Catalytic Asymmetric Aminolysis of 3,5,8-Trioxabicyclo[5.1.0]octane Providing an Optically Pure 2-Amino-1,3,4-butanetriol Equivalent

Full text of DOI:10.1021/jo9900883

4962
J . Org. Chem. 1999, 64, 4962-4965
Ta ble 1. Am in olysisa of Ep oxid es w ith Ben zyla m in e
Ca ta lytic Asym m etr ic Am in olysis of
cat.b
(mol %)
convn (%)
after 24 h
3,5,8-Tr ioxa bicyclo[5.1.0]octa n e P r ovid in g
a n Op tica lly P u r e 2-Am in o-1,3,4-bu ta n etr iol
Equ iva len t
entry
epoxide
product
ee (%)c
1
2
3
4
5
6
7
8
1a
1b
1c
1c
1c
3
1.0
1.0
1.0
0.75
0.5
1.0
1.0
1.0
2a d
2bd
2cd
2c
55.7e
79.9e
94.3e
93.7e
85.5e
<5f
20.0
92.0
93.2
90.6
86.2
Shoichi Sagawa, Hiroyuki Abe, Yasunori Hase, and
Takashi Inaba*
2c
Central Pharmaceutical Research Institute, J apan Tobacco,
Inc., 1-1, Murasaki-cho Takatsuki Osaka, 569-1125, J apan
4
5
<5f
< 5f
a
All reactions were performed using 1.0 equiv of benzylamine
Received J anuary 19, 1999
b
in the presence of 10 mol % of water in toluene at 40 °C. The
The catalytic asymmetric ring opening of meso ep-
oxides1 is becoming a powerful tool in the field of
synthetic chemistry, and there have been many reports
of success using thiols,2 carboxylic acids,3 azides,4 cya-
nides,5 halides,6 phenols,7 and anilines2b,8 as nucleophiles.
To the best of our knowledge, however, there has been
no report employing an alkylamine as a nucleophile
irrespective of the utility of the resulting chiral â-amino
alcohol for various synthetic purposes.9 A likely obstacle
to achieving a catalytic asymmetric aminolysis of meso
epoxides is deactivation of the chiral Lewis acid catalyst,
by stable complex formation with an amine used as a
nucleophile and/or with a generated â-amino alcohol,
resulting in the inability to activate the epoxide.4a Nu-
cleophiles employed previously are generally less basic
than an alkylamine. Here, we report the catalytic de-
symmetrization of 3,5,8-trioxabicyclo[5.1.0]octane deriva-
tives 1 using R-substituted or unsubstituted benzyl-
amines as nucleophiles, in which 1 coordinates to a chiral
Lewis acid so efficiently in the presence of the amines
that very low catalyst loading (<1%) is sufficient for high
ee and conversion. The reaction provides optically pure
2-amino-1,3,4-butanetriol equivalents (ABTE), versatile
chiral C4 building blocks bearing a substituted or un-
substituted amino group and three hydroxyl groups.
catalyst was composed of (S)-1,1′-bi-2-naphthol and Ti(O-i-Pr)4 (1:1
d
ratio). c Determined by HPLC. The absolute configuration was
determined as described in the Supporting Information. e HPLC
peak area ratio (2/(2+benzylamine) at 214 nm), calibrated by the
molar UV intensity ratio, was used for the conversion. f No new
peak was observed in the 300 MHz 1H NMR spectrum of the crude
extract.
When epoxides 1a -c possessing a ketal group were
treated with benzylamine in the presence of catalytic
amounts of (S)-1,1′-bi-2-naphthol and Ti(O-i-Pr)4 and 10
mol % of water10 relative to epoxide in toluene,11 smooth
reactions took place affording the corresponding amino
alcohols 2a -c (Table 1). Particularly, excellent asym-
(1) Review: Hodgson, D. M.; Gibbs, A. R.; Lee, G. P. Tetrahedron
1996, 52, 14361.
(2) (a) Yamashita, H.; Mukaiyama, T. Chem. Lett. 1985, 1643. (b)
Yamashita, H. Bull. Chem. Soc. J pn. 1988, 61, 1213. (c) Iida, T.;
Yamamoto, N.; Sasai, H.; Shibasaki, M. J . Am. Chem. Soc. 1997, 119,
4783. (d) Wu, J .; Hou, X. L.; Dai, L. X.; Xia, L. J .; Tang, M. H.
Tetrahedron: Asymmetry 1998, 9, 3431.
(3) J acobsen, E. N.; Kakiuchi, F.; Konsler, R. G.; Larrow, J . F.;
Tokunaga, M. Tetrahedron Lett. 1997, 38, 773.
(4) (a) Yamashita, H. Chem. Lett. 1987, 525. (b) Emziane, M.;
Sutowardoyo, K. I.; Sinou, D. J . Organomet. Chem. 1988, 346, C7. (c)
Hayashi, M.; Kohmura, K.; Oguni, N. Synlett 1991, 774. (d) Nugent,
W. A. J . Am. Chem. Soc. 1992, 114, 2768. (e) Martinez, L. E.; Leighton,
J . L.; Carsten, D. H.; J acobsen, E. N. J . Am. Chem. Soc. 1995, 117,
5897. (f) Adolfsson, H.; Moberg, C. Tetrahedron: Asymmetry 1995, 6,
2023. (g) Annis, D. A.; Helluin, O.; J acobsen, E. N. Angew. Chem., Int.
Ed. Engl. 1998, 37, 1907.
(5) (a) Cole, B. M.; Shimizu, K. D.; Krueger, C. A.; Harrity, J . P. A.;
Snapper, M. L.; Hoveyda, A. H. Angew. Chem., Int. Ed. Engl. 1996,
35, 1668. (b) Shimizu, K. D.; Cole, B. M.; Krueger, C. A.; Kuntz, K.
W.; Snapper, M. L.; Hoveyda, A. H. Angew. Chem., Int. Ed. Engl. 1997,
36, 1704.
(6) Nugent, W. A. J . Am. Chem. Soc. 1998, 120, 7139.
(7) Iida, T.; Yamamoto, N.; Matsunaga, S.; Woo, H. G.; Shibasaki,
M. Angew. Chem., Int. Ed. Engl. 1998, 37, 2223.
(8) (a) Fu, X. L.; Wu, S. H. Synth. Commun. 1997, 27, 1677. (b) Hou,
X. L.; Wu, J .; Dai, L. X.; Xia, L. J .; Tang, M. H. Tetrahedron:
Asymmetry 1998, 9, 1747.
(9) A catalytic amount of lanthanide(III) was reported to promote
aminolysis of epoxides and was used in the regioselective reactions,
see: (a) Chini, M.; Crotti, P.; Favero, L.; Macchia, F.; Pineschi, M.
Tetrahedron Lett. 1994, 35, 433. (b) Meguro, M.; Asao, N.; Yamamoto,
Y. J . Chem. Soc., Perkin Trans. 1 1994, 2597.
metric induction was observed in the reactions of ep-
oxides (1b and 1c) having alkyl substituents (R) on the
ketal carbon (entries 2 to 5, Table 1). In sharp contrast
to 1, when 3, 4, and 5 were used as the substrates, no
reaction was observed (entries 6-8, Table 1). Molecular
modeling studies12 of 1c and 5 provided insight into
possible reasons for this dramatic difference. The most
(10) The importance of water in the Lewis acid promoted reaction
was previously reported, see: Pitchen, P.; Dunach, E.; Deshmukh, M.
N.; Kagan, H. B. J . Am. Chem. Soc. 1984, 106, 8188. When 1c was
subjected to the similar conditions of entry 3 in Table 1 with exclusion
of water (0.01 mol %, Karl Fischer), the conversion of the reaction and
ee of product 2c were 84% and 46%, respectively.
(11) Using an aprotic solvent was crucial to prevent the noncatalyzed
reaction. No reaction was observed when the epoxides (1a , 1c, and 5)
were treated with the amines (benzylamine and 6) in the absence of
the catalyst in toluene or heptane.
(12) Computational results were obtained using software programs
from MSI of San DiegosMD and MM calculations were conducted with
the Discover program, using the CFF91 (for 1c and 5) and ESFF (for
1c+Ti) force fields. Simulated annealing calculation (via molecular
dynamics) was carried out for each compound starting from 100 initial
conformations generated by randomizing atom coordinates of the
molecules. The structure of 1c+Ti was obtained by the calculation of
the 1c-Ti(OMe)4 complex. The methoxyl groups were omitted for
clarity in Figure 1.
10.1021/jo9900883 CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/29/1999

Notes
J . Org. Chem., Vol. 64, No. 13, 1999 4963
F igu r e 1. The arrows in 5 are directions of the lone pairs. The structure 1c+Ti was obtained though the calculation of the
1c-Ti(OMe)4 complex. The four methyl groups were omitted for clarity.
Ta ble 2. Am in olysisa of 1c w ith 6
stable conformer, which was obtained through simulated
annealing calculation of 1c, was the twist-boat form
(Figure 1 1c). It was rational to assume that the structure
of 1c was suitable for interaction with a Ti atom, because
the most stable Ti-coordinated structure of 1c (Figure 1,
1c+Ti) was close to that of 1c in conformation.13 The
structure of 1c+Ti indicated that the direction of the lone
pair on one of the two ketal oxygen atoms and that of
the epoxide oxygen fitted well in the geometry of chelat-
ing titanium atom, which should assume a typical
tetragonal bipyramidal structure.14 Hence 1c was con-
sidered to preferentially coordinate to a titanium atom
in a bidentate fashion to form an activated complex of
1c+Ti leading to 2c. However, the unreactive compound
3 could not assume such a bidentate coordination because
of the absence of a supplementary oxygen atom. The
epoxide 4, which possesses both oxygen atoms could be
unreactive because it was speculated not to assume a
conformation having an effective bidentate coordination
without a severe decrease in entropy. Calculation of
unreactive 5 showed that a pseudo-chair form was the
most stable conformer in which the directions of the
oxygen lone pairs significantly deviated from the as-
sumed O-Ti bonds (Figure 1, 5+Ti). These findings allow
us to speculate that bidentate titanium complex forma-
tion with the assistance of a ketal oxygen atom is
necessary to promote the catalytic asymmetric aminoly-
sis.15 It is proposed that enantiomeric discrimination of
a
All reactions were performed with 1.0 equiv of 6 in the
one of two acetal oxygen atoms by the (S)-1,1′-bi-2-
b
presence of 10 mol % of water at 40 °C. The catalyst was
naphthol-titanium complex is responsible for the asym-
composed of (S)-1,1′-bi-2-naphthol and Ti(O-i-Pr)4 (1:1 ratio).
metric induction in the reaction of 1. When R in 1 was a
hydrogen atom (1a ), the discrimination of one of two
acetal oxygen atoms seemed to be less effective, resulting
in the lower ee of product 2a (entry 1, Table 1). This fact
implies that the alkyl substituents on the ketal carbon
favorably interact with the chiral Lewis acid to enhance
the stereoselectivity.
c HPLC peak area ratio (7/(7+6) at 214 nm), calibrated by the
d
molar UV intensity ratio, was used for the conversion. Deter-
mined by HPLC. e The absolute configuration was determined as
described in the Supporting Information. f The ee was that of crude
g
8 derived from the reaction mixture. When the reaction was
performed in the presence of 0.1 mol % of the catalyst, 67.6% de
and 6.6% conversion were obtained after 24 h.
composed of heptane-toluene (9/1) was effective in
retaining excellent conversion (entries 4, 5, and 6, Table
2). It is worth noting that the reaction of 1c with (R)-
and (S)-6c gave 7c16 and 7d , respectively, with the same
stereochemical mode at the newly formed asymmetric
centers in >97% de and excellent yield using the 0.5 mol
% catalysts (entries 5 and 6, Table 2). These results
indicated that the enantioselectivity was not affected by
the configuration of 6c but rather by that of (S)-1,1′-bi-
2-naphthol. Use of racemic 6c gave the same result.
Removal of the substituted benzyl group from the product
(a 1:1 diastereomeric mixture of 7c and 7d ) followed by
recrystallization gave optically pure 816 in 84% from
racemic 6c (entry 7, Table 2).17
With the intention of increasing the stereoselectivity,
the reaction of 1c with R-substituted and disubstituted
benzylamines (6a -c) was investigated (Table 2). Enan-
tioselectivity decreased when 1c was opened using di-
methylbenzylamine (6a , 70.8% ee) or benzhydrylamine
(6b, 89% ee). However, employment of the R-methylbenz-
ylamine ((R)- or (S)-6c) brought about an increase of
stereoselectivity (entry 3, Table 2), and using a solvent
(13) The RMS distance between the heavy atoms of 1c and 1c in
1c+Ti was 0.26 Å.
(14) Williams, I. D.; Pedersen, S. F.; Sharpless, K. B.; Lippard, S.
J . J . Am. Chem. Soc. 1984, 106, 6430.
(15) The stable conformations of the products (2c and the amino
alcohol derived from 5) indicated that there was no significant
difference in the conformation of their â-amino alcohol portions.
Therefore, a difference in the dissociation rate of Ti from the products
was ruled out as the cause of the dramatic difference in reactivity.
(16) Inaba, T.; Birchler, A. G.; Yamada, Y.; Sagawa, S.; Yokota, K.;
Ando, K.; Uchida, I. J . Org. Chem. 1998, 63, 7582.

4964 J . Org. Chem., Vol. 64, No. 13, 1999
Notes
3.88 (dd, 1H, J ) 2.2, 10.3 Hz), 3.84 (dd, 1H, J ) 2.6, 10.3 Hz),
3.81 (d, 1H, J ) 13.2 Hz), 3.77-3.64 (m, 2H), 3.61 (dt, 1H, J )
2.6, 5.1 Hz), 2.86 (dt, 1H, J ) 2.2, 5.1 Hz), 2.02 (brs, 1H); IR
(KBr) 3134, 1132, 1049 cm-1; [R]25 +10.3 (c 1.0, MeOH). Anal.
D
Calcd for C12H17NO3: C, 64.55; H, 7.67; N, 6.27. Found: C, 64.64;
H, 7.91; N, 6.28. tR(2a ) ) 8.0 min (60.0%), tR(anti-2a ) ) 10.1
min (40.0%), Daicel Chemical Industries, Ltd., CHIRAL AGP,
35 °C, 2-propanol/phosphate buffer (pH 7) ) 0.5/99.5, 214 nm,
0.5 mL/min. There was 55.7% conversion after 24 h according
to HPLC (AM-302 (achiral), 45 °C, 2-propanol/phosphate buffer
(pH 7) ) 17/83, 214 nm, 1.0 mL/min, tR(benzylamine) ) 2.1 min,
tR(2a ) ) 5.1 min). The % conversion was calibrated by the
relative molar UV intensity ratio of 0.70 (benzylamine/2a ).
(5R,6S)-2,2-Cycloh exyliden e-6-ben zylam in o-1,3-dioxepan -
5-ol (2b). This compound was similarly obtained from the
reaction of 1b with benzylamine as pale yellow crystals (82%
yield, 92.0% ee): mp 96-98 °C; 1H NMR (CDCl3) δ 7.37-7.24
(m, 5H), 3.94 (d, 1H, J ) 13.2 Hz), 3.82-3.77 (m, 3H), 3.59-
3.48 (m, 3H), 2.56 (br t, 1H, J ) 4.0 Hz), 1.90 (brs, 1H), 1.58-
In summary, excellent stereoselectivity and yield were
observed in the catalytic aminolysis of meso epoxides 1b
and 1c, which effectively coordinate to a titanium atom.
Moreover, this reaction is a practical preparation of
optically pure ABTEs, because no hazardous reagents are
used and chiral Lewis acid of 0.5 mol % is sufficient to
achieve >97% ee as well as excellent conversion. ABTE
is a chiral C4 unit, which is expected to be useful in the
production of chiral synthetic targets of biological inter-
est.18
1.39 (m, 10H); IR (KBr) 3167, 1118, 1066 cm-1; [R]25 +45.4 (c
D
1.0, MeOH). Anal. Calcd for C17H25NO3: C, 70.07; H, 8.65; N,
4.81. Found: C, 70.37; H, 9.05; N, 4.92. tR(2b) ) 11.7 min
(96.0%), tR(anti-2b) ) 15.5 min (4.0%), CHIRAL AGP, 35 °C,
2-propanol/phosphate buffer (pH 7) ) 5/95, 214 nm, 0.5 mL/min.
There was 79.9% conversion after 24 h according to HPLC (AM-
302 (achiral), 45 °C, 2-propanol/phosphate buffer (pH 7) ) 25/
75, 214 nm, 1.0 mL/min, tR(benzylamine) ) 2.0 min, tR(2b) )
24.0 min). The % conversion was calibrated by the relative molar
UV intensity ratio of 0.69 (benzylamine/2b).
Exp er im en ta l Section
Gen er a l. All melting points are uncorrected. The 1H NMR
spectra were recorded at 300 MHz. Solvents, starting materials,
and reagents were used as purchased without further purifica-
tion. Silica gel 60 (230-400 mesh, Merck) was used for column
chromatography. The enantioselectivity and diastereoselectivity
were obtained by HPLC analyses. When reactions became
heterogeneous because of the crystallization of products, CH2-
Cl2 was added to obtain clear solutions for HPLC analysis. All
HPLC conditions shown below afforded good separations for the
determination of ee and de.
(5R,6S)-2,2-Dim eth yl-6-(1-m eth yl-1-p h en yleth yla m in o)-
1,3-d ioxep a n -5-ol (7a ). This compound was similarly obtained
(42 h reaction) from the reaction of 1c with 6a as pale yellow
crystals (5% yield, 70.8% ee): mp 62-64 °C; 1H NMR (CDCl3) δ
7.49-7.46 (m, 2H), 7.33-7.18 (m, 3H), 3.92 (dd, 1H, J ) 0.9,
13.2 Hz), 3.61 (dd, 1H, J ) 1.5, 13.2 Hz), 3.46 (ddd, 1H, J ) 0.9,
4.2, 12.9 Hz), 3.32 (br t, 1H, J ) 4.2 Hz), 3.08 (dd, 1H, J ) 4.2,
12.9 Hz), 2.38 (br t, 1H, J ) 4.2 Hz), 1.83 (brs, 1H), 1.48 (s, 3H),
1.47 (s, 3H), 1.32 (s, 3H), 1.27 (s, 3H); IR (KBr) 3497, 1054, 1030
(5R,6S)-2,2-Dim et h yl-6-b en zyla m in o-1,3-d ioxep a n -5-ol
(2c). To a solution of (S)-1,1′-bi-2-naphthol (19.9 mg, 0.0694
mmol) in toluene (2.0 mL) was added titanium tetraisopropoxide
(19.7 mg, 0.0694 mmol) under a nitrogen atmosphere at room
temperature. After this mixture was stirred for 10 min, benzy-
lamine (744 mg, 6.94 mmol), water (10 mg), and 1c (1.00 g, 6.94
mmol) were successively added to the solution and the reaction
mixture was stirred at 40 °C for 24 h. There was 94.3%
conversion after 24 h according to HPLC (YMC Co., Ltd., AM-
302 (achiral), 45 °C, 2-propanol/phosphate buffer (pH 7) ) 17/
83, 214 nm, 1.0 mL/min, tR(benzylamine) ) 2.1 min, tR(2c) )
11.7 min). The % conversion was calibrated by the relative molar
UV intensity ratio of 0.72 (benzylamine/2c). The reaction
mixture was washed with aqueous 1.0 M NaOH (1.8 mL) and
concentrated to dryness. Purification by silica gel column
chromatography (hexane-AcOEt) gave 2c (1.60 g, 92% yield,
93.2% ee) as a pale yellow oil: 1H NMR (CDCl3) δ 7.33-7.23
(m, 5H), 3.92 (d, 1H, J ) 13.2 Hz), 3.82-3.76 (m, 2H), 3.78 (d,
1H, J ) 13.2 Hz), 3.59-3.49 (m, 3H), 2.56 (m, 1H), 2.04 (br s,
cm-1; [R]25 +24.6 (c 1.0, MeOH). Anal. Calcd for C16H25NO3:
D
C, 68.79; H, 9.02; N, 5.01. Found: C, 68.82; H, 9.31; N, 5.10.
There was 7.4% conversion after 42 h according to HPLC (AM-
302 (achiral), 45 °C, 2-propanol/phosphate buffer (pH 7) ) 25/
75, 214 nm, 1.0 mL/min, tR(6a ) ) 2.5 min, tR(7a ) ) 13.0 min).
The % conversion was calibrated by the relative molar UV
intensity ratio of 0.67 (6a /7a ). The compound 7a thus obtained
was converted to 919 via 8 to measure its enantiomeric purity:
tR (9) ) 22.6 min (85.4%), tR (anti-9) ) 30.5 min (14.6%), Daicel
Chemical Industries, Ltd., CHIRALCEL OD, 35 °C, hexane/
2-propanol ) 95/5, 210 nm, 1.0 mL/min.
(5R,6S)-2,2-Dim eth yl-6-d ip h en ylm eth yla m in o-1,3-d iox-
ep a n -5-ol (7b). This compound was similarly obtained (36 h
reaction) from the reaction of 1c with 6b as pale yellow crystals
(95% yield, 89.8% ee): mp 106-108 °C; 1H NMR (CDCl3) δ 7.44-
7.17 (m, 10H), 5.00 (s, 1H), 4.02 (ddd, 1H, J ) 3.0, 14.4, 17.1
Hz), 3.82 (d, 1H, J ) 11.4 Hz), 3.77 (dd, 1H, J ) 1.8, 13.2 Hz),
3.60-3.51 (m, 3H), 2.54 (dt, 1H, J ) 1.2, 4.5 Hz), 1.82 (br s,
1H), 1.33 (s, 3H), 1.32 (s, 3H); IR (CHCl3) 3416, 1219, 1047 cm-1
;
[R]25 +50.6 (c 0.5, MeOH); HRMS (Fab) calcd for C14H22NO3
1H), 1.32 (s, 3H), 1.31 (s, 3H); IR (KBr) 3345, 1218, 1050 cm-1
;
D
(M+ + 1) 252.1594, found 252.1591. tR(2c) ) 25.5 min (96.6%),
tR(anti-2c) ) 23.9 min (3.4%), Shinwa Chemical Industries, Ltd.,
ULTRON ES-PhCD, 45 °C, MeCN/phosphate buffer (pH 7) )
20/80, 214 nm, 1.0 mL/min. The compound 2c thus obtained was
converted to 9 via 8 to determine its absolute configuration.19
(5R,6S)-6-Ben zyla m in o-1,3-d ioxep a n -5-ol (2a ). This com-
pound was similarly obtained from the reaction of 1a with
benzylamine as pale yellow crystals (51% yield, 20.0% ee): mp
56-58 °C; 1H NMR (CDCl3) δ 7.34-7.24 (m, 5H), 4.75 (d, 1H, J
) 4.8 Hz), 4.73 (d, 1H, J ) 4.8 Hz), 3.94 (d, 1H, J ) 13.2 Hz),
[R]25D +42.8 (c 0.5, MeOH). Anal. Calcd for C20H25NO3: C, 73.37;
H, 7.70; N, 4.28. Found: C, 73.61; H, 8.05; N, 4.24. tR(7b) ) 16.2
min (94.9%), tR(anti-7b) ) 14.1 min (5.1%), CHIRAL AGP, 35
°C, 2-propanol/phosphate buffer (pH 7) ) 5/95, 214 nm, 0.5 mL/
min. There was 97.0% conversion after 36 h according to HPLC
(AM-302 (achiral), 45 °C, 2-propanol/phosphate buffer (pH 7) )
25/75, 214 nm, 1.0 mL/min, tR(6b) ) 10.2 min, tR(7b) ) 41.8
min). The % conversion was calibrated by the relative molar UV
intensity ratio of 0.91 (6b/7b). The compound 7b thus obtained
was converted to 9 via 8 to determine its absolute configura-
tion.19
(5R,6S)-2,2-Dim eth yl-6-[(R)-1-p h en yleth yla m in o]-1,3-d i-
oxep a n -5-ol (7c).16 To a 1 L four-necked round-bottomed flask
were added (S)-1,1′-bi-2-naphthol (993 mg, 3.47 mmol) and
heptane-toluene (9:1, 350 mL). To the suspension was added
titanium tetraisopropoxide (986 mg, 3.47 mmol) under a nitrogen
atmosphere at room temperature; after the solution was stirred
for 10 min, (R)-6c (84.1 g, 694 mmol), water (1.25 g, 69.4 mmol),
and 1c (100 g, 694 mmol) were added to the solution successively.
After stirring at 40 °C for 24 h, toluene (280 mL) was added to
(17) To avoid the problems related to the preferential hydrogenation
of some diastereomer, the completion of the reaction was confirmed
by TLC.
(18) The ABTE, 8, was used for the practical synthesis of nelfinavir,
a potent HIV protease inhibitor, see ref 16.
(19) (a) The compound 9, (5R,6S)-6-benzyloxycarbonylamino-2,2-
dimethyl-5-hydroxy-1,3-dioxepan, was obtained by treatment of 8 with
Cbz-Cl. See the Supporting Information. (b) 2c, 7a , 7b, and 7d were
converted into 8 using the similar conditions described in the prepara-
tion of 8 from the mixture of 7c and 7d .

Notes
J . Org. Chem., Vol. 64, No. 13, 1999 4965
the reaction mixture, which was then stirred at 40 °C for 1 h
and at 5 °C for 1 h. Filtration of the deposited crystals gave crude
7c as bright yellow crystals, which contained trace amounts of
(S)-1,1′-bi-2-naphthol and titanium. This material was sus-
pended in toluene (526 mL) and was heated to reflux in the
presence of 1 M aqueous sodium hydroxide (175 mL) for 1 h.
After removal of the aqueous layer while hot (>50 °C), the
organic layer was refluxed in the presence of charcoal (3.0 g)
for 1 h. The insoluble material in the mixture was removed by
filtration while hot (>50 °C), and then the filtrate was concen-
trated to dryness, affording 7c (157.1 g, 90% yield and >99%de
(HPLC)) as pale yellow crystals. The analytical sample was
prepared by recrystallization from heptane/2-propanol: mp 108-
109 °C; 1H NMR (CDCl3) δ 7.33-7.22 (m, 5H), 3.95 (q, 1H, J )
6.5 Hz), 3.75 (dd, 1H, J ) 1.8, 12.1?Hz), 3.74 (dd, 1H, J ) 2.0,
12.5?Hz), 3.52 (dd, 1H, J ) 5.5, 12.5 Hz), 3.48 (ddd, 1H, J )
0.5, 5.9, 12.1 Hz), 3.37 (dt, 1H, J ) 1.4, 5.6 Hz), 2.44 (brs, 1H),
2.34 (dt, 1H, J ) 1.7, 5.5 Hz), 1.34 (d, 3H, J ) 6.5 Hz), 1.34 (s,
ep a n -6-yla m m on iu m Aceta te (8). To a 500 mL three-necked
round-bottomed flask were added (S)-1,1′-bi-2-naphthol (283 mg,
0.99 mmol) and heptane-toluene (9:1, 105 mL). To the suspen-
sion was added titanium tetraisopropoxide (281 mg, 0.99 mmol)
under a nitrogen atmosphere at room temperature, and the
solution was stirred for 10 min. Racemic 6c (24.0 g, 198 mmol),
water (300 mg, 16.7 mmol), and 1c (28.6 g, 198 mol) were added
to the solution successively. After stirring at 40 °C for 24 h, the
reaction mixture was diluted with toluene (84 mL) and heated
to reflux in the presence of 1 M aqueous sodium hydroxide (29
mL) for 1 h. The aqueous layer was removed, and the organic
layer was stirred in the presence of charcoal (300 mg) for 1 h.
The insoluble material in the mixture was removed by filtration,
and the filtrate was concentrated to give a crude mixture
containing 7c and 7d (1:1, 7c and 7d had the same molar UV
intensity at 214 nm) as a pale yellow oil. To a suspension of 5%
Pd(C) (5.30 g, 55% wet type) in 2-propanol (316 mL) were added
the mixture thus obtained, AcOH (11.3 mL, 198 mmol), and
water (9.8 mL). Then the mixture was stirred at room temper-
ature for 24 h under an H2 atmosphere (3 atm). After the catalyst
was removed, the filtrate was concentrated to 1/2 volume, diluted
with heptane (263 mL), and then stirred at room temperature
for 0.5 h. Filtration of the deposited crystals followed by drying
at 50 °C in vacuo gave 8 (36.7 g, 84% yield, ee >99%) as colorless
3H), 1.31 (s, 3H); IR (KBr) 3214, 1219, 1052 cm-1; [R]25 +96.2
D
(c 1.00, MeOH). Anal. Calcd for C15H23NO3: C, 67.90; H, 8.74;
N, 5.28. Found: C, 67.90; H, 9.01; N, 5.31. The de (97.0%) and
conversion ratio (99.5%) were determined by HPLC analysis of
an independent smaller run using identical conditions to those
shown above, in which the reaction mixture was converted to a
clear solution by addition of CH2Cl2 after 24 h. tR(7c) ) 16.4
min (98.5%), tR(corresponding diastereomer of 7c) ) 17.7 min
(1.5%), AM-302 (achiral), 45 °C, 2-propanol/phosphate buffer (pH
7) ) 17/83, 214 nm, 1.0 mL/min. The % conversion was calibrated
by the relative molar UV intensity ratio of 0.65 ((R)-6c (tR ) 2.5
min)/7c). The relative molar UV intensity ratio (7c/correspond-
ing diastereomer of 7c) was 1.00.
1
crystals: mp 133-134 °C; H NMR (CD3OD) δ 3.84 (dd, 1H, J
) 2.5, 12.7 Hz), 3.74 (dd, 1H, J ) 2.5, 12.5 Hz), 3.67-3.53 (m,
3H), 2.98 (dt, 1H, J ) 2.4, 6.5 Hz), 1.91 (s, 3H), 1.33 (s, 6H); IR
(KBr) 3178, 1561, 1087 cm-1; [R]25D +29.6 (c 1.05, MeOH). Anal.
Calcd for C9H19NO5: C, 48.86; H, 8.66; N, 6.33. Found: C, 48.98;
H, 8.70; N, 6.36. The crude 8 obtained before recrystallization
was converted into 919a to measure the stereoselectivity of the
reaction: tR(9) ) 23.7 min (98.6%), tR(anti-9) ) 31.7 min (1.4%),
CHIRALCEL OD, 35 °C, hexane/2-propanol ) 95/5, 210 nm, 1.0
mL/min.
(5R,6S)-2,2-Dim eth yl-6-[(S)-1-p h en yleth yla m in o]-1,3-d i-
oxep a n -5-ol (7d ). This compound was similarly obtained from
the reaction of 1c with (S)-6c as a pale yellow oil (94% yield,
97.6% de): 1H NMR (CDCl3) δ 7.32-7.20 (m, 5H), 3.92-3.83
(m, 2H), 3.67 (dd, 1H, J ) 2.0, 12.5 Hz), 3.59-3.40 (m, 2H), 3.30
(dd, 1H, J ) 5.3, 12.5 Hz), 2.52 (br t, 1H, J ) 4.0 Hz), 2.10 (brs,
1H), 1.34 (d, 1H, J ) 6.5 Hz), 1.31 (s, 6H); IR (KBr) 3416, 1218,
Ack n ow led gm en t. The authors would like to thank
Drs. I. Uchida, H. Cho, J . Haruta, and A. G. Birchler
for helpful discussions throughout this work. The au-
thors also express their appreciation to Mr. Masami
Tsuboshima for his encouragement.
1048 cm-1; [R]25D +12.0 (c 1.0, MeOH). HRMS Calcd for C15H24
-
NO3 (M+ + 1): 266.1757,found 266.1749. tR(7d ) ) 17.0 min
(98.8%), tR(corresponding diastereomer of 7d ) ) 15.9 min (1.2%),
AM-302 (achiral), 45 °C, 2-propanol/phosphate buffer (pH 7) )
17/83, 214 nm, 1.0 mL/min. There was 96.9% conversion after
24 h according to HPLC under identical conditions (tR((S)-6c) )
2.8 min). The % conversion was calibrated by the relative molar
UV intensity ratio of 0.65 ((S)-6c/7d ). The relative molar UV
intensity ratio (7d /corresponding diastereomer of 7d ) was 1.00.
The compound 7d thus obtained was converted to 9 via 8 to
determine its absolute configuration.19
Su p p or tin g In for m a tion Ava ila ble: Absolute stereo-
chemical assignment of all new chiral compounds, molecular
modeling procedure and coordinates of the most stable con-
former of 1c, 5, and 1c+Ti, HPLC charts from Table 1, entry
1
6 and from Table 2, entry 4, and 300 MHz H NMR spectra of
2c and 7d . This material is available free of charge via the
Internet at http://pubs.acs.org.
P r ep a r a tion of a Mixtu r e of 7c a n d 7d a n d Su ccessive
Con ver sion to (5R,6S)-2,2-Dim eth yl-5-h yd r oxy-1,3-d iox-
J O9900883