Dynamic kinetic resolution of 2,3-dihydrobenzo[be]furans: Chemoenzymatic synthesis of analgesic agent BRL 37959

Article abstract of DOI:10.1002/chem.201200683

An efficient asymmetric synthesis of (S)-2,3-dihydrobenzo[be]furan-3- carboxylic acid (8 ) and (S)-5-chloro-2,3-dihydrobenzo[be]furan-3-carboxylic acid (8a b) was established. Key to the success was the highly stereoselective enzymatic kinetic resolution

Full text of DOI:10.1002/chem.201200683

FULL PAPER
DOI: 10.1002/chem.201200683
Dynamic Kinetic Resolution of 2,3-Dihydrobenzo[b]furans:
Chemoenzymatic Synthesis of Analgesic Agent BRL 37959
Patrick Bongen, Jçrg Pietruszka,* and Robert Christian Simon[a]
Abstract: An efficient asymmetric syn-
esters that was further developed into
a dynamic process. As a reliable and
fast tool for analysing the enantiomeric
excess, HPLC coupled with a CD de-
thesis of (S)-2,3-dihydrobenzo[b]furan-
3-carboxylic acid (8a) and (S)-5-
chloro-2,3-dihydrobenzo[b]furan-3-car-
boxylic acid (8b) was established. Key
to the success was the highly stereose-
lective enzymatic kinetic resolution of
the corresponding methyl or ethyl
tector was utilized. The route was com-
pleted by a Friedel–Crafts acylation of
ethyl (S)-5-chloro-2,3-dihydrobenzo[b]-
furan-3-carboxylate (7c) followed by
saponification leading to (S)-5-chloro-
2,3-dihydrobenzo[b]furan-3-carboxylic
acid (2), an analgesic agent.
Keywords: chiral resolution · enan-
tioselectivity · enzymes · hydrolases ·
dynamic kinetic resolution
Introduction
metal-based chemocatalysts, additional biocatalysts, reduc-
tion and oxidation processes, bases/acids, or ion-exchange
resins.[7,8] Of course, certain issues are required to reach suf-
ficient levels of efficiency in DKR: The resolution process
needs to be highly selective (E>50)[9] and the racemization
constant (krac) must be higher than the resolution constant
(kA). Moreover, the final reaction must be irreversible and
the product should be stable under certain reaction condi-
tions and not undergo (unwanted) side reactions.[2,6,7]
Racemisation of esters can be especially challenging. Ad-
ditional activation is often essential to increase the acidity
of the substrate, for example, the use of thioesters instead of
the required methyl ester was often found to be benefi-
cial.[10] However, additional reaction steps are required to
obtain the desired enantiomerically pure target compound.
Although successful applications have been reported for un-
modified esters, for example, for the synthesis of ibupro-
fen,[11] the rate for the overall transformation was usually
low and hence prolonged reaction times were required.
Herein, we present a detailed study on the synthesis of opti-
cally pure 3-substituted 2,3-dihydrobenzo[b]furan through
enzymatic KR and enzymatic DKR that overcomes these re-
strictions.
The 2,3-dihydrobenzo[b]furan structural motif has been
identified in an increasing number of natural products and
used in (potential) pharmacological drugs exhibiting a broad
range of biological activities (Figure 1). For example, imida-
zole derivative 1 serves as a novel gamma-secretase modula-
tor developed by Pfizer in 2011. It relates to the treatment
of Alzheimerꢀs disease and other neurodegenerative disor-
ders, and has IC50 values in the sub-micromolar range.[12]
The racemic form of 5-chloro-substituted 2,3-dihydro-ben-
zo[b]furan 2 (BRL 37959) combines potent analgesic activity
with low gastric irritancy, likewise at low concentrations.[13]
Novel prostaglandin (PG) D2 receptor antagonists, like
compound 3, were also synthesized and biologically evaluat-
ed recently.[14]
Catalytic processes are of major significance in current (or-
ganic) research because they enable production of the de-
sired molecules in a most economical and environmental
benign way. Nevertheless, the demand to create more effi-
cient routes to fulfil the concept of sustainability increases
steadily.[1] Biocatalysis, as such,[2] meets those requirements
and has proven to be remarkably capable, especially in the
pharmaceutical and chemical industries.[3] A vast number of
different techniques have been elaborated to provide opti-
cally pure products, but kinetic resolution (KR)[4] still repre-
sents the most frequently used and powerful method devel-
oped so far.[5] Although the theoretical maximum overall
yield is limited to only 50%, this method holds the advant-
age of open access to both stereoisomers, which can be im-
portant, for example, for enantiocomplementary biological
evaluation studies. Nevertheless, because only one enantio-
mer is required in most synthetic applications, highly sophis-
ticated methods have been established in which a racemisa-
tion of the undesired enantiomer is coupled with a chemical
or enzymatic kinetic resolution process. The result, a dynam-
ic kinetic resolution (DKR), allows the 50% barrier to be
exceeded and can approach a theoretical yield of 100%.[2,6]
Popular racemisation strategies include the use of transition-
[a] Dipl.-Chem. P. Bongen, Prof. Dr. J. Pietruszka, Dr. R. C. Simon
Institut fꢁr Bioorganische Chemie
der Heinrich-Heine Universitꢂt Dꢁsseldorf
im Forschungszentrum Jꢁlich
Stetternicher Forst, Geb. 15.8
52426 Jꢁlich (Germany)
Fax : (+49)2461-616196
E-mail: j.pietruszka@fz-juelich.de
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201200683.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

With the substrates in hand, we focused on the forthcom-
ing biotransformations of esters 7. Whereas in the case of
methyl esters 7a and 7b, Candida antarctica Lipase B
(CAL-B) and enantiocomplementary Bacillus subtilis ester-
ase (BS3) were thought to be appropriate; both enzymes
have already been shown to be superior in the enzymatic
(D)KR of structurally related methyl 2,3-dihydro-1H-
indene-3-carboxylate and methyl indoline-3-carboxylate in
our previous studies[20] (see also the Supporting Informa-
tion). To identify further potential hydrolytic enzymes that
could be used with ethyl ester 7c, a HPLC-CD selectivity
assay, developed by Reetz and co-workers, was used.[21] The
approach is based on HPLC analysis with an achiral station-
ary phase and a CD-detector, which allows direct determi-
nation of the enantiomeric composition (by using a circular
dichroism CD-detector [De]) of a defined sample at a fixed
wavelength. Interestingly, although this ought to be a time-
saving evaluation method for enzymes in general, it has only
been recently applied in an ADH-screening for the first
time after the initial report.[22] The main application still in-
volves determination of the absolute configuration of com-
pounds with unknown chirality.[23]
We examined the use of 22 hydrolases (for tested enzymes
see the Supporting Information) in the reaction and the ee
of the substrate was monitored by using HPLC-CD (for de-
tails on the method see the Supporting Information) to de-
termine the enantioselectivity. As a result CAL-B and BS3
could be detected as stereoselective hydrolases (Table 1),
with the former showing (S)-selectivity against the substrate,
and the latter showing (R)-selectivity. These results were
used for the following kinetic resolution studies.
Figure 1. Bioactive substances containing the 2,3-dihydrobenzo[b]furan
ring structure.
Numerous methods have been reported for the synthesis
of 2,3-dihydrobenzo[b]furans, including dehydrative meth-
ods, cycloadditions, radical-, anionic- and electrochemical
cyclisation or transition-metal-catalysed processes,[15] but ste-
reoselective methods are rare. Although asymmetric catalyt-
ic hydrogenation of the corresponding benzofuran seems the
most obvious route, to the best of our knowledge, only
a few examples are known to date.[16] The reason for this
paucity of successful outcomes is due to partial cleavage of
the furan ring leading to the formation of 2-ethylcyclohexa-
nol and other alcohols.[17]
Results and Discussion
Substrate synthesis for the current study was straightforward
(Scheme 1). According to a method described by Hossain
and co-workers,[18] salicylaldehyde (4a) and its chloro-deriv-
ative 4b were activated with HBF4·Et2O and treated with
ethyl diazoacetate to give intermediates 5a and 5b after aryl
migration; the semi-acetals were not isolated but directly
converted. Subsequent water elimination upon treatment
with concentrated H2SO4 afforded the corresponding benzo-
furans 6a and 6b in 90 and 75% yield, respectively (two
steps). Reduction of the double bond under simultaneous
transesterification was achieved with magnesium in
MeOH[19] through a single electron transfer (SET) process.
The resulting 2,3-dihydrobenzo[b]furans 7a and 7b were ob-
tained in very good yield (90%). Because Boyle et al. re-
ported a higher yield of the ethyl ester in comparison to the
methyl ester 7b in the late stage Friedel–Crafts aryolation
(FCA) of analgesic agent BRL 37959 (2),[13] compound 7c
was also prepared by a common transesterification.
The kinetic resolution of esters 7a–c were finally per-
formed on a preparative scale [rac-7a: 2.00 g (11.2 mmol);
rac-7b: 2.00 g (9.4 mmol); rac-7c: 1.00 g (4.4 mmol)] using
an immobilised form of the enzyme CAL-B (Novo 435)
under the optimised conditions used for structurally related
compounds (Table 2).[20] Due to the insolubility of the sub-
Table 1. Reagents and Conditions: a) CAL-B, 368C, KPi buffer (100 mm,
pH 8.5). eeE =enantiomeric excess of the ester.
Entry
Enzyme
1 h
5 h
10 h
eeE [%]
eeE [%]
eeE [%]
1
2
CAL-B
BS3
33 (R)
36 (S)
66 (R)
50 (S)
81 (R)
78 (S)
Scheme 1. Synthesis of 2,3-dihydrobenzo[b]furans 7a–c. a) HBF4·Et2O, ethyl diazoacetate, CH2Cl2, RT; b) conc. H2SO4, CH2Cl2, RT (R=H: 6a 90%
over two steps; R=Cl: 6b 73% over two steps); c) Mg turnings, MeOH (R=H: 7a 90%; R=Cl: 7b 90%); d) EtOH, conc. H2SO4, 308C (94%).
&
2
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Resolution of 2,3-Dihydrobenzo[b]furans
FULL PAPER
Table 2. Kinetic resolution of esters 7a–c. Reagents and conditions:
a) CAL-B (Novo 435), RT, KPi buffer (100 mm, pH 8.5). eeE =enantio-
meric excess of the ester; eeA =enantiomeric excess of the corresponding
acid; c=conversion; t=reaction time.
Entry
R
R1
t
c
[%]
eeE [%]
(yield [%])
eeA [%]
(yield [%])
E value
[h]
N
ACHTUNGTRENNUNG
1
2
3
H
Cl
Cl
Me
Me
Et
4
6
6
50
50
50
>99 (46)
>99 (49)
>99 (47)
ꢀ97 (50)
>99 (49)
> 99 (48)
>200
>200
>200
strates in water, they were emulsified in pure buffer by vigo-
rous stirring to generate a maximum surface and fast con-
version. After the theoretical amount of base was consumed
to maintain the pH at approximately 8.5 (4–6 h), the reac-
tions were stopped by removing the enzyme by filtration.
Subsequent purification afforded the (R)-configured esters
7a–c and the corresponding (S)-configured acids 8a and 8b
in remarkable optical purity (more than 97% ee in all cases)
and excellent yields (greater than 46%) at 50% conversions
(Table 2, entry 1–3). The enantioselectivity was therefore
E>200 in all reactions. Moreover, considering the fact that
only 2% of the net weight of Novo 435 is pure protein,[24]
real catalytic ratios of 1:1000 (for rac-7a and rac-7b) and
1:2000 (rac-7c) for the hydrolysis may be calculated.
Regarding the enzymatic DKR, a reliable and efficient
protocol for the racemisation was needed to overcome the
yield limitation of 50%. Based on our experiences with
base-catalysed racemisation and epimerisation,[20] enantio-
pure starting material (R)-[7a–c] was initially reacted with
the sterically demanding, non-nucleophilic guanidine-base
Figure 2. a) Controlled racemisation of methyl ester (R)-7a (11.2 mm) as
a function of the amount of base amount (BEMP; analytical scale). Base
~
!
&
*
^
concentrations: ( ) 35 mm, ( ) 68 mm, ( ) 133 mm, ( ) 196 mm, (
256 mm, (3) 342 mm; b) Linear dependency of the interconversion con-
)
triazabicyclo-1,5,7ACHTUNGTRENNUNG[4.4.0]-dec-5-ene (TBD) in n-heptane. The
&
*
stants kinv of esters 7b ( ) and 7c ( ) as a function of the amount of
loss of ee could be monitored by chiral HPLC, but TBD
proved to be unsatisfactory for our intention; no racemisa-
tion could be detected, even at elevated temperatures of
808C or when an excess of base was used (more than 10
equivalents). Upon seeking an alternative, the Schwesinger
base tert-butylimino-2-diethylamino-1,3-dimethylperhydro-
1,2,3-diazaphosphorine (BEMP), which is a phosphorous-
based organic “super base” with a pKa of 27.5 (MeCN), was
found to be convenient.[25] This base allowed the rate of rac-
emisation to determine as a function of base concentration
(Scheme 2).
base. (c) Calculated values.
time and analysed by chiral HPLC. Figure 2a shows the
time-dependent loss of optical purity for ester 7a as a func-
tion of the amount of base (for racemisation curves of esters
7b and 7c, see the Supporting Information). As expected,
increasing BEMP concentrations resulted in a faster de-
crease in the ee values and, hence, faster racemisation. Nota-
bly, all experimental findings are in quantitative agreement
with the theoretical results, which are also reflected in the
interconversion constants (kinv) of esters 7b and 7c (Fig-
ure 2b). By using the data from Figure 2a and by consider-
ing the base concentration, the racemisation constant
(2krac =kinv) can be determined from the ratio of eeS/eeS0
(eeS0 =initial ee value of the enantiomer, eeS at a defined
time).[26,27] For the nonhalogenated ester 7a, a krac/cBEMP
value of 2.8ꢄ10À3 LmmolÀ1 hÀ1 and for the 5-chloro-substi-
tuted analogues 7b and 7c, 43.7ꢄ10À3 and 22.2ꢄ
10À3 LmmolÀ1 hÀ1 were calculated, respectively. The reasons
for the significant differences in the magnitude of the calcu-
From a practical point of view, the enantiopure esters
(R)-[7a–c] were treated with different base concentrations
and small samples were withdrawn after defined periods of
Scheme 2. Controlled racemisation of the esters (R)-[7a–c] in the pres-
ence of the Schwesinger base BEMP on an analytical scale.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

J. Pietruszka et al.
Table 3. Results of the dynamic kinetic resolution of the esters rac-7a–
c on preparative scale (reaction setup is depicted in Scheme 3).
lated values may lie in the stereoelectronic nature of the
compounds themselves; the chlorine atom with its electron-
withdrawing properties decreases the electron-density at C-
3, resulting in faster racemisation. For comparison, it was
found that, for complete racemisation of ester 7a within two
hours, approximately 30 equivalents of BEMP were needed,
whereas for 7b and 7c only 3–4 equivalents were sufficient.
With respect to the enzyme-mediated dynamic kinetic res-
olution, a highly enantioselective enzyme (CAL-B) and
a suitable base (BEMP) for a controlled racemisation were
established. To combine both data sets, inactivation of the
base through aqueous media needed to be considered. As
such, an aqueous one-pot system was ineligible for our pur-
poses. Alternatively, based on a two-compartment system
described in 2005 by Bornscheuer and co-workers,[28]
a simple reaction setup was designed in which a spatial sepa-
ration of both processes is realized (Scheme 3).[20a] A stan-
Entry
R
R1
t
BEMP
[equiv]
eeA
[%]
Yield
[%]
[h]
N
1
2
3
4
5
H
H
H
Cl
Cl
Me
Me
Me
Me
Et
26
26
24
24
24
0.88
1.00
1.50
2.00
1.00
>84
ꢀ90
>95
>99
>99
91
95
92
71
82
hance the enantiomeric excess of the product, the amount
of base was increased, keeping the other reaction parame-
ters constant. As a result, the yield (95%) and the ee (ca.
90%) could be further increased (Table 3, entry 2). Further
increasing the amount of base gave even better results; acid
(S)-8a could be isolated after 24 h in a remarkable yield of
92% with an ee value of more than 95% (Table 3, entry 3).
Transferring the reaction conditions to the chlorinated esters
also proved to be convenient; in both cases, the acid (S)-8b
was obtained in perfect optical purity (ee>99%) and in
good yield (Table 2, entries 4 and 5). Hence, enzyme-medi-
ated hydrolysis in combination with a chemical racemisation
has been developed as a new approach to DKR. It should
be noted that the influence of water content in n-heptane
was not systematically investigated; however, when com-
pared to the racemisation in anhydrous n-heptane, no
change of the overall performance of the base was observed.
Markwell and co-workers described in the mid 1980ꢀs the
synthesis of various organic compounds that were evaluated
for analgesic activity in the mouse phenyl-p-quinone-in-
duced writhing test. Their studies revealed that racemic 7-
benzoyl-5-chloro-2,3-dihydrobenzo[b]furan carboxylic acid
(BRL 37959) combines potent analgesic activity with low
gastric irritancy at low concentrations.[13] Interestingly, de-
spite the promising biological feature, no enantioselective
synthesis has yet been realized; therefore, enantiopure ethyl
ester (R)-7c (either from the KR or the DKR after esterifi-
cation) was treated with the Lewis acid AlCl3 in CS2 at
room temperature, before benzoyl chloride was added. Pu-
rification afforded the R-configured ethyl ester (R)-9a in
only moderate yield of 30% after 72 h. Several attempts
were made to increase the yield of this step (e.g., variation
of ester functionality, the solvent, the reaction time, and
amount of Lewis acid), but all failed (see the Supporting In-
formation for optimization studies). Nevertheless, subse-
quent hydrolysis gave the desired bioactive compound
BRL 37959 (R)-2 in good yield (93%) for the first time
(Scheme 4).[29]
Scheme 3. Process flow sheet of the reaction setup, and basic principles
of the dynamic kinetic resolution for various substrates.
dard reaction flask was connected with a peristaltic pump to
a pre-packed column containing an immobilised form of
BEMP. The application of a basic two-phase system (buffer
and n-heptane as organic solvent) allows the enzyme and
produced acid to be kept in the aqueous layer, whilst the
ester (in the organic layer) was continuously pumped
through the racemisation column. As a result, a constant
racemisation of the ester can be obtained as the enantiomer-
ically pure acid accumulates in the aqueous phase. To avoid
unwanted side reactions or inactivation of the enzyme
through base leaching, the racemisation column was
equipped with a second layer of ion-exchange resin to trap
the base.
Initial results using methyl ester 7a as substrate (prepara-
tive scale; 500 mg, 2.81 mmol) were promising; the corre-
sponding acid (S)-8a was isolated after 26 h in 91% yield
with a high optical purity (84%ee), thus already demon-
strating dynamic kinetic resolution (Table 3, entry 1). To en-
&
4
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Resolution of 2,3-Dihydrobenzo[b]furans
FULL PAPER
H2SO4 (0.8–1.0 mL) under vigorous stirring for 15 min, then the reaction
was diluted with CH2Cl2 (15.0 mL) and neutralised with solid NaHCO3.
The residue was filtered and concentrated under reduced pressure. Sub-
sequent flash column chromatography (petroleum ether/EtOAc) afforded
the product 6a (2.50 g, 13.1 mmol, 80%) as a colourless liquid. Spectro-
scopic data are full in agreement with those reported previously.[18]
1H NMR (600 MHz, CDCl3) d=1.43 (t, 3J
(q, 3J
(1’,2’)=7.2 Hz, 2H; 1’-H), 7.36 (mc, 2H; Ar-H), 7.53 (mc, 1H; Ar-
ACHTUNGTNER(NUNG 2’,1’)=7.2 Hz, 3H; 2’-H), 4.41
Scheme 4. Synthesis of enantiopure analgesic agent BRL 37959 (R)-2.
a) CS2, AlCl3, 408C, 72 h (30%); b) H2O, conc. H2SO4, RT (93%).
AHCTUNGTRENNUNG
H), 8.07 (mc, 1H; Ar-H), 8.26 ppm (s, 1H, 2-H); 13C NMR (151 MHz,
CDCl3): d=14.4 (C-2’), 60.6 (C-1’), 111.7 (CHAr), 112.7 (C-9), 114.8
(CHAr), 122.1 (CHAr), 124.1 (CHAr),124.7 (CHAr), 125.3 (Cipso), 150.9 (C-
2), 155.6 (Cipso), 163.5 ppm (COOEt); GC-MS (EI, 70 eV): m/z (%): 190
(40) [M+], 165 (26) [C9H5O3+], 145 (100) [C9H5O2+].
Conclusions
Ethyl 5-chlorobenzo[b]furan-3-carboxylate (6b): According to the litera-
ture,[18] aldehyde 4b (3.00 g, 19.2 mmol) was diluted with anhydrous
CH2Cl2 (15.0 mL) and HBF4·OEt2 (54%, 270 mL, 1.71 mmol) was added
at RT. The solution was cooled to 08C and ethyl diazoacetate (20 vol%
in CH2Cl2, 3.00 g, 26.2 mmol) was added dropwise to the reaction within
1 h. When nitrogen formation had ceased, the remaining solvent was re-
moved under reduced pressure. The residue was treated with conc.
H2SO4 (0.5 mL) under vigorous stirring for 15 min, then the reaction was
diluted with CH2Cl2 (15.0 mL) and neutralised with solid NaHCO3. The
residue was filtered and concentrated under reduced pressure. Subse-
quent flash column chromatography (petroleum ether/EtOAc) afforded
the product 6b (3.25 g, 14.5 mmol, 73%) as a colourless solid. Spectro-
scopic data are full in agreement with those reported previously.[18] M.p.
We have demonstrated that kinetic resolution is a powerful
tool that can be used to synthesise enantiopure products
with excellent ee and yields. Insights from kinetic resolution
of methyl ester 7b could be transferred to the ethyl ester
7c. After enzyme screening by using HPLC-CD, a fast assay
could also be established for the ethyl ester 7c. Dynamic ki-
netic resolution can overcome disadvantages associated with
kinetic resolution such as yield limitations. With the set-up
described here, acids (S)-8a and (S)-8b could be obtained in
92 and 71% yield, respectively, and the enantiomeric excess
was determined to be more than 95 and 99%. Acylation of
ester (R)-7c and subsequent hydrolysis gave the active sub-
stance BRL 37959 (R)-2. Emerging difficulties encountered
during the acylation of methyl ester 7b were avoided by
synthesising ethyl ester 7c.
588C; 1H NMR (600 MHz, CDCl3) d=1.43 (t, 3J
H), 4.41 (q, 3J(1’,2’)=7.1 Hz, 2H; 1’-H), 7.32 (dd, 3J
(6,4)=2.0 Hz, 1H; 6-H), 7.44 (d, 3J(7,6)=8.8 Hz, 1H; 7-H), 8.03 (d, 4J-
(4,6)=2.1 Hz, 1H; 4-H), 8.26 ppm (s, 1H; 2-H); 13C NMR (151 MHz,
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
CDCl3): d=14.2 (C-2’), 60.8 (C-1’), 112.7 (CHAr), 114.6 (CHAr), 121.8
(CHAr), 125.6 (CHAr),125.8 (Cipso), 130.0 (CClAr), 152.0 (C-1’), 153.9
(Cipso), 162.8 ppm (COOEt); GC-MS (EI, 70 eV): m/z (%): 224 (41) [M+
], 196 (33) [C9H4ClO3+], 179 (100) [C9H4ClO2+], 123 (20) [C7H4Cl+].
Methyl 2,3-dihydrobenzo[b]furan-3-carboxylate (7a): Mg turnings
(1.38 g, 56.8 mmol) were added in small portions to a stirred solution of
ester 6a (2.20 g, 11.3 mmol) in anhydrous MeOH (80 mL). After all the
Mg had reacted and no staring material could be detected by TLC, the
mixture was concentrated under reduced pressure and the residue was
extracted against saturated NH4Cl (300 mL) with CH2Cl2 (4ꢄ20 mL) and
EtOAc (50 mL). The combined organic layer was dried over MgSO4, fil-
tered and concentrated under reduced pressure. Subsequent flash column
chromatography (petroleum ether/EtOAc acetate, 95:5) afforded the
Experimental Section
All starting materials were obtained from commercial suppliers and used
as received unless stated otherwise. Enzyme-catalysed reactions conduct-
ed on preparative scale were carried out with Novozyme 435 (Candida
antarctica lipase B immobilized on acrylic resin from Sigma-Aldrich, ac-
tivity according to specification: ꢁ10.000 UgÀ1, EC: 3.1.1.3; no change of
activity or selectivity was observed during the investigation). Synthetic
transformations were performed by using standard Schlenk techniques
under an Ar/N2 atmosphere using oven-dried (1208C) glassware. Solvents
were dried and purified either by conventional methods prior to use or
by using a Solvent Purification System (MBraun). Preparative chromato-
graphic separations were performed by column chromatography on
Merck silica gel 60 (0.063–0.200 mm). Solvents for flash chromatography
(petroleum ether/ethyl acetate) were distilled before use. Petroleum
ether refers to the fraction with a boiling point between 40–608C. TLC
analysis was carried out with pre-coated plastic sheets (Polygram SIL G/
UV, Macherey–Nagel) with detection by UV (254 nm) and/or by staining
with cerium molybdenum solution [phosphomolybdic acid (25 g), Ce-
product 7a (1.81 g, 10.2 mmol, 90%) as
(600 MHz, CDCl3): d=3.78 (s, 3H; OMe), 4.34 (dd, 3J
(2a,3)=6.6 Hz, 1H; 2-Ha), 4.67 (dd, 3J(2b,2a)=9.8 Hz, 3J
1H; 2-Hb), 4.93 (dd, 3J(3,2b)=9.2 Hz, 3J
(3,2a)=6.6 Hz, 1H; 3-H), 6.82
a
colourless oil. 1H NMR
(2a,2b)=9.8 Hz, 3J-
(2b,3)=9.2 Hz,
AHCTUNGTRENNUNG
A
R
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
(mc, 1H; ArH), 6.89 (mc, 1H; ArH), 7.18 (mc, 1H; ArH), 7.37 ppm (mc,
1H; ArH); 13C NMR (151 MHz, CDCl3): d=47.1 (C-3), 52.6 (COOCH3),
52.2 (C-2), 110.0, 120.7 (Cipso), 124.1, 125.4, 129.5 (CHAr), 159.8 (Cipso),
171.6 ppm (COOH); IR (ATR film): n˜ =2954, 1735 (C=O), 1596, 1482,
1460, 1435, 1330, 1233, 1204, 1172, 1016, 973, 749 cmÀ1; GC-MS (EI,
70 eV): m/z (%): 178 (61) [M+], 119 (100) [C8H7O+], 91 (63) [C7H7+].
Determination of the enantiomeric excess by chiral HPLC [column: 4.6ꢄ
250 mm, Chiracel OD-H (Daicel); flow rate=0.5 mLminÀ1; detection l=
210 nm; solvent=n-hexane/2-PrOH (98:2)]: Rt =14.76 [(R)-7a],
18.36 min [(S)-7b].
ACHTUNGTRENNUNG(SO4)2·H2O (10 g), conc. sulfuric acid (60 mL), H2O (940 mL)]. Optical
rotation was measured at 208C with a PerkinElmer Polarimeter 241 MC
against sodium D-line. 1H and 13C NMR spectra were recorded at 208C
with a Bruker Avance/DRX 600 spectrometer in CDCl3 with TMS as in-
ternal standard. Chemical shifts are given in ppm relative to the Me4Si
(1H: Me4Si=0 ppm) or relative to the resonance of the solvent (13C:
CDCl3 =77.0 ppm or 13C: MeOD=50.4 ppm).
Ethyl benzo[b]furan-3-carboxylate (6a): According to the literature,[18]
aldehyde 4a (2.00 g, 16.4 mmol) was diluted with anhydrous CH2Cl2
(4.0 mL) and HBF4·OEt2 (54%, 224 mL, 1.64 mmol) was added at RT.
The solution was cooled to 08C and ethyl diazoacetate (20 vol% in
CH2Cl2, 3.00 g, 26.2 mmol) was added dropwise to the reaction within
1 h. When nitrogen formation had ceased, the remaining solvent was re-
moved under reduced pressure. The residue was treated with conc.
Methyl 5-chloro-2,3-dihydrobenzo[b]furan-3-carboxylate (7b): Mg turn-
ings (1.78 g, 73.4 mmol) were added in small portions to a stirred solution
of ester 6b (3.00 g, 13.4 mmol) in anhydrous MeOH (150 mL). After all
the Mg had reacted and no staring material could be detected by TLC,
the mixture was concentrated under reduced pressure and the residue
was extracted against saturated NH4Cl (300 mL) with CH2Cl2 (4ꢄ20 mL)
and EtOAc (50 mL). The combined organic layer was dried over MgSO4,
filtered and concentrated under reduced pressure. Subsequent flash
column chromatography (petroleum ether/ethyl acetate 90:10) afforded
the methyl ester 7b (2.55 g, 12.0 mmol, 90%) a yellowish oil. 1H NMR
3
3
(600 MHz, CDCl3): d=3.80 (s, 3H; OCH3), 4.32 (dd, J
N
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

J. Pietruszka et al.
A
E
E
Enzymatic kinetic resolution of methyl 5-chloro-2,3-dihydrobenzo[b]fur-
an-3-carboxylate (rac-7b): In a two-necked reaction flask, ester rac-7b
(2.00 g, 9.40 mmol) was emulsified in potassium phosphate buffer
(60 mL, 100 mm, pH 8.0). The flask was equipped with a pH electrode
and a magnetic stirring bar (stirring at >1000 rpm). After immobilized
Candida antarctica (Novo 435) (100 mg) was added, the pH was main-
tained at pH 8.0 during the reaction by addition of small portions of
NaOH (1m). When an ee of more than 95% (determined by HPLC anal-
ysis, approx. 6 h) was reached for substrate (R)-7b, the mixture was fil-
tered through a plug of cotton wool and the organic phase was separated.
The remaining aqueous layer was acidified with 1m HCl to pH 1.0 and
extracted with EtOAc (4ꢄ40 mL). The combined organic layer was dried
over MgSO4 and concentrated under reduced pressure. The resulting oil
was purified by flash column chromatography (petroleum ether/EtOAc
90:10 + 1 vol% AcOH) to give ester (R)-7b (995 mg, 4.68 mmol, ee>
99%, 49%) as a colourless oil and the acid (S)-8b (921 mg, 4.70 mmol,
ee>99%, 49%) as colourless solid. Spectroscopic data of the ester (R)-
7b are full in agreement with those reported for its racemate (see
above). Optical rotation for (R)-7b [a]2D0 =+29 (c 0.91, CHCl3, ee>
99%).
G
ACHTUNGTRENNUNG
(d, J=8.5 Hz, 1H; ArH), 7.14 (dd, J=8.4 Hz, J=2.0 Hz, 2H, ; ArH),
7.34 ppm (mc, 1H; ArH); 13C NMR (151 MHz, CDCl3): d=47.0 (C-3),
52.8 (C-1’), 73.0 (C-2), 110.9 (CHAr), 125.4 (Cispo), 125.5, 125.9, 129.4
(CHAr), 158.6 (Cipso), 171.1 ppm (COOH); IR (ATR film) n˜ =2954, 1736
(C=O), 1605, 1328, 1295, 1238, 1204, 1110, 710 cmÀ1; GC-MS (EI, 70 eV):
m/z (%): 212 (63) [M+], 153 (100) [C8H6ClO+], 125 (81) [C6H3ClO+]; el-
emental analysis calcd (%) for C10H10ClO3 (213): C 56.49, 4.27; found: C
56.32, H 4.32. Determination of the enantiomeric excess by chiral HPLC
[column: 4.6ꢄ250 mm, Chiralpak IA (Daicel); flow rate=0.5 mLminÀ1
detection l=254 nm; solvent=n-heptane/2-PrOH (98:2)]: Rt =14.70
[(R)-7b], 15.42 min [(S)-7b].
Ethyl 5-chloro-2,3-dihydrobenzo[b]furan-3-carboxylate (7c): Conc.
H2SO4 (0.5 mL) was added to a stirred solution of methyl ester 7b
(200 mg, 0.94 mmol) in ethanol (50 mL) and the reaction was stirred for
24 h at 308C. EtOH was removed and the H2SO4 was neutralised by ad-
dition of saturated NaHCO3. The residue was extracted with CH2Cl2 (5ꢄ
5 mL) and the organic layer was dried over MgSO4, the solvent was re-
moved under reduced pressure and the product was purified by flash
column chromatography (petroleum ether/EtOAc, 99:1). Ester 7c
(199 mg, 0.88 mmol, 94%) was obtained as a brownish oil. 1H NMR
Spectroscopic data for acid (S)-8b: M.p. 758C; [a]2D0 =À7 (c 0.59, CHCl3
1
3
3
ee>99%); H NMR (600 MHz, CDCl3) d=4.36 (dd, JCAHTUNGTRENNNUG
A
(2a,2b)=9.5 Hz, 2J
ACHUTGTNRENNUG CAHTUNGTRENNUNG
(600 MHz, CDCl3) d=1.29 (t, 3J
(1’,2’)=7.1 Hz, 2H; 1’-H), 4.27 (dd, 3 J
1H; 3-H), 4.65 (dd, 2J(2a,2b)=9.6 Hz, 1H; 2-Ha), 4.92 (dd, 2J
9.2 Hz, 3J(2b,3)=6.9 Hz 1H; 2-Hb), 6.69 (d, 3J
(7,6)=8.5 Hz, 1H; 7-H),
7.10 (dd, 3J(6,7)=8.5 Hz, 4J
(6,4)=1.7 Hz, 1H; 6-H), 7.32 ppm (s, 1H; 4-
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
G
R
A
R
ACHTUNGTRENNUNG
T
1H; 6-H), 7.36–7.41 (m, 1H; 4-H), 11.36 ppm (s, 1H; COOH); 13C NMR
(151 MHz, CDCl3): d=46.9 (C-3), 72.7 (C-2), 111.0 (CHAr), 125.1 (Cipso),
125.6 (CHAr), 125.6 (CClAr), 129.8 (CHAr), 158.5 (Cipso), 176.7 ppm
(COOH); MS (ESI): m/z (%): 198 (100) [M+].
E
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
H); 13C NMR (151 MHz, CDCl3): d=14.2 (C-2’), 47.1 (C-3), 61.7 (C-1’),
73.0 (C-2), 110.8 (CHAr), 125.3 (Cipso), 125.4 (CHAr), 126.2 (CClAr), 129.3
(CHAr), 158.6 (Cipso), 170.4 ppm (COOEt); GC-MS (EI, 70 eV): m/z (%):
226 (63) [M+], 153 (100) [C8H6ClO+], 125 (51) [C6H3ClO+]. Determina-
tion of the enantiomeric excess by chiral HPLC [column: 4.6ꢄ250 mm,
Chiralpak IA (Daicel); flow rate=0.5 mLminÀ1; detection l=300 nm;
solvent=n-heptane/2-PrOH 99:1]: Rt =13.1 [(R)-7c], 13.9 min [(S)-7c].
Enzymatic kinetic resolution of ethyl 5-chloro-2,3-dihydrobenzo[b]furan-
3-carboxylate (rac-7c): In
a typical run, the ester rac-7c (1.00 g,
4.41 mmol) was emulsified in potassium phosphate buffer (100 mL,
100 mm, pH 8.5) at 08C. The flask was equipped with a pH electrode and
a magnetic stirring bar (stirring at >1000 rpm). After immobilized Candi-
da antarctica (Novo 435) (50 mg) was added, the pH was maintained at
pH 8.5 during the reaction by addition of small portions of NaOH (1m).
When an ee of >95% (determined by HPLC analysis, approx. 6 h) was
reached for substrate (R)-7c, the mixture was filtered through a plug of
cotton wool and the organic phase was separated. The remaining aqueous
layer was acidified with 1N HCl to pH 1.0 and extracted with EtOAc
(4ꢄ40 mL). The combined organic layer was dried over MgSO4 and con-
centrated under reduced pressure. The resulting oil was purified by flash
Enzymatic kinetic resolution of methyl 2,3-dihydrobenzo[b]furan-3-car-
boxylate (rac-7a): In a typical run, ester rac-7a (2.00 g, 11.3 mmol) was
emulsified in potassium phosphate buffer (30 mL, 100 mm, pH 8.5) at
08C (magnetic stirring; >1000 rpm). The reaction was initiated by the
addition of Novozyme 435 (100 mg) and the pH was monitored with
a pH electrode. The pH was maintained at pH 8.5 during the reaction by
the addition of small portions of NaOH solution (2m) and was stopped
after the consumption of 95% of the theoretical amount of base. The
aqueous layer was filtered through a plug of cotton wool (to remove the
enzyme beads), acidified with aqueous 2m HCl to pH 2.0, and extracted
with EtOAc (5ꢄ20 mL). The combined organic layer was dried over
MgSO4, filtered, and concentrated under reduced pressure. Subsequent
flash column chromatography (petroleum ether/ethyl acetate, 95:5 to
50:50) afforded the methyl ester (R)-7a (918 mg, 5.14 mmol, ee >99%,
46%) as a clear liquid, and the acid (S)-8a (915 mg, 5.15 mmol, ee
<97%, 50%) as a colourless solid. Spectroscopic data of the ester 7a are
full in agreement with those reported for its racemate (see above). Opti-
cal rotation for (R)-7a: [a]2D0 =À85.5 (c 0.81, CHCl3, ee>99%) {Lit.:[30]
[a]2D0 =À8.76 (c 0.69, CHCl3, ee 63%)}.
column chromatography (petroleum ether/EtOAc, 90:10
+ 1 vol.%
AcOH) to give the ester (R)-7c (412 mg, 2.07 mmol, ee>99%, 47%) as
a colourless oil and the acid (S)-8b (480 mg, 2.11 mmol, ee>99%, 48%)
as colourless solid. Spectroscopic data of ester (R)-7c are full in agree-
ment with those reported for its racemate (see above). Optical rotation
for (R)-7c [a]2D0 =+17.0 (c 0.25, CHCl3, ee>99%). Spectroscopic data of
acid (S)-8b are in full agreement with those reported (see above).
Representative procedure for the determination of racemisation constant
krac of [(R)-7a]: A stock solution (1 mL) of enantiopure (R)-7a (2 mL in
1 mL n-heptane) and BEMP (10, 20, 40, 60, 80 and 110 mL) were mixed
in an Eppendorf vial. All reactions were shaken at 258C and 1400 rpm,
and aliquots (50 mL) were removed after defined periods of time (20, 40,
60, 90 and 120 min). The removed samples were diluted with n-heptane
(300 mL) and extracted against NH4Cl (500 mL). The organic layer was
dried over MgSO4 and a sample (200 mL) of each was analysed by chiral
HPLC (see above).
Spectroscopic data for acid (S)-8a: M.p. 948C; 1H NMR (600 MHz,
CDCl3): d=4.37 (dd, 3J(3,2b)=9.7 Hz, 3J
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
(dd, 2J
9.3 Hz, 3J
(2a,2b)=9.4 Hz, 3J
E
ACHTUNGTRENNUNG
6.90 (td, J=7.5, 1.0 Hz, 1H; ArH), 7.20 (mc, 1H; ArH), 7.41 ppm (mc,
1H; ArH); 13C NMR (151 MHz, CDCl3): d=47.0 (C-3), 72.2 (C-2), 110.1,
120.8 (CHAr), 123.5 (Cipso), 125.5, 129.8, 159.8 (Cipso), 176.6 ppm (COO);
IR (ATR film): n˜ =2898, 1720 (C=O), 1625, 1481, 1381, 1231, 970,
755 cmÀ1; GC-MS (EI, 70 eV): m/z (%): 164 [M+] (65), 119 [C8H7O+]
(100), 91 [C7H7+] (88); elemental analysis calcd (%) for C9H8O3: C 65.85,
4.91; found: C 66.17, H 5.13. Optical rotation for (S)-8a: [a]2D0 =+79.6 (c
0.66, CHCl3, ee<97%). Determination of the ee was performed after de-
rivatisation with etherical diazomethane solution to the corresponding
methyl ester (see above).
Enzymatic dynamic kinetic resolution of methyl 2,3-dihydrobenzo[b]fur-
an-3-carboxylate (rac-7a): Before starting DKR, a 5 mL syringe was pre-
pared as follows: cotton wool was placed on the bottom, overlaid with
cation exchange resin (3.00 g; Merck, Amberlite, IRC-50, mesh 20–30).
A second layer of cotton wool followed and the immobilised base BEMP
on PS (2.00 g) was then added. The syringe was capped with a third layer
of cotton. The ester rac-7a (500 mg, 2.81 mmol) was dissolved in n-hep-
tane (20 mL) and placed in a three-necked flask with potassium phos-
phate buffer (65 mL, 100 mm, pH 8.5) and immobilized Candida antarcti-
ca lipase B (50 mg, Novo 435) was added. After 5 min, a Teflon tube con-
&
6
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Resolution of 2,3-Dihydrobenzo[b]furans
FULL PAPER
nected to
a
peristaltic pump (Pharmacia Bioscience P1, flow
Ethyl ester (R)-8c (400 mg, 1.76 mmol) was diluted with CS2 (2 mL) and
transferred into the reaction vessel. The mixture was allowed to warm to
ambient temperature and stirred for a further 72 h. The solution was
poured onto ice with 5m HCl and extracted with CH2Cl2 (3ꢄ40 mL). The
combined organic layer was dried over MgSO4 and the solvent was re-
moved under reduced pressure. The remaining brown oil was purified by
flash column chromatography (petroleum ether/EtOAc, 99:1) to afford
the product (175 mg, 0.53 mmol, 30%) as a colourless oil. Spectroscopic
data are full in agreement with those reported previously.[13] [a]D20 =À43
0.5 mLminÀ1) was dipped into the organic layer. The other side behind
the peristaltic pump was connected to the syringe. The pump was run
continuously until the end of the reaction. The pH value was monitored
with a pH electrode and maintained at pH 8.8 by the addition of 1m
NaOH. When the theoretical amount of NaOH was consumed, the reac-
tion was stopped (ca. 24 h). The mixture was filtered through cotton wool
to remove the enzyme. The organic layer was separated and the aqueous
layer was acidified with aqueous 2m HCl to pH 2.0. The aqueous layer
was extracted with EtOAc (5ꢄ20 mL). The combined organic layer was
dried over MgSO4 and concentrated under reduced pressure. The residue
was purified by flash column chromatography (petroleum ether/EtOAc
90:10+1 vol% AcOH) to give the acid (S)-8a (424 mg, 2.58 mmol, ee>
95%, 92%) as a colourless solid. Spectroscopic data are full in agree-
ment with those reported in the kinetic resolution (see above).
(c 0.32, CHCl3); 1H NMR (600 MHz, CDCl3): d=1.34 (t, 3J
ACHTUNGTRENNUNG
7.2 Hz, 3H; 2’-H), 4.28 (q, 3J
(3,2a)=9.4 Hz, 2J
3J
ACHTUNGTRENNUNG
G
R
ACHTUNGTRENNUNG
G
N
ACHTUNGTRENNUNG
7.4 Hz, 1H; 2-Hb), 7.41 (s, 1H; 6-H), 7.47 (mc, 2H; ArH), 7.52 (s, 1H; 4-
H), 7.59 (mc, 1H, ArH), 7.76–7.84 ppm (m, 2H; ArH); 13C NMR
(151 MHz, CDCl3): d=14.2 (C-2’), 46.5 (C-3), 62.0 (C-1’), 73.8 (C-2),
122.3 (C-8), 125.5 (Cipso), 128.1 (CClAr), 128.3 (CHAr), 128.6 (CHAr), 128.6
(CHAr), 128.9 (CHAr), 128.9 (CHAr), 129.6 (CHAr), 129.9 (CHAr), 137.1
(C-1’’’), 157.2 (Cipso), 169.9 (C-10), 192.9 ppm (COOEt); GC-MS (EI,
70 eV): m/z (%): 330 (67) [M+], 257 (63) [C15H10ClO2+], 179 (23)
[C9H5ClO2+], 105 (100) [C7H5O+], 77 (44) [C6H5+].
Enzymatic dynamic kinetic resolution of methyl 5-chloro-2,3-dihydroben-
zo[b]furan-3-carboxylate (rac-7b): Before starting DKR, a 5 mL syringe
was prepared as follows: cotton wool was placed on the bottom, overlaid
with cation exchange resin (1.00 g; Merck, Amberlite, IRC-50, mesh 20–
30). A second layer of cotton wool followed and the immobilised base
BEMP on PS (1.00 g) was then added. The syringe was capped with
a third layer of cotton. The ester rac-7b (500 mg, 2.35 mmol) was dis-
solved in n-heptane (40 mL) and placed with potassium phosphate buffer
(60 mL, 100 mm, pH 8.0) in a three necked flask. Immobilised Candida
antarctica Lipase B (25 mg, Novo 435) was then added. After 5 min,
a Teflon tube connected to a peristaltic pump was dipped into the organic
layer. The other side behind the peristaltic pump was connected to the
syringe. The pump was run continuously at 0.5 mLminÀ1 until the end of
the reaction. The pH value was monitored with a pH electrode and was
maintained at pH 8.8 by the addition of 1m NaOH. After the theoretical
amount of NaOH was consumed, the reaction was stopped (ca. 35 h).
The mixture was filtered through cotton wool to remove the enzyme.
The organic layer was separated and the aqueous layer was acidified with
1m HCl to pH 1.0. The aqueous layer was extracted with EtOAc (4ꢄ
40 mL) and the combined organic layer was dried over MgSO4 and con-
centrated under reduced pressure. The residue was purified by flash
7-Benzoyl-5-chloro-2,3-dihydrobenzofuran-3-carboxylic acid [(R)-2]:
Ethyl ester (R)-9a (200 mg, 0.60 mmol) was dissolved in THF (5 mL) and
conc. HCl (10 mL) was added to demineralised H2O (1 mL). The aqueous
solution was transferred to the reaction mixture that was stirred for 24 h
at RT. THF was removed by rotary evaporation and the residue was
quenched with NH4Cl and extracted with CH2Cl2 (3ꢄ20 mL). The com-
bined organic layer was dried over MgSO4 and the solvent was removed
by under reduced pressure. The crude product was purified by flash
column chromatography (petroleum ether/EtOAc, 90:10+1 vol.%
AcOH) to afford the product (170 mg, 0.56 mmol, 93%) as a slightly
yellow oil. Spectroscopic data are full in agreement with those reported
in literature.[13] [a]D20 =+40 (c 0.58, CHCl3); 1H NMR (600 MHz, CDCl3):
d=4.41 (dd, 3J
(2a,2b)=9.5 Hz, 3J
AHCTUNGTREG(NNNU 2b,3)=6.9 Hz, 1H; 2-Hb), 7.13–7.19 (m, 1H; ArH), 7.44–7.52 (m, 2H;
A
ACHTUNGTRENNUNG
A
R
ACHTUNGTRENNUNG
ArH), 7.54–7.67 (m, 2H; ArH), 7.73–7.85 ppm (m, 2H; ArH); 13C NMR
(151 MHz, CDCl3): d=21.4 (C-3), 73.5 (C-2), 122.4 (C-7), 125.3 (Cipso),
125.7 (CClAr), 127.3 (CHAr), 128.3 (CHAr), 128.4 (CHAr),129.1 (CHAr),
129.1 (CHAr), 129.9 (CHAr), 130.6 (CHAr), 133.3 (C-1’), 137.0 (C-2’), 157.2
(Cipso), 192.9 ppm (COOH); MS (ESI, cation): m/z (%): 303 [M+].
column chromatography (petroleum ether/EtOAc 90:10
+ 1 vol%
HOAc) to give the acid (S)-8b (329 mg, 1.66 mmol, ee>99%, 71%) as
a colourless solid. Spectroscopic data are full in agreement with those re-
ported in the kinetic resolution (see above).
Enzymatic dynamic kinetic resolution of ethyl 5-chloro-2,3-dihydroben-
zo[b]furan-3-carboxylate (rac-7c): Before starting DKR, a 5 mL syringe
was prepared as follows: cotton wool was placed on the bottom, overlaid
with cation exchange resin (2.00 g; Merck, Amberlite, IRC-50, mesh 20–
30). A second layer of cotton wool followed and the immobilised base
BEMP on PS (1.00 g) was then added. The syringe was capped with
a third layer of cotton. The ester rac-7c (1.00 g, 4.41 mmol) was dissolved
in n-heptane (20 mL) and placed with potassium phosphate buffer
(40 mL, 100 mm, pH 8.0) in a three-necked flask. Immobilised Candida
antarctica Lipase B (40 mg, Novo 435) was then added. After 5 min,
a Teflon tube connected to a peristaltic pump was dipped into the organic
layer. The other side behind the peristaltic pump was connected to the
syringe. The pump was run continuously (0.5 mLminÀ1) until the end of
the reaction. The pH value was monitored with a pH electrode and was
maintained at pH 8.5 by addition of 1m NaOH. When the theoretical
amount of NaOH was consumed, the reaction was stopped (ca. 24 h).
The mixture was filtered through cotton wool to remove the enzyme.
The organic layer was separated and the aqueous layer was acidified with
1m HCl to pH 1.0. The aqueous layer was extracted with EtOAc (4ꢄ
40 mL) and the combined organic layer was dried over MgSO4 and con-
centrated under reduced pressure. Subsequent flash column chromatogra-
phy (petroleum ether/EtOAc 90:10 + 1 vol% AcOH) afforded the acid
(S)-8b (720 mg, 3.62 mmol, ee>99%, 82%). Spectroscopic data are full
in agreement with those reported in the kinetic resolution (see above).
Acknowledgements
We gratefully acknowledge the Federal Ministry of Education and Re-
search (GenoMik-Transfer project “ExpresSys”), the Ministry of Innova-
tion, Science and Research of the German federal state of North Rhine-
Westphalia (NRW) (technology platform “ExpressO” within the “Ziel 2-
Programm 2007-2013, NRW–EFRE”), the Deutsche Forschungsgemein-
schaft, the Heinrich Heine University Dꢁsseldorf, and the Forschungs-
zentrum Jꢁlich GmbH for ongoing support of our projects.
[1] R. Noyori, Nat. Chem. 2009, 1, 5–6.
[2] For general reading and multiple applications of biotransformations,
see: a) K. Faber, Biotransformations in Organic Chemistry, Springer,
Berlin, 2011, 7th ed.; b) V. Gotor, I. Alfonso, E. Garcꢅa-Uridales,
Asymmetric Organic Synthesis with Enzymes, Wiley-VCH, Wein-
heim, 2008; c) U. T. Bornscheuer, R. J. Kazlauskas, Hydrolases in
Organic Synthesis-Regio- and Stereoselective Biotransformations,
Wiley-VCH, Weinheim, 2006, 2nd ed..
[3] Selected reviews: a) A. Schmid, J. S. Dordick, B. Hauer, A. Kiener,
M. Wubbolts, B. Witholt, Nature 2001, 409, 258–268; b) R. N. Patel,
ACS Catal. 2011, 1, 1056–1074; c) S. Wenda, S. Illner, A. Mell, U.
Kragl, Green Chem. 2011, 13, 3007–3047; d) T. Fischer, J. Pietrusz-
ka, Top. Curr. Chem. 2010, 297, 1–43; e) D. MuÇoz Solano, P.
Hoyos, M. J. Hernꢆiz, A. R. Alcꢆntara, J. M. Sꢆnchez-Montero, Bio-
Ethyl 7-benzoyl-5-chloro-2,3-dihydrobenzofuran-3-carboxylate [(R)-9a]:
According to the literature,[13] AlCl3 (1.60 g, 12.0 mmol) was suspended
in CS2 (10 mL) under an atmosphere of dry nitrogen. The mixture was
cooled to 08C and benzoyl chloride (1.68 g, 12.00 mmol) was added.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

J. Pietruszka et al.
resour. Technol. 2012, 115, 196–207; f) A. Liese, C. Seelbach, C.
Wandrey in Industrial biotransformations, 2nd ed., Wiley-VCH,
Weinheim, 2006.
[13] E. A. Boyle, F. R. Mangan, R. E. Markwell, S. A. Smith, M. J. Thom-
son, R. W. Ward, P. A. Wyman, J. Med. Chem. 1986, 29, 894–898.
[14] M. Iwahashi, A. Shimabukuro, T. Onoda, Y. Matsunga, Y. Okada,
R. Matsumoto, F. Nambu, H. Nakai, M. Toda, Bioorg. Med. Chem.
2011, 19, 4574–4588.
[15] For a review on the recent progress on the synthesis of 2,3-dihydro-
benzo[b]furans, see: F. Bertolini, M. Pineschi, Org. Prep. Proced.
Int. 2009, 41, 385–418.
[16] a) N. Ortega, S. Urban, B. Beiring, F. Glorius, Angew. Chem. Int.
Ed. 2012, 51, 1710–1713; Angew. Chem. 2012, 124, 1742–1745; b) S.
Kaiser, S. P. Smidt, A. Pfalz, Angew. Chem. 2006, 118, 5318–5321;
Angew. Chem. Int. Ed. 2006, 45, 5194–5197.
[4] For reviews dealing exclusively with non-enzymatic chemical kinetic
resolution processes, see: a) H. Pellissier, Adv. Synth. Catal. 2011,
353, 1613–1666; b) E. Vedejs, M. Jure, Angew. Chem. 2005, 117,
4040–4069; Angew. Chem. Int. Ed. 2005, 44, 3974–4001;
c) D. E. J. E. Robinson, S. D. Bull, Tetrahedron: Asymmetry 2003,
14, 1407–1446.
[5] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keßeler, R.
Stꢁrmer, T. Zelinski, Angew. Chem. 2004, 116, 806–843; Angew.
Chem. Int. Ed. 2004, 43, 788–824.
[6] For selected reviews dealing with DKR, see: a) H. Pellissier, Tetra-
hedron 2011, 67, 3769–3802; b) H. Pellissier, Tetrahedron 2008, 64,
1563–1601; c) O. Pꢇmies, J.-E. Bꢂckvall, Chem. Rev. 2003, 103,
3247–3262; d) B. Martꢅn-Matute, J.-E. Bꢂckvall, Curr. Opin. Chem.
Biol. 2007, 11, 226–232; e) M.-J. Kim, Y. Ahn, J. Park, Curr. Opin.
Biotechnol. 2002, 13, 578–587; f) N. J. Turner, Curr. Opin. Chem.
Biol. 2004, 8, 114–119.
[7] For reviews dealing with different racemisation techniques and ap-
plication in DKR, see: a) F. F. Huerta, A. B. E. Minidis, J.-E. Bꢂck-
vall, Chem. Soc. Rev. 2001, 30, 321–331; b) B. Schnell, K. Faber, W.
Kroutil, Adv. Synth. Catal. 2003, 345, 653–666.
[8] For recent examples of enzymatic DKR, see: a) P. Vongvilai, M.
Linder, M. Sakulsombat, M. S. Humble, P. Berglund, T. Brinck, O.
Ramstrçm, Angew. Chem. 2011, 123, 6722–6725; Angew. Chem. Int.
Ed. 2011, 50, 6592–6595; Angew. Chem. 2011, 123, 6722–6725;
b) A. Rioz-Martꢅnez, A. Cuetos, C. Rodrꢅguez, G. de Gonzalo, I.
Lavandera, M. W. Fraaije, V. Gotor, Angew. Chem. 2011, 123, 8537–
8540; Angew. Chem. Int. Ed. 2011, 50, 8387–8380; c) H. Kim, Y. K.
Choi, J. Lee, E. Lee, J. Park, M.-J. Kim, Angew. Chem. 2011, 123,
11136–11140; Angew. Chem. Int. Ed. 2011, 50, 10944–10948; d) D.
Koszelewski, B. Grischek, S. M. Glueck, W. Kroutil, K. Faber,
Chem. Eur. J. 2011, 17, 378–383; e) R. Lihammar, R. Millet, J.-E.
Bꢂckvall, Adv. Synth. Catal. 2011, 353, 2321–2327; f) F. Poulhꢈs, N.
Vanthuyne, M. P. Bertrand, S. Gastaldi, G. Gill, V. Gotor, J. Org.
Chem. 2011, 76, 7281–8276; g) F. J. Quijada, V. Gotor, F. Rebolledo,
Org. Lett. 2010, 12, 3602–3605; h) Y. Cheng, G. Xu, J. Wu, C.
Zhang, L. Yang, Tetrahedron Lett. 2010, 51, 2366–2369.
[17] N. I. Shuikin, I. I. Dmitriev, T. P. Dobrynina, Chem. Abstr. 1941, 35,
2508.
[18] M. E. Dudley, M. M. Morshed, M. M. Hossain, Synlett 2006, 1711–
1714.
[19] a) Y. I. Kwon, Y. G. Hwan, P. C. Siek, Tetrahedron Lett. 1986, 27,
2409–2410; b) G. H. Lee, I. K. Youn, E. B. Choi, H. K. Lee, G. H.
Yon, H. C. Yang, C. S. Pak, Curr. Org. Chem. 2004, 8, 1263–1287.
[20] For a detailed investigation on the kinetic and dynamic kinetic reso-
lution of structurally related methyl 2,3-dihydro-1H-indenecarboxy-
late, see: a) J. Pietruszka, R. C. Simon, F. Kruska, M. Braun, Eur. J.
Org. Chem. 2009, 6217–6224; for transfer-studies to methyl indo-
line-3-carboxylate, see: b) J. Pietruszka, R. C. Simon, ChemCatChem
2010, 2, 505–508; c) J. Pietruszka, R. C. Simon, Chem. Eur. J. 2010,
16, 14534–14544.
[21] M. T. Reetz, K. M. Kꢁhling, H. Hinrichs, A. Deege, Chirality 2000,
12, 479–482.
[22] H. Hamzic, J. Pietruszka, D. Sandkuhl, Chirality 2011, 23, E110–
E115.
[23] For examples of the determination of the absolute configuration by
using the HPLC-CD technique, see: a) W. Krenn, P. Verdino, G.
Uray, K. Faber, O. Kappe, Chirality 1999, 11, 659–662; b) G. Bring-
mann, T. A. M. Gulder, M. Reichert, T. Gulder, Chirality 2008, 20,
628–642.
[24] F. Secundo, G. Carrea, C. Soregaroli, D. Varinelli, R. Morrone, Bio-
technol. Bioeng. 2001, 73, 157–163.
[25] T. Ishikawa, Superbases for Organic Synthesis, Wiley-VCH, Wein-
heim, 2009.
[9] The dimensionless E value describes the selectivity of a resolution
and remains constant throughout the reaction, see: a) C. S. Chen, Y.
Fujimoto, G. Girdaukas, C. J. Sih, J. Am. Chem. Soc. 1982, 104,
7294–7299; b) W. Kroutil, A. Kleewein, K. Faber, Tetrahedron:
Asymmetry 1997, 8, 3251–3261.
[26] The rate of racemisation can be described either as interconversion
of the current enantiomers or as the rate of formation of the race-
mate; for further details, see: E. J. Ebbers, G. J. A. Ariaans, J. P. M.
Houbiers, A. Bruggink, B. Zwanenburg, Tetrahedron 1997, 53, 9417–
9476.
[10] a) J. A. Pesti, J. Yin, L.-h. Zhang, L. Anzalone, R. E. Waltermire, P.
Ma, E. Gorko, P. N. Confalone, J. Fortunak, C. Silverman, J. Black-
well, J. C. Chung, M. D. Hrytsak, M. Cooke, L. Powell, C. Ray, Org.
Process Res. Dev. 2004, 8, 22–27; b) L. W. Wang, Y. C. Cheng, S. W.
Tsai, Bioprocess Biosyst. Eng. 2004, 27, 39–49; c) P. D’Arrigo, L.
Cerioli, S. Servi, F. Viani, D. Tessaro, Catal. Sci. Technol. 2012, DOI:
10.1039/c2cy20106b; d) D. Arosio, A. Caligiuri, P. DꢀArrigo, G. Ped-
erocchi-Fantoni, C. Rossi, C. Saraceno, S. Servi, D. Tessaro, Adv.
Synth. Catal. 2007, 349, 1345–1348; e) D. Tessaro, L. Cerioli, S.
Servi, F. Viani, P. D’Arrigo, Adv. Synth. Catal. 2011, 353, 2333–
2338; f) P. DꢀArrigo, D. Arosio, D. Moscatelli, S. Servi, F. Viani, D.
Tessaro, Tetrahedron: Asymmetry 2011, 22, 851–856.
[27] Note: lnCATHNUGETRN(UNNG eeS/eeS0)=2krac·t whereas 2krac =kinv.
[28] P. ꢉdman, L. A. Wessjohann, U. T. Bornscheuer, J. Org. Chem.
2005, 70, 9551–9555.
[29] The ee of product (R)-9a could not be established by conventional
chromatographic methods or derivatisation techniques. However,
during incomplete conversion under Friedel–Crafts conditions no
racemisation of starting material (R)-7a was observed. Furthermore,
when resubmitting the product to the reaction conditions (AlCl3,
RT, 72 h) no change in the optical ration was found and, hence, it
was assumed that the product was configurationally stable under the
given conditions.
[30] H. M. L. Davies, M. V. A. Grazini, E. Aouad, Org. Lett. 2001, 3,
1475–1477.
[11] D. Chavez-Flores, J. M. Salvador, Tetrahedron: Asymmetry 2012, 23,
237–239; and references cited.
[12] C. William, D. S. Johnson, J. D. O’Donnel, M. J. Pettersson, C. Sub-
ramanyam, International Patent: WO 2011/048525 A1, 2011.
Received: February 29, 2012
Published online: && &&, 2012
&
8
&
www.chemeurj.org
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Resolution of 2,3-Dihydrobenzo[b]furans
FULL PAPER
BRiLliant synthesis! With an enantio-
selectivity assay based on a HPLC-CD
protocol as a starting point, first the
kinetic and then the dynamic kinetic
enzymatic resolution of the title com-
pounds was performed (see scheme).
For the first time, access to the enan-
tiomerically pure active agent
Kinetic Resolution
P. Bongen, J. Pietruszka,*
R. C. Simon ................... &&&& — &&&&
Dynamic Kinetic Resolution of 2,3-
Dihydrobenzo[b]furans: Chemoenzy-
matic Synthesis of Analgesic Agent
BRL 37959
BRL 37959 was established in a short
and concise manner.
Chem. Eur. J. 2012, 00, 0 – 0
ꢃ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!