The Dehydrogenation and Isomerization of Cyclohexane Over a Platinum Alumina Mordenite Catalyst

Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1970 The ehD ydrogenation and Isomerization of Cyclohexane Over a Platinum Alumina Mordenite Catalyst. David Erskine Allan Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Allan, David Erskine, "The eD hydrogenation and Isomerization of Cyclohexane Over a Platinum Alumina Mordenite Catalyst." (1970). LSU Historical Dissertations and Theses. 1768. https://digitalcommons.lsu.edu/gradschool_disstheses/1768

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. r

71 - 31+00

ALLAN, David Erskine, 1939- THE DEHYDROGENATION AND ISOMERIZATION OF CYCLOHEXANE OVER A PLATINUM ALUMINA MORDENITE CATALYST.

The Louisiana State University and Agricultural and Mechanical College, Ph.D., 1970 Engineering, chemical

University Microfilms, Inc., Ann Arbor, Michigan

aaAiaoan sv m o v x a aawiuoHDiw Naaa s v h NoiiviaassiG sihj, Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. THE DEHYDROGENATION AND ISOMERIZATION OF CYCLOHEXANE OVER A PLATINUM ALUMINA MORDENITE CATALYST

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

The Department of Chemical Engineering

by David Erskine Allan B.S., University of Tulsa, 1962 M.S., University of Tulsa, 1964 June, 1970

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dedicated to my parents who taught me the importance of education and to my wife for her sacrifices in its undertaking.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGEMENTS

The author wishes first to particularly acknowledge Professor

Alexis Voorhies, Jr. who provided the opportunity to do this research

and served as faculty advisor. His advice, guidance and unbounded

optimism are greatly appreciated.

Special acknowledgement is expressed to the Esso Research and

Engineering Company for sponsoring this project. Thanks are also

given to the staff of the Esso Research Laboratories, Baton Rouge,

Louisiana, who provided the equipment, catalysts and certain analytical

work. In particular, the author wishes to give his thanks to Dr.

H. E. Merrill of the Esso Research Laboratories who prepared the

catalysts.

Appreciation is expressed to the Petroleum Processing Laboratory

of the Chemical Engineering Department for the use of their equipment

and facilities.

The author wishes to offer his sincere thanks to Dr. Phillip A.

Bryant for many helpful and stimulating discussions on heterogeneous

catalysis. His enthusiasm and willingness to help at all times are

greatly appreciated. Thanks are also given to Dr. J. R. Hopper, a

co-worker on this project who was above all, a friend.

Special appreciation and thanks are expressed to Mrs. Claudia

Ainsworth who typed this dissertation. Her cheerfulness, diligence

and skill made a difficult task much less so.

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Finally, the author wishes to acknowledge the unnamed faculty,

graduate students and staff at Louisiana State University who gave

help and advice when it was needed.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

PAGE

LIST OF TABLES ...... xi

LIST OF FIGURES ...... xiv

ABSTRACT ...... xvi

CHAPTER

I. INTRODUCTION ...... 1

List of References - Chapter I ...... 5

II. LITERATURE SURVEY ...... 6

A. Introduction ...... 6

B. Non-Zeolite Catalysts ...... 10

1. Cycloparaffin Dehydrogenation Catalysts .... 10

2. Non-Zeolite Cycloparaffin Isomerization

Catalysts ...... 11

3. Non-Zeolite Bifunctional Catalysts...... 14

C. Crystalline Zeolites ...... 21

1. Introduction ...... 21

2. Physical Properties of Zeolites ...... 23

a. General ...... 23

b. Mordenite ...... 28

3. Sorption, Diffusion and Acidity in Zeolites . 30

4. Catalytic Properties of Zeolites ...... 34

a. General ...... 34

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PAGE

b. Mordenite ...... 36

D. Catalytic Mechanisms ...... 39

1. Dehydrogenation of Naphthenes ...... 39

2. Isomerization of Naphthenes ...... 41

3. Combined Dehydrogenation and Isomerization

of Naphthenes ...... 43

List of References - Chapter II ...... 46

III. EXPERIMENTAL EQUIPMENT AND PROCEDURE ...... 58

A. General ...... 58

B. Experimental Equipment ...... 58

1. Reaction System ...... 58

a. Liquid Feed S y s t e m ...... 60

b. Gas Feed System ...... 60

c. Reactor ...... 61

d. Temperature and Pressure Control ...... 63

e. Product Recovery System ...... 64

2. Analytical System ...... 64

C. Materials ...... 64

1. Gases ...... 64

2. Liquids ...... 65

3. Catalysts ...... 65

D. Experimental Procedure ...... 65

1. Catalyst Activation ...... 65

2. Hydrocarbon Reaction Material Balance ...... 66

3. Material Balance Calculations ...... 68

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PAGE

IV. CATALYST PREPARATION AND ANALYSES ...... 69

A. Introduction ...... 69

6 . Catalyst Preparation ...... 70

C. Analytical Measurements ...... 71

List of References - Chapter IV ...... 76

V. A KINETIC MODEL FOR THE SIMULTANEOUS DEHYDROGENATION

AND ISOMERIZATION OF CYCLOHEXANE ...... 77

A. Introduction ...... 77

B. Cyclohexane - Benzene and Cyclohexane -

Methylcyclopentane Equilibrium ...... 78

C. Reactor Contacting ...... 80

D. Derivation of Kinetic Equations ...... 82

1. Rate Limiting Processes in Heterogeneous

Catalysis ...... 83

2. Derivation of the Rate Equations ...... 85

3. Simultaneous Solution of the Differential

Rate Equations ...... 91

E. Solution of the Kinetic Equations for the

Reaction Rate Constants ...... 97

1. Solution of Non-Linear Equations ...... 97

2. Pattern Search ...... 99

3. Solution of Kinetic Equations by Pattern

Search ...... 105

List of References - Chapter V ...... 110

VI. EXPERIMENTAL DATA AND ITS ANALYSIS ...... 113

vii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PAGE

A. Introduction ...... 113

B. Pore Diffusion and Bulk Mass Transfer Limitations 114

1. Diffusional Effects in Experimental Catalysis 114

2. Search for Operating Conditions - Pore

Diffusion and Mass Transfer Limitations .... 119

C. Test for Mass Transfer Limitations at 775°F and

85 psia - Pt-ALgOg -Mordenite Catalyst ...... 129

D. Test for Pore Diffusion Limitations at 775°F

and 85 psia - Pt-AlgO^ -Mordenite and Pt-Al^Og

Catalysts ...... 133

E. Reproducibility of the Experimental Data on the

Pt-Alg03 -Mordenite Catalyst at 775°F and 85 psia 141

F. Test of the Reaction Model for the Dehydro

genation and Isomerization of Cyclohexane -

Pt-AlgOg-Mordenite Catalyst at 775?F and 85 psia 143

G. The Effect of Temperature on the Cyclohexane

Dehydrogenation and Isomerization Rate

Constants - Pt-Als03 -Mordenite Catalyst ...... 150

H. The Effect of Pressure oh the Cyclohexane

Dehydrogenation and Isomerization Rate Constants

at 775PF - Pt-ALjOs-Mordenite Catalyst ...... 158

1. The Effect of a Variation in Total Pressure

at a Constant Hydrogen to Cyclohexane Feed

Diluent Ratio ...... 158

2. The Effect of Independently Varying Hydro

carbon and Hydrogen Partial Pressure ...... 161

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PAGE

I. Correlation of the Cyclohexane Dehydrogenation

/ and Isomerization Rate Constants at 775?F with

the Observed Pressure Effects - Pt-Alg0 3 -

Mordenite Catalyst ...... -172

1. Introduction ...... 172

2. Single and Dual Site Surface Reaction

Adsorption Models ...... 177

3. Simplification of the Surface Reaction

Adsorption Models ...... 180

4. Correlation of Experimental Results with

the Simplified Surface Reaction Models ..... 182

J. Effectiveness of the Overall Model for the

Pt-ALjOa-Mordenite Catalyst ...... 185

K. Reactivity of n-Heptane and Methylcyclopentane

over the Pt-ALjO^ -Mordenite Catalyst at 925^F

and 305 psia ...... 187

List of References - Chapter VI ...... 192

VII. CONCLUSIONS AND RECOMMENDATIONS ...... 196

A. Conclusions ...... 196

B. Recommendations ...... 199

APPENDICES

A. NOMENCLATURE ...... 202

B. DETAILED EXPERIMENTAL DATA ...... 211

C. SAMPLE CALCULATIONS ...... 281

D. ANALYTICAL SYSTEM ...... 289 »

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. PAGE

E. COMPUTER PROGRAM LISTING ...... 295

VITA ...... 337

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

TABLE PAGE

1 Dehydrogenation Activity Values of Some Supported Metal and Oxide Catalysts...... 17

2 Some Zeolites and Their Properties...... 26

3 Possible Applications of Zeolite Catalysts...... 37

4 Description of Catalyst Analytical Tests...... 73

5 Catalyst Analytical Measurements...... 74

6 Convergence of the Search Routine for Different Initial Estimates of the Rate Constants, Run 35A...... 109

7 Exploratory Data for Cyclohexane over the Pt-Al3 03 - Mordenite Catalyst at 90CPF and 225 psia...... 121

8 Test for Mass Transfer Limitations at Low Cyclo hexane Conversion Levels over the Pt-Al^Og- Mordenite Catalyst at 90(PF and 225 psia...... 124

9 Effect of Catalyst Particle Size on Cyclohexane Dehydrogenation and Isomerization Rate Constants for the Pt-Al^Og -Mordenite Catalyst...... 127

10 Equilibrium Composition of the Cyclohexane- Benzene-Methylcyclopentane System as Determined from Free Energy Data...... 130

11 Test for Mass Transfer Limitations at 775°F and 85 psia - Cyclohexane Dehydrogenation and Isomer ization over the Pt-ALjOg-Mordenite Catalyst 131

12 Test for Pore Diffusion Limitations at 775°F and 85 psia - Cyclohexane Dehydrogenation and Isomer ization over the Pt-A^Og-Mordenite Catalyst 135

13 Test for Pore Diffusion Limitations at 775°F and 85 psia - Cyclohexane Dehydrogenation and Isomer ization over the Pt-ALjOg Catalyst...... 138

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE PAGE

14 Reproducibility of the Experimental Data on the Pt-AlaQ,-Mordenite Catalyst at 775°F and 85 psia...... 142

15 Statistical Parameters Calculated from the Reproducibility Study on the Pt-AL,Oa -Mordenite Catalyst...... 144

16 Test of the Reaction Model for the Dehydrogenation and Isomerization of Cyclohexane - Variation of Space Time over the Pt-Alg03 -Mordenite Catalyst at 775°F and 85 psia...... 147

17 The Effect of Temperature on the Cyclohexane Dehydrogenation and Isomerization Rate Constants - Pt-ALgOg-Mordenite Catalyst at 85 psia...... 153 1 18 Reported Activation Energies for the Dehydrogena tion of Cyclohexane to Benzene...... 156

19 The Effect of Total Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F - Pt-ALgOg -Mordenite Catalyst...... 159

20 The Effect ot Total Hydrocarbon Partial Pressure on the Cyclohexane Dehydrogenation and Isomeri zation Rate Constants at 775°F and ~80 psia Hydrogen Partial Pressure - Pt-AlgOg-Mordenite Catalyst...... 163

21 The Effect of Hydrogen Partial Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F and ~5.6 psia Hydro carbon Partial Pressure - Pt-ALgOg-Mordenite Catalyst...... 166

22 The Effect of Hydrogen Partial Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F and ~2.7 psia Hydro carbon Partial Pressure - Pt-ALg03 -Mordenite Catalyst...... 169

23 Summary of Rate and Adsorption Constants for the Dehydrogenation and Isomerization of Cyclohexane over the Pt-ALgOg-Mordenite Catalyst at 775°F.... 184

24 Effectiveness of the Overall Model for the Pt- AlgOg -Mordenite Catalyst...... 186

xii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE PAGE

25 Reactivity of n-Heptane over the Pt-Al2 03 - Mordenite Catalyst at 925°F and 305 psia...... 188

26 Reactivity of Methylcyclopentane with and without Nitrogen Addition over the Pt-AlgOa-Mordenite Catalyst at 925°F and 305 psia...... 190

D-l Calibration Factors for the Gas Chromatograph.... 294

xiii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

FIGURE PAGE

1 Basic Structure of Zeolite Crystals...... 24

2 Cage Structure for Faujasite...... 27

3 Cross-Sectional View of Mordenite...... 29

4 Reforming C6 Hydrocarbons with a Bifunctional Catalyst...... -...... 44

5 Simplified Flow Diagram of the Reaction System... 59

6 Cross Sectional View of Reactor in Fluidized Heating Bath...... 62

7 Response Surface for Run 35A...... 100

8 Cross Section of the Response Surface for Run 35A 101

9 Illustration of Pattern Search for an Arbitrary Response Surface...... 103

10 Schematic Flow Diagram for Computer Program SLAVEZ...... 106

11 The Binodal Pore Size Distribution...... 117

12 Effect of Cyclohexane Conversion Level on the Degree of Bulk Mass Transfer Limitation for the Exploratory Data on the Pt-Al^Og-Mordenite Catalyst at 900PF and 225 psia...... 123

13 Test for Mass Transfer Limitations at Low Cyclo hexane Conversion Levels over the Pt-AlgOg- Mordenite Catalyst at 900PF and 225 psia...... 125

14 Test for Mass Transfer Limitations at 775°F and 85 psia - Cyclohexane Dehydrogenation and Isomer ization over the Pt-AlgOg-Mordenite Catalyst 132

15 Test for Pore Diffusion Limitations at 77fPF and 85 psia - Cyclohexane Dehydrogenation and Isomer ization over the Pt-Alg03 -Mordenite Catalyst..... 136

xiv

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. FIGURE PAGE

16 Test for Pore Diffusion Limitations at 775°F and 85 psia - Cyclohexane Dehydrogenation and Isomer ization over the Pt-ALjOg Catalyst...... 139

17 Test of the Reaction Model for the Dehydrogena tion and Isomerization of Cyclohexane - Variation of Space Time over the Pt-Al2 03 -Mordenite Catalyst at 775°F and 85 psia...... 151

18 The Effect of Temperature on the Cyclohexane Dehydrogenation Rate Constant - Pt-Alg03 - Mordenite Catalyst at 85 psia...... 154

19 The Effect of Temperature on the Cyclohexane Isomerization Rate Constant - Pt-Alg03 -Mordenite Catalyst at 85 psia...... 157

20 The Effect of Total Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F - Pt-AlgO^-Mordenite Catalyst...... 160

21 The Effect of Total Hydrocarbon Partial Pressure on the Cyclohexane Dehydrogenation and Isomeri zation Rate Constants at 775°F and ~80 psia Hydrogen Partial Pressure - Pt-Al3 03 -Mordenite Catalyst...... 164

22 The Effect of Hydrogen Partial Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F and ~5.6 psia Hydrocarbon Partial Pressure - Pt-ALjOg-Mordenite Catalyst... 167

23 The Effect of Hydrogen Partial Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F and ~2.7 psia Hydrocarbon Partial Pressure - Pt-Alg03 -Mordenite Catalyst... 170

24 The Effect of Hydrocarbon Partial Pressure on The Cyclohexane Dehydrogenation and Isomerization Rate Constants at 775°F - Pt-Al2 03 -Mordenite Catalyst...... 173

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

A halogen-free Pt-AljOg -mordenite catalyst has been investigated

for the simultaneous dehydrogenation and isomerization of cyclohexane

in an integral fixed-bed reactor system. These results on cyclohexane

are supplemented by a limited study on a halogen-free Pt-AlgOg catalyst.

A mathematical model for the simultaneous dehydrogenation and isomeri

zation of cyclohexane has been developed assuming that the surface

reactions were rate controlling and first order with respect to the

hydrocarbon components. Integration of the differential rate equations

from the model yielded two nonlinear equations which were solved

simultaneously for the rate constants by an optimization technique.

A study of macropore diffusion on both catalysts revealed that

significant diffusional effects existed in the dehydrogenation and

isomerization of cyclohexane. The rates of cyclohexane dehydrogenation

and isomerization at 775?F on the Pt-AlgOg -mordenite catalyst were

shown to be independent of particle size below a certain critical size.

Data on the Pt-AlgOg catalyst at 775°F showed that the cyclohexane

dehydrogenation rate was dependent on catalyst particle size throughout

the entire range investigated. The rate of cyclohexane isomerization

over the Pt-ALj03 catalyst was very low at 775° F and not dependent on

particle size.

The rates of cyclohexane dehydrogenation and isomerization over

the Pt-AlgOg-mordenite catalyst at 775°F were shown, by a variation

in contact time, to be consistent with the assumed reaction model.

xvi

with permission of the copyright owner. Further reproduction prohibited without permission. The activation energy for cyclohexane dehydrogenation over the Pt-

AlgOa-mordenite was evaluated in the 724°-775°F temperature range and

found to be 14.4 kcal/gm mole. This value agrees with values reported

in the literature for similar catalysts. The activation energy for

cyclohexane isomerization was found to be 12.5 kcal/gm mole which is

low in comparison to reported values for mordenite catalysts. This

suggests a diffusional limitation in the micropores of the mordenite

component.

The effects of pressure on the rates of cyclohexane dehydrogena

tion and isomerization at 77^*F over the Pt-A^C^ -mordenite catalyst

were investigated by varying total pressure and reactant partial

pressures. Both rates were found to decrease with increasing hydro

carbon partial pressure and to be independent of hydrogen partial

pressure. The observed pressure effects on the Pt-Al^Og -mordenite

catalyst were adequately correlated by using the conventional Langmuir

adsorption approach and assuming the dynamic adsorption coefficients

of the hydrocarbons were equal. An exact knowledge of the reaction

mechanism is not implied because of certain limitations of the

Langmuir adsorption theory when applied to mixtures and reaction

environments.

The effectiveness of the model for cyclohexane dehydrogenation

and isomerization over the Pt-ALj03 -mordenite catalyst was examined

by comparing predicted and experimentally determined rate constants.

The average % error for the dehydrogenation rate constant was 8.4%

while that for the isomerization rate, constant was 10.5%. These

values correspond closely to the measured precision of the data and

xvii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the model is considered to effectively represent the observed

experimental behavior.

The cyclohexane isomerization activity of the Pt-ALjOg-mordenite

catalyst was found to be approximately eighteen times that of the

Pt-ALjQj catalyst at 775°F. The difference in isomerization activity

can be qualitatively explained by the presence of the mordenite which

was found to be twelve times more acidic than the alumina by ammonia

chemisorption measurements. These data confirm that catalyst acidity

is directly related to isomerization activity.

The dehydrocyclization and dehydroisomerization capabilities

of the Pt-AlgOg-mordenite were also investigated by limited experi

mentation on n-heptane and raethylcyclopentane.

xviii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER I

INTRODUCTION

The study of chemical reaction phenomena has assumed a role of

ever increasing importance in the physical sciences. This importance

stems not only from the insight gained into the fundamental nature of

molecules and bonds but from a commercial standpoint as well. Few raw

materials can be made into useful products by purely physical transfor

mations. The heart of most industrial processes is a chemical reaction.

The use of a catalyst to influence a reaction is a key area in the

study of chemical reactions, both academically and industrially. Pro

ducts from catalytic processes account for approximately 20% of the (3) gross national product of the United States and it has been postu

lated that a nation's usage of catalytic technology provides a good

index to the state of its industrial’ development.^ Nowhere does

industry depend more heavily on catalysis than in the petroleum and

petrochemical industry. The development of the modern catalytic pro

cesses for cracking and reforming are crowning achievements in catalytic

technology.

The position of catalytic reforming as an industrial process of

importance is evident from the present day U. S. capacity of about 2.5 4 million barrels per day. ( ) Initial impetus for catalytic reforming

on a large scale was provided by the need for synthetic toluene from

petroleum to supply the demands for TNT in World War II. Though reformer

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. construction lagged immediately following the war, the octane require

ments of post-war automobiles and the development of new economical

catalysts caused growth to spiral to the present day figure. Catalytic

reforming has recently been stimulated by the need of high pressure

hydrocracking for reformer by-product hydrogen. If governmental air

pollution requirements require the withdrawal of tetraethyl lead from

gasoline, future growth of catalytic reforming will undoubtedly be

stimulated.

The primary objective in catalytic reforming of gasoline for octane

number improvement is to convert naphthenes and paraffins to aromatics.

The usual feed is an olefin-free naphtha in the C6 - Cg boiling range

consisting of cyclohexane homologues, cyclopentane homologues, normal

paraffins, isoparaffins and single ring aromatics. Generally, the

ratio of cyclohexanes to cyclopentanes is not far from unity and the

ratio of isoparaffins to normal paraffins is less than the equilibrium

ratio at reforming conditions. There are at least five different

reactions that occur in catalytic.reforming:

a. Simple dehydrogenation of cyclohexane homologues to

aromatics.

b. Isomerization of cyclopentane homologues to cyclohexanes

with subsequent dehydrogenation to aromatics.

c. Dehydrocyclization of paraffins to aromatics.

d. Isomerization of paraffins.

e. Hydrocracking, especially of paraffins.

The principal reforming catalyst for toluene production during

World War II was molybdenum oxide supported on a high-surface-area

alumina. This type of catalyst is still being used to a limited extent

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3

in making high octane gasoline. However, in about 95% of present (2) catalytic reforming capacity in the U. S., the catalyst employed is

a post-war development. In this post-war reforming catalyst, a small

amount of platinum metal is supported on a high-surface-area alumina

which may be modified (e.g., by halogen activation) to provide additional

acidity. Catalysts of this type are much more active than the MoOg-AljOa

catalysts, and somewhat more selective; but both general types are

remarkably similar in their promotion of the five typical reactions

discussed in the preceding paragraph. Both types of catalysts are what

are known as bifunctional catalysts which possess a strong hydrogena-

tion-dehydrogenation activity and an adequate isomerization activity.

For example, Pt and Mo02 are known as active species for hydrogenation-

dehydrogenation and in both cases the alumina base contributes to the

isomerization activity by reason of its acidity. However, too much

acidity will result in an undesirable increase in hydrocracking and a

superior reforming catalyst requires a careful balancing of the two

functions.

During the past decade, crystalline aluminosilicates or zeolites

have received considerable attention as catalytic materials. Synthetic

zeolites have high crystalline regularity with an acidic functionality

and early applications were directed toward their selective adsorption

properties. Intensive research is presently being conducted toward

application of zeolites to many important industrial catalytic processes

and commercialization has already been achieved for hydrocracking and

hydroisomerization.

The use of a zeolite as the acidic component of a bifunctional

reforming catalyst presents interesting possibilities but has received

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. little attention academically or industrially. This fact, coupled

with the importance of dehydrogenation and isomerization in catalytic

reforming has provided the motivation for this research.

This dissertation is a study of the dehydrogenation and isomeriza

tion properties of a bifunctional platinum-alumina-zeolite catalyst.

The zeolite component is synthetic mordenite which has a high isomeri

zation activity for aliphatic compounds. For contrast, a platinum

alumina catalyst with no acidic functionality is also studied. The

greater part of this research is directed toward the simultaneous dehy

drogenation and isomerization of cyclohexane supplemented by limited

studies on n-heptane and methylcyclopentane.- The primary objectives of

this research are to (1) investigate the use of the synthetic zeolite

mordenite as the isomerization promoting part of a bifunctional reform

ing catalyst and (2) develop a mathematical model for the simultaneous

dehydrogenation and isomerization of cyclohexane from the experimental

data on this catalyst over a wide range of conditions.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5

LIST OF REFERENCES - CHAPTER I

1. Anonymous, "Why the Soviet Union Often Failed to Capitalize on its Excellent Studies of Catalysis," Chemical Engineering, 76 (No. 20), 66 (1969).

2. Ciapetta, F. G., "Special Report - Catalytic Reforming," Petro/Chem Engineer, 33 (No. 5), C-19 (1961).

3. Haensel, V., "Catalysis in the Petroleum Industry," presented at the 31st Midyear Meeting of the American Petroleum Institute's Division of Refining, Houston, Texas, May 9, 1966.

4. Stormont, D. H., "Modernization-Refinings Record of 1968," Oil and Gas Journal, 67 (No. 12), 108 (1969).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER II

LITERATURE SURVEY

A. Introduction

A catalyst is a substance which, without appearing in the final

product, influences the velocity of a chemical reaction. The first

scientific observation of a catalytic transformation is due to Kirchhoff,

who in 1811 showed that mineral acids catalyze the starch-sugar reaction /oo\ without themselves being altered. During the period 1811-1900, many /oo\ catalytic reactions were investigated but it was not until shortly

after the turn of the century that catalysis was accorded world recog

nition. In 1909 and 1912 Ostwald and Sabatier respectively were awarded

Nobel Prizes in Chemistry for their work in reaction rates, chemical

equilibria and catalysis.

The petroleum and petrochemical industry are unmatched in applying

the achievements of research and development in catalysis to practical

industrial processes. The "catalytic era" in the petroleum industry

emerged with the catalytic cracking process which was developed largely

through the observations of Houdry from 1924 to 1928.^®’^®^ The first / loft} catalytic cracking unit went on stream in 1936 ' and since that time

the application of catalysis in the petroleum industry has grown at a

fantastic rate. Products from catalysis currently account for 20% of

the gross national product of the U. S. The petroleum and petrochemical

industry is almost completely catalysis oriented. In 1965 the catalytic

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7

processes of cracking, reforming, alkylation and hydrogen treating were (43) alone responsible for 9.6 million barrels per day capacity. The

primary reason for this orientation in the petroleum industry is that

catalytic methods are much more satisfactory for control of selectivity

than are straight thermal reactions.

One of the most interesting and important heterogeneous catalytic

processes is the reforming of virgin and cracked naphthas to produce pure

aromatic hydrocarbons and high octane gasoline. It is the most important

refinery technique for raising octane number and is second in capacity

only to catalytic cracking. As in catalytic cracking, catalytic

reforming was preceded by thermal methods which have virtually been

discarded in favor of the catalytic process for reasons of product

quality.

The first catalytic reforming unit went on stream in 1940 and (28) employed a process using a molybdena-alumina catalyst. This process,

which was used until 1950, was initially applied to virgin naphtha for (8$ production of aromatics to be used in explosives and aviation gasoline.

After World War II reformer construction lagged and by 1950 ’.nstalled

capacity was less than 80,000 barrels per day, some of which was not used

after the war due to high operating cost and lack of demand for high

octane fuels.

During the late 1940's catalysis investigations by petroleum

scientists led to the discovery of a new class of hydrocarbon reforming

catalysts which were more active and selective than the molybdena-alumina

catalysts. These new dual-function catalysts were obtained by the

deliberate incorporation of a hydrogenation-dehydrogenation component

such as platinum on a surface containing acid sites such as halogen

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. /or 28) activated alumina. ’ ' These catalysts are called bifunctional since

they simultaneously promote dehydrogenation and isomerization of saturated

hydrocarbons. The first process to use the new bifunctional catalyst

was the Platforming process of Universal Oil Products which was announced

in 1949. During the next five years nine new catalytic reforming pro

cesses were announced indicating that parallel investigations were being (27) carried out in many petroleum laboratories. ' The growth of catalytic

reforming in the past twenty years has been phenomenal. The present

U. S. capacity is about 2.5 million barrels per d a y ^ ^ of which 95%

is based on the platinum-containing catalysts, and the remainder on the (27) molybdena-alumina catalysts.

The usual catalytic reforming feed is an olefin-free naphtha in the

C6 -Cg boiling range consisting of cyclohexane homologues, cyclopentane

homologues, paraffins and single ring aromatics. Although any straight

run naphtha is a potential reformer feed, the product quality over

platinum catalysts will be governed by the feed composition. It has

been shown that the relative amounts of the hydrocarbons in a reformer

feed is a function of the crude source.The aromatics concentration

is usually less than 20% but the naphthene concentration may vary from (27) 12 to 70% depending on the crude source. There are at least five

different reactions that occur in catalytic reforming.(®®>64>28,113)

a. Dehydrogenation of cyclohexane homologues to aromatics.

0 — 0 + 3 ^

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9

b. Dehydroisomerization of cyclopentane homologues to

aromatics.

+ 3 H j

c. Dehydrocyclization of paraffins to aromatics.

nGjrfye «»- Y

d. Isomerization of paraffins.

nc7H16 ;==£ C%CH(CH 3)3C%

e. Hydrocracking of paraffins.

Other possible reactions that may occur are isomerization of substituted

aromatics, hydrodesulfurization of thiophenes and hydroisomerization of

olefins to isoparaffins. The extent to which each of these reactions

takes place depends on the catalyst, composition of the feed, and the

operating conditions.

The most important reactions in catalytic reforming from an octane

upgrading standpoint are dehydrogenation of cyclohexanes, dehydroisom

erization of cyclopentanes and dehydrocyclization of paraffins. This

importance lies in the fact that naphthenes and paraffins have lower

octane numbers than the corresponding aromatics to which they can be (281 converted. The naphthene dehydrogenation and isomerization reactions

form the basis for this investigation. The pertinent literature with

respect to these reactions is covered in detail in a latter part of this

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. chapter. The dehydrocyclization reaction has been the subject of many . ■ . (115,62,74,63) . _ . (28) . „ ,(46) pure compound studies, and Ciapetta et al and Hansch

give complete reviews of the literature. One current dehydrocyclization

study of interest is reported by L y s t e r ^ ^ who studied the dehydro

cyclization of n-heptane over twenty-four reforming type catalystsi

B. Non-Zeolite Catalysts

Non-zeolite catalysts have been used extensively for cycloparaffin

dehydrogenation and isomerization. Cycloparaffin dehydrogenation cata

lysts employed are usually a transition group metal or transition group

metal oxide. Non-zeolite cycloparaffin isomerization catalysts employed

are usually acidic halides or acidic chalcides. Bifunctional catalysts

for the simultaneous dehydrogenation and isomerization of cycloparaffins

contain a hydrogenation-dehydrogenation component and an acidic component.

1. Cycloparaffin Dehydrogenation Catalysts

Dehydrogenation-hydrogenation reactions are those in which a

hydrogen-hydrogen bond is broken or formed and in which hydrogen is a

reactant or product. Active catalysts tc\ reactions involving hydrogen

can be classified into three groups: (a) transition and bordering

metallic elements, (b) transition oxides or sulfides, and (c) other

oxides.

Pure metals and oxide supported pure metals have been found

to be active cycloparaffin dehydrogenation catalysts. The most common

pure metals reported to be active for dehydrogenation of cyclohexane (122 134) and other monocyclic naphthenes are platinum and palladium, *

although cobalt,rhenium,^ and nickel^have also been listed.

Examples of oxide-supported metal catalysts used for the dehydrogenation

of cyclohexane to benzene include Pt-AlgOs , (*06,23) Pt-TiOa , ^ ^

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Re-A]g03 , ^ ^ and Pd-AlgOg.^3^ Innes^33^ indicates that metal dehydro

genation catalysts must be capable of forming chemical bonds (chemi-

sorption) with the reactants, but bonding must not be so strong as to

inhibit the easy release of the reaction products. The strength of

bonding with hydrogen increases with the number of vacant d orbitals of

the metal. Catalytic activity, however, commonly reaches a maximum with

about one vacant d orbital per atom so that the elements Co, Ni, Rh, xr,

and Pt which have this characteristic are generally the most active

metals for reactions involving hydrogen.

Elements such as V, Cr, Mo, Ta, and W have a large number of

vacant d orbitals and are relatively inactive for reactions involving

hydrogen, which is probably due to their strong adsorption for reactants

and products.During dehydrogenation-hydrogenation conditions, the

lower oxides and sulfides of -.hese metals undergo partial or complete

reduction to the metal at the surface. After reduction of the oxide or

sulfide, Innes^33^ postulates that the electrons of the residual oxygen

or sulfur lessen the strong adsorptive properties of the metals by fill

ing some of the vacant orbitals, so that the adsorption and catalytic

properties would be similar to elements with fewer empty orbitals. This

theory may help to explain why the compounds Cr2 03 ,^ ^ * 23^ V2 03 , ^ ^

- Cr2 03 -AL,03 , MoOj-ALjOj, ,(124) M o S a / 8^ and NiW04 -ALjOg (23^ have

been reported active for the dehydrogenation of cyclohexane whereas the

pure metals are usually inactive for reactions involving hydrogen.

2. Non-Zeolite Cycloparaffin Isomerization Catalysts

Isomerization of hydrocarbons can be defined as a rearrange

ment of the molecular structure of the hydrocarbon without a change in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12

molecular weight. Non-zeolite catalysts for the isomerization of cyclo-

paraffins are usually acidic halides or acidic chalcides (sulfides or

oxides).(32)

Acidic halides were the first catalysts commercialized for the

isomerization of saturated hydrocarbons and came into prominence during

the nineteen forties. Although the acidic halides as ordinarily prepared

are effective catalysts for the isomerization of saturated hydrocarbons,

it has been well established that these catalysts are ineffective in the (32) absence of a promoter or initiator. This promoter or initiator of

catalytic activity in the acidic haldies is a source of protons such as

a hydrogen halide or water^33,3^ ,^2,^ >*''^ and is usually present in

comparatively small amounts.For example, Stevenson and B e e c k ^ ^

found that moist aluminum chloride is catalytically active for the isom

erization of cyclohexane but has no effect when it is anhydrous. Alumi

num chloride and aluminum bromide are the most active and widely used (32) of the acidic halides for isomerization of saturated hydrocarbons

and have been used extensively for the isomerization of cyclohexane and

methylcyclopentane.(117,68,92,94) other acidic halides reported active

for the isomerization of paraffins are boron trifluoride plus hydrogen (54) (32) fluoride and zinc chloride. Although acidic halides have the

advantage of being active for isomerization at relatively low temperatures

(~ 250°F), they have the disadvantages of promoting undesirable side

reactions and being corrosive. They also have the disadvantage of being

poisoned by relatively small amounts of water and sulfur.

The elements oxygen, sulfur, selenium, and tellurium of Group

VIA of the Periodic Table have been called "chalcogens," a term analogous

to "halogens" used for the elements of Group VIIA. The compounds of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13

these elements of Group VIA are called acidic chalcides. Although the

acidic chalcides include a great variety of solid oxides and sulfides, (32) the most widely used are silica, alumina and their mixtures.

Chalcide catalysts such as silica-alumina or alumina are not particularly

effective for isomerization of saturated hydrocarbons unless they possess

electronic (oxidative-reductive, hydrogenactive) as well as acidic

(electron pair accepting) properties. Hydrogenation-dehydrogenation

activity may be imparted to an acidic chalcide by the addition of a (29,77) transition metal or transition metal oxide such as platinum, ’ , . , (29,98) , (29) . . . (29,31,41) „ ,, (29) nickel, cobalt, molybdena, or tungstic oxide.

Acidic chalcides of the silica-alumina variety reported effective for (98 26 29) the isomerization of naphthenes and paraffins are Ni-SiOs -AlgOg, ’ *

Pt-SiOg-ALjOg Mo03 -Si02 -AL3O3 Th03 -Si03 -ALjOg ,*21^ and (21) ZrOg-SiOg-ALgOa . Other synthetic acidic chalcide catalysts used for

the isomerization of cyclohexane or cyclohexene to methylcyclopentane or

methylcyclopentene are B e O , ^ ALjOg-F^03 plus H C l , ^ ^ AlgOg-V3 03 ,

MoSg,^2 ’*^2^ and MoS3 -CoS. Although the acidic chalcides are non-

corrosive and selective for isomerization of hydrocarbons, their low

activity requires the use of high temperature (~ 70CPF) which is undesir

able from the standpoint of the thermodynamic equilibrium of the paraffin

isomers.

One recent development in non-zeolite isomerization catalysts

for saturated hydrocarbons is the combination of an acidic halide such (139) as aluminum chloride with an acidic chalcide such as platinum alumina.

This combination acidic halide-acidic chalcide type catalyst operates in

a moderate temperature range (200-40GPF) and apparently does not have

corrosion problems.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14

3. Non-Zeolite Bifunctional Catalysts

The development of bifunctional catalysts for catalytic reform

ing of petroleum naphthas represents one of the most outstanding appli

cations of heterogeneous catalysis over' the past two decades. These

bifunctional catalysts usually consist of a dehydrogenation-hydrogenation

component such as platinum or palladium on an acidic support such as

silica-alumina or halogen promoted alumina.Such catalysts are

termed bifunctional since they have been found to simultaneously promote

dehydrogenation and isomerization of saturated hydrocarbons.

The earliest catalysts employed in catalytic reforming were

transition metal oxides such as molybdena or chromia supported on

alumina.Molybdena-alumina is still employed commercially in cata

lytic reforming, but only to a limited extent as compared to platinum-

alumina. It is a much less active catalyst. Also, due to side-reactions

leading to carbonaceous deposits, the molybdena-alumina catalyst must be /Og\ frequently regenerated. Early pure compound studies of the funda

mental reforming reactions over Mo02 -Al^Os and CrgOa-AlgO^ catalysts are (42) reported by Grosse et al who studied the dehydrocyclization of n- (41) heptane, and Greensfelder et al who studied the aromatization of

n-heptane, raethylcyclohexane, and methylcyclopentane. Although the

molybdena-alumina reforming catalyst is truly bifunctional since it

promotes both dehydrogenation and isomerization, the distinction between

the two types of sites is not clear since both types of catalytic centers (28) are present in this transition metal oxide. Since the development

of the bifunctional metal-acidic oxide reforming catalysts, the transition

metal oxide-alumina catalysts have little commercial importance and were (27) used in only 5% of U. S. reforming capacity in 1961. In spite of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15

the lack of industrial stimulus MoOg-Al^t^ and CrgOa-A^C^ catalysts are

still being studied^^’^ ’^? as catalysts for typical pure component

reforming reactions to learn more about the mechanisms in catalytic

reforming and the chemistry of these catalysts.

During the mid 1940's, petroleum catalytic scientists began

to investigate the catalytic properties of active metal dehydrogenation-

hydrogenation catalysts as possible components of reforming catalysts.

These investigations led to the discovery of a new class of reforming

catalysts which were more active and selective than the transition metal

oxide catalysts. These catalysts consisted of a dehydrogenation component

such as platinum or palladium supported on an acidic isomerization pro

moting component such as silica-alumina or halogen promoted alumina. The

first commercial process to successfully employ the platinum-acidic oxide (141 142) type catalyst ’ was the Platforming process of Universal Oil (27) Products which was announced in 1949. The Platformir.g process was

an immediate success since its announcement coincided with the early

stages of the present octane race, and by 1951 twenty-five units were (28) operating, under construction, or under contract. During this time,

similar investigations were being carried out in other petroleum labora

tories and by 1956 twelve new catalytic reforming processes.had been (27) announced. All of these modern catalytic reforming processes employ

ing a platinum type catalyst are characterized by moderate temperatures

(850-950PF) , moderate hydrogen partial pressures (10-30 atm.) and low /or} regeneration frequency. As of 1961, 95% of catalytic reforming

capacity in the United States was based on the platinum-acidic oxide (27) type reforming catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16

As already noted, the most common bifunctional reforming

catalysts today consist of an active metal dispersed on an acidic support.

A highly selective reforming catalyst must contain a very active dehydro

genation component. Several experimental procedures have been used to

determine the dehydrogenation activity of various metals and metal (27 130 30) (27) oxides. * ’ Ciapetta measured the rate of converting cyclo

hexane to benzene in a differential flow reactor under reforming con

ditions on several supported metal and oxide catalysts and this data is

shown in Table 1. These results show clearly why platinum has emerged

as the most desirable metal for imparting dehydrogenation activity to

reforming catalysts. Most platinum-acidic oxide catalysts have platinum

concentrations ranging from 0.1 to 1.0 wt. It has been s h o w n ^ ^

.that platinum concentrations higher than 0.5 to 0.7 wt. % do not increase

the activity of the catalyst. The acidic support for the platinum may

be either halogen promoted alumina or silica-alumina^^ although most (27) commercial platinum reforming catalysts use halogen promoted alumina.

The characterization of platinum on reforming-type catalysts

has received much attention. Platinum is usually incorporated on alumina (81) supports by chloroplatinic acid. Mills et alv ' have described the

distinct chemical steps in the preparation and use of the catalyst.

These steps are:

a. Impregnation,

HgPtClg (soln) -• HgPtClg (ionic adsorption)

b. Drying and calcining,

c. Reduction to metal from original salt,

HjPtClg + 2Hg - Pt + 6HC1

d. Catalytic use,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dehydrogenation Activity Values of Some Supported Table 1. ^etal an

Dehydrogenation Activity Index* (micromoles Catalyst (wt. %) benzene/gm catalyst-sec)

34% Cr2 03 cogelled with AlgOg 0.5

10% M o 0 3 coprecipitated with Alg03 3

57» Ni on ALg03 or Si02 -AlgOg 13

5% Co on Alg03 13

0.5% Ir on Alg03 190

1% Pd on ALg03 200

5% Ni on Si02 320

1% Rh on Alg03 890

0.5% Pt on Alg03 or Si02 -Alg03 1400-4000

*fc Determined with cyclohexane in a differential flow reactor, 80CPF, 100 psig, and 6 moles Hg/mole cyclohexane after 30 min. pretreated in flowing Hj at the same conditions.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18

e. Regeneration by burning off coke,

Pt + 02 - Pt03

f. Reduction from oxide.

Pt02 + 2E, -* Pt + E,0

These authors studied steps c., e. and f. and found that the

platinum was highly dispersed and that loss of platinum effectiveness was

paralleled by growth in platinum crystallite size. The high degree of

dispersion of platinum on alumina was also noted by Spenadel and

B o u d a r t ^ ^ who studied platinum dispersion by hydrogen chemisorption. I This study indicated that the crystal size must be less than 10 X

and that the platinum may be dispersed as two-dimensional patches but not (71) as isolated atoms. Maat and Moscou studied the influence of platinum

crystallite size on the n-heptane dehydrocyclization reaction and it was O found that at increased platinum crystallite sizes (> 100 A) the dehydro

cyclization reaction rate decreased. The effect of platinum dispersion (33) on cyclohexane dehydrogenation has been investigated by Cusumano et al. (74) McHenry et al suggested that platinum on alumina exists as a complex

which is soluble in hydrofluoric acid or acetylacetone.

The alumina or alumina-silica supports used in bifunctional

reforming catalysts have been shown to be acidic in nature by their

chemical affinity for basic nitrogen compounds such as ammonia, n-butyla- (18 87 108) mine, pyridine and quinoline. ’ ’ As discussed previously, the

function of the acidity is to give isomerization activity to the support.

It has been suggested that acidic nature of silica-alumina arises from

the replacement of silicon by aluminum atoms at the surface, yielding

sites that may act as Lewis or Br^nsted acids.Although alumina

itself is slightly acidic, it is much less acidic than the silica-alumina

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19

combination.The most common way to incorporate acidity and hence

isomerization activity to alumina is by treatment with halogens such as u, • n (141,142,145,151,144) . _ . .. , chlorine or fluorine. The halogen may be added

during preparation of the catalyst or by the addition of organic halogens (27) to the reformer feed. The concentration of halogen on platinum-

alumina reforming catalysts are commonly in the range, 0 .1 -1.0 wt.

Although the simple term alumina has been used in this discussion,

several phases of alumina e x i s t ^ * ^ which have been shown to have quite

different catalytic properties.^"*’^

One of the most important considerations in bifunctional plati

num-acidic oxide catalysts is the correct balance of the dual-functions.

It has been shown that the platinum sites and acidic sites can act (52 132} (52} independently. ' Hinden et alv ' found that mechanical mixtures of

Pt-Si02 and SiOg-AlgOa were active for the dehydroisomerization of methyl-

cyclopentane to benzene. Since this stepwise reaction requires both

isomerization and dehydrogenation, it appears that the olefin intermediates

diffuse through the gas phase from one type of catalytic site to the other. (132) Similarly, Weisz and Swegler found that mechanical mixtures, in which

the acidic and dehydrogenation functions were present on different parti

cles were active for the isomerization of n-paraffins. The independent

action of the dehydrogenation and acidic functions in platinum-acidic

oxide catalysts makes it possible, to a certain extent, to exert control

over these functions and obtain the correct balance in the catalysts. It

has been shown previously that increasing the platinum content beyond

0.5 wt. % provides no increase in dehydrogenation activity. From a cost

viewpoint, it is desirable to keep the platinum concentration as low as

possible. The main problem, however, is correct adjustment of the acidity

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of the support. Too much acidity will cause excessive hydrocracking

while too little acidity will not sufficiently promote the desired (27) (27) isomerization reactions. Ciapetta ' has shown that although

Pt-Si02 -ALgOg has a high activity for dehydrogenation and isomerization

of saturated hydrocarbons at 75CPF, operation at 90(fF gives excessive

hydrocracking due to high acidity of the Si02 -Alg03 . Acidity of Pt-Si02 - (27) A1203 catalysts can be controlled by reduction in surface area, (27) addition of controlled amounts of alumina to silica gel or by addition

of basic metal oxides.A similar problem in acidity control exists

with Pt-AlgOjj catalysts promoted by halogens. A study of the reactivity

of a n-heptane-cyclohexane mixture at 800PF over a Pt-Alg03 catalyst with (27) varying amounts of chloride was reported by Ciapetta. 7 The results

indicate that while catalysts with less than 0.06 wt. % chloride have

virtually no heptane isomerization activity, an increase to 0.49% gives

a tremendous increase in isomerization activity. Increasing the chloride

content to 1.2 wt. % gives increased isomerization activity but the

selectivity to heptane isomers is decreased due to excessive hydrocrack

ing. All of these results indicate that control of acidity in bifunc

tional reforming catalysts is essential.

Pure compound studies over bifunctional platinum-acidic oxide

catalysts at reforming conditions have been numerous. Heinemann et al^ ^

studied the reactions of cyclohexane, methylcyclohexane, methylcyclopentane /Og\ and n-heptane over the Houdry platinum on alumina catalyst at 95CPF

and 300-600 psig. A companion paperproposes a reaction mechanism

for the C6 hydrocarbons. Hettinger et al^ ^ report extensive data on

the reactions of n-heptane, alkylcyclohexanes, and other pure compounds

over the Sinclair-Baker RD-150 platinum on alumina catalyst^ ' over

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21

/ temperature and pressure ranges of commercial interest. Data are also

presented on the effect of particle size, platinum concentration and (44) catalyst poisons. Haensel and Donaldson show the effectiveness of

the Platforming catalyst^®^ (Pt-AlgOg-Cl) to promote typical reforming

reactions using n-heptane, n-pentane, cumene, and methylcyclohexane. The

reactivities of thirty individual Cg -C8 naphthenes over a platinum-alumina-

halogen catalyst were investigated by Keulemans and V o g e , ^ ^ and a

detailed reaction mechanism is proposed which accounts for the influence

of structure on reactivity. Data on the reactions of cyclohexane and

methylcy.clopentane over Pt-Alg03 -Cl and Pt-ALjOg catalysts at very high (45) space velocities are reported by Haensel et al. Cyclohexane was

observed as a primary product and a reaction mechanism explaining the

experimental data is presented. Detailed literature surveys on pure

component studies over bifunctional reforming catalysts are given by

S i n f e l t ^ ^ and Ciapetta.

Bifunctional platinum-acidic oxides catalysts are very sensi

tive to certain poisons. These poisons may be of a permenent nature such

as arsenic or lead or of a temporary variety such as nitrogen or sul- (27) fur. The effects of the temporary poisons on cycloparaffin reactions

over platinum alumina catalyst are reported by Meisel et al , ^ ^ Minachev

and Isagulyants^®^ and Haensel.Hettinger et al^ ^ report the

effect of the permanent poison arsenic for typical reforming reactions in

addition to a study of the temporary poisons.

C. Crystalline Zeolites

1. Introduction

During the past decade, the unique physical and chemical prop

erties of crystalline aluminosilicates, called zeolites, have attracted

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22

widespread attention in scientific circles and in the chemical process

industry. The basic building block in any zeolite crystal is a tetra

hedron of four oxygen atoms surrounding a silicon or aluminum atom. The

different geometrical patterns in which these Si04 and A104 tetrahedra

are arranged form the basis for the different physical and chemical

properties of the various classes of zeolites. Smith has recently defined

a zeolite as "an aluminosilicate with a framework structure enclosing

cavities occupied by large ions and water molecules, both of which have

considerable freedom of movement, permitting ion exchange and reversible

dehydration."*22^' There are over thirty naturally occurring zeolites and

many synthetic variants have been made which have quite different prop

erties from their natural counterparts.*^

Zeolite minerals were discovered and named in 1756 by Baron

Cronstedt, the Swedish mineralogist.*22) Early research (1840-1925) on

zeolites was concentrated on purely physical phenomena such as ion (22) exchange, reversible dehydration and adsorption properties. The term

"molecular sieve" was proposed for crystalline zeolites in 1932 by

McBain*^) in a discussion of the significance of the early adsorption

studies. The period 1930-1945 produced significant studies on the struc

ture of natural zeolites***^ and their sorptive properties.* ) During

the late nineteen forties, research was directed toward synthesis of

zeolites; and by 1952, the Linde Company had successfully prepared several

synthetic varieties.*22) The early synthetic zeolites were applied

primarily to the selective separation of gases.

The past decade has seen the emergence of zeolites as important

catalytic materials. The regular, ordered arrangement of the zeolite

crystals permits catalysis to be treated as an exact science. Although

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23

zeolites are still important in adsorption, ion exchange and water soften

ing, their future use in catalysis is expected to overshadow all other

commercial applications.^^

2. Physical Properties of Zeolites

a. General

There are over thirty naturally occurring zeolites; and in

addition, many have been synthesized in the laboratory.^ All zeolites’,

natural or synthetic, are composed by Si04 and A104 tetrahedra as shown

in Figure 1, having the basic form o f : ^ ^ ^

0 -1 r 1 -1 r I 1 0 0 < - J % O 0 1 1 1 1 l+ 1 1 M- 0 0 and 1 1 0 0 L I • I -

The oxygen atoms are mutually shared between tetrahedral units, contrib

uting one of the two valence charges of each oxygen to each tetrahedron.

The Si04 tetrahedra containing the tetravalent silicon are electrically

neutral while the A104 tetrahedra containing the trivalent aluminum are

negatively charged. Electrical neutrality is typically maintained by the

presence of an alkali metal or alkaline earth ion such as Na+ , K+ , CaP+ ,

BaP+ , or Sr3 + . >m>90) g£nce each A104 tetrahedron requires only a

monovalent cation for electrical neutrality, a divalent cation may be

-shared between two A104 tetrahedra. Figure 1 shows the spatial arrange

ment of simple zeolite crystal with examples of monovalent and divalent

cations.

A general empirical formula of a unit cell for all zeolites, (22 121) natural or synthetic, is ’

Mex/N £(A102)X (Si02)v J • Zl^O

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24

Electrically neutral 0 Si04 tetrahedron f

.Si

A104 tetrahedron uses a Al monovalent sodium atom 0 to balance charge

Ca Al

Two A104 tetrahedra share a divalent calcium atom which balances charge

(79) Figure 1. Basic Structure of Zeolite Crystals

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25

The notation Me represents a metal cation of valence N while X, Y and

Z give the number of moles of alumina, silica and water respectively per

unit cell. Table 2 gives the compositions and some structural properties

for several zeolite groups.

The Si04 and A104 tetrahedra in zeolites may be joined

together in an almost infinite variety of geometrical patterns. This

variation in geometry, plus the metal cations included, determine the

physical and chemical characteristics of any particular zeolite. A

recent structural classification of zeolites by x-ray diffraction divides

them into seven groups: analcite, natrolite, chabazite, harmoi.one, (121) heulandite, mordenite, and faujasite. This structural classification

can be broken down into three general classes: (1) chainlike or fibrous

crystals, (2) layer structures and (3) semi-rigid three dimensional (79) structures. Mordenite is an example of a chainlike structure and

will be discussed in detail in the next section. Zeolites in the chaba

zite group have frameworks best represented in terms of sheets or layers,

and are an example of the second general structural class. An

example of semi-rigid three dimensional zeolites is the sodalite group (7 9 ) which contains faujasite and Linde's A, X, and Y zeolites. The

sodalite group of zeolites are all based on frameworks which are simple (22) arrangements of truncated octahedra. This arrangement of truncated

octahedra forms periodic inner cavities or cages in the crystal structure,

and an example of the cage structure for faujasite is shown in Figure 2.

Each vertex in the cage structure represents a silicon or aluminum atom

and the edges represent oxygen atoms linking the tetrahedra. The inner

cavity has a diameter of about 11 %. and access is by four twelve-sided 9 (79) windows having diameters of about 9 A each.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ro CTv

6.6 2.6 8 8 4.0 8 4.0 2.8 4.2 ~ 3.0x4.3 3.6x5.2 Size, Size, A Aperture * 4c

0.22 0.15 0.22 0.29 0.21 0.30 0.099 0 2.0 cnP /gm cnP Void Volume, Symmetry Hexagonal OrthorhombicMonoclinic 0.14, Orthorhombic Rhombohedral CubicCubic 0.35 0.35 Orthorhombic Cubic 0.36 Cubic Cubic Cubic 1011,0 ] ] 2711,0 • ) )i32 ] 260HgO * )i3B ] 264HgO * 22 2 )12 2 2 ( )i B ] * ] 26 IIjO * B )i (Si02 )26] • (Si02 )26] 2811,0 • 2 (Si0 (Si02 )32] (Si02 )32] 16IfeO • 1 0 (Si0 ) (Si02 )4O] • )4O] (Si02 2411,0 • (Si03>6 3 • (Si03>6 3 2C1 • 2 1 6 5 6 8 )ag(Si02)1O6 3 264HjO * ) ) 6 ) ) 2 )12 2 2 ) Q (S10 Q ) 2 2 2 102 [(A10 4 L(A102 )g (Si02 )2.jp j * j 27HgO * )2.jp (Si02 L(A102 )g L(A [(A10 L(A10 .6 1(A10 Ca

.2 2 8 6 4 1 6 a4 Naa[(A10 C C^ ^a Kg L(4102)b (SiOg)!! ] • ] • (SiOg)!! L(4102)b Kg Na Source Composition Idaho Synthetic ] 2411,0 • )4O (Si02 t(A102)e Nag Italy Synthetic Synthetic Naj Germany C^oCCAtaajeo(Si0 Many

Name Table 2. Some Zeolites Theirand Properties. Stilbite Many Na Erionite Oregon Philipsite Sodalite Many Naell(A10 ZeoliteA Mordenite Mordenite Chabazite Nova Scotia Faujasite ZeoliteX Zeolite Y Synthetic C(A10 Analcime Based on amount the water of contained per gram of dehydrated zeolite. Unknown Structures Sodalite Group Phillipsite Group Mordenite Group Chabazite Group Analcime Group

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27

Figure 2. Cage Structure for Faujasite

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28

In general, zeolites are characterized by a narrow,

uniform continuous channel system that becomes available after the water

of hydration is driven off by heating and evacuation.It is this

property that gives the zeolites their "sieving property" since materials

held in channels may be driven off without collapse of the crystal struc

ture. The empty channels are then in a condition to receive other mole

cules and the rigidity and fixed dimensions of the crystal structure limit

the size of the molecules that can be accomodated.Excellent reviewsLews (22) of the structural characteristics of zeolites have been given by Breck,

B a r r e r , ^ ’^ Venuto et al, ^ ^ ^ and Meier.

b. Mordenite

A unit cell of the zeolite mordenite has the empirical .... . (22,127,121) chemical formula

Na0 [(A102 )8 (Si03)4O] • 24HgO

Mordenite has been found in such diverse parts of the world as Iceland,

Nova Scotia,^39^ Soviet Union,andUni Idaho.It can also be (10,135,34,138) synthesized.

Structurally, mordenite is characterized by a two dimen

sional, tubular pore system which is parallel to the fiber axis of the (127) crystal. ' A cross-sectional view of mordenite is shown in Figure 3.

The silicon and aluminum atoms are represented by the points of inter

section while the linking oxygen atoms are represented by the lines. The

crystal structure consists of chains of silica and alumina tetrahedra

linked laterally so that a system of large elliptical parallel channels (127) interconnected by smaller cross channels is created. As reported

by Meier,in his structural investigations, mordenite is orthorhombic

with unit cell dimensions of: a = 18.13 X, b = 20.49 X, and c = 7.52 X.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. b=20.49A

(37) Figure 3. Cross-Sectional View of Mordenite.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30

A unit cell of mordenite contains two of the large parallel elliptical

tubes and extends a distance of 7.52 A perpendicular to the plane of the

.page in Figure 3. Ideally, the dimensions of the large elliptical

channels are 6.95 A by 5.81 A while the small elliptical cross channels

have major and minor diameters of 4.72 A and 3.87 The effec

tive channel diameters may be reduced by stacking faults which are (127 75) periodic displacements of the lattice network. *

The first successful synthesis of mordenite was achieved

by B a r r e r ^ ^ who crystallized mordenite from aqueous suspensions of

sodium aluminosilicate gels. The optimal temperature for crystallization

was 265-29fPc at a pH of 8-10. The resulting mordenite product was found

to readily undergo cation exchange and was a good gas adsorbent. Recent (34) synthesislesis iof mordenite have been reported by Domine and Quobex and (135) Sand.

Mordenite has a high Si/Al ratio and a good thermal and

acid stability. Sodium mordenite can be readily cation exchanged

or converted to hydrogen mordenite by exchange with ammonium ions with

subsequent calcining, or by dilute acid treatment. Mordenite is one of

the few zeolites that will undergo complete hydrogen ion exchange in

acidic media without destruction of the crystal structure.

3. Sorption, Diffusion and Acidity in Zeolites

The capacity of porous solids to sort mixtures of molecules

according to their size and shape has been exhibited in a variety of (7) sorbents such as coals, active carbon, and silica gels. Nowhere,,

however, is molecular sieve separation so specific and quantitative as in

the naturally occurring zeolites or their synthetic variants. Once

zeolitic water has been removed by heating and degassing, crystalline

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31

zeolites are able to sorb guest molecules in great variety and yet in a

highly selective manner. Zeolitic crystals are permeated by continuous

channel systems and entry is regulated by windows located periodically

through the structures. Free access to the structure is controlled by

the dimensions of the cavity windows and not by the free dimensions of

the cavities themselves.

The first report of the molecular sieve behavior of zeolites

was in 1925 when Weigel and Steinhoff studied the sorption properties of (22) an anhydrous chabazite. The affinity of dehydrated zeolites for a

wide variety of guest molecules was recognized at an early date by (11) Barrer who in 1938 reported on the sorption of argon, nitrogen,

hydrogen, and ammonia by five natural zeolites. Since 1938, the scien

tific literature has contained many reports on the sorption characteris

tics of natural and synthetic zeolites. Sorption studies on such zeolites

as erionite,^ Linde 5A,^^ gmelinite,^^ montmorillonite,^^ and (9 109 37 39 13) mordenite ’ ’ * * have led to the development of several important

commercial separation, drying, and purification processes. (133) (131) .As discussed by Wheeler and Weisz and Prater, the

surface reaction in heterogeneous catalyzed reactions may not be rate

controlling. In many solid catalyzed reaction systems, the processes

of bulk mass transfer, pore diffusion, and catalyst adsorption/desorption

•may determine or affect the overall reaction. Pore diffusion in catalyst

particles may occur by one or more of three possible mechanisms:

ordinary diffusion, Knudsen diffusion, and surface diffusion. Ordinary O diffusion usually occurs in pores with a diameter greater than 10,000 A;

and under these conditions, a molecule within the pore structure will (133) collide with other molecules far more often than with the pore wall.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ' 32

O Knudsen diffusion or molecular flow usually occurs in pores of 20-1000 A

diameter; and in this case, the molecules collide with the pore wall much

more frequently than with each other.Flow through very small

channels (5-20 A) is known as surface d i f f u s i o n ^ ^ * ^ ^ and is charac

teristic of the molecular movement in the channels of crystalline zeolites.

Prior to 1960, little research had been done toward the investi

gation of surface diffusion effects.(133,110) & resuit of the growing

interest in zeolites as catalytic agents, some work on the diffusion of /Og\ hydrocarbons in zeolite structures has begun to appear. Eberly

studied adsorption and diffusion of n-pentane through n-octane on erionite

and 5A molecular sieve. Measurements were made at temperatures from 93°

to 207°F and the diffusivities, calculated from adsorption and desorption

curves using Pick's second law, were quite different. Measured diffusion

coefficients were in the order of 10"7 cn?/sec. Satterfield and

Frabetti^**^ found that diffusion coefficients for desorption were from

3 to 60 times smaller than those for adsorption in their study of the

diffusion of -C4 .paraffin gases in single sodium mordenite crystals.

Measured diffusion coefficients at 0 to 20 cm. Hg and 25^ to 14CPC were

in the order of 10”9to 10~10 cnP/sec. Beecher et al^ ^ measured the

diffusivities of n-decane, decalin, toluene and various inert gases in

synthetic mordenites. Aluminum deficient mordenite prepared by acid

leaching was found to have a much lower diffusion resistance than conven

tional mordenite.

Acidity in zeolites may arise in two different ways depending

on how the zeolite is prepared or chemically treated. If the monovalent

cations in zeolites are exchanged for divalent cations such as CaP*,

then in certain parts of the zeolite crystal, each of the A104 tetrahedra

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33

will contain a net positive charge because of the charge imbalance and

the inability of the rigid crystal network to move and accomodate the

imbalance.This exchange can be represented b y ^

Na+ Na+ w v v v y / \ / \ / \ /■ \ / \ / \ Exchange N 0 0 0 0 0 '

o r o o oj /V V V V * 'V Vs/ V a/\

This net positive charge is the site of the acid activity of the catalyst,

and the smaller the divalent cation the greater the acid activity since

a larger electrostatic field results.

If, however, the monovalent cation in the original zeolite is

ion exchanged with mineral acids or exchanged with with subsequent

calcining, the hydrogen form (decationated) of the zeolite results.

Venuto et al^ ^ suggest that there exists a possible analogy between

hydrogen zeolites and certain heteropoly acids. Depending on temperature,

the acidity in hydrogen zeolites may take one of two possible forms.

At temperatures in the order of 3 0 C P c ^ ^ zeolites that have been Nl^ (127) exchanged will evolve NI^ and the following mechanism is postulated:'

4. *4* n n j ht o yOv 0 m 0 ,0. V °V 0H y ° \ l \ i ( ^ / A i ^ a t ( ?± A1 \si 0 \/ 0 0 0(/ o o o / o

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34

The equilibrium pair formed by loss of ammonia is viewed as a BrjJnsted

a c i d ^ ^ ^ and infrared studies by Liengme and H a l l ^ ^ and Uytterhoeven

et confirm their existence. At about 50CPC, a second process (127 123) called dehydroxylation, becomes important in decationated zeolites. ’

In this process, the protionic form exhibits instability and decomposes

losing one mole of water per pair of A104 tetrahedra to form the defect . . (127,123) structure.

2 °\ A1 °H)S1( /° - .feO + °\)Alx /°\ )S1 /° ♦ °\ Al Si +/° 0 00 o 0 00 0 0 0 0 0

This structure is viewed as a Lewis acid, and under certain activation (123) (19) conditions both Lewis and Br0nsted sites may be present. Benesi

studied the reactions of toluene, n-butane, and n-pentane on mordenite

and synthetic faujasite and determined the catalytic activity as a func

tion of the decomposition of the ammonia zeolites to their hydrogen

(BrfSnsted acid) and hydrogen-free (Lewis acid) forms.

4. Catalytic Properties of Zeolites

a. General

The first patent concerning the use of a zeolite catalyst

was issued in 1917 and reported the use of palladium-exchanged chabazite

for hydrogenation-dehydrogenation reactions.From 1917 to 1960,

industrial and academic interest in zeolites was directed primarily

toward structural identification and use as selective adsorbents. Since

I960, however, the open literature on zeolite catalysis has become

voluminous, and it is postulated that zeolite catalysts will be far more

important commercially than molecular sieve adsorbents were ever expected (79) to be. The unique properties of crystalline zeolites permit

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35

catalysis to be treated as an exact science with catalysts tailor-made

to fit the properties of the desired reactant(s) . The unique properties

that set zeolite catalysts apart from conventional ones a r e / 9^*

a. Their ionic surfaces provide unusually strong

carbonium ion activity.

b. Their exchangeable cations facilitate the

introduction of catalytically active metals in the highest degree of

dispersion.

c. They are completely crystalline with uniform

internal pores of molecular dimension.

Since 1960, crystalline zeolites have been found to

catalyze numerous organic reactions of interest. Catalytic cracking was (79) the first commercial breakthrough for zeolite catalysis. Catalysts

investigated for cracking activity include natural and synthetic

u,orda„ite/60-129*W 6 -137-59-78> fa»jaslte,<129'99’78> x-aeollte/4®

grnelimta/129-146-78) stilblta,'129’78’ chabaalte,<78> and oKretite<78>

Zeolites investigated for high pressure hydrocracking typically are

impregnated with an active dehydrogenation-hydrogenation component and

include mordenite^17* and faujasite/47*

Crystalline zeolites have frequently been reported as

active isomerization catalysts and may or may not have an active hydro-

genation-dehydrogenation component. Zeolites used for isomerization of

n-paraffins/39,104,103,25,53,16* naphthenes/53,39* and alkylaromat-

ics<150> include mordenite,<25,53,16,39,150> X-zeolite,<103> and

Y-zeolite/1^4,1^3* Other reported applications of zeolites are alkyla-

t i o n / 126* catalytic reforming/147,136’149’148* olefin polymerization^86*

and dehydrohalogenation/123*

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Due to their unique properties, crystalline zeolites have

great potential for catalysis of commercially important reactions. Table

3 lists some of the processes that can look to zeolites for calculated

catalysis.

b. Mordenite

Natural mordenite has good thermal stability, acid resis

tance and ideally has a fairly open structure.In spite of these

advantages, natural mordenite is subject to stacking f a u l t s ^ ^ which can o o (127) reduce its effective channel size from about 7 A to about 4 A. * This

reduction in effective channel size bars molecules such as n-heptane,

cyclohexane, or benzene from the crystalline structure. Recently devel

oped synthetic mordenites seem to be more free of stacking faults and

have considerable capacity for large molecules.

One of the first reports on catalysis over mordenite was

published in 1961 by Keough and S a n d . ^ ^ These authors studied cracking

of n-decane and hexadecane over the hydrogen form of synthetic sodium

mordenite and found it to be much more active than conventional Si02 - (59) Al^Og cracking catalyst. Keough reports data on the cracking of

n-decane and cumene over hydrogen mordenite and various cation exchanged

synthetic mordenites. Hydrogen mordenite had the greatest cracking

activity, and the data obtained are compatible with a carbonium ion ( 129) mechanism. Data obtained by Weisz and Miale on the catalytic crack

ing of n-hexane confirm the high activity of mordenite in relation to

conventional silica-alumina cracking catalyst. (47) Extensive data were obtained by Hatcher' on hydrocrack

ing of cyclohexane and n-hexane at typical hydrocracking conditions over

synthetic hydrogen mordenite impregnated with palladium. A reaction

with permission of the copyright owner. Further reproduction prohibited without permission 37

(79^ Table 3. Possible Applications of Zeolite Catalysts.

Probable Process Zeolite Catalysts Competitive Features

Catalytic Cracking X and Y Improved Yields; Reduced Light Gas and Coke.

Isomerization Y, Mordenite Troublesome Activators Unnecessary; High Selectiv ity and Resistance to S Poisoning.

Catalytic Reforming Y, Mordenite Activators Unnecessary; Gasoline with Reduced Sensitivity.

Polymerization Y Noncorrosive.

Alkylation Y Noncorrosive; Feed Pretreat ment Minimized.

Hydrodealkylation X and Y High Activity and Improved Selectivity.

Hydrogenation X and Y Resistance to S Poisoning.

Hydrogenation of X and Y High Selectivity and Low Fats and Oils Isomerization.

Selective Hydro A Separation Problems genation Minimized.

Methanation X and Y High Yields; Resistance to Poisons.

Dehydrogenation X and Y Improved Selectivity.

Dehydration A Improved Rates and Yields.

Dehydrohalogenation A Molecular Size; Selectivity.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38

mechanism Is presented and results on the mordenite are compared to the

results obtained on a synthetic faujasite. Beecher et al^ ^ report data

on hydrocracking of n-decane and decalin over palladium hydrogen mordenite

catalysts. An aluminum deficient Pd-H mordenite prepared by acid leaching

was found to have superior hydrocracking activity to the conventional

mordenite, which was accounted for in part by lowered diffusion resis

tance .

Isomerization of hydrocarbons via zeolite catalysts is a

subject of much current interest. Reported advantages of zeolite base

catalysts for isomerization are: high activity, lack of corrosion

problems, and resistance to poisons. Extensive data on the isomeri

zation of n-pentane, n-hexane, and cyclohexane have been reported by _ (25) „ . (16) . „ (53) Bryant, ' Beecher, ' and Hopper. (25) Bryant ' studied the isomerization of n-pentane over

synthetic hydrogen mordenite with and without palladium impregnation.

The overall conversion rate was found to be independent of concentration

of the dispersed metal, indicating that the hydrogenation-dehydrogenation

functions of the catalyst were not rate limiting. The conversion rate,

however, was significantly influenced by the acidity of the cations in

the crystal lattice.

Beecherstudied the isomerization of n-hexane plus

other normal paraffins over synthetic hydrogen mordenite and faujasite.

Both catalysts were prepared with and without palladium impregnation.

The mordenites were found to be much more active for paraffin isomeri

zation than the faujasites. A detailed reaction mechanism for the

isomerization of n-hexane is presented.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 39

A series of palladium hydrogen mordenite catalysts of (5 3 ) varying SiOa/AlgOa ratio were evaluated by Hopper for the isomeri

zation of n-pentane and cyclohexane. The relative activity for cyclo

hexane isomerization was found to be independent of the SiOjg/ALgOs ratio

while the activity for pentane decreased with decreasing SlOg/AlgS^ ratio.

Although most research on mordenite-containing catalysts (136. has been directed toward cracking and isomerization, recent patents 149 148) * indicate that mordenite may find application in catalytic

reforming. Mobil Oil Corporation^claims a naphtha reforming catalyst

can be made using an acid extracted synthetic mordenite. Standard Oil (149 148) Company of Indiana' ' 7 describes a process whereby a platinum

alumina mordenite catalyst is active for naphtha reforming if basic % nitrogen compounds are added to the feed.

D. Catalytic Mechanisms t 1. Dehydrogenation of Naphthenes

The mechanism of naphthene dehydrogenation has been studied (4) extensively by Balandin who has postulated the multiplet theory for

dehydrogenation catalysis. Balandin postulates that catalytic dehydro

genation occurs when a group of surface atoms appropriately spaced and

of necessary activity adsorb the reactant in a definitely oriented

position. Cyclohexane is assumed to adsorb on a hexagoned metal crystal

face such that all six hydrogen atoms are lost simultaneously. Balandin

has shown that those metals possessing the correct crystal structure and

interatomic distances are effective for cyclohexane dehydrogenation and

vice versa. Furthermore, the multiplet theory predicts that: (a) cyclo-

hexene and cyclohexadiene should not be obtainable from partial dehydro

genation of cyclohexane and (b) cyclopentane and cycloheptane should not

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. AO

be dehydrogenated by metal catalysts. While this theory satisfactorily

explained experimental data at that time, more recent data on oxide and

metal acidic catalysts essentially refute the multiplet hypothesis of

dehydrogenat ion.

Herrington and Rideal^^ studied the dehydrogenation of cyclo

hexane to benzene over Cr2 03 supported on alumina. Cyclohexene was

observed and the proposed mechanism was

0 * O * 0 * o + % + ^5 +

No cyclohexadiene was observed experimentally and it was concluded that

this component existed only on the catalyst surface. It was also con

cluded from the kinetic measurements that the loss of the first pair of

hydrogen atoms was the rate controlling step.

Hills et al^ t y studied the reactions of cyclohexane, cyclo

hexene, methylcyclopentane, and methylcyclopentene on three different

types of catalysts. These hydrocarbons were tested using a single func

tion isomerization catalyst, a single function dehydrogenation catalyst,

and a bifunctional Houdriforming catalyst (noble metal-acidic oxide) . It

was found that the dehydrogenation of cyclohexane to benzene proceeded

as well over the dehydrogenation catalyst as it did over the bifunctional

catalyst but did not take place over the catalyst containing only acidic

functionality. Olefin formation was observed with cyclohexane on both

dehydrogenation and bifunctional catalysts. Like Herrington and

R i d e a l / ^ these authors state that the dehydrogenation of cyclohexane

to benzene may proceed stepwise through the olefin and diolefin

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41

» intermediates. Additionally, they postulate that the stepwise dehydro

genation may take place on a single dehydrogenation site which can

account for low observed olefin concentrations in the gas phase.

2. Isomerization of Naphthenes

Two general types of catalytic mechanisms have been proposed

for the isomerization of saturated hydrocarbons. One has been postulated

for isomerization of saturated hydrocarbons over acidic halide catalysts

and the other for supported metal catalysts.

Isomerization of saturated hydrocarbons over acidic halide

catalysts is presently thought to occur by a carbonium ion chain mecha

nism first proposed by Bloch et al^ ^ in their study of n-butane isom

erization over an AlClg-HCl catalyst. These a u t h o r s ^ ^ state that the

mechanism is also applicable to cycloparaffins, and confirming evidence

has been obtained for the isomerization of methylcyclopentane and cyclo- (93 91 92 94) hexane. ’ * ’ To start the ionic chain reaction, an initiator is

needed.Trace amounts of oxygen,alkyl halides,olefins, (92) o r light have been found to be effective initiators. The mechanism (91) can be demonstrated for the isomerization of methylcyclopentane as:

a. Carbonium ion formation

,+ + R

b. Carbonium ion rearrangement

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42

c. Chain propagation O * O'- O * d CHj

A different type of mechanism has been proposed for the isom

erization of saturated hydrocarbons catalyzed by metals on acidic

supports. This bifunctional mechanism, as discussed by Sinfe l t ^ ^ and /£1\ Keulemans and Voge, involves both hydrogenation-dehydrogenation and

isomerization sites with gas phase migration of the reaction intermediates

between the sites.. This mechanism can be demonstrated for the isomeri

zation of cyclohexane as:

a. Dehydrogenation to olefin

Metal Site

o ■ □ * '

b. Carbonium ion formation

Acid Site + H+ 0

c. Carbonium ion isomerization

Acid + Acid + P X Site | |CHg Site .----- .CHj

U s o = o

d. Olefin formation

Acid p ’ " " c r * H+

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. A3

e. Olefin hydrogenation

Metal ji CH3 Site i JCH3 C3 - q Confirming evidence in support of this bifunctional mechanism for naph-

thene isomerization over metal-acidic oxide catalysts has been obtained

by Hinden et al^ ^ andMills et al. ^ ^ Keulemans and Voge^^ propose

that the carbonium ion isomerization is rate limiting.

3. Combination Dehydrogenation and Isomerization of Naphthenes

Mills et al^*^ studied the reactions of cyclohexane, cyclo-

hexene, methylcyclopentane, and methylcyclopentene on three types of

catalysts. These catalysts were a single function dehydrogenation

catalyst, a single function isomerization catalyst and a bifunctional

dehydrogenation-isomerization catalyst. It was found that dehydrogenation

of cyclohexane to benzene was essentially the same over the bifunctional

and dehydrogenation catalysts but did not take place over the catalyst

containing only isomerization functionality. The isomerization and

dehydroisomerization of methylcyclopentane to cyclohexane and benzene

only occurred to a significant extent over the bifunctional catalyst

while the isomerization of cyclohexane to methylcyclopentene required

only an acidic functionality. To explain these observations, Mills

et al^*^ proposed the reaction scheme shown in Figure 4 for reforming

C6 hydrocarbons. The vertical paths in Figure 4 take place on the

dehydrogenation-hydrogenation sites while the horizontal paths take place

on the isomerization sites. These authors propose that methylcyclopentane

conversion to benzene proceeds by first converting to methylcyclopentene

on a dehydrogenation site. This methylcyclopentene then migrates to an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44

n-Hexane i-Hexanes It iT C% s: n-Hexene i-Hexenes o a CO ij

Isomerization Sites

, Reforming C« Hydrocarbons with a Bifunctional Figure 4. B ®* Catalyst (8°*

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45

isomerization site where cyclohexene is formed. The cyclohexene migrates

to a dehydrogenation site where further dehydrogenation to cyclohexadiene

and finally benzene takes place. ' (45) Haensel et al studied the dehydrogenation of cyclohexane

to benzene over a Pt-ALgOg catalyst at 520^0, 300 psig, and space

velocities from 1000 to 32,000. Cyclohexene was observed to be a primary

product. The dehydroisomerization of methylcyclopentane was also studied

over a Pt-AlgOg-Cl catalyst, and methylcyclopentene was observed as a

primary intermediate product. Based on these observations, Haensel and

co-workers proposed that the mechanism for the dehydrogenation and

isomerization of cyclohexane is

0 = 0-0 ^ - jCHg

which is essentially the same as that proposed by Mills et al. ^ ^

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. L IS T OF REFERENCES - CHAPTER I I

1. Andreev, A. A., Shopov, D. M . , and Kiperman, S. L., "Kinetics of Cyclohexane Dehydrogenation in a Gradientless System," Kinet. Katal. (U.S.S.R.) 7, 1092 (1966); Chemical Abstracts, 6£, 85291 (1967) .

2. Anonymous, "Molecular Sieves Enter the Catalyst Market," Chemical Engineering 7 1 , (No. 26), 52 (1964).

3. Arbuzov, Yu. A., Batuev, M. I., and Zelinskii, N. D., "Contact Isom erization of Cyclohexene," Bulletin of the Academy of Science (U.S.S.R.), 665 (1945); Chemical Abstracts, 42, 5857 (1948).

4. Balandin, A. A., "The Theory of Heterogeneous Catalytic Reactions. The Multiplet Hypothesis. Model for Dehydrogenation Catalysis," Z. Physik. Chem., 2^, 289 (1929); Chemical Abstracts, 23, 2872 (1929) .

5. Balandin, A. A., Karpeiskaya, E. I. and Tolstopyatova, A. A., "Catalytic Dehydrogenation of Hydrocarbons and Alcohols over Metallic Rhenium," Doklady Akad. Nauk (U.S.S.R.) 122, 227 (1958); Chemical Abstracts, 53, 837 (1959). 5 6 . Balandin, A. A. and Kostin, F. L., "Kinetic Study on the Dehydro genation of Cyclohexane," Acta Physicochim. (U.S.S.R.), JL7, 211 (1942); Chemical Abstracts, 3 7 , 4615 (1943).

7. Barrer, R. M., "Molecular Sieves," Endeavor, 23 (No. 90), 122 (1964).

8 . Barrer, R. M., "Some Aspects of Molecular Sieve Science and Tech nology," 2. 1203 (1968) .

9. Barrer, R. M., "Sorption by Gmelinite and Mordenite," Transactions of the Faraday Society, 40, 555 (1944).

10. Barrer, R. M., "Synthesis and Reactions of Mordenite," Journal of the Chemical Society, Part 2, 2158 (1948).

11. Barrer, R. M., "The Sorption of Polar and Non-Polar Gases by Zeolites Proceedings of the Royal Society of London, A-167, 393 (1938) .

12. Barrer, R. ,M. and MacLeod, D. M., "Intercalation and Sorption by Montmorillonite," Transactions of the Faraday Society, 50 , 980 (1954).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 7

13.. Barrer, R. M. and Peterson, D. L., "Intracrystalline Sorption by Synthetic Mordenites," Proceedings of the Royal Society of London, A-280, 466 (1964).

14. Barrer, R. M. and Peterson, D. L., "Kinetics of N-Paraffin Sorption in the Natural Zeolite Erionite," Journal of Physical Chemistry, 68 (No. 11), 3427 (1964).

15. Barrett, W. T., Sanchez, M. G., and Smith, J. G., "Phase Transfor mations in Silica Alumina Catalysts," Proceedings of the First International Congress of Catalysis, 551 (1956).

16. Beecher, R. G., "Hexane Isomerization," Ph.D. dissertation, Department of Chemical Engineering, Louisiana State University, 1967.

17. Beecher, R. G., Voorhies, A., Jr., and Eberly, P. E., Jr., "Hydro cracking and Diffusion of Pure Compounds on Mordenite Catalysts," Industrial and Engineering Chemistry Product Research and Development, 7_, 203 (1968).

18. Benesi, H. A., "Acidity of Catalyst Surfaces. II. Amine Titration Using Hammett Indicators," Journal of Physical Chemistry, 61, 970 (1957) .

19. Benesi, H. A., "Relationship Between Catalytic Activity and Nature of Acidity of the Crystalline Zeolites, Mordenite, and Y Faujasite," Journal of Catalysis, j5, 368 (1967).

'20. Bloch, H. S., Pines, H., and Schmerling, L., "The Mechanism of Paraffin Isomerization," Journal of the American Chemical Society, 6 8 , 153 (1946).

21. Bloch, H. S. and Thomas, C. L., "Hydrocarbon Reactions in the Presence of Cracking Catalysts. III. Cyclohexene, Decalin, and Tetralin," Journal of the American Chemical Society, 6 6 , 1589 (1944).

22. Breck, D. W . , "Crystalline Molecular Sieves," Journal of Chemical Education, 41 (No. 12), 678 (1964).

23. Bridges, J. M. and Houghton, G., "The Evaluation of Activation Energies Using a Rising Temperature Flow Reactor. The Dehydro genation of Cyclohexane over WS3 , Pt/AlgOg , CrgOg/Al^Og, NiW04 / Al^Og , and Cr2 03 ," Journal of the American Chemical Society, 81, 1334 (1959).

24. Briggs, R. A. and Taylor, H. S., "The Dehydrogenation of Normal Heptane and Cyclohexane on Cerium, Vanadium, and Thorium Oxide Catalysts," Journal of the American Chemical Society, 63, 2500 (1941) .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 8

25. Bryant, P. A., "Hydroisomerization of Normal Pentane over a Zeolite Catalyst," Ph.D. dissertation, Department of Chemical Engineering, , Louisiana State University, 1966.

26. Ciapetta, F. G., "Isomerization of Saturated Hydrocarbons - Cyclo- alkanes," Industrial and Engineering Chemistry, 45 (No. 1), 159 (1953).

27. Ciapetta, F. G., "Special Report - Catalytic Reforming," Petro/ Chem Engineer, 33 (No. 5), C-19 (1961).

'28. Ciapetta, F. G., Dobres, R. M. and Baker, R. W.., "Catalytic Reform ing of Pure Hydrocarbons and Petroleum Naphthas," Catalysis, (>, 495, Reinhold Publishing Corp., New York, New York, 1958.

29. Ciapetta, F. G. and Hunter, J. B., "Isomerization of Saturated Hydrocarbons in the Presence of Hydrogenation - Cracking Catalysts," Industrial and Engineering Chemistry, 45 (No. 1), 147 (1953).

30. Clark, A., "Oxides of the Transition Metals as Catalysts," Industrial and Engineering Chemistry, 45 (No. 7), 1476 (1953).

31. Clark, A., Matuszak, M. P., Carter, N. C., and Cromeans, J. S., "Isomerization of N-Pentane in the Presence of Molybdena Alumina Catalyst at Low Hydrogen-Hydrocarbon Ratio," Industrial and Engineering Chemistry, 45, 803 (1953).

32. Condon, F. E., "Catalytic Isomerization of Hydrocarbons," Catalysis, <5, 43, Reinhold Publishing Corp., New York, New York, 1958.

33. Cusumano, J. A., Dembinski, G. W., and Sinfelt, J. H . , "Chemisorption and Catalytic Properties of Supported Platinum," Journal of Catalysis, 5, 471 (1966).

34. Domine, D. and Quobex, J., "Synthesis of Mordenite," Conference on Molecular Sieves, 78, Society of Chemical Industry, London, 1967

35. Eastam, A. M., "Co-Catalysis in Friedel-Crafts Reactions I. Boron Fluoride-Water," Journal of the American Chemical Society, 78, 6040 (1956).

36. Eberly, P. E., Jr., "Adsorption of Normal Paraffins in Erionite and 5A molecular Sieve," American Chemical Society Preprints, Division of Petroleum Chemistry, 13 (No. 3), 216 (1968).

37. Eberly, P. E., Jr., "Hydrocarbon Adsorption Studies at Low Pressures on the Sodium and Acid Forms of Synthetic Mordenite," Journal of Physical Chemistry, 67, 2404 (1963).

38. Emmett, P. H., Sabatier, P., and Reid, E. E., Catalysis Then and Now, Franklin Publishing Company, Englewood, N. J., 1965.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4 9

39. Frilette, V. J. and Rubin, M. K., "Sorption and Catalytic Properties of Natural Mordenite," Journal of Catalysis, 4, 310 (1965) .

40. Frilette, V. J., Weisz, P. B., and Golden, R. L., "Catalysis by Crystaline Aluminosilicates. I. Cracking of Hydrocarbon Types over Sodium and Calcium X Zeolites," Journal of Catalysis, 1_, 301 (1962).

41. Greensfelder, B. S., Archibald, R. C., and Fuller, D. L., "Catalytic Reforming - Fundamental Hydrocarbon Reactions of Petroleum Naphthas with Molybdena-Alumina and Chromia-Alumina Catalysts," Chemical Engineering Progress, 43 (No. 10), 561 (1947).

42. Grosse, A. V., Morrell, J. C., and Mattox, W. J., "Catalytic Cyclization of Aliphatic Hydrocarbons to Aromatics," Industrial and Engineering Chemistry, 32 (No. 4), 528 (1940).

43. Haensel, V., "Catalysis in the Petroleum Industry," presented at the 31st Midyear Meeting of the American Petroleum Institute's Division of Refining, Houston, Texas, May 9, 1966.

44. Haensel, V. and Donaldson, G. R., "Platforming of Pure Hydrocarbons," Industrial and Engineering Chemistry, 43 (No. 9), 2102 (1951) .

45. Haensel, V., Donaldson, G. R., and Riedl, F. J., "Mechansims of Cyclohexane Conversion Over Platinum-Alumina Catalysts," Proceedings of the Third International Congress of Catalysis, 294 (1964).

46. Hansch, C., "The Dehydrocyclization Reaction," Chemical Reviews, 53, 353 (1953) .

47. Hatcher, W. J., Jr., "Hydrocracking of Normal Hexane and Cyclo hexane over Zeolite Catalysts," Ph.D. dissertation, Department of Chemical Engineering, Louisiana State University, 1968.

48. Heinemann, H . , Mills, G. A., Hattman, J. B., and Kirsch, F. W., "Houdriforming Reactions - Studies with Pure Hydrocarbons," Industrial and Engineering Chemistry, 45 (No. 1), 130 (1953).

49. Herbo, C., "Research on the Mechanism of Catalytic Reactions. II Kinetic Studies on the Dehydrogenation of Cyclohexane," Bulletin of the Chemical Society Belgium, 51, 44 (1942); Chemical Abstracts, 37, 5307 (1943).

50. Herington, E. F. G. and Rideal, E. K . , "The Catalytic Dehydrogena tion of Naphthenes. I. Kinetic Study," Proceedings of the Royal Society of London, A-190, 289 (1947^ .

51. Hettinger, W. P., Keith, C. D., Gring, J. L., and Teter, J. W. , "Hydroforming Reactions - Effect of Certain Catalyst Properties and Poisons," Industrial and Engineering Chemistry, 47 (No. 4), 719 (1955).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50

52. Hindin, S. G., Weller, S. W., and Mills, G. A., "Mechanically Mixed Dual-Function Catalysts," Journal of Physical Chemistry, 62, 244 (1958) .

53. Hopper, J. R., "A Study of the Catalytic Hydroisomerization Reactions of N-Pentane and Cyclohexane over Structurally Modified Mordenites," Ph.D. dissertation, Department of Chemical Engineering, Louisiana State University, 1969.

54. Hughes, E. C. and Darling, S. M . , "Lower Paraffin Hydrocarbons, Catalytic Conversion by Boron Fluoride with Hydrogen Fluoride," Industrial and Engineering Chemistry, 43, 746 (1951) .

55. Innes, W. B., "Classification of Heterogeneous Catalytic Vapor Phase Reactions," Catalysis, ,2, 1, Reinhold Publishing Corp., New York, New York, 1955.

56. Ipatieff, V. N. and Grosse, A. V., "Polymerization of Ethylene with Aluminum Chloride," Journal of the American Chemical Society, 58, 915 (1936).

57. Isagluyants, G. V., Ryashentseva, M. A., Derbentsev, Yu. I., Minachev, Kh. M . , and Balandin, A. A., "The Role of Cyclohexane in Cyclohexane Dehydrogenation and Isomerization Under Reforming Conditions," Neftekhimiya .(U.S.S.R.), 4 (No. 2), 229 (1964); Chemical Abstracts, 61, 8103 (1964).

58. John, G. S., Den Herder, M. J., Mikovsky, R. J., and Waters, R. F., "Physicochemical Studies of Molybdena Reforming Catalysts," Proceedings of the First International Congress of Catalysis, 252 (1956).

59. Keough, A. H., "Catalytic Cracking of Hydrocarbons with Open Synthetic Mordenites," American Chemical Society Preprints, Division of Petroleum Chemistry, 8 (No. 1), 65 (1963).

60. Keough, A. H.. and Sand, L. B., "A New Intracrystalline Catalyst," Journal of the American Chemical Society, 83, 3536 (1961).

61. Keulemans, A. I. and Voge, H. H., "Reactivities of Naphthenes over a Platinum Reforming Catalyst by a Gas Chromatographic Technique," Journal of Physical Chemistry, 63, 476 (1959).

62. Kira, G. and Krieger, K. A., "Catalytic Activity of Chromia-Silica for the Dehydro-Aromatization of N-Heptane," Proceedings of the Second International Congress of Catalysis, Section I, No. 28, 1 (1960).

63. Komarewsky, V. I. and Shand, W. C., "Catalytic Aromatization of Branched Chain Aliphatic Hydrocarbons," Journal of the American Chemical Society, 6 6 , 1118 (1944).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51

64. Krane, H. G., Groh, A. B., Schulman, B. L., and Slnfelt, J. H . , Reactions in Catalytic Reforming of Naphthas," Proceedings Fifth World Petroleum Congress, Section III, 39 (1959).

65. Lawrence, P. A. and Rawlings, A. A., "Advances in Isomerization," Proceedings Seventh World Petroleum Congress, 4, 135 (1967).

6 6 . Lazier, W. A. and Vaughen, J. V., "Non-Metallic Catalysts for Hydrogenation and Dehydrogenation. II. The Catalytic Properties of Chromium Oxide," Journal of the American Chemical Society, 54, 3080 (1932).

67. Leighton, P. A. and Heldman, J. D., "The Catalytic Isomerization of Paraffin Hydrocarbons. I. Butanes," Journal of the American Chemical Society, 65, 2276 (1943).

6 8 . Lien, A. P., D'Ouville, E. L., Evering, B. L., and Grubb, H. M., "Rate of Isomerization of Cyclohexane," Industrial and Engineer ing Chemistry, 44 (No. 1) , 351 (1952).

69. Liengme, B. V. and Hall, W. K., "Studies of Hydrogen Held by Solids," Transactions of the Faraday Society, 62, 3229 (1966).

70. Lyster, W. N., "Kinetics of Chemical Reactions - Dehydrocyclization of N-Heptane," Ph.D. dissertation, Department of Chemical Engineering, University of Houston, 1964.

71. Maat, H. J. and Moscou, L., "A Study of the Influence of Platinum Crystallite Size on the Selectivity of Platinum Reforming Catalysts," Proceedings of the Third International Congress of Catalysis, Section II. 5, 1277 (1964).

72. Maslyanskii, G. N., "Kinetics of Isomerization of Cyclohexane at High Pressures," Journal of General Chemistry (U.S.S.R.), 13, 540 (1943); Chemical Abstracts, 39, 454 (1945).

73. McBain, J. N., The Sorption of Gases and Vapors by Solids, Rutledge and Sons, Ltd., London, 1932.

74. McHenry, K. W., Bertolacini, R. J., Brennan, H. M . , Wilson, J. L., and Seelig, H. S., "The Nature of the Platinum Dehydrocyclization Catalysts," Proceedings of the Second International Congress of Catalysis, Section II, No 117, 1 (1960).

75. Meier, W. M., "The Crystal Structure of Mordenite (Ptilolite)," Zeitschrift fur Kristallographie, 115, 439 (1961).

76. Meier, W. M., "Zeolite Structures," Conference on Molecular Sieves,' 10, Society of Chemical Industry, London, 1967.

77. Meisel, S. L., Koft, E., Jr., and Ciapetta, F. G., "Effect of Nitrogen Compounds on Platinum Acidic Oxide Catalysts," American Chemical Society Preprints, Division of Petroleum Chemistry, (No. 4), A-45 (1957).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52

78. Miale, J. N., Chen, N. Y., and Weisz, P. B., "Catalysis by Crystalline Aluminosilicates IV. Attainable Catalytic Cracking Rate Constants, and Superactivity," Journal of Catalysis, 6 , 278 (1966).

79. Miller, R., "Molecular Sieve Catalysts," Chemical Week, 95 (No. 20), 77 (1964).

80. Mills, G. A., Heinemann, H., Milliken, T. H., and Oblad, A. G., "Houdriforming Reactions - Catalytic Mechanism," Industrial Engineering Chemistry, 45 (No. 1), 134 (1953).

81. Mills, G. A., Weller, S., and Cornelius, E. B., "The State of Platinum in a Reforming Catalyst," Proceedings of the Second International Congress of Catalysis, Section II, No. 113, 1 (1960).

82. Minachev, Kh. M. and Isagulyants, G. V . , "Investigation of Catalyst Poisoning and Hydrocarbon Conversion Mechanism in Reforming Process," Proceedings of the Third International Congress of Catalysis, 309 (1964).

83. Minachev, Kh. M . , Shuikin, N. I., and Rozhdestrenskaya, I. D., "Hydro- and Dehydrogenation of Hydrocarbons in the Presence of Ruthenium and Rhodium Catalysts with a Low Content of the Metal," Izvest. Akad. Nauk (U.S.S.R.), 338 (1954); Chemical Abstracts, 48, 10259 (1954).

84. Moldavskii, B. L., Kamusher, G. D., and Livshits, S. E., "Catalytic Dehydrogenation of Hydrocarbons. I Dehydrogenation of Cyclohexane in the Presence of Sulfide and Oxide Catalysts," Journal of General Chemistry (U.S.S.R.), 7, 131 (1937); Chemical Abstracts, 31, 4282 (1937).

85. Nehring, D. and Dreyer, H . , "Influence of Various Catalyst Supports on the Catalytic Properties of Pt," Chemical Technology, (Berlin), 12, 343 (1960); Chemical Abstracts, 5 5 , 3159 (1961).

8 6 . Norton, C. J., "Olefin Polymerization over Synthetic Molecular Sieves," Industrial and Engineering Chemistry Process Development, 3 (No. 3), 230 (1964).

87. Oblad, A. G., Milliken, T. H., and Mills, G. A., "Chemical Charac teristics and Structure of Cracking Catalysts," Advances in Catalysis, 3_, 199, Academic Press, New York, New York, 1951.

8 8 . Oblad, A. G., Shalit, H., and Tadd, H. T., "Catalytic Technology in the Petroleum Industry," Proceedings of the First International Congress of Catalysis, 510 (1956).

89. Pauling), L., "The Structure of Some Sodium and Calcium Aluminosili cates]," Proceedings of the National Academy of Science (U.S.), 16, 453 (1930).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53

90. Pickert, P. E., Bolton, A. P., and Lanewala, M. A., "Molecular Sieve Zeolites: Trendsetters in Heterogeneous Catalysis," Chemical Engineering, 75 (No. 16), 133 (1968).

91. Pines, H., Abraham, B. M . , and Ipatieff, V. N., "Isomerization of Saturated Hydrocarbons. V. The Effect of Cyclohexene Upon the Isomerization of Methylcyclopentane and Cyclohexane," Journal of the American Chemical Society, 70, 1742 (1948).

92. Pines, H., Aristoff, E., and Ipatieff, V. N., "Isomerization of Saturated Hydrocarbons. VII. The Effect of Light Upon the Isomerization of Methylcyclopentane in the Presence of Aluminum Bromide-Hydrogen Bromide," Journal of the American Chemical Society, 72, 4055 (1950).

93. Pines, H., Aristoff, E., and Ipatieff, V. N . , "Isomerization of Saturated Hydrocarbons. VIII. The Effect of Oxygen and Light Upon the Isomerization of Methylcyclopentane in the Presence of Aluminum Bromide," Journal of the American Chemical Society, 72, 4304 (1950).

94. Pines, H., Aristoff, E., and Ipatieff, V. N., "Isomerization of Saturated Hydrocarbons. XII. The Effect of Experimental Variables, Alkyl Bromides and Light Upon the Isomerization of Methylcyclopentane in the Presence of Aluminum Bromide," Journal of the American Chemical Society, 75, 4775 (1953).

95. Pines, H. and Chen, C. T., "Alumina: Catalyst and Support. VII. Aromatization of N-Octane-l-C14 and Cyclo-Octane over Chromia- Alumina Catalysts," Proceedings of the Second International Congress of Catalysis, 11 (I960).

96. Pines, H. and Csicsery, S. M., "Alumina: Catalyst and Support. XVI. Aromatization and Dehydroisomerization of Branched C6 -C8 Hydro carbons over Nonacidic Chromia-Alumina Catalyst," Journal of Catalysis, 1, 313 (1962).

97i Pines, H. and Haag, W. 0., "Alumina: Catalyst and Support. I. Alumina, Its Intrinsic Acidity and Catalytic Activity," Journal of the American Chemical Society, 82, 2471 (1960).

98. Pines, H. and Shaw, A. W., "Isomerization of Saturated Hydro carbons. XV. The Hydro-isomerization of Ethyl-a-C1 4 -Cyclo hexane," Journal of the American Chemical Society, 79, 1474 (1957).

99. Plank, C. J . , Rosinski, E. J., and Hawthorne, U. P., "Acidic Crystalline Aluminosilicates," Industrial and Engineering Chemistry Product Research and Development, 3^ (No. 3), 165 (1964).

100. Plate, A. F., "Mechanism of Catalytic Transformations of Hydro carbons on a Vanadium Catalyst. V. Simultaneous Hydrogenation and Dehydrogenation of Cyclic Olefins," Journal of General Chemistry (U.S.S.R.), 15, 156 (1945); Chemical Abstracts, 40, 3409 (1946).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54

101. Prokopetz, E. I. and Filaratov, A. N., "Isomerization of Cyclohexane and Methylcyclopentane in High-Temperature Hydrogenation," Journal of Applied Chemistry (U.S.S.R.), JL1, 1631 (1938); Chemical Abstracts, 33, 5817 (1939).

102. Puchkov, P. V. and Nikolaeva, A. F., "Transformation of Cyclohexane under Hydrogen Pressure," Journal of General Chemistry (U.S.S.R.), 8 , 1153 (1938); Chemical Abstracts, 33, 3766 (1939).

103. Rabo, J. A., Pickert, P. E., Stamires, D. N., and Boyle, J. E., "Molecular Sieve Catalysts in Hydrocarbon Reactions," Proceedings of the Second International Congress of Catalysis, Section II, No. 104, 1 (1960).

104. Rabo, J. A., Schomaker, V., and Pickert, P. E., "Sulfur Resistant Isomerization Catalyst: Study of Atomic Platinum Dispersions on a Zeolite Support," Proceedings of the Third International Congress of Catalysis, 1264 (1964) .

105. Rengarten, N. V., "A Zeolite from the Mordenite Group in the Upper Cretaceous and Paleogene Marine Deposits of the Eastern Slope of the Urals," Compt. Rend. Acad. Sci. (U.S.S.R.), 48, 591 (1945); Chemical Abstracts, 40, 4625 (1946) .

106. Ritchie, A. W. and Nixon, A. C., "Dehydrogenation of Monocyclic Naphthenes over a Platinum on Alumina Catalyst without Added Hydrogen," American Chemical Society Preprints, Division of Petroleum Chemistry, 11 (No. 3), 93 (1966).

107. Rossini, F. D. and Mair, B. J., Hydrocarbons from Petroleum, Reinhold Publishing Corp., New York, New York, 1953.

108. Ryland, L. B. Tamele, M. W., and Wilson, J. N., "Cracking Catalysts," Catalysis, 7, 1, Reinhold Publishing Corp., New York, New York, 1960.

109. Satterfield, C. N. and Frabetti, A. J., Jr., "Sorption and Diffusion of Gaseous Hydrocarbons in Synthetic Mordenite," American Institute of Chemical Engineering Journal, 13 (No. 4), 731 (1967) .

110. Satterfield, C. N., and Sherwood, T. K., The Role of Diffusion in Catalysis, Addison - Wesley Publishing Co., Inc., Reading, Mass., 1963.

111. Shalit, H. and Conner, J. E., Jr., "New Developments in Catalysis," Chemical Engineering, 72 (No. 7), 73 (1965).

112. Shuikin, N. I., "The Contact-Catalytic Isomerization of Six-Membered into Five-Membered Rings, "Bulletin of the Academy of Science (U.S.S.R.), 440 (1944); Chemical Abstracts, 3 9 , 4319 (1945).

113. Sinfelt, J. H., "Bifunctional Catalysis," Advances in Chemical Engineering, 5, 37, Academic Press, New York, New York, 1964.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 55

114. Spenadel, L. and Boudart, M., "Dispersion of Platinum on Supported Catalysts," American Chemical Society Preprints, Division of Petroleum Chemistry, 4 (No. 2-C), C-5 (1959) .

115. Steiner, H., "The Catalytic Dehydrocyclization of Paraffins," "Journal of the American Chemical Society, 67, 2052 (1945).

116. Stevenson, D. P. and Beeck, 0., "The Isomerization of Cyclohexane and Methylcyclopentane in the Presence of Aluminum Halides. I. The Nature of the Catalyst," Journal of the American Chemical Society, 70, 2890 (1948).

117. Stevenson, D. P. and Morgan, J. H., "The Isomerization of Cyclo hexane and Methylcyclopentane in the Presence of Aluminum Halides. II. Equilibrium and Side Reactions," Journal of the American Chemical Society, 70, 2773 (1948).

118. Stormont, D. H., "Modernization - Refinings Record for 1968," Oil and Gas Journal, 67 (No. 12), 108 (1969).

119. Stumpf, H. C., Russell, A. S., Newsome, J. W., and Tucker, C. M., "Thermal Transformations of Aluminas and Alumina Hydrates," Industrial and Engineering Chemistry, 42 (No. 7), 1398 (1950).

• 120. Tamele, M. W., "Chemistry of the Surface and the Activity of Alumina-Silica Cracking Catalyst," Discussions of the Faraday Society, {J, 270 (1950) .

121. Turkevich, J., "Zeolites as Catalysts," Catalysis Reviews, JL, 1, Marcel Dekker, Inc., New York, New York, 1968.

122. Tutumi, S., "Differences Among the Catalytic Actions of Cobalt, Palladium, and Platinum," Science Papers Institute Physics Chemical Research (Tokyo) , 3jj, 352 (1939) ; Chemical Abstracts, 33, 9104 (1939).

123. Uytterhoeven, J. B., Christner, L. G., and Hall, W. K., "Studies of Hydrogen Held by Solids. VIII. The Decationated Zeolite," Journal of Physical Chemistry, 69 (No. 6), 2117 (1965).

124. Varga, J., Rabo, G., and Zalai, A., "Comparative Study of the Promoters for Dehydrogenation Catalysts," Acta Chim. Hung., 1_, 137 (1951); Chemical Abstracts, 45, 10022 (1951).

125. Venuto, P. B., Givens, E. N., Hamilton, L. A., and Landis, P. S., "Organic Reactions Catalyzed by Crystalline Aluminosilicates. Dehydrohalogenation and Related Reactions," American Chemical Society, Division of Petroleum Chemistry, 11 (No. 3), 73 (1966).

126. Venuto, P. B., Hamilton, L. A., Landis, P. S., and Wise, J. J., "Alkylation Reactions Catalyzed by Crystalline Alumino-silicates," American Chemical Society, Division of Petroleum Chemistry, 10 (No. 4), B-71 (1965).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56

127. Venuto, P. B. and Landis, P. S., "Organic Catalysis over Crystalline Aluminosilicates," Advances in Catalysis, 18, 259, Academic Press, New York, New York, 1968.

128. Voge, H. H., "Catalytic Cracking," Catalysis, (5, 407, Reinhold Publishing Corp., New York, New York, 1958.

129. Weisz, P. B. and Miale, J. N., "Superactive Crystalline Alumino- silicate Hydrocarbon Catalysts," Journal of Catalysis, 4, 527 (1965) .

130. Weisz, P. B. and Prater, C. D., "Basic Activity Properties for Pt-Type Reforming Catalysts," Proceedings of the First Inter national Congress of Catalysis, 575 (1956) .

131. Weisz, P. B. and Prater, C. D . , "Interpretation of Measurements in Experimental Catalysis," Advances in Catalysis, 6_, 143, Academic Press, Inc., New York, New York, 1954.

132. Weisz, P. B. and Swegler, E. W., "Stepwise Reactions on Separate Catalytic Centers: Isomerization of Saturated Hydrocarbons," Science, 126, 31 (1957).

133. Wheeler, A., "Reaction Rates and Selectivity in Catalyst Pores," Advances in Catalysis, 3^ 249, Academic Press, Inc., New York, New York, 1951.

134. Zelinskii, N. D. and Balandin, A. A., "Kinetics of Dehydrogenation Catalysis," Bulletin of the Academy of Science (U.S.S.R.), 29 (1929); Chemical Abstracts, 2 4 , 774 (1930).

Patents

135. British Patent 992,872.

136. British Patent 1,051,890.

137. British Patent 1,074,496.

138. French Patent 1,369, 377.

139. South African Patent 67 06,642.

140. U. S. Patent 1,215,391.

141. U. S. Patent 2,479,109.

142. U. S. Patent 2,479,110.

143. U. S. Patent 2,651,598.

144. U. S. Patent 2,689,208.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 145. U. S. Patent 2,818,394

146. U. S. Patent 3,173,855

147. U. S. Patent 3,247,099

148. U. S. Patent 3,376,214

149. U. S. Patent 3,376,215

150. U. S. Patent 3,409,685

151. U. S. Patent 3,410,789

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER III

EXPERIMENTAL EQUIPMENT AND PROCEDURE

A. General

The experimental equipment used in this research is located in the

Petroleum Processing Laboratory of the Chemical Engineering Department

at Louisiana State University. The project is supported by Esso Research

and Engineering Company and all of the equipment was fabricated by or

purchased through Esso Research Laboratories, Baton Rouge, Louisiana.

The equipment was planned and designed by P. A. Bryant, an original

member of the project. Modifications were carried out by another

member of the project, W. J. Hatcher, Jr., and the author.

B . Experimental Equipment

1. Reaction System

The reaction system used in this research consists of a liquid

feed system, a gas feed system, a reactor, and a product recovery sys

tem. The reaction system with the exception of the liquid feed pump

and the gas cylinders is located within a walk-in hood. This hood is

provided with a 3000 ft.3 /min. exhaust fan and sliding glass safety

panels. The liquid feed pump is located adjacent to the hood and the

gas cylinders are located outside the building. The equipment housed

within the walk-in hood is supported on a steel frame, three feet wide,

three feet deep, and six feet high. A simplified flow diagram of the

reaction system is shown in Figure 5.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59

Ruska Pump

Back Pressure Liquid Feed Regulator Temperature Control^— 1 1 Controller Valve r ^ \ Needle Valve Pressure Gauge

DP Cell Flow Controller

Regulator Reactor Wet Test Meter {X]— Fluidized Bath I Bath Heating'' Elements Driers

Back-Pressure Regulator Gas Sampling Point Regulators

Liquid Ice Condenser Bath

Water Saturator Hydrogen Nitrogen Cylinder Cylinder Liquid Sampling Point

Figure 5. Simplified Flow Diagram of the Reaction System.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. a. Liquid Feed System

The liquid feed is metered to the reaction system by a

high pressure, positive displacement pump manufactured by the Ruska

Instrument Corporation, Houston, Texas. The pump acts on the principle

of a syringe with the piston being driven by a synchronous electric

motor. Feed rates are adjusted by changing gear arrangements to give

rates of 2 to 240 cnf3 /hr. The precision of the Ruska pump is 0.15

cnP/hr and its capacity is 250 cut3. The pump is charged by a 250 cirt3

burette.

b. Gas Feed System

The bottled gases used in this study were hydrogen and

nitrogen. Hydrogen was used as a feed diluent while nitrogen was used

as a reactor and liquid feed system purge. Both gases are independently

routed from the high pressure cylinders through regulators and molecular

sieve driers to a manifold in the reaction system. The gas being used

flows from the manifold through another regulator and then through a

flow control system.

The primary gas flow control system consists of a Foxboro

orifice differential pressure cell, Foxboro flow controller and a

Research Control % inch control valve joined together in a simple feed

back control network. The orifice and valve trim were sized to give

flow rates from 2 to 20 ft3 /hr. A % inch micrometer needle valve for

manual flow control is in parallel with the control valve as a back-up

in case of failure in the automatic system.

After passing through the flow control system the gas

passes through a Grove back pressure regulator. The regulator is used

to provide a constant pressure drop across the gas flow control system

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61

regardless of reactor pressure. The gas then goes by a pressure gauge

for measuring reactor pressure to a tee where mixture with the liquid

occurs. This mixture then flows into the reactor,

c. Reactor

A cross-sectional view of the reactor in the fluidized

heating bath is shown in Figure 6 . The reactor is constructed of %

inch schedule 80 Inconel pipe. While designed to hold a maximum of 30-

40 cnt3 of catalyst, 15 cnt3 was the standard volume used in all of these

experiments. This corresponded to a cylindrical catalyst bed % inch in

diameter by 4 inches in length. A flange opening is provided in the

reactor for introducing and removing the catalyst. A hollow steel

0-ring forms a pressure tight seal in the V-groove seat of the flange

when the temperature compensating coupling is attached. The complete

flange assembly is manufactured by the D. S. D. Company of East Granby,

Connecticut.

The catalyst is positioned in the reactor between two

micrometallic porous frits. These frits prevent catalyst particles

from leaving the reactor.

The temperature in the catalyst bed is measured by a 1/16

diameter, metallic sheathed, iron-constantan thermocouple which is

positioned midway with respect to the length and diameter of the catalyst

bed. The thermocouple enters the reactor through a \ inch Conax fitting.

Temperature readings of the catalyst bed and fluidized heating bath were

taken on a Leeds and Northrup Speedomax-H temperature indicator. This

temperature indicator was calibrated by an external potentionmeter and

was found to be accurate within one degree Fahrenheit.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62

Bath Thermocouple Line Product LineFeed

Porous Sheathed Exit _ Frit Thermocouple for v Fluidizing Air

Thermocouple Well for Bath Temperature Control Sand Bath

Reactor

Catalyst

Glass Wool

Temperature Compensated Coupling

Inlet for Fluidizing Air Air Distributor

Figure 6. Cross Sectional View of Reactor in Fluidized Heating Bath.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63

The feed mixture of gas and liquid is heated to reaction

conditions before entering the reactor by coiling the feed line several

turns around the reactor before connecting to the reactor. The feed

mixture then passes downflow through the catalyst bed and out the

product line.

d. Temperature and Pressure Control

The fluidized heating bath as shown in Figure 6 is

designed for isothermal control of the reactor temperature. This is

accomplished by using a fluidized bed to transfer heat to the reactor.

This bed consists of spent silica-alumina cracking catalyst fluidized

by a continuous stream of air. The fluidized bed is approximately 6

inches in diameter and 25 inches high. Air enters the bed through a

distributor and a micro-metallic porous frit. Entrained material from

the fluidized bed is retained by a circular micrometallic porous frit

at the top of the bed. The fluidized bed is heated by six 500 watt

strip heater elements. Two inches of insulation cover the heating

elements.

Control of the fluidized bath temperature is achieved by

a simple feedback control network. The bath temperature is sensed by

an iron-constantan thermocouple inserted in a well adjacent to the

reactor. This temperature signal goes to a Leeds and Northrup Electro

max C. A. T. (current-adjusting-type) solid state temperature controller.

This controller, which is equipped with proportional and reset control

action generates a direct current control signal after comparison of the

sensed bath temperature and controller setpoint. This direct current

control signal is sent to a Phaser power controller manufactured by R-I

Controls of Minneapolis, Minnesota. The output of the Phaser is an a-c

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64

voltage to the strip heater, the RMS value of which is varied between

zero and maximum line value in direct proportion to the magnitude of

the direct current control signal.

The reactor pressure is controlled by a Grove back

pressure regulator placed at the reactor outlet. Reactor pressure drop

at the maximum gas rate (20 ft3 /hr) was about 5 psi.

e. Product Recovery System

The reactor effluent, after leaving the back pressure

regulator, passes through an ice-water bath condenser. Those components

not condensed pass through a water saturator and a wet-test meter for

measurement. The condensed liquid product is drawn through a valve on

the bottom of the condenser. Due to the large amount of diluent hydro

gen and volatility of the feed and products, liquid product recovery

was not required for most of the experiments in this study.

2. Analytical System

Product analyses were made with an F & M Model 810R dual

column gas chromatograph. A ten-foot column of 10% silicone rubber

(SE-30) on 90% white chromosorb (80-100 mesh) was used for separation

of the hydrocarbon components. Integration of the peak areas was per

formed by an Infotronics Model CRS-110 Digital Integrator and a disk

integrator. A detailed discussion of this system is given in Appendix

C. Materials

1. Gases

High pressure cylinders of electrolytic hydrogen (99.95%) and

prepurified nitrogen (99.99%) were used. Both gases passed through a

bed of platinum on alumina particles, a bed of Linde 3A molecular sieve

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65

and a bed of indicating Drierite. The platinum on alumina is used to

combine with any traces of oxygen which might be present and the

Drierite is used to remove any water present.

2. Liquids

All liquid feeds were Phillips pure grade (99 mole %). These

feeds were stored over Linde 13x molecular sieve to remove any traces

of water which might be present.

3. Catalysts

Pt-AlgOg -mordenite and Pt-Al2 03 catalysts were used in this

research. These catalysts were prepared at the Esso Research Labora

tories in Baton Rouge, Louisiana. The exact preparation steps are

given in Chapter IV.

D. Experimental Procedure

1. Catalyst Activation

Solid catalysts are normally activated by heating. This heat

ing drives off any physically sorbed water and any ammonia which may be

present in the mordenite. This ammonia which is present in the morden

ite catalyst is a result of the sodium exchange step in the preparation

and is discussed in detail in Chapter IV. The catalyst activation

procedure is as follows:

a. Catalyst samples of approximately 15 cut3 are put in Vycor

glass "boats" and placed in a Pyrex cylinder inside a tube furnace.

b. A continuous stream of purified air is passed over the

catalysts during activation. This air is purified by removing any

traces of hydrocarbons or water which might be present. The trace

hydrocarbons are removed by passing the air stream through a tube in a

muffle furnace which contains a copper oxide catalyst at IOOQPf . This

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66

converts the trace hydrocarbons to carbon dioxide and water. The

effluent air from the hydrocarbon removal step is then passed through

beds of indicating Drierite and Linde 13x molecular sieve. This step

removes any water produced by the hydrocarbon removal step plus any

water initially present.

c. The temperature of the furnace containing the catalyst

samples is increased to 350PF at a rate of approximately lOO^F/hour.

This temperature is maintained for about 16 hours. This low temperature

is used to prevent rupture of the catalyst crystal structure when the

bulk of the water is driven off.

d. The furnace temperature is increased at a rate of 10CPF/

hour to the final activation temperature.

e. The final activation temperature of IOOCPf is maintained

for three hours.

f. The furnace temperature is then reduced to a temperature

of about 450PF at a rate of 20CPF/hour.

g. The catalyst, samples are removed from the furnace and

transferred to stoppered glass vials.

h. The stoppered glass vials are cooled to room temperature,

weighed to the nearest milligram, and stored in a dessicator filled

with indicating Drierite.

2. Hydrocarbon Reaction Material Balance

The procedure for making an experimental run is as follows:

a. After cleaning the catalyst space in the reactor, the

reactor is inverted and the catalyst sample is transferred to the

reactor from the stoppered glass vial. A one inch plug of glass wool

is inserted behind the catalyst to act as a support. The reactor is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67

then sealed with the temperature compensating coupling using a new

0 -ring.

b. The reactor is purged of air using prepurified nitrogen

and pressure tested with hydrogen at 500 psig and room temperature.

c. The reactor is then placed in the fluidized heat bath,

connected to the reaction system and again purged with nitrogen.

d. Hydrogen flow over the catalyst is started and maintained

until the catalyst has been at 600-700°F for about one hour. This

procedure is used to reduce the platinum to the metallic form.

e. The reactor is again pressure-trested at 500 psig and

system temperature.

f. The system pressure and temperature are set on the back

pressure regulator and the temperature controller.

g. The approximate hydrogen flow rate is set by adjustment

of the flow controller. The exact gas rate is measured by timing the

wet-test meter through several revolutions.

h. The Ruska pump is charged with liquid hydrocarbon and the

rate is set by selecting the proper gear arrangement.

i. The experiment is started by opening the pump outlet valve,

turning the pump on and starting a timer.

j. During the initial line-out period of about 90 minutes, .

small adjustments are made in the set point of the temperature control

ler to bring the reactor temperature to the desired value.

k. When the reactor temperature is at steady state, gas

samples are taken at intervals of approximately 15 minutes.

1. When the gas composition reaches steady state, a material

balance is taken.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. m. A material balance consists of a timed period (usually

30 minutes) during which the gas and liquid rates are measured and

products analyzed.

n. At the completion of a material balance, the system is

changed to the conditions for the next balance or the gas and liquid

flows are discontinued, the system depressured, and the reactor

removed.

3. Material Balance Calculations

The product composition, reactor temperature and pressure,

inlet and outlet flow rates, and weight of catalyst charged were used

to calculate the results from a material balance run. All of these

calculations were made by the computer program listed in Appendix E.

Sample calculations for a typical material balance are described in

Appendix C.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER IV

CATALYST PREPARATION AND ANALYSES

A. Introduction

The majority of current commercial catalytic reforming processes use

bifunctional platinum-alumina-halogen catalysts. The use of a crystalline

zeolite as the isomerization promoting component of a platinum-alumina

reforming catalyst presents interesting possibilities but has received

little attention in the open literature. Apart from possible benefits

gained in activity and thermal stability, zeolite promoted reforming

catalysts would not require the presence of a halogen. The absence of a

halogen is an important advantage over prior-art catalysts since corrosion

and halogen level maintenance problems are avoided. (2) A recent U.S. patent issued to the Standard Oil Company of

Indiana claims an improved naphtha reforming process by contacting a

combined-nitrogen-containing naphtha with a halogen-free platinum-alumina-

H-mordenite catalyst. Experimental results shown in the patent example

indicate that this catalyst is highly active and selective for naphtha

reforming if a small amount of combined nitrogen is added to the feed.

These results are contrary to teachings of the prior-art since nitrogen

usually acts as a poison to platinum reforming catalysts.^ Although

the function of the nitrogen is not stated in the patent, it presumably

acts to suppress hydrocracking. No claims or statements are made about

the isomerization-promoting abilities of the synthetic mordenite component.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70

The primary objectives of this research are to study and model the

dehydrogenation and isomerization properties of a platinum-alumina-

zeolite catalyst. The halogen-free platinum-alumina-mordenite described (2) in the Indiana patent ' fits the catalyst requirements for this work and

was the primary catalyst type selected and prepared for this investigation.

In addition, a halogen free platinum-alumina was also prepared. This

chapter discusses the preparation and pertinent analyses of these two

catalysts.

B. Catalyst Preparation

The intended compositions of the two catalysts selected for this

investigation were:

a. 0.6% Pt on halogen-free 95% Als 03 -5% hydrogen

mordenite

b. 0.6% Pt on halogen-free ALjOa

The concentrations of platinum, alumina, and mordenite in these catalysts

were selected to give values that fell approximately midway in the compo- (2) sition ranges claimed by the Standard Oil Company of Indiana patent.

These catalysts were prepared at the Esso Research Laboratories, Baton

Rouge, Louisiana. A description of these preparations is given in the

following discussion.

The alumina-base used in the preparation of the halogen-free plati-

num-alumina-mordenite catalyst was a commercially available y-alumina

made in the form of extruded particles by the American Cyanamid Company.

Before starting the catalyst preparation, the water content of the

alumina extrudate was estimated by the weight loss of a small portion

during calcination. An estimation of the water content is necessary so

that the final composition of the catalysts will be correct on a dry

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71

basis since the uncalcined alumina is used in the catalyst preparation.

The uncalcined alumina extrudate was ground to a particle size smaller

than 0.35 mm and suspended in an aqueous solution of acetic acid. The

resulting solution was heated overnight, and the aluminum monoacetate

sol was formed by esterification of the alumina. Norton synthetic

hydrogen-mordenite in the form of 1-5 micron crystallites was slurried

with water to form a paste and added to the alumina sol. Simultaneously,

an aqueous solution of platinum-diamino-dinitrite [PtCNl^ )3 (N08 );j ] was

stoichiometrically added to the alumina sol with the mordenite paste and

total solution well mixed. The resulting platinum-alumina-H-mordenite

gel was washed with water to remove unabsorbed reactants and then dried

to form a hard glasslike material. This dried gel was crushed and sized

to form the final catalyst ready for activation.

The preparation of the halogen-free platinum-alumina catalyst is

similar to that of the platinum-alumina-mordenite catalyst except for

non-inclusion of the hydrogen mordenite component. Uncalcined Cyanamid

y-A^Oa was converted to the sol by addition of aqueous acetic acid and

heating. Platinum was incorporated by stoichiometric addition of

platinum-diamino-dinitrite to the alumina sol with subsequent gelation

using excess ammonium hydroxide. The catalyst preparation was completed

by washing, drying, and sizing the platinum-alumina gel.

C . Analytical Measurements

Several analytical measurements were made on the finished platinum-

alumina-mordenite and platinum-alumina catalysts. Where pertinent,

analytical measurements for platinum, sodium, alumina, silica, surface

area, and ammonia adsorption were performed on these two catalysts. In

addition, certain analyses were also performed on the pure hydrogen

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72

mordenite and fused silica-alumina used as a catalyst diluent. These

measurements were performed at the Esso Research Laboratories in Baton

Rouge, Louisiana. A brief description of the analytical method used for

each test is given in Table 4.

The values of the catalyst analytical measurements for the calcined

catalysts are shown in Table 5. The platinum concentrations of the

Pt-ALgOg-mordenite and Pt-AlgOg were 0.15 and 0.18 wt. %. These platinum

concentrations are less than the intended value of 0.6 wt. % and indicate

that either the Pt(NI^)2 (N02)s was not completely absorbed by the alumina

sol or that the water content of the uncalcined alumina was underesti

mated. These platinum concentrations are sufficient, however, since the (2) "Indiana" Patent claims that concentrations as low as 0.01 wt. % are

effective. The intended concentration of mordenite in the Pt-ALgOg-

mordenite catalyst was 5 wt. 7®. The actual mordenite concentration can

be approximately computed by using the silica concentration in the pure

mordenite and finished Pt-ALg03 -mordenite. The actual concentration of

mordenite in the Pt-ALg03 -mordenite was computed to be 3.1 wt. %. As

with the platinum, the low mordenite concentration in the calcined

catalyst probably indicates that water content of the alumina used in

the preparation was underestimated. Since the patent claims mordenite

concentrations as low as 0 .1% are effective, this mordenite concentra

tion was deemed sufficient. The Langmuir surface area of the Pt-AlgOg-

mordenite was 489 n?/gm while that of the Pt-ALgOg was 353 n?/gm.

Although H-mordenite has a higher surface area than ordinary alumina gel,

the quantity of H-mordenite is so small (~3%) that the observed increase

in surface area must be due to other factors, e.g., in the peptization

of Cyanamid y-ALgOg with acetic acid. The Langmuir surface area' of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 73

Table 4. Description of Catalyst Analytical Tests.

Quantity Measured Description of Analytical Test

Platinum Content Platinum extracted with perchloric acid to form chloroplatinic acid which is reduced to chloroplatinous acid with stannous chloride. Extraction with amyl acetate followed by colorimetric analysis.

Sodium Content Acid extraction of sodium followed by flame photometry.

Silica Content Acid extraction followed by reaction with hydrofluoric acid. Silica is determined from the evolution of hydrofluosilicic acid.

Alumina Content Acid extraction followed by a titrimetric determination with potassium fluoride.

Surface Area Adsorption of nitrogen at liquid nitrogen temperatures. Surface area computed via Langmuir adsorption isotherm.

Ammonia Adsorption Sample is evacuated and allowed to reach a constant level of nitrogen adsorption at 200PF. Sample is heated to IOOCPf under an ammonia atmosphere and allowed to equilibrate. Ammonia is removed and samples are cooled to 200PF under flowing nitrogen and allowed to equili brate. Chemisorbed ammonia is the loss or gain of weight at 20CPF.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74

Table 5. Catalyst Analytical Measurements.

Sample Description Pt-ALjOg - Fused Analytical Test H-Mord. Pt-AL,0, H-Mordenite SiOo -AL,0q

Pt, wt. % 0.15 0.18

Na, wt. % 0.55 0.43 0.68

AlgOg , w t . % 94.0 95.6 11.9 64.7

Si02 , wt. % 2.7 86.5 34.9

Surface Area, n?/gm 489 353 0

NRj Chemisorption, 0.0063 0.0046 gm NHj/gm catalyst

Samples were submitted for analysis after calcining.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75

fused SiOg-ALgOs was so low that no measurable area could be detected.

This material is used as a catalyst diluent in this investigation and

was found to have no catalytic activity.

The capabilities of the Pt-A^Og-mordenite and Pt-AlgOg catalysts

to chemisorb ammonia were measured by the procedure outlined in Table 4.

This ability to chemisorb ammonia should be a measure of the acidity of

the catalyst surface or its ability to promote isomerization reactions.

The measured chemisorption value for the Pt-AlgOg catalyst was

0.0046 gm NHj /gm catalyst while that for the Pt-AlgOg-mordenite was

0.0063 gm NHj/gm catalyst. The lower ammonia chemisorption value

obtained for the Pt-AlgOg catalyst indicates that the halogen-free

alumina is weakly acidic. The 38% increase in ammonia chemisorption

capacity by the Pt-AlgOg -mordenite catalyst reflects the presence of

the 3.1 wt. % mordenite. By using the actual mordenite concentration

in the Pt-A^Og-mordenite and the chemisorption capacities of the two

catalysts, the mordenite was computed to be approximately twelve times

more acidic than the halogen-free alumina. This indicates that the Pt-

AlgOg-mordenite catalyst should have a much greater isomerization

promoting ability and this is confirmed by experimental results discussed

in Chapter VI.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. L IS T OF REFERENCES - CHAPTER IV

1. Ciapetta, F. G., Dobres, R. M . , and Baker, R. W . , "Catalytic Reforming of Pure Hydrocarbons and Petroleum Naphthas," Catalysis, 6_, 495, Reinhold Publishing Corp., New York, New York, 1958.

2. U. S. Patent 3,376,214.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER V

A KINETIC MODEL FOR THE SIMULTANEOUS

DEHYDROGENATION AND ISOMERIZATION OF CYCLOHEXANE

A. Introduction

The quantitative study of heterogeneous catalysis usually involves

at some point, a development of a mathematical model that represents

the reaction rate data. This model can be empirical or purport to

represent the actual reaction mechanism. Powerful arguments for and / 1g\ against each philosophy have been discussed by Levenspiel. . Purely

empirical approaches are usually used where design is the main consider

ation and extrapolation over wide ranges of conditions is not required.

The mechanistic approach, however, is the only alternative when a

better understanding of how catalysts promote specific reactions is

required.

One of the objectives of this investigation is to mathematically

model the solid catalyzed simultaneous dehydrogenation and isomerization

of cyclohexane

in an integral reactor using a simplified reaction mechanism. The

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 78

development of the mathematical model for this system is discussed in.

this chapter.

A large proportion of catalytically promoted reactions reported in

the literature are of the single reaction unimolecular type. In an

integral reactor the forward rate constant for a unimolecular reaction

can be expressed as an explicit function of all the other variables.

The determination of the rate constants from the composition data for

complex reaction systems however, is not a trivial problem. Complex

reactions carried out in integral reactors usually require the solution

of simultaneous differential equations either analytically or numeri

cally with subsequent application of iterative techniques to determine

the rate constants.

Exact integration of the simultaneous differential equations for

the dehydrogenation and isomerization of cyclohexane resulted in a set

of non-linear equations. These equations could not be solved explicitly

for the two forward rate constants. Application of optimization tech

niques to solve for these rate constants is also discussed in this

chapter.

B . Cyclohexane-Benzene and Cyclohexane-Methylcyclopentane Equilibrium

The determination of reaction rate constants in reversible reactions .

requires a knowledge of the equilibrium constants involved. In general,

reaction equilibrium constants can be determined from free energy data

or from experimental data reported in the literature. Both methods

can give erroneous values for the equilibrium constant depending on

the application. Use of free energy data to determine equilibrium

constants can be subject to question when the free energy change or heat

of reaction is small. This results in an exponentiation of a small

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79

difference between two large numbers and is subject to large percentage

errors. Use of actual experimental data for equilibrium constants can

result in error if the approach to equilibrium is not sufficient or if

side reactions are not suppressed.

The isomerization of cyclohexane to methylcyclopentane is a rela

tively slow reaction and is accompanied by a small free energy change.

The experimental equilibrium constants for this reaction reported by

H o p p e r ^ ^ and o t h ers^’^ ’^ ’^ ’^^ were 10-20% below those calculated (23) from API-44 free energy data in the 100 to 600^F temperature range.

The data of Ciapetta,^"^ however, were almost coincidental with the free

energy predicted values at 700PF; and the general trend of all the

experimental data indicated convergence with API-44 values at tempera

tures above 70CP'F. Since the temperature range of interest for this (23) investigation was 700-93CPF, API-44 free energy data' were used to

calculate the isomerization of cyclohexane. The linear equation repre

senting the isomerization equilibrium constant is

In Ke = -2875/T(°R) + 4.430

The dehydrogenation of cyclohexane to benzene is a very fast

reaction and is accompanied by a large free energy change. The liter

ature contained very few experimental equilibrium determinations for

the dehydrogenation of cyclohexane to benzene or the hydrogenation of (3) (33) benzene to cyclohexane. Burrows' ' and Zharkova reported equilibrium

data for the hydrogenation of benzene to cyclohexane in the temperature

range 450-590PF. N o data, however, could be found in the temperature

region of interest, 700-93CPF. This is quite understandable since the

magnitude of the dehydrogenation equilibrium constant in the region of

interest is 104 -107 and compositions would be very difficult to measure.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80

(23) For these reasons the API-44 free energy data' 7 were used to calculate

the equilibrium constant for the dehydrogenation of cyclohexane. The

linear equation representing the dehydrogenation equilibrium constant is

In Ka = -46350/T(°R) +4 6 . 5 2

Confidence in the equilibrium constants calculated from the free

energy data was strengthened by the results of an experimental equilib

rium run. Cyclohexane was reacted over the Pt-ALjOa -mordenite catalyst

at 775°F, 85 psia and a long contact time to insure a close approach to

equilibrium. The experimental and. calculated equilibrium compositions

for the simultaneous dehydrogenation and isomerization are shown below.

Product Mole Fraction free)_____

Run Cyclohexane Methylcyclopentane Benzene

26A* 0.033 0.112 0.855

Calculated 0.020 0.140 0.840 Equilibrium

The calculated equilibrium composition was determined by assuming

K = Kp and solving the equilibria simultaneously. The agreement between

the experimental and calculated equilibrium compositions is good. The

small discrepancy in the methylcyclopentane composition is probably due

to insufficx. nt contact time ‘since it was shown previously that the free

energy and experimental equilibria become coincident above 70CPF.

C. Reactor Contacting

The calculation of reaction rate constants in a flow system requires (2) a knowledge of the type of flow pattern involved. Bryant 7 made a

study of the flow pattern in a duplicate of the reactor used in this

work. Tracer tests methods were used and the results of this study

Detailed run data are given in Appendix B.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81

Indicated that plug flow essentially existed for Peclet numbers above

60. The Peclet number is a measure of the directed motion of the bulk

stream to the random motion in the diffusive eddies and describes the

mixing behavior of a fluid system. Correspondingly, a Peclet number

of zero represents perfect mixing (i.e., an ideal backmix reactor) while

a Peclet number of infinity represents the situation of ideal plug flow.

Some investigators^^ used the "reactor dispersion number" as an index

of reactor, contacting which is simply the reciprocal of the Peclet

number. (2) The contacting data obtained by Bryant' are in good agreement with

a correlation published by Levenspiel and Bischoff^^ which relates the

"mixing intensity" to the particle Reynold's number. This correlation

was used to estimate the Peclet number for the reactor used in this

investigation. The conditions used in the calculation represent the

average flow conditions for the majority of the work. This estimation

is shown below.

Data

Catalyst Porosity, € = 0 . 7 (estimated)

Average Particle Diameter, dp = 0.000589 ft

Reactor Cross Sectional Area, A = 0.001625 ft

Catalyst Bed Length, L = 0.326 ft

Cyclohexane Flow Rate = 0.0700 lb/hr

Hydrogen Flow Rate = 0.0336 lb/hr

Total Mass Flow Rate, W = 0.1036 lb/hr

Viscosity of Mixture at 775°F and 85 psia, p, = 0.0358 lb/ft-hr

(estimated)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82

Calculation of Peclet No. dpW Particle Reynolds Number; Rep = = 1.01

"Mixing Intensity" Estimated from Correlation:

= 2.0 at Rep = 1.01 Pe = 194 Pe dp p

The estimated value of 194 for the Peclet number indicates that the

experiments in this investigation were performed in the plug flow

region.

D. Derivation of Kinetic Equations

The vast majority of chemical reactions are believed to proceed

through reactive intermediates which are present in very small concen

trations.^ In solid catalyzed chemical reactions these intermediates

have been postulated to be chemically adsorbed species at the surface of

the catalyst. The spots or centers of the catalyst which have the

ability to form chemically reactive intermediates are called active (29) centers,- a term coined by Taylor in 1925.

The active center concept has dominated thinking in heterogeneous

catalysis during the past forty years. Though it may never be possible

to see active centers directly, they can be developed in certain situa

tions such that they are visible under electron microscopy.^This

viewing of developed active centers by an electron microscope earned

Ziegler and Natta the 1963 Nobel Prize in Chemistry for the polymeri

zation of propylene on a titanium trichloride-aluminum alkyl catalyst.

The combination of the active center concept and adsorption theory

in the case of fluid reactions catalyzed by solids has been summarized (13) by Hougen and Watson and is generally referred to as Langmuir-

Hinshelwood kinetics. The application of Langmuir-Hinshelwood kinetics

to solid catalyzed chemical reactions usually results in a large number

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83

of arbitrary parameters for each rate controlling mechanism. The large

number of parameters makes it difficult to reach absolute conclusions

with regard to the mechanism or rate controlling step on the basis of

best fit alone.Investigators still disagree^’^^ or are (14) undecided about the exact mechanism of ortho-parahydrogen conversion

and it may never be possible to determine the correct mechanism from

kinetic measurements alone.

In spite of the difficulty of deducing exact mechanisms from

analysis of kinetic data by Langmuir-Hinshelwood kinetics, it is

believed that it still represents the most powerful tool available for

modeling solid catalyzed fluid reactions.

1. Rate Limiting Processes in Heterogeneous Catalysis

The quantitative relations involved in the catalysis of fluid (13) reactions by solid catalysts are summarized by Hougen and Watson.

It is postulated that the reaction occurs on the surface of the catalyst

which involves molecules chemisorbed on the active centers. The catalyst

increases the rate of reaction by adsorbing the reactants in a form such

that the activation energy is reduced below that required for the

uncatalyzed reaction.

The catalytic conversion of a reactant in the bulk stream to

a product in the bulk stream can be visualized as taking place by the

following stepwise process:

a. Mass transfer of the reactants from the main fluid phase

to the exterior surface of the catalyst.

b. Diffusion of the reactants into the pores of the catalyst.

c. Activated adsorption of the reactants on the active sites

of the catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. d. Surface reaction to form products.

e. Desorption of the products from the active sites of the

catalyst.

f. Diffusion of the products out the pores of the catalyst.

g. Mass transfer of the products from the exterior surface

of the catalyst to the main fluid phase.

The rate at which each of these steps occur influences the

overall rate and the slowest step is termed the "rate controlling step.

Although all the steps are dependent on the concentration gradients of

the system they additionally depend on several widely different factors

The mass transfer steps (a. and g.) depend on the flow characteristics

of the bulk stream such as mass velocity and diffusivity. The pore

diffusion steps (b. and f.) depend on the porosity of the catalyst,

geometry of the pores, and the rate of reaction at the interface. The

chemical steps (c., d. and e.) depend on the nature of the catalyst

surface and the activation energies involved. Although the mass trans

fer steps and chemical steps occur in series and can be treated

separately, the pore diffusion steps are not related in any simple way

to the others.

The response of each of the steps to changes in system

parameters varies widely. This fact can be used to adjust the system

parameters such that step(s) under consideration are rate controlling

and all the other steps are at equilibrium. The mass transfer steps

can be eliminated as rate controlling by increasing the velocity of the

bulk stream to a point where there is no change in reactant conversion

with a further increase in velocity. The pore diffusion step is gener

ally eliminated as rate controlling by two methods. The first is to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85

decrease the catalyst particle size to a point where a further decrease

causes no change in reaction rate. The second is to decrease the

reaction temperature to a value where the surface chemical steps become

rate controlling.

Since one of the objectives of this investigation is to study

and model the surface kinetics, mass transfer and macropore diffusion

were eliminated as rate controlling steps by adjustment of the system

parameters. The procedure followed is to first develop the rate equations

assuming mass transfer and macropore diffusion resistances are negligible.

Then using the kinetic equations it is shown that changes in the signif

icant independent variables (i.e., mass velocity and catalyst particle

size) cause no change in the rate constants. The experimental data

confirming that mass transfer is not rate controlling and the macropore

diffusion resistance is negligible are shown in Chapter VI.

2. Derivation of the Rate Equations

Once it has been shown that mass transfer and macropore

diffusion contribute negligible resistance to the rate of reaction, the

chemical steps of adsorption/desorption and surface reaction must be

examined. The usual procedure assuming Langmuir-Hinshelwood kinetics is:

a. Postulate a mechanism as a series of stepwise reactions.

b. Derive a reversible rate equation for each step.

c. Assume one of the steps is rate controlling and all the

rest are at equilibrium.

d. Using the results of a., b., and c. derive an equation

for the overall rate.

e. Apply the rate equation to the experimental data and

calculate the arbitrary parameters (i.e., rate and adsorption constants).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86

£. Repeat steps a. through e. for each possible mechanism.

g. Discard those mechanisms with physically unrealizable

constants.

h. Of the remaining set choose the mechanism with the best

fit.

As mentioned before, this procedure has received criticism not for its

ability to correlate kinetic data but for reaching absolute conclusions

about the mechanism from best fit alone.

The complexity of the situation increases rapidly with the (13) complexity of the reaction considered. Hougen and Watson in their

analysis of the hydrogenation of mixed iso-octenes considered seventeen /£\ possible mechanisms. Cochrane in the analysis of the steam-methane

reforming reactions eliminated all but six out of over one hundred

possible mechanisms. Corrigan^^ has pessimistically stated that there

is no instance in the literature where a complete kinetic analysis of a

complex reaction has been made. It should also be noted that the kinetic

studies mentioned in this paragraph used differential reactors to obtain

the reaction rate data. The use of an integral reactor multiplies the

complexity of the problem .immeasurably. To find the rate in an integral

reactor the data must be graphically or numerically differentiated or

the rate equation must be integrated. Graphical or numerical differ

entiation is very hard to do accurately and exact integration is impos

sible with many assumed mechanisms.

Therefore, faced with the problem of kinetic analysis of a

complex reaction in an integral reactor a course has been selected which

allows the reaction to be modeled within the framework of Langmuir-

Hinshelwood kinetics. It is assumed that the simultaneous

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87

dehydrogenation and isomerization of cyclohexane can be represented by

a simplified reversible reaction model which is

where kj and kg are the forward reaction rate constants and k^ and kj

are the reverse reaction rate constants.

This model assumes that:

a. The surface dehydrogenation of cyclohexane to benzene and

hydrogen is the rate controlling step in the dehydrogenation reaction.

b. The surface isomerization of cyclohexane to methylcyclo-

pentane is the rate controlling step in the isomerization reaction.

c. The rate constants are independent of composition.

d. Hydrocracking of cyclohexane to form lower molecular

weight products is negligible.

The examination and justification of these assumptions here

would pre-empt the purpose of Chapter VI. This discussion has been

deferred to Chapter VI where they can be examined in light of the

experimental data.

To derive the rate equations, consider an elementary section

of the reactor catalyst bed containing an amount of catalyst AWC .

Flow AW,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88

The reaction zone is of constant cross section and it is assumed that

the catalyst is packed uniformly throughout the reaction zone. At steady

state, assuming constant temperature and negligible pressure drop, the

material balances on benzene and methylcyclopentane are

Nb (Wc) + rBAWc - Nb (Wc + AWC) = 0 (1)

and

H,(WC) + r„AWc - H,(WC + AWC) = 0 (2)

where ■ Ng = gm moles benzene/min,

= gm moles methylcyclopentane/min,

W c = gm of catalyst,

rB = gm moles benzene produced/min-gm catalyst,

rH = gm moles methylcyclopentane produced/min-gm catalyst.

If equations (1) and (2) are rearranged and divided by AWC , the result is

Nb (Wc + AWC) - Nb (Wc) AW, (3)

and

H,(WC + AWC) - H,(WC) • tH = AWC • W

Taking the limit of equations (3) and (4) as AWC-* 0 forms the definitions

of the first derivatives, and the equations become

r= - 3 5 7 (5)

and

r« - % ■ < «

In terms of simplified reaction model given on page 87, the rates of

formation of benzene and methylcyclopentane may be written as

rs ° ^Pc - WftPka

and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89

rB “ H b Pc " . <8>

where pc = partial pressure of cyclohexane, atm,

pB = partial pressure of benzene, atm,

p^, = partial pressure of methylcyclopentane, atm,

pHg = partial pressure of hydrogen, atm,

kj, ^2 = 8m moles/gm catalyst-atra-min.

Eliminating rB and rH between equations (5), (6), (7) and (8) yields

g k - - k^pB PHg (9)

and

= kgPc - k2 pH . (10)

At the conditions of this study (~800PF, 85 psia) the fugacity coeffi

cients of all the components are very close to unity. It seems reason

able, therefore, to assume Dalton's law is applicable which is

Pi » nyi (11)

where pt = partial pressure of component i, atm,

tt = total pressure, atm,

yt = mole fraction of component i.

Substituting equation (11) into equations (9) and (10) for each compo

nent yields

= kitryc - kJ'(TTyB )(TTyH2)3 (12)

and

= ksnyc - kg'nyn . (13)

Equations (12) and (13) cannot be integrated in their present

form unless some provision is made to counteract the severe volume

change caused by the dehydrogenation reaction. To make these equations

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90

integrable and simplify the analysis of the experimental data, a very

high hydrogen diluent was used. At the average conditions of this study,

the dilution ratio was 20 moles Hg/mole cyclohexane feed. This diluent

ratio kept the system volume change under 10% for the highest conversion

levels with the average volume change being around 5%. These percent

ages are within the accuracy of the experimental data taken and the

assumption of constant system volume is considered reasonable. Another

consequence of the high hydrogen diluent is that the hydrogen concentra

tion can be considered uniform with respect to reactor length. Mathe

matically, the assumptions of constant volume (or moles) and constant

hydrogen concentration are

Nt = NTy (14)

with

dNj = NTdyi (15)

and

y»a * m (l6)

where NT - total gm-moles gas/min, a constant,

R = gm-moles Hj diluent/mole cyclohexane feed, a constant.

Substituting equations (15) and (16) into(12) and (13) gives

= kiTTyc - k^(nyB ) (R- ^ (17)

and

c = " kaTT3fo ‘ (18)

For ease in interpreting the experimental results the mole fractions are

converted to a hydrogen-free basis by using

y X c = yiNT (19)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91

where y{ - mole fraction of component i on a hydrogen-free basis,

Nhc *» total moles hydrocarbons/min.

Substituting equation (19) into equations (17) and (18) and noting that

the total moles of hydrocarbon remain constant gives, after simplification

d y B _ -i,f ^ye /_5_\3 j /on\ dwT yc ^ 157 R + 1

and

= £aZ yp' - Id HZ”. dW,~ic T n t yc ^ . nt . (21)

Defining a space time 0 as

6n = -r- , (22) F

which can be transformed to

dWc = Fd0 (23)

since F, the mass feed rate of cyclohexane to the reactor, is a constant.

Equations (20) and (21) can be put into final form by substituting

equation (23) and noting that

yc “ 1 - yB - yM • (24)

Performing these substitutions gives the final differential form of the

simultaneous rate equations which are

dy£ _ I^ttF , , kfrF / _ R _ v 3 _3 v / (25) d0 N t ( ~ N t (R + l'

and

dyH _ ^ttF , , Icj'ttF , do u T (1 - yB - yH) - — y» • (26)

3. Simultaneous Solution of the Differential Rate Equations

Once the rate equations have been obtained they can be used

to calculate values for the rate constants. Himmelblau et al^ ^ have

reviewed the methods available for estimati.ng rate constants for complex

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. reactions from experimental data. These methods generally fall into the

following four primary classes:

a. Analytical (exact or approximate) integration of the set

of differential equations with subsequent application of iterative non

linear least squares techniques.

b. Differentiation of the experimental data with subsequent

application of linear least squares techniques.

c. Numerical integration of the differential equations

using the experimental data followed by iterative non-linear least

squares techniques.

d. Matching of the experimental data on an analog computer

by a trial and error search.

All of the methods listed above and their variations have

advantages and disadvantages depending on the application. Category

(a) is used frequently where a simple mechanism is applicable and is the

general method used in this investigation since the simultaneous

differential equations can be solved exactly.

The equations that must be solved simultaneously are

(25)

and

(26)

which were derived in the previous section. The boundary conditions are

yl - 0 and yH' = 0 at 9 = 0 .

Jungers e t a l ^ ^ have solved the simultaneous differential

rate equations resulting from the general system

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93

B sc •which are

dx — = kL (a - x - y) - k.j(b + x) (27)

and

^ = kTT (a - x - y) - k.n (c + y) (28) dt *1 1 '

where t = time,

ki = reaction rate constants,

a, b, c = initial concentrations of components A, B, and C,

x = concentration of component B,

y = concentration of component C.

If the substitutions

t = 0 (29)

a = 1 (30)

b = c = 0 (31)

x = yB' (32)

(33) y = Yh

*1 Nt (34)

k = ( . . R . ) 3 n 3 (35) k-l Nt R + 1

k - v * (36) *i “ i i T

kg'nF k. (37) 11 = ~

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94

are made in equations (25) and (26) the resulting equations are identical

to equations (27) and (28) which are Jungers starting equations. Since

the two sets of equations are equivalent, Jungers solution was used and

then transformed to the nomenclature of this problem and checked by

differentiation.

Equations (27) and (28) are solved simultaneously by first

taking the derivative of equation (27) with respect to t which yields

. dF = ' kidT “ k-idF " kidT • (38)

Substituting equation (28) into equation (38) for gives

= -(kx+ k-i)f^ - k-l[kji(a - x) - k_li;c - (kn + k_n )y] . (39)

Equation (39) can be put in the final form for solution by solving

equation (27) for y, which upon substitution in (39) yields

dF + (ki+ k-i+ kn+ k-n>dt + (kik-n+ k-ikn+ k-ik-u)x

- kIk.I 1 (c + a) + k_jb (kjj+ k_lx) = 0 . (40)

By an analogous procedure equation (28) can be transformed to give

+ (kL+ k.j+ kn + k _IX) ~ 2 + (kI£k_j+ k£Ikx+ k.IIk_I )y

~klIk_X (b + a) + k_j£c (kj+ k_£) = 0 . (41)

Equations (40) and (41) are ordinary second order differential equations

with constant coefficients and have the general solutions

X - P I er't + Qler= t + ^ (42)

and

y = Pn eri,: + 0 1 1 ^ 3 C +*Et. (43)

where Pj, Pxx, Q j , QIX , Yj /0 and Yj-j /P are functions of the constant

terms in the coefficients.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95

Equations (42) and (43) are also solutions to equations (25)

and (26), the rate equations for this investigation. The final solution

of equations (25) and (26) is obtained by applying the transformation

equations (29) (37) to Jungers solution. In addition, the reverse

reaction rate constants are eliminated in the final solution by intro

ducing the definitions of the equilibrium constants

*» - sj- ■ «*> •

and

Kb = rr (45)

K - £ <«>

and

Vi - • (47) k b

Therefore, the final solutions are

Ye = Bj e*1 0 + Bs e*3 9 + B3 (48)

and

y^ = Mieri 0 + Mge^ 6 + ^ (49)

where

(50)

(51)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96

r ( h ) + iaJlZ: = 2l_ , (52)

r l ~ r a

r (_ 3 s + V* B2 - .*3.+ HK° +. 3 ____ »I_ , (53) r 3 - r l

^ = KA + HKb + H *

r ( HKg , + I^ttF Nt_ t (55) rl - r2

r ( »*» >, . KaTTF Ms = k a + + H; ^ N t (56) r3 - rl

M HKfl ^ KA + HKe + H * ^ ^

H B (r~tt)3ti3 • (58)

Equations (48) and (49) give the outlet composition for the

assumedreaction model in an integral reactor as a explicit function of

the space time 0 , reaction rate constants and other process variables or

yB' = f(kl , kg, 0 , process variables) (59)

and

y» = g(^i> Hb > 9> process variables) . (60)

To test the assumed model and correlate the rate constants with temper

ature and pressure it would be convenient to solve equations (59) and

(60) simultaneously giving

kj = f'(yB', yM', 0 , process variables) (61)

and

kg s'g /(yB/» Yh > 9» process variables) . (62)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97

Due to the non-linearity of the system, this algebraic transformation is

impossible and it is necessary to use iterative techniques to solve for

the reaction rate constants.

The simultaneous solution of equations (48) and (49) for the

reaction rate constants kj and kg by iterative techniques is discussed

in the next section.

E. Solution of the Kinetic Equations for the Reaction Rate Constants

1. Solution of Non-Linear Equations

The numerical solution of simultaneous algebraic and transcen

dental equations is one of the tasks most frequently encountered in

applied mathematics. The methods available for the exact solution of

simultaneous linear equations are numerous and have been developed to a

high degree. In contrast there are no general procedures for the solu

tion of simultaneous non-linear equations.

Most methods for solving systems of non-linear equations are

iterative. These iterative methods require an initial guess to the

solution and a new point is computed by the algorithm used. The process

is continued by successively calculating points that give improved

approximations to the solution. A good example of such a procedure is

the Newton-Raphson method which is described in many texts on numerical (24 25 27 18) analysis. ’ ’ * The Newton-Raphson method and many others use

the linear terms of a Taylor series expansion to approximate the function

and require initial guesses fairly near the solution.

The problem of locating maxima or minima is very similar to

that of finding a root. It is convenient many times in the solution of

simultaneous non-linear equations to formulate the problem as a

minimization.<22,31)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98

Consider the set of p simultaneous non-linear equations

(59)

One way to solve this set of equations is to compute the function

P (60)

i=l

and search for the minimum of $. The values of the independent variables

Xjl , x^ , ..., Xp which makes $ = 0 is the solution to the set since all of

the ft must be zero. If these equations are non-linear they will have

a number of solutions and care must be taken in the selection of the

initial guesses. Frequently natural limitations often ensure that the

equations have only a single solution which is physically meaningful.

These limitations are usually taken into account by the use of constraints.

To carry out the solution by minimization, a search plan or

optimization routine is needed. Many methods of steepest descent for (31 3: multivariable problems are described in modern optimization texts. ’

The choice of the optimization technique depends on the complexity and

nature of the problem. Reasonably simple methods can be used where the

response surface is strongly unimodal but very complex ones are usually

required where saddle points or ridges are encountered.

Therefore, the solution of the non-linear kinetic equations

Vb = Bieri0 + Bge^ 0 + Rj (61)

and

yH' = M ^ l ® + K Jer2 0 + M, (62)

for the reaction rate constants kj^ and Vq can be formulated as a

minimization problem. If equations are rearranged to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99

\ e ri e + E^ 9 + Rj - yB' = Bj (63)

and

M ^ i 0 + ^ e r2 0 + M, - yH' = Rg , (64) I the problem is to

minimize: $ = I^ 3 + Rg2 (65)

subject to: kj > 0 , kg > 0

where Rj and Kg are called the residuals. The values of and kg that

make $ = 0 are the solution of the set.

The search plan used to accomplish this minimization and its

implementation is discussed in the following two sections.

2. Pattern Search

It was pointed out previously that the choice of a search

technique depends on the nature of the response surface. Usually it is

advisable to graphically examine the response surface so that the

characteristics of the surface can be matched to the advantages of the

search technique. The response surface for a typical experimental run

is shown in Figure 7 with the surface cross section shown in Figure 8 .

These curves were generated by computing $ from equation (65) for many

values of kj and kg and joining points of constant $ to form the contours.

Figures 7 and 8 clearly indicate that the response surface generated

from equation (65) is strongly unimodal with ellipsoidal like contours

and no apparent saddle points or ridges. Most optimization techniques

will work well with a surface like this and the choice is usually one

of convenience rather than necessity.

The optimization routine used in this investigation to find

the solution of equations (61) and (62) by minimization is pattern

search developed by Hooke and Jeeves.Wilde^^ indicates that the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 100

0.30

0.06 0.20

0.15

Contours are lines 0 .10, 0.05 of constant $.

c •H e ■ 4JB 0.04

4J flj u B 00

r—4 0.03 o B 0.03 B oo

0.01 0.02 0.005

0.001 0.01

0.0 0.02 0.04 0.06 0.08

kx , gm moles/gm cat-atm-min

*/c Detailed Run Data are given in Appendix B.

Figure 7. Response Surface for Run 35A.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101

0.24

0.20

0.16 kg =0.02

0.12 u o o 0.08

0.04

0.04 0.08 0.12 0.16 k j , gm moles/gm cat-atm-min

Detailed Run Data are given in Appendix B.

Figure 8 . Cross Section of the Response Surface for Run 35A.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102

pattern search strategy is very well adopted to non-linear curve fitting

problems involving minimization of a sum of squares. Another advantage

of pattern search is that the computation time is proportional to only

the first power of the number of variables. This is in contrast to the

computing time of many classical multivariable search techniques which

grows with the cube of the dimensionality.

Pattern search is based on the premise that any adjustments

of the independent variable which have been successful in the early part

of the search will be worth trying again. Although the method starts

cautiously, the steps grow with repeated success. Subsequent failure

causes the steps to grow shorter and if a change in direction is indicated,

a new pattern will be formed and the technique will start over. In the

vicinity of the optimum, the steps become very small.

Illustration of the pattern search technique can be accom- (31 32) plished by the use of a simple generalized example after Wilde. *

Consider the problem of a two variable search as shown in Figure 9

where the contours represent the criterion function. The search starts

at an arbitrary base point a± , and a step size Ai is chosen for each

independent variable. After measuring the criterion function at aj, an

observation is taken at aj + Ax • If this point is better than % > % + Ai

is called the temporary head hlx which indicates that the first pattern

is being developed and the first variable has been perturbed. It is

possible that % + Aj may not be as good as in which case % + Ai I®

discarded and aj^ - Ai is tried. If this point is better than a^ , it is

called the temporary head. Therefore, the temporary head, h ^ , is

given by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103

O — A r / O

r s / h4 0 ” % " a6 0 *’P k " 0 /

+A;

Illustration of Pattern Search for an Arbitrary Figure 9. Response surface.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104

% + Ax if y( % + A x) > y(aj)

% - Ax if y(ax - Ax) > y(ai) (66)

a^ if y(a*) > max [y(a1+ Ax), y^ - Ax)]

The next independent variable x^ is perturbed in a similar manner, this

time about the temporary head hj^ instead of the base aj. The next

temporary head h12 is determined by

h u + Ag if y(hxx'F Ag) > y(hjj)

hxj - Ag if y(hlx- Ag) > y(hn) (67)

hxl if y(hxl) > max [>(1^+ Ag) , y(hxX- Ag)]

The original base and the new base ag together form the

first pattern. Pattern search reasons that if a similar exploration were

conducted at ag the results would be the same. Therefore, the local

explorations at ag are skipped and an arrow is extended from through

3g establishing a new temporary head, hgg, for the second pattern at ag .

The new temporary head, hg0 is given by

hg0 “ "b (2flg “ 3j) = 2ag - a^ . ( 6 8 )

The subscript 20 means that a second pattern is being started but the

variables have not yet been perturbed. An exploration is conducted at

hj0 to correct the tenative pattern. The equations which control the

establishment of new temporary heads are similar to equations (66) and

(67) with the appropriate subscripts being used. A new temporary head,

hj0 , is established by extrapolating from ag through % ,

tbo e 2as - % • <69)

Repeated success in this direction causes the pattern to grow; and if

this general trend continues, the pattern will go to the left still

growing in length.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 105

Suppose, however, that for the fourth pattern none of the

explorations about the temporary head,,h4 0 , improve the criterion func

tion but that y(h40) > y(a4) . Also, suppose that none of the temporary

heads extending from hi0 are any better than ag . Then % = a6 and the

pattern is destroyed. This can mean that this is the vicinity of the

optimum or a ridge is being crossed.

Since the old pattern cannot be continued, a new one is built

using % as the base point. If local exploration about ag produces a

better point, the process is continued. If, however, a better point

cannot be found, the steps are shortened to break the deadlock. If this

reduction allows an improvement, a new, smaller pattern is begun. A l

though this pattern is small, success will cause it to grow rapidly.

The whole process is continued until a point is reached where

repeated reduction in step size yields no improvement in the criterion

function. Pattern search, being unable to advance further, calls this

point the optimum.

3. Solution of Kinetic Equations by Pattern Search

The pattern search minimization of

$ * Rj3 + Rg2 (70)

to obtain the reaction rate constants for each experimental run is a very

long and tedious numerical calculation. Iterative computations of this

kind are easily programmed for digital computer solution which gives a

degree of accuracy and consistency not obtainable manually. The pattern

search minimization routine for this investigation is included as a part

of the overall yield-material balance computer program SLAVEU which is

shown schematically in Figure 10 and listed in Appendix E. Subroutine

PATERN is a generalized pattern search variable adjustment routine and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106

SLAVEU

START

PROGRAM MAIN - SUBROUTINES MULSVR CSTGEN, FEEDX, SORTX, GASX, LIQX

Calculation of yields and material balances. Accessory calculations and parameter set-up for minimization routine. Printing of final answers

SUBROUTINE PATERN

Adjustment of independent variables for minimization.

SUBROUTINE PROC SUBROUTINE BOUNDS

Evaluation of sum Check for constraint of squares of violation. residuals.

Figure 10. Schematic Flow Diagram for Computer Program SLAVEZ

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107

was written by Dr. C. F. Moore for use in the Chemical Engineering

Department of Louisiana State University.

The calculation of the rate constants for an experimental run

requires the following data:

a. Run identification

b. Liquid feed and hydrogen diluent rates

c. . Reactor temperature and pressure

d. Catalyst weight

e. Product gas and liquid rates

f. Product analysis from gas chromatograph.

The main program loads the data and calculates the product yields,

material balance, and constant terms necessary for the minimization.

Control is passed to subroutine PATERN which directs the search for

the minimum of the objective function. During the search subroutine,

PROC evaluates §, the sum of the squares of the residuals while BOUNDS

checks for constraint violations. When subroutine PATERN finds the

minimum, control is passed back to the main program which prints the

final answers.

The pattern search routine requires an initial estimate of

both rate constants. While the routine is able to find the minimum from

almost any pair of positive numbers, computing time can be saved by

making good estimates. It was found that a good estimate of the dehydro

genation forward rate constant could be made by assuming that only the

dehydrogenation part of the reaction is significant. If only the dehy

drogenation of cyclohexane is significant then an estimate of the for

ward reaction rate constant is given by

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108

n o = - M ____ In (71) ^ E S T ttF O C K ^ 1) K* + H. ( ) ye J Ka

Equation (71) is obtained by integration of the dehydrogenation rate

equation assuming methylcyclopentane is not present. An estimate of the

isomerization forward rate constant is found by the product of (kj,)EST

and the ratio of the product mole fractions of methylcyclopentane and

benzene or

= (^i )est • (72)

Equations (71) and (72) are used to provide initial estimates for the

pattern search. The initial estimates of the reaction rate constants

were usually within 20-307, of the final values.

The possibility of multiple solutions and local minima which

do not represent solutions are always present in non-linear optimization

problems. Local minima which do not represent a solution to the set of

equations can be easily recognized by the presence of a non-zero value

for the sum of squares of the residuals. Multiple feasible solutions,

if present, can be a real problem since all of them must mathematically

be considered valid. Fortunately, neither local minima nor multiple

feasible solutions were encountered in this investigation.

The presence of multiple feasible solutions can be detected

by a variation in the final answer with the initial estimate. Each

experimental run in this investigation was optimized for seven sets of

initial estimates. These estimates ranged from one fifth to five times ♦ the estimates calculated from equations (71) and (72) . In all experi

mental runs convergence to the same solution was obtained. The set of

initial estimates and final solution for a typical experimental run is

given in Table 6 .

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109

0.007495 0.007494 0.007494 0.007494 0.007494 0.007494 0.007494 Isomerization ^ cat-atm-min ’gm . . moles gm

0.04366 0.04367 0.04366 0.04366 0.04366 0.04366 0.04366 FinalOptimized Rate Constants Dehydrogenation ^ cat-atm-min ’gm . . gmmoles

0.3355xl0_xo 0.1947x10"X1 0.9607xl0"xx of.Criterion 0.4448x10_11 Function, 4> Function, Final Value Different Initial Estimates of the

74 0.6789x10-® 75 0.3372xl0-9 97 86 117 112 No. No. of Iterations

0.03535 0.02121 0.01414 0.001414 0.002350 0.003535 81 0.1417xl0-1° 0.007071 Isomerization ^ cat-atm-min ’gm . . moles gm of Rateof Constants

* Rate Constants, Run 35A.* Detailed RunData are givenAppendix in B. 6 * . . - Convergence of the SearchRoutine for 0.2062 0.1237 . 0.01373 0.04124 0.08248 0.00825 0.02062 a a Initial Estimate Dehydrogenat ion Dehydrogenat ^ cat-atm-min ’gm . . moles gm

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF REFERENCES - CHAPTER V

1. Boudart, M . , "Four Decades of Active Centers," American Scientist, 57 (No. 1), 97 (1969).

2. Bryant, P. A., "Hydroisomerization of Normal Pentane over a Zeolite Catalyst," Ph.D. dissertation, Department of Chemical Engineering, Louisiana State University, 1966.

3. Burrows, G. H. and Lucarini, J. F., "The Equilibrium Between Benzene, Hydrogen, and Cyclohexane," Journal of the American Chemical Society, 49, 1157 (1927).

4. Chou, C. H., "Least Squares," Industrial and Engineering Chemistry, 50 (No. 5), 799 (1958).

5. Ciapetta, F. G., "Isomerization of Saturated Hydrocarbons, Cycloalkanes," Industrial and Engineering Chemistry, 45, 159 (1953).

6 . Cochrane, T. J., Jr., "A Kinetic Study of the Steam-Methane Reaction," M.S. thesis, Department of Chemical Engineering, West Virginia University, 1551.

7. Corrigan, T. E., "Chemical Engineering Fundamentals, Interpretation of Kinetic Data-II," Chemical Engineering, 62, 203 (1955).

8 . Couper, A. and Eley, D. D., "The Reversible Dissociation of Hydrogen Molecules and Para-Hydrogen Conversion," Proceedings of the Royal Society of London, A-211, 536 (1952).

9. Glazebrook, A. L. and Lovell, W. G., "The Isomerization of Cyclo hexane and Methylcyclopentane," Journal of the American Chemical Society, 61, 1717 (1939).

10. Himmelblau, D. M., Jones, C. R. and Bischoff, K. B., "Determination of Rate Constants for Complex Kinetic Models," Industrial and Engineering Chemistry Fundamentals, 6_ (No. 4) , 539 (1967) .

11. Hooke, R. and Jeeves, T. A., "Direct Search Solution of Numerical and Statistical Problems," Journal Association of Computing Machinery, 8 (No. 2), 212 (1961).

12. Hopper, J. R., "A Study of the Catalytic Hydroisomerization Reactions of N-Pentane and Cyclohexane over Structurally Modified Mordenites," Ph.D. dissertation, Department of Chemical Engineering, Louisiana State University, 1969.

110

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Ill

13. -Hougen, 0. A. and Watson, K M., Chemical Process Principles, Part III, Kinetics and Catalysis, John Wiley and Sons, New York, N. Y., 1947.

14. Hutchinson, H. L., Barrick, P. L. , and Brown, L. F., "Langmuir Kinetics and the Catalytic Para-Orthohydrogen Shift Reaction," Chemical Engineering Progress Symposium Series, 63 (No. 72), 18

15. Jungers, J. C. et al, Cine'teque Chimique Applique'e, Technip, Paris, France, 1958. ,

16. Levenspiel, 0., Chemical Reaction Engineering, John Wiley and Sons, New York, N. Y., 1962.

17. Levenspiel, 0. and Bischoff, K. B., "Patterns of Flow in Chemical Process Vessels," Advances in Chemical Engineering, 4, 95, Academic Press, New York, N. Y., 1963.

18. McCracken, D. D. and Dorn, W. S., Numerical Methods and Fortran Programming, John Wiley and Sons, New York, N. Y . , 1964.

19. Mizushima, S., Morino, Y. , and Fiyisiro, R . , Science Papers Institute Physics Chemical Research (Tokyo), 3J5, Nos. 1034-1035 (1941).

20. d'Ouville, E. L., Lien, A. P., Evering, B.L.,and Grobb, H. M . , "Rate of Isomerization of Cyclohexane," Industrial and Engineer ing Chemistry, 4 4 , 351 (1952).

21. Rideal, E. K. and Trapnell, B. M. W., "The Mechanism and Temperature Coefficient of the Para-Hydrogen Conversion," Discussions of the Faraday Society, No. 8 , 114 (1950).

22. Rosenbrock, H. H. and Storey, C., Computational Techniques for Chemical Engineers, Pergamon Press, London, 1966.

23. Rossini, F. D., e£ al, Selected Values of Physical and Thermodynamic Properties of Hydrocarbons and Related Compounds, Carnegie Press, Pittsburgh, Pa., 1953.

24. Saaty, T.L., and Bram J., Nonlinear Mathematics, McGraw-Hill, New York, N. Y., 1964.

25. Salvadori, M. G. and Baron, M. L., Numerical Methods in Engineering, Prentice-Hall, Englewood Cliffs, N. J., 1961.

26. Schuit, G. C. A., Hoog,H.,and Verheus, J., "Investigations Into the Isomerizations of Aliphatic and Alicyclic Hydrocarbons," Recueil des Traveux Chimiques de Pays-Bas, 59, 793 (1940) .

27. Stanton, R. G., Numerical Methods for Science and Engineering, Prentice-Hall, Englewood Cliffs, N. J., 1961.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 112

28. Stevenson, D. P. and Morgan, J. H., "The Isomerization of Cyclo hexane and Methylcyclopentane in the Presence of Aluminum Halides. II. Equilibrium and Side Reactions," Journal of the American Chemical Society, 70, 2773 (1948).

29. Taylor, H. S., "A Theory of the Catalytic Surface," Proceedings of the Royal Society of London, A108, 105 (1925) .

30. Weller, S., "Analysis of Kinetic Data for Heterogeneous Reactions," American Institute of Chemical Engineering Journal, 2_ (No. 1), 59 (1956).

31. Wilde, D. J., Optimum Seeking Methods, Prentice-Hall, Englewood Cliffs, N. J., 1964.

32. Wilde, D. J., and Beightler, C. S., Foundations of Optimization, Prentice-Hall, Englewood Cliffs, N. J., 1967.

33. Zharkova, V. R. and Frost, A. V., "Chemical Equilibrium, of Reactions of Hydrocarbons," Journal of General Chemistry U.S.S.R., 2, 534 (1932); Chemical Abstracts, 27, 888 (1932).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER V I

EXPERIMENTAL DATA AND ITS ANALYSIS

A. Introduction

This investigation is a study of the dehydrogenation and isomeri

zation properties of a halogen-free, bifunctional platinum-alumina-

mordenite catalyst. The importance of dehydrogenation and isomerization

in catalytic reforming plus the selective catalysis potential offered by

crystalline zeolites have provided the motivation for this research. The

majority of the experimentation has been directed toward the simultaneous

dehydrogenation and isomerization of cyclohexane over the platinum-

alumina-mordenite catalyst. These results are supplemented by a limited

study of cyclohexane dehydrogenation over a halogen-free platinum-alumina

catalyst. The dehydrocyclization and dehydroisomerization capabilities

of the platinum-alumina-mordenite catalyst are also investigated by

limited experimentation on n-heptane and methylcyclopentane. The prepar

ation and physical properties, of these two catalysts have been discussed

in Chapter IV.

The primary objectives of this research are to (1) investigate the

use of mordenite as the isomerization-promoting part of a bifunctional

reforming catalyst and (2 ) develop a mathematical model for the simulta

neous dehydrogenation and isomerization of cyclohexane from the experi

mental data on this catalyst over a wide range of conditions. The

experimental data presented in this chapter have been obtained in a

113

with permission of the copyright owner. Further reproduction prohibited without permission. 114

fixed-bed integral reactor system which is discussed in Chapter III. A

mathematical model for the dehydrogenation and isomerization of cyclo

hexane has been developed in Chapter V and applied to the experimental

data. The applicability of this model and the influence of catalyst

composition on cyclohexane reactivity are examined in this chapter. Data

from sixty-nine experimental runs are presented.

B. Pore Diffusion and Bulk Mass Transfer Limitations

1. Diffusional Effects in Experimental Catalysis

The kinetic equations derived in Chapter V for the solid

catalyzed dehydrogenation and isomerization of cyclohexane were developed

assuming that the pore diffusion and/or bulk mass transfer steps were not

rate-controlling. This is equivalent to assuming that the concentration

of the gaseous components seen by the catalytic surface is identical to

that measured in the bulk gas phase. This assumption may not be correct

for all experimental conditions due to the possible competition of the

reaction rate on the catalyst surface with the natural transport rate of

reactants and products to and from the surface. In general, either the

assumption that pore diffusion and bulk mass transfer are not rate

controlling must be proved valid experimentally or the rate equation(s)

must be corrected for any positive influence of diffusional phenomena.

In a system where a study of the surface kinetics is desired, it is

generally preferable to eliminate diffusional phenomena as rate-control

ling processes by adjustment of the system parameters rather than includ

ing their influence in the rate equations. This philosophy is in keeping

with the objectives of this investigation, and the following discussion

will outline how pore diffusion and bulk mass transfer may be eliminated

as rate controlling steps. Detailed reviews on the diffusion processes

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115

in porous catalysts and their effect on reaction rates have been

/Og\ Z O O \ presented by Wheeler and Weisz and Prater.

The mass transport processes of interest in heterogeneous

catalysis include diffusion of mass in the bulk fluid phase surrounding a

catalyst particle and in the pore network of the catalyst. These events

have been called interparticle and intraparticle transport, respec- (9) tively. Interparticle or bulk mass transfer is the transport of mass

from the bulk fluid phase to the exterior surface of the catalyst particle.

The total mass flux in bulk or ordinary mass transfer can be separated

into a component resulting from ordinary molecular diffusion and a

component resulting from bulk or convective transport.^ Frequently,

bulk mass transport is described in terms of the film concept.The

film concept assumes that mass transfer takes place across a laminar

film at the fluid-solid interface by molecular diffusion under the

influence of a concentration gradient. Where gas film resistance is /on 1 important, conversion will vary with changing gas flow rate. *

Experimentally, the effect of film resistance can be determined by making

a series of experimental runs at varying bulk gas velocities holding all

other conditions constant. If gas film resistance is not controlling

the reactant, conversion should remain the same at all gas flow rates.

The limits of operation where gas film resistance becomes rate controlling

is the point where conversion just begins to drop.

Since heterogeneous catalysis is a surface phenomenon, porous

catalysts having a high surface area per unit mass are frequently

employed. In porous catalysts, the reaction extends to the interior

surface; and the gross external area is generally a negligible fraction

of the total effective interfacial a r e a . ^ ^ To reach this large

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116

internal surface reactants and products must diffuse through the catalyst

pore network. This phenomena is called intraparticle or pore diffusion

transport. The availability and effectiveness of the interior catalyst

surface depends on the size, shape, and permeability of the pore structure

and upon the rate and nature of the reaction. ■

Amorphous, porous catalysts such as the Pt-Al2 03 catalyst

employed in this investigation are characterized by a distribution of o (26) macropores which typically vary in' size between 10 and 10,000 A.

Diffusion in the macropore network can be described by ordinary molecular /Qg\ diffusion or Knudsen diffusion depending on the macropore size. When

a catalyst contains both amorphous and crystalline components such as a

Pt-AlgC^ -mordenite, a binodal or macro-micropore distribution (26 291 results. ’ This resulting binodal pore distribution consists of a

micropore network within each of the particles of the crystalline material

plus a macropore structure formed by the passageways around the crystal

line material and in the amorphous material. Depending on the crystalline O zeolite employed, micropore diameters may range typically from 3 to 10 A. O Micropore diameters for mordenite are usually about 5 to 7 A. Unlike the

macropore network, a distribution of sizes does not occur for the micro

pore structure since all of the fine channels in the crystallites are the

same size. Diffusion in the micropores can be described by surface

diffusion which has characteristics of both adsorption and ordinary

diffusion. A typical binodal pore size distribution is illustrated in

Figure 11. Several zeolites in addition to mordenite are shown in

Figure 11 with a typical macropore distribution being represented by

activated carbon.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 117

LEGEND

(a) Molecular sieve type 3A (b) Molecular sieve type 5A (c) Mordenite (d) Molecular sieve type 13X (e) Typical activated carbon

100

CO a) M o

o u c tv u M 0) (u

1 5 10 50 100 500 1000 5000 10,000 o Pore Diameter, A .

/OA\ Figure 11. The Binodal Pore Size Distribution.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 118

Experimentally, pore diffusion can never be studied alone since (22) it is not in series with the chemical reaction step. As mentioned

previously, pore diffusion is a function of the catalyst geometry and

nature of the reaction. Diffusion in porous catalysts may be investi

gated by either changing pore length or pore width and observing the rate (17 22) of reaction. * An increase in the pore width or a decrease in the

pore length makes the interior surface more accessible and should speed

up the reaction rate if pore diffusional effects are present. Conversely,

if it should be found in such an investigation that the activity of the

catalyst does not depend on pore width or particle size, the conclusion

can be drawn that pore diffusional effects are absent. A change in

particle size for a catalyst such as Pt-ALjOg-mordenite changes the

macropore length but does not affect the micropores. The only way to

investigate diffusion in the micropores is to vary the crystal size, if

variable crystal sizes are available. Investigation of pore diffusional

effects in this research was limited to the macropores since variable

mordenite crystal sizes were not available. If diffusion in the micro

pores is significant, then its effect is included in the overall reaction

rate.

The effect of pore diffusion may be accounted for in the rate (32) equation by the effectiveness factor which was developed by Thiele (2) and later studied by Aris. The effectiveness factor is defined as

the ratio of the average rate of reaction within a pore to the maximum

reaction rate if pore diffusion is absent.Carberry^^’^ has

studied the influence of nonisothermal conditions on the effectiveness

f a c t o r ^ ^ and has developed a micro-macro effectiveness f a c t o r ^ ^ for

reversible first order reactions over catalysts with a binodal pore

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 119

distribution. Very fast reactions such as the solid catalyzed dehydro

genation of cyclohexane to benzene may be accompanied by significant (34) pore diffusion effects. Weisz and Swegler studied the dehydrogena

tion of cyclohexane over a chromia-alumina catalyst and found that the

ratio of cycloliexene to benzene was dependent on catalyst particle size. (4) Barnett et al studied the dehydrogenation of cyclohexane to benzene

over a platinum-alumina catalyst and a wide range of experimental

conditions. Pore diffusional limitations were encountered and the data

were correlated with an effectiveness factor model.

2. Search for Operating Conditions-Pore Diffusion and Mass

Transfer Limitations

One of the early experimental phases of this investigation was

concerned with the determination of suitable operating conditions for

the dehydrogenation and isomerization of cyclohexane over the Pt-AlgOg-

mordenite catalyst. Although the initial operating conditions could be

chosen more or less arbitrarily, typical operating ranges for commercial

catalytic reforming can serve as a good guide. Temperatures employed

in modern catalytic reforming processes are typically between 85CP and /io\ 975°F. Higher temperatures cause excessive hydrocracking and lower

temperatures are generally unfavorable from an equilibrium viewpoint.

Pressures in catalytic reforming may vary from 200 psig to 600 psig

depending on whether regeneration of the catalyst is frequent or (13) infrequent. Higher pressures are unfavorable due to equilibrium

limitations as well as a tendency to promote undesirable hydrocracking

reactions. Very low pressures can lead to rapid catalyst deactivation

through increased coke deposition. The conditions used for naphtha

reforming over the Pt-Al^Og-mordenite catalyst described in the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120

Standard Oil Company of Indiana Patent example were 9O0PF and 250 /OQ) psig. Hydrogen to fresh feed molar diluent ratios in commercial (13) reforming processes are generally between five and eight. Low

hydrogen recycle rates are avoided because of increased coke deposition

on the catalyst. High hydrogen recycle rates cause the equilibrium

yield of aromatics to change unfavorably but this effect is small and (13) high hydrogen recycle rates may be used for high boiling feedstocks.

No generalizations can be made with respect to space velocity or contact

time. The effects of space velocity and temperature are to some extent

interchangeable, and the space velocity chosen will depend on activity

of the catalyst and severity desired.

Using the aforementioned operating ranges as a guide, a tem

perature of 900Pf and a pressure of 225 psia were chosen for the initial

experimental work on cyclohexane over the Pt-Al3 03 -mordenite catalyst.

In this and all subsequent work, a hydrogen diluent ratio in the order

of 20 moles Hg per mole cyclohexane was used. This high hydrogen diluent

ratio was chosen so that volume changes are minimized and constant

volume can be assumed. A discussion of the assumption of constant

volume and its effect on the rate equations for the dehydrogenation and

isomerization of cyclohexane can be found in Chapter V.

The initial exploratory data obtained on cyclohexane over, the

Pt-ALgOg-mordenite catalyst at 900PF and 225 psia are shown in Table 7.

The space time was varied between 1.07 and 3.22 gm cat-min/gm feed and

all other conditions except superficial gas velocity were kept constant.

The space time-cyclohexane conversion data in Table 7 do not appear to

be consistent, in that cyclohexane conversion does not increase smoothly

with increasing space time. As far as runs 8D and 8B are concerned, an

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121

Exploratory Data for Cyclohexane over the Pt-Al2 03 - Table 7. Mordenite Catalyst at. 90CPf and 225 psia

Operating Conditions

Feed Cyclohexane Catalyst Pt-Al^Og -mordenite Catalyst Particle Size, irrni 0.147-0.833 Temperature, °F 900 Pressure, psia 224.7 Mole Ratio, Hg /cyclohexane ~ 2 0

* Experimental Results

Superficial Cyclohexane Space Time, 0 Gas Velocity Conversion Run No. gm cat-min/gm feed cm/sec Wt. %

7A 1.07 5.3 37.1

8C 1.61 8/5 51.0 8D 1.61 8.5 50.2 7B 1.77 3.2 44.6

8B 2.14 6.4 40.4

7C 2.67 2.1 54.4

9A 3.19 6.4 91.8 8A 3.22 4.2 73.5

Detailed run data are given in Appendix B.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 122

increase in space time actually causes a decrease in cyclohexane conver

sion. The data seem to be correlated with superficial gas velocity in

addition to space time, indicating that mass transfer is rate-limiting.

The effect of gas velocity on conversion seems to be more pronounced at

high conversion levels than at low conversion levels. For example, at

a space time of 1.7 gm cat-min/gm feed a large change in velocity

produces only a small change in cyclohexane conversion while at a space

time of 3.2 gm cat-min/gm feed a small change in gas velocity causes a

large change in conversion. The dependence of the degree of mass trans

fer limitation on cyclohexane conversion level can be seen more easily

in Figure 12 which is a crossplot of the data in Table 7. Figure 12 is

a plot of cyclohexane conversion versus superficial gas velocity with

lines of constant space time. Figure 12 clearly shows that at these

conditions cyclohexane reactivity over the Pt-Algt^ -mordenite catalyst

is severely mass transfer limited at high levels of conversion or space

time. The dependence of conversion on velocity at constant space time,

however, seems to disappear at low conversion levels. These data indi

cate that at a space time of about 1.0 gm cat-min/gm feed the reaction

is no longer mass transfer limited.

An evaluation of the mass transfer effect at low conversion

levels was obtained by making a series of experimental runs over the

Pt-ALj03 -mordenite catalyst at 900PF, 225 psia and a constant space

time (1.07 gm cat-min/gm feed) with varying superficial gas velocity.

These data are summarized in Table 8 and plotted in Figure 13. The

relative constancy of cyclohexane conversion with increasing superficial

gas velocity indicates that mass transfer is not limiting at this level

of space time. If fluid to particle mass transfer had been rate

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123

Runs 7,8,9

100

0=3.2

0= 2.0

0= 1.0 ' — -G

4 8 12 16

Reactor Velocity, cm/sec

Effect of Cyclohexane Conversion Level on the Degree Figure 12. of Bulk Mass Transfer Limitation for the Exploratory Data on the Pt-AL,0, -Mordenite Catalyst at 900°F and 225 psia.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 124

Test for Mass Transfer Limitations at Low Cyclohexane Table 8 . Conversion Levels over the Pt-AlgOg -Mordenite Catalyst at 90CPF and 225 psia.

Operating Conditions

Feed Cyclohexane Catalyst Pt-Alg03 -Mordenite Catalyst Particle Size, tmn 0.147-0.833 Temperature, °F 900 Pressure, psia 224.7 Space Time 0, gm cat-min/gm feed 1.07 Mole Ratio, Hj/cyclohexane ~ 2 0

* Experimental Results

Superficial Gas Cyclohexane Run No. Velocity, cm/sec Conversion, %

7A 5.3 37.1

10A 6.8 38.7

10B 6.8 38.7

13A 8.5 39.7

•At Detailed run data are given in Appendix B.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

9.0 8.0 7.0 6.0 Reactor Velocity, cm/sec Runs 7A,10,13A 5.0 20 60 40 >> c o u > u o Test Mass for TransferLimitations Lowat Cyclohexane ConversionLevels over the ure ure • pt-AlgOa-Mordenite and 225 psia. Catalyst at 90(f>F 8

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126

limiting, the high velocity run should have given a significantly higher

conversion.

The particle size range of the Pt-AlgOg-mordenite used for the

mass transfer study at 900PF, 225 psia, and low space time was 0.147-

0.833 mm. To investigate the effects of a smaller particle size and

possible pore diffusion limitations at these conditions, a 0.147-0.351

mm Pt-ALgOg-mordenite was used for cyclohexane dehydrogenation and

isomerization at 900PF, 225 psia, and low 0. These results are shown in

the first two runs (10A and 14A) of Table 9. This small decrease in

particle size causes the cyclohexane dehydrogenation rate constant to

increase by a factor of seven while the cyclohexane isomerization rate

constant is approximately doubled. These data indicate that this system

is severely pore diffusion limited at these conditions.

Severe pore diffusion limitations can sometimes be overcome by

appropriate reductions in particle size or reaction temperature. Condi

tions must be selected where the diffusional resistance is unimportant

relative to the resistance of the chemical reaction. The data on the Pt-ALjOg -mordenite at 90CPF and 225 psia indicate that cyclohexane conversion is severely pore diffusion limited. A change in reaction

temperature seemed in order so that the surface reaction step could be

slowed down relative to the pore diffusional process. A corresponding

reduction in pressure was also necessary so that a favorable equilibrium for dehydrogenation could be maintained. The second part of Table 9 (runs 14B and 15A) shows the experimental data from two particle size runs with cyclohexane over the Pt-Al2 03 -mordenite catalyst at 80CPF and 85 psia. All conditions except particle size range were held con stant in these runs. These data at 800PF show that there is a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 127 Ks

0.00487 0.00706 0.00500 0.00564 0.00260 Isomerization

Rate Constant, ki

0.02263 0.04309 0.00461 0.03522 0.00563 gm mols/gmgm cat-atm-min

Dehydrogenation 1.07 2.1

~20 Cyclohexane Pt-AlgOg -Mordenite Pt-AlgOg 68.7 56.2 0.03089 47.0 92.1 38.7 Conv., % Cyclohexane

Size, Size, mm Catalyst 0.074-0.147 0.074-0.147 0.147-0.351 0.147-0.351 0.147-0.351. 49.1 0.02229 0.147-0.833

224.7 psia Pressure, /cyclohexane \ °F 750 84.7 750 84.7 800 84.7 800 84.7 900 224.7 Temperature, Detailed run data are givenAppendix in B. Effect Catalystof Particle Size on Cyclohexane Dehydrogenation and Isomerization Rate Constants for Pt-Al^-Mordenite the Catalyst. Space Time 9, gm cat-min/gm feed Reactor Gas Velocity, cm/sec Catalyst Mole Ratio, Feed 15B 15A 16A 14B 10A 14A 900 ExperimentalResults Operating Conditions RunNo. Table 9.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128

significant increase in both rate constants by going from 0.147-0.351 mm

to 0.074-0.147 mm particles. As at 9QCPF, these data show that pore

diffusional limitations exist for cyclohexane dehydrogenation and isomer

ization over Pt-AlgOg-mordenite at 800PF and 85 psia. The experimental

results of a second reduction in temperature to 75CPF are shown in the

third part of Table 9 (runs 16A and 15B). These data at 75GPF and 85

psia show no real positive effect on the rate constants by decreasing the

Pt-AlgC^ -mordenite particle size from 0.147-0.351 mm to 0.074-0.147 mm.

It can be concluded from these results that macropore diffusion is not

rate limiting for cyclohexane over the Pt-ALgOa -mordenite catalyst at

75GPF and 85 psia.

The experimental data on cyclohexane dehydrogenation and

isomerization over the Pt-ALgC^-mordenite catalyst show that macropore

diffusion becomes rate controlling somewhere between 750° and 800PF when

using 0.147-0.351 mm particles. This particle size range is significant

because it is about the smallest particle size range that can be handled

conveniently. Smaller particle size ranges are difficult to handle with

out dust formation and subsequent catalyst loss. A great deal of the

experimental work in this investigation on cyclohexane dehydrogenation

and isomerization is performed at constant temperature. This constant

temperature at which variations in space time, total pressure, and

partial pressure are made can be called the base case temperature.

Although 75CPF could be safely chosen for the base case temperature,

7 7 ^ F is a better choice from the standpoint of a more favorable cyclo

hexane dehydrogenation equilibrium. The equilibrium yield of benzene

from cyclohexane dehydrogenation decreases with decreasing temperature

and increasing total pressure. Since a wide variation in pressure is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 129

desirable for studying the reaction kinetics, the higher temperature of

775°F allows the favorable dehydrogenation equilibrium to be maintained

over a wider range of pressure. The reaction equilibria between cyclo

hexane, benzene, and methylcyclopentane at 75(P and 775°F as a function

of pressure is given in Table 10. The equilibria in Table 10 were

calculated using the equilibrium constants calculated from free energy

data and assuming ideal gas behavior.

The choice of 775°F as the base case temperature for the

kinetic study of the dehydrogenation and isomerization of cyclohexane

over the Pt-Al2 03 -mordenite catalyst cannot be made arbitrarily. The

data show that pore diffusion ceases to become rate limiting somewhere

between 750P and 80CPF. There are no assurances that this transformation

does not occur between 750P and 775°F. For this reason, tests for mass

transfer and pore diffusional limitations were made at 775°F and 85 psia

for the dehydrogenation and isomerization of cyclohexane. These results

are discussed in the next two sections of this chapter.

C. Test for Mass Transfer Limitations at 775?F and 85 psia - Pt-A12 Q3 -

Mordenite Catalyst

The effect of mass transfer on the dehydrogenation and isomerization

of cyclohexane over the Pt-Al^Og-mordenite catalyst at 775°F and 85 psia

was evaluated by making a series of experimental runs with varying gas

velocities. Other experimental conditions such as space time, catalyst

particle size, and hydrogen diluent ratio were held constant. These

runs are summarized in Table 11 and the rate constants are plotted

against superficial gas velocity in Figure 14.

The experimental data presented in Table 11 and Figure 14 show no

consistent positive trend in the reaction rate constants or cyclohexane

conversion with increasing superficial gas velocity. Although there is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 130

Equilibrium Composition of the Cyclohexane-Benzene- Table 10. Methylcyclopentane System as Determined from Free Energy Data.

Temperature, Pressure, Equilibrium Mole Fraction______psia Cyclohexane Benzene Methylcyclopentane

750 50 0.004 0.920 0.076 100 0.035 0.600 0.365 150 0.064 0.316 0.620 200 0.089 0.161 0.750

775 50 0.004 0.961 0.035 100 0.026 0.757 0.211 150 0.055 0.482 0.464 200 0.076 0.283 0.642

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 131

84.7 775 ~ 2 0 Rl Rl Rn 0.04873 0.03617 0.00836 0.00654 0.04450 0.04706 0.00579 0.00756 0.147-0.351 Cyclohexane Dehydrogenation Isomerization Pt-AlÔa -Mordenite Pt-AlÔa Rate Constant, gmmols/gm cat-atm-min % 50.7 6.86 3.14 56.7 5.30 60.7 4.16 59.6 cm/sec Conv., mm , , Superficial Gas Velocity Cyclohexane * Space Time, 9 Catalyst Particle Size Temperature, °F Pressure, psia Catalyst Mole Ratio /cyclohexane Feed * and Isomerization over Pt-AlÔa-Mordenite the Catalyst. 6 Detailed run data are given Appendix in B. Experimental Results Operating Conditions 18A 18A 0.766 21A 0.765 22A 23A 0.766 0.766 No. No. gm feed Run gm cat-min/ . . .. Test Mass for TransferLimitations at 775?F and psia85 Cyclohexane- Dehydrogenation

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 132

LEGEND Dehydrogenation, kx Q Isomerization, kg A

0.05

0.04

E 4J 0.03 03I C B i 4-» u 0.02

I(0 H o B E 6 0 ft £ M O 0.01 0.009

0.007

0.005

3.0 5.04.0 6.0 7.0

Velocity, cm/sec

Test for Mass Transfer Limitations at 775°F and 85 psia- Figure 14. Cyclohexane Dehydrogenation and Isomerization over the Pt-Al3 03 -Mordenite Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 133

more variability in this particular set of experimental data than is

generally desired, the conclusion that mass transfer is not rate limiting

is the only one that can be reached that is consistent with the experi

mental data. If bulk mass transfer had been rate limiting for the Pt-

AlgOg —mordenite catalyst, the high velocity run would have shown a

significantly higher cyclohexane conversion than was observed at the

lower gas velocities. On the basis of these experimental data, it is

concluded that bulk mass transfer to the Pt-Alg03 -mordenite surface is

not rate controlling for cyclohexane dehydrogenation and isomerization

at 775°F and 85 psia and the range of gas velocities investigated.

The Pt-Alg03 catalyst was not examined for possible mass transfer

limitations in cyclohexane dehydrogenation and isomerization. On the

basis of the data obtained for the Pt-Alg03 -mordenite catalyst, it was

assumed that the Pt-Alg03 catalyst would also show no mass transfer

limitations at these experimental conditions.

D. Test for Pore Diffusion Limitations at 775°F and 85 psia - Pt-AL,0B -

Mordenite and Pt-AlgOg Catalysts

Once it has been shown that bulk mass transfer contributes a negli

gible resistance to the overall reaction rate, the effect of possible

pore diffusion limitations must be investigated. The effect of pore

diffusion on cyclohexane dehydrogenation and isomerization reaction rates

has been determined for both catalysts by changing catalyst particle

size with all other conditions held constant. For a completely amorphous

material such as the Pt-Al^Os catalyst, a variation in particle size

provides a complete test for possible pore diffusional limitations. As

was pointed out previously, a variation in catalyst particle size for a

material containing a crystalline component such as the Pt-AlgOg-mordenite

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 134

catalyst gives information only on diffusion within the macropores.

Since it was not possible to vary the mordenite crystallite size, any

diffusional limitations imposed by the micropore structure would be

included in the overall reaction rate.

The effect of macropore diffusion in the Pt-Alg03 -mordenite catalyst

at 775?F and 85 psia was examined by making experimental runs with cyclo

hexane over four different contiguous size ranges of catalyst. The size

ranges examined for the Pt-AlgOg-mordenite catalyst were 0.833-1.397 mm,

0.351-0.833 tnrn, 0.147-0.351mm, and 0.074-0.147 mm. All process variables

except particle size were kept constant in these experimental runs. The

experimental results that were obtained from this study are presented in

Table 12. Figure 15 shows a plot of the dehydrogenation and isomeri

zation rate constants as a function of average particle diameter of the

Pt-AlgOg-mordenite. It is readily apparent from Table 12 and Figure 15

that the dehydrogenation and isomerization of cyclohexane over the Pt-

ALjOg-mordenite at 775°F is pore diffusion limited at large particle

sizes. A change in the particle size range from 0.833-1.397 mm to

0.147-0.351 mm causes the dehydrogenation rate constant to undergo a

fourfold change while the isomerization rate constant is tripled. A

further reduction in catalyst particle size range to 0.074-0.147 mm,

however, causes no positive change in the rate constants or cyclohexane

conversion. It must be concluded, therfore, that at 775°F macropore

diffusion is rate controlling for this system at particle size ranges

larger than 0.147-0.351 mm. It can also be concluded that at 775°F

macropore diffusion in the Pt-AlgOg-mordenite contributes a negligible

resistance to the overall reaction rate at a particle size of 0.147-

0.351 mm or smaller. All further experimentation with cyclohexane on the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 135

0.00801 0.00267 0.00836 0.00302

0.77 84.7 775 ~ 2 0 Cyclohexane - Cyclohexane- Dehydrogenation Pt-ALgOg -Mordenite Pt-ALgOg

k-i k-i Ka 0.04393 0.00999 0.03064 0.04873 Dehydrogenation Isomerization Rate Constant, mols/gm gm cat-atm-min

* 7o

57.0 20.4 42.6 Conv., Conv., Cyclohexane

Catalyst Size

Detailed run data are givenAppendix in B. Space Time 0, gm cat-min/gm feed Pressure, psia Catalyst Temperature, °F Mole Ratio, Hg/cyclohexane Feed 17A 0.074-0.147 25A 0.351-0.833 •k 24A 0.833-1.397 ‘ and Isomerization over Pt-Al^Og-Mordenite the Catalyst. Experimental Results • • 18A 0.147-0.351 60.7 Run No. tnm , Range 6 Operating Conditions a a T ble 12 ^est Pore f°r DiffusionLimitations at 7 7 ^ F and 85 psia

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 136

______LEGEND______Dehydrogenation, k: O Isomerization, kg A

0.05

0.04

0.03

0.02 «o u o B 00 co o B 0.01 B 00 0.008

Vl o 0.006 r~<

0.004

0.1 0.2 0.4 0.8 1.0

Average Particle Diameter, mm

Test for Pore Diffusion Limitations at 775°F and 85 Figure 15. psia - Cyclohexane Dehydrogenation and Isomerization over the Pt-Alg03 -Mordenite Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 137 4

Pt-AlgOg -mordenite catalyst was performed at temperatures of 7 7 ^ F or

less and a particle size range of 0.147-0.351 mm. These restrictions

assured that no pore diffusional limitations would be encountered.

The effect of particle size on cyclohexane reaction rate at 775?F

and 85 psia was also evaluated for the Pt-ALg03 catalyst. Four

contiguous particle size ranges were studied and all experimental con

ditions except Pt-AlgOg catalyst particle size were kept constant. The

experimental data from these runs are given in Table 13 while Figure 16

shows a plot of the dehydrogenation rate constant versus average particle

diameter. These data on the Pt-AlgOg catalyst indicate that cyclohexane

dehydrogenation is pore diffusion limited at all of the particle size

ranges investigated. Furthermore, since the dehydrogenation rate con

stant shows no tendency of leveling off even at the smallest particle

size investigated (0.044-0.074 mm), it must be assumed that the particle

size at which pore diffusion is not rate controlling is very small. The

rate constants in Table 13 indicate that the isomerization of cyclo

hexane to methylcyclopentane over the Pt-AlgOg catalyst is practically

non-existent. Although finite isomerization rate constants were obtained

for all of the runs in the Pt-AlgOg particle size study the mole fraction

of methylcyclopentane in the product was less than 0.01 for all of the

runs. The constancy of the cyclohexane isomerization rate constants for

the Pt-AlgOg catalyst with decreasing particle size in Table 13 indicates

that the isomerization reaction over this catalyst is not pore diffusion

limited at any of the particle sizes investigated.

The smallest particle size investigated for the Pt-AlgOg catalyst

was 0.044-0.074 mm which corresponds to a screen size of 200-325 mesh.

Catalyst particles in this size range are very difficult to handle and

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 138 0.00045 0.00042 0.00118 0.00040 0.00037 0.00051 3 0.55 84.7 775 ~ 2 0 Pt-AL,0 Cyclohexane gm mols/gmgm cat-atm-min - Cyclohexane- Dehydrogenation

0.1061 0.09159 0.06408 0.05946 0.05101 0.01637 Dehydrogenation Isomerization Rate Constant, at 775? F and psia 85 at 775? Catalyst.

66.1 53.9 71.4 50.9 18.7 44.3 Conv., Conv., % Cyclohexane

Catalyst Size

Detailed run data are given Appendix in B. Space Time, 9, gm cat-min/gm feed Catalyst Temperature, °F Mole Ratio, Hg/cyclohexane Pressure, psia Feed 39A 0.044-0.074 38A 40A 0.074-0.147 0.044-0.074 37A 0.074-0.147 36A 0.147-0.351 43A 0.351-0.833 Run No. Range, mm Experimental Results Operating Conditions ' and Isomerization over P the t - A L ^ 6 . - - . Test Porefor DiffusionLimitations

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 139

0.20

0.10

0.05 60

0.02

0.01 0.05 0.1 0.2 0.4. 0.6

Average Particle Diameter, mm

Test for Pore Diffusion Limitations at 775°F and 85 ■Figure 16. psia - Cyclohexane Dehydrogenation and Isomerization over the Pt-Al2 03 Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 140

cause high reactor pressure drop. For these reasons, Pt-A]g03 particle

sizes smaller than 0.044 mm were not investigated.

The difference in behavior for the Pt-Alg03 -mordenite and the Pt-

AlgOg catalysts for cyclohexane dehydrogenation and isomerization can be

partly explained in terms of differences in composition of the two

catalysts. Examples of the cyclohexane dehydrogenation and isomerization

rate constants for these two catalysts at the smallest particle sizes

investigated are shown below.

Rate Constant @ 775°F, 85 psia, gm moles/gm cat-atm-min Particle Size Dehydrogenation Isomerization ______Catalyst_____ mm______kj______kg__

Pt-AlgOg-mordenite 0.074-0.147 0.04393 0.00836

Pt-AlgOg 0.044-0.074 0.09159 0.00045

The cyclohexane isomerization reaction rate has been shown to be indepen

dent of particle size for each catalyst in this particle size region and

hence not limited by macropore diffusion. A ratio of the isomerization

rate constants for the two catalysts shows that the Pt-Alg03 -mordenite

catalyst is approximately eighteen times as active for isomerization of

cyclohexane as the Pt-Alg03 catalyst. The ammonia chemisorption data

presented in Chapter IV for these two catalysts showed that the mordenite

was approximately twelve times more acidic than the halogen-free alumina.

Comparison of the isomerization activity ratio and the relative acidities

of the two catalysts demonstrates that isomerization promoting ability is

directly related to the degree of acidity. It is especially significant

to note that this isomerization promoting ability in a bifunctional

reforming catalyst may be supplied by the addition of a small quantity

of a crystalline zeolite.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 141

Another difference in behavior of the two catalysts is their rela

tive ability to promote cyclohexane dehydrogenation. Although the Pt~

Alg03 catalyst is pore diffusion limited at 775PF, it is approximately

twice as active for dehydrogenation as the Pt-AlgOg -mordenite which is

not pore diffusion limited. The platinum concentrations of both cata

lysts are quite close and both catalysts are derived from the same

alumina base. On the basis of these composition similarities, it might

be expected that these catalysts would show relatively close dehydro

genation activities. The lower dehydrogenation activity exhibited by

the Pt-AlgOg-mordenite may indicate that the mordenite addition has

unfavorably modified some of the metal sites causing this catalyst to

be less active for dehydrogenation. Catalyst manufacture is still, to

some extent, an art and duplication of catalysts is sometimes difficult

even in the most carefully controlled preparations.

E . Reproducibility of the Experimental Data on the Pt-AljO^-Mordenite

Catalyst at 775?F and 85 psia

A knowledge of data reproducibility in an experimental study is

helpful in deciding whether or not observed trends in the data are

statistically significant. During the course of the experimental inves

tigations on cyclohexane dehydrogenation and isomerization over the

Pt-AlgOg-mordenite catalyst, several runs were periodically repeated at

a constant set of conditions. These runs not only provided data for

calculation of statisitcal information but ensured that the basic cata

lyst activity and equipment calibrations had not shifted.

Table 14 gives the experimental results for the six reproducibility

runs with cyclohexane over the Pt-AlgOg-mordenite catalyst at 775°F and

85 psia. All experimental conditions for these runs were held relatively

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 142 ks 0.00881 0.00871 0.00750 0.00743 0.00746 0.00760 cat-atm-min at 775? F at 775?

84.7 -Mordenite 775 775 ' 3 ~ 2 0 0.147-0.351 Cyclohexane Pt-A1^0 fc, 0.03670 0.04185 0.04270 0.04862 0.04370 0.04988 Dehydrogenation Isomerization Rate Constant, mols/gm gm y« 0.080 0.084 0.070 0.069 0.067 0.068 Fraction • b Y

gm cat-min/ ProductMole Space Time, 9 and psia. 85 Reproducibility of Experimentalthe Data on Pt-ALgOj-Mordenitethe Catalyst Catalyst Catalyst Size Range, min Temperature, °F Mole Ratio, Hj/cyclohexane Pressure, psia Feed Detailed run data are givenAppendix in B 33D 0.551. 0.349 33A 0.551 0.386 28C 28C 0.547 0.438 35A 0.552 0.407 41G 0.551 0.399 28A 0.547 0.444 Run No. gm feed Experimental Results Operating Conditions oe i 6 TclDl

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 143

constant and five of the runs represented different catalyst batches.

The data show some variability, but it is not excessive for data

obtained on solid catalyzed reactions. The first five runs in Table 14

(28A-33A) show good reproducibility in the product mole fractions and

rate constants. The last run in Table 14 (33D) is somewhat separated

from the rest of the data and may represent catalyst deactivation or an

error in product analysis.

The data in Table 14 were used to calculate the mean values and

standard deviations of the rate constants and product mole fractions.

These statistical parameters are presented in Table 15. In addition,

the standard deviation as a percentage of the mean value for each

parameter is also shown. These percentage standard deviations for all of

the parameters are between 7 and 10%. The standard deviations calculated

for these runs seem to be perfectly reasonable for the type of data

obtained.

F. Test of the Reaction Model for the Dehydrogenation and Isomerization

. of Cyclohexane - Pt-AL,0,,-Mordenite Catalyst at 775?F and 85 psia

In Chapter V of this investigation, a kinetic model for the simul

taneous, reversible dehydrogenation and isomerization of cyclohexane

was developed. This model assumed that the surface reactions were rate

controlling and that the dehydrogenation and isomerization steps were

first order in both directions. The kinetic model was developed for

constant temperature and it was also assumed that the simplified rate

constants were independent of composition and that hydrocracking of

cyclohexane was negligible. Simultaneous solution of the kinetic rate

equations yielded two non-linear equations which expressed the outlet

concentrations of benzene and methylcyclopentane as functions of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

. ♦ ♦ 0.000598 7.55 0.0060 0.00438 8.23 9.96 0.032 7.92 Standard Standard Deviation, Deviation Mean of 7» calculated from the data

y„' y„' 0.073 yB' yB' 0.404 Measured or Calculated Variable MeanValue in Table in 14. Statistical parameters in this table are kg, kg, mols/gm gm cat-atm-min 0.00792 kx, kx, mols/gmgm cat-atm-min 0.04391 , . . , StatisticalParameters Calculated from Reproducibilitythe Study on the Table 15. .* _Mordenite Catalyst

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 145

rate constants, space time i»d process variables:

yg = f (kj, kg, 0 , process variables) (1)

yH = S (^i> ^2 > 0 > process variables) (2)

where

yB' = outlet mole fraction benzene on a hydrogen-free basis,

yH' = outlet mole fraction methylcyclopentane on a hydrogen-

free basis.

An optimization technique was applied to equation (1) and (2) so that

the rate constants could be solved for, given the outlet composition,

space time and other process variables.

The validity of the kinetic model for the assumed reaction order

may be examined by varying the space time at constant temperature,

pressure, and hydrogen diluent ratio. The rate constants obtained from

a variation in space time at constant temperature and pressure should

remain relatively constant since they are not functions of the space time.

The validity of a kinetic model for simple first order unimolecular

reactions may also be examined by plotting the experimental data in such

a way that a linear relation should result if the assumed model is

applicable. Unfortunately, the integrated rate equations that result

from complex reactions may be of such a form that it is impossible to

plot a function of the composition versus the space time so that a

linear relation is obtained if the assumed model is applicable. This

restriction in complex reactions may be circumvented by plotting the

experimental composition data versus space time and comparing to a

reaction model curve generated by varying space time and using the

average values of the rate constants.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 146

The validity of the kinetic model for cyclohexane dehydrogenation

and isomerization over the Pt-ALgOg-mordenite catalyst was tested by

varying space time with all other conditions held constant. Thirteen

experimental runs were made at 775PF, 85 psia, and 20 moles Hg/mole

cyclohexane feed. The space time was varied over a tenfold range ranging

from 0.184 to 1.521 gm cat-min/gm feed. The product composition and

calculated rate constants for each of these runs are shown in Table 16.

The cyclohexane conversion ranged between 21.9 and 78.0 wt. %. The

degree of hydrocracking for all of the experimental runs shown in Table

16 was less than 1.5 wt. % which can be considered negligible.

The dehydrogenation rate constants shown in Table 16 all fall within

the range of 0.037 to 0.054 gm moles/gm cat-atm-min. The majority of

the dehydrogenation rate constants (8), however, fall in the range of

0.042 to 0.050 gm moles/gm cat-atm-min. The mean value of the dehydro

genation rate constant for these thirteen runs was 0.0453 gm moles/

gm cat-atm-min and the calculated standard deviation was 1 1 .6% of the

mean. The percentage standard deviation calculated for the dehydro

genation rate constant in previously discussed reproducibility study

was 10.0%. These two values for the standard deviation of the dehydro

genation rate constant are close enough so that the variation in kj may

be attributed to experimental error.

The isomerization rate constants shown in Table 16 all fall within

the range of 0.00546 to 0.00881 gm moles/gm cat-atm-min and the value

of the isomerization rate constant for run 31A seems to be inordinately

low. The mean value for the isomerization rate constant was 0.00750

gm moles/gm cat-atm-min and the calculated standard deviation for all of

the runs was 12.6% of the mean. As pointed out previously, the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 147

k* 0.00743 0.00881 0.00871 0.00750 0.00546 0.00631 0.00756 0.00617 0.00760 0.00830 0.00801 0.00836 Isomerization 84.7 775 ~ 2 0 mols/gm cat-atm-min 0.147-0.351 Cyclohexane Pt-AlgOg -Mordenite Pt-AlgOg h, 0.04370 0.04274 0.03670 0.03931 0.048620.05398 0.00746 0.04706 0.05339 0.03890 0.04988 0.04393 0.04873 Dehydrogenation Rate Constant, gm y« 0.069 0.084 0.070 0.080 0.04185 0.035 0.061 0.023 0.085 0.108 0.068 0.090 yB ' 0.399 0.349 0.386 0.252 9.394 0.200 0.438 0.067 0.444 0.493 0.530 0.091 * SpaceTime ProductMole 9, 9, gm cat-min/ Fraction Test of Reactionthe Model for Dehydrogenationthe and Isomerization Cyclohexaneof - Detailed run data are given Appendix in B Catalyst Size Range, mm Temperature, °F Catalyst Pressure, psia Mole Ratio, /cyclohexane Feed Experimental Results Operating Conditions 30A 0.184 35A 0.552 0.407 33D 0.551 41G 0.551 33A 0.551 31A 0.330 18A 18A 0.766 28C 28C 0.547 23A 0.766 0.527 27A 28A 1.521 0.547 0.671 17A 0.766 28B 0.438 Run No. gm feed a a e ' Variation of Space Time over Pt-Al^Og-Mordenite the Catalyst at 775PF and 85 psia.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. isomerization rate constant for run 31A seems to fall outside of the

general population. If run 31A is excluded, the percentage standard

deviation of the isomerization rate constants for the remaining twelve

runs drops to 10.4%. The percentage standard deviation calculated for

the isomerization rate constant in the reproducibility study was 7.6%.

Although these two standard deviation are more widely separated than

those for the dehydrogenation rate constant, it is felt that they are

close enough so that the scatter in the isomerization rate constant can

be attributed to experimental error. The range of allowable methylcyclo

pentane compositions at these conditions is rather narrow and limited by

equilibrium to a mole fraction of 0.147 methylcyclopentane. The runs in

the reproducibility study were made at an intermediate space time with

the mean methylcyclopentane mole fraction being 0.073 (see Table 15) .

The majority of the scatter in the isomerization rate constant shown in

Table 16 occur at methylcyclopentane concentrations which are very low

or very close to equilibrium. If the reproducibility study on the isom

erization rate constant had been performed closer to these two limiting

conditions, it is highly possible that a standard deviation greater than

7 .6% would have been obtained.

In addition to an examination of the constancy of the rate constants

with varying space time, the validity of the kinetic model may also be

examined graphically. Due to the complexity and extreme non-linearity

of the integrated rate equations, there is no way to rearrange them so

that functions of the experimental data may be plotted that are indepen

dent of the rate constants. The experimental composition-space time

data, however, may be plotted and compared to curves that are generated

from the integrated rate equations by varying space time and using

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 149

average or best fit values of the rate constants. The generation of the

theroretical composition-space time curves and their comparison to the

experimental data is discussed in the following paragraphs.

The first step in the generation of the theoretical curves from the

integrated rates equations was the selection of values to be used for

the reaction rate constants for the Pt-Alg03 -mordenite at 775°F and 85

psia. The integrated rate equations may be put in the form of residuals

Ri = f (1^ , kg, 0, process variables) - yB' (3)

and

®3 = g (ki > hg > 9» process variables) - y„' (4)

where

Rx = residual difference between a predicted and an experi

mental value of yj,

Rg = residual difference between a predicted and an experi

mental value of .

The values of kx and kg that best fit a set of experiments data where

only 0 varies can be found by minimizing a function § which is formed by

summing the squares of the residuals for each data point. The problem

can be formulated mathematically as

m m minimize: $ = ^ x2 + Rls3 , (5) i=l i=l

subject to: kx > 0 , kg > 0

where m is the number of data points considered. The values of and

kg that minimize § are the best fit values. This least squares type

technique was applied to the experimental data in Table 16 to find the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 150

values of kx and kg that best represented the data. The function $ was

minimized by application of the pattern search method discussed in

Chapter V. This proceedure yielded best fit values of kj and of

0.04484 and 0.00777 gm moles/gm cat-atm-min, respectively. These values

are very close to the mean values of kx and kg from Table 16 of 0.0453

and 0.00750 gm moles/gm cat-atm-min, respectively. The theoretical model

curves for the Pt-ALg03 -mordenite catalyst at 775^F and 85 psia were

calculated from the integrated rate equations by using the best fit

values of kx and kg and varying 0 .

A comparison of the experimental composition-space time data from

Table 16 and the model curves generated from the integrated rate equa

tions is shown in Figure 17. The experimental y B', yH', data are plotted

as single points. The curves shown in Figure 17 represent the artifi

cially generated curves from the integrated rate equations. The agree

ment between the reaction model curves and the experimental data is

excellent over the total variation in space time. This agreement between

the experimental data and the model generated curves confirms the

validity of the reversible first order kinetic model developed for cyclo

hexane dehydrogenation and isomerization.

G . The Effect of Temperature on the Cyclohexane Dehydrogenation and

Isomerization Rate Constants - Pt-AlaCh -Mordenite Catalyst

A limited series of experiments were performed with cyclohexane on

the Pt-AlgOg-mordenite catalyst to determine the effect of temperature

on the simplified dehydrogenation and isomerization rate constants.

Because of pore diffusional limitations on cyclohexane dehydrogenation

and isomerization over the Pt-Alg03 -mordenite catalyst, temperatures

above 775^F were not included in the temperature variation. Experimental

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 151

1.4 1.2

1.0 Curve MCP for ReactionModel ReactionModel Curve Benzene for 0.8 0.6

Space Time 0, gm cat-min/gm feed

0.4

LEGEND k, k, = 0.00777 0.2 for kx for kx = 0.04484 Exp. y„ Exp. y„ £ Exp. yB' Q Exp. yB' Model Curves generated 0.2 0.6 0.4 0.8 Variation of Space-Mordenite Time Catalyst over Pt-Al^Og at the 775°F and 85 psia. c o o JJ V5 Figure 17 Test Reaction the Model for Dehydrogenation the and Isomerization Cyclohexaneof -

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 152

runs were made at 85 psia and temperatures of 77iP , 75CP, and 724°F with

all other process conditions being held constant. The experimental data

and calculated reaction rate constants for these runs are shown in Table

17. The dehydrogenation rate constant varied from 0.02730 to 0.04274

gm moles/gm cat-atm-min while the isomerization rate constant varied

from 0.00502 to 0.00743 gm moles/gm cat-atm-min over the temperature

range.

The temperature dependency of experimentally determined reaction

rate constants has been found in practically all cases to be well (22) represented by the Arrhenius equation,

k = , (6)

where

k = arbitrary reaction rate constant,

k,j = frequency factor,

E ® activation energy,

R = gas constant,

T = absolute temperature.

If the Arrhenius equation is applicable, a plot of the logarithm of the

rate constant versus the reciprocal of the absolute temperature should

yield a linear relation.

An Arrhenius plot of the cyclohexane dehydrogenation rate constants

in Table 17 is shown in Figure 18. These data on the dehydrogenation

rate constant appear to be well correlated with an Arrhenius type

relationship. The activation energy for the dehydrogenation of cyclo

hexane as determined from the slope of the Arrhenius plot in Figure 18

is 14.4 kcal/gra mole. Some reported activation energies for the dehydro

genation of cyclohexane to benzene over various catalysts are shown in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

0.551 84.7 ~ 2 0 ki 0.03435 0.02730 0.00637 0.00502 0.04274 0.00743 ■ 0.147-0.351 Cyclohexane Pt-AlgOg -mordenite Pt-AlgOg Rate Constant, mols/gm gm cat-atm-min m Y Product

Ya Ya Mole Fraction Dehydrogenation Isomerization given Appendix in B 0.276 0.052 0.399 0.069 0.337 0.063

* Temperature, Detailed run data are 4IF 750 41H 724 41G 775 Rate Constants Pt-ALgC^-Mordenite- Catalyst at 85 psia Space Time 0, gm cat-min/gm feed Catalyst SizeRange, mm Catalyst Pressure, psia Mole Ratio, Hg/cyclohexane Feed Run No. °F Experimental Results Operating Conditions Table 17. T^e E^fect °f Temperature on Cyclohexanethe Dehydrogenation and Isomerization

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 154

Run Nos. 41F,G,H

Ex=14.4 kcal/gm mole 0.06 k0 1 =1.66xlOP gm mole/ gm cat-atm-min

0.05

B i B cd i 4J 0.04 cdu B.

o e

a 0.03

0.02 8.1 8.2 8.3 8.4 8.5

10000/T(°R)

The Effect of Temperature on the Cyclohexane Figure 18. Dehydrogenation Rate Constant - Pt-Al2 03 -Mordenite Catalyst at 85 psia.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 155

Table 18. Depending on the catalyst employed, these dehydrogenation

activation energies range from 9.9 to 26.0 kcal/gm mole. The activation

energies for platinum-alumina catalysts in Table 18 vary from 11.4 to

17.6 kcal/gm mole. The activation energy of 14.4 kcal/gm mole obtained

in this investigation for cyclohexane dehydrogenation over a Pt-ALgOg-

mordenite catalyst appears to compare favorably with reported values for

similar catalysts.

An Arrhenius plot of the cyclohexane isomerization rate constants

in Table 17 is shown in Figure 19. These data on the isomerization rate

constant also appear to be well correlated with an Arrhenius type

relationship. The activation energy for the isomerization of cyclohexane

as determined from the slope of the Arrhenius plot in Figure 19 is 12.5

kcal/gm mole. From the data discussed previously, it can be concluded

that the cyclohexane isomerization reaction on the Pt-ALgOg -mordenite

catalyst takes place almost exclusively on the mordenite component. (19) Hopper studied the isomerization of cyclohexane on structurally

modified mordenites in the 50CP -67CPF temperature range and reported

isomerization activation energies from 23.9 to 35.5 kcal/gm mole. In

comparison, the isomerization activation energy of 12.5 kcal/gm mole

found in this investigation appears to be low and could possibly repre

sent a pore diffusional limitation due to the higher temperature

employed. For the particle size used in this temperature study, it has

been shown previously that mass transfer or macropore diffusion are not

rate controlling at or below’ 775°F. If the low isomerization activa

tion energy represents a pore diffusional limitation, then it must occur

in the micropores of the mordenite crystallites. Since only one size of

with permission of the copyright owner. Further reproduction prohibited without permission. 156

Reported Activation Energies for the Dehydrogenation a 8 ' o f Cyclohexane to Benzene.

Activation Energy E Catalyst kcal/gm mole Investigator (37) Ni-Charcoal 9.9 Zelinskii and Balandin (27) P t - A l ^ 11.4 Popescu et al (3) Ni-ALjOg 13.6 Balandin and Rubinstein

Pt-AL,03 15.2 Ritchie and Nixon^®^ (37) Pd-Charcoal 15.8 Zelinskii and Balandin

Pt-ALjOa -Cl 16.5 P'ang et al^"*^

Ni-ZnO 17.0 Andreev et al^^

Pt-ALjOa -Cl 17.6 Bridges and Houghton^^ (37) Pt-Charcoal 18.7 Zelinskii and Balandin

Crg O3 -AI3 O3 20 . 0 Bridges and Houghton^^

NiW04 -Al2 03 26.0 Bridges and Houghton^^

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 157

Run Nos. 41F,G,H

0.008

0.007 •rl

0.006

0.005

Eg = 12.5 kcal/gm mole ko2 = 72.4 gtn moles/gm cat-atm-min 0.004

8.1 8.2 8.3 8.4 8.5

10000/T(° R)

The Effect of Temperature on the Cyclohexane Figure 19. Isomerization Rate Constant - Pt-Alg03 -Mordenite Catalyst at 85 psia.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 158

mordenite crystallites was available, it was not possible to investigate

the effect of micropore diffusion in the mordenite.

H. The Effect of Pressure on the Cyclohexane Dehydrogenation and

Isomerization Rate Constants at 775°F - Pt-AL,Oa -Mordenite Catalyst

The final variable investigated in the kinetic study of cyclohexane

dehydrogenation and isomerization over the Pt-AlgOg-mordenite catalyst

was pressure. All of the experimental runs reported thus far on cyclo

hexane over the Pt-AlgOg -mordenite catalyst have been made at a constant

total pressure and a constant hydrogen to cyclohexane feed diluent ratio.

A study of the effect of pressure is necessary to complete the modeling

of the reaction. The experimental results concerning the effect of

pressure on the cyclohexane dehydrogenation and isomerization rate con

stants are presented in this section. The mathematical correlation of

the rate constants with pressure using an adsorption-type model is

discussed in the following section of this investigation.

1. The Effect of a Variation in Total Pressure at a Constant

Hydrogen to Cyclohexane Feed Diluent Ratio

The effect of total pressure on the cyclohexane dehydrogenation

and isomerization rate constants at 775°F and constant feed diluent ratio

was evaluated by varying the total pressure from 55 to 135 psia. All

other experimental conditions were held constant for these runs. The

experimental compositions and calculated rate constants for tfie variation

of total pressure at constant temperature and feed diluent on the Pt-

ALgOg-mordenite catalyst are presented in Table 19. A plot of the

dehydrogenation and isomerization rate constants as a function of total

reactor pressure is shown in Figure 20.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 159

775 -20 0.551-0.552 0.147-0.351 Cyclohexane ki ki hp 0.05206 0.04274 0.00558 0.00743 0.03721 0.04267 0.00713 0.00790 0.04372 0.03530 0.03458 0.00750 0.00609 0.00679 0.04185 0.00871 Pt-AlgOg -Mordenite Pt-AlgOg Dehydrogenation Isomerization Rate Constant, mols/gm gm cat-atm-min m Product Y b Y MoleFraction 0.399 0.069 0.340 0.036 0.483 0.095 0.467 0.087 0.409 0.081 0.407 0.070 0.472 0.083 0.386 0.080 ^-Mordenite Catalyst. Pt-ALj03 - 775 * 84.7 54.7 psia 109.7 109.7 134.7 9, 9, gm cat-min/gm feed , Hg/cyclohexane, Results Detailed run data are givenAppendix in B. Total Pressure SpaceTime Temperature, °F Catalyst Size Range, mm Catalyst Mole Ratio Feed The Effect Totalof Pressure on Cyclohexanethe Dehydrogenation and Isomerization 41G 35C 35A 84.7 35B 41B 41C 41A 134.7 33A 84.7 Run No. Experimental Operating Conditions Table 19. Rate Constants at

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 160

140120 A LEGEND Isomerization, Dehydrogenation,Q kx 100 80 Total Pressure, psia Run Nos. 33A,35,41A,B,C,G 60 40 0.0 0.06 H i i o o E S 0.02 oo CO E B d 3 0.04 E u •Si H The Effect Total of Pressure on Cyclohexanethe Dehydrogenation and Isomerization Rate Sure ‘ Constants at 775°F Pt-Al^Oa-Mordenite- Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 161

The cyclohexane dehydrogenation rate constants listed in

Table 19 range from 0.03721 to 0.05206 gm moles/gm cat-atm-min. Figure

20 indicates that the dehydrogenation rate constant decreases with

increasing total pressure at constant hydrogen to cyclohexane diluent

ratio. The effect of total pressure on the dehydrogenation rate con

stants seems to diminish at the higher pressure levels investigated.

The cyclohexane isomerization rate constants listed in Table

19 range from 0.00558 to 0.00871 gm moles/gm cat-atm-min. The isomeri

zation rate constants plotted in Figure 20 show that the effect of total

pressure, if any, on this rate constant is very small.

The experimental results obtained by varying total pressure

with other conditions held constant indicates that pressure does have •

some effect on cyclohexane reactivity over the Pt-ALjOg -mordenite cata

lyst. A better understanding of the effects of pressure may be obtained

by varying the hydrogen and hydrocarbon partial pressures independently.

Experimental results on the variation of the partial pressures are given

in the following subsection.

2. The Effect of Independently Varying Hydrocarbon and Hydrogen

Partial Pressure

Since the data in this investigation were obtained in an

integral reactor, the partial pressures of the components varied with

catalyst bed length during the course of each run. Although the individ

ual partial pressures vary, the sum of the cyclohexane, benzene, and

methylcyclopentane partial pressures remains constant with respect to.

reactor length. This is because both dehydrogenation and isomerization

reactions are unimolecular with respect to the hydrocarbon components,

and the total moles of hydrocarbon at any point in the reactor is a

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 162

constant. Therefore, the total hydrocarbon partial pressure for any

particular experimental run is a constant and is equal to the initial

partial pressure of the cyclohexane in the feed hydrogen-cyclohexane

mixture. It should also be noted that although hydrogen participates

in the dehydrogenation reaction, the high hydrogen feed diluent ratios

used ensure that the partial pressure remains nearly constant with

respect to reactor length.

The effects of a variation in hydrocarbon partial pressure on

the cyclohexane dehydrogenation and isomerization rate constants for the

Pt-AlgOa -mordenite catalyst at 775°F are shown in Table 20. These

experimental results were achieved by varying the feed diluent ratio

from 15 to 30 moles 1^/mole cyclohexane at constant total pressure. All

other process conditions were held constant for these experimental runs.

The hydrocarbon partial pressure varied between 2.70 and 5.24 psia. Due

to the high hydrogen concentration, even at the lower feed diluent

ratio, the hydrogen partial pressure remained almost constant and only

varied from 79.5 to 81.9 psia. Therefore, the results in Table 20 can

be taken as the effect of hydrocarbon partial pressure on the rate

constants at constant hydrogen partial pressure. As a result of the £ varying hydrocarbon partial pressure, the dehydrogenation rate constant

ranged from 0.03028 to 0.06190 gm moles/gm cat-atm-min while the isomeri

zation rate constant ranged from 0.00643 to 0.01052 gm moles/gm cat-atm-

min.

The rate constants listed in Table 20 are plotted in Figure 21

as a function of the hydrocarbon partial pressure. The correlation of

both rate constants with hydrocarbon partial pressure in Figure 21

appears to be very good. Both rate constants decrease with increasing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. K 0.00643 0.01052 0.00743 0.00750 Isomerization

______

0.04274 0.03473 0.054210.030280.06190 0.00974 0.00669 0.04372

84.7 775 Dehydrogenation Rate Constant, gmmols/gm cat-atm-min 79.5-81.9 0.551-0.552 0.147-0.351 Cyclohexane Pt-AlgC^ -Mordenite Pt-AlgC^ L

j 0.067 0.077 0.082 775 F ~80775 and psia HydrogenParital Pressure -

Product

~ r Mole Fraction JSL. 0.396 0.418 0.399 0.069 0.373 0.386 0.080 0.04185 0.00871

2.74 5.30 2.70 0.355 0.064 3.96 4.01 Pressure, psia -Mordenite Catalyst. 3 Hydrocarbon Partial Detailed run data are given Appendix in B. Space Time 0, gm cat-min/gm feed HydrogenPartial Pressure, psia Feed Catalyst Catalyst Size Range, mm Temperature, °F Pressure, psia The Effect Totalof HydrocarbonPartial Pressure on Cyclohexanethe Dehydrogenation Pt-Alg0 Experimental Results Operating Conditions 33B 33C35A 5.24 4.02 0.407 0.070 33A ID 4 IE 4 41G Run No. Table 20. and IsomerizationRate Constants at

—i O Q. C o CD Q. CD with permission of the copyright owner. Further reproduction prohibited without permission. T3 164

Run Nos. 33A,B,C ,35A,41D,E,G

______LEGEND Dehydrogenation, O Isomerization, kg A

c 0.06 B > 4JB « i u <0 u g, 0.04 to H O B §> J) 0.02 n o

J --

2.0 3.0 4.0 5.0 6.0

p0 , Hydrocarbon Partial Pressure, psia

The Effect of Total Hydrocarbon Partial Pressure on the , Cyclohexane Dehydrogenation and Isomerization Rate gure . const;ants at 775^F and ~80 psia Hydrogen Partial Pressure - Pt-A^Og-Mordenite Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 165

hydrocarbon partial pressure and the effect seems to diminish somewhat at

the higher hydrocarbon partial pressure levels. The percentage reduction

in the dehydrogenation rate constant with increasing hydrocarbon partial

pressure is greater than that for the isomerization rate constant.

The effect of a variation in hydrogen partial pressure at

constant hydrocarbon partial pressure was investigated at two levels of

hydrocarbon partial pressure. These two levels were approximately 5.6

and 2.7 psia and represented the upper and lower limits of the total

hydrocarbon partial pressure variation previously presented in Table 20.

The experimental results on the variation in hydrogen partial pressure

at 5.6 psia hydrocarbon partial pressure and 775^F for the Pt-Al^Og-

mordenite catalyst are shown in Table 21. All other experimental con

ditions were held constant. The hydrogen partial pressure was varied

between 79.4 and 128.3 psia. This variation was achieved by changing

both total pressure and feed diluent ratio. Although there was a slight

variation in the hydrocarbon partial pressure around 5.6 psia, the

results can be interpreted as the effect of hydrogen partial pressure at

constant hydrocarbon partial pressure. The dehydrogenation rate con

stants in Table 21 ranged from 0.03028 to 0.04267 gm moles/gm cat-atm-min

while the isomerization rate constants varied between 0.00609 and 0.00790

gm moles/gm cat-atm-min. The rate constants listed in Table 21 are

plotted in Figure 22 as a function of the hydrogen partial pressure. At

this level of hydrocarbon partial pressure, there seems to be no effect

of hydrogen partial pressure on the isomerization rate constant. The

dehydrogenation rate constant shows a slight upward trend with increasing

hydrogen partial pressure but the effect, if any, is very small. The

percentage standard deviation of the dehydrogenation rate constants in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 166

fe 0.00609 0.00713 0.00790 Isomerization

ki 0.03530 0.034580.03721 0.00679 ' 0.03028 0.00669 775 5.15-6.40 0.147-0.351 0.551-0.552 Dehydrogenation Cyclohexane Pt-ALgOg -Mordenite Pt-ALgOg yM' 0.087 . 0.04267 0.083 0.081 0.095 0.082 Fraction Product Rate Constant, gmmols/gm cat-atm-min 7 y B Mole 0.472 0.418 0.077 0.03473 0.00643 --

79.5 0.373 79.4 104.5 0.467 128.3 104.5128.3 0.409 0.483 Pressure, psia Detailed run data are givenAppendix in B. HydrogenPartial k The Effect of HydrogenPartial Pressure on Cyclohexanethe Dehydrogenationand Pt-Al^Og -Mordenite Catalyst. Pt-Al^Og Space Time 9, gm cat-min/gm feed HydrocarbonPartial Pressure, psia Catalyst Catalyst Size Range, mm Feed Temperature, °F 35C 35B 33C 41A IB 4 IE 4 Experimental Results Operating Conditions RunNo. Table 21. Isomerization Rate Constants at ~5.6 775^F and psia HydrocarbonPartial Pressure

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1 6 7

Run Nos. 33C,35B,C,41A,B,E

______LEGEND______Dehydrogenation, kj Isomerization, kg A

c 0.06 •H Bi B 4J cd i 4J

& 0.02

0.0 60 80 100 120 140

pH , Hydrogen Partial Pressure, psia 3

The Effect of Hydrogen Partial Pressure on the Cyclohexane Dehydrogenation and Isomerization Rate Figure 22. Constants at 775°F and ~5.6 psia Hydrocarbon Partial Pressure - Pt-AlgOa-Mordenite Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 168

Figure 22 about the mean value is 10.7% while that for the isomerization

rate constants is 8.4%. The standard deviation of the dehydrogenation

rate constant calculated for the replicate runs in the reproducibility

study was 1 0 .0% of the mean while that for the isomerization rate con

stant was 7.6%. These two sets of standard deviations are so close that

the variation in both rate constants in Figure 22 may be attributed to

experimental error. It can be concluded, therefore, that at 5.6 psia

hydrocarbon partial pressure, a variation in hydrogen partial pressure

over the range discussed has no effect on either rate constant.

Experimental results on the variation of hydrogen partial

pressure at 2.7 psia hydrocarbon partial pressure and 775°F for the Pt-

AlgOg-mordenite catalyst are shown in Table 22. All other experimental

conditions were held constant for the five runs. The hydrogen partial

pressure was varied between 52.1 and 130.1 psia. The hydrocarbon partial

pressure only varied between 2.60 and 2.74 psia and results can be

interpreted as the effect of hydrogen partial pressure at a constant

hydrocarbon partial pressure of 2.7 psia. The dehydrogenation rate con

stants in Table 22 ranged from 0.05206 to 0.06970 gm moles/gm cat-atm-

min while the isomerization rate constant varied between 0.00558 and

0.01283 gm moles/gm cat-atm-min. The rate constants listed in Table 22

are plotted in Figure 23 as a function of the hydrogen partial pressure.

At this level of hydrocarbon partial pressure, the dehydrogenation rate

constant in Figure 23 shows a slight upward trend with increasing hydro

gen partial pressure. The standard deviation of the dehydrogenation

rate constants plotted in Figure 23 is 10.2% of the mean. Since the

standard deviation of the dehydrogenation rate constant found in the

reproducibility study was 1 0 .0%, it can be concluded that the upward

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 169 -

K 0.01119 0.01283 0.00558 0.01052 0.00974 Isomerization ki 0.06160 0.06970 0.06190 0.05206 0.05421 775 2.60-2.74 0.551-0.552 Dehydrogenation 0.147-0.351 Cyclohexane Rate Constant, mols/gm gm cat-atm-min Pt-AlgOa -Mordenite Pt-AlgOa psia HydrocarbonPartial Pressure - ~2.7 F and Product 775° yB yB yM MoleFraction 0.379 0.069 0.396 0.410 0.067 0.077 0.355 0.340 0.064 0.036 are givenAppendix in B. , psia , * 82.0 52.1 82.0 Detailed run data * Pressure, psia HydrogenPartial Conditions IsomerizationRate Constants at The Effect of HydrogenPartial Pressure on Cyclohexanethe Dehydrogenation and Pt-ALgOa-Mordenite Catalyst. Space Time 0, gm cat-min/gm feed HydrocarbonPartial Pressure Temperature, °F Catalyst Catalyst Size Range, mm Feed 44B 105.1 44A 130.1 33B 41C 41D Experimental Results Operating 22. 22. RunNo. Table

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

140

120 100 80 3 p„ p„ Hydrogen , Partial Pressure, psia 60 Run Nos. Run Nos. 33B,41D,C,44A,B 40 0.0 0.06 0.08 0.04 I I o e E o B m Cft 00 c E E cti u XJ Pt-Al^Og-Mordenite Catalyst. The Effect of Hydrogen Partial Pressure on Cyclohexane the Dehydrogenation and LEGEND Isomerization, kg Dehydrogenation, kj Figure 23. Isomerization Rate Constants at 775°F ~2.7 and psia Hydrocarbon Partial Pressure -

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 171

trend of the dehydrogenation rate constant in Figure 23 is not really

significant. The majority of the isomerization rate constants plotted

in Figure 23 show no effect of hydrogen partial pressure. The isomeri

zation rate constant at the lowest hydrogen partial pressure studied,

however, has a value of about one half of the isomerization rate con

stants over the rest of the pressure range. Although this point could (31) simply represent experimental error, Sinfelt indicates that at very

low hydrogen partial pressures, isomerization rates may increase with

increasing hydrogen partial pressure. Sinfelt suggests that this

phenomena arises when the hydrogen partial pressure is not sufficient

to keep the catalyst surface free of coke-like hydrocarbon residues.

The standard deviation of all of the isomerization rate constants plotted

in Figure 23 is 24.0% of the mean. However, if the isomerization rate

constants at the lowest hydrogen partial pressure is excluded, the

percentage standard deviation falls to 8.3% of the mean. Since there

exists a good possibility that the isomerization rate constant at the

lowest hydrogen partial pressure is low due to catalyst deactivation,

it is felt that the 8.3% standard deviation is more representative. The

value of 8.3% for the standard deviation suggests that a variation in

hydrogen partial pressure from 80 to 130 psia has no effect on the

isomerization rate constant at 2.7 psia hydrocarbon partial pressure.

The effect of pressure on the cyclohexane dehydrogenation and

isomerization rate constants at 775°F on the Pt-ALjOg-mordenite catalyst

has been evaluated by a series of fourteen experimental runs. It can

be concluded qualitatively from these data that:

a. The dehydrogenation rate constant decreases with increas

ing total pressure at a fixed hydrogen to cyclohexane feed diluent ratio.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 172

b. The isomerization rate constant shows no effect of a

change in total pressure at a fixed hydrogen to cyclohexane feed diluent

ratio.

c. Both rate constants decrease with increasing hydrocarbon

partial pressure at a constant hydrogen partial pressure.

d. At constant hydrocarbon partial pressure, the variation

in both rate constants with changing hydrogen partial pressure is

approximately equal to the normal scatter of the data.

It can be concluded, therefore, that pressure effects on the

dehydrogenation and isomerization rate constant may be correlated with

hydrocarbon partial pressure. Figure 21 showed the effect of hydro

carbon partial pressure on the rate constants at a constant hydrogen

partial pressure. Since it has been shown that a variation in hydrogen

partial pressure does not significantly affect the rate constants, this

plot has been extended to include all the pressure data, regardless of

hydrogen partial pressure. Figure 24 shows both rate constants plotted

as a function of hydrocarbon partial pressure for all of the pressure

data on the Pt-AlgOg -mordenite catalyst at 775°F. As with Figure 21,

Figure 24 indicates that both rate constants decrease with increasing

hydrocarbon partial pressure. This effect seems to diminish at the

higher levels of hydrocarbon partial pressure.

I . Correlation of the Cyclohexane Dehydrogenation and Isomerization

Rate Constants at 775?F with the Observed Pressure Effects -

Pt-AL,03 -Mordenite Catalyst

1. Intjcoduction

During the past twenty-five years, it has become popular to

analyze the kinetic data for solid catalyzed gas phase reactions in terms

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Co

A 6.0 LEGEND 5.0 Isomerization, kg Dehydrogenation, Q 1^ 4.0 3.0 p0 , p0Hydrocarbon , Partial Pressure, psia Run Nos. 33,35,41A,B,C,D,E,G,44 2.0

0.0 0.02 i t o e B 00 W o u & DO 0.04 B cd B td c c 0.06 4J r—I The Effect Total of HydrocarbonPartial Pressure on Cyclohexanethe Dehydrogenation and Figure 4. isomerizationRate Constants at 775°F Pt-Al^-Mordenite - Catalyst.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 174

(3 5 ) of the Langmuir-Hinshelwood adsorption theory. The application of

the Langmuir adsorption theory to the adsorption-desorption and surface

reaction steps in heterogeneous catalysis has been discussed in detail

v „ . tt *_ (20) . _ . (14,15) by Hougen and Watson and Corrigan.

The adsorption phenomena that occur in gas-phase heterogeneous

catalytic reactions involve gas mixtures rather than one gas alone. The (21) Langmuir theory for simple unimolecular adsorption can be extended (8) to describe adsorption of gas mixtures by solid surfaces. The

assumptions necessary to extend the Langmuir adsorption isotherm to cover (35 8} mixed adsorption are: * (a) molecules adsorb on the solid surface

in a unimolecular layer, and (b) no interaction occurs between adsorbed

molecules, which implies that the heat of adsorption is a constant and

independent of the amount of gas adsorbed. If the extended Langmuir

theory is applied, for example, to a binary mixture of gases, then the

amount of gas A that is adsorbed from the mixture of A and B is given (8) by an expression of the form

*... ■ — = S|,‘ . O) 1 + Pa + Kb pa

where

Aftds = quantity of gas A adsorbed by the solid surface,

pA = partial pressure of component A,

pB = partial pressure of component B,

Ka - adsorption coefficient of component A,

Kg = adsorption coefficient of component B,

a - quantity of gas A necessary to completely cover the

solid surface with a monolayer.

Equation (7) predicts that the addition to one gas of a second gas will

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 175

always decrease the amount of the first gas adsorbed. Enhancement of

adsorption of one gas by another cannot be explained on the basis of the

Langmuir theory.

If the Langmuir adsorption theory is applied to a solid cata

lyzed first order reversible reaction in which the surface reaction is

rate controlling, then the rate of reaction is given by an expression

of the f o r m ^ ^

« (Pa “ Pb /K) '"(iVtTJ~ 1 <8)

where

r = reaction rate,

kg = a constant dependent on temperature and the catalyst,

A = refers to a reactant,

B = refers to a product,

i = refers to a reactant, product or an inert,

K = adsorption coefficient,

p = partial pressure,

K = reaction equilibrium constant, ,

n = a constant dependent on the reaction mechanism.

Expressions having the general form of equation (8) have often

been used for the successful correlation of kinetic data from solid

catalyzed reactions. This general theory, however, is not without

certain limitations. Although the Langmuir theory always predicts a

decrease in adsorption when one gas is added to another, numerous (8) examples of physical adsorption have been cited by Brunauer where an (35) enhancement of adsorption is obtained. Weller states that adsorp

tion of this type corresponds to having a negative adsorption

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 176

coefficient in the denominator of the Langmuir equation. In the usual

application of the Langmuir approach to kinetic data, this result is

considered to be physically impossible and any reaction mechanism which

leads to a negative adsorption coefficient is automatically discarded.

Another difficulty in the application of the adsorption theory

to kinetic data concerns the problem of finding the correct reaction

mechanism. The rate equation illustrated by equation (8) assumes that

the surface reaction is rate controlling with the other steps on the

catalyst surface being at equilibrium. Other assumed mechanisms result

in different equations. The occurrence of a better fit of the data to

one equation than to another is frequently employed as a sufficient

criterion of the reaction mechanism. As discussed previously, this

proceedure has received criticism, not for its ability to correlate the

kinetic data, but for reaching absolute conclusions about the mechanism

from the basis of best fit o n l y / 35^ The statistical analyses by

C h o u ^ ^ on the codimer example in Hougen and Wa t s o n ^ ^ illustrates

the difficulty in finding the correct kinetic mechanism from kinetic

data, even in very carefully conducted experimental programs.

Levenspiel^^ suggests that it is hardly ever possible to determine

from kinetic measurements alone which is the correct mechanism.

Thus, the kinetics investigator in heterogeneous catalysis is

faced with either using the Langmuir adsorption approach with its

recognized limitations or resorting to an empirical method such as the (351 one suggested by Weller. In spite of the difficulties inherent in

the Langmuir adsorption method, Boudart^^ indicates that the limitations

are often more than compensated for by the added insight into the reac

tion mechanism which it can provide without undue complexity. Another

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 177

argument in favor of the general Langmuir adsorption approach is that if

a mechanism is found that reasonably represents the data, extrapolation (22) to new operating conditions is more safely done.

The general philosophy used in this investigation has been to

correlate the observed pressure dependency of the reaction rate constants

within the framework of the general Langmuir adsorption model as presented

by Hougen and Watson. The restrictions and limitations of this

approach are recognized and where simplifications in the equations could

be made without unduly compromising the theory, they have been made.

2. Single and Dual Site Surface Reaction Adsorption Models

Simple reversible unimolecular or unimolecular-bimolecular

surface reactions such as the ones studied in this investigation generally

may proceed by either single or dual site mechanisms. The single site

mechanism postulates that only one active site is involved per reactant

molecule. The reactant molecule is adsorbed on the single site and

undergoes surface reaction to form the product(s) . If the reaction is

unimolecular, the product molecule remains on the single site until it is

desorbed. If the reaction is unimolecular-bimolecular, one of the two

products is released directly from the site upon reaction. The other

product molecule remains on the single site until it is desorbed. The

condition of neighboring active sites does not affect the reaction rate

predicted by the single site mechanism.

The dual site mechanism postulates that an adsorbed reactant

molecule can undergo surface reaction only if it is on an active site

that has a neighboring unoccupied site that is of the same type. For a

unimolecular reaction, the dual site surface reaction proceeds by a com

plex formed by the reactant molecule and the two adjacent, active sites.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 178

This complex decomposes to form an adsorbed product molecule on one of

the sites which is then desorbed. If the reaction is unimolecular-

bimolecular, the dual site mechanism also predicts the formation of a

complex between the reactant molecule and both active sites. After

reaction occurs, the two neighboring sites each have adsorbed one of the

two product molecules which are then subsequently desorbed from each

site.

The differential rate equations given in Chapter V for the

surface reaction controlled rates of dehydrogenation and isomerization

of cyclohexane are

i=klPc-k'pBpHJ (9)

and

dWc ° ^ P c ~ Hb'pm - <10>

where

Nb = gm moles benzene/min,

= gm moles methylcyclopentane/min,

W c = weight of catalyst, gm

Pc» Pb » Ph 2 » Pm ~ partial pressures of cyclohexane, benzene, hydrogen,

and methylcyclopentane, atm,

kj, kj = forward and reverse rate constants for dehydrogenation,

gm moles/gm cat-atm-min,

kg, k^ = forward and reverse rate constants for isomerization,

gm moles/gm cat-atm-min.

If the reverse rate constants are eliminated by introducing the defini

tions of the equilibrium constants, equations (9) and (10) become

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 179

dNa „ . = k* (PC - PbPh23/Ka) (11) c

and

a w c = ^3 (Pc " R i /^b ) (I2)

where

K* = equilibrium constant for the dehydrogenation of cyclohexane

to benzene,

Ka = equilibrium constant for the isomerization of cyclohexane

to methylcyclopentane.

The observed or simple forward reaction rate constants in

equations (11) and (12) can be related to the adsorption model rate

constants.For the dehydrogenation reaction, the relationship

between the simplified forward rate constant and the constants of the

single site model is

kolKc ^ = (1 + Kcpc + KaPa + ^ + ^ sPh2) (13)

The relationship between the simplified dehydrogenation rate constant

and the constants of the dual site model is

Kdi ^c______^ = (1 + Kcpc + KaPa +• Kh -P^, + K„*, 2 prH„ 2s,)s ' (14)

Similarly, the simplified isomerization rate constant is

related to the constants of the single site model by

lc, = ------J p s ik .-.------, (15) (1 + KcPc + ^ p B + + Kh2Ph2)

and to the constants of the dual site model by

kjj = ----- . (16) ' (1 + Kcpc + KaPa + ^.Ph + ^ 3Ph2)3

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 180

3. Simplification of the Surface Reaction Adsorption Models

As discussed previously, in an integral reactor such as the one

used in this investigation, the partial pressures of the reactants and

products vary with reactor length. Although the individual partial

pressures vary, the sum of the cyclohexane, benzene, and methylcyclo-

pentane partial pressures or total hydrocarbon partial pressure remains

constant with respect to reactor length. Also, the high hydrogen feed

diluent ratios used ensure that the hydrogen partial pressure remains

nearly constant with respect to reactor length.

When the differential rate equations for cyclohexane dehydro

genation and isomerization were integrated with respect to catalyst

bed length in Chapter V, it was assumed that the simplified forward rate

constants were independent of composition. Equations (13), (14), (15),

and (16) which relate the simplified rate constants to the adsorption

model constants would violate this assumption if used in their present

form since they are functions of the individual partial pressures.

These equations, however, may be simplified by certain reasonable

assumptions to a form which is not concentration-dependent and which

still retains the essential features of the adsorption model.

If it is assumed that all of the hydrocarbon adsorption constants

in equations (13), (14), (15), and (16) are essentially equal, then

these equations are reduced to the form

k, = ------V * .KP_ , (17) (1 + %V0 +Kh2Ph2)“ .

where

i = 1 for dehydrogenation, 2 for isomerization,

Ko <= hydrocarbon adsorption coefficient,

w V* ^ v v

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 181

p0 = total hydrocarbon partial pressure,

Kh = hydrogen adsorption coefficient,

pHg «= hydrogen partial pressure,

n = 1 for a single site model, 2 for a dual site model.

The assumption of equal adsorption coefficients for cyclohexane and

methylcyclopentane seems to be perfectly reasonable since these two

compounds are geometric isomers. In terms of individual adsorption

measurements, the adsorption coefficient of benzene might be expected

to be different from those of the two saturated hydrocarbons. There

is evidence, however, which indicates that the adsorption characteristics

of reacting systems described by Langmuir models are quite different

from the adsorption characteristics as determined from.individual

adsorption measurements. In a study of solid catalyzed pentane hydro- (24) isomerization, Lyster found that the kinetically determined adsorp

tion coefficient of hydrogen was only 30% larger than that for i-pentane.

This difference is much less than would be expected from single component

adsorption measurements. B o u d a r t ^ and S c h w a b ^ ^ point out that the

adsorption constants determined in reacting environments usually do not

agree with those obtained in separate experiments. This lack of

correspondence in adsorption characteristics undoubtedly reflects

deficiencies in the assumptions of the Langmuir adsorption theory when

-applied to real surfaces and mixtures of gases. In light of the uncer

tainties in application of the Langmuir adsorption theory to reacting

systems it is felt that the assumption of nearly equal adsorption

coefficients for the hydrocarbon components is not a serious limitation.

Previously, it was observed that variations in hydrogen partial

pressure had no significant effect on the simplified rate constants for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 182

the dehydrogenation or isomerization of cyclohexane on the Pt-Alg03 -

mordenite catalyst. This indicates that the adsorption coefficient-

partial pressure product for hydrogen is much less than that for the

hydrocarbons or Pn «K q P0 . In light of this observation, equation 2 2 (17) reduces to

t --- . (is) (1 + ^,p0)"

Equation (18) relates the simplified rate constant for dehydro

genation or isomerization to the adsorption coefficient of the hydro

carbon components and the total hydrocarbon partial pressure. This

equation assumes that the adsorption coefficients of the hydrocarbon

components are essentially equal and that the effect of hydrogen adsorp

tion on the rate constants is negligible. Since the total hydrocarbon

partial pressure does not vary with catalyst bed length, equation (18)

is independent of composition and satisfies the assumption made in

Chapter V.

4. Correlation of Experimental Results with the Simplified Surface

Reaction Models

Equation (18) which relates the simplified rate constants and

the adsorption characteristics can be rearranged to a linear form. For

a single site model, equation (18) may be rearranged to give

+ ■ (19>

For a dual site model equation (18) may be rearranged to give

+ • (20)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 183

The experimental data shown in Tables 19, 20, 21, and 22 giving

the effect of pressure variations on the rate constants for the Pt-AlgOg-

mordenite catalyst at 7 7 ^ F have been used to determine the constants in

equations (19) and (20) . These constants were determined by a least

squares analysis of the experimental data. A summary of the least

squares rate and adsorption constants for each reaction and each model

is given in Table 23.

The data in Table 23 shows that positive rate and adsorption

constants were obtained for each reaction and each model. The multiple

correlation coefficients indicate that statistically, both single and

dual site models describe the data equally well. The multiple corre

lation coefficients also show that the adsorption models explain about

90% of the variation in the rate constant-pressure data. The remaining

10% of the variation may be attributed to experimental error and this

figure agrees with the standard deviations obtained in the reproduci

bility study.

A ratio of the rate constants for dehydrogenation and isomeri

zation in Table 23 (either model) indicates that the dehydrogenation

rate of cyclohexane is about five times faster than the isomerization

rate for the Pt-ALjOg-mordenite catalyst at 775°F. The hydrocarbon

adsorption coefficients calculated from dehydrogenation or isomerization

rates (either model) show very good agreement. It can be concluded that

the simplified adsorption model adequately describes the variation of the

rate constants with pressure within the accuracy of the experimental

data.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 184

0.896 0! 107 0! 0.162 Dual Site JS(Y„,-Y)S 0.873 0.899 0.115 0.362 0.841 0.0566 Dual Site Single Site - )2 )2

)3 Dehydrogenation Isomerization^ Single Site r Variation (Total \ ** ** 1 0.288 0.406 q ^Correlation excludes run41C. Correlation Coefficient, ^ * Variation /(Explained k,,! Multiple 0.871 K Summary of Rate and Adsorption Constant for Dehydrogenation the and Isomerization of Table Z3. over Pt-ALjO^-Mordenite the cycl.0hexane Catalyst at 775?F.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 185

J. Effectiveness of the Overall Model for the Pt-AlgOg-Mordenite

Catalyst

A mathematical model for an experimental process has value only

Insofar as it is able to predict the result that would be obtained if

actual experimentation were performed. The effectiveness of the mathe

matical model developed for the simultaneous dehydrogenation and isom

erization of cyclohexane over the Pt-AlgOg -mordenite catalyst has been

examined by comparing the calculated and experimental values of the

rate constants. Values of the dehydrogenation and isomerization rate

constants were calculated from the model for the complete range of

variation in space time, temperature, and pressure. Predicted values of

the rate constants were compared to the experimental values by calcu

lating the percent difference between them. The average percent errors

between the experimental and model predicted values for the dehydrogena

tion and isomerization rate constants are shown in Table 24. The average

percent error between experimental and calculated values for the dehydro

genation rate constant for the total model is 8.4% while that for the

isomerization rate constant is 10.57». These results indicate that the

agreement between experimental and calculated values is good.

It was assumed in the development of the mathematical model

for the simultaneous dehydrogenation and isomerization of cyclohexane

that the surface reactions were rate controlling. Since the percent

errors between calculated and experimental values of the rate constants

are about equal to the normal scatter of the data, it can be concluded

that this model gives a satisfactory representation of the data.

Although it cannot be concluded absolutely that this model represents

the controlling mechanism, the agreement of experimental and calculated

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 186

T hi 9 A. Effectiveness of the Overall Model for the a 6 ' Pt-AlgOj -Mordenite Catalyst

Average % Error Between Experimental and Calculated Rate Constant Dehydrogenat ion Isomerization Correlation k. ka

Space Time 10.0 13.5

Temperature 0.8 2.6

Pressure^ 8.1 . 9.3

Total Model 8.4 10.5

experimental calculated ik. x 1 0 0 , E l ^experimental______Average % Error = Number of Comparisons

^Predicted rate constants for both reactions use a single site model.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 187

rate constants makes it highly probable. If a more satisfactory physi

cal model did exist, it probably could not be distinguished from this

one due to the normal experimental error in the data.

K. Reactivity of n-Heptane and Methylcyclopentane over the Pt-AlgO, -

Mordenite Catalyst at 925°F and 305 psia

A limited amount of experimental data was obtained for n-heptane

and methylcyclopentane over the Pt-Alg03 -mordenite catalyst at 92^ ’f

and 305 psia. The experimental results for n-heptane at these conditions

are shown in Table 25. The space time was varied between 1.51 and 6.04

gm cat-min/gm feed and the n-heptane conversion ranged from 49.8 to

98.1 wt. %. At these conditions with n-heptane, hydrocracking was the

predominant reaction with very little dehydrocyclization to toluene

taking place. The degree of selectivity to hydrocracking was 98% or

above for all the experimental runs shown in Table 25. (23) Lyster has presented a fairly complete survey of the literature

on proposed mechanisms for the dehydrocyclization of n-heptane to toluene.

Most mechanisms presented in this survey envision the first step in the

dehydrocyclization reaction as the dehydrogenation of n-heptane to a

heptene molecule. The heptene molecule is thought to desorb from the

catalyst surface and readsorb at the double bond in such a manner that

ring closure to methylcyclohexane is favored. The methylcyclohexane

subsequently dehydrogenates to form toluene. The step involving ring

closure is usually viewed as rate controlling. The high degree of

hydrocracking activity exhibited by the Pt-AlgO^ -mordenite catalyst

indicates that at these conditions the hydrocracking rate is much higher

than the rate of ring closure. The hydrocracking reaction may be

moderated by decreasing the temperature, decreasing the acidity of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99.5 98.0 98.5 98.5 Selectivity to Hydrocracking, % ~5 305 925 n-Heptane 0.147-0.833 Pt-AlgG^ -Mordenite Pt-AlgG^ Toluene 7 Cs “ C 76.45 76.45 22.35 0.86 97.66 1.92 0.28 68.99 29.96 0.74 48.88 50.17 0.44 Product Composition, wt. 7„ the Pt-ALgQg-Mordenite Catalyst at 925°F and 305 psia. SpaceTime 9, Detailed rundata are givenAppendix in B. Catalyst Size Range, mm Temperature, °F Pressure, psia Mole Ratio, Hg/n-Heptane Catalyst Feed 1C 1C 6.04 IB IB 1A 1.51 2.58 ID ID 2.58 Run No. gm cat-min/gm feed Experimental Results Operating Conditions Table 25. Reactivity of n-Heptane over

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 189

catalyst or addition of an Inhibiting agent such as nitrogen. All of

these actions may also decrease the rates of the desired reactions and ( 38} care must be exercised in their uses. The patent on which the

Pt-ALg03 -mordenite catalyst used in this investigation is based, claims

a beneficial effect from the addition of nitrogen. Presumably the nitro

gen acts to suppress hydrocracking. The effect of nitrogen addition on

the dehydroisomerization of methylcyclopentane has been investigated and

is discussed in the following paragraphs.

The experimental results for the dehydroisomerization of methyl

cyclopentane at 9 2 ^ F and 305 psia over the Pt-Alg03 -mordenite catalyst

are shown in Table 26. The space time was varied between 10.6 and 21.3

gm cat-tnin/gm feed and the methylcyclopentane conversions ranged from

43.8 to 81.2 wt. %. Experimental runs with and without addition of 200

ppm nitrogen to the methylcyclopentane feed were made at each of the

three space times investigated.

The data in Table 26 show that the addition of nitrogen at constant

space time decreases the conversion of methylcyclopentane and increases •> the selectivity to benzene. This action is undoubtably due to the

inhibiting effect of nitrogen on the hydrocracking reaction. At the low

space time, the addition of 200 ppm nitrogen causes about a 30% reduction

in conversion and a 507» increase in selectivity to benzene. At the

highest space time, however, addition of 200 ppm nitrogen causes only

about a 13% decrease in conversion while the selectivity to benzene is

nearly doubled. The benzene selectivity for the runs with no nitrogen

shows a tendency to decrease slightly with increasing conversion. The

benzene selectivity for the runs with the addition of nitrogen, however,

/ l g \ increases with increasing conversion. Heinemann et al in a study

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

66.4 35.6 60.5 35.3 58.5 40.2 Benzene, % Selectivity to

28.94 47.31 38.86 27.37 25.56 25.06 ~5 305 Benzene 925

0.147-0.833 Pt-Al^Og -Mordenite Pt-Al^Og Me thylcyclopentane Me MCP 18.79 28.56 24.63 36.24 37-69 56.19 Partial Product Composition, Wt. % ZL 15.77 16.38 39.23 35.62 11.57 24.80 3

21.3 21.3 17.0 17.0 10.6 10.6 gm feed gm cat-min/ SpaceTime 9,

0 0 0 200 200 200 PPm in Feed, in Nitrogen^ Pt-AI^Oa -Mordenite Catalyst Pt-AI^Oa at 925^F and 305 psia. Reactivity Methylcyclopentaneof with and without NitrogenAddition over the Catalyst Size Range, mm Temperature, °F Pressure, psia MoleRatio, Hj/Methylcyclopentane Catalyst Feed Experimental Results Operating Conditions As n-butylamine. Detailed run data are given Appendix in B. a 3B 3G 3F 3C 3D 3E RunNo. Table 26.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 191

of methylcyclopentane dehydroisomerization with a Houdriforming catalyst

at 950PF and 315 psia found that the benzene selectivity increased with

increasing conversion. It must be concluded that the Pt-AlgOt,-mordenite

catalyst is an active catalyst for dehydroisomerization of methylcyclo

pentane if nitrogen is added to suppress hydrocracking.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. L IS T OF REFERENCES - CHAPTER V I

1. Andreev, A. A., Shopov, D. M . , and Kiperman, S. L., "Kinetics of Cyclohexane Dehydrogenation in a Gradientless System," Kinet. Ratal. (U.S.S.R.), ]_ (No. 6), 1092 (1966): Chemical Abstracts, 6 6 , 85291 (1967).

2. Aris, R., "On Shape Factors .for Irregular Particles, I. - The Steady State Problem. Diffusion and Reaction," Chemical Engineering Science, j>, 262 (1957) .

3.- Balandin, A. A. and Rubinstein, A. M . , "The Effect of Methyl Substitution on the Catalytic Dehydration of Six-Membered Cyclic Hydrocarbons," Z. Physik Chem (U.S.S.R.), A167, 431 (1934); Chemical Abstracts, 28, 2255 (1934).

4. Barnett, L. G., Weaver, R. E. C., and Gilkeson, M. M., "Effect of Mass Transfer on Solid-Catalyzed Reactions: The Dehydrogenation of Cyclohexane to Benzene," American Institute of Chemical Engineering Journal. 1_ (No. 2), 211 (1961).

5. Bird, R. B., Stewart, W. E., and Lightfoot, E. N., Transport Phenomena, John Wiley and Sons, New York, New York, 1962.

6 . Boudart, M . , "Kinetics on Ideal and Real Surfaces," American Institute of Chemical Engineering Journal, £ (No. 1), 62 (1956).

7. Bridges, J. M. and Houghton, G., "The Evaluation of Activation Energies Using a Rising Temperature Flow Reactor. The Dehydrogenation of Cyclohexane over WS2 , Pt/Al203 , CrgOg/AlgOg, NiW04 /ALj03 and Cr2 03 ," Journal of the American Chemical Society, 81, 1334 (1959).

8 . Brunauer, S., The Adsorption of Gases and Vapors, Volume I, Physical Adsorption, Princeton University Press, Princeton, New Jersey, 1943.

9. Carberry, J. J., "Mass Diffusion and Isothermal Catalytic Selectivity," Chemical Engineering Science, 17, 675 (1962) .

10. Carberry, J. J., "The Catalytic Effectiveness Factor Under Nonisothermal Conditions," American Institute of Chemical Engineering Journal, 7 (No. 2), 350 (1961).

11. Carberry, J. J., "The Micro-Macro Effectiveness Factor for the Reversible Catalytic Reaction," American Institute of Chemical Engineering Journal, 8 (No. 4), 557 (1962).

192

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 193

12. Chou, C., "Least Squares," Industrial and Engineering Chemistry, 50 (No. 5), 799 (1958).

13. Ciapetta, F. G., "Special Report - Catalytic Reforming," Petro/ Chero Engineer, 33 (No. 5), C-19 (1961).

14. Corrigan, T. E., "Chemical Engineering Fundamentals. Catalytic Vapor Phase Reactions - I," Chemical Engineering, 62 (No. 1), 199 (1955).

15. Corrigan, T. E., "Chemical Engineering Fundamentals, Catalytic Vapor Phase Reactions - II," Chemical Engineering, 62 (No. 2), 195 (1955).

16. Corrigan, T. E., "Chemical Engineering Fundamentals, Interpre tation of Kinetic Data - I," Chemical Engineering, 62 (No. 4), 199 (1955).

17. Erkelens, J., Rozendaal, A., Linsen, B. G., and Okkerse, C., "Elimination of Diffusional Effects in the Study of Catalytic Reactions in Flow Systems," Chemistry and Industry, 2159 (December 30, 1967).

18. Heinemann, H., Mills, G. A., Hattmann, J. B., and Kirsch, F. W., "Houdriforming Reactions - Studies with Pure Hydrocarbons, Industrial and Engineering Chemistry, 45 (No. 1), 130 (1953).

19. Hopper, J. R . , "A Study of the Catalytic Hydroisomerization Reactions of n-Pentane and Cyclohexane over Structurally Modified Mordenites," Ph.D. Dissertation, Department of Chemical Engineering, Louisiana State University, 1969.

20. Hougen, 0. A. and Watson, K. M . , Chemical Process Principles, Part III, Kinetics and Catalysis, John Wiley and Sons, New York, New York, 1947.

21. Langmuir, I., "The Adsorption of Gases on Glass, Mica, and Platinum," Journal of the American Chemical Society, 4 0 , 1361 (1918).

22. Levenspiel, 0., Chemical Reaction Engineering, John Wiley and Sons, New York, New York, 1962.

23. Lyster, W. N., "Kinetics of Chemical Reactions - Dehydrocycli zation of N-Heptane," Ph.D. dissertation, Department of Chemical Engineering, Univeristy of Houston, 1964.

24. Lyster, W. N . , Hubbs, J. L., and Prengle, H. W., Jr., "Isomeriza tion of n-Pentane over Platinum Alumina Catalysts of Different Activity," American Institute of Chemical Engineering Journal, 10 (No. 6), 907 (1964).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 194

25. P'ang, L. et al, "Kinetics of Cyclohexane Dehydrogenation on Platinum Reforming Catalysts by the Flow-Circulating Method," Acta Foculio-Chim. Sinica (China), (5 (No. 2), 175 (1965); Chemical Abstracts, 65, 18379 (1966).

26. Pickert, P. E., Bolton, A. P., and Lanewala, M. A., "Molecular Sieve Zeolites: Trendsetters in Heterogeneous Catalysis," Chemical Engineering, 75 (No. 16), 133 (1968).

27. Popescu, A., Negoita, N . , and Baiulescu, E., "Activity of Pt-AlgC^ Catalysts," Analele Univ. Bucuresti, Ser. Stiint Nat. (Romania), JjL, 137 (1963); Chemical Abstracts, 64, 12571 (1966) .

28. Ritchie, A. W. and Nixon,.A. C., "Dehydrogenation of Monocyclic Naphthenes over a Platinum on Alumina Catalyst without Added Hydrogen," American Chemical Society Preprints. Division of Petroleum Chemistry, 11 (No. 3), 93 (1966).

29. Satterfield, C. N. and Sherwood, T. K., The Role of Diffusion in Catalysis, Addison - Wesley Publishing Company, Reading, Mass., 1963.

30. Schwab, G. M . , "About the Mechanism of Contact Catalysis," Advances in Catalysis, 2, Academic Press, Inc., New York, New York, 1950.

31. Sinfelt, J. H . , "Bifunctional Catalysis," Advances in Chemical Engineering, 5, 37, Academic Press, New York, New York, 1964.

32. Thiele, E. W., "Relation Between Catalytic Activity and Size of Particle," Industrial and Engineering Chemistry, 31 (No. 7), 916 (1939).

33. Weisz, P. B. and Prater, C. D., "Interpretation of Measurements in Experimental Catalysis," Advances in Catalysis, 6, 143, Academic Press, Inc-., New York, New York, 1954.

34. Weisz, P. B. and Swegler, E. W., "Effect of Intra-Particle Diffusion on the Kinetics of Catalytic Dehydrogenation of Cyclohexane," Journal of Physical Chemistry, 59, 823 (1955).

35. Weller, S., "Analysis of Kinetic Data for Heterogeneous Reactions," American Institute of Chemical Engineering Journal, (No. 1), 59 (1956).

36. Wheeler, A., "Reaction Rates and Selectivity in Catalyst Pores," Advances in Catalysis, 3^, 249, Academic Press, Inc., New York, New York, 1951.

37. Zelinskii, N. D. and Balandin, A. A., "Kinetics of Dehydrogenation Catalysts," Bulletin of the Academy of Science (U.S.S.R.), 29 (1929); Chemical Abstracts, 24, 774 (1930).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 195

38. U. S. Patent 3,376,214.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER V I I

CONCLUSIONS AND RECOMMENDATIONS

A. Conclusions

A halogen-free Pt-ALgOa-mordenite catalyst has been investigated

for the simultaneous dehydrogenation and isomerization of cyclohexane

in an integral fixed-bed reactor system. These results on cyclohexane

are supplemented by a limited experimental study on a halogen-free

Pt-ALgOa catalyst. A mathematical model for the simultaneous dehydro

genation and isomerization of cyclohexane has been developed and

applied to the experimental data from these two catalysts to determine

the rate constants. The dehydrocyclization and dehydroisomerization

capabilities of the Pt-ALgOa-mordenite catalyst were also investigated

by limited experimentation on n-heptane and methylcyclopentane.

The results from these experimental investigations have been

summarized in the following paragraphs.

1. A study of macropore diffusion on the Pt-Al2 03 -mordenite

catalyst revealed that significant diffusional effects existed for

certain particle sizes in the dehydrogenation and isomerization of

cyclohexane between 775? F and 900? F. The rates of cyclohexane dehydro

genation and isomerization at 775?F were shown to be independent of

particle size below a certain critical size. Diffusion in the micro

pores of the mordenite was not evaluated since different crystallite

196

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 197

sizes were not available. The effects of micropore diffusion, if

they are present, are included in the overall reaction rate.

2. The mass transfer of cyclohexane from the bulk fluid phase

to the surface of the Pt-AlgOg-mordenite catalyst was shown to be

not rate-limiting for dehydrogenation or isomerization at 775°F.

3. The rate of cyclohexane dehydrogenation over the Pt-AlgOg

catalyst was shown to be dependent on catalyst particle size for all

of the particle sizes investigated at 775°F. The rate of cyclohexane

isomerization over the Pt-Al^Og catalyst was almost non-existent at

775°F and not dependent on particle size.

4. The rates of cyclohexane dehydrogenation and isomerization

over the Pt-AlgOg -mordenite catalyst at 775°F were shown to be

consistent with first-order reversible reactions.

5. Activation energies for cyclohexane dehydrogenation and

isomerization over the Pt-A^Og -mordenite catalyst were determined in

the 724°-775°F temperature range. The activation energy for cyclo

hexane dehydrogenation was found to be 14.4 kcal/gm mole which is

consistent with values reported in the literature over similar cata

lysts. The activation energy for cyclohexane isomerization was found

to be 12.5 kcal/gm mole. The isomerization activation appeared to be

low in comparison to values previously reported for mordenite cata

lysts and may indicate a diffusional limitation in the micropores of

the mordenite.

6 . The effect of pressure on the cyclohexane dehydrogenation

and isomerization rate constants at 775°F on the Pt-AlgOg -mordenite

catalyst may be summarized as follows:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 198

a. The dehydrogenation rate constant decreased with

Increasing total pressure at a fixed hydrogen to cyclohexane feed

diluent ratio. The isomerization rate constant showed no effect

of a change in total pressure at a constant hydrogen to cyclohexane

feed diluent ratio.

b. Both rate constants decreased with increasing hydro

carbon partial pressure at a constant hydrogen partial pressure.

c. At constant hydrocarbon partial pressure, the variation

in both rate constants with changing hydrogen partial pressure was

approximately equal to the normal scatter of the data.

7. The observed pressure effects on cyclohexane dehydrogenation

and isomerization over the Pt-ALgGlj -mordenite catalyst at 775°F were

adequately correlated by using the conventional Langmuir adsorption

approach. The experimental data from both reactions were found to be

compatible with either dual or single site surface reaction mechanisms

in which the dynamic adsorption coefficients of hydrocarbons are equal

and influence of hydrogen adsorption is negligible. An exact know

ledge of the reaction mechanism is not implied because of certain

limitations in the Langmuir adsorption theory when applied to mixtures

and reaction environments.

8 . The effectiveness of the mathematical model for cyclohexane

dehydrogenation and isomerization over the Pt-Al^Ck -mordenite was

evaluated by comparing the experimental and predicted rate constants

over the range of conditions investigated. The average percentage

error between the experimental and predicted values of the dehydro

genation rate constant was 8.4%. The average percentage error between

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 199

the experimental and predicted values of the isomerization rate

constant was 10.5%.

9. The cyclohexane isomerization activity of the Pt-AlgOg-

mordenite catalyst was shown to be approximately eighteen times that

of the Pt-AlaOg catalyst at 775°F. This difference in isomerization

promoting ability may be explained by presence of the mordenite which

was shown to be approximately twelve times more acidic than the

halogen-free alumina. These data confirm that the degree of catalyst

acidity is directly related to the isomerization activity.

10. The Pt-AlgOg -mordenite catalyst was found to have a negli

gible n-heptane dehydrocyclization activity at 925°F due to a high

rate of hydrocracking.

11. The Pt-AlgOg-mordenite catalyst was found to have a good

methylcyclopentane dehydroisomerization activity at 925°F if nitrogen

was added to the feed to suppress hydrocracking.

B. Recommendations

One of the most important and significant facets of any scientific

investigation is the recommendations for further research. Scientific

research is somewhat analogous to a chain reaction in which the results

of any single investigation may generate a whole series of interesting

topics for further evaluation. Three recommendations for further

research are made in the following paragraphs. Two of these recommen

dations are direct extensions of this investigation while the third

concerns measurements in heterogeneous catalysis in general.

This investigation has shown that isomerization activity may be

imparted to a platinum-alumina dehydrogenation catalyst by the

addition of a small quantity of a crystalline zeolite to the catalyst

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 200

matrix. In addition, it has been shown that the degree of isomeri

zation activity is directly related to the degree of acidity of the

catalyst. The use of a bifunctional platinum-alumina-mordenite could

have broad implications in commercial catalytic reforming due to the

high stability and non-corrosive nature of the zeolite component. It

is necessary, however, to establish a quantitative relation between

the degree of acidity and the isomerization activity so that the opti

mum balance between the dual functions may be obtained. Conceivably,

the acidity of the platinum-alumina-mordenite catalyst may be changed

by varying either the concentration of the mordenite or by altering

the structural characteristics of the mordenite.

The second recommendation concerns diffusional characteristics

within the micropores of the mordenite component. Certain results

obtained in this study indicate micropore diffusion may contribute

significant resistance to the rate of reaction in the absence of

macropore diffusional limitations. An obvious first step in the

evaluation of micropore diffusional effects is the investigation of

different crystallite sizes. In conjunction with this first step,

a satisfactory theory for the surface type diffusion that occurs within

the micropores must be developed.

The final recommendation for further investigation concerns the

conventional methods for describing reaction mechanisms in hetero

geneous catalysis and their possible improvements. The determination

of reaction mechanisms in heterogeneous catalysis is not just a

theoretical pursuit but can have far-reaching commercial implications.

Once specific reaction mechanisms are known, the problem of tailor-

making catalysts to fit any desired requirement is much less difficult.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 201

I ■ Two deficiencies in the conventional methods for describing reaction

rates over solid catalysts are obvious. The first deficiency concerns

the Langmuir single component adsorption theory and its application to

reacting systems. It has been repeatedly demonstrated that the

extension of this theory to multicomponent reacting systems often

violates the theories assumptions and makes interpretation of results

difficult. The second problem concerns concentration measurements in

solid catalyzed reactions and the actual events on the catalyst surface.

The limitations of present technology make it necessary to make

concentration measurements in the bulk fluid phase and extrapolate to

what happens on the catalyst surface. This approach may disguise the

actual kinetics so long as we do not actually follow events on the

catalyst surface.

Both of the aforementioned deficiencies in present catalysis

theory make it difficult or impossible to pinpoint exact reaction

mechanisms. The development of a suitable adsorption theory for

reacting environments and measurement of concentrations on catalyst

surface are formidable problems, theoretically and experimentally.

This author feels, however, that they should be suitable and prime

topics for further research in heterogeneous catalysis.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A

NOMENCLATURE

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX A

NOMENCLATURE

A - Reactor cross-sectional area

A - Representation of an arbitrary reactant

Aadg - Quantity of an arbitrary gas A adsorbed by a

solid surface

a - Quantity of an arbitrary gas A necessary to

completely cover a solid surface with a monolayer

a - Initial concentration of component A

a - Unit cell dimension for mordenite

at - Designation of an arbitrary base point in the

pattern search technique

B - Representation of an arbitrary product

Bi - Exponential coefficient in the solution for the

concentration of benzene

B2 - Exponential coefficient in the solution for the

concentration of benzene

Rj - Constant term in the solution for the concen

tration of benzene

b - Initial concentration of component B

b - Unit cell dimension for mordenite

C - Representation of an arbitrary product

c - Initial concentration of component C

203

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 204

c - Unit cell dimension for mordenite

d — Differential operator

dp - Average particle diameter

E - . Activation energy

e - Base of natural logarithms

F - Mass feed rate of cyclohexane

f — Denotes general functional relation

fj - Denotes i*^ functional relation

£' - Denotes general functional realtion

g - Denotes general functional relation

g' - Denotes general functional relation

H - Function of the diluent ratio and total

pressure

hjj — Temporary base point in the pattern search

technique

K - Thermodynamic equilibrium constant for an

arbitrary first-order reversible reaction

KA - Thermodynamic equilibrium constant for the

dehydrogenation of cyclohexane to benzene

Kg — Thermodynamic equilibrium constant for the

isomerization of cyclohexane to methylcyclo

pentane

Ka - Adsorption coefficient of an arbitrary gaseous

component A

Kg - Adsorption coefficient of an arbitrary gaseous

component B

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 205

Kb - ' Adsorption coefficient for benzene

Kg - Adsorption coefficient for cyclohexane

- Adsorption coefficient for hydrogen

Kj - Adsorption coefficient for component i

K,, - Adsorption coefficient for methylcyclopentane

K,) - Adsorption coefficient for cyclohexane, benzene,

and methylcyclopentane

k - Arbitrary reaction rate constant

k,j - Frequency factor

k„ - First-order forward rate constant in the

adsorption model for an aribtrary reaction

kQj - First-order forward rate constant in the

adsorption model for the dehydrogenation of

cyclohexane to benzene

ko3 - First-order forward rate constant in the

adsorption model for the isomerization oc

cyclohexane to methylcyclopentane

kj^ - First-order forward rate constant for the

dehydrogenation of cyclohexane to benzene

k^ - First-order reverse rate constant for the

dehydrogenation of cyclohexane to benzene

kg - First-order forward rate constant for the

isomerization of cyclohexane to methylcyclo

pentane

k^ - First-order reverse rate constant for the

isomerization of cyclohexane to methylcyclo

pentane

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 206

k| - First-order forward rate constant for the

reaction of A to B

k_, - First-order reverse rate constant for the

reaction of A to B

k, , - First-order forward rate constant for the

reaction of A to C

k_, j - First order reverse rate constant for the

reaction of A to C

(ki^EST “ Estimate of the first-order forward rate

constant for the dehydrogenation of cyclohexane

to benzene

" Estimate of the first-order forward rate

constant for the isomerization of cyclohexane

to methylcyclopentane

L - Catalyst bed length

Mj^ - Exponential coefficient in the solution for the

concentration of methylcyclopentane

Mg - Exponential coefficient in the solution for the

concentration of methylcyclopentane

Mj - Constant term in the solution for the concen

tration of methylcyclopentane

m - Number of data points for the least squares

minimization

- Molar flow rate of benzene

l^c - Total molar flow rate of hydrocarbons

Nj - Molar flow rate of component i

- Molar flow rate of methylcyclopentane

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 207

Nt - Total molar flow rate

n A constant in the adsorption model dependent

on the reaction mechanism

P, - Exponential coefficient in the general solution

for the concentration of component B

Pj ( - Exponential coefficient in the general solution

for the concentration of component C

p - Number of simultaneous nonlinear equations

p^ - Partial pressure of an arbitrary gaseous

component A

Pp - Partial pressure of an aribtrary gaseous

component B

pB - Partial pressure of benzene

pc - Partial pressure of cyclohexane

p^ - Partial pressure of hydrogen

Pj - Partial pressure of component i

p^ - Partial pressure of methylcyclopentane

p0 - Total hydrocarbon partial pressure

Q( - Exponential coefficient in the general solution

for the concentration of component B

Qj t - Exponential coefficient in the general solution

for the concentration of component C

P. - Feed diluent ratio

R - Universal gas constant

Re,, - Particle Reynolds number

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 208

Rx - Residual difference between a predicted and an

experimental value of the benzene concentration

Rg - Residual difference between a predicted and an

experimental value of the methylcyclopentane

concentration

r - Multiple correlation coefficient

r - Reaction rate for an arbitrary first-order

reversible reaction

rB - Rate of reaction for the dehydrogenation of

cyclohexane to benzene

rH - Rate of reaction for the isomerization of

cyclohexane to methylcyclopentane

Tj - Root of the auxiliary equation in the general

parallel reaction solution

rj - Root of the auxiliary equation in the parallel

reaction solution for cyclohexane

r2 - Root of the auxiliary equation in the general

parallel reaction solution

r3 - Root of the auxiliary equation in the parallel

reaction solution for cyclohexane

T - Absolute temperature

t - Time

W - Mass flow rate

Wc - Weight of catalyst

x - Concentration of component B at any time

- Arbitrary independent variable

Xj - Arbitrary independent variable

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 209

Y - Experimental value of an arbitary dependent

variable

Yest - Predicted value of an arbitrary dependent % variable

Y - Mean value of an arbitrary dependent variable

y - Concentration of component C at any time

yB - Mole fraction of benzene

yc • - Mole fraction of cyclohexane

yH - Mole fraction of hydrogen s yl - Mole fraction of component i

yH - Mole fraction of methylcyclopentane

yB - • Mole fraction of benzene on a hydrogen-free

basis

y( - Mole fraction of component i on a hydrogen-free

basis

y^ - Mole fraction of methylcyclopentane on a

hydrogen-free basis

y - Phase designation of alumina

y i ] $ - Constant term in the general solution for the

concentration of component B

yil/jj - Constant term in the general solution for the

concentration of component C

A - Difference operator

Aj - Step size in the pattern search technique

6 - Void fraction

0 - Space time

|i - Viscosity

-V. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. tt — Total pressure

$ — General criterion function

$ - Sun of the squares of the residuals

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX B

DETAILED EXPERIMENTAL DATA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number LA

Feed n-Heptane

Catalyst Type Pt-ALjOg -Mordenite Code FJH-617-2RD Size, mm 0.147-0.833 Wt., gm 2.1042

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vo l ., cnP ~ 1 2

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 49.5 Hydrogen, psia 255.2

w/hr/w, gm feed/gm cat-hr 23.21

Feed Diluent, moles l^/mole feed 5.16

Feed Rate, cni3/hr 71.4

Minutes on Feed 270

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 76.45 i-Heptane 2.56 n-Heptane 19.79 Methylcyclohexane 0.00 Benzene 0.23 Toluene 0.86 Cg and Cg Aromatics 0.11

Hydrogen Balance, % 100.09

Space Time 0, gm cat-min/gm feed 2.580

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number IB

Feed n-Heptane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-2RD Size, mm 0.147-0.833 Wt., gm . 2.1042

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol.-, cnf3 ~ 1 2

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 50.6 Hydrogen, psia 254.1

w/hr/w, gm feed/gm cat-hr 39.79

Feed Diluent, moles Hj/mole feed 5.02

Feed Rate, ait3/hr 122.4

Minutes on Feed 400

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 48.88 i-Heptane 2.88 n-Heptane 47.29 Methylcyclohexane 0.00 Benzene 0.31 Toluene 0.44 C8 and Cg Aromatics 0.10

Hydrogen Balance, % 100.28

Space Time 9, gm cat-min/gm feed 1.510

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 1C

Feed n-Heptane

Catalyst Type Pt-AlgOa-Mordenite Code FJH-617-2RD Size, mm 0.147-0.833 Wt., gm 2.1042

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnt3 ~12

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 47.5 257.2 Hydrogen, psia i

w/hr/w, gm feed/gm cat-hr 9.95

Feed Diluent, moles Hg/mole feed 5.41

Feed Rate, cnP/hr 30.6

Minutes on Feed 560

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (CB-) 97.66 i-Heptane 0.36 n-Heptane 1.56 Methylcyc1ohexane 0.00 Benzene 0.14 Toluene 0.28 Ce and C9 Aromatics 0.00

Hydrogen Balance, % 99.29

Space Time 0, gm cat-min/gm feed 6.040

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number ID

Feed n-Heptane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-2RD Size, mm 0.147-0.833 Wt., gm 2.1042

Catalyst Diluent Mullite Size, ram 0.147-0.495 Vol., cnf3 ~12

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 48.9 Hydrogen, psia 255.8

w/hr/w, gm feed/gm cat-hr 23.21

Feed Diluent, moles Hg/mole feed 5.23

Feed Rate, cirP/hr 71.4

Minutes on Feed 680

Product Composition, wt. % (Hydrogen Free Basis)

Cracked Gas (C6 -) 68.99 i-Heptane 1.68 n-Heptane 28.28 Methylcyclohexane 0.00 Benzene 0.20 Toluene 0.74 Ce and Cg Aromatics 0.11

Hydrogen Balance, % 99.37

Space Time 0, gm cat-min/gm feed 2.580

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number

Feed Methylcyclopentane + 200 ppm Nitrogen*

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-2RE Size, mm 0.147-0.833 Wt., gm 10.6155

Catalyst Diluent None Size, mm Vol., cn?

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 51.0 Hydrogen, psia 253.7

w/hr/\-7, gm feed/gm cat-hr 2.82

Feed Diluent, moles I^/mole feed 4.98

Feed Rate, cnP/hr 40.8

Minutes on Feed 175

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 19.19 Methylcyclopentane 21.50 Methylcyclopentene 0.50 Cyclohexane 0.85 Cyclohexene 0.06 Benzene 50.79 Toluene 4.11 Ce and Cg Aromatics 3.00

Hydrogen Balance, % 99.11

Space Time 0, gm cat-min/gm feed 21.304

*As n-Butylamine.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 3B

Feed Methylcyclopentane + 200 ppm Nitrogen* Catalyst Type Pt-Al203 -Mordenite Code FJH-617-2RE Size, irnn 0.147-0.833 W t ., gin 10.6155 Catalyst Diluent None Size, mm Vol., cnf5

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 51.0 Hydrogen, psia 253.7

w/hr/w, gm feed/gm cat-hr 2.82

Feed Diluent, moles Hg/mole feed 4.98

Feed Rate,, cnf3 /hr 40.8

Minutes on Feed 290

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 15.77 Methylcyclopentane 28.56 Methylcyclopentene 0.81 Cyclohexane 1.13 Cyclohexene 0.20 Benzene 47.31 Toluene 1.64 Ce and Cg Aromatics 4.65

Hydrogen Balance, % 99.47

Space Time 0, gm cat-min/gm feed 21.304

*As n-Butylamine.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number

Feed Methylcyclopentane + 200 ppm Nitrogen

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-2RE Size, mm 0.147-0.833 Wt., gm 10.6155

Catalyst Diluent None Size, mm Vol., cm3

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 51.2 Hydrogen, psia 253.5

w/hr/w, gm feed/gm cat-hr 3.53

Feed Diluent, moles Hj/mole feed 4.96

Feed Rate, cn?/hr 51.0

Minutes on Feed 595

Product Composition, wt. % (Hydrogen Free Basis)

Cracked Gas (C6-) 16.38 Methylcyclopentane 36.24 Methylcyclopentene 0.64 Cyclohexane 1.35 Cyclohexene 0.22 Benzene 38.86 Toluene 2.31 C8 and Cg Aromatics 4.05

Hydrogen Balance, % 100.35

Space Time 0, gm cat-min/gm feed 17.001

*As n-Butylamine.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 3D

Feed Methylcyclopentane + 200 ppm Nitrogen*

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-2RE Size, mm 0.147-0.833 wt., gm 10.6155

Catalyst Diluent None Size, mm Vol., cnf3

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 50.5 Hydrogen, psia 254.2

'4 w/hr/w, gm feed/gm cat-hr 5.65

Feed Diluent, moles Hj/mole feed 5.03

Feed Rate, cn?/hr 81.6

Minutes on Feed 815

Product.Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 11.57 Methylcyclopentane 56.19 Methylcyclopentene 1.41 Cyclohexane 1.88 Cyclohexene 0.35 Benzene 25.56 Toluene 1.24 C8 and Cg Aromatics 1.79

Hydrogen Balance, % 100.32

Space Time 0, gra cat-min/gm feed 10.600

*As n-Butylamine.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 3E

Feed Methylcyclopentane

Catalyst Type Pt-AlgOs -Mordenite Code FJH-617-2RE Size, mm 0.147-0.833 Wt., gm 10.6155

Catalyst Diluent None Size, mm Vol., cnP

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 45.9 Hydrogen, psia 258.8

w/hr/w, gm feed/gm cat-hr 5.65

Feed Diluent, moles llg/mole feed 5.12

Feed Rate, cm3 /hr 81.6

Minutes on Feed 965

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 24.80 Methylcyclopentane 37.69 Methylcyclopentene 1.07 "Cyclohexane 2.18 Cyclohexene 0.94 Benzene 25.06 Toluene 3.06 Ca and Cg Aromatics 5.21

Hydrogen Balance, % 100.33

Space Time 0, gm cat-min/gm feed 10.600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 3F

Feed Methylcyclopentane

Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-2RE Size, iron 0.147-0.833 Wt., gm 10.6155

Catalyst Diluent None . Size, mm Vol., cnP

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 50.4 Hydrogen, psia 254.3

w/hr/w, gm feed/gm cat-hr 3.53

Feed Diluent, moles llg/mole feed 5.05

Feed Rate, cnf5/hr 51.0

Minutes on Feed 1190

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C5 -) 35.62 Methylcyclopentane 24.63 Methylcyclopentene 0.24 Cyclohexane 1.68 Cyclohexene 0.32 Benzene 27.37 Toluene 3.91 Ce and Cg Aromatics 6.21

Hydrogen Balance, % 100.50

Space Time 0, gm cat-min/gm feed 17.001

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 3G

Feed Methylcyclopentane

Catalyst Type Pt-ALjOs -Mordenite Code FJH-617-2RE Size, mm 0.147-0.833 Wt., gm 10.6155

Catalyst Diluent None Size, mm Vol., cnt3

Temperature, °F 925

Pressure, psia 304.7 Feed, psia 51.4 Hydrogen, psia 253.3

w/hr/w, gm feed/gm cat-hr 2.82

Feed Diluent, moles 1^/mole feed 4.93

Feed Rate, cnt3 /hr 40.8

Minutes on Feed 1355

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 39.23 Methylcyclopentane 18.79 Methylcyclopentene 0.70 Cyclohexane 1.44 Cyclohexene 0.93 Benzene 28.94 Toluene 4.49 Ca and Cg Aromatics 5.49

Hydrogen Balance, % 99.61

Space Time 0, gm cat-min/gm feed 21.304

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 7A

Feed Cyclohexane

Catalyst Type Pt-ALgt^ -Mordenite Code FJH-617-3RH Size, mm 0.147-0.833 Wt., gm 0.7068

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnt3 ~14

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.3 Hydrogen, psia 214.4

w/hr/w, gm feed/gm cat-hr 56.21

Feed Diluent, moles Hj/mole feed 20.80

Feed Rate, cnt3 /hr 51.0

Minutes on Feed 170

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 3.21 Methylcyclopentane 13.63 Methylcyclopentene 0.00 Cyclohexane 62.94 Cyclohexene 0.00 Benzene 19.99 Toluene 0.00 Ce and Cg Aromatics 0.22

Hydrogen Balance, % 100.09

Space Time 0, gm-cat-min/gm feed 1.067 Rate Constant kj , gm moles/gm cat-atm-min 0.00432 Rate Constant gm moles/gm cat-atm-min 0.00275

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 7B

Feed Cyclohexane

Catalyst Type Pt-AlgOj-Mordenite Code FJH-617-3RH Size, mm 0.147-0.833 Wt., gm 0.7068

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnt3 ~14

Temperature, °F 901

Pressure, psia 224.7 Feed, psia 10.3 Hydrogen, psia 214.4

w/hr/w, gm feed/gm cat-hr 33.84

Feed Diluent, moles Hg/mole feed 20.81

Feed Rate, cnf3/hr 30.7

Minutes on Feed 320

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 2.62 Methylcyclopentane 18.24 Methylcyclopentene 0.00 Cyclohexane 55.44 Cyclohexene 0.00 Benzene 22.99 Toluene 0.03 Cq and Cg Aromatics 0.67

Hydrogen Balance, % 100.13

Space Time 0, gm cat-min/gm feed 1.773 Rate Constant kj, gm moles/gm cat-atm-min 0.00317 Rate Constant kg, gm moles/gm cat-atm-min 0.00236

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 7G

Feed Cyclohexane

Catalyst Type Pt-ALg03 -Mordenite Code FJH-617-3RH Size, mm 0.147-0.833 W t . , gm 0.7068

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnt3 ~14

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.4 Hydrogen, psia 214.3

w/hr/w, gm feed/gm cat-hr 22.48

Feed Diluent, moles 1^/mole feed 20.70

Feed Rate, cut3 /hr 20.4

Minutes on Feed 495

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Cs -) 4.53 Methylcyclopentane 21.78 Methylcyclopentene 0.00 Cyclohexane 45.35 Cyclohexene 0.00 Benzene 27.77 Toluene 0.02 Cs and Cg Aromatics 0.53

Hydrogen Balance, % 98.64

Space Time 8 , gm cat-min/gm feed 2.669 Rate Constant kj, gm moles/gm cat-atm-min 0.00279 Rate Constant kg, gm moles/gm cat-atm-min 0.00206

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 8 A

Feed Cyclohexane

Catalyst Type Pt-ALgOa -Mordenite Code FJH-617-3RI Size, m m 0.147-0.833 Wt., gm 1.7042

Catalyst Diluent Mullite Size, mm 0.147-0.495 •Vol., cnf3

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.7 Hydrogen, psia 214.0

w/hr/w, gm feed/gm cat-hr 18.65

Feed Diluent, moles E^/mole feed 20.08

Feed Rate, cm3/hr 40.8

Minutes on Feed 180

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 4.61 Methylcyclopentane 18.60 Methylcyclopentene 0.00 Cyclohexane 26.51 Cyclohexene 0.00 Benzene 49.37 Toluene 0.03 Cg and Cg Aromatics 0.87

Hydrogen Balance, % 98.87

Space Time 8 , gm cat-min/gm feed 3.217 Rate Constant 1^ , gm moles/gm cat-atm-min 0.00499 Rate Constant kg , gm moles/gm cat-atm-min 0.00177

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 8 B

Feed Cyclohexane

Catalyst Type Pt-ALg03 -Mordenite Code FJH-617-3RI Size, mm 0.147-0.833 Wt., gm 1.7042

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnt3 ~13

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.4 Hydrogen, psia 214.3

w/hr/w, gm feed/gm cat-hr 27.97

Feed Diluent, moles l^/mole feed 20.70

Feed Rate, cnf3 /hr 61.2

Minutes on Feed 350

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.66 Methylcyclopentane 17.86 Methylcyclopentene 0.00 Cyclohexane 39.61 Cyclohexene 0.00 Benzene 41.13 Toluene 0.03 Cg and Cg Aromatics 0.71

Hydrogen Balance, % 99.68

Space Time 0, gm cat-min/gm feed 2.145 Rate Constant k^ , gm moles/gm cat-atm-min 0.00532 Rate Constant kg, gm moles/gm cat-atm-min 0.00217

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 8 C

Feed Cyclohexane

Catalyst Type P t - A L ^ -Mordenite Code FJH-617-3RI Size, mm 0.147-0.833 Wt., gm 1.7042

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cni3

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.5 Hydrogen, psia 214.2

w/hr/w, gm feed/gm cat-hr 37.30

Feed Diluent, moles Hg/mole feed 20.47

Feed Rate, cnf3/hr 81.6

Minutes on Feed 480

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 3.47 Methylcyclopentane 16.48 Methylcyclopentene 0.00 Cyclohexane 49.02 Cyclohexene 0.00 Benzene 30.85 Toluene 0.00 Ce and Cg Aromatics 0.17

Hydrogen Balance, % 100.03

Space Time 0, gm cat-min/gm feed 1.609 Rate Constant , gm moles/gm cat-atm-min 0.00487 Rate Constant kg, gm moles/gm cat-atm-min 0.00244

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 229

TABLE B

RUN DATA

Run Number 8D

Feed Cyclohexane

Catalyst Type Pt-Alg0 3 -Mordenite Code FJH-617-3RI Size, mm 0.147-0.833 Wt., gm 1.7042

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnP ~13

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.5 Hydrogen, psia 214.2

w/hr/w, gm feed/gm cat-hr 37.30

Feed Diluent, moles 1^/mole feed 20.47

Feed Rate, cut3/hr 81.6

Minutes on Feed 505

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Cs -) 2.63 Methylcyclopentane 16.27 Methylcyclopentene 0.00 Cyclohexane 49.81 Cyclohexene 0.00 Benzene 30.77 Toluene 0.01 Ce and C^ Aromatics 0.52

Hydrogen Balance, % 99.77

Space Time 0, gm cat-min/gm feed 1.609 Rate Constant kj^ , gm moles/gm cat-atm-min 0.00481 Rate Constant , gm moles/gm cat-atm-min 0.00238

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 9A

Feed Cyclohexane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-3RJ Size, mm 0.147-0.833 Wt., gm 2.5382

Catalyst Diluent Mullite Size, mm 0.147-0.495 V o l ., cnf3 ~12

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.4 Hydrogen, psia 214.3

w/hr/w, gm feed/gm cat-hr 18.78

Feed Diluent, moles 1^/mole feed 20.57

Feed Rate, cnt3/hr 61.2

Minutes on Feed 130

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Cs-) 4.23 Methylcyclopentane 15.01 Methylcyclopentene 0.00 Cyclohexane 8.20 Cyclohexene 0.00 Benzene 71.64 Toluene 0.04 C8 and Cg Aroma tics 0.88

Hydrogen Balance, % 99.88

Space Time 0, gm cat-min/gm feed 3.194 Rate Constant kj , gm moles/gm cat-atm-min 0.01140 Rate Constant , gm moles/gm cat-atm-min 0.00224

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 10A

Feed Cyclohexane

Catalyst Type Pt-ALgOg-Mordenite Code FJH-617-3RL Size, mm 0.147-0.833 Wt., gm 0.9202

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnt3 ~14

Temperature, °F 900

Pressure, psia 224.7 . Feed, psia 10.5 Hydrogen, psia 214.2

w/hr/w, gm feed/gm cat-hr 56.13

Feed Diluent, moles Hg/mole feed 20.42

Feed Rate, cnt3/hr 66.3

Minutes on Feed 125

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 4.19 Methylcyclopentane 12.88 Methylcyclopentene 0.00 Cyclohexane 61.26 Cyclohexene 0.00 Benzene 21.35 Toluene 0.00 Cs and Cg Aromatics 0.31

Hydrogen Balance, % 100.11

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant kj^ , gm moles/gm cat-atm-min 0.00461 Rate Constant k^> gm moles/gm cat-atm-min 0.00260

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 232

TABLE B

RUN DATA

Run Number 10B

Feed Cyclohexane

Catalyst Type Pt-AlgOg-Mordenite Code FJH-617-3RL Size, mm 0.147-0.833 Wt., gm 0.9202

Catalyst Diluent Mullite Size, mm 0.147-0.495 V o l ., ctiP ~14

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.5 Hydrogen, psia 214.2

w/hr/w, gm feed/gm cat-hr 56.13

Feed Diluent, moles /mole feed 20.42

Feed Rate, cut3 /hr 66.3

Minutes on Feed 165

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 3.74 Methylcyclopentane 10.65 Methylcyclopentene 0.00 Cyclohexane 63.29 Cyclohexene 0.00 Benzene 22.00 Toluene 0.00 C8 and Cg Aromatics 0.31

Hydrogen Balance, % 100.02

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant kj , gm moles/gm cat-atm-min 0.00467 Rate Constant kg , gm moles/gm cat-atm-min 0.00211

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 11A

Feed Cyclohexane

Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-3RM Size, m m 0.147-0.833 Wt., gm 0.5043

Catalyst Diluent Mullite Size, m m 0.147-0.495 Vol., cuP ~14

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.2 Hydrogen, psia 214.5

w/hr/v, gm feed/gm cat-hr 55.15

Feed Diluent, moles I^/mole feed 21.04

Feed Rate, cn?/hr 35.7

Minutes on Feed 120

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.39 Methylcyclopentane 13.65 Methylcyclopentene 0.00 Cyclohexane 36.10 Cyclohexene 0.00 Benzene 49.61 Toluene 0.00 C8 and Cg Aromatics 0.26

Hydrogen Balance, % 99.35

Space Time 0, gm cat-min/gm feed 1.088 Rate Constant 1^ , gm moles/gm cat-atm-min 0.01327 Rate Constant kg , gm moles/gm cat-atm-min 0.00341

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 234

TABLE B

RU N DATA

Run Number 12A

Feed Cyclohexane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-3RN Size, mm 0.147-0.833 W t ., gm 0.4273

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cnf3 ~14

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.6 Hydrogen, psia 214.1

w/hr/w, gm feed/gm cat-hr 55.97

Feed Diluent, moles Hg/mole feed 20.24

Feed Rate, cnt3 /hr 30.7

Minutes on Feed 160

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 2.71 Methylcyclopentane 15.49 Methylcyclopentene 0.00 Cyclohexane 31.96 Cyclohexene 0.00 Benzene 49.33 Toluene 0.00 CQ and Cg Aromatics 0.50

Hydrogen Balance, % 99.40

Space Time 9, gm cat-min/gm feed 1.072 Rate Constant kj , gm moles/gm cat-atm-min 0.01378 Rate Constant , gm moles/gm cat-atm-min 0.00405

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 235

TABLE B

RUN DATA

Run Number 13A

peed Cyclohexane

Catalyst Type P t - A l ^ -Mordenite Code FJH-617-3RO Size, mm 0.147-0.833 Wt., gm 1.1329

Catalyst Diluent Mullite Size, mm 0.147-0.495 Vol., cd3 ~13

Temperature, °F 900

Pressure, psia 224.7 Feed, psia 10.6 Hydrogen, psia 214.1

w/hr/w, gm feed/gm cat-hr 56.11

Feed Diluent, moles 1^/mole feed 20.14

Feed Rate, cd3/hr 81.6

Minutes on Feed 190

Product Composition, wt. % (Hydrogen Free Basis)

Cracked Gas (C6 -) 2.64 Methylcyclopentane 11.70 Methylcyclopentene 0.00 Cyclohexane 60.27 Cyclohexene 0.00 Benzene 25.23 Toluene 0.00 C8 and Cg Aromatics 0.14

Hydrogen Balance, % 101.07

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant kj, gm moles/gm cat-atm-min 0.00536 Rate Constant k^, gm moles/gm cat-atm-min 0.00232

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 236

TABLE B

RUN DATA

Run Number l^A

Feed Cyclohexane

Catalyst Type P t - A l ^ -Mordenite Code FJH-617-4RE Size, mm 0.147-0.351 Wt., gm 0.9201

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP

Temperature, °F . 900

Pressure, psia 224.7 Feed, psia 10.4 Hydrogen, psia 214.3

v/hr/w, gm feed/gm cat-hr 56.13

Feed Diluent, moles Hg/mole feed 20.62

Feed Rate, cnf3/hr 66.3

Minutes on Feed 215

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 7.77 Methylcyclopentane 12.18 Methylcyclopentene 0.00 Cyclohexane 7.89 Cyclohexene 0.00 Benzene 71.11 Toluene 0.03 CB and Cg Aromatics 1.03

Hydrogen Balance, % 101.01

Space Time 9, gm cat-min/gm feed 1.069 Rate Constant kx , gm moles/gm cat-atm-min 0.03522 Rate Constant kg, gm moles/gm cat-atm-min 0.00563

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 237

TABLE B

RUN DATA

Run Number

I*ee

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-4RE Size, mm 0.147-0.351 Wt., gm 0.9201

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cut3

Temperature, °F 800

Pressure, psia 84.7 Feed, psia 4*0 Hydrogen, psia 80.7

v/hr/w, gm feed/gm cat-hr 56.13

Feed Diluent, moles Hg/mole feed 20.42

Feed Rate, cnP/hr 66.3

Minutes on Feed 355

Product Composition, wt. % (Hydrogen Free Basis)

Cracked Gas (C6-) 2.90 Methylcyclopentane 9.83 Methylcyclopentene 0.00 Cyclohexane 50.93 Cyclohexene 0.00 Benzene 36.27 Toluene 0.00 CB and Cg Aromatics 0.07

Hydrogen Balance, % 99.96

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant , gm moles/gm cat-atm-min 0.02229 Rate Constant kg, gm moles/gm cat-atm-min 0.00564

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 238

TABLE B

RUN DATA

Run Number 15A

Feed Cyclohexane

Catalyst Type Pt-ALgOg-Mordenite Code FJH-617-4RF Size, mm 0.074-0.147 Wt., gm 0.9201

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cnt3 ~14

Temperature, °F 800

Pressure, psia 84.7 Feed, psia 3.9 Hydrogen, psia 80.8

w/hr/w, gm feed/gm cat-hr 56.13

Feed Diluent, moles Hg/mole feed 20.56

Feed Rate, cnP/hr 66.3

Minutes on Feed 145

Product Composition, wt. \ (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.75 Methylcyclopentane 10.10 Methylcyclopentene 0.00 Cyclohexane 31.27 Cyclohexene 0.00 Benzene 57.58 Toluene 0.00 Ce and Cg Aromatics 0.30

Hydrogen Balance, % 100.16

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant k j , gm moles/gm cat-atm-min 0.04309 Rate Constant kg, gm moles/gm cat-atm-min 0.00706

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 239

TABLE B

RUN DATA

Run Number 15B

Feed Cyclohexane

Catalyst Type Pt-Al2 03 -Mordenite Code FJH-617-4RF Size, nnn 0.074-0.147 Wt., gm 0.9201

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cnt3 ~14

Temperature, °F 750

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 56.13

Feed Diluent, moles Hg/mole feed 20.30

Feed Rate, cnt3 /hr 66.3

Minutes on Feed 310

Product Composition, w t . % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.00 Methylcyclopentane 9.06 Methylcyclopentene 0.00 Cyclohexane 53.01 Cyclohexene 0.00 Benzene 37.87 Toluene 0.00 C8 and Cg Aromatics 0.07

Hydrogen Balance, % 100.12

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant , gm moles/gm cat-atm-min 0.02263 Rate Constant kg, gm moles/gm cat-atm-min 0.00500

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 240

TABLE B

RUN DATA

Run Number 16A

Feed Cyclohexane

Catalyst Type Pt-Als03 -Mordenite Code FJH-617-4RG Size, mm 0.147-0.351 Wt., gm 0.9204

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cnP ~14

Temperature, °F 749

Pressure, psia 84.7 Feed, psia 4*0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 56.11

Feed Diluent, moles Hg/mole feed 20.36

Feed Rate, cnt3/hr 66.3

Minutes on Feed 155

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.97 Methylcyclopentane 8.07 Methylcyclopentene 0.00 Cyclohexane 43.82 Cyclohexene 0.00 Benzene 47.01 Toluene 0.00 C8 and Cg Aromatics 0.13

Hydrogen Balance, % 100.96

Space Time 0, gm cat-min/gm feed 1.069 Rate Constant kj, gm moles/gm cat-atm-min 0.03089 Rate Constant kg, gm moles/gm cat-atm-min 0.00487

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 241

TABLE B

R U N DATA

Run Number 17A

Feed Cyclohexane

Catalyst Type Pt-Al8 03 -Mordenite Code FJH-617-4RH Size, nnn • 0.074-0.147 Wt., gm 0.5072

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cnt3

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 3.9 Hydrogen, psia 80.8

. w/hr/w, gm feed/gm cat-hr 78.33

Feed Diluent, moles 1^/mole feed 20.84

Feed Rate, cnt3/hr 51.0

Minutes on Feed 160

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.39 Methylcyclopentane 9.26 Methylcyclopentene 0.00 Cyclohexane 42.97 Cyclohexene 0.00 Benzene 47.15 Toluene 0.00 CB and Cg Aromatics 0.23

Hydrogen Balance, % 99.91

Space Time 0, gm cat-min/gm feed 0.766 Rate Constant 1^, gm moles/gm cat-atm-min 0.04393 Rate Constant kg, gm moles/gm cat-atm-min 0.00801

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 242

TABLE B

RUN DATA

Run Number 18A

Feed Cyclohexane Catalyst Type Pt-AL,03 -Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 Wt . , gm 0.5072 Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnt* ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 3.9 Hydrogen, psia 80.8

w/hr/w, gm feed/gm cat-hr 78.33

Feed Diluent, moles Hg/mole feed 20.59

Feed Rate, cn?/hr 51.0

Minutes on Feed 155

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.00 Methylcyclopentane 9.44 Methylcyclopentene 0.00 Cyclohexane 39.27 Cyclohexene 0.00 Benzene 51.03 Toluene 0^.00 C8 and Cg Aromatics 0.27

Hydrogen Balance, % 100.50

Space Time 0, gm cat-min/gm feed 0.766 Rate Constant kj, gm moles/gm cat-atm-min 0.04873 Rate Constant kg, gm moles/gm cat-atm-min 0.00836

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 19A

Feed Cyclohexane

Catalyst Type Pt-ALg03 -Mordenite Code FJH-617-4RJ Size, mm 0.147-0.351 Wt., gm 0.4500

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cn? ~14

Temperature, °F 800

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 105.94

Feed Diluent, moles Hg/mole feed 20.38

Feed Rate, cnt3/hr 61.2

Minutes on Feed 155

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.39 Methylcyclopentane 8.08 Methylcyclopentene 0.00 Cyclohexane' 42.23 Cyclohexene 0.00 Benzene 49.30 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, % 100.09

Space Time 0, gm cat-min/gm feed 0.566 Rate Constant kj, gm moles/gm cat-atm-min 0.06082 Rate Constant kg, gm moles/gm cat-atm-min 0.00929

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 244

TABLE B

RUN DATA

Run Number 20A

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-4RK Size, mm 0.074-0.147 Wt., gm 0.4501

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cut3 ~14

Temperature, °F 800

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 105.92

Feed Diluent, moles H g / m o l e 20.33

Feed Rate, cnt3/hr 61.2

Minutes on Feed 135

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.00 - Methylcyclopentane 8.30 Methylcyclopentene 0.00 Cyclohexane 36.86 Cyclohexene 0.00 Benzene 54.81 Toluene 0.00 C8 and Cg Aromatics 0.03

Hydrogen Balance, % 99.59

Space Time 0, gm cat-min/gm feed 0.566 Rate Constant kj^ , gm moles/gm cat-atm-min 0.07119 Rate Constant kg, gm moles/gm cat-atm-min 0.01004

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 21A

Feed Cyclohexane

Catalyst Type Pt-AlgOa-Mordenite Code FJH-617-4RL Size, ntm 0.147-0.351 Wt., gm 0.7095

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cni3 ~14

Temperature, °F 774

Pressure, psia 84.7 Feed, psia 3.9 Hydrogen, psia 80.8

w/hr/w, gm feed/gm cat-hr 78.39

Feed Diluent, moles 1^/mole feed 20.51 . .

Feed Rate, cm3 /hr 71.4

Minutes on Feed 125

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.71 Methylcyclopentane 8.13 Methylcyclopentene 0.00 Cyclohexane 49.34 Cyclohexene 0.00 Benzene 41.78 Toluene 0.00 Ce and Cg Aromatics 0.03

Hydrogen Balance, % 99.39

Space Time 0, gm cat-min/gm feed 0.765 Rate Constant , gm moles/gm cat-atm-min 0.03617 Rate Constant kg, gm moles/gm cat-atm-min 0.00654

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 22A

Feed Cyclohexane

Catalyst Type Pt-AlgOg-Mordenite Code FJH-617-4EM v Size, mm 0.147-0.351 Wt., gm 0.3050

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 78.33

Feed Diluent, moles Hj/mole feed 20.33

Feed Rate, cnt3 /hr 30.7

Minutes on Feed 190

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.51 Methylcyclopentane 6.89 Methylcyclopentene 0.00 Cyclohexane 43.25 Cyclohexene 0.00 Benzene 48.99 Toluene ' 0.00 CB and Cg Aromatics 0.36

Hydrogen Balance, % 99.40

Space Time 0, gm cat-min/gm feed 0.766 Rate Constant kj , gm moles/gm cat-atm-min 0.04449 Rate Constant kg, gm moles/gm cat-atm-min 0.00579

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 23A

Feed Cyclohexane

’ Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-4RN Size, mm 0.147-0.351 Wt., Gm 0.4060

Catalyst Diluent Mullite Size, mm * 0.147-0.208 Vol., ctrP ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 78.28

Feed Diluent, moles I^/mole feed 20.23

Feed Rate, cnP/hr 40.8

Minutes on Feed 145

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Ce -) 0.79 Methylcyclopentane 8.73 Methylcyclopentene 0.00 Cyclohexane 40.03 Cyclohexene 0.00 Benzene 50.42 Toluene 0.00 Ce and Cg Aromatics 0.03

Hydrogen Balance, % 99.94

Space Time 0, gm cat-min/gm feed 0.766 Rate Constant kj, gm moles/gm cat-atm-min 0.04706 Rate Constant kg, gm moles/gm cat-atm-min 0.00756

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 24A

Feed Cyclohexane

Catalyst Type P t - A L ^ -Mordenite Code FJH-617-4R0 Size, mm 0.833-1.397 Wt., gm 0.5066

Catalyst Diluent Mullite Size, mm 0.495-0.833 Vol., cnt3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 78.42

Feed Diluent, moles Hj/mole feed 20.37

Feed Rate, cnt3/hr 51.0

Minutes on Feed 170

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 1.34 Methylcyclopentane 4.67 Methylcyclopentene 0.00 Cyclohexane 79.59 Cyclohexene 0.00 Benzene 14.35 Toluene 0.00 Cg and Cg Aromatics 0.05

Hydrogen Balance, % 99.39

Space Time 0, gm cat-min/gm feed 0.765 Rate Constant kj , gm moles/gm cat-atm-min 0.00999 Rate Constant kg, gm moles/gm cat-atm-min 0.00302

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 25A

Feed Cyclohexane

Catalyst Type Pt-AlgOg-Mordenite Code FJH-617-4RP Size, mm 0.351-0.833 Wt., gm 0.5069

Catalyst Diluent Mullite Size, mm 0.351-0.495 Vol., cnt3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 78.38

Feed Diluent, moles Hg/mole feed 20.34

Feed Rate, cnt3/hr 51.0

Minutes on Feed 175

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.77 Methylcyclopentane 3.60 Methylcyclopentene 0.00 Cyclohexane 57.42 Cyclohexene 0.00 Benzene 38.18 Toluene 0.00 C8 and Cg Aromatics 0.04

Hydrogen Balance, % 99.27

Space Time 0, gm cat-min/gm feed 0.766 Rate Constant , gm moles/gm cat-atm-min 0.03064 Rate Constant kg, gm moles/gm cat-atm-min 0.00267

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B RUN DATA

Run Number 26A

Feed Cyclohexane

Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-6RC Size, mm 0.147-0.351 W t ., gm 9.4286

Catalyst Diluent None Size, mm Vol., cnt3

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 3.37

Feed Diluent, moles Ilg/mole feed 20.14

Feed Rate, cm3/hr 40.8

Minutes on Feed 180

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 5.38 Methylcyclopentane 11.23 Methylcyclopentene 0.00 Cyclohexane 3.33 Cyclohexene 0.00 Benzene 79.33 Toluene 0.02 CB and Cg Aromatics 0.70

Hydrogen Balance, % 99.82

Space Time 0, gm cat-min/gm feed 17.799

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 251

TABLE B

RUN DATA

Run Number . 27A

Feed Cyclohexane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-6RD Size, mm 0.147-0.351 Wt., gm 0.8057

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnf3 <^13

Temperature, °F 775

Pressure, psia 84.7 Fee’d, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 39.45

Feed Diluent, moles I^/mole feed 20.00

Feed Rate, cnf3/hr 40.8

Minutes on Feed 160

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 1.47 Methylcyclopentane 11.16 Methylcyclopentene 0.00 Cyclohexane 22.05 Cyclohexene 0.00 Benzene 65.21 Toluene 0.00 C8 and Cg Aromatics 0.10

Hydrogen Balance, % 99.39

Space Time 0, gm cat-min/gm feed 1.521 Rate Constant kj , gm moles/gm cat-atm-min 0.03890 Rate Constant k^ , gm moles/gm cat-atm-min 0.00617

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 252

TABLE B

RUN DATA

Run Number 28A

Feed Cyclohexane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-6RF Size, mm 0.147-0.351 Wt., gm 0.2897

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., ctiP <—14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 109.71

Feed Diluent, moles I^/mole feed 20.14

Feed Rate, cnP/hr 40.8

Minutes on Feed 165

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.78 Methylcyclopentane 6.92 Methylcyclopentene 0.00 Cyclohexane 50.00 Cyclohexene 0.00 Benzene 42.11 Toluene 0.00 C8 and Cg Aromatics 0.18

Hydrogen Balance, % 100.65

Space Time 9, gm cat-min/gm feed 0.547 Rate Constant kj , gm moles/gm cat-atm-min 0.04988 Rate Constant kg-, gm moles/gm cat-atm-min 0.00760

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 253

TABLE B

RUN DATA

Run Number s 28B

Feed Cyclohexane

Catalyst Type Pt-ALg03 -Mordenite Code FJH-617-6RF Size, mm 0.147-0.351 Wt., gm 0.2897

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnf3 **'14

. Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 137.14

Feed Diluent, moles Hg/mole feed 20.41

Feed Rate, cnt3/hr 51.0

Minutes on Feed 330

Product Composition, wt. % (Hydrogen Free Base)______

Cracked Gas (C6-) 0.68 Methylcyclopentane 6.27 Methylcyclopentene 0.00 Cyclohexane 55.67 Cyclohexene 0.00 Benzene 37.38 Toluene 0.00 CB and Cg Aromatics 0.00

Hydrogen Balance, % 100.47

Space Time 0, gm cat-min/gm feed 0.438 Rate Constant , gm moles/gm cat-atm-min 0.05339 Rate Constant kg, gm moles/gm cat-atm-min 0.00830

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 254

TABLE B

RUN DATA

Run Number 28C

Feed Cyclohexane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-6RF Size, mm 0.147-0.351 Wt., gm 0.2897

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnt3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia .80.7

w/hr/w, gm feed/gm cat-hr 109.71

Feed Diluent, moles Hg/mole feed 20.01

Feed Rate, cnt3/hr 40.8

Minutes on Feed 490

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.22 Methylcyclopentane 6.92 Methylcyclopentene 0.00 Cyclohexane «. 50.97 Cyclohexene 0.00 Benzene 41.80 Toluene 0.00 C8 and Cg Aromatics 0.09

Hydrogen Balance, % 99.80

Space Time 0, gm cat-min/gm feed 0.547 Rate Constant kj, gm moles/gm cat-atm-min 0.04862 Rate Constant kg, gm moles/gm cat-atm-min 0.00746

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 29A

Feed Cyclohexane

Catalyst Type Pt-AlgOg Code FJH-618-1RF Size, mm 0.147-0.833 Wt., gm , 1.5245

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnt3 ~13

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 20.85

Feed Diluent, moles Hj/mole feed 20.05

Feed Rate, cnf3/hr 40.8

Minutes on Feed 145

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Ce -) 0.00 Methylcyclopentane 0.00 Methylcyclopentene 8.26 Cyclohexane 0.00 Cyclohexene 0.00 Benzene 91.74 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, 7» 99.74 « Space Time 0, gm cat-min/gm feed 2.878

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 3QA

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-6RG Size, mm 0.147-0.351 Wt., gm 0.0973

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnf3 ~15

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.70

w/hr/w, gm feed/gm cat-hr 326.65

Feed Diluent, moles Hj/mole feed 20.20

Feed Rate, cnf3/hr 40.8

Minutes on Feed 165

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Ce -) 0.78 Methylcyclopentane 2.36 Methylcyclopentene 0.00 Cyclohexane 78.14 Cyclohexene 0.00 Benzene 18.68 Toluene 0.00 Ce and Cg Aromatics 0.05

Hydrogen Balance, % 100.41

Space Time 0, gm cat-min/gm feed 0.184 Rate Constant 1^, gm moles/gm cat-atm-min 0.05398 Rate Constant kg, gm moles/gm cat-atm-min 0.00631

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 31A

Peed Cyclohexane

Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-6RJ Size, mm 0.147-0.351 Wt., gm 0.1750

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 181.62

Feed Diluent, moles Hj/mole feed 20.14

Feed Rate, cut3/hr 40.8

Minutes on Feed 160

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Ce -) 0.83 Methylcyclopentane 3.54 Methylcyclopentene 0.00 Cyclohexane 72.00 Cyclohexene 0.00 Benzene 23.63 Toluene 0.00 C8 and Cg Aromatics 0.01

Hydrogen Balance, % 100.57

Space Time 0, gm cat-min/gm feed 0.330 Rate Constant 1^ , gm moles/gm cat-atm-min 0.03931 Rate Constant 1^, gm moles/gm cat-atm-min 0.00546

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 33A

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-6RM Size, mm 0.147-0.351 Wt., gm 0.3647

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.94

Feed Diluent, moles Hj/mole feed 20.38

Feed Rate, cut3/hr 51.0

Minutes on Feed 150

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 1.20 Methylcyclopentane 8.15 Methylcyclopentene 0.00 Cyclohexane 54.25 Cyclohexene 0.00 Benzene 36.40 Toluene 0.00 C8 and Cg Aromatics 0.00

Hydrogen Balance, % 100.56

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kj, gm moles/gm cat-atm-min 0.04185 Rate Constant kg, gm moles/gm cat-atm-min 0.00871

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 33B

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-6RM Size, mm 0.147-0.351 Wt., gm 0.3647

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnt3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 2.7 Hydrogen, psia 82.0

w/hr/w, gm feed/gm cat-hr 108.94

Feed Diluent, moles i^/mole feed 30.34

Feed Rate, cnP/hr 51.0

Minutes on Feed 300

Product Composition, wt. 7, (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 1.55 Methylcyclopentane 6.44 Methylcyclopentene 0.00 Cyclohexane 58.75 Cyclohexene 0.00 Benzene 33.26 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, % 100.12

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kj, gm moles/gm cat-atm-min 0.05421 Rate Constant kg, gm moles/gm cat-atm-min 0.00974

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 33C

Peed Cyclohexane

Catalyst Type Pt-Al203 -Mordenite Code FJH-617-6RM Size, mm 0.147-0.351 Wt,, gm 0.3647

Catalyst Diluent Mullite Size, irnn 0.147-0.208 Vol., cut* '"'15

Temperature, °F 775

Pressure, psia 84.7 Feed, psia' 5.2 Hydrogen, psia 79.5

w/hr/v, gm feed/gm cat-hr 108.94

Feed Diluent, moles .Hg/mole feed 15.17

Feed Rate, cni3/hr 51.0

Minutes on Feed 455

Product Composition, wt. % " (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.94 Methylcyclopentane 8.36 Methylcyclopentene 0.00 Cyclohexane 55.48 Cyclohexene 0.00 Benzene 35.22 Toluene 0.00 C8 and Cg Aromatics 0.00

Hydrogen Balance, % 100.04

Space Time 6 , gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.03028 Rate Constant kg, gm moles/gm cat-atm-min 0.00669

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 33D

Feed Cyclohexane

Catalyst Type Pt-ALg03 -Mordenite Code FJH-617-6RM Size, mm 0.147-0.351 Wt., gm 0.3647

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.94

Feed Diluent, moles Hg/mole feed 20.33

Feed Rate, cntVhr 51.0

Minutes on Feed 590

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 1.31 Methylcyclopentane 8.47 Methylcyclopentene 0.00 Cyclohexane 57.46 Cyclohexene 0.00 Benzene 32.76 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, % 100.68

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.03665 Rate Constant kg, gm moles/gm cat-atm-min 0.00881

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 35A

Feed Cyclohexane

Catalyst Type Pt-AlgO^ -Mordenite Code FJH-617-4RS Size, nnn 0.147-0.351 Wt., gm 0.2926

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnf3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr ■ 108.62

Feed Diluent, moles Hg/mole feed 20.06

Feed Rate, cnf3/hr 40.8

Minutes on Feed 160

Product Composition, wt. % (Hydrogen Free Basis)

Cracked Gas (C6 -) 0.73 Methylcyclopentane 7.12 Methylcyclopentene 0.00 Cyclohexane 53.47 Cyclohexene 0.00 Benzene 38.55 Toluene 0.00 Ce and Cg Aromatics 0.14

Hydrogen Balance, % 99.70

Space Time 0, gm cat-min/gm feed 0.552 Rate Constant , gm moles/gm cat-atm-min 0.04372 Rate Constant 1^, gm moles/gm cat-atm-min 0.00750

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 35B

Feed Cyclohexane

Catalyst Type Pt-ALjOg -Mordenite Code FJH-617-4RS Size, mm 0.147-0.351 Wt., gm 0.2926

Catalyst Diluent Mulllte Size, mm 0.147-0.208 Vol., cnf3 ~14

Temperature, °F 775

Pressure, psia 134.7 Feed, psia 6.4 Hydrogen, psia 128.3

w/hr/w, gm feed/gm cat-hr 108.62

Feed Diluent, moles 1^/mole feed 20.05

Feed Rate, cnf3/hr 40.8

Minutes on Feed 320

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.54 Methylcyclopentane 8.53 Methylcyclopentene 0.00 Cyclohexane 45.54 Cyclohexene 0.00 Benzene 44.81 Toluene 0.00 Ce and Cg Aromatics 0.58

Hydrogen Balance, % 99.35

Space Time 0, gm cat-min/gm feed 0.552 Rate Constant kj, gm moles/gm cat-atm-min 0.03530 Rate Constant kg, gm moles/gm cat-atm-min 0.00609

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 35C

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-4RS Size, mm 0.147-0.351 Wt., gm 0.2926

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cm3 ''■'14-

Temperature, °F 776

Pressure, psia 109.7 Feed, psia 5.2 Hydrogen, psia 104.5

w/hr/w, gm feed/gm cat-hr 108.62

Feed Diluent, moles Hg/mole feed 20.04

Feed Rate, cnf3/hr 40.8

Minutes on Feed 475

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.61 Methylcyclopentane 8.26 Methylcyclopentene 0.00 Cyclohexane 52.16 Cyclohexene 0.00 Benzene 38.79 Toluene 0.00 Ce and Cg Aromatics 0.17

Hydrogen Balance, % 99.11

Space Time 0, gm cat-min/gm feed 0.552 Rate Constant kx, gm moles/gm cat-atm-min 0.03458 Rate Constant kg, gm moles/gm cat-atm-min 0.00679

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 36A

Feed Cyclohexane

Catalyst Type I’t-ALgtk Code FJH-618-7RD Size, nnn 0.147-0.351 Wt., gm 0.2926

Catalyst Diluent Miillite Size, mm 0.147-0.208 Vol., cm3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.62

Feed Diluent, moles I^/mole feed 20.04

Feed Rate, cnP/hr 40.8

Minutes on Feed 140

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 ~) 0.61 Methylcyclopentane 0..38 Methylcyclopentene 0.00 Cyclohexane 53.74 Cyclohexene 0.00 Benzene 45.26 Toluene 0.00 C8 and Cg Aromatics 0.00

Hydrogen Balance, % 99.69

Space Time 0, gm cat-min/gm feed 0.552 Rate Constant kx, gm moles/gm cat-atm-min 0.05101 Rate Constant , gm moles/gm cat-atm-min 0.00040

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 37A

Feed Cyclohexane

Catalyst Type Pt-AlgOg Code FJH-618-7RE Size, mm 0.074-0.147 Wt., gm 0.2920

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cni3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.85

Feed Diluent, moles Hg/mole feed 20.33

Feed Rate, cm3/hr 40.8

Minutes on Feed 145

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.64 Methylcyclopentane 0.33 Methylcyclopentene 0.00 Cyclohexane 49.11 Cyclohexene 0.00 Benzene 49.92 Toluene 0.00 C8 and Cg Aroma tics 0.00

Hydrogen Balance, % 100.61

Space Time 9, gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.05946 Rate Constant kg, gm moles/gm cat-atm-min 0.00037

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 38A

Feed Cyclohexane

Catalyst Type Pt-Al203 Code FJH-618-7RF Size, mm 0.074-0.147 Wt., gm 0.2913

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cni3 • ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 109.11

Feed Diluent, moles 1^/mole feed 20.09

Feed Rate, cnf3 /hr 40.8

Minutes on Feed 150

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C8-) 0.72 Methylcyclopentane 0.37' Methylcyclopentene 0.00 Cyclohexane 46.11 Cyclohexene 0.00 Benzene 52.80 Toluene 0.00 C8 and Cg Aromatics 0.00

Hydrogen Balance, % 100.23

Space Time 0, gm cat-min/gm feed 0.550 Rate Constant 1^, gm moles/gm cat-atm-min 0.06408 Rate Constant kj , gm moles/gm cat-atm-min 0.00042

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 39A

Feed Cyclohexane

Catalyst Type Pt-Al203 Code FJH-618-10RD Size, mm 0.044-0.074 W t., gm 0.2922

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cnt3 ~14

Temperature, °F 774

Pressure, psia 84.7 Feed, psia 3.9 Hydrogen, psia 80.3

w/hr/w, gm feed/gm cat-hr 108.77

Feed Diluent, moles B^/mole feed 20.68

Feed Rate, cm3/hr 40.8

Minutes on Feed 140

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.74 Methylcyclopentane 0.85 Methylcyclopentene 0.00 Cyclohexane 28.63 Cyclohexene 0.00 Benzene 69.78 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, % 100.20

Space Time 0, gm cat-min/gm feed 0.552 Rate Constant kj, gm moles/gm cat-atm-min 0.10613 Rate Constant kg, gm moles/gm cat-atm-min 0.00118

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 40A

Feed Cyclohexane

Catalyst Type Pt-AlgOg Code FJH-618-10RF Size, mm 0.044-0.074 Wt., gm 0.2922

Catalyst Diluent Mullite Size, mm 0.074-0.147 Vol., cni3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.77

Feed Diluent, moles Rj/mole feed 20.05

Feed Rate, cni3/hr 40.8

Minutes on Feed 120

Product Composition, wt. 7» (Hydrogen Free Basis)_____

Cracked Gas (Ce -) 0.68 Methylcyclopentane 0.35 Methylcyclopentene 0.00 Cyclohexane 33.08 Cyclohexene 0.00 Benzene 65.89 Toluene 0.00 C8 and Cg Aromatics 0.00

Hydrogen Balance, % 100.30

Space Time 0, gm cat-min/gm feed 0.552 Rate Constant kj, gm moles/gm cat-atm-min 0.09159 Rate Constant kg, gm moles/gm cat-atm-min 0.00045

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41A

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnt* ~14

Temperature, °F 775

Pressure, psia 134.7' Feed, psia 6.4 Hydrogen, psia 128.3

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles l^/mole feed 20.07

Feed Rate, cnP/hr 40.8

Minutes on Feed 150

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.59 Methylcyclopentane 9.78 Methylcyclopentene 0.00 Cyclohexane 43.44 Cyclohexene 0.00 Benzene 46.19 Toluene 0.00 C8 and Cg Aroma tics 0.00

Hydrogen Balance, % 100.92

Space Time 6 , gm cat-min/gm feed 0.551 Rate Constant kj , gm moles/gm cat-atm-min 0.03721 Rate Constant kg, gm moles/gm cat-atm-min ' 0.00713

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41B

Feed Cyclohexane

Catalyst Type Pt-ALgGtj -Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol. , cut3 ~14

Temperature, °F 775

Pressure, psia 109.7 Feed, psia 5.2 Hydrogen, psia 104.5

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles H^/mole feed 20.29

Feed Rate, cnt3/hr 40.8

Minutes on Feed 355

Product Composition, wt. 7> (Hydrogen Free Basis)_____

Cracked Gas (Ce-) 0.61 Methylcyclopentane 8.94 Methylcyclopentene 0.00 Cyclohexane 45.73 Cyclohexene 0.00 Benzene 44.45 Toluene 0.00 Ca a n d Cg Aromatics 0.26

Hydrogen Balance, % 99.97

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.04267 Rate Constant kg , gm moles/gm cat-atm-min 0.00790

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41C

Feed Cyclohexane

Catalyst Type Pt-Al803 -Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 W t ., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cut3 ~14

Temperature, °F -775

Pressure, psia 54.7 Feed, psia 2.6 Hydrogen, psia 52.1

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles Hj/mole feed 20.06

Feed Rate, cnP/hr 40.8

Minutes on Feed 505

Product Composition, wt. (Hydrogen Free Basis)_____

Cracked Gas (C6 -) 0.53 Methylcyclopentane 3.71 Methylcyclopentene 0.00 Cyclohexane 63.52 Cyclohexene 0.00 Benzene 32.14 Toluene 0.00 C8 and Cg Aroma tics 0.10

Hydrogen Balance, % 99.43

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kj, gm moles/gm cat-atm-min 0.05206 Rate Constant kg, gm moles/gm cat-atm-min 0.00558

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41D

Feed Cyclohexane

Catalyst Type Pt-ALgOa-Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cm3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 2.7 Hydrogen, psia 82.0

w/hr/w, gm feed/gm cat-hr 108.81

FeedDiluent, Moles Hg/raole feed 29.90

Feed Rate, cnP/hr 40.8

Minutes on Feed 650

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.77 Methylcyclopentane 6.88 Methylcyclopentene 0.00 Cyclohexane 54.80 Cyclohexene 0.00 Benzene 37.55 Toluene 0.00 Ca and Cg Aromatics 0.00

Hydrogen Balance, % . 100.58

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kj, gm moles/gm cat-atm-min 0.06190 Rate Constant kg , gm moles/gm cat-atm-min 0.01052

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41E

Feed Cyclohexane

Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-4RI Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnf3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 5.3 Hydrogen, psia 79.4

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles Hj/mole feed 14.97

Feed Rate, cnt3 /hr 40.8

Minutes on Feed 795

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (C6-) 0.48 Methylcyclopentane 7.93 Methylcyclopentene 0.00 Cyclohexane 51.76 Cyclohexene 0.00 Benzene 39.83 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, % 100.48

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kj, gm moles/gm cat-atm-min 0.03473 Rate Constant , gm moles/gm cat-atm-min 0.00643

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41F

Feed Cyclohexane

Catalyst Type Pt-Alg03 -Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cut3 ~14

Temperature, °F _ 750

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles Hj/mole feed 20.14

Feed Rate, cnP/hr 40.8

Minutes on Feed 945

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Ce -) 1.20 Methyicyclopentane 6.36 Methylcyclopentene 0.00 Cyclohexane 60.80 Cyclohexene 0.00 Benzene 31.64 Toluene 0.00 C8 and Cg Aroma tics 0.00

Hydrogen Balance, % 100.24

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant , gm moles/gm cat-atm-min 0.03435 Rate Constant kg, gm moles/gm cat-atm-min 0.00637

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41G

Feed Cyclohexane

Catalyst Type Pt-AlgOg -Mordenite Code FJH-617-4RT Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnf3 ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia ' 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles Hj/mole feed 20.14

Feed Rate, cnf3/hr 40.8

Minutes on Feed 1085

Product Composition, wt.% (Hydrogen Free Basis)

Cracked Gas (Cg -) 0.67 Methylcyclopentane 7.09 Methylcyclopentene 0.00 Cyclohexane 54.42 Cyclohexene 0.00 Benzene 37.82 Toluene 0.00 Cg and Cg Aromatics 0.00

Hydrogen Balance, % 99.64

Space Time 6 , gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.04274 Rate Constant kg, gm moles/gm cat-atm-min 0.00743

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 41H

Feed Cydohexane

Catalyst Type Pt-AlgOa -Mordenite Code FJH-617-4RT Size, m m 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~14

Temperature, 0 P 724

Pressure, psia 83.7 Feed, psia 4.0 Hydrogen, psia 79.7

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles /mole feed 20.16

Feed Rate, cnP/hr 40.8

Minutes on Feed 1230

Product Composition, wt. % (Hydrogen Free Basis)______

Cracked Gas (C6 -) 0.55 Methylcyclopentane 5.23 Methylcyclopentene 0.00 Cyclohexane 68.22 Cvclohexene 0.00 Benzene 26.00 Toluene 0.00 Ce and Cg Aromatics 0.00

Hydrogen Balance, % 98.89

Space Time 9, gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.02730 Rate Constant Ife , gm moles/gm cat-atm-min 0.00502

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 43A

Feed Cydohexane

Catalyst Type Ft-AlgOa Code FJH-618-7RG Size, mm 0.3J1-0.833 Wt., gm 0.2930

Catalyst Diluent Mullite Size, mm 0.351-0.495 Vol., cnP ~14

Temperature, °F 775

Pressure, psia 84.7 Feed, psia 4.0 Hydrogen, psia 80.7

w/hr/w, gm feed/gm cat-hr 108.47

Feed Diluent, moles Ife /mole feed 20.10

Feed Rate, cnP/hr 40.8

Minutes on Feed 120

Product Composition, wt.% (Hydrogen Free Basis)

Cracked Gas (Cg -) 0.57 Methylcyclopentane 0.59 Methylcyclopentene 0.00 Cyclohexane 81.34 Cyclohexene 0.00 Benzene 17.50 Toluene 0.00 Cg and Cg Aromatics 0.00

Hydrogen Balance, % 100.57

Space Time 0, gm cat-min/gm feed 0.553 Rate Constant kx , gm moles/gm cat-atm-min 0.01637 Rate Constant , gm moles/gm cat-atm-min 0.00051

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 44A

Feed Cydohexane

Catalyst Type Pt-AIg O3 -Mordenite Code FJH-617-6RN Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~14

Temperature, °F 775

Pressure, psia 132.7 Feed, psia 2.6 Hydrogen, psia 130.1

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles /mole feed 49.48

Feed Rate, cnP/hr 40.8

Minutes on Feed 140

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Cg -) 0.76 Methylcyclopentane 7.87 Methylcyclopentene 0.00 Cyclohexane 52.49 Cyclohexene 0.00 Benzene 38.89 Toluene 0.00 Cq and C9 Aromatics 0.00

Hydrogen Balance, % 100.79

Space Time 0, gm cat-min/gm feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.06970 Rate Constant leg , gm moles/gm cat-atm-min 0.01283

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE B

RUN DATA

Run Number 44B

Feed Cydohexane

Catalyst Type Pt-Alg Cfe-Mordenite Code FJH-617-6RN Size, mm 0.147-0.351 Wt., gm 0.2921

Catalyst Diluent Mullite Size, mm 0.147-0.208 Vol., cnP ~ 1 4

Temperature, °F 775

Pressure, psia 107.7 Feed, psia 2.6 Hydrogen, psia 105.1

w/hr/w, gm feed/gm cat-hr 108.81

Feed Diluent, moles lfc/mole feed 40.04

Feed Rate, cut3/hr 40.8

Minutes on Feed 235

Product Composition, wt. % (Hydrogen Free Basis)_____

Cracked Gas (Cg -) 0.54 Methylcyclopentane 7.09 Methylcyclopentene 0.00 Cyclohexane 56.36 Cyclohexene 0.00 Benzene 36.01 Toluene 0.00 Cq and Cg Aromatics 0.00

Hydrogen Balance, % 101.18

Space Time 8 , gm cat-min/gra feed 0.551 Rate Constant kx , gm moles/gm cat-atm-min 0.06160 Rate Constant Ife , gm moles/gm cat-atm-min 0.01119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX C

SAMPLE CALCULATIONS

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX G

. SAMPLE CALCULATIONS

All material balance and rate constant calculations for this

investigation were made using the computer program listed in Appendix

E. All computations using this program were made at the Louisiana

State University Computing Center using either the IBM-7044 or OS/360

computers. The sample calculations listed in this Appendix illustrate

the calculation of yields and material balance for a typical experi

mental run. Any slight discrepancies noted in the final results

listed here and those listed in Appendix B for this run can be attrib

uted to manual versus computer calculation.

A detailed explanation of the calculational method for the rate

constants from the experimental data is given in Chapter V and not

included here.

Process Experimental Data

Run number 35A

Feed Cyclohexane

Catalyst Pt-AlgC^ -Mordenite

Catalyst size range 0.147-0.351 mm

Catalyst weight 0.2926 gm

Catalyst diluent Mullite

Catalyst diluent size range 0.147-0.208 mm

Catalyst diluent volume ~14 cnt3

282

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 283

Temperature 775° F

Pressure 84.7 psia

Length of material balance period 20 min

Feed in 13.59,cn?

Hydrogen in 2.281 ft?

Temperature of hydrogen in 80PF

Product gas out 2.524 ft?

Temperature of product gas out 80Pp

Liquid product out 0.0139 gm

The quantity of liquid product obtained in this and almost all

runs on cyclohexane was too small to be analyzed on the gas chromato

graph. For all of the runs where a liquid product was obtained that

was too small to be analyzed, it was called Cg residue for the computer

calculation. Duplicate computer workups on all experimental runs

where liquid products were obtained that were too small for analysis

showed no discernible difference in the yields or rate constants.

For the purposes of this sample calculation, the liquid product obtained

in this run was neglected although it was included in the final workup

listed in Appendix B.

The degree of hydrocracking in all of the cyclohexane runs was

small. Excluding the equilibrium run (26A), the highest hydrocracking

yield was 1.5% with most runs having about 0.5% hydrocracking. Since

the hydrocracking yields were so small for the cyclohexane runs, all

of the components produced from hydrocracking cyclohexane were lumped

together into a single component called cracked gas. It was assumed

ic Cx through C6 paraffins.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 284 \

that this cracked gas component could be represented on the average

by i-pentane. In the runs using n-heptane or methylcyclopentane as

feeds where hydrocracking was appreciable, all components from hydro

cracking were treated separately.

B. Product Gas Analysis from Chromatograph

Component Peak Area

Cracked gas 0.10

Cyclohexane 6 .77

Methylcyclopentane 0.94

Benzene 4.89

C. Yield and Material Balance Calculations

Basis: 20 minute balance period

Weight of cyclohexane fed =» (13.59 cut3)(0.779 gm/cnt3) ** 10.59 gm

Gram moles of cyclohexane fed *» (10.59 gm)/(84.16 gm/gm mole) =

0.1258 gm moles

Grams of carbon in cyclohexane feed = (0.1258 gm moles) x

(72.06 gm C/mole cyclohexane) = 9.065 gm

Grams of in cyclohexane feed = (0.1258 gm moles) x

(12.10 gm l^/mole cyclohexane) = 1.522 gm

Weight hourly space velocity = (10.59 gm/20 min)(60 min/hr)/

(0.2926 gm) = 108.6 gm/hr/gm

Space Time 0 ** (60 min/hr)/108.6 gm/hr/gm) ** 0.555 gm cat-min/gm

feed

Volume of gas in at 6 GPF and 1 atm = (2.281 ft?)(0.926) **

2.112 ft?

Gram moles of gas in = (2.112 ft?)(1.20 gm moles/ft?) =

2.534 gm moles

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 285

Partial pressure of cyclohexane = (84.7 psia)(0.1258)/

(0.1258 + 2.534) = 4.01 psia

Partial pressure of hydrogen = 84.7 - 4.01 = 80.69 psia

Hydrogen to cyclohexane mole ratio = (2.534 gm moles)/

(0.1258 gm moles) = 20.14

Volume of product gas out at 6 CPF and 1 atm =

(2.524 ft3)(0.926) - 2.337 ft3

Gram moles of product gas out = (2.337 ft3 ) (1.20 gm moles/ft3 ) =

2.804 gm moles

Assume grams of carbon in cyclohexane feed = grams of carbon

in product = 9.065 gm

Gas chromatograph molar weighting factors for hydrocarbons in

gas out:

Peak Area x

Component GC Calibration Factor Holes (unsealed)

Cracked gas (0.10)(1.580) = 0.1580

Cyclohexane (6.77)(1.471) = 9.9587

Methylcyclopentane (0.94)(1.411) - 1.3263

Benzene (4.89)(1.582) = 7.7360

£ Moles (unsealed) = 19.1790

Hydrogen-free mole fractions of hydrocarbons:

Moles (unsealed)/ Component E Moles (unsealed) Mole Fraction

Cracked gas 0.1580/19.1790 = 0.0080

Cyclohexane 9.9587/19.1790 = 0.5192

Methylcyclopentane 1.3263/19.1790 = 0.0692

Benzene 7.7360/19.1790 - 0.4036

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 286

Grams of carbon per gram mole of hydrocarbon gas out:

(Hj-free mole

fraction) x (grams of Grams carbon per

Component carbon per gram mole) gram mole gas out

Cracked gas (0.0080)(60.05) 0.480

Cyclohexane (0.5192)(72.06) 37.413

Methylcyclopentane (0.0692)(72.06 4.987

Benzene (0.4036)(72.06) 29.083 E = 71.963 gm C/ gm mole

Actual gram moles of hydrocarbon out = actual grams of carbon

out/grams of carbon per gm mole of hydrocarbon gas out = 9.065

gm C/71.963 gm C/gm mole = 0.1260 gm mole.

Gram moles of hydrocarbons in gas out:

(Hj -free ,mole gm moles of hydro

fraction) x (gm moles carbon component

hydrocarbon out)Component in gas out

Cracked gas (0.0080)(0.1260) 0.00101

Cyclohexane (0.5192)(0.1260) 0.06542

Methylcyclopentane (0.0692)(0.1260) 0.00872

Benzene (0.4036)(0.1260) 0.05085

0.12600

gm moles

Grams of hydrocarbon in gas out:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 287

(gram moles) x Grams of

Component (molecular weight) hydrocarbon out

Cracked gas (0.00101)(72.15) 0.073

Cyclohexane (0.06542)(84.16) 5.506

Methylcyclopentane (0.00872)(84.16 0.734

Benzene (0.05085)(78.11) 3.972

10.285

Actual grams hydrocarbon out

Grams of hydrogen in hydrocarbon gas out:

(gram moles) x Grams of

Component (grams of Ife per gram mole) Ha out

Cracked gas (0.00101)(12.10) 0.0122

Cyclohexane (0.06542)(12.10) 0.7916

Methylcyclopentane (0.00872)(12.10) 0.1055

Benzene (0.05085)(6.06) 0.3076

1.2169

Actual grams Hg out

Material balance = [(gm hydrocarbons out + gm Hg as gas out)/

(gm cyclohexane in + gm as gas in)] x 100

- | [10.285 + (2.804 - 0.126)(2.016)]/

[10.59 + (2.534)(2.016)] J x 100 = 99.91%

Hydrogen balance = [(gm Hg in hydrocarbon out + gm 1^ as gas out)/

(gm Hg in hydrocarbon in + gm Hg as gas in)] x 100

« |[l.2169 + (2.804 - 0.126)(2.016)]/

[1.522 + (2.534)(2.016)] | x 100 = 99.78%

Product yield on a hydrogen-free basis:

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 288

Component (grams/total grams) x 100 Wt. %

Cracked gas (0.073/10.285) x 100 0.71

Cyclohexane (5.506/10.285) x 100 53.53

Methylcyclopentane (0.734/10.285) x 100 7.14

Benzene (3.972/10.285) x 100 38.62

Z » 100.00

Mole fractions of benzene and methylcyclopentane on a hydrogen

and cracked gas-free basis:

yB' - (0.05085) gm moles/(0.06542 + 0.00872 + 0.05085)

gm moles

= 0.4068

yH' = (0.00872) gm moles/(0.06542 + 0.00872 + 0.05085)

gm moles

= 0.0698

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX D

ANALYTICAL SYSTEM

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX D

ANALYTICAL SYSTEM

A. Introduction

The product analyses in this investigation were made on a F & M

Model 810R dual-column gas chromatograph. This gas chromatograph

employed thermal conductivity detection and was equipped with tempera

ture-programming for optimum elution time and peak sharpness. The

column used was 107» silicone rubber (SE-30) and 90% white chromosorb

(80-100 mesh) and was ten feet long. Peripheral equipment used for

peak area integration included a ball and disk mechanical integrator

and an Infotronics Model CRS-110 digital Integrator. The following

sections describe the operation of the gas chromatograph, calculation

of component concentrations and the calibration factors used.

B. Operation of Gas Chromatograph

The first step in a typical product analysis is the sample

injection via a hypodermic syringe into the helium carrier gas stream.

The helium carrier gas, containing the hydrocarbon sample then passes

through the silicone-chromosorb column where the individual hydro

carbons are selectively adsorbed. After leaving the column, the

carrier gas and any hydrocarbons present pass through the thermal

conductivity detector cell. Since each hydrocarbon has a different

column retention time due to varying adsorptivity, the gas passing

through the thermal conductivity cell normally consists of helium plus

a single hydrocarbon component.

290

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 291

The thermal conductivity cell is used to detect the difference

in thermal conductivity between the pure helium carrier gas and other

gases present in the mixture. This difference in thermal conductivity

is proportioned to the instantaneous concentration of the non-helium

components present and is recorded on a strip-chart as function of

time. To obtain the total concentration of any component it is neces

sary to integrate the detector output with respect to time. The

primary integrator used was the Infotronics Model CRS-110 integrator

which uses the detector output directly. This integrator prints out

peak areas and retention times on a Victor totalizing printer. The

Infotronics integrator senses changes between peaks by a change in

the detector output slope sign from negative to positive. Retention

times are given at the peak maxima which are recognized by a change

in detector output slope sign from positive to negative. The secondary

or back-up integration system is a ball and disk mechanical integrator

which operates directly off of the slide wire for the strip-chart pen.

Temperature programming was used to provide optimum elution time

and peak sharpness. All product analyses from the cyclohexane studies

used the same temperature programming schedule. The initial tempera

ture of the oven which held the silicone rubber-chromosorb column was

set at 4(PC. At the instant of sample injection the temperature

program was started. This program consisted of an increase in tempera

ture at the rate of 30PC/minute to a final temperature of lOOPc. The

final portion of the program consisted of holding the column isother

mal ly at lOQPC for seven minutes at which time the analysis was

complete.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 292

C. Calculation of Component Concentrations from the Dectector Output

The response from the thermal conductivity detector is a linear

function of the instantaneous hydrocarbon concentration in the helium

carrier gas stream. The instantaneous concentration of an arbitrary

component j is

yj = c /d j

where

yj e instantaneous mole fraction of component j,

Cj *» a constant, characteristic of component j,

D, = detector output for component j .

Associated with each component is an area which is the integrated

detector output with respect to time for that component. This area

is proportional to the total concentration of that component in the

carrier gas stream and is related to the hydrogen-free mole fraction

in the sample by

y o - -C^ J ____ n

k=l

where

Y ** hydrogen-free mole fraction,

C «* experimentally determined calibration constant,

A = peak area,

n ° total number of components in the hydrogen-free sample.

The gas chromatograph has an electronic attenuator so that the detector

output may [be scaled down by selected factors. The final equation

relating pelak areas and concentration is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 293

^ 'j k=l

where F is the attenuator factor.

D. Calibration Factors

The calibration factors used in this study are those used daily

in the Petroleum Processing Laboratory at Louisiana State University

and were initially determined by Dr. P. A. Bryant. During the cali

bration program, a precise volume of the component in question was

injected into the chromatograph and the corresponding peak area noted.

The integrated response (including the attenuation factor) is a cali

bration factor since it represents the area/cnt3 of the sample. A more

convenient form of this number, is the reciprocal, which corresponds

to the gram-moles of the given component per unit integrator reading.

The calibration factors listed in Table D-l were determined as

discussed above and are defined as

r - 1000 ‘ ' V i The factor os 1000 was used for convenience and does not affect final

results since it cancels out in the calculation of the hydrogen-free

concentration.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 294

Table D-l. Calibration Factors for the Gas Chromatograph.

C, Component gram-moles per unit area x 10?

Methane 3.765 Ethane '2.496 Propane 2.006 i-Butane 1.707 n-Butane 1.685 i-Pentane 1.580 n-Pentane 1.495 2 ,2 -Dime thylbu tane 1.352 2 ,3-Dime thylbu tane 1.352 2-Methylpentane 1.308 3-Methylpentane 1.318 n-Hexane 1.278 Methylcyclopentane 1.411. 1-Methylcyclopentene 1.436 Cyclohexane 1.4714. Cyclohexene 1.497* Benzene 1.582 i-Heptanes 1.147 n-Heptanes 1.147 C0-Aromatics 1.000 Unidentified Cg 1.000 i

•fa Determined in the present investigation.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. APPENDIX E

COMPUTER PROGRAM LISTING

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

COMPUTER PROGRAM LISTING n i - Oo ou o o uO O o X X X x o — or U. 03 0t-— X — - t 00 % © < - _J u _J - < 4-4< o © r o m m — Z X to 0 0 X X X DC X © w H o e o to X to m m JO U - l O Z < CO© < O < CLI UICK — © «. E Ul*v % to X < \ mo < o jm _ O D m o u \ x o < m > % o c \ u < ~ i _ "DO C3 • H r* X •

— cu Kl < x 2 > to © »■» u < m < Q I— L l a ~ * % •* l o ~ u . a O o o zz z z m © ««*» >0 M Z>00 X < m m ^ m m o U w u w * - I O — O o w O m ^ CL o x ru < x 13 J >- 0J O o o to < © © «*CD 1 o x o ? r - t u a — u 4 * •4 * * (3 © © tvj U- o CL- X i— o m m X i X Ui a o X © © m m o CL X X —1 X X * 4-4 a ■» x n < in L X 0J zz z z L CD 2 .-« U — CU CU X L W 3 — nin o u o r ~ " CO "3 O u © u © © 4-4 x o o o •» © CD X L -1 © X X X UJ — >- >- UI «* o _ o X X > _ j to j _ X X o a: o H X c X l l •> O' <* * %H- l L l s * * ** 2 j •* X % < c o X X r o l * •» "4 I X 1 U UJ cc X CD DC I— < X c QC u c © DC r. o to X EC X - h to X X c o X or DC ' O - X V ■4 % «o V o oo \X\ \ X \ o to o a a CVJ >- o u 2 m ■4- X © © m m © © 2 «»_* -—. X um — m cu X M CO to m UJ z z o z z t—4 X X X or u X X CL 2 X \ % . a *—« 4—4 w i u -4 1- 1- * X « 2 H X — n ~ in nin n — in X X * © o © M N H © ru X — X O CO 4-4\ \ x 4- « 4— 0 I-XHOW O H X Is- \0 H 2 . s X CVJ IV X V X 4 r V, O s rv •4- % — X \ X ■!*- O X \ CU © vO V rv x X H 2 X \ c \ Q — Q - r n- in x ■4- ~ < ^ — *■< 2 V X N T x in x CVJ \ 4-4 2 2 4V • — • - \ --- • 4 • •4 4-- • . ■ «. * • •* • % © \D H CVJ \ 2 4-4 2 2 • % % ^ cu 4 (U 00 r-- H tH 4—4 Is- cu m 0 X © s m © \ 4^ _ vO H 4-4 4—4 X X 4—1 \ 4- vO 4 4—4 \ 4—4 \ CU\ X 2 © 2 4_4 \ w 2 W © X 4—4 * s o > 2 X 2 X -— • • ♦ % % —4 • _ 4 * N \ + \ N in m CU 4-4 tH 4-4 »-4 n v n i 4—4 4-4 rs. > m « o> \ *•« fly © H 4-4 © 4 \ © X O O \ \ m \ 0 X 0 -—4 4—4 ^4 \ CU X cu x © X O cu X X 2 2 X CU X 2 \ m \ 0 2 X CU X 2 X X X X -- » • • Si »

vO u \ ru 4-4 cu 4-1 0 \ H 4 ^ uS cu O 4- 4-4 \ 2 \ 4-4 <0 cu \ 4 0 0 CU \ \ © 4—4 v \ ov —4 \ X 2 X X X 2 2 . r X X 2 X —4 © • • • • • cu CV> n cu XX X CU «-4 m r ^ in \ -m 4- 4 4—4 4 4—. O - iv «-» 0 4*^ 2 2 2 X • b • H % • ^ m CU 44 4- V © CU ■* -fl- © 4- 4—4 cu cu © cu \ CTV 4-4 4-4 \ 4—* ■ 0 CUv0 —■ s 4—. « Q w 4—4 w \ • \ X < t- 2 X < cu 2 4 4—4 2 O X \ \ \ X X 4 _ • —. • * • • - © - 4 m

4-4 4—4 - 0— -*• — 40 >- \ O 4 \ 0 X 2 -- » in 4-4 4 •4- 4—4 ■4- 4- * m - > >- ■4- © © CU O cu < m >- CO X © 0 cu 0 0 >- m r- X X -— \ 0 2 < \ 2 2 X \ X 2 \ >- \ 2 \ -- • • » • vrv W W W W N M N M •V — V • % 3 0 0 X 1 0 • 2 44 40 0 4 H 0 0 3 4 S 4-4 X - ' O 4-4 O X w X ^ cu >- rv — © 0 — 40 tv V 4-4 40 V tv •« © tv or ~ N O W X O X cu X rv — 4-4 2 0 - V M >- 40 N • .N X I R 0 >- >- \0 40 4H v0 © CU 2 >- CU >- 0 n v v in 4 \ + X \ 0 2 U ( 2 - O X «-4 O O • x tv 2 X m \ \ \ 2 • W r- x - • >- O V 4- V O W 4 H \ “ "Vtv ■» \ 0 v . . v w O 0 v v 2 \ >4 H >4 2 o v x \ •—» X CU 2 © CU 2 X V» . H4 O N. X • - © >- • > *4* • % ■4 % fv 4-1 • ^ V - u32XS X 2 3 ru — w - CU 2 >- V V c m in m cu - • « >- ■4- —

4-4 nin ^ n > 2 U ( 0 v "4 tv 0 • 40 v u• - 4-4 w -V • cu rv v *4 rv 40 40 — CU— 0 4- — — - ' 2 V x IV 0 c c *4 m cu cu x X T *4 ■T rv rv cu X - 2 — X X 4-4 X X U C X 4 ■4 O • X • X X • — * ■4CU ^ • • «v © r — ru \ V w X 4-4 «H 4-4 4-4 s 0 0 m — © 0 © u t x 0 2 2 h 4

\ O

- 4— 0 X 40 — «4 >- >- >- 4-4 O 4-4 44 4-4 4-4 V. 4-4 - s 4-4 n x i © in x in 0 0 O CU O - 2 4- — — 0 >- 40 0 > i X in x >- \ — O » 4» »-• • v» 0 * c < -* \ X • V 2 X — X — V ~ 2 >- « m \ \ — \ 2 x H « r x © x © H •* -—■ • *— • » «» O n h s h in n w CU

— © « IV

n r ru tn 4-4 40 0 V X 40 0 x 40 -1 .- DC v ,4 vO ,-4 tv 4-4 4-4 OC v ru v — x m X O X • — n — in nin DC OC \\ \ H • X >4 4-4© >4 • X cu \ "V t^ — \ \ I V X V 4-4 V — C m \ X v or v X or or -4 O 440 «4 V o x o V V X O 0 oXX X ro x V V V X tv X © © v VV V v V X • • • • • X • •4 n x in % — N fv • V X \ ^

© |V c o c o '-- X © 4-1 ru 4 0 0 0 0 • - 4 IV rv if < © h © DC *— c o c o 40 « cu c o — 0 0 X X 0 X 0 0 X X X • • » • % » •4 4» * rv H © © H 0 ucu cu © in +*+ ■4- V -— c o ru tv or -—' c o X © V m © 40 X < * V X 40 0 \ 0 0 1— c o 0 X \ V © X \ X • » 4-4 40 4-4 4-4 © 40 © © © c o m c o 4-4 V in ^4 — 0 a rv or V V 0 V 40 0 0 V 0 0 X X V 0 V X I V X

TABLE E X 4- cn — v rv CU vrv m oc «-< cn «-< cn O 0% VC 00 o o s or — 1 V v r cu —o v 0 x n cucn 4- CU00 CUN X 0 00 CU N % cu cnoc i — ~ o \ X 4- \ ~ “ V U>X OC 1 V x \ V V 00 X O V \ X X cu v • X o X V o x V CVJ V x oo x *• • X • cn • v in v * ■* V V

o — »» 00 oc oc 00

s m m » in \ o «-•o o Ll Llv ~ — m

^ v ^ L vo vo — — — O cn cn l r \0 L • so o in o N4- v L l l l l l l ■sf'

L L •» cu L Ll - * on N cn cn cn * w-i w-i m cu cu w-i l l l l

N • N cn— cn o V V — cn L — on L on v m • v ^ H ~ N cuv • V L ON— L — Ll (l— • V V « cnv CU o o Ll V ~ % H v. V cu o cu o cn-*• \0 rv 00 H ^ H rv IL — — V cu o o •* Ll O lL V — V • V Ll ~ \ cu o tn — tn cnu. rtV Ll h V — V cu o cn v Ll v tn v v cn v l cn • • -a- cu • * cn * o • v cn v • »-« l v cnv — «- —

in CU 00 l i CU v CU L V Ll O Ll o O l l l i

V ~ — cn V I cnv v t V — — ^ — 00 ■* OC — o L © 0N\«- vO • vO I CU I O U L CU n «-* in U I I CU V ~ — u ^t- • cu L cn v — cn v on — 00 V L CU V — v — in V L •-< O — 00 V L V 00 — in CU L n c o H L O O cn o o o o cn o o — O V U v CU V ON L O — cnv V L L L L oo * h- ' — O'. V l Ll — • l l l N < ^ l L ^ • ON V v L • ON V v I — • l V O v l l u. oo u.

3 O -3- • V

in • • < • V . m U. cn • N v i U V — v cn 00— ON V «-• • o V L 00 h cn cn o cn O K I I L v 50 v l v r • l l l l m CU OC fv O 00 L 00 00 00 O l l l l l l v l v

HI I—I •—IO M HI _J L4HIHI I—I © o - x o o o x x x o o x x r o o L O O O C 0 3 X O U X 3 X O 2 M llllH l r cu H | U C ( 2 >- cnx 2 x H I U O O O o H o o m o i h II I M H • • v X ^-J ^ J - ^ v X X X v • • • s 2 2 0 0 - 1 = • n rv in .=*• 3

•—o I3 IhD» D I « I h I 3 H i h HI I H M M 2 0 0 - 1

h --- 2 "3 3 " 2 . un O O cu n o

i h « h

— «-• * — o O h 2 2 l • ) C 2 O U

298 h e p t a n e

TABLE E n o r m a l COMPUTER PROGRAM L IS T IN G

o d)*0.0 XDATA(I)*0*0 D02I=1/21 Z P M T)*0.0 d ZPMG dZPMHM )*0»0 DO DO 6 J=l/2 DO 6 1=1/5 0.0F E X d /= J ) AGCC(I)*0.0 A G C A>=0.0 d ZPML(I)* B G C C)*0.0 d B G C A)«0.0 d FMOLXd)«0.0 XMPL(I> *0»0 XMPL(I> X M P G)«0.0 d AUT(I)*0.0 B U T d )=0*0D U T)*0.0 d READC5/3000) KDEA/KPT DO 998 KJRH=1/KDEA D01I*1/30 CONTINUE 8 6 6 FDX(I/J)=0.0 1 1 CONTINUE C C KPT-5/DEHYDR0CYCLIZATI0N OF C C C C KPT*2,LEAST SQUARES KPT»3,GENERATION PATTERN /KDEA OFPTS KPT«4,IS0M-DEHYDR0GENATION OBJECTIVE FUNCTION CONTOURS OF METHYLCYCLOPENTANE C C KPT-1/N0RMAL PATTERN SEARCH SINGLE PT.

TABLE E CU u u o u n < Z ~» ^ in i u >-« « > U O D O c D < ( m I—z I—I— o o ------o o M n • •

to c o j c •—I j u 1-0 in in in z < << < UI UJ

o 0 o o: o (D UJ C r-i o o 1-4 -—' S H

O O C .o o o u. ~ < Li. U UI cn uu. u — I— Li- H m o CD < 0> O < < < — > U O w w 3 OJ Cn !-•Z — d c »II ll 3 3 •

K Q U in s i u —OO O l— z (D CD Ui <■ m z »-i s 3 c o CO Z m in o c o o o r 3 o 3 O z

o ui o ui u cn cn tn I- (D »-4 13 U < U CD ~ O -• 3 n U a 3 a (D cn CD J*- ~ z • (3 • r U

Z< Z H D D U

CO -4-CO o o sO O cn CO ID CD ui »- a

OX O O 4-1s Z in < «- < in C O o CD u CD ui C • CO Z m sin u in 3 in OUI O *4 Q cn cj I- Z (D < ~ 3

z z . — CD CD CJ CJ *-> CD CD O O ■>■' ID CD < U w CD UJ UI U CD CD v w 3 3 3 X X X < r ao r- a x < u Z H-W - H Z Z u u u u ' 3 s s X 3 ■'» m

in o uj z h«h h«h m o O u n vO sO H CD CD CD H D '' CD CO O t I U CO CD c o •in in < «- <

•—> 00 CJ o CJ ■* -=t■* tn in C O C O s •»s 299 sO

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

TABLE E n n o o in in (VI # r n n o o in in o o «- - cvj . - i u u u . a x u K t t in v r M M D CDCD u u M W J *»-XlC DU< < D I Z < D l < Z U.<* C tD J*-»»-«X0 U 3 3 CU + ri CD 3 O X . 0 CD X CD X 1 CD< _ X X ~ 3 X 3 D - h (D n ++ + tn D + CD ID CD * DCD ID X ll*— - 1 • X X C CD CD 1 o + u N. x X D X CD < o — o z o CD CD z tn CD t—« i " Z - h o ** — -J • M 0 CD J - - h _J + + 1 o + « - - \ 4 3 X X CO 3 X < 1 ID _J a 1 - i _J + It - -

+m0 2 < \ J H \ 1 II - I- O 3 II • +90 H H X < X in -J > o X X nrcd \tn o \ II • CD CD cn X _J cn uj —4 9— X— r B • 300 rl «—i «-» o cn tn o U. Q. X < \ o CD Q- NO X _i * or + N • • 301

< CVJ a cn x (TV O + cn 00 o o vO cu o » 0\ a IO x ■a- UJ 00 + + o oo < cn o m w o 0 cn o i— o 1 vO o 2 :0 ar-* v0 NJ * • i «-* 00 o + cu + cn 0 0 H o * »v in ~ cn ■*• * cn — J— o * •O vO 0 0 rv x 2: C ■U* m in rv _ i a o 00 X h- r x t - rv 00 m T-t CD (D NJ o cn N J Z a _j -4- o < CD — a in CD h- 2; + _ i ID CD a rvJ CD o D h < a — 1- CD a ~ CD I— X M + 2: o X CD h - a + hi ID CD in 21 id a + w CD < 1 « «— o « ^ «-* cn a O X ' ' 00 x 3C CD ^ 00 _ J '■ 'O 00 ^ _ i x ~ ~ m a * O X cn ~ 2 ^ O • < ~ r UI IsJ " x a • a ^ < x • QrlOH^tnDI-l- » a X z o 1— m a a — n •— x o x h 1- a cn X - II 11 hi a z < < > M X 11 CD VO < > M X H > X X M 11 m < u in — U 2: * - 4 W JT *h Q Q z ll 11 s: U Q Z n 11 x z 11 n a i n i o u i J - a x a o i - a i - x x

rv O r v 00

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 302

TABLE E

COMPUTERPROGRAM LISTING XIS0M*(ZPMT(11)/FMBLS)*100.0 IC4PC4» ( ZPMT( 3)/( ZPMT(3) +ZPMT(4)))*100*0ZPMT(3) 3)/( ZPMT( IC4PC4» ( C0NVs(1*0-(ZPMT(13)/FM8LS))*100»0 G8TB160 DEC8NV=(1.0»(ZPMT(26)/FMBLX(26)))*100.0XIS0Ma((ZPMT(24)+ZPMT(25))/FM8LX(26))*100.0DAC0NV=(l.O-((ZPMT(22)+ZPMT(23))/FMBLX(27)))*100.0 CDEFAX=AL0G1O(1.O/(1.O-(DECBNV/1OO.O)))CDAFAXaAL8GlO(1.0/(1.0-(DACBNV/100*0))) XI SBti* (ZPMT ( 3) /FM8LS) *100.0/FM8LS) 3) ( (ZPMT SBti* XI XIP=(ZPMT(3)/(ZPMT(3)+ZPMT(4)))*100,0 XEIP6=XHEXE*100.0 RATEHKa((1•O+RM0LS>/WMHRH0)*XHEXE*(WHRW/3600.0)*AL0G(CNHVF) GBT016O C8NV=(1»O-(ZPMT(4)/FM0LS))*100.0 ' CRK=CBNV-XIS8M SELEsZPMT(3)/C9NV GBTB150 G0T016O CNPVF»l*0/<1.0-«1.0/C1.0-(XHEX/XHEXE))CNHVF RATEPK=((l.O+R'1BLS)/WMPRH0)*XPENE*(WHRW/36OO.O)*AL8G(CNPVF) G0T075 l(8)+459.6)-.4271))) 79 IF(XDATA(20)»NE»7*0)G8T081 78 IF(XDATA(20).NE»8.0)30T979 76 C3WT=BUT(1)*BUT<2) 75 WMHRH0»(1#17699E-1)*((XDATA(9)+14.69)/(X0ATA(8)+459*69)) 74 XHEXE«EXP<1986.4/<(XDATA(8)+459*6)-.4271))/(1»0+EXp(1986.4/((XDATA 81 81 A(20)*NE.6*0)G8T882 IF(XDAT

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 303

(.8)

TABLE E

COMPUTER PROGRAM L IS T IN G IF(XDATA(8)*GT*899*6)G8T01O1 IF(XDATA(20)*NE*2*O)G0T811O IF(XDATA(20)*NE*3*O)30T81OO IF(XDATA(20)*NE*4*O)30T09O XE»EXP(1986*4/((XDATA(8)+459*6)-*4271))/(I*0+EXp(1986*4/((XDATA(8) XE = EXP(2483*6/(8)+459*6)-*9358)(XDATA( )/(l*0+EXP(2483*6/(= XE (XDATA G0T01O2 XIS0MM(ZPMT(7)+ZPMT(8)+ZPMT(9)+ZPMT(10))/FMGLS)*100*0 CRK=C0NV-XIS0M C0NV3 (1*0*((ZPMT(22)+ZPMT(23))/FM0LS))*100*0CFAX=AL031O(1.0/(1.0-(C8NV/100-0))> XIS0M=((ZPMT(24)+ZPMT(25)J/FM0LS)*100*0 G8 G8 T0 160 C0NV3 (1*0-(ZPMT(14)/FM0LS))*100*0 C0NV3 (1*0-(ZPMT(26)/FM0LS))*100*0 CFAX=AL0G1O(1.0/(1*0-(CRK/100*0))) G0 T0 160 XCYHEX»ZPMT(11)/(ZPMT(11)+ZPMT(13)> CRK-C0NV-XIS0M WMRH8»(1•2575E-1)*((XDATA(9)>/(XDATA{8)+459•69))14•69 + CARK»-AL8G(1 * 0-(XCYHEX/XE))CARK»-AL8G(1 * CNVF=1*0/(1.0-(XCYHEX/XE))RATEK*((1*O+RM0LS)/WMRH0)*XE*(WHRW/3600.0)*AL0G(CNVF)G0T016O XE=EXP(-48lO*/(XDATA(8)+459*6)+6*114)/(1*0+EXP(-4810*/(XDATA(8)+45 DUAL=1.0/SQRT(RATEK) CRK*C0NV»XIS0M 1+459*6)-*4271))) 1+459*6)-*9358))) 19*6)+6*114)) 101 100 90 82

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 304

TABLE E.

)) *100*0 )) COMPUTER PROGRAM L IS T IN G /F.M0LS

AUT(I)=ZPMT(I)*WMT(I)REFSM»REFSM+AUT(I) 002101=1/30 AUT(I )*0.0 B U T (I )=0.0 REFSM=0.0 002001=1/30 ZPMHM(I)=ZPMT(I>* XIS0M=(ZPMT(5)/FM0LS)*100.0 XE»0.4*((XDATA(8)/1000.0)**2.0>-0.66*(XDATA(8>/10Q0.0)+0.932WMRH0*(1.07699E-1)*((XDATA(9)+14*69)/(XDATA(8)+459.69)) GO GO T0 160 CRK=C0NV-XIS8M RATEKM(1.O+RM0LS)/VHRH0>*XE*(WHRW/36OO.O)*AI_8G(CNVF> DUAL»1.0/SG)RT(RATEK) XIHEX={ZPMT(7)+ZPMT(8)+ZPMT(9)+ZPMT(10))XHEX=XIHEX/(XIHEX+ZPMT(14)) CNVF=1.0/(1.0-(XHEX/XE)) 190 CONTINUE 150 001901=1/30 170 CONTINUE 200 CONTINUE 160 D017OI*1/MC 110 C0NV= (1 *0- (ZPMT(6) 110 *0- (1 C0NV= • • 102 WMRH0-(1«17699E-1>*((XDATA(9)+14.69)/(XDATA(8>+459.6))

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 305

TABLE E

6,570)

COMPUTER PROGRAM L IS T IN G WRITE(

,EG).3.0)

IF(XDATA(20).EQ.8.0)WRITE(6/593)IF(XDATA(20).E3«9.0)WRITE(6/594)IF(XDATA(20).E3*10.0)WRITE(6/595) IF(XDATA(20).E3.6.0)WRITE(6/591)IF(XDATA(20).EQ»7.0)WRITE(6/592) IF(XDATA(20)*EQ»5»0)WRITE(6/590)(FDX(I/l)/I*l/5) IF(XDATA(20).EQ.8.0)G8T876 IF(XDATA(20).E3»1.0)WRITE(6/550)IF(XDATA(20),EQ.2*0)WRITE(6/560)(XDATA(20) IF IF(XDATA(20).E D •4.0)WRITE(6/580) WRITE(6/640) WRITE(6/620)XDATA(7) WRITE(6/630) IF(FDX(I/1)»E0*0»0> GO TO 211 WRITE(6/600)XDATA(1) WRITE(6/610)XDATA(2) IF(KNDX*NE«18) GO TO 212 WRITE(6/549) GO GO TO 211 FDX{I/2)»((FDX(I/2}-3UT(15>-BUT(18))/FDX(1/2)>*100*0 KNDX=FDX(1/1)+0.1 F E X (I/1)-FDX(I/1) BUT(I}*100«0*(AUT(IJ/REFSM> FEX(I/2)-FDX(1/2) DO 211 1-1/5 180 HPR8D-(H0UTG/2.016)*(100•O/FM0LS) 211 CONTINUE 212 FDX(I/2)=((FDX)/FDX(1/2)>*100.0 210 CONTINUE

COMPUTER PROGRAM LISTING X - h - l h - l - t h - | ^ - l w - ( w - w ( w h - w » - ^ t w - ( w - | w - I w W X - ^ l W - W l - V ( W - t h h - l V - w l W - l W W h h ^ w H V W W W w a 0 0 ' n i o - 3 - o r i a * - « o o o o o o o o o o o o o o o o o o o o o o —« —« —« ^ M M H » 4 H H > 4 H M M M M < H M H » M H » « •— * W M M M t H 4 H « •— < 4 H I H M « •— M UIUIUIUJUJ0 NO m NO NO m NO >0 vO vO ^*Ni o o t- X X X o < OO r( *■** h < < t- < h* > cn r- cn . 0 1 1 . 0 0 < O H + • XU 3x x U. 3 X > •. < < a. nr or nr a. < — ^^^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ^ ' n O NO NO O 3 n O n a. N —I— NO O o: • _i X ID O coin in o (D CU (D UJ ID • • n O n a X X X X X X X X o CD rXXX XX X X X X X X or O vO vO N M N N N N N N N N N C LC LC LQ _C LC L x x x x x x x x x x x n »-tcucoj*-in'or«»coo>»-«w ^ ^ ^ ^ O n O NO NO \0 \0 NO NO O n O C LC LC LC L n O NO NO O ^ rr r r w w w w O w w w w n — O -4- in oo in -4- — O O n O vO xO vfl xO vO t— CL X X X X NJ NJ • * * 0 - X X

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

COMPUTER PROGRAM LISTING ^ w ^ - o o > o > 0£ 0£ l l— X UJ UJ UJ O H o rH fO H l N l N O I I I - cj n cvj in cvj»-* O O O < vO vOvO — M »—« M < CO «-* 00 ^ H H L C N • N CL (L O O — ^ 3 3 • x x x L UJ CL X X XX. XX. ------— C. OC 1 — ( < Q O v cvicvj cu (xi vO o a o ov cvi O O f- I- w - j j- < < o o uiui CD CD o o (D (D - X < < CD CD CD Q O

CVI CVJ 0 0 DCD CD 0 . u M 0 ID JZ ( Q Q. Q. (L Z UJ CD vO <. ft.X (D CD O —1- t U \ U -CL CU H- \D ftO U) - 1 1— J— _ • • • • • CU CU cu O O O O U. < < w*

___ 3 1 — 3 a—• a 3- w 1 — ftfl < id id 1 x — UX CU O OJ H CVI cvj . u CD * »o ^ «* •> --- ftft—ft ---- - CVJ in o OV 0 1- U) 3 ftH ••ft 4 % i in in - -t CU•-ft H O O vOvO CJ> \ 0 o cn co »- 3 •—* ) = h- * : 307 1 3 1 — — Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

TABLE E in in m Ld LU UJ Id CO LlI CO Id UJ LU Ld CD CO I CO H-i M HI H-i HI xO HI H-iM f— I— fr— «oI— II

D U 3 CD 3 “ 3 ~ cn in r>-~ in >o cn + in O v - l O v£>3 I :> CD 1- —< ^ n r cn 3 00 3 CO •f q h ;

in m rv O O CO CO ^ W W W W w w w w w w w ^ w ^ w ^ w w w w ii w L J U Q w J U U L w iJ L lU W J U iJ L W iJ L iJ L W liJ J U W iJ L I— t— l— I— I— I I— IjJ I— I—I— I— X I— I— H J l— ^ U J U W V cn cn 3 3 H H * 4 oc a: oc vO vO 3 3 3 ^ % ^ cn ^

vOvO'O'OvO'O'OsOvOvO'O'OI— a: or or or a: I i - H I— 3 H o in o oo

vO VO cn o cn ■— X — 4 \ \0 -4 ■4- o 3 3 d * * * : o dc <

h Q • • Q Q i-»— < * < < [ID.Q.ZD - HiH -■ < -—■ H-i H H-i CL Q_ — • • O O 4 4 « ^ ■% > y y y n u c 3 3 3 - l - I O O v k X X W JC ZC W W V W C O d Id Id vOvC <. Hi H-i i - H i - H i H 3 3 - i

___

a 00 *»-» — •*l— 3 r ^ or o ^II ^ 4 CO ^ V a. in in 3 CD V • ID V or• < 0 • 4 • • CQ li) • O •

•-* h C 0 CO y y < 308 3 CO o: V a (D 3 21 CO CD Q HH u

309

ro co

Z3 o ro to

co * * cu ru CD >- ON * kfi X X X 3 < X X ~ \ o «-* cu cu • cn ^ >o X + • < cn •* D 3 — I «H CD CD (D + r Z « o cr cn \ < w X < D CO Q X X s X x X cn 3 3 S3 UJ > X XX 13 cn 5T • CD in 3 ♦ H H _J < ♦ H H < M rvl M Ql O k> w * CD CD CD O (Ti H X X X M M M in u. • NO CO z> 3 o ; c r c r cd cd cc Ut CD O • * CD CD * * * X X X CO cr» no •*• * X o ro o cr cr cr (0 < ^ (T\ CT 3 X X < ^ + < cn • j|r N ^ ^ ***% ^ - " * * < cn O «-f “ w X - ~ -k X O V —- cn cu nO x CD 00 00 X «-• < XO IU 'krt'k — • < X ON H w • ( - • # • • • • • > • • *■» Q_ II J- Cn •H »krl'»X H k»X 00 U X C 3*“*C3C3C3(3C3 a a X o o CU UJ ~ < < JH ------+ + X * •n ro X Ul UJ UI UJ Ui UI ui ui Z • I ^ I— cn j— o o«i-(*)hn^yyy \Q w < » o • • •• * Cl cn — < x X a dhdh - * ^ u jh x it z: cr cr or cr— cr CD k . k . k^ V W k. (VI U I U *-- f- CS n n n X(D32:3XXXX cr id XU.lDU-UUU.U- U U ~ i— t- Q_ cn x u c xi->-x>-x

TABLE E z o CL CL UJ J- I H — u UJ UJ OC H Z cu CL rC o CD cn < u < o. cn n I- CL < • • < CD Ul or «.or •*M * * • o c o cp

OJ U! *»UJ CL US U — U o CL c o rz— z cr Z a l

cn O O O CD CL H-I III H-I 2 H4 >0 o o O Nl CD ' O

o o o o o o m w h o o ro c cr cl oror UJUl UJ UJ v O v O v O v O 3 3 3 3 3 vO VO vO vO ui a. ; u a x r r - >- > »- cr * ^

o * y y c s y * >o y n i CD U U. U U U U >- - U > U. - CD > U - > - CD > — D D CD (DI— CD (D CD (D CU *-■ CU CO -4- If) CU %0 “>~5 ^ ^ CD ^ ^ w w w w w ^ ^ c v c o c o c o r c I I — l I_ J _ X X X X X >->-»-<< >->-»-<< X X X X X M M u M N M a X X X X Ul Ul OC

u u u x x X v >• cr crcr cr CD CD vO N X X CP < X X n n *•» > c o cr n

L V X 3 “ “3 - >- >- X X r cr cr c o X X cr m cr X X CD CD 00 II D * ^ l . u. CD cn m

u u u O .-I O x > - x i ~ c o >- X >- _i \ cd - >- >- crcr CD 00 >- — >- CD CD CD CD _J_J • x l * Kl X X X X CD X u * N n % % o cn > - cr X x

hh cn h cu— CUH- < — • Ul • — cn • — < V J- “3 — X - < I- I < Q — < H-I “3cu o l — I H ll H X O < cr x —cu x a • a o ^ u_o •CD X cr _i >- u u u >- JXQ _J X ~ o cr CD *- X • -> X 00 • ID X CD CD »- * * o UJ H C , w < «, 1 • cn • • o m «o i— cn CU o CD c\ ID o 111 CD ---

C9 cd

cu o I- X I- «C w o o — 3 CU w — o Ulo cr u. u. cr • z Ul • U X UJ x U cn • z CO CD> H-I vO H D C — ' Ul O CD O' CD — n < cn CD CD cn cn O CD 1 — • X •

cu cu o o •HCU cr cr u u u crcr til UJtil w w o o o 3 — — 3 S >0 ^ * vO

»-» (0 u cu cu H-| 00 0 O 0 »- CD .*•cn < < < < o o CD cncu CD 03 X X O Q I- »- CD H- • 310 • — CD 0

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

TABLE E o o o o o " 4" 4 4" Ui Ui Ul oo o n-o cno 0 0 2 2 3 « O J 3 3 « 3 3 I 3 2 x co x u — — %♦W *—• V-«— M • .—. *—* ^ W w W M W h %—♦ M _ •—* *—* M H 07Q7Q7Q7U.Q7Q7Q7CDU.Q7Q7Q7CDU.Q7Q7Q7Q707U.U.U.Q7Q7Q7Q7Q707ID *-4 W „ H >o«oo «N ^ ^

ui ui ui ro ui 12 - y

I— X l Q Ul 4" cu o «-•o vO H- o • tc cu V orz t- ■N

cucu o o ro f~.< Ui Ui o o _! Ui O NO NO or V < «N % o 4 z> -

cu o o o o I— IDlucnouujuicnoujujujujujxooujuiujuiuiujcn h h X I— (D h I—X I— H-I— I—I— '-'XXI— H-*— H- CD «X) o 07 4 - w w c v i4 - C V J 4 - Q7 •- V < N N N N < % o OH- roo - h CD 0 0 0 ^ ov O (D t- cn cn ID NO z or K Q z 70ui < I- i < i- • • V ui y j n O n 4 O - «-• n or OOOI—

o o o o o o o o o ~

Oov NO O O O O O IO V C U -'C O O O O O U C ov ov r4iNr(Jou in>o^ + oro4-inNOrt(\Jooium cn < N N N N N H < < N N N N N N - 4 - 4 4 - 4 - 4 - 4 ~ - ~ < - 4 j \ c j n u r j v r ~ ov ov h d l O 13 CD I— LU<2.l-UJnUiUIUICJ x>x > x > o o * o zcrouirD; zcrouirD; *11120000007 03 • i - < o : i or C3 • • u u u i u c or or c u i u u u • • C3 or i : o < - i • v • - n 3 3 2 3 3 O ------n • vO y (D N U.K s (D y j O n O n O n <—I—O

312

in ui H • z Ul • 0 < z >0 «-t 1— •-I (jl u z 3 _l < » - t- ID CD a § -• — z z in •* OC DC *-< ► H oc 4- U l Ul V V in c z z U l LU m *—«-* u u >- Ui 0 • O ID 1D ID ID _i _JI- 1— «-• a OC OC a < 3 Ui CL u 0 O O 0 r- > V > X r u X 0 -* > 3- >- >- < Z in o_ z >0 X V X X X X X u n X X X XX u in *-• I— U M> — t - DC < H- 0 _i _i OC ^ H O 4- in w uj •-> UJ < UJ «-• Ul < *- u a 1- « t < 4 - cn m w w w CL X XKXX X OC DC o cn u u • ^ » •« •* in 1. in in in x x X X XX o «-<( U O O r l WOO -4- • x • • • in ox ox ox ox ox o o o o o o O O «-« o «-♦ ■r^ cvj m • 0 *0 0 0 0 •* •» * % s * H H «-« ru ro co CD -4- O o o • » • O r l ^ r l H H fH \ \ s \ X ■4- 4- + + + 4- 4- 4- 4- 4- OHO U l U. 4- U U L l X X \ \ N X % •» ^ ^ % * UJ UJ UJ UJ • i0 < W W W H\ \ \ X X O vO \0 *0 O *0 \0 <3 '0 3 * • • N- w w *—» w w -war w ■» i Z h h l - CL 1— 1— 1— 1— 1— 1— 1— H- 1- 1- Ul UJ UJ UJ UJ UJ Ul UJ Ui UJ 111 *-* tt. D. £L w < < < < < < < < < < < 1— 1— H- (D I- (- h !C X X ll - ( L r L L L z z Z z frH »—4 • ^ (D OC OC OC CC oc a OC a oc oc c l a: cl or or a: a o a K d L U - L j ; U. I—

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 313 5Xj 16HMBLSWE 5Xj 5X, HYDR0CARB8N, E/, D TABLE E ,/, 3H15») COMPUTER PROGRAM L IS T IN G 16X>4H0#00) 16X* 14X#F6«2) j j j 19X/F8»0) j 9X*27HHYDR8ISBMERIZATIBN BF C5/C6) 5X/10HRESEARCHER/15X/A4) 9X# 33HHYDR8IS8MERIZAT10N 9X# 0F CYCL8HEXANE) 7HVBLCC 4HDATE j H) 6 // / > ///» (///* 1 (8X/8HN-BUTANJE/ 12X#F7.2)(8X/8HN-BUTANJE/ t 5X /// FBRMAT 1 820 830 F8RMAT(8X*9HI-PENTANE*11X#F7»2) 750 FBRMAT(5X/16HM0LS760 16HPR8DUCT*M0LS/1OO (5X, FBRMATH2/M0L FEED>7X*F6.2///)770 FBRMAT(8X,8HHYQR0GEN,12X,F7.2)780 F0RMAT(8X/7HMETHANE 800 FBRMAT(8X/7HPR0PAME/13X/F7.2)810 FBRMAT(8X#8HI-BUTANE*12X/F7«H) 790 FBRMAT(8X/6HETHANE#14X#2) F7» 59^ FBRMAT 600 F0RMAT( 650 3A4) FBRFAT(8X/ 66014X> 4HCBDE* FBRMAT(8X/7HSIZE/MM)670 FBRMAT(8X# 680 FBRMAT(8X>6HWT.,GM#14X,F7.4)690 FBRMAT(5X^17HTEMPERATURE/DEG*F/6X#F6*0)700 FBRMAT(5X/12HFEED FBRMAT(5X/13HPRESSURE/PS710 • A#10X/F8*2) I PP/PSIA*11X,F8*2>720 F0RMAT(8X/6HV/HR/V#14XiF6.2)730 F0RMAT(8X#6HW/HR/W 740 F8RMAT(5X/16HHYDR6GEN PP,PSIA,7X>F8t2) 592 FORMAT{////9X*593 31HHYDRBCRACKING F8RMAT(////9X/30HHYDRBIS8MERIZATI 8F DECANE-DECALIN) BN 0F N-BUTANE) 595 31HHYDR8IS8MERIZATI0N FBRMAT(////9X.» 0F N-HEPTANE) 610 FBRMAT(5X*6203HRUN/22X/A4) FBRMAT(5X630 F8RMAT(5X/8HN6TEB88K)640 FBRMAT(5X*13HCATALYST TYPE) 591 FORMATC

TABLE E ^"X 1- ooooooooooooovvvvv v ' ov cooocoooooooooooooooooooooovavovovov .c-inininininin' L X lH■rt Ul H o o o o o o u i H o o - n i ^ n w H O cu crv rx 000 00 0 0 X X L CD ZUl z Z < h- cn X z z X X l l X • X 1 «l x * OV 00 OV L h- x^ < X U H X H rx x x x CD CD X CD CD X X X L- X X X t - * < l • X x x u. 1 U. U - f CU rl c ct cu cu *-« < C < Z 0 1 % % • UJ X ^ '—. x^» - 1 CU IX \DUjrxooavo.-icuro 1— IX u u ror or < CL • X X X ^•x L OV rC rH X X X H- L UJ X X < X 00 < CD X X Z l • X X l X X X rx 0 0 L 1— CU <■ L o C_) LJ CD X or X c a X X X X + • l l X X H 00 L rx X CU X 0r» r 00 Va# h- L CD vO _J CU u >- ID < UJ UJ u u ror or l X X X l • M CO «-! X I rx cu CU h- JU.H . UJ U J U lL L— CD < L cn a Z rx < _J X X X 00 X z or < X • X X X l «-» 4" —x• L 1— CJV " CO OV UJ C Xx-* l L XUJ X U X rx •—i fx CU CJ IX 000 00 0 0 CD CD or cr or < < n a • z •— XX X X u X X X l x x CU • X x LJ U. L 1— •»->■ X CU 00 u. •—1 ►— rx X X u CD < < o z X 1 l • X X tenfO X CU ni ni ni nin in in in in in in in o o oo in oo HUN L x x x OJ X X Ll. • : r r x z z x x x x x CU (D • • (D CU N » < C < < < C < < C < < n x cn x t T O D x (D UJ ovovcvo\ovooo

o«HCuro4-tn'Oix ______L L l l

00 n u cn l L IX cu * cu CVJ ^ H ^ H L X X X 0 0 H 0 0 u x U_ lx X x x • N- X X l % % X %

00 ro L < l

CU (U (U cu cu cu cu (U (U CU OO OO o c r n cn on ro cn ro U. Ll U. Ll LlLl Ll U. U. X X X X X X X X XX X X X X N N N N N N N L u u u u u --Z X X X X X X X X X X X X X X ^ ^ ^ ^ ^ ^ ^ ^ % % ^ ^ » • • • • • • •» l ( in H in vo \o H H I n l . L 00 2 l oo oo — — L l n L 2 l -

L • IX CU CVJ — CU OHCU C H O O L «-* 00 cu ro x u X |x a . . a x x o x X X X % X l * X l X L X cu • IX

00 L 00 o o o o o l l x

00 Ll Ll jx rx ro ro L X X X z — U U IX |x X X X IX IX X X X X X x X l x x 314

00 L l

315

TABLE E COMPUTER PROGRAM L IS T IN G 4102 FORMAT(5X/15HIC5/PC5/PCT/EQ*/8X/F7*2) 4101 FORMAT(5X/15HIC5/PC5/PCT/ACT/8X/F7»2) 1080 /5X/20HMATERIAL FORMAT(/ BALANCE/PCT,3X/F7*2) > 5 F9 • 9X/ / ) -X ) (1 (1/ 12HL0G 5X/ 4000 (./// FORMAT 4010 FORMAT(5X/11H1000/T(ABS)/12X/F7*3) 4040 FBRMAT(5X/19HL0G(1/(1-X))/DECANE/4X,F7.2)4060 FORMAT(5X/20HL0G(1/(1-X))/DECALIN/3X/F7*2> 1060 FORMAT(8X/7HT0LUENE/13X/F7*2)1070 FORMAT(8X/12HC8 AROMATICS/8X/F7*2) 2060 FORMAT(5X/13Hl/(W/HR-W)/HR/10X/F7.3)3000 FORMAT(13/II) 4020 FORMAT(//5X/16HDECANE4030 F0RMAT(5X/17HDECANE CONV./PCT/7X/F7.2) CRACK*/PCT/6X/F7«2)4050 FORMAT(//5X/17HDECALIN C0NV./PCT/6X,F7.2) 1030 FORMAT(8X/3HMCH/17X/F7*2)1050 FORMAT{8X/7HBENZENE/13X/F7.2> 1090 FORMAT(5X/20HHYDRBGEN BALANCE/PCT/3X/F7*2) 2040 FORMAT(5X/9H1/SQ»RT.

TABLE E n in in X X L tvrv L - Q CL X X X % H H in in cn *—• w vO vO i— i— a o 00 I i noooooooo o o o o o o o in 4-^-4-4-4-\0%0\0N0v0ininin -in 4 n c n c < < d x c d < c x d c < x d c d x c < d c x d < c d x c d < x c d c x < d i d x o < d c x < x < x x < < ov oh uco hc ho vo oo o •-ICUOVOOOOOOOOOO CVJ (VI a c l l c a o o u o LJ < 1 X X X LJ U u u ororororororororororororor ^ \ \ ^ * • • l •» *•». l VV H % V -• L 0 0 L l l

I l

in n in — s—I tv n cu in - o o «-• cvj 111 Q. t— cn c n in x —• > CL «H «H CL z u X oo * or — U. L lL X «. 2 V X CD X J OV CJ U — «* «-• v l

H UL v

" - 4 OJ Ll Ll ■» i n »-•inin - X »-t vO DC ori- 2 U cd CL CD v Z 0 v 00 < u i i u < - » % cn MUJ 00 X «-• x x < x CJ UJ h Ll X I— z ii. » • » I I CVJ II X II •» v ^ X 4 4-O v * U ** X v «. in l cn • cn L CD l

H H I-• CO U ll Z 00 00 X X h- •* X •-< Nl L X O l _ Z < Ll z X D CD CD u X X V V •» u l

< n 00 u 4 I \0 or. l -

tv — n z cn — •UJ X L Hrl OV H H UJ v OVO Ll X CL Z tv X 0 0 V U 0V>- X U O - l . U - 4 L O i I h oo I in I Q U h J x v a. x ^ U. U >- J (D 7. i t U C z u X X CD UJ >- •-• X X X X X X X or. » CVJ * •» ItI— v • l v v UJ • Ll V V^ V CDv UJ «-* l Ul l _ X • n Z • J — t O

_j x _j (VI ohcu L 00 l

fv cvj 00 L 1 l

oo I l

U J C U J C U U J X / C U OCDXXXXXXX ZtUCDCDCDCDCDCDCD D CD o Z Z Z Z Z Z Z D Z x x x x x x x r o OCDCDCDCDCDCDCD X Z - i UJ cn u x cn H > in - i Q < - i u in CD OJ L < L - I < X o 33 Z> cn o n cn cn u ~ - * cn cn o o ^ < cn — o o zd «c X u I— u x in cn o o < X O O CD < CDZD D v CD CD - I CD x _i u CD X U CL CD < v v — ^ O v O o o L l OJ Nl o X N! lL CD O _J — —_J CVJ — D * CD U l N INI II in ~ o cu o (D I— X v X n - * cn o X cn CU O CD >- a X X cn ^ l _ L i C CL x x — _J —in U CL cn v L CL O CD X n v cn X X l % v v V * w h *-> >- >- x

L v CL CDUJ x a CD 3 “ cn x X X v x cn CD X or v a _i ■* x t- < j u N C or o >■ or - t «> < CD 3 X X X X < % X u LJ X >- nr x >~ r o ori- I—CL < V X O >-x 3 orv l U ■. ■» cn v % V CD v %< ■s X X v «s v x «.

uj or or or or or

' tv cn ft—i Ul z UJ X z cn cn x CD CD ZD cd Z z •—• OJ X X _ * o OJ cn cl w 1- UJ CL r—« X % OJ o o in »- CD lL UJ •-I <3 X CD CD • • • z X CL t- 4 LCL CL L— UJ U CL J_l _J u u u as U. X X < CD • - 1— in •— L— X CL o <3 L -— _J < CD l • • « nr o or Ul CD i- z ZD«-l tv »H in o o cd u M • 316 cn O o ZD X e Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

CU CVJ I- =3 — - • * % »-<•=*-n D DD O It cno o O < 5Tu D ID ID O O «-- CD z cnz z + - iz % r

d *-• id

cu o

> cn

Z » HI % % ■% x I- H »• Z Z '-'3cncn»HCucL'->“ '-'3cncn»HCucL'->“ — H t r H i i i w n i o H(VJ c O x K 3 D Q''4-in D D Z W U C N 'O Q Q w n ( 3 2 3 U ) W n ( a . Q cncn z z oo o o z z z —

--- \ x a \ \ D O cn •-* »-• > n cncn »- > ___ xO cu u ID uj CL cn * % CL x I—I I— I—

cu uj a. ■»

cn V *—i u z u ►—* z — z ii y y o o UJ i u UJ i u Ifi a in 4- I- < < cl I x ■ a •• • • x a h- r-t

O JI- I _J CO CL O w % »-»o i r l CJ CJ l r i 3 U O id UJ O »-• X O B «x •>-cn < cr < CL «- «- CL o Cl- I- Ul DC -J •-* cnx

xD O l UJCO Ul ac< ca c n o cn o J X X UJ n in • a • UJ z < < < < Ct I- I— CD I— I— z % + + •x H a a H •x

J UJUJ y y X CL - I — i a a u u ■a- in UJUJ X I • • » • h

. u c o u 111 CO O D J ID UJ ID H- z o H- —< Z crZ < z i- cr ID rt X • u n > to (D (D UJ Ul X CD or Z z j cr z ID lL \ \ in X UJ < O Ul CO «-• cu o o o o CO H :

CO CU «-* t - • OV z n ^ in O -X CO x x s in x X-I I— ov cu X «• y L U in x -x O L X O D ID ID cr cr < - • * % \ •I x X •x •X UJ l

cu « cu < X >- o X - h < l x

318

to > I- < o U- « k

to u I- c o z r o cn r x x tz co N X O * Ul _1 X oc r-f O © N etc % ru iu ro X o ~d m U. x m X c »- n w * cn *-» UJ r - z d cn uj IN 1— < o ~ O © j x r l cn o * o ro X © •» N X — CO w >- X X »- ■% o X CD OC > X ~ ro < X N N * 3 »- o ~ U X 3 CO \ X z »» w ro h- CD CL o or > Ul < ro — z d 33 Nl CVI I •> IN h- J— Ll CD N 3 QC U i c n * U «* ru * X N z c n a CD -* O O «—i O X r - © © n O CO to w u j cn ' cu u X a ■4- — ro _ l - >- V *■«* to X u — — o * n CD C CD Nl Ll < < ■4- • =3 X r l X • -— X X «k % X ■» * - t CD cn ZD *** © *-« CO o o r ~ «-» *—-—. O X M + + U) X 1—«+ * N « * o o o CVJ >- NC b - X + •k _ i cn * ■4- — ro co ro ■k ■s © »■* l - X _ i X © M _J • o — to © vO ro cd < CD UJ X ov 3 \ «H »H X in • «kC3 X o U. w CL U X X CD V X vC N '— «-l X X o o UJ O' • < . N U X CD X Ll) X < Ul • V- - + X o o c H b - X X X X < NJ U- * N J - -*- •—* • r -f X 1- CO + * z Nl N + X Ul rv U 3 •» <* •» N 1- < cn «-< 3 X _J © m c n CD U X x X b - •» \ 1— b - V 3 © _1 3 V i UJ J V V > h— •a- l - ~ O o o OJ X UJ ZD =3 -—. V X © X Q x © a n h a • 0 . 0 ( 0 ro co N ■k X OJ ov c n m CU — X X X UJ u « X Ui '— CD CO w w to X 1- o + + 1—1 H w X 3 X X \ X X X Ll — _ l U _ l w X •«. X < »-« X X w *H S \ »» CX3 x x x n UJ 1- CL U X X •k X t - * u U 1— w Q. \ o o UJ* \ m o X Z X X CD X o U «k z < O h j X D h U © • o X cn cn + X V X — 3 : X < NJ u .X < © o • H Z> UJ ZD © D X _J H H 1— Ul —J X N N b - »-4 X o •— cn ZD CO © U © '— x © a *-i cvj cn h - OZZ z z z z z cn H II n Z N U u ©X II © c n x X Ul C © CD CD © © © © © z cn x o x —< X © X — X < « *-< X X _1 _1 _I X OC X X X X X X X UJ Ul U rIU I— U J J « —' 1— o » © © © a o o D z r X X XX XX X X XZ X © © X u Ul < t«r o X X X (DZDOD CD © © © © *-< X Z3 © Z3 © ZD X X u © X Ul C3 n M u >- >- >- Il UKOUU u u u CJ u Q X cn q cn >- 1- X Ul < X < CD O CD o co

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 319

k m m # k j r h , r m o l s

, w h r w , TABLE E a e q , y t , ZPMT(30),ZPMHM(30),FM0LX(30) t COMPUTER PROGRAM L IS T IN G y c p ,

k d e a

k p t , m r , KJRHJ-STEP(1) m c S t SUBROUTINE BPTNCH COMMON ZPML(30),ZPMG{30) COMMON END C0MM0N WMT(30),WMC(30),RHO(30),GCF(30)C0MM0NFDATC(5),FDATV(5) XMPL(30),XMPG< 30),AUT(30),BUT(30),DUT(30),XDATA(21)/ C0MM0N AGCC(30),A3CA<30),BGCC{30),BGCA(30) WRITE (6,130) THETA WRITE(6/150) YCP COMMON FDX(5/2),FEX(5,2),FYMOL(12,20)*GYM0L(10,10) RETURN CALL PATERN(2,P,STEP/NRD,10,COST)WRITE(6,110) P(l) WRITE(6,120) P(2) WRITE(6,140) Y3 GYM0L(lO,KJRH)=STEP(2)WRITE(6/160) A,H,AEQ,SUMCP,XM0LB,XM0LM GYM0L(8/K J R H )-P(2) GYM0L( GYM0L (4, K JRH > *A > JRH K (4, GYM0L GYM0L(6/KJRH > aAEQ GYM0L(6/KJRH> GYM0L(7,KJRH)»P(1) GYM0L<5/KJRH)*H 150 FORMAT(8X,16HM3L,FRACT.,CRPR=,2X,F7.4) 120 F0RMAT(8X,15HRATE130 F0RMAT(8Xil8KC0NTACT C0NST,CPR*,3X,F14.9)140 FORMAT(8X,16HM0L*FRACT*,BENZ*/2X,F7.4) TIME MIN.=,F10.5) 100 F0RMAT(F1O*5,U,9X,Il,9X,4FlO*5)110 FORMAT(/,8X,16HRATE C0NSTiBENZ»/2XJFl4»9) 160 F0RMAT(5X,6H0PTMCP,6F10»5)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 320

RESMM j

TABLE E ) COMPUTER PROGRAM L IS T IN G I0/P (1)/P (2)/STEP(1>/STEP(2)

AKR/BKR#AKRR/BKRR/YB/YM/RESBB

NRD/ j

THETA j *K JRH)*STEP(1) 3 CALL PATERN(2,P,STEP*NRD/I0/C0ST>WRITE(6i110) P (1) GYMOL(10/KJRH)3STEP(2) A«((XPRES*1OO.2O)/(RM0LS+1.O)) AEQ3(EQK+H)/EQ< GYM0L(3/KJRH)=THETA.GYMBL(4/KJRH)3A GYM0L(5/K JRH)3H GYM0L(6/K J R H )3AEQ GYM0L(7/K JRH)3P (1) GYM0L(8/K JRH)3PGYM0L( (2) H»<(RM0LS/(RM0LS+1.O))**4)*(XPRES**4)GYM0L(I*KJRH)aYTKJRH)aYCP 2* GYM0LI THETA«(1.0/WHRW)*60*0 XM0L7*(BUT(15)+BUT(18))/WMT(18) =(XMOLT/(SUMCP+XM8LT+XM0L7))YT READ(5,100) EQK/ XMRLT 3 BUT(17)/WMT(17) 3 XMRLT SUMCP»SUMCP+BUT(I)/WMT(I YCP=(SUMCP/(SUMCP+XM9LT+XM0L7)) XPRES=XDATA{9)/14»69SUMCP=0*0 D0 10 I«l/10 SUMCP'SUMCP+BUT<> <14 14)/WMT DIMENSION P(2),STEP(2> C0MM0N A/H 10 10 CONTINUE C C EQK IS THE EQUILIBRIUM C0NST F0R NC-7 -T0L

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

F14*9) j TABLE E COMPUTER PROGRAM L IS T IN G

* * H-» A = SUM( 1J/DIVS SUM( ‘ = A to DB 2 1=4/12 SUM(I-3>=SUM(I*3)+FYM8l(1/J) DB 2 J=1/KDEA SUBR8UTINE MULSVR CBMMBN A/H/THETA/AKR/BKR/AKRR/BKRR/YB/YM/RESBB/RESMMDIMENSION P(2)/STEP(2)/SUM(9) DIVS=KDEA S U M (I)=0* 0 S U M (I)=0* END CBMMBN WMT(30)/WMC(30)/RHB(30)/BCF(30)/XDATA(21)/FDATC(5)/FDATV(5)COMMON XMPL(30),XMPG(30)/AUT(30)/BUT(30)/OUT(30)COMMON AGCC(30),AGCA(30)/5GCC(30),BGCA(30)CBMMBN ZPML(30)/ZPMG(30)/ZPMT(30)/ZPMHM(30)/FMBLX(30)CBMMBN FDX(5/2)/FEX(5/2)/FYMBL(12/20)/GYM8L(10/10)YT/AEU/WHRW/RMBLS/KDEA/YCP/C9MM0NMR/KPT* KJRH/KMM MC/ DB 1 1=1/9 WRITE(6/140) YT WRITE(6/150) YCP RETURN WRITE (6/130) THETA WRITE(6/120) P(2) 1 1 CONTINUE 2 2 CONTINUE 100 FBRMAT(F10»5/I1/9X/I1/9X/4F10»5) 140 FORMAT(8X/16HM0L«FRACT./T8LU2/2X/F7.4) 150 FORMAT(8X/16HM8L.FRACT./CRPR=/2X/F7.4) 120 FBRMAT(8X/15HRATE130 FBRMAT(8X/18HC0NTACT C0NST/CPR*/3X/F14.9) TIME MIN»»/F10.5) 110 FORMAT(//8X/16HRATE C0NST/T0LU2*2X

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 322

TABLE E COMPUTER PROGRAM L IS T IN G

P/ STEP / 6/ 2/C0ST) 6/ / STEP P/

('2/

COMMON AGCC(30)/AGCA(30)/BGCC(30)/BSCA(30) COMMON /ZPMT(30)/ZPMHM(30)/ZPMG(30) ZPML(30) /FM0LX(30) COMMON WMT(30)/WMC(30),RHO(30),QCF130),XDATA(21),FDATC(5)/FDATV(5)COMMON XMPL(30)/XMPG(30)/AUT(30)/BUT(30)> DUT(30 / SUBROUTINE CSTGEN THETA»0»10 130) THETA/RESBB/RESMM 6/ ( ITE WP. END DO 10 I»l/17 CALL PR0C (P/COST) 1 THETA*THET.A+0» RETURN WRITEI6/110) P (1 j / P ( 2) ( P WRITE( / 100) j WRITEI6/110) (1 P WRITE(6/120) 31 KPT P(2)3SUM(7)/DIVS CALL PATERN STEP(1)“SUM(8)/DIVS STEP(2)*SUM(9)/DIVS AKRRaSUM(A)/DIVS BKRR3SUM(5)/DIVS P(1)=SUM(6)/DIVS AKR»SUM(2)/DIVS BKR»SUM(3)/DIVS 10 10 CONTINUE 130 FBRMAT(4X/F7#4/7X/F6»4/6X/F6»4) 120 FORMAT(///5X/5HTHETA/10X/2HYB/10X/HHYM//) 100 FBRMAT(1H1/////5Xj33HGENERATI0N110 FORMAT(//5X/3HKl3/FlO* 6/2X/3HK23/FlO* 0F CONTACT 6) TIME CURVES)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 323

in > i- < o u.

in u 1- X X c o x X O CO ~ V CO Ll o •»UJ X »-« I or — 3 «. or h «-l O ID O CO CU CO X H V CD w I I " % CO < h - «% _J cn UJ \ 1 - 3 ^ CD _ l oc < 0 ^ 0 x CU »- Q > 0 ( 0 >- 51 X X (0 X ~ CO ~ (3 QC >- ID O ^ X ■* > z (J CO < X ^ 3 to ID CO I • o w u X O O' 5- •—• •* CO 1— CO CL tU X * 3 tO Nl ■H 2 OC c X U. CD •» •» OJ «. QC OC 1—1 CO — U < n a X O «* H CO CD ~ o o UJ CD z cu COCO > vO _J «»O CO CO _ l < * X UJ o CU • co w W CD *s OC 3 z x in X o — (J h X t - oc CO z _J w x » -« Ll CO 1— CJ X > - 5- X •—t CD < X 1- (U UJ 3 CD CL u . *>< > * co * "> J O < f l ) N «. CL * CD OC < CD X X Z X > > > ~ CJ CJC CJ t - cd CJ CJ o in QC CVI >- X z M z z * •> X •> o o o *> % cu ►HIH CU •» in s ~ CO 10 CO in < X CO ►— X cu 4—4 r l • J o ------W — UJ O' * w QC * X D- Z CO CD < CD X o X w < cu CL X I U l M W CL U X UJ V < V CD 1— X CO cu ^ U U 5" to CL u . ■* % w CO X % X ^ X X < r>J •> I - < z H r l cu x in X 2 > > > — Q. 1- CD «-• UJ H (/) < CO •—•fs i o %•--■ 4^ 4—> CVI V UJ X QC V cd a . u u •« > in LU o o o % «* X CVJ oc u . 3 U 3 X ■»X •—< X O * - i * * cu + + 4-1 UJ I - O l u X Q CJ * z CL X O H H H U H I l IX Z X X CD Q. U. X < CD CO h- H •» H CU If It ID * H UJ i—1 CU UJ ■4- r - i in h 2 X < N M o * vO X X - J cr »o—- 3 X 2 '— h- z z z CO z in ~ X X a. «-»Q_ Z X CL Z Z I— 1— I— 3 Z Z Z Z ID ID D Z < CO < — UJ II II o o UJ tt • n It M QC < < < c o O O ID X X X UJ CJ 1—* Q h- CVJ _J 1— 1------1- 3 X X X X X X X X X X X X z < «-i CVJ _J Z ^ CVJ Z 1— QC Ot QC O CD X x x x CD CD ID •—# 1—1 < X UJ QC w» w CD (D < or w CD CD UJ CD ID CD Z 3 ID ID ID CD u u U Q XM X QC 2 0.1Q O U 3 CL U CL CL U QC U . I l U . LU CO U CJ U O o o o o (VI OrlftJ

u o u

TABLE E asc H !Ui X «-• v m cuyui . z CDU. ■* z < X -ft X •»3 z ^ X in or»- in < E D O o x c t y:~ co ■" U CL CU •» CL 3 CO^ u u u u -ft UJ £0 CD CDz z z CD z z z o o «. o o C/J — * o Ul CD CD CD CU >- V u.* <. U-■» * ''O.l- sZ s >- o z z r - >- ■»V O tL_1 J X U UJ X < cd _ Z 1— CL y o x •*** £D ■ft CL% ■» or* ^ «*^ * **■» X •ft H- - < •» < UJ CL z z 3 z z o: >■ % : U O M U 3 U O O «.» -—. cn " J O JH30JQ. Q J 0 3 H UJ _J O u »- Z U O H Z •—*CD>I— CD^DCD>U _l — z — inz o no in in CD > M o H • o o •H I3II 3 • 1—1 UJ z 0 ino — f- Z CU CD M • ^- H Q »— o ft— j- M o < < D < X H \ J- * • w* H ft—ft U. u o Q »- z + U • a O o u Z CL 1 ft—ftu —ft CD —• UJ > uCLcu t- CD CD —ft • • « *— 3 U — — ft— M >— r o D h u z D 3U.CD _jCL 3 _J h- Q 3 C J C Z X D C D ►H CL —ft I CD * ii — i UJ t- CJ 3 1- H 3 —ft C/i *-« ft- ►H Z 3 I— CJ + II — »—« — ll. U. U. c s II _ . i *— -ft J— z ft— toX Z z U ft—ft ft- ft— ft— CDZ -I CL Z CJ 3 CJ •- 3 z z Z o z \3 \ 3 ft—i CJ \ II ft -J cn ft— ft- W a z u i- u. t- + N *>m+ z I U Hi u _l — *— H u ft—ft T M ft— — u 1_1 _1 CJ Z *—• w ft-ft -I—»- Vft 3 z LCJCL CL + 3 z z ii ♦ CL U . L O U I ft- ►ft f t —f t C D C D Cl C D ft- CL 3 ft— CJ ft— ft—ft ft— Z o ft— rft ft—« CJ — UJZ ft— I— _J J- CJ \ ft-ft 3 z z — -»• II cn * m 1 cu o ft-ft D n3 in CO O H m u « OI D UJ^ ft— J- — 3 ft— (j O o o »— -—■ »— X -ft— N ►—« Z 3) N -ft z • 1 ft CO o ft—1 z U U U W 1 U K ; m o i- o too z o z z a; C I C O D Z I U o:z z z z ^ z 3 z Z < X T 3 W < x : s - l~ o o o ~ cl ■> 3 -> X . o 0. o o * ~ I— V CJ CD5" V i COCO Ui i r z 111 0. u I- COCO CO (D O —— CO H CO X CD C t- . ~ ~0. u ~ *-» ro w -► to •« JU J - ft— CD ~ O «-l O H COU H cu cn ( « ~ u 0 COI— CD ~ O «. x " O - O C - »- i c II c •»cn u u — o o d l * * too •% o — < cn < ------3 Z —ft 3 3 Z

** cd 324 ao O co

325

0 z 1 : CO Z i t ~ ~ V C/l X 0 •» Ul j ^ x q : ", cr * II O “J CD u - w y 00 ~ > cn — _icn u j 0 «d _ j a (OIOS >- IT ST Z CO oc > - X * ^ * Z 3 CO w D . O K > - M OJ X n w •» •* z a -» CVJ *4 QC O 4-1 3 yc (0 w UJ CO I ~ _i c •» 1— id ^ or z z 1- a. a . >- > - v M u •» < o o CL % • » ^ - u o r o o 0 ru >- v o n. n CO % CD rv 4- i n < * 03 CO UJ on O f- z x a v CO 03 < < o . u i y < 03 co CJ (J < M U. *4 •> h- CO CO o •4 > 1- < (D < CO co --- a i - CO < 0 ru v u j ^4 ^ 4 ^4 O f- CO *4, •» X 0J 4-4 4—4 + 13 «-t CVJ — i n oc \- a ^ CD _ 1 ~ Z 1 0 0 < u < ui m N (3 C3 O Z X ■* I 0 0 0 O U J U U u u c j c j u • U Ui Ul U Ul CL 0 CJ *4 CO • • • • Z CO CO IT CO CD r Z • • I • M U. Z < 4 Cl O O Ul Ul 4 < < O •« CD CO Ul * *-«3 Q- Ul Ul Ul Ul % UJ 4-t III II » D I L H 11 n ID H H II TD 4-1 — < Z 3HQ.I1. 3r(Q. Z Z Z Z II «-• 44 « Z > " # — ^ M Z II |— n U Z » >• >- 2 # > ID CD CD ID *-* —* 4-4 4—4 1—4 |-- 4—4 4—4 4—4 T> *-4 hH 4—4 »—* 4—I h V 4—4 4—4 I f— »—4 4—4 |— z z z z 0 — ------1 - 0 --- »- o < CL w h O ^ 1C 1— O M Z Z Z Z J - 1— l— Z w cu 1—» - t— I - Z + 1- t- Z 'O '- I! t - Z 00 *- *-■ Z tn *- (DO(D(D(D3D3(DU.(DDD (D (D D D ID D Ii. U ID (DU UUUUO

o o o o o o o (VI CO -3- in (v 00

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

TABLE E TABLE

COMPUTER PROGRAM L IS T IN G

ZPMG(I)“CUT(I)/WMT(I) IF(C0GAS.LT.O.O) RETURN FN0RM = C0GAS/WS’JM = FN0RM MC DO30I»1/ BUT(I)*FN0RM*AUT(I)C U T (I )=BUT(I)*(W MT(I )/WMC(I )) WStJM = 0.0 = WStJM AUT(I)«AGCA{I)#GCP(I)*WMC(I)WSUM*WSUM+AUT(I) 1F(KTYPE.EQ.2)BGCA(I)»DUT(I) COMMON A/H/THETA/AKR/BKRMKRR/BKRR/YB/YM/RESBB/RESMM AU T (I)=0*0 BU T (I)=0*0 C U T (I)=0»0 DB20I=1/MC DUT(I)»0«0 CBMM0N MC/MR/KPT/'

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 327

TABLE E COMPUTERPROGRAM LISTING

XMOLL»XMOLL+ZPML(I) W T H Q “XDATA(17)/(XDATA{18)/60t0) BUT(I)=FN0RM*AUT(I) WSUM=WSUM+AUT FN0RM“WTLIQ/WSUM B0 U T (I)=0* COMMON MC/MR/KPT/KDEA/YCP/YT/AEQ/WHRW/RMOLS/KJRH/KMM COMMON /ZPMT(30)/ZPMHM(30)/FMOLX(30) ZPML(30)/ZPMG< 30) COMMON FDX<5/2)/FEX(5/2)/FYM0L(12/20)/GYM0L(10/10) DO 101“1/30 DO CBMMBN A/H/THETA/AKR/BKR/AKRR/BKRR/yB/YM/RESBB/RESMM RETURN END SUBROUTINE LIOX(COUTL/HOUTL/XM0LL)COMMON WMT(3O),WMC(3O),RH0(3O>/GCFC3O>,XDATA(H1>/FDATCC5>/FDATV<5)C0MM0NCOMMON AGCC(30)/AGCA(30)/BGCC(3U)/BGCA(30) XMPL(30)/XMPG(30)/AUT(30)/BUT(30)/DUT(30) XM0LG*XM0LG+ZPMG{I)DUT(I)"CUT(I)•BUT(I>HBUTP*H0UTP+DUTU) ' ZPML(I)*BUT(I)/WMT(I>' 10 10 CONTINUE 20 CONTINUE 30 CONTINUE

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 328

TABLE E STEP# NPASSjIB#C0ST)STEP# COMPUTERPROGRAM LISTING j END SUBROUTINE PATERN(NP#P RETURN H0UTLsH0UTL+BUT(I)*(1*0-(WMC(I)/WMT

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 329

********** * * *. in

* X * X 3 * < < UJ * X < O 3 3 X o * Z Z C3 3 * CD Ll 3 < X < z * d x > * 3 cn c d . x z in * O LU < ♦ x x cn < u * o x o LU * x x cn x XX c o z z * t-t UJ < * CD o ro z z X • NJ 3 X ~ V to * U X X 3 * o O - LU LU 3 0 u X ■ H I K * X X O X 3 * z «-• o d «» X «. X U 0 w < • d cvi oo z O "J CD * LU X _J U X * L4 • U l n y a i cn cd o cn < cn x < x — -* c/> * 3 X UJ X 3 d n * z z » - 3 _i cn lu cu > 3 0 x cn X < o O CD I CL * X X U LU X * z ** O ' o ro Z d * cn CD U l LU U X LU • - cn X ~ ro ^ >- z z * cn i o x UJ X * O X •»o — z id x > Z 3 X U l cn lu X ~ oo c x •i % OJ * LU z cn 3 Z * X X o ~ u z ~ 3 CD X X X CD < CD X oo »- CD OL O QC > w * X b o w i n X X * z >" 13 33 N l CVJ I - % •—1 X X CD < > < Ll CD N % 3 cr * DC Z LU < 3 X * 3 X L) *<• oj - cr OJ 3 UJ 3 3 CJ z z < CO ~ o o H C3 v 52I * X 3 O CJ ID X * d X ■S O ro oo — UJ CD OJ 3 LU x c *-< d x d ^ ro 3 < •»CD * CD - - > > X * O X 0 U X d «* cr X z M X X d ro x u z z x x * X 3 X < O X LU * X — 3 CD X x x v OJ cn < *-• 3 cn Z X 0 3 < CD NJ L l «* < * 3 UJ 3 Z X * X 1 «* Q_ ^ -• 3 X 3 ~ ro ro ro CD *CJ 3 X ►-< X 3 * •» z o w . in «r •» OJ LL CD X CD > 3 CJ cu ro cd < CD - u o : *X x X CD 3 X < N J Ll Z < o cn ♦ CD s •—« CD X ** z ►-> < z X X CJ UJ u cn * z z z z z z z cn x * ►-4 — C 3 Z •* CD * * I— 3D CD O d O d d z z 1— cn 3 3 d X u X K x r z z z z z LU n * X O d d * X < z z z z z z z Z O «-• d Z - 4 X d CJ X d h 0 (D d cn x CD d * z W ( J u u u u u u O Z 3 X d X * I * 3 CD X * * I cn * I * * * * * * * * * * J* * ** t * I u u

TABLE E m w D ( - I U oj a UJ a c cr — O t- ii a in a. Q z i OJ -i « a. ti a. O CD CO l u ll — 1 ID < I— ID 10 I . Q — I ' - ' t—i LU I II I(D• —i «-• 0 in (0 .

u y U c => 3 0 Ui ■< ozco z UJ O 3 0 • CO ZD c LU > u •—• (J z ~ o ~ z u. < z LU CD -"H " |- l f L ^tJ IDZ t_J CL^ • o ~ i r o fiu w ZJ -nn I- _J D ** 33 ID * *-* 3 j id •

1 UJ *. o o o toin ID vO >- _J vO ID id • H • UJ - w ~3 ~3 o O a CL o ~3 M ——>0 1 Ul z a n - «*>- % % or ac h- ZD z UJ «-i •-« 3 3 U. QC 3 «-i•-« 3 o a. t_» — U QC CC Q u u QC—U Q CC t_> as •< oc V * CVJ h a H iiu r ''0 U^OUJ « OC «*U -H o o O O O -Ho «C o ID w «— * 1— ( Q_ w »H w M w«—* 1—( Q_ EH -) U -) EH -or M H - — — < »-<1— •-*1— 1— 1— j- i 0 1/1 0 “3H- tL z UJn %< ID w Q > > ^ ^ w Q Z n Z o cn to Z tK — ^ D <1 OL

cvj ~ cn o 03 • in

^ n •

^ o o o ^ id ID^ ^ — 3 3 —3 UJ to o - — h- J'OvO D “J ID »Hrt co o q - ir ~3 CL Z ■ * J I OJ MQ- Ll UJ —• H- z> l _ O oc < too I-O ■oc • < I u »-< «c (D CO CD z Z3 I I I I

1 o ~ o o I— «—I ►-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

TABLE E •H •H J ( i h C - J D c n l U < U. < Cl J CD CJ D CL CD 3 + . ~ Q. J— O O CL • CO h • > z — CD O CD - J Z - J »-O O0 O M J J OJ %O - O - M J T C —• CO CJ CJ U CD !o _! _1 •* H _ + j o o o ^ cuo a ' U M CU OJ CL co o o coo

CO CO H u c u c OJ cu ru co CU uCD ru I H . u u 1 % % OJ »— UJ —* ►— s CDCO 0> •s- OJ •—# D+ CD HI- Cl . u DCOCD cn • • A OCDCO f—I hH u 1 CU _J »- CD L S3 *—• H-I OJ u *-• *-• H _ Jt- < H-I < II l j I H - i r r < z ti O h- CL ►—1 w w ll CD CO Du - h CD CD CVJ H-I tn H-I - t n

- CU l r U It m ro o U Ul - J Ul UJ Ul J - Ul |— H-I H-I I—I 7 ^ —— —^~7 CD ir> CO L CD l CD_ OJIL J - CD CD CD J—CU «-< • • l _ U U - U H-I H-I H-I . U. U U. CT CT Z — i D iUi P)U rt(UH CO O O CO O 0 * J _ •—i U • -h .4 h- h- : 3 a CL O vO 4- : 3 z w t c %

. a u c o r H o r o r u c co co «-* cu H O U . U u H4 u » * i f f Q M m (0 » n c H * M z M % 331 CU # w M CD II Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

TABLE E CD w OL w M •—• cu M m w — nu oco o »— in— u CD OJ co H I M II « OJ OJ o O h nc J ( l J .l_ U .(D lJU C cD tn H « D C •o u o ii coin •4- in « h m »-i z 0. M «* »-4 w J- w H- II CO H H 10 H in CO CO m J (D OJ u ~ cr> ~ m (L « < CD — CD UJ o — < co + - Z «- * II *-» I z CD to CO ON H- n rl CD UJ z f- I H I H hi a . V W. • • a O |— l r l |— r • |—HH h h -OO O O - K UJ • — i— H- •-< _J vO _J \0 •-< UJ COO O UKC K U h o; + • N O H J C ■4- O UJ H-»-- w M HI M w ID r H I H ^ ^ - | T h D ( cd

r~ u OH ^ H 2 2 uj % “5 ~

0 uj Z * -

< u UJ ID 0 u < UJ -I 10 UJ u cc z I- I I I I I . o «-»in } — ■ o COOOI-Q.IOCJ—• 4 M • -

rl UJ a HI H-l ►I • • 1

in \ \ > j- < o co U z X CD % in X o U CVJ »- (VI * < o r x vO < a CO 3: x CL > u. v co * in UJ * X o x Ul x a Ul * _l H 0 CD x cr % X < CU CO r 0 ~3 m Q. z CO H- u »-i V CC * ■ CD CO O < 1— X x (O 0 1— •—«or z H- 3 H- _J CO UJ CO I— Ul 3 < O *"XO CD _l CL #-x» CD or CJ 1- CD Q X O CO X CD x + Q. U UJ Z UJ CD X -- CO X- >- X X Z X H- 3 X X O X ID a- >- CVJ in UJ u m vO or CU CL H- x^ 3 CD M CU X x or % v • < X 3 u CO X X X 3 or uj so CD -x or z H Q. X U CO •—< or H~H CO XH _J < c x CD cvj j*r arO N < HH l- Ui 0 H—H u J— CD x or — 03 z> * • J- UJ x CL 1- CO J— u X X t- or CL * H- _l CD co CD UJ 3 CDCL > >- x 1 » x UJ Z CDor or X 3. CD C CD NJ U x «t in or cvj a: -t O U CD Ul 00 < X X X X X CL x x — O O - Z X u 1— h-i or or ^X ^X 0 or a < CL 0 O O N (D in u x < O 0 0 CU >- NT UJ ♦ + • H rH X ►—« X UJ < X CL J— CO CO CO x x 03 *x or UJ O h- CO IxJ O CO 0 H-H x- in < x H V _j sO • < HH to CVJ J CD CO CD < co — u j or ^ Q_ — < w co nr X m 0: X cr ■ ►- 7 H Z U X -- ^X a j- x CD H- D M 2 jC H l U J O co < 1 I 1 HH CD *-* O 0 0 CU v UJ or — i— Z X X X X CL O CO CO (0 x x X cu *-»a * a. in •rH Ul X 00 CO u in CL CO -- x- in or j- co ^ z x X X CO cu iH H- CO _l u j — x » 0. • 4> • % X «-< < X XX CO TH UJ H- CL u X X X x a x CVJ fr-4 H O N CL O 1—1 X CD X Z X X l!J0. a CJ x Z Ul • X w UJ H M hH N < h- i n in u »-« -i3 X < N u X < CD • cu < oj rD M H CL H— X — 1- S I CD z 1- O Z h h X H- 1— 1— ir> »— 3 Z Z z z z < — 9 O II U or < < O < c

TABLE E CO H M 8 O O u. u. i- u. u. CL < < < < (U (U < < O O «-tQ L L O CL CL cl CL CL CL CL CL CL + _| | _ u n

cu UJ UJ cu cl I I »

(U - 1 O O O J - m J CDUJ *-» - h O 2 N I O h M n a % w o UJ O VC M II • •

U L - .u U. ID ID cc ->- >- X H • • • L JUl UJ CL - h 3 C5 cu >■ x -1 CL 1- H- X X H l • CU M CO * c .CL u. . 0 - h V ±J l± 1- iL 3 - h u 2 < CD II • • •» : H CU«H CU z n / C L c i / c i / t r JUICL

ro Q w H o X j u r o L u a CL _l < »- Ul CU _J ' O \ UJ < < l c Ui . 0 CL + + CL * + * * * * * i y ui u y iii i u y O UX CD X CO CU CD D 1X >1 CD CD X l J O l o CO Ul UJ — Ul - > O LCO O CO CL L LU 3 Ul CL CL ~ I H II CD 1 V . a o H \ o X CL u j- CL CO CU CL -—. u»- » cu < ' O •**% < Ul (VJ uX w* cu i _ UJ UJ V CL O' CL + + + * * * * q c d - uto cu >- X L O Z CL H 1 o *** UJ CO * + CL * * + •1 CU O - O £iL 2£ O ID t i JUJ UJ z UJ X >- < M a • • • • • • • < CL a L »—« »- j— H '— V >~ a UJ □ '— Z X >- X < 334 l • • ■ • • Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

TABLE E u 1- n UJ G G y y 0. •3- in oUJ X 1- G ro ro ro x x X o ro ro J- X X B * * • • • h- JO + UJ OJ < _ a. hh u JUJ OJ o * in -1— l- * * u • • _ • < t o H yC U G to hh UJ w UJ X ID o V B • • • • • • JOJ OJ hhh hh u Ul U) c X X y y B- • • OJ O 4—1 5U JUJ UJ UJ "5 - • • • «-H ID O 0) II IIH HI hh u u “5 hh >- ye V rH :aa a: a a a G ID ..—.»—. X X X X X _J ID >- C >- X ll • • • JOJ OJ X OJ “J ID JX _J >- u X CD n * ~— u HH HH to >- J< \ < UJ < 1- UJ UJ < u u V -) _J X a h- 1- h H I ID < n «% T* T-i '— X X X X X X UJ UJ X uj HH *—*. X X * * OJ oj co in co

hh uj h- * m xX x m m DO a X CD O HH CL OJ X H HH HH vH CD HH a UJ UJ UJ Ul UJ UJ 1- Ul _J * -♦ >- * II n r UJ DO cu x CD OJ XXX X X X OXco o CO X U rj UJ X X X < 1 _J H- OJ H c OJ o Ul r-o "—■ - < UJ X X I- \ Ul CL CL OJ o X - + + X + * * n * * * Ul CO Ul o O X OJ > X a z n / i + u n • cu Ul z X CO X CJ * X * * * II OJ o HH X U CJ U O o o ID 1 CO — a • iH t HH CO HH CO ID II * H X CO OZH- Z CO z o '— 1— J— _J HH »—• (D 1— D3 ID + II ro Ul o ID UJ 335 X Z 336

in > t- < Q li.

in u 1- ro < o z z o TO z z u or ro SL TO * X o o. Ul .—'ro _ J X OH rl o CD * DC s. CU TO Z o 3 CD or .— u «-• s l CD < 1- % TO h- 3 ro _J TO Ul < o ro O CD_ J o: O O TO z ID •» X ro TO — X z Z % o •*- z CD DC X TO C X s o or u z 3 CD W TO 1- i d a O DCX 3 m m OJ X N W U CO o o 3 OH U ro ro OJ o OH CD ro o o 1—1 3 X o TO TO or UJ CD a ..— or or TO _1 < O or U H- CD * a: TO 1—o z z J- CL —' 3 CD CL X X X (D < CD Isl U. < X o o o CL % i— oh ro ro ro U OH 3 «. O O O OJ X X (D * - TO TO TO N CD »-• O or or or in < o n n •» TO CD s . x < M U z < l - 3 Z z z z z z z W O r t O J Z CD CD CD c d

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA

David Erskine Allan was born on August 6, 1939, in Tulsa,

Oklahoma. He is the oldest of the two children of Jane and the late

Edgar Allan. In 1957, he was graduated from Birdville High School

in Fort Worth, Texas. After a two-year period at Texas A & M

University, he transferred to the University of Tulsa and in 1962

received the Bachelor of Science Degree in chemical engineering from

that institution. In 1964, he subsequently received the Master of

Science Degree in chemical engineering from the University of Tulsa.

Upon completion of the requirements for the Master of Science

Degree, he was employed by the Esso Research and Engineering Company

in Baytown, Texas. In 1967, he was transferred to the Esso Research

Laboratories in Baton Rouge, Louisiana. He was granted educational

leave from the Esso Research Laboratories in 1968 to complete the

requirements for the Ph.D. degree in chemical engineering at Louisiana

State University. After receiving his doctorate, he returned to the

Esso Research Laboratories in Baton Rouge.

He is married to the former Mary Margaret Dunaway and they have

a daughter, Laura Ann.

3 3 7

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EXAMINATION AND THESIS REPOET

Candidate: David Erskine Allan

Major Field: Chemical Engineering

Title of Thesis: The Dehydrogenation and Isomerization of Cyclohenxane Over a Platinum Alumina Mordenite Catalyst Approved:

/fyiutXo UoirtJuvfa Major Professor and Chairman

/ h . - ^ " Dean of the Graduate School

EXAMINING COMMITTEE:

~\) /*udL

&j o a .

-mM

Date of Examination:

January 26, 1970

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.