Bubble hydrodynamic influence on oxygen transfer rate at presence of cationic and anionic surfactants in electroflotation process*

KOTTI Mariam, KSENTINI Issam, BEN MANSOUR Lassaad

Laboratory of Applied Fluid Mechanics, Process Engineering and Environment, Science Faculty of Sfax, B.P.1171, 3000 Sfax, Tunisie, E-mail: kotti_mariam@yahoo.fr

(Received February 14, 2013, Revised July 21, 2013)

Bubble hydrodynamic influence on oxygen transfer rate at presence of cationic and anionic surfactants in electroflotation process*

KOTTI Mariam, KSENTINI Issam, BEN MANSOUR Lassaad

Laboratory of Applied Fluid Mechanics, Process Engineering and Environment, Science Faculty of Sfax, B.P.1171, 3000 Sfax, Tunisie, E-mail: kotti_mariam@yahoo.fr

(Received February 14, 2013, Revised July 21, 2013)

In this work, the effects of the presence of surfactants in the liquid phase and the hydrodynamic regime of the bubble flow on the oxygen transfer rate were investigated in an electroflotation process in batch mode. The volumetric mass transfer coefficientKLaand the oxygenation capacity were evaluated to improve the performances of the electroflotation process in terms of oxygenation. In order to evaluate the liquid-side mass transfer coefficientKL, the volumetric mass transfer coefficientKLawas dissociated intoKLand the specific interfacial area (a) since the last one was obtained from the gas hold-up and the bubble diameter. The effect of Reynolds number which define the hydrodynamic of the bubble flow has been also studied. Models ofKLaandKLhave been established to show the effects of the hydrodynamic parameters and liquid phase characteristics on the oxygen transfer rate.

electroflotation, mass transfer coefficient, hydrodynamic parameters, surfactant

Introduction

The effects of surfactants in wastewater on the oxygen transfer rate have been studied because of their industrial relevance. In fact these surfactants are present in the wastewater issued from diverse applications ranging from paint technology, lubrication, paper making, oil recovery and biochemistry of proteins[1].

It is generally recognized that small amounts of surfactant additives as contaminants affect markedly mass transfer rate from the gas to the liquid phase[2-5]. Mass transfer effectiveness is most frequently assessed in gas-liquid contactors by measuring the volumetric mass transfer coefficientKLa[6]. This coefficient is a key parameter in electroflotation process which can be used as aeration system.

Electroflotation (EF) is a simple process that floats pollutants (or other substances) by their adhesion onto tiny bubbles of hydrogen and oxygen generated from electrolysis of aqueous solutions[7-9]. The chemical reactions taking place at the cathode and the anode are given as follows:

Anodic oxidation

Chen[9]has shown that EF is more competitive than other flotation technologies such as dissolved air flotation and dispersed air flotation. In fact, the electroflotation process gives the highest oxygenation efficiency[10].

TheKLavalues are often global and thus insufficient to understand the gas-liquid mass transfer mechanisms[11]. For this purpose, it becomes essential to separate the parameters, especially the liquid-side mass transfer coefficientKLand the interfacial areaa[12-14].

Fig.1 Experimental set-up

1. Theory

The volumetric mass transfer coefficient of dissolved oxygen can be derived from the two-film theory. Assume that the diffusion rate of oxygen through gas film is much higher than the diffusion rate through liquid film, then the resistance of gas film can be neglected. For a complete mixed system, Eq.(3) is obtained[15]

where dC/dtis the rate of change of oxygen concentration with time.

Equation (3) can be readily integrated to yield the following expression forCas a function of time

whereC0is the initial dissolved oxygen concentration att=0 andC*is the equilibrium oxygen concentration in liquid phase.

A nonlinear regression analysis based on the Gauss-Newton method was recommended by American Society of Civil Engineers (ASCE) to fit Eq.(4) to experimental data usingKLa,C*andC0as three adjustable model parameters[16].

The volumetric mass transfer coefficient must be corrected to a standard reference temperature (T) of 20oC by using the Arrhenius relationship[6]

A generally-accepted value of the temperature correction factor,θis 1.024.

The Oxygenation Capacity (OC) presents the mass of oxygen that can be transferred by the aeration system per m3and per hour to evaluate the dissolved oxygen concentration in water.

2. Experimental set-up and mesuring techniques

2.1Electroflotation cell

The electroflotation cell, shown in Fig.1, is used for batch mode. It is a cylindrical plexiglas vessel and is 0.092 m in diameter and 0.71.5 m in height. It is provided with two electrodes: titanium coated with ruthenium oxide anode and a stainless steel cathode. These two electrodes are supplied by a generator of DC current which enables the variation of current density. It is also noticed that the gap between anode and cathode is maintained at 0.005 m to minimize the ohmic loss. The cathode compared to the anode is perforated and occupies the top position .This perforation allows the evacuation of bubbles produced at the anode.

2.2Image acquisition and treatment method

The equipments used for the determination of the bubble size distributions by image analysis are a microscopic zoom digital video camera (model NV-A3E from Panasonic, Japan), an acquisition card (model Pinnacle PCTV PRO version 4.02 from Pinnacle systems), a PC (model Pentium 4, from Fujitsu Siemens) with a digital image analysis programs namely: Photofiltre (Version 6.2.6), Photoshop (CS2), Ulead Photo Impact (Version 11 Pro) and 700 watt power halogen spot.

Table 1 Chemical characteristics of liquid phases at 20oC

A wire of a known diameter (1.49×10–4m) is videotaped for use as the calibration factor for the bubble size. Then, we obtain a video file in which the number of frames per second is set. We extract all frames (photos) from this video. We apply a series of filters which lead to clear bubbles as showed. For getting a sufficiently representative bubble size, 50 bubbles were at least measured in each experimental condition. The confidence level for reproducibility of experiments was 95%[17].

2.2.1 Gas hold-up

Gas hold-up is a dimensionless key parameter for design purposes that characterizes hydrodynamic phenomena of bubble column systems[18]. It is basically defined as the volume fraction of gas phase occupied by the gas bubbles[19].

The gas hold-up is calculable in the following way using image treatment system[6]

where ΔHis the increase in liquid level after gassing andHLis the ungased liquid height.

2.2.2 Bubble rise velocity

Thanks to the image treatment system, the bubble rise velocity can be estimated by measuring the bubble displacement between two shifted images during precise time.

where ΔDis the bubble course in a laps timet. In fact, series of single bubbles were identified and recorded in their ascension. Then, images were treated and superposed in order to calculate the bubble rise velocity[17].

2.2.3 Reynolds number

Historically, the most classical methods used for the characterization of regime transitions in bubble columns have consisted in directly observing the value of Reynolds number (Re). In fact,Rehas been used as a key parameter for flow regime analysis: homogenous regime, transition region and heterogeneous regime. The end of the homogeneous regime is generally defined asRe=1[20].

whereLμis the liquid viscosity andLρis the liquid density.

2.2.4 Volumetric mass transfer coefficient

The volumetric mass transfer coefficientKLawas measured using the unsteady state method with an oxygen probe (Consort C932) placed mid-way in the electroflotation cell. The oxygen concentration was reduced to zero by adding 150 mg/L of sodium sulphite (Na2SO3) and 2 mg/L of cobalt ions.

Experiments were conducted with different model cationic and anionic surfactant solutions (Table 1) at current density ranging from 60 A/m2 to 260 A/m2.

2.2.5 The specific interfacial area

The specific interfacial area is one of the most important parameters for gas-liquid reactor design. Generally, the specific interfacial area depends on the size of the unit, the operating parameters and the physical and chemical properties of the liquid[21].

The gas hold-up and bubble size are measured, allowing the specific interfacial area (a) to be determined using the following equation[21]

2.2.6 The liquid -side mass transfer coefficientKL

The liquid-side mass transfer coefficientKLdepends on the turbulence created in the liquid phase[12]. Measurement of mass transfer coefficientKLaand the specific interfacial area (a) then allows determination of the liquid-side mass transfer coefficientKLwhich is calculated from

Fig.2 Variation of conductivity as a function of anionic (a) and cationic (b) surfactant concentration

3. Liquid phase characterization

Table1 presents the variety of liquid phase characteristics which allows understanding the effect of surfactant solutions on the mass transfer efficiency.

It is well known that surfactants are characterized by the Critical Micelle Concentrations (CMC). The CMC is the concentration where surfactant molecules arrange themselves into organized molecular assemblies known as micelles[22].

The critical micelle concentration of each surfactant in water is determined from measurements of the specific conductivity versus the surfactant concentration by Consort C932 at 20oC[23].

According to Fig.2, the conductivity varies linearly with increasing surfactant concentrationCST, but with two different slopes. The CMC represents the break point of the slope[24]. The micellization process changes the slope of the curves because the charged micelles are less mobile than the monomeric ions[23].

Fig.3 Impact of liquid phase solution on the volumetricass transfer coefficient

4. Results and discussion

The result shown in Fig.3 indicate that, whatever the liquid phases, the volumetric mass transfer coefficientKLaincreases with the increase of the current density (J)[25]. It is also noted that the volumetric mass transfer coefficientsKLain the case of all surfactant solutions are significantly smaller than those in the case of tap water. The trend obtained is summed up as

Following this tendency we can say that the screeneffect[26]due to the anionic surfactant solutions is higher than that for the cationic surfactant solutions: Actually, the anionic surfactant has an additional effect which results in an increase in viscosity (Table 1), depressing the mobility of the oxygen bubbles.

Fig.4 Impact of liquid phase solution on the oxygenation capacity

4.1Effect of surfactants on the oxygenation capacity

The results presented in the Fig.4 shows that the oxygenation capacity increases as a function of current density for different liquid solutions since it is in direct relation withKLa.

It is also noted that the oxygenation capacity of all surfactant solutions are significantly smaller than those in tap water.

Fig.5 Variation of the volumetric mass transfer coefficient as a function of Reynolds number for cationic (a) and anionic (b) surfactant solution

4.2Effect of Reynolds number on KLa

In order to study the effect of hydrodynamic regime of the bubble flow on the volumetric mass transfer coefficient in presence of surfactants,KLawas represented as a function of Reynolds number.

As is shown in Fig.5, an increase in the volumetric mass transfer coefficientKLaaccording to the Reynolds number is observed for water and different cationic and anionic surfactant concentrations. OnceRereaches a value around unity, we notice a significant change in the slope of the curve which corresponds to the transition from laminar regime to turbulent one.

This behavior is confirmed by correlations that expressKLaas a function of Reynolds number and liquid surface tension (σL) using an exponential multiple regressions in different surfactant solutions. In the case of cationic surfactant solution:

Figure 6 presents the comparison between the experimental values and the values predicted by Eqs.(13)-(16). The average difference is about ±12%, which is considered satisfactory. Thus, this correlation confirms thatKLaincreases with the Reynolds number whatever the surfactant concentrations. This effect is similar to that of liquid surface tension before CMC, but besides this one we notice that the surface tension has a negative effect onKLa[27,28].

Fig.6 Comparison of calculated and experimentalKLa

As is shown in Fig.7 whatever the liquid phases, the specific interfacial area increases with the current density. Overall, the following trend is found

Fig.7 Impact of liquid phase solution on the interfacial area

4.3Effect of Reynolds number on KL

According to the double film theory, the variation ofKLis mainly due to the film thickness, as this one is influenced by the hydrodynamics of the liquid phase, and thus it is interesting to study the variation ofKLaccording toRe.

Fig.8 Variation of the liquid-side mass transfer coefficient as a function of Reynolds number for cationic (a) and anionic (b) surfactant solution

As is shown in Fig.8 an increase in the liquidside mass transfer coefficientKLaccording to the Reynolds number is observed for water and different cationic and anionic surfactant concentrations. OnceRereaches a value around unity we notice a significant change in the slope of the curve which corresponds to the transition regime.

According to Fig.8 we can also deduce that the variation of liquid-side mass transfer coefficientKLwith the Reynolds number is similar to that ofKLafor different cationic and anion surfactant concentrations (see Fig.5), this confirms that the effect of the hydrodynamic behavior on the volumetric mass transfer coefficient due to mainly on the influence of this hydrodynamic regime on the liquid- side mass transfer.

According to the literature, Higbie’s theory and Frosslig’s equation are indicated as follows[29]:

In our case the treatment of the results using a multiple regressions allows that theKLvalues related to CST≤CMC are deduced from Higbie’s theory butKLvalue related to CST>CMC are deduced from Frossling’s equation with a quite difference relative to the constant values. In fact, in the case of cationic surfactant (CST) solution

In case of CST solution, the transition from laminar to turbulent flows is obtained with a current density of 150 A/m2 whatever the concentration of surfactant[17]. In the case of Anionic Surfactant Tolution (AST):

In case of AST solution, the transition is significantly moved to higher current density value (>200 A/m2) whatever the concentration of surfactant[17].

Fig.9 Comparison of calculated and experimentalKL

Figure 9 presents the comparison between the experimental values and the values predicted by Eqs.(17)-(20), the average difference is about ±10%. This difference is considered satisfactory. Thus, this correlation can be used to determineKLin the presence of cationic and anionic surfactant.

5. Conclusions

A comparative study between cationic and anionic surfactants as contaminant in aqua solution confirms that:

(1) The values ofKLaand OC for both surfactants solution are significantly smaller than those of the tap water.

(2) The specific interfacial area tends to increase by the addition of cationic surfactant, but it decreases with the anionic surfactant.

(3) The effects of Reynolds number on the volumetric mass transfer coefficient and the liquid-side mass transfer coefficient are similar.

(4) The simple model for estimation ofKLabased on the Reynolds number and the liquid surface tension are established. Good correlations of theKLwith the hydrodynamics parameter are also found.

[1] GHIACI M., KALBASI R. J. and ABBASPOUR A. Adsorption isotherms of non-ionic surfactants on Nabentonite (Iran) and evaluation of thermodynamic parameters[J]. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 2007, 297(1-3): 105-113.

[2] RUZICKA M. C., VECER M. M. and ORVALHO S. et al. Effect of surfactant on homogeneous regime stability in bubble column[J]. Chemical Engineering Science, 2008, 63(4): 951- 967.

[3] MOHAMMED T. J., HANNA F. Z. and HAMAWAND I. B. The effect of surfactants on characteristics of hydrodynamic and mass transfer coefficient in gasliquid dispersion column[J]. Engineering and Technology, 2007, 25: 591-607.

[4] SAJJADI B., MORAVEJI M. K. and DAVARNEJAD R. Investigation of surfactant effect on the operational characteristics of a packed bed internal loop airlift reactor[J]. World Applied Sciences Journal, 2010, 11(8): 1004-1014.

[5] LEE R. J., SAYLOR J. R. The effect of a surfactant monolayer on oxygen transfer across an air/water interface during mixed convection[J]. International Journal Heat and Mass Transfer, 2010, 53(17-18): 3405-3413.

[6] VASCONCELOS J. M. T., RODRIGUES J. M. L. and ORVALHO S. C. P. et al. Effect of contaminants on mass transfer coefficients in bubble column and airlift contactors[J]. Chemical Engineering Science, 2003, 58(8): 1431-1440.

[7] JUNG Y. K., HAN M. Y. Simultaneous removal of cadmium and turbidity in contaminated soil-washing water by electroflotation[J]. Water Science and Technology, 2002, 46(11-12): 225-230.

[8] CASQUEIRA R. G., TOREM M. L. and CUNHA F. O. Removal of heavy toxic metal by electroflotation[J]. Saneamento Ambiental, 2002, 85: 46-51(in Portuguese).

[9] CHEN G. Electrochemical technologies in wastewater treatment[J]. Separation and Purification Technology, 2004, 38(1): 11- 41.

[10] BEN MANSOUR L., KOLSI K. and KSENTINI I. Influence of current density on oxygen transfer in an electroflotation cell[J]. Journal of Applied Electrochemistry, 2007, 37(8): 887-892.

[11] AKOSMAN C., ORHAN R. and DURSUN G. Effects of liquid property on gas hold-up and mass transfer in co-current downflow contacting column[J]. Chemical Engineering and Processing, 2004, 43(4): 503-509.

[12] BOUAIFI M., HéBRARD G. and BASTOUL D. et al. A comparative study of gas hold-up bubble size interfacial area and mass transfer coefficients in stirred gasliquid reactors and bubble columns[J]. Chemical Engineering and Processing, 2001, 40(2): 97-111.

[13] ZHAO B., WANG J. and YANG W. et al. Gas-liquid mass transfer in slurry bubble systems: I. Mathematical modeling based on a single bubble mechanism[J]. Chemical Engineering Journal, 2003, 96(1-3): 23-27.

[14] YANG W., WANG J. and ZHAO B. et al. Gas-liquid mass transfer in slurry bubble systems: II. Mathematical modeling based on a single bubble mechanism[J]. Chemical Engineering Journal, 2003, 96(1-3): 29-35.

[15] CHENA J. H., HSU Y. C. and CHENB Y. F. et al. Application of gas-inducing reactor to obtain high oxygen dissolution in aeration process[J]. Water Research, 2003, 37(12): 2919-2928.

[16] ASCE. Standard measurement of oxygen transfer in clean water[M]. American Society of Civil Engineers, 1984.

[17] KSENTINI I., KOTTI M. and BEN MANSOUR L. Effect of liquid phase physicochemical characteristics on hydrodynamics of an electroflotation column[J]. Desalination and Water Treatment, 2013, 51: 1-8.

[18] LUO X., LEE D. J. and LAU R. et al. Maximum stable bubble size and gas hold-up in high-pressure slurry bubble columns[J]. AIChE Journal, 1999, 45(4): 665-680.

[19] KANTARCI N., BORAK F. and ULGEN K. O. Bubble column reactors[J]. Process Biochemistry, 2005, 40(7): 2263-2283.

[20] KENDOUSH A. A., MOHAMMED T. J. and ABID B. A. et al. Experimental investigation of the hydrodyna-mic interaction in bubbly two-phase flow[J]. Chemical Engineering and Processing: Process Intensification, 2004, 43(1): 23-33.

[21] MACEIRAS R., áLVAREZ E. and CANCELA M. A. Experimental interfacial area measurements in a bubble column[J]. Chemical Engineering Journal, 2010, 163(3): 331-336.

[22] BHATTARAI A., SHAH S. K. and YADAV A. K. et al. Effect of solvent composition on the critical micelle concentration of sodium deoxycholate in ethanol water mixed solvent media[J]. BIBECHANA, 2013, 9: 63-68.

[23] KHAN A. M., SHAH S. S. Determination of Critical Micelle Concentration (CMC) of Sodiaum Dodcyl Sulfate (SDS) and the effect of low concentration of pyrene on its cmc using ORIGIN software[J]. Journal of the Chemical Society of Pakistan, 2008, 30: 186-191.

[24] INOUE T., MISONO T. and LEE S. Comment on “determination of the critical micelle concentration of dodecylguanidine monoacetate (dodine)”[J]. Journal of Colloid and Interface Science, 2007, 314(1): 334-336.

[25] KOTTI M., DAMMAK N. and KSENTINI I. et al. Effects of impurities on oxygen transfer rate in the electroflotation process[J]. Indian Journal of Chemical Technology, 2009, 16(6): 513-518.

[26] KOTTI M., KSENTINI I. and BEN MANSOUR L. Impact of anionic surfactants on oxygen transfer rate in the electroflotation process[J]. Desalination and Water Treatment, 2011, 36(1-3): 34-40.

[27] LIU Zhen-hua, GAO Yi-pu. Effect of surfactant on twophase flow patterns of water-gas in capillary tubes[J]. Journal of Hydrodynamics, Ser. B, 2007, 19(5): 630-634.

[28] XIA Guo-dong, CHAI Lei. Influence of surfactant on two-phase flow regime and pressure drop in upward inclined pipes[J]. Journal of Hydrodynamics, 2012, 24(1): 39-49.

[29] PAINMANAKULA P., LOUBIèREA K. and HéBRARDA G. et al. Effect of surfactants on liquid-side mass transfer coefficients[J]. Chemical Engineering Science, 2005, 60(22): 6480-6491.

10.1016/S1001-6058(13)60421-7

* Biography: KOTTI Mariam (1982-), Female, Ph. D., Assistant Researcher

BEN MANSOUR Lassaad, E-mail: lassaadbenmansour@yahoo.fr