Asymmetric oxidation of 3-alkyl-1,2-cyclopentanediones. Part 2: Oxidative ring cleavage of 3-alkyl-1,2-cyclopentanediones: Synthesis of 2-alkyl-γ-lactone acids

Article abstract of DOI:10.1016/S0957-4166(03)00309-4

Ti(OiPr)₄/diethyl tartrate/tBuOOH system oxidizes 3-alkyl-1,2-cyclopentanediones resulting in hydroxylated ring cleavage products 2-alkyl-γ-lactone acids, in high enantioselectivity (～95% ee) and satisfactory isolated yields (up to 55%).

Full text of DOI:10.1016/S0957-4166(03)00309-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 1565–1573
Asymmetric oxidation of 3-alkyl-1,2-cyclopentanediones. Part 2:
Oxidative ring cleavage of 3-alkyl-1,2-cyclopentanediones:
synthesis of 2-alkyl-g-lactone acids
Anne Paju,a To˜nis Kanger,a To˜nis Pehk,b Rasmus Lindmaa,a Aleksander-Mati Mu¨u¨riseppa and
Margus Loppa,*
aDepartment of Chemistry, Tallinn Technical University, Ehitajate tee 5, Tallinn 19086, Estonia
bNational Institute of Chemical Physics and Biophysics, Akadeemia tee 23, Tallinn 12618, Estonia
Received 19 February 2003; accepted 3 April 2003
Abstract—Ti(OiPr)4/diethyl tartrate/tBuOOH system oxidizes 3-alkyl-1,2-cyclopentanediones resulting in hydroxylated ring
cleavage products 2-alkyl-g-lactone acids, in high enantioselectivity (ꢀ95% ee) and satisfactory isolated yields (up to 55%).
© 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
hydroxylation/ring cleavage, and on the synthesis of
2-alkyl-g-lactonic acids via that reaction.
The unique Ti(OiPr)4/tartaric ester/tBuOOH complex
introduced by Sharpless for the asymmetric epoxidation
of allylic alcohols1 has found use in the other asymmet-
ric oxidation processes such as oxidation of sulfides,2,3
Baeyer–Villiger reaction with cyclobutanones4 and 2-
hydroxylation of 3-hydroxyketones.5,6 We have recently
found that the Sharpless system oxidizes readily 3-
alkyl-1,2-cyclopentanediones, resulting in enantiomeric
3-hydroxy- and ring cleavage oxidation products in
excellent stereoselectivity.7
2. Results and discussion
2.1. Oxidation of 3-methyl-1,2-cyclopentanedione
Our previous experiments revealed that the substrate
1a, when subjected to the oxidation, using the
Ti(OiPr)4-tartaric ester-tert-BuOOH (TBHP) system,
resulted in up to 40% 3-hydroxylation (mono-oxidation
products), and up to 31% 3-hydroxylated ring cleavage
products (double-oxidation products−hydroxylation+
ring cleavage).8 In order to shift the reaction towards
3-hydroxylation/ring cleavage we investigated the oxi-
dation of 3-methyl-1,2-cyclopentanedione 1a as a model
compound in different conditions. The main ring cleav-
age product identified was lactone 2a with ester 3a
being found as a minor product; Scheme 1. The results
obtained are presented in Table 1.
We have already demonstrated that the 3-hydroxylation
of 1,2-cyclopentanediones can be used for a preparative
synthesis of 3-hydroxy-1,2-cyclopentanediones.8 The
non-asymmetric cleavage of cyclic 3-alkyl-1,2-diones
resulting in diacids or keto acids is known earlier.9–12
We are not aware of any literature references available
connected with the asymmetric 3-hydroxylation/ring
cleavage of 3-alkyl-cyclopentane-1,2-diones except for
our preliminary report:7 the products of that ring cleav-
age provide non-racemic lactones and esters which are
of importance because of their interesting biological
properties.13,14 Also, the method equips us with several
chirons which can be used in asymmetric synthesis.15–17
Herein, we focus on the possibility of directing the
oxidation of 3-alkyl-1,2-cyclopentanediones towards 3-
The data from Table 1 show that the amount of double
oxidized products 2a and 3a varies from 31 to 50%. In
each case, a certain amount of the non-cleaved 3-
hydroxylation product (mono-oxidized product) was
also isolated. The reaction conditions best suited for
3-hydroxylation (40% of the corresponding product 4a)
gave 31% of the ring cleaved products as a sum of 2a
and 3a; Table 1, no 1.8 The ee of the isolated com-
pounds 2a, 3a and 4a was ꢀ95% in the best case. In
order to increase the amount of ring cleavage products,
a larger quantity of Ti(OiPr)4 and oxidant was used.
* Corresponding author. E-mail: lopp@chemnet.ee
0957-4166/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0957-4166(03)00309-4

1566
A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
Scheme 1. 3-Hydroxylation/ring cleavage of 3-alkyl-1,2-cyclopentanediones.
Table 1. Products of the asymmetric oxidation of 3-methyl-1,2-cyclopentanedione 1a*
No Ti/DET/TBHP
2a yield %
3a yield %
Sum isolated ring cleavage
3-Hydroxylation Sum of isolated oxidation
time (h)
products 2a+3a
product 4a
products 2a+3a+4a
18
2
1/1.6/1.5
42
2/2.4/2.5
42
3/3.6/4.0
42
1/1.6/2.5
68
1/3/2.5
68
1/1.6/4
68
1/1.6/2.5
10 days
23
48
42
44
23
31
43
8
31
50
47
50
48
38
49
40
19
23
15
15
13
15
71
69
70
65
63
51
64
2
3
5
4
6
5
25
7
6
7
6
* Conditions: reaction temperature −20°C; solvent dichloromethane (see Section 3).
Thus 2- and 3-fold excess of the Ti(OiPr)4 increased the
amount of double oxidized (hydroxylated+ring cleaved)
products (as a sum of 2a and 3a) (Table 1, nos. 2 and
3). The same increase was achieved also with a stoichio-
metric Ti(OiPr)4 amount (tartaric ester 1.6 equiv.) by
increasing only the oxidant ratio in the catalyst (TBHP
2.5 fold excess in respect to Ti(OiPr)4; Table 1, no 4).
Additional increase of the Ti(OiPr)4/TBHP ratio in
favour of TBHP reduced the amount of the ring cleav-
age product (Table 1, no 6). The maximum amount of
ring cleavage (isolated yield 50%, as a sum of 2a and
3a) was obtained at the ratio 1/1.6/2.5 of the reagents.
(2a+3a+4a) reduced. The prolonged reaction time did
not reduce the relative quantity of the 3-hydroxylation
product 4a among the reaction products either (Table
1, no 7). The observed quite stable ratio of the 3-
hydroxylation product towards ring cleaved products
(4a versus 2a+3a) in all experiments implies that part of
the intermediate 3-hydroxylated product does not par-
ticipate in the following oxidation and does not enter
the subsequent oxidation step. It might be possible that
only the primary 3-hydroxylation product in the form
of acetal 5a (after the first oxidation step) is capable of
further oxidation: The corresponding enolic form 4a
remains un-oxidized (or vice versa). The slow transfor-
mation of the initial isopropyl acetal 5a in the chloro-
form solution to 4a has been observed by us already
previously.7
We were concerned about the considerable quantity of
3-hydroxylation product 4a that remained un-cleaved in
these experiments (13–23%; Table 1, nos 2–7). If you
assume that the oxidation proceeds via the 3-hydroxyla-
tion process, two reasons can be considered for the
incomplete oxidation: the quantity of the oxidant is not
sufficient or, the reaction time is not sufficient. In
checking these assumptions, however, we found that a
considerable increase of the oxidant (up to Ti(OiPr)4/
TBHP ratio 1:4) had no substantial influence on the
quantity of the hydroxylated product 4a (15% versus
13%; Table 1, no 4 and no 6, respectively). On the
contrary, the yield of the ring cleavage as well as the
total amount of the asymmetric oxidation products
To verify this assumption, 4a and 5a were oxidized
separately under the Sharpless conditions. The oxida-
tion of 4a did not lead to the expected ring cleavage
product lactone acid 2a (different products formed),
while the acetal 5a resulted in the ring cleavage product
2a (the isolated yield, however, was quite moderate).
Also, when 4a and 5a were oxidized with achiral oxi-
dants such as Jones’ reagent, peracetic acid or
Ti(OiPr)4/TBHP, different types of oxidation products
were obtained. Thus, from 5a the lactone acid 2a was

A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
1567
easily obtained with peracetic acid. Compound 4a
2.2. Oxidation of 3-ethyl-, 3-benzyloxyethyl- and 3-
hydroxymethyl-1,2-cyclopentanediones
remained unchanged with peracetic acid, however, with
Ti(OiPr)4/TBHP, hydroxy ketoacid 6 was isolated in low
yield. The latter finding indicates that during the oxida-
tion a-hydroxylation at non-branched carbon also occurs
to a certain extent. This pathway, leading to a chiral
hydroxylated compound 6, is, however, marginal
(Scheme 2).
The optimum reaction conditions found for double
oxidation were used in the case of the substrates 1b–d.
Thus, the substrates 1b–d were subjected to the oxidation
at the catalyst ratio Ti(OiPr)4/DET/TBHP 1/1.6/2.5 and
Ti(OiPr)4/substrate ratio 1:1 (in some cases 2:1, depend-
ing on the presence of alcohol functionalities in the
substrate). The results obtained are presented in Table
2.
The results obtained support the proposed assumption
that 5a (or a similar intermediate) is the substrate for the
second oxidation step resulting in the ring cleavage
products while the enol 4a is further oxidized in a smaller
amount. The oxidative ring cleavage product also con-
tains a certain amount of monoester 3a coming from the
esterification of 2a with the hydroxyl moiety of tartaric
acid diethyl ester component of the catalyst system. The
extent of formation of the esters depends on the relative
amount of the diethyl tartrate in respect to the substrate
(Table 1, nos 4 and 5; 3a: 6% versus 25%). However, the
amount of the tartaric acid ester did not influence greatly
the overall yield of the ring cleavage product (Table 1,
no 5: 2a+3a 48% versus no 4: 2a+3a 50%).
The oxidation of the substrates 1b–d resulted in a product
profile similar to that of 1a. Also, the yields of the
reaction products were quite similar. The total yield of
the ring cleavage products (presented as a sum of
products 2 and 3) was about 50% (Table 2). Only in the
case of hydroxymethyl substrate 1d was the yield of
oxidation lower when one equivalent of the Ti-complex
towards substrate was used (Table 2, no 6 versus no 4:
55% versus 42%). It is noteworthy that the enantioselec-
tivity of the oxidation was also higher when two equiv-
alents of the catalyst were used (Table 2, no 6 versus no
Scheme 2. Ring cleavage of acetal 5a and enol 4a.
Table 2. Products of the asymmetric oxidation of 3-alkyl-1,2-cyclopentanediones 1a–d
No Substrate
Ti/DET/TBHP
time (h)
Sum, ring cleavage products 2
and 3 (%)
2
3
4 yield %
7 yield %
Yield (%)
Ee%
>95a
Yield (%)
1
2
3
4
5
6
1a
1b
1c
1d
1d
1d
1/1.6/2.5
68
1/1.6/2.5
68
1/1.6/2.5
68
1/1.6/2.5
68
2/3/1.5
42
2/3/2.5
68
50
51
53b
42
31
55
44
44
28
42
31
55
6
15
6
>95a
98c
7
15
6
16
18
9
69d
97d
94d
N.d.e
N.d.e
N.d.e
N.d.
N.d.
N.d.
Traces
Traces
Traces
a Determined by NMR from the (−)-menthol esters 9, the other diastereomer was not observed (Scheme 4).
b Dimeric ester 8c (Fig. 2) in 9% yield was also separated, and that amount added to the sum of ring cleavage products.
c Determined by HPLC as spirodilactone 11 using a chiral column (Daicel Chiralcel ODH; Scheme 5).
d Determined by HPLC as p-bromophenacyl esters of 2d using a chiral column (Daicel Chiralcel ODH).
e Not detected.

1568
A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
Figure 1. Simplified model of the double co-ordinated complex (see also Part 1.8)
4: ee 69% versus 94%). It means that both chiral
complex molecules that are bound to the two hydroxyl
groups of the substrate are favouring the same oxida-
tion face in the first hydroxylation step (i.e. they are a
‘matched’ pair) (Fig. 1). When a smaller amount of the
oxidant was used the isolated yield of ring-cleavage
product was lower, however, the enantioselectivity was
slightly higher (Table 2, no 5).
In the case of the substrate 1c, together with other
esterified ring cleaved oxidation product 3c, the forma-
tion of triacid monoester 8c was observed (Fig. 2).
Scheme 3. By-products of the oxidative cleavage of 3-alkyl-
1,2-cyclopentanediones.
2.3. Determination of enantiomeric purity and the abso-
lute configuration of the products
The enantiomeric purity of the compounds was deter-
mined by use of different techniques.
Thus, lactones 2a and 2b were converted to (1R,2S,5R)-
(−)-menthol esters 9 and the ee was determined by
NMR from the diastereomeric ratio of the correspond-
ing esters (Scheme 4).
Figure 2.
The formation of the quite unusual dimeric ester 8c,
and also the finding that in most cases monoesters 3
(easily lactonize on standing) are formed during the
oxidation (depending on the ratio of the tartaric acid in
the reaction media, Table 1, no 5) refer to the catalytic
ability of the Sharpless complex in the esterification
reaction (analogous to that described in18 for a TAD-
DOL complex).
In all cases, the decarbonylated products were detected
among the reaction products. Thus, the achiral 4-oxo-
carboxylic acids 7 were isolated in 6–9% yield (Scheme
3). The highest amount (9%) of the decarbonylated
product was observed in the case of the substrate 1c.
Another decarbonylated product 6 was also detected
among the reaction products in the case of substrate 1a
(discussed in Section 2.1).
Scheme 4. Diastereomeric (−)-menthol esters from ring
cleaved products.
1
In H NMR spectra from these diastereomeric esters,
different chemical shifts are clearly seen from menthol

A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
1569
carbinol proton and from some of its methyl groups.
Differences in the shieldings of the remaining protons
were obtained using the 2D FT correlation diagrams.
In 13C NMR spectra from racemic 2 nearly all lines
from the corresponding atoms of the diastereoisomers
9a and 9b are resolved. These differences are only in the
range of tens of ppb, as a rule, however, in few cases
the differences exceed 100 ppb. There are no known
rules to use these differences for the assignment of the
absolute configuration of the menthol esters. It is
remarkable that in the present case for the esters of 2a
and 2b the 13C chemical shieldings are about 0.5 ppm
less than in their corresponding diastereoisomers. The
high field 13C chemical shifts generally indicate a higher
role of the nonbonded interactions that is characteristic
to the thermodynamically less stable isomer. The
molecular mechanics (MM, Merck and Sybyl force
fields) and semiempirical (AM1) calculations (Spartan
5.0, Wavefunction, Inc.) of diastereoisomeric (−)-men-
thol esters of 2a and 2b reveal that S-2a and S-2b are
less stable and, thus, 13C chemical shifts support the
proposed S configuration of the lactones.
The obtained compound 2a has the same absolute
configuration that the compound obtained from 3-
hydroxylation product 4a by non-asymmetric oxidation
(see Part 18). It proves the assumption that the 3-
hydroxylation reaction (generation of an asymmetry)
occurs before the ring cleavage reaction.
The absolute configurations of other ring cleaved prod-
ucts were assigned by analogy. The enantioselectivity
(presented as ee) of the obtained ring cleavage oxida-
tion reaction was high (ee ꢀ95%, comparable with that
of the corresponding 3-hydroxylated products). This
means that the ring cleavage reaction may proceed via
3-hydroxylation without any loss in enantiomeric excess
in the second oxidation step.
3. Experimental
1H and 13C NMR spectra were determined in deuter-
ated solvents on a Bruker AMX-500 spectrometer. Full
1
assignment of all H and 13C chemical shifts are based
on 2D FT NMR and selective INEPT experiments. The
solvent peaks of CHCl3 (l 7.27 ppm for 1H) and CDCl3
(l 77.0 ppm for 13C) were used as internal references.
Mass spectra were measured on a Hitachi M80B spec-
trometer using the EI (70 eV), CI (isobutane) and
HRMS mode. Optical rotations were obtained using a
A. Kru¨ss Optronic GmbH polarimeter P 3002. TLC
was performed using DC-Alufolien Kieselgel 60 F254
(Merck) or Silufol® UV 254 silica gel plates. Merck
silica gel 60 (0.063–0.200 mm) or Chemapol silica gel L
40/100 was used for column chromatography. All the
reactions sensitive to oxygen or moisture were con-
ducted under argon atmosphere in oven-dried glass-
ware. Commercial reagents were generally used as
received. CH2Cl2 was distilled from CaH2 and stored
The enantiomeric purity of the compound 2c was deter-
mined by chiral HPLC after converting it to the
spirodilactone 11 (Scheme 5).
The lactone acid 2c was deprotected using H2 and
Pd/C. The resulting hydroxyethyl carboxylic acid 10
was lactonized with p-TsOH to afford spirodilactone
11. The enantiomers of the spirodilactone are separable
on a HPLC chiral column that enables us to estimate
the ee of the initial 2c (ee of 11 was found 98%). The ee
of the compound 2d was determined directly by HPLC
after it was converted into p-bromophenacyl ester by
p-bromophenacyl bromide which provides excellent res-
olution of enantiomers on a Chiralcel ODH column
(Daicel). The enantiomeric purity of the obtained ring
cleavage products was high in all cases. Only in the case
of 1d, does the oxidation enantioselectivity depends on
the Ti(OiPr)4/substrate ratio (Table 2, nos 4–6).
,
over 3 A molecular sieve pellets. THF and ether were
distilled from LiAlH4 before use, DMF and Et3N from
CaH2.
3.1. Substrates
The absolute configuration of the compounds was
determined from the ring cleavage product 2a that has
(−) optical rotation, which is identical with the sign of
the known (S)-(−)-2-hydroxypentane-1,5-dienoic acid
2,5-lactone.15 It means that the absolute configuration
of 2a is S and according to the specific rotation its ee is
ꢀ95%. The value of the specific rotation of the 2a was
[h]D25 −15.5; c 2.35, water ([h]D23 −16.2; c 1.86, water15).
The synthesis (or origin) of the substrates 1a–c is
described in the Part 1.8 Compound 1d was prepared
from 2-cyclopentene-1-methanol19 by protecting the
hydroxy group as a tert-butyldimethylsilyl ether, dihy-
droxylating the double bond with KMnO4, oxidising
the vicinal diols according to Amon et al.,20 followed by
deprotection.
Scheme 5. Conversion of lactone acid 2c to spirodilactone 11.

1570
A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
3.2. Oxidative ring cleavage of 3-alkyl-1,2-
cyclopentanediones
3.1.1. 2-Hydroxy-3-(hydroxymethyl)-cyclopent-2-en-1-
one, 1d. A solution of 2-cyclopentene-1-methanol (373
mg, 2.8 mmol), TBDMSCl (630 mg, 4.2 mmol) and
imidazole (378 mg, 5.6 mmol) in dry DMF (9 mL)
was stirred at rt for 22 h. Ether (100 mL) was then
added, the mixture was washed with water, with 5%
NaHCO3 solution, water, brine, dried (MgSO4) and
ether evaporated. The residue was purified by flash
chromatography (silica gel, hexanes/acetone 100:1) to
give of silyl-protected 2-cyclopentene-1-methanol (574
mg, 96%).
3.2.1. Typical procedure for synthesis of lactone-acids
by asymmetric oxidation of 3-alkyl-1,2-cyclopentane-
diones. To a solution of Ti(OiPr)4 (0.3 mL, 1 mmol)
,
and 4 A powdered molecular sieves (100 mg) in
CH2Cl2 (6 mL) (+)-DET (0.27mL, 1.6 mmol) was
added and the mixture was stirred for 15 min at
−20°C. After addition of cyclopentanedione (1 mmol)
in CH2Cl2 (2 mL) the mixture was stirred for 30 min.
Then TBHP (0.4 mL, 2.5 mmol, 6.25 M solution in
decane) was added and the mixture was kept at
−20°C for 68 h. The reaction was quenched by stir-
ring with a solution of citric acid (192 mg, 1 mmol in
a mixture of 10% MeOH in CH2Cl2 or 10% acetone
in ether) at rt for 1 h. The reaction mixture was
filtered through a path of Celite and purified by
column chromatography on silica gel (Chemapol silica
gel L40/100).
To a stirred solution of protected 2-cyclopentene-1-
methanol (574 mg, 2.7 mmol) in t-BuOH (16 mL)
and water (11 mL), cooled in an ice bath, was added
over a 15 min period a cooled solution of KMnO4
(469 mg, 3 mmol) and NaOH (162 mg, 4 mmol) in
water (18 mL). After completion of the addition, the
solution was stirred at 0°C for 20 min, and then
Na2SO3 (136 mg, 1.1 mmol) in water (3 mL) was
added. The reaction mixture was filtered and the
filtrate was extracted three times with EtOAc. The
combined extracts were washed with water, brine,
dried (MgSO4) and the solvents evaporated. Flash
chromatography (silica gel, hexanes/acetone 10:1.5)
yielded 3-(tert-butyldimethylsilyloxymethyl)-1,2-cyclo-
pentanediols (420 mg, 63%).
3.2.2. (S)-2-Methyl-5-oxotetrahydrofuran-2-carboxylic
acid, 2a. Diketone 1a was oxidized according to the
typical procedure and purified by column chromatog-
raphy (petroleum ether/acetone 10:3 to 10:6) to afford
compounds 2a together with minor amounts of 3a
and 7a. Compound 2a as a white solid (63 mg, 44%);
ee 95%, [h]2D0=−15 (c 1.66, CH2Cl2). 1H NMR (500
MHz CDCl3): l 1.66 (s, 3H, 2-Me), 2.20 (dt, J=13.3,
9.8, 9.8 Hz, 1H, H-3), 2.60 (ddd, J=13.3, 9.8, 3.6
Hz, 1H, H-3), 2.62 (ddd, J=18.0, 9.8, 3.6 Hz, 1H,
H-4), 2.70 (dt, J=18.0, 9.8, 9.8 Hz, 1H, H-4), 10.37
(bs, 1H, COOH); 13C NMR (125 MHz CDCl3): l
23.44 (2-Me), 28.32 (C-4), 32.90 (C-3), 83.55 (C-2),
176.31 (2-COOH), 176.71 (C-5); EI (m/z, %): 145
(M+H+, 0.34), 127 (0.27), 100 (5.6), 99 (M−COOH+,
90.1), 71 (12.7), 43 (100); CI (m/z, %): 145 (M+H+,
90); HRMS calcd. for (M−OH)+ C6H7O3: 127.0394;
found: 127.0397.
To a solution of DMSO (0.71 mL, 10.1 mmol) in
CH2Cl2 (43 mL) TFAA (1.28 mL, 9.1 mmol) was
added dropwise at −60°C. The mixture was stirred for
10 min, followed by addition of the above diol (777
mg, 3.16 mmol) in CH2Cl2 (4 mL). After stirring at
−60°C for 1.5 h, Et3N (2.9 mL, 21 mmol) was added
at −60°C and the mixture was stirred for 1.5 h at
that temperature. The reaction mixture was allowed
to warm up to ca. 5°C, poured into a cold 1N HCl
solution (100 mL) and extracted twice with CH2Cl2.
The extract was washed with water, brine, dried
(MgSO4) and concentrated. The residue was chro-
matographed (silica gel, hexanes/EtOAc 10:1) to give
371 mg (48.5%) of 3-(tert-butyldimethylsilyloxy-
methyl)-1,2-cyclopentanedione.
(S)-2-Hydroxy-2-methylpentanedioic acid (+)-diethyl-
tartrate ester 3a as colourless oil (21 mg, 6%). 1H
NMR (500 MHz CDCl3): l 1.29 (t, J=7.2 Hz, 3H,
4%-Et Me), 1.31 (t, J=7.2 Hz, 3H, 1%-Et Me), 2.04
(ddd, J=14.0, 10.1, 6.1 Hz, 1H, H-3), 2.12 (ddd, J=
14.0, 10.0, 5.6 Hz, 1H, H-3), 2.29 (ddd, J=16.4, 10.1,
5.6 Hz, H-4), 2.53 (ddd, J=16.4, 10.0, 6.1 Hz, H-4),
4.26 (m, 2H, 4%-Et CH2), 4.28 (m, 2H, 1%-Et CH2),
4.80 (d, J=2.1 Hz, 1H, H-3%), 5.51 (d, J=2.1 Hz,
1H, H-2%); 13C NMR (125 MHz CDCl3): l 13.98 (1%-
Et Me), 14.03 (4%-Et Me), 26.03 (2-Me), 28.56 (C-4),
33.93 (C-3), 62.46 (1%-Et CH2), 62.85 (4%-Et CH2),
70.53 (C-3%), 73.84 (C-2%), 74.29 (C-2), 166.10 (C-1%),
170.69 (C-4%), 174.88 (C-1), 178.24 (C-5).
To a solution of the above diketone (371 mg, 1.53
mmol) in THF (15 mL) 1N HCl solution (6 mL) was
added. After stirring at rt for 1.5 h the mixture was
diluted with water (12 mL) and THF was removed at
25°C under vacuum. The water phase was extracted
12 times with dry EtOAc, the combined extracts were
dried (Na2SO4) and concentrated. The residue was
purified by flash chromatography (silica gel, hexanes/
acetone 10:5), yielding 174 mg (89%) of 2-hydroxy-3-
(hydroxymethyl)cyclopent-2-en-1-one 1d as a white
solid; 1H NMR (500 MHz CDCl3+CD3OD): l 2.21
(m, 2H, H-5), 2.31 (m, 2H, H-4), 4.30 (s, 2H, 3%-
CH2O); 13C NMR (125 MHz CDCl3+CD3OD): l
22.25 (C-4), 31.77 (C-5), 58.58 (C-3%), 145.47 (C-3),
148.65 (C-2), 204.06 (C-1); EI (m/z, %): 128 (M+,
12.3), 111 (3.2), 110 (3.2), 100 (8.4), 99 (100); HRMS
calcd. for C6H8O3: 128.0473; found: 128.0463.
4-Oxopentanoic acid 7a (7 mg, 6%). 1H NMR (500
MHz CDCl3): l 2.19 (s, 3H, Me), 2.62 (t, J=6.4 Hz,
2H, H-2), 2.75 (t, J=6.4 Hz, 2H, H-3); 13C NMR
(125 MHz CDCl3): l 27.78 (C-2), 29.75 (Me), 37.63
(C-3), 178.60 (C-1), 206.77 (C-4).
3.2.3.
(S)-2-Ethyl-5-oxotetrahydrofuran-2-carboxylic
acid, 2b. Diketone 1b was oxidized according to the

A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
1571
typical procedure and purified by column chromatogra-
phy (CH2Cl2/MeOH 10:0.2 to 10:1) to afford 2b
together with minor amounts of 3b and 7b. Compound
2b as a white solid (70 mg, 44%); ee 95%, [h]2D2=−23 (c
1H NMR (500 MHz CDCl3): l 1.28 (t, J=7.2 Hz, 3H,
4%-Et Me), 1.29 (t, J=7.2 Hz, 3H, 1%-Et Me), 2.06 (m,
2H, H-3), 2.13 (m, 2H, 2-CH2), 2.23 (m, 1H, H-4), 2.56
(ddd, J=16.4, 10.0, 6.1 Hz, 1H, H-4), 3.72 (dt, J=9.5,
2×4.7 Hz, 1H, 2-CH2O), 3.80 (td, J=2×9.5, 4.7 Hz, 1H,
2-CH2O), 4.18 (dq, J=10.7, 3×7.2 Hz, 1H, 4%-Et CH2),
4.24 (q, J=7.2 Hz, 2H, 1%-Et CH2), 4.28 (dq, J=10.7,
3×7.2 Hz, 1H, 4%-Et CH2), 4.44 and 4.51 (both d,
J=11.7 Hz, 2H, Bn CH2), 4.77 (d, J=2.2 Hz, 1H,
H-3%), 5.49 (d, J=2.2 Hz, 1H, H-2%), 7.26–7.35 (m, 5H,
Bn o, m, p); 13C NMR (125 MHz CDCl3): l 13.95
(1%-Et Me), 13.97 (4%-Et Me), 28.18 (C-4), 33.69 (C-3),
37.72 (2-CH2), 62.21 (1%-Et CH2), 62.79 (4%-Et CH2),
66.92 (2-OCH2), 70.51 (C-3%), 73.31 (Bn CH2), 73.55
(C-2%), 76.97 (C-2), 127.71 (p), 127.74 (o), 128.34 (m),
137.43 (s), 166.09 (C-1%), 170.67 (C-4%), 173.69 (C-1),
178.15 (C-5).
1
1.49, CH2Cl2). H NMR (500 MHz, CDCl3): l 1.02 (t,
J=7.5 Hz, 3H, 2-Me), 1.89 and 2.12 (both dq, J=14.3,
7.5 Hz, 2H, 2-CH2), 2.23 (dt, J=13.5, 9.8 Hz, 1H, H-3),
2.53 (ddd, J=13.5, 9.9, 3.9 Hz, 1H, H-3), 2.58–2.72 (m,
2H, H-4), 9.90 (bs, 1H, COOH); 13C NMR (125 MHz
CDCl3): l 8.03 (C-2¦), 28.08 (C-4), 30.33 (C-2%), 30.97
(C-3), 86.97 (C-2), 176.11 (C-1), 176.39 (C-5); EI (m/z,
%): 159 (M+H+, 0.33), 141 (0.41), 129 (1.8), 114 (7.9),
113 (M−COOH+, 100); CI (m/z, %): 159 (M+H+, 100);
HRMS calcd for (M−OH)+ C7H9O3: 141.0551; found:
141.0580.
(S)-2-Hydroxy-2-ethylpentanedioic acid (+)-diethyltar-
1
trate ester 3b as a colourless oil (26 mg, 7%). H NMR
Lactone from 3c—(R)-tetrahydro-2-benzyloxyethyl-5-
oxo-2-furancarboxylic acid (+)-diethyltartrate ester,
(500 MHz CDCl3): l 0.91 (t, J=7.4 Hz, 3H, 2-Et Me),
1.27 (t, J=7.2 Hz, 3H, 4%-Et Me), 1.29 (t, J=7.2 Hz,
3H, 1%-Et Me), 1.71 (dq, J=14.0, 3×7.4 Hz, 1H, 2-Et
CH2), 1.86 (dq, J=14.0, 3×7.4 Hz, 1H, 2-Et CH2), 2.03
(m, 1H, H-3), 2.04 (m, 1H, H-3), 2.24 (ddd, J=16.4,
10.1, 5.6 Hz, H-4), 2.51 (ddd, J=16.4, 9.9, 6.1 Hz,
H-4), 4.14 (dq, J=10.7, 3×7.2 Hz, 1H, 4%-Et CH2), 4.27
(q, J=7.2 Hz, 2H, 1%-Et CH2), 4.28 (dq, J=10.7, 3×7.2
Hz, 1H, 4%-Et CH2), 4.79 (d, J=2.2 Hz, 1H, H-3%), 5.50
(d, J=2.2 Hz, 1H, H-2%); 13C NMR (125 MHz CDCl3):
l 7.40 (2-Et Me), 13.95 (2C, 1%,4%-Et Me), 28.48 (C-4),
32.14 (2-Et CH2), 33.09 (C-3), 62.40 (1%-Et CH2), 62.81
(4%-Et CH2), 70.60 (C-3%), 73.87 (C-2%), 77.59 (C-2),
166.06 (C-1%), 170.63 (C-4%), 174.44 (C-1), 178.44 (C-5).
1
atom numbering as in 3c. H NMR (500 MHz CDCl3):
l 1.25 (t, J=7.2 Hz, 3H, 4%-Et Me), 1.29 (t, J=7.2 Hz,
3H, 1%-Et Me), 2.19 (m, 1H, 2-CH2), 2.38 (m, 1H, H3),
2.47 (m, 1H, 2-CH2), 2.52 (m, 1H, H-4), 2.53 (m, 1H,
H-3), 2.75 (m, 1H, H-4), 3.64 (t, J=6.4 Hz, 2H, 2-
CH2O), 4.18 (dq, J=10.7, 3×7.2 Hz, 1H, 4%-Et CH2),
4.24 (q, J=7.2 Hz, 2H, 1%-Et CH2), 4.28 (dq, J=10.7,
3×7.2 Hz, 1H, 4%-Et CH2), 4.45 and 4.47 (both d,
J=12.3 Hz, 2H, Bn CH2), 4.74 (d, J=2.3 Hz, 1H,
H-3%), 5.42 (d, J=2.3 Hz, 1H, H-2%), 7.26–7.35 (m, 5H,
Bn o, m, p); 13C NMR (125 MHz CDCl3): l 13.95
(1%-Et Me), 13.97 (4%-Et Me), 27.74 (C-4), 31.85 (C-3),
36.26 (2-CH2), 62.34 (1%-Et CH2), 62.72 (4%-Et CH2),
64.98 (2-OCH2), 70.22 (C-3%), 72.99 (Bn CH2), 73.86
(C-2%), 84.52 (C-2), 127.58 (o), 127.62 (p), 128.32 (m),
137.75 (s), 165.84 (C-1%), 170.17 (C-1), 170.20 (C-4%),
176.03 (C-5).
4-Oxohexanoic acid 7b (8 mg, 6%). 1H NMR (500 MHz
CDCl3): l 1.08 (t, J=7.3 Hz, 3H, H-6), 2.48 (q, J=7.3
Hz, 2H, H-5), 2.65 (t, J=6.4 Hz, 2H, H-2), 2.74 (t,
J=6.4 Hz, 2H, H-3); 13C NMR (125 MHz CDCl3): l
7.72 (C-6), 27.73 (C-2), 35.83 (C-5), 36.36 (C-3), 177.83
(C-1), 209.34 (C-4).
4-Oxo-6-benzyloxyhexanoic acid 7c (18 mg, 9%). 1H
NMR (500 MHz CDCl3): l 2.64 (t, J=6.4 Hz, 2H,
H-2), 2.75 (t, J=6.2 Hz, 2H, H-5), 2.78 (t, J=6.4 Hz,
2H, H-3), 3.76 (t, J=6.2 Hz, 2H, H-6), 4.52 (s, 2H, Bn
CH2), 7.27–7.37 (m, 5H, o, m, p); 13C NMR (125 MHz
CDCl3): l 27.57 (C-2), 37.39 (C-3), 42.80 (C-5), 65.11
(C-6), 73.22 (Bn CH2), 127.68 (p), 127.70 (o), 128.38
(m), 137.91 (s), 178.34 (C-1), 207.03 (C-4).
3.2.4. (R)-2-Benzyloxyethyl-5-oxotetrahydrofuran-2-car-
boxylic acid, 2c. Diketone 1c was oxidized according to
a typical procedure (using 195 mg, 0.84 mmol of the
substrate) and purified by column chromatography
(petroleum ether/acetone 10:1.5 to 10:6) to afford com-
pounds 2c together with minor amounts of 3c, 7c and
8c. Compound 2c as a colourless oil (62 mg, 28%); ee
2-{[4-Carboxy-2-hydroxy-2-(2-benzyloxyethyl)buta-
noyl]oxy}-2-(2-benzyloxyethyl)pentanedioic acid 8c (21
mg, 9%). 1H NMR (500 MHz CDCl3): l 1.90 (dt,
J=14.4, 2×5.1 Hz, 1H, H-1§), 1.98 (dt, J=15.5, 2×9.4
Hz, 1H, H-3%), 2.19 (ddd, J=14.4, 8.4, 5.1 Hz, 1H,
H-1§), 2.30 (m, 1H, H-3), 2.35 (m, 1H, H-3), 2.36 (m,
2H, H-3%, H-1¦), 2.43 (m, 1H, H-4%), 2.45 (m, 1H, H-4),
2.52 (m, 1H, H-1¦), 2.59 (m, 1H, H-4%), 2.60 (m, 1H,
H-4), 3.21 (dt, J=10, 5, 2×5.3 Hz, 1H, H-2¦), 3.47 (m,
1H, H-2§), 3.50 (m, 1H, H-2¦), 3.57 (m, 1H, H-2§),
4.33 and 4.41 (both d, J=11.9 Hz, 2H, b% or b¦), 4.37
and 4.39 (both d, J=11.8 Hz, 2H, b% or b¦), 7.24–7.35
(m, 10H, o%, o¦, m%, m¦, p%, p¦); 13C NMR (125 MHz
CDCl3): l 27.80 (C-4%), 28.16 (C-4), 30.45 (C-3), 32.03
(C-1¦), 33.51 (3%), 38.44 (C-1§), 64.92 (C-2¦), 65.61
(C-2§), 72.87 and 73.47 (b% and b¦), 75.80 (C-2%), 82.40
(C-2), 127.56 and 128.01 (o% and o¦), 127.66 and 127.82
1
98%, [h]2D0=−24 (c 2.96, CHCl3). H NMR (500 MHz
CDCl3): l 2.12 (dt, J=14.8, 2×5.5 Hz, 1H, H-2%), 2.32
(dt, J=13.5, 2×9.7 Hz, H-3), 2.46 (m, 1H, H-2%), 2.47
(m, 1H, H-3), 2.54 and 2.56 (m, 2H, H4), 3.67 (m, 2H,
H-2¦), 4.45 and 4.49 (2d J=11.8 Hz, 2H, Bn CH2),
7.25–7.34 (m, 5H, Bn o, m, p); 13C NMR (125 MHz
CDCl3): l 27.74 (C-4), 31.80 (C-3), 36.78 (C-2%), 65.10
(C-2¦), 73.08 (Bn CH2), 84.27 (C-2), 127.56 (o), 127.62
(p), 128.29 (m), 137.55 (s), 175.68 (2-COOH), 176.08
(C-5); EI (m/z, %): 264 (M+, 0.4), 219 (M−COOH+,
1.7), 158 (17.1), 140 (26.2), 107 (118.1), 91 (100); CI
(m/z, %): 265 (M+H+, 23); HRMS calcd for (M−
COOH)+ C13H15O3: 219.1020; found: 219.1053.
(R)-2-Hydroxy-2-benzyloxyethylpentanedioic acid (+)-
diethyltartrate ester 3c as colourless oil (65 mg, 16%).

1572
A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
(p% and p¦), 128.38 (m% and m%), 137.58 and 137.95 (s%
and s¦), 174.84 (C-1), 176.59 (C-1%), 179.54 (C-5%),
179.73 (C-5).
175.89 (C-5), 30.59, 30.54 (2-CH2), 8.11, 8.14 (2-Me),
171.07, 170.96 (COO), menthyl: 76.40, 76.24 (C-1),
46.73, 46.78 (C-2), 23.06, 23.18 (C-3), 34.07, 34.07
(C-4), 31.38, 31.38 (C-5), 40.49, 40.53 (C-6), 21.92,
21.92 (C-7), 26.22, 26.25 (C-8), 15.87, 15.98 (C-9),
20.75, 20.70 (C-10).
3.2.5. (R)-2-Hydroxymethyl-5-oxotetrahydrofuran-2-car-
boxylic acid, 2d. Diketone 1d was oxidized according to
a typical procedure and purified by column chromatog-
raphy (CHCl3/MeOH 20:1 to 10:1) to afford compound
2d as a colourless oil (88 mg, 55%); ee 94%, [h]1D9=+16
3.3.2. Enantiomeric purity of lactone acid 2c by HPLC
via 1,7-dioxaspiro[4.4]nonane-2,6-dione, 11. To a stirred
solution of lactone acid 2c (20 mg, 0.076 mmol) in
MeOH (2 mL) 10% Pd/C (10 mg) was added and the
mixture was stirred under an atmospheric pressure of
H2 at rt for 1 h. After filtration and removing the
solvents the residue was dissolved in CH2Cl2 (4 mL)
and a crystal of p-TsOH was added. After stirring at rt
for 3.5 h the mixture was washed with saturated
NaHCO3 solution, brine, dried (Na2SO4) and concen-
trated. The residue was purified by flash chromatogra-
phy (silica gel, petroleum ether/acetone 10:3) giving 9.4
mg (79%) of spirodilactone 11. The isolated product
was analysed by HPLC (column: Daicel, Chiralcel
ODH; hexane/iPrOH 7:3; flow rate 0.7 mL/min, detec-
tion at 206 nm). The peaks of (+)- and (−)-spirolactones
were detected at tR 14.7 min (major) and 16.5 min
(minor), respectively, (compared with the correspond-
ing racemic product).
1
(c 1.99, MeOH). H NMR (500 MHz CDCl3+CD3OD):
l 2.36 (m, 2H, H-3), 2.61 (m, 2H, H-4), 3.76 (d, J=12.4
Hz, 1H, H-2%), 3.94 (d, J=12.4 Hz, 1H, H-2%); 13C
NMR (125 MHz CDCl3+CD3OD): l 27.58 (C-3), 28.65
(C-4), 64.89 (C-2%), 87.78 (C-2), 172.77 (COOH), 178.05
(C-5); EI (m/z, %): 161 (M+H+, 2.9), 130 (38.9), 129
(22.4), 115 (M−COOH+, 92.3), 101 (46.0); CI (m/z, %):
161 (M+H+, 100); HRMS calcd for (M+H)+ C6H9O5:
161.0449; found: 161.0450.
3.3. Determination of the enantiomeric purity of the
lactone acids 2
3.3.1. Lactone acids 2a and 2b by NMR as (−)-menthol
esters, 9a and 9b. A mixture of 2 (0.1 mmol),
(1R,2S,5R)-(−)-menthol (31.2 mg, 0.2 mmol), DCC
(24.7 mg, 0.12 mmol) and DMAP (6.2 mg) in THF (1
mL) was stirred at rt for 5 h. The workup was per-
formed as described in Part 18 for derivatization of
primary hydroxylation products with methoxyphenyl-
acetic acid. Flash chromatography on silica gel (petrol
ether/ethyl acetate 15:1 to 10:1) yielded the (−) menthol
esters of the corresponding lactone-acids. Analogously
the diastereomeric esters from racemic 2a-rac and 2b-
rac were prepared. NMR chemical shifts of
diastereomeric esters in CDCl3 solution are reported
from the racemic mixture, values for nonracemic 2a
and 2b derivatives are given in italics.
3.3.3. Lactone acid 2d by HPLC as p-Br-phenacyl esters.
An analytical sample 0.2–0.5 mg of 2d in CH3CN was
treated with p-Br-phenacyl bromide and N,N-diiso-
propylethylamine. The obtained p-Br-phenacyl ester
was analysed by HPLC (column: Chiralcel ODH,
Daicel; hexane/iPrOH 6:4; flow rate 0.65 mL/min,
detection at 254 nm). The peaks of (+)-2d (major) and
(−)-2d (minor) were detected at tR 15.1 and 29.2 min,
respectively.
Compound 9a. 1H NMR (500 MHz CDCl3): l acid
part: 2.13/2.50, 2.13/2.49 (H-3), 2.58/2.63, 2.56/2.64
(H-4), 1.650, 1.646 (CH3), menthyl: 4.74/4.73 (H-1),
1.45/1.44 (H-2), 1.05/1.05, 1.70/1.69 (H-3), 0.87/0.87,
1.69/1.69 (H-4), 1.50/1.50 (H-5), 1.02/1.00, 1.96/1.96
(H-6), 0.906/0.906 (H-7), 1.79/1.83 (H-8), 0.749/0.749
(H-9), 0.894/0.890 (H-10); 13C NMR (125 MHz
CDCl3): l acid part: 83.86, 83.92 (C-2), 33.01, 33.10
(C-3), 28.46, 28.50 (C-4), 175.82, 175.89 (C-5), 23.61,
23.60 (2-Me), 171.18, 171.18 (COO), menthyl: 76.28,
76.25 (C-1), 46.78, 46.81 (C-2), 23.09, 23.25 (C-3),
34.00, 34.02 (C-4), 31.32, 31.32 (C-5), 40.36, 40.45
(C-6), 21.90, 21.90 (C-7), 26.29, 26.26 (C-8), 15.97,
16.10 (C-9), 20.74, 20.67 (C-10).
Acknowledgements
Support from Estonian Science Foundation (Grants
Nos 4976, 5133 and 5628) and from Estonian Ministry
of Education (Grant 0350312s01) is acknowledged.
References
1. Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980,
102, 5974–5976.
2. Pitchen, P.; Kagan, H. B. Tetrahedron Lett. 1984, 25,
1049–1052.
3. Di Furia, F.; Modena, G.; Seraglia, R. Synthesis 1984,
325–326.
4. Lopp, M.; Paju, A.; Kanger, T.; Pehk, T. Tetrahedron
Lett. 1996, 37, 7583–7586.
Compound 9b. 1H NMR (500 MHz CDCl3): l acid
part: 1.88/2.10, 2.17/2.44 (H-3), 2.57, 2.57 (H-4), 1.88/
2.10, 1.87/2.12 (CH2 of Et), 1.003, 1.000 (CH3), men-
thyl: 4.76/4.77 (H-1), 1.48/1.47 (H-2), 1.07/1.07,
1.72/1.71 (H-3), 0.89/0.89, 1.71/1.71 (H-4), 1.52/1.52
(H-5), 1.04/1.02, 2.00/1.99 (H-6), 0.923/0.923 (H-7),
1.84/1.86 (H-8), 0.765/0.768 (H-9), 0.911/0.909 (H-10);
13C NMR (125 MHz CDCl3): l acid part: 87.34, 87.38
(C-2), 31.19, 31.49 (C-3), 28.23, 28.27 (C-4), 175.77,
5. Lopp, M.; Paju, A.; Kanger, T.; Pehk, T. Tetrahedron
Lett. 1997, 38, 5051–5054.
6. Paju, A.; Kanger, T.; Pehk, T.; Lopp, M. Tetrahedron
2002, 58, 7321–7326.
7. Paju, A.; Kanger, T.; Pehk, T.; Lopp, M. Tetrahedron
Lett. 2000, 41, 6883–6887.

A. Paju et al. / Tetrahedron: Asymmetry 14 (2003) 1565–1573
1573
8. Paju, A.; Kanger, T.; Pehk, T.; Mu¨u¨risepp, A.-M.; Lopp,
M. Tetrahedron: Asymmetry 2002, 13, 2439–2448.
9. Payne, G. B. J. Org. Chem. 1959, 24, 719–720.
10. Utaka, M.; Nakatani, M.; Takeda, A. J. Org. Chem.
1986, 51, 1140–1141.
15. Mori, K. Tetrahedron 1975, 31, 1381–1384.
16. Donohoe, T. J.; Stevenson, C. A.; Helliwell, M.; Irshad,
R.; Ladduwahetty, T. Tetrahedron: Asymmetry 1999, 10,
1315–1322.
17. Partridge, J. J.; Shiuey, S.-J.; Chadha, N. K.; Baggiolini,
E. G.; Blount, J. F.; Uskokovic, M. R. J. Am. Chem. Soc.
1981, 103, 1253–1255.
11. Kwon, B.-M.; Foote, C. S. J. Org. Chem. 1989, 54,
3878–3882.
18. Jaeschke, G.; Seebach, D. J. Org. Chem. 1998, 63, 1190–
1197.
12. Abu-Omar, M. M.; Espenson, J. H. Organometallics
1996, 15, 3543–3549.
19. Branner-Jørgensen, S.; Berg, A. Acta Chem. Scand. 1966,
20, 2192–2194.
13. Chandrasekhar, M.; Chandra, K. L.; Singh, V. K. Tetra-
hedron Lett. 2002, 43, 2773–2775.
20. Amon, C. M.; Banwell, M. G.; Gravatt, G. L. J. Org.
Chem. 1987, 52, 4851–4855.
14. Chang, S.-W.; Hung, C.-Y.; Liu, H.-H.; Uang, B.-J.
Tetrahedron: Asymmetry 1998, 9, 521–529.