http://www.paper.edu.

cn

J. Microbiol. Biotechnol. (2005), 15(1), 4047

Modeling and Simulation of Simultaneous Saccharification and Fermentation

of Paper Mill Sludge to Lactic Acid
1

LIN, JIAN-QIANG , SANG-MOK LEE , AND YOON-MO KOO

State Key Lab of Microbial Technology, School of Life Science, Shandong University, Jinan 250100, China
Department of Biological Engineering, ERC for Advanced Bioseparation Technology, Inha University, Incheon 402-751, Korea

Received: February 2, 2004

Accepted: May 11, 2004

Abstract Modeling and simulation for simultaneous

saccharification and fermentation (SSF) process in bioconversion
of paper mill sludge to lactic acid was carried out. The SSF
process combined the enzymatic hydrolysis of paper mill
sludge into glucose and the fermentation of glucose into lactic
acid in one reactor. A mathematical modeling for cellulose
hydrolysis was developed, based on the proposed mechanism
of cellulase adsorption deactivation. Another model for
simple lactic acid fermentation was also made. A whole
mathematical model for SSF was developed by combining
the above two models for cellulose hydrolysis and lactic acid
fermentation. The characteristics of the SSF process were
investigated using the mathematical model.
Key words: Cellulase, cellulose, paper mill sludge, lactic
acid, modeling

Lactic acid has wide applications in food and pharmaceutical

industries. In addition, the potential use as the source for
polylactate polymers in manufacturing biodegradable plastics
forms even a bigger market for lactic acid. About half of
the world production of lactic acid is made from fermentation,
while the remainder is produced synthetically [19]. Utilization
of industry wastes, like paper mill sludge, as the substrate
for fermentation of lactic acid makes the production
process more attractive in both environmental preservation
and production cost reduction [5, 11, 13].
About 106 tons of paper mill sludge are produced in
Korea in each year. The dried materials comprising 40% of
this sludge are composed of 30 to 60% of cellulose, which
can be hydrolyzed enzymatically to produce glucose. The
glucose produced can be fermented to produce lactic acid.
*Corresponding author
Phone: 82-32-860-7513; Fax: 82-32-872-4046;
E-mail: [email protected]

Compared with the natural cellulosic materials like wood

[16] and rice straws [7, 8], paper mill sludge needs no
pretreatment before enzymatic hydrolysis, reducing the
production cost. It has already been treated for removing
lignin and hemicellulose in industry and is much more
susceptible towards enzymatic hydrolysis [13].
The major hindrance for the commercial application of
bioconversion of paper mill sludge to lactic acid was the
high cost of enzymes. Thermal deactivation was ever
considered a main reason [2, 3]. However, cellulase is not
easily deactivated in the optimal enzymatic reaction condition
of 50oC, pH 4.8, and moderate agitation. Cellulase needs to
adsorb to cellulose molecules in order to react with the
solid substrate. The cellulase-cellulose adsorption complex
can be changed into an irreversible adsorption complex,
which makes the cellulase fixed to a certain place of the
surface of cellulose molecule, unable to reach new substrates
and hence becomes apparently deactivated. The irreversible
adsorption of cellulase to cellulose should be the main
reason for the high cellulase consumption. Glucose and
cellobiose inhibitory effects may be another reason for the
high enzyme consumption. SSF can relieve the inhibitory
effects by decreasing glucose and cellobiose concentrations
and is a good choice for bioconversion of paper mill sludge
to lactic acid [1].
With several enzymatic and one microbial reactions going
on simultaneously, SSF is a relatively sophisticated process.
Modeling and simulation are a powerful tool in mechanism
investigation, feasibility evaluation, and process design and
optimization. In this research, a mathematical model for
the SSF process to bioconvert paper mill sludge into lactic
acid was developed. It comprises simple cellulose hydrolysis
based on the mechanism of cellulase adsorption deactivation
and simple lactic acid fermentation. The characteristics of
SSF were investigated by simulation, using the mathematical
model.

http://www.paper.edu.cn

BIOCONVERSION OF PAPER MILL SLUDGE TO LACTIC ACID

MATERIALS AND METHODS

Strains, Medium, and Enzymes
Raw cellulase powder [18] was produced by Trichoderma
reecei Rut C-30 (ATCC 56765). -Glucosidase used was
the commercial product of Novozym 188 (Novo Nordisk,
Denmark). Cellulose powder of type 101-F (Sigma, U.S.A.)
was used in the cellulose hydrolysis experiment. Paper
mill sludge (sludge for short) was obtained from Samwha
Paper Co., Korea. Lactobacillus rhamnosus (ATCC 10863)
producing L-lactic acid was used. A medium containing
sodium acetate 1 g/l, K2HPO4 0.3 g/l, KH2PO4 0.15 g/l,
MgSO47H2O 0.15 g/l, and glucose 85 g/l was used for
lactic acid fermentation. In the SSF process, paper mill
sludge was used instead of glucose.
Operation Methods
Cellulose hydrolysis was done in a 300-ml flask containing
100 ml of the mixture of cellulose power at the final
concentration of 45 g/l with Na-citrate buffer (pH 4.8).
After autoclaving at 121oC for 30 min and cooling down
to room temperature, cellulase, measured by CMCase
activity, and -glucosidase were added to the final
activities of 5 and 2 IU/ml, respectively. Oxymycin was
added to the final concentration of 25 mg/ml, as an
antibiotic for prevention of contamination. The flasks were
incubated at 42oC and shaken at 150 rpm in a shaking
water bath (KMC1205KW1, KMC Vision Co., Korea).
Samples were taken for analysis of cellulose, cellobiose,
and glucose.
Flask lactic acid fermentation was done using flasks
of 100 ml, containing 50 ml of medium with 5% of
inoculation. It was cultivated at 42oC, shaking at 150 rpm
in a rotary shaking incubator of Model KMC-8480 SF
(Vision Scientific Co., Korea). Four percent CaCO3 were
added in the medium for buffering pH.
Fermenter lactic acid fermentation was done using a 3-l
fermenter (KFC, Korea) containing 1 l of medium with
inoculation volume of 5%, cultivated at 42oC and agitated
at 90 rpm. pH was maintained at 5.0 using 5 M NaOH
solution.
SSF for lactic acid production from sludge was operated
in fed-batch mode in a 3-l fermenter (KFC, Korea),
containing 1 l of medium with initial sludge concentration
of 15 g/l, inoculation volume of 5%, initial cellulase (measured
by CMCase activity) and -glucosidase activities of 5 and
2 IU/ml, respectively, cultivated at 42oC and agitated at
90 rpm. Sludge was fed five times at 8, 14, 20, 26, and
32 h, respectively, to increase 15 g/l sludge concentration
in every feeding. Cellulase and -glucosidase were fed two
times at 14 and 26 h, respectively, to increase CMCase and
-glucosidase activities of 5 and 2 IU/ml, respectively, in
every enzyme feeding. pH was maintained at 5.0 using
5 M NaOH solution.

41

Analytical Methods
Cellulose concentration in cellulose hydrolysis experiment
was measured by the dry weight. After thorough shaking
of the flask, three 1 ml aliquots of broth were accurately
sampled, centrifuged, and the supernatant removed and dried
at 110oC overnight until the weight was measured constant
for three times. The cellulose concentration was the average
of triple samples with the net error less than 5%.
Cellobiose concentration was analyzed by HPLC (Waters
Co., U.S.A.) using the C18 column (Waters Co., U.S.A.)
and Evaporate Laser Scattering detector (Alltech Co.,
U.S.A.). Glucose and lactic acid concentrations were
analyzed using glucose oxidase-peroxidase and lactic acid
oxidase-peroxidase methods, respectively, with an autoanalyzer
(Biochemistry Analyzer 2700; YSI, Ohio, U.S.A.). Protein
concentration was measured by the Lowry method [4].
Endo and exo-type glucanase activities were measured
by CMCase and avicelase activities, respectively. CMCase,
avicelase, and -glucosidase activities were measured
according to the method of the International Union of Pure
and Applied Chemistry (IUPAC). One unit of the enzyme
activity was defined as the amount of enzyme releasing
1 mol of glucose per minute.
Cell concentration was determined by absorbance using
spectrophotometry (Spectronic Instruments Co., U.S.A.)
at 600 nm wavelength. Samples were diluted from 5 to
50 times to keep the value of absorbance below 0.7. The
sample containing CaCO3 were diluted with 0.5 M HCl
solution in order to overcome the interference of CaCO3 in
absorbance in the measurement.

MODEL DEVELOPMENT
Cellulose Saccharification
Cellulase Adsorption It was confirmed that cellulase
adsorbed to cellulose in both reversible and irreversible
ways, while -glucosidase was not adsorbed to cellulose
(data not shown). Cellulase adsorption is described using
the Langmuir equation:
K p E ads.m E L
E ads = ------------------------------1+K p E L

(1)

where EL is cellulase concentration in the liquid phase, mg/l;

Eads is amount of cellulase adsorbed per gram of cellulose,
mg/g; Eads,m is maximum amount of cellulase adsorbed per
gram of cellulose, mg/g; and Kp is a constant.
E L =E 0 E ads C

(2)

where, E0 is the total cellulase concentration, mg/l; and C is

cellulose concentration, g/l.
The adsorbed cellulase, Eads, is converted to the irreversibly
adsorbed cellulase-cellulose complex. The reversibly
adsorbed cellulase, Ea, decreases in the following way:

42

http://www.paper.edu.cn

LIN et al.

dE a
-------- = k d E ads C
dt

(3)

where kd is a constant, 1/h.

Enzymatic Reaction Kinetics The irreversibly adsorbed
cellulase was fastened to a fixed place in the surface of the
cellulose molecule, unable to reach new substrate of
cellulose, and was apparently deactivated. Only reversibly
adsorbed cellulase could react with cellulose and the
reaction rate is as follows:
K 1 Ea C
r 1 = -----------------------------------------------K 1s ( 1+B K 1B )+C

(4)

where r1 is the reaction rate of cellulose, K1, K1s, K1B are

constants, and B and C are the concentrations of cellobiose
and cellulose.
The fact that a small amount of cellulase liquefied cellulose
with little glucose produced (data not shown) showed that
the substrate of exoglucanase, which is an oligosaccharide,
was soluble. For soluble substrate, the adsorbed cellulase
was a kind of immobilized enzyme, having much decreased
activity compared with the soluble state enzyme. As a
result, only the soluble state cellulase was counted on the
oligosaccharide conversion with the reaction rate as
follows:
K 2 E L OS
r 2 = ---------------------------------------------------K 2s ( 1+B K 2B )+OS

(5)

where r2 is the reaction rate of oligosaccharide, OS is the

concentration of oligosaccharide. K2, K2s, K2B are constants.
V 3m B
r 3 = -----------------------------------------------K 3s ( 1+G K 3G )+B

(6)

where r3 is the reaction rate of cellobiose, V3m, K3s, and K3G

are constants, and G is the concentration of glucose.
Mass Balances for the Reactions
dC
------- = r 1
dt
dOS
----------- =r 1 r 2
dt
dB
------- =1.056 r 2 r 3
dt
dG
------- =1.053 r 3
dt

(7)
(8)
(9)
(10)

Lactic Acid Fermentation

The specific growth rate, , and the specific lactic acid
production rate, qP, are described using Equations (11) and
(12), respectively:
max G
k iG
P n
X
= ------------------ -------------- 1 ------- 1 ----------
k m +G k iG +G
P cri
X max

(11)

q P = +

(12)

The byproducts of oligosaccharides and other kinds of

organic acids in addition to lactic acid were found in
fermentation broth (data not shown). The specific
byproduct production rate, q'P, was found to be growth
related:
q'P =

(13)

The specific glucose consumption rate for byproduct

synthesis, q'G, is as follows:
q'P 1
q'G = ----------- = ----------- = ---
Y'P G Y'P G

(14)

where =Y'P/G/. The total specific glucose consumption

rate, qG, is:
1 1

qP
qP
q G = ---------- ---------- q'G = ---------- + --- ---------

YX G
Y X G YP G
YP G

(15)

In complex medium, carbon source is used for only

energy production and products synthesis, not for cell
synthesis, which was also confirmed using Lactobacillus
rhamnosus [15, 20]. In this research, complex medium was
utilized, therefore, the term of glucose consumption for
cell synthesis in Equation (15) was omitted in the following
simulations.
The mass balances for cell, lactic acid, and glucose are
expressed by Equations (16)- (18):
dX
------- = X
dt

(16)

dP
------ =qP X
dt

(17)

dG
------- = q G X
dt

(18)

Simultaneous Saccharification and Fermentation

In SSF, glucose produced in sludge hydrolysis was
simultaneously consumed by the microorganisms. Combining
the equations for glucose production (10) and glucose
consumption (15) and (18), Equation (19) was built for
modeling the dynamics of glucose concentration in SSF:
dG
X
1 dP
------- =1.053 r 3 ------------ ---------- -----dt

Y P G dt

(19)

In modeling SSF, Equations (1)- (9), (11), (12),

(15)- (17), and (19) were used. Besides, the SSF was
operated in fed-batch mode in experiments, so that
the equations were modified for the dilution effect caused
by the substrate feeding in the SSF simulation. However,
the dilution effect caused by enzyme feeding was
neglected.

http://www.paper.edu.cn

BIOCONVERSION OF PAPER MILL SLUDGE TO LACTIC ACID

43

Table 1. Parameter values for mathematical model of cellulose

hydrolysis.
Parameter

Value

Unit

Eads.m
Kp
kd
K1
K2
V3m
K1s
K2s
K3s
K1B
K2B
K3G

98.2900
00.1855
01.4698
04.1711
00.4156
02.0000
09.4647
00.1161
00.3078
05.9097
06.7788
00.9455

mg/g
l/mg
1/h
g/mg/h
g/mg/h
g/l/h
g/l
g/l
g/l
g/l
g/l
g/l

Fig. 1. Time course of cellulose hydrolysis.

The scattered symbols are experimental data, and the lines are model
predictions.

RESULTS AND DISCUSSION

Cellulose Hydrolysis
Cellulose hydrolysis experiment was performed and
the time courses of cellulose, cellobiose, and glucose
concentrations were measured (Fig. 1). Glucose concentration
reached about 24 g/l at around 50 h from the start of
hydrolysis, and did not increase much from 50 to 80 h.
Cellobiose concentration was maintained at low level all
the time.
The model Equations (1)- (10) were used in the simulation
of cellulose hydrolysis. The differential equations of the
mathematical model were solved using the fourth-order
Runge-Kutta method [10]. The parameter Eads.m was measured
under various cellulase concentrations, and KP was obtained
by linear plots using the experimental data. For the
determination of the values of V1m, K1S, K1B,V3m, K3S, and
K3G, experiments on cellulose and cellobiose hydrolysis
using cellulase and -glucosidase, respectively, were done
using various substrate concentrations, and the experimental
data were plotted to obtain the values of the above
parameters. K1 was calculated using V1m and the protein
concentration of cellulase. For the determination of the
values of K1B and K3G, experiments on cellulose and cellobiose
hydrolysis were performed at various concentrations of the
inhibition of cellobiose and glucose, respectively. The data
of r1 and r3 versus the respective inhibition concentration
were fitted with Equations (4) and (6), respectively, using
nonlinear fitting to obtain the constant values. The values
of K2, K2S, K2B, and kd were obtained using parameter
optimization using the experimental data of the time
course of cellulose hydrolysis. Parameter optimization
was done automatically by computer through search of
the parameter values that minimized the sum of the
squared errors between model prediction and experimental

data using genetic algorithm (GA) [6]. All the parameter

values were then refined using GA from the experimental
data of SSF for lactic acid production from sludge,
as described later in this paper. The final values of model
parameters are shown in Table 1. All the calculations
were carried out on an IBM compatible computer, using
self programmed Visual Basic (Microsoft Co., U.S.A.)
programs.
The mathematical model fitted satisfactorily with
the experimental data (Fig. 1), supporting the hypothesis
of the mechanism of cellullase adsorption deactivation.
This model can be used to predict the effects of both
reversible and irreversible adsorption of cellulase and
the effects of the portions of endo and exo-type glucanase
as well as -glucosidase on cellulose hydrolysis, to provide
useful information for process optimization.
Lactic Acid Fermentation
Lactic acid fermentation at various initial glucose
concentrations, ranging from 10 to 50 g/l, was done using
flask culture, and the max, km, and kiG parameters were
calculated. The result showed that glucose inhibition was
moderate. Lactic acid fermentation at various initial lactic
acid concentrations from 0 to 90 g/l was done using flask,
culture and the specific growth rates were calculated and
plotted (Fig. 2). It shows that lactic acid inhibition was
severe: Initial lactic acid concentration of 30 g/l decreased
the specific growth rate by 21 percent from 0.48 to 0.38 1/
h and was compared with the case with no initial lactic
acid. Almost no cell growth was found in the case of
initial lactic acid concentration of 70 g/l, which was
defined as the critical lactic acid concentration, Pcri. The
specific growth rate was fitted with the experimental data,
and the parameter value of n of Equation (11) was
determined to be 0.7. The parameter values of , , , and
YX/S were calculated using the data of the above two
experiments at various initial glucose and lactic acid

44

http://www.paper.edu.cn

LIN et al.

Table 2. Parameter values for mathematical model of lactic acid

fermentation.

Fig. 2. Effects of initial lactic acid concentration on the specific

growth rate of Lactobacillus.
The scattered symbols are experimental data, and the lines are model
predictions.

concentrations. The value of varied very much with the

cultivation conditions. It was about 0.7 for cultivation
using flasks and was about 0.3 for cultivation using
fermenter. The value of YP/S was obtained by theoretical
calculation, based on the fact that two lactic acid molecules
are produced from one glucose molecule (YP/S=(2MLac)/
MGluc=1g/g). The parameter value of Xmax was directly
measured.
Lactic acid fermentation was done using fermenter, and
the experimental data were used in refinement of the
parameter values of the mathematical model for lactic acid
fermentation using GA (Fig. 3). Then, the parameter
values were further refined using the experimental data of
SSF for lactic acid production from sludge. However, no
refinement was made for the theoretical parameter value of
YP/S. The parameter values for lactic acid fermentation

Parameter

Value

Unit

max
km
Pcri
kiG
n
YP/S

00.576
00.300
70.000
00.200
00.700
01.000
05.000
00.090
00.286

1/h
g/l
g/l
g/l
g/g
g/g

mathematical model are shown in Table 2. Simulation for

fermenter lactic acid fermentation was made, and the
simulation results were compared with the experimental
data, as shown in Fig. 3. The simulation fitted satisfactorily
with the experimental data.
In modeling cell growth, the term of (1- X/Xmax) from
logistic equation was used for the stationary growth phase,
even if there was other kind of equations [12]. The
Luedeking-Piret model [14] was used in modeling lactic
acid production. This model was simple and practical.
Simultaneous Saccharification and Fermentation
Lactic acid production from sludge using SSF was done in
fed-batch mode using fermenter. During the SSF process,
the enzymes were fed 2 times at 8 and 14 h, respectively,
and sludge was fed 5 times at 8, 14, 20, 26, and 32 h,
respectively. The results are shown in Fig. 4. The final
lactic acid concentration reached the highest value of 63 g/l
at about 40 h from the start of cultivation. Glucose
concentration had a small increase at 4 h, reaching the

Fig. 3. Time course of lactic acid fermentation using glucose.

Fig. 4. Time course of fed-batch SSF in production of lactic

acid using sludge.

The scattered symbols are experimental data, and the lines are model
predictions.

The scattered symbols are experimental data, and the lines are model
predictions.

http://www.paper.edu.cn

BIOCONVERSION OF PAPER MILL SLUDGE TO LACTIC ACID

Fig. 5. Time course of predicted reversibly adsorbed cellulase

during the fed-batch SSF.

highest value of 5 g/l, but soon decreased with the cell

growth and remained at low level during the later period of
the cultivation. Simulation of the SSF process was made
using the parameter values listed in Tables 1 and 2, and the
results are shown in Figs. 4- 6. The model simulation fitted
satisfactorily with the experimental data (Fig. 4).
The simulation results showed that the reversibly
adsorbed cellulase decreased quickly due to conversion to
irreversibly adsorbed form (Fig. 5). The curve of the
reversibly adsorbed cellulase concentration of a saw-tooth
shape resulted from the feedings of sludge as well as
enzyme. Simulation of the rates of cellulose hydrolysis is
shown in Fig. 6. The time course of r1 was also of sawtooth shape, almost parallel with the curve of the concentration
of reversibly adsorbed cellulase as expected. r2 and r3 were
catalyzed by soluble state enzymes, as analyzed above,
therefore, the curves of r2 and r3 were quite different from

45

r1. The simulation showed that the reaction rates, especially

r1, were greatly decreased due to irreversible adsorption of
cellulase. It was expected that increase of the reversible
and decrease of the irreversible adsorption of cellulase by
modification of the binding domain through recombinant
gene technology could greatly increase the cellulose hydrolysis
efficiency.
Conventionally, it is regarded that the higher the affinity
of cellulase to cellulose the better for the reaction. As
discussed earlier, as cellulase has no mechanism to move
along the surface of cellulose, it needs to adsorb, desorb,
and re-adsorb to or from the cellulose surface, so that it can
reach new substrates and react continuously. Irreversibly
adsorbed cellulase fails to reach new substrates and is
apparently deactivated, even though the enzyme reaction
domain still has activity. It was confirmed experimentally
that about 8.8% of the total adsorbed cellulase were
changed into irreversibly adsorbed form (data not shown).
Using atomic force microscopy, the above hypothesis was
supported by the experimental results that many holes
were left on the surface of cellulose molecules after cocultivation with the reaction domain deactivated cellulase
[9]. It showed that the cellulase binding domain penetrated
deeply into the cellulose molecule, which might lead to the
formation of the tightly combined cellulase-cellulose complex
and irreversible cellulase adsorption.
Lactobacillus also has the ability to ferment cellobiose
to produce lactic acid. Experiments on lactic acid fermentation,
using cellobiose and the mixture of cellobiose and glucose,
showed that cellobiose utilization was repressed by
glucose (data not shown). Furthermore, -glucosidase
activity was high, and the result of the cellulose hydrolysis
experiment showed that cellobiose concentration was at a
low level (Fig. 1). Therefore, cellobiose fermentation to
lactic acid by Lactobacillus was omitted in the SSF
mathematical model.
In conclusion, SSF is a good choice for bioconversion of
cellulosic materials to lactic acid. The conditions for cellulose
hydrolysis and lactic acid fermentation are similar, which
favors the two reactions to be carried out simultaneously in
the same reactor and greatly simplifies the production
process. Besides, SSF can relieve inhibition effects of
glucose and cellobiose [1]. Utilization of paper mill sludge
instead of natural cellulosic materials saves not only the
pretreatment cost, but also relieves the environmental pollution
problem of paper industries. Modeling and simulation are
useful in investigation of characteristics, analysis, as well
as optimization of the SSF process.

Acknowledgment
Fig. 6. Time course of predicted enzymatic reaction rates of r1,
r2, and r3 during the fed-batch SSF.

This work has been supported by the ERC for Advanced

Bioseparation Technology, KOSEF.

46

http://www.paper.edu.cn

LIN et al.

NOMENCLATURE

REFERENCES

B
C
E0
Ea
Eads

1. Abe, S. and M. Takagi. 1991. Simultaneous saccharification

and fermentation of cellulose to lactic acid. Biotechnol.
Bioeng. 37: 93- 96.
2. Carminal, G., J. Lopez Santin, and C. Sola. 1985. Kinetic
modeling of the enzymatic hydrolysis of pretreated
cellulose. Biotechnol. Bioeng. 27: 1282- 1290.
3. Gonzalez, G., G. Carminal, C. Demas, and J. Lopez Santin.
1989. A kinetic model for pretreated wheat-straw saccharification
by cellulase. J. Chem. Techn. Biotechnol. 44: 275- 288.
4. Dawson, R. M. C., D. C. Elliott, W. H. Elliott, and
K. M. Jones. 1986. Data for Biochemical Research, Third
Ed. Clarendon Press, Oxford.
5. Ge, C. M., S. B. Gu, X. H. Zhou, J. M. Yao, R. R. Pan, and
Z. L. Yu. 2004. Breeding of L(+)-lactic acid producing strain
by low-energy ion implantation. J. Microbiol. Biotechnol.
14: 363- 366.
6. Goldberg, D. E. 1989. Genetic Algorithm in Search,
Optimization and Machine Learning. Addison Wesley
Publishing Co., U.S.A.
7. Kaur, P. P., J. S. Arneja, and J. Singh. 1998. Enzymatic
hydrolysis of rice straw by crude cellulase from Trichoderma
reesei. Bioresourse Technol. 66: 267- 269.
8. Koh, J. H., J. M. Kim, and H. J. Suh. 2003. Immune enhancing
effect by orally-administered mixture of Saccharomyces
cerevisiae and fermented rice bran. J. Microbiol. Biotechnol.
13: 196- 201.
9. Lee, I., B. R. Evans, and J. Woodward. 2000. The mechanism
of cellulase action on cotton fibers: Evidence from atomic
force microscopy. Ultramicroscopy 82: 213- 221.
10. Shoup, T. E. 1979. A Practical Guide to Computer Methods
for Engineers. Prentice-Hall, Englewood Cliffs, N.J.
11. Lee, S. M., J. Q. Lin, and Y. M. Koo. 2002. Chapter 10.
Hydrolysis of paper mill sludge using mixed cellulose
system: Enzymatic hydrolysis of paper sludge, pp. 121138. In M. R. Marten, T. H. Park, and T. Nagamune
(eds.), Biological Systems Engineering. Oxford Univ Press,
England.
12. Lin, J. Q., S. M. Lee, H. J. Lee, and Y. M. Koo. 2000.
Modeling of typical microbial cell growth in batch culture.
Biotechnol. Bioprocess Eng. 5: 382- 385.
13. Lin, J. Q., S. M. Lee, and Y. M. Koo. 2001. Hydrolysis of
paper mill sludge using an improved enzyme system. J.
Microbiol. Biotechnol. 11: 362- 368.
14. Luedeking, R. and E. L. Piret. 1959. A kinetic study of
the lactic acid fermentation. J. Biochem. Microbiol. 1: 393412.
15. Olmos-Dichara A., F. Ampe, J. Uribelarrea, A. Pareilleux,
and G. Goma. 1997. Growth and lactic acid production by
Lactobacillus casei ssp. rhamnosus in batch and membrane
bioreactor. Biotechnology Lett. 8: 709- 714.
16. Parajo, J. C., J. L. Alonso, and V. Santos. 1996.
Development of a generalized phenomenological model
describing the kinetics of the enzymatic hydrolysis of
NaOH-treated pine wood. Appl. Biochem. Biotechnol. 56:
289- 299.

Eads.m
EL
G
K1
K1B
K1s
K2
K2B
K2s
K3G
K3s
kd
kiG
km
Kp
MLac
MGluc
n
OS
P
Pcri
qG
q'G
qP
qP
r1
r2
r3
X
Xmax
YP/G
YX/G

: cellobiose concentration (g/l)

: cellulose concentration (g/l)
: total cellulase concentration (mg/l)
: reversibly adsorbed cellulase concentration (mg/l)
: amount of cellulase adsorbed per gram of cellulose
(mg/g)
: maximum amount of cellulase adsorbed per gram
of cellulose (mg/g)
: cellulase concentration in liquid phase (mg/l)
: glucose concentration (g/l)
: constant (g/mg/h)
: cellobiose inhibition constant for r1 (g/l)
: cellulose saturation constant (g/l)
: constant (g/mg/h)
: cellobiose inhibitory constant for r2 (g/l)
: oligosaccharide saturation constant (g/l)
: glucose inhibitory constant for r3 (g/l)
: glucose saturation constant (g/l)
: conversion constant of adsorbed cellulase from
reversible to irreversible form (1/h)
: glucose inhibitory constant for Lactobacillus cell
growth (g/l)
: glucose saturation constant for Lactobacillus cell
growth (g/l)
: constant of Langumir equation (l/mg)
: molecular weight of lactic acid (g/mol)
: molecular weight of glucose (g/mol)
: constant (- )
: oligosaccharide concentration (g/l)
: lactic acid concentration (g/l)
: critical lactic acid concentration for inhibition of
glucose consumption (g/l)
: total specific glucose consumption rate (g/g/h)
: specific glucose consumption rate for byproduct
synthesis (g/g/h)
: specific lactic acid production rate (g/g/h)
: specific byproduct production rate (g/g/h)
: reaction rate of cellulose (g/l/h)
: reaction rate of oligosaccharide (g/l/h)
: reaction rate of cellobiose (g/l/h)
: cell concentration (g/l)
: maximum cell concentration (g/l)
: lactic acid yield from glucose (g/g)
: cell yield from glucose (g/g)

Greek Symbols

: constant (g/g)

: constant (g/g/h)

: constant (g/g)

: constant (g/g)

: specific growth rate (1/h)

max : maximum specific growth rate (1/h)

http://www.paper.edu.cn

BIOCONVERSION OF PAPER MILL SLUDGE TO LACTIC ACID

17. South, C. R., D. A. L. Hogsett, and L. R. Lynd. 1995.

Modeling simultaneous saccharification and fermentation of
lignocellulose to ethanol in batch and continuous reactors.
Enzyme Microb. Technol. 17: 797- 803.
18. Yu, X., H. S. Yun, and Y. M. Koo. 1998. Production of
cellulase by T. reesei Rut C30 in a batch fermentation. J.
Microbiol. Biotechnol. 8: 575- 580.
19. Vickroy, T. B. 1985. Lactic acid, pp. 761- 776. In Blanch,
H. W., Drew, S. and Wang, D. I. C. (eds.), The Practice

47

of Biotechnology: Commodity Products. Pergamon Press,

Elmsford, NY, U.S.A.
20. Youssef, C. B., V. Guillou, and A. Olmos-Dichara. 2000.
Modelling and adaptive control strategy in a lactic acid
fermentation process. Control Engineering Practice 8: 12971307.