Evaluating the Behavior of Electrolytic Gas Bubbles and Their Effect

Oct 1, 2012 - Centre for Energy (M473), The University of Western Australia, 35 Stirling Highway, ... International Journal of Energy Research 2017 10...
0 downloads 0 Views 1005KB Size
Article pubs.acs.org/IECR

Evaluating the Behavior of Electrolytic Gas Bubbles and Their Effect on the Cell Voltage in Alkaline Water Electrolysis Dongke Zhang* and Kai Zeng Centre for Energy (M473), The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia ABSTRACT: The behavior of electrolytic gas bubbles and their effect on the cell voltage in water electrolysis were studied theoretically and experimentally. A fundamental force analysis was employed to predict the critical diameter for the departure of the electrolytic gas bubbles. Good agreement between the predictions and observations was obtained. It was found that increasing the electrode potential strengthened the force due to the interfacial tension and increased the critical diameter, while increasing the electrolyte concentration led to a reduction. This was explained by the changes in both the contact angle and surface tension. Many more fine gas bubbles were observed at high current density, which was explained by that the enhanced natural convection forced bubbles to depart prematurely. The cell voltage was only slightly reduced by the electrolyte circulation, which reduced the critical diameter for bubble departure. This confirmed that the layer of fine bubbles represented a significant energy barrier, that is, the additional resistance due to the bubble curtain formed on the electrodes, for alkaline water electrolysis.



can be made.9,10Some excellent literature reports examined the possible forces acting on a growing bubble in boiling.11−13 In addition to the predictions of the critical departure diameters of electrolytic bubbles, force analysis was also used to correlate bubble coverage on an electrode with electrolyte flow.14,15 One of the most important forces in the force balance analysis for electrolytic gas bubbles is the interfacial tension force. Different from the bubbles in boiling, the electrolytic gas bubbles experience a different interfacial tension in the presence of an electrical charge layer, electrolyte concentration gradient, and possible temperature gradient.16 All these have yet to be considered in the modeling of electrolytic gas bubble behavior. Despite many theoretical models or empirical studies describing the gas bubble behavior, little theoretical work, backed by our own experimental observations, has quantitatively examined the behavior of these electrolytic gas bubbles and correlated them with the behavior of an electrolysis cell. This paper attempts to analyze the forces acting on an electrolytic gas bubble by developing a model to explain the gas bubble behavior in alkaline water electrolysis and correlate the gas bubble behavior with the resistance effects of gas bubbles on the cell voltage.

INTRODUCTION Water electrolysis to split water into hydrogen and oxygen has been known for centuries, and it can be used as a means of energy storage, especially when combined with renewable sources for remote areas or distributed energy systems.1 However, one of the issues holding back the widespread applications of alkaline water electrolysis is its high energy consumption per unit volume of gas produced. Among the causes of high energy consumption, the evolution of hydrogen and oxygen bubbles poses both electrochemical resistance and mass transfer barriers to the electrode reactions.2 Currently, research activities on the electrolytic gas bubble phenomena share a common goal, that is, to minimize the resistance of electrolytic gas bubbles as a means to improve the efficiency of alkaline water electrolysis.1 Several techniques, including ultrasonic field,3 magnetic field,4 and super gravity,5 have shown the ability to facilitate bubble removal and reduction in the cell voltage. The bubble behavior and characterization in water electrolysis were also examined under microgravity.6,7 Although the resistance of electrolytic gas bubbles can be reduced by applying various techniques, the evolution of electrolytic gas bubbles in solutions is a poorly understood complex phenomenon. Similar to gas bubble evolution caused by boiling or desorption, electrolytic gas bubble formation involves nucleation, growth, and departure.8 All these steps determine both the residence time and the diameter of the gas bubbles, which are important parameters for determining the resistance effect. It is thus important to gain a detailed understanding of the electrolytic gas bubble behavior which will help to alleviate the resistance of the electrolytic bubbles. The departure of the electrolytic gas bubbles is one of the most influential steps in determining the resistance effects of the gas bubbles. A thorough force analysis is critical to gaining a better understanding and to forecasting the electrolytic gas bubble behavior and their departing from the electrode. Keeping the difference between electrolytic gas bubbles and bubble evolution caused by boiling in mind, a careful analogy © 2012 American Chemical Society



THEORETICAL ANALYSIS A force balance analysis around an electrolytic bubble was performed, based on two types of electrolytic gas evolution models in this work. Both models were based on a vertical electrode: the first was an electrolytic gas bubble evolution model simply involving the bubble formation on an electrode and was noted as the stagnant model; the second was a model that involved electrolyte flow, by adding an upward electrolyte Received: Revised: Accepted: Published: 13825

April 20, 2012 September 20, 2012 October 1, 2012 October 1, 2012 dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

flow over the electrode, and this bubble evolution model was noted as the f low model. Figure 1a shows a gas bubble on an electrode surface in the stagnant model. In the flow model, the gas bubble experiences

ρL is the density of the liquid. σ is the surface tension between the gas and the electrolyte liquid. θ is the contact angle between the liquid−gas and liquid−solid interfaces, R and r are the radius of the gas bubble and its circular contact area with the electrode, respectively, as shown in Figure 1b. ex and ey are vectors representing the force on the directions of the x and y coordinates, respectively. Expansion Force. Due to the growth of the gas bubble with the gas production, the pressure of the bubble experiences a dynamic change, which can be described by the well-known Rayleigh equation.12 The gas bubble radius can be expressed18 as a function of time by R(t) = kt1/2, where k is an empirical coefficient. There is a force present due to the expansion of the bubble which is estimated by FG = −2πρLR2Ṙ 2ey,11,12 where Ṙ 2 denotes the growth rate of the radius with respect to time, Ṙ 2(t) = (1/2)kt−1/2 = k2/2R, so the expansion force can be simplified as eq 4. FG = −πρL k 4ey

where the coefficient k is a constant determined experimentally. Interfacial Tension Force. The interfacial tension force exerted on the gas bubble exists along the circular contact area where the gas, liquid, and solid phases are in contact with each other. It can be expressed by eq 512,13 in the directions of the x and y coordinates, respectively.

Figure 1. Schematic diagram of a gas bubble on an electrode surface (a) and the forces acting on the bubble (b).

an upward electrolyte flow. The x coordinate is in the direction against gravity, and the y coordinate is normal to the x coordinate and pointing away from the electrode, respectively. The electrolyte flows in the direction of the x coordinate. Figure 1b shows the forces acting on the gas bubble originating from various sources. The buoyancy and surface tension exist due to the density difference between liquid and gas and the property of the solution, respectively.17 In the presence of the electrolyte flow, a drag force and a lift force also come into play.16 In this work, the force incurred by the temperature field effect is neglected by maintaining the temperature of the electrolyte constant. The forces acting on a gas bubble can be decomposed into components along the x and y coordinates, resulting in possible movements of gas bubbles in the corresponding directions. These movements are noted as departure and liftoff, respectively. Buoyancy. The buoyancy, FB, is composed of the force from the pressure and the gravity on the mass of the gas bubble, which is expressed by eq 1. FB =

∫A

B

pL (x) dA +

∫A

pB dA − C

∫V ρB g dV B

FS = rσ

(1)



cos θ cos ϕ dϕ ex − rσ

∫0



sin θ dϕ ey

(5)

σ = σ1 + k1Δc

(6)

σ = σ2 + k 2Δp

(7)

σ = σ3 + k 3ΔT

(8)

σ = σ4 + k4ΔU

(9)

All these empirical equations provide quantitative information on the trend of the interfacial tension force for the prediction of the critical diameter. Drag and Lift Forces. Due to the flow velocity distribution shown in Figure 1a, the drag and lift forces acting on a bubble attached to a wall can be derived. By definition, they are in the directions of the x and y coordinates12 and can be expressed as

(2)

where VB = 1/3πR3(1 + cos θ )2 (2 − cos θ )

∫0

where ϕ is the circumferential angle as in Figure 1b. Equation 5 includes the situation where the gas bubble experiences an inclination due to the buoyancy during its growth, where the contact angle θ is a function of ϕ. In this simplified model, the electrolytic gas bubble is assumed to grow symmetrically. The interfacial tension force can be expressed as FS = −2πrσ sin αey, where σ is the surface tension between the electrolyte liquid and the gas. The surface tension and the contact angle are the two key components determining the interfacial tension force. However, they have been approximated as properties of the liquid. Due to the fact that the origin of the interfacial tension is from the contact of the three phases, more factors need to be considered in modeling the bubble behavior in water electrolysis, for example, the electrode potential and surface roughness of the electrode. Equations 6−9 show that the surface tension can be found as a function of the electrolyte concentration gradient,19 pressure gradient,20 temperature gradient,12 and voltage gradient,20 respectively.

where AB and AC are the surface area of the gas bubble contacting the electrolyte and the contact area between the gas bubble and the electrode, respectively. pL and pB are the pressures of the liquid and gas bubble, respectively. VB is the gas bubble volume, and ρB is the gas bubble density. The gravitational force on the gas bubble mass is assigned to be negative as it acts in the opposite direction of the x coordinate. After rearrangement and integration,12 the buoyancy can be simplified15 as eq 2. FB = (ρL − ρB )gVBex + [(ρB − ρL )gR + 2σ /R ]πr 2ey

(4)

(3) 13826

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

FD = (1/2)ζDρLv2Aaex and FL = (1/2)CLρLv2Aaey, respectively, where ζD and CL are the drag and lift coefficients, respectively. v is the velocity of flow. Aa is the projected area of the bubble on the vertical plane, that is, the maximum bubble cross section perpendicular to the flow direction; Aa = πR2[1 − (θ − cos α sin θ)/π].15 Bubble Departure Diameter Predictions. When the gas bubble is attached to the electrode surface as shown in Figure 1, both ∑Fx = 0 and ∑Fy = 0 are satisfied. Once one of these conditions is broken, the gas bubble departs or lifts off from the electrode, respectively. In the stagnant model, when the gas bubble attaches on the electrode surface, ∑Fx = Fb > 0. Therefore, the buoyancy force is not balanced with the sum of the other forces. This unbalance will force the bubble to tilt upward to reach a new force balance and then departs when the bubble can no longer tilt after sufficient growth in size. The upward tilting of the bubble causes so-called advancing and receding angles,12 which are denoted as α and β in Figure 2. Therefore, there will be an interfacial tension force in the x

R= 6 sin θσ(α − β)[sin α + sin β ] (ρL − ρB )[π 2 − (α − β)2 ]g (1 + cos θ )2 (2 − cos θ ) (13)

where ρL and ρB can be considered as constants. Combining eqs 6−9 with eq 13, it can be predicted that varying the electrolyte concentration and the electrode potential will alter the interfacial tension force and thus the critical diameter for the bubble departure. These changes can be estimated in theory, and thus the critical diameter for the bubble departure can be calculated. Experimental data can be recorded to verify the dependence of the critical diameter on parameters such as the electrode potential and electrolyte concentration. The theoretical values can then be compared to validate the predictions. To calculate the critical diameter for bubble departure, a few parameters in eq 13 need to be obtained independently and analyzed, such as the contact angle and surface tension. The advancing and receding contact angles are usually measured experimentally.7 They characterize the flexibility of the gas bubble in distortion. It is determined by the surface tension, thus the electrode surface defects,21 and the property of the gas and the liquid.22 Although the understanding of phenomenon is still poor, it can be ascertained that the bubble size is a function of θ, α, β, and the interfacial tension σ. The contact angles for both hydrogen bubbles and oxygen bubbles had been experimentally determined by Matsushima et al.7 and adapted for our model predictions. The average contact angles for hydrogen bubbles and oxygen bubbles were 43 and 50°, respectively. The advancing angle and the receding angle can be written as eqs 14 and 15, and Δθ was reported23 with values of not more than 10°. The critical diameters of the hydrogen bubble and the oxygen bubble can be calculated according to eq 13.

Figure 2. Advancing and receding angles of a gas bubble attached to a vertical electrode surface.

coordinate direction. This force will be balanced with the buoyancy force to maintain the bubble attachment on the electrodes. Assuming a linear relationship between the contact angles and the circumferential angle,13 that is, θ = β + (α − β)(ϕ/π), substitution of θ into the interfacial tension force equation results in FS, x = −2rσ

π (α − β ) [sin α + sin β ] π 2 − (α − β)2

(10)

FS, y = −2rσ

π [cos β − cos α] α−β

(11)

(14)

β = θ − Δθ

(15)

Take a 0.5 M KOH solution for an example: its density is 1.02 g·cm−3 at the temperature of 20 °C, and its surface tension is 72.4 mN·m−1. Assuming Δθ is 6°, the critical diameters for the departure of hydrogen and oxygen gas bubbles were predicted to be 0.50 and 0.57 mm, respectively. Increasing the potential applied to the electrodes enhanced the wettability of the electrodes,24 which led to an increase in the surface tension and Δθ. On the other hand, increasing electrolyte concentration led to an increase of electrolyte viscosity and surface tension. The increase in the electrolyte viscosity caused not only less coalescence of bubbles25 but also a reduction in Δθ. In the flow model, the drag force was added to the force balance so that

Replacing eqs 1, 2, and 4 in the force balance FB,x + FS,x = 0 yields eq 12.

(ρL − ρB )gπR3(1 + cos θ )2 (2 − cos θ )

1/3(ρL − ρB )gπR3(1 + cos θ )2 (2 − cos θ ) − 2rσ

+

3 π (α − β ) [sin α + sin β ] − 2rσ 2 π − (α − β)2

π (α − β ) [sin α + sin β ] π − (α − β)2 2

=0

α = θ + Δθ

(12)

=0

Therefore, the critical diameter of a gas bubble, above which the bubble departs, in the stagnant model, denoted by D, can be approximated as eq 13.

ζDρL v 2A a 2

(16)

Therefore, applying electrolyte flow will bring down the critical diameter for bubble departure. 13827

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

Figure 3. Schematic of the experimental setup of the alkaline water electrolysis cell and the associated circulation system.



EXPERIMENTAL SECTION

An Autolab Potentiostat 302N electrochemistry workstation was used for generating and recording potential and current signals. The potentiostat technique was chosen for recording the relationship between current and voltage. Due to the fact that the electrolyte convection caused by the bubble movements and electrode potential may have a significant effect on the departure of the bubbles, the water electrolysis was conducted at a very low current density to minimize the effect of the bubble induced convection. When the alkaline water electrolysis was started, the behavior of electrolytic gas bubbles on the Ni electrodes was filmed using cameras from the sides of the cell at a frame rate of 30 frames/s. The video records were analyzed offline on a computer so that the size of gas bubbles on frames of interest could be determined. The KOH concentrations used were 0.5, 1, 2, and 4 M, respectively. The electrolyte was maintained at 22 ± 1 °C to minimize the effect of temperature. Polarization curves were plotted as the cell voltage against the current density. These polarization curves were employed to characterize the resistance due to the gas bubbles according to the stagnant model and the flow model. The average electrolyte flow velocity was set to 0.26 m·s−1, and the corresponding Reynolds number (with respect to the cubic cell width or depth, i.e., 1 cm) was calculated to be 2521, 2548, 2447, and 2002 at the electrolyte concentrations of 0.5, 1.0, 2.0, and 4.0 M, respectively.

According to the theory discussed above, the interfacial tension force and drag force are dominant factors in determining the gas bubble detachment. The interfacial tension force is determined by the factors related to the interfaces between the gas, liquid, and solid phases. Several factors to be considered are the electrolyte concentration, cell voltage, temperature, gas property, and solid property. For comparison, the electrolyte concentration and cell voltage can be varied to examine their effects, while the temperature, gas property, and electrode surface property are maintained constant or considered to be unchanged during operation. On the other hand, the lift and drag forces will affect the departure diameter by introducing the electrolyte flow. The experimental setup, illustrated in Figure 3, consisted of an electrolysis cell and an electrolyte circulating system. The electrolysis cell was a custom-made cubic glass cell. The inner dimensions of the cubic cell were 1 cm × 1 cm × 20 cm by depth, width, and height, respectively. The anode and cathode employed were both Ni plates of 1 cm × 1 cm. The Ni plates were soldered on copper wires for good electrical connection. A polish treatment was applied to the electrodes using sandpaper with an average sand grain size of 5 μm. One side of the Ni plate with the copper wire soldering joint was masked with epoxy resin, exposing the other side to the electrolyte. Two Ni electrodes were placed in the electrolysis cell back to back, facing the cameras located outside the cell to capture the dynamic images of the H2 and O2 bubbles. This arrangement enables the behaviors of the bubbles to be observed simultaneously. The electrolyte circulation system contained an electrolyte tank of 20 L capacity, a diaphragm pump, and a flow meter, as shown in Figure 3, where “V”, “P”, and “F” stand for the valve, the pump, and the flow meter, respectively. The electrolyte was potassium hydroxide (KOH) solutions with concentrations up to 4 M. The electrolyte could be pumped at a rate up to 6 L·min−1, giving a superficial velocity of the electrolyte in the vertical flow cell in the range of 0−1 m·s−1. By passing through the vertical flow cell, the electrolyte was circulated back to the electrolyte tank.



RESULTS AND DISCUSSION Dependence of Critical Diameter for Bubble Departure on Cell Voltage. Figure 4 shows typical images of hydrogen bubbles in 0.5 M KOH at different current densities. To avoid the coalescence of bubbles, the current density applied was set at very low values so that the growth and detachment of a single bubble could be observed and analyzed individually. Each image of Figure 4 shows a number of individual hydrogen bubbles of different diameters. These bubbles were at different stages of growth. The images were extracted immediately before the largest bubble detached. Therefore, the largest bubble on each image represented the critical diameter under the particular current density. 13828

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

The dependence of the critical diameters on cell voltage for hydrogen and oxygen bubble departure is presented in Table 1. Table 1. Critical Diameters for Hydrogen and Oxygen Bubble Departure at Different Cell Voltages in 0.5 M KOH at 22 ± 1 °C critical diameter, mm current density, mA·cm−2

cell voltage, V

0.30 0.45 0.60 0.75

1.72 1.83 1.88 1.93

hydrogen 0.59 0.88 1.09 1.03

± ± ± ±

0.02 0.04 0.07 0.02

oxygen 0.60 0.89 1.08 0.96

± ± ± ±

0.04 0.06 0.07 0.09

The trend of increasing the critical diameters with increasing cell voltage can be explained by the increased interfacial tension at higher electrode potentials.20 At a higher current density, the cell voltage was also higher and therefore the interfacial tension force in the x coordinate direction was greater. Furthermore, the bubble buoyancy force had to be large enough to overcome the interfacial tension force by increasing the bubble diameter. However, at the highest current density of 0.75 mA·cm−2, the critical diameter for both hydrogen and oxygen bubbles decreased a little. This could also be attributed to the high current density. In this case, high current densities can lead to two possible changes. The first is that the high voltage applied would result in local heating due to high ohmic loss. The localized heating will cause a temperature gradient on the electrode surface. The second is the number of gas bubbles. It was observed that more bubbles formed and departed at the high current density as shown in Figures 4d and 5d. Both the temperature field and the bubble movement resulted in enhanced natural convection or microconvection of the electrolyte. This would have caused some bubbles to depart prematurely. It should be noted that the critical diameter concept as mathematically formulated can only be applied to ideal situations where the temperature throughout the electrolysis cell is uniform, the electrolyte is stagnant, and the electrode surfaces are smooth so that the physical properties of the nucleation sites (such as the surface roughness) for the bubble growth are uniform for all bubbles. In experimentation and practical water electrolysis systems, it would be rather difficult to attain the ideal situation and, as such, the observed bubble departure may not be uniform. It was interesting to note that, during the experiments at the high cell voltages, at 0.75 mA·cm−2 for instance, some fine bubbles much smaller than the observed critical diameters for hydrogen and oxygen, respectively, also departed. It can also be explained as the natural convection of the electrolyte caused by the bubble movement and temperature field due to local heating. The natural convection may prevent the small bubbles at the nucleation site from growing. Therefore, small bubbles beyond the detection limit departed from the electrode. Dependence of Critical Diameter for Bubble Departure on Electrolyte Concentration. Since the electrolyte concentration also plays an important role in the interfacial force, its effect on the critical diameter for bubble departure was also examined. Figure 6 shows typical images of hydrogen bubbles at the current density of 0.60 mA·cm−2 in KOH solutions of different concentrations. It was found that increasing the KOH concentration from 0.5 to 4 M dramatically decreased the critical diameter for hydrogen

Figure 4. Typical images of hydrogen bubbles in 0.5 M KOH, at 22 ± 1 °C, at different current densities (mA·cm−2): (a) 0.3; (b) 0.45; (c) 0.6; (d) 0.75.

The critical diameter for hydrogen bubble departure increased with increasing current density. When the current density increased from 0.3 to 0.60 mA·cm−2, the critical diameter also increased from 0.59 to 1.09 mm. The number of hydrogen bubbles at a higher current density was also larger than that at a lower current density. Figure 5 shows typical images of oxygen bubbles in 0.5 M KOH at different current densities. Note that the critical diameter for oxygen gas bubble departure increased from 0.60 to 1.08 mm when the current density increased from 0.3 to 0.60 mA·cm−2.

Figure 5. Typical images of oxygen bubbles in 0.5 M KOH, at 22 ± 1 °C, at different current densities (mA·cm−2): (a) 0.3; (b) 0.45; (c) 0.6; (d) 0.75. 13829

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

the critical bubble diameter decreased. This led to only one possible explanation; that is, the increase in viscosity results in an adverse effect. It is thought that the change in viscosity led to the decrease in Δθ. The increased viscosity made it harder for the bubble to tilt or stretch, which would result in the decrease in Δθ. According to eq 10, the interfacial force in the direction of the x coordinate would also decrease. Ultimately, a decrease in the critical diameter for bubble departure occurred as indicated by eq 13. Bubble Behavior at High Cell Voltages. The emergence of fine bubbles departing from the electrode had been identified for both hydrogen and oxygen bubbles when the cell voltage was 1.88 V. It was reasonable to predict that more fine oxygen and hydrogen bubbles would form if the cell voltage increased further. As shown in Figure 7, as the cell voltage increased from

Figure 6. Typical images of hydrogen bubbles at 0.6 mA·cm−2, at 22 ± 1 °C, in KOH solutions of different concentrations (M): (a) 0.5; (b) 1; (c) 2; (d) 4.

bubble departure and the number of bubbles was reduced as well. A similar decrease in the critical diameter was also found for the oxygen bubbles. At a given current density, 0.60 mA·cm−2 in this case, an increase in the electrolyte concentration resulted in a decrease in cell voltage from 1.88 to 1.77 V and an increase in electrolyte surface tension from 73 to 84 mN·m−1,26 which would have opposite effects on the critical diameter according to eqs 6, 9, and 13. These two factors need to be examined separately for their effects on the critical diameter. Table 2 lists the critical diameters for hydrogen bubbles and corresponding cell voltages at all concentrations tested. The cell Figure 7. Typical images of hydrogen bubbles in the water electrolysis at different cell voltages in 0.5 M KOH electrolyte at 22 ± 1 °C: (a) 2.2, (b) 2.4, (c) 2.6, and (d) 2.8 V, where the current densities were 20, 32, 48, and 60 mA·cm−2, respectively.

Table 2. Critical Diameters for Hydrogen Bubbles in Different KOH Concentrations at 22 ± 1 °C and Current Density of 0.6 mA·cm−2 KOH concn, M

cell voltage, V

0.5 1.0 2.0 4.0

1.88 1.80 1.78 1.77

crit diam (hydrogen), mm 1.08 0.50 0.36 0.24

± ± ± ±

2.2 to 2.8 V, increasingly more fine bubbles detached from the electrode. The diameters of these fine bubbles in Figure 7c,d could not be detected by the current imaging technique. One possible explanation to this phenomenon is that at high cell voltages, and therefore high current densities, gases were produced at high rates. New gas bubbles repelled the adjacent gas bubbles formed earlier away from the electrode surfaces. In the meantime, the natural convection caused by the upward electrolyte motion due to the rapidly rising bubbles and the temperature field caused by the field localized heating would have also enhanced the departure of the rapidly forming gas bubbles.4 Effect of Electrolyte Circulation. The images in Figure 8 were taken immediately before the largest bubble departed. Figure 8a and Figure 8b show typical images of hydrogen bubbles in the 0.5 M KOH electrolyte solution at the current density of 0.75 mA·cm−2 without and with electrolyte circulation, respectively. The Reynolds number of the electrolyte flow was 2521. As can be seen, the diameter of the largest bubble departing the electrode was dramatically reduced as expected when the electrolyte circulation was applied due to

0.07 0.05 0.03 0.02

voltages decreased from 1.88 to 1.77 V when the concentration of KOH varied from 1 to 4 M. The decreasing trend of the critical diameter was in accordance with its dependence on the cell voltage. Assuming the decreasing cell voltage was the only cause of the bubble size reduction, the critical diameter should be located between 0.59 and 1.08 mm according to Table 1. However, the critical diameter of the hydrogen bubbles was reduced dramatically when the electrolyte concentration increased. Therefore, it can be ruled out that the cell voltage change alone caused the reduction in the hydrogen bubble critical diameter. On the other hand, as the KOH concentration increased, the viscosity of the electrolyte increased and the surface tension would have also increased.26 The increase in surface tension would result in the increase in the critical diameter. However, 13830

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

The polarization curves with filled symbols were recorded with electrolyte flows in 0.5, 1.0, 2.0, and 4.0 M KOH, respectively. There were small reductions in cell voltage in the polarization curves with electrolyte flows compared to those without electrolyte flow. This may be attributed to the enhanced removal of the growing bubbles at the electrode surface by the electrolyte circulation. However, due to the fact that the bubble curtain still existed on the electrode, especially at high current densities, it can be inferred that this bubblingfroth layer represented most of the resistance caused by the bubbles.1



COMPARISON OF MODEL PREDICTIONS WITH EXPERIMENTAL OBSERVATIONS The prediction of the critical diameter for bubble departure was based on eq 13, where σ, θ, α, and β are unknown parameters. The values or ranges of these parameters were sourced from the literature.7,23,26 Table 3 summarizes the values used in the Table 3. Summary of Parameters Used in the Prediction of the Critical Diameter for Bubble Departure

Figure 8. Typical images of hydrogen bubbles in 0.5 M KOH at the current density of 0.75 mA·cm−2 (a) without circulation and (b) with circulation, and at the current density of 200 mA·cm−2 (c) without circulation and (d) with circulation, where the Reynolds number for both cases was 2521.

current density, mA·cm−2

the drag and lift forces of the bubble-enhanced convention. At a current density of 200 mA·cm−2, it was found that a large number of small gas bubbles were generated so that the fine bubbles formed a bubble curtain on the electrode surface. The application of the electrolyte circulation reduced the critical diameter for departure as the circulation velocity increased. However, the bubble curtain was still on the surface of the electrode which still posed a resistance barrier for water electrolysis. Figure 9 shows the polarization curves of water electrolysis in the stagnant and flow models at different KOH concentrations.

concentration, M

0.30 0.45 0.60 0.75 0.5 1.0 2.0 4.0

σ, mN·m−1

θH2, deg

θO2, deg

Δθ, deg

84.0 89.0 94.0 100.0 94.0 94.0 94.0 94.0

50 50 50 50 50 50 50 50

43 43 43 43 43 43 43 43

6 8 9 10 8 4 3 1.5

prediction of the critical diameter of bubble departure. The surface tension of the KOH solution was found to be 73.3, 78.5, 81.7, and 86.4 mN·m−1 for the electrolyte solutions of 0.5, 1.0, 2.0, and 4.0 M KOH, respectively.26 Due to the effect of the electrode potential, the values of surface tension were increased accordingly. The contact angles for oxygen and hydrogen were independent of the electrolyte and electrode potential, and were reported to be 50 and 43° according to Matsushima et al.7 α and β were assumed through eqs 14 and 15, where Δθ was chosen23 in the range between 0 and 10°, and the values of Δθ and σ chosen were in agreement with the previous analysis of the effect of the current density and KOH concentration.7,23,26 With the predicted values for the contact angle and surface tension, the critical diameter for bubble departure can be calculated. Figure 10 compares the predicted and measured values of the critical diameters under different conditions. As can be seen from Figure 10, generally good agreement is evident for both hydrogen and oxygen gas bubbles. Unfortunately, due to the lack of data, it was difficult to quantify the relationship between the drag and lift forces acting on the bubble in the presence of electrolyte circulation. Therefore, the predicted values were not presented in Figure 10. However, Figure 8 shows a significant decrease in the critical bubble diameter, which is also in qualitative agreement with expectations from eq 16.

Figure 9. Polarization curves at different KOH concentrations with and without electrolyte circulation.

The flow rate was maintained at 1.6 L·min−1 and the Reynolds numbers were 2521, 2548, 2447, and 2002 for electrolytes of 0.5, 1.0, 2.0, and 4.0 M KOH, respectively. The polarization curves with open symbols were recorded in different electrolyte concentrations without electrolyte flows. The polarization curve for 4.0 M KOH showed the lowest cell voltages, followed by the polarization curves for 2.0, 1.0, and 0.5 M KOH.



CONCLUSIONS A detailed force analysis was applied to analyze the behavior of hydrogen and oxygen gas bubbles formed on smooth electrode surfaces during alkaline water electrolysis. The critical diameter 13831

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832

Industrial & Engineering Chemistry Research

Article

(2) Hine, F.; Murakami, K. Bubble effects on the solution IR drop in a vertical elctrolyzer under free and force convection. J. Electrochem. Soc. 1980, 127, 292−297. (3) Li, S. D.; Wang, C. C.; Chen, C. Y. Water electrolysis in the presence of an ultrasonic field. Electrochim. Acta 2009, 54, 3877−3883. (4) Iida, T.; Matsushima, H.; Fukunaka, Y. Water electrolysis under a magnetic field. J. Electrochem. Soc. 2007, 154, E112−E115. (5) Wang, M.; Wang, Z.; Guo, Z. Water electrolysis enhanced by super gravity field for hydrogen production. Int. J. Hydrogen Energy 2010, 35, 3198−3205. (6) Kiuchi, D.; Matsushima, H.; Fukunaka, Y.; Kuribayashi, K. Ohmic resistance measurement of bubble froth layer in water electrolysis under microgravity. J. Electrochem. Soc. 2006, 153, E138−E143. (7) Matsushima, H.; Fukunaka, Y.; Kuribayashi, K. Water electrolysis under microgravity. Part II. Description of gas bubble evolution phenomena. Electrochim. Acta 2006, 51, 4190−4198. (8) Jones, S. F.; Evans, G. M.; Galvin, K. P. Bubble nucleation from gas cavities - a review. Adv. Colloid Interface Sci. 1999, 80, 27−50. (9) Darby, R.; Haque, M. S. The dynamics of electrolytic hydrogen bubble evolution. Chem. Eng. Sci. 1973, 28, 1129−1138. (10) Vogt, H.; Aras, O.; Balzer, R. J. The limits of the analogy between boiling and gas evolution at electrodes. Int. J. Heat Mass Transfer 2004, 47, 787−795. (11) Thorncroft, G. E.; Klausner, J. F. Bubble forces and detachment models. Multiphase Science and Technology 2001, 13, 35−76. (12) Van Helden, W. G. J.; Van Der Geld, C. W. M.; Boot, P. G. M. Forces on bubbles growing and detaching in flow along a vertical wall. Int. J. Heat Mass Transfer 1995, 38, 2075−2088. (13) Klausner, J. F.; Mei, R.; Bernhard, D. M.; Zeng, L. Z. Vapor bubble departure in forced convection boiling. Int. J. Heat Mass Transfer 1993, 36, 651−662. (14) Vogt, H.; Balzer, R. J. The bubble coverage of gas-evolving electrodes in stagnant electrolytes. Electrochim. Acta 2005, 50, 2073− 2079. (15) Eigeldinger, J.; Vogt, H. The bubble coverage of gas-evolving electrodes in a flowing electrolyte. Electrochim. Acta 2000, 45, 4449− 4456. (16) Cole, R.; Papazian, J. M.; Wilcox, W. R. Bubble departure radii at solidification interfaces. Int. J. Heat Mass Transfer 1980, 23, 219− 224. (17) Kulkarni, A. A.; Joshi, J. B. Bubble formation and bubble rise velocity in gas-liquid systems: A review. Ind. Eng. Chem. Res. 2005, 44, 5873−5931. (18) Zeng, L. Z.; Klausner, J. F.; Mei, R. A unified model for the prediction of bubble detachment diameters in boiling systems– I. Pool boiling. Int. J. Heat Mass Transfer 1993, 36, 2261−2270. (19) Weissenborn, P. K.; Pugh, R. J. Surface-Tension and Bubble Coalescence Phenomena of Aqueous-Solutions of Electrolytes. Langmuir 1995, 11, 1422−1426. (20) Lubetkin, S. The motion of electrolytic gas bubbles near electrodes. Electrochim. Acta 2002, 48, 357−375. (21) Drelich, J.; Miller, J. D.; Good, R. J. The effect of drop (bubble) size on advancing and receding contact angles for heterogeneous and rough solid surfaces as observed with sessile-drop and captive-bubble techniques. J. Colloid Interface Sci. 1996, 179, 37−50. (22) Tadmor, R. Line energy and the relation between advancing, receding, and young contact angles. Langmuir 2004, 20, 7659−7664. (23) Yeoh, G. H.; Cheung, S. C. P.; Tu, J. Y.; Ho, M. K. M. Fundamental consideration of wall heat partition of vertical subcooled boiling flows. Int. J. Heat Mass Transfer 2008, 51, 3840−3853. (24) Brussieux, C.; Viers, P.; Roustan, H.; Rakib, M. Controlled electrochemical gas bubble release from electrodes entirely and partially covered with hydrophobic materials. Electrochim. Acta 2011, 56, 7194−7201. (25) Craig, V. S. J.; Ninham, B. W.; Pashley, R. M. The effect of electrolytes on bubble coalescence in water. J. Phys. Chem. 1993, 97, 10192−10197. (26) Dunlap, P. M.; Faris, S. R. Surface Tension of Aqueous Solutions of Potassium Hydroxide. Nature 1962, 196, 1312−1313.

Figure 10. Comparison of predicted and measured critical diameters for hydrogen and oxygen gas bubbles.

for electrolytic gas bubble departure was highly dependent on the electrolyte concentration and cell voltage at low current densities. The critical diameters for the hydrogen and oxygen bubble departure increased from 0.59 to 1.03 mm and from 0.60 to 1.08 mm as the electrode potential increased from 1.72 to 1.93 V, respectively. The critical diameter for hydrogen bubble departure decreased to 0.27 mm as the KOH concentration increased from 0.5 to 4 M, respectively. Similar findings were also found to be true for the oxygen gas bubbles. This was explained by the fact that an increase in the electrode potential resulted in an increase in the interfacial tension, while an increase in the KOH concentration had an adverse effect by reducing Δθ. The convection caused by the local heating, bubble detachment, and electrolyte flow also influenced the critical diameter of the electrolytic gas bubbles. The predicted critical diameters were generally in good agreement with the measured values. At high cell voltages or electrolyte flow, the corresponding convections induced by the upward flowing bubbles caused the other bubbles to depart prematurely. The application of electrolyte circulation dramatically reduced the bubble sizes. It was also found that only a small reduction in the cell voltage was caused by the electrolyte circulation. This was explained by the gas bubble curtain formed on the electrode surface, which presented a significant energy barrier for alkaline water electrolysis.



AUTHOR INFORMATION

Corresponding Author

*Tel.: 61-8-6488 7600. Fax: 61-8-6488 7235. E-mail: Dongke. [email protected]. Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This research was supported under the Australian Research Council’s Linkage Projects Scheme (Project Nos. LP0669575 and LP100200135).



REFERENCES

(1) Zeng, K.; Zhang, D. Recent progress in alkaline water electrolysis for hydrogen production and applications. Prog. Energy Combust. Sci. 2010, 36, 307−326. 13832

dx.doi.org/10.1021/ie301029e | Ind. Eng. Chem. Res. 2012, 51, 13825−13832