- Email: [email protected]
Effects of ultrasound on adsorption–desorption of p-chlorophenol on granular activated carbon
Ultrasonics Sonochemistry 10 (2003) 109–114 www.elsevier.com/locate/ultsonch
Effects of ultrasound on adsorption–desorption of p-chlorophenol on granular activated carbon O. Hamdaoui a, E. Naffrechoux
b,*
, L. Tifouti a, C. Petrier
b
a
b
Department of Process Engineering, Faculty of Engineering, University of Annaba, P.O. Box 12, 23000 Annaba, Algeria Laboratoire de Chimie Mol eculaire et Environnement, ESIGEC, Universit e de Savoie, BP 1104, 73376 Le Bourget du Lac Cedex, France Received 17 June 2002; accepted 23 September 2002
Abstract The aim of this work is the evaluation of the effects of ultrasound on p-chlorophenol adsorption–desorption on granular activated carbon. Adsorption equilibrium experiments and batch kinetics studies were carried out in the presence and the absence of ultrasound at 21 kHz. Results indicate that the adsorption of p-chlorophenol determined in the presence of ultrasound is lower than the adsorption observed in the absence of ultrasound. Desorption of p-chlorophenol from activated carbon with and without the application of ultrasound was studied. The desorption rates were favoured by increased ultrasound intensity. This rise is more noticeable as temperature increases. The addition of ethanol or NaOH to the system causes an enhancement of the amount of p-chlorophenol desorbed, especially in the presence of ultrasound. A synergetic enhancement of the desorption rate was observed when ultrasonic irradiation was coupled with ethanol chemical regeneration. Ó 2002 Elsevier Science B.V. All rights reserved. Keywords: Ultrasound; Adsorption; Desorption; Activated carbon; p-chlorophenol
1. Introduction Granular activated carbon (GAC) is used in many processes for the treatment of drinking water and industrial wastewaters. It provides a convenient technology for removing a broad range of organic pollutants, which are generally of concern because of their toxicity to human health. Phenols are regarded as priority pollutants because of their toxicity in environment at low concentrations. Many of them have been classified as hazardous pollutants because of their potential harm to human health. There is a considerable amount of data in the literature concerning adsorption of phenols onto GAC [1–4]. Although the adsorption of phenolic compounds on GAC is easy to manage, the process of regenerating the adsorbent poses a major challenge because of the high affinity of the phenolic compounds to the sorbent sur* Corresponding author. Tel.: +33-479-758830; fax: +33-479758674. E-mail address: emmanuel.naff[email protected] (E. Naffrechoux).
face. The most common techniques for desorption of phenols from activated carbon are thermal regeneration, chemical regeneration, and to a smaller extent, bioregeneration and regeneration under supercritical condition. As these techniques are presenting drawbacks, it is of interest to explore other methods of desorption. Currently, one of the regeneration way which is considered is the desorption by ultrasound. Qin et al. investigated the effect of ultrasound on the desorption of phenol from NKA II resin, and pointed out that ultrasound had ‘‘spot energy effects’’ which would enhance the process [5]. Ultrasound (1440 kHz) was used by Rege et al. [6] to realize the desorption of phenol from activated carbon and polymeric resin adsorbents. It was found that the desorption rates were helped by temperature decrease, air saturation of the liquid medium and increase of the ultrasonic intensity. The rate enhancement was attributed to acoustic vortex microstreaming. Breitbach et al. [7] found that ultrasound promotes the desorption of benzoic acid from polymeric resin. An important factor influencing ultrasound enhanced desorption was the temperature rise due to ultrasonication.
1350-4177/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved. PII: S 1 3 5 0 - 4 1 7 7 ( 0 2 ) 0 0 1 3 7 - 2
110
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
Another ultrasonic effect is supposed to occur because regeneration is more effective at 610 kHz than at 204 kHz for a same operating temperature. Using NKA II resin, Li et al. [8] found that the adsorption of phenol determined in the presence of ultrasound (39 kHz, ultrasonic bath) is less than that in the absence of ultrasound. It was also observed that the stronger the acoustic power delivered to the system, the lower the adsorption capacity of phenol on the resin was. Chlorophenols are produced during water chlorination and can only be removed from water during the last treatment stage by adsorption. It seems of importance to test the alternative acoustic regeneration of GAC when the adsorbant is saturated with the micropollutant. Object of this work was to compare the adsorption of p-chlorophenol on GAC in the absence and in the presence of 21 kHz ultrasound at ambient temperature. In a second part, the ultrasound effect on the desorption of p-chlorophenol from GAC is supplied to chemical regeneration method in order to evaluate the efficiency of the combined process.
2. Materials and methods 2.1. Reagent and activated carbon p-chlorophenol (p-CP) with purity greater than 99.5% (Sigma) and ultra-high quality water (Elga elgastat) were used to prepare the aqueous solutions for the different tests. The GAC (Prolabo) has a BET surface area of 929 m2 g1 and a mean granulometry of 3 mm. The carbon was pretreated by boiling in ultra-high quality (UHQ) water for 1 h and washed thoroughly with UHQ water until the UV absorbance intensity was equal to zero. Finally, the washed GAC was dried in an oven at 110 °C to constant weight before use. The particle size analysis of GAC was realised before and after treatment. A suspension of GAC (0.6 g) in pure water (90 ml) was exposed to ultrasound (15.2 and 38.3 W) or stirred at 400 rpm during 7 h. The suspension was then filtered (0.45 lm) and dried at 110 °C for 24 h. The particle size distribution was performed with a laser granulometer (Mastersizer 2000––Malvern Instruments). The size measurement range and detection limit were 0.02–5000 lm and 0.01 lm respectively.
2.3. Apparatus Batch adsorption–desorption experiments were carried out using 100 ml (fluid volume) reactor as shown in Fig. 1. It consists of a double cylindrical jacket, allowing water cooling of the reactor. This reactor is made of a glass cylinder of 7.9 cm height and internal diameter 4.1 cm. The top of the glass reactor received the ultrasonic transducer. The 21 kHz ultrasonic wave was emitted from a titanium horn (diameter 3.5 cm) connected to a commercial supply Sinaptec Nexus II. Determination of the acoustical energy absorbed in the reactor was achieved following the calorimetric method [9]. The volume treated was 90 ml for a liquid height of 6.8 cm. 2.4. Adsorption procedure Adsorption isotherm under normal condition was carried out in batch mode with the following procedure. Various amounts of GAC (0.05–1 g l1 ) were continuously stirred at 21 °C with 90 ml of p-CP aqueous solution (100 mg l1 ). The stirring rate was 400 rpm. Equilibrium was reached after a contact time of 4 days. Samples of 2 ml were withdrawn for analysis using UV– visible spectrophotometer to determine the concentration of p-CP remaining in the fluid phase, and the amounts of p-CP adsorbed on the GAC were inferred from the mass balance. For the kinetic studies 1 g of activated carbon was continuously stirred at 21 °C with 1 l of aqueous p-CP solution (100 mg l1 ). The analyses of the supernatant solution were achieved as previously described. In order to determine the isotherm of p-CP on the GAC under ultrasonic field, the system was conducted to equilibrium without ultrasound. The ultrasonic field was applied on the system for 180 min to shift to the new adsorption equilibrium. Determination of the adsorption kinetic under ultrasonic field was conducted with
2.2. Analysis of p-CP A diode array spectrophotometer (Hewlett Packard 8453) was used for UV absorbance spectroscopy measurement. The wavelength resolution and the bandwidth were respectively 1 and 0.5 nm. The length of the optical path in suprasil quartz cell was 1 cm.
Fig. 1. Scheme of the experimental set-up for adsorption–desorption under ultrasonic field.
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
111
0.09 g of GAC, added to 90 ml of aqueous solution at 21 °C (initial water contains 100 mg l1 of p-CP). 2.5. Desorption test For adsorption experiments, GAC (1 g) was added to 1 l of 100 mg l1 of p-CP solution at 21 °C in sealed conical flask. The flask was stirred (400 rpm) for 4 days. The carbon was then separated from the solution. The supernatant was analysed by absorbance measurement. The activated carbon loaded with p-CP was dried under air. After adsorption experiments, about 0.6 g of GAC was tested for regeneration by adding 90 ml UHQ water and placing the sample in the ultrasonic reactor. Ethyl alcohol or sodium hydroxide solution was added to the aqueous solution to determine their effects on the desorption under ultrasonic field. Identical experiments were repeated in the absence of ultrasound using a magnetic stirrer with a stirring rate of 400 rpm.
3. Results and discussion 3.1. Degradation of p-chlorophenol by ultrasound Hoffman et al. have studied decomposition of p-CP by ultrasound at 20 kHz [10,11]. Yields were low in comparison with decomposition rates observed in the high frequency range. In our conditions (21 kHz, 15.2–38.3 W, 1 h irradiation) p-CP degradation was not noticeable. 3.2. Toughness of the activated carbon to ultrasound In order to propose a reasonable process alternative to existing regeneration processes, it is essential that no attrition of the activated carbon occurs when treated with ultrasound. To determine the stability, the activated carbon was exposed to ultrasound at two different intensities for 7 h in the same set-up than the adsorption–desorption experiments. GAC particle size distributions showed same characteristics with and without the application of ultrasound. The mean granulometry has only shifted of 3% for 38.3 W ultrasonic power evidencing stiffness of the activated carbon towards ultrasonic application. 3.3. Effect of ultrasound on the adsorption isotherms The comparison between the isotherms of p-CP on the activated carbon obtained separately in the absence of ultrasound and in the presence of ultrasound with different calorimetric powers at a frequency of 21 kHz is shown on Fig. 2. The plateau of GAC saturation ob-
Fig. 2. Adsorption isotherms of p-chlorophenol on activated carbon at 21 °C in silent conditions and with 21 kHz ultrasound at various calorimetric powers.
tained in the presence of ultrasound is lower than the one obtained in the absence of ultrasound. In both cases, the isotherms exhibit a Langmuir shape. Ultrasound does not appear to modify adsorption process but shifts the equilibrium towards lower sorption concentrations. It indicated that after the ultrasound was exerted into the reactor for 180 min, it caused some parts of the p-CP adsorbed on the carbon to be desorbed and enter into the fluid phase. Finally, it made the concentration of p-CP rise in the fluid phase and hence made the adsorption system reach a new equilibrium state. We explain this phenomenon by a superimposed dynamic process on microscopic level in addition to the molecular dynamic process of adsorption. Cavitation bubbles act like energy transformers by growing over several sound cycles until they reach a critical size. Then they collapse during a fraction of a cycle and a lot of energy is set free. When the bubble is collapsing near the solid surface, which is several orders of magnitude larger than the cavitating bubble, symmetric cavitation is hindered and collapse occurs asymmetrically. As the bubble collapsed, localized areas of high temperatures and pressures are generated in the fluid. The former would make the temperature of the system increase slightly and the latter would have microjets of solvent to be formed. In addition, shock waves are also produced as the bubbles collapsed, which has the potential of creating microscopic turbulence within interfacial film surrounding nearby solid particles. As a result, the acoustic cavitation could produce not only high-speed microjets but also high-pressure shock waves that impinged incessantly on the surface. This action leads to enhancing the breaking of bonds between the adsorbate and the adsorbent surface, and causes more molecules of p-CP adsorbed on the adsorbent to get into the liquid phase. The stronger the acoustic power delivered to the adsorption system, the lower is the corresponding isotherms, which means that the equilibrium amount
112
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
adsorbed on the activated carbon decreases (Fig. 2). It indicates that the isotherms of p-CP adsorption on GAC are dependent on the intensity of ultrasonic field. Indeed, the number of cavitation events, the intensity of the high-speed microjets and high-pressure shock waves produced by acoustic cavitation are mostly dependent on the power delivered to the system. 3.4. Effect of ultrasound on the adsorption kinetics The adsorption data for the uptake of p-chlorophenol versus contact time in silent conditions as well as with different acoustic power are presented in Fig. 3. The comparison of the kinetic curves shows that the rate of p-CP adsorption is much higher under ultrasound than is obtained by simple stirring. In the presence of ultrasound, differences between adsorption rates seem less marked at the beginning of adsorption; the observed variation increases with the time. The amount of p-CP adsorbed onto the GAC decreases with increasing intensity, probably because more cavitation events occur and more molecules are desorbed. The adsorption kinetics can be described by the Adam–Thomas relation: dq ¼ K1 Cðqm qÞ K2 q dt
ð1Þ
where q is the adsorption capacity (mg g1 ), C is the solution concentration (mg l1 ), K1 is the adsorption kinetic constant (l mg1 s1 ), K2 is the desorption kinetic constant (s1 ), qm is the maximal surface concentration (mg g1 ) and t the time (s). At the initial stage of the adsorption reaction, when t ! 0, then q ! 0 and C ! C0 . Then, Eq. (1) could be rewritten: dq V dðC C0 Þ ¼ ¼ K1 C0 qm ð2Þ dt t!0 m dt t!0
where C0 is the initial concentration (mg l1 ), V the volume of solution (l) and m the weight of activated carbon (g). It is then possible to calculate the initial adsorption kinetic coefficient c: V dC ð3Þ c ¼ K1 qm ¼ C0 m dt t!0 The values of c, calculated taking into account the initial slopes of the kinetic curves, are given in Table 1. The adsorption velocity increases in the presence of ultrasonic field. Initial adsorption rates in the presence of ultrasound are 5–7 times higher than without ultrasound, due to reduced mass transfer resistances. The initial adsorption coefficient decreases with increasing the acoustic power of ultrasound. In general, adsorption may be described as a series of steps: mass transfer from the fluid to the particle surface across the boundary layer thickness, diffusion within the porous particle and adsorption itself onto the surface. It has been previously observed [12] that ultrasound waves, and the associated microdisturbances of cavitation bubbles near the surface of solid, reduce the mass transfer boundary layer and therefore give rise to an efficient increase of the mass transfer. If the process is controlled by external resistance, the plot ln C versus time must be linear [13]. In the presence of ultrasound, the relation is not linear and proves that the pore diffusion is the transport limiting step. In the absence of ultrasound, the plot of adsorbate uptake versus the square root of time can be represented by such a linear relationship but it does not pass through the origin. This indicates that intraparticle diffusion was involved in the sorption process but this was not the only rate-limiting mechanism: some other mechanisms were involved [14]. The intraparticle diffusion coefficient can be determined using Weber and Morris model [15,16]: C ¼ C0 KW t1=2
ð4Þ
or q¼
KW 1=2 t w
ð5Þ
with Table 1 Intraparticle diffusion coefficient and initial adsorption rates with different calorimetric powers of 21 kHz ultrasound
Fig. 3. Dynamics of p-chlorophenol uptake by activated carbon in the absence and in the presence of ultrasound at 21 °C.
Calorimetric power (W)
R2
KW 103 (mg l1 min1=2 )
c 105 (l mg1 min1 )
No ultrasoundstirring 15.2 21.5 31.1 38.3
0.972
1771.8
0.197
0.972 0.968 0.999 0.993
8955.3 8318.9 8216.8 7862.7
1.324 1.050 0.966 0.906
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
w¼
C0 C q
113
3.6. Effect of temperature of the regenerating solution ð6Þ
where C0 is the initial concentration (mg l1 ), C is the concentration at any time t (mg l1 ), t is the time (min), q is the amount adsorbed at any time t (mg g1 ), w is the weight of adsorbent per volume of reactor (g l1 ), KW is the Weber intraparticle diffusion coefficient (mg l1 min1=2 ). To determine the intraparticle diffusion coefficient, (wq) is fitted against t1=2 . The values of KW determined from the slope of plots are tabulated in Table 1. The values of the intraparticle diffusion coefficient obtained in the presence of ultrasound are greater than that obtained in the absence of ultrasound, i.e. ultrasound enhances the mass transport in the pores. We attribute this behavior to the induced turbulence and additional convective mass transport inside the pores caused by microjets. The intraparticle diffusion coefficient decreases slightly with increasing ultrasound intensities.
The results of the desorption of p-CP in the presence of ultrasound at 21 and 63 °C with calorimetric powers of 21.5 and 38.3 W are shown in Fig. 5. It can be seen that the rates of desorption at 63 °C are much higher than the corresponding rates observed at lower temperature (21 °C). It is interesting to note that the enhancement in desorption rate at 63 °C due to the ultrasound is appreciably greater at high acoustic power. In general, cavitating bubbles are more easily produced at high temperature because of the decrease of the liquid tensile stress and viscosity. 3.7. Effect of addition of alcohol or NaOH on the desorption
The concentration of p-CP in the UHQ water regenerating the activated carbon was monitored with time, both in the presence of ultrasound and in silent conditions with stirring. The results of the desorption experiments are shown in Fig. 4. The rate of p-CP desorption is significantly increased in the presence of ultrasound. With increasing ultrasound intensities the amount of p-CP desorbed increases, because with high intensities more cavitation events occur and more molecules are desorbed. Desorption, which is an endothermic process, is promoted if such a bubble collapse occurs in the vicinity of the adsorbent surface wherefore adsorbed molecules at this spot go into solution. Thus, it was concluded that high intensity of ultrasound leads to the breaking of bonds formed between p-CP and the adsorbent surface.
Ethyl alcohol or NaOH (1 M) was added into the desorption system (UHQ water þ GAC) in the absence and presence of the ultrasonic field. The dosage was 5% and 20% (volume ratio fluid) for NaOH and ethanol, respectively. Experimental results show that the addition of ethanol or NaOH would obviously cause the enhancement of the amount of p-CP desorbed, especially in the presence of ultrasound, as shown in Figs. 6 and 7. Ethanol, a surfactant substance, can reduce the tensile stress of the liquid and thus reduce the cavitation threshold and facilitate the generation of bubbles. The generation of more transient cavitation bubbles helps to produce easily the high-speed microjets and high-pressure shock waves of solvent as they collapse. Ultrasound and ethanol produce a synergistic effect: the desorption of p-CP from the activated carbon is greater than the sum of the two separate processes. The addition of NaOH to the desorption system increases the pH value, and thus the fraction of phenolate ion. Therefore, p-CP, a weak acid (pKa ¼ 9:20), will be desorbed to a greater extent due to the repulsive forces prevailing at high pH values. The final result is that the
Fig. 4. Amount of p-chlorophenol desorbed from activated carbon in UHQ water versus time at 21 °C in stirring and ultrasound conditions.
Fig. 5. Effect of temperature on the desorption of p-chlorophenol from activated carbon in the presence of ultrasound.
3.5. Regeneration of activated carbon
114
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
microdisturbances near the surface of solid, modify the boundary layer and increase the intraparticle diffusion. The ultrasonic effects increase with increasing acoustic intensity. The rate of desorption increases with the temperature (between 21 and 63 °C). The addition of ethanol or NaOH to the system obviously increases the desorbed amount of p-chlorophenol. Contrary to the NaOH addition, ultrasound and ethanol produce a synergetic effect that can be explained by the lowering of the cavitation threshold.
Fig. 6. Effect of addition of 20% (v/v) ethanol on the desorption of pchlorophenol in the presence and absence of ultrasound at 21 °C.
Fig. 7. Effect of addition of NaOH (1 M––5% v/v) on the desorption of pchlorophenol in the presence and in the absence of ultrasound at 21 °C.
desorbed amounts of p-CP increase with NaOH addition (Fig. 7). In this case, the desorbed p-CP amount under ultrasound with alkaline condition (0.76 mg g1 ) is exactly the sum of desorbed p-CP quantity under ultrasound (0.58 mg g1 ) and the desorbed p-CP quantity at alkaline pH under stirring (0.18 mg g1 ).
4. Conclusions The effects of 21 kHz ultrasound on the adsorption– desorption of p-chlorophenol on GAC were studied. Adsorption of p-chlorophenol on GAC in the presence of ultrasound is lower than in the absence of ultrasound. The stronger the power intensity of the ultrasonic field, the smaller is the adsorption capacity of p-chlorophenol on the activated carbon. However, low frequency ultrasound enhances pore diffusion and initial rate of adsorption. Ultrasound promotes the desorption of adsorbed molecules. The cavitation bubbles, and the associated
References [1] J.L. Sotelo, G. Ovejero, J.A. Delgado, I. Martinez, Comparison of adsorption equilibrium and kinetics of four chlorinated organics from water onto GAC, Water Res. 36 (2002) 599–608. [2] M.W. Jung, K.H. Ahn, Y. Lee, K.P. Kim, J.S. Rhee, J.T. Park, K.J. Paeng, Adsorption characteristics of phenol and chlorophenols on granular activated carbons (GAC), Microchem. J. 70 (2001) 123–131. [3] K. Laszlo, A. Szucs, Surface characterization of polyethleneterephthalate (PET) based activated carbon and the effect of pH on its adsorption capacity from aqueous phenol and 2,3,4tricholorophenol solutions, Carbon 39 (2001) 1945–1953. [4] M.L. Zhou, These de doctorat, Universite de Rennes I, France, 1992. [5] W. Qin, Y. Yuan, Y.Y. Dai, Study on the spot energy effect of ultrasound––influence of ultrasound on desorption equilibrium for hydrogen association system, J. Tsinghua Univ. (Sci. Technol.) 38 (1998) 84. [6] S.U. Rege, R.T. Yang, C.A. Cain, Desorption by ultrasound: phenol on activated carbon and polymeric resin, AIChE J. 44 (1998) 1519–1528. [7] M. Breitbach, D. Bathen, H. Schmidt-Traub, Desorption of a fixed-bed adsorber by ultrasound, Ultrasonics 40 (2002) 679–682. [8] Z. Li, X. Li, H. Xi, B. Hua, Effects of ultrasound on adsorption equilibrium of phenol on polymeric adsorption resin, Chem. Eng. J. 86 (2002) 375–379. [9] T.J. Mason, J.P. Lorimer, D.M. Bates, Quantifying sonochemistry: casting some light on a black art, Ultrasonics 30 (1992) 40. [10] M.R. Hoffman, I. Hua, R. H€ ochemer, Applications of ultrasound irradiation for the degradation of chemical contaminants in water, Ultrason. Sonochem. 3 (1996) 163. [11] C. Petrier, M.F. Lamy, A. Francony, A. Benhacene, B. David, V. Renaudin, N. Gondrexon, Sonochemical degradation of phenol in dilute aqueous solutions: comparison of the reaction rates at 20 and 487 kHz, J. Phys. Chem. 98 (1994) 10514. [12] R. Penn, E. Yeager, F. Hovorka, Effect of ultrasonic waves on concentration gradients, J. Acoust. Soc. Am. 10 (1959) 1372–1376. [13] D.M. Nevskaia, A. Santianes, V. Munoz, A. Guerrero-Ruiz, Interaction of aqueous solutions of phenol with commercial activated carbons: an adsorption and kinetic study, Carbon 37 (1999) 1065–1074. [14] F.A. Banat, B. Al-Bashir, S. Al-Asheh, O. Hayajneh, Adsorption of phenol by bentonite, Environ. Pollut. 107 (2000) 391–398. [15] W.J. Weber, J.C. Morris, Kinetics of adsorption on carbon from solution, J. San. Eng. Div. 89 (1963) 31–59. [16] G. McKay, M.J. Bino, A. Altememi, External mass transfer during the adsorption of various pollutants onto activated carbon, Water Res. 20 (1986) 435–442.
Effects of ultrasound on adsorption–desorption of p-chlorophenol on granular activated carbon O. Hamdaoui a, E. Naffrechoux
b,*
, L. Tifouti a, C. Petrier
b
a
b
Department of Process Engineering, Faculty of Engineering, University of Annaba, P.O. Box 12, 23000 Annaba, Algeria Laboratoire de Chimie Mol eculaire et Environnement, ESIGEC, Universit e de Savoie, BP 1104, 73376 Le Bourget du Lac Cedex, France Received 17 June 2002; accepted 23 September 2002
Abstract The aim of this work is the evaluation of the effects of ultrasound on p-chlorophenol adsorption–desorption on granular activated carbon. Adsorption equilibrium experiments and batch kinetics studies were carried out in the presence and the absence of ultrasound at 21 kHz. Results indicate that the adsorption of p-chlorophenol determined in the presence of ultrasound is lower than the adsorption observed in the absence of ultrasound. Desorption of p-chlorophenol from activated carbon with and without the application of ultrasound was studied. The desorption rates were favoured by increased ultrasound intensity. This rise is more noticeable as temperature increases. The addition of ethanol or NaOH to the system causes an enhancement of the amount of p-chlorophenol desorbed, especially in the presence of ultrasound. A synergetic enhancement of the desorption rate was observed when ultrasonic irradiation was coupled with ethanol chemical regeneration. Ó 2002 Elsevier Science B.V. All rights reserved. Keywords: Ultrasound; Adsorption; Desorption; Activated carbon; p-chlorophenol
1. Introduction Granular activated carbon (GAC) is used in many processes for the treatment of drinking water and industrial wastewaters. It provides a convenient technology for removing a broad range of organic pollutants, which are generally of concern because of their toxicity to human health. Phenols are regarded as priority pollutants because of their toxicity in environment at low concentrations. Many of them have been classified as hazardous pollutants because of their potential harm to human health. There is a considerable amount of data in the literature concerning adsorption of phenols onto GAC [1–4]. Although the adsorption of phenolic compounds on GAC is easy to manage, the process of regenerating the adsorbent poses a major challenge because of the high affinity of the phenolic compounds to the sorbent sur* Corresponding author. Tel.: +33-479-758830; fax: +33-479758674. E-mail address: emmanuel.naff[email protected] (E. Naffrechoux).
face. The most common techniques for desorption of phenols from activated carbon are thermal regeneration, chemical regeneration, and to a smaller extent, bioregeneration and regeneration under supercritical condition. As these techniques are presenting drawbacks, it is of interest to explore other methods of desorption. Currently, one of the regeneration way which is considered is the desorption by ultrasound. Qin et al. investigated the effect of ultrasound on the desorption of phenol from NKA II resin, and pointed out that ultrasound had ‘‘spot energy effects’’ which would enhance the process [5]. Ultrasound (1440 kHz) was used by Rege et al. [6] to realize the desorption of phenol from activated carbon and polymeric resin adsorbents. It was found that the desorption rates were helped by temperature decrease, air saturation of the liquid medium and increase of the ultrasonic intensity. The rate enhancement was attributed to acoustic vortex microstreaming. Breitbach et al. [7] found that ultrasound promotes the desorption of benzoic acid from polymeric resin. An important factor influencing ultrasound enhanced desorption was the temperature rise due to ultrasonication.
1350-4177/02/$ - see front matter Ó 2002 Elsevier Science B.V. All rights reserved. PII: S 1 3 5 0 - 4 1 7 7 ( 0 2 ) 0 0 1 3 7 - 2
110
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
Another ultrasonic effect is supposed to occur because regeneration is more effective at 610 kHz than at 204 kHz for a same operating temperature. Using NKA II resin, Li et al. [8] found that the adsorption of phenol determined in the presence of ultrasound (39 kHz, ultrasonic bath) is less than that in the absence of ultrasound. It was also observed that the stronger the acoustic power delivered to the system, the lower the adsorption capacity of phenol on the resin was. Chlorophenols are produced during water chlorination and can only be removed from water during the last treatment stage by adsorption. It seems of importance to test the alternative acoustic regeneration of GAC when the adsorbant is saturated with the micropollutant. Object of this work was to compare the adsorption of p-chlorophenol on GAC in the absence and in the presence of 21 kHz ultrasound at ambient temperature. In a second part, the ultrasound effect on the desorption of p-chlorophenol from GAC is supplied to chemical regeneration method in order to evaluate the efficiency of the combined process.
2. Materials and methods 2.1. Reagent and activated carbon p-chlorophenol (p-CP) with purity greater than 99.5% (Sigma) and ultra-high quality water (Elga elgastat) were used to prepare the aqueous solutions for the different tests. The GAC (Prolabo) has a BET surface area of 929 m2 g1 and a mean granulometry of 3 mm. The carbon was pretreated by boiling in ultra-high quality (UHQ) water for 1 h and washed thoroughly with UHQ water until the UV absorbance intensity was equal to zero. Finally, the washed GAC was dried in an oven at 110 °C to constant weight before use. The particle size analysis of GAC was realised before and after treatment. A suspension of GAC (0.6 g) in pure water (90 ml) was exposed to ultrasound (15.2 and 38.3 W) or stirred at 400 rpm during 7 h. The suspension was then filtered (0.45 lm) and dried at 110 °C for 24 h. The particle size distribution was performed with a laser granulometer (Mastersizer 2000––Malvern Instruments). The size measurement range and detection limit were 0.02–5000 lm and 0.01 lm respectively.
2.3. Apparatus Batch adsorption–desorption experiments were carried out using 100 ml (fluid volume) reactor as shown in Fig. 1. It consists of a double cylindrical jacket, allowing water cooling of the reactor. This reactor is made of a glass cylinder of 7.9 cm height and internal diameter 4.1 cm. The top of the glass reactor received the ultrasonic transducer. The 21 kHz ultrasonic wave was emitted from a titanium horn (diameter 3.5 cm) connected to a commercial supply Sinaptec Nexus II. Determination of the acoustical energy absorbed in the reactor was achieved following the calorimetric method [9]. The volume treated was 90 ml for a liquid height of 6.8 cm. 2.4. Adsorption procedure Adsorption isotherm under normal condition was carried out in batch mode with the following procedure. Various amounts of GAC (0.05–1 g l1 ) were continuously stirred at 21 °C with 90 ml of p-CP aqueous solution (100 mg l1 ). The stirring rate was 400 rpm. Equilibrium was reached after a contact time of 4 days. Samples of 2 ml were withdrawn for analysis using UV– visible spectrophotometer to determine the concentration of p-CP remaining in the fluid phase, and the amounts of p-CP adsorbed on the GAC were inferred from the mass balance. For the kinetic studies 1 g of activated carbon was continuously stirred at 21 °C with 1 l of aqueous p-CP solution (100 mg l1 ). The analyses of the supernatant solution were achieved as previously described. In order to determine the isotherm of p-CP on the GAC under ultrasonic field, the system was conducted to equilibrium without ultrasound. The ultrasonic field was applied on the system for 180 min to shift to the new adsorption equilibrium. Determination of the adsorption kinetic under ultrasonic field was conducted with
2.2. Analysis of p-CP A diode array spectrophotometer (Hewlett Packard 8453) was used for UV absorbance spectroscopy measurement. The wavelength resolution and the bandwidth were respectively 1 and 0.5 nm. The length of the optical path in suprasil quartz cell was 1 cm.
Fig. 1. Scheme of the experimental set-up for adsorption–desorption under ultrasonic field.
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
111
0.09 g of GAC, added to 90 ml of aqueous solution at 21 °C (initial water contains 100 mg l1 of p-CP). 2.5. Desorption test For adsorption experiments, GAC (1 g) was added to 1 l of 100 mg l1 of p-CP solution at 21 °C in sealed conical flask. The flask was stirred (400 rpm) for 4 days. The carbon was then separated from the solution. The supernatant was analysed by absorbance measurement. The activated carbon loaded with p-CP was dried under air. After adsorption experiments, about 0.6 g of GAC was tested for regeneration by adding 90 ml UHQ water and placing the sample in the ultrasonic reactor. Ethyl alcohol or sodium hydroxide solution was added to the aqueous solution to determine their effects on the desorption under ultrasonic field. Identical experiments were repeated in the absence of ultrasound using a magnetic stirrer with a stirring rate of 400 rpm.
3. Results and discussion 3.1. Degradation of p-chlorophenol by ultrasound Hoffman et al. have studied decomposition of p-CP by ultrasound at 20 kHz [10,11]. Yields were low in comparison with decomposition rates observed in the high frequency range. In our conditions (21 kHz, 15.2–38.3 W, 1 h irradiation) p-CP degradation was not noticeable. 3.2. Toughness of the activated carbon to ultrasound In order to propose a reasonable process alternative to existing regeneration processes, it is essential that no attrition of the activated carbon occurs when treated with ultrasound. To determine the stability, the activated carbon was exposed to ultrasound at two different intensities for 7 h in the same set-up than the adsorption–desorption experiments. GAC particle size distributions showed same characteristics with and without the application of ultrasound. The mean granulometry has only shifted of 3% for 38.3 W ultrasonic power evidencing stiffness of the activated carbon towards ultrasonic application. 3.3. Effect of ultrasound on the adsorption isotherms The comparison between the isotherms of p-CP on the activated carbon obtained separately in the absence of ultrasound and in the presence of ultrasound with different calorimetric powers at a frequency of 21 kHz is shown on Fig. 2. The plateau of GAC saturation ob-
Fig. 2. Adsorption isotherms of p-chlorophenol on activated carbon at 21 °C in silent conditions and with 21 kHz ultrasound at various calorimetric powers.
tained in the presence of ultrasound is lower than the one obtained in the absence of ultrasound. In both cases, the isotherms exhibit a Langmuir shape. Ultrasound does not appear to modify adsorption process but shifts the equilibrium towards lower sorption concentrations. It indicated that after the ultrasound was exerted into the reactor for 180 min, it caused some parts of the p-CP adsorbed on the carbon to be desorbed and enter into the fluid phase. Finally, it made the concentration of p-CP rise in the fluid phase and hence made the adsorption system reach a new equilibrium state. We explain this phenomenon by a superimposed dynamic process on microscopic level in addition to the molecular dynamic process of adsorption. Cavitation bubbles act like energy transformers by growing over several sound cycles until they reach a critical size. Then they collapse during a fraction of a cycle and a lot of energy is set free. When the bubble is collapsing near the solid surface, which is several orders of magnitude larger than the cavitating bubble, symmetric cavitation is hindered and collapse occurs asymmetrically. As the bubble collapsed, localized areas of high temperatures and pressures are generated in the fluid. The former would make the temperature of the system increase slightly and the latter would have microjets of solvent to be formed. In addition, shock waves are also produced as the bubbles collapsed, which has the potential of creating microscopic turbulence within interfacial film surrounding nearby solid particles. As a result, the acoustic cavitation could produce not only high-speed microjets but also high-pressure shock waves that impinged incessantly on the surface. This action leads to enhancing the breaking of bonds between the adsorbate and the adsorbent surface, and causes more molecules of p-CP adsorbed on the adsorbent to get into the liquid phase. The stronger the acoustic power delivered to the adsorption system, the lower is the corresponding isotherms, which means that the equilibrium amount
112
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
adsorbed on the activated carbon decreases (Fig. 2). It indicates that the isotherms of p-CP adsorption on GAC are dependent on the intensity of ultrasonic field. Indeed, the number of cavitation events, the intensity of the high-speed microjets and high-pressure shock waves produced by acoustic cavitation are mostly dependent on the power delivered to the system. 3.4. Effect of ultrasound on the adsorption kinetics The adsorption data for the uptake of p-chlorophenol versus contact time in silent conditions as well as with different acoustic power are presented in Fig. 3. The comparison of the kinetic curves shows that the rate of p-CP adsorption is much higher under ultrasound than is obtained by simple stirring. In the presence of ultrasound, differences between adsorption rates seem less marked at the beginning of adsorption; the observed variation increases with the time. The amount of p-CP adsorbed onto the GAC decreases with increasing intensity, probably because more cavitation events occur and more molecules are desorbed. The adsorption kinetics can be described by the Adam–Thomas relation: dq ¼ K1 Cðqm qÞ K2 q dt
ð1Þ
where q is the adsorption capacity (mg g1 ), C is the solution concentration (mg l1 ), K1 is the adsorption kinetic constant (l mg1 s1 ), K2 is the desorption kinetic constant (s1 ), qm is the maximal surface concentration (mg g1 ) and t the time (s). At the initial stage of the adsorption reaction, when t ! 0, then q ! 0 and C ! C0 . Then, Eq. (1) could be rewritten: dq V dðC C0 Þ ¼ ¼ K1 C0 qm ð2Þ dt t!0 m dt t!0
where C0 is the initial concentration (mg l1 ), V the volume of solution (l) and m the weight of activated carbon (g). It is then possible to calculate the initial adsorption kinetic coefficient c: V dC ð3Þ c ¼ K1 qm ¼ C0 m dt t!0 The values of c, calculated taking into account the initial slopes of the kinetic curves, are given in Table 1. The adsorption velocity increases in the presence of ultrasonic field. Initial adsorption rates in the presence of ultrasound are 5–7 times higher than without ultrasound, due to reduced mass transfer resistances. The initial adsorption coefficient decreases with increasing the acoustic power of ultrasound. In general, adsorption may be described as a series of steps: mass transfer from the fluid to the particle surface across the boundary layer thickness, diffusion within the porous particle and adsorption itself onto the surface. It has been previously observed [12] that ultrasound waves, and the associated microdisturbances of cavitation bubbles near the surface of solid, reduce the mass transfer boundary layer and therefore give rise to an efficient increase of the mass transfer. If the process is controlled by external resistance, the plot ln C versus time must be linear [13]. In the presence of ultrasound, the relation is not linear and proves that the pore diffusion is the transport limiting step. In the absence of ultrasound, the plot of adsorbate uptake versus the square root of time can be represented by such a linear relationship but it does not pass through the origin. This indicates that intraparticle diffusion was involved in the sorption process but this was not the only rate-limiting mechanism: some other mechanisms were involved [14]. The intraparticle diffusion coefficient can be determined using Weber and Morris model [15,16]: C ¼ C0 KW t1=2
ð4Þ
or q¼
KW 1=2 t w
ð5Þ
with Table 1 Intraparticle diffusion coefficient and initial adsorption rates with different calorimetric powers of 21 kHz ultrasound
Fig. 3. Dynamics of p-chlorophenol uptake by activated carbon in the absence and in the presence of ultrasound at 21 °C.
Calorimetric power (W)
R2
KW 103 (mg l1 min1=2 )
c 105 (l mg1 min1 )
No ultrasoundstirring 15.2 21.5 31.1 38.3
0.972
1771.8
0.197
0.972 0.968 0.999 0.993
8955.3 8318.9 8216.8 7862.7
1.324 1.050 0.966 0.906
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
w¼
C0 C q
113
3.6. Effect of temperature of the regenerating solution ð6Þ
where C0 is the initial concentration (mg l1 ), C is the concentration at any time t (mg l1 ), t is the time (min), q is the amount adsorbed at any time t (mg g1 ), w is the weight of adsorbent per volume of reactor (g l1 ), KW is the Weber intraparticle diffusion coefficient (mg l1 min1=2 ). To determine the intraparticle diffusion coefficient, (wq) is fitted against t1=2 . The values of KW determined from the slope of plots are tabulated in Table 1. The values of the intraparticle diffusion coefficient obtained in the presence of ultrasound are greater than that obtained in the absence of ultrasound, i.e. ultrasound enhances the mass transport in the pores. We attribute this behavior to the induced turbulence and additional convective mass transport inside the pores caused by microjets. The intraparticle diffusion coefficient decreases slightly with increasing ultrasound intensities.
The results of the desorption of p-CP in the presence of ultrasound at 21 and 63 °C with calorimetric powers of 21.5 and 38.3 W are shown in Fig. 5. It can be seen that the rates of desorption at 63 °C are much higher than the corresponding rates observed at lower temperature (21 °C). It is interesting to note that the enhancement in desorption rate at 63 °C due to the ultrasound is appreciably greater at high acoustic power. In general, cavitating bubbles are more easily produced at high temperature because of the decrease of the liquid tensile stress and viscosity. 3.7. Effect of addition of alcohol or NaOH on the desorption
The concentration of p-CP in the UHQ water regenerating the activated carbon was monitored with time, both in the presence of ultrasound and in silent conditions with stirring. The results of the desorption experiments are shown in Fig. 4. The rate of p-CP desorption is significantly increased in the presence of ultrasound. With increasing ultrasound intensities the amount of p-CP desorbed increases, because with high intensities more cavitation events occur and more molecules are desorbed. Desorption, which is an endothermic process, is promoted if such a bubble collapse occurs in the vicinity of the adsorbent surface wherefore adsorbed molecules at this spot go into solution. Thus, it was concluded that high intensity of ultrasound leads to the breaking of bonds formed between p-CP and the adsorbent surface.
Ethyl alcohol or NaOH (1 M) was added into the desorption system (UHQ water þ GAC) in the absence and presence of the ultrasonic field. The dosage was 5% and 20% (volume ratio fluid) for NaOH and ethanol, respectively. Experimental results show that the addition of ethanol or NaOH would obviously cause the enhancement of the amount of p-CP desorbed, especially in the presence of ultrasound, as shown in Figs. 6 and 7. Ethanol, a surfactant substance, can reduce the tensile stress of the liquid and thus reduce the cavitation threshold and facilitate the generation of bubbles. The generation of more transient cavitation bubbles helps to produce easily the high-speed microjets and high-pressure shock waves of solvent as they collapse. Ultrasound and ethanol produce a synergistic effect: the desorption of p-CP from the activated carbon is greater than the sum of the two separate processes. The addition of NaOH to the desorption system increases the pH value, and thus the fraction of phenolate ion. Therefore, p-CP, a weak acid (pKa ¼ 9:20), will be desorbed to a greater extent due to the repulsive forces prevailing at high pH values. The final result is that the
Fig. 4. Amount of p-chlorophenol desorbed from activated carbon in UHQ water versus time at 21 °C in stirring and ultrasound conditions.
Fig. 5. Effect of temperature on the desorption of p-chlorophenol from activated carbon in the presence of ultrasound.
3.5. Regeneration of activated carbon
114
O. Hamdaoui et al. / Ultrasonics Sonochemistry 10 (2003) 109–114
microdisturbances near the surface of solid, modify the boundary layer and increase the intraparticle diffusion. The ultrasonic effects increase with increasing acoustic intensity. The rate of desorption increases with the temperature (between 21 and 63 °C). The addition of ethanol or NaOH to the system obviously increases the desorbed amount of p-chlorophenol. Contrary to the NaOH addition, ultrasound and ethanol produce a synergetic effect that can be explained by the lowering of the cavitation threshold.
Fig. 6. Effect of addition of 20% (v/v) ethanol on the desorption of pchlorophenol in the presence and absence of ultrasound at 21 °C.
Fig. 7. Effect of addition of NaOH (1 M––5% v/v) on the desorption of pchlorophenol in the presence and in the absence of ultrasound at 21 °C.
desorbed amounts of p-CP increase with NaOH addition (Fig. 7). In this case, the desorbed p-CP amount under ultrasound with alkaline condition (0.76 mg g1 ) is exactly the sum of desorbed p-CP quantity under ultrasound (0.58 mg g1 ) and the desorbed p-CP quantity at alkaline pH under stirring (0.18 mg g1 ).
4. Conclusions The effects of 21 kHz ultrasound on the adsorption– desorption of p-chlorophenol on GAC were studied. Adsorption of p-chlorophenol on GAC in the presence of ultrasound is lower than in the absence of ultrasound. The stronger the power intensity of the ultrasonic field, the smaller is the adsorption capacity of p-chlorophenol on the activated carbon. However, low frequency ultrasound enhances pore diffusion and initial rate of adsorption. Ultrasound promotes the desorption of adsorbed molecules. The cavitation bubbles, and the associated
References [1] J.L. Sotelo, G. Ovejero, J.A. Delgado, I. Martinez, Comparison of adsorption equilibrium and kinetics of four chlorinated organics from water onto GAC, Water Res. 36 (2002) 599–608. [2] M.W. Jung, K.H. Ahn, Y. Lee, K.P. Kim, J.S. Rhee, J.T. Park, K.J. Paeng, Adsorption characteristics of phenol and chlorophenols on granular activated carbons (GAC), Microchem. J. 70 (2001) 123–131. [3] K. Laszlo, A. Szucs, Surface characterization of polyethleneterephthalate (PET) based activated carbon and the effect of pH on its adsorption capacity from aqueous phenol and 2,3,4tricholorophenol solutions, Carbon 39 (2001) 1945–1953. [4] M.L. Zhou, These de doctorat, Universite de Rennes I, France, 1992. [5] W. Qin, Y. Yuan, Y.Y. Dai, Study on the spot energy effect of ultrasound––influence of ultrasound on desorption equilibrium for hydrogen association system, J. Tsinghua Univ. (Sci. Technol.) 38 (1998) 84. [6] S.U. Rege, R.T. Yang, C.A. Cain, Desorption by ultrasound: phenol on activated carbon and polymeric resin, AIChE J. 44 (1998) 1519–1528. [7] M. Breitbach, D. Bathen, H. Schmidt-Traub, Desorption of a fixed-bed adsorber by ultrasound, Ultrasonics 40 (2002) 679–682. [8] Z. Li, X. Li, H. Xi, B. Hua, Effects of ultrasound on adsorption equilibrium of phenol on polymeric adsorption resin, Chem. Eng. J. 86 (2002) 375–379. [9] T.J. Mason, J.P. Lorimer, D.M. Bates, Quantifying sonochemistry: casting some light on a black art, Ultrasonics 30 (1992) 40. [10] M.R. Hoffman, I. Hua, R. H€ ochemer, Applications of ultrasound irradiation for the degradation of chemical contaminants in water, Ultrason. Sonochem. 3 (1996) 163. [11] C. Petrier, M.F. Lamy, A. Francony, A. Benhacene, B. David, V. Renaudin, N. Gondrexon, Sonochemical degradation of phenol in dilute aqueous solutions: comparison of the reaction rates at 20 and 487 kHz, J. Phys. Chem. 98 (1994) 10514. [12] R. Penn, E. Yeager, F. Hovorka, Effect of ultrasonic waves on concentration gradients, J. Acoust. Soc. Am. 10 (1959) 1372–1376. [13] D.M. Nevskaia, A. Santianes, V. Munoz, A. Guerrero-Ruiz, Interaction of aqueous solutions of phenol with commercial activated carbons: an adsorption and kinetic study, Carbon 37 (1999) 1065–1074. [14] F.A. Banat, B. Al-Bashir, S. Al-Asheh, O. Hayajneh, Adsorption of phenol by bentonite, Environ. Pollut. 107 (2000) 391–398. [15] W.J. Weber, J.C. Morris, Kinetics of adsorption on carbon from solution, J. San. Eng. Div. 89 (1963) 31–59. [16] G. McKay, M.J. Bino, A. Altememi, External mass transfer during the adsorption of various pollutants onto activated carbon, Water Res. 20 (1986) 435–442.