New Route to Synthesize Surface Organometallic Complexes (SOMC): an Approach by Alkylating Halogenated Surface Organometallic Fragments

New Route to Synthesize Surface Organometallic Complexes (SOMC): An Approach by Alkylating Halogenated Surface Organometallic Fragments.

Dissertation by

Ali Hamieh

In Partial Fulfillment of the Requirements

For the Degree of

Doctor of Philosophy

King Abdullah University of Science and Technology

Thuwal, Kingdom of Saudi Arabia

February, 2017

EXAMINATION COMMITTEE PAGE

The dissertation of Ali Hamieh is approved by the examination committee.

Committee Chairperson: Professor Jean-Marie Basset Committee Members: Professor Kazuhiro Takanabe, Professor Udo Schwingenschlogl, Professor Joumana Toufaily.

Ali Hamieh

ABSTRACT

New Route to Synthesize Surface Organometallic Complexes (SOMC): An Approach by Alkylating Halogenated Surface Organometallic Fragments.

Ali I. Hamieh

The aim of this thesis is to explore new simpler and efficient routes for the preparation of surface organometallic complexes (SOMC) for the transformation of small organic molecules to valuable products. The key element in this new route relies on surface alkylation of various halogenated surface coordination complexes or organometallic fragments (SOMF). The first chapter provides an overview on the origin of organometallic compounds, their classical synthesis, characterization and some of their applications

In the second chapter, novel silica-supported tungsten oxo-trimethyl complex [(≡Si-O-

)W(=O)Me3] was synthesized using the new SOMC synthetic approach. WOCl4 was grafted on the surface of silica, partially dehydroxylated at 700°C (SiO2-700), and [(≡Si-O-

)W(=O)Cl3] was produced. The supported complex methylated with ZnMe2 and transformed into [(≡Si-O-)W(=O)Me3], which was fully characterized. It was found that complex [(≡Si-O-)W(=O)Me3] has two conformational isomers at room temperature.

This complex was found to be active in cyclooctane metathesis and direct transformation of ethylene to propylene.

The third chapter elaborates the effect of substituents on catalysis. In this chapter we

Dipp replace one methyl from [(≡Si-O-)W(=O)Me3] by a bulkier ligand (Im = 1,3-bis(2,6- diisopropylphenyl)imidazolin-2-iminato). Synthesis, structure, and olefin metathesis 5

Dipp activity of complex [(≡Si-O-)W(=O)(CH3)2-Im N] is reported. This complex proved to be an active pre-catalyst for self-metathesis of terminal olefins such as propylene and 1- hexene, showing better selectivity as compared to [(≡Si-O-)W(=O)Me3].

In the fourth chapter, surface alkylation strategy was used as a scalable method to synthesize the highly important complex [(≡Si–O−)WMe5] that obtained by in situ alkylation of the surface-grafted tungsten chloride [(≡Si–O−)WCl5] (1). Upon alkylation, a mixture of [(≡Si–O−)WMe5] and [(≡Si–O−)WMe2(≡CH)] was identified. The latter might have been generated by partial decomposition of the tungsten methyl chloride compound; this was confirmed by DFT calculations and experimental results.

v In the fifth chapter, we isolated the first tetramethyl niobium complex [(≡SiO)Nb Me4]

v (2) through the surface alkylation of [(≡SiO-)Nb Cl3Me] (1). Complex (1) was found to be active in the ethylene oligomerization by producing olefins up to C30, while surprisingly

(2) selectively dimerizes ethylene into 1-butene and in both cases reaction occurs in the absence of any co-catalyst.

ACKNOWLEDGEMENTS

I would like to extend thanks to the many people, who so generously contributed to the work presented in this thesis. Special mention goes to my great supervisor, Professor

Jean-Marie Basset. I am highly honored and delighted to be his student. My PhD has been an amazing experience and I thank Professor Basset wholeheartedly, not only for his tremendous academic support, but also for giving me so many wonderful opportunities. Greatest gratitude to my committee members Professor Kazuhiro

Takanabe, Professor Udo Schwingenschlogl and Professor Joumana Toufaily.

So many people I need to thank especially Dr Manoja Samantaray, Dr Raju Dey, Dr Yin

Chen, Dr. Ziyauddin Qureshi, Professor Valerio D’Elia, Dr Jérémie Pelletier and Dr.

Santosh Kavitake. I thank them for the knowledge they share with me. I need to thank

Dr. Edy Abou-Hamad for his contribution in lot of solid-state NMR experiments and for teaching me so many techniques. Special mention to Dr Youssef Saih for his fantastic training on dynamic flow catalytic reactions using PID reactors. I would like to thank Dr

Safwat Abdellatif and Professor Luigi Cavallo for their contribution in the DFT calculations we perform.

Finally, special mention goes to my friends Dr. Bilel Hamzaoui, Dr. Walid Almaksoud,

Baraa Werghi, Reem Alshareef, Dr. Kushal Bhatte, Dr Samir Barman, Dr Niladri Maity, Dr.

Janet Mohandas, Dr. Umesh Patil, Dr. Eva Pump, Dr. Anissa BendjeriouSedjerari, Gabriel

Jeantelot, Dr. Michael J. Kelly, Dr. Jullian Vittenet, Dr. Serena Goh, Dr. Jeff Espinas and

Dr.Julien SofackKreutzer., 7

DEDICATION

To the biggest contributors since I opened my eyes to this world, to Mom and Dad.

To the lady who I loved the most, my wife Rowaida and to the God’s gift for us, our son

Hassan.

To my special uncle Haj Mohammad Hamieh, I will always remember your advices, wisdom, and your white hands.

To my family and all the beloved ones…

TABLE OF CONTENTS

EXAMINATION COMMITTEE PAGE ...... 2 ABSTRACT ...... 4 DEDICATION ...... 7 TABLE OF CONTENTS ...... 8 LIST OF ABBREVIATIONS ...... 12 LIST OF SYMBOLS ...... 13 LIST OF ILLUSTRATIONS ...... 14 LIST OF SCHEMES...... 19 LIST OF TABLES ...... 21 CHAPTER 1: Bibliography ...... 23 Introduction...... 24 1.1. Overview on Catalysts ...... 24 1.2. Two main Types of catalysts: ...... 25 1.2.1. Homogeneous Catalysis ...... 25 1.2.2. Heterogeneous Catalysis ...... 25 1.3. Heterogeneous vs. Homogenous Catalysis ...... 25 1.4. Bridging the gap by SOMC ...... 26 1.5. History of SOMC ...... 27 1.6. SOMC complexes ...... 27 1.6.1. Support ...... 28 1.6.2. Grafted Metals ...... 30 1.6.3. Functional and spectator ligands ...... 30 1.7. Grafting a complex ...... 30 1.7.1 Reaction of Metal-Alkyl Complexes ...... 30 1.7.2. Reactivity of Metal Alkoxide ...... 32 1.7.3. Reactivity of Metal halide ...... 33 1.8. Characterization ...... 33 9

1.9. Some Applications of SOMC based complexes ...... 34 1.9.1. Metathesis of alkane ...... 34 1.9.2. Conversion of ethylene into propylene ...... 34 1.9.3. Olefin metathesis ...... 36 1.10. Transition metal alkylation ...... 37 1.10.1. Alkylating agents rise ...... 37 1.10.2. Importance of organometallic compounds ...... 37 1.10.3. Early transition metals coordinated to methyl ligands ...... 38 1.10.4. Early transition metals coordinated to neopentyl ligands ...... 40 1.11. The Novel approach to synthesize SOMC complexes ...... 41 1.12. References ...... 42

CHAPTER 2: Well-Defined Surface Species [(≡Si-O-)W(═O)Me3] Prepared by Direct

Methylation of [(≡Si-O-)W(═O)Cl3], a Catalyst for Cycloalkane Metathesis and Transformation of Ethylene to Propylene...... 52 2.1. Introduction ...... 53 2.1.1. Olefin metathesis ...... 53 2.1.2. History of Olefin Metathesis ...... 53 2.1.3. The Active species and mechanism ...... 54 2.1.4. Mimicking and enhancing the catalyst’s active site ...... 56 2.1.5. Homogenous catalysts ...... 56 2.1.6. Heterogeneous catalysts ...... 58 2.1.7. Selecting the route ...... 60 2.1.8. Surface alkylation strategy ...... 61 2.2. RESULTS AND DISCUSSION ...... 62

2.2.1. Preparation and characterization of [(≡Si-O-)W(=O)Me3] on SiO2-700: ...... 62 2.2.2. NMR Study ...... 64 2.2.3. DFT NMR calculations ...... 69 2.2.4. CATALYTIC STUDY ...... 72 2.3. CONCLUSION ...... 76 2.4. EXPERIMENTAL SECTION ...... 76 10

2.5. References ...... 83 CHAPTER 3: Preparation of Silica Supported Tungsten Oxo Imidazolin-2-iminato Methyl Precatalyst for Olefin Metathesis Reactions ...... 87 3.1 Introduction...... 88 3.1.1 Choosing the Ligand ...... 88 3.1.2 Preparation Path ...... 89 3.2 RESULTS AND DISCUSSION ...... 90 3.2.1. Preparation and characterization of complex 2...... 90 3.2.2. Preparation and characterization of complex 3...... 92 3.2.3. Preparation and characterization of complex 4...... 95 3.2.4. EXAFS Study ...... 98 3.3. Catalytic studies in olefin metathesis: ...... 101 3.3.1. Propene metathesis...... 101 3.3.2. 1-hexene metathesis...... 103 3.4. Conclusion ...... 105 3.5. EXPERIMENTAL SECTION ...... 105 3.5.1. NMR Spectra ...... 112 3.5.2. SXRD analysis of 2 ...... 115 3.5.3. Buried Volume Computational Details ...... 123 3.5.4. EXAFS Studies ...... 124 3.5.5. Catalysis Data ...... 126 3.5.6. NMR and GC Analysis of the Product Mixture of 1-hexene Self metathesis Reactions Catalyzed by 4, 5 and 6 Under Different Catalytic Conditions ...... 127 3.6. References ...... 133 CHAPTER 4: In Situ Generation of a Silica-Supported Tungsten Methyl Catalyst for Cyclooctane Metathesis ...... 136 4.1 Introduction...... 137 4.2 RESULTS AND DISCUSSION ...... 138 4.2.1 Methylation of 1 ...... 139 4.2.2. Optimizing the methylating condition of complex ...... 140 11

4.2.3. FTIR spectroscopy ...... 145 4.2.4. NMR Study ...... 145 4.2.5. Thermal Decomposition of 2 (2T) ...... 152 4.2.6. Hydrogenolysis of 2 (2H) ...... 152 4.2.7. DFT calculations ...... 154 4.2.8. Cyclooctane metathesis ...... 162 4.3 CONCLUSIONS ...... 165 4.4 EXPERIMENTAL SECTION ...... 165 4.5 References ...... 172 CHAPTER 5: Isolation of Silica Supported Niobium tetramethyl Complex by the SOMC Strategy: Synthesis, Characterization and Structure-Activity Relationship in Ethylene Oligomerization Reaction ...... 175 5.1 Introduction...... 176 5.1.1 Ethylene oligomerization: ...... 177 5.1.2 Ethylene oligomerization using SOMC-based catalyst ...... 177 5.1.3 Mechanism of ethylene oligomerization ...... 178 5.2 Results and Discussion ...... 180 v 5.2.1. Synthesis and Characterization of [(≡SiO-)Nb Cl3Me] (1) ...... 180 v 5.2.2. Synthesis and Characterization of complex 2 [(≡SiO)Nb Me4] ...... 183 5.2.3. Catalytic activity: ethylene oligomerization ...... 186 5.2.4. Catalytic Study: 1-octene isomerization ...... 188 5.3 Conclusion ...... 191 5.4 General Procedure ...... 191 5.5 Preparation procedure ...... 194 5.6 GC-Spectra after ethylene oligomerization using complex (1) & (2) ...... 197 5.7. References ...... 201 CAPTER 6: Conclusions and Outlooks ...... 203 6.1. Achievements ...... 204 6.2 Future Outlook ...... 206

LIST OF ABBREVIATIONS

Symbol /abbreviation Definition

CP/MAS Cross Polarization/ Magic Angle Spinning DCM DiChloroMethane Eq. equivalent EXAFS Extended X-Ray Absorption Fine Structure GC Gas Chromatography GC-MS Gas Chromatography mass spectrometry h hour(s) IR Infra-Red Me methyl mg milligram(s) MHz megahertz min minute(s) mL milliliter(s) mmol millimole(s) mol mole(s) Mw weight average molecular weight NMR Nuclear Magnetic Resonance Np Neopentyl (-CH2C(CH3)3) Ph Phenyl ppm part per million rt Room temperature SOMC Surface OrganoMetallic Chemistry SXRD Single Chrystal X-Ray Diffraction T Temperature TOF Turn over frequency TON Turn over number

LIST OF SYMBOLS

° degree δ Chemical shift

LIST OF ILLUSTRATIONS

Figure 1.1: Generic potential energy diagram showing the effect of a catalyst in a hypothetical exothermic chemical reaction X + Y to give Z. The presence of the catalyst opens a different reaction pathway (shown in red) with a lower activation energy.1 ...... 24 Figure 1.2: Model of Surface organometallic complexes...... 26 Figure 1.3: The general structure of surface organometallic compounds and the possibility of bearing different possible supports, transition metals, functional and spectator ligands...... 28

Figure 2.4: (A) SiO2-700 (B) after grafting of WOCl4 (C) after alkylation with ZnMe2...... 64 Figure 2.5: 1D 1H MAS solid-state NMR spectra of 3* acquired at 600 MHz at different temperatures with a 22 kHz MAS frequency and a repetition delay of 5 s and 8 scans...... 66 Figure 2.6: (A) 2D 1H-1H DQ/SQ and (B) 1H-1H TQ/SQ NMR spectra of 3* at room temperature (both acquired with number of scans per increment = 32 per t1 increment, repetition delay = 5 s, number of individual t1 increments = 128 ...... 67 Figure 2.7: One-dimensional 13C CP/MAS NMR spectra of 3* at different temperatures with a 10 kHz MAS frequency (a repetition delay = 5 s, contact time = 2 ms, line broadening = 80 Hz and number of scans = 10000.) ...... 68 Figure 2.8: 2D CP/MAS HETCOR NMR spectra of 3* acquired with short contact times of 0.2 ms under 8.5 kHz MAS frequency (number of scans per increment = 4000, repetition delay = 4 s, number of t1 increments = 32, line broadening = 80 Hz.) ...... 69 Figure 2.9: Ball-and-stick representations of the DFT molecular model of the -

W(=O)(Me)3 complex; two optimized conformations (A, B). Hydrogen atoms are omitted for clarity. Color code; W: cyan, C: gray, O: red and Si: orange. Geometric parameters are reported in Å for the bond distances and degrees for angles...... 70 Figure 2.10: GC chromatogram of cyclooctane metathesis products that been catalyzed by 3. Reaction condition: batch reactor, compound 3 (40 mg, 8.3 µmol, W loading: 4.83 wt%), cyclooctane (0.5 ml, 3.7 mmol), 160 h, 150oC. Conversion=60%, TON=260.

Products formed are ring-contracted (cC6), as well as ring-expanded species ranging from cC13 to cC29...... 72 Figure 3.11:Raman spectroscopy analysis of 2 ...... 91 Figure 3.12: Single crystal X-ray structure of 2 ( llipsoids at 50% probability level). Selected bond distances (Å) and angels (º): W(1)-Cl(1) = 2.361(2), W(1)-Cl(2) = 2.364(2), W(1)-Cl(3) = 2.357(2), W(1)-O(1) = 1.729(5), W(1)-N(1) = 1.782(5); O(1)-W(1)-N(1) = 108.7(3), O(1)-W(1)-Cl(3) = 95.2(2), N(1)-W(1)-Cl(3) = 95.5(2), O(1)-W(1)-Cl(1) = 94.0(2), N(1)-W(1)-Cl(1) = 92.9(2), Cl(3)-W(1)-Cl(1) = 165.08(7), O(1)-W(1)-Cl(2) = 123.0(2), N(1)- W(1)-Cl(2) = 128.3(2), Cl(3)-W(1)-Cl(2) = 82.11(7), Cl(1)-W(1)-Cl(2) = 83.02(7)...... 92 15

Figure 3.13: IR spectroscopy of complex 3, 4, and SiO2-700 ...... 94 Figure 3.14: (a) One-dimensional (1D) 1H MAS solid-state NMR spectrum of 4 acquired at 600 MHz (14.1 T) with a 22 kHz MAS frequency, a repetition delay of 5 s, and 8 scans. (b) Two-dimensional (2D) 1H−1H double-quantum (DQ)/single-quantum (SQ) and (c) 1H−1H triple- quantum (TQ)/SQ NMR spectra of 4 (both acquired with 32 scans per t1 13 increment, 5 s repetition delay, 32 individual t1 increments). (d) C CP/MAS NMR 1 spectrum of 4 acquired at 9.4 T (ʋo ( H) = 400 MHz) with a 10 kHz MAS frequency, 10000 scans, a 5 s repetition delay, and a 2 ms contact time. An exponential line broadening of 80 Hz was applied prior to Fourier transformation. (e) 2D 1H−13C CP/MAS dipolar

HETCOR spectrum of 4 (acquired at 9.4 T with 10 kHz MAS frequency, 4000 scans per t1 increment, a 4 s repetition delay, 64 individual t1 increments, and a 0.2 ms contact time)...... 97 3 Figure 3.15: Tungsten W LIII-edge k -weighted EXAFS (up) and corresponding Fourier transform (down, modulus and imaginary part) with a comparison to simulated curves Dipp Dipp for [(≡Si-O-)W(=O)Cl2-Im N] and [(≡Si-O-)W(=O)(CH3)2-Im N]. Solid lines: experimental; dashed lines: spherical wave theory...... 100 Figure 3.16: Product conversion and selectivity vs. time plot for the propene metathesis Dipp by [(≡Si-O-)W(=O)(CH3)2-Im N] (3050 equivalents/W) (TON=1165)...... 102 Dipp Figure 3.17: [(≡Si-O-)W(=O)(CH3)2-Im N] reaction with different amount of propene at 150°C, TON is calculated at different times...... 103 Figure 3.18: 1H (top) and 13C (bottom) NMR spectra of 1...... 112 Figure 3.19: 1H (top) and 13C (bottom) NMR spectra of 2...... 113 Figure 3.20: HSQC NMR spectrum of 2 ...... 114 Figure 3.21: Solid state 13C CP/MAS NMR spectrum (400 MHz, spinning rate 10 kHz, Dipp contact time = 1 ms, recycle delay = 2 sec, ns = 80 k) of [(≡Si-O-)W(=O)Cl2-Im N] (3)...... 114 3 Figure 3.22: Fourier transform of the k -weighted EXAFS at tungsten W LIII-edge (right) and its corresponding back Fourier transform (left) in the [1.5, 5.7] Å range with comparison to simulated curves for a tungsten metallic foil. Open symbol: experimental data; full line: fit ...... 124 Figure 3.23: Fourier transform of the k3-weighted EXAFS at tungsten W LIII-edge (right) and its corresponding back Fourier transform (left) in the [0.54, 3.9] Å range with comparison to simulated curves for compound 4. Models with and without chlorine shell are compared. Open symbol: experimental data; full line: fit...... 125 Figure 3.24: 1H (top) and the 13C (bottom) NMR spectra of the product mixture from 1- hexene self-metathesis reaction catalyzed by 4 at 80°C for 24 h...... 127 Figure 3.25: 1H (top) and the 13C (bottom) NMR spectra of the product mixture from the 1-hexene self-metathesis reaction catalyzed by 6 at 80 °C for 4.5 h...... 128 Figure 3.26: 1H NMR spectrum of pure (E)-Dec-5-ene...... 129 16

Figure 3.27: 1H NMR spectrum of pure (Z)-Dec-5-ene...... 129 Figure 3.28: 1H NMR spectrum of the mixture of pure (E)- and (Z)-Dec-5-ene...... 130 Figure 3.29: GC chromatogram of the product mixture from the 1-hexene self- metathesis reaction catalyzed by 4 at 80°C for 24 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product...... 130 Figure 3.30: GC chromatogram of the product mixture from the 1-hexene self- metathesis reaction catalyzed by 5 at 80 °C for 24 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product...... 131 Figure 3.31: GC chromatogram of the product mixture from the 1-hexene self- metathesis reaction catalyzed by 6 at 80°C for 24 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product...... 131 Figure 3.32: GC chromatogram of the product mixture from the 1-hexene self- metathesis reaction catalyzed by 6 at 80°C for 4.5 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product...... 132 Figure 4.33: One-dimensional 13C CP/MAS NMR spectra (600 MHz 1H frequency) of 1 after treating with carbon enriched MeLi, peaks at -1.7 and -7.4 ppm are attributed to -

Si-Me and -Si(Me)2...... 141 Figure 4.34: One-dimensional 13C CP/MAS NMR spectra (600 MHz 1H frequency) of 1 after treating with 2.5 equivalents of ZnMe2 in pentane...... 141 Figure 4.35: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with 5 equivalents of ZnMe2 in diethyl ether...... 142 Figure 4.36: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with 2.5 equivalents of ZnMe2 in diethyl ether...... 142 Figure 4.37: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with 2.5 equivalents of 97% carbon enriched ZnMe2 in dichloromethane for 15 min...... 143 Figure 4.38: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with 2.5 equivalents of carbon enriched ZnMe2 (97 %) in dichloromethane for 4 hours...... 143 Figure 4.39: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with >5 equivalents of carbon enriched ZnMe2 (97 %) in dichloromethane for 4 hours...... 144

Figure 4.40: (i) SiO2-700 (ii) WCl6 grafted with SiO2-700 (1) (iii), after alkylation of 1 with

ZnMe2 (2)...... 145 Figure 4.41: a) 1H MAS/ NMR of 2 . b) 2D 1H-1H DQ/SQ and 1H-1H TQ/SQ NMR spectra of

2 (both acquired with number of scans per increment = 32 per t1 increment, repetition 13 delay = 5 s, number of individual t1 increments = 128. c) C CP/MAS NMR spectra of 2* 17 d) 2D CP/MAS HETCOR NMR spectra of 2* acquired with short contact times of 0.2 ms under 8.5 kHz MAS frequency ...... 148 Figure 4.42: Comparison of 13C-SS NMR of (i) 4, obtained by thermal decomposition at

150°C of well-defined [(≡Si-O-)WMe5], 3. (ii) Reaction of 4 with dichloromethane. (iii)

The reaction of 4 with ZnMe2 in dichloromethane. (iv)The reaction of 4 with ZnCl2 in dichloromethane...... 149

Figure 4.43: Molecular models used for the (A) [(≡Si-O-)WMe4Cl] and (B) [(≡Si-O-)WMe5] species in DFT calculation...... 150 Figure 4.44: (A) 1D 13C CP/MAS SS-NMR spectrum of 2*, prepared by 13C-labeled dimethyl zinc (97% 13C-enriched) with a 10 kHz MAS frequency (a repetition delay = 5 s, contact time = 2 ms, line broadening = 80 Hz and number of scans = 10000). (B) 1D 13C CP/MAS NMR spectrum of 2* after heating at 150°C for 2 hours...... 153 Figure 4.45: FTIR spectrum of (i) 2 (ii) 2H (iii) 2H - 2...... 154

Figure 4.46: Interconversion of [(≡Si-O-)WMe4Cl] to [(≡Si-O-)WMe5] calculated using PBE/Def2 TZVP level of theory...... 155 Figure 4.47: FTIR spectrum of the evolved gasses during the reaction of 1 with excess

ZnMe2 in dichloromethane for one hour at room temperature ...... 156

Figure 4.48: Molecular models for the carbene [(≡Si-O-)W(=CH)Me3] (C) and carbyne

[(≡Si-O-)W(≡CH)Me2] (D) species used in our DFT calculations of NMR spectra and the reaction kinetics...... 158 Figure 4.49: (A) Energy profile calculated for the formation of carbene [(≡Si-O-

)W(=CH2)Me3] from [(≡Si-O-)WMe5] or [(≡Si-O-)WMe4Cl] (B) and the formation of the tungsten carbyne complex [(≡Si-O-)W≡CH)Me2] from its analogue carbene [(≡Si-O-

)W(=CH2)Me3]...... 159 Figure 4.50: GC-FID chromatogram of cyclooctane metathesis products that have been catalyzed by 2, 2T and 2H (See Table 4.18). Reaction conditions: batch reactor, catalyst 2 (0.035 g, W loading, 0.17 mmol g-1), cyclooctane (0.5 mL, 3.7 mmol), 120 h, 150 °C. Conversion = 13%, TON = 74. Products formed are ring-contracted (cC7) as well as ring- expanded species ranging from cC12 to cC30. For better understanding response for C7 was divided by factor of 4 in the chart...... 163

Figure 5.51: FTIR spectra of SiO2-700 (purple), complex 1 (yellow) and complex 2 (red)...... 181 Figure 5.52: a) 1H MAS NMR spectra of complex 1 b) 2D 1H−1H DQ/SQ and (c) 1H−1H TQ/SQ NMR spectra of complex 1 c) One-dimensional 13C CP/MAS NMR spectra of complex 1 d) 2D CP/MAS HETCOR NMR spectra of complex 1...... 182 Figure 5.53: a) 1H MAS NMR spectra of complex 2 b) 2D 1H−1H DQ/SQ and (c) 1H−1H TQ/SQ NMR spectra of complex 2 c) One-dimensional 13C CP/MAS NMR spectra of complex 2 d) 2D CP/MAS HETCOR NMR spectra of complex 2...... 184 18

Figure 5.54: a) 1H MAS NMR spectra of complex 2’ b) 2D 1H−1H DQ/SQ and (c) 1H−1H TQ/SQ NMR spectra of complex 2’ c) One-dimensional 13C CP/MAS NMR spectra of complex 2’ d) 2D CP/MAS HETCOR NMR spectra of complex 2’...... 186 Figure 5.55: Kinetic study of the isomerization and dimerization of 1-octene using complex 1 at room temperature (5 mg, 2µmol of Nb complex reacts with 5 mL of dried 1-octene in an ampoule inside the glovebox)...... 189 Figure 5.56: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #1 of [(≡SiO-)Nb Cl3Me] after addition to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 197 Figure 5.57: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #2 of [(≡SiO-)Nb Cl3Me] after addition to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 197 Figure 58: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #3 of [(≡SiO-)Nb Cl3Me] after addition to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 198 Figure 5.59:GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #4 of [(≡SiO-)Nb Cl3Me] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 198 Figure 5.60: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #1 of [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 199 Figure 5.61: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #2 of [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 199 Figure 5.62: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #3 of [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 200 Figure 5.63: GC spectrum of the liquid phase products after a catalytic oligomerization v reaction using 100 mg from batch #4 of [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC...... 200

LIST OF SCHEMES Scheme 1.1: Various hydroxyl groups on the surface of silica...... 29

Scheme 1.2: Synthesis of mono and bis-grafted group (IV) metal alkyls on SiO2-500 ...... 31

Scheme 1.3: Grafting of group-VI metal alkylidyne on the surface of SiO2-700...... 31

Scheme 1.4: Grafting of Ta(CH3)5 on SiO2-700...... 32

Scheme 1.5: Grafting of W(CH3)6 on silica partially dehydroxylated at 700 °C ...... 32

Scheme 1.6: Grafting of WCl6 on SiO2-700...... 33 Scheme 1.7: Mechanism of direct transformation of ethylene to propylene and its deactivation ...... 35

Scheme 1.8: Grafting of WO(Np)4 on SiO2-700 and then the formation of the active alkylidene by the release of one equiv. of neopentane per tungsten...... 36

Scheme 1.9: tungsten or molybdenum-imido-alkylidene complexes grafted on SiO2-700...... 37 Scheme 2.10: Overall olefin metathesis ...... 53 Scheme 2.11: Proposed mechanism for olefin metathesis...... 55

Scheme 2.12: Proposed active site on the WO3/SiO2 metathesis catalyst ...... 55

Scheme 2.13: a) Synthesis of (WO)Cl2(=CHR)(PR3)2 by tantalum tungsten ligand exchange reaction b) Synthesis of (WO)Cl2(=CHR)(PR3) by abstracting one PR3 by adding palladium complex that scavenges phosphine...... 56 Scheme 2.14: Synthesis of tungsten and molybdenum-oxo neopentyl by Osborn and Kress...... 57 Scheme 2.15: Synthesis of Tungsten oxo tetra neopentyl...... 57 Scheme 2.16: Group of different tungsten oxo alkylidene complexes for olefin metathesis...... 58

Scheme 2.17: Grafting of tungsten oxo tetra neopentyl on SiO2-700...... 59

Scheme 2.18: Grafting 1 on SiO2-700 yielding [(≡SiO)W(=O)(=CHCMe3)(OAr) and small amount of [(≡SiO)W(=O)(CHCMe3)(OAr)2 ...... 59 Scheme 2.19: Substituting aryl oxide by strongly -donating thiophenoxide in silica supported tungsten oxo catalyst improve catalyst efficiency in 1-alkene metathesis...... 60

Scheme 2.20: Structure of (Me4WO)2Mg(THF)4 ...... 60

Scheme 2.21: Synthesis of [(≡Si-O-)W(=O)Cl3] (2) by reaction of (1) with partially o dehydroxylated silica at 700 C (SiO2-700) ...... 61

Scheme 2.22: Synthesis of supported [(≡Si-O-)W(=O)Me3] (3) by surface methylation of (2) with 1.5 equivalents of dimethylzinc ...... 62 Scheme 2.23: Proposed Mechanism for the Direct Conversion of Ethylene to Propylene by [(≡Si-O-)W(=O)Me3] ...... 74

Scheme 3.24:(a) Proposed active site of WO3/SiO2 metathesis catalyst; (b) The tungsten oxo imidazolin-2-iminato precatalyst prepared in this chapter...... 90

Scheme 3.25: Synthesis of complex 2 [ImDippNWOCl3] ...... 91 20

Scheme 3.26: Grafting 2 on SiO2-700...... 93 Dipp Scheme 3.27: Synthesis of 4 [(≡Si-O-)W(=O)(CH3)2-Im N] by surface methylation of 3 with 1.5 equiv of ZnMe2...... 95

Scheme 4.28: Reported synthesis of [(≡Si-O-)WMe5] complex and the new proposed synthetic strategy developed on SiO2-700...... 138

Scheme 4.29: Synthesis of 1 by grafting WCl6 on partially dehydroxylated silica at 700 °C

(SiO2-700)...... 139

Scheme 4.30: Optimized condition for the alkylation of [(≡Si-O-)WCl5] ...... 144 Scheme 4.31: Proposed mechanism for secondary product formation...... 157 V Scheme 5.32: Synthesis of [(≡SiO)Ta Cl2Me2] (1)...... 178 Scheme 5.33: Proposed mechanism pathways for the formation of butane, ethane, methane, propylene, and active TaIII surface species ...... 179 V Scheme 5.34: Synthesis of [(≡SiO)Nb Cl3Me] (1)...... 180 V V Scheme 5.35: Synthesis of [(≡SiO)Nb Me4] (2) by the methylation of [(≡SiO)Nb Cl3Me] (1)...... 183 Scheme 5.36: Synthesis of complex (2’) by using 5 equivalent of ZnMe2 to methylate V complex [(≡SiO)Nb Cl3Me] (1)...... 185 Scheme 5.37: Proposed mechanism pathway for the isomerization of 1-octene with v complex 1 [(≡SiO)Nb Cl3Me]...... 190

LIST OF TABLES Table 1.1: Comparison of main advantages/ disadvantages of homogeneous vs. heterogeneous catalysts7-15 25

Table 2.2: DFT Calculated 13C Nuclear Magnetic Shielding and Chemical Shifts for the two possible geometries of complex [(≡Si-O-)W(=O)Me3] ...... 71

Table 2.3: Ethylene to Propylene Transformation Using [(≡Si-O-)W(=O)Me3] as Compared to the Analogue Tungsten Hydride and Tungsten Pentamethyl ...... 75 Dipp Table 3.4: EXAFS parameters obtained for [(≡Si-O-)W(=O)Cl2-Im N] and[(≡Si- Dipp O)W(=O)(CH3)2-Im N]...... 99 Dipp Table 3.5: Homo-coupling of 1-hexene using [(≡Si-O-)W(=O)(CH3)2-Im N] (4) and 15 16 compared with [(≡Si-O-)W(CH3)5] (5) and [(≡Si-O-)W(=O)(CH3)3] (6) ...... 104 Dipp Table 3.6: Crystal data and structure refinement for [Im NWOCl3] (2)...... 115 Table 3.7: Atomic coordinates (x 104) and equivalent isotropic displacement parameters (Å2x 103) ...... 116 Table 3.8: Bond lengths [Å] and angles [°] for [ImDippNWOCl3] (2)...... 117 Table 3.9: Anisotropic displacement parameters (Å2x 103) for material 2. The anisotropic displacement factor exponent takes the form: -2p2[ h2 a*2U11 + ... + 2 h k a* b* U12 ] ...... 121 Table 3.10: Hydrogen coordinates (x 104) and isotropic displacement parameters (Å2x Dipp 10 3) for [Im NWOCl3] (2)...... 122 Table 3.11: EXAFS parameters obtained for a tungsten metallic foil (Im-3m). A total of 19 scattering paths with 5 single and 14 scattering paths were used to fit the data in the [1.5, 5.7] Å R-range (multiple scattering paths are omitted for the sake of clarity). The cell parameter was determined equal to 3.157 ± 0.004 Å. The error propagation for the half length of each scattering path is not referenced in the table and underlined characters denote fixed parameters...... 124 Table 3.12: . EXAFS parameters obtained for compound 4 for two different models including or excluding a chlorine shell...... 125 Table 4.13: Optimization of methylating agents...... 140

Table 4.14: NMR-DFT calculations of the partial substitution of [(≡Si-O-) WCl5] ...... 151 13 Table 4.15: DFT-NMR C chemical shift of the carbene adduct species with ZnCl2 and

ZnMe2...... 160 13 Table 4.16: DFT-NMR C chemical shift of the carbyne adduct species with ZnCl2 and

ZnMe2...... 161 Table 4.17: Comparison of the activity of different precursors in cyclooctane metathesis reaction expressed by TON...... 163 22

Table 4.18: Summarized table is showing the GC area of the reaction of different complexes with cyclooctane inside an ampule at 150°C, TON was calculated. The area of each product by was added under its carbon number...... 164 Table 5.19: Oligomerization of ethylene catalyzed with different batches of complex v v [(≡SiO)Nb Cl3Me] 1, complex [(≡SiO)Nb Me4] 2 and reported tantalum complex v [(≡SiO)Ta Cl2Me2] ...... 187

CHAPTER 1: Bibliography

Introduction 1.1. Overview on Catalysts Catalysis is one of the most important fields of chemistry. The catalyst usually play a role in lowering the activation energy of a reaction (Figure 1.1). It is a science taught by nature, where many biocatalysts lead vital functions in living beings. In the last several decades, mimicking biocatalysts and designing new ones leads to an industrial revolution. Catalysts are categorized into three groups, bio, homogeneous and heterogeneous catalysts.

1.2. Two main Types of catalysts: 1.2.1. Homogeneous Catalysis It is named so because the catalyst and the substrate share the same phase. Usually, it is either in a gaseous state or liquid one. The industrial use of homogeneous catalysis is aimed to produce pharmaceuticals.2

1.2.2. Heterogeneous Catalysis The catalyst and the substrates occupy different phases. Usually, the catalyst in the form of a solid and the reactants are either in the gaseous or liquid state. Chemical and petrochemical industries rely mostly on heterogeneous catalysis,3 due to easiness to regenerate the highly expensive materials (Rhodium, Platinium…), and to reduce the cost of separating the products from the reaction mixture.4-6

Table 1.1: Comparison of main advantages/ disadvantages of homogeneous vs. heterogeneous catalysts7-

Character Heterogenous Homogeneous

Thermal stability Good Poor

Catalyst recovery Easy and cheap Difficult and expensive

Selectivity Good/poor Excellent/good

Active site Multiple active sites Single active site

Catalyst modification Relatively difficult Relatively easy

Reaction mechanism Hard to find out Easy to find out

1.3. Heterogeneous vs. Homogenous Catalysis The two types of catalysts exhibit different characters. Industrialized heterogeneous catalyst show better thermal stability and easier catalyst recovery but it have poor 26 selectivity due to the presence of multiple active sites, which are also different to modify.16 Here we make a table that summarizes the advantage and disadvantage of each type (Table 1.1). Knowing that exceptional cases could be available.

1.4. Bridging the gap by SOMC

Figure 1.2: Model of Surface organometallic complexes.

Chemical industry usually favors heterogeneous over homogeneous catalysis17 due to its increased thermal stability18-19 (reaction can be done at high temperature) and the facile catalyst recovery20-21 (re-use expensive metals and cleaner products). However, heterogeneous catalysis can display many active sites22-23 that cause lesser selectivity and make them harder to comprehend and to improve.24-26

To bring the concepts of homogeneous catalysis into heterogeneous catalysis, a new field of catalysis was developed and named surface organometallic chemistry27-32

(SOMC) (Figure 1.2).33-37 Organometallic complexes can react with the surface of metal oxides in controlled fashion and lead to new materials exhibiting single site grafted 27 metals.31, 38-41 The use of advanced characterization technique (SS-NMR and EXAF) had allowed to describe the coordination sphere of the grafted metals complexes thoroughly.42-45 This field had allowed to establish a structure-activity relationship and to propose reaction mechanisms.44-46

1.5. History of SOMC In the early 1960s, scientists suggested that the surface of the heterogeneous catalyst has a molecular character since molecular reactants transform into molecular products which need a molecular catalyst. Those investigations increase the acceptance of the “ organometallic character of surface intermediates involved in heterogeneous catalysis .”

This lead to the development of hybrid materials that combine organometallic fragments with support, this concept was highly investigated by molecular and surface science tools. Rising from this simple idea, a novel scientific approach emerge and was named “ surface organometallic chemistry ” as fruit from different intuitions and experimental overlaps.47

1.6. SOMC complexes Complexes prepared using SOMC comprise four different parts: the support, the metal, the functional ligands and the spectator ligands (Figure 1.3).

Figure 1.3: The general structure of surface organometallic compounds and the possibility of bearing different possible supports, transition metals, functional and spectator ligands.

1.6.1. Support Organometallic complexes can be grafted conveniently on a variety of supports, such as metal oxides (silicas,48 alumina,49-51 silica-alumina,52-53 zeolites.54 Silica, one of the most documented SOMC support, can be found in different types: silica gel, flame silica,

55 56 porous, and mesoporous silica (MCM-41, SBA-15 ). It is understood to contain SiO4 tetrahedra linked by siloxane bridges ≡Si-O-Si≡ and surface silanols (≡Si-OH). Hydrophilic hydroxyl groups are displayed on the surface of silica and are available for covalent grafting of organometallic complexes. Heating silica until 150°C under vacuum results in 29 dehydration but in dehydroxylation at higher temperatures. In this process, siloxane bridges form from adjacent silanols accompanied by the release of water.57-59

Scheme 1.1: Various hydroxyl groups on the surface of silica.

Dehydroxylated silica presents different types of hydroxyl groups on the surface. The surface population consists mainly of hydrogen-bonded silanol groups (-OH). Elimination of the H-bonding by condensation will leave a mixture of three types of silanol groups: germinal, vicinal, and isolated (Scheme 1.1). The active sites of the support are usually hydroxyl groups; what differs is their distribution depending on the treatment and the nature of the oxide under study. The formation of a covalent bond between the organometallic fragment and the support will then pass in most cases through a reaction with these hydroxyl groups. After dehydroxylation at 200°C, the surface hydroxyl groups exist primarily as vicinal pairs, while dehydroxylation at higher temperatures (450-500°C) leads to further condensation and only isolated hydroxyl groups remains on the surface. FTIR spectroscopy had allowed the identification of the surface silanols: isolated hydroxyl groups (free from interaction with water or other silanols) vibrates at 3747 cm-1 (v(SiO-H)). While hydroxyl groups bearing hydrogen bonds linking them to other surface hydroxyl groups have lower infrared frequencies, down to 30

3200 cm-1.60-62 Geminal OH groups with an IR absorption in a very close range to the isolated silanols (3745-3736 cm-1) are a minor component, but persist on the surface up to 800oC ( do not undergo internal condensation).

1.6.2. Grafted Metals Changing the transition metal will also change the character of catalyst, depending on the size of the metal, its oxidation state, coordination sphere, and many other factors.

Mainly transition metals were used in the preparation of SOMC based compounds.

1.6.3. Functional and spectator ligands A variety of functional and spectator ligands have been reported and have direct control over the grafted metal reactivities (Figure 1.3).63

1.7. Grafting a complex

Organometallic precursors can be grafted with an M1–X bond for which the reaction M–

OH + M1–X  [M1]–O–[M] + HX will be favored thermodynamically, where M–OH being the hydroxyl group of the support. Homoleptic organometallic complexes usually lead to only one surface complex to produce single site well-defined species over the entire support.3 Structural information can be gathered by solid-state NMR, gas quantification methods as well as elemental analysis.

1.7.1 Reaction of Metal-Alkyl Complexes

One of the first examples in this series was tris-neopentyl zirconium surface complex

64 [≡SiO-Zr(Np)3]. It was synthesized by sublimation of tetra neopentyl zirconium complex 31

[Zr(Np)4] onto the surface of partially dehydroxylated silica at 500°C. The reaction in the

65 case of zirconium or titanium leads to a monopodal species on SiO2-500 (Scheme 1.2).

Scheme 1.2: Synthesis of mono and bis-grafted group (IV) metal alkyls on SiO2-500 Similar to group-V metal alkyls, group-VI metal alkyls can undergo reaction with dehydroxylated silica, alumina, and silica-alumina. For instance, [W(≡CtBu)(CH2tBu)3] or

66 [Mo(≡CtBu)(CH2tBu)3] were grafted on SiO2-700. In general a pentane solution of an excess of [M(≡CtBu)(CH2tBu)3] (M = W, Mo) was added to SiO2-700 at room temperature to obtain [(≡SiO-)W(≡CtBu)(CH2tBu)2] or [(≡SiO-)Mo(≡CtBu)(CH2tBu)2] (Scheme 1.3).

Scheme 1.3: Grafting of group-VI metal alkylidyne on the surface of SiO2-700.

67 The surface can stabilize very reactive complexes such as [Ta(CH3)5], which is known to be very unstable at room temperature in solution. Grafting on SiO2-700 surface allowed a 32

68 similar observation with [W(Me)6]/SiO2-700. [Ta(CH3)5] reacts with SiO2-700 at -20°C and generates [(≡SiO-)Ta(CH3)4] monopodal species (Scheme 1.4).

Scheme 1.4: Grafting of Ta(CH3)5 on SiO2-700.

It was found that highly unstable [W(CH3)6] can be grafted on the oxide support to generate the corresponding stable grafted surface organometallic compound (Scheme

o 1.5). A pentane solution of [W(CH3)6] was allowed to react with SiO2-700 at -50 to -30 C for a couple of hours where golden yellow solid power is formed.

Scheme 1.5: Grafting of W(CH3)6 on silica partially dehydroxylated at 700 °C

1.7.2. Reactivity of Metal Alkoxide

The reaction of tantalum methoxide Ta(OMe)5 with silica, dehydroxylated at 700°C at room temperature.69 Characterization using solid-state carbon NMR shows clearly two signals for methoxy species (that means two different species). One is the tantalum ligand, and the second methoxy is the one tamed to the silica support (due to the reaction of methanol with strained siloxane bridge or silanols). In this case, the metal 33 plays a role in catalyzing this side reaction at RT, which results in a non-clean surface species.

1.7.3. Reactivity of Metal halide

This reaction is not preferred since released hydrogen chloride byproduct can further react with the grafted complex in a reversible manner or even open siloxane bridge in the case of dehydroxylated silica. Recently, many groups were developing simpler methods to tackle this problem, such as small dynamic vacuum during the grafting reaction or under argon flow with an outlet to release the gas generated. Many

70 71 72 73 complexes were grafted such as WCl6 (Scheme 1.6). WOCl4, VOCl3, NbCl5, TiOCl4,

74 CrO2Cl2 )

Scheme 1.6: Grafting of WCl6 on SiO2-700. 1.8. Characterization SOMC employed a range of analytical tools to characterize the active sites at a molecular level. To investigate the electronic states and the structures of the active sites that allow improvement in the catalysts, plenty of spectroscopic methods are extensively exploited. To attain this goal, infrared (IR), Raman, solid state NMR, UV-Vis and X-ray absorption (XAS) spectroscopies along with other surface science techniques and molecular modeling have been used.47, 75 34

1.9. Some Applications of SOMC based complexes 1.9.1. Metathesis of alkane Alkane Metathesis is a catalytic reaction involving successive breaking and formation of

C-H and C-C bonds of alkanes to give lower and higher homologs alkanes.76 The following general equation can describe metathesis reaction:

Since 1997, Basset et al. reported catalytic transformation of acyclic alkanes into their lower and higher homologs using silica supported metal hydrides77 or metal alkyls relatively at low temperature (150°C). This reaction was found to pass through three different steps, which are (1) activation C-H bond (2) formation of metallocarbene (3) olefin metathesis and finally (4) hydrogenation of the double bonds to form alkanes.

Different types of alkanes have been used in this reaction such as linear (gas, liquid), branched, and cyclic alkanes. Cross-metathesis between two different alkanes was successfully demonstrated by Basset et al. in 2001 using a mixture of toluene, ethane, propane and methane catalyzed by silica-supported tantalum hydride.78

1.9.2. Conversion of ethylene into propylene An increased demand for propylene mainly to be used in polypropylene (PP) manufacturers. Its current annual production is >80 million tons and is projected to grow over the next decade to reach more than 130 million tons.79 PP is widely used in automotive industry,80 construction,81 durables, and non-durables consumer,82 35 packaging83, and electronics.84 The primary source of propylene (PP’s building block) is by steam or catalytic cracking of natural gas and oil.85 However, these processes require large amounts of energy in addition to the high amount of byproducts. Alternative routes to generate propylene become more and more important to maintain the market supply.

In 2007, Basset et al. reported on purpose catalysis that generates propylene directly from ethylene by W-hydrides grafted on an alumina support.86 This tri-functional single- site catalyst operated via three consecutive reaction steps: (i) dimerization of ethylene to 1-butene (via the tungsten hydride moiety), (ii) 1-butene isomerization to 2-butene

(iii) cross-metathesis between ethylene and 2-butenes to give two molecules of propylene (Scheme 1.7).

Scheme 1.7: Mechanism of direct transformation of ethylene to propylene and its deactivation

This process is highly selective (95% of the products) with a cumulative turnover number of 1120 obtained after 120 h of reaction. Nevertheless, the tungsten hydride on alumina 36 catalyst found to undergo catalyst deactivation with time on stream because the blocking of catalyst’s active sites by polymers formation

1.9.3. Olefin metathesis

Motivated by classical olefin metathesis catalyst (WO3/SiO2 system), Basset, Gauvin and

Taoufik have proposed a similar, but the well-defined model to understand the nature of the active sites. Grafting of W(O)Np4 on dehydroxylated silica generate [(≡Si-O-

)(W)(O)Np3] that was highly active in propene metathesis reaching a 22000 TON in the dynamic reactor (Scheme 1.8).87

Scheme 1.8: Grafting of WO(Np)4 on SiO2-700 and then the formation of the active alkylidene by the release of one equiv. of neopentane per tungsten.

Similarly, tungsten- or molybdenum-imido-alkylidene complex were also grafted on silica. One example is the reaction of Schrock tungsten-carbene with silica partially

88 dehydroxylated at 700°C that yields W(=NAr)(=CHtBu)(CH2tBu)2 (Scheme 1.9). This complex showed high activity towards propene metathesis with TON equals 16,000 within 100 hours. 37

Scheme 1.9: tungsten or molybdenum-imido-alkylidene complexes grafted on SiO2-700. 1.10. Transition metal alkylation 1.10.1. Alkylating agents rise

It all started by the mid of 1850s, air-sensitive metal alkyl complexes such as ZnEt2,

89 HgEt2, SnEt4, and BMe3 was synthesized by Frankland. Those complexes are used as precursors for the synthesis of many organometallic compounds such as organochlorosilanes (by Friedel and Crafts).

Victor Grignard had established a reaction of magnesium with an organic halide,90 which leads into Grignard reagent RMgX. This reagent had an enormous impact on the preparation of transition metal organometallic chemistry. As an example, Grignard reagent (PhMgX) allowed Hein to prepare polyphenyl chromium derivative -

91 Ph)n].

1.10.2. Importance of organometallic compounds “Chemical compounds in which there is a single bond between a saturated carbon atom and a transition metal atom are of unusual importance,” Geoffery Wilkinson said in his

Nobel lecture in 1973. “During the time taken to deliver this lecture, tens of thousands of tons of chemical compounds are being transformed or synthesized industrially in the processes which at some stage involve a transition metal to carbon bond” he added. 38

During many catalytic reactions (Ziegler-Natta, Philipp’s, etc...), a metal-carbon single bond was formed, where the catalytic process requires a labile carbon ligand readily able to undergo a chemical reaction. Chemists, start looking for simple, stable metal compounds that have single bonds to saturated carbon.92 In the beginning, researcher failed to prepare and isolate stable metal-alkyl molecules. This failure was due to the presence of β-hydrogen that can transfer from the ethyl group bound to the metal to the metal center. A metal hydride coordinated by an olefin may be obtained in the case of transition metals. A solution was then to synthesize complexes that have no β- hydrogen, as in the case of methyl [-CH3], neopentyl [CH2C(CH3)3], and methyl-trimethyl silane [CH2Si(CH3)3].

1.10.3. Early transition metals coordinated to methyl ligands 1.10.3.1. Group 4 metals The first trials were at 1963 by Berthold and Groh, where they prepare a highly air- and moisture- sensitive tetramethyl titanium that undergoes thermal decomposition at temperatures higher than -78°C to form titanium metal. 93

MCl4 + 4 LiCH3  M(CH3)4 +4 LiCl (M=Ti, Zr, Hf)

Similarly, the same group prepared tetramethylzirconium which was stable up to -

15°C.94 Following the same strategy, Morrow reported the synthesis of tetramethylhafnium at 1970. Three years later, Wilkinson reported the tetramethylcromium.95 Apart from those unstable tetramethyl metal complexes (Ti, Zr,

Hf, Cr), octahedral adducts with stabilizing neutral ligands L were reported.96 This 39 stability was attributed to the role of the ligand in preventing inter or intramolecular decomposition by blocking coordination site and increasing metal electron density.

1.10.3.2. Group 5 and 6 metals Octahedral transition metals coordinated to methyl groups would be more stable than the tetramethyl-transition metal complexes. In 1973, starting from WCl6, Wilkinson

97 reported the synthesis of W(CH3)6 close to a 50% yield. This complex was thermally unstable and decomposed slowly at room temperature and had to be stored below -

40°C either under nitrogen or under vacuum. It is worth to add that hexamethyl tungsten reacts violently with atmospheric oxygen and may also detonate, under certain circumstances, in vacuo or under nitrogen or argon.

WCl6 + 3 LiCH3  W(CH3)6 + LiCl

Several years later Geoffrey Wilkinson published a new synthesis protocol of hexamethyl tungsten (VI) with increased yield.98-99

WCl6 + 3 Al(CH3)3  W(CH3)6 + 6 Al(CH3)2Cl

Richard Schrock, the 2005 Nobel laureate, investigated high oxidation state peralkyl complexes and had chosen to explore the organometallics chemistry of tantalum after his arrival at DuPont in 1972. The starting point for tantalum pentalkyl complex was the tantalum trimethyl dichloride species prepared by Juvinall.100

TaCl5 + 2.5 Zn(CH3)2  Ta(CH3)3Cl2+ ZnCl2 40

Schrock improved the synthesis by decreasing the amount of dimethyl zinc added to the

TaCl5 in pentane, followed by treating the product with two equivalents of methyl lithium in ether to produce the volatile, yellow, crystalline pentamethyl tantalum.101 This species was less stable than W(CH3)6 but much more stable than tetramethyl complexes

(Hf, Zr, Cr, Ti) as a consequence of simple steric reasons.

1.10.4. Early transition metals coordinated to neopentyl ligands Despite that methyl ligand lacks beta-hydrogen atom, transition metal-methyl complexes can decompose in a bimolecular fashion. The interaction is usually between the metal center and C-H bond in another metal complex, this is the reason behind the more stability of W(CH3)6 as compared to Ta(CH3)5.

To increase the stability of the pentalkyl complexes, Schrock studied compounds with neopentyl ligands instead of the small methyl ligands. The preparation of Ta(CH2CMe3)5 was targeted by the addition of two equivalents of [LiCH2CMe3] to Ta(CH2CMe3)3Cl2.

Instead of this complex, an orange, crystalline, and thermally stable 41

102 [(Me3CCH2)3Ta=CHCMe3] was formed, more probably due to α-hydrogen abstraction.

Those previously described complexes were reacted with different supports, showing some various activities, selectivity, as well as more stability as we earlier described in this chapter.

1.11. The Novel approach to synthesize SOMC complexes As described earlier, surface grafted metal alkyls represent the most important family among SOMCs, due to their ability to differentiate into many functional groups

(carbene, carbyne and hydride). Those complexes requires the synthesis of the organometallic precursor prior to grafting, thus limit its production to advanced organometallic labs. A new simpler synthetic strategy might have a high impact in improving the accessibility to this catalysis field. To achieve the goal, a synthetic method that comprises two stages is designed. The first step is grafting a metal halide complex on the support, knowing that those metal halide complexes are much cheaper, readilty available and have a better stablity as compared to the metal alkyls. It is important to note that, metal halide complexes could have any spectator ligand such as oxo, amido, aryl, alkoxy, etc. The grafting step will be followed by a selective alkylation to transform the metal-halide into a metal-alkyl group. In this thesis and using this strategy we will synthesize many new catalysts. Those new catalysts cannot be prepared by the classical synthesis of SOMCs, due to the fact that some of them are not isolated or unstable. The synthesized complexes will be characterized, and then used in various catalytic reactions. Also, we will show the limitations of this technique, and we will propose some suggestions to enhance it. 42

1.12. References 1. Masel, R. I., Chemical kinetics and catalysis. Wiley-Interscience New York: 2001.

2. Behr, A.; Neubert, P., Applied homogeneous catalysis. John Wiley & Sons: 2012.

3. Lefebvre, F., Synthesis of Well-defined Solid Catalysts by Surface Organometallic Chemistry. 2014.

4. Parravan, G., Thomas, J.M. Introduction to Principles of Heterogeneous Catalysis. Ind. Eng. Chem. 1967, 59 (10), 14.

5. Austin, J. M.; Groenewald, T.; Spiro, M., Heterogeneous Catalysis in Solution .18. The Catalysis by Carbons of Oxidation-Reduction Reactions. J. Chem. Soc. Dalton 1980, (6), 854-859.

6. Takehira, K.; Hayakawa, T.; Ishikawa, T., Heterogeneous Catalysis in the Liquid- Phase Oxidation of Olefins .4. The Activity of a Supported Vanadium or Chromium-Oxide Catalyst in the Decomposition of Tert-Butyl Hydroperoxide. Bull. Chem. Soc. Jpn. 1980, 53 (8), 2103-2110.

7. de Vries, J. G.; Jackson, S. D., Homogeneous and heterogeneous catalysis in industry. Catal. Sci. Technol. 2012, 2 (10), 2009-2009.

8. Gosselain, P. A.; Hardy, J. C.; Slegers, J.; Verbelen, R.; Simoens, G., Homogeneous and Heterogeneous Catalysis - Comparison of 2 Processes during Ethylene Hydrogenation in Presence of Same Catalytic Species. Bull. Soc. Chim. Belg. 1972, 81 (1- 2), 29.

9. Bond, G. C., Homogeneous and Heterogeneous Catalysis by Noble Metals. Adv. Chem. Ser. 1968, (70), 25.

10. Arai, M.; Zhao, F. Y., Metal Catalysts Recycling and Heterogeneous/Homogeneous Catalysis. Catalysts 2015, 5 (2), 868-870.

11. Luz, I.; Xamena, F. X. L. I.; Corma, A., Bridging homogeneous and heterogeneous catalysis with MOFs "Click" reactions with Cu-MOF catalysts. J. Catal. 2010, 276 (1), 134- 140.

12. Luz, I.; Xamena, F. X. L. I.; Corma, A., Bridging homogeneous and heterogeneous catalysis with MOFs: Cu-MOFs as solid catalysts for three-component coupling and cyclization reactions for the synthesis of propargylamines, indoles and imidazopyridines. J. Catal. 2012, 285 (1), 285-291.

13. Panigrahi, S.; Basu, S.; Praharaj, S.; Pande, S.; Jana, S.; Pal, A.; Ghosh, S. K.; Pal, T., Synthesis and size-selective catalysis by supported gold nanoparticles: Study on 43 heterogeneous and homogeneous catalytic process. J. Phys. Chem. C 2007, 111 (12), 4596-4605.

14. Bartz, M.; Kuther, J.; Seshadri, R.; Tremel, W., Colloid-bound catalysts for ring- opening metathesis polymerization: A combination of homogenous and heterogeneous properties. 1998, 37 (18), 2466-2468.

15. Astruc, D.; Lu, F.; Aranzaes, J. R., Nanoparticles as recyclable catalysts: The frontier between homogeneous and heterogeneous catalysis. Angew. Chem. Int. Ed. 2005, 44 (48), 7852-7872.

16. Leach, B., Applied industrial catalysis. Elsevier: 2012.

17. Hayes, M., Commercial Heterogeneous Catalysis HANDBOOK OF COMMERCIAL CATALYSTS - HETEROGENEOUS CATALYSTS. Platin Met Rev 2001, 45 (2), 83-83.

18. Haruta, M.; Tsubota, S.; Kobayashi, T.; Kageyama, H.; Genet, M. J.; Delmon, B., Low-Temperature Oxidation of Co over Gold Supported on TiO2, Alpha-Fe2O3, and Co3O4. J. Catal. 1993, 144 (1), 175-192.

19. Haruta, M.; Date, M., Advances in the catalysis of Au nanoparticles. Appl. Catal. a-Gen. 2001, 222 (1-2), 427-437.

20. Koukabi, N.; Kolvari, E.; Khazaei, A.; Zolfigol, M. A.; Shirmardi-Shaghasemi, B.; Khavasi, H. R., Hantzsch reaction on free nano-Fe2O3 catalyst: excellent reactivity combined with facile catalyst recovery and recyclability. Chem. Commun. 2011, 47 (32), 9230-9232.

21. Yang, H. Q.; Wang, Y. W.; Qin, Y.; Chong, Y. Z.; Yang, Q. Z.; Li, G.; Zhang, L.; Li, W., One-pot preparation of magnetic N-heterocyclic carbene-functionalized silica nanoparticles for the Suzuki-Miyaura coupling of aryl chlorides: improved activity and facile catalyst recovery. Green Chem. 2011, 13 (5), 1352-1361.

22. Wang, Q.; Weng, J. H.; Xu, L.; Fan, Z. Q.; Feng, L. X., Multiple active sites model of ethylene polymerization with the Cp2ZrCl2-aluminoxanes catalytic system. Polymer 1999, 40 (7), 1863-1870.

23. Dufaud, V.; Davis, M. E., Design of heterogeneous catalysts via multiple active site positioning in organic-inorganic hybrid materials. J. Am. Chem. Soc. 2003, 125 (31), 9403-9413.

24. Joglekar, M.; Nguyen, V.; Herring, A. M.; Trewyn, B. G., Single-site catalysts tethered onto monodispersed mesoporous carbon nanoparticles as active electrocatalyts for fuel-cell applications. Abstr. Pap. Am. Chem. S 2014, 247. 44

25. Maiwald, S.; Sommer, C.; Muller, G.; Taube, R., Highly active single-site catalysts for the 1,4-cis polymerization of butadiene from allylneodymium(III) chlorides and trialkylaluminiums - A contribution to the activation of tris(allyl)neodymium(III) and the further elucidation of the structure-activity relationship. Macromol. Chem. Physic. 2002, 203 (7), 1029-1039.

26. Tanaka, K. I.; Okuhara, T., Molecular Mechanisms of Catalytic Isomerization and Hydrogen-Exchange of Olefins over the MoS2 Single-Crystal Catalysts - Regulation of Catalytic Processes by the Conformation of Active-Sites. J. Catal. 1982, 78 (1), 155-164.

27. Efimov, O. A.; Minkov, A. I.; Zakharov, V. A.; Ermakov, Y. I., Catalysts Obtained by Reaction of Organometallic Compounds of Transition-Elements with a Support Surface - Deposited Catalysts for Polymerization of Ethylene Based on Tetrabenzyl Titanium. Kinet. Catal. 1976, 17 (4), 864-867.

28. Ryndin, Y. A.; Kuznetsov, B. N.; Moroz, E. M.; Tripolskii, A. A.; Ermakov, Y. I., Catalysts Obtained by Reaction of Transition-Metal Organometallic Compounds with Surface of Supports - Catalyst (W+Pt)-SiO2. Kinet. Catal. 1977, 18 (6), 1255-1259.

29. Ermakov, Y. I.; Kuznetsov, B. N.; Ryndin, Y. A., Catalysts Obtained by Reaction of Organometallic Compounds of Transition-Metals with Support Surface - Special Features in Hydrogenolysis of Ethane over Supported (W+Pd)-SiO2 and (Mo+Pd)-SiO2 Catalysts. Kinet. Catal. 1978, 19 (1), 131-137.

30. Lisitsyn, A. S.; Kuznetsov, V. L.; Ermakov, Y. I., Catalysts Obtained by the Reaction of Transition-Element Organometallic Compounds with Oxide-Support Surfaces - Catalytic Properties of Systems Prepared by the Pyrolysis of Co2(CO)6 on Oxide Supports in the Reaction Co+H2 Depending on Their Composition and Pretreatment. Kinet. Catal. 1982, 23 (4), 783-787.

31. Candy, J. P.; Ferretti, O. A.; Mabilon, G.; Bournonville, J. P.; Elmansour, A.; Basset, J. M.; Martino, G., Surface Organometallic Chemistry on Metals .2. Characterization of New Bimetallic Catalysts Generated by Reaction of Sn(n-C4H9)4 with Silica-Supported Rhodium. J. Catal. 1988, 112 (1), 210-220.

32. Ferretti, O. A.; Bournonville, J. P.; Mabilon, G.; Martino, G.; Candy, J. P.; Basset, J. M., Surface Organometallic Chemistry on Metals .4. Selective Hydrogenation of Ethyl- Acetate to Ethanol on Rh-Sn/SiO2 Bimetallic Catalysts - a Mechanistic Study. J. Mol. Catal. 1991, 67 (3), 283-294.

33. Dossi, C.; Fusi, A.; Psaro, R.; Roberto, D.; Ugo, R., Surface Organometallic Chemistry - a Molecular Approach to Heterogeneous Catalysis and a New Perspective for the Synthesis of Metal-Carbonyl Clusters. Mater. Chem. Phys. 1991, 29 (1-4), 191- 199. 45

34. Cariati, E.; Roberto, D.; Ugo, R.; Lucenti, E., The surface of inorganic oxides or zeolites as a nonconventional reaction medium for the selective synthesis of metal carbonyl complexes and clusters. Chem. Rev. 2003, 103 (9), 3707-3732.

35. Anwander, R., SOMC@PMS. Surface organometallic chemistry at periodic mesoporous silica. Chem. Mater. 2001, 13 (12), 4419-4438.

36. Psaro, R.; Ugo, R.; Zanderighi, G. M.; Besson, B.; Smith, A. K.; Basset, J. M., Surface-Supported Metal Cluster Carbonyls - Chemisorption, Decomposition and Reactivity of Os3(CO)12, H2OS3(CO)10 and Os6(CO)18 Supported on Silica and Alumina and the Investigation of the Fischer-Tropsch Catalysis with These Systems. J. Organomet. Chem. 1981, 213 (1), 215-247.

37. Smith, A. K.; Besson, B.; Basset, J. M.; Psaro, R.; Fusi, A.; Ugo, R., Surface Supported Metal Cluster Carbonyls - the Surface Organometallic Chemistry of Polymetallic and Monometallic Osmium Carbonyl Species Formed by Depositing Various Osmium Clusters on Silica and Alumina. J. Organomet. Chem. 1980, 192 (2), 31-34.

38. Didillon, B.; Candy, J. P.; Elmansour, A.; Houtmann, C.; Basset, J. M., The Impact of Surface Organometallic Chemistry in Heterogeneous Catalysis - a New Class of Highly Chemoselective Hydrogenation Catalysts, RhsSn(n-C4H9)2/SiO2. J. Mol. Catal. 1992, 74 (1- 3), 43-49.

39. Candy, J. P.; Didillon, B.; Smith, E. L.; Shay, T. B.; Basset, J. M., Surface Organometallic Chemistry on Metals - a Novel and Effective Route to Custom-Designed Bimetallic Catalysts. J. Mol. Catal. 1994, 86 (1-3), 179-204.

40. Ferretti, O. A.; Lucas, C.; Candy, J. P.; Basset, J. M.; Didillon, B.; Lepeltier, F., Surface Organometallic Chemistry on Metals - Kinetics of the Hydrogenolysis of Tetra N- Butyl Tin on Silica and Alumina-Supported Rhodium Catalysts. J. Mol. Catal. a-Chem. 1995, 103 (2), 125-132.

41. Niccolai, G. P.; Toyir, J.; Candy, J. P.; Leconte, M.; Basset, J. M., Surface Organometallic Chemistry - High Selectivity for the Dimerization of Propene on the Silica-Supported Bimetallic Catalyst, RhSNx/SiO2. Abstr. Pap. Am. Chem. S 1995, 210, 403.

42. Coperet, C.; Basset, J. M., Strategies to immobilize well-defined olefin metathesis catalysts: Supported homogeneous catalysis vs. surface organometallic chemistry. Adv. Synth. Catal. 2007, 349 (1-2), 78-92.

43. Millot, N.; Santini, C. C.; Lefebvre, F.; Basset, J. M., Surface organometallic chemistry: a route to well-defined boron heterogeneous co-catalyst for olefin polymerisation. Cr. Chim. 2004, 7 (8-9), 725-736. 46

44. Millot, N.; Baudouin, A.; Santini, C. C.; Basset, J. M.; Vautier, M., Design of heterogeneous polymerization catalysts by surface organometallic chemistry: Selective preparation of the well-defined highly electrophilic metallocenium cation [≡SiO- - + B(C6F5)3] [Cp*ZrMe2] supported on silica. Abstr. Pap. Am. Chem. S 2003, 225, U90-U91.

45. Chabanas, M.; Baudouin, A.; Coperet, C.; Basset, J. M., A highly active well- defined rhenium heterogeneous catalyst for olefin metathesis prepared via surface organometallic chemistry. J. Am. Chem. Soc. 2001, 123 (9), 2062-2063.

46. Jezequel, M.; Dufaud, V.; Ruiz-Garcia, M. J.; Carrillo-Hermosilla, F.; Neugebauer, U.; Niccolai, G. P.; Lefebvre, F.; Bayard, F.; Corker, J.; Fiddy, S.; Evans, J.; Broyer, J. P.; Malinge, J.; Basset, J. M., Supported metallocene catalysts by surface organometallic chemistry. Synthesis, characterization, and reactivity in ethylene polymerization of oxide-supported mono- and biscyclopentadienyl zirconium alkyl complexes: Establishment of structure/reactivity relationships. J. Am. Chem. Soc. 2001, 123 (15), 3520-3540.

47. Basset, J. M.; Psaro, R.; Roberto, D.; Ugo, R., Modern surface organometallic chemistry. John Wiley & Sons: 2009.

48. Bilhou, J. L.; Bilhoubougnol, V.; Graydon, W. F.; Basset, J. M.; Smith, A. K., Surface-Linked Metal Cluster Carbonyls - Fixation, Behavior and Reactivity of Rh6(CO)16 Bonded to a Phosphinated Silica. J.Mol. Catal. 1980, 8 (4), 411-429.

49. Nedez, C.; Lefebvre, F.; Basset, J. M., Surface organometallic chemistry of germanium: Infrared, CP-MAS C-13 solid state NMR, and analytical evidence for the formation of highly thermally stable -O-Ge(n-C4H9)3 grafted fragments by reaction of Ge(n-C4H9)4 with the surface of partially dehydroxylated alumina and silica-alumina. Langmuir 1996, 12 (4), 925-929.

50. Dufour, P.; Houtman, C.; Santini, C. C.; Basset, J. M., Surface Organometallic Chemistry on Oxides - Reaction of Hydrogen with Bis(Allyl)Rhodium Grafted to Silica, Titania and Alumina. J. Mol. Catal. 1992, 77 (3), 257-272.

51. Theolier, A.; Smith, A. K.; Leconte, M.; Basset, J. M.; Zanderighi, G. M.; Psaro, R.; Ugo, R., Surface Supported Metal Cluster Carbonyls - Chemisorption Decomposition and Reactivity of Rh4(CO)12 Supported on Silica and Alumina. J. Organomet. Chem. 1980, 191 (2), 415-424.

52. Le Roux, E.; Taoufik, M.; Baudouin, A.; Coperet, C.; Thivolle-Cazat, J.; Basset, J. M.; Maunders, B. M.; Sunley, G. J., Silica-alumina-supported, tungsten-based heterogeneous alkane metathesis catalyst: Is it closer to a silica or an alumina- supported system? Adv. Synth. Catal. 2007, 349 (1-2), 231-237. 47

53. Hugues, F.; Basset, J. M.; Bentaarit, Y.; Choplin, A.; Primet, M.; Rojas, D.; Smith, A. K., Surface Organo-Metallic Chemistry - Formation of HFe3(CO)11- from Fe3(CO)12 and Fe(CO)5 on Silica, Alumina, Magnesia, and Zinc-Oxide. J. Am. Chem. Soc. 1982, 104 (25), 7020-7024.

54. Theolier, A.; Custodero, E.; Choplin, A.; Basset, J. M.; Raatz, F., A Surface Organometallic Approach for Controlling the Pore Opening Size of Zeolites - the Case of Mordenite Modified Via Grafting of Dineopentylmagnesium. Angew. Chem. Int. Ed. 1990, 29 (7), 805-807.

55. Wang, X. X.; Lefebvre, F.; Patarin, J.; Basset, J. M., Synthesis and characterization of zirconium containing mesoporous silicas I. Hydrothermal synthesis of Zr-MCM-41- type materials. Micropor. Mesopor. Mat. 2001, 42 (2-3), 269-276.

56. Bendjeriou-Sedjerari, A.; Pelletier, J. D. A.; Abou-Hamad, E.; Emsley, L.; Basset, J. M., A well-defined mesoporous amine silica surface via a selective treatment of SBA-15 with ammonia. Chem. Commun. 2012, 48 (25), 3067-3069.

57. Volkov, A. V.; Kiselev, A. V.; Lygin, V. I.; Ryabenko, E. A.; Chalumov, B. Z.; Shimichev, V. S., Study of Silica Dehydroxylation and Sintering by Means of Infrared Spectroscopy. Colloid. J. USSR. 1979, 41 (2), 260-263.

58. Kiselev, A. V.; Kuznetsov, B. V.; Lanin, S. N., Chemical Modification of Silica-Gels with Different Degrees of Surface Dehydroxylation. Colloid. J. USSR. 1976, 38 (1), 142- 146.

59. Aristov, B. G.; Kiselev, A. V., Effect of the Dehydroxylation of Silica Surface on the Heats of Adsorption of Nitrogen and Avgon Vapors. Zh. Fiz. Khim. 1964, 38 (8), 1984- 1989.

60. dos Santos, J. H. Z.; Krug, C.; da Rosa, M. B.; Stedile, F. C.; Dupont, J.; Forte, M. D., The effect of silica dehydroxylation temperature on the activity of SiO2-supported zirconocene catalysts. J. Mol. Catal. a-Chem. 1999, 139 (2-3), 199-207.

61. Dyakonov, S. S.; Lygin, V. I.; Shalumov, B. Z.; Khlopova, Z. G.; Shimicheva, P. A., Ir- Spectroscopic Investigation of the Dehydroxylation of Silicas Alloyed with Aluminum- Oxide. Colloid J. USSR 1985, 47 (1), 119-122.

62. Kazanskii, V. B., Mechanism of Dehydroxylation of Naturally-Occurring High-Silica Zeolites Involving the Formation of Lewis Acid Sites. Kinet. Catal. 1987, 28 (3), 482-489.

63. Samantaray, M. K.; Dey, R.; Kavitake, S.; Basset, J. M., New Concept of C–H and C–C Bond Activation via Surface Organometallic Chemistry. In C-H Bond Activation and Catalytic Functionalization II, Springer: 2015; 155-187. 48

64. Quignard, F.; Lecuyer, C.; Bougault, C.; Lefebvre, F.; Choplin, A.; Olivier, D.; Basset, J. M., Surface Organometallic Chemistry - Synthesis and Characterization of a Tris(Neopentyl)Zirconium(IV) Complex Grafted to the Surface of a Partially Dehydroxylated Silica. Inorg. Chem. 1992, 31 (5), 928-930.

65. Rosier, C.; Niccolai, G. P.; Basset, J. M., Catalytic hydrogenolysis and isomerization of light alkanes over the silica-supported titanium hydride complex [(≡SiO)3TiH]. J. Am. Chem. Soc. 1997, 119 (50), 12408-12409.

66. Le Roux, E.; Taoufik, M.; Chabanas, M.; Alcor, D.; Baudouin, A.; Coperet, C.; Thivolle-Cazat, J.; Basset, J. M.; Lesage, A.; Hediger, S.; Emsley, L., Well-defined surface tungstenocarbyne complexes through the reaction of [W(=CtBu)(CH)2tBu)3] with silica. Organometallics 2005, 24 (17), 4274-4279.

67. Chen, Y.; Abou-hamad, E.; Hamieh, A.; Hamzaoui, B.; Emsley, L.; Basset, J. M., Alkane Metathesis with the Tantalum Methylidene [(≡SiO)Ta(=CH2)Me2]/[(≡SiO)2)Ta(=CH2)Me] Generated from Well-Defined Surface Organometallic Complex [(≡SiO)(TaMe4)Me]. J. Am. Chem. Soc. 2015, 137 (2), 588-591.

68. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Rossini, A. J.; Widdifield, C. M.; Dey, R.; Emsley, L.; Basset, J. M., WMe6 Tamed by Silica: [(≡Si-O-)WMe5] as an Efficient, Well-Defined Species for Alkane Metathesis, Leading to the Observation of a Supported W-Methyl/Methylidyne Species. J. Am. Chem. Soc. 2014, 136 (3), 1054-1061.

69. Saint-Arroman, R. P.; Didillon, B.; de Mallmann, A.; Basset, J. M.; Lefebvre, F., Deperoxidation of cyclohexyl hydroperoxide by silica-supported alkoxo-tantalum complexes. Appl. Catal. a-Gen. 2008, 337 (1), 78-85.

70. Hamieh, A.; Chen, Y.; Abdel-Azeim, S.; Abou-hamad, E.; Goh, S.; Samantaray, M.; Dey, R.; Cavallo, L.; Basset, J. M., Well-Defined Surface Species [(≡Si-O)W(=O)Me3)] Prepared by Direct Methylation of [(≡Si-O-)W(=O)Cl3)], a Catalyst for Cycloalkane Metathesis and Transformation of Ethylene to Propylene. Acs. Catal. 2015, 5 (4), 2164- 2171.

71. Zhu, H. B.; Ould-Chikh, S.; Dong, H. L.; Llorens, I.; Saih, Y.; Anjum, D. H.; Hazemann, J. L.; Basset, J. M., VOx/SiO2 Catalyst Prepared by Grafting VOCl3 on Silica for Oxidative Dehydrogenation of Propane. Chem. cat. chem. 2015, 7 (20), 3332-3339.

72. D'Elia, V.; Dong, H. L.; Rossini, A. J.; Widdifield, C. M.; Vummaleti, S. V. C.; Minenkov, Y.; Poater, A.; Abou-Hamad, E.; Pelletier, J. D. A.; Cavallo, L.; Emsley, L.; Basset, J. M., Cooperative Effect of Monopodal Silica-Supported Niobium Complex Pairs Enhancing Catalytic Cyclic Carbonate Production. J. Am. Chem. Soc. 2015, 137 (24), 7728-7739. 49

73. Aronson, B. J.; Blanford, C. F.; Stein, A., Solution-phase grafting of titanium dioxide onto the pore surface of mesoporous silicates: Synthesis and structural characterization. Chem. Mater. 1997, 9 (12), 2842-2851.

74. Demmelmaier, C. A.; White, R. E.; van Bokhoven, J. A.; Scottt, S. L., Nature of SiOCrO2Cl and [(≡SiO)2CrO2] sites prepared by grafting CrO2Cl2 onto silica. J. Phys. Chem. C 2008, 112 (16), 6439-6449.

75. Coperet, C.; Comas-Vives, A.; Conley, M. P.; Estes, D. P.; Fedorov, A.; Mougel, V.; Nagae, H. N ez-Zarur, F.; Zhizhko, P. A., Surface Organometallic and Coordination Chemistry toward Single-Site Heterogeneous Catalysts: Strategies, Methods, Structures, and Activities. Chem. Rev. 2016, 116 (2), 323-421.

76. Basset, J. M.; Coperet, C.; Soulivong, D.; Taoufik, M.; Thivolle-Cazat, J., From olefin to alkane metathesis: A historical point of view. Angew. Chem. Int. Edit. 2006, 45 (37), 6082-6085.

77. Vidal, V.; Theolier, A.; ThivolleCazat, J.; Basset, J. M., Metathesis of alkanes catalyzed by silica-supported transition metal hydrides. Science 1997, 276 (5309), 99- 102.

78. Coperet, C.; Maury, O.; Thivolle-Cazat, J.; Basset, J. M., sigma-bond metathesis of alkanes on a silica-supported tantalum(V) alkyl alkylidene complex: First evidence for alkane cross-metathesis. Angew. Chem. Int. Edit. 2001, 40 (12), 2331-2334.

79. Tullo, A. H., Propylene on demand - A tight market for propylene globally is breathing new life into propane dehydrogenation. Chem. Eng. News 2003, 81 (50), 15- 16.

80. Hong, C. H.; Lee, Y. B.; Bae, J. W.; Jho, J. Y.; Nam, B. U.; Hwang, T. W., Performance of polypropylene/clay nanocomposites for automotive parts applications. J. Ind. Eng. Chem. 2005, 11 (1), 76-82.

81. Cowperthwait, J. D.; Congdon, J. P.; Phillips, E. M., Polypropylene based resins designed for the building and construction industry. Vinyl. Retec. '97 1997, 41.

82. Wu, J. J.; Lee, G. J.; Chen, Y. S.; Hu, T. L., The synthesis of nano- silver/polypropylene plastics for antibacterial application. Curr. Appl. Phys. 2012, 12, S89-S95.

83. Khalaj, M. J.; Ahmadi, H.; Lesankhosh, R.; Khalaj, G., Study of physical and mechanical properties of polypropylene nanocomposites for food packaging application: Nano-clay modified with iron nanoparticles. Trends Food Sci. Tech. 2016, 51, 41-48.

84. Galashina, N. M.; Nedorezova, P. M.; Shevchenko, V. G.; Tsvetkova, V. I.; Klyamkina, A. N.; Chmutin, I. A.; Ponomarenko, A. T.; Dyachkovskii, F. S., 50

Polymerizationally Filled Electricity and Heat-Conducting Polypropylene Graphite Composites - Some Specific Features of Their Properties. Vysokomol. Soedin. 1993, 35 (8), A1315-A1319.

85. Zhu, X. L.; Jiang, S.; Li, C. Y.; Chen, X. B.; Yang, C. H., Residue Catalytic Cracking Process for Maximum Ethylene and Propylene Production. Ind. Eng. Chem. Res. 2013, 52 (40), 14366-14375.

86. Taoufik, M.; Le Roux, E.; Thivolle-Cazat, J.; Basset, J. M., Direct transformation of ethylene into propylene catalyzed by a tungsten hydride supported on alumina: Trifunctional single-site catalysis. Angew. Chem. Int. Ed. 2007, 46 (38), 7202-7205.

87. Mazoyer, E.; Merle, N.; de Mallmann, A.; Basset, J. M.; Berrier, E.; Delevoye, L.; Paul, J. F.; Nicholas, C. P.; Gauvin, R. M.; Taoufik, M., Development of the first well- defined tungsten oxo alkyl derivatives supported on silica by SOMC: towards a model of WO3/SiO2 olefin metathesis catalyst. Chem. Commun. 2010, 46 (47), 8944-8946.

88. Rhers, B.; Salameh, A.; Baudouin, A.; Quadrelli, E. A.; Taoufik, M.; Coperet, C.; Lefebvre, F.; Basset, J. M.; Solans-Monfort, X.; Eisenstein, O.; Lukens, W. W.; Lopez, L. P. H.; Sinha, A.; Schrock, R. R., A well-defined, silica-supported tungsten imido alkylidene olefin metathesis catalyst. Organometallics 2006, 25 (15), 3554-3557.

89. Seyferth, D., Zinc alkyls, Edward Frankland, and the beginnings of main-group organometallic chemistry. Organometallics 2001, 20 (14), 2940-2955.

90. Grignard, V., Regarding some new organometallic combinations of magnesium and their application to alcohol synthesis and hydrocarbides. Cr. Hebd. Acad. Sci. 1900, 130, 1322-1324.

91. Hein, F.; Eissner, W., Concerning the tetraphenyl chromium (C6H5)4Cr. (Communication concerning organic chromium compounds). Ber. Dtsch. Chem. Ges. 1926, 59, 362-366.

92. Wilkinson, G., Long Search for Stable Transition-Metal Alkyls. Science 1974, 185 (4146), 109-112.

93. Berthold, H. J.; Groh, G., Uber Die Isolierung Von Tetramethyltitan. Z. Anorg. Allg. Chem. 1963, 319 (5-6), 230-235.

94. Berthold, H. J.; Groh, G., Preparation of Tetramethylzirconium Zr(CH3)4. Angew. Chem. Int. Ed. 1966, 5 (5), 516.

95. Mowat, W.; Shortland, A. J.; Hill, N. J.; Wilkinson, G., Elimination Stabilized Alkyls .2. Neopentyl and Related Alkyls of Chromium(IV). J. Chem. Soc. Dalton 1973, (7), 770- 778. 51

96. Cuenca, T.; Galakhov, M.; Jimenez, G.; Royo, E.; Royo, P.; Bochmann, M., Dinuclear cationic zirconium complexes with the fulvalene ligand. Synthesis and reactivity. J. Organomet. Chem. 1997, 543 (1-2), 209-215.

97. Shortlan, A. J.; Wilkinson, G., Preparation and Properties of Hexamethyltungsten. J. Chem. Soc. Dalton 1973, (8), 872-876.

98. Galyer, A. L.; Wilkinson, G., New Synthesis of Hexamethyltungsten(VI) - Octamethyltungstate(VI) Ion. J. Chem. Soc. Dalton 1976, (21), 2235-2238.

99. Galyer, L.; Mertis, K.; Wilkinson, G., New Synthesis of Hexamethyltungsten(VI) - Hexamethylrhenium(VI) and Dioxotrimethylrhenium(VII). J. Organomet. Chem. 1975, 85 (3), C37-C38.

100. Juvinall, G. L., Sigma-Bonded Alkyl Compounds of Niobium + Tantalum . Trimethyldichloroniobium + Trimethyldichlorotantalum. J. Am. Chem. Soc. 1964, 86 (19), 4202-4203.

101. Schrock, R. R.; Meakin, P., Pentamethyl Complexes of Niobium and Tantalum. J. Am. Chem. Soc. 1974, 96 (16), 5288-5290.

102. Schrock, R. R., Alkylcarbene Complex of Tantalum by Intramolecular Alpha- Hydrogen Abstraction. J. Am. Chem. Soc. 1974, 96 (21), 6796-6797.

CHAPTER 2: Well-Defined Surface Species [(≡Si-O-)W(═O)Me3] Prepared by Direct Methylation of [(≡Si-O-)W(═O)Cl3], a Catalyst for Cycloalkane Metathesis and Transformation of Ethylene to Propylene.

2.1. Introduction

The following work was done on the methylation of silica supported WOCl4. The objective was to tune the ligands (halogen) of the grafted precursor (WOCl4) in order to generate a well-defined tungsten oxo trimethyl complex. This grafting is carried out by

SOMC approach and then trials of surface methylation are attempted. The resulted complex was used for catalyzing several reactions, such as alkane and alkene metathesis in addition to the direct transformation of ethylene into propylene.

2.1.1. Olefin metathesis “Olefin metathesis” reaction is one of the most important and fundamental tools for organic chemistry that was discovered a half century back (Scheme 2.10).1

Scheme 2.10: Overall olefin metathesis It has opened new industrial routes in different fields of petro-chemistry2, energy3, polymers4, oleochemicals, drugs5, etc. Yves Chauvin, Robert H. Grubbs, and Richard R.

Schrock were collectively awarded the 2005 Nobel Prize in Chemistry due to their contribution in clarifying the reaction mechanism as well as the synthesis of a variety of highly active and selective catalysts.

2.1.2. History of Olefin Metathesis Olefin metathesis was accidentally discovered at Phillips Petroleum Company in the

1960s using impregnated metal oxo compounds on a support. Their initial investigations 54

6 focused on systems such as MoO3/Al2O3, Mo(CO)6/Al2O3, WO3/Al2O3 and W(CO)6/Al2O3.

WO3/SiO2 was the first heterogeneous catalyst used in the industrial application of

7 olefin metathesis (Phillips Triolefin process) . WO3/SiO2 was favored catalyst in industrial processes and is operating since 1966 due to (i) a unique poison resistance. (ii) The catalyst can be regenerated at the operating temperature. (2) The catalyst can be continuously regenerated.(iv) Isomerization activity toward equilibrated components.

To prepare this catalyst, water solutions of ammonium tungstate were impregnating on various silicas. The catalyst was then pretreated with air at a temperature ranging between 700 to 810°K followed by nitrogen flushing. Propylene metathesis was carried out in a dynamic reactor at high temperature (around 650°K) to convert into ethylene and 2-butenes.

2.1.3. The Active species and mechanism In 1971 Hérisson and Chauvin proposed a mechanism8 for the olefin metathesis using a mixture of metal halide (WOCl4) and an alkylating agent (Sn(C4H9)4, or Al(C2H5)2Cl) in which a metal carbene is an active species (Scheme 2.11). 55

Scheme 2.11: Proposed mechanism for olefin metathesis. Relying on the mechanism of Chauvin, Grünert et al. proposed a structure9 for the active isolated sites of the WO3/Al2O3 catalyst, which consist of a W(VI) bearing two aluminoxy, an oxo, and a carbene ligand. It was then postulated that the isolated active species of the WO3/SiO2 catalyst may thus have a similar structure (Scheme 2.12).

Scheme 2.12: Proposed active site on the WO3/SiO2 metathesis catalyst

It was noticed that the complexes with tungsten loading exceed 10 wt% shows WO3 crystallites (by XRD, RAMAN spectroscopy).10 Those crystallites are not involved in the olefin metathesis reaction. It seems that the preparation method (aqueous 56 impregnation) is behind the poor dispersion and the small number of active sites of the catalyst on the support.

2.1.4. Mimicking and enhancing the catalyst’s active site In order to increase the number of active sites and extend catalyst lifetime, researcher’s efforts moved rapidly to isolate the metal’s active site (structure-activity relationship).

Molybdenum and tungsten alkylidenes in the form of oxo complexes have been synthesized and used since then.

2.1.5. Homogenous catalysts Parallel to what been developed in heterogeneous catalysis early in 1980s Schrock et al. reported the first Tungsten oxo-alkylidene complex which is

Scheme 2.13: a) Synthesis of (WO)Cl2(=CHR)(PR3)2 by tantalum tungsten ligand exchange reaction b) Synthesis of (WO)Cl2(=CHR)(PR3) by abstracting one PR3 by adding palladium complex that scavenges phosphine. synthesized by metal-metal ligand substitution (Scheme 2.13), and have been used in olefin metathesis. Other complexes were then synthesized by tuning the ligands to modify activity and selectivity. In a direct study of the solutions of the most reactive catalysts in olefin metathesis, Osborn and Kress found11 that the presence of oxo-ligands on the metal has a dramatic effect on catalytic activity. They investigated the reaction of 57

WOCl4 and MoOCl4 with dineopentylmagnesium where they isolated the WOCl(Np)3 after reacting an ether solution of WOCl4 with 1.5 equiv dineopentylmagnesium(dioxane) ( Scheme 2.14).

Scheme 2.14: Synthesis of tungsten and molybdenum-oxo neopentyl by Osborn and Kress.

WO(Np)4 was also isolated after the reaction with two equiv. of MgNp2(dioxane), similarly, MoCl(Np)3 and Mo(Np)4 were isolated and characterized in terms of NMR, a mass spectrometer, and IR spectroscopy. But Osborn and Kress found that WONP3(ONp)

12 was mistakenly assigned as WONp4. Three decades later Basset et al. inspired by

Schrock’s synthesis13 of

Scheme 2.15: Synthesis of Tungsten oxo tetra neopentyl.

W(NAr)(Np)2(CHtBu) from W(NAr)(Np)3Cl and NpLi succeeded to synthesize and isolate

WO(Np)4 by reacting WOCl(Np)3 with NaNp to afford a tractable product WO(Np)4 as a colourless, light- and highly moisture- sensitive oil (Scheme 2.15).14 Since the report of 58

Osborn scientists have been preparing complexes with an oxo ligand to enhance the activity in olefin metathesis (Scheme 2.16).

Scheme 2.16: Group of different tungsten oxo alkylidene complexes for olefin metathesis.

2.1.6. Heterogeneous catalysts Parallel to the development of complexes with oxo ligand as homogeneous catalysts for olefin metathesis, well-defined heterogeneous catalyst were established in a slower manner. In 2010 Taoufik et al developed the first well-defined tungsten oxo alkyl

14 derivatives supported on silica by SOMC. WONp4 was grafted on dehydroxylated silica at 700°C (SiO2-700) to give a monopodal well-defined and single-site complex

[(≡SiO)(W=O)Np3] (Scheme 2.17). Remarkably, its catalytic performance surpasses that of the parent imido derivatives, underlying the importance of the oxo ligand to the design of robust catalysts. Using [(≡SiO)(WO)Np3] a cumulated TON of 22000 was achieved in the dynamic reactor for propene metathesis. 59

Scheme 2.17: Grafting of tungsten oxo tetra neopentyl on SiO2-700.

Several years later Copéret et al. prepared SOMC based complex using recently reported molecular defined oxo alkylidene (Scheme 2.18).15

Scheme 2.18: Grafting 1 on SiO2-700 yielding [(≡SiO)W(=O)(=CHCMe3)(OAr) and small amount of [(≡SiO)W(=O)(CHCMe3)(OAr)2 The resulted catalyst showed unprecedented activity in alkene of cis-4-nonene, 1- nonene and ethyl oleate metathesis at 30°C. Later in 2015 the same group used another tungsten oxo complex with a strong σ donating thiophenoxide that improved the efficiency for the metathesis of terminal olefins (Scheme 2.19).16 60

Scheme 2.19: Substituting aryl oxide by strongly -donating thiophenoxide in silica supported tungsten oxo catalyst improve catalyst efficiency in 1-alkene metathesis.

On the other hand our group prepared WMe6 tamed by silica as an efficient, well- defined species for alkane metathesis.17 This novel catalyst opens the desire to modify some of the methyl ligands to enhance the activity or to make longer lifetime catalyst.

Connecting to previous explanation of the importance of the oxo ligand we decided to prepare the grafted WOMe4.

2.1.7. Selecting the route

The alkylation of WOCl4 by various methylating agents, including HgMe2, ZnMe2, MgMe2 and MgMeI, has been thoroughly investigated by both Santini-Scampucci and

Scheme 2.20: Structure of (Me4WO)2Mg(THF)4

18 Riess with no success. But partially alkylated complex (MeWOCl3.OEt2) has been isolated and characterized at low temperature by Muetterties and Band.19 More than a 61 decade later, Wilkinson found that alkylating tungsten oxo chloride in THF produces an orange-red crystals that were identified by single-crystal X-Ray as (Me4WO)2Mg(THF)

- (Scheme 2.20), the obtained product could be due a reduction of WOCl4 into [WCl4O]

2- 20 and [WCl5O] followed by bimolecular interaction and then alkylation.

2.1.8. Surface alkylation strategy In this work, inspired by the reactivity of surface metal-methyl complexes17, 21, added to oxo ligand importance, synthetic new approach having two main steps is developed: grafting inexpensive commercially available metal halides and then selectively alkylate

(methylate) the grafted complexes to form catalytically active metal-CH3 groups. Here we describe a simple method for the synthesis of a novel silica-supported tungsten oxo- trimethyl complex [(≡Si-O-)W(=O)Me3]. Our synthesis protocol follows a sequence of grafting the commercially available tungsten (VI) oxy tetrachloride on silica dehydroxylated at 700°C (SiO2-700) under high vacuum, then alkylation of the grafted complex. A similar strategy was recently published, however, using SiO2-200 to produce

22 bipodal species that is different [(≡Si- O)2W(=O)Me2].

Scheme 2.21: Synthesis of [(≡Si-O-)W(=O)Cl3] (2) by reaction of (1) with partially dehydroxylated silica at o 700 C (SiO2-700) 62

We fully characterized the silica-supported tungsten oxo- trimethyl using solid-state

NMR, FT-IR spectroscopy, microanalysis, mass balance and validated by NMR-DFT calculations.

Scheme 2.22: Synthesis of supported [(≡Si-O-)W(=O)Me3] (3) by surface methylation of (2) with 1.5 equivalents of dimethylzinc

The new tungsten oxo-trimethyl catalyst showed significant activity in various organic transformations, especially for cyclooctane metathesis and direct transformation of ethylene to propylene.

2.2. RESULTS AND DISCUSSION

2.2.1. Preparation and characterization of [(≡Si-O-)W(=O)Me3] on SiO2-700: [(≡Si-O-

)W(=O)Cl3] 2 was used as the precursor for the synthesis of 3. Surface compound 2 was

® prepared by grafting WOCl4 on silica Aerosil 200 partially dehydroxylated overnight

-5 under 10 bar at 700°C (SiO2-700) (Scheme 2.21). The reaction was performed in dichloromethane at room temperature for 4 hours. A partial vacuum was used to prevent any secondary reactions by gaseous hydrogen chloride liberated during the reaction with the support. Excess physisorbed molecular complexes were removed by 63 washing the solid trice with dichloromethane (3 x 20 ml). The resulting light green material was dried under high vacuum (10-5 bar).

The infrared spectrum showed that the characteristic peak of free silanol at 3743 cm-1

(Figure 2.4A) completely disappeared (Figure 2.4B). The elemental analysis of the light green powder (product 2) found 4.84% W and 2.59% Cl with a molar ratio of

W/Cl=1.0/2.7±0.2 (theoretical ratio for monopodal 1.0/3.0).

Mass balance determination was performed on [(≡Si-O-)W(=O)Cl3] by hydrolysis with water; the amount of released hydrogen chloride was quantified using silver nitrate. A ratio of 2.8±0.2 chlorides per grafted tungsten atom was found. Mass balance and gas quantification studies confirmed that there are three chlorides per tungsten. Direct methylation of [(≡Si-O-)W(=O)Cl3] was achieved using various alkylating agents such as

ZnMe2, SnMe4, MeLi and Al2Me6. Partially methylated species were produced with

SnMe4, while MeLi and Al2Me6 caused violent reactions that formed insoluble salts. The best result was obtained after treatment of [(≡Si-O-)W(=O)Cl3] with 1.5 equivalents of

ZnMe2 in diethyl ether at room temperature for 3 hours (Scheme 2.22). This treatment provided us high selectivity for fully methylated surface complex [(≡Si-O-)W(=O)Me3] 3. 64

Figure 2.4: (A) SiO2-700 (B) after grafting of WOCl4 (C) after alkylation with ZnMe2. The IR spectrum (Figure 2.4C) revealed new bands in the range of 3000 cm-1, ν(C-H); and

1454 cm-1, δ(C-H), assigned to methyl fragments. Microanalysis associated with gas quantification results revealed the ratio of carbon to tungsten (C/W) in complex 3 (2.8 ±

0.2) (theoretical value 3). The elemental analysis gave 4.81% W, 1.2% C, 1.01% Zn and

1.56% Cl, with a ratio of C/W=3.5 (theoretical value 3) and a ratio of Cl/Zn=2.1; these results suggest that we have c.a. three methyl groups bonded to tungsten. Some remaining diethyl ether coordinated to zinc chloride (vide infra) accounts for the higher carbon to tungsten ratio.

2.2.2. NMR Study The prepared surface complex 3* was characterized by solid-state NMR. The 1H magic- angle spinning (MAS) NMR spectra of 3 at room temperature (Figure 2.5) exhibits a main 65 signal at 1.4 ppm with a shoulder at 1.9 ppm and a broad signal at 3.6 ppm. The signals at 1.4 and 1.9 ppm auto-correlate in two-dimensional (2D) double-quantum (DQ) (2.8 and 3.8 ppm in F1) (Figure 3A) and triple-quantum (TQ) (4.2 and 5.7 ppm in F1) (Figure

2.6B) spectra and are assigned to methyl groups connected to tungsten. Proton resonance at 3.6 ppm with less than an 11% intensity ratio was assigned to the CH2 group of the coordinated ether and supported by an auto-correlation in the DQ (7.2 ppm in F1) (Figure 2.6A) but not in the TQ spectra (Figure 2.6B). Furthermore, a correlation between the CH3 and CH2 groups of the coordinated ether was observed in both DQ and TQ NMR experiments (Figure 2.6A and 2.6B). Also, a small auto-correlation for a proton signal at -0.7 ppm in both DQ and TQ spectra indicates the presence of trace amounts of remaining dimethylzinc. 66

Figure 2.5: 1D 1H MAS solid-state NMR spectra of 3* acquired at 600 MHz at different temperatures with a 22 kHz MAS frequency and a repetition delay of 5 s and 8 scans.

13 13 To obtain a better C MAS NMR resolution, an enriched C carbon Zn*(Me)2 was prepared and used in synthesizing complex 3* (13C 97% enriched). The 13C CP/MAS NMR spectra (Figure 2.7) at room temperature of 3* displays a main signal at 46 ppm with a shoulder at 51 ppm and small peak around 36 ppm. Additionally, the 2D 1H-13C HETCOR

NMR spectrum (Figure 2.8) with a short contact time (0.2 ms) shows a correlation between the methyl protons 1.4-1.9 ppm and carbon atoms 36, 46 and 51 ppm, which allows their assignment to methyl groups. 67

Figure 2.6: (A) 2D 1H-1H DQ/SQ and (B) 1H-1H TQ/SQ NMR spectra of 3* at room temperature (both acquired with number of scans per increment = 32 per t1 increment, repetition delay = 5 s, number of individual t1 increments = 128 68

Furthermore, a small correlation between carbon atoms 15 and 66 ppm and protons 1.1 and 3.6 ppm, respectively, are assigned to CH3 and CH2 of coordinated diethyl ether.

Figure 2.7: One-dimensional 13C CP/MAS NMR spectra of 3* at different temperatures with a 10 kHz MAS frequency (a repetition delay = 5 s, contact time = 2 ms, line broadening = 80 Hz and number of scans = 10000.) 69

Figure 2.8: 2D CP/MAS HETCOR NMR spectra of 3* acquired with short contact times of 0.2 ms under 8.5 kHz MAS frequency (number of scans per increment = 4000, repetition delay = 4 s, number of t1 increments = 32, line broadening = 80 Hz.)

2.2.3. DFT NMR calculations We modeled tungsten species grafted on silica using the molecular model depicted in

Figure 2.9. We considered two conformations that differ with respect to their relative orientation of the SiO-W and W=O bonds. In the former, the oxo ligand is trans oriented to the siloxy ligand, creating a bipyramidal trigonal geometry around the metal center

(Figure 2.9A). In the second conformation, the oxo and the siloxy ligand are cis-oriented, 70 producing an overall geometry around the metal center that is best described as a distorted square pyramid with the W=O bond at the apex (Figure 2.9B).

Figure 2.9: Ball-and-stick representations of the DFT molecular model of the -W(=O)(Me)3 complex; two optimized conformations (A, B). Hydrogen atoms are omitted for clarity. Color code; W: cyan, C: gray, O: red and Si: orange. Geometric parameters are reported in Å for the bond distances and degrees for angles.

According to calculations, the structure with the trans-oriented W-siloxy and W=O bonds (Figure 2.9A) is only 2.1 kcal/mol more stable than the structure with the same bonds cis-oriented (Figure 2.9B). For this reason, the following DFT NMR analysis was performed on both geometries. The magnetic shielding values for 13C reported in (Table

2.2) show a high level of agreement between the calculated and experimental carbon chemical shifts. The DFT models of the supported species indicate that the three carbons on the trigonal bipyramid model in (Figure 2.9A) possess magnetic shielding 71 values of 142.2, 143.3 and 143.7 ppm for C1, C2 and C3, respectively. Overall the average value of the chemical shift of these three carbons is 44.5 ppm, which is in accordance with the experimental main resonance of 46 ppm. Figure 2.9B illustrates a model with square pyramid geometry where C1 is predicted to be shielded by 7.8 and

8.6 ppm more than C2 and C3, respectively (Table 2.2). Therefore, the experimentally observed chemical shifts at 36 can be assigned to C1 while 51 ppm peak attribute to C2 and C3 (see the model of Figure 2.9B). Thus, the single peak at 48 ppm, 318 K is more likely a result of a rapid isomerization between the trigonal bipyramid and the square pyramid geometries, averaging the chemical shift of carbon atoms in the different geometries

Table 2.2: DFT Calculated 13C Nuclear Magnetic Shielding and Chemical Shifts for the two possible geometries of complex [(≡Si-O-)W(=O)Me3]

Model of Figure 2.9A: trans-oriented W=O bonds

Methyl group σiso (ppm) δiso (ppm) δiso (ppm)Average δiso (ppm) Exper.

C1 142.19 44.67

C2 143.26 43.60 44.51 46

C3 143.70 43.16

Model of Figure 2.9B: cis-oriented W=O bonds

C1 146.76 40.10 40.10 36

C2 138.95 47.91 48.32 51

C3 138.13 48.73

2.2.4. CATALYTIC STUDY Cyclooctane metathesis We predicted that complex 3 would be able to generate a tungsten methyl/methylidene surface molecule at elevated temperature (around 80°C), which we suspected to be responsible for the metathesis reaction; therefore, we applied complex 3 for the metathesis of cyclooctane. We focused on cyclooctane that can be used as a building block for higher macrocycles and are crucial in the preparation of pharmaceutical intermediates23, polymers24, and fragrances.25

Figure 2.10: GC chromatogram of cyclooctane metathesis products that been catalyzed by 3. Reaction condition: batch reactor, compound 3 (40 mg, 8.3 µmol, W loading: 4.83 wt%), cyclooctane (0.5 ml, 3.7 o mmol), 160 h, 150 C. Conversion=60%, TON=260. Products formed are ring-contracted (cC6), as well as ring-expanded species ranging from cC13 to cC29. In a typical reaction, 400 equivalents of cyclooctane were mixed with 40 mg of [(≡Si-O-

)W(=O)Me3] 3 at 150°C in a sealed ampoule for seven days. At the end of the reaction, 73 the ampoule was frozen with liquid nitrogen; the reaction mixture was quenched with dichloromethane, and the liquid fraction was filtrated and analyzed by gas chromatography. Cyclooctane was converted into cycloheptane by ring contraction

(68%) and higher macrocycles by ring expansion ranging between cC13 to cC30 (30%) with a turn over number (TON) of 260 as determined by GC and GC-MS (Figure 2.10). We know from previous results on cyclooctane metathesis that ring contraction is becoming more favorable as contact time increases; knowing that the higher and lower homologs are cyclic with TON exceeds 350.26

Direct conversion of ethylene to Propylene Polypropylene (PP) is becoming more and more important as a chemical for everywhere applications.27 Therefore there is a high demand by industry for propylene manufacture.

More than 80 million tons are produced each year worldwide; this production is going to increase over the next decade to reach 130 million tons annually.28 The primary source of this highly valuable building block in polymer chemistry is the steam or catalytic cracking of natural gas and oil29; however, these processes require large amounts of energy. In this work, we found that [(≡Si-O-)W(=O)Me3] can directly convert ethylene into propylene under milder reaction conditions than previously documented.

In an earlier study, we used tungsten hydride on alumina to demonstrate that a hydride catalyst could perform ethylene dimerization to 1-butene via the metal hydride, isomerization of 1-butene to 2-butene and cross-metathesis between ethylene and 2- butene via a metallocarbene.30

Scheme 2.23: Proposed Mechanism for the Direct Conversion of Ethylene to Propylene by [(≡Si-O-)W(=O)Me3] 75

Our more recent research documented here is likely following the same process, but via a different initiation step. Tungsten–methyl could insert an ethylene to give tungsten-n- propyl, followed by a beta-hydride elimination, which would generate propylene and a tungsten hydride that is necessary to propagate the dimerization of ethylene and the isomerization of the 1-butene. (Scheme 2.23) Cross metathesis could be achieved with an equilibrium between W-CH3 and W(H)(=CH2), which would be sufficient to promote the full catalytic cycle. We introduced ethylene (11.65 mmol) to a 300 ml glass reactor loaded with 100 mg of [(≡Si-O-)W(=O)Me3] to produce selectively propylene after 18 hours reaction at 150°C. We monitored the products using GC-FID and found that propylene represents more than 90% of the products (Table 2.3). This silica-supported catalyst is more active and selective when compared to its analog tungsten hydride or tungsten pentamethyl grafted on SiO2-700 after 18 hours reaction at 150°C.

Table 2.3: Ethylene to Propylene Transformation Using [(≡Si-O-)W(=O)Me3] as Compared to the Analogue Tungsten Hydride and Tungsten Pentamethyl

Product Selectivitya (%)

Catalyst TONb Propylene Butenes Pentenes

[(≡Si-O-)W(=O)Me3] 375 93.2 6.45 0.28

[(≡Si-O-)WMe5] 75 51.6 46.3 2.01

[(≡Si-O-)W(H)x] 77 32.4 64.85 2.73 aThe selectivity is defined as the amount of a product over the total amount of products. bTON is expressed in moles of ethylene transformed per mole of W.

2.3. CONCLUSION

In this chapter we reported the facile synthesis of monopodal [(≡Si-O-)W(=O)Me3] by grafting commercially available WOCl4 onto SiO2-700. The resultant [(≡Si-O-)W(=O)Cl3] was then methylated by ZnMe2 in diethyl ether solution to selectively give [(≡Si-O-

)W(=O)Me3]. 3 was fully characterized by SS-NMR, IR, microanalysis and mass balance techniques. Two different conformations of 3 were proposed depending on the relative orientation of the SiO-W and W=O bonds. Experimental and DFT-calculated NMR supports these models. [(≡Si-O-)W(=O)Me3] 3 was applied as a catalyst for the metathesis of cyclooctane, demonstrating significant activity in producing a wide range of ring contracted and expanded cycloalkanes. On the other hand, the tungsten oxo- trimethyl catalyst was tested for the direct transformation of ethylene to propylene and was found to have a higher activity and selectivity compared to the analogs tungsten hydride/silica and tungsten pentamethyl/silica.

2.4. EXPERIMENTAL SECTION General procedure: All experiments were performed by using standard Schlenk and glovebox techniques under an inert nitrogenous atmosphere. Syntheses and treatments of the surface species were completed using high-vacuum lines (<10−5 mbar) and glovebox techniques. All solvents were dried, degassed and freshly distilled before use according to classical methods (over CaH2 for dichloromethane, over Na/benzophenone for diethyl ether). Ethylene was dried and deoxygenated before use by passage through a mixture of freshly regenerated molecular sieves (3 Å) and R3-15 catalysts (BASF). IR spectra were recorded on a Nicolet 6700 FT-IR spectrometer using a DRIFT cell equipped with CaF2 windows. The IR samples were prepared under argon within a glovebox. 77

Typically, 64 scans were collected for each spectrum (resolution 4 cm−1). Elemental analyses were performed at Mikroanalytisches Labor Pascher (Germany). Gas-phase analysis of alkanes was performed using an Agilent 6850 gas chromatography column with a split injector coupled with an FID (flame ionization detector). An HP-PLOT Al2O3

KCl 30 m × 0.53 mm, 20.00 mm capillary column coated with a stationary phase of aluminum oxide deactivated with KCl was used with helium as the carrier gas at 32.1 kPa. Each analysis was executed under the same conditions: a flow rate of 1.5 ml/min and an isotherm at 80°C.

Solid State Nuclear Magnetic Resonance Spectroscopy. One-dimensional 1H MAS and 13C

CP/MAS solid-state NMR spectra were recorded on Bruker AVANCE 2 spectrometers operating at 400 MHz or 600 MHz resonance frequencies for 1H. 400 MHz experiments employed a conventional double resonance 4mm CP/MAS probe, while experiments at

600 MHz utilized a 3.2 mm HCN triple resonance probe. In all cases, the samples were packed into rotors under inert atmosphere inside gloveboxes. Dry nitrogen gas was utilized for sample spinning to prevent degradation of the samples. NMR chemical shifts are reported with respect to the external references, TMS, and adamantane. 13C

CP/MAS NMR experiments used the following sequence–90 0 pulses on the proton

(pulse length 2.4 s), then a cross-polarization step with a contact time of typically 2 ms, and finally acquisition of the 13C signal under high-power proton decoupling. The delay between the scans was set to 5 s to allow the complete relaxation of the 1H nuclei and the number of scans ranged between 5000 and 10000 for 13C and was 32 for 1H. An 78 exponential apodization function corresponding to a line broadening of 80 Hz was applied prior to the Fourier transformations.

The 2D 1H-13C heteronuclear correlation (HETCOR) solid-state NMR spectroscopy experiments were conducted on a Bruker AVANCE 2 spectrometer operating at 600 MHz using a 3.2 mm MAS probe. These experiments were performed according to the

0 13 following scheme–90 proton pulse, t1 evolution period, CP to C and detection of the

13C magnetization under TPPM decoupling. During the cross-polarization step, a ramped radio frequency (RF) field centered at 75 kHz was applied to the protons, while the 13C channel RF field was matched to obtain an optimal signal. A total of 32 t1 increments with 4000 scans each was collected; sample-spinning frequency was 8.5 kHz. Using a short contact time (0.2 ms) for the CP step, the polarization transfer in the dipolar correlation experiment was verified to be selective for the first coordination sphere around tungsten, such that correlations occurred only between pairs of attached 1H-13C spins (C-H directly bonded).

1H-1H multiple-quantum spectroscopy. Two-dimensional double-quantum (DQ) and triple-quantum (TQ) experiments were recorded on a Bruker AVANCE 2 spectrometer operating at 600 MHz with a conventional double resonance 3.2 mm CP/MAS probe, according to the following general scheme–excitation of DQ coherences, t1 evolution, z- filter, and detection. The spectra were recorded in a rotor-synchronized fashion in t1 such that the t1 increment was set equal to one rotor period (45.45 µs). One cycle of the standard back-to-back (BABA) recoupling sequences was used for the excitation and 79

reconversion period. Quadrature detection in w1 was achieved using the States-TPPI method. A spinning frequency of 22 KHz was used. The 90 ° proton pulse length was 2.5

µs, while a recycle delay of 5 s was used. A total of 128 t1 increments with 32 scans per increment were recorded. The DQ frequency in the w1 dimension corresponds to the sum of two single quanta (SQ) frequencies of the two coupled protons and correlates in the w2 dimension with the two corresponding proton resonances. The TQ frequency in the w1 dimension corresponds to the sum of the three SQ frequencies of the three- coupled protons and correlates in the w2 dimension with the three individual proton resonances. Conversely, groups of less than three equivalent spins will not give rise to diagonal signals in the spectrum.

Preparation of the silica partially dehydroxylated at 700oC. Typically, 4.000 g of Degussa

Aerosil 200 silica was treated in a quartz reactor fitting a tubular furnace under high vacuum (10-5 Torr) pressure at 700oC for 16 h. The temperature program was set to 90oC

/h. IR (cm−1): 3747 (ν(SiO−H)).

Preparation of [(≡Si-O-)W(=O)Cl3]. In a double Schlenk, a solution of WOCl4 (342 mg, 1 mmol, 1.1 equivalent with respect to the amount of surface-accessible silanols) in 25 ml

o of dichloromethane was allowed to react with 3.0 g of SiO2-700 at 25 C for 4 h. This reaction was performed under mild dynamic vacuum conditions to remove any of the

HCl produced during the reaction. At the end of the reaction, the resulting light green solid was washed with dichloromethane (3×20 ml) and dried under a dynamic vacuum

(<10−5 Torr, 2 h). Elemental analysis found 4.81% W; 2.59% Cl. 80

13 Preparation of Zn( CH3)2 in ether. In a 200 ml Schlenk, 15 mL, 1.6 M BuLi was taken under argon, and to that 3.4g (24 mmol) 13MeI diluted in 25 ml of pentane was added drop wise at -20°C with stirring. A white precipitate formed immediately and the solution was stirred for another 30 minutes after the addition of 13MeI. The precipitate was then filtered and dried under a vacuum to produce a white solid (13MeLi). A small amount of 13MeLi was taken and titrated with deoxygenated water to quantify the amount of methyl lithium present by quantifying the release of methane (around 55%

13MeLi was found). In a double Schlenk, 13MeLi (22 mmol, 484 mg) was added to 10 ml of diethyl ether, and to that a solution of anhydrous ZnCl2 (10 mmol in10 ml) was added at -20°C. The reaction was allowed to reach room temperature slowly, and remained at that temperature for another hour. The final product was filtered and the dimethylzinc was collected from the filtrate.

Preparation of [(≡Si-O-)W(=O)Me3]. In a double Schlenk [(≡Si-O-)W(=O)Cl3] (1.0 g, 0.3 mmol) was added to 20 mL of ether and to that (0.45 mmol) ZnMe2 solution in diethyl ether (1.5 equivalents with respect to the amount of tungsten oxo chloride on the surface) was added to react at 25°C for 4 h. At the end of the reaction, the resulting gray solid was filtered and thoroughly washed with diethyl ether. The gray solid was then dried under dynamic vacuum (<10−5 Torr) conditions for two hours. 1H MAS SS-NMR

13 (400 MHz, 298 K): δ (ppm) 1.4 (CH3), 1.9 (CH3), 3.6 (CH2). C CP-MAS SS-NMR (400 MHz,

1 298K): δ (ppm) 36 (CH3), 46 (CH3), 51 (CH3). H MAS SS-NMR (400 MHz, 318K K): δ (ppm)

13 1.7 (CH3), 3.6 (CH2). C CP-MAS SS-NMR (400 MHz, 318K): δ (ppm) 48 (CH3). The 81 hydrolysis of the solid with a vapor pressure of degassed water generated 0.74±0.05

−1 mmol CH4•g (2.8 ± 0.2 CH4/W). Anal. 4.81% W, 1.2% C, 1.01% Zn, 1.56% Cl.

Cyclooctane metathesis using [(≡Si-O-)W(=O)Me3]. The reactions were performed using an ampoule filled with the catalyst (40 mg, 0.0083 mmol, W loading: 4.8 wt%) inside a glovebox where cyclooctane (0.5 ml, 3.7 mmol) was added. Substrate/Catalyst=450. The ampoule was sealed under vacuum pressure, immersed in an oil bath and heated at

150°C for 7 days. At the end of the reaction, the ampoule was cooled using liquid nitrogen. The mixture was quenched by adding a fixed amount of dichloromethane after filtration; the resulting solution was analyzed by GC and GC-MS. Reaction conversion=60%

Ethylene to propylene using [(≡Si-O-)W(=O)Me3]. A 100 mg of complex 3 was placed in a

300 ml batch reactor and after evacuation to 10-5 mbar, 0.95 atm of dry ethylene was introduced (Substrate/catalyst ratio around 500). The reactor was heated at 150°C for

18 h. The final products were analyzed by GC.

DFT computational Details. All quantum chemical computations were performed using the Amsterdam Density Functional (ADF)31 and Gaussian 09 programs. NMR calculations were made for each of the two different conformations depicted in Figure 6. The two different models represent the position of the oxygen atoms around Tungsten; in the first conformation (A) the oxygen atom is perpendicular to the plane of the three methyl groups. In the second conformation (B) the position of the oxygen atom exchanged with one methyl group. 82

First, geometry optimizations were performed using Gaussian 09 with the generalized gradient approximation (GGA) exchange-correlation function developed by Perdew,

Burke, and Ernzerhof (PBE)32 an all electron Def2_TZVP 31,32 basis set computational level. Geometry optimizations were carried out without any constraints. Optimized geometries were used as input for the ADF NMR calculations. For the magnetic shielding tensor calculations (1H and 13C), the zeroth-order regular approximation (ZORA)33 was used to include relativistic effects both scalar and spin−orbit contributions.34 The NMR calculations used an all-electron triple-ζ basis set which included two polarization functions (i.e., TZ2P) and a PBE functional.

2.5. References 1. (a) Lozano-Vila, A. M.; Monsaert, S.; Bajek, A.; Verpoort, F., Ruthenium-Based Olefin Metathesis Catalysts Derived from Alkynes. Chem. Rev. 2010, 110 (8), 4865-4909; (b) Vougioukalakis, G. C.; Grubbs, R. H., Ruthenium-Based Heterocyclic Carbene- Coordinated Olefin Metathesis Catalysts. Chem. Rev. 2010, 110 (3), 1746-1787; (c) Samojlowicz, C.; Bieniek, M.; Grela, K., Ruthenium-Based Olefin Metathesis Catalysts Bearing N-Heterocyclic Carbene Ligands. Chem. Rev. 2009, 109 (8), 3708-3742.

2. Mol, J. C., Industrial applications of olefin metathesis. J. Mol.Cata. A: Chem. 2004, 213 (1), 39-45.

3. Goldman, A. S.; Roy, A. H.; Huang, Z.; Ahuja, R.; Schinski, W.; Brookhart, M., Catalytic alkane metathesis by tandem alkane dehydrogenation olefin metathesis. Science 2006, 312 (5771), 257-261.

4. (a) Lee, H. K.; Bang, K. T.; Hess, A.; Grubbs, R. H.; Choi, T. L., Multiple Olefin Metathesis Polymerization That Combines All Three Olefin Metathesis Transformations: Ring-Opening, Ring-Closing, and Cross Metathesis. J. Am. Chem. Soc. 2015, 137 (29), 9262-9265; (b) Risse, W.; Grubbs, R. H., Block and Graft-Copolymers by Living Ring- Opening Olefin Metathesis Polymerization. J. Mol. Catal. 1991, 65 (1-2), 211-217; (c) Toreki, R.; Farooq, A.; Schrock, R. R., Olefin Metathesis and Ring-Opening Metathesis Polymerization by Well-Defined Rhenium Catalysts. Abstr. Pap. Am. Chem. S. 1990, 200, 345-Inor; (d) Risse, W.; Grubbs, R. H., Block and Graft-Copolymers by Living Ring- Opening Olefin Metathesis Polymerization. Abstr. Pap. Am. Chem. S. 1989, 197, 99-Poly; (e) Farona, M. F.; Tsonis, C., Polymerization of Cyclic Olefins by Chloropentacarbonylrhenium, an Olefin Metathesis Catalyst. Abstr. Pap. Am. Chem. S. 1977, 173, 62-62.

5. Schreiber, S. L., Target-oriented and diversity-oriented organic synthesis in drug discovery. Science 2000, 287 (5460), 1964-1969.

6. Banks, R. L.; Bailey, G. C., Olefin Disproportionation - New Catalytic Process. Ind. Eng. Chem. Prod. Rd. 1964, 3 (3), 170.

7. Heckelsb, L. F; Banks, R. L.; Bailey, G. C., A Tungsten Oxide on Silica Catalyst for Phillips Triolefin Process. Ind. Eng. Chem. Prod. Rd. 1968, 7 (1), 29.

8. Herisson, J. L.; Chauvin, Y., Transformation Catalysis of Olefins by Tungsten Complexes .2. Telomerization of Cyclic Olefins in Presence of Acyclic Olefins. Makromolekul. Chem. 1971, 141, 161.

9. Grunert, W.; Feldhaus, R.; Anders, K.; Shpiro, E. S.; Minachev, K. M., Reduction Behavior and Metathesis Activity of WO3/Al2O3 Catalysts .3. The Activation of WO3/Al2O3 Catalysts. J. Catal. 1989, 120 (2), 444-456. 84

10. Thomas, R.; Moulijn, J. A.; Kerkhof, F. P. J. M., Raman-Spectra of Metathesis Catalysts Tungsten-Oxide on Silica and Tungsten-Oxide on Gamma-Alumina. Recl. Trav. Chim. Pay. B 1977, 96 (11), M134-M135.

11. Kress, J. R. M.; Russell, M. J. M.; Wesolek, M. G.; Osborn, J. A., Tungsten(VI) and Molybdenum(VI) Oxo-Alkyl Species - Their Role in the Metathesis of Olefins. J. Chem. Soc. Chem. Comm. 1980, (10), 431-432.

12. Kress, J.; Wesolek, M.; Leny, J. P.; Osborn, J. A., Molecular-Complexes for Efficient Metathesis of Olefins - the Oxo-Ligand as a Catalyst-Cocatalyst Bridge and the Nature of the Active Species. J. Chem. Soc. Chem. Comm. 1981, (20), 1039-1040.

13. Lopez, L. P. H.; Schrock, R. R., Formation of dimers that contain unbridged W(IV)/W(IV) double bonds. J. Am. Chem. Soc. 2004, 126 (31), 9526-9527.

14. Mazoyer, E.; Merle, N.; de Mallmann, A.; Basset, J. M.; Berrier, E.; Delevoye, L.; Paul, J. F.; Nicholas, C. P.; Gauvin, R. M.; Taoufik, M., Development of the first well- defined tungsten oxo alkyl derivatives supported on silica by SOMC: towards a model of WO3/SiO2 olefin metathesis catalyst. Chem. Commun. 2010, 46 (47), 8944-8946.

15. Conley, M. P.; Mougel, V.; Peryshkov, D. V.; Forrest, W. P.; Gajan, D.; Lesage, A.; Emsley, L.; Coperet, C.; Schrock, R. R., A Well-Defined Silica-Supported Tungsten Oxo Alkylidene is a Highly Active Alkene Metathesis Catalyst. J. Am. Chem. Soc. 2013, 135 (51), 19068-19070.

16. Mougel, V.; Pucino, M.; Coperet, C., Strongly sigma Donating Thiophenoxide in Silica-Supported Tungsten Oxo Catalysts for Improved 1-Alkene Metathesis Efficiency. Organometallics 2015, 34 (3), 551-554.

17. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Rossini, A. J.; Widdifield, C. M.; Dey, R.; Emsley, L.; Basset, J. M., WMe6 Tamed by Silica: Si-O-WMe5 as an Efficient, Well- Defined Species for Alkane Metathesis, Leading to the Observation of a Supported W- Methyl/Methylidyne Species. J. Am. Chem. Soc. 2014, 136 (3), 1054-1061.

18. Stavropoulos, P.; Wilkinson, G.; Motevalli, M.; Hursthouse, M. B., Oxo methyls of molybdenum(V), tungsten(V) and rhenium(V): X-ray crystal structure of (Me4WO)2Mg(Thf)4. Polyhedron 1987, 6 (5), 1081-1087.

19. Muetterties, E. L.; Band, E., Olefin Metathesis Reaction - Characterization of an Active Catalyst Precursor, CH3WOCl3.O(C2H5)2, from the WOCl4(CH3)2Mg Reaction. J. Am. Chem. Soc. 1980, 102 (21), 6572-6574.

20. Stavropoulos, P.; Wilkinson, G.; Motevalli, M.; Hursthouse, M. B., Oxo Methyls of Molybdenum(V), Tungsten(V) and Rhenium(V) - X-Ray Crystal-Structure of (Me4WO)2Mg(Thf)4. Polyhedron 1987, 6 (5), 1081-1087. 85

21. (a) Chen, Y.; Abou-hamad, E.; Hamieh, A.; Hamzaoui, B.; Emsley, L.; Basset, J. M., Alkane Metathesis with the Tantalum Methylidene [(≡SiO)Ta(=CH2)Me2]/[(≡SiO)2Ta(=CH2)Me] Generated from Well-Defined Surface Organometallic Complex [(≡SiO)(TaMe4)]. J. Am. Chem. Soc. 2015, 137 (2), 588-591; (b) Chen, Y.; Ould-Chikh, S.; Abou-Hamad, E.; Callens, E.; Mohandas, J. C.; Khalid, S.; Basset, III J. M., Facile and Efficient Synthesis of the Surface Tantalum Hydride (≡SiO)2Ta H and III Tris-Siloxy Tantalum (≡SiO)3Ta Starting from Novel Tantalum Surface Species [(≡SiO)TaMe4] and [(≡SiO)2TaMe3]. Organometallics 2014, 33 (5), 1205-1211; (c) Riache, N.; Callens, E.; Samantaray, M. K.; Kharbatia, N. M.; Atiqullah, M.; Basset, J. M., Cyclooctane Metathesis Catalyzed by Silica-Supported Tungsten Pentamethyl [(≡SiO)W(Me)5]: Distribution of Macrocyclic Alkanes. Chem.-Eur. J. 2014, 20 (46), 15089- 15094; (d) Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Dey, R.; Basset, J. M., Well- defined silica supported W-alkyl/alkylidyne precursor for metathesis of propane. Abstr. Pap. Am. Chem. S. 2014, 247.

22. Bouhoute, Y.; Garron, A.; Grekov, D.; Merle, N.; Szeto, K. C.; De Mallmann, A.; Del Rosal, I.; Maron, L.; Girard, G.; Gauvin, R. M.; Delevoye, L.; Taoufik, M., Well-Defined Supported Mononuclear Tungsten Oxo Species as Olefin Metathesis Pre-Catalysts. ACS Catal. 2014, 4 (11), 4232-4241.

23. Driggers, E. M.; Hale, S. P.; Lee, J.; Terrett, N. K., The exploration of macrocycles for drug discovery - an underexploited structural class. Nat. Rev. Drug Discov. 2008, 7 (7), 608-624.

24. Voter, A. F.; Tillman, E. S., An Easy and Efficient Route to Macrocyclic Polymers Via Intramolecular Radical-Radical Coupling of Chain Ends. Macromolecules 2010, 43 (24), 10304-10310.

25. Zou, Y.; Mouhib, H.; Stahl, W.; Goeke, A.; Wang, Q. R.; Kraft, P., Efficient Macrocyclization by a Novel Oxy-Oxonia-Cope Reaction: Synthesis and Olfactory Properties of New Macrocyclic Musks. Chem. Eur. J. 2012, 18 (23), 7010-7015.

26. Riache, N.; Callens, E.; Samantaray, M. K.; Kharbatia, N. M.; Atiqullah, M.; Basset, J. M., Cyclooctane Metathesis Catalyzed by Silica-Supported Tungsten Pentamethyl [(≡SiO)W(Me)5]: Distribution of Macrocyclic Alkanes. Chem. Eur. J. 2015, 21 (24), 8656- 8656.

27. Siracusa, V.; Rocculi, P.; Romani, S.; Dalla Rosa, M., Biodegradable polymers for food packaging: a review. Trends Food Sci. Tech. 2008, 19 (12), 634-643.

28. Tullo, A. H., Propylene on demand - A tight market for propylene globally is breathing new life into propane dehydrogenation. Chem. Eng. News 2003, 81 (50), 15- 16. 86

29. Zhu, X. L.; Jiang, S.; Li, C. Y.; Chen, X. B.; Yang, C. H., Residue Catalytic Cracking Process for Maximum Ethylene and Propylene Production. Ind. Eng. Chem. Res. 2013, 52 (40), 14366-14375.

30. Taoufik, M.; Le Roux, E.; Thivolle-Cazat, J.; Basset, J. M., Direct transformation of ethylene into propylene catalyzed by a tungsten hydride supported on alumina: Trifunctional single-site catalysis. Angew. Chem. Int. Ed. 2007, 46 (38), 7202-7205.

31. Te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Guerra, C. F.; Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T., Chemistry with ADF. J. Comput. Chem. 2001, 22 (9), 931- 967.

32. (a) Perdew, J. P.; Burke, K.; Ernzerhof, M., Generalized gradient approximation made simple Phys. Rev. Lett. 1997, 78 (7), 1396-1396; (b) Perdew, J. P.; Burke, K.; Ernzerhof, M., Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77 (18), 3865-3868.

33. (a) Van Lenthe, E.; Ehlers, A.; Baerends, E. J., Geometry optimizations in the zero order regular approximation for relativistic effects. J. Chem. Phys. 1999, 110 (18), 8943- 8953; (b) Vanlenthe, E.; Baerends, E. J.; Snijders, J. G., Relativistic Regular 2-Component Hamiltonians. J. Chem. Phys. 1993, 99 (6), 4597-4610.

34. VanLenthe, E.; Snijders, J. G.; Baerends, E. J., The zero-order regular approximation for relativistic effects: The effect of spin-orbit coupling in closed shell molecules. J. Chem. Phys. 1996, 105 (15), 6505-6516.

CHAPTER 3: Preparation of Silica Supported Tungsten Oxo Imidazolin-2- iminato Methyl Precatalyst for Olefin Metathesis Reactions

3.1 Introduction In the previous chapter, we focused on the synthesis of tungsten oxo trimethyl [(≡Si-O-

)W(=O)Me3] through surface alkylation of [(≡Si-O-)W(=O)Cl3], this catalyst was active in olefin metathesis, alkane metathesis and direct transformation of ethylene into propylene. One of the drawbacks of this catalyst is the deactivation process, in which tungsten could be reduced through α–H elimination or by the formation of polyethylene chain that covers the tungsten active sites.1 To prepare a more robust catalyst we need to tune the ligands of the catalyst with keeping the W-CH3 functionality, substitution of one of the methyl groups into a bulkier ligand could be a solution. In this chapter, mainly we will be focusing on the effect of the electronic environment around (W=O) center on selectivity in olefin metathesis reaction. To achieve the selectivity we tune the ligand system and stabilize it on the silica surface.

3.1.1 Choosing the Ligand Bulky ligands record a high enhancement in stability as well as selectivity. Recently, for instance, Schrock et al. have reported a room-temperature active and Z-selective homogeneous tungsten oxo metathesis catalyst that is analogous to the metal imido compounds coordinated to monoaryl oxide Monopyrrolide (MAP) complexes.2 Similarly,

Schowner et al. reported a highly active and versatile NHC-based (NHC: N-heterocyclic carbene) cationic tungsten oxo complex for the metathesis of various olefins while displaying poor E-Z selectivity.3 Likewise, imidazolin-2-iminato ligand,4 a highly basic nitrogen donor ligand showed the ability to release π-electron toward early transition metals.5 It has been used to modulate the electronic properties of the metal centers in ethylene polymerization,6 hydroamination,7 and C-H activation.5 In this regard, strong 89 electron donor ligands (phosphines, thiophenoxide, NHCs) have often shown the ability to tune the electronic properties of the W center leading to highly active homogeneous and heterogeneous olefin metathesis catalysts.8 Moreover, the functionalization of the imidazoline core with the bulky 2,6-diisopropylphenyl group should satisfy the steric requirements for robust as well as a selective catalyst.

3.1.2 Preparation Path Two leading synthetic strategies have emerged for the preparation of silica supported tungsten oxo complexes. The first relies on grafting organometallic tungsten oxo alkylidene or oxo alkyl precursor; the latter pre-catalyst can generate an alkylidene moiety upon thermal treatment.9 The second strategy employs easy-to-handle tungsten oxo halides that can be grafted on the support surface and subsequently alkylated by suitable reagents. The latter pathway is advantageous as the precursors are much more stable than the analogous tungsten oxo alkyl or alkylidene complexes and their preparation does not require elaborate synthetic procedures. Moreover, as we will show in this chapter, these precursors can be tailored before the grafting to afford a robust catalyst. At this purpose, we have synthesized molecular, and silica supported well-defined tungsten oxo imidazolin-2-iminato methyl complex [(≡Si-O)W(=O)(CH3)2-

(L)] with L= ImDippN (1,3-bis(2,6-diisopropylphenylimidazolin-2-iminato ligand, b) by

Dipp grafting of tungsten oxo imidazolin-2-iminato chloride complex [Im NWOCl3] on the silica surface. This was followed by the selective methylation of the grafted compounds to afford the crucial (W–CH3) fragment that will generate the active carbene (W=CH2)

(Scheme 3.24). 90

Scheme 3.24:(a) Proposed active site of WO3/SiO2 metathesis catalyst; (b) The tungsten oxo imidazolin-2- iminato precatalyst prepared in this chapter.

3.2 RESULTS AND DISCUSSION 3.2.1. Preparation and characterization of complex 2. Silylated iminophosphoranes have been widely used for complexation reactions with various metal halides to yield phosphoraneiminato complexes together with the elimination of trialkylsilyl halides.10

Accordingly, tungsten(VI) oxytetrachloride solution in dichloromethane was treated with one equiv. of N-silylated 2-iminoimidazoline imine 1 at reflux temperature results in the rapid precipitation of complex 2 as an orange crystalline solid in nearly quantitative yield

(84 %, Scheme 3.25). Desilylation and coordination to the metal center resulted in a pronounced down-field shift of the 1H NMR resonances compared to compound 1, and to a minor extent, it was same for the protons of the aromatic rings as expected due to high electron release towards the metal center (Figure 3.18 & Figure 3.19). The 1H and

13C NMR spectra of 2 are in good agreement with those of previously reported complex bearing the imidazoline-2-iminato ligand such as the (imido)vanadium complexes reported by Nomura et al.6c 91

Scheme 3.25: Synthesis of complex 2 [ImDippNWOCl3] . Moreover, the structure of 2 is supported by HSQC 1H, 13C correlation NMR spectra

(Figure 3.20), and elemental analysis (Calc. for 2: 45.75% C; 5.12% H;, 5.93% N; 2.26%

O,15% Cl; 25.94% W. Raman spectroscopy analysis of 2 revealed characteristic absorption bands at 946 cm-1 and at 360 cm-1 (Figure 3.11) attributable to W=O11 and to the W-Cl12 bond vibration respectively. The signals at about 1041 and 1061 cm-1 results from the phenyl rings of the imidazolin-2-iminato ligand.13

Figure 3.11:Raman spectroscopy analysis of 2 92

Figure 3.12: Single crystal X-ray structure of 2 ( llipsoids at 50% probability level). Selected bond distances (Å) and angels (º): W(1)-Cl(1) = 2.361(2), W(1)-Cl(2) = 2.364(2), W(1)-Cl(3) = 2.357(2), W(1)-O(1) = 1.729(5), W(1)-N(1) = 1.782(5); O(1)-W(1)-N(1) = 108.7(3), O(1)-W(1)-Cl(3) = 95.2(2), N(1)-W(1)-Cl(3) = 95.5(2), O(1)- W(1)-Cl(1) = 94.0(2), N(1)-W(1)-Cl(1) = 92.9(2), Cl(3)-W(1)-Cl(1) = 165.08(7), O(1)-W(1)-Cl(2) = 123.0(2), N(1)-W(1)-Cl(2) = 128.3(2), Cl(3)-W(1)-Cl(2) = 82.11(7), Cl(1)-W(1)-Cl(2) = 83.02(7).

At this level, a single crystal of 2 suitable for X-ray crystal structure determination was obtained from a hexane-dichloromethane solvent mix at -30 °C. The crystal X-ray structure of the molecule is shown in Figure 3.12 with selected bond distances and angles; the crystal structure confirms the formation of a monomeric tungsten oxo imidazolin-2-iminato chloride compound with a slightly distorted trigonal bipyramidal geometry around W(1) with oxygen and N-ligand in the equatorial position. The W-N bond distance is found to be 1.782(5) Å; the W-Cl bond length determined to be ca. 2.36

Å and the W=O distance is 1.729(5) Å. The imidazole ring is in the equatorial plane O(1)-

N(1)-Cl(2).

3.2.2. Preparation and characterization of complex 3. After the full characterization of complex 2, the next step was grafting it on silica support to generate surface compound 93

3. Silica partially dehydroxylated at 700°C (SiO2-700) was chosen to generate a well- defined and single-site species (Scheme 3.26). The reaction was performed in dichloromethane under a mild dynamic vacuum in order to remove gaseous HCl and prevent its reaction with the support. The product was then washed with dichloromethane (3 × 20 mL). The resulting light orange material was dried under high vacuum (10-5 mbar).

Scheme 3.26: Grafting 2 on SiO2-700. FTIR and mass balance

The IR spectrum of the grafted complex 3 revealed nearly complete consumption of the surface silanols, as indicated by the disappearance of the silanol peak at 3747 cm-1

(Figure 3.13). IR bands associated with ν (C-H), ν (C-C) and δ (C-H) of the ligand appear in the 2970-2877 region and the 1550-1350 cm-1 region respectively. A characteristic ν

(C=N) appears at 1653 cm-1. 94

Figure 3.13: IR spectroscopy of complex 3, 4, and SiO2-700 Elemental analysis of 3 confirmed by elemental analysis of C, Cl, H, and N indicating the presence of 26.35% C, 2.70% Cl, 3.27% N and 43.24% H per W center (theoretical 27% C,

3% Cl, 3% N and 36% H expected for 3). The elemental analysis of 3 shows the presence of 2.17% W, 1.13% Cl with a ratio of W/Cl equals 1.0/2.70. Mass balance (Cl/W) determination for 3 was performed by suspending it (0.4 g) in a NaHCO3 solution (0.005

M, 100 mL). The resulting chloride ion was titrated with a silver nitrate solution (0.01 M,

90 mL) using potassium chromate as an indicator. A ratio of 2.0 ± 0.2 chlorine atoms per grafted tungsten was found thus confirming the proposed composition. 95

NMR

The 1H magic-angle spinning (MAS) solid-state NMR spectrum of 3 produced broad signals at 1.1, 2.9, 3.6, 5.1, 6.3, and 7.4 ppm analogously to its homogeneous counterpart. The 13C cross polarization magic angle spinning (CP/MAS) NMR spectrum of

3 shows distinct signals at δ = 22, 29, 116, 124, 129, 137 and 147 ppm from the imidadzoline-2-iminato ligand and aromatic groups that are in a good agreement with the NMR spectrum of 2 in solution (Figure 3.19).

3.2.3. Preparation and characterization of complex 4. In order to generate the pre- catalytic methyl derivative of 3, readily available ZnMe2 was used as an alkylating agent.

A color change from yellow to light gray was obtained after treating 3 with 1.5 equiv of

ZnMe2 in dichloromethane at an initial temperature of -50°C for one hour and kept at r.t. for another two hours (Scheme 3.27).

Dipp Scheme 3.27: Synthesis of 4 [(≡Si-O-)W(=O)(CH3)2-Im N] by surface methylation of 3 with 1.5 equiv of ZnMe2.

FTIR spectroscopy and mass balance

The FT-IR analysis of 4 displayed new bands in the 3000 cm-1 (ν (C-H)) and 1300-1500 cm-1 (δ (C-H), (ν (C-H)) regions overlapping those present in the parent compound 3. A 96 significant shift for ν (C=N) was observed at 1612 cm-1 (vs. 1653 cm-1 in 3), which clearly demonstrate a change in the electronic environment around the tungsten center (Figure

3.13). The elemental analysis of 4 shows 2.09% W, 4.07% C, 2.56% Zn, and 1.33% Cl. Gas quantification results revealed the ratio of carbon to tungsten (C/W) in complex 4 equals to 29.85 ± 0.2, which is in close agreement with the expected value (C/W : 29).

This result suggests that both chlorides associated with tungsten got completely exchanged with -CH3.

NMR

Further spectroscopic analysis of 4 was carried out by solid-state NMR. The 1H (MAS)

NMR spectrum of 4 displays several large signals observed in parent compound 3 at 1.1,

2.9, 3.6, 5.1, 6.3, and 7.4 ppm and a new, peak at 1.9 ppm. The signals at 1.1, 2.9 and 3.6 ppm auto-correlate in 2D double-quantum (DQ) and triple-quantum (TQ) 1H–1H homonuclear dipolar correlation spectra, and are assigned to the different methyl groups of the ligand (Figure 3.14b&c). The new peak at 1.9 ppm also auto correlates in the DQ and TQ NMR experiments in agreement with its attribution to the -CH3 group formed after methylation (Figure 3.14b&c). The proton resonances at 5.1, 6.3 and 7.6 ppm display no auto-correlation in TQ but auto-correlates in DQ spectra (Figure

3.14b&c) which confirms that these protons belong to the aromatic from imidazole rings. The broad signal at -0.1 ppm is assigned to ZnMe2 impurities, as supported by autocorrelation in DQ and TQ (Figure 3.14b&c) and as reported in the previous chapter.

The 13C CP/MAS NMR spectrum (Figure 3.14d) displays eight signals at 22, 29, 116, 124, 97

136, 147, and 153 ppm similar to 3 and a new peak at 50 ppm that can be attributed to

W-CH3 after methylation

Figure 3.14: (a) One-dimensional (1D) 1H MAS solid-state NMR spectrum of 4 acquired at 600 MHz (14.1 T) with a 22 kHz MAS frequency, a repetition delay of 5 s, and 8 scans. (b) Two-dimensional (2D) 1H−1H double-quantum (DQ)/single-quantum (SQ) and (c) 1H−1H triple- quantum (TQ)/SQ NMR spectra of 4 13 (both acquired with 32 scans per t1 increment, 5 s repetition delay, 32 individual t1 increments). (d) C 1 CP/MAS NMR spectrum of 4 acquired at 9.4 T (ʋo ( H) = 400 MHz) with a 10 kHz MAS frequency, 10000 scans, a 5 s repetition delay, and a 2 ms contact time. An exponential line broadening of 80 Hz was applied prior to Fourier transformation. (e) 2D 1H−13C CP/MAS dipolar HETCOR spectrum of 4 (acquired at 9.4 T with 10 kHz MAS frequency, 4000 scans per t1 increment, a 4 s repetition delay, 64 individual t1 increments, and a 0.2 ms contact time). Additionally, the 2D 1H–13C HETCOR NMR spectrum (Figure 3.14e) with a short contact time (0.2 ms) shows a correlation between the methyl protons of the ligand (1.1, 2.9, and 3.6 ppm) with their carbon atoms at (22 and 29 ppm) respectively, similarly the aromatic proton and imidazole proton also correlates in HETCOR spectra. Furthermore, 98 the strong correlation between the carbon and proton signals at 50 ppm and 1.9 ppm, respectively, supports the assignment of a methyl moiety of imidazolin-2-iminato

Dipp complex [(≡Si-O-)W(=O)(CH3)2-Im N] as the main surface species (Scheme 3.27).

3.2.4. EXAFS Study The structures of the grafted tungsten complexes 3 and 4 were further investigated by extended X-ray absorption fine structure (EXAFS) spectroscopy at W LIII-edge (Figure

3.15 & Table 3.4). The previous crystallographic structure of 2 as determined by single crystal X-ray diffraction was used to generate the scattering paths with their related theoretical atomic scattering factors, photoelectron mean free paths and phase shifts.

The initial model for a first shell fitting included W-O, W-N, W-O, and W-Cl scattering paths with respective half path lengths of 1.73 Å, 1.79 Å, 1.9 Å and 2.35 Å. Attempts to fit the EXAFS data of compound 3 with the latter model failed because the W-N and W-

O paths were found to be merged into one single frequency. A more reasonable model was thus restricted to a single W-N scattering path modeling both the bond to the imidazoline-2-iminato ligand and the bond to a siloxide ligand. The total coordination number of the metal center was re-strained to 5 in order to decrease the correlation between parameters. The best fit resulted in 1.1 ± 0.1 oxygen atom at 1.72 ± 0.01 Å, assigned to the shortest W=O bond; 2.4 ± 0.2 nitrogen/oxygen atoms at 1.95 ± 0.02 Å, and 1.6 ± 0.1 chlorine atoms at 2.36 ± 0.01 Å. The coordination calculated for tungsten suggests, on average, a predominantly monopodal grafting of 2 and is thus consistent

Dipp with the formula [(≡Si-O-)W(=O)Cl2-Im N] describing the structure of 3. 99

A qualitative examination of the Fourier transform of 4, obtained after methylation of 3, depicted a large decrease of the peak at 2.0 Å (Figure 3.15). The latter observation is indeed expected when chlorine atoms are exchanged by carbon atoms which have a much weaker atomic scattering factor. A first model consisting of W-O (W=O), W-N/O

(ImDippN, siloxide), and W-C (methyl) scattering paths was used to fit the EXAFS data. The resultant fit, shown in (Table 3.11, Figure 3.22 & Figure 3.23), provided an unrealistic total coordination number (≈7) and energy shift (>10 eV), together with a lesser statistical agreement with experimental data.

Dipp Dipp Table 3.4: EXAFS parameters obtained for [(≡Si-O-)W(=O)Cl2-Im N] and[(≡Si-O)W(=O)(CH3)2-Im N].

Paths N d 103 x 2 ∆E R (%) (Å) (Å2) (eV) Dipp [(≡Si-O-)W(=O)Cl2-Im N] W→O 1.2 ± 0.2 1.72 ± 0.01 1 5 ± 2 0.2 W→N/O 2.4 ± 0.1 1.95 ± 0.02 5 ± 3 W→Cl 1.6 ± 0.1 2.36 ± 0.01 3.5 Dipp [(≡Si-O-)W(=O)(CH3)1.4Cl0.6-Im N] W→O 1.0 ± 0.2 1.71 ± 0.01 1.5 6 ± 4 0.5 W→C/O/N 3.4 ± 0.2 1.95 ± 0.05 11 ± 3 W→Cl 0.6 ± 0.2 2.28 ± 0.03 3.5 100

3 Figure 3.15: Tungsten W LIII-edge k -weighted EXAFS (up) and corresponding Fourier transform (down, Dipp modulus and imaginary part) with a comparison to simulated curves for [(≡Si-O-)W(=O)Cl2-Im N] and Dipp [(≡Si-O-)W(=O)(CH3)2-Im N]. Solid lines: experimental; dashed lines: spherical wave theory.

This gave a hint that chlorine atoms might still be coordinated to the tungsten center.

Indeed, the best model used to fit the data included W-O (W=O), W-C (ImDippN, siloxide, methyl) and W-Cl scattering paths. The parameters extracted from the fit are consistent 101 with 1.0 ± 0.2 oxo ligand at 1.71 ± 0.01 Å, 3.4 ± 0.2 carbon atoms at 1.95 ± 0.05 modelling siloxide, imidazoline-2-iminato and methyl ligands, and 0.6 ± 0.2 chlorine atoms at 2.28 ± 0.03 Å. Comparison of the chlorine coordination number of 3 and 4 (1.6 vs 0.6 respectively) shows that the Cl/Me exchange is occurring, but it is not complete, upon reaction with 1.5 equiv of ZnMe2. To conclude EXAFS results suggests that complex

Dipp 4 is better described on average by the formula [(≡Si-O-)W(=O)(CH3)1.4Cl0.6-Im N].

3.3. Catalytic studies in olefin metathesis: 3.3.1. Propene metathesis. Material 4 was found to be an active metathesis pre-catalyst in the presence of an excess (3050 equivalents, catalyst loading: 0.03 mol %) of propene in a batch reactor at 150°C (Figure 3.17). Under these conditions, the formation of an equilibrated mixture of metathesis products (ethene, 2-butenes, conversion 38.2%,

TON: 1165, See Figure 3.16) was observed after about two hours with a trans/cis-butene ratio of ca. 1.8. By allowing the reaction to proceed for a prolonged time, limited quantities of 1-butene could also be detected. When 4 was exposed to 150 equivalent of propene at 150°C in a batch reactor, the formation of an equilibrated mixture of metathesis products containing exclusively propene, ethene, and 2-butene (trans/cis ratio of 1.93) was observed within one hour (conversion 38%, Figure 3.17). In this case,

GC analysis showed the formation of 1.5 equivalents of methane per tungsten. This is in a rough agreement with the potential formation of the expected catalytically active

Dipp carbenic species [(≡Si-O-)W(=O)(=CH2)-Im N] by α-H-transfer from a methyl ligand to generate the catalytically relevant W-methylidene intermediate.14 However, attempts to observe an alkylidene species by solid state NMR after heating material 4 to 150°C prior 102 to substrate addition were not successful; no resonances attributable to carbenic species could be detected.

Figure 3.16: Product conversion and selectivity vs. time plot for the propene metathesis by [(≡Si-O- Dipp )W(=O)(CH3)2-Im N] (3050 equivalents/W) (TON=1165). 103

Dipp Figure 3.17: [(≡Si-O-)W(=O)(CH3)2-Im N] reaction with different amount of propene at 150°C, TON is calculated at different times.

3.3.2. 1-hexene metathesis. The catalytic activity of 4 towards the self-metathesis of higher terminal olefins was also investigated. In particular, the homo-coupling of 1-hexene by precatalyst 4 took place slowly at room temperature (2% conversion in 24 h, TON: 70). However, the rate of the reaction could be significantly increased by performing it at 80°C leading to the conversion of about 19% of the substrate in 24 h with selectivity for Dec-5-ene of ca.

64% as determined by gas chromatographic analysis (Table 3.5 & Figure 3.29). We have compared the performance of 4 in the metathesis of 1-hexene with tungsten

15 pentamethyl grafted on SiO2-700 (≡SiO-WMe5, 5) and with its oxo analog [(≡Si-O-

)W(=O)Me3], 6. After 24 h of reaction at 80°C, we found catalyst 4 to be much more selective than 5 and 6 under the same reaction conditions, but at the cost of a drop in 1- 104 hexene conversion when compared to the latter catalysts. The reduction in the rate of

1-hexene conversion for 4 when compared to 5 and 6 could be attributed to the steric and/or electronic effect of the iminato ligand. As the selectivity in olefin metathesis is generally linked to the occurrence of consecutive metathesis processes, we investigated, as well, the selectivity of 6 at a degree of conversion comparable to that of

4 that was afforded after just 4.5 h of reaction. Even under these conditions the selectivity observed for the unsubstituted W-oxo system (6) is lower than that for 4. This reveals the high potential of this synthetic strategy towards the preparation of catalysts with enhanced selectivity for olefin metathesis.

The dec-5-ene formed in the presence of 4 is mostly found as the thermodynamic E- isomer product (Table 3.5). The preparation of supported Z-selective W-oxo catalysts has so far remained elusive in spite of the application of other W-oxo complexes carrying bulky ligands at the metal center such as dAdPO, OSitBu3, and HMTO. Indeed, it appears that the search for the suitable coordination environment leading to efficient

W-oxo, Z-selective olefin metathesis catalyst is still open. We believe that the preparation technique displayed in this study will be advantageous for the straightforward synthesis of judiciously designed tungsten-oxo catalysts with enhanced selectivity for olefin metathesis.

Dipp Table 3.5: Homo-coupling of 1-hexene using [(≡Si-O-)W(=O)(CH3)2-Im N] (4) and compared with [(≡Si-O- 15 16 )W(CH3)5] (5) and [(≡Si-O-)W(=O)(CH3)3] (6) .

105

Catalyst Reactant/Catalyst Conversion (%) Selectivity (%) Selectivity (%) TON Dec-5-ene (Z)-Dec-5-ene

Dipp a [(≡Si- O)W(=O)(CH3)2Im N] (4) 3517 18.6 64 30 654 a [(≡Si-O-)W(CH3)5] (5) 3311 84.6 9 -- 2801 a [(≡Si-O-)W(=O)(CH3)3] (6) 1981 78.5 34 -- 1555 b [(≡Si-O-)W(=O)(CH3)3] (6) 1981 19.7 48 31 391 a reaction condition: 24 h at 80 °C in vacuum sealed tube; breaction condition: 4.5 h at 80 °C in a vacuum-sealed tube.

3.4. Conclusion We have elaborated a straightforward synthetic strategy for the preparation of a functionalized and well-defined silica-supported W-oxo complex [(≡Si-O-)W(=O)(CH3)2-

ImDippN] (4), incorporating a bulky imidazolin-2-iminato ligand. The complex was systematically characterized via elemental analysis, FT-IR, solid state NMR and EXAFS spectroscopies. Under mild condition, and without the need of co-catalyst, 4 displayed good catalytic activity in the metathesis of terminal olefins such as propene and 1- hexene and was more selective towards dec-5-ene than other W-oxo catalysts under the same reaction conditions. Further efforts will be directed to the application of this synthetic approach to the synthesis of analog complexes of 4 able to promote the Z- selective metathesis of terminal olefins by choice of the suitable substitution pattern at the W center.

3.5. EXPERIMENTAL SECTION General procedure: All experiments were performed by using standard Schlenk and glovebox techniques under an inert nitrogenous atmosphere. Syntheses and treatments of the surface species were completed using high-vacuum lines (<10−5 mbar) and glovebox techniques. All solvents were dried, degassed and freshly distilled prior to use according to classical methods (over CaH2 for dichloromethane, over Na/benzophenone 106 for diethyl ether). Ethylene was dried and deoxygenated before use by passage through a mixture of freshly regenerated molecular sieves (3 Å) and R3-15 catalysts (BASF). IR spectra were recorded on a Nicolet 6700 FT-IR spectrometer using a DRIFT cell equipped with CaF2 windows. The IR samples were prepared under argon within a glovebox.

Solid State Nuclear Magnetic Resonance Spectroscopy. One-dimensional 1H MAS and 13C

CP/MAS solid-state NMR spectra were recorded on Bruker AVANCE III spectrometers operating at 400 MHz or 600 MHz resonance frequencies for 1H. 400 MHz experiments employed a conventional double resonance 4mm CP/MAS probe, while experiments at

600 MHz utilized a 3.2 mm HCN triple resonance probe. In all cases the samples were packed into rotors under inert atmosphere inside gloveboxes. Dry nitrogen gas was utilized for sample spinning to prevent degradation of the samples. NMR chemical shifts are reported with respect to the external references, TMS and adamantane. 13C CP/MAS

NMR experiments used the following sequence–90 0 pulse on the proton (pulse length

2.4 s), then a cross-polarization step with a contact time of typically 2 ms, and finally 107 acquisition of the 13C signal under high-power proton decoupling. The delay between the scans was set to 5 s to allow the complete relaxation of the 1H nuclei and the number of scans ranged between 5000 and 10000 for 13C and was 32 for 1H. An exponential apodization function corresponding to a line broadening of 80 Hz was applied prior to the Fourier transformations.

The 2D 1H-13C heteronuclear correlation (HETCOR) solid-state NMR spectroscopy experiments were conducted on a Bruker AVANCE III spectrometer operating at 600

MHz using a 3.2 mm MAS probe. These experiments were performed according to the

13 following sequence–90° proton pulse, t1 evolution period, CP to C and detection of the

13C magnetization under TPPM decoupling. During the cross-polarization step, a ramped radio frequency (RF) field centered at 75 kHz was applied to the protons, while the 13C channel RF field was matched to obtain an optimal signal. A total of 32 t1 increments with 4000 scans each were collected; sample-spinning frequency was 8.5 kHz. Using a short contact time (0.2 ms) for the CP step, the polarization transfer in the dipolar correlation experiment was verified to be selective for the first coordination sphere around tungsten, such that correlations occurred only between pairs of attached 1H-13C spins (C-H directly bonded).

1H-1H multiple-quantum spectroscopy. Two-dimensional double-quantum (DQ) and triple-quantum (TQ) experiments were recorded on a Bruker AVANCE III spectrometer operating at 600 MHz with a conventional double resonance 3.2 mm CP/MAS probe, according to the following general scheme–excitation of DQ coherences, t1 evolution, z- 108

filter and detection. The spectra were recorded in a rotor-synchronized fashion in t1 such that the t1 increment was set equal to one rotor period (45.45 µs). One cycle of the standard back-to-back (BABA) recoupling sequences was used for the excitation and reconversion period. Quadrature detection in w1 was achieved using the States-TPPI method. A spinning frequency of 22 KHz was used. The 90 ° proton pulse length was 2.5

µs, while a recycle delay of 5 s was used. A total of 128 t1 increments with 32 scans per increment were recorded. The DQ frequency in the w1 dimension corresponds to the sum of two single quantum (SQ) frequencies of the two coupled protons and correlates in the w2 dimension with the two corresponding proton resonances. The TQ frequency in the w1 dimension corresponds to the sum of the three SQ frequencies of the three- coupled protons and correlates in the w2 dimension with the three individual proton resonances. Conversely, groups of less than three equivalent spins will not give rise to diagonal signals in the spectrum.

Preparation of N-silylated 2-iminoimidazoline (1): N-silylated 2-iminoimidazoline was prepared according to the procedure reported in the previous literature.6a A solution of

1,3-bis(2,6-diisopropylphenylimidazolin-2-ylidene) (10 mmol) in toluene (20 mL) was treated dropwise with trimethylsilyl azide (14 mmol) at ambient temperature, and the resulting reaction mixture was subsequently refluxed for 72 h. Filtration and evaporation of the solvent afforded the imine product as yellowish solid that could be

o 1 purified by bulb to bulb distillation at 180 C/9 mbar. Yield 94%; H NMR (400 MHz, C6D6)

δ (ppm) -0.18 (9H, s, SiCH3), 1.17-1.19 (12H, d, CH3), 1.36-1.38 (12H, d, CH3), 3.10-3.21

13 (4H, sept, CHCH3), 5.93 (2H, s, NCH), 7.12 (2H, s, p-H), 7.19-7.23 (4H, m, m-H); C NMR 109

(70 MHz, C6D6) δ (ppm) 3.15 (SiCH3), 23.13 (CHCH3), 24.0 (CHCH3), 28.5 (CHCH3), 113.39

(CH), 123.54 (m-CH), 129.03 (p-CH), 134.70 (ipso-CH), 140.98 (NCN), 147.61 (o-CH). See

Figure below.

Dipp Dipp Preparation of Im NWOCl3 (Im = 1,3-bis(2,6-diisopropylphenylimidazolin-2- iminato) (2): To a solution of tungsten(VI) oxytetrachloride 1.44 g (4.20 mmol) in dichloromethane (40 mL) was added a solution of 2.0 g (4.2 mmol) of 1 in dichloromethane (10 mL) under stirring. The reaction mixture was stirred and refluxed overnight. The solvent was evaporated and the crude material was extracted three times with n-hexane (20 mL) to remove the unreacted 1. The crude material was crystallized from an n-hexane, dichloromethane mixture at -30°C to yield 2.5 g of

Dipp complex 2 [Im NWOCl3] (3.54 mmol, 84%). Anal. Calcd. for C27H36Cl3N3OW:, 44.75%

-1 -1 C;, 5.12% H; 15.00% Cl, 5.93% N; 25.94% W. Raman: 946 cm νs (W=O), 360 cm νs (W-

1 Cl); H NMR (400 MHz, C6D6) δ (ppm) 1.23-1.25 (12H, d, CH3), 1.37-1.39 (12H, d, CH3),

2.68-2.75 (4H, sept, CHCH3), 7.15 (2H, s, NCH), 7.36-7.38 (4H, m, CH), 7.53-5.56 (2H, m,

13 CH); C NMR (70 MHz, C6D6) δ (ppm) (70 MHz, CDCl3) 22.98 (CHCH3), 25.05 (CHCH3),

29.34 (CHCH3), 121.44 (CH), 124.70 (m-CH), 129.49 (p-CH), 131.91 (ipso-CH), 145.79

(NCN), 149.44 (o-CH).

Preparation of silica support partially dehydroxylated at 700°C (SiO2-700) The oxide support used in this study is Aerosil 200 silica from Degussa. Aerosil 200 is a non-porous flame silica in the form of regular particles of an average diameter of 150 Å and free of halides impurities. It was dehydroxylated under vacuum (10-5 torr) at 700°C for a 110

minimum of 15 hours leading to SiO2-700. The specific surface area for SiO2-700 was 200 ±

2 m2/g. IR (cm-1): 3745 cm-1, ν (Si-OH).

Dipp Grafting of 2 on SiO2-700 to produce [(≡Si-O-)Im NWOCl2] (3): In a double Schlenk, a solution of 2 (212 mg, 0.3 mmol, 1.1 equiv with respect to the amount of surface- accessible silanol) in 25 mL of dichloromethane was allowed to react with 1.0 g of SiO2-

700 at 25°C for 4 h. The reaction was performed under mild dynamic vacuum conditions to remove the HCl produced during the reaction. At the end of the reaction, the solvent was filtered and the resulting yellow solid was washed three times with dichloromethane (20 mL) and dried under a dynamic vacuum (<10−5 Torr, 2 h).

Elemental analysis found, 3.73% C;, 1.13% Cl; 0.5% N; 2.17% W. IR (cm-1): 2970-2877 ν

(C-H), 1653 cm-1 ν (C=N) and 1550-1350 cm-1 δ (C-H). 13C CP-MAS SS NMR (400 MHz, 298

K) δ 22 (CH3), 29 (CHCH3), 116 (CH), 124 (m-CH); 129 (p-CH); 137 (ipso-CH); 147 (NCN) ppm.

Dipp Alkylation of the grafted material 3 to yield [(≡Si-O-)W(=O)(CH3)2-Im N] (4) : In a double Schlenk 3 (0.7 g, 0.21 mmol) was suspended in 20 mL of dichloromethane and a heptane solution of ZnMe2 (0.315 mmol, 1.5 equiv. with respect to the amount of tungsten oxo chloride in 3) was added at -50oC. The mixture was allowed to warm to room temperature and stirred for further 2 h. At the end of the reaction, the resulting light gray solid was filtered and thoroughly washed with dichloromethane. The light gray solid was then dried under dynamic vacuum (<10−5 Torr) for 4 h. 1H MAS SS NMR (400

13 MHz, 298 K) δ 1.1 (WCH3), 1.9 (CH3), 3.6 (CH); 5.1 (CH); 6.3 (m-CH); 7.4 (p-CH) ppm; C

CP-MAS SS NMR (400 MHz, 298 K) δ 22 (CH3), 29 (CH), 50 (WCH3); 116 (CH); 124 (m-CH); 111

129 (p-CH), 133 (ipso-CH), 147 (NCN). Elemental analysis, found: 4.07% C; 1.33% Cl;

0.63% N; 2.09% W; 2.56% Zn,: Carbon to tungsten ratio (C/W) in complex 4: found, 29.85

± 0.2 expected value, 29.

Decomposition of 4 was carried out by placing 100 mg of material in a 270 mL reactor equipped with a septum. It was then heated at 150°C for 12 hours. Gas samples were withdrawn by an air-tight 0.5 mL gas syringe and injected into a GC instrument equipped with an FID detector. Methane was the only product detected in the gas phase.

Methane calibration was carried out by withdrawing 0.5 mL samples of pure methane at different, known pressure from a high vacuum Schlenck line and by building a graphic correlating the partial pressure of methane with its GC area. The amount of methane

−1 liberated by 4 at 150°C was quantified as 0.028 ± 0.05 mmol CH4.g (1.58 ± 0.2 CH4/W).

112

3.5.1. NMR Spectra

Figure 3.18: 1H (top) and 13C (bottom) NMR spectra of 1. 113

Figure 3.19: 1H (top) and 13C (bottom) NMR spectra of 2. 114

Figure 3.20: HSQC NMR spectrum of 2

Figure 3.21: Solid state 13C CP/MAS NMR spectrum (400 MHz, spinning rate 10 kHz, contact time = 1 ms, Dipp recycle delay = 2 sec, ns = 80 k) of [(≡Si-O-)W(=O)Cl2-Im N] (3).

115

3.5.2. SXRD analysis of 2

Dipp Table 3.6: Crystal data and structure refinement for [Im NWOCl3] (2).

Dipp Identification code 2 [Im NWOCl3] ∙ CH2Cl2

Empirical formula C28H38Cl5N3OW Formula weight 793.71 Temperature 200(2) K Wavelength 0.71073 Å Crystal system Triclinic Space group P-1 Unit cell dimensions a = 9.5406(5) Å b = 10.9846(6) Å c = 16.8508(9) Å Volume 1717.92(16) Å3 Z 2 Density (calculated) 1.534 Mg/m3 Absorption coefficient 3.775 mm-1 F(000) 788 Crystal size 0.200 x 0.200 x 0.200 mm3 Theta range for data collection 1.904 to 29.209°. Index ranges -13<=h<=13, -14<=k<=13, -23<=l<=23 Reflections collected 16470 Independent reflections 8462 [R(int) = 0.0261] Completeness to theta = 25.000° 94.0 % 116

Absorption correction Numerical Max. and min. transmission 0.5252 and 0.3755 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 8462 / 4 / 363 Goodness-of-fit on F2 1.116 Final R indices [I>2sigma(I)] R1 = 0.0455, wR2 = 0.1335 R indices (all data) R1 = 0.0550, wR2 = 0.1459 Extinction coefficient n/a Largest diff. peak and hole 4.012 and -2.005 e.Å-3

Table 3.7: Atomic coordinates (x 104) and equivalent isotropic displacement parameters (Å2x 103)

x Y z U(eq) N(1) 3978(5) 1592(5) 2747(3) 29(1) O(1) 6142(6) 2165(5) 4029(3) 48(1) W(1) 5827(1) 2106(1) 3005(1) 29(1) Cl(1) 5600(2) 4238(2) 3136(1) 48(1) Cl(2) 7636(2) 2680(2) 2152(2) 56(1) Cl(3) 6482(2) 120(2) 2630(1) 44(1) C(1) 2579(6) 1279(6) 2666(3) 26(1) N(2) 1788(5) 997(6) 1952(3) 33(1) C(2) 379(7) 735(8) 2106(4) 40(2) C(3) 295(6) 846(7) 2919(4) 37(1) N(3) 1674(5) 1196(5) 3257(3) 28(1) C(4) 2335(7) 953(8) 1159(4) 40(2) C(5) 2793(8) 2071(9) 892(4) 47(2) C(6) 3258(11) 1967(12) 107(5) 65(3) C(7) 3276(12) 856(13) -350(5) 78(3) C(8) 2836(11) -256(11) -71(5) 66(3) C(9) 2352(8) -230(9) 711(4) 49(2) C(10) 2773(10) 3344(9) 1389(5) 53(2) C(11) 3929(14) 4370(12) 1214(7) 78(3) C(12) 1344(14) 3736(12) 1364(8) 81(3) C(13) 1950(9) -1427(9) 1028(5) 59(2) C(14) 3240(20) -2075(19) 1097(19) 86(4) 117

C(15) 790(20) -2170(20) 409(16) 86(4) C(13') 1950(9) -1427(9) 1028(5) 59(2) C(14') 3200(30) -1730(30) 1540(20) 86(4) C(15') 1180(40) -2610(30) 480(20) 86(4) C(16) 2107(6) 1440(6) 4101(3) 31(1) C(17) 2332(7) 2679(7) 4498(4) 36(1) C(18) 2774(9) 2876(8) 5318(4) 48(2) C(19) 2971(9) 1902(9) 5696(4) 52(2) C(20) 2742(9) 698(9) 5284(4) 47(2) C(21) 2292(7) 424(7) 4475(4) 34(1) C(22) 2053(9) 3756(7) 4081(5) 47(2) C(23) 2899(11) 5007(9) 4466(6) 63(2) C(24) 455(11) 3835(10) 4041(7) 64(2) C(25) 2044(8) -909(7) 4031(4) 41(2) C(26) 1117(9) -1800(9) 4484(6) 53(2) C(27) 3450(12) -1362(10) 3869(8) 75(3) C(28) 7895(12) 5923(9) 1940(5) 58(2) Cl(4) 8085(3) 5211(2) 946(1) 63(1) Cl(5) 9561(5) 6471(5) 2452(2) 61(1) C(28') 7895(12) 5923(9) 1940(5) 58(2) Cl(4') 8085(3) 5211(2) 946(1) 63(1) Cl(5') 6863(5) 7088(3) 1957(2) 48(1) N(1) 3978(5) 1592(5) 2747(3) 29(1) O(1) 6142(6) 2165(5) 4029(3) 48(1)

Table 3.8: Bond lengths [Å] and angles [°] for [ImDippNWOCl3] (2).

N(1)-C(1) 1.321(8) N(1)-W(1) 1.782(5) O(1)-W(1) 1.729(5) W(1)-Cl(3) 2.3568(18) W(1)-Cl(1) 2.3606(19) W(1)-Cl(2) 2.3640(19) C(1)-N(3) 1.349(7) C(1)-N(2) 1.360(7) N(2)-C(2) 1.374(8) 118

N(2)-C(4) 1.449(8) C(2)-C(3) 1.363(9) C(3)-N(3) 1.385(8) N(3)-C(16) 1.436(7) C(4)-C(5) 1.390(12) C(4)-C(9) 1.395(11) C(5)-C(6) 1.404(10) C(5)-C(10) 1.512(13) C(6)-C(7) 1.336(17) C(7)-C(8) 1.389(17) C(8)-C(9) 1.410(11) C(9)-C(13') 1.501(14) C(9)-C(13) 1.501(14) C(10)-C(12) 1.493(14) C(10)-C(11) 1.525(14) C(13)-C(14) 1.523(16) C(13)-C(15) 1.532(16) C(13')-C(15') 1.527(18) C(13')-C(14') 1.553(18) C(16)-C(17) 1.397(9) C(16)-C(21) 1.398(9) C(17)-C(18) 1.402(10) C(17)-C(22) 1.522(11) C(18)-C(19) 1.364(13) C(19)-C(20) 1.371(12) C(20)-C(21) 1.383(9) C(21)-C(25) 1.511(10) C(22)-C(23) 1.513(12) C(22)-C(24) 1.540(13) C(25)-C(27) 1.520(13) C(25)-C(26) 1.533(11) C(28)-Cl(5) 1.751(11) C(28)-Cl(4) 1.757(9) C(28')-Cl(5') 1.728(10) C(28')-Cl(4') 1.757(9) 119

C(1)-N(1)-W(1) 171.6(4) O(1)-W(1)-N(1) 108.7(3) O(1)-W(1)-Cl(3) 95.1(2) N(1)-W(1)-Cl(3) 95.18(17) O(1)-W(1)-Cl(1) 94.0(2) N(1)-W(1)-Cl(1) 93.01(18) Cl(3)-W(1)-Cl(1) 165.08(7) O(1)-W(1)-Cl(2) 123.0(2) N(1)-W(1)-Cl(2) 128.28(18) Cl(3)-W(1)-Cl(2) 82.11(7) Cl(1)-W(1)-Cl(2) 83.02(7) N(1)-C(1)-N(3) 127.4(5) N(1)-C(1)-N(2) 125.1(5) N(3)-C(1)-N(2) 107.4(5) C(1)-N(2)-C(2) 108.6(5) C(1)-N(2)-C(4) 125.9(5) C(2)-N(2)-C(4) 125.4(5) C(3)-C(2)-N(2) 108.0(6) C(2)-C(3)-N(3) 106.6(5) C(1)-N(3)-C(3) 109.3(5) C(1)-N(3)-C(16) 124.1(5) C(3)-N(3)-C(16) 126.5(5) C(5)-C(4)-C(9) 124.4(7) C(5)-C(4)-N(2) 118.7(7) C(9)-C(4)-N(2) 116.9(7) C(4)-C(5)-C(6) 116.1(9) C(4)-C(5)-C(10) 123.5(7) C(6)-C(5)-C(10) 120.4(8) C(7)-C(6)-C(5) 121.7(10) C(6)-C(7)-C(8) 121.6(8) C(7)-C(8)-C(9) 120.0(9) C(4)-C(9)-C(8) 116.1(9) C(4)-C(9)-C(13') 123.8(7) C(8)-C(9)-C(13') 120.0(8) C(4)-C(9)-C(13) 123.8(7) 120

C(8)-C(9)-C(13) 120.0(8) C(12)-C(10)-C(5) 113.1(8) C(12)-C(10)-C(11) 111.3(10) C(5)-C(10)-C(11) 114.4(9) C(9)-C(13)-C(14) 110.4(11) C(9)-C(13)-C(15) 103.3(13) C(14)-C(13)-C(15) 113.2(13) C(9)-C(13')-C(15') 121(2) C(9)-C(13')-C(14') 111.7(14) C(15')-C(13')-C(14') 111(2) C(17)-C(16)-C(21) 123.9(6) C(17)-C(16)-N(3) 118.0(6) C(21)-C(16)-N(3) 118.1(6) C(16)-C(17)-C(18) 116.1(7) C(16)-C(17)-C(22) 122.4(6) C(18)-C(17)-C(22) 121.4(7) C(19)-C(18)-C(17) 121.2(7) C(18)-C(19)-C(20) 120.9(7) C(19)-C(20)-C(21) 121.5(7) C(20)-C(21)-C(16) 116.4(7) C(20)-C(21)-C(25) 120.7(7) C(16)-C(21)-C(25) 122.8(6) C(23)-C(22)-C(17) 113.6(8) C(23)-C(22)-C(24) 110.0(7) C(17)-C(22)-C(24) 110.5(7) C(21)-C(25)-C(27) 110.7(7) C(21)-C(25)-C(26) 112.3(6) C(27)-C(25)-C(26) 110.2(7) Cl(5)-C(28)-Cl(4) 110.6(6) Cl(5')-C(28')-Cl(4') 111.2(5) C(20)-C(21)-C(25) 120.7(7) Symmetry transformations used to generate equivalent atoms:

121

Table 3.9: Anisotropic displacement parameters (Å2x 103) for material 2. The anisotropic displacement factor exponent takes the form: -2p2[ h2 a*2U11 + ... + 2 h k a* b* U12 ]

U11 U22 U33 U23 U13 U12 N(1) 32(2) 23(3) 34(2) 8(2) 3(2) 6(2) O(1) 61(3) 45(3) 37(2) 11(2) -18(2) 4(2) W(1) 27(1) 31(1) 29(1) 4(1) 0(1) 4(1) Cl(1) 59(1) 30(1) 53(1) 1(1) 9(1) 6(1) Cl(2) 50(1) 50(1) 77(1) 24(1) 28(1) 13(1) Cl(3) 41(1) 37(1) 56(1) 10(1) 9(1) 13(1) C(1) 30(3) 25(3) 26(2) 4(2) 2(2) 9(2) N(2) 28(2) 43(3) 26(2) 2(2) -1(2) 6(2) C(2) 25(3) 59(5) 36(3) 9(3) 0(2) 7(3) C(3) 26(3) 51(4) 33(3) 7(3) 4(2) 8(2) N(3) 27(2) 29(3) 27(2) 5(2) 2(2) 7(2) C(4) 31(3) 59(5) 27(3) 5(3) 2(2) 6(3) C(5) 49(4) 64(6) 32(3) 19(3) 1(3) 12(3) C(6) 75(6) 93(8) 35(4) 24(4) 13(4) 18(5) C(7) 80(7) 122(11) 33(4) 15(5) 16(4) 16(7) C(8) 71(6) 83(7) 37(4) -11(4) 10(4) 6(5) C(9) 42(4) 66(5) 31(3) -7(3) 4(3) 1(3) C(10) 61(5) 56(6) 46(4) 21(4) -1(3) 11(4) C(11) 92(8) 71(8) 73(7) 36(6) -5(6) -3(6) C(12) 88(8) 71(8) 85(8) 9(6) -15(6) 29(6) C(13) 59(5) 53(5) 55(5) -12(4) 10(4) 0(4) C(14) 88(8) 42(8) 122(11) 3(7) 4(8) -3(6) C(15) 88(8) 42(8) 122(11) 3(7) 4(8) -3(6) C(13') 59(5) 53(5) 55(5) -12(4) 10(4) 0(4) C(14') 88(8) 42(8) 122(11) 3(7) 4(8) -3(6) C(15') 88(8) 42(8) 122(11) 3(7) 4(8) -3(6) C(16) 34(3) 39(4) 22(2) 6(2) 5(2) 10(2) C(17) 37(3) 36(4) 34(3) 2(2) 10(2) 7(2) C(18) 53(4) 48(5) 37(3) -9(3) 5(3) 7(3) C(19) 57(5) 70(6) 27(3) 3(3) -3(3) 16(4) C(20) 51(4) 56(5) 38(3) 16(3) -2(3) 15(3) C(21) 41(3) 36(4) 29(3) 9(2) 5(2) 122

C(22) 65(5) 33(4) 43(4) 3(3) 18(3) 11(3) C(23) 78(6) 37(5) 71(6) -2(4) 23(5) 3(4) C(24) 71(6) 48(6) 77(6) 10(4) 0(5) 25(4) C(25) 58(4) 29(4) 39(3) 11(3) 5(3) 11(3) C(26) 48(4) 46(5) 65(5) 14(4) 7(4) 6(3) C(27) 79(7) 42(6) 109(9) 17(5) 47(6) 11(4) C(28) 93(7) 43(5) 41(4) 10(3) 14(4) 13(4) Cl(4) 80(2) 58(1) 49(1) 4(1) 12(1) 12(1) Cl(5) 71(3) 62(3) 40(2) -10(2) -23(2) 5(2) C(28') 93(7) 43(5) 41(4) 10(3) 14(4) 13(4) Cl(4') 80(2) 58(1) 49(1) 4(1) 12(1) 12(1) Cl(5') 96(3) 28(2) 32(1) 7(1) 8(2) 43(2)

Table 3.10: Hydrogen coordinates (x 104) and isotropic displacement parameters (Å2x 10 3) for Dipp [Im NWOCl3] (2)

x y z U(eq) H(2A) -401 515 1714 48 H(3A) -545 709 3199 44 H(6A) 3567 2706 -106 78 H(7A) 3597 824 -878 94 H(8A) 2862 -1034 -407 80 H(10A) 2981 3253 1961 63 H(11A) 3980 5112 1632 117 H(11B) 3704 4586 687 117 H(11C) 4846 4074 1211 117 H(12A) 646 3114 1555 121 H(12B) 1062 3802 810 121 H(12C) 1386 4550 1712 121 H(13A) 1550 -1252 1567 71 H(14A) 4012 -1493 1421 130 H(14B) 3549 -2342 558 130 H(14C) 2985 -2807 1358 130 H(15A) -15 -1716 394 130 H(15B) 461 -2991 561 130 123

H(15C) 1168 -2295 -125 130 H(13B) 1248 -1222 1435 71 H(14D) 3634 -978 1914 130 H(14E) 3916 -2013 1179 130 H(14F) 2846 -2392 1840 130 H(15D) 178 -2525 385 130 H(15E) 1226 -3333 747 130 H(15F) 1626 -2718 -30 130 H(18A) 2938 3702 5616 57 H(19A) 3272 2059 6252 62 H(20A) 2896 37 5560 56 H(22A) 2338 3571 3515 56 H(23A) 3905 4929 4533 95 H(23B) 2546 5272 4995 95 H(23C) 2795 5628 4121 95 H(24A) 297 4553 3792 96 H(24B) 122 3936 4587 96 H(24C) -72 3066 3718 96 H(25A) 1534 -925 3499 49 H(26A) 198 -1519 4565 79 H(26B) 1595 -1803 5009 79 H(26C) 964 -2646 4169 79 H(27A) 3979 -853 3518 112 H(27B) 3267 -2239 3604 112 H(27C) 4009 -1286 4380 112 H(28A) 7367 6629 1931 70 H(28B) 7340 5311 2226 70 H(28C) 7448 5284 2245 70 H(28D) 8845 6285 2205 70

3.5.3. Buried Volume Computational Details All the DFT geometry optimizations were performed at the GGA B3LYP level with the Gaussian09 package.17The electronic configuration of the systems was described with the SVP basis set18 for H, C, Cl and O while for W we adopted the quasi-relativistic SDD effective core potential.19 All geometries were characterized as minimum through frequency calculations. The %VBur reported in 124

20 the paper have been computed using the SambVca package developed by some of us. The complexes are oriented placing the metal center at the origin of the sphere with the W-ligand distance fixed at 1.8 Å in order to have a comparison of the three ligands not biased on different W- ligand distances.

3.5.4. EXAFS Studies 4.0 data data 3.5 fit 2 fit 3.0

1 2.5

) -4 ) 2.0

-3 0

1.5 (R)| (A

(q)] (A (q)] 

 |

Im[ 1.0 -1

0.5

-2 0.0

0 5 10 15 20 0 1 2 3 4 5 6 k (Å-1) R [Å] 3 Figure 3.22: Fourier transform of the k -weighted EXAFS at tungsten W LIII-edge (right) and its corresponding back Fourier transform (left) in the [1.5, 5.7] Å range with comparison to simulated curves for a tungsten metallic foil. Open symbol: experimental data; full line: fit

Table 3.11: EXAFS parameters obtained for a tungsten metallic foil (Im-3m). A total of 19 scattering paths with 5 single and 14 scattering paths were used to fit the data in the [1.5, 5.7] Å R-range (multiple scattering paths are omitted for the sake of clarity). The cell parameter was determined equal to 3.157 ± 0.004 Å. The error propagation for the half length of each scattering path is not referenced in the table and underlined characters denote fixed parameters.

Paths So2 N D 103 x 2 E R (%) (Å) (Å2) (eV) W→W 0.94 ± 0.09 8 2.73 2.9 ± 0.3 5 ± 1 0.3 W→W 6 3.16 3.6 ± 0.5 W→W 12 4.31 4.1 ± 0.6 W→W 24 5.23 4.7 ± 0.6 W→W 8 5.47 6 ± 1

125

6 Fit without chlorine shell Fit with chlorine shell

Fit without chlorine shell 5 4 Fit with chlorine shell

3 4

) -4 2 3

)

-3

1 (R)| (A

 |

(q)] (A (q)] 2



Im[ 0 1 -1

0 -2 0 5 10 15 0 1 2 3 4 5 k (Å-1) R [Å] Figure 3.23: Fourier transform of the k3-weighted EXAFS at tungsten W LIII-edge (right) and its corresponding back Fourier transform (left) in the [0.54, 3.9] Å range with comparison to simulated curves for compound 4. Models with and without chlorine shell are compared. Open symbol: experimental data; full line: fit.

Table 3.12: . EXAFS parameters obtained for compound 4 for two different models including or excluding a chlorine shell.

Paths N D 103 x 2 E R (%) (Å) (Å2) (eV) Fit without chlorine shell W→O 1.7 ± 0.1 1.72 ± 0.01 1.5 12 ± 4 0.6 W→N/O 2.4 ± 0.2 1.95 ± 0.02 2.3 W→C 2.5 ± 0.3 2.36 ± 0.02 2.7 Fit with chlorine shell W→O 1.0 ± 0.2 1.71 ± 0.01 1.5 6 ± 4 0.5 W→C/O/N 3.4 ± 0.2 1.95 ± 0.05 11 ± 3 W→Cl 0.6 ± 0.2 2.28 ± 0.03 3.5

126

3.5.5. Catalysis Data Dipp Propene metathesis catalyzed by [(≡Si-O-)W(=O)(CH3)2-Im N] 4 in a batch reactor: In a

Dipp glove box, complex 4 [(≡Si-O-)W(=O)(CH3)2-Im N] (50 mg, 0.0057 mmol W) was introduced in a batch reactor (160 mL). After evacuation of Argon, about 150 (or 3050) equivalents (with respect to tungsten) of dry purified propene were added at room temperature and the reactor was heated to 150°C (heating rate: 4°C/min). The composition of the gas phase was monitored by GC in below figures.

Self-metathesis of 1-hexene: In a glovebox, scintillation vials containing a wing shaped magnetic stirred were charged with different tungsten supported catalysts on silica. A solution of 4.0 mmol of 1-hexene was added. The flask was removed from the glovebox and connected to a vacuum line where it was cooled to -78°C and evacuated to ca. 300-

500 mTorr (usually the onset of bubbling). The flask was sealed under static vacuum, allowed to warm to room temperature. The reaction mixture was stirred at 600 rpm and heated to 80°C. Aliquots of the reaction mixture were quenched by exposure to air and addition of dichloromethane and subjected to gas chromatography analysis to determine the conversion and E/Z selectivity. The resulting solution was analyzed by

GC/FID (Agilent 6850) equipped with a fused silica column. The identity of the metathesis products was assessed by the injection of pure standards.

127

3.5.6. NMR and GC Analysis of the Product Mixture of 1-hexene Self metathesis Reactions Catalyzed by 4, 5 and 6 Under Different Catalytic Conditions

Figure 3.24: 1H (top) and the 13C (bottom) NMR spectra of the product mixture from 1-hexene self- metathesis reaction catalyzed by 4 at 80°C for 24 h.

128

Figure 3.25: 1H (top) and the 13C (bottom) NMR spectra of the product mixture from the 1-hexene self- metathesis reaction catalyzed by 6 at 80 °C for 4.5 h. 129

Figure 3.26: 1H NMR spectrum of pure (E)-Dec-5-ene.

Figure 3.27: 1H NMR spectrum of pure (Z)-Dec-5-ene. 130

Figure 3.28: 1H NMR spectrum of the mixture of pure (E)- and (Z)-Dec-5-ene.

Figure 3.29: GC chromatogram of the product mixture from the 1-hexene self-metathesis reaction catalyzed by 4 at 80°C for 24 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product. 131

Figure 3.30: GC chromatogram of the product mixture from the 1-hexene self-metathesis reaction catalyzed by 5 at 80 °C for 24 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product.

Figure 3.31: GC chromatogram of the product mixture from the 1-hexene self-metathesis reaction catalyzed by 6 at 80°C for 24 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product.

132

Figure 3.32: GC chromatogram of the product mixture from the 1-hexene self-metathesis reaction catalyzed by 6 at 80°C for 4.5 h. The numbers in the chromatogram refer to the number of carbon atoms in the metathesis product.

133

3.6. References 1. Mazoyer, E.; Szeto, K. C.; Merle, N.; Thivolle-Cazat, J.; Boyron, O.; Basset, J. M.; Nicholas, C. P.; Taoufik, M., Insights into the deactivation mechanism of supported tungsten hydride on alumina (W-H/Al2O3) catalyst for the direct conversion of ethylene to propylene. J. Mol. Catal. a-Chem. 2014, 385, 125-132.

2. Peryshkov, D. V.; Schrock, R. R.; Takase, M. K.; Müller, P.; Hoveyda, A. H., Z- Selective Olefin Metathesis Reactions Promoted by Tungsten Oxo Alkylidene Complexes. J. Am. Chem. Soc. 2011, 133 (51), 20754-20757.

3. Schowner, R.; Frey, W.; Buchmeiser, M. R., Cationic Tungsten-Oxo-Alkylidene-N- Heterocyclic Carbene Complexes: Highly Active Olefin Metathesis Catalysts. J. Am. Chem. Soc. 2015, 137 (19), 6188-6191.

4. Tamm, M.; Randoll, S.; Bannenberg, T.; Herdtweck, E., Titanium complexes with imidazolin-2-iminato ligands. Chem. Commun. 2004, (7), 876-877.

5. Nomura, K.; Bahuleyan, B. K.; Tsutsumi, K.; Igarashi, A., Synthesis of (Imido)vanadium(V) Alkyl and Alkylidene Complexes Containing Imidazolidin-2-iminato Ligands: Effect of Imid Ligand on ROMP and 1,2-C-H Bond Activation of Benzene. Organometallics 2014, 33 (22), 6682-6691.

6. (a) Tamm, M.; Randoll, S.; Herdtweck, E.; Kleigrewe, N.; Kehr, G.; Erker, G.; Rieger, B., Imidazolin-2-iminato titanium complexes: synthesis, structure and use in ethylene polymerization catalysis. Dalton T. 2006, (3), 459-467; (b) Shoken, D.; Sharma, M.; Botoshansky, M.; Tamm, M.; Eisen, M. S., Mono(imidazolin-2-iminato) Titanium Complexes for Ethylene Polymerization at Low Amounts of Methylaluminoxane. J. Am. Chem. Soc. 2013, 135 (34), 12592-12595; (c) Nomura, K.; Bahuleyan, B. K.; Zhang, S.; Sharma, P. M. V.; Katao, S.; Igarashi, A.; Inagaki, A.; Tamm, M., Synthesis and Structural Analysis of (Imido)vanadium(V) Dichloride Complexes Containing Imidazolin-2-iminato- and Imidazolidin-2-iminato Ligands, and their Use as Catalyst Precursors for Ethylene (Co)polymerization. Inorg. Chem. 2014, 53 (1), 607-623.

7. Trambitas, A. G.; Melcher, D.; Hartenstein, L.; Roesky, P. W.; Daniliuc, C.; Jones, P. G.; Tamm, M., Bis(imidazolin-2-iminato) Rare Earth Metal Complexes: Synthesis, Structural Characterization, and Catalytic Application. Inorg. Chem. 2012, 51 (12), 6753- 6761.

8. (a) Mougel, V.; Coperet, C., Isostructural Molecular and Surface Mimics of the Active Sites of the Industrial WO3/SiO2 Metathesis Catalysts. ACS. Catal. 2015, 5 (11), 6436-6439; (b) Mougel, V.; Pucino, M.; Coperet, C., Strongly sigma Donating Thiophenoxide in Silica-Supported Tungsten Oxo Catalysts for Improved 1-Alkene Metathesis Efficiency. Organometallics 2015, 34 (3), 551-554. 134

9. Mazoyer, E.; Merle, N.; de Mallmann, A.; Basset, J. M.; Berrier, E.; Delevoye, L.; Paul, J. F.; Nicholas, C. P.; Gauvin, R. M.; Taoufik, M., Development of the first well- defined tungsten oxo alkyl derivatives supported on silica by SOMC: towards a model of WO3/SiO2 olefin metathesis catalyst. Chem. Commun. 2010, 46 (47), 8944-8946.

10. (a) Dehnicke, K.; Greiner, A., Unusual complex chemistry of rare-earth elements: Large ionic radii-small coordination numbers. Angew. Chem. Int. Ed. 2003, 42 (12), 1340- 1354; (b) Dehnicke, K.; Krieger, M.; Massa, W., Phosphoraneiminato complexes of transition metals. Coordin. Chem. Rev. 1999, 182, 19-65; (c) Dehnicke, K.; Weller, F., Phosphorane iminato complexes of main group elements. Coordin. Chem. Rev. 1997, 158, 103-169; (d) Dehnicke, K.; Strahle, J., Phosphorane Iminato Complexes of Transition-Metals. Polyhedron 1989, 8 (6), 707-726.

11. Paulsen, A. L.; Kalampounias, A. G.; Berg, R. W.; Boghosian, S., Raman Spectroscopic Study of Tungsten(VI) Oxosulfato Complexes in WO3-K2S2O7-K2SO4 Molten Mixtures: Stoichiometry, Vibrational Properties, and Molecular Structure. J. Phys. Chem. A 2011, 115 (17), 4214-4222.

12. Mcdowell, R. S.; Kennedy, R. C.; Asprey, L. B.; Sherman, R. J., Infrared-Spectrum and Force-Field of Tungsten Hexachloride. J. Mol. Struct. 1977, 36 (1), 1-6.

13. Ferrari, A. C.; Robertson, J., Resonant Raman spectroscopy of disordered, amorphous, and diamondlike carbon. Phys. Rev. B 2001, 64 (7).

14. (a) Chauvin, Y., Olefin metathesis: The early days (Nobel lecture). Angew. Chem. Int. Ed. 2006, 45 (23), 3740-3747; (b) Schrock, R. R., High oxidation state multiple metal- carbon bonds. Chem. Rev. 2002, 102 (1), 145-179.

15. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Rossini, A. J.; Widdifield, C. M.; Dey, R.; Emsley, L.; Basset, J. M., WMe6 Tamed by Silica: [≡Si-O-WMe5] as an Efficient, Well-Defined Species for Alkane Metathesis, Leading to the Observation of a Supported W-Methyl/Methylidyne Species. J. Am. Chem. Soc. 2014, 136 (3), 1054-1061.

16. Hamieh, A.; Chen, Y.; Abdel-Azeim, S.; Abou-hamad, E.; Goh, S.; Samantaray, M.; Dey, R.; Cavallo, L.; Basset, J. M., Well-Defined Surface Species [( Si-O)W(=O)Me3) Prepared by Direct Methylation of [(≡Si-O-)W(=O)Cl3), a Catalyst for Cycloalkane Metathesis and Transformation of Ethylene to Propylene. Acs Catal. 2015, 5 (4), 2164- 2171.

17. Gomperts, R.; Frisch, M.; Panziera, J. P., Scalability of Gaussian 03 on SGI Altix: The Importance of Data Locality on CC-NUMA Architecture. Lect. Notes Comput. Sc. 2009, 5568, 93-103. 135

18. Weigend, F.; Ahlrichs, R., Balanced basis sets of split valence, triple zeta valence and quadruple zeta valence quality for H to Rn: Design and assessment of accuracy. Phys. Chem. Chem. Phys. 2005, 7 (18), 3297-3305.

19. Dolg, M.; Stoll, H.; Preuss, H., A Combination of Quasi-Relativistic Pseudopotential and Ligand-Field Calculations for Lanthanoid Compounds. Theor. Chim. Acta. 1993, 85 (6), 441-450.

20. Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.; Scarano, V.; Cavallo, L., SambVca: A Web Application for the Calculation of the Buried Volume of N-Heterocyclic Carbene Ligands. Eur. J. Inorg. Chem. 2009, (13), 1759-1766.

136

CHAPTER 4: In Situ Generation of a Silica-Supported Tungsten Methyl Catalyst for Cyclooctane Metathesis 137

4.1 Introduction The concept of surface organometallic fragments (SOMF) has recently emerged as a simple and powerful tool to understand and design heterogeneous catalysts for several transformations. For instance, it has firmly established that metal hydrido carbenes (M-

H(=CHR)) or metal alkyl carbenes (M-R(=CHR)) are required to promote alkane metathesis reaction.1 In addition to the formation of surface active complexes, the support plays a role of a stabilizing ligand. For instance, highly sensitive tungsten hexamethyl2 was successfully grafted on different supports (silica3, alumina, and silica alumina4). In the case of silica grafted tungsten hexamethyl a well-defined and single site species was formed and characterized [(≡SiO)WMe5]. Despite its stability,

[(≡SiO)WMe5] shows a significant activity in the wide range of catalysis reaction, such as the metathesis of cyclic5 and acyclic6 alkanes at moderate temperatures, olefin metathesis, alkyne trimerization,7 and many other reactions. This put the complex under high focus and demand. However, the challenging synthesis of tungsten hexamethyl precludes its application outside specialized academic labs.

Synthesis of this highly desirable complex with our surface alkylation strategy will bring great benefits for the surface organometallic field and its commercialization (Scheme

4.28). Herein, we are reporting the syntheses of [(≡Si-O-)WMe5] by direct surface methylation of [(≡Si-O-)WCl5] using ZnMe2. The latter is obtained by grafting commercially available tungsten (VI) hexachloride on silica partially dehydroxylated at

700°C (SiO2-700). The methylated silica-supported tungsten complex was fully characterized by solid-state NMR spectroscopy, FTIR spectroscopy, microanalysis, and 138 mass balance techniques. Our experimental results are supported by DFT calculation for molecular modeling and the simulation of NMR chemical shifts.

Scheme 4.28: Reported synthesis of [(≡Si-O-)WMe5] complex and the new proposed synthetic strategy developed on SiO2-700.

4.2 RESULTS AND DISCUSSION

Preparation and characterization of [(≡Si-O-)WCl5] (1): Surface compound 1 was prepared by treating Aerosil 200® silica that was partially dehydroxylated under vacuum

-05 (10 mbar) at 700°C (SiO2-700) with WCl6 (Scheme 4.29). The reaction was performed in dichloromethane at room temperature for 2 hours. A slight dynamic vacuum was simultaneously applied to prevent side reactions by the liberated gaseous hydrogen chloride. Excess of unreacted WCl6 was efficiently washed with dichloromethane (4 x 10 mL). The resultant orange material was dried under high vacuum (10-5 bar) for 2 hours. 139

The infrared spectrum showed that the peak characteristic of free silanols at 3747 cm-1

(Figure 4.40, i) almost disappeared (Figure 4.40, ii). The elemental analysis of 1 found

3.97% W and 3.80% Cl with a molar ratio of Cl/W = 4.9 ± 0.2 (theoretical ratio for the monopodal surface complex is 5). Mass balance of 1 was completed by quantifying the

HCl gas released after water washing of the complex in a Schlenk tube equipped with a septum. The amount of HCl determined, in agreement with 1 having the formula [(≡Si-

O-)WCl5].

Scheme 4.29: Synthesis of 1 by grafting WCl6 on partially dehydroxylated silica at 700 °C (SiO2-700). 4.2.1 Methylation of 1

Several alkylating agents such as SnMe4, MeLi, Al2Me6, and ZnMe2 were tested to methylate 1. However, it was observed that tetramethyl tin was weakly reactive toward complex 1 at room temperature, whereas MeLi and Al2Me6 both react violently with silica support to produce a large amount of Si-Me and Si-Me2. Thus, SnMe4, MeLi, and

Al2Me6 were not further considered as a methylating agent. On the other hand, dimethylzinc was found to be more reactive than tetramethyl tin and milder than MeLi and Al2Me6 toward the support and thus consider for a methylating agent. Several other parameters e.g. temperature, duration of the reaction, the amount of methylating agent 140 and solvents were tested to find out the optimum reaction conditions. Here we report some of the conducted reactions with different conditions where we only reveal carbon

NMR for simplicity (Figure 4.33-Figure 4.39). We found that the reaction is best using 2.5 equivalents of ZnMe2 in dichloromethane at room temperature or at a lower temperature (-40°C). The reaction was completed in a very short time (15 minutes)

(Scheme 4.30). The resulting dark gray solid was washed several times using dichloromethane and dried under high vacuum (10-5 mbar).

4.2.2. Optimizing the methylating condition of complex Table 4.13: Optimization of methylating agents.

Methylating agent Remarks

SnMe4 weakly reactive toward starting complex MeLi react violently with the silica support Al2Me6 react violently with the silica support ZnMe2 found reactive with metal-halide and very less reactivity with the support

141

Figure 4.33: One-dimensional 13C CP/MAS NMR spectra (600 MHz 1H frequency) of 1 after treating with carbon enriched MeLi, peaks at -1.7 and -7.4 ppm are attributed to -Si-Me and -Si(Me)2.

Figure 4.34: One-dimensional 13C CP/MAS NMR spectra (600 MHz 1H frequency) of 1 after treating with

2.5 equivalents of ZnMe2 in pentane. 142

Figure 4.35: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with 5 equivalents of ZnMe2 in diethyl ether.

Figure 4.36: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with

2.5 equivalents of ZnMe2 in diethyl ether. 143

Figure 4.37: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with

2.5 equivalents of 97% carbon enriched ZnMe2 in dichloromethane for 15 min.

Figure 4.38: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with 2.5 equivalents of carbon enriched ZnMe2 (97 %) in dichloromethane for 4 hours. 144

Figure 4.39: One-dimensional 13C CP/MAS NMR spectra of 1 (600 MHz 1H frequency) after treating with >5 equivalents of carbon enriched ZnMe2 (97 %) in dichloromethane for 4 hours.

Scheme 4.30: Optimized condition for the alkylation of [(≡Si-O-)WCl5]

145

Figure 4.40: (i) SiO2-700 (ii) WCl6 grafted with SiO2-700 (1) (iii), after alkylation of 1 with ZnMe2 (2). 4.2.3. FTIR spectroscopy The FTIR spectrum of 2 (Figure 4.40, iii) revealed new sets of signals compared to the starting material 1 (Figure 4.40, ii). The new bands in the range of 3019-2859 cm-1, ν(C-

H), and 1459 cm-1, δ(C-H), were assigned to methyl fragments.

4.2.4. NMR Study 1H magic-angle spinning (MAS) SS-NMR spectrum of 2 (Figure 4.41a) exhibits one major signals at 2.0 ppm with a shoulder at 1.3 ppm and another small peak around -0.1 ppm.

All the signals auto-correlate in the two-dimensional (2D) double-quantum (DQ) (0.0,

2.6, 4.0, in F1) and triple-quantum (TQ) (0.0, 3.9, 6.0 ppm in F1) spectra (Figure 4.41b), indicating that all three resonances can be assigned to methyl groups. The peak at -0.1 146 ppm has a chemical shift characteristic to that of zinc methyl fragment (fitting either

8 ZnMeCl or ZnMe2). The other resonances are tentatively attributed to methyl groups coordinated to tungsten surface species.

13 13 C-enriched complex 2* (97% C-enriched) Zn*Me2 was prepared to enhance carbon

NMR sensitivity; the preparation protocol is added to the experimental part. NMR line shapes of 13C CP/MAS SS-NMR spectra observed in 2, and 2* (Figure 4.41c) are consistent with the presence of several species of four signals at -18, 46, 52 and 82 ppm.

Also, two-dimensional CP/MAS HETCOR NMR spectra (Figure 4.41d) indicate that the signals in 1H MAS SS-NMR at 1.3, 1.8 and 2.0 ppm are correlated with 13C CP/MAS SS-

NMR spectra at 46, 52 and 82 ppm. As mentioned above, DQ and TQ clearly show that only methyl groups are present. It is worth noting that the absence of cross peaks between these methyl signals indicates that the three peaks belong to three separate surface species. For the sake of clarity, these three species included in 2 will be named

2a, 2b, and 2c.

First, the signals at 82 ppm in 13C CP/MAS and 2.0 ppm in 1H NMR are assigned to [(≡Si-

O-)WMe5] that can be estimated to be in a significant amount; it will be named 2a.

Second, the peaks at 18 ppm in 13C CP/MAS and that at -0.1 ppm in 1H NMR are assigned to the methyl groups of ZnMeCl and ZnMe2. These species will collectively be referred as

(2b) and can be qualitatively estimated to be a minor species. Finally, the remaining 13C signals at 46 and 51 ppm are assigned to surface tungsten complex (2c). The resonances of complex (2c) are reminiscent to that of silica-supported tungsten bearing one carbyne 147

3 and two methyl groups [(≡Si-O-)WMe2(≡CH)]. This can be explained by the conversion of a significant fraction of [(≡Si-O-)WMe5] into [(≡Si-O-)WMe2(≡CH)] during the alkylation reaction, a process known for some transition metals.9 The absence of carbyne signal was investigated by studying the effect of the two zinc compounds

(ZnMe2 and ZnCl2). A well-defined tungsten complex [(≡Si-O-)WMe5] (3) was prepared according to our previous report.3 It was then thermally decomposed by treatment at

150°C to generate the analog carbyne [(≡Si-O-)WMe2(≡CH)] (4). It was then treated with

ZnMe2 (1M solution in heptane) or ZnCl2 (5 equiv. in CH2Cl2) in two separate experiments. In both cases after treatment, the carbyne peak at 299 ppm in 13C CP/MAS

NMR clearly disappeared (Figure 4.42). This may be interpreted by the formation of a tungsten carbyne adduct with the Lewis acid zinc moieties (Scheme 4.31). Examples of complexations of a carbyne by a Lewis acid center have already been reported.10 148

Figure 4.41: a) 1H MAS/ NMR of 2 . b) 2D 1H-1H DQ/SQ and 1H-1H TQ/SQ NMR spectra of 2 (both acquired with number of scans per increment = 32 per t1 increment, repetition delay = 5 s, number of individual t1 increments = 128. c) 13C CP/MAS NMR spectra of 2* d) 2D CP/MAS HETCOR NMR spectra of 2* acquired with short contact times of 0.2 ms under 8.5 kHz MAS frequency

149

Figure 4.42: Comparison of 13C-SS NMR of (i) 4, obtained by thermal decomposition at 150°C of well- defined [(≡Si-O-)WMe5], 3. (ii) Reaction of 4 with dichloromethane. (iii) The reaction of 4 with ZnMe2 in dichloromethane. (iv)The reaction of 4 with ZnCl2 in dichloromethane.

150

The formation of partially alkylated tungsten species was considered to explain the presence of several surface species. DFT calculations were carried out to investigate the origin of these two other broad peaks at 46 and 52 ppm observed in 13C SS-NMR.

Figure 4.43: Molecular models used for the (A) [(≡Si-O-)WMe4Cl] and (B) [(≡Si-O-)WMe5] species in DFT calculation.

Several surface species corresponding to the different degree of alkylation, [(≡Si-O-

)WMenCl5-n] were considered (Figure 4.43 & Table 4.14). The NMR-DFT calculation of the 13C chemical shift was performed on all the possible species and conformations.

151

Table 4.14: NMR-DFT calculations of the partial substitution of [(≡Si-O-) WCl5] Species

[(≡Si-O-)WMeCl4] _conf_1 [(≡Si-O-)WMe2Cl3]_conf2 [(≡Si-O-) [(≡Si-O-) WMeCl4] [(≡Si-O-) WMe Cl ]_conf1 _conf_2 2 3 WMe2Cl3]_conf3

Chemical C1=96.49 C1=77.47 C1=92.14 C1=86.20 C1=80.11 shift (ppm) C2=101.33 C2=85.04 C2=75.42

[(≡Si-O-) [(≡Si-O-) WMe3Cl2]_conf1 WMe3Cl2]_conf2 [(≡Si-O-) WMe4Cl]_conf1 [(≡Si-O-) WMe4Cl]_conf2 [(≡Si-O-)WMe5]

Chemical C1=91.30 C1=82.14 C1=73.59 C1=82.62 C1=72.38 shift (ppm) C2=88.80 C2=86.08 C2=69.97 C2=81.13 C2=84.21 C3=96.13 C3=72.10 C3=74.02 C3=84.16 C3=63.34 C4=91.78 C4=95.35 C4=85.38 C5=63.79

152

The calculated chemical shifts were scaled with the TMS chemical shielding determined using the same method. In all cases, the 13C-DFT chemical shifts (ranging from 70 to 100 ppm) are far from that of the experimental peaks.

This reduces significantly the possibility of attributing the two peaks (at 46 and 52 ppm) to [(≡Si-O-)WMenCl5-n] species. From those data, it appears that [(≡Si-O-)WMe5] exhibits

(depending on conformation) 13C resonances ranging from 85 to 75 ppm, in line with experimental results.

4.2.5. Thermal Decomposition of 2 (2T) Heating 2 at 150°C leads to the formation of 2T. 13C-CP/MAS SS-NMR spectroscopy shows the disappearance of the peak at 82 ppm and the appearance of broad resonances centered below 40 ppm. The two resonances at 46 and 52 ppm remain but with a different pattern as the former seems more intense (Figure 4.44). However, repeated attempts failed to detect the peaks expected to the carbyne groups at 298 ppm in the spectrum 13C (W≡C-H) and the peak at 7.8 ppm in 1H-NMR spectrum (W≡C-

H) is more probably due to complexations with zinc moieties as explained earlier.

4.2.6. Hydrogenolysis of 2 (2H)

Complex 2 was found to react readily with molecular H2 even at temperatures as low as

-78°C (1 bar, -78°C, 10 min) to afford corresponding hydride (2H). The reaction was monitored by FTIR spectroscopy showing that the ν(C-H) bands characteristic of the [W-

-1 -1 CH3] group at 3018 and 2990 cm decreased. New peaks around 1958, and 1978 cm were observed (Figure 4.45) and are consistent with the ν(W-H) bands of tungsten 153 hydride. Methane gas was detected by GC-FID after the analysis of the gas phase (0.23 ±

0.02 mmol.g-1).

Figure 4.44: (A) 1D 13C CP/MAS SS-NMR spectrum of 2*, prepared by 13C-labeled dimethyl zinc (97% 13C- enriched) with a 10 kHz MAS frequency (a repetition delay = 5 s, contact time = 2 ms, line broadening = 80 Hz and number of scans = 10000). (B) 1D 13C CP/MAS NMR spectrum of 2* after heating at 150°C for 2 hours. This is in agreement with the well-documented hydrogenolysis of surface tungsten alkyl complexes.11 It is important to add that the amount of tungsten hydride observed by 154

FTIR spectrum is qualitatively smaller than that prepared by the hydrogenation of [(≡Si-

O-)WMe5] (3) at the same temperature.

Figure 4.45: FTIR spectrum of (i) 2 (ii) 2H (iii) 2H - 2. 4.2.7. DFT calculations DFT calculations were performed to model the carbyne species, which may have formed from intermediate carbenes during the methylation of 1. The formation of the carbyne was considered from the fully alkylated [(≡Si-O-)WMe5] and the partial alkylated [(≡Si-O-

)WMe4Cl] species (Figure 4.49). A relaxed energy scan of the W-Cl bond was performed to approach the transition state for carbene formation from [(≡Si-O-)WMe4Cl] via elimination of gaseous HCl (traces were identified experimentally (Figure 4.47). The fully 155 optimized transition state corresponds to a proton transfer from one of the W-Me groups to the dissociating chloride ion (Figure 4.49).

Figure 4.46: Interconversion of [(≡Si-O-)WMe4Cl] to [(≡Si-O-)WMe5] calculated using PBE/Def2 TZVP level of theory. This transition state is much lower in energy (29.6 kcal/mol) than the transition state corresponding to carbene formation from the penta-methylated [(≡Si-O-)WMe5] species, 39.7 kcal/mol, (Figure 4.49A). The next step consists in the evolution of the intermediate carbene [(≡Si-O-)W(=CH2)Me3] to the carbyne [(≡Si-O-)W(≡CH)Me2] plus elimination of a methane molecule via an energy barrier of 28.8 kcal/mol (Figure 4.49B).

It is worth noting that the interconversion between [(≡Si-O-)WMe4Cl] and [(≡Si-O-

)WMe5] assisted by an external ZnMeCl molecule is easy and reversible since it requires the overcoming of a barrier of only 16.2 kcal/mol (Figure 4.46). 156

Figure 4.47: FTIR spectrum of the evolved gasses during the reaction of 1 with excess ZnMe2 in dichloromethane for one hour at room temperature

The free energy difference between the two states is only 0.3 kcal/mol, which indicates that the two species can be considered to be in equilibrium. Hence, it is possible that

[(≡Si-O-)WMe4Cl] is a significant route to the formation of carbene and later to carbyne.

NMR-DFT calculations were performed to determine whether the carbene ([(≡Si-O-

)W(=CH2)Me3]) and carbyne ([(≡Si-O-)W(≡CH)Me2]) species are compatible with the two

13 experimental C SS-NMR peaks at 46 and 52 ppm. The adducts with either ZnMe2 or

ZnCl2 were also included in our studies. The results reported in (Table 4.15 & Table 4.16) indicate that the peaks mentioned above can be assigned to the methyls of the tungsten carbyne species. 157

Scheme 4.31: Proposed mechanism for secondary product formation.

The carbyne complex formed from [(≡Si-O-)WMe4Cl] and [(≡Si-O-)WMe5] have peaks at

46 ppm while the peaks of its adducts (ZnMe2, ZnMeCl or ZnCl2) should be detected at

50 ppm. We can, therefore, assign the two experimental peaks at 46 and 52 ppm to the methyl groups of the carbyne complexes [(≡Si-O-)W(≡CH)Me2]) and its possible adducts formed with ZnMe2 and ZnCl2. Formation of those adducts is thermodynamically favored in case of ZnMe2, and its interaction with carbyne species is a spontaneous exothermic reaction. The energy released by the alkylation reaction and the zinc methyl interaction with the carbyne could be sufficient for the reaction of ZnCl2 with the carbyne.

158

Figure 4.48: Molecular models for the carbene [(≡Si-O-)W(=CH)Me3] (C) and carbyne [(≡Si-O-)W(≡CH)Me2] (D) species used in our DFT calculations of NMR spectra and the reaction kinetics.

159

Figure 4.49: (A) Energy profile calculated for the formation of carbene [(≡Si-O-)W(=CH2)Me3] from [(≡Si-O- )WMe5] or [(≡Si-O-)WMe4Cl] (B) and the formation of the tungsten carbyne complex [(≡Si-O-)W≡CH)Me2] from its analogue carbene [(≡Si-O-)W(=CH2)Me3].

160

13 Table 4.15: DFT-NMR C chemical shift of the carbene adduct species with ZnCl2 and ZnMe2.

Carbene-adduct species with ZnCl2 and ZnMe2

Species

Chemical C1=49.96 C1=48.28 C1=44.14 shift (ppm) C2=62.94 C2=66.38 C2=66.00 C3=224.14 C3=222.57 C3=-6.52 C4=227.81

Species

Chemical C1=38.40 C1=38.85 C1=42.96 shift (ppm) C2=38.15 C2=39.79 C2=38.95 C3=37.78 C3=38.77 C3=39.13 C4=212.70 C4=235.18 C4=-6.45 161

13 Table 4.16: DFT-NMR C chemical shift of the carbyne adduct species with ZnCl2 and ZnMe2.

Carbyne species with ZnCl2 and ZnMe2

Species

Chemical C1=45.21 C1=46.37 C1=44.95 shift (ppm) C2=287.69 C2=313.01 C2=-12.32 C3=341.58

Species

Chemical C1=48.08 C1=53.09 C1=48.71 shift (ppm) C2=47.49 C2=53.42 C2=48.15 C3=289.49 C3=308.59 C3=-12.35 C4=342.47

162

4.2.8. Cyclooctane metathesis One interesting reaction in alkane metathesis is the formation of higher cyclic alkanes starting from lower ones such as cyclooctane and cyclodecane. The produced macrocycles can serve as building blocks for highly useful intermediates in the field of drug,12 fragrance,13 dye,14 and polymer15 synthesis. Surface complexes 2, 2T and 2H, were tested for the metathesis of cyclooctane using a protocol similar to that reported previously.5 In a typical experiment, cyclooctane was added to 2 (ratio between substrate and total tungsten present in the catalyst is 570), frozen, evacuated and flame-sealed in an ampule. The mixture was stirred at 150°C for five days. At the end of the reaction, the ampule was frozen in liquid nitrogen. The liquid fraction after the run was extracted and analyzed by GC-FID.

More than 13% of initial cyclooctane was converted into cycloheptane (cC7, 68%) in addition to other smaller cycles and higher macrocyclic compounds ranging between cC13 and cC30 (32% out of total conversion) with an overall turnover number (TON) of

74 as determined by GC-FID (Table 4.18 & Figure 4.50). In comparison, employing [≡SiO-

W(Me)5] as catalysts resulted in a TON of 210. However, this is not surprising given the fact that 2 is a mixture (Table 4.17).

In another run using 2T, a slight increase in reactivity was observed (TON = 84). Using 2H as a catalyst gave an increase of conversion was revealed (TON = 144), which is much closer to the values observed with [≡SiO-W(Me)5], again it really means that 2 contains

W-Me bonds which can be transformed into W-H and likely WH(=CH2).

163

Figure 4.50: GC-FID chromatogram of cyclooctane metathesis products that have been catalyzed by 2, 2T and 2H (See Table 4.18). Reaction conditions: batch reactor, catalyst 2 (0.035 g, W loading, 0.17 mmol g- 1), cyclooctane (0.5 mL, 3.7 mmol), 120 h, 150 °C. Conversion = 13%, TON = 74. Products formed are ring- contracted (cC7) as well as ring-expanded species ranging from cC12 to cC30. For better understanding response for C7 was divided by factor of 4 in the chart.

Table 4.17: Comparison of the activity of different precursors in cyclooctane metathesis reaction expressed by TON.

Precursor TON 2 74 2T 84 2H 144

[≡SiO-W(Me)5] (3) 210 4 225

[≡SiO-W(Me)5] + ZnCl2 195

[≡SiO-W(Me)5] + ZnMe2 0

164

C4 C5 C6 C7 C9 C10 C11 C12 C13 C14 C15 8 4 5 6 7 9 10 11 12 13 14 15 Mass Complex (Days) TON Conv 15000 7500 9375 11250 13125 16875 18750 20625 22500 24375 26250 28125 B Complex2(1d) 18 3.2 98.7 14520 31.0 176.9 32.4 19.1 10.1 Complex2(5d) 70 12.3 96.6 13150 21.8 37.5 17.0 672.7 32.2 18.6 6.8 12.7 22.0 27.5 43.1 Complex2(8d) 74 13.1 94.2 13041 9.2 38.4 10.5 548.8 32.4 18.2 0 10.5 17.6 21.7 36.5 Complex 2H (5d) 84 14.7 95.6 12790 17.8 105.6 25.2 1471.2 40.5 23.5 0 25.8 44 30.1 108.1 Complex 2T(5d) 144 25.3 94.3 11200 14.0 60.4 30.2 840.0 37.4 21.6 0 13.4 23.5 55.7 50.6

C16 C17 C18 C19 C20 C21 C22 C23 C24 C25 C26 C27 C28 C29 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Complex (Days) 30000 31875 33750 35625 37500 39375 41250 43125 45000 46875 50625 52500 54375 54375 Complex2(1d) 12.0 12.8 11.8 10.7 Complex2(5d) 56.0 59.9 55.4 49.5 41.1 33.5 27.0 22.7 19.4 17.7 15.6 14.0 12.3 10.9 Complex2(8d) 49.1 51.9 46.9 40.9 33.2 26.3 20.7 17.5 15.0 15.7 12.2 11.0 9.6 8.2 Complex 2H (5d) 142.3 152.1 131.9 114.9 90.2 74.2 57.2 50 46.5 42.3 38.1 34.3 30.4 26.3 Complex 2T(5d) 62.8 65.4 57.8 50.7 41.6 34.1 27.8 24.0 21.0 23.7 17.5 15.7 13.8 12.4

165

Complementary tests were carried out to determine the influence of the zinc compounds over the catalytic performance. The addition of an excess amount of dimethylzinc to 3 yields a black solution once heated at 150°C (N.B. without ZnMe2 catalysis reaction solution is dark orange).

No catalytic activity was observed even after repeating reaction several times. By contrast, the addition of excess amount of ZnCl2 to [≡SiO-W(Me)5] did not alter its activity significantly (Table

4.17). Further investigations are underway to understand the mechanism for the deactivation of the tungsten pentamethyl moiety by ZnMe2.

4.3 CONCLUSIONS The current work describes a convenient method for the direct methylation of silica bound tungsten chloride complex [(≡Si-O-)WCl5] (1) at room temperature. The resulting complex (2) was characterized by advanced solid-state NMR spectroscopy, FTIR spectroscopy, and mass balance techniques. The target product [≡SiO-W(Me)5] was identified alongside other products which were likely to be a decomposed form of tungsten alkylidene complex. NMR-DFT calculation validated the proposed surface complexes. Finally, this complex was found to be catalytically active in the metathesis of cyclooctane at 150°C. Based on experimental evidence and theoretical calculation a possible pathway for catalyst deactivation was proposed.

4.4 EXPERIMENTAL SECTION General procedure: All experiments were performed by using standard Schlenk and glovebox techniques under an inert argon atmosphere. Syntheses and treatments of the surface species were completed using high vacuum lines (<10−5 mbar) and glovebox techniques. All solvents were dried, degassed and freshly distilled prior to use according to classical methods (over CaH2

166 for dichloromethane, over Na/benzophenone for diethyl ether and pentane). Hydrogen was dried and deoxygenated before use by passage through a mixture of freshly regenerated molecular sieves (3 Å) and R3-15 catalysts (BASF). IR spectra were recorded on a Nicolet 6700

FTIR spectrometer using a DRIFT cell equipped with CaF2 windows. The IR samples were prepared under argon within a glovebox. Typically, 32 scans were collected for each spectrum

(resolution 4 cm−1). Elemental analysis was performed at Mikroanalytisches Labor Pascher

(Germany). Gas-phase analysis of alkanes was performed using an Agilent 6850 gas chromatography column with a split injector coupled with an FID (flame ionization detector).

An HP-PLOT Al2O3 KCl 30 m × 0.53 mm, 20.00 mm capillary column coated with a stationary phase of aluminum oxide deactivated with KCl was used with helium as the carrier gas at 32.1 kPa. Each analysis was executed under the same conditions: a flow rate of 1.5 mL/min and an isotherm at 80°C.

Solid State Nuclear Magnetic Resonance Spectroscopy. One-dimensional 1H MAS and 13C

3.2 mm HCN triple resonance probe. In all cases, the samples were packed into rotors under argon atmosphere inside glove boxes. Dry nitrogen gas was utilized for sample spinning to prevent degradation of the samples. NMR chemical shifts are reported with respect to the external references, TMS, and adamantane. 13C CP/MAS NMR experiments used the following sequence–90° pulses on the proton (pulse length 2.4 s), then a cross-polarization step with a contact time of typically 2 ms, and finally acquisition of the 13C signal under high-power proton

167 decoupling. The delay between the scans was set to 5 s to allow the complete relaxation of the

1H nuclei and the number of scans ranged between 5 000 and 10 000 for 13C and was 32 for 1H.

An exponential apodization function corresponding to a line broadening of 80 Hz was applied to the Fourier transformations.

The 2D 1H-13C heteronuclear correlation (HETCOR) solid-state NMR spectroscopy experiments were conducted on a Bruker AVANCE III spectrometer operating at 600 MHz using a 3.2 mm

MAS probe. These experiments were performed according to the following scheme –90° proton

13 13 pulse, t1 evolution period, CP to C and detection of the C magnetization under TPPM decoupling. During the cross-polarization step, a ramped radio frequency (RF) field centered at

75 kHz was applied to the protons while the 13C channel RF field was matched to obtain an optimal signal. A total of 32 t1 increments with 4000 scans each was collected; sample-spinning frequency was 8.5 kHz. Using a short contact time (0.2 ms) for the CP step, the polarization transfer in the dipolar correlation experiment was verified to be selective in the first coordination sphere around tungsten, such that correlations occurred only between pairs of attached 1H-13C spins (C-H directly bonded).

1H-1H multiple-quantum spectroscopy. Two-dimensional double-quantum (DQ) and triple- quantum (TQ) experiments were recorded on a Bruker AVANCE III spectrometer operating at

600 MHz with a conventional double resonance 3.2 mm CP/MAS probe, according to the following general scheme–excitation of DQ coherences, t1 evolution, z-filter, and detection. The spectra were recorded in a rotor-synchronized fashion in t1 such that the t1 increment was set equal to one rotor period (45.45 µs). One cycle of the standard back-to-back (BABA) recoupling

168

sequences was used for the excitation and reconversion period. Quadrature detection in w1 was achieved using the States-TPPI method. A spinning frequency of 22 KHz was used. The 90 ° proton pulse length was 2.5 µs while a recycle delay of 5 s was used. A total of 128 t1 increments with 32 scans per increment were recorded. The DQ frequency in the w1 dimension corresponds to the sum of two single quanta (SQ) frequencies of the two coupled protons and correlates in the w2 dimension with the two corresponding proton resonances. The TQ frequency in the w1 dimension corresponds to the sum of the three SQ frequencies of the three-coupled protons and correlates in the w2 dimension with the three individual proton resonances. Conversely, groups of less than three equivalent spins will not give rise to diagonal signals in the spectrum.

DFT Computational Details. All quantum chemical computations were performed using the

Amsterdam Density Functional (ADF)16 and Gaussian 09 programs. First, geometry optimizations were performed using Gaussian 09 with the generalized gradient approximation

(GGA) exchange-correlation function developed by Perdew, Burke, and Ernzerhof (PBE)17an all electron Def2TZVP2918 basis set computational level. Geometry optimizations were carried out without any constraints. Thermodynamics and reaction kinetics calculations were carried out using single point at M06/Def2TZVP30 level19 and frequency calculations at PBE/Def2TZVP level of theory. The optimized geometries were used as input for the ADF NMR calculations. For the magnetic shielding tensor calculations (13C), the zeroth-order regular approximation (ZORA)20 was used to include relativistic effects both scalar and spin−orbit contributions.21 The NMR calculations used an all-electron triple-ζ basis set which included two polarization functions

(i.e., TZ2P) and a PBE functional.

169

Preparation of the silica partially dehydroxylated at 700°C. Typically, 4.0 g of Degussa Aerosil

200 silica was introduced into a quartz reactor fitting with a tubular furnace connected with high vacuum (10-5 mbar) line at 700 °C for 16 h. The temperature program was set to 90°C /h. IR

(cm−1): 3747 (ν(SiO−H)).

Preparation of [(≡Si-O-)WCl5] (1). In a double Schlenk, a solution of WCl6 (0.396 g, 1 mmol, 1.1 equivalent with respect to the amount of surface-accessible silanols) in 25 mL of dichloromethane was allowed to react with (3.0 g, 0.9 mmol) of SiO2-700 at 25 °C for 2 h. This reaction was performed under mild dynamic vacuum conditions to remove any of the HCl produced during the reaction. At the end of the reaction, the resulting orange solid was washed with dichloromethane (3×20 mL) and dried under a dynamic vacuum (<10−5 Mbar, 2 h).

Elemental analysis found 3.97% W; 3.8% Cl.

13 Preparation of Zn( CH3)2. In a 200 mL Schlenk, 15 mL of 1.6 M BuLi was taken under argon, and to that was added dropwise 3.4 g (24 mmol) of 13MeI diluted in 25 mL of pentane at −20°C with stirring. A white precipitate formed immediately, and the solution was stirred for another 3 hours. The precipitate was then filtered and dried under a vacuum to produce a white solid

(13MeLi). A small amount of 13MeLi was taken and titrated with deoxygenated water to quantify the amount of methyl lithium present by quantifying the release of methane (around 55%

13MeLi was found). In another double Schlenk, 13MeLi (22 mmol, 484 mg) was added with 20 mL of pentane, and to that, a solution of anhydrous ZnCl2 (10 mmol in 10 mL) was added at -20°C.

The reaction was allowed to reach room temperature slowly and remained at that temperature

170 for another 2 days. The volatiles was then trapped into a schlenk, and the dimethylzinc was collected as pentane solution.

Preparation of complex 2. (1.0 g, 0.3 mmol) of [(≡Si-O-)WCl5] was introduced into a double

Schlenk along with 20 mL of dichloromethane. A ZnMe2 (0.75 mmol) solution in hexane was added dropwise at 25°C and stirred for defined time period (15 min). After completion of the reaction, the resulting solid was filtered and thoroughly washed with dichloromethane. The complex was then dried under dynamic vacuum (<10−5 mbar) conditions for 2 h. 1H MAS SS-

13 NMR (400 MHz, 298 K): δ (ppm) -0.1 (CH3), 1.3 (CH3), 2.0 (CH3), 2.2 (CH3). C CP-MAS SS-NMR

(400 MHz, 298 K): δ (ppm) 46 (CH3), 52 (CH3), 82 (CH3). Gas quantification of 2 was carried out

−1 by its hydrolysis with the evolution of methane (0.57±0.2 mmol CH4•g ) in a ratio of carbon to tungsten equal 2.9±0.2. Elemental analysis of complex 2 gave 3.63% W, 0.87% C, 2.34% Zn,

0.13% H and 2.07% Cl, with a ratio of C/W=3.7 and a ratio of Cl/Zn=1.7 Zn/W=1.8.

Thermal decomposition of 2. In a glass reactor, (1.0 g, 0.3 mmol) of 3 was taken and heated at

150°C with ramp equals 60°C/h for 12 h to produce a dark gray powder.

Hydrogenation of 2. IR pellet of 3 (0.050 g, 0.015 mmol) was taken into a 150 mL IR cell equipped with CaF2 windows then evacuated. The IR cell tube was immersed into a solution of liquid nitrogen/acetone solution for 5 minutes. Dried hydrogen was introduced through it at -

78°C (800 mbar) for 10 minutes and then evacuated to before recording the infrared spectrum.

Similar results were obtained when hydrogenolysis reaction was repeated at room temperature.

171

Cyclooctane metathesis using complex 2. The reactions were performed using an ampoule filled with the catalyst (35 mg, W loading, 0.17 mmol g-1) inside a glovebox where cyclooctane (0.5 mL, 3.7 mmol) was added. The ampoule was sealed under vacuum, immersed in an oil bath and heated at 150°C for 5 days. At the end of the reaction, the ampoules were frozen using liquid nitrogen. The mixture was quenched by adding a fixed amount of dichloromethane, after filtration; the resulting solution was analyzed by GC-FID.

172

4.5 References 1. Pelletier, J. D. A.; Basset, J. M., Catalysis by Design: Well-Defined Single-Site Heterogeneous Catalysts. Accounts Chem. Res. 2016, 49 (4), 664-677.

2. Shortlan, A. J.; Wilkinson, G., Preparation, and Properties of Hexamethyltungsten. J. Chem. Soc. Dalton 1973, (8), 872-876.

3. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Rossini, A. J.; Widdifield, C. M.; Dey, R.; Emsley, L.; Basset, J. M., WMe6 Tamed by Silica: Si-O-WMe5 as an Efficient, Well-Defined Species for Alkane Metathesis, Leading to the Observation of a Supported W- Methyl/Methylidyne Species. J. Am. Chem. Soc. 2014, 136 (3), 1054-1061.

4. Samantaray, M. K.; Dey, R.; Abou-Hamad, E.; Hamieh, A.; Basset, J. M., Effect of Support on Metathesis of n-Decane: Drastic Improvement in Alkane Metathesis with WMe5 Linked to Silica-Alumina. Chem. Eur. J. 2015, 21 (16), 6100-6106.

5. Riache, N.; Callens, E.; Samantaray, M. K.; Kharbatia, N. M.; Atiqullah, M.; Basset, J. M., Cyclooctane Metathesis Catalyzed by Silica-Supported Tungsten Pentamethyl [(≡SiO)W(Me)5]: Distribution of Macrocyclic Alkanes. Chem. Eur. J. 2015, 21 (24), 8656-8656.

6. Samantaray, M. K.; Callens, E.; Abou-Hamad, E.; Dey, R.; Basset, J. M., Well-defined silica supported W-alkyl/alkylidyne precursor for the metathesis of propane. Abstr. Pap. Am. Chem. S 2014, 247.

7. Riache, N.; Dery, A.; Callens, E.; Poater, A.; Samantaray, M.; Dey, R.; Hong, J. H.; Lo, K.; Cavallo, L.; Basset, J. M., Silica-Supported Tungsten Carbynes ( SiO)xW(CH)(Me)y (x=1, y=2; x=2, y=1): New Efficient Catalysts for Alkyne Cyclotrimerization. Organometallics 2015, 34 (4), 690- 695.

8. Hamieh, A.; Chen, Y.; Abdel-Azeim, S.; Abou-hamad, E.; Goh, S.; Samantaray, M.; Dey, R.; Cavallo, L.; Basset, J. M., Well-Defined Surface Species [(≡Si-O)W(=O)Me3)] Prepared by Direct Methylation of [(≡Si-O-)W(=O)Cl3)], a Catalyst for Cycloalkane Metathesis and Transformation of Ethylene to Propylene. ACS. Catal. 2015, 5 (4), 2164-2171.

9. Herrmann, W. A.; Felixberger, J. K.; Anwander, R.; Herdtweck, E.; Kiprof, P.; Riede, J., Multiple Bonds between Transition-Metals and Main-Group Elements .73. Synthetic Routes to Rhenium(V) Alkyl and Rhenium(VII) Alkylidyne Complexes - X-Ray Crystal-Structures of (η-5- 5 C5Me5)Re(=O)(CH3)[CH2C(CH3)3] and (η -C5Me5)(Br)3Re=CC(CH3)3. Organometallics 1990, 9 (5), 1434-1443.

10. (a) Rietveld, M. H. P.; Lohner, P.; Nijkamp, M. G.; Grove, D. M.; Veldman, N.; Spek, A. L.; Pfeffer, M.; vanKoten, G., A novel synthetic route to tantalum-zinc neophylidyne complexes stabilized by ortho-chelating arylamine ligands; The x-ray structure of [TaCl2(µ- C6H4CH2NMe2)(µ-CCMe2Ph)ZnCl(THF)]. Chem. Eur. J. 1997, 3 (5), 817-822; (b) Abbenhuis, H. C. L.; Feiken, N.; Haarman, H. F.; Grove, D. M.; Horn, E.; Spek, A. L.; Pfeffer, M.; Vankoten, G., A

173

Bimetallic Tantalum Zinc Complex with an Ancillary Aryldiamine Ligand as Precursor for a Reactive Alkylidyne Species - Alkylidyne-Mediated C-H Activation and a Palladium-Mediated Alkylidyne Functionalization. Organometallics 1993, 12 (6), 2227-2235.

11. Maity, N.; Barman, S.; Callens, E.; Samantaray, M. K.; Abou-Hamad, E.; Minenkov, Y.; D'Elia, V.; Hoffman, A. S.; Widdifield, C. M.; Cavallo, L.; Gates, B. C.; Basset, J. M., Controlling the hydrogenolysis of silica-supported tungsten pentamethyl leads to a class of highly electron deficient partially alkylated metal hydrides. Chem. Sci. 2016, 7 (2), 1558-1568.

12. Driggers, E. M.; Hale, S. P.; Lee, J.; Terrett, N. K., The exploration of macrocycles for drug discovery - an underexploited structural class. Nat. Rev. Drug Discov. 2008, 7 (7), 608-624.

13. Zou, Y.; Mouhib, H.; Stahl, W.; Goeke, A.; Wang, Q. R.; Kraft, P., Efficient Macrocyclization by a Novel Oxy-Oxonia-Cope Reaction: Synthesis and Olfactory Properties of New Macrocyclic Musks. Chem. Eur. J. 2012, 18 (23), 7010-7015.

14. Dsouza, R. N.; Pischel, U.; Nau, W. M., Fluorescent Dyes and Their Supramolecular Host/Guest Complexes with Macrocycles in Aqueous Solution. Chem. Rev. 2011, 111 (12), 7941- 7980.

15. Voter, A. F.; Tillman, E. S., An Easy and Efficient Route to Macrocyclic Polymers Via Intramolecular Radical-Radical Coupling of Chain Ends. Macromolecules 2010, 43 (24), 10304- 10310.

16. Te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Guerra, C. F.; Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T., Chemistry with ADF. J. Comput. Chem. 2001, 22 (9), 931-967.

17. Perdew, J. P.; Burke, K.; Ernzerhof, M., Generalized gradient approximation made simple Phys. Rev. Lett. 1997, 78 (7), 1396-1396.

18. Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun, L. S.; Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L., Basis set exchange: A community database for computational sciences. J. Chem. Inf. Model 2007, 47 (3), 1045-1052.

19. Zhao, Y.; Truhlar, D. G., The M06 suite of density functionals for main group thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, and transition elements: two new functionals and systematic testing of four M06-class functionals and 12 other functionals. Theor. Chem. Acc. 2008, 120 (1-3), 215-241.

20. (a) van Lenthe, E.; Ehlers, A.; Baerends, E. J., Geometry optimizations in the zero order regular approximation for relativistic effects. J Chem Phys 1999, 110 (18), 8943-8953; (b) Vanlenthe, E.; Baerends, E. J.; Snijders, J. G., Relativistic Regular 2-Component Hamiltonians. J. Chem. Phys. 1993, 99 (6), 4597-4610.

174

21. VanLenthe, E.; Snijders, J. G.; Baerends, E. J., The zero-order regular approximation for relativistic effects: The effect of spin-orbit coupling in closed shell molecules. J. Chem. Phys. 1996, 105 (15), 6505-6516.

175

CHAPTER 5: Isolation of Silica Supported Niobium tetramethyl Complex by the SOMC Strategy: Synthesis, Characterization and Structure-Activity Relationship in Ethylene Oligomerization Reaction

176

5.1 Introduction

In the previous chapters, our studies was focusing on tungsten based compounds:

Dipp [(≡SiO-)W(=O)Me3], [(≡SiO-)W(=O)Me2Im N], [(≡SiO-)WMe5]. The main uses of those compounds were olefin and paraffin metathesis. In order to expand the scope of the new synthetic process, we started using our surface alkylation route to prepare group

(V) metal-alkyl compounds, mainly niobium based ones for two reasons:

1-Pentaalkyl niobium and tetraalkyl niobium halide compounds are less studied and not isolated yet in homogeneous phase,1-3 noting that some new potential applications could be discovered if we can stabilize the alkyl complexes of niobium on surface through SOMC approach.

2- Supported niobium alkyl complexes could play a pivotal role in the reactivity and selectivity of some catalytic reactions (polymerization, oligomerization, oxidation, etc.) which are performed by other group (V) metals (vanadium and tantalum).4-5

v Recently, we observed that silica supported tantalum compounds [(≡SiO-)Ta Cl2Me2] showed a high selectivity towards trimerization of ethylene into 1-hexene.6

Oligomerization of ethylene is a highly important reaction and activity could be enhanced by using niobium instead of tantalum as niobium is more acidic. In this chapter we aim to navigate through the chemistry of niobium alkyl compounds and its uses in ethylene oligomerization and other possible reactions.

177

5.1.1 Ethylene oligomerization: The catalytic oligomerization of ethylene to produce linear alpha olefins (LAOs) is one of the most commercially successful applications of catalysis in the petrochemical industry, with a production that exceeds a million ton annually.7-8 Usually, most of the catalysts provide a wide distribution of products primarily consisting of C14−C20 olefins. On the other hand, the high need of linear low-density polyethylene (LLDPE) has led to an increasing demand for short LAOs that will be used as a co-monomers in LLDPE production.8-11 Due to their high importance as monomers (1-butene, 1-hexene, and 1- octene), they gain an intense research interest in both academia and industry in order to develop new catalysts capable of selective dimerization, trimerization, and tetramerization of ethylene12-19. Some of the developed catalysts were industrialized as in the case of titanium-based catalysts that is used for the dimerization of ethylene

(Alpha-butol, IFP), chromium-based catalysts for the trimerization of ethylene (Phillips

Petroleum),19-20 or much recently the chromium-bearing PNP ligand for ethylene tetramerization (SASOL).21-22 Most of those classical systems require the use of co- catalysts such as MAO.

5.1.2 Ethylene oligomerization using SOMC-based catalyst Another possible catalyst for ethylene oligomerization was reported by Sen,16, 23 using a combination of TaCl5 with an alkylating agent (MeLi, SnMe4, AlMe3 or ZnMe2), the pre-

III catalyst in this case could be TaMe2Cl3 which generate Ta active species after its reduction by ethylene during the reaction, but it is worth to note that a co-catalyst is needed to proceed in this system. Our group started a study that aims to design and

178

isolate an efficient catalyst following the previous report of Sen. TaMe3Cl2 was prepared

o and grafted on SBA-15 dehroxylated at 700 C (SBA-15(700)). Analyzing the formed material by elemental analysis, solid state NMR, IR-spectroscopy and mass balance proved the structure as a silica supported tantalum dichloro dimethyl compound

V [(≡SiO)Ta Cl2Me2](Scheme 5.32). It was found that this complex is active for ethylene oligomerization without a co-catalyst or any other additives. On the other hand the activity was comparable to the homogenous system forming 1-hexene as a major product.

V Scheme 5.32: Synthesis of [(≡SiO)Ta Cl2Me2] (1). 5.1.3 Mechanism of ethylene oligomerization It is generally accepted that the mechanism for the selective ethylene transformation into α-olefins goes through a four center transition state. The two main steps in this mechanism are i) propagation through ethylene insertion and ii) the termination step, mainly by β-hydrogen abstraction that releases of the olefin. Thus, the olefin selectivity depends on those two steps, changing the metal or ligands of a catalyst will affect its ability to grow a chain and terminate it at a selective point. In the study of silica grafted tantalum, it was found that its exposure to dynamic flow of ethylene using a flow bed reactor release a mixture of methane, propylene, ethane and butane.

179

Scheme 5.33: Proposed mechanism pathways for the formation of butane, ethane, methane, propylene, and active TaIII surface species After a couple of hours it starts to dimerize ethylene into 1-butene. Then the reaction was terminated. The process probably starts with the coordination of ethylene; it can insert into the

Ta-Me bond to generate 3, which by a β-hydrogen elimination would produce first 4 and then 5 via a subsequent reductive elimination of methane (path c).25 As ethane and butane been

180 observed, which are presumably formed by reductive elimination from the intermediate 2 (path a) and 3 (path b) respectively, and which also leads to 5 (Scheme 5.33).

5.2 Results and Discussion

v 5.2.1. Synthesis and Characterization of [(≡SiO-)Nb Cl3Me] (1) In the present chapter, we will report the synthesis of silica supported tetramethyl niobium

v v complex [(≡SiO)Nb Me4] via the surface methylation of niobium monomethyl [(≡SiO-)Nb Cl3Me]

(1) SOMC complex. These surface niobium compounds were systematically characterized and structure was determined.

V Scheme 5.34: Synthesis of [(≡SiO)Nb Cl3Me] (1).

V Niobium complex 1 was prepared by the grafting of Nb Cl3Me2 on partially dehydoxylated silica (partial dehydroxylation was carried out at 700℃ under high

V vacuum to form SiO2-700). Nb Cl3Me2 was reacted with SiO2-700 at room temperature in dichloromethane or pentane for two hours under argon flow, followed by washing with pentane and drying by high vacuum (10−5 bar) in the absence of light, resulted 1 as a pale yellow powder (Scheme 5.34). The grafting reaction leads to the release of one equivalent of methane and the formation of a well-defined niobium complex [(≡SiO-

v )Nb Cl3Me] 1. The structure of 1 was confirmed precisely by the means of FTIR,

181

1 13 elemental analysis and solid state NMR ( H, C, DQ, TQ, and HETCOR). Infrared spectrum showed that the characteristic peak of free

−1 silanol in SiO2-700 at 3743 cm (Figure 5.51) completely disappeared (Figure1, yellow) as a sign of a complete grafting.

Figure 5.51: FTIR spectra of SiO2-700 (purple), complex 1 (yellow) and complex 2 (red).

182

Figure 5.52: a) 1H MAS NMR spectra of complex 1 b) 2D 1H−1H DQ/SQ and (c) 1H−1H TQ/SQ NMR spectra of complex 1 c) One-dimensional 13C CP/MAS NMR spectra of complex 1 d) 2D CP/MAS HETCOR NMR spectra of complex 1.

On the other hand, elemental analysis gave 2.69% Nb, 0.38% C, 2.87% Cl with a ratio of

v Nb/C/Cl=1/1.1/2.8 (Theoretical: Nb/C/Cl=1/1/3) with [(≡SiO-)Nb Cl3Me] represents the major product. Complex 1 was then characterized by solid state NMR. The 1H magic- angle spinning (MAS) NMR spectra of complex 1 exhibit one main signal at 2.5 ppm with a small shoulder at 2.0 ppm. All the signals auto-correlate in the two-dimensional (2D)

183

double-quantum (DQ) and triple-quantum (TQ) proving that they both belong to a CH3 group (Figure 5.52).

Similarly, 13C CP/MAS NMR spectrum of 1 displays only a single peak at 69 ppm. The 2D

HETCOR spectrum of 1 with short contact time (0.5 ms) shows a clear correlation between the main methyl proton and the carbon peak at 69 ppm, allowing the assignment of the carbon–proton pairs to the individual methyl groups.

v 5.2.2. Synthesis and Characterization of complex 2 [(≡SiO)Nb Me4] We then used the surface methylation strategy to methylate compound 1 [(≡SiO-

v v )Nb Cl3Me] with dimethylzinc aiming to form the full alkylated complex [(≡SiO-)Nb Me4].

v Complex 1 [(≡SiO-)Nb Cl3Me] was taken in dichloromethane and then reacted with 1.5 equivalents (compared to niobium count) of dimethylzinc (solution in pentane) at room temperature for an hour (Scheme 5.35). The product was washed up three time with pentane and then dried under high vacuum line for 30 min (<10-5bar) to form complex 2.

V V Scheme 5.35: Synthesis of [(≡SiO)Nb Me4] (2) by the methylation of [(≡SiO)Nb Cl3Me] (1).

The FTIR characteristic shows an increases in the range of 3000 cm-1, (C-H), and 1540 cm-1, δ(C-H) assigned to an increase in the amount of methyl fragments(Figure 5.51, red). Elemental analysis of complex 2 gave 2.59% Nb, 1.28% C, 1.32% Cl, 1.11% Zn with a ratio of Nb/C =1/3.8 (Theoretical: Nb/C =1/4). In addition, adding degassed water to 100 mg of complex 2 releases 0.114 mmol of methane (4.2CH4/Nb) after analyzing gas

184 samples by gas chromatography (GC). This can be explained by the presence of [(≡SiO-

v )Nb Me4] as a major species.

Figure 5.53: a) 1H MAS NMR spectra of complex 2 b) 2D 1H−1H DQ/SQ and (c) 1H−1H TQ/SQ NMR spectra of complex 2 c) One-dimensional 13C CP/MAS NMR spectra of complex 2 d) 2D CP/MAS HETCOR NMR spectra of complex 2.

v 1 Additionally, formulation of SOMF 2 [(≡SiO)Nb Me4] was elucidated by solid state NMR. The H- solid state NMR spectrum of 2 displays a main peak at 2.0 ppm along with a small peak at 0.3 ppm that could be methane (Figure 5.53) This peak shows a strong autocorrelation peak on the

185 diagonal of both 2D 1H-1H double and triple quantum (DQ, TQ) (Figure 5.53). Thus, the peak at

2.0 ppm can be assigned for a methyl group.

Besides, 13C cross polarization magic-angle spinning (CP-MAS) of 2 shows a single peak at 54 ppm, 1H-13C HETCOR spectrum of 2 shows a correlation between the carbon peak at 54 ppm and the proton peak at 2.0 ppm (Figure 5.53).

V Scheme 5.36: Synthesis of complex (2’) by using 5 equivalent of ZnMe2 to methylate complex [(≡SiO)Nb Cl3Me]

(1).

To understand the complex 2, the alkylation of the reaction was repeated with the use of higher amount of dimethylzinc (5 equivalents compared to Nb) (Scheme 5.36). After analyzing the formed product 2’, the proton peak at 2.0 and the carbon peak at 54 are still intact (Figure

5.54), as well as new proton and carbon peaks appears at 0.1 and -20 ppm as expected due to the formation of [(≡SiO-)ZnMe].

186

Figure 5.54: a) 1H MAS NMR spectra of complex 2’ b) 2D 1H−1H DQ/SQ and (c) 1H−1H TQ/SQ NMR spectra of complex 2’ c) One-dimensional 13C CP/MAS NMR spectra of complex 2’ d) 2D CP/MAS HETCOR NMR spectra of complex 2’. 5.2.3. Catalytic activity: ethylene oligomerization After systematic characterization, the reactivity of the surface niobium compounds was tested in ethylene oligomerization. Complex 1 or 2 was taken inside a special glass reactor in dry toluene, followed by a high pressure introduction of ethylene (at constant pressure of 50 bars) at 100℃. Using complex 1, oligomerization of ethylene into olefins ranging between C4 and C32 (with a difference of two carbons between two consecutive products) is observed (Figure 5.56-Figure 5.59).

187

v Table 5.19: Oligomerization of ethylene catalyzed with different batches of complex [(≡SiO)Nb Cl3Me] 1, complex v v [(≡SiO)Nb Me4] 2 and reported tantalum complex [(≡SiO)Ta Cl2Me2]

Complex(a) Butenes Hexenes Activity(b)

Selectivity (%) Selectivity (%)

v [(≡SiO)Nb Cl3Me] (batch 38.8 36.5 1150

v [(≡SiO)Nb Cl3Me] 40.1 34.5 1316

(batch 2)

v [(≡SiO)Nb Me4] 77.2 20.5 970

(batch 1)

v [(≡SiO)Nb Me4] 91.9 7.8 1050

(batch 2)

v [(≡SiO)Ta Cl2Me2] 9.6 82.7 375

Reaction conditions: [a] 100 mg of complex 1 or 2, 7 mL of toluene, 60 min. [b] Turnover number expressed in mol (product formed)/mol(Nb or Ta).

In this reaction C4 (38.8%) and C6(36.5%) share equally up to 76 % of the total amount of the formed oligomers. However, the same catalytic test was carried out with 2, a dramatic change in the selectivity was observed (Figure 5.60-Figure 5.63); Only butenes and hexenes are formed (Table 5.19). In contrast to previously reported tantalum

v complex [(≡SiO-)Ta Cl2Me2] that forms hexenes (>90%), complex 2 forms butenes selectively in the range between 80-92%. This difference of selectivity between 1 and 2 is likely due to different coordination sphere and their ability to easily abstract the β- hydrogen from the growing chain.

188

5.2.4. Catalytic Study: 1-octene isomerization We predicted that complex 1 would be able to generate a niobium hydride surface complex. Therefore, we applied this complex for isomerization of terminal olefins. We choose 1-octene as simple substrate as it is liquid in experimental condition. In a typical reaction, 2,000 equivalent of 1-octene was mixed with catalyst at room temperature in an ampule and reaction was monitored for 2 days. At the end of the reaction, the ampule was frozen with liquid nitrogen, the reaction mixture was quenched with dichloromethane, and the liquid fraction was filtered and analyzed by gas chromatography (GC). 1-octene was converted into mixture of its regio-isomers by isomerization process. After the reaction, a mixture of cis and trans, 2-, 3- and 4- octene

(80%) was obtained as a product along with 20% dimerized products and their isomers

(C16) with an overall turnover number (TON) of 2000 as determined by GC and GC-MS

(Figure 5.55). Nevertheless, catalyst 1 was found to be highly active for isomerization of

1-octene (it can reach 25000 TON) in comparison to its homogeneous counterpart

Nb(CH3)2Cl3 (TON=500). Finally we proposed a mechanism for the initiation and isomerization mechanism of complex 1 after contacting with1-octene (Scheme 5.37).

Initiation step was confirmed by the presence of stoichiometric amounts of C9 olefins as detected by GC & GC-MS.

189

Figure 5.55: Kinetic study of the isomerization and dimerization of 1-octene using complex 1 at room temperature (5 mg, 2µmol of Nb complex reacts with 5 mL of dried 1-octene in an ampoule inside the glovebox).

190

v Scheme 5.37: Proposed mechanism pathway for the isomerization of 1-octene with complex 1 [(≡SiO)Nb Cl3Me].

191

5.3 Conclusion

In conclusion, we have succeeded to prepare the first niobium tetramethyl complex on a

v support [(≡SiO-)Nb Me4] using SOMC-approach combined by surface alkylation tool.

v v Complex [(≡SiO-)Nb Me4] 2 was synthesized by alkylating [(≡SiO-)Nb Cl3Me] 1 with dimethylzinc. The silica supported complexes 1 and 2 were fully characterized by advance solid state NMR, FTIR, elemental analysis and mass balance. Finally, we revealed that complex 2 showed good activity and selectivity in dimerization of ethylene to 1-butene. On the other hand, complex 1 showed an ability to isomerize olefins and dimerize them at room temperature through a niobium hydride active species.

5.4 General Procedure

All experiments were performed by using standard Schlenk and glovebox techniques under an inert nitrogenous atmosphere. Syntheses and treatments of the surface species were completed using high-vacuum lines (<10−5 mbar) and glovebox techniques. All solvents were dried, degassed, and freshly distilled prior to use according to classical methods (over CaH2 for dichloromethane and pentane, and over Na/ benzophenone for toluene). Ethylene was dried and deoxygenated before use by passage through a mixture of freshly regenerated molecular sieves (3 Å) and R3-15 catalysts (BASF). IR spectra were recorded on a Nicolet 6700 FT-IR spectrometer using a DRIFT cell equipped with CaF2 windows The IR samples were prepared under argon within a glovebox. Typically, 64 scans were collected for each spectrum (resolution

= 4 cm−1). Elemental analysis were performed at Mikroanalytisches Labor Pascher (Germany) and KAUST core labs. Gas-phase analysis of alkanes was performed using an Agilent 6850 gas

192 chromatography column with a split injector coupled with a flame ionization detector (FID). An

HP-PLOT Al2O3 KCl 30 m × 0.53 mm, 20.00 mm, capillary column coated with a stationary phase of aluminum oxide deactivated with KCl was used with helium as the carrier gas at 32.1 kPa.

Each analysis was executed under the same conditions: a flow rate of 1.5 mL/ min and an isotherm at 80°C.

Solid-State Nuclear Magnetic Resonance Spectroscopy. One-dimensional 1H MAS and 13C

CP/MAS solid-state NMR spectra were recorded on Bruker AVANCE III spectrometers operating at 400 or 600 MHz resonance frequencies for 1H. The 400 MHz experiments employed a conventional double-resonance 4 mm CP/MAS probe, whereas experiments at 600 MHz utilized a 3.2 mm HCN triple resonance probe. In all cases the samples were packed into rotors under inert atmosphere inside gloveboxes. Dry nitrogen gas was utilized for sample spinning to prevent degradation of the samples. NMR chemical shifts are reported with respect to the external references TMS and adamantane. 13C CP/MAS NMR experiments used the following sequence: 90° pulse on the proton (pulse length = 2.4 s), then a cross-polarization step with a contact time of typically 2 ms, and finally acquisition of the 13C signal under high-power proton decoupling. The delay between the scans was set to 5 s to allow the complete relaxation of the

1H nuclei, and the number of scans ranged between 5 000 and 10 000 for 13C and was 32 for 1H.

An exponential apodization function corresponding to a line broadening of 80 Hz was applied prior to the Fourier transformations.

The 2D 1H−13C heteronuclear correlation (HETCOR) solid-state NMR spectroscopy experiments were conducted on a Bruker AVANCE III spectrometer operating at 600 MHz using a 3.2 mm

193

MAS probe. These experiments were performed according to the following scheme: 90° proton

13 13 pulse, t1 evolution period, CP to C, and detection of the C magnetization under TPPM decoupling. During the cross-polarization step, a ramped radio frequency (RF) field centered at

75 kHz was applied to the protons, whereas the 13C channel RF field was matched to obtain an optimal signal. A total of 32 t1 increments with 4000 scans each were collected; sample spinning frequency was 8.5 kHz. Using a short contact time (0.2 ms) for the CP step, the polarization transfer in the dipolar correlation experiment was verified to be selective for the first coordination sphere around tungsten, such that correlations occurred only between pairs of attached 1H−13C spins (C−H directly bonded).

1H−1H Multiple-Quantum Spectroscopy. Two-dimensional double-quantum (DQ) and triple- quantum (TQ) experiments were recorded on a Bruker AVANCE III spectrometer operating at

600 MHz with a conventional double-resonance 3.2 mm CP/MAS probe, according to the following general scheme: excitation of DQ coherences, t1 evolution, z-filter, and detection. The spectra were recorded in a rotor-synchronized fashion in t1 such that the t1 increment was set equal to one rotor period (45.45 μs). One cycle of the standard back-to-back (BABA) recoupling sequences was used for the excitation and reconversion period. Quadrature detection in w1 was achieved using the States-TPPI method. A spinning frequency of 22 kHz was used. The 90° proton pulse length was 2.5 μs, whereas a recycle delay of 5 s was used. A total of 128 t1 increments with 32 scans per increment were recorded. The DQ frequency in the w1 dimension corresponds to the sum of two single-quantum (SQ) frequencies of the two coupled protons and correlates in the w2 dimension with the two corresponding proton resonances. The TQ frequency in the w1 dimension corresponds to the sum of the three SQ frequencies of the

194

three-coupled protons and correlates in the w2 dimension with the three individual proton resonances. Conversely, groups of fewer than three equivalent spins will not give rise to diagonal signals in the spectrum.

5.5 Preparation procedure

Synthesis of Nb(CH3)2Cl3. Nb(CH3)2Cl3 was prepared according to the reported procedure.[4]

Inside a glovebox, and to a double schlenk containing a yellow suspension of NbCl5 (4.564 g,

16.87 mmol) in hexane (30 mL) was added Zn(CH3)2 (2.217 g, 21.84 mmol) at room temperature. After stirring for 14 hours at room temperature in the absent of light, the reaction mixture changed to a yellowish orange suspension. The reaction mixture was filtered through a celite plug supported by a medium porosity glass frit of the double schlenck, then transferred into another schlenk. The yellowish orange filtrate was evaporated to dryness to give

Nb(CH3)2Cl3 as yellowish orange solid. Yield = 64% (2.493 g, 10.85 mmol). Nb(CH3)2Cl3 was used

1 without further purification. H NMR (25°C, 76.8 MHz, CD2Cl2) 2.95 ppm

(13MeLi). A small amount of 13MeLi was taken and titrated with deoxygenated water to quantify the amount of methyl lithium present by quantifying the release of methane (around 55%

13MeLi was found). In another double Schlenk, 13MeLi (22 mmol, 484 mg) was added with 20 mL of pentane, and to that a solution of anhydrous ZnCl2 (10 mmol in 10 mL) was added at -20°C.

The reaction was allowed to reach at room temperature slowly and remained at that

195 temperature for another 2 days. The volatiles were then trapped into a schlenk, and the dimethylzinc was collected as pentane solution.

Preparation of [(≡SiO-)NbCl3CH3] . 230 mg NbMe2Cl3 in slight excess (1.1 equivalent) with respect to the amount of surface accessible silanols (0.3 mmol silanol groups per gram) was reacted with 3.0 g of SiO2-700 at room temperature in pentane for 2 h. After filtration and three washing cycles, all volatile compounds were condensed to quantify methane evolved during the grafting. And the solid was dried under dynamic vacuum (< 10-5 Torr, 2 h). The volatile components were analysed by GC and 0.28 ±0.1 mmol of CH4 per gram of SiO2-700 was determined. The reaction was monitored by FT-IR showing that 95% of isolated silanol disappeared with a concomitant appearance of weak bands at 2988-2875 cm-1 and 1500-1300 cm-1 region. Elemental analysis gave 2.69% Nb, 0.38% C, 2.87% Cl consistent with [(≡SiO-

V )Nb Cl3Me] 1. This was further confirmed by the hydrolysis of 1 at 298 K which produces 1.1 ±

0.1 mol of methane per mol Nb (expd: 1.0), and further analyses gave the ratio of

Nb/C/Cl=1/1.1/2.8 (Theoretical: Nb/C/Cl=1/1/3).

V Hydrolysis of solid [(≡SiO-)Nb Cl3Me] 1. 100 mg of 1 (29 μmol of metal) was transferred in a

Schlenk tube and H2O vapor (2 mL was added and reacted with 1 at room temperature for 4 h to give methane exclusively (34 μmol of methane was quantified by GC to 1.1 ±0.1 per niobium atom).

V Preparation of [(≡SiO)Nb(CH3)4] . To a suspension of 1.0 g of [(≡SiO)Nb Cl3Me] in dichloromethane or pentane 0.45 mL of dimethyl zinc solution in hexane (1 M) was added slowly to it in side an argon atmosphere. Reaction was kept for 2 hours at room temperature.

After filtration and three washing cycles the solid was dried under dynamic vacuum (<10-5 Torr,

196

2 h). The reaction was monitored by FT-IR shows an increase in the signals at 2988-2875 cm-1 and 1500-1300 cm-1 region. Elemental analysis of complex 2 gave 2.59% Nb, 1.28% C, 1.32% Cl,

1.11% Zn with a ratio of Nb/C =1/3.8 (Theoretical: Nb/C =1/4). This was further confirmed by the hydrolysis of 2 at 298°K which produces 4.2 ±0.1 mol of methane per mol Nb (expected:

4.0)

V Oligomerization of ethylene with [(≡SiO-)Nb Cl3Me] 1 & [(≡SiO-)Nb(CH3)4] 2 in semi-batch reactor. 100 mg of each catalyst in dry toluene (7 mL) were added into a glass vial equipped with a magnetic stirrer, and the glass vial was sealed in the glovebox. The vial was put into the autoclave of the ILS Premex parallel reactor system under pressure of 50 bar of ethylene at

100°C. The reaction was carried out for 60 min with vigorously stirring, and cooled to -10°C to condense all the products generated. An aliquot of the mixture was analyzed by GC to quantify the oligomers.

197

5.6 GC-Spectra after ethylene oligomerization using complex (1) & (2)

Toluene

1-butene

Hexenes

Figure 5.56: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #1 of v [(≡SiO-)Nb Cl3Me] after addition to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

Figure 5.57: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #2 of v [(≡SiO-)Nb Cl3Me] after addition to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

198

Figure 58: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #3 of v [(≡SiO-)Nb Cl3Me] after addition to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

Figure 5.59:GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #4 of v [(≡SiO-)Nb Cl3Me] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

199

Figure 5.60: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #1 of v [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

Figure 5.61: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #2 of v [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

200

Figure 5.62: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #3 of v [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

Figure 5.63: GC spectrum of the liquid phase products after a catalytic oligomerization reaction using 100 mg from batch #4 of v [(≡SiO-)Nb Me4] after adding to a semi-batch reactor, solubilized in 7 ml of toluene and feed by ethylene pressure of 50 bar at 100oC.

201

5.7. References 1. Schrock, R. R.; Meakin, P., Pentamethyl Complexes of Niobium and Tantalum. J. Am. Chem. Soc. 1974, 96 (16), 5288-5290.

2. Wu, Y. D.; Chan, K. W. K.; Xue, Z., Transition Structures of Methane Elimination in Pentamethylniobium and Pentamethyltantalum. J. Am. Chem. Soc. 1995, 117 (36), 9259-9264.

3. Schrock, R. R., Preparation and Characterization of M(CH3)5 (M=Nb or Ta) and Ta(CH2C6H5)5 and Evidence for Decomposition by Alpha-Hydrogen Atom Abstraction. J. Organomet. Chem. 1976, 122 (2), 209-225.

4. Parker, K. D. J.; Fryzuk, M. D., Synthesis, Structure, and Reactivity of Niobium and Tantalum Alkyne Complexes. Organometallics 2015, 34 (11), 2037-2047.

5. VenkatRamani, S.; Roland, C. D.; Zhang, J. G.; Ghiviriga, I.; Abboud, K. A.; Veige, A. S., Trianionic Pincer Complexes of Niobium and Tantalum as Precatalysts for ROMP of Norbornene. Organometallics 2016, 35 (16), 2675-2682.

6. Chen, Y.; Callens, E.; Abou-Hamad, E.; Merle, N.; White, A. J. P.; Taoufik, M.; Coperet, C.; Le Roux, E.; Basset, J. M., [(≡Si-O)(TaCl2Me2)Cl]: A Well-Defined Silica-Supported Tantalum(V) Surface Complex as Catalyst Precursor for the Selective Cocatalyst-Free Trimerization of Ethylene. Angew. Chem. Int. Ed. 2012, 51 (47), 11886-11889.

7. McGuinness, D. S., Olefin Oligomerization via Metallacycles: Dimerization, Trimerization, Tetramerization, and Beyond. Chem. Rev. 2011, 111 (3), 2321-2341.

8. Forestiere, A.; Olivier-Bourbigou, H.; Saussine, L., Oligomerization of Monoolefins by Homogeneous Catalysts. Oil Gas Sci. Technol. 2009, 64 (6), 649-667.

9. Speiser, F.; Braunstein, P.; Saussine, W., Catalytic ethylene dimerization and oligomerization: Recent developments with nickel complexes containing P,N-chelating ligands. Accounts Chem. Res. 2005, 38 (10), 784-793.

10. Mukherjee, S.; Patel, B. A.; Bhaduri, S., Selective Ethylene Oligomerization with Nickel Oxime Complexes. Organometallics 2009, 28 (10), 3074-3078.

11. Killian, C. M.; Johnson, L. K.; Brookhart, M., Preparation of linear alpha-olefins using cationic nickel(II) alpha-diimine catalysts. Organometallics 1997, 16 (10), 2005-2007.

12. Emrich, R.; Heinemann, O.; Jolly, P. W.; Kruger, C.; Verhovnik, G. P. J., The role of metallacycles in the chromium-catalyzed trimerization of ethylene. Organometallics 1997, 16 (8), 1511-1513.

13. Manyik, R. M.; Walker, W. E.; Wilson, T. P., Soluble Chromium-Based Catalyst for Ethylene Trimerization and Polymerization. J. Catal. 1977, 47 (2), 197-209.

202

14. Kohn, R. D.; Haufe, M.; Kociok-Kohn, G.; Grimm, S.; Wasserscheid, P.; Keim, W., Selective trimerization of alpha-olefins with triazacyclohexane complexes of chromium as catalysts. Angew. Chem. Int. Ed. 2000, 39 (23), 4337.

15. Schrock, R.; Mclain, S.; Sancho, J., Tantalacyclopentane Complexes and Their Role in the Catalytic Dimerization of Olefins. Pure Appl. Chem. 1980, 52 (3), 729-732.

16. Andes, C.; Harkins, S. B.; Murtuza, S.; Oyler, K.; Sen, A., New tantalum-based catalyst system for the selective trimerization of ethene to 1-hexene. J. Am. Chem. Soc. 2001, 123 (30), 7423-7424.

17. Deckers, P. J. W.; Hessen, B.; Teuben, J. H., Catalytic trimerization of ethene with highly active cyclopentadienyl-arene titanium catalysts. Organometallics 2002, 21 (23), 5122-5135.

18. Blok, A. N. J.; Budzelaar, P. H. M.; Gal, A. W., Mechanism of ethene trimerization at an ansa-(arene)(cyclopentadienyl) titanium fragment. Organometallics 2003, 22 (13), 2564-2570.

19. Dixon, J. T.; Green, M. J.; Hess, F. M.; Morgan, D. H., Advances in selective ethylene trimerisation - a critical overview. J. Organomet. Chem. 2004, 689 (23), 3641-3668.

20. Carter, A.; Cohen, S. A.; Cooley, N. A.; Murphy, A.; Scutt, J.; Wass, D. F., High activity ethylene trimerisation catalysts based on diphosphine ligands. Chem. Commun. 2002, (8), 858- 859.

21. Bollmann, A.; Blann, K.; Dixon, J. T.; Hess, F. M.; Killian, E.; Maumela, H.; McGuinness, D. S.; Morgan, D. H.; Neveling, A.; Otto, S.; Overett, M.; Slawin, A. M. Z.; Wasserscheid, P.; Kuhlmann, S., Ethylene tetramerization: A new route to produce 1-octene in exceptionally high selectivities. J. Am. Chem. Soc. 2004, 126 (45), 14712-14713.

22. McGuinness, D. S.; Rucklidge, A. J.; Tooze, R. P.; Slawin, A. M. Z., Cocatalyst influence in selective oligomerization: Effect on activity, catalyst stability, and 1-hexene/1-octene selectivity in the ethylene trimerization and tetramerization reaction. Organometallics 2007, 26 (10), 2561-2569.

23. Yu, Z. X.; Houk, K. N., Why trimerization? Computational elucidation of the origin of selective trimerization of ethene catalyzed by [TaCl3(CH3)2] and an agostic-assisted hydride transfer mechanism. Angew. Chem. Int. Ed. 2003, 42 (7), 808-811.

203

CAPTER 6: Conclusions and Outlooks

204

6.1. Achievements

Surface organometallic compounds (SOMCs) shows an ability to bridge the gap between homogeneous and heterogeneous catalysts by forming a well-defined and single site heterogeneous complexes. Those complexes are able to catalyze various reactions for instance alkane metathesis, olefin metathesis, direct ethylene transformation into propene, ethylene oligomerization without co-catalyst and many others and in all these cases the path depends on the metal and its coordinating ligands,. The classical way to synthesize surface organometallic complexes path through classical homogeneous synthesis of organometallic precursors followed by grafting on a support, thus limit their use to well-equipped organometallic labs.

This thesis described a new route to prepare some surface organometallic complexes, this route is easier, faster, cheaper and safer than classical synthesis. The key element of this synthetic approach depends on alkylating surface supported metal halide fragments. The outcome of this study leads to new and unprecedented catalysts. Starting by chapter 2 that discussed the synthesis of the monopodal [(≡Si-O-)W(=O)Me3] by grafting the commercially available WOCl4 onto SiO2-700. The resultant [(≡Si-O-)W(=O)Cl3] was then methylated by ZnMe2 in diethyl ether solution to produce the single site complex [(≡Si-O-)W(=O)Me3]. This complex was then fully characterized by SS-NMR, IR, microanalysis and mass balance techniques. Depending on the results, two different conformations of [(≡Si-O-)W(=O)Me3] were proposed depending on the relative orientation of the oxo group. This was further proved by experimental and DFT- calculated NMR. This complex shows ability in producing valuable alkane macrocycles by the metathesis of cyclooctane. On the other hand, the tungsten oxo-trimethyl catalyst was used to transform ethylene into propylene, the activity and selectivity of the complex in this reaction

205 was superior to that of tungsten pentamethyl and tungsten hydride supported onto silica, perhaps as a result of the oxo ligand . On the other side, changing one of the methyl ligands by

Dipp imidazoline-2-iminato ligand leads to the formation of [(≡Si-O-)W(=O)(CH3)2-Im N] that was described in chapter 3. The synthesis started by the preparation and isolation of tungsten trichloro imidizaoline-2-iminato, followed by its grafting on SiO2-700 and finalized by alkylation of

Dipp [(≡Si-O-)W(=O)(Cl)2-Im N]. This complex was fully characterized by SS-NMR, IR, microanalysis, mass balance and EXAFS study. It was found that this complex shows a better selectivity in the metathesis of 1-hexene by forming higher percentage of 5-decene when comparing to the silica supported tungsten oxo trimethyl complex [(≡Si-O-)W(=O)(CH3)3] more probably due to the influence of the imidazoline-2-iminato ligand bulkiness. Similarly, in chapter 5 we succeeded to prepare the first grafted niobium tetramethyl complex [(≡Si-O-)Nb(CH3)4] by using surface alkylation approach, without being limited to the ability of isolating the homogeneous niobium penatmethyl precursor. This opened the door for a catalyst that can perform ethylene oligomerization and dimerization. In other cases, the new surface alkylation approach allows an easier, cheaper and scalable synthesis of already known complexes such as silica supported tungsten pentamethyl complex [(≡SiO-)WMe5] by the surface methylation of [(≡SiO-)WCl5]. It is important to note that alkylation step was the most challenging step in this approach, and the required reaction conditions varies depending on the metal-halide complex and its surrounding ligands. For instance using diethyl ether was crucial to prepare a well-defined silica supported tungsten oxo methyl complexes [(≡SiO-)W(=O)Me3], while in the case of preparing [(≡SiO-

)WMe5] it leads to the formation of more undesired product on the surface. In addition, in chapter 4 we propose a better understanding to the alkylation step through a kinetic study

206

(theoretical and experimental) this allows us to figure out the strong and weak points of this synthesis path and allowing a better understanding of the factors controlling the surface alkylation mechanism.

6.2 Future Outlook

The new easy synthetic approach facilitates the production of surface organometallic compounds, but as a side effect, less well-defined systems were produced. If we take the synthesis of silica supported tungsten pentamethyl [(≡Si-O-)W(Me)5] as an example, we can find that another species co-exist on the surface. A study was established and we found that in this particular case that some the complex decomposes in the presence of the alkylating agent itself. Another Side effect is the by-product generated from the alkylation reaction, in which some stay attached to the surface or even to the surface complexes. Here many factors could affect the final product, the support, the metal-halide complex it-self, the alkylating agent, the generated salt and the energy released during the reaction. Usually in homogeneous alkylation by-products are formed, but there is ability to separate them from the aimed complex through washing, filtration, crystallization or even sublimation. While on the support, we lack this ability because the formed by-products are chemically bonded with the surface. In addition, characterization of the formed complexes is challenging, especially for new complexes due to the absence of proper references. More work has to be done for better understanding of the alkylation process to reach a point that this strategy could replace the classical way of synthesizing surface metal alkyl complexes. Ultimately, this will lead to more use of this highly important type of catalysts (SOMCs).