i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Electrochemical investigation of the Bunsen reaction in the sulfureiodine cycle

Zhi Ying, Yanwei Zhang*, Qiaoqiao Zhu, Pingan Peng, Zhihua Wang, Junhu Zhou, Jianzhong Liu, Kefa Cen
State Key Laboratory of Clean Energy Utilization, Zhejiang University, Hangzhou 310027, China

article info
Article history: Received 1 July 2013 Received in revised form 29 August 2013 Accepted 5 September 2013 Available online 2 October 2013 Keywords: Sulfureiodine cycle Electrochemical Bunsen reaction Electrolysis Cation exchange membrane Hydrogen production

abstract
The Bunsen reaction, as a part of the sulfureiodine thermochemical cycle, was studied using an electrochemical cell. The effects of current density, operating temperature, H2SO4 concentration in the anolyte, HI concentration and I2/HI molar ratio in the catholyte were explored. Both the H2SO4 in anolyte and the HI in catholyte were concentrated during electrolysis. Increasing current density amplied this H2SO4 and HI concentration, while the other operating parameters also varied the anolyte and catholyte concentration. The transport properties of the cation exchange membrane were examined. The electrode current efciency remained close to 100% for most runs except those at high current density. Both the average cell voltage and the heat equivalent of electric energy were determined at different conditions. Crown Copyright 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

The sulfureiodine (SI) thermochemical cycle, rst proposed by General Atomics [1], is considered as one of the most promising routes for massive hydrogen production. The SI process consists of the Bunsen Reaction (1) and the hydriodic acid and sulfuric acid decomposition Reactions (2, 3), listed as follows: SO2 I2 2H2 O 2HI
573 K773 K 293 K393 K

H2 SO4 2HI

(1) (2) (3)

H2 I2 SO2 H2 O 0:5O2

H2 SO4

1073 K1273 K

Reaction (1), known as the Bunsen reaction, produces sulfuric acid (as an aqueous H2SO4 solution) and hydriodic acid

(as a HI solution). Both acids are puried, concentrated, and then decomposed into hydrogen and oxygen, as shown in Reactions (2) and (3). The traditional direct contact mode of the Bunsen reaction requires a large excess of both iodine and water with respect to their stoichiometric values [1]. The excess iodine induces a separation of the solution produced into two immiscible phases [2,3] and the excess of water makes the reaction thermodynamically more favorable [4]. One of the disadvantages of the traditional SI cycle is that a large amount of energy is required to remove the excess iodine and water before the decomposition of HI [5]; moreover the circulation of iodine in the whole cycle is undesirable, due to its corrosive characteristics. Therefore, the overall thermal efciency of the SI cycle is greatly reduced. In recent years, some alternative ways for carrying out the Bunsen reaction have been proposed in order to reduce the

* Corresponding author. Tel.: 86 571 87952040; fax: 86 571 87951616. E-mail address: zhangyw@zju.edu.cn (Y. Zhang). 0360-3199/$ e see front matter Crown Copyright 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijhydene.2013.09.016

14392

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

excess amounts of iodine and water [6]. The addition of a precipitating agent, such as lead sulfate, was proposed to remove iodide and thus induce separation of the two acids without the excess of iodine [6]. Reaction solvents, such as liquid SO2, tributylphosphate (TBP), ionic liquids, and molecular solvents, were used to extract the HI produced in the Bunsen reaction [6e9]; therefore, the presence of excess water can be avoided. Application of an electrochemical cell to the Bunsen reaction [10] is also a potential alternative. The electrolysis cell consists of two electrodes and two compartments separated by a cation exchange membrane. Gaseous SO2 dissolved in H2SO4 solution and solid iodine dissolved in HI solution is used as anolyte and catholyte, respectively. The reactions at the anode and cathode are: Anode : SO2 2H2 O/H2 SO4 2H 2e E0;a 0:172 V (4) Cathode : I2 2H 2e /2HI E0;c 0:535 V (5)

operating conditions. The transport properties of the cation exchange membrane strongly affect the cell performance, yet concrete data has not been reported in literature. In this work, the effects of current density, temperature, initial H2SO4 concentration in the anolyte, initial HI concentration and I2/HI molar ratio in the catholyte on the evolution of anolyte and catholyte concentration were studied. Cell voltage was recorded during electrolysis. The transport properties of the membrane were determined, and for different runs the current efciency and energy consumption during electrolysis were also calculated.

2.
2.1.

Experimental
Experimental setup

This method offers several advantages: (1) the required amount of iodine and water is reduced and the liquideliquid separator can be dropped; (2) it increases the concentration of H2SO4 and HI simultaneously; (3) the recirculation of unreacted reactant is reduced in the overall process. Disadvantage of this electrochemical method is that it requires electric power for electrolysis. The Bunsen reaction was studied by Nomura et al. in an electrochemical membrane reactor operated at room temperature [10]. Currentevoltage characteristics were evaluated and this study showed that the total resistance between the electrodes was independent of the H2SO4 concentration and decreased with declining I2/HI molar ratio. The overpotential at the anode was reduced by addition of a small amount of HI to the anolyte. In another experiment [11], conducted in a temperature range of 290e333 K, the resistance between the electrodes decreased with rising operating temperature, but it increased with rising H2SO4 concentration. Thermal efciency diminished with rising current density and H2SO4 concentration. The same group [12] found that thermal efciency grew with rising HI concentration and a maximum thermal efciency of 38.9% was obtained. In recent years, a group at the Indian Institute of Technology Delhi performed studies on the electrochemical Bunsen reaction. Cross-contamination through the membrane was measured [13] and showed that the electroosmotic ow induced by the current enlarged the concentration of sulfate ions and SO2 in the catholyte and reduced the concentration of iodide ions in the anolyte. The same group characterized platinum and graphite electrodes using linear sweep voltammetry (LSV) and cyclic voltammetry (CV) [14]. Areal-resistance of the membrane was measured at different operating conditions [15]. Effects of anolyte pressure, catholyte pressure, temperature, H2SO4 concentration in the anolyte, HI concentration and I2/HI molar ratio in the catholyte on cell voltage were investigated [16]. The concentration of HI and H2SO4 is one of the important advantages for the electrochemical Bunsen reaction. Moreover, it is desirable to study the variation of anolyte and catholyte concentration during electrolysis at different

The Bunsen reaction was carried out in the experimental setup shown in Fig. 1. It mainly consists of an electrochemical cell, anolyte and catholyte reservoirs, the solution recirculation section between cell and reservoir, and a D.C. power supply. Fig. 2 shows a schematic diagram of the electrochemical cell (Hephas Energy Co., Ltd.), consisting of two graphite electrodes and a Naon 117 membrane (DuPont Co., USA); the effective membrane area was 12.25 cm2 (3.5 cm 3.5 cm). Two rubber gaskets (170 mm thickness) were used at both sides of the membrane to prevent the intermixing of anolyte and catholyte. The temperature of the cell was controlled by means of a temperature controller (Hephas Energy Co., Ltd.). Two double-jacketed glass reservoirs operating at atmospheric pressure were placed on magnetic stirrers, and temperature was controlled using a thermostatic water bath (GX-2010, Shanghai Bilang Instrument Co.). Teon tubes were employed to connect the reservoirs with each compartment of the cell. Two peristaltic pumps (BT100-2J, Baoding Longer Precision Pump Co., Ltd.) were used for circulating the solutions. In order to avoid heat dissipation, the teon tubes were wrapped with heat insulation cotton. A D.C. power supply (PAN16-10A, KIKUSUI ELECTRONICS CORP., Japan) provided a constant current to the cell and monitored the cell voltage.

2.2.

Electrolysis

Table 1 lists the operating conditions for the electrochemical Bunsen reaction. The total amount of water (the sum of water in the acids and the extra feed water) was xed at 50 g for both anolyte and catholyte. The anolyte reservoir was initially lled with known amounts of sulfuric acid (about 98 wt. %, Quzhou Juhua reagent Co., Ltd.) and deionized water, while the catholyte reservoir was lled with known amounts of hydriodic acid (approximately 56 wt%, Shanghai Kefeng Industrial Co., Ltd.), iodine (>99.9%, Chinasun Specialty Products Co., Ltd.) and deionized water. Both solutions were recirculated through their respective loops by two peristaltic pumps, the ow rate of both anolyte and catholyte was xed at 10 ml/min. To start the test, the stirrer, the thermostatic water bath, and the temperature controller were switched on. After the system had reached the desired temperature, gaseous SO2 was continuously fed at a constant ow rate of 20 ml/min into the H2SO4 solution of the anolyte reservoir. This bubbling

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

14393

D.C supply

MFC A node Cathode

SO 2 H 2SO 4 solution Magnetic stirrer T wo gaskets & a Magnetic stirrer cation exchange membrane Peristaltic pumps HI-I 2 solution

Fig. 1 e Experimental setup used for the electrochemical Bunsen reaction.

procedure was conducted for 1 h before the start of electrochemical reactions. Then, electrolysis was performed in the constant current mode supported by the D.C. power. The Bunsen reaction was carried out for 4 h, and the cell voltage was continuously recorded. Anolyte and catholyte were sampled periodically and analyzed by means of an autotitrator (ZDJ-5, Shanghai Leici Instrument Co.). Solutions of 0.5 N NaOH, 0.05 N KIO3, and 0.1 N Na2S2O3 were used to titrate H, I, and I2, respectively. To minimize the errors resulted from sampling and analysis, three samples were taken for measuring each ion; then, the average value and deviation of the H, I, and I2 concentration were determined.

3.
3.1.

Results and discussion

Theoretical

the reduction rate of I2 at cathode exceeds the transport rate of H through membrane, the I molality will become higher than the H molality, then HI concentration depends on the H concentration, otherwise, the molality of H is higher than that of I, and HI molality is determined by I molality. (2) The concentration of H in the anolyte is much higher than that of H in the catholyte. Hence, permeation of H from the anode to the cathode compartment may occur even without current passing. Even more H can migrate from the anode compartment to the cathode if current is applied. (3) Electroosmotic ow from anode to cathode occurs along with the migration of hydronium ions, leading to rising and SO2 in the catholyte and concentrations of SO2 4 diminution of I and I2 in the anolyte [13]. As a result, the weight of catholyte may increase and side reactions may occur. All effects above contribute to differences between the H and the I molality in the catholyte. The standard electrode potential (vs. standard hydrogen electrode (SHE)) of anode E0,a is 0.172 V, and that of cathode E0,c is 0.535 V. Then the equilibrium voltage of the cell is 0.535 V 0.172 V 0.363 V if the cell works as a voltaic cell. However, in practice, electrode and electrolyte overpotentials require that some electric power must be supplied to the system. The practical anode potential Ea and cathode potential Ec are as follows (using Nernst equation) [17]: E E
a 0;a

The theoretical moles of H2SO4 and HI formed during electrolysis can be calculated using Faradays law [13,14]: DN0 i It nF i H2 SO4 ; HI (6)

where I, t, and F are the constant current (A), the electrolytic time (s), and the Faraday constant (96,485 C/mol); n is the number of electrons required per mole of H2SO4 or HI formed in the half cell reaction. In theory, SO2 loses two electrons at the anode, consumes a water molecule, and produces SO2 4 and H ions; thus the H2SO4 concentration should steadily increase during electrolysis. Meanwhile, a portion of the H generated at the anode transports to the cathode through the membrane, and then constitutes HI with I ions reduced from I2; therefore, the HI concentration should also rise, while the I2 concentration is lessening during electrolysis. However, in practice the H and I molality in the catholyte obtained through titration is not equal; this is due to the three following effects: (1) The reduction rate of I2 at the cathode is not equal to the transport rate through the Naon 117 membrane, and once

aSO2 $a4 RT H 4 ln 2F pSO2

aSO2 $a4 8:31T H 4 ln 0:172 2 96485 pSO2

(7) RT aI ln 22 2F aI ! 8:31T aI ln 22 0:535 2 96485 aI ! (8)

E E
c

0;c

Here, aSO2 , aH are the activity of species and pSO2 is the 4 relative pressure SO2 in the anolyte, dimensionless. aI2 , aI are

14394

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

concentration increased with the electrolysis, so hc (less than 0) diminished during electrolysis, but its absolute value amplied. Therefore, the charge transfer resistance of both anode and cathode increased with the proceeding of electrochemical reactions, so the reaction rate steadily decreased. However, since more H was produced at anode, the concentration difference between two compartments was enlarged, thus the transport rate of H was increased, and more H could migrate from anode compartment to the cathode. According to the above analyses, the electrode reaction and the transport of H tend to be equilibrium with the electrolysis. Fig. 2 e Schematic diagram of the electrochemical cell. (i) cation exchange membrane, (ii) rubber gasket, (iii) graphite electrode, (iv) conductive plate, (v) aluminum plate, (vi) insulator, (vii) heater.

3.2. 3.2.1.

Concentration of H2SO4 and HI solution Effect of current density

the activity of species in the catholyte, dimensionless. R, T, and F are gas constant (8.31 J mol1 K1), temperature (K), and Faraday constant (96,485 C/mol). Then, the anode/cathode overpotential (ha/hc) is: ha Ea E0;a aSO2 $a4 RT H 4 ln 2F pSO2 RT aI ln 22 2F aI ! (10) ! (9)

hc Ec E0;c

The overpotential relates to the charge transfer resistance of the electrode, and thereby affects the electrode reaction rate. The activity of species is also closely related to their concentration in the solution. Both SO2 4 and H concentration in the anolyte increased during electrolysis, and thereby improved the activity of both ions, while SO2 kept at atmospheric pressure. Then ha increased during electrolysis. I2 concentration in the catholyte decreased, while I

Table 1 e Operating conditions tested for the electrochemical Bunsen reaction. Run Current Temperature Conc. Conc. I2/HI density molar (K) H2SO4 HI (A/dm2) mol=kgH2 O mol=kgH2 O ratio
1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2.5 3.5 5 15 25 5 5 5 5 303 10 6 0.5

323 343 303 303 303

10 13 16 10 10

6 6 7 8 6

0.5 0.5 0.5 0.2 0.8

Fig. 3 shows the variation of anolyte and catholyte concentration during electrolysis at different current densities. The symbols in the gure indicate the average of three measurements conducted at each condition, and the error bars show the calculated standard deviation, which remain less than 1.5%. For most current densities, the H2SO4 concentration in the anolyte augmented in the course of electrolysis (Fig. 3a). For a current density of 1 A/dm2, the H2SO4 molality amplied to a maximum value of 5.147 mol/kg before declining at the last hour. This can be attributed to rise in cell resistance, and the electrode reaction of SO2 became slower at low current density. However, the total weight of anolyte increased, along with the dissolution of SO2, leading to a decrease in H2SO4 molality. When the current density rose to 2.5e3.5 A/dm2, the molality of H2SO4 amplied at 1 h and then diminished before rising again at 3e4 h. The cell resistance was low and the electrode reaction was fast at the start of electrolysis, with a large quantity of H generated at the anode, but the transport of H from anode to cathode was slower than the anode reaction; hence, at the rst hour the molality of H was higher than that of SO2 4 . During electrolysis more H moves from the anolyte to the catholyte while the electrode reaction became slower; then, at 2 h the H2SO4 concentration declined. However, as the anode reaction and H migration tended to equilibrium with the electrolysis, leading to rising H2SO4 concentration again at 3e4 h. When the current density increased to values above 5 A/dm2, the H2SO4 molality increased steadily during electrolysis, indicating the equilibrium between the electrode reaction and the transport of H. However, for the current density higher than 15 A/dm2, the anode-side graphite electrode could react with H2O into CO2 or CO over 3 V of the cell voltage [11]; some carbon scraps were found in the anolyte, which affected the H2SO4 concentration. At 25 A/dm2 of the current density, the electrolysis was performed only 3 h according to the theoretical calculation that the completion of reaction was no more than 3 h, so only three points were obtained. The HI concentration in the catholyte increased with the electrolysis, and increase in current density increased HI concentration, as shown in Fig. 3b. At the rst hour of reaction, the measured molality of H in catholyte was lower than that of I, which was mainly due to the fact that the cathode reaction rate was higher than the proton transport rate,

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

14395

6.4 6.2

3.6

(a)

H2SO4 molality (mol/kg)

6.0 5.8 5.6 5.4 5.2

HI molality (mol/kg)

1 A/dm2 2.5 A/dm2 2 3.5 A/dm 5 A/dm2 15 A/dm2 25 A/dm2

(b)
3.4 3.2

1 A/dm2 2.5 A/dm2 3.5 A/dm2 5 A/dm2 15 A/dm2 25 A/dm2

3.0

2.8

2.6 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Electrolytic time (h)

0.85 0.80 0.75

Electrolytic time (h)

1 A/dm2 2.5 A/dm2 3.5 A/dm2 2 5 A/dm 15 A/dm2 2 25 A/dm

(c)

I2 molality (mol/kg)

0.70 0.65 0.60 0.55 0.50 0.45 1.0 1.5 2.0 2.5 3.0

3.5

4.0

Electrolytic time (h)

Fig. 3 e Evolution of anolyte and catholyte concentration during electrolysis at different current densities. (a) H2SO4 concentration in anolyte, (b) HI concentration in catholyte, (c) I2 concentration in catholyte (Anolyte: H2SO4 [ 10 mol=kgH2 O , Catholyte: HI [ 6 mol=kgH2 O and I2/HI [ 0.5, T [ 303 K).

therefore, the HI concentration at 1 h depended on H molality. Then, more H migrated from anode to cathode compartment during electrolysis along with the permeation of H2O, SO2 4 , and SO2 driven by electroosmotic ow, besides, rising charge transfer resistance resulted in the slowdown of electrode reaction, as a result, the measured H molality was higher than I molality, I molality determined the concentration of HI. For the current density lower than 2.5 A/dm2, the HI concentration increased at the rst hour because the cathode reaction was fast, then the cathode reaction slowed down and the catholyte weight increased with the permeation of hydronium ions, leading to the decrease of HI molality at 2 h. At the last two hours, HI concentration increased again because the equilibrium was reached between electrode reaction and proton transport. For current density above 3.5 A/ dm2, the HI concentration steadily increased during electrolysis, which suggested that increase in current density might favor the match between proton transport and cathode reaction. As seen in Fig. 3c, the I2 concentration decreased during the rst 3 h and then increased at 4 h. This might be due to the occurrence of side reactions at cathode, permeated SO2 4 and SO2 reacted with HI in the catholyte. At low current density, white particles were found to deposit on the membrane and cathode, which was also found by Nomura

et al. [11], indicating the formation of sulfur, two reactions might occurred at cathode compartment: H2SO4 6HI / S 3I2 4H2O, SO2 4HI / S 2I2 2H2O. At higher current density, H2S was detected, the chemical reactions might be H2SO4 8HI / H2S 4I2 4H2O, and/or SO2 6HI / H2S 3I2 2H2O. Increase in current density contributed to rising H2SO4 and HI concentration in anolyte and catholyte, respectively. However, a higher current density caused the corrosion of anode-side carbon electrode and increased the consumption of electric energy. In view of this, effects of other operating parameters were studied at current density of 5 A/dm2.

3.2.2.

Effect of temperature

Fig. 4 shows the evolution of anolyte and catholyte concentration at different temperatures. As seen in Fig. 4a, at each temperature H2SO4 was concentrated during electrolysis. The molality of H2SO4 decreased for rising temperature from 303 K to 323 K and then increased from 323 K to 343 K. This trend might be due to two factors [16]. The rst one is that an increase in temperature decreases the solubility of SO2, so that the conversion of SO2 to H2SO4 diminished when the temperature increased to 323 K, the nal H2SO4 molality decreased from 5.455 mol/kg at 303 K to 5.356 mol/kg at 323 K. The second one is that electrode reaction rates increase with rising

14396

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

5.60 5.55 5.50

(a)

303 K 323 K 343 K

3.3 3.2

(b)

303 K 323 K 343 K

H2SO4 molality (mol/kg)

HI molality (mol/kg)
1.0 1.5 2.0 2.5 3.0 3.5 4.0

3.1 3.0 2.9 2.8 2.7 2.6 1.0 1.5 2.0 2.5 3.0 3.5 4.0

5.45 5.40 5.35 5.30 5.25 5.20

Electrolytic time (h)

1.2

Electrolytic time (h)

(c)
1.1 1.0

303 K 323 K 343 K

I2 molality (mol/kg)

0.9 0.8 0.7 0.6 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Electrolytic time (h)

Fig. 4 e Evolution of anolyte and catholyte concentration during electrolysis at different temperatures. (a) H2SO4 concentration in anolyte, (b) HI concentration in catholyte, (c) I2 concentration in catholyte (Anolyte: H2SO4 [ 10 mol=kgH2 O , Catholyte: HI [ 6 mol=kgH2 O and I2/HI [ 0.5, Current density [ 5 A/dm2).

temperature, and the activities of solutions are higher at higher temperatures, as a result, the nal H2SO4 concentration increased again to 5.530 mol/kg at 343 K. The evolution of the HI concentration at different temperatures is shown in Fig. 4b. At 303 K, the HI molality increased steadily from 2.750 mol/kg to 3.066 mol/kg during electrolysis. For the temperature of 323 K and 343 K, HI molality increased to a high value at the rst hour, which was due to the high electrode reactions rates. Then the molality of HI decreased to a minimum value at 2 h before increasing again at the last two hours. This can be attributed to the decrease of cathode reaction rate and increase of proton transport rate. The molality of HI increased for rising temperature from 303 K to 323 K and then decreased for the further increase of temperature to 343 K. The highest HI concentration of 3.193 mol/kg was obtained at 323 K after 4 h electrolysis. As shown in Fig. 4c, increase in temperature increased I2 concentration in catholyte. Though the activity of I2 was improved at higher temperature, rising temperature inhibited the conversion of I2, leading to rising I2 concentration.

3.2.3.

Effect of H2SO4 concentration

Fig. 5 shows the evolution of anolyte and catholyte concentration for three initial H2SO4 concentrations of 10, 13, and

16 mol=kgH2 O in anolyte. Increase in the initial H2SO4 concentration reduced the solubility of SO2, then the change in concentrations of both the H2SO4 and the dissolved SO2 affected the activity of anolyte. The performance of the entire electrochemical cell, including the cathode reaction, was inuenced as well. Fig. 5a shows that the H2SO4 molality drastically increased at the rst hour and then changed slowly with the electrolysis, which suggested that the anode reaction was fast at the beginning of the electrolysis and then slowed down. Increasing the initial H2SO4 concentration amplied the H2SO4 molality and the difference between H2SO4 molality at the same time was comparable. As shown in Fig. 5b and c, the nial HI molality increased from 3.066 mol/kg to 3.243 mol/ kg for rising H2SO4 concentration from 10 mol=kgH2 O to 13 mol=kgH2 O , but then slightly lessened with further increase of H2SO4 concentration to 16 mol=kgH2 O . On the contrary, I2 molality declined as the concentration of H2SO4 rose from 10 mol=kgH2 O to 13 mol=kgH2 O and then amplied as the H2SO4 concentration increased to 16 mol=kgH2 O . The cell resistance might be the lowest and the conversion of reactants was highest at 13 mol=kgH2 O of H2SO4 concentration. At higher H2SO4 concentration, the membrane tended to dehydrate and its conductivity decreased. Therefore, a H2SO4 concentration of 13 mol=kgH2 O might be an optimal choice.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

14397

6.75 6.50 6.25

(a)
HI molality (mol/kg)

3.3 3.2 3.1 3.0 2.9 2.8 2.7 4

(b)

H2SO4 concentration 10 mol/kgH2O 13 mol/kgH2O 16 mol/kgH2O

H2SO4 molality (mol/kg)

6.00 5.75 5.50 5.25 5.00 4.75 4.50 0 1 2 3

H2SO4 concentration 10 mol/kgH2O 13 mol/kgH2O 16 mol/kgH2O

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Electrolytic time (h)

Electrolytic time (h)

0.8

(c)

H2SO4 concentration 10 mol/kgH2O 13 mol/kgH2O 16 mol/kgH2O

0.7

I2 molality (mol/kg)

0.6

0.5

0.4 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Electrolytic time (h)

Fig. 5 e Evolution of anolyte and catholyte concentration during electrolysis for different H2SO4 concentrations. (a) H2SO4 concentration in anolyte, (b) HI concentration in catholyte, (c) I2 concentration in catholyte (Catholyte: HI [ 6 mol=kgH2 O and I2/HI [ 0.5, T [ 303 K, Current density [ 5 A/dm2).

3.2.4.

Effect of HI concentration

Fig. 6 shows the variation of anolyte and catholyte concentration for three different initial HI concentrations in catholyte (6, 7, 8 mol=kgH2 O ). The I2 concentration was also altered, to keep the I2/HI molar ratio at 0.5. Rising initial HI concentration improved the activities of the catholyte and thereby increased the cathode reactions rates. Meanwhile, the anode reaction could be affected as well. As shown in Fig. 6a and b, the molality of both H2SO4 and HI enlarged as the initial HI concentration rose from 6 mol=kgH2 O to 7 mol=kgH2 O . However, when the initial HI concentration further increased to 8 mol=kgH2 O , the HI molality continued growing because of the high initial concentration, while the H2SO4 molality diminished, which may be due to rising cell resistance and a reduction of anode reaction rate. The variations of both the HI molality and the I2 molality were signicant at the rst hour and then became slow at the last 3 h. The I2 molality declined during electrolysis, and it increased with rising initial HI concentration since the initial I2 concentration was increased as well. But the side reactions occurred at 4 h for the initial HI concentration of 6 mol=kgH2 O contributed to the increase of I2 molality, so the lowest I2 molality of 0.634 mol/kg was obtained at initial HI concentration of 7 mol=kgH2 O . Then, a HI concentration of

7 mol=kgH2 O was advantageous as far as the nal solution concentration was concerned.

3.2.5.

Effect of I2/HI molar ratio

Fig. 7 shows the variation of solution concentration for three different I2/HI molar ratios in catholyte (0.2, 0.5, 0.8). A rise in I2/HI molar ratio amplied the I2 concentration and therefore improved the activity of I2 in the catholyte; then the overpotential at the cathode for reduction of I2 to I was reduced, but the resistance of catholyte solution could increase [13]. As shown in Fig. 7a, the H2SO4 molality rose with increasing I2/HI ratio, which might be due to the improvement in electrode kinetics and the slowdown of proton permeation at higher I2 concentration. In Fig. 7b, the HI molality continuously increased during electrolysis, and it was higher at the lower I2/HI molar ratio because of the corresponding higher initial HI concentration. Increase in I2/HI molar ratio amplied the total weight of catholyte and thereby reduced the initial HI concentration. For an I2/HI molar ratio of 0.5 and 0.8, the I2 molality increased at the last hour, possibly related to the occurrence of side reactions in the catholyte, since white particles were found to deposit on the membrane and cathode. For the traditional direct contact mode of Bunsen

14398

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

5.8

3.3

5.7

(a)

HI concentration 6 mol/kgH2O 7 mol/kgH2O 8 mol/kgH2O

3.2 3.1

(b)

H2SO4 molality (mol/kg)

5.6

HI molality (mol/kg)

3.0 2.9 2.8 2.7 2.6 2.5 2.4

5.5

5.4

HI concentration 6 mol/kgH2O 7 mol/kgH2O 8 mol/kgH2O

5.3

5.2 1.0 1.5 2.0 2.5 3.0 3.5 4.0

2.3 0 1 2 3 4

Electrolytic time (h)

1.4

Electrolytic time (h)

(c)
1.2

HI concentration 6 mol/kgH2O 7 mol/kgH2O 8 mol/kgH2O

I2 molality (mol/kg)

1.0

0.8

0.6

Electrolytic time (h)

Fig. 6 e Evolution of anolyte and catholyte concentration during electrolysis for different HI concentrations. (a) H2SO4 concentration in anolyte, (b) HI concentration in catholyte, (c) I2 concentration in catholyte (Anolyte: H2SO4 [ 10 mol=kgH2 O , Catholyte: I2/HI [ 0.5, T [ 303 K, Current density [ 5 A/dm2).

reaction, the optimal I2/HI value is in the range of 2e3 [18]. However, in the electrochemical Bunsen reaction, two acids with high purity can be obtained with a lower excess amount of iodine.

3.3.

Proton transport and water permeability

In the electrochemical cell the transport properties of the cation exchange membrane can be characterized by two parameters: the transport number of proton (t) and the water permeability coefcient (b) [19e21]. These parameters are dened as follows: t FDNc HI It (11)

DNc H2 O DNc HI

(12)

c Here, DNc HI and DNH2 O are the molar change of HI and of H2O in catholyte (mol). I, t, and F respectively denote the electric current (A), the electrolytic time (s), and the Faraday constant (96,485 C/mol). At each operating condition, the weight of anolyte and catholyte were measured after electrolysis by collecting the solution from the glass reservoirs. Small

amounts of solutions remained in the cell and tubes, however, which caused some errors in weighting the solutions. The parameters t and b obtained designate the entire process of the electrochemical Bunsen reaction. Table 2 lists the t and b values for different runs; Qe denotes the total amount of electricity consumed during electrolysis (C). In runs 1e4, values of t close to 1 indicate that the transport of H was awless whenever the current density remained below 5 A/dm2. However, the value of t decreased to less than 0.5 as current density further amplied to 15 and 25 A/dm2. The variation of b was contrary to t, an increase in current density decreasing the permeation of water. For runs 7e14, t was higher than 0.95, which indicated that the proton transport rate was high and that the HI in catholyte was fully concentrated. When the temperature increased from 303 K to 323 K and to 343 K, t increased while b drastically decreased to less than 0.15, indicating rising temperature promoted the concentration of HI in catholyte, which was in agreement with the result in Section 3.2.2. The parameter t increased while b rst increased and then decreased with rising H2SO4 concentration in anolyte and HI concentration in catholyte. An increase in the I2/HI molar ratio increased the permeation of water into cathode compartment.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

14399

5.7

4.0

(a)
5.6

I2/HI

(a)
HI molality (mol/kg)

H2SO4 molality (mol/kg)

0.2 0.5 0.8

(b)
3.6 3.2

I2/HI 0.2 0.5 0.8

5.5

2.8

5.4

2.4

5.3 2.0 5.2 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0 1 2 3 4

Electrolytic time (h)

Electrolytic time (h)

1.6

(c)

I2/HI 0.2 0.5 0.8

1.2

I2 molality (mol/kg)

0.8

0.4

0.0 0 1 2 3 4

Electrolytic time (h)

Fig. 7 e Evolution of anolyte and catholyte concentration during electrolysis for different I2/HI molar ratios. (a) H2SO4 concentration in anolyte, (b) HI concentration in catholyte, (c) I2 concentration in catholyte (Anolyte: H2SO4 [ 10 mol=kgH2 O , Catholyte: HI [ 6 mol=kgH2 O , T [ 303 K, Current density [ 5 A/dm2).

3.4. Evaluation of current efciency and energy consumption

For each run, the current efciency for both electrode reactions was calculated by comparing the change in the amount of acids with the amount expected from the amount of charge supplied by the D.C. power [13]. It is dened as follows: Anode ha I
DNa H It=2F
2 SO4

electric energy consumed per mole of H2 at each run is calculated using Eq. (15) [16]. UH2 2FVe hE hc I (15)

(13)

Cathode hc I

DNc HI It=F

(14)

c Here, ha I and hI denote the current efciency of anode and a cathode. DNH2 SO4 and DNc HI are the measured molar increase of H2SO4 in anolyte and HI in catholyte, respectively. It/2F and It/ F are the theoretical moles of H2SO4 and HI formed during electrolysis, 2 indicates that two mole electrons are required per mole of H2SO4 formed. I, t, and F denote the constant current (A), the electrolytic time (s), and the Faraday constant (96,485 C/mol). Since the nal desired product of the SI cycle is H2, the energy consumed in the electrochemical cell should be reported in terms of hydrogen produced. Heat equivalent of

Here Ve is the average cell voltage (V), which is calculated using the integral of the voltage with electrolytic time, and the obtained integral value is then divided by the total electrolytic time. F is the Faraday constant (96,485 C/mol), hE is the efciency for conversion of heat to electricity, which is assumed 0.4, and hc I is the current efciency of cathode. Table 3 shows the data for current efciency of anode and cathode and heat equivalent of electric energy consumed per mole of H2 after electrolysis. For current density of 1 A/dm2, ha I was less than 50%, which was due to the decrease of H2SO4 molality at the last hour (Section 3.2.1). For the low current density of 2.5e3.5 A/dm2, ha I was slightly above 100%. This may be explained by three factors: the rst one is the uncertainties in sampling and titration, the second is that a small amount of residual SO2 remained in anolyte and thereby interferes the titration of H, and the last one is that the electroosmosis ow from anode to cathode decreased the weight of anolyte and thus increased the concentration of H2SO4 in anolyte [13]. When the current density further increased to 15 and 25 A/

14400

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

Table 2 e tD and b at different operating conditions. Run

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Qe (C)
1728 4464 6192 8784 26496 33048 8784 8784 8784 8784 8784 8784 8784 8784

t
0.928 0.985 0.925 0.926 0.379 0.462 0.999 0.953 0.990 1.002 0.985 1.002 0.959 0.954

b
7.751 6.866 4.743 2.696 2.579 1.658 0.146 0.025 4.861 1.805 5.322 4.704 0.629 5.292

c dm2, ha I and hI drastically decreased to less than 80%. For the c temperature of 303, 323, and 343 K, ha I and hI reached the maximum value at 343 K and 323 K, respectively, agreeing well with the results in Section 3.2.2. Increase in H2SO4 concentration in anolyte or HI concentration in catholyte increased the current efciency of anode and cathode. Value of electrode current efciency close to 100% for Run 9e12 suggested that no side reactions occurred and there were no homogeneous reactions between the reactants that could lead to change in the concentration of H2SO4 and HI [13]. Therefore, a higher initial H2SO4 and HI concentration were preferred for anolyte and catholyte to suppress the side reaction of sulfur and c hydrogen sulde formation. ha I increased while hI rst decreased and then increased as the I2/HI molar ratio rose from 0.2 to 0.8. As shown in Table 3, the lowest UH2 of 108.92 kJ/mol-H2 was obtained at current density of 1 A/dm2, while value of UH2 at 15 and 25 A/dm2 were much higher. A lower UH2 can be obtained at lower current densities and will lead to a higher thermal efciency of SI cycle. However, the H2 production rate may be reduced at a lower current density, then the required size of electrochemical cell and membrane area must increase for a given production rate, leading to rising capital cost [14].

Increase in temperature increased the activity of anolyte and catholyte and thereby increased electrode reaction rates, the cell resistance was reduced and Ve decreased, leading to the decrease of UH2 . Ve slightly decreased with rising H2SO4 concentration, and the reduction of UH2 was also small. Increase in HI concentration increased the activity of catholyte and decreased the cell resistance, as a result, both Ve and UH2 diminished signicantly. I2/HI molar ratio has two effects on cell voltage [13]. An increase in I2/HI molar ratio increases I2 concentration and therefore reduces the required overpotential at cathode for reduction of I2 to I. On the other hand, increase in I2 concentration also increases ohmic resistance of the catholyte and hence increases the cell voltage. As I2/HI ratio increased from 0.2 to 0.5, the ohmic voltage drop may be the major effect, thus Ve increased. But the voltage decreased with further increase of I2/HI molar ratio to 0.8, which was due to the reduction of overpotential at cathode was more than the increase in ohmic voltage drop. As the I2/HI molar ratio increased from 0.2 to 0.8, UH2 increased from 240.07 kJ/mol-H2 to 869.07 kJ/mol-H2 and then decreased to 281.68 kJ/mol-H2, which was in accordance with the variation of voltage.

4.

Conclusions

Table 3 e Current efciency and energy consumption at different operating conditions. Run
1 2 3 4 5 6 7 8 9 10 11 12 13 14

ha I (%)
42.46 100.09 100.08 95.25 67.41 76.04 90.09 98.29 99.84 99.98 100.11 100.06 93.61 98.92

hc I (%)
92.77 98.51 92.50 92.60 37.88 46.25 99.86 95.32 99.03 100.25 98.45 100.22 95.90 95.41

Ve (V)
0.21 0.33 0.46 1.67 3.46 3.69 1.08 0.93 1.55 1.45 0.56 0.82 0.48 0.56

UH2 (kJ/mol-H2)
108.92 161.10 237.93 869.07 4412.60 3849.30 523.04 468.32 756.51 699.16 276.84 393.24 240.07 281.68

The Bunsen reaction was successfully carried out using an electrochemical cell featuring a Naon 117 membrane and two graphite electrodes. Both H2SO4 in the anolyte and HI in the catholyte were concentrated during electrolysis. Increase in the current density contributed to elevating the H2SO4 and HI concentration. However, a higher current density also caused corrosion of the anode-side graphite electrode and increased power consumption. As temperature increased from 303 K to 343 K, the highest H2SO4 in anolyte and HI molality in catholyte were respectively obtained at 343 K and 323 K. The H2SO4 molality in anolyte increased with rising initial H2SO4 concentration while the HI molality in catholyte reached its highest value at 13 mol=kgH2 O of H2SO4 concentration. As the initial HI concentration amplied, the HI molality in catholyte increased while the H2SO4 molality in anolyte rst increased and then decreased. The H2SO4 concentration increased with rising I2/HI molar ratio, whereas a higher HI concentration can be obtained at a lower I2/HI molar ratio. Compared with the traditional Bunsen reaction, the electrochemical method generated two acids with high purity even at low excess iodine values. The value of t was close to 1 for most runs except at higher current densities, but an increase in current density reduced the permeation of water (b). The electrode current efciency drastically decreased to less than 80% as the current density augmented to 15 and 25 A/dm2, while the electrode current efciency was close to 100% at other conditions. A lower UH2 can be obtained at lower current densities, and the lowest UH2 of 108.92 kJ/mol-H2 was obtained at a current density of 1 A/ dm2. An increase in temperature, H2SO4 concentration, and HI concentration lessened the value of UH2 . But UH2 increased as the I2/HI molar ratio increased from 0.2 to 0.5 and then decreased for further rises of I2/HI to 0.8.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 4 3 9 1 e1 4 4 0 1

14401

Acknowledgments
This work has been nancially supported by the National Natural Science Foundation of China (51276170). The authors gratefully acknowledge this support.

Nomenclature

E0, j DN0 i I t F n Ej ai pSO2 R T hj t b j DNi Qe j hI UH2 Ve hE

Standard j-electrode potential (vs. standard hydrogen electrode (SHE)), V theoretical moles of i-component formed during electrolysis, mol electric current, A electrolytic time, s Faraday constant, 96,485 C/mol number of electrons required in the half cell reaction, dimensionless j-electrode potential, V activity of i-species in the solution, dimensionless relative pressure of SO2 in the anolyte, dimensionless gas constant, 8.31 J mol1 K1 temperature, K j-electrode overpotential, V transport number of proton, dimensionless water permeability coefcient, dimensionless molar change of i-component in j-electrolyte, mol total amount of electricity, C current efciency of j-electrode, dimensionless heat equivalent of electric energy consumed per mole of H2, kJ/mol-H2 average cell voltage, V efciency for conversion of heat to electricity, dimensionless

references

[1] Norman JH, Besenbruch GE, Brown LC, OKeefe DR, Allen CL. Thermochemical water-splitting cycle, bench-scale investigations, and process engineering. Final report for the period February 1977 through December 1981. GA-A16713; 1982. [2] Sakurai M, Nakajima H, Onuki K, Ikenoya K, Shimizu S. Preliminary process analysis for the closed cycle operation of the iodineesulfur thermochemical hydrogen production process. Int J Hydrogen Energy 1999;24(7):603e12. [3] Sakurai M, Nakajima H, Onuki K, Shimizu S. Investigation of 2 liquid phase separation characteristics on the iodineesulfur thermochemical hydrogen production process. Int J Hydrogen Energy 2000;25(7):605e11. [4] Goldstein S, Borgard JM, Vitart X. Upper bound and best estimate of the efciency of the iodine sulphur cycle. Int J Hydrogen Energy 2005;30(6):619e26.

[5] Kasahara S, Kubo SJ, Onuki K, Nomura M. Thermal efciency evaluation of HI synthesis/concentration procedures in the thermochemical water splitting IS process. Int J Hydrogen Energy 2004;29(6):579e87. [6] Giaconia A, Caputo G, Sau S, Prosini PP, Pozio A, De Francesco M, et al. Survey of Bunsen reaction routes to improve the sulfureiodine thermochemical water-splitting cycle. Int J Hydrogen Energy 2009;34(9):4041e8. [7] Taylor ML, Elder RH, Styring P, Allen RWK. Improved solvation routes for the Bunsen reaction in the sulphur iodine thermochemical cycle: part I e ionic liquids. Int J Hydrogen Energy 2013;38(4):1765e74. [8] Taylor ML, Elder RH, Allen RWK. Improved solvation routes for the Bunsen reaction in the sulphur iodine thermochemical cycle: part II e molecular solvent properties. Int J Hydrogen Energy 2013;38(4):1775e83. [9] Taylor ML, Elder RH, Allen RWK. Improved solvation routes for the Bunsen reaction in the sulphur iodine thermochemical cycle: part III e Bunsen reaction in molecular solvents. Int J Hydrogen Energy 2013;38(4):1784e94. [10] Nomura M, Fujiwara S, Ikenoya K, Kasahara S, Nakajima H, Kubo S, et al. Application of an electrochemical membrane reactor to the thermochemical water splitting IS process for hydrogen production. J Membr Sci 2004;240:221e6. [11] Nomura M, Nakao SI, Okuda H, Fujiwara S, Kasahara S, Ikenoya K, et al. Development of an electrochemical cell for efcient hydrogen production through the IS process. AIChE J 2004;50(8):1991e8. [12] Nomura M, Okuda H, Kasahara S, Nakao SI. Optimization of the process parameters of an electrochemical cell in the IS process. Chem Eng Sci 2005;60(24):7160e7. [13] Immanuel V, Gokul KU, Shukla A. Membrane electrolysis of Bunsen reaction in the iodineesulphur process for hydrogen production. Int J Hydrogen Energy 2012;37(4):3595e601. [14] Gokul KU, Immanuel V, Sant S, Shukla A. Membrane electrolysis for Bunsen reaction of the SI cycle. J Membr Sci 2011;380(1e2):13e20. [15] Immanuel V, Parvatalu D, Bhardwaj A, Prabhu BN, Bhaskarwar AN, Shukla A, et al. Properties of Naon 117 in highly acidic environment of Bunsen reaction of IeS cycle. J Membr Sci 2012;409e410(0):137e44. [16] Immanuel V, Shukla A. Effect of operating variables on performance of membrane electrolysis cell for carrying out Bunsen reaction of IeS cycle. Int J Hydrogen Energy 2012;37(6):4829e42. [17] Tanaka N, Onuki K. Equilibrium potential across cation exchange membrane in HIeI2eH2O solution. J Membr Sci 2010;357(1e2):73e9. [18] Lee BJ, No HC, Yoon HJ, Kim SJ, Kim ES. An optimal operating window for the Bunsen process in the IeS thermochemical cycle. Int J Hydrogen Energy 2008;33(9):2200e10. [19] Onuki K, Hwang GJ, Arifal, Shimizu S. Electro-electrodialysis of hydriodic acid in the presence of iodine at elevated temperature. J Membr Sci 2001;192(1e2):193e9. [20] Yoshida M, Tanaka N, Okuda H, Onuki K. Concentration of HIx solution by electro-electrodialysis using Naon 117 for thermochemical water-splitting IS process. Int J Hydrogen Energy 2008;33(23):6913e20. [21] Chen SZ, Wang RL, Zhang P, Wang LJ, Xu JM, Ke YC. HIx concentration by electro-electrodialysis using stacked cells for thermochemical water-splitting IS process. Int J Hydrogen Energy 2013;38(8):3146e53.