Top Banner
Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting Cycle PI: C. Huang (Florida Solar Energy Center) N. Muradov (Florida Solar Energy Center) A. Raissi (Florida Solar Energy Center) A. Adebiyi (Florida Solar Energy Center) Abstract Solar-driven thermochemical water splitting cycles (TCWSCs) provide an energy efficient and environmentally attractive method for generating hydrogen. Solar-powered TCWSCs utilize both thermal (i.e. high temperature heat) and light (i.e. quantum energy) components of the solar resource, thus boosting the overall solar-to-hydrogen energy conversion efficiency compared to those with heat-only input. At the Florida Solar Energy Center (FSEC), a new solar-powered TCWSC, sulfur dioxide (SO2)/sulfuric acid cycle, is under research and development. FSEC's cycle - a novel hybrid photo- thermochemical sulfur-ammonia (S-A) cycle, is a modification of the well-known Bowman-Westinghouse (B-W) hybrid cycle wherein the electrochemical step is replaced by a photocatalytic process. The main reaction (unique to FSEC's S-A cycle) is the light- induced photocatalytic production of hydrogen and ammonium sulfate from an aqueous ammonium sulfite solution. Ammonium sulfate product is processed to generate oxygen and recover ammonia and SO 2 . Ammonia and sulfur dioxide are then recycled and reacted with water to regenerate ammonium sulfite. Introduction Thermochemical water splitting cycles (TCWSCs) can achieve high overall heat-to- hydrogen energy conversion efficiencies. Presently, the prospective high temperature heat sources suitable for thermochemical process interface include solar thermal concentrator and central receiver systems, and nuclear power plants (i.e. high temperature gas-cooled reactors, HTGR). U.S. DOE's Nuclear Energy Research Initiative program has funded several efforts aimed at hydrogen production using nuclear power. For example, the General Atomics (GA) Corp. has been developing the sulfur- iodine (S-I) cycle for many years. The GA cycle belongs to a group of TCWSCs known as the "sulfur family cycles." The objective of this research is to investigate a new sulfur ammonia cycle, which utilizes solar energy both for heating and for the photocatalytic oxidation process. It is potentially applicable to the production of solar hydrogen through thermochemical processes. The rationale developing a new solar powered TWSPC is to utilize both thermal and quantum energies to increase the solar to hydrogen energy conversion efficiency. For the B-W cycle, the solution pH can have a detrimental effect on the performance. In particular, at low pH, sulfur may form instead of hydrogen. To prevent sulfur formation, one must maintain a relatively high pH, which requires a reduction in acid concentration. However, high pH results in low hydrogen evolution rates. Additionally, low sulfuric acid concentration in the outlet of an electrochemical reactor would require a more energy- intensive and costly acid separation and concentration step. In order to mitigate these problems, a novel sulfur-ammonia (S-A) cycle has been developed at FSEC. In the
41

Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Apr 04, 2020

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting Cycle

PI: C. Huang (Florida Solar Energy Center) N. Muradov (Florida Solar Energy Center)

A. Raissi (Florida Solar Energy Center) A. Adebiyi (Florida Solar Energy Center)

Abstract Solar-driven thermochemical water splitting cycles (TCWSCs) provide an energy efficient and environmentally attractive method for generating hydrogen. Solar-powered TCWSCs utilize both thermal (i.e. high temperature heat) and light (i.e. quantum energy) components of the solar resource, thus boosting the overall solar-to-hydrogen energy conversion efficiency compared to those with heat-only input. At the Florida Solar Energy Center (FSEC), a new solar-powered TCWSC, sulfur dioxide (SO2)/sulfuric acid cycle, is under research and development. FSEC's cycle - a novel hybrid photo-thermochemical sulfur-ammonia (S-A) cycle, is a modification of the well-known Bowman-Westinghouse (B-W) hybrid cycle wherein the electrochemical step is replaced by a photocatalytic process. The main reaction (unique to FSEC's S-A cycle) is the light-induced photocatalytic production of hydrogen and ammonium sulfate from an aqueous ammonium sulfite solution. Ammonium sulfate product is processed to generate oxygen and recover ammonia and SO2. Ammonia and sulfur dioxide are then recycled and reacted with water to regenerate ammonium sulfite.

Introduction Thermochemical water splitting cycles (TCWSCs) can achieve high overall heat-to-hydrogen energy conversion efficiencies. Presently, the prospective high temperature heat sources suitable for thermochemical process interface include solar thermal concentrator and central receiver systems, and nuclear power plants (i.e. high temperature gas-cooled reactors, HTGR). U.S. DOE's Nuclear Energy Research Initiative program has funded several efforts aimed at hydrogen production using nuclear power. For example, the General Atomics (GA) Corp. has been developing the sulfur-iodine (S-I) cycle for many years. The GA cycle belongs to a group of TCWSCs known as the "sulfur family cycles." The objective of this research is to investigate a new sulfur ammonia cycle, which utilizes solar energy both for heating and for the photocatalytic oxidation process. It is potentially applicable to the production of solar hydrogen through thermochemical processes. The rationale developing a new solar powered TWSPC is to utilize both thermal and quantum energies to increase the solar to hydrogen energy conversion efficiency. For the B-W cycle, the solution pH can have a detrimental effect on the performance. In particular, at low pH, sulfur may form instead of hydrogen. To prevent sulfur formation, one must maintain a relatively high pH, which requires a reduction in acid concentration. However, high pH results in low hydrogen evolution rates. Additionally, low sulfuric acid concentration in the outlet of an electrochemical reactor would require a more energy-intensive and costly acid separation and concentration step. In order to mitigate these problems, a novel sulfur-ammonia (S-A) cycle has been developed at FSEC. In the

Page 2: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

FSEC's S-A cycle, the Gibbs energy input to the reaction of sulfite oxidation is via photons (at wavelengths greater than 350 nm), thus making it compatible with a solar power source.

Background Despite the difficulties that challenge efficient electrolytic oxidation of sulfur dioxide, the Westinghouse hybrid sulfur cycle still remains as one of the most studied and promising thermochemical water splitting cycles for the production of hydrogen from water. The cycle is written as follows: SO2(g) + H2O(l) = H2(g) + H2SO4(aq) (electrolysis) 77oC (1) 2H2SO4(g) = 2SO2(g) + 2H2O(g) + O2(g) 850oC (2) The many advantages of the Westinghouse cycle (WC) have been widely reported and discussed in the literature. However, it is also well known that the cycle suffers from the low solubility of sulfur dioxide in water and challenges presented by the acidic environment during SO2 electrolytic oxidation process1. To date, many efforts have been made to improve the efficiency of the electrolytic process for oxidation of sulfur dioxide. Past activities have involved the use of a depolarized electrolyzer as well as the addition of a third process step - examples include S-I, S-Br and S-Fe cycles below: Ispra Mark 13 sulfur-bromine cycle: Br2(l) + SO2(g) + H2O(l) = 2HBr(aq) + H2SO4(aq) 77oC (3) 2H2SO4(g) = 2SO2(g) + 2H2O(g) + O2(g) 850oC (4) 2HBr(aq) = Br2(aq) + H2(g) (electrolysis) 77oC (5) General Atomics' sulfur-iodine cycle: I2 + SO2(g) + H2O(l) = 2HI(a) + H2SO4(aq) 100oC (6) 2H2SO4(g) = 2SO2(g) + 2H2O(g) + O2(g) 850oC (7) 2HI = I2(g) + H2(g) 450oC (8) Sulfur-iron cycle2: 2Fe2(SO4)3(aq)+SO2(g) + 2H2O(l) = 2FeSO4(aq) + 3H2SO4(aq) 25oC (9) 2H2SO4(g) = 2SO2(g) + 2H2O(g) + O2(g) (electrolysis) 850oC (10) 2FeSO4(aq) + 2HSO4(aq) = 2Fe2(SO4)3(aq) + H2(g) 25oC (11) Although these cycles solve some of the WC problems, especially with regard to the solubility of SO2 in water, they have many issues of their own. For example, separation of sulfuric acid from the reaction products such as HI, HBr or FeSO4 presents a challenge. Additionally, the pH factor remains a serious problem. In fact, this problem becomes more acute due to the generation of other acidic species such as HI and HBr. As an alternative to direct electrolysis of aqueous SO2, we have developed a new sulfur-ammonia (S-NH3) cycle3. This cycle was presented at the 15th World Hydrogen Energy Conference (WHEC-15), held in Yokohama, Japan in 2004 and received an Innovative Technology Award from the WHEC-15 scientific organizing committee - the only award given to the researchers from the United States of America.

Page 3: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

(http://dbase.fsec.ucf.edu/pls/operation/press_display?pressid=2158). We have carried out extensive experimental and analytical (i.e. thermodynamic and process flowsheet analyses using AspenPlus™ and HYSYS chemical process simulation platforms) work on this concept and a patent application based on our findings is in the works. The following provides a brief description of the concept. We have modified WC by adding ammonia to reaction (1). The reaction steps of our new cycle can be written as follows: SO2(g) + 2NH3(g) + H2O(l) → (NH4)2SO3(aq) 25oC (chemical absorption) (12) (NH4)2SO3(aq)+H2O → (NH4)2SO4(aq) + H2(g) 80oC (photochemical step) (13) (NH4)2SO4(aq) →2NH3(g) + H2SO4(l) 300oC (thermochemical step) (14) H2SO4(l) → SO2(g) + H2O(g) + 1/2O2(g) 850oC (thermochemical step) (15) Overall Reaction: H2O + solar energy = H2 + 0.5 O2 The main advantages of the S-NH3 are as follows: Reaction (12) produces only one product (ammonium sulfite) so no separation stage is necessary. And it can be viewed as the step for extracting and separating oxygen from SO2. This reaction is more efficient than the conventional O2 separation process used in the WC. Unlike SO2, both (NH4)2SO3 and (NH4)2SO4 have very high solubility (about 6M at 30oC) in water and thus more suitable (than SO2) for the production of hydrogen using a depolarized electrolyzer. The pH of (NH4)2SO3(aq) and (NH4)2SO4(aq) are about 8.50 and 6.00, respectively, indicating that reactions (12) and (13) can proceed in neutral pH. Additionally, (NH4)2SO3 and (NH4)2SO4 can easily decompose at temperatures of 80oC and 300oC, respectively. Furthermore, S-NH3 cycle employs readily available and low cost chemicals. The flow diagram is shown in Figure 1. We have proposed and developed five variations of the reaction (13) according to the form of the energy input into the process (i.e. electrical energy versus photonic energy) and the use of photocatalysts as follows: a. (NH4)2SO3 + H2O + electricity → H2 + (NH4)2SO4

b. (NH4)2SO3 + H2O + photocatalyst + sunlight → H2 + (NH4)2SO4

c. (NH4)2SO3 + H2O + UV light → H2 + (NH4)2SO4

d. (NH4)2SO3 + H2O + UV light + TiO2 → H2 + (NH4)2SO4

e. (NH4)2SO3 + H2O + photoelectrode + sunlight → H2 + (NH4)2SO4

Page 4: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

photocatalytic reactor

absorb

gas-

liquid

NH3 recovery mixunit

NH3

sulfuric acid decomp.

dish

Figure 1. Flow diagram of solar powered S-NH3 thermochemical water splitting cycle

Experimentally we have verified the viability of the reactions (b-d). In particular, we have developed and successfully tested an efficient photocatalyst for the reaction (b) that absorbs in the visible region of the solar spectrum. Further experimental work is still in progress for full characterization of reactions (a) and (e). To address the application of the S-NH3 both on lunar surface and on earth surface, we focus our efforts on the practical realization of the reaction (b), (c) and (d) – photolytic and photocatalytic processes of ammonium sulfite aqueous solution for the production of hydrogen. Potentially, this approach will allow utilization of the ultraviolet and visible light on the lunar surface for directly convert solar photonic energy into hydrogen chemical energy. The solar thermal energy is used for the production of oxygen through the decomposition of sulfuric acid shown in Reaction (15). In short, we propose to combine FSEC developed technology described above with a solar concentrator to develop a state-of-the-art cycle for the cost-effective production of hydrogen from water.

Part I: UV Light Photolytic Hydrogen Production from Ammonium Sulfite Aqueous Solution

Introduction One of unique features of FSEC’s Sulfur-Ammonia (S-A) thermochemical water splitting cycle is that it provides a platform for utilizing a variety of energy sources for the production of hydrogen. Hydrogen production step of the S-A cycle can be interfaced

Page 5: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

with available energy sources including off-peak electricity from the grid, solar and nuclear energy, which makes the cycle more versatile. One of its important applications is the utilization of available UV and visible light on the lunar surface for production of oxygen from lunar soil. The production of hydrogen by irradiating the ammonium sulfite solution with ultra violet light is one of the several alternatives FSEC has proposed to study. The chemical reaction for the hydrogen production step using ultra violet light is depicted in equation 2.1 below. c. (NH4)2SO3 + H2O + UV light = H2 + (NH4)2SO4 Although UV light accounts for only ~4% of solar energy on earth, its portion on the lunar surface will be dramatically increased. Additionally, there are several reasons that necessitate the study for the hydrogen production from UV light photolysis. One major reason for this study is that due to the high energy present in photons of ultraviolet light, there is no need for the use of a semiconductor photocatalyst; this will in turn eliminate a potentially difficult separation step required to remove photocatalyst powders from the reaction solution. The production of O2 on the moon utilizing indigenous materials and solar energy will be essential for successful lunar colonization. Several processes have been put forth to accomplish lunar O2 generation. Among these technologies the H2 reduction of ilmenite (FeTiO3) has received the most study to date, and is considered a feasible and promising scheme. The process includes the following two reactions: FeTiO3 + H2 + electricity/heat = Fe + TiO2 + H2O (16) H2O + electricity = H2 + 0.5O2 (17) Overall reaction: FeTiO3 + electricity = Fe + TiO2 + 0.5O2 Reaction (16) requires 700 to 900 oC to achieve suitable reaction rates and acceptable conversion efficiencies. H2O produced from Reaction (16) is electrolyzed into H2 and O2 and H2 is recycled in the reduction process. The critical issue involved in the lunar O2 production using solar energy is that the process requires an intermediate step to convert solar energy into electrical energy with efficiency about 15%. Although ilmenite reduction is a highly endothermic reaction and the degradation of electrical energy into thermal heat results in an exergy loss, water splitting through electrolysis requires higher amount of electrical energy with about 10% of solar to H2 efficiency. Therefore, increasing the efficiency of lunar O2 production demands eliminating the intermediate step. Direct water splitting via solar powered thermal chemical cycles can approximate 40 to 50% efficiency, which can be further enhanced by the higher level of available UV photonic energy in the solar spectra on the lunar surface. The FSEC’s S-NH3 cycle is a purely solar powered thermochemical water splitting process for the production of H2 and O2. The unique feature of the cycle is that solar photonic energy is used for the production of H2 while solar thermal energy is used for the production of O2. Integrating ilmenite reduction into the process constitutes a direct solar powered O2 production as follows: FeTiO3 + H2 = Fe + TiO2 + H2O Solar thermal energy 800 oC (18) SO2 + 2NH3 + H2O → (NH4)2SO3 Chemical absorption 25oC (19) (NH4)2SO3+H2O → (NH4)2SO4 + H2 Solar photonic energy 80oC (20)

Page 6: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

(NH4)2SO4 →2NH3 + SO2 + H2O + 0.5O2 Solar heat 850 oC (21) Overall reaction: FeTiO3 + solar energy = Fe + TiO2 + 0.5O2 In a preliminary step, we used a UV light to simulate UV light spectra on the lunar surface and to investigate hydrogen production under UV light. The objective of this research is to investigate the efficiency enhancement of photolytic hydrogen production from ammonium sulfite aqueous solution in the presence of UV light, (Process (c)). The success of the project will promote Research & Development of innovative lightweight solar concentrators applicable for in lunar environment. In this project, a low-pressure mercury lamp is used as a UV light source for the photolytic oxidation of ammonium sulfite without requiring a photocatalyst. Main advantages of this process are: 1) no catalysts or electrodes are needed to affect the reaction, 2) no separation step to remove the catalyst from the solution is required, 3) process is simple and no costly reagents are needed, 4) UV lamps are inexpensive and have a very long lifetime, 5) configuration of the photoreactor using UV lamps is simple, and 6) approach can potentially be used in space or on the moon where UV light is readily available. The results indicate that sulfite ions in aqueous solution could be photo-oxidized to form sulfate ions while water is reduce to hydrogen under UV irradiation. The mechanism for UV photolytic hydrogen production from aqueous solutions of ammonium sulfite involves two reaction pathways as follows: 1) sulfite directly oxidizes to form sulfate, and 2) sulfite oxidizes to generate dithionate, and the dithionate further reacts to form sulfate. The first route is the dominant reaction pathway. Experimental Reagent grade ammonium sulfite ((NH4)2SO3H2O, Aldrich) and sodium dithionate (Na2S2O6, Pfaltz & Bauer) were used for the photolytic reactions. UV-Vis measurement was performed with a UV-Vis light spectrophotometer (Shimadzu UV-2401). A 60 W germicidal lamp (Atlantic Ultraviolet Company) was applied as a light source. The spectral output of the light source is mainly at 254 nm. Figure 2 depicts a continuous flow system for the ammonium sulfite photolytic reactions. The total volume of solution used for each run was 1500 mL. Prior to irradiation, the system was purged with instrumental grade argon gas for approximately 2.5 hrs. pH measurements were carried out with an Oakton 1100 pH meter under carefully calibrated conditions.

Page 7: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Figure 2. A cyclic reaction system for the photolytic

decomposition of ammonium sulfite Results and Discussion Section 1.1. Kinetics of ammonium sulfite oxidation -- A kinetic study was carried out on the ammonium sulfite solution. A 0.0432M ammonium sulfite solution was subjected to UV light irradiation to produce hydrogen. Hydrogen evolution was measured volumetrically as shown in Figure 2 and the purity of the gas generated was determined by a gas chromatograph equipped with a 5 Å molecular sieve packed column (Shimadzu 14B). The presence of sulfite, sulfate and dithionate ions was determined by a High Pressure Liquid Chromatograph (Dionex DX-500). The experiment was carried out to determine the mechanism of the reaction that leads to the production of hydrogen and carry out a material balance on the reaction. Figure 3 shows the hydrogen production after 1400 minutes of irradiation. The initial production rate of hydrogen is relatively high at approximately 2 mL per minute; however, after 500 minutes of irradiation, the reaction rate slows to less than half the initial rate and finally comes to a halt after 1300 minutes. The reaction mechanism for the curve is represented as follows: Irradiation: (22) *2

323

−− →+ SOhvSOOxidation: (23) −−−− ++→+ eOHSOOHSO 22 2

24

*23

Reduction: (24) −− +→+ OHHeOH 222 22

Total: (25) 2242

23 HSOOHSO +→+ −−

Page 8: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0

100

200

300

400

500

600

700

800

900

1000

0 200 400 600 800 1000 1200 1400

Time (min)

Hyd

roge

n (m

L)

0

1

2

3

4

5

6

7

8

9

pH

Hydrogen

pH

Figure 3. Kinetic Curve for hydrogen production from the UV light irradiation of 0.0432 M ammonium sulfite (pH = 7.721, T = 50 oC,

Total Conc.= 0.0432 M, solution V=1500 mL)

Initially, the sulfite ions present in the ammonium sulfite solution are excited by ultra-violet light to form sulfite ion radicals (SO3

2-*). The radicals are then hydrolyzed with hydroxyl ions forming sulfate ions, water and two electrons. The electrons then combine with H+ ions present in H2O to evolve hydrogen gas and hydroxyl ions, which react with sulfite ions to close the cycle. Figures 4 and 5 show changes of sulfite and sulfate ions before and after reaction. The ammonium sulfite solution initially contains both sulfite and sulfate ions at a ratio of 60:40. The presence of sulfate ions in the ammonium sulfite solution is attributed to the oxidation of the ammonium sulfite. After the reaction, sulfite ions are completely consumed and converted into sulfate ions as shown in Figure 5.

Page 9: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00-1.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0 µS

min

6-9-05 Standardization Runs SO3 DX-500 #16 ECD_1

1 - 0

.913

2 - S

ULF

ITE

- 2.9

27

3 - S

ULF

ATE

- 3.

987

Figure 4: HPLC spectra of ammonium sulfite and sulfate species before photolytic reaction (pH = 7.721, T=50 oC, Total Conc.=0.0432 M)

0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00-1.0

2.0

4.0

6.0

8.0

10.0

12.0 6-9-05 Standardization Runs SO3 DX-500 #18 ECD_1µS

min

1 - 0

.903

2 - S

ULF

ATE

- 4.

000

Figure 5: HPLC results of ammonium sulfite and sulfate species photolysis (pH =

7.721, T=50 oC, Total Conc.=0.0432 M, Reaction t = 21 hr) In parallel to the production of sulfate ions, two sulfite ions can also combine to form a dimer called dithionate. The dithionate ion is an intermediate of the reaction and forms at a slower rate than sulfate ions. Huang1 proposed the following mechanism for dithionate production from sulfite ions under UV light irradiation.

Page 10: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Oxidation: 2SO3

2- S2O62- + 2e- (26)

Reduction: 2H2O + 2e- H2 + 2OH- (27) Overall: 2SO3

2- + 2H2O S2O62- + H2 + 2OH- (28)

The dithionate ion can be oxidized to produce sulfate ions resulting in the evolution of hydrogen. The mechanism proposed by Huang4 for the oxidation of dithionate is shown as follows: Oxidation: (29) −+−−− ++→+ eHSOOHOS 2422 2

4*2

62

Reduction: (30) −− +→+ OHHeOH 222 22

Overall: (31) +−− ++→+ HHSOOHOS 222 2242

262

HPLC measurements were carried out on a sulfite sample after 8.5 hours’ irradiation of UV-light and the results are shown in Figure 6. After 8.5 hours, a dithionate ion peak was detected.

0.0 1.3 2.5 3.8 5.0 6.3 7.5 8.8 10.0 12.0-10.0

0.0

12.5

25.0

37.5

50.0

70.0 6-3-05 STANDARDIZATION RUNS S2O6 DX-50 ECD_1µS

min

1 - 0

.793

2 - 0

.897

3 - D

ITH

ION

ATE

- 8.

927

Figure 6. HPLC spectra of ammonium sulfite and sulfate species after 8.5 hrs of photolytic reaction (pH = 7.721, T=50 oC, Total Conc.=0.0432 M)

Figure 7 depicts that after 21 hours of UV light irradiation, the dithionate ion peak was no longer detectable, suggesting that dithionate ions were totally converted into sulfate ions. These results are in accordance with the mechanism proposed in Reactions (29 to 30).

Page 11: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0.0 1.3 2.5 3.8 5.0 6.3 7.5 8.8 10.0 12.0-10

20

40

60

80

100

120 6-3-05 STANDARDIZATION RUNS S2O6 DX-500 # ECD_1µS

min

1 - 0

.793

2 - 0

.913

Figure 7. HPLC spectra of ammonium sulfite solution after 21 hours of UV light irradiation, no dithionate peak is detectable (pH=7.721, T=50 oC, Total

Conc.=0.0432 M) To further verify this mechanism, an experiment was carried out using a 0.025M sodium dithionate aqueous solution as a starting reactant and underwent UV light irradiation to produce hydrogen. In this experiment, the reaction intermediate, sodium dithionate, was used to replace ammonium dithionate because ammonium dithionate could not be purchased. It was assumed that there was little or no cationic involvement during UV light photolytic processes. The reaction was conducted over a period of 3 hours with 20 mL of hydrogen evolving from the solution. HPLC analysis was conducted on the sample before and after the experiment to ascertain the products in solution. It is found that dithionate ions were completely converted into sulfite and sulfate ions. Based on this result a detailed mechanism of the reaction can then be formulated as shown in Figure 8. Ammonium sulfite under the irradiation of UV light is oxidized to produce ammonium sulfate while water is reduced to produce hydrogen. This is the fundamental rout of the reaction with higher kinetic rate constant k1. The sulfite ions can also polymerize to form dithionate ions and in the process hydrogen is also produced. The rate constant of hydrogen production during this process (k2) is unknown because the process is mixed with the principle reaction. But from the amount of dithionate detected, k2 should be less than k1. The dithionate ions readily produce sulfate ions and hydrogen under uv-light irradiation. The results shown above indicate that the rate constant k1 is much higher than k2, k3 and k4. Additionally, sulfite ions can be completely converted into sulfate ions in the course of UV light photolytic process.

Page 12: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

SO32- + hv SO42- + H2

S2O62- + hv + H2

k1

k2 k3

k4

H2O +

Figure 8. Reaction mechanism for UV-light irradiation of ammonium sulfite (k1 >> k2, k3, k4)

To further verify the mechanism discussed above, a material balance was conducted on the reaction to determine if the system could attain a balance. A 0.0432M ammonium sulfite solution was used for the reaction. The 0.0432M represents the apparent concentration of the solution based on the mass of the salt (8.7167g) measured for the reaction, however, this does not represent the actual concentration due to the presence of sulfate impurities in the sulfite sample as shown using HPLC analysis. A more precise value of concentration can be achieved from HPLC analysis as was done for this experiment. The reaction was run until no visible production of hydrogen occurred, signifying the end of the reaction. Liquid samples of the reaction solution before and after the reaction are subjected to HPLC analyses. Table 1 shows the results of the material balance. The HPLC analysis shows that the sulfite ions inlet into the reaction was consumed fully and the only species remaining in the solution was sulfate ions. Further HPLC analysis carried out during the course of the reaction show the formation of dithionate ions, however, no dithionate ions were detected at the end of the reaction, indicating that the dithionate ions formed were totally consumed as the reaction progressed.

Table 1. Material balance on 0.0432M ammonium sulfite irradiated for 21 hours

Component Initial (mmol) Final (mmol) Produced (mmol)

Consumed (mmol)

SO32- 40.61 0.00 0.00 40.61

SO42- 14.43 54.66 40.23 0.00

>1% difference

H2

Expected H2 (mL) Actual H2 (mL) Difference (mL) % Difference

986.9 909.0 77.9 7.9 Section 1.2. Rate dependency on ammonium sulfite concentration -- Hydrogen formation rates at different ammonium sulfite concentrations were studied. Figure 9 reveals that the hydrogen formation rate remains practically the same for all concentrations after 8 hrs of irradiation. However, Figure 9 also shows that solutions with lower concentrations have a slightly higher initial rate, and as time goes on the rates begin to level out because of the consumption of ammonium sulfite.

Page 13: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0

100

200

300

400

500

600

700

800

0 50 100 150 200 250 300 350 400 450 500

Reaction Time (min)

Hyd

roge

n (m

L)0.0216M

0.0432M

0.216M

0.432M

pH = 7.8 Temperature = 35oC

Solution Volume = 1500mL

Figure 9. Hydrogen formation rates at varying ammonium sulfite concentrations

Section 1.3. Rate dependency on solution’s pH -- The pH of the reacting solution is one of the important parameters that influence the hydrogen evolution rate. Increasing or decreasing the pH of the solution by introducing H+ ions or OH- ions may have a significant effect on hydrogen evolution rates. This study was carried out to determine an optimal pH range for maximizing the rate of hydrogen production. Figure 10 shows that the solution with a pH of approximately 7.8 had the highest hydrogen production rate. However, as the pH was reduced to a value of 6.0, the rate of reaction slowed down tremendously and eventually stopped after only 200 minutes. The initial pH of the ammonium sulfite solution (0.0216 M) is between 7.8 – 7.9, and in this range the percentage of sulfur species in the form of sulfite ions is approximately 50% - 60% as illustrated 1n Figure 11. The remainder of the sulfur species is in the form of HSO3

- ions. At a pH range of 9 and above almost 100% of any sulfur species are in the form of sulfite ions (SO3

2-). When a solution’s pH is greater than 9.0, all the species present in the solution are sulfite ions; the expectancy is the amount of sulfite ions oxidized to sulfate ions will be benefited, however, the high pH value inhibits water reduction. In a case where water reduction is a rate-determining step a higher pH will slow down the reaction rate; the rate of sulfite ions being converted to sulfate ions should increase, but the rate of production of hydrogen will be inhibited. As depicted in Figure 2.3.1, the rate of reaction is reduced as the pH value is increased to a value of 9 and above leading to the conclusion that the rate determining step for the sulfite photochemical oxidation is the water reduction step. One explanation for the dependency of hydrogen production rate on solution’s pH is that the potential required to generate hydrogen varies with solution pH. Direct water splitting requires a potential of 1.23 volts. The potential required to oxidize sulfite ions to sulfate ions is 0.92 volts under standard conditions. However, the oxidation of bisulfite

Page 14: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

ions (HSO32)- requires approximately 0.1175 volts. As mentioned earlier, at a pH of 7.8

the amount of sulfite ions and bisulfite ions present in the ammonium sulfite solution are virtually equal, so the average potential of this solution is about 0.5 volts as depicted in Figure 12. If the pH of the ammonium sulfite solution is lower than 6, virtually all the sulfur species present in the solution are bisulfite ions.

0

100

200

300

400

500

600

0 50 100 150 200 250 300

Time (min)

Hyd

roge

n (m

L)

pH 6.04pH 7.8pH 9.007pH 10.1pH 11

Figure 10. Hydrogen formation rates at varying pH levels

(conc.=0.0216 M, T= 35 oC, solution volume = 1500 mL)

Page 15: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0.000

0.005

0.010

0.015

0.020

0.025

1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0pH

Co

nce

ntr

ati

on

(M

)

0.030

H2SO3HSO3-SO32-

Figure 11. SO3

2-, HSO3- and H2SO3

concentrations vs. pH (0.025 M Na2SO3 solution)

0.0

5.0

10.0

15.0

20.0

25.0

0.00 0.25 0.50 0.75 1.00 1.25

Voltage (V)

Cur

rent

(mA

)

Figure 12. Electric potential required generating hydrogen from ammonium sulfite solution

Hydrogen production rate for the solution at pH 6 is substantially less than the maximum rate achieved with a solution of pH 7.8 (Figure 10); this is because bisulfite ions can be oxidized to produce elemental sulfur. The solution with initial pH of 6 is seen to produce hydrogen for the first few minutes, but the rate of reaction decreases rapidly, accompanied by a sharp decrease in the pH of the solution as indicated in Figure 13. Elemental sulfur is observed at this stage of reaction.

Page 16: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

3.000

3.500

4.000

4.500

5.000

5.500

6.000

6.500

7.000

0 50 100 150 200 250 300 350

Time (min)`

pH

Figure 13. pH change of ammonium sulfite solution (Initial pH = 6.000) during the UV light photolytic reaction

Section 1.4. Rate dependency on reaction temperature: Temperature is a very important parameter in determining hydrogen production rate. The result of inputting thermal energy to the system can be measured by the system’s temperature changes. If higher solution temperatures can increase the hydrogen production rate, it indicates that thermal energy can be employed to enhance the hydrogen production rate during a UV light photolytic process. Hydrogen formation rates at three different temperatures were carried out and the results are shown in Figure 14.

0

100

200

300

400

500

600

700

800

900

0 100 200 300 400 500

Time (min)

Hyd

roge

n (m

L)

35 C

50 C

75 C

Figure 14. Hydrogen formation rates at varying temperatures (Initial conc. = 0.0432 M, solution volume = 1500 mL)

Page 17: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

It is indicated in Figure 14 that as temperature increases from 35oC to 50oC there is a significant increase in the hydrogen production rate. However, as temperature further increases from 50oC to 75oC the rate of hydrogen production decreases. The initial increase in production rate between the temperatures of 35oC to 50oC can be attributed to the fact that the reaction is endothermic and the increase in temperature enhances the forward reaction. However, at a temperature of 75oC, the solution’s pH plays a more important role than temperature. As the temperature is increased to 75oC, the pH of the solution decreases from an initial value at room temperature of 8.080 to a final value of 7.534. As discussed above, the increase of solution’s pH has an adverse effect on hydrogen evolution rate. An experiment was carried out to test the relationship between the solution’s pH and the temperature. The results are shown in Figure 15. Ammonium sulfite solution was prepared in concentrations of 0.5M and 0.05M. The two solutions were heated at a steady rate over a period of time, and the pH values for each solution were measured by a pH meter (Oakton 1100 pH meter).

6.000

6.500

7.000

7.500

8.000

8.500

20.0 30.0 40.0 50.0 60.0 70.0 80.0

Temperature(oC)

pH

0.05M0.5M

Figure 15. Dependence of ammonium sulfite solution’s pH on its temperature The rate dependence on pH experiments (Figure 10) indicates that there is an optimal range (7.8-8.0) of pH for hydrogen production rate. If the pH value of the solution exists either below or above this range, the production rate of hydrogen will decrease; the results are shown in Figure 10. In the case studied in Figure 14, the pH value of 7.534 is slightly below the optimal pH range as outlined earlier, so the rate of hydrogen production is reduced. As shown in Figure 15, pH is dependent on the temperature, and hydrogen production rate depends on pH, so when considering temperature effect one should take into account the dependency of pH on temperature to optimize pH and temperature ranges. Another important factor is related to the light absorption of the photoreactor material and lamp glass. When temperature increases, the absorption of UV light penetration through the lamp glass and photoreactor will increase, hence, the

Page 18: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

total available UV light photons absorbed by the solution will be reduced resulting in hydrogen production rate decreases. In our future work, we will measure light intensity measurement at different temperatures to verify this assumption and find an optimal temperature for operating the photolytic reaction. Conclusions Aqueous ammonium sulfite solution was irradiated under a 60-W low-pressure mercury ultraviolet light. The light spectrum is mainly in a wavelength of 254 nm. It was observed that highly efficient hydrogen could be produced from an ammonium sulfite aqueous solution with the absence of a semiconductor photocatalyst. The highest rate of production attained for hydrogen was 1.7 mL/min under standard conditions. High Pressure Liquid Chromatograph (HPLC) analysis indicated that the final product of the photochemical reaction was ammonium sulfate. Ammonium dithionate was an intermediate of the reaction and was eventually converted into ammonium sulfate. The concentration of the ammonium sulfite solution did not have a significant effect on the rate of hydrogen production. This may be primarily due to the fact that the photon number available for the photolysis of ammonium sulfite solution was a limiting factor for the current experimental set-up. The number of photons generated by the 60-W germicidal UV lamp can only generate enough number of photons to produce certain amount of hydrogen. Therefore, in this experiment, the change of solution’s concentration did not affect hydrogen production rate. Mass transfer in this tubular photoreactor also played a role in preventing the high concentration ammonium sulfite solutions from generating high rates of evolution for hydrogen. The solution circulation rate was controlled so that the flow in the reaction (Figure 2) was a laminar flow. As the concentration of the solution increased, the solution’s viscosity also increased and in the process, reduced the mass transfer rate in the reactor. The slow mass transfer rate is also one factor in explaining the reason that the hydrogen production rate is a weak function of the concentration of ammonium sulfite solution. The pH of the ammonium sulfite solution proved to have a significant effect on the rate of hydrogen production. The solution’s optimal pH range for hydrogen production was between 7.8 and 8.0; at that pH range the amount of sulfite ions and bisulfite ions present in the aqueous solutions are equal, which in turn leads to the reduction of the potential required by the solution. Higher pH solutions contain predominantly sulfite ions whose potential for oxidation to sulfate is higher than that of lower pH solutions. Hence, from an energy requirement point of view, the hydrogen production rate for a solution with pH ranging between 7.8 and 8.0 would be favorable. The pH lowering than that optimal range would lead to the production of elemental sulfur, which would result in significantly reducing the rate of hydrogen evolution. Temperature of the ammonium sulfite also had an effect on the rate of production of hydrogen. As temperature increased from 35oC to 50oC, the rate of hydrogen production also increased. The rate increase as a result of temperature increase can be attributed to the increase of kinetic rate constant and pH changes. However, as temperature increased further to 75oC, the rate of hydrogen production dropped slightly in comparison with that at temperature of 50 oC. The reason for the rate decrease is partially due to the change in pH caused by temperature change. The other possible reason is that higher temperature will increase the UV light absorption by the UV lamp glass and photoreactor wall - quartz.

Page 19: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Part II: Visible Light Photocatalytic Hydrogen Production

from Ammonium Sulfite Aqueous Solution Introduction The term of photocatalysis consists of the combination of photochemistry and catalysis and implies that light and a catalyst are necessary to bring about or to accelerate a chemical transformation5. This definition includes photosensitization, a process by which photochemical alteration occurs in one molecular entity as a result of initial absorption of radiation by another molecular entity called a photsensitizer, but it excludes the photoacceleration of a stoichiometric thermal reaction irrespective of whether it occurs in homogenous solution or at the surface of an illuminated electrode5. Most photocatalysts used in photocatalysis are semiconductors. Semiconductors are particularly useful for such applications because of a favorable combination of electronic structure, light absorption properties, charge transport characteristics, and excited-state lifetimes. Semiconductor particles such as TiO2, provide very high oxidative power upon illumination with light. Thereby they can oxidize even very stable organic compounds all the way to CO2 in air. As shown in Figure 2.1, when absorbing a photon with energy greater than its band gap energy, an electron is excited from a semiconductor’s valence band to its conductive band. The electron is capable of reducing species A and as a result of this there remains a positive charge (defect electron or hole) in the valence band capable of oxidizing species D 6. In the UV light photolytic experimental section, it was observed that hydrogen production occurred without the use of a semiconductor photocatalyst. In the case of visible light photochemical processes, photocatalysts are needed to speed up the hydrogen evolution rate. The visible light spectrum ranges from about 400 nm – 700 nm. Ammonium sulfite is a clear solution that does not absorb light in the visible spectrum range. In order to oxidize sulfite ions, photocatalyst powders are suspended in the solution to absorb visible light and generate electron and hole pairs that are used for sulfite ion oxidation and water reduction. Semiconductors under irradiation of light energy higher than their band gap energy (Eg), are capable of generating electron and hole pairs. If Eg is greater than the energy required for water splitting, the electron hole pairs are capable of decomposing water into hydrogen and oxygen. The theoretical minimum potential needed for water decomposition is 1.23 V. However, the potential for the oxidation of sulfite ions is 0.92 V, lower than that for water splitting. Furthermore, this potential can be reduced through adjusting the solution’s pH. For example, as discussed in the previous chapter; when the pH of ammonium sulfite solution is 7.8, the potential is reduced to 0.50 V. Low potential requirement makes the oxidation easier than direct water splitting. For an ideal semiconductor photocatalyst, conduction band potential level should be well above (more negative than) the hydrogen reduction level and valence band edge should be well below (more positive than) the water oxidation level for an efficient production of hydrogen and oxygen from water by photocatalysis7. As mentioned earlier, since aqueous ammonium sulfite has a lower minimum photovolatge than that for water splitting, the valence band edge of a semiconductor photocatalyst to be used for this process only needs to be well below the sulfite oxidation

Page 20: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

level in order to oxidize sulfite ions to sulfate ions. Figure 2.2 shows the conduction and valence band edges for a number of semiconductors.

Figure 2.1. Energy scheme of a semiconductor particle in which an electron is excited by light absorption. A + e- = A* -, D + h+ = D*+ 6

Page 21: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Figure 2.2. The electrochemical energy position of the valence and conduction band edges for several semiconductors with aqueous electrolyte

Experimental System Set-up -- Photocatalytic hydrogen production from ammonium sulfite solution was carried out using a 1000-Watt Xenon arc lamp (Spectral Energy) as a solar simulator providing a visible light source. A water tank was used to filter out infrared radiation as shown in Figure 2.3.

Figure 2.3. Experimental set-up for the production of hydrogen via photochemical

oxidation of ammonium sulfite aqueous solution A custom-made borosilicate glass reactor 250 mL in volume was used as the reaction vessel. The reaction temperature was controlled by a water circulation jacket paired with a Polyscience MDL 612 Analog Recirculator. The reaction solution was stirred by a magnetic stirrer. The glass reactor can be seen in Figure 3.4. A Pyrex glass window was used for the reactor where most of UV light portion from the xenon lamp radiation is

Page 22: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

absorbed. Since the red portion of xenon light with wavelength greater than 800 nm will not affect the photoreaction, the light source applied in this experimental work can be considered an ideal solar light simulator in comparison to the solar spectra.

Figure 2.4. Photoreactor configuration

ydrogen was collected in a 1000 ml Pyrex glass-measuring cylinder. On line pH

ght Intensity Measurement -- The main purpose of the light measurement is to

Hmeasurements were carried out with an Oakton 1100 pH meter under carefully calibrated conditions. Prior to irradiation, the reactor was purged with Argon gas for two hours to avoid oxygen-involved in reactions. UV-Visible light spectral analysis was carried out using a SHIMADZU UV-2401 spectrophotometer. Reagent ammonium sulfite ((NH4)2SO3.H2O, Aldrich) was used as purchased without further purification. Reagent grade cadmium sulfide (99.999% purity, Aldrich) was used as a photocatalyst. Pt2Cl3*3H2O (Aldrich) was used as purchased for preparing Pt loaded photocatalyst, Pt/CdS. Lidetermine the amount of light energy incident on the photoreactor used to facilitate hydrogen production. The light intensity measured can be used for determining solar energy to hydrogen chemical energy efficiency of the system. In this study, we used a radiometer and a pyranometer to measure light intensity. The Radiometer can measure xenon light irradiance as a function of its wavelength. The measurement was carried out using a well-calibrated LI-1800 portable spectroradiometer. The light meter was placed 25 inches away from the light source and the light irradiance distribution of the xenon light source is given in Figure 2.5. The light measurements included irradiance with and without a Pyrex glass window across the detector. The reason for including a Pyrex glass window was to filter out wavelengths UV light portion in the spectrum to simulate solar slight radiation because solar light consists about 4% of UV light, which is less than the amount in a xenon arc light source. After filtration, the spectra of both the xenon arc lamp and solar light are similar.

Page 23: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

20.00

300 400 500 600 700 800 900 1000 1100

Wavelength (nm)

Spec

tral

Irra

dian

ce (W

/m2 /n

m)

ArcArc/Glass

Figure 2.5. Spectral irradiance for xenon short arc lamp (1000 W) The light intensity (Watt) within wavelength ranges was calculated by integrating areas under the irradiance curve. The results obtained from the radiometer measurements are shown in Table 2.1. The detailed process can be described as follows: Utilizing the spectral irradiance distribution plot (Figure 2.5), segments of wavelength ranges were cut out and weighed to determine the ratios between the different wavelength ranges. The weight of each wavelength range was translated to an intensity value and then used to calculate the overall radiation energy for each wavelength range. The pyranometer used in this case was a LICOR LI-200 pyranomter (LICOR). The pyranometer was placed at distances of 12 inches and 25 inches from the xenon light source. The average light intensity was taken at 9 different positions on the elliptical shape formed by the light impression. The total solar energy incident on the reactor was calculated as 25.07 Watt. However, the light intensity ranging from 300 nm to 450 nm is a most important parameter for this experiment; this is because the photocatalyst employed in this study is a cadmium sulfide semiconductor with a band gap of 2.4 eV, responding to a light wavelength of 516 nm or below. The total light radiation energy within the wavelength of 300 nm – 450 nm was determined as 5.20 Watt; this value will be used to calculate both the energy efficiency (converting solar photonic energy to hydrogen chemical energy) and the quantum efficiency of the photocatalytic process.

Table 2.1. Light Intensity Calculation at Different Wavelength Ranges Wavelength (nm)

Weight (once)

Light Intensity (W/m2)

Intensity Average (W/m2)

Power (W)

300-400 0.0072 96.91 59.89 0.804 350-400 0.00158 212.66 131.42 1.766 400-450 0.00235 316.30 195.47 2.627 450-500 0.003065 412.54 254.95 3.426

Page 24: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

500-600 0.005425 730.19 451.26 6.065 600-700 0.005225 703.27 434.62 5.841 700-800 0.004055 545.79 337.30 4.533 Total 0.02242 3017.66 1864.91 25.066 Visible Spectral Analysis -- A UV visible spectral analysis was conducted using a spectral meter (SHIMADZU UV-2401) on six different samples to determine the composition of the reacted solution and if at all ammonium dithionate, a dimer of ammonium sulfite was produced during the course of the reaction. The sample solutions used for the analysis were, 0.5M Na2SO3, 0.5M Na2SO4, 0.5M Na2S2O6 (dithionate), 0.5M (NH4)2SO3, 0.5M (NH4)2SO4, and the reacted solution after 64 hours of irradiation. It was observed that there was virtually no cation effect on the absorbance of the solutions. Both ammonium sulfite and sodium sulfite solutions had almost identical UV-Vis light spectra. Figure 3.10 depicts that the peak for the reacted solution falls between the peaks of sulfite ions and dithionate. It may suggest that that after 64 hours irradiation, the reacted solution contains a mixture of sulfate, sulfite and dithionate ions. The quantity analysis was carried out with a HPLC measurement.

Figure 2.7. UV-Visible spectra of sulfate, sulfite and dithionate ions

esults and Discussion

reliminary Phase, 64 hours irradiation, Photocatalyst Lifespan Test -- This

(2.1)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

200 225 250 275 300 325 350Wavelength (nm)

Abs

orba

nce

SO32-

SO42-

S2O62-

Reacted SolutionSO3

2- S2O62- SO3

2-

0.5M solutions

R Pexperiment was carried out using a 0.5 g Pt/CdS suspended in 250 mL 0.5M solution of ammonium sulfite solution. The Pt loading was 1.0 wt.% of 0.5g CdS catalyst. The purpose of the experiment was to determine the stability of the cadmium sulfide catalyst under extended irradiation from the xenon lamp light source. Cadmium sulfide has been reported to be photocorrosive and undergo the following decomposition reaction when exposed to light 10. SCdCdSh +→+ ++ 22

Page 25: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Photoplatinization of CdS particles reduces the photocorrosion and the use of nafion and clay also prevents this effect to some extent 11. With the presence of sulfite ions, the oxidation of SO3

2- to SO42- competes with Reaction (3.2.1). If the oxidation of sulfite ions

is easier than the photo-decomposition of cadmium sulfide, the photocatalyst Pt/CdS would be stable under irradiation. The experiment was run on and off for a total of 9 days, approximately 7 hours per day. Figure 2.8 shows hydrogen production versus time for the 9-day period. The hydrogen production rate has substantially increased in comparison with that of non-Pt doped CdS catalyst. The hydrogen production rate from days 1 thru 4 is not linear compared to the production rate from days 5 thru 7. There is evidence of a steep increase in production rate after day 4. The inconsistencies in the hydrogen production rate is a result of the fact that the reaction system was turned on and off over the 9 day period hence the pH value varied at beginning of each day due to the reaction extent, causing the reaction rate to change as time progressed. Figure 2.9 shows how the pH of the solution changed from day to day. More detailed kinetic study is shown in the following sections. A total of 2.5 L of hydrogen was generated over the total 64 hours. There is no significant deactivation of Pt/CdS photocatalyst. The HPLC analysis on the reacted solution shows that all ammonium sulfite was totally converted into ammonium sulfate.

0

500

1000

1500

2000

2500

3000

0 500 1000 1500 2000 2500 3000 3500 4000

Time(min)

Volu

me(

mL)

0.00

22.32

44.64

66.96

89.28

111.60

133.92

Hyd

roge

n (m

mol

)

Day 1

2.6hrs

Day 2

7.98 hrs

Day 3

7.48 hrsDay 4

7.16 hrs

Day 5

8.21 hrs

Day 6

8.16 hrs

Day 7

8.15 hrs

Day 8

8.00 hrs

Day 9

7.18 hrs

Figure 2.8. Catalyst lifespan for hydrogen production from photocatalysis of 0.5M ammonium sulfite solutions

Page 26: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

7

7.2

7.4

7.6

7.8

8

8.2

8.4

8.6

0 500 1000 1500 2000 2500 3000 3500 4000

Time (min)

pHDay 2

7.98 hrs

Day 3

7.48 hrs

Day 4

7.16 hrs

Day 5

8.21 hrs

Day 6

8.16 hrs

Day 7

8.15 hrs

Day 8

8.00 hrs

Day 9

7.18 hrs

Figure 2.9. pH change versus time for catalyst lifespan run over 9day period Temperature dependence of Hydrogen production -- In the UV light photolysis, it has been shown that the temperature has a significant effect on the hydrogen production rate. The major difference between the visible light photocatalysis and the ultra-violet light photolysis is that the former requires a photocatalyst. Platinized cadmium sulfide is a typical photocatalyst applicable in the visible light range. As temperature was increased from 50oC to 70oC, the rate of hydrogen production also increased significantly as shown in Figure 10. However, as temperature was further increased from 70oC to 90oC the rate of hydrogen production dropped significantly. These results may seem to infer that the reaction is not favored at high temperatures. However; similarly to the temperature dependency experiments for the ultra-violet light photolysis, the effect on temperature on pH must be considered. Increasing the solution’s temperature to 90oC will result in a low solution pH, which in turn decreases the rate of photoreaction. Adjusting the solution pH to a favorable value at a given temperature will increase hydrogen production rate. Buhler and co-workers [13] mentioned that changing the temperature of the solution leads to shifts in the energy of both the semiconductor and electrolyte and causes changes in the electrochemical reaction rates. It has been shown that only small shifts of the flatband and redox potentials of the electrolyte species are observed with increasing temperature. Therefore the enhancement of photoactivity must be attributed to the increase in the exchange rate of the electrolyte species at the cadmium sulfide surface [13]. The aforementioned study is further validated by results obtained in this study. Results from both the UV light photolysis and the visible light photocatalysis show that the hydrogen evolution rate increases with increase of temperature within a certain temperature range. The temperature dependency is more significant in for the visible light photocatalysis than that achieved in UV light photolytic process. Since the UV light photolysis of ammonium sulfite solution does not include a semiconductor photocatalyst, any rate

Page 27: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

enhancement by temperature increase would only result from the change of redox potentials of the ammonium sulfite.

0

200

400

600

800

1000

1200

1400

0 50 100 150 200 250 300 350 400 450

T ime (min)

Hydr

ogen

(mL

50C70C90C

Figure 2.10. Hydrogen production rates at various temperatures Rate dependency on platinum loading on cadmium sulfide catalyst -- The preliminary experiments carried out have confirmed the importance of platinum loading on the semiconductor phototocatalyst. There is a need to determine the extent to which the semiconductor photocatalyst may be doped. Platinum metal acts as an electron bridge for the photocatalyst. However, if too much platinum is deposited on the surface of the photocatalyst, it may block the photocatalyst from the light source. Experiments have been carried out for four different values of weight percent for platinum in the cadmium sulfide photocatalyst. Figure 2.11 shows the results obtained from the experiment. Note that the percentage of Pt refers to the weight percent of Pt to 0.5000 gram of CdS. The Pt loading percentage is an apparent percentage. It is observed that the highest rate of hydrogen production was achieved at 1% weight of platinum in cadmium sulfide. Decreasing the amount of platinum loading to 0.5% weight led to a decrease in the production rate of hydrogen indicating that there was not enough platinum loaded onto the photocatalyst to maximize electron transfer from the photocatalyst to the sulfite ions. The hydrogen production rate for 2% weight platinum was slower than both 1% and 0.5% weight meaning the optimal platinum loading weight percent lies within the 0.5% to 1.0% range. Increasing the platinum loading to 3% weight showed a further decrease in the production rate of hydrogen. Based on the experimental findings, it was ascertained that there is an optimal value for the extent of platinum loading. Figure 2.12 shows the relationship between the rate of hydrogen production and platinum loading extent of cadmium sulfide. It is seen that the optimal platinum loading value is approximately at 1% weight of cadmium sulfide.

Page 28: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

0

50

100

150

200

250

0 50 100 150 200 250 300 350

Time (min)

Hyd

roge

n (m

L)

0.50%

1%

2%

3%

Figure 2.11. Platinum loading effect on hydrogen production rate

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 0.5 1 1.5 2 2.5 3 3.

CdS Platinum Loading %

Rate

(mL/

min

5

)

Figure 2.12. Optimal platinum loading on hydrogen production rate

Rate dependency on concentration of ammonium sulfite solution -- The hydrogen production rate dependency on the solution concentration has been previously studied by Buhler and co-workers [13]. Buhler’s work focused on hydrogen production utilizing both sodium sulfite and sodium sulfide solutions. The results of the work show that as the solution concentration increased, the rate of hydrogen production increased logarithmically for the sodium sulfite solution. As mentioned earlier in the paper, sodium sulfite and ammonium sulfite have very similar reaction characteristics. Both solutions

Page 29: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

undergo very little or no cationic influence. In Buhler’s work however, the concentration of the solution was increased to a maximum value of 1.0M. The purpose of this paper was to determine the concentration effect on hydrogen production rate using an ammonium sulfite solution. One of the reasons ammonium sulfite was selected for the proposed thermochemical cycle was its high solubility in aqueous solution. If a higher concentration of ammonium sulfite can be used in the cycle without affecting the rate of production of hydrogen, the efficiency of the cycle would be increased due to more efficient mass transfer. Three different concentrations were used for this study. Figure 13 shows that as the concentration increases from 0.432M to 0.866M, the rate of hydrogen production increases significantly, however, as the concentration of solution was increased from 0.866M to 1.731M, the rate of production of hydrogen remained the same, indicating that the optimal concentration value lies somewhere between 0.432M and 0.866M.

0

100

200

300

400

500

600

700

800

0 50 100 150 200 250 300 350 400

Reaction Time (min)

Hyd

roge

n (m

L)

0.432M0.866M1.731

Figure 13. Hydrogen production rate dependency on

ammonium sulfite concentration Catalyst screening -- It has been pointed out in this study that the cadmium sulfide photocatalyst needs to be doped with platinum metal to increase its activity. The technique for doping platinum metal onto the surface of the cadmium sulfide powder has a significant effect on hydrogen evolution rates. Five different methods were studied in this thesis work to determine their effect on the hydrogen production rate. Preparation Method A: 0.5000 g of Cadmium Sulfide powder (99.9%, Alfa Aesar) was placed in a 15 mL beaker. 2.67925g of 0.4 wt% H2PtCl6 solution (Aldrich) was added to the CdS powder. 0.11 g of NaBH4 (Aldrich) powder was added to 50mL of deionized water; the resulting solution will act as a reducing agent due to the hydrogen evolved when NaBH4 reacts with water. The hydrogen evolved from the borohydride water mixture reduces the H2PtCl6 solution into platinum metal and HCl. The platinum particles

Page 30: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

can then deposit onto the surface of the CdS powder. The NaBH4 aqueous solution was added drop wisely to the CdS/H2PtCl6 slurry under vigorously mixed conditions until the bright yellow slurry turns dark yellow, indicating the formation of platinum metal and its subsequent deposition on the surface of the CdS powder. A total of 10 – 15 drops of the NaBH4 solution was needed. Preparation Method B: 6.000 g of CdS powder (99.9%, Alfa Aesar) was heated at 400 oC in the presence of air for one hour. After heating the CdS powder was poured into 180 mL of deionized water and 180 drops of concentrated HNO3. The slurry was stirred for five minutes and then filtered. After filtration, the solid is rinsed with deionized water three times; it is then dried at 110 oC under a pressure of 0.3 atm. Followed by the pre-treatment, Method A was used for preparation of Pt/CdS photocatalyst. Preparation Method C: 0.5000 g of CdS powder (99.9%, Alfa Aesar) was placed in a 15 mL beaker. 2.6868 g of 0.4 wt% H2PtCl6 was added to the CdS powder. The slurry was diluted with deionized water to a volume of 25 mL. Hydrogen gas is bubbled through the slurry vigorously for 1.5 hours and the slurry was allowed to age for 16 hours, Figure 2.14 shows the color change before and after hydrogen bubbling of the slurry.

Figure 3.14. Colors of CdS slurry before and after bubbling hydrogen for 1.5 hours

Preparation Method D: 5.3643 g of 0.4 wt% H2PtCl6 was added to a 15 ml beaker. 3 mL of 0.442 g/mL of Polyvinlylpyrolidone (PVP) (Fluka, MW 30,000) was diluted by 9 mL of deinoized water and added to the beaker, bringing the total solution volume to 20 mL. Sodium borohydride solution prepared based on Method A was added drop wisely to the solution. The number of drops of sodium borohydride required was monitored by UV-Visible spectrometry measurement to determine the number of drops at which all the Pt(IV) ions were reduced to platinum metal. Figure 2.15 shows the results from the spectrometric analyses. With the addition of sodium borohydride, the H2PtCl6 peak, 270 nm[25] gradually reduced; after 22 drops of sodium borohydride solution was added to the PVP + H2PtCl6 solution, the H2PtCl6 peak completely disappeared, suggesting that it had been reduced to produce Pt nanosized particles. A total of 25 drops of sodium borohydride solution was used in this experiment. The resulting polymerized solution was dark brown. The pH of the solution was measured as 2.421. The highly acidic property of the solution was due to the formation of HCl during the reduction of the

Page 31: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

platinic acid. The pH of the solution was adjusted to 5.300 by adding ammonium sulfite solution. The CdS powder was then added to the as prepared solution. The slurry was sonicated with an ultrasonic shaker for 30 minuets to increase the deposition of platinum particles onto the surface of the CdS powder. The mixture was filtered and rinsed several times to remove PVP, which may interfere with the photocatalytic reaction. Preparation Method E: Method B was used to pretreat the CdS catalyst. 0.9996g of CdS was measured out and added to 5.3585g of H2PtCl6, which represents 1 wt% of platinum on the CdS catalyst. The mixture was dried at 110oC and 0.3 atm for 1 hr. The dried sample was then added to a glass tubular reactor where it was purged with hydrogen gas for 2hrs at 200oC. The aim was to reduce the H2PtCl6 by means of the hydrogen gas flowing over the catalyst.

0

1

2

3

4

5

200 250 300 350 400 450

Wavelength

Abs

orba

nce

PVP + Platinic acid

PVP + Platinic acid +10 drops borohydride

PVP + Platinic acid + 22 drops borohydride

PVP + Platinic acid + 25 drops borohydridde

Figure 2.15. UV-Visible spectrometry of the PVP + H2PtCl6 solution before and after the addition of NaBH4

Page 32: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Table 2.2. Effect of photocatalyst preparation methods on the hydrogen formation rates (0.866 M (NH4)2SO3, pH = 7.800, catalyst weight = 0.5g, Temperature = 70oC

Reactor volume 250 mL)

Description of Methods

Apparent Platinum Loading (wt%)

Hydrogen Formation Rates (mL/min)

Advantages

Disadvantages

A) CdS powder added to H2PtCl6 solution with NaBH4 added drop wise

1 1++

1.25+

0++

Process is simple and quick

Less repeatability Extended exposure leads to corrosion of CdS

B) CdS powder thermally treated and etched with water and HNO3

1

1.26

Eliminates presence of CdO

Time consuming

C) CdS powder added to H2PtCl6 and water with hydrogen gas bubbled through the slurry

1

1.24

Reduces CdS corrosion by eliminating the need for NaBH4

Slow reduction process

D) PVP added to H2PtCl6 solution and water with NaBH4 added drops wise

1

1.06

Protects platinum particles from agglomeration with one another

Time intensive and complicated process

D) PVP added to H2PtCl6 solution and water with NaBH4 added drops wise

2

2.37

Protects platinum particles from agglomeration with one another

Time intensive and complicated process

D) PVP added to H2PtCl6 solution and water with NaBH4 added drops wise

3

1.80+++

---

---

E) CdS added to H2PtCl6 then dried and purged with H2

1

0

Reduces CdS corrosion by eliminating the need for NaBH4

Difficult process. No hydrogen was produced.

Page 33: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

+The rate was taken as seen in Figure 3.8.1, but several months prior, the rate was 2.6 mL/min using this same method. This is due the lamp aging and other conditions change, such as focus, unstable power supply and reactor window blockage etc. ++Catalysts slurry was allowed to stir overnight under sealed conditions before being added to the reaction mixture. No visible hydrogen production was observed. Corrosion of the CdS photocatalyst by the HCL present in solution may have led to its inactivity. +++The solution was heated to 70oC before the reaction commenced to attain isothermal conditions as opposed to prior reactions that were allowed to reach that temperature gradually as the reaction progressed.

0

100

200

300

400

500

600

700

800

0 50 100 150 200 250 300 350 400

Time(min)

Hyd

roge

n (m

L)

PVP + Borohydride +Filtering

Etched Cadmium Sulfide

Hydrogen Bubbling through CdS Slurry

Borohydride directly to CdS/Pt

Figure 3.6.3: Hydrogen evolution rates with varying photocatalyst preparation methods

Catalyst screening results -- Hydrogen evolution rates with different catalyst preparation methods are depicted in Figure 2.16. Table 2.2 summarizes the five technologies applied for the Pt/CdS photocatalyst preparation and shows that method A resulted in one of the lowest hydrogen production rates of the four methods studied. Method A involved directly adding NaBH4, which acts as a reducing agent to the CdS/H2PtCl6 slurry, forming platinum metal and HCl. Due to the fact that the reaction occurs instantaneously, platinum metal particles formed immediately seek out hosting sites by which they can either be deposited on or other platinum particles to agglomerate with. Since CdS is in-situ of the slurry, the platinum particles deposit directly to the surface of the CdS. The low hydrogen production rate experienced when using method A can be attributed to a number of reasons. One of the reasons can be CdS corrosion due to the direct addition of a high pH solution such as NaBH4 to the surface of the catalyst hence

Page 34: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

reducing its activity. Another reason is that the method of preparation is not a precise procedure because it is difficult to control the amount of sodium borohydride required for the reduction of H2PtCl6. Adding too many drops of sodium borohydrite will lead to the production of large platinum particles that can be seen floating on the surface of the slurry; the CdS powder cannot utilize these particles because they are too large to deposit on the fine powder. It should be noted that a rate of up to 2.6mL/min of hydrogen was obtained using method A, however, this experiment was conducted several months prior to the experiment being discussed (1.25mL/min). The drastic drop-off in the rate of hydrogen production can be a result of the lamp aging and other condition changes, such as focus, unstable power supply and reactor window blockage etc. Method B was carried out primarily to eliminate any cadmium oxide (CdO) that formed on the surface on the CdS powder resulting from the exposure of CdS to air. It was believed that increasing the purity of the CdS powder would subsequently increase its photoactivity8. Figure 2.17 shows a comparison of the CdS powder before and after chemical etching. The CdS powder has a bright orange color before etching, outlined in Method B. After the etching process the CdS powder has a bright yellow color. Hydrogen formation rates obtained from experimental runs showed that chemically etching the CdS powder did not significantly increase the hydrogen evolution rates from the previous rate obtained using method A.

Figure 2.17. CdS powder samples before and after chemical etching with HNO3

Method C adopted a different approach to reduce the hexachloroplanic acid by directly bubbling hydrogen through the slurry mixture as opposed to the addition of sodium borohydride. It was evident that the rate of reduction of the hexachloroplatinic acid was significantly slower when bubbling hydrogen through the mixture. The slurry required more than an hour of vigorous hydrogen bubbling before any significant color change occurred. However, the hydrogen evolution rates obtained using method C were not impressive. One possible reason is for the slow rate of hydrogen production is that since the reduction reaction occurs slowly when directly bubbling hydrogen through the slurry, the amount of hydrogen bubbled was not sufficient, hence the hexachloroplatinic acid solution was not fully reduced to platinum metal particles.

Page 35: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Preparation method D resulted in the highest hydrogen evolution rates. The concept behind the preparation method was reported by Tang and co-workers9. In the process they proposed, a polymer, Polyvinlypyrrolidone (PVP), was employed as a stabilizer keeping Pt particles from aggregation. The H2PtCl6 reduction agents used in the process were hydrogen gas and methanol. The size of Pt particles prepared using this method was in the range of 2 to 4 nm. The function of PVP is obvious because if a stabilizing agent is not present in solution, the nanosized platinum particles formed will be quickly agglomerated to form large Pt particles and finally become Pt black. Large platinum particles are difficult to attach to the fine CdS powder surface, which is also a nanosized particle. One problem facing this method of preparation is that once the platinum nano-particles are being protected by PVP, they are very difficult to be deposited on the surface of the CdS powder. A method has to be devised to destroy the PVP protection, which will then reduce the stability of PVP and allow the platinum particles to attach to the CdS powder. One way is to reduce the concentration of PVP so that Pt particles are less stabilized and need to be relocated. The other alternative is to increase the concentration of H2PtCl6 so to increase the number of Pt particles to be deposited to the CdS surface. In Method D, we introduced a polymer-protected synthesis of nanosized Pt particles and combined this technology with a Method A for the preparation of highly efficient Pt/CdS photocatalysts. Sodium borohydride was used as a reduction agent to replace hydrogen and methanol. The uniqueness of the Method D is that it separates the Pt particle synthesis and Pt particle deposition on CdS into two independent steps. The process will allow us to control and measure the solution’s pH and Pt particle size. The potential loss of platinum particles from filtration using method D led to the increase of the amount of hexachloroplatinic acid used. Three different platinum weight percents (1%, 2%, and 3%) were studied using preparation method D. It was observed that the 2% wt sample resulted in the highest hydrogen production rate of the three. As noted earlier, one reason for the drop in hydrogen production rate when increasing from 2 wt% to 3 wt% can be the fact that the initial temperature for the 3 wt% run was set at 70oC as opposed to 25oC for all prior runs; the reason for this was to maintain isothermal conditions throughout the experiment. For all other runs, the solution temperature is allowed to gradually increase as a result of light irradiation from room temperature to 70oC. Heating the solution to 70oC prior to light irradiation will cause a pH effect, hence leading to lower hydrogen evolution rates. In summary, photoactivity of Pt/CdS photocatalysts depend upon the size of Pt particles loaded on CdS surface. Two issues need to be addressed in the preparation technique employed. One is the CdS corrosion issue caused by HCl generated after reduction of sodium borohydride. An experiment was carried out to address this issue. Utilizing Method A, we prepared Pt/CdS catalyst slurry and kept it overnight before conducting a photocatalytic reaction; we discovered that the hydrogen production rate was close to zero. This result may suggest that CdS reacted with HCl present in the slurry to form CdCl2 causing the loss of catalyst activity. In other methods presented in the study, all the Pt/CdS catalysts were used for the photocatalysis immediately after they were prepared. The second issue is Pt particle size control. It has been found that small particle size will promote the rate of hydrogen formation. Future endeavors will focus on how to efficiently deposit Pt nanosized particle onto CdS surface. . Efficiency of Photocatalysis of Ammonium Sulfite -- The overall efficiency is defined as the ratio of the hydrogen chemical energy to the total energy input for hydrogen

Page 36: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

generation. In this calculation, hydrogen high heating value, 285.9 kJ/mol, was used. The feasibility of the S-A thermochemical cycle is mainly depended on the efficiency of the photocatalytic step. The conversion of solar energy to hydrogen chemical energy has proven to be an arduous task; hence efficiency levels for processes such as Photovoltaic Electrolysis of water only reach as high as 6 – 7%. The main goal of this project is to attain an overall efficiency value of 20 to 30% for the hydrogen production step of the thermochemical cycle. In this work both the overall efficiency and the quantum efficiency of the process were calculated based on the highest hydrogen production rate obtained over an 8-hour period. The highest rate obtained for hydrogen production over an 8-hour period was 3.2 mL/min. The number of moles of hydrogen generated was calculated and this number was used to determine the hydrogen chemical energy. The light intensity incident on the photoreactor was determined using the measurement methods outlined above. The overall efficiency obtained is 12%, which is substantially higher than current efficiency values being reported for solar hydrogen from water or Photovoltaic electrolysis of water. The quantum yield of a given process is defined as the ratio of moles of hydrogen produced per Einstein of photons received by the process. The highest hydrogen production rate (3.2 mL/min) was once again used to calculate the quantum efficiency of the process. The total number of moles of hydrogen atoms generated in one hour was calculated. In order to determine the total number of photons generated by the xenon lamp per hour, a radiant energy was required. Due to the fact that a spectrum of xenon light contains a variety of wavelengths, an average of the wavelength (365 nm) was determined from the spectrum measurement (Figure 2.5) between 300 nm – 450 nm. Once the radiant energy of the light source was obtained, the number of photons generated by the xenon lamp was calculated in Einstein/hr; this value was then used to calculate the quantum efficiency per hydrogen atom. The quantum efficiency per hydrogen atom generated was determined to be 27.0%. The value indicates that the photocatalysis of ammonium sulfite for the production of hydrogen is a very promising process. The highest quantum efficiency value reported for photocatalytic processes of sodium sulfite and sodium sulfate solution is 40% [8]. Kinetic Study -- A kinetic study was conducted to determine the reaction rate and mechanism of the visible light photocatalysis of ammonium sulfite solution for the production of hydrogen. Material balances are also carried out to confirm the reaction mechanism proposed. A 0.866 M ammonium sulfite solution was used for the kinetic experiment. The reactor temperature was maintained at 35oC and the initial pH of the solution was approximately 7.9. 0.5000 g of CdS platinized with 1 wt% platinum was used as the photocatalyst that was prepared using preparation Method A outlined in Section 2.5. The reaction time was determined when no hydrogen bubbles were observed with a total of 32 hours and the kinetic curve is depicted in Figure 2.18, indicating a constant hydrogen production rate of 1.79 mL/min. The rate begins to slow down after about 1500 minutes then approaches zero after 2000 minutes. The online pH measurement shows that the solution’s pH increases slightly from about 7.9 to 8.6. The increase in the pH solution is indicative of the formation of dithionate dimer as discussed in previous sections. In order to confirm the presence of the dithionate ion and to determine the initial and end products of the reaction, a high-pressure liquid chromatograph (HPLC) (Dionex DX-500)

Page 37: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

was employed for the measurement of species in solutions before and after reaction. The HPLC analysis shows that before photocatalytic reaction there already exists ammonium sulfate in the ammonium sulfite solution. This indicates that the ammonium sulfite has been oxidized during before shipping to our lab. The mechanism of photocatalysis of ammonium sulfite solution is almost identical to that from the UV light photolysis except that it requires a semiconductor photocatalyst. CdS served as a photocatalyst to generate electron hole pairs upon visible light irradiation according to the following. (2.2)

−+ +hv

ehCdS → vbvb

0

500

1000

1500

2000

2500

3000

0 500 1000 1500 2000

Reaction Time (min)

Hyd

roge

n (m

L)

0.000

2.000

4.000

6.000

8.000

10.000

12.000

pH

HydrogenpH

Figure 2.18. Kinetic and pH curves for the production of hydrogen from visible

light photocatalytic process of aqueous ammonium sulfite solution Electrons generated by the irradiated photocatalyst can reduce H+ ions to generate hydrogen, while sulfite ions are oxidized by the holes to produce sulfate ions. At the same time, part of the sulfite ions can be polymerized to form dithionate ions. These processes are shown in Reactions 2.3 to 2.5.

+−+− +→++ HSOhOHSO 22 242

23 (2.3)

−−− +→ eOSSO 22 262

23 (2.4)

222 HeH →+ −+ (2.5) After reaction, no sulfite ions were detected by the HPLC analysis, indicating that all the sulfite ions had totally been consumed over the 32-hour reaction period. As mentioned

Page 38: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

earlier, the increase of pH as the reaction progressed is an indicator of the formation of dithionate ions; this assumption was confirmed by the HPLC analysis for dithionate ions. HPLC spectra show the presence of a dithionate ion peak after 32 hours of reaction. The formation of dithionate ions was also confirmed in the UV light photolytic experiments as a reaction intermediate, however, with the reaction progressing, dithionate ions were further oxidized to generate sulfate ions and hydrogen or dissociated into sulfite ions. It was observed, however, that in the case of the visible light photocatalytic process, a significant amount of dithionate ions were detected after 32 hours of reaction. As outlined earlier, Huang4 proposed the following mechanism for sulfite polymerization to form a dithionate dimer while reducing water to produce hydrogen: Oxidation: 2SO3

2- = S2O62- + 2e- (2.6)

Reduction: 2H2O + 2e- = H2 + 2OH- (2.7) Overall: 2SO3

2- + 2H2O = S2O62- + H2 + 2OH- (2.8)

The initial pH increase as the reaction progresses indicates that the formation of dithionate ions occurs immediately after light irradiation begins. As a comparison, the pH increase under UV light irradiation is not observed until 3 hours after light irradiation begins; indicating that under UV light photolysis, sulfite oxidation to sulfate is much faster than the sulfite to dithionate reaction. Table 2.3 shows the material balance carried out on 32 hours photocatalytic reaction for 0.866M of the ammonium sulfite solution. 0.866M represents the apparent concentration of the solution based on the mass of salt (29.035g) measured for the reaction; this however, does not represent the actual concentration of the sample due to the presence of sulfate ions as a result of oxidation in air during storage as mentioned earlier in the section. An accurate concentration for the solution was determined using HPLC analysis. The results show that there is only a 5% difference between the amount of sulfite ions consumed and the sum of sulfate ions and dithionate ions produced. The margin of error is acceptable because the HPLC measurement itself has errors. From the results obtained in the material balance, it can be seen that the ammonium sulfite sample contained 30 % of sulfate ions prior to reaction. The presence of the sulfate ions can be attributed to sulfite oxidation in air during storage. A total of 123 mmol of sulfite ions were consumed during the reaction. In liquid phase, 47 mmol of sulfate ions and 69 mmol of dithionate ions were produced and, in gaseous phase, a total of 2651 mL of hydrogen was produced over the 32 hours of reaction. The amount of hydrogen produced by the reaction was only 7.3% less than the theoretical value which was calculated based on the number of moles of sulfite ions consumed by the reaction because the production of the sum of sulfate and dithionate ions and the production of hydrogen gas are at 1:1 ratio. The theoretical volume of hydrogen produced was calculated based on the ideal gas law. The difference of 7.3% between the theoretical and measured volume of hydrogen gas may result from hydrogen dissolution in water and/or hydrogen adsorption on the photocatalyst. Furthermore, hydrogen produced may partially be consumed to reduce Pt2Cl3*3H2O in case it is not fully reduced during the catalyst preparation. The errors encountered from HPLC quantification will also directly affect the results obtained in the gaseous phase material balance calculation.

Page 39: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

Table 2.3. Material balance for 0.866M ammonium sulfite solution irradiated with visible light over a 32 hour period

Component Initial (mmol) Final (mmol) Produced (mmol)

Consumed (mmol)

SO32- 123.38 0.00 0.00 123.38

SO42- 39.59 86.96 47.37 0.00

S2O62- 0.00 69.62 69.62 0.00

H2

Expected H2 (mL) Actual H2 (mL) Difference (mL) % Difference

2858.6 2651.0 207.6 7.3 Conclusions The objective of this research was to implement a lab scale experiment to study solar hydrogen production through a thermochemical cycle. The research focused on the photocatalytic step in the Sulfur-Ammonia thermochemical process proposed by the Florida Solar Energy Center, which involves the use of visible light to irradiadiate an aqueous ammonium sulfite solution for hydrogen production. The visible light photocatalytic oxidation of ammonium sulfite solution was carried out to simulate solar hydrogen production. The focus of the research was to optimize the reaction conditions and to conduct a photocatalyst screening. It was observed that the concentration of the ammonium sulfite solution and reaction temperature had significant effects on the rates of hydrogen production. There existed optimal ranges for both concentration and temperature. It was also determined that optimal range for the weight percentage of platinum doping on CdS photocatalyst existed between 0.5 to 1.0 wt% for the maximum hydrogen production rate. Experimental results showed that reaction conditions such as temperature and pH of the ammonium sulfite solution had a significant effect on hydrogen evolution rates. No significant rate change was observed as the solution concentration was increased, indicating that other than the solution’s concentration, light intensity and mass transfer rate were limiting factors in the photoreaction system. Four methods of preparation for Pt/CdS photocatalyst were studied and the best technique was identified. The synthesis of nanosized Pt particles via a polymer protection process was found to be a critically important step in the preparation of Pt/CdS photocatalysts. Two important factors were defined for preparation of highly efficient photocatalysts; one was the corrosion of CdS particles and the other was the size of platinum particles. Nanoscaled Pt particles loaded onto the CdS surface can increase the hydrogen production rate. Kinetic studies and a material balance were performed for the visible light photocatalytic process. Similar reaction routs with the UV light photolysis of ammonia sulfite solution were determined in the visible light photocatalytic process. The material balance calculation indicated that in the course of UV light photolytic process, sulfite ions were completely converted into sulfate ions and no dithionate ions were detected as an end product. Dithionate ions were generated as an intermediate and were further converted to sulfate ions and hydrogen as the reaction progressed. However, in the visible light

Page 40: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

photocatalytic reaction, both sulfate and dithionate ions were detected as end products. The dithionate ions detected may further be converted to sulfate ions, but the process will require longer time than that experienced in the UV light photolytic process. The energy and the quantum efficiencies of the visible light photocatalytic process were determined based on the highest rate of hydrogen production recorded (3.2 mL/min with 0.500 g of Pt/CdS photocatalyst and a solution volume of 250 mL). The process has achieved 12.0% energy efficiency and 27.0% quantum efficiency. These efficiencies are higher or comparable to the efficiency values reported in literature publications. Potentially, the energy efficiency can be expected to reach up to 20 to 30%. Results show that both UV light photolytic and visible light photocatalytic processes are highly efficient photochemical processes for the production of hydrogen via oxidation of ammonium sulfite solution. Future Work While it has been shown that both UV light photolysis and visible light photocatalysis of ammonium sulfite can produce 100% pure hydrogen at relatively high efficiencies, further research must be conducted to improve upon current efficiency values obtained for the photochemical step of the sulfur-ammonia cycle and incorporating this step to the overall cycle. Some areas of this study that still require additional research include:

• Identification of more effective photocatalysts capable of absorbing light wider visible light wavelength.

• Preparing nanosize photocatalysts to enhance system efficiency. • Designing novel photoreactors that will increase the efficiency of solar light

harvesting. • Accurately simulating solar spectra both on lunar and earth surfaces for the

production of hydrogen from ammonium sulfite aqueous solution. • Performing closed loop experiments by combing all four steps of the sulfur

ammonia cycle.4 1Lu, P. W. T. (1983). Technological aspects of sulfur dioxide depolarized electrolysis for hydrogen production. Int. J. Hydrogen Energy, 8(10), 773-781. 2Takehara, Z., Nogami, M., & Shimizu, Y. (1989). Int. J. Hydrogen Energy, 14(4), 233-239. 3Huang, C., & T-Raissi, A. (2004, June). A new solar thermochemical water-splitting cycle for hydrogen production. Proceedings of the 15th World Hydrogen Energy Conference, Yokohama, Japan. 4 Huang C, Linkous C., & Adebiyi, O. (2006). Hydrogen production via photolytic oxidation of sodium sulfite aqueous solution. Paper submitted for publication. 5 Sepone, N. & Pelizzetti, E. (1989). Photocatalysis: Fundamentals and applications. New York: John Wiley and Sons. 6 http://netserv.ipc.unilinz.ac.at/~dieter/DsWeb/Research/Detox/Photocatalysis.htm

Page 41: Solar Powered Hydrogen Production via a High …hydrogenresearch.org/FR04/FSEC--Huang et al--Solar...Solar Powered Hydrogen Production via a High Temperature Photocatalytic Water Splitting

7Ashokkumar M. (1998). An overview on semiconductor particulate systems for photoproduction of hydrogen. Int. J. Hydrogen Energy, 23(6), 427-438. 8Linkous C, Huang C, & Fowler J. R. (2004). UV photochemical oxidation of aqueous sodium sulfide to produce hydrogen and sulfur. Journal of Photochemistry and Photobiology A: Chemistry, 168, 153-160 9 Tang Z, Geng D, & Lu G. (2005). (2005). Size-controlled synthesis of colloidal platinum nanoparticles and their activity for the electrocatalytic oxidation of carbon monoxide. Journal of Colloid and Interface Science, 287, 159-166. 10Buhler N, Meier K, & Reber J. F. (1984). Photochemical hydrogen production with cadmium sulfide suspensions. Journal of Physical Chemistry, 88(15), 3261-3268. 11Ashokkumar M. (1998). An overview on semiconductor particulate systems for photoproduction of hydrogen. Int. J. Hydrogen Energy, 23(6), 427-438. March 2006