A practical high through-put continuous process for the synthesis of chiral cyanohydrins

Article abstract of DOI:10.1021/jo020432n

A practical high through-put continuous process for the synthesis of chiral cyanohydrins is reported. Pretreated almond meal (or other solid raw enzyme sources) was loaded in a column to form a reactor, to which were attached a supplying system to deliver

Full text of DOI:10.1021/jo020432n

A P r a ctica l High Th r ou gh -P u t Con tin u ou s
P r ocess for th e Syn th esis of Ch ir a l
Cya n oh yd r in s
Peiran Chen, Shiqing Han, Guoqiang Lin,* and Zuyi Li
Shanghai Insistitute of Organic Chemistry,
Chinese Academy of Sciences, 354 Fenglin Road,
Shanghai 200032, China
lingq@pub.sioc.ac.cn
Received J une 25, 2002
Abstr a ct: A practical high through-put continuous process
for the synthesis of chiral cyanohydrins is reported. Pre-
treated almond meal (or other solid raw enzyme sources)
was loaded in a column to form a reactor, to which were
attached a supplying system to deliver a solution of substrate
and HCN in solvent on one end and a collecting-separating
system on the other end. By controlling the flowing rate,
optimal conditions were achieved for the hydrocyanation of
various aromatic carboxaldehydes in a “micro-aqueous”
medium to produce chiral cyanohydrins in high yields and
high enantiomeric excess (ee) with high substrate/catalyst
ratio.
F IGURE 1. A schematic flow sheet of the continuous process.
reactor due to their swelling by absorption of the water
in aqueous medium.
In the previous papers,4 we described an efficient
catalytic process using a defatted crude almond meal
(peach or loquat kernel meal works as well) in a micro-
aqueous organic solvent system for the formation of chiral
cyanohydrins in high enantioselectivities and high yields.
The enzyme was retained in the intact cells and kept as
in the natural environments. Furthermore, the organic
medium used for the reaction caused no swelling prob-
lem. Therefore, we envisioned that this reaction condition
could be applied to a column or pipe-like reactor to
continuously produce the desired product in high through-
put. This would not be applicable to the biphasic system,
because the plant cells would absorb the water contained
in the system and become swollen, finally clogging the
column or pipe reactor.
Cyanohydrins not only play an important role at the
interface between chemistry and biology, but they also
have considerable synthetic potential as chiral building
blocks, especially in a wide range of pharmaceutical and
agrochemical applications. With regard to the prepara-
tion of chiral cyanohydrins, it can be mentioned that, in
addition to the asymmetric synthetic methods such as
being catalyzed by cyclic peptides or chiral Lewis acid,1
the biocatalytic hydrocyanations with oxynitrilases pro-
vide a competitive access in the formation of (R)- and (S)-
cyanohydrins. This enzymatic approch has been success-
fully developed since the pioneering work of Effenberger.2
Application of an almond meal as a solid crude enzyme
source3a and a polymer-entrapped (R)-oxynitrilase as an
immobilized enzyme3b have been reported in the hydro-
cyanation reaction in a biphasic (aqueous buffer solution
and organic solvent) reaction system.
In the case where enzyme preparation is in the solid
form (e.g. the crude enzyme source preparation and the
immobilized enzyme as mentioned above), enzymatic
reactions can be efficiently performed in a continuous
way in a tubular reaction vessel to give a higher through-
put than in a batch process. However, reactions in
aqueous buffer solution have to be carried out batchwise,
because the solid enzyme preparations may clog the
This process involves (1) preparing meal of an enzyme
source material (for example, almond kernel) by soaking
the material in water, air-drying, pulverizing the swollen
material in a homogenizer, and then defatting the
obtained meal, which contains a micro-amount of water
(8.96% (w/w)) after the above-mentioned treatment; (2)
loading the resultant meal into a column; and (3) carring
out the reaction in an organic solvent such as isopropyl
ether (IPE, with a water content of 0.3% v/v). Benzalde-
hyde was hydrocyanated in a column filled with ca. 15 g
of the defatted almond meal to produce mandelonitrile
in both good yields and enantiomeric excess. (Table 1).
As shown in Table 1, after in total 2 mol of the
substrate was treated, the column still retained its high
catalytic activity. The through-put of the column is 3630
g/(L‚day) (grams of the cyanohydrin per liter of the
almond meal in a day). According to van der Gen et al.,6
two grams of pure oxynitrilase could be obtained from 1
kg of almond kernel. Therefore, the column efficiency can
(1) For examples, see: Abiko, A.; Wang, G. J . Org. Chem. 1996, 61,
2264. Hayashi, M.; Miyamoto, Y.; Inoue, T.; Oguni, N. J . Chem. Soc.,
Chem. Commun. 1991, 1752.
(2) For the development of the enzymatic asymmetric synthesis of
cyanohydrins, see: (a) Effenberger, F. Angew. Chem., Int. Ed. 1994,
33, 1555. (b) Effenberger, F.; Fo¨ster, S.; Wajant, H. Curr. Opin.
Biotechnol. 2001, 532. (c) Gro¨ger, H. Adv. Synth. Catal. 2001, 343, 547.
(d) Gregory, R. J . H. Chem. Rev. 1999, 99, 3649.
(4) (a) Han, S.; Lin, G.; Li, Z. Tetrahedron: Asymmetry 1998, 9, 1835.
(b) Lin, G.; Han, S.; Li, Z. Tetrahedron 1999, 55, 3531. (c) Han, S.;
Chen, P.; Lin, G.; Huang, H.; Li, Z. Tetrahedron: Asymmetry 2001,
12, 843. (d) Chen, P.; Han, S.; Lin, G.; Huang, H.; Li, Z. Tetrahedron:
Asymmetry 2001, 12, 3273.
(5) Brussee, J .; Loos, W. T.; Kruse, C. G.; van der Gen, A.
Tetrahedron 1990, 46, 979.
(3) (a) Huuhtanen, T. T.; Kanerva, L. T. Tetrahedron: Asymmetry
1992, 3, 1223. (b) Gro¨ger, H.; Capan, E.; Barthuber, A.; Vorlop, K.-D.
Org. Lett. 2001, 3, 1969.
(6) Smitskamp-Wilms, E.; Brussee, J .; van der Gen, A. Recl. Trav.
Chim. Pays-Bas 1991, 110, 209.
10.1021/jo020432n CCC: $22.00 © 2002 American Chemical Society
Published on Web 10/19/2002
J . Org. Chem. 2002, 67, 8251-8253
8251

TABLE 1. Hyd r ocya n a tion of Ben za ld eh yd e in Colu m n
(in sid e d ia m eter 0.7 cm × len gth 80 cm , w ith ca . 15 g of
th e d efa tted a lm on d m ea l, flow r a te ) 3 m L/m in , T ) 15
°C)
substrate
(mmol)
total substrate
(mmol)
yield
(%)
ee
run
(%)a
4
13
25
32
50
110
150
200
10
10
10
10
10
10
10
10
40
130
250
320
500
110
1500
2000
91
92
95
94
91
89
90
89
99
>99
>99
>99
>99
>99
>99
>99
F IGURE 2. Time courses of cyanohydrins for five aldehydes:
(a) benzaldehyde, (b) p-fluorobenzaldehyde, (c) furanyl-2-
carboxaldehyde, (d) thienyl-2-carboxaldehyde, and (e) 5-me-
thylfuranyl-2-carboxaldehyde.
a
The enantiomeric excess (ee) was obtained by comparing the
determined specific optical rotation to the reported data.5
TABLE 3. Hyd r ocya n a tion of F ive Heter oa r yl
Ca r boxa ld eh yd es in Colu m n (in sid e d ia m eter 0.7 cm
len gth 80 cm , w ith ca . 15 g of th e d efa tted a lm on d m ea l,
T ) 15 °C)
TABLE 2. Hyd r ocya n a tion of F u r a n yl-2-ca r boxa ld eh yd e
(flow r a te ) 1 m L/1 m in , T ) 15 °C)
substrate
(mmol)
total substrate
(mmol)
ee
run
yield (%)
(%)a
1
12
17
24
30
38
50
60
90
10
10
10
10
10
10
10
10
10
10
10
120
170
240
300
380
500
600
900
1200
94
93
99
98
100
95
98
97
94
96
99
98
99
99
99
99
98
97
97
98
120
a
The enantiomeric excess (ee) was obtained by comparing the
determined specific optical rotation to the reported data.7
a
The S configurations are due to the Cahn-Ingold-Prelog
rules; no reversal of the absolute configuration on the chiral carbon
occurs. Calculated based on the starting material used. The
b
be described by S/C ) 66.7 mol/g (moles of the substrate/
gram of the pure enzyme protein).
reaction mixture collected from the column reactor contained the
cyanohydrin as a sole product, no starting material remained.
c-f The enantiomeric excesse (ee) was obtained by comparing the
determined specific optical rotations to the reported data.4c,5,7,8 For
details, see the Experimental Section.
Hydrocyanation of furanyl-2-carboxaldehyde performed
under the same conditions but with slower flow rate also
gave sastisfactory results (Table 2). The through-put is
calculated to be 1176 g/(L‚day) and S/C ) 40 mol/g.
In most cases, the resulting cyanohydrins are used as
intermediates for further reaction. The yields and ee
values shown in the above tables indicate that the
products have sufficient purity for further reactions. No
purification operation is needed.
following equation to be a:b:c:d:e ) 100:66:44:28:22.
Therefore, the flow rate for these individual substrates
was set in accordance with their relative reaction rates
as shown in Table 3.
In our experiments, it is found that higher flow rate
favors the enhancement of the enantiomeric excess but
compromises the chemical yields. Because different al-
dehydes undergo hydrocyanation in different reaction
rates, it is necessary to find an optimal flow rate suitable
for individual substrate to achieve a balance between the
yield and the enantiomeric excess. For this purpose, five
aldehydes were subjected to hydrocyanation in separate
flasks under the same conditions. Figure 2 shows the
time-course of the reaction.
flow rate (mL/min) ) 3 ×
relative reaction rate of aldehyde
relative reaction rate of benzaldehyde
In conclusion, by taking advantage of the unique
feature of the defatted almond meal in the micro-aqueous
medium,4 we are able to establish a continuous process
for asymmetric hydrocyanation with high yields and ee
values in high through-put. No purification operation is
needed in this process. Although the preparation of five
chiral cyanohydrins is described for illustration of the
efficiency and scope of the method, more cyanohydrins
The relative reaction rates for substrates (a, b, c, d,
and e, respectively) were calculated according to the
8252 J . Org. Chem., Vol. 67, No. 23, 2002

4-F lu or om a n d elon itr ile (2): clear oil; yield 92%; [R]20D +32.6
(c 6.8, CHCl3), ee 84% (lit.4c [R] +36.4 (c 6.38, CHCl3), ee 94.2%);
1H NMR (CDCl3) δ 3.39 (br, s, 1H, OH), 5.52 (s, 1H, CH), 7.13
(m, 2H, Ar-H), 7.49 (m, 2H, Ar-H) ppm; 19F NMR (CDCl3/TFA,
ppm) -33.93 (m, Ar-F).
can be prepared with the technique by adjusting the
column length, the reaction temperature, and the flow
rate.
Exp er im en ta l Section
(S)-(+)-2-Hyd r oxy-2-(2-fu r yl)a ceton itr ile (3): clear oil;
yield 100%; [R]21 +50.0 (c 1.60, CHCl3), ee >99% (lit.7 [R] +14
D
Ca u tion : HCN is extremely toxic and must be handled with
high caution. Safety instruction is given in the Material Safety
Data Sheet (MSDS) of HCN. For further safety information, see
the International Chemical Safety Card of HCN (ICSC0492),
available from http://www.cdc.gov/niosh/ipcsneng/neng0492.html.
Gen er a l. Almond kernels were soaked in distilled water for
2 h, peeled, air-dried, and then powdered in cold ethyl acetate
with a homogenizer. The powder was defatted by a further three
washes with ethyl acetete, filtered, and stored in a refrigerator.
After 1 g of sea sand was added to a glass column with a sintered
glass frit, the column was loaded with approximately 15 g of
the crude almond meal powder. Finally, 1 g of sea sand was
added to complete the preparation of the enzyme column. A
solution of 10 mmol of the substrate in 50 mL of IPE and a
solution of HCN (1.5 equiv)/IPE were mixed and the mixture
was pumped passing through the enzyme column under the
reaction conditions mentioned in the above tables. The eluate
was pooled and extracted with saturated aqueous ferric chloride
solution in a separatory funnel for several times to remove the
residual HCN until the brown color of ferric chloride solution
no longer changes to blue. The organic phase was worked up in
the conventional manner (washed with water, dried over anhy-
drous Na2SO4, and concentrated in vacuo) to give the product.
Ma n d elon itr ile (1): clear oil; yield 95%; [R]20D +47.5 (c 1.89,
CHCl3), ee >99% (lit.5 [R] +45 (c 1, CHCl3), ee >99%); 1H NMR
δ 3.70 (s, 1H, OH), 5.50 (s, 1H, CH), 7.50 (m, 5H, Ar-H).
(c 1.87, CHCl3), ee 79%); 1H NMR (CDCl3) δ 3.17 (br, s, 1H, OH),
5.55 (s, 1H, CH), 6.43 (dd, 1H, J 1 ) 1.9 Hz, J 2 ) 3 Hz, Ar-H),
6.61 (m, 1H, Ar-H), 7.49 (m, 1H, Ar-H) ppm.
(S)-(+)-2-Hyd r oxy-2-(2-th ien yl)a ceton itr ile (4): clear oil;
yield 70%; [R]21 +62.4 (c 2.3, CHCl3), ee >99% (lit.7 [R] +61.6
D
1
(c 1.6, CHCl3), ee 99%); H NMR (CDCl3) δ 3.0 (s, 1H, OH), 5.8
(s, 1H, CH), 6.9-7.4 (m, 3H, Ar-H).
(S)-(+)-2-H yd r oxy-2-(5-m et h yl-2-fu r yl)a cet on it r ile (5):
clear oil; yield 99%; [R]21 +60.1 (c 1.75, CHCl3), ee 97% (lit.8
D
[R] +61.0 (c 1.60, CHCl3), ee 99%); 1H NMR (CDCl3) δ 2.5 (s,
3H, CH3), 3.0 (s, 1H, OH), 5.5 (s, 1H, CH), 6.0-7.2 (m, 2H, Ar-
H) ppm.
Ack n ow led gm en t . This work was supported by
The Chinese Academy of Sciences (KJ 951A150605),
The Major Basic Research Development Program
(2000077506), and The National Nature Science Foun-
dation of China (29790125).
J O020432N
(7) (a) Effenberger, F.; Ziegler, T.; Fo¨ster, S. Angew. Chem., Int. Ed.
Engl. 1987, 26, 458. (b) Effenberger, F.; Eichhorn, J . Tetrahedron:
Asymmetry 1997, 8, 469.
(8) Zandbergen, P.; van der Linden, J .; Brussee, J .; van der Gen, A.
Synth. Commun. 1991, 21,1387.
J . Org. Chem, Vol. 67, No. 23, 2002 8253