Chemo- and stereoselective monobenzoylation of 1,2-diols catalyzed by organotin compounds

Article abstract of DOI:10.1021/jo991394j

A new facile method for monoacylation of diols has been developed. A variety of cyclic and acyclic diols, in particular 1,2-diols, were selectively monobenzoylated in good yields by the reaction with benzoyl chloride in the presence of a catalytic amount of dimethyltin dichloride and inorganic bases such as potassium carbonate. Furthermore, the method was successfully applied to a kinetic resolution of racemic 1-phenyl-1,2- ethanediol using a chiral organotin catalyst. The ee was dependent on the kind of base, water as an additive, and the reaction temperature.

Full text of DOI:10.1021/jo991394j

996
J . Org. Chem. 2000, 65, 996-1002
Ch em o- a n d Ster eoselective Mon oben zoyla tion of 1,2-Diols
Ca ta lyzed by Or ga n otin Com p ou n d s
Fumiaki Iwasaki,† Toshihide Maki,‡ Osamu Onomura,‡ Waka Nakashima,‡ and
Yoshihiro Matsumura*,§
Tsukuba Research Laboratory, Tokuyama Corporation, 40 Wadai, Tsukuba 300-4247, J apan, Faculty of
Pharmaceutical Sciences, Nagasaki University, 1-14 Bunkyo-machi, Nagasaki 852-8131, J apan, and
Institute for Fundamental Research of Organic Chemistry, Kyushu University, 6-10-1 Hakozaki,
Higashi-ku, Fukuoka 812-8581, J apan
Received September 2, 1999
A new facile method for monoacylation of diols has been developed. A variety of cyclic and acyclic
diols, in particular 1,2-diols, were selectively monobenzoylated in good yields by the reaction with
benzoyl chloride in the presence of a catalytic amount of dimethyltin dichloride and inorganic bases
such as potassium carbonate. Furthermore, the method was successfully applied to a kinetic
resolution of racemic 1-phenyl-1,2-ethanediol using a chiral organotin catalyst. The ee was dependent
on the kind of base, water as an additive, and the reaction temperature.
In tr od u ction
Selective protection of a hydroxyl group of polyols is
very important in organic synthesis.1-4 Furthermore,
provided that the protecting method has some kind of
stereoselectivities, the method would be more useful in
organic synthesis. It is well-known that dibutyltin oxide
readily reacts with 1,2-diols 1 to form stannylene acetals
2, which can be benzoylated to give monobenzoylated
products 3 (eq 1).5-9
F igu r e 1. Racemic 1-phenyl-1,2-ethanediol (1j).
the use of more than an equimolar amount of dibutyltin
oxide, which complicates the purification process of the
products and makes the large-scale production of mono-
benzoylated products 3 difficult. Recently, a microwave
irradiation method has been reported to reduce the
amount of dibutyltin oxide. However, this method still
possesses disadvantages in that a microwave apparatus
is required and the irradiation conditions might be
drastic.10-12 Moreover, dibenzoates 4 are often formed as
byproducts in the benzoylation of 1 using dibutyltin
oxide.11
We preliminarily reported a very convenient method
for the monobenzoylation of 1, which was achievable in
a one-pot procedure with less than 0.01 molar equivalents
of tin catalysts under mild conditions.13,14 The report
briefly described the characteristics of the method such
as the high yields of 3, an excellent 1,2-diol-selectivity
with a minimized formation of 4. Furthermore, as an
application of the method, we succeeded in a first
catalytic system for nonenzymatic kinetic resolution of
racemic 1-phenyl-1,2-ethanediol (1j, Figure 1).15 The
method was based on the selective activation of 1,2-diols
by a chiral organotin catalyst, and therefore the process
possessed not only a high enantioselectivity but also a
high chemoselectivity.
The method, however, possesses some drawbacks.
Namely, 2 must be prepared by heating a mixture of 1
and dibutyltin oxide for extended periods of time prior
to the benzoylation because of the slow dehydration
reaction between 1 and dibutyltin oxide.5 This imposes
† Tokuyama Corporation.
‡ Nagasaki University.
§ Kyushu University
(1) Wilkinson, S. G. In Comprehensive Organic Chemistry; Stoddart,
J . F., Ed; Pergamon Press: Oxford, 1979; Vol. 1, Chapter 4.1, pp 579-
706.
(2) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, second ed.; J ohn Wiley & Sons: New York, 1991.
(3) Orita, A.; Mitsutome, A.; Otera, J . J . Org. Chem. 1998, 63, 2420-
2421.
(4) Tanino, K.; Shimizu, T.; Kuwahara, M.; Kuwajima, I. J . Org.
Chem. 1998, 63, 2422-2423.
(5) Hanessian, S.; David, S. Tetrahedron 1985, 41, 643-663 and
references therein.
(10) Morcuende, A.; Valverde, S.; Herrado´n, B. Synlett 1994, 89-
91.
(11) Herrado´n, B.; Morcuende, A.; Valverde, S. Synlett 1995, 455-
458.
(12) Morcuende, A.; Ors, M.; Valverde, S.; Herrado´n, B. J . Org.
Chem. 1996, 61, 5264-5270.
(13) The result has been preliminarily reported: Maki, T.; Iwasaki,
F.; Matsumura, Y. Tetrahedron Lett. 1998, 39, 5601-5604.
(14) More recently, a dibutyltin-catalyzed monotosylation of alcohols
involving diols has been reported after our preliminary report: Mar-
tinelli, M. J .; Nayyar, N. K.; Moher, E. D.; Dhokte, U. P.; Pawlak, J .
M.; Vaidyanathan, R. Org. Lett. 1999, 1, 447-450.
(15) Iwasaki, F.; Maki, T.; Nakashima, W.; Onomura, O.; Mat-
sumura, Y. Org. Lett. 1999, 1, 969-972.
(6) Ricci, A.; Roelens, S.; Vannucchi, A. J . Chem. Soc., Chem.
Commun. 1985, 1457-1458.
(7) Roelens, S. J . Chem. Soc., Perkin Trans. 2 1988, 1617-1625.
(8) Reginato G.; Ricci, A.; Roelens, S.; Scapecchi, S. J . Org. Chem.
1990, 55, 5132-5139.
(9) Roelens, S. J . Org. Chem. 1996, 61, 5257-5263.
10.1021/jo991394j CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/26/2000

Monobenzoylation of 1,2-Diols
J . Org. Chem., Vol. 65, No. 4, 2000 997
Ta ble 1. Mon oben zoyla tion of 1a w ith P h COCl
Ca ta lyzed by Or ga n otin Com p ou n d sa
Ta ble 2. Mon oben zoyla tion of Va r iou s Diols 1b-h
a
Ca ta lyzed by Me2Sn Cl2
yield (%)b,c
organotin mol
reaction
run compound
%
base
Et3N
solvent time (h)
3a
4a
1
2
3
none
none
none
THF
THF
CH2Cl2
THF
THF
THF
THF
THF
THF
24
24
24
24
12
12
12
24
12
12
12
12
12
12
24
72
3
34
56
30
37
83
78
55
0
>99
>99
>99
>99
>99
>99
93
82
93
54
trace
6
DMAP
DMAP
K2CO3
K2CO3
K2CO3
K2CO3
none
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
31
34
0
0
0
0
0
0
0
0
0
0
0
0
0
4d none
5
6
7
8
9
Bu2SndO
1
1
1
1
Bu2SnCl2
Ph2SnCl2
Me2SnCl2
Me2SnCl2 10
10 Me2SnCl2
11 Me2SnCl2
12 Me2SnCl2
13 Me2SnCl2
14 Me2SnCl2
1
1
1
1
1
THF
CH2Cl2
MeCN
Et2O
Na2CO3 THF
15 Me2SnCl2 0.5 K2CO3
16 Me2SnCl2 0.1 K2CO3
THF
THF
THF
THF
THF
17 Me2SnCl2
18 Me2SnCl2
19 Me2SnCl2
20 Me2SnCl2
21 Bu3SnCl
22 BuSnCl3
1
1
1
1
1
1
Et3N
DMAP
KOAc
NaHCO3 THF
K2CO3
K2CO3
trace
32
0
0
0
24
24
24
12
12
72
33
37
THF
THF
0
a
General conditions: 1a (1.0 mmol), Me2SnCl2 (0.01 mmol),
K2CO3 (2.0 mmol) in THF (5 mL), then PhCOCl (1.2 mmol); see
a
General conditions: 1b-h (1.0 mmol), Me2SnCl2 (0.01 mmol),
b
Experimental Section. Isolated yield. c Starting diol remained in
K2CO3 (2.0 mmol) in THF (5 mL), then PhCOCl (1.2 mmol); see
d
the cases of low yields. Equimolar amount of PhCOCl was used.
Experimental Section. The corresponding dibenzoate. c Starting
b
diol remained.
We present herein the full detail of the chemo- and
stereoselective monobenzoylation of 1,2-diols using or-
ganotin compounds.
yltin catalyst (runs 2, 3 and 18). This may be due to the
effective acylation catalysis ability of DMAP. Sodium
bicarbonate resulted in a relatively moderate yield (run
20). Although triethylamine afforded 93% of 3a within 3
h by using 1 mol % of dimethyltin dichloride, a trace
amount of 4a was observed (run 17), whereas potassium
or sodium carbonate required longer reaction time but
showed a complete selectivity. Thus, potassium or sodium
carbonate was found to be a suitable base for the reaction.
Tetrahydrofuran (THF) was recommended as a solvent,
while dichloromethane (run 11), acetonitrile (run 12), and
ether (run 13) were also usable.16
Resu lts a n d Discu ssion
Mon oben zoyla tion of tr a n s-1,2-Cycloh exa n ed iol.
To exploit an efficient method for the monobenzoylation
of 1,2-diols 1 using a catalyst, we first studied a benzo-
ylation of trans-1,2-cyclohexanediol (1a ) as a representa-
tive of 1 under various conditions (eq 2).
A typical experimental procedure is as follows: Diol
1a (1.0 mmol) was dissolved in THF (5 mL) with a
catalytic amount of dimethyltin dichloride (0.01 mmol)
in the presence of potassium carbonate (2.0 mmol).
Benzoyl chloride (1.2 mmol) was added to the mixture,
and the solution was stirred at room temperature until
1a disappeared (checked by thin-layer chromatography).
After a usual workup, monobenzoylated product 3a was
obtained in a quantitative yield.
The acylation of diols also took place in a similar way
to benzoylation by using other acyl chlorides, though the
yields were less satisfactory. For example, 1a was acyl-
ated by acetyl chloride to give 3a ′ in 54% yield (eq 3).
We surveyed a variety of organotin compounds as
catalysts and examined the effect of the amount of the
catalysts, the kind of bases and solvents on the yield of
monobenzoylated product 3a , and the ratio of 3a to
dibenzoylated product 4a . The results are shown in Table
1.
The results clearly show the effect of dialkyltin com-
pounds (runs 5-20 except 8 and 19). The absence of
organotin compounds (runs 1-4) and the presence of
trialkyl- (run 21) or monoalkyltin compounds (run 22)
resulted in low yields of 3a . Among dialkyltin compounds,
dimethyltin dichloride gave the best result (compare runs
10-14 with runs 5-7). The yield of 3a decreased as the
amount of dimethyltin dichloride decreased (runs 9, 10,
15, and 16).
Mon oben zoyla tion of 1,n -Diols. Under the best
reaction conditions described above, 1,2-diols 1b-d and
The results also show the importance of the bases used.
Namely, in the absence of a base, attempted monoben-
zoylation in the presence of dimethyltin dichloride did
not afford 3a (runs 8 and 19). When (4-dimethylamino)-
pyridine (DMAP) was used as a base, a large amount of
4a was generated irrespective of the presence of dimeth-
(16) Although the chemical yields in the monobenzoylation of 1a
using dimethyltin dichloride were almost similar for these solvents,
the use of THF gave the highest ee of 3j in the benzoylation of racemic
1j.

998 J . Org. Chem., Vol. 65, No. 4, 2000
Iwasaki et al.
F igu r e 2.
1,n-diols 1e-h were benzoylated (Table 2). Cyclic and
acyclic 1,2-diols 1b-d were monobenzoylated with almost
perfect selectivity in the presence of 1 mol % of dimeth-
yltin dichloride (runs 1-3). 1,3-Propanediol (1e) was also
monobenzoylated with a high selectivity, but the yield
was less than those in cases of 1,2-diols (run 4). Further-
more, the yields of 3 decreased as the number of n in
1,n-diols increased (runs 3-8).
In a case of an unsymmetrical 1,3-diol 1l, a selectivity
similar to that of 1j was observed (eq 7).
In contrast to the high selectivity observed in those 1,2-
and 1,n-diols with respect to mono- and dibenzoylation,
the benzoylation of diethyl L-tartrate (1i) gave the
monobenzoylated product 3i in 70% yield with a signifi-
cant amount (16%) of the dibenzoylated product 4i (eq
4). The low selectivity concerning the mono- and diben-
zoylation may be due to the high acidity of the hydroxyl
groups of 1i.17
Ch em oselectivities; Com p etitive Rea ction . To see
the chemoselectivity in the tin-catalyzed benzoylation
reaction, we carried out competitive reactions between
1,2-diol 1a and n-butanol and between 1,n-diols. In a
competitive reaction between 1a and n-butanol, 1a was
selectively benzoylated even in the presence of 5 molar
equivalents of n-butanol to give 3a in 89% yield (eq 8).
Thus, n-butanol was found to be actually inert for the
benzoylation under the reaction conditions.
Regioselectivity of Un sym m etr ica l Diols 1j-l. Our
tin-catalyzed benzoylation was applied to unsymmetrical
diols 1j-l. 1-Phenyl-1,2-ethanediol (1j) afforded the
monobenzoylated product 3j in good yield, though a small
amount of the regioisomer 3j′ was formed in low yield
(eq 5).
Furthermore, when an equimolar amount of 1,2-
ethanediol (1d ) and 1,3-propanediol (1e) was subjected
to the monobenzoylation reaction, 1d was predominantly
monobenzoylated to give 3d (eq 9). On the other hand,
the difference of yields between 3e and 3f was not
significant in a competitive reaction between 1,3-diol 1e
and 1,4-diol 1f under the same reaction condition (eq 9).
On the other hand, a benzoylation of a glucose deriva-
tive 1k selectively occurred at the 2-hydroxyl group (eq
6).
(17) J ohnson, R.-A.; Sharpless, K. B. In Comprehensive Organic
Synthesis; Trost, B. M., Fleming, I., Eds.; Pergamon Press: New York,
1991; Vol. 7, Chapter 3.2, pp 390-432.
Interestingly, when an equimolar mixture of trans-1,2-
cyclohexanediol (1a ) and diethyl L-tartrate (1i) was
subjected to the benzoylation reaction, only 1i was
benzoylated to afford a mixture of 3i in 64% and 4i in
14%, respectively (eq 10).
(18) Noyori, R.; Kitamura, M.; Takemoto, K. J pn. Kokai Tokkyo Koho
J P 1992, 04-91093.
(19) Nanni, D.; Curran, D. P. Tetrahedron: Asymmetry 1996, 7,
2417-2422.
(20) Maigrot, N.; Mazaleyrat, J .-P. Synthesis 1985, 317-320.
(21) Brown, J . M.; Murrer, B. A. J . Chem. Soc., Perkin Trans. 2 1982,
489.
(22) Kagan, H. B.; Fiaud, J . C. Top. Stereochem. 1988, 18, 249-
330.
There was no significant difference between cyclic-1,2-
diols and acyclic-1,2-diols as exemplified by the benzo-

Monobenzoylation of 1,2-Diols
J . Org. Chem., Vol. 65, No. 4, 2000 999
F igu r e 3.
Ta ble 3. Kin etic Resolu tion of Diol 1j by Ben zoyla tion
w ith a Ca ta lyst Aa
H2O
temp
(°C) (yields %e) of 3jf
%eec,d
run A (mol %) (µL)
baseb
Et3N
Na2CO3 r.t.
sg
1
2
3
4
5
6
7
8
9
0
0
0
0
0
0
r.t.
∼0 (40)
∼0 (23)
∼0 (43)
32 (31)
58 (42)
23 (16)
13 (19)
75 (54)
86 (38)
84 (41)
79 (28)
32 (10)
ylation of a mixture of an equimolar amount of 1,2-diols
1a and 1d (eq 11).
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
Et3N
-20
Na2CO3 r.t.
2.2
5.6
1.7
100 Na2CO3 r.t.
1000 Na2CO3 r.t.
100 Et3N
r.t.
0
1.3
100 Na2CO3
18.6
22.4
20.7
11.5
2.0
100 Na2CO3 -10
100 Na2CO3 -20
100 Na2CO3 -30
100 Na2CO3 -40
10
11
12
a
b
General conditions: see Experimental Section. 1.5 equiv of
1j was used. c Determined by CSP HPLC analysis (for 3j, Chiralcel
d
OB column; 254 nm; n-hexane:i-PrOH ) 9:1). Corrected value
based on the ee of used A (91%). e Isolated yield based on starting
These observations strongly suggest that the reaction
proceeds via formation of dimethylstannylene acetal 5
(Figure 2), and a formation of the five-membered inter-
mediate (n ) 0) should be kinetically more favored than
that of the other membered stanylene acetals (n > 0).
This feature might lead to the high 1,2-diol selectivity
that is very useful for synthetic utilization of this
benzoylation method.
1j. f (S)-isomer was obtained. Selectivity: see ref 22.
g
results are summarized in Table 3. The main product was
the (S)-enantiomer-enriched 2-benzoyloxy-1-phenyletha-
nol ((S)-3j), of which an ee was determined by CSP HPLC
analysis. The absolute configuration of (S)-3j was con-
firmed by its alkaline hydrolysis followed by comparison
of the specific rotation of the resulting 1j with that of
authentic sample.21 The recovered 1j was enriched with
(R)-enantiomer in 56-80% yields. There appeared a
small amount of 1-benzoylated diol 3j′ (0∼10%) but no
dibenzoylated product.
Ster eoselectivity. There was almost no difference in
the benzoylation between trans- and cis-1,2-cyclohex-
anediols 1a ,b (eq 12).
The results showed the remarkable effects of a chiral
tin catalyst A, the kind of base, and water on the ee of
3j. Namely, the absence of A in the reaction system did
not cause any resolution of 1j (runs 1 and 2), though 3j
was formed in moderate yields. Theoretically, the maxi-
mum yield is 50% since the amount of benzoyl chloride
was 0.5 equiv to that of 1j. On the other hand, the
presence of A afforded a result with a resolution of 1j
(runs 4-12). The use of sodium carbonate was one of the
critical factors for getting a satisfactory result. Although
triethylamine gave an unsuccessful result with respect
to the ee of 3j (run 3), sodium carbonate as a suspended
state gave an appreciable degree of ee (32%) of 3j (run
4). The other factor for the kinetic resolution was the
presence of water. The selectivity observed in run 4 was
improved to 58% ee if a small amount of water was added
(run 5). However, the ee decreased in the presence of a
large amount of water (run 6), where sodium carbonate
was completely dissolved to give a homogeneous solution.
In contrast with sodium carbonate, triethylamine was not
effective even though water was present (run 7).
However, an interesting result was obtained in the
benzoylation of 1j (eq 13) using a chiral tin catalyst, (S)-
4,4-dibromo-4,5-dihydro-3H-dinaphtho[2,1-c:1′,2′-e]stan-
nepin (A, Figure 3),18 which was synthesized according
to the reported method.19,20
Although the selectivity (s value)22 was not much
varied in the temperature range of 0 to -30 °C (runs
8-11), the yield of 3j dropped when the reaction was
carried out at -30 °C (run 11), and the reaction at -40
°C resulted in both low ee and low yield (run 12). The
best selectivity was obtained when the reaction was
The benzoylation of racemic 1j using a chiral catalyst
A was carried out under a variety of conditions, and the

1000 J . Org. Chem., Vol. 65, No. 4, 2000
Iwasaki et al.
Ta ble 4. Effects of Solven ts on th e Mon oben zoyla tion of
1ja
Ta ble 5. Kin etic Resolu tion of Diol 1n -q by
Ben zoyla tion w ith a Ca ta lyst Aa
PhCOCl (0.5 equiv)
A (0.25 mol %)
1j
8 3j + recovered 1j
water (100 µL)
Na2CO3 (1.5 equiv)
-10 °C, 14 h
run
solvent (5 mL)
ee %b,c (yield %)d of 3j
se
1
2
3
4
THF
DME
MeCN
CH2Cl2
84 (41)
56 (39)
74 (22)
13 (12)
20.7
5.0
8.2
1.3
a
b
General conditions: See Experimental Section. Determined
by Chiral Column HPLC. c Corrected based on the ee of used A
(91%). Isolated yield. e Selectivity; see ref 22.
d
performed using sodium carbonate in the presence of a
small amount of water at -10 °C (run 9).
The other factor affecting the selectivity of the kinetic
resolution was the kind of solvent. The results carried
out in several solvents are shown in Table 4.
a
b
General conditions: see Experimental Section. Determined
by CSP HPLC analysis. c (S)-Isomer was enriched. Isolated yield.
d
e See ref 22.
The benzoylation of 1j in dichloromethane was sluggish
and afforded poor results in both the ee and chemical
yield of 3j (run 4), whereas relatively high ee was
observed by the use of polar solvents such as acetonitrile,
THF, or DME (runs 1-3). The best result was obtained
by using THF (run 1). Although the reason the ee of 3j
was enhanced by polar solvents has not been clear yet,
an ability of solvents to coordinate with the tin atom of
A may be important.
Sch em e 1. En a n tioselective Ben zoyla tion Usin g a
Ch ir a l Ca ta lyst A
The catalyst A also showed a high 1,2-diol selectivity
as dimethyltin dichloride did. That is, 1j was efficiently
resolved even in the presence of an equimolar amount of
cyclohexanol (1m ), and the ee of 3j obtained in the
reaction was almost not affected by the presence of 1m ,
though a small amount of cyclohexyl benzoate (3m ) was
formed (eq 14).
various acid chlorides,5 the first step may be responsible
for the enantio discrimination of racemic 1j, which may
occur on the surface of sodium carbonate.
Thus, we propose a mechanism for the kinetic resolu-
tion of 1j using A as depicted in Scheme 2. That is, when
amine is used as a base, stannylene acetals 2j may be
formed in a solution phase (THF) in a manner in which
A approaches the primary hydroxyl group of 1j from a
direction as far away as possible from a bulky phenyl
group (route a in Scheme 2). According to this hypothesis,
there seems to be no significant difference between the
accessibility of A toward (S)-1j and (R)-1j.
On the contrary, when sodium carbonate is used as a
base, the hydrophilic 1,2-diol moiety of 1j may be ad-
sorbed on the surface of the sodium carbonate, directing
the hydrophobic phenyl group toward the solution phase
(THF) (route b in Scheme 2). In the presence of a small
amount of water, a thin aqueous phase is formed on the
surface of the sodium carbonate and, as the result, the
adsorption of 1j is assisted by hydrogen bonding as
depicted in Scheme 2. Provided that this hypothesis is
true, A should approach from the solution phase in a
manner in which a steric repulsion between A and phenyl
group is as small as possible. Thus, the discrimination
of (S)-1j from (R)-1j takes place since a steric repulsion
between the methylene group of A and phenyl group of
(S)-1j is less than that between A and (R)-1j.
This method could be applied to various 1-substituted
1,2-ethanediols 1n -q. The results are summarized in
Table 5.
A-catalyzed monobenzoylation of terminal 1,2-diols
1n -q afforded (S)-enantiomer-enriched 2-benzoylated
diols 3n -q in moderate selectivity. Only a trace amount
of 1-benzoylated diols was observed under the experi-
mental conditions. Other patterns of 1,2-diols such as 1a
afforded disappointing results under similar reaction
conditions.
Mech a n ism for Kin etic Resolu tion . The resolution
reaction may involve two crucial steps: a formation of
stannylene acetal 2 by the reaction of 1 with a chiral tin
catalyst A (first step), and a benzoylation of 2 with
benzoyl chloride (second step) (Scheme 1). A formation
of 2 is supported by the high chemoselective formation
of 3j with a high ee from a mixture of 1j and 1m (eq 14).
Since the observed ee was strongly dependent on the
base (Table 3) and it has been well-documented that
stannylene acetals are easily acylated without bases by
The role of THF on the kinetic resolution is not clear.
One of working hypotheses is that a coordination of THF
on the tin atom of A may increase the bulkiness of A,
which works to increase the enantioselectivity in the
monobenzoylation of 1j.

Monobenzoylation of 1,2-Diols
J . Org. Chem., Vol. 65, No. 4, 2000 1001
possible with a mill. Monobenzoylated products 3a ,23 3b,24 3c,24
3d ,25 3e,26 3f,25 3g,27 3h ,28 3i,10 3j,8 3j′,8 3k ,29 3l,10 and 3l′,10
monoacetylated product 3′a ,30 and dibenzoylated products 4a ,31
Sch em e 2. En a n tioselective Ben zoyla tion of 1j on
th e Su r fa ce of Na 2CO3
1
4g,32 4h ,33 and 4i10 are known compounds, and only H NMR
and/or IR spectra data for monobenzoylated compounds 3 are
presented here.
Gen er a l P r oced u r e (Mon oben zoyla tion of 1,2-Diol). To
a solution of diol (1 mmol) in tetrahydrofuran (THF, 5 mL)
were added a catalytic amount of dimethyltin dichloride (0.1-
10 mol %), K2CO3 (2 mmol), and benzoyl chloride (1.2 mmol),
successively, at room temperature. The mixture was stirred
at room temperature for 3-72 h. Then the mixture was poured
into water and extracted three times with CH2Cl2 (30 mL).
The organic layers were combined, and the solvent was
evaporated in vacuo to give a residue, which was almost a pure
monobenzoylated product. Further purification of the crude
products was carried out by silica gel short-column chroma-
tography to afford the benzoylated products.
tr a n s-2-Hyd r oxycycloh exyl Ben zoa te (3a ).23 IR (neat)
3450, 2940, 2863, 1719, 1451, 1321, 1275, 1113, 1071, 712
cm-1; 1H NMR (300 MHz, CDCl3) δ 1.22-1.53 (m, 4H), 1.60-
1.84 (m, 2H), 2.00-2.25 (m, 2H), 2.50 (br s, 1H), 3.67-3.80
(m, 1H), 4.79-4.91 (m, 1H), 7.43-7.60 (m, 3H), 8.07-8.10 (m,
2H).
cis-2-Hydr oxycycloh exyl Ben zoate (3b).24 IR (neat) 3480,
1
2940, 2865, 1711, 1451, 1316, 1287, 1111, 984, 716 cm-1; H
NMR (300 MHz, CDCl3) δ 1.31-1.55 (m, 2H), 1.56-1.90 (m,
4H), 1.90-2.16 (m, 3H), 3.95-3.97 (m, 1H), 5.21-5.23 (m, 1H),
7.42-7.44 (m, 2H), 7.54-7.57 (m, 1H), 8.04-8.08 (m, 2H).
cis-2-Hyd r oxycyclop en tyl Ben zoa te (3c).24 IR (neat)
3490, 2971, 1694, 1451, 1271, 1117, 1026, 710 cm-1; 1H NMR
(300 MHz, CDCl3) δ 1.55-2.14 (m, 6H), 2.25 (br s, 1H), 4.25-
4.33 (m, 1H), 5.19-5.25 (m, 1H), 7.48-7.48 (m, 2H), 7.55-
7.58 (m, 1H), 8.04-8.07 (m, 2H).
2-Hyd r oxyeth yl Ben zoa te (3d ).25 IR (neat) 3440, 2953,
1
1722, 1603, 1452, 1315, 1124, 1070, 1028, 907, 718 cm-1; H
NMR (300 MHz, CDCl3) δ 2.40 (br s, 1H), 3.95 (t, 2H, J ) 4.7
Hz), 4.45 (t, 2H, J ) 4.7 Hz) 7.40-7.47 (m, 2H), 7.50-7.59
(m, 1H), 8.01-8.08 (m, 2H).
In conclusion, we reported a chemo- and stereoselective
monobenzoylation of diols. The extremely high 1,2-diol
selectivity of this reaction may be very useful in organic
synthesis. Especially, the reaction required only a cata-
lytic amount of organotin compound. Furthermore, it was
demonstrated that a chiral oragnotin compound A could
work as an efficient catalyst for kinetic resolution of
terminal 1,2-diols with the assistance of metal carbonate
and a small amount of water. Since the catalyst A works
with not only 1,2-diol selective but also stereoselective
manner, this method would be a promising protocol for
kinetic resolution of diols.
3-Hyd r oxyp r op yl Ben zoa te (3e).26 IR (neat) 3420, 2961,
2890, 1725, 1452, 1391, 1316, 1177, 1100, 718 cm-1; 1H NMR
(300 MHz, CDCl3) δ 2.01 (quint, 2H, J ) 6.1 Hz), 2.23 (br s,
1H), 3.78 (t, 1H, J ) 6.1 Hz), 4.49 (t, 1H, J ) 6.1 Hz) 7.44-
7.47 (m, 2H), 7.55-7.57 (m, 1H), 8.03-8.06 (m, 2H).
4-Hyd r oxybu tyl 1-Ben zoa te (3f).25 IR (neat) 3410, 2951,
1722, 1453, 1316, 1177, 1070, 943, 712 cm-1 1H NMR (300
;
MHz, CDCl3) δ 1.57-2.01 (m, 4H), 2.48, (br s, 1H), 3.73 (t,
2H, J ) 6.4 Hz), 4.37 (t, 2H, J ) 6.4 Hz) 7.44-7.47 (m, 2H),
7.55-7.57 (m, 1H), 8.03-8.10 (m, 2H).
5-Hyd r oxyp en tyl 1-Ben zoa te (3g).27 IR (neat) 3410, 2950,
1725, 1453, 1180, 1075, 945, 710 cm-1 1H NMR (300 MHz,
;
Although the speculated mechanism can explain the
observed stereoselectivity of the reaction, further study
has to be carried out to make the mechanism more clear.
Designing of new organotin reagents and application of
this catalytic system to further complex polyols are now
under investigation.
CDCl3) δ 1.40-1.71 (m, 4H), 1.71-1.92 (m, 2H), 2.15 (s, 1H),
3.66 (t, 2H, J ) 6.4 Hz), 4.33 (t, 2H, J ) 6.4 Hz) 7.40-7.47
(m, 2H), 7.50-7.60 (m, 1H), 8.00-8.10 (m, 2H).
6-Hyd r oxyh exyl 1-Ben zoa te (3h ).28 IR (neat) 3410, 2955,
1722, 1453, 1180, 1075, 943, 712 cm-1 1H NMR (300 MHz,
;
(23) Fujita, M.; Hiyama, T. J . Org. Chem. 1988, 53, 5405-5415.
(24) Glass, B. D.; Goosen, A.; McCleland, C. W. J . Chem. Soc., Perkin
Trans. 2 1993, 2175-2181.
Exp er im en ta l Section
(25) J ain, S.; J ain, R.; Singh, J .; Anand, N. J . Indian Inst. Sci. 1994,
74, 117-133.
GC analyses were obtained by using a GC-12A from Shi-
1
madzu Seisakusho Inc. H NMR spectra were measured on a
(26) Murahashi, S.; Naota, T.; Oda, Y. J pn. Kokai Tokkyo Koho J P
1992, 283-540.
Varian Gemini 200, 300, or 500 spectrometer with TMS as an
internal standard. Mass spectra were obtained on a J EOL
IMS-DX 300 instrument. IR spectra were obtained on a
Shimadzu FTIR-8100A.
Ma ter ia ls. MeCN and CH2Cl2 were distilled over P2O5.
Freshly distilled THF and ether were used after drying over
sodium metal. Organotin compounds (Bu2SndO, Bu2SnCl2,
Ph2SnCl2, Me2SnCl2, Bu3SnCl, and BuSnCl3), benzoyl chloride,
acetyl chloride, Et3N, and 4-(dimethylamino)pyridine (DMAP)
were commercially available and used without further puri-
fication. Diols 1a -l were also commercially available. K2CO3
and Na2CO3 were used after being pulverized as small as
(27) Ito, H.; Taguchi, T.; Hanzawa, Y. J . Org. Chem. 1993, 58, 774-
775.
(28) Toyama, T.; Tomiyama, I.; Wakabayashi, S.; Yokota, M. J pn.
Kokai Tokkyo Koho J P 93-140, 064.
(29) Munavu, R. M.; Szmant, H. H. J . Org. Chem. 1976, 41, 1832-
1836.
(30) Hodson, L. F.; Parker, T. M.; Whittaker, D. J . Chem. Soc.,
Chem. Commun. 1993, 1427-1428.
(31) Chen, Y.; Sciao, M.-T. Bull. Chem. Soc. J pn. 1992, 65, 3423-
3429.
(32) Shiseido, Co. Ltd. J pn. Kokai Tokkyo Koho J P 80-133, 305.
(33) Meth-Cohn, O.; Smith, D. I. J . Chem. Soc., Perkin Trans. 1
1982, 261-270.

1002 J . Org. Chem., Vol. 65, No. 4, 2000
Iwasaki et al.
CDCl3) δ 1.35-1.68 (m, 6H), 1.71-1.90 (m, 2H), 2.09 (s, 1H),
3.64 (t, 2H, J ) 6.4 Hz), 4.32 (t, 2H, J ) 6.4 Hz) 7.40-7.49
(m, 2H), 7.50-7.60 (m, 1H), 8.00-8.10 (m, 2H).
(S)-2-Ben zoyloxy-1-p h en yleth a n ol (3j). [R]26D +8.5 (c 1.0,
MeOH); DAICEL Chiralcel OB (0.46 cm × 25 cm); n-hexane:
2-propanol ) 9:1; wavelength, 254 nm; flow rate, 1.0 mL/min;
retention time, 13, 25 min, 78% ee (uncorrected). The (S)
configuration of 3j was confirmed by comparison of the specific
rotation of authentic samples of 1j21 after an alkaline hydroly-
sis of 3j.
Mon oben zoyla ted Dieth yl L-Ta r tr a te (3i).10 IR (neat)
3500, 3010, 1732, 1260 cm-1 1H NMR (300 MHz, CDCl3) δ
;
1.20 (t, 3H, J ) 7.1 Hz), 1.31 (t, 3H, J ) 7.1 Hz), 2.40-3.30
(br, 1H), 4.17-4.38 (m, 4H), 4.87 (d, 1H, J ) 2.3 Hz), 5.67 (d,
1H, J ) 2.3 Hz), 7.37-7.52 (m, 2H), 7.53-7.65 (m, 1H), 7.99-
8.09 (m, 2H).
(S)-(2-Na p h th yl)-2-h yd r oxyeth yl Ben zoa te (3n ). White
solid, mp 90-97 °C; [R]21D +4.9 (c 0.48, MeOH); IR (KBr) 2361,
2341 cm-1; 1H NMR (CDCl3) δ 4.51 (1H, dd, J ) 8.0, 11.6 Hz),
4.62 (1H, dd, J ) 3.6, 11.6 Hz), 5.28 (1H, dd, J ) 3.6, 8.0 Hz),
7.40-7.65 (6H, m), 7.81-8.09 (6H, m); DAICEL Chiralpak AS
(0.46 cm × 25 cm); n-hexane:2-propanol ) 9:1; wavelength,
254 nm; flow rate, 0.8 mL/min; retention time, 16, 19 min;
64% ee; HRMS calcd for C19H16O3 292.1099, found 292.1129.
Anal. Calcd for C19H16O3: C, 78.06; H, 5.52. Found: C, 78.12;
H, 5.62.
2-Hyd r oxy-2-p h en yleth yl Ben zoa te (3j).8 IR (neat) 3470,
1
1705, 1451, 1318, 1291, 1127, 1065, 708 cm-1; H NMR(300
MHz, CDCl3) δ 2.60 (br s, 1H), 4.43 (dd, 1H, J ) 7.0, 11.0 Hz),
4.55 (dd, 1H, J ) 4.0, 11.0 Hz), 5.13 (dd, 1H, J ) 4.0, 7.0 Hz),
7.30-7.65 (m, 8H), 8.04-8.10 (m, 2H).
2-Hyd r oxy-1-p h en yleth yl Ben zoa te (3j′).8 IR (neat) 3430,
1
2934, 1717, 1453, 1318, 1117, 1026, 711 cm-1; H NMR (300
MHz, CDCl3) δ 2.04 (br s, 1H), 3.90-4.20 (m, 2H), 6.11 (dd,
1H, J ) 7.4, 7.7 Hz), 7.30-7.65 (m, 8H), 8.10-8.15 (m, 2H).
(S)-(1-Na p h th yl)-2-h yd r oxyeth yl Ben zoa te (3o). White
solid, mp 99-103 °C; [R]D24 -6.5 (c 0.5, MeOH); IR (KBr) 3510,
3404, 3059, 1705, 1722 cm-1; 1H NMR (CDCl3) δ 4.51 (1H, dd,
J ) 8.0, 11.6 Hz), 4.63 (1H, dd, J ) 3.6, 11.6 Hz), 5.28 (1H,
dd, J ) 3.6, 8.0 Hz), 7.37-7.54 (6H, m), 7.81-7.89 (4H, m),
8.02-8.06 (2H, m); DAICEL Chiralpak AS (0.46 cm ø × 25
cm); n-hexane:2-propanol ) 9:1; wavelength, 254 nm; flow rate,
1.0 mL/min; retention time, 16, 20 min; 72% ee; HRMS calcd
Met h yl
2-Ben zoyl-4,6-O-b en zylid en e-r-D-glu cosid e
(3k ).29 IR (neat) 3498, 2986, 1723, 1453, 1379, 1281, 1109,
1071, 995, 700 cm-1; 1H NMR (500 MHz, CDCl3) δ 2.57 (s, 1H),
3.39 (s, 3H), 3.62 (t, 1H, J ) 9.4 Hz), 3.79 (t, 1H, J ) 10.3 Hz),
3.88-3.93 (m, 1H), 4.31-4.37 (m, 2H), 5.02-5.08 (m, 2H), 5.57
(s, 1H), 7.36-7.40 (m, 3H), 7.43-7.47 (m, 2H), 7.50-7.52 (m,
2H), 7.56-7.59 (m, 1H), 8.08-8.11 (m, 2H).
3-Hyd r oxybu tyl 1-Ben zoa te (3l).10 1H NMR (300 MHz,
CDCl3) δ 1.27 (d, 3H, J ) 6.2 Hz), 1.77-2.00 (m, 2H), 2.35 (br
s, 1H), 3.91-4.04 (m, 1H), 4.32-4.43 (m, 1H), 4.52-4.65 (m,
1H), 7.39-7.47 (m, 2H), 7.51-7.57 (m, 1H), 8.00-8.07 (m, 2H).
4-Hyd r oxybu tyl 2-Ben zoa te (3l′).10 1H NMR (300 MHz,
CDCl3) δ 1.41 (d, 3H, J ) 7.9 Hz), 1.76-1.99 (m, 2H), 2.53 (br
s, 1H), 3.62-3.73 (m, 2H), 5.31-5.42 (m, 1H) 7.47-7.50 (m,
2H), 7.55-7.60 (m, 1H), 8.01-8.07 (m, 2H).
for
C19H16O3 292.1099, found 292.1134. Anal. Calcd for
C
19H16O3: C, 78.06; H, 5.52. Found: C, 78.18; H, 5.61.
(S)-2-Hyd r oxyd od ecyl Ben zoa te (3p ). White solid, mp
35-37 °C; [R]24 +1.9 (c 0.49, MeOH); IR (KBr) 3501, 2953,
D
2920, 2849, 1697 cm-1; 1H NMR (CDCl3) δ 0.88 (3H, t, J ) 6.5
Hz), 4.23 (2H, s), 7.42-7.50 (2H, m), 7.55-7.60 (1H, m), 8.04-
8.09 (2H, m); DAICEL Chiralpak AS (0.46 cm × 25 cm);
n-hexane:2-propanol ) 98:2; wavelength, 254 nm; flow rate,
1.0 mL/min; retention time, 11, 16 min; 59% ee; HRMS calcd
tr a n s-2-Hyd r oxycycloh exyl Aceta te (3a ′).30 IR (neat)
1
3450, 2940, 1710 cm-1; H NMR (300 MHz, CDCl3) δ 1.20-
for
C19H30O3 306.2195, found 306.2203. Anal. Calcd for
1.53 (m, 4H), 1.63-1.82 (m, 2H), 1.90-2.15 (m, 2H), 2.10 (s,
3H), 2.40 (br s, 1H), 3.48-3.65 (m, 1H), 4.52-4.65 (m, 1H).
Gen er a l P r oced u r e for Com p etitive Rea ction s. To a
solution of diols (1.0 mmol and 1.0 mmol) in tetrahydrofuran
(THF, 5 mL) were added a catalytic amount of dimethyltin
dichloride (0.01 mmol), K2CO3 (2 mmol), and benzoyl chloride
(1.0 mmol), successively, at room temperature. The reaction
was carried out in a similar way to that in the monobenzo-
ylation of a 1,2-diol. The product ratios were determined by
C19H30O3: C, 74.47; H, 9.87. Found: C, 74.70; H, 9.71.
(S)-2-Hyd r oxybu tyl Ben zoa te (3q). Colorless oil, [R]20
D
+3.7 (c 1.0, MeOH); IR (neat) 3450, 2966, 2880, 1718 cm-1
;
1H NMR (CDCl3) δ 1.03 (3H, t, J ) 7.5 Hz), 1.54-2.27 (2H,
m), 3.92 (1H, dd, J ) 3.4, 6.8 Hz), 4.24 (1H, dd, J ) 6.8, 11.8
Hz), 4.40 (1H, dd, J ) 3.4, 11.8 Hz), 7.40-7.48 (2H, m), 7.54-
7.58 (1H, m), 8.03-8.08 (2H, m); DAICEL Chiralcel OJ (0.46
cm × 25 cm); n-hexane:2-propanol ) 98:2; wavelength, 254
nm, flow rate, 1.0 mL/min; retention time, 45, 48 min; 40%
ee; HRMS calcd for C11H14O3 194.0943, found 194.0918. Anal.
Calcd for C11H14O3: C, 68.02; H, 7.26. Found: C, 68.02; H,
7.43.All benzoylated products 3j and 3n -q were converted to
corresponding diols by alkaline hydrolysis. Then the absolute
configurations of the diols were confirmed by comparison of
their specific rotation with authentic data (1j,21 1n ,34 1o,35 1p ,36
and 1q34).
1
means of H NMR spectra.
(S)-4,4-Dibr om o-4,5-d ih yd r o-3H-d in a p h th o[2,1-c:1′,2′-
e]sta n n ep in (A).18 A was prepared according to the reported
method (91% ee).19,20
Kin etic Resolu tion of 1-P h en yl-1,2-eth a n ed iol (1j).
1-Phenyl-1,2-ethanediol (1 mmol) and catalyst A (0.0025 mmol)
were dissolved in 5 mL of THF. After a pulverized sodium
carbonate (1.5 mmol) was suspended in the solution, water
(100 µL) and benzoyl chloride (0.5 mmol), successively, were
added at -10 °C. The reaction mixture was stirred at -10 °C
for 14 h. Then, the suspension was poured into water, and the
organic portion was extracted three times with ethyl acetate.
The combined organic layer was dried over magnesium sulfate.
After removal of solvent, the residue was separated by silica
gel column chromatography (n-hexane:ethyl acetate ) 3:1).
Ack n ow led gm en t. This study was supported by
Grant-in-Aid for Scientific Research on Priority Areas,
No. 706: Dynamic Control of Stereochemistry and
No.11771414 from the Ministry of Education, Science,
and Culture, J apan, and one of the authors (T.M.) also
thanks Daicel Chemical Industries for general financial
support.
(34) Cho, B. T.; Chun, Y. S. J . Org. Chem. 1998, 63, 5280-5282.
(35) Miao, G.; Rossiter, B. E. J . Org. Chem. 1995, 60, 8424-8427.
(36) Ko, K.-Y.; Eliel, E. L. J . Org. Chem. 1986, 51, 5353-5362.
J O991394J