USF Libraries
USF Digital Collections

Supercritical carbon dioxide aided preparation of nickel oxide/alumina aerogel catalyst

MISSING IMAGE

Material Information

Title:
Supercritical carbon dioxide aided preparation of nickel oxide/alumina aerogel catalyst
Physical Description:
Book
Language:
English
Creator:
Li, Haitao
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Sol-gel
Aging
Extraction and drying
Impregnation
Calcination
Functionalization
Porosity
Surface area
Dissertations, Academic -- Chemical Engineering -- Masters -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The strength, thermal stability, pore structure and morphology are keys to success for wider deployment of aerogels. Furthermore, co or subsequent functionalization of the surfaces are equally, if not more important. This study addresses these issues through a new method. The path involves successful use of surfactant templating, supercritical extraction and drying, and supercritical fluid aided functionalization of the surface. Alumina support and alumina supported nickel catalyst particles are used to evaluate the approach. Initially thermally stable surfactant alumina was synthesized. The surfactant template was removed completely with the aid of a supercritical solvent mixture. Surfactant-templated alumina aerogel showed remarkable thermal stability and gave specific surface area above 500m2/g both before and after calcination. The alumina support is subsequently impregnated with nickel.
Thesis:
Thesis (M.S.Ch.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Haitao Li.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 131 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001681037
oclc - 62708158
usfldc doi - E14-SFE0001090
usfldc handle - e14.1090
System ID:
SFS0025411:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681037
003 fts
005 20060215071141.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051219s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001090
035
(OCoLC)62708158
SFE0001090
040
FHM
c FHM
049
FHMM
090
TP145 (Online)
1 100
Li, Haitao.
0 245
Supercritical carbon dioxide aided preparation of nickel oxide/alumina aerogel catalyst
h [electronic resource] /
by Haitao Li.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (M.S.Ch.)--University of South Florida, 2005.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 131 pages.
3 520
ABSTRACT: The strength, thermal stability, pore structure and morphology are keys to success for wider deployment of aerogels. Furthermore, co or subsequent functionalization of the surfaces are equally, if not more important. This study addresses these issues through a new method. The path involves successful use of surfactant templating, supercritical extraction and drying, and supercritical fluid aided functionalization of the surface. Alumina support and alumina supported nickel catalyst particles are used to evaluate the approach. Initially thermally stable surfactant alumina was synthesized. The surfactant template was removed completely with the aid of a supercritical solvent mixture. Surfactant-templated alumina aerogel showed remarkable thermal stability and gave specific surface area above 500m2/g both before and after calcination. The alumina support is subsequently impregnated with nickel.
590
Adviser: Dr. Aydin Sunol.
Co-adviser: Dr. John Wolan
653
Sol-gel.
Aging.
Extraction and drying.
Impregnation.
Calcination.
Functionalization.
Porosity.
Surface area.
690
Dissertations, Academic
z USF
x Chemical Engineering
Masters.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1090



PAGE 1

Supercritical Carbon Diox ide Aided Preparation of Nickel Oxide/Alumina Aerogel Catalyst by Haitao Li A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Chemical Engineering Department of Chemical Engineering College of Engineering University of South Florida Co-Major Professor: Aydin Sunol, Ph.D., P.E. Co-major Professor: John Wolan, Ph.D. Sermin Sunol, Ph.D. Date of Approval: February 15, 2005 Keywords: Sol-gel, Aging, Extraction and drying, Impregnation, Calcination, Functionalization, Porosity, Surface area Copyright 2005, Haitao Li

PAGE 2

ACKNOWLEDGEMENTS Special sincere gratitude and thanks go to professors Aydin Sunol, Sermin Sunol and John Wolan for their continuous guidance, moral support, especially in my toughest time in USF. It is doubtful that I could have finished my study and research, or even been able to stay here without their help and support. An Additional thanks goes to Dr. Aydin Sunols research group, Al-Ahmad, Raquel Carvallo, Brandon Smeltzer, Naveed Aslam, Ying Zhang, and Dr. John Wolans research group, especi ally Benjamin Grayson for XPS analysis, and my friends at the Nanomaterials an d Nanomanufacturing Research Center (NNRC). Special thanks are also given to Peiyao Cheng. And last but not least, I would like to thank my parents and my families for their support.

PAGE 3

i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES vi LIST OF ABBREVIATIONS x ABSTRACT xi CHAPTER ONE INTRODUCTION AND OBJECTIVES 1 1.1 Introduction 1 1.2 Objectives 4 CHAPTER TWO BACKGROUND 5 2.1 Supercritical Fluids Technology 5 2.1.1. Properties of Supercritical Fluids 6 2.1.2. Supercritical Fluids Extraction (SFE ) and Drying 8 2.1.3. Supercritical Fluids Impregnation (SFI) 10 2.1.4. Solubility of Organometallic Complexes and Template Surfactants in Supercritical Fluids 11 2.2 Introduction to Sol-gel Technology 18 2.2.1. Sol-gel Chemistry 19 2.2.2. Aging 21 2.2.3. Heat Treatment 22 2.3 Surfactant Templates 23 2.4 Alumina-Nickel/Alumina Catalyst System 25 2.4.1. Alumina Catalyst Support 26 2.4.2. Nickel/Alumina Catalyst System 26 CHAPTER THREE EXPERIMENTAL WORK 28 3.1 Sol-gel Preparation 28 3.1.1. Synthesis of Sol-gel for Templated Alumina 28 3.1.2. Synthesis of Sol-gel for Regular Co-precipitate Nickel Oxide/Alumina 30 3.1.3. Synthesis of Sol-gel for Templated Co-precipitate Nickel Oxide/Alumina 34

PAGE 4

ii 3.2 Sol-gel solvent replacement 37 3.3 Supercritical Extraction and Drying 37 3.3.1. SCF Extraction and drying Experimental Setup 37 3.3.2. Procedure of SCF Extraction and Drying 39 3.4 Supercritical Impregnation 41 3.4.1. Supercritical Impregnation Setup 41 3.4.2. Procedure of SCF Impregnation 42 3.4.2.1. Continuous Impregnation 42 3.4.2.2. Batch Impregnation 44 3.5 Calcination and Heat Treatment 46 3.6 Catalyst and Catalyst Support Characterization Analysis Setups 48 3.6.1. Surface and Structural Characterizations 48 3.6.2. Scanning Electron Microscopy (SEM) and Energy Dispersion Spectroscopy (EDS) 51 3.6.3. X-ray Photoelectron Spectroscopy (XPS) 52 3.7 Reagents 53 CHAPTER FOUR RESULTS AND DISCUSSION 54 4.1 Alumina Aerogel Preparation 54 4.1.1. Effect of Composition 54 4.1.2. Effect of Aging 57 4.1.3. Effect of SCF Extraction/Drying Conditions 61 4.1.3.1 Flow Rate 61 4.1.3.2 SCF Extraction Time 64 4.1.4. Effect of Degassing 67 4.2 Co-precipitate Nickel/Alumina System 69 4.3 Supercritical Fluid Impregnation 76 4.3.1. Continuous Supercritical Impregnation 77 4.3.1.1. Effect of Nickel Complex 77 4.3.1.2. Effect of Initial Ni Amount 81 4.3.2. Batch Supercritical Impregnation 81 4.4 Comparison between Continuous and Batch SCF Impregnation 85 4.5 Comparison between SCF Impregnation and Co-precipitate 86 CHAPTER FIVE CONCLUSIONS AND RECOMMENDATIONS 88 5.1 Conclusions 88 5.2 Recommendations 89

PAGE 5

iii REFERENCES 91 APPENDICES 95 Appendix A: The BET Analysis Results 96 Appendix B: The SEM Images 108 Appendix C: The EDS Spectra 112 Appendix D: The XPS Elemental Analysis Results 116

PAGE 6

iv LIST OF TABLES Table 2.1 Comparison of Physical Properties of Gases, Supercritical Fluids and Liquids 7 Table 2.2 Mole Fraction Solubility (x) of Transition Metal Complexes in Supercritical CO 2 12 Table 3.1 Specifications for the Quantachrome NOVA 2000 (High Speed Gas Sorption Analyzer Version 6.11 50 Table 3.2 Specifications for SEM-EDS Hitachi Model S-800 Machine 51 Table 3.3 Main Reagent Information 53 Table 4.1 Textural Properties for Templated Alumina Aerogel with Different Composition 56 Table 4.2 Textural Properties for Templated Alumina Aerogel with Different Aging Period 60 Table 4.3 Textural Properties for Templated Alumina Aerogels with Different SCF Extraction Flow Rates 61 Table 4.4 Textural Properties for Templated Alumina Aerogels with Different SCF Extraction Times 64 Table 4.5 Textural Properties for Templated Alumina Aerogel with Degassing and Without Degassing 67 Table 4.6 The Textural Properties and Element Composition of The Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Catalyst System 71 Table 4.7 The Textural Properties and Element Composition of Continuous SCF Impregnation Nickel Oxide/Alumina Catalyst System 79

PAGE 7

v Table 4.8 The Constant Process Condition of Batch SCF Impregnation 82 Table 4.9 The Textural Properties and Element Composition of Batch SCF Impregnation Nickel Oxide/Alumina Catalyst System 83 Table 4.10 Comparison between Two Continuous Impregnation Samples and One Batch Impregnation Sample 85 Table 4.11 Comparison between Co-Precipitated Samples and Batch Impregnation Sample 86

PAGE 8

vi LIST OF FIGURES Figure 2.1 Phase Diagram of Carbon Dioxide and Water 7 Figure 2.2 Schematic Diagram for Sol-gel Routes, Preparation of Metal Oxide Gel through Sol-gel Method 20 Figure 3.1 Schematic Diagram for Synthesizing and Processing Templated Alumina Sol-gel 31 Figure 3.2 Schematic Diagram for Synthesizing and Processing Regular Nickel Oxide-Alumina Sol-gel 33 Figure 3.3 Schematic Diagram for Synthesizing and Processing Templated Nickel Oxide-Alumina Sol-gel 36 Figure 3.4 Schematic Diagram for The Supercritical Extraction and Drying Setup 38 Figure 3.5 Schematic Diagram for The Supercritical Impregnation Setup 43 Figure 3.6 Schematic Diagram for Processing Supercritical Batch Impregnation 45 Figure 3.7 Designed Sample Calcination Device 47 Figure 4.1 Schematic Diagram of Templated Alumina Aerogel Preparation with Different Composition 55 Figure 4.2 SSA, APV and APD of Two Alumina Aerogel with Different Composition (Before Calcinations and After Calcinations) 56 Figure 4.3 Pore Size Distribution of Two Alumina Aerogel With Different Composition (Before Calcination and After Calcination) 58

PAGE 9

vii Figure 4.4 Schematic Diagram of Templated Alumina Aerogel Preparation with Aging Periods 59 Figure 4.5 SSA, APV and APD of Two Alumina Aerogel with Different Aging Periods (Before Calcinations and After Calcinations) 60 Figure 4.6 Schematic Diagram of Templated Alumina Aerogel Preparation with Different SCF Flow Rates 62 Figure 4.7 SSA, APV and APD of Three Alumina Aerogel with Different SCF Extraction Flow Rates (Before Calcinations and After Calcinations) 63 Figure 4.8 The Mean Values and Standard Deviations of SSA, APV and APD of Two Series of Alumina Aerogel with Different SCF Extraction Time (Before Calcinations and After Calcinations) 65 Figure 4.9 Schematic Diagram of Templated Alumina Aerogel Preparation with Different SCF Extraction Time 66 Figure 4.10 SSA, APV and APD of One Alumina Aerogel with Degassing and Without Degassing (Before Calcinations and After Calcinations) 67 Figure 4.11 Pore Size Distribution of the Sample after Calcination Without Degassing 68 Figure 4.12 Schematic Diagram of Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Preparation 70 Figure 4.13 SSA, APV and APD of Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Catalyst System (Before Calcinations and After Calcinations) 72 Figure 4.14 SEM-EDS Spectrum of Regular Co-Precipitated Nickel/Alumina Aerogel (Ni:Al=0.15:1) 73 Figure 4.15 SEM Images of Regular and Templated Co-Precipitated Nickel/Alumina Aerogel before Calcination (Ni:Al=0.15:1) 74

PAGE 10

viii Figure 4.16 SEM-EDS Elements Mapping Image of Regular (A) and Templated(B) Co-Precipitated Nickel/Alumina Aerogel Before Calcination 75 Figure 4.17 Three Basic Steps of Continuous SCF Impregnation 77 Figure 4.18 XPS Elemental Analysis Result of Sample D 80 Figure 4.19 Three Basic Steps of Batch SCF Impregnation 81 Figure 4.20 SEM-EDS Elements Mapping Image of Two Batch SCF Impregnation Nickel Oxide/Alumina Catalyst System after Calcination 84 Figure 4.21 Comparison Between Two Continuous Impregnation Samples and One Batch Impregnation Sample 85 Figure 4.22 Comparison between Co-Precipitated Sample and Batch Impregnation Sample 87 Figure B.1 The SEM Images of Continuous Impregnation Sample A (Nickel acetate : Alumina=2:1) 108 Figure B.2 The SEM Images of Continuous Impregnation Sample B (Nickel nitrate : Alumina=2:1) 108 Figure B.3 The SEM Images of Continuous Impregnation Sample C (Nickel acetate : Alumina=1:1) 108 Figure B.4 The SEM Element Mapping Images of Continuous Impregnation Sample A (Nickel acetate : Alumina=2:1) 109 Figure B.5 The SEM Element Mapping Images of Regular Co-precipitated Nickel /Alumina before Calcination (Nickel acetate : Alumina=0.15:1) 109 Figure B.6 The SEM Element Mapping Images of Templated Co-precipitated Nickel /Alumina before Calcination (Nickel acetate : Alumina=0.15:1) 110

PAGE 11

ix Figure B.7 The SEM Element Mapping Images of Templated Co-precipitated Nickel /Alumina after Calcination (Nickel acetate : Alumina=0.15:1) 110 Figure B.8 The SEM Element Mapping Images of Batch Impregnation Nickel /Alumina before Calcination (NiAC:Alumina=3:1 at 2500 psig & 40 o C) 111 Figure B.9 The SEM Element Mapping Images of Batch Impregnation Nickel /Alumina after Calcination (Nickel acetate: Alumina= 3:1 at 1500 psig & 60 o C) 111 Figure C.1 The SEM-EDS Spectrum of Regular Co-precipitated Nickel /Alumina after Calcination (Nickel acetate : Alumina=0.15:1) 112 Figure C.2 The SEM-EDS Spectrum of Templated Co-precipitated Nickel /Alumina after Calcination (Nickel acetate : Alumina=0.15:1) 113 Figure C.3 The SEM-EDS Spectr um of Batch Impregnation Nickel /Alumina before Calcination (Nickel acetate : Alumin a=3:1 at 2500psig & 40 o C) 114 Figure C.4 The SEM-EDS Spectr um of Batch Impregnation Nickel /Alumina after Calcination (Nickel acetate : Alumin a=3:1 at 2500psig & 40 o C) 115 Figure D.1 XPS Elemental Analysis Result of Sample A 116 Figure D.2 XPS Elemental Analysis Result of Sample B 117 Figure D.3 XPS Elemental Analysis Result of Sample C 118

PAGE 12

x LIST OF ABBREVIATIONS ASB ASM APD APV BET CAL EDS HPLC M.W. MSDS SSA SCF SEM S.W. TSA TPV XPS Aluminum tri-sec-butoxide As-made Average Pore Diameter Average Pore Volume BET Analysis After Calcination Energy Dispersion Spectrum High Performance Liquid Chromatography Molecular Weight Material Safety Data Sheet Specific Surface Area Supercritical Fluid Secondary Electron Microscopy Sample Weight Total Surface Area Total Pore Volume X-ray Photoelectron Spectroscopy

PAGE 13

xi SUPERCRITICAL CARBON DIOXIDE AIDED PREPARATION OF NICKEL OXIDE/ALUMINA AEROGEL CATALYST HAITAO LI ABSTRACT The strength, thermal stability, pore structure and morphology are keys to success for wider deployment of aerogels. Furthermore, co or subsequent functionalization of the surfaces are equally, if not more important. This study addresses these issues through a new method. The path involves successful use of surfactant templating, supercritical extraction and drying, and supercritical fluid aided functionalizatio n of the surface. Alumina support and alumina supported nickel catalyst particles are used to evaluate the approach. Initially thermally stable surfactant alumina was synthesized. The surfactant template was removed completely with the aid of a supercritical solvent mixture. Surfactant-templated alumina aerogel showed remarkable thermal stability and gave specific surface area above 500m 2 /g both before and after calcination. The alumina support is subsequently impregnated with nickel. BET and BJH method (Nitrogen adsorption-desorption isotherms) were used to follow the removal of solvents and templates as well as tracking the textural properties for the synthesized gel. Meanwhile, co-precipitated nickel oxide/alumina system was also synthesized for comparison with the supercritical impregnation nickel ox ide/alumina system. SEM-EDS and XPS were employed to study the distribution of the nickel on the alumina support and the percentage was compared with the initial mixture of the sol gels.

PAGE 14

1 CHAPTER ONE INTRODUCTION AND OBJECTIVES 1.1 Introduction The catalyst industry is one of the most important components of the entire worldwide industry with a steadily improving market. The US catalyst market is likely to increase to $2.7 bi llion dollars by 2005, based on an average growth rate of 3.9%/y from the year 2000. Among hundreds of catalyst support materials, transition alumina became popular recently due to its highly thermal and chemical stability and higher porosity which are key requirements for catalyst supports, thermal & acoustic insulators and adsorbents. High thermal stability means that transition alumina used as catalyst support material has the ability to with stand severe reaction conditions without losing a great deal of its physical properties, such as, textural and morphological properties. High porosity means that a transition alumina catalyst support is capable to house more active catalyst component in its highly porous matrix so that it can provide more accessible and active reaction sites for catalytic reaction. Commercial activated alumina products typically exhibit specific surfac e areas (SSA) between 185-250 m 2 /g with an average pore volume (APV) less than 0.5cc/g. Those textural parameters, to a large extent, determine the performance of alumina in catalysis and adsorption applications. Sol-gel chemistry is currently appli ed as one of the most widely used methods for synthesis of alumina supports (Scherer and Brinker, 1990). The sol-gel method offers several advantages in making catalysts and catalyst supports. This process provides aerogels with favorable properties such as high

PAGE 15

2 purity, high porosity, high surface area in addition to homogeneity at the molecular level (Suh,D.J. et al. 1998). However, different drying processes after synthesis of sol-gel system determines the final textural and functional properties of the alumina support. When the liquid in sol-gel system is dried at atmosphere pressure or vacuum conditio ns, capillary forces produce shrinkage and cracking which causes lose of the major porous structure of the gel and a tremendous reduction of porosity and surface area. Aerogel was first invented by Kistler (Kistler, 1932) and was defi ned as gel dried at a temperature and pressure higher than the critical poin t of the pore fluid. Eliminating the iquid-vapor interface and capillary force, minimize the shrinkage associated with drying; the hypercritical drying process largely increases the porosity and surface area. With newly accessible supercritical regimes (T c = 31.1 o C, P c = 72.8 atm), carbon dioxide, when used as supercritical drying ingredient, enables synthesis of aerogel at relati ve lower temperatures and pressures (Brinker, 1986), than previously. Controlled pore size distribution is an other favorable property for catalysts. Recently, new pathways have been studied to prepare porous base materials, such as alumina, allowing better contro l over the properties of the porous materials through the introduction of surf actants to the synthesis of the sol-gel (Gonzales-Pena, V, et al. 2001). A refined and optimized non-ionic surfactant-templated pathway was ut ilized to produce activated alumina supports with mesoporous structure in this research. The basic advantages of the surfactant-templated porous material are high porosity and tunable unimodal nanoporosity (Mark, T. Anderson, et al. 1998). Also, removal of the solvent and surfactant using supercritical fluids ensures retaining the pore structure (Sunol, A.K. and Sunol, S.G., 2004).

PAGE 16

3 Currently, Supercritical fluid (SCF) ai ded impregnation was widely applied for various product functionalization pu rposes including catalysts, polymers, thin films, porous semiconductors and wood products. The low viscosities of SCFs and high diffusivities of solutes in SCFs may result in superior mass transfer characteristics compared to conventional solvents. This is accomplishment through controlling the solu tes solubility in different operating conditions solutes can be precipitated into the pores. In this study, the impregnation of nickel onto the alumina support was achieved by using nitrate and acetate precursors with the aid of supercritical methanol carbon dioxide mixtures. After the processes of sol-gel production, supercritical extraction dr ying, supercritical impregnation and further heat treatment step proceed in order to convert the nickel precursors to NiO X and enhance the thermal, chemical and morphological stability of the alumina catalyst in the same simple process. In this thesis, thermally stable surfactant-templated alumina aerogels and impregnation of nickel on the alumina supports including their synthesis, calcination, textural, morphological char acterization, and metal dispersion are presented. A literature review presented in Chapter two provides background on alumina and nickel/alumina systems, sol-gel chemistry, templates, supercritical extraction and drying, supe rcritical impregnation, solubility of non-ionic surfactant and nickel complexes, properties of supercritical fluids and catalyst characterization. Chapter three describes the experimental setups, procedures employed, and characterization methods used. Chapter four presents and discusses the experime ntal results. Conclusion and recommendations are presented in Chapter five.

PAGE 17

4 1.2 Objectives The specific objectives of this study are: 1) Preparing non-ionic surfactant-templated alumina sol-gels by a refined and optimized pathway; 2) Removing the surfactant and solvent of synthesized templated sol-gels by using supercritical flui d aided extraction and drying to form aerogels which are highly porous, mesoporous and thermally stable; 3) Preliminarily functionalizing the po rous matrices and surfaces of aerogels with nickel using nickel nitrate or nickel acetate as precursors through novel supercritical fluid aided impregnations, and comparing it with co-precipitation nickel oxide/alumina catalyst systems; 4) Characterizing alumina and nickel oxide/alumina aerogels with a variety of analysis methods including BET, SEM-EDS and XPS.

PAGE 18

5 CHAPTER TWO BACKGROUND In this chapter, a literature review of supercritical fluid (SCF) technology including properties of SCF, supercr itical fluid extraction and drying, supercritical fluid impregnation and solubility of organometallic complexes and non-ionic surfactant in SCFs is given. Also, sol-gel chemistry, sol-gel aging, drying, and calcination are discussed in the second section. The third section mainly reviews the new pathway of intr oducing surfactant templates. In the last section, methods of characterization of alumina and nickel/alumina catalysts are presented. 2.1 Supercritical Fluids Technology For any pure substance, there is a tran sition state called the critical state: for temperatures below th e critical temperature T C or for pressures below the critical pressure P C two phases co-exist, liquid and vapor. When temperature and (or) pressure exceed the critical temperature and pressure ( T>T C ,P>P C ) namely the critical point (CP), the two phases, liquid and vapor become indistinguishable. This is called the su percritical phase. Compared with liquid solvents, supercritical fluids reveal several unique properties including a greater diffusivity ( D ), a tunable equivalent density ( ), and a lower viscosity () (Michel Perrut, 1994). Supercritica l fluid technology has become very attractive for these stated reasons. Even though the first plant using supercritical fluid technology was built over 20 years ago in North America, it is only now that we see its wide applications. Supercritical fluid technology offers numerous advantages compared to conventional processes, such as high mass transfer rate, reduced unit, and lower operating costs and environmentally

PAGE 19

6 benign processing (Chordia, L. and Martinez, J.L., 2002). Extraction and purification, particle production, impreg nation, analytical application, solvent replacement and green chemistry are the most popular attributes of supercritical processing (Teja, A.S., and Eckert, C.A., 2000). 2.1.1. Properties of Supercritical Fluid Properties of SCFs are different from those of ordinary liquids and gases being tunable simply by changing the pressure and temperature. In particular, density and viscosity change drastically at conditions close to the critical point which greatly affects the solvent activity. Mass transport properties such as viscosity and diffusivity are similar to those of gases; therefore, improved mass transport rates are achieved in supercritical fluids. Indeed, the viscosities of supercritical fluids are lowe r than those of liquid solvents, but higher than those of gases; the diffusivities of the solute in SCFs are also intermediate between those of gases and liquids. Both of thes e properties reveal that the density of SCFs is also between those of gases and liquids, therefore, SCFs is capable of dissolving many compounds which might not be soluble in gases at ambient temperature. Furthermore, the solvating power of SCFs is tunable because the density of SCFs is also ea sily tunable with merely changing the pressure at constant temperature. In general, supercritical fluids have the mobility of gases and the solvating power of liquid solvents resulting in ef ficient permeation into porous matrices, high mass transport rates and high solvency. A general comparison of the magnitude of some of the important properties for liquids, gases, and supercritical fluids in the near critical region is listed (Taylor, 1996) in Table 2.1. The phase diagram of carbon dioxide and water is shown in Figure 2.1.

PAGE 20

Table 2.1 Comparison of Physical Properties of Gases, Supercritical Fluids and Liquids Physical Property Gas Supercritical fluid Liquid Density (kg/m 3 ) 0.6-2.0 200-500 600-1600 Diffusivity, D, (10 -6 m 2 /s) 10-40 0.07 0.0002-0.002 Dynamic Viscosity, (mPa s) 0.01-0.3 0.01-0.03 0.2-3 Kinematics Viscosity, (10 -6 m 2 /s) 5-500 0.02-0.1 0.1-5 The thermodynamic properties of supercritical mixtures are very complex. A simple equation of state such as Peng-Robinson (Equation 2.1) does very well in correlating the experimental data and representing the phase behaviors. )()(bVmbbVmVmabVmRT (2.1) Figure 2.1 Phase Diagram of Carbon Dioxide and Water The three most popular supercritical fluids are carbon dioxide, propane and water. Among them, special attention and research have been paid to 7

PAGE 21

8 carbon dioxide (CO 2 ) because of its convenient critical parameters (T c =31.1 o C and P c =7.38 MPa), non-toxicity, inflammability, availability, physiological compatibility, low cost and environmental benignancy. At near ambient temperature and under acceptable pressure, carbon dioxide, operates as an excellent supercritical solvent and is typically used to process low-volume and high-value products. This includes food, pharmaceuticals, fine chemicals, and cosmetics which should be carried out at temperatures as close as possible to ambient temperature to avoid thermal degradation and without the use of hazardous chemicals (Perrut Michel, 1994). There are also a number of practical advantages associated with the use of supercritical carbon dioxide as a processing solvent. For example, pr oduct isolation to total dryness is achieved by simple evaporation. This coul d prove to be particularly useful in the final steps of pharmaceutical syntheses where even trace amounts of solvent residues are considered problematic. 2.1.2. Supercritical Fluid Extraction (SFE) and Drying The tunable density and compressibility make supercritical fluids adjustable solvents with a continuous transition between excellent solvents under supercritical conditions and poor solvents in the state of a compressed gas; most processes are based on these solvent power variations (Perrut, Michel. 1994). Supercritical fluid extraction has been traditionally applied in food and in pharmaceutical areas, especially for nutra ceutical area when extraction is done with scCO 2 which does not leave any toxic residues in the matrix. Nutraeuticals are natural extracts from plants or natural products that contain physiological or health benefits. The first industrial application of supercritical fluids was a coffee decaffeination plant 1978, followed by a hops extraction plant in 1982.

PAGE 22

During the 80s and 90s industrial and lab research in supercritical fluids were mainly focused on extraction processes from liquid and solid matrices. In the method of extraction and drying using supercritical fluids, the selectivity and extractability of the objectives are highly determined by their solubility in supercritical fluids. This is accomplished by tuning the density of the supercritical fluid through minutely adjusting the processing temperature or pressure, or by adding co-solvents to enhance the solvating power for specific compounds. The process of supercritical drying is a specific example of supercritical fluid extraction and it utilizes another property of SCF -low surface tension. Low surface tension accompanied by high diffusivity allows the supercritical fluids to penetrate into porous matrix very affectively. When compared with supercritical fluids, traditional evaporative extraction and drying uses solvents to separate and extract certain solutes of interest under ambient pressure or vacuum either with or without heating. As a liquid-vapor interfacial phenomenon, the evaporation of solvent from a sample surface or matrix is associated with severe capillary pressures which could cause permanent collapsing of fragile sample such as sol-gel porous structures. The best approach to minimize the capillary pressure acting on the sol-gel network during drying is to examine the following Equation 2.2 rP)cos(2 (2.2) where P is capillary pressure, is surface tension, is the contact angle between liquid and solid and r the pore radius. For a given pore size the capillary pressure can be reduced by using a solvent with a lower surface tension than the original solvent in the gel network (Anderson, 1997, Brinker, 1992) and/or eliminating the liquid-vapor interface with supercritical drying 9

PAGE 23

10 (Kistler, 1932, Tewari, 1985). On the other hand, supercritical carbon dioxide offers considerable potential as a replacement for solvents in many extraction and drying processes, such as, extrac tion of solvents and monomers from polymer solutions. High purity CO 2 is the supercritical fluid medium of choice for most extractions. For reasons already stated, in supercritical drying, the supercritical fluid being used as the drying medium can be tuned to perform different level of extractions in supercritical fractionation because the supercritical solvating power is practically tuned and adjusted. One of those pathways is varying the density of the supercritical fluid throug h either or both of temperature and pressure during the process, the higher the density of the supercritical fluid is, the higher the solvating power of that supercritical fluid an d hence the higher drying rate achieved. Another method to tune the solubility of the supercritical fluid is to modify the supercritical fluids to be able to entrain the solute from the solid phase in the drying process. For example, enhanced non-polar supercritical CO 2 produced by adding small amount of misc ible polar modifier such as ethanol and/or methanol is able to dissolve polar molecules. The added fluid is referred to in the literatures as entrainer, co-sol vent or modifier. The diffusion coefficient of the solute in the supercritical fluid with a modifier is much smaller than without the use of the modifier. In this study, the second pathway was used in supercritical extraction an d supercritical impregnation with different entrainers in each process. 2.1.3. Supercritical Fluids Impregnation (SFI) Impregnation with supercritical fluids, such as impregnating the matrices with dopants, is the reverse process of SFE. Impregnating active reagents in porous media or polymer system is more effective and timesaving with

PAGE 24

11 supercritical fluids than traditional liqui d solvents due to the higher diffusivity of supercritical fluids. Moreover, the grea ter diffusivity of supercritical fluids allows one to reach a homogeneous dist ribution of active compounds inside various porous matrices. Supercritical impregnation occurs in two steps. First, the matrix is permeated by a supercritical solution containing the compound. In the second step, a fast decompression causes an extreme decrease in the solubility of the supercritical fluid and permits the solute deposition inside the porous matrix while the supercritical solvent is rapidly vented out in the gas phase. Examples of this concept are impregnation of colo rants, aromas and pesticides in various matrices (Saus et al., 1993). Most of the applications in this area relate to polymeric materials. Impregnation of organic dyes into glassy polymers (Kazarian S.G. et al., 1997) and impregnation of organometallic compounds into polymer (PVC, PE or PVA) films (Res t, A.J. 1990) were achieved recently. Similarly, supercritical fluid dyes of textiles, especially polyesters, are considered to substitute classical aqueous dyeing with related water pollution. 2.1.4. Solubility of Organometallic Complexes and Template Surfactants in Supercritical Fluids By increasing the pressure of the gas ab ove the critical point, it is possible to give the solvent liquid-like densities and solvating strengths. Near the critical point, the density of the gas will increase rapidly with increasing pressure. Therefore, the solubility of many compounds is several orders of magnitude greater than predicted from the classical thermodynamics of ideal gases. As the average distance between molecules decreases, non-ideal gas behavior will begin to govern the interactions between the solvent and the sample accounting for a tremendous enhancement in solubility. In the supercritical region, solvating strength is a direct function of density, which in turn is

PAGE 25

12 dependent on system pressure (at constant temperature). Solvating strengths can be finely tuned by minutely adjustin g the pressure and/or temperature. Because of the noncompressibility of conventional liquids, this phenomenon is unique to supercritical fluids. It is even possible, by adding small quantities of miscible co-solvent, to customize a supercritical fluid for a specific application. Polar co-solvents, also known as modifiers or entrainers in different research reports, the most frequently used ones being ethanol and methanol; lead to increases the selectivity and solvating power because of solute molecules interacting with the co-solvent. Much work has been done on the modification of supercritical CO 2 by adding one or more modifiers to increase the solvating power and selectivity for specific solutes such as surfactants and organometallic complexes. A summary of recent research work on the mole fraction solubility of transition metal complexes in supercritical CO 2 is shown in Table 2.2. In this research, a non-ionic surfactant templateTriton X-114 was used to create a highly porous matrix in solgel during the preparation of an alumina aerogel which was removed as effectivel y as possible through supercritical extraction. Meanwhile, nickel acetate an d nickel nitrate employed as catalyst (NiO X ) precursors were impregnated in alumina aerogel throughsupercritical impregnation. The affinity study on solu bility of these non-ionic surfactant and nickel complexes in different supercritic al mixtures and conditions has been done experimentally and accurate models for calculating the solubility of these compounds were also achieved (Smeltzer, Brandon, 2005).

PAGE 26

Table 2.2 Mole Fraction Solubility (x) of Transition Metal Complexes in Supercritical CO 2 Metal complex P, atm T, o C Density, g/ml Solubility, x Reference Available/Supplier/Price GROUP 4 TiCl 4 100 76 0.23 0.09 [1]. Bartle, K. D.; Clifford, A. A.; Jafar, S. A.; Shilstone, G. F. J. Phys. Chem. Ref. Data 1991, 20, 713-756. SA, 89541,500ml,>98%,$35.4 GROUP 6 Cr(acac) 3 200 60~40 0.73~0.85 1.2~1.110 -4 [2]. M. Ashraf-Khorassani, M.T. Combs and L.T. Taylor, Solubility of metal chelates and their extraction from an aqueous environment via supercritical CO 2 Talanta 44 (1997), p. 755-763.. [3]. Lagalante, A. F.; Hansen, B. N.; Bruno, T. J.; Sievers, R. E. Inorg. Chem. 1995, 34, 5781-5785. SA,202231,100g, >97%,$31.2 13 175 40 0.81 9.110 -5 [3] Cr(acacBr) 3 200 40 0.85 1.310 -5 [3] Synthesized [3] 175 40 0.81 1.010 -5 [3] Cr(thd) 3 200 40 0.85 4.510 -3 [3] SA, 468223-1G,$57.20 trans-Cr(thd) 3 175 40 0.81 4.2110 -3 [3] Synthesized [3] mer-Cr(tfa) 3 200 40 0.85 2.010 -3 [3] Synthesized [3] fac-Cr(tfa) 3 200 40 0.85 1.410 -3 [3] Synthesized [3] cis-Cr(tfa) 3 175 40 0.81 1.13910 -3 [3] Synthesized [3] Cr(hfa) 3 200 60 0.73 >8.010 -3 [2] Mo(CO) 6 103 51 0.39 9.210 -3 [4]. Warzinski, R. P.; Lee, C.-H.; Holder, G. D. J. Supercrit. Fluids, 1992, 5, 71. Pressure Chemicals, Pittsburgh, PA, 1

PAGE 27

Table 2.2 Continued GROUP 7 Mn(acac) 3 290 60 0.83 1.910 -7 [5]. Saito, N.; Ikushima, Y.; Goto, T. Bull. Chem. Soc. Jpn. 1990, 63, 1532-1534. F-A, ac34378-0250,25g,$42.1 Mn(acac) 2 290 60 0.83 8.510 -8 [5] CpMn(CO) 3 100 40 0.62 8.110 -3 [6]. Fedotov, A. N.; Simonov, A. P.; Popov, V. K.; Bagratashvili, V. N. J. Phys. Chem. B 1997, 101, 2929-2932. SA, 288055-1G,$31.6 GROUP 8 Fe(C 5 H 5 ) 2 241 50 0.83 4.010 -3 [7]. Cowey, C. M.; Bartle, K. D.; Burford, M. D.; Clifford, A. A.; Zhu, S.; Smart, N. G.; Tinker, N. D. J. Chem. Eng. Data 1995, 40, 1217-1221. [8]. Bartle, K. D.; Burford, M. D.; Clifford, A. A.; Cowey, C. M. Measurement of the solubility of metal complexes in supercritical fluids; Bartle, K. D., Burford, M. D., Clifford, A. A., Cowey, C. M., Eds.; International Society for the Advancement of Supercritical Fluids: Strasbourg, 1994; Vol. 1, pp 419-422. SA GROUP 9 Co(fddc) 2 100 50 0.41 8.810 -5 [9]. Lin, Y. H.; Smart, N. G.; Wai, C. M. Trends Anal. Chem. 1995, 14, 123-133. [10]. Laintz, K. E.; Wai, C. M.; Yonker, C. R.; Smith, R. D. J. Supercrit. Fluids, 1991, 4, 194-198. 14 2

PAGE 28

Table 2.2 Continued 100 50 0.40 5.310 -5 [10] Co(ddc) 2 100 50 0.41 2.610 -7 [9],[10] 100 50 0.40 1.610 -7 [10] Co(acac) 3 290 60 0.83 9.410 -8 [5] SA,C83902,25g,98%,$34.4 Co(acac) 2 290 60 0.83 5.210 -8 [5] GROUP 10 Ni(fddc) 2 100 50 0.41 7.910 -5 [9], [10] Synthesized [15] 100 50 0.40 4.810 -5 [10] Ni(ddc) 2 100 50 0.41 9.410 -8 [9], [10] Synthesized [15] 100 50 0.40 6.010 -8 [10] Ni(hfa) 2 200 60 0.73 4.910 -4 [2] SA, 339709-5G, 98%,$43.5 200 60 0.73 8.010 -3 [2] NiCl 2 [P(C 6 H 5 ) 3 ] 2 220 45 0.83 3.610 -6 [11]. Palo, D. R.; Erkey, C. J. Chem. Eng. Data 1998, 43, 47-48. SA, 15245-5G,>98%,$14.6 GROUP 11 Cu(fddc) 2 100 50 0.41 1.010 -4 [9], [10] Synthesized [15] 15 230 60 0.77 4.010 -3 [13]. C.M. Wai, S. Wang and J.-J. Yu, Solubility parameters and solubilities of metal dithiocarbamates in supercritical carbon dioxide. Anal. Chem. 68 (1996), p. 3516. [15]. Laintz, K. E.; Wai, C. M.; Yonker, C. R.; Smith, R. D. Anal. Chem., 1992, 64, 2875-2878. Cu(ddc) 2 100 50 0.41 1.210 -7 [9], [10] Synthesized [15] 230 60 0.77 1.110 -5 [13] Cu(bdc) 2 230 60 0.77 7.210 -4 [13] Synthesized [12] Cu(hdc) 2 230 60 0.77 2.810 -3 [13] Synthesized [12] 3

PAGE 29

Table 2.2 Continued Cu(acac) 2 200~290 40~60 0.85~0.83 1.810 -5 ~4.310 -8 [3], [5] FA,AC11065-1000,100g 175 40 0.81 1.610 -5 [3] Cu(bzac) 2 200 40 0.85 6.010 -6 [3] Commercial supplier 175 40 0.81 5.010 -6 [3] Cu(tfbzm) 2 200 40 0.85 2.110 -5 [3] Synthesized [3] 175 40 0.81 1.410 -5 [3] Cu(dmhd) 2 200 40 0.85 2.110 -4 [3] Synthesized [3] 175 40 0.81 1.510 -4 [3] Cu(dibm) 2 200 40 0.85 4.610 -4 [3] Synthesized [3] 175 40 0.81 3.910 -4 [3] Cu(thd) 2 200 40 0.85 5.810 -4 [3] SA, 345083-1G,$27 175 40 0.81 4.510 -4 [3] Cu(tod) 2 200 40 0.85 1.110 -3 [3] Synthesized [3] 175 40 0.81 7.810 -4 [3] Cu(hfa) 2 200 40 0.85 3.810 -3 [3] Aldrich, 335193-5G,$32.3, 175 40 0.81 3.5410 -3 [3] Cu(tfa) 2 200 40 0.85 4.210 -4 [3] Commercial supplier 175 40 0.81 3.510 -4 [3] Cu(hfa) 2 H 2 O 100 40 0.62 8.110 -3 [6] Aldrich, 335193-5G,$32.3, Cu(pdc) 2 230 60 0.77 4.010 -6 [13] Synthesized [13] Cu(p3dc) 2 230 60 0.77 1.210 -4 [13] Synthesized [13] Cu(p5dc) 2 230 60 0.77 1.810 -3 [13] Synthesized [13] Cu Kelex 100 200 60 0.74 5.010 -6 [14]. N.G. Smart, T.E. Carleson, S. Elshani, S. Wang and C.M. Wai,. Ind. Eng. Chem. Res. 36 (1997), p. 1819. Synthesized [14] 16 4

PAGE 30

5 Table 2.2 Continued Cu Cyanex 301 200 60 0.74 4.110 -4 [14] Cytec Industries Inc. Cu Cyanex 302 200 60 0.74 1.510 -4 [14] Cytec Industries Inc. Cu Cyanex 272 200 60 0.74 8.010 -6 [14] Cytec Industries Inc. GROUP 12 Zn[SCSN(n-C 4 H 9 ) 2 ] 2 237 55 0.80 3.010 -5 [12]. Wang, J.; Marshall, W. D. Anal. Chem. 1994, 66, 1658-1663. Synthesized [12] Zn[SCSN(C 2 H 5 ) 2 ] 2 237 55 0.80 1.910 -6 [12] Synthesized [12] Zn(SCSNC 4 H 8 ) 2 237 55 0.80 2.810 -7 [12] Synthesized [12] Zn(acac) 2 290 60 0.83 2.110 -7 [5] F,ICN219908,100G,$1175.8 Zn(pdc) 2 230 60 0.77 9.010 -6 [13] Synthesized Zn(ddc) 2 230 60 0.77 2.410 -5 [13] SA,329703-25g, 98%,$17.6 Zn(p3dc) 2 230 60 0.77 1.510 -4 [13] Synthesized [13] Zn(bdc) 2 230 60 0.77 6.910 -4 [13] F, ICN219924,$587.05 Zn(p5dc) 2 230 60 0.77 3.210 -3 [13] Synthesized [13] Zn(hdc) 2 230 60 0.77 5.810 -3 [13] Synthesized [13] Zn(fddc) 2 230 60 0.77 9.010 -3 [13] Synthesized Hg(fddc) 2 100 50 0.41 5.510 -4 [9] Synthesized 230 60 0.77 1.410 -2 [13] Hg(ddc) 2 100 50 0.41 9.010 -7 [9] Synthesized 230 60 0.77 5.310 -5 [13] Hg(pdc) 2 230 60 0.77 3.410 -6 [13] Synthesized Hg(p3dc) 2 230 60 0.77 2.310 -4 [13] Synthesized[12] Hg(bdc) 2 230 60 0.77 5.610 -4 [13] Synthesized [12] Hg(p5dc) 2 230 60 0.77 2.010 -3 [13] Synthesized [12] Hg(hdc) 2 230 60 0.77 3.810 -3 [13] Synthesized [12] 17

PAGE 31

18 Abbreviations : acac pentane-2,4-dionate; acacBr 3-bromopentane-2,4-dionate; bdc, dibutyldithiocarbamate; bzac, 1-phenylpentane-1,3-dionate; ddc diethyldithiocarbamate; dibm 2,6-dimethylheptane-3,5-dionate; dmhd, 1,1-dimethylhexane-3,5-dionate; fddc bis(trifluoroethyl)dithiocarbamate; hdc, dihexyldithiocarbamate; hfa, 1,1,1,6,6,6-hexafluor opentane-2,4-dionate; pdc, pyrrolidinedithiocarbamate; p3dc, dipropyldithiocarbamate; p5dc, dipentyldithiocarbamate; tfa 1,1,1-trifluoropentane-2,4-dionate; tfbzm 1,1,1-trifluoro-4-phenylbutane-2,4-dionate; thd 2,2,6,6-tetramethyl heptane-3,5-dionate; tod 2,2,7-trimethyloctane-3,5-dionate; tta 1-thienyl-4,4,4-triflu oropentane-1,3-dionate. Cyanex 272, di-2,4,4-trimethylpentyl phosphinic acid; Cyanex 301, bis(2,4,4-trimethylpentyl) dithiophosphinic acid; Cyanex 302, bis(2,4,4-trimethylpentyl) thiophosphinic acid; Kelex-100, 7-(4-ethyl-1-methyloctyl)-8-hydroxyquinoline; CpMn(CO) 3 Cyclopentadienylmanganese tricarbonyl SA: Sigma-Aldrich FA: Fisher-Acros 2.2 Introduction to Sol-gel Technology Sol-gel process generally involves the use of inorganic or organic salts as well as metal alkoxides or nitrates as precursors. Hydrolysis and polycondensation reactions occur when the precursors are mixed with water and catalyst. The further condensation of sol particles into a three-dimensional network produces a gel, which is a solid phase encapsulating solvent. The gel materials are referred as aqua-gel when water is used as the solvent and alco-gel when alcohol is used. The encapsulated solvent can be removed from a gel by either evaporative drying or supercritical drying. The resulting solid products are known as a xerogel and an aerogel, respectively.

PAGE 32

19 The advantages of the sol-gel route for making catalytic materials are: 1) The ability to maintain high purity materials due to the purity of the starting materials; 2) The ability to impact some of the physical characteristics such as pore size distribution and pore volume; 3) The ability to vary compositional homogeneity at a molecular level; 4) The ability to prepare samples at low temperatures; 5) The ability to introduce several components in a single step; 6) The ability to produce sample s in different physical forms. 2.2.1. Sol-gel Chemistry Metal alkoxides have been widely used as precursor in sol-gel preparation because they are commercially available in high purity and their solution chemistry has been documented (Bradley, 1989).The sol-gel synthesis of metal oxides is based on the poly-condensation of metal alkoxides M (OR) z in which R is usually an alkyl group (R=CH 3 C 2 H 5 ,...) and z the oxidation state of the metal atom M Z+ Sol-gel chemistry with metal alkoxides can be described in terms of two classes of reactions known as hydrolysis reaction and condensation reactions as shown in equat ions 2.3, 2.4 and 2.5. Metal alkoxides are not miscible with water and have to be dissolved in another solvent, currently the parent alcohol prior to hydrolysis. (1) Hydrolysis reaction: -M-OR + H 2 O -M-OH + ROH (2.3) where M: Si, Ti, Zr, Hf, Ta, Nb, and Al, etc. and R: CH 3 C 2 H 5 C 3 H 7 etc. (2) Condensation reaction: leading to the formation of bridging oxygen -M-OH + XO-M -M-O-M+ XOH (2.4) or

PAGE 33

-M-OH + -M-OH -M-O-M+ H 2 O (2.5) where X can either be H or R (an alkyl group). Figure 2.2 Schematic Diagram for Sol-gel Routes, Preparation of Metal Oxide Gel through Sol-gel Method 20

PAGE 34

21 The condensation of hydroxyl and or alkoxy groups can be considered as a type or class of inorgani c polymerization. When the extent of polymerization and cross-linking of polymeric molecules become extensive, the entire solution becomes rigid and a gel is formed. The size and degree of branching and cross-linking of the inorganic polymer st rongly impact the porosity of the gel, and later the surface area, pore volume, pore size distribution, and thermal stability of the final oxide after heat treatment. The relative rates of hydrolysis and condensation of the hydrolysis and condensation determine the extent of branching and cross-linking of the inorganic polymer and colloidal aggregation in the gelation mixture (Livage, 1988). When the condensation reaction is relatively faster than the hydrolysis reaction, the resulting sol-gel is highly branched and the corresponding gel is more mesoporous in structure. Whereas, if the hydrolysis reaction is relatively faster than the condensation reaction. The resulting sol-gel is weakly branched and the corresponding gel is more microporous in structure. Both relative reaction rates are functions of many parameters including temperature, pH va lue, concentration of precursor, and the amount of water present. 2.2.2. Aging Aging is the process between the formation of a gel network and the removal of solvent from that solid gel ne twork. When the pore liquid remains in the matrix, a gel is not static and can undergo many changes (Scherer and Brinker, 1990). During aging, hydrolysis and condensation continue to occur. Variables such as temperature and pH value of the gel affect the properties of the sol-gel. This process can actually be desirable because it leads to an enhanced cross-linking networ k that is mechanically stronger and easier to handle and also increases the viscosity of the mixture.

PAGE 35

22 Gel aging is an extension of the gelation step in which the gel network is reinforced through further condensation. Syneresis, the expulsion of solvent due to gel matrix shrinkage can occur during gel aging. During the following step drying, solvent trapped in the porous gel matrix is removed. As the solvent evaporates, surface tension or capillary force accompanying evaporation pulls on the wa lls of the pores, therefore, the pore structure collapses during drying. In this research, supercritical drying was used for the drying of the aged sol-gel ma terial to free it from its solvents in order to maintain the pore size and volu me. Supercritical drying is covered in detail in a separate section in this chapter. 2.2.3. Heat Treatment After the wet gel system has been dried and the solid phase has emerged from the drying process, further heat treatment, also called calcination, is needed to stabilize the physical and chemical properties of the solid gel and to make it suitable for future applications at severe temperature conditions (Yao, 2001). Calcination is carried out in an oxidizing or inert atmosphere at a temperature slightly above the projected operating temperature of the catalyst. The aim of calcination is to stabilize the physical, chemical, and catalytic properties of the catalyst. During calcination several reactions and processes can occur (Wijngaarden, 1998). 1) Thermally unstable compounds (carbo nates, nitrates, hydroxides, and organic salts) decompose under gas evolution and are usually converted to oxides. Relevant reactions are shown in equations 2.6 and 2.7 2) Decomposition products can form new compounds by solid-state reactions. 3) Amorphous regions can become crystalline at very high temperatures.

PAGE 36

23 4) Various crystalline modifications can undergo reversible conversions. 5) The pore structure and mechanical st rength of precipitated catalysts can change. 2AlOOH Al 2 O 3 +H 2 O (2.6) Ni(OH) 2 NiO + H 2 O (2.7) Various textural and chemical change s of gel are associated with the calcination process. All organic groups such as alkyl group (C n H 2n+1 ) will react with oxygen in air and finally change to water, carbon dioxide and residual carbon. The nickel complex which had been impregnated into the aerogel sample in supercritical impregnation experiments would also react with oxygen; nickel oxide would be generated and kept in the porous network. The by-product for nickel acetate and nickel nitrate will be different because the acetate group will convert to water and carbon dioxide. Nitrogen dioxide will be produced from the reaction between the nitrate group and oxygen oppositely. Some components inside the gel react with oxygen in the air flow or decompose to other compounds which are more st able at this high temperature. The solid gel undergoes phase transformations under different temperature conditions resulting in different crystal and morphological structures that occur during different heating temperatures (Keysar, 1997, Osaki, 1998). The phase transitions of alumina during calcination are non-reversible. Thus, the physical char acteristics of the final product depend on several parameters such as heating temperature, time, and gaseous environment. It is common practice to subject the sample to a more severe heat-treatment than it is likely to encounter in a reactor to ensure the stability of its structural and textural properties during industrial operations in the future.

PAGE 37

24 Due to viscous sintering, micro-structural pores were found to be diminishing (Yao, 2001) or disappearing completely (Gonzalez-Pena, 2001) during calcination as determined by the nitrogen adsorption isotherms. Pore size distribution becomes narrower after calcination (Suh, 1996) due to the reduction or elimination of the micropores in the solid gel. 2.3 Surfactant Templates The exploration of the supramolecular templating technique has created the ability to synthesize an array of mesoporous metal oxide materials with potential application in various areas that include catalysis, separation, chemical sensing, environmental remediation, and optics (Ying, J.Y. et al., 1999). The use of template effects in th e synthesis of new materials was first demonstrated by the pioneering work of Busch on metal template effect exploration in the synthesis of macr o-compounds in the 1960s (Gerbeleu, 1999). Many materials prepared by the sol-gel technique, especially the aerogels, are mesoporous with large surface area and high pore volume. However, those materials usually have a broad pore size distribution as well. Since the discovery of the M41S family by Mobile scientists in 1992, synthesis and application of ordered mesoporo us materials using surfactants as pore-directing agents has attracted wide attention. 1. Templates Silica porous material was synthesi zed through using a neutral chemical templating material for the first time in 1995 (Pinnavaia, 1995). This templating methodology had some advantages over the former templating methodology used for M41S material due to the fact that the neutral templates are less attached to the solid struct ure and therefore can be removed easily. Easy removal for the nonionic template from the templated solid network

PAGE 38

25 exhibits many advantages for the templated material, such as, improvement of the textural mesoporosity, sharper pore size distribution and thicker framework walls. Using the neutral template, it was possible to recycle the cost-intensive neutral chemical template. 2. Surfactants Surfactants are classified on the nature of the hydrophobic groups. There are four general groups of surfactants, which are defined as: anionic, cationic, nonionic and amphoteric. In all four cl asses the hydrophobic group is generally a long-chain of hydrocarbon radicals which will influence the size and sharp of pores in gel. The hydrophilic group will be an ionic or highly polar group that can impart some solvent solubility to surfactant molecules. Surfactants in the sol-gel process help to make molecular assembling, which is defined as a spontaneous association of molecules into thermodynamically or kinetically stable, structurally well-defined aggregates joined by non-covalent bonds. Surfactant can be used in many different ways for templating. However, using nonionic surfactants seem to have more advantages than others, such as, a significantly lower sensitivity to the pres ence of electrolyte in the system and a small affect on the system pH value. Due to the neutral electrical property, nonionic surfactants do not to provide th e precursor with extra electrical charge which will limit the crosslinking of the units and compromise the thermal stability of the porous framework. 2.4 AluminaNickel/Alumina Catalyst System Alumina and nickel/alumina cata lyst systems formed by sol-gel techniques have been significantly studied and used as catalyst within the last twenty years, this is due to (Janosovits, 1997) their unique properties, such as, porosity, thermal stability, chemical activity, availability and affordability.

PAGE 39

26 2.4.1. Alumina Catalyst Support Alumina support synthesized through the sol-gel method shows superior properties, such as high purity and homogeneity at molecular scale as well as allowing low temperature preparation (Janosovits, 1997). Alumina is formed through a hydrolysis reaction of alum inum precursors which can be either inorganic or organometall ic aluminum precursors. The final properties of alumina are affected by a variety of preparation parameters, especially the stirring and pH value. The synthesis step is followed by aging, solvent washing, drying and dehydration which are also going to impact the final properties of alumina. The templated alumina sol-gel is obtained through hydrolysis of alumina in the presence of nonionic surfactant (Pinnavaia, 1998, Gonzalez-Pena, 2001, Pinnavaia, 2002). Templated alumina gel shows a higher pore volume, more uniform pore size distribution, larger su rface area and higher thermal stability than non-templated alumina gels. The aver age pore diameter was correlated to the chain length of the surfactant. 2.4.2. Nickel/ Alumina Catalyst System There are two basic pathways with different initial preparation processes to synthesize nickel/alumina catalyst system. During the co-precipitation pathway, two metal precursors (organomet allic aluminum and a nickel complex) can be employed to synthesize a co-precipitated gel as starting materials. The resulting metal oxide system is highly dispersed throughout the support after drying and calcination (Osaki, 1998). Nickel-alumina aerogel ca talysts were prepared by the sol-gel method and dried under supercritical condition. In a similar way, nickel-alumina xerogel

PAGE 40

27 was made by drying the gel under ambien t conditions (Suh, 1998). The aerogel catalysts exhibited improved catalytic activity and excellent stability for long reaction times compared with the alumina xerogel with same nickel content. In another pathway, the nickel complex is dissolved to form a solution and then impregnated into alumina system fo llowed by drying and heat treatment. Supercritical fluid impregnation has taken more focu s recently due to its excellent effectiveness and tunable controlling properties compared to the conventional liquid impregnation, especial ly for the highly mesoporous alumina catalyst support system.

PAGE 41

CHAPTER THREE EXPERIMENTAL WORK In this chapter, the experimental methods, experimental equipment setups and procedures, materials and analysis procedures applied for this research will be described. There are seven major sections included in this chapter: Sol-gel preparation, Sol-gel solvent replacement, supercritical extraction and drying, calcination and heat treatment, supercritical impregnation, catalyst and catalyst support characterization. Detailed information and description of instruments and chemicals used in this research are listed in the appendices. 3.1 Sol-gel Preparation Preparation procedures for synthesizing templated alumina gel catalyst support, regular co-precipitate nickel-oxide/alumina (NiO/Al 2 O 3 ) gel and templated co-precipitate nickel-oxide/alumina (NiO/Al 2 O 3 ) gel are described in sections 3.1.1, 3.1.2 and 3.1.3 respectively. Regular gel in this study refers to the synthesis of the alumina gel without using surfactants as template agent (or template) in the sol-gel step. On the contrary, a templated gel means the alumina gel used surfactants as the template. 3.1.1. Synthesis of Sol-gel for Templated Alumina Preparation procedure for templated alumina will be discussed in this section. After the synthesis of sol-gel, xerogels and aerogels are obtained. Xerogel is achieved through ambient drying methods and the aerogel is achieved through supercritical fluid extraction (SFE), respectively. A schematic diagram for the steps involved in the synthesis of sol-gel for templated alumina is shown in Figure 3.1. 28

PAGE 42

In preparation procedure, first of all, two solutions were prepared. The first solution contains ASB (aluminum tri-sec-butoxide) (Aldrich) with the surfactant dissolved in sec-butanol (Aldrich). The surfactant used in this solution was Triton X-114 (polyoxyethylene (8) isooctylphenyl ether) (Sigma) which is a non-ionic surfactant. The required amount of the surfactant was dissolved in one half of the required amount of the sec-butanol, and then ASB was added to the solution. The second solution contains the required amount of de-ionized water which was mixed with the other half of the required sec-butanol. The latter solution was then added to the former drop by drop under continuous stirring of the mixture by using a magnetic stirring device. The mixture was stirred for another hour at room temperature resulting in spontaneous gelation. The gel was stored and covered by parafilm at room temperature for 24 hours for aging purposes. The molar ratio of ASB: sec-butanol: Triton X-114: H 2 O used in making the sol-gel here was 1:20:X:2, respectively. The required amount of water in this sol-gel preparation was according to the ASB hydrolysis reaction shown in Equation 3.1 The affect of surfactant concentration, X, on aerogel pore structure was studied and optimized in a former study (Al-ahamdi, 2004). In this study, surfactant to ASB ratio X was fixed at 0.1. The effect of the sec-butanol amount used in the aerogel preparation is discussed in chapter 4. (CH 3 -CH 2 -CH(CH 3 )-O) 3 Al + 2 H 2 O AlOOH + 3 CH 3 -CH 2 -CHOH-CH 3 (3.1) The aged sol-gel sample was then filtered using a vacuum filtering setup and washed with the required amount of ethanol (Aldrich) before performing the supercritical extraction. 29

PAGE 43

Procedure for making templated alumina sol-gel: Solution one for the sol-gel: 1. The Surfactant Triton X-114 (5.66 g) was dissolved in sec-butanol (74 g). The solution was stirred until it formed one liquid phase. 2. ASB (24.65 g) was added to the solution, which was stirred until the ASB dissolved completely in the solution. Solution two for the sol-gel: 1. H 2 O (3.6 g) was added to the other half sec-butanol (74 g). 2. The solution was mixed to form one liquid phase solution refer to Fig. 3.1. Forming the templated sol-gel solution: 1. Under continuous stirring, solution 2 was added to solution 1 dropwise (about 25-30 drops per minute), resulting in aluminum hydroxide gel (AlOOH). 2. The mixed sol-gel solution was covered with parafilm and stirred for one hour. 3. The formed sol-gel was stored for aging purpose at room temperature for 24 hours 3.1.2. Synthesis of Sol-gel for Regular Co-precipitate Nickel Oxide/Alumina For synthesis of regular nickel oxide catalyst supported by alumina, anickel complex and aluminum oxide were co-gelled or co-precipitated in one step without a template as described in this section. A schematic diagram for synthesizing regular nickel oxide-alumina is shown in Figure 3.2. 30

PAGE 44

Figure 3.1 Schematic Diagram for Synthesizing and Processing Templated Alumina Sol-Gel 31

PAGE 45

In this research, the basic molar ratio of ASB: sec-butanol: nickel complex: H 2 O: methanol used in making this co-precipitated sol-gel here was 1:20:0.15:4:4.5, respectively. Nickel acetate (NiAc) and nickel nitrate (Ni(NO 3 ) 2 ) (Sigma-Aldrich) were chosen as the nickel complexes during impregnation because both can be easily converted to nickel oxide by calcination. However, nickel acetate was used during co-precipitation, because it could be hydrolyzed easily into nickel hydroxide. Methanol was used as a co-solvent for dissolving the nickel complex and the molar amount of methanol was 30 times that of the nickel complex. The required amount of water in this sol-gel preparation was according to the ASB hydrolysis reaction; the nickel acetate hydrolysis reaction are shown in Equation 3.1 and Equation 3.2, respectively. (CH 3 -COO) 2 Ni + 2 H 2 O Ni(OH) 2 + 2 CH 3 COOH (3.2) Procedure for making regular nickel oxide-alumina sol-gel: Solution one for the sol-gel: 1. ASB (aluminum secondary butoxide) (4.92 g) was dissolved in sec-butanol (29.65 g). 2. The solution is then stirred until the ASB dissolved completely in the sec-butanol. Solution two for the sol-gel: 1. The designed amount of nickel acetate tetrahydrate crystals ((CH 3 -COO) 2 Ni 4H 2 O) (Aldrich) (i.e., the designed amount to give a certain Ni/Al molar ratio) was mixed with methanol in a molar ratio of 1:30. 2. The mixture solution was then stirred continuously until the nickel acetate tetrahydrate crystals dissolved completely. 32

PAGE 46

Figure 3.2 Schematic Diagram for Synthesizing and Processing Regular Nickel Oxide-Alumina Sol-gel 33

PAGE 47

3. The amount of H 2 O needed for the hydrolysis of the solution process (i.e., H 2 O needed for both of the nickel acetate and for the ASB hydrolysis reactions) was added. The solution was further stirred to ensure miscibility and the formation of a transparent green liquid phase solution. Forming the regular nickel oxide-alumina sol-gel solution: 1. Under continuous stirring, solution 2 was added to the solution dropwise. 2. After the addition of solution 2 to solution 1 was completed, the resulting sol solution was covered with parafilm to prevent evaporation of the solvents and condensation from the air-humidity into the sol solution. The sol solution was further stirred for one hour. The formed sol-gel is then stored and kept for aging purpose at room temperature for 24 hours. 3.1.3. Synthesis of Sol-gel for Templated Co-precipitate Nickel Oxide/Alumina For the synthesis of templated nickel oxide catalyst supported by alumina, the nickel complex and alumina were co-precipitated in one step with surfactant template as described in this section. The surfactant Triton X-114 was first dissolved in the sec-butanol, and then ASB was added to the solution slowly under continuous stirring. Simultaneously, the nickel complex was dissolved in methanol with a molar ratio of 1:30 and the resulting solution was then mixed with de-ionized water. The second solution was added to the first dropwise under continuous stirring for one hour at room temperature. The gel was then stored at room temperature for 24 hours to age. The molar ratio of ASB: sec-butanol: Triton X-114: H 2 O used for preparing this sol-gel was 1: 20: 0.1: 2, respectively. A schematic diagram for synthesizing templated nickel oxide-alumina is shown in Figure 3.3. 34

PAGE 48

Procedure for making templated nickel oxide-alumina sol-gel Solution one for templated nickel oxide-alumina sol-gel: 1. The surfactant X-114 (1.12 g) was dissolved in sec-butanol (29.65 g). 2. The solution was stirred until the components mixed together and formed one liquid phase. ASB (aluminum sec-butoxide) (4.92 g) was then added to the solution and stirred until the ASB dissolved completely. Solution two for templated nickel oxide-alumina sol-gel: 1. The desired amount of nickel acetate tetrahydrate crystals ((CH 3 -COO) 2 Ni4H 2 O) (0.75 g) (the desired amount used in this section is to give a Ni/Al=0.15/1 ratio) was mixed with methanol (2.88g) in a molar ratio of 1:30. 2. The solution of the nickel acetate tetrahydrate and the methanol was then stirred continuously until all the nickel acetate tetrahydrate crystals were dissolved completely. 3. The amount of H 2 O (1.44g) needed for the hydrolysis for the solution process (i.e., H 2 O needed for the nickel acetate and for the ASB hydrolysis reactions) was added to the solution in this step. The solution was further stirred to ensure the formation of one liquid phase. Forming the templated nickel oxide-alumina sol-gel solution: 1. Under continuous stirring, solution 2 was added to solution dropwise. 2. After the addition of solution 2 to solution 1 was completed, the resulting sol solution was covered with a parafilm to prevent evaporation of the solvents or condensation from the air-humidity to the sol solution. The sol solution was further stirred for one hour. 3. The formed sol-gel is then stored and kept to age at room temperature for 24 four hours. 35

PAGE 49

Figure 3.3 Schematic Diagram for Synthesizing and Processing Templated Nickel Oxide-Alumina Sol-gel 36

PAGE 50

3.2 Sol-gel Solvent Replacement The sol-gel solvent replacement began after the sol-gel has been synthesized and aged in one of the three methods discussed in section 3.1. The sol-gel solvent replacement procedure includes vacuum filtration and washing. The aged sol-gel sample was placed on a vacuum filtration device to remove the visible part of the sol-gel solvent. Then the sol-gel was washed with 20-30 mL of ethanol and the solution was filtrated under vacuum to replace the sol-gel solvent with ethanol. 3.3 Supercritical Extraction and Drying The templated alumina sol-gel, regular and templated co-precipitate nickel oxidealumina sol-gel were dried under supercritical condition to form the aerogel been after being synthesized as described in section 3.1; then filtered and washed with ethanol as described in section 3.2. Experimental procedures and set up used in this research are discussed in this section. 3.3.1. SCF Extraction and Drying Experimental Setup The supercritical fluid (SCF) apparatus used for supercritical extraction and drying is shown in Figure 3.4. The SCF extraction and drying apparatus can be divided into three main sections, feeding section, drying section and gas-liquid separation and collection section. The maximum design pressure for the entire system is 3000 psi. All parts, tubing and Swagelock fittings are made of stainless steel, (SS-316). The tubing used in the main fabrication of the SCF setup is mainly 1/4 inch in diameter. The feeding section consists of a compressed carbon dioxide tank (AIRGAS, CD-50), pressure regulator (Matheson), refrigerating circulator (Ecoline RE120, Lauda-Brinkmann), HPLC pump (600E, Waters), entrainer container, syringe pump (ISCO model 500DX), in-line mixer, magnetic pump (2330-802, Ruska), and number of plug and needle valves. 37

PAGE 51

Figure 3.4 Schematic Diagram for the Supercritical Extraction and Drying Set-up 38

PAGE 52

The drying section consists of a 100mL controlled heat-jacketed autoclave (Autoclave Engineer), water circular bath (HAKKE, model B81), several electrical heating tapes (Omega), pressure gauge (Matheson), and thermocouple Probe (Omega, K-type). The gas-liquid separation and collection consists of a compress nitrogen tank (AIRGAS, NI200), backpressure regulator, and low-pressure gas-liquid separator and venting system. 3.3.2. Procedure for SCF Extraction and Drying When the alcogel sample is dried at ambient pressure or vacuum to produce xerogel, the solvent remained in alcogel sample evaporates and then the porous structure of xerogel collapses due to the capillary forces (CF). Whereas, with particular permeability and transport properties, enhanced dissolvability, supercritical fluid (SCF) extraction and drying is aimed at eliminating the liquid-vapor interface which opposes further diffusion and the accompanying capillary pressure associated with it. SCF extraction drying was employed on the alcogel samples to produce aerogel having unique properties such as high porosity, high specific surface areas (SSA), uniform pore size distribution and good structural stability. The procedure of supercritical extraction and drying is composed of three major stages: supercritical pre-drying by ScCO 2 only, supercritical extraction (surfactant removal step) by ScCO 2 and entrainer, and final supercritical drying. First, most of solvent in the alcogel is disengaged and removed from the system in the first stage. After that, the mixture of ScCO 2 and entrainer is able to extract most of surfactant template from alcogel sample without disturbing the porous structure. Finally, ScCO 2 is able to remove most of the residual solvent and surfactant from aerogel in the final supercritical drying stage. The trace amount of surfactant and solvent in aerogel is totally eliminated by calcination and heat treatment which is discussed in section 3.5. 39

PAGE 53

The procedure of Supercritical Extraction and Drying Stage one: Pre-drying with ScCO 2 only 1. The alcogel sample was placed into a heat-jacketed autoclave which had been pre-heated to a set-point temperature of 62.5 o C. The autoclave was then closed tightly and purged with carbon dioxide gas. All the vents and outlet valves were then securely closed and the system pressure was first raised to the compressed CO 2 tank pressure 850 psi. Meanwhile, backpressure system was turned on and set at 1800 psi. 2. Carbon dioxide was compressed and cooled down to -4 o C passing through syringe pump, and pressure in the system was kept increasing till it reached the desired set point pressure which was determined and entered into the backpressure regulator at the start of each experimental run. The designed processing pressure for this section was 1800 psi. ScCO 2 would go to an in-line mixer unit just before it was introduced to the autoclave from the bottom section. When the setting pressure was achieved, the feeding flow rate of CO 2 was kept at 10 ml/min for 2 hours. 3. The recycle pump was then engaged between the input (bottom) and the output (top) of the autoclave. At the beginning of the run, the recycled fluid was passed through the vessel, named silica bed, which contained silica gel particles as the adsorbent. The mixture of carbon dioxide, part of solvents and residual template was evacuated from the SCF experimental setup through the backpressure regulator. The electrical heating element was used to keep the effluent supercritical mixture from freezing, hence preventing blocking of the system. Fluid leaving the backpressure regulator went to an ambient pressure vessel inside which the effluent mixture was separated into gas and liquid mixture phases. 40

PAGE 54

Stage Two: supercritical extraction with ScCO 2 and entrainer Under the experimental condition on stage one, the entrainer feed (5%) was introduced to the experimental system through the HPLC pump to be mixed with ScCO 2 In this experiment, ethanol (Sigma-Aldrich, HPLC, 200 proof) was used as the entrainer. Experimental running time for this condition was 4 hours. Stage Three: Final Supercritical Drying, ScCO 2 only 1. Silica Bed containing the adsorbent silica gel particles was isolated from the rest of the system. The recycle flow was redirected through parallel tubing to prevent the adsorbed solvent and template content from desorbing at depressurization step of the SCF drying process and reaching back to the dried aerogel. 2. Entrainer flow was then terminated. Only the ScCO 2 flow was kept running for drying purpose for 2 more hours. 3. After stopping the feed flow, all inlet valves were closed and the vent valve was turned on slowly in order to depressurize the system gradually. 3.4 Supercritical Impregnation After the supercritical extraction and drying process described in section 3.3, most of the template surfactant in aerogel porous network system was removed resulting in a highly porous structure. Dried alumina aerogels were then impregnated with the nickel complex under supercritical conditions to form a catalytically functionalized aerogel. The experimental procedures and setup used in this research are discussed in this section. 3.4.1. Supercritical Impregnation Setup The similar experimental setup which had been shown in Figure 3.4 was also used for the supercritical fluid (SCF) impregnation setup except the absorbent cell, silica bed, was replaced with a nickel complex cell and another 41

PAGE 55

system vent was added to the bottom of the autoclave. The diagram for the supercritical impregnation apparatus setup is shown in Figure 3.5. 3.4.2. Procedure of SCF Impregnation In this research, two kinds of experimental methods, continuous flow and batch experiments were used to achieve supercritical impregnation. Both of these experimental procedures are discussed in sections 3.4.2.1 and 3.4.2.2 respectively. 3.4.2.1. Continuous Impregnation The continuous impregnation pathway is a time dominated procedure. The adsorption of solute on the interior wall of pores was dependent on how soon and how long the concentration equilibrium could be reached. The excess amount of nickel complex as solute was loaded in the nickel complex cell. Solute was then dissolved in a supercritical solvent mixture which was composed of carbon dioxide flow at 10ml/min and methanol flow at 0.5 ml/min. The supercritical solution then passed into autoclave where the supercritical dried aerogel was placed. With outstanding permeability, the supercritical solution was able to fill out and permeate the meso-pore aerogel structure completely. After several hours of continuous experimental running at the designed temperature and pressure with the aid of the recycle pump, the solute concentration equilibrium was finally achieved. After stopping all feed flow and closing all inlet valves, depressurizing the entire system quickly caused the solubility of solute to decrease and the solute to precipitate in the pores with no solvent left inside the pores structure. 42

PAGE 56

Figure 3.5 Schematic Diagram for the Supercritical Impregnation Setup 43

PAGE 57

3.4.2.2. Batch Impregnation The batch impregnation pathway is a diffusion dominated procedure. The adsorption of solute on the interior wall of pores was dependent on diffusivity of the solute transferring into the pores. Solubility of the nickel complex in various conditions is the key parameter to control the entire impregnation procedure. The initial experimental setup and procedure was the same as the one used for the continuous experiments. The system was pressurized to achieve the desired pressure first. After running for 30 minutes with high mixture flow rates (CO 2 10ml/min, MeOH 0.5ml/min), the solvent mixture flow was reduced to 2ml/min for carbon dioxide flow and 0.1 mL/min for the entrainer methanol flow; the experiment ran for another 30 minutes in order to keep the system pressure constant. During the low flow period, a relative static homogeneous supercritical environment was maintained in the autoclave. Solute was able to diffuse into the pores uniformly. After stopping all feeding flows and closing all inlet valves, depressurizing the entire system quickly caused the solubility of solute to decrease dramatically and the nickel complex to precipitate in pores without solvent being left inside the pores. The procedure was repeated 2 to 3 more times in order to achieve high solute precipitation in the pores. The operating pressures and temperatures were varied in each experiment. High pressures and low temperatures caused higher solute solubility than lower pressures and higher temperatures. Hence, the preferred operation condition for batch experiment was fixed at 40 o C and 2500 psi. The schematic diagram for processing the supercritical batch impregnation is shown in Figure 3.6 44

PAGE 58

Figure 3.6 Schematic Diagram for Processing Supercritical Batch Impregnation 45

PAGE 59

3.5 Calcination and Heat Treatment Both of samples that were prepared as described in sections 3.3 and section 3.4 need to be further heat treated. After disengaging the liquid from the porous structure of the aerogel sample in the supercritical extraction and drying process, further heat treatment is necessary to convert the aerogel into a catalytically useful form and to achieve structural stability. Heating is done in the presence of an air flow in order to burn off any residual solvent and surfactant from the aerogel (Caruso, 2000). A calcination device was designed to meet the following objectives: (1) holding enough quantity of the sample, (2) with standing the high calcination temperatures, (3) allowing a direct temperature measurement through an inserted electronic temperature probe, (4) preventing the sample particle from being carried out of the device by the flowing air, and (5) loading and unloading the sample easily. The designed schematic diagram of calcination device is shown in Figure 3.7 for horizontal and vertical position settings respectively. The dried aerogel or impregnated nickel/alumina aerogel were placed in the calcination device. Samples are placed in the middle of a 1-inch diameter and 10-inch long stainless steel (SS-316) tube (Swagelock) in a horizontal setting, otherwise at one end of the calcination tube when it was set vertically. The tubing was then connected to the airflow lines and to a secondary temperature measurement probe device that works in addition to the temperature displayed on the control panel of the tube furnace for confirmation and better accuracy. After the whole calcination assembly was placed into the tube fFurnace (Thermodyne model 21100), the airflow was introduced first from a compress air tank (AIRGAS, breathing quality) regulated at 10 psig; then the tube furnace temperature was raised gradually the desired calcination temperature. The sample was calcined for three hours at 500 o C. 46

PAGE 60

(A) Designed Sample Calcination Device in Horizontal Position Figure 3.7 (B) Designed Sample Calcination Device in Vertical Position 47

PAGE 61

3.6 Catalyst and Catalyst Support Characterization and Analysis Setups All samples that were synthesized with the aid of supercritical fluids were characterized and analyzed before and after heat treatment which determined their properties and the influence of the various synthesis parameters and procedures. Structural properties, including specific surface area (SSA), average pore volume (APV), average pore size or pore diameter (APD), pore size distribution and meso-porosity were determined by using nitrogen adsorption-desorption isotherms (NOVA 2000, Quantachrome). The surface morphology, surface images, and elemental composing were characterized by secondary electron microscopy & energy dispersion spectrum (SEM-EDS) (HITACHI S800) analysis and x-ray photoelectron spectroscopy (XPS) analysis. 3.6.1. Surface and Structural Characterizations The surface area, average pore size and pore size distribution were the vital results of the structural and textural measurements needed to be characterized by using the Nova 2000 in this section. Nitrogen was the only gas to be applied in the analysis of all samples in this system. The Brunauer-Emmett-Teller (BET) method (Brunauer,S. et al, 1938) is the most widely used procedure for the determination of surface area of solid materials and involves the use of the BET equation 3.3 00111/1PPCWCCWPPWmm 3.3 It results in a linear plot of 1/[W/(P 0 /P)-1] vs. P/P 0 for most solids using nitrogen as the adsorbate. Nitrogen adsorption-desorption data at 77 K which is the standard boiling point of nitrogen were collected and analyzed. The total specific surface areas (SSA) for as-made and calcined samples were obtained by using a multi-point BET method which is restricted to a limited region of the 48

PAGE 62

adsorption isotherm, usually in the P/P o range of 0.05 to 0.30. The results are presented and discussed in Chapters 4. Porosity of powders and other porous solids can be conveniently characterized by gas adsorption studies. Two common techniques for describing porosity are the determination of total pore volume and pore size distribution. The total pore volume is derived from the amount of vapor adsorbed at a relative pressure P/P o close to unity, by assuming that the pores are then filled with liquid adsorbate. For evaluation of the porosity of most solid materials, nitrogen at 77K is the most suitable adsorbate. Pore volumes, pore sizes and pore size distributions of the alumina support network before and after heat-treatments were calculated by using the NOVA data reduction system through BJH (Barrett, Joyner and Halenda, 1951) method utilizing the nitrogen desorption isotherm between P/P o range of 0.99 and 0.05. The results are presented and discussed in Chapters 4. The existence of micropore structure in samples was investigated using the t-test (de Boer method) for confirmation. When a sample contains micropores, the t-test gives the micropores specific surface area and the pore volume. NOVA is an acronym for NO Void Analysis. The NOVA performs rapid and accurate physical adsorption/desorption measurements of nitrogen gas (or any other non-corrosive gas) on solid surfaces. Specification for the equipment and the experimental adsorption/desorption conditions is shown in Table 3.1. Measurements that can be made by the NOVA include: multipoint BET surface area, single point BET surface area, external surface area (STSA), 100 point adsorption isotherms, 100 point desorption isotherm, total pore volume, 49

PAGE 63

average pore radius, BJH pore size distribution based on the adsorption or desorption isotherms, approximate sample volume and density, and microporous surface area and volume. Table 3.1 Specifications for the Quantachrome NOVA 2000 (High Speed Gas Sorption Analyzer Version 6.11 Instrument Nova Model 2000, from Quantachrome Corporation Data analysis Enhanced Data Reduction Software, Version 2.13 Adsorption gas Nitrogen, purity 99.99% Bath temperature 74K, liquid nitrogen Pressure range P/P o = 0.05 to 1.0 Sample size Between 0.05 and 0.5 g (Enough sample to have total surface area between 2 and 50 m 2 ) Sample cells Standard 9 mm cell Vacuum machine 1.5 hp rotary pump, Edwards The procedure for performing a textural characterization using the NOVA 2000 involves the following steps. 1. Sample Degassing. An amount of the sample (sufficient quantity to give total surface area between 2-50 m 2 ) is loaded into the specific sample cell, then the cell(s) is placed in the pouch of the heating mantel, the top of sample cell is inserted into a fitting to the degassing station in the NOVA machine. Temperature is set to the required degas temperature which was 80 o C in this research and the heating mantel was switched on. The degas station then automatically operated under vacuum conditions for 3 hours. 2. Sample Analysis After degassing, the degassed sample was cooled down gradually to room temperature and re-weighed to get the actual degassed sample weight. Then the sample was re-loaded into the same sample cell and connected to the analysis stations. The sample cell was placed in the liquid nitrogen Dewar flask 50

PAGE 64

that was filled with an adequate amount of liquid nitrogen in advance. The desired analysis type, e.g., BET, BJH, analysis condition and the data reduction parameters are selected step by step. Pre-selected sample run conditions and sample cell calibration files are chosen from the system database for both of the sample and the sample cell that contains the sample. The real analysis time was dependent on the analysis setting and the total surface area of the samples. The sample weight was put in prior to the analysis. At the end of the analysis, data were saved to a floppy disk and were ready to be analyzed by the Enhanced Data Reduction software. 3.6.2. Scanning Electron Microscopy(SEM) and Energy Dispersion Spectroscopy (EDS) Scanning electron microscope (SEM) and energy dispersion spectroscopy (EDS) were used to obtain surface images, elemental composition and to study distribution of nickel on the alumina catalyst support. The Hitachi, Model S-800 SEM specifications for analysis are listed in Table 3.2. Table 3.2 Specifications for SEM-EDS Hitachi Model S-800 Machine Instrument Hitachi Model S-800 Data analysis Genesis software Primary electron beam voltage Vo 130 kv Emission extracting voltage V1 0 6.3 kv Emission current 10 A. System pressure 1.13 x 10-9 torr Sample pretreatment 10 nm coating film with Ag Pd filament. Vacuum system Two rotary pumps, a turbo pump and three ion pumps. 51

PAGE 65

A small amount (< 0.01 g) of the sample was placed on the top of the SEM-EDS sample holder using carbon tape as the sample holding media. The sample was coated with gold (Ag) and palladium (Pd) material via a Hummer X sputtering instrument. The coating process was highly recommended when super fine and clear surface images are required because alumina sample and nickel/alumina sample are both electrically non-conductors. In this way the samples are grounded and to not become highly electrically charged which results in blurry SEM images. However, the carbon tape and metallic coating can introduce other elemental signals when the SEM-EDS analysis is processed. The details of the SEM-EDS setups and analysis protocols are discussed in Appendix B and C. 3.6.3. X-ray Photoelectron Spectroscopy (XPS) X-ray Photoelectron Spectroscopy (XPS) is a relatively non-destructive technique that: (1) exposes the sample to ultra high vacuum (UHV) pressures with base pressures of 1 x 10 -10 Torr, (2) increases the temperature by approximately 10-20 o C, and (3) exposes the sample to soft X-rays between approximately 1000 -1500 eV. These conditions are considered nondestructive for most materials and systems with few exceptions. The uses of XPS are elemental analysis of near-surface regions of all elements except hydrogen and chemical state information of near-surface species. Some examples of the information gathered through the use of XPS are the determination of oxidation states of metal atoms in metal oxide surface films and the identification of their surface concentrations. The average analysis depth for XPS is 10-100 Qualitative analysis can be performed in 5 to 10 minutes. Quantitative analysis requires 1 hour to several hours depending on the information desired. The details of XPS setups and analysis protocols were discussed in Appendix D. 52

PAGE 66

3.7 Reagents All reagent used in this research are available from major chemical reagent manufacturers and retailer such as Fisher Scientific Inc., Sigma-Aldrich Inc., and Airgas. Some reagent information is shown in Table 3.3. Ethanol and methanol (200proof or HPLC grade) are provided by Sigma-Aldrich. Deionized (or nano-pure) water is obtained from the Environmental Lab in the Civil Engineering department at the University of South Florida. The purity of compressed CO 2 is breath grade; compressed Air is industrial grade. All gases tanks were provided by Airgas. Table 3.3 Main Reagent Information Reagent Molecular Formula F.W. Density Chemical Structure Aluminum tri-sec-butoxide [C 2 H 5 CH(CH 3 )O] 3 Al 246.33 0.967 g/ml Sec-Butanol C 4 H 10 O 74.12 0.808 g/ml Triton X-114 C 8 H 17 (C 6 H 4 )(OC 2 H 4 ) n OH 426-558 1.022-1.058 g/ml Nickel(II) acetate tetrahydrate C 4 H 6 NiO 4 4H 2 O 248.8 1.798 g/ml Nickel(II) nitrate hexahydrate crystal N 2 NiO 6 6H 2 O 290.8 2.05 g/ml 53

PAGE 67

54 CHAPTER FOUR RESULTS AND DISCUSSION In this chapter, the results are presented and discussed in four main related sections: alumina aerogel preparation, co-precipitated nickel/ alumina catalyst system, supercritical impreg nation, and comparison between SCF impregnation and co-precipitated templated nickel oxide/ alumina system. 4.1 Alumina Aerogel Preparation The details of the procedure for alum ina aerogel preparation are described in Chapter 3. In this study, Nonionic surfactant Triton X-114 was the only surfactant being used in alumina aero gel preparation, and co-precipitated nickel/alumina aerogel preparation was used as the template to create a highly uniform mesoporous structure. Furthermore, supercritical extraction and drying was the unique extraction/drying method used. Some operation or analysis conditions or parameters had been studied and optimized in previous work done by Ahmad Al-Ghamdi or suggested in the instruments user manual such as the NOVA 2000 BET analyzer; othe rs are the subjects of this study. 4.1.1. Effect of Composition The initial composition of reagents used in preparing alumina aerogel obviously impacts the final textural an d chemical properties of the alumina aerogel, including the surface area, porosity, chemical and thermal stability, average pore volume, etc. because different compositions cause directly different gelation rate, hydrolysis and condensation reaction rate, and cross-linking degree.

PAGE 68

Figure 4.1 Schematic Diagram of Templated Alumina Aerogel Preparation with Different Composition In the study described in the following sections, only the molar ratio of sec-butanol was varied. Recipes for two different alumina sol-gel preparations are shown in Figure 4.1. Here, all other conditions for preparation of the aerogels are kept constant. The molar ratios of each component in the first and second recipe are ASB: Sec-Butanol: X-114: H 2 O= 1:20:0.1:2 and 1:10:0.1:2, 55

PAGE 69

respectively. The amount of water was chosen based on the mechanism of hydrolysis reaction. The textural properties of alumina aerogels prepared with different composition were determined before and after calcination via nitrogen adsorption/ desorption isotherms using NOVA 2000 high speed gas sorption analyzer. The comparison results are shown in Table 4.1 and Figure 4.2. The Pore size distribution results for four samples characterized through BJH dV/dD desorption method are also shown in Figure 4.3. Table 4.1 Textural Properties for Templated Alumina Aerogel with Different Composition sample SSA (m 2 /g) TPV (cc/g) APD () 1 st As-made 504.9465 1.10494 87.530 Calcined 518.4409 1.15593 89.185 2 nd As-made 305.0506 0.61783 81.014 Calcined 371.8001 0.81095 87.245 NOTE: TSA: Total surface area SSA: Specific surface area TPV: Total pore volume APD: Average pore diameter 0100200300400500600 SSA APV APD Figure 4.2 SSA, APV and APD of two Alumina Aerogel with Different Composition (Before Calcinations and After Calcinations) 56

PAGE 70

57 Specific surface area (SSA) is a vital property of porous catalyst support. The higher SSA the catalyst support has, the greater the number of actual reacting sites is available for catalyti c reaction. As shown in Figure 4.2, 1 st sample had approximately 50% higher specific surface areas than the 2 nd sample which only showed 305 m 2 /g SSA before calcination and 371 m 2 /g SSA after calcination. This phenomena indicated that enough solvent will provide more space and volume which allows gelation and cross-linking to be carried throughout completely thus generating a highly porous matrix. Use of same surfactant in both samples resulted in almost the same average pore diameters. However, after calcination, the residual surfactant inside the pore was decomposed and removed which can provide more total pore volume of the aerogel than before calcination. The pore size distribution shown in Figure 4.3 indicated the pore volume of the pores which diameters were less than 1.5 nm were strongly reduced after calcination. Meanwhile, the average pore size of samples was increased a little due to the removal of micropores. 4.1.2. Effect of Aging After the sol-gel was prepared for two similar samples, one was left to age for 24 hours before extraction and drying, the other sample aged for 120 hours in order to study the effects of aging on the properties of the aerogels. Sol-gel samples were prepared in the same composition and under same drying and calcination conditions except for the time of aging. The schematic diagram of the comparison is shown in Figure 4.4. A comparison of results is shown in Table 4.2 and Figure 4.5, respectively.

PAGE 71

05101520250100200300400500600700800Pore diameter [A]Pore Volume [cc/g] *10e-3 1st Sample As made 1st Sample After Calcination 2nd Sample As-made 2nd Sample After Calcination Figure 4.3 Pore Size Distribution Of Two Alumina Aerogel With Different Composition (Before Calcination And After) 0 58

PAGE 72

Figure 4.4 Schematic Diagram of Templated Alumina Aerogel Preparation with Aging Periods 59

PAGE 73

Table 4.2 Textural Properties for Templated Alumina Aerogel With Different Aging Period Aging sample SSA (m 2 /g) TPV (cc/g) APD () 24 hrs As-made 142.1738 0.36923 103.882 Calcined 368.7771 0.82643 89.640 120 hrs As-made 356.8457 1.06364 119.226 Calcined 233.3519 0.81110 139.034 Figure 4.5 SSA, APV and APD of Two Alumina Aerogel with Different Aging Periods (Before Calcinations and After Calcinations) With the same SCF extraction and drying time, a longer aging time resulted in larger cross-linking which caused larger specific surface area and greater average pore sizes in the sample before calcination. However, after calcination, the SSA decreased by about 35%; which is even lower than that of the sample with 24 hours aging which means longer aging time is not really necessary for templated alumina sol-gel preparation. Also, it needs to be noted that effect of aging was only tested for the sample with an ASB: BuOH ratio equal to 1:10. The effect of aging was not tested on the sample that had the optimized composition (ASB: BuOH=1:20). 60

PAGE 74

61 4.1.3. Effect of SCF Extraction/Drying Conditions During SCF extraction and drying, the solvent and surfactant template was removed by both scCO 2 and scCO 2 with entrainer. The aerogel porous matrix was left behind after this process with only small amounts of residual solvent and surfactant. Various process parameters influence the textural properties of final alumina aerogel including operating temperature and pressure, SCF flow rates, the percentage of entrainer in SCF flow and the extraction/drying time. 4.1.3.1. Flow Rate In SCF extraction, the surfactant template within the gel solution was first dissolved in a continuous supercritic al fluid mixture which contained a pre-cooled scCO 2 flow pumped by a 500mL syringe pump and an entrainer flow (in this section, ethanol was used) pumped by a HPLC pump. Then the template was carried out of the aerogel matrix and the closed operating environment through the vent system. Three templated alumina aerogel samples were prepared through the same SCF extraction process but with different SCF flow rates, the schematic diagram of the comparison is shown in Figure 4.6. The comparison of the results is shown in Table 4.3 and Figure 4.7. The results are obtained with using the same characterization method described in section 4.1.1. Table 4.3 Textural Properties for Templated Alumina Aerogel with Different SCF Extraction Flow Rates Flow rate (ml/min) Sample SSA (m 2 /g) TPV (cc/g) APD () 1 st CO 2 : 2.5 As-made 196.9852 0.41673 84.662 EtOH: 0.125 Calcined 232.6541 0.40650 69.889 2 nd CO 2 : 5.0 As-made 153.3509 0.33243 86.710 EtOH: 0.25 Calcined 310.1901 0.47969 46.777 3 rd CO 2 : 10.0 As-made 305.0506 0.61783 81.014 EtOH: 0.50 Calcined 371.8001 0.81095 87.245

PAGE 75

Figure 4.6 Schematic Diagram of Templated Alumina Aerogel Preparation with Different SCF Flow Rates 62

PAGE 76

050100150200250300350400 CO22.5ml/min+EtOH0.125ml/min ASMCO22.5ml/min+EtOH0.125ml/min CALCO25ml/min+EtOH0.25ml/min ASMCO25ml/min+EtOH0.25ml/min CALCO210ml/min+EtOH0.5ml/min ASMCO210ml/min+EtOH0.5ml/min CAL SSA TPV APD Figure 4.7 SSA, APV and APD of Three Alumina Aerogel with Different SCF lcination) aerogel increas Ru222 Extraction Flow Rates (Before Calcinations and After Calcinations) The specific surface area of the templated alumina (before ca ed from 196 m2/g to 305 m2/g when the SCF flow rate increased. When flow rate increased theeynolds nmber (Re) of the supercritical fluid mixture was increased, and the flow regime might have changed. Due to the unknown properties of the packed autoclave with alumina gel, the exact value of the Reynolds number was not able to be calculated; the change of flow regime was not finally concluded. During the same running period, at the higher flow rate, more solvent was into SCF extraction system. When the scCO flow rate was equal to 2.5mL/min and the SCF extraction running time was 4 hours (the effective extraction time was 80% of the real running time due to syringe pump refill), around 480 mL scCO was pumped into the autoclave, whereas almost 1920mL of scCO was pumped into autoclave when the flow rate was 10ml/min. At higher flow rate, more surfactant was dissolved in the solvent and carried out of the pores. Therefore, the porosity of alumina aerogel 63

PAGE 77

64 CF extraction and drying in this study is to remo deteifferent Data SSA (m2/g)TPV (cc/g) APD () increased with rising flow rate. 4.1.3.2. SCF Extraction Time The main objective of using S ve the surfactant template and sol-gel solvent from sol-gel matrix as much as possible by taking advantages of the outstanding permeability and solubility of supercritical fluids. However, making the entire process more effective and time-saving is also an important goal for aerogel preparation. Two different SCF extraction time series experiments were done in order to rmine the optimized extraction time for a specific designed aerogel preparation. The schematic diagram of this comparison is shown in Figure 4.9 and the comparison of results is shown in Table 4.4 and Figure 4.8. Table 4.4 Textural Properties for Templated Alumina Aerogels with D SCF Extraction Times Extraction Time Sample 2 hours As-made ge 0 10 Avera 127.7086 .32496 4.0437 STDEA 25.37539 0.03836 19.55500 C alcined Average 363.8187 0.80212 88.50633 STDEA 32.98376 0.03108 5.514599 4 hours s-made A Average 449.7571 1.22791 109.4965 STDEA 17.53766 0.18948 13.76702 Calcined Average 385.3981 1.05453 110.3485 STDEA 42.29524 0.04125 16.39286

PAGE 78

0501001502002503003504004505002 hours sampleAs-made2 hours sampleAfter Calcination4 hours sampleAs-made4 hours sampleAfter Calcination SSA TPV APD 65 Figure 4.8 The Mean Values and Standard Deviations of SSA, APV And APD of Two Series of Alumina Aerogel with Different SCF Extraction Time (Before Calcinations and After Calcinations) As shown in Fig 4.9, when the extraction time was 2 hours, the SSA of the sample after calcination was larger than before calcination which can be attributed to the fact that the solvent and surfactant were not completely extracted from the pores. The solvent or the surfactant covered most of the internal surface area before calcination. After heat treatment, most of the residual solvent and surfactant template evaporated and decomposed so that more surface area in the matrix was available. On the other hand, there may also be some loss of surface area due to sintering. Apparently, the amount of surface area that became available after calcination was greater than the surface area that was lost due to micropore sintering. Whereas, during the 4 hours extraction time, most of the residual solvent was removed during extraction and only a very small amount of surfactant that was left was removed during calcination. After losing the micropores during calcination, the total surface area of the sample decreased. This follows the same theory as that

PAGE 79

in the last section; the longer SCF extraction time allows contact of the surfactant with more SCF mixture solution which enables the dissolving and removal of more surfactant. Figure 4.9 Schematic Diagram of Templated Alumina Aerogel Preparation with Different SCF Extraction Time 66

PAGE 80

4.1.4. Effect of Degassing For BET analysis, nitrogen adsorption desortption experiments were carried out on the NOVA 2000. Prior to each experiment, samples were degassed at 80 o C for 3 hours (Chapter 3). During degassing volatile solvents and gases deposited on sample surface or even inside the pores to a degree are extracted and removed under vacuum and high temperature. It is not a mandatory process for all samples. In this section, the textural properties of one sample were characterized in the BET analyzer with and without degassing. The selected sample was sample one, described and discussed in section 4.1.1. The comparison of the results is shown in Table 4.5 and Figure 4.10. Table 4.5 Textural Properties for Templated Alumina Aerogel with Degassing and Without Degassing Sample SSA (m2/g) TPV(cc/g) APD () As-made with degassing 504.9465 1.10494 87.530 After Calcination with degassing 518.4409 1.15593 89.185 As-made without degassing 301.1644 0.35063 46.570 After Calcination without degassing 253.9237 0.74802 117.834 0100200300400500600 As-made withdegassingAfterCalcination withdegassingAs-madewithoutdegassingAfterCalcinationwithoutdegassing SSA TPV APD Figure 4.10 SSA, APV and APD of One Alumina Aerogel with Degassing and Without Degassing (Before Calcinations and After Calcinations) 67

PAGE 81

Figure 4.10 clearly shows that degassing provided a more porous sample which has a larger SSA and a larger pore volume associated with it for both cases before calcination and after calcination. Without degassing, solvent residues in the as-made aerogel occupied much more pore space and SSA, TPV and APD were all relatively lower than those of degassed samples. After calcination, some by-products such as lightweight hydrocarbons, H 2 O, CO 2 still remained inside the pores and blocked some mesopores (20500) would take a larger percentage in the average pore diameter in undegassed samples so that the undegassed sample had a higher APD These findings suggest that highly porous aerogel catalyst support should be degassed under vacuum before doing further processing such as impregnation. Pore size distribution of the sample after calcination without degassing is shown in Figure 4.11 Figure 4.11 Pore Size Distribution of The Sample After Calcination Without Degassing 68

PAGE 82

69 4.2 Co-precipitated Nickel/Alumina System Regular and templated co-precipitated nickel/alumina aerogel were prepared as described in relevant sections in Chapter 3. The textural properties and nickel distribution were determined and reported in the section below. Two samples with and without the template were prepared in an identical fashion. They were aged for the same period of time and underwent the same solvent replacement. Regular co-precipitated nickel/ alumina sol-gel was only dried by supercritical CO 2 (10mL/min) for 3 hours, whereas, the templated co-precipitate nickel/ alumina sol-gel underwent a supercritical extraction process (4 hours with scCO 2 (10mL/min) and entrainer ethanol (0.5mL/min)) and drying process (2 hours, scCO 2 only (10mL/min)). Nickel acetate tetrahydrate (C 4 H 6 O 4 Ni.4H 2 O) was chosen as the nickel complex for preparing co-precipitated samples and the component molar ratio of Ni:Al was 0.15:1. The analysis of the textual properties was performed by the BET analyzer and topographical profile and element quan titative analysis was achieved by SEM-EDS. The schematic diagram of regular and templated co-precipitated nickel/alumina aerogel preparation is shown in Figure 4.12. The analysis results from both samples before and after calcination are shown in Table 4.6. As shown in Table 4.6, and Figure 4. 13, both of the regular and templated co-precipitated nickel/alumina aerogel samples have higher specific surface area before calcination. A reasonable explanation for the phenomena would be that the nickel acetate hydrolysis reac tion also enhanced the cross-linking degree of the co-precipitated nickel/alumina aerogel system so that the SSA was increased. However, after calcination, nickel hydroxide lost its hydroxyl group and was converted to nickel oxide. This caused drastic loss of SSA because most of the nickel compound was embedded in the bulk of the sample. The average pore size of the templated sample was more stable than the

PAGE 83

regular one both before and after calcination due to the aid of surfactant templates. This result also suggests the successful removal of surfactant template during SCF extraction. Furthermore, the total pore volume also increased by using surfactant templates. Figure 4.12 Schematic Diagram of Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Preparation 70

PAGE 84

71 Sample Ni:Al Weight% Atomic% Ni:Al Initial molar ratio C K O K Ni L Al K C K O K Ni L Al K molar ratio Regular co-precipitate sol-gel 15% NiAc/Al2O3 ASM 0.15:1 22.21 43.77 8.24 25.79 32.55 48.16 2.47 16.83 0.1468 Regular co-precipitate sol-gel 15% NiAc/Al2O3 CAL 0.15:1 5.93 41.49 12.73 39.85 10.33 54.24 4.54 30.89 0.1470 Templated co-precipitate sol-gel 15% NiAc/Al2O3 ASM 0.15:1 19.86 36.48 10.04 33.62 30.91 42.61 3.19 23.29 0.1370 Templated co-precipitate sol-gel 15% NiAc/Al2O3 CAL 0.15:1 4.76 38.86 14.75 41.63 8.58 52.59 5.44 33.4 0.1629 Running time (hours) SSA(m 2 /g) TPV(cc/g) APD() Regular co-precipitate sol-gel 15% NiAc/Al2O3 ASM 3 (CO 2 only, 10ml/min) 700.4091 1.02836 58.729 Regular co-precipitate sol-gel 15% NiAc/Al2O3 CAL 3 (CO 2 only, 10ml/min) 470.8939 0.74692 63.447 Templated co-precipitate sol-gel 15% NiAc/Al2O3 ASM 2+4+2 (CO 2 10ml/min,MeOH 05.ml/min) 622.661 1.27473 81.889 Templated co-precipitate sol-gel Ni15% NiAc/Al2O3 CAL 2+4+2 (CO 2 10ml/min,MeOH 05.ml/min) 443.1932 0.90209 81.417 71 Table 4.6 The Textural Properties and Element Composition of The Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Catalyst System

PAGE 85

0100200300400500600700800 Regular co-preASMRegular co-preCALTemplated co-preASMTemplated co-preCAL SSA TPV APD Figure 4.13 SSA, APV and APD of Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Catalyst System (Before Calcinations and After Calcinations) During calcination, the co-precipitated nickel/alumina aerogel was converted from Ni(OH) 2 /AlOOH to NiO X /-Al 2 O 3 and at the same time some by-products and solvent which contain groups composed of C, O, and H are eliminated and carried out of the samples by air flow. Therefore, the carbon ratio in the sample was reduced to almost 60% after calcination. The reason why carbon was still present in the elemental composition analysis is that samples were held by or mounted on carbon tape, the C peak in the EDS spectrum always appears when carbon tape is used to hold sample no matter what sample is analyzed. One example of SEM-EDS quantitative elemental analysis spectrums is shown in Figure 4.14. Some of the SEM images of these samples are presented in Figure 4.15; SEM-EDS element mapping images give a vivid view of the distribution of each element on the surface and in the bulk of the sample. The nickel particles on the sample surface are present as bright yellow dots due to stronger elemental signal emission, nickel deeper in the sample are seen as relative dark yellow dots due to weaker elemental signal emission. Some of these SEM-EDS mapping images are shown in Figure 4.16. 7272

PAGE 86

More SEM images, SEM-EDS quantitative elemental analysis spectra and SEM-EDS elements mapping images are listed in Appendix B and C. Some samples analyzed through SEM-EDS were not coated by Au/Pt plasma as usual in order to minimize the disturbing effects from other exotic elements. Hence, the SEM images of those samples will be blurred compared to those of the coated samples. Figure 4.14 SEM-EDS Spectrum of Regular Co-Precipitated Nickel/Alumina Aerogel (Ni:Al=0.15:1) 7373

PAGE 87

Mesopore Figure 4.15 SEM Images of Regular and Templated Co-Precipitated Nickel/Alumina Aerogel Before Calcination (Ni:Al=0.15:1) 7474

PAGE 88

(A) (B) Figure 4.16 SEM-EDS Elements Mapping Image of Regular (A) and Templated(B) Co-Precipitated Nickel/Alumina Aerogel Before Calcination 7575

PAGE 89

76 76 the distributions of nickel shown in Figure 4.16 are compared to the element mapping images of supercritical impr egnated nickel oxide/alumina aerogel system in section 4.4. 4.3 Supercritical Fluid Impregnation Compared to conventional liquid impregnation, supercritical impregnation takes advantages of the enhanced per meability, diffusivity and tunable solubility of supercritical solvent to impregnate the solute into porous material matrices in much more effective and manipulated pathway. On the other hand, the co-precipitated nickel alumina catalyst which was prepared in section 3.1 has a uniform nickel distribution in the entire sample matrix, however, quite a number of nickel particles can be embedded into the bulk not on the surface. Catalytic reactions take place on the su rface and near surface regions of the catalyst only. The surface of the catalyst support includes two parts: external surface and interior surface, the latter takes the bigger percentage of the total surface area. Hence, locating nickel complex particle uniformly on the interior surface instead of only on the external su rface as in traditional pathways is the objective of supercritical impregnation. Referring to another study on solubility of nickel complexes, nickel acetate was completely dissolved in the supercritical mixture which contained carbon dioxide and methanol as the entrainer when the temperature was 35 o C or above and the pressure is 1300 psig or above, but not in pure supercritical carbon dioxide at any condition (Semltzer, Brandon,2005). In this section, all alumina cataly st supports were prepared using the same recipe and preparation conditions as the 1 st sample described in section 4.1.1. Two different experimental operating routes were used to perform supercritical impregnation: (1) cont inuous experiment and (2) batch experiment which have been described in Chapter 3. Two kinds of nickel

PAGE 90

complexes nickel acetate and the nickel nitrate (Ni(NO 3 ) 2 ) were utilized as nickel oxide precursors. 4.3.1. Continuous Supercritical Impregnation The continuous experiment is a time domination process. The longer running time the process holds, the more chance the solute has to be introduced into porous matrix. Basically, with continuous SCF solution flow, there are 3 steps that take place in this SCF impregnation which are shown is Figure 4.17. Running time, solubility of nickel complex, process pressure and temperature are the key factors influencing the efficiency of the continuous impregnation process. Figure 4.17 Three Basic Steps of Continuous SCF Impregnation 4.3.1.1. Effect of Nickel complex Nickel acetate and nickel nitrate were both used as nickel oxide precursors in this section for comparison purpose. At the same time, different initial component ratios of the nickel complex were used which will be discussed in next section. SCF impregnation was carried out at 1500 psig and 50 o C where scCO 2 density is 0.426 g/mL and viscosity is 31.017 uPa*S. Continuous scCO 2 and entrainer methanol flow rates were 2.0 mL/min and 0.1 mL/min respectively. After 6 hours continuous SCF impregnation, extra 2 hours SCF drying (scCO 2 only at 10 ml/min) was applied to precipitate solute in the pores by reducing solubility of the nickel complex and to remove the residual entrainer. All 7777

PAGE 91

Based on the elemental analysis results shown in Table 4.7, using different nickel complex precursors did not cause obvious difference in the success of impregnating a nickel complex in the alumina catalyst support. Recently, an associated study conducted by Brandon Smeltzer (2005) shows both nickel nitrate and nickel acetate can be easily dissolved in a SCF mixture which consists of CO 2 plus 5% entrainer at a relative lower pressure (1400 psig) and temperature (35 o C). However, the nickel precursors are not soluble in pure scCO 2 under any condition. A broad impregnation rate of 0.61% to 22.15% was observed in this study. Additional future work needs to be conducted to explain the mechanism behind this phenomenon. An XPS elemental survey was also utilized to characterize sample A through sample D. Although a specific peak indicating the existence of elemental Ni could be detected, the analysis could not give the precise quantitative result for the trace amount of Ni which existed at a mole ratio percentage less than 1%. The XPS elemental survey showed that these samples consisted mainly of Al, C, and O. The Al seems to be in the form of Al 2 O 3 and the carbon in various alcohol, ketone and ether groups. An example of an XPS survey of sample A is shown in Figure 4.18. Additional date are shown in Appendix D. alumina catalyst supports were prepared using the same recipe and preparation conditions as the 1 st sample described in section 4.1.1. The BET analysis and SEM-EDS quantitative element analysis results are shown in Table 4.7. The textural properties of SCF impregnated nickel/alumina samples do not show significant difference when compared to those of the templated alumina catalyst support. After calcination, the SSA and TPV were decreased due to micropores sintering; on the other hand, the APD was almost same both before and after calcination. 7878

PAGE 92

Sample Initial Ni:Al Weight % Atomic % Ni:Al S.W. Elemt mole C K O K Al K Ni K C K O K Al K Ni K mole ratio Impreg. yield A Ni(NO 3 ) 2 /r-Al 2 O 3 After Calcination 2:1 0.35:1 N/A 49.06 47.21 3.73 N/A 62.84 35.86 1.30 0.0363 10.37 B NiAC/r-Al 2 O 3 After Calcination 2:1 0.40:1 N/A 52.04 47.45 0.50 N/A 64.80 35.03 0.17 0.0049 1.23 C Ni(NO 3 ) 2 /r-Al 2 O 3 After Calcination 1:1 0.18:1 N/A 52.48 47.28 0.13 N/A 65.10 34.85 0.04 0.0011 0.61 D NiAC/r-Al 2 O 3 After Calcination 1:1 0.20:1 N/A 48.74 46.76 4.50 N/A 62.73 35.69 1.58 0.0443 22.15 Running conditions SSA(m 2 /g) TPV(cc/g) APD() r-Al 2 O 3 Same as 1st sample in section 4.1.1 434.1407 0.72211 66.532 Ni(NO 3 ) 2 /r-Al 2 O 3 1500psig 50 o C 6+2hrs CO 2 2ml/min MeOH 0.1ml/min Ni:Al=2:1 383.6538 0.69690 72.659 Ni(NO 3 ) 2 /r-Al 2 O 3 After CAL 2hrs 500 o C, 375.6281 0.67589 71.975 NiAc/r-Al 2 O 3 1500psi 50 o C 6+2hrs CO 2 2ml/min MeOH 0.1ml/min Ni:Al=2:1 432.0571 0.78371 72.556 NiAc/r-Al 2 O 3 After CAL 2hrs, 500 o C 410.8509 0.74761 72.786 r-Al 2 O 3 Same as 1st sample in section 4.1.1 405.4360 0.85407 84.261 Ni(NO 3 ) 2 /r-Al 2 O 3 1500psi 50 o C 6+2hr, CO 2 2ml/min MeOH 0.1ml/min Ni:Al=1:1 438.8713 0.91930 83.787 Ni(NO 3 ) 2 /r-Al 2 O 3 After CAL 3hr 500 o C, 389.9513 0.72336 74.200 NiAc/r-Al 2 O 3 1500psi 50 o C 6+2hrs, CO 2 2ml/mi MeOH 0.1ml/min Ni:Al=1:1 434.8802 0.83530 76.831 NiAc/r-Al 2 O 3 After CAL 3hrs 500 o C, 366.2574 0.76618 83.677 79 Table 4.7 The Textural Properties and Element Composition of Continuous SCF Impregnation Nickel Oxide/Alumina Catalyst System 7979

PAGE 93

8080 Intensity (cts) Intensity (cts) Intensity (cts) Intensity (cts) Intensity (cts) Intensity (cts) Bindin g ener gy ( eV ) Bindin g ener gy ( eV ) Bindin g ener gy ( eV ) Bindin g ener gy ( eV ) Fi g ure 4.18 XPS Elemental Anal Bindin g ener gy ( eV ) y sis Result Of Sam p le D Bindin g ener gy ( eV )

PAGE 94

4.3.1.2. Effect of Initial Ni Amount One of most important reasons for using SCF impregnation instead of co-precipitation is to minimize the consumption of catalyst precursor and to maximize the amount of catalyst precursor inside the porous alumina matrix at same time. Excess amount of nickel complex is kept inside the nickel complex cell and is reusable. Whereas, in co-precipitated nickel/ alumina catalyst systems, the nickel catalyst embedded in the bulk of catalyst support has no chance to participate in future catalytic reactions and is useless and un-reusable for this purpose. As the results show in Table 4.7indicate, there is no apparent correlation between the amount of the initial nickel complexes and the textural properties of finalized nickel oxide/alumina catalyst system as well as the successful impregnation yield. 4.3.2. Batch Supercritical Impregnation A batch experiment is diffusion dominated mass transfer process. The diffusivity of solute determines how much solute is able to be introduced into the porous matrix and finally depositd inside the pores. Similar to continuous supercritical impregnation, there are 3 steps in the progress of this SCF impregnation which are shown in figure 4.19. The solubility of the nickel complex in SCF, process pressure, temperature, system homogeneity are the key influence factors of the batch process. During the release of the high pressure, the solubility of nickel complex will rapidly decrease; the nickel complex will then precipitate the in pores. Figure 4.19 Three Basic Steps of Batch SCF Impregnation 8181

PAGE 95

82 82 Diffusivity of solute, in large degree, depends on the solubility of solute in the solvent, in other words, it is determined by the dissolving properties of the solvent. Different temperature and pressure will result in different SCF density and different dissolving strength s. In this section, two different SCF impregnation temperatures and pressures were used. Other operation conditions including the use of the initial nickel precursor, the experiment running time, SCF flow rates, calcinatio n conditions, etc., were kept constant. The process conditions are listed in Table 4.8. Table 4.8 The Constant Process Condition of Batch SCF Impregnation Parameter C ondition Nickel complex Nickel(II) acetate tetrahydrate, C 4 H 6 NiO 4 4H 2 O Initial Nickel/alumina weight ratio Nickel acetate: Alumina=3:1 Impregnation: 3 runs=3(0.5+0.5) hours 30 minutes 30minutes Drying 2 hours Flow rate And running time CO 2 :10ml/min MeOH:0.5ml/min CO 2 : 1ml/min MeOH: 0.05ml/min CO 2 : 10ml/min Drying P and T 1800 psig, 62.5 o C Calcination condition 500 o C 3 hours The BET analysis results and SEM-ED S quantitative element analysis results are shown in Table 4.9. SEM-EDS elemental analysis results indicated that with different SCF impregnation conditions-CO2 =0.42610g/mL and CO2 =0.81073g/mL, the final Ni: Al ratio did not appear to differ greatly. Only 1.65% to 2.7% of initial amount of nickel acetate was impregnated into the alumina porous system. Howe ver, there is one intensity result shown in Table 4.9 as well. More than 27% of the initial nickel acetate was successfully impregnated at 2500 psig and 40 o C. Unfortunately, no reasonable explanation can be issued at this time. Examples of SEM-EDS element distribution mapping images are shown in Figure 4.20.

PAGE 96

Sample Initial Ni:Al Weight % Atomic % Ni:Al S.W. Elemt mole C K O K Ni L Al k C K O K Ni L Al K mole ratio Impreg. yield NiAc/r-Al 2 O 3 After Calcination 3:1 0.60:1 4.96 48.58 1.58 44.88 8.04 59.07 0.52 32.36 0.0161 2.68 1500 psig 50 o C 3:1 0.60:1 N/A 49.56 1.71 48.73 N/A 62.8 0.59 36.61 0.0161 2.68 NiAc/r-Al 2 O 3 Before Calcination 3:1 0.60:1 14.03 46.85 10.52 28.6 21.89 54.88 3.36 19.87 0.1691 28.18 2500 psig 40 o C 3:1 0.60:1 N/A 49.72 13.16 37.12 N/A 66.02 4.76 29.22 0.1629 27.15 NiAc/r-Al 2 O 3 After Calcination 3:1 0.60:1 6.23 45.13 1.01 47.63 10.13 55.07 0.34 34.46 0.0099 1.65 2500 psig 40 o C 3:1 0.60:1 N/A 46.28 1.11 52.61 N/A 59.51 0.39 40.1 0.0097 1.62 NiAc/r-Al 2 O 3 Before Calcination 3:1 0.60:1 13.26 45.97 0.84 39.93 20.18 52.52 0.26 27.04 0.0096 1.60 NiAc/r-Al 2 O 3 After Calcination 3:1 0.60:1 4.76 47.78 1.23 46.23 7.75 58.36 0.41 33.48 0.0122 2.03 Running conditions SSA(m 2 /g) TPV(cc/g) APD() r-Al 2 O 3 Same as 1st sample in section 4.1.1 480.3005 0.93043 77.488 NiAc/r-Al 2 O 3 Before calcination 1500 psi 50 o C CO2 =0.42610g/ml 334.0292 0.73200 87.657 NiAc/r-Al 2 O 3 After Calcination 1500 psi 50 o C CO2 =0.42610g/ml 356.3589 0.69605 78.129 r-Al 2 O 3 Same as 1st sample in section 4.1.1 593.6372 1.17838 79.401 NiAc/r-Al 2 O 3 Before calcination 2500 psi 40 o C CO2 =0.81073g/ml 465.0874 0.90239 77.610 NiAc/r-Al 2 O 3 Before calcination 2500 psi 40 o C CO2 =0.81073g/ml 705.1931 1.25454 71.160 NiAc/r-Al 2 O 3 After Calcination 2500 psi 40 o C CO2 =0.81073g/ml 227.4617 0.69258 121.794 NiAc/r-Al 2 O 3 Before calcination 2500 psi 40 o C CO2 =0.81073g/ml 408.1979 1.13044 110.774 NiAc/r-Al 2 O 3 After Calcination 2500 psi 40 o C CO2 =0.81073g/ml 242.2806 0.93225 153.913 83 Table 4.9 The Textural Properties and Element Composition of Batch SCF Impregnation Nickel Oxide/Alumina Catalyst System 8383

PAGE 97

(a) Batch Impregnation NiO/Al 2 O 3 after Calcination (1500psig&50 o C) (b) Batch Impregnation NiO/Al 2 O 3 After Calcination (2500psig&40 o C) Figure 4.20 SEM-EDS Elements Mapping Image of Two Batch SCF Impregnation Nickel Oxide/Alumina Catalyst System after Calcination 8484

PAGE 98

Uniform nickel distributions were indicated in both (a) and (b) of figure 4.20. These will be further investigated and compared with those mapping images of co-precipitate nickel/alumina system in section 4.5. 4.4 Comparison between Continuous and Batch SCF impregnation Key results of the SEM-EDS elemental analysis and BET analysis of both pathways are listed in Table 4.10 for comparison purpose. SCF impregnation conditions for chosen comparison were: Pressure and temperature: 1500 psig and 50 o C. Nickel complex: nickel acetate Catalyst system status: Finalized after calcination. Table 4.10 Comparison Between Two Continuous Impregnation Sample And One Batch Impregnation Sample Sample Initial Ni:Al (wt) Pathway Impreg. molar ratio (Ni/Al) SSA(m 2 /g) TPV(cc/g) APD() Ni :Al=2 :1 Continuous 0.0049 410.8509 0.74761 72.786 Ni :Al=1 :1 Continuous 0.0443 366.2574 0.76618 83.677 Ni :Al=3 :1 Batch 0.0161 356.3589 0.69605 78.129 Impregnation molar ratio SSA TPV APD Figure 4.21 Comparison Between Two Continuous Impregnation Sample And One Batch Impregnation Sample 8585

PAGE 99

86 86 The comparison indicates that both continuous and batch SCF impregnation have no apparent impact on the textural properties of alumina catalyst support. The specific surface area was reduced sl ightly due to micropore sintering and nickel oxide particle occupation on the interior pore surface. The impregnation molar ratios of both pathways are difficult to be compared to each other because of the unstable and nonrepeatable results. Further work needs to be conducted to obtain stable and repeatable results. So far, in both pathways, nickel can be impregnated evenly in a highly porous alumina matrix. But, batch impregnation is preferred because it is more time-s aving and more material-saving compared to continuous impregnation. 4.5 Comparison between SCF Impregnation and Co-precipitation The results of SEM-EDS elemental an alysis and BET analysis of SCF impregnation and Co-precipitated nickel/alumina are listed in Table 4.11 for comparison in this study. SCF batch impregnation conditions for comparison used: Pressure and temperature: 2500 psig and 40 o C. Nickel complex: nickel acetate Catalyst system status: Finalized after calcination. Table 4.11 Comparison between Co -Precipitated Samples and Batch Impregnation Sample Sample after calcination Molar ratio (Ni/Al) SSA(m 2 /g) TPV(cc/g) APD() Regular co-precipitated 0.1470 470.8939 0.74692 63.447 Templated co-precipitated 0.1629 443.1932 0.90209 81.417 Batch impregnation 0.0161 356.3589 0.69605 78.129

PAGE 100

Figure 4.22 Comparison between Co-Precipitated Samples and One Batch Impregnation Sample Co-precipitated nickel/alumina samples have higher specific surface area than batch impregnation sample. The reason is that the co-precipitated sample has nickel hydroxide cross-linking with alumina during sol-gel preparation. Higher degree of cross-linking causes higher specific surface area of the sample. Whereas, the SSA of batch the impregnation sample is mainly determined by the templated alumina catalyst support which always possesses a little lower SSA than the co-precipitated sample. However, most nickel in the batch impregnation sample was available and accessible for future catalytic reaction since the nickel was only deposited onto sample surface and not embedded into the bulk like co-precipitated process sample. Even with lower SSA and lower Ni/Al ratio, it still has more effective catalytic sites than the co-precipitated sample. Furthermore, batch impregnation sample have more uniform nickel distribution on the samples surface as the comparison between figure 4.16 and figure 4.20 reveals. 8787

PAGE 101

88 CHAPTER FIVE CONCLUSIONS AND RECOMMENDATIONS This study was conducted to utilize novel supercritical fluid aided methods in order to synthesize high ly porous alumina catalyst supports which were then functionalized by impregnating with nick el by another new supercritical fluid aided approach. This chapter provides some concluding remarks and major accomplishments as well as recommendations for further research. 5.1 Conclusions Two major thrusts have been taken and their feasibility have been demonstrated in this study: supercritic al fluid aided synthesis of alumina aerogel, functionalization of the alumina support by supercritical fluid aided impregnating with nickel. The influences of various key factors in each work were evaluated through characterization techniques such as nitrogen adsorption/desorption isotherm, SEM-EDS and XPS. The major conclusions are: (1) High surface area alumina catalyst supports with thermally stable structure were prepared using surfactant templated sol-gel synthesis. (2) The non-ionic surfactant was extrac ted with supercritical ethanol carbon dioxide mixture. The solvent used in the sol-gel process (alcohol) was removed with supercritical carbon dioxide. (3) The composition of sol-gel preparation has profound impact on the textural properties of alumina aerogel. Longer SCF extraction time and higher SCF mixture flow rates induce improved textural properties of alumina aerogels.

PAGE 102

89 (4) Mesoporous alumina aerogel with uniform pore sizes with diameter equal to 7-8 nm were obtained throug h the use of a non-ionic surfactant template, and above 500 m 2 /g Specific Surface Areas (SSA) and over 0.7cc/g Average Pore Volume (APV) were also achieved before and after heat treatment. (5) Impregnation of nickel onto the alumina support was obtained with the aid of supercritical methanol ca rbon dioxide mixture in both batch and continuous pathways. Outstanding uniform nickel distribution on alumina support was achieved thro ugh supercritical impregnation compared to co-precipitated nickel oxide/ alumina system. (6) The impregnation yields (1~3%) were lower than what was expected for the experimental conditions studi ed, warranting extended analysis and studies. 5.2 Recommendations Since the methods used in this research are novel and exploratory, only a study of few major experimental variables was preformed. Future work needs to be conducted in order to optimize these techniques and explain the mechanisms more thoroughly. (1) The pH value of sol-gel will strongly impact the future structural properties of alumina aerogel. The effect of pH values of alcogel before supercritical extraction an d before drying on synthesized material should be investigated further. (2) The SCF extraction and drying times and flow rates should be calculated based on fundamental phenomena such as the solubility of surfactant in the carrier gas.

PAGE 103

90 (3) Retrofitting the experimental setup and further instrumentation of the set-up will enable more controlled environment that will facilitate more reproducible results. (4) The mechanism of nickel complex deposition into porous alumina matrix should be studied and coupled with solubility of nickel complex in supercritical fluid mixture, sorption, and diffusion. (5) The catalytic activity of material synthesized through supercritical impregnation of nickel oxide /alumina catalyst system should be investigated and compared with that of relevant co-precipitated materials. (6) Various characterization techniques such as XRD, FTIR should be included in future evaluations.

PAGE 104

91 REFERENCES Al-Ghamdi, A.M., Synthesis of Surfactant Templated Alumina Support with Tunable Nanoporosity, Ph.D. Candidates Dissertation, University of South Florida, 2003. Anderson, Mark T.; Martin, James E.; Odinek, Judy G.; Newcomer, Paula P. and Wilcoxon, Jess P., Monolithic Periodic Mesoporous Silica Gels, Microporous Materials, 10, 1997, 13-24. Anderson, Mark T.; Sawyer, Patricia S. and Rieker, Thomas, Surfactanat-Templated Silica Aeroge ls, Microporous and Mesoporous Materials, 20, 1998, 53-65. Barrett, E.P.; Joyner, L.G.; Halenda, P.P., J. Amer. Chem. Soc., 73,1951, 373. Bradley, D. C., Chemical Reviews, Metal alkoxides as precursors for electronic and ceramic materials, vol. 89, 1989, 1317-1322. Brinker, C. Jeffery; Smith, Douglas M.; Desphande, Ravindra, Preparation of low-density aerogels at ambient pressure, Materials Research Society, 271, 1992, 567-572. Brinker, C. J.; Ward, K. J.; Keefer, K. D.; Holupka, E.; Bray, P.J. and Pearson, Aerogels, J. Fricke (Springer-Verl ag, New York,, 1986), pp 57-67. Brunauer, S.; Emmett, P.; Teller, E., J. Amer. Chem. Soc., 60, 1938, 309. Caruso, R. A. and Schattka, J., Cellulose acetate templates for porous inorganic network fabrication, Advanced Materials, vol. 12, No. 24, 2000, 1921-1923. Chordia, L; Martinez, J.L., Whats So Hot About Supercritical Fluids? Laboratory Focus, Vol. 6, No. 1, 2002,

PAGE 105

92 Gerbeleu, N.V.; Arion, V.B. and Burgess, J., Template Synthesis of Macrocyclic Compounds, Wiley-VCH, 1999. Gonzalez-Pena, V.; Diaz, I.; Marquez-Alvarez, C.; Sastre, E.; Perez-Pariente, J., Thermally Stable Mesoporous Alumina Synthesized with Non-Ionic Surfactants in The Presence Of Amines, Mieroporous and Mesoporous Materials vol. 44-45, 2001, 203-210. Janosovits, Ute.; Ziegler, Gunter; Scharf, Uta and Wokaun, Alexander, Structural characterization of interm ediate species during synthesis of Al 2 O 3 -aerogels, Journal of Non-Crystalline Solids, vol. 210, 1997, 1-13. Kazarian, S.G., Applications Of FTIR Spec troscopy to Supercrit ical Fluid Drying, Extraction and Impregnation, Applied Spectroscopy Reviews, 32(4), 1997, 301-348. Keene, Matthew T. J.; Gougeon, Regis D. M.; Denoyel, Renaud; Harris, Robin K.; Rouquerol, Jean and Llewellyn, Philip L., Calcination of MCM-41 mesophase: mechanism of surfactant thermal degradation and evolution of the porosity, Journal of materials chemistry, vol. 9, 1999, 2843-2850. Keysar, Shani; Shter, Gennady; Hazan, Yoram; Cohen, Yachin and Grader, Gideon, Heat Treatment of Alumina Aerogels, Chemistry of Materials, vol. 9, 1997, 2464-2467. Kistler, S. S., Coherent Expanded Aerogels, Journal of Physical Chemistry, vol. 36, 1932, 52-64. Livage, J.; Henry, M. and Sanchez, C., Sol-gel chemistry of transition metal oxides, Progress in Solid State Chemistry, vol. 18, 1988, 259-341. Osaki, Toshihiko; Horiuchi, Tatsuro; Sugiyama, Toyohiko; Suzki, Kenzi; Mori, Toshiaki, NiO-Al 2 O 3 aerogel from (CH 2 O) 2 Ni and AlOOH sol, Journal of Non-Crystalline Solids, vol. 225, 1998, 111-114. Perrut, Michel, Advances in Supercrit ical Fluid Chromatographic Process, Journal of Chromatography A, 658, 1994, 293-313. Perrut, Michel, Supercritical Fluid Applications: Industrial Developments and Economic Issues, Ind. Eng. Chem. Res., 2000, 39, 4531-4535.

PAGE 106

93 Pinnavaia, Thomas and Liu, Yu , Assembly of hydrothermally stable aluminosilicate foams and large pore hexagonal mesostructures from zeolite seeds, Chemical materials communications, vol. 14, 2002, 3-5. Pinnavaia, Thomas and Tanev, P. T., A neutral templating route to mesoporous molecular sieves, Science, 267, 1995, 865-567. Pinnavaia, Thomas and Zhang, Wenzhong, Rare earth stabilization of mesoporous alumina molecular sieves assembled through an N 0 I 0 pathway, Chemical Communications, 1998, 1185-1186. Pinnavaia, Thomas; Zhang, Zhaorong; Hick s, Randall W. and Pauly, Thomas R., Mesostructured forms of Al 2 O 3 , Journal of American Chemical Society, 124, No. 8, 2002. Rest, A.J., Infra Dig Matrix Media, J. Mol. Struct., 1990, 222, 87-93. Scherer, George W. and Brinker, C. Jeffer, Sol-Gel Science The Physical and Chemistry of Sol-Gel Processing, Academic press, New York, 1990. Smeltzer, Brandon, Solubility of Surfactant Templates and Inorganic Complexes in Supercritical Carbon Dioxide Solutions Masters Thesis, University of South Florida, 2005 Suh, Dong Jin and Park, Tae-Jin, sol-gel strategies for pore control of high-surface-area transition-metal oxide aerogels, Chemistry of Materials, vol. 8, 1996, 509-513. Suh, Dong Jin; KLM, Jim-Hong; Park, Tae-Jin and KLM, Kyung-Lim, Improved stability of nickel-alumina aerogel catalysts for carbon dioxide reforming of methane, Natural Gas Conversion V, vol. 119, 1998, 771-776. Suh, Dong Jin; Park, Tae-Jin; Ki m, Jin-Hong and Kim, Kyung-Lim, Nickel-Alumina Aerogel Catalysts Prepared by Fast Sol-Gel Synthesis, Journal of Non-Crystalline Solids, vol. 225, 1998, 168-172. Suh, Dong Jin; Park, Tae-Jin; Lee, Seo-Ho; Kim, Kyung-Lim, Nickel-alumina composite aerogel as liquid-phase hydrogenation catalysts, Journal of Non-Crystalline Solids, vol. 285, 2001, 309-316.

PAGE 107

94 Sunol, A. K., Sunol, S. G., "Templating Aerogels for Tunable Nanoporosity" in first edition of the Dekker Encyclopedia of Nanoscience and Nanotechnology, edited by James A. Schwarz, Cristian Contescu, and Karol Putyera, in press 2004. Taylor, L. T., Supercritical Fluids Ex traction, John Wiley and Sons, New York, 1996. Teja, A.S.,; Eckert, C.A., Commentary on Supercritical Fluids: Research and Applications, Ind. Eng. Chem. Res., 2000, 39, 4442-4444. Tewari, Param H.; Hunt, Arlon J. and Lofftus, Kevin D., Ambient-Temperature Supercritical Drying of Transparent Silica Aerogels, Materials letters, 3, No.9-10, 1985, 363-367. Wijngaarden, R. J.; Kronberg A. and Westerterp K. R., Industrial catalysis Optimizing Catalysts and Processes, Wiely-VCH, 1998. Yao, Nan; Xiong, Guoxing; Yuhong; He, Mingyuan and Yang, Weishen, Preparation of novel uniform mesoporous alumina catalysts by the sol-gel method, Catalysis Today, vol. 68, 2001, 97-109. Ying, J.V.; Mehnert, C.P.; Wong, M.S. Synthesis and Applications of Supramolecular-templated Mesoporous Materials Angew. Chem., Intl. Ed., 1999, 38, 56-77.

PAGE 108

95 APPENDICES

PAGE 109

Appendix A: The BET Analysis Results 96

PAGE 110

Appendix A (Continued) 97

PAGE 111

Appendix A (Continued) 98

PAGE 112

Appendix A (Continued) 99

PAGE 113

Appendix A (Continued) 100

PAGE 114

Appendix A (Continued) 101

PAGE 115

Appendix A (Continued) 102

PAGE 116

Appendix A (Continued) 103

PAGE 117

Appendix A (Continued) 104

PAGE 118

Appendix A (Continued) 105

PAGE 119

Appendix A (Continued) 106

PAGE 120

Appendix A (Continued) 107

PAGE 121

Appendix B: The SEM Images Figure B.1 The SEM Images of Continuous Impregnation Sample A (Nickel acetate : Alumina=2:1) Figure B.2 The SEM Images of Continuous Impregnation Sample B (Nickel nitrate : Alumina=2:1) Figure B.3 The SEM Images of Continuous Impregnation Sample C (Nickel acetate : Alumina=1:1) 108

PAGE 122

Appendix B (Continued) Figure B.4 The SEM Element Mapping Images of Continuous Impregnation Sample A (Nickel acetate : Alumina=2:1) Figure B.5 The SEM Element Mapping Images of Regular Co-precipitated Nickel /Alumina before Calcination (Nickel acetate : Alumina=0.15:1) 109

PAGE 123

Appendix B (Continued) Figure B.6 The SEM Element Mapping Images of Templated Co-precipitated Nickel /Alumina before Calcination (Nickel acetate : Alumina=0.15:1) Figure B.7 The SEM Element Mapping Images of Templated Co-precipitated Nickel /Alumina after Calcination (Nickel acetate : Alumina=0.15:1) 110

PAGE 124

Appendix B (Continued) Figure B.8 The SEM Element Mapping Images of Batch Impregnation Nickel /Alumina before Calcination (NiAC:Alumina=3:1 at 2500 psig & 40 o C) Figure B.9 The SEM Element Mapping Images of Batch Impregnation Nickel /Alumina After Calcination (Nickel acetate: Alumina=3:1 at 1500 psig & 60 o C) 111

PAGE 125

Appendix C: The EDS Spectra Figure C.1 The SEM-EDS Spectrum of Regular Co-precipitated Nickel /Alumina after Calcination (Nickel acetate : Alumina=0.15:1) 112

PAGE 126

Appendix C (Continued) Figure C.2 The SEM-EDS Spectrum of Templated Co-precipitated Nickel /Alumina after Calcination (Nickel acetate : Alumina=0.15:1) 113

PAGE 127

Appendix C (Continued) Figure C.3 The SEM-EDS Spectrum of Batch Impregnation Nickel /Alumina before Calcination (Nickel acetate : Alumina=3:1 at 2500psig & 40 o C) 114

PAGE 128

Appendix C (Continued) Figure C.4 The SEM-EDS Spectrum of Batch Impregnation Nickel /Alumina after Calcination (Nickel acetate : Alumina=3:1 at 2500psig & 40 o C) 115

PAGE 129

Appendix D: The XPS Elemental Analysis Results Figure D.1 XPS Elemental Analysis Result of Sample A 116

PAGE 130

Appendix D (Continued) Figure D.2 XPS Elemental Analysis Result of Sample B 117

PAGE 131

Appendix D (Continued) Figure D.3 XPS Elemental Analysis Result of Sample C 118