Asymmetrie synthesis of bicyclic diol derivatives through metal and enzyme catalysis: Application to the formal synthesis of sertraline

Article abstract of DOI:10.1002/chem.200903114

Enzyme- and ruthenium-catalyzed dynamic kinetic asymmetric transformation (DYKAT) of bicyclic diols to their diacetates was highly enantio- and diastereoselective to give the corresponding diacetates in high yield with high enantioselectivity (99.9% ee). The enantiomerically pure diols are accessible by simple hydrolysis (NaOH, MeOH), but an alternative enzyme-catalyzed ester cleavage was also used to give the trans-diol (R,R)1b in extremely high diastereomeric purity (trans/cis = 99.9:0.1, >99.9% ee). It was demonstrated that the diols can be selectively oxidized to the ketoalcohols in a ruthenium-catalyzed Oppenauer-type reaction. A formal enantioselective synthesis of sertraline from a simple racemic cis/trans diol 1b was demonstrated.

Full text of DOI:10.1002/chem.200903114

FULL PAPER
DOI: 10.1002/chem.200903114
Asymmetric Synthesis of Bicyclic Diol Derivatives through Metal and
Enzyme Catalysis: Application to the Formal Synthesis of Sertraline
Patrik Krumlinde,[a] Krisztiꢀn Bogꢀr,[a, b] and Jan-E. Bꢁckvall*[a]
Abstract: Enzyme- and ruthenium-cat-
alyzed dynamic kinetic asymmetric
transformation (DYKAT) of bicyclic
diols to their diacetates was highly
enantio- and diastereoselective to give
the corresponding diacetates in high
yield with high enantioselectivity
(99.9% ee). The enantiomerically pure
diols are accessible by simple hydroly-
sis (NaOH, MeOH), but an alternative
enzyme-catalyzed ester cleavage was
also used to give the trans-diol (R,R)-
1b in extremely high diastereomeric
purity (trans/cis=99.9:0.1, >99.9% ee).
It was demonstrated that the diols can
be selectively oxidized to the ketoalco-
hols in a ruthenium-catalyzed Oppen-
auer-type reaction. A formal enantiose-
lective synthesis of sertraline from a
simple racemic cis/trans diol 1b was
demonstrated.
Keywords: alcohols
· asymmetric
catalysis · dynamic resolution · en-
zyme catalysis · metal catalysis
Introduction
has recently been involved in combined enzyme- and transi-
tion-metal-catalyzed DKR of secondary alcohols,[5a] primary
amines,[6b,c] and primary alcohols[7] as well as the dynamic ki-
netic asymmetric transformation (DYKAT)[8] of acyclic
diols.[9] In the DYKAT of acyclic diols, coupled resolution
and epimerization of a dl/meso mixture of the diol occur in
situ, thus resulting in one enantiomer of the diol diacetate.
The DYKAT of cyclic diols has been less studied and previ-
ous attempts to obtain enantiomerically pure diol diacetates
from 1,3-cyclohexanediol were moderately successful.[10] It
occurred to us that bicyclic diols 1 may be more suitable
substrates for enzymatic transesterification, and we there-
fore decided to study them in DYKAT reactions.
Mixtures of cis- and racemic trans-diols 1 are readily ac-
cessible from the corresponding diones. Subsequent
DYKAT reactions of these diols would be an efficient route
toward enantiomerically pure diols 1. Herein, we report the
DYKAT reactions of diols 1 in high yield with 99.9% ee and
apply diol 1b to the formal synthesis of sertraline.
Chiral bicyclic diols of type 1 are potentially important syn-
thetic intermediates.[1,2] For example, diol 1b and derivatives
thereof are useful precursors in
the synthesis of sertraline, its
analogues,[1] and other biologi-
cally active 1,4-disubstituted
tetralins.[2]
Recently, Kꢀndig et al. re-
ported
a procedure for the
enantioselective reduction of tetralin-1,4-dione to give
(R,R)-tetralin-1,4-diol ((R,R)-1b) in 72% yield with
99% ee.[3]
During the past decade dynamic kinetic resolution
(DKR) has evolved as an efficient method in catalytic asym-
metric synthesis[4] for the preparation of enantiomerically
pure secondary alcohols[5] and primary amines.[6] Our group
[a] P. Krumlinde, Dr. K. Bogꢁr, Prof. J.-E. Bꢂckvall
Department of Organic Chemistry
Arrhenius Laboratory, Stockholm University
10691 Stockholm (Sweden)
Results and Discussion
Fax : (+46)8-15-49-08
E-mail: jeb@organ.su.se
Many transition-metal catalysts have been developed for the
racemization of alcohols. Our group has successfully em-
ployed racemization catalysts 2 and 3 (Scheme 1)[11] in DKR
and DYKAT processes, and ruthenium complex 2 is one of
the best catalysts known today in dynamic processes with al-
cohols. Many catalysts have been shown to racemize alco-
hols quickly, but only a few are compatible with enzymes.
[b] Dr. K. Bogꢁr
Medicinal Chemistry Discovery
AstraZeneca R&D
15185 Sçdertꢂlje (Sweden)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200903114.
Chem. Eur. J. 2010, 16, 4031 – 4036
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4031

enantiopure. Indeed, this outcome was obtained when carry-
ing out the KAT reaction with diol 1b with CALB as the
lipase at ambient temperature for 24 hours. Prolonged reac-
tion times produced a small amount of meso-diacetate.
Because the KAT of diol 1b was very selective at ambient
temperature, DYKAT with CALB and Ru catalyst 2 was
tested at this temperature. However, epimerization cata-
lyzed by 2 was too slow at ambient temperatures, which has
previously been observed for diols.[9c–e] Diols 1a,b gave ex-
cellent results at 508C with high yields of around 90%
(Table 1, entries 2–5). Diol 1a reached full conversion be-
tween 24 and 48 hours (Table 1, entry 1; 94 versus >99%
conversion), whereas diol 1b took at least 48 h. A by-prod-
uct in the DYKAT of diols 1a–c corresponds to acetoxy ke-
tones 8a–c, which are difficult to remove by column chro-
matography. The yields stated in Table 1 include both 8 and
cis diastereomer cis-7.
Scheme 1. Ruthenium racemization catalysts 2 and 3.
Catalyst 2 in combination with Candida antarctica lipase B
(CALB) often provides efficient routes to enantio- and dia-
stereomerically pure alcohols. In this study, DYKAT pro-
cesses of diols 1a–c have been developed by using the com-
bination of catalyst 2 and CALB.
The chemistry and synthesis of diols 1a–c have not been
very well studied. However, they are readily accessible from
the reduction of the corresponding diketones. Diol 1a was
prepared in high yield, as previously described, by using Ru-
catalyzed transfer hydrogenation.[12,13] Diol 1b was prepared
from the corresponding diketone according to a procedure
developed by Enriquez-Garcia and Kꢀndig.[14] Diol 1c is
known,[15] but has not been explored to any great extent. We
chose a new route for the preparation of 1c involving a re-
duction (NaBH4/MeOH) of the corresponding diketone,
which in turn was prepared from dimethylphthalate (4;
Scheme 2).
Table 1. DYKAT of diols 1a–c.[a]
Entry
Diol 1
Time
[h]
Conv.
[%]
Yield[b]
[%]
8
[%]
ee
[%]
d.r.
1
2
1a
1a
1a
1b
1b
1c
1c
1c
24
48
96
48
96
60
120
168
94
>99
>99
98
98
>99
83
85
94
98
96
98
88
n.d.
91
8a (9)
8a (10)
8a (9)
8b (9)
8b (8)
8c (29)
8c (3)
8c (4)
n.d.
99.9
99.9
99.9
99.9
99.9
n.d.
99.9
97:3
97:3
95:5
97:3
97:3
99:1
98:2
98:2
3[c]
4
5[c]
6[d]
7[e]
8[e]
>99
[a] Unless otherwise stated the reactions were carried out under the fol-
lowing conditions: ruthenium catalyst 2 (5 mol%), tBuOK (5 mol%), iso-
propenyl acetate (4 equiv), Na2CO3 (0.2 equiv), CALB (12.5 mgmmol of
substrateÀ1, toluene (0.5m), 508C. [b] Yields of the isolated products, in-
cluding acetoxy ketone 8. [c] Toluene was the solvent (0.2m). [d] Condi-
tions: Na2CO3 (1 equiv), CALB (25 mgmmol of substrateÀ1), 0.1m, 808C.
[e] Conditions: Na2CO3 (1 equiv), CALB (25 mgmmol substrateÀ1),
0.13m, 508C.
Scheme 2. Synthesis of diol 1c.
The condensation of dimethylglutarate and 4 gave dicar-
boxylate 5 according to a known procedure,[16] and subse-
quent decarboxylation of diester 5 afforded diketone 6 in
83% yield.[17]
Diols 1a–c are all hygroscopic, so the diols were dried
over P2O5 under vacuum and then stored over P2O5 in a des-
iccator because the DKR and DYKAT reactions involving
Ru-catalyst 2 are water sensitive.
A kinetic asymmetric transformation (KAT) of a 1:1 cis/
trans mixture of symmetric diols 1 by using a highly selective
lipase and isopropenyl acetate for the transesterification
should result in diacetate, monoacetate, and diol products in
yields of 25, 50, and 25%, respectively, all of which are
Diol 1c is a more troublesome substrate than 1a,b. The
temperature of the reaction had to be increased to 808C to
obtain a reasonable reaction time. At this temperature, the
reaction is complete after 60 hours (Table 1, entry 6). The
drawback with such high temperatures is that a large
amount of 8c is formed (29%). The reaction can be run at
508C to avoid ketone formation, but in this case the reac-
tion has to be stirred for at least 7 days (Table 1, entry 8).
Under these conditions, 1c afforded the product (R,R)-7c in
high yield with excellent stereoselectivity (>99.9% ee, d.r.
>98:2). Furthermore, only 4% of ketone 8c was observed
in the crude product.
4032
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4031 – 4036

Enantiomeric Catalysis of Bicyclic Diols
FULL PAPER
With a working protocol in hand, we decided to scale up
the DYKAT reactions of 1a,b (Table 2). The loading of the
Ru-catalyst 2 and tBuOK was decreased from 5 to 1 mol%.
vent, whereas we found that toluene gave the best result.
Also, the loading of CALB could be decreased from 377 to
only 25 mgmmolÀ1.
This reaction was sensitive to the amount of iPrOH
added. The reaction was severely slowed down with
16 equivalents of iPrOH; therefore, a suitable amount was
8 equivalents with respect to the diacetate. By using this
method, pure (R,R)-1b was obtained in 76% overall yield
from racemic 1b in a two-step procedure (preparative scale)
with d.r.=99.9:0.1 and >99.9% ee. The enzymatic hydroly-
sis of diacetate 7a was less efficient, and the classical hydrol-
ysis reaction (NaOH, MeOH) was employed, thus affording
(R,R)-1a with d.r.=97:3 and >99.9% ee.
During our investigations, we found that oxidation of
diols 1a,b to hydroxy ketones 9a[19a] and 9b,[14,19] respective-
ly, through Ru-catalyzed hydrogen transfer with catalyst 3[20]
in acetone (transfer dehydrogenation)[21] is much faster than
the corresponding oxidation of the hydroxy ketones to dike-
tones 10a,b, respectively (Scheme 4). We decided to take
Table 2. DYKAT of diols 1a-b on a preparative scale.[a]
Entry
1
Time
[h]
Conv.
[%]
Yield[b]
[%]
8
[%]
ee
[%]
d.r.
1[c]
2[d]
3[d]
1a
1b
1b
67
72
96
>99
88
99
95
n.d.
94
8a (4)
8b (6)
8b (5)
99.9
n.d.
99.9
92:8
96:4
95:5
[a] The reactions were run under the following conditions: ruthenium cat-
alyst
2 (1 mol%), tBuOK (1 mol%), isopropenyl acetate (4 equiv),
Na2CO3 (0.2 equiv), CALB (12.5 mgmmol of substrateÀ1), toluene
(0.5m), 508C. [b] Yields of the isolated products, including acetoxy
ketone 8. [c] Scale=8 mmol. [d] Scale=10 mmol.
As a consequence of the decrease in catalyst loading, the re-
action time had to be extended to ensure complete conver-
sion. The yields in the scaled-up reactions of 1a,b (95 and
94%, respectively) were in the same range as in the case of
the small-scale reactions (see Table 1). In both cases the
d.r. value slightly decreased, probably due to slower epime-
rization as a consequence of the lower catalyst loading.
However, the decrease in catalyst loading also gave a lower
amount of acetoxy ketone 8.
Hydrolysis of diacetates 7a,b (Scheme 3) can be per-
formed in a solution of NaOH in methanol (1m) to afford
enantiopure and diastereoenriched diols (R,R)-1a and
Scheme 4. Mono-oxidation of diol (R,R)-1a,b catalyzed by 3.
advantage of the difference in rate between the two oxida-
tion steps for the preparation of enantioenriched hydroxy
ketones (R)-9a and (R)-9b. Complex 3 is also known to cat-
alyze the racemization of sec-alcohols. However, we argued
that because acetone is in large excess there should be little
or no racemization.
Scheme 3. Ester cleavage of diacetates 7a,b.
(R,R)-1b. Because it is difficult to remove the cis-diacetate,
present in a small percentage, from the product, the diol will
have the same d.r. value as the diacetate. In an effort to im-
prove the d.r. value of diols 1a,b, we carried out the ester
cleavage with CALB. In this case, only the acetate group
with the R configuration will be cleaved, and because the
cis-diacetate has an R,S configuration it will only be trans-
formed into a monoacetate by the enzyme and is easily sep-
arated from the diol product. In this way, the d.r. value
should be dramatically increased. To our surprise, the
CALB-catalyzed hydrolysis of 7b was more difficult than
we had expected. Very low conversions were achieved in
phosphate buffers of pH 7.0–7.9 under different concentra-
tions and at different temperatures. The addition of organic
solvents such as acetone and methanol to improve the solu-
bility of 7b did not improve the reaction. Instead, a method
developed by Kim and co-workers[18] was employed in which
CALB is used to catalyze the transesterification of 7b in an
organic solvent. Kim and co-workers[18] used THF as the sol-
The oxidation of diols (R,R)-1a and (R,R)-1b was carried
out at 358C. At higher temperatures (i.e., ꢀ508C), the
mono-oxidation is fast (i.e., ~1 h); therefore, it is difficult to
stop the reaction without overoxidation to a diketone. Oxi-
dation of enantio- and diastereomerically pure diol (R,R)-1a
(99.9% ee, >99% trans)[22] gave the corresponding hydroxy
ketone (R)-9a in 85% yield with 98% ee after 20 h at 358C.
This outcome indicates that only a small amount of racemi-
zation takes place in the mono-oxidation of (R,R)-1a, which
makes it an attractive method to obtain enantioenriched
(R)-9a. Compound (R)-9a should be a potential precursor
for the synthesis of indatralin.[23]
In the mono-oxidation of diol (R,R)-1b, which starts from
99.9% ee and d.r.=99.9:0.1, the reaction yields the maxi-
mum amount of hydroxy ketone (R)-9b after around
1
7 hours (86% according to H NMR spectroscopic analysis).
Compound (R)-9b was isolated by chromatography on silica
gel in 82% yield with 97% ee. Hydroxy ketone (R)-9b
should be a useful starting material for the synthesis of ser-
Chem. Eur. J. 2010, 16, 4031 – 4036
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4033

J.-E. Bꢂckvall et al.
tralin,[1] as also pointed out by Enriquez-Garcia and
Kꢀndig.[14] The reaction profiles of the oxidation of (R,R)-
1a and (R,R)-1b are given in Figures 1 and 2, respectively.
1
These data were recorded with H NMR spectroscopy; the
concentration in the experiments was 0.1m in [D6]acetone
for solubility reasons. Hydroxy ketone (R)-9b has been pre-
pared before,[14,19] although the present method gives a
higher ee value than previous methods and compares very
well with the best of those approaches.[14]
Scheme 5. Formal total synthesis of sertraline. DCM=dichloromethane,
DMAP=4-dimethylaminopyridine, MsCl=methanesulfonyl chloride,
TBDPS=tert-butyldiphenylsilyl.
major isomer. The isolated yield of 13 was 85%. Due to sep-
aration problems of 11 and the major diastereomer of 13,
the mixture of 11 and 13 was used in the consecutive elimi-
nation step to give 14. It has been observed that silyl ether
14 is labile,[24a] which was also noted during our investiga-
tions. All attempts to use acidic dehydration failed,[26] thus
leading to an aromatized product, presumably through desi-
lylation and elimination of H2O. Therefore, alcohol 13 was
transformed into its mesylate, which undergoes an elimina-
tion reaction in situ at room temperature if excess triethyla-
mine is added. This method gave 14 in 84% yield (97%
yield based on 86% conversion) with 97% ee. After the
elimination reaction, 11 can be recycled after separation
from 14 by using column chromatography.
^
Figure 1. Reaction profile for the oxidation of diol (R)-1a. : 1a; &: 9a;
~: 10a. Conditions: catalyst 2 (0.02 equiv), [D6]acetone (0.1m), 358C.
^
Figure 2. Reaction profile for the oxidation of diol (R)-1b. : 1b; &: 9b;
~: 10b. Conditions: Catalyst 2 (0.02 equiv), [D6]acetone (0.1m), 358C.
Conclusions
In summary, we have developed a DYKAT protocol for
diols 1a–c that give high enantio- and diastereoselectivities.
These procedures could also be used on 8- and 10-mmol
scales for diols 1a,b, respectively. We have also demonstrat-
ed that the hydrolysis of (R,R)-7b catalyzed by CALB gave
enantio- and diastereomerically pure diols. The optically
pure diols (R,R)-1a and (R,R)-1b were transformed into hy-
droxy ketones (R)-9a,b with high selectivity and in high
yield. This mono-oxidation was catalyzed by 3 with only a
small amount of racemization. Finally, (R)-9b was used in a
formal total synthesis of sertraline without loss of enantio-
meric excess.
The enantiomerically pure bicyclic diols (R,R)-1, their
acetate derivatives (R,R)-7, and, in particular, chiral hydroxy
ketones (R)-9 should be useful synthetic intermediates for
the preparation of various biologically active compounds.
Compound 14 is an important precursor for the synthesis of
sertraline and has been transformed into sertraline in a few
steps.[24,25] We now demonstrate that (R)-9b can be trans-
formed into 14 in three steps in high yield (Scheme 5), thus
constituting a formal total synthesis of sertraline.
The alcohol group in (R)-9b was protected with a TBDPS
group in high yield (97%) followed by a Grignard addition
to the ketone. The competing reaction to the addition reac-
tion is deprotonation at the a-position, which gives back the
ketone in the workup. However, the selectivity is 85:15, thus
favoring addition to the ketone. The remaining ketone 11 is
also easily recycled in the final elimination step. The addi-
tion product 13 is a diastereomeric mixture of 4:1, presuma-
bly with the aryl group trans to the silyl ether moiety as the
Experimental Section
Procedure for the DYKAT of diols 1a,b: Base tBuOK (0.5m, 0.05 mmol)
in THF (100 mL) was added to a flame-dried Schlenk flask under argon.
THF was removed under vacuum and the flask refilled with argon. Ru
4034
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4031 – 4036

Enantiomeric Catalysis of Bicyclic Diols
FULL PAPER
catalyst 2 (32 mg, 0.05 mmol) and toluene (2 mL) were added to the reac-
tion mixture, which was stirred for 3 min at 508C before the diol
(1 mmol) was added. After the mixture had been stirred at 508C for a
further 30 min, CALB (12.5 mg), Na2CO3 (21 mg, 0.2 mmol), and isopro-
penyl acetate (440 mL, 4 mmol) were charged to the flask. The reaction
mixture was stirred at 508C until no more diol or monoacetate remained
(the conversion was checked by 1H NMR spectroscopic analysis with
CD3OD as the solvent). The reaction mixture was allowed to reach ambi-
ent temperature before filtration through a short plug of silica gel (elu-
ated with EtOAc) to remove the inorganic material and the enzyme. The
solvent was evaporated and the crude product was purified by chroma-
tography on silica gel (eluent: CH2Cl2). (For yields see Table 1): An ana-
lytical sample of (R,R)-7a was obtained by refined chromatography.
(R,R)-7a: [a]2D0 =+131.4 (c=1.1, EtOAc; ꢀ99.9% ee, 99% d.r.);
1H NMR (400 MHz, CD3OD): d=7.44–7.33 (4H, m), 6.28 (2H, app t, J=
5.4 Hz), 2.47 (2H, app t, J=5.4 Hz), 2.04 ppm (6H, s); 13C NMR
(100 MHz, CD3OD): d=172.7, 143.0, 130.5, 126.6, 77.4, 40.9, 21.0 ppm;
the analytical data for 7b was in accordance with those previously report-
ed.[27]
cording to method D (see the Supporting Information). [a] 2D0 =+ 42.5
(c=0.8, EtOAc; 98% ee); H NMR (400 MHz, CDCl3): d=8.00 (1H, dd,
1
J=7.7, 1.5 Hz), 7.76–7.72 (2H, m), 7.62–7.58 (2H, m), 7.50–7.32 (8H, m),
7.29–7.24 (1H, m), 4.96 (1H, dd, J=7.2, 3.5 Hz), 2.97 (1H, ddd, J=17.6,
8.3, 4.7 Hz), 2.45 (1H, ddd, J=17.6, 8.1, 4.8 Hz), 2.23–2.13 (1H, m), 2.12–
2.03 (1H, m), 1.08 ppm (9H, s); 13C NMR (100 MHz, CDCl3): d=198.0,
145.6, 136.0, 136.0, 134.0, 133.7, 133.4, 131.4, 130.1, 130.0, 128.1, 127.9,
127.8, 127.6, 127.1, 69.6, 34.9, 32.2, 27.1, 19.6 ppm.
4-(tert-Butyldiphenylsilyloxy)-1-(3,4-dichlorophenyl)-1,2,3,4-tetrahydro-
naphthalen-1-ol (13): A solution of 12 (0.4m) was prepared by slowly
adding a solution of 3,4-dichloroiodobenzene (1.1 g, 4 mmol) in dry Et2O
(10 mL) to Mg turnings (144 mg, 6 mmol), which was stirred for 2 h at
room temperature under argon. The Grignard solution of 12 (3.3 mL,
1.32 mmol) was charged into a solution of 11 in Et2O (0.2m, 241 mg,
0.6 mmol). The reaction mixture was stirred for 17 h at room temperature
before quenching the reaction with saturated aqueous NH4Cl. The aque-
ous layer was acidified with HCl (1m) to pH~1 and extracted with Et2O
(3ꢄ20 mL). The combined organic layers were washed with saturated
aqueous NaHCO3 and dried over MgSO4. Evaporation of the solvents
yielded a yellow sticky solid (348 mg). The crude mixture contained 11
and product 13 in a ratio of 15:85. Compound 13 was obtained as a dia-
stereomeric mixture of 4:1. Chromatography on silica gel (pentane/ethyl
acetate 96:4) easily separated 11 from the minor diastereomer, but it was
difficult to fully separate the major diastereomer from 11. For practical
purposes, the 15:85 mixture of 11 and 13 was employed in the subsequent
reaction. The combined fractions from the chromatography on silica gel
yielded a colourless sticky solid (313 mg). Based on NMR spectroscopic
analysis, the amount of 13 was estimated to be 277 mg (85%). For de-
tailed analysis, a small amount of product was purified by chromatogra-
phy on silica gel (pentane/ethyl acetate 96:4). Major diastereomer of 13:
1H NMR (400 MHz, CDCl3): d=7.78 (4H, m), 7.58–7.54 (1H, m), 7.50–
7.36 (7H, m), 7.30 (1H, d, J=8.3 Hz), 7.29 (1H, app td, J=7.6, 1.5 Hz),
7.20 (1H, app td, J=3.8, 1.4 Hz), 7.02 (1H, dd, J=8.4, 2.2 Hz), 6.96 (1H,
dd, J=7.8, 1.3 Hz), 4.90 (1H, dd, J=8.3, 4.8 Hz), 2.38 (1H, bs), 2.14–2.01
(1H, m), 2.19 (1H, ddd, J=13.8, 6.9, 3.0 Hz), 1.88 (1H, ddd, J=13.8,
10.6, 3.2 Hz), 1.83–1.73 (1H, m), 1.13 ppm (9H, s); 13C NMR (100 MHz,
CDCl3): d=148.8, 140.8, 140.5, 136.1, 134.3, 133.6, 132.1, 130.8, 130.0,
129.9, 129.8, 128.6, 128.6, 128.4, 128.3, 127.9, 127.8, 127.8, 126.1, 74.7,
70.7, 38.2, 29.5, 27.3, 19.6 ppm. Minor diastereomer of 13: 1H NMR
(400 MHz, CDCl3): d=7.75–7.70 (2H, m), 7.65 (1H, d, J=2.2 Hz), 7.63–
7.59 (2H, m), 7.49–7.33 (7H, m), 7.22 (1H, dd, J=8.4, 2.2 Hz), 7.19 (1H,
app td, J=3.8, 1.5 Hz), 7.11 (1H, app td, J=3.7, 1.5 Hz), 6.94 (2H, ddd,
J=9.4, 7.7, 1.5 Hz), 4.84 (1H, app t, J=3.7 Hz), 2.67 (1H, ddd, J=14.2,
12.3, 3.0 Hz), 2.06–1.83 (4H, m), 1.06 ppm (9H, s); 13C NMR (125 MHz,
CDCl3): d=149.8, 141.0, 138.6, 136.2, 136.1, 134.1, 134.0, 132.2, 130.8,
130.0, 129.9, 129.9, 129.7, 128.9, 128.8, 128.6, 128.2, 127.9, 127.7, 126.1,
74.7, 69.2, 36.3, 28.6, 27.1, 19.6 ppm.
Procedure for the DYKAT of diol 1c: Base tBuOK (0.5m, 0.02 mmol) in
THF (40 mL) was added to a flame-dried Schlenk flask under argon.
THF was removed under vacuum and the flask refilled with argon. Ru
catalyst 2 (13 mg, 0.02 mmol) and toluene (3 mL) were added to the reac-
tion mixture, which was stirred for 3 min at 508C before the diol
(0.4 mmol) was added. After the mixture had been stirred at 508C for a
further 30 min, CALB (10 mg), Na2CO3 (42 mg, 0.4 mmol), and isopro-
penyl acetate (176 mL, 1.6 mmol) were charged into the flask. The reac-
tion mixture was stirred at 508C for 7 days. The reaction mixture was al-
lowed to reach ambient temperature before filtration through a short
plug of silica gel (eluated with EtOAc) to remove the inorganic material
and the enzyme. The solvent was evaporated and the crude product was
purified by chromatography on silica gel (eluent: CH2Cl2). (For yields see
Table 1): An analytical sample of (R,R)-7c was obtained by refined chro-
matography. (R,R)-7c. [a]2D0 =+100.6 (c=1.1, EtOAc; ꢀ99.9% ee,
>99.9% d.r.); 1H NMR (400 MHz, CDCl3): d=7.38–7.31 (m, 2H), 7.29–
7.22 (2H, m), 6.19–6.09 (2H, m), 2–16 (6H, s), 2.07–1.94 (4H, m), 1.94–
1.81 ppm (2H, m); 13C NMR (100 MHz, CDCl3): d=170.0, 138.5, 127.8,
127.0, 75.3, 32.8, 22.0, 21.3 ppm.
ACHTUNGTRENNUNG(1R,4R)-1,2,3,4-Tetrahydronaphthalene-1,4-diol ((R,R)-1b): Compound
(R,R)-7b from the large-scale DYKAT reaction of 1a (1.0 g, 4 mmol;
99.9% ee, dr=95:5; 5% ketone 8b) and CALB (100 mg) were suspended
in dry toluene (16 mL) in an argon atmosphere. iPrOH (HPLC grade
from a freshly opened bottle; 2.45 mL, 32 mmol) was added to the reac-
tion mixture, which was stirred 3.5 days at 508C. The solvent was evapo-
rated, the residue was added to MeOH to make a slurry, and the solids
were removed by filtration. The product was evaporated onto silica gel
and purified by chromatography (CH2Cl2/EtOAc 1:1) to yield a white
solid (535 mg, 81%, 99.9% ee, d.r.=99.9:0.1). The ee value was deter-
mined according to method B (see the Supporting Information) and the
(R)-tert-Butyl(4-(3,4-dichlorophenyl)-1,2-dihydronaphthalen-1-yloxy)di-
phenylsilane (14): Compound 13 (65 mg mixture of 13 and 11 (85:15);
0.1 mmol with respect to 13), DMAP (0.6 mg, 0.005 mmol), and Et3N
(84 mL, 0.6 mmol) were dissolved in CH2Cl2 (1 mL) at ambient tempera-
ture. MsCl (23 mL, 0.3 mmol) was added to the reaction mixture, which
was stirred for 4 h. The reaction mixture was put directly onto a prepre-
pared column of silica gel and chromatography (pentane/EtOAc 99:1),
which yielded the pure product 14 (44.5 mg, 84%, 97% ee; see method E
in the Supporting Information). The NMR spectroscopic data were slight-
ly different from those reported previously.[24] The major differences in
1
d.r. value was measured by using H NMR spectroscopic analysis.
(R)-4-Hydroxy-3,4-dihydronaphthalen-1ACTHNUTRGNE(NUG 2H)-one ((R)-9b): Compound
(R,R)-1b (246 mg, 1 mmol, ꢀ99.9% ee, d.r.=ꢀ99.9:0.1) and the Shvo
catalyst 3 (32 mg, 0.02 mmol) were suspended in acetone (PA grade;
7.5 mL) in an atmosphere of air and stirred at 358C for 7 h. The solvent
was removed under vacuum (Tꢁ358C). Purification by chromatography
on silica gel (CH2Cl2/EtOAc 4:1) yielded the desired product (201 mg,
82%, 97% ee). The analytical data were in accordance with previously
reported data.[14,19] The ee value was determined according to method B
(see the Supporting Information).
1
the H NMR spectra were that the reported shift at d=7.38 (1H, m) was
not found by us.[28] The control experiments given below confirmed that
we have the correct product 14. First, desilylation of 14 with the same de-
silylation protocol as described in ref. [24] yielded the alcohol (compound
10 in ref. [24]), which had spectral data identical to those reported in
ref. [24]. We also quenched the Grignard reagent 12 (prepared in the syn-
thesis of 13) with CO2 to yield 3,4-dichlorobenzoic acid to ensure that the
Grignard position or any of the chlorine atoms did move on the ring.
1H NMR (400 MHz, [D6]DMSO) of 3,4-dichlorobenzoic acid: d=8.06
(1H, d, J=2.0 Hz), 7.88 (1H, dd, J=8.3, 2.0 Hz), 7.78 ppm (1H, d, J=
8.3 Hz). The coupling pattern of 3,4-dichlorobenzoic acid is difficult to
find in 14 if the 1H NMR spectra is recorded in C6D6; however, the pat-
(R)-4-(tert-Butyldiphenylsilyloxy)-3,4-dihydronaphthalen-1ACTHUNTGRENNG(U 2H)-one
(11): Compound (R)-9b (122 mg, 0.75 mmol), imidazole (128 mg,
1.88 mmol), and DMAP (6.5 mg, 0.05 mmol) were dissolved in CH2Cl2
(3.75 mL, 0.2m with respect to (R)-9b). t-Butyldiphenylchlorosilane
(212 mL, 0.83 mmol) was added to the reaction mixture, which was stirred
2 h at ambient temperature. The reaction mixture was diluted with
CH2Cl2, silica gel was added, and the solvent was removed under
vacuum. Chromatography on silica gel (pentane/ethyl acetate 98:2) af-
forded product (293 mg, 97%, 98% ee). The ee value was measured ac-
Chem. Eur. J. 2010, 16, 4031 – 4036
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4035

J.-E. Bꢂckvall et al.
tern can be found if CDCl3 is used instead (see the attached COSY spec-
tra: d=7.43 (1H, d, J=8.3 Hz), 7.43 (1H, d, J=2.0 Hz), 7.16 ppm (1H,
dd, J=8.3, 2.0 Hz)). Additionally, the 13C NMR spectrum is easier inter-
preted if it is recorded in CDCl3 due to the lack of overlap in the aromat-
ic region (relative to C6D6). Therefore, we chose to report our spectra re-
corded in CDCl3, although the spectra recorded in C6D6 are attached for
comparison purposes. 1H NMR (400 MHz, CDCl3): d=7.77 (2H, m),
7.70–7.63 (2H, m), 7.50–7.33 (7H, m), 7.43 (1H, d, J=8.3 Hz), 7.43 (1H,
d, J=2.0 Hz), 7.25 (1H, app td, J=7.5, 1.3 Hz), 7.20 (1H, app td, J=7.5,
1.6 Hz), 7.16 (1H, dd, J=8.3, 2.0 Hz), 6.96 (1H, dd, J=7.5, 1.3 Hz), 5.86
(1H, dd, J=5.3, 4.0 Hz), 4.99 (1H, dd, J=9.9, 5.5 Hz), 2.49 (1H, ddd, J=
16.5, 9.9, 4.0 Hz), 2.26 (1H, app dt, J=16.5, Hz), 1.11 ppm (9H, s);
13C NMR (100 MHz, CDCl3): d=140.5, 138.9, 137.8, 136.1, 135.9, 134.6,
133.7, 133.5, 132.5, 131.3, 130.6, 130.4, 129.9, 129.9, 128.2, 127.9, 127.8,
127.8, 127.5, 125.9, 125.8, 125.3, 69.8, 33.0, 27.2, 19.6 ppm.
Acad. Sci. USA 2004, 101, 5761–5766; c) B. Martꢆn-Matute, M.
Edin, J. E. Bꢂckvall, Chem. Eur. J. 2006, 12, 6053–6061; d) K. Lei-
jondahl, L. Borꢇn, R. Braun, J.-E. Bꢂckvall, Org. Lett. 2008, 10,
2027–2030; e) K. Leijondahl, L. Borꢇn, R. Braun, J.-E. Bꢂckvall, J.
Org. Chem. 2009, 74, 1988–1993.
[10] A. B. L. Fransson, Y. Xu, K. Leijondahl, J.-E. Bꢂckvall, J. Org.
Chem. 2006, 71, 6309–6316.
[11] G. Csjernyik, K. Bogꢁr, J.-E. Bꢂckvall, Tetrahedron Lett. 2004, 45,
6799–6802.
[12] K. Leijondahl, A. B. Fransson, J.-E. Bꢂckvall, J. Org. Chem. 2006,
71, 8622–8625.
[13] The reduction of 1,3-diketones with NaBH4 generally suffer from
moderate yields of diol, and therefore the Ru-catalyzed method in
ref. [12] was developed; however, diol 1a can be prepared in 70–
80% yield (this work) through reduction with NaBH4 in MeOH rel-
ative to 93% yield by using the Ru 3-catalyzed method.
[14] A. Enriquez-Garcia, E. P. Kꢀndig, Beilstein J. Org. Chem. 2008, 4,
No. 37.
[15] a) G. L. Buchanan, J. M. McCrae, Tetrahedron 1967, 23, 279–282;
b) G. L. Buchanan, J. Chem. Soc. 1954, 1060–1063.
Acknowledgements
[16] B. Danieli, G. Lesma, D. Passarella, A. Silvani, Synth. Commun.
1997, 27, 69–77.
[17] A modified procedure was used from W. E. Hahn, B. Kryczka, Pol.
J. Chem. 1979, 53, 1751–1764.
[18] Y. K. Choi, J. H. Suh, D. Lee, I. T. Lim, J. Y. Jung, M. J. Kim, J. Org.
Chem. 1999, 64, 8423–8424.
The Swedish Research Council, the Berzelius Center EXSELENT, the
European FP7 network INTENANT (“Integrated synthesis and purifica-
tion of single enantiomers”), and the K&A Wallenberg foundation are
gratefully acknowledged for financial support.
[19] a) S. Joly, M. S. Nair, Tetrahedron: Asymmetry 2001, 12, 2283–2287;
b) E. M. Ferreira, B. M. Stoltz, J. Am. Chem. Soc. 2001, 123, 7725–
7726.
[20] For mechanistic studies on hydrogen transfer with 3, see: a) C. P.
Casey, S. W. Singer, D. R. Powell, R. K. Hayashi, M. Kavana, J. Am.
Chem. Soc. 2001, 123, 1090–1100; b) J. S. M. Samec, J. E. Bꢂckvall,
P. G. Andersson, P. Brandt, Chem. Soc. Rev. 2006, 35, 237–248.
[21] M. L. S. Almeida, M. Beller, G-Z. Wang, J.-E. Bꢂckvall, Chem. Eur.
J. 1996, 2, 1533–1536.
[22] A sample was purified by column chromatography to give enantio-
and diastereomically pure (R)-1a.
[23] K. P. Bøgesø, A. V. Christensen, J. Hyttel, T. Liljefors, J. Med.
Chem. 1985, 28, 1817.
[24] a) M. Lautens, T. Rovis, J. Org. Chem. 1997, 62, 5246–5247; b) M.
Lautens, T. Rovis, Tetrahedron 1999, 55, 8967–8976.
[25] Sertralin is a serotonin-uptake inhibitor used for the treatment of
depression; see: W. M. Welch, A. R. Kraska, R. Sarges, K. B. Coe, J.
Med. Chem. 1984, 27, 1508–1515.
[26] Catalytic para-toluenesulfonic acid (PTSA) in toluene at 1008C and
ambient temperature gave the aromatized product; no reaction at
all was seen on changing to acetic acid at ambient temperature,
whereas the only product that could be observed at 508C was the ar-
omatized elimination product.
[27] S. Yamada, H. Katsumata, J. Org. Chem. 1999, 64, 9365–9373.
[28] Prof. Lautens has confirmed that the peak at d=7.38 ppm does not
belong to the compound but to an impurity that may have arisen
from some decomposition; a 1H NMR spectrum in C6D6 that lacks
the peak at d=7.38 ppm was reported in the thesis by Tom Rovis
(T. Rovis, Dissertation, University of Toronto, 1998); our 1H NMR
spectrum of 14 in C6D6 is in agreement with that spectrum.
[1] Z. Han, S. G. Koenig, H. Zhao, X. Su, S. P. Singh, R. Bakale, Org.
Process Res. Dev. 2007, 11, 726–730.
[2] E. Quesada, M. Stockley, J. P. Ragot, M. E. Prime, A. C. Whitwood,
R. J. K. Taylor, Org. Biomol. Chem. 2004, 2, 2483–2495.
[3] E. P. Kꢀndig, P. D. Chaudhuri, D. House, G. Bernardinelli, Angew.
Chem. 2006, 118, 104–107; Angew. Chem. Int. Ed. 2006, 45, 98–101.
[4] a) F. F. Huerta, A. B. E. Minidis, J.-E. Bꢂckvall, Chem. Soc. Rev.
2001, 30, 321–331; b) M.-J. Kim, Y. Ahn, J. Park, Curr. Opin. Bio-
technol. 2002, 13, 578–587; c) O. Pꢅmies, J.-E. Bꢂckvall, Chem. Rev.
2003, 103, 3247–3262; d) B. Martꢆn-Matute, J.-E. Bꢂckvall, Curr.
Opin. Chem. Biol. 2007, 11, 226–232.
[5] a) B. Martꢆn-Matute, M. Edin, K. Bogꢁr, F. B. Kaynak, J.-E. Bꢂck-
vall, J. Am. Chem. Soc. 2005, 127, 8817–8825; b) J. H. Choi, Y. K.
Choi, Y. H. Kim, E. S. Park, E. J. Kim, M. J. Kim, J. J. Park, J. Org.
Chem. 2004, 69, 1972–1977; c) A. Trꢂff, K. Bogꢁr, M. Warner, J. E.
Bꢂckvall Org. Lett. 2008, 10, 4807–4810; d) for a recent example of
chemoenzymatic DKR of chloroalcohols that leads to chiral epox-
ides, see: R. M. Haak, F. Berthiol, T. Jerphagnon, A. J. A. Gayet, C.
Tarabiono, C. P. Postema, V. Ritleng, M. Pfeffer, D. B. Janssen, A. J.
Minnaard, B. L. Feringa, J. G. de Vries, J. Am. Chem. Soc. 2008, 130,
13508–13509.
[6] a) M. T. Reetz, K. Schimossek, Chimia 1996, 50, 668–669; b) J. Paet-
zold, J.-E. Bꢂckvall, J. Am. Chem. Soc. 2005, 127, 17620–17621;
c) L. K. Thalꢇn, D. Zhao, J. B. Sortais, J. Paetzold, C. Hoben, J.-E.
Bꢂckvall, Chem. Eur. J. 2009, 15, 3403–3410; d) M. J. Kim, W. H.
Kim, K. Han, Y. K. Choi, J. Park, Org. Lett. 2007, 9, 1157–1159;
e) A. N. Parvulescu, P. A. Jacobs, D. E. De Vos, Chem. Eur. J. 2007,
13, 2034–2043.
[7] D. Strꢀbing, P. Krumlinde, J. Piera, J.-E. Bꢂckvall, Adv. Synth. Catal.
2007, 349, 1577–1581.
[8] The concept of DYKAT was recently reviewed; see: J. Steinreiber,
K. Faber, H. Griengl, Chem. Eur. J. 2008, 14, 8060–8072.
[9] a) B. A. Persson, F. F. Huerta, J.-E. Bꢂckvall, J. Org. Chem. 1999, 64,
5237–5240; b) M. Edin, J. Steinreiber, J.-E. Bꢂckvall, Proc. Natl.
Received: November 13, 2009
Published online: March 1, 2010
4036
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4031 – 4036