Self-association-free dimeric cinchona alkaloid organocatalysts: Unprecedented catalytic activity, enantioselectivity and catalyst recyclability in dynamic kinetic resolution of racemic azlactones

Article abstract of DOI:10.1039/b917882a

Self-association-free, bifunctional, squaramide-based dimeric cinchona alkaloid organocatalysts show unprecedented catalytic activity, enantioselectivity and catalyst recyclability in the dynamic kinetic resolution (DKR) reaction of a broad range of racemic azlactones. The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b917882a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Self-association-free dimeric cinchona alkaloid organocatalysts:
unprecedented catalytic activity, enantioselectivity and catalyst
recyclability in dynamic kinetic resolution of racemic azlactonesw
Ji Woong Lee,a Tae Hi Ryu,a Joong Suk Oh,a Han Yong Bae,a Hyeong Bin Janga and
Choong Eui Song*ab
Received (in Cambridge, UK) 1st September 2009, Accepted 6th October 2009
First published as an Advance Article on the web 16th October 2009
DOI: 10.1039/b917882a
Self-association-free, bifunctional, squaramide-based dimeric
cinchona alkaloid organocatalysts show unprecedented catalytic
activity, enantioselectivity and catalyst recyclability in the
dynamic kinetic resolution (DKR) reaction of a broad range of
racemic azlactones.
azlactones. However, all of these methods suffer from either
a narrow substrate scope and/or very long reaction times and
unsatisfactory enantioselectivity. Recently, the groups of
Berkessel7 and Connon8 demonstrated that the use of
bifunctional (thio)ureas as organic chiral catalysts can promote
the DKR reactions of a range of azlactones using an alcohol
nucleophile, affording the corresponding N-protected amino
esters. However, their catalytic activity and enantioselectivity
are still unsatisfactory for synthetic use for the preparation of
enantiomerically pure a-amino acids. Moreover, it is known
that these urea or thiourea based organocatalysts can form
H-bonded aggregates, due to their bifunctional nature, resulting
in the strong dependency of the reactivity and enantio-
selectivity on the concentration and temperature.9,10 Due to
the self-association phenomena of this type of catalysts,
the enantioselectivity generally decreases with increasing
concentration or decreasing temperature, which can hamper
their practical use.9 This may well be a general problem, since
such organocatalysts are used for a wide variety of reactions.11
Thus, it would be highly desirable to develop a new class of
highly active and enantioselective bifunctional organocatalysts
that do not self aggregate in the solution state.
The development of practical methods for the synthesis of
enantiomerically pure natural and non-natural a-amino acids
is a challenge of considerable practical importance, due to
their numerous applications in the synthesis of chiral drugs,
peptides, chiral ligands, chiral catalysts and many other
valuable target molecules. Besides the (bio)catalytic enantio-
selective approach employing prochiral substrates, the catalytic
dynamic kinetic resolution (DKR) of racemic a-amino acid
derivatives is another potentially very practical route for the
preparation of enantiomerically pure amino acid derivatives.1
Azlactones 1 are attractive substrates for the DKR process due
to their configurational lability (pKa of a-hydrogen B9, H2O,
25 1C).2 In the presence of chiral catalysts with sufficient
basicity to enable the rapid racemization of azlactones, the
asymmetric alcoholytic ring opening of racemic azlactones
leads to the formation of the enantiomerically enriched N-acyl
amino esters 3 in 100% theoretical yield (Scheme 1).
To prevent self-aggregation, we designed the squaramide-
based dimeric cinchona alkaloids 4 (Fig. 1), in the hope that
the steric bulk of the two alkaloid moieties would suppress
their self-aggregation. The squaramide-based (4a–e) dimeric
cinchona alkaloid catalysts were simply prepared in one step
by the reaction of the corresponding 9-amino-(9-deoxy)-
epi-cinchona alkaloids12 with dimethyl squarate (see ESIw).
Several strategies involving enzymes,3 Ti-based complexes,4
cyclic dipeptides,5 and chiral DMAP nucleophilic catalysts6
have been employed to bring about the efficient DKR of
Scheme 1
a Department of Chemistry, Sungkyunkwan University,
300 Cheoncheon, Jangan, Suwon, Gyeonggi, 440-746, Korea
b Department of Energy Science, Sungkyunkwan University,
300 Cheoncheon, Jangan, Suwon, Gyeonggi, 440-746, Korea
w Electronic supplementary information (ESI) available: Full experi-
mental details and spectroscopic analysis. See DOI: 10.1039/b917882a
Fig. 1 Structures of Cinchona alkaloid organocatalysts.
ꢀc
This journal is The Royal Society of Chemistry 2009
7224 | Chem. Commun., 2009, 7224–7226

View Article Online
Table 1 Catalytic DKR of the racemic valine-derived azlactone 1a
with allyl alcohola
Catalyst
(mol%)
Yieldb
(%)
eec
(%)
Main
product
Entry
T/1C
Time
Fig. 2 Effect of concentration on enantioselectivity of dimeric catalyst
4b (squares) versus that of monomeric catalyst 5a (triangles).
1
2
3
4
5
6
7
8
9
10
4a (10)
4b (10)
4b (5)
rt
rt
rt
rt
6 h
3 h
6 h
24 h
8 h
6 h
3 h
48 h
24 h
12 h
99
95
92
98
98
96
98
97
86
95
93
94
94
94
97
93
94
91
86
75
(S)-3a
(S)-3a
(S)-3a
(S)-3a
(S)-3a
(S)-3a
(S)-3a
(R)-3a
(S)-3a
(S)-3a
Furthermore, the robust nature of the squaramide-based
dimeric catalysts, (Bis-HQN-SQA (4b) and Bis-HQD-SQA
(4e)), toward chemicals allowed for a convenient ‘‘one-pot’’
process starting from the N-toluoyl racemic valine 2b, which
gave the corresponding enantio-enriched allyl ester 3b in nearly
quantitative yields and excellent enantioselectivities (97% ee
for (S)-3b and 90% ee for (R)-3b) (Scheme 2). It is noteworthy
that the product was obtained in almost pure form after the
filtration of the produced dicyclohexyl urea, followed by the
treatment of the filtrate with aqueous HCl and subsequent
evaporation of the volatiles in the organic layer. Moreover,
after a single recrystallization (from methyl cyclohexane), the
product N-toluoyl amino acid allyl ester (S)-3b was obtained
in enantiomerically pure form (>99% ee) in 91% yield.
We assumed that the unprecedented catalytic efficiency of
the squaramide-based dimeric catalysts 4 was due to the
efficient suppression of their self-aggregation. To prove this
assumption, we conducted DKR reactions of the azlactone 1a
with allyl alcohol (2 equiv.) in the presence of the hydroquinine
derived dimeric catalyst 4b (10 mol%) and the corresponding
monomeric catalysts 5a (10 mol%) at various concentrations
([1a] = 0.1–1.0 M in CH2Cl2). As depicted in Fig. 2, the
enantioselectivity of the dimeric catalyst 4b (square symbols)
was not significantly dependent on the concentration, unlike in
the case of the corresponding monomeric catalyst 5a (triangle).
On the basis of these experimental results, it is clear that the
squaramide-based dimeric catalysts 4 do not self aggregate to
any appreciable extent, as was anticipated.
4b (2)
4b (10)
4c (10)
4d (10)
4e (20)
5a (10)
5b (10)
À20
rt
rt
rt
rt
rt
a
The reactions were carried out with 1a (0.5 mmol), allyl alcohol
(2 equiv., 1 mmol) in the presence of catalyst in CH2Cl2 (1.0 mL).
c
Isolated yields after chromatographic purification. Determined by
b
chiral HPLC (see the ESIw).
To investigate the catalytic activity and enantioselectivity of
the readily synthesized, new dimeric alkaloid catalysts (4), we
initially examined the DKR reactions of the racemic valine-
derived azlactone 1a in the presence of the catalyst
(2–10 mol%) and allyl alcohol (2 equiv.) in CH2Cl2 (0.5 M).
The results are summarized in Table 1, together with the data
obtained using the previously reported monomeric catalysts,
HQN-SQA (5a)13 and HQN-TU (5b)12 (Fig. 1). As shown in
entries 1, 2, 6 and 7 in Table 1, regardless of the substituent at
the C6’ and C10’ positions of alkaloids 4a–d, the DKR
reactions of 1a proceeded unprecedentedly fast, with the
reaction being completed within 3–6 h affording the (S)-a-amino
allyl ester in nearly quantitative yields and excellent ees
(93–94% ee).14 A lower catalyst loading of up to 2 mol% still
resulted in excellent enantioselectivity (94% ee, entries 3 and 4).
The enantioselectivity could be further increased to 97% ee by
lowering the reaction temperature to À20 1C (entry 5). The
non-natural (R)-configured a-amino ester (R)-3a could also
be obtained using the hydroquinidine-derived catalyst,
Bis-HQD-SQA (4e), in 97% yield and with 91% ee (entry 8).
To the best of our knowledge, this level of enantioselectivity
is unprecedented in the organocatalytic DKR of racemic
azlactones. In contrast to these results, the previously reported
cinchona-based monomeric catalysts 5a and 5b (entries 9
and 10, respectively) showed much inferior reactivity and
enantioselectivity (75–86% ee) compared to the squaramide-
based dimeric ones 4a–e.
Further direct evidence was obtained by experiments using
the DOSY (Diffusion Ordered Spectroscopy) technique, which
is regarded as an invaluable tool for studying self-association
phenomena in solution.15 The monomer/dimer (or higher
aggregates) equilibrium can be monitored by measuring the
diffusion coefficients D at different concentrations.
As shown in Table 2, the diffusion coefficients of the
monomeric catalyst, HQN-SQA (5a), significantly decreased
on increasing the concentration from 10 mM to 20 mM (from
4.33 Â 10À10 m2 sÀ1 to 3.50 Â 10À10 m2 sÀ1).17 However, in
studies using the same concentration gradient, the diffusion
Table 2 Diffusion coefficients D [10À10 m2sÀ1] (500 MHz, CDCl3,
16
25 1C) of 4b and 5a at different concentrations cM
D/10À10 m2 sÀ1 of
Bis-HQN-SQA (4b)
D/10À10 m2 sÀ1 of
HQN-SQA (5a)
cM
10 mM
20 mM
5.12
5.08
4.33
3.50
Scheme 2
ꢀc
This journal is The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7224–7226 | 7225

View Article Online
coefficients of the dimeric catalyst, Bis-HQN-SQA (4b), did
not show any appreciable dependence on the concentration
(DD = À0.04 Â 10À10 m2 sÀ1), in contrast to the monomeric
catalyst 5a. On the basis of the experimental (Fig. 2) and
DOSY (Table 2) results, it is now clear that the self-association
phenomenon is negligible in the case of the dimeric catalyst
Bis-HQN-SQA (4b).
allowing their repeated recycling without any loss of turnover time
or enantioselectivity (experimental details, see ESIw) (Scheme 3).
In summary, we developed self-association-free, bifunctional,
squaramide-based dimeric cinchona alkaloid organocatalysts
which showed unprecedented catalytic activity and the highest
levels of enantioselectivity reported to date in the dynamic
kinetic resolution (DKR) reaction of a broad range of racemic
azlactones, affording a variety of natural and non-natural
a-amino acid derivatives. We also showed that our DKR
protocol can be applied to the stereoinversion of chiral
a-amino acids. Moreover, the robust nature of these catalysts
toward chemicals allowed for a convenient ‘‘one-pot’’ process
starting from the racemic N-protected a-amino acids. Further-
more, the poor solubility of these catalysts in organic solvents
enabled their easy recovery by a simple precipitation method,
allowing their repeated recycling without any loss of turnover
time or enantioselectivity.
Having established Bis-HQN-SQA (4b) and Bis-HQD-SQA
(4e) as the highly enantioselective and self-association free
catalysts for the preparation of both enantiomers of a-amino
acids via the DKR process, we undertook to explore the scope
of the substrate and nucleophile (R3OH) (Table S-3 in ESIw).
All reactions were carried out in one-pot starting from the
N-protected racemic amino acids 2, derived from valine,
leucine, tert-leucine, 2-aminobutyric acid, 2-amino-4-pentenoic
acid, cyclohexylalanine, alanine, phenylalanine, DOPA, tyrosine
and phenylglycine in the presence of the catalyst 4b or 4e in
CH2Cl2 (0.5 M) at rt or À20 1C (for experimental details,
see ESIw). As shown in Table S-3, in all cases (entries 1–15 of
Table S-3) except phenylglycine, the squaramide-based
dimeric catalyst, Bis-HQN-SQA (4b), was insensitive to the
substituents (R1 and R2) of 2 as well as the nucleophiles
R3OH, thus allowing the DKR reaction to proceed with nearly
quantitative yields and unprecedentedly high levels of
enantioselectivity. Highly enantioenriched (R)-configured
amino esters (up to 92% ee) could also be obtained using
the hydroquinidine derived catalyst Bis-HQD-SQA (4e)
(entries 17–20 of Table S-3). It should also be noted that our
DKR protocol can be applied to the stereoinversion of chiral
a-amino acids. For example, the optically pure N-benzoyl-(R)-
valine ((R)-2a) gave the N-benzoyl-(S)-valine allyl ester
((S)-3a) with 96% ee after alcoholytic DKR with catalyst
Bis-HQN-SQA (4b) (entry 22 of Table S-3). The corresponding
R-isomer was also obtained in 91% ee with the catalyst
Bis-HQD-SQA (4e) from N-benzoyl-(S)-valine ((S)-2a)
(entry 21 of Table S-3). Our DKR products, N-protected amino
esters, was also successfully transformed into more valuable
N-protected amino acids without racemization by hydrolysis
with LiOH, showing the utility of the DKR products.
We gratefully acknowledge the KRF-2008-005-J00701
(MOEHRD), NRF-20090063002 (SRC program) and
R31-2008-000-10029-0 (WCU program) for financial support.
Notes and references
1 (a) K. Faber, Chem.–Eur. J., 2001, 7, 5004–5010; (b) H. Pellissier,
Tetrahedron, 2003, 59, 8291–8327; (c) N. J. Turner, Curr. Opin.
Chem. Biol., 2004, 8, 114–119.
2 J. De Jersey and B. Zerner, Biochemistry, 1969, 8, 1967–1974.
3 S. A. Brown, M.-C. Parker and N. J. Turner, Tetrahedron: Asymmetry,
2000, 11, 1687–1690 and references therein.
4 K. Gottwald and D. Seebach, Tetrahedron, 1999, 55, 723–738.
5 L. Xie, W. Hua, A. S. C. Chan and Y.-C. Leung, Tetrahedron:
Asymmetry, 1999, 10, 4715–4728.
6 J. Liang, J. C. Ruble and G. C. Fu, J. Org. Chem., 1998, 63,
3154–3155.
7 (a) A. Berkessel, F. Cleemann, S. Mukherjee, T. N. Muller and
¨
J. Lex, Angew. Chem., Int. Ed., 2005, 44, 807–811; (b) A. Berkessel,
S. Mukherjee, F. Cleemann, T. N. Muller and J. Lex, Chem.
Commun., 2005, 1898–1900.
¨
8 A. Peschiulli, C. Quigley, S. Tallon, Y. K. Gun’ko and
S. J. Connon, J. Org. Chem., 2008, 73, 6409–6412.
9 H. S. Rho, S. H. Oh, J. W. Lee, J. Y. Lee, J. Chin and C. E. Song,
Chem. Commun., 2008, 1208–1210.
´ ´ ´ ´
10 G. Tarkanyi, P. Kiraly, S. Varga, B. Vakulya and T. Soos,
Chem.–Eur. J., 2008, 14, 6078–6086.
11 (a) S. J. Connon, Chem. Commun., 2008, 2499–2510; (b) X. Yu and
W. Wang, Chem.–Asian J., 2008, 3, 516–532.
´ ´
12 B. Vakulya, S. Varga, A. Csampai and T. Soos, Org. Lett., 2005, 7,
Finally, the recyclability of the catalysts was also examined.
The squaramide-based dimeric cinchona alkaloid catalysts
4a–e are poorly soluble in all organic solvents,14 and
thus could readily be recovered using a simple precipitation
method. Upon the completion of the reaction, the addition of
hexane induced the complete precipitation of the catalysts,
1967–1969.
13 The monomeric hydroquinine squaramide 5a was prepared according
to the reported procedure for cinchonine analogue; J. P. Malerich,
K. Hagihara and V. H. Rawal, J. Am. Chem. Soc., 2008, 130,
14416–14417.
14 Although the squaramide-based dimeric catalysts are poorly
soluble in all organic solvents, to our delight, they dissolved into
the clear solution during the reaction.
15 (a) C. S. Johnson Jr, Prog. Nucl. Magn. Reson. Spectrosc., 1999, 34,
203–256; (b) Magn. Reson. Chem., ed. G. Morris, 2002, 20, S2
(special issue on ‘NMR and diffusion’).
16 Diffusion coefficient value (D) was corrected by employing TMS as
an internal standard. See also; E. J. Cabrita and S. Berger,
Magn. Reson. Chem., 2001, 39, S142–S148.
17 Due to the poor solubility of squaramide catalysts in organic
solvents, we were not able to determine the diffusion coefficients
at higher concentrated conditions.
Scheme 3 Catalyst recycling experiments.
ꢀc
This journal is The Royal Society of Chemistry 2009
7226 | Chem. Commun., 2009, 7224–7226