Design of Solar Thermo-Chemical Reactor and Bench Scale Testing of Ferrites for Syngas Production

MISSING IMAGE

Material Information

Title:
Design of Solar Thermo-Chemical Reactor and Bench Scale Testing of Ferrites for Syngas Production
Physical Description:
1 online resource (65 p.)
Language:
english
Creator:
Mishra, Rishi
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Mechanical Engineering, Mechanical and Aerospace Engineering
Committee Chair:
Klausner, James F
Committee Co-Chair:
Mei, Renwei
Committee Members:
Auyeung, Nicholas

Subjects

Subjects / Keywords:
ferrites -- reactor -- solar -- thermochemical
Mechanical and Aerospace Engineering -- Dissertations, Academic -- UF
Genre:
Mechanical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
With a tremendous increase in demand for energy, the search for renewable sources has gained momentum in the past decade. Fossil fuel reserves are limited and the exploration and extraction has become increasingly difficult. The UF Solar Fuel Team is developing a technology to produce a sustainable fuel supply. Solar fuels will not only solve the problem of increased demand of energy but also reduce the harmful impact of burning fuels in the environment. Research efforts are in progress to produce syngas which is a precursor to hydrocarbon fuels. A two-step thermochemical cycle consisting of oxidation of ferrite material with water and CO2 splitting followed by thermal reduction to release oxygen and regenerate the material is one method of producing syngas. Under atmospheric pressure, the temperature at which thermal reduction occurs spontaneously is above 1500 °C for ferrites. A high temperature increases the chances of sintering of the material. By decreasing the partial pressure of oxygen evolved during reduction step, the reduction temperature can be decreased. Thermal reduction of the material by creating vacuum in the reaction tube and also by flowing inert gas was successfully tested on a bench scale reactor. The material showed no loss in reactivity after 18 cycles. Even after the use of vacuum and inert, the reduction temperature is still very high, in excess of 1400°C, which creates a need for special reactor with high service temperature. In addition, there are several other design challenges such as achieving uniformity of flux on tubes, chemical inertness of the reactor, and minimizing heat losses that govern the design of the reactor. A solar reactor that meets the above requirements was successfully designed, built and tested in front of the solar simulator. The reactor gave a service temperature of around 1600 °C without any major thermal cracking of the components. It was also efficient enough to reach 1600°C with less than 10 kW power at the aperture.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Rishi Mishra.
Thesis:
Thesis (M.S.)--University of Florida, 2013.
Local:
Adviser: Klausner, James F.
Local:
Co-adviser: Mei, Renwei.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-11-30

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2013
System ID:
UFE0045537:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 DESIGN OF SOLAR THERMO CHEMICAL REACTOR AND BENCH SCALE TESTING OF FERRITES FOR SYNGAS PRODUCTION By RISHI MISHRA A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2013

PAGE 2

2 2013 Rishi Mishra

PAGE 3

3 To my p arents

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank Dr. James F. Klausner for giving me the opportunity to work on this project. This is my first research experience and it has been a pleasure working under his guidance. I would particularly like to thank Dr. Nicholas AuYeung for being such a wonderful mentor. I cannot imagine completing this thesis without his support. I would like to thank him for teaching me everything from setting up a rig to analyzing data. I would also like to thank Dr. Renwei Mei and Dr. David Hahn for providing valuable inputs on my work. Special thanks go to Nikhil Sehgal for getting me started on the project. I would also like to express my deepest gratitude to Dr. Ayyoub Mehdizade h for helping me out with the reactor design. Finally I want to thank my parents for helping me to pursue my dreams.

PAGE 5

5 TABLE OF CONTENTS p age ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF FIGURES .......................................................................................................... 7 ABSTRACT ..................................................................................................................... 9 CHAPTER 1 LITERATURE REVIEW .......................................................................................... 11 Solar Thermochemical Materials ............................................................................ 11 Solar Thermochemical Reactors ............................................................................. 14 2 BENCH SCALE TESTING OF REACTIVE MATERIAL .......................................... 17 Need for Bench Scale Testing ................................................................................ 17 Vacuum and Inert Reduction .................................................................................. 17 Experimental Facility ............................................................................................... 18 Analytical Instrumentation ....................................................................................... 19 Material Preparation and Structure Formation ........................................................ 19 Co Precipitation ...................................................................................................... 20 Experimental Procedure (5 g S ample) .................................................................... 20 Experimental Procedure (18g S ample) ................................................................... 21 Results (5g S ample) ............................................................................................... 21 Results (18g S ample) ............................................................................................. 23 3 SOLAR REACTOR DESIGN .................................................................................. 31 Maximum Service Temperature .............................................................................. 31 Heat Loss ................................................................................................................ 31 Thermal Shock ........................................................................................................ 32 Flux D istribution ...................................................................................................... 32 Chemical Stability ................................................................................................... 32 Ease of Manufacture ............................................................................................... 33 Material Selection ................................................................................................... 33 Aperture Radius Selection ...................................................................................... 33 Cavity Diam eter and Length Calculation ................................................................. 33 Reactor Components .............................................................................................. 34 120 Bent Tube Concept ......................................................................................... 34 Tube within a Tube ................................................................................................. 35 4 THERMAL TESTING OF SOLAR REACTOR ......................................................... 50 Experimental Facility ............................................................................................... 50

PAGE 6

6 Reactor Thermal Testing ........................................................................................ 50 Analysis of Tube Temperature ................................................................................ 51 Analysis of Cavity Temperature .............................................................................. 51 Analysis of Back Lid ................................................................................................ 52 Observations ........................................................................................................... 52 Recommendations .................................................................................................. 52 LIST OF REFERENCES ............................................................................................... 62 BIOGRAPHICAL SKETCH ............................................................................................ 65

PAGE 7

7 L IST OF FIGURES Figure P age 2 1 Graph showing variation of reduction temperature with partial pressure of O2 ... 24 2 2 Experimental facility for bench scale testing ....................................................... 25 2 3 Picture of the bench scale rig ............................................................................. 25 2 4 Hydrogen production following thermal reduction at vacuu m holding for 25 and 60 minutes at 1450C. ................................................................................. 26 2 5 Comparison subsequent H2 production after inert purge and vacuum environments for a 45 min thermal reduction at 1400C ..................................... 27 2 6 Oxygen production during a 45 min thermal reduction at 1400 ......................... 27 2 7 Hydrogen Production in Cycle 1 and Cycle 18 ................................................... 28 2 8 Hydrogen production achieved after several reduction schemes ....................... 29 2 9 Specific H2 yield .................................................................................................. 30 3 1 Solar Simulator ................................................................................................... 36 3 2 Minimum diameter calculation (All dimensions in inches) ................................... 36 3 3 Maximum length calculation (All dimensions in inches) ...................................... 36 3 4 1 Reactor Assembly (All dimensions in inches) ................................................. 37 3 5 Front lid for 1 reactor design (All dimensions in inches) .................................... 38 3 6 Back lid for 1 reactor design (All dimensions in inches) .................................... 39 3 7 Cavity for 1reactor design (All dimensions in inches) ........................................ 40 3 8 1 Tube (All dimensions in inches) ...................................................................... 41 3 9 Front lid for 2 reactor (All dimensions in inches) ................................................ 42 3 10 Cavity for 2 reactor design (All dimensions in inches) ....................................... 43 3 11 Back lid for 2 reactor design (All dimensions in inches) .................................... 44 3 12 120 tube design................................................................................................. 45 3 13 120 Bent tube (All dimensions in inches) .......................................................... 45

PAGE 8

8 3 14 Front lid for 120 bent tube design (All dimensions in inches) ............................ 46 3 15 Back lid for 120 bent tube design (All dimensions in inches) ............................ 47 3 16 Cavity for 120 bent tube design (All dimensions in i nches) ............................... 48 3 17 Assembly showing tube within a tube design ..................................................... 49 4 1 XY Table ............................................................................................................. 53 4 2 Shutter in front of the mirrors .............................................................................. 54 4 3 Aluminum framework on the X Y table ............................................................... 55 4 4 R eactor with alumina tube in the cavity .............................................................. 56 4 5 Variation of tube temperature with time .............................................................. 56 4 6 Variation of temperatures on top and side of the cavity ...................................... 57 4 7 Variation of Back lid Temperature ...................................................................... 57 4 8 Crack developed on the Back Lid ....................................................................... 58 4 9 Erosion of material at the aperture of front lid ..................................................... 59 4 10 Picture showing shri nkage of top half of the cavity ............................................. 60 4 11 Discoloration of the insulation Blanket ................................................................ 61

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the Univer sity of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DESIGN OF SOLAR THERMO CHEMICAL REACTOR AND BENCH SCALE TESTING OF FERRITES FOR SYNGAS PRODUCTION By Rishi Mishra May 2013 Chair: James F. Kl ausner Cochair: Renwei Mei Major: Mechanical Engineering With a tremendous increase in demand for energy, the search for renewable sources has gained momentum in the past decade. Fossil fuel reserves are limited and the exploration and extraction has beco me increasingly difficult. The UF Solar Fuel Team is developing a technology to produce a sustainable fuel supply. Solar fuels will not only solve the problem of increased demand of energy but also reduce the harmful impact of burning fuels in the environm ent. Research efforts are in progress to produce syngas which is a precursor to hydrocarbon fuels. A twostep thermochemical cycle consisting of oxidation of ferrite material with water and CO2 splitting followed by thermal reduction to release oxygen and regenerate the material is one method of producing syngas Under atmospheric pressure, t he temperature at which thermal reduction occurs spontaneously is above 1500 C for ferrites A high temperature increases the chances of sintering of the material. By decreasing the partial pressure of oxygen evolved during reduction step, the reduction temperature can be decreased. Thermal reduction of the material by creating vacuum in the reaction tube and also by flowing inert gas was successfully

PAGE 10

10 tested on a bench scale reactor. The material showed no loss in reactivity after 18 cycles. Even after the use of vacuum and inert, the reduction temperature is still very high, in excess of 1400C, which creates a need for special reactor with high service temperature. I n addition, there are several other design challenges such as achieving uniformity of flux on tubes, chemical inertness of the reactor, and minimizing heat loss es that govern the design of the reactor. A solar reactor that meets the above requirements was successfully designed, built and tested in front of the solar simulator. The reactor gave a service temperature of around 1600 C without any major thermal cracking of the components. It was also efficient enough to reach 1600C with less than 10 kW power at the aperture

PAGE 11

11 CHAPTER 1 LITERATURE REVIEW Solar Thermochemical Materials The demand for fossil fuels has been rising every day despite limited resources It is expected that coal, natural gas and oil will run out in 130, 60 and 42 years respectively a t present rate of consumption [1]. Another problem associated with the use of fossil fuels is air pollution. Fossil fuels are known to emit greenhouse gases and other obnoxious pollutants that harm the environment. With the exploration and extraction of new fossil fuel resources becoming increasingly difficult and in consideration of their illeffects on environment, technologies for fuel production using solar energy provide an attractive alternative. The production of synthetic fuels requires no changes i n the present distribution system and uses CO2 and water as the raw material, both of which are available in relative abundance. The most obvious way of producing hydrogen fuel is by direct splitting of water. But it requires a temperature of over 2500C which is difficult to achieve with solar energy and creates gas sealing and recombination problems [2]. Also, manufacturing a reactor that can provide such high service temperature is difficult with present solar reactor and ceramic technologies. Two step thermochemical cycles present an alternative to direct thermolysis. This process splits water and/or carbondioxide over a metal oxide in different oxidation and reduction steps thus eliminating the need to separate gases at high temperatures. These cycles operate at temperatures much lower than direct water splitting thus making reactor development more feasible. The first two step metal oxide thermochemical cycle was the Fe3O4/FeO redox pair, which was proposed and thermodynamically evaluated by Nakamur a [3].

PAGE 12

12 However, the temperature at which spontaneous thermal reduction of Fe3O4 occurs was reported to be above 2000C. This temperature can be decreased via partial substitution of iron by other metals [4] thus forming mixed metal oxides. Different metal s (Co, Ni, Zn etc) in different compositions have been studied [5] and among the studied samples NiFe2O4 was found to be the most reactive material. However, at high temperatures, ferrites much like pure ironbased oxides, lose their chemical surface area considerably due to sintering [6]. This causes a decrease in hydrogen production in subsequent cycles. Kodama et al. [7] first demonstrated the use of monoclinic zirconia supported ferrites in the two step thermochemical looping process. It was also shown that the depth of reduction increased with higher thermal reduction temperatures [8]. Other metal oxide pairs (Mn3O4/MnO, Co3O4/CoO, ZnO/Zn, SnO2/Sn) [9 11] have also been investigated for two step thermochemical water splitting. Thermodynamic calculations [12] show that Mn3O4 and Co3O4 can be decomposed at 1537C and 902C respectively, which is less than thermal reduction temperature of Fe3O4. But the overall hydrogen yield was 0.002% for manganese oxide and 4107% for cobalt oxide. Another redox pair Nb2O5/NbO2 was also studied and the H2 yield was found to be 99.7% at 627C but the temperature at which it decomposes was 3327C in air which is above the melting point of the redox pair. Another redox pair In2O3 was studied [12] but it too required a hig her temperature for decomposition as compared to ferrites. CdO has a lower thermal reduction temperature (below 1400C) but its oxidation was limited by melting. Among these single metal oxide cycles, ZnO/Zn was the only potential redox pair and was extens ively studied by many research groups [13-

PAGE 13

13 14]. It is a volatile cycle and is thermodynamically favorable due to increase in the entropy from solid to gas phase change, but presents a problem of cooling the product gases to prevent recombination and separat ion of the gases at high temperatures [15]. The reduction of ZnO2 at atmospheric pressures occurs at temperatures over 2200C. To decrease the reduction temperature the pressure of product gases can be decreased thus making the reduction thermodynamically favorable. Another redox pair involving CeO2/Ce2O3 has been studied recently [1619] that involves nonstoichiometric conversion. One advantage of using ceria over zinc is that it does not require quenching of the products as its melting point is above the reduction temperature. However, the specific yield of H2 for ceria based cycles is low as compared to zinc and ferrites. The lower reactivity of ceria is compensated by the high thermal stability of ceria, which at the temperatures of interest does not necessarily require an inert matrix to avoid sintering. A novel solution for making stabilized iron cobalt structure by atomic layer deposition (ALD) was proposed by Arifin et al. [2021]. The structure was formed by coating alumina over a polymer substrate burning out the polymer and then coating the cobalt ferrite material. The ALD material had lower reduction temperature (200C 300C less) as compared to cobalt doped ferrites. Also, the material showed less deactivation after several cycles as compared t o ferrites.

PAGE 14

14 Solar Thermochemical Reactors Many solar thermochemical reactors have been built to facilitate high temperature processes. They can be classified into two categories: Directly irradiated and indirectly irradiated reactors [22]. The direc tly irradiated reactors have a transparent window from which the solar radiation enters the reactor cavity. The material receives the solar radiation directly in such kind of reactors. In an indirectly irradiated reactor, the material is heated inside a tube that receives the solar radiation. Heat is transferred to the material via conduction and indirect radiation from the reactor tube surface. This type of reactor does not require transparent windows for its operation. Directly irradiated rea ctors have faster heat transfer to the material but they are not 100% transparent to the incoming solar radiation. Also, dust can accumulate on the windows over a period of time which can further decrease transparency. Furthermore, quartz or glass windows are likely to develop crack during temperature change thus making it even difficult to scale up the reactor. The German Aerospace Center (DLR) has built a multi channeled ceramic honeycomb ceramic support coated with ferrites in a solar reactor [23]. The ferrite coate d honeycombs directly receive the radiation and get heated to 1200C to 1300C in the thermal reduction step. Steam is passed during the oxidation step and the solar power input is reduced to bring down the temperature for oxidation. In this reactor configuration the problem of recombination was eliminated but another problem of reaction of ferrites with SiC was encountered at high temperatures. However, the reaction with SiC can be stopped by having a layer of alumina or zirconia on the SiC cylinder.

PAGE 15

15 In S andia National Laboratories, United States, a Counter Rotating Ring Receiver/Reactor/Recuperator (CR 5) design was developed [24]. It has two sets of ferrite materials in close proximity that move in opposite directions thus facilitating heat recuperation between oxidation and reduction. The reactor has counter rotating fins with ferrite on support. The rings are directly irradiated by radiation through a transparent window. Both the DLR and CR5 designs have a disadvantage in terms of maximum loading of reactant mass. Thus the chemically active surface area in both the reactors is also limited which means that many reactors have to be run in parallel to get a mass production of hydrogen. This problem can be solved by use of volumetric gas particle solar reactor receiver [25]. Fine particles have a larger surface area than a honeycomb structure and can provide better hydrogen productivity as compared to other reactors using metal oxide structures. However, this reactor also has some disadvantages as it is dif ficult to keep the particles separated from the glass window. Also, the radiation has to be distributed well inside the reactor to take all the particles to reduction temperatures. For ZnO/Zn cycle another cavity receiver reactor design called ROCA was tes ted at Paul Scherrer Institute [26]. The rotating design of the reactor enables it to effectively use the solar radiation on all the particles. However, the drawback of this reactor is in its inefficiency in recovering a large percentage of products from t he reactor. An interesting reactor was designed and developed by Meier et al. [27] with a multi tube reaction chamber. In this design the solar radiation enters through a circular aperture and heats the absorber tubs that are regularly arranged inside the reaction

PAGE 16

16 chamber. Limestone particles are continuously fed from the back and as the reactor rotates they get heated and transported. Both the directly and indirectly irradiated reactor offer advantages and disadvantages depending on the application. The design and material selection of reactor is governed by the reaction kinetics, chemical compatibility of the reactant, thermal cycling rates and most importantly the scalability of reactor design.

PAGE 17

17 CHAPTER 2 BENCH SCALE TESTING OF RE ACTIVE MATERIAL Need for Bench Scale Testing The feasibility of a material to produce hydrogen can be identified by using thermogravimetry. However, the amount of sample that is used in a typical thermogravimetry is in fractions of a gram. With such a small sample it is difficult to perceive how well the material will behave when filled in a reactor tube. Due to sintering, the material might shrink in the tube causing the reactant gases to move around it instead of diffusing through the bulk material. This will reduce the effective chemical surface area available for oxidation and thus making the material unusable, although it showed good results in the thermo gravimeter. Also, in a thermogravimeter, the amount of inert gas flow on sample is proportionally large to the point where the partial pressure of product gas is at a level that is not achievable in an actual reactor (e.g. less than 104 bar). Hence, the bench scale testing is vital in finding the reactivity of the material in real conditions. Vac uum and Inert R eduction The change in Gibbs free energy for a reaction to proceed spontaneously must be less than or equal to zero. Iron oxide (Fe3O4) does not reduce spontaneously below 2500C at 1 atm pressure as shown in Fig ure 2 1 The problem compounds, with its melting point at 1566C as handling liquids in a reactor can be far more challenging than solids. One of the methods to decrease the reduction temperature is to lower the pressure of the product gas. The equilibrium quotient of the forward reaction for reduction is directly proportional to the partial pressure of oxygen. As the partial

PAGE 18

18 pressure decreases, the reaction goes further in the forward direction according to Le Chateliers principle thus reducing more material. In ironsteam looping t his can be achieved by lowering the partial pressure of oxygen evolved. In order to reduce the partial pressure two methods were analyzed. One is by reducing the partial pressure of oxygen by flowing large volumes of an inert gas through the system. The ot her is by reduction of overall pressure of the system by pulling vacuum in the system. Both the methods were experimented in the bench scale reactor as presented in the following sections. Experimental Facility Fig ure 2 2 and Figure 23 show the experiment al facility with different components marked with a number. There are three gas tanks (1) for argon, carbondioxide and helium. Gases flow in separate pipes to the flow controllers (2). The flow controllers for argon, carbondioxide and helium can have a maximum flow rate of 1 SLPM, 5 SLPM and 100 SCCM respectively. The outlet from the flow controllers is connected to a common line. The common line has a ball valve to shut off the flow as required. Steam is delivered via a syringe pump (4) flowing into heated stainless steel tubing, where temperature is kept at 200C using PID controlled resistive heating tape (5). The steam generator has been tested up to 3 g/min successfully A valve (6) is installed in a section tapped from the main steam generator line to test the uniformity of the steam flow. Another ball valve is connected right below the alumina tube (17) to shut off the flow. The tube is put in a high t emperature furnace (11) capable of reaching uum A vacuum pump (10) is connected to the alumina tube with an inline valve ( 7) and a pressure transducer (9). The upper end of alumina tube is connected to a pressure transducer (12). A ball valve

PAGE 19

19 (13) is connected in the outlet line which goes to condenser (14) then trap (15) and flow meter (16). The outlet of flow meter is connected to the exhaust line. From the exhaust line two separate pipe lines are tapped for mass spectrometer and gas chromatograph. All data is acquired by interfacing instrumentati on with National Instruments DAQ board and Labview software. Hydrogen production is monitored by mass flow meter readings and can be confirmed via mass spectrometer. A Type B thermocouple measures the internal temperature of the furnace. Current efforts ar e focused on reaching vacuum levels below 100 mTorr for the system. Every fitting and valve contributes to decreasing pumping speed. Furthermore, pulling vacuum through a bed of powdered material negatively influences the achievable level of vacuum. Analyt ical Instrumentation A Hiden HPR20 Mass Spectrometer equipped with a Faraday and Scanning Electron Multiplier detectors can be used to monitor the composition of product gases at near atmospheric pressure. Depth of reduction was determined instead by the extent of oxidation in the following step. Although mass spectrometers can achieve fast sample rates of roughly 20 seconds, the identification of gases can be problematic due to similar fragmentation patterns and molecular weights. Furthermore, apparent changes in gas composition can also be attributed to total pressure changes rather than actual composition changes. To avoid such problems, a gas chromatograph equipped with a thermal conductivity detector (TCD), molecular sieve column, and Hayesep D column was used in initial experiments to verify mass spectrometer readings. Material Preparation and Structure Formation A co precipitation method was used to create a homogeneous mixture of iron, cobalt and YSZ The sample was then rigorously calcined at 1350 C for 36 h followed

PAGE 20

20 by 4 h at 1450C under oxidative conditions (air). The sample was removed and mixed thoroughly with 3.6 g and 7.2g of 300 mesh and 100 mesh graphite, respectively. It was then brought up to 1500C under inert (Argon) atmosphere. The tem perature was held at 1500C for 1 h to promote sintering of the material. After cooling to 1330C, the material was then subjected to oxygen for about 4 h to produce voids in the sintered structure. The pores were formed from graphite oxidizing and leaving the structure as CO2 or CO gas. Product gas was monitored via GC, and procedure was carried out until there was O2 breakthrough and no CO or CO2 present. Co Precipitation For a 10 g sample of 10% Coferrite (CoFe2O4) in 8 mol% YSZ sample, 1.24 g of cobal t (III) nitrate hexahydrate, 3.44 g of iron (III) nitrate nonahydrate, 4.20 g of yttrium (III) nitrate hexahydrate and 41.62 g of zirconyl nitrate solution were dissolved in deionized water. Dilute aqueous ammonia was then added gradually dropwise at 60 C to the mixture solutions, with vigorous stirring; unti The precipitate was then washed and removed from the supernatant liquid. The supernatant liquid was checked for clarity. The precipitate was then oven dried at 60 C for ~24 h. Experimental Procedure (5 g S ample) The 5 g sample was subjected to several experiments to ascertain the effectiveness of thermal reduction under vacuum. Typical pressure readings during vacuum reduction were 100200 mTorr (0.1 0.2 Torr). A 115 sccm purge of argon and 1 sccm of helium were used during inert purge reduction. During oxidation, 0.1 mL/min of water were injected via syringe pump with 115 sccm argon. A condenser/water trap was used to condense excess steam. The 1 sccm flow of helium served as an internal

PAGE 21

21 calibration standard for the m ass spectrometer, which sampled the gas environment after the water trap. Both dynamic and isothermal temperature profiles were used. Experim ental Procedure (18g S ample) After structure formation, the material was put through testing at thermal reduction t emperatures ranging from 1350 to 1500C, with exposure to either a 100 sccm inert purge (50 sccm Ar and 50 sccm He) or vacuum conditions. Helium was used as the internal calibration gas for the mass spectrometer. During vacuum pumping, the vertically oriented system was closed from the top and open to a vacuum pump at the bottom. Temperature was ramped at 10C/min from 1200C to the desired reduction temperature, where it was then held for 12 minutes before cooling at 10C/min back to 1200C. Once a stable gas composition under the 50 sccm argon and 50 sccm helium purge was reached (typically several minutes), distilled H2O was pumped via syringe at 0.2 mL/min to commence oxidation. Results (5g S ample) An investigation was performed to get an estimate of the time scales required for thermal reduction. Furnace temperature was brought up to 1450C at 10C/min. In the 25 minute reduction, temperature was ramped from 1200 to 1450C followed then by oxidation starting at 1450C without any hold. For the two 60 min reductions, temperature was ramped from 1200 to 1450C and then held for an additional 35 minutes before oxidation was started. In both cases, once oxidation was started, temperature was ramped to 1200C at 10C/min. This 60 minute time frame was sufficie nt for near complete conversion in trials where an inert purge was used and O2 production was monitored. As show n in Fig ure 2 4 the reduction reaction is clearly aided by holding the temperature at 1450C for 35 minutes, as the hydrogen yield of the

PAGE 22

22 60 mi nute trials (9.28 and8.23 cc/g) were both over twice that of the 25 minute trial (3.71 cc/g). A comparison of the hydrogen production in the subsequent oxidation step following inert purge and vacuum r eduction is shown in Fig ure 2 5 In each case, the temp erature was increased to 1400C under steam (0.1 mL/min) at a faster ramp rate of 20C/min. This procedure allowed for negligible oxygen production during the ramp period. Once 1400C was reached, the steam was stopped and the sample subjected to either vacuum or iner t purge. As shown in Fig ure 2 6 the O2 production reaches a peak value of roughly 0.08 cc/min/g Ferrite, at which point the partial pressure of O2 w as roughly 0.3 Torr. In Fig ure 2 6 the subsequent H2 production rate and yield for the inert pur ge was found to be greater than that of vacuum, where the vacuum level was between 0.1 Torr downstream of the reactor and 2 Torr upstream of the reactor. This discrepancy between pressures is likely due to the decrease in conductance as gas moves through t he porous structure. Although the vacuum trial showed less yield, it is a good illustration of how vacuum pumping can be an effective way of decreasing the partial pressure of O2 during the thermal reduction step. From both an economic and energy standpoint, vacuum pumping is likely much less costly than inert gas separation. After 18 cycles of varying conditions, the hydrogen production of the 5 g sample was compared to that of the first cycle, which the same conditions of reduction were used (115 sccm arg on purge). The hydrogen produc tion results are shown in Fig ure 2 7 and Fig ure 2 8 respectively. Both the yield and peak rates are comparable. No further cycling was done on the material, and these results are promising in that the material appears as thoug h it did not degrade.

PAGE 23

23 Results (18g S ample) Fig 2 9 shows the results of the specific H2 production rate following reduction at several different temperatures. As expected, higher reduction temperatures lead to higher specific H2 production rates. Fig 2 10 also shows the calculated specific H2 yield (per g Ferrite) for each experiment. In general, the 100 sccm inert purge appears to achieve both higher subsequent H2 production rates and yields than vacuum reduction at the same temperature, with the exception of the trials at 1500C. This finding could mean that the partial pressure of O2 ( pO2) during a 100 sccm inert purge is lower than that achieved during vacuum pumping, thus enabling a greater depth of reduction as governed by thermodynamics. Vacuum pumping resulted in a pressure reading of roughly 0.2 Torr downstream of the reactor and 3 Torr upstream of the reactor. The result of the similar performance of both the 100 sccm inert purge and the vacuum reduction trials at 1500C could be largely due to a si gnificant proportion of the reduction occurring during the long ramp durations required to cycle between 1200C to 1500C; if indeed the system reached a pseudoequilibrium level of reduction, the differences between the two purging and vacuum pumping due to different partial pressures environments would be lessened, if not eliminated.

PAGE 24

24 Figure 21. Graph showing variation of reduction temperature with partial pressure of O2

PAGE 25

25 Fig ure 2 2. Experimental facility for bench scale testing Fig ure 2 3. Pict ure of the bench scale rig

PAGE 26

26 Fig ure 2 4 Hydrogen production following thermal reduction at vacuum holding for 25 and 60 minutes at 1450C.

PAGE 27

27 Figure 25 Comparison subsequent H2 production after inert purge and vacuum environments for a 45 min thermal reduction at 1400C Fig ure 2 6 Oxygen production during a 45 min thermal reduction at 1400

PAGE 28

28 Fig ure 2 7 Hydrogen Production in Cycle 1 and Cycle 18

PAGE 29

29 Fig ure 2 8 Hydrogen production achieved after several reduction schemes

PAGE 30

30 Fig ure 2 9 Spe cific H2 yield

PAGE 31

31 CHAPTER 3 SOLAR REACTOR DESIGN The central idea behind the design of solar reactor is to allow the reactive material to be heated using UFs solar simulator Figure 3 1 The solar simulator has 7 xenon arc lamps, each surrounded by an ellipsoidal mirror that focus es the radiation on a circular spot. This flux is similar in nature to the solar radiation and can be used as a source of heat. The reactor was chosen to be indirectly heated as it eliminates the use of a quartz window at t he aperture. Two different paths were chosen for designing the cavity and lids one based on a 25.4mm ( 1 inch ) reactor tube and other based on a 50.8mm (2 inch) tube. Both of them offer distinct advantages that need to be considered. The 25.8mm reactor is smaller in size and develops less thermal stress. The 50.8mm reactor design can have greater amounts of material with little increase in overall cavity dimensions. Furthermore two reactor designs were developed; one with tubes bent at a 90 angle and the other with concentric tubes (tube within a tube). The reactor was modeled on seven parameters as presented in the following sections. Maximum Service Temperature The maximum service temperature of reactor was chosen to be 1700C. Although the maximum operating temperature is 1500C, it was noted that there may be local hot spots in some zones based on the angles at which light is incident from the simulator thus creating a zone of 200C higher than the average temperature on the reactor tube. Heat Loss Heat convection in the air and radiation to the surroundings were identified as the two major modes of heat loss from the reactor. Surface area and the temperature are

PAGE 32

32 the governing factor for heat loss. In order to reduce the surface area a minimum radius of the reactor was calculated. The temperature of outer surface is directly proportional to the thermal conductivity of cavity material T hus ceramics were reasoned as the material of choice. Also, the tubes were placed along the cavity wall to create a barrier for the radiation. Thermal Shock In order to make the production of syngas efficient, the time required to change the temperatures must be minimized. This meant faster ramp rates going from room temperature to reaction temperatures as well as fas ter heating and cooling when cycling between oxidation and reduction. Higher ramp rates mean high thermal shock due to thermal stresses developed in the material as it expands and contracts. The design specification was such that the material should be abl e to take thermal shocks associated with rates of over 30C/min. Flux D istribution A uniform flux distribution not only provides similar thermal conditions for the reactants in different tubes leading to similar kinetics, but also simplifies the numerical coding of reactor dynamics. A horizontal orientation of the reactor was chosen to that effect as opposed to a vertical orientation which has a non uniform distribution among the tubes. Chemical Stability Since the material is filled inside the tubes, the tube material must be chemically inert to the reactive material. Also, it is essential that the tube material does not react with the reactant or the product gases. This eliminates the use of SiC tubes as it oxidizes during the thermal reduction step by r eacting with oxygen evolved

PAGE 33

33 Ease of Manufacture Complex ceramic components are expensive hence the reactor design was kept simple and modular so that if one component fails it does not affect rest of the reactor. Material Selection Alumina was chosen as the reactor and tube material due to several reasons. It maximizes the internal reflections due to its low emissivity. Also, it is capable of providing high service temperatures of 1700C and has a good resistance to thermal shock. SiC has a better therma l shock capability than alumina but it reacts with ferrites and is unsuitable as t he tube material. Furthermore, alumina is chemically inert to H2, CO2, CO, steam. In experimentation with alumina tubes, cobalt was noticed to diffuse into the tube walls, leaving a blue color attributed to CoAlO4 (cobalt aluminate, also known as cobalt blue). However, a lumina tubes were identified as the best option due to their availability, and since the depth of migration of impurities was estimated to be minimal. For 50.8mm model, a Thermcraft model VF heater was chosen as the cavity. It has heating elements embedded in the ceramic fiber that allow slow heating and cooling to prevent thermal shocks. Aperture Radius Selection The elliptical mirrors focus the radiation on a circular spot of 5cm radius. Thus the aperture of the cavity was chosen as 5 cm. A diameter less than that would capture the rays only partially and greater than 5cm would i ncrease the reflections out of the aperture. Cavity Diameter and L ength C alcula tion The diameter of the cavity was calculated such that the rays do not hit the tube fittings. Figure 3 2 shows the calculation of diameter for a 25.4 mm tube cavity. The

PAGE 34

34 rays are incident in a conical volume of half angle 50 degrees. The thickness of th e cavity was kept within 25.4 mm in order to avoid thermal shock. An extra 13.2 mm gap was provided between tube and inner cavity wall to make tube insertion in the cavity and lids easier. The length of the cavity was calculated such that the rays coming b etween 25 through 50 angle do not hit the back wall ; ( Figure 3 3 ) Reactor C omponents Figures 34 through 311 show the Solidworks model of different components of the reactor for 25.4mm and 50.8 mm tubes. There are 4 components in the reactor. The disc facing the aperture is the front lid. There are holes on the front with center at a calculated distance from the center of the lid. Tubes with go inside these holes. An angular cut of 30 is given at the aperture as shown in Figure 35 to further avoid ray s hitting the lid wall. Behind the lids is an open ended cylinder ( Figure [3 6]) that serves as the cavity. The other end of the cavity is covered with a back lid having equal number of holes as the front lid. 120 Bent T ube Concept In order to avoid havi ng the radiation hitting tubes or fittings another design with tubes bent at 120 ( Fig ures 3 12 through 3 16 ) was considered as an alternative to the straight tube. Instead of coming out the front lid facing the aperture, the tubes in this design come out from the side of the front lid as shown in Figure 312 The back lid design (Figure 3 15) changes with the holes at different distance from the center. This design eliminates the restriction on a minimum diameter. However, it involves a problem during fabrication as it is more likely to develop a crack at the edges during sintering of the ceramic. Also, it is more difficult to assemble as the tubes first have to be inserted in the front lid and then put inside the cavity. While in the straight tube design the tubes

PAGE 35

35 can directly be inserted in the lids once all the holes are aligned. The bent tube concept also makes the sealing easier as the length of the bent portion can be increased which decreases the temperature at the sealing region. Tube within a T ube Another concept for avoiding tubes coming from the front was to have a set of concentric tubes or tube within a tube (Figure 3 17). In this tube design a 6.35mm OD tube was inserted in a 25.4mm tube. The length of the inner tube was kept shorter than the outer tube. The idea was to fl ow the reactant gases in the inner tube which would then enter the outer tube and react with the material located in the annular gap between the two tubes. The product gases, being pushed by reactant gases, would then exit from the outer tube.

PAGE 36

36 Fig ure 3 1 Solar S imulator Figure 32. Minimum diameter calculation (All dimensions in inches) Figure 33. Maximum length calculation (All dimensions in inches)

PAGE 37

37 Fig ure 3 4. 1 Reactor Assembly (All dimensions in inches )

PAGE 38

38 Fig ure 3 5. Front lid for 1 reactor design (All dimensions in inches)

PAGE 39

39 Fig ure 3 6. Back lid for 1 reactor design (All dimensions in inches)

PAGE 40

40 Fig ure 3 7. Cavity for 1 reactor design (All dimensions in inches)

PAGE 41

41 Fig ure 3 8. 1 Tube (All dimensions in inches)

PAGE 42

42 Fig ure 3 9. Front lid for 2 reactor (All dimensions in inches)

PAGE 43

43 Fig ure 3 10. Cavity for 2 reactor design (All dimensions in inches)

PAGE 44

44 Fig ure 3 11. Back lid for 2 reactor design (All dimensions in inches)

PAGE 45

45 Fig ure 3 12. 120 tube design Fig ure 3 13. 120 Bent tube (All dimensions in inches)

PAGE 46

46 Fig ure 3 14. Front lid for 120 bent tube design (All dimensions in inches)

PAGE 47

47 Fig ure 3 15. Back lid for 120 bent tube design (All dimensions in inches)

PAGE 48

48 Fig ure 3 16. Cavity for 120 bent tube design (All dimensions in inches)

PAGE 49

49 Fig ure 3 17. Assembly showing tube within a tube design

PAGE 50

50 CHAPTER 4 THERMAL TESTING OF SOLAR REACTOR Experimental Facility The solar park facility is equipped with a solar simulator that can provi de high heat flux in excess of 4 000 kW/m2. The solar simulator consists of 7 Xenon arc lamps of 6 kW each. The lamps are placed at the focus of 3D ellipsoidal mirrors that focus all the radiations on a single focal s pot. An X Y table ( Figure 41 ) is installed in front of solar simulator to allow the flux targets positions to be controlled remotely through a computer program. The table has approximately an 80 centimeter range parallel to the face of the simulator and a 40 centimeter range in the axial direction. To measure temperatures, 8 K type, 3 B type and 2E type thermocouples were put at different locations. A data acquisition unit (DAQ) from National Instruments is installed in the solar simulator room. The communication with DAQ is done via a Labview program. A shutter assembly (Figure 4 2 ) is also mounted on the X Y table frame to cut off the radiation to bring the temperature down. It can be operated remotely with a controller. On the X Y table a frame was buil t using 4040 extruded aluminum (Figure 4 3 ). On the frame a layer of fire bricks was placed to create a base for the reactor. The lower half of the cavity was put on three layers of insulation blanket and positioned at the front edge of the firebricks. Th e front lid and back lid were then carefully positioned and secured with help of brackets. After aligning the holes of the front and back lid a tube was inserted in the hole as shown in Figure 44 Reactor Thermal T esting In order to find the thermal stability of the tube and cavity material under high heating and cooling rates; the reactor was tested in front of the solar simulator. The Z

PAGE 51

51 position (height) of the reactor was fixed to be on the focal plane with the help of positioning lasers. 4 lamps were turned on in succession after a near steady state temperature was achieved at thermocouple18 each time. Analysis of T ube T emperature A type B thermocouple was placed in the tube resting against the insulation stuffed inside. The tip of thermocouple was 3 outside of the front lid. For the first 30 minutes, with lamp 7 on, the ramp rate was observed to be 16.5 C/min. Lamp 7 directly irradiates back lid and thus there was not a rapi d temperature ramp in the tube ( Figure 4 5 ) As the ramp rate fell to 5C /minute, lamp 2 was turned on. Since lamp 2 directly irradiates on the tube, the temperature increased in excess of 50C/min for first 4 minutes. The rate fell down to ~5C/min in 50 minutes, at which point lamp 6 was turned on. The temperature was noted at 986C in the tube at this point in time. As the ramp rate decreased after few minutes, lamp 4 was turned on. As the rate started to fall, the current was increased from 140 to 151 Amperes. The temperature reached ~1450C in the tube in 3 hours at an aver age rate of 8.05C/min. Lamp 2 and 6 were then turned off to bring the temperature to 1200C. The initial rate of decrease of temperature was 21C/min which slowly decreased with time. The temperature dropped to 1200C at an average rate of 10.41C/min. Lamp 2 and 6 were turned on again to study the effect of cycling on the reactor tube and rest of the reactor components. Analysis of Cavity Temperature Figure 46 shows the variation of the temperature of cavity with respect to time. A thermocouple was put on the top center of the cavity and one side center of the cavity. Due to symmetry of the flux with respect to cavity it was assumed that the temperature

PAGE 52

52 profile on both these thermocouples should remain constant. From the graph it is clearly visible that the temperature did remain same with a minor difference of less than 50 C. Analysis of Back Lid Figure 47 shows the variation of temperature at the center of the back lid. It was expected that it will be the highest temperature as compared to the tube and the cavity since Lamp 7 shines directly on it. But as seen in the graph, the maximum temperature at the back lid was 1280 C which was roughly 170 C less than the tube temperature. Since the back lid was very well insulated, the only reason for it coul d be reradiation in the cavity. Also since Lamp 2 was directly shining on the tube that could also have contributed to increase in the tube temperature as compared to back lid Observations Cracks were observed in the back lid as shown in Figure 4 8 caused due to high rate of heating when lamp 2 was turned on or high rate of cooling when all the lamps were turned off. Material close to the aperture eroded (Figure 4 9 ) due to local hot spots as the focus of the radiation was not exactly aligned with the aper ture. The top half of the cavity shrunk in diameter (Figure 4 10) due to thermal expansion and cooling. The insulation blanket discolored (Figure 4 11) due to oxidation of its impurities. Interestingly, the tube did not crack even with such high thermal sh ock Recommendations Since the 25.4 mm tube performed very well in the thermal test without any failure, it is possible that 50.8 mm tube will perform well without developing any cracks. Using a bigger tube will also increase the efficiency of the reactor since it can hold larger amounts of material. Also, to prevent the fracture of material on the back lid, a 5 mm thick disc made of SiC can be put in front of the back lid as SiC has a better resistance

PAGE 53

53 t o thermal shock as compared to alumina. A framework should be built to support tubes since the end plates will not be able to handle the weight of 29 tubes and material. The diameter of the holes on the lids for the 25.4 mm reactor design should be increased to 28 mm to facilitate easy installation of tubes. The temperature of the outer insulation was more than 200C, thus one more layer of insulation should be wrapped around the reactor to bring down the temperature under 100C. Thermocouples inside the insulations were found misplaced from their intended position thus they should be secured in position by using nichrome wire tied around the aluminum framework. Fig ure 4 1 XY Table

PAGE 54

54 Fig ure 4 2. Shutter in front of the mirrors

PAGE 55

55 Fig ure 4 3. Aluminum framework on the X Y table

PAGE 56

56 Fig ure 4 4. Reactor with alumina tube in the cavity Fig ure 4 5. Variation of tube temperature with time

PAGE 57

57 Fig ure 4 6. Variation of temperatures on top and side of the cavity Fig ure 4 7. Variation of Back lid Temperature

PAGE 58

58 Fig ure 4 8 Crack developed on the Back Lid Crack on Back lid

PAGE 59

59 Fig ure 4 9 Erosion of material at the aperture of front lid

PAGE 60

60 Fig ure 4 10. Picture showing shrinkage of top half of the cavity Shrunk top half

PAGE 61

61 Fig ure 4 11. Discoloration of the insulation Blanket

PAGE 62

62 LIST OF REFERENCES [1] Lichty, P., Liang, X., Muhich, C., Evanko, B., Bingham C., and Weimer, A. W. Atomic layer deposited thin film metal oxides for fuel produc tion in a solar cavity reactor. Internati onal Journal of Hydrogen Energy 2012; 37 : 1 688816894. [2] Romero M., and Steinfeld, A. Concentrating solar thermal power and thermochemical fuels Energy & Environmental Science 2012; 5 : 92349245. [3] Nakamura, T. H ydrogen production from water utilizing solar heat at hightemperatures Solar Energy 1977;19 : 467 475. [4] Kodama, T., and Gokon, N. Thermochernical cycles for hightemperature solar hydrogen production. Chemical Reviews 2007; 107 : 40484077. [5] Fresno, F., Fernandez Saavedra, R., Belen Gomez Mancebo, M., Vidal, A., Sanchez, M., Rucandio, M. I., Queji do, A. J., and Romero, M. Sola r hydrogen production by twostep thermochemical cycles: Evaluation of the activity of commercial ferrites. International Journal of Hydrogen Energy 2009; 34: 29182924. [6] Mehdizadeh, A. M., Klausner, J. F., Barde, A., and Mei, R. Enhancement of thermochem ical hydrogen production using an ironsilica magnetical ly stabilized porous structure. International Journal of Hydrogen Energy 2012; 37 : 89548963. [7] Kodama, T. Nakamuro, Y., and Mizuno, T. A two step thermochemical water splitting by iro n oxide on stabilized zirconia. Journal of Solar Energy Engineering Transactions of the Asme 2006; 128 : 3 7. [8] Gokon, N., Murayama, H., Umeda, J., Hatamachi, T., and Kodama, T. Monoclinic zirconiasupported Fe3O4 for the twostep water splitting thermochemical cycle at high thermal reduction temperatures of 14001600 degrees C. International Journal of Hydrogen Energy 2009; 34 : 12081217. [9] Xu, B. J ., Bhawe, Y., and Davis, M. E. Low temperature, manganese oxidebased, thermo chemical water splitting cycle. Proceedings of t he National Academy of Sciences of the United States of America 2012; 1 09 : 9260 9264. [10] J ones, W. M., and Bowman, M. G. The use of oxides in thermochemical water splitting cycles for sola r heat sources. Cobalt oxides. Hydrogen energy progress IV(Vol. 2) 1 982 [11] Abanades, S., and Chambon, M. CO(2) Dissociation and Upgrading from TwoStep Solar Thermochemical Processes Based on ZnO /Zn and SnO(2)/SnO Redox Pairs. Energy & Fuels 2010;24: 66676674. [12] Lundberg, M. Model calculation on some feasible 2step water splitting processes. International Journal of Hydrogen Energy 1993; 18 : 369 376.

PAGE 63

63 [13] Loutzenhiser, P. G., and Steinfeld, A. Solar syngas production from CO2 and H2O in a twostep thermochemical cycle via Zn/ZnO redox reactions: Thermodynamic cycle analysis. International Journal of Hydrogen Energy 2011; 36 : 12141 12147. [14] Chambon, M., Abanades, S., and Flamant, G. Solar thermal reduction of ZnO and SnO(2): Characterization of the recombination reaction with O(2). Chemical Engineering Science 2010; 65: 3671 3680. [15] Steinfeld, A. Solar hydrogen production via a twostep water splitting thermochemical cycle based on Zn/ZnO redox reactions. International Journal of Hydrogen Energy 2002; 2 7 : 611 619. [16] Chueh, W. C., and Haile, S. M. Ceria as a Thermochemical Reaction Medium for Selectively Generating Syng as or Methane from H2O and CO2. Chemsuschem 2009 ; 2 : 735 739. [17] Furler, P., Scheffe, J., and Steinfeld, A. Two step solar thermochemical cycle for splitting H(2)O and CO(2) via ceria redox reactions: E xperimental investigation with a 3 kW solar reactor. Abstracts of Papers of the American Chemical Society 2011 ; 241 [1 8] Singh, P., and Hegde, M. S. Ce(0.67)Cr(0.33)O(2.11): A New Low Temperature O(2) Evolution Material and H(2) Generation Catalyst by Ther mochemical Splitting of Water. Chemistry of Materials 2010 ; 22 : 762 768. [19] Abanades, S., and Flamant, G. Thermochemical hydrogen production from a twostep solar driven water splitting cycle based on cerium oxides. Solar Energy 2006; 80: 16111623. [20] Sch effe, J. R., Li, J., and Weimer, A. W. A spinel ferrite/hercyni te water splitting redox cycle. International Journal of Hydrogen Energy 2010; 35 : 3333 3340. [21] Scheffe, J. R., Allendorf, M. D., Coker, E. N., Jacobs, B. W., McDa niel, A. H., and Weimer, A. W Hydrogen Production via Chemical Looping Redox Cycles Using Atomic Layer DepositionSynthesized Iron Oxide and Cob alt Ferrites. Chemistry of Materials 2011; 23 : 20302038. [22] Romero, M., and Steinfeld, A. Concentrating solar thermal power and thermochemi cal fuels. Energy & Environmental Science 2012 ; 5 [23] Roeb, M., Sattler, C., Kluser, R., Monnerie, N., de Oliveira, L., Konstandopoulos, A. G., Agrafiotis, C., Zaspalis, V. T., Nalbandian, L., Steele, A., and Stobbe, P. Solar hydrogen production by a twostep cycle based on mixed iron oxides. Journal of Solar Energy Engineering Transactions of the Asme 2006; 1 28 [24] Diver, R. B., Miller, J. E., Allendorf, M. D., S iegel, N. P., and Hogan, R. E. Solar thermochemical water splitting ferrite cycle heat engines. Journal of Solar Energy Engineering Transactions of the Asme 2008; 130.

PAGE 64

64 [25] Dahl, J. K., Buechler, K. J., Weimer, A. W., Le wandowski, A., and Bingham, C. Solar thermal dissociation of methane in a f luid wall aerosol flow reactor. International Journal o f Hydrogen Energy 2004; 29 : 725 736. [26] Haueter, P., Moeller, S., Palumbo R., and Steinfeld, A. The production of zinc by thermal dissociation of zinc oxide Solar chemical reactor design. Solar Energy 1999; 67: 161 167. [27] Meier, A., Bonaldi, E., Celia, G. M., and Lipinski, W. Multitube rotary kiln for the indus trial solar production of lime. Journal of Solar Energy Engineering Transactions of the Asme 2005 ; 127

PAGE 65

65 BIOGRAPHICAL SKETCH Rishi Mishra was born in Bihar, India in 1986. He received his Bachelor of Science in M echanical E ngineering from Birla Institute of Technology in 2008. He joined Suzuki Powertrain India Limited after the undergraduate studies as an equipment maintenance engineer. Rishi was admitted to the University of Florida in the fall of 2011. He started working on this project in January 2012. His work involved the design solar reactor and bench scale testing of reactive materials He received his MS from the University of Florida in the spring of 2013