Optimization of the lipase mediated epoxidation of monoterpenes using the design of experiments—Taguchi method

Article abstract of DOI:10.1016/j.procbio.2016.07.005

This work deals with the optimization of the Candida antartica lipase B (CALB) mediated epoxidation of monoterpenes by using the design of experiments (DoE) working with the Taguchi Method. Epoxides are essential organic intermediates that find various industrial applications making the epoxidation one of the most investigated processes in chemical industry. As many as 8 parameters such as the reaction medium, carboxylic acid type, carboxylic acid concentration, temperature, monoterpene type, monoterpene concentration, hydrogen peroxide concentration and amount of lipase were optimized using as less as 18 runs in triplicates (54 runs). As a result, the hydrogen peroxide concentration used was found to be the most influential parameter of this process while the type of monoterpene was least influential. Scaling up of the reaction conditions according to the findings of the optimization achieved full conversion in less than 6?h. In addition, a purification process for the epoxides was developed leading to an isolated yield of ca. 72.3%, 88.8% and 62.5% for α-pinene, 3-carene and limonene, respectively.

Full text of DOI:10.1016/j.procbio.2016.07.005

Accepted Manuscript
Title: Optimization of the lipase mediated epoxidation of
monoterpenes using the design of experiments—Taguchi
method
Author: Sumanth Ranganathan Johannes Tebbe Lars O.
Wiemann Volker Sieber
PII:
S1359-5113(16)30248-3
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.procbio.2016.07.005
PRBI 10736
To appear in:
Process Biochemistry
Received date:
Revised date:
Accepted date:
12-5-2016
24-6-2016
5-7-2016
Please cite this article as: Ranganathan Sumanth, Tebbe Johannes, Wiemann
Lars O, Sieber Volker.Optimization of the lipase mediated epoxidation of
monoterpenes using the design of experiments—Taguchi method.Process Biochemistry
http://dx.doi.org/10.1016/j.procbio.2016.07.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Optimization of the lipase mediated epoxidation of monoterpenes using the design of
experiments – Taguchi method
Sumanth Ranganathan1, Johannes Tebbe2, Lars O Wiemann2,3, Volker Sieber1,2
1- Chair of Chemistry of Biogenic Resources, Technical University of Munich, Schulgasse
16, Straubing 94315, Germany
2- Fraunhofer Institute of Interfacial Engineering and Biotechnology (IGB) Bio-, Electro
and Chemo Catalysis BioCat Branch Straubing, Schulgasse 11a, Straubing 94315,
Germany
3- Current address: evocatal GmbH, Alfred-Nobel-Str. 10, Monheim am Rhein 40789,
Germany
Email: sieber@tum.de
Phone: +499421187300
Fax: +499421187310
1

Parameters and levels
identified for this process
Optimization performed
Optimized values tested in
reaction
Performance evaluated at
large scale

Highlights
 Traditional lipase mediated epoxidation was extended for monoterpenes
 Design of experiments – Taguchi method used to optimize the process
 Hydrogen peroxide was determined to be the most important parameter
 Simple and easy product purification protocol developed
2

ABSTRACT
This work deals with the optimization of the Candida antartica lipase B (CALB) mediated
epoxidation of monoterpenes by using the design of experiments (DoE) working with the
Taguchi Method. Epoxides are essential organic intermediates that find various industrial
applications making the epoxidation one of the most investigated processes in chemical
industry. As many as 8 parameters such as the reaction medium, carboxylic acid type,
carboxylic acid concentration, temperature, monoterpene type, monoterpene concentration,
hydrogen peroxide concentration and amount of lipase were optimized using as less as 18
runs in triplicates (54 runs). As a result, the hydrogen peroxide concentration used was found
to be the most influential parameter of this process while the type of monoterpene was least
influential. Scaling up of the reaction conditions according to the findings of the optimization
achieved full conversion in less than 6 hours. In addition, a purification process for the
epoxides was developed leading to an isolated yield of ca. 72.3%, 88.8% and 62.5 % for α-
pinene, 3-carene and limonene, respectively.
Keywords: epoxidation, lipase, design of experiments, Taguchi method, monoterpenes,
process optimization.
3

1. INTRODUCTION
Epoxides possess high polarities and ring strains making them a highly reactive species and
very useful building blocks in organic synthesis. They are predominantly synthesized with the
Prileschajew epoxidation method using peroxycarboxylic acids, that in turn attack the double
bonds of alkenes [1, 2]. Peroxycarboxylic acids are extremely reactive, possess high
oxidation potentials and are therefore recommended to be produced in-situ for safe operation
of the epoxidation process [3]. The most commonly used substance for Prileschajew
epoxidation is meta-chloroperbenzoic acid - a strong electrophile prone to detonation when
exposed to shocks in the environment. In addition to the explosive nature, these reactions
should be performed at a temperature range of 0-25°C [4].
Owing to the aforementioned operational hazards of using high amounts of this substance and
the subsequent cleaning steps involved thereafter, chemo-enzymatic in-situ generation of
peroxycarboxylic acids was developed by Fredrik Björkling and his co-workers in the early
1990s using lipases (glycerol ester hydrolases, E.C. 3.1.1.3). The process (Scheme 1[5]) was
the first of its kind and subsequent works have been carried out using this protocol [6-10]; to
name a few. Variations of this process have been reported by the works of Ankudey
et.al. [11], when they used ethyl acetate as the solvent and acid donor for the epoxidation
process. Another modification of the Bjökling process was carried out by Klass &
Warwel [12], where the researchers used dimethyl carbonate to epoxidize alkenes and carbon
dioxide was obtained as the by-product. In addition to this, Baeyer Villiger Oxidation has also
been done using the mechanism explained by Björkling and his co-workers [13-15].
Scheme 1 to be inserted here
4

Every process needs to be optimized for good yields and the process shown in Scheme 1 is no
exception. On optimizing this process at a small scale (laboratory and pilot), the industrial
production could be achieved with pure products being formed and less waste being
generated. The outcome of an experiment highly depends on the careful design of the
experimental process [16]. Generally, in the design of a statistically based experiment the first
step is the choice of the performance characteristic or the response variable, which will be
closely monitored. The second step is the identification of variables or factors that contribute
to this response variable, which will be studied. The next step is the choice of different
treatment stages or levels, at which these factors will be tested for individual experiments.
The final step is the identification of uncontrollable factors or noise factors that may influence
the process in any way [17]. The usage of statistical procedures follows the general principles
of randomization, replication and duplication to predict the actual behavior of a process.
Generally, Plackett-Burmann Design (PBD), Central Composite Design (CCD) and Box-
Benkhen Design (BBD) have already been used to optimize several processes.
The optimization of the above mentioned lipase mediated epoxidation of alkenes has already
been carried out with the traditional „alteration of one variable at a time‟ [18, 19] and also
using the response surface methodology approach [20, 21]. The disadvantage of the one
variable at a time approach is that it generates large amounts of samples and waste, is
extremely time consuming and also expensive. Although, the response surface methodology
system is advantageous in minimizing the number of trials and predicting interactions of the
variables used, the Taguchi method with orthogonal array design predicts a mean performance
characteristic value close to the target value, instead of just adhering to traditional limits,
which in turn improves the quality of the process/product [22] The present work deals
exclusively with the optimization of this lipase mediated epoxidation process for such
monoterpene substrates, esp. α-pinene, 3-carene and limonene using the Taguchi approach.
5

Once the process has been tested for these three substrates, the procedure will be expanded to
other terpenes and alkenes as well. Monoterpenes are simple plant products that are found
predominantly in essential oils, but also in waste streams of pulp and paper industries and are
widely used in the food, paint and pharmaceutical industries. Their oxygenated versions, viz.
monoterpene epoxides and the corresponding diols are building blocks and synthetic
intermediates [10]. Another important aspect to consider is that in classical chemical
epoxidation approach, various unwanted side products are generated [23]. Seven parameters
at three different settings and one parameter at two different settings were tested for obtaining
the maximum conversion. Scale-up of the optimized runs obtained from Taguchi method was
investigated and found out to comply with the results.
2. MATERIALS & METHODS
2.1 Introduction – Taguchi method of experimental design
Many of the industrial processes of today use the technique that was developed by Dr.
Genichi Taguchi [24]. The Taguchi method was developed on the foundations of robust
design introduced in the 1950s and 1960s. Robust design can be defined as “an engineering
methodology for improving productivity during research and development so that high-
quality products can be produced quickly and at low cost”. This method can be applied to a
range of problems and has already been used in the field of electronics, automotives,
photography and many others [25].
On designing a process based on robustness strategy, the following approach is to be
followed:
6

 Drafting of the P-diagram and classification of variables into noise (uncontrollable),
signal (input) and response (output) factors (Figure 1)
 Use of orthogonal arrays for gathering usable information about the control factors by
carrying out a minimal amount of experiments
 Determination of signal to noise ratio for determining the field quality through
laboratory experiments. Because, with a decreasing mean, the standard deviation also
decreases. The standard deviation cannot be reduced first and mean brought to the
target value [26]. Hence, the signal to noise ratio is used.
 Use of this ratio in the specified way (larger the better, minimal the better and nominal
the best) to determine the outcome of the process.
2.2 Determining the signal to noise ratio
There are many ways to define the signal to noise ratio. The three most important ones are
described below.
Nominal is best
Smaller the better
Larger the better
(
)
(
)
Where, is the mean of the data observed, is the variance calculated for y (observed data),
n is the number of observations.
Nominal is most suitable when the output value needs to be around a certain value, e.g. ratio
of nitric acid and hydrochloric acid in aqua regia mixture. Smaller the better is to be used
7

when an output characteristic needs to be minimized, e.g. electromagnetic radiations from
telecommunication equipment. Larger the better is to be used when a response needs to be
maximized without compromising the process reliability, e.g. yield of a certain chemical
process [25].
The approach of this method is primarily focused on determination of the optimal variable
settings of process, thus achieving improved performance, in addition to reducing variability
in the process with the help of orthogonal arrays [27]. The Taguchi method considers three
stages in the development of a process- system design, parameter design and tolerance design.
During the system design stage, the experimenter determines the basic configuration of the
process. In the parameter design stage, values specific to the system are assigned in a nominal
manner, so that the variability from uncontrollable variables (noise variables) is minimized.
Tolerance design is used to indicate the best tolerances for the selected parameters [28, 29]
Figure 1 to be inserted here
2.3 Chemicals
3-Carene, α-pinene was obtained from Sigma Aldrich Co. LLC. Toluene was purchased from
Chem Solute, Germany. Ethyl acetate and acetonitrile were purchased from Carl Roth,
Germany. Hydrogen peroxide (35%) was bought from Avantomaterials, Netherlands. Lipase
enzyme (CALB, 7500 TBU/g) for the reaction was purchased from Chiral Vision,
Netherlands (Order No: CALB-T2-150XL).
2.4 Reaction Conditions
The reaction was carried out in 1 mL gas chromatography vials. There were 8 parameters that
were to be optimized in this work and are described in (Table 1). The parameters and the
8

various levels listed below in Table 1 were later used in an orthogonal array for process
optimization using the Taguchi Method. Final run of the process using optimized parameters
was carried out in 100 mL round bottomed flasks. Temperature was controlled in an oil bath
using Heidolph Magnetic Stirrers (MR series) fitted with a Pt 1000 temperature sensor
purchased from Heidolph industries, Germany. The reaction contents were stirred using a
magnetic stirrer at 500 rpm.
Table 1 to be inserted here
2.5 Analytics
The analytics of the epoxidation process was monitored using gas chromatography coupled
with mass spectrometry. The gas chromatograph (GC-QP 2010, Shimadzu) was coupled with
an autoinjector (AOC-5000, Jain Compipal) and was fitted to a mass spectrometer (GC-MS-
QP2010 Plus, Shimadzu). The column that was used for measurements was a BPX5 column
(SGE Analytical Science, Australia) of 0.25 mm diameter with a thickness of 0.25 µm and a
total length of 30 m. Helium was used as the carrier gas and the temperature profile used for
the analysis is given below:
 Gas chromatography: Start at 60°C with a holding time of 1 min, increase to 170°C at
a rate of 10°C/min and finally increasing the temperature to a maximum of 270°C with
a holding time of 3 min at a rate of 70°C/min
 Mass spectrometer: Ion source was maintained at a temperature of 200°C and the
interface temperature was 250°C
The analysis was done using the software provided by Shimadzu, GC-MS Postrun analysis
and the mass to charge ratio (m/Q) of all compounds used in this work were compared to the
9

database of National Institute of Standard and technology (NIST) library-version 08. The
samples for analysis were prepared in ethyl acetate of LC-MS grade and to avoid the
saturation of the MS detector, a solvent cut at 3.9 min was implemented with the help of the
software. The different retention times for the monoterpenes and their corresponding epoxides
on using the above mentioned procedure is given below in Table 2.
Table 2 to be inserted here
The epoxides of alpha-pinene, 3-carene and limonene after purification were also analyzed
using proton (1H) Nuclear Magnetic Resonance (NMR). It was performed on a Bruker
Avance 400 (1H: 400.13 MHz, 13C: 101 MHz, T=300 K). The residual peak of the solvent
(δCDCl3: H7.26; C77.0) was used as the internal reference and chemical shifts have been
reported in δ [ppm]. All resonance multiplicities are designated as s (singlet), d (doublet),
t (triplet), m (multiplet) and coupling constants J in Hertz [Hz].
3. RESULTS & DISCUSSION
3.1 Choice of reaction medium for optimization
To test the best reaction medium for the chemo-enzymatic epoxidation process, all other
parameters except the reaction medium were kept constant and the best functioning system,
i.e. the reaction medium, was fed into the design. The following reaction conditions were used
for the initial screening of various solvents: Substrate– limonene, substrate amount– 100 mM,
hydrogen peroxide (35%) amount- 150 mM, acid type- octanoic acid at 50 mM, lipase– 30
mg, temperature- 60°C. The different solvents tested for this purpose and their logP values
10

[30] are given in Table 3. A reaction time of 24 h was used with sampling done at 1, 3 and 6 h
and the reaction was followed for the production of limonene epoxide.
Table 3 to be inserted here
Although all solvents yielded a conversion of more than 90%; acetonitrile,
methylcyclohexane, n-heptane, xylene and toluene had the maximum (i.e. full) conversion
after 24 h (Figure 2).
Figure 2 to be inserted here
In the work of Björkling et al. [5] in 1992, it was reported that toluene, xylene and
nitromethane were ideal for high yields. However, in order to choose the two best solvents,
the conversion of those that generated 100% conversion was analyzed in more detail in this
work (Figure 3). Acetonitrile and toluene were shown to have the best conversion after 3 h,
complying only partially with the findings of Björkling et al. Hence, these two solvents were
chosen to be fed into the Taguchi design. Björkling et al. had also shown high concentrations
of hydrogen peroxide being important for the formation of the peroxycarboxylic acid. This
information was helpful in choosing the hydrogen peroxide concentration (levels) for the
Taguchi design that was appropriate for the process described in this work.
Figure 3 to be inserted here
3.2 Results of the Taguchi Method
The first step in the Taguchi method as mentioned in Section 2 was the identification of
controllable and uncontrollable factors. The controllable factors are given in the materials and
methods section (Table 2), while the uncontrollable factors could be attributed to water
content in the immobilized lipase, type of immobilization of the lipase etc. The reactions were
11

carried out according to the Taguchi method‟s orthogonal array design and were selected
using the software Minitab v 17.0.
For 1 parameter at 2 levels and 7 parameters at 3 levels, the array selector suggested a L18
array, which means 18 experiments were to be carried out in triplicates to obtain the optimum
settings for the process (Table 4).
Table 4 to be inserted here
The experiments were done in the same sequence as described in the table and were repeated
three times; at the end of which conversion was monitored for each. Conversion values for the
different set of experiments obtained is shown in Figure 4.
Figure 4 to be inserted here
The main effects for the various parameters and levels are given below. The conversion
values were used to calculate the signal to noise ratios. The characteristic used here was the
“larger the best”, and the signal to noise ratios obtained is given in Table 5. The signal to
noise ratio and the various ranks determined for the process were done using the software
“Minitab® 17.1.0” and are shown in Figure 5.
Table 5 to be inserted here
Rank 1 implies maximum impact on the process and rank 8 implies minimum impact of the
process. So, from the above table, it can be concluded that the maximum impact on the
process in decreasing order is: hydrogen peroxide concentration > substrate concentration >
type of carboxylic acid used > temperature > enzyme amount > carboxylic acid amount >
reaction medium used > substrate type used.
Figure 5 to be inserted here
12

When choosing the right level for each of the parameter, the one with the highest value needs
to be chosen; as it affects the process and maintain it at the maximum production efficiency.
On using this concept, the final conditions for testing at optimum conditions would be:
Hydrogen peroxide concentration-500 mM, substrate concentration-100 mM, substrate type-
limonene, temperature- 40°C, carboxylic acid type- octanoic acid (C8), enzyme amount-
60 mg, solvent type-toluene and carboxylic acid concentration- 70 mM. The test run with the
optimized conditions was done using these parameters and levels but in a scaled up fashion.
If a full factorial experiment were to be carried out with the same amount of parameters, then
more than 4000 experiments were to be done, and on making it three times to minimize the
error the numbers keep increasing and so does the costs and the amount of waste. Hence, the
design of experiments using the Taguchi method could be considered as a great way to extract
information on process optimization by reducing costs and by making the process more robust
even when considering variations in the operating conditions with regard to uncontrollable
factors.
3.3 Scale-up based on the results of the Taguchi method
The scale up was done for the optimized reaction conditions in a volume of 100 mL and based
on the results from the Taguchi method explained in the previous section. In order to test the
variability of the process, all three monoterpenes were used in the epoxidation process. The
concentration of limonene, 3-carene and α-pinene was 200 mM. The conversions obtained
with the optimized parameters along with changed monoterpene concentrations are shown in
Figure 6.
Figure 6 to be inserted here
13

It can be inferred that the process can be carried out with monoterpenes at 200 mM
concentration as well. At the end of the reaction, hydrogen peroxide concentration was
checked for all three processes and residual amounts could be discovered.
In order to test for the adaptability of the process further, 3-carene (300 mM) was tested using
the same set of optimized values. The results showed incomplete conversion (up to 70%) to
its corresponding epoxide and the reaction was left to run for another 16 h. In the end, there
was no characteristic improvement in the conversion; even though hydrogen peroxide was
detectable in the medium. A possible explanation could be the inactivation of the enzyme,
however, a reusability test with the same enzyme showed a complete conversion of starting
material (Results not shown). The exact reason for this phenomenon is still not known and
needs to be investigated further. Furthermore, control reactions with no enzyme, acid or
hydrogen peroxide showed no conversion after 16 h reaction time.
3.4 Purification of Monoterpene epoxide
The scaled up process of monoterpene epoxidation was then subjected to a purification step.
According to Arata & Tanabe [31], the epoxides of terpenes are highly sensitive compounds
in a basic medium. On using a strong base such as 10 M sodium hydroxide solution, they
immediately undergo ring opening to form diols. Hence a weaker base such as sodium
bicarbonate (saturated amounts), was used for more than 5 to 7 times to completely neutralize
the residual acid concentration that was used in this process, viz. the octanoic acid (C8) as a
sodium salt. On developing this process to industrial efficiency, this salt could be used as a
valuable by-product of the process. The exact procedure of carrying out this purification step
is shown in Figure 7.
Figure 7 to be inserted here
14

The isolated yields of the whole process of epoxidation are 72.3% for alpha pinene, 88.8% for
3-carene and a combined isolated yield (mono and di-epoxides) of 62.5% for limonene. In the
case of limonene, the ratio of mono to di-epoxide was 80: 20 (%). The ratio of the mono-
epoxide isomers was 55 % cis and 45% trans. For limonene di-epoxide, 4 different
diastereomers could be obtained theoretically. The ratio of these four predicted from GC-MS
software Postrun analysis, is: 4, 40, 19 and 37%. Although all three reactions yielded a 100%
conversion (GC-MS and NMR spectra attached as supplementary information), the
subsequent steps involving neutralization of the octanoic acid with saturated sodium
bicarbonate as well as non-optimized manual handling led to the loss of some product as well.
4. CONCLUSION
The process conditions of the lipase mediated epoxidation of monoterpenes were optimized
successfully in the tested range using Taguchi method of robust design. A total of 8
parameters (1 parameter at 2 levels and 7 parameters at 3 levels) were successfully optimized
in this research. However, a point of concern is that interactions were not accounted for when
using the Taguchi method. Nevertheless, the efficiency of the system was tested for volumes
of up to 100 mL and was found to comply with the findings of optimization. With our
optimization we were able to reach full conversion of substrates after 4 – 6 h, compared to the
6 - 24 h required in the process by Björkling et al., which even showed incomplete conversion
in certain cases. A simple and efficient purification method for the epoxides was developed
and carried out using the two phase extraction setup using a weak base such as saturated
amounts of sodium bicarbonate. Though it can be argued that a stronger base such as sodium
or potassium hydroxide needs to be used, the consequences of the epoxide ring being opened
to a diol cannot be overlooked, even if it had to be at the expense of product(s) loss. The
purification system is to be tested for other alkene epoxides and the monoterpene epoxides as
well. The idea of this optimization procedure was to make the product robust in such a way
15

that by changing the substrate alone, the process could be used to produce its corresponding
epoxide.
CONFLICT OF INTEREST
The authors wish to declare no financial or commercial conflict of interest.
16

List of Figures
Scheme 1: Lipase mediated epoxidation of monoterpenes using peroxycarboxylic acids
according to the method of Björkling et.al, 1992. (1 – α-pinene, 2- 3-carene and 3-
limonene; 1a- α-pinene epoxide, 2a- 3-carene epoxide, 3a and b- limonene monoepxoide
and 3c- limonene diepoxide)
17

Figure 1: Parameter diagram for product/process system
Figure 2: Results of solvent screening for the conversion of monoterpenes to
monoterpene epoxide over a period of 24 h.
18

Figure 3: Kinetics of limonene conversion over a time period of 6 h using acetonitrile
(square), methylcyclohexane (circle), n-heptane (triangle), xylene (inverted triangle)
and toluene (diamond)
Figure 4: Conversion obtained by trying out different formulations of parameters and
levels as suggested by the orthogonal array design (L18).
19

Figure 5: Signal to noise ratios for various parameters and levels tested using the L18
orthogonal array and "larger the better" characteristic. Dashed line implies the mean
signal to noise ratio for all 18 trials. (A-H parameters tested at different levels 1, 2 and 3
(Table 1))
20

Figure 6: Conversion profile of limonene-200 mM (triangle), alpha-pinene-200 mM
(square) and 3-carene-200 mM (sphere) using optimized parameters from Taguchi
method. (Note: Conversion for the limonene reaction refers to the epoxidation of the
molecule as such with two olefin bonds being epoxidized.)
21

Figure 7: Epoxide purification process after the use of optimized lipase mediated
epoxidation systems
22

List of Tables
Table 1: List of parameters and the levels for the optimization of lipase mediated
epoxidation of monoterpenes
Parameter
Level 1
toluene
octanoic acid
(C8)
Level 2
Level 3
-
A
B
Reaction Medium
Carboxylic acid type
Acetonitrile
decanoic acid
(C10)
lauric acid
(C12)
C
Carboxylic acid
concentration
Temperature
30 mM
50 mM
70 mM
D
E
F
20°C
40°C
60°C
Monoterpene type
Monoterpene
limonene
100 mM
3-carene
200mM
α-pinene
300 mM
concentration
Hydrogen peroxide
concentration
Lipase amount
G
H
100 mM
20 mg
300 mM
40 mg
500 mM
60 mg
Table 2: Retention times of the various monoterpenes and their subsequent epoxides
obtained by GC-MS analysis
S.No Compound
Retention Time (min)
1
2
Limonene
5.735
7.272
Limonene epoxide
23

3
4
5
6
7
Limonene-diepoxide 9.554
3-carene
5.809
7.945
4.392
6.802
3-carene epoxide
α-pinene
α-pinene epoxide
Table 3: Different solvents tested and their log P values according to [32].
S.No. Solvent
Log P
0.89
-0.34
3.44
3.88
1,26
4.66
1
2
3
4
5
6
7
8
2-methyl-2-butanol
Acetonitrile
Cyclohexane
Methylcyclohexane
Methyl tetrahydrofuran
n-heptane
Xylene (isomeric mixture) 3.12 – 3.2
Toluene 2.73
Table 4: L18 orthogonal array for the optimization of 1 parameter at 2 levels and 7
parameters at 3 levels for the optimization of lipase mediated epoxidation of
monoterpenes (For detailed account of the parameters and levels, please refer Table 1)
Trial. Solvent Acid Acid Temperature Substrate Sub.
H2O2 Enzyme
No.
type concn.
Concn. concn. amount
1
1
1
1
1
1
1
1
1
24

2
3
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
1
1
2
2
2
3
3
3
1
1
1
2
2
2
3
3
3
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
2
3
1
2
3
2
3
1
3
1
2
2
3
1
3
1
2
2
3
2
3
1
1
2
3
3
1
2
3
1
2
2
3
1
2
3
2
3
1
3
1
2
2
3
1
1
2
3
3
1
2
2
3
3
1
2
2
3
1
2
3
1
3
1
2
1
2
3
2
3
3
1
2
3
1
2
1
2
3
2
3
1
2
3
1
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Table 5: S/N ratio for different trials calculated using the "larger the better"
characteristic. A-H all parameters tested (Table 1) (Values obtained from Minitab 17.0
software)
Level
A
B
C
D
E
F
G
H
-1.95
-3.16
1
-2.13* -0.83* -2.66
-2.50 -2.76 -2.60
-2.06
-2.21* -0.73* -6.32
2
-1.69* -2.33 -2.54 -0.53
25

3
-
-3.36
2.54
3
-1.70* -3.20
-2.41
0.20
8
-3.68
2.95
2
-0.09* -1.84*
Delta
Rank
*
0.37
0.96
1.51
6.23
1.32
7
6
4
1
5
Optimized set of values for the process
5. REFERENCES
[1] G. Sienel, R. Rieth, K.T. Rowbottom, Epoxides, Ullmann's Encyclopedia of Industrial
Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA 2000.
[2] Z. Wang, Prilezhaev Reaction, Comprehensive Organic Name Reactions and Reagents,
John Wiley & Sons, Inc. 2010.
[3] H. Klenk, P.H. Götz, R. Siegmeier, W. Mayr, Peroxy Compounds, Organic, Ullmann's
Encyclopedia of Industrial Chemistry, Wiley-VCH Verlag GmbH & Co. KGaA 2000.
[4] A.S. Rao, H.R. Mohan, A. Charette, m-Chloroperbenzoic Acid, Encyclopedia of Reagents
for Organic Synthesis, John Wiley & Sons, Ltd 2001.
[5] F. Björkling, H. Frykman, S.E. Godtfredsen, O. Kirk, Lipase catalyzed synthesis of
peroxycarboxylic acids and lipase mediated oxidations, Tetrahedron 48 (22) (1992) 4587-
4592.
[6] V. Skouridou, H. Stamatis, F.N. Kolisis, Lipase-mediated epoxidation of α-pinene, Journal
of Molecular Catalysis B: Enzymatic 21 (1–2) (2003) 67-69.
[7] O. Wiemann Lars, C. Faltl, V. Sieber, Lipase-mediated Epoxidation of the Cyclic
Monoterpene Limonene to Limonene Oxide and Limonene Dioxide, Zeitschrift für
Naturforschung B, 2012, p. 1056 - 1060.
[8] C. Orellana-Coca, S. Camocho, D. Adlercreutz, B. Mattiasson, R. Hatti-Kaul, Chemo-
enzymatic epoxidation of linoleic acid: Parameters influencing the reaction, European Journal
of Lipid Science and Technology 107 (12) (2005) 864-870.
26

[9] S. Warwel, M. Rüsch gen. Klaas, Chemo-enzymatic epoxidation of unsaturated carboxylic
acids, Journal of Molecular Catalysis B: Enzymatic 1(1) (1995) 29-35.
[10] V. Skouridou, H. Stamatis, F.N. Kolisis, A Study on the Process of Lipase-catalyzed
Synthesis of α-pinene Oxide in Organic Solvents, Biocatalysis and Biotransformation 21(6)
(2003) 285-290.
[11] E.G. Ankudey, H.F. Olivo, T.L. Peeples, Lipase-mediated epoxidation utilizing urea-
hydrogen peroxide in ethyl acetate, Green Chemistry 8(10) (2006) 923-926.
[12] M. Rüsch gen. Klaas, S. Warwel, Chemoenzymatic Epoxidation of Alkenes by Dimethyl
Carbonate and Hydrogen Peroxide, Organic Letters 1(7) (1999) 1025-1026.
[13] P. Carlqvist, R. Eklund, K. Hult, T. Brinck, Rational design of a lipase to accommodate
catalysis of Baeyer–Villiger oxidation with hydrogen peroxide, Journal of Molecular
Modeling 9(3) (2003) 164-171.
[14] B. K. Pchelka, M. Gelo-Pujic, E. Guibe-Jampel, Chemoenzymatic autocatalytic Baeyer-
Villiger oxidation, Journal of the Chemical Society, Perkin Transactions 1 (17) (1998) 2625-
2628.
[15] S.C. Lemoult, P.F. Richardson, S.M. Roberts, Lipase-catalysed Baeyer-Villiger
reactions, Journal of the Chemical Society, Perkin Transactions 1 (2) (1995) 89-91.
[16] S.M. Bhat, A. Prabhakar, R.R.K. Rao, G.M. Madhu, Optimization of Process Variables
for Amylase Production Using Agro Residual Wastes by Taguchi Methodology, Chemical
Product and Process Modeling, 2010.
[17] R.S. Rao, C.G. Kumar, R.S. Prakasham, P.J. Hobbs, The Taguchi methodology as a
statistical tool for biotechnological applications: A critical appraisal, Biotechnology Journal
3(4) (2008) 510-523.
[18] E. Abdulmalek, M. Arumugam, M. Basri, M.B.A. Rahman, Optimization of Lipase-
Mediated Synthesis of 1-Nonene Oxide Using Phenylacetic Acid and Hydrogen Peroxide,
International Journal of Molecular Sciences 13(10) (2012) 13140-13149.
27

[19] M. Alves Moreira, T. Bergler Bitencourt, M. da Graça Nascimento, Optimization of
Chemo‐Enzymatic Epoxidation of Cyclohexene Mediated by Lipases, Synthetic
Communications 35(15) (2005) 2107-2114.
[20] B.M. Abdullah, N. Salih, J. Salimon, Optimization of the chemoenzymatic mono-
epoxidation of linoleic acid using D-optimal design, Journal of Saudi Chemical Society 18(3)
(2014) 276-287.
[21] S. Sun, X. Ke, L. Cui, G. Yang, Y. Bi, F. Song, X. Xu, Enzymatic epoxidation of
Sapindus mukorossi seed oil by perstearic acid optimized using response surface
methodology, Industrial Crops and Products 33(3) (2011) 676-682.
[22] S.P. Kondapalli, S.R. Chalamalasetti, N.R. Damera, Application of Taguchi based
Design of Experiments to Fusion Arc Weld Processes: A Review, 2013.
[23] K. John, Epoxidation process, Google Patents, 1967.
[24] R.N. Kacker, E.S. Lagergren, J.J. Filliben, Taguchi‟s orthogonal arrays are classical
designs of experiments, Journal of research of the National Institute of Standards and
Technology 96(5) (1991) 577-591.
[25] M.S. Phadke, Quality Engineering Using Robust Design, Prentice Hall PTR1995.
[26] J.A. Ghani, I.A. Choudhury, H.H. Hassan, Application of Taguchi method in the
optimization of end milling parameters, Journal of Materials Processing Technology 145(1)
(2004) 84-92.
[27] J. Antony, F.J. Antony, Teaching the Taguchi method to industrial engineers, Work
Study 50(4) (2001) 141-149.
[28] G.P. Syrcos, Die casting process optimization using Taguchi methods, Journal of
Materials Processing Technology 135(1) (2003) 68-74.
[29] R.S. Rao, R.S. Prakasham, K.K. Prasad, S. Rajesham, P.N. Sarma, L.V. Rao, Xylitol
production by Candida sp.: parameter optimization using Taguchi approach, Process
Biochemistry 39(8) (2004) 951-956.
28