Microwave-assisted conversion of d-glucose into lactic acid under solvent-free conditions

Article abstract of DOI:10.1039/b922286c

The microwave-assisted alkaline degradation of d-glucose with alumina supported potassium hydroxide towards its conversion into lactic acid is reported. An experimental design approach was used to optimize the variables involved in this transformation. The reaction succeeded in yielding 75C-% of lactic acid starting from d-glucose using 1.5 equiv of KOH at 180 °C. The optimal conditions were applied to d-fructose, d-mannose and d-sucrose.

Full text of DOI:10.1039/b922286c

View Article Online / Journal Homepage / Table of Contents for this issue
Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
PAPER
www.rsc.org/greenchem | Green Chemistry
Microwave-assisted conversion of D-glucose into lactic acid under
solvent-free conditions
Ge´raldine Epane,a Jean Claude Laguerre,a Anne Wadouachi*b and Delphine Marek*a
Received 28th October 2009, Accepted 26th November 2009
First published as an Advance Article on the web 29th January 2010
DOI: 10.1039/b922286c
The microwave-assisted alkaline degradation of D-glucose with alumina supported potassium
hydroxide towards its conversion into lactic acid is reported. An experimental design approach
was used to optimize the variables involved in this transformation. The reaction succeeded in
yielding 75C-% of lactic acid starting from D-glucose using 1.5 equiv of KOH at 180 ◦C. The
optimal conditions were applied to D-fructose, D-mannose and D-sucrose.
conditions such as power, time and concentration of alkali on the
Introduction
reaction of monosaccharide were investigated. An experimental
design was used to obtain optimal conditions for conversion of
D-glucose into lactic acid.
Lactic acid (2-hydroxypropanoic acid) is a feedstock with
a growing market. It is used in food and pharmaceutical
industries and has the potential to be used for the production of
biodegradable polymers, such as poly(lactic acid).1
Currently, lactic acid is produced by a fermentative route
starting from different sources of sugar, such as starch, sucrose,
lactose, sugarcane bagass or apple pomace.2 However, the
biotechnology process presents some disadvantages, such as
limited space–time yield and requires control and regulation
of the microorganism pool.
Alternative chemical processes exist starting from monosac-
charides treated in aqueous alkaline solution.3 These processes
are less constraining than the fermentative route. However, their
limits are the release of a broad excess of by-products and the
high concentration of alkaline used.
Recently, Onda et all studied the conversion of D-glucose
into lactic acid using noble metal supported catalyst with an
alkaline in aqueous solution. This method allowed the reduction
of the amount of complex by-products and lactic acid was
obtained with 57%-C yield. Nevertheless, the reaction requires
a broad alkali excess compared to the starting sugar (20 equiv
compared to sugar).4 In the last decades, microwave power has
taken an undeniable place in chemical laboratory practice as a
very effective and environmentally friendly process, particularly
when it is carried out in solvent-less conditions. Microwave
heating makes it convenient to perform reactions very efficiently
in dry media conditions, and thus within the frame of the
green chemistry concept. The advantages of using dry media
conditions provide reaction rate enhancements with different
selectivity compared to conventional conditions.5
Results and discussion
The CEM Discover MW allows for irradiation either by
assigning the power with continuous measurement of the
resulting temperature by an optic fiber captor or by assigning the
temperature with continuous adjustment of irradiation power.
In this study, we decided to employ the first mode of irradiation,
in solventless media with assigned power in order to obtain a
temperature profile. First, we investigated a microwave-assisted
protocol in solvent-free conditions with an experiment field
starting from the identification of the boundary values (i.e.,
minima and maxima) for three variables, specific power applied
SP (W g-1 of product), alkali concentration (equiv) and time
(min). We used the Box–Benkhen design to determine the
reaction conditions with the three factors which could have a
significant effect on the outcome of the reaction. The D-glucose
mixed with previously crushed alumina supported potassium
hydroxide was loaded in a glass tube and irradiated.
Optimization of reaction conditions
The execution of the experimental design yielded approximately
50% of lactic acid. The experimental matrix of the executed
Box–Benkhen design with the coded values (-1, 0, +1) is given
in Table 1. The corresponding real values are given in the Table 2.
We employed a full three-level quadratic design giving 15
experiments, which lead to the formation of up to 50% of lactic
acid according to reaction conditions. Experiment 5 represents
the alkaline degradation of glucose supported on alumina at
level-0 of potassium hydroxide level-0 of specific power and level-
0 for the contact time. This reaction provided approximately 50%
of lactic acid.
In this study, we present the microwave-assisted conversion of
D-glucose into lactic acid in dry media conditions using alumina
supported potassium hydroxide. The effect of experimental
aInstitut Polytechnique Lasalle Beauvais, 19 rue Pierre Waguet, BP
30313, 60026, Beauvais cedex. E-mail:
The identified quadratic model giving the lactic acid yield
according to the experimental conditions is given in eqn (1)
delphine.marek@lasalle-beauvais.fr
bLaboratoire des Glucides UMR CNRS 6219, 33 rue Saint-Leu, 80039,
Amiens cedex. E-mail: anne.wadouachi@u-picardie.fr; Fax: +33 3 22 82
75 27
Y = 31.25 + 8.875x1 - 1.375x2 + 6.5x3 - 3.5x1x2
2
2
2
- 7.25x1x3 - 4.75x2x3 - 13.624x1 - 6.125x2 + 8.625x3
(1)
502 | Green Chem., 2010, 12, 502–506
This journal is
The Royal Society of Chemistry 2010
©

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Table 1 Full Experimental values of the Box-Benkhen design
Lactic acid Total
Exp. x1
x2
x3
C-% yield mass/g Power/W Final T/◦C
1
2
3
4
5
6
7
8
-1 -1
+1 -1
-1 +1
+1 +1
0
0
0
0
0
0
24
6
16
49
0
8
8
16
32
24
49
31
20
41
20
41
30
24
37
24
37
31
52
199
68
280
230
62
220
174
244
214
117
280
165
280
218
12
12
10
10
10
10
10
10
8
12
8
12
10
0
-1
+1
-1
+1
0
0
0
0
0
0
0
-1
-1 33
+1 34
+1 38
9
10
11
12
13
14
15
0
39
0
-1 -1 28
+1 -1 33
-1 +1 44
+1 +1 30
0
0
Fig. 2 Contour plot of lactic acid yield from regression analysis of
experimental data at level-0 of alkali (equiv/initial glucose).
0
0
0
0
37
x1: microwave specific power SP (W g-1). x2: potassium hydroxide
(equiv/initial glucose). x3: contact time (min).
fixed at level-0 (i.e. 2 equiv/initial glucose) it is necessary to
work around level-0.5 of the specific power (i.e. 2.5 W g-1) and
level-1 for the time (i.e. 30 min) to produce 50% of lactic acid.
Plots at other levels of alkali and power are qualitatively similar
but the yields are lower. Analysis of variance indicates that the
response models are suitable for lactic acid yields. Indeed the
correlation coefficient (R2 = 0.85) indicated that the yield of
lactic acid depended on the factors studied.
Table 2 Lactic acid yields starting from glucose at 3.1 W g-1 with
1.5 equiv of alkali at different reaction times in comparison to predicted
experiment
Coded
Predicted
Observed
values Real values of C-% yield in C-% yield in % Conversion
of time time/min
lactic acid
Lactic acid
of glucose
In the experimental design, the maximum product yield was
obtained for a specific power of 3 W g-1, 2 equiv of potassium
hydroxide and 30 min reaction time conditions. Using these
conditions, lactic acid yielded 49%. However, However, this
result was obtained or accomplished along the boundary of
the time parameter (i.e. after 30 min of reaction) in the design.
Experimental predictions were investigated using the
quadratic model at different levels of specific power, concen-
tration of potassium hydroxide and time. It appears that by
working under 3.1 W g-1 and 1.5 equiv of alkali for 40 min, it
can be possible to produce 82% of lactic acid. For this reason,
additional experiments were carried out at different times while
specific power and alkali concentration were set at 3.1 W g-1 and
1.5 equiv, respectively, to determine if mathematical simulations
matched authentic experiments. The predicted values of the
response under these experimental conditions and the values
obtained experimentally are given in Table 2.
+1
+1.5
+2
30
35
40
48
63
82
49
59
75
99
99
99
Different response surfaces were calculated from eqn (1) by
setting alternatively one factor at a time at different levels (0, -1,
+1). Only level-0 of both the specific power (i.e. 3 W g-1) and the
concentration of KOH (i.e. 2 equiv/initial glucose) are shown in
Fig. 1 and 2, respectively.
We observed that optimal conditions for the production of
lactic acid based on these additional experiments occurred when
glucose was irradiated at 3.1 W g-1 in the presence of 1.5 equiv
of alkali for 40 min. Moreover, between 30 min and 40 min,
the predictions fit the experiments. But working beyond 40 min
proved to be harmful for the production of lactic acid.
Impact of alkali concentration and the temperature on reaction
conditions
Investigation of the experiments enabled us to study the effects
of the temperature and the quantity of alkali on the output
of the formed products. Fig. 3 demonstrated that lactic acid
formation was enhanced by temperatures ranging between 180
and 190 ◦C for 1.5 equiv of alkali compared to glucose. Under
these conditions, it was then possible to achieve more than 50%
of lactic acid.
Fig. 1 Contour plot of lactic acid yield from regression analysis of
experimental data at level-0 of the specific power.
As indicated by these plots, when the specific power was fixed
at level-0 (i.e. 3 W g-1), it was possible to produce 30% of lactic
acid at level-0 of KOH (i.e. 2 equiv/initial glucose) and level-1
for the time (i.e. 30 min). When the alkali concentration was
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 502–506 | 503
©

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Table 5 Lactic acid yields obtained from different carbohydrates with
optimized reaction conditions determined with D-glucose
Carbohydrates
% Conversion of glucose
%-C Lactic acid yield
D-Glucose
D-Mannose
D-Fructose
D-ribose
D-arabinose
D-Sucrose
99
99
99
99
99
99
75
41
36
43
35
23
Fig. 3 Effect of alkali concentration and the temperature on the lactic
acid yield.
Using the optimal conditions, we realized a reaction without
solid support. In comparison with the reaction including solid
support, the yield of lactic acid was four times less, as shown in
Table 3. Solid support is thus an important component in the
reaction.
In order to elucidate the importance of microwaves on these
reactions, similar experiments were realized using conventional
heating (Table 4).
Fig. 4 Temperature profile of glucose irradiation at continuous power
(28 W) during 40 min.
As shown in Table 4, the use of microwave irradiation at
fixed power is the best method for the production of lactic acid
under dry media conditions, the yield reached 75 C-%. Indeed,
when the reaction was accomplished in an oil bath, only 36% of
lactic acid was obtained. Proceeding with fixed temperature with
microwaves induced a lower yield in lactic acid, probably caused
by a reduction of the power during the reaction. In contrast,
executing the reaction at fixed power managed to stabilize the
temperature during the reaction (Fig. 4).
Similar studies have been conducted in the presence of
hydrotalcite-type catalysts to initiate the formation of lactic acid
from glucose. These catalysts have Brønsted-base sites and are
able to catalyze aldolization reaction.6 The reaction mechanism
of alkaline degradation of glucose into lactate has been described
as a reverse aldol condensation and studied using 13C-labeled
glucose.7
The optimized conditions were applied to other sugars such
D-mannose, D-fructose, D-ribose, D-arabinose and D-sucrose to
compare the production of lactic acid (Table 5).
The yields obtained are lower for other sugars under the
optimal conditions found for producing lactic acid from D-
glucose. This means that it is necessary to perform another
experimental designs for each type of sugar in order to maximize
the lactic acid yield.
Table 3 Effect of solid support on the degradation of glucose using
optimal conditionsa
C-% Lactic % Conversion
Time/min Final T/◦C acid yield
of glucose
Without solid 40
support
180
180
19
75
99
99
With solid
support
40
a Reactions were carried out with 11.1 mmol of D-glucose, 16.65 mmol
of KOH and 200% w/w of alumina at 3.1 W g-1 as specific power.
Conclusion
Solvent-free organic syntheses combined with supported reagent
and microwave irradiation is an effective process to obtain
lactic acid starting from sugar source. Ideal temperature for
the production of lactic acid is between 180 and 190 ◦C with
a specific power of 3.1 W g-1. Compared to classical heating,
MW accelerates the conversion rate of glucose and consequently
enhance the yield of lactic acid, when the optimal temperature
for this conversion is assigned, a lower yield in lactic acid is
obtained.
These results suggest that beneficial rate enhancements are
probably due to microwave effects, by comparing conventional
and microwave heating. Pilot scales for industrial applications
are being studied.
Table 4 Comparison between microwave and conventional heatinga
C-% Lactic
acid yield
% Conversion
of glucose
Time/min
Final T/◦C
Oil bath
MWb
40
40
40
180
180
180
36
75
21
99
99
99
MWc
a Reactions were carried out with 11.1 mmol of D-glucose, 16.65 mmol
of KOH and 200% w/w of alumina at 3.1 W g-1 as specific power. b With
assigned power (specific power: 3 W g-1). c With assigned temperature:
180 ◦C.
504 | Green Chem., 2010, 12, 502–506
This journal is
The Royal Society of Chemistry 2010
©

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
View Article Online
Table 6 Box–Benkhen design reaction conditions
Boundary values
Experimental
Material
Parameter
-1
0
1
D-Glucose anhydrous and KOH (90%) were purchased from
Verfilco and used without further purification.
x1: Microwave specific power SP (W g-1)
x2: Alkali concentration (equiv/initial
glucose)
2
1
3
2
4
3
˚
Neutral alumina (150 mesh, 58 A) was purchased from
Sigma–Aldrich. Microwave irradiation was performed in a CEM
x3: Time (min)
10
20
30
R
Discoverꢀ System. This system operates at a frequency of 2.45
GHz with continuous MW irradiation power up to 300 W.
The reactions are carried out in round bottom pyrex tubes
with a diameter of 38 mm and a height of 200 mm. Mechanical
agitation is used.
of the minima and maxima levels for all the retained factors
of the experimental design (Table 5). That makes it possible to
carry out the experiments under the conditions stipulated and to
correlate the answers of the design (here the yield in lactic acid)
to the experimental conditions (the factors of the experimental
design) thanks to the quadratic model presented below (eqn
Reaction procedure
Glucose (11.1 mmol) and potassium hydroxide (16.65 mmol)
were introduced into the reactor with alumina (200% w/w
compared to initial glucose and alkali). The mixture (total
mass of 9 g) was stirred for a minute then was irradiated by
MW at the desired power and time. A mechanical system of
agitation was used to homogenize the powder. After reaction,
a solution of H2SO4 0.2 M was added so as to obtain a pH of
approximately 2–3. The aqueous solution was separated from
alumina by filtration for analysis.
R
(4)).The linear regression function on Excelꢀ software was used
in order to identify the coefficients ai, aij and aii of this quadratic
model.
Y = a0 + a1x1 + a2x2 + a3x3 + a12x1x2 + a13x1x3 + a23x2x3
2
2
2
+ a11x1 + a22x2 + a33x3
(4)
Where:
Y is the response (lactic acid C-%),
a0 is the value of the response at the center point response of
the design,
ai, aii and aij are the coefficients of linear, quadratic and
interactive effects respectively.
Product analysis
The filtrate was analyzed by HPLC (Surveyor photodiode array
detector from ThermoFinnigan) with a UV detector (wavelength
220 nm). A Prevail Organic Acid column from Alltech (250 ¥
4.6 mm, 5 mm) was used with an aqueous solution of formic
acid 0.15% as eluant at 0.5 mL min-1 flow rate for 40 min.8 The
remaining sugar was quantified by HPLC analysis (Surveyor LC
pump from ThermoFinnigan) coupled with an evaporative light
scattering detector. A Prevail carbohydrates ES column, also
from Alltech (250 ¥ 4.6 mm, 5 mm), was used with a solution
of 75% acetonitrile in water as eluant at 1 mL min-1 flow rate
for 30 min. In addition, a guard column of the same packing
material was also used for both sugar and acid analysis.
Output in lactic acid and conversion of glucose were estimated
on a carbon percentage basis (C-%) from eqn (2) and (3) in the
following way:
In our study, three reaction conditions designated by x1, x2,
and x3 for specific power applied (W g-1 of product), alkali
concentration (equiv/initial sugar) and time (min), respectively,
were selected as being the experimental factors. For economic
reasons, we have chosen to work between a specific power range
of 2 W g-1 to 4 W g-1. The coefficients (ai, aij) in the model
equation were determined using the linear regression function
R
on Excelꢀ software. The determination of optimum conditions
and investigation of lactic acid output behaviour are illustrated
by contour plots produced from eqn (1).
Table 6 presents the boundary values of reaction conditions
in the experimental field.
(molar concentration of lactate formed×3)×100
C-% yield of lactic acid=
Acknowledgements
(molar concentration of introduced sugar×6)
The authors thank the “Industries and Agroresources” com-
petitiveness cluster and the “Conseil Re´gional de Picardie” for
financial support.
(2)
{1−(molar concentration of remained sugar)}×100
% glucose conversion =
(molar concentration of introduced sugar)
(3)
Notes and references
1 (a) T. B. VickRoy, Lactic acid in: Comprenhensive Biotehnology,
Moo-Young Ed., Pergamon press, 1985, pp761-776; (b) Y. B. Wee,
J. N. Kim and H. W. Ryu, Food Technol. Biotechnol., 2006, 44, 163;
(c) R. P. John, K. M. Nampoothiri and A. Pandey, Appl. Microbiol.
Biotechnol., 2007, 74, 524.
2 (a) B. Gullon, R. Yanez, J. L. Alonso and J-C Parajo, Bioresour.
Technol., 2008, 99, 308; (b) Y. J. Wee and H. W. Ryu, Bioresour.
Technol., 2009, 100, 4262; (c) R. P. John, G. S. Anisha, K. M.
Nampoothiri and A. Pandey, Biotechnol. Adv., 2009, 27, 145;
(d) M. G. Adsul, A. J. Varma and D. V. Gokhale, Green Chem.,
2007, 9, 58.
Optimization method
A response surface methodology (RSM) was used to find
optimum operating conditions for the formation of lactic acid.
Thus, a Box–Benkhen design, with three factors was chosen. The
execution of such a design is a rapid and effective technique for
finding the optimal operating conditions in order to improve the
results.9,10 The implementation of the design includes the setting
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 502–506 | 505
©