Development of Tailored Homogeneous Ruthenium Catalysts for the Application in the Hydrogenation of Biogenic Substrates

Entwicklung maßgeschneiderter homogener Ruthenium Katalysatoren für die Anwendung in der Hydrierung biogener Substrate

Von der Fakultät für Mathematik, Informatik und Naturwissenschaften der RWTH Aachen University zur Erlangung des akademischen Grades einer Doktorin der Naturwissenschaften genehmigte Dissertation

vorgelegt von

Celine Laura Jung M.Sc. RWTH

aus Püttlingen

Berichter: Prof. Dr. rer. nat. Jürgen Klankermayer Prof. Dr.-Ing. Andreas Jupke

Tag der mündlichen Prüfung: 17.06.2020

Diese Dissertation ist auf den Internetseiten der Universitätsbibliothek verfügbar.

Die experimentellen Arbeiten zur vorliegenden Dissertation wurden im Zeitraum von Januar 2016 bis November 2019 unter der wissenschaftlichen Anleitung von Prof. Dr. rer. nat. Jürgen Klankermayer am Institut für Technische und Makromolekulare Chemie (ITMC) der RWTH Aachen University angefertigt.

Eidesstattliche Erklärung

Ich, Celine Laura Jung erkläre hiermit, dass diese Dissertation und die darin dargelegten Inhalte die eigenen sind und selbstständig, als Ergebnis der eigenen originären Forschung, generiert wurden.

Hiermit erkläre ich an Eides statt

1. Diese Arbeit wurde vollständig oder größtenteils in der Phase als Doktorand dieser Fakultät und Universität angefertigt; 2. Sofern irgendein Bestandteil dieser Dissertation zuvor für einen akademischen Abschluss oder eine andere Qualifikation an dieser oder einer anderen Institution verwendet wurde, wurde dies klar angezeigt; 3. Wenn immer andere eigene- oder Veröffentlichungen Dritter herangezogen wurden, wurden diese klar benannt; 4. Wenn aus anderen eigenen- oder Veröffentlichungen Dritter zitiert wurde, wurde stets die Quelle hierfür angegeben. Diese Dissertation ist vollständig meine eigene Arbeit, mit der Ausnahme solcher Zitate; 5. Alle wesentlichen Quellen von Unterstützung wurden benannt; 6. Wenn immer ein Teil dieser Dissertation auf der Zusammenarbeit mit anderen basiert, wurde von mir klar gekennzeichnet, was von anderen und was von mir selbst erarbeitet wurde; 7. Ein Teil oder Teile dieser Arbeit wurden zuvor veröffentlicht und zwar in:  Gordon Research Seminar and Gordon Research Conference 2018 Green Chemistry, Castelldefels, Spanien, 28. Juli–3. August 2018. Posterpräsentation “Novel Ruthenium Triphos Catalysts for the Hydrogenation of Methyl-γ-butyrolactones to 2-Methyl- butane-1,4-diol”, C. L. Jung, J. Klankermayer.  6th International Conference of the Cluster of Excellence “Tailor-Made Fuels from Biomass”, Aachen, Deutschland, 19.–21. Juni 2018. Posterpräsentation “Novel Ru- thenium Triphos Catalysts for the Hydrogenation of Itaconic Acid and Methyl-γ-butyrolactones”, C. L. Jung, A. J. Schmitz, J. Klankermayer.

Datum, Unterschrift

Abstract

The present thesis describes the development of tailored molecular ruthenium catalyst systems for the selective hydrogenation of biogenic platform chemicals. Chapter 1 provides an introduction on the state of the art in the synthesis of the two bio-based compounds levulinic and itaconic acid and their potential as renewable building blocks for future chemical value chains. Complementary, a short overview of homogeneous catalytic systems describes the challenges linked to the hydrogenation of acids and esters with special focus on ruthenium triphos systems. In Chapter 2, the development of a catalyst lead structure is explained, enabling the tailoring of molecular complexes for the catalytic application. In detail, the preparation of a series of ruthenium triphos-xyl complexes bearing acetato, naphtholato or phenolato ligands from [Ru(triphos-xyl)(tmm)] is described and completed with the characterization of the complex molecules.

The subsequent results of the catalytic hydrogenation of levulinic acid and γ-valerolactone to 1,4-pentanediol by selected ruthenium triphos-xyl complexes are reported in Chapter 3. High activities at low catalyst concentrations could be observed in batch experiments and multiple substrate reloading cycles revealed a correlation of catalyst deactivation with thermal stress in the absence of substrate. Moreover, the successful dehydrogenation of 1,4-pentanediol to γ-valerolactone showed the presence of a dynamic lactone-diol equilibrium. NMR-spectroscopic investigations revealed the formation of a bis(levulinato) complex on dissolving the catalysts in levulinic acid prior to the hydrogenation.

The challenging hydrogenation of itaconic acid, itaconates and 2-methyl-γ-butyrolactone to 2-methyl-1,4-butanediol with adapted ruthenium triphos-xyl complexes is reported in Chapter 4. Selected fluorinated phenolato and naphtholato ruthenium triphos-xyl catalyst systems showed significantly improved activities compared to established systems. Besides the fluorinated phenols, also hexafluorobenzene could be identified as beneficial additive leading to a closer investigation on its reactivity with [Ru(triphos-xyl)(tmm)] in Chapter 5. As a result, the dimeric fluorine-bridged complex [{Ru(triphos-xyl)}2(µ-F)3]+ could be isolated and successfully applied in the hydrogenation of selected acids, esters and amides. In particular, it was found to be active at very low catalyst concentrations in the hydrogenation of levulinic acid and γ-valerolactone. Moreover, an unprecedented complex with coordinated PF6− anion to the ruthenium center could be developed and characterized.

iii

Zusammenfassung

Die vorliegende Arbeit beschreibt die Entwicklung maßgeschneiderter molekularer Ruthenium Katalysatoren für die selektive Hydrierung biogener Plattformchemikalien. In Kapitel 1 erfolgt eine Einführung in den aktuellen Stand der Herstellung der beiden biobasierten Verbindungen Levulinsäure und Itaconsäure und deren Potential als erneuerbare Bausteine für zukünftige chemische Wertschöpfungsketten. Ergänzend beschreibt ein kurzer Überblick über homogene Katalysatorsysteme, mit besonderem Fokus auf Ruthenium Triphos Systemen, die mit der Hydrierung von Säuren und Estern einhergehenden Herausforderungen. In Kapitel 2 wird die Entwicklung einer Katalysator-Leitstruktur beschrieben, welche das Maßschneidern von molekularen Komplexen für die Anwendung als Katalysatoren ermöglicht. Im Detail wird die Darstellung einer Reihe von Ruthenium Triphos-xyl Komplexen mit Acetato-, Naphtholato- oder Phenolatoliganden ausgehend von [Ru(triphos-xyl)(tmm)] beschrieben und mit der Charakterisierung der Komplexmoleküle komplettiert.

Die folgenden Ergebnisse der katalytischen Hydrierung von Levulinsäure und γ-Valerolacton zu 1,4-Pentandiol mit ausgewählten Ruthenium Triphos-xyl Komplexen werden in Kapitel 3 berichtet. In Batch-Experimenten konnten hohe Aktivitäten bei niedriger Katalysatorkonzentration beobachtet werden und in mehreren Substratbeladungs-Zyklen wurde eine Korrelation zwischen der Katalysatordeaktivierung und der thermischen Belastung in Abwesenheit von Substrat beobachtet. Zudem zeigte die erfolgreiche Dehydrierung von 1,4-Pentandiol zu γ-Valerolacton das Auftreten eines dynamischen Lacton-Diol Gleichgewichts. NMR-spektroskopische Untersuchungen offenbarten die Bildung einer bis(levulinato) Spezies, die durch Lösen der Katalysatoren in Levulinsäure vor der Hydrierung gebildet wird. Die anspruchsvolle Hydrierung von Itaconsäure, Itaconaten und 2-Methyl-γ-butyrolacton zu 2-Methyl-1,4-butandiol mit geeigneten Ruthenium Triphos-xyl Komplexen wird in Kapitel 4 berichtet. Ausgewählte fluorierte Phenolato- und Naphtholato Ruthenium Triphos-xyl Katalysatoren zeigten signifikant verbesserte Aktivitäten im Vergleich zu etablierten Systemen. Neben den fluorierten Phenolen wurde auch Hexafluorobenzol als vorteilhaftes Additiv identifiziert, was zu einer näheren Betrachtung seiner Reaktivität mit [Ru(triphos-xyl)(tmm)] in Kapitel 5 führte. In Folge konnte das Fluor-verbrückte Dimer

[{Ru(triphos-xyl)}2(µ-F)3]+ isoliert und erfolgreich in der Hydrierung ausgewählter Säuren, Ester und Amide eingesetzt werden. Im Besonderen konnte seine Aktivität in sehr niedrigen Katalysatorkonzentrationen in der Hydrierung von Levulinsäure und γ-Valerolacton genutzt werden. Zudem konnte ein neuartiger Komplex mit einem an das Rutheniumzentrum koordinierten

PF6− Anion entwickelt und charakterisiert werden. iv

Table of Contents

Eidesstattliche Erklärung ...... ii Abstract ...... iii Zusammenfassung ...... iv Table of Contents ...... v List of Abbreviations ...... vii List of Displayed Substances ...... x List of Utilized or Observed Complexes / Catalytic Systems ...... xiii 1 Introduction and Motivation ...... 1 1.1 Bio-Based Carboxylic Acids as Renewable Platform Chemicals for Chemical Building Blocks and Fuels ...... 1 1.1.1 Levulinic Acid – Production from Renewable Resources and Utilization as Starting Material in the Chemical Value Chain...... 3 1.1.2 Itaconic Acid – Production and Application ...... 5 1.2 Homogeneous Catalytic Systems for the Hydrogenation of Carboxylic Acids and Esters ...... 8 1.2.1 Development of Ruthenium Triphos Based Catalysts for Selective Hydrogenation Reactions ...... 8 1.2.2 Mechanistic Insights on the Hydrogenation of Carboxylic Acids and Esters with Ruthenium Triphos Catalysts ...... 11 1.3 Motivation and Aim ...... 13 2 Development of Ruthenium Triphos-xyl Based Complexes ...... 14 3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes ...... 25 3.1 Hydrogenation of Levulinic Acid and γ-Valerolactone – State of the Art ...... 25 3.2 Application of Novel Catalytic Systems in the Hydrogenation of γ-Valerolactone and Levulinic Acid ...... 27 3.3 Mechanistic Insights into the Hydrogenation of γ-Valerolactone and Levulinic Acid using Ruthenium Triphos-xyl Complexes ...... 51 4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes ...... 56 4.1 Hydrogenation of Itaconic Acid – State of the Art ...... 56 4.2 Application of Novel Catalytic Systems in the Hydrogenation of 2-Methyl-γ-butyrolactone, Itaconic Acid and Itaconates ...... 57 4.3 Mechanistic Insights into the Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Ruthenium Triphos-xyl Complexes ...... 76 5 Ruthenium Triphos-xyl Based Complexes Derived from Perhalogenatobenzenes ...... 84 5.1 Reactivity of Ru(triphos-xyl)(tmm) with Perhalogenatobenzenes ...... 84 5.2 Hydrogenation of Acids, Esters and Amides with a Trifluorine Bridged Dimeric Ruthenium Triphos-xyl Complex...... 96 6 Summary ...... 101 7 Experimental Section...... 104 7.1 General Experimental Procedures ...... 104 7.2 Analytical Methods...... 106 7.3 Synthetic Procedures ...... 111 7.3.1 Synthesis of Bis(3,5-dimethylphenyl)phosphine oxide ...... 111 7.3.2 Synthesis of Bis(3,5-dimethylphenyl)phosphine ...... 112 7.3.3 Synthesis of 1,1,1-Tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane ...... 113 7.3.4 Synthesis of Trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 114 7.3.5 Synthesis of Trimethylenemethane-1,1,1-tris(diphenylphosphinomethyl)propaneruthenium(II) ...... 115 7.3.6 Synthesis of Trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)propaneruthenium(II) ...... 116 7.3.7 Synthesis of Carbonyldihydrido-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 117 7.3.8 Synthesis of Bis(levulinato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 118

Table of Contents

7.3.9 Synthesis of Bis(acetato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 119 7.3.10 Synthesis of Bis(trifluoroacetato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 120 7.3.11 Synthesis of Bis(1-naphtholato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 121 7.3.12 Synthesis of Bis(2-naphtholato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 122 7.3.13 Synthesis of Naphthalene-1,8-bis(olato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 123 7.3.14 Synthesis of Naphthalene-2,3-bis(olato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 124 7.3.15 Synthesis of Bis(2-ortho-phenylolphenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 125 7.3.16 Synthesis of Bis(2-fluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 126 7.3.17 Synthesis of Bis(2,5-difluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)...... 127 7.3.18 Synthesis of Bis(2,6-difluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)...... 128 7.3.19 Synthesis of Bis(3,5-difluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)...... 129 7.3.20 Synthesis of Bis(2,4,6-trifluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 130 7.3.21 Synthesis of Bis(3,4,5-trifluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 131 7.3.22 Synthesis of Bis(2,3,5,6-tetrafluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 132 7.3.23 Synthesis of Bis(2,3,4,5,6-pentafluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 133 7.3.24 Synthesis of Bis(1,3,4,5,6,7,8-heptafluoronaphtholato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 134 7.3.25 Synthesis of Bis(3,5-dichlorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 136 7.3.26 Synthesis of Bis(3,5-dibromophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 137 7.3.27 Synthesis of Bis(3,5-bis(trifluoromethyl)phenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 138 7.3.28 Synthesis of Tri-µ-fluorido-bis(1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane)diruthenium(II) ...... 139 7.3.29 Synthesis of η3-hexafluorophosphosphato-(1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ruthenium ...... 141 7.3.30 Synthesis of Dichlorido-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 142 7.3.31 Synthesis of Dibromido-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) ...... 144 7.3.32 Synthesis of Tri-µ-bromido-bis(1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane)diruthenium(II) bromide ...... 145 7.3.33 Synthesis of Heptafluoro-2-naphthol ...... 147 7.4 Procedures of Catalytic Experiments ...... 148 7.4.1 General Procedure for the Hydrogenation of Levulinic Acid, γ-Valerolactone and α-Methyl-γ-butyrolactone with Air-sensitive Catalytic Systems...... 148 7.4.2 General Procedure for the Hydrogenation of Itaconic Acid with Air-sensitive Catalytic Systems ...... 148 7.4.3 General Procedure for the Hydrogenation of Itaconates with Air-sensitive Catalytic Systems ...... 148 7.4.4 General Procedure for the Hydrogenation of High-boiling Substrates with Air-stable Catalytic Systems ...... 149 7.4.5 General Procedure for the Hydrogenation of Low-boiling Substrates with Air-stable Catalytic Systems ...... 149 7.4.6 General Procedure for the Dehydrogenation of Diols ...... 149 8 Appendix ...... 150 8.1 SC-XRD Data ...... 150 9 Bibliography ...... 157 Danksagung ...... 165

List of Abbreviations

2-ortho-pp 2-ortho-phenylolphenolato A. Aspergillus AA acetanilide ABBt Aachen Biology and Biotechnology Ac acetyl acac acetylacetonato AN aniline APT attached proton test BA benzoic acid BAL benzyl alcohol BAM benzamide BHT butylated hydroxytoluene (2,6-di-tert-butyl-4-methylphenol) BL butyl levulinate BP bromophenolato Bu butyl BuOH butanol cat catalyst cod cyclooctadiene CP chlorophenolato DBI dibutyl itaconate DCM dichloromethane DEAN diethyl aniline DFT density functional theory DIBAL-H diisobutylaluminium hydride DMI dimethyl itaconate DMSO dimethyl sulfoxide DOE Department of Energy EAN ethyl aniline EL ethyl levulinate EPR electron paramagnetic resonance ESI electron spray ionization est. estimated Et ethyl EtOH ethanol eq equivalents

vii

List of Abbreviations

FAB fast atom bombardment FDCA 2,5-furan dicarboxylic acid FID flame ionization detector FP fluorophenolato GC gas chromatography GVL γ-valerolactone hf-2-naph heptafluoro-2-naphtholato HPA hydroxypropionic acid HR high resolution IA itaconic acid IR infrared Institut für Technische und Makromolekulare Chemie (engl.: Institute of ITMC Technical and Macromolecular Chemistry) LVA levulinic acid MBDO methylbutane-1,4-diol Me methyl MeOH methanol MF methylfuran MGBL methyl-γ-butyrolactone ML methyl levulinate MS mass spectrometry MSA methyl succinic acid MTHF methyl tetrahydrofuran MV methyl valerate n-1,8-bo naphthalene-1,8-bis(olato) n-2,3-bo naphthalene-2,3-bis(olato) naph naphtholato NMR nuclear magnetic resonance OTf triflate PDO pentanediol Pe pentyl PeOH pentanol PMMA poly(methyl methacrylate) Pr propyl S solvent/substrate STR stirred tank reactor THF tetrahydrofuran

viii

List of Abbreviations tmm trimethylenemethane tol tolyl triFMP trifluoromethylphenolato triphos 1,1,1-tris(diphenylphosphinomethyl)ethane triphos-xyl 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane xyl 3,5-dimethylphenyl

List of Displayed Substances

1 succinic acid 2 fumaric acid 3 malic acid 4 2,5-furandicarboxylic acid (FDCA) 5 3-hydroxypropionic acid (3-HPA) 6 aspartic acid 7 glucaric acid 8 glutamic acid 9 levulinic acid (LVA) 10 itaconic acid (IA) 11 3-hydroxybutyrolactone 12 glycerol 13 sorbitol 14 xylitol 15 arabinitol 16 γ-valerolactone (GVL) 17 1,4-pentanediol (1,4-PDO) 18 2-methyltetrahydrofuran (2-MTHF) 19 diphenolic acid 20 δ-aminolevulinic acid (δ-ALA) 21 methyl succinic acid (MSA) 22 2-methyl-γ-butyrolactone (2-MGBL) 23 3-methyl-γ-butyrolactone (3-MGBL) 24 2-methyl-1,4-butanediol (2-MBDO) 25 3-methyltetrahydrofuran (3-MTHF) 26 methacrylic acid 27 itaconic anhydride 28 1,3,5-trihydroxybenzene (phloroglucinol) 29 8-hydroxyquinoline 30 2,6-di-tert-butyl-4-methylphenol (BHT) 31 2,6-dimethoxyphenol (syringol) 32 phenol 33 menthol 34 1,2,4-trihydroxybenzene (hydroxyhydroquinone) 35 1,2,3-trihydroxybenzene (pyrogallol)

List of Displayed Substances

36 3,5-dimethoxyphenol 37 5,6,7,8-tetrahydro-1-naphthol 38 5,6,7,8-tetrahydro-2-naphthol 39 1,3-dihydroxynaphthalene 40 4-fluorophenol 41 3,5-diiodophenol 42 2,2’-methylenediphenol 43 1,1’-bi-2-naphthol 44 acetic acid 45 trifluoroacetic acid 46 1-naphthol 47 2-naphthol 48 1,8-dihydroxynaphthalene 49 2,3-dihydroxynaphthalene 50 2,2’-biphenol 51 2-fluorophenol 52 2,5-difluorophenol 53 2,6-difluorophenol 54 3,5-difluorophenol 55 2,4,6-trifluorophenol 56 3,4,5-trifluorophenol 57 2,3,5,6-tetrafluorophenol 58 2,3,4,5,6-pentafluorophenol 59 heptafluoro-2-naphthol 60 3,5-dichlorophenol 61 3,5-dibromophenol 62 3,5-bis(trifluoromethyl)phenol 63 hexafluorobenzene 64 dimethyl itaconate 65 dibutyl itaconate 66 hexachlorobenzene 67 hexabromobenzene 68 methyl levulinate 69 ethyl levulinate 70 butyl levulinate 71 methyl valerate 72 1-pentanol

List of Displayed Substances

73 acetanilide 74 aniline 75 ethylaniline 76 diethylaniline 77 benzoic acid 78 benzamide 79 methyl benzoate 80 benzyl alcohol

xii

List of Utilized or Observed Complexes / Catalytic Systems cat-I-acac [Ru(acac)3]/triphos

+ cat-I-CO2H [Ru(triphos)(CO)2H] cat-I-COCl2 [Ru(triphos)(CO)Cl2] cat-I-COH2 [Ru(triphos)(CO)H2] cat-I-H2 [{Ru(triphos)}2(µ-H)2] cat-I-H2H2 [{Ru(triphos)H}2(µ-H)2]

+ cat-I-H3 [{Ru(triphos)}2(µ-H)3] cat-II-44 [Ru(triphos-xyl)(OOCCH3)2] cat-II-45 [Ru(triphos-xyl)(OOCCF3)2] cat-II-46 [Ru(triphos-xyl)(1-naph)2] cat-II-47 [Ru(triphos-xyl)(2-naph)2] cat-II-48 [Ru(triphos-xyl)(n-1,8-bo)] cat-II-49 [Ru(triphos-xyl)(n-2,3-bo)] cat-II-50 [Ru(triphos-xyl)(2-ortho-pp)2] cat-II-51 [Ru(triphos-xyl)(2-FP)2] cat-II-52 [Ru(triphos-xyl)(2,5-diFP)2] cat-II-53 [Ru(triphos-xyl)(2,6-diFP)2] cat-II-54 [Ru(triphos-xyl)(3,5-diFP)2] cat-II-55 [Ru(triphos-xyl)(2,4,6-triFP)2] cat-II-56 [Ru(triphos-xyl)(3,4,5-triFP)2] cat-II-57 [Ru(triphos-xyl)(2,3,5,6-tetraFP)2] cat-II-58 [Ru(triphos-xyl)(pentaFP)2] cat-II-59 [Ru(triphos-xyl)(hf-2-naph)2] cat-II-60 [Ru(triphos-xyl)(3,5-diCP)2] cat-II-61 [Ru(triphos-xyl)(3,5-diBP)2] cat-II-62 [Ru(triphos-xyl)(3,5-bis(triFMP)2]

+ cat-II-63a [{Ru(triphos-xyl)}2(µ-F)3]

3 cat-II-63b [Ru(triphos-xyl)(η -PF6)] cat-II-66a [Ru(triphos-xyl)Cl2] cat-II-66b [{Ru(triphos-xyl)}2(µ-Cl)3]Cl cat-II-67a [Ru(triphos-xyl)Br2] cat-II-67b [{Ru(triphos-xyl)}2(µ-Br)3]Br cat-II-9 [Ru(triphos-xyl)(OOCCH2CH2COCH3)2]

+ cat-II-CO2H [Ru(triphos-xyl)(CO)2H] cat-II-COH2 [Ru(triphos-xyl)(CO)H2]

xiii

List of Utilized or Observed Complexes / Catalytic Systems

+ cat-II-H3 [{Ru(triphos-xyl)}2(µ-H)3] cat-III-tmm [Ru(pr-triphos)(tmm)] cat-II-tmm [Ru(triphos-xyl)(tmm)] cat-I-MeCN3 [Ru(triphos)(MeCN)3](OTf)2 cat-I-OAcCl [Ru(triphos)(OAc)Cl] cat-I-OCHOOCHO [Ru(triphos)(κ1-OCHO)(κ2-OCHO)] cat-I-SH2 [Ru(triphos)(S)H2]

+ cat-I-SHH2 [Ru(triphos)(S)H(H2)] cat-I-tmm [Ru(triphos)(tmm)] cat-IV-tmm [Ru(pr-triphos-xyl)(tmm)] cat-VIII-tmm [Ru(triphos-xyl(OMe)2)(tmm)] cat-VII-tmm [Ru(triphos-xyl(CF3)2)(tmm)] cat-VI-tmm [Ru(triphos-tol(CF3))(tmm)] cat-V-tmm [Ru(triphos-tol)(tmm)]

xiv

1 Introduction and Motivation

1.1 Bio-Based Carboxylic Acids as Renewable Platform Chemicals for Chemical Building Blocks and Fuels

Driven by the continuous depletion of fossil resources, the development and establishment of sustainable systems for the utilization of renewable resources is of increasing importance. To overcome the dependency on petroleum, sustainable pathways to produce fuels and chemical building blocks have to be designed and integrated into the chemical value chain. In this context, both the establishment of sustainable processes for the production of already established petrochemical building blocks from renewable resources as well as the introduction of renewable substitutes for established petrochemicals are valuable strategies.[1]

In 2004, the US Department of Energy (DOE) selected 15 bio-based target molecules that can be obtained from biorefinery carbohydrates (Figure 1). These compounds – organic acids, esters, alcohols or ketones – have been selected due to a number of criteria, including their potential to act as starting materials and building blocks for the chemical value chain.[2] They are therefore also often referred to as bio(mass)-based or bio(mass)-derived platform chemicals.[3] Further criteria address the status of the compounds for the synthesis of consumer products as well as the technology readiness level of the production of the compounds from renewable sources.[2]

Figure 1: Sugar based target molecules selected by the DOE, report published in 2004.[2a]

1 Introduction and Motivation

Organic acids, esters and alcohols derived from biorefinery processes already contain important functional groups, designating them as valuable starting materials for the synthesis of various chemical products. The platform chemicals of the petrochemical industry – mainly olefins and aromatics – are largely produced by thermal cracking of longer chain hydrocarbons and must undergo energy-intensive oxidative processes to obtain hydroxyl, carbonyl, carboxyl or alkoxy functional groups. However, for the application as fuels, the oxygen content of biomass derived compounds must be reduced. Here, the bio-based chemicals must be deoxygenated to obtain the required combustion properties.[4]

Biomass based fuels can be classified into different generations. 1st generation biofuels, e.g. bioethanol and biodiesel, are mainly produced from oils and fats as well as sugars and starches. Grains and seeds are used as the main raw materials, which facilitates the conversion into biofuels due to the high sugar or oil content of the feedstock.[4a] Despite coming from renewable resources, the production of 1st generation biofuels also implicates environmental drawbacks such as preceding land clearing and greenhouse gas emissions.[5] As the use of raw materials for the production of 1st generation biofuels is also in competition with food industry, social and ethical concerns give additional rise to the motivation for finding pathways to produce 2nd generation biofuels that do not compete with food industry, e.g. from plants that can be grown on land not suitable for food production or from organic waste material.[4a,6] A prominent examples for a 2nd generation biofuel is ethanol, obtained by fermentation of carbohydrates from lignocellulose, but recently also furanoids such as 2-methylfuran (2-MF) and 2-methyltetrahydrofuran (2-MTHF) derived from furfural and levulinic acid have been found to exhibit promising combustion properties.[7]

The utilization of inedible biomass material as feedstock is certainly not only preferable for the production of biofuels, but biogenic building blocks and starting materials in general. A suitable feedstock for the sustainable production of renewable chemical compounds and 2nd generation biofuels is lignocellulosic material, which can be obtained from plants as well as from agricultural or wood waste.[6b] By hydrolytic treatment, lignocellulose is converted to its main components hemicellulose, cellulose and lignin, which form a mixture with triglycerides and proteins. Cellulose and hemicellulose can then be converted into C5 and C6 sugars by hydrolysis, which are transformed into chemical building blocks by fermentation or catalytic conversions.[8]

1 Introduction and Motivation

1.1.1 Levulinic Acid – Production from Renewable Resources and Utilization as Starting Material in the Chemical Value Chain 4-Oxopentanoic acid or levulinic acid (LVA) 9, one of the DOE’s top 15 potential building block candidates, has been comprehensively studied in both academic and industrial research.[9] Containing both a carbonyl and a carboxyl group, LVA is a promising building block for a variety of chemical transformations. Already in 1840, the synthesis of LVA from fructose by boiling in hydrochloric acid was reported by the Dutch chemist Mulder.[10] Most synthetic approaches proceed via the intermediate hydroxymethylfurfural (HMF), which in turn can be derived most easily from acidic treatment of C6 sugars, mainly fructose and glucose (Figure 2). As fructose can be converted into HMF more easily, but glucose is more abundantly available, upstream isomerization of glucose to fructose can be applied to improve the efficiency of the process.[11] In the course of improving the sustainability of LVA production, the utilization of lignocellulosic biomass as feedstock instead of pure sugars has been studied intensively. Lignocellulose contains cellulose and hemicellulose, which are composed of hexoses and are therefore suitable starting materials for the synthesis of HMF and LVA. Other sustainable synthetic approaches for LVA include the production from pentoses by acidic treatment and subsequent reduction, from furfural by ring cleavage or by hydrolysis of furfuryl alcohol.[11b,12] However, the production of LVA from lignocellulose has received most academic and industrial attention due to its high sustainability potential when using residues or waste material.

The Biofine process represents an established technology, where lignocellulosic feedstock can be converted to LVA, furfural and formic acid in two stages.[13] The material is shredded to particle sizes of 0.5 to 1 cm to increase the efficiency of the hydrolysis step. The particulates are then mixed with diluted sulfuric acid. The first reaction step is conducted in a plug flow reactor with a short residence time of only 12 s. Here, the hydrolysis of the polysaccharides to HMF and other soluble intermediates takes place at a pressure of 25 bar and temperatures of 210–220 °C.

Figure 2: Schematic reaction pathway from lignocellulose to levulinic acid (LVA) 9.

1 Introduction and Motivation

Figure 3: Selected products from levulinic acid (LVA) 9 transformations.

To obtain high yields of LVA, the second step is performed under operating conditions of 190– 200 °C and 14 bar at the same acid concentrations in a back mix reactor. The residence time amounts to 20 min and formed furfural and formic acid can already be removed due to their higher volatility. LVA can then be separated from the solid byproducts and is obtained in up to 98% purity after purification. Moreover, the process also enables the recycling of the utilized sulfuric acid.[11b,13]

LVA (9) can act as starting material for a variety of promising industrial products (Figure 3). γ-Valerolactone (GVL) 16 can be obtained from LVA by catalytic hydrogenation or transfer hydrogenation. GVL was tested as a sustainable organic solvent for biomass processing and has been discussed as fuel additive.[14] GVL based ionic liquids have been evaluated in solvent- applications and α-methylene-γ-valerolactone as well as a number of ring-opening products derived from GVL have been found to be potential polymer-building blocks.[15] Among these monomeric compounds, 1,4-pentanediol (1,4-PDO) 17 is regarded as promising starting material in polyester synthesis.[2a,14b] Condensation of 17 leads to the formation of 2-methyltetrahydrofuran (2-MTHF) 18, a structural relative to tetrahydrofuran, that can be applied as renewable solvent and is under consideration as 2nd generation biofuel.[7] It is of specific interest as solvent in biphasic reactions due to its miscibility gap with water.[16] The conversion of LVA with alcohols yields levulinate esters, which have been studied mainly for application as solvents and fuel additives.[17] Acid-catalyzed conversion of LVA with phenols yields diphenolic acid 19, which is discussed as renewable replacement for bisphenol A due to its structural similarity.[18] δ-Aminolevulinic acid (δ-ALA) 20 is an established fertilizer in agriculture. In medical applications, it is used to visualize tumorous tissue and is approved for photodynamic therapy.[19]

1 Introduction and Motivation

The successful establishment of LVA and its derivatives in chemical industry is proceeding, but mainly dependent on customer acceptance and cost competitiveness in production processes and applications. Academic and industrial focus is therefore laid on the development and improvement of viable process concepts and evaluation of the scope of application.

1.1.2 Itaconic Acid – Production and Application

Methylene butanedioic acid or itaconic acid (IA) 10 is a C5 dicarboxylic acid containing a methylene group as additional functionality.[20] It was found to be generated from citric acid pyrolysis by Baup already in 1836.[21] In 1932, Kinoshita reported the formation of IA as product from sugars using the osmophilic fungus Aspergillus itaconicus.[22] Intensive research on the production of IA by fungi from glucose in the middle of the 20th century found Aspergillus terreus to be the most suitable candidate.[23] Since then, this synthesis route is the commercially predominant production pathway, as the indirect synthesis via citric acid is less efficient due to lower overall yield.[24]

The formation of IA from glucose using A. terreus under aerobic fermentation conditions proceeds via the subsequent formations of pyruvate, oxaloacetate, malate and citric acid and cis-aconitate, which is then decarboxylated to IA.[24c,25] This process is conventionally conducted in a stirred tank reactor (STR), a well-established reactor type for fermentation processes. As STRs implicate high investment/operating costs and the mechanical stirring is harmful to the used filamentus fungi, other reactor types such as bubble columns, tubular reactors and air-lift reactors have been investigated as well.[26] Due to the absence of undesired moving parts, its easy construction and lower energy demand, the air-lift reactor has been studied more closely for the production of IA as promising alternative to the STR. Comparison of the two reactor types showed an increased production rate at lower power input when using the air-lift reactor.[26-27] Extensive research has also been carried out concerning the variation and improvement of the utilized fungi.

Figure 4: Schematic reaction pathway from lignocellulose to itaconic acid (IA) 10.

1 Introduction and Motivation

Initial screening of over 300 strains and optimization of the fermentation conditions for the most promising A. terreus strains led to significantly enhanced production of IA and different approaches for the immobilization of A. terreus have been pursued to facilitate the fermentation process.[23a,24c,25c,28] As A. terreus is very sensitive to variable substrate and product concentrations, pH value, temperature as well as nitrogen, phosphorous and trace metal ion levels, other fungi were closely investigated.[24a,24c,28a] Among a number of studied candidates, strains of Ustilago produced high amounts of IA from glucose and also strains of the yeast Candida showed promising activities.[29]

After the fermentation step, IA is isolated in the downstream process from the fermentation broth, which still contains unreacted biomass, fermentation media and side products. After filtration to discard the solid components, the filtrate is concentrated, and pure IA can be obtained in two crystallization steps. To achieve higher purities, even further purification steps are necessary.[23b,24b,26] Other recovery techniques include precipitation as itaconate salts, ultrafiltration, reverse osmosis, ion exchange, adsorption and reactive extraction.[24a,24b,30] The use of electrodialysis for the recovery of IA was shown to significantly lower investment costs and is suitable for continuous recovery when using a bipolar membrane, enabling an economic continuous production of IA.[24a,29e,31]

When IA was listed in the DOE’s report in 2004, it was stated that even though its market potential was significant, the existing production pathway was too cost-intensive and therefore needed to be optimized to be profitable.[2a] Approaches to improve the economics of the process do not only include the described efforts in optimization of reactor design, utilized fungi and downstream processing, but also the implementation of cheaper feedstocks. Whereas utilization of the pure sugars glucose or sucrose leads to the best yields for IA, also sugarcane or beet molasses, hydrolyzates of corn syrup or starch have been tested early on.[24a,32] However, utilization of these sources is in competition with the food industry and therefore unfavored, while the utilization of inedible biomass would be desired. In this context, beech wood, corn stover, corn cob and switchgrass have been successfully applied as lignocellulosic feedstock (Figure 4) using Neurospora crassa or different Aspergillus species as fungi.[33] Another promising substrate for sustainable production of IA is glycerol, which is generated as a by-product in biodiesel production in large quantities. It was successfully fermented to IA using A. terreus MJL05 strain.[34]

The multiple functionalities of the unsaturated dicarboxylic acid render IA a versatile building block in the chemical industry (Figure 5). It has been established as monomeric building block and additive for a variety of specialized polymers.[35] Copolymers of IA and acrylic acid show promising

1 Introduction and Motivation

properties for application as dental materials of high strength.[36] Some polymeric biocompatible hydrogels from the copolymerization of IA and acrylic acid have shown antibacterial activity and the potential for drug delivery and release.[37] pH-sensitive hydrogels synthesized by subsequent polymerization of IA with ethylene glycol and acrylic acid could be successfully applied in dye removal.[38] IA can be used as monomer for a number of bio-based epoxy resins such as flame retardants, high-performance epoxy resins or trifunctional epoxy resins.[39] Copolyamides with IA were found to be more cost effective in respect to conventional polyamides and exhibit improved mechanical and thermal properties. Moreover, due to their solubility in solvents like water or ethanol, these polyamides could also be environmentally degradable and some show promising potentials in medical applications.[40] IA can also be used as additive to vinylidene chloride coatings, emulsion paints and styrene-butadiene lattices with improved adhesion.[24a] The hydrogenation of IA leads to the successive formation of methyl succinic acid (MSA) 21, 2- and 3-methyl-γ-butyrolactone (2- and 3-MGBL) 22 and 23, and 2-methyl-1,4-butanediol (2-MBDO) 24.[41] 24 is structurally resemblant to the important established petrochemical 1,4-butanediol and is therefore regarded as promising bio-based starting material for polymers.[42] It has also been successfully applied in the synthesis of cyclic carbonates to be used in ring-opening polymerization reactions.[43] Acid catalyzed cyclization of 24 leads to the formation of 3-methyltetrahydrofuran (3-MTHF) 25, a structural isomer to 2-MTHF and structural relative to the established petrochemical THF.[44] 25 is therefore considered as promising renewable solvent and biofuel. It shows a miscibility gap with water, which is a promising feature for the use as solvent in biphasic reaction systems.[45] Methacrylic acid 26 is obtained from IA by decarboxylation and can be esterificated to methyl methacrylate, the starting material for poly(methyl methacrylate) (PMMA).[46]

Figure 5: Selected products from itaconic acid (IA) 10 transformations. 7

1 Introduction and Motivation

This pathway thereby enables a production route for bio-based PMMA. Itaconic anhydride 27 is applicable as starting material for different polymers and can be synthesized from IA by different procedures, among them refluxing IA with thionyl chloride or the catalytic conversion using magnesium chloride and a dialkyl dicarbonate as catalytic system.[47]

Whereas polymer syntheses from IA and derivatives are already well-studied, the application of the promising hydrogenation products 2-MBDO and 3-MTHF has up to date not received remarkable academic attention. This can certainly be partly attributed to the challenges in efficient supply of the materials as the catalytic hydrogenation of IA remains highly challenging.

1.2 Homogeneous Catalytic Systems for the Hydrogenation of Carboxylic Acids and Esters

The selective hydrogenation of the biogenic acids LVA and IA to valuable chemical products has been intensively studied and was found to require tailored catalytic systems to achieve high yield and selectivities. In the 1980s and 90s, different ruthenium based complexes have been considered for the hydrogenation of carboxylic acids and esters.[48] Among them, the catalytic system composed of the precursor [Ru(acac)3] and the tridentate phosphine ligand 1,1,1-tris(diphenylphosphinomethyl)ethane (triphos) enabled the conversion of dimethyl oxalate to ethylene glycol with 95% yield.[48g] This in situ system was also found to hydrogenate dimethyl phthalate, dimethyl maleate, methyl palmitate and benzyl benzoate in the presence of triethyl amine.[48h] In 2006, Milstein showed the successful hydrogenation of non-activated esters using ruthenium pincer complexes.[49] Driven by these results, a number of different ruthenium pincer complexes were developed and introduced for the hydrogenation of esters in the following years.[50] Ruthenium complexes with NP, PNNP or PNNN type ligands as well as ruthenium carbene complexes were also found to be suitable catalysts.[51] Besides ruthenium and other precious-metal catalysts, also a number of non-precious metal complexes could be applied in the hydrogenation of carboxylic acids and esters, albeit with generally lower activities.[52]

1.2.1 Development of Ruthenium Triphos Based Catalysts for Selective Hydrogenation Reactions Alongside the development of catalytic systems with complex phosphorous and nitrogen based pincer ligands, the [Ru(acac)3]/triphos in situ system cat-I-acac (Figure 6) was studied intensively. In 2003, the application of cat-I-acac in the hydrogenation of free carboxylic acids was described in a patent by Crabtree and coworkers at Davy Process Technologies Ltd..[53] Later, Cole-Hamilton applied cat-I-acac in the reduction of amides in 2007.[54] 8

1 Introduction and Motivation

Figure 6: Structures of [Ru(acac)3)]/triphos cat-I-acac, [Ru(triphos)(tmm)] cat-I-tmm and [Ru(triphos-xyl)(tmm)] cat-II-tmm.

In 2010, Leitner and Klankermayer used the system to enable the selective hydrogenation of LVA and IA to the corresponding lactones, diols or methyltetrahydrofurans.[41a,55] These findings entailed intensive studies on the use of the triphos ligand and its derivatives in catalytic systems for the hydrogenation of a large variety of substrates such as carboxylic acids and esters, lactones and carbon dioxide.[56] In 2012, Leitner and Klankermayer introduced the molecular defined catalyst [Ru(triphos)(tmm)] cat-I-tmm (Figure 6) for the first homogeneously catalyzed direct hydrogenation of carbon dioxide to methanol, which was found to be enhanced in the presence of acidic additives.[57] cat-I-tmm can be derived from the conversion of the ruthenium precursor

[Ru(cod)(methylallyl)2] with triphos in a one-step synthesis.[58] Utilization of triphos leads to a very robust structure of the formed ruthenium complex due to its facial coordination to the ruthenium center. The anionic trimethylenemethane (tmm) ligand is readily detached under suitable reaction conditions providing an easily accessible coordination site at the metal center. These features render cat-I-tmm a very stable yet highly active catalyst. In 2014, cat-I-tmm also enabled the hydrogenation of carboxylic acids and esters, amides, imides, lactones, carbonates and anhydrides as well as the methylation of amines or imines using carbon dioxide.[59] It was also successfully applied in the catalytic cleavage of lignin model compounds, the methylation of ammonia and the synthesis of dialkoxymethane ethers using hydrogen and carbon dioxide.[58,60]

Figure 7: Structures of different reported ruthenium triphos complexes.[55,61]

1 Introduction and Motivation

Alongside the development of cat-I-tmm bearing a tmm ligand, several other molecular defined ruthenium triphos based complexes were characterized, isolated and applied in catalytic reactions (Figure 7). In 2008, Dyson synthesized the acetate complex [Ru(triphos)(OAc)Cl] cat-I-OAcCl from the corresponding tris(triphenylphosphine) species.[61a] In 2010, Dutta applied the monocarbonyl complex [Ru(triphos)(CO)Cl2] cat-I-COCl2 in the transfer hydrogenation of carbonyl compounds.[61b] Beller and Gonsalvi performed the dehydrogenation of formic acid with different ruthenium triphos systems, among them the tris(acetonitrile) complex [Ru(MeCN)3](OTf)2 cat-I-MeCN3. In 2014, Cantat observed the formation of [Ru(triphos)(κ1-OCHO)(κ2-OCHO)] cat-I-OCHOOCHO in the disproportionation of formic acid using cat-I-tmm.[61c]

To increase the activity of the catalytic ruthenium triphos system, various modifications on the structure of the triphos ligand have been performed by changing the sterical and electronic properties by alteration of the backbone or phosphine groups. In 2014 and 2015, Long and Miller synthesized N-centered triphos ligand (N-triphos) based ruthenium catalysts, which were successfully applied in the selective hydrogenation of LVA to 17. Trapp utilized ruthenium complexes with N-triphos, P-triphos and Si-triphos for the reduction of carbon dioxide to dimethoxymethane and methyl formate.[62] In situ systems of N-triphos derivatives with a ruthenium precursor were applied by Palkovits in 2016 in the hydrogenation of carboxylic acids.[56j]

Driven by the interest on cheaper and more abundant metal catalysts, the research on non-precious metal triphos based catalytic systems has also significantly increased in the past years. In 2015, Elsevier and de Bruin showed the hydrogenation of carboxylic acids with a cobalt triphos based catalytic system.[56e] Lately, other cobalt catalyzed hydrogenation reactions with triphos-derivatives were reported by Beller and Sundararaju and recently Roemelt and Apfel reported the solvent- controlled reduction of carbon dioxide to either CO or formate using an iron triphos complex.[56m,56p,56q]

A large scope of derivatives was obtained by alteration of the moieties at the three phosphorous atoms of the triphos ligand.[63] Among the investigated structures, [Ru(triphos-xyl)(tmm)] cat-II-tmm (Figure 6) bearing sterically demanding meta-xylyl rings showed significantly increased activity in the hydrogenation of methyl benzoate to benzyl alcohol as compared to cat-I-tmm.[64] The triphos-xyl ligand was also used in combination with cobalt precursors to successfully catalyze the hydrogenation of carbon dioxide to methanol or dialkoxymethane ethers and the reductive alkylation of anilines with carboxylic acids.[56l,56n,56p] Recently, Klankermayer reported the base-free hydrogenation of CO2 to methyl formate using ruthenium complexes bearing N-triphos, P-triphos and Si-triphos ligands with alkyl, aryl or cycloalkyl moieties at the three phosphorous atoms.[65] The

1 Introduction and Motivation

reductive depolymerization of waste plastics composed of the polymers polylactic acid, polycaprolactone, bisphenol A based polycarbonate, polyethylene terephthalate and polybutylene terephthalate was recently reported using cat-I-tmm and cat-II-tmm, demonstrating the high tolerance of the catalytic systems against substrate impurities such as dyes and metals.[66]

1.2.2 Mechanistic Insights on the Hydrogenation of Carboxylic Acids and Esters with Ruthenium Triphos Catalysts The ongoing development and tailoring of ruthenium triphos systems in catalytic hydrogenation reactions is supported by detailed mechanistic studies, which have paved the way for remarkable enhancements of the activities and stabilities of the catalytic systems as well as the selectivities of the catalytic reactions.

When using cat-I-acac or cat-I-tmm as catalysts, [Ru(triphos)(S)H(H2)]+ cat-I-SHH2 was postulated as catalytically active species under acidic hydrogenation conditions and

[Ru(triphos)(S)H2] cat-I-SH2 under neutral conditions (Figure 8).[55,59b] The dimeric structure

[{Ru(triphos)H}2(µ-H)2] cat-I-H2H2 has been postulated to form an equilibrium with cat-I-SH2 according to NMR-spectroscopical investigations.[59b] Based on DFT-studies, the reduction of substrates such as aldehydes, ketones, lactones and carboxylic acids with cat-I-SH2 as active species starts with a hydride transfer from the ruthenium center to the carbonyl/carboxyl carbon atom of the substrate (migratory insertion). This step is followed by a protonation of the Ru-O unit by a coordinated hydrogen molecule (σ-bond metathesis). After the hydrogenation of LVA using cat-I-acac, the monocarbonyl dihydrido species [Ru(triphos)(CO)H2] cat-I-COH2 could be isolated from the reaction mixture and was identified as catalytically inactive under neutral conditions. However, slightly lower to similar catalytic activities as compared to cat-I-acac in the presence of acids have been observed. NMR-spectroscopical studies revealed the formation of a monohydride ruthenium triphos species with a coordinated LVA (9) molecule showing the reactivation of cat-I-COH2 with LVA at room temperature.[55]

In the presence of acids and CO or by decarbonylation reactions of reactants, cat-I-SH2, cat-I-SHH2 and cat-I-COH2 can react to the cationic dicarbonyl monohydrido complex [Ru(triphos)(CO)2H]+ cat-I-CO2H, which was isolated and found to be catalytically inactive.[67] Another deactivation pathway of the catalytically active species occurs by irreversible dimerization reactions, in which cationic or neutral hydride-bridged dimers are formed. The defined structures of these inactive dimers are not fully revealed, but the two structures [{Ru(triphos)}2(µ-H)2] cat-I-H2 and

+ [59b,68] [{Ru(triphos)}2(µ-H)3] cat-I-H3 have been proposed based on NMR-spectroscopical studies.

1 Introduction and Motivation

Figure 8: Structures and mechanistical relations between monomeric and dimeric ruthenium triphos species under hydrogenation reaction conditions (P3 = triphos).

In this manner, efforts to suppress or minimize these dimerization tendencies have led to the utilization of more sterically demanding ligands such as triphos-xyl (Chapter 1.2.1). Further studies by Markus Meuresch, Stefan Westhues and Andrey Charkovskiy have demonstrated the challenges to tailor triphos derivatives which form stable, yet catalytically active ruthenium complexes with low dimerization tendencies.[68b,69]

1 Introduction and Motivation

1.3 Motivation and Aim

One major challenge in the integration of sustainable processes into the chemical value chain is the development of highly active and selective catalysts that enable the transformations of bio-based platform chemicals to valuable chemicals in continuous processes.

Based on earlier work on the hydrogenation of bio-based acids and esters with ruthenium triphos systems, tailored ruthenium triphos-xyl complexes should pave the way to optimized performances.

The targeted complexes of the general structure [Ru(triphos-xyl)(OR)2] can be obtained by the conversion of [Ru(triphos-xyl)(tmm)] with selected phenols, naphthols and acids. The assembled portfolio of targeted species will then allow the study of the influence of varying sterical and electronic effects on the catalytic performance.

The synthesized complexes can be applied as catalysts in the hydrogenation of LVA and GVL, both in batch reactions and recycling experiments. Gas chromatographic analyses as well as kinetic experiments will allow an evaluation of reactivity of the new catalysts. With regard to the application in a scaled-up continuous process, the influence of temperature on catalyst deactivation will be examined and the performance of selected catalysts can be compared to the corresponding precursor/ligand in situ systems. NMR spectroscopic studies and evaluation of the pressure courses will also allow deeper insights into the reaction pathway.

The hydrogenation of 2-methyl-γ-butyrolactone, IA and itaconate esters will be investigated using the tailored homogeneous catalytic systems. Besides the defined complexes, also in situ systems of different [Ru(triphos-y)(tmm)] catalysts with selected fluorophenols, -naphthols and perhalogenatobenzenes will be investigated. NMR spectroscopic investigations will aid the understanding of different processes in the reaction course, finally leading to the development of a highly active molecular catalyst system for the hydrogenation of selected substrates.

2 Development of Ruthenium Triphos-xyl Based Complexes

The enhancement of catalytic activities in hydrogenation reactions by addition of Brønsted or Lewis acids to ruthenium triphos based catalytic systems has been shown for various substrates.[57a,59b,64a] However, the use of strong acids as additives to the hydrogenation of LVA 9 and IA 10 leads to the formation of 2-MTHF 18 and 3-MTHF 25 as consecutive products to 1,4-PDO 17 and 2-MBDO 24 and is therefore disadvantageous for the selective conversion to these diols (Figure 9).[41a]

Figure 9: Challenges in enhancing the selective hydrogenation of LVA to 1,4-PDO.

Hence, the incorporation of additives that enhance the catalytic activity but do not promote the subsequent acid catalyzed cyclization of the diol products is desirable and was already subject to earlier studies. Dominik Limper synthesized different ruthenium triphos acetato and phenolato complexes by conversion with the corresponding acetic acid or phenol derivatives. These complexes already showed promising activities in the hydrogenation of esters.[67]

Scheme 1: Conversion of cat-II-tmm with selected acids and alcohols.

Based on these findings, selected acids, phenols, naphthols as well as a terpene were tested in the conversion with [Ru(triphos-xyl)(tmm)] cat-II-tmm (Scheme 1). In a first approach, stoichiometric amounts of the investigated substances were stirred with cat-II-tmm at 105 °C in toluene overnight. A successful conversion to the desired complex was usually indicated by color change of the reaction solution or even precipitation of the formed complex. NMR spectroscopic measurements of the reaction solution after cooling to room temperature as well as of the obtained solids after the work-up were conducted to identify the synthesized species. A detailed overview on the scope of the reactants is systematically shown in Figure 10. 14

2 Development of Ruthenium Triphos-xyl Based Complexes

Some of the targeted complexes could not be obtained as no conversion was observed when even a large excess of the alcohol did not lead to observable product formation. In detail, with twenty equivalents of 1,3,5-trihydroxybenzene 28, two equivalents of 8-hydroxyquinoline 29 or BHT 30, four equivalents of syringol 31 and ten equivalents of phenol 32 or menthol 33, no product could be isolated. This can generally be attributed to either the large sterical demand or the insufficient acidity of the utilized substrates necessary to protonate the tmm ligand. Addition of two equivalents of the isomers 1,2,4-trihydroxybenzene 34 or 1,2,3-hydroxybenzene 35 resulted in more than one product species, according to 31P{1H}-NMR spectroscopy, due to the presence of more than one accessible hydroxyl group in the molecules.

Figure 10: Investigated ligands for the synthesis of [Ru(triphos-xyl)(OR)2] complexes, pKa values of the ROH ligands in parentheses.[70] 15

2 Development of Ruthenium Triphos-xyl Based Complexes

The same could be observed for twenty equivalents of 1,3-dihydroxynaphthalene 36. Using twenty equivalents of 3,5-dimethoxyphenol 37 led to the formation of an unsymmetric complex, which resulted in a triplet and doublet in the 31P{1H}-NMR spectrum. However, the species could not be successfully isolated due to very similar solubility properties of 37 and the formed complex. Same applied for the reaction of cat-II-tmm with twenty equivalents of 5,6,7,8-tetrahydro-1-naphthol 38, which resulted in 90% conversion to a new species, which could not be successfully separated from remaining 38. The analogous conversion with 5,6,7,8-tetrahydro-2-naphthol 39 led to the formation of multiple species according to 31P{1H}-NMR spectroscopy, which were not further identified. The conversion of cat-II-tmm with ten equivalents of 4-fluorophenol 40 was found to be successful according to the measured 31P{1H}-NMR spectrum. However, the corresponding complex could not be separated from the remaining excess of 40 due to very similar solubility properties of 40 and the formed complex. The complex derived from the conversion of cat-II-tmm with four equivalents of 3,5-diiodophenol 41 could be obtained with a purity of only 90%. Using 1.5 equivalents of 2,2’-methylenediphenol 42 resulted in the formation of a bridged ruthenium triphos-xyl dimer complex according to 31P{1H}-NMR spectroscopy. However, this complex could also not be isolated in sufficient purity. The conversion of cat-II-tmm with twenty equivalents of 1,1’-bi-2-naphthol 43 led to the formation of the corresponding complex, but due to extensive washing required to remove remaining 43, this species could only be obtained in marginal quantities.

The successful reactions towards the envisaged complexes are summarized in Figure 10 (right column). In detail, two equivalents of acetic acid 44 or trifluoroacetic acid 45 could be successfully converted with cat-II-tmm to yield the corresponding bis(acetato) and bis(trifluoroacetato) complexes. After the removal of toluene followed by dissolving the obtained solid in DCM, subsequent solvent removal and washing of the solid residue with n-pentane, the yellow powders could be isolated after drying in vacuo. The synthesis of [Ru(triphos-xyl)(OOCCF3)2] cat-II-45 was already reported by Dominik Limper by a similar procedure.[67] The bis(acetato) and bis(trifluoroacetato) complexes cat-II-44 and cat-II-45 could be obtained with yields of 58% and 71%, respectively and were found to generate singlets in the recorded 31P{1H}-NMR spectra (Table 1). The observed low-field shifts of cat-II-44 and cat-II-45 compared to cat-II-tmm (32.3 ppm in

THF-d8 and 31.7 ppm in CD2Cl2) can be attributed to the lower magnetic shielding of the phosphorous nuclei. The electron withdrawing effect of the acetato ligands lowers the electron density at the phosphorous atoms. As anticipated, the singlet in the 31P{1H}-NMR spectrum generated by cat-II-45 is shifted more downfield than the corresponding signal generated by cat-II-44, as the fluoro-substituted acetato ligand is more electron-withdrawing.

2 Development of Ruthenium Triphos-xyl Based Complexes

To generate the bis(naphtholato) complexes cat-II-46 and cat-II-47, cat-II-tmm had to be converted with a twenty-fold excess of 1-naphthol 46 or 2-naphthol 47, respectively. The purple product solutions were cooled to −18 °C to crystallize the corresponding complexes as dark red crystals, which were separated from the mother liquor and washed with n-pentane. The crystals of cat-II-47 had to be additionally washed with diethyl ether to remove all remaining 47. After drying in vacuo, cat-II-46 and cat-II-47 could be isolated as dark purple powders with yields of 62% and 37%, respectively and were found to generate more downfield shifted singlets than cat-II-44 and cat-II-45 in the measured 31P{1H}-NMR spectra (Table 1), which is contrary to the trend expected from the higher pKa values.

Table 1: Structures and 31P{1H}-NMR spectroscopical shifts of the complexes cat-II-44 to cat-II-50 measured at rt.

complex δ (31P{1H}) [ppm] complex δ (31P{1H}) [ppm]

38.4 (in THF-d8) 45.3 (in THF-d8) 38.0 (in CD2Cl2) 44.8 (in CD2Cl2)

cat-II-44 cat-II-48

39.9 (in THF-d8) 48.7 (in THF-d8) 40.5 (in CD2Cl2) 48.1 (in CD2Cl2)

cat-II-45 cat-II-49

50.9 (in THF-d8) 44.0 (in THF-d8) 50.0 (in CD2Cl2) 44.6 (in CD2Cl2)

cat-II-46 cat-II-50

51.2 (in THF-d ) 8 50.2 (in CD2Cl2)

cat-II-47 17

2 Development of Ruthenium Triphos-xyl Based Complexes

The naphthalene-bis(olato) complexes cat-II-48 and cat-II-49 could be synthesized by conversion of cat-II-tmm with one equivalent of 1,8-dihydroxynaphthalene 48 or 2,3-dihydroxynaphthalene 49, respectively and isolated following the same work-up procedure as for cat-II-44 and cat-II-45. cat-II-48 was obtained as dark green powder with a yield of 60% generating a high-field shifted singlet in the 31P{1H}-NMR spectrum as compared to the bis(naphtholato) complexes. cat-II-49 was isolated as bright red powder with a yield of 61% generating a slightly high-field shifted singlet as compared to cat-II-46. The conversion of cat-II-tmm with two equivalents of 2,2’-biphenol 50 led to the formation of the bis(phenolato) complex cat-II-50. The orange product solution was cooled to

−18 °C to crystallize cat-II-50 as orange crystals, which were separated from the mother liquor, washed with n-pentane and dried in vacuo to obtain an orange powder with a yield of 69%. cat-II-50 generates the most high-field shifted singlet in the 31P{1H}-NMR spectrum compared to complexes cat-II-46–cat-II-49. The trend of the chemical shifts of the complexes cat-II-46– cat-II-49 is in general alignment with the acidity of the corresponding naphthols and dihydroxynaphthalenes. However, cat-II-50 derived from 50 with a higher pKa value than 48 is shifted more downfield than cat-II-48, which could be attributed to an electron releasing effect of the free OH-groups of the 2-ortho-phenylolphenolato ligands. cat-II-49 could also be isolated as single crystals and analyzed by X-ray structure analysis. The obtained crystal structure shows a coordination number of five for ruthenium (Figure 11).

Figure 11: Single crystal structure of cat-II-49 (hydrogen atoms are omitted for clarity). 18

2 Development of Ruthenium Triphos-xyl Based Complexes

Table 2: Bond-lengths and angles within cat-II-49 determined by single-crystal X-ray diffraction.

bond bond-length [Å] angle [°]

Ru–O 2.055 / 2.058 - Ru–P 2.178 / 2.255 / 2.261 - O–Ru–O - 79.67 Ru–O–C - 113.15 / 113.26

Pax–Ru–O - 105.00 / 110.36

Peq–Ru–O - 91.77 / 96.38 / 161.21 / 166.19

The P–Ru–O angles for the equatorial P atoms of 91.77° and 96.38° for the adjacent P and O atoms are slightly larger but close to the values of an ideal planar arrangement (Table 2). The P–Ru–O angles of 161.21° and 166.19° for the opposite P and O atoms are smaller than the value expected for an ideal planar geometry. The P–Ru–O angles of 105.00° and 110.36° for the axial phosphorous atom are in turn larger than a right angle. Consequently, compex cat-II-49 exhibits a distorted tetragonal pyramidal structure with an Addison parameter of τ = 0.23.[71] The Ru–P bonds are shorter than the average value of 2.282 Å reported for cat-II-tmm.[64a]

The fluorinated bis(phenolato) and bis(naphtholato) complexes cat-II-51–cat-II-59 were synthesized according to the general procedure using different amounts of the corresponding fluorophenols or -naphthol. The work-ups and isolations were conducted analogously to the bis(acetato) complexes. The syntheses of the mono- and difluorinated phenolato complexes cat-II-51, cat-II-52 and cat-II-53 were achieved using an excess of the corresponding phenols 2-fluorophenol 51 (10 eq), 2,5-difluorophenol 52 (4 eq) or 2,6-difluorophenol 53 (4 eq), respectively. These three complexes could be isolated as red powders with yields of 50–73%. The other di-, tri-, tetra- and pentaflurophenols 54–58 as well as the perfluorinated naphthol heptafluoro-2-naphthol 59 were used in stoichiometric amounts to achieve the synthesis of the corresponding complexes cat-II-54–cat-II-59. These were isolated as red, orange or yellow powders with yields of 67–85%. The complexes cat-II-52, cat-II-54, cat-II-56, cat-II-57 and cat-II-58 bearing meta-substituted fluorophenolato ligands generate more low-field shifted singlets in the measured 31P{1H}-NMR spectra than the complexes cat-II-51, cat-II-53 and cat-II-55 bearing only ortho- and/or para-substituted fluorophenolato ligands (Table 3). This observation can be attributed to the stronger electron withdrawing effect of fluoro-substituents in meta-position. Theoretical studies on the effect of fluorine substitution on phenols by Urban and Taft found significantly higher phenol acidities in the case of meta-substitution.[72] No value for the pKa of 59 could be obtained from literature, but due to its structure of a perfluorinated naphthol, the value is expected to be similar or lower than the pKa of 58. 19

2 Development of Ruthenium Triphos-xyl Based Complexes

Table 3: Structures and 31P{1H}-NMR spectroscopical shifts of the complexes cat-II-51 to cat-II-59 measured at rt.

complex δ (31P{1H}) [ppm] complex δ (31P{1H}) [ppm]

50.2 (in THF-d8) 52.8 (in THF-d8) 50.0 (in CD2Cl2) 51.5 (in CD2Cl2)

cat-II-51 cat-II-56

51.1 (in THF-d8) 51.7 (in THF-d8) 49.8 (in CD2Cl2) 50.2 (in CD2Cl2)

cat-II-52 cat-II-57

49.6 (in THF-d8) 52.7 (in THF-d8) 48.2 (in CD2Cl2) 51.1 (in CD2Cl2)

cat-II-53 cat-II-58

52.4 (in THF-d8) 53.4 (in THF-d8) 51.3 (in CD2Cl2)

cat-II-54 cat-II-59

50.8 (in THF-d ) 8 49.3 (in CD2Cl2)

cat-II-55

2 Development of Ruthenium Triphos-xyl Based Complexes

Figure 12: Single crystal structure of cat-II-54 (hydrogen atoms and solvent molecules are omitted for clarity). As consequently anticipated, the perfluorinated bis(naphtholato) complex cat-II-59 generates the most low-field shifted singlet in the 31P{1H}-NMR spectrum. It could be observed that not all aromatic carbon nuclei of the phenolato and naphtholato ligands could be detected when measuring the 13C{1H}-APT-NMR spectra of cat-II-51–cat-II-59 in CD2Cl2, whereas all expected signals could be observed using THF-d8. This observation suggests an effect of THF-d8 on dynamics of the molecules in solution, possibly by coordination through the Lewis basic functionality. For cat-II-54 (Figure 12) and cat-II-58 (Figure 13), single crystals could be isolated and analyzed by X-ray structure analysis. The P–Ru–O angles for the equatorial phosphorous atoms of 86.43° and 96.42° for the adjacent P and O atoms are slightly deviant, but close to the values of an ideal planar arrangement (Table 4). The P–Ru–O angles of 155.21° and 172.20° for the opposite P and O atoms are smaller than the value expected for an ideal planar geometry.

Table 4: Bond-lengths and angles within cat-II-54 determined by single-crystal X-ray diffraction.

bond bond-length [Å] angle [°]

Ru–O 2.106 / 2.073 - Ru–P 2.194 / 2.260 / 2.275 - O–Ru–O - 86.28 Ru–O–C - 127.25 / 138.28

Pax–Ru–O - 97.42 / 118.54

Peq–Ru–O - 86.43 / 96.42 / 155.21 / 172.20

2 Development of Ruthenium Triphos-xyl Based Complexes

Figure 13: Single crystal structure of cat-II-58 (hydrogen atoms and solvent molecules are omitted for clarity).

The values for the P–Ru–O angles for the axial phosphorous atom are bigger than 90°, but significantly smaller than 180°. Consequently, the structure of complex cat-II-54 classifies as a distorted tetragonal pyramidal structure with an Addison parameter of τ = 0.37.[71]

The obtained data for cat-II-58 (Table 5) disclose the structure to be in between a tetragonal pyramidal structure and a trigonal bipyramidal structure with an Addison parameter of τ = 0.43.[71]

In contrast to cat-II-54, a Pax–Ru–Oax angle close to 180° can be observed. This geometry is the result of a π-stacking of the two aromatic perfluorinated rings, which is not observed for the less substituted aromatic rings in cat-II-54.

Table 5: Bond-lengths and angles within cat-II-58 determined by single-crystal X-ray diffraction.

bond bond-length [Å] angle [°]

Ru–O 2.104 / 2.111 - Ru–P 2.209 / 2.219 / 2.260 - O–Ru–O - 85.78 Ru–O–C - 127.60 / 146.39

Pax–Ru–Oax - 171.06

Pax–Ru–Oeq - 86.96

Peq–Ru–Oeq - 128.46 / 145.32

Peq–Ru–Oax - 91.50 / 100.27 22

2 Development of Ruthenium Triphos-xyl Based Complexes

Whereas the P–Ru–O angles for the axial phosphorous atoms are slightly smaller than the ideal values of 90° and 180° for the ideal trigonal bipyramidal geometry, the values for the P–Ru–O angles for the equatorial phosphorous atoms are bigger than the ideal values of 120° due to the rigidity of the triphos ligand. cat-II-54 and cat-II-58 show similar geometries as the analogous ruthenium triphos complexes reported by Dominik Limper with the triphos-xyl derivatives exhibiting longer Ru–O bonds.[67] In comparison to cat-II-49, the Ru–O bonds of the fluorophenolato complexes are also longer, suggesting more labile structures, which are expected to facilitate the activation of the complexes under hydrogenation conditions.

The chloro- and bromo-substituted analogs of cat-II-54 were obtained by the conversion of cat-II-tmm with four equivalents of 3,5-dichlorophenol 60 or 3,5-dibromophenol 61, respectively. The isolation procedure of the resulting complexes cat-II-60 and cat-II-61 were performed mostly similarly to those of the fluorinated phenols, but the obtained orange powders had to be additionally washed with diethyl ether to remove remaining 60 or 61. cat-II-60 and cat-II-61 were isolated with yields of 81% and generate slightly downfield shifted singlets in the 31P{1H}-NMR spectra in comparison to cat-II-54, which is in alignment with the different basicities of the ligands. The conversion of cat-II-tmm with two equivalents of 3,5-bis(trifluoromethyl)phenol 62 resulted in the formation of cat-II-62, which could be isolated as an orange powder with a yield of 78%. cat-II-62 generates a downfield shifted singlet in the 31P{1H}-NMR spectrum as compared to cat-II-60 and cat-II-61 (Table 6), which is in line with the higher acidity of 62.

Table 6: Structures and 31P{1H}-NMR spectroscopical shifts of the complexes cat-II-60 to cat-II-62 measured at rt.

complex δ (31P{1H}) [ppm] complex δ (31P{1H}) [ppm]

52.9 (in THF-d8) 53.9 (in THF-d8) 51.8 (in CD2Cl2) 52.9 (in CD2Cl2)

cat-II-60 cat-II-62

53.1 (in THF-d ) 8 52.0 (in CD2Cl2)

cat-II-61

2 Development of Ruthenium Triphos-xyl Based Complexes

Table 7: Structures and 31P{1H}-NMR spectroscopical shifts of the complexes cat-III-tmm and cat-IV-tmm measured at rt.

complex δ (31P{1H}) [ppm] complex δ (31P{1H}) [ppm]

32.7 (in CD2Cl2) 30.0 (in CD2Cl2)

cat-III-tmm cat-IV-tmm To study the effect of an elongated alkyl chain at the triphos backbone, [Ru(pr-triphos)(tmm)] cat-III-tmm and [Ru(pr-triphos-xyl)] cat-IV-tmm were synthesized by conversion of

[Ru(cod)(methylallyl)2] with the corresponding triphos derivatives supplied by HYBRID CATALYSIS according to the literature procedures for the syntheses of cat-I-tmm and cat-II-tmm (Table 7).[58,69a]

Analogously to cat-II-tmm and cat-I-tmm, cat-IV-tmm generates a slightly high-field shifted singlet in the corresponding 31P{1H}-NMR spectrum as compared to cat-III-tmm due to the higher electron density at the phosphorous atoms induced by the positive inductive effect of the meta-alkyl substituents at the aromatic rings.[68b]

The wide portfolio of synthesized ruthenium triphos-xyl based complexes to be applied in the hydrogenation of 9, 10 and their derivatives allows the study of different sterical and electronic effects of the introduced ligands on the catalytic performance.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

3.1 Hydrogenation of Levulinic Acid and γ-Valerolactone – State of the Art

The catalyzed hydrogenation of levulinic acid (LVA) 9 to 1,4-pentanediol (1,4-PDO) 17 using molecular hydrogen proceeds in two basic reaction steps. First, 9 undergoes an intramolecular transformation to γ-valerolactone (GVL) 16 under the elimination of water by hydrogenation of the keto group. In the second step, 16 reacts to 17 under ring-opening by hydrogenation of the ester functionality. In the presence of acids, 17 undergoes a cyclic condensation to 2-methyltetrahydrofuran (2-MTHF) 18 (Scheme 2).

Scheme 2: Catalytic conversion of LVA (9) to GVL (16), 1,4-PDO (17) and 2-MTHF (18).

The selective hydrogenation of 9 has been achieved using both homogeneous and heterogeneous catalytic systems. There are numerous examples for the successful hydrogenation or transfer hydrogenation of 9 to 16 with high yields, turn over numbers (TONs) and turn over frequencies (TOFs),[56c,73] but only few reported catalysts are able to catalyze the hydrogenation of 16 to 17 or even the consecutive hydrogenation of 9 to 17. In 1947, Hixon achieved 83% yield for 17 from 16 using a heterogeneous copper-chromium oxide catalyst.[74] In 2011, Corma used a Ru/C catalyst for the hydrogenation of 9, achieving a 1,4-PDO (17) yield of 10%.[75] In the same year, Hwang and

Chang achieved yields of 11% for 17 and 64% for 18 using a Cu/SiO3 catalyst.[76] A significantly higher 1,4-PDO (17) yield of 82% was obtained by Pinel in 2013 with a Ru-Re/C catalyst.[77] One year later, Li and Dong used a molybdenum modified Rh/SiO2 catalyst to achieve a 1,4-PDO (17) yield of 70%.[78] Kaneda studied different Pt-Mo catalysts in the hydrogenation of 9 to reach a 1,4-PDO (17) yield of 93% in 2015 and a 2-MTHF yield of 86% in 2016.[79] In 2017, a selectivity of 99% for 17 by hydrogenation of 16 was achieved in continuous flow using a Cu/ZnO catalyst.[80] A molybdenum modified Ru/AC catalyst was reported in 2018 by Zhu and Cheng to achieve yields of over 95% 17. Recently, Hu and Hu achieved a yield of 53.6% of 17 from 9 using copper-based catalysts.[81] 25

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Alongside the development of heterogeneous catalysts, the design and study of homogeneous catalytic systems has revealed their high potentials in catalyzing these challenging transformations (Table 8). In 2010, Leitner and Klankermayer achieved 95% yield of 17 from 9 after 18 h at 160 °C and 100 bar H2 using 0.1 mol% [Ru(acac)3] and 0.2 mol% triphos as homogeneous in situ system cat-I-acac (entry 1).[41a] A mechanistic DFT-study revealed an energetic span for the hydrogenation of 9 to 16 of 17.1 kcal/mol and 7.0 kcal/mol for the subsequent formation of 17 from 16 under acidic conditions.[55] Using 2 mol% of the molecular defined complex [Ru(triphos)(tmm)] cat-I-tmm led to a yield of 99% for 17 after a reaction time of 16 h at 140 °C and 50 bar H2 (entry 2).[59b] In 2015, Elsevier and de Bruin used a cobalt triphos in situ system for the hydrogenation of 9 and 16 to

17. Using 10 mol% catalyst, 80 bar H2 and 100 °C, they achieved yields of 47% 17 and 14% 18 for the hydrogenation of 9 in THF after 22 h reaction time (entry 3). The hydrogenation of 16 in MeOH under otherwise identical conditions led to yields of 63% 17 and 25% 18.[56e] In the same year, Long and Miller applied different ruthenium N-triphos complexes in the hydrogenation of 9. A catalyst amount of 0.5 mol% at 150 °C and 65 bar H2 led to a 1,4-PDO (17) yield of 99% after 25 h reaction time (entry 4). The addition of HNTf2 resulted in a 2-MTHF (18) yield of 87%.[56g] In 2016, Palkovits applied unsymmetrical ruthenium N-triphos in situ systems in the hydrogenation of 9 to 17 achieving a yield of 98% after 18 h at 160 °C and 70 bar H2 using 0.5 mol% ruthenium precursor and 0.75 mol% ligand (entry 5).[56j]

Homogeneous catalytic systems with other ligands than triphos for the hydrogenation of 9 to 16 and 17 have also been reported.

Table 8: Selected reported catalytic systems with phosphine ligands for the hydrogenation of LVA (9) to GVL (16) and 1,4-PDO (17).

pinitial Y Y n% cat. n (9) T entry catalytic system (H2) t [h] (16) (17) ref. [mol%] [mmol] [°C] [bar] [%] [%]

1 [Ru(acac)3]/triphos 0.1/0.2 10 160 100 18 3 95 [41a]

2 [Ru(triphos)(tmm)] 2 1 140 50 16 - 98 [59b]

3 Co(BF4)2·6H2O/triphos 10 3 100 80 22 n/a 47 [56e]

4 [Ru(N-triphos)H2(PPh3)] 0.5 10 150 65 25 1 99 [56g]

[a] 5 [RuH2(PPh3)4]/N-triphos 0.5/0.75 10 160 70 18 2 98 [56j]

[b] [c] 6 [Ru(acac)3]/PBu3 0.24/8.4 10 200 ~83 7x6 37 63 [82] [a] Unsymmetrical N-triphos ligand. [b] 1200 psi. [c] The procedure was repeated seven times, reported are the total yields of GVL (16) and 1,4-PDO (17).

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

In 2008, Horváth achieved full conversion of 9 to 16 using the [Ru(acac)3] precursor in combination with a tenfold excess of PBu3 and NH4PF6 as additive after 8 h at 135 °C and 100 bar H2.[82] At longer reaction times, higher temperatures and higher ratios of catalyst/substrate and acid/substrate, this catalytic system also enabled the full conversion of 9 to 18. In the absence of NH4PF6 as acidic additive, yields of 37% for 16 and 63% for 17 could be achieved at 200 °C (entry 6).[82] In 2013,

Beller used in situ systems of [Ru(p-cymene)Cl2] with bidentate carbene ligands and KOtBu as additive. A GVL (16) yield of 80% from 9 was achieved at 100 °C and 50 bar H2 after 6 h using 1 mol% ruthenium precursor and 2 mol% ligand. Applying a different ligand enabled the conversion of 16 to 17 with 54% yield.[51d] In 2017, Jones achieved the conversion of 16 to 17 with 98% yield after 5 h at 120 °C under 55 bar H2 using 0.1 mol% of a cobalt PNP-pincer ligand.[83]

As the condensation reaction of diols to tetrahydrofurans does not essentially require highly specialized catalytic systems but can take place in the presence of various homogeneous and heterogeneous acids,[44] this work will focus on the development and tailoring of catalytic systems for hydrogenation of 9 and 16 to the corresponding diol. As ruthenium triphos based systems are up to now the most promising candidates for the homogeneously catalyzed hydrogenation of 9 to 17, the work discussed in this chapter discloses the investigations of catalytic activity of tailor-made ruthenium triphos-xyl based systems.

3.2 Application of Novel Catalytic Systems in the Hydrogenation of γ-Valerolactone and Levulinic Acid

The hydrogenation of LVA 9 and GVL 16 with different ruthenium triphos complexes was performed without any additional solvents, as all components are liquid under operating conditions. In a general reaction procedure, the catalyst was dissolved in the liquid substrate and the reaction mixture transferred into a previously evacuated stainless-steel finger autoclave equipped with a glass inlet and a magnetic stir bar under argon counter-current flow. The autoclave was pressurized with hydrogen at room temperature and heated to reaction temperature in a preheated aluminum cone fitted on a magnetic stir plate. In most cases, the corresponding pressure course was recorded using LabView. After completion, the autoclave was cooled down in an ice bath and carefully vented. The crude product solution was subsequently analyzed by GC.

To gain basic insights into the progression of the two-step hydrogenation, the reaction was performed at a low catalyst-loading while recording the pressure course (Figure 14). Under the selected reaction conditions, a temperature-dependent pressure increase can be observed at the beginning, followed by a pressure drop due to the consumption of hydrogen.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

140

120 99% GVL (16)

100

[bar]

p 60

40 LVA → GVL GVL → 1,4-PDO 9 16 16 17 20 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300

t [min]

Figure 14: Pressure course for the hydrogenation of LVA (9) using cat-II-tmm. Reaction conditions: 1 mL LVA (9), 0.17 mol% cat-II-tmm, 160 °C, 100 bar initial H2 pressure at rt, 5 h. Starting pressure was normalized to 100 bar for clarity. After ~60 min, a short pressure plateau is reached, which is then followed by a shallower decrease. The reach of an equilibrium state is indicated by a terminal pressure plateau. Performing the same experiment with a reaction time of one hour resulted in a GVL (16) yield of 99%, which reveals that full conversion of 9 to 16 has to take place before the hydrogenation of 16 to 17 begins. This observation indicates the different nature of the two sequential hydrogenation reactions and thereby highlights the challenges in performing both steps using one tailored catalytic system. Table 9: Hydrogenation of LVA (9) to 1,4-PDO (17) using cat-I-tmm, cat-II-tmm, cat-III-tmm and cat-IV-tmm.[a]

entry catalyst abbreviation catalyst structure Y (17) [%][b]

1 cat-I-tmm 91

2 cat-II-tmm 98

3 cat-III-tmm 84

4 cat-IV-tmm 85

[a] General conditions: 1 mL LVA (9), 0.17 mol% catalyst, 160 °C, 100 bar initial H2 pressure at rt, 5 h. [b] Determined by GC using 1-hexanol as internal standard. 28

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

In a first approach, [Ru(triphos)(tmm)] cat-I-tmm and [Ru(triphos-xyl)(tmm)] cat-II-tmm as well as the analogous complexes bearing propane-backbones, [Ru(pr-triphos)(tmm)] cat-III-tmm and [Ru(pr-triphos-xyl)(tmm)] cat-IV-tmm, were applied in the hydrogenation of 9. The comparison shows that cat-I-tmm and cat-II-tmm perform superior to cat-III-tmm and cat-IV-tmm, resulting in higher 1,4-PDO (17) yields, and nearly full conversion to 17 was reached using cat-II-tmm (Table 9). The difference in performance could be attributed to small changes in steric and electronic properties of the active metal center due to the elongation of the triphos backbone, but also to differences in solubility of the catalytic species in the reaction mixture.

The corresponding recorded pressure courses disclose the superior activity of cat-I-tmm and cat-III-tmm in the first hydrogenation step of 9 to 16, but a faster hydrogenation of 16 to 17 using cat-II-tmm and cat-IV-tmm (Figure 15), which can be interpreted from the steepness of the respective pressure drop areas. The initially higher activity of the two triphos complexes (cat-I-tmm, cat-III-tmm) suggests the superior performance of the corresponding active species as compared to the triphos-xyl complexes (cat-II-tmm, cat-IV-tmm). The second hydrogenation step, in turn, proceeds faster using the triphos-xyl complexes, indicating a higher activity of the catalysts in the conversion of 16 to 17. cat-I-tmm and cat-II-tmm were further compared concerning their performance in substrate reloading cycles (Figure 16). At a higher catalyst concentration, the hydrogenations of 9 were run until a pressure plateau could be observed in the recorded pressure courses, indicating full conversion to 17.

140

120

100

80 [bar]

cat-I-tmm p 60 cat-II-tmm

40 cat-III-tmm cat-IV-tmm 20 0 50 100 150 200 250 300 t [min]

Figure 15: Pressure courses for the hydrogenation of LVA (9) using cat-I-tmm, cat-II-tmm, cat-III-tmm and cat-IV-tmm. Reaction conditions: 1 mL LVA (9), 0.17 mol% catalyst, 160 °C, 100 bar initial H2 pressure at rt, 5 h. Starting pressures were normalized to 100 bar for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Figure 16: Schematic procedure for the hydrogenation of LVA (9) in two substrate loading runs.

Then, the reactions were stopped, the same amount of 9 as in the first run was added and the hydrogenations were started again. Under these conditions, both catalysts complete the reaction in similar time in the first run as can be interpreted from the almost concurrent reach of the pressure plateau after 70–75 min (Figure 17). Using cat-II-tmm, the conversion to 17 in the second run is completed after ~130 min as indicated by a pressure plateau. In contrast, the pressure drop in the second run using cat-I-tmm is significantly less steep, which indicates a much lower reaction rate. Within 225 min, no completion of the reaction could be observed as the pressure plateau was not reached suggesting a significant degree of catalyst deactivation. The results clearly show the better performance and stability of the triphos-xyl complex cat-II-tmm. It can be observed that due to the very high reaction rate at high catalyst concentrations, the pressure plateau indicating the completion of the hydrogenation of 9 to 16 could only be observed for the very slow second hydrogenation run using cat-I-tmm (at ~23 min).

[bar]

p 55

50 cat-I-tmm 1st run cat-I-tmm 2nd run 45 cat-II-tmm 1st run 0 50 100 150 200 t [min] cat-II-tmm 2nd run

Figure 17: Pressure course for two hydrogenation runs of LVA (9) to 1,4-PDO (17) with preceding substrate loadings using cat-I-tmm and cat-II-tmm. Reaction conditions: 1 mol% catalyst (1st run), 0.1 mL LVA (9) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Table 10: Hydrogenation of 9 to 17 using cat-II-tmm employing different reaction parameters.[a]

T t n (catalyst) n (LVA 9) catalyst:substrate n% catalyst entry Y (17) [%][b] [°C] [h] [µmol] [mmol] n:n [mol%]

1 160 5 16.37 9.82 1:600 0.17 98 2 160 5 12.28 9.82 1:800 0.13 90 3 160 5 9.82 9.82 1:1000 0.10 80 4 160 5 8.18 9.82 1:1200 0.08 73 5 160 5 9.82 5.88 1:600 0.17 99 6 160 5 9.82 7.84 1:800 0.13 97 7 160 5 9.82 11.76 1:1200 0.08 61 8 180 5 9.82 9.82 1:1000 0.1 88 9 200 5 9.82 9.82 1:1000 0.1 78 10[c] 120 5 9.82 9.82 1:1000 0.1 0 11 140 5 9.82 9.82 1:1000 0.1 9 12[d] 100 24 9.82 9.82 1:1000 0.1 0 13 120 24 9.82 9.82 1:1000 0.1 20 14 140 24 9.82 9.82 1:1000 0.1 97

[a] General conditions: 5.88–11.76 mmol LVA (9), 0.08–0.17 mol% cat-II-tmm, 100–200 °C, 100 bar initial H2 pressure at rt, 5–24 h. [b] Determined by GC using 1-hexanol as internal standard. [c] Y (16) = 32%. [d] Y (16) = 30%. To study the effect of different amounts of substrate, catalyst:substrate ratios, reaction temperature and reaction time on the product compositions, the corresponding parameters were varied in the hydrogenation reaction of 9 using cat-II-tmm (Table 10). As expected, a decrease of catalyst amount at constant amount of substrate and a constant temperature of 160 °C led to a decrease of 1,4-PDO (17) yield (entries 1–4). When increasing the amount of substrate at constant catalyst amount, the same trend can be observed (entries 3, 5–7). At a constant catalyst:substrate ratio of 1:1000, the reaction temperature was varied between 100 and 200 °C (entries 3, 8–14). A reaction temperature of 180 °C led to a slight increase in 1,4-PDO (17) yield as compared to 160 °C (entry 8). Increasing the temperature even further to 200 °C caused a decrease in 1,4-PDO (17) yield (entry 9), which can supposably be attributed to either a decreased catalyst stability under thermic stress or temperature dependent changes in the lactone-diol equilibrium, which will be studied more closely in chapter 3.3. At 120 °C, only 16 was formed in 32% yield after 5 h, whereas running the reaction for 24 h afforded 20% 1,4-PDO (17) yield (entries 10 and 13). At 140 °C, 9% 1,4-PDO (17)

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

yield was obtained, which could be increased to almost full conversion to 17 after 24 h (entries 11 and 14). Performing the reaction at 100 °C for 24 h led to a GVL (9) yield of only 30% without any subsequent hydrogenation to 17 (entry 12).

To compare the catalytic performances of selected [Ru(triphos-xyl)(OR)2] type catalysts and cat-II-tmm, they were first applied in the hydrogenation of 16 to 17 at different reaction times (Figure 18). After 1 h, the complexes show different activities indicated by varying 1,4-PDO (17) yields in the order cat-II-48 > cat-II-50 > cat-II-tmm ~ cat-II-46 > cat-II-45 > cat-II-47 > cat-II-44, which are most likely attributed to different time periods for the formation of the active catalytic species. The lowest yield could be observed when applying the bis(acetato) complex cat-II-44, while a slightly higher yield was achieved using the trifluorinated anlogon cat-II-45 derived from an acid with a lower pKa value. However, this trend does not proceed as complexes cat-II-46, cat-II-47, cat-II-48 and cat-II-50 all led to higher 1,4-PDO (17) yields, while the respective protonated ligands are all less acidic than 44. Surprisingly, cat-II-48 bearing a ligand coordinated by two oxygen atoms shows the highest 1,4-PDO (17) yield of 23%. Due to an expectedly stable six membered Ru–O–C3–O ring in cat-II-48, a slower activation of the complex would rather be anticipated. Therefore, the higher activity after 1 h reaction time is more likely attributed to the higher acidity of 48.

100

60 1 h

3.5 h

PDO) PDO) [%] -

40 5 h

(1,4

Y 20

cat-II- -tmm -44 -45 -46 -47 -48 -49 -50

Figure 18: Hydrogenation of GVL (16) to 1,4-PDO (17) using cat-II-tmm and cat-II-44–cat-II-50. General conditions: 1 mL GVL (16), 0.1 mol% catalyst, 160 °C, 70 bar initial H2 at rt. 1,4-PDO (17) yield determined by GC using 1-hexanol as internal standard. 32

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

The OR-type complexes with aromatic moieties reflect the trend of 1,4-PDO (17) yields, correlated with the basicity of the ligands. The second highest 1,4-PDO (17) yield was achieved using cat-II-50 with 50 having a higher pKa value than 46 and 47. After 3.5 h, cat-II-44 and cat-II-45 led to significantly lower 1,4-PDO (17) yields than all other complexes, which in turn generally follow the trend observed after 1 h reaction time, but with less significant variation. This observation supports the assumption that the differences in activities observed after 1 h can be attributed to variable induction periods required for the formation of the catalytically active species. As anticipated by the previous pressure courses, the data show a significantly lower additional conversion between 3.5 and 5 h as compared to shorter reaction times. After 5 h, cat-II-44, cat-II-47 and cat-II-48 provided slightly lower yields than after 3.5 h, most likely to the error of reproducibility. The formation of 17 using cat-II-44 is slowed down while cat-II-45 led to a significantly higher yield, which indicates the completion of the catalyst activation between 3.5 and 5 h, followed by an increased reaction rate. A slight increase in yield to 85% and 90% could be observed using cat-II-50 and cat-II-46, respectively. A 1,4-PDO (17) yield of only 21% after 5 h was achieved using cat-II-49 indicating a challenging activation of the very stable complex under the applied reaction conditions. Consequently, the hydrogenation of 16 to 17 using cat-II-49 was not studied at shorter reaction times.

100

60 1,4-PDO [%] GVL Y 40

0 1 3.5 5 1 3.5 5 1 3.5 5 1 3.5 5 1 3.5 5 1 3.5 5 1 3.5 5 1 3.5 5 t [h]

cat-II- -tmm -44 -45 -46 -48 -47 -49 -50 Figure 19: Hydrogenation of LVA (9) to 1,4-PDO (17) using cat-II-tmm and cat-II-44–cat-II-50. General conditions: 1 mL LVA (9), 0.1 mol% catalyst, 160 °C, 100 bar H2. GVL (16) and 1,4-PDO (17) yields determined by GC using 1-hexanol as internal standard. 33

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Using the applied complexes in the hydrogenation of 9 led to an incomplete conversion to 16 after 1 h in most cases, while consequently no formation of 17 could be observed (Figure 19). With cat-II-45 and cat-II-49, already full conversion of 9 to 16 and a 1,4-PDO (17) yield of 2–9% was obtained. The other complexes show different activities indicated by varying GVL (16) yields in the order cat-II-50 > cat-II-48 > cat-II-44 ~ cat-II-tmm > cat-II-46 > cat-II-47. The lowest activity was observed for the two bis(naphtholato) complexes obtained from naphthols of comparable pKa values. In contrast to the hydrogenation of 16 to 17, the bis(acetato) complex cat-II-44 did not show a significant decrease in activity under the applied reaction conditions and shows a similar activity as cat-II-tmm. The hydrogenation of 16 to 17 at short reaction time showed that cat-II-48 and cat-II-50 provide the highest activities (Figure 18). However, starting from 9, cat-II-50 performs superior (Figure 19). This is likely attributed to a longer induction period of cat-II-48 bearing the previously discussed stable six membered ring. Interestingly, after a longer reaction time of 3.5 h, complete conversion of 9 to 16 was achieved in all cases (Figure 19). The observed activities for the formation of 17 are mostly in alignment with the trend observed for the formation of 16 after 1 h with a significant approximation of the performance of cat-II-45 to the other catalysts. After 5 h reaction time, cat-II-44, cat-II-45, cat-II-46, cat-II-47 and cat-II-49 led to very similar 1,4-PDO (17) yields of 74–76%. This result is in contrast to the observation made for the hydrogenation of 16 to 17 using cat-II-49 and indicates a highly improved activation of the complex using the acidic substrate 9. The observed differences in activity of the employed catalysts are less prominent for the two-step reaction of 9 to 17 than in the hydrogenation of 16, which indicates a non-similar influence of the different ligands in the two-step reaction starting from the acidic substrate 9.

Figure 20: Schematic procedure for the hydrogenation of GVL (16) in three substrate loading and hydrogenation runs. 34

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

The reactivity of complexes cat-II-tmm and cat-II-44–cat-II-50 was also studied in multiple substrate loading and hydrogenation runs for the conversion of 9 and 16 to 17. At a higher catalyst concentration, the hydrogenations were run until a plateau, indicating full conversion to 17. Then, the reactions were stopped, the autoclaves were cooled in an ice bath and depressurized. Fresh substrate was added under argon counter-current flow and the hydrogenations were started again. One to three substrate loading and hydrogenation runs were performed using 16 as substrate (Figure 20).

The slight differences in hydrogen pressure at the pressure plateaus can be attributed to different starting pressures or variations in the amounts of reaction mixture components. In the first run, significant differences in the performance of the catalysts for the hydrogenation of 16 to 17 could be observed (Figure 21). The recorded pressure courses show different activities indicated by the time required to reach a pressure plateau in the order cat-II-50 > cat-II-46 ~ cat-II-tmm ~ cat-II-47 > cat-II-48 > cat-II-45 > cat-II-44 >> cat-II-49. The results are in general alignment with the trend observed for the performed batch experiments with short reaction time (Figure 18). As expected from the obtained results for the batch experiments, no decrease in pressure and consequently no conversion of 16 was observed using cat-II-49 as catalyst.

75 cat-II-tmm 70 cat-II-44

cat-II-45

[bar]

p cat-II-46 55 cat-II-47 50 cat-II-48 45 cat-II-49 0 50 100 t [min] cat-II-50

Figure 21: Pressure course for the hydrogenation of 1st load of GVL (16) to 1,4-PDO (17) using cat-II-tmm and cat-II-44– cat-II-50. Reaction conditions: 0.1 mL GVL (16), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

75 cat-II-tmm 70 cat-II-44

60 cat-II-45

[bar]

p 55 cat-II-46

50 cat-II-47

45 cat-II-48 0 100 200 t [min] cat-II-50

Figure 22: Pressure course for the hydrogenation of 2nd load of GVL (16) to 1,4-PDO (17) using cat-II-tmm, cat-II-44– cat-II-48 and cat-II-50. Reaction conditions: 0.1 mL GVL (16), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

Hence, no further substrate loading and hydrogenation runs were performed using this catalyst. cat-II-45 shows a faster reaction rate in the hydrogenation of 16 than its non-fluorinated analogon cat-II-44. All other complexes provide higher activities than cat-II-45. As suggested earlier, the significantly prolonged reaction time until the beginning of the pressure drop using cat-II-44 is likely attributed to a longer induction period of the catalyst. Using cat-II-50, the shortest reaction time was required to reach a pressure plateau after only ~20 min.

In the second substrate loading run of 16, longer reaction times were required for the approximation of a pressure plateau in all cases (Figure 22).

70 cat-II-tmm

65 cat-II-44

[bar] cat-II-46

p 55 cat-II-47 50 cat-II-48 45 0 100 200 300 400 cat-II-50 t [min]

Figure 23: Pressure course for the hydrogenation of 3rd load of GVL (16) to 1,4-PDO (17) using cat-II-tmm, cat-II-44– cat-II-48 and cat-II-50. Reaction conditions: 0.1 mL GVL (16), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Similar to the first run, cat-II-50 shows the highest activity indicated by the steepest pressure drop and the shortest reaction time of ~30 min required to reach the pressure plateau. cat-II-44, cat-II- 46, cat-II-47 and cat-II-48 perform very similar leading to pressure plateaus after ~62 min. This observation also supports the previous assumption of a lower activity of cat-II-44 due to a longer induction period. Lower activities could be observed using cat-II-tmm and cat-II-45, while the latter shows a significantly slower reaction rate indicated by a shallower pressure drop. Consequently, no third run using cat-II-45 will be discussed. As before, longer reaction times are required to reach a pressure plateau in all cases in the third run (Figure 23), which is most likely attributed to the earlier mentioned factors. No significant differences in the activities of cat-II-44– cat-II-48 and cat-II-50 could be observed, while cat-II-tmm showed a lower reaction rate indicated by a slightly shallower pressure drop. This observation indicates an enhancing effect of the respective introduced OR-type ligands on the activity of the catalytic systems. The most significant loss in activity was observed for cat-II-50 from the second to the third hydrogenation run.

As can be observed by the recorded pressure course, the second cycle using cat-II-50 was run for another 45 min after the pressure plateau had been reached. To investigate the influence of additional stress on the catalyst by prolonged upkeep of the reaction conditions after completion of the hydrogenation, the procedure was repeated with a shorter reaction time for the first or second hydrogenation run. The recorded pressure courses clearly show the influence of prolonged upkeep of the reaction conditions after full conversion of 16 to 17 on the catalyst activity.

[bar] p p 55 3rd run 50 2nd run 1st run 45 0 100 200 300 400 500 t [min]

Figure 24: Pressure courses for three hydrogenation runs of GVL (16) to 1,4-PDO (17) with preceding substrate loadings and prolonged 2nd hydrogenation run using cat-II-50. Reaction conditions: 1 mol% cat-II-50 (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

[bar]

p 55 3rd run 2nd run 50 1st run 45 0 50 100 150 t [min]

Figure 25: Pressure courses for three hydrogenation runs of GVL (16) to 1,4-PDO (17) with preceding substrate loadings and prolonged 1st hydrogenation run using cat-II-50. Reaction conditions: 1 mol% cat-II-50 (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

Interruption of the first hydrogenation run directly after the reach of a pressure plateau (Figure 24) enhances the catalyst activity in the second hydrogenation run, as the respective pressure plateau was reached after ~30 min as compared to ~65 min when allowing a prolonged first run (Figure 25). Analogously, stopping the second hydrogenation run immediately after the reach of a pressure plateau (Figure 25) led to a significantly shorter reaction time of ~64 min to reach the pressure plateau in the third run as compared to ~380 min when allowing a prolonged second run (Figure 24).

The respective 31P{1H}-NMR spectra of the postreaction mixtures recorded after the third hydrogenation run clearly reveal the formation of [Ru(triphos-xyl)(CO)H2] cat-II-COH2 generating a triplet at 33.2 / 31.4 ppm and a doublet at 25.4 / 23.5 ppm with a coupling constant of 31.8 Hz (Figure 26).[68b,69a] As ruthenium triphos monocarbonyl dihydrido species can be generated by decarbonylation of esters and alcohols and were found to be inactive in the absence of acids, the increased formation of cat-II-COH2 in case of the prolonged second run supports the interpretations described above.[41a,55] The observed effects show a significantly enhanced catalyst deactivation under prolonged stress, resulting from the applied reaction conditions in the absence of substrate. It can therefore be concluded that the activity of the catalyst can be significantly maintained, when reducing additional time after the completion of the hydrogenation. A promoting effect would also be expected under continuous addition of substrate to the reaction mixture to circumvent the catalyst deactivation.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

P 2 P

Figure 26: 31P{1H}-NMR spectra after the hydrogenation of the third load of GVL (16) using cat-II-50. Reaction conditions: 1 mol% cat-II-50 (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure at rt. Measured in THF-d8 at rt. Upper spectrum: with prolonged second run. Lower spectrum: with prolonged first run.

The effect of treating the catalyst under reaction conditions in the absence of hydrogen was studied using cat-II-48. After the second hydrogenation run, cooling to room temperature and depressurizing the autoclave, the reaction mixture was stirred for 30 min at reaction temperature in the absence of hydrogen before conducting the third substrate loading and hydrogenation run (Figure 27). The very shallow pressure drop for the third hydrogenation run shows a similar course as the third run using cat-II-50 after a prolonged 2nd run (Figure 28).

Figure 27: Schematic procedure for the hydrogenation of GVL (16) in three substrate loading and hydrogenation runs with additional stirring of the reaction mixture for 30 min at 160 °C without hydrogen after the 2nd hydrogenation run.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

[bar]

p 55 3rd run 2nd run 50 1st run 45 0 100 200 300 400 t [min]

Figure 28: Pressure courses for three hydrogenation runs of GVL (16) to 1,4-PDO (17) with preceding substrate loadings using cat-II-48 and additional stirring of the reaction mixture for 30 min at 160 °C without hydrogen after the 2nd hydrogenation run. Reaction conditions: 1 mol% cat-II-48 (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

Consequently, this result suggests that the increasing catalyst deactivation under reaction conditions in the absence of substrate can be primarily assigned to thermal stress while the presence of hydrogen does not significantly contribute to the decreased activity, which is in alignment with the previous observation of catalyst deactivation by increased formation of inactive carbonyl complexes.

The performance of the catalysts in multiple substrate loading and hydrogenation runs was also investigated for the conversion of 9 to 17 (Figure 29). Due to the very high reaction rate at high catalyst concentrations, the pressure plateau indicating the completion of the hydrogenation of 9 to 16 could not be observed for the first hydrogenation run (Figure 30). The recorded pressure courses show slightly different activities indicated by the time required to reach a pressure plateau in the order cat-II-50 > cat-II-48 ~ cat-II-tmm ~ cat-II-47 > cat-II-44 > cat-II-46 > cat-II-45 >> cat-II-49.

Figure 29: Schematic procedure for the hydrogenation of LVA (9) in two substrate loading and hydrogenation runs.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

70 cat-II-tmm

65 cat-II-44

60 cat-II-45

[bar]

p cat-II-46 55 cat-II-47 50 cat-II-48 45 0 50 100 cat-II-49 t [min] cat-II-50

Figure 30: Pressure course for the hydrogenation of 1st load of LVA (9) to 1,4-PDO (17) using cat-II-tmm and cat-II-44– cat-II-50. Reaction conditions: 0.1 mL LVA (9), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

Only a marginal pressure drop could be observed using cat-II-49 and the pressure plateau could not be reached within 170 min. This observation is contrary to the results obtained at lower catalyst concentration in the batch experiments (Figure 19). Consequently, no further substrate loading and hydrogenation runs were performed using this catalyst. While cat-II-50 shows the highest activity reaching the plateau in the shortest reaction time of ~74 min, cat-II-45 led to a plateau after ~110 min. Using the other complexes, reaction times between 82 and 93 min were required to reach the equilibrium state.

75 cat-II-tmm 70 cat-II-44

cat-II-45

[bar]

p cat-II-46 55 cat-II-47 50 cat-II-48 45 0 50 100 150 200 250 cat-II-50 t [min]

Figure 31: Pressure course for the hydrogenation of 2nd load of LVA (9) to 1,4-PDO (17) using cat-II-tmm, cat-II-44– cat-II-48 and cat-II-50. Reaction conditions: 0.1 mL LVA (9), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

In the second hydrogenation run, the corresponding pressure courses show very similar activities for complexes cat-II-tmm, cat-II-44, cat-II-46, cat-II-47 and cat-II-48, while cat-II-45 and cat-II-50 resulted in significantly longer reaction times (Figure 31). While the performance of cat-II-45 is in accordance with previous observations, the loss in activity of cat-II-50 is rather surprising. It cannot be ruled out that the complex would show higher activity in the second run if treated shorter with the applied reaction conditions after full conversion of 9 in the first run as it was disclosed for the hydrogenation of 16.

The hydrogenation of 9 and 16 was also performed using in situ systems with a selection of non- fluorinated phenol and naphthol additives that could not be converted with cat-II-tmm to the corresponding isolated complexes (Chapter 2). For the conversion of 16 to 17, the addition of two equivalents of 5,6,7,8-tetrahydro-1-naphthol 38 or -2-naphthol 39 led to similar activities in the first hydrogenation run, but significantly decreased activities in the second run as compared the isolated bis(naphtholato) complexes cat-II-46 and cat-II-47 (Figure 32). The hydrogenation of 9 to 17 using two equivalents of 38 or 39 also revealed a significantly lower catalyst activity in the second run as compared to cat-II-46 and cat-II-47 (Figure 33). An approach to the pressure plateau could only be observed after over 200 min in both cases.

[bar]

p 55

50 cat-II-tmm / 38 1st run 45 cat-II-tmm / 38 2nd run 0 200 400 cat-II-tmm / 39 1st run t [min] cat-II-tmm / 39 2nd run

Figure 32: Pressure courses for two hydrogenation runs of GVL (16) to 1,4-PDO (17) with preceding substrate loadings using cat-II-tmm/38 or cat-II-tmm/39. Reaction conditions: 1 mol% cat-II-tmm, 2 mol% additive (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

[bar]

p 55

50 cat-II-tmm / 38 1st run 45 cat-II-tmm / 38 2nd run 0 200 400 cat-II-tmm / 39 1st run t [min] cat-II-tmm / 39 2nd run

Figure 33: Pressure courses for two hydrogenation runs of LVA (9) to 1,4-PDO (17) using cat-II-tmm/38 or cat-II-tmm/39. Reaction conditions: 1 mol% cat-II-tmm, additive (1st run), 0.1 mL LVA (9) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

While the addition of two equivalents of 1,3,5-trihydroxybenzene 28 was found to suppress the hydrogenation of 16 as no significant pressure drop could be observed, performing the hydrogenation with two equivalents of 3,5-dimethoxyphenol 37 led to similar results as using the isolated complexes cat-II-44 and cat-II-46–cat-II-48 even after three substrate loading and hydrogenation runs (Figure 34).

60 p[bar] 55

50 cat-II-tmm / 28 1st run cat-II-tmm / 37 3rd run 45 0 100 200 300 cat-II-tmm / 37 2nd run t [min] cat-II-tmm / 37 1st run

Figure 34: Pressure courses for one to three hydrogenation runs for GVL (16) to 1,4-PDO (17) with preceding substrate loadings using cat-II-tmm/28 or cat-II-tmm/37. Reaction conditions: 1 mol% cat-II-tmm, 2 mol% additive (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

cat-II-51 70 cat-II-52

65 cat-II-53

cat-II-54

[bar]

p cat-II-55 55 cat-II-56

50 cat-II-57

cat-II-58 45 0 20 40 60 cat-II-59 t [min]

Figure 35: Pressure course for the hydrogenation of 1st load of GVL (16) to 1,4-PDO (17) using cat-II-51–cat-II-59. Reaction conditions: 0.1 mL GVL (16), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

The hydrogenation of 16 to 17 in multiple substrate loading and hydrogenation runs was also performed using the bis(fluorophenolato) and bis(fluoronaphtholato) complexes cat-II-51– cat-II-59. Already in the first hydrogenation run, significant differences in the performance of the applied catalysts could be observed. The shortest reaction times until the approach of the pressure plateau were obtained using cat-II-58 and cat-II-59 bearing perfluorinated ligands (Figure 35). Whereas cat-II-53–cat-II-56 showed similar activities, significantly lower reaction velocities were observed using the bis(tetrafluorophenolato) complex cat-II-57, the bis(monofluorophenolato) complex cat-II-51 and the 2,5-substituted bis(difluorophenolato) complex cat-II-52.

The second hydrogenation run revealed mostly similar trends with more assimilated performances of complexes cat-II-54–cat-II-57 and cat-II-59, while cat-II-58 still showed the highest activity reaching the pressure plateau after ~37 min and cat-II-51 the lowest activity with a pressure plateau after ~100 min (Figure 36). cat-II-52 and cat-II-53 also showed comparable activities to cat-II-51.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

cat-II-51 70

cat-II-52 65

cat-II-53

[bar] cat-II-54

cat-II-55 55

cat-II-56 50 cat-II-57

45 cat-II-58 0 50 100 t [min] cat-II-59

Figure 36: Pressure course for the hydrogenation of 2nd load of GVL (16) to 1,4-PDO (17) using cat-II-51–cat-II-59. Reaction conditions: 0.1 mL GVL (16), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

In the third hydrogenation run, the inferior performance of complexes cat-II-51–cat-II-53 was found to be even more significant (Figure 37). The catalysts cat-II-54–cat-II-59 showed fairly similar activities reaching the pressure plateaus between 48–87 min after comparably steep pressure drops. The significantly shallower approach to a plateau using cat-II-51–cat-II-53 could only be observed after over 200 min. Except for these complexes, generally higher activities of the fluorophenolato and -naphtholato complexes were observed compared to the non-fluorinated phenolato, naphtholato, acetato and fluoroacetato complexes. This observation is most significant in the third hydrogenation run (Figure 31 and Figure 37). According to the obtained results, the performances of cat-II-51–cat-II-59 seem to be dependent on the degree of fluoro-substitution and the substitution pattern. While the perfluorinated complexes show the highest activities, the monosubstituted bis(fluorophenolato) complex cat-II-51 and the 2,5- and 2,6-substituted bis(fluorophenolato) complexes cat-II-52 and cat-II-53 are significantly less active under the applied reaction conditions. However, the activity of cat-II-54 bearing 3,5-substituted ligands is significantly higher as compared to cat-II-52 and cat-II-53.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

cat-II-51 70 cat-II-52

cat-II-53

60 cat-II-54

[bar]

p cat-II-55 55 cat-II-56 50 cat-II-57

45 cat-II-58 0 50 100 150 200 250 t [min] cat-II-59

Figure 37: Pressure course for the hydrogenation of the 3rd load of GVL (16) to 1,4-PDO (17) using cat-II-51–cat-II-59. Reaction conditions: 0.1 mL GVL (16), 1 mol% catalyst, 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

As anticipated, the 31P{1H}-NMR spectra recorded after the third hydrogenation run reveal a general decrease of the amount of formed monocarbonyl dihydrido complex cat-II-COH2 (♦) and increase of amount of dicarbonyl monohydrido complex cat-II-CO2H (♠) with decreasing pKa value of the corresponding fluorophenol or •naphthol (Figure 38). This observation consequently indicates the presence of free phenol or naphthol during the hydrogenation reaction inducing different degrees of acidity to the reaction mixture.

♠ ♠

(♦) ♦ ♦

(♠)

Figure 38: 31P{1H}-NMR spectra of the product solution after the 3rd hydrogenation run of GVL (16) to 1,4-PDO (17) using cat-II-51–cat-II-59 (in ascending order) measured at rt in THF-d8. 46

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

75 75

70 70 3rd 3rd

65 run 65 run

2nd 2nd

60 60 [bar]

[bar] run run

p p 55 1st 55 1st run run 50 50

45 45 0 50 100 0 50 t [min] t [min]

Figure 39: Pressure courses for three hydrogenation runs of GVL (16) to 1,4-PDO (17) with preceding substrate loadings using cat-II-58 with prolonged 2nd (left) or prolonged 1st (right) hydrogenation run. Reaction conditions: 1 mol% cat-II-58 (1st run), 0.1 mL GVL (16) each run, 160 °C, 50 bar initial H2 pressure. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

Analogously to the non-fluorinated bis(phenolato) complex cat-II-50, the activity of cat-II-58 in multiple substrate loading and hydrogenation runs of 16 to 17 was also decreased when running the reaction for longer time after the full conversion of 16 in the first or second run (Figure 39). However, the overall activity of cat-II-58 was found to be higher and the dependency of the activity in the third run on the reaction time of the second run less significant.

75 160 °C 3rd run 160 °C 2nd run 70

160 °C 1st run 65

180 °C 3rd run

[bar]

p 60 180 °C 2nd run 55 180 °C 1st run 50 200 °C 3rd run 45 200 °C 2nd run 0 20 40 60 80 100 200 °C 1st run t [min]

Figure 40: Pressure course for multiple hydrogenation runs of GVL (16) to 1,4-PDO (17) with preceding substrate loadings using cat-II-58 at different reaction temperatures. Reaction conditions: 0.1 mL GVL (16), 1 mol% cat-II-58 (1st run), 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

The reaction time required to reach the pressure plateau in the third run using cat-II-58 was increased from ~48 to ~73 min when prolonging the reaction time of the second run, which is still a significantly lower value than observed using cat-II-50 (Figure 24).

To gain a deeper insight into the influence of thermal stress on the catalytic performance, the hydrogenation of 16 to 17 was carried out at temperatures from 160–200 °C in multiple substrate loading and hydrogenation runs using cat-II-54 and cat-II-58. As similar observations were made in both cases, only the results using cat-II-58 will be discussed. The recorded pressure courses clearly show the influence of thermal stress on the catalytic performance (Figure 40). Whereas the pressure plateaus in the first and second hydrogenation runs at 180 °C and 200 °C are reached earlier than at 160 °C indicating an increasing activity with increasing reaction temperature, this trend does not proceed for the third run. Here, significant losses in activity at 180 °C and 200 °C could be observed indicated by shallower pressure drops and significantly longer reaction times.

To investigate whether the same observations can be observed using 9 as substrate, the performance of cat-II-58 was studied in the hydrogenation of 9 to 17 in multiple substrate loading and hydrogenation runs at different reaction temperatures. Analogously to the observations made for the hydrogenation of 16 to 17, an increase in temperature led to an initial improvement of the catalytic activity using cat-II-58 in the hydrogenation of 9 (Figure 41). As the activity at 160 °C was found to be significantly lower, no third run was conducted in this case. The recorded pressure courses also show the significant influence of thermal stress on the catalytic activity indicated by slowed down reaction velocities in the third cycle at 180 °C and 200 °C.

75 160 °C 2nd run

70 160 °C 1st run

65 180 °C 3rd run [bar]

180 °C 2nd run

p 60 180 °C 1st run 55 200 °C 3rd run 50 200 °C 2nd run 45 200 °C 1st run 0 50 100 150 t [min]

Figure 41: Pressure course for multiple hydrogenation runs of LVA (9) to 1,4-PDO (17) with preceding substrate loadings using cat-II-58 at different reaction temperatures. Reaction conditions: 0.1 mL LVA (9), 1 mol% cat-II-58 (1st run), 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

[bar]

p 55 cat-II-47 1st run 50 cat-II-47 2nd run cat-II-59 1st run 45 0 50 100 150 200 cat-II-59 2nd run t [min]

Figure 42: Pressure course for two hydrogenation runs of LVA (9) to 1,4-PDO (17) with preceding substrate loadings using cat-II-47 and cat-II-59. Reaction conditions: 0.1 mL LVA (9), 1 mol% catalyst (1st run), 160 °C, 50 bar initial H2 pressure at rt. Starting pressures were normalized to 50 bar and selected courses were shortened for clarity.

The bis(heptafluoro-2-naphtholato) complex cat-II-59 was compared to its non-fluorinated analogon cat-II-47 in the hydrogenation of 9 to 17 in two runs to investigate whether the perfluoro- substitution also significantly enhances the catalytic performance for this substrate. Surprisingly, a marginally higher activity of the fluorinated complex in the first hydrogenation run but a significantly lower activity in the second run could be observed from the corresponding pressure courses (Figure 42). This result once again emphasizes the differences of the catalytic reactions using 9 or 16 as substrate and the associated different influences of the applied catalytic systems.

Table 11: Hydrogenation of LVA (9) to 1,4-PDO (17) using selected catalytic systems.[a]

entry [Ru] ligand additive Y (17) [%][b]

1 [Ru(cod)(methylallyl)2] triphos-xyl - 78

2 [Ru(cod)(methylallyl)2] triphos-xyl 50 80 3 cat-II-tmm - - 80 4 cat-II-tmm - 50 76 5 cat-II-50 - - 79 6 cat-II-50 - 50 79

[a] General conditions: 1 mL LVA (9), 0.1 mol% catalyst, 0.1 mol% triphos-xyl, 0.2 mol% additive, 160 °C, 100 bar initial H2 pressure at rt, 5 h. [b] Determined by GC using 1-hexanol as internal standard.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

The employment of 9 introduces an acidic reaction environment until full conversion of 9 has taken place. Additionally, water is formed as a coupling product in significant amounts during the conversion of 9 to 16, which does not occur when starting directly from 16.

To compare the performance of the investigated complexes with the corresponding in situ systems, the hydrogenations of 9 and 16 to 17 were performed with different catalytic systems composed of ruthenium precursors, ligands and additives addressing complexes cat-II-tmm and cat-II-50. No significant differences in the yields of 17 were observed when applying the different catalytic systems in the hydrogenation of 9 (Table 11). The successful application of cat-II-50 in the hydrogenation of levulinic acid in larger scale and with subsequent catalyst recycling was recently demonstrated in a study by Pascal Albrecht.[84] The results for the hydrogenation of 16 reveal a significant influence of the applied systems on the achieved yield (Table 12). The in situ systems using [Ru(cod)(methylallyl)2] corresponding to cat-II-tmm and cat-II-50 showed significantly lower 1,4-PDO (17) yields of 14% and 22% (entries 1 and 3) as compared to the isolated complexes providing 78% and 79% 17 (entries 5 and 8).

Table 12: Hydrogenation of GVL (16) to 1,4-PDO (17) using selected catalytic systems.[a]

entry [Ru] ligand additive Y (17) [%][b]

1 [Ru(cod)(methylallyl)2] triphos-xyl - 14

[c] 2 [Ru(cod)(methylallyl)2] triphos-xyl - 9

3 [Ru(cod)(methylallyl)2] triphos-xyl 50 22

[c] 4 [Ru(cod)(methylallyl)2] triphos-xyl 50 16 5 cat-II-tmm - - 78 6 cat-II-tmm - 50 73 7[d] cat-II-tmm - 50 61 8 cat-II-50 - - 79 9 cat-II-50 - 50 75 10[d] cat-II-50 - 50 71

[a] General conditions: 1 mL GVL (16), 0.1 mol% catalyst, 0.1 mol% triphos-xyl, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 3.5 h. [b] Determined by GC using 1-hexanol as internal standard. [c] Components of the catalytic systems were mixed 3 weeks prior to application. [d] 0.1 mol% 50.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

When premixing the components three weeks prior to application, slightly lower 1,4-PDO (17) yields were achieved (entries 2 and 4), which is an important observation to consider when designing the catalyst storage. The in situ systems of cat-II-tmm with one or two equivalents of 50 also showed lower yields for 17 than the isolated complex cat-II-50 (entries 6 and 7). Adding one or two equivalents of 50 to cat-II-50 also decreased the resulting 1,4-PDO (17) yield (entries 9 and 10). The lower performance using additional 50 to cat-II-tmm and cat-II-50 can likely be attributed to a hindrance of substrate coordination to the active metal center in the presence of additional ligand.

The results demonstrate that the use of molecularly defined complexes significantly enhances the catalytic performance in the hydrogenation of 16 to 17. The hydrogenation of 9 to 17 in turn can be performed with both in situ systems and isolated complexes with similar results. The inferior performance of the in situ systems in the hydrogenation of 16 suggests slower or incomplete catalyst activation under the applied conditions, while the hydrogenation of 9 using in situ systems led to similar performances and reactivity compared to isolated systems.

3.3 Mechanistic Insights into the Hydrogenation of γ-Valerolactone and Levulinic Acid using Ruthenium Triphos-xyl Complexes

The presented results raise several questions on details of the hydrogenation of LVA (9) and GVL (16) to 1,4-PDO (17) using the applied cat-II- complexes, which will be investigated in detail in the following part.

The recorded pressure courses of the hydrogenation of 16 to 17 reveal a significant flattening of the pressure drop towards the end of the reaction indicating a successive slowdown in reaction rate. This could be attributed to a dynamic equilibrium between 16 and 17. The dehydrogenation of 17 to 16 was achieved before, using homogeneous and heterogeneous catalytic system.[85] In 2012, Tobias Weigand reported the successful formation of 16 from 17 in a NMR spectroscopic experiment using a ruthenium complex bearing a triphos derivative ligand.[86]

The dehydrogenation of 17 to 16 was studied in a batch experiment, stirring 1 mL of 17 and 0.1 mol% cat-II-tmm or cat-II-50 at 160 °C in a stainless steel 10 mL finger autoclave for 5 h. The crude product mixture was analyzed by gas chromatography and the reaction was followed by recording the corresponding pressure course.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Table 13: Dehydrogenation of 1,4-PDO (17) to GVL (16) using cat-II-tmm and cat-II-50.[a]

catalyst catalyst structure Y (16) [%][b]

cat-II-tmm 22

cat-II-50 23

[a] General conditions: 1 mL 1,4-PDO (17), 0.1 mol% catalyst, 160 °C, 5 h. [b] Determined by GC using 1-hexanol as internal standard.

The gas chromatographic analysis reveals similar activities of the two catalysts in the dehydrogenation of 17, leading to GVL (16) yields of over 20% with a slightly higher value obtained using cat-II-tmm (Table 13). The corresponding pressure courses reveal a short period of shallow pressure increase until ~6 min (Figure 43), which can be attributed to the time required for the heating-up period to the targeted reaction temperature in the applied reaction setup. Until ~31 min, a steep rise in pressure can be observed, which is then followed by a shallower further increase. Nevertheless, the obtained results clearly show the high activities of the utilized catalysts in the dehydrogenation of 17 to 16, supporting the corresponding assumptions disclosed in chapter 3.2.

18 16 14

10 [bar]

8 cat-II-tmm p 6 cat-II-50 4 2 0 0 50 100 150 200 250 300 t [min]

Figure 43: Pressure course for the dehydrogenation of 1,4-PDO (17) to GVL (16). Reaction conditions: 1 mL 1,4-PDO (17), 0.1 mol% catalyst, 160 °C, 5 h.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

Figure 44: 31P{1H}-NMR spectrum of the reaction mixture obtained by dissolving cat-II-tmm in LVA (9) measured using a DMSO-d6 capillary as external NMR solvent at rt.

In 2011, Leitner and Klankermayer studied the mechanism of the hydrogenation of 9 using the in situ system cat-I-acac.[55] The described DFT study was performed starting with [Ru(triphos)H]+ as cationic fragment, which was presumed as the common active species under acidic conditions. To obtain a closer insight on the formation of possible precursors prior to the start of the hydrogenation, different NMR spectroscopic experiments were performed. Dissolving cat-II-tmm in 9 at room temperature and subsequent measurement of the solution led to the generation of a singlet at 40.6 ppm in the corresponding 31P{1H}-NMR spectrum (Figure 44). The same behavior was observed for complexes cat-II-44–cat-II-48 and cat-II-50. These results suggest the formation of a levulinic acid derived complex at the stage of premixing of the reaction components before transferring the mixture into the autoclave reactors.

Consequently, attempts were made to isolate this complex by stoichiometric conversion of cat-II-tmm with 9. By stirring two equivalents of 9 with cat-II-tmm in toluene at 50 °C overnight, subsequent removal of the solvent in vacuo, washing of the obtained yellow residue with n-pentane and drying in vacuo, a yellow powder was obtained. By NMR spectroscopic analyses, the resulting complex could be identified as the bis(levulinato) complex [Ru(triphos-xyl)(OOCCH2CH2COCH3)2] cat-II-9.

Scheme 3: Formation of cat-II-9 from ruthenium triphos-xyl complexes in LVA (9).

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

It can therefore be stated, that cat-II-9 is formed by dissolving the applied triphos-xyl complexes in 9 at room temperature and represents the precursor in the hydrogenation of 9 (Scheme 3).

The 1H-NMR spectrum after the conversion of cat-II-50 with two equivalents of 9 reveals the presence of four OH groups that can be assigned to free 50. Consequently, cat-II-9 is formed by the protonation of the two phenolato ligands by 9. It could also be observed that no mono-levulinato species was formed when converting cat-II-50 with one equivalent of 9, but the reaction solution contained a mixture of cat-II-50 and cat-II-9.

To obtain information on the precursor of the hydrogenation of 16, the utilized catalysts were dissolved in 16 at room temperature and the corresponding 31P{1H}-NMR spectra were recorded. Dissolving cat-II-45 in 16 led to the generation of a singlet at 29.2 ppm (Figure 45). Using cat-II-tmm, cat-II-44, cat-II-46, cat-II-47 or cat-II-50 resulted in the generation of an additional broad signal at ~39 ppm which might be attributed to a complex with 16 coordinated to the metal center in a ring-opened form due to the similar chemical shift as cat-II-9. Dissolving cat-II-48 in 16 led to the selective generation of this signal. Unfortunately, the described species could not be characterized as their isolation from the reaction mixture was unsuccessful and they could not be generated by using stoichiometric amounts of 16. However, it can be stated that the applied complexes undergo a conversion with 16 in the preparation of the reaction solution prior to the performance of the hydrogenation reaction.

As the in situ systems were found to perform inferior to the isolated complexes in the hydrogenation of 16 (Chapter 3.2), 31P{1H}-NMR spectroscopic studies were performed to investigate whether the application of these in situ systems leads to the formation of inactive species. Therefore, mixtures of catalyst precursors and ligands were heated to 160 °C in 16.

cat-II-45

cat-II-50

cat-II-48

Figure 45: 31P{1H}-NMR spectra of the reaction mixture obtained by dissolving cat-II-48, cat-II-50 or cat-II-45 in GVL (16) measured using DMSO-d6 in a capillary as external NMR-solvent at rt.

3 Hydrogenation of Levulinic Acid and γ-Valerolactone using Tailored Ruthenium Triphos-xyl Complexes

cat-II-tmm/ 2 eq 50

[Ru(cod)(methylallyl)2]/triphos-xyl/ 2 eq 50

[Ru(cod)(methylallyl)2]/triphos-xyl

Figure 46: 31P{1H}-NMR spectra of the reaction mixture obtained by heating [Ru(cod)(methylallyl)2]/triphos-xyl, [Ru(cod)(methylallyl)2]/triphos-xyl/2 eq 50 or cat-II-tmm/2 eq 50 at 160 °C in GVL (16) measured using DMSO-d6 in a capillary as external NMR-solvent at rt.

In all three cases, both a broad signal at ~39 ppm and a sharp singlet at ~28 ppm were generated in the 31P{1H}-NMR spectrum (Figure 46). The intensity of the singlet decreases in the order

[Ru(cod)(methylallyl)2/triphos-xyl > [Ru(cod)(methylallyl)2]/triphos-xyl/2 eq 50 > cat-II-tmm/2 eq 50. In all cases, both previously observed signals could be observed, which suggests that similar species were formed. Even though this finding indicates a similar suitability of the in situ systems, the batch experiments have shown a significantly diminished activity as compared to the isolated complexes (Chapter 3.2).

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

4.1 Hydrogenation of Itaconic Acid – State of the Art

The catalyzed conversion of itaconic acid (IA) 10 to 2-methyl-1,4-butanediol (2-MBDO) 24 proceeds in three hydrogenation steps. In a first step, methyl succinic acid (MSA) 21 is formed by hydrogenation of the methylene group of 10. In a second step, intramolecular esterification of 21 affords 2- and 3-methyl-γ-butyrolactone (2- and 3-MGBL) 22 and 23. 24 is then generated by ring- opening hydrogenation of the lactones. In the presence of acids, 24 can further react to 3-methyltetrahydrofuran (3-MTHF) 25 in a condensation reaction (Scheme 4).

Scheme 4: Catalytic conversion of IA (10) to MSA (21), 2-MGBL (22), 3-MGBL (23), 2-MBDO (24) and 3-MTHF (25).

The catalytic transformation of 10 to 24 and 25 has so far only received little academic attention. However, there are a number of patents claiming different synthetic strategies for the synthesis of 25 from alcohols, acids, esters, epoxides and lactones using heterogeneous catalysts.[87] In 1975, Du Pont described the acid catalyzed cyclization of 24 to 25 and in 1988, the company achieved selectivities of 98% for MGBL and 4% for 24 from 10 using Pd-Re/TiO2 catalysts.[88] In 1992, BASF showed the hydrogenation of 10 using copper and other group 7-10 metals resulting in yields of up to 85% for 24 and 91% for 25.[89] In 1996, BASF disclosed the copper catalyzed hydrogenation of 10 to 25 and observed the formation of 22 and 23 as intermediates.[90] In the last decade, many academic research groups have investigated the development of catalytic systems for the transformation of 10 to 22, 23, 24 and 25. In 2011, Corma used Ru/TiO2 catalysts to obtain a MGBL yield of 90% from the hydrogenation of 10.[75] In 2015, Han and Mu achieved a selectivity of 89.5% of MGBL using Pd/C catalysts.[91] One year later, 24 was obtained with over 80% yield using 56

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Pd-3ReOx/C catalysts.[41c] In 2016, Dauenhauer obtained 82% yield of MGBL from 10 using a Pd/C catalyst and a 2-MBDO (24) yield of almost 80% using Ru/C catalysts for the hydrogenation of MGBL.[41b] One year later, Pd-Re/C catalysts were utilized for the synthesis of 25 from 10 with a yield of 80%.[92]

In 2010, Leitner and Klankermayer disclosed the selective hydrogenation of 10 to 22, 23, 24 or 25 using the homogeneous [Ru(acac)3]/triphos in situ system (cat-I-acac).[41a] The hydrogenation of

2.4 mmol 10 in 1 mL 1,4-dioxane using 0.5 mol% [Ru(acac)3] and 0.6 mol% triphos at 195 °C and

100 bar initial H2 pressure resulted in a 93% yield of 24 after 18 h reaction time. In the mechanistic DFT-study, high energetic spans for the reactions steps were calculated, resulting in 31.0 kcal/mol for the hydrogenation of 21 to MGBL and 29.7 kcal/mol for the initial hydrogenation transfer in MGBL hydrogenation under acidic conditions.[55] The application of a ruthenium N-triphos system for the hydrogenation of 10 in 2016 by Palkovits resulted in a MGBL yield of 95%.[56j]

Analogously to the studies on LVA (9) hydrogenation disclosed in this work (Chapter 3), the catalyst development for the hydrogenation of 10 and 22 will focus on the selective conversion to 24 using tailored ruthenium triphos-xyl complexes.

4.2 Application of Novel Catalytic Systems in the Hydrogenation of 2-Methyl-γ-butyrolactone, Itaconic Acid and Itaconates

The performance of the different tailored catalysts and catalytic systems was first investigated for the hydrogenation of 2-MGBL (22). The reactions were performed without any additional solvents, as all components are liquid under operating conditions. The general reaction procedure was carried out analogously to the hydrogenation of LVA (9) or GVL (16) (Chapter 3.1).

In a first approach, cat-I-tmm and cat-II-tmm, as well as the bis(fluorophenolato) complexes cat-II-51–cat-II-58 and the bis(heptafluoro-2-naphtholato) complex cat-II-59 were tested in the hydrogenation of 22 at a catalyst concentration of 0.1 mol%, an initial H2 pressure of 70 bar and a reaction time of 1 h (Table 14). As expected, the obtained results for the yield of 24 show the significantly higher activity of the triphos-xyl complex cat-II-tmm as compared to cat-I-tmm. The application of complexes cat-II-51–cat-II-59 led to yet higher 2-MBDO (24) yields of up to 67% using cat-II-59 (entry 11). The partly significant formation of 3-MGBL (23) suggests a dehydrogenation reaction of 24 to 22 and 23 and a dynamic equilibrium between the lactones and the diol, which will be investigated more closely in chapter 4.3. As for comparison of the different

[Ru(triphos-xyl)(OR)2] type complexes, no clear trend concerning the obtained yields and structural or electronic properties could be observed. 57

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 14: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-I-tmm, cat-II-tmm, cat-II-51–cat-II-59.[a]

entry catalyst catalyst structure Y (24) [%][b] Y (23) [%][b]

1 cat-I-tmm 17 0

2 cat-II-tmm 38 2

3 cat-II-51 58 3

4 cat-II-52 63 7

5 cat-II-53 65 5

6 cat-II-54 54 4

7 cat-II-55 60 10

8 cat-II-56 66 6

9 cat-II-57 63 12

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

entry catalyst catalyst structure Y (24) [%][b] Y (23) [%][b]

10 cat-II-58 65 14

11 cat-II-59 67 14

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% catalyst, 160 °C, 70 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard. The corresponding pressure courses also clearly demonstrate the higher activity of the triphos-xyl complexes, as cat-I-tmm led to a significantly shallower pressure drop and a higher final pressure indicating a slower and lower hydrogen consumption and consequently less product formation (Figure 47). As expected from the obtained GC data, cat-II-tmm shows a higher activity than cat-I-tmm but a lower activity than complexes cat-II-51–cat-II-59.

100 cat-I-tmm

90 cat-II-tmm cat-II-51 80

cat-II-52

70 cat-II-53

[bar]

p cat-II-54 60 cat-II-55

50 cat-II-56

cat-II-57 40 0 20 40 60 cat-II-58 t [min] cat-II-59

Figure 47: Pressure courses for the hydrogenation of 2-MGBL (22) using cat-I-tmm, cat-II-tmm and cat-II-51–cat-II-59. Reaction conditions: 1 mL 2-MGBL (22), 0.1 mol% catalyst, 160 °C, 70 bar initial H2 pressure at rt, 1 h. Starting pressures were normalized to 70 bar.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

The pressure course recorded in the hydrogenation using cat-II-tmm shows a continuous linear decrease after the maximum pressure, whereas the courses using cat-II-51–cat-II-59 show a steep pressure drop in the beginning. This was found to be less prominent for the less active complexes cat-II-51 and cat-II-52, which also provided the lowest 2-MBDO (24) yields among the bis(fluorophenolato) complexes (Table 14, entries 3 and 4). The flattening of the pressure course indicates the approach of the dynamic lactone-diol equilibrium and can only be observed if the catalysts enable the reach of high diol concentrations within the reaction time.

Table 15: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-tmm and 0.2 mol% of additive after 1 h.[a]

entry additive additive structure Y (24) [%][b] Y (23) [%][b]

1 - - 38 2

2 40 45 2

3 51 50 3

4 52 68 7

5 53 74 9

6 54 77 8

7 55 74 13

8 56 75 11

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

entry additive additive structure Y (24) [%][b] Y (23) [%][b]

9 57 68 17

10 58 69 15

11 59 73 11

12 63 72 11

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard. To compare the activities of the isolated catalysts with the activities of their in situ systems, the reactions were performed using 0.1 mol% cat-II-tmm and 0.2 mol% of the respective fluorophenols or -naphthol under otherwise identical reaction conditions. 4-Fluorophenol 40 was also used as additive, but was found to provide the lowest 2-MBDO (24) yield of only 45% among the systems using fluorophenols as additives (Table 15, entry 2) and only a slightly higher yield than cat-II-tmm without additive (entry 1). With the exception of 2-fluorophenol 51, all in situ systems led to higher yields than the respective isolated complexes. The highest yield of 77% 24 was achieved using 3,5-difluorophenol 54 (entry 6). Varying amounts of 23 could be observed in all cases. Besides the fluorophenols and -naphthol, also hexafluorobenzene 63 was tested as additive to study whether the increased activity of the catalytic system could also be increased by the presence of perfluorinated benzene without an alcohol function. The application of the corresponding in situ system led to comparable results (entry 12).

As the high 3-MGBL (23) yield in some cases indicates an approach to the lactone-diol equilibrium after 1 h, the hydrogenation reactions were repeated at halved reaction time to compare the activity of the catalysts under conditions limiting the influence of the dehydrogenation reaction. The shorter reaction time was not applied using 40 and 51 due to the significantly lower performance of the corresponding in situ systems at 1 h reaction time. The results reveal the lowest 2-MBDO (24) yields of 32% and 39% using 2,5-difluorophenol 52 and hexafluorobenzene 63 (Table 16, entries 1 and 9). The highest 2-MBDO (24) yield of 53% with a 3-MGBL (23) yield of 8% was achieved using pentafluorophenol 58, denoting this catalytic system as the most active under the applied reaction conditions. 61

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 16: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-tmm and 0.2 mol% additive after 0.5 h.[a]

entry additive additive structure Y (24) [%][b] Y (23) [%][b]

1 52 32 1

2 53 46 3

3 54 46 3

4 55 52 5

5 56 50 4

6 57 43 4

7 58 53 8

8 59 47 9

9 63 39 3

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 0.5 h. [b] Determined by GC using ethyl heptanoate as internal standard.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

100 52

90 53

80 54

70 55

[bar]

p 56 60 57 50 58 40 59 0 5 10 15 20 25 30 t [min] 63

Figure 48: Pressure courses for the hydrogenation of 2-MGBL (22) using cat-II-tmm and 2 eq additive. Reaction conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 0.5 h. Starting pressures were normalized to 70 bar.

The corresponding pressure recordings (Figure 48) indicate similar activities for the in situ systems using 55, 56, 58 and 59 with similar courses of the pressure drops and the lowest final pressures for 55 and 58, which is in line with the obtained GC results. In contrast, the pressure courses of the hydrogenations using the in situ systems with 52, 53, 54, 57 and 63 show a longer pressure plateau at maximum pressure followed by a shallower pressure drop. In the case of 3,5-difluorophenol 54, the pressure course shows an acceleration of reaction rate after ~22 min, indicated by an increase in steepness of the curve, whereas the pressure courses of the other systems show a constant or even decreasing reaction rate. This indicates a slower activation of the catalyst using 54 as additive.

Driven by the obtained results, the in situ system cat-II-tmm/58 was studied more closely. Performing the reaction with a decreased concentration of 0.1 mol% 58 led to slightly higher 2-MBDO (24) yields than using 0.2 mol% 58 after both 0.5 and 1 h reaction time (Table 17, entries 1–4). Consequently, there is no necessity to add stoichiometric amounts of 58 in respect to cat-II-tmm, which would be required to form the corresponding complex cat-II-58 to achieve an increased activity in the hydrogenation of 22. A higher concentration of 0.4 mol% 58 did not significantly influence the yield (entry 5). Decreasing the reaction temperature from 160 °C to 140 °C led to a lower 2-MBDO (24) yield after 1 h and a significantly lower 3-MGBL (23) yield (entry 6). Increasing the temperature from 160 °C to 180 °C did not significantly change the obtained yield of 24, but slightly increased the 3-MGBL (23) yield (entries 7 and 8). An increase to 200 °C decreased the yield of 24 while leading to a 3-MGBL (23) yield of 25% after 0.5 h reaction time (entry 9).

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 17: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation to 3-MGBL (23) using cat-II-tmm and 58.[a]

entry n% 58 [mol%] T [°C] p [bar][b] t [h] Y (24) [%][c] Y (23) [%][c]

1 0.1 160 70 1 73 13 2 0.2 160 70 1 69 15 3 0.1 160 70 0.5 56 14 4 0.2 160 70 0.5 53 8 5 0.4 160 70 0.5 55 14 6 0.2 140 70 1 52 2 7 0.2 180 70 1 66 19 8 0.2 180 70 0.5 55 20 9 0.2 200 70 0.5 43 25 10 0.2 160 90 1 87 7 11 0.2 180 90 1 77 7 12 0.1 160 90 1 78 8 13 0.2 160 90 1.5 91 3 14 0.2 160 110 1 89 4 15 0.2 180 110 1 85 6 16[d] 0.02 160 90 24 1 0 17[d] 0.02 160 90 72 1 0 18[d] 0.02 180 90 72 3 0

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm. [b] initial H2 pressure at rt. [c] Determined by GC using ethyl heptanoate as internal standard. [d] 0.01 mol% cat-II-tmm. These results clearly show a shift of the lactone-diol equilibrium with increasing temperature to the lactone side counteracting the benefits of an increased reaction rate at higher reaction temperatures.

With the objective of shifting the lactone-diol equilibrium to the side of the desired product 24, the initial hydrogen pressure was varied from 70 to 110 bar, resulting in an increased 2-MBDO (24) yield of 89% using 0.2 mol% 58 after 1 h (entries 10 and 14), while decreasing the yield of 23 to 4%. Increasing the reaction temperature to 180 °C or decreasing the concentration of 58 to 0.1 mol% at 90 bar resulted in decreased 2-MBDO (24) yields at constant 3-MGBL (23) yields (entries 11 and 12). A slightly decreased yield for 24 was also obtained at 180 °C under 110 bar 64

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

initial hydrogen pressure (entry 15). Elongation of the reaction time to 1.5 h increased the obtained 2-MBDO (24) yield to 91%, while the formation of only 3% 23 was observed (entry 13). The comparably small increase in 2-MBDO (24) yield after a reaction time elongated by 50% shows the significant slowdown of reaction rate between 1 and 1.5 h. Lowering the catalyst concentration to 0.01 mol% cat-II-tmm and 0.02 mol% 58, to test the performance of the catalytic system under diluted conditions, led to no significant formation of 24 after reaction times of up to 72 h (entries 16–18).

Table 18: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using different catalysts and 0.2 mol% 58.[a]

entry catalyst catalyst structure Y (24) [%][b] Y (23) [%][b]

1 cat-II-tmm 87 7

2 cat-V-tmm 47 2

3 cat-VI-tmm 31 0

4 cat-VII-tmm 49 8

5 cat-VIII-tmm 78 11

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% catalyst, 0.2 mol% 58, 160 °C, 90 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Different derivatives of ruthenium triphos tmm complexes synthesized[69a] and kindly supplied by Stefan Westhues were also tested as in situ systems with two equivalents of 58 as additive and compared to the performance of the cat-II-tmm/58 in situ system under optimized reaction conditions. As expected, significantly lower 2-MBDO (24) and 3-MGBL (23) yields were obtained using [Ru(triphos-tol)(tmm)] cat-V-tmm (Table 18, entry 2), which showed lower activity than cat-II-tmm in the hydrogenation of methyl benzoate and was found to deactivate by the formation of the corresponding hydrogen-bridged dimer in the same study.[64a] The complexes bearing

CF3-substituted triphos ligands [Ru(triphos-tol(CF3))(tmm)] cat-VI-tmm and

[Ru(triphos-xyl(CF3)2)(tmm)] cat-VII-tmm also performed inferior to cat-II-tmm, with cat-VI-tmm leading to a more significantly decreased 2-MBDO (24) yield (entries 3 and 4). The in situ system using [Ru(triphos-xyl(OMe)2)(tmm)] cat-VIII-tmm led to a lower 2-MBDO (24) yield of 78% and a slightly higher 3-MGBL (23) yield than cat-II-tmm, indicating a higher activity of the system in the dehydrogenation reaction (entry 5).

The in situ system of cat-II-tmm and 58 was also studied and compared to cat-II-58 concerning its performance in multiple substrate loading and hydrogenation runs at unchanged reaction time for the second hydrogenation run. After the first run, the reactions were stopped, the autoclaves cooled in an ice bath and depressurized. Fresh 22 was added under argon counter-current flow and the hydrogenations were performed again (Figure 49).

Figure 49: Schematic procedure for the hydrogenation of 2-MGBL (22) in two substrate loading and hydrogenation runs.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 19: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-tmm and 0.2 mol% 58 or cat-II-58 in two substrate loading and hydrogenation runs.[a]

Y (24) Y (23) entry catalyst n% 58 [mol%] run t [h] [%][b] [%][b]

1st 1 87 7 1 cat-II-tmm 0.2 2nd 1 56 16 1st 1.5 91 3 2 cat-II-tmm 0.2 2nd 1.5 60 15 1st 1.5 80 5 3 cat-II-58 - 2nd 1.5 52 16

[a] General conditions: 1 mL 2-MGBL (22) each run, 0.1 mol% catalyst, 0.2 mol% 58, 160 °C, 90 bar initial H2 pressure at rt. [b] Determined by GC using ethyl heptanoate as internal standard. The obtained results for cat-II-tmm and 0.2 mol% 58 and a reaction time of 1 h each for the first and second run show a significant decrease in 2-MBDO (24) and an increase in 3-MGBL (23) yield after the second run, while increasing the reaction time to 1.5 h each run only slightly increased the yield of 24 obtained after the second run as compared to 1 h reaction time (Table 19, entries 1 and 2). Using cat-II-58 led to slightly decreased 2-MBDO (24) yields after the first and the second run, while the obtained 3-MGBL (23) yields were found to be similar (entry 3). As the reaction media at the beginning of the second hydrogenation run already contains significant amounts of 24, the back reaction to 22 and 23 is benefited and consequently a lower 2-MBDO (24) yield can be achieved under the applied reaction conditions. Additionally, the lower yields could be attributed to a certain degree of catalyst deactivation due to the formation of inactive species as well.

After optimization of the reaction conditions for the cat-II-tmm/58 in situ system, the scope was extended to IA (10) and the two itaconate esters dimethyl itaconate (DMI) 64 and dibutyl itaconate (DBI) 65. As 64 and 65 are liquid under the applied reaction conditions, no additional solvent was used, whereas 1 mL 1,4-dioxane was added to dissolve 1 mmol of 10. Nearly no 24 could be detected after a reaction time of 15 h using 10 as substrate with 1 mol% cat-II-tmm and 2 mol% 58 at 160 °C, while a 2-MBDO (24) yield of 89% could be achieved after 20 h (Table 20, entries 1 and 2).

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 20: Hydrogenation of IA (10) to 2-MGBL (22), 3-MGBL (23) and 2-MBDO (24), using cat-II-tmm and 2 eq 58.[a]

t entry n% cat-II-tmm [mol%] T [°C] Y (22) [%][b] Y (23) [%][b] Y (24) [%][b] [h]

1 1 160 15 20 34 1 2 1 160 20 3 4 89 3 0.1 160 72 24 34 1 4 0.1 195 18 27 31 42 5 0.2 195 18 14 15 67 6 0.3 195 18 12 14 70

[a] General conditions: 1 mmol IA (10), 0.1–1 mol% cat-II-tmm, 2 eq 58, 50 bar initial H2 pressure at rt, 1 mL dioxane. [b] Determined by GC using ethyl heptanoate as internal standard. This result indicates a similar reaction course to the hydrogenation of 9 where no diol formation could be observed before full conversion of the acid to the lactone was achieved (Chapter 3.2). As more 23 than 22 was obtained, the formation of the 3-substituted lactone seems to be favored. Lowering the catalyst concentration to 0.1 mol% cat-II-tmm and 0.2 mol% 58 led to almost no formation of 24 after 72 h, indicating a significantly slowed down reaction rate (entry 3). To increase the rate, the temperature was increased to 195 °C resulting in a 2-MBDO (24) yield of 42% after 18 h (entry 4).

Table 21: Hydrogenation of DMI (64) and DBI (65) to 2-MGBL (22), 3-MGBL (23) and 2-MBDO (24) using cat-II-tmm and 58.[a]

p entry substrate t Y (22) [%][b] Y (23) [%][c] Y (24) [%][c] [bar][b]

1 64 100 190 min 27 35 30 2 64 150 190 min 6 8 80 3 65 150 22 h 7 9 78

[a] General conditions: 10 mmol substrate, 0.1 mol% cat-II-tmm, 0.2 mol% 58, 160 °C. [b] initial H2 pressure at rt. [c] Determined by GC using ethyl heptanoate as internal standard.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

200 200

150 150

100 100

[bar]

p p p

50 50

0 0 0 100 200 0 500 1000 1500 t [min] t [min]

Figure 50: Pressure courses for the hydrogenation of 64 (left) and 65 (right). Reaction conditions: 10 mmol substrate, 0.1 mol% cat-II-tmm, 0.2 mol% 58, 160 °C, 150 bar initial H2 pressure at rt. Starting pressures were normalized to 150 bar.

Increasing the concentration to 0.2 mol% cat-II-tmm led to a 2-MBDO (24) yield of 67%, while a further increase to 0.3 mol% did not result in a significantly higher yield (entries 5 and 6). The observed stagnation in 2-MBDO (24) yield can most likely be attributed to the earlier discussed shift of the lactone-diol equilibrium at higher reaction temperatures. Surprisingly, the increased formation of 23 as compared to 22 is less significant at increased temperatures.

The hydrogenation of 64 was carried out at 160 °C and an initial hydrogen pressure of 100 bar using 10 mmol of substrate with a catalyst concentration of 0.1 mol% cat-II-tmm and 0.2 mol% 58. As the obtained results with a 2-MBDO (24) yield of 30% after 190 min were promising (Table 21, entry 1), but the applied hydrogen pressure was found to be insufficient for 10 mmol of substrate, the initial pressure was increased to 150 bar. Consequently, an increased 2-MBDO yield of 80% was achieved in the hydrogenation of 64 and 78% of 24 were obtained using 65 as substrate after 22 h (entries 2 and 3). The corresponding pressure courses clearly demonstrate a much faster hydrogenation for 64 (Figure 50). The rise in pressure due to the temperature increase at the beginning of the pressure course is less prominent for 64 with a maximum pressure of ~180 bar as compared to ~185 bar for 65. Consequently, hydrogen is already being consumed faster during the heating-up period when hydrogenating the dimethyl ester. The pressure course also shows the significantly earlier approach of the final pressure plateau using 64 as substrate, which indicates the reach of the lactone-diol equilibrium. Analogously to the observations made in the hydrogenation of 9, the courses show a pressure plateau after a first pressure drop, followed by a second decline of pressure, which can be attributed to the different reaction steps.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 22: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-COH2 with or without fluorophenolic/-naphtholic additives.[a]

entry additive (pKa) Y (24) [%][b] Y (23) [%][b]

1 - 2 0 2 54 (8.4)[70m] 2 0 3 58 (5.41/5.48)[70n,70p] 70 12 4 59 82 7

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-COH2, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 3 h. [b] Determined by GC using ethyl heptanoate as internal standard. Whereas the very fast hydrogenation of the double bond of the itaconate esters leading to the formation of the corresponding methylsuccinate esters is assumed to proceed within the time range of the temperature-dependent pressure increase, the first pressure decline can be attributed to the hydrogenation of the methylsuccinate esters to 22 and 23. The second pressure drop corresponds to the formation of 24 from the lactones. In contrast to the hydrogenation of 10, the coupling products methanol or butanol are formed in the hydrogenation of 64 or 65, respectively.

The main deactivation pathway of cat-II-tmm in hydrogenation reactions under neutral conditions was found to be the formation of [Ru(triphos-xyl)(CO)H2] cat-II-COH2.[64a] As the corresponding ruthenium triphos monocarbonyl dihydrido complex was successfully reactivated in the presence of acids, cat-II-COH2 was applied in the hydrogenation of 22 with fluorophenolic and -naphtholic additives of different pKa values. Performing the hydrogenation without any additive led to a 2-MBDO (24) yield of only 2% (Table 22, entry 1). The same yield was observed when adding two equivalents of 54, a fluorophenol with a comparably low acidity (entry 2). As expected, the use of 58 and 59 as additives with higher acidic strengths significantly increased the 2-MBDO (24) yield to 70% and 82%, respectively (entries 3 and 4). These results show the successful activation of cat-II-COH2 in the presence of a fluorophenol or -naphthol with a certain acidity.

Besides the bis(3,5-difluorophenolato) complex cat-II-54, also cat-II-60 and cat-II-61 bearing two 3,5-dichloro- or 3,5-dibromophenolato ligands as well as cat-II-62 bearing two 3,5-trifluoromethylphenolato ligands were applied in the hydrogenation of 22.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 23: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-54, cat-II-60–cat-II-62.[a]

entry catalyst catalyst structure Y (24) [%][b] Y (23) [%][b]

1 cat-II-54 54 4

2 cat-II-60 20 0

3 cat-II-61 1 0

4 cat-II-62 57 6

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% catalyst, 160 °C, 70 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard. The results show a similar performance of cat-II-54 and cat-II-62, while the chloro- and bromosubstituted complexes led to significantly lower 2-MBDO (24) yields with nearly no reactivity when using cat-II-61 (Table 23).

To investigate whether these inferior activities can be attributed to a slower activation of the complexes, 3,5-dichlorophenol 60 and 3,5-dibromophenol 61 were also applied as additives in the hydrogenation of 22 using cat-II-tmm. The results show the same trend as observed for the preformed complexes with a significantly lower 2-MBDO (24) yield using 60 as additive compared to using 54 and only 4% 2-MBDO (24) yield using 61 (Table 24, entries 2–4). The in situ system using 60 shows a more significantly increased activity than cat-II-60 compared to the rise in activity of the in situ system using 54 or cat-II-54. This suggests a negative influence of the chloro- substituted ligand on the catalyst activation as compared to the fluoro-substituted ligands.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 24: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-tmm and different additives.[a]

entry additive additive structure Y (24) [%][b] Y (23) [%][b]

1 - 38 2

2 54 77 8

3 60 59 4

4 61 4 0

5 32 39 1

6 47 5 tris(pentafluorophenyl)borane

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard. Another approach for an explanation of the inferior activities of the chloro- and bromo-substituted ligands in both the molecular defined complexes and the in situ systems is their higher steric demand, which can hinder the substrate coordination if the ligands stay in proximity to the metal center. This can occur due to a coordination of one or two of the ligands, but also a π-stacking of the aromatic rings with the xylyl-substituents at the phosphorous atoms of the triphos-xyl ligands or the formation of inactive ruthenium chlorido or bromido complexes is possible.

The addition of unsubstituted phenol 32 had no influence on the catalytic activity and led to very similar 2-MBDO (24) and 3-MGBL (23) yields as using cat-II-tmm without any additive (entry 5). The application of the Lewis acid tris(pentafluorophenyl)borane resulted in a 2-MBDO (24) yield of 47% and consequently increased the activity of cat-II-tmm, but not to the same degree as most of the investigated fluorophenols (entry 6).

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 25: Salvation effect on the hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-54.[a]

entry solvent Y (24) [%][b] Y (23) [%][b]

1 - 54 4 2 toluene 55 4 3 1,4-dioxane 46 3 4 THF 46 2 5 benzene 31 1

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-54, 1 mL solvent, 160 °C, 70 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard. The bis(3,5-difluorophenolato) complex cat-II-54 was also applied in the hydrogenation of 22 with different solvents to study their influence on the outcome of the reaction as compared to neat conditions. None of the investigated solvents did significantly increase the obtained 2-MBDO (24) yield (Table 25). While adding 1 mL of toluene led to similar results (entry 2), the more polar solvents 1,4-dioxane, THF and benzene had a negative influence on the formation of 24 (entries 3– 5). However, the 2-MBDO (24) yield does not consistently decrease with increasing polarity of the solvents as benzene being less polar than 1,4-dioxane and THF provides a lower yield. The observed trend could also be attributed to a lower hydrogen solubility in 1,4-dioxane, THF and benzene as compared to toluene.[93]

As discussed earlier, the application of hexafluorobenzene 63 as additive to cat-II-tmm was also found to increase the 2-MBDO (24) yield (Table 15). In contrast to the utilized fluorophenols and -naphthol, the enhancement of the catalytic activity cannot be attributed to Brønsted acidic properties of the additive. A theoretical study by Tsuzuki and Uchimaru disclosed a Lewis acid-base interaction of the electron deficient π-system of 63 with the electron donor ammonia.[94] Therefore, the Lewis-acidic properties of 63 supposably enhance the activity of the cat-II-tmm/63 in situ system. Application of the electron-rich π-system benzene as additive led to the same 2-MBDO yield as using cat-II-tmm without additive (Table 26). Using hexachlorobenzene 66 led to a significant decrease in 2-MBDO (24) yield as compared to 63. However, the observed yield is still higher than without any additive (entry 4). Hexabromobenzene 67 as additive was found to entirely suppress the formation of 24 under the applied reaction conditions (entry 5).

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 26: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-tmm and different benzene additives.[a]

entry additive structure additive Y (24) [%][b] Y (23) [%][b]

1 - - 38 2

2 benzene 37 1

3 63 72 11

4 66 43 2

5 67 0 0

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm, 0.2 mol% additive, 160 °C, 70 bar initial H2 pressure at rt, 1 h. [b] Determined by GC using ethyl heptanoate as internal standard. These findings support the earlier discussed assumption of the inhibition of catalytic activity in the presence of sterically demanding aromatic rings, likely due to a π-stacking of the aromatic xylyl rings with the halogenated benzenes or the formation of inactive ruthenium chlorido or bromido complexes. In a theoretical study by Tsuzuki, intermolecular interaction energies between 63 and benzene were calculated. It was found that the electron rich and the electron poor molecule form stable sandwich and slipped-parallel complexes.[95] As the xylyl rings bear electron donating groups, the electron density of the aromatic ring is increased and the formation of stable complexes with perhalogenatobenzenes is even more supposable. Deactivation of the catalysts by the formation of inactive chlorido or bromide complexes also has to be taken into consideration and will be studied more closely in chapter 5.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Table 27: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-tmm and 63.[a]

[b] [b] entry n% 63 [mol%] p [bar] ttotal [h] Y (24) [%] Y (23) [%]

1 0.1 70 1 41 1 2 0.2 70 1 72 11 3 0.2 70 0.5 39 3 4 0.4 70 1 70 11 5 0.4 70 0.5 50 5 6 1.0 70 1 61 15 7 10 70 1 55 19 8a 0.2 90 1 85 7 8b[c] 0.2 90 2 51 9 9[d] 0.02 90 72 5 0

[a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% cat-II-tmm, 0.02–10.0 mol% 63, 160 °C, 70–90 bar initial H2 pressure at rt. [b] Determined by GC using ethyl heptanoate as internal standard. [c] Second hydrogenation run. [d] 0.01 mol% cat-II-tmm. Using 0.1 mol% cat-II-tmm with 0.1 mol% 63 significantly decreased the obtained 2-MBDO (24) yield as compared to the application of 0.2 mol% 63 (Table 27, entries 1 and 2). Doubling of the 63 amount to 0.4 mol% led to similar results as 0.2 mol% after 1 h reaction time (entry 4). To minimize the effect of the approach of the lactone-diol equilibrium, the reaction was also carried out at halved reaction time leading to a higher 2-MBDO (24) yield for 0.4 mol% 63 (entries 3 and 5). A further increase of 63 to 1.0 or 10 mol% significantly decreased the 2-MBDO (24) yield while increasing the 3-MGBL (23) yield (entries 6 and 7). As expected, a higher initial hydrogen pressure of 90 bar led to a higher 2-MBDO (24) yield of 85% (entry 8a). A second hydrogenation run after the addition of fresh 22 led to a lower 2-MBDO (24) yield of 51% (entry 8b), which was also observed in the corresponding experiment with 58 as additive (Table 19). In contrast, no significant increase in 3-MGBL (23) yield could be observed. Analogously to the investigations on the cat-II-tmm/58 in situ system (Table 17), decreasing the catalyst concentration to 0.01 mol% cat-II-tmm and

0.02 mol% 63 led to no significant formation of 24 even after 72 h reaction time (entry 9).

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

4.3 Mechanistic Insights into the Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Ruthenium Triphos-xyl Complexes

The formation of significant amounts of 23 in the hydrogenation of 22 could be attributed to a dynamic diol-lactone equilibrium between 24 and 22/23. To evaluate the direct isomerization of 22 to 23, 22 was stirred with 0.1 mol% cat-II-58 at 160 °C for 1 h in the absence of hydrogen. In this experiment, no formation of 23 could be observed.

Table 28: Dehydrogenation of 2-MBDO (24) to 2-MGBL (22) and 3-MGBL (23) using cat-II-58.[a]

entry p [bar][b] Y (22) [%][c] Y (23) [%][c]

1 0 11 15 2 70 0 1

[a] General conditions: 1 mL 2-MBDO (24), 0.1 mol% cat-II-58, 160 °C, 3 h. [b] initial H2 pressure at rt. [c] Determined by GC using ethyl heptanoate as internal standard. On the other hand, heating 24 at 160 °C for 3 h in the initial absence of hydrogen with 0.1 mol% cat-II-58 led to the formation of 15% 23 and 11% 22 (Table 28, entry 1). The higher yield of 23 as compared to 22 can be attributed to the sterically more favored formation of a 3-substituted lactone which was observed before by Ishii using different homogeneous ruthenium catalysts.[96]

[bar]

0 0 20 40 60 80 100 120 140 160 180 t [min]

Figure 51: Pressure course for the dehydrogenation of 2-MBDO (24). Reaction conditions: 1 mL 2-MBDO (24), 0.1 mol% cat-II-58, 160 °C, 3 h.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

As anticipated, the corresponding pressure course shows a steep increase in pressure after the heating-up phase, which flattens with proceeding reaction time (Figure 51). Performing the same reaction with 70 bar initial hydrogen pressure consequently suppressed the lactone formation (entry 2). These experiments clearly show the direct equilibrium pathway between 22, 23 and 24. cat-II-tmm was converted with one or two equivalents of 10 to obtain information about the formation of catalyst species prior to the hydrogenation. The 31P{1H}-NMR spectrum measured after dissolving the components in THF-d8 showed a 34:66 ratio of a new species generating a multiplet at 40.2–39.6 ppm (Figure 52). Measuring the same sample after 1 day increased the ratio to 78:22. After 4 days, more newly generated signals between 41.7 and 38.9 ppm and a ratio of 83:17 of the sum of these signals to the singlet generated by cat-II-tmm could be observed. Heating the mixture at 50 °C led to the generation of two singlets at 41.0 and 39.1 ppm. The ratio of the sum of new signals to the singlet of cat-II-tmm was found to be 91:9, which could be increased to 93:7 after additional 16 h at 50 °C. Performing the analogous experiment with two equivalents of 10 led to full conversion to the species generating the multiplet at 40.2–39.6 ppm after 1 day at room temperature. No significant changes could be observed after 4 days (Figure 54).

heated to 50 °C for 16 h

rt, measured after 4 days

rt, measured after 1 day

rt, measured directly

Figure 52: 31P{1H}-NMR spectra of the reaction mixture obtained by conversion of cat-II-tmm with one equivalent IA (10) at rt, at rt after one day, at rt after four days and heated to 50 °C for 16 h in THF-d8 measured at rt.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

♠ ♦

THF-d8 residual signals

♠ ♦ ♦ ♦ ♦

Figure 53: 1H-NMR spectrum of the reaction mixture obtained by conversion of cat-II-tmm with two equivalents IA (10) at rt after 4 days in THF-d8 , measured at rt.

The corresponding 1H-NMR spectrum shows the characteristic signals generated by the triphos-xyl ligand (♦) and reveals the formation of isobutene (♠) by protonation of the tmm ligand, which generates a singlet at 4.52 ppm representing the olefinic protons (Figure 53). The signal generated by the corresponding CH3 groups is generated at 1.70 ppm. Furthermore, multiple signals occur in the region between 6.10–5.20 ppm where the signals generated by the olefinic protons of 10 would be expected.

Figure 54: 31P{1H}-NMR spectrum (left) and 1H-31P-HMBC-NMR spectrum (right) of the reaction mixture obtained by conversion of cat-II-tmm with two equivalents IA (10) at rt after 4 days in THF-d8 measured at rt.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

The two signal groups (6.10–5.85 ppm and 5.54–5.20 ppm) amount to an intensity of two protons each. A group of signals between 3.18–3.05 ppm can be found in the region where the signal generated by the alkylic CH2 group of 10 would be expected. Additionally, a broad signal at 12.61 ppm representing two protons can be observed, which can be assigned to the carboxylic acid groups of 10 or itaconato ligands. It can therefore be assumed, that multiple ruthenium triphos-xyl species bearing itaconato ligands have been formed by the conversion of cat-II-tmm with two equivalents of 10, during which the tmm ligand has been protonated by the acid function of 10. The structure of 10 being an unsymmetric diacid allows several different possible itaconato complexes.

The 1H-31P-HMBC-NMR spectrum reveals that the phosphorous atoms only show coupling with the protons attributed to the triphos-xyl ligand as presumably the corresponding nuclei of the itaconato ligands are too far distanced (Figure 54).

One equivalent 10 was also converted with cat-II-tmm in a toluene-d8/1,4-dioxane mixture. After heating to 100 °C, full conversion of cat-II-tmm to several species generating signals between 41.1 and 38.5 ppm in the 31P{1H}-NMR spectrum similar to those observed in THF-d8 was obtained (Figure 55).

Figure 55: 31P{1H}-NMR spectra of the reaction mixture obtained by conversion of cat-II-tmm with one equivalent of IA (10) heated to 100 °C for 2 h in toluene-d8/1,4-dioxane (left) and by conversion of cat-II-54 with one equivalent IA (10) at rt (a) and heated to 100 °C for 2 h (b) in toluene-d8/dioxane (right), measured at rt.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Figure 56: 31P{1H}-NMR spectra of cat-II-54 dissolved in 2-MGBL (22) measured using a DMSO-d6 filled capillary as external NMR solvent at rt.

Adding one equivalent of 10 to the bis(3,5-difluorophenolato) complex cat-II-54 led to full conversion of cat-II-54 already at room temperature with concurrent formation of species generating signals in the same range. Small changes could be observed when heating the mixture to 100 °C. These observations reveal that also monoitaconato complexes can be formed.

Heating cat-II-tmm or cat-II-54 with one equivalent 22 in toluene-d8/1,4-dioxane at up to 100 °C led to no conversion according to 31P{1H}-NMR spectroscopy. Dissolving cat-II-54 in 22 led to the generation of a singlet at 27.2 ppm and multiplets in the region of 42.0–37.7 ppm, while cat-II-54 was still present in the solution generating a singlet with a ratio of 38% (Figure 56). Consequently, only a small amount of species bearing a 22 derived ligand is present in the short time of preparation of the substrate solution prior to the hydrogenation reaction.

To gain a deeper insight on the behavior of the in situ system with the highest observed activity, a solution of cat-II-tmm and two equivalents 58 was measured directly after dissolving the components in THF-d8.

b b

a a

Figure 57: 31P{1H}-NMR spectra (left) and 19F{1H}-NMR spectra (right) of the reaction mixture obtained by conversion of cat-II-tmm with two equivalents 58 at rt (a) and heated to 160 °C for 30 min with an initial H2 pressure of 50 bar (b) in THF-d8, measured at rt. 80

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

The solution was then transferred into an autoclave, which was then pressurized with 50 bar H2 and heated at 160 °C for 30 min. The obtained 31P{1H}-NMR spectra show a 48:52 ratio of cat-II-58 and cat-II-tmm at room temperature and the formation of various new signals after the described treatment in the autoclave (Figure 57). Two singlets are generated at 49.3 and 25.1 ppm and a broad signal occurs at 46.0–44.0 ppm. A pseudo-pentet with a coupling constant of 21.3 Hz can be observed at 48.0 ppm and a multiplet is generated at 35.5–33.6 ppm. The corresponding 19F{1H}- NMR spectrum reveals only one singlet at −141.58 ppm in the measured range of 25–−291 ppm (Figure 57). This observation leads to the conclusion that all pentafluorophenol 58 and pentafluorophenolato ligands have reacted to other fluorinated species, which could not be isolated and identified in this experiment but will be further investigated in detail in chapter 5 using 63. Conducting the same experiment in 22 led to the generation of two broad signals at 42.2–40.5 ppm and 30.4–29.4 ppm in the 31P{1H}-NMR spectrum. In the 19F{1H}-NMR spectrum, the previously observed singlet at −141.58 ppm was not generated and several new signals with different coupling patterns occurred (Figure 58). Even though the signals occur at different chemical shifts, some of them exhibit very similar coupling patterns to those of 58 or pentafluorophenolato ligands. This finding indicates that the applied reaction environment has a stabilizing effect and circumvents the decomposition of pentafluorophenol/-ato to entirely different compounds.

b b

a a

Figure 58: 31P{1H}-NMR spectra (left) and 19F{1H}-NMR spectra (right) of the reaction mixture obtained by conversion of cat-II-tmm with two equivalents 58 at rt (a) and heated to 160 °C for 30 min with an initial H2 pressure of 50 bar (b) in 2-MGBL (22) measured using a DMSO-d6 filled capillary as external NMR solvent at rt.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

cat-II-61

cat-II-60

cat-II-54

Figure 59: 31P{1H}-NMR spectra of the reaction mixture obtained by the hydrogenation of 2-MGBL (22) with 1 mol% cat-II-54, cat-II-60 or cat-II-61 after 15 min at 160 °C with an initial H2 pressure of 50 bar, measured in THF-d8 at rt.

The same experiment was performed with cat-II-tmm and two equivalents of 63 in 22 and led to similar results. According to the 31P{1H}-NMR spectrum, traces of cat-II-COH2 were also formed in this case and the corresponding 19F{1H}-NMR spectrum showed no conversion of 63 to other fluorinated compounds.

Significantly different activities for the application of the fluoro-, chloro- and bromo-substituted complexes cat-II-54, cat-II-60 and cat-II-61 have been observed in the hydrogenation of 22 (Chapter 4.2, Table 23). Similar observations were made using perhalogenated benzenes as additives (Chapter 4.2, Table 26). Their interaction with cat-II-tmm will be studied closer in chapter 5. To investigate why cat-II-60 leads to a lower 2-MBDO (24) yield than cat-II-54, and cat-II-61 suppresses any hydrogenation, the reaction mixture was analyzed by NMR spectroscopy after 15 min reaction time at 160 °C using 1 mol% of the respective catalysts with an initial hydrogen pressure of 50 bar. Using cat-II-54 or cat-II-60, the obtained 31P{1H}-NMR spectra show the characteristic triplet and doublet generated by the monocarbonyl dihydrido complex (Figure 59). Using cat-II-54, also a broad signal at ~42.5 ppm and a sharp singlet at ~29.5 ppm were generated, which were not further identified. Using cat-II-60, this singlet can be observed with a lower intensity and additionally, two other singlets at ~31.5 and ~27 ppm are generated of which the latter is of higher intensity. Using cat-II-61, only one sharp singlet at ~29.5 ppm can be detected. The corresponding species was not further identified, but as no conversion was observed in the corresponding batch reaction it can be presumed that this species is inactive in the hydrogenation of 22.

4 Hydrogenation of 2-Methyl-γ-butyrolactone and Itaconic Acid using Tailored Ruthenium Triphos-xyl Complexes

Figure 60: 19F{1H}-NMR spectra of cat-II-54 (a), cat-II-54 dissolved in 2-MGBL (22) (b) and the reaction mixture obtained after the hydrogenation of 2-MGBL (22) with 1 mol% cat-II-54 after 15 min at 160 °C with an initial H2 pressure of 50 bar (c), measured in THF-d8 at rt.

The obtained results show that the different catalytic performances could be attributed to the formation of different active or inactive species under hydrogenation conditions. In the case of cat-II-54, the mixture was also analyzed using 19F{1H}-NMR spectroscopy and compared to the spectrum of cat-II-54 in THF-d8. The substrate solution containing dissolved cat-II-54 shows the additional generation of a broad signal at −111.73 ppm as well as of a multiplet at −117.13– −117.37 ppm (Figure 60). After the hydrogenation, a multiplet at −111.65–−111.78 ppm is the only detectable signal in the measured range, which is generated by an intact 3,5-difluorophenolato moiety as can be stated due to the very similar coupling pattern of the signal as compared to 54.

5 Ruthenium Triphos-xyl Based Complexes Derived from Perhalogenatobenzenes

The hydrogenation of 2-MGBL (22) using the in situ system [Ru(triphos-xyl)(tmm)]/hexafluorobenzene (cat-II-tmm/63) has led to a significant improvement of the obtained 2-MBDO (24) yield as compared to using cat-II-tmm without additives (Chapter 4.2, Table 26). Using hexachlorobenzene HCB (66) led to a decreased activity as compared to HFB (63) and the application of hexabromobenzene HBB (67) entirely suppressed the hydrogenation reaction. The interaction between cat-II-tmm and the perhalogenated benzenes will be studied more closely in this chapter.

In literature, several ruthenium complexes have been reported to catalyze the hydrodefluorination of HFB using silanes or isopropanol together with a sodium base as fluorine scavengers.[97] In the absence of fluorine scavengers, the conversion of two ruthenium dihydrido complexes with HFB has been observed, leading to the formation of Ru-C6F5 species and the generation of HF.[98] In one of these cases, the released HF was reported to lead to the formation of a ruthenium species bearing a F–H–F ligand.[99]

5.1 Reactivity of Ru(triphos-xyl)(tmm) with Perhalogenatobenzenes

In a first approach, cat-II-tmm was dissolved with two equivalents of 63 in toluene-d8 or THF-d8.

Heating the sample in toluene-d8 up to 100 °C led to no changes in the measured 31P{1H}-NMR spectra. The analogous sample in THF-d8 was first heated to 50 °C without any observable changes according to 31P{1H}-NMR spectroscopy. The solution was then transferred into a previously evacuated stainless-steel autoclave equipped with a borosilicate glass inlet and a magnetic stir bar under argon countercurrent flow. The autoclave was pressurized with 50 bar H2 at rt and heated at 160 °C for 30 min (Scheme 5). After cooling, the product solution was removed from the autoclave under argon countercurrent flow. The precipitation of a yellow solid from the brown solution could be observed.

Scheme 5: Conversion of cat-II-tmm with HFB (63) and H2 in THF-d8. 84

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Figure 61: 31P{1H}-NMR (left) and 19F{1H}-NMR (right) spectra of the reaction mixture obtained after the conversion of cat-II-tmm with two equivalents of 63 in THF-d8 after 30 min at 160 °C and an initial H2 pressure of 50 bar, measured at rt.

According to the 31P{1H}-NMR spectrum, cat-II-tmm was fully converted to different species. Two singlets of low intensities were generated at 51.0 and 26.8 ppm and the main species generate multiplets at 50.5–49.4 ppm and 37.4–35.5 ppm (Figure 61). A very similar spectrum was observed for the conversion of cat-II-tmm with pentafluorophenol (58) under the same conditions (Chapter 4.3). The corresponding 19F{1H}-NMR spectrum shows the generation of several signals with different multiplicities at different chemical shifts suggesting the formation of multiple different fluorinated species. The main signals are found at −139.81 ppm and −164.51 ppm, the latter being generated by residual 63. No formation of hydrido ligands could be observed according to the 1H-NMR spectrum.

To review the necessity of the applied reaction parameters on the conversion of cat-II-tmm to the observed species, the reaction was also conducted in the absence of H2 at 160 °C and under H2 pressure at room temperature (Scheme 6). In both cases the corresponding 1H-, 31P{1H}- and 19F{1H}-NMR spectra indicate no conversion of cat-II-tmm with 63. After the treatment of 63 at

160 °C in the absence of H2 as well as under H2 pressure at room temperature and at 160 °C, no significant changes were observed in the 19F{1H}-NMR spectrum either. These experiments show that both elevated temperatures and hydrogen are necessary for the conversion of cat-II-tmm with 63.

Scheme 6: Conversion of cat-II-tmm with HFB (63) under different reaction conditions (left) and treatment of HFB (63) with hydrogen at rt or at 160 °C in the absence of hydrogen (right) in THF-d8.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

The applied reaction temperature, time and hydrogen pressure were varied to investigate if the observed main species could be isolated. Under most of the tested conditions, both species were present in the reaction mixture. Generally, the species generating the multiplet at 50.5–49.4 ppm was obtained in larger quantities than the species generating the multiplet at 37.4–35.5 ppm at higher reaction temperatures, longer reaction times and higher hydrogen pressures. These observations lead to the conclusion that the species generating the multiplet at 37.4–35.5 ppm is formed prior to its subsequent transformation into the species generating the multiplet at 50.5– 49.4 ppm.

Scheme 7: Conversion of cat-II-tmm with HFB (63) to [{Ru(triphos-xyl)}2(µ-F)3]+ (cat-II-63a).

Fine-tuning of the reaction conditions successfully enabled the selective synthesis and isolation of three different species. The conversion of 100 mg cat-II-tmm with 39.1 mg 63 (two equivalents) in

5 mL THF at 140 °C for 30 min with 30 bar initial H2 pressure (Scheme 7) led to the precipitation of a yellow powder, which was washed with THF and n-pentane and dried in vacuo. The obtained substance was found to be air-stable in solid form over months stored in a glass vial at room temperature on the laboratory bench and as CD2Cl2 solution for days according to 31P{1H}-NMR spectroscopy. The triphos-xyl ligand generates a multiplet at 37.4–35.5 ppm in the 31P{1H}-NMR spectrum (Figure 62). A very similar signal was observed by Perutz for trifluorine bridged cationic ruthenium complexes with three phosphine ligands per ruthenium center.[100] These complexes were synthesized by conversion of cis-[RuH2(PR3)4] with NEt3·3HF in THF. This analogy suggests that HF is formed from 63 under the applied reaction conditions resulting in the formation of the cationic trifluorine bridged ruthenium triphos-xyl complex [{Ru(triphos-xyl)}2(µ-F)3]+ cat-II-63a.

Figure 62: 31P{1H}-NMR spectrum of cat-II-63a measured in CD2Cl2 at rt. 86

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

BF4− (µ-F)3

Figure 63: Excerpts from the 19F{1H}-NMR spectrum of cat-II-63a measured in CD2Cl2 at rt.

The 19F{1H}-NMR spectrum shows a multiplet at −337.47–−339.52 ppm generated by the bridging fluorine atoms (Figure 63), which is in alignment with the observations by Perutz.[100] Two signals at

−153.47 and −153.52 ppm are generated by a BF4− counter anion and represent the 10B and 11B isotopes. BF4− is assumed to form by the reaction of the generated HF with the borosilicate glass inlet in the autoclave. An additional signal at −139.54 ppm could not yet be successfully assigned.

Single-crystal X-ray diffraction measurements revealed F− and PO2F2− as counter anions (Figure 64).

PO2F2− has been reported to be generated from PF6− with traces of water.[101] PF6− is likely to be formed by partial decomposition of the triphos-xyl ligand as will be described for cat-II-63b. Consequently, cat-II-63a is generated as a mixture with different anions.

Figure 64: Single crystal structure of cat-II-63a with PO2F2− as counter anion (hydrogen atoms and solvent molecules are omitted for clarity).

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Figure 65: 1H-NMR spectrum of cat-II-63a measured in CD2Cl2 at rt.

In the 1H-NMR spectrum, the 72 protons of the xylyl-CH3 groups generate a singlet at 1.75 ppm and a broad singlet at 1.54 ppm (Figure 65). Cooling of the sample to 273 K leads to a merging of these signals indicating that the split of the signals at room temperature occurs due to a temperature- related decrease in symmetry (Figure 66). The three CH2 groups of the triphos-xyl backbone generate two groups of multiplets at 2.37–2.19 ppm and 1.88–1.80 ppm. A multiplet at 1.36–

1.29 ppm is generated by the CH3 group of the triphos-xyl backbone. The ortho-CH groups generate a broad signal at 7.81–7.18 ppm and the para-CH groups a multiplet at 6.74–6.66 ppm. The split of most of the signals, which was also reported for the µ-H3 dimer,[69a] suggests an unsymmetric structure of the dimeric complex. Additionally, for samples of less purity, a very broad signal at 14.03 ppm can be observed, which could be generated by residual HF or a reaction product thereof. High-resolution mass spectrometry revealed a m/z value of 1848.626, which corresponds to cat-II-63a+ (theoretical value: m/z = 1848.637).

The species cat-II-63b generating the multiplet at 50.5–49.4 ppm in the 31P{1H}-NMR spectrum could be isolated by the conversion of 200 mg cat-II-tmm with 78.2 mg 63 (two equivalents) in

5 mL THF at 160 °C for 24 h with 50 bar initial H2 pressure (Scheme 8).

Figure 66: 1H-NMR spectrum of cat-II-63a measured in CD2Cl2 at 273 K. 88

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Scheme 8: Conversion of cat-II-tmm with HFB (63) and H2.

A yellow powder precipitated, which was worked up analogously to cat-II-63a. The 1H- and 13C{1H}- APT-spectra revealed no additional signals to those generated by the triphos-xyl ligand. The 19F{1H}- NMR spectrum shows traces of a signal at −133.56 ppm and a broad signal at −155.10–−165.00 ppm (Figure 67). The attempt to diminish the apparent dynamics by cooling of the sample led to crystallization of the substance impeding the measurement. The corresponding 31P{1H}-NMR spectrum reveals a septet with a coupling constant of 20.1 Hz suggesting a coupling of the phosphorous nuclei with six magnetically equivalent fluorine nuclei. According to the chemical shift of this signal, it is most likely generated by the triphos-xyl phosphorous atoms. At 47.7–46.9 ppm, a multiplet of very low intensity is found, which represents traces of a second generated species.

A 1H-DOSY-NMR spectrum was measured for a mixture of the two 63 derived compounds (Figure 68). The results reveal diffusion coefficients of 5.62·10−10 m2/s for cat-II-63a and 6.86·10−10 m2/s for cat-II-63b. Using the Stokes-Einstein equation, the corresponding hydrodynamic radii were determined. As expected, the hydrodynamic radius of 9.0 Å of cat-II-63a is in the range of a dimeric species. In contrast, a hydrodynamic radius of 7.4 Å for cat-II-63b suggests a monomeric structure.[102] It was also investigated whether the isolated compound cat-II-63a could be converted into this monomeric species as the previous observations suggest the formation of cat-II-63a prior to the formation of cat-II-63b.

Figure 67: 19F{1H}-NMR spectrum (left) and 31P{1H}-NMR spectrum (right) of cat-II-63b measured in CD2Cl2 at rt. 89

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Figure 68: 1H-DOSY-NMR spectrum of a mixture of cat-II-63a and cat-II-63b measured in CD2Cl2 at rt.

However, the treatment of cat-II-63a in THF at 160 °C with an initial hydrogen pressure of 50 bar led to no observable conversion according to 31P{1H}-NMR spectroscopy. Consequently, excess 63 is mandatory to allow the formation of cat-II-63b.

3 Single-crystal X-ray diffraction revealed the structure of the complex to be [Ru(triphos-xyl)(η -PF6)]

(Figure 69). Consequently, the PF6-ligand is generated by partial decomposition of the triphos-xyl ligand.

Figure 69: Single crystal structure of cat-II-63b (hydrogen atoms and solvent molecules are omitted for clarity).

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

The NMR spectroscopic signal representing the phosphorous nucleus of PF6− of cat-II-63b could not be detected by 31P{1H}-NMR measurements in chemical shift ranges between 1350 and −1350 ppm.

While cat-II-63a was found to be stable under air in solid form for months and as CD2Cl2 solution for several days according to 31P{1H}-NMR spectroscopy, cat-II-63b showed rapid decomposition in solution indicated by color change.

When the reaction temperature was raised to 180 °C for the conversion of 100 mg cat-II-tmm with 39.1 mg 63 (2 eq) in 5 mL THF for 2 h, a red powder precipitated, which could be isolated by washing with THF and n-pentane and drying in vacuo. It was identified as the hydrogen bridged dimer [{Ru(triphos-xyl)}2(µ-H)3]+ according to 1H- and 31P{1H}-NMR spectroscopy.[69a]

The conversion of cat-II-tmm with two equivalents of 66 or 67 was first approached analogously to the NMR scale experiments performed with 63. No conversions could be observed by heating the reaction mixtures in toluene-d8 to up to 100 °C or in THF-d8 at up to 50 °C.

Scheme 9: Conversion of cat-II-tmm with HCB (66) or HBB (67) to cat-II-66a or cat-II-67a.

In contrast, the treatment of the THF-d8 solution at 160 °C for 30 min with 50 bar initial H2 pressure led to the precipitation of a yellow solid in the case of 66 and a brown solid when using 67. Analogously to the results using 63, no conversions were observed when treating cat-II-tmm and

66 or 67 at room temperature with 50 bar H2 or at 160 °C in the absence of hydrogen. Additionally, the treatment of 66 or 67 at room temperature with an initial hydrogen pressure of 50 bar showed no impact on the perhalogenatobenzenes. The precipitates could be obtained in larger quantities by conversion of 100 mg cat-II-tmm with two equivalents of 66 or 67 at 160 °C in THF for 2 h under an initial hydrogen pressure of 50 bar in 5 mL THF (Scheme 9). The solids were washed with THF and n-pentane and subsequently dried in vacuo. The obtained powders were found to be air-stable in solid form over months stored in glass vials at room temperature on the laboratory bench according to 1H-NMR spectroscopy.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Figure 70: 31P{1H}-NMR spectra of cat-II-66a (left) and cat-II-67a (right) measured in CD2Cl2 at rt.

Among the standard deuterated solvents, the solids were found to be only soluble in CD2Cl2. However, the conversion to different compounds both generating singlets at ~27 ppm in the

31P{1H}-NMR spectra could be observed when dissolving the powders in CD2Cl2. The transformation of the 66 derived complex cat-II-66a was found to proceed slower than of cat-II-67a. Consequently, the NMR spectra could only be measured for very short times after dissolving the complexes and no pure 13C{1H}-NMR spectra could be obtained. In both cases, no signals additional to those generated by the triphos-xyl ligand could be observed in the 1H-NMR spectrum and one singlet at 50.2 ppm was generated in the 31P{1H}-NMR spectra obtained by fast measurements (Figure 70). The elemental analysis for cat-II-66a resulted in irreproducible results as very different values were obtained in three measurements.

As the formation of a ruthenium dichlorido species seems likely by decomposition of hexachlorobenzene, [Ru(triphos-xyl)Cl2] was synthesized by a different procedure for comparison (Scheme 10). The substitution of three triphenylphosphine ligands by triphos-xyl led to the formation of the desired dichlorido species according to elemental analysis. When performing NMR measurements in CD2Cl2, the compound shows the same behavior as cat-II-66a, generating the same signals in the 1H- and 31P{1H}-NMR spectrum with the tendency to further react to a species generating a singlet at 27.1 ppm over time.

Scheme 10: Synthesis of [Ru(triphos-xyl)Cl2] from [Ru(PPh3)3Cl2].

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Figure 71: 1H-DOSY-NMR spectrum of cat-II-66a (left) and cat-II-67a (right) and the corresponding reaction products cat-II-66b and cat-II-67b measured in CD2Cl2.

Due to these analogies, the complex cat-II-66a is very likely to be identical with

[Ru(triphos-xyl)Cl2]. Due to the similar NMR spectra of cat-II-67a compared to cat-II-66a,

[Ru(triphos-xyl)Br2] is suggested as structure of the complex, which is in accord with the obtained elemental analysis values with a slightly deviating value for the carbon content.

As mentioned before, both compounds react when dissolved in DCM. For both cat-II-66a and cat-II-67a, 1H-DOSY-NMR spectra were recorded in CD2Cl2 to determine their diffusion coefficients as well as those of the resulting species cat-II-66b and cat-II-67b (Figure 71). The recorded spectra reveal diffusion coefficients of 7.59·10−10 m2/s for cat-II-66a and 5.92·10−10 m2/s for cat-II-66b translating to hydrodynamic radii of 6.7 Å and 8.6 Å, respectively. Diffusion coefficients of 1.07·10−9 m2/s and 6.17·10−10 m2/s were calculated for cat-II-67a and cat-II-67b, respectively translating to hydrodynamic radii of 4.7 Å and 8.2 Å. These results suggest monomeric structures of cat-II-66a and cat-II-67a, while cat-II-66b and cat-II-67b are indicated to be dimeric compounds. The conversion of cat-II-66a DCM resulted in only 50% conversion to cat-II-66b after 6 days, while stirring cat-II-67a in DCM at room temperature for 26 h led full conversion to cat-II-67b according to 31P{1H}-NMR spectroscopy (Scheme 11). A yellow solution was obtained, from which a white solid precipitated. After removal of the precipitate, a yellow solid could be isolated by removal of the solvent in vacuo. The solid was washed with n-pentane and dried in vacuo to yield a yellow powder. The isolated species cat-II-67b generates a singlet at 27.3 ppm in the 31P{1H}-NMR spectrum (Figure 72).

Scheme 11: Conversion of cat-II-tmm with HBB (67) and subsequent conversion of the obtained product in DCM at rt to cat-II-67b.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Figure 72: 1H- (left) and 31P{1H}-NMR (right) spectra of cat-II-67b measured in CD2Cl2 at rt.

Similar to cat-II-63a, two signals appear for the CH2- and the xyl-CH3-groups of the triphos-xyl ligand in the 1H-NMR spectrum, indicating a dimeric structure. Single-crystal X-ray diffraction revealed the structure of cat-II-67b to be the tribromido bridged complex

[{Ru(triphos-xyl)}2(µ-Br)3]Br (Figure 73). Analogously, it is likely that cat-II-66a reacts in solution to form the corresponding dimer [{Ru(triphos-xyl)}2(µ-Cl)3]Cl cat-II-66b. The 31P{1H}-NMR spectroscopic data for cat-II-66a and cat-II-66b are also in accordance with spectra recorded by Andrey Charkovskiy for the corresponding complexes.[69b]

The catalytic performances of the five isolated perhalogenatobenzene derived compounds were compared for the hydrogenation of 2-MGBL (22) using 1 mL substrate and 0.1 mol% catalyst for 24 h (Table 29).

Figure 73: Single crystal structure of cat-II-67b (hydrogen atoms are omitted for clarity). 94

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Table 29: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using catalysts cat-II-63a, cat-II-63b, cat-II-66a, cat-II-67a and cat-II-67b[a]

entry [Ru] Y (24) [%][b] Y (23) [%][b]

1 cat-II-63a 95 2 2 cat-II-63b 90 2 3 cat-II-66a 53 3 4 cat-II-67a 0 0 5 cat-II-67b 0 0 [a] General conditions: 1 mL 2-MGBL (22), 0.1 mol% [Ru] (n% [Ru] equals 0.5 n% cat-II-63a and cat-II67b), 160 °C, 24 h. 90 bar initial H2 pressure at rt. [b] Determined by GC using ethyl heptanoate as internal standard. The catalyst concentration is referred to the equivalents of ruthenium in the complexes, therefore n% [Ru] equals 0.5 n% for cat-II-63a and cat-II-67b. The obtained results show over 90% yield of 2-MBDO (24) using the HFB (63) derived complexes cat-II-63a and cat-II-63b (entries 1 and 2). A significant drop in activity could be observed when using the dichlorido complex cat-II-66a (entry 3) and no product formation was observed for the two HBB (67) derived complexes cat-II-67a and cat-II-67b (entries 4 and 5). Using cat-II-63a, the corresponding pressure course reveals a very shallow pressure drop for ~40 min after the reach of the plateau before the reaction significantly speeds up indicated by a very steep drop in pressure for ~160 min (Figure 74).

120

100

[bar]

80 cat-II-63a p cat-II-63b 60

40 0 100 200 300 400 t [min]

Figure 74: Excerpts from the pressure courses for the hydrogenation of 2-MGBL (22) using cat-II-63a and cat-II-63b. Reaction conditions: 1 mL 2-MGBL (22), 0.1 mol% [Ru], 160 °C, 90 bar bar initial H2 pressure at rt, 24 h. n% [Ru] equals 0.5 n% cat-II-63a. Starting pressure was normalized 90 bar.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Consequently, an induction period is required to activate the dimeric complex. In contrast, cat-II-63b is readily active under the applied hydrogenation conditions. In both cases, the pressure courses reach a plateau significantly earlier than the applied reaction time, indicating the completion of the hydrogenation reaction.

5.2 Hydrogenation of Acids, Esters and Amides with a Trifluorine Bridged Dimeric Ruthenium Triphos-xyl Complex

The trifluorine bridged complex cat-II-63a was generated by the conversion of cat-II-tmm with 63 and hydrogen, a combination which was shown to provide improved activity in the hydrogenation of 2-MGBL (22) (Chapter 4.2). cat-II-63a was found to be air-stable in solid form over months and as CD2Cl2 solution for days according to 31P{1H}-NMR spectroscopy. The highly stable complex could consequently be weighed into the glass inlet under air together with high-boiling substrates. After insertion of the glass inlet and a magnetic stir bar, the autoclave was evacuated, pressurized with hydrogen and inserted into a preheated aluminum cone for the applied reaction time. Low-boiling substrates were added to the autoclave under argon countercurrent flow after the evacuation. To most solid substrates, 1,4-dioxane was added as solvent. As cat-II-63a is a ruthenium dimer complex, catalyst concentrations are reported referring to ruthenium centers abbreviated as [Ru] for better comparison with the performance of monomeric complexes.

In contrast to the cat-II-tmm/58 and the cat-II-tmm/63 in situ systems (Chapter 4.2, Table 17), cat-II-63a was able to successfully catalyze the hydrogenation of 22 to 24 at catalyst concentrations as low as 0.02 mol% leading to a 2-MBDO (24) yield of 77% and 0.01 mol% with 63% 24 after 72 h (Table 30, entries 1 and 2).

Table 30: Hydrogenation of 2-MGBL (22) to 2-MBDO (24) and dehydrogenation of 24 to 22 and 3-MGBL (23) using cat-II-63a.[a]

entry n% [Ru] [mol%][b] t [h] Y (24) [%][c] Y (23) [%][c]

1[d] 0.01 72 63 10 2 0.02 72 77 8 3 0.1 1 0 0 4 0.1 24 95 2

[a] General conditions: 1 mL 2-MGBL (22), cat-II-63a, 160 °C, 90 bar initial H2 pressure at rt. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using ethyl heptanoate as internal standard. [d] 10.0 mmol 22. 96

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Increasing the concentration to 0.1 mol% resulted in no conversion after 1 h, but nearly full conversion after 24 h (entries 3 and 4). This observation supports the assumptions drawn from the corresponding pressure course (Chapter 5.1, Figure 74).

Table 31: Hydrogenation of IA (10) to 2-MGBL (22), 3-MGBL (23) and 2-MBDO (24) using cat-II-63a.[a]

entry n% [Ru] [mol%][b] t [h] T [°C] Y (22) [%][c] Y (23) [%][c] Y (24) [%][c]

1 0.1 72 160 32 61 3 2 0.1 20 200 18 12 57 3 0.2 72 160 11 11 78

[a] General conditions: 1 mmol IA (10), cat-II-63a, 50 bar initial H2 pressure at rt, 1 mL 1,4-dioxane. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using ethyl heptanoate as internal standard. cat-II-63a was also found to enable good to high 2-MBDO (24) yields in the hydrogenation of IA (10). A yield of 3% 24 at a concentration of 0.1 mol% after 72 h at 160 °C could be increased to 57% by increasing the temperature to 200 °C or to 78% by doubling of the catalyst concentration at unchanged temperature (Table 31).

Table 32: Hydrogenation of 64 and 65 to 2-MGBL (22), 3-MGBL (23) and 2-MBDO (24) using cat-II-63a.[a]

entry substrate n% [Ru] [mol%][b] t [h] T [°C] Y (22) [%][c] Y (23) [%][c] Y (24) [%][c]

1 64 0.02 72 160 2 1 0 2 64 0.01 72 200 2 4 0 3[d] 64 0.1 20 160 2 2 66 4[e] 64 0.1 20 160 0 0 0 5 65 0.02 72 160 1 1 0 6 65 0.01 72 200 2 1 0 7 65 0.1 20 160 9 13 71 8[e] 65 0.1 20 160 0 0 0

[a] General conditions: 10 mmol substrate, cat-II-63a, 160 °C, 150 bar initial H2 pressure at rt. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using ethyl heptanoate as internal standard. [d] Reproducibly, only a n%-balance of 69– 71% could be obtained. [e] 1 mmol substrate, 50 bar initial H2 pressure at rt.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

The hydrogenation of the itaconate esters DMI (64) and DBI (65) also required a [Ru] concentration of 0.1 mol% to achieve yields of 66% and 71% 24 at 160 °C after 20 h (Table 32, entries 3 and 7). Lowering the catalyst concentration to 0.01 mol% at 200 °C or 0.02 mol% at 160 °C led to no formation or only traces of 24 (entries 1, 2, 4 and 5). Surprisingly, also no diol formation could be observed when lowering the substrate amounts to 1 mmol at a catalyst concentration of 0.1 mol% (entries 4 and 8). For both substrates, only the hydrogenation of the C-C double bond could be observed in this case.

The hydrogenation of 9 and 16 using cat-II-63a was successful at a low catalyst concentration of 0.01 mol% leading to 1,4-PDO (17) yields of 89% and 96% after 48 h at 160 °C (Table 33, entries 1 and 2). In contrast, using the methyl, ethyl and butyl levulinate esters as substrates required an increase in temperature or catalyst concentration to reach good 1,4-PDO (17) yields. Using 0.01 mol% cat-II-63a for the hydrogenation of 68 at 160 °C led to no formation of 17 after 72 h (entry 3). An increase in temperature to 200 °C increased the yield to 10% (entry 4). A high yield of 81% could be achieved with a [Ru] concentration of 0.1 mol% at 160 °C after 20 h (entry 5).

Table 33: Hydrogenation o LVA (9), GVL (16), ML (68), EL (69) and BL (70) to GVL (16) and 1,4-PDO (17) using cat-II-63a.[a]

entry substrate n% [Ru] [mol%][b] t [h] T [°C] Y (16) [%][c] Y (17) [%][c]

1 9 0.01 48 160 6 89 2 16 0.01 48 160 - 96 3 68 0.01 72 160 74 0 4 68 0.01 72 200 79 10 5 68 0.1 20 160 8 81 6 69 0.01 72 160 69 15 7 69 0.01 72 200 18 41 8 69 0.1 20 160 5 85 9 70 0.01 72 160 81 6 10 70 0.1 20 160 6 86

[a] General conditions: 10 mmol substrate, cat-II-63a, 100 bar initial H2 pressure at rt. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using 1-hexanol as internal standard.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

Using 69, a 1,4-PDO (17) yield of 15% was obtained with 0.01 mol% catalyst concentration at 160 °C after 72 h, which could be increased to 41% at 200 °C or to 85% at a [Ru] concentration of 0.1 mol% at 160 °C after 20 h (entries 6–8). Analogously, 70 could be converted to 24 with a yield of 86% at 0.1 mol% cat-II-63a at 160 °C after 20 h, but only 6% were achieved at 0.01 mol% [Ru] concentration after 72 h (entries 9 and 10).

Table 34: Hydrogenation of MV (71) to 1-PeOH (72) using cat-II-63a.[a]

entry n% [Ru] [mol%][b] t [h] T [°C] Y (72) [%][c]

1 0.01 72 160 25 2 0.01 72 200 4 3 0.1 20 160 95

[a] General conditions: 10 mmol MV (71), cat-II-63a, 160 °C, 100 bar initial H2 pressure at rt. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using 1-hexanol as internal standard.

The hydrogenation of methyl valerate (MV) 71 with 0.01 mol% [Ru] at 160 °C led to a yield of 25% 1-pentanol (1-PeOH) 72 after 72 h, while an increase in temperature to 200 °C suppressed the reaction (Table 34, entries 1 and 2). An increase in [Ru] concentration to 0.1 mol% led to 95% yield after 20 h at 160 °C (entry 3).

Table 35: Hydrogenation of AA (73) to AN (74), EAN (75) and DEAN (76) using cat-II-63a.[a]

entry n% [Ru] [mol%][b] t [h] T [°C] Y (74) [%][c] Y (75) [%][c] Y (76) [%][c]

1 0.01 72 160 36 17 1 2 0.01 72 200 10 3 0 3 0.1 20 160 66 32 1

[a] General conditions: 10 mmol AA (73), cat-II-63a, 100 bar initial H2 pressure at rt, 1 mL 1,4-dioxane. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using n-dodecane as internal standard.

5 Ruthenium Triphos-Xyl Based Complexes Derived from Perhalogenatobenzenes

The hydrogenation of acetanilide (AA) 73 with 0.01 mol% cat-II-63a at 160 °C resulted in 36% aniline (AN) 74 and 17% ethylaniline (EA) 75 yield (Table 35, entry 1). An increase in temperature to 200 °C decreased both yields while an increase in [Ru] concentration to 0.1 mol% at unchanged temperature increased the yields to 66% 74 and 32% 75 (entries 2 and 3). In all cases only traces of diethylaniline (DEAN) 76 were detected.

Table 36: Hydrogenation of BA 77, BAM 78 and MB 79 to BAL 80 using cat-II-63a.[a]

entry substrate n% [Ru] [mol%][b] t [h] T [°C] Y (80) [%][c]

1[d,e] 77 0.1 20 160 0 2[e] 78 0.1 20 160 45 3 79 0.1 20 160 98 4 79 0.01 72 160 1 5 79 0.01 72 200 6

[a] General conditions: 10 mmol substrate, cat-II-63a, 100 bar initial H2 pressure at rt. [b] n% [Ru] equals 0.5 n% cat-II-63a. [c] Determined by GC using n-dodecane as internal standard. [d] 1 mmol substrate, 50 bar initial H2 pressure at rt. [e] 1 mL 1,4-dioxane.

The hydrogenation of 1 mmol benzoic acid (BA) 77 with 0.1 mol% [Ru] and 50 bar initial H2 pressure was unsuccessful after 20 h at 160 °C (Table 36, entry 1). The analogous reaction with 10 mmol benzamide (BAM) 78 as substrate and an initial hydrogen pressure of 100 bar led to a yield of 45% for benzyl alcohol (BAL) 80 (entry 2). Using methyl benzoate (MB) 79 as substrate led to a BAL (80) yield of 98% with 0.1 mol% catalyst at 160 °C after 20 h (entry 3). A decrease in catalyst concentration to 0.01 mol% resulted in no significant formation of 80 and only 6% were obtained when increasing the temperature to 200 °C (entries 4 and 5).

100

6 Summary

The design and tailoring of highly active and selective catalytic systems for the hydrogenation of biogenic platform chemicals enables the development of more sustainable production routes in the chemical industry.

Within this work, the successful isolation and characterization of different ruthenium triphos-xyl catalysts was achieved. A number of [Ru(triphos-xyl)(OR)2] type complexes were synthesized by conversion of [Ru(triphos-xyl)(tmm)] cat-II-tmm with different acids, phenols or naphthols and characterized.

Figure 75: Synthesis of [Ru(triphos-xyl)(OR)2] complexes and application as catalysts in the hydrogenation of LVA (9), IA (10) as well as the corresponding lactones and selected itaconate esters. R=H: IA (10), R=Me: DMI (64), R=nBu: DBI (65).

101

6 Summary

The complexes were applied in the hydrogenation of the biogenic platform chemicals levulinic acid (9) and itaconic acid (10) as well as the corresponding lactones and itaconate esters (Figure 75). The performance of the different catalysts was investigated by gas chromatographic analysis of the product mixtures and evaluation of detailed kinetic experiments. The superior performance of cat-II-tmm as compared to the established cat-I-tmm was observed for the hydrogenation of 9 both in batch as well as reaction cycles.

Testing the acetato, phenolato and naphtholato complexes in the hydrogenation of γ-valerolactone 16 revealed a very low activity for the 2,3-dihydroxynaphthalene derived complex cat-II-49. However, all other complexes successfully catalyzed the reaction, leading to higher yields for 1,4-pentanediol (17) at 0.1 mol% catalyst concentration and 160 °C after 5 h with an initial hydrogen pressure of 100 bar. In the substrate reloading experiments, the best performance could be observed using cat-II-50 obtained from 2,2’-biphenol. Additionally, a correlation of the loss in activity in multiple hydrogenation runs and thermal stress on the catalysts was discovered, an important factor to be considered in an optimization of the reaction and process conditions in continuous systems. The acetato, phenolato and naphtholato complexes led to comparable or higher yields of 17 in the hydrogenation of 9 as compared to cat-II-tmm in batch experiments. cat-II-tmm and cat-II-50 could also be applied in the dehydrogenation of 17 to 16, which revealed the existence of a lactone-diol hydrogenation-dehydrogenation equilibrium. Shifting this equilibrium to the desired species is of major importance when optimizing the reaction conditions and designing a scaled-up process. It was found that dissolving cat-II-tmm and cat-II-44–cat-II-50 in 9 led to the generation of the bis(levulinato) complex cat-II-9, which could also be successfully isolated and characterized. Selected fluorinated phenolato and naphtholato complexes performed superior in comparison to the previously investigated species in multiple hydrogenation cycles of 16.

Application of cat-II-tmm in the hydrogenation of 2-methyl-γ-butyrolactone (22) led to a significant improvement in activity as compared to cat-I-tmm. An even further increase was achieved by using the isolated fluorophenolato and -naphtholato complexes. Importantly, the in situ systems of cat-II-tmm with the corresponding fluorophenols and -naphthol improved the performance even more. In the course of these experiments, also hexafluorobenzene 63 was discovered to raise the achieved yield of 2-methyl-1,4-butanediol (24). The presence of 3-methyl-γ-butyrolactone (23) could be observed in most product mixtures, suggesting the existence of a lactone-diol equilibrium, which was confirmed by the successful dehydrogenation of 24 using cat-II-58. The equilibrium could be shifted to the side of 24 by increasing the initial hydrogen pressure. Further optimization of the reaction conditions led to a yield of 91% after 1.5 h at 160 °C and 90 bar initial hydrogen

102

6 Summary

pressure using 1 mmol 22, 0.1 mol% cat-II-tmm and 0.2 mol% 58. The in situ system could also be successfully applied in the hydrogenation of 10 and two itaconate esters. NMR-spectroscopic studies revealed the formation of several itaconato species by conversion of selected catalysts with 10 in the absence of hydrogen. A reactivation of the otherwise inactive monocarbonyl dihydrido complex cat-II-COH2 was achieved by the addition of acidic fluorophenols and –naphthol to the hydrogenation of 22. The application of complexes bearing chloro- or bromosubstituted phenols or the corresponding in situ systems led to decreased activities as compared to the fluorinated species. The same applied when using hexachlorobenzene 66 or hexabromobenzene 67 as additives.

The conversion of cat-II-tmm with the perhalogenated benzenes 63, 66 and 67 and hydrogen at elevated temperatures in the presence of hydrogen led to the synthesis and isolation of different halogenated complexes. Optimization of the reaction conditions for the conversion of cat-II-tmm with 63 and hydrogen led to the selective formation of two compounds which could be characterized as [{Ru(triphos-xyl)}2(µ-F3]+ cat-II-63a, with BF4−, F− and PO2F2− as possible counter anions, and [Ru(triphos-xyl)(η3-PF6)] cat-II-63b. cat-II-63a was subsequently applied in the hydrogenation of various compounds (Figure 76). It was found to successfully catalyze the hydrogenation of 9 and 16 to 17 at low catalyst concentrations achieving yields of 89% and 96%, respectively. Three levulinate esters could also be hydrogenated to over 80% 17 using cat-II-63a. Good to high yields of 24 were also achieved when using cat-II-63a in the hydrogenation of 10, 22 and two itaconate esters. Moreover, 95% yield of 1-pentanol (72) could be achieved in the hydrogenation of methyl valerate (71) with cat-II-63a. The hydrogenation of acetanilide (81) led to the generation of 66% aniline (82) and 32% ethylaniline (83) and a yield of 98% benzyl alcohol (80) was achieved for the hydrogenation of methyl benzoate (79) using cat-II-63a.

Figure 76: Synthesis of [{Ru(triphos-xyl)(µ-F)3]+ and application as catalyst in the hydrogenation of selected substrates.

Based on these important findings, the application of the novel catalysts on larger scale can be envisioned. Moreover, the novel catalyst lead structures pave the way to effective reduction reactions of challenging substrates.

103

7 Experimental Section

7.1 General Experimental Procedures

Experimental Procedures

All synthetic and catalytic procedures with air sensitive reaction components were carried out under inert gas and performed in previously evacuated glassware and stainless steel autoclaves that were refilled with inert gas at least three times unless otherwise stated. Argon 4.8 by Westfalen was used as inert gas. In case of water sensitive components, the glassware was also heated to 650 °C while being evacuated. Solids to be handled under inert gas atmosphere were transferred in an argon filled glove box. Liquids to be handled under inert gas atmosphere were transferred in an argon filled glove box or transferred in argon counter current flow using syringes that have been rinsed with argon at least three times. The experimental procedures reported in chapters 2, 3 and 4 were partly carried out by Julia Nowacki in the course of an apprenticeship as lab technician and by Alexander Schmitz and Eren Temur in the course of research internships under the guidance of Celine Jung.

Preparation, storage and handling of chemicals and organic solvents

Unless otherwise stated, all chemicals were purchased from the chemical suppliers Sigma Aldrich, abcr, TCI, Acros Organics, Carl Roth, Fisher Scientific, Fluka and Alfa Aesar and used without further purification. Dimethyl itaconate was sublimed and stored under argon atmosphere prior to application. Chemicals to be handled under inert gas atmosphere were degassed by freeze-pump- thaw cycling or by removal of air in vacuo followed by flushing with argon at least three times. Bis(2-methylallyl)(1,5-cyclooctadiene)ruthenium(II) was kindly donated by Umicore. 1,1,1-tris(diphenylphosphinomethyl)ethane was purchased or supplied by HYBRID CATALYSIS in the course of the GreenSolRes Horizon 2020-BBI-PPP project. Trimethylenemethane-1,1,1-tris(diphenylphosphinomethyl)ethaneruthenium(II) (cat-I-tmm) was synthesized according to literature procedure.[59b] 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane was either supplied by HYBRID CATALYSIS in the course of the GreenSolRes Horizon 2020-BBI-PPP project or synthesized according to literature procedures (see Chapter 7.3). 1,1,1-tris(diphenylphosphinomethyl)propane and 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)propane were supplied by HYBRID CATALYSIS in the course of the GreenSolRes Horizon 2020-BBI-PPP project. Trimethylene- methane-1,1,1-tris(di(4-methylphenyl)phosphinomethyl)ethaneruthenium(II) (cat-V-tmm), tri-

104

7 Experimental Section

methylenemethane-1,1,1-tris(di(4-(trifluoromethyl)phenyl)phosphinomethyl)ethaneruthenium(II) (cat-VI-tmm), trimethylenemethane-1,1,1-tris(di(3,5-di(trifluoromethyl)phenyl)phosphinomethyl)ethaneruthenium(II) (cat-VII-tmm) and trimethylenemethane-1,1,1-tris(di(3,5-dimethoxy- phenyl)phosphinomethyl)ethaneruthenium(II) (cat-VIII-tmm), were prepared and kindly supplied by Stefan Westhues. Larger quantities of cat-II-tmm, cat-II-44–cat-II-48, cat-II-50–cat-II-53, cat-II-55, cat-II-56 and cat-II-59 were prepared and supplied by Sandra Brosinski under the guidance of Celine Jung.

Organic solvents were distilled, degassed by flushing argon through a glass lance equipped with a frit tip and stored over activated molecular sieves (size 4 Å) under argon atmosphere. Deuterated organic solvents to be used under inert gas atmosphere were purchased from Eurisotop, degassed by freeze-pump-thaw cycles and stored over activated molecular sieves (size 4 Å) under argon atmosphere. Molecular sieves were heated to 500 °C for at least 4 hours in a muffle furnace and cooled down to room temperature in vacuo prior to use.

Handling of High Pressure Gases

All reactions under high pressure were carried out with appropriate high pressure equipment and performed under rigorous safety precautions. H2 (5.0) used as reaction gas was purchased from PraxAir. All high pressure experiments were performed in 10 and 20 mL finger autoclaves manufactured by the in-house mechanical workshop from stainless steel (material number 1.45.71) (Figure 77).

Figure 77: Profile drawing of utilized 10/20 mL finger autoclaves by the mechanical workshop of the ITMC, RWTH Aachen University. 105

7 Experimental Section

The autoclaves were equipped with needle valves and digital or analogous manometers. For all the reactions, the autoclaves were equipped with glass inlets from borosilicate glass and magnetic stir bars. After every reaction, the inlets and stir bars were cleaned and replaced if necessary. The autoclaves were heated in aluminum cones that were fitted on magnetic stir plates. For digital pressure recordings, digital manometers were equipped and the pressure courses recorded using the software LabView 14 by National Instruments.

7.2 Analytical Methods

Nuclear Magnetic Resonance Spectroscopy

1H-, 11B-, 13C-, 19F- and 31P-NMR spectroscopic measurements were carried out using Bruker AV-300 (1H-resonance: 300 MHz), Bruker AV-400 (1H-resonance: 400 MHz), Bruker AS-400 (1H-resonance:

400 MHz) and Bruker AV-600 (1H-resonance: 600 MHz) spectrometers. CDCl3, CD2Cl2, THF-d8, toluene-d8 and DMSO-d6 were used as deuterated solvents. The chemical shifts are reported in ppm relative to tetramethylsilane. The signal of the corresponding deuterated solvent is used as calibration frequency.[103] The NMR spectra were evaluated using the software TopSpin 3.6.1 by Bruker. The multiplicities are reported as s (singlet), brs (broad singlet), d (doublet), t (triplet), q (quartet), sept (septet) and m (multiplet).

Mass Spectrometry

ESI-MS and GC-MS analyses were performed by the in-house analytical department. HR-MS analyses were performed with a Thermo Scientific LTQ Orbitrap XL by Thermo ScientificTM at the ABBt at RWTH Aachen University or at the Max-Planck-Institut für Kohlenforschung. FAB-MS analyses were performed at the Institute of Inorganic Chemistry at RWTH Aachen University.

Infrared Spectroscopy

The IR absorbance spectra were acquired under argon atmosphere using a FT IR Alpha P spectrometer by Bruker. The evaluation of the spectra was performed using the software Opus 6.5 by Bruker.

Elemental Analysis

Elemental analyses were performed by Mikrolab Kolbe for the elements C, H, Ru, P, F, Cl, Br, B.

106

7 Experimental Section

Single Crystal X-ray Diffraction

SC-XRD measurements for cat-II-49, cat-II-54 and cat-II-58 were performed by the group of Prof. Dr. Ulli Englert in the department of Inorganic Chemistry at RWTH Aachen University using a D8-Goniometer by Bruker equipped with an APEX-II CCD detector and an Incoatec Microsource X-ray souce (Mo-Kα (λ = 0.71073 Å) radiation). The temperature was controlled using an Oxford Cryostream 700. SC-XRD measurements for cat-II-63a, cat-II-63b and cat-II-67b were performed by the group of Prof. Kari Rissanen in the Department of Chemistry at Jyväskylä University, Finland using a Rigaku SuperNova dual-source Oxford diffractometer equipped with an Eos detector using mirror-monochromated Mo-Kα (λ = 0.71073 Å) radiation. The data evaluation for cat-II-49, cat-II-54 and cat-II-58 was carried out by Dr. Markus Hölscher using SHELXS to solve and SHELXL-97 to refine the crystal structures and by Dr. Khai-Nghi Truong for cat-II-63a, cat-II-63b and cat-II-67b using (SHELXT)2 to solve and SHELXL-2015 to refine the crystal structures.

Gas Chromatography

For quantitative analyses of the products from the catalytic reactions, gas chromatographic analyses were performed. The measurements and method developments were performed by the in-house GC department. The method key parameters, the retention times and correction factors are sorted by reaction and listed below.

Hydrogenation of Levulinic Acid, γ-Valerolactone, Methyl-, Ethyl- and Butyllevulinate

Measurements were partly performed on a Focus-4 GC device by Thermo Fisher Scientific with a FID at 250 °C vaporization temperature, a 50 m CP-Wax-57 column, a He-flow of 2.0 mL/min and a split of 66 mL/min. The injection volume was 1 µL and the measurements were performed under a temperature ramp from 50 to 200 °C. 1-Hexanol was used as internal standard.

Table 37: Retention times and correction factors for the GC-analysis of the compounds from the conversion of levulinic acid and γ-valerolactone using a Focus-4 device.

compound retention time [min] correction factor 1-hexanol (standard) 11.22 - γ-valerolactone 16.33 1.16 1,4-pentanediol 19.85 1.63 levulinic acid 24.89 2.31

107

7 Experimental Section

Measurements were also partly performed on a Trace-3 GC device by Thermo Fisher Scientific with a FID at 250 °C vaporization temperature, a 60 m CP-Wax-52 column, a He-flow of 1.5 mL/min and a split of 50 mL/min. The injection volume was 1 µL and the measurements were performed under a temperature ramp from 50 to 200 °C. 1-Hexanol was used as internal standard.

Table 38: Retention times and correction factors for the GC-analysis of the compounds from the conversion of levulinic acid, levulinates and γ-valerolactone using a Trace-3 device.

compound retention time [min] correction factor methanol 4.72 2.31 ethanol 5.13 1.69 1-butanol 8.99 1.10 1-hexanol (standard) 13.38 - methyl levulinate 17.15 1.89 ethyl levulinate 17.71 1.66 γ-valerolactone 18.30 1.56 butyl levulinate 20.34 1.36 1,4-pentanediol 21.73 1.59 levulinic acid 28.08 2.52

Hydrogenation of Itaconic Acid, Dimethyl Itaconate and Dibutyl Itaconate

Measurements were performed on a Trace-3 GC device by Thermo Fisher Scientific with a FID at 250 °C vaporization temperature, a 60 m CP-Wax-52 column, a He-flow of 1.5 mL/min and a split of 50 mL/min. The injection volume was 1 µL and the measurements were performed under a temperature ramp from 50 to 200 °C. Ethyl heptanoate was used as internal standard.

Table 39: Retention times and correction factors for the GC-analysis of the compounds from the conversion of itaconic acid, dimethyl itaconate and dibutyl itaconate.

compound retention time [min] correction factor ethyl heptanoate (standard) 13.22 - dimethyl methylsuccinate 16.93 1.80 2-methyl-γ-butyrolactone 18.17 1.40 3-methyl-γ-butyrolactone 18.52 1.40 (est.) dimethyl itaconate 18.72 1.77 dibutyl methylsuccinate 22.62 1.31 2-methyl-1,4-butanediol 23.04 1.27 dibutyl itaconate 23.78 1.34

108

7 Experimental Section

Hydrogenation of Methyl Valerate

Table 40: Retention times and correction factors for the GC-analysis of the compounds from the conversion of methyl valerate.

compound retention time [min] correction factor methanol 4.72 2.31 methyl valerate 7.97 1.34 1-pentanol 11.21 1.02 1-hexanol (standard) 13.21 -

Hydrogenation of Methyl Benzoate, Benzoic Acid and Benzamide

Measurements were performed on a Trace-4 GC device by Thermo Fisher Scientific with a FID at 260 °C vaporization temperature, a 50 m Rtx-1-Pona column, a He-flow of 1.5 mL/min and a split of 80 mL/min. The injection volume was 1 µL and the measurements were performed under a temperature ramp from 50 to 250 °C. n-Dodecane was used as internal standard.

Table 41: Retention times and correction factors for the GC-analysis of the compounds from the conversion of methyl benzoate, benzoic acid and benzamide.

compound retention time [min] correction factor methanol 1.92 3.07 benzyl methyl ether 12.75 1.20 cyclohexylmethanol 13.18 1.22 benzyl alcohol 13.73 1.13 methyl cyclohexanecarboxylate 14.42 1.46 methyl benzoate 15.07 1.36 benzamide 16.92 1.55 benzoic acid 17.53 1.51 n-dodecane (standard) 17.60 - dibenzyl ether 24.68 1.11

109

7 Experimental Section

Hydrogenation of Acetanilide

Measurements were performed on a Trace-6 GC device by Thermo Fisher Scientific with a FID at

250 °C vaporization temperature, a 60 m Optima-5 column, a N2-flow of 1.5 mL/min and a split of 50 mL/min. The injection volume was 1 µL and the measurements were performed under a temperature ramp from 50 to 280 °C. n-Dodecane was used as internal standard.

Table 42: Retention times and correction factors for the GC-analysis of the compounds from the conversion of acetanilide.

compound retention time [min] correction factor aniline 17.48 1.17 ethyl aniline 20.86 1.13 n-dodecane (standard) 21.99 - diethyl aniline 22.70 1.10 acetanilide 25.61 1.40

110

7 Experimental Section

7.3 Synthetic Procedures

7.3.1 Synthesis of Bis(3,5-dimethylphenyl)phosphine oxide

41.6 g (225 mmol, 1.00 eq) 1-bromoxylol in 120 mL 2-MTHF was added dropwise to a dispersion of 6.20 g (255 mmol, 1.13 eq) magnesium chips in 75 mL 2-MTHF at 50 °C for 2 h and the reaction mixture stirred for another 30 min after complete addition. Afterwards, it was cooled to 0 °C and 9.32 g (67.4 mmol, 0.30 eq) diethyl phosphite in 65 mL 2-MTHF was added dropwise. The reaction mixture was allowed to slowly warm-up to room temperature overnight. After addition of 75 mL HCl-solution (5 M), the reaction mixture was filtered and the phases were separated. The aqueous phase was extracted with 150 mL Et2O and the combined organic phases were washed with

2x75 mL water and 75 mL brine. The aqueous phase was again extracted with 150 mL Et2O and the resulting organic phase was washed with 2x75 mL water and 75 mL brine. All organic phases were combined, drying over MgSO4 and filtrated. The product was obtained by subsequent evaporation of the solvent as yellow oil with a purity of 90% according to 31P{1H}-NMR spectroscopy.

Yield: 17.83 g ≙ 16.05 g pure product (92.2%)

1H-NMR (300 MHz, CDCl3): δ = 8.66 (m, 1 H, PH), 7.35–7.17 (m, 2 H, CarH), 7.15–7.01 (m, 2 H, CarH), 7.01–6.84 (m, 2 H, CarH), 2.29–2.21 (m, 12 H, CH3) ppm.

31P{1H}-NMR (121 MHz, CDCl3): δ = 22.8 (s, 1 P) ppm.

The synthesis was performed following literature procedures.[69a] The analytic data is in agreement with literature.[104]

111

7 Experimental Section

7.3.2 Synthesis of Bis(3,5-dimethylphenyl)phosphine

19.9 g (77.0 mmol, 1.00 eq) bis(3,5-dimethylphenyl)phosphine oxide was dispersed in 130 mL 2-MTHF and 10 mL THF and added dropwise to a solution of 21.9 mL (122 mmol, 1.58 eq) DIBAL-H in 100 mL 2-MTHF for 1 h. The reaction mixture was stirred for another 2 h, then 200 mL of aqueous NaOH-solution (1 M) was added dropwise. After addition of 60 mL Et2O, the phases were separated and the organic phase was washed with 2x100 mL water. The solvent was removed in vacuo and the product obtained after vacuum distillation at 180 °C as colorless liquid.

Yield: 2.63 g (14.1%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.01 (d, 3JPH = 7.7 Hz, 4 H, Car,orthoH), 6.86 (s, 2 H, Car,paraH), 4.99 (d, 1JPH = 219.1 Hz, 1 H, PH), 2.18 (s, 12 H, CH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = −39.2 (s, 1 P) ppm.

The synthesis was performed following literature procedures.[69a] The analytic data is in agreement with literature.[104]

112

7 Experimental Section

7.3.3 Synthesis of 1,1,1-Tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane

2.00 g (8.25 mmol, 3.30 eq) bis(3,5-dimethylphenyl)phosphine and 1.04 g (9.25 mmol, 3.70 eq) potassium tert-butoxide were dissolved in 20 mL DMSO. After stirring for 30 min, 0.44 g (2.50 mmol, 1.00 eq) 1,1,1-tris(chloromethyl)ethane in 5 mL DMSO was added dropwise. The orange dispersion was stirred overnight at room temperature. After addition of 50 mL water, a white solid precipitated that was separated, dissolved in 20 mL DCM and washed with 10 mL water. The combined aqueous phases were extracted with 2x20 mL DCM. After removal of the solvent from the combined organic phases in vacuo, the white residue was washed with 20 mL EtOH at 60 °C. After removal of the solvent and drying in vacuo, the product was obtained as colorless powder.

Yield: 1.67 g (84.4%)

1H-NMR (300 MHz, CDCl3): δ = 7.00 (d, 3JPH = 7.9 Hz, 12 H, Car,orthoH), 6.85 (s, 6 H, Car,paraH), 2.49–2.40 (m, 6 H, (CH2)3CCH3), 2.24 (s, 36 H, Car,xylCH3), 0.93 (s, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (121 MHz, CDCl3): δ = −26.4 (s, 3 P) ppm.

The synthesis was performed following literature procedures.[64a,68b,69a] The analytic data is in agreement with literature.[64a,68b,69a]

113

7 Experimental Section

7.3.4 Synthesis of Trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-tmm

242 mg (0.76 mmol, 1.00 eq) bis(2-methylallyl)(1,5-cyclooctadiene)ruthenium(II) and 600 mg (0.76 mmol, 1.00 eq) 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane were dissolved in 15 mL toluene and stirred for 15 h at 115 °C. After cooling to room temperature, the solvent was removed from the brown solution in vacuo and the yellow solid dissolved in 3 mL DCM. After removal of the solvent in vacuo, the yellow powder was washed with 2x10 mL n-pentane and dried in vacuo.

Yield: 405 mg (56.2%)

1H-NMR (400 MHz, CD2Cl2): δ = 6.77–6.66 (m, 18 H, CarH), 2.27–2.18 (m, 6 H, (CH2)3CCH3), 1.99 (s, 36 H, Car,xylCH3), 1.72–1.67 (m, 6 H, C(CH2)3), 1.46–1.39 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 31.7 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 32.3 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, toluene-d8): δ = 32.8 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 142.9–141.9 (m, Car,ipso,xyl), 137.5–136.3 (m, Ca,meta,xyl), 130.4– 129.9 (m, Car,ortho,xylH), 129.9–129.3 (m, Car,para,xylH), 107.1–106.9 (m, C(CH2)3), 43.9–42.9 (m, C(CH2)3), 39.7–39.1 (m, (CH2)3CCH3), 38.2–37.9 (m, (CH2)3CCH3), 35.4–34.8 (m, (CH2)3CCH3), 21.3 (s, Car,xylCH3) ppm.

The synthesis was performed following literature procedures.[64a,68b,69a] The analytic data is in agreement with literature.[64a,68b,69a]

114

7 Experimental Section

7.3.5 Synthesis of Trimethylenemethane-1,1,1-tris(diphenylphosphinomethyl)propaneruthenium(II)

Abbreviation: cat-III-tmm

250 mg (0.78 mmol, 1.00 eq) bis(2-methylallyl)(1,5-cyclooctadiene)ruthenium(II) and 500 mg (0.78 mmol, 1.00 eq) 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)propane were dissolved in 15 mL toluene and stirred for 2.5 h at 115 °C. After cooling to room temperature, a yellow solid precipitated from the orange solution by addition of 20 mL n-pentane under stirring. After removal of the supernatant, the yellow powder was washed with 3x5 mL and 1x10 mL n-pentane and dried in vacuo.

Yield: 317 mg (51.2%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.14–7.02 (m, 18 H, CarH), 7.01–6.85 (m, 12 H, CarH), 2.27–2.16 (m, 6 H, (CH2)3CCH2CH3), 1.71–1.57 (8 H, C(CH2)3, (CH2)3CCH2CH3), 1.00 (t, 3JHH = 7.3 Hz, 3 H, (CH2)3CCH2CH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 32.7 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 142.6–141.8 (m, Car), 132.8–132.2 (m, CarH), 128.4–127.9 (m, CarH), 127.9–127.6 (m, CarH), 107.1–106.9 (m, C(CH2)3), 43.8–42.7 (m, (CH2)3CCH2CH3) and C(CH2)3), 41.5–41.3 (m, (CH2)3CCH2CH3), 33.7–33.2 (m, (CH2)3CCH2CH3), 8.2 (s, (CH2)3CCH2CH3) ppm.

HR-MS: m/z = 795.197 (M++H+ calculated: 795.200).

IR: ṽ = 3048.60, 2960.53, 1577.97, 1477.95, 1430.81, 1307.84, 1272.60, 1215.74, 1180.89, 1083.51, 1024.96, 992.34, 946.25, 876.58, 835.00, 798.45, 737.47, 693.25, 546.70, 517.55, 482.43, 462.07 cm−1.

115

7 Experimental Section

7.3.6 Synthesis of Trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)propaneruthenium(II)

Abbreviation: cat-IV-tmm

238 mg (0.74 mmol, 1.00 eq) bis(2-methylallyl)(1,5-cyclooctadiene)ruthenium(II) and 600 mg (0.74 mmol, 1.00 eq) 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)propane were dissolved in 15 mL toluene and stirred for 15 h at 115 °C. After cooling to room temperature, the solvent was removed in vacuo and the resulting yellow solid dissolved in 5 mL DCM. After removal of the solvent in vacuo, the yellow powder was washed with 2x5 mL n-pentane and dried in vacuo.

Yield: 449 mg (63.1%)

1H-NMR (400 MHz, CD2Cl2): δ = 6.77–6.63 (m, 18 H, CarH), 2.23–2.14 (m, 6 H, (CH2)3CCH2CH3), 1.98 (s, 36 H, Car,xylCH3), 1.73–1.62 (m, 8 H, C(CH2)3 and (CH2)3CCH2CH3), 1.06 (t, 3JHH = 7.3 Hz, 3 H, (CH2)3CCH2CH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 30.0 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 142.7–141.8 (m, Car,ipso,xyl), 137.5–136.3 (m, Car,meta,xyl), 130.3– 129.9 (m, Car,ortho,xylH), 129.8–129.4 (m, Car,para,xylH), 107.2–106.9 (m, C(CH2)3), 43.9–42.6 (m, C(CH2)3 and (CH2)3CCH2CH3), 41.2–40.8 (m, (CH2)3CCH2CH3), 33.1–32.0 (m, (CH2)3CCH2CH3), 21.4 (s, Car,xylCH3), 8.2 (s, CH2CH3) ppm.

HR-MS: m/z = 963.388 (M+H+ calculated: 963.388).

IR: ṽ = 2983.07, 2910.43, 2855.28, 1589.35, 1451.52, 1413.58, 1373.67, 1265.21, 1211.23, 1118.35, 1035.84, 997.86, 944.76, 878.53, 838.94, 795.52, 728.16, 691.79, 561.02, 519.66, 479.57, 447.14 cm−1.

116

7 Experimental Section

7.3.7 Synthesis of Carbonyldihydrido-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-COH2

A solution of 147 mg (0.16 mmol, 1.00 eq) carbonyldihydridotris(triphenylphosphine)ruthenium(II), 127 mg (0.16 mmol, 1.00 eq) 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane and sodium methanolate (~5 mg) in 15 mL toluene was stirred at 115 °C for 20 h. After cooling to room temperature, 50 mL n-pentane was added under stirring. After settling of the off-white solid overnight, the brown supernatant was removed and the solid separated into two Schlenk centrifuge tubes. After centrifugation, the supernatant was removed and the solids washed with 2x10 mL n-pentane each, separating solid and supernatant by centrifugation and subsequent removal of the supernatant each time. The solids were combined and dried in vacuo.

Yield: 64.0 mg (43.3%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.33–7.08 (m, 9 H, CarH), 6.99–6.89 (m, 3 H, CarH), 6.79–6.65 (m, 6 H, Car,xylH), 2.26–2.10 (m, 6 H, (CH2)3CCH3), 2.09–2.00 (m, 36 H, Car,xylCH3), 1.50–1.43 (m, 3 H, (CH2)3CCH3), −7.27–−7.64 (m, 2 H, RuH2) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 32.6 (t, 2JPM,PA = 31.9 Hz, 1 PA), 25.0 (d, 2JPM,PA = 31.6 Hz, 2 PM) ppm.

The synthesis was performed following literature procedures.[69a] The analytic data reported for the 31P{1H}-NMR spectrum is in agreement with literature.[69a] The signal generated by the two hydrido ligands could not be highly resolved. In literature, a signal corresponding to the AA’XX’Y system is reported.[105]

117

7 Experimental Section

7.3.8 Synthesis of Bis(levulinato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-9

A solution of 300 mg (0.32 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 73.2 mg (0.63 mmol, 2.00 eq) levulinic acid in 15 mL toluene was stirred at 50 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the yellow solid was dissolved in 5 mL DCM. After removal of the solvent in vacuo, the yellow powder was washed with 2x10 mL n-pentane and dried in vacuo.

Yield: 295 mg (83.1%)

1H-NMR (600 MHz, THF-d8): δ = 7.23–7.10 (m, 12 H, Car,orthoH), 6.74–6.66 (m, 6 H, Car,paraH), 2.63 (t, 3JHH = 6.8 Hz, 4 H, COOCH2), 2.46 (t, 3JHH = 6.8 Hz, 4 H, COCH2), 2.27–2.19 (m, 6 H, (CH2)3CCH3), 2.12 (s, 6 H, COCH3), 2.03 (s, 36 H, Car,xylCH3), 1.54–1.43 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 38.9 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 38.1 (s, 3 P) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 207.2 (s, COO), 182.0 (s, COCH3) 138.1–137.6 (m, Car,ipso,xyl), 137.3–136.9 (m, Car,meta,xyl), 131.0–130.8 (m, Car,para,xylH), 131.0–130.5 (m, Car,ortho,xylH), 39.8 (s, COOCH2), 38.6–38.4 (m, (CH2)3CCH3) , 38.1–37.7 (m, (CH2)3CCH3), 33.7 (s, COCH2), 33.5–32.9 (m, (CH2)3CCH3), 30.0 (s, COCH3), 21.2 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1109.358 (M+−H+ calculated: 1109.374).

IR: ṽ = 2913.15, 2858.88, 1710.93, 1611.41, 1526.90, 1450.52, 1414.18, 1357.68, 1271.18, 1217.05, 1160.74, 1130.73, 1086.25, 1038.98, 922.01, 842.56, 744.40, 692.44, 563.14, 519.04, 481.63, 452.31 cm−1.

118

7 Experimental Section

7.3.9 Synthesis of Bis(acetato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-44

12 µL (0.21 mmol, 2.00 eq) acetic acid in 0.6 mL toluene was added dropwise to a solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) in 9.4 mL toluene. The solution was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the yellow solid was dissolved in 4 mL DCM. After removal of the solvent in vacuo, the yellow powder was washed with 2 mL n-pentane and dried in vacuo.

Yield: 64.1 mg (57.6%)

1H-NMR (600 MHz, THF-d8): δ = 7.25–7.15 (m, 12 H, Car,orthoH), 6.74–6.66 (m, 6 H, Car,paraH), 2.28– 2.18 (m, 6 H, (CH2)3CCH3), 2.02 (s, 36 H, Car,xylCH3), 1.88 (s, 6 H, COCH3), 1.53–1.44 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 38.4 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, toluene-d8): δ = 38.9 (s, 3 P) ppm.

31P{1H}-NMR (121 MHz, CD2Cl2): δ = 38.0 (s, 3 P) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 180.6 (s, CO) 138.5–137.8 (m, Car,ipso,xyl), 137.2–136.8 (m, Car,meta,xyl), 131.3–130.5 (m, Car,para,xylH and Car,ortho,xylH), 38.7–38.4 (m, (CH2)3CCH3) , 38.2–37.7 (m, (CH2)3CCH3) , 33.9–33.1 (m, (CH2)3CCH3), 25.7 (s, COCH3), 21.2 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1012.344595 (M+ calculated: 1012.344287).

IR: ṽ = 2913.60, 2858.65, 1719.99, 1614.42, 1577.57, 1528.10, 1454.63, 1413.87, 1363.40, 1317.09, 1280.55, 1214.11, 1133.12, 1090.60, 1038.04, 1012.08, 994.54, 942.76, 843.80, 819.22, 740.93, 693.89, 675.81, 563.91, 520.70, 479.07, 458.64, 415.73 cm−1.

119

7 Experimental Section

7.3.10 Synthesis of Bis(trifluoroacetato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-45

20 µL (0.21 mmol, 2.00 eq) trifluoroacetic acid was added dropwise to a solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) in 10 mL toluene. The solution was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the yellow solid was dissolved in 4 mL DCM. After removal of the solvent in vacuo, the yellow powder was washed with 3x4 mL n-pentane and dried in vacuo.

Yield: 87.6 mg (71.1%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.12–6.90 (m, 12 H, Car,orthoH), 6.87–6.76 (s, 6 H, Car,paraH), 2.36–2.23 (m, 6 H, (CH2)3CCH3), 2.04 (s, 36 H, Car,xylCH3), 1.66–1.56 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 40.5 (s, 3 P) ppm.

31P{1H}-NMR (121 MHz, THF-d8): δ = 39.9 (s, 3 P) ppm.

19F{1H}-NMR (377 MHz, CD2Cl2): δ = −75.99 (s, 6 F) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 165.5–164.1 (m, CO), 138.1–137.6 (m, Car,meta,xyl), 136.1–135.0 (m, Car,ipso,xyl), 131.9–131.3 (m, Car,para,xylH), 130.3–129.6 (m, Car,ortho,xylH), 116.0 (q, 1JCF = 290.2 Hz, COOCF3), 38.5–38.0 (m, (CH2)3CCH3), 38.0–37.4 (m, (CH2)3CCH3), 33.4–32.2 (m, (CH2)3CCH3), 21.2 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1007.300 (M+−C2F3O2 calculated = 1007.304).

IR: ṽ = 2917.22, 2861.24, 1691.82, 1606.51, 1451.52, 1408.64, 1197.36, 1134.94, 1090.77, 1042.65, 847.51, 785.11, 727.48, 691.39, 561.83, 520.41, 480.59, 455.30, 423.30 cm−1.

The synthesis was performed following literature procedures.[67] The analytic data is in agreement with literature.[67]

120

7 Experimental Section

7.3.11 Synthesis of Bis(1-naphtholato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-46

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 303 mg (2.10 mmol, 20.0 eq) 1-naphthol in 10 mL toluene was stirred at 105 °C for 19 h. After cooling to room temperature, the mixture was cooled to −18 °C and dark red crystals precipitated. The crystals were separated, washed with 2x5 mL n-pentane and dried in vacuo to yield a purple powder.

Yield: 80.0 mg (61.6%)

1H-NMR (400 MHz, CD2Cl2): δ = 8.31 (d, 3JHH = 8.2 Hz, 2 H, C8ar,olatoH), 7.41 (d, 3JHH = 8.0 Hz, 2 H, C5ar,olatoH), 7.23–7.12 (m, 14 H, Car,xylH and C7ar,olatoH), 7.07 (t, 3JHH = 7.1 Hz, 2 H, C6ar,olatoH), 6.83–6.71 (m, 6 H, Car,para,xylH), 6.62 (d, 3JHH = 7.5 Hz, 2 H, C1ar,olatoH), 6.55 (t, 3JHH = 7.8 Hz, 2 H, C2ar,olatoH), 6.43 (d, 3JHH = 7.8 Hz, 2 H, C3ar,olatoH), 2.55–2.37 (m, 6 H, (CH2)3CCH3), 1.94 (s, 36 H, Car,xylCH3), 1.76–1.69 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 50.0 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 50.9 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 164.6 (s, OCar,olato), 137.7–137.1 (m, Car,meta,xyl), 137.1–136.4 (m, Car,ipso,xyl), 135.3 (s, C9ar,olato), 131.4–131.1 (m, Car,para,xylH), 130.8–130.4 (m, Car,ortho,xylH), 130.4–130.1 (m, C4ar,olato), 127.2–126.9 (m, C2ar,olatoH), 126.9–126.5 (m, C5ar,olatoH), 125.8–125.4 (m, C8ar,olatoH), 124.3 – 123.9 (m, C7ar,olatoH), 121.6–121.2 (m, C6ar,olatoH), 112.8–112.4 (m, C1ar,olatoH), 111.1–110.6 (m, C3ar,olatoH), 40.6–40.2 (m, (CH2)3CCH3), 38.1–37.5 (m, (CH2)3CCH3), 36.2–35.4 (m, (CH2)3CCH3), 21.3 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1037.363 (M+−C10H7O calculated = 1037.368).

IR: ṽ = 3029.14, 2915.68, 2857.93, 1593.59, 1560.86, 1499.52, 1445.78, 1397.03, 1370.43, 1293.93, 1237.20, 1127.64, 1087.93, 1047.47, 1015.83, 887.97, 844.34, 763.46, 726.73, 691.60, 597.11, 560.66, 475.95, 432.53 cm−1.

121

7 Experimental Section

7.3.12 Synthesis of Bis(2-naphtholato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-47

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 303 mg (2.10 mmol, 20.0 eq) 2-naphthol in 10 mL toluene was stirred at 105 °C for 19 h. After cooling to room temperature, the mixture was cooled to −18 °C and dark red crystals precipitated. The crystals were separated, washed with

2x5 mL n-pentane and 2x5 mL Et2O and dried in vacuo to yield a purple powder.

Yield: 47.7 mg (36.7%)

1H-NMR (400 MHz, THF-d8): δ = 7.25–7.17 (m, 4 H, Car,olatoH), 7.17–7.12 (m, 12 H, Car,ortho,xylH), 7.11– 7.05 (m, 4 H, Car,olatoH), 7.01–6.96 (s, 2 H, Car,olatoH), 6.93–6.84 (m, 2 H, Car,olatoH), 6.72–6.67 (m, 6 H, Car,para,xylH), 6.67–6.61 (m, 2 H Car,olatoH), 2.50–2.40 (m, 6 H, (CH2)3CCH3), 1.94 (s, 36 H, Car,xylCH3), 1.69–1.64 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 51.2 (s, 3 P) ppm.

31P{1H}-NMR (121 MHz, CD2Cl2): δ = 50.2 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, THF-d8): δ = 167.3 (s, OCar,olato), 138.3–138.2 (m, Car,olato), 138.0–136.9 (m, Car,ipso,xyl and Car,metha,xyl), 131.5–131.2 (s, Car,para,xylH), 131.2–130.7 (m, Car,ortho,xylH), 127.7–127.4 (m, Car,olatoH), 127.4–127.1 (m, Car,olatoH), 126.8–126.5 (m, Car,olato), 126.4–126.0 (m, Car,olatoH), 125.6–125.1 (m, Car,olatoH), 124.4–123.9 (m, Car,olatoH), 119.0–118.5 (m, Car,olatoH), 112.8–112.2 (m, Car,olatoH), 41.3– 41.0 (m, (CH2)3CCH3), 37.5–37.0 (m, (CH2)3CCH3), 36.2–35.5 (m, (CH2)3CCH3), 21.2 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1037.368 (M+−C10H7O calculated = 1037.368).

IR: ṽ = 2992.54, 2912.28, 2855.90, 1612.24, 1585.94, 1493.44, 1458.85, 1428.97, 1373.85, 1337.36, 1277.10, 1217.40, 1170.80, 1123.25, 1085.72, 1038.07, 948.68, 911.94 ,836.18, 735.94, 688.58, 623.34, 558.16, 508.44, 472.50, 435.26 cm−1.

The signals generated by the 2-naphtholato ligand could not be assigned due to partial overlapping of the signals in the 1H-NMR spectrum.

122

7 Experimental Section

7.3.13 Synthesis of Naphthalene-1,8-bis(olato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-48

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 16.8 mg (0.11 mmol, 1.00 eq) 1,8-dihydroxynaphthalene in 10 mL toluene was stirred at 105 °C for 16 h. After cooling to room temperature, the solvent was removed in vacuo and the yellow solid dissolved in 3 mL DCM. After removal of the solvent in vacuo, the green powder was washed with 5 mL n-pentane and dried in vacuo.

Yield: 69.3 mg (59.9%)

1H-NMR (600 MHz, THF-d8): δ = 7.31 (s, 12 H, Car,ortho,xylH), 6.91 (t, 3JHH = 7.7 Hz, 2 H, C2ar,olatoH), 6.75– 6.70 (m, 2 H, C3ar,olatoH), 6.66 (s, 6 H, Car,para,xylH), 6.43–6.63 (m, 2 H, C1ar,olatoH), 2.35–2.24 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.65–1.56 (s, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 45.3 (s, 3 P) ppm.

31P{1H}-NMR (121 MHz, CD2Cl2): δ = 44.8 (s, 3 P) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 164.2 (s, OCar), 140.5–140.3 (m, C4ar,olato), 138.1–137.6 (m, Car,ipso,xyl), 137.5–137.2 (m, Car,meta,xyl), 131.4–130.8 (s, Car,para,xylH), 130.5–130.0 (m, Car,ortho,xylH), 126.8– 126.6 (m, C2ar,olatoH), 124.1–124.0 (m, C5ar,olato), 114.2–114.0 (m, C3ar,olatoH), 110.0–109.7 (m, C1ar,olatoH), 39.9–39.5 (m, (CH2)3CCH3) , 37.8–37.2 (m, (CH2)3CCH3), 32.6–31.9 (m, (CH2)3CCH3), 21.1 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1052.354765 (M+ calculated = 1052.354186).

IR: ṽ = 3031.56, 2948.49, 2911.76, 2854.52, 1594.01, 1546.12, 1457.47, 1425.94, 1374.91, 1277.70, 1216.10, 1167.68, 1128.96, 1088.85, 1047.69, 840.62, 777.92, 747.52, 717.53, 689.97, 609.53, 561.09, 500.73, 483.72, 469.98, 455.93 cm−1.

123

7 Experimental Section

7.3.14 Synthesis of Naphthalene-2,3-bis(olato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-49

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 16.8 mg (0.11 mmol, 1.00 eq) 2,3-dihydroxynaphthalene in 10 mL toluene was stirred at 105 °C for 24 h. After cooling to room temperature, the solvent was removed in vacuo and the yellow solid dissolved in 4 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 2x4 mL n-pentane and dried in vacuo.

Yield: 83.8 mg (60.8%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.77–7.56 (m, 2 H, C2ar,olatoH), 7.46–7.27 (m, 2 H, C1ar,olatoH), 7.26–7.10 (m, 12 H, Car,ortho,xylH), 7.10–7.02 (m, 2 H, C3ar,olatoH), 6.85–6.58 (m, 6 H, Car,para,xylH), 2.58–2.24 (m, 6 H, (CH2)3CCH3), 2.05 (s, 36 H, Car,xylCH3), 1.80–1.60 (s, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 48.1 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 48.7 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 166.4 (s, OCar,olato), 138.0–137.0 (m, Car,meta,xyl and Car,ipso,xyl), 131.5–130.6 (m, Car,para,xylH), 130.0–129.5 (m, Car,ortho,xylH), 129.5–129.3 (m, Car,olato) 125.5–125.1 (m, C2ar,olatoH), 120.4–120.0 (m, C3ar,olatoH), 108.9–108.3 (s, C1ar,olatoH), 39.9–39.5 (m, (CH2)3CCH3), 38.2– 37.6 (m, (CH2)3CCH3), 32.8–32.0 (m, (CH2)3CCH3), 21.4 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1052.355 (M+ calculated = 1052.354).

IR: ṽ = 3033.03, 2912.39, 2858.27, 1972.10, 1591.76, 1450.97, 1416.60, 1373.92, 1278.42, 1212.70, 1162.41, 1129.19, 1084.66, 1037.07, 934.49, 839.45, 776.44, 738.68, 690.49, 646.29, 622.38, 559.28, 541.00, 481.37, 451.40 cm−1.

124

7 Experimental Section

7.3.15 Synthesis of Bis(2-ortho-phenylolphenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-50

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 39.2 mg (0.21 mmol, 2.00 eq) 2,2’-biphenol in 10 mL toluene was stirred at 105 °C for 24 h. After cooling to room temperature, the mixture was cooled to −18 °C and orange crystals precipitated. The crystals were separated, washed with 2x5 mL n-pentane and dried in vacuo to yield an orange powder.

Yield: 95.9 mg (68.9%)

1H-NMR (400 MHz, THF-d8): δ = 8.67 (brs, 2 H, OH), 7.32–7.03 (m, 12 H, Car,ortho,xylH and 6 H, Car,olatoH), 6.98–6.75 (m, 6 H, Car,olatoH), 6.73–6.66 (s, 6 H, Car,para,xylH), 6.66–6.55 (m, 4 H, Car,olatoH), 2.37–2.23 (m, 6 H, (CH2)3CCH3), 1.97 (s, 36 H, Car,xylCH3), 1.60–1.51 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 44.0 (s, 3 P) ppm.

31P{1H}-NMR (121 MHz, CD2Cl2): δ = 44.6 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, THF-d8): δ = 165.2 (s, OCar), 155.5–155.0 (s, HOCar), 138.4–137.8 (m, Car,ipso,xyl), 137.5–137.1 (m, Car,meta,xyl), 135.0–134.8 (m, Car,olato), 132.2–132.0 (m, Car,olatoH), 131.7– 131.5 (m, Car,olatoH), 131.2 (s, Car,para,xylH), 131.1–130.8 (m, Car,ortho,xylH), 129.0–128.8 (m, Car,olatoH), 127.5–127.3 (m, Car,olato), 127.3–127.1 (m, Car,olatoH), 124.1–123.9 (m, Car,olatoH), 120.5–120.1 (m, Car,olatoH), 117.5–117.1 (m, Car,olatoH), 116.1–116.0 (m, Car,olatoH), 39.6–39.5 (m, (CH2)3CCH3), 37.5–37.0 (m, (CH2)3CCH3), 34.7–34.2 (m, (CH2)3CCH3), 21.3 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1079.37832 (M+−C12H9O2 calculated = 1079.379260).

IR: ṽ = 3028.90, 2914.19, 2858.81, 1591.22, 1507.05, 1474.26, 1434.97, 1380.70, 1306.73, 1281.33, 1226.77, 1127.32, 1087.12, 1039.63, 1000.42, 939.53, 843.72, 817.10, 754.39, 693.28, 609.11, 560.83, 520.21, 475.49, 452.87 cm−1.

Due to partial overlapping of the signals in the 1H-NMR spectrum, not all signals generated by the 2-ortho-phenylolphenolato ligand could be assigned.

125

7 Experimental Section

7.3.16 Synthesis of Bis(2-fluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-51

A solution of 100 mg (0.11 mmol, 1.00 eq trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 118 mg (10.5 mmol, 10.0 eq) 2-fluorophenol in 10 mL toluene was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 4 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 3x4 mL n-pentane, 2 mL diethyl ether and 4 mL n-pentane and dried in vacuo.

Yield: 61.4 mg (50.0%)

1H-NMR (600 MHz, THF-d8): δ = 7.25–7.15 (m, 12 H, Car,ortho,xylH), 6.78–6.73 (m, 2 H, Car,olato), 6.69– 6.63 (m, 6 H, Car,para,xylH), 6.49–6.43 (m, 2 H, Car,olatoH), 6.42–6.37 (m, 2 H , Car,olatoH), 5.95–5.89 (m, 2 H, Car,olatoH), 2.50–2.40 (m, 6 H, (CH2)3CCH3), 1.96 (s, 36 H, Car,xylCH3), 1.70–1.65 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 50.2 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 49.0 (s, 3 P) ppm.

19F{1H}-NMR (565 MHz, THF-d8): δ = −136.63 (s, 2 F) ppm.

13 1 1 2 C{ H}-NMR (151 MHz, THF-d8): δ = 156.4 (d, JCF = 236.9 Hz, Car,olatoF), 156.4 (d, JCF = 11.6 Hz, OCar,olato), 137.6–136.9 (m, Car,ipso,xyl and Car,meta,xyl), 131.3–131.0 (m, Car,para,xylH), 131.0–130.8 (m, Car,ortho,xylH), 123.6–123.3 (m, Car,olatoH), 122.9–122.6 (m, Car,olatoH), 113.9 (d, 2JCF = 19.3 Hz, Car,olatoH), 112.0–111.7 (m, Car,olatoH), 40.8–40.6 (m, (CH2)3CCH3), 37.5–37.0 (m, (CH2)3CCH3), 35.6–35.0 (m, (CH2)3CCH3), 21.1 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1006.346 (M+H+−C6H4FO calculated = 1006.351).

IR: ṽ = 2913.36, 2856.99, 1594.07, 1487.94, 1447.66, 1415.94, 1302.47, 1238.43, 1180.68, 1127.44, 1091.74, 1030.14, 840.53, 735.81, 688.28, 560.06, 517.16, 475.58 cm−1. Not all NMR-spectroscopic signals generated by the 2-fluorophenolato ligand could be assigned to the respective specific nuclei.

126

7 Experimental Section

7.3.17 Synthesis of Bis(2,5-difluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-52

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 55.0 mg (0.42 mmol, 4.00 eq) 2,5-difluorophenol in 10 mL toluene was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 4 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 2x4 mL n-pentane and dried in vacuo.

Yield: 92.6 mg (73.1%)

1H-NMR (600 MHz, THF-d8): δ = 7.31–7.03 (m, 12 H, Car,ortho,xylH), 6.74–6.61 (m, 6 H, Car,para,xylH), 6.49–6.38 (m, 4 H, Car,olatoH), 5.69–5.62 (m, 2 H, Car,olatoH), 2.51–2.42 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.72–1.68 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 51.1 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 49.8 (s, 3 P) ppm.

19F{1H}-NMR (565 MHz, THF-d8): δ = −121.76–−122.86 (m, 2 F), −142.24–−142.84 (m, 2 F) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 160.0 (d, 1JCF = 236.4 Hz, Car,olatoF), 157.7–157.2 (m, OCar,olato), 153.1 (d, 1JCF = 230.6 Hz, Car,olatoF), 137.5–137.2 (m, Car,meta,xyl), 137.1–136.6 (m, Car,ipso,xyl), 131.5–131.1 (m, Car,para,xylH), 131.0–130.6 (m, Car,ortho,xylH), 113.4–112.8 (m, Car,olatoH), 108.8–108.4 (m, Car,olatoH), 97.5–97.0 (m, Car,olatoH), 40.9–40.7 (m, (CH2)3CCH3), 37.3–37.0 (m, (CH2)3CCH3), 35.4–34.9 (m, (CH2)3CCH3), 21.0 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1023.331 (M+−C6H3F2O calculated = 1023.334).

IR: ṽ = 2917.56, 2856.21, 1613.70, 1565.55, 1492.19, 1424.55, 1375.42, 1325.36, 1283.24, 1221.47, 1178.42, 1134.82, 1090.99, 1039.60, 970.89, 945.42, 836.60, 822.42, 772.14, 741.61, 689.98, 616.03, 562.88, 520.98, 478.27, 454.80, 445.08, 416.41 cm−1.

Not all NMR-spectroscopic signals generated by the 2,5-difluorophenolato ligand could be assigned to the respective specific nuclei.

127

7 Experimental Section

7.3.18 Synthesis of Bis(2,6-difluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-53

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 55.0 mg (0.42 mmol, 4.00 eq) 2,6-difluorophenol in 10 mL toluene was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the remaining red solid dissolved in 4 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 2x4 mL n-pentane and dried in vacuo.

Yield: 88.5 mg (69.8%)

1H-NMR (600 MHz, THF-d8): δ = 7.27–7.15 (m, 12 H, Car,ortho,xylH), 6.69–6.61 (m, 6 H, Car,para,xylH), 6.33–6.19 (m, 4 H, C1ar,meta,olatoH), 5.83–5.74 (m, 2 H, C2ar,olatoH), 2.45–2.36 (m, 6 H, (CH2)3CCH3), 1.97 (s, 36 H, Car,xylCH3), 1.69–1.63 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 49.6 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 48.2 (s, 3 P) ppm.

19F{1H}-NMR (565 MHz, THF-d8): δ = −132.66 (s, 4 F) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 157.0 (dd, 1JCF = 238.2 Hz, 3JCF = 9.4 Hz, Car,olatoF), 145.9 (t, 2JCF = 15.1 Hz, OCar,olato), 137.5–136.6 (m, Car,ipso,xyl and Car,meta,xyl), 131.4–131.1 (m, Car,ortho,xylH), 131.1–130.9 (m, Car,para,xylH), 110.6–110.1 (m, C1ar,olatoH), 109.47–109.4 (m, C2ar,olatoH), 40.3–40.0 (m, (CH2)3CCH3), 37.6–37.2 (m, (CH2)3CCH3), 35.5–34.9 (m, (CH2)3CCH3), 21.2 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1023.328 (M+−C6H3 F2O calculated = 1023.334).

IR: ṽ = 2916.79, 2859.69, 1599.55, 1495.64, 1475.42, 1376.32, 1307.41, 1216.33, 1132.22, 1086.86, 1057.26, 983.49, 848.50, 745.84, 695.86, 560.36, 528.87, 474.13 cm−1.

128

7 Experimental Section

7.3.19 Synthesis of Bis(3,5-difluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-54

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 27.3 mg (0.21 mmol, 2.00 eq) 3,5-difluorophenol in 10 mL toluene was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 2 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 2 mL n-pentane and dried in vacuo.

Yield: 98.2 mg (77.5%)

1H-NMR (600 MHz, THF-d8): δ = 7.08–6.94 (m, 12 H, Car,ortho,xylH), 6.78–6.65 (m, 6 H, Car,para,xylH), 6.17–6.04 (m, 4 H, C1ar,olatoH), 5.67–5.57 (m, 2 H, C2ar,olatoH), 2.49–2.42 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.71–1.66 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 52.4 (s, 3 P) ppm.

31P{1H}-NMR (243 MHz, CD2Cl2): δ = 51.3 (s, 3 P) ppm.

19F{1H}-NMR (565 MHz, THF-d8): δ = −116.55 (s, 4 F) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 171.0 (t, 3JCF = 14.6 Hz, OCar,olato), 164.2 (dd, 1JCF = 239.8 Hz, 3JCF = 18.7 Hz, Car,olatoF), 137.4–137.1 (m, Car,meta,xyl), 136.9–136.4 (m, Car,ipso,xyl), 131.4–131.1 (m, Car,para,xylH), 130.7–130.4 (m, Car,ortho,xylH), 102.7 (dd, 2JCF = 18.8 Hz, 4JCF = 4.1 Hz, C1ar,olatoH), 88.0 (t, 2JCF = 27.0 Hz, C2ar,olatoH), 41.1–41.0 (m, (CH2)3CCH3), 37.1–36.7 (m, (CH2)3CCH3), 35.5–35.0 (m, (CH2)3CCH3), 20.9 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1023.333 (M+−C6H3F2O calculated = 1023.334).

IR: ṽ = 2915.88, 2859.60, 1612.45, 1564.17, 1461.23, 1354.86, 1279.27, 1145.71, 1089.46, 1042.02, 1011.66, 976.60, 821.84, 766.94, 740.28, 686.79, 636.28, 558.87, 511.80, 475.09 cm−1.

129

7 Experimental Section

7.3.20 Synthesis of Bis(2,4,6-trifluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-55

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 31.1 mg (0.21 mmol, 2.00 eq) 2,4,6-trifluorophenol in 10 mL toluene was stirred at 105 °C for 24 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 2 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 3x5 mL n-pentane and dried in vacuo.

Yield: 88.0 mg (67.3%)

1H-NMR (400 MHz, THF-d8): δ = 7.26–7.11 (m, 12 H, Car,ortho,xylH), 6.71–6.63 (m, 6 H, Car,para,xylH), 6.27–6.10 (m, 4 H, C2ar,olatoH), 2.53–2.37 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.70–1.65 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 50.8 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 49.3 (s, 3 P) ppm.

19F{1H}-NMR (377 MHz, THF-d8): δ = −129.39–−130.03 (m, 4 F), −134.19–−134.73 (m, 2 F) ppm.

13C{1H}-NMR (101 MHz, THF-d8): δ = 156.2–153.8 (m, C1ar,olatoF), 151.3–148.4 (m, C3ar,olatoF), 142.5– 142.0 (m, OCar,olato), 137.4–136.4 (m, Car,meta,xyl and Car,ipso,xyl), 131.3–130.7 (m, Car,para,xylH and Car,ortho,xylH), 99.1–98.0 (m, C2ar,olatoH), 40.4–40.1 (m, (CH2)3CCH3), 37.4–36.7 (m, (CH2)3CCH3), 35.6– 34.9 (m, (CH2)3CCH3), 21.0 (s, Car,xylCH3) ppm.

IR: ṽ = 2914.87, 2858.63, 1592.23, 1496.79, 1443.73, 1414.68, 1360.04, 1266.04, 1167.18, 1129.71, 1098.36, 1019.43, 989.32, 827.80, 770.55, 741.29, 689.96, 560.50, 530.79, 473.20 cm−1.

MS analysis revealed no characteristic fractionation.

The CF-coupling of the Car,olatoF signals could not be fully resolved. The following coupling constants could be determined: C1ar,olatoF: 1JCF = 239.7 Hz, 3JCF = 13.3 Hz; C2ar,olatoF: 1JCF = 230.2 Hz, 3JCF = 15.0 Hz.

130

7 Experimental Section

7.3.21 Synthesis of Bis(3,4,5-trifluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-56

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 31.1 mg (0.21 mmol, 2.00 eq) 3,4,5-trifluorophenol in 10 mL toluene was stirred at 105 °C for 21 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 4 mL DCM. After removal of the solvent in vacuo, the red powder was washed with 2x4 mL n-pentane and dried in vacuo.

Yield: 91.4 mg (69.9%)

1H-NMR (400 MHz, THF-d8): δ = 7.04–6.94 (m, 12 H, Car,ortho,xylH), 6.76–6.70 (m, 6 H, Car,para,xylH), 6.22–6.11 (m, 4 H, Car,olatoH), 2.50–2.42 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.71–1.66 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 52.8 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 51.5 (s, 3 P) ppm.

19F{1H}-NMR (377 MHz, THF-d8): δ = −141.05–−142.11 (m, 4 F), −184.11–−184.33 (m, 2 F) ppm.

13C{1H}-NMR (101 MHz, THF-d8): δ = 164.7–164.2 (m, OCar,olato), 153.2–150.3 (m, Car,olatoF), 137.6– 137.2 (m, Car,meta,xyl), 137.1–136.3 (m, Car,ipso,xyl), 131.6–131.3 (m, Car,para,xylH), 130.9–130.4 (m, Car,ortho,xylH), 129.6–128.7 (m, Car,olatoF), 103.1–102.6 (m, Car,olatoH), 41.4–41.1 (m, (CH2)3CCH3), 37.1– 36.5 (m, (CH2)3CCH3), 35.8–35.0 (m, (CH2)3CCH3), 21.0 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1041.326 (M+−C6H2F3O calculated = 1041.324).

IR: ṽ = 2915.82, 2857.02, 1628.05, 1590.80, 1498.68, 1394.50, 1263.37, 1213.36, 1151.73, 1125.25, 1085.46, 1023.09, 987.23, 840.17, 820.47, 778.75, 741.57, 687.59, 639.25, 558.74, 511.61, 473.46, 437.43 cm−1.

The CF-coupling of the Car,olatoF signals could not be fully resolved and not all signals of the 3,4,5-trifluorophenolato ligand could be assigned. The following coupling constants could be determined for the multiplet at 153.2–150.3 ppm: 1JCF = 240.9 Hz, JCF = 9.0 Hz.

131

7 Experimental Section

7.3.22 Synthesis of Bis(2,3,5,6-tetrafluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-57

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 34.9 mg (0.21 mmol, 2.00 eq) 2,3,5,6-tetrafluorophenol in 10 mL toluene was stirred at 105 °C for 18 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 4 mL DCM. After removal of the solvent in vacuo, the orange powder was washed with 3x4 mL n-pentane and dried in vacuo.

Yield: 106 mg (78.7%)

1H-NMR (600 MHz, THF-d8): δ = 7.17–7.07 (m, 12 H, Car,ortho,xylH), 6.76–6.67 (m, 6 H, Car,para,xylH), 5.87–5.75 (m, 2 H, Car,para,olatoH), 2.53–2.42 (m, 6 H, (CH2)3CCH3), 1.99 (s, 36 H, Car,xylCH3), 1.72–1.66 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 51.7 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 50.2 (s, 3 P) ppm.

19F{1H}-NMR (565 MHz, THF-d8): δ = −147.77–−148.00 (m, 4 F), −162.84–−163.04 (m, 4 F) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 149.1–148.6 (m, OCar,olato), 148.5–146.3 (m, Car,olatoF), 142.5– 140.5 (m, Car,olatoF), 137.5–137.2 (m, Car,meta,xyl), 136.7–136.2 (m, Car,ipso,xyl), 131.7–131.3 (m, Car,para,xylH), 131.2–130.8 (m, Car,ortho,xylH), 87.6–87.0 (m, Car,olatoH), 40.5–40.4 (m, (CH2)3CCH3), 37.2– 36.8 (m, (CH2)3CCH3), 35.3–34.9 (m, (CH2)3CCH3), 21.1 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1059.315 (M+−C6HF4O calculated = 1059.315).

IR: ṽ = 2916.15, 2860.16, 1635.46, 1583.48, 1496.31, 1418.69, 1376.71, 1274.12, 1163.59, 1132.09, 1101.95, 1041.07, 932.70, 844.38, 768.62, 741.60, 719.22, 690.44, 624.13, 561.02, 529.31, 474.96, 439.52 cm−1.

The CF-coupling of the Car,olatoF signals could not be fully resolved and not all signals of the 2,3,5,6 tetrafluorophenolato ligand could be assigned. The following coupling constants could be determined for the multiplet at 148.5–146.3 ppm: 1JCF = 239.4 Hz; for the multiplet at 142.5– 140.5 ppm: 1JCF = 243.0 Hz; for the multiplet at 87.6–87.0 ppm: 2JCF = 24.2 Hz.

132

7 Experimental Section

7.3.23 Synthesis of Bis(2,3,4,5,6-pentafluorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-58

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 38.7 mg (0.21 mmol, 2.00 eq) pentafluorophenol in 10 mL toluene was stirred at 105 °C for 20 h. After cooling to room temperature, the solvent was removed in vacuo and the brown solid dissolved in 2 mL DCM. After removal of the solvent in vacuo, the orange powder was washed with 2 mL n-pentane and dried in vacuo.

Yield: 118 mg (85.1%)

1H-NMR (400 MHz, THF-d8): δ = 7.17–7.02 (m, 12 H, Car,ortho,xylH), 6.79–6.68 (m, 6 H, Car,para,xylH), 2.53–2.44 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.80–1.60 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 52.7 (s, 3 P) ppm.

31P{1H}-NMR (243 MHz, CD2Cl2): δ = 51.1 (s, 3 P) ppm.

19F{1H}-NMR (377 MHz, THF-d8): δ = −163.7–−164.0 (m, 4 F), −171.9–−172.3 (m, 4 F), −184.9– −185.2 (m, 2 F) ppm.

13C{1H}-NMR (101 MHz, THF-d8): δ = 144.4–143.8 (m, Car,olato), 142.9–140.2 (m, Car,olatoF), 140.2– 139.6 (m, Car,olato), 137.9–137.1 (m, Car,meta,xyl), 136.8–135.9 (m, Car,ipso,xyl), 131.9–131.3 (m, Car,para,xylH), 131.2–130.7 (m, Car,ortho,xylH), 129.2–128.6 (m, Car,olato), 40.9–40.5 (m, (CH2)3CCH3), 37.3–36.6 (m, (CH2)3CCH3), 35.7–34.8 (m, (CH2)3CCH3), 21.1 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1077.30499 (M+−C6F5O calculated = 1077.306850).

IR: ṽ = 2915.78, 2859.48, 1643.79, 1596.10, 1497.03, 1419.46, 1376.69, 1305.82, 1266.99, 1169.24, 1128.75, 1089.29, 1006.15, 981.24, 842.94, 768.70, 739.77, 689.02, 627.55, 559.37, 471.49 cm−1.

The CF-coupling of the Car,olatoF signals could not be fully resolved and not all signals of the pentafluorophenolato ligand could be assigned. The following coupling constant could be determined for the multiplet at 142.9–140.2: 1JCF = 236.2 Hz.

133

7 Experimental Section

7.3.24 Synthesis of Bis(1,3,4,5,6,7,8-heptafluoronaphtholato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-59

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 56.8 mg (0.21 mmol, 2.00 eq) heptafluoro-2-naphthol in 10 mL toluene was stirred at 105 °C for 22 h. After cooling to room temperature, the solvent was removed in vacuo and the brown solid dissolved in 5 mL DCM. After removal of the solvent in vacuo, the yellow powder was washed with 5x5 mL n-pentane and dried in vacuo.

Yield: 118 mg (74.9%)

1H-NMR (600 MHz, THF-d8): δ = 7.24–7.06 (m, 12 H, Car,ortho,xylH), 6.79–6.69 (m, 6 H, Car,para,xylH), 2.70–2.46 (m, 6 H, (CH2)3CCH3), 2.00 (s, 36 H, Car,xylCH3), 1.78–1.74 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 53.4 (s, 3 P) ppm.

19F{1H}-NMR (565 MHz, THF-d8): δ = −147.17–−147.82 (m, 2 F), −148.19–−148.98 (m, 2 F), −150.77–−151.44 (m, 2 F), −151.44–−152.53 (m, 2 F), −153.75–−154.23 (m, 2 F), −162.46–−163.61 (m, 2 F), −168.52–−169.26 (m, 2 F) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 148.2–147.7 (m, Car,olato), 147.5–145.5 (m, Car,olatoF), 144.6– 142.6 (m, Car,olatoF), 142.7–140.4 (m, Car,olatoF), 140.1–138.1 (m, Car,olatoF), 139.1–137.0 (m, Car,olatoF), 137.7–137.4 (m, Car,meta,xyl), 136.5–135.9 (m, Car,ipso,xyl), 135.9–134.0 (m, Car,olatoF), 131.9–131.5 (m, Car,para,xylH), 131.4–130.4 (m, Car,ortho,xylH), 108.4–107.9 (m, Car,olato), 99.5–99.0 (m, Car,olato), 41.0–40.7 (m, (CH2)3CCH3), 37.1–36.6 (m, (CH2)3CCH3), 35.6–35.0 (m, (CH2)3CCH3), 21.1 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1163.30180 (M+−C10F7O calculated = 1163.404000).

IR: ṽ = 2916.88, 2861.20, 1662.62, 1634.61, 1594.54, 1505.57, 1465.58, 1397.33, 1353.45, 1272.43, 1207.00, 1103.39, 1037.50, 993.93, 940.58, 842.92, 796.66, 740.27, 688.43, 557.65, 474.34, 438.45 cm−1.

The CF-coupling of the Car,olatoF signals could not be fully resolved and not all signals of the heptafluoro-2-naphtolato ligand could be assigned. The following coupling constants could be determined for the multiplet at 148.2–147.7 ppm: JCF = 15.1 Hz; for the multiplet at 147.5– 134

7 Experimental Section

145.5 ppm: 1JCF = 247.5 Hz; for the multiplet at 144.6–142.6 ppm: 1JCF = 242.5 Hz; for the multiplet at 142.7–140.4 ppm: 1JCF = 246.6 Hz; for the multiplet at 140.1–138.1 ppm: 1JCF = 260.2 Hz; for the multiplet at 139.1–137.0 ppm: 1JCF = 248.0 Hz, JCF = 14.1 Hz; for the multiplet at 135.9–134.0 ppm: 1JCF = 245.4 Hz, JCF = 15.9 Hz.

135

7 Experimental Section

7.3.25 Synthesis of Bis(3,5-dichlorophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-60

A solution of 200 mg (0.22 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 137 mg (0.84 mmol, 4.00 eq) 3,5-dichlorophenol in 10 mL toluene was stirred at 105 °C for 21 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 3 mL DCM. After removal of the solvent in vacuo, the orange powder was washed with 5 mL n-pentane, 5 mL diethyl ether and 5 mL n-pentane and dried in vacuo.

Yield: 217.1 mg (81.0%)

1H-NMR (600 MHz, THF-d8): δ = 7.04–6.95 (m, 12 H, Car,ortho,xylH), 6.76–6.67 (m, 6 H, Car,para,xylH), 6.51–6.44 (m, 4 H, C1ar,olatoH), 6.17–6.11 (m, 2 H, C2ar,olatoH), 2.52–2.41 (m, 6 H, (CH2)3CCH3), 1.99 (s, 36 H, Car,xylCH3), 1.71–1.67 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 52.9 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, tol-d8): δ = 52.6 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 51.8 (s, 3 P) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 170.1 (s, OCar,olato), 137.7–137.4 (m, Car,meta,xyl), 137.2–136.5 (m, Car,ipso,xyl), 134.0 (s, Car,olatoCl), 131.7–131.5 (m, Car,para,xylH), 131.0–130.6 (m, Car,ortho,xylH), 119.5–119.1 (m, C1ar,xylH), 113.0 (s, C2ar, olatoH), 41.5–41.2 (m, (CH2)3CCH3)), 37.4–36.7 (m, (CH2)3CCH3), 35.8–35.3 (m, (CH2)3CCH3), 21.2 (s, Car,xylCH3) ppm.

IR: ṽ = 2914.27, 2857.69, 1556.53, 1443.24, 1375.90, 1295.78, 1213.17, 1126.47, 1087.52, 1039.42, 978.28, 941.67, 838.38, 785.25, 740.55, 686.17, 598.51, 557.62, 473.14, 437.20 cm−1.

MS analysis revealed no characteristic fractionation.

136

7 Experimental Section

7.3.26 Synthesis of Bis(3,5-dibromophenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-61

A solution of 200 mg (0.22 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 212 mg (0.84 mmol, 4.00 eq) 3,5-dibromophenol in 10 mL toluene was stirred at 105 °C for 21 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 3 mL DCM. After removal of the solvent in vacuo, the orange powder was washed with 5 mL n-pentane, 5 mL diethyl ether and 5 mL n-pentane and dried in vacuo.

Yield: 247.8 mg (80.7%)

1H-NMR (600 MHz, THF-d8): δ = 7.05–6.94 (m, 12 H, Car,ortho,xylH), 6.76–6.70 (m, 6 H, Car,para,xylH), 6.69–6.63 (m, 4 H, C1ar,olatoH), 6.46–6.39 (m, 2 H, C2ar,olatoH), 2.51–2.42 (m, 6 H, (CH2)3CCH3), 2.00 (s, 36 H, Car,xylCH3), 1.71–1.67 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (243 MHz, THF-d8): δ = 53.1 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 52.0 (s, 3 P) ppm.

13C{1H}-NMR (151 MHz, THF-d8): δ = 170.3 (s, OCar,olato), 137.8–137.3 (m, Car,meta,xyl), 137.2–136.5 (m, Car,ipso,xyl), 131.8–131.3 (m, Car,para,xylH), 131.0–130.5 (m, Car,ortho,xylH), 122.7 (s, C1ar,olatoH), 122.3 (s, Car,olatoBr), 118.4 (s, C2ar,olatoH), 41.6–41.3 (m, (CH2)3CCH3), 37.3–36.8 (m, (CH2)3CCH3), 35.8–35.3 (m, (CH2)3CCH3), 21.3 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1145.172 (M+–C6H3Br2O calculated = 1145.172).

IR: ṽ = 2914.21, 2856.82, 1550.75, 1513.43, 1442.28, 1287.86, 1216.02, 1168.78, 1126.12, 1086.12, 1035.21, 969.64, 915.50, 838.67, 785.11, 734.69, 686.25, 589.94, 556.80, 472.31, 436.38 cm−1.

137

7 Experimental Section

7.3.27 Synthesis of Bis(3,5-bis(trifluoromethyl)phenolato)-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-62

A solution of 100 mg (0.11 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 48.4 mg (0.21 mmol, 2.00 eq) 3,5-bis(trifluoromethyl)phenol in 10 mL toluene was stirred at 105 °C for 16 h. After cooling to room temperature, the solvent was removed in vacuo and the red solid dissolved in 3 mL DCM. After removal of the solvent in vacuo, the orange powder was washed with 3x10 mL n-pentane and dried in vacuo.

Yield: 116.3 mg (78.2%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.00–6.89 (m, 12 H, Car,ortho,xylH), 6.89–6.83 (m, 4 H, C1ar,olatoH), 6.79– 6.72 (m, 6 H, Car,para,xylH), 6.63–6.57 (m, 2 H, C2ar,olatoH), 2.47–2.36 (m, 6 H, (CH2)3CCH3), 1.98 (s, 36 H, Car,xylCH3), 1.77–1.67 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 52.9 (s, 3 P) ppm.

31P{1H}-NMR (162 MHz, THF-d8): δ = 53.9 (s, 3 P) ppm.

19F{1H}-NMR (377 MHz, CD2Cl2): δ = −63.17 (s, 12 F) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 168.3 (s, OCar,olato), 137.8–137.4 (m, Car,meta,xyl), 136.4–135.7 (m, Car,ipso,xyl), 131.8–131.4 (m, Car,para,xylH), 131.3–130.5 (m, Car,olatoCF3), 130.4–129.9 (m, Car,ortho,xylH), 124.7 (q, 1JCF = 272.4 Hz, CarCF3), 119.9 (s, C1ar,olatoH), 105.7 (s, C2ar,olatoH), 41.2–41.0 (m, (CH2)3CCH3), 38.1–37.5 (m, (CH2)3CCH3), 35.8–35.2 (m, (CH2)3CCH3), 21.2 (s, Car,xylCH3) ppm.

HR-MS: m/z = 1352.339 (M+ calculated = 1352.336).

IR: ṽ = 2918.35, 1595.85, 1462.65, 1384.77, 1269.73, 1160.02, 1116.12, 989.17, 951.18, 839.62, 730.98, 685.97, 593.81, 558.03, 474.22 cm−1.

138

7 Experimental Section

7.3.28 Synthesis of Tri-µ-fluorido-bis(1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane)diruthenium(II)

Abbreviation: cat-II-63a

A solution of 200 mg (0.21 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 78.2 mg (0.42 mmol, 2.00 eq) hexafluorobenzene in 10 mL THF was divided into two 20 mL stainless steel finger autoclaves equipped with a glass inlet and a magnetic stir bar. The autoclaves were pressurized with hydrogen to 30 bar and heated at 140 °C in an aluminum cone fitted on a magnetic stir plate for 0.5 h. The autoclaves were cooled down in an ice bath and carefully vented. The two product mixtures were combined. After settling of the yellow solid, the brown supernatant was removed. The solid was washed with 2x5 mL THF and 3x5 mL n-pentane and dried in vacuo.

Yield: 101.2 mg (52%, calculated for a molecular weight of 1864.15 g mol-1)

1H-NMR (400 MHz, CD2Cl2): δ = 7.81–7.18 (m, 24 H, Car,ortho,xylH), 6.74–6.66 (m, 12 H, Car,para,xylH), 2.37–2.19 (m, 6 H, (CH2)3CCH3), 1.88–1.80 (m, 6 H, (CH2)3CCH3), 1.75 (s, 38 H, Car,xyl-CH3), 1.54 (br s, 34 H, Car,xylCH3), 1.36–1.29 (m, 6 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 37.4–35.5 (m, 6 P) ppm.

19F{1H}-NMR (377 MHz, CD2Cl2): δ = −139.54 (br s), −153.47 (s, 10BF4−), −153.52 (s, 11BF4−), −337.47–−339.52 (m, µ-F) ppm.

11B-NMR (96 MHz, CD2Cl2): δ = −1.1 (br s, 1 B) ppm

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 139.2–138.5 (m, Car,ipso,xyl), 138.5–138.1 (m, Car,meta,xyl), 138.1– 137.8 (m, Car,meta,xyl), 137.4–136.7 (m, Car,ipso,xyl), 131.7–131.5 (m, Car,para,xylH), 131.5–131.4 (m, Car,para,xylH), 131.4–131.0 (m, Car,ortho,xylH), 38.4–37.8 (m, (CH2)3CCH3), 36.4–36.1 (m, (CH2)3CCH3), 35.5–34.7(m, (CH2)3CCH3), 21.6 (s, Car,xylCH3). 20.2 (br s, Car,xylCH3) ppm.

HR-MS: m/z = 1848.626 (M+ calculated = 1848.637).

IR: ṽ = 2990.98, 2945.15, 2916.36, 2860.02, 2181.36, 2048.14, 2003.55, 1990.68, 1784.47, 1598.64, 1581.65, 1448.56, 1414.21, 1378.99, 1316.53, 1272.74, 1128.44, 1097.59, 1037.17, 994.25, 860.94, 835.67, 787.64, 748.04, 698.20, 685.18, 564.45, 519.45, 487.14, 448.85, 432.86, 422.04 cm−1.

139

7 Experimental Section

Elemental analysis (%): C 65.92, H 6.61, P 9.59, Ru 10.41, F 6.84, B 0.57. (calculated value with BF4− as counter anion: C 65.90, H 6.57, P 9.63, Ru 10.46, F 6.88, B 0.56; calculated value with PO2F2− as counter anion: C 65.42, H 6.53, P 11.14, Ru 10.39, F 4.88, O 1.64; calculated value with F− as counter anion: C 68.30, H 6.81, P 9.97, Ru 10.84, F 4.08).

140

7 Experimental Section

7.3.29 Synthesis of η3-hexafluorophosphosphato-(1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ruthenium

Abbreviation: cat-II-63b

A solution of 400 mg (0.42 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 156 mg (0.84 mmol, 2.00 eq) hexafluorobenzene in 10 mL THF was divided into two 20 mL stainless steel finger autoclaves equipped with a glass inlet and a magnetic stir bar. The autoclaves were pressurized with hydrogen to 50 bar and heated at 160 °C in an aluminum cone fitted on a magnetic stir plate for 24 h. The autoclaves were cooled down in an ice bath and carefully vented. After settling of the yellow solids, the brown supernatants were removed. As both products showed the same signals in the 31P{1H}-NMR spectra, the dispersions were combined. After washing with 2x3 mL THF and 2x3 mL n-pentane, the yellow solid was dried in vacuo.

Yield: 239.6 mg (55%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.23–7.05 (m, 12 H, Car,ortho,xylH), 6.88–6.79 (m, 6 H, Car,para,xylH), 2.34– 2.21 (m, 6 H, (CH2)3CCH3), 2.10 (s, 36 H, Car,xylCH3), 1.75–1.65 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 49.8 (sept, JFP = 20.1 Hz, 3 P) ppm.

19F{1H}-NMR (377 MHz, CD2Cl2): δ = −133.56 (s), −155.10–−165.00 (m) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 138.8–138.4 (m, Car,meta,xyl), 134.0–133.1 (m, Car,ipso,xyl), 132.5– 131.9 (m, Car,para,xylH), 129.2–128.5 (m, Car,ortho,xylH), 39.0–38.7 (m, (CH2)3CCH3), 37.6–37.0 (m, (CH2)3CCH3), 29.8–29.0 (m, (CH2)3CCH3), 21.3 (s, Car,xylCH3) ppm.

IR: ṽ = 2913.95, 2858.13, 1598.78, 1415.98, 1374.02, 1323.73, 1272.05, 1135.26, 1089.97, 1057.89, 993.24, 886.99, 842.90, 827.22, 785.00, 764.63, 695.03, 685.14, 609.21, 579.44, 563.80, 520.71, 497.34, 485.67, 462.20, 438.22, 420.12 cm−1.

Elemental analysis (%): C 61.09, H 6.06, P 11.86, Ru 9.67, F 10.91. (calculated: C 61.27, H 6.11, P 11.92, Ru 9.73, F 10.97).

MS analysis revealed no characteristic fractionation.

141

7 Experimental Section

7.3.30 Synthesis of Dichlorido-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) Route 1:

Abbreviation: cat-II-66a

200 mg (0.21 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 119.6 mg (0.42 mmol, 2.00 eq) hexachlorobenzene were dispersed in 5 mL THF and transferred into a 20 mL stainless steel finger autoclave equipped with a glass inlet and a magnetic stir bar. The autoclave was pressurized with hydrogen to 50 bar and heated at 160 °C in an aluminum cone fitted on a magnetic stir plate for 2 h. The autoclave was cooled down in an ice bath and carefully vented. After settling of the brown solid, the brown supernatant was removed. The solid was washed with 10 mL THF and 2x10 mL n-pentane, then dried in vacuo.

Yield: 106.7 mg (53%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.10–7.00 (m, 12 H, Car,ortho,xylH), 6.81–6.72 (m, 6 H, Car,para,xylH), 2.38– 2.24 (m, 6 H, (CH2)3CCH3), 2.05 (s, 36 H, Car,xylCH3), 1.66–1.58 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 50.2 (s, 3 P) ppm.

FAB-MS: m/z = 929.3 (M+−Cl calculated = 929.3).

IR: ṽ = 2991.42, 2954.20, 2912.56, 2857.45, 1598.70, 1584.55, 1453.53, 1412.84, 1374.77, 1320.38, 1273.90, 1230.76, 1169.89, 1127.91, 1107.61, 1095.29, 1037.82, 991.84, 942.02, 849.04, 838.03, 770.31, 743.45, 692.68, 686.94, 564.15, 555.84, 533.32, 518.31, 481.69, 469.73, 447.95, 411.89 cm−1.

Elemental analysis (%): C 67.15, H 6.84, P 10.09, Ru 10.75, Cl 3.67, B 0.61; C 65.91, H 6.33, P 11.74, Ru 11.59, Cl 3.48; C 67.05, H 6.44, P 7.55, Ru 8.35, Cl 5.97. (calculated: C 65.97, H 6.58, P 9.63, Ru 10.47, Cl 7.35).

Due to fast conversion into a different species, no pure 13C-NMR spectrum could be recorded.

142

7 Experimental Section

Route 2:

100 mg (0.10 mmol, 1.00 eq) dichloridotris(triphenylphosphine)ruthenium(II) and 82.7 mg (0.10 mmol, 1.00 eq) 1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane were dissolved in 10 mL toluene and stirred for 17 h at 115 °C. After cooling of the reaction solution to room temperature, the orange supernatant was removed and the orange solid washed with 10 mL toluene and 3x10 mL n-pentane, then dried in vacuo.

Yield: 25 mg (25%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.10–7.00 (m, 12 H, Car,ortho,xylH), 6.82–6.72 (m, 6 H, Car,para,xylH), 2.32– 2.25 (m, 6 H, (CH2)3CCH3), 2.04 (s, 36 H, Car,xylCH3), 1.66–1.58 (m, 3 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 50.2 (s, 3 P) ppm.

Elemental analysis (%): C 65.02, H 6.51, P 6.51, Ru 10.21, Cl 7.51. (calculated: C 65.97, H 6.58, P 9.63, Ru 10.47, Cl 7.35).

Due to fast conversion to a different species, no pure 13C-NMR spectrum could be recorded.

143

7 Experimental Section

7.3.31 Synthesis of Dibromido-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II)

Abbreviation: cat-II-67a

200 mg (0.21 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) and 231.6 mg (0.42 mmol, 2.00 eq) hexabromobenzene were dispersed in 5 mL THF and transferred into a 20 mL stainless steel finger autoclave equipped with a glass inlet and a magnetic stir bar. The autoclave was pressurized with hydrogen to 50 bar and heated at 160 °C in an aluminum cone fitted on a magnetic stir plate for 2 h. The autoclave was cooled down in an ice bath and carefully vented. After settling of the brown solid, the brown supernatant was removed. The solid was washed with 10 mL THF and 2x10 mL n-pentane and then dried in vacuo.

Yield: 169.0 mg (76%)

31P{1H}-NMR (243 MHz, CD2Cl2): δ = 50.2 (s, 3 P) ppm.

FAB-MS: m/z = 973.1 (M+−Br calculated = 973.2).

IR: ṽ = 2990.74, 2952.49, 2911.09, 2857.60, 1598.31, 1584.22, 1451.68, 1412.75, 1374.38, 1290.89, 1230.86, 1169.87, 1127.90, 1106.57, 1095.52, 1037.88, 992.09, 942.84, 893.98, 847.99, 837.48, 768.87, 742.81, 687.08, 563.32, 553.83, 532.86, 517.70, 480.52, 469.05, 446.96, 423.36, 409.44 cm−1.

Elemental analysis (%): C 59.05, H 6.00, P 8.69, Ru 9.41, Br 15.33, B 0.53. (calculated: C 60.40, H 6.03, P 8.82, Ru 9.59, Br 15.16).

Due to fast conversion to a different species, no pure 1H- and 13C-NMR spectra could be recorded.

144

7 Experimental Section

7.3.32 Synthesis of Tri-µ-bromido-bis(1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethane)diruthenium(II) bromide

Abbreviation: cat-II-67b

231.6 mg (0.42 mmol, 2.00 eq) hexabromobenzene were weighed into a glass inlet, which was inserted into a 20 mL stainless steel autoclave together with a magnetic stir bar. The autoclave was evacuated and then filled with argon. Under argon counter current flow, a solution of 200 mg (0.21 mmol, 1.00 eq) trimethylenemethane-1,1,1-tris(di(3,5-dimethylphenyl)phosphinomethyl)ethaneruthenium(II) in 5 mL THF was transferred into the autoclave. The autoclave was pressurized with hydrogen to 50 bar and heated at 160 °C in an aluminum cone fitted on a magnetic stir plate for 2 h. The autoclave was cooled down in an ice bath and carefully vented. After settling of the brown solid, the supernatant was removed. The solid was washed with 2 mL THF and 2x10 mL n-pentane and then dried in vacuo. 12 mL DCM were added and the reaction mixture stirred at room temperature for 26 h. The yellow solution was separated from the colorless precipitate. After removal of the solvent in vacuo, the yellow solid was washed with 3x10 mL n-pentane and dried in vacuo.

Yield: 144.7 mg (33%)

1H-NMR (400 MHz, CD2Cl2): δ = 7.53–7.27 (m, 12 H, Car,ortho,xylH), 7.23–7.07 (m, 12 H, Car,ortho,xylH), 6.88–6.75 (m, 12 H, Car,paraxylH), 2.51–2.35 (m, 6 H, (CH2)3CCH3), 2.34–2.22 (m, 6 H, (CH2)3CCH3), 1.87 (s, 36 H, Car,xylCH3), 1.81 (s, 36 H, Car,xylCH3), 1.52–1.43 (m, 6 H, (CH2)3CCH3) ppm.

31P{1H}-NMR (162 MHz, CD2Cl2): δ = 27.3 (s, 3 P) ppm.

13C{1H}-NMR (101 MHz, CD2Cl2): δ = 138.6–137.9 (m, Car,ipso,xyl), 137.9–137.4 (m, Car,meta,xyl), 137.3– 136.7 (m, Car,meta,xyl), 136.1–135.2 (m, Car,ipso,xyl), 132.3–131.7 (m, Car,ortho,xylH), 131.6–131.3 (m, Car,para,xylH), 131.1–130.7 (m, Car,ortho,xylH), 38.6–37.9 (m, (CH2)3CCH3), 36.6–36.2 (m, (CH2)3CCH3), 35.1–34.0 (m, (CH2)3CCH3), 21.1 (s, Car,xylCH3), 20.8 (s, Car,xylCH3) ppm.

FAB-MS: m/z = 2027.5 (M+ calculated = 2027.4).

IR: ṽ = 2992.65, 2911.35, 2856.63, 1598.05, 1581.87, 1443.13, 1411.23, 1376.70, 1290.83, 1220.03, 1123.28, 1099.05, 1037.69, 994.57, 838.38, 783.74, 740.33, 694.75, 688.20, 562.69, 542.65, 519.55, 484.31, 451.20, 429.19, 413.76 cm−1.

145

7 Experimental Section

Elemental analysis (%): C 58.38, H 5.91, P 8.36, Ru 9.12, Br 14.87. (calculated: C 60.40, H 6.03, P 8.82, Ru 9.59, Br 15.16).

146

7 Experimental Section

7.3.33 Synthesis of Heptafluoro-2-naphthol

957 mg (3.52 mmol, 1.00 eq) octafluoronaphthalene and 464 mg (8.45 mmol, 2.40 eq) potassium hydroxide were combined in 7 mL tert-butanol. The reaction mixture was refluxed for 4 h. After cooling to room temperature, 17 mL water was added under stirring. After removal of tert-butanol by distillation, the yellow solution was extracted with 3x6 mL diethyl ether. The aqueous phase was acidified with 2.5 mL HCl-solution (5 M) and extracted with 2x6 mL diethyl ether. The combined organic phases were washed with 10 mL water and dried with sodium sulfate. After removal of diethyl ether in vacuo, the residual yellow solid was washed with 1x8 mL and 1x5 mL n-pentane. After drying in vacuo, the product was obtained as yellow solid.

Yield: 274 mg (29%)

1H-NMR (400 MHz, THF-d8): δ = 10.56 (br s, OH) ppm.

19F{1H}-NMR (377 MHz, THF-d8): δ = −148.46–−148.84 (m, 1 F), −148.84–−149.15 (m, 1 F), −149.53–−149.87 (m, 1 F), −150.30–−150.70 (m, 1 F), −153.49–−153.85 (m, 1 F), −158.78–−159.18 (m, 1 F), −161.41–−161.71 (m, 1 F) ppm.

The synthesis was carried out following literature procedures.[106]

147

7 Experimental Section

7.4 Procedures of Catalytic Experiments

7.4.1 General Procedure for the Hydrogenation of Levulinic Acid, γ-Valerolactone and α-Methyl-γ-butyrolactone with Air-sensitive Catalytic Systems Under argon atmosphere, the catalyst and additive were dissolved in the liquid substrate and the reaction mixture was transferred under argon countercurrent flow into a previously evacuated 10 mL stainless steel finger autoclave equipped with a glass inlet and a magnetic stir bar. The autoclave was pressurized with hydrogen and heated at reaction temperature in a preheated aluminum cone fitted on a magnetic stir plate. After completion, the autoclave was cooled down in an ice bath and carefully vented. 1,4-Dioxane was added via the top screw plug of the autoclave, then the product mixture was retrieved. The glass inlet was rinsed with 1,4-dioxane and the rinsing solution combined with the product mixture, which was then weighed and analyzed by gas chromatography.

7.4.2 General Procedure for the Hydrogenation of Itaconic Acid with Air-sensitive Catalytic Systems Under argon atmosphere, the catalyst and additive were dissolved in 1,4-dioxane and the solution transferred under argon countercurrent flow into a previously evacuated 10 mL stainless steel finger autoclave equipped with a glass inlet containing itaconic acid and a magnetic stir bar. The autoclave was pressurized with hydrogen and heated at reaction temperature in a preheated aluminum cone fitted on a magnetic stir plate. After completion, the autoclave was cooled down in an ice bath and carefully vented. 1,4-Dioxane was added via the top screw plug of the autoclave, then the product mixture was retrieved. The glass inlet was rinsed with 1,4-dioxane and the rinsing solution combined with the product mixture, which was then weighed and analyzed by gas chromatography.

7.4.3 General Procedure for the Hydrogenation of Itaconates with Air-sensitive Catalytic Systems Under argon atmosphere, the catalyst and additive were weighed into a glass inlet and covered with the itaconates. The glass inlet was quickly inserted into a 10 mL stainless steel finger autoclave together with a magnetic stir bar. The autoclave was evacuated and then filled with argon. It was pressurized with hydrogen and heated at reaction temperature in a preheated aluminum cone fitted on a magnetic stir plate. After completion, the autoclave was cooled down in an ice bath and carefully vented. 1,4-Dioxane was added via the top screw plug of the autoclave, then the product

148

7 Experimental Section

mixture was retrieved. The glass liner was rinsed with 1,4-dioxane and the rinsing solution combined with the product mixture, which was then weighed and analyzed by gas chromatography.

7.4.4 General Procedure for the Hydrogenation of High-boiling Substrates with Air-stable Catalytic Systems The catalyst and substrate were weighed into a glass inlet under air, inserted in a 10 mL stainless steel autoclave together with a magnetic stir bar. The autoclave was evacuated and then filled with argon. The autoclave was pressurized with hydrogen and heated at reaction temperature in an aluminum cone fitted on a magnetic stir plate. After completion, the autoclave was cooled down in an ice bath and carefully vented. 1,4-Dioxane was added via the top screw plug of the autoclave, then the product mixture was retrieved. The glass liner was rinsed with 1,4-dioxane and the rinsing solution combined with the product mixture, which was then weighed and analyzed by gas chromatography.

7.4.5 General Procedure for the Hydrogenation of Low-boiling Substrates with Air- stable Catalytic Systems The catalyst was weighed into a glass inlet under air and inserted in a 10 mL stainless steel autoclave together with a magnetic stir bar. The autoclave was evacuated and then filled with argon. The substrate was added under argon countercurrent flow. The autoclave was pressurized with hydrogen and heated at reaction temperature in an aluminum cone fitted on a magnetic stir plate. After completion, the autoclave was cooled down in an ice bath and carefully vented. 1,4-Dioxane was added via the top screw plug of the autoclave, then the product mixture was retrieved. The glass liner was rinsed with 1,4-dioxane and the rinsing solution combined with the product mixture, which was then weighed and analyzed by gas chromatography.

7.4.6 General Procedure for the Dehydrogenation of Diols A suspension of the catalyst in the respective diol was transferred under argon countercurrent flow into a previously evacuated 10 mL stainless steel finger autoclave equipped with a glass inlet and a magnetic stir bar. The autoclave was closed and heated at reaction temperature in a preheated aluminum cone fitted on a magnetic stir plate. After completion, the autoclave was cooled down in an ice bath and carefully vented. The product mixture was retrieved, the glass liner was rinsed with 1,4-dioxane and the rinsing solution combined with the product mixture, which was then weighed and analyzed by gas chromatography.

149

8 Appendix

8.1 SC-XRD Data

Table 43: Single crystal X-ray diffractometric data for cat-II-49.

Bond precision: C-C = 0.0044 A Wavelength=0.71073 Cell: a=12.0486(9) b=14.0264(11) c=18.2150(14) alpha=95.594(1) beta=104.093(1) gamma=113.154(1) Temperature: 100 K Calculated Reported Volume 2680.0(4) 2680.0(4) Space group P -1 P -1 Hall group -P 1 -P 1 Moiety formula C63 H69 O2 P3 Ru Sum formula C63 H69 O2 P3 Ru C63 H69 O2 P3 Ru Mr 1052.16 1052.16 Dx, g cm-3 1.304 1.304 Z 2 2 Mu (mm-1) 0.426 0.426 F000 1104.0 1104.0 F000’ 1102.43 h, k, lmax 17, 20, 26 17, 19, 25 Nref 16455 15450 Tmin, Tmax 0.950, 0.979 Tmin’ 0.888

Correction method= Not given

Data completeness= 0.939

Theta(max)= 30.576

R(reflections)= 0.0517( 11920) wR2(reflections)= 0.1540( 15450)

S = 0.956

Npar= 622

150

8 Appendix

Table 44: Single crystal X-ray diffractometric data for cat-II-54.

Bond precision: C-C = 0.0045 A

Wavelength=0.71073

Cell: a=12.2050(8) b=23.1908(14) c=25.1013(16) alpha=90 beta=90.872(1) gamma=90

Temperature: 100 K

Calculated Reported Volume 7103.9(8) 7103.9(8) Space group P 21/n P21/n Hall group -P 2yn -P 2yn C65 H69 F4 O2 P3 Ru, 3 (C6 Moiety formula H6) Sum formula C83 H87 F4 O2 P3 Ru C83 H87 F4 O2 P3 Ru Mr 1386.51 1386.58 Dx, g cm-3 1.296 1.296 Z 4 4 Mu (mm-1) 0.346 0.334 F000 2904.0 2400.0 F000’ 2901.32 h, k, lmax 14, 28, 30 14, 28, 30 Nref 13215 13158 Tmin, Tmax 0.953, 0.967 0.635, 0.745 Tmin’ 0.929

Correction method= #

Reported T Limits: Tmin=0.635 Tmax=0.745

AbsCorr = MULTI-SCAN

Data completeness= 0.996

Theta(max)= 25.492

R(reflections)= 0.0424( 10391) wR2(reflections)= 0.1372( 13158)

S = 0.933

Npar= 838

151

8 Appendix

Table 45: Single crystal X-ray diffractometric data for cat-II-58.

Bond precision: C-C = 0.0047 A

Wavelength=0.71073

Cell: a=13.4342(9) b=15.1001(9) c=19.6127(12) alpha=109.048(1) beta=91.127(1) gamma=114.421(1)

Temperature: 100 K

Calculated Reported Volume 3368.6(4) 3368.6(4) Space group P -1 P -1 Hall group -P 1 -P 1 C65 H63 F10 O2 P3 Ru, 2 (C6 Moiety formula H6) Sum formula C77 H75 F10 O2 P3 Ru C77 H75 F10 O2 P3 Ru Mr 1416.35 1416.14 Dx, g cm-3 1.396 1.396 Z 2 2 Mu (mm-1) 0.379 0.370 F000 1464.0 1296.0 F000’ 1462.83 h, k, lmax 18, 21, 27 18, 21, 27 Nref 19539 18982 Tmin, Tmax 0.952, 0.982 0.655, 0.746 Tmin’ 0.915

Correction method= #

Reported T Limits: Tmin=0.655 Tmax=0.746

AbsCorr = MULTI-SCAN

Data completeness= 0.971

Theta(max)= 29.933

R(reflections)= 0.0527( 14565) wR2(reflections)= 0.1605( 18982)

S = 0.967

Npar= 838

152

8 Appendix

Table 46: Single crystal X-ray diffractometric data for cat-II-63a with F– as counter anion.

Bond precision: C-C = 0.0057 A

Wavelength= 0.71073

Cell: a= 32.9003(6) b= 27.8932(6) c= 25.2891(5)

Alpha= 90 beta= 90 gamma= 90

Temperature: 120 K

Calculated Reported Volume 23207.7(8) 23207.7(8) Space group C m c a C m c a Hall group -C 2bc 2 -C 2bc 2 Moiety formula C106 H126 F3 P6 Ru2, C6 H12 C106 H126 F3 P6 Ru2, C6 H12 [+ solvent] Sum formula C112 H138 F3 P6 Ru2 [+ C112 H138 F3 P6 Ru2 solvent] Mr 1929.18 1929.18 Dx, g cm-3 1.104 1.104 Z 8 8 Mu (mm-1) 0.388 0.388 F000 8120.0 8120.0 F000’ 8107.25 h, k, lmax 39, 33, 30 39, 33, 30 Nref 10702 10689 Tmin, Tmax 0.876, 0.890 0.956, 1.000 Tmin’ 0.876

Correction method= #

Reported T Limits: Tmin= 0.956 Tmax= 1.000

AbsCorr = MULTI-SCAN

Data completeness= 0.999

Theta(max)= 25.247

R(reflections)= 0.0454 ( 7206) wR2(reflections)= 0.1227 ( 10689)

S = 1.038

Npar= 571

153

8 Appendix

Table 47: Single crystal X-ray diffractometric data for cat-II-63a with PO2F2– as counter anion.

Bond precision: C-C = 0.0056 A

Wavelength= 0.71073

Cell: a= 31.8866(5) b= 27.9692(3) c= 25.5036(3)

Alpha= 90 beta= 90 gamma= 90

Temperature: 120 K

Calculated Reported Volume 22745.2(5) 22745.2(5) Space group P b c n P b c n Hall group -P 2n 2ab -P 2n 2ab Moiety formula 2 (C106 H126 F3 P6 Ru2), 4 2 (C106 H126 F3 P6 Ru2), 4 (C6 H5 Cl), 2 (C6 H12), F2 O2 (C6 H5 Cl), 2 (C6 H12), F2 O2 P [+ solvent] P Sum formula C248 H296 Cl4 F8 O2 P13 Ru4 C248 H296 Cl4 F8 O2 P13 [+ solvent] Ru4 Mr 4409.55 4409.52 Dx, g cm-3 1.288 1.288 Z 4 4 Mu (mm-1) 0.459 0.459 F000 9244.0 9244.0 F000’ 9234.49 h, k, lmax 38, 33, 30 38, 33, 30 Nref 20508 20479 Tmin, Tmax 0.856, 0.871 0.927, 1.000 Tmin’ 0.856

Correction method= #

Reported T Limits: Tmin= 0.927 Tmax= 1.000

AbsCorr = MULTI-SCAN

Data completeness= 0.999

Theta(max)= 25.200

R(reflections)= 0.0431 ( 14622) wR2(reflections)= 0.1445 ( 20479)

S = 1.085

Npar= 1277

154

8 Appendix

Table 48: Single crystal X-ray diffractometric data for cat-II-63b.

Bond precision: C-C = 0.0041 A

Wavelength= 0.71073

Cell: a= 9.9212(3) b= 11.8676(4) c= 24.1183(8) alpha= 103.956(2) beta= 96.157(1) gamma= 101.248(1)

Temperature: 120 K

Calculated Reported Volume 2667.20(15) 2667.20(15) Space group P -1 P -1 Hall group -P 1 -P 1 Moiety formula C53 H63 F6 P4 Ru, C H2 Cl2 C 53 H63 F6 P4 Ru, C H2 Cl2 Sum formula C54 H65 Cl2 F6 P4 Ru C54 H65 Cl2 F6 P4 Ru Mr 1123.91 1123.91 Dx, g cm-3 1.400 1.399 Z 2 2 Mu (mm-1) 0.570 0.570 F000 1162.0 1162.0 F000’ 1161.35 h, k, lmax 12, 14, 29 12, 14, 29 Nref 10789 10722 Tmin, Tmax 0.783, 0.843 0.968, 0.989 Tmin’ 0.783

Correction method= #

Reported T Limits: Tmin= 0.968 Tmax= 0.989

AbsCorr = GAUSSIAN

Data completeness= 0.994

Theta(max)= 26.250

R(reflections)= 0.0372 ( 9065) wR2(reflections)= 0.0946 ( 10722)

S = 1.042

Npar= 617

155

8 Appendix

Table 49: Single crystal X-ray diffractometric data for cat-II-67b.

Bond precision: C-C = 0.0105 A

Wavelength= 0.71073

Cell: a= 33.8637(4) b= 27.7948(6) c= 25.3137(6) alpha= 90 beta= 90 gamma= 90

Temperature: 120 K

Calculated Reported Volume 23826.1(8) 23826.1(8) Space group C m c a C m c a Hall group -C 2bc 2 -C 2bc 2 Moiety formula C106 H124 Br3 P6 Ru2, Br [+ C106 H124 Br3 P6 Ru2, Br solvent] [+solvent] Sum formula C106 H124 Br 4 P6 Ru2 [+ C106 H124 Br4 P6 Ru2 solvent] Mr 2105.61 2105.64 Dx, g cm-3 1.174 1.174 Z 8 8 Mu (mm-1) 1.714 1.714 F000 8624.0 8624.0 F000’ 8601.78 h, k, lmax 40, 33, 30 40, 33, 30 Nref 10990 10951 Tmin, Tmax 0.614, 0.710 0.641, 0.746 Tmin’ 0.602

Correction method= #

Reported T Limits: Tmin= 0.641 Tmax= 0.746

AbsCorr = MULTI-SCAN

Data completeness= 0.996

Theta(max)= 25.250

R(reflections)= 0.0591 ( 6308) wR2(reflections)= 0.1673 (10951)

S = 1.010 Npar= 547

156

9 Bibliography

[1] C. H. Christensen, J. Rass-Hansen, C. C. Marsden, E. Taarning, K. Egeblad, ChemSusChem 2008, 1, 283-289. [2] a) T. A. Werpy, J. Holladay, J. White, Top Value Added Chemicals From Biomass: I. Results of Screening for Potential Candidates from Sugars and Synthesis Gas, 2004; b) J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539-554. [3] T. J. Farmer, M. Mascal, in Introduction to Chemicals from Biomass, second ed. (Eds.: J. Clark, F. Deswarte), John Wiley & Sons, Ltd., 2015. [4] a) F. Cherubini, Energy Convers. Manage. 2010, 51, 1412-1421; b) G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044-4098. [5] a) S. Guran, F. A. Agblevor, M. Brennan-Tonetta, in Emerging Areas in Bioengineering (Ed.: H. N. Chang), Wiley‐VCH Verlag GmbH & Co. KGaA, 2018; b) T. Searchinger, R. Heimlich, R. Houghton, F. Dong, A. Elobeid, J. Fabiosa, S. Tokgoz, D. Hayes, T.-H. Yu, Science 2008, 319, 1238-1240; c) T. Hertel, A. Golub, A. Jones, M. O'Hare, R. Plevin, D. Kammen, Effects of US Maize Ethanol on Global Land Use and Greenhouse Gas Emissions: Estimating Market-Mediated Responses, Vol. 60, 2010; d) J. Fargione, J. Hill, D. Tilman, S. Polasky, P. Hawthorne, Science 2008, 319, 1235-1238. [6] a) D. Mitchel, A Note On Rising Food Prices, 2008; b) F. Lenk, S. Bröring, P. Herzog, J. Leker, Biotechnol. J. 2007, 2, 1497-1504. [7] a) M. Boot, Biofuels from Lignocellulosic Biomass: Innovations Beyond Bioethanol, Wiley‐VCH Verlag GmbH & Co. KGaA. , 2016; b) W. Leitner, J. Klankermayer, S. Pischinger, H. Pitsch, K. Kohse-Höinghaus, Angew. Chem. Int. Ed. 2017, 56, 5412-5452. [8] R. A. Sheldon, Green Chem. 2014, 16, 950-963. [9] F. D. Pileidis, M. M. Titirici, ChemSusChem 2016, 9, 562-582. [10] G. J. Mulder, J. prakt. Chem. 1840, 21, 203-240. [11] a) D. Garcés, L. Faba, E. Diaz, S. Ordóñez, ChemSusChem 2019, 12, 924-934; b) A. Mukherjee, M. J. Dumont, V. Raghauan, Biomass Bioenerg. 2015, 72, 143-183. [12] a) M. Patel, M. Crank, V. Dornburg, B. Hermann, L. Roes, B. Hüsing, L. Overbeek, F. Terragni, E. Recchia, Utrecht University: European Commission’s GROWTH Programme (DG Research), Utrecht, 2006; b) B. V. Timokhin, V. A. Baransky, G. D. Eliseeva, Russ. Chem. Rev. 1999, 68, 73-84. [13] D. J. Hayes, S. Fitzpatrick, M. H. B. Hayes, J. R. H. Ross, in Biorefineries-Industrial Processes and Products, Wiley‐VCH Verlag GmbH & Co. KGaA. , 2006. [14] a) I. T. Horváth, H. Mehdi, V. Fábos, L. Boda, L. T. Mika, Green Chem. 2008, 10, 238- 242; b) D. M. Alonso, S. G. Wettstein, J. A. Dumesic, Green Chem. 2013, 15, 584-595. [15] a) A. Stradi, M. Molnar, P. Szakal, G. Dibo, D. Gaspar, L. T. Mika, RSC Adv. 2015, 5, 72529-72535; b) M. Chalid, H. J. Heeres, A. A. Broekhuis, Procedia Chem. 2012, 4, 260- 267; c) L. E. Manzer, Appl. Catal., A 2004, 272, 249-256. [16] a) D. F. Aycock, Org. Process Res. Dev. 2007, 11, 156-159; b) T. Eifert, M. A. Liauw, React. Chem. Eng. 2016, 1, 521-532; c) S. Kar, R. Sen, A. Goeppert, G. K. S. Prakash, J. Am. Chem. Soc. 2018, 140, 1580-1583. [17] a) A. Démolis, N. Essayem, F. Rataboul, ACS Sustain. Chem. Eng. 2014, 2, 1338-1352; b) K. Y. Nandiwale, V. V. Bokade, Environ. Prog. Sustain. Energy 2015, 34, 795-801; c) L. Lomba, B. Giner, I. Bandrés, C. Lafuente, M. R. Pino, Green Chem. 2011, 13, 2062- 2070. [18] a) H.-F. Liu, F.-X. Zeng, L. Deng, B. Liao, H. Pang, Q.-X. Guo, Green Chem. 2013, 15, 81-84; b) S. Van de Vyver, J. Geboers, S. Helsen, F. Yu, J. Thomas, M. Smet, W. Dehaen, B. F. Sels, Chem. Commun. 2012, 48, 3497-3499.

157

9 Bibliography

[19] a) W. Stummer, U. Pichlmeier, T. Meinel, O. D. Wiestler, F. Zanella, H. J. Reulen, Lancet Oncol. 2006, 7, 392-401; b) I. Y. Eyüpoglu, M. Buchfelder, N. E. Savaskan, Nat. Rev. Neurol. 2013, 9, 141-151; c) R. Pokluda, S. M. Shehata, T. Kopta, Gesunde Pflanzen 2018, 70, 21-29. [20] T. Klement, J. Büchs, Bioresour. Technol. 2013, 135, 422-431. [21] S. Baup, Annalen der Pharmacie 1836, 19, 29-38. [22] K. Kinoshita, Acta Phytochim. 1932, 5, 271-287. [23] a) L. B. Lockwood, M. D. Reeves, Arch. Biochem. 1945, 6, 455-469; b) G. E. N. Nelson, D. H. Traufler, S. E. Kelley, L. B. Lockwood, Ind. Eng. Chem. 1952, 44, 1166-1168; c) V. F. Pfeifer, C. Vojnovich, E. N. Heger, Ind. Eng. Chem. 1952, 44, 2975-2980; d) T. Kobayashi, I. Nakamura, J. Ferment. Technol. 1964, 44, 264-274; e) T. Kobayashi, Process Biochem. 1967, 2, 61 - 65. [24] a) T. Willke, K. D. Vorlop, Appl. Microbiol. Biotechnol. 2001, 56, 289-295; b) A. I. Magalhães, Jr., J. C. de Carvalho, J. D. C. Medina, C. R. Soccol, Appl. Microbiol. Biotechnol. 2017, 101, 1-12; c) B. C. Saha, J. Ind. Microbiol. Biotechnol. 2017, 44, 303- 315. [25] a) W. Jaklitsch, C. Kubicek, M. C. Scrutton, Microbiology 1991, 137, 533-539; b) P. Bonnarme, B. Gillet, A. M. Sepulchre, C. Role, J. C. Beloeil, C. Ducrocq, J. Bacteriol. 1995, 177, 3573-3578; c) X. Huang, X. Lu, Y. Li, X. Li, J. J. Li, Microb. Cell Fact. 2014, 13, 119. [26] M. Okabe, D. Lies, S. Kanamasa, E. Y. Park, Appl. Microbiol. Biotechnol. 2009, 84, 597- 606. [27] a) M. Okabe, N. Ohta, Y. S. Park, J. Ferment. Bioeng. 1993, 76, 117-122; b) Y. S. Park, M. Itida, N. Ohta, M. Okabe, J. Ferment. Bioeng. 1994, 77, 329-331; c) K. Yahiro, T. Takahama, S. R. Jai, Y. S. Park, M. Okabe, Biotechnol. Lett. 1997, 19, 619-621. [28] a) H. Hajian, W. M. W. Yusoff, Curr. Res. J. Biol. Sci. 2015, 7, 37-42; b) A. Hevekerl, A. Kuenz, K. D. Vorlop, Appl. Microbiol. Biotechnol. 2014, 98, 10005-10012; c) N. Ju, S. S. Wang, Appl. Microbiol. Biotechnol. 1986, 23, 311-314; d) H. Kautola, Appl. Microbiol. Biotechnol. 1990, 33, 7-11. [29] a) R. H. Haskins, J. A. Thorn, B. Boothroyd, Can. J. Microbiol. 1955, 1, 749-756; b) T. Tabuchi, T. Sugisawa, T. Ishidori, T. Nakahara, J. Sugiyama, Agric. Biol. Chem. 1981, 45, 475-479; c) E. Geiser, S. K. Przybilla, A. Friedrich, W. Buckel, N. Wierckx, L. M. Blank, M. Bolker, Microb. Biotechnol. 2016, 9, 116-126; d) R. Bafana, S. Sivanesan, R. A. Pandey, Indian J. Microbiol. 2017, 57, 322-328; e) F. Carstensen, T. Klement, J. Büchs, T. Melin, M. Wessling, Bioresour. Technol. 2013, 137, 179-187. [30] a) T. Kobayashi, I. Nakamura, Process for Recovering Itaconic Acid and Salts Thereof from Fermented Broth, 1971, US000003621053A; b) D. Kreyenschulte, B. Heyman, A. Eggert, T. Massmann, C. Kalvelage, R. Kossack, L. Regestein, A. Jupke, J. Büchs, Biochem. Eng. J. 2018, 135, 133-141; c) K. L. Wasewar, D. Shende, A. Keshav, Ind. Eng. Chem. Res. 2011, 50, 1003-1011; d) K. L. Wasewar, D. Shende, A. Keshav, J. Chem. Technol. Biot. 2011, 86, 319-323; e) K. Schute, C. Detoni, A. Kann, O. Jung, R. Palkovits, M. Rose, ACS Sustain. Chem. Eng. 2016, 4, 5921-5928; f) J. J. Gulicovski, L. S. Čerović, S. K. Milonjić, I. G. Popović, J. Serb. Chem. Soc. 2008, 73, 825-834; g) A. I. Magalhães, J. C. de Carvalho, E. N. M. Ramírez, J. D. C. Medina, C. R. Soccol, J. Chem. Eng. Data 2016, 61, 430-437. [31] a) T. Kobayashi, I. Nakamura, N. Mitutoshi, in Ferment. Technol. Today, Proc. Int. Ferment. Symp., 4th (Ed.: T. Gyozo), Soc. Ferment. Technol., Jpn. Osaka, Japan, 1972, pp. 215-221; b) M. Fidaleo, M. Moresi, J. Membr. Sci. 2010, 349, 393-404; c) J. Stodollick, R. Femmer, M. Gloede, T. Melin, M. Wessling, J. Membr. Sci. 2014, 453, 275- 281. [32] a) J. Grislis, V. Zamalh, P. Kaulins, V. Lukso, S. P. A., Itaconic Acid, 1976, SU 507633; b) J. H. Kane, A. C. Finlay, P. F. Amann, Production of Itaconic Acid, 1945, 158

9 Bibliography

US000002385283A; c) R. C. Nubel, E. J. Ratajak, Process for Producing Itaconic Acid, 1962, US000003044941A; d) P. Cros, D. Schneider, Microbiological Production of Itaconic Acid, 1993, US000005231016A; e) K. Yahiro, S. Shibata, S. R. Jia, Y. Park, M. Okabe, J. Ferment. Bioeng. 1997, 84, 375-377; f) M. Petruccioli, V. Pulci, F. Federici, Lett. Appl. Microbiol. 1999, 28, 309-312. [33] a) L. Regestein, T. Klement, P. Grande, D. Kreyenschulte, B. Heyman, T. Maßmann, A. Eggert, R. Sengpiel, Y. Wang, N. Wierckx, L. M. Blank, A. Spiess, W. Leitner, C. Bolm, M. Wessling, A. Jupke, M. Rosenbaum, J. Büchs, Biotechnol. Biofuels 2018, 11, 279; b) A. Jiménez-Quero, E. Pollet, M. Zhao, E. Marchioni, L. Averous, V. Phalip, J. Microbiol. Biotechnol. 2017, 27, 1-8; c) N. Tippkötter, A. M. Duwe, S. Wiesen, T. Sieker, R. Ulber, Bioresour. Technol. 2014, 167, 447-455; d) C. Zhao, S. Chen, H. Fang, Appl. Microbiol. Biotechnol. 2018, 102, 9577-9584. [34] M. I. Juy, J. A. Orejas, M. E. Lucca, Rev. Colomb. Biotecnol. 2010, 12, 187-193. [35] a) for a comprehensive review on IA-derived polyesters see: T. Robert, S. Friebel, Green Chem. 2016, 18, 2922-2934; b) for a comprehensive review on IA-derived resins and polyesters see: S. Kumar, S. Krishnan, S. K. Samal, S. Mohanty, S. K. Nayak, Polym. Int. 2017, 66, 1349-1363. [36] a) B. M. Culbertson, D. Xie, W. M. Johnston, in Specialty Monomers and Polymers, Vol. 755, American Chemical Society, 2000; b) D. Xie, W. Wu, A. Puckett, B. Farmer, J. W. Mays, Eur. Polym. J. 2004, 40, 343-351; c) B. M. Culbertson, M. H. Dotrong, S. R. Schicker, in Macromolecules Containing Metal and Metal‐Like Elements: Biomedical Applications, Volume 3, John Wiley & Sons, Inc. , 2004. [37] a) N. M. Ranjha, E. Doelker, pH-sensitive Hydrogels for Site-specific Drug Delivery: I. Swelling Behaviour of Crosslinked Copolymers of Acrylic and Methacrylic Acid, Vol. 9, 1999; b) A. Bal, F. E. Çepni, O. Çakir, I. Acar, G. Güçlü, Braz. J. Chem. Eng. 2015, 32, 509-518; c) M. Sakthivel, D. S. Franklin, S. Sudarsan, G. Chitra, S. Guhanathan, RSC Adv. 2016, 6, 106821-106831; d) J. C. Carrillo-Rodríguez, H. I. Meléndez-Ortiz, B. Puente-Urbina, G. Padrón, A. Ledezma, R. Betancourt-Galindo, Polym. Compos. 2018, 39, 171-180. [38] M. Sakthivel, D. S. Franklin, S. Guhanathan, Ecotoxicol. Environ. Saf. 2016, 134, 427- 432. [39] a) S. Q. Ma, X. Q. Liu, Y. H. Jiang, Z. B. Tang, C. Z. Zhang, J. Zhu, Green Chem. 2013, 15, 245-254; b) S. Ma, X. Liu, L. Fan, Y. Jiang, L. Cao, Z. Tang, J. Zhu, ChemSusChem 2014, 7, 555-562; c) S. Q. Ma, X. Q. Liu, Y. H. Jiang, L. B. Fan, J. X. Feng, J. Zhu, Sci. China Chem. 2014, 57, 379-388. [40] Z. Wang, T. Wei, X. Xue, M. M. He, J. J. Xue, M. Song, S. Z. Wu, H. L. Kang, L. Q. Zhang, Q. X. Jia, Polymer 2014, 55, 4846-4856. [41] a) F. M. Geilen, B. Engendahl, A. Harwardt, W. Marquardt, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2010, 49, 5510-5514; b) C. S. Spanjers, D. K. Schneiderman, J. Z. Wang, J. Wang, M. A. Hillmyer, K. Zhang, P. J. Dauenhauer, ChemCatChem 2016, 8, 3031-3035; c) X. R. Liu, X. C. Wang, Q. Liu, G. Q. Xu, X. M. Li, X. D. Mu, Catal. Today 2016, 274, 88-93. [42] F. H. Isikgor, C. R. Becer, Polym. Chem. 2015, 6, 4497-4559. [43] P. Brignou, M. P. Gil, O. Casagrande, J. F. Carpentier, S. M. Guillaume, Macromolecules 2010, 43, 8007-8017. [44] G. A. Olah, A. P. Fung, R. Malhotra, Synthesis 1981, 1981, 474-476. [45] D. Dechambre, J. Thien, A. Bardow, Fuel 2017, 209, 615-623. [46] a) J. C. Lansing, R. E. Murray, B. R. Moser, ACS Sustain. Chem. Eng. 2017, 5, 3132- 3140; b) X. H. Hao, A. Yoshida, J. Nishikido, Tetrahedron Lett. 2004, 45, 781-785; c) T. Qui, J. Dong, X. Wang, C. Ye, H. Wang, Z. Huang, C. Yang, J. Yang, L. Li, Rectification Method and Apparatus for Producing Methyl Methacrylate, 2017, CN 107056617.

159

9 Bibliography

[47] a) A. V. Shlyakhtin, I. E. Nifant'ev, D. A. Lemenovsky, D. P. Krut'ko, V. V. Bagrov, P. S. Timashev, V. K. Popov, V. N. Bagratashvili, Russ. J. Phys. Chem. B 2014, 8, 1049-1053; b) C. Robert, F. de Montigny, C. M. Thomas, ACS Catal. 2014, 4, 3586-3589; c) S. Aoki, A. Fukui, Polym. J. 1998, 30, 295-299; d) S. Chakraborty, S. G. Ramkumar, S. Ramakrishnan, Macromolecules 2017, 50, 5004-5013. [48] a) M. Bianchi, G. Menchi, F. Francalanci, F. Piacenti, U. Matteoli, P. Frediani, C. Botteghi, J. Organomet. Chem. 1980, 188, 109-119; b) R. A. Grey, G. P. Pez, A. Wallo, J. Corsi, Chem. Commun. 1980, 783-784; c) G. P. Pez, R. A. Grey, J. Corsi, J. Am. Chem. Soc. 1981, 103, 7528-7535; d) R. A. Grey, G. P. Pez, A. Wallo, J. Am. Chem. Soc. 1981, 103, 7536-7542; e) U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, J. Organomet. Chem. 1986, 299, 233-238; f) U. Matteoli, G. Menchi, M. Bianchi, F. Piacenti, S. Ianelli, M. Nardelli, J. Organomet. Chem. 1995, 498, 177-186; g) H. T. Teunissen, C. J. Elsevier, Chem. Commun. 1997, 667-668; h) H. T. Teunissen, Chem. Commun. 1998, 1367-1368. [49] J. Zhang, G. Leitus, Y. Ben-David, D. Milstein, Angew. Chem. Int. Ed. 2006, 45, 1113- 1115. [50] a) E. Fogler, E. Balaraman, Y. Ben-David, G. Leitus, L. J. W. Shimon, D. Milstein, Organometallics 2011, 30, 3826-3833; b) Y. Sun, C. Koehler, R. Tan, V. T. Annibale, D. Song, Chem. Commun. 2011, 47, 8349-8351; c) I. Carpenter, S. C. Eckelmann, M. T. Kuntz, J. A. Fuentes, M. B. France, M. L. Clarke, Dalton Trans. 2012, 41, 10136-10140; d) W. Kuriyama, T. Matsumoto, O. Ogata, Y. Ino, K. Aoki, S. Tanaka, K. Ishida, T. Kobayashi, N. Sayo, T. Saito, Org. Process Res. Dev. 2012, 16, 166-171; e) D. Spasyuk, S. Smith, D. G. Gusev, Angew. Chem. Int. Ed. 2012, 51, 2772-2775; f) D. Spasyuk, S. Smith, D. G. Gusev, Angew. Chem. Int. Ed. 2013, 52, 2538-2542; g) T. Otsuka, A. Ishii, P. A. Dub, T. Ikariya, J. Am. Chem. Soc. 2013, 135, 9600-9603; h) C. Ziebart, R. Jackstell, M. Beller, ChemCatChem 2013, 5, 3228-3231; i) G. A. Filonenko, E. Cosimi, L. Lefort, M. P. Conley, C. Copéret, M. Lutz, E. J. M. Hensen, E. A. Pidko, ACS Catal. 2014, 4, 2667-2671. [51] a) L. A. Saudan, C. M. Saudan, C. Debieux, P. Wyss, Angew. Chem. Int. Ed. 2007, 46, 7473-7476; b) S. Takebayashi, S. H. Bergens, Organometallics 2009, 28, 2349-2351; c) W. W. N. O, A. J. Lough, R. H. Morris, Chem. Commun. 2010, 46, 8240-8242; d) F. A. Westerhaus, B. Wendt, A. Dumrath, G. Wienhöfer, K. Junge, M. Beller, ChemSusChem 2013, 6, 1001-1005; e) W. Li, J. H. Xie, M. L. Yuan, Q. L. Zhou, Green Chem. 2014, 16, 4081-4085; f) X. Tan, Y. Wang, Y. Liu, F. Wang, L. Shi, K. H. Lee, Z. Lin, H. Lv, X. Zhang, Org. Lett. 2015, 17, 454-457; g) G. A. Filonenko, M. J. Aguila, E. N. Schulpen, R. van Putten, J. Wiecko, C. Müller, L. Lefort, E. J. Hensen, E. A. Pidko, J. Am. Chem. Soc. 2015, 137, 7620-7623. [52] G. A. Filonenko, R. van Putten, E. J. M. Hensen, E. A. Pidko, Chem. Soc. Rev. 2018, 47, 1459-1483. [53] M. Kilner, D. V. Tyers, S. P. Crabtree, M. A. Wood, Homogeneous Process for the Hydrogenation of Carboxylic Acids and Derivatives Using Water, 2003, WO 03/093208 A1. [54] A. A. Núñez Magro, G. R. Eastham, D. J. Cole-Hamilton, Chem. Commun. 2007, 3154- 3156. [55] F. M. Geilen, B. Engendahl, M. Hölscher, J. Klankermayer, W. Leitner, J. Am. Chem. Soc. 2011, 133, 14349-14358. [56] a) L. Rosi, M. Frediani, P. Frediani, J. Organomet. Chem. 2010, 695, 1314-1322; b) Y. Li, T. Yan, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2014, 53, 10476-10480; c) A. Dutta Chowdhury, R. Jackstell, M. Beller, ChemCatChem 2014, 6, 3360-3365; d) Y. Li, C. Topf, X. Cui, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2015, 54, 5196-5200; e) T. J. Korstanje, J. I. van der Vlugt, C. J. Elsevier, B. de Bruin, Science 2015, 350, 298-302; f) X. Cui, Y. Li, C. Topf, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2015, 54, 10596- 160

9 Bibliography

10599; g) A. Phanopoulos, A. J. P. White, N. J. Long, P. W. Miller, ACS Catal. 2015, 5, 2500-2512; h) I. Sorribes, J. R. Cabrero-Antonino, C. Vicent, K. Junge, M. Beller, J. Am. Chem. Soc. 2015, 137, 13580-13587; i) J. R. Cabrero-Antonino, E. Alberico, K. Junge, H. Junge, M. Beller, Chem. Sci. 2016, 7, 3432-3442; j) L. Deng, B. Kang, U. Englert, J. Klankermayer, R. Palkovits, ChemSusChem 2016, 9, 177-180; k) Z. He, H. Liu, Q. Qian, L. Lu, W. Guo, L. Zhang, B. Han, Sci. China Chem. 2017, 60, 927-933; l) B. G. Schieweck, J. Klankermayer, Angew. Chem. Int. Ed. 2017, 56, 10854-10857; m) B. Emayavaramban, P. Chakraborty, B. Sundararaju, ChemSusChem 2019, 12, 3089-3093; n) F. K. Scharnagl, M. F. Hertrich, G. Neitzel, R. Jackstell, M. Beller, Adv. Synth. Catal. 2019, 361, 374-379; o) F. Christie, A. Zanotti-Gerosa, D. Grainger, ChemCatChem 2018, 10, 1012-1018; p) W. Liu, B. Sahoo, A. Spannenberg, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2018, 57, 11673-11677; q) L. Iffland, A. Khedkar, A. Petuker, M. Lieb, F. Wittkamp, M. van Gastel, M. Roemelt, U.-P. Apfel, Organometallics 2019, 38, 289-299. [57] a) S. Wesselbaum, T. Vom Stein, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2012, 51, 7499-7502; b) S. Wesselbaum, V. Moha, M. Meuresch, S. Brosinski, K. M. Thenert, J. Kothe, T. V. Stein, U. Englert, M. Hölscher, J. Klankermayer, W. Leitner, Chem. Sci. 2015, 6, 693-704; c) J. Klankermayer, W. Leitner, Science 2015, 350, 629- 630. [58] T. vom Stein, T. Weigand, C. Merkens, J. Klankermayer, W. Leitner, ChemCatChem 2013, 5, 439-441. [59] a) K. Beydoun, T. vom Stein, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2013, 52, 9554-9557; b) T. vom Stein, M. Meuresch, D. Limper, M. Schmitz, M. Hölscher, J. Coetzee, D. J. Cole-Hamilton, J. Klankermayer, W. Leitner, J. Am. Chem. Soc. 2014, 136, 13217-13225; c) K. Beydoun, G. Ghattas, K. Thenert, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2014, 53, 11010-11014. [60] a) T. vom Stein, T. den Hartog, J. Buendia, S. Stoychev, J. Mottweiler, C. Bolm, J. Klankermayer, W. Leitner, Angew. Chem. Int. Ed. 2015, 54, 5859-5863; b) K. Beydoun, K. Thenert, E. S. Streng, S. Brosinski, W. Leitner, J. Klankermayer, ChemCatChem 2016, 8, 135-138; c) K. Thenert, K. Beydoun, J. Wiesenthal, W. Leitner, J. Klankermayer, Angew. Chem. Int. Ed. 2016, 55, 12266-12269. [61] a) A. B. Chaplin, P. J. Dyson, Inorg. Chem. 2008, 47, 381-390; b) B. J. Sarmah, D. K. Dutta, J. Organomet. Chem. 2010, 695, 781-785; c) S. Savourey, G. Lefèvre, J. C. Berthet, P. Thuéry, C. Genre, T. Cantat, Angew. Chem. Int. Ed. 2014, 53, 10466-10470. [62] a) A. Phanopoulos, N. J. Brown, A. J. White, N. J. Long, P. W. Miller, Inorg. Chem. 2014, 53, 3742-3752; b) A. Phanopoulos, N. Long, P. Miller, J. Vis. Exp. 2015; c) M. Siebert, M. Seibicke, A. Siegle, S. Kräh, O. Trapp, J. Am. Chem. Soc. 2019, 141, 334-341; d) M. Seibicke, M. Siebert, A. F. Siegle, S. M. Gutenthaler, O. Trapp, Organometallics 2019, 38, 1809-1814. [63] a) J. Scherer, G. Huttner, O. Walter, B. C. Janssen, L. Zsolnai, Chem. Ber. 1996, 129, 1603-1615; b) A. Jacobi, G. Huttner, U. Winterhalter, J. Organomet. Chem. 1998, 571, 231-241. [64] a) M. Meuresch, S. Westhues, W. Leitner, J. Klankermayer, Angew. Chem. Int. Ed. 2016, 55, 1392-1395; b) J. Klankermayer, W. Leitner, M. Meuresch, Method for the Reduction of Organic Molecules 2016, WO002016128044A1. [65] N. Westhues, M. Belleflamme, J. Klankermayer, ChemCatChem 2019, 11, 5269-5274. [66] S. Westhues, J. Idel, J. Klankermayer, Sci. Adv. 2018, 4, eaat9669. [67] D. Limper: Hydrierung biogener Substrate mit maßgeschneiderten Ruthenium-Triphos- Katalysatoren: vom mechanistischen Verständnis bis zur Anwendung, RWTH Aachen University (Aachen), 2016. [68] a) J. Coetzee, D. L. Dodds, J. Klankermayer, S. Brosinski, W. Leitner, A. M. Slawin, D. J. Cole-Hamilton, Chemistry 2013, 19, 11039-11050; b) M. Meuresch: Maßgeschneiderte

161

9 Bibliography

Ruthenium-Triphos Katalysatoren für die homogenkatalysierte Hydrierung von Carbonsäurederivaten, RWTH Aachen University (Aachen), 2016. [69] a) S. Westhues: Development of organometallic catalysts for the homogeneously catalyzed hydrogenation of carboxylic acid derivatives and the depolymerization of polycondensation plastics, RWTH Aachen University (Aachen), 2020; b) A. Charkovskiy: Valorisation of lignocellulose using homogeneous catalysts : ligand design and application, RWTH Aachen University (Aachen), 2020. [70] a) A. Bryson, R. W. Matthews, Aust. J. Chem. 1963, 16, 401-410; b) C. H. Rochester, B. Rossall, Trans. Faraday Soc. 1969, 65, 1004-1013; c) CRC Handbook of Chemistry and Physics, 85th Edition, CRC Press, 2004; d) L. M. Tolbert, J. E. Haubrich, J. Am. Chem. Soc. 1990, 112, 8163-8165; e) F. Terrier, G. Moutiers, L. Xiao, E. Le Guével, F. Guir, J. Org. Chem. 1995, 60, 1748-1754; f) R. M. Tarkka, I.-H. Um, H.-Q. Xie, A.-P. Chatrouse, F. Terrier, S. Hoz, E. Buncel, J. Phys. Org. Chem. 1998, 11, 847-852; g) M. Forconi, N. H. Williams, Angew. Chem. Int. Ed. 2002, 41, 849-852; h) J. Mohanty, A. V. Sapre, R. D. Saini, T. Mukherjee, H. Pal, Res. Chem. Intermediat. 2005, 31, 47-61; i) H. Imura, T. Namai, K.-I. Ishimori, S. Hayashi, A. Ohashi, K. Ohashi, Bull. Chem. Soc. Jpn. 2005, 78, 2146-2151; j) J. Han, H. Lee, F.-M. Tao, J. Phys. Chem. A 2005, 109, 5186-5192; k) D. A. Kraut, P. A. Sigala, B. Pybus, C. W. Liu, D. Ringe, G. A. Petsko, D. Herschlag, PLoS Biol. 2006, 4, e99; l) A. U. Kulikov, M. N. Galat, J. Sep. Sci. 2009, 32, 1340-1350; m) P. Hanoian, P. A. Sigala, D. Herschlag, S. Hammes-Schiffer, Biochemistry 2010, 49, 10339- 10348; n) E. Kwiatkowska, I. Majerz, Chem. Phys. Lett. 2010, 484, 134-138; o) F. De Vleeschouwer, W. Yang, D. N. Beratan, P. Geerlings, F. De Proft, Phys. Chem. Chem. Phys. 2012, 14, 16002-16013; p) C. Gasbarri, G. Angelini, RSC Adv. 2014, 4, 17840- 17845. [71] C. Janiak, H.-J. Mayer, D. Gudat, R. Alsfasser, Riedel Moderne Anorganische Chemie, De Gruyter, 2012. [72] a) M. Fujio, R. T. McIver, R. W. Taft, J. Am. Chem. Soc. 1981, 103, 4017-4029; b) J. J. Urban, R. L. Vontersch, G. R. Famini, J. Org. Chem. 1994, 59, 5239-5245. [73] a) W. Li, J. H. Xie, H. Lin, Q. L. Zhou, Green Chem. 2012, 14, 2388-2390; b) W. R. Wright, R. Palkovits, ChemSusChem 2012, 5, 1657-1667; c) M. Selva, M. Gottardo, A. Perosa, ACS Sustain. Chem. Eng. 2012, 1, 180-189; d) J. M. Tukacs, R. V. Jones, F. Darvas, G. Dibó, G. Lezsák, L. T. Mika, RSC Adv. 2013, 3, 16283-16287; e) J. Deng, Y. Wang, T. Pan, Q. Xu, Q. X. Guo, Y. Fu, ChemSusChem 2013, 6, 1163-1167; f) J. M. Tukacs, M. Novák, G. Dibó, L. T. Mika, Catal. Sci. Technol. 2014, 4, 2908-2912; g) Z. Rong, Z. Sun, L. Wang, J. Lv, Y. Wang, Y. Wang, Catal. Lett. 2014, 144, 1766-1771; h) V. Fábos, L. T. Mika, I. T. Horváth, Organometallics 2014, 33, 181-187; i) C. Ortiz- Cervantes, M. Flores-Alamo, J. J. García, ACS Catal. 2015, 5, 1424-1431; j) J. J. Tan, J. L. Cui, X. J. Cui, T. S. Deng, X. Q. Li, Y. L. Zhu, Y. W. Li, ACS Catal. 2015, 5, 7379- 7384; k) J. Tan, J. Cui, T. Deng, X. Cui, G. Ding, Y. L. Zhu, Y. Li, ChemCatChem 2015, 7, 508-512; l) J. Zhang, J. Chen, Y. Guo, L. Chen, ACS Sustain. Chem. Eng. 2015, 3, 1708-1714; m) W. P. Wu, Y. J. Xu, S. W. Chang, J. Deng, Y. Fu, ChemCatChem 2016, 8, 3375-3380; n) J. Fu, D. Sheng, X. Lu, Catalysts 2016, 6, 6; o) S. Wang, H. Huang, V. Dorcet, T. Roisnel, C. Bruneau, C. Fischmeister, Organometallics 2017, 36, 3152-3162; p) S. Wang, V. Dorcet, T. Roisnel, C. Bruneau, C. Fischmeister, Organometallics 2017, 36, 708-713; q) S. Song, S. Yao, J. Cao, L. Di, G. Wu, N. Guan, L. Li, Appl. Catal., B 2017, 217, 115-124; r) Y. Liu, Z. Sun, C. Huang, T. Tu, Chem. Asian J. 2017, 12, 355- 360; s) C. A. M. R. van Slagmaat, S. M. A. De Wildeman, Eur. J. Inorg. Chem. 2018, 2018, 681-681. [74] R. V. Christian, H. D. Brown, R. M. Hixon, J. Am. Chem. Soc. 1947, 69, 1961-1963. [75] A. Primo, P. Concepción, A. Corma, Chem. Commun. 2011, 47, 3613-3615. [76] P. P. Upare, J. M. Lee, Y. K. Hwang, D. W. Hwang, J. H. Lee, S. B. Halligudi, J. S. Hwang, J. S. Chang, ChemSusChem 2011, 4, 1749-1752. 162

9 Bibliography

[77] L. Corbel-Demailly, B. K. Ly, D. P. Minh, B. Tapin, C. Especel, F. Epron, A. Cabiac, E. Guillon, M. Besson, C. Pinel, ChemSusChem 2013, 6, 2388-2395. [78] M. Li, G. Li, N. Li, A. Wang, W. Dong, X. Wang, Y. Cong, Chem. Commun. 2014, 50, 1414-1416. [79] a) T. Mizugaki, Y. Nagatsu, K. Togo, Z. Maeno, T. Mitsudome, K. Jitsukawa, K. Kaneda, Green Chem. 2015, 17, 5136-5139; b) T. Mizugaki, K. Togo, Z. Maeno, T. Mitsudome, K. Jitsukawa, K. Kaneda, ACS Sustain. Chem. Eng. 2016, 4, 682-685. [80] D. Sun, T. Saito, Y. Yamada, X. Chen, S. Sato, Appl. Catal., A 2017, 542, 289-295. [81] Y. Shao, K. Sun, Q. Li, Q. Liu, S. Zhang, Q. Liu, G. Hu, X. Hu, Green Chem. 2019, 21, 4499-4511. [82] H. Mehdi, V. Fábos, R. Tuba, A. Bodor, L. T. Mika, I. T. Horváth, Top. Catal. 2008, 48, 49-54. [83] J. Yuwen, S. Chakraborty, W. W. Brennessel, W. D. Jones, ACS Catal. 2017, 7, 3735- 3740. [84] P. Albrecht: Optimization of the Catalytic Hydrogenation of Levulinic Acid and Development of a Concept for Product Separation, RWTH Aachen University (Aachen), 2020. [85] a) T. Suzuki, K. Morita, M. Tsuchida, K. Hiroi, Org. Lett. 2002, 4, 2361-2363; b) H. B. Ji, T. Mizugaki, K. Ebitani, K. Kaneda, Tetrahedron Lett. 2002, 43, 7179-7183; c) A. Sølvhøj, R. Madsen, Organometallics 2011, 30, 6044-6048; d) A. Diaz-Rodríguez, I. Lavandera, S. Kanbak-Aksu, R. A. Sheldon, V. Gotor, V. Gotor-Fernandez, Adv. Synth. Catal. 2012, 354, 3405-3408; e) A. Bhatia, M. Kannan, S. Muthaiah, Synlett 2019, 30, 721-725. [86] T. Weigand: Mechanistic Investigations on the Transfer Hydrogenation with Ruthenium- Trimethylenemethane-Triphos catalysts, RWTH Aachen (Aachen), 2012. [87] a) V. P. Kurkov, Dihydrofuran Preparation 1976, US000003932468A; b) R. Fischer, C. Sigwart, H.-J. Weyer, Hydrogenation of Citric Acid 1995, US000005391771A; c) T. Abe, H. Nitobe, M. Takemoto, F. Tanaka, Process for Producing 3-Methyltetrahydrofuran 1996, EP000000727422A1; d) T. Abe, H. Nitobe, M. Takemoto, F. Tanaka, Process for Producing 3-Methyltetrahydrofuran 1997, US000005618953A; e) T. Abe, Y. Shima, M. Takemoto, Process for Preparing 3-Methyltetrahydrofuran 1997, EP000000773215A1; f) T. Abe, M. Takemoto, Process for Producing 3-Methyltetrahydrofuran 1998, EP000000869125A1; g) T. Abe, M. Takemoto, Process for Producing 3- Methyltetrahydrofuran 1999, US000005990324A; h) H. Iwasaki, Process for Producing 3- Methyltetrahydrofuran, 2000, EP000001057823A2; i) W. A. Beavers, Process for the Preparation of 3-Methyltetrahydrofuran 2000, US000006147233; j) L. E. Manzer, Manufacture of 3-Methyl-Tetrahydrofuran from Alpha-Methylene-Gamma-Butyrolactone in a Two Step Process, 2003, US000006664402B2; k) W. A. Beavers, A. V. Ignatchenko, Z. Liu, Process for the Preparation of 3-Methyltetrahydrofuran 2003, US000006521765; l) W. A. Beavers, I. A. V., Process for Preparing Alpha- and Beta- Methyl-gamma- butyrolactone and 3-Methyltetrahydrofuran 2004, US020040119052A1. [88] V. N. M. Rao, Process for Preparing Butyrolactones and Butanediols., 1988, EP000000276012A2. [89] R. Fischer, J. Frank, J. Henkelmann, F. Merger, T. Ruehl, H. Siegel, H.-J. Weyer, Verfahren zur Herstellung von 2-Methyl-1,4-butandiol und 3-Methyltetrahydrofuran 1992, DE000004231782A1. [90] a) H. B. Copelin, Process for the Production of 2-Methyl-1,4-butanediol 1975, US000003859369A; b) R. Fischer, J. Frank, J. Henkelmann, M. F., T. Ruehl, H. Siegel, H.-J. Weyer, Preparation of 2-Methyl-1,4-butanediol and 3-Methyltetrahydrofuran 1996, US000005536854A. [91] S. Li, X. Wang, X. Liu, G. Xu, S. Han, X. Mu, Catal. Commun. 2015, 61, 92-96. [92] O. A. Abdelrahman, D. S. Park, K. P. Vinter, C. S. Spanjers, L. M. Ren, H. J. Cho, K. C. Zhang, W. Fan, M. Tsapatsis, P. J. Dauenhauer, ACS Catal. 2017, 7, 1428-1431. 163

9 Bibliography

[93] a) M. W. Cook, D. N. Hanson, B. J. Alder, J. Chem. Phys. 1957, 26, 748-751; b) E. Brunner, J. Chem. Eng. Data 1985, 30, 269-273. [94] S.-I. Kawahara, S. Tsuzuki, T. Uchimaru, Chem.: Eur. J. 2005, 11, 4458-4464. [95] S. Tsuzuki, T. Uchimaru, M. Mikami, J. Phys. Chem. A 2006, 110, 2027-2033. [96] Y. Ishii, K. Osakada, T. Ikariya, M. Saburi, S. Yoshikawa, J. Org. Chem. 1986, 51, 2034- 2039. [97] a) S. P. Reade, M. F. Mahon, M. K. Whittlesey, J. Am. Chem. Soc. 2009, 131, 1847- 1861; b) L. M. Guard, A. E. W. Ledger, S. P. Reade, C. E. Ellul, M. F. Mahon, M. K. Whittlesey, J. Organomet. Chem. 2011, 696, 780-786; c) S. A. Macgregor, D. McKay, J. A. Panetier, M. K. Whittlesey, Dalton Trans. 2013, 42, 7386-7395; d) M. K. Cybulski, I. M. Riddlestone, M. F. Mahon, T. J. Woodman, M. K. Whittlesey, Dalton Trans. 2015, 44, 19597-19605; e) V. H. Mai, G. I. Nikonov, ACS Catal. 2016, 6, 7956-7961; f) M. K. Cybulski, J. E. Nicholls, J. P. Lowe, M. F. Mahon, M. K. Whittlesey, Organometallics 2017, 36, 2308-2316. [98] a) M. K. Whittlesey, R. N. Perutz, M. H. Moore, Chem. Commun. 1996, 787-788; b) S. P. Reade, A. L. Acton, M. F. Mahon, T. A. Martin, M. K. Whittlesey, Eur. J. Inorg. Chem. 2009, 2009, 1774-1785. [99] M. K. Whittlesey, R. N. Perutz, B. Greener, M. H. Moore, Chem. Commun. 1997, 187- 188. [100] N. A. Jasim, R. N. Perutz, S. J. Archibald, Dalton Trans. 2003, 2184-2187. [101] K. Matsumoto, Y. Sano, M. Kawano, H. Uemura, M. J., T. Sato, Bull. Chem. Soc. Jpn. 1997, 70, 1239-1244. [102] D. Limper: Mechanistische Untersuchungen zur Ruthenium-katalysierten Reduktion biogener Ester, RWTH Aachen University (Aachen, Germany), 2012. [103] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman, B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010, 29, 2176-2179. [104] R. Mede, S. Blohm, H. Görls, M. Westerhausen, Z. anorg. allg. Chem. 2016, 642, 508- 514. [105] V. I. Bakhmutov, E. V. Bakhmutova, N. V. Belkova, C. Bianchini, L. M. Epstein, D. Masi, M. Peruzzini, E. S. Shubina, E. V. Vorontsov, F. Zanobini, Can. J. Chem. 2001, 79, 479- 489. [106] a) J. M. Birchall, R. N. Haszeldine, J. Chem. Soc. (Resumed) 1959, 13-17; b) B. Gething, C. R. Patrick, J. C. Tatlow, J. Chem. Soc. (Resumed) 1962, 186-190.

164

Danksagung

An dieser Stelle möchte ich mich zunächst bei meinem Doktorvater Prof. Dr. Jürgen Klankermayer bedanken. Ich konnte meine Forschung unter Ihrer Betreuung unter hervorragenden Bedingungen durchführen. Die anregenden Diskussionen haben mir stets geholfen meine Forschung zu fokussieren und immer wieder neue Wege und Lösungsansätze zu finden.

Meinem Zweitberichter Prof. Dr. Andreas Jupke danke ich für seine ansteckende Begeisterung für die nachhaltige Chemie und die Umsetzung von Projekten, welche mich stets in meiner eigenen Motivation bestätigt hat.

Ich danke Dominik Limper dafür bei mir die Faszination für die homogene Katalyse geweckt zu haben. Dass unter Deiner Anleitung damals aus interessanter Theorie spannende praktische Anwendung wurde hat sicherlich mit dazu geführt, dass ich diesen Forschungsbereich auch für meine Promotion gewählt habe.

Meiner Laborkollegin Sandra Brosinski danke ich für eine tolle Arbeitsatmosphäre, viele lustige Momente, ihre Unterstützung im Labor, ihre unerschöpflich gute Laune, lebhafte und aufschlussreiche Diskussionen und eine stets hervorragende Musikauswahl. Auch bei Alexander Hell möchte ich mich für ein gemeinsames halbes Jahr Laborzeit in einer entspannten, lustigen und produktiven Atmosphäre bedanken.

Ines Bachmann-Remy, Sandra Brosinski und Meike Emondts danke ich für ihren motivierten Einsatz und ihre Geduld bei vielen Stunden NMR-Messungen, die auch mit vielen anregenden Diskussionen sowie oft auch mit lustigen Gesprächen verbunden waren. Ein großer Dank gilt der GC-Abteilung und damit Hannelore Eschmann, Elke Biener und Heike Fickers-Boltz für die Erstellung vieler Methoden und die Messung unzähliger Proben. Für die Messung meiner MS-Proben möchte ich mich bei Wolfgang Falter, Philipp Jürling-Will, Benjamin Schieweck und den Angestellten der instituts- externen Analytikeinrichtungen bedanken.

Ich danke den Arbeitsgruppen von Prof. Dr. Ulli Englert und Prof. Dr. Kari Rissanen für die Messung meiner SC-XRD-Proben. Khai-Nghi Truong und Markus Hölscher danke ich für das Lösen und Verfeinern der Kristallstrukturen.

Ich danke Ralf Thelen und der mechanischen Werkstatt des ITMC für die stets hervorragende und schnelle Hilfe bei sämtlichen großen und kleinen Projekten und Reparaturen. Thomas Müller und Stefan Aey danke ich ebenfalls für ihre Hilfe bei Bestellungen und Reparaturen. Für sämtliche IT- Unterstützung danke ich Günter Wirtz, Christian Westhues, Dennis Weidener und Daniel Geier. Ich

165

Danksagung

möchte mich ganz herzlich bei Sascha Schulze und dem gesamten Lager-Team für die zuverlässige Versorgung mit Labormaterial, Lösungsmitteln und die Bestellung von Chemikalien bedanken.

Ich danke Stefan Westhues für seine Unterstützung bei den von ihm optimierten Synthesen und der Bereitstellung von Katalysatormaterial. Julia Nowacki, Alexander Schmitz, Eren Temur und Michael Wittpahl danke ich für ihren Einsatz bei der Laborarbeit.

Ich danke den Partnern innerhalb des GreenSolRes Horizon 2020-BBI-PPP Projekts und des Exzellenzclusters Tailor-Made Fuels from Biomass für die gute Zusammenarbeit. Ich möchte mich vor allem bei Hybrid Catalysis für die Bereitstellung von Ligandenmaterial bedanken.

Den Mitgliedern und Partnern des interdisziplinären Netzwerks SusChemSys 2.0 und vor allem den Organisatoren Stefanie Gottuck, Tobias Klement und Klara Krämer-Klement danke ich für viele interessante Workshops und Exkursionen.

Ich bedanke mich bei Markus Hölscher, Giancarlo Franciò, Claudia Kohnen, Marion Sieprath und Margarete Rosen für die Unterstützung bei Bestellungen und administrativen Angelegenheiten sämtlicher Art.

Bei meinen Kollegen und ehemaligen ITMC-Kollegen möchte ich mich für die großartige Zeit im, um und auch fern des Instituts bedanken. Danke für viele tolle Erlebnisse mit euch. Ich danke Kassem Beydoun, Sandra Brosinski, Jasmine Idel und Niklas Westhues für das Gegenlesen meiner Arbeit.

Meiner Familie und meinen Freunden danke ich für die Unterstützung während meines Studiums und meiner Promotion.

166