Efficient synthesis and hydrolysis of cyclic oxalate esters of glycols.

Article abstract of DOI:10.1248/cpb.50.346

Based on the mechanism postulated for the formation of the cyclic carbonates 3 in the reactions of glycols 1 with oxalyl chloride in the presence of triethylamine, we present here three efficient syntheses of the cyclic oxalates 2 of various glycols 1 by controlling the formation of 3: replacement of the base by pyridine markedly diminishes yields of 3 in all reactions, realizing dramatic reversals of the product ratios in the reactions with the (R*,R*)-compounds 1g-i,q,r and pinacol (1k); although considerable amounts of the oxalate polymers are formed in the reactions with some (R*,S*)-glycols, this drawback can be removed by the use of 2,4,6-collidine instead of pyridine; 1,1'-oxalyldiimidazole is useful for the synthesis of two selected cyclic oxalates 2e,f. The cyclic oxalates 2 other than trisubstituted and tetrasubstituted ones were found to be very reactive: kinetic studies on the hydrolysis of 1,4-dioxane-2,3-dione (2a) as well as its mono- and some selected 5,6-disubstituted derivatives 2 have revealed that they undergo hydrolysis 260-1500 times more rapidly than diethyl oxalate (12) in acetate buffer-acetonitrile (pH 5.69) at 25 degrees C. Although the cyclic oxalate 21 from cis-1,2-cyclopentanediol (11) was 1.5 times more reactive than 2a, it has been shown with other substrates that increasing number of the alkyl substituents decreases the rate of hydrolysis. On the contrary, the phenyl group was found to have somewhat accelerative effect.

Full text of DOI:10.1248/cpb.50.346

346
Chem. Pharm. Bull. 50(3) 346—353 (2002)
Vol. 50, No. 3
Efficient Synthesis and Hydrolysis of Cyclic Oxalate Esters of Glycols
b
Taisuke ITAYA, ,a Takehiko IIDA,a Yasuko GOMYO,a Itaru NATSUTANI,a and Masashi OHBA
*
Faculty of Pharmaceutical Sciencesa and Center for Instrumental Analysis,b Kanazawa University, Takara-machi,
Kanazawa 920–0934, Japan. Received October 12, 2001; accepted December 3, 2001
Based on the mechanism postulated for the formation of the cyclic carbonates 3 in the reactions of glycols 1
with oxalyl chloride in the presence of triethylamine, we present here three efficient syntheses of the cyclic ox-
alates 2 of various glycols 1 by controlling the formation of 3: replacement of the base by pyridine markedly di-
minishes yields of 3 in all reactions, realizing dramatic reversals of the product ratios in the reactions with the
(R*,R*)-compounds 1g—i, q, r and pinacol (1k); although considerable amounts of the oxalate polymers are
formed in the reactions with some (R*,S*)-glycols, this drawback can be removed by the use of 2,4,6-collidine in-
stead of pyridine; 1,1؅-oxalyldiimidazole is useful for the synthesis of two selected cyclic oxalates 2e, f. The cyclic
oxalates 2 other than trisubstituted and tetrasubstituted ones were found to be very reactive: kinetic studies on
the hydrolysis of 1,4-dioxane-2,3-dione (2a) as well as its mono- and some selected 5,6-disubstituted derivatives 2
have revealed that they undergo hydrolysis 260—1500 times more rapidly than diethyl oxalate (12) in acetate
buffer–acetonitrile (pH 5.69) at 25 °C. Although the cyclic oxalate 2l from cis-1,2-cyclopentanediol (1l) was 1.5
times more reactive than 2a, it has been shown with other substrates that increasing number of the alkyl sub-
stituents decreases the rate of hydrolysis. On the contrary, the phenyl group was found to have somewhat acceler-
ative effect.
Key words glycol cyclocondensation; oxalyl chloride; 1,4-dioxane-2,3-dione hydrolysis; oxalate hydrolysis kinetics; 1,1Ј-oxa-
lyldiimidazole; stereoelectronic effect
It has been established that for simple carboxylic esters the slow reaction at room temperature to afford the carbonate 3k
basic ester skeletal framework CC(O)OC is planar or near together with a small amount of the oxalate 2k.13) Although
planar and that the (Z)-form is much more stable than the the relationship between the product patterns and the config-
(E)-form.1) It has been also proposed that diesters of oxalic urations of cis-1,2-cycloalkanediols 1l—n, trans-1,2-cyclo-
acid mainly exist as the s-trans conformers.2) Accordingly, a heptanediol (1q), and trans-1,2-cyclooctanediol (1r) was
noteworthy feature of cyclic oxalate esters 2 formed from analogous to that obtained for 1,2-disubstituted acyclic ethyl-
glycols 1 is that two less stable (E)-esters are linked to form ene glycols 1d—i, trans-1,2-cyclopentanediol (1o) and trans-
the unfavorable s-cis structure of the 1,2-dicarbonyl system. 1,2-cyclohexanediol (1p) were exceptions to give no traces
Probably it accounts for that compounds with a 1,4-dioxane- of the cyclic carbonates 3o, p.14) As most of the cyclic ox-
2,3-dione framework are few notwithstanding that they are alates 2 described above were quite unstable under conditions
predictable to be normal products from reactions between ox- employed for separation by chromatography or distillation,
alyl chloride and glycols. Examples of this ring system can we failed to isolate 2e—i, o, q, r. Although the cyclic ox-
be seen as a part structure of the cyclic oxalates of unsubsti- alates 2e, h of hydrobenzoins (1e, h) were once reported to be
tuted3) and substituted 1,2-benzenediols.4) Some other com- produced by White,8) we suppose these were in fact the
pounds with four sp2-hybridized endocyclic carbons have cyclic carbonates 3e, h (see below). We thus desired to devise
been reported.5) Fewer examples of compounds with two sp3- a general procedure for the highly selective preparation of
hybridized ring carbons have been reported. The prototype of the cyclic oxalates 2 for further investigation of chemical
these compounds, 1,4-dioxane-2,3-dione (2a), and its 5- properties of this class of compounds. A preliminary account
methyl derivative 2b have been well investigated.6) The of a part of this work has been published.15)
cyclic oxalates 2 of meso-2,3-butanediol (1d),7) (Ϫ)-2,3-bu-
According to our proposed mechanism illustrated in Chart
tanediol,7) meso- and (Ϯ)-hydrobenzoins (1e, h),8) cis- and 2,13b) the formation of tetramethylethylene carbonate (3k)
trans-1,2-cyclohexanediols (1m, p),9) and some other cyclic from pinacol (1k) in the presence of oxalyl chloride and tri-
1,2-diols10) were reported to be prepared by acylating the cor-
responding glycols with oxalyl chloride. Tetrachloroethylene
oxalate was prepared by chlorination of 2a.11) A reaction
using (Ϯ)-5-vinyl-1,4-dioxane-2,3-dione was reported with-
out characterization and description of a method of prepara-
tion of this compound.12)
We have already disclosed that oxalyl chloride generally
reacts with acyclic glycols in tetrahydrofuran (THF) in the
presence of triethylamine at 0 °C to form the cyclic oxalates
2 together with the cyclic carbonates 3: unsubstituted (1a),
monosubstituted (1b, c), and erythro-1,2-disubstituted ethyl-
ene glycols 1d—f produce 2a—f and/or the polymeric ox-
alates as the major products, while the threo-isomers 1g—i
afford 3g—i as the major products. Pinacol (1k) underwent
Chart 1
To whom correspondence should be addressed. e-mail: itaya@dbs.p.kanazawa-u.ac.jp
© 2002 Pharmaceutical Society of Japan

March 2002
347
Chart 2
Table 1. Temperature Effect on the Reactions of Oxalyl Chloride with Hydrobenzoins (1e, h) in THF in the Presence of Triethylaminea)
Yield (%)b)
Temperature
(°C)
Reaction time
(min)
Substrate
2
3
Polymers
meso-Hydrobenzoin (1e)
(Ϯ)-Hydrobenzoin (1h)
0c)
Ϫ78d)
Ϫ78d)
0c)
15
195
435
15
195
435
67
45
29
36
47
47
33 (29)
37
34
64 (58)
53
53 (44)
0
18
37
0
0
0
Ϫ78d)
Ϫ78d)
a) Hydrobenzoin (1 mmol) was treated with oxalyl chloride (1.1 mmol) in the presence of triethylamine (2.2 mmol) in THF (12 ml). b) The yields were estimated by 1H-
NMR spectroscopy (CDCl3) on the basis of the intensity of the signal of the benzylic protons. Figures in parentheses denote isolated yields. c) Taken from ref. 13b. d) Carried
out in a dry ice–acetone bath.
ethylamine is interpreted in terms of stereochemically con- tiating polymerization as shown in Chart 3. Involvement of
trolled formation of the tetrahedral intermediate 6k·H from 2e in the polymerization was suggested by the decreasing
the initially formed s-trans intermediate 5k, followed succes- yield of 2e with increasing reaction time. It is not surprising
sively by deprotonation and stereoelectronically controlled that no polymerization occurred in the reaction of 1h, be-
cleavage (two nonbonding electron pairs contributing to bond cause the rotamer CЈ to undergo polymerization is the most
cleavage of 6Ϫk are shown as shaded lobes) of the C–C bond. unfavorable one.
The cyclic oxalate 2k can be formed only after 6Ϫk confor-
The solvent effects on the reactions of 1e, h with oxalyl
mationally changes into 7Ϫk. The almost exclusive formation chloride are summarized in Table 2. It was found that the
of 3k is attributable to the large rotational barrier between threo-compound 1h did not provide the oxalate polymers at
the tetrasubstituted tetrahedral intermediates (6Ϫk and 7Ϫk) all in every solvent tested. On the other hand, the erythro-
compared with the activation energy for the decay of 6Ϫk. compound 1e produced the polymers when the solvents other
However, the barriers to interconversion for the correspond- than ethers were employed. The polymerization was probably
ing disubstituted intermediates (6Ϫe, h and 7Ϫe, h) from hy- promoted by protonation of 2e with triethylamine hydrochlo-
drobenzoins (1e, h) are comparable to the activation energies ride. In the ethereal solvents such as THF, 1,4-dioxane, and
for their decay.13b) Furthermore, the transition-state energy diethyl ether, protonation would occur on the solvent mole-
level for the breakdown of the intermediates 6Ϫe, h is much cules rather than 2e, preventing acid-catalyzed polymeriza-
higher than that for the breakdown of the other 7Ϫe, h.13b) Ac- tion. It should be noted that the yield of the carbonate 3e
cordingly, we can expect to obtain 2e, h exclusively if we suf- from the erythro-compound 1e exceeded that of the total ox-
ficiently retard the fragmentation of the species 6e, h com- alates in the reaction in a polar solvent acetonitrile. Table 2
pared to their conformational change into 7e, h.
finally shows that no better solvent than THF was found for
preferential production of 2. Thus we failed in isolation of
pure cyclic oxalates 2e, h, notwithstanding that White re-
Results and Discussion
Table 1 shows that lowering the reaction temperature in- ported the preparation of these compounds, without any
creased the yield of the cyclic oxalate 2h from the threo- characterization, by the reactions of 1e, h with oxalyl chlo-
compound 1h to a slight extent, but resulted in the formation ride in the presence of triethylamine in ether.8)
of considerable amounts of the oxalate polymers from the
The cyclic oxalates 2m, p of cis- and trans-1,2-cyclohex-
erythro-compound 1e at the cost of the cyclic oxalate 2e. The anediols (1m, p) have already been prepared by azeotropic
formation of the polymers is probably due to slow intercon- distillation of water from a mixture of the diol, oxalic acid,
version among the rotamers (A, B, and C) of the half-ester 5e and benzene.9) As shown in Chart 4 for the reaction of 1p,
at lower temperature: the rotamer C, which is unable to un- the successful result is conceivable by postulating the tetra-
dergo intramolecular cyclization, would react with either an- hedral intermediate 9,14) which would lead to 2p through
other molecule of the rotamer C or the cyclic oxalate 2e, ini- 9·Hϩ. We indeed obtained 2p in 79% yield according to this

348
Vol. 50, No. 3
Chart 3
Table 2. Reactions of Oxalyl Chloride with Hydrobenzoins (1e, h) in the Presence of Triethylamine in Various Solventsa)
meso-Hydrobenzoin (1e)
(Ϯ)-Hydrobenzoin (1h)
Yield (%)b)
Yield (%)b)
Solvent
2e
3e
Polymers
2h
3h
Polymers
Benzenec)
48
43
20
26
52
67
55
20
33 (30)
32 (31)
42 (29)
40 (31)
48 (41)
33 (29)
30 (30)
54 (45)
19
23
38
34
0
0
0
26
26
25
16
17
24
36
31
22
74 (60)
75 (61)
81 (72)
83 (69)
68 (62)
64 (58)
65 (59)
78 (58)
0
0
0
0
0
0
0
0
Toluened)
Chloroformd)
Dichloromethaned)
Diethyl etherd)
THFd,e)
1,4-Dioxanef)
Acetonitriled)
a) Hydrobenzoin (1 mmol) was treated with oxalyl chloride (1.1 mmol) in the presence of triethylamine (2.2 mmol) in the solvent (12 ml) for 20 min. b) The yields were es-
timated by 1H-NMR spectroscopy. Figures in parentheses denote isolated yields. c) Carried out at 5 °C. d) Carried out at 0 °C. e) Taken from ref. 13b. f) Carried out at
room temperature.
Chart 4
procedure, but 2h could not be obtained in a similar man-
ner.16)
Assuming that the neutral intermediate 6h·H from 1h does
not undergo fragmentation, but 7h·H changes into 2h as il-
lustrated in Chart 5, we might effectively suppress the forma-
tion of 3h by preventing the dissociation of 6h·H into the
anion 6Ϫh. In the absence of a base, however, we already
found that 1k provided 3k probably through 6k·H2ϩ.13b) Our
next task was accordingly to test an appropriately weaker
base than triethylamine. Table 3 summarizes the base effects
on the product patterns of the reactions of 1e, h. It was found
that the use of pyridine most effectively controlled the forma-
tion of 3h. We thus obtained 2h in 69% yield for the first
Chart 5
time. Although the formation of 3e was also effectively in-

March 2002
349
hibited by the use of pyridine, considerable amounts of the col (1k). Thus compound 2g, i, q, r were obtained in 50%,
oxalate polymers were formed at the cost of the yield of 2e. 75%, 80%, and 82% yields, respectively, for the first time.
The formation of the polymers might be accounted for in Compound 2k, which had been obtained in only 0.8%
view of stronger acidity of pyridine hydrochloride than that yield,13b) was produced in 85% yield by the present method.
of triethylamine hydrochloride. Although it was possible to Relatively slow reaction of the tertiary diol 1k evoked our in-
suppress the formation of the polymers by increasing the vol- terest in the reaction products from (Ϯ)-2-methyl-2,3-bu-
ume of the solvent (Table 4, entry 5) or by the use of 2,4,6- tanediol (1j). This compound underwent reaction with oxalyl
collidine instead of pyridine, we failed to remove a small chloride at 0 °C, giving the cyclic oxalate 2j and presumably
amount of 3e by means of recrystallization.
a diastereomeric mixture of bis(2-hydroxy-1,2-dimethyl-
The results reported in Table 4 stress the importance of the propyl) oxalates. Although the main product 2j is not so sta-
use of pyridine in place of triethylamine for dramatic rever- ble as 2k, it could be obtained by flash chromatography on
sals of the product ratios in the reactions with other threo- silica gel in 16% yield together with the latter (20%). Unfor-
compounds 1g, i, trans-1,2-cycloalkanediols 1q, r and pina- tunately, the reactions of the erythro-compounds 1d, f were
Table 3. Reactions of Oxalyl Chloride with Hydrobenzoins (1e, h) in the Presence of Different Bases in THFa)
meso-Hydrobenzoin (1e)
(Ϯ)-Hydrobenzoin (1h)
Yield (%)b)
Yield (%)b)
Base
2e
3e
Polymers
2h
3h
Polymers
Triethylamine
67c)
67
—
33 (29)c)
31 (26)
—
0c)
0
36
16
26
5
64 (58)
84 (70)
74 (71)
91 (86)
87 (70)
0
0
0
0
0
0
0
0
0
—
—
—
—
—
Ethyldiisopropylamine
Cyclohexyldiethylamine
1,2,2,6,6-Pentamethylpiperidine
N,N-Diethylaniline
Pyridine
—
0
49
—
39
—
d)
d)
d)
—
13
e)
27
—
95
—
0
2 (2)
—
5
—
0
—
92 (69)
87 (38)
87 (45)
82 (33)
—
—
—
—
—
7
(5)
2,6-Lutidine
2,4,6-Collidine
—
0
—
0
12 (12)
13 (8)
10 (9)
—
2-Methoxypyridine
DBNf)
Tributylphosphine
Butyllithium
Sodium hydride
Potassium carbonate
47
—
—
0
—
—
6
—
—
—
d)
d)
d)
d)
d)
d)
—
d)
d)
d)
d)
d)
d)
—
0
0
0
—
a) Hydrobenzoin (0.5 mmol) was treated with oxalyl chloride (0.55 mmol) in the presence of base (2.5 mmol) in THF (12 ml) at 0 °C for 15 min. b) The yields were esti-
mated by 1H-NMR spectroscopy. Figures in parentheses denote isolated yields. c) Taken from ref. 13b (2.2 mol eq of triethylamine was used). d) A complex mixture of prod-
ucts. e) Not determined. f) DBN: 1,5-diazabicyclo[4.3.0]non-5-ene.
Table 4. Reactions of Oxalyl Chloride and Glycols 1 in THF in the Presence of Pyridine
Reaction conditionsa)
Yield (%)b)
Entry
Glycol
R1
R2
R3
R4
Solventc)
(ml)
Temp.
(°C)
Time
(min)
2
3
Polymers
d)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1a
H
H
H
H
H
H
H
H
Me
Ph
Ph
H
H
H
H
Me
H
H
H
H
H
H
H
H
H
H
Me
Ph
5
5
110
110
113
24
110
24
26
0
0
0
0
0
0
0
0
0
0
r.t.
0
0
0
0
0
0
0
20
25
30
40
60
15
40
15
15
60
480
40
40
40
40
40
40
40
—
d) (90e))
0
0
—
1bf)
1cf)
1d
26 (71e))
81 (48)
74
19
51
8
74
8
0
49 (49e))
88 (—h))
25 (—h))
92 (50e))
92 (69)
0g)
4
1e
Ph
Me2CH
Me
Ph
Me2CH
Me
1ff)
1gf)
1hf)
1if)
1jf)
1k
H
1
0
7 (5)
4 (2)
0
Me
Ph
Ph
Me
Me
0
95 (75)
0
d)
Me
Me
25
24
—
—
d) (16i,j))
d) (85i))
—
d)
Me
—
d) (0.7)
—
1l
R1ϭR3ϭH, R2R4ϭ(CH2)3
R1ϭR3ϭH, R2R4ϭ(CH2)4
R1ϭR3ϭH, R2R4ϭ(CH2)6
R2ϭR3ϭH, R1R4ϭ(CH2)3
R2ϭR3ϭH, R1R4ϭ(CH2)4
R2ϭR3ϭH, R1R4ϭ(CH2)5
R2ϭR3ϭH, R1R4ϭ(CH2)6
110
110
110
110
110
110
110
88 (31e))
79 (58e))
64 (38e))
10 (—h))
78 (74e))
87 (80)
0g)
9
14
33
90
9
1m
1n
0
3
0
0
1
1of)
1pf)
1qf)
1rf)
12
5
92 (82e))
3 (0.9)
a) Except for entries 4, 7, 12, 15, and 17, the diols 1 were treated with 1.1 mol eq of oxalyl chloride and 5 mol eq of pyridine; compounds 1o, q were treated with 1.3 mol eq of
oxalyl chloride and 5 mol eq of pyridine; compounds 1d, g, l were treated with 1.5 mol eq of oxalyl chloride and 7.5 mol eq of pyridine. b) The yields were estimated by 1H-NMR
spectroscopy. Figures in parentheses denote isolated yields. c) Volume per mmol of 1. d) Not determined. e) Obtained by Kugelrohr distillation of a mixture of the products.
f) Racemic modification. g) Determined by means of thin-layer chromatography. h) Could not be isolated. i) Obtained by flash chromatography on silica gel. j) A di-
astereomeric mixture of bis(2-hydroxy-1,2-dimethylpropyl) oxalates was obtained in 20% yield.

350
Vol. 50, No. 3
again disturbed by the formation of the polymers. Although composed on silica gel almost instantly. As had been ob-
2d was obtainable by pyrolytic distillation,13b) isolation of the served with the cyclic oxalates of meso-2,3-butanediol (1d)
benzylic compound 2f was unsuccessful. However, this com- and (Ϫ)-2,3-butanediol,7) 2a—i, l—r were all highly sen-
pound 2f was obtained in 76% yield by the use of 2,4,6-colli- sitive to hydrolysis even in plain water at room tempera-
dine. On the other hand, neither pyridine (entry 15) nor 2,4,6- ture. When a solution of 2h in water–acetone (1 : 1, v/v) was
collidine17) had the advantage over triethylamine14) for the kept at room temperature for 24 h, (R*,R*)-(2-hydroxy-1,2-
formation of the cyclic oxalate 2o from trans-1,2-cyclopen- diphenyl)ethyl hydrogen ethanedioate (11h) was obtained in
tanediol (1o). We have already reported the preparation of 94% yield. The diastereomer 11e was not formed, showing
the cyclic oxalates 2m, p from cis- and trans-1,2-cyclohexa- that unusual SN1 mechanism did not work in the hydrolysis
nediols (1m, p) by the reaction with oxalyl chloride in the of 2h. The half-ester 11h did not undergo further reaction
presence of triethylamine.14) These compounds 2m, p were under these conditions. The rate of the first hydrolysis of 2h
also obtainable in the presence of pyridine (entries 13 and
16). It should be stated in this connection that syntheses of
the cyclic oxalates have already been reported by the reac-
tions of cyclohexanediols such as 1-benzyloxy-5,6,7,8-
tetrahydronaphthalene-6,7-diol10a,b) and methyl 4,6-O-ben-
zylidene-a-D-glucopyranoside10c) with oxalyl chloride in the
presence of pyridine.
The strategy we finally adopted to obtain hitherto unavail-
able 2e was to retard the C–C bond cleavage by canceling the
stereoelectronic effect from the chlorine atom in the interme-
diate (type 6·H or 6Ϫ in Chart 2), which is formed in the re-
action with oxalyl chloride. The reagent we selected was
1,1Ј-oxalyldiimidazole. Because the nonbonding electron
pair on the N(1) atom of the putative intermediate 10 should
Chart 6
be employed for aromatization of the imidazole ring, it is of
no assistance in breaking the C–C bond as shown in Chart 6.
Indeed, no trace of the carbonate 3e was formed from 1e in
the reaction with 1,1Ј-oxalyldiimidazole, and the pure cyclic
oxalate 2e was obtained in 64% yield. Compound 2f was also
obtainable in 69% yield according to this procedure.
The cyclic oxalates other than 2j, k were found to be de-
Table 5. Pseudo-First-Order Rate Constants for the First Hydrolysis of the Oxalate Esters in 0.07 M Acetate Buffer (pH 5.69) in Aqueous Acetonitrile at
25 °C
Entry
1
Substrate
k1 (minϪ1
)
Entry
8
Substrate
k1 (minϪ1
)
12
2a
1.5ϫ10Ϫ4
2c
(2.1Ϯ0.1)ϫ10Ϫ1
(2.2Ϯ0.1)ϫ10Ϫ1
2
(1.5Ϯ0.1)ϫ10Ϫ1
9
2h
3
4
2b
2d
(1.1Ϯ0.0)ϫ10Ϫ1
(7.8Ϯ0.8)ϫ10Ϫ2
10
11
2i
2l
(9.0Ϯ0.1)ϫ10Ϫ2
(2.3Ϯ0.1)ϫ10Ϫ1
5
2g
(8.6Ϯ0.2)ϫ10Ϫ2
12
2m
(7.5Ϯ0.0)ϫ10Ϫ2
6
7
2j
7.4ϫ10Ϫ3
1.5ϫ10Ϫ4
13
14
2p
2n
(6.4Ϯ0.3)ϫ10Ϫ2
(4.7Ϯ0.0)ϫ10Ϫ2
2k
15
2r
(3.9Ϯ0.2)ϫ10Ϫ2

March 2002
351
stirred at room temperature for 3 h. Sodium metabisulfite (2.7 g, 14 mmol)
was added to the resulting mixture, and stirring was continued for a further
30 min. The mixture was diluted with water (50 ml), brought to pH 5 with
10% hydrochloric acid, saturated with sodium chloride, and extracted with
dichloromethane (3ϫ20 ml) and ethyl acetate (7ϫ20 ml). The combined or-
was determined in aqueous acetonitrile because the substrate
is hardly dissolved in plain water. As it was difficult to follow
too rapid hydrolysis of 2h at pH 6 or above, the pseudo-first-
order rate constant was determined for the reaction in 0.07 M
acetate buffer in aqueous acetonitrile (pH 5.69) at 25 °C. The ganic layers were dried over magnesium sulfate and concentrated in vacuo to
leave a brown oil (662 mg). Flash chromatography [hexane–ethyl acetate
result is shown in Table 5 together with those obtained for
the reactions of 2a—d, g, i—n, p, r.
(1 : 4, v/v)] of this product provided 1j19) (549 mg, 53%) as a colorless oil;
neat
IR n
cmϪ1: 3400 (OH). 1H-NMR (CDCl3) d: 1.16 (3H, d, Jϭ6.6 Hz,
The pseudo-first-order rate constant (1.5ϫ10Ϫ1 min, t1/2
4.6 min) for the first hydrolysis of the prototype 2a of a
cyclic oxalate (entry 2) was 1000 times larger than that
(1.5ϫ10Ϫ4 min) of the acyclic analogue 12 (entry 1) under
identical conditions. Among 2a—d, g—i, l—n, p, r the cy-
clopentane derivative 2l underwent hydrolysis fastest (entry
11), but not more than only six times faster than the most un-
reactive one 2r (entry 15). It can be seen from entries 8 and 9
that phenyl substituent(s) rather accelerate the reaction, while
alkyl substituents (entries 3—7, 10, 12—15) other than
trimethylene (entry 11) retard the reaction. Entries 2—7
show that increasing number of methyl substituents de-
creases the rate of hydrolysis. Thus the cyclic oxalate 2j with
three methyl groups underwent hydrolysis 20 times more
slowly than 2a. The rate for the hydrolysis of fully substi-
tuted one 2k was as small as that of acyclic oxalate 12. This
max
CHMe), 1.16, 1.21 (3H each, s, CMe2), 1.96 (1H, s, Me2COH), 2.19 (1H, d,
Jϭ4.0 Hz, CHOH), 3.62 (1H, dq, Jϭ4.0, 6.6 Hz, CHMe).
Reactions of Glycols with Oxalyl Chloride in the Presence of Pyridine
The procedures for the synthesis of new compounds 2g—j, q, r are de-
scribed below as typical examples. Other compounds 2a—d, k13b) and 2l—
n, p14) obtained under conditions specified in Table 4 were identical (by com-
1
parison of the IR and H-NMR spectra) with their corresponding authentic
samples.
(؎)-trans-5,6-Dimethyl-1,4-dioxane-2,3-dione (2g) A solution of 1g
(270 mg, 3 mmol) and pyridine (1.85 ml, 22.9 mmol) in THF (300 ml) was
cooled with ice-water and then a solution of oxalyl chloride (0.38 ml,
4.5 mmol) in THF (30 ml) was added dropwise over a period of 30 min. The
mixture was stirred at 0 °C for a further 10 min and concentrated in vacuo to
ca. 20 ml. The resulting precipitate was filtered off and washed with THF
(20 ml). The filtrate and washings were combined and concentrated in vacuo.
The residue was submitted to Kugelrohr distillation. The distillate (50 mg)
obtained at 120—140 °C and 0.8 mmHg was recrystallized from carbon
tetrachloride to give a first crop of 2g (18 mg), mp 80.5—82 °C (softened
below 40 °C). Pure 2g (151 mg, the total yield was 50%) was obtained as the
exceptionally low reactivity of 2k may be interpreted in distillate running at 140—180 °C and 0.8 mmHg, mp 80.5—82 °C (softened
below 40 °C). Recrystallization of this sample from carbon tetrachloride
terms of steric hindrance of the methyl groups: as shown in
Chart 2, two of the four methyl groups of 2k are always situ-
ated almost perpendicular to the plane consisting of the two
carbonyl groups,18) blocking either side of the carbonyl
plane from nucleophilic attack.
gave an analytical sample of 2g as colorless pillars, mp 82.5—84.5 °C (soft-
MeCN
ened at 30—35 °C). MS m/z: 145 (Mϩϩ1). UV l
nm (e): 272 (54). IR
max
Nujol
max
CHCl3
cmϪ1: 1780 (CϭO). 1H-NMR
max
n
cmϪ1: 1771, 1748 (CϭO); n
(CDCl3) d: 1.46 (6H, m, two Me’s), 4.59 (2H, m, two CH’s). 13C-NMR
(CDCl3) d: 16.4 (Me), 78.8 (CH), 153.3 (CϭO). Anal. Calcd for C6H8O4: C,
50.00; H, 5.59. Found: C, 49.78; H, 5.59.
In conclusion, various cyclic oxalates 2 have become ac-
cessible by the reactions between oxalyl chloride and glycols
1 by controlling the formation of the cyclic carbonates 3 by
the use of pyridine instead of triethylamine. 2,4,6-Collidine
was shown superior to pyridine in suppressing the formation
of polymers for the reactions with erythro-1,2-disubstituted
ethylene glycols 1e, f, and imidazole was a good choice of
bases for the selective formation of the cyclic oxalates 2e, f.
Preliminary kinetic studies with some selected cyclic ox-
alates 2 have revealed that these compounds other than
trisubstituted and tetrasubstituted ones undergo hydrolysis
260—1500 times more rapidly than diethyl oxalate (12). It
also has been suggested that the number and the nature of the
substituents largely affect the rate of hydrolysis. More de-
tailed information about the structure–reactivity relationship
has to await further studies.
(؎)-trans-5,6-Diphenyl-1,4-dioxane-2,3-dione (2h) A solution of oxa-
lyl chloride (0.19 ml, 2.2 mmol) in THF (8 ml) was added to an ice-cooled
solution of 1h (429 mg, 2 mmol) and pyridine (0.80 ml, 9.9 mmol) in THF
(40 ml) over a period of 5 min. The mixture was stirred at 0 °C for 10 min
and concentrated in vacuo. The residue was partitioned between ethyl ac-
etate (50 ml) and water (20 ml). The organic layer was washed successively
with 5% aqueous citric acid and saturated aqueous sodium bicarbonate
(20 ml each), dried over magnesium sulfate, and concentrated to afford a
solid residue (526 mg), mp 138—154 °C. The molar ratio of 2h and 3h in
this product was 93 : 7. Recrystallization of this material from hexane–
dichloromethane (2 : 1, v/v) afforded 2h (370 mg, 69%) as colorless prisms,
mp 160.5—163.5 °C. The mother liquor was concentrated and the residue
was purified by flash chromatography [hexane–dichloromethane (1 : 10,
v/v)] followed by thin-layer chromatography on silica gel [hexane–ethyl ac-
etate (2 : 1, v/v)] to afford 3h13b) (26 mg, 5.4%), mp 107.5—109 °C.
Further recrystallization of crude 2h from ether–dichloromethane (5 : 1,
v/v) afforded an analytical sample of 2h as colorless prisms, mp 164—
MeCN
165 °C. MS m/z: 268 (Mϩ). UV l
nm (e): 251 (560), 257 (620), 262
max
Nujol
CHCl3
(580), 267 (400). IR n
cmϪ1: 1782, 1755 (CϭO); n
cmϪ1: 1782
max
max
(CϭO). 1H-NMR (CDCl3) d: 5.69 (2H, s, two CH’s), 7.07 (4H), 7.21—7.36
(6H) (m each, two Ph’s). 13C-NMR (CDCl3) d: 84.3 (CH), 127.3, 128.7,
130.0, 131.5 (Ph), 153.0 (CϭO). Anal. Calcd for C16H12O4: C, 71.64; H,
4.51. Found: C, 71.75; H, 4.41.
Experimental
General Notes All melting points were determined using a Yamato MP-
1 or Büchi model 530 capillary melting point apparatus and values are cor-
rected. Spectra reported herein were recorded on a JEOL JMS-SX102A
mass spectrometer, a Hitachi model 320 UV spectrophotometer, a Shimadzu
FTIR-8100 or a FTIR-8400 IR spectrophotometer, a JEOL JNM-EX-270 or
a JEOL JNM-GSX-500 NMR spectrometer (measured at 25 °C with Me4Si
as an internal standard). For the measurements of pH values, a Toa HM-
18ET pH meter equipped with a Toa type GST-5211C glass electrode was
employed. Elemental analyses and MS measurements were performed by Dr.
M. Takani and her associates at Kanazawa University. The following abbre-
viations are used: brϭbroad, dϭdoublet, ddϭdoublet-of-doublets, dddϭ
doublet-of-doublets-of-doublets, dqϭdoublet-of-quartets, dqqϭdoublet-of-
quartets-of-quartets, mϭmultiplet, qϭquartet, sϭsinglet, shϭshoulder.
(؎)-2-Methyl-2,3-butanediol (1j) A 2.3% (w/v) osmium tetroxide so-
lution (1 ml) in 1,1-dimethylethanol was added to a mixture of 2-methyl-2-
butene (700 mg, 10 mmol), N-methylmorpholine-N-oxide monohydrate
(1.62 g, 12 mmol), water (5 ml), and acetone (50 ml), and the whole was
(؎)-trans-5-Isopropyl-6-phenyl-1,4-dioxane-2,3-dione (2i) Compound
1i (83 mg, 0.46 mmol) was treated with oxalyl chloride in a manner similar
to that described for the preparation of 2h, and the reaction mixture was con-
centrated in vacuo. The residue was partitioned between ethyl acetate
(20 ml) and water (10 ml). The organic layer was washed successively with
5% aqueous citric acid and saturated aqueous sodium bicarbonate (10 ml
each), dried over magnesium sulfate, and concentrated to afford a solid
residue (106 mg), mp 96—103 °C. The molar ratio of 2i and 3i in this prod-
uct was 96 : 4. Recrystallization of this material from hexane–ether (1 : 1,
v/v) afforded 2i (81 mg, 75%) as colorless prisms, mp 113.5—116 °C. The
mother liquor of recrystallization was concentrated and the residue was puri-
fied by thin-layer chromatography on silica gel [hexane–ethyl acetate (4 : 1,
v/v)] to afford 3i13b) (2 mg, 2.1%) as a colorless viscous oil.
Further recrystallization of 2i afforded an analytical sample of 2i as color-
MeCN
max
less prisms, mp 118—119.5 °C. MS m/z: 234 (Mϩ). UV l
nm (e): 250

352
Vol. 50, No. 3
Nujol
max
(280), 257 (320), 262 (300), 267 (210). IR n
max
cmϪ1: 1790, 1761 (CϭO);
liquor of recrystallization was concentrated and the residue was purified by
thin-layer chromatography on silica gel [hexane–ethyl acetate (2 : 1, v/v)] to
afford 3f13b) (7 mg, 6%), mp 60—61.5 °C.
Further recrystallization of crude 2f from carbon tetrachloride afforded an
analytical sample of 2f as colorless prisms, mp 95.5—96.5 °C. MS m/z: 234
CHCl3
n
cmϪ1: 1784 (CϭO). 1H-NMR (CDCl3) d: 1.02, 1.06 (3H each, d,
Jϭ6.8 Hz, Me2), 1.66 (1H, dqq, Jϭ6.8, 6.8, 2.4 Hz, CHMe2), 4.67 (1H, dd,
Jϭ2.4, 8.8 Hz, CHCHMe2), 5.57 (1H, d, Jϭ8.8 Hz, PhCH), 7.36—7.42
(2H), 7.44—7.51 (3H, m each, Ph). 13C-NMR (CDCl3) d: 14.8, 19.4 (Me2),
28.4 (CMe2), 82.1 (CCMe2), 85.8 (PhC), 127.3, 129.4, 130.5, 132.2 (Ph),
153.2, 153.5 (CϭO). Anal. Calcd for C13H14O4: C, 66.66; H, 6.02. Found: C,
66.67; H, 5.99.
CH3CN
(Mϩ). UV l
(310). IR n
nm (e): 250 (sh) (510), 257 (470), 262 (430), 268 (sh)
max
Nujol
max
CHCl3
cmϪ1: 1790, 1761 (CϭO); n
cmϪ1: 1784, 1765 (CϭO).
max
1H-NMR (CDCl3) d: 0.98, 1.07 (3H each, d, Jϭ6.8 Hz, Me2), 1.88 (1H, m,
CHMe2), 4.67 (1H, dd, Jϭ2.9, 7.8 Hz, CHCHMe2), 5.80 (1H, d, Jϭ2.9 Hz,
PhCH), 7.32—7.37 (2H), 7.43—7.48 (3H) (m each, Ph). 13C-NMR (CDCl3)
d: 18.3, 19.2 (Me2), 28.6 (CMe2), 80.9 (CCMe2), 85.2 (PhC), 127.1, 129.3,
129.9, 131.8 (Ph), 153.0, 153.4 (CϭO). Anal. Calcd for C13H14O4: C, 66.66;
H, 6.02. Found: C, 66.43; H, 5.98.
5,5,6-Trimethyl-1,4-dioxane-2,3-dione (2j) A solution of oxalyl chlo-
ride (0.27 ml, 3.2 mmol) in THF (12 ml) was added to an ice-cooled solution
of 1j (297 mg, 2.85 mmol) and pyridine (1.15 ml, 14.2 mmol) in THF (60 ml)
over a period of 10 min. The mixture was stirred at 0 °C for 50 min and con-
centrated in vacuo. The residue was partitioned between ethyl acetate
(30 ml) and water (10 ml). The organic layer was washed successively with
water, 5% aqueous citric acid, water, and saturated aqueous sodium bicar-
bonate (10 ml each), dried over magnesium sulfate, and concentrated to af-
ford a colorless oil (272 mg). This was a mixture of 2j and bis(2-hydroxy-
1,2-dimethylpropyl) oxalates (the molar ratio was 3 : 1). Flash chromatogra-
phy [hexane–ethyl acetate (1 : 1, v/v)] of this product gave 2j (74 mg, 16%)
Reaction of meso-Hydrobenzoin (1e) with Oxalyl Chloride in the
Presence of Imidazole Oxalyl chloride (0.17 ml, 1.99 mmol) was added
dropwise to an ice-cooled solution of imidazole (545 mg, 8.01 mmol) in
THF (20 ml), and the mixture was stirred at 0 °C for 10 min under nitrogen.
To the resulting yellow suspension was added a solution of 1e (429 mg,
2 mmol) in THF (20 ml) over a period of 10 min. The mixture was then
stirred at room temperature under nitrogen for 24 h. The resulting precipitate
was filtered off and washed with THF (10 ml). The filtrate and washings
were combined and concentrated in vacuo. The residue was partitioned be-
tween ethyl acetate (70 ml) and water (30 ml). The organic layer was washed
successively with 5% aqueous citric acid, water, saturated aqueous sodium
bicarbonate, and water (30 ml each), dried over magnesium sulfate, and con-
centrated in vacuo, leaving cis-5,6-diphenyl-1,4-dioxane-2,3-dione (2e)
(342 mg, 64%) as a colorless solid, mp 125—128 °C. Recrystallization of
this product from hexane–ethyl acetate (1 : 1, v/v) afforded an analytical
sample of 2e as colorless prisms, mp 128—129.5 °C. MS m/z: 268 (Mϩ).
neat
max
1
as a colorless oil, MS m/z: 159 (Mϩϩ1). IR n
cmϪ1: 1779 (CϭO). H-
NMR (CDCl3) d: 1.45 (3H, d, Jϭ6.6 Hz, CHMe), 1.49 (6H, s, CMe2), 4.70
(1H, q, Jϭ6.6 Hz, CHMe). 13C-NMR (CDCl3) d: 15.7, 20.5, 24.8 (Me), 80.1
(CHMe), 84.0 (CMe2), 153.1, 153.5 (CϭO).
Further elution of the column afforded a 1 : 1 mixture of meso- and (Ϯ)-
bis(2-hydroxy-1,2-dimethylpropyl) oxalates (76 mg, 20%) as a colorless oil,
MS m/z: 245 (MϩϪOH). IR n
cmϪ1: 3409 (OH), 1765, 1732 (CϭO). 1H-
neat
max
NMR (CDCl3) d: 1.25 (24H, s, four CMe2’s), 1.33 (12H, d, Jϭ6.3 Hz, four
CHMe’s), 1.85, 2.48 (2H each, br, four OH’s), 4.91, 4.92 (2H each, q,
Jϭ6.3 Hz, four CHMe’s). 13C-NMR (CDCl3) d: 14.6, 14.7, 24.2, 24.4, 26.2
(Me), 71.79, 71.83 (CHMe), 80.4 (CMe2), 157.4, 157.5 (CϭO).
CH3CN
max
Nujol
cmϪ1
:
max
UV l
nm (e): 257 (930), 262 (800), 267 (sh) (570). IR n
CHCl3
1777, 1771, 1765 (CϭO); n
cmϪ1: 1788, 1769 (CϭO). 1H-NMR
(؎)-trans-Hexahydro-5H-cyclohepta-1,4-dioxin-2,3-dione (2q) A so-
lution of 1q20) (260 mg, 2 mmol) and pyridine (0.8 ml, 10 mmol) in THF
(200 ml) was cooled with ice-water and then a solution of oxalyl chloride
(0.22 ml, 2.6 mmol) in THF (20 ml) was added dropwise over a period of
30 min under nitrogen. The mixture was stirred at 0 °C for a further 10 min,
and the resulting precipitate was filtered off. The filtrate was concentrated in
vacuo, and the solid residue was recrystallized from carbon tetrachloride to
give 2q (301 mg, 80%). Further recrystallization from carbon tetrachloride
and drying over phosphorus pentoxide at 2 mmHg and room temperature for
17 h afforded an analytical sample of 2q as colorless needles, mp 80—
max
(CDCl3) d: 6.01 (2H, s, two CH’s), 6.88—6.94 (4H), 7.25—7.32 (4H),
7.33—7.38 (2H) (m each, two Ph’s). 13C-NMR (CDCl3) d: 82.0 (CH), 126.8,
128.6, 129.6, 130.7 (Ph), 153.1 (CϭO). Anal. Calcd for C16H12O4: C, 71.64;
H, 4.51. Found: C, 71.67; H, 4.46.
(R*,S*)-2-Hydroxy-1,2-diphenylethyl Hydrogen Ethanedioate (11e)
Water (15 ml) was added to a solution of a mixture of 2e and 3e (the molar
ratio was 2 : 1, 514 mg) in acetone (15 ml), and the resulting solution was al-
lowed to stand at room temperature for 24 h. The solution was concentrated
to half the initial volume in vacuo, and saturated aqueous sodium bicarbon-
ate (30 ml) was added to the resulting suspension. The solution thus obtained
was washed successively with chloroform (30 ml) and ethyl acetate
(2ϫ30 ml), brought to pH 1 by addition of 10% hydrochloric acid, and ex-
tracted with ether (2ϫ20 ml). The ethereal extracts were dried over magne-
sium sulfate and concentrated in vacuo, leaving 11e (233 mg) as a colorless
solid, mp 117—120 °C (dec.). Recrystallization of this product from
Nujol
max
CHCl3
cmϪ1: 1767, 1749 (CϭO); n
max
84.5 °C. MS m/z: 185 (Mϩϩ1). IR n
cmϪ1: 1778 (CϭO). 1H-NMR (CDCl3) d: 1.42—1.93 (8H), 2.09—2.17 [2H,
m each, (CH2)5], 4.53 (2H, m, two CH’s). 13C-NMR (CDCl3) d: 21.8, 26.3,
28.8 (CH2), 83.4 (CH), 153.6 (CϭO). Anal. Calcd for C9H12O4·1/4H2O: C,
57.29; H, 6.68. Found: C, 57.26; H, 6.69.
(؎)-trans-Octahydrocycloocta-1,4-dioxin-2,3-dione (2r) A solution of
1r (288 mg, 2 mmol) and pyridine (0.81 ml, 10 mmol) in THF (200 ml) was
cooled with ice-water and then a solution of oxalyl chloride (0.19 ml,
2.2 mmol) in THF (20 ml) was added dropwise over a period of 30 min. The
mixture was stirred at 0 °C for a further 10 min and concentrated in vacuo to
ca. 20 ml. The resulting precipitate was filtered off and washed with THF
(30 ml). The filtrate and washings were combined and concentrated in vacuo.
The residue was submitted to Kugelrohr distillation. Compound 2r (325 mg,
82%) was collected at 200—250 °C and 0.8 mmHg, mp 111—117 °C. Re-
crystallization of this product from carbon tetrachloride gave an analytical
sample of 2r as colorless pillars, mp 119—120 °C. MS m/z: 199 (Mϩϩ1).
hexane–ethyl acetate (1 : 1, v/v) afforded an analytical sample of 11e as col-
Nujol
orless prisms, mp 129.5—130.5 °C (dec.). MS m/z: 286 (Mϩ). IR n
max
cmϪ1: 3328 (OH), 1761, 1698 (COCO2H). H-NMR [(CD3)2SO] d: 3.4, 5.8
1
(1H each, br, OH and CO2H), 4.95 (1H, d, Jϭ5.9 Hz, CHOH), 5.86 (1H, d,
Jϭ5.9 Hz, CHOCOCO2H), 7.19—7.32 (10H, m, two Ph’s). 13C-NMR
[(CD3)2SO] d: 74.0 (CHOH), 80.4 (CHCHOH), 127.1, 127.4, 127.6, 127.8,
128.0, 136.4, 140.9 (Ph), 158.2, 159.1 (COCO2H). Anal. Calcd for
C16H14O5: C, 67.13; H, 4.93. Found: C, 66.95; H, 4.87.
(R*,R*)-2-Hydroxy-1,2-diphenylethyl Hydrogen Ethanedioate (11h)
Water (7 ml) was added to a solution of 2h (188 mg, 0.701 mmol) in acetone
(7 ml), and the resulting solution was allowed to stand at room temperature
for 24 h. The solution was concentrated to half the initial volume in vacuo,
and saturated aqueous sodium bicarbonate (20 ml) was added to the resulting
suspension. The solution thus obtained was washed with dichloromethane
(2ϫ10 ml), brought to pH 1 by addition of 10% hydrochloric acid, and ex-
tracted with ethyl acetate (3ϫ30 ml). The ethyl acetate layers were com-
bined, dried over magnesium sulfate, and concentrated in vacuo, leaving 11h
(188 mg, 94%), mp 126—127.5 °C (dec.). Recrystallization of this product
from hexane–ether (1 : 1, v/v) afforded an analytical sample of 11h as color-
less prisms, mp 139.5—140.5 °C (dec.). Chemical ionization (CI)-MS m/z:
CH3CN
max
Nujol
cmϪ1: 1779, 1771, 1759, 1750
max
UV l
nm (e): 273 (56). IR n
CHCl3
max
(CϭO); n
cmϪ1: 1779 (CϭO). 1H-NMR (CDCl3) d: 1.61 (4H), 1.75
(4H), 1.99 (2H), 2.11 (2H) [m each, (CH2)6], 4.71 (2H, m, two CH’s). 13C-
NMR (CDCl3) d: 22.8, 25.6, 28.3 (CH2), 82.3 (CH), 153.9 (CϭO). Anal.
Calcd for C10H14O4: C, 60.60; H, 7.12. Found: C, 60.53; H, 7.06.
Reaction of (؎)-erythro-3-Methyl-1-phenylbutane-1,2-diol (1f) with
Oxalyl Chloride in the Presence of 2,4,6-Collidine A solution of oxalyl
chloride (0.056 ml, 0.66 mmol) in THF (2.5 ml) was added to an ice-cooled
solution of 1f (108 mg, 0.599 mmol) and 2,4,6-collidine (0.40 ml, 3 mmol) in
THF (12 ml) over a period of 5 min. The mixture was stirred at 0 °C for
10 min and concentrated in vacuo. The residue was partitioned between ethyl
acetate (30 ml) and water (10 ml). The organic layer was washed succes-
sively with 5% aqueous citric acid, water, saturated aqueous sodium bicar-
bonate, and water (10 ml each), dried over magnesium sulfate, and concen-
trated to afford a colorless solid (135 mg), mp 89.5—95 °C. The molar ratio
of 2f and 3f in this product was 94 : 6. Recrystallization of this material from
carbon tetrachloride afforded (Ϯ)-cis-5-isopropyl-6-phenyl-1,4-dioxane-2,3-
dione (2f) (107 mg, 76%) as colorless prisms, mp 93—94.5 °C. The mother
269 (MϩϪOH). UV lMmeaCx N nm (e): 247 (sh) (330), 252 (380), 258 (440), 263
Nujol
max
(360). IR n
cmϪ1: 3338 (OH), 1759, 1715 (COCO2H). 1H-NMR
[(CD3)2SO] d: 3.4, 5.8 (1H each, br, OH and CO2H), 4.94 (1H, d, Jϭ6.8
Hz, CHOH), 5.87 (1H, d, Jϭ6.8 Hz, CHOCOCO2H), 7.12—7.26 (10H,
m, two Ph’s). 13C-NMR [(CD3)2SO] d: 74.6 (CHOH), 81.4 (CHCHOH),
127.2, 127.4, 127.5, 127.7, 127.9, 128.1, 136.4, 140.7 (Ph), 158.5, 159.3
(COCO2H). Anal. Calcd for C16H14O5: C, 67.13; H, 4.93. Found: C, 67.04;
H, 4.88.

March 2002
353
2) a) Stäglich P., Thimm K., Voß J., Justus Liebigs Ann. Chem., 1974,
671—689; b) Katon J. E., Lin S.-L., J. Mol. Struct., 48, 335—342
(1978); c) Jones G. P., Cornell B. A., Horn E., Tiekink E. R. T., J.
Crystallogr. Spectrosc. Res., 19, 715—723 (1989).
3) Schmidt S. P., Schuster G. B., J. Org. Chem., 43, 1823—1824 (1978).
4) a) Wimmer T., Ger. Offen. 2,263,814 [Chem. Abstr., 79, 79182y
(1973)]; b) Marhold A., Klauke E., Ger. Offen. DE 3,329,126 [Chem.
Abstr., 103, 22602v (1985)]; c) Nakayama S., Ikeda F., Eur. Pat. Appl.
EP 261,977 [Chem. Abstr., 109, 54784w (1988)].
(؎)-trans-2-Hydroxycyclohexyl Hydrogen Ethanedioate (11p) Water
(24 ml) was added to a solution of 2p (285 mg, 1.67 mmol) in acetonitrile
(6 ml), and the resulting solution was allowed to stand at room temperature
for 15 h. The solution was brought to pH 2 by addition of 10% hydrochloric
acid, saturated with sodium chloride, and extracted with ether (8ϫ15 ml).
The organic layers were combined, dried over magnesium sulfate, and con-
centrated in vacuo, leaving crude 11p (270 mg), mp 130—137 °C. Recrys-
tallization of this product from hexane–ethyl acetate (5 : 1, v/v) afforded 11p
(210 mg, 67%), mp 137.5—139 °C. Further recrystallization from hexane–
5) a) Verter H. S., Potter H., Dominic R., Chem. Commun., 1968, 973—
974; b) Forgione P. S., U. S. 3,431,278 [Chem. Abstr., 70, 96396c
(1969)]; c) Steglich W., Schmidt H., Hollitzer O., Synthesis, 1978,
622—624.
6) a) Carothers W. H., Arvin J. A., Dorough G. L., J. Am. Chem. Soc., 52,
3292—3300 (1930) and earlier references cited therein; b) Davies A.
G., Hua-De P., Hawari J. A.-A., J. Organomet. Chem., 256, 251—260
(1983).
ethyl acetate (2 : 1, v/v) afforded an analytical sample of 11p as colorless
MeCN
prisms, mp 140.5—141.5 °C. CI-MS m/z: 189 (Mϩϩ1). UV l
nm (e):
max
Nujol
max
1
258 (sh) (44). IR n
cmϪ1: 3455 (OH), 1761, 1732 (COCO2H). H-NMR
[(CD3)2SO] d: 1.16—1.41 (4H), 1.61 (2H), 1.86 (2H) [m each, (CH2)4], 3.44
(1H, m, CHOH), 4.60 (1H, ddd, Jϭ4.4, 8.8, 10.3 Hz, CHOCOCO2H). 13C-
NMR [(CD3)2SO] d: 23.1, 23.3, 29.2, 33.1 (CH2), 70.1 (CHOH), 79.2
(CHOCOCO2H), 158.7, 159.5 (COCO2H). Anal. Calcd for C8H12O5: C,
51.06; H, 6.43. Found: C, 50.93; H, 6.61.
7) Ripley L. G., Watson R. W., Can. J. Chem., 29, 970—973 (1951).
8) White R. C., Tetrahedron Lett., 21, 1021—1022 (1980).
9) Lloyd W. D., Navarette B. J., Shaw M. F., Org. Prep. Proced. Int., 7,
207—210 (1975).
10) a) Hauck F. P., U. S. 3,943,149 [Chem. Abstr., 85, 62847v (1976)]; b)
Idem, U. S. 4,082,773 [Chem. Abstr., 89, 108823x (1978)]; c) Rose-
man S., Link K. P., Carbohydr. Res., 69, 301—304 (1979).
11) Ellingboe E. K., Melby L. R., U. S. 2,816,287 [Chem. Abstr., 52,
P12899g (1958)].
12) Jäger V., Schohe R., Paulus E. F., Tetrahedron Lett., 24, 5501—5504
(1983).
13) a) Itaya T., Iida T., Eguchi H., Chem. Pharm. Bull., 41, 408—410
(1993); b) Iida T., Itaya T., Tetrahedron, 49, 10511—10530 (1993).
14) Itaya T., Iida T., Natsutani I., Ohba M., Chem. Pharm. Bull., 50, 83—
86 (2002).
Kinetic Procedure The solvent for the hydrolysis of 2 was prepared by
diluting 0.1 M acetate buffer (pH 5.00 at 25 °C, 14 ml) with acetonitrile to
make the whole volume 20 ml, showing pH 5.69 at 25 °C. The reactions of 2
were followed for 1.48—1.56ϫ10Ϫ3 M solutions by means of UV spec-
trophotometric analysis through at least 85% completion of the reaction with
at least nine measurements and were found to obey pseudo-first-order kinet-
ics. Each reaction of 2a—d, g—i, l—n, p, r was carried out in duplicate, and
the averaged rate constants are collected in Table 5. The pH values of the re-
action mixtures after the reactions decreased by not more than 0.03. The hy-
drolysis of 2h is described below as a typical example.
Compound 2h (4.07 mg) was dissolved as quickly as possible in the sol-
vent described above in an ultrasonic bath maintained at 25 °C to make the
whole volume 10 ml. The solution (1.52ϫ10Ϫ3 M) was transferred to a cu-
vette, which was placed in a cell compartment maintained at 25 °C, without
delay. Absorbances (At) of the mixture at selected times were determined at
262 nm. The absorbance (A∞) on completion of the reaction was reached in
ca. 40 min (12 half-lives). A plot of ln(AtϪA∞) against time gave a straight
line (rϭ0.999 for 16 determinations) and k1ϭ2.1ϫ10Ϫ1 minϪ1 was estimated
by linear regression analysis.21) Compound 11h (4 mg) was obtained from
the reaction mixture.
15) Itaya T., Iida T., J. Chem. Soc., Perkin Trans. 1, 1994, 1671—1672.
16) This method was also unsuccessful in the preparation of 2l from cis-
1,2-cyclopentanediol (1l).
17) When 2,4,6-collidine was employed, the estimated yields of 2o and the
oxalate polymers were 29% and 71%, respectively.
18) This structure was supported by X-ray crystallographic analysis: Iida
T., Ph. D. Dissertation, Kanazawa University, January 1994.
19) Buist G. J., Bunton C. A., Miles J. H., J. Chem. Soc., 1957, 4567—
4574.
20) Hayashi M., Terashima S., Koga K., Tetrahedron, 37, 2797—2803
(1981).
21) For example, see Bunnett J. F., “Investigation of Rates and Mech-
anisms of Reactions,” 3rd ed., Part I, ed. by Lewis E. S., John Wiley
and Sons, New York, 1974, pp. 145—153.
Compounds 2a, b, d, g, j, k, l—n, p, r having no phenyl group have ab-
sorption maxima at 272—273 nm in acetonitrile. Hydrolyses of these com-
pounds were monitored by measuring the decrease in absorbance at 275 nm.
The UV spectrum of every reaction mixture at completion was coincident
with that of 11p.
The slow reaction of diethyl oxalate (12) was monitored with a solution
(1.9ϫ10Ϫ3 M) by measuring the increase in absorbance at 250 nm. It was
shown by thin-layer chromatography that the product was ethyl hydrogen
ethanedioate and that oxalic acid was absent in the reaction mixture.22)
22) For neutral hydrolysis of diethyl oxalate (12) in aqueous organic sol-
vents, see Geheb D., Kasanskaja N. F., Beresin I. W., Ber. Bunsenges.
Phys. Chem., 76, 160—169 (1972).
References and Notes
1) Jones G. I. L., Owen N. L., J. Mol. Struct., 18, 1—32 (1973).