R. A.; Carreira, E. M. J. Am. Chem. Soc. 1995, 117, 12360. (s)
Brunner, H.; Henning, F.; Weber, M. Tetrahedron: Asymmetry 2002, 13,
37. (t) Tang, W.; Hu, X.; Zhang, X. Tetrahedron Lett. 2002, 43, 3075. (u)
Van Veldhuizen, J. J.; Garber, S. B.; Kingsbury, J. S.; Hoveyda, A. H. J.
Am. Chem. Soc. 2002, 124, 4954. For the corresponding methyl ether and
imines derived from it, see the following: (v) Hattori, T.; Hotta, H.; Suzuki,
T.; Miyano, S. Bull. Chem. Soc. Jpn. 1993, 66, 613. (w) Kno¨lker, H.-J.;
Hermann, H. Angew. Chem., Int. Ed. Engl. 1996, 35, 341. (x) Hungerhoff,
B.; Metz, P. Tetrahedron 1999, 55, 14941.
the activity of the catalysts, a significant negative charge on
the N-acyl oxygen atom appears to be a prerequisite for the
catalytic activity. The rigid mutual orientation of OH and RCO
groups in 32b-f (Figure 8) might facilitate efficient interactions
with the enolate of 1 in the transition state of the reaction by
simultaneous coordination to the Ni(II) center (by the amide
carbonyl) and Na+ (by the phenolate). π-Stacking interaction
of the aromatic moieties of the catalysts and the substrate may
also contribute to the overall stabilization of the transition state.
Interestingly, two rotamers A and B (Figure 8), related by a
2-fold screw axis, were observed in the crystal of acetamide
(R)-32b, with the OH and CdO oxygens pointing in the same
and in the opposite direction, respectively. By contrast, only
one rotamer was observed in each of the crystalline amides (S)-
32d and (R)-32f.
Conclusions
We have developed a new series of efficient nucleophilic
substrates for the synthesis of R-amino acids, both chiral and
achiral, employing very simple and easily reproducible modes
of operation. The amino acids can be prepared on a 100 g scale
and further scale-up does not present any difficulty. Combined
with the use of the chiral catalysts, (S)- and (R)-NOBIN,19 (S)-
and (R)-iso-NOBIN,7b and their derivatives (31 and 32), this
method opens a convenient synthetic route to the enantiomeri-
cally pure R-amino acids of both configurations. The additional
bonus of this protocol is that the enantiomerically pure catalysts
(20) (a) Fedorova, O.; Zaitsev, V.; Kuznetsova, O.; Ametamey, S.; Belokon,
Y.; Vyskocˇil, Sˇ.; Nader, M.; Schubiger, P.; Krasikova, R. Abstracts of the
9th Turku PET Symposium, May 25-28, 2002; Turku, Finland; p 17. (b)
Zaitsev, V.; Kuznetsova, O.; Fedorova, O.; Mosevich, I.; Ametamey, S.;
Belokon, Y.; Vyskocˇil, Sˇ.; Schubiger, P.; Krasikova, R. Abstracts of the
9th Turku PET Symposium, May 25-28, 2002; Turku, Finland; p R17.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12871