Optical resolution of acyclic α-hydroxy ketone derivatives by inclusion complexation

Article abstract of DOI:10.1246/bcsj.77.2051

A new method for the preparation of optically active acyclic α-hydroxy ketone derivatives by thermodynamic resolution using a chiral host compound is described. We examined the resolution of racemic 2-benzyloxy-3-pentanone with chiral host compounds in va

Full text of DOI:10.1246/bcsj.77.2051

Ó 2004 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 77, 2051–2056 (2004) 2051
Optical Resolution of Acyclic ꢀ-Hydroxy Ketone Derivatives
by Inclusion Complexation
ꢀ
Kazutsugu Matsumoto, Tomomi Okamoto, and Keiko Otsuka
Department of Chemistry, Meisei University, Hodokubo 2-1-1, Hino, Tokyo 191-8506
Received May 7, 2004; E-mail: mkazu@chem.meisei-u.ac.jp
A new method for the preparation of optically active acyclic ꢀ-hydroxy ketone derivatives by thermodynamic re-
solution using a chiral host compound is described. We examined the resolution of racemic 2-benzyloxy-3-pentanone
with chiral host compounds in various solvents. After optimization of the reaction conditions, it has become apparent
that the optical resolution could occur only under certain strict conditions about solvents, host compounds, and the con-
centration of the substrate. Stirring the suspension of the ketone with (ꢁ)-TADDOL bearing a cyclohexyl ring, derived
from ethyl L-tartrate, in MeOH–H2O (1:1) at room temperature produces the inclusion complex with high enantioselec-
tivity to give (S)-ketone with up to 99% ee. On the other hand, the ee of (R)-ketone obtained from the filtrate could be
improved by the same procedure using (þ)-TADDOL; finally, both enantiomers of the ketone as an almost optically pure
form were afforded. Analysis of the formed complex by ESI-TOFMS supposed that the host–guest ratio would be 1:1.
This resolution procedure was also applied to other substrates to afford the corresponding optically active ketones.
Optically active acyclic ꢀ-substituted ketones are very im-
portant synthons in organic synthesis. Naturally occurring ꢀ-
Results and Discussion
hydroxy and amino acids are important chiral sources for such
classes of compounds, but are hardly sufficient to supply a va-
riety of compounds. In previous papers, we reported on en-
zyme-mediated kinetic resolution via the hydrolysis of enol es-
ters as masked ketones.1 Although the reaction directly gives
optically active acyclic ꢀ-substituted ketones, it does not al-
ways satisfactory work in terms of moderate enantioselectivity
and a difficulty of regioselective preparation of the enol form
as the substrate.
On the other hand, the thermodynamic resolution through
the formation of inclusion complexes with chiral host com-
pounds has been studied as an attractive approach to obtaining
optically active compounds.2–4 In particular, separations have
been accomplished with compounds bearing amino and hy-
droxy groups, which can readily undergo hydrogen bonding
with host compounds. Recently, a few studies of the optical
resolution of ꢀ-substituted cycloalkanones have been report-
ed,5,6 and thus we have noticed that the method can be poten-
tially useful for the preparation of optically active acyclic ke-
tones, although the stereocontrol of the acyclic compounds
should be more difficult than that of the cyclic ones.
We now report on a useful procedure for the preparation of
optically active acyclic ketones with an ꢀ-substitution by op-
tical resolution using the inclusion complexation methodology
(Scheme 1).
A racemic 2-benzyloxy-3-pentanone (dl-1) was chosen as
the representative guest substrate, which was easily prepared
from racemic lactic acid, as shown in Scheme 2.
We first examined the resolution of dl-1 with a (ꢁ)-
TADDOL derivative bearing a cyclohexyl ring ((4R,5R)-(ꢁ)-
2,2-pentamethylene-ꢀ,ꢀ,ꢀ0,ꢀ0-tetraphenyl-1,3-dioxolan-4,5-
dimethanol, (ꢁ)-4a),7 derived from ethyl L-tartrate, as a typical
chiral host compound for inclusion complexation (Scheme 3).
After a suspension of dl-1 (ca. 100 mg; sub. conc., 50 mM) and
(ꢁ)-4a in a solvent (10 mL) was stirred for 72 h at room tem-
perature, the solid was filtered. Then, after dissolving the solid
into AcOEt, the host and guest were separated by column chro-
matography on silica gel. The results are summarized in
Table 1. While 1 was not obtained from the solid in the cases
of the reactions using hydrophobic (hexane, entry 1; cyclohex-
ane, entry 2), and hydrophilic (MeOH, entry 3) organic sol-
vents, the complexation of 1 with (ꢁ)-4a occurred in an aque-
O
O
R3
OR1
chiral host
R3
R2
*
R2
OR1
the inclusion complex
Scheme 1.
O
O
O
NaH, BnBr
EtMgBr
N
N
( )-lactic acid
THF, r.t.
91%
THF, r.t.
97%
OBn
dl-1
O
OH
O
OBn
2
3
Scheme 2.
Published on the web November 11, 2004; DOI 10.1246/bcsj.77.2051

2052 Bull. Chem. Soc. Jpn., 77, No. 11 (2004)
Optical Resolution of Acyclic Ketones
O
the solid
the inclusion complex
of (−)-4a · (S)-1
OBn
(−)-4a
(−)-4a
(S)-1
dl-1
MeOH-H2O (1:1)
O
the filtrate
OBn
(R)-1
OH HO
Ph
Ph
Ph
Ph
OH
OH
OH
OH
H
H
O
O
R1 R2
(−)-4a: R1, R2 = -(CH2)5-
(−)-4b: R1, R2 = -(CH2)4-
(−)-4c: R1 = R2 = Me
(+)-5
(−)-6
Scheme 3.
Table 1. Optical Resolution of ꢀ-Hydroxy Ketone Derivative dl-1 with (ꢁ)-4aaÞ
(R)-1 from the filtrate
(S)-1 from the solid
Entry
Solvent
Yield/%
Ee/%
Yield/%
Ee/%
1
2
3
4
Hexane
Cyclohexane
MeOH
98
92
97
16
68
73
54
0
0
0
29
38
29
59
0
0
0
79
26
18
34
—
—
—
6
97
94
99
H2O
5
MeOH–H2ObÞ
MeOH–H2ObÞ
MeOH–H2Ob,dÞ
6cÞ
7
a) Unless otherwise specified, dl-1 (ca. 100 mg) was mixed with (ꢁ)-4a (1 equiv.) in a solvent
(10 mL) for 72 h at room temperature. b) MeOH/H2O ¼ 1=1. c) Using 2 equiv. of (ꢁ)-4a. d)
Using 5 mL of the mixed solvent.
ous suspension (entry 4). However, the ee of 1 recovered in
79% yield from the solid was very low (6% ee, (S)-form).
To our surprise, the highly enantioselective resolution pro-
ceeded only in a 1:1 mixed solvent of MeOH and H2O (entries
5–7). After optimization of the reaction conditions, the reac-
tion of dl-1 (ca. 100 mg; sub. conc., 100 mM) with 1 equiv.
of (ꢁ)-4a in MeOH–H2O (1:1, 5 mL) (entry 7) afforded almost
optically pure (S)-1 (34%, 99% ee) from the solid. The ee of
(S)-1 was determined by an HPLC analysis of the compound
with CHIRALCEL OJ (Daicel Chemical Industries, Ltd).
The absolute configuration was confirmed by comparing the
OH HO
Ph
Ph
Ph
Ph
H
H
O
O
(+)-4a
(R)-1
(R)-1
MeOH-H2O (1 : 1)
(59% ee)
46%, 98% ee
Scheme 4.
22
obtained optical rotation value of 1 (½ꢀꢂD ¼ ꢁ45:0 (c 0.98,
should improve the ee of (R)-1. We then examined the resolu-
tion of (R)-1 (59% ee) with (þ)-4a (Scheme 4). As expected,
the sequential experiment delivered (R)-1 with 98% ee
24
CHCl3)) with the reported one; lit.,1b ½ꢀꢂD ¼ ꢁ45:9 (c
1.16, CHCl3), (S)-form. On the other hand, (R)-1 was reason-
ably obtained from the filtrate with a moderate ee (54%, 59%
25
(½ꢀꢂD ¼ þ40:9 (c 0.92, CHCl3)) in 46% yield from the solid,
22
ee), ½ꢀꢂD ¼ þ25:1 (c 1.07, CHCl3). It is noteworthy that the
although the uncomplexed 1 with 28% ee ((R)-form) was also
recovered in 31% yield.
complexation of 1 did not occur in examinations using not on-
ly (R)-(þ)-1,10-bi-2-naphthol (5) and (S,S)-(ꢁ)-hydrobenzoin
(6), but also the other TADDOL derivatives ((ꢁ)-4b and
(ꢁ)-4c) as chiral host compounds under the same conditions.
In all cases, the chiral host was quantitatively recovered.
If the enantioselective resolution proceeded thermodynami-
cally, repeating the reaction for obtaining (R)-1 with the (þ)-
antipode 4a derived from ethyl D-tartrate instead of (ꢁ)-4a
In order to study the relationship of (S)-1 and (ꢁ)-4a, the
collected solid after the reaction was analyzed employing
ESI-TOFMS. As a result, we detected the peak of ½1 þ 4a þ
Naꢂþ ion (m=z, 721.34928; 721.35051 calcd for C46H50O6Na)
as well as the peaks of ½4a þ Naꢂþ and ½4a þ 4a þ Naꢂþ
(Fig. 1). Thus, we suppose that the ratio of (ꢁ)-4a to (S)-1
of the inclusion complex would be 1:1, although there is no

K. Matsumoto et al.
Bull. Chem. Soc. Jpn., 77, No. 11 (2004) 2053
[1+4a+Na]+
721.3
100
[4a+Na]+
529.2
O
R3
R2
OR1
75
50
25
0
[4a+4a+Na]+
1035.4
7: R1 = Benzyloxymethyl (BOM), R2 = Et, R3 = Me
8: R1 = p-Methoxybenzyl (MPM), R2 = Et, R3 = Me
9: R1 = Bn, R2 = Me, R3 = Me
Irel
10: R1 = Bn, R2 = n-Pr, R3 = Me
11: R1 = Bn, R2 = Bu, R3 = Me
12: R1 = Bn, R2 = -CH2CH2CH=CH2, R3 = Me
13: R1 = Bn, R2 = Ph, R3 = Me
14: R1 = Bn, R2 = Et, R3 = Et
Fig. 2.
200
400
600
800
1000
1200
m/z
Fig. 1.
Table 2. Optical Resolution of ꢀ-Hydroxy Ketone Derivatives with (ꢁ)-4aaÞ
Ketone from the filtrate
bÞ
Ketone from the solid
bÞ
Entry
Substrate
Yield/%
½ꢀꢂD
Ee/%cÞ
Yield/%
½ꢀꢂD
Ee/%cÞ
1
2
3
4
5
6
7
8
9
10
dl-7
(S)-7 (59% ee)
dl-8
dl-9
dl-10
(S)-10 (70% ee)
dl-11
dl-12
72
72
70
39
50
45
33
43
35
70
—
—
—
—
—
—
22 (R)
48 (S)
40 (S)
27 (S)
71 (R)
37 (S)
44 (R)
44 (R)
90 (R)
20 (R)
26
20
29
45
50
53
61
57
58
30
—
ꢁ18:0 (c 0.73)
þ35:0 (c 1.02)
þ11:2 (c 1.04)
—
ꢁ46:7 (c 0.57)
—
—
59 (S)
81 (S)
90 (R)
27 (R)
70 (S)
98 (S)
23 (S)
33 (S)
54 (S)
42 (S)
þ16:3 (c 1.00)
þ17:8 (c 1.07)
þ68:9 (c 0.97)
—
dl-13
dl-14
—
ꢁ29:1 (c 1.00)
a) Unless otherwise specified, the ketone (ca. 100 mg) was mixed with (ꢁ)-4a (1 equiv.) in MeOH–H2O (1:1, 5
mL) for 72 h at room temperature. b) Measured in CHCl3 at r.t. c) Determined by HPLC analysis (CHIRALCEL
OJ or OB-H).
conclusive evidence of the host–guest ratio.
substituent (dl-12, R2 ¼ {CH2CH2CH=CH2) did not affect
the enantioselectivity (entry 8). Interestingly, the (R)-form of
9 was obtained from the solid as well as the case of the com-
pound 8. In the case of compound dl-10, having a propyl group
at the R2 position (entry 5), the complexation proceeded with
good enantioselectivity to give (S)-10 with 70% ee from the
solid and (R)-10 with 71% ee from the filtrate in 50% yields.
After a sequential reaction for obtaining (S)-10, as in the meth-
od mentioned above, the ee of (S)-10 was up to 98% ee (entry
6). On the other hand, in the case of dl-13 (R2 ¼ Ph; entry 9),
compound (R)-13 with a high ee (90% ee) was produced from
the filtrate in 35% yield and (S)-13 with 54% ee from the solid
was yielded in 58%. The relatively high yield of 13 from the
solid is supposed to be due to the good affinity of the phenyl
group with (ꢁ)-4a.
We applied the reaction to other acyclic ꢀ-hydroxy ketone
derivatives. Reactions of substrates bearing different R1, R2,
and R3 groups (Fig. 2) with the host (ꢁ)-4a were performed
under the same conditions as in the case of dl-1; the results
are summarized in Table 2. First, we tried to change the pro-
tecting group (R1) of the hydroxy group. The complexation
of the dl-7 bearing a benzyloxymethyl group proceeded with
moderate enantioselectivity (entry 1). As expected, the sequen-
tial reaction of (S)-7 (59% ee) with (ꢁ)-4a also succeeded to
improve the ee to 81% ee (entry 2). An excellent thermody-
namic resolution was observed in the case of dl-8 bearing a
p-methoxybenzyl group to give (R)-8 with 90% ee from the
solid in 29% yield (entry 3). Although the details are not yet
clear, the presence of a methoxy group on the benzene ring
for 8 could change the enantioselective mode, and (ꢁ)-4a cap-
tured (R)-8 with the opposite absolute configuration of those of
the other substrates.
Conclusion
We have disclosed a facile method to prepare optically ac-
tive acyclic ketones bearing an ꢀ-alkoxy group by thermody-
namic optical resolution with a chiral host. The procedure is a
quite simple and the chiral source could be quantitatively re-
covered. Thus, this convenient method is expected to be poten-
tially useful in preparing optically active acyclic carbonyl
compounds, although the details of the molecular-recognition
Second, we focused on changing the R2 and R3 groups. It
was found that the carbon number of the substituents of the
substrates reflected the enantioselectivity. The reactions of
dl-9 (R2 ¼ Me), dl-11 (R2 ¼ Bu), and dl-14 (R3 ¼ Et) showed
low-to-moderate enantioselectivities (entry 4, 7, and 10, re-
spectively), and the C=C double bond at the terminus of the

2054 Bull. Chem. Soc. Jpn., 77, No. 11 (2004)
Optical Resolution of Acyclic Ketones
30.6, 71.8, 80.5, 127.7, 127.9, 128.5, 137.6, 213.5; MS m=z (EI)
192 (Mþ, 1.2%), 135 (100), 105 (47), 91 (100), 58 (100); HRMS
m=z (EI) 192.1151 (192.1150 calcd for C12H16O2, Mþ).
The other substrates were prepared by the same procedure men-
tioned above from the corresponding starting materials.
process are still unclear. Further investigations for applying the
method and the analysis of the crystal structure of the inclusion
complex are now in progress.
Experimental
2-Benzyloxymethoxy-3-pentanone (dl-7): IR (neat) 2856,
1653, 1437, 1234, 1115, 1028, 741, 700 cmꢁ1 1H NMR (300
;
General Procedure and Instruments.
1H (300 and 500
MHz) and 13C (75 and 125 MHz) NMR spectra were measured
on JEOL JNM AL-300 and ꢀ-500, respectively, with tetramethyl-
silane (TMS) as the internal standard. IR spectra were recorded
with a Shimadzu IR Prestige-21 spectrometer. Mass spectra were
obtained with a JEOL EI/FAB mate BU25 instrument (EI meth-
od) and a JEOL AccuTOF MS-50010BU instrument (ESI-
TOFMS). The optical rotations were measured with a Jasco
DIP-1000 polarimeter. HPLC data were obtained on Shimadzu
LC-10ADVP, SPD-10AVP, and sic 480II data station (System In-
struments Inc.). Kieselgel 60 F254 Art.5715 (E. Merck) was used
for analytical TLC. Preparative TLC was performed on a Kiesel-
gel 60 F254 Art.5744 (E. Merck). Flash column chromatography
was performed with Silica Gel 60N (63–210 mm, Kanto Chemical
Co. Inc.). The melting points were obtained on a Yanako melting-
point apparatus, and were not corrected. The hosts ((ꢁ)-4a, (ꢁ)-
4b, and (ꢁ)-4c) were prepared from diethyl L-tartrate by the same
procedure as that reported.7 The compound (þ)-4a was derived
from D-tartaric acid. The spectral data of these compounds were
in full agreement with the reported data.7 On the other hand, com-
pounds (þ)-5 and (ꢁ)-6 were purchased from Wako pure Chemi-
cal Industries, Ltd. and Kanto chemical Co. Inc., respectively. All
other chemicals were also obtained from commercial sources.
Preparation of Racemic 2-Benzyloxy-3-pentanone (dl-1).
Under an argon atmosphere, to a suspension of NaH (60% in
oil, 830 mg, 20.8 mmol) in THF (30 mL) were added a solution
of 2-hydroxy-1-(morpholin-4-yl)-1-propanone (2, 2.99 g, 18.8
mmol) in THF (30 mL) and benzyl bromide (2.0 mL, 18.8 mmol)
at 0 ꢃC. The mixture was stirred for 2.5 h at room temperature, and
the reaction was stopped with 0.1 M phosphate buffer (pH 6.5).
The products were extracted with AcOEt (ꢄ 3), and the organic
layer was washed with brine, and dried over Na2SO4. After evap-
oration under reduced pressure, the residue was purified by flash
column chromatography (hexane/AcOEt = 1/1 ! AcOEt) to
give 2-benzyloxy-1-(morpholin-4-yl)-1-propanone (3) as a color-
less oil (4.26 g, 91%); IR (neat) 2857, 1645, 1435, 1271, 1234,
MHz, CDCl3) ꢁ 1.04 (3H, t, J ¼ 7:5 Hz, CH3CH2), 1.34 (3H, d,
J ¼ 7:0 Hz, CH3CH), 2.46–2.66 (2H, m, CH3CH2), 4.21 (1H, q,
J ¼ 7:0 Hz, CH3CH), 4.62 (2H, s, OCH2O), 4.77 (1H, d, J ¼
11:5 Hz, PhCHH), 4.83 (1H, d, J ¼ 11:5 Hz, PhCHH), 7.24–
7.40 (5H, m, Ph); 13C NMR (75 MHz, CDCl3) ꢁ 7.3, 17.6, 31.3,
70.0, 78.1, 93.9, 127.8, 128.4, 137.5, 212.1; MS m=z (EI) 192
(30%), 165 (Mþ ꢁ CH3CH2CO, 37), 135 (100), 107 (100), 91
(100), 58 (100); HRMS m=z (EI) 222.1246 (222.1256 calcd for
C13H18O3, Mþ).
2-p-Methoxybenzyloxy-3-pentanone (dl-8): IR (neat) 2978,
1719, 1612, 1514, 1458, 1250, 1105, 1035, 822 cmꢁ1 1H NMR
;
(300 MHz, CDCl3) ꢁ 1.04 (3H, t, J ¼ 7:5 Hz, CH3CH2), 1.32
(3H, d, J ¼ 7:0 Hz, CH3CH), 2.51–2.65 (2H, m, CH3CH2), 3.80
(3H, s, CH3O), 3.92 (1H, q, J ¼ 7:0 Hz, CH3CH), 4.44 (1H, d,
J ¼ 11:5 Hz, C6H4CHH), 4.46 (1H, d, J ¼ 11:5 Hz, C6H4CHH),
6.85–6.92 (2H, m, C6H2H2CHH), 7.23–7.30 (2H, m,
C6H2H2CHH); 13C NMR (75 MHz, CDCl3) ꢁ 7.3, 17.5, 30.5,
55.3, 71.5, 80.2, 113.9, 129.4, 129.7, 159.4, 213.6; MS m=z (EI)
222 (Mþ, 22%), 178 (50), 137 (100), 122 (100), 107 (21), 91
(58), 77 (94), 58 (64); HRMS m=z (EI) 222.1250 (222.1256 calcd
for C13H18O3, Mþ).
3-Benzyloxy-2-butanone (dl-9): IR (neat) 2986, 1717, 1456,
1356, 1240, 1124, 748, 698 cmꢁ1; 1H NMR (300 MHz, CDCl3) ꢁ
1.34 (3H, d, J ¼ 7:0 Hz, CH3CH2), 2.20 (3H, s, CH3CO), 3.91
(1H, q, J ¼ 7:0 Hz, CH3CH), 4.50 (1H, d, J ¼ 11:5 Hz, PhCHH),
4.55 (1H, d, J ¼ 11:5 Hz, PhCHH), 7.25–7.39 (5H, m, Ph);
13C NMR (75 MHz, CDCl3) ꢁ 17.2, 25.0, 71.8, 80.8, 127.7,
127.9, 128.5, 137.5, 211.2; MS m=z (EI) 178 (Mþ, 16%), 135
(56), 107 (34), 91 (100), 77 (100), 58 (77); HRMS m=z (EI)
178.1039 (178.0994 calcd for C11H14O2, Mþ).
2-Benzyloxy-3-hexanone (dl-10): IR (neat) 2962, 1717, 1456,
1
1369, 1117, 739, 698 cmꢁ1; H NMR (300 MHz, CDCl3) ꢁ 0.92
(3H, t, J ¼ 7:5 Hz, CH3CH2), 1.34 (3H, d, J ¼ 7:0 Hz, CH3CH),
1.50–1.68 (2H, m, CH3CH2), 2.42–2.64 (2H, m, CH2CO), 3.93
(1H, q, J ¼ 7:0 Hz, CH3CH), 4.49 (1H, d, J ¼ 11:5 Hz, PhCHH),
4.55 (1H, d, J ¼ 11:5 Hz, PhCHH), 7.26–7.40 (5H, m, Ph);
13C NMR (75 MHz, CDCl3) ꢁ 13.8, 16.6, 17.4, 39.2, 71.8, 80.6,
127.7, 127.8, 128.5, 137.6, 212.9; MS m=z (EI, rel intensity)
205 (Mþ, 8.3%), 135 (14), 105 (52), 91 (100), 71 (44), 58 (18);
HRMS m=z (EI) 205.1211 (205.1229 calcd for C13H17O2, Mþ
ꢁ H).
1
1115, 1030, 741, 700 cmꢁ1; H NMR (300 MHz, CDCl3) ꢁ 1.45
(3H, d, J ¼ 7:0 Hz, CH3), 3.50–3.75 (8H, m, NCH2CH2O), 4.33
(1H, q, J ¼ 7:0 Hz, CH), 4.47 (1H, d, J ¼ 11:5 Hz, PhCHH),
4.59 (1H, d, J ¼ 11:5 Hz, PhCHH), 7.26–7.39 (5H, m, Ph);
13C NMR (75 MHz, CDCl3) ꢁ 17.8, 42.5, 45.6, 66.7, 67.0, 71.1,
75.3, 127.8, 127.9, 128.5, 137.4, 170.5.
Under an argon atmosphere, to a solution of 3 (2.56 g, 10.3
mmol) in THF (30 mL) was ꢃadded ethylmagnesium bromide
(1.6 M in THF, 12.6 mL) at 0 C. After stirring for 2 h at room
temperature, the mixture was poured into a sat. NH4Cl aqueous
solution and the products were extracted with Et2O (ꢄ 3). The or-
ganic layer was washed with brine and dried over Na2SO4. After
evaporation, the residue was purified by flash column chromatog-
raphy (hexane/AcOEt = 10/1) to give dl-1 as a colorless oil
(1.91 g, 97%); IR (neat) 2980, 1717, 1456, 1369, 1105, 739,
2-Benzyloxy-3-heptanone (dl-11):
IR (neat) 2959, 1717,
1456, 1369, 1117, 737, 698 cmꢁ1 1H NMR (300 MHz, CDCl3)
;
ꢁ 0.91 (3H, t, J ¼ 7:5 Hz, CH3CH2CH2), 1.24–1.49 (2H, m,
2H, CH3CH2CH2), 1.33 (3H, d, J ¼ 7:0 Hz, CH3CH), 1.49–
1.65 (2H, m, CH3CH2CH2), 2.45–2.66 (2H, m, CH2CO), 3.93
(1H, q, J ¼ 7:0 Hz, CH3CH), 4.49 (1H, d, J ¼ 11:5 Hz, PhCHH),
4.51 (1H, d, J ¼ 11:5 Hz, PhCHH), 7.25–7.41 (5H, m, Ph);
13C NMR (75 MHz, CDCl3) ꢁ 13.9, 17.4, 22.4, 25.3, 37.0, 71.8,
80.6, 127.7, 127.8, 128.5, 137.6, 213.0; MS m=z (EI, rel intensity)
220 (Mþ, 2.4%), 135 (100), 114 (100), 91 (100), 85 (100), 58
(100); HRMS m=z (EI) 221.1534 (221.1542 calcd for C14H21O2,
Mþ + H).
698 cmꢁ1
;
1H NMR (500 MHz, CDCl3) ꢁ 1.05 (3H, t, J ¼ 7:5
Hz, CH3CH2), 1.35 (3H, d, J ¼ 7:0 Hz, CH3CH), 2.52–2.67
(2H, m, CH3CH2), 3.95 (1H, q, J ¼ 7:0 Hz, CH3CH), 4.50 (1H,
d, J ¼ 11:5 Hz, PhCHH), 4.54 (1H, d, J ¼ 11:5 Hz, PhCHH),
7.27–7.40 (5H, m, Ph); 13C NMR (75 MHz, CDCl3) ꢁ 7.3, 17.5,
2-Benzyloxy-6-hepten-3-one (dl-12): IR (neat) 2980, 1717,
1454, 1369, 1113, 914, 737, 698 cmꢁ1 1H NMR (300 MHz,
;

K. Matsumoto et al.
Bull. Chem. Soc. Jpn., 77, No. 11 (2004) 2055
CDCl3) ꢁ 1.34 (3H, d, J ¼ 7:0 Hz, CH3CH), 2.28–2.36 (2H, m,
CHH=CHCH2), 2.60–2.74 (2H, m, CH2CO), 3.94 (1H, q, J ¼
7:0 Hz, CH3CH), 4.50 (1H, d, J ¼ 11:5 Hz, PhCHH), 4.55 (1H,
d, J ¼ 11:5 Hz, PhCHH), 4.98 (1H, tdd, J ¼ 2:0, 2.0, 10.5 Hz,
CHH=CHCH2), 5.04 (1H, tdd, J ¼ 2:0, 2.0, 17.0 Hz,
CHH=CHCH2), 5.81 (1H, tdd, J ¼ 6:5, 10.5, 17.0 Hz,
CHH=CHCH2), 7.26–7.39 (5H, m, Ph); 13C NMR (125 MHz,
CDCl3) ꢁ 17.3, 27.2, 36.5, 71.8, 80.6, 115.2, 127.7, 127.9,
128.5, 137.2, 137.5, 2120; MS m=z (EI, rel intensity) 218 (Mþ,
5.5%), 188 (9.6), 135 (52), 112 (47), 105 (100), 91 (100), 77
(100), 56 (100); HRMS m=z (EI) 218.1295 (218.1307 calcd for
C14H18O2, Mþ).
7: column, OJ; eluent, hexane/2-propanol = 2000/1; flow
rate, 0.5 mL minꢁ1; (S) 119 min, (R) 133 min.
8: column, OJ; eluent, hexane/2-propanol = 99/1; flow rate,
0.5 mL minꢁ1; (S) 78 min, (R) 98 min.
9: column, OJ; eluent, hexane/2-propanol = 180/1; flow rate,
0.5 mL minꢁ1); (R) 39 min, (S) 61 min.
10: column, OJ; eluent, hexane/2-propanol = 180/1; flow
rate, 0.5 mL minꢁ1; (R) 26 min, (S) 28 min.
11: column, OJ; eluent, hexane/2-propanol = 180/1; flow
rate, 0.3 mL minꢁ1; (R) 36 min, (S) 40 min.
12: column, OB-H; eluent, hexane/2-propanol = 92/8; flow
rate, 0.5 mL minꢁ1; (S) 14 min, (R) 31 min.
2-Benzyloxy-1-phenyl-1-propanone (dl-13): IR (neat) 2984,
2868, 1693, 1448, 1227, 1105, 961, 739, 698 cmꢁ1; 1H NMR (300
MHz, CDCl3) ꢁ 1.53 (3H, d, J ¼ 7:0 Hz, CH3CH), 4.44 (1H, d,
J ¼ 11:5 Hz, PhCHH), 4.65 (1H, d, J ¼ 11:5 Hz, PhCHH), 4.79
(1H, q, J ¼ 7:0 Hz, CH3CH), 7.24–7.37 (5H, m, PhCH2O),
7.40–7.61 (3H, m, PhCO), 8.01–8.07 (2H, m, PhCO); 13C NMR
(125 MHz, CDCl3) ꢁ 18.8, 71.6, 78.2, 127.8, 128.0, 128.4,
128.6, 128.8, 133.4, 134.9, 137.6, 200.7; MS m=z (EI, rel intensi-
ty) 240 (Mþ, 6.3%), 210 (47), 134 (99), 122 (100), 107 (72), 91
(100), 77 (100); HRMS m=z (EI) 240.1188 (240.1150 calcd for
C16H16O2, Mþ).
13: column, OJ; eluent, hexane/2-propanol = 90/10; flow
rate, 0.5 mL minꢁ1; (S) 23 min, (R) 29 min.
14: column, OJ; eluent, hexane/2-propanol = 2000/1; flow
rate, 0.5 mL minꢁ1; (R) 41 min, (S) 49 min.
The absolute configurations of 7, 8, 10, and 14 were determined
by comparing the optical rotation signs with those of the published
data.1b In other cases, the absolute configurations were determined
by comparing the optical rotation signs with those of synthesized
authentic samples, which were prepared from optically active
starting materials by the same procedure of preparing racemic
substrates.
4-Benzyloxy-3-hexanone (dl-14): Derived from 2-hydroxy-
butanoic acid; IR (neat) 2972, 1715, 1454, 1331, 1111, 737, 698
Analysis of the Inclusion Complex of (À)-4a with (S)-1 by
ESI-TOFMS.
After stirring a mixture of (ꢁ)-4a (104 mg,
cmꢁ1
;
1H NMR (500 MHz, CDCl3) ꢁ 0.95 (3H, t, J ¼ 7:5 Hz,
0.21 mmol) and (S)-1 (153 mg, 0.80 mmol) in MeOH–H2O
(1:1, 8 mL) for 72 h at room temperature, the suspension was
filtered. The methanol solution of the collected solid was injected
directly to the ESI-TOFMS system. The exact mass of the sample
was determined by the same system with PEG as an exter-
nal standard. HRMS ½1 þ 4a þ Naꢂþ ion (m=z, 721.34928;
721.35051 calcd for C46H50O6Na).
CH3CH2CH), 1.05 (3H, t, J ¼ 7:5 Hz, CH3CH2CO), 1.65–1.79
(2H, m, CH3CH2CH), 2.55 (3H, dq, J ¼ 2:5, 7.5 Hz,
CH3CH2CO), 3.75 (1H, dd, J ¼ 5:5, 7.0 Hz, CH3CH2CH), 4.44
(1H, d, J ¼ 11:5 Hz, PhCHH), 4.56 (1H, d, J ¼ 11:5 Hz,
PhCHH), 7.27–7.40 (5H, m, Ph); 13C NMR (75 MHz, CDCl3) ꢁ
7.2, 9.6, 25.3, 31.0, 72.3, 86.1, 127.76, 127.83, 128.43, 137.7,
213.7; MS m=z (EI, rel intensity) 207 (Mþ + H, 3.6%), 149
(100), 105 (28), 100 (100), 91 (199), 58 (100); HRMS m=z (EI)
206.1212 (206.1307 calcd for C13H18O2, Mþ).
We thank JEOL Ltd. for the ESI-TOFMS spectroscopic
analysis, and Material Science Research Center (Meisei uni-
versity) for the NMR analysis.
Typical Procedure of Enantioselective Thermodynamic
To a mixture of 101 mg
Resolution of dl-1 with (À)-4a.
(0.524 mmol) of dl-1 and 265 mg (0.524 mmol) of (ꢁ)-4a were
added MeOH (2.65 mL) and H2O (2.65 mL), and the suspension
was stirred for 72 h at room temperature. After filtration, the solid
was dissolved into AcOEt, and the solvent was removed under re-
duced pressure. The residue was purified by flash column chroma-
tography on silica gel (hexane/AcOEt = 20/1) to afford (S)-1
(34.3 mg, 34%, 99% ee) and (ꢁ)-4a (264 mg, quantitative). On
the other hand, the compounds were extracted from the filtrate
with AcOEt (ꢄ 3), and the combined organic layer was washed
with brine and dried over Na2SO4. After evaporation, the residue
was purified by the same procedure as mentioned above to give
(R)-1 (54.4 mg, 54%, 59% ee). The ee of 1 was determined by
HPLC analysis with CHIRALCEL OJ (Daicel Chemical Indus-
tries, Ltd.); eluent, hexane/2-propanol = 180/1; flow rate, 0.5
mL minꢁ1; retention time, 40 (R) and 50 (S) min. The absolute
configuration of 1 was determined by comparing the optical rota-
tion sign with that of the published data.1b
Enantioselective reactions of the other cases were carried out
by the same procedure. All of the spectral data (1H NMR,
13C NMR, IR, and MS) were in full agreement with those of the
racemates. The optical rotations of the obtained compounds are
given in the text and Table 2. The ee’s of the ketones were deter-
mined by HPLC analysis with CHIRALCEL OJ or OB-H (Daicel
Chemical Industries, Ltd.). The conditions of the analyses are giv-
en below:
References
1
a) K. Matsumoto and H. Ohta, Chem. Lett., 1989, 1109. b)
K. Matsumoto, N. Suzuki, and H. Ohta, Tetrahedron Lett., 31,
7159 (1990). c) K. Matsumoto, N. Suzuki, and H. Ohta, Tetra-
hedron Lett., 31, 7163 (1990).
2
Recent reviews for inclusion complexation with chiral
hosts: a) F. Toda, Aust. J. Chem., 54, 573 (2001). b) F. Toda,
Cryst. Eng. Chem., 4, 215 (2002).
3
A review for TADDOLs as chiral reagents: D. Seebach, A.
K. Beck, and A. Heckel, Angew. Chem., Int. Ed., 40, 92 (2001).
Recent representative examples for enantioselective inclu-
sion complexations: a) T. Olszewska, M. J. Milewska, M.
4
´
Gdaniec, and T. Polonski, Chem. Commun., 1999, 1385. b) K.
Nishikawa, H. Tsukada, S. Abe, M. Kishimoto, and N. Yasuoka,
Chirality, 11, 166 (1999). c) K. Tanaka, S. Honke, Z.
Urbanczyk-Lipkowska, and F. Toda, Eur. J. Org. Chem., 2000,
3171. d) J. Deng, Y. Chi, F. Fu, X. Cui, K. Yu, J. Zhu, and Y.
Jiang, Tetrahedron: Asymmetry, 11, 1729 (2000). e) H. Miyamoto,
M. Sakamoto, K. Yoshioka, R. Takaoka, and F. Toda, Tetra-
hedron: Asymmetry, 11, 3045 (2000). f) K. Tanaka, T. Fujiwara,
and Z. Urbanczyk-Lipkowska, Org. Lett., 4, 3255 (2002). g) M.
Shizuma, Y. Kadoya, Y. Takai, H. Imamura, H. Yamada, T.
Takeda, R. Arakawa, S. Takahashi, and M. Sawada, J. Org.
Chem., 67, 4795 (2002). h) O. Bortolini, M. Fogagnolo, G. Fantin,

2056 Bull. Chem. Soc. Jpn., 77, No. 11 (2004)
and A. Medici, Chem. Lett., 32, 206 (2003).
5 Optical resolutions of ꢀ-substituted cycloalkanones: a) F.
Toda, Jpn. Kokai Tokkyo Koho, JP08059539 (1996). b) F. Toda
Optical Resolution of Acyclic Ketones
Okuyama, Tetrahedron Lett., 38, 7759 (1997). b) H. Kaku, S.
Ozako, S. Kawamura, S. Takatsu, M. Ishii, and T. Tsunoda,
Heterocycles, 55, 847 (2001). c) H. Kaku, S. Takaoka, and T.
Tsunoda, Tetrahedron, 58, 3401 (2002).
and T. Shinyama, J. Chem. Soc., Perkin Trans. 1, 1997, 1759.
6 Deracemizations of ꢀ-substituted cyclohexanone deriva-
tives utilizing inclusion complexation with chiral hosts under
7 D. Seebach, A. K. Beck, R. Imwinkerlried, S. Roggo, and
A. Wonnacott, Helv. Chim. Acta, 70, 954 (1987).
basic conditions: a) T. Tsunoda, H. Kaku, M. Nagaku, and E.