Dynamic kinetic resolution of hemiaminals using a novel DMAP catalyst

Article abstract of DOI:10.1016/j.tetlet.2007.11.026

We describe the first catalytic dynamic kinetic resolution of hemiaminals mediated by an organocatalyst. A 0.1-1 mol % catalyst loading is effective for the dynamic kinetic resolution of hemiaminals to produce esters up to 88% ee in high yields. A 10 mol % catalyst loading resulted in a decreased selectivity, whereas the selectivity increased at 50 °C. The absolute configuration is assigned on the basis of the empirical Cotton effect rule.

Full text of DOI:10.1016/j.tetlet.2007.11.026

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 32–35
Dynamic kinetic resolution of hemiaminals using a novel
DMAP catalyst
Shinji Yamada* and Kaori Yamashita
Department of Chemistry, Faculty of Science, Ochanomizu University, Bunkyo-ku, Tokyo 112-8610, Japan
Received 6 October 2007; revised 1 November 2007; accepted 5 November 2007
Available online 7 November 2007
Abstract—We describe the first catalytic dynamic kinetic resolution of hemiaminals mediated by an organocatalyst. A 0.1–1 mol %
catalyst loading is effective for the dynamic kinetic resolution of hemiaminals to produce esters up to 88% ee in high yields. A
10 mol % catalyst loading resulted in a decreased selectivity, whereas the selectivity increased at 50 °C. The absolute configuration
is assigned on the basis of the empirical Cotton effect rule.
Ó 2007 Elsevier Ltd. All rights reserved.
Enzymatic dynamic kinetic resolution (DKR) of sec-
alcohols by way of asymmetric acylation has attracted
significant interest because it provides optically active
esters in up to 100% yield,1 and therefore, this procedure
has been applied to a variety of organic syntheses. As
substrates for this procedure, hemiacetals and hemiami-
nals are promising candidates due to their capability of
in situ racemization through tautomerism.2,3 In addi-
tion, the product esters have been used as chiral building
blocks for various organic syntheses.4
we describe that our catalyst is also effective for the
DKR of hemiaminals. To the best of our knowledge,
this is the first example of the organocatalyst mediated
DKR of hemiaminals.
We previously reported the nonenzymatic DKR of
hemiaminals by acylation with chiral acylating
reagents.5 Moreover, we recently developed a novel
dimethylaminopyridine (DMAP) catalyst, which was
effective for the kinetic resolution of racemic sec-alco-
hols6,7 and desymmetrization of meso-diols.7 These find-
ings prompted us to investigate the catalytic kinetic
resolution of hemiaminals (Scheme 1). In this Letter,
Scheme 1. DKR of hemiaminals through asymmetric acylation.
*
Corresponding author. Tel./fax: +81
yamada.shinji@ocha.ac.jp
3 5978 5349; e-mail:
0040-4039/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.11.026

S. Yamada, K. Yamashita / Tetrahedron Letters 49 (2008) 32–35
Table 1. Catalytic dynamic kinetic resolution of 1–3
33
Entry
Substrate
Cata
Solv.
Product
Time (h)
Yieldb (%)
eec (Config.)d (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1a
1b
1c
1d
1e
1c
1c
1c
1b
1b
1b
1b
2a
2b
3
4
4
4
4
4
5
6
7
4
4
4
4
4
4
4
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
t-BuOMe
Toluene
CH2CI2
THF
c-PentylOMe
t-BuOMe
t-BuOMe
t-BuOMe
8a
8b
8c
8d
8e
8c
8c
8c
8b
8b
8b
8b
9a
9b
10
23
11
8
8
23
8
98
>99
>99
98
95
>99
99
98
99
>99
99
99
82
84 (R)
88 (R)
87
66 (R)
82 (R)
87 (R)
70 (R)
79 (R)
65 (R)
82 (R)
84 (R)
49 (R)
24 (R)
40 (R)
8
8
11
11
11
11
8
94
86
>99
24
8
a 1 mol % of catalyst was used.
b Isolated yield.
c Determined by HPLC with a chiral column.
d Absolute configuration was assigned based on the Cotton effect of the CD spectra.
The dynamic kinetic resolution of hemiaminals 1 and 2
and hemiacetal 3 by catalytic acylation with 1 mol % of
catalysts 4–7 was investigated under various reaction
conditions as shown in Table 1. The acylation of hemi-
aminals 1a–e with isobutyric anhydride and triethylamine
in the presence of catalysts 4 gave the corresponding
esters in high yields with 66–88% ee, clearly suggesting
that the DKR effectively occurred in this system (entries
1–5).8 The selectivity depends on the substituent at the
N-acyl group of the hemiaminals; the acylation of 1c
and 1d having isobutyroyl and pivaloyl groups, respec-
tively, resulted in the high selectivities (entries 3 and
4). This indicates that the steric bulkiness around the
N-acyl group would be important for the discrimination
of racemic alcohols. Among the DMAP derivatives 4–7,
4–6 serve as effective catalysts (entries 3 and 6–8). Since
a similar trend is observed in the kinetic resolution of
sec-alcohols, the importance of the C@S group for the
stereoselectivity would be a common feature in the
The absolute configurations of the product esters were
determined by a CD spectral analysis according to the
empirical rule proposed by Feringa and co-workers10
in which the signs of the Cotton effects of the n–p*
and p–p* transitions are correlated to the absolute con-
figuration. The CD spectrum of 8c shows positive n–p*
(245 nm) and negative p–p* (208 nm) Cotton effects,
which allowed assigning the R configuration to the
stereogenic center (Fig. 1). The absolute configurations
of esters 8 were all assigned in a similar manner. On
the other hand, the opposite sign pattern observed for
9 allowed assigning the R configuration.
Next, the effect of the catalyst amount on the selectivity
was investigated for 1b and 2b using catalyst 4 (Table 2).
The DKR of 1b with 0.1 mol % of catalyst 4 quantita-
tively gave 8b in 75% ee (entry 1). The reaction at 0 °C
required 139 h for completion of the reaction, but the
stereoselectivity was improved to 83% ee (entry 2). A
1 mol % loading of the catalyst significantly accelerates
the reaction and slightly increased the selectivity com-
pared to the 0.1% loading (entry 3). Interesting is the
fact that the increasing amount of the catalyst from 1
catalytic acylations with this series of catalysts.7
A
survey of solvents revealed bulky ethers to be effective
(entries 2 and 9–11).
The catalytic acylation of monocyclic hemiaminals 2a
and 2b9 gave the corresponding esters in good yields
with 49% and 24% ee, respectively (entries 13 and 14).
The relatively lower stereoselectivities in comparison
with the case of hemiaminals 1 indicated that the con-
densed aromatic ring would have a significant effect on
the enantioselectivity. It is worthwhile noting that the
catalytic DKR of hemiacetal 3 resulted in a 40% ee with
the opposite stereoselectivity compared to the case of 1,
suggesting the importance of the N-acyl moiety on the
selectivity (entry 15).
Figure 1. Determination of the absolute configurations of the products
by CD Cotton effects.

34
S. Yamada, K. Yamashita / Tetrahedron Letters 49 (2008) 32–35
Table 2. Effects of catalyst amount and temperature on the selectivity
Entry
Aminal
Cat (mol %)
Temperature
Time (h)
Yielda (%)
eeb (Config.)c (%)
1
2
3
4
5
6
7
1b
1b
1b
1b
1b
2b
2b
0.1
0.1
1
10
10
1
rt
0
72
139
9
>99
>99
96
96
>99
86
75 (R)
83 (R)
79 (R)
63 (R)
74 (R)
24 (R)
45 (R)
rt
rt
50
rt
50
2
0.5
24
24
1
90
a Isolated yield.
b Determined by HPLC with chiral columns.
c Absolute configuration was assigned based on the Cotton effect of the CD spectra.
to 10 mol % caused a significant decrease in the selectiv-
ity (entry 4). However, surprisingly, the selectivity was
much improved to 74% ee by raising the temperature
to 50 °C. A similar phenomenon was observed in the
DKR of 2b; the selectivity at 50 °C is higher than that
at rt (entries 6 and 7). These unusual observations where
the selectivity decreased with the increasing amount of
the catalyst while the selectivity increased by raising
the reaction temperature are attributable to the changes
in the relative rates between the acylation and racemiz-
ation steps. Thus, a 10 mol % catalyst loading caused
much faster acylation than the racemization of the sub-
strate, and as a result, the selectivity decreased. On the
contrary, raising the temperature to 50 °C would more
effectively enhance the racemization than the acylation,
which resulted in the increased ee.
following transition state models regarding the stable
conformations of the catalyst and the substrate (Scheme
2). In our previous studies,7 the catalyst is predicted to
have a self-complexed conformation as shown in
Scheme 2. On the other hand, the X-ray crystallographic
analysis11,12 and DFT calculations13 for 1c suggested the
existence of a hydrogen bond between the oxygen atom
of the hydroxy group and the N-acylcarbonyl group
˚
with the distance of 2.988 and 2.743 A, respectively
(Fig. 2). Each enantiomeric hemiaminal would approach
the unblocked side of the pyridinium face with a face-
to-face orientation between the pyridinium and the
aromatic rings through an intermolecular cation–p
interaction. As shown in TS-I and TS-II, the (R)-alcohol
preferentially attacks the N-acyl group rather than the
(S)-alcohol because the (S)-alcohol receives a consider-
able steric repulsion by the carbonyl and the isopropyl
group, and as a result, the (R)-ester was predominantly
produced via TS-I. This can explain the higher selectiv-
Although exact mechanism for this DKR is still unclear,
the R-selectivity in this reaction can be explained by the
Scheme 2. Plausible mechanism for DKR of 1.

S. Yamada, K. Yamashita / Tetrahedron Letters 49 (2008) 32–35
35
Kellogg, R. M.; Feringa, B. L. Tetrahedron Lett. 1997, 38,
1655–1658.
3. (a) Sharfuddin, M.; Narumi, A.; Iwai, Y.; Miyazawa, K.;
Yamada, S.; Kakuchi, T.; Kaga, H. Tetrahedron: Asym-
metry 2003, 14, 1581–1585; (b) Cuiper, A. D.; Kouwijzer,
M. L. C. E.; Grootenhuis, P. D. J.; Kellogg, R. M.;
Feringa, B. L. J. Org. Chem. 1999, 64, 9529.
4. (a) Dijkink, J.; Cintrat, J.-C.; Speckamp, W. N.; Hiemstra,
H. Tetrahedron Lett. 1999, 40, 5919–5922; (b) Cuiper, A.
D.; Kellogg, R. M.; Feringa, B. L. Chem. Commun. 1998,
655.
5. Yamada, S.; Noguchi, E. Tetrahedron Lett. 2001, 42, 3621.
6. Yamada, S.; Misono, T.; Iwai, Y. Tetrahedron Lett. 2005,
46, 2239–2242.
7. Yamada, S.; Misono, T.; Iwai, Y.; Masumizu, A.; Akiy-
ama, Y. J. Org. Chem. 2006, 71, 6872–6880.
Figure 2. ORTEP drawing for 1c at the 50% probability level.
ity for hemiaminals 1b–d having both a condensed aro-
matic ring and a bulky N-acyl group.
8. A typical procedure for the dynamic kinetic resolution of
hemiaminals: To a solution of hemiaminal 1c (30 mg) and
1 mol % of catalyst 4 (447 lL of 3.1 mmol LÀ1 solution) in
1 mL of t-BuOMe were added Et3N (21 lL) and isobutyric
anhydride (25 lL) at 0 °C. The solution was stirred for 8 h
at rt. MeOH was added to the reaction mixture, and the
solution was stirred for 20 min. The reaction mixture was
concentrated and the residue was purified by preparative
TLC (hexane/AcOEt = 4/1) to give the product ester 8c
(40.0 mg). The enantiomeric excess of the ester was
determined by HPLC analysis using Daicel CHIRALPAK
AD column using 9:1 mixture of hexane and i-PrOH as an
eluent solvent.
In summary, we described the first catalytic dynamic
kinetic resolution of hemiaminals mediated by an organ-
ocatalyst. A 0.1–1 mol % catalyst loading is effective for
the DKR of hemiaminals to give esters up to 88% ee in
high yields. Controlling the relative rates between the
racemization and the acylation steps was essential to
the success of this DKR. The proposed transition state
model can explain the resulting stereochemical outcome,
in which the cation–p interaction might play a key role.
9. Bourguignon, J. J.; Wermuth, C. G. J. Org. Chem. 1981,
46, 4889–4894.
10. (a) Gawronski, J. K.; van Oeveren, A.; Deen, H. van der.;
Leung, C. W.; Feringa, B. L. J. Org. Chem. 1996, 61,
1513–1515; (b) Cuiper, A. D.; Brzostowska, M.; Gawron-
ski, J. K.; Smeets, W. J. J.; Spek, A. L.; Hiemstra, H.;
Kellogg, R. M.; Feringa, B. L. J. Org. Chem. 1999, 64,
2567–2570.
Acknowledgment
This work was supported by a Grant-in-Aid for Scien-
tific Research (B) (No. 17350046) from the Japan Soci-
ety for the Promotion of Science.
11. X-ray crystal data for 1c: C12H13NO3, M = 219.24, mono-
clinic, P21/c, a = 11.1726(6), b = 7.1147(4) c = 13.6512(6)
3
˚
References and notes
A, b = 98.083(3)°, V = 1074.36(9) A , T = 298 K, Z = 4,
Dc = 1.256 g cmÀ1. A total of 11711 reflections were
collected and 1944 are unique (Rint = 0.052). R1 and wR2
are 0.0365 [I > 2r(I)] and 0.1050 (all data), respectively.
CCDC 661154.
1. (a) Pellissier, H. Tetrahedron 2003, 59, 8291–8327; (b)
Huerta, F. F.; Minidis, A. B. E.; Backvall, J. E. Chem.
Soc. Rev. 2001, 30, 321–331; (c) Strauss, U. T.; Felfer, U.;
Faber, K. Tetrahedron: Asymmetry 1999, 10, 107–117; (d)
Ebbers, E. J.; Ariaans, G. J. A.; Houbiers, J. P. M.;
Bruggink, A.; Zwanenburg, B. Tetrahedron 1997, 53, 9417;
(e) Stecher, H.; Faber, K. Synthesis 1997, 1.
12. The distance between the hydrogen atom of HO(1) and
˚
O(3) atom is ca. 2.6 A and that for O(1) and O(3) is
2.998 A which is within a range of general O–HÁ Á ÁO
˚
hydrogen bond.
13. The DFT calculations were performed at the B3LYP/
6-31G* level by using Spartan 06’. The calculations
predicted only one stable conformer in which two
carbonyl groups have an opposite direction with a
hydrogen bonding between the hydroxy group and the
carbonyl group.
2. (a) Deen, H. van der.; Cuiper, A. D.; Hof, R. P.; van
Oeveren, A.; Feringa, B. L.; Kellog, R. M. J. Am. Chem.
Soc. 1996, 118, 3801–3803; (b) Thuring, J. W. J. F.;
Klunder, A. J. H.; Nefkens, G. H. L.; Wegman, M. A.;
Zwanenburg, B. Tetrahedron Lett. 1996, 37, 4759–4760;
(c) Heuvel, M. van den.; Cuiper, A. D.; Deen, H. van der.;