Ultrasonics Sonochemistry 12 (2005) 373384

www.elsevier.com/locate/ultsonch

Eects of ultrasonic intensity and reactor scale on kinetics

of enzymatic saccharication of various waste papers
in continuously irradiated stirred tanks
Chengzhou Li, Makoto Yoshimoto, Haruki Ogata, Naoki Tsukuda, Kimitoshi Fukunaga,
Katsumi Nakao *
Faculty of Engineering, Department of Applied Chemistry and Chemical Engineering, Yamaguchi University,
2-16-1 Tokiwadai, Ube, Yamaguchi, 755-8611, Japan
Received 22 July 2003; received in revised form 16 December 2003; accepted 6 February 2004
Available online 15 April 2004

Abstract
Based on the enzymatic saccharication of the various pulps in the previous 0.8 l ultrasonic stirred tank reactor, the ultrasoundenhanced saccharication of waste papers such as newspaper, carton paper, oce paper etc. was carried out in the same reactor as
well as larger scale stirred tank reactors of size 3.2 and 6.4 l. The saccharication of each waste paper was less enhanced in the larger
reactor at a given ultrasonic intensity. This could be attributed to the decrease in the ultrasonic intensity per reaction volume, i.e.,
the specic ultrasonic intensity. Most waste papers were more eciently hydrolyzed with increasing specic ultrasonic intensities,
although newspaper was less eciently done for a too high specic intensity. Such an adverse eect might be due to the fact that
some impurities in the newspaper such as lignin were activated by an intensive ultrasonic irradiation to form a rigid and closed
network, which inhibited the access and adsorption of cellulase on to the substrate surface. The previous kinetic model was found to
be applicable to analyze and simulate the saccharication of each waste paper in the dierent ultrasonic reactors. The ultimate
conversion of a substrate based on the total sugar concentration estimated for an innite reaction time could be correlated as a
function of the ratio of initial substrate to enzyme concentrations at a xed specic ultrasonic intensity. Either the apparent rate
constant or the ultimate conversion increased and tended to approach a constant with an increase in the specic ultrasonic intensity
except for the case of newspaper, while neither the apparent Michaelis constant, product inhibition constant nor glucose formation
equilibrium constant was inuenced by the specic ultrasonic intensity.

2004 Elsevier B.V. All rights reserved.
Keywords: Enzymatic saccharication; Waste paper; Ultrasonic stirred tank; Scale-up; Kinetics

1. Introduction
Waste reutilization has become a matter of great
interest since the increased waste emission is threatening
the limited resources and living spaces on the Earth.
Among the various components of the municipal solid
wastes, waste papers rank rst with a percentage of 38%
[1]. Waste papers can be reutilized several times through
manufacturing recycled paper before cellulose bers

Corresponding author. Tel.: +81-836-85-9271; fax: +81-836-859201.

E-mail address: knakao@yamaguchi-u.ac.jp (K. Nakao).
1350-4177/$ - see front matter
2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ultsonch.2004.02.004

become too short and weak to make paper [2]. On the

other hand, waste papers may be typical of abundant
renewable cellulosic resources for producing useful
bioproducts such as soluble sugars [3]. Moreover, bioethanol that is ethanol made microbially from biomass,
such as cellulose has been approved as a promising
alternative energy source for increasing energy security
and reducing air pollution from contaminants such as
NOx [4]. To achieve an ecient conversion of waste
cellulose to soluble sugars, the enzymatic hydrolysis of
cellulose is suggested to be preferred to the various acidcatalyzed processes using inorganic acid, subcritical or
supercritical waters, etc. since the former not only oers
a bioconversion process under the simpler and milder

374

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

operating conditions but also produces no by-products

detrimental to fermentative microorganisms [5].
Despite the numerous researches on the enzymatic
hydrolysis of cellulose over the last few decades, the
process has remained of low eciency and not been fully
understood because of the heterogeneity of substrate as
well as enzyme [6,7]. Many processes have been developed to improve the conversion of cellulose hydrolysis,
such as ball mill reactors [8,9], aqueous two-phase systems [10,11], various explosion pretreatment processes
[2,12] and so on. The steam explosion pretreatment
process is reported to be the most economical process
among these processes because it makes the substrate
more susceptible to enzymes with a lower energy
requirement and environmental protection cost [2,3].
However, the process cannot prevent cellulase from
adsorbing irreversibly onto the substrate during the
hydrolysis, which has been proposed to be one of the
mechanisms responsible for cellulase deactivation [13
15].
In our previous work on the enzymatic hydrolysis of
alkali-treated lignocellulosics [16], it was suggested
that the continuous ultrasonic irradiation enhanced not
only the accessibility of the substrate for enzyme
adsorption but also the desorption of the inactively
adsorbed enzyme to cause its reactivation resulting in an
enhanced hydrolysis. Recently, we have observed a
remarkable enhancement for the enzymatic hydrolysis
of the paper pulps in a 0.8 l stirred tank reactor with
continuous ultrasonic irradiation. The saccharication
of the pulp called NUKP under ultrasonic irradiation
proceeded as a two-staged reaction, and the enhancement eect of the ultrasonic irradiation on enzymatic
hydrolysis of cellulose was evaluated based on the simplied kinetic model proposed. It was also found that
the variation in the property of pulp as well as ultrasonic
intensity exerted an eect on the apparent rate constant
and ultimate conversion of the substrate, but no eect
on the apparent Michaelis and competitive product
inhibition constants [5]. Although many researches have
focused on the ultrasound-enhanced microbial reaction
during the last decade [17], there have been few kinetic
studies on the eects of ultrasonic intensity and reactor
size on the enzymatic reaction enhanced by ultrasonic
irradiation for design and scale-up of ultrasonic bioreactors.
The purpose of this work is (a) to extend the enzymatic saccharication of the pulps enhanced by the
continuous ultrasonic irradiation to that of the waste
papers including newspaper, carton paper and oce
paper, (b) to scale-up the stirred tank reactor for the
saccharication of waste papers with continuous ultrasonic irradiation, (c) to examine the applicability of
the previous simplied kinetic model to the present
process in the varied scale reactors and (d) to evaluate
the eects on the saccharication of ultrasonic irradia-

tion, scale-up and substrate property based on the

kinetic model.

2. Materials and methods

2.1. Enzyme and substrates
Cellulase (Meicelase , 238 FPU/g) from Trichoderma
viride was provided by Meiji Seika Kaisha Ltd. (Tokyo,
Japan) and used without further purication. The cellulase activity was assayed according to the method
suggested by Mandels et al. [18]. Newspaper, carton
paper and oce paper, which had been shredded into
pieces (6 12 mm) and then washed with deionized
water and dried at 50 C, were used as model substrates
of waste cellulose. NUKP pulp supplied by Daio Paper
Corporation (Tokyo, Japan) was used as a reference.
2.2. Determination of cellulose availability of
waste papers used
A piece of a waste paper with a size of 20 60 mm
was weighed, shaped into a spiral sheet and hydrolyzed
in 6 ml of 0.001 M acetate buer solution of pH 4.8 in a
borosilicate test tube at 45 C for 144 h under an initial
cellulase concentration of 4 g l
1 , which was 10 times
higher than that of 0.4 g l
1 for the enzymatic saccharication of the waste paper in the ultrasonic stirred tank
reactors described below. All the enzymatic hydrolyses
of the various waste papers under such a high cellulase
concentration were assumed to give the available
amount of cellulose present in the respective substrates
used. The cellulose availability of a waste paper was
therefore calculated as 90 Tf =S0 %, where Tf (g l
1 ) and
S0 (g l
1 ) represent the observed nal total sugar concentration and initial substrate concentration, respectively.
2.3. Ultrasonic stirred tank reactors employed
Fig. 1(a) shows a schematic diagram of the 0.8 l
ultrasonic reactor made of semitransparent PVC resin,
which was the same as that used previously for the paper
pulps [5]. Fig. 1(b) represents larger scale ultrasonic
reactors having a reaction volume of 3.2 and 6.4 l. The
larger stirred tank reactors made of PVC resin having
a round cross-section are equipped with the same design
of impeller and ultrasonic horn system as in the case
of the 0.8 l ultrasonic reactor, being installed vertically
into the tank. The ultrasonic generator used in this
work was Sonier Model 250D (Branson Ultrasonics
Corporation, USA), which is the same as that in the
previous work [5] and can produce monofrequency
ultrasound of 20 kHz with an adjustable output power
up to 250 W to irradiate variable ultrasonic intensities.

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

375

Fig. 1. Schematic diagrams of (a) small scale stirred tank reactor of 0.8 l in working volume and (b) larger scale reactors of 3.2 and 6.4 l in volume for
enzymatic saccharication of waste paper with continuous ultrasonic irradiation.

A negligible eect of continuous ultrasonic irradiation

on cellulase activity was observed under the present
experimental conditions. The power applied to the
reaction solution was used to evaluate the ultrasonic
intensity since it is considered to be approximately
proportional to the heat generated under the present
condition of 1560 W.
2.4. Enzymatic saccharication of waste paper in
small scale ultrasonic reactors
To examine the eect of the continuous ultrasonic
irradiation on the kinetics of the enzymatic saccharication of the waste papers, the reactions were carried
out in the 0.8 l stirred tank reactor. The initial enzyme
concentration was xed at 0.4 g l
1 for each experiment.
The initial substrate concentration was varied from 7.5
to 25 g l
1 to observe the dierent time courses, from
which the kinetic model parameters could be determined
based on the previous kinetic analysis [5]. The ultrasonic
intensity of 30 W was employed to enhance the enzymatic saccharication of the waste papers according to
our previous report on the paper pulps. A known volume of the suspension of a substrate in 0.001 M acetate
buer solution of pH 4.8 was stirred at 45C for 24 h to
form a well dispersed slurry before starting the hydrolysis by adding a given amount of concentrated cellulase
solution. Then the enzymatic hydrolysis was carried out

at 45C for 48 h with a stirring speed of 300 rpm under

an ultrasonic intensity of 30 W. The liquid samples were
withdrawn periodically and analyzed to follow the time
courses of total sugar and glucose formed during the
hydrolysis.
2.5. Enzymatic saccharication of waste paper
in larger scale ultrasonic reactors
In the larger scale ultrasonic stirred tank reactors of
3.2 and 6.4 l, the various waste papers were hydrolyzed
to examine the eect of reactor scale as well as ultrasonic
intensity on the kinetics of the enzymatic saccharications. Oce paper A among the waste papers used in the
small scale reactor was replaced by oce paper B in the
larger scale ones as the former paper was found to be a
recalcitrant substrate for the enzymatic hydrolysis. The
initial substrate concentration S0 (g l
1 ) was xed at 7.5
g l
1 in each experiment while the ultrasonic intensity Is
(W) was varied from 0 to 60 W. The enzymatic hydrolysis was started and followed in the same way as above.
2.6. Sampling and analysis
The samples withdrawn were instantly boiled by
microwaves and kept in boiled water for 10 min to
inactivate cellulase and then ltered to remove the unreacted insoluble substrate. The total sugar and glucose

376

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

in the ltrate were analyzed by the phenolsulfuric acid

and enzymatic methods, respectively.

3. Results and discussion

3.1. Cellulose availability of waste papers used
The cellulose availability of waste papers dened
above as an indicator of substrate accessibility to the
enzyme is considered to inuence the reaction proles of
their enzymatic hydrolysis. The values of cellulose
availability of the waste papers used were determined
and are shown in Table 1.
The results suggest that the cellulose availability for
the waste papers except oce paper A and carton paper
is approximately equal to the inherent cellulose content
Table 1
Cellulose availability of substrates used
Substrate

Cellulose availability (%)

NUKP
Newspaper
Carton paper
Oce paper A, unprinted
Oce paper A, printed
Oce paper B, unprinted
Oce paper B, printed

94.8
75.1
28.6
7.04
5.08
97.4
93.5

of the substrates. With respect to the very low values

determined for oce paper A and carton paper, their
cellulose availability should be regarded as the cellulose
content that might be enzymatically hydrolyzed as
extensively as possible under the present condition.
Oce paper A and carton paper are considered to
contain some kinds of additives, which resist the enzyme
action, to improve their own quality. Ink printed on the
waste papers was found to exert negligible inuence on
the sugar production within the range of the present
enzymatic saccharications. The oce paper shown
below was an unprinted one.
3.2. Enzymatic saccharication of waste paper in a
small scale ultrasonic reactor
Fig. 2(a) shows the observed time courses of both
total sugar and glucose concentrations, T (g l
1 ) and G
(g l
1 ), respectively, in the saccharications of the various waste papers including newspaper, carton paper,
oce paper A and the paper pulp NUKP as reference in
the 0.8 l ultrasonic reactor with continuous irradiation
of 30 W. The time courses of the saccharications
without ultrasonic irradiation as a control are shown in
Fig. 2(b).
It is clearly found in Fig. 2 that the continuous
ultrasonic irradiation exerted an enhancement eect on
the enzymatic saccharications of all the waste papers

(b) Without irradiation

(a) With irradiation of 30 W

T(g l-1)

T22=8.1 g l-1

T11=6.9 g l-1
T=4.8 g l-1

Newspaper
Carton
Office paper A
NUKP

10
T (g l-1)

Newspaper
Carton
Office paper A
NUKP

10

T=6.7 g l-1
5

T=4.2 g l-1

-1

T=3.5 g l

T=2.9 g l-1
-1

T =1.7 g l

T=1.2 g l-1

5
G (g l-1)

G (g l-1)

0
0

12

24
t (h)

36

48

12

24

36

48

t (h)

Fig. 2. Observed time courses of total sugar and glucose concentrations, T and G, in enzymatic saccharications of various waste paper in 0.8 l
ultrasonic stirred tank reactor (S0 7:5 g l
1 ) (a) with ultrasonic irradiation of 30 W and (b) without irradiation. Value of ultimate total sugar
concentration T1 was determined by the previous extrapolation method [5]. Solid lines show the calculated time courses from Eq. (5).

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

3.2.1. Kinetic analysis and simulation

Table 2 shows the simplied kinetic model proposed
previously. The details on the model are available in our
previous papers [5,8].
The kinetic model parameters such as T1 , k, KM , KI
and K were determined from the observed time course of

the total sugar and glucose concentrations. At rst, for

each waste paper, the value of the ultimate total sugar
concentration T1 corresponding to the dierent value of
S0 was determined according to the extrapolation
method described previously [5]. In the initial stage of
the reaction (t ! 0), the total sugar formed can be ignored so that Eq. (2) can be simplied as follows:


dT
kE0 T1

or
dt t!0 KM 0:9T1
1
KM 1
0:9

4
dT =dtt!0 kE0 T1 kE0
Fig. 3 shows the determination of the initial rates
dT =dtt!0 from the corresponding time course of total
sugar produced during the enzymatic saccharication of
newspaper and also the results of T1 for the dierent
values of S0 . For the kinetic analysis of the saccharications in the small scale ultrasonic reactor, the values of
the apparent rate constant k and Michaelis constant KM
for each substrate under a given ultrasonic intensity

15
With ultrasonic irradiation of 30 W

T=12.2 g l-1

Slope=(dT/dt)t0
t 0
10

T (g l-1)

used. The extremely low level of sugar production is

observed in the case of oce paper A due to the very low
cellulose availability shown in Table 1. NUKP in the
case of ultrasonic irradiation shows the two stage time
course as reported previously [5].
The two stage time course is considered to be due to
the following mechanisms [8]. The two cellulase components EG and CBH diuse into the cellulose ber
forming the paper substrate and adsorb onto the surface
layer of the microbrils composing the ber to form a
reaction zone from the outer surface toward the inner
part of the cellulose ber. The rst stage starts with the
formation of the reaction zone described above accompanied by the hydrolysis of the cellulose ber on its
outer surface. As the hydrolysis proceeds, the reaction
rate decreases since the soluble total sugars are accumulated and competitively inhibit the enzymes EG and
CBH desorbed from the substrate. Especially, the enzymes which are adsorbed on to the more deeper inner
layers of microbrils network in the reaction zone, may
be inactivated due to locally concentrated oligosaccharides. The inhibited enzyme is considered to be incapable of adsorbing on to the substrate. The second stage of
the reaction starts with the appearance of the remaining
part of the cellulose ber which contains neither cellulase nor total sugars. At the instant when the fresh
substrate surface appears, the inhibited EG and CBH
are complexed with the fresh substrate to form a reaction zone similar to that in the rst stage above.
The behavior of the two-stage reaction is considered
to be more remarkable when the eect of degradation of
the microbril network by the enzymatic action is
comparable to that of its loosening by such an external
shear stress as is caused by continuous ultrasonic irradiation.

377

T=8.0 g l-1

T=4.8 g l-1

S0 [g l-1]
7.5
15
25

0
0

12

24
t (h)

36

48

Fig. 3. Determination of various initial rates from time courses observed for dierent initial newspaper concentrations in 0.8 l reactor
with ultrasonic intensity of 30 W. Solid lines represent the empirical
correlation curves.

Table 2
Simplied kinetic model for enzymatic hydrolysis of cellulose
EG=CBH

BG

Celluose S ! Oligosaccharides O
Glucose G

Rate of production of total sugar

dT
kE0 T1
T

dt
KM 1 1=KI T  0:9T1
T

Equilibrium constant K G=O G=T
G

where EG, CBH and BG denote endoglucanase, cellobiohydrolase and b-glucosidase, respectively, T1 , k, KM and KI mean ultimate total sugar
concentration, apparent rate constant, apparent Michaelis constant and apparent inhibition constant, respectively, and T O G

378

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

were determined by plotting of 1=dT =dtt!0 vs. 1=T1

according to Eq. (4). To determine the parameter KI ,
integrating Eq. (2) in Table 2 under the initial and nal
conditions (T T0 at t 0 and T T at t t, respectively) gives the following equations:
t
0:9T
T0

lnT1
T0 =T1
T 

c
0:9T
T0

where
b

KM 1
KM
T1
kE0 KI
kE0

1 KM 1
1


0:9 kE0 KI kE0

The left-hand side of Eq. (5) was then plotted against

the rst term of the right-hand side except b to obtain
a series of straight lines with the same intercept corresponding to a constant value of c for the various
values of S0 . Then the values of b and c were obtained
from the slope and intercept of the plot, respectively.
Thus, a reasonable value of the apparent inhibition
constant KI was determined to satisfy Eqs. (6) and (7)
simultaneously. Finally, to determine the value of the
equilibrium constant K between oligosaccharide O and
glucose G, Eq. (3) was rearranged into the following
form:
G=T K=K 1

The equilibrium constant K for each substrate was

calculated from the slope of a linear plot of G vs. T
according to Eq. (8). The results on the kinetic model
parameters k, KM ; KI and K obtained for the waste
papers as well as NUKP are shown in Table 3.
To examine the applicability of the kinetic model, the
time courses of the various saccharications were calculated by applying Eqs. (5) and (8) with the parameters
determined above. The solid lines shown in Fig. 2 are
the calculated time courses, which agree well with the
observed ones. The two stage time course in the saccharication of NUKP shown in Fig. 2(a) was not so
appreciable as reported previously [5] due to the lower

values of S0 in this work. Nevertheless, the kinetic

analysis and simulation for the present enzymatic saccharication of NUKP with the continuous ultrasonic
irradiation were carried out based on the method proposed previously for the two stage time courses [5]. At
rst, the two ultimate total sugar concentrations T11
and T12 for the rst and second stage saccharications,
respectively, were determined by the extrapolation
mentioned above, then the kinetic model parameters
were determined from the time course of the rst stage
and employed to simulate the second stage saccharication with T12 under the empirically found condition
that the time course of the second stage of saccharication started when the total sugar concentration
reached 75% of T11 . When the dierence between T11
and T12 is small enough (e.g. less than 5%) to be neglected considering the experimental errors, the time
course can be approximated as the one stage time course
for simplication. The time courses except for NUKP
shown in Fig. 2(a) and those in Figs. 2(b) and 3 were
regarded as one stage saccharications.
3.2.2. Eects of substrate property and ultrasonic
irradiation on kinetic model parameters
The present kinetic model assumes that the apparent
Michaelis constant KM and competitive inhibition constant KI represent the anity of the enzyme to the
substrate and to the total sugar produced, respectively,
and hence predicts that the values of KM and KI can be
determined as the respective constant irrespective of the
substrate property, ultrasonic intensity and reactor type
and scale. The results for the KM and KI values shown in
Table 3 are found to be consistent with the kinetic
model. On the other hand, the model suggests that the
apparent rate constant k representing binding frequency
between substrate and enzyme as well as the ultimate
total sugar concentration T1 representing the amount of
cellulose available for enzymatic hydrolysis depend on
the substrate property and the operating conditions such
as ultrasonic intensity, reactor type and scale, initial
substrate concentration, etc. and also that the equilibrium constant K is aected by the substrate property
and unaected by such operating conditions as those

Table 3
Kinetic model parameters for waste papers and NUKP in the 0.8 l ultrasonic stirred tank reactor
Substrate

Is (W)

k (h
1 )

KM (g l
1 )

KI (g l
1 )

K ()

Newspaper

0
30

13
16

13.4

0.704

2.0

Carton paper

0
30

9.0
10

13.4

0.704

1.0

Oce paper A

0
30

2.5
3.7

13.4

0.704

0.0056

NUKP

0
30

11
24

13.4

0.704

2.7

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

mentioned above. The values of k and K shown in Table

3 and those of T1 inserted in Fig. 2 are considered to
support the kinetic model.
It is seen in Table 3 that the value of K except for
oce paper A ranges from 1.0 to 2.7 being within a
narrow limit. Too low a value of K for oce paper A
may be due to an inhibition of b-glucosidase by some
additives in the paper.
3.2.3. Apparent ultimate conversion of substrate
inuenced by ratio of initial substrate to enzyme
concentrations
As shown in Fig. 4, the solid curve for the saccharication of a pure cellulose CC31 without ultrasonic
irradiation, demonstrates that the apparent ultimate
conversion Xf (dened as 0.9 T1 =S0 ) was found to be a
unique function of the ratio of the initial substrate to
enzyme concentrations S0 =E0 [8]. In the case of pure
cellulose, the value of Xf is found to be almost unity
when the ratio of S0 =E0 is less than a certain value, to
decrease sharply with increasing S0 =E0 within a certain
range of S0 =E0 and nally to tend to approach a constant. The Xf values obtained for NUKP and waste
papers in the 0.8 l ultrasonic reactor with the various
initial substrate concentrations and ultrasonic intensity
are also shown in Fig. 4. On the whole, the trend of the
data shown for each waste paper with any ultrasonic
intensity is found to be similar to the result of CC31. It is
also revealed in the gure that the enhancement eect of
ultrasonic irradiation on Xf increases with increasing
S0 =E0 for each substrate, which means that the ultrasonic irradiation becomes more ecient for the higher
initial substrate concentration at a constant initial enzyme concentration. Once such a correlation of Xf vs.
S0 =E0 is empirically determined, the correlation curve
can be used to predict the value of Xf , i.e., the value of
T1 from the predetermined value of S0 =E0 .

379

3.3. Enzymatic saccharication of waste papers in

larger scale ultrasonic reactors
Since the ultrasound-enhanced saccharications of
the various waste papers were conrmed in the 0.8 l
reactor with the continuous irradiation, the reactor size
was scaled-up to the 3.2 and 6.4 l. It was found in the
scale-up that only a jacket attached to the outer wall of
the larger reactors could keep the reaction temperature
constant at 45 C, as opposed to the case of the 0.8 l
reactor having an additional coil inside it. It is noted
that the heat generated by ultrasonic irradiation can be
more easily controlled in the larger scale ultrasonic
bioreactor in which the lower ultrasound energy is
usually applied compared to common sonochemical
reactors [17].
Fig. 5(a)(d) show the time courses of the total sugar
and glucose concentrations, T and G, in the enzymatic
saccharications of the three types of waste papers,
newspaper, carton paper and oce paper B and the
paper pulp NUKP as reference substrate, respectively,
with dierent reactor sizes and ultrasonic intensities.
With respect to the eect of reactor size on the time
courses of the saccharications, it is found from the
results shown in Fig. 5 together with Fig. 2 that the
concentration levels of total sugar and glucose produced
are only slightly decreased with increasing reactor sizes
from 0.8 to 6.4 l for any given substrate without ultrasonic irradiation. This may be due to the very low rate of
enzymatic hydrolysis of cellulose in spite of the higher
degree of non-homogeneity in dispersion of insoluble
substrate for the larger scale reactor. Concerning the
eect of the ultrasonic intensity on the total sugar and
glucose production, it is revealed from Fig. 5(a) for
newspaper that the 30 W irradiation enhances sugar
production in both the 3.2 and 6.4 l reactors, whereas
the 60 W irradiation retards it in both reactors. The

0 W 30 W

Xf (-)

Newspaper
Carton
Office paper A
NUKP (T 22)
NUKP (T 11)
Newspaper
Carton
Office paper A
NUKP
Pure cellulose
(CC31) without
ultrasonic
irradiation

0.5

10

100

1000

S0/E0 (-)
Fig. 4. Apparent ultimate conversion of substrate Xf as a function of the ratio of initial substrate to enzyme concentrations S0 =E0 in 0.8 l ultrasonic
stirred tank reactor for various substrates (E0 0:4 g l
1 ; S0 7:5; 15; 25 g l
1 ). Lines represent the empirical correlation curves.

380

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

10

(a) Newspaper
0 W (T=4.1 g l-1)
30 W (T =5.2 g l-1)
60 W (T =4.6 g l-1)

6.4 l

6.4 l

3.2 l
0

0
0

10

12

24
36
t (h)

48 0

12

24
36
t (h)

48

10

(c) Office paper B

T (g l- 1)

T (g l- 1)

0 W (T=2.4 g l-1)
30 W (T=2.8 g l-1)
60 W (T=3.2 g l-1)

G (g l- 1)

G (g l- 1)

0 W (T=2.8 g l-1)
30 W (T=3.1 g l-1)
60 W (T =3.6 g l-1)

3.2 l

3.2 l

6.4 l

-1

-1

0 W (T=6.7 g l )
30 W (T=7.5 g l-1)
60 W (T=8.1 g l-1)

12

24
36
t (h)

48 0

12

24
36
t (h)

48

(d) NUKP

6.4 l

3.2 l
0 W (T=6.6 g l-1)
30 W (T=7.8 g l-1)
60 W (T=8.0 g l-1)

0 W (T=6.6 g l )
30 W (T=7.3 g l-1)
60 W (T=7.5 g l-1)

G (g l- 1)

G (g l- 1)

(b) Carton

0 W (T=3.6 g l-1)
30 W (T=4.9 g l-1)
60 W (T=4.7 g l-1)
T (g l- 1)

T (g l- 1)

10

0 W ((T=5.9 g l-1)
30 W (T=6.4 g l-1)
60 W (T=7.3 g l-1)

0
0

12

24
36
t (h)

48 0

12

24
36
t (h)

48

12

24
36
t (h)

48 0

12

24
36
t (h)

48

Fig. 5. Observed time courses of total sugar and glucose concentrations, T and G, in enzymatic saccharications of (a) newspaper, (b) carton paper,
(c) oce paper B and (d) NUKP in 3.2 and 6.4 l stirred tank reactors with dierent ultrasonic intensities Is (S0 7:5 g l
1 ). Solid lines show the
calculated time courses from Eq. (5).

reason for this is discussed later. For either carton paper

(Fig. 5(b)) or oce paper B (Fig. 5(c)), a steadily
increasing enhancement eect with increase in the
ultrasonic intensity from 30 to 60 W is clearly seen in
both the 3.2 and 6.4 l reactors although for the latter
substrate a greater enhancement is obtained when the
ultrasonic intensity is increased from 0 to 30 W than
when increased from 30 to 60 W in both reactors. As for
the reference substrate NUKP (Fig. 5(d)), even in the
case of the 3.2 l reactor, the ultrasonic irradiation of 30
W exerts almost the same enhancement eect on sugar
production as that of 60 W while in the case of 6.4 l
reactor, the intensity of 30 W has a slight enhancement
eect although increasing the ultrasonic intensity from
30 to 60 W results in an appreciable eect.

3.3.1. Kinetic analysis and simulation

For the kinetic analysis of the saccharications in
the larger scale reactors with a xed initial substrate
concentration of 7.5 g l
1 shown in Fig. 5, plotting
1=dT = dtt!0 vs. 1=T1 to determine the values of k
and KM is not applicable since it requires the kinetic data
on dierent values of T1 resulting from the dierent
initial substrate concentrations S0 .
Thus, the kinetic determination was made as follows:
At rst, the value of T1 for each time course was
determined using the extrapolation method described
earlier. Then, the linear plot of the time course of the
total sugar concentration T for a given substrate based
on Eq. (5) gave the values of b and c from its slope and
intercept, respectively. Hence, the denitions of b and c

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

expressed as Eqs. (6) and (7), respectively, yielded an

optimal value of k with the known values of KM and KI
so that the value of k should be consistent with both
slope and intercept of the linear plot. The same values of
KM and KI as those shown in Table 3 were assumed for
all the substrates including oce paper B since either
parameter was shown to be an intrinsic constant. Finally, the equilibrium constant K for each substrate was
determined by the same method as described above. The
results for the values of k and K obtained for the substrates under the various operating conditions are
summarized in Table 4.
It is revealed from the parameters determined that the
value of K is unaected by the ultrasonic intensity and
that both k and T1 values increase with increasing
ultrasonic intensities for all the substrates except newspaper for which increase in the ultrasonic intensity from
30 to 60 W in the 3.2 l reactor results in a decrease in
both k and T1 values. Each observed time course of the
saccharication in the larger reactors can be calculated
in the same way as that in the small scale reactor. The
calculated time courses shown as solid lines in Fig. 5 are
found to be in a good agreement with the observed ones,
which conrms the applicability of the kinetic model to
the scaled-up saccharication process and the assumption of the constant values for both KM and KI independent of the substrate property and reactor scale. The
much less remarkable two stage time course for NUKP
was observed in the larger scale reactors. This is considered to be due to the decrease in the ultrasonic
intensity per reaction volume leading to a reduced eect
of loosening of the microbrils network based on the
mechanism mentioned earlier.
3.3.2. Eects of reactor scale and ultrasonic intensity
on kinetic model parameters for various substrates
As shown in Table 4, the equilibrium constant K
governing the glucose production is found to be a constant inherent in each substrate irrespective of the reactor
scale as well as the ultrasonic intensity. The data on the
values of T1 obtained for newspaper (Figs. 2 and 5(a))
showed that the value of T1 increased steadily with
increasing ultrasonic intensities up to 30 W in any scale
of reactor, whereas it tended to decrease with a further

381

increase in the ultrasonic intensity up to 60 W. The

reason for such an adverse eect of the higher intensity
might be that some impurities in newspaper such as lignin were activated by the intense ultrasonic irradiation to
ow plastically and diuse extensively among the microbrils to form a rigid and close network of cellulose
ber which is much more recalcitrant to cellulase. As
seen in Table 4, the value of k for newspaper showed a
similar trend with increasing ultrasonic intensities in the
3.2 l reactor although it did show a dierent trend in the
6.4 l reactor, in which increasing ultrasonic intensity up
to 60 W resulted in the highest value of k. This suggests
that the detrimental eect on the k value of the intense
ultrasonic irradiation was relaxed with increasing reactor
scales. Thus, for newspaper, it was found that there was
an optimal ultrasonic intensity depending on the reactor
scale in order to make the saccharication most ecient.
Due to the lower cellulose availability of carton paper
compared to that of newspaper, the values of k and T1
for carton paper were lower than those for newspaper
and increased moderately by the ultrasonic irradiation
in any scale reactor. Oce paper B and NUKP were
also more easily hydrolyzed than carton paper due to
their high cellulose availability. For these three substrates, the common trend was found to be that both k
and T1 values increased with increasing ultrasonic
intensities in the same reactor scale and decreased with
scale-up at the same ultrasonic intensity. Therefore, the
specic ultrasonic intensity Is =V dened as the ultrasonic intensity Is divided by the reaction volume V was
proposed as the key factor for quantifying the
enhancement eect of the ultrasonic irradiation on
the enzymatic hydrolysis of a cellulosic substrate in the
dierent scale reactor.
3.4. Relation of apparent rate constant and ultimate
conversion to specic ultrasonic intensity for
various substrates
Based on the discussion mentioned above, the
apparent rate constant k and apparent ultimate conversion Xf were plotted as a function of the specic
ultrasonic intensity Is =V as shown in Fig. 6(a)(d) for
newspaper, carton paper, oce paper B and NUKP,

Table 4
Kinetic model parameters for waste paper and NUKP in 3.2 and 6.4 l stirred tank reactors with dierent ultrasonic intensities (S0 7:5 g l
1 )
Is (W)

3.2 l reactor
0

6.4 l reactor
30

60

1

k (h )
Newspaper
Carton paper
Oce paper B
NUKP

13
8.0
14
10

KM 13:4 g l
1 , KI 0:704 g l
1 .

18
9.6
17
16

15
11
19
19

060

30

60

15
7.4
16
12

16
9.2
18
14

1

K ()

k (h )

2.0
1.0
3.0
2.7

12
5.6
14
8.5

060
K ()
2.0
1.0
3.0
2.7

382

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

(b) Carton

(a) Newspaper
20

10

10

0.8 l 3.2 l 6.4 l

0W
30 W
60 W

0
1

Xf (- )

Xf (- )

0
1

0.5

0.5

0
0

10

20
Is /V (W l- 1)

30

40

(c) Office paper B

10

20
Is /V (W l- 1)

30

40

(d) NUKP
20
k (h- 1)

k (h- 1)

20

0.8 l 3.2 l 6.4 l

10

10

0.8 l 3.2 l 6.4 l

0W
30 W
60 W

Xf (- )

0.5

0W
30 W
60 W

0
1

0
1

Xf (- )

0.8 l 3.2 l 6.4 l

0W
30 W
60 W

k (h- 1)

k (h- 1)

20

10

20
Is /V (W l- 1)

30

40

0.5

0
0

10

20

30

40

Is /V (W l- 1)

Fig. 6. Relation of apparent rate constant k and apparent ultimate conversion Xf to specic ultrasonic intensity Is =V for (a) newspaper, (b) carton
paper, (c) oce paper B and (d) NUKP in various scale reactors. Solid lines represent the empirical correlation curves.

respectively. It is seen from the gures that for any

substrate except newspaper the apparent rate constant k
increases with increasing specic ultrasonic intensities
Is =V and the increment reaches a plateau above a certain
specic ultrasonic intensity of around 20 W l
1 , and that
the apparent ultimate conversion Xf of each substrate
except newspaper shows a trend similar to the above
result for k with increasing Is =V . The results on newspaper shown in Fig. 6(a) show a unique behavior which
is characterized by the presence of the optimal value of
Is =V giving the maximal values of k and Xf . This
behavior arises from an adverse eect of the intense
irradiation on the enzymatic hydrolysis of newspaper as
described earlier.

According to the mechanism proposed in Section 3.2,

the value of T1 corresponding to Xf reects the thickness of the reaction zone. The value of k represents an
overall hydrolysis rate of the microbril network of pulp
ber under a synergistic action of EG and CBH with
their adsorption and desorption. The enhancement effects of the continuous ultrasonic irradiation on T1 and
k are considered to be due mainly to an enhanced diffusion of EG and CBH and to an increased frequency of
adsorption and desorption of EG and CBH, respectively, through the ultrasonic eect on loosening of the
microbril network. The value of k should approach a
constant with increasing ultrasonic intensities due to the
recalcitrance of the microbril network as well as the

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

limitation of the adsorption/desorption frequency of EG

and CBH.
3.5. Prediction of time courses of enzymatic hydrolysis
of waste papers in the ultrasonic stirred tank reactor
For a substrate shown in Fig. 6, the values of k and
Xf , i.e., T1 can be estimated at a given value of the
specic ultrasonic intensity Is =V . It should be noted that
the relations of Xf to Is =V shown in Fig. 6 are applicable
to the case of S0 =E0 7:5=0:4 18:75, whereas those of
k to Is =V are independent of S0 =E0 . Fig. 4 is an example
of the values of Xf expressed as a function of both S0 =E0
and Is =V . An analysis on Xf except for newspaper in
terms of the two variables above resulted in the following empirical correlation:
Xf
Xf0 =Xf0

fXf
Xf0 =Xf0 gmax Is =V

a Is =V

383

where Xf0 means Xf at Is =V 0, Xf
Xf0 Xf0 shows the

fractional increase of Xf due to Is =V , fXf
Xf0 =Xf0 gmax
is the maximum fractional increase of Xf for Is =V 1
and a corresponds to the Is =V value for one half of the
maximum Xf increase. The maximum Xf increase was
found to be a linear function of logS0 =E0 . Thus, using
such generalized correlations of k and Xf as those above,
the time courses of the total sugar T and glucose G
produced during enzymatic hydrolysis of a given cellulosic substrate can be predicted under any operating
condition such as initial enzyme and substrate concentrations, ultrasonic intensity and reactor scale. Fig. 7
shows the comparison of the time courses of T calculated using the method above with the observed ones,
which were not used in the kinetic analysis described
earlier. The predicted time courses are seen to agree
fairly with the observed time courses.

9
4. Conclusions
In this work, the enzymatic hydrolyses of waste
papers such as newspaper, carton paper, oce paper in
the various scale ultrasonic stirred tank reactors were
studied and the eects of the property of waste paper,
ultrasonic intensity and reactor scale were examined
based on our previous kinetic model. The results obtained in this work are summarized as follows:

15
(a) Carton
Is=45 W
V=0.8 l
S0=25 g l-1

T (g l- 1)

10

k: estimated from Fig. 6(b)

T : estimated from Eq. (9)
0
0

12

24
t (h)

36

48

20
(b) NUKP
Is=30 W
V=0.8 l
S0=25 g l-1

T (g l- 1)

15

10

k: estimated from Fig. 6(d)

T : estimated from Eq. (9)
0

12

24
t (h)

36

48

Fig. 7. Prediction of time course of total sugar T in enzymatic saccharication of (a) carton paper and (b) NUKP in the 0.8 l ultrasonic
stirred tank reactor (S0 25 g l
1 ). Solid lines show the time courses
predicted from Eq. (5) with kinetic model parameters estimated.

(1) The saccharication of the waste papers used was

eciently enhanced by an ultrasonic intensity of 30
W even in scaling up from 0.8 to 3.2 l, but less eciently enhanced in scaling up from 3.2 to 6.4 l. This
eect could be attributed to the decrease in specic
ultrasonic intensity with scale-up. It was also found
that heat generated by ultrasonic irradiation was
more easily controlled in the larger scale reactors.
(2) The waste papers except newspaper were more eciently hydrolyzed with increasing specic ultrasonic
intensities. The reason for the adverse eect of the
higher specic intensity on the hydrolysis of newspaper was considered to be that some impurities such
as lignin were activated by the intense ultrasonic
irradiation to form a rigid and closed network,
which inhibited the access and adsorption of cellulase to the substrate leading to retarded hydrolysis.
(3) The kinetic model developed previously was applicable to the enzymatic saccharication of various
waste papers in dierent scale ultrasonic reactors.
The apparent ultimate conversion of each substrate
could be correlated as a function of the ratio of the
initial substrate to enzyme concentrations with the
specic ultrasonic intensity xed.
(4) The apparent rate constant k as well as the apparent
ultimate conversion Xf increased and then tended
to approach a constant with increasing specic

384

C. Li et al. / Ultrasonics Sonochemistry 12 (2005) 373384

ultrasonic intensities Is =V for all the substrates except newspaper. For newspaper, there existed an
optimal value of Is =V giving the maximal values of
k and Xf .
(5) The correlation of k with Is =V and that of Xf with Is =V
and S0 =E0 could be used to predict the time courses of
total sugar and glucose produced during enzymatic
hydrolysis of the substrates examined in the ultrasonic reactor under a given operating condition.
References
[1]
[2]
[3]
[4]
[5]

J.P.H. van Wyk, Trends Biotechnol. 19 (2001) 172.

S.J.B. Du, W.D. Murray, Bioresour. Technol. 55 (1996) 1.
L.P. Walker, D.B. Wilson, Bioresour. Technol. 36 (1991) 3.
J. Sheehan, M. Himmel, Biotechnol. Prog. 15 (1999) 817.
K. Nakao, C. Li, N. Tsukuda, Y. Yasuda, M. Yoshimoto, K.
Fukunaga, in: Proceedings of the 9th APCChE Congress (2002),
No. 889.
[6] A.A. Klyosov, Biochemistry 29 (1990) 10577.

[7] S.D. Manseld, C. Mooney, J.N. Saddler, Biotechnol. Prog. 15

(1999) 804.
[8] K. Nakao, K. Fukunaga, Y. Yasuda, Y. Tejima, M. Kimura,
Kagaku Kogaku Ronbunshu 17 (1991) 882.
[9] U. Mais, A.R. Esteghlalian, J.N. Saddler, S.D. Manseld, Appl.
Biochem. Biotechnol. 98100 (2002) 815.
 Albertsson, B. Hahn-Haegerdal,
[10] F. Tjerneld, I. Persson, P.A.
Biotechnol. Bioeng. 27 (1985) 1044.
[11] F. Tjerneld, I. Persson, J.M. Lee, Biotechnol. Bioeng. 37 (1991)
876.
[12] Y. Zheng, H.M. Lin, J. Wen, N. Cao, X. Yu, G.T. Tsao,
Biotechnol. Lett. 17 (1995) 845.
[13] J.A. Howell, M. Mangat, Biotechnol. Bioeng. 20 (1978)
847.
[14] G. Beldman, G. Voragen, F.M. Rombouts, M.F. Searle-van
Leeuwen, W. Pilnik, Biotechnol. Bioeng. 30 (1987) 251.
[15] J.W. Park, K. Park, H. Song, H. Shin, J. Biotechnol. 93 (2002)
203.
[16] K. Nakao, K. Fukunaga, Y. Yasuda, M. Kimura, Ann. N.Y.
Acad. Sci. 613 (1990) 802.
[17] Y. Chisti, Trends Biotechnol. 21 (2003) 89.
[18] M. Mandels, R. Andreotti, C. Roche, Biotechnol. Bioeng. Symp.
6 (1976) 21.