<%BANNER%>

High Temperature Synthesis of Cerium Sulfides and Kinetic Modeling

Permanent Link: http://ufdc.ufl.edu/UFE0010471/00001

Material Information

Title: High Temperature Synthesis of Cerium Sulfides and Kinetic Modeling
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010471:00001

Permanent Link: http://ufdc.ufl.edu/UFE0010471/00001

Material Information

Title: High Temperature Synthesis of Cerium Sulfides and Kinetic Modeling
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0010471:00001


This item has the following downloads:


Full Text

PAGE 1

HIGH TEMPERATURE SYNTHESIS OF CERIUM SULFIDES AND KINETIC MODELING By KEVIN GIBBARD A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Kevin Gibbard

PAGE 3

iii ACKNOWLEDGMENTS The author would like to ta ke this opportunity to th ank those who gave their support during my time at the Un iversity of Florida both pe rsonally and professionally. Thanks go to Dr. David Kolman of the Los Alamos National Laboratory (LANL) for financial support and project ov ersight. The University of Florida Materials Science and Engineering Department also provided s upport and necessary laboratory facilities. Characterization analysis was done using instruments at the Major Analytical Instrumentation Center (MAIC) whic h was essential to this project. In addition the author thanks Dr. Kerry Allahar for the guidance and help he provided during the duration of this project. Most impor tantly, the thoughtfulness and leadership of committee chairman Dr. Darryl Bu tt were instrumental in the success of this project. The author would also like to thank th e members of Dr. Butt’s research group for their help and support in ways to various to list: Abby Queale, Matt Hofrichter, Edgardo Pabit, Jairaj Payyapilly, Soraya Benitez, Samantha Crane, JongSang Lee, and particularly Steven Crane who helped tremendously dur ing the early stages of the project. Finally I thank my family for their love and support, especially my fiance Jennifer who helped keep me from getting overwhelmed on a day to day basis.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii TABLE.......................................................................................................................... .....vi LIST OF FIGURES..........................................................................................................vii ABSTRACT....................................................................................................................... ix CHAPTER 1 INTRODUCTION........................................................................................................1 Properties of the Cerium Sulfides.................................................................................2 Ce2S3......................................................................................................................2 Ce3S4......................................................................................................................3 CeS........................................................................................................................3 Research Objective.......................................................................................................4 2 LITERATURE REVIEW.............................................................................................5 Cerium Monosulfide Production..................................................................................5 Aluminothermic Reduction...................................................................................5 Carbothermic Reduction........................................................................................6 Electrolytic Reduction...........................................................................................6 Reduction by Cerium Hydride...............................................................................7 Cerium Sesquisulfide Production.................................................................................8 Sulfidation by CS2.................................................................................................8 Sulfidation by H2S.................................................................................................8 Thermodynamic Modeling...........................................................................................9 Cerium Sesquisulfide Production........................................................................10 Cerium Monosulfide Production.........................................................................10 3 EXPERIMENTAL APPARATUS DESIGN..............................................................11 Ceria Sulfidation Reactor...........................................................................................11 Cerium Hydrogenation Reactor..................................................................................14 Cerium Monosulfide Reactor.....................................................................................16

PAGE 5

v 4 EXPERIMENTAL PROCEDURES...........................................................................19 Ceria Sulfidation Procedure........................................................................................19 Cerium Hydrogenation Procedure..............................................................................21 Cerium Monosulfide Production Procedure...............................................................23 5 EXPERIMENTAL RESULTS...................................................................................26 X-Ray Diffraction Analysis........................................................................................26 Cerium Sesquisulfide Results..............................................................................26 Cerium Hydride Results......................................................................................29 Cerium Monosulfide Results...............................................................................31 Qualitative X-ray Diffraction Analysis........................................................31 Quantitative X-ray Diffraction Analysis......................................................33 Scanning Electron Microscopy Analysis....................................................................34 6 KINETIC ANALYSIS................................................................................................38 Kinetic Data................................................................................................................38 Kinetics Discussion....................................................................................................40 7 CONCLUSIONS........................................................................................................44 LIST OF REFERENCES...................................................................................................46 BIOGRAPHICAL SKETCH.............................................................................................48

PAGE 6

vi TABLE Table page 6-1 Cerium monosulfi de rate constants............................................................................40

PAGE 7

vii LIST OF FIGURES Figure page 1-1 Cerium-sulfur phase diagram.......................................................................................1 2-1 Overall process flow.....................................................................................................9 3-1 Photograph of ceria sulfidation reactor......................................................................12 3-2 Photograph of the induction power suppl y and cooling water system used for the ceria sulfidation and cerium monosulfide reactors.....................................................13 3-3 Schematic diagram of the ceria sulfidation reactor column interior..........................14 3-4 Photograph of glove box contai ning cerium hydrogenation reactor..........................15 3-5 Photograph of ceriu m hydrogenation reactor.............................................................16 3-6 Photograph of glove box used fo r cerium monosulfide reactor setup........................17 3-7 Schematic diagram of the cerium monosulfide reactor column interior....................18 3-8 Photograph of cerium monosulfide reactor................................................................18 5-1 XRD scan of cerium sesquisu lfide center and edge samples.....................................28 5-2 XRD scan of cerium sesquisulfide cen ter and edge samples when flowing hydrogen sulfide before heating.................................................................................................28 5-3 XRD scans of cerium sesquisulfide samples before and after purification................29 5-4 XRD scan of cerium hydride sample..........................................................................31 5-5 XRD scans of cerium monosulfide samples...............................................................32 5-6 Calibration curve fo r quantitative analysis.................................................................34 5-7 SEM micrograph of the re actant powder at 600X, 50 wt% Ce2S3 and 50 wt% CeH235 5-8 SEM micrograph of the re actant powder at 2500X, 50 wt% Ce2S3 and 50 wt% CeH236 5-9 SEM micrograph of CeS powder................................................................................37

PAGE 8

viii 6-1 Weight fraction of CeS produced ve rsus time at diffe rent temperatures...................39 6-2 Weight fraction of CeS produced ve rsus reaction temperature at 20 min..................39 6-3 Jander equation F versus time at 1300 C and 1500 C.............................................40 6-4 ln K versus 1/T for ce rium monosulfide production..................................................42

PAGE 9

ix Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science HIGH TEMPERATURE SYNTHESIS OF CERIUM SULFIDES AND KINETIC MODELING By Kevin Gibbard May 2005 Chair: Darryl Butt Major Department: Materials Science and Engineering Research has shown that the sulfides of cerium are poorly understood compounds with a wide range of interesting prop erties. While the four sulfides CeS2, Ce2S3, Ce3S4, and CeS show a wide range of potentia l applications from high temperature semiconductors to pigments, the most promisi ng application is in th e area of refractory crucibles for molten metal processing. Ceri um monosulfide (CeS) is the sulfide best suited for this application and a synthesis procedure for this material was devised. Reactors were built to produce th e reactants required to synthesize cerium monosulfide. High purity cerium sesquisulfide (Ce2S3) was produced by reacting ceria (CeO2) and hydrogen sulfide (H2S) in an induction furnace. Cerium hydride (CeH2) was synthesized from cerium metal and hydrogen gas in a glove box containing an inert environment. These materials were then reacted together in an induction furnace at temperatures above 1700 C to produce cerium monosulfide. X -ray diffraction was used to analyze the samples produced and kinetic studies were done on the cerium monosulfide synthesis

PAGE 10

x reaction. The reaction kinetic s were modeled as a diffusi on-limited reaction and the activation energy for the pro cess was estimated to be 190 kJ/mol cerium monosulfide.

PAGE 11

1 CHAPTER 1 INTRODUCTION Recent research has shown that the su lfides of cerium are poorly understood compounds with a wide range of interesting pr operties. Four sulfides of cerium exist with chemical compositions CeS2, Ce2S3, Ce3S4, and CeS. These phases can be seen on the phase diagram of the cerium-sulfur syst em in Figure 1-1. Although these compounds are somewhat difficult to produce their in teresting properties and high temperature capabilities warrant further inve stigation of the synt hesis procedures that could be used for their production. Figure 1-1. Cerium-sulfur phase diagram ( Samsonov, 1964 )

PAGE 12

2 Properties of the Cerium Sulfides Three of the four cerium sulfides have b een investigated in the past and have properties that can be used in a wide variety of applica tions. Cerium disulfide (CeS2) has not been studied in detail as it has no know n applications and is not encountered when dealing with the other sulfides. The characteristics of the other sulfides of cerium will be discussed in the following sections. Ce2S3 Cerium sesquisulfide (Ce2S3) has many applications stemming from its high temperature stability and electr onic configuration. It is a semiconductor with a band gap of 2.06 eV and a high Seebeck coefficient ( Ryan et al. 1962 ). In addition, it can be used as a high temperature pigment due to its bright red color ( Marrot et al. 1997 ). Cerium sesquisulfide is also an essential compound th at is often required for the synthesis of cerium monosulfide. The sesquisulfide has other properties that allow it to be considered as a crucible material for processing molten metals. Along with its high temperature capability it has good stability in air up to 600 C and a low coefficient of thermal expansion. This allows for heating rates up to 1000 C/minute. Cerium sesquisulfide and the other cerium sulfides show low reactivity to many reactive molten metals such as alkali metals and actinides ( Hirai et al. 1998 ). Low interaction between the crucible and metal allows for crucibles to be reused, elim inating economic and environmental waste from replacing broken or corroded crucibles. The properties of cerium sesquisulf ide would make it an especially good candidate material for use in crucibles for processing alkali metals ( Hogan, 2002 ).

PAGE 13

3 Ce3S4 Ce3S4 has properties similar to those of the other cerium sulfides having a melting temperature intermediate to those of ceriu m sesquisulfide and cerium monosulfide. It also forms a complete solid solution with cerium sesquisulfide. The thermal shock resistance of Ce3S4 is exceptional exceeding that of cer ium sesquisulfide and shows lower reactivity with molten metals when used as a crucible ( Eastman et al. 1951 ). Despite these desirable characteristi cs cerium monosulfide still has better properties than Ce3S4 for use as a crucible material. CeS Cerium monosulfide (CeS) at 18.6 wt% sulf ur is a cubic crystalline compound with the sodium chloride crystal structure and a brassy-yellow color. It has the best high temperature characteristics of the cerium sulf ides with volatilization occurring only at 2450 C and a vapor pressure of 10-3 mmHg at 1900 C ( Eastman et al. 1950 ). Below 2500 C the vapor is produced as molecule s leaving the stoichiometry of the bulk material intact. It has good thermal shock resistance like the other sulfides and can be cooled at 1,000 C/minut e without fracturing ( Samsonov, 1964 ). While performing better than the other sulfides in many areas cerium monosulfide does have lower oxidation resistance than the others and will oxidize in air above 200 C. This can be avoided however by heating in an oxygen free environment. Cerium monosulfide is an excellent crucible material due to the wetting characteristics it has with other materials. It shows little if any interaction with nearly all metals except for platinum. Molten platinum reacts vigorously with cerium monosulfide to produce the intermetallic compound CePt whic h destroys the crucible. This is true even for actinide metals which are known for heavily corroding the crucibles used to

PAGE 14

4 process and handle them maki ng cerium monosulfide a good mate rial choice for crucibles in this application ( Condon & Holcombe 1977 ). Research Objective Since there is little existing literature on the synthesis of the sulfides of cerium, especially cerium monosulfide, this proj ect was designed to de velop a production method and procedure for cerium monosulfide. In addi tion the kinetics of the synthesis reaction will be studied. This sulfide proved to have the most beneficial properties and potential to be used in an industrial setting. Initial thermodynamic modeling and process selection for the project were done by Hogan in addition to early cerium sesquisulf ide production experiments. This project was continued by designing and constructing the required experimental apparatus then developing the synthesis procedure for cerium monosulfide. Once the cerium monosulfide was successfully produced the kine tics of its synthesis reaction would be investigated.

PAGE 15

5 CHAPTER 2 LITERATURE REVIEW A review of available scientific lit erature was done focusing on the production methods of the various sulfides of cerium and similar rare earth sulfid es. This search was done in order to gain an unde rstanding of the steps and processes involved in the reactions used to synthesize cerium sulfid e. Thermodynamic modeling of the chosen reactions was done previously and will also be discussed. Cerium Monosulfide Production Several notable methods have been found for the production of cerium monosulfide. Each one of these methods i nvolves the reaction of cerium sesquisulfide with another cerium bearing material to produ ce cerium monosulfide. Since all of the methods require cerium sesquisulfide this co mpound is essential to the overall synthesis procedure. Details of the different me thods to produce both cerium monosulfide and cerium sesquisulfide are described below. Aluminothermic Reduction Combining aluminum metal with cerium sesquisulfide and cerium (III) oxide can produce cerium monosulfide when pressed a nd heated to 1600 C under vacuum for 2 hours. The reaction 2 Ce2S3 + Ce2O3 + 2 Al 6 CeS + Al2O3 detailing this process was originally described by Flahaut ( Flahaut, 1956 ). This process is al so used frequently in the production of different rare-earth su lfides such as lanthanum monosulfide, praseodymium monosulfide, ne odymium monosulfide, and samarium monosulfide as well as several others.

PAGE 16

6 While the reaction temperature is relativel y low and the reactants are inexpensive there are several disadvantages to this process. The products are both solids and result in a mixture of cerium monosulfide and aluminum oxide in addition to any excess unreacted aluminum metal. Vacuum heating reduces th e amount of impurities pr esent in the cerium monosulfide but this process also lowers the yield by removing some of the cerium bearing material from the product. Carbothermic Reduction Cerium monosulfide can also be produced from the following reaction as described by Radzikovskaya: Ce2S3 + CeO2 + 2 C 3 CeS + 2 CO ( Radzikovskaya, 1961 ). Carbon is used to reduce th e cerium compounds to cerium monosulfide by heating the finely powered reactants to 1500-1700 C under vacuum. Othe r rare-earth sulfides can be produced using this method including lanthanum monosulfide, praseodymium monosulfide, and neodymium monosulfide. This reaction is also advantageous due to the relatively low temperature used and the increased product purity due to the re moval of carbon and oxygen as gaseous carbon monoxide. Even with the evolution of the gaseous product the amount of impurities remaining in the product is still significant with up to 0.2 % oxygen and 0.3 % carbon present. Electrolytic Reduction Electrolysis can also be used as a me thod to produce cerium monosulfide from cerium sesquisulfide. In 1962 Didchenko and Litz described this process following the reaction: 2 Ce2S3 + 2 CeCl3 6 CeS + 3 Cl2 ( Samsonov, 1964 ). The electrolysis was conducted in a graphite vessel with a mo lybdenum cathode at 800 C using a melt of CeCl5, Ce2S3, Na2S, and eutectic NaCl-KCl.

PAGE 17

7 This process requires vacuum purification when complete to remove excess salts and metallic cerium. Even after purification up to 1 % impurities remain and the stoichiometry of the cerium monosul fide ranges from 0.95-1.00 sulfur. Reduction by Cerium Hydride Cerium hydride can be used to reduce cerium sesquisulfide to cerium sulfide following the reaction: Ce2S3 + CeH2 3 CeS + H2. The experimental conditions indicated for this reaction included a temp erature of 2200 C and a pressure of 10-4 mmHg ( Eastman et al. 1950 ). Only a few minutes were needed for the reaction to proceed to completion. A vari ation of this method was used by Krikorian who produced cerium monosulfide by arc melting Ce2S3 with cerium metal in an argon atmosphere ( Krikorian, 1988 ). Advantages for this process include the use of additional cerium bearing material in the process to increase yield and the high purit y of the resulting cerium sulfide. A slight excess of cerium hydride is used when comb ining the reactants to eliminate impurities like oxygen from the final product. The r eaction also takes place quickly, going to completion in only a few minutes. Problems w ith this reaction pro cess include the high reaction temperature and the production a nd handling of the cerium hydride which oxidizes readily when exposed to air. A glove box containing an inert gas atmosphere is required during these steps to prevent the de composition of the cerium hydride to various cerium oxides. This method of cerium sulfide producti on was chosen due to the high purity achieved. It also does not require any furt her purification steps which cause a loss of product and decreased reaction yield. Additi onally, only a few minutes are required at

PAGE 18

8 high temperature for the reaction to go to comp letion as compared to the hours needed for the other types of reduction processes. Cerium Sesquisulfide Production As mentioned before, all of the methods used for the production of cerium monosulfide involve the reducti on of cerium sesquisulfide. Synthesis of this compound is then also important to the overall pr oduction of cerium monosulfide. Two major methods have been found to produce cerium sesquisulfide from ceria and they are detailed in the sections below. Sulfidation by CS2 Carbon sulfide can be used to produce cerium sesquisulfide according to the reaction: 6 CeO2 + 5 CS2 3 Ce2S3 + 5 CO2 + SO2 ( Hirai et al. 1998 ). Hirai noted that graphite powder was also incl uded with the reactants to ai d in conversion. While this process proceeds at temperatures as low as 700 C it can take up to 8 hours for the cerium sesquisulfide to be prod uced in large amounts. The low reaction temperature required for th is reaction process is desirable but the long reaction time required is disadvantageous. The final product from this method also tends to have significant amounts of carbon and oxygen impurities present which eliminate it as a useful procedure. Sulfidation by H2S In 1950 Eastman produced cerium sesquisu lfide through the reduction of ceria by hydrogen sulfide. It was speculated that the process involved more th an one reaction step and followed the equations: 6 CeO2 + 4 H2S 3 Ce2O2S + 4 H2O + SO2 followed by Ce2O2S + 2 H2S + 2 C Ce2S3 + 2 CO + 2 H2. Eastman also indi cated that the second reaction could involve a gaseous CS species as another intermediate. The reaction

PAGE 19

9 required at least 2 hours at a temperat ure of 1500-1700 C to produce the cerium sesquisulfide. This process produced high purity cerium ses quisulfide in a much shorter time than the alternative process. The drawbacks of this method include the use of corrosive hydrogen sulfide gas and higher reaction temperat ures. It was ultimately decided to use this process due to the low amount of im purities in the product produced and the short reaction time required. A flow diagram of th e overall process flow that was chosen can be seen in Figure 2-1. Figure 2-1. Overall process flow Thermodynamic Modeling The thermodynamic modeling of the sele cted synthesis reactions was done previously ( Hogan, 2002 ). Each of the three synthesis reactions was examined to verify the experimental conditions f ound in literature to ensure th at the desired compounds were thermodynamically stable. The sulfide reac tions were found to have complications involving impurities that remained in the pr oduct depending on the reactants provided during the reaction. 6 CeO2 + 4 H2S 3 Ce2O2S + 4 H2O + SO2 2 H2S + 2 C + Ce2O2S Ce2S3 + 2 CO + 2 H2 CeH2 + Ce2S3 3 CeS + H2 Ce + H2 CeH2

PAGE 20

10 Cerium Sesquisulfide Production It was found that oxygen impurities present du ring both sulfide reactions would be retained in the sample as Ce2O2S. When synthesizing cerium sesquisulfide it was also important to include an excess of hydrogen su lfide. If there was inadequate hydrogen sulfide many forms of cerium oxide woul d remain in the product such as CeO2 and Ce6O11 in addition to Ce2O2S. Cerium Monosulfide Production The cerium monosulfide synthesis reaction was found to be very sensitive to oxygen in the system. Even small amounts of oxygen remaining in the reaction vessel would be converted to Ce2O2S during the reaction. These im purities could be avoided if the cerium sesquisulfide was r eacted with a large excess of cerium hydride at very high temperatures above 2300 C to remove them.

PAGE 21

11 CHAPTER 3 EXPERIMENTAL APPARATUS DESIGN The method chosen for the production of cerium monosulfide involved three separate processes each needing its own specifi cally designed reactor. Reactors for each of these processes were designed based on e xperimental conditions found in literature and available equipment used in the previous work done. Small reac tors were developed for simplicity since there was only a need to produce a few grams of the cerium monosulfide at one time. For the first reaction cerium sesquisulf ide was made by the reduction of ceria by hydrogen sulfide at 1500 C in the pres ence of carbon. Next cerium hydride was produced from the reaction of hydrogen gas w ith cerium metal at 100 C in a vacuum purged tube furnace. The two cerium compoun ds that were produced were then mixed and reacted together at 1 800 C and a pressure of 10-2 Torr to synthesize cerium monosulfide. The details of the reactor for each of these processes are described in the following sections. Ceria Sulfidation Reactor The original design used by Hogan to pr oduce cerium sesquisulfide used molten sulfur and flowing hydrogen gas to produce hydrogen sulfide ( Hogan, 2002 ). A 2 inch inner diameter glass column was used as the reaction vessel. Modifi cations were made to the original reactor design in order to increa se the purity of the pr oduct and the efficiency of the reaction. The hydrogen sulfide producti on vessel was replaced with a tank of pure

PAGE 22

12 H2S and a 1 inch inner diameter reactor colu mn was used in place of the 2 inch inner diameter column. The ceria sulfidation reaction was housed in the 1 inch inner diameter jacketed chromatographic column with a 2 inch outer diameter. Cooling water was circulated through the jacketing to preven t overheating of the colum n. The reactor column was situated in the fume hood insi de of a three turn induction co il made from .25 inch copper tubing. A 5 kilowatt induction power supply and its associated cooling water system were situated next to the fu me hood. Pictures of the experi mental apparatus are shown in Figures 3-1 and 3-2. Figure 3-1. Photograph of ceria sulfidation reactor The ends of the reactor column were sealed with PTFE caps that contained polymeric o-ring seals that fit against the gl ass reactor column. The end caps contained threaded channels through them allowing for stainless steel gas lines to be attached at

PAGE 23

13 both ends. A .375 inch stainless steel sight tube was connected to the top end cap to give a clear line of sight for temperature measur ement by a pyrometer and a .25 inch stainless steel gas line was used to bring gasses into the column through a T-joint. The sight tube was fitted with a carbon tip to enhance the r eaction since it is belie ved that the reaction may proceed through a gaseous ca rbon sulfide intermediate ( Eastman et al. 1950 ). The bottom end cap contained .25 inch stainless steel fittings to allow for gas to be exhausted into a cold water bath before being ve nted into the back of the fume hood. Figure 3-2. Photograph of the induction power supply and cooling water system used for the ceria sulfidation and cerium monosulfide reactors Crucibles to hold the reactants were made from .625 inch diameter carbon rods which were cut into .5 inch segments and had a .375 inch hole drilled in the center. Each crucible was surrounded with zirconia felt for insulation when it was used. A stand made from rolled zirconia felt was used to hold the sample in the center of the reaction column

PAGE 24

14 where it would be surrounded by the induction co il. Several layers of carbon felt were also used as insulation on above and below the crucible. A schematic diagram of the setup inside the column is shown in Figure 3-3. Figure 3-3. Schematic diagram of the ceria sulfidation reactor column interior Cerium Hydrogenation Reactor Both cerium metal and cerium hydride ar e reactive in ai r and will oxidize spontaneously to produce cerium oxides when exposed. In order to prevent this from taking place it was necessary to construct a gl ove box where an inert atmosphere could be maintained when handling these two materials. The glove box used an argon atmosphere and was connected to an oxygen analyzer to quantify the amount of oxygen inside the glove box while handlin g reactive species. The reacting gas used in the procedure was pure hydrogen so a stainless steel dilution chamber was set up inside the glove box where the exhaust gas from the reactor could be mixed with argon before being vented out of the system. This was a modification made from initial experiments that used 5 % hydrogen in argon reacting gas. Kinetic data showed that this setup did not provide sufficient hydrogen for the reaction to proceed at an acceptable rate ( Brill et al. 1995, Sarussi et al. 1993 ). A dual stage rotary vane vacuum pump was used when purging the tube furnace and was also needed when Crucible Zirconia Gra p hite

PAGE 25

15 cycling gas through the glove box to reduce the oxygen concentration present. A picture of the glove box is shown in Figure 3-4. Figure 3-4. Photograph of glove box containing cerium hydrogenation reactor The glove box contained the equipment n ecessary to prepare the cerium metal before using it as well as a small tube furnace where the reaction took place. The furnace used a 1 inch diameter stainless steel tube w ith fittings to .25 inch st ainless steel tubes at each end for the entry and exhaust of reaction gasses. A copper boat was made to hold the cerium metal while in the tube furnace wi th a copper retaining wire to move the boat in and out of the furnace. A picture of the tube furnace and the exhaust mixing chamber is shown in Figure 3-5.

PAGE 26

16 Figure 3-5. Photograph of cerium hydrogenation reactor Cerium Monosulfide Reactor The production of cerium monosulfide was done in a setup similar to that of the ceria sulfidation reactor. Another jacketed chromatographic column was used and placed in the same induction system that was used in the cerium sesquisu lfide production. In this setup the bottom end cap of the reacto r was a solid PTFE plug and the top end cap contained a stainless steel valve to isolate th e contents of the column from atmosphere. The addition of the valve allowed for an iner t argon atmosphere inside the column to be maintained while it was moved and connected to a vacuum pump during the setup of the experiment. In order to keep the cerium hydride in an inert atmos phere throughout the preparation of the r eactor a small glove box was set up where the cerium sesquisulfide and cerium hydride could be handled. This glove box used an argon atmosphere and

PAGE 27

17 contained a microbalance and mortar and pest le for the measuring and mixing of the two compounds before loading the mixture into reac tor columns. Two columns were able to be prepared at one time in order to re duce the amount of time spent purging the oxygen out of the glove box. A picture of the small preparation glove box can be seen in Figure 3-6. Figure 3-6. Photograph of glove box used for cerium monosulfide reactor setup The vacuum condition used during this reac tion allowed for less insulation to be used inside the column. Zirconia felt was again used as a stand for the molybdenum or molybdenum-lined carbon crucible which was th en wrapped with more zirconia felt. A small ring of carbon felt was also included above the crucible to act as an oxygen getter in an effort to reduce the effects of any oxygen remaining in the system. A schematic of the setup inside the reactor column can be s een in Figure 3-7 and a picture of the cerium monosulfide reactor can be seen in Figure 3-8.

PAGE 28

18 Figure 3-7. Schematic diagram of the ceriu m monosulfide reactor column interior Figure 3-8. Photograph of cerium monosulfide reactor Crucible Zirconia Gra p hite

PAGE 29

19 CHAPTER 4 EXPERIMENTAL PROCEDURES The three individual reactions used during the scope of this project each involved the development of a detailed experiment al procedure to accompany the new reactor design for each process. Each proce dure was refined over the course of many experiments in order to determine the most efficient method available for the production of each material. The finalized procedures th at were used are detailed in the sections below. Ceria Sulfidation Procedure To begin, 2.0 grams of cerium oxide powder and .14 grams of carbon powder were weighed and placed into a mortar and pestle and ground together to thoroughly mix the two reactants. Reactions done without car bon powder present did not go to completion and only resulted in the intermediate compound Ce2O2S ( Samsonov, 1964 ). Approximately 1.75 grams of the mixture was th en weighed into a graphite crucible and the crucible was then wrapped in two spirals of 2 millimeter thick zirconia felt. Three layers of .5 inch thick graphite felt with a hole in the center were placed on top of the crucible and two layers of solid graphite felt were placed under it. The entire assembly of the crucible and insulation was then wrapped in enough zirconia felt so that it fit snugly into the chromatographic column. This assembly was placed in the reactor column so that it fit against another spiral of zirconia felt acting as a stand located in the bottom of the column.

PAGE 30

20 The sight tube assembly was then screwed into the top of the column making sure that the tip of the tube did not touch the t op of the insulation. Next the gas supply was connected and the cooling water systems fo r the reactor column and induction system were turned on. The alignment of the induction coil was also ch ecked to make sure that it was at the same level as the crucible and was not touching itself or th e glass of the reactor column. Next the hydrogen sulfide supply valve and reactor exhaust valves were opened for one minute allowing any residual oxygen in th e hydrogen sulfide line to flow out of the system. The hydrogen sulfide supply valve wa s then closed and the argon supply valve was opened allowing argon to flow for approxi mately ten minutes purging the reactor. The pyrometer was then aligned with the cont ents of the crucible and the induction coil power supply was plugged in and turned on at the circuit breaker. The argon supply valve was then closed and the hydrogen sulfide supply valve was opened allowing hydrogen sulfide to flow thr ough the reactor for five minut es before the heating was started. The power level was set at 350 a nd the induction coil was started, gradually increasing the power level until the pyrometer read th e desired temperature. The hydrogen sulfide supply valve was closed during the experiment to prevent any intermediate gases from being exhausted out of the system. Eastman proposed that the reaction between Ce2O2S and H2S may involve a gaseous CS intermediate. It was found that if the hydrogen sulfide flowed constant ly during the reaction no cerium sesquisulfide would be formed. This may be due to the in ability of CS to form and react with the Ce2O2S before being driven away by the flow of H2S. To accommodate this effect hydrogen sulfide was flowed through the syst em for five minutes of every hour during

PAGE 31

21 the reaction to maintain a high concentration of the reactant gas in the system without a constant flow of gas. After the desired r eaction time had been reached the induction coil power supply was shut down and the hydrogen sulfide cyli nder was closed and the argon supply valve was opened allowing argon to flow through the reactor. After approximately five minutes the induction coil cooling water was shut off. Fifteen minutes later the reactor cooling water was also turned off and the argon cylinder was closed. The sample was then allowed to cool for one hour before being removed. The gas supply and the bottom plug of the reactor column were removed in order to retrieve the sample assembly containing the crucible and the insulation from the bottom of the column. Once the crucible was removed from the surrounding insula tion its contents were stored in a small glass vial for analysis. Cerium Hydrogenation Procedure The oxygen analyzer was turned on and a llowed to preheat before checking the oxygen content of the glove box. If the oxyge n analyzer read over 300 parts per million oxygen the oxygen analyzer valves were closed and the vacuum pump was turned on. The argon inlet valve and the vacuum valves to the glove box were opened and the valves were adjusted to maintain a slight over pressure while flowing argon through the box. After five to ten minutes of purging with argon and vacuum, the glove box valves were closed and the concentration of oxyge n was checked again. Once the oxygen concentration in the glove box read below 300 parts per million oxygen the vacuum pump was turned off and the vacuum port was switched from the glove box vacuum line to the reactor vacuum line.

PAGE 32

22 Chips of the cerium rod with a diameter of 6.35 millimeter were removed from the storage vial and sanded on all surfaces with 600 grit SiC san dpaper to remove any oxide that may have formed. The cerium pieces were then rinsed with acetone and allowed to dry. The copper rod used as a sample holder was also sanded with 600 grit SiC sandpaper to remove any surface oxide. Each ce rium piece was then placed in one of the holes in the copper rod before inserting the c opper retaining wire and copper rod into the tube of the tube furnace. Care was taken to maintain the holes in the copper rod upright and leave a 1.5 centimeter section of the copper retaining wire extendi ng out of the end of the tube. The end fitting of the tube fu rnace was then screwed in firmly and it was plugged into the internal glove box socket. The furnace was turned on and set to 25 C along with the digital pressu re gauge and vacuum pump. Next the vacuum valve was opened and th e pressure decrease in the reactor was monitored. When the pressure had reached a constant valu e the cooling water inlet and outlet valves were opened and the setpoint on the tube furnace was increased to 300 C. After the reactor had purged for one hour at 300 C the tube furnace setpoint was adjusted to the desired reaction temperature. Once the furnace had cooled to the reacti on temperature the reactor was backfilled with argon and the vacuum pump was then shut off. Once the pressure had increased to atmospheric pressure the hydrogen supply valv es were opened. The hydrogen flow meter was adjusted to achieve the desired flow of hydrogen for the reaction and the argon flow meter for diluting the exhaust gas was opened to a flow rate that gave a ratio of argon to hydrogen in the exhaust mixing chamber of 20:1. The reaction was allowed to run for the desired reaction time while monitoring th e temperature in the tube furnace.

PAGE 33

23 Once the reaction had run for the desired am ount of time the setpoint of the tube furnace was decreased to room temperature and the two hydrogen supply valves and the hydrogen flow meter were closed. The argon fl ow meter was also closed before opening the argon valves to purge the reactor. On ce the tube furnace had cooled to room temperature the furnace was turned off and unplugged and the cooling water inlet and outlet valves were closed. The purge argon was stopped and th e end fitting of the tube furnace was removed. Next the copper retain ing wire was pulled carefully out of the furnace to retrieve the copper rod from inside The samples were then removed from the holes in the copper rod and stored in a glass vial for analysis. Cerium Monosulfide Production Procedure The smaller glove box was used for the r eactor setup in this experiment. Two reaction columns were prepared together by inserting a roll of 2 mm thick zirconia felt into the top of the column with a round layer flat on top of th e roll to create a stand. Two molybdenum crucibles were then wrapped in enough 2 mm thick zirconia felt to fit snugly inside a column and were placed on th e felt stands in the top of both reactor columns. For experiments requiring temperatures above 1400 C the molybdenum crucibles were substituted with carbon crucibles contai ning an internal molybdenum lining made from .25 mm thick molybdenum foil. This was required due to the weight and wall thickness of the molybdenum crucibles which kept them from being heated above 1400 C at full power. The molybdenum lining was also important to keep any cerium monosulfide from coming into contact with carbon above 1900 C above which the cerium monosulfide would react to form cerium carbide.

PAGE 34

24 The columns were placed in the glove box through the side access door along with their end fittings, additional zirconia felt and the cerium se squisulfide and cerium hydride samples. The glove box doors were then clos ed and the box exhaust line was connected to the oxygen analyzer inlet line. The oxyge n analyzer was turned on and allowed to preheat and reach operating temperature. The glove box valves were then opened and argon was allowed to flow through the box until the oxygen analyzer read below 300 parts per million oxygen. At this point all valves were closed and the cerium sesqui sulfide and cerium hydride were weighed out and placed into a mo rtar and pestle. They were then ground together to thoroughly mix the two reactants. The mixture was then weighed and .5 grams of the mixture was loaded into each of the two crucibles. The crucibles and insulation were pushed down into the column s until the zirconia felt stand reached the bottom of the column and the bottom end fittings were inserted and tightened. The top fittings of the columns were then inserted a nd tightened with the included valve in the closed position. The top door of the glove box was then opened allowing for the columns to be removed vertically. For an individual run a reactor column wa s removed from the glove box and taken to the fume hood where it was inserted into th e induction coil and attached to the stand, pyrometer sight glass, vacuum line, and c ooling water lines. The vacuum pump and digital vacuum gauge were then turned on. Once the pressure reach ed a constant value the reactor vacuum valve was opened. Afte r allowing the vacuum gauge to stabilize again the reactor column cooling water valves were opened and the cooling water for the induction coil was also started. The alignmen t of the induction coil was also checked to

PAGE 35

25 make sure that it was at the same level as the crucible and was not touching the glass of the reactor column. The pyrometer was then a ligned with the contents of the crucible and the hood was moved to its lowest possible po sition. Next the induc tion coil power supply was turned on and set to full power. The i nduction coil was started and the power level was reduced to maintain the desired reaction temperature. Once the reaction time was up the induction coil power supply and the vacuum pump were turned off after closing the reacto r valve. After approximately five minutes the induction coil coolin g water was shut off and unplugge d. Fifteen minutes later the reactor cooling water was turned off. The sample was then allowed to cool for one hour before being removed. Next the vacuum lin e and the bottom plug of the reactor column were removed. The crucible and insulation were then removed from the bottom of the column and the sample was stored in a small glass vial for analysis.

PAGE 36

26 CHAPTER 5 EXPERIMENTAL RESULTS The samples produced in the experiments conducted were analyzed through the use of x-ray diffraction and scanning electron microscopy. X-ray diffraction (XRD) was used because of its sensitivity to the phase s present and the relative amounts of these phases. Both qualitative and quantitative studies were done on various samples. Scanning electron microscopy (SEM) was used to produce images of the powders made during the experiments and to determine the sizes of individual particles in those powders. X-Ray Diffraction Analysis Samples were prepared for x-ray diffraction analysis by placing the sample in a mortar and pestle and grinding the material in to a fine powder. A glass slide was labeled with the designation of the sample and a pi ece of double sided tape approximately 1 cm square was placed in the center of the slide. A portion of the sample powder was placed on the tape and spread with a small spatula in order to cover the entire area of the tape. Samples were typically analyzed w ithin four days of being produced. Cerium Sesquisulfide Results Many experiments were run focusing on the production of cerium sesquisulfide samples that were homogeneous and free from oxygen impurities during the development of the experimental procedure for the ceria sulfidation reactor. Initially the samples produced were heterogeneous and containe d large amounts of oxygen impurities in the form of cerium thiosulfide. Though cerium sesquisulfide was also present in these

PAGE 37

27 samples the cerium thiosulfide dominated the x-ray diffraction patterns and was concentrated towards the center of the sample. To determine the homogeneity of the sa mples, pellets of solid product were removed from the carbon crucibles, encas ed in epoxy, and cross sectioned. X-ray diffraction analysis was then done on powders removed from the cente r and edge regions of the pellet. In samples where the hydrogen sulfide flow was not started until the sample reached its reaction temperature the peak intens ities showed that the edges of the pellet contained more cerium sesquisulfide than th e center. X-ray diffraction scans taken from a sample produced by this method can be s een in Figure 5-1. The x-ray diffraction patterns obtained when the hydrogen sulfide flow was started before heating the sample show similar peak intensities between both compounds indica ting that the samples were more homogeneous. Figure 5-2 contains xray diffraction scans taken from a sample produced while flowing the hydrogen sulfide both before and during the initial heating of the sample. Eastman described the reaction of ceria and hydrogen sulfide in the absence of carbon and showed that under these circum stances cerium thiosulfide was the only product phase with no production of cerium sesqui sulfide. From this observation it was decided to increase the carbon content available to react with the ceria powder. Powdered carbon black was mixed with the ce ria powder before bei ng loaded into the crucibles and heated. Samples produced that contained oxygen impurities were also able to be purified by adding carbon and rerunni ng the samples at 1500 C under flowing hydrogen sulfide. The x-ray diffraction scans taken from cerium sesquisulfide samples before and after purification are shown in Figure 5-3.

PAGE 38

28 Ce2S3-0127040 100 200 300 400 500 600 2030405060708090 Angle (2theta)Intensity (counts) Edge Center Figure 5-1. XRD scan of cerium sesquisu lfide center and edge samples with Ce2S3 peaks labeled Ce2S3-0128040 100 200 300 400 500 600 2030405060708090 Angle (2theta)Intensity (counts) Edge Center Figure 5-2. XRD scan of cerium sesquisulfid e center and edge samples when flowing hydrogen sulfide before heating with Ce2S3 peaks labeled Ce2S3 Peaks Ce2S3 Peaks

PAGE 39

29 Ce2S3-0719040 50 100 150 200 250 300 350 400 450 2025303540455055606570758085 Angle (2theta)Intensity (counts) Before Purification After Purification Figure 5-3. XRD scans of cerium sesquisulf ide samples before and after purification Samples produced after incr easing the carbon content in th e crucible were shown to be cerium sesquisulfide with its characteristic re d color. Even if a slight stoichiometric surplus of carbon was added to the reactants no carbon peaks were observed in the resulting x-ray diffraction spectra. This ma y indicate that excess carbon is reacted with oxygen present to produce gaseous carbon mono xide or with the hydrogen sulfide to produce a gaseous car bon sulfide compound. Cerium Hydride Results Cerium hydride is a non-stoichiometric compound with a chemical formula of CeHx where x varies from 2 to 3. This compou nd shows a large extent of disorder in the crystal lattice and the amount of hydrogen inco rporated can cause shifts in the peak locations and changes their intensities. Thes e factors as well as th e reactivity of cerium hydride in air complicated the use of x-ray di ffraction as a character ization technique.

PAGE 40

30 Cerium hydride samples were analyzed within minutes of being removed from the inert atmosphere they were stored in allowing fo r the composition to be analyzed before a large extent of oxi dation could occur. Initial experiments done using only 5% hydr ogen gas in argon showed the presence of cerium metal and its oxides when analyzed. A kinetic analysis was done on the rate of hydrogenation using data obtained by Brill and showed that the partial pressure of hydrogen was too low for the reaction to proceed at an acceptable rate. The 5% hydrogen gas was replaced with pur e hydrogen in the experimental setup to speed up the kinetics of the reaction. The resulting samples that were produced took a black rough appearance and the x-ray spectra showed peaks indicating the presence of cerium hydride. An x-ray diffraction scan taken from a cerium hydride sample can be seen in Figure 5-4. The weak intensity of the peaks and the large amount of noise in the spectrum is indicative of the hi ghly disordered structure of cerium hydride. If samples were allowed to sit for several weeks, even in an environment with a low oxygen concentration, the resulting powder took on the characterist ic yellow color of cerium oxide. This observation was also ve rified by x-ray diffraction spectra. The cerium hydride was also analyzed to determine the stoichiometry of the compound. Samples were weighed and allowed to oxidize in air for many days. Once the samples had taken on a uniform yellow co lor they were reweighed and the difference in mass was used to calculate the amount of hydrogen that had been incorporated into the cerium hydride. Results showed that the ce rium hydride produced had a composition of approximately CeH2.7.

PAGE 41

31 CeH2-0917040 20 40 60 80 100 120 140 2025303540455055606570758085 Angle (2theta)Intensity (counts) Figure 5-4. XRD scan of cerium hydride sample Cerium Monosulfide Results Since the production of cerium monosulfid e was the focus of this project the samples produced were examined using seve ral methods. Qualitative data including xray diffraction patterns were ta ken comparable to the anal ysis of the other compounds produced. When analyzing the cerium monosulfide quantitativ e data was also collected in addition to the qualitative da ta so that the kinetics of the reaction involved could be investigated. Quantitative x-ray diffraction wa s used to determine the amount of cerium monosulfide produced during the experiments done in order to relate that amount to the degree of completion of the reaction. Qualitative X-ray Diffraction Analysis Samples taken from the cerium monosulfide reactor were a brassy yellow color and were shown be pure cerium monosulfide when the reaction was allowed to go to

PAGE 42

32 completion. X-ray diffraction scans taken from two different cerium monosulfide samples are shown in Figure 5-5. The tw o samples of cerium monosulfide were synthesized from independently produced samples of both cerium sesquisulfide and cerium hydride. CeS Samples 0 50 100 150 200 250 300 350 400 450 2535455565758595105115 Angle (2theta)Intensity (counts) CeS-102904 CeS-100604 Figure 5-5. XRD scans of cerium monosulfide samples The cerium monosulfide synthesis reaction wa s also shown to be sensitive to the presence of oxygen in the reac tants. If the cerium hydride had oxidized partially before being reacted the resulting cerium monosulfide contains oxygen impuriti es in the form of cerium thiosulfide. The use of carbon crucib les also changed the structure of the final product. To prevent this cr ucibles used for this reacti on were either molybdenum or carbon with a molybdenum foil lining to isolate the contents of the crucible from the carbon.

PAGE 43

33 The experiments run to analyze the kinetic s of the cerium monosulfide production reaction were done at lower temperatures a nd shorter times than those required for the reaction to finish. The samples produced under these conditions showed peaks from cerium monosulfide as well as cerium ses quisulfide and were analyzed further to determine the extent of the reaction. Quantitative X-ray Diffraction Analysis Quantitative x-ray diffraction was used to de termine the extent of reaction from the relative amounts of the phases present. This analysis was done using the internal standard method where reference samples are made with a known weight of a reference powder before analyzing experimental samples ( Bish & Chipera, 1988, Connolly, 2003 ). This method avoids complications involving th e determination of th e x-ray absorption coefficient and density of the material. Al pha-aluminum oxide was used as a reference material in this experiment because its ma jor diffraction peaks do not overlap with those of any of the encountered cerium compounds. Powders with compositions ranging from 0 to 100 % cerium monosulfide were mixed with equal weights of aluminum oxide to produce reference samples. These samples were then analyzed using x-ray diffrac tion and the ratio of the peak area of the primary cerium monosulfide peak to the area of the primary aluminum oxide peak was calculated. The area ratio was then plotte d as a function of sample composition to determine the calibration constant that would be used to calculate the amount of cerium monosulfide present in samples with an unde termined composition using the equation: (ACeS / AAl2O3) = k (XCeS / XAl2O3). The calibration curve produced can be seen in Figure 5-6.

PAGE 44

34 y = 0.8652x + 0.0281 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 00.20.40.60.81 XCeS / XAl2O3ACeSAAl2O3 Figure 5-6. Calibration curve for quantitative analysis Samples produced experimentally were mixed with an equal mass of aluminum oxide and this mixture was analyzed usi ng x-ray diffraction. The amount of cerium monosulfide present was then determined by comparing the relative intensity of the cerium monosulfide peak to the intensity of the aluminum oxide peak in the sample against the calibration constant calculated from known mixtures in the reference samples. The data taken from this analysis will be discussed further in the Kinetic Analysis chapter. Scanning Electron Microscopy Analysis Scanning electron microscopy was used to examine the reactant powder used to produce the cerium monosulfide and the cerium monosulfide powder produced. After mixing and grinding the cerium sesquisulfid e and cerium hydride samples a portion of the mixture was placed on a carbon glue tab attached to an aluminum mount and stored in a sealed jar. The mount was then taken to the electron microscope and loaded into the sample stage as quickly as possible to pr event any reaction of the powder with the environment. The powder was analyzed to de termine the particle sizes of the different

PAGE 45

35 materials and the homogeneity of the powder. Pictures of the reactant powders can be seen in Figures 5-6 and 5-7. Figure 5-7. SEM micrograph of the r eactant powder at 600X, 50 wt% Ce2S3 and 50 wt% CeH2 The particle size distribution of the powder was determined to be approximately 10 m based on the scanning electron micrographs. Particles appeared to be faceted but shaped roughly like spheres. Though some va riation existed in the size of the particles the value of 10 m for the particle diameter was judge d a reasonable estimate of the size distribution. Pure cerium monosulfide produced from a high temperature run was removed from the crucible and ground with a mortar and pest le to separate the particles. At high temperatures the cerium monosulfide tended to sinter together in to a solid mass that could be broken apart using moderate force in the mortar and pestle. This powder was

PAGE 46

36 also analyzed on a carbon sticky tab on an al uminum mount to determine the size of the powder particles. The resulting micrograph of the cerium monosulfide powder can be seen in Figure 5-8. Figure 5-8. SEM micrograph of the r eactant powder at 2500X, 50 wt% Ce2S3 and 50 wt% CeH2 The particle size distribution of the ce rium monosulfide powder appears to be approximately 10 m. This distribution is sim ilar to the particle size distribution of the reactant powder. The major differen ce between the reactant powder and cerium monosulfide powder is the particle morphology. While the reactant powder appeared to be more spherical the cerium monosulfide pow der is more irregularly shaped with an increased number of facets and edges.

PAGE 47

37 Figure 5-9. SEM micrograph of CeS powder

PAGE 48

38 CHAPTER 6 KINETIC ANALYSIS The data taken from the x-ray diffraction an alysis was used to analyze the kinetics of the reaction of cerium sesquisulfide with cerium hydride to produce cerium monosulfide. Samples were produced times a nd temperatures less than those required for the reaction to go to completion. Experiments were run at several temperatures for 20 minutes and again at 1300 C and 1500 C for longer times. The extent of the reaction was determined from quantitative x-ray di ffraction data and analyzed using existing models for the rate and type of reaction. Kinetic Data The extent of reaction as determined by qua ntitative x-ray diffraction for all of the samples analyzed for kinetics is shown in Figure 6-1. The reaction coordinate was defined as the weight percen t of cerium monosulfide that had been produced during the reaction. Pure cerium monosul fide was obtained from the sa mple run for 20 minutes at 1750 C meaning that the reac tion had gone to completion an d the reaction coordinate was 100 wt% cerium monosulfide. There is no way to determine exactly when the reaction reached this point so subsequent e xperiments were run at lower temperatures where the extent of reaction could be determ ined for the reaction time. The data from experiments run for 20 minutes is shown in gr eater detail in Figure 6-2 as a function of reaction temperature. This curve displays th e increasing rate of reaction as temperature increases.

PAGE 49

39 Reaction Coordinate vs. Time0 10 20 30 40 50 60 70 80 90 100 0102030405060 Time (min.)Reaction Coordinate 1200 1300 1385 1500 1750 Figure 6-1. Weight fraction of CeS produced versus time at different temperatures Reaction Coordinate vs. Temperature0 10 20 30 40 50 60 70 80 90 100 100011001200130014001500160017001800 Temperature (C)Reaction Coordinate Figure 6-2. Weight fraction of CeS produced versus reac tion temperature at 20 min. The data taken from these experiments wa s analyzed using the Jander equation for parabolic kinetics by plotti ng the values of F versus time where F = (1 3 (1))2 and is the reaction coordinate ( Jander, 1927 ). The data is then mode led as a straight line using the equation F = Kt where K is the rate constant and t is time. This plot can be seen in Figure 6-3. The Jander equation was devel oped to model the ra te of reaction of diffusion-limited processes in spherical partic les. When taking the particle size into account the rate constant can be expressed as Kt = 2kt/r2 where k is the parabolic rate constant and r is the particle radius wh ich was determined from scanning electron

PAGE 50

40 microscopy to be approximately 10 m. From this expression rate constants have been calculated and displayed in Table 6-1. Rate Data-0.02 0 0.02 0.04 0.06 0.08 0.1 010203040506070 Time (min.)F 1300 1500 Figure 6-3. Jander equation F versus time at 1300 C and 1500 C Table 6-1. Cerium monosulfide rate constants Temperature (C) Rate Constant (m2/s) 1300 3x10-4 1500 2x10-3 Kinetics Discussion As expected while the time allowed for the reaction or the temperature of the reaction increases the extent of reaction incr eases. The rate of reaction appears to decrease as time increases due to the incr eased distance required for diffusion to form cerium monosulfide. The Jander equation us ed for modeling diffusion-limited reactions fits the data reasonably well al though there is some variation that could be caused by the particle size distribution and non-spherical particle morphology. Other models have been developed that take a large particle size distribution into consideration but the sensitivity of the cerium hydride to air and moisture prev ent a more detailed an alysis of the powder size distribution ( Carter, 1961 ).

PAGE 51

41 Impurities may also play a role in the ra te of the cerium monosulfide synthesis reaction. It was indicated that an excess of cerium hydr ide could be used in the preparation of the reac tants in an attempt to reduce oxyge n impurities in the form of CeO ( Eastman et al. 1950 ). This procedure works best at high temperatures where the vapor pressure of the CeO is high, allowing it to be vaporized and removed from the sample by the vacuum system. At lower temperatures such impurities may not be removed completely and could slow the reaction rate by acting as additional barriers to diffusion. It is presumed that between 1500 C and 1750 C there is a large amount of CeO vaporized as the temperature raises the vapor pressure of CeO above the pressure inside the reaction column. This may lead to the large increase in both the reaction rate and purity of the cerium monosu lfide that is produced. As the reactant mixture is heated thro ugh 575 C the cerium hydride releases its hydrogen and the cerium reverts back to metallic form. The mixture of cerium and cerium sesquisulfide then reacts to form cer ium monosulfide. The rate-limiting step in this reaction would be the diffusion of either cerium atoms into cerium sesquisulfide or of sulfur atoms into the metallic cerium. This could be verified by measuring the diffusion rates of cerium and sulfur atoms through the cerium sulfides and comparing the activation energy for this process to the ac tivation energy for the cerium monosulfide production reaction. The x-ray diffraction data show that th e cerium sesquisulfide peak disappears before the reaction is completed which may be due to an increasing amount of disorder in its structure as diffusion takes place. This could be caused by li quid cerium coating the particles since there is an excess of cerium hydride used in the reaction.

PAGE 52

42 Since experiments were done at more th an one temperature it was possible to analyze the results and estimat e the activation energy for the rate-limiting step of the reaction. This is done by using the Jander equation to find the ra te constant of the reaction which is the slope of the F versus time curve. This value varies with temperature according to the equation: K = Ce-q/RT where K is the rate constant, q is the activation energy, T is absolute temperature, R is the ga s constant, and C is a constant. Figure 6-4 contains a plot of ln K versus 1/T, the sl ope of which is -q/R. Although only two points can be produced for this plot since only two te mperatures were investigated it is useful to get a rough estimate of q. The value of the activation energy was found to be approximately 190 kJ/mol cerium monosulfide. -8.5 -8 -7.5 -7 -6.5 -6 0.000560.000570.000580.000590.00060.000610.000620.000630.00064 1/T (K)ln(K) Figure 6-4. ln K versus 1/T fo r cerium monosulfide production Potential drawbacks of using the Jander model include the assumptions that the particles are spherical and the particle surface area remains constant. These assumptions are not strictly true as can be seen in the scanning electron micrographs of the reactant and product powders. The Jander model also depends on the reaction being diffusionlimited. While it is assumed that this reac tion involves only the di ffusion of cerium or sulfur atoms, other processes could take pl ace complicating the kinetics and invalidating

PAGE 53

43 the model. Reactions taking place with impur ities that are present could also cause the model to fail.

PAGE 54

44 CHAPTER 7 CONCLUSIONS This project was able to produce the requi red apparatuses and a detailed procedure for the production of cerium monosulfide from cerium sesquisulfide and cerium hydride. The kinetics of this reaction was also inve stigated using quantita tive x-ray diffraction. Since cerium sesquisulfide was required fo r the production of the cerium monosulfide an experimental procedure was also develope d for its synthesis from cerium oxide and hydrogen sulfide. Cerium hydride was also pr oduced based on previ ous literature from cerium metal and hydrogen ga s to participate in the reaction to produce cerium monosulfide. It was found that the cerium monosulfide synthesis reaction fit reasonably well with the Jander model of a diffusion-limited re action. Most likely the rate-limiting step of the reaction involves the di ffusion of either cerium atoms or sulfur atoms. The activation energy for this process was determin ed to be approximately 190 kJ/mol cerium sulfide. Additional work with cerium monosulfide will involve larger scale production followed by pressing the resulting powder in to test crucibles a nd running interaction studies with various molten metals. With the synthesis procedure available it will be possible to tailor the chemistry of the cer ium monosulfide using additives such as thorium sulfide in order to further enhance it s properties as a crucible material for use with actinide metals.

PAGE 55

45 In addition more extensive st udies could be done into th e kinetics of the synthesis reactions of not only cerium monosulfide but also cerium sesquisulfide. The cerium sesquisulfide reaction appears to involve several intermed iate compounds and is still poorly understood. As this compound is e ssential to the production of cerium monosulfide further investigation of its pr oduction would be beneficial to the overall synthesis procedure.

PAGE 56

46 LIST OF REFERENCES Bish D, Chipera S, 1988, Problems and solutio ns in quantitative an alysis of complex mixtures by x-ray powder diffraction. A dvances in X-ray Analysis 31:295-308. Brill M, Bloch D, Shmariahu D, Mintz M, 1995, The incipient kinetics of hydride growth on cerium surfaces. Journal of Alloys and Compounds 231:368-375. Carter R, 1961, Kinetic model for solid-state reactions. The Journal of Chemical Physics 34 [6]:2010-2015. Condon J, Holcombe C, 1977, Crucible materi als to contain molten uranium. Report Y2084 prepared for the U.S. Energy Resear ch and Development Administration, Oak Ridge, TN. Connolly J, April 2003, Introduction quantitat ive x-ray diffraction methods, pp. 1-14, http://epswww.unm.edu/xrd/ xrdclass/09-Quant-intro.pdf last accessed March 18, 2005. Eastman E, Brewer L, Bromley L, Gilles P, Lorgren N, 1950, Preparation and properties of refractory cerium sulfides. Journal of the American Ceramics Society 72 [5]:2248-2250. Eastman E, Brewer L, Bromley L, Gilles P, Lorgren N, 1951, Preparation and tests of refractory sulfide crucibles. Journal of the American Ceramics Soceity 34 [4]:128134. Flahaut J, Guittard M, Patrie M, 1956, Sur les sulfures de terres rares ceriques S3Me4. Comptes Rendus de l’Acadmie des Sciences 243 [25]:2074-2076. Hirai S, Shimakage K, Saitou Y, Nishimura T, Uemura Y, Mitomo M, Brewer L, 1998, Synthesis and sintering of cerium(III) powders. Journal of the American Ceramics Society 81 [1]:145-151. Hogan P, 2002, High temperature synthesis of sulfides of cerium and thermodynamic system modeling. Masters thesis presented to the University of Florida, Gainesville, FL. Jander W, 1927, Reactions in solid state at high temperatures. Zeitschrift fr Anorganische und Allgemeine Chemie 163:1.

PAGE 57

47 Krikorian O, Curtis P, 1988, Synthesis of Ce S and interactions with molten metals. High Temperatures-High Pressures 20:9-18. Marrot F, Mosset A, Trombe J, Macaudiere P, Maestro P, 1997, The stabilization of gamma-Ce2S3 at low temperature by heavy rare earths. Journal of Alloys and Compounds 259:145-152. Radzikovskaya S, 1961, The chemistry of rare -earth and actinide sulphides. Russian Chemical Review 30 [1]:28-41. Ryan F, Greenberg I, Carter F, Miller R, 1962, Thermoelectric prope rties of some cerium sulfide semiconductors from 4 to 1300K. Journal of Applied Physics 33 [3]:864868. Samsonov G, 1964, High-temperature compounds of rare earth metals with nonmetals. Metallurgiya Press, Moscow, Russia. Sarussi D, Jacob I, Bloch J, Shamir N, Mintz M, 1993, The kinetics and mechanism of cerium hydride formation. Journal of Alloys and Compounds 191:91-99.

PAGE 58

48 BIOGRAPHICAL SKETCH Kevin Brent Gibbard was born in Michigan on April 18, 1981, and moved to Florida in 1985. From an early age his l ove of math, science, and Legos always facilitated his interest in e ngineering. During high school he had an internship using chemistry to make sunscreen and was a memb er of Mu Alpha Theta. Kevin remained well rounded as he was on his high school sw im team and won many backstroke and distance events. He graduated from hi gh school in 1999 from the International Baccalaureate program at Palm Harbor Un iversity High School. He attended the University of Florida for four years as an undergraduate student and earned the degree of Bachelor of Science in materials science a nd engineering in May 2003. He was invited to become a member of the Golden Key Nationa l Honor Society and Tau Beta Pi. Kevin continued his education at the master’s level at the University of Florida. He received his Master of Science in materials science a nd engineering in May 2005. Kevin accepted an engineering position with Bechtel Bettis at its atomic power laboratory in Pittsburgh, Pennsylvania.