Journal of Colloid and Interface Science 249, 253–255 (2002)doi:10.1006/jcis.2001.8178, available online at http://www.idealibrary.com on

NOTE

Coagulation–Flocculation of Beech Condensate:Particles Size Distribution

Beech wood (Fagus sylvatica L.) condensate from a steaming op-eration was studied. The objective of our work was to study theprecipitation of these wood extracts in presence of calcium ion afterautoxidation at basic pH (8). The autoxidation was carried out at250 rpm for 30 min, and flocculation was followed up for 30 min.An investigation with a laser sizer Mastersizer of Malvern has beendone in order to study the influence of the agitation on the state ofaggregation of the condensate. A negative correlation was observedbetween the mean size of particles and the agitation rate. Withoutstirring, flocculation rapidly occurred and the mean size of particleswas high. Calcium-induced aggregation of the condensate was alsofound to be reversible toward agitation. C© 2002 Elsevier Science (USA)

Key Words: condensate; beech wood; particle sizing; flocculation;aggregation.

INTRODUCTION

Steaming of green sawing is largely used by wood industries in order to changethe color of wood, to facilitate tooling of tough wood or to permit a betterpenetration of preservation products (1–6). However, companies are worriedabout the environmental impact of the condensate resulting from this technology.In fact, pollution charges such as COD and BOD5 of this condensate are high.Wood extracts are a complex mixture of molecules with varying molecularweight and chemical nature. Some compounds such as carbohydrates, phenolicacids, aldehyde, lignin carbohydrates complex (LCC), and condensed tanninsare present in the beech wood condensate (7, 8).

Thus, the phenolic compounds may exert significant toxicity toward the mixedmicrobial communities and present serious treatment difficulties as they resistbiological treatment (9). The disposal of condensate from steaming or hot watertreatment represents a hazard to the environment (10). The detoxification ofaqueous bark extracts by oxidation was studied (11). It is reported that chemi-cal coagulation and precipitation is a phenomenon that is highly pH dependent.Randtke (12) summarized the influence of pH on chemical coagulation. At ahigher pH, the OH− ion mixed with organic compounds for metal adsorptionsites, and the precipitation of metal hydroxides occurs by coprecipitation. Thenthe coagulating species become less positively charged, diminishing their attrac-tion to the anionic organic compounds (13). The objective of this work is to studythe influence of agitation on the aggregation of particles after autoxidation at pH8. The size of particles was determined with a laser sizer Mastersizer of Malvern.

MATERIALS AND METHODS

Oxidation and Flocculation

One liter of waste waters was brought to pH 8 by NaOH addition. The autox-idation was conducted at 250 rpm for 30 min. After addition of CaCl2 (0.4 g),

25

waste waters were introduced in the stirring cell of the laser sizer Mastersizerof Malvern and flocculation was carried out for 30 min under variable agitationconditions.

Determination of Mean Particles Size

The mean size of particles was determined with a laser sizer Mastersizerof Malvern. The installation involved a measurement cell 2.5 mm thick and astirring cell, which was a Plexiglas pot of 1 liter with an anti-vortex system. Apipe circuit connected the two cells. A pump, located on the backward way tothe stirring cell, ensured the suspension circulation. The lens focal distance was300 µm. This allowed the studying of a particle size range from 1 to 600 µm.The influence of the pump rate and the stirring rate on the aggregation statewas investigated. Indirectly, the measurement inside the cell takes into accountthe hydrodynamics in the stirring system and in the tubular. We consider twostirring processes as comparable if the wasted energies are comparable. Thespecific dissipation energy in the cell (Ea) may be obtained from the averagevelocity gradient G (14)

G =√

P/Vc

µ1, [1]

where P is total wasted energy

Ea = P/Vc, [2]

Vc is the cell volume, and µ1 is the liquid dynamic viscosity, considered equalto that of water: µ1 = 10−3 kg m−1 s−1.

The specific energy (Et) wasted in the connection pipes was calculated ac-cording to Filippov (15) by the formula

Et = 8ν

π2 R6· V 2

s , [3]

where Vs is the pump rate, ν is the liquid cinematic viscosity, and R is the piperadius.

The use of a peristaltic pump with rate in the range 0 to 537.5 cm3/minallows the specific dissipation energy in the pipe connections to vary from 0.0 to0.266 W/kg. Note Poiseille’s assumption (Re ≤ 2300) is verified so long asQp < 540 cm3/min, thus, for the whole experience.

RESULTS AND DISCUSSION

Effect of Pump Rate on Aggregation State

Figure 1 shows the influence of the pump rate on the aggregation state.A negative correlation was observed between the mean size of the particlesand the agitation rate. A monomodal distribution of particles was observedfor pump rates at 43 and 35 ml/min with a population of particles with rel-atively low mean size. A bimodal distribution was formed for a lower pump

3 0021-9797/02 $35.00C© 2002 Elsevier Science (USA)

All rights reserved.

254 NO

FIG. 1. Effect of pump rate on size distribution of particles formed after au-toxidation and addition of CaCl2. Pump rate was 43 (- -�- -), 35 (—�—), 27 ml/min (—�—), and 19 ml/min ( ).

FIG. 2. Size distribution of particles formed with (thin line) and withoutstirring (thick line). Stirring rate was 48 rpm and pump rate was 19 ml/min.

FIG. 3. Evolution of particle size after stirring has stopped. Stirring ratewas 48 rpm, and pump rate was 19 ml/min.

TE

FIG. 4. Size distribution of particles formed before (solid line) and after(dashed line) redispersion of flocculated waste waters.

rate 27 ml/min. A second population of particles appeared with a higher meansize. Decreasing the pump rate (19 ml/min) led to a rise in the populationof particles with higher mean size, indicating coagulation–flocculation of thecondensate.

Effect of Stirring Rate on Aggregation State

Experiments were done with a constant pump rate of 19 ml/min. No significantflocculation was observed with a stirring rate at 200 rpm. However, a stirringrate at 48 rpm makes it possible to obtain a monomodal distribution of particles(Fig. 2). An increase of the mean particles size was observed (Figs. 2 and 3)without stirring due to the decreasing specific dissipation energy in the cell.Thus, sedimentation occurs in the stirring cell. Figure 4 shows flocculated wastewaters behavior after a redispersion cycle. The size distributions of particlesformed before and after redispersion were similar.

CONCLUSION

A laser size Mastersizer of Malvern was used to study the influence of agi-tation on the state of particle aggregation. A negative correlation between themean size of particles and the agitation rate was observed. Without stirring,flocculation rapidly occurred and the mean size of particles was high. Calcium-induced aggregation of the beech condensate was also found to be reversibletoward agitation. On the other hand, the particle size distribution observed in anaqueous suspension can be connected with transformation of the extracts andsubsequent rearrangement due to the oxidation process and the presence of cal-cium ion. In the end, precipitation of these chemical compounds by autoxidationfollowed by flocculation can be involved in order to purify these beech woodcondensates.

ACKNOWLEDGMENTS

The authors acknowledge the financial support provided by the ANVARLoraine and Thanry establishment. Authors thank Dr. L. O. Philippov for criticalreading of the manuscript.

REFERENCES

1. Karakas, J. A., and Voulgaris, E. V., Holz-als-Roh-und-Werkstoff 50, 74(1992).

2. Kubinsky, E., and Ifju, G., Wood Sci. 6, 87–94 (1973).3. Chen, P. Y. S., Wood Fiber 7, 222 (1975).

T

Received September 4, 2001; accepted December 10, 2001

NO

4. Chen, P. Y. S., and Workman, E. C., Jr., Wood Fiber 11, 218 (1980).5. Inoue, M., Norimoto, M., Tanahashi, M., and Rowell, R. M., Wood Fiber

Sci. 25, 224 (1993).6. Hayashi, K., Nakamura, K., Kanagawa, Y., Yasujima, M., and Aoki, K.,

J. Soc. Mater. Sci. Jpn. 44, 279 (1995).7. Irmouli, M., and Haluk, J. P., in “4eme colloque des Sciences et Industries

du bois,” p. 527, 1996.8. Irmouli, M., Haluk, J. P., Kamdem, D. P., and Charrier, B., J. Wood Chem.

Technol., in press.9. Field, J. A., Leyendeckers, M. J. H., Sierra Alvarez., R., Lettinga, G., and

Habets, L. H. A., Water Sci. Tech. 20, 219 (1988).10. Wanli, D. A., Kamdem, D. P., Loconto, P., Yanlyang pan, D., Gage, D., and

Dawson-Andoh, B. E., Wood Fiber Sci. 31, 370 (1999).11. Field, J. A., and Lettiga, G., J. Chem. Tech. Biotechnol. 49, 15

(1989).

12. Randtke, S. J., J. Am. Water Works Assoc. 80, 40 (1988).13. Ching, H.-W., Tanaka, T. S., and Elimelech, M., Water Res. 28, 559 (1994).

E 255

14. Camp, T. R., and. Stein, P. C., J. Boston Sci. Civil Eng. 147, 407 (1943).15. Filippov, L. O. Ph.D. thesis. INPL, Nancy, France, 1996.

Mohammed Irmouli∗,1

Jean Pierre Haluk†∗Ecole Superieure du Bois44306 NantesFrance†Lermab54500 Vandoeuvre les NancyFranceE-mail: mohammed.irmouli@ecolesupbois.asso.fr

1 To whom correspondence should be addressed.