Analysis of liquid circulation and mixing in a partitioned electrolytic tank

Analysis of liquid circulation and mixing in a partitioned electrolytic tank

International Journal of Multiphase Flow 37 (2011) 1191–1200 Contents lists available at ScienceDirect International Journal of Multiphase Flow j o ...

2MB Sizes 0 Downloads 0 Views

Recommend Documents

No documents
International Journal of Multiphase Flow 37 (2011) 1191–1200

Contents lists available at ScienceDirect

International Journal of Multiphase Flow j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / i j m u l fl o w

Analysis of liquid circulation and mixing in a partitioned electrolytic tank B. Ashraf Ali, S. Pushpavanam ⇑ Department of Chemical Engineering, IIT-Madras, Chennai 600 036, TN, India

a r t i c l e

i n f o

Article history: Received 16 February 2011 Received in revised form 3 June 2011 Accepted 6 June 2011 Available online 20 July 2011 Keywords: Gas evolution electrodes PIV Turbulence Electrolysis of water Mixing of acid–base chemicals

a b s t r a c t The performance of an electrochemical process depends critically on the mobility of the reacting species or ions towards the electrode surface. In this work, a partitioned electrolytic cell is studied. Here the fluid flow is induced by gases which evolve at the electrode surface. The liquid circulation induced by the rising bubbles is primarily responsible for mixing. In this study, the liquid circulation in a cell where an alkaline solution of water is electrolyzed using different Nickel designs of electrodes is investigated using PIV. For each electrode, the optimum operating conditions such as voltage and concentration of electrolyte which resulted in good mixing are found. The flow-field is quantified by calculating time averaged velocity profiles along the horizontal line and by analyzing the temporal variation of liquid velocity at a point. It is found that there are differences in the circulation and hence vorticity in the two compartments, anode and cathode. The effect of gas evolution on mixing between the two chambers is studied by taking uric acid in the cathode half and NaOH in the anode half. The flow induced by the evolved gas bubbles leads to convective mixing in the two chambers. The mixing time is calculated by measuring the variation of current with time under potentiostatic conditions. This is verified by measuring the pH in anode and cathode compartments during the electrolysis. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction In several electrolytic processes such as chlorine production, water electrolysis, there is gas evolution at the electrode surface (Philippe et al., 2005). The performance of such electrolytic processes is intrinsically related to mass transfer effects which are influenced by hydrodynamics. This has led to an increase in investigations on the role of hydrodynamics in electrochemistry. The dispersed phase (gas bubbles) modifies the electrical and thermal properties of the electrolyte, diffusive transport of electro active species and current density. This in turn, affects the macroscopic cell performance. Fluid flow in the cell depends on the release of gas bubbles, which in turn depends on the cell design. Thus, the understanding of gas–liquid flows in electrolytic systems is very important from the view-point of enhancing mass transport, system optimization and improving efficiency (Mat and Aldasb, 2005). The gases released at the electrode surfaces move upward due to buoyancy. The presence of the gas phase at the electrodes can be detrimental to the process as it blocks the active surface area of the electrodes and increases the resistance of the electrolyte (Boissonneau and Byrne, 2000). The presence of gas, in the form of small bubbles, and its motion has a significant impact on the performance of the electrolytic cell (Dahlkild, 2001).

⇑ Corresponding author. Tel.: +91 44 22574161; fax: +91 44 22570509/22570545. E-mail address: [email protected] (S. Pushpavanam). 0301-9322/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.ijmultiphaseflow.2011.06.006

Boissonneau and Byrne (2000) investigated the velocity field in an electrolytic cell with vertical electrodes through Laser Doppler Anemometry and Particle Image Velocimetry. Their experiment relates gas evolution to the hydrodynamics of electrolyte flow through a narrow vertical channel. They showed that although the flow is laminar in terms of the Reynolds number range considered, the bubbles induce local turbulence which causes velocity fluctuations due to the interactions with continuous phase. Sasaki et al. (2003) analyzed the behavior of the gas bubbles for optimizing the electrode spacing through PIV. They have shown that when the electrode spacing is shortened, the average rising velocity of bubbles decreases considerably due to wall friction. Their results also indicate that the hydrogen bubbles induce a faster flow locally near the electrode surface, while the oxygen bubbles rise at a relatively slow rate. Philippe et al. (2005) discussed the hydrodynamic and electrical properties in a laboratory scale electrolysis cell through CFD simulations based on a Lagrangian approach. They concluded that the fluid flow induced by bubble release from the electrodes had an important influence on the cell performance. Jupudi et al. (2009) studied the effect of bubbles on the current density distribution in an alkaline electrolyzer using a twodimensional Euler–Eulerian model. They found that their CFD model captures the effect of cell voltage and inter-electrode gap on current density accurately. Sun et al. (2009) investigated the flow in an electrolyzer using a three-dimensional CFD model. The computational model was validated for electrolysis of a zinc sulfate solution from PIV data.

1192

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

Aldas (2004) investigated gas evolution in a vertical electrolytic cell with a two-phase flow model developed in PHOENICS. The model showed that gas release is enhanced at high current density. However, a layer of gas accumulates on the electrode surface, decreasing the active area of reaction and adversely affecting the reaction rate. It was concluded that for electrolysis to proceed efficiently, the gas released should be removed from the reaction sites to increase surface area available for reaction. Most of the electrolysis studies so far were carried out using a porous or a membrane partition separating the cathode and the anode compartments. This allows transportation of ions between the chambers. In several systems such as electrolysis of magnesium chloride, there is a non porous barrier between the two chambers and a small gap at the bottom through which ion transport takes place. In such systems, it is possible to investigate the effect of gas evolution on the hydrodynamics in the two halves. The two halves are essentially decoupled except for the small opening below the partition. The gas evolution at the surface of the electrodes creates a two-phase flow in each half. In such a system, the reactions at each electrode can be studied in view of the partition. The two halves behave differently as the gases liberated at the two electrodes are different. This design can also be used to determine macroscopic information on mixing between two halves of such electrolytic systems.

Our objective is to understand the liquid flow-field in water electrolysis for different electrode designs using PIV. Nickel is used as the electrode material and the electrolysis of an alkaline solution of water is studied. Based on the hydrodynamics, a suitable electrode design and optimum operating conditions such as voltage and concentration of electrolyte is chosen. To understand mixing in the system, electrolysis is then carried out by keeping uric acid solution in the cathode half and NaOH solution in the anode half of the partitioned cell. This helps to understand the effect of gas evolution on fluid flow in each half and mixing. The paper is structured as follows. We first describe the experimental setup for carrying out hydrodynamic as well as mixing studies. The results are quantified in Section 3. We summarize our findings in the last section.

2. Experimental description In this work, focus is on an electrolytic system where gas evolution occurs from the surface of the electrode. For this, electrolysis of a solution of a 1 M NaOH was conducted in a transparent rectangular acrylic tank 30 cm high, 20 cm wide and 2.5 cm deep. The tank was partitioned at the middle using a Perspex sheet (Fig. 1a) and there was a gap of 1 cm at the bottom through which

Fig. 1a. Experimental setup for PIV analysis of water electrolysis system.

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

1193

Fig. 1b. Schematic representation of field of view by the camera, shaded domain. The open rectangle along which the circulation is calculated is also depicted. The lines along which the velocity is determined are also indicated.

the two compartments communicate, i.e., ions can flow. The partition helps to analyze the effect of released hydrogen and oxygen bubbles on the flow in each half of the cell. It helps to examine how the hydrodynamics in both halves is different. Nickel electrodes were used for electrolysis. These were mounted on a Teflon plate to ensure that the electrodes are positioned at the centre of the tank (along the depth). Hydrogen is liberated at the cathode and oxygen at the anode. From the stoichiometry, the amount of H2 liberated is twice that of O2 and the former is lighter than the latter. Particle Image Velocimetry (PIV) is used to quantitatively analyze the flow field in the system. Here, the liquid is seeded with tiny, neutrally buoyant rhodamine coated particles of poly methyl metha acrylate, PMMA (q = 1000 kg/m3, size 10 lm). The particles are assumed to follow the local flow faithfully so that the particle velocity directly measures the liquid velocity. A laser sheet of thickness 1 mm is formed by passing a double pulsed Nd–YAG

(532 nm, 120 mJ) laser beam through an optical arrangement consisting of cylindrical and spherical lenses. The sheet illuminates the plane of interest. The field of view is a 13.3 cm by 10.7 cm region and its location is indicated in Figs. 1a and 1b. It is chosen to span both halves of the electrolytic cell simultaneously. The particles in the flow field are illuminated twice at a time interval of 5000– 6000 ls. The light emitted from these particles is at a higher wavelength (k = 560 nm, red) than the incident green light (k = 532 nm). A CCD camera is positioned perpendicular to the plane of the light sheet to capture the light scattered from the rhodamine coated particles. An optical filter is placed in front of the camera which allows only the emitted red light from the fluorescent particles to enter the camera. The filter helps to capture the reflection from the particles in the liquid by filtering out the incident green light. This helps to obtain the velocities of the particles. Thus the liquid phase velocity is measured as only the liquid contains the fluorescent particles.

Fig. 2. Schematic diagram of various electrode designs: (a) 1-SS, (b) 3-SS, (c) 1-SL, (d) 3-SL and (e) rod.

1194

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

Fig. 3. (a–e) Time averaged flow field in the field of view of camera for 6 V, 1 M solution using (a) 1-SL, (b) 1-SS, (c) 3-SL, (d) 3-SS and (e) rod.

The total area under investigation is divided into small regions called interrogation windows. The displacement of particles in the time between the laser pulses in each interrogation window is

recorded by capturing the image of the particles in each pulse. The average displacement of particles position in each area is obtained using a cross correlation technique (Kompenhans et al.,

1195

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

1998). This gives the displacement in an interrogation window and hence the velocity. Three hundred images were taken for different electrode designs by varying operating conditions such as (i) voltage (ii) concentration of electrolyte (NaOH). This corresponds to a measurement time of 75 s. These were taken after the initial transients decayed and processed to get the liquid velocity vector field in the entire plane. Five different nickel electrode designs were tested and analyzed. A schematic diagram of these electrode designs is depicted in Fig. 2. 1SS refers to small single electrode and 3SS refers to three small electrodes in parallel. Here small (large) refers to the width 1(2) cm. 1(SL) and 3(SL) refer to a single electrode and three electrodes in parallel of width 2 cm each. The thickness of each of the flat electrodes was 1 mm. The spacing between the electrodes in the 3-SS and 3-SL designs is 3 mm. These electrode designs are mounted on a Teflon lid of the electrolytic tank. The position of the electrodes between the sidewalls is 8 mm for 3-SS, 3-SL and 12 mm for 1-SS, 1-SL designs. To understand the effect of gas evolution on mixing between the two electrolytes separated by partitioned plate, experiments were carried out with NaOH in the anode compartment and uric acid (UA) in the cathode compartment. The experiments were conducted at a fixed voltage and the evolution of current with time was measured. Here at the beginning of the experiment the two halves were completely separated i.e., the bottom opening was closed. The two halves were filled with the liquid containing different solutes. Initially, the fluids in the two compartments have different conductivities. The gap below the partition was opened and the electrolysis was allowed to proceed. As time progresses, the two fluids mix and the composition in the two halves becomes uniform and a steady current is obtained. To quantitatively analyze mixing, pH was measured at a depth of 5 cm in the system from the top surface. The points from where sample was withdrawn are marked by (H) in Fig. 1a. Initially, the pH was measured at four points equispaced (2.5 cm) from the left side. No appreciable change in pH was observed across the points in each half at each time instant. This confirmed that each half was well-mixed. Hence, the pH was measured at a distance of 5 cm from the two walls as a function of time to monitor mixing in each half. An equimolar concentration 0.1 M of UA and NaOH was first used. The electrolysis was found to be slow for this concentration. To increase the rate of electrolysis, concentration of NaOH was changed to 1.0 M in the anode chamber and that of uric acid was increased to 0.5 M in the cathode chamber.

3. Results and discussion The theoretical voltage for electrolysis of water is 1.23 V. But at this voltage, no electrolysis was observed for 1 M NaOH solution due to resistances in the system. The voltage was increased gradually and first signs of electrolysis (liberation of gas bubbles) were observed at 6 V. Measurement of velocity field in a planar region along the mid plane z = 0.01 m in the rectangular tank was carried out using PIV. The voltage was varied from 6 V to 18 V and the velocity field was measured. The time averaged x and y components of velocities (u and v) were analyzed in the planar region where images were taken. The instantaneous planar velocity fields were time averaged and plotted for different electrode designs. The time averaged flow field obtained experimentally for each kind of electrode was compared first. The velocity vectors are predominantly in the upward direction along the sides of electrodes in both compartments because the liquid is dragged up by the rising gas bubbles produced at the electrodes. Away from the electrode towards the central baffle, the velocity vectors are in the downward direction. The streamlines are in the form of closed curves

in clockwise direction in the cathode (left) compartment where H2 is released and counter clockwise direction in the anode (right) compartment where O2 is released. These are depicted in (Fig. 3a– e) for the different electrode designs used. The circulation of the flow field was determined to quantify the irreversibility induced by different electrode designs. It is estimated from a line integral around a closed curve of the fluid velocity and is given by (Fox et al., 2003)



I

v  dl

ð1Þ

c

where ‘‘v’’ is the fluid velocity on a small differential element, ‘‘dl’’ of the closed curve ‘‘C’’ in the flow-domain. The closed curve chosen for calculation is the perimeter of two rectangles depicted in Fig. 1b. As the electrode design is changed from 1-SS to 1-SL, i.e., when the width of the electrode surface increases, the circulation value decreases (Table 1, compare C of 1-SL with 1-SS). When the number of flat surfaces in parallel is increased for the smaller size electrode, the circulation value increases marginally (Table 1, compare C of 1-SS with 3-SS). On the other hand, for the larger size flat electrode (Table 1, compare C of 1-SL with 3-SL), an opposite effect, i.e., a decrease in circulation was observed with an increase in the number of parallel electrodes. When a rod was used as electrode for electrolysis, the amount of circulation induced is less in comparison with other electrode designs. The trends observed in the above behavior can be explained as arising due to the interaction of different variables such as surface area available for gases released, gap between the electrodes and the reactor wall. The corresponding flow fields for those different electrodes are shown in Fig. 3. From this analysis, it is concluded that the 3-SS electrode has the highest circulation. This electrode (3-SS) was chosen for further analysis. The variation in circulation for different applied voltages and concentrations of electrolyte for the 3SS electrode is now studied. Fig. 4a–c depicts the time averaged flow field obtained for different applied voltages at 1 M concentration of NaOH for 3-SS electrode. The corresponding circulation values are given in Table 2. It is observed that there is an increase in circulation as voltage increases and it results in a considerable change in size and shape of the circulation cells (Fig. 4a–c). The negative (positive) sign in the anode (cathode) indicates that flow is in counter clock wise (clock wise) direction. The maximum circulation was found at 18 V. It is seen that the voltage has a significant effect in deciding the flow pattern. Fig. 4d–f depicts the effect of voltage on the flow field when the concentration of NaOH was 2 M for 3S-S electrode design. The corresponding circulation values are given in Table 3. When the circulation values for 2 M solution were compared with those of 1 M solution, it was found that there is an increase in circulation values in anode side whereas there is a decrease in the cathode side. An increase in the concentration of solution (from 1 M to 2 M) increases the current passing through the solution which in turn produces more gas bubbles at both electrode surfaces. It was observed that the produced O2 gas bubbles at the anode electrode side leave the system whereas H2 bubbles circulate inside the tank and do not Table 1 H Circulation ðC ¼ c v  dlÞ values for different electrode designs, 1 M, 6 V. S. No.

Electrode

Circulation in the cathode half (m2/s) 105

Circulation in the anode half (m2/s) 105

1 2 3 4 5

1-SL 1-SS 3-SL 3-SS Rod

20.9 109.6 6.2 124.2 12.1

61 75 20 86 46

1196

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

Fig. 4. (a–e) Time averaged flow field in the field of view of camera for 3-SS: (a) 6 V, 1 M, (b) 12 V, 1 M, (c) 18 V, 1 M, (d) 6 V, 2 M, (e) 12 V, 2 M and (f) 18 V, 2 M.

leave the system. This is attributed to impurities in water which get washed away to the rear of the gas bubbles. This leads to a surface tension gradient along the bubble. This gradient creates an

additional stress in the tangential direction, the Marangoni stress (Clift et al., 1978). This Marangoni stress increases the drag force which is dominant for the small sized hydrogen bubbles. They

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200 Table 2 H Circulation ðC ¼ c v  dlÞ values for different applied voltage, 1 M, 3-SS. S. No.

Voltage (V)

Circulation in the cathode half (m2/s) 105

Circulation in the anode half (m2/s) 105

1 2 3

6 12 18

124.2 149.4 191.4

86 149 253

Table 3 H Circulation ðC ¼ c v  dlÞ values for different applied voltage, 2 M, 3-SS. S. No.

Voltage (V)

Circulation in the cathode half (m2/s) 105

Circulation in the anode half (m2/s) 105

1 2 3

6 12 18

108.5 64.6 25.4

134 151 282

cause the smaller hydrogen bubbles to follow the liquid. They circulate inside the tank itself without escaping to the atmosphere, as the concentration of the solution increases. This decreases the circulation in the anode half as the voltage increases. From the above studies, it is observed that the maximum liquid circulation is obtained while using a 3-SS flat electrode design, 1 M NaOH concentration for 18 V. This electrode design was investigated quantitatively by analyzing the experimentally obtained velocity vectors. For this, a horizontal line (Y = 0.174 m) was chosen and velocity vectors were analyzed along that line for different voltages and concentrations of electrolyte. In Fig. 5a–b, the effect of different electrode designs on time averaged electrolyte velocity profiles for 1 M NaOH are compared. There is some kind of symmetry for the Vx profile around (0.11, 0) and for Vy profile around x = 0.11. The symmetry is not exact as the gases liberated on the two sides are different. Fig. 6a–b depicts the time averaged x and y components of velocity along the horizontal line Y = 0.174 m for 3-SS electrode design. It is observed that the magnitude of liquid velocity increases with an increase in voltage as the production of gas bubbles increases at higher voltages in both the compartments. The liquid velocity obtained at the cathode side is higher in comparison with that at the anode side. This is attributed to the higher bubble production in cathode compartment as compared to the anode

1197

compartment. In the cathode compartment, the lighter hydrogen bubbles are released and this rise up faster inducing a higher velocity. In Fig. 7a–b, the time averaged liquid velocity components are compared for different concentrations of solution using 3-SS flat electrodes at 18 V. It is observed that as the concentration of NaOH solution increases, the spatial variation of liquid velocity in cathode and anode side decreases due to higher production of smaller size bubbles. These circulate inside the tank, affect the flow and reduce the liquid velocity. To analyze the nature of flow field, velocity components at a point in cathode and anode compartments were extracted from the instantaneous velocity fields. It is seen that both the velocity components exhibit variations with time due to turbulent fluctuations around a mean value (Fig. 8a–d). The temporal variation of liquid velocity at a point shows a steady state in the time averaged sense. This implies that the gas bubbles evolved from the electrodes rise straight without oscillations in the liquid domain. Here the flow is turbulent in both the compartments. In order to further analyze the turbulence level produced by the evolution of bubbles, turbulent intensity profiles were determined along the line Y = 0.174 m for 3-SS electrode. This is depicted in Fig. 9. This is calculated experimentally as the ratio of the RMS value of velocity to the average value of velocity. It is observed that even though the production of H2 bubbles is twice that of O2 bubbles, the turbulent intensity increased with voltage in both compartments in a similar manner. The effect of gas evolution on mixing when NaOH is in one half and a solution of uric acid is in the other is discussed now. Fig. 10 depicts the variation of current through the circuit and pH at the points (5 cm and 25 cm) and (15 cm and 25 cm) with time during electrolysis. Here, the origin is at the left lower corner of the tank. The current reached a steady state value (152 mA) at 80 s for an applied voltage of 6 V. This was verified by measuring pH value at the two points, one in each compartment as described. Initial pH in the uric acid chamber was found to be 5.7 and that of NaOH chamber was found to be 11.8. As gas evolved from the surface of electrode, a drop in pH value of NaOH was observed in the anode compartment. This was attributed to the neutralization of alkali by uric acid. The time to attain steady state was determined as the time when the curves attain a plateau. It is seen that the pH as well as current attains a constant value after 80sec. This is the time at which the solutions mix uniformly and the composition in both

Fig. 5. (a and b) Time averaged velocity for different electrodes, 1 M, 18 V: (a) Vx and (b) Vy.

1198

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

Fig. 6. (a and b) Time averaged velocity along a line Y = 0.174 for different voltage in the case of 3-SS flat electrode design: (a) Vx, 1 M and (b) Vy, 1 M.

Fig. 7. (a and b) Time averaged velocity along the line Y = 0.174 m for different electrolyte concentration, 18 V for 3-SS: (a) Vx and (b) Vy.

halves becomes equal and there is no change in current. Here, the compositions are uniform and the electrolyte properties are constant resulting in a steady current. As the voltage increases from 6 V to 18 V (Fig. 11), the time to attain steady state decreases from 80 s to 30 s. This is attributed to an increase in electrolysis rate, which is accompanied by an increase of gas evolution at the electrodes. This increases the rate of mixing through the bottom opening by convection. The effect of increase in concentration of NaOH for a fixed concentration of uric acid (0.5 M) at 12 V is depicted in Fig. 12. It is observed that with an increase in the concentration of NaOH from 0.5 M to 2.0 M, the time taken to attain steady state or uniform mixing decreases from 70 s to 30 s. Here, the conductivity of the solution increases and this results in an increase in the electrolysis rate. The effect of increase in concentration of uric acid for a fixed concentration of NaOH (1.0 M) at 12 V is shown in Fig. 13. It is

observed that with an increase in the concentration of uric acid from 0.5 M to 2.0 M, for a fixed NaOH concentration, the mixing time decreases from 80 s to 50 s. This confirms that the electrolytically evolved gas bubbles from the surface of the electrodes helps to mix the two electrolytes.

4. Summary and conclusion The effect of gas evolution on hydrodynamics of water electrolysis was studied in a partitioned electrolytic system using PIV for various electrode designs and different operating conditions such as voltage and concentration. The optimum condition for mixing was determined based on liquid circulation through circulation. It was found that maximum liquid circulation was obtained for a 3-SS electrode (smaller in dimension) in parallel connection at 18 V when a 1 M solution is used. Circulation was computed as a

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

1199

Fig. 8. (a–d) Temporal variation of liquid velocity at point [(a), (c), H2, (0.0678, 0.1743)]  Vx, [(b), (d), O2, (0.1538, 0.1743)]  Vy.

Fig. 9. Time averaged turbulent intensity profile along a line Y = 0.174 m for different voltage, 3S-S electrode, 1 M.

Fig. 10. Variation of pH in the two chambers and current with time at 6 V.

1200

B. Ashraf Ali, S. Pushpavanam / International Journal of Multiphase Flow 37 (2011) 1191–1200

Fig. 11. Variation of current for different voltages (concentration of uric acid = 0.5 M, NaOH = 1.0 M).

Fig. 13. Variation of current with time for different concentrations of uric acid at 12 V (NaOH = 1.0 M).

This was attributed to the increase in conductivity of the solution and a faster electrolysis rate due to a lowered resistance. The mixing time decreases when the concentration of uric acid increases again as the resistance is lowered. Thus the electrolytically evolved gas bubbles from the surface of the electrode induce appreciable mixing of electrolytes in the two chambers by convective transport across the gap below the baffle. Acknowledgements We thank Mr. Narayanan Rao, Technical officer, Chemical Engineering Department, IIT-Madras for helpful discussions and Industrial Consultancy and Sponsored Research (ISP/08-09/019/ICSR/ SPUS), IIT-Madras for their financial support. References

Fig. 12. Variation of current for different concentration of NaOH at 12 V (concentration of uric acid = 0.5 M).

measure of irrotationality or irreversibility in the system. This was further verified quantitatively by analyzing time averaged velocity profiles along a line. The temporal variation of liquid velocity at a point was also analyzed. It was found that velocity components exhibit turbulent fluctuations around a mean value. To understand the effect of gas evolution on mixing in such electrolytic system was studied using uric acid and NaOH in the two chambers. The variation of voltages, concentration of NaOH and uric acid on mixing was studied. Mixing time was estimated by measuring the variation of current with time and the results were verified with pH measurement. The results from these two measurements were consistent. An increase in applied voltage of electrolysis results in a decrease in mixing time accompanied by an increase in gas evolution at the electrodes. The effect of increase in concentration of NaOH was found to decrease the mixing time.

Aldas, K., 2004. Application of a two-phase flow model for hydrogen evolution in an electrochemical cell. Appl. Math. Comput. 154, 507–519. Boissonneau, P., Byrne, P., 2000. An experimental investigation of bubble-induced free convection in a small electrochemical cell. J. Appl. Electrochem. 30, 767– 775. Clift, R., Grace, J.R., Weber, M.E., 1978. Bubbles, Drops and Particles. Academic Press, San Diego. Dahlkild, A.A., 2001. Modeling the two-phase flow and current distribution along a vertical gas-evolving electrode. J. Fluid Mech. 428, 249–272. Fox, R.W., McDonald, A.T., Pritchard, P.J., 2003. Introduction to Fluid Mechanics. Wiley. Jupudi, R.S., Zhang, H., Zappi, G., Bourgeois, R., 2009. Modeling bubble flow and current density distribution in an alkaline electrolysis cell. J. Comput. Multiphase Flows 1, 341–348. Kompenhans, J., Raffle, M., Willert, 1998. Particle Image Velocimetry – A Practical Guide. Springer, Berlin. Mat, M.D., Aldasb, K., 2005. Application of a two-phase flow model for natural convection in an electrochemical cell. Int. J. Hydrogen Energy 30, 411–420. Philippe, M., Jerome, H., Sebastien, B., Gerard, P., 2005. Modeling and calculation of the current density distribution evolution at vertical gas-evolving electrodes. Electrochim. Acta 51, 1140–1156. Sasaki, T., Nagai, N., Murai, Y., Yamamoto, F., 2003. Particle Image Velocimetry measurement of bubbly flow induced by alkaline water electrolysis. In: Proceedings of PSFVIP-4, France. Sun, Z., Zhang, H., Li, P., Li, B., Lu, G., Yu, J., 2009. Modeling and simulation of the flow field in the electrolysis of magnesium. J. Miner. Met. Mater. Soc. 61, 29–33.