Citation
Low-temperature homogeneous oxidation of alkanes using hydrogen peroxide

Material Information

Title:
Low-temperature homogeneous oxidation of alkanes using hydrogen peroxide
Creator:
Gonzalez, Michael A., 1969-
Publication Date:
Language:
English
Physical Description:
xv, 157 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Acetates ( jstor )
Alcohols ( jstor )
Alkanes ( jstor )
Catalysts ( jstor )
Hydrogen ( jstor )
Methane ( jstor )
Oxidation ( jstor )
Oxygen ( jstor )
Peroxides ( jstor )
Propane ( jstor )
Alkanes ( lcsh )
Chemistry thesis, Ph. D ( lcsh )
Coordination compounds ( lcsh )
Cryochemistry ( lcsh )
Dissertations, Academic -- Chemistry -- UF ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1998.
Bibliography:
Includes bibliographical references (leaves 151-156).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Michael A. Gonzalez.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
029250268 ( ALEPH )
41372613 ( OCLC )

Downloads

This item has the following downloads:


Full Text










LOW-TEMPERATURE HOMOGENEOUS OXIDATION
OF ALKANES USING HYDROGEN PEROXIDE












By

MICHAEL A. GONZALEZ


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA































To my dear wife, Stacy Michelle, for all your support and belief in me.
You complete me.



















ACKNOWLEDGMENTS


No one has ever said life way easy, but it has been said that it can be extremely

rewarding. My decision to attend college and graduate school has made my life very

fulfilling, and I have only lived less than one-third of it. Although college and graduate

school has been filled with late night study sessions, overnighters at the lab and numerous

classes (as well as migraines), I would do it all again if given the opportunity.

My undergraduate freshman year at the University of Texas El Paso (UTEP) was

a time of indecisiveness and uncertainty about my future career aspirations. In my

sophomore year all these worries had changed for the better, this was the result of

receiving a scholarship and being given the opportunity to conduct undergraduate

research in the area of inorganic structural and polymeric chemistry. For this I would like

to thank Dr. Phillip Goodell, Dr. L.W. ter Haar and the Research Careers for Minority

Students (RCMS) committee for the opportunity to allow me to demonstrate my

academic and research wares. My adventures at UTEP would not have been possible

without the aid of Tony "Hubcap" Sanchez, Manuel "Manny" Garbalena, Anthony

"Skate" Guarero, Joseph "Little Joe" Janisheck and Gerardo "Ardo" Fuentes. To my

friends I say thanks for the memories and the friendship; you will never be forgotten.








From Texas my quest for higher learning landed me in sunny Florida, where

graduate life began at the University of Florida. I chose to work for Dr. Russell "Doc"

Drago, the second best decision I have made in my life. Doc is the reason I remain in

chemistry; he is the most intelligent person I know, and his fervor for chemistry is

undaunting. I can only hope to achieve this level in my chemistry career. "Doc, thank

you for all your patience and instilling a portion of your scientific knowledge in me." I

would also like to offer thanks to Ruth Drago, she has made my four-year stay

memorable. I would like to thank her and Doc for taking Stacy, Mikey and me into their

"scientific" as well as personal family and making us feel like family. This, in turn, made

our transition to Florida effortless.

My graduate school life was also made memorable by the friendships I formed

with members of the Drago group, both past and present. To the past members, Mike

"Bevis" Robbins, Todd "I can fix it" Lafrenz, Phil Kaufmnan, Garth "Ogre" Dahlen, Chris

Chronister, Don Ferris "Bueller" and Mike Naughton, I offer my thanks for showing me

how to survive in graduate school on both an academic and personal level. The current

Drago group members, especially the oxidation sub-group, are the greatest. Ken "Kato"

Lo, has been a great friend and always listened to me, plus I will never forget our Friday

lunches. Alfredo "Alf" Mateus was a great help with understanding heterogeneous

oxidation chemistry. Without Ben "Hubcap" Gordon I would not have completed the last

three months of research. Cheng "Tiger" Xu helped me understand the iron-oxo system.

I also want to thank Steve "Jorg" Joerg for bailing me out of trouble, answering my

philosophical and ethical questions, and running my NMR's. I would also like to thank

Krystoff Jurzyk and Nick "Corn on the" Kob for insight on high temperature








heterogeneous chemistry. I would also like to say thanks to Maribel and Diane for their

help in office matters.

The biggest thanks go to Stacy, my best decision, for standing by me, believing in

me and giving me the best things in life a person could ask for: love and a child. I thank

her for putting up with me on late nights of writing research summaries, being at the lab

and tutoring. This dissertation is as much hers as it is mine.

I thank my parents for their love and support. Dad always told me to set high

goals and never settle for second best. And mother was always there when I needed an

ear, and for this I say thanks.

No one knows what the future holds, but with my wife, family and friends, I know

I can approach it and control my own destiny.




























v














TABLE OF CONTENTS


a.gg


ACKNOWLEDGMENTS ............................................................ ......................... iii

LIST OF TABLES........................................ ...................................................... ix

LIST OF FIGURES ................................. ..............xii

A B ST R A C T .............................................................................. ....................... .......... xiv

CHAPTERS

1 GENERAL INTRODUCTION TO ACTIVATION OF ALKANES ..........................
Importance of Homogeneous Catalysis .............................................. ........................1
General Classification of Homogeneous Oxidation Reactions...................................3
General Classification of Alkane Activation Reactions ..............................................8
Examples of Alkane Activation by Homogeneous Catalysts ......................................9
Designing a Homogeneous Catalyst......................................................................... 12
Five Classes of Metal-Oxo Reactions..................................................................13
Homolytic or Heterolytic Pathway?.......................... ...........................19
Effect of Ligand on Metal Center .......................................................................21
H202 as an Oxidant ....................................................... ............................23

2 HOMOGENEOUS CATALYZED PARTIAL OXIDATION OF METHANE
WITH HYDROGEN PEROXIDE AND OXYGEN ..................................................27
Introduction.......................................................................... ..................................27
Experim ental................................................................................. ...................................30
Materials and Methods............................ ...... ..........................30
Physical Measurements...................... ...............................................33
Synthesis of Com pounds .......................................................................................33
Oxidation Procedure .................................. ... ..........................34
Results and D discussion ....................................................... .. .........................37
A union M odification.............................................................. ...............................37
Ruthenium Catalyzed Oxidation of Methane with Hydrogen Peroxide ................40
Peracid Formation and Reactivity....................................................................44
Oxidation of Methane with Molecular Oxygen.....................................................48
C onclusions............................. ...... ................................................................ 51









3 OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING A
RUTHENIUM METAL-OXO CATALYST............................................................53
Introduction......................................................................... ...................................53
Experim ental............................................................................... .................................55
Materials and Methods............................ ...................... ..........................55
Physical Measurements....................... .............................................55
Synthesis of Com pounds ..................................................... ............................... 55
Oxidation Procedure ................... .. ..... ............................56
Results and Discussion .......................... ............................ .............................58
Alkane Oxidation.......................... ................................... ..........................58
Mechanism of Oxidation.......................... ....... ..........................62
A addition of CuC ........................................... ............... ................... ...............64
Effect of Temperature ....................... .....................................72
C onclusions.......................................................................... ..................................77

4 SYNTHESIS AND CHARACTERIZATION OF IRON DIMETHYL
PHENANTHROLINE COMPLEXES.......................... ...........................78
Introduction.......................................................................... ..................................78
Experim ental........................................................................... ...............................81
Materials and Methods............................ ....... ...........................81
Physical Measurements....................................................... ...........................81
Synthesis of Compounds ............................................................82
Results and Discussion ......................................................... ...........................83
Characterization .............................................................. ...........................83
Single Crystal X-ray Diffraction............................ .................................. 83
FAB Analysis.................... ........................................................94
IR Analysis..................................... ...................................................... 94
N M R Analysis ........................................ ............................................97
High Valent Iron-Oxo Formation Studies..............................................103
C onclusions........................................................................ ..................................106

5 OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING AN
IRON METAL-OXO CATALYST.......................................... ...........................109
Introduction ......................................................................... ................................. 109
Experim ental............................................... ...................................................110
M materials and M ethods..................................... ......... ............. 10
Physical Measurements.................................................... 11
Synthesis of Com pounds ............................................... .......................... 111
Oxidation Procedure ........................................................... ..........................1. 12
Oxidation of Methane ........................ .....................................................112
Oxidation of Higher Alkanes..................................................................... 113
Results and Discussion ........................................................................................ 114
Oxidation of Methane with H202 ............................... ............................... 114
Mechanism for Oxidation of Methane............................................................ 118
Oxidation of Methane with 02 ........................................................................ .... 118


vii









Alkane Oxidations with cis-[Fe(dmp)2(H20),](CF3SO3), ....................................122
Alkane Oxidation with [Fe(dmp)C12]............................. ................................126
Mechanism for Higher Alkane Oxidation...................................................... 130
Addition of CuC ........................................................... ...............................131
Effect of Temperature ................................................................................... 141
Conclusions......................................................................... .................................146

6 CONCLUSIONS...........................................................................................148

GLOSSARY .......................................................................... ....................................150

REFERENCES ......................................................................... .................................151

BIOGRAPHICAL SKETCH ........................ ............................157





































viii













LIST OF TABLES


Table pag

1-1: Advantages of a Homogeneous Catalysts .......................... .......................... 14

1-2: Disadvantages of a Homogeneous Catalyst........................ .......................... 15

2-1: Oxidation Results for Methane @ 750C with cis-[Ru(dmp)2(H20)2](CF3S03)2 Using
H 20 2................................... ..................................................................................43

2-2: Oxidation Results for Methane @ 750C with cis-[Ru(dmp),(H20)23(CF3SO,)2 Using
H 20 and 0 ............................................ ......................................................49

3-1: Oxidation Results for Ethane, Propane and Butane @ 750C with cis-
[Ru(dmp),(H20)](CF3S03)2 using H202........................... .......................... 59

3-2: Oxidation Results for Iso-Butane and Pentane @ 750C with cis-
[Ru(dmp)2(HO0)](CF3S03)2 using H20....................................................... 60

3-3: Oxidation Results for Ethane, Propane and Butane @ 75C with cis-
[Ru(dmp)2(H20)](CF3SO3)2 and CuCI2 using H202 ...................................... ...67

3-4: Oxidation Results for Iso-Butane and Pentane @ 75C with
[Ru(dmp),(H20)](CFSO03)2 and CuCI2 using H202 ...................................... ...68

3-5: Oxidation Results for Propane @ 75C with cis-[Ru(dmp),(H20)](CF3S03)2 and
Varying Mole Equivalents of CuC12 using H202 .....................................................70

3-6: Oxidation Results for Propane @ Varying Temperatures with cis-
[Ru(dmp)2(H20)](CF3SO3)2 using H202.......................... ............................73

3-7: Oxidation Results for Propane @ Varying Temperatures with cis-
[Ru(dmp)2(H20)](CF3S03)2 and CuC12 using H202 ............................... .............75

4-1: Crystal Data and Structure Refinement for Fe(dmp)Cl2...................................85

4-1: (C ont'd). ................................................................................. .............................86

4-2: Atomic Coordinates (x 104) and Equivalent Isotropic Displacement Parameters (A2 x
10 ) for Fe(dm p)C 2........................................ ........................... ................... 87








4-3: Bond Lengths [A] for Fe(dmp)C12.................................................... 89

4-4: Bond Angles [o] for Fe(dmp)C2. .................................. ........................90

4-4: (Cont'd) ........................................... .................... 91

4-5: Anisotropic Displacement Parameters (A2 x 103) for Fe(dmp)Cl. ...........................92

4-6: Hydrogen Coordinates (x 104) and Isotropic Displacement Parameters
(A2 x 10 ) for Fe(dmp)C .................................................. ......................... 93

5-1: Oxidation Results for Methane @ 750C with cis-[Fe(dmp)2(H20)2](CF3SO,), Using
H 20 2 ................ ........................... .. .............................115

5-2: Oxidation Results for Methane @ 750C with cis-[Fe(dmp)2(H20)2](CF3SO3)2 Using
H 20 2 and 0 ............................................ ............................ ......... ...............119

5-3: Oxidation Results for Ethane, Propane and Butane @ 750C with cis-
[Fe(dmp),(H20)](CF3SO3), using H2................. ............................ 123

5-4: Oxidation Results for Iso-Butane and Pentane @ 750C with cis-
[Fe(dmp)(H2O)](CF3SO3)2 using H202.................................... ..................124

5-5: Oxidation Results for Ethane, Propane and Butane @ 750C with Fe(dmp)Cl2 using
H 20 2 ....... ........................................ .................................127

5-6: Oxidation Results for Iso-Butane and Pentane @ 750C with Fe(dmp)C12 and H202128

5-7: Oxidation Results for Ethane, Propane and Butane @ 750C with cis-
[Fe(dmp)2(H20)](CF3SO3)2 and CuCI2 using H202..................................................132

5-8: Oxidation Results for Iso-Butane and Pentane @ 750C with cis-
[Fe(dmp)2(H20)](CF3SO3)2 and CuCI2 using H ,0...............................................133

5-9: Oxidation Results for Ethane, Propane and Butane @ 75C with Fe(dmp)C1, and
CuCl2 using H 20 2 .......................................... .............................................135

5-10: Oxidation Results for Iso-Butane and Pentane @ 750C with Fe(dmp)C1, and CuC12
using H 20 ....... ........ ...... .... ....................................... 136

5-11: Oxidation Results for Propane @ 750C with cis-[Fe(dmp),(H20)](CF3SO3), and
Varying Mole Equivalents of CuCI2 using H202 ............. ..................... 137

5-12: Oxidation Results for Propane @ 750C with Fe(dmp)Cl2 and Varying Mole
Equivalents of CuCl2 using H202............................................................................139








5-13: Oxidation Results for Propane @ Varying Temperatures with cis-
[Fe(dmp),(H20)](CF3SO3)2 using H ........................... ........................142

5-14: Oxidation Results for Propane @ Varying Temperatures with cis-
[Fe(dmp)2(H20)](CF3SO3)2 and CuCI2 using H202 ...............................................143

5-15: Oxidation Results for Propane @ Varying Temperatures with Fe(dmp)CI2 using
H202........................... .. .... ..... ................................. 144

5-16: Oxidation Results for Propane @ Varying Temperatures with Fe(dmp)C1, and
CuC12 using H 20 2 .................................................... ............................ 145













LIST OF FIGURES


Fgigue page

1-1: Catalytic Cycle for the Hydroformylation Reaction....................................................11

1-2: Desired Reaction Pathway for Substrate Oxidation .................................................. 20

1-3: Detailed Reaction Pathways of Metal-Oxygen Species...........................................22

1-4: Methods for Activating Hydrogen Peroxide ........................ .......................26

2-1: Structure for 2,9-dimethyl-l,10-phenanthroline........................................................27

2-2: Proposed Catalytic Cycle for Alkane Oxidation via cis-[Ru(dmp)2(O)2]2 ...............29

2-3: Detailed Catalytic Cycle for Ruthenium Analogue....................................................31

2-4: Distribution of Worldwide Natural Gas Resources....................................................32

2-5: Diagram of Batch Type Hydrogenation Reactor................................................35

2-6: 'H NMR Spectrum for cis-[Ru(dmp)2(H20)2](PF6)2.............. ...........................38

2-7: 'H NMR Spectrum for cis-[Ru(dmp)2(H20)2](CF3SO3)2 ..........................................39

3-1: Proposed Hydrogen Atom Abstraction Mechanism for cis-[Ru(dmp)2(H20)2]2
C om plex ................................................................ ..............................................63

3-2: Proposed Oxygen Atom Insertion Mechanism for cis-[Ru(dmp)2(H20)212+
C om plex ......................................................... .....................................................65

3-3: Oxidation of Propane with cis-[Ru(dmp)2(H20)](CF3S03)2 and Varying Mole
Equivalents of CuC1, using H202........................................ .............................. 71

3-4: Oxidation of Propane @ Varying Temperatures with cis-[Ru(dmp)2(H20)](CF3SO3)2
using H 20 .......................................................... ..........................................74

3-5: Oxidation of Propane @ Varying Temperatures with cis-[Ru(dmp)2(H20)](CF3SO3)2
and CuC using H202 ........................................ ..........................................76

4-1: Crystal Structure for Fe(dmp)C2.......................... ..... ......................84








4-2: FAB' Spectral Results for cis-[Fe(dmp)2(H20)2](CFSO3)2.......................................95

4-3: Infrared Spectrum for cis-[Fe(dmp)2(H20),](CF3SO3). ............................................96

4-4: 'H NMR Spectrum for Fe(dmp)C .................................... ......................98

4-5: 'H NMR Spectrum for cis-[Fe(dmp)2(H20)2](CF3S3)2...........................................100

4-6: Crystal Structure for cis-Fe(dmp)2(NCS)...........................................................101

4-7: UV-VIS Spectra for the Addition of H202 to cis-[Fe(dmp)2(H20)2](CF3SO3). ........105

4-8: UV-VIS Spectra for the Addition of H202 to Fe(dmp)Cl2......................................107

5-1: Oxidation of Propane with cis-[Fe(dmp)2(H20)](CF3SO3)2 and Varying Mole
Equivalents of CuC12 using H202 ................................ .............138

5-2: Oxidation of Propane with Fe(dmp)C12 and Varying Mole Equivalents of CuCl2
using H 20 2......................... ............... ............. ........................140














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


LOW-TEMPERATURE HOMOGENEOUS OXIDATION
OF ALKANES USING HYDROGEN PEROXIDE

By

Michael A. Gonzalez

May 1998


Chairman: Dr. Russell S. Drago
Major Department: Chemistry

The activation of saturated hydrocarbons by metal complexes in the liquid phase

became an active research area towards the end of the 1960. Stringent conditions such as

the use of powerful oxidants, corrosive superacids, elevated pressures and high

temperatures are used for the selective oxidation of low molecular weight paraffins. As

products from these oxidations gain increased industrial importance, environmentally

friendly and cost-efficient catalysts and processes are sought.

Natural gas reserves, which are approximately 90% methane, are abundant and a

source of a valuable feedstock. The direct oxidation of methane to methanol provides the

first step towards a route to an alternative large-scale transportation fuel replacing

dwindling petroleum reserves. For this reason, the development of a direct one-step

process for the oxidation of methane to methanol becomes of interest. Also of








importance is the ability to synthesize valuable organic compounds selectively (alcohols,

aldehydes, ketones and carboxylic acids) from higher alkane feed stocks.

Previous reports from this laboratory have demonstrated the use of the cis-

[Ru(dmp)2(H20),](PF6), precursor, where dmp is 2,9-dimethyl-l,10-phenanthroline, for

the activation of methane in acetonitrile. The nitrogen-based dmp ligand not only

increases the electrophillicity of the metal center, but also imparts a steric hindrance

about the metal, forcing a cis geometry, allowing the catalyst to be a more potent oxidant.

Using the properties from this complex, a number of derivatives have been successfully

synthesized and characterized.

The synthesis and characterization of each newly synthesized catalyst: cis-

[Ru(dmp)2(H20)2](CF3SO3)2, cis-[Fe(dmp)2(H20)2](CF3SO3)2 and [Fe(dmp)C12], as well as

the ability of each to hydroxylate C2-C5 alkanes in acetonitrile and methane in a glacial

acetic acid/acetic anhydride solvent mixture at 750C and 40 psi are presented. Also

investigated is the ability to modify and increase the selectivity to the alcohol product

upon addition of a metal chloride, and the effect of temperature on each catalyst

precursor. Preliminary mechanistic data suggesting a pathway for methane and higher

alkane oxidation are also presented.


XV













CHAPTER 1
GENERAL INTRODUCTION TO ACTIVATION OF ALKANES

Importance of Homogeneous Catalysis


The end of the 1960s confronted the field of homogeneous catalysis with a

potential problem. This obstacle was the inability for a soluble transition-metal complex

to effectively activate a saturated hydrocarbon in the liquid phase.' As large resources of

natural gas were discovered, the need for improvements to the current homogeneous

catalytic systems became apparent, as well as the demand for alternative catalysts.

One solution to this issue was the development of new classes of metal

complexes, which were capable of undergoing an oxidative addition across the C-H bond

of a paraffin (Equation 1-1). The application of this catalyst would allow for alkanes to

be utilized as feedstocks in a number of industrial chemical reactions. This oxidative

addition process produces the desired oxidized products in selectivity ratios, which are

more favorable than those achieved using typical free radical reagents.2,3,4



(Equation 1-1) M + R-H R-M-H



As advances in coordination chemistry and catalysis were developed, reports of

activating hydrogen, olefins, aromatics, carbon monoxide, and molecular nitrogen with

soluble transition-metal complexes became commonplace.' The term activation is used








to imply "the molecule or its part becoming a ligand in the coordination sphere of the

complex and, as a result, undergoing a subsequent chemical transformation"(p. 2).1

The primary deterrent one is confronted with in the activation of alkanes is the

chemical inertness these molecules possess. Alkanes are benign, to the point they can be

safely used as a non-interacting solvent. Therefore, any such activation of an alkane

would require notably stringent conditions. These conditions typically require active

particles, which include strong oxidants, superacids, free atoms, radicals, carbenes, high

temperatures (500-1000C) and other sources of energy (i.e. radiation chemistry). As

environmental and monetary constraints become more significant, a more desirable

oxidation technology needed to be sought.2-3,4

Interest in the activation of methane, a primary component in natural gas, also

heightened prompting the development of a new catalyst that was able to impart

selectivity in its oxidized product under relatively mild conditions. The typically inert

methane molecule, along with its alkane counterparts, required designing a catalyst that

was able to effectively activate these substrates which did not contain double or triple

bonds or lone electron pairs and possessing relatively strong covalent C-H and C-C

a-bonds.

Shilov' had presented evidence for the activation of alkanes; he expressed that if

other more selective reactions of alkanes under "comparatively mild conditions" exist,

there is an inherent ability to oxidize the alkanes in question. The statements that follow

provide a basis for his proclamation:








Numerous examples of homogeneous H2 activation are reported.2 One example

is the H-D exchange for hydrogenation. Yet, the cr bond in molecular

hydrogen is not weaker than the a C-H bond in alkanes.

Many metal complexes (ML, or M) are capable of reacting with substrates

containing "activated" C-H bonds.2 One example is the C-H bond in aromatics

or in the a-position to double bonds. As the C-H bond is broken, formation of

a M-C bond results. However, the C-H bond energy which is greater in

aromatics than alkanes does not hinder the reaction.

Also demonstrated is the involvement of a "nonactivated" aliphatic C-H bond

within the coordination sphere. This occurs when a suitable position on the

ligand of the complex becomes available.2 This results in a process defined as

"cyclometallation" and provides evidence for the possibility of a reaction

occurring with alkanes, possibly at elevated temperatures.

Hydroxylation of C-H bonds in saturated hydrocarbons catalyzed by metal

enzymes is known.2 One published example is the oxidation of methane to

methanol (primarily) by methane monooxygenase.

General Classification of Homogeneous Oxidation Reactions


This research is directed towards developing a series of homogeneous catalysts

able to activate saturated hydrocarbons, olefins and aromatics with mild oxidants. In

order to do so, an understanding of the fundamental chemical reactions must be

understood. These liquid phase (homogeneous) transition metal catalyzed oxidations can

be categorized into three areas.5 These categories include the following:








1. Free Radical Autoxidation

The Mid-Century/Amoco process for the conversion ofp-xylene to terepthalic

acid is one example of this category. This reaction is exhibited in Equation 1-2.

CH3 CO2H

Co(OAc)2
(Equation 1-2) + 02 -a 0
NaBr
200 OC
H3 02H

In this process air (20psi) is utilized as the oxidant, with a bromide-promoted

cobalt salt as the catalyst in an acetic acid solvent. The primary oxidation step, i.e.

p-xylene to p-toluic acid, occurs quite readily in the presence of a small amount of the

cobalt salt. The secondary oxidation step, formation of the di-acid, necessitates the

presence of higher concentrations of the catalyst and/or addition of the bromide

promoter.2

Oxidations of alkyl aromatics involving a cobalt (III) catalyst appear to proceed

almost exclusively via an electron transfer mechanism.' The reaction of the alkyl

aromatic with Co(III) results in formation of [ArCH3]*. and a Co(III) radical. Loss of a

proton results in producing ArCH2, and HW. In the presence of air or oxygen, a peroxy

species is formed producing ArCH2O2', which then reacts to produce the normal

oxidation products. The Co(II)* catalyst can be regenerated either by reacting with 0,,

combined with ArCH2O,*, or by reacting with the hydroperoxy species, ArCH202H, to

produce ArCHO, Co(III) and OH-.2








The corresponding aromatic acids are produced by subsequent aldehyde oxidation

via a peroxy acid intermediate. The addition of the bromide allows for formation of

bromine atoms (Br-) via electron transfer oxidation of bromide by Co(III). The bromine

atom has been demonstrated to be extremely efficient at hydrogen atom abstraction,

which can allow for rapid formation of the benzyl radical, thus initiating the autoxidation

sequence.'

2. Nucleophillic Attack on Coordinated Substrates

An example of this classification is the Wacker Process for oxidizing ethylene to

vinyl acetate. This reaction is provided in Equation 1-3.



(Equation 1-3) H,C=CH2 + 2 02 + HOAc H2C=CHOAc + H,O



PdCI2 and CuCI2 are two catalysts necessary for the above reaction to proceed in

an acetic acid (CH3COOH) solvent. The initial stage of the oxidation involves the t-

coordination of the ethene substrate with concurrent or subsequent loss of chloride to

form a sixteen-valence electron palladium (II) anion, which in turn undergoes a ligand

replacement reaction to yield the neutral palladium (II) aquo species. Coordination of

ethene to the palladium (II) metal center results in a decrease in the double bond electron

density, allowing it to become more susceptible to nucleophillic attack by either OH or

H2O,.2

The palladium metal is re-oxidized to palladium chloride by the co-catalyst, either

CuCl2 or FeCl3, as described above. In this case, the co-catalyst, CuCI2 is the reduction








product of CuCl that is readily oxidized with either air or oxygen. Therefore, this catalyst

combination gives rise to a catalytic system which allows for air or oxygen oxidation of

ethene to ethanal.2

3. Metal-Catalyzed Oxygen Atom Transfer Reactions from Coordinated
Hydroperoxides or Metal-Oxo Species to Organic Substrates

The reaction ofpropylene with alkylhydroperoxides to yield propylene oxide falls

within this category. This reaction is illustrated in Equation 1-4.



(Equation 1-4) Mo Cat O
H3CHC=CH2 + ROOH H3CHC H2 + ROH



This process involves the reaction of a homogeneous metal catalyst, molybdenum,

with an organic hydroperoxide and an alkene. As a result this reaction produces an

epoxide product in a relatively high yield. This oxidation process proceeds with the

hydroperoxide becoming "activated by coordination" to the metal center. As a result, the

peroxide's oxygen experiences a decrease in electron density, rendering it susceptible to

nucleophillic attack by the alkene. Additional metal catalysts for this reaction are those

belonging to the second row, positioned early in the series and possessing high attainable

oxidation states. Metals with these characteristics include Mo(VI), W(VI) and Ti(VI).2

Transfer of the oxygen from the metal hydroperoxide complex to the alkene is

suggested to occur via a cyclic transition state, as detailed in Equation 1-5.









(Equation 1-5) ---- + MRO,H

R



The success of the Mo(VI), W(VI) and Ti(VI) catalysts for this type of oxidation

reaction is attributed to their being weak oxidants with poor one electron redox potentials:

-0.21, -0.03 and -0.37eV respectively. Therefore, they are poor catalysts for homolytic

hydroperoxide decomposition, as exhibited in Equation 1-6, which is detrimental to the

desired epoxidation reaction.'



(Equation 1-6) [M"ROH] M'"" + RO,. + H'.



It is important to emphasize that only the autoxidation, category number I,

utilizes the direct reaction of molecular oxygen with the organic substrate.6

Homogeneous catalysis is now a relatively mature field with a number of diverse

reactions being investigated for informative studies as well as mechanism and theoretical

analysis. Along with the growth in this field, a relationship to other important areas has

been firmly established. These areas include heterogeneous catalysis, organometallic

chemistry and bio-catalysis.

Homogeneous catalysis is a resource, which is virtually untapped, and its

importance cannot be overestimated. In the age of industry becoming oriented towards

specialty chemicals, methods to catalyze functional group transformation, hydrocarbon








activation, polymerization and asymmetrical catalysis, the need and importance for

effective homogeneous catalytic systems will continue to grow.

The activation of C-H bonds in alkanes by transition metal complexes, not long

ago thought to be the most difficult challenges facing chemists, is now almost

commonplace. Many, examples, often under remarkably mild conditions, have

appeared.',3,7 Problems arise in the transformation of these fundamentals into the

practical alkane conversion process. These reasons include thermodynamics, a reaction

being uphill at ambient conditions, catalyst stability, and catalyst incompatibility with 02

to activate alkanes. Other problems include the issue of selectivity. Regioselectivity is

an obvious problem, one such case is if terminal alcohols are preferred over their isomeric

products. More important, the products (alcohols and aldehydes) tend to be considerably

more reactive than the starting alkane. This then places a limit on the achievable yield of

the desired hydroxylated product.

General Classification of Alkane Activation Reactions


Three categories of remarkably facile alkane activation can be described:

oxidative addition (Equation 1-7), c-bond metathesis (Equation 1-8) and electrophillic

substitution (Equation 1-9).1.3.7



(Equation 1-7) L,M + R-H s L,M(H)(R)

(Equation 1-8) LM-X + R-H LnM-R + H-X

(Equation 1-9) LM"~ + R-H LM-R"(-' + H'








The first in this classification of reactions, oxidative addition reactions, has been

found to occur in low valent electron rich, coordinatively unsaturated metal centers

towards the right end of the transition metal series (Groups VI-X).3

Sigma-bond metathesis reactions, the second classification, are observed for

complexes containing Group III transition metals (including lanthanides and actinides), as

well as metals belonging to Group IV. Also exhibited in this category are examples of

intramolecular reactions with metals belonging to Group V. These reactions lead to

stable organometallic products; however, very limited examples of alkane

functionalization have been achieved. Each of the active species involved is highly

sensitive to 02 and the oxygenated products that are produced.3

Finally, electrophillic substitution reactions have been observed with the

"traditional" or "classic" coordination complexes. An example of a such complex is

PtClx(H20)4,,1-'2.89 In the duration of the mechanistic process, a stable organometallic

species is not obtained, but net functionalization of the alkane is achieved. This category

of catalysis appears attractive in that the species involved will utilize dioxygen. This

property allows in principle for the closing of the catalytic cycle.

Examples of Alkane Activation by Homogeneous Catalysts


Many examples of homogeneous catalytic reactions are published in literature,

although this number is too large to provide one example for each area of activation.

Two important examples are exhibited to provide a general idea of the catalytic chemistry

being performed on the industrial and laboratory scale with homogeneous catalysts.








The hydroformylation or Oxo reaction discovered in 1938 by Roelen is utilized on

a large industrial scale.10.1,12 This process employs a homogeneous catalyst based on

cobaltl0'1 ,12 or rhodium.13 Most commonly the HCo(CO)4 pre-catalyst is employed, the

product generated by the in-situ hydrogenation of CO2(CO),.

In 1961, Heck and Breslow14 proposed the mechanism for the cobalt-based

Hydroformylation process, as illustrated in Figure 1-1. Five elementary processes

comprise the complete mechanism.

The first step involves the dissociation of the carbonyl ligand from HCo(CO)4 to

produce the catalytically active species tricarbonylhydridocobalt, [HCo(CO)3]. This is

followed by the combination of the active catalyst with the olefin to generate a n-olefin

complex. The migratory insertion of the olefin into the Co-H bond is the third step,

proceed by the alkyl undergoing a migratory insertion into the Co-CO bond. In the final

step an H2 induced aldehyde elimination occurs, producing the desired oxidized product

and regeneration of the active catalyst.

Another cobalt based catalytic system is the reductive carbonylation of methanol

(Equation 1-10). This example illustrates how cobalt-based catalysts have dominated this

area of industrial chemistry for the past fifty years.15-22 Alternative metals are capable of

catalyzing the identical reaction; however, these are generally inferior to those of the

cobalt-based systems.


(Equation 1-10) CH3OH + CO + H2 CH3CHO + H20














HCo(CO)4

1 CO
RCOH H ,CCH2




5 2
HCo(CO)3 -- --)



H-H--




R-CO-Co(CO)3 OCH,





4
3


RCo(CO)4


Figure 1-1: Catalytic Cycle for the Hydroformylation Reaction








Improvements to the cobalt-based catalyst have been achieved through the use of

co-catalysts and promoters. Such promoters employed include iodide, phosphines and

transition metals.18 Iodide, by far the most important of this group, is almost always

utilized as a promoter. Despite advances in catalyst performance, high pressure and

temperatures are required. Conditions for these catalysts are typically operated at

pressures of 4000-8000 psi and temperatures of 175-200*C. The increased pressures

required can pose obvious difficulties for reactor design and operation, as well as for

safety considerations. The elevated temperatures required can lead to the formation of

heavy by-products via the Aldol condensation reaction of acetaldehyde.19,20

It is shown both rhodium and cobalt are capable of catalyzing a variety of

carbonylation reactions.23,24 Rhodium catalysts have been demonstrated to be

significantly more active than their cobalt counterparts and allow reactions to proceed at

much lower extremes of temperature and pressure. However, iodide is necessary as a

promoter. In the presence of this promoter and CO, rhodium is an extremely proficient

catalyst for the carbonylation of methanol to acetic acid. This process is known as the

Monsanto Acetic Acid Process.25-28

Designing a Homogeneous Catalyst


The focal point of our research is directed towards designing and developing a

catalyst able to undergo chemistry involving a transition-metal catalyzed oxygen atom

transfer from a metal-oxo complexes or species. The activation of alkanes, primarily

methane, utilizing this newly developed metal-oxo complex is our primary goal. The

development of this catalyst can be hindered due to the following conditions:








1. The role of the transition metal complex for activation of molecular oxygen

can be complicated by numerous competing reaction pathways which are

present.

2. In order to selectively oxidize a variety of substrates, a number of different

conditions may be required.

3. The lack of any obvious pattern or role portrayed by the metal, as well as lack

of a detailed reaction mechanism.

A catalyst can either be homogeneous or heterogeneous in nature. A homo-

geneous catalyst was selected on the basis of its advantages in a reaction mixture and

underlying chemistry. The advantages and disadvantages of utilizing a homogeneous

catalyst are listed in Table 1-1 and Table 1-2 respectively.

Five Classes of Metal-Oxo Reactions

Oxidation pathways have been exhibited to occur in a number of general reaction

pathways. Drago29 has formulated five classes of metal-oxo reactions, which are based

on the role of the transition metal complex. In order to effectively categorize these

reactions, the mechanism for substrate oxidation was omitted. The five classes are listed

below and each reaction mechanism summarized. An example of each reaction class is

also provided.



Class I: Metal Bound 0,

(Equation 1-11) M + 02 M-O-O

(Equation 1-12) M-O-O + RH M-O-O-H + R-

























Table 1-1: Advantages of Homogeneous Catalysts
Advantage Description
Homogeneity Uniform active structure
Efficiency Theoretically all the atoms or molecules of the catalyst are
available for reactants
Reproducibility A result of well-controlled active sites
High Selectivity Often more specific than a heterogeneous catalyst even for
asymmetric induction
Mild Reaction High activity often achieved under mild conditions
Conditions
Controllability If chemical information is readily available, modification
allowing the control of electronic and steric properties of the
metal center can be attained.
Reaction Mechanism Investigation of the molecular reaction can be performed.

























Table 1-2: Disadvantages of a Homogeneous Catalyst
Advantage Description
Separation The separating of an expensive catalyst from the reaction
mixture poses a problem and can be costly, and may also require
special treatment, which usually destroys the catalyst.
Solubility The range of solvents suitable for a homogeneous catalyst is
often limited by the solubility of the catalyst; a compatible
solvent is not necessarily the most suitable for a high reaction
rate.








(Equation 1-13) M-O-O-H +R- M + ROOH

(Equation 1-14) ROOH ROH or R=O



Molecular oxygen is coordinated to the metal (Equation 1-11) to increase its

basicity and radical reactivity. The substrate then undergoes hydrogen atom abstraction

by the metal-peroxo species to generate a metal hydroperoxo species and an alkyl radical

(Equation 1-12). The metal hydroperoxo species and the alkyl radical then undergo

recombination (Equation 1-13) to generate an alkylhydroperoxide species and the reduced

metal species. The final step (Equation 1-14) involves decomposition of the

alkylhydroperoxide to yield the corresponding alcohol and ketone. Cobalt (II)

coordination compounds that reversibly bind dioxygen are examples of this group.



Class II: Metal-Oxo via 0,

(Equation 1-15) M + 02 M-O-O

(Equation 1-16) M-O-O + M M-O-O-M

(Equation 1-17) M"N-O-O-M"* 2 M""(O)

(Equation 1-18) M=O + S SO + M



Once again the metal coordinates dioxygen (Equation 1-15); however, in this

series another metal enters the coordination sphere and forms a L-peroxo dimer (Equation

1-16). This p-peroxo dimer then undergoes cleavage to form two high-valent metal-oxo

complexes (Equation 1-17). The metal-oxo complex is then able to undergo oxygen atom








transfer or another similar type reaction (Equation 1-18), thus regenerating the lower

oxidation-state of the metal and producing the oxidized species of the substrate. Metal

oxidation states normally involved in this cycle are two to four, three to five or four to

six. One system to be discussed, cis-[Ru(dmp)2(H2O)2]2', is an example of this category.



Class III: Metal-Oxo via Peroxides

(Equation 1-19) M"' + H20, M7+2(0) + H,O

(Equation 1-20) M"+(O) + S SO + M"'



This class of reactions takes advantage of the more powerful oxidizing ability of

hydrogen peroxide to oxidize lower-valent metal complexes to metal-oxo complexes.

The low-valent metal complex is oxidized with hydrogen peroxide (Equation 1-19) to a

higher valency producing a metal-oxo complex and a benign by-product, water. The

metal-oxo species is then able to undergo oxygen atom transfer or a similar reaction

(Equation 1-20), thus regenerating the lower oxidation-state of the metal and producing

the oxidized species of the substrate. A candidate for this class is a monooxygenase

enzyme, which involve metal-oxo oxidants.



Class IV: Metal-Peroxo Systems

IVa- Metal Catalyzed Peroxide Decomposition

(Equation 1-21) M"' + H,02 M" + OH + OH-

(Equation 1-22) M" + H202 M" + I+ + HO2-








IVb: Metal-Peroxo Formation

(Equation 1-23) M=O + H202 MO2 + H20

(Equation 1-24) MO, + S I SO + M=O



Metal peroxo complexes are the reactive intermediates for this group. Several

reactions with substrates yield further divisions of this class. In class IVa, the metal

complex reacts with hydrogen peroxide to generate radical species, which are the result of

peroxide decomposition (Equations 1-23 and 1-24). An example of this class is Haber-

Weiss and Fenton Chemistry. The radial species, the active oxidant, are then able to

oxidize the substrate. Class IVb is the result of the bound metal peroxo or alkylperoxo

complex attacking the substrate. The metal peroxo is generated by the reaction of a

metal-oxo species with hydrogen peroxide. The alkylhydroperoxo species is generated in

the same fashion as illustrated in Equation 1-22. Asymmetric epoxidation of alkenes

with alkylperoxides are examples of this class of chemistry.



Class V: Metal Centered Oxidizing Agents

Metal center is active electron acceptor in the oxidation.

Often involves coordination of substrate to metal center, "activation".

The Wacker process, Equation 1-3, is an example of this class.

The reduced complex is re-oxidized by 02 or peroxide.








The approach taken in this research is to develop a catalyst that possesses

properties pertaining to either Class II or III. The developed catalyst could also possess a

mechanism of regeneration with dioxygen to generate a high-valent metal-oxo complex,

as in Class V, necessary for alkane oxidation. This catalyst must be able to perform the

key step, that being the generation of a strongly oxidizing high-valent metal center.30

The desired reaction pathway, illustrated in Figure 1-2, exhibits the need for a

metal center, which can be activated by hydrogen peroxide or molecular oxygen to form

the mono-oxo species. This activated complex may be a powerful enough oxidant to

activate the substrate or be further oxidized by hydrogen peroxide or dioxygen to generate

a di-oxo species, a complex with greater oxidizing strength than the mono-oxo species.

Upon delivery of its one or two oxygen atoms, the reduced metal center could be

re-oxidized by hydrogen peroxide or oxygen to regenerate the mono or di-oxo species,

thereby closing the catalytic cycle

Homolvtic or Heterolvtic Pathway?

There are two pathways, which exist for a metal center to be activated by

dioxygen or hydrogen peroxide. These pathways are known as homolytic and heterolytic

oxidation. The former31 normally involves the following transition metal couples:

VV/V", Crv/Crv, Mn'n/Mn", Fe"/Fe1, Co '/Con and Cu1/Cu'. Characteristics of this

reaction include the production of a free radical intermediate, oxidation occurring in the

outer-sphere via bimolecular steps, non-coordination of the substrate. Oxidation products

are generally not very selective or stereospecific, and the metal center undergoes a one-

electron oxidation and reduction step. Examples of relevant oxidations employing this















Substrate Oxidized Substrate



[M(lig)2(H20)2]x+
S H202 or
( 02
OH2

[M(lig)2(0)2]x+ [M(lig)2(H20)(O)]x+


OH2 H202
or
02


Figure 1-2: Desired Reaction Pathway for Substrate Oxidation








oxidation pathway include the non-stereoselective epoxidation of olefins occurring with

transition metal complexes of vanadium, manganese and iron and the hydroxylation of

alkanes and arenes occurring with the transition metal complexes of vanadium,

chromium, manganese, iron, cobalt and copper.

The latter, heterolytic oxidations,31 involves the following transition metals: Ti",

V", Cr", Mo"1, Mo", Mnv". Ru", Os"v'", Rhm/Rh, Irm/Ir', Pd"/Pdo, pt/Pt.

Characteristics of this reaction pathway include the inability to produce a free radical

intermediate, the reaction occurring in the coordination sphere of the metal, substrate

coordination resulting in activation, high selectivity and stereospecificity of the oxidized

products and no net change in the oxidation state of the metal. However if a change in

the oxidation state occurs, it is via a two electron step, i.e. M" to M4. Examples of

heterolytic oxidations include the stereoselective epoxidation of olefins with transition

metal complexes of titanium, vanadium and tungsten and the ketonization of olefins with

transition metal complexes of rhenium, iridium, palladium and platinum. Mimoun30 has

also categorized a reaction coordinate pathway involving metal-oxygen species, which is

illustrated in Figure 1-3.

Effect of Ligand on Metal Center

Numerous and thorough studies for polypyridyl ligands with ruthenium metal

centers by Drago31, Meyer32, Takeuchi33 and Che34 have been performed. These studies

have revealed certain features that make these complexes capable of catalyzing oxidations

of hydrocarbons with molecular oxygen. These complexes display a series of reversible

















hydroxo

Mn+2-OH


+ 02
0.
Q + Mn

0M^


superoxc









peroxo

peroxo


O Mn+1
M \ / 2 M+2=O
Mn~l 0


'N


o 1I-peroxo oxo

+ HO H20



S+ R M+2-0OR +0ROOH
alkylperoxo -HX Mn2
+ H++ H 0
Mn2-00H -
hydroperoxo


+ M"



M nloM n+

Pl-OXo


Figure 1-3: Detailed Reaction Pathways of Metal-Oxygen Species








and accessible oxidation states (i.e. from 2' to 6') to reduce dioxygen completely.

The poylpyridyl ligand provides the stability for the metal center to achieve

higher oxidation states and impart stability to the complex. It also forms stable

ruthenium oxo-species, {Ru(O)}2 and {Ru(0),}2', which are known to be excellent metal

oxygen atom-transfer reagents towards a variety of substrates. The activation of methane

and other saturated alkanes is the field where the majority of our interest lies.

HO, as an Oxidant

The need for an inexpensive, environmentally friendly and benign by-product oxidant is

of top concern. Many oxidants are available which include dioxygen, hydrogen peroxide,

alkyl hydroperoxides, peracids, sodium sulfate, oxone, nitric acid, dichromate, bromine,

manganate and Caro's acid (H2SO,). However, the last five are costly, and involve

expensive disposal problems. Nitric acid also produces noxious NO, compounds as by-

products.

Peroxygen reagents have been utilized in chemical synthesis for a number of

years.35 As the public, chemical producers, and governmental agencies direct their efforts

towards "greener" pathways, the interest in the use ofperoxygen oxidants has increased.

A reagent, hydrogen peroxide (H202), can offer environmental and economic benefits.

Hydrogen peroxide is available for chemical synthesis as an aqueous solution in

concentrations ranging from 35% to 90% by weight. Stabilizers are normally present in

the part per million levels to prevent decomposition. With no additional additives being

present, a minimization in unwanted side reactions is an additional advantage.








Hydrogen peroxide can also be stabilized for extended periods of times under a

variety of conditions. As an aqueous solution, hydrogen peroxide will lose less than one

percent of its active oxygen content per year, when stored in compatible containers at

ambient temperatures. This stability allows hydrogen peroxide to be a more efficient

oxidant than dichromate or permanganate on a weight for weight basis.

Reactions involving hydrogen peroxide are traditionally performed under mild

conditions of temperature and pressure, preventing an increase in peroxide

decomposition. By performing under these conditions the need for a large capital

investment is eliminated as well as a decrease in safety concerns. Reactions, which use

hydrogen peroxide as an oxidant, can also proceed in either an aqueous, organic or neat

solvent. This allows for reaction and reactor design to be engineered allowing one to

avoid the use of solvents, thus eliminating costly start-up expenses and production of

waste streams.36

The major benefit of using hydrogen peroxide is its environmental acceptability.

When the oxidizing power of the peroxide is spent, only water remains as the by-product,

eliminating the need for expensive effluent disposal and treatment. Hydrogen peroxide is

also relatively inexpensive as a reagent, when compared on the basis of oxidizing power

and by-product production. Although chlorine and oxygen oxidants are less expensive,

since these are gases, phase separation concerns are raised as well as the inability to

achieve selectivity towards oxidation products.

Hydrogen peroxide can be utilized in a variety of methods: direct activation,

catalytic activation and activation via peroxides. All require the use of hydrogen

peroxide to be activated to an even more potent oxidant. Figure 1-4 summarizes methods








for activating hydrogen peroxide. Our use of hydrogen peroxide falls under the area of

catalytic activation. Hydrogen peroxide is activated with a metal complex by forming a

metal-oxo or metal-peroxo species.36

The Einchem Corporation has demonstrated the feasibility of using hydrogen

peroxide as an oxidant on a commercial scale. Einchem has developed a heterogeneous

titanium silicate zeolite catalyst, commercial tradename of TS-1.37 This catalyst, totally

heterogeneous in nature, has a wide synthetic use for hydrogen peroxide oxidations.

Einchem has commercialized this catalyst for the hydroxylation of phenol to

hydroquinone and catechol.

The catalysts used in this research will employ hydrogen peroxide and dioxygen.

Hydrogen peroxide was chosen as the primary oxidant for hydrocarbons because of its

ease of handling and oxidizing strength. Dioxygen is the desired oxidant but hydrogen

peroxide can be used safely with hydrocarbons, while dioxygen requires one to remain

within the explosion limits.
















































Figure 1-4: Methods for Activating Hydrogen Peroxide













CHAPTER 2
HOMOGENEOUS CATALYZED PARTIAL OXIDATION OF METHANE WITH
HYDROGEN PEROXIDE AND OXYGEN

Introduction


Cis-ruthenium oxo complexes were previously reported to be effective catalysts to

activate hydrogen peroxide and molecular oxygen for the selective oxidation ofalkenes38

and alkanes.39,40 The sterically hindered complex cis-[Ru(dmp)2(H0)J](PF6,) (II), where

dmp is 2,9-dimethyl-l,10-phenanthroline (Figure 2-1), is a precursor that can be oxidized

with hydrogen peroxide to form cis-[Ru(dmp)2(O)(HO2)](PF,), or cis-

[Ru(dmp)(O)j2(PF6)2 as shown in Equations 2-1 and 2-2.






N N
H3C CH3


Figure 2-1: Structure for 2,9-dimethyl-l,10-phenanthroline


Unsuccessful attempts to isolate these oxo complexes led to their characterization based



(Equation 2-1) cis-[Ru"(dmp),(S)2](PF6)2 2

cis-[Ru'(dmp)2(S)(O)](PF6)2








(Equation 2-2) cis-[RuV(dmp)2(S)(O)](PF,),

cis-[Ru"'(dmp)2(O)2(PF )2



on isobestic points in the electronic spectra and changes in solution NMR as hydrogen

peroxide is added to II. The electronic spectra of the oxidized complex are similar to

those of analogous ruthenium-oxo complexes.41 Cis-{Ru(O),}2 complexes isomerize to

the trans-complexes which are weaker oxidants than the cis analogues.41,42,4344.45 The

novel aspect of the dmp ligand is its steric requirement which prevents isomerization of

the cis-[Ru(O),]2* complex and inhibits formation of both tris complexes and stable t-oxo

dinuclear species. The intra-ligand repulsions and neutral ligand charge also make the

high oxidation state ruthenium center electropositive.4'

Preliminary results describe the use of this complex to catalyze the oxidation of

alkanes using H202 as the oxidant.40 Figure 2-2 summarizes the proposed46 catalytic

cycle for alkane oxidation via cis-[Ru(dmp),(0)2 2. The cis-[Ru(II)(dmp)2(H20)2]2+

precursor is oxidized rapidly by H202 to form the (III),(IV) and (VI) oxidation state

complexes. In alkane oxidations, cis-[Ru(III)(dmp)2(H20)(OH)]2 reacts with dioxygen to

produce the cis-[Ru(IV)(dmp),(HO0)(O)]2' but cis-[Ru(II)(dmp)2(H20)2]2 is reported to

be unreactive with molecular oxygen.46 When appreciable concentrations of Ru(IV)

exist, the Ru(II) complex reacts with it to form cis-[Ru(III)(dmp)2(H20)(OH)]2*. For

kinetic reasons, alkane oxidations require the electrophillic cis-[Ru(VI)(dmp),(O)2]2+

















Oxidized Substrate
Substrate [Ru(dmp)2(H20)]2
H202

1< H20
[Ru(dmp)20,J]2 [Ru(dmp)2(H20)0]2*




H2O H202


Figure 2-2: Proposed Catalytic Cycle for Alkane Oxidation via cis-[Ru(dmp),(O)2]2-







complex. The cis-[Ru(IV)(dmp)2(H20)(Oh)2+ formed after oxygen atom transfer must be

oxidized back to the Ru(VI) complex with hydrogen peroxide40 to make the system

catalytic in CH3CN solvent. Figure 2-3 illustrates the catalytic cycle of the ruthenium

analogue.

Methane was selected as the substrate for study because the oxidation of natural

gas into a liquid product is of great industrial and economic importance.47 Figure 2-4

depicts the worldwide distribution of natural gas resources. Currently, methanol is

produced from methane by the steam reforming of methane into synthesis gas, followed

by the catalytic conversion of the syn gas into methanol. In order to increase efficiency,

the direct partial oxidation of methane to methanol is the desired pathway.48 However,

hydrogen atom abstraction, commonly involved in methane oxidation, occurs more

rapidly with methanol than methane resulting in the over-oxidation of methanol to CO,.

This research extends earlier reports of the use ofcis-[Ru(VI)(dmp)2(0)2]2' as a

catalyst for methane oxidation. Greatly improved conversions to oxygenates result by

trapping methanol and slowing its over-oxidation. Also demonstrated is an improved

catalyst precursor, evidence for oxidation of methane by 02 with this catalyst, and

demonstrate oxidation of methane by peracids.

Experimental


Materials and Methods

RuC13 x HO0, 2,9-dimethyl-l,10-phenanthroline, LiCI, NaPF6, 60% HPF6,

NH4PF6, NaCF3SO3, were all used as received from Aldrich. Ethylene glycol,

acetonitrile, glacial acetic acid, acetic anhydride, 4A molecular sieves and hydrogen



















H202 (Fast) H202 (Fast)
02 (V. Slow) or 02 H202 (Fast)


Ru" Ru" + RuV Ru + Ru Rv() Ru(0)
Ru Rui "' RuV(O) RuV(O),






Alkane
Epoxide Alkene Oxidized
Alkane


Figure 2-3: Detailed Catalytic Cycle for Ruthenium Analogue






















Africa
Middle East 10%
25%




Russia
30%

Asia-Oceania
13%

N. America Am W Europe
12% 5% 5%


Figure 2-4: Distribution of Worldwide Natural Gas Resources








peroxide (30%) were all used as received from Fisher Scientific (ACS grade).

Acetonitrile was distilled over P205 under dinitrogen and was stored over 4A activated

molecular sieves. Methane (99.99%) was purchased from Matheson and used as

received.

Physical Measurements

UV-Vis measurements employed a Perkin Elmer lambda-6 spectrophotometer.

The pH measurements were made with a Fisher Accumet model 630 pH meter. NMR

spectra were recorded on a Varian VXR300 spectrometer. FAB mass spectral data were

obtained by Dr. David Powell (U.F.) in a m-nitrobenzyl alcohol matrix. U.F. Analytical

Services performed elemental analyses.

Synthesis of Compounds

Cis-Ruthenium(II) Bis(chloride)bis(2,9-dimethyl-l,10-phenanthroline)

Monohydrate, [Ru(dmp),C1,] H2O (I). Cis-[Ru(dmp)2Cl2] H,O was synthesized as

reported.40 Analysis: Calculated for C2,H26N4OC12Ru; C, 55.45; H, 4.29; N, 9.24. Found:

C, 55.65; H, 4.35; N, 9.29.

Cis-Ruthenium(II) Bis(aquo)bis(2,9-dimethyl-l,10-phenanthroline)

bis(hexafluorophosphate), Cis-[Ru(dmp),(HO0)2](PF6) (II). A 1.0 g (1.7 mmol)

portion of I was dissolved in 150 ml of deionized H20, under N2 by heating 500C for 30

minutes. After adding 50 ml of a saturated NaPF, (aq), the solution is cooled, placed in

an ice bath for two hours, and filtered. The precipitate is redissolved in H20 by heating to

500C; a saturated NaPF6 (aq) solution is added and reprecipitated as above. Re-

crystallization and exchange of Cl for PF6 is repeated until the filtrate affords a negative







chloride test with AgNO3. The product is dried in vacuum at 600C overnight. Analysis:

Calculated for C2,H2,N402P2F12Ru: C, 39.85; H, 3.32; N, 6.64. Found: C, 39.46; H 3.29;

N, 6.72.

For comparison purposes cis-[Ru(dmp)2(H20)2](PF6)2 was also synthesized using

the previously reported procedure.40 Analysis: Calculated for C28H2,N4O2P2F,2Ru: C,

39.85; H, 3.32; N, 6.64. Found: C, 39.57; H 3.27; N, 6.63.

Cis-Ruthenium(II) Bis(aquo)bis(2,9-dimethyl-l,10-phenanthroline)

bis(trifluoromethanesulfonate), Cis-[Ru(dmp)2(HI0),](CF3SO3)2 (III). A 1.0g (1.7

mmol) portion of I is slowly dissolved in H20 as above. Following the addition of 50 ml

of a saturated aqueous solution of NaCF3SO,, the resulting solution is cooled, and placed

in an ice bath to complete precipitation. The product is collected, the filtrate tested for

chloride, and recrystallized with CF3SO3" until the filtrate gives a negative chloride test.

The resulting solid is dried under vacuum at 600C overnight. Analysis: Calculated for

C30H34N4OSSFRu, C, 42.25; H, 3.31; N, 6.57. Found: C, 41.40, H, 3.26; N 6.39.

Oxidation Procedure

The pressurized oxidations were carried out as previously described49, in glass,

batch hydrogenation reactors. Figure 2-5 provides an illustration of such batch reactor.

The reaction mixtures were varied as described in the table footnotes. Blank runs

omitting certain reactants and solvent components are also described in the tables.

Oxidations with oxygen use 30 psig of a 5% 02 in helium mixture, to remain outside the

explosion limits, and 20 psig methane. Reaction temperatures normally were maintained

















Pressure
Gauge


Gas Inlet/Purge


Gas Centrifuge
Bottle


Figure 2-5: Diagram of Batch Type Hydrogenation Reactor







between 75 to 770C. To remove air initially present, nitrogen gas is purged through the

reactor, the apparatus is pressurized with the substrate, which is then released, and re-

pressurized to the desired pressure.

The oxygenated products of the reaction were analyzed and quantified with a

Hewlett-Packard 5890 Gas Chromatograph equipped with a FID detector and outfitted

with a 30m Alltech RSL 160 column (5im thickness). Helium was employed as the

carrier gas. Carbon dioxide and carbon monoxide analyses were performed with a Varian

3700 Gas Chromatograph equipped with a TCD detector outfitted with a 15' Carboxen

Column. Methane was quantified by gas chromatography (TCD) using N2 as an internal

standard. Three chromatograms were measured for each sample using injection volumes

of 0.lml for gas and 0.1Al for liquid samples.

The following definitions describe terms used in the presentation of the results.

Selectivity to any product is the moles of a given product divided by the total moles of all

products formed expressed in percent. The percent CO2 produced is moles of CO2

divided by the total moles of all products. The selectivity to oxygenates is the moles of

HCO, CH3OH and CH3C(O)OCH3 divided by the total moles of products in percent.

Traces of CH,(OCH3),, HCOOCH3 are formed, but not quantified. Percent peroxide

efficiency is the moles of H202 needed to account for all oxidized products including CO2

divided by the moles ofH202 consumed. The percent conversion ofCH4 is the moles

carbon in the oxidized products divided by the moles of CH4 added to the reactor.

Safety Precautions, the combination of molecular oxygen with organic

compounds and solvents at elevated temperatures and pressures are potentially explosive.







Extreme caution should be taken during the charging and disassembly of the

experimental apparatus. Equipment which can generates sparks must be avoided, a safety

shield and cooling of the batch reactor in an ice bath for 30 minutes prior to disassembly

is recommended.

Results and Discussions


Anion Modification

Reproducibility for the oxidation of CH4 cis-[Ru(dmp),(H20)2](PF6)2 catalyst

precursor40 depends on the purity of the complex. Chloro complexes are inactive41 and

excess AgPF, used to remove the chloride from the dichloro precursor inhibits3839,40 the

reaction. Using NaPF, instead of AgPF, requires a large amount of NaPF6 and repeated

recrystallizations. The hexafluorophosphate anion of the final product is hydrolytically

unstable and extensive etching of the glass vial occurs during storage. With more than

20% fluorine, verification of the complex purity by elemental analysis is difficult.

Complex degradation is evident as well as the inability to obtain an appropriate NMR

spectrum for complex II as illustrated in Figure 2-6.

In an attempt to overcome these problems, the anion was changed to

trifluoromethanesulfonate, a poorly coordinating anion. A shortened synthesis time, need

for less anion in the synthesis, more reliable elemental analyses, and stability towards

hydrolysis are immediate advantages of this new complex. The direct effects of altering

the anion are exhibited in the lack of complex degradation and the ability to obtain an

identifiable NMR spectrum. A NMR for this newly synthesis complex III is provided in

Figure 2-7.
















9 8 7


3 2 1 PPM

Figure 2-6: 'H NMR Spectrum for cis-[Ru(dmp)2(H20)(PF6)2





39







,,
m 7






0 7


Figure 2-7: 'H NMR Spectrum for cis-[Ru(dmp)2(H0)2](CF3SO,)2


1 3 2 PPH


a


-








ivUII"IWlUII 4Yi IvLVU JAIULkVUII Ui IVIcUIOAUi WIUI IIYUIUFII FUULVAIU


The trifluoromethanesulfonate derivative, complex III, is a potent catalyst for the

oxidation of CH4 by hydrogen peroxide in an acetonitrile solvent. Within 48 hours, at

75C, CO,, CO and traces of both methanol and formaldehyde appear which correspond

to 60% of the initial concentration of methane. A more active catalyst, less methanol and

more CO2 result with III than with the PF." derivative (II).

Previous reports suggest the di-oxo complex is necessary for the activation of the

C-H bond in methane40. In water, the hydrogen peroxide potential46 gives a negative free

energy for the formation of the di-oxo species only when the solution pH is less than 5.

However, at a pH of 2 in water, the oxidation of CH4 by H202 with this catalyst is not

observed at 750C.46 The relevant potentials are not known in acetonitrile but spectral

studies indicate a high oxidation state complex is formed upon addition of H202.

The best condition reported for methane oxidation40 with complex II in

acetonitrile yield only 5 turnovers for methanol accompanied by extensive over-oxidation

of methane to CO2 occurs. This prompted an experiment to determine the activity of this

catalyst for methanol oxidation to CO2. Methanol, catalyst (6.6x10" moles) and

hydrogen peroxide (5.0x10"2 moles) reacted in 4 hours at 750C to convert 97% of the

methanol to carbon dioxide and carbon monoxide. Thus, the main challenge to the use of

these ruthenium catalysts for the selective oxidation of methane to methanol is to inhibit

the over-oxidation. The successful oxidation of methane in HS0450,51 and

CF3C(O)OHS2,53,54 to large quantities of available methanol can be attributed to trapping

methanol as CH3OSO3H and CF3C(O)OCH3 respectively, and stabilizing it from over-








oxidation. This led to an attempt to trap methanol in our system as an ester by reaction

with acetic acid and acetic anhydride. Acetic acid was used since it is an acidic solvent,

and relatively difficult to oxidize. With aqueous H202 as the oxidant and water formed in

the reaction, the anhydride can function to keep the water concentration low enhancing

ester formation, as demonstrated in Equations 2-3, 2-4 and 2-5.



(Equation 2-3) CH3OH + CH3C(O)OH CH3C(O)OCH3 + H20

(Equation 2-4) H20 + CH3C(O)OC(O)CH3 2 CH3C(O)OH

(Equation 2-5) CH3C(O)OC(O)CH3 + CH3OH CH3C(O)OCH3 +

CHC(O)OH



The anticipated product, methyl acetate, could be subsequently hydrolyzed to produce

methanol and regenerate acetic acid.

The oxidative stability of methyl acetate to over-oxidation by our catalyst was

shown with an experiment in which methyl acetate, catalyst (6.6x10-5 moles) and

hydrogen peroxide (5.0x10-2 moles) dissolved in acetonitrile were reacted for 4 hours at

750C. An 8% decrease in methyl acetate and formation of small amounts of carbon

monoxide and carbon dioxide result, indicating that methyl acetate has the necessary

stability to be an effective trap for methanol. The 8% oxidation observed could proceed

through methanol formed from the equilibrium shown in Equation 2-3.

Using catalyst III and a solvent mixture composed of equal volumes of

CH3CN/CH3COOH/(CHCO)20, 7.4 millimoles of oxygenates formed in 24 hours







representing a significant improvement over the previously reported 0.5 millimoles in

acetonitrile.3839,40 Methyl acetate, methanol, formaldehyde, formic acid, methyl format,

carbon monoxide and carbon dioxide are all detected as products. As mentioned above,

in the absence of acetic acid and acetic anhydride, only CO and CO, and trace amounts of

formaldehyde are obtained with III as the catalyst.

Increasing the reaction time and hydrogen peroxide concentration in a series of

experiments, gave increased amounts of CO2, but led to an upper limit of 1.5x102 M

methyl acetate in this solvent mixture. Apparently when methyl acetate approaches this

concentration, the equilibrium concentration of CH3OH reaches a level at which its rate of

oxidation becomes equal to its rate of formation from methane. Decreasing the water in

solution should produce a lower steady state concentrations of methanol (i.e. the position

of the equilibrium in Equations 2-3 and 2-5 is shifted towards ester formation) and

increase the trapping efficiency leading to decreased CO, and increased methyl acetate.

To remove water, 4A molecular sieves were added to the reaction mixture and shown to

have a significant effect on the conversion of methane to methyl acetate. In comparable

24 hour runs, the conversion to oxygenates increased from 3 millimoles to 7 millimoles

(Table 2-1, Experiment 1) with the addition of sieves even though some peroxide

decomposition by the sieves occurs, vide infra. This result reinforces the proposal that

efficient trapping is the main challenge for effective methanol synthesis.

Variation of the reaction time in Experiments 2 and 3 show that in four hours the

maximum amount of methyl acetate and minimum amount of CO2 form. The shorter

time also led to an increase in the amount of formaldehyde. The methyl acetate formed

corresponds to 61 turnover numbers (5 mmoles, 6.6x102 M) with a selectivity of 55%















Table 2-1: Oxidation Results for Methane @ 750C with cis-[Ru(dmp)2(H20)2(CF3SO,3)
Using H202
Experiment MeCOOMe CO, Total CH,
Number mmol,(%) mmol,(%) Oxygenates Consumed
mmol,(%) (%)
1" 5,(44) 5,(44) 7,(62) 30

2 4,(40) 3,(30) 7,(69) 26

3V 5,(57) 1,(11) 7,(80) 19

4Af 10,(49) 9,(44) 10,(49) 62

5bs 4,(70) 1,(17) 5,(87) 25

6b- 9,(80) 3,(27) 8,(71) 49

7b-1 2,(100) 0 2,(100) N/A

8i 3,(47) 0 4,(62) 28

a. Reactions 1-4 used 6.6 x 10' moles cis-[Ru(dmp),(H20)2](CF3SO,)2 catalyst, 5 ml 30% H20,
(5.0xl0-2 moles). The solvent mixture is 20 ml acetonitrile, 20 ml glacial acetic acid, 20 ml
Acetic Anhydride, Initial methane pressure was 40 psig corresponding to 23 millimoles. 3.5g
4A molecular sieves (MS) were used. The percent products are based on total products seen
from all sources.
b. Reactions 5-8 used 2.4 x 10'5 moles cis-[Ru(dmp)2(H20)](CF3SO,), catalyst, 1 ml 35% H20,
(l.lxl0"' moles). The solvent mixture is 3 ml a-dichlorobenzene, 5 ml glacial acetic acid, 10
ml Acetic Anhydride, Initial methane pressure was 40 psig corresponding to 23 millimoles.
1.0g 4A molecular sieves (MS) were used.
c. Reaction Time 24 hours. The percent products are based on total products from all sources.
d. Reaction Time 12 hours. The percent products are based on total products from all sources.
e. Reaction Time 4 hours. The percent products are based on total products from all sources.
f. Blank experiment, performed in the absence of catalyst, Reaction Time 1 hour.
g. Blank experiment, performed in the absence of catalyst and acetonitrile, Reaction Time 4 hours.
h. Reaction Time 4 hours.
i. Blank experiment, performed in the absence of catalyst and methane, Reaction Time 4 hours.
j. Blank experiment, performed in the absence of catalyst, Reaction time 4. Hours.








while formaldehyde corresponds to 24 turnover numbers (2 mmoles) with a selectivity of

28%. The CO and CO2 produced correspond to 9 and 67 turnover numbers (0.2 mmoles

and 5 mmoles) respectively. The total turnover numbers for all the products is 198,

which if they all arose from methane would correspond to a methane conversion of 38%.

The total moles of oxidant used are 1.3x10"2, so 21% of the peroxide has decomposed.

Peracid Formation and Reactivity

In view of the difficulty of oxidizing methane and its low concentration in

solution, the importance of running control experiments to determine the source of

products cannot be overemphasized. When the catalyst was eliminated from the reaction

mixture, methyl acetate formed in significant amounts, Experiment 4 of Table 2-1. In

earlier research from this laboratory it was shown that hydrogen peroxide is catalytically

activated by forming peracids with organic acids55. To determine if peracetic acid forms,

acetic acid, acetic anhydride, and 35% aqueous hydrogen peroxide were stirred at 250C

for thirty minutes in the absence of sieves. The difference in an iodometric titration of the

products for total oxidant, and a cerium titration, for hydrogen peroxide, indicates by

difference that 20% of the HO,0 was converted to peracid (1.0xl0O2 moles). At this point

methane (40 psig) was introduced and the reaction vessel placed in an oil bath at 750C.

Experiment 4 (Table 2-1) shows that, in one hour of reaction time 10 mmoles of methyl

acetate (49% selectivity), traces of formaldehyde, 0.4 mmoles of carbon monoxide and a

large amount of carbon dioxide (9 mmoles or 44% of the products) were formed. This

reaction led to about 19 mmoles of total products which, if they all arose from methane

would correspond to 89% of the methane. The decrease observed in the amount of








methane by TCD is approximately 60% (Table 2-1) implying some of the oxygenates

arise from peracid decomposition.

To limit over-oxidation by peracid, the reaction time was decreased to twenty

minutes. The reaction produced 6.4x10" moles of methyl acetate (69% selectivity)

(1.1x10-' M), 1.6x0l moles of formaldehyde (0.04% selectivity), 3.9x10" moles of

carbon monoxide (0.06% selectivity) and 2.9x103 moles of carbon dioxide (31.1%). The

peroxide utilization was 97% and 1.0x10.2 moles of products were observed. The

decrease in reaction time leads to less products, but increased selectivity to methyl

acetate.

Blank runs were performed with and without sieves to determine if hydrogen

peroxide and acetonitrile form a peracid, CH3C(OOH)NH, which oxidizes methane to

produce methyl acetate. In these experiments, acetonitrile, hydrogen peroxide and

methane were added together and allowed to react for 48 hours at 750C. Iodometric

titrations also confirmed the absence of peracid.

The next concern is the source of the products in the blank, as shown in

Experiment 5. Oxidations of paraffins by 3,5-dinitroperbenzoic acid, perbenzoic acid and

perfluoroperacetic acid result56 in more than 50% conversion of the paraffin. A radical

mechanism is suggested57.58 for the formation of the hydroxylation products that is

initiated by the decomposition of peracids as shown in Equations 2-6, 2-7, and 2-8.



(Equation 2-6) R-CO3H RCO2 + 'OH

R'+ CO2 +'OH








(Equation 2-7) R-H + R" 'R" + R-H

(Equation 2-8) 'R + R-CO3H 'R-OH + R-CO,'



The R-CO," radical decomposes to form CO, and the alkyl radical (R') which

abstracts a hydrogen atom from the paraffin R'H to form the 'R' alkyl radical which

reacts with the peracid to form the alcohol generating R'OH and another RCO2". In our

system, with CH4 as the alkane and peracetic acid as the oxidant, decomposition of

CH3CO3H would generate methyl radicals which may exchange with methane in

Equation 2-7 and would have to react as shown in Equation 2-8 to produce CHOH.

Methanol formation would in effect arise from peracid decomposition.

Alternatively, RCO2*can react with R'H to form R" which then reacts as shown

in Equation 2-8. In this case CH4 would be converted to methanol by CH3CO2, formed

in Equation 2-5, reacting with methane, as shown in Equation 2-6, to form methanol via

Equation 2-8. Methane oxidation by peracid results, in contrast to the literature

mechanism where the products would derive from the solvent. In view of the possible

exchange of methyl radical with labeled methane, Equation 2-7, labeling experiments will

not be definitive. One also notes that less methyl acetate is produced with the ruthenium

complex present (Experiment 3, Table 2-1), than with it absent in Experiment 4,

suggesting metal complex decomposition of H202. Crude estimates of the methane that

disappeared in the experiments described above, indicate the metal complex is involved

in Experiments 1-3 but more compelling evidence is required.








The question concerning the reactivity of methane with peracid raised above

impacts not only on the metal catalyzed oxidations in acetic acid-acetic anhydride solvent

but also on peracid oxidation ofalkenes.56 Blank runs at 75*C, with catalyst, hydrogen

peroxide, acetic acid/acetic anhydride, acetonitrile and no methane, produced one

millimole of methane and traces of methanol after four hours of reaction. A similar

experiment in a-dichlorobenzene, showed only traces of methane and methanol. The

source of methane is thought to arise from solvent facilitated peracid decomposition in

acetonitrile. This proposal was not investigated further, but the solvent was switched

from acetonitrile to a-dicholorbenzene and Experiment 5-8 were carried out. Since

labeling experiments are uninformative, efforts were expended to accurately determine

the methane disappearance to provide a material balance. To facilitate the metal

catalyzed path, the catalyst and hydrogen peroxide concentrations were increased by

20%. Experiments 5 and 7 indicate that the beneficial effect of the sieves removing water

is accompanied by the deleterious decomposition of hydrogen peroxide. In Experiment 6,

methane oxidation is catalyzed by III, giving an amount of products equivalent to 49%

conversion to oxygenates. Nine millimoles of methyl acetate, two millimoles of other

oxygenates and 3 millimoles of CO, were formed. In Experiments 5 and 6 a good

material balance for methane disappearance and product formation resulted.

When an identical blank experiment with no catalyst or methane was carried out,

two millimoles of methyl acetate and no other products were formed from peracid

decomposition, Experiment 7. When no catalyst was used and methane added,

Experiment 8, three millimoles of methyl acetate and one millimole of other oxygenates





48

were formed. Experiment 7 indicates that methyl acetate can arise from the solvent via

peracid decomposition. Since the amounts of oxygenates produced are doubled when

methane is added, Experiment 8, methyl acetate is produced by the reaction of the peracid

with methane. However, in Experiment 6 when the ruthenium catalyst is present, the

almost three-fold increase in oxygenates compared to Experiment 7 establishes the metal

catalyzed path.

All methane conversions given in Table 2-1 is based on the amount of methane

consumed in the reaction. In the absence of a material balance for runs 1-4, an upper

estimate results for conversions. Since peracid formation is not expected for oxidations

with 02, trapping with the acetic acid-acetic anhydride mixture was investigated next with

this oxidant.

Oxidation of Methane with Molecular Oxygen

The direct conversion of methane to methanol using molecular oxygen is even a

greater challenge than using hydrogen peroxide as the oxidant. Table 2-2 provides results

for experiments using 30 psig of 5% 0O in helium as the oxidant. The moles of oxidized

products formed clearly exceed the moles hydrogen peroxide (2.5x105) added to initiate

the reaction by oxidizing III. After 48 hours (Experiment 9), methyl acetate (5 turnover

numbers, 0.32 mmoles, 19% selectivity), and relatively large amounts of formaldehyde

(20 turnover numbers, 0.65 mmoles, 38% selectivity) methanol and methyl format were

obtained. GC analysis (TCD) shows carbon monoxide (9 turnover numbers, 0.19

mmoles, 11% selectivity) and carbon dioxide (32 turnover numbers, 0.52 mmoles, 30%

selectivity). At most a total of 0.4 turnovers (2.5x10" moles) for methyl


























Table 2-2: Oxidation Results for Methane @ 750C with cis-[Ru(dmp)2(H20)](CF3SO3)2
Using H202 and 02


a. 6.6 x 10-5 moles cis-[Ru(dmp)2(HO0)](CF,SO,)2, 20 ml acetonitrile, 20 ml glacial acetic
acid, 20 ml Acetic Anhydride, 20 psig CH, (1.0x102 moles), 3.5g Molecular Sieves (MS).,
750C,
b. Reaction time 48 hours 2.5 MI 30% H,20 (2.5x105 moles).
c. Reaction time 24 hours 2.0 ui 30% H20, (2.0x 10' moles).
d. Blank Experiment, performed in the absence of H20, 48 hours.
e. Blank Experiment, performed in the absence of catalyst 48 hours.







acetate could have resulted from the 2.5x105 moles ofH202 assuming 100% peroxide

efficiency. Clearly this catalytic system utilizes 02 for the oxidation of methane at 75C.

The total moles of products correspond to 1.7 mmoles, corresponding to a 17% percent

conversion of methane. The formation of methyl acetate indicates the trapping of

methanol is occurring.

The reaction of methane with 0O was carried out for 24 hours (Experiment 10).

Less CO2 and less methane conversion results. The products are methyl acetate (5

turnover numbers, 0.30 mmoles, 50% selectivity), carbon monoxide (3 turnover numbers,

0.06 mmoles, 10% selectivity) and carbon dioxide (16 turnover numbers, 0.26 mmoles,

40% selectivity). A total of 0.63 mmoles are formed and the percent methane conversion

was 6%. Since the same amount of methyl acetate forms in 24 or 48 hours, the extended

time results in the oxidation of methanol that is in equilibrium with methyl acetate

(Equation 2-2).

Experiment 11 was performed to determine if di-oxygen could catalyze the

reaction without hydrogen peroxide to initiate formation of the ruthenium-oxo species.

After 48 hours no oxidation products were observed. The [Ru(dmp)2(H20),2+ catalyst

precursor utilizes molecular oxygen in epoxidations38.39 only after an initiation period in

which an alkyl hydroperoxide is generate to oxidize the complex to the (IV) oxidation

state. Upon oxygen atom transfer to the substrate from the (IV) oxidation state the (II)

oxidation state forms, which reacts with the Ru (IV) to generate Ru (III). The (III)

oxidation state reacts with molecular oxygen to regenerate the higher oxidation state

metal-oxo species.46 Since 02 will not form methylhydroperoxide from methane, the

inactivity of the ruthenium (II) complex with 02 (Experiment 11), and the required







initiation by H202 suggests that a metal-oxo species generated from ruthenium (III) by 02

is involved by either reacting with ruthenium (III) to form a metal-oxo species, catalyzing

the reduction of O to H202 by methane or inhibiting radical chain termination steps.

Conclusions


The use of the sterically hindered cis-[Ru(dmp)2(H20)2,] complex to catalyze the

oxidation of methane has been investigated. Using the poorly coordinating

trifluoromethanesulfonate anion in place ofhexafluorophosphate an improvement in

reproducibility and catalytic activity of the complex has been demonstrated. This

complex is a precursor to the proposed di-oxo catalyst that oxidizes methane to CO2 with

hydrogen peroxide at 750C in acetonitrile. In order to stop the reaction at the methanol

intermediate, an acetic acid/acetic anhydride mixture was chosen to trap methanol as

methyl acetate. This ester, which is not as readily oxidized as the alcohol, increases the

selectivity and inhibits the over-oxidation of methanol.

In oxidations with hydrogen peroxide, control experiments indicate the formation

of peracetic acid from the solvent mixture of acetic acid/acetic anhydride. In the absence

of methane, the peracid decomposes to produce methyl acetate. When the reaction is

carried out in identical conditions under methane, a doubling of the methyl acetate,

produced shows that methane is oxidized by peracid to form methyl acetate. A

substantial increase in methyl acetate, for identical reaction conditions with metal

complex added, clearly demonstrates that metal catalyzed oxidation occurs. The material

balance shows that the predominant source of the methyl acetate is methane with at most

a small contribution from peracid decomposition.





52

The cis-[Ru(dmp)2(H20)2]2* complex is a catalytic precursor for the oxidation of

methane with dioxygen in acetonitrile at 750C. It is necessary to initiate the reaction by

oxidation of the complex with trace amounts of hydrogen peroxide. With molecular

oxygen as the oxidant, a 6% conversion of methane and 50% selectivity to methyl acetate

can be obtained after 24 hours of reaction.














CHAPTER 3
OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING A
RUTHENIUM METAL-OXO CATALYST

Introduction


The transformation of saturated hydrocarbons into their oxygentated derivatives

(i.e. alcohols, aldehydes, ketones and carboxylic acids) has been the subject of intense

investigation over the past two decades. Alkanes given their great abundance offer a cost

effective and ideal feedstock for an industrial process.59 However, the chemical inertness

of saturated hydrocarbons make activation extremely difficult at mild conditions. The

oxygenation products, vital intermediates in many industrial processes, are then converted

into commercial products. In order for a process to be feasible on the industrial scale the

catalyst must demonstrate two properties: selectivity towards the partial oxygenates and

exhibit a specific regioselectivity. To achieve these requirements a commercial industrial

application must maintain high temperatures and pressures.60 Currently, research efforts

are being directed towards the development of new efficient catalytic systems, which are

able to oxygenate, saturated hydrocarbons under mild conditions using hydrogen

peroxide and/or molecular oxygen.

Reagents with the capacity to oxidize paraffins and arylalkanes have been known

for well over a century.61,62 Two such compounds are chromyl chloride (CrO2-Cl) and

permanganate (MnO").61.62 However, these oxidants are stoichiometric and a catalytic







one is desired. Numerous researchers63,64.65 have reported organic radical mechanisms

via homogeneous or heterogeneous oxidation. The essential properties that allow these

oxidants to oxidize a relatively inert C-H bond are not fully understood.

A number ofmetal-oxo and metal-oxide surfaces perform as reagents or catalysts

for the oxidation of hydrocarbons, on industrial and laboratory scales.1,66,67,68 Metallo-

enzyme sites also activate hydrocarbons, two well-documented examples being

Cytochrome P-45069 and methane monooxygenase47 (MMO).

Current research in the area of hydrocarbon activation has utilized a variety of

oxidants, catalysts and reaction conditions. Barton et al.70,71 has developed the Gif

system, as well as a number of variations (i.e. Barton-Gif). Gif uses a pyridine/acetic

acid solution, substrate, Fe catalyst and an oxidant, usually t-BuOOH. The mechanism of

oxidation has been reported as proceeding through the involvement of free radicals.

Que72 has been investigating the use of di-iron complexes, in this case a high valent

Fe2(9-O), moiety. This complex has been proposed as the key oxidizing species for

methane monooxygenase, as well as for other non-heme di-iron enzymes.

Ribonucleotide reductase and fatty acid desaturase are mentioned.

This research demonstrates the cis-[Ru(dmp)2(H20)2(CF3SO,)2 as a precursor for

the catalytic oxidation of higher linear and branched alkanes with hydrogen peroxide in

acetonitrile at 750C. The complex generates a large fraction of products oxygenated at

the primary carbon position. An increase in selectivity towards the alcohol product is

observed upon addition of CuCI2. The effect of temperature on reactions with this catalyst







is also investigated to determine the effect on overall hydrocarbon conversion, selectivity

distribution of oxygenates.

Experimental


Materials and Methods

RuCI x H20, 2,9-dimethyl-l,10-phenanthroline, LiCI, NaCF3SO3, hydrogen

peroxide (35%) and pentane (99.9%) were all used as received from Aldrich. Ethylene

glycol and acetonitrile were all used as received from Fisher Scientific (ACS grade).

Acetonitrile was distilled over P205 under dinitrogen and was stored over 4A activated

molecular sieves. Ethane (99.9%), propane (99%), butane (99%) and iso-butane (99.5%)

was purchased from Matheson and used as received.

Physical Measurements

UV-Vis measurements employed a Perkin Elmer lambda-6 spectrophotometer.

The pH measurements were made with a Fisher Accumet model 630 pH meter. NMR

spectra were recorded on a Varian VXR300 spectrometer. FAB mass spectral data were

obtained by Dr. David Powell (U.F.) in a m-nitrobenzyl alcohol matrix. U.F. Analytical

Services performed elemental analyses.

Synthesis of Compounds

cis-Ruthenium(I) Bis(chloride)bis(2,9-dimethyl-l,10-phenanthroline)

Monohydrate, [Ru(dmp)2C1, H,0 (I). Cis-[Ru(dmp),C1l] H20 was synthesized as

reported. Analysis: Calculated for C2,H,6N4OC12Ru; C, 55.45; H, 4.29; N, 9.24. Found:

C, 55.65; H, 4.35; N, 9.29.







cis-Ruthenium(II) Bis(aquo)bis(2,9-dimethyl-l,10-phenanthroline)

bis(trifluoromethanesulfonate), Cis-[Ru(dmp),(HzO)2](CF3SO3), (II). A 1.0g (1.7

mmol) portion of I is slowly dissolved in 150 ml of deionized H20 under N2 by heating at

500C for 30 minutes. After adding 50 ml of a saturated NaCF3SO3 (aq), the solution

cooled, placed in an ice bath for two hours and filtered. The precipitate is redissolved in

H20 by heating at 500C, a saturated NaCF3SO3 (aq) (9g NaCF3SO3/ 25ml deionized H20)

solution is added and reprecipitated as above. Recrystallization and exchange of Cl for

CF3SO3 is repeated until the filtrate affords a negative chloride test with AgNO3. The

product is dried in vacuum at 60C overnight. Analysis: Calculated for

C3oH34N40OS2,FRu, C, 42.25; H, 3.31; N, 6.57. Found: C, 41.40, H, 3.26; N 6.39.

Oxidation Procedure

The pressurized oxidations were carried out as previously described49, in batch

type hydrogenation reactors, Figure 2-5. The reaction mixtures were varied as described

in table footnotes. Blank runs were performed omitting certain reactants and are also

described in the text. Reaction temperatures were normally maintained at 750C unless

stated otherwise. To remove air initially present, nitrogen gas is purged through the

reactor, the apparatus is pressurized with the substrate, which is then released, and re-

pressurized to the desired pressure.

The oxygenated products of the reaction were analyzed and quantified with a

Hewlett-Packard 5890 Series II gas chromatograph equipped with an FID detector and

outfitted with a 30m HP 50+ (50% Ph Me Silicone Gum; 1 im thickness). Helium was

employed as the carrier gas. Carbon dioxide and carbon monoxide analysis were







performed with a Varian 3700 gas chromatograph equipped with a TCD detector outfitted

with a 15' Carboxen Column, I prm thickness. Helium was utilized as the carrier gas.

Concentrations of the substrate and oxidized products were quantified using acetonitrile

as an internal standard. Three chromatograms were measured for each sample using

injection volumes of 0.1 ml for gas and 0.1 gl for liquid samples.

A typical reaction mixture consisted of 60 ml of acetonitrile, 40 psig total pressure

for gaseous reactants, 1.6x10' moles of catalyst and 5ml of 35% hydrogen peroxide

(5.0x10'" moles). Experiments in the absence of oxidants were performed as blanks.

The following definitions describe terms used in the presentation of the results.

Selectivity to any product is the moles of a given product divided by the total moles of all

products formed expressed as a percent. The percent peroxide efficiency is the moles of

H20, needed to account for all oxidized products divided by the moles of H20,

consumed. The percent conversion of alkane is the moles carbon in the oxidized products

divided by the moles ofalkane added to the reactor.

Safety Precautions, the combination of molecular oxygen with organic

compounds and solvents at elevated temperatures and pressures are potentially explosive.

Extreme caution should be taken during the charging and disassembly of the

experimental apparatus. Equipment which can generates sparks must be avoided, a safety

shield and cooling of the batch reactor in an ice bath for 30 minutes prior to disassembly

is recommended.








Results and Discussion


Alkane Oxidation

Attempts to oxidize alkanes with hydrogen peroxide in H20 as the solvent at 750C

were unsuccessful after 48 hours with or without catalyst at solution pH values of 1 to 7.

As suggested previously46, the Ru(VI) oxidation state is necessary for the activation of

the C-H bond of alkanes. The [Ru(dmp)2(H20)2(CF3SO3)2 complex II has been shown to

be a potent catalyst in an acetonitrile solvent.73 Consequently alkane oxidations were

performed in acetonitrile.

Each alkane (ethane, propane, butane, iso-butane and pentane) was reacted in

acetonitrile (60ml) with hydrogen peroxide (5.0x 10mole) and catalyst (1.6x10" moles)

for 15 hours at 750C. The results are provided in Table 3-1 and Table 3-2. Propane, iso-

butane and pentane allowed us to determine catalyst activity in terms of selectivity and

regioselectivity.

After 15 hours, a 20.8% conversion of propane was observed corresponding to 3.4

mmoles of oxidized products. The selectivities of the oxidized products obtained are of

interest. Oxidation at the primary carbon position accounts for 65.1% or 2.2 mmoles of

the total oxidized products (1-propanol: 0.1 mmoles, 1.6% selectivity, and

propionaldehyde: 2.1 mmoles, 63.6% selectivity). Trace amounts ofpropionoic acid

were detected by gas chromatography analysis, however peak broadening and tailing

make quantification difficult. Oxidation at the secondary carbon position was also

observed, accounting for 34.9% or 1.2 mmoles of the total oxidized products (2-propanol:











Table 3-1: Oxidation Results for Ethane, Propane and Butane @ 750C with cis-
[Ru(dmp)2(H20)](CF3SO3)2 using H202
Oxidized Total Percent
Substrate" Products mmoles Peroxide Percent
mmoles Oxidant Efficiency Conversionb
(Selectivity)
Ethanol:
0.08 (2.3)
Ethane Acetaldehyde: 8.10 16.2 20.1
2.50 (69.3)
Acetic Acid:
1.00 (28.4)
1-Propanol:
0.05 (1.6)
2-Propanol:
0.07(1.9)
Propane Propanal: 6.80 13.5 20.8
2.10 (63.6)
Acetone:
1.20 (32.9)
Propionoic Acid:
Trace
1-Butanol:
0.08 (4.8)
2-Butanol:
0.94 (5.3)
Butane Butanal: 5.40 10.7 26.2
0.10 (5.5)
2-Butanone:
0.31 (18.1)
Butanoic Acid:
1.20 (66.3)
a: 1.6x10" mole cis-[Ru(dmp)2(H20)](CFSOz), 60ml acetonitrile, 5.0xl02 moles 35% HO2,
40psi substrate (1.9x10'2 moles), 15 hours at 750C.
b: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.
















Table 3-2: Oxidation Results for Iso-Butane and Pentane @ 750C with cis-
[Ru(dmp)2(H20)](CF3SO3)2 using H202

Oxidized Total Percent
Substrate' Products mmoles Peroxide Percent
mmoles Oxidant Efficiency Conversion'
(Selectivity)
Iso-Butanol:
Iso-Butane 3.90 (99.0) 3.90 7.8 20.5
Iso-Butyl-ol:
0.04 (0.9)
1-Pentanol:
0.58 (5.7)
2-Pentanol:
1.20 (11.5)
3-Pentanol:
Pentaneb 1.20 (17.1) 16.10 32.4 23.4
Pentanal:
2.20(21.7)
2-Pentanone:
1.90 (18.4)
3-Pentanone:
2.60 (25.7)
Pentanoic Acid:
Trace
a: 1.6xl0" mole cis-[Ru(dmp),(H20)](CFSO3), 60ml acetonitrile, 5.0x 10 moles 35% H202,
40psi substrate (1.9x 10" moles), 15 hours at 750C.
b: 2ml pentane (2.0x10"2 moles) used.
c: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.







0.1 mmoles, 1.9% selectivity and acetone: 1.1 mmoles, 32.9% selectivity). Activation of

propane by this catalyst produces a selectivity ratio of primary:secondary carbon

oxidation of 2:1. A peroxide efficiency of 13.5% and 43 turnover numbers (TON's) was

obtained for this reaction.

Ethane and butane were oxidized under identical conditions giving a 20.1% and

26.2% conversion respectively. The results are shown in Tables 3-1 and 3-2. Over-

oxidation of the hydroxylated products to the corresponding aldehyde, ketone and

carboxylic acid were also exhibited. As a result selectivity, overall conversion of the

alkane, peroxide efficiency and catalyst lifetime are decreased by over-oxidation.

Iso-butane was investigated to demonstrate the ability of this catalyst to produce

oxidation products at the tertiary and primary carbon position. After 15 hours of reaction,

3.9 mmoles of oxidized products were obtained. Oxidation at the tertiary carbon

accounted for 99.0% (3.9 mmoles) of the oxidized products. Also detected were trace

amounts (0.04 mmoles, 0.9% selectivity) ofisobutyl alcohol (2-methyl-l-propanol),

demonstrating some oxidation at the primary position.

Pentane can lead to oxidation at the CI, C2 and C3 positions. This is also the first

liquid alkane investigated. After 15 hours, a 23.4% conversion ofpentane was observed

accounting for 10 mmoles of oxidized products. Oxidation at the primary carbon position

(Ci) afforded: 1-pentanol (0.6 mmoles, 5.7% selectivity), valeraldehyde (pentanal) (2.2

mmoles, 21.7% selectivity) and a trace amount of valeric acid (pentanoic acid), once

again quantification was difficult due to peak broadening and tailing. Oxidation at the

secondary carbon (C,) position produced: 2-pentanol (1.2 mmoles, 11.5% sel.) and 2-

pentanone (1.9 mmoles, 18.4% sel.). Activation at the C3 position produces: 3-pentanol








(1.2 mmol, 17.1% sel.) and 3-pentanone (2.6 mmoles, 25.7% sel.). A total of 16 mmoles

of oxidant was consumed in this reaction yielding a peroxide efficiency of 32.4% and 50

TON.

Mechanism of Oxidation

Two mechanisms, hydrogen abstraction or "rebound" and oxygen atom insertion,

have been proposed for the oxidation of hydrocarbons with cis-[Ru(dmp),(H2O)j]2.40o74

These mechanisms are illustrated in Figures 3-1 and 3-2 respectively.

The hydrogen abstraction or "rebound" mechanism produces an alkyl free radical.

The free radical produced in the first step of the mechanism is the result of the cis-

{Ru(0)2)2 moiety abstracting a hydrogen atom from the alkane to produce a reduced

ruthenium-hydroxide species. The active metal center, Rum)=O functions as a free

radical4o which is capable of abstracting a hydrogen atom from the hydrocarbon. The

second step proceeds with the transfer of the hydroxo ligand to the formed radical to yield

the hydroxylated product. The catalytic cycle involves regeneration of the high valent

metal center by oxidation with 02 or H202. Other ruthenium complexes have been

proposed to proceed through this or a similar mechanism.75,76,77

The oxygen atom insertion mechanism has been proposed for the epoxidation of

olefins.40'78,79 It has also been reported that ruthenium-oxo complexes are able to

epoxidize alkenes via a non-radical mechanism.40 The first step in this mechanism for

alkane oxidation is the insertion of the oxygen bound to ruthenium into the C-H bond of

the substrate. This adduct forms a three coordinate oxygen and a five coordinate carbon





63





















0 0 0
Ru:O + H-CH RuoH + C + HO-CH3
3ORu:OH HCH3 Ruu


O
II
- Ru:O


+ H202


+ H20


Figure 3-1: Proposed Hydrogen Atom Abstraction Mechanism for cis-
[Ru(dmp)2(H20)]2 Complex








similar to CH,5 obtained in an S,2 mechanism. The next step is scission of the C-H bond

forming a coordinated alcohol followed by displacement of the alcohol from the

ruthenium complex. The final step is the regeneration of the high valent metal-oxo as

previously described in the hydrogen abstraction mechanism.

Results from this laboratory4o,73 indicate that the activation ofalkanes with cis-

[Ru(dmp)Oj2+ is proceeding via the hydrogen abstraction mechanism.

Addition of CuCl,

The majority of the oxidized products obtained are aldehydes, ketones and

carboxylic acids. In order to obtain increased amounts of the partially oxidized products

its direct over-oxidation must be prevented. Addition of a metal chloride could offer a

solution. Addition of NiCl2 to a methanol solution produces complexes with methanol

bound to the nickel.80 In this investigation formation of this metal-alcohol adduct was

determined by IR analysis. Coordination of the alcohol produced in our oxidations to a

metal could possibly retard over-oxidation and improve selectivity for the hydroxylated

product. To determine if an increase in selectivity towards the hydroxylated products can

be obtained one mole equivalent of CuCl2 (1.6xl04) was added to the previously

described reaction with all other conditions held constant. The results are shown in Table

3-3 and Table 3-4.

After 15 hours at 750C, an 18.2% conversion of propane was observed in the

presence of CuCI. The addition of CuCI2 led to a 2.4% decreased in conversion.

However, an increase in alcohol selectivity to 6.1% for 1-propanol and a 18.5% for























H-CH3


H TCH3
0+
Ru:O


0
____ I
Ru


O
II
------ Ru:O


+ HO-CH3


+ H,O


Figure 3-2: Proposed Oxygen Atom Insertion Mechanism for cis-
[Ru(dmp)2(H2O0)2] Complex


0
II
Ru:O +



H CH,


Ru:O


+ H202







2-propanol is observed, compared to selectivities of 1.6% and 1.9% when CuCI2 is

omitted. With the increase in alcohol selectivity a decrease in selectivity to the aldehyde

and ketone results. A 61.0% selectivity to propionaldehyde and 14.4% selectivity for

acetone is exhibited with the addition of CuCl,. CuC12 addition leads to a decrease in the

peroxide efficiency attributed to known peroxide decomposition by first row transition

metal salts.29 Increases in selectivity to the alcohol are also found for each of the alkanes

studied. These results are presented in Table 3-3 and 3-4.

To determine the role of CuCI2 a series of blank experiments were performed.

When 60 ml of acetonitrile, 1.6x104 moles CuCI2, 5.0x10" mole H20, and 1.9x10'2 moles

of propane were reacted for 15 hours at 750C no oxygenates were observed suggesting

CuCI2 is not participating in the oxidation of propane.

Experiments were performed to determine the stability of alcohol under reaction

conditions. Ethanol (Iml, 1.9x102 mol), acetonitrile (60ml), 1.6x10' moles catalyst and

5.0x102 moles H202 were reacted for 15 hours at 750C. A 77% decrease in ethanol and

formation of acetaldehyde and acetic acid was observed. TCD gas chromatography

analysis exhibited the trace amounts of carbon monoxide and carbon dioxide. The

quantity of oxidized products corresponds to the decrease in moles of ethanol. Analysis

of the resulting solution for peroxides with sodium metavanadate and iodometric titration

showed none was present. An identical experiment was performed with the addition

CuC12 (1.6x10" mol) to the reaction solution. After 15 hours the amount of alcohol

present in the final reaction solution corresponded to only a 43% decrease. Aldehyde and

carboxylic acid also formed in decreased amounts according to mass balance calculations.











Table 3-3: Oxidation Results for Ethane, Propane and Butane @ 75C with cis-
[Ru(dmp)2(H20)](CF3SO3)2 and CuCl2 using H20,

Oxidized Total Percent
Sub- Products mmoles Peroxide Percent
state' mmoles Oxidant Efficiency Conversionb
(Selectivity)
Ethanol:
0.18 (4.31)
Ethane Acetaldehyde: 9.10 18.1 23.2
3.10(74.7)
Acetic Acid:
0.88 (21.0)
1-Propanol:
0.21 (6.1)
2-Propanol:
0.64 (18.5)
Propane Propanal: 6.10 12.1 18.2
2.10 (61.0)
Acetone:
0.49 (14.4)
Propionoic Acid:
Trace
1-Butanol:
0.26 (12.5)
2-Butanol
0.15(7.1)
Butane Butanal: 4.90 9.7 30.9
0.11 (5.3)
2-Butanone:
0.46 (21.8)
Butanoic Acid:
1.10(53.2)
a: 1.6x10' mole cis-[Ru(dmp)2(H20)](CF3S03),, 1.6x10" mole CuCl, 2 H20,
60ml acetonitrile, 5.0xl0-2 moles 35% H202, 40psi substrate (1.9xl02 moles),
15 hours at 750C.
b: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.
















Table 3-4: Oxidation Results for Iso-Butane and Pentane @ 750C with
[Ru(dmp)2(HO0)](CF3SO3)2 and CuC12 using H202
Oxidized Total Percent Percent
Substrate' Products mmoles Peroxide Conversion'
mmoles Oxidant Efficiency
(Selectivity)
Iso-Butanol:
Iso-Butane 3.50(99.2) 3.50 7.1 18.8
Iso-Butyl ol:
0.01 (0.8)
1-Pentanol:
0.94(10.4)
2-Pentanol:
1.60 (17.5)
3-Pentanol:
Pentaneb 2.70 (30.1) 13.10 25.5 20.9
Pentanal:
1.30 (13.8)
2-Pentanone:
1.30(13.7)
3-Pentanone:
1.30 (14.5)
Pentanoic Acid:
Trace
a: 1.6x 10 mole ci-[Ru(dmp),(H20)](CF3SO3)2, 1.6x 10 mole CuC 2 H,0,
60ml acetonitrile, 5.0x10"2 moles 35% H202, 40psi substrate (1.9xl0" moles),
15 hours at 75C.
b: 2ml pentane (2.0xl10" moles) used.
c: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.







calculations. No peroxide was found in the resulting solution. These experiments

confirm the role of CuC12 in producing more alcohol in the oxidation of alkanes.

In the absence of the catalyst and CuCIl, a mixture of H202 (5.0x02 moles) and

ethanol (1.9x102 moles) in acetonitrile (60 ml) leads to a 21% decrease of ethanol after

15 hours. Hydrogen peroxide an alcohol under these type of reactive conditions. When

the reaction is carried out with CuCI (1.6x10' moles) added, the hydrogen peroxide

(5.0xl0-2 moles) in acetonitrile (60 ml) oxidizes only 9% of the ethanol to acetaldehyde

and acetic acid after 15 hours. This experiment demonstrates CuCI2 either complexes the

alcohol preventing oxidation or increases the rate of decomposition of the peroxide

preventing oxidation of the products.

The effect of increased amounts of CuC12 on selectivity to the alcohol was

investigated using the oxidation of propane at 750C. The results of these experiments are

given in Table 3-5 and illustrated in Figure 3-3.

Successive additions of CuC12 increased selectivity towards the alcohol with the

accompanying decrease in selectivity to the aldehyde and ketone. With increased

selectivity, a decrease in overall conversion is observed. This is most likely attributed to

peroxide decomposition by the increased copper concentration present. Without CuCI2 a

20.80% conversion is achieved. Addition of 1 mole equivalent decreases the conversion

decreases to 18.12%. Additional CuCI2 continues the trend reaching a 6.52% conversion

with 5 mole equivalents added. This decrease in conversion is the result of increased

peroxide decomposition, leading to a lower concentration of peroxide present to perform

catalysts and over-oxidize the products.


















Table 3-5: Oxidation Results for Propane @ 750C with cis-[Ru(dmp)2(H20)](CF3SO3)2
and Varying Mole Equivalents of CuCI2 using H202

Mole mmoles mmoles mmoles Total Percent
Equivalents Alcohol Aldehyde Ketone mmoles Conversionb
CuCl,"' ___Products

0 0.12 2.14 1.18 3.44 20.80


1 0.59 1.45 0.34 2.38 18.24


2 0.84 0.14 0.33 1.30 7.19


3 0.84 0.13 0.32 1.29 7.14


4 0.90 0.11 0.22 1.22 6.73


5 0.91 0.10 0.18 1.18 6.52

a: 1.6x0L mole cis-[Ru(dmp),(H0O)](CF3SO3)2, 1.6x10" x mole eq. moles CuCl, 2 H20, 60ml
acetonitrile, 5.0x102 moles 35% H202, 40psi substrate (1.9x l02 moles), 15 hours at 750C.
b: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.




















Effect of CuCl2 on Products with the
cis-[Ru(dmp)2(H20)J2] Precursor


-
14.00

0
11.00
g.


0 1 2 3 4 5
Mole Equivalents CuCIl
---Alcohol ---Aldehyde --Ketone -A-Total Moles -- Percent Conversion (2nd)


Figure 3-3: Oxidation of Propane with cis-[Ru(dmp),(H20)](CF3SO3)2 and Varying Mole
Equivalents of CuClz using H202







Effect of Temperature

Increased temperature is expected to increase the rate on an oxidation reaction.

Increase in substrate reactivity and solubility of the catalyst is also expected with an

increase in reaction temperature. Disadvantages include an increased rate of peroxide

decomposition. A series of experiments was performed at 25, 50, 75 and 1000C with and

without CuCI2 were performed to determine the effect of temperature on this oxidation

reaction. The results provided in Tables 3-6 and 3-7, are illustrated in Figures 3-4 and

3-5.

Propane was again selected as the substrate for these experiments. A mixture of

catalyst (1.6x 10 moles), acetonitrile (60 ml), H02, (5.0xl0" moles) and propane

(1.9xl0"2) were allowed to react for 15 hours at each temperature. Even at 250C the

ruthenium produces oxygenates from propane with a 3.71% conversion. The moles of

aldehyde and ketone produced are greater than the moles of alcohol generated. As the

temperature increases, overall conversion increases along with an increase in the relative

amounts of the aldehyde and ketone. The over-oxidation of alcohol present is also more

rapid with the catalyst as the temperature is raised.

An identical series of experiments were performed with the addition of 1 mole

equivalent of CuC12 (1.6x10" mole). The result are given in Table 3-7 and graphically

represented in Figure 3-5. The conversions obtained at high temperatures with CuCI,

present are decreased when compared to those in the absence of CuC12. Again, a higher

selectivity to the alcohol is obtained.























Table 3-6: Oxidation Results for Propane @ Varying Temperatures with cis-
[Ru(dmp),(H20)](CF3SO3)2 using HAO2

Total
Temp. Mmoles mmoles mmoles mmoles Percent
(C)' Alcohol Aldehyde Ketone Products Conversionb


25 0.14 0.28 0.26 0.67 3.71


50 0.13 0.55 0.40 1.08 5.52


75 0.12 1.58 0.49 2.19 12.09


100 0.10 2.04 0.89 3.03 16.72

a: 1.6xl0' mole cis-[Ru(dmp),(H1O)](CFSO3),, 60ml acetonitrile, 5.0xl0" moles 35% HO22,
40psi substrate (1.9x10"2 moles), 15 hours.
b: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.






74


















Temperature Dependence of cis-[Ru(dmp)2(H20)2]2+


18.00


15.00


12.00 .


9.00 0


6.00


3.00


U.UU II 0.00
0 25 50 75 100
Temperature (Degrees Celcius)

---Alcohol ----Aldehyde -Ar-Ketone --Total Moles --- Percent Conversion (2nd)






Figure 3-4: Oxidation of Propane @ Varying Temperatures with cis-
[Ru(dmp),(HzO)](CF3SO3)2 using H202


















Table 3-7: Oxidation Results for Propane @ Varying Temperatures with cis-
[Ru(dmp),(H20)](CFSO), and CuC1, using H202

Total
Temp. Mmoles mmoles mmoles mmoles Percent
(C) Alcohol Aldehyde Ketone Products Conversionb


25 0.23 0.28 0.40 0.91 5.03


50 0.50 0.99 0.43 1.93 9.89


75 0.58 1.45 0.54 2.58 14.23


100 0.66 1.70 0.67 3.03 16.72

a: 1.6x104 mole cis-[Ru(dmp)2(H,0)](CFSO3)2, 1.6x10' mole CuCl 2 HO, 60ml acetonitrile,
5.0x10' moles 35% HzO2, 40psi substrate (1.9x10-2 moles), 15 hours.
b: Percent conversion is based on the total of oxidized products divided by the initial amount
of substrate.























Temperature Dependence of
cis-[Ru(dmp)2(H20)]2* with CuCl2 (1 mol eq)


0 25 50 75 100
Temperature (Degrees Celcius)
---Alcohol ---Aldehyde ---Ketone ---Total Moles -- Percent Conversion (2nd)


Figure 3-5: Oxidation of Propane @ Varying Temperatures with cis-
[Ru(dmp)2(H20)](CF3SO3), and CuC12 using H202








Conclusions


The oxidizing ability of the cis-[Ru(dmp)2(H20)21(CF3SO3)2 complex to activate

linear and branched saturated hydrocarbons has been investigated. Using hydrogen

peroxide as the oxidant, production of the respective alcohols and their over-oxidized

products are seen after 15 hours at 750C in an acetonitrile solution. An example of the

oxidizing power of this system is shown by the 20.1% conversion of ethane to ethanol

(2.3% selectivity, 0.08 mmoles), acetaldehyde (69.3% selectivity, 2.5 mmoles) and acetic

acid (28.4% selectivity, 1.0 mmoles).

To deter production of the over-oxidized products and increase the selectivity of

the hydroxylated products addition of one mole equivalent of CuC12 was performed. The

addition of CuCIl, illustrates a pronounced effect towards increasing the selectivity to the

alcohols. Successive additions of CuCl2 mole equivalents also exhibit a further increase

in alcohol selectivity. Blank experiments performed also illustrate the ability of CuC1 to

retard further oxidation of the alcohol. Additional blank experiments performed in the

absence of catalyst, produce small quantities of oxidized products demonstrating the

catalyst is responsible for the majority of the oxidation.

Temperature dependence studies were also performed with and without the

addition of CuCl. The effect observed is the ability of this catalyst to produce

oxygenates of propane at 250C (3.2% conversion) to 1000C (16.7% conversion). The

addition of CuCI2 to the oxidations performed at variable temperatures also allowed for

increased selectivity of the alcohol.













CHAPTER 4
SYNTHESIS AND CHARACTERIZATION OF IRON DIMETHYL
PHENANTHROLINE COMPLEXES

Introduction


As previously discussed, a high valent metal-oxo complex is required for the

activation of alkanes. The cis-[Ru(dmp)2(H20)2]2+ precursor is converted to cis-

[Ru(0)2)2+to attain this high oxidation state moiety.40,74 This oxidant effectively oxidizes

alkanes with hydrogen peroxide, and methane with hydrogen peroxide and dioxygen.73

The success with this complex prompted research directed towards the synthesis of

analogous complexes in an attempt to prepare a more robust catalyst. This complex

should be able to utilize dioxygen without the need of a co-oxidant and be less expensive.

First row transition metals might produce a complex with these advantages, therefore the

synthesis of an iron based catalyst was investigated.

Iron containing complexes including heme and non-heme enzymes are reported to

effectively oxidize alkenes and alkanes.81,82,83 The development of catalytic systems to

mimic a variety of biological systems has been the subject of intense investigation.

Cytochrome P-450, peroxidases, catalysases and high valent iron-oxo porphyrin

complexes, all involve a 2-electron oxidation producing the reactive intermediate in

hydrocarbon oxidation reactions.69,84








The use of iron porphyrins as model catalysts has allowed for further

understanding of significant steps, which are involved in many enzymatic oxidation

reaction mechanisms.85 More recently, Watanabe and Morishima82 have demonstrated

the use of a (TMP)Fem(RCO2), where TMP is 5,10,15,20-tetramesityl porphyrin, complex

for the epoxidation of norborylene and a-methylstyrene at -780C with a variety of

peracids. This research has suggested the O=Fe"(TMP) 7-cation radical as being the

active oxidant for this oxygenation reaction.

Catalysts based on iron complexes in the absence of porphyrins have also been

investigated. Que83 has demonstrated the reactivity of (j-oxo) di-ferric complexes with

t-BuOOH for the activation of cyclic hydrocarbons in acetonitrile. The p-oxo di-ferric

complex was synthesized as an effort to model the dinuclear sites found in non-heme

iron-enzymes. This [Fe(TPA)20(OAc)](C103)3, where TPA is tris-(2-pyridyl)-methyl

amine and OAc is acetate, complex is demonstrated to be a robust catalyst for the

activation of cyclohexane at ambient temperature and pressure. The oxidation products

produced from this reaction are cyclohexanol, cyclohexanone and (t-butylperoxy)-

cyclohexane. Que has suggested the formation of the alcohol and ketone occur by

catalyst initiated decomposition of t-BuOOH to afford a high valent metal-oxo complex

via a heterolytic pathway. Additional research presented by Que72 has been directed

towards the further investigation of these non-heme iron centers. Furthermore, Que has

proposed a high valent Fe2(P-O)2 diamond core structure, which is believed to be the key

oxidizing species of methane monooxygenase (MMO).








Further research using iron coordinated ligand type complexes have also been

reported. Sawyers' has reported a number of iron (II) complexes [FenL.); Fe(DPAH)2,

where DPAH2 is 2,6-dicarboxylpyridine, Fe"(PA),, where PA is picolinic acid, and

Fe"(bpy)2', where bpy is 2,2'-bipyridine. Each of these complexes with the addition of a

reductant, [DH2:PhNHNHPh] for example, are able to catalytically activate 02 (1 atm) for

the hydroxylation of phenol and substituted phenol. Results provided indicate the

mechanism of oxidation is proceeding via a Fenton like intermediate.

The synthesis of an iron "analogue" based on the cis-[Ru(dmp)2(H20)2](CF3SO3)2

complex is the focus of our research. By altering the metal center it is anticipated a more

robust catalyst will result. The ligand was not altered due to its ability to impart a cis

geometry40, with respect to the two remaining coordination sites, when the metal is

coordinated by two dmp ligands. This geometry, the result of steric hindrance created by

the methyl groups a to the nitrogen atoms results in the cis isomer being a more powerful

oxidant when compared to the trans counterpart.74 The lesser oxidizing power of the

trans isomer, measured to be on the order of 35 kcal mol', 86 is attributed to the lower

energy of the HOMO d,.40 Additional benefits of this ligand include its ability to

prevent the condensation of hydroxo or oxo bridged dinuclear species when two ligands

are present on the metal center.

Based on accounts reported in literature and the success of the cis-

[Ru(dmp)2(H2O),]2 complex, the synthesis of the iron analogues was attempted.

Characterization of these newly formed complexes, as well as investigation for the







production of a high valent iron-oxo species, for the eventual use as an oxidation catalyst,

was also performed.

Experimental


Materials and Methods

FeCI2 4 HO0, 2-9-dimethyl-l,10-phenanthroline, NaCF3SO3 and LiCI were all

used as received from Aldrich. Acetonitrile, acetone and 30% hydrogen peroxide (aq)

were all used as received from Fisher Scientific (ACS Grade). Acetonitrile was distilled

over P20, and stored over 4A molecular sieves.

Physical Measurements

UV-Vis measurements employed a Perkin Elmer lambda-6 spectrophotometer, all

spectra were background corrected. Infrared Spectroscopy (IR) analysis were all

collected either as Nujol mulls or KBr pellets on a Nicolet 5DXB spectrometer and were

background corrected. Nuclear Magnetic Resonance (NMR) measurements were

recorder on a multinuclear Varian VXR 300 MHz spectrometer or a multinuclear Gemini

300 MHz spectrometer. All samples analyzed by NMR were performed in deutrated

solvents with 1% w/w TMS. Electrochemical studies were performed with a PAR model

175 Universal Programmer connected to a model 173 potentiostat/galvanostat. The

electrodes used were a platinum working and auxiliary electrode, with an Ag/AgCI

reference electrode. The pH measurements were made with a Fisher Accumet model 630

pH meter. FAB (+ and -) mass spectral data were obtained by Dr. David Powell (U.F.) in

a m-nitrobenzyl alcohol matrix. Single crystal X-ray diffraction analysis were performed

by Dr. Kalih Aboud using a Nicolet diffractometer equipped with a graphite-








monochromated Mo-Ka radiation source. The Nicolet Structure Determination Package

was used for data collection, data recovery and structure elucidation. U.F. Analytical

Services performed elemental analysis.

Synthesis of Compounds

Iron(II)bis(chloride)mono(2,9-dimethyl-1,10-phenanthroline), [Fe(dmp)Cl2]

(I). A 2.25 g (12 mmol) of 2,9-dimethyl-l,10-phenanthroline was dissolved in 100 ml of

acetonitrile under nitrogen at 700C. Next, 1.25 g (6 mmol) FeClI 4H20 was added to

the solution. Upon addition of the metal a precipitate is formed immediately. The

resulting mixture is allowed to react for 30 minutes, the solution cooled and filtered. The

product is then dried under vacuum at 600C overnight. Analysis: Calculated for

C,4H,2N2C12Fe: C, 50.29; H, 3.59; N, 8.38. Found: C, 50.12; H, 3.46; N, 8.23.

Cis-Iron(II)bis(aquo)bis(2,9-dimethyl-1,10-phenanthroline)bis(trifluoro-

methanesulfonate), cis-[Fe(dmp),(HIO):(CF3SO3)2 (II). This complex was

synthesized using modification of a prior method.86 A 4.5g (24 mmol) of 2,9-dimethyl-

1,10-phenanthroline was added to 150 ml of deionized H20 at 900C and allowed to stir

vigorously for 20 minutes under nitrogen. Upon complete dissolution of the ligand, 2.5g

(12 mmol) of FeC12 4 H20 was added, the resulting solution was then stirred for 2 hours.

The solution is then filtered (hot) and the filtrate immediately added dropwise to a chilled

saturated NaCF3SO3 (aq) solution (25ml H20 / 9g NaCF3SO3). The resulting precipitate

is allowed to stand in ice for 2 hours and filtered. The product is dried under vacuum at

600C overnight. Analysis: Calculated for C301HN40,F6S,Fe: C, 44.66; H, 3.47; N, 6.94.

Found: C, 44.87; H, 3.33; N, 7.01.








Results and Discussion


Characterization

A number of analytical techniques were utilized to characterize the complexes (I

and II) synthesized in this chapter. The conditions employed for each spectroscopic

technique are described within the experimental section of this chapter. Any variation in

analysis will be described in the appropriate section. Elemental values obtained are

provided at the end of each synthesis procedure.

Single Crystal X-ray Diffraction

Single crystal X-ray diffraction analysis was collected for the mono-2,9-dimethyl-

1,10-phenanthroline (dmp) complex, Fe(dmp)Cl2 (I) to provide structural information and

confirm results obtained from elemental analysis. A brown-orange needle of I (0.26 x

0.19 x 0.16mm) was mounted on the end of a glass capillary tube and analysis performed.

The crystal structure and crystal data obtained are provided in Figure 4-1 and Table 4-1.

Additional information for this structure is given in Tables 4-2, 4-5 and 4-6.

In this structure iron has a four coordinate tetrahedral geometry with two

coordination sites occupied by the nitrogen atoms of the dmp ligand and the remaining

bound to the chlorine atoms leading to a 2* oxidation state. The coordination of only one

dmp ligand leads to the possibility of u-oxo dimer formation when this complex is

oxidized in the presence of others.
















































Figure 4-1: Crystal Structure for Fe(dmp)C1,.








Table 4-1: Crystal Data and Structure Refinement for Fe(dmp)Cl2.


Empirical formula

Formula weight

Temperature

Wavelength

Crystal system

Space group

Unit cell dimensions



Volume, Z

Density (calculated)

Absorption coefficient

F(000)

Crystal size

Theta range for data collection

Limiting indices

Reflections collected

Independent reflections

Absorption correction

Max. and min. transmission


C14H12CI2FeN2

335.01

173(2) K

0.71073 A

Orthorhombic

Pnma

a =11.2265(7) A a =90*
B = 7.4630(5) A p = 90"
C= 17.788(1)A y =90"

1490.3(2) A3, 4

1.493 Mg/m3

1.356 mm-1

680

0.26 x 0.19 x 0.16 mm

2.15 to 27.50 deg.

-5<=h<=15, -4<=k<=10, -24<=1<=18

5761

1841 [R(int) = 0.0451]

Semi-empirical from psi-scans

0.9660 and 0.7631




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ES8G3KDW7_6E7VZ6 INGEST_TIME 2013-01-23T16:07:33Z PACKAGE AA00012951_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES



PAGE 1

LOW-TEMPERATURE HOMOGENEOUS OXIDATION OF ALKANES USING HYDROGEN PEROXIDE By MICHAEL A. GONZALEZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1998

PAGE 2

To my dear wife Stacy Michelle for all your support and belief in me. You complete me

PAGE 3

ACKNOWLEDGMENTS No one has ever said life way easy but it has been said that it can be extremely rewarding. My decision to attend college and graduate school has made my life very fulfilling and I have only lived less than one-third of it Although college and graduate school has been filled with late night study sessions overnighters at the lab and numerous classes (as well as migraines) I would do it all again if given the opportunity. My undergraduate freshman year at the University of Texas El Paso (UTEP ) was a time of indecisiveness and uncertainty about my future career aspirations In my sophomore year all these worrie s had changed for the better this was the result of receiving a scholarship and being given the opportunity to conduct undergraduate research in the area of inorganic structural and polymeric chemistry. For this I would like to thank Dr Phillip Goodell Dr L W ter Haar and the Research Careers for Minority Students (RCMS ) committee for the opportunity to allow me to demonstrate my academic and research wares. My adventures at UTEP would not have been possible without the aid of Tony 'Hubcap '' Sanche z Manuel '' Manny '' Garbalena Anthony '' Skate '' Guarnera Joseph 'Little Joe '' Janisheck and Gerardo ' Ardo '' Fuentes. To my friends I say thanks for the memories and the friendship ; you will never be forgotten 111

PAGE 4

From Texa s my quest for higher learning landed me in sunny Florida where graduate life began at the University of Florida. I cho s e to work for Dr Russell '' Doc '' Drago the second best d eci s ion I have made in my life Doc is the reason I remain in chemistry ; he i s the most intelligent person I know and his fervor for chemistry is undaunting I can only hope to achieve this level in my chemistry career. ' Doc thank you for all your patience and instilling a portion of your s cientific knowledge in me. '' I would also like to offer thank s to Ruth Drago she has made my four-year stay memorable I would like to thank her and Doc for taking Stacy Mikey and me into their 'scientific' as well as personal family and making us feel like family. This in turn made our transition to Florida effortless My graduate school life was also made memorable by the friendships I formed with members of the Drago group both past and pre s ent To the past members Mike '' Bevis' Robbins Todd '' I can fix it ' Lafrenz Phil Kaufman, Garth 'Ogre '' Dahlen C hris Chronister Don Ferris ' Bueller '' and Mike Naughton I offer my thanks for s howing me how to survive in graduate school on both an academic and personal level The curr e nt Drago group members especially the oxidation sub group are the greatest. Ken '' Kato '' Lo, has been a great friend and always listened to me plus I will never forget our Friday lunches. Alfredo '' Alf Mateus was a great help with understanding heterogeneou s oxidation chemistry. Without Ben '' Hubcap '' Gordon I would not have completed the last three months of research Cheng '' Tiger '' Xu helped me understand the iron-oxo sy s tem I al s o want to thank Steve '' Jorg '' Joerg for bailing me out of trouble answering my philosophical and ethical questions and running my NMR s. I would also like to thank Krystoff Jurzyk and Nick ' Com on the '' Koh for insight on high temperature lV

PAGE 5

heterogeneous chemistry. I would also like to say thanks to Maribel and Diane for their help in office matters. The biggest thanks go to Stacy, my best decision, for standing by me, believing in me and giving me the best things in life a person could ask for: love and a child I thank her for putting up with me on late nights of writing research summaries, being at the lab and tutoring. This dissertation is as much hers as it is mine. I thank my parents for their love and support. Dad always told me to set high goals and never settle for second best. And mother was always there when I needed an ear, and for this I say thanks. No one knows what the future holds, but with my wife, family and friends, I know I can approach it and control my own destiny V

PAGE 6

TABLE OF CONTENTS page ACKNOWLEDGMENTS ........ ... .................. . ... ... ... ................. ... . .... .... .. ............... .. i ii LIST OF TABLES .. .. ...... ... .. . .. .. . . .. ...... ..... .......... ... ................... ........ .................. .. .. ix LIST OF FIGURES ......... .. ....... ... ... .. .... . .. ..... .. . ..... .... ... . ....... .... .. .. . . . ... .... . .. .. .. xii ABSTRACT .. . .... ... .. ............ ..... . . .. . .. .. .. .. .. ..... .... .. ... .. . ...... ... .................. .. ... ........ xiv CHAPTERS 1 GENERAL INTRODUCTION TO ACTIVATION OF ALKANES .... ... .. .. ............ 1 Importance of Homogeneous Catalysis ....... .. .... .... ... . ............ ... ..... .. ............ .. .... .... .... 1 General Classification of Homogeneous Oxidation Reactions .. .. . . ... ............... ..... .. . 3 General Classification of Alkane Activation Reactions ................................................ 8 Examples of Alkane Activation by Homogeneous Catalysts . .... .... ... .. .... ................. 9 Designing a Homogeneous Catalyst ...... ........ ......... .... ... ...... ... ... ... ... ... .. ............... .. 12 Five Classes of Metal Oxo Reactions . .. ..... ................ .... ..... ................... .. .... .. .... 13 Homolytic or Heterolytic Pathway? .................................. .. .. ......... .... .. . . .......... 19 Effect of Ligand on Metal Center .. .... ....... .. ..... .. ............. .. ....... .. .. . .. . .. .. ......... ... 21 H 2 0 2 as an Oxidant ................................................. ... ..... ... ...... . ... . .. ... . .......... 23 2 HOMOGENEOUS CATALYZED PARTIAL OXIDATION OF METHANE WITH HYDROGEN PEROXIDE AND OXYGEN .... .......... ... .. ... ... .. .. ... ..... .. .. ....... 27 Introduction .... ... .. .. .. .. ...... ... .... ... . ... ..... .. .............. . ... ... ..... .. ... .. . .. ..... ...... .. ... ... .... . . .. ............. .. 27 Experimental ........................................................................................ .. . .... ......... ...... ..................... ... ... 3 0 Materials an.d Methods .. .. ..... .... . ... ...... .... .. .... .. .. .............. .................... .... ... ... .. .......... .. ...... .. . 30 Physical Meas1.1Temen ts . .. .. ..... .. .. .. ....... .. . .... .... . ... ..... . .. .... ..... .. ...... .... . .. .... ..... ........................ 3 3 Synthesis of Compounds . .. ........ .. ........................................ ................ .. .. ... ..... ........... ....... ...... ... 33 Oxidation Procedl.lfe ... .. .... .... .. . .. ....... .. .... .. .... .. .... .. ........... ....................................... ... .. ........ .... .... ................ ........ .. 34 Results and Discussion ........................................................................................ .. .......... .. ... .... . .. ..... ..... ... .... .... 3 7 Anion Modification ...................................................................................................................................... 3 7 Ruthenium Catalyzed Oxidation of Methane with Hydrogen Peroxide ... ..... .... . .. 40 Peracid Formation and Reactivity ...... .. ........... .......... ............. .... .... .. ... ......... .. .... 44 Oxidation of Methane with Molecular Oxygen . ............. . .. ... .................. . .. .. .. .. 48 Conclusions ..... .... ................... ... ........................ .. ........ ... .. .... ............. ... ......... .. .. ............ .. ..... ........ .. .... .. ..... ... .. ....... .. 51 Vl

PAGE 7

3 4 5 OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING A R UT HENIUM MET AL-OXO CATALYST .... .... ... ............ .. ... . .. . ... .. .................... 53 Introduction . .. .... .. .... .. .. .... .. .. .. ... .... .. ........ .. ... .... .. ...... ...... .. ........ .. .. .. .. .. .. .. .... .. .... .. .... .... .... ...... .... .. ........ .. .. .. .... .. .. .. .. .... .. .. .. ........... .. .. .. .. .......................... 53 Experimental .......... ... .... ... ..... ...... .... .. . . ........... .... .. .......... ................................... .... 5 5 Materials an.d Methods .. .. .......... . ... .......... .............. ... ... ........ . ......... .. ... .. . . ..... 5 5 Physical Measurements ..... .................................................. ... .. ... .. ... ....... .. .. .. ..... . 5 5 Synthesis of Compounds . ... .. .. . .. . .. . ... .. . ..... .. .. .. .. ... . . . ...... . . . .. .. ... .. . .... ......... 5 5 Oxidation Procedure .................................................................................................................................... 56 Results and Discussion ........... .. .. .. .... ... ..... ... .... .. ......... ..... .. .. .. ..... . .. .. .... .. .... ..... .. ............... .. ............ ..... .. .............. .58 Alkan.e Oxidation ... ... .. .............. .. . ................ ............... .. ........ ........................................... .. ......................................... .58 Mechanism of Oxidation. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . .. .. .. .. .. .. .. .. .. .. . .. .. .. .. .. .62 Addition of CuC1 2 .... ........ .. ................ ... .......... .... ....... .. .... .... .................. .... .... .. .... .... .. .. .... .. ...... .. .. ........ .. .................. .... .. .. .. .. .. ..... .. .. .. 64 Effect of Temperature .... .. .... .. ... .. .... .. .... .. .. .. .. .. .. .... .. .. .. .... .... .......................... .... .. .... .... .. .... ....... ......... .. .. .. .. ..... .. .... ................. .. ........... .. 72 Conclusions ................................................................................................................................................. .. ...... ... ............ .. .... ... ..... ......... .... 77 SYNTHESIS AND CHARACTERIZATION OF IRON DIMETHYL PHENANTHROLINE COMPLEXES .... ......... ..... ............... .. .... .. .... .. ........ .... .................. ...................................................... .78 Introduction .. .. .. .. .... .. ... .. ..... .... ........ .. ....... .. .. .... .......... .. .................................... .. ................. .. .............. ... . .............. .. ........... ....... ................ .. 78 Experimental .... . . ... .. . .. ... ................ .. ...... .... .... .. .. .. .... .. ...... ...... .... .. ........ .......... .. .. ...................... .. .. .. .. .... .. .. .. ........ ............... .. .... ................. .. .... .. 81 Materials and Methods. ............................................................................................................................................. .. ............ ...... .. 81 Physical Measurements ... .............. .. ................ .... .. .. .. .. .... ...... ...... .. ........ .... ... .. ........................... .. ........ .... ... ...... ........... . .. .... .. ................ 81 Synthesis of Compounds .. .. ... .. .... .. ... .... .. ...... .. ............ .. .. ........... .... .. .. .. .................................................................... ...... .... .. .... 82 Results an.d Di s cussion .. ........ .... ... .. .. .......... .... .. .. .... .. .. .... . .. .. ...... .. .... ........ ................ ... .. .. .... .. .... .. .... .. ............ .. ......... ....... .... .. ....... .. ... .. .. .... ....... 83 Characterization . ............... .. .............. .... ........ ... ....... .. ...... .. ..... .. ....... .... ....... .. .............................. .. ........................ .. ...... ..... .... ....... .... .. .. .. 83 Single Crystal X-ray Diffraction ... . . .................... ... .. .. . .............. ...... .. ..... .. 83 F AB Analys i s .. .. ... ............. .. ........................... .... ... ................ .... ...................... .. .... ........ .. .... ..... .. ... .............. .. ................ 94 IR Analysis ........................................................................................................................................... .. ...................... .. .. 94 NMR Analysis .. .. .. . .. .. .. .. .. .. .. .. .. . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . .. . . . .. . .. .. .. .. . .. .. . .. .. .. .. .. .. .. .. . .. . .. .. . .. .. . .. .. .. .97 High Valent Iron-Oxo Formation Studies . . .. .. . .. .. . .. .. .. .. .. .. .. .. .. .. .. .. . .. .. .. . .. .. .. .. .. .. .. . .. .. .. .. 1 0 3 Conclusions ........................................................................................................................... .. ............................................ .. ......... .... .. 106 OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING AN IRON MET AL-OXO CATALYST ... ... .. .. ...... .. .. ..... .. .... .. .... .... .. .... .... .... ........... .. ..... .. .................................................. .. 109 Introduction .. ............... ......... .... .. .. .. ....... ......... .. .. ....... .. .. .. .. .. ..... .. .... .. .... .. .... .. .... .......... ... .. ...................................................... ... .... .... I 09 Experimental ...................................................................................................................................................................... 110 Materials and Methods .. .. .... ... .. ....... ...... .. ..... .... .... .. .. ... ..... .. .. ... .. .. ..... .. .... .. ... .. .. ..... ...... .. .. ... .......... ... .......................... 110 Physical Measurements .. .. ... ......... .. ...................... .. ..... .. ................................. ........ ..... ................................ ........ ... 111 Synthesis of Compo'Ullds ...... ........ ....... .. ................................... .. ............ ........ ....... .. ... ... .... ........... .. .................... 111 Oxidation Procedure .. ....... .. ......... .. .. .... ............. ............... .. ........... .. ...................... ...... ...... ................ .......... .. ..... .. 112 Oxidation of Methan.e ................ .. .... ..... .. ..... .... .. .... ....... ... .. .. .. .... ..... .... .... .. .... .. .. ......................................... 112 Oxidation of Higher Alkanes ............................... .. ....... .. .. .. .... . .. . ..... .. ..... ...... .... ............ ... .. ... .. .............. .. Results an.d Discussion ....... .... ........ . ................................................................................ . ................. .. ... ... ..... ... ........... ........... .... ... .. .... ..... .... ... .......... .. .... . ......................... .. ................ ..... .. .... .. .. ..... Oxidation of Methan.e with H 2 0 2 Mechanism for Oxidation of Methane .. .. .. .. ........ .. ..... .... ....................... .. .... .. .. ......................... .. ... ...................... 113 114 114 118 Oxidation of Methane with 0 2 ...... ....... .. .. ..... .... ...... ........ .. . .. .. ..... .. .. .... .. 118 Vll

PAGE 8

Alkane Oxidations with c is-[Fe( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 )2 ................................... Alkane Oxidation with [Fe( dmp )Cl 2 ] .. .. ..... . .. .. . .... ................................... .... Mechanism for Higher Alkane Oxidation ....... .. .................................... ..... ..... Addition of CuC1 2 E ffect of Temperature ......................... .. ..... ......... .. . .. .. .......... ..... .... ..... ..... ..... ... ... Conclusions .. .... .. .. ..... .. .. .......... .. . . .. .. . . .. .... . ... .. . .. .... .. .. . .... . ... ............. ... .. .... .. . 122 126 130 131 141 146 6 CONCLUSIONS ........................................................................................................ 148 GLOSSARY .......................................................................................................... ... ............ 150 REFERENCES ..................... .. ... .. .. ..... ........... .. .... .... .. ..... ...... .. ..... ... .. ... ...... .. .... .. . ..... .. .. ..... . .. .. ...... ..... .......... .. ............. 151 BIOG RA.PHI CAL SKETCH .. . .. .. .. .. .. . .. .. ... .... .. ............ .. .... .......... ...... .. ...... .. .... ... .. ... ............... .. ... .. ... ... .. ...... .. ..... 157 Vlll

PAGE 9

LIST OF TABLES Table page 1-1: Advantage s of a Homogeneous Catalysts .. ......... .. .................................................. 14 1-2: Disadvantage s of a Homogeneous Catalyst .. ..... ........ .. . .. ... ... .... .. .. ... .... ...... . .. .. .. 15 H 2 0 2 .... . .... .. .. .. . . .... .. .. .................................................................................................................................... 43 2 2: Oxidation Results for Methane @ 75 C with c is-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 Using H 2 0 2 and 0 2 . 49 3 1: Oxidation Re s ults for Ethane, Propane and Butane @ 75 C with cis[Ru( dmp ) 2 (H 2 0) ]( CF 3 S0 3 ) 2 using H 2 0 2 .. ......................... . .... ....... ... ... ..... ....... .. .... 59 3-2: Oxidation Re s ults for ]so-Butane and Pentane @ 75C with cis[Ru( dmp )2(H20)](CF3S03)2 using H 2 0 2 ... .. . .. .. ... ... .. ... .... . .. ........ ... .. ..... ........ ..... ... 60 3-3: Oxidation Results for Ethane, Propane and Butane @ 75 C with cis[Ru( dmp ) 2 (H 2 0)](CF 3 SQ 3 ) 2 and CuCl 2 using H 2 0 2 .............................. .. .... .............. 67 3-4: Oxidation Re s ults for ]so-Butane and Pentane @ 75C with [Ru(dmp) 2 ( H 2 0)](CF 3 S0 3 ) 2 and CuCl 2 using H 2 0 2 . ....... ... .... .. . .. . ....... ....... .. ..... 68 3-5: Oxidation Re s ults for Propane @ 75 C with cis-[Ru(dmp) 2 {H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 u s ing H 2 0 2 . . .. .. ...... ...................................... 70 3-6: Oxidation Results for Propane @ Varying Temperatures with cis[Ru(dmp) 2 (H 2 0)](CF 3 S0 3)2 using H 2 0 2 ............................... ...... .. . ... . . ... .. ............... 73 3 7: Oxidation Re s ults for Propane @ Varying Temperatures with cis[Ru(dmp) 2CH2 0)](CF 3 SQ 3)2 and CuCl 2 using H 2 0 2 .. .... .... . . .. ..... ...... ................... 75 4-1: Crystal Data and Structure Refmement for Fe(dmp)Cl 2 85 4-1: (Cont'd) .. ... ...... ........ .. ...... .............. .... .. .. . .. .... ........ .. . ... .. . ......... ... .. .. .. .. .. ... ..... .... ........... .. .. . 86 4-2 : Atomic Coordinates (x 10 4 ) and Equivalent Isotropic Displacement Parameters (A 2 x 10 3 ) for Fe( dmp )Cl 2 87 1X

PAGE 10

4-3 : Bond Lengths [A] for Fe(dmp)Cl 2 89 4-4: Bond Angles [ 0 ] for Fe( dm p )Cl 2 90 4-4: (Cont'd) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91 4-5 : Anisotropic Displacement Parameter s (A 2 x 10 3 ) for Fe( dmp )C l 2 . . .. 92 4-6: Hydrogen Coordinates (x 10 4 ) and Isotropic Displacement Parameters (A 2 x 10 3 ) for Fe(dmp)Cl 2 93 5-1: Oxidation Results for Methane @ 75C with cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 Using H 2 0 2 ............................................................................................................................ 115 H 2 0 2 and 0 2 .............................................................................................................. 1 l 9 5-3: Oxidation Results for Ethane, Propane and Butane @ 75C with cis[F e( dmp ) 2 (H 2 0) ](CF 3 S0 3 ) 2 using H 2 0 2 .... . .... ....................... .... ...... . .. ..... . . .. .... ..... 123 5-4: Oxidation Results for ]so-Butane and Pentane @ 75C with cis[Fe( dmp ) 2 (H 2 0)](CF 3 S0 3 ) 2 using H 2 0 2 124 5-5: Oxidation Results for Ethane, Propane and Butane @ 75C with Fe(dmp)Cl 2 using H 2 0 2 ..... .. .. ............ .................... ....................................................... ........................ 127 5-6: Oxidation Results for ]so-Butane and Pentane @ 75C with Fe(dmp)Cl 2 and H 2 0 2 128 5-7 : Oxidation Results for Ethane, Propane and Butane @ 75C with cis[F e( dmp ) 2 (H 2 0)](CF 3 S0 3 ) 2 and CuCl 2 using H 2 0 2 .... 132 5-8: Oxidation Re s ults for ]so-Butane and Pentane @ 75 C with cis[Fe( dmp )2(H20)](CF 3 S0 3)2 and CuC1 2 using H 2 0 2 .... ..... ......... ... ... ... ... . ....... .. ... 133 5-9: Oxidation Results for Ethane, Propane and Butane @ 75C with Fe(dmp)Cl 2 and CuC1 2 using H 2 0 2 13 5 5-10 : Oxidation Results for ]so-Butane and Pentane @ 75C with Fe(dmp)Cl 2 and CuC1 2 usin_g H 2 0 2 136 5-11: Oxidation Results for Propane @ 75 C with cis-[Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 .... ... ........ .... .. .......... ... .... .... ........ 13 7 5-12: Oxidation Results for Propane @ 75C with Fe(dmp)Cl 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 .... .... . .. ... .. . ...................... .. . .. ... .............. ..... ... .... 139 X

PAGE 11

5-13 : Oxidation Results for Propane @ Varying Temperatures with cis[Fe( dmp ) 2 (H 2 0)](CF 3 S0 3 ) 2 using H 2 0 2 142 5-14 : Oxidation Results for Propane @ Varying Temperatures with cis[Fe( dmp ) 2 (H 2 0)](CF 3 S0 3 ) 2 and CuCI 2 using H 2 0 2 143 5-15: Oxidation Results for Propane @ Varying Temperatures with Fe(dmp)Cl 2 using H 2 0 2 . .. .. .. .. . ... ... .. ........................ ............ ...... .......... .. .......... .. ...... .. ........ ...... ...... ............ .. .... .. .... ........ .. ........ .. .... .. ............ .. .. .... ........ ...................... ....... l 4 4 5-16: Oxidation Re s ults for Propane @ Varying Temperatures with Fe(dmp)Cl 2 and Cu C 1 2 using H 2 0 2 .. .. 14 5 X1

PAGE 12

LIST OF FIGURES Figure page 1-1: Catalytic Cycle for the Hydroformylation Reaction ... ..... ................ ......... .. .. .. ... .. 11 1-2: Desired Reaction Pathway for Substrate Oxidation ......... ....... ...... . ......... ... ......... 20 1-3: Detailed Reaction Pathways of Metal-Oxygen Species .. ...................... .. . ....... ...... 2 2 1-4 : Methods for Activating Hydrogen Peroxide . ....................... . .. ... .............. .. . .. . . .... 26 2-1: Structure for 2 9-dimethyl-1 10-phenanthroline .. ... ............... . ........................ .. ... . 27 2-2: Proposed Catalytic Cycle for Alkane Oxidation via cis-[Ru ( dmp ) 2 ( 0) 2 ] 2 + 29 2-3: Detailed Catalytic Cycle for Ruthenium Analogue .. .. ............ . .... ........ .... ..... .... ..... 31 2-4: Distribution of Worldwide Natural Gas Resources ... .. ............................ .. . .. .. .. .... 3 2 2-5: Diagram of Batch Type Hydrogenation Reactor .. .. . .. . ..... .... .................... .. ..... . .... 35 2-6: 1 H NMR Spectrum for cis-[Ru(dmp) 2 (H 2 0) 2 ](PF 6 ) 2 38 27: 1 H NMR Spectrwn for cis-[Ru( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 39 3-1: Proposed Hydrogen Atom Abstraction Mechanism for cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + Complex .. ........ . ...... .... ... .... .. .. ..... .... .......................... ....... .. . . .... .... . .... .. .. ..... 6 3 3-2: Proposed Oxygen Atom Insertion Mechanism for cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + Complex .... .. ... .... ........ . ..... ........ ......... ........ .. .. . ........ .... ..... ..... . .... ......... ...... 6 5 3-3: Oxidation of Propane with cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 71 3-4: Oxidation of Propane @ Varying Temperatures with cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 using H 2 0 2 7 4 3-5: Oxidation of Propane @ Varying Temperature s with cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and CuC1 2 using H 2 0 2 7 6 4-1: Crystal Structure for Fe( dmp )Cl 2 84 Xll

PAGE 13

4-2: F AB + Spectral Results for cis-[Fe( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 95 4-3: Infrared Spectrum for cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 SQ 3 ) 2 96 4-4: 1 H NMR Spectrum for Fe(dmp)Cl 2 98 4-5: 1 H NMR Spectrum for cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 SQ 3 ) 2 lOO 4-6: Crystal Structure for cis-Fe( dmp ) 2 (NCS) 2 101 4-7: UV-VIS Spectra for the Addition ofH 2 0 2 to cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 105 4-8: UV-VIS Spectra for the Addition of H 2 0 2 to Fe(dmp)Cl 2 107 5-1: Oxidation of Propane with cis-[Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 13 8 5-2: Oxidation of Propane with Fe(dmp)Cl 2 and Varying Mole Equivalents of CuCl 2 using H 2 0 2 140 Xlll

PAGE 14

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy LOW-TEMPERATURE HOMOGENEOUS OXIDATION OF ALKANES USING HYDROGEN PEROXIDE Chairman: Dr. Russell S. Drago Major Department: Chemistry By Michael A. Gonzalez May 1998 The activation of saturated hydrocarbons by metal complexes in the liquid phase became an active research area towards the end of the 1960. Stringent conditions such as the use of powerful oxidants, corrosive superacids, elevated pressures and high temperatures are used for the selective oxidation of low molecular weight paraffms. As products from these oxidations gain increased industrial importance, environmentally friendly and cost-efficient catalysts and processes are sought. Natural gas reserves, which are approximately 90% methane are abundant and a source of a valuable feedstock. The direct oxidation of methane to methanol provides the first step towards a route to an alternative large-scale transportation fuel replacing dwindling petroleum reserves. For this reason, the development of a direct one-step process for the oxidation of methane to methanol becomes of interest. Also of XlV

PAGE 15

importance is the ability to synthesize valuable organic compounds selectively (alcohols aldehydes ketones and carboxylic acids) from higher alkane feed stocks. Previous reports from this laboratory have demonstrated the use of the cis[Ru( dmp ) 2 (H 2 0) 2 ](PF 6 ) 2 precursor, where dmp is 2 9-dimethyl-1,10-phenanthroline, for the activation of methane in acetonitrile. The nitrogen-based dmp ligand not only increases the electrophillicity of the metal center but also imparts a steric hindrance about the metal forcing a cis geometry allowing the catalyst to be a more potent oxidant. Using the properties from this complex, a number of derivatives have been successfully synthesized and characterized. The synthesis and characterization of each newly synthesized catalyst: cis[Ru( dmp ) 2 (H 2 0 ) 2 ](CF 3 S0 3 ) 2 cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 SQ 3 ) 2 and [Fe(dmp)Cl 2 ] as well as the ability of each to hydroxylate C 2 -C 5 alkanes in acetonitrile and methane in a glacial acetic acid/acetic anhydride solvent mixture at 75 C and 40 psi are presented. Also investigated is the ability to modify and increase the selectivity to the alcohol product upon addition of a metal chloride, and the effect of temperature on each catalyst precursor. Preliminary mechanistic data suggesting a pathway for methane and higher alkane oxidation are also presented. xv

PAGE 16

CHAPTER 1 GENERAL INTRODUCTION TO ACTIVATION OF ALKANES Importance of Homogeneous Catalysis The end of the 1960s confronted the field of homogeneous catalysis with a potential problem. This obstacle was the inability for a soluble transition-metal complex to effectively activate a saturated hydrocarbon in the liquid phase. 1 As large resources of natural gas were discovered, the need for improvements to the current homogeneous catalytic systems became apparent, as well as the demand for alternative catalysts. One solution to this issue was the development of new classes of metal complexes, which were capable of undergoing an oxidative addition across the C-H bond of a paraffm (Equation 1-1 ). The application of this catalyst would allow for alkanes to be utilized as feedstocks in a number of industrial chemical reactions. This oxidative addition process produces the desired oxidized products in selectivity ratios, which are more favorable than those achieved using typical free radical reagents. 2 3 4 (Equation 1-1) M + R-H R-M-H As advances in coordination chemistry and catalysis were developed, reports of activating hydrogen olefins, aromatics, carbon monoxide, and molecular nitrogen with soluble transition-metal complexes became commonplace. 1 The ter1n activation is used 1

PAGE 17

2 to imply "the molecule or its part becoming a ligand in the coordination sphere of the complex and as a result undergoing a subsequent chemical transfonnation"(p. 2) 1 The primary deterrent one is confronted with in the activation of alkanes is the chemical inertness these molecules possess. Alkanes are benign to the point they can be safely used as a non-interacting solvent. Therefore any such activation of an alkane would require notably stringent conditions. These conditions typically require active particles, which include strong oxidants, superacids free atoms, radicals carbenes high temperatures ( 500-1000C) and other sources of energy (i.e. radiation chemistry) As environmental and monetary constraints become more significant a more desirable oxidation technology needed to be sought. 2 3 4 Interest in the activation of methane, a primary component in natural gas al s o heightened prompting the development of a new catalyst that was able to impart selectivity in its oxidized product under relative!):' mild conditions The typically inert methane molecule along with its alkane counterparts required designing a catalyst that was able to effectively activate these substrates which did not contain double or triple bonds or lone electron pairs and possessing relatively strong covalent C-H and C-C cr-bonds. 1 Shilov 1 had presented evidence for the activation of alkanes ; he expressed that if other more selective reactions of alkanes under "comparatively mild conditions" exi s t there is an inherent ability to oxidize the alkanes in question. The s tatements that follow provide a basis for his proclamation:

PAGE 18

. 3 Nwnerous examples of homogeneous H 2 activation are reported. 2 One example is the H-D exchange for hydrogenation. Yet, the cr bond in molecular hydrogen is not weaker than the cr C-H bond in alkanes. Many metal complexes CMLn or M) are capable of reacting with substrates containing "activated" C-H bonds. 2 One example is the C-H bond in aromatics or in the a-position to double bonds. As the C-H bond is broken formation of a M-C bond results However, the C-H bond energy which is greater in aromatics than alkanes does not hinder the reaction. Also demonstrated is the involvement of a "nonactivated" aliphatic C-H bond within the coordination sphere. This occurs when a suitable position on the ligand of the complex becomes available. 2 This results in a process defined as ''cyclometallation" and provides evidence for the possibility of a reaction occurring with alkanes possibly at elevated temperatures. Hydroxylation of C-H bonds in saturated hydrocarbons catalyzed by metal enzymes is known. 2 One published example is the oxidation of methane to methanol (primarily) by methane monooxygenase. General Classification of Homogeneous Oxidation Reactions This research is directed towards developing a series of homogeneous catalysts able to activate saturated hydrocarbons, olefins and aromatics with mild oxidants. In order to do so, an understanding of the fundamental chemical reactions must be understood. These liquid phase (homogeneous) transition metal catalyzed oxidations can be categorized into three areas. 5 These categories include the following:

PAGE 19

4 1. Free Radical Autoxidation The Mid-Century / Amoco process for the conversion of p-xylene to terepthalic acid is one example of this category This reaction is exhibited in Equation 1-2. (Equation 1-2) Co(OAc)2 NaBr 200c In this process air (20psi) is utilized as the oxidant, with a bromide-promoted cobalt salt as the catalyst in an acetic acid solvent The primary oxidation s tep i e. p-xylene to p-toluic acid occurs quite readily in the presence of a small amount of the cobalt salt. The secondary oxidation step, fonnation of the di-acid, necessitates the presence of higher concentrations of the catalyst and/or addition of the bromide promoter. 2 Oxidations of alkyl aromatics involving a cobalt (III) catalyst appear to proceed almost exclusively via an electron transfer mechanism 1 The reaction of the alkyl aromatic with Co(III) results in formation of [ArCH 3 ] + and a Co(III) radical. Loss of a proton results in producing ArCH 2 and H + In the presence of air or oxygen a peroxy species is formed producing ArCH 2 0 2 which then reacts to produce the nortnal oxidation product s The Co(II) catalyst can be regenerated either by reacting with 0 2 combined with ArCH 2 0 2 or by reacting with the hydroperoxy species ArCH 2 0 2 H to produce ArCHO Co(III) and OH. 2

PAGE 20

5 The corresponding aromatic acids are produced b y s ubsequent aldehyde oxidation via a peroxy acid intermediate. The addition of the bromide allows for for111ation of bromine atoms ( Br) via electron transfer oxidation of bromide by Co ( III). The bromine atom has been demonstrated to be extremely efficient at hydrogen atom abstraction which can allow for rapid fortnation of the benzyl radical thus initiating the autoxidation sequence. 1 2. Nucleophillic Attack on Coordinated Substrates An example of this classification is the Wacker Process for oxidizing ethylene to vinyl acetate. This reaction is provided in Equation 1-3. (Equation 1-3) H 2 C = CH 2 + 0 2 + HOAc PdC1 2 and CuC1 2 are two catalysts necessary for the above reaction to proceed in an acetic acid (CH 3 COOH ) solvent. The initial stage of the oxidation involves then coordination of the ethene substrate with concurrent or subsequent loss of chloride to fotm a sixteen-valence electron palladium (II) anion which in turn undergoes a ligand replacement reaction to yield the neutral palladium (II ) aquo species. Coordination of ethene to the palladium (II) metal center results in a decrease in the double bond electron density allowing it to become more susceptible to nucleophillic attack by either OH or The palladium metal is re-oxidized to palladium chloride by the co-catalyst either CuCI 2 or FeC1 3 as described above. In this case the co-catalyst CuC1 2 is the reduction

PAGE 21

6 product of CuCl that is readily oxidized with either air or oxygen. Therefore thi s catalyst combination gives rise to a catalytic system which allows for air or oxygen oxidation of ethene to ethanal 2 3. Metal-Catalyzed Oxygen Atom Transfer Reactions from Coordinated Hydroperoxides or Metal-Oxo Species to Organic Substrates The reaction of propylene with alkylhydroperoxides to yield propylene oxide falls within this category. This reaction is illustrated in Equation 1-4. (Equation 1-4) Mo Cat 0 H 3 CHC=CH 2 + ROOH H3CHC~CH 2 + ROH This process involves the reaction of a homogeneous metal catalyst molybdenum with an organic hydroperoxide and an alkene. As a result this reaction produces an epoxide product in a relatively high yield. This oxidation process proceeds with the hydroperoxide becoming ''activated by coordination'' to the metal center. As a result, the peroxide's oxygen experiences a decrease in electron density rendering it susceptible to nucleophillic attack by the alkene. Additional metal catalysts for this reaction are those belonging to the second row positioned early in the series and possessing high attainable oxidation states. Metals with these characteristics include Mo(VI) W(VI) and Ti(VI). 2 Transfer of the oxygen from the metal hydro peroxide complex to the alkene is suggested to occur via a cyclic transition state as detailed in Equation 1-5.

PAGE 22

7 (Equation 1-5) The success of the Mo(VI) W(VI) and Ti(VI) catalysts for this type of oxidation reaction is attributed to their being weak oxidants with poor one electron redox potentials : -0.21, -0.03 and -0.37eV respectively Therefore, they are poor catalysts for homolytic hydroperoxide decomposition as exhibited in Equation 1-6, which is detrimental to the desired epoxidation reaction. 1 It is important to emphasize that only the autoxidation, category number 1, utilizes the direct reaction of molecular oxygen with the organic substrate. 6 Homogeneous catalysis is now a relatively mature field with a number of diverse reactions being investigated for infortnative studies as well as mechanism and theoretical analysis. Along with the growth in this field a relationship to other important area s has been finnly established. These areas include heterogeneous catalysis, organometallic chemistry and bio-catalysis. Homogeneous catalysis is a resource, which is virtually untapped and its importance cannot be overestimated. In the age of industry becoming oriented towards specialty chemicals methods to catalyze functional group transformation, hydrocarb o n

PAGE 23

8 activation, polymerization and asymmetrical catalysis, the need and importance for effective homogeneous catalytic systems will continue to grow. The activation of C-H bonds in alkanes by transition metal complexes, not long ago thought to be the most difficult challenges facing chemists, is now almost commonplace. Many, examples, often under remarkably mild conditions, have appeared. 1 3 7 Problems arise in the transformation of these fundamentals into the practical alkane conversion process. These reasons include thermodynamics, a reaction being uphill at ambient conditions, catalyst stability, and catalyst incompatibility with 0 2 to activate alkanes. Other problems include the issue of selectivity. Regioselectivity is an obvious problem, one such case is if terminal alcohols are preferred over their isomeric products. More important, the products (alcohols and aldehydes) tend to be considerably more reactive than the starting alkane. This then places a limit on the achievable yield of the desired hydroxylated product. General Classification of Alkane Activation Reactions Three categories of remarkably facile alkane activation can be described: oxidative addition (Equation 17), cr-bond metathesis (Equation 1-8) and electrophillic substitution (Equation 1-9).1 3,7 (Equation 17) (Equation 1-8) (Equation 1-9) LM +R-H n LM-X n + R-H L MY + + R-H D LnM-R + H-X

PAGE 24

9 The first in this classification of reactions oxidative addition reactions, ha s been found to occur in low valent electron rich coordinatively unsaturated metal center s towards the right end of the transition metal series (Groups VI-X). 3 Sigma-bond metathesis reactions the second classification, are observed for complexes containing Group III transition metals (including lanthanides and actinides) as well as metals belonging to Group IV Also exhibited in this category are examples of intramolecular reactions with metals belonging to Group V. These reactions lead to stable organometallic products; however very limited examples of alkane functionalization have been achieved. Each of the active species involved is highly sensitive to 0 2 and the oxygenated products that are produced 3 Finally, electrophillic substitution reactions have been observed with the ''traditional'' or '' classic '' coordination complexes. An example of a such complex is PtCl (H 2 0) 4 _x.s 9 In the duration of the mechanistic process a stable organometallic species is not obtained but net functionalization of the alkane is achieved. This category of catalysis appears attractive in that the species involved will utilize dioxygen This property allows in principle for the closing of the catalytic cycle. Examples of Alkane Activation by Homogeneous Catalysts Many examples of homogeneous catalytic reactions are published in literature although this number is too large to provide one example for each area of activation. Two important examples are exhibited to provide a general idea of the catalytic chemistry being perfo11ned on the industrial and laboratory scale with homogeneous catalysts.

PAGE 25

10 The hydrofor1nylation or Oxo reaction discovered in 1938 by Roelen is utilized on a large industrial scale. 1 0 11 12 This process employs a homogeneous catalyst based on cobaltio 1 1 12 or rhodium. IJ Most commonly the HCo(C0) 4 pre-catalyst is employed, the product generated by the in-situ hydrogenation of C0 2 (C0) 8 In 1961, Heck and Breslow 14 proposed the mechanism for the cobalt-based Hydrofo11nylation process, as illustrated in Figure 1-1. Five elementary processes comprise the complete mechanism. The first step involves the dissociation of the carbonyl ligand from HCo(C0) 4 to produce the catalytically active species tricarbonylhydridocobalt, [HCo(C0) 3 ]. This is followed by the combination of the active catalyst with the olefID to generate a n-olefm complex. The migratory insertion of the olefm into the Co-H bond is the third step, proceed by the alley 1 undergoing a migratory insertion into the Co-CO bond. In the fmal step an H 2 induced aldehyde elimination occurs, producing the desired oxidized product and regeneration of the active catalyst. Another cobalt based catalytic system is the reductive carbonylation of methanol (Equation 1-10). This example illustrates how cobalt-based catalysts have dominated this area of industrial chemistry for the past fifty years. 15 22 Alternative metals are capable of catalyzing the identical reaction; however, these are generally inferior to those of the cobalt-based systems. (Equation 1-10) CH 3 0H + CO + H 2

PAGE 26

11 HCo(C0) 4 RCOH 1 co .----.,. HCo(C0) 3 5 2 H-H---R-CO-Co(C0) 3 4 3 RCo(C0) 4 co Figure 1-1: Catalytic Cycle for the Hydrofonnylation Reaction

PAGE 27

12 Improvements to the cobalt-based catalyst have been achieved through the use of co-catalysts and promoters. Such promoters employed include iodide, phosphines and transition metals. 18 Iodide, by far the most important of this group, is almost always utilized as a promoter. Despite advances in catalyst perfonnance, high pressure and temperatures are required. Conditions for these catalysts are typically operated at pressures of 4000-8000 psi and temperatures of l 75-200C. The increased pressures required can pose obvious difficulties for reactor design and operation, as well as for safety considerations. The elevated temperatures required can lead to the for1nation of heavy by-products via the Aldol condensation reaction of acetaldehyde. 1 9 20 It is shown both rhodium and cobalt are capable of catalyzing a variety of carbonylation reactions. 23 24 Rhodium catalysts have been demonstrated to be significantly more active than their cobalt counterparts and allow reactions to proceed at much lower extremes of temperature and pressure. However, iodide is necessary as a promoter. In the presence of this promoter and CO, rhodium is an extremely proficient catalyst for the carbonylation of methanol to acetic acid. This process is known as the Monsanto Acetic Acid Process. 2s-2s Designing a Homogeneous Catalyst The focal point of our research is directed towards designing and developing a catalyst able to undergo chemistry involving a transition-metal catalyzed oxygen atom transfer from a metal-oxo complexes or species. The activation of alkanes, primarily methane, utilizing this newly developed metal-oxo complex is our primary goal. The development of this catalyst can be hindered due to the following conditions:

PAGE 28

13 1. The role of the transition metal complex for activation of molecular oxygen can be complicated by numerous competing reaction pathways which are present . 2. In order to selectively oxidize a variety of substrates, a number of different conditions may be required. 3. The lack of any obvious pattern or role portrayed by the metal, as well as lack of a detailed reaction mechanism. A catalyst can either be homogeneous or heterogeneous in nature. A homo geneous catalyst was selected on the basis of its advantages in a reaction mixture and underlying chemistry. The advantages and disadvantages of utilizing a homogeneous catalyst are listed in Table 1-1 and Table 1-2 respectively. Five Classes ofMetal-Oxo Reactions Oxidation pathways have been exhibited to occur in a number of general reaction pathways. Drago29 has fo11nulated five classes of metal-oxo reactions, which are based on the role of the transition metal complex. In order to effectively categorize these reactions, the mechanism for substrate oxidation was omitted. The five classes are listed below and each reaction mechanism summarized. An example of each reaction class is also provided. Class I: Metal Bound 0 2 (Equation 1-11) M + 0 2 M-0-0 (Equation 1-12) M-0-0 + RH M-0-0-H + R

PAGE 29

14 Table 1-1: Advantages of Homogeneous Catalysts Advantage Description Homogeneity Uniform active structure Efficiency Theoretically all the atoms or molecules of the catalyst are available for reactants Reproducibility A result of well-controlled active sites High Selectivity Often more specific than a heterogeneous catalyst even for asymmetric induction Mild Reaction High activity often achieyed under mild conditions Conditions Controllability If chemical info11nation is readily available modification allowing the control of electronic and steric properties of the metal center can be attained Reaction Mechanism Investigation of the molecular reaction can be performed.

PAGE 30

15 Table 1-2: Disadvantages of a Homogeneous Catalyst Advantage Description Separation The separating of an expensive catalyst from the reaction mixture poses a problem and can be costly, and may also require special treatment, which usually destroys the catalyst. Solubility The range of solvents s uitable for a homogeneous catalyst is often limited by the solubility of the catalyst; a compatible solvent is not necessarily the most suitable for a high reaction rate.

PAGE 31

(Equation 1-13) M-0-0-H + R (Equation 1-14) ROOH 16 M + ROOH ROH or R = O Molecular oxygen is coordinated to the metal (Equation 1-11) to increase its basicity and radical reactivity. The substrate then undergoes hydrogen atom abstraction by the metal-peroxo species to generate a metal hydroperoxo species and an alkyl radical (Equation 1-12). The metal hydroperoxo species and the alkyl radical then undergo recombination (Equation 1-13) to generate an alkylhydroperoxide species and the reduced metal species. The final step (Equation 1-14) involves decomposition of the alkylhydroperoxide to yield the corresponding alcohol and ketone. Cobalt (II) coordination compounds that reversibly bind dioxygen are examples of this group. Class Il: Metal-Oxo via 0 2 (Equation 1-15) M + 0 2 (Equation 1-16) M-0-0 + M (Equation 1-17) + -O-O-M1 + (Equation 1-18) M = O + S M-0-0 M-0-0-M SO + M Once again the metal coordinates dioxygen (Equation 1-15) ; however, in this series another metal enters the coordination sphere and fortns a -peroxo dimer (Equation 1-16). This -peroxo dimer then undergoes cleavage to for1n two high-valent metal-oxo complexes (Equation 1-17). The metal-oxo complex is then able to undergo oxygen atom

PAGE 32

17 transfer or another similar type reaction (Equation 1-18) thus regenerating the lower oxidation-state of the metal and producing the oxidized species of the substrate. Metal oxidation states normally involved in this cycle are two to four three to five or four to six. One system to be discussed cis-[Ru(dmp) 2 (H 2 0) 2 ] 2 +, is an example of this category. Class III: Metal-Oxo via Peroxides (Equation 1-19) M" + + H 2 0 2 (Equ ation 1-20) M" + 2 (0) + S so + M" + This class of reactions takes advantage of the more powerful oxidizing ability of hydrogen peroxide to oxidize lower-valent metal complexes to metal-oxo complexes The low-valent metal complex is oxidized with hydrogen peroxide (Equation 1-19) to a higher valency producing a metal-oxo complex and a benign by-product, water. The metal-oxo species is then able to undergo oxygen atom transfer or a similar reaction (Equation 1-20), thus regenerating the lower oxidation state of the metal and producing the oxidized species of the substrate. A candidate for this class is a monooxygenase enzyme, which involve metal-oxo oxidants. Class IV: Metal-Peroxo Systems /VaMetal Catalyzed Peroxide Decomposition (Equation 1-21) (Equation 1-22) M" + M" + ) + OH HO 2

PAGE 33

18 /Vb: Metal-Peroxo Formation (Equation 1-23) M=O + H 2 0 2 (Equation 1-24) M 0 2 + S SO + M = O Metal peroxo complexes are the reactive inter1nediates for this group Several reactions with substrates yield further divisions of this class. In class IV a, the metal complex reacts with hydrogen peroxide to generate radical species which are the result of peroxide decomposition (Equations 1-23 and 1-24). An example of this class is HaberWeiss and Fenton Chemistry. The radial species the active oxidant, are then able to oxidize the substrate. Class IVb is the result of the bound metal peroxo or alkylperoxo complex attacking the substrate. The metal peroxo is generated by the reaction of a metal-oxo species with hydrogen peroxide. The alkylhydroperoxo species is generated in the same fashion as illustrated in Equation 1-22 . Asymmetric epoxidation of alkene s with alkylperoxides are examples of this class of chemistry Class V: Metal Centered Oxidizing Agents Metal center is active electron acceptor in the oxidation Often involves coordination of substrate to metal center "activation". The Wacker process, Equation 1-3, is an example of this class. The reduced complex is re-oxidized by 0 2 or peroxide.

PAGE 34

19 The approach taken in this research is to develop a catalyst that possesses properties pertaining to either Class II or III. The developed catalyst could also possess a mechanism of regeneration with dioxygen to generate a high-valent metal-oxo complex as in Class V, necessary for alkane oxidation. This catalyst must be able to perforin the key step, that being the generation of a strongly oxidizing high-valent metal center 30 The desired reaction pathway, illustrated in Figure 1-2, exhibits the need for a metal center, which can be activated by hydrogen peroxide or molecular oxygen to form the mono-oxo species. This activated complex may be a powerful enough oxidant to activate the substrate or be further oxidized by hydrogen peroxide or dioxygen to generate a di-oxo species, a complex with greater oxidizing strength than the mono-oxo species Upon delivery of its one or two oxygen atoms the reduced metal center could be re-oxidized by hydrogen peroxide or oxygen to regenerate the mono or di-oxo species, thereby closing the catalytic cycle Homolytic or Heterolytic Pathway? There are two pathways, which exist for a metal center to be activated by dioxygen or hydrogen peroxide These pathways are known as homolytic and heterolytic oxidation. The foimer 31 normally involves the following transition metal couples: reaction include the production of a free radical intermediate, oxidation occurring in the outer-sphere via bimolecular steps, non-coordination of the substrate. Oxidation products are generally not very selective or stereospecific, and the metal center undergoes a one electron oxidation and reduction step. Examples of relevant oxidations employing this

PAGE 35

20 Substrate Oxidized Substrate Figure 1-2; Desired Reaction Pathway for Substrate Oxidation

PAGE 36

21 oxidation pathway include the non-stereoselective epoxidation of olefins occurring with transition metal complexes of vanadium, manganese and iron and the hydroxylation of alkanes and arenes occurring with the transition metal complexes of vanadium, chromium, manganese, iron, cobalt and copper. The latter heterolytic oxidations, 31 involves the following transition metals: Tiv Characteristics of this reaction pathway include the inability to produce a free radical intermediate, the reaction occurring in the coordination sphere of the metal, substrate coordination resulting in activation, high selectivity and stereospecificity of the oxidized products and no net change in the oxidation state of the metal. However if a change in the oxidation state occurs, it is via a two electron step i.e. M 2 + to M 4 + Examples of heterolytic oxidations include the stereoselective epoxidation of olefins with transition metal complexes of titanium vanadium and tungsten and the ketonization of olefin s with transition metal complexes of rhenium, iridium, palladium and platinum. Mimoun 30 has also categorized a reaction coordinate pathway involving metal-oxygen species, which is illustrated in Figure 1-3. Effect of Ligand on Metal Center Numerous and thorough studies for polypyridyl ligands with ruthenium metal centers by Drago 3 1 Meyer 32 Takeuchi 33 and Che 34 have been perfor1ned. These studies have revealed certain features that make these complexes capable of catalyzing oxidations of hydrocarbons with molecular oxygen. These complexes display a series of reversible

PAGE 37

22 hydroxo Q. 0/ + Mn+ 0 Mn+1 / '/ Mn+1 0 2 Mn+2:Q n+1 superoxo -peroxo oxo Mn+1 0 _Mn+1 -oxo +R~, Mn+2 OOR +ROOH / -alkylperoxo -HX Mn+2 Mn+2 X 'o + H+ + H202 ---Mn+2 OOH C __.. -HX peroxo hydroperoxo Figure 1-3: Detailed Reaction Pathways of Metal-Oxygen Species

PAGE 38

23 and accessible oxidation states (i e from 2 + to 6 + ) to reduce dioxygen completely. The poylpyridyl ligand provides the stability for the metal center to achieve higher oxidation states and impart stability to the complex It also fonns stable ruthenium oxo-species, {Ru(0)} 2 + and {Ru(0) 2 } 2 + which are known to be excellent metal oxygen atom-transfer reagents towards a variety of substrates. The activation of methane and other saturated alkanes is the field where the majority of our interest lies. The need for an inexpensive environmentally friendly and benign by-product oxidant is of top concern. Many oxidants are available which include dioxygen, hydrogen peroxide, alkyl hydroperoxides peracids sodium sulfate, oxone nitric acid dichromate, bromine manganate and Caro s acid (H 2 S0 5 ) However the last five are costly and involve expensive disposal problems. Nitric acid also produces noxious NO x compounds as byproducts. Peroxygen reagents have been utilized in chemical synthesis for a number of years. 3 5 As the public, chemical producers, and governmental agencies direct their efforts towards '' greener'' pathways the interest in the use of peroxygen oxidants has increased. A reagent hydrogen peroxide (H 2 0 2 ), can offer environmental and economic benefit s. Hydrogen peroxide is available for chemical synthesis as an aqueous solution in concentrations ranging from 35% to 90% by weight. Stabilizers are nonnally present in the part per million levels to prevent decomposition. With no additional additives being present, a minimization in unwanted side reactions is an additional advantage

PAGE 39

24 Hydrogen peroxide can also be stabilized for extended periods of times under a variety of conditions. As an aqueous solution, hydrogen peroxide will lose less than one percent of its active oxygen content per year, when stored in compatible containers at ambient temperatures. This stability allows hydrogen peroxide to be a more efficient oxidant than dichromate or pertnanganate on a weight for weight basis. Reactions involving hydrogen peroxide are traditionally perfortned under mild conditions of temperature and pressure, preventing an increase in peroxide decomposition. By perfo11ning under these conditions the need for a large capital investment is eliminated as well as a decrease in safety concerns. Reactions, which use hydrogen peroxide as an oxidant, can also proceed in either an aqueous, organic or neat solvent. This allows for reaction and reactor design to be engineered allowing one to avoid the use of solvents, thus eliminating costly start-up expenses and production of waste streams. 36 The major benefit of using hydrogen peroxide is its environmental acceptability. When the oxidizing power of the peroxide is spent, only water remains as the by-product, eliminating the need for expensive effluent disposal and treatment. Hydrogen peroxide is also relatively inexpensive as a reagent, when compared on the basis of oxidizing power and by-product production. Although chlorine and oxygen oxidants are less expensive, since these are gases, phase separation concerns are raised as well as the inability to achieve selectivity towards oxidation products. Hydrogen peroxide can be utilized in a variety of methods: direct activation, catalytic activation and activation via peroxides. All require the use of hydrogen peroxide to be activated to an even more potent oxidant. Figure 1-4 summarizes methods

PAGE 40

25 for activating hydrogen peroxide. Our use of hydrogen peroxide falls under the area of catalytic activation. Hydrogen peroxide is activated with a metal complex by fo1ning a metal-oxo or metal-peroxo species. 36 The Einchem Corporation has demonstrated the feasibility of using hydrogen peroxide as an oxidant on a commercial scale. Einchem has developed a heterogeneous titanium silicate zeolite catalyst, commercial tradename of TS-1. 37 This catalyst, totally heterogeneous in nature, has a wide synthetic use for hydrogen peroxide oxidations. Einchem has commercialized this catalyst for the hydroxylation of phenol to hydroquinone and catechol. The catalysts used in this research will employ hydrogen peroxide and dioxygen. Hydrogen peroxide was chosen as the primary oxidant for hydrocarbons because of its ease of handling and oxidizing strength. Dioxygen is the desired oxidant but hydrogen peroxide can be used safely with hydrocarbons, while dioxygen requires one to remain within the explosion limits.

PAGE 41

M ~=O y 26 Catalytic 'Direct' Activation Activation M-OOH MY OH(x+n)+ M(n+) [OH+] M+ or UV HX H2 RC(O)OOH 04 RC(NH)OOH H 2 S0 5 Activation Via Peracids [HO.] Figure 1-4: Methods for Activating Hydrogen Peroxide

PAGE 42

CHAPTER2 HOMOGENEOUS CATALYZED PARTIAL OXIDATION OF METHANE WITH HYDROGEN PEROXIDE AND OXYGEN Introduction Cis-rutheni11m oxo complexes were previously reported to be effective catalysts to activate hydrogen peroxide and molecular oxygen for the selective oxidation of alkenes 38 and alkanes. 39 40 The sterically hindered complex cis-[Ru(dmp) 2 (H 2 0) 2 ](PF 6 ) 2 (II), where dmp is 2,9-dimethyl-1, 10-phenanthroline (Figure 2-1 ), is a precursor that can be oxidized with hydrogen peroxide to fo11n cis-[Ru(dmp) 2 (0)(H 2 0)](PF 6 ) 2 or cis [Ru(dmp) 2( 0) 2 ](PF 6)2 as shown in Equations 2-1 and 2-2. N--< Figure 2-1: Structure for 2,9-dimethyl-1, I 0-phenanthroline Unsuccessful attempts to isolate these oxo complexes led to their characterization based (Equation 2-1) 27

PAGE 43

28 (Equation 2-2) cis-[Ruiv ( dmp ) 2 (S)(O)](PF 6 ) 2 on isobestic points in the electronic spectra and changes in solution NMR as hydrogen peroxide is added to II. The electronic spectra of the oxidized complex are similar to those of analogous ruthenium-oxo complexes. 41 Cis-{Ru(0) 2 } 2 + complexes isomerize to the trans-complexes which are weaker oxidants than the cis analogues. 4 I 42 4 3 44 4 s The novel aspect of the dmp ligand is its steric requirement which prevents isomerization of the cis-[Ru(0) 2 ] 2 + complex and inhibits fortnation of both tris complexes and stable -oxo dinuclear species. The intra-ligand repulsions and neutral ligand charge also make the high oxidation state ruthenium center electropositive. 41 Preliminary results describe the use of this complex to catalyze the oxidation of alkanes using H 2 0 2 as the oxidant. 4 Figure 2-2 s ummarizes the proposed 4 6 catalytic cycle for alkane oxidation via cis-[Ru(dmp) 2 (0) 2 ] 2 +. The cis-[Ru(II)(dmp) 2 {H 2 0) 2 ] 2 + precursor is oxidized rapidly by H 2 0 2 to f onn the (III) (IV) and (VI) oxidation state complexes. In alkane oxidations, cis-[Ru(III)( dmp ) 2 (H 1 0)(0H)] 2 + reacts with dioxygen to be unreactive with molecular oxygen. 46 When appreciable concentrations ofRu ( IV ) exist, the Ru(II) complex reacts with it to fo1111 cis-[Ru(III)(dmp) 2 (H 2 0 )( 0H)] 2 + For kinetic reasons alkane oxidations require the electrophillic cis-[Ru ( VI)( dmp ) 2 ( 0 ) 2 ] 2 +

PAGE 44

29 Oxidized Substrate Substrate [Ru( dmp ) 2 (H 2 0) 2 ] 2 H202 H 2 0 [Ru( dmp ) 2 (H 2 0)0] 2 Figure 2-2: Proposed Catalytic Cycle for Alkane Oxidation via cis-[Ru(dmp) 2 (0) 2 ] 2 +

PAGE 45

30 complex. The cis-[Ru(IV)(dmp) 2 (H 2 0)(0) 2 ] 2 + fonned after oxygen atom transfer must be oxidized back to the Ru(VI) complex with hydrogen peroxide 40 to make the s ystem catalytic in CH 3 CN solvent. Figure 2-3 illustrates the catalytic cycle of the ruthenium analogue. Methane was selected as the substrate for study because the oxidation of natural gas into a liquid product is of great industrial and economic importance. 47 Figure 2-4 depicts the worldwide distribution of natural gas resources. Currently, methanol is produced from methane by the steam ref or1ning of methane into synthesis gas followed by the catalytic conversion of the syn gas into methanol. In order to increase efficiency, the direct partial oxidation of methane to methanol is the desired pathway. 4 8 However hydrogen atom abstraction commonly involved in methane oxidation occurs more rapidly with methanol than methane resulting in the over-oxidation of methanol to CO 2 This research extends earlier reports of the use of cis-[Ru(Vl)( dmp ) 2 (0 ) 2 ] 2 + as a catalyst for methane oxidation. Greatly improved conversions to oxygenates result by trapping methanol and slowing its over-oxidation. Also demonstrated is an improved catalyst precursor evidence for oxidation of methane by 0 2 with this catalyst, and demonstrate oxidation of methane by peracids Experimental Materials and Methods RuCl 3 x H 2 0 2 9-dimethyl-1 10-phenanthroline Li Cl, NaPF 6 60% HPF 6 NH 4 PF 6 NaCF 3 S0 3 were all used as received from Aldrich. Ethylene glycol acetonitrile glacial acetic acid acetic anhydride, 4A molecular sieves and hydrogen

PAGE 46

H 2 0 2 (Fast) 0 2 (V. Slow) Ru 11 + Ru 1 v Ru 11 Epoxide Ru 111 31 Ru" + Ru'v Ru'V(Q) Alkene Oxidized Alkane Figure 2-3: Detailed Catalytic Cycle for Ruthenium Analogue Alkane

PAGE 47

Middle East 25% Asia-Oceania 13% N America 12/o 32 Africa 10% Russia 30o/o W. Europe S. America S% 501o Figure 2-4: Distribution of Worldwide Natural Gas Resources

PAGE 48

33 peroxide (30%) were all used as received from Fisher Scientific (ACS grade). Acetonitrile was distilled over P 2 0 5 under dinitrogen and was stored over 4A activated molecular sieves. Methane (99.99%) was purchased from Matheson and used as received. Physical Measurements UV-Vis measurements employed a Perkin Elmer lambda-6 spectrophotometer. The pH measurements were made with a Fisher Accumet model 630 pH meter. NMR spectra were recorded on a Varian VXR300 spectrometer. FAB mass spectral data were obtained by Dr. David Powell (U.F.) in a m-nitrobenzyl alcohol matrix. U F Analytical Services perfonned elemental analyses. Synthesis of Compounds Cis-Ruthenium(II) Bis(chloride)bis(2,9-dimethyl-1,10-phenanthroline) Mono hydrate, [Ru( dmp ) 2 Cl 2 ] H 2 0 (I). Cis-[Ru( dmp ) 2 Cl 2 ] H 2 0 was synthesized as reported. 40 Analysis : Calculated for C 28 H 26 N 4 0Cl 2 Ru; C, 55 45; H, 4.29; N 9.24. Found: C, 55.65; H, 4.35; N 9.29. Cis-Ruthenium(II) Bis(aquo)bis(2,9-dimethyl-1,10-phenanthroline) bis(hexafluorophosphate), Cis-[Ru(dmp) 2 (H 2 0) 2 ](PF 6 ) 2 (II) A 1.0 g (1.7 mmol) portion of I was dissolved in 150 ml of deionized H 2 0, under N 2 by heating 50 C for 30 minutes. After adding 50 ml of a saturated NaPF 6 (aq), the solution is cooled placed in an ice bath for two hours and filtered. The precipitate is redissolved in H 2 0 by heating to 50 C ; a saturated NaPF 6 (aq) solution is added and reprecipitated as above Re crystallization and exchange of c1for PF 6 is repeated until the filtrate affords a negative

PAGE 49

34 chloride test with AgN0 3 The product is dried in vacuum at 60 C overnight. Analysis: N 6.72. For comparison purposes cis-[Ru(dmp) 2 (H 2 0) 2 ](PF 6 ) 2 was also synthesized using 39.85; H 3.32; N 6.64. Found: C 39.57; H 3.27; N, 6.63. Cis-Ruthenium(II) Bis(aquo)bis(2,9-dimethyl-1,10-phenanthroline) bis(trifluoromethanesulfonate), Cis-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 (III). A l.Og ( 1 7 mmol) portion of I is slowly dissolved in H 2 0 as above. Following the addition of 50 ml of a saturated aqueous solution ofNaCF 3 S0 3 the resulting solution is cooled and placed in an ice bath to complete precipitation. The product is collected, the filtrate tested for chloride and recrystallized with CF 3 S0 3 until the filtrate gives a negative chloride test. The resulting solid is dried under vacuum at 60 C overnight. Analysis: Calculated for Oxidation Procedure The pressurized oxidations were carried out as previously described 4 9 in glas s, batch hydrogenation reactors. Figure 2-5 provides an illustration of such batch reactor The reaction mixtures were varied as described in the table footnotes. Blank runs omitting certain reactants and solvent components are also described in the tables. Oxidations with oxygen use 30 psig of a 5% 0 2 in helium mixture, to remain outside the explosion limits and 20 psig methane. Reaction temperatures nonnally were maintained

PAGE 50

Pressure Gauge 1/4" NPT 90 Elbow Safety Cage 35 Sampling Port 1/4 11 NPT Cross Silicone Stopper On/Off 1 / 8" NPT Ball Valve 1/4" NPT Screw Valve Gas Inlet/Purge Gas Centrifuge Bottle Figure 2-5: Diagram of Batch Type Hydrogenation Reactor

PAGE 51

36 between 75 to 77 C. To remove air initially present nitrogen gas is purged through the reactor, the apparatus is pressurized with the substrate which is then released and re pressurized to the desired pressure. The oxygenated products of the reaction were analyzed and quantified with a Hewlett-Packard 5890 Gas Chromatograph equipped with a FID detector and outfitted with a 30m Alltech RSL 160 column (Sm thickness) Helium was employed as the carrier gas. Carbon dioxide and carbon monoxide analyses were perfo1n1ed with a Varian 3700 Gas Chromatograph equipped with a TCD detector outfitted with a 15 Carboxen Column. Methane was quantified by gas chromatography (TCD) using N 2 as an internal standard. Three chromatograms were measured for each sample using injection volumes of 0.1ml for gas and 0.1 l for liquid samples. The foil owing definitions describe ter111s used in the presentation of the results. Selectivity to any product is the moles of a given product divided by the total moles of all products for1ned expressed in percent. The percent CO 2 produced is moles of CO 2 divided ~y the total moles of all products. The selectivity to oxygenates is the moles of H 2 CO CH 3 0H and CH 3 C{O)OCH 3 divided by the total moles of products in percent Traces of CH 2 (0CH 3 ) 2 HCOOCH 3 are for111ed but not quantified. Percent peroxide efficiency is the moles of H 2 0 2 needed to account for all oxidized products including CO 2 divided by the moles of H 2 0 2 consumed. The percent conversion of CH 4 is the mole s carbon in the oxidi z ed products divided by the moles of CH 4 added to the react o r. Safety Pre c autions the combination of molecular oxygen with organic compounds and sol v ents at elevated temperatures and pressures are potentially explo s ive.

PAGE 52

37 Extreme caution should be taken during the charging and disassembly of the experimental apparatus. Equipment which can generates sparks must be avoided a safety shield and cooling of the batch reactor in an ice bath for 30 minutes prior to disassembly is recommended Results and Discussions Anion Modification Reproducibility for the oxidation of CH 4 cis-[Ru( dmp ) 2 (H 2 0) 2 ](PF 6 ) 2 catalyst precursor4 depends on the purity of the complex. Chloro complexes are inactive 41 and excess AgPF 6 used to remove the chloride from the dichloro precursor inhibits 3 8 39 4 0 the reaction. Using NaPF 6 instead of AgPF 6 requires a large amount ofNaPF 6 and repeated recrystallizations. The hexafluorophosphate anion of the fmal product is hydrolytically unstable and extensive etching of the glass vial occurs during storage. With more than 20% fluorine, verification of the complex purity by elemental analysis is difficult. Complex degradation is evident as well as the inability to obtain an appropriate NMR spectrum for complex II as illustrated in Figure 2-6. In an attempt to overcome these problems, the anion was changed to trifluoromethanesulfonate, a poorly coordinating anion. A shortened synthesis time need for less anion in the synthesis, more reliable elemental analyses and stability towards hydrolysis are immediate advantages of this new complex. The direct effects of altering the anion are exhibited in the lack of complex degradation and the ability to obtain an identifiable NMR spectrum. A NMR for this newly synthesis complex III is provided in Figure 2-7.

PAGE 53

38 C"'., .. ,o '# ,,..._ V ,... "1 cc ,... I co I I I I I I I I I I I I I I I I I I I I I I I J I 1 I I t 8 7 9 V t.n mo oco co NO lO "'co ,.... 0 0 CI) "' 0 t"'I "' ''''l''''f'''''''''l'''''''''l''''I 3 2 1 PPM

PAGE 54

~en t'\J 10 QO al-..0 I/') ~( Cl) I I I I I I I I I I 0 (liiiJiiiiJ I 3 r r.n
PAGE 55

40 Ruthenium Cataly z ed Oxidation of Methane with Hydrogen Peroxide The trifluoromethanesulfonate derivative complex III is a potent catalyst for the oxidation of CH 4 by hydrogen peroxide in an acetonitrile solvent. Within 48 hours at 75C CO 2 CO and traces of both methanol and formaldehyde appear which correspond to 60% of the initial concentration of methane. A more active catalyst less methanol and more CO 2 result with III than with the PF 6 derivative (II). Previous reports suggest the di-oxo complex is necessary for the activation of the C-H bond in methane 40 In water the hydrogen peroxide potential 4 6 gives a negative free energy for the fonnation of the di-oxo species only when the solution pH is less than 5. However at a pH of 2 in water the oxidation of CH 4 by H 2 0 2 with this catalyst is not observed at 75 C. 4 6 The relevant potentials are not known in acetonitrile but spectral studies indicate a high oxidation state complex is forrned upon addition of H 2 0 2 The best condition reported for methane ~xidation 4 0 with complex II in acetonitrile yield only 5 turnovers for methanol accompanied by extensive over-oxidation of methane to CO 2 occurs. This prompted an experiment to deter111ine the activity of this catalyst for methanol oxidation to CO 2 Methanol catalyst (6.6xlQ 5 moles) and hydrogen peroxide ( 5.0xI0 2 moles) reacted in 4 hours at 75 C to convert 97 % of the methanol to carbon dioxide and carbon monoxide. Thus, the main challenge to the use of these ruthenium catalysts for the selective oxidation of methane to methanol is to inhibit the over-oxidation. The successful oxidation of methane in H 2 S0 4 5 0 51 and CF 3 C(O)OH 52 5 3 5 4 to large quantities of available methanol can be attributed to trapping methanol as CH 3 0S0 3 H and CF 3 C(O)OCH 3 respectively and s tabilizing it from over

PAGE 56

41 oxidation. This led to an attempt to trap methanol in our system as an ester by reaction with acetic acid and acetic anhydride. Acetic acid was used since it is an acidic solvent, and relatively difficult to oxidize. With aqueous H 2 0 2 as the oxidant and water fo11ned in the reaction, the anhydride can function to keep the water concentration low enhancing ester fo11nation, as demonstrated in Equations 2-3, 2-4 and 2-5. (Equation 2-3) CH 3 0H + CH 3 C(O)OH CH 3 C(O)OCH 3 + H 2 0 (Equation 2-4) H 2 0 + CH 3 C(O)OC(O)CH 3 2 CH 3 C(O)OH CH 3 C(O)OCH 3 + CH 3 C(O)OH The anticipated product, methyl acetate could be subsequently hydrolyzed to produce methanol and regenerate acetic acid. The oxidative stability of methyl acetate to over-oxidation by our catalyst was shown with an experiment in which methyl acetate, catalyst (6.6xl 05 moles) and hydrogen peroxide (S.Oxlo 2 moles) dissolved in acetonitrile were reacted for 4 hours at 75C. An 8% decrease in methyl acetate and for1nation of small amounts of carbon monoxide and carbon dioxide result, indicating that methyl acetate has the necessary stability to be an effective trap for methanol. The 8% oxidation observed could proceed through methanol fo11ned from the equilibrium shown in Equation 2-3. Using catalyst III and a solvent mixture composed of equal volumes of CH 3 CN / CH 3 COOH/(CH 3 C0) 2 0 7.4 millimoles of oxygenates forrned in 24 hours

PAGE 57

42 representing a significant improvement over the previously reported 0.5 millimoles in acetonitrile. 3 s 3 9 4 0 Methyl acetate methanol, fortnaldehyde formic acid methyl for111ate, carbon monoxide and carbon dioxide are all detected as products. As mentioned above, in the absence of acetic acid and acetic anhydride, only CO and CO 2 and trace amounts of fortnaldehyde are obtained with III as the catalyst. Increasing the reaction time and hydrogen peroxide concentration i n a s eries of experiments, gave increased amounts of CO 2 but led to an upper limit of l.5xI0 2 M methyl acetate in this solvent mixture Apparently when methyl acetate approaches this concentration, the equilibrium concentration of CH 3 0H reaches a level at which its rate of oxidation becomes equal to its rate of fottnation from methane. Decreasing the water in solution should produce a lower steady state concentrations of methanol ( i.e. the position of the equilibrium in Equations 2-3 and 2-5 is shifted towards ester fotmation ) and increase the trapping efficiency leading to decreased CO 2 and increased methy I acetate To remove water 4A molecular sieves were added to the reaction mixture and shown to have a s~gnificant effect on the conversion of methane to methyl acetate In comparable 24 hour runs the conversion to oxygenates increased from 3 millimoles to 7 millimoles {Table 2-1, Experiment 1) with the addition of sieves even though some peroxide decomposition by the sieves occurs vide infra. This result reinforces the proposal that efficient trapping is the main challenge for effective methanol synthesis. Variation of the reaction time in Experiments 2 and 3 show that in four hour s the maximum amount of methyl acetate and minimum amount of CO 2 fotm. The s horter time also led to an increase in the amount of fonnaldehyde. The methyl acetate fortned corresponds to 61 turnover numbers (5 mmoles 6.6x10 2 M ) with a selectivity of 55 %

PAGE 58

43 Table 2-1: Oxidation Results for Methane@ 75C with cis-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 Using H 2 0 2 a Experiment MeCOOMe co 1 Total CH 4 Number mmol,( 0 /4) mmol,(%) Oxygenates Consumed mmol,( 0 /4) (%) 1 a,c 5,(44) 5,(44) 7,(62) 30 2a.d 4,(40) 3,(30) 7,(69) 26 3a,e 5,(57) 1,(11) 7,(80) 19 4a.r 10,(49) 9,(44) I 0,(49) 62 5b g 4,(70) 1,(17) 5,(87) 25 6b h 9,(80) 3,(27) 8,(71) 49 7b l 2,(100) 0 2 (100) NIA 8bJ 3,(47) 0 4,(62) 28 a Reactions 1-4 used 6.6 x 10 5 moles cis-[Ru{dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 catalyst 5 ml 30% H 2 0 2 (5.0xio 2 moles). The solvent mixture is 20 ml acetonitrile, 20 ml glacial acetic acid, 20 ml Acetic Anhydride, Initial methane pressure was 40 psig corresponding to 23 millimoles 3 5g 4A molecular sieves (MS) were used. The percent products are based on total products seen from all sources b. Reactions 5-8 used 2.4 x 10 5 moles cis-[Ru { dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 catalyst 1 ml 35% H 2 0 2 ( 1 1 x I 0 2 moles). The solvent mixture is 3 ml cr-dichlorobenzene 5 ml glacial acetic acid 1 O ml Acetic Anhydride, Initial methane pressure was 40 psig corresponding to 23 millimoles. l Og 4A molecular sieves (MS) were used. c Reaction Time 24 hours. The percent products are based on total products from all sources. d. Reaction Time 12 hours. The percent products are based on total products from all sources. e Reaction Time 4 hours The percent products are based on total products from all sources f. Blank experiment, perfonned in the absence of catalyst Reaction Time I hour. g. Blank experiment, perfonned in the absence of catalyst and acetonitrile Reaction Time 4 hours. h. Reaction Time 4 hours i Blank experiment perfonned in the absence of catalyst and methane Reaction Time 4 hour s. j Blank experiment, perfonned in the absence of catalyst, Reaction time 4 Hours

PAGE 59

44 while fo11naldehyde corresponds to 24 turnover numbers (2 mmoles) with a selectivity of 28%. The CO and CO 2 produced correspond to 9 and 67 turnover numbers (0.2 mmoles and 5 mmoles) respectively. The total turnover numbers for all the products is 198 which if they all arose from methane would correspond to a methane conversion of 38%. The total moles of oxidant used are 1.3xlo2 so 21 % of the peroxide has decomposed. Peracid For111ation and Reactivity In view of the difficulty of oxidizing methane and its low concentration in solution the importance of running control experiments to determine the source of products cannot be overemphasized. When the catalyst was eliminated from the reaction mixture, methyl acetate fortned in significant amounts, Experiment 4 of Table 2-1. In earlier research from this laboratory it was shown that hydrogen peroxide is catalytically activated by fortning peracids with organic acids 55 To determine if peracetic acid fo11ns, acetic acid acetic anhydride and 35% aqueous hydrogen peroxide were stirred at 25 C for thirty minutes in the absence of sieves. The difference in an iodometric titration of the products for total oxidant, and a cerium titration, for hydrogen peroxide, indicates by difference that 20% of the H 2 0 2 was converted to peracid (l.Oxlo2 moles). At this point methane (40 psig) was introduced and the reaction vessel placed in an oil bath at 75 C Experiment 4 (Table 2-1) shows that in one hour of reaction time 10 mmoles of methyl acetate (49% selectivity), traces of fo1111aldehyde, 0.4 mmoles of carbon monoxide and a large amount of carbon dioxide (9 mmoles or 44% of the products) were fotmed. This reaction led to about 19 rnmoles of total products which, if they all arose from methane would correspond to 89% of the methane. The decrease observed in the amount of

PAGE 60

45 methane by TCD is approximately 60% (Table 2-1) implying some of the oxygenates arise from peracid decomposition. To limit over-oxidation by peracid, the reaction time was decreased to twenty minutes. The reaction produced 6.4x10 3 moles of methyl acetate (69% selectivity) (l.lxlo 1 M), l.6x10-6 moles offor1naldehyde (0.04% selectivity), 3.9xl0-6 moles of carbon monoxide (0.06% selectivity) and 2.9xl 0 3 moles of carbon dioxide (31.1 %). The peroxide utilization was 97% and 1.0x10" 2 moles of products were observed. The decrease in reaction time leads to less products, but increased selectivity to methyl acetate. Blank runs were performed with and without sieves to deter1nine if hydrogen peroxide and acetonitrile for111 a peracid, CH 3 C(OOH)NH, which oxidizes methane to produce methyl acetate. In these experiments, acetonitrile, hydrogen peroxide and methane were added together and allowed to react for 48 hours at 75C. Iodometric titrations also confi11ned the absence of peracid. T}le next concern is the source of the products in the blank, as shown in Experiment 5. Oxidations of paraffins by 3,5-dinitroperbenzoic acid perbenzoic acid and perfluoroperacetic acid result 56 in more than 50% conversion of the paraffin. A radical mechanism is suggested 57 58 for the fo11nation of the hydroxylation products that is initiated by the decomposition of peracids as shown in Equations 2-6, 2-7, and 2-8. (Equation 2-6) R-C0 3 H

PAGE 61

(Equation 2-7) R-H + R (Equation 2-8) 'R + R-C0 3 H 46 'R. + R-H 'R-OH + R-C0 2 The R-Co 2 radical decomposes to fonn CO 2 and the alkyl radical ) which abstracts a hydrogen atom from the paraffm R H to fortn the R alkyl radical which reacts with the peracid to for1n the alcohol generating R OH and another RC0 2 In our system with CH 4 as the alkane and peracetic acid as the oxidant decomposition of CH 3 C0 3 H would generate methyl radicals which may exchange with methane in Equation 2-7 and would have to react as shown in Equation 2-8 to produce CH 3 0H Methanol f or1nation would in effect arise from peracid decomposition. Alternatively RC0 2 can react with R H to for1n R' which then reacts as shown in Equation 2 8. In this case CH 4 would be converted to methanol by CH 3 C0 2 forrned in Equation 2-5 reacting with methane as shown in Equation 2-6 to fonn methanol via Equation 2-8. Methane oxidation by peracid results in contrast to the literature mechanism where the products would derive from the solvent. In view of the possible exchange of methyl radical with labeled methane Equation 27 labeling experiments will not be definitive One also notes that less methyl acetate is produced with the ruthenium complex present ( Experiment 3, Table 2-1 ), than with it absent in Experiment 4 suggesting metal complex decomposition of H 2 0 2 Crude estimates of the methane that disappeared in the experiments described above indicate the metal complex is involved in Experiments 1-3 but more compelling evidence is required

PAGE 62

47 The question concerning the reactivity of methane with peracid raised above impacts not only on the metal catalyzed oxidations in acetic acid-acetic anhydride solvent but also on peracid oxidation of alkenes. 56 Blank runs at 75C, with catalyst, hydrogen peroxide, acetic acid/acetic anhydride, acetonitrile and no methane, produced one millimole of methane and traces of methanol after four hours of reaction. A similar experiment in cr-dichlorobenzene, showed only traces of methane and methanol. The source of methane is thought to arise from solvent facilitated peracid decomposition in acetonitrile. This proposal was not investigated further, but the solvent was switched from acetonitrile to cr-dicholorbenzene and Experiment 5-8 were carried out. Since labeling experiments are uninformative, efforts were expended to accurately deterrnine the methane disappearance to provide a material balance. To facilitate the metal catalyzed path, the catalyst and hydrogen peroxide concentrations were increased by 20%. Experiments 5 and 7 indicate that the beneficial effect of the sieves removing water is accompanied by the deleterious decomposition of hydrogen peroxide. In Experiment 6, methane oxidation is catalyzed by III, giving an amount of products equivalent to 49% conversion to oxygenates. Nine millimoles of methyl acetate, two millimoles of other oxygenates and 3 millimoles of CO 2 were fo1rned. In Experiments 5 and 6 a good material balance for methane disappearance and product fo1mation resulted. When an identical blank experiment with no catalyst or methane was carried out, two millimoles of methyl acetate and no other products were forrned from peracid decomposition, Experiment 7. When no catalyst was used and methane added, Experiment 8, three millimoles of methyl acetate and one millimole of other oxygenates

PAGE 63

48 were formed. Experiment 7 indicates that methyl acetate can arise from the solvent via peracid decomposition. Since the amounts of oxygenates produced are doubled when methane is added, Experiment 8, methyl acetate is produced by the reaction of the peracid with methane. However, in Experiment 6 when the ruthenium catalyst is present, the almost three-fold increase in oxygenates compared to Experiment 7 establishes the metal catalyzed path. All methane conversions given in Table 2-1 is based on the amount of methane consumed in the reaction. In the absence of a material balance for runs 1-4, an upper estimate results for conversions. Since peracid formation is not expected for oxidations with 0 2 trapping with the acetic acid-acetic anhydride mixture was investigated next with this oxidant. Oxidation of Methane with Molecular Oxygen The direct conversion of methane to methanol using molecular oxygen is even a greater challenge than using hydrogen peroxide as the oxidant. Table 2-2 provides results for experiments using 30 psig of 5% 0 2 in helium as the oxidant. The moles of oxidized products for111ed clearly exceed the moles hydrogen peroxide (2.5xl 0" 5 ) added to initiate the reaction by oxidizing m. After 48 hours (Experiment 9), methyl acetate (5 turnover numbers, 0.32 mmoles, 19% selectivity), and relatively large amounts of for1naldehyde (20 turnover numbers, 0.65 mmoles, 38% selectivity) methanol and methyl forrr1ate were obtained. GC analysis (TCD) shows carbon monoxide (9 turnover numbers, 0.19 mmoles, 11 % selectivity) and carbon dioxide (32 turnover numbers, 0.52 mmoles, 30% selectivity). At most a total of0.4 turnovers (2.5x10 5 moles) for methyl

PAGE 64

49 Table 2-2: Oxidation Results for Methane@ 75C with cis-[Ru(dmp) 2 (H 2 0) 2 ]{CF 3 S0 3 ) 2 Using H 2 0 2 and 0 2 Experiment Sel. to CO 2 Sel. To H202 CH 4 Numbe.-8 MeAcetate Formation Oxygenates Efficiency Conversion (%) (O/o) (%) (%) (Ofo) 9b 19.2 30.1 57.6 >100 16.8 10c 50.1 40.2 50.1 >100 6.3 11 0.0 0.0 0.0 0.0 0.0 12e 0.0 0.0 0.0 0.0 0.0 a. 6.6 x 10-s moles cis-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 20 ml acetonitrile, 20 ml glacial acetic acid, 20 ml Acetic Anhydride, 20 psig CH 4 (l.OxI0 2 moles), 3 5g Molecular Sieves (MS)., 1s 0 c, b Reaction time 48 hours 2.5 % H 2 0 2 (2.5xI0 5 moles). c Reaction time 24 hours 2.0 % H 2 0 2 (2.0xio-s moles) d Blank Experiment, performed in the absence of H 2 0 2 48 hours e. Blank Experiment, performed in the absence of catalyst 48 hours

PAGE 65

50 acetate could have resulted from the 2.5x 10 5 moles of H 2 0 2 assuming 100% peroxide efficiency Clearly this catalytic system utilizes 0 2 for the oxidation of methane at 75C. The total moles of products correspond to 1. 7 mmoles, corresponding to a 17% percent conversion of methane. The formation of methyl acetate indicates the trapping of methanol is occurring. The reaction of methane with 0 2 was carried out for 24 hours (Experiment 10). Less CO 2 and less methane conversion results. The products are methyl acetate ( 5 turnover numbers, 0.30 mmoles, 50% selectivity), carbon monoxide (3 turnover numbers, 0.06 mmoles, 10% selectivity) and carbon dioxide (16 turnover numbers, 0.26 mmoles, 40% selectivity). A total of0.63 mmoles are formed and the percent methane conversion was 6%. Since the same amount of methyl acetate forms in 24 or 48 hours, the extended time results in the oxidation of methanol that is in equilibrium with methyl acetate (Equation 2-2). Experiment 11 was perfor1ned to dete1n1ine if di-oxygen could catalyze the reaction without hydrogen peroxide to initiate formation of the ruthenium-oxo species After 48 hours no oxidation products were observed. The [Ru( dmp ) 2 (H 2 0) 2 ] 2 + catalyst precursor utilizes molecular oxygen in epoxidations 38 39 only after an initiation period in which an alkyl hydroperoxide is generate to oxidize the complex to the (IV) oxidation state. Upon oxygen atom transfer to the substrate from the (IV) oxidation state the (II) oxidation state for1ns, which reacts with the Ru (IV) to generate Ru (III). The (III) oxidation state reacts with molecular oxygen to regenerate the higher oxidation state metal-oxo species. 46 Since 0 2 will not form methylhydroperoxide from methane, the inactivity of the ruthenium (II) complex with 0 2 (Experiment 11), and the required

PAGE 66

51 initiation by H 2 0 2 sug g e s ts that a metal-oxo species generated from ruthenium (III ) by 0 2 is involved by either reacting with ruthenium (III ) to foxrn a metal-oxo species catalyzing the reduction of 0 2 to H 2 0 2 by methane or inhibiting radical chain terxnination steps Conclusions The use of the sterically hindered cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + complex to catalyze the oxidation of methane has been investigated. Using the poorly coordinating trifluoromethanesulf onate anion in place of hexafluorophosphate an improvement in reproducibility and catalytic activity of the complex has been demonstrated. This complex is a precursor to the proposed di-oxo catalyst that oxidizes methane to CO 2 with hydrogen peroxide at 75 C in acetonitrile. In order to stop the reaction at the methanol inter1nediate, an acetic acid/acetic anhydride mixture was chosen to trap methanol as methyl acetate. This ester which is not as readily oxidized as the alcohol increases the selectivity and inhibits the over-oxidation of methanol. In oxidations with hydrogen peroxide, control experiments indicate the f 01mation of peracetic acid from the solvent mixture of acetic acid/acetic anhydride. In the absence of methane the peracid decomposes to produce methyl acetate. When the reaction i s carried out in identical conditions under methane a doubling of the methyl acetate produced shows that methane is oxidized by peracid to fo11n methyl acetate. A substantial increase in methyl acetate for identical reaction conditions with metal complex added clearly demonstrates that metal catalyzed oxidation occurs. The material balance shows that the predominant source of the methyl acetate is methane with at most a small contribution from peracid decomposition

PAGE 67

52 The cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + complex is a catalytic precursor for the oxidation of methane with dioxygen in acetonitrile at 75C. It is necessary to initiate the reaction by oxidation of the complex with trace amounts of hydrogen peroxide. With molecular oxygen as the oxidant, a 6% conversion of methane and 50% selectivity to methyl acetate can be obtained after 24 hours of reaction.

PAGE 68

CHAPTER3 OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING A RUTHENIUM METAL-OXO CATALYST Introduction The transforrnation of saturated hydrocarbons into their oxygentated derivatives (i e. alcohols, aldehydes ketones and carboxylic acids) has been the subject of intense investigation over the past two decades. Alkanes given their great abundance offer a cost effective and ideal feedstock for an industrial process. 59 However the chemical inertness of saturated hydrocarbons make activation extremely difficult at mild conditions. The oxygenation products vital intennediates in many industrial processes are then converted into commercial products. In order for a process to be feasible on the industrial scale the catalyst must demonstrate two properties: selectivity towards the partial oxygenates and exhibit a specific regioselectivity. To achieve these requirements a commercial industrial application must maintain high temperatures and pressures. 6 Currently research efforts are being directed towards the development of new efficient catalytic systems which are able to oxygenate s aturated hydrocarbons under mild conditions using hydrogen peroxide and/or molecular oxygen. Reagent s with the capacity to oxidize paraffms and arylalkanes have been known for well over a century. 61 62 Two such compounds are chromyl chloride ( Cr0 2 -Cl :J and per111anganate ( Mn0 4 -). 6 1 62 However these oxidants are stoichiometric and a catalytic 53

PAGE 69

54 one is desired. Numerous researchers 63 64 65 have reported organic radical mechanisms via homogeneous or heterogeneous oxidation. The essential properties that allow these oxidants to oxidize a relatively inert C-H bond are not fully understood. A number of metal-oxo and metal-oxide surfaces perfo11n as reagents or catalysts for the oxidation of hydrocarbons, on industrial and laboratory scales. 1 2 6 6 6 7 ,6 8 Metallo enzyme sites also activate hydrocarbons, two well-documented examples being Cytochrome P-4506 9 and methane monooxygenase 47 (MMO). Current research in the area of hydrocarbon activation has utilized a variety of oxidants, catalysts and reaction conditions. Barton et al. 70 71 has developed the Gif system, as well as a number of variations (i.e. Barton-Git). Gif uses a pyridine / acetic acid solution, substrate, Fe catalyst and an oxidant, usually t-BuOOH. The mechanism of oxidation has been reported as proceeding through the involvement of free radicals. Que 72 has been investigating the use of di-iron complexes, in this case a high valent Fe 2 (-0) 2 moiety. This complex has been proposed as the key oxidizing species for methane monooxygenase, as well as for other non-heme di-iron enzymes . Ribonucleotide reductase and fatty acid desaturase are mentioned. the catalytic oxidation of higher linear and branched alkanes with hydrogen peroxide in acetonitrile at 75 C. The complex generates a large fraction of products oxygenated at the primary carbon position. An increase in selectivity towards the alcohol product is observed upon addition of CuC1 2 The effect of temperature on reactions with this catalyst

PAGE 70

55 is also investigated to detertnine the effect on overall hydrocarbon conversion selectivity distribution of ox y genates Experimental Materials and Methods RuCl 3 x H 2 0 2,9-dimethyl-1,10-phenanthroline LiCl NaCF 3 S0 3 hydrogen peroxide (35%) and pentane (99.9 % ) were all used as received from Aldrich. Ethylene glycol and acetonitrile were all used as received from Fisher Scientific ( ACS grade ). Acetonitrile was distilled over P 2 0 5 under dinitrogen and was stored over 4A activated molecular sieves. Ethane (99.9 %), propane (99 % ) butane (99 % ) and iso-butane ( 99.5% ) was purchased from Matheson and used as received. Physical Measurements UV-Vis measurements employed a Perkin Elmer lambda-6 spectrophotometer The pH measurements were made with a Fisher Accuroet model 630 pH meter. NMR spectra were recorded on a Varian VXR300 spectrometer F AB mass spectral data were obtained by Dr. David Powell (U.F. ) in a m-nitrobenzyl alcohol matrix. U F. Analytical Services perfo11ned elemental analyses. Synthesis of Compounds cis-Ruthenium(II) Bis( chloride)bis(2,9-dimethyl-l,10-phenanthroline) reported 4 0 Analysis: Calculated for C 28 H 26 N 4 0Cl 2 Ru ; C 55 45 ; H 4 29 ; N 9. 2 4. F o und : C 55.65 ; H 4.35 ; N 9.29.

PAGE 71

56 cis'-Ruthenium(II) Bis( aquo )bis(2,9-dimethyl-1,10-phenanthroline) bis(trifluoromethanesulfonate), Cis'-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 (II). A l .Og (1.7 mmol) portion of I is slowly dissolved in 150 ml of deionized H 2 0 under N 2 by heating at 50C for 30 minutes. After adding 50 ml of a saturated NaCF 3 S0 3 (aq), the solution cooled, placed in an ice bath for two hours and filtered. The precipitate is redissolved in solution is added and reprecipitated as above. Recrystalliz.ation and exchange of c1 for CF 3 SQ 3 is repeated until the filtrate affords a negative chloride test with AgN0 3 The product is dried in vacuum at 60C overnight. Analysis: Calculated for C 30 H 3 4 N 4 0 8 S 2 F 6 Ru, C, 42.25; H, 3.31; N, 6.57. Found: C, 41.40, H, 3.26; N 6.39 Oxidation Procedure The pressurized oxidations were carried out as previously described 4 9, in batch type hydrogenation reactors, Figure 2-5. The reaction mixtures were varied as described in table footnotes. Blank runs were perfor1ned omitting certain reactants and are also described in the text. Reaction temperatures were normally maintained at 75C unless stated otherwise. To remove air initially present, nitrogen gas is purged through the reactor, the apparatus is pressurized with the substrate, which is then released, and re pressurized to the desired pressure. The oxygenated products of the reaction were analyzed and quantified with a Hewlett-Packard 5890 Series 11 gas chromatograph equipped with an FID detector and outfitted with a 30m HP 50+ (50% Ph Me Silicone Gum; 1 m thickness). Helium was employed as the carrier gas. Carbon dioxide and carbon monoxide analysis were

PAGE 72

57 performed with a Varian 3700 gas chromatograph equipped with a TCD detector outfitted with a 15' Carboxen Column, 1 m thickness Helium was utilized as the carrier gas. Concentrations of the substrate and oxidized products were quantified using acetonitrile as an internal standard. Three chromatograms were measured for each sample using injection volumes of 0.1 ml for gas and 0.1 l for liquid samples. A typical reaction mixture consisted of 60 ml of acetonitrile, 40 psig total pressure for gaseous reactants, 1.6xl 0 4 moles of catalyst and 5ml of 35% hydrogen peroxide (5.0xl 02 moles). Experiments in the absence of oxidants were perfortned as blanks. The following defmitions describe te1ns used in the presentation of the results. Selectivity to any product is the moles of a given product divided by the total moles of all products formed expressed as a percent. The percent peroxide efficiency is the moles of H 2 0 2 needed to account for all oxidized products divided by the moles of H 2 0 2 consumed. The percent conversion of alkane is the moles carbon in the oxidized products divided by the moles of alkane added to the reactor. ~afety Precautions, the combination of molecular oxygen with organic compounds and solvents at elevated temperatures and pressures are potentially explosive. Extreme caution should be taken during the charging and disassembly of the experimental apparatus. Equipment which can generates sparks must be avoided, a safety shield and cooling of the batch reactor in an ice bath for 30 minutes prior to disassembly is recommended.

PAGE 73

58 Results and Discussion Alkane Oxidation Attempts to oxidize alkanes with hydrogen peroxide in H 2 0 as the solvent at 75C were unsuccessful after 48 hours with or without catalyst at solution pH values of 1 to 7. As suggested previously4 6 the Ru(VI) oxidation state is necessary for the activation of the C-H bond of alkanes. The [Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex II has been shown to be a potent catalyst in an acetonitrile solvent. 73 Consequently alkane oxidations were performed in acetonitrile. Each alkane ( ethane, propane, butane, iso-butane and pentane) was reacted in acetonitrile (60ml) with hydrogen peroxide (5.0x102 mole) and catalyst (l.6x10-4 moles) for 15 hours at 75C. The results are provided in Table 3-1 and Table 3-2. Propane, iso butane and pentane allowed us to dete1111ine catalyst activity in ter1ns of selectivity and regioselectivity. After 15 hours, a 20.8% conversion of propane was observed corresponding to 3.4 mmoles of oxidized products. The selectivities of the oxidized products obtained are of interest. Oxidation at the primary carbon position accounts for 65.1 % or 2.2 mmoles of the total oxidized products (1-propanol: 0.1 mmoles, 1.6% selectivity, and propionaldehyde: 2.1 mmoles, 63.6% selectivity). Trace amounts of propionoic acid were detected by gas chromatography analysis, however peak broadening and tailing make quantification difficult. Oxidation at the secondary carbon position was also observed, accounting for 34.9% or 1.2 mmoles of the total oxidized products (2-propanol:

PAGE 74

59 Table 3-1: Oxidation Results for Ethane, Propane and Butane @ 75C with cis [Ru(dmp) 2 (H20)](CF 3 S0 3 ) 2 using H 2 0 2 Oxidized Total Percent Substrate 1 Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) Ethanol: 0.08 (2.3) Ethane Acetaldehyde: 8.10 16.2 20.1 2.50 (69.3) Acetic Acid: 1.00 (28.4) 1-Propanol: 0.05 (1.6) 2-Propanol: 0 07 (1.9) Propane Propanal: 6.80 13 5 20.8 2.10 (63.6) Acetone: 1.20 (32.9) Propionoic Acid : Trace 1-Butanol: 0.08 (4.8) 2-Butanol: 0.94 (5.3) Butane Butanal : 5.40 10.7 26.2 0.10 (5.5) 2-Butanone: 0 31 (18.1) Butanoic Acid: 1.20 (66.3) a : 1 6xl 0 -4 mole cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 60ml acetonitrile, 5.0xl 0 2 mole s 35% H 2 0 2 40psi sub s trate (I 9x 102 moles ), I 5 hours at 75C b: Percent conversion is ba sed on the total of oxidized product s divided by the initial amount of sub s trat e.

PAGE 75

60 Table 3-2: Oxidation Results for !so-Butane and Pentane @ 75 C with cis [Ru( dmp )2(H20)](CF3S03)2 using H 2 0 2 Oxidized Total Percent Substrate 8 Products mmoles Peroxide mmoles Oxidant Efficiency (Selectivity) /so-Butanol: /so-Butane 3.90 (99.0) 3.90 7.8 /so-Butyl-ol: 0.04 (0.9) 1-Pentanol: 0.58 (5.7) 2-Pentanol: 1.20 (11.5) 3-Pentanol: Pentaneb 1.20 (17.1) 16.10 32.4 Pentanal: 2.20(21.7) 2-Pentanone: 1.90 (18.4) 3-Pentanone: 2.60 (25.7) Pentanoic Acid: Trace Percent Conversionc 20.5 23.4 a: l .6x l0 -4 mole c is-[Ru ( dmp) 2 ( H 2 0) ](CF 3 S0 3 ) 2 60ml acetonitrile, 5.0x lo 2 mole s 35% H 2 0 2 40psi substrate (1.9xlo2 moles) 15 hours at 75C. b: 2ml pentane (2. 0x I 02 moles) used c: Percent conversion is based on the total of oxidized product s divided by the initial amount of substrate.

PAGE 76

61 0.1 mmoles, 1.9% selectivity and acetone: 1.1 mmoles 32.9% selectivity). Activation of propane by this catalyst produces a selectivity ratio of primary:secondary carbon oxidation of2:l. A peroxide efficiency of 13.5% and 43 turnover numbers (TON s ) was obtained for this reaction. Ethane and butane were oxidized under identical conditions giving a 20.1 % and 26.2% conversion respectively The results are shown in Tables 3-1 and 3-2. Over oxidation of the hydroxylated products to the corresponding aldehyde ketone and carboxylic acid were also exhibited. As a result selectivity overall conversion of the alkane peroxide efficiency and catalyst lifetime are decreased by over-oxidation. !so-butane was investigated to demonstrate the ability of this catalyst to produce oxidation products at the tertiary and primary carbon position. After 15 hours of reaction 3.9 mmoles of oxidized products were obtained. Oxidation at the tertiary carbon accounted for 99.0 % (3.9 mmoles) of the oxidized products. Also detected were trace amounts (0.04 mmoles 0.9% selectivity) of isobutyl alcohol (2-methyl-1-propanol) demonstrating some oxidation at the primary position. Pentane can lead to oxidation at the C 1 C 2 and C 3 positions. This is also the fust liquid alkane investigated. After 15 hours a 23.4% conversion of pentane was observed accounting for 10 mmoles of oxidized products. Oxidation at the primary carbon position (C 1 ) afforded: 1-pentanol (0 6 mmoles 5.7% selectivity ), valeraldehyde (pentanal) ( 2.2 mmoles 21.7% selectivity) and a trace amount of valeric acid (pentanoic acid) once again quantification was difficult due to peak broadening and tailing. Oxidation at the secondary carbon (C 2 ) position produced: 2-pentanol ( 1.2 mmoles, 11.5% sel. ) and 2 pentanone (1.9 mmoles 18.4% sel.) Activation at the C 3 position produces: 3-pentanol

PAGE 77

62 (1.2 mmol, 17 1 % sel.) and 3-pentanone (2.6 mmoles, 25. 7% sel.). A total of 16 mmoles of oxidant was consumed in this reaction yielding a peroxide efficiency of 32.4% and 50 TON. Mechanism of Oxidation Two mechanisms, hydrogen abstraction or 'rebound'' and oxygen atom insertion have been proposed for the oxidation of hydrocarbons with cis-[Ru(dmp) 2 (H 2 0) 2 ] 2 + 4 0 1 4 These mechanisms are illustrated in Figures 3-1 and 3-2 respectively. The hydrogen abstraction or 'rebound'' mechanism produces an alkyl free radical. The free radical produced in the frrst step of the mechanism is the result of the cis{Ru( 0 ) 2 } 2 + moiety abstracting a hydrogen atom from the alkane to produce a reduced ruthenium-hydroxide species. The active metal center Ru(VI)=O functions as a free radical 40 which is capable of abstracting a hydrogen atom from the hydrocarbon. The second step proceeds with the transfer of the hy~oxo ligand to the fotmed radical to yield the hydroxy lated product. The catalytic cycle involves regeneration of the high valent metal center by oxidation with 0 2 or H 2 0 2 Other ruthenium complexes have been proposed to proceed through this or a similar mechanism. 75 76 7 7 The oxygen atom insertion mechanism has been proposed for the epoxidation of olefins. 40 78 7 9 It has also been reported that ruthenium-oxo complexes are able to epoxidize alkenes via a non-radical mechanism. 40 The first step in this mechanism for alkane oxidation is the insertion of the oxygen bound to ruthenium into the C-H bond of the substrate. This adduct f onns a three coordinate oxygen and a five coordinate carb o n

PAGE 78

0 11 Ru = O + H-CH 3 0 I I Ru + 63 0 11 Ru = OH + 0 11 Ru : Q 0 11 + Ru Figure 3-1: Proposed Hydrogen Atom Abstraction Mechanism for cis [Ru( dmp ) 2 (H 2 0) 2 ] 2 + Complex

PAGE 79

64 similar to CH 5 + obtained in an S 0 2 mechanism. The next step is scission of the C-H bond for1ning a coordinated alcohol followed by displacement of the alcohol from the ruthenium complex The final step is the regeneration of the high valent metal-oxo as previously described in the hydrogen abstraction mechanism. Results from this laboratory 40 73 indicate that the activation of alkane s with c is [Ru( dmp )O:J 2 + is proceeding via the hydrogen abstraction mechanism. Addition of CuCl 2 The majority of the oxidized products obtained are aldehydes ketones and carboxylic acids. In order to obtain increased amounts of the partially oxidized products its direct over-oxidation must be prevented. Addition of a metal chloride could offer a solution. Addition ofNiC1 2 to a methanol solution produces complexes with methanol bound to the nickel B O In this investigation formation of this metal-alcohol adduct was detern1ined by IR analysis. Coordination of the a)cohol produced in our oxidations to a metal could possibly retard over-oxidation and improve selectivity for the hydroxylated product . To deter111ine if an increase in selectivity towards the hydroxylated products can be obtained one mole equivalent of CuC1 2 (l .6x I 0-4) was added to the previously described reaction with all other conditions held constant. The results are shown in Table 3-3 and Table 3-4. After 15 hours at 75 C an 18.2% conversion of propane was observed in the presence of CuC1 2 The addition of CuC1 2 led to a 2.4% decreased in conversion. However an increase in alcohol selectivity to 6.1 % for 1-propanol and a 18 5 % for

PAGE 80

65 0 I I Ru=O 0 11 + Ru 0 I I Ru + 0 I I Ru:Q Figure 3-2: Proposed Oxygen Atom Insertion Mechanism for cis [Ru( dmp ) 2 (H 2 0) 2 ] 2 + Complex

PAGE 81

66 2-propanol is observed compared to selectivities of 1.6% and 1.9% when CuC1 2 is omitted. With the increase in alcohol selectivity a decrease in selectivity to the aldehyde and ketone result s A 61 0% selectivity to propionaldehyde and 14 4% selectivity for acetone is exhibited with the addition of CuCI 2 CuC1 2 addition leads to a decrease in the peroxide efficiency attributed to known peroxide decomposition by first row transition metal salts. 29 Increases in selectivity to the alcohol are also found for each of the alkanes studied. These results are presented in Table 3-3 and 3-4. To deter1nine the role of CuC1 2 a series of blank experiments were performed. When 60 ml of acetonitrile l 6xl0 4 moles CuC1 2 5.0x102 mole H 2 0 2 and l.9xlo 2 moles of propane were reacted for 15 hours at 75 C no oxygenates were observed suggesting CuC1 2 is not participating in the oxidation of propane. Experiments were performed to dete1n1ine the stability of alcohol under reaction conditions. Ethanol (1ml l .9xl 02 mol), acetonitrile (60ml) 1.6xl 0 4 moles catalyst and 5.0x102 moles H 2 0 2 were reacted for 15 hours at 75 C. A 77% decrease in ethanol and formation of acetaldehyde and acetic acid was observed. TCD gas chromatography analysis exhibited the trace amounts of carbon monoxide and carbon dioxide. The quantity of oxidized products corresponds to the decrease in moles of ethanol Analy s is of the resulting solution for peroxides with sodium metavanadate and iodometric titration showed none was present An identical experiment was performed with the addition CuC1 2 ( 1.6xl 0 4 mol ) to the reaction solution. After 15 hours the amount of alcohol present in the final reaction solution corresponded to only a 43 % decrease. Aldehyde and carboxylic acid also formed in decreased amounts according to ma s s balance calculati o ns

PAGE 82

67 Table 3-3: Oxidation Results for Ethane, Propane and Butane@ 75C with cis [Ru(dmp)2(H 2 0)](CF 3 S0 3 ) 2 and CuC1 2 using H 2 0 2 Oxidized Total Percent SubProducts mmoles Peroxide Percent stratea mmoles Oxidant Efficiency Conversionb (Selectivity) Ethanol: 0.18 (4.31) Ethane Acetaldehyde: 9.10 18.1 3.10 (74.7) Acetic Acid: 0.88 (21.0) 1-Propanol: 0.21 (6.1) 2-Propanol: 0.64 (18.5) Propane Propanal: 6.10 12.1 2.10 (61.0) Acetone: 0.49 (14.4) Propionoic Acid: Trace 1-Butanol: 0.26 (12.5) 2-Butanol 0.15 (7.1) Butane Butanal: 4.90 9.7 0.11 (5.3) 2-Butanone: 0.46 (21.8) Butanoic Acid: 1.10 (53.2) a: l.6xl0-4 mole cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 l.6xl0 -4 mole CuCl 2 2 H 2 0 60ml acetonitrile, 5.0x10 2 moles 35% H 2 0 2 40psi substrate (l 9xI0 2 moles) 15 hours at 75 C. 23.2 18.2 30 9 b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate

PAGE 83

68 Table 3-4: Oxidation Results for /so-Butane and Pentane @ 75 C with [Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and CuCI 2 using H 2 0 2 Oxidized Total Percent Substrate 1 Products mmoles Peroxide mmoles Oxidant Efficiency (Selectivity) Jso-Butanol: !so-Butane 3.50 (99.2) 3.50 7.1 /so-Butyl ol: 0.01 (0.8) 1-Pentanol: 0.94 (10.4) 2-Pentanol: 1.60 (17.5) 3-Pentanol: Pentaneb 2.70 (30 1) 13.10 25.5 Pentanal: 1.30 (13.8) 2-Pentanone: 1.30(13.7) 3-Pentanone: 1.30 (14.5) Pentanoic Acid: Trace a: l 6xJ0 -4 mole c is-[Ru(drnp) 2 (H 2 0)](CF 3 S0 3 ) 2 l 6xl0 -4 mole CuCl 2 2 H 2 0 60ml acetonitrile, 5.0x I 0 2 moles 35% H 2 0 2 40psi s ubstrate ( I 9x 10 2 moles ), 15 hours at 75 C b : 2ml pentane (2.0xl 2 moles) used. Percent Conversionc 18.8 20.9 c: Percent conversion is based on the total of oxidized products divided by the initial amount of sub s trate

PAGE 84

69 calculations. No peroxide was found in the resulting solution. These experiments confitm the role o f CuC1 2 in producing more alcohol in the oxidation of alkanes. In the ab s ence of the cataly s t and CuC1 2 a mixture ofH 2 0 2 ( 5.0x10 2 moles ) and ethanol (l.9x10 2 moles) in acetonitrile (60 ml) leads to a 21 % decrease of ethanol after 15 hours. Hydrogen peroxide an alcohol under these type of reactive conditions. When the reaction is carried out with CuC1 2 (l .6xl0 4 moles) added the hydrogen peroxide (5.0x102 moles) in acetonitrile (60 ml) oxidizes only 9% of the ethanol to acetaldehyde and acetic acid after 15 hours. This experiment demonstrates CuC1 2 either complexes the alcohol preventing oxidation or increases the rate of decomposition of the peroxide preventing oxidation of the products. The effect of increased amounts of CuC1 2 on selectivity to the alcohol was investigated using the oxidation of propane at 75 C. The results of these experiment s are given in Table 3-5 and illustrated in Figure 3-3. Successive additions of CuC1 2 increased selectivity towards the alcohol with the accomp~ying decrease in selectivity to the aldehyde and ketone With increased selectivity a decrease in overall conversion is observed. This is most likely attributed to peroxide decomposition by the increased copper concentration present. Without CuCl 2 a 20.80% conversion is achieved. Addition of 1 mole equivalent decreases the conversion decreases to 18 12 %. Additional CuC1 2 continues the trend reaching a 6.52% conversion with 5 mole equivalents added. Thi s decrease in conversion is the result of increased peroxide decomposition leading to a lower concentration of peroxide present to perfor1n catalysts and over-oxidize the products.

PAGE 85

70 Table 3-5: Oxidation Results for Propane@ 75 C with cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 Mole mmoles mmoles mmoles Total Percent Equivalents Alcohol Aldehyde Ketone mmoles Conversionb CuC1 2 Products 0 0.12 2.14 1.18 3.44 20.80 1 0.59 1.45 0.34 2.38 18.24 2 0 84 0.14 0.33 1.30 7.19 3 0.84 0.13 0.32 1.29 7.14 4 0.90 0.11 0.22 1.22 6.73 5 0.91 0.10 0.18 1.18 6.52 a: 1.6xl0 -4 mole cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 l.6x10-4 x mole eq. moles CuCl 2 2 H 2 0 60ml acetonitrile, 5.0xJo 2 moles 35% H 2 0 2 40psi substrate (l.9xI0 2 moles), 15 hours at 75 C b : Percent conversion is based on the total of oxidized products divided by the initial amount of substrate

PAGE 86

CJ) (1) 0 E E CJ) 0 :::J "O 0 L... a.. 3 00 2 50 2 00 1 50 1 00 0 50 71 Effect of CuCl 2 on Products with the cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + Precursor 17 00 14 00 11 00 8 00 0 00 .,__ ___ ...a.._ _______ __._ ___ __,_ ___ ___,i 5 00 0 1 2 3 4 5 Mole Equivalents CuCl 2 Alcohol --II-Aldehyde O Ketone Total Moles I Percent Conversion (2nd ) C 0 CJ) L... (1) > C 0 () C (1) (1) a.. Figure 3-3: Oxidation of Propane with cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2

PAGE 87

72 Effect of Temperature Increased temperature is expected to increase the rate on an oxidation reaction. Increase in substrate reactivity and solubility of the catalyst is also expected with an increase in reaction temperature. Disadvantages include an increased rate of peroxide decomposition. A series of experiments was perfonned at 25, 50, 75 and 100C with and without CuC1 2 were perfor1ned to detetmine the effect of temperature on this oxidation reaction. The results provided in Tables 3-6 and 3-7, are illustrated in Figures 3-4 and 3-5. Propane was again selected as the substrate for these experiments. A mixture of catalyst (1.6xl0 4 moles), acetonitrile (60 ml), H 2 0 2 (5.0x10 2 moles) and propane ( 1. 9x 10 2 ) were allowed to react for 15 hours at each temperature. Even at 25 C the ruthenium produces oxygenates from propane with a 3. 71 % conversion. The moles of aldehyde and ketone produced are greater than the moles of alcohol generated As the temperature increases, overall conversion increases along with an increase in the relative amounts of the aldehyde and ketone. The over-oxidation of alcohol present is also more rapid with the catalyst as the temperature is raised. An identical series of experiments were perforr11ed with the addition of 1 mole equivalent ofCuC1 2 (l.6xl0 4 mole). The result are given in Table 3-7 and graphically represented in Figure 3-5. The conversions obtained at high temperatures with CuC1 2 present are decreased when compared to those in the absence of CuCl 2 Again, a higher selectivity to the alcohol is obtained

PAGE 88

73 Table 3-6: Oxidation Results for Propane @Varying Temperatures with cis [Ru( dmp ) 2 (H20)](CF3S03)2 using H 2 0 2 Total Temp. Mmoles mmoles mmoles mmoles Percent (oc) Alcohol Aldehyde Ketone Products Conversionb 25 0.14 0.28 0.26 0.67 3.71 50 0.13 0.55 0.40 1.08 5.52 75 0.12 1.58 0.49 2.19 12.09 100 0.10 2.04 0.89 3.03 16.72 a: l.6xl0-4 mole cis-[Ru(dmp) 2 (H 2 0 ) ](CF 3 S0 3 ) 2 60ml acetonitrile, 5.0xI0 2 moles 35% H 2 0 2 40psi substrate {l.9xI0 2 moles) 15 hours b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate.

PAGE 89

74 Temperature Dependence of cis-[Ru(dmp) 2 (H 2 0) 2 ] 2 + 3.00 15 00 2 50 C "' 12 00 ~ Q) 0 ._ E 2 00 Q) > E C 9 00 0 (/) (.) 0 1 50 ::::, C Q) ""C 0 6 00 ._ Q) a.. 1 00 a.. 0 50 3 00 0 00 "-____ ....__ ____ __.__ ____ --'____ _... 0 00 0 25 50 75 100 Temperature (Degrees Celcius) Alcohol -a-Aldehyde Ketone Total Moles -ePercent C o nvers i on ( 2nd ) I Figure 3-4 : Oxidation of Propane @ Varying Temperatures with c is [Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 using H 2 0 2

PAGE 90

75 Table 3-7: Oxidation Results for Propane@Varying Temperatures with cis [Ru(dmp)2(H20)](CF 3 S03)2 and CuCl 2 using H 2 0 2 Total Temp. Mmoles mmoles mmoles mmoles Percent (OC) Alcohol Aldehyde Ketone Products Conversionb 25 0.23 0.28 0.40 0.91 5.03 50 0.50 0.99 0.43 1.93 9.89 75 0 58 1.45 0.54 2.58 14.23 100 0.66 1.70 0.67 3.03 16 72 a: l.6xl0 4 mole cis-[Ru(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 l.6xl0-4 mole CuCJ 2 2 H 2 0 60ml acetonitrile, 5 0xI0 2 moles 35% H 2 0 2 40psi substrate (I 9xI0 2 moles), 15 hours b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate

PAGE 91

C/J Q) 0 E E C/J 0 ::J "O 0 ,._ Q.. 3 00 2 50 2 00 1 50 1 00 0 50 76 Temperature Dependence of cis-[Ru(dmp) 2 (H 2 0) 2 ] 2 + with CuCl 2 (1 mol eq) 15 00 10 00 5 00 0 00 '---------1 ____ __., _________ _. 0 00 0 25 50 75 100 Temperature (Degrees Celcius) C 0 C/J ,._ Q) > C 0 (.) C Q) Q) Q.. 0 Alcohol -a-Aldehyde Ketone Total Moles Percent Conversion (2 nd ) Figure 3-5: Oxidation of Propane@Varying Temperatures with cis [Ru(dmp) 2 (H 2 0)](CF3S03) 2 and CuCI 2 using H 2 0 2

PAGE 92

77 Conclusions The oxidizing ability of the cis-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex to activate linear and branched saturated hydrocarbons has been investigated. Using hydrogen peroxide as the oxidant, production of the respective alcohols and their over-oxidized products are seen after 15 hours at 75C in an acetonitrile solution. An example of the oxidizing power of this system is shown by the 20.1 % conversion of ethane to ethanol (2.3% selectivity, 0.08 rnmoles), acetaldehyde (69 3% selectivity, 2.5 rnmoles) and acetic acid (28.4% selectivity, 1.0 mmoles). To deter production of the over-oxidized products and increase the selectivity of the hydroxy lated products addition of one mole equivalent of CuC1 2 was perforrned The addition of CuC1 2 illustrates a pronounced effect towards increasing the selectivity to the alcohols. Successive additions of CuC1 2 mole equivalents also exhibit a further increase in alcohol selectivity. Blank experiments perfor111ed also illustrate the ability of CuC1 2 to retard further oxidation of the alcohol. Additional blank experiments performed in the absence of catalyst produce small quantities of oxidized products demonstrating the catalyst is responsible for the majority of the oxidation. Temperature dependence studies were also perfo1111ed with and without the addition of CuC1 2 The effect observed is the ability of this catalyst to produce oxygenates of propane at 25C (3.2% conversion) to 100C (16.7% conversion). The addition of CuC1 2 to the oxidations perf orrned at variable temperatures also allowed for increased selectivity of the alcohol.

PAGE 93

CHAPTER4 SYNTHESIS AND CHARACTERIZATION OF IRON DIMETHYL PHENANTHROLINE COMPLEXES Introduction As previously discussed a high valent metal-oxo complex is required for the activation of alkanes. The cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + precursor is converted to c i s [Ru(0) 2 ) 2 + to attain this high oxidation state moiety 40 74 This oxidant effectively oxidizes alkanes with hydrogen peroxide and methane with hydrogen peroxide and dioxygen. 7 3 The success with this complex prompted research directed towards the synthe s is of analogous complexes in an attempt to prepare a more robust catalyst. Thi s complex should be able to utilize dioxygen without the need of a co-oxidant and be less expensive First row transition metals might produce a complex with these advantages therefor the synthesis of an iron based catalyst was investigated Iron containing complexes including heme and non-heme enzymes are reported to effectively oxidize alkenes and alkanes. 81 82 8 3 The development of catalytic s ystems to mimic a variety of biological systems has been the subject of intense investigation. Cytochrome P-450 peroxidases catalysases and high valent iron-oxo porph y rin complexes all involve a 2-electron oxidation producing the reactive intertnediate in hydrocarbon oxidation reactions. 69 8 4 78

PAGE 94

79 The use of iron porphyrins as model catalysts has allowed for further understanding of significant steps, which are involved in many enzymatic oxidation reaction mechaoisms. 85 More recently, Watanabe and Morishima 82 have demonstrated the use of a (TMP)Fem(RC0 2 ), where TMP is 5, 10, 15,20-tetramesityl porphyrin complex for the epoxidation of norborylene and a-methylstyrene at 78C with a variety of peracids. This research has suggested the O=FeIV(TMP) n-cation radical as being the active oxidant for this oxygenation reaction. Catalysts based on iron complexes in the absence of porphyrins have also been investigated. Que 83 has demonstrated the reactivity of (-oxo) di-ferric complexes with t-BuOOH for the activation of cyclic hydrocarbons in acetonitrile. The -oxo di-ferric complex was synthesized as an effort to model the dinuclear sites found in non-heme iron-enzymes. This [Fe(TPA) 2 0(0Ac)](Cl0 3 ) 3 where TPA is tris-(2-pyridyl)-methyl amine and OAc is acetate, complex is demonstrated to be a robust catalyst for the activation of cyclohexane at ambient temperature and pressure. The oxidation products produce~ from this reaction are cyclohexanol cyclohexanone and (t-butylperoxy) cyclohexane. Que has suggested the for1nation of the alcohol and ketone occur by catalyst initiated decomposition of t-BuOOH to afford a high valent metal-oxo complex via a heterolytic pathway. Additional research presented by Que 72 has been directed towards the further investigation of these non-heme iron centers. Furthermore Que has proposed a high valent Fe 2 (-0) 2 diamond core structure, which is believed to be the key oxidizing species of methane monooxygenase (MMO).

PAGE 95

80 Further research using iron coordinated ligand type complexes have also been reported. Sawyer 8 1 has reported a number of iron (II) complexes [FerrL ); Fe ( DP AH) 2 where DP AH 2 is 2 6-dicarboxylpyridine, Fen(P A) 2 where PA is picolinic acid and Ferr(bpy) 2 + where bpy is 2,2'-bipyridine. Each of these complexes with the addition of a reductant [DH 2 :PhNHNHPh] for example, are able to catalytically activate 0 2 (1 atm) for the hydroxylation of phenol and substituted phenol. Results provided indicate the mechanism of oxidation is proceeding via a Fenton like intermediate. The synthesis of an iron ''analogue'' based on the cis-[Ru( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex is the focus of our research. By altering the metal center it is anticipated a more robust catalyst will result. The ligand was not altered due to its ability to impart a cis geometry 40 with respect to the two remaining coordination sites when the metal is coordinated by two dmp ligands. This geometry, the result of steric hindrance created by the methy I groups a. to the nitrogen atoms results in the cis isomer being a more powerful oxidant when compared to the trans counterpart. 74 The lesser oxidizing power of the trans isomer, measured to be on the order of 35 kcal mo1 1 86 is attributed to the lower energy of the HOMO ,, 40 Additional benefits of this ligand include its ability to prevent the condensation of hydroxo or oxo bridged dinuclear species when two ligands are present on the metal center. Based on accounts reported in literature and the success of the cis[Ru( dmp ) 2 {H20 )2] 2+ complex, the synthesis of the iron analogues was attempted. Characterization of these newly for1ned complexes, as well as investigation for the

PAGE 96

81 production of a high valent iron-oxo species, for the eventual use as an oxidation catalyst, was also perf 01111ed. Experimental Materials and Methods FeC1 2 4 H 2 0, 2-9-dimethyl-l, 10-phenanthroline, NaCF 3 S0 3 and Li Cl were all used as received from Aldrich. Acetonitrile, acetone and 30% hydrogen peroxide (aq) were all used as received from Fisher Scientific (ACS Grade). Acetonitrile was distilled over P 2 0 5 and stored over 4A molecular sieves. Physical Measurements UV-Vis measurements employed a Perkin Elmer ]amhda-6 spectrophotometer, all spectra were background corrected. Infrared Spectroscopy (IR) analysis were all collected either as Nujol mulls or KBr pellets on a Nicolet 5DXB spectrometer and were background corrected. Nuclear Magnetic Resonance (NMR) measurements were recorder on a multinuclear Varian VXR 300 MHz spectrometer or a multinuclear Gemini 300 MHz spectrometer. All samples analyzed by NMR were performed in deutrated solvents with 1% w / w TMS. Electrochemical studies were perfonned with a PAR model 175 Universal Programmer connected to a model 173 potentiostat/galvanostat. The electrodes used were a platinum working and auxiliary electrode, with an Ag / AgCl reference electrode. The pH measurements were made with a Fisher Accumet model 630 pH meter FAB (+and ) mass spectral data were obtained by Dr. David Powell (U.F.) in a m-nitrobenzyl alcohol matrix. Single crystal X-ray diffraction analysis were perfortned by Dr. Kalih Aboud using a Nicolet diffractometer equipped with a graphite

PAGE 97

82 monochromated Mo-Ka radiation source. The Nicolet Structure Deterrnination Pac k a g e was used for data collection data recovery and structure elucidation. U.F. Analytical Services perforrned elemental analysis. Synthesis of Compounds Iron(II)bis( chloride )mono(2,9-dimethyl-1,10-phenanthroline ), [Fe( dmp )Cl 2 ] (I). A 2.25 g (12 mmol) of2 9-dimethyl-l 10-phenanthroline was dissolved in 100 ml of acetonitrile under nitrogen at 70 C. Next 1.25 g (6 mmol ) FeC1 2 4H 2 0 was added to the solution. Upon addition of the metal a precipitate is formed immediately. The resulting mixture is allowed to react for 30 minutes the solution cooled and filtered The product is then dried under vacuum at 60 C overnight. Analysis: Calculated for C 14 H 1 2 N 2 Cl 2 Fe: C, 50.29; H, 3.59 ; N, 8.38. Found: C 50.12; H 3.46; N 8.23. Cis-Iron(ll)bis(aquo)bis(2,9-dimethyl-1,10-phenanthroline)bis(trifluoro methanesulfonate), cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S9 3 ) 2 (II). This complex was synthesized using modification of a prior method. 8 6 A 4.5g (24 mmol) of 2 9-dimethyl1, 10-phenanthroline was added to 150 ml of deionized H 2 0 at 90 C and allowed to s tir vigorously for 20 minutes under nitrogen. Upon complete dissolution of the ligand 2 .5g (12 rnmol) ofFeCI 2 4 H 2 0 was added the resulting solution was then s tirred for 2 hours . The solution is then filtered (hot) and the filtrate immediately added dropwise to a chilled saturated NaCF 3 S0 3 (aq) solution (25ml H 2 0 / 9g NaCF 3 S0 3 ). The resulting precipitate is allowed to stand in ice for 2 hours and filtered. The product is dried under vacuum at 60 C overnight Analysis: Calculated for C 30 H 28 N 4 0 8 F 6 S 2 Fe: C 44 66 ; H 3 47 ; N 6 .9 4. Found: C, 44.87 ; H 3 33; N 7.01

PAGE 98

83 Results and Discussion Characterization A nwrtber of analytical techniques were utilized to characterize the complexes (I and II) synthesized in this chapter. The conditions employed for each spectroscopic technique are described within the experimental section of this chapter. Any variation in analysis will be described in the appropriate section. Elemental values obtained are provided at the end of each synthesis procedure. Single Crystal X-ray Diffraction Single crystal X-ray diffraction analysis was collected for the mono-2,9-dimethyl l, l O-phenanthroline (dmp) complex, Fe(dmp)Cl 2 (I) to provide structural info11nation and confirn1 results obtained from elemental analysis. A brown-orange needle of I (0.26 x 0.19 x 0.16mm) was mounted on the end of a glass capillary tube and analysis perfor1ned. The crystal structure and crystal data obtained are provided in Figure 4-1 and Table 4-1. Additional information for this structure is given in Tables 4-2, 4-5 and 4-6. In this structure iron has a four coordinate tetrahedral geometry with two coordination sites occupied by the nitrogen atoms of the dmp ligand and the remaining bound to the chlorine atoms leading to a 2 + oxidation state. The coordination of only one dmp ligand leads to the possibility of -oxo dimer forn1ation when this complex is oxidized in the presence of others.

PAGE 99

84 (.) 0 (.) Figure 4-1: Crystal Structure for Fe( dmp )Cl 2

PAGE 100

85 T able 4-1: Cryst al Data and Structure Refinement f or F e ( dmp ) Cl 2 Empirical formula Formula weight Temperature Wavelength Crystal system Space group Unit cell dimensions Volume Z Density ( calculated) Absorption coefficient F(OOO) Crystal size Theta range for data collection Limiting indices Reflections collected Independent reflections Absorption correction Max and min. transmission 335 01 173(2) K o.71073 A Orthorhombic Pnma a = 11.2265(7) A B = 7 4630(5) A C = 17 788(1) A 1490 3(2) A 3 4 1.493 Mg/m 3 1.356 mm1 680 a = go o ~=go o y = go o 0.26 x 0.19 x 0.16 mm 2 15 to 27.50 deg -5<=h<=15 -4<=k<=10 -24<=1<=18 5761 1841 [R(int) = 0 0451] Semi-empirical from psi-scans 0 9660 and 0.7631

PAGE 101

Table 4-1 (Cont'd) Refinement method Data I restraints / parameters Goodness-of-fit on F 2 Final R indices [1>2sigma(I)] R indices (all data) Extinction coefficient Largest diff. Peak and hole 86 Full-matrix least-squares on F"2 1823 / 0 / 145 1.083 R1 = 0.0343, WR2 = 0.0794 R1 = 0.0475, WR2 = 0.0912 0.0070(7) 0.346 and -0.323 e.A 3

PAGE 102

87 Table 4-2: Atomic Coordinates (x 10 4 ) and Equivalent Isotropic Displacement Parameters ( A 2 x 10 3 ) for Fe(dmp)Cl 2 X y z U(eq) Fe 2624(1) 2500 3878(1) 27(1) Cl 3133(1) -114(1) 3367(1) 42(1) N(1) 860(2) 2500 4289(1) 26(1) N(2) 2936(2) 2500 5050(1) 26(1) C(1) -151(3) 2500 3890(2) 33(1) C(2) -1268(3) 2500 4257(2) 43(1) C(3) -1333(3) 2500 5019(2) 39(1) C(4) -277(3) 2500 5451 (2) 28(1) C(5) 799(3) 2500 5057(2) 24(1) C(6) 1910(3) 2500 5461 (2) 24(1) C(7) 1892(3) 2500 6250(2) 29(1) C(8) 3011 (3) 2500 6613(2) 41 (1) C(9) 4032(3) 2500 6201(2) 43(1) C(10) 3977(3) 2500 5411(2) 34(1) C(11) -263(3) 2500 6258(2) 35(1) C(12) 775(3) 2500 6640(2) 35(1) C(13) -59(4) 2500 3051 (2) 49(1) C(14) 5085(3) 2500 4946(3) 46(1) Note: U(eq) is defined as one third of the trace of the orthogonalized Uij tens o r

PAGE 103

88 The bond lengths and angles for complex I are provided in Table s 4-3 and 4-4 respectively. An important criterion for the determination of a crystal structure s accuracy is the amount of uncertainty in the bond lengths and angles. The small Rv alue of 4.8% obtained for this structure in the crystal data indicates a great deal of certainty A bond length of 2 .111(3)A was obtained for the Fe-N(l ) bond and 2.114(3)A for the Fe N ( 2) bond, for an average of2.112A. These bond lengths are slightly larger than those obtained for the cis-[Ru( dmp ) 2 {H 2 0) 2 ] 2 + complex. In the ruthenium complex the four Ru N bond lengths are reported as 2.085, 2.092, 2.063 and 2.094A for an average length of 2.084A 8 7 Also of interest is the 79.3 N(l)-Fe-N(2) bite angle observed for the Fe(dmp)Cl 2 complex which is almost identical to the 78.9 angle exhibited by the N ( l ) Ru-N(2) and N(3 ) -Ru-N(4) in the cis-[Ru(dmp) 2 (H 2 0) 2 ] 2 + complex. Attempts to obtain a crystal structure for the cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex (II ) were unsuccessful Crystal growth was attempted in water ( acidic and basic conditions) acetonitrile ethanol methanol dimethylforamide (DMF) tetrahyrdofuran (THF) methyle!le chloride and propylene carbonate, however a suitable crystal could not be obtained. Crystal growth attempted in acetone led to the formation of small orange brown crystals which were submitted for single crystal X-ray diffraction. The structure collected illustrated the presence of two protonated dmp ligands which were not coordinated to iron The FeC1 2 and H 2 0 molecules were present in the crystal lattice indicating the instability of this complex. To allow for confi11nation of this complex ( II) an alternative spectroscopic method was needed

PAGE 104

89 Table 4-3: Bond Lengths [A] for Fe ( dmp)Cl 2 Fe-N(1) Fe-N(2) Fe-Cl#1 Fe-Cl N(1)-C(1) N(1)-C(5) N(2)-C(10) N(2)-C(6) C(1)-C(2) C(1)-C(13) C(2)-C(3) C(2)-H(2) C(3)-C(4) C(3)-H(3) C(4)-C(5) C(4)-C(11) C(5)-C(6) C(6)-C(7) C(7)-C(8) C(7)-C(12) C(8)-C(9) C(8)-H(8) C(9)-C( 10) C(9)-H(9) C(10)-C(14) C(11)-C(12) C(11 )-H(11) C(12)-H(12) C(13)-H(13A) C(13)-H(138) C(14)-H(14A) C(14)-H(148) 2.111 (3) 2.114(3) 2.2273(6) 2.2273(6) 1.338(4) 1.368(4) 1.334(4) 1.364(4) 1.414(5) 1.497(5) 1.357(6) 0.86(5) 1.412(5) 0.96(4) 1.396(4) 1 437(5) 1.440(4) 1 403(4) 1.412(5) 1.434(5) 1 361 (5) 0.99(4) 1.407(5) 0.95(5) 1.493(5) 1 348(5) 0.94(4) 0.97(4) 0.86(6) 0.98(4) 0.84(8) 0.89(5) Note: Symmetry transformations used to generate equivalent atoms: #1 x,-y+1/2 z

PAGE 105

Table 4-4: Bond Angles [ 0 ] for Fe(dmp)Cl 2 N(1 )-Fe-N(2) N(1 )-Fe-Cl#1 N(2)-Fe-Cl#1 N(1)-Fe-CI N(2)-Fe-CI Cl#1-Fe-CI C(1)-N(1)-C(5) C( 1 )-N( 1 )-Fe C(5)-N(1 )-Fe C(1 O)-N(2)-C(6) C(1 O)-N(2)-Fe C(6)-N(2)-Fe N(1 )-C(1 )-C(2) N(1 )-C(1 )-C(13) C(2)-C(1 )-C(13) C(3)-C(2)-C( 1) C(3)-C(2)-H(2) C( 1 )-C(2)-H(2) C(2)-C(3)-C(4) C(2)-C(3)-H(3) C(4 )-C(3)-H(3) C(5)-C(4)-C(3) C(5)-C(4)-C(11) C(3)-C(4)-C(11) N(1 )-C(5)-C(4) N(1 )-C(5)-C(6) C(4)-C(5)-C(6) N(2)-C(6)-C(7) N(2)-C(6)-C(5) C(7)-C(6)-C(5) C(6)-C(7)-C(8) C(6)-C(7)-C( 12) C(8)-C(7)-C( 12) C(9)-C(8)-C(7) 90 79.28(9) 112.48(3) 111.10(3) 112.48(3) 111.10(3) 122.33(4) 119.1(3) 127.7(2) 113.1(2) 118.9(3) 128.3(2) 112.9(2) 120.5(3) 118.1(3) 121.4(3) 120.6(3) 124(3) 116(3) 119.8(3) 124(2) 117(2) 117.0(3) 119.5(3) 123.5(3) 123.0(3) 117.1(3) 119.9(3) 123.2(3) 117.6(3) 119.1(3) 116.4(3) 119.8(3) 123.8(3) 120 2(3)

PAGE 106

Table 4-4 (Cont d) C(9).;C(8)-H(8) C(7)-C(8)-H(8) C(8)-C(9)-C( 10) C(8)-C(9)-H(9) C(1 O)-C(9)-H(9) N(2)-C(1 O)-C(9) N(2)-C(1 O)-C(14) C(9)-C(1 O)-C(14) C(12)-C(11)-C(4) C(12)-C(11 )-H(11) C(4)-C(11)-H(11) C( 11 )-C( 12)-C(?) C(11)-C(12)-H(12) C(7)-C(12)-H(12) C(1 )-C(13)-H(13A) C(1 )-C(13)-H(138) H(13A)-C(13)-H(138) C(1 O)-C(14)-H(14A) C(1 O)-C(14)-H(148) H(14A)-C(14)-H(148) 91 124(2) 116(2) 120.1(3) 124(3) 116(3) 121.2(3) 117.7(3) 121.1(3) 120.9(3) 124(2) 115(2) 120.8(3) 122(2) 117(2) 113(4) 112(2) 113(3) 112(5) 119(3) 105(4) Note: Symmetry transformations used to generate equivalent atoms : #1 x,-y+1 /2,z

PAGE 107

92 Table 4-5: Anisotropic Displacement Parameters (A 2 x 10 3 ) for Fe(dmp)Cl 2 U11 U22 U33 U23 U13 Fe 30(1) 26(1) 24(1) 0 4(1) Cl 47(1) 30(1) 48(1) -8(1) 14(1) N(1) 26(1) 26(1) 26(1) 0 -3(1) N(2) 21(1) 29(1) 27(1) 0 -1 (1) C(1) 34(2) 29(2) 35(2) 0 -13(1) C(2) 27(2) 51(2) 49(2) 0 -16(2) C(3) 20(2) 44(2) 51(2) 0 -2(2) C(4) 21(1) 29(2) 34(2) 0 0(1) C(5) 23(1) 23(1) 26(1) 0 -3(1) C(6) 20(1) 25(2) 27(1) 0 0(1) C(7) 28(2) 34(2) 26(2) 0 -2(1) C(8) 36(2) 61(3) 25(2) 0 -7(1) C(9) 27(2) 63(3) 38(2) 0 -9(1) C(10) 20(2) 43(2) 37(2) 0 -3(1) C(11) 29(2) 44(2) 31(2) 0 9(1) C(12) 39(2) 45(2) 20(2) 0 5(1) C(13) 51(2) 60(3) 36(2) 0 -19(2) C(14) 22(2) 62(3) 53(2) 0 6(2) Note: The anisotropic displacement factor exponent ta.lees the fortn: -2 1t 2 [ h 2 a 2 Ul 1 + ... + 2 h k a* b* U12] U12 0 0(1) 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

PAGE 108

93 Table 4-6: Hydrogen Coordinates (x 10 4 ) and Isotropic Displacement Parameters (A 2 x 10 3 ) for Fe(dmp)Cl 2 X y z U(eq) H(2) -1880(41) 2500 3965(24) 52(13) H(3) -2070(34) 2500 5293(21 ) 35(10) H(8) 2993(31) 2500 7167(21) 32(9) H(9) 4811 (41) 2500 6416(24) 54(12) H(11) -1016(35) 2500 6489(21) 40(11) H(12) 803(33) 2500 7186(22) 39(10) H(13A) -747(55) 2500 2833(32) 94(20) H(138) 461 (35) 1536(50) 2870(21) 92(13) H(14A) 5698(69) 2500 5213(41) 129(28) H(148) 5197(41) 1613(64) 4623(24) 124(19)

PAGE 109

94 F AB Analysis The analytical method chosen for further characterization of complex II was fast atom bombardment (F AB) (positive ( +) and negative (-)). F AB mass spectroscopy spectral analysis permits the determination of a complex's atomic weight including the identification of cations (F AB + ) or anions (F AB-) specifically. The results obtained from F AB mass spectrometry of complex II are illustrated in Figure 4-2. The F AB + spectrum contains a number of peaks each of which were correlated to fragmentation of the title complex yielded peaks at 507.2 amu (cis-[Fe(dmp) 2 (H 2 0) 2 ] 2 + ), 299.0 amu ([Fe(dmp)(H 2 0) 2 ] 2 + ) and 209.1 amu (dmp + ). The inability to identify the complete complex is due to its overall neutral charge. To further characterize complex II, FABspectroscopy was perfonned. In this spec~ a peak at 155 amu which corresponds to the CF 3 S0 3 anion was found. IR Analysis Although there is limited variability for the coordination of the two dmp ligands, IR spectroscopy was employed to further investigate coordination of the ligands in complex II. The IR spectra obtained are illustrated in Figure 4-3. Sauvage and Collin88 have reported the infrared spectral bands for the cis-[Ru( dmp ) 2 (H 2 0) 2 ] 2 + complex. Since

PAGE 110

I 100 80 60 40 20 107.1 7 100 154.1 I 209.l 281. J 200 299.0 I 301.1 335.1 I 300 95 430.1 400 lx20 607 2 638.3 621.2 519.3 I I 727.5 545.3 I 500 600 699.5 7 700 I 728.5 753.6 ,800

PAGE 111

SAMl'LE 67.0 j 10 r 1 -, ID .. ..... ID ..... I I I I 96 en N 11 N u, CD I ..... in CD 0 ID--4 ID ..... N ,.,, '" 0 .. ... N .. .... I ... r r ~ r -1~0 r NAVENUMBERS N tO ..... ,.. 0 ..... ,.,, N ..... I I 0 CD IO I N in CD ,.,, co I f 0 0 .. '" II') CD 01 en ID in i sJo l 400 0

PAGE 112

97 metal center, the spectra obtained should be similar with only a small shift in the bands for the iron complex. The following bands were reported for the ruthenium complex: 1630, 1580 1500 1280 1220 1050 860 730,560 530 and 500 cm '. IR analysis performed on complex II exhibited the following bands: 1622, 1564 1464 1257, 1153 1032, 862 771 571 543 and 518 cm-I. Upon comparison of the two sets observed a small shift is exhibited suggesting the iron complex synthesized is analogous in coordination to the ruthenium complex. NMR Analysis A number ofNMR experiments were perfor1ned to further characterize the newly spectrum of complex I Fe(dmp)Cl 2 in deutrated acetonitrile is provided in Figure 4-4. In this spectrum, a single methyl resonance at 2.0 ppm is expected since both ancillary methyl groups on the phenanthroline ligand are chemically equivalent. In the aromatic region 7 to 9 ppm the presence of two doublets (7.6 and 8.3 ppm) and one singlet ( 7.8 ppm) ar~ found from the coordination of only one dmp ligand . difficult. The use of I H NMR spectroscopy would allow for further deter111ination of the dmp ligand geometry about the metal center. Spectra could not be obtained in deutrated water acetonitrile and methylene chloride due to the slight solubility of the complex Greater solubility of this complex (II) occurs in acetone. The resulting solution in deutrated acetone was passed through a syringe filter to remove any undissolved

PAGE 113

9 &~ & (\J M 00 00 8 OI ID (\J & 00 ID I'I'00 I' ll) I'98 7 N N N I\J 1 Figure 4-4: 1 H NMR Spectrum for Fe(dmp)Cl 2 1/) ..,. I\J Cl> Cl> CD CD -e Cl>

PAGE 114

99 particulates, and added to the NMR sample tube. This NMR spectrum is shown in Figure 4-5. Two non-equivalent methyl peaks, the result of each methyl being chemically non equivalent, at 2.0 and 3.1 ppm indicate the expected cis geometry. However a detailed spectrum in the aromatic region was not obtained. The expected 3 sets of AB systems (3.4; 5,6; 7,8) characteristic of a di-substituted phenanthroline are not observed because of broadening of the peaks in the aromatic region. This broadening is attributed to iron (III) impurities in the complex. Several 1 H NMR spectra on different batches of purified complexes each produced the same result. Upon review of the literature the synthesis and crystal structure for the Fe(dmp) 2 (NCS) 2 H 2 0 complex was found, which proceeds through a membrane 89 In one compartment FeC1 2 4 H 2 0 (1 mmol) and dmp (1 9 mmol) are mixed into a 30 % H 2 0 I 70% ethanol solution. The other comparttnent contains an aqueous solution of 15 ml NaSCN (2.4 mmol) Crystal for111ation occurs at the interface after several days Single crystal analysis confir1ned the synthesis of this compound and is provided in Figure 4-6.8 9 In this structure the octahedral iron center is bound to the four nitrogen atoms of the two dmp ligands and the two nitrogen atoms from the thiocynate ligands in a cis orientation with respect to the two thiocynate ligands. The dmp ligands also impart a distortion on the octahedral metal center which is greater than the 76.1 angle observed for the Fe(phen) 2 (NCS) 2 complex where phen is 1,10-phenanthroline.89 The phen anthroline complex gives a trans orientation producing a much larger N-Fe-N angle. The smaller N-Fe-N angles, the result of the ligands trying to remain on different planes

PAGE 115

10 9 rin M
PAGE 116

101 phcr\ 1 S (I) C( 16) Cll41 C( I ~) ClZl> C.(241 c,lo, phtn 2 C(2) S(21 Figure 4-6: Crystal Structure for cis-Fe(dmp) 2 (NCS) 2 8 9

PAGE 117

102 exhibited by the cis-Fe(dmp) 2 (NCS) 2 complex are attributed to the methyl groups on this ligand Thus inhibiting fonnation of the trans isomer. The average bite angle (N-Fe-N) for the dmp complex is 73.0 ; identical to that obtained for the Fe(dmp)Cl 2 complex (I) This angle is also smaller than those observed for iron complexes with other non substituted ligands. Such examples are Fe(phen) 2 (NCS) 2 (76.1 ) 89 Fe(bpy) 2 (NCS) 2 (75.1 )90 and Fe(bt) 2 (NCS) 2 (73.8). 91 The substituted ligand forces a distortion about the metal center. The effect of distortion is the result of the inability of the ~.vo ligands to reside on the same plane. As the ligands are forced to opposite planes the N-Metal-N bite angle becomes smaller. Heber related the Fe-N bond distances to crystal field arguments. The two Fe-N cs distances observed in this complex average out to 2.067 A and the four Fe-N 0 MP bond distances average out to 2.27 A. The Fe-N 0 MP bond distance average is reported to be the longest average reported in literature for an iron (II) complex with a bi-dentate nitrogen donor ligand.90 9t,92, 93 This elongated iron-nitrogen bond explains the difficulty we encount~red in crystal growth. Heber attributes the abnonnally elongated Fe-N to steric crowding from each dmp ligand Each dmp ligand is bending away from the other causing elongation of the bond. In addition a large dihedral angle (153.5 is observed between each dmp ligand, much larger than the 86.9 angle observed in the Fe(phen) 2 (NCS) 2 complex. The deviation of Fe from the dmp plane suggests then-back donation from the ligand to the iron metal center should be significantly reduced when compared to the unsubstituted Fe(phen) 2 (NCS) 2 complex.

PAGE 118

103 IR NMR and F AB + spectral analysis confirms the synthesis of the cisHigh Valent Iron-Oxo Formation Studies The generation of a high valent iron-oxo species must occur in order for these complexes to be efficient catalysts for the activation of alkanes. The characterization of a such species in this case cis-[Fe( dmp ) 2 (0)(H 2 0)] 2 + or cis-[Fe( dmp ) 2 (0) 2 ) ] 2 + was attempted with Ultraviolet-Visible spectroscopy The formation of a high valent iron (V)-oxo complex with a porphorin ligand by oxidation of an iron (III) complex with hydrogen peroxide has been reported by Larpent 9 2 Equation 4-1 illu strates this reaction. The newly fo11ned Fe M =O species is then able to hydroxylate the alkane producing an alcohol and the reduced Fe(III) complex. Upon addition of hydrogen peroxide the Fe M =O complex is regenerated closing the catalytic cycle. (Equation 4-1 ): V -Fe=O / + R I -C-H I V -Fe : =O / + I -C-OH I complex problems were encountered. An UV-VIS spectrum was taken with 1.6x 10 -4 moles of complex ( II) dissolved in 60 ml of acetonitrile. Next hydrogen peroxide was

PAGE 119

104 added in 4.0x10 5 mol aliquots until a total of 3.2xl0 4 moles had been added with a UV VIS spectrum taken after addition of each aliquot. The full range of spectra acquired is illustrated in Figure 4-7. It was anticipated upon addition of each hydrogen peroxide aliquot a spectral change would be observed, resulting in f or1nation higher oxidation state species. These new species would allow for identification in change of the oxidation state of the complex. Analysis of these spectra illustrates minimal formation of other species. Analysis of the initial spectra (faint yellow solution) resulted in a peak at 276nm and a small shoulder at 295nm. Upon addition of each aliquot of hydrogen peroxide the solution became orange brown ( dark) in color resulting in an a small shift and increase in absorbance for each peak. After complete addition of the hydrogen peroxide, the peak at 276nm has become split into two smaller peaks. Analysis of each peak was unable to provide any additional infonnation Although this UV-VIS experiments perfor1ned resulted in the inconclusive evidence for the for1nation of a high valent iron-oxo. Results published by Larpent 92 have demonstrated the for1nation of such iron-oxo compound with the addition of hydrogen peroxide. Que 8 3 has also reported the generation of an iron-oxo species upon addition of t-BuOOH. In our situation the resulting spectra for the newly for1ned iron oxo species could be similar to that for the cis-[Ru( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex. To investigate the ability of the cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex to react with hydrogen peroxide the following experiment was performed. To 60 ml of acetonitrile l.6xl0 4 moles this complex was added followed by addition of 3.2xl0 4 moles of

PAGE 120

105 4 0 3 5 3 0 2.5 A 2 0 1 5 1 0 0 5 o o L---.---.-----r----..--~---.---...----.---.---_,:_===:==:::::=:::;:==:;:::=:::::;=~~~ ~~~~~ 240 2lO 260 270 280 290 300 310 320 330 340 3,0 360 370 380 390 400 410 420 430 440 45< NM

PAGE 121

106 hydrogen peroxide. Upon addition of the hydrogen peroxide effervescence was exhibited, the result of hydrogen peroxide decomposition which may led to for1nation of the desired high valent iron-oxo species. The solution was allowed to stir for IO hours at room temperature, at this point an iodometric titration of the solution was taken. The results obtained indicated an 88% decrease in the peroxide concentration due to decomposition. This result suggests the ability of our complex to be oxidized by H 2 0 2 however the inability to identify this newly for1ned species still exist. An identical set of UV-VIS experiments were also perfor1ned for the Fe( dmp )Cl 2 complex. The experiment for the stepwise addition of hydrogen peroxide produced the spectra, which is illustrated in Figure 47. In these spectra for1nation of the high valent iron-oxo species did not occur. What did result was the forn1ation of a new peak (392nm) after addition of one hydrogen peroxide aliquot as well as a shift of the original peak (333 to 369nm). Upon additional aliquots a small shift of the new peak is observed. This peak could possibly be a mono-oxo iron species, [Fe(dmp)H 2 0(0)] 2 +. However exact confmnation of this could not be obtained. The experiment for the generation of a high valent iron-oxo also suggested formation of such species. Iodometric titration results exhibited a 95% decrease of hydrogen peroxide after ten hours at room temperature. A difficulty in quantification of any iron-oxo species generated is attributed to the uncertainty of the amount of peroxide lost to decomposition. Conclusions The synthesis and characterization of two additional ''analogues '' Fe( dmp )Cl 2 (I)

PAGE 122

1 07 3 5 3 0 2.15 A 2.0 1 5 1 0 0 5 0 1 L-~--.-------.---_;_::::::;:======;:== pa---~ --,----, 300 3S O soo sso NM 600 6SO 700 7SO 800 Figure 4-8: UV-VIS Spectra for the Addition of H 2 0 2 to Fe(dmp)Cl 2

PAGE 123

108 complex I by X-ray crystallography analysis exhibits iron in a tetrahedral geometry complexed by only one dmp ligand. Further analysis of this complex by IR, F AB +, F AB, and NMR were also perf or1ned. Although a crystal structure for complex II was not obtained, characterization was deter1nined using F AB +, F AB, IR and NMR spectroscopy. Complex (II) has iron in an octahedral geometry, complexed by two dmp ligands, which enforce the desired cis geometry. Attempts to generate and characterize high valent iron-oxo species for the two newly synthesized complexes were also attempted, however data obtained does not allow for conclusive evidence to be drawn at this time

PAGE 124

CHAPTERS OXIDATION OF ALKANES WITH HYDROGEN PEROXIDE USING AN IRON METAL-OXO CATALYST Introduction A substantial amount of interest for use of non-heme complexes as catalysts for the activation of alkanes has occurred in recent years. These complexes, similar to methane monooxygenase or P-450, serve as models to the reactivity of non-heme iron centers in enzymes 70, 9 4 9 5 9 6 97 9 8 Of these complexes, those described as being the most active no11nally employ nitrogen based pyridine ligands. These ligands are preferred due to the enhance electrophillicity they provide to the metal center. Some examples of complexes which utilize the pyridine based ligands include Bartons's 70 "Gif'' catalyst, Fontecave's 95 [Fe 2 0(byp) 4 ] 4 +/ t-BuOOH complex and Que's 97 Fe(TPA) / t-BuOOH system where TP A is tris(2-pyridylmethyl)amine. The newly synthesized iron complexes: however hydrogen peroxide is the oxidant used in our catalytic system. All of the above mentioned catalytic systems have precursors metal complexes which have the ability to generate a metal-peroxide inter1nediate or a high valent iron-oxo species. Either of these species is necessary for the successful functionalization of an alkane as shown in Equation 5-1. 109

PAGE 125

110 (Equation 5-1) RH + LFeII 1 -00R or LFe v = o __ ,., R-OH + LFem for the activation of methane at 75 C in an acetic acid/acetic anhydride s olvent mixture. To further demonstrate the ability of this complex to activate methane oxidation experiments using molecular oxygen were also performed at ambient and high pres s ures. Furthern1ore, complex (II) and the [Fe(dmp)Cl 2 ] (I) precursor were also investigated for the activation of C 2 -C 5 linear and branched alkane s with hydrogen peroxide at 75 C in acetonitrile. Although significant selectivity profiles to products were achieved attempts to further increase selectivity to the hydroxylated product with addition of a mole equivalent of a metal chloride was also perfortned The effects of additional mole equivalents and temperature dependence on the catalyst s activity as well as mechanistic considerations were also investigated Experimental Materials and Methods FeC1 2 4 H 2 0 2-9-dimethyl-1,10-phenanthroline NaCF 3 S0 3 and LiCI were a ll used as received from Aldrich. Acetonitrile acetone and 30 % hydrogen peroxide ( aq)

PAGE 126

111 were all used as received from Fisher Scientific (ACS Grade). Acetonitrile was distilled over P 2 0 5 and stored over 4A molecular sieves. Physical Measurements The pH measurements were made with a Fisher Accumet model 630 pH meter. U .F. Analytical Services perfor1ned elemental analysis. Synthesis of Compounds Iron(II)bis( chloride )mono(2,9-dimethyl-l,l 0-phenanthroline ), [Fe( dmp )Cl 2 ] (I). A 2.25 g (12 mmol) of 2,9-dimethyl-1,10-phenanthroline was dissolved in 100 of acetonitrile under nitrogen at 70C. Next, 1.25 g (6 mmol) FeC1 2 4H 2 0 was added to the solution. Upon addition of the metal a precipitate is for1ned immediately. The resulting mixture is allowed to react for 30 minutes, the solution cooled and filtered. The product is then dried under vacuum at 60C overnight. Analysis: Calculated for C 14 H 12 N 2 Cl 2 Fe: C, 50.29; H, 3.59; N, 8.38. Found: C, 50.12; H, 3.46; N, 8.23. Cis-Iron(II)bis(aquo)bis(2,9-dimethyl-1,10-phenanthroline)bis(trifluoro methanesulfonate), cis-(Fe(dmp) 2 {H 2 0) 2 ](CF 3 S0 3 ) 2 (II). This complex was synthesized using modification of a prior method. 86 A 4.5g (24 mmol) of 2,9-dimethyl1, 10-phenanthroline was added to 150 ml of deionized H 2 0 at 90C and allowed to stir vigorously for 20 minutes under nitrogen. Upon complete dissolution of the ligand, 2.5g (12 mmol) of FeCl 2 4 H 2 0 was added, the resulting solution was then stirred for 2 hours. The solution is then filtered (hot) and the filtrate immediately added dropwise to a chilled saturated NaCF 3 S0 3 (aq) solution (25ml H 2 0 / 9g NaCF 3 S0 3 ). The resulting precipitate is allowed to stand in ice for 2 hours and filtered. The product is dried under vacuum at

PAGE 127

112 60C overnight. Analysis: Calculated for C 30 H 2 8 N 4 0 8 F 6 S 2 Fe: C, 44.66; H, 3 47; N, 6.94. Found: C, 44.87; H, 3.33; N, 7.01. Oxidation Procedure The pressurized oxidations were carried out as previously described 49 in glass, batch hydrogenation reactors. Figure 2-5 provides an illustration of such batch reactor. The reaction mixtures were varied as described in the table footnotes. Blank runs omitting certain reactants and solvent components are also described in the tables. Oxidations with oxygen use 30 psig of a 5% 0 2 in helium mixture, to remain outside the explosion limits, and 20 psig alkane. Reaction temperatures normally were maintained between 75 to 77C. To remove air initially present, nitrogen gas is purged through the reactor, the apparatus is pressurized with the substrate, which is then released, and re pressurized to the desired pressure Oxidation of Methane The oxygenated products of the reaction were analyzed and quantified with a Hewlett~Packard 5890 gas chromatograph equipped with a F ID detector and outfitted with a 30m Alltech RSL 160 column (5m thickness). Helium was employed as the carrier gas Carbon dioxide and carbon monoxide analyses were performed with a Varian 3700 gas chromatograph equipped with a TCD detector outfitted with a 15' Carboxen Column. Methane was quantified by gas chromatography (TCD) using N 2 as an internal standard. Three chromatograms were measured for each sample using injection volumes of 0.1ml for gas and 0.1 for liquid samples

PAGE 128

113 The following definitions describe terms used in the presentation of the results. Selectivity to an y product is the moles of a given product divided by the total moles of all products f or1ned expressed in percent. The percent CO 2 produced is moles of CO 2 divided by the moles of all products. The selectivity to oxygenates is the moles of H 2 CO CH 3 0H and CH 3 C(O)OCH 3 divided by the total moles of products in percent. Traces of CH 2 (0CH 3 ) 2 HCOOCH 3 are formed, but not quantified. Percent peroxide efficienc y is the moles of H 2 0 2 needed to account for all oxidized products including CO 2 divided by the moles of H 2 0 2 consumed. The percent conversion of CH 4 is the moles carbon in the oxidized products divided by the moles of CH 4 added to the reactor. Oxidation of Higher Alkanes The oxygenated products of the reaction were analyzed and quantified with a Hewlett-Packard 5890 Series II gas chromatograph equipped with an FID detector and outfitted with a 30m HP 50 + (50% Ph Me Silicone Gum; 1 m thickness ) Helium was employed as the carrier gas Carbon dioxide and carbon monoxide analysis were perfor1ned with a Varian 3700 gas chromatograph equipped with a TCD detector outfitted with a I 5 Carboxen Column 1 m thickness. Heljum was utilized as the carrier ga s. Concentrations of the substrate and oxidized products were quantified using acetonitrile as an internal standard. Three chromatograms were measured for each sample using injection volumes of 0.1 ml for gas and 0.1 for liquid samples A typical reaction mixture consisted of 60 ml of acetonitrile 40 psig total pressure for gaseous reactants l .6xl 0 -4 moles of catalyst and 5ml of 35 % hydrogen peroxide (5.0xl 0 2 moles) E xperiments in the absence of oxidants were perfortned as blanks

PAGE 129

114 The following defmitions describe terrns used in the presentation of the results. Selectivity to any product is the moles of a given product divided by the total moles of all products f or1ned expressed as a percent. The percent peroxide efficiency is the moles of . H 2 0 2 needed to account for all oxidized products divided by the moles of H 2 0 2 consumed. The percent conversion of alkane is the moles carbon in the oxidized products divided by the moles of alkane added to the reactor. Safety Precautions, the combination of molecular oxygen with organic compounds and solvents at elevated temperatures and pressures are potentially explosive. Extreme caution should be taken during the charging and disassembly of the experimental apparatus. Equipment which can generates sparks must be avoided, a safety shield and cooling of the batch reactor in an ice bath for 30 minutes prior to disassembly is recommended. Results and Discussion Oxidation of Methane with H 2 0 2 Although two iron analogues have been synthesized, the cis[F e( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 (II) complex was chosen for activation of methane studies. This was done to allow for a direct comparison to the results obtained for the cis[Ru( dmp ) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 catalyzed oxidation of methane under similar conditions. The results obtained from this oxidation study are provided in Table 5-1. Experiment 1 proceeded for 24 hours, after which 12 mmol of methyl acetate (69% selectivity) was produced, the largest quantity obtained to date. Carbon dioxide (4 mmol, 24% selectivity) and the corresponding intennediate oxygenates (i.e.

PAGE 130

115 Table 5-1: Oxidation Results for Methane @ 75C with cis-[Fe(dmp ) 2 (H 2 0) 2 ] ( CF 3 S0 3 ) 2 Using H 2 0 2 Experiment MeCOOMe CO 2 Total CH 4 Number8 mmol,( 0 /o) mmol,( 0 /o) Oxygenates Consumed mmol,( 0 /o) (%) 1 a,b 12,(69) 4, ( 24 ) 13,(76) 72 2 a,b c 6 (50) 4 ,( 31) 9,(69) 60 3 a.b d 0 (0) 5,(99) Trace,(!) 20 4 a.e 11 (71) 3 ,( 20) 13,(80) 70 5 a. t 10 (73) 2 ( 16) 12 (84) 67 6a l 8 5 (54) 3 (32) 6,(67) 42 6 6 x 10 5 mole s c is-[Fe(dmp ) 2 (H 2 0 ) 2 ](CF 3 S0 3 ) 2 catalyst, 5 ml 30% H 2 0 2 (5 0xI0 2 mole s). The s olvent mixture is 20 ml glacial acetic acid, 40 ml acetic anhydride initial m e thane pre ss ure was 40 psig corresponding to 23 millimoles 3.5g 4A molecular sieve s ( MS) were u s ed The percent products are based on total products s een from all s our c e s b. Reaction time 24 hours The percent product s are based on total products from a ll s ource s. c. 0 1 mJ concentrated H 2 S0 4 added. d Solvent mixture 60ml acetic acid e. Reaction Time 12 hours The percent products are based on total product s from all s ource s. f Reaction Time 4 hour s. The percent products are based on total product s from all s our ces. g. Molecular s ieve s omitted

PAGE 131

116 formaldehyde fortnic acid and carbon dioxide) in a quantity corresponding to 1 mmol (7% selectivity) were also produced. Furthermore a 72 % conversion of methane was obtained exhibit~ng the robust oxidizing power of this iron analogue (II) when compared to the ruthenium catalyst. The continued ability of the solvent system to trap methanol as methyl acetate with this iron-dmp analogue is also demonstrated. In an effort to enhance the esterification reaction 0 1 ml of concentrated H 2 S0 4 was introduced In addition to a larger quantity of ester produced, the acidic environment may influence the redox potentials of this catalyst. The results provided in Experiment 2 of Table 5-1 exhibit the decreased activity of the catalyst when exposed to acidic conditions. Overall a 60% conversion was obtained a decrease when compared to 72% obtained in the absence of acid. The increased acidity also exhibits a detrimental effect on the trapping reaction as evidenced by the decrease in selectivity to the methyl ester (50% selectivity). A subsequent increase in selectivity towards carbon dioxide (31 % selectivity) also results. However, the quantity of carbon dioxide obtained in both Experiments 1 and 2 remains at 4 mmol. The decrease in the selectivities obtained provides evidence for the decreased efficiency of the trapping reaction. A decrease in the quantity of methyl acetate obtained (6 mmol) is also exhibited a result of decreased catalytic activity Experiment 3 performed in the absences of trapping agents was allowed to react in 60 ml of glacial acetic acid for 24 hours. This resulted in the mineralization of methane to carbon dioxide and an overall conversion of 20 %. A selectivity proftle of 99% for CO 2 and 1 % for the intertnediate oxygenates was exhibited demonstrating the need for trapping the alcohol as an ester

PAGE 132

117 To increase the selectivity of methyl acetate the reaction time was decreased to 12 hours The result Experiment 4, is an increase in methyl acetate selectivity from 69% in Experiment 1 to 72%. However, a decrease in the quantity of methyl acetate obtained (from 12 to 11 mmol) results, the function of a decrease in reaction time. When comparing the 24 and 12 hour runs, the quantity (4 and 3 mmol) and selectivity (24 and 20%) of CO 2 obtained exhibits the ability to alter reaction conditions to produce a selectivity towards a desired product. An increase in production of the intermediate oxygenates from 1 mmol for Experiment 1 to 2 mmol for Experiment 4 was also demonstrated. The shorter reaction time had a minimal effect on the overall conversion, only resulting in a decrease of2% (from 72 to 70%). To further investigate the effect of decreased reaction time, Experiment 5 was allowed to proceed for four hours. As a result, the anticipated increase in selectivity towards methyl acetate (73%) when compared to the 71 % obtained for the 12 hour run (Experiment 4) was achieved The lesser reaction time also increases the quantity of intermediate oxygenates produced from 2 to 3 mmol. The quantity (2 mmol) and selectivity (20%) of carbon dioxide obtained was also favorable, as well as a the minimal decrease in the overall conversion of methane to 67% which was observed. Experiment 6 was perfortned in the absence of 4A molecular sieves with a reaction time of four hours The results obtained indicate a need for a dehydrating environment, which facilitates the esterification reaction. Although methyl acetate (5 mmol, 54% selectivity) is produced, the q11antity is decreased from the 10 mmol and 73% selectivity obtained with the sieves present. The over-oxidation of methanol to carbon

PAGE 133

118 dioxide (3 mmol, 32% selectivity) is also observed. Also affected by the absence of the molecular sieves is a decrease in overall conversion to 42%. Mechanism for Oxidation of Methane Preliminary mechanistic data suggests the activation of methane with the cis [Fe( dmp )2(H20)2](CF 3 S03)2 complex is parallel to the mechanism suggested for the cisreaction conditions present in both systems, as well as the established literature which suggests ruthenium and iron as proceeding through identical reaction mechanism when each are perfon11ed under analogous conditions. The for1nation of a peracid in the ruthenium system was demonstrated and explained in Chapter 2, as well as its ability to oxidize methane in the absence of the catalyst precursor Therefore, the role of the metal complex must be explained. As with the ruthenium catalyst a decrease in overall conversion is exhibited when the iron complex is added to the reaction system This related decrease could be attributed to the metal complex facilitating decomposition of the peracid, which is generated in situ, along with decomposition of the hydrogen peroxide oxidant. However, further investigation is warranted. Oxidation of Methane with 0 2 catalyst with hydrogen peroxide, initiated the investigation of this catalyst's ability to activate methane with molecular oxygen. A series of experiment provided in Table 5-2,

PAGE 134

119 Table 5-2: Oxidation Results for Methane@ 75C with cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 Using H 2 0 2 and 0 2 Experiment Sel. To CO 2 Sel. To H202 CH 4 Number MeAcetate Formation Oxygenates Efficiency Conversion (0/4) (o/o) (0/4) (Ofo) (%) 7a,b 36.2 62.8 37.2 >100 5.9 8a,c 47.4 52.4 47.6 >100 4.5 9ct.e 78.1 18.1 79.9 >100 9.5 1oa.e 0.0 0.0 0.0 0.0 0.0 11 a,t 0.0 0.0 0.0 0.0 0.0 a. 6 6 x 10 5 moles cis-[Fe(dmp} 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 20 ml glacial acetic acid 40 ml acetic anhydride 20 psig CH 4 (l.Oxt0 2 moles), 30 psig HeOx (l OxI0 2 moles), 3.5g molecular sieves (MS), 75 C. b. Reaction time 24 hours 2 0 l 30% H 2 0 2 (2.0xlo -s moles) c Reaction time 12 hours 2.0 l 30% H 2 0 2 (2 0xto 5 moles). d. Parr bomb reactor used. 250psi CH 4 (6 6xt0 2 moles), 250psi 0 0 (6.6xlo 2 moles), 500 p s i He (l.8xlo moles), 1 3 x 10-4 moles cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 20 ml glacial acetic acid, 40 ml acetic anhydride, 3.5g molecular sieves (MS), 100 C. e. Performed in the absence of H 2 0 2 24 hours. f Blank experiment, performed in the absence of catalyst 24 hours

PAGE 135

120 system investigated in Chapter 2 Once again reaction parameters (reaction time solvent and gas mixtures) were held constant allowing for direct comparison of results obtained. Table 5-2 illustrates the results for the direct conversion of methane to methanol using molecular oxygen (5% 0 2 in He) as the oxidant. As with the ruthenium analogue the quantities of products obtained exceed the amount of oxidant (2.0xl0 5 ) provided to initiate the iron complex (II) which suggests oxygen is contributing to the production of oxygenates. After 24 hours (Experiment 7), a 5.9% conversion of methane is obtained producing a 36.2% selectivity to methyl acetate. The presence of deep oxygenates (carbon dioxide) 62.8 % selectivity was also detected. A total of 6 0xl 0 4 moles of oxidized products were obtained an amount which exceeds the 2.0xl 5 moles of oxidant supplied by hydrogen peroxide. Clearly demonstrating the ability of this complex to utilize molecular oxygen in the oxidation process. Experiment 8 was perfor1r1ed as an attempt to further increase selectivity of methyl acetate After 12 hours a 4.5% conversion of methane to oxygenates was observed. A selectivity profile of 47 4% for methyl acetate and 47 6 for carbon dio x ide was achieved exhibiting the ability to allow for control of reaction condition to alter product selectivity It was apparent one of limiting factors is the solubility of each gas in this solvent matrix To allow for an improved s olubility oxidation experiments were performed in a high-pressure environment To do this a Parr High-Pressure apparatus was employed. This configuration allows for the use of high pressures of oxygen and methane as well as being able to operate at higher temperatures. Helium was also introduced to serve as a dilutant allowing ourselves to remain within the safety limits for a hydrocarbon-oxy g en

PAGE 136

121 mixture. The reactor body, head and stirrer were constructed out of titanium for corrosive resistivity. To allow for a further increase in gas solubility, a novel stirrer was used. The stirrer referre~ to as the gasperator is designed to allow for use of the convective flow which exists over the liquid in the head gas space. This stirrer is equipped with vertical slits in a bored out shaft and holes in the stirrer blades which are connected to the hollow shaft The vertical slits in the shaft, which are above the liquid level, draw the gas from the headspace into the shaft, where it is then passed through the blades. The gas is then released from the blades as bubbles and then disperses through the solvent, thus improving solubility of the gas. In order for this stirrer to function properly a revolution rate of 500 rpm must be attained. An additional benefit of the increased stirring rate is a more favorable interaction between the catalyst and gaseous substrate. Experiment 9, performed in the Parr Bomb used 250 psi of methane (6.6xlo 2 moles), 250 psi 0 2 (6.6xlo2 moles), 500 psi helit (l.8xI0 1 moles), 1.3xl0 4 moles cis [Fe(dmp)2(H20) 2 ](CF 3 S0 3 ) 2 (II), 20 ml of glacial acetic acid and 40 ml of acetic anhydride. After 1 hour of reaction time, a 9 .5% conversion of methane to oxygenates was observed. This oxidation allowed for a high selectivity to methyl acetate to be achieved (78.1 %), while limiting the amount of carbon dioxide (18.1 % selectivity) to be produced. With the oxidant being molecular oxygen a lower concentration of hydrogen peroxide is present within the system, therefore subsequent oxidation of the methanol to carbon dioxide does not occur as readily allowing for a higher selectivity of methyl acetate to be observed. This result is extremely encouraging from the standpoint of the minimal reaction time and low temperature necessary to achieve a large product selectivity and relatively high overall conversion with molecular oxygen as the oxidant.

PAGE 137

122 Experiments 10 and 11 detail blank experiments perfor1ned. A detailed explanation of each experiment is provided in Chapter 2. Alkane Oxidations with cis-[Fe(dmp) 2 .ili 2 ill 2 ](CF 3 S03)2 -Attempts to oxidize C 2 -C 5 saturated hydrocarbons with hydrogen peroxide in deionized water at 75C were unsuccessful after 48 hours with and without catalyst at pH solution values of 1 to 7. The inability to activate the Fe(II) center to a higher oxidation state, which is necessary for C-H bond activation, can be explained in a manner similar to that for the cis-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 precursor described in Chapter 2. As a result, alkane oxidation experiments were performed in acetonitrile. Each alkane ( ethane, propane, butane, iso-butane and pentane) was reacted in acetonitrile (60ml) with hydrogen peroxide (5.0xl 02 mole) and catalyst (l.6x10-4 mole) for 15 hours at 75C. The results obtained are provided in Tables 5-3 and 5-4. Again propane iso-butane and pentane allow ourselves to dete1111ine the catalyst's activity in ter1ns of selectivity and regioselectivity. After 15 hours, a 33.2% conversion of propane was obtained corresponding to 6.3 mmoles of oxidized products. Of note is the selectivity profile of the products produced. Oxidation at the primary carbon position accounts for 45.4% or 2.9 mmoles of the total oxidized products: 1-propanol (0.1 mmoles, 1.8% selectivity) and propionaldehyde (2.8 mmoles, 43.6% selectivity). Trace amounts of propionoic acid were detected, however peak broadening and tailing made quantification difficult. Oxidation at the secondary carbon was also observed, totaling to 54.6% selectivity or 3.4 mmoles of the oxidized products: 2-propanol (0.8 mmoles, 13.1% selectivity) and acetone (2.6 mmoles 41.5

PAGE 138

123 Table 5-3: Oxidation Results for Ethane, Propane and Butane@ 75C with cis [Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 using H 2 02 Oxidized Total Percent Substratea Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) Ethanol: 0.15 (2.20) Ethane Acetaldehyde: 17.8 35.7 36.2 2.47 (36.0) Acetic Acid: 4.25 (61.8) 1-Propanol: 0.11 (1.80) 2-Propanol: 0.83 (13.1) Propane Propanal: 11.7 23 3 33.2 2.77 (43.6) Acetone: 2.59 (41.5) Propionoic Acid: Trace 1-Butanol: 0.15 (6.30) 2-Butanol: 0.33 (14.3) Butane c Butanal: 4.74 9.48 24.2 0.35 (15.4) 2-Butanone: 0.85 (36.9) Butanoic Acid: 0.62 (27.1) a: l 6x10 -4 mole [Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 60ml acetonitrile 5.0xI0 2 moles 35% H 2 0 2 40psi substrate ( I 9x 10 2 moles), 15 hours at 7 5 C. b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate. c : 20psi (9 .5x 10 3 moles) butane used .

PAGE 139

124 Table 5-4: Oxidation Results for !so-Butane and Pentane @ 75C with cis [Fe(dmp) 2( H 2 0)](CF 3 S0 3)2 using H 2 0 2 Oxidized Total Percent Substrate Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) /so-Butanol: !so-Butane 3.0 (97.1) 3.09 6.18 19.5 !so-Butyl ol: 0.09 (2.90) 1-Pentanol: 0.36 (3.90) 2-Pentanol: 0.94 (10.0) 3-Pentanol: Pentanec 1.19 (12. 7) 16.2 32.4 47 .8 Pentanal: 1.63 (17.5) 2-Pentanone: 2.58 (27.6) 3-Pentanone: 2.64 (28.3) Pentanoic Acid: Trace a: 1 .6 xl0 4 mole [Fe(drnp) 2 ( H 2 0) ](CF 3 S0 3 ) 2 60ml acetonitrile, 5.0x10 2 moles 35% H 2 0 2 40psi substrate ( l .9x I 2 mole s), 15 hours at 7 5C. b : Percent conversion is based on the total of oxidized products divided by the initial amount of substrate. c: 2ml (2.0x10 2 moles) pentane used

PAGE 140

125 mmoles ). A selectivity ratio of primary:secondary carbon oxidation of 1: 1 was achieved by this catalyst. A peroxide efficiency of 23.3% and 48 turnover numbers (TON's ) was obtained for this reaction. Ethane and butane were also oxidized under the previously described conditions accounting for a 36.2% and 24.2% overall conversion respectively. These results are also provided in Table 5-3. As with the other catalytic system, over-oxidation of the products to the corresponding aldehyde ketone and carboxylic acid is also exhibited. The deep oxidation of the desired alcohol results in a decrease in alcohol selectivity overall conversion peroxide efficiency and catalyst lifetime. Oxidation of iso-butane with this catalyst resulted in the production of oxidation products at the primary and tertiary position. After 15 hours of reaction 3 .1 mmole s of oxidized products were obtained corresponding to a 19 .5% conversion. Oxidation at the tertiary carbon accounts for 97 1 % ( 3.0 mmoles) pf the total products. Trace amounts (0 09 mmoles 2.9% selectivity) of 2-methyl-1-propanol was also detected demonstrating the ability of this catalyst to activate the primary carbon position. Pentane the first liquid alkane investigated, was activated by this catalyst to lead to oxidation products at the C 1 C 2 and C 3 carbon position. After 15 hours a 47.8 % conversion was obtained, accounting for 9 .3 mmoles of oxidized products. Primary carbon oxidation (C 1 ) led to production of: 1-pentanol (0.3 mmoles 3.9% selectivity ), valeraldehyde (pentanal) (1.6 mmoles 17.5% selectivity) and trace quantities of valeric acid (pentanoic acid ), which was not quantified due to peak broadening and tailing. Oxidation at the s econdary carbon position (C 2 ) position afforded: 2-pentanol ( 0.9 mmoles 10.1 % selectivity) and 2-pentanone (2.6 mmoles 27.6% selectivity ) C 3 carbon

PAGE 141

126 activation produced 3-pentanol (1.2 mmoles 12. 7% selectivity) and 3-pentanone (2.6 mmoles, 28.3% selectivity). A total of 16 mmoles of oxidant was consumed yielding a peroxide efficiency of32.4 % and accounting to 51TON s . Alkane Oxidation with [Fe(dmp)Cl 2 } saturated hydrocarbons (C 2 -C 5 ) in deionized H 2 0 were also unsuccessful under a number of conditions. This resulted in oxidation reactions to proceed in an acetonitrile sol v ent matrix. Reaction conditions are identical to those described in the previous section. Results obtained for alkane oxidation with the [Fe(dmp)Cl 2 ] catalyst precursor are detailed in Tables 5-5 and 5-6. Ethane and butane oxidation resulted in a 36 7% and 22.3% conversion respectively. In this reaction mixture subsequent over-oxidation of the hydroxylated product was also observed. Result s detailing the effect of over-oxidation for these substrates are provided in Table 5-5. After 15 hours, a 18.8 % conversion of propane was observed, corresponding to 3.6 rnmoles of oxidized products Products resulting from oxidation at the primary carbon position accounted for I. 7 mmoles or a 48.2% selectivity: 1-propanol ( 0.01 mmoles 2.5% selectivity) and propionaldehyde: (1 7 mmoles 45.7% selectivity). Trace amounts of propionoic acid were also observed but not quantified. Secondary carbon oxidation products were also observed totaling 1.9 mmoles and 51.8% selectivity : 2propanol (0.7 mmoles 18.9% selectivity) and acetone (1.2 mmoles, 32.9% selectivity ). A selectivity ratio of primary : secondary carbon oxidized products of 1: 1 was also

PAGE 142

' 127 Table 5-5: Oxidation Results for Ethane, Propane and Butane@ 75 C with Fe(dmp)Cl 2 using H 2 0 2 Oxidized Total Percent Substratea Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) Ethanol: 0.13 (1.80) Ethane Acetaldehyde: 16.8 33.6 36.7 3.85 (55.3) Acetic Acid: 2.99 (42.9) 1-Propanol: 0.01 (2.50) 2-Propanol: 0.68 (18.9) Propane Propanal: 6.47 12.9 18.8 1.68 (45.7) Acetone: 1.21 (32.9) Propionoic Acid: Trace 1-Butanol: 0.09 (10.7) 2-Butanol: 0.10 (12.1) Butane c Butanal: 5.48 11.0 22.3 0.19 (22.5) 2-Butanone: 0.31 (37.5) Butanoic Acid: 1.43 (17.2) a: l.6x10-4 mole Fe(dmp) 2 Cl 2 60ml acetonitrile, 5 0xI0 2 moles 35% H 2 0 2 40psi substrate (1.9xI0 2 moles), 15 hours at 75 C. b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate. c : 20psi (9.5x10 3 moles) butane used.

PAGE 143

128 Table 5-6: Oxidation Results for /so-Butane and Pentane @ 75 C with Fe(dmp)Cl 2 using H 2 0 2 Oxidized Total Percent Substratea Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) /so-Butanol: !so-Butane 2.91 (99.1) 2.94 5.86 15.4 /so-Butyl ol: 0.03 (0.90) 1-Pentanol: 0.15 (2.20) 2-Pentanol: 0.56 (8.00) 3-Pentanol: Pentane c 1.47 (20.9) 11.8 23.7 36.9 Pentanal: 1.61 (22.9) 2-Pentanone: 1.97 (28.2) 3-Pentanone: 1.25 (17.8) Pentanoic Acid: Trace a : l 6xl0 4 mole Fe(dmp) 2 Cl 2 60ml acetonitrile 5 0x10 2 moles 35 % H 2 0 2 40p s i substrate (l.9xI0 2 moles) 15 hours at 75 C. b: Percent conversion is based on the total of oxidized products divided by the initial amount of sub s trate c: 2ml (2 0xio 2 moles) pentane u s ed

PAGE 144

129 observed for this iron analogue. A peroxide efficiency of 12.9% and 39 TON's was also obtained for this oxidation. Jso-butan e oxidation by this catalyst also resulted in the production of oxidized products at the primary and tertiary carbon positions. At the end of 15 hours, a 15.4% conversion was obtained accounting for 2.9 mmoles of oxidized products. Oxidation at the tertiary carbon accounts for the majority of products, identical to that observed for the selectivity. Limited quantities of 2-methyl-1-propanol were also detected, accounting for 0.03 mmoles of the total products (0.09% selectivity). Pentane oxidation with the [Fe( dmp )Cl:J catalyst precursor also occurred. After 15 hours of reaction, a 36.9% conversion was demonstrated totaling 7.01 mmoles of oxidized products. Primary carbon (C 1 ) oxidation (1.8 mmoles, 25.1% selectivity) led to the production of: 1-pentanol (0.2 mmoles, 2.2% selectivity), valeraldehyde (pentanal) (1.6 mmoles, 22.9% selectivity) and trace amounts of valeric acid (pentanoic acid) (not quantified). Oxidation at the secondary carbon (C 2 ) position (2.5 mmoles, 36.2o/o selectivity) was also observed: 2-pentanol (0.6 mmoles, 8.0% selectivity) and 2pentanone (2.0 mmoles, 28.2% selectivity). Products arising from oxidation at the C 3 position (2.8 mmoles, 38.7% selectivity) were 3-pentanol (I .5 mmoles, 20.9% selectivity) and 3-pentanone (1.3 rnmoles, 17.8% selectivity). A total of 11.8 mmoles of oxidant was consumed as well as the catalyst providing 35 TON's for this reaction. Upon comparison of the results obtained for the activation of alkanes with the two iron derivatives (Complexes I and II) and the ruthenium analogue, the mono dmp complex, [Fe(dmp)Cl 2 ] (I), is the least active. Nearly identical results in ter1ns of

PAGE 145

130 activity, selectivity regioselectivity and overall conversion were obtained for the two complexes that exhibited cis geometry. The decreased activity of the [Fe ( dmp)Cl 2 ] precursor may be attributed dimerization thus leading to the lack of available sites need for oxidation. Additional factors include the metal center being less electrophillic due to the presence of one ligand and also the instability of this complex which may lead to an increase in complex degradation. Mechanism for Higher Alkane Oxidation The ''rebound '' mechanism has been proposed for the activation of higher alk~es for both iron derivatives This free radical pathway is constant with reports published in literature for substrate activation with iron complexes under similar conditions. In the hydrogen abstract mechanism '' Rebound ', an alkyl free radical is produced. The alkyl free radial is the result of a high valent iron-oxo species abstracting a hydrogen from the substrate molecule. This results in formation of an iron-hydroxide Feiv-OH, s pecie s The hydroxyl ligand is then donated to the radical alkyl species to product the hydroxylated species. The catalytic cycle is then completed with the regeneration of the Feu 1 species to Fe v = O with molecular oxygen or hydrogen peroxide. Collins 99, Que 98 and Groves 8 5 have suggested similar mechanisms. Other supporting evidence for this proposed mechanism is obtained upon analysis of final reaction mixture when radical initiators and inhibitors have been added. Upon addition of AIBN 15 mole equivalents (a free radical initiator, azo-bis( isobutyronitrile )) to the reaction mixture no change in catalyst activity was observed. With addition of 15 mole equivalents of BQ (benzoquinone a free radical inhibitor ), no

PAGE 146

131 oxidation products were observed within the first 3 hours. After this time period which allows for consumption of the radical inhibitor product fonnation is then observed. Based on these results, additional evidence is provided to further suggest radical formation is involved in the induction steps necessary for alkane activation with these iron catalysts. An additional mechanism which cannot be ignored is the Fenton Mechanism. In this mechanism hydrogen peroxide can be decomposed by an iron complex into a hydroxyl radical and hydroxyl anion as well as the oxidized species of the iron complex. The hydroxyl species generated which is extremely reactive only second to elemental fluorine may be able to partake in the oxidation reaction. Resulting in the production of oxygenated products. The effect of this mechanism in our system was not studied extensively, so additional experimentation is needed Addition of CuC1 2 As exhibited in Chapter 2 the addition of CuC1 2 to the reaction mixture demonstrated a pronounced effect on the ability to retain the alcohol and retard further oxidation of the desired product. To determine if this increase in alcohol s electivity could also be obtained with the two iron derivatives, one mole equivalent ( l .6x 10 -4) of CuC1 2 was added to each of the previously described reactions. All other conditions were held constant allowing for a direct comparison. Tables 57 and 5-8 provide the results obtained with the c is-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 precur s or while Tables 5-9 and 5-10 exhibit the results obtained with the [Fe(dmp)Cl 2 ] complex.

PAGE 147

132 Table 5-7: Oxidation Results for Ethane, Propane and Butane @ 75C with cis[Fe(dmp)2(H 2 0)](CF 3 S0 3 )2 and CuC1 2 using H 2 0 2 Oxidized Total Percent Substratea Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) Ethanol: 0.17 (2.20) Ethane Acetaldehyde: 17.7 35.4 43.1 6.45 (78.9) Acetic Acid: 1.55 (18.9) 1-Propanol: 0.67 (14.9) 2-Propanol: 1. 71 (38.4) Propane Propanal: 10.7 21.4 34.4 2.26 (25.4) Acetone: 1.90 (21.3) Propionoic Acid: Trace 1-Butanol: 0.15 (6.30) 2-Butanol: 0.33 (14.3) Butane c Butanal: 4.74 9.48 24 2 0.35 (15.4) 2-Butanone: 0.85 (36.9) Butanoic Acid: 0 62 (27.1) a: l 6x10-4 mole [Fe(dmp ) 2 (H 2 0)](CF 3 S0 3 ) 2 1 6xl0 -4 mole CuC1 2 60ml acetonitrile 5.0x10 2 moles 35% H 2 0 2 40psi s ubstrate (1.9xio 2 moles) 15 hours at 75 C b : Percent conversion is based on the total of oxidized products divided by the initial amount of substrate c : 20psi (9 5x I 0 3 moles) butane used

PAGE 148

133 Table 5-8: Oxidation Results for /so-Butane and Pentane@ 75C with cis [Fe( dmp ) 2 (H 2 0)](CF 3 S0 3 )2 and CuCl 2 using H 2 0 2 Oxidized Total Percent Substratea Products mmoles Peroxide mmoles Oxidant Efficiency (Selectivity) /so-Butanol: /so-Butane 3.6 (99.0) 3.64 7.28 !so-Butyl ol: 0.04 (1.00) 1-Pentanol: 0.36 (4.60) 2-Pentanol: 1 67 (21.4) 3-Pentanol: Pentanec 1.99 (25.6) 11.6 23.1 Pentanal: 0.96 (12.3) 2-Pentanone: 1.45 (18.6) 3-Pentanone: 1.36 (17.5) Pentanoic Acid: Trace Percent Conversionb 19.2 41.0 a: l 6xl0-4 mole [Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 l.6xl0 4 mole CuC1 2 60ml acetonitrile 5 0xI0 2 moles 35% H 2 0 2 40psi substrate (l 9xl0 2 moles), 15 hours at 75 C b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate c: 2ml (2 0x10 2 moles) pentane used

PAGE 149

134 Oxidation experiments with propane cis-[Fe(dmp ) 2 (H 2 0) 2 ] ( CF 3 S0 3 ) 2 and CuCI 2 resulted in a 34.3 % overall conversion after 15 hours. An increase in overall conversion of 1.2% was obt~ined when compared to an oxidation perfor1ned in the absence of C uC1 2 Also exhibited was the desired increase in alcohol selectivity to 14.1 % for 1-propanol and 38.4% for 2-propanol both increased from the 1.8% and 13.1% selectivities achieved when CuC1 2 is omitted. The accompanying decrease in selectivity to the aldehyde and ketone also result s Similar results were obtained for the remaining alkanes, as shown in Tables 5-7 and 5-8. Oxidation of propane experiments performed with the [Fe( dmp )Cl 2 ] complex and CuC1 2 resulted in a 20.1 % overall conversion of the alkane after 15 hours. Once again a minimal increase in conversion ( 1.2 % ) is achieved when compared to the results obtained in the absence of CuC1 2 Alcohol selectivity increases once again from 2.5 to 3.8% forl propanol and from 18 9 to 34.9 % for 2-propanol. Aldehyde and ketone selectivities were also decreased to 29 9% for propanal and 31.4% for acetone. Tables 5-9 and 5-10 further demonstrate the desired increases in alcohol selectivities for the remaining alkanes. The role of CuC1 2 in this reaction mixture has been investigated through blank analysis and control experiments. An explanation and discussion of the result s are provided in Chapter 2. The effect of increased mole equivalents of CuC1 2 on the s electivity of the alcohol were also investigated using both precursors cis-[Fe ( dmp) 2 (H 2 0 ) 2 ](CF 3 S0 3 ) 2 and [Fe(dmp)Cl 2 ] at 75 C. The results for the oxidation of propane with the cis [Fe(dmp) 2( H 2 0) 2 ](CF 3 S0 3 ) 2 complex are provided in Table 5-11 and illustrated in Figure

PAGE 150

135 Table 5-9: Oxidation Results for Ethane, Propane and Butane@ 75C with Fe(dmp)Cl 2 and CuCl 2 using H 2 0 2 Oxidized Total Percent Substratea Products mmoles Peroxide Percent mmoles Oxidant Efficiency Conversionb (Selectivity) Ethanol: 0.51 (6.50) Ethane Acetaldehyde: 16.5 32.9 40.8 5.76 (74.3) Acetic Acid: 1.48 (19.2) 1-Propano 1: 0.14 (3.80) 2-Propanol: 1.31 (34.9) Propane Propanal: 6.03 12.1 20.1 1.12(29.9) Acetone: 1.17 (31.4) Propionoic Acid: Trace 1-Butanol: 0.09 (5.60) 2-Butanol: 0.07 (4.20) Butane c Butanal: 4.21 8.42 17.4 0.17 (10.2) 2-Butanone: 0.25 (15.2) Butanoic Acid: 1.07 (64.8) a: l.6xl0 -4 mole Fe(dmp) 2 Cl 2 l.6xl0-4 mole CuCl 2 60ml acetonitrile, 5.0x10 2 mole s 35 % H 2 0 2 40psi substrate (l 9xI0 2 moles) 15 hours at 75 C. b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate c: 20psi (9 5xJo 3 moles) butane used.

PAGE 151

136 Table 5-10: Oxidation Results for !so-Butane and Pentane @ 75 C with Fe(dmp)Cl 2 and CuCI 2 using H 2 0 2 Oxidized Total Percent Substrate Products mmoles Peroxide Percent moles Oxidant Efficiency Conversionb (Selectivity) lso-B utanol: !so-Butane 3.5 (98.3) 3.56 7.12 18.7 !so-Butyl ol: 0.06 (1.7) 1-Pentanol: 0.15 (2.50) 2-Pentanol: 0.77 (13.1) 3-Pentanol: Pentane c 1.37 (23.3) 9.51 19.1 31.1 Pentanal: 1.61 (27.3) 2-Pentanone: 1.23 (20.8) 3-Pentanone: 0.77 (13.0) Pentanoic Acid: Trace a: l 6xl0 -4 mole Fe(dmp) 2 Cl 2 l 6xl0-4 mole CuC1 2 60ml acetonitrile 5.0xJ0 2 moles 35% H 2 0 2 40psi substrate (t.9xio 2 moles) 15 hours at 75C. b : Percent conversion is based on the total of oxidized product s divided by the initial amount of substrate. c: 2ml (2.0xl 0 2 moles) pentane u se d

PAGE 152

137 Table 5-11: Oxidation Results for Propane@ 75C with cis-[Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 Mole mmoles mmoles mmoles Total Percent Equivalents Alcohol Aldehyde Ketone mmoles Conversion b CuC1 2 a ( 0 /o Sel.) ( 0 /o Sel.) (o/o Sel.) Products 0 0.94 2.77 2.59 6.30 33 2 (14.9) (43.6) ( 41.5) 1 2.38 2.26 1.90 6 54 34 .4 ( 53.3) (25.4) (21.3) 2 1.08 1.49 0.52 3.08 17.10 (3 5.0) (48.3) (16.7) 3 1.86 1.09 0.46 3.41 18.40 ( 54.2) (32.1) (13.7) 4 2.35 0.69 0.54 3.58 19 .79 ( 65.6) (19.2) (15.2) 5 2.75 0.60 0.55 3.90 21.50 ( 70.6) (15.4) (14.0) a: 1 .6x l0 -4 mole cis -[Fe { dmp ) 2 ( H 2 0) ](CF 3 S0 3 ) 2 l.6xl0 -4 x mole eq. mole s CuCl 2 60ml acetonitrile 5.0x10 2 moles 35 % H 2 0 2 40psi substrate (l.9xlo 2 moles ), 15 hours at 75 C. b: Percent conversion is based on the total of oxidized product s divided b y the initial amount of substrate

PAGE 153

> 0 Q) Q) Cl) C Q) 0 'Q) a. 70 00 60 00 50 00 40 00 30 00 20 00 10 00 138 Effect of CuCl 2 on Product Selectivity with the cis-[Fe(dmp) 2 (H 2 0) 2 ] 2 + Precursor 6.00 5 00 4 00 Cl) Q) 0 :lE 3 00 2 00 1 00 0 00 .._ __ __._ ______ __._ _________ 0 00 0 1 2 3 4 5 Mole Equivalents CuCl 2 Alcohol -a-Aldehyde a Ketone Percent Convers i on O Total Moles (2nd ) Figure 5-1: Oxidation of Propane with cis-[Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 and Varying Mole Equivalents of CuCI 2 using H 2 0 2

PAGE 154

139 Table 5-12: Oxidation Results for Propane@ 75C with Fe(dmp)Cl 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2 Mole mmoles mmoles mmoles Total Percent Equivalents Alcohol Aldehyde Ketone mmoles Conversionb CuC1 2 (o/o Se).) (% Sel.) ( 0 /o Sel.) Products 0 0.69 1.68 1.21 3.58 18.8 (21.4) (45.7) (32.9) 1 1.45 1.12 1.17 3.74 19.7 > (38.7) (29.9) (31.4) 2 1.27 0.50 0.52 2.29 12.66 (55.6) (21.9) (22.5) 3 1.83 0.48 0.35 2.66 14.70 (68 8) (17.9) (13.3) 4 2.30 0.54 0.41 3.25 17.95 (70.9) (16.5) (12.6) 5 2.93 0.54 0.59 4.05 21.28 (76.0) (14.0) (10.0) a: l.6x10 4 mole cis-[Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 l.6xl0-4 x mole eq. moles CuCl 2 60ml acetonitrile, 5.0xI0 2 moles 35% H 2 0 2 40psi substrate (l.9xlo 2 moles), 15 hours at 75 C. b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate

PAGE 155

> .... (J Q) Q) Cl) .... C Q) u Q) a.. 70 00 60 00 50 00 40 00 30 00 20 00 10 00 140 Effect of CuCl 2 on Product Selectivity with the Fe( dmp )Cl 2 Precursor 4 00 3 00 2 00 1 00 0 00 ._ __ --1, ___ ...,__ ___ i,..._ __________ 0 00 0 1 2 3 4 5 Mole Equivalents CuCl 2 (/) Q) 0 t Alcohol -II-Aldehyde B Ketone Percent Conversion O Total Moles ( 2nd) Figure 5-2: Oxidation of Propane with Fe(dmp)Cl 2 and Varying Mole Equivalents of CuC1 2 using H 2 0 2

PAGE 156

141 5-1. The result s f or the [Fe(dmp ) Cl 2 ] complex are given in Table 5-12 and graphically represented in Figure 5-2. As with the cis-[Ru(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 analogue, successive addition of CuC1 2 mole equivalents to both iron derivatives resulted in an increase in selectivity to the alcohol along with the decreased selectivity to the aldehyde and ketone. Although the trend exhibited by the iron complexes are not as smooth as that observed for the ruthenium analogue the effect of the added copper is exhibited. Variations in results for the iron complexes can be attributed to the presence of two first row transition metals (Cu and Fe) which lead to an increased rate of peroxide decomposition thus the varied results. Peroxide decomposition observed in the ruthenium system is not as rapid when compared to the iron analogue. The decrease in overall conversion exhibited can be attributed to a lower concentration of peroxide present due to decomposition. Effect of Temperature A series of experiments were perfo1n1ed on both iron analogues with and without one mole equivalent of CuC1 2 at 25 50 75 and 100 C to detertnine the effect of temperature on each complex. The increase in temperature is expected to allow for increased substrate reactivity, solubility of the catalyst and rate of oxidation The results for the oxidation of propane at each temperature for both catalysts are provided in Tables 5-13, 5-14 5-15 and 5-16. A mixture of catalyst ( l .6x 10-4 moles), acetonitrile ( 60ml) hydrogen peroxide (5.0x10 2 moles) and propane (1.9xt0 2 moles) were allowed to reaction for 15 hours at

PAGE 157

142 Table 5-13: Oxidation Results for Propane @ Varying Temperatures with ci s [Fe(dmp) 2 (H20)](CF 3 S0 3 ) 2 using H 2 0 2 mmoles mmoles mmoles Total Temp. Alcohol Aldehyde Ketone mmoles Percent (oc) (% Sel.) ( 0 /o Sel.) ( 0 /o Sel.) Products Conversionb 25 0.16 0.55 0.54 1.25 5 90 ( 12 7) ( 44.4) (42.9) 50 0.64 1.96 1.97 4.56 23.1 ( 14.0) (43.0) (43 10) 75 0.94 2.77 2.59 6.30 34 9 ( 14.9) (44.0) (41.1) 100 1.15 1.57 2.26 4.98 29.5 (23.1) (31 5) (45.8) a: l.6xl0-4 mole c i s -[Fe(dmp) 2 ( H 2 0)](CF 3 S0 3 ) 2 60ml acetonitrile 5.0x10 2 mole s 3 5 % H 2 0 2 40 psi sub s trate (l.9xI0 2 mole s), 15 hour s. b : Percent conversion i s ba s ed on the total of oxidized product s divided by the initial a mount of substrate.

PAGE 158

143 Table 5-14: Oxidation Results for Propane@ Varying Temperatures with cis [Fe(dmp) 2 (H20)](CF3S03)2 and CuCI 2 using H 2 0 2 mmoles mmoles mmoles Total Temp. Alcohol Aldehyde Ketone mmoles Percent (oc) (% Sel.) ( 0 /o Sel.) (% Sel.) Products Conversionb 25 0.86 0.50 0.50 1.86 8.82 ( 46.2) (26.9) (26.9) 50 0 69 4.06 0.96 5.71 29.3 (12.1) (71.1) (16.8) 75 2.38 2.26 1.90 4.46 24.9 (53.3) (25.4) (21.3) 100 1.67 1.36 0.91 3.94 23.3 (42.4) (34.5) (23.1) a: l.6x10 4 mole cis-[Fe(dmp) 2 (H 2 0)](CF 3 S0 3 ) 2 l 6xl0 4 mole CuC1 2 60ml acetonitrile, 5.0xio 2 moles 35% H 2 0 2 40psi substrate (l.9xI0 2 moles), 15 hours. b : Percent conversion is based on the total of oxidized products divided by the initial amount of substrate

PAGE 159

144 Table 5-15: Oxidation Results for Propane @ Varying Temperatures with Fe ( dmp ) Cl 2 using H 2 0 2 mmoles mmoles mmoles Total Temp. Alcohol Aldehyde Ketone mmoles Percent (oc) (% Sel.) (% Sel.) (o/o Sel.) Products Conversionb 25 0 17 0.70 0.34 1.21 5 .7 2 ( 13 9) (57 9 ) (28 3) 50 0 98 2.07 2.37 5.43 27.85 ( 18.1) ( 38.2) (43.8) 75 0.79 1.68 1.21 3.68 20 5 (21.5) (45.7) (32.9) 100 2.15 1.57 1 44 4.66 27.6 ( 46.2 ) (22.9) (30 9) a: l.6xl0 -4 mole Fe(dmp ) Cl 2 60ml acetonitrile 5 0 x I0 2 moles 35 % H 2 0 i, 40p s i s ub s trat e (l 9xJ0 2 mole s ) 15 hour s. b : Percent conversion is based on the total of oxidi z ed product s divided by the initial amount of sub s trate

PAGE 160

145 Table 5-16: Oxidation Results for Propane@ Varying Temperatures with Fe(dmp)Cl 2 and CuC1 2 using H 2 0 2 mmoles mmoles mmoles Total Temp. Alcohol Aldehyde Ketone mmoles Percent (oc) (% Sel.) ( 0 /o Sel.) (% Sel.) Products Conversionb 25 0.59 0.29 0.32 1.20 5.70 (49.4) (24.2) (26.5) 50 1.97 1.66 1.29 4.94 25.4 (35.3) (33.6) (26.1) 75 1.45 1.12 1.17 3.74 20.9 (38.8) (29.9) (31.4) 100 3.38 1.63 0.99 6.00 35.6 (56.4) (27.1) (16.6) a: l.6xl0 4 mole Fe(dmp)Cl 2 l 6xl0 4 mole CuC1 2 60ml acetonitrile 5 0x10 2 moles 3 5% H 2 0 2 40psi substrate (1 9x I 0 2 moles), 15 hours b: Percent conversion is based on the total of oxidized products divided by the initial amount of substrate.

PAGE 161

146 proceeded at a temperature as low as 25C (5.9% overall conversion) and as high as 100C (29.5% overall conversion). Addition of one mole equivalent of CuC1 2 resulted in the anticipated increase in alcohol selectivity Experiments at each temperature using the [Fe(dmp)Cl 2 ] complex also followed a similar trend when compared to cis-[Fe(dmp) 2 (H 2 0) 2 ](CF 3 S0 3 ) 2 complex. However upon addition of CuC1 2 no increase in overall conversion was obtained, but an increase in selectivity to the alcohol was achieved. Conclusions [Fe(dmp)Cl 2 ] (I) for the activation of linear and branched alkanes has been perfonned. Oxidation of methane experiments conducted with the sterically hindered cisproduction of methyl acetate in an acetic acid/acetic anhydride solvent mixture in four hours at 75C. Methyl acetate, produced from the hydrolysis of methanol and acetic acid, is generated as prevention to the over-oxidation of methanol. As a result, selectivity to the alcohol increases. To further demonstrate the activity of this complex, methane activation experiments were conducted with molecular oxygen. This resulted in the production of methyl acetate in reaction times as low as one hour at 100C. Functionalization of C 2 -C 5 alkanes were also investigated using each complex and hydrogen peroxide in an acetonitrile at 75C. In an effort to increase selectivity to the hydroxy lated product, CuC1 2 ( one mole equivalent) was added to the reaction mixture. This resulted in a selectivity increase to the alcohol without a significant decrease in the

PAGE 162

147 catalytic activity of the precursor. Also exhibited was the associated decrease in selectivity to the aldehyde and ketone. The effec~ of subsequent additions of CuC1 2 mole equivalents was also investigated resulting in a further increase in alcohol selectivity, however catalyst activity and lifetime were decreased. Temperature dependence studies performed demonstrated the ability of each catalyst precursor to activate propane at temperatures as low as 25 C with comparable selectivities to those obtained at 75 C.

PAGE 163

CHAPTER6 CONCLUSIONS A series of homogeneous catalysts able to activate saturated hydrocarbons under mild and ambient conditions have been developed synthesized, characterized and oxidation catalysis perfor1ned. The ability to design a such catalyst has been the subject of ongoing research throughout the industrial sector. This family of catalyst is offered as a substitute to processes, which operate under stringent conditions as well as a catalyst able to utilize alternative feedstocks for the production of viable synthetic compounds. via a peracid assisted mechanism provides a direct pathway to the production of methanol. The reaction conditions (75 C four hours of reaction and a mild oxidant H 2 0 2 ) offers an attractive reaction from an industrial standpoint. Attempts to introduce aerobic conditions to this oxidation were also successful, allowing for a more detailed investigation. An additional novel feature of this oxidation system is the trapping o f the methanol product as methyl acetate The ester for1ned is resistant to further oxidation thereby allowing for higher selectivities to be attained. Hydrolysis of the ester allow s for recovery of methanol and regeneration of the trapping agent This ruthenium complex also functionalized linear and branched alkanes at 7 5 C using hydrogen peroxide. The peroxide allows for generation of the high valent ruthenium-oxo species necessary for alkane functionalization This reaction proceeds via 148

PAGE 164

149 a free radical mechanism and offers the selective partial oxidation of each alkane investigated while maintaining a relatively high overall conversion. Further modification of alcohol selectivity can be obtained upon addition of CuC1 2 . Attempts to synthesize additional ''analogues '' based on the cis[Ru( dmp ) 2 (H 2 0 ) 2 ](CF 3 S0 3 ) 2 architecture but with greater oxidizing potential were performed. As a result the synthesis of two additional catalyst precursors ci s characterized by IR NMR and F AB and X-ray crystallography providing insight into this family of homogeneous catalysts. Investigation into the catalytic activity for cis-[Fe( dmp ) 2 (H 2 0 ) 2 ](CF 3 S0 3 ) 2 complex for hydroxylation of methane with hydrogen peroxide and molecular oxygen at 75 C was also perfor111ed. The iron complex offers a higher catalytic activity when compared to the ruthenium analogue, yet initial mechanistic studies suggest it proceeds via a peracid mechanism identical to that of the ruthenium system. The mono-dmp complex [Fe(dmp)Cl 2 ], and the sterically hindered ci s alkanes Again each offered a more robust activity when compared to the ruthenium analogue and their selectivities could be modified upon addition of CuC1 2 Preliminary mechanistic studies suggest oxygenates are f ortned via a free radical hydrogen ab s traction mechanism.

PAGE 165

AIBN BQ Catalyst Conversion Coordination DMP Heterolysis Homolysis Inhibitor Initiator Oxidation Selectivity Yield GLOSSARY A radical chain initiator. Azo-bis-(iso-Butyrolnitrile). A radical chain inhibitor. Benzoquinone. A complex which allows for increase in rate for a chemical process. Amount of reaction consumed in a chemical process, usually expressed as a percentage. The addition of ancillary ligands which are directly bonded to the metal center. 2,9-dimethyl-1, 10-phenanthroline, also known as neocouprine Fragmentation of a neutral compound into an anion and cation species. Fragmentation of a neutral compound into two identically electronically charge species. A compound which prevents the propagation of a radical or catalytic cycle. A compound which begins propigation of a radical or catalytic cycle. Electron removal from a chemical species. The relative rates of two or more simultaneous processes occurring on the same substrate. Usually expressed as a percentage. The amount of a particular product for1ned divided by the amount of reactant provided. Usually expressed as a percentage. 150

PAGE 166

REFERENCES 1. Shilov, A.E., '' Historical Evolution of Homogeneous Alkane Activation Systems, in Activation and Functionalization of Alkanes; Hill, C., Ed., John Wiley and Sons New York, 1989 2 . 2. Shilov A.E., Activation of Saturated Hydrocarbons by Transition Metal Complexes; Reidel Publishing Co. Dordrecht, 1984. 3. Crabtree, R.H., Chem. Rev., 1985, 85,245. 4. Bergman R.G. Science, 1984 223, 902. 5. Goldstein, A.S. Ph.D. Dissertation, University of Florida, 1991. 6. Lyons J.E., 'Transition Metal Complexes as Catalysts for the Addition of Oxygen to Reactive Organic Substrates,'' in Aspects of Homogeneous Catalysis Volume 3 ; Ugo R., Ed., Reidel Publishing Co ., Dordrecht 1977 10. 7. Shilov A.E., Activation of Saturated Hydrocarbons by Transition Metal Complexes ; Reidel Publishing Co. Dordrecht 1984 15-20. 8. Shilov A.E. Activation of Saturated Hydrocarbons by Transition Metal Complexes; Reidel Publishing Co., Dordrecht, 1984 163-182. 9. Shilov, A.E. 'Historical Evolution of Homogeneous Alkane Activation Systems ' in Activation and Functionalization of Alkanes; Hill.C., Ed., John Wiley and Sons New York ; 1989 3-11. 10. Heck R.F., Adv Organomet. Chem 1966 4, 2431. 11. Orchin, M.; Ruplius, W., Cata! Rev 1972, 6 85. 12. Orchin, M. Acc Chem Res 1981, 14, 25. 13. Pino, R.; Piacenti F.; Bianchi M., Organic Synthesis via Metal Carbonyls. Volume II; Wonder I. ; Pino. P., Eds.: John Wiley and Sons, New York; 1977, 43-143. 14 Heck R.F; Breslow D.S., J. Am Chem Soc ., 1961 83 4023 15. Roper M.; Loevenich, H. Catalysis in C 1 Chemistry, Keim, W. Ed. Reidel Publishing Co .,: Dordrecht, 1983 105. 16. Chen M ~ J. ; Rathke, J.W. Organometallics 1987 6 1833. 151

PAGE 167

152 17. Wegman, R.W.; Busby, D.C.; Letts, J.B., Industrial Chemicals via C 1 Processes; Fahey, D.R., Ed., ACS Symposium Series 328; American Chemical Society, Washington, D.C., 1987, 125. 18. Pretzer, W.R:; Kobylinski, T.P., Ann. NY. Acad. Sci., 1980, 33, 58. 19. Keim, W., Industrial Chemistry via C 1 Processes; Fahey, D.R., Ed., ACS Symposium Series 328; American Chemical Society, Washington, D.C., 1987, 1. 20. Fakley, M.E.; Head, R.A., Appl. Cata/., 1983, 5, 3. 21. Rizkalla, N.; Goloaszewski, A., Industrial Chemicals via C, Processes; Fahey, D.R., Ed., ACS Symposium Series 328; American Chemical Society, Washington, D.C., 1987, 136. 22. Gauthier-Lafaye, J.; Perron, R. Methanol and Carbonylation; Editions Technip Paris, 1987, 136. 23. Sneeden, R.P.A, Comprehensive Organometallic Chemistry; Wilkinson, G.; Stone, F G.A.; Able, E.W., Eds., Pergamon Press, Oxford, 1982; Vol. 8, Chapter 50.2. 24. Tkatchenko, I., Comprehensive Organometallic Chemistry; Wilkinson, G.; Stone, F.G.A.; Able, E.W., Eds., Pergamon Press, Oxford, 1982; Vol. 8, Chapter 50.3. 25. Dumas, H.; Levisalles, J.; Rudler, H., J. Orgdnomet. Chem., 1979, 177,239. 26. Forster, D., Adv. Organomet. Chem., 1979, 17, 255 . 27. Adamson, G.W.; Daly, J.J.; Forster, D.J., Organomet. Chem., 1974, 71, C17. 28. Adams, H.; Bailey, N.A.; Mann, B.E.; Manuel, C.P.; Spencer, C.M.; Kent, A.G., J. Chem. Soc., Dalton Trans., 1988, 489. 29. Drago, R.S., Coordination Chem Rev., 1992, 117, 185-213 . 30. Mimoun, H., Homogeneous Metal-Oxo Complexes; Elsevier, New York, 1985. 31. Drago, R.S.; Beer, R.H., lnorganica Chimica Acta, 1992, 198-200, 359-367. 32. Meyer, T J, ''Metal-Oxo Complexes and Oxygen Activation," in Oxygen Complexes and Oxygen Activation by Transition Metal Complexes; Martell, A.E.; Saywer, D.T. Eds., Proceedings at the Fifth Annual JUCCP Symposium; Plenum Press, New York, 1988.

PAGE 168

153 33. Leising R.A. ; Takenchi K.J. Jnorg Chem., 1987, 26 3833. 34. Lau, T.C. ; Che C.M. ; Lee W O; Poon, C.K, J. Chem Soc Chem Commun. 1988, 1406. 35. Schir1nann J P.; Delavarenne S.Y., Hydrogen Peroxide in Organic Chemistry; Edition et Documentation Industrielle Paris, 1979. 36. Wilson, S., Peroxygen Technology in the Chemical Industry, Internal Publication Solvay Interox Ltd. R&D, New York, 1993. 37. Einchem Anic Spa European Patent EPA 100, 119 1984. 38. Bailey, C.; Drago, R.S.; J. Chem. Soc Chem. Commun. 1987 179. 39. Goldstein, A.S. ; Drago R.S ; J. Chem. Soc. Chem Commun., 1991, 21 40. Goldstein, A.S.; Beer R.H.; Drago, R.S. J. Am. Chem Soc. 1994 116 2424-2429. 41. Ahn K.H.; Groves, J.T., lnorg. Chem., 1987, 26, 3833. 42. Dobson J.C.; Seok, W.K.; Meyer, T.J., lnorg. Chem 1986 25, 1514. 43. Courthey, J L. ; Swanborough K F. Rev Pure Appl Chem 1972 47 22 44. Lee D.G. ; Van den Engh M., Org Chem., 1973, 58 177. 45. Dobson, J.C.; Meyer, T.J. Jnorg. Chem., 1988 27 3283 . 46. Robbins M H .; Drago R.S., J. Chem Soc ., Dalton Trans 1996, 105-110 47 Parkyns, N.D. Chem. Brit., September 1990, 841-844. 48. Foulds G.A.; Gray, B.F., Fuel Proc. Tech ., 1995, 42, 129-150. 49. Hamiliton, D.E.; Drago R.S.; Zombeck A., J. Am Chem. Soc ., 1987 109 314. 50. Periana, R.A.; Taube D.J.; Evitt, E.R.; Loffer, D.G.; Wentrch P.R.; Voso G. ; Moesudai T Science 1993 259 340. 51. Periana R.A.; Taube D.J. ; Evitt E.R. U.S. Patent 1994 5,306 855. 52. Gretz, E. ; Oliver T.F.; Sen A. J. Am Chem Soc. 1987 109 8109. 53. Kao L.C. ; Hutson A.C. ; Sen, A. J. Am Chem. Soc ., 1991, 113 700

PAGE 169

154 54. Varagaftik, M.N.; Stolaror, I.P.; Moiseec, I.I. J Chem. Soc. Chem. Commun, 1990, 1049. 55. Drago, R.S.; Mateus, A.L., J Org Chem., 1996, 61, 5693-5696. 56. Schneider ~ H .J.; Muller, W., J Org. Chem., 1985, 50, 4609. 57. Murahashi, S.; Oda, Y.; Komiya, N.; Nasta, T. Tetr. Lett ., 35, (43), 1994 7953-7956. 58. Schneider, H.J.; Muller, W., J Org. Chem., 1985, 50, 4609. 59. Sheldon, RA.; Koehl, J.K., Metal Catalyzed Oxidations of Organic Compounds; Academic Press, New York, 1981. 60. Mimoun, H ., Comprehensive Coordination Chemistry: Wilkinson, G. Gillard R.D.; Mc Cleverty, J.A., Eds., Pergamon Press: Oxford, 1987; Vol. 6. 61. Hartford, W.H.; Darrin, M., Chem. Rev., 1958, 58, 1-61. 62. Stewart, R., Oxidation Mechanisms, Benjamin, New York, 1964 63. Rosenzwieg, A.C.; Frederick C.A.; Lippard, S.J.; Nordlund, P., Nature, 1993, 366, 537-543. 64. Liu, K.E.; Johnson, C.C.; Newcomb, M.; Lippard, S.J., J Am. Chem., Soc., 1993 115 939-947. 65 Cook, G.K.; Meyer, J.M., J Am. Chem. Soc., 1995, 117, 7139-7156 . 66. Haber, J., ''Studies in Surface Sciences," In Dioxygen Activation and Homogeneous Catalytic Oxidation; Simadi L.I., Ed, Elsevier, New York, 1991; 66. 67. Roelofs, M.G.; Wasse11nan, E.; Jensen, J.K., J Am. Chem. Soc ., 1987, 109 42074217. 68. Comprehensive Organic Synthesis; Trost, B.M., Ed., Pergamon, New York 1991 Vol. 7. 69. Cytochrome P-450: Structure, Mechanism and Biochemsity: Ortiz de Montellano P.R., Ed., Plenum, New York, 1985. 70 Barton, D.H.R.; Doller, D Acc. Chem. Res., 1992 25, 504. 71. Barton, D.H.R.; Wang, T.L. Tetrahedron, 1994 50, 1011.

PAGE 170

155 72. Shu, L.; Nesheim, J.C.; Kauffinann, K.; Munck, E.; Lipscomb, J.D.; Que L. Science, 275, 1997, 515-518. 73. Gonzalez, M. ; Drago, R.S.; Gordon, B.W.F. Submitted to J. Am. Chem. Soc., December, 1997. 74. Beer, R.H., Drago, R.S. Symposium on Natural Gas Upgrading II, Co-Chairmen, Huff, G.A.; Scarpiello, D.A, Feb. 1992, 37, (1), 239. 75. Che, C.M.; Leung, W.H. J. Chem. Soc. Chem. Commun., 1987, 1376. 76. Che, C.M.; Tang, W.T.; Wong, W.T.; Lai, T.F. J. Am Chem. Soc., 1989, 111, 9048. 77. Che, C.M.; Yorn, V.; Mak, T.C.W. J. Am. Chem. Soc., 1990, 112, 2284. 78. Holm, R.H. Chem. Rev., 1987, 87, 1401. 79. Griffith, W.P. Transit. Met Chem., 1990, 15,251. 80 Imoff, V.; Drago, R.S. lnorg. Chem., 1965, 4, (3), 6921. 81. Hague, J.P.; Sawyer, D.T. J. Am Chem. Soc., 1995, 117, 5617-5621 82. Machii, K.; Watanabe, Y.; Morishima, I. J. Am Chem Soc., 1995, 117, 6691-6697. 83. Leising, R.A.; Kin, J.; Perez, M.A.; Que, L. J Am. Chem Soc., 1993, 115, 95249530. 84. Wa~abe, Y.; Groves, J.T. The Enzymes, Sigman, D.S., Ed., Academic Press, San Diego, 1992; Vol. 20, 405-452. 85. Groves, J.T.; Nemo, T.E.; Meyers, R.S. J. Am. Chem. Soc., 1979, 101, 1032-1033 86. LaMar, G.N.; Van Hecke, G.R. lnorg. Chem., 1970, 9, 1547-1549. 87. Goldstein, A.S., Ph.D. Dissertation, University of Florida, 1991, 63-64. 88. Goldstein, A.S., Ph.D. Dissertation, University of Florida, 1991, 65-66. 89. Larpent, L. Briffand, T.; Patin, H. Chem. Soc Chem. Commun., 1990, 1193. 90. Konig, E.; Watson, K. Chem. Phys. Lett ., 1990, 6, 457. 91. Ozarowski, A.; McGarvey, B.R.; Sarkar, H.B.; Drake, J.E. lnorg. Chem., 1988, 27, 628.

PAGE 171

156 92. Johansson, L.; Molund, M.; Oskarsson, A., Jnorg. Chim. Acta, 1978, 31, 117. 93. Goodwin, H.A.; Kucharski, E.S.; White, A.H., Aust. J. Chem 1983, 36, 1115. 94. Vincent, J.B.~ Huffman, J.C.; Christou, G.; Li, Q.; Nanny, M.A.; Hendrickson, D.N.; Fong, R.H.; Fish, R.H., J. Am. Chem. Soc ., 1988, 110, 6898-6900. 95. Menage, S.; Vincent,, J.-M.; Lambeaux, C.; Chottard, G.; Grand, A.; Fontecave, M., lnorg. Chem ., 1993, 32, 4766-4773. 96. Kojima, T.; Leising, R.A.; Zang, Y. Que, L., J. Am. Chem Soc., 1991, 113, 85558557. 97. Kim, J.; Harrison, R.G.; Kim, C.; Que, L., J. Am. Chem. Soc 1996, 118, 4373-4379. 98. Kauffmann, K. E.; Bartos, M. J.; Kidwell, C.; Gordon-Wylie, S. W.; Collins, T.J; Clark, R.; Weintraub, S. T.; Miinck, E., J. lnorg. Biochem., 1995, 59, 317

PAGE 172

BIOGRAPfilCAL SKETCH Michael A. Gonzalez was born in El Paso, Texas, on November 17, 1969, to Wilfredo and Armida Gonzalez. He attended Andress High School in El Paso, Texas, and graduated in June 1987. While in high school, Michael was a member of the National Honor Society, treasurer of the Student Council, and two-time member of the varsity golf team. In the Fall of 1987 he began his college career at the University of Texas El Paso as a pre-medicine major. In his sophomore year, he was awarded a Research Careers for Minority Students (RCMS) academic scholarship, at which time he changed his major to chemistry. After three and one half years of undergraduate research, he graduated in 1992 under the direction of Dr. Leonard W. terHaar. An undergraduate honors thesis ''Synthesis and Characterization of Inorganic Polymers'' was the result of this research. In the Fall of 1992 he began graduate school at the University of Florida to attain his doctorate in chemistry. His research focused on the development and application of homogeneous transition metal-oxo catalysts for the activation of alkanes under the direction of Dr. Russell S. Drago Upon completion of his doctorate he will be employed with the United States Environmental Protection Agency in Cincinnati, Ohio 157

PAGE 173

. I certify that I have read this study and that in my opinion it conforrns to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philo so y. Russell S. Drago, Chair Graduate Research Professor of Chemistry I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy Daniel R. Talham Associate Professor of Chemistry I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Rick A. Yos Professor of Chemistry I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Robert T. Kenneay Associate Professor o emistry I certify that I have read this study and that in my opinion it confor1ns to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Michael D. Sacks Professor of Material Science

PAGE 174

This dissertation was submitted to the Graduate Faculty of the Department of Chemistry in the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. May 1998 Dean Graduate School

PAGE 175

LD 1780 199 g ,,6b't3 l