Resolution of Alcohol Racemate by Oil

doi:10.4236/gsc.2011.11002

Paper Menu >>

Journal Menu >>

Green and Sustainable Chemistry, 2011, 1, 7-11

doi:10.4236/gsc.2011.11002 Published Online February 2011 (http://www.SciRP.org/journal/gsc)

Resolution of Alcohol Racemate by Oil

Adi Wolfson*, Nadia Komyagina, Christina Dlugy, Janine Blumenfeld

Green Processes Centre, Chemical Engineering Department, Sami Shamoon College of Engineering,

Bialik/Basel Sts. Beer-Sheva, Israel

E-mail: [email protected]

Received January 7, 2011; revised January 28, 2011; accepted January 30, 2011

Abstract

Vegetable oil was successfully used as solvent and acyl donor in the kinetic resolution of several secondary

alcohol racemates yielding high enantioselectivities. Using vegetable oil as solvent and acyl donor allowed

easy separation of the pure alcohol enantiomer by extraction with methanol, by distillation under reduced

pressure, or by column chromatography.

Keywords: Oil, Kinetic Resolution, Lipase, Separation

1. Introduction

Optically pure chiral compounds are important building

blocks in the production of fine chemicals for pharma-

ceuticals, agrochemicals, and food ingredients [1,2]. Pure

enantiomers can be produced by using the “chiral pool”

of readily available natural compounds as starting mate-

rials, by resolving a racemic mixture, or by generating

asymmetry through asymmetric synthesis [3,4].

As starting materials, racemates can be obtained from

natural sources or by ‘classical’ chemical synthesis [1,2].

Since enantiomers exhibit different chemical activities in

a chiral environment, when a racemic mixture reacts with

a chiral (bio- or chemo-) catalyst, the different conversion

rates of the two enantiomers lead to the enrichment of

the less active enantiomer in the mixture. Other methods

of enriching enantiomers are preferential crystallization

and separation by chiral HPLC and chiral membrane.

However, these methodologies typically return low

theoretical yields (maximum 50%) of the pure product.

Chiral alcohols are useful intermediates and auxiliaries

in the production of various fine chemicals [5,6]. Pure

alcohol enantiomers can be produced by asymmetric

hydrogenation or transfer hydrogenation of the corres-

ponding carbonyl compounds using bio- and chemo-

catalysts [7-10] or by resolution of alcohol racemate

[1,2]. Lipase catalyzed kinetic resolution of alcohol ra-

cemate is the oldest and most commonly used method of

obtaining pure alcohol enantiomers [11-15]. Several

other methods-such as direct esterification using carbo-

xylic acid, transesterification using an ester as acyl donor,

and acylation by acyl chloride, anhydride, or vinyl alka-

noates-which differ in the type of acyl donor, can be used

for the resolution.

While water is the solvent of choice for most biocata-

lysis synthesis reactions, it cannot be used for alcohol

resolution, as it directs the reverse reaction, ester hy-

drolysis. Yet, as enzymes require no more than a

monolayer or so of water around them to preserve their

conformation and thus their activity, it can also be used

in an organic medium. It was found that the most hydro-

phobic solvents were the best for maintaining active en-

zymes mainly since the non-polar solvent cannot strip

the water from the protein and as such, does not change

its native structure [15]. The kinetic resolution of alco-

hols in an organic medium was previously reported with

several free and immobilized lipases. In those studies,

Candida antarctica lipase B (CAL-B) and its supported

analogues were mostly used [11-15].

As is typical of many organic reactions, in the kinetic

resolution of alcohols, the choice of the solvent is critical.

Besides activity and enantioselectivity, solvent selection

should also consider environmental aspects and product

separation procedure. We recently showed that using

glycerol triacetate (triacetin) as both solvent and acyl

donor in the kinetic resolution of alcohol resulted in high

conversions and enantioselectivities, and that the product

was easily separated by simple extraction with diethyl

ether [15]. Not only is triacetin environmentally friendly

and non-toxic, using it as acyl donor yielded by-products

that remained in the reaction mixture, and thus, only the

converted and unconverted enantiomers were separated.

8 A. WOLFSON ET AL.

In this paper we report on our study about the kinetic

resolution of several secondary alcohol racemates with im-

mobilized C. antarctica lipase B (CAL-B) in vegetable oil

(Figure 1). The oil was used not only as solvent, but also

as acyl donor, yielding pure (S)-alcohol enantiomer and

ester of (R)-alcohol enantiomer and fatty acid. The effects

of oil source and reaction condition on resolution activity

and enantioselectivity were studied together with separa-

tion techniques for (S)-alcohol enantiomer purification.

2. Experimental

All chemicals were purchased from Aldrich, except some

of the vegetable oils, which were bought at a supermarket.

2.1. Kinetic Resolution

In a typical reaction, 2.3 mmol of alcohol racemate were

added to 5 mL vegetable oil together with 0.043 g of C.

antarctica lipase B immobilized on acrylic resin. The

reaction mixture was heated in an oil bath to the required

temperature (80˚C) and mixed with stirrer for 5 h. At the

end of the reaction, the reaction mixture was cooled and

analyzed by TLC to determine alcohol conversion. In

addition, the products were extracted with 2 mL of

methanol for GC-analysis.

For the analytical procedure, the reaction mixture was

analyzed by GC using an Astec Chiraldex G-TA chiral

column (30 m × 0.25 mm, 0.25 μm thickness) to deter-

mine the reaction conversion and the enantiomeric ex-

cess. The injector temperature was 220˚C, and the FID

detector temperature was 250˚C. The temperature pro-

gram initialized at 80˚C for 5 min and then increased up

to 180˚C at a rate of 10˚C/min.

2.2. Alcohol Separation

Pure alcohol enantiomer separation was tested by several

techniques. First, kinetic resolution of 2-octanol was

performed by adding 0.305 g immobilized CAL-B and

16.1 mmol 2-octanol to 35 mL canola oil (purchased

from Aldrich). Then the reaction mixture was mixed in

an oil bath at 80˚C for 5 h, and the immobilized lipase

was filtrated from the reaction mixture at the end of the

reaction.

R1 R2

'/R

Immobilized CAL-B

(R)-Acetate

(S)-Alcohol

Alcohol

racemate

Figure 1. Kinetic resolution of a secondary alcohol race-

mate in oil.

Two methanol-based alcohol enantiomer extraction

procedures were used. In the first procedure, 150 mL of

methanol were added to the reaction mixture in a 250 mL

Erlenmeyer flask and mixed for 30 min, after which

phase separation was performed. The methanol was then

evaporated under reduced pressure in a Rotavapor. The

second procedure entailed a six-step extraction done us-

ing 25 mL of methanol for each step. The extracts from

all the steps were then joined and evaporated under re-

duced pressure in a Rotavapor. For both of the proce-

dures, the resultant crude was weighed and analyzed by

TLC to determine oil and methyl ester contents and by

GC to detect the total amount of (S)-2-octanol in the

crude. The extraction yield was calculated by dividing

the weight of the crude by half of the weight of the

2-octanol, and the purity was calculated by dividing the

total amount of (S)-2-octanol by half the weight of the

2-octanol.

Distillation of (S)-2-octanol from the reaction mixture

was tested under reduced pressure at 120˚C or 90˚C. At

the end of each process, the distillate was weighed and

analyzed by TLC and GC, and the total alcohol yield and

content were calculated as detailed above.

Column chromatography was run in a glass column

measuring 1 m long by 6 cm in diameter. A portion of 5 g

glass wood was fixed at the bottom of the column, and

above it was a 0.5 cm layer of sand. The column was filled

with silica gel as the stationery phase while dichloro-

methane was used as the mobile phase. The elution was

collected at the end of the column and analyzed by TLC to

determine when the oil and the methyl ester exited the

column relative to the alcohol enantiomer (the oil and the

methyl ester were eluted out before the alcohol). The frac-

tion of (S)-2-octanol was collected, and dichloromethane

was evaporated under reduced pressure in a Rotavapor. At

the end of the process, the crude was weighed and ana-

lyzed by TLC and GC, and the total alcohol yield and

content were calculated as detailed above.

3. Results and Discussion

The main motivation to perform the kinetic resolution of

alcohol racemate by oil was to test an alternative to the

toxic and expensive petroleum based organic solvents

usually employed for this purpose. Moreover, in addition

to its function as a solvent, the oil also acted as resolu-

tion agent, which resulted in two different enantiomers,

an alcohol and an ester of alcohol, and a fatty acid. As

the two different enantiomers have different molecular

structure and thus have different polarities, solubilities,

and boiling points in oil, the pure alcohol enantiomer can

be separated using techniques in which there is no need

to first separate the alcohol and ester enantiomers from

the reaction mixture before separating between them.

A. WOLFSON ET AL.

The investigation began by testing the performance of

immobilized CAL-B in the kinetic resolution of racemic

mixtures of 2-octanol as a representative compound us-

ing several different types of oil (e.g., sunflower, corn,

soybean, and canola oils) that were bought from a su-

permarket and some canola oil purchased from Aldrich

(Table 1). The reactions were run at 80˚C for 5 h in two

scales (2.3 mmol 2-alcohol in 5 mL oil and 16.1 mmol

2-alcohol in 35 mL oil). In all cases, full alcohol conver-

sion was achieved.

As illustrated in Table 1, in the kinetic resolution of

2-octanol, the source of the oil slightly affected the enan-

tiomeric excess (%ee) value of the resultant alcohol en-

antiomer (entries 1-4), and the highest %ee was observed

in canola oil. Cooking oil comprises mainly triglycerides,

but it also contains other chemicals, such as odors and

flavors, as well as antioxidants [17]. Therefore, from all

oil types, alcohol extraction also tuned up some unrec-

ognized organic compounds, which were also detected

when canola oil was bought from Aldrich. Replacing 2-

octanol by 2-butanol or 2-phenylethanol resulted in similar

conversions and enantiomeric excess values (entries

8-11). Performing the resolution with larger amounts of

secondary alcohol, oil, and enzyme yielded the same

results (entries 7 and 9).

Tests of the effect of reaction temperature on reaction

progress over time showed that the enantioselectivity of

2-octanol, which also depended on the reaction conver-

sion, increased with both the reaction temperature and

time, as expected. At temperatures below 60˚C, the reac-

tion progressed slowly with time, while at temperatures

above 60˚C, the reaction reached full conversion and

%ee after one hour (Figure 2). This effect may be attrib-

utable to an energy barrier that can be overcome faster at

higher temperatures and to the decrease, as the tempera-

ture increased, in the viscosity of oil, which served to

increase both mass and heat transfer.

The separation of pure alcohol enantiomer was tested

by several techniques after running the kinetic resolution

Table 1. Lipase catalyzed kinetic resolution of 2-alcohola.

Entry Oil Source Alcohol type ee, (S)

-alcohol (%)

1 Sunflower 2-Octanol 95

2 Corn 2-Octanol 97

3 Soybean 2-Octanol 93

4 Canola 2-Octanol 98

5 Canola (Aldrich) 2-Octanol 98

6b Canola (Aldrich) 2-Octanol 98

7 Canola (Aldrich) 2-Butanol 98

8 Canola (Aldrich) 2-Phenylethanol 99

9 b Canola (Aldrich) 2-Phenylethanol 99

aReaction conditions: 5 mL oil, 0.043 g immobilized CAL-B, 2.3 mmol 2-

alcohol, 80˚C, 5 h; bReaction conditions: 35 mL oil, 0.305 g immobilized

CAL-B, 16.1 mmol 2-alcohol, 80˚C, 5 h.

100

0510 1520 25

Tim e (h)

E nan tio meric e xcess (% )

Figure 2. Reaction progress over time at different temperatures.

Reaction conditions: 35 mL oil, 0.305 g immobilized lipase

CAL-B, 16.1 mmol 2-octanol. () 25˚C; (■) 40˚C; () 60˚C;

(▲) 80˚C.

of 2-octanol according to the procedure of Table 1 (entry

6) and filtration of the immobilized lipase from the reac-

tion mixture at the end of the reaction (Table 2). First,

extraction of the product was examined. As oil is a hy-

drophobic organic liquid, several hydrophilic extraction

solvents may be considered. Water, an oil immiscible

solvent that forms a biphasic system with oil but that

poorly dissolves most secondary alcohols, cannot be em-

ployed for the extraction. On the other hand, polar sol-

vents like acetonitrile and DMSO were dissolved in the

oil phase. It was found that methanol and vegetable oil

have a low mutual solubility, while methanol efficiently

dissolved secondary alcohols, identifying it as the most

suitable solvent for the extraction.

The extraction of (S)-2-octanol from the reaction

mixture by methanol was tested using two procedures.

In the first procedure, 150 mL of methanol were added

to the reaction mixture in a 250 mL Erlenmeyer flask

and mixed for 30 min followed by phase separation

(Table 2, entry 1). Then the methanol was evaporated

under reduced pressure in a Rotavapor, and the resul-

tant crude was weighed and analyzed in TLC and GC.

It was found that the overall yield, based on

(S)-2-octanol, was above 100%, as the extraction mix-

ture also contained some oil and small amounts of un-

characterized organic molecules. Based on GC analysis,

the percentage of (S)-2-octanol in the extraction mix-

ture was 89.7%, which implies that all the alcohol was

extracted from the reaction mixture. Performing the

extraction in six extraction steps using 25 mL of

methanol in each step resulted in a lower extraction

yield and a slightly higher purity of the crude after-

methanol evaporation, mainly as the oil content in the

extract decreased (entry 2).

A. WOLFSON ET AL.

Entry yield (%)

hol

purity (%)

Table 2. Separation of (S)-2-octanol from oila.

Separation technique Total

Alco

1 Extr

(150 mL methanol)

action-one step 115 89.7

2 Extraction-six step

(6  25 mL methanol)

4 89 95.2

109 92.4

3 Distillation-120˚C 150 67.7

Distillation-90˚C

5 Column-dichloromethane 72 99

aReon il, 0.305 g immobi CAL-B,

2-ocol, l.

rom the reaction mixture

as also tested (Table 2, entries 3 and 4). Because the

Ta-

ons well as solvent and acyl donor in

e kinetic resolution of secondary alcohol racemates.

, “Chirotechnology: Industrial Synthesis of

Optically Active Compounds,” Marcel Dekker, New

[2]

ommercial Manufacture and Application

Design of Novel Chirally Modified Platinum

and A.

ch to the Prepara-

ai, “An Efficient Synthesis of

l. 45, No. 11, 1989, pp. 3233-3298.

, M.

tal Complexes

xes,” Plenume Press, New York, 1983.

ence the Enantioselectivity

ectivity in

Kinetic

enium(II) and Lipase

acticonditions: 35 mL olized 16.1 mmol

tion of Enantiomerically Pure Chiral Building Blocks,”

Chemical Reviews, Vol. 92, No. 5, 1992, pp. 1071-1140.

tan 80˚C, 5 h; bBased on 2-octano

Distillation of (S)-2-octanol f

iling point of 2-octanol is relatively high, the distilla-

tions were done under reduced pressure. Performing the

distillation at a high temperature of 120˚C (entry 3) re-

sulted in large amounts of various organic compounds,

probably from the destruction of the oil molecules, and

therefore, the total yield was above 100%. Lowering the

distillation temperature to 90˚C produced lower yield of

(S)-2-octanol and a low organic compounds content.

Finally, column chromatography was also used for the

separation of the alcohol from the oil and the ester (

e 2, entry 5). Several solvents, such as ethyl acetate,

petroleum ether, or dichloromethane, alone or in mix-

tures and with methanol, were tested. It was found that

using dichloromethane produced the best difference be-

tween the retention factors (Rf) of the oil and ester mix-

ture and of (S)-2-octanol. Ethyl acetate had high elution

strength, which resulted in a high level of transport of all

materials through the column. When using petroleum

ether, their transport through the column was very slow.

Performing the column separation with dichloromethane

yielded 72% pure (S)-2-octanol (entry 5).

4. Conclusions

Vegetable oil functi

Novo

Various oil types can be used for the transesterification

step, yielding high enantioselectivities. Increasing either

reaction temperature or time increased the reaction rate.

Separation of the pure alcohol enantiomer from the reac-

tion mixture at the end of the reaction can be done by

extraction with methanol, distillation under reduced

pressure, or by column chromatography.

5. References

[1] R. A. Sheldon

Cata

York, 1993.

A. N. Collins, G. Sheldarke and J. Crosby, “Chirality in

Industry: The C

of Optically Active Compounds,” John Wiley, New York,

1995.

[3] A. Baiker, “Progress in Asymmetric Heterogeneous Ca-

talysis:

Metal Catalysts,” Journal of Molecular Catalysis A:

Chemical, Vol. 115, No. 3, 1997, pp. 473-493.

doi:10.1016/S1381-1169(96)00352-4

[4] E. Santaniello, P. Ferrabosc, P. Grisenti

Manzocchi, “The Biocatalytic Approa

doi:10.1021/cr00013a016

[5] D. M. Tsachen, L. M. Fuentes J. E. Lynch, W. L. Laswell,

R. P. Volante and I. Shink

4-benzoyloxyazetidinone: An Important Carbapenem In-

termediate,” Tetrahedron Letters, Vol. 29, No. 23, 1988,

pp. 2779-2782.

[6] K. Mori, “Synthesis of Optically Active Pheromones,”

Tetrahedron, Vo

doi:10.1016/S0040-4020(01)81007-3

[7] C. Dhenaut, I. Ledoux, I. D. W. Samuel, J. Zyss

Bourgault and H. Le Bozec, “Chiral Me

with Large Octupolar Optical Nonlinearities,” Nature,

Vol. 374, 1995, pp. 339-342. doi:10.1038/374339a0

[8] L. Poppe and L. Novak, “Selective Biocatalysis,” VCH,

Weinheim, 1992.

[9] L. H. Pignolet, “Homogeneous Catalysis with Metal

Phosphine Comple

[10] R. Noyori, “Asymmetric Catalysis in Organic Synthesis,”

John Wiley, New York, 1994.

[11] E. E. Jacobsen, L. S. Andresen and T. Anthonsen,

“Immobilization Does Not Influ

of CAL-B Catalyzed Kinetic Resolution of Secondary

Slcohols,” Tetrahedron: Asymmetry, Vol. 16, No. 4, 2005,

pp. 847-850. doi:10.1016/j.tetasy.2004.11.081

[12] E. E. Jacobsen, E. W. van Hellemond, A. R. Moen, L. C.

V. Prado and T. Anthonsen, “Enhanced Sel

zym 435 Catalyzed Kinetic Resolution of Secon-

dary Alcohols and Butanoates Caused by the

(R)-alcohols,” Tetrahedron Letters, Vol. 44, No. 46, 2003,

pp. 8453-8455. doi:10.1016/j.tetlet.2003.09.105

[13] N. Kim, S. B. Ko, M. S. Kwon, M. J. Kim and J. Park,

“Air-Stable Racemization Catalyst for Dynamic

Resolution of Secondary Alcohols at Room Tempe-

rature,” Organic Letters, Vol. 7, No. 20, 2005, pp.

4523-4526. doi:10.1021/ol051889x

[14] B. Martin-Matute, E. Michaela, K. Bogár, F. B. Kaynak

and J. E. Bäckvall, “Combined Ruth

lysis for Efficient Dynamic Kinetic Resolution of

Secondary Alcohols. Insight into the Racemization

Mechanism,” Journal of the American Chemical Society,

Vol. 127, No. 24, 2005, pp. 8817-8825.

doi:10.1021/ja051576x

A. WOLFSON ET AL.

lyse Glycer

acemates,” Bioprocess and

ticle Enzy

Oils,”

ty, Vol. 53,

[15] C. Dlugy and A. Wolfson, “Lipase Cata

for Kinetic Resolution of R

olysis

in Non-Conventional Phases,”

Biosystems Engineering, Vol. 30, No. 5, 2006, pp.

327-330. doi:10.1007/s00449-007-0128-x

[16] A. Ballesteros, U. Bornscheuer, A. Capewell, D. Combes,

J. S. Condoret and K. Koenig, “Review Armes N

Biocatalysis and Biotrans-

formation, Vol. 13, No. 1, 1995, pp. 1-42.

doi:10.3109/10242429509040103

[17] E. R. Sherwin, “Antioxidants for Vegetable

Journal of the American Oil Chemists’ Socie

o. 6, 1976, pp. 430-436. doi:10.1007/BF02605739