Epstein N., Grace J.R. Spouted and Spout-Fluid Beds: Fundamentals and Applications

Подождите немного. Документ загружается.

Powder–particle spouted beds 177

annulus. The discharge pipe should be on the opposite side from the feeding pipe to

prevent short-circuiting of fresh particles. The position of the discharge pipe strongly

affects the composition of the steady bed, and this must be taken into consideration.

Also, for a Group B–D particle system, the freeboard height at which the elutriation

rate becomes constant, often called the transport disengaging height (TDH), is usually

achieved within the PPSB.

The effects of the apparatus structure on the elutriation rate

have been conﬁrmed.

For the application of a conventional spouted bed, as in combustion and gasiﬁcation,

attrition is caused by particle collisions, especially in the spout, so coarse particles

become smaller and ﬁnes tend to be elutriated from the spouted bed. The elutriation

of ﬁnes creates problems of decreasing thermal efﬁciency and increasing air pollution.

Moreover, when the spouted bed is used as a reactor, the freeboard above the bed provides

space not only for the disengagement of particles but also for additional reaction between

the ejected particles and the gas.

10.2 Fundamentals

To design and operate a PPSB stably for chemical or physical processes, it is important

to understand how ﬁne powders behave in a PPSB.

10.2.1 Elutriation phenomena

If the gas velocity exceeds the terminal velocity of a ﬁne powder, then the ﬁnes are

completely carried out of the bed. This phenomenon is called elutriation. Estimation

of the elutriation rate of ﬁnes from a PPSB is crucial to determining the operating gas

ﬂowrate, to estimating the reaction yield, and to recovering valuable materials such as

reactants or catalysts.

For a Group C–D particle system, Xu et al.,

who investigated the effects of operating

parameters on the holdup of ﬁne powders, interpreted their results from the viewpoint

of the cohesive force of the ﬁnes. The holdup of ﬁnes in the bed (X) was deﬁned as the

mass fraction of the ﬁnes in the mixture of ﬁne and coarse particles. This was found

to increase with increasing feed rate of the ﬁnes and their cohesive force, and with

decreasing gas velocity and particle diameter of the ﬁnes. Similar results were reported

by Ishikura et al.

Taking these factors into account, it was considered that the holdup

of ﬁnes within a PPSB is controlled by a balance between the adhesion rate of the ﬁnes

to the s urface of the coarse particles in the annulus and fountain, and the separation rate

of the ﬁnes in the spout and fountain.

An elutriation rate constant, K (units: kg/m

s), for a PPSB has been proposed based

on the mass balance of the ﬁne and coarse particles and a ﬁrst-order model for a batch

system,

3,6,7

such that the elutriation rate is

V = K · A · X/(1 − X), (10.1)

178 Toshifumi Ishikura and Hiroshi Nagashima

Coarse particles : Silica sand

Aluminium oxide

U = 0.7 m/s

3.50

1.96

1.33

0.84

2.93

2.10

1.75

0.87

0.51

Aluminium hydroxide

[μm]

= 920 μm

[μm]

–1

–2

–1

Holdup of fine powders in bed, X, wt.%

Elutriation rate constant, K, kg/(m

M = 0.55 kg

Figure 10.3. Relationship between elutriation rate constant and holdup of ﬁnes in bed with

diameter of ﬁnes varied for Group C–D particle system and bottom feeding of ﬁnes; data from

Xu.

where A is the cross-sectional area of the column and X is the holdup of ﬁnes in the bed.

Eq. (10.1) has also been applied to a continuous system using the holdup of ﬁnes at the

discharge pipe.

3,6,7

For X  0.1, Eq. (10.1) can be approximated

5,8

V = K · A · X. (10.2)

In an operation such as the one shown in Figure 10.2a, the steady-state elutriation rate

of ﬁnes (V ) is equal to the feed rate of the ﬁnes. Eq. (10.2) implies that V is proportional

to X. The elutriation rate constant (K ) at a steady state was calculated from Eq. (10.1)

or (10.2) by measuring both X and V.

From Figure 10.3, for a Group C–D particle system,

K is hardly affected by X as long

as the superﬁcial gas velocity (U) is kept constant, if the diameter of the ﬁnes is larger

than 1 μm. This ﬁnding is consistent with results

3,6,7

using ﬁne powders of diameter

greater than 100 μm, as shown in Figure 10.4. Figure 10.5 shows the effect of the

diameter of ﬁnes on the elutriation rate constant (K ) for several kinds of ﬁne powders. K

decreases with decreasing diameter of ﬁnes for the Group C–D particle system, because

the adhesive force of the ﬁnes increases, so the ﬁnes adhere to the surface of the coarse

particles more readily and do not easily separate from them. This behavior is completely

contrary to the elutriation phenomenon using ﬁnes of size greater than 100 μm,

3,6,7

also shown in Figure 10.5. Thus, for the Group B–D particle system, K increases with

decreasing diameter of the ﬁnes because the driving force for elutriation increases owing

to a decrease in the terminal velocity of the particles. Wang et al.

investigated the effect

of hydrodynamic factors on the elutriation rate of wet calcium carbonate powder from a

coarse particle bed with aeration of the annulus.

For a liquid–solid system, Ishikura et al.

10,11

investigated the effects of the operating

conditions on the elutriation rate coefﬁcient based on a ﬁrst-order model, and correlated

the elutriation rate coefﬁcient for a column height above the TDH with the modiﬁed

Powder–particle spouted beds 179

–1

–3

–2

2.07

1.87

1.71

1.56

1.47

Feed rate of solid particles, F, kg/s

Elutriation rate constant, K, kg/(m

Coarse particles : Soma sand

Fine powders : Toyoura sand

= 718 μm

= 180 μm

= 1.31 m/s

U [m/s]

= 0.5

Figure 10.4. Relationship between elutriation rate constant and feed rate of solid particles with gas

velocity varied for Group B–D particle system and top feeding of ﬁne component, above TDH.

Data from Ishikura et al.

–1

–2

–1

Group C

Group B

Aluminium oxide

Aluminium hydroxide

Toyoura sand

Glass beads

U = 0.70 m/s

U = 1.85 m/s

Particle size, d

, μm

Elutriation rate constant, K, kg/(m

Figure 10.5. Effect of diameter of ﬁnes on elutriation rate constant with several kinds of ﬁnes.

(,: data from Xu

; ♦,, : data from Ishikura et al.

3,5

)

Froude number of the ﬁne particles, deﬁned as (U–U

)

ρ/g ·d

(ρ

– ρ). The elutriation

of ﬁnes was closely related to the stratiﬁed length of the particle transport bed for the

Group B–D particle system. Al-Jabari et al.

12,13

described the experimental aspects of

the elutriation process from a conical spouted bed and used an existing ﬁrst-order model

to determine the elutriation rate coefﬁcient of ﬁnes from a pulp ﬁber suspension.

For a Group B–D particle system, the particle entrainment rate from the top of the

spout in a PPSB provides a basis for estimating the elutriation rate of the ﬁner component

180 Toshifumi Ishikura and Hiroshi Nagashima

from the column. Therefore, Ishikura et al.

14,15

attempted to estimate the entrainment

rate of a binary particle mixture f rom the top of the spout, using the mass and momentum

balance for the gas and par ticles within the spout. To interpret the decrease in elutriation

rate of ﬁnes along the freeboard in Figure 10.2b, the holdup of ﬁnes, their local velocity,

and the ﬂuctuating gas velocity at an arbitrary height of the freeboard were examined by

using a shutter plate, a ﬁber optic probe, and a hot-wire anemometer, respectively.

16,17

The exponential decrease of the elutriation rate of the ﬁnes along the freeboard was

veriﬁed.

From those experimental results, Ishikura et al.

14–16

suggested the following three

elutriation mechanisms in the freeboard:

(1) This process occurs in the fountain region. Sufﬁcient column height is required

for the formation of a spouted bed with ample freeboard for particle transport

disengagement. The velocity distribution of the gas jet leaving the top of the spout

is spread out radially, imparting a relatively large radial velocity to both the ﬁne

and coarse particles. Hence, the axial component of the particle velocity decreases

sharply.

(2) There are no coarse particles present in the freeboard, and the elutriation rate of

the ﬁnes decreases exponentially with increasing freeboard height. This decrease is

presumably caused by the exponential dissipation of the effective gas velocity in the

freeboard. It is considered that turbulent radial migration of ﬁnes occurs.

(3) The elutriation rate becomes constant above the TDH.

Based on this model, the elutriation height coefﬁcients for each process were deter-

mined, and the values could be used to estimate the elutriation rate of ﬁnes for any

column height.

7,18

The residence time of ﬁnes to be treated is an important factor in a PPSB. For example,

the conversion of reactants is determined by the residence time of the reactant, and the

residence time of the solid materials affects their treatment efﬁciency. Xu

showed that

the residence time of ﬁnes increases with decreasing diameter of ﬁnes and decreasing

gas velocity, and with increasing static bed depth and diameter of the coarse particles

in the bed. Similar results were obtained by Ishikura et al.

found that the mean

residence time of ﬁnes in the bed is 20 to 1200 times longer than that of the spouting

gas for a Group C–D particle system.

10.2.2 Other hydrodynamic characteristics

The minimum spouting velocity with ﬁne powders entrained in spouting gas (U

msF

)is

an important parameter in the design and operation of a PPSB because it signiﬁes the

minimum velocity that can be steadily processed in the bed.

For a PPSB of the type shown in Figure 10.2a, U

msF

was measured by Zhu et al.

in the same way as the minimum spouting velocity (U

) for an ordinary spouted bed

is reproducibly determined (see Chapter 2). First they spouted the bed steadily with

Powder–particle spouted beds 181

1.4

1.2

1.0

0.8

0.6

0.4

0 0.1 0.2

Fine powders : FCC particles

= 3690 μm, r

= 1259 kg/m

= 66 μm, r

= 1700 kg/m

0.145

0.200

0.290

0.340

Stability limit

[m]

Coarse particles : PVC particles

0.3 0.4 0.5 0.6 0.7 0.8

Loading ratio, R

, -

Minimum spouting velocity, U

msF

, m/s

Figure 10.6. Effect of loading ratio (based on minimum spouting velocity of coarse particles

alone) on minimum spouting velocity with ﬁne powder entrainment for different values of bed

depth (D = 144 mm, D

= 19.1 mm); data from Zhu et al.

coarse particles alone at a constant gas velocity, then introduced the ﬁnes steadily, and

ﬁnally decreased the gas ﬂow slowly until the bed collapsed. The velocity at which

the bed collapsed was then taken as U

msF

. For convenience, they deﬁned a loading

ratio based on the minimum spouting condition with coarse particles alone as R

F/(ρ ·A ·U

). Figure 10.6 shows that as the loading of the ﬁne powders increases, U

msF

decreases and becomes progressively smaller than U

for spouting with the coarse

particles alone. This trend resembles the behavior observed for continuous feeding of

the coarse particles alone to the bottom of the bed.

Finally, the increase in loading

leads to unstable spouting or slugging, and the critical points are joined by stability limit

lines, as shown in Figure 10.6.

Zhu et al.

investigated the change of spouting pressure drop with the feeding of

powders in a PPSB. At any gas velocity and bed depth, the pressure drop increased

with time during powder feeding, then the pressure drop stabilized. The stable pressure

drop increased as the feed rate of the powders, bed depth, and gas ﬂowrate increased.

However, pressure drop stabilization ceased when the feed rate became so high that

slugging occurred instead of spouting, behavior that was accompanied by large pressure

drop ﬂuctuations. In the future, the effect of ﬁnes holdup on the pressure drop of the bed

should be investigated in detail.

Maximum spoutable bed depth (H

) is an important parameter in the design and

operation of a PPSB because it is directly related to the amount of material that can be

processed batchwise in a bed. Above this particular bed depth, bubbling and slugging

begin in a PPSB. The maximum spoutable bed depth was examined by Zhu et al.

in the

same way as H

for an ordinary s pouted bed is determined reproducibly. They indicated

that H

decreases with an increase in the loading of powders fed to the bed bottom with

spouting gas. In contrast, Fane et al.

reported that H

increases with an increase in

loading of the coarse particles fed to the bottom of the column where coarse particles

are spouted alone.

182 Toshifumi Ishikura and Hiroshi Nagashima

Slurry drop

Stage A

Stage B

Stage C

Stage D

Ca(OH)

Figure 10.7. Mechanisms at work in a PPSB desulfurizer. Stage A: slurry dropping; Stage B:

reaction and drying; Stage C: diffusion controlling; Stage D: entraining. Model of Guo et al.

10.3 Applications

A PPSB can be applied to such chemical processes as desulfurization and the devel-

opment of particulate products, and to such physical processes as the separation of

ﬁnes.

10.3.1 Semidry ﬂue gas desulfurization

The desulfurization efﬁciency of the semidry ﬂue gas desulfurization (FGD) process has

not reached the level of wet scrubbers. The major reason is that the residence time of

the SO

sorbent is very short in existing processes such as spray drying. Accordingly,

the development of low-cost high-efﬁciency FGD technologies is signiﬁcant.

Guo et al.

developed a new type of semidry FGD process using a PPSB as a main

reactor. In this process, droplets of ﬁne SO

sorbent slurry were fed continuously into

a spouted bed, in which coarse inert particles were spouted with hot gas containing

. This process enabled the long time reaction of sorbent powders with SO

in the

liquid and gas phases, as well as in drying the spent sorbent powders, because they

adhered strongly to the surface of the coarse particles. Finally, the dried spent sorbent

was elutriated from the spouted bed and collected by a bag ﬁlter, and effective SO

removal was thus achieved.

As shown in Figure 10.7, a mechanism of spouted bed desulfurization was indicated

by Guo et al.,

who suggested the following four stages related to SO

removal in a

Powder–particle spouted beds 183

PPSB. Stage A: Slurry drops collide with coarse particles, and a ﬁlm of sorbent solution

begins to react with the ﬂue gas. Stage B: Slurry on the surface of the coarse particles

begins to evaporate and the sorbent powders adhering to the coarse particles react with

in a humidiﬁed ﬂue gas. Stage C: Water moves to the outside surface of the sorbent

and SO

diffuses from the outside to the inside of the sorbent. New surface of the sorbent

is exposed owing to collision and separation of particles. Stage D: Reacted and dried

sorbent particles are entrained.

From this description, the authors pointed out that using a PPSB in the semidry FGD

process offers the following advantages:

(1) The properties of the coarse particles allow the system to operate at a high gas

velocity, as shown in Figure 10.1.

(2) As t he coarse particles are spouted, the ﬁne SO

sorbent is well dispersed.

(3) The mean residence time of the sorbent in the reactor is much longer.

(4) The drying of the slurry, as well as the reaction between SO

and the sorbent, occur

simultaneously in the reactor.

For suitable operation, design, and scaleup of a PPSB, it is important to understand

the hydrodynamics and the chemical reaction phenomena in terms of the main operating

variables. Many researchers

21–27

have examined the effect of such operating parameters

as types of SO

sorbent, particle diameter of SO

sorbent, apparent mean residence time

of gas i n the bed, sorbent-to-SO

stoichiometric ratio, approach to saturation temperature,

static bed depth, and gas velocity on the SO

removal efﬁciency – that is, the removal

limit of SO

in the ﬂue gas.

Ma et al.

tried to use limestone to remove SO

from the ﬂue gas in the semidry FGD

process with a PPSB. Reactivity of SO

was compared with limestone and hydrated

lime based on their structural properties. Limestone was indicated as an attractive SO

sorbent by choosing the appropriate conditions in terms of removal efﬁciency, price,

and availability of sorbent. Nakazato et al.

investigated semidry FGD in a PPSB with

several kinds of SO

sorbents, including slaked lime, limestone, magnesium hydroxide,

and industrial alkaline waste. They pointed out that an increase in the apparent mean

residence time of the gas in the bed improves SO

removal efﬁciency. However, the

apparent mean residence time of gas is a function of bed depth and gas velocity. There

is a minimum spouting velocity (U

msF

) and a maximum spoutable bed depth (H

)for

any speciﬁc PPSB, as mentioned previously. This implies that there is a limited operat-

ing range of possible contacting times for a speciﬁc PPSB. Moeini and Hatamipour

examined two models for SO

removal f rom ﬂue gas in a PPSB with hydrated lime as the

sorbent. Their analysis was based on one-dimensional

and streamtube models

proposed previously for a conventional spouted bed. Predictions from these two models

were compared with published experimental data.

It was found that the two models

are both valid, as they are indeed equivalent.

For practical applications, Xu et al.

proposed a novel design of a multiple large-scale PPSB suitable for commercial uti-

lization, as a small PPSB is limited in throughput. In the future, it will be important to

examine the applicability of such a design.

184 Toshifumi Ishikura and Hiroshi Nagashima

10.3.2 Other applications

Nakazato et al.

developed a new simple process for semidry production of very ﬁne

calcium carbonate powder using a PPSB. The PPSB provided a multitask reactor in which

the dispersion of the hydrated lime slurry, the reaction of the hydrated lime with carbon

dioxide, the drying of the produced calcium carbonate, and its pulverization occurred

simultaneously in the bed. This semidry PPSB process produced calcium carbonate

powders with a mean particle diameter of ∼1 μm and hydrated lime conversion of

about 90 percent. Similar research was reported by Lin et al.

In a liquid–solid system, Al-Jabari et al.

12,13

developed a simple new technique for

separating ﬁne powders from pulp ﬁber suspension, using a conical spouted bed, in a

process called elutriation spouting. At suitable operating conditions, the ﬁbers remained

in the circulating zone, whereas ﬁne powders moved upward with the liquid and were

elutriated efﬁciently.

PPSBs show various ﬂow patterns depending on the properties of particles, such as

the diameters of the coarse and the ﬁne particles. Therefore, PPSBs are likely to be used

for a wide variety of chemical and physical processes in the future.

Chapter-speciﬁc nomenclature

F feed rate of solid particles, kg/s

FGD ﬂue gas desulfurization

K elutriation rate constant of ﬁne powders, kg/(m

·s)

loading ratio based on minimum spouting of coarse particles alone

TDH transport disengagement height, m

msF

minimum spouting velocity with ﬁne powders entrainment, m/s

terminal velocity of a particle, m/s

V elutriation rate of ﬁne powders, kg/s

X holdup of ﬁne powders in bed

holdup of ﬁne powders in feed

References

1. D. Geldart. Types of gas ﬂuidization. Powder Technol., 7 (1973), 285–292.

2. L. A. P. Freitas and J. T. Freire. Analysis of ﬂuid dynamics in a spouted bed with continuous

solids feeding. Drying Technol., 16 (1998), 1903–1921.

3. T. Ishikura, I. Tanaka, and H. Shinohara. Elutriation of ﬁnes from spouted bed: effects of

apparatus structure, J. Soc. Powder Technol. Japan, 13 (1976), 535–540.

4. J. Xu, Y. Washizu, N. Nakagawa, and K. Kato. Hold-up of ﬁne particles in a powder-particle

spouted bed. J. Chem. Eng. Japan, 31 (1998), 61–66.

5. T. Ishikura, H. Nagashima, and M. Ide. Behaviour of cohesive powders in a powder-particle

spouted bed. Can. J. Chem. Eng., 82 (2004), 102–109.

Powder–particle spouted beds 185

6. T. Ishikura, S. Suizu, I. Tanaka, and H. Shinohara. Elutriation of ﬁnes from a small spouted

bed in solid-gas system. Kagaku Kogaku Ronbunshu, 5 (1979), 149–154.

7. T. Ishikura and H. Shinohara. Estimation of elutriation rate of ﬁne particles from a continuous

spouted bed in solid-gas system. Fukuoka Daigaku Kogaku Shuho, 40 (1988), 245–252.

8. J. Xu. Investigation on the powder-particle spouted bed. Ph.D thesis, Gunma University, Japan

(1998).

9. G. Wang, L. Ding, and W. Guo. Dr ying and elutriation of wet powder particle in spouted bed

with aeration. J. Shenyang Inst. Chem. Technol., 15 (2001), 8–12.

10. T. Ishikura, I. Tanaka, and H. Shinohara. Elutriation of particles from a batch spouted bed in

solid-liquid system. J. Chem. Eng. Japan, 12 (1979), 148–151.

11. T. Ishikura and I. Tanaka. Behavior and removal of ﬁne particles in a liquid-solid spouted bed

consisting of binary mixtures. Can. J. Chem. Eng., 70 (1992), 880–886.

12. M. Al-Jabari, M. E. Weber, and T. G. M. Van de Ven. Particle elutriation from a spouted bed

of recycled pulp ﬁbres. J. Pulp and Paper Sci., 22 (1996), J231–J236.

13. M. Al-Jabari, M. E. Weber, and T. G. M. Van de Ven. Modeling ﬁnes elutriation from a

spouted bed of pulp ﬁbers. Chem. Eng. Comm., 148–150 (1996), 456–476.

14. T. Ishikura, I. Tanaka, and H. Shinohara. Elutriation of ﬁnes from a continuous spouted bed

in solid-gas system: Effects of column height on elutriation rate of ﬁnes. Kagaku K o gaku

Ronbunshu., 6 (1980), 448–454.

15. T. Ishikura, H. Shinohara, and I. Tanaka. Elutriation of ﬁnes from a continuous spouted bed

containing a solid-gas system: Effect of column height on the ﬁnes elutriation rate. Internat.

Chem. Eng., 22 (1982), 346–354.

16. T. Ishikura, H. Shinohara, and I. Tanaka. Behavior of ﬁne particles in a spouted bed consisting

of ﬁne and coarse particles. Can. J. Chem. Eng., 61 (1983), 317–324.

17. T. Ishikura and K. Funatsu. Behavior of ﬁne particles and gas in freeboard of a gas-solid

spouted bed consisting of ﬁne and coarse particles. Fukuoka Daigaku Kogaku Shuho, 35

(1985), 103–115.

18. T. Ishikura and H. Shinohara. Estimation of elutriation rate of ﬁne particles from a continuous

spouted bed in solid-gas system: Based on particle transport mechanism in the freeboard

region. Micromeritics, 32 (1988), 42–48.

19. Q. Zhu, C. J. Lim, N. Epstein, and H. T. Bi. Hydrodynamic characteristics of a powder-particle

spouted bed with powder entrained in spouting gas. Can. J. Chem. Eng., 83 (2005), 644–651.

20. A. G. Fane, A. E. Firek, and C. W. P. Wong. Spouting with a solids-laden gas stream.

CHEMECA, 85 (1985), 465–469.

21. Q. Guo, N. Iwata, and K. Kato. Process development of effective semidry ﬂue gas desulfur-

ization by powder-particle spouted bed, Kagaku Kogaku Ronbunshu, 22 (1996), 1400–1407.

22. Q. Guo and K. Kato. The effect of operating condition on SO

removal in semidry desul-

furization process by powder-particle spouted bed. Kagaku Kogaku Ronbunshu, 24 (1998),

279–284.

23. X. Ma, T. Kaneko, Q. Guo, G. Xu, and K. Kato. Removal of SO

from ﬂue gas using a new

semidry ﬂue gas desulfurization process with a powder-particle spouted bed, Can. J. Chem.

Eng., 77 (1999), 356–362.

24. X. Ma, T. Kaneko, T. Tashimo, T. Yoshida, and K. Kato. Use of limestone for SO

removal

from ﬂue gas in the semidry FGD process with a powder-particle spouted bed. Chem. Eng.

Sci., 55 (2000), 4643–4652.

25. G. Xu, Q. Guo, T. Kaneko, and K. Kato. A new semidry desulfurization process using a

powder-particle spouted bed. Adv. Environ. Res., 4 (2000), 9–18.

186 Toshifumi Ishikura and Hiroshi Nagashima

26. T. Nakazato, Y. Liu, and K. Kato. Removal of SO

in semidry ﬂue gas desulfurization process

with a powder-particle spouted bed. Can. J. Chem. Eng., 82 (2004), 110–115.

27. M. Moeini and M. S. Hatamipour. Flue gas desulfurization by a powder-particle spouted bed.

Chem. Eng. Technol., 31 (2008), 71–82.

28. K. B. Mathur and C. J. Lim. Vapour phase chemical reaction in spouted bed: a theoretical

model, Chem. Eng. Sci., 29 (1974), 789–797.

29. C. J. Lim and K. B. Mathur. A ﬂow model for gas movement in spouted beds. AIChE J., 22

(1976), 674–680.

30. C. M. H. Brereton and J. R. Grace. A note on comparison of spouted bed reactor models,

Chem.Eng.Sci., 39 (1984), 1315–1317.

31. T. Nakazato, Y. Liu, K. Sato, and K. Kato. Semi-dry process for production of very ﬁne

calcium carbonate powder by a powder-particle spouted bed. J. Chem. Eng. Japan, 35 (2002),

409–414.

32. C. Lin, T. Zhu, Y. Zhu, and J. Zhang. Experiments and modeling of the preparation of ultraﬁne

calcium carbonate in spouted beds with inert particles. Chinese J. Chem. Eng., 11 (2003),

726–730.