Kinetic resolution of epoxy alcohols with the Sharpless Ti-isopropoxide/tartaric ester complex

Article abstract of DOI:10.1016/j.tetasy.2016.05.007

When investigating the Sharpless epoxidation of enol-protected 4-hydroxy-1,2-cyclopentanediones, the ability of the asymmetric Ti(OiPr)₄/tartaric ester complex to discriminate between enantiomeric epoxides formed in situ was discovered, leading to the epoxide opening reaction of only one enantiomer. This observation was used in the kinetic resolution of racemic substituted 2,3-epoxy-4-hydroxy-cyclopentanol, to afford enantiomerically enriched epoxyalcohols in good yields and with ees up to 96%.

Full text of DOI:10.1016/j.tetasy.2016.05.007

Tetrahedron: Asymmetry 27 (2016) 608–613
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Kinetic resolution of epoxy alcohols with the Sharpless
Ti-isopropoxide/tartaric ester complex
a
a
a
b
a,
⇑
Karolin Maljutenko , Anne Paju , Ivar Järving , Tõnis Pehk , Margus Lopp
a
Department of Chemistry, Tallinn University of Technology, Akadeemia tee 15, 12618 Tallinn, Estonia
National Institute of Chemical Physics and Biophysics, Akadeemia tee 23, 12618 Tallinn, Estonia
b
a r t i c l e i n f o
a b s t r a c t
Article history:
When investigating the Sharpless epoxidation of enol-protected 4-hydroxy-1,2-cyclopentanediones, the
ability of the asymmetric Ti(OiPr) /tartaric ester complex to discriminate between enantiomeric epoxides
formed in situ was discovered, leading to the epoxide opening reaction of only one enantiomer. This
observation was used in the kinetic resolution of racemic substituted 2,3-epoxy-4-hydroxy-cyclopentanol,
to afford enantiomerically enriched epoxyalcohols in good yields and with ees up to 96%.
Ó 2016 Elsevier Ltd. All rights reserved.
Received 28 April 2015
Accepted 27 May 2015
Available online 11 June 2016
4
1
. Introduction
We have previously found that the asymmetric oxidation of
3
-alkyl-1,2-cyclopentanediones with the Sharpless complex
The kinetic resolution of allylic alcohols via Sharpless asymmet-
results in lactone carboxylic acids in good yield and with high ee
1
4
ric epoxidation is a widely used reaction in the synthesis of various
values (Scheme 1, A, example a). We have also reported that a
1
15,16
enantiomerically pure compounds and in the total synthesis of
hydroxyl group in the allylic or homoallylic
position plays an
2
,3
several natural products.
important role in determining the selectivity of the reaction. Thus,
3-benzyl-1,2-cyclopentanedione 5 with 4-silylprotected OH group
gave lactone carboxylic acids as a mixture of two diastereomers in
a 1:1 ratio with excellent enantioselectivities for both diastere-
omers 8 and 9 (ee values 97% and 95%), while the corresponding
It has previously been found that cyclic secondary allylic alco-
hols are poor substrates for Sharpless kinetic resolutions.4 In this
respect, cyclohexenol has been found to be the worst compound,
while substituted cyclohexenols5 and cyclopentenols6 afford
slightly better results in the oxidations. Because of this knowledge,
Sharpless asymmetric epoxidation is not usually used for the syn-
thesis of enantiomerically pure cyclic secondary epoxy alcohols.
However, there is a need for these compounds because they are
important intermediates in the various regio- and stereoselective
unprotected 4-hydroxy-3-benzyl-1,2-cyclopentanedione
6
afforded reduced enantiopurities of 70% ee and 80% ee for lactone
acids 10 and 11, but improved diastereoselectivity 3:1 (Scheme 1,
A, example b). In the first case, only the highly stereoselective
1
7
cascade oxidation may occur, while in the second case the
highly stereoselectivity cascade oxidation and low stereoselectivity
allylic oxidation of cyclic allylic alcohol compete, reducing the
enantiopurity of the resulting product.
2
,7
ring-opening reactions
intermediates.
that afford chiral building blocks and
8
Usually, the in situ ring opening of the epoxy alcohols formed
during Sharpless kinetic resolutions is the reaction that has to be
When the enol OH in 4-hydroxy-1,2-cyclopentanedione sub-
strate is protected, only the epoxidation of the allyl alcohol moiety
may occur. The Baeyer–Villiger-type oxidation does not proceed;
instead, the ring opening reaction of the resulting epoxide may
9
avoided, thus making the synthesis of certain epoxy alcohols dif-
ficult using standard asymmetric epoxidation processes. 2-Alkyl-
10
allylic alcohols often suffer from this limitation.
1
8
Although the kinetic resolution of substituted 2,3-epoxy alco-
hols has been extensively studied, only a few efficient cases have
been found: the resolution of meso-epoxy alcohols11 (the opening
of meso-epoxy alcohols with amines leads to chiral beta-amino
occur (Scheme 1, B).
Herein, the oxidation of allylic systems with the subsequent
epoxide opening is investigated. First, the obtained data on the oxi-
dation of 4-hydroxy-1,2-cyclopentanedione enol ether 1 led us to
the understanding that the enantiomers of the formed epoxide
behave differently in the presence of the Sharpless complex:
the epoxide openings proceed at different rates. This opens up
the possibility of the kinetic resolution of these epoxides. Thus,
1
2
13
alcohol units ) and terminal epoxy alcohols.
⇑
Corresponding author. Tel.: +372 513 6083.
E-mail address: margus.lopp@ttu.ee (M. Lopp).
http://dx.doi.org/10.1016/j.tetasy.2016.05.007
957-4166/Ó 2016 Elsevier Ltd. All rights reserved.
0

K. Maljutenko et al. / Tetrahedron: Asymmetry 27 (2016) 608–613
609
O
O
O
4
Ti(OiPr) /(+)-DET/
TBHP
OH
A
O
+
O
COOH
COOH
a) 1:1.6:2.5
1
1
1
R O
Bn
R O
Bn
R O
Bn
1
a) 8 R
1
= TBS
a) 9 R1 = TBS
27%, ee 95%
b) 11 R = H
11%, ee 80%
5
6
R
R
= TBS
= H
1
27%, ee 97%
b) 10 R = H
3%, ee 70%
1
1
3
O
O
O
OTBS
OTBS
OR
OTBS
Ti(OiPr) /(+)-DET/
4
Ti(OiPr)
4
/(+)-DET
B
2
TBHP
TBHP
O
OH
Bn
1
:1.6:1.5
HO
Bn
HO
Bn
HO
rac-1
(-)-2
1%,ee 46%
-)-3 R2 = iPr
(
4
2
4
R
= DET
Scheme 1. Reactions of 4-hydroxy-1,2-cyclopentanediones.
the epoxide opening of rac-2, cyclohexane epoxyalcohol rac-15 and
aliphatic epoxyalcohol rac-13 were investigated.
O
O
O
OTBS
OR
Ti(OiPr)
4
/(+)-DET/
OTBS
OTBS
TBHP
Herein we have demonstrate the possibility of the kinetic reso-
lution of racemic epoxides from 4-hydroxy-1,2-cyclopentanedione
by using the Sharpless titanium isopropoxide/tartaric ester com-
plex. We also found that with other common epoxy alcohols, the
resolution is not efficient enough for preparative purposes.
+
O
OH
Bn
HO
Bn
HO
Bn
HO
rac-1
(-)-2
(-)-3 R = iPr
4 R = DET
Scheme 3. Sharpless kinetic resolution of substrate rac-1.
2
2
. Results and discussion
2
.2. Allylic oxidation of cyclopentenols
.1. Synthesis of substrates
The preliminary results of the oxidation of rac-1 with (+)-DET
Scheme 3) showed that the substrate with an enol-protecting silyl
(
The 4-silyloxy substituted 3-benzyl-1,2-cyclopentanedione 5
group reacts, as expected, as an allylic alcohol to give epoxide (À)-2
as a single diastereomer, with modest ee values. Additionally, the
ring opening of the resulting epoxide also occurs to a certain
extent, to give isopropyl acetal 3 and tartaric ester acetal 4
was prepared according to a procedure recently described by
1
4
us. The protecting silyl group was removed with 3 M HCl in
THF to afford 6 in 70% yield. The enol hydroxyl group of 6 was
selectively protected to afford rac-1 (yield 66%). For the synthesis
of enantiomeric alcohols, the allylic alcohol was subjected to
enzymatic acylation to afford (À)-1 (yield 60%, ee 88%) and
enriched acetate 7 in 30% yield. The other enantiomer was
obtained by enzymatic deacetalization of enriched acetate 7 with
Novozyme 435, resulting in (+)-1 (yield 79%, ee 99%) (see
Scheme 2).
(
as single diastereomers) (Fig. 1). This result suggests stereodiffer-
entiation in the epoxide opening (Table 1, No. 1).
O
O
OEt
HO
OTBS
O
OEt
O
O
OEt
OEt
HO
O
O
OH
+
OTBS
OH
Bn
O
HO
Bn
HO
4
a
4b
Figure 1. Tartaric ester acetals.
O
O
O
OH
OH
OTBS
THF/HCl/H
2
O
TBS/imidazole
The epoxidation of highly enantioenriched (À)-1 (ee = 96%) with
Ti(OiPr)
/(À)-DET/TBHP complex gave epoxide (À)-2 in 18% yield
4
TBSO
Bn
HO
Bn
HO
Bn
(
ee 90%), isopropyl acetal (À)-3 in 2% isolated yield (ee 98%) and
5
6
rac-1
DET acetal 4 in 24% isolated yield. There was 17% of unreacted
substrate left, with ee 96%.
O
O
O
OTBS
At the same time, the epoxidation of (À)-1 (88% ee) with the
OTBS
OTBS
Novozyme 435
EtOAc
Ti(OiPr)
4
/(+)-DET/TBHP complex gave epoxide (À)-2 in 31%
isolated yield and with 88% ee, as expected. The unreacted (À)-1
retained its initial stereoisomeric state (88% ee). It is notable that in
this case, no isolatable amounts of acetals 3 and/or 4 were formed.
We looked at this interesting phenomenon in more detail.
When oxidizing substrates 1 with different enantiopurities with
HO
Bn
HO
Bn
AcO
Bn
rac-1
(-)-1
7
O
O
OTBS
OTBS
Novozyme 435
MeCN/MeOH
(
+)-DET/TBHP complex, we observed that the product profile (ratio
of acetals 3 + 4/epoxide 2) depended on the enantiomeric purity of
the initial substrate (Fig. 2). The lower the excess of one
enantiomer of the substrate, the higher the amount of formed
acetals (Fig. 2).
AcO
Bn
HO
Bn
7
(+)-1
Scheme 2. Synthesis of substrates.

6
10
K. Maljutenko et al. / Tetrahedron: Asymmetry 27 (2016) 608–613
Table 1
Sharpless epoxidation of 3-benzyl-1,2-cyclopentanedione 1
*
No.
DET stereoisomer
Substrate 1
Epoxide 2
iPrOH acetal 3
DET acetals 4 (%)
Unreacted substrate 1
1
2
3
(+)-DET
(À)-DET
(+)-DET
rac
(À)-2, 29%, ee = 46%
(À)-2, 18%, ee = 90%
(À)-2, 31%, ee = 88%
(À)-3, 7%, ee = 14%
(À)-3, 2%, ee = 98%
—
21
24
—
(À)-1, 12%, ee = 16%
(À)-1, 17% ee = 96%
(À)-1, 59%, ee = 88%
(À)-1, ee = 96%
(À)-1, ee = 88%
*
Conditions: À20 °C; solvent CH
2
Cl
2
; overnight.
Acetals/epoxide change with rise in substrate ee
.20
If this conclusion is correct, there must be the possibility of
separating the racemic 1,2-diketone epoxyalcohols by using
Sharpless Ti/tartaric ester complexes.
1
1
0
0
0
0
0
.00
.80
.60
.40
.20
.00
2.3. Kinetic resolution of allylic epoxides with the Sharpless
complex
Our assumption was tested by reacting racemic epoxyalcohol
rac-2 with the (+)-DET Sharpless complex. The selective epoxide
opening should eliminate (+)-2. Indeed, after a 1-day reaction there
was 66% of epoxide (À)-2 left with 60% ee (see Scheme 5).
20
40
60
80
100
Substrate ee (%)
Figure 2. Illustration of the decrease in the ratio of acetals 3 + 4/epoxide 2 with rise
in substrate 1 ee value for reactions with (+)-DET (Table 1).
O
O
O
OTBS
OTBS
OR
OTBS
Figure 1 demonstrates that when the ee value of substrate 1 was
higher, the yields of the acetals decreased. Also, in the experiment
with enantiomerically enriched substrates, the combination of
Ti(OiPr)
4
/(+)-DET
O
+
O
OH
Bn
1:1.6
HO
Bn
HO
Bn
HO
(
À)-1 with (À)-DET complex (Table 1, No. 2) produced more acetals
rac-2
(
-)-2
(
3
-)- R = iPr
than (À)-1 with (+)-DET (Table 1, No. 3). All of the results
presented above suggest a possible different behaviour of the
enantiomers of the substrate and the formed epoxides with the
chiral reagent. We propose that the obtained data may be rational-
ized as follows: in the two different reactions, epoxidation and
epoxide opening, match and mismatch pairs of the substrate/
reagent form, reacting at different rates and selectivities. First,
the kinetic resolution in the course of the epoxidation of allylic
cyclopentenol occurs with low selectivity; then, the second kinetic
resolution occurs in the epoxide opening reaction (acetal forma-
tion). This reaction proceeds with high enantioselectivity and in
this case the matched and mismatched pairs clearly cause different
reaction rates. The (À)-2 and (À)-DET combination results in the
matched pair for epoxide opening (Table 1, No. 2). This conclusion
is illustrated in Scheme 4.
4
R = DET
Scheme 5. Kinetic resolution of racemic epoxy alcohol rac-2.
The obtained DET-acetal 4 was formed in 29% and iPrOH-acetal
was formed in 4% yield (ee 50%). To obtained better conversion,
the reaction time was prolonged to 3 days, after which 54% of
3
highly enantioenriched unreacted epoxide (À)-2 was left (ee
9
6%). DET-acetal 4 was formed in 36% yield and acetal 3 as a single
diastereomer in 12% yield (ee 10%) (Table 2).
These experiments demonstrate that the kinetic resolution of
epoxide rac-2 in the ring opening proceeded with high selectivity.
Of the formed acetals, isopropyl acetal 3 had lower enantiopurity,
while tartaric ester acetal 4 gave high enantioselectivity in the
epoxide opening reaction.
O
OTBS
O
Bn
O
2.4. Kinetic resolution of other epoxy alcohols
HO
OTBS
Ti(OiPr)
4
/
(
-)-2
(
+)-DET /
Sharpless had previously noted that after 100% conversion in
the kinetic resolution of allylic alcohols, the erythro/threo ratio of
the obtained epoxy alcohols increases due to an enantioselective
HO
Bn
TBHP
O
O
rac-1
OTBS
OR
OTBS
4
opening process. Because of this knowledge, we were further
O
OH
encouraged to try our conditions with other epoxyalcohols. To elu-
cidate the scope of the reaction, attempts at the kinetic resolution
of an aliphatic epoxyalcohol 13 from hex-2-en-1-ol and cyclic
epoxy alcohol 15 were made. In the first case, the Sharpless oxida-
tion led to epoxy alcohol (À)-13 in 63% yield and with 92% ee. The
HO
Bn
HO
Bn
(
+)-2
(-)-3 R = iPr
4
R = DET
Scheme 4. Sharpless kinetic resolution of rac-1 and the following epoxide opening.
Table 2
*
Experiments with racemic epoxyalcohol 2
No.
Time (days)
DET stereoisomer
iPrOH acetal 3
DET acetals 4 (%)
Unreacted substrate
1
2
3
1
3
4
(+)-DET
(+)-DET
(À)-DET
(À)-3, 4%, ee 50%
(À)-3, 10%, ee 12%
(+)-3, 12%, ee 10%
29
37
33
(À)-2, 66%, ee 60%
(À)-2, 53%, ee 96%
(+)-2, 55%, ee 93%
*
Yields from RP-HPLC.

K. Maljutenko et al. / Tetrahedron: Asymmetry 27 (2016) 608–613
611
formation of a small amount of isopropyl ether 14 (4%) was also
observed, which indicated that the kinetic resolution of rac-12
should be possible (Scheme 6).
modest stereoselectivity. Two subsequent processes occur: allylic
epoxidation, which proceeds with low stereoselectivity, and the
epoxide opening, which is a highly stereoselective process. We
found that the Sharpless complex is able to discriminate between
enantiomeric in epoxides 2, thus catalysing their epoxide opening
at different rates. As a result, a kinetic resolution occurs, leading to
acetals from predominantly one enantiomer only. The unreacted
epoxide 2 can thus be obtained with high enantiomeric purity. This
method is also applicable to simple racemic epoxy alcohols, such as
13 and 15, but with less impressive results. We have demonstrated
that the Sharpless complex can be used not only for epoxidation
but also, as in the present case, for the kinetic resolution of racemic
epoxides.
4
Ti(OiPr) /
(
+)-DET/
OiPr
TBHP
HO
HO
+
HO
1/1.6/1.5
O
OH
1
2
(-)-13
14
6
3%
4%
Scheme 6. Sharpless kinetic resolution of allylic alcohol 12.
Compound rac-13 was prepared by using an mCPBA oxidation
of allylic alcohol 12, resulting in 88% of the target product.
When rac-13 was subjected to a reaction with the (+)-DET
Sharpless complex with a substrate/reagent ratio of 1:1.6, after
two days 56% of unreacted substrate (+)-13 was left with ee 14%
4. Experimental
The full assignment of the H and 13C chemical shifts is based on
the 1D and 2D FT NMR spectra measured on a Bruker Avance III
400 MHz or Bruker Avance III 800 MHz instrument. High resolu-
tion mass spectra were recorded by using an Agilent Technologies
6540 UHD Accurate-Mass Q-TOF LC/MS spectrometer by using
AJ-ESI ionization. Elemental analyses were done by using Elemen-
tar vario Micro. IR spectra were recorded on a Bruker Tensor 27 FT
infrared spectrophotometer. MS spectra were measured on a
Shimadzu GSMS-QP2010 spectrometer on a 70 eV EI. Optical
rotations were obtained using an Anton Paar GWB Polarimeter
MCP 500. Chiral GC was performed using Shimadzu GC-2010
1
(Scheme 7). After 8 days of reaction, the enantiomeric purity of
(
+)-13 increased to 42%. At the same time, several by-products
formed, and we were able to isolate 9% of iPrOH-ether 14 (Table 3,
Nos. 1–2).
OiPr
4
Ti(OiPr) / (+)-DET
HO
HO
HO
O
O
OH
13
(+)-13
14
TM
Supelco b-DEX 225 column (30 m  0.25 mm). Chiral HPLC was
2
8
days 3%
days 9%
performed using Chiralpak AD-H (250 Â 4.6 mm), Chiralcel OJ-H
(
(
250 Â 4.6 mm), Chiralcel OD-H (250 Â 4.6 mm), Chiralpak AS-H
250 Â 4.6 mm) or Lux 3u Amylose-2 (250 Â 4.6 mm) columns.
Scheme 7. Kinetic resolution of epoxy alcohol 13 with the Sharpless complex.
Precoated silica gel 60 F254 plates from Merck were used for
TLC, whereas for column chromatography silica gel Kieselgel
Table 3
Kinetic resolution of epoxy alcohols 13 and 15 with the Sharpless complex
4
0–63 lm was used. Purchased chemicals and solvents were used
as received. DCM was distilled over phosphorous pentoxide.
Petroleum ether has a boiling point 40–60 °C. The reactions were
performed under air atmosphere without additional moisture
elimination unless stated otherwise.
No.
Substrate
Time (days)
Unreacted substrate
iPrOH-ether
1
2
3
4
rac-13
rac-13
rac-15
rac-15
2
8
3
7
13; 56%, ee 14%
13; 30%, ee 42%
15; 44%, ee 6%
15; 56%, ee 10%
14; 3%
14; 9%
16; 1%
16; 2%
4
.1. Synthesis of 3-benzyl-2,4-dihydroxycyclopent-2-enone 6
To a solution of diketone 5 (1.017 g, 3.2 mmol) in THF (20 mL), a
M HCl solution (7.7 mL) was added, and the reaction mixture was
We also prepared epoxyalcohol 15 from cyclohex-2-ene-1-ol
with mCPBA in 73% yield. The kinetic resolution of the product
3
stirred for 3 h at room temperature. Next, H
and the mixture was extracted 3 times with EtOAc. The extracts
were dried over MgSO , and the crude product was purified by
column chromatography (heptane:acetone 10:3). 394 mg (60%) of
2
O (40 mL) was added
with Ti(OiPr)
starting epoxide, less than in the case of the aliphatic compound
3 (Scheme 8). Prolonging the reaction time led to small improve-
ments: after three days, 15 was isolated with 6% ee, and after
days, with 10% ee. We were able to isolate small amounts of
4
/(+)-DET led to a small enantioenrichment of the
4
1
1
diketone
400 MHz, MeOD) d 7.37–7.07 (m, 5H, Ph), 4.53 (d, J = 6.1 Hz, 1H,
H-4), 3.94 and 3.54 (2d, J = 14.2 Hz, 2H, Ph-CH ), 2.74–2.62 (m,
H, H-5), 2.21–2.10 (m, 1H, H-5). C NMR (101 MHz, MeOD) d
01.10 (C-1), 149.88 (C-2), 145.42 (C-3), 137.87 (s-Ph), 129.10
6
were obtained as
a
colourless oil. 6:
H NMR
7
(
iPrOH-ether 16 from a mixture of several by-products (Table 3,
Nos. 3–4).
2
13
1
2
(
Ph), 128.66 (Ph), 126.54 (p-Ph), 65.96 (C-4), 42.68 (C-5), 31.16
OH
OH
OH
À1
Ti(OiPr)
+)-DET
4
/
(Ph-CH
). IR (film, cm ): 3265, 1702, 1658, 1385, 1049, 980,
2
OH
(
7
7
5
62, 637. MS (m/z %): 186, 158, 129 (main peak), 115, 105, 91,
7, 65, 51. Elemental analysis calcd for C12 C, 70.57; H,
.92; found C, 69.76; H, 6.19.
O
O
12 3
H O
OiPr
rac-15
15
16
4
.2. Synthesis of 3-benzyl-2[tert-butyldimethylsilyl]oxy-4-
Scheme 8. Kinetic resolution of epoxy alcohol 15 with the Sharpless complex.
hydroxycyclopent-2-enone rac-1
3
. Conclusion
2 2
Diketone 6 (110 mg, 0.54 mmol) was dissolved in CH Cl (5 mL),
after which imidazole (52 mg, 0.77 mmol) and TBSCl (89 mg,
0.59 mmol) were added. The mixture was stirred at room temper-
ature for 1.5 h, after which H O (10 mL) was added. The mixture
2
The kinetic resolution of secondary allylic cyclopentenol rac-1
gives epoxides and acetals with excellent diastereoselectivity, but

6
12
K. Maljutenko et al. / Tetrahedron: Asymmetry 27 (2016) 608–613
was extracted 3 times with CH
2
Cl
2
, and the extracts were dried
1 ml/min, 210.8 nm, ee = 58%, major 9.7 min, minor 8.4 min. HRMS
(ES) m/z calcd for [MÀH] 333.1528; found 333.1524.
À
over MgSO . The obtained crude product was purified by column
4
chromatography (heptane:EtOAc 20:1–20:3). 269 mg (66%) of
diketone rac-1 was obtained as a colourless oil. rac-1: 1H NMR
4.5.1. 3-Benzyl-2-((tert-butyldimethylsilyl)oxy)-3,4-dihydroxy-
(
400 MHz, CDCl
3
) d 7.35–7.19 (m, 5H, Ph), 4.60 (d, J = 5.5 Hz, 1H,
), 2.73–2.63 (m,
H, H-5), 2.26–2.18 (m, 1H, H-5), 1.92 (br s, 1H, OH), 0.98 (s, 9H,
2-isopropoxycyclopentanone (+)-3
2
5
1
H-4), 3.94 and 3.61 (2d, J = 14.4 Hz, 2H, Ph-CH
1
t-Bu), 0.26 and 0.25 (s, 6H, CH
2
[
a]
D
= +127 (c 0.07, CHCl
7.32–7.17 (m, 5H, Ph), 4.44–4.34 (m, 1H, H-4), 4.03 (hept,
J = 6.2 Hz, 1H, iPr CH), 3.06 and 2.97 (2d, J = 13.8 Hz, 2H, Ph-CH ),
3
) H NMR (400 MHz, chloroform-d) d
13
3
-Si-CH
3
). C NMR (101 MHz, CDCl
3
)
2
d 199.95 (C@O), 150.83 (C-2), 150.28 (C-3), 137.77 (s-Ph), 129.09
2.69 (dd, J = 19.4, 8.7 Hz, 1H, H-5), 2.48 (s, 1H, 3-OH), 1.98 (dd,
J = 19.4, 7.4 Hz, 1H, H-5), 1.21 (s, 1H, 4-OH) 1.16 and 0.92 (2d,
(
(
(
Ph), 128.94 (Ph), 126.84 (p-Ph), 66.25 (C-4), 42.89 (C-5), 31.65
Ph-CH
Si-CH
2
), 25.86 (t-Bu CH
). IR (film, cm ): 3252, 2856, 1748, 1496, 1389, 1258,
3
), 18.52 (t-Bu C), À3.71 (Si-CH
3
), À3.90
J = 6.2 Hz, 6H, iPr 2 Â CH
3
), 0.79 (s, 9H, t-Bu), 0.20 (s, 3H, Si-CH
3 3
). C NMR (101 MHz, CDCl ) d 208.24 (C@O),
3
),
À1
13
3
À0.00 (s, 3H, Si-CH
1
6
064, 837, 779, 701. Elemental analysis calcd for C17
7.88; H, 8.23; found C, 67.38; H, 8.36.
26
H O
3
Si C,
137.32 (s-Ph), 130.39 (Ph), 129.06 (Ph), 127.07 (p-Ph), 101.88
(C-2), 78.97 (C-3), 69.95 (C-4), 65.61 (iPr CH), 40.45 (C-5), 37.15
(
Ph-CH
2
3 3 3
), 26.03 (t-Bu CH ), 24.89 (iPr CH ), 23.16 (iPr CH ), 18.81
4
4
.3. Synthesis of (4S)-3-benzyl-2[tert-butyldimethylsilyl]oxy-
-hydroxycyclopent-2-enone (À)-1
(t-Bu C), À2.94 (Si-CH
3
), À3.32 (Si-CH
3
). HPLC: AD-H Hex:iPrOH
97:3, 1 ml/min, 210 nm, ee = 74%, major 4.9 min, minor 5.9 min.
À
HRMS (ES) m/z calcd for [MÀH] 393.2103; found 393.2116.
Novozym SP 435 (195 mg) was added to a solution of diketone 1
(
195 mg, 0.61 mmol) in EtOAc (4 mL). The reaction mixture was
4.5.2. 3-Benzyl-2-((tert-butyldimethylsilyl)oxy)-2,3,4-trihydroxy-
then stirred overnight after which it was filtered and concentrated.
The products were separated by column chromatography
cyclopentane-1-one diethyl tartrate acetal 4a
1
H NMR (800 MHz, chloroform-d) d 7.33–7.28 (m, 5H, Ph), 4.75
(
Heptane:EtOAc 10:1). Product (À)-1 (151 mg, 77%) was obtained
(d, J = 3.6 Hz, 1H, DET CHO), 4.45 (br dd, J = 8.0, 3.6 Hz, 1H, DET
as a colourless oil and acylated diketone 7 (39 mg, 18%) was
CHOH), 4.34–4.22 (m, 4H, DET OCH
4.18–4,15 (br m, 1H, H-4), 3.76 (s, 1H, 3-OH), 3.35 (d, J = 8.9 Hz,
1H, DET OH), 3.17 (d, J = 14.2 Hz 1H, Ph-CH ), 2.67 (dd, J = 20.0,
7.9 Hz, 1H, H-5), 2.52 (d, J = 14.2 Hz, 1H, Ph-CH ), 2.36 (dd,
J = 20.0, 4.2 Hz, 1H, H-5), 1.34 and 1.32 (2t, J = 7.2 Hz, 6H, DET
CH ), 0.93 (s, 9H, t-Bu), 0.41 (s, 3H, Si-CH ), 0.09 (s, 3H, Si-CH ).
C NMR (201 MHz, chloroform-d) d 208.44 (C@O), 171.28 (DET
C@O), 169.65 (DET C@O), 136.22 (s-Ph), 130.68 (o-Ph), 128.62
m-Ph), 127.08 (p-Ph), 102.26 (C-2), 79.24 (C-3), 74.17 (DET
CHO), 72.44 (DET CHO), 68.50 (C-4), 62.36 (DET OCH ), 62.02
(DET OCH ), 41.34 (C-5), 38.07 (Ph-CH ), 25.89 (t-Bu CH ) 18.70
(t-Bu C), 14.22 (DET CH ), 14.11 (DET CH ), À4.13
), À2.60 (Si-CH
(Si-CH ). HRMS (ES) m/z calcd for [M+Na] 563.2283; found
2
), 4.23 (br s, 1H, 4-OH),
2
5
obtained as a colourless oil. (À)-1: [
a]
D
= À89.6 (c 1.27, CHCl
3
);
À1
IR (film, cm ): 3421, 3064, 1720, 1642, 1365, 1253, 1117, 1053,
2
8
40, 699; HPLC: AD-H Hex–iPrOH 97:3, 1 ml/min, 210 nm,
2
ee = 88%, major 9.1 min, minor 11.9 min.
3
3
3
1
3
4
4
.4. Synthesis of (4R)-3-benzyl-2[tert-butyldimethylsilyl]oxy-
-dihydroxycyclopent-2-enone (+)-1
(
Novozym SP 435 enzyme (225 mg) was added to a solution of
2
acylated diketone
(
reaction mixture was filtered and concentrated. The crude product
was purified by column chromatography (Heptane:EtOAc 10:1).
Diketone (+)-1 (104 mg, 79%) was obtained as a colourless oil.
7
(150 mg, 0.42 mmol) in MeCN/MeOH
2
2
3
7.5 mL, 0.3 mL of MeOH and 7.2 mL of MeCN). After 4 days the
3
3
3
+
3
563.2298.
2
5
(
+)-1: [
a]
D
3
= +122.8 (c 0.36, CHCl ); HPLC: AD-H Hex–iPrOH
4.5.3. 3-Benzyl-2-((tert-butyldimethylsilyl)oxy)-2,3,4-trihydroxy-
9
7:3, 1 ml/min, 210 nm, ee = 99%, major 11.9 min, minor 8.9 min.
cyclopentane-1-one diethyl tartrate diacetal 4b
1
H NMR (800 MHz, chloroform-d) d 7.29–7.24 (m, 5H, Ph), 4.55
4
.5. Synthesis of 1-benzyl-5-[tert-butyldimethylsilyl]oxy-2-
(d, J = 9.3 Hz, 1H, DET CHO), 4.47 (d, J = 9.3 Hz, 1H, DET CHO), 4.27–
hydroxy-6-oxabicyclo[3.1.0]hexane-4-one (+)-2
4.22 (m, 2H, DET OCH
OCH ) 3.77 (s, 1H, 3-OH) 3.74 (br m, 1H, H-4), 3.15 (br s, 1H,
1-OH), 3.06 (d, J = 14.7 Hz, 1H, Ph-CH ), 2.90 (dm, J = 14.7 Hz, 1H,
Ph-CH ), 2.85 (br d, J = 10.4 Hz, 1H, 4-OH), 2.55 (dd, J = 16.5,
7.9 Hz, 1H, H-5), 2.26 (dm, J = 16.5 Hz, 1H, H-5), 1.30 and 1.29
(2t, J = 7.2 Hz, 6H, DET CH ), 1.00 (s, 9H, t-Bu), 0.47 (s, 3H,
2
), 4.12 and 4.11 (2q, J = 7.2 Hz, 2H, DET
2
At first, (À)-DET (0.14 mL, 0.78 mmol) was added at À20 °C
2
under an argon atmosphere to a solution of Ti(O-i-Pr)
4
(0.15 mL,
(3 mL). After
5 min of stirring, compound 1 (156 mg, 0.49 mmol) was added
Cl (1 mL). After stirring for 30 min, TBHP
0.13 mL, 0.78 mmol) was added. After an overnight reaction,
diethyl ether (10 mL) and a saturated Na SO solution (0.49 mL)
2
0
2 2
.49 mmol) and molecular sieves (49 mg) in CH Cl
1
3
1
3
as a solution in CH
2
2
3 3
Si-CH ), 0.29 (s, 3H, Si-CH ). C NMR (201 MHz, chloroform-d) d
(
167.67 (DET C@O), 167.15 (DET C@O), 136.49 (s-Ph), 131.12
(o-Ph), 128.17 (m-Ph), 126.73 (p-Ph), 100.24 (C-2), 99.34 (C-1),
80.49 (C-3), 72.57 (DET CHO), 69.41 (C-4), 69.11 (DET CHO),
2
4
were added at À20 °C and the mixture was stirred for 2 h at room
temperature. The mixture was filtered through Celite and concen-
trated. The crude mixture was purified by column chromatography
62.38 (DET OCH
26.24 (t-Bu CH
CH
), À1.56 (Si-CH
2
), 62.23 (DET OCH
), 18.75 (t-Bu C), 14.08 (DET CH
), À3.11 (Si-CH ).
2
), 40.46 (C-5), 37.93 (Ph-CH
2
),
3
3
), 14.02 (DET
(
Heptane:EtOAc 15:1–10:3) to afford 60 mg (37%) of epoxide (+)-2
3
3
3
as an amorphous solid, 19 mg (10%) of acetal (+)-3 as a colourless
oil and 49 mg (18%) of acetals 4 as a colourless oil. (+)-2:
4.6. General procedure for the synthesis of racemic epoxy
alcohols
2
5
1
[
7
(
a
]
D
= +72.7 (c 1.7, CHCl
.43–7.11 (m, 5H, Ph), 4.10 (d, J = 6.0 Hz, 1H, H-2), 3.44 and 3.04
), 2.55–2.04 (m, 2H, H-3), 0.99 (s, 9H,
3
), H NMR (400 MHz, chloroform-d) d
2d, J = 14.2 Hz, 2H, Ph-CH
t-Bu), 0.34 (s, 3H, Si-CH
101 MHz, CDCl ) d 203.67 (C@O), 135.55 (s-Ph), 129.62 (Ph),
29.09 (Ph), 127.35 (p-Ph), 88.27 (C-5), 73.81 (C-1), 65.43 (C-2),
2
2 2
To a solution of allylic alcohol (1 mmol) in CH Cl (3.5 mL),
m-CPBA (1.2 equiv) was added at 0 °C. The mixture was allowed
to gradually warm up to room temperature. After 30 min, 1.5 mL
13
3
), 0.24 (s, 3H, Si-CH
3
).
C
NMR
(
1
4
3
of saturated NaHCO
the mixture was stirred for 30 min. The organic layer was sepa-
rated and the mixture was extracted with CH Cl , then washed
with saturated NaHCO and brine, and dried with Na SO . The
3 2 2 3
and 1.5 ml of 10% Na S O were added and
1.33 (C-3), 33.23 (Ph-CH
2
), 25.71 (t-Bu CH
3
), 18.14 (t-Bu C),
). IR (KBr, cm ): 3000, 1764, 1608,
À1
À3.84 (Si-CH
3
), À4.03 (Si-CH
3
2
2
1
496, 1254, 1065, 858, 702. HPLC: AS-H Hex–iPrOH 97.5:2.5,
3
2
4

K. Maljutenko et al. / Tetrahedron: Asymmetry 27 (2016) 608–613
613
filtered and concentrated crude product was purified by column
chromatography (Petroleum ether:Acetone 10:2).
HPLC analysis was performed after benzoylation of the product.
1
15: H NMR (400 MHz, CDCl
3
) d 4.04–3.95 (m, 1H), 3.36–3.28 (m,
2
H), 2.20–2.11 (m, 1H), 1.91–1.68 (m, 2H), 1.61–1.49 (m, 2H),
4
.7. Procedure for the kinetic resolution of rac-2, rac-13 and
1.50–1.37 (m, 1H), 1.32–1.18 (m, 1H). HPLC AD-H 95:5, 1 ml/min,
rac-15
230 nm, major 7.9 min, minor 9.1 min, ee = 10%.
At first, (À)-DET (0.16 mL, 0.82 mmol) solution in CH
0.5 mL) was added at À20 °C under an argon atmosphere to a
(0.16 mL, 0.51 mmol) and molecular sieves
(2.5 mL). After 15 min of stirring, the epoxyalco-
Cl
2 mL). After an overnight reaction, diethyl ether (11 mL) and
saturated Na SO
the mixture was stirred for 2 h at room temperature. The mixture
was then filtered through Celite and concentrated. The crude mix-
ture was purified by column chromatography with Hept:EtOAc
2
Cl
2
Acknowledgements
(
(
solution of Ti(OiPr)
51 mg) in CH Cl
2 2
4
This work has been partially supported by graduate school
‘Functional materials and technologies’ receiving funding from
the European Social Fund under project 1.2.0401.09-0079 in
Estonia and Estonian Ministry of Education and Research grants
IUT 19-32 and IUT23-7.
hol substrate (0.55 mmol) was added as a solution in CH
(
2
2
2
4
solution (0.51 mL) were added at À20 °C and
References
1.
2.
3.
4.
Shibuya, H.; Kawashima, K.; Narita, N.; Ikeda, M.; Kitagawa, I. Chem. Pharm.
Bull. 1992, 40, 1154–1165.
Li, X.; Jiao, X.; Liu, X.; Tian, C.; Dong, L.; Yao, Y.; Xie, P. Tetrahedron Lett. 2014, 55,
1
5:1–10:3 mixture to yield 89 mg (53%) of epoxide (À)-2, 6 mg
(
4%) of acetal 3 and 43 mg (16%) of acetals 4.
6
324–6327.
Rodriguez-López, J.; Ortega, N.; Martin, V. S.; Martin, T. Chem. Commun. 2014,
685–3688.
4
.8. Kinetic resolution of 2,3-epoxyhexanol 13
3
Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.; Ikeda, M.; Sharpless, B. K. J.
Am. Chem. Soc. 1981, 103, 6237–6240.
From the epoxyalcohol substrate 13 (87 mg, 0.75 mmol) with
5
6
.
.
Yamamoto, H.; Oritani, T. Biosci., Biotechnol., Biochem. 1994, 58, 992–993.
Take, K.; Okumura, K.; Tsubaki, K.; Taniguchi, K.; Shiokawa, Y. Chem. Pharm.
Bull. 2000, 48, 1903–1907.
(
CH
À)-DET (0.23 mL, 1.2 mmol), Ti(OiPr)
4
(0.23 mL, 0.75 mmol) in
(7 mL), 49 mg (56%) of epoxyalcohol (+)-13 and 4 mg (3%)
2
Cl
2
of isopropyl ether 14 were obtained. The data for the compounds
7. Caron, M.; Sharpless, K. B. J. Org. Chem. 1985, 50, 1560–1563.
8. Krishna, P. R.; Nomula, R.; Krishna, V. S. R. Tetrahedron Lett. 2014, 55, 3381–
3383.
1
9
25
were in accordance with the literature. 13 [
CHCl ), H NMR (400 MHz, CDCl
3 3
(
4
a
]
D
= +1.7 (c 0.49,
) d 3.94–3.86 (m, 1H), 3.66–3.57
m, 1H), 2.98–2.89 (m, 2H), 2.02–1.87 (m, 1H), 1.59–1.40 (m,
1
9. Ko, S. Y.; Sharpless, K. B. J. Org. Chem. 1986, 51, 5413–5415.
1
1
1
0. Lu, L. D.-L.; Johnson, A.; Finn, M. G.; Sharpless, K. B. J. Org. Chem. 1984, 49, 731–
733.
1. Belonok, Y. N.; Chusov, D.; Peregudov, A. S.; Yashkina, L. V.; Timofeeva, G. I.;
Maleev, V. I.; North, M.; Kagan, H. B. Adv. Synth. Catal. 2009, 351, 3157–3167.
2. Kumar, M.; Kureshy, R. I.; Ghosh, D.; Khan, N. H.; Abdi, S. H. R.; Bajaj, H. C.
ChemCatChem 2013, 5, 2336–2342.
TM
H), 0.95 (t, J = 7.3 Hz, 3H). Chiral GC b-DEX 225 column:
ee = 14% major 11.0 min, minor 10.8 min. 3-Isopropoxyhexane-
1
1
3
1
3
,2-diol 14: H NMR (400 MHz, CDCl ) d 3.83–3.75 (m, 1H),
.75–3.68 (m, 1H), 3.65–3.58 (m, 2H), 3.51 (dt, J = 7.0, 3.9 Hz,
H), 2.71 (br, 2H), 1.60–1.26 (m, 4H), 1.15 and 1.14 (2d,
1
1
3. Luo, L.; Yamamoto, H. Org. Biomol. Chem. 2015, 13, 10466–10470.
4. Paju, A.; Oja, K.; Matkevitš, K.; Lumi, P.; Järving, I.; Pehk, T.; Lopp, M.
Heterocycles 2014, 88, 981–995.
J = 6.1 Hz, 2 Â 3H), 0.93 (t, J = 7.0 Hz, 3H).
15. Paju, A.; Kanger, T.; Pehk, T.; Lindmaa, R.; Müürisepp, A.-M.; Lopp, M.
Tetrahedron: Asymmetry 2003, 14, 1565–1573.
4
.9. Kinetic resolution of 2,3-epoxycyclohexanol 15
16. Paju, A.; Kanger, T.; Niitsoo, O.; Pehk, T.; Müürisepp, A.-M.; Lopp, M.
Tetrahedron: Asymmetry 2003, 14, 2393–2399.
From the epoxyalcohol substrate 15 (69 mg, 0.7 mmol), (À)-DET
17. Paju, A.; Kanger, T.; Pehk, T.; Lopp, M. Tetrahedron Lett. 2000, 41, 6883–6887.
1
1
2
8. Caron, M.; Sharpless, K. B. J. Org. Chem. 1985, 50, 1560–1563.
9. Wang, C.; Yamamoto, H. J. Am. Chem. Soc. 2014, 136, 1222–1225.
0. Elhalem, E.; Comin, M. J.; Rodriguez, J. B. Eur. J. Org. Chem. 2006, 4473–4482.
(
(
4 2 2
0.21 mL, 1.12 mmol), Ti(OiPr) (0.2 mL, 0.7 mmol) in CH Cl
6.5 mL), 35 mg (44%) of epoxyalcohol 15 were obtained. The data
2
0
for the epoxyalcohol were in accordance with the literature.