Research Collection
Doctoral Thesis
Catalytic C-H activation of benzene by plantinum(II) a mechanistic study
Author(s): Gerdes, Gerd
Publication Date: 2004
Permanent Link: https://doi.org/10.3929/ethz-a-004830916
Rights / License: In Copyright - Non-Commercial Use Permitted
This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.
ETH Library DISS. ETH No. 15631
Catalytic C-H Activation of Benzene by Platinum(II):
A Mechanistic Study
A dissertation submitted to the SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZÜRICH for the degree of Doctor of Natural Sciences
presented by
Gerd Gerdes Diplom - Chemiker, Universität Heidelberg born July 3, 1973 citizen of Germany
accepted on the recommendation of
Prof Dr Peter Chen, examiner Prof Dr Antonio Togni, co-examiner
2004
Für Anja
Es gibt viel zu verlieren, Du kannst nur gewinnen Genug ist zuwenig - oder es wird so wie es war Stillstand ist der Tod, geh voran, bleibt alles anders Der 1. Stein fällt aus der Mauer Der Durchbruch ist nah
Herbert Grönemeyer
Danksagung
Mein besonderer Dank gilt Professor Dr. Peter Chen für das Überlassen des überaus spannenden Projekts, zahlreiche, inspirierende Diskussionen und wertvolle Anregungen. Zudem gewährte er mir stets den notwendigen wissenschaftlichen Freiraum, den ich sehr schätzte.
Des Weiteren danke ich Professor Dr. Dietmar A. Plattner sehr herzlich für gründliche Diskussionen, sorgfältige Durchsicht des Manuskripts und gute Freundschaft.
Bei Professor Dr. Antonio Togni bedanke ich mich herzlich für die Übernahme des Korreferats.
Diese Arbeit enthielte nur halb so viele Resultate, hätte ich nicht stets ein offenes Ohr bei meinem Laborkollegen Dr. Christian Adlhart gefunden. Seinen Anregungen folgten einige wichtige Experimente und Auswertungen.
Herrn Dr. Erich Meister bin ich zu Dank für Hilfe bei der Durchführung der UV/VIS Experimente sowie Diskussionen über korrekte Kinetiken verpflichtet. Genauso half mir Guido Grassi, die GC- MS Experimente durch zu führen. Weitere eingehende Diskussionen über die Geheimnisse der physikalischen Chemie verdanke ich Achim Sieben.
André Müller danke ich für einige zum Teil verzwickte Synthesen, praktische Anregungen im Labor und Gesellschaft in der Mensa.
Die Arbeit wäre gänzlich unmöglich gewesen, wenn nicht Rolf Dietiker meinen und alle anderen Computer am Laufen gehalten hätte. Für weiteres Korrekturlesen des Manuskripts bedanke ich mich bei Dr. Andreas Bach. Die gesamte Chen group hat mich mit einer freundlichen Atmosphäre unterstützt. Dafür danke ich Dr. Harold Baumann, Dr. Ruedi Hartmann, Dr. Daniel Lührs, Dr. Chrisian Hinderling, Dr. Derek Feichtinger, Dr. Thomas Gilbert, Prof. Dr. Ingo Fischer, Dr. Joëlle Vialon, Dr. Changkun Liu sowie den aktuellen Mitgliedern der Gruppe Eva Schön, Sanja Narancic, Dr. Loubna Hammad, Martin Jufer, Marc Bornand, Peter Zimmermann, Xueyi Chen, Dr. Xiangyang Zhang, Luca Cereghetti, Luca Castiglioni, Fabio di Lena, Claudio Gandolfo, Jonas Hostettler und nicht zuletzt Annette Ryter.
Den Kristallographen des Instituts, Dr. Bernd Schweizer und Paul Seiler, danke ich für das Messen und Lösen der Kristallstrukturen.
Mit ideenreichen Ratschlägen standen mir stets Prof. Dr. Bernhard Jaun als Leiter der NMR Abteilung sowie Dr. Walter Amrein als Leiter des MS Service zur Seite.
Einige MS und UV/VIS Experimente wären nicht möglich gewesen, hätte René Dreier mir nicht die entsprechende Geräte gefertigt.
Alle meine Arbeit verlöre ihren Sinn, wenn ich abends nicht meine wunderbare Frau Anja sehen könnte. Dank ihrer tatkräftigen Unterstützung und des ehrlichen Rückhalts konnte ich mich eingehend der Chemie widmen.
Dem Fonds der Chemischen Industrie danke ich für die zwei Jahre währende Finanzierung.
Parts of this work were published:
Gerd Gerdes and Peter Chen, "Cationic Pt(II) Carboxlyato Complexes are Competent in Catalytic C-H Activation under Mild Conditions" Organometallics 2004, 23, 3031-3036.
Gerd Gerdes and Peter Chen, "Comparative Gas-Phase and Solution-Phase Investigations of the Mechanism of C-H Activation by [(N-N)Pt(CH3)(L)]+" Organometallics 2003, 22, 2217-2225.
Oral Presentations and Poster Presentations:
2003 XIV. Massenspektrometrische Diskussionsveranstaltung, Vienna, A (OP). Heidelberg Forum of Molecular Catalysis, University of Heidelberg, D (PP). XVth FECHEM Conference on Organometallic Chemistry, Zürich, CH (PP). 4th International School of Organometallic Chemistry, Camerino, I (OP).
2002 International Conference on Reactive Intermediates and Reaction Mechanism, Ascona, CH (PP). SFC Eurochem, Toulouse, F (OP). Gordon Conference on Organometallic Chemistry, Newport, USA (PP). 16th IUPAC Conference on Physical Organic Chemistry, UC San Diego, USA (OP).
2001 Meeting of scholarship holders of the FCI, University of Tübingen, Tübingen, D (OP). Fall Meeting of the New Swiss Chemical Society, University of Zürich, Zürich, CH (PP). Heidelberg Forum of Molecular Catalysis, University of Heidelberg, Heidelberg, D (PP).
Contents
Contents
Abbreviations and Conventions IV
Summary V
Zusammenfassung VII
1 Introduction 1 1.1 Motivation 1 1.2 Platinum-mediated C-H Activation 1 1.3 Transition Metal Acetates in Hydrocarbon C-H Activation 6 1.4 Functionalization of Benzene and Methane through Heterobimetallic Catalysis 7 1.5 Electrospray Ionization Mass Spectrometry in Organometallic Chemistry 8 1.6 Gas-Phase Investigation of Pt-mediated C-H Activations 13 1.7 Density Functional Theory as a Tool in Organometallic Chemistry 14
2 The Solvent in Pt-mediated C-H Activations 16 2.1 Introduction 16 2.1.1 Syntheses 17 2.1.2 X-ray Crystallography 20 2.1.3 Quantum Chemical Methods 22 2.2 Gas-Phase Experiments 23 2.2.1 Intramolecular C-H, and C-F Activation in Tilset-like Platinum Complexes 24 2.2.2 Collision-Induced Dissociation of Isolated Cationic PtII Solvento Complexes 25 2.2.3 Coordination of Benzene to a Cationic PtII Center 29 2.2.4 Reaction of a cationic PtII Complex with Benzene in the Gas-Phase 30 2.2.5 Collision-Induced Dissociation of Cationic PtII Benzene Adducts with Argon, 2,2,2- Trifluoroethanol, and 1,1,1,2-Tetrafluoroethane 32 2.2.6 Scaling a Pirani Gauge for Diverse Collision Gases 33 2.2.7 Quantitative Comparison of PtII-mediated C-H Activation Reactions in the Gas Phase 34 2.2.8 Discussion 35 2.3 Solution-Phase Experiments 38 2.3.1 Benzene C-H Activation by PtII complexes in the Presence of 2,2,2-Trifluoroethanol 39 2.3.2 Benzene C-H Activation by PtII complexes in the Absence of 2,2,2-Trifluoroethanol 43 2.3.3 Activation Parameters for the C-H Activation of Benzene by a PtII complex 45
I Contents
2.3.4 Discussion 46 2.4 Conclusions 47
3 Scrambling in a Cationic PtII Complex in the Gas Phase 48 3.1 Deuterium Isotopic Scrambling in PtII Benzene Adducts in the Gas Phase 49 3.2 Installation of a Collision Cell in an Ion Trap 51 3.3 Theoretical Study of the Deuterium Isotopic Scrambling 52 3.4 Discussion 55 3.5 Conclusion 56
4 PtII-Acetato Complexes in Catalytic C-H Activation 57 4.1 Syntheses and Materials 57 4.2 Observation of Catalytic C-H Activation of Benzene by PtII 58
4.3 Acetic Acid-d4 as Deuterium Source 59 4.4 Kinetics of the Catalytic C-H Activation of Benzene by a PtII Complex 60 4.5 Gas-Phase Investigations of the Deuteration of Benzene catalyzed by PtII 62 4.6 Theoretical Study of the Intermediates of the Catalytic Deuteration 65 4.7 Discussion 66 4.8 Conclusion 70
5 Pt/Cu-catalyzed Functionalization of Benzene 71 5.1 Syntheses and X-Ray Crystallography 72 5.2 Variations on the Ullmann Reaction 75 5.3 Attempts to Fuse Catalytic Cycles 76 5.4 Gas-Phase Experiments with PtII/CuII Heterobimetallic Clusters 78 5.5 Discussion 82 5.6 Conclusion 83
6 Experimental Section 84 6.1 Instruments 84 6.2 Chemicals and Experimental Procedures 85 6.3 Ligand Syntheses. 86 6.4 Complex Syntheses. 87 6.5 Deuterated compounds and HBArF 91 6.6 Copper compounds 92 6.7 Crystallographic Data 93
II Contents
7 Appendix 95 7.1 Derivation of eq. 2.2 and Lambert-Beer's law 95 7.2 Principle of Linear Initial Rate 96
8 References 97
III Abbreviations and Conventions
Abbreviations and Conventions
API atmospheric pressure ionization (first region of all ESI instruments) CID collision-induced dissociation Da Dalton, equals unified atomic mass unit dau daughter scan mode DFT density functional theory ESI-MS electrospray ionization mass spectrometry eV electron Volt FT ICR-MS Fourier-transform ion cyclotron resonance mass spectrometry FWHM full width – half maximum GC-MS gas chromatography - mass spectrometry HiRes high resolution m/z mass-to-charge ratio NMR nuclear magnetic resonance r.t. room temperature rfd radio frequency voltage only daughter mode solv unspecified solvent molecule TFA trifluoroacetic acid TFE 2,2,2-trifluoroethanol THF tetrahydrofuran UV/Vis ultraviolet-visible spectroscopy VE valence electron
Arabic numerals denote compounds that were synthesized by the author or used in syntheses. Lower case letters denote compounds that were computed by DFT methods. Capital letters denote chemical compounds that are quoted from literature. Roman numerals denote proposed or identified intermediates in reactions and catalytic cycles.
For all compounds synthesized in the present work the lab journal page where the synthesis of the compound is described is given in the experimental section (GG n or André n).
IV Summary
Summary
Hydrocarbons from oil and natural gas are the main feedstocks in the chemical industry. Thus, a direct, catalytic transformation of alkanes and arenes to functionalized compounds via C-H activation is of considerable interest and remains a challenge to chemists. A seminal achievement in this area was the homogeneous PtII-catalyzed C-H activation of benzene developed by Garnett and Hodges in the late 1960s. This reaction holds some mechanistic questions that have been addressed in the present work. The work is related to the better-known Shilov chemistry, the C-H activation of saturated hydrocarbons with concomitant formation of aliphatic alcohols, in that comparable intermediates and transition states are involved. This reaction has initiated an ever-growing interest in C-H activation reactions. However, a detailed investigation of the reaction mechanism has always been hampered by the harsh reaction conditions. In this work, the intrinsic reactivity of isolated intermediates in C-H activation cycles has been addressed in the gas phase by ion-molecule reactions. The relevance of the results from these gas-phase experiments was confirmed by solution-phase studies on authentic catalysts, monitored mainly by UV/Vis spectroscopy and GC- MS measurements.
In previous studies modeling individual C-H activation steps, solvent effects were found to have a major influence on the course of the reaction. At first, the role of the solvent 2,2,2-trifluoroethanol (TFE) in the stoichiometric C-H activation of benzene by cationic [(diimine)Pt(Me)(solv)]+ complexes (solv = solvent molecule) was elucidated because it had been supposed that TFE accelerated the reaction. The first intermediate investigated in the gas phase was the benzene π- complex [(diimine)Pt(Me)(benzene)]+ formed upon exchange of the solvent molecule by benzene + which is in equilibrium with [(diimine)Pt(Ph)(CH4)] . One of the results was that CID (collision- induced dissociation) with TFE or 1,1,1,2-tetrafluoroethane as collision gas did neither alter product yield nor product distribution. From this observation it was concluded that TFE is not associated with the rate-determining transition state. This hypothesis was confirmed by reactions of two [(diimine)Pt(Me)(solv)]+ complexes with benzene in solution. The reactions were monitored by UV/Vis spectroscopy. Identical rate constants were found with and without TFE being present, showing that the C-H activation of benzene by cationic PtII methyl complexes is zero-order in TFE. In addition, kinetic isotope effects for individual reaction steps in the gas phase as well as activation parameters for the C-H activation of benzene in solution were determined. In gas-phase experiments with benzene-d6 incomplete deuterium isotopic scrambling around the phenyl and methyl ligands of the isomerizing benzene adduct was observed. In order to rule out possible isotope effects a + complementary experiment with [(diimine)Pt(CD3)] and benzene-d3 was conducted. Although previous work suggested that the scrambling process should go to statistical completion, a non- statistical distribution of the deuterium and hydrogen atoms was found for both cases. One possible explanation for this observation is that an agostic interaction of the ligand with the coordinatively unsaturated metal center formed in the course of the reaction terminates the scrambling process in the gas phase.
In a second project, the mechanism of the PtII-catalyzed H/D exchange between benzene and acetic acid-d4 was elucidated. A combination of ESI-MS experiments (electrospray ionization mass spectrometry) and kinetic analyses showed that the reaction proceeded with high turnover numbers + in acetic acid solution. The [(diimine)Pt(acetato)(CH3COOH)] complex was identified as the resting state of the catalytic cycle; dissociative ligand exchange of the reactants was found to be the turnover-limiting transition state. The rate law and rate constant of H/D exchange were determined. Finally, it was attempted to link the transient PtII phenyl complex of this catalytic cycle to a CuII-
V Summary
catalyzed coupling protocol with acetic acid in order to synthesize phenyl acetate. An ESI-MS study was conducted to identify factors that govern the reactivity of the PtII/CuII heterobimetallic system. Well-defined, cationic PtII/CuII heterobimetallic species were detected with different ligands at PtII and CuII. Upon CID, these cations formed a small number of fragments that could be structurally assigned. By systematic variations of the ligand spheres of the two metals it was possible to achieve the formation of a CuII compound containing both an acetato and a phenyl ligand, a prerequisite for reductive coupling to phenyl acetate.
VI Zusammenfassung
Zusammenfassung
Kohlenwasserstoffe aus Erdöl und Erdgas sind die wichtigsten Rohstoffe der chemischen Industrie. Deshalb ist die direkte, katalytische Überführung von Alkanen und Aromaten in funktionalisierte Verbindungen von großem Interesse und stellt eine bedeutende Herausforderung für Chemiker dar. Eine wegweisende Entdeckung in diesem Gebiet war die von Garnett und Hodges in den 1960er Jahren entwickelte katalytische C-H Aktivierung von Benzol. Diese Reaktion birgt einige mechanistische Fragen, die in der vorliegenden Arbeit aufgegriffen wurden. Diese stehen mit der besser bekannten Shilov Chemie - der C-H Aktivierung von gesättigten Kohlenwasserstoffen unter Bildung aliphatischer Alkohole - insofern in Zusammenhang, als ähnliche Zwischenprodukte und Übergangszustände im Verlauf der Reaktion auftreten. Seit der Entdeckung dieser Reaktion ist das Interesse an C-H Aktivierungsreaktionen ständig gewachsen. Indes sind detaillierte Untersuchungen des Reaktionsmechanismus aufgrund der drastischen Reaktionsbedingungen schwierig. In der vorliegenden Arbeit wurde die intrinsische Reaktivität von isolierten Zwischenstufen des C-H- Aktivierungszyklus mittels Ionen-Molekül-Reaktionen in der Gasphase untersucht. Die Bedeutung der Ergebnisse dieser Gasphasenexperimente wurde durch Untersuchungen mit realen, homogenen Katalysatoren in Lösung bestätigt. Dafür wurden hauptsächlich UV/VIS Spektroskopie und GC-MS Messungen eingesetzt.
In früheren Untersuchungen, in denen einzelne C-H-Aktivierungenschritte simuliert wurden, stellte man fest, dass das Lösungsmittel eine wichtigen Einfluss auf den Verlauf der Reaktion hatte. Zuerst wurde die Rolle von 2,2,2-Trifluorethanol (TFE) in der stöchiometrischen C-H Aktivierung von Benzol durch kationische [(Diimin)Pt(Me)(solv)]+-Komplexe (solv = Lösungsmittelmolekül) aufgeklärt. Als erstes Intermediat wurde der π-Benzolkomplex [(Diimin)Pt(Me)(Benzol)]+, der durch den Austausch des Lösungsmittelmoleküls mit Benzol erhalten wird und mit + [(Diimin)Pt(Ph)(CH4)] im Gleichgewicht vorliegt, in der Gasphase isoliert. Ein Ergebnis war, dass CID (stossinduzierte Fragmentierung) mit TFE oder 1,1,1,2-Tetrafluorethan weder Unterschiede in der Ausbeute noch in der Produktverteilung ergab. Daraus wurde der Schluss gezogen, dass TFE nicht am geschwindigkeitsbestimmenden Schritt beteiligt ist. Diese Hypothese wurde durch den Verlauf der Reaktion von zwei [(Diimin)Pt(Me)(solv)]+-Komplexen mit Benzol in Lösung bestätigt. Die Reaktionen wurden mit UV/VIS Spektroskopie verfolgt. Da in Gegewart bzw. Abwesenheit von TFE identische Geschwindigkeitskonstanten gefunden wurden, konnte bestätigt werden, dass die C-H Aktivierung von Benzol durch kationische Pt(II)-Methylkomplexe tatsächlich nullter Ordnung bezüglich TFE ist. Des Weiteren wurden kinetische Isotopeneffekte für einzelne Reaktionsschritte in der Gasphase sowie die Aktivierungsparameter der C-H Aktivierung von Benzol in Lösung bestimmt. In Gasphasenexperimenten mit [D6] Benzol wurde unvollständiger Deuteriumaustausch zwischen dem Methyl- und Phenylliganden des isomerisierenden Benzoladdukts beobachtet. Um denkbare Isotopeneffekte auszuschließen wurde ein + komplementäres Experiment mit [(Diimin)Pt(CD3)] und [D3] Benzol durchgeführt. Obwohl frühere Arbeiten erwarten ließen, dass die Deuteriumverteilung in den Liganden statistisch verlaufen sollte, wurde eine signifikant nicht-statistische Verteilung der Deuterium- und Wasserstoffatome beobachtet. Eine mögliche Erklärung dieser Beobachtung ist eine agostische Wechselwirkung des Liganden mit dem im Verlauf der Reaktion zeitweise koordinativ ungesättigten Zentralatom, die den Austauschprozess in der Gasphase effektiv beendet.
In einem zweiten Projekt wurde der Mechanismus des Pt(II)-katalysierten H/D Austausches zwischen Benzol und deuterierter Essigsäure aufgeklärt. Mit Hilfe einer Kombination von ESI-MS Experimenten (Elektrospray-Ionisations-Massenspektrometrie) und kinetischen Untersuchungen
VII Zusammenfassung
konnte gezeigt werden, dass die Reaktion in Essigsäure mit hohen Wechselzahlen ablief. Der + [(Diimin)Pt(Acetat)(CH3COOH)] -Komplex wurde als Ruhezustand des Katalysezyklus identifiziert. Als geschwindigkeitsbestimmender Schritt wurde der dissoziative Ligandenaustausch zwischen den Edukten identifiziert. Das Geschwindigkeitsgesetz sowie die Geschwindigkeitskonstante des H/D Austausches wurde bestimmt. Zuletzt wurde versucht, den in diesem Zyklus intermediär auftretende Pt(II)-Phenyl-Komplex in eine Cu(II)-katalysierte Kupplung mit Essigsäure einzubinden, um Phenylacetat herzustellen. Mittels einer ESI-MS-Untersuchung wurde versucht, diejenigen Faktoren zu bestimmen, welche die Reaktivität des heterobimetallischen Pt(II)/Cu(II) Systems beeinflussen. Massenspektrometrisch wurden wohldefinierte, kationische heterobimetallische Pt(II)/Cu(II) Spezies mit verschiedenen Liganden an Pt(II) und Cu(II) beobachtet. Diese bildeten nach CID wenige Fragmente, denen jeweils eine Struktur zugewiesen werden konnte. Durch systematische Variation der Liganden an Pt(II) und Cu(II) konnte die Bildung einer Cu(II) Verbindung, die sowohl einen Phenyl- als auch einen Acetatliganden enthielt, herbeigeführt werden. Dies ist eine der Grundvorrausetzungen für die reduktive Kupplung von Phenylacetat.
VIII 1.1 Motivation
1 Introduction
1.1 Motivation
For more than a century transition metals have been known to catalyze a variety of organic and inorganic reactions. The discovery and development of homogeneous transition metal catalysts has been fundamental, among others, to the progress in organic synthesis. Today, these catalysts are employed in reactions in all areas of organic chemistry, ranging from reactions on the laboratory bench, the synthesis of pharmaceutical fine chemicals and the production of bulk chemicals.
However, in the majority of transition metal-catalyzed reactions the role of the metal center is only partially understood. This is a result of the enormous structural and electronic diversity of organometallic compounds, which makes these reactions extremely sensitive to the slightest changes in any reaction parameter. In addition, elementary reaction steps, i.e. coordination, dissociation of ligands, structural rearrangements, bond formation and bond rupture, are usually extremely fast. This makes monitoring of elementary reaction steps difficult and precludes the isolation of many intermediates.
By means of electrospray ionization mass spectrometry (ESI-MS) it is possible to isolate and to identify highly reactive intermediates from reaction mixtures. Furthermore, stabilities and intrinsic reactivities of individual compounds can be investigated by gas-phase ion molecule reactions in absence of solvent- and counter ion effects, or ligand exchange equilibria. These studies are ideally supplemented by theoretical investigations based on density functional theory (DFT) calculations, which treat isolated molecules in the gas phase.
The goal of this work was to gain insight into the mechanism of stoichiometric, platinum-mediated C-H activations of benzene leading mainly to platinum-phenyl compounds by combining ESI-MS with other analytical techniques. From the results of these investigations, solution-phase experiments were designed in which benzene was activated catalytically. Finally, a route to functionalize benzene was examined. Therein the transient platinum-phenyl species was incorporated into a copper-catalyzed coupling cycle, similar to the Ullmann coupling.
1.2 Platinum-mediated C-H Activation
Thermodynamically viable reactions (∆G < 0) often proceed so slowly under ambient conditions that their synthetic use is strongly limited. Raising the reaction temperature enhances the rate of reaction. However, this causes significant energy costs and normally also leads to the formation of undesired by-products or to the decomposition of the starting material and products. These problems can be circumvented partly by using a catalyst. Its role is either to destabilize the reactants or to stabilize the transition state of a reaction or reaction step. This lowers the activation barrier of the reaction and leads to an increased rate.
Since the days of Döbereineri and Berzeliusii, the use of transition metals as catalyst is well established. In particular, homogeneous transition metal complexes are often employed as catalysts i Johann Wolfgang Döbereiner (*13.12.1780-†24.03.1849) demonstrated in 1823 the first catalysis by reacting hydrogen and oxygen under ambient conditions with the help of platinum (Döbereiner lighter). ii Jöns Jakob Berzelius (*20.08.1779-†07.08.1848) was the first to use the term "catalysis".
1 1 Introduction
in the synthesis of organic molecules. These catalysts usually work under milder conditions and provide higher selectivity than their heterogeneous equivalents. In addition, their reaction mechanisms are better understood, they can be investigated partly by standard, or modified analytical techniques and modifications on the ligand sphere are readily accomplished. On the other hand, homogeneous catalysts are typically more expensive and their separation from the reaction mixture is difficult. In order to justify its high costs, each catalyst molecule must catalyze the same reaction step exceedingly often before it decomposes (turn over number, TON).iii1
A very promising field in this regard is the selective functionalization of C-H bonds under mild conditions, which has been referred to as the "holy grail" in chemistry.2 (Scheme 1)
R H R X
Scheme 1. Functionalization of hydrocarbons (for X ≠ D).
Aliphatic hydrocarbons are readily available from oil and natural gas,3 and, together with aromatic, olefinic and acetylenic hydrocarbons form the feedstocks for chemical industry. But their transformation as shown in Scheme 1 is problematic because of the low reactivity of most C-H 4 bonds. For example, the pKa of methane is 48, while the C-H bond energy of the first C-H bond is 439 kJ/mol, and 473 kJ/mol for benzene.5 Thus, highly reactive species are needed to react them. Since super acids and free radicals allow only little control over product selectivity, an important amount of research is focused on coordinatively unsaturated transition metal complexes.6 ,7,8,9,10,11,12 Pt-mediated reactions are especially promising as they seem to allow control over chemo- and regioselectivity.13,14
In 1954 Leitch observed that heterogeneous Pt-metal is able to catalyze the exchange of deuterium 15 from D2O with protons from benzene. More than a decade later, Garnett and Hodges seized this II idea and showed that homogeneous Pt -salts dissolved in D2O/acetic acid displayed similar behavior towards benzene.16 Since these reactions only exchanged a hydrogen atom with one of its isotopes, the C-H activation does not lead to a functionalized molecule. Shilov transferred this catalysisiv to the less reactive methane,17 and developed a catalytic funtionalization/oxidation of methane using catalytic amounts of PtII, and stoichiometric amounts of PtIV.18
In 1983 Shilov proposed a mechanism for platinum-catalyzed alkane oxidation consisting of three basic transformations (Scheme 2):19 a) activation of the alkane by PtII to generate an alkylplatinumII intermediate, b) two-electron oxidation of the alkylplatinum(II) intermediate to generate an alkylplatinum(IV) species, and c) reductive elimination of RX (X = Cl or OH) to liberate the oxidized alkane and the PtII catalyst. In principle, this cycle applies to the oxidation of all hydrocarbons and although its general concept has been quickly accepted, many features of the individual steps have only recently been identified.
iii A very simplified example for the transformation CH4 → CH3OH assumes costs of 0.00046 $/mol(CH4), 0.0045 $/mol(MeOH) and 28500 $/kg(Pt). If conditions were found which require no other additives such as ligands, solvents or stabilizers and if oxygen from the air could be used without purification, break-even would be reached at 1.4 x 106 TON.1 iv In his paper Shilov mentions that the idea to use Pt(II) as a catalyst came from Garnett and Hodges who presented their work at a congress in Moscow in 1968.
2 1.2 Platinum-mediated C-H Activation
a)
II + Pt II + RH Pt + H R
ROH + H+ RCl Pt IV
c) b)
H2O
- II Cl Pt IV Pt R
Scheme 2. Shilov's cycle for the oxidation of alkanes reproduced from ref. 19 where no complete stoichiometry is given.
Important details of the catalytic cycle have remained elusive for a long time because it is difficult to observe elementary steps under the harsh reaction conditions (>100 °C). In addition, the C-H activation step is often accompanied by the formation of Pt(0). Therefore, model compounds were used that mimic individual reaction steps, allowing milder reaction conditions so that the reaction steps could be investigated with standard analytical techniques.6,20 The actual C-H activation step seemed to control most reaction parameters, such as catalyst activity, regio- and chemoselectivity of the reaction. But even with these model compounds, this step was not observable directly. That is why most of the knowledge about the C-H activation step comes from the microscopic reverse reaction, the protonation of a Pt-alkyl species that induces reductive coupling and elimination.
By the protonolysis of (tmeda)PtMe2, Bercaw and co-workers were able to demonstrate the intermediacy of Pt-hydrides and Pt-σ-alkane complexes in the C-H activation of methane (e.g. A – C).21
H H H LnPt R LnPt R C LnPt H H H R H H H η2 - C,H; A η2 - H,H; B η3 - H,H,H; C Subsequently, Pt-hydrides were synthesized through protonation of PtII-complexes containing tridentate ligands capable of fac-coordination, such as hydridotris(pyrazolyl)borate (Tp) or hydridotris(3,5-dimethylpyrazolyl)borate (Tp').22 Pt-hydrides could also be obtained from C-H activation of arenes,23 and saturated hydrocarbons.24 The resulting six-coordinate Pt-complexes are so stable that they were isolated and characterized by x-ray crystallography.25 The kinetic and thermodynamic stability of six coordinate PtIV compounds is general and contrasts strongly the stability of, for instance, PdIV complexes.26 Templeton and co-workers determined the activation barriers of the oxidative addition of the C-H bond in methane (25.7 kcal/mol),25 benzene (12.7 – 12.9 kcal/mol),27 toluene (13.6 kcal/mol) and xylene (14.2 kcal/mol) to Pt.28 Although a complete characterization of the σ-bonded alkane complexes remains elusive, the lower bound for the bond energy of Pt-methane was estimated to be 9 kcal/mol.25 This combined evidence ruled out a C-H activation mechanism by σ-bond metathesis. (Scheme 3)
3 1 Introduction
X R RH + LnPt X LnPt XH+ LnPt R H
X
RH + LnPt X LnPt H XH+ LnPt R R
Scheme 3. Oxidative addition and σ-bond metathesis.
The redox behavior of the intermediate PtII-species was investigated by Bercaw and co-worker by variation of the ligand of the model compounds.29 In their studies PtII could be oxidized reversibly only by chemical oxidation, contrary to electrochemical oxidation. (compare the role of Cu in Scheme 6)
From the stability of the six-coordinate hydrides it was evident that before reductive coupling and elimination of the hydrocarbon and the functionalized hydrocarbon, respectively, another intermediate had to be on the reaction coordinate. This conclusion was strongly supported by observations of several other groups. If deuterium was present in one of the ligands, e.g. from the protonolysis by DOTf, it was isotopically scrambled in all hydrocarbon ligands before loss of the ligands.25,30,31
TpPtMe2H, for instance, was heated in methanol-d4 to 70 °C for hours without elimination of methane, while at the same time deuterium isotopic scrambling was observed.32 The rates for 25 - deuterium isotopic scrambling for Tp'PtMe2H, and Tp'PtPh2H were determined to be k = 4.1 x 10 5 s-1 (339 K), and k = 53 s-1 (256 K), respectively.28 (Scheme 4) The observation of isotopic scrambling in Pt-complexes shows that the barrier of elimination of the hydrocarbon from the σ- complex is higher than that for C-H activation.
D H N CH N CH D Pt 3 Pt 2 N CH3 N CH3 N N N N B N B N H H
25 Scheme 4. Deuterium scrambling in Tp'PtMe2D.
The six-coordinate PtIV first has to lose one ligand before C-H reductive coupling occurs from the five-coordinate intermediate.22,33 A Si-analog was isolated first,34 followed by isolation of a true five coordinated PtIV species by Goldberg, Kaminsky, and Fekl.35 The final elimination of the hydrocarbon from the coordination sphere of the platinum, however, is subject of discussion. Evidence for dissociative loss of the ligand was found,25 along with the expected inverse deuterium isotope effect for the process.32 This result is supported by a vast number of ligand exchange,36,37,38,39,40 and isomerization41,42 experiments by Romeo et al. that provide evidence for the intermediacy of 14 valence electron (VE) species.43 Such Pt-compounds were characterized by
4 1.2 Platinum-mediated C-H Activation
means of x-ray crystallography,44,45,46 and NMR spectroscopy.47,48,49 However, associative loss of the hydrocarbon was also proposed,50 and is accepted as the usual pattern for ligand exchange on 8 51 d -ML4 fragments.
For the coordination of the hydrocarbon (step a, Scheme 2), on the other hand, an associative mechanism is generally accepted. Recently, this was confirmed by measurements of the volume of activation for C-H activation of benzene on a Pt-model compound.52 While benzene coordination was found to be the rate-determining step of the reaction,53 slight modifications on the diimine ligand turned the subsequent C-H activation into the rate-limiting step.54
A general inconvenience in the early studies of intermolecular C-H activation was the choice of a solvent that is not involved in the reaction. Bercaw et al. found that the solvents diethyl ether and thf were C-H activated to form Pt-Fischer-carbenes.55 In order to circumvent this problem they used pentafluoropyridine as a solvent.30 Another alternative to water as solvent in model studies of the Shilov system was introduced by Tilset and co-workers. They used 2,2,2-trifluorethanol (TFE) in the investigation of the C-H activation of benzene, and methane by [Pt(Me)(solv)]+-cations (A) ligated with diimine ligands.56 (Scheme 5) For the investigation of C-H activations the complex constituted a new class of model compounds, which they had used before in electrochemical investigations of Pt.57
R' + R' + R' R' R R R R N N CH3 , -CH4 Pt Pt
N OH2 F3C OH N OH2 R R R R R' R' R' R' A: R = H, R' = CF3; B: R = CH3, R' = H Scheme 5. Stoichiometric C-H activation of benzene by [(diimine)PtMe(solv)]+.
In a subsequent study they showed that the TFE had weaker coordinating abilities towards [(diimine)PtMe]+ than water.58 At the same time, a detailed study of the kinetics of the activation of + benzene by [(diimine)PtMe] (diimine = ArN=CMeCMe=NAr, Ar = 2,6-(CH3)2-C6H3) (B) confirmed that TFE for water exchange was a preequilibrium to the C-H activation.31 The reaction followed the rate law
ν = k[B][benzene]/[water] eq 1.1 with k = 1.98 x 10-5 s-1 at 298 K. The authors were puzzled by the unprecedented high rate of reaction. Later, they attributed the high rate of C-H activation of toluene, and xylene to a solvent assisted mechanism with TFE associated to the rate-determining transition state.59
5 1 Introduction
1.3 Transition Metal Acetates in Hydrocarbon C-H Activation
In the original reports on the activation of benzene the role of the solvent mixtures could not be determined unambiguously; neither in the heterogeneous version of this deuteration,15,60,61,62 nor in 16 its homogeneous analog. In the homogeneous reaction, benzene is mixed with D2O, K2PtCl4, DCl II to stabilize Pt , and with significant amounts of acetic acid-d4 to ensure homogeneity of the mixture. Upon heating to 80 to 110 °C deuteration of benzene occurs. Although detailed mechanistic and kinetic studies were performed under these reaction conditions, no particular role other than ensuring homogeneity was assigned to the acetic acid.63,64,65,66,67 This is surprising because an influence of the acetic acid concentration on the kinetics and on the mechanism of the reaction had been established.68 Furthermore, even weak acids such as water tend to react with alkyl and aryl transition metal complexes by protonation.69 At present, the Pt-protocol is sporadically used for the heterogeneous tritiation of aromatics,70,71,72 as well as for the homogeneous deuteration of a number of aromatics.73
Bercaw concluded in his 1998 review: " The solvent appears to play a crucial role in these catalytic systems. In all examples discussed, the solvent is highly polar and in many cases a strong acid such as … trifluoroacetic acid (TFA). … Clearly, the role of the solvent cannot be ignored in these reactions."10 Organic acids are not only used for the catalytic H/D-exchange, as described above, but they are also used in many very promising functionalizations of arenes and alkenes.74,75 In these reactions the precursor for the catalyst is usually a transition metal acetate, often Pd(OAc)2, which contains no metal carbon bond as most model systems do.
In 1985 a method to produce phenyl acetate and phenol from benzene and acetic acid by oxidation with oxygen at a heterogeneous Pd/Sb/Cr catalyst was described.76 Later, a homogeneous analog was presented by Sen and co-workers with the PdII acetate mediated trifluoroacetoxylation of adamantane and methane in TFA.77 They improved the reaction by replacing the acid by its anhydride, which was supposed to react with residual water in the reaction.78 In 1993 Periana and co-workers oxidized methane to methanol with HgII acetate in sulfuric acid,79 before they directed their attention towards PtII catalysis.80 Recently, Bergman et al. synthesized an Ir benzoate and posed the question of whether the complex would be able to catalyze the H/D exchange of various 81 aliphatic alcohols, ethers, and aromatic acids with D2O.
Fujiwara and co-workers used ligated PdII acetates as catalysts in the oxidation of benzene with oxygen in acetic acid to obtain phenol.82 Essential to the reaction was the presence of carbon monoxide, which they assumed to accept one oxygen atom from dioxygen. They gained evidence that Pd4 species were involved in the reaction. As a reasonable model they used an acetato bridged diphenyltripalladiumII complex C to elucidate the reaction mechanism.83 As a result they found that the oxygen of the phenol stemmed from dioxygen.
RR'S OOOOPh Pd Pd Pd C Ph O O O O SRR'
6 1.4 Functionalization of Benzene and Methane through Heterobimetallic Catalysis
1.4 Functionalization of Benzene and Methane through Heterobimetallic Catalysis
In addition to monometallic catalysis, reactions were found in which to different metal centers were involved in the overall transformation.84 Fujiwara and co-workers found that this concept applied also in reactions involving C-H activation, oxidation, and reductive elimination in one reaction. For the catalytic reaction of methane or benzene with carbon monoxide to form acetic acid and benzoic acid, respectively, they found that the best bimetallic pair is obtained from Pd(OAc)2 and Cu(OAc)2 85 when K2S2O8 is used as an oxidant. In the reactions performed in TFA they found evidence for Pd/Cu heterobimetallic species. However, they could not derive a general reaction mechanism. Other groups followed the concept of bimetallic catalysis for the functionalization of benzene. Sasson and co-workers oxidized benzene with oxygen to form biphenyl.86 They coupled the Pd catalysis to Zr, Co, and Mn cycles in acetic acid mixtures. For the same reaction Yamaji and Okamoto used Li, Be, Mg, Al, Pr, Zr, Mo, and Ti salts as co-catalysts.87
Almost 100 years earlier, in 1903, Ullmann discovered a related chemistry with the formation of diaryl amines from aryl chlorides and aryl amines in the presence of copper.88 A year later, he observed a similar reaction with the coupling of aryl chloride with aryl alcohols to form diaryl ethers.89 Since that time the copper promoted coupling of functionalized arenes with amines, alcohols, and sulfides experienced an enormous progression because these functionalities are part of many compounds in pharmaceutical and agrochemical industries.90
B(OH)2 B(OR)2
R R
Si(OMe)3 Sn(Me)3 Pb(OAc)3 Bi(Ph)2(OAc)2
R R R R
- + + - BF3 K I BF4
R R 2
Figure 1. Substrates for the copper-promoted coupling of arnenes with amines, alcohols, and sulfides.
Today, a huge number of functional groups on the phenyl residue, ranging from halides91,92,93 and boronic acids,94 esters95 or anhydrides to arylstannanes,96 arylsiloxanes97 and arylbismuth98,99 compounds (Figure 1), are used for the stoichiometric coupling. In addition ionic aryl compounds 100 101 like BF3-salts or iodonium salts are employed. The first catalytic version of the reaction was presented by Evans in 1999 who used 10 mol-% Cu(OAc)2 for the coupling of 4-tBu-phenol with 4- tolylboronic acid.102 The method was improved by Lam who used various co-oxidants, the best being dioxygen, to increase the yield.103 In 2000, Collman and Zhong introduced the catalytic N- 104 arylation of imidazole with boronic acids through catalysis by [{Cu(OH)(tmeda)}2]Cl2. Subsequently, they optimized the reaction parameters by varying ligand and counter ion of the catalyst, its amount, the boronic acids and the imidazoles, solvents, reaction time, oxidant, molecular sieves, and temperature.105 They found that under their reaction conditions, CuI was oxidized to CuII and that the actual catalyst was the same if they started from CuII. The catalytic cycle they proposed for the reaction is depicted in Scheme 6. Transmetallation of the phenyl group
7 1 Introduction
from boron to CuII (a) is followed by coordination of imidazole to CuII (b), oxidation of CuII to CuIII I II by O2 (c), reductive elimination of the product (d), and reoxidation of Cu to Cu (e).
0.5 O2 H2O
N c) N 2 Cu II 2 Cu III N N N X N X N N N N b) 2 N NH d)
N N 2 Cu II 2 Cu I X N X N
a) e) 0.5 O2 + H2O B(OH)3
N OH N B(OH)2 Cu II II Cu X2 N OH N
Scheme 6. Catalytic cycle for the coupling of imidazole with phenylboronic acid (reproduced from ref. 105).
In a detailed mechanistic study Combes and Finet showed that when organobismuth compounds were employed as aryl source, no radicals were involved in the CuII catalyzed reaction.106 Finally, Buchwald and co-workers used myristic acid (n-C13H27COOH) as an additive in Cu(OAc)2- catalyzed reaction of anilines with boronic acids.107
1.5 Electrospray Ionization Mass Spectrometry in Organometallic Chemistry
A fairly new method to conduct chemical analysis is electrospray ionization mass spectrometry (ESI-MS).108 It originated in the paint and coatings industry and was first applied to mass spectrometry by Fenn and co-workers who used the method for the mass spectrometry of biomolecules and polyethers.109 Its value, especially to biological chemistry was acknowledged with the Nobel Prize for John B. Fenn in 2002. In his Nobel-lecture he gave a very comprehensible description of the method as well as survey of the complicated ion formation process.110
An electrospray source transfers molecules from a dilute solution directly to the gas phase by a complicated process involving charged droplet formation, fission, and field desorption. At the same time electrospray produces ions only very rarely. It usually transfers existing ions from solution to
8 1.5 Electrospray Ionization Mass Spectrometry in Organometallic Chemistry
the gas phase. Only when neutral compounds with very low ionization energies (. 7 eV), e.g. N,N,N',N'-tetramethyl-1,4-phenylenediame and 2,3-benzanthracene,111 metalloporphyrins112 or ferrocenes,113 are electrosprayed, they are oxidized during the electrospray process. Preformed ions exist either, the usual case in orgonametallic chemistry, because the molecule or complex is inherently charged, or, as in biomolecules and polymers, because the molecule coordinates ions from solution to heteroatoms, e.g. H+, Na+, or K+. In large molecules this leads to multiply charged species and the spectra demand deconvolution to obtain the mass of a hypothetic singly charged species (Figure 2). The major benefit of electrospray ionization is its gentleness. It allows transferring large ions or even nonconvalent receptor-ligand complexes without decomposition into the gas phase. Recently, Aquilina and Robinson observed the Ca2+ bound 30mer of the human protein SAP (serum amyloid P component) with 761 kDa by means of ESI time-of-flight MS.114
Figure 2. Multiple charging as observed for the pentamer and decamer of SAP.114
Because an electrospray source transfers ions continuously to the gas phase it is best coupled to a continuously working mass spectrometer. From the time of the first successful applications of this technique, linear quadrupoles proved to be of great practical use.109 Today, usually a linear multipole (Q0) guides the ions to a first quadrupole mass filter (Q1), followed by a second ion guide (Q2) leading to a second quadrupole mass filter (Q3). Finally, an ion multiplier detects the ions. (Figure 3)
Figure 3. Block diagram of the modified Finnigan MAT TSQ 700 Tandem ESI-MS (side view).
Differantial pumping reduces the pressured from atmospheric pressure (API region, ESI source) to 10-3 mbar in Q1. Q0 and Q2 can be equipped with gas inlets, which allows carrying out ion-
9 1 Introduction
molecule reactions in the gas phase. The kinetic energy distribution of the ions gets narrower with exp(2n-2) where n is the number of a pair of multipole rods.115 Collisions with a thermalizing gas, usually in Q0, further equalize the kinetic energies of the ions. Q1 and Q3 analyze the reaction products or isolate single compounds (Q1). In newer triple quad instruments the ESI source may be set orthogonal to Q0 in order to reduce the neutral gas load (solvent and uncharged impurities) in the instrument. In addition, Q2 is bent by 90° (Figure 4). This makes the whole instrument smaller and further reduces the signal noise in Q3.
Figure 4. Photograph of a Finningan TSQ-Quantum Tandem ESI-MS (side view).
Recently the coupling of ESI sources to quadrupole ion traps (Paul-trap) has been realized (Figure 5).116 Again, multipoles guide the ions from the API region to the low-pressure region of the ion trap. These instruments have spread all over the field of biomolecular analytics as they provide easy handling, good signal intensities, and high mass resolution in combination with low costs compared to linear instruments. Furthermore, ESI-MS instruments are easily coupled to liquid chromatograghic techniques. However, the use of ion traps in organometallic chemistry is limited for a number of reasons. First, in an ion trap ions are stored under collisional conditions before being scanned out. Thus labile species decompose before analysis. Second, due to the complexity of the electronics and the packed assembly of commercial ion traps, modifications of the set-up are
10 1.5 Electrospray Ionization Mass Spectrometry in Organometallic Chemistry
difficult to accomplish, e.g. introduction of a device to allow the use of collision gases other than He. Third, an ion trap works only in the low-pressure regime (<7 x 10-4 mbar) with a very limited range of damping gases.
Figure 5. Photograph of the multipole and mass analyzer region of a Finnigan MAT LCQ ion trap (top view).
The use of ESI-MS instruments in biomolecular sciences is widespread,117,118,119 but rare in organometallic chemistry. Recent reviews describe gas-phase experiments performed with genuine catalyst complexes,120,121 as well as the potential of the method for organometallic chemistry.122 A few major traits of organometallic reactions and catalysis are challenging for conventional analytical techniques. Reactions at a transition metal center include a vast number of reaction channels that a substrate may undergo. This means a great number of intermediates, transition states, and reaction steps are involved in each possible reaction coordinate. In addition these steps
11 1 Introduction
sometimes proceed on a millisecond time scale.v The development of new reactions or the improvement of known reactions strongly depends on the knowledge of these factors.123
By means of ESI-MS some of the inherent difficulties may be circumvented and the corresponding questions may be addressed directly. ESI works well in the concentration range 0.001-0.1 mM that also corresponds to the concentration of many catalysts under realistic conditions (compare to footnote iii, p. 2). Therefore, by direct sampling from a running catalysis a number of intermediates that are present in solution in small concentrations and with short life times can be isolated and identified in Q1. Isolated, intact catalyst ions or ions of any other species present in solution in the gas phase provide the opportunity to investigate these compounds in the absence of unwanted ligand exchange reactions or solvent influence. Ion-molecule reactions can be performed by addition of a reagent gas in Q0 or Q2, to determine for instance the intrinsic reactivity towards different substrates, or the relative stabilities of a metal ligand bonds. If successive reactions are performed in Q0 and Q2, intramolecular reactions can be investigated by the analysis of the products of the second reaction, often collision-induced dissociation (CID).
CID-like-experiments marked the first utilization of ESI-MS as a tool for the analysis of ionic transition metal complexes. Chait and co-workers fragmented RuII bipyridyl and 1,10-phenanthrolin complexes by ion-source collisions to observe loss of undecomposed ligands such as acetonitrile, and bipyridine.124 One of the first nonanalytical applications of ESI-MS was presented by Posey and co-workers who performed ion molecule reactions between unsaturated FeII complexes and ligands inside the mass spectrometer.125 Peter Chen's group used these techniques in combination with other analytical tools to investigate the mechanism of Ir-mediated C-H activation,126,127 catalytic epoxidations with [Mn(salen)] complexes,128,129,130,131 Ziegler-Natta-polymerizations by Zr-132 and Pd-catalysts,133,134,135 catalytic Rh-136,137 and Ru-mediated hydrogenations,138 olefin metathesis by Grubbs-type catalysts,139,140,141,142,143,144 and aldehyde olefination by high-valent Re- compounds (Table 1).145,146 The present work focuses on the catalytic, Pt-mediated C-H activation of benzene, and its functionalization.147,148
v For example: Ru-catalyzed transfer hydrogenations attain a turn over frequency (TOF) of 156 s-1, zirconocenes produce up to 11400 kg of polypropylene per (mole x h) (TOF = 75 s-1). Each turn over step involves a complete catalytic cycle which is comprised of a number of elementary steps reaching a time scale of 10-5 s for a single reaction step.123
12 1.6 Gas-Phase Investigation of Pt-mediated C-H Activations
Table 1. Transition metal catalyses investigated by Chen and co-workers by means of ESI-MS. Reaction Catalyst Reference
RH RIr Ir 126, 127
O 128, 129, Mn 130, 131
Zr Zr 132 n Pd 133, 134, 135 n H H Rh 136, 137
O OH * Ru 138 R R' R R'
RR' R R' 139, 140, + + Ru 141, 142, R'' R''' R'' R''' 143, 144
O CR'R'' Re 145, 146 R H R H
H R Pt 147, 148
Recently, Pfaltz and Markert investigated the efficiency of chiral Pd catalysts in the kinetic resolution of allylic esters by means of ESI-MS. With the help of a Finnigan LCQ ion trap they screened a number of catalyst with homochiral substrates in parallel. From the ratio of the observed intensities for the two diastereomeric intermediates they predicted the selectivity of a particular catalyst.149
1.6 Gas-Phase Investigation of Pt-mediated C-H Activations
Armentrout's group carried out fundamental studies in the gas phase by using guided ion beam mass spectrometry to determine metal ion ligand bond dissociation energies (BDE).150 Although the method is limited to small complexes and ordinary ligands, highly valuable thermodynamic data can be determined from the experiments by elaborate treatment of the data. Recently, they published results of the investigation of the C-H activation of methane by Pt+.151 They showed that C-H activation occurred through oxidative addition of a C-H bond to Pt+. The BDE of Pt+-H and + Pt -CH3 were determined to be 271 ± 5, and 258 ± 8 kJ/mol, respectively.
13 1 Introduction
Comparable studies have been performed by Helmut Schwarz's group who have used sector-field mass spectrometry and Fourier transform-ion cyclotron resonance mass spectrometry (FT-ICR-MS) to investigate, among other reactions, the gas-phase C-H activation and functionalization of methane by Pt+.152 Pt+ served as a model for the heterogeneous Pt-metal in the Degussa and 153 Andrussow-process, which produces hydrogen cyanide from CH4, NH3, and water. As a result + they found that the first step in the reaction is C-H activation to PtCH2 followed by attack of ammonia to the Pt-methylene group. Subsequent to this step, two intermediates containing C-N bonds were postulated. More recently, they found that C-N coupling of CH4 and NH3 was possible + 154 + 155 with the help of PtmAun clusters (m + n ≤ 4) , and PtM clusters (M = Cu, Ag, Au) (Scheme 7). From their results they concluded that C-H activation occurred at the Pt+ center whereas N-H + activation and C-N coupling only occurred subsequently at the PtMCH2 heterobimetallic species. + Importantly, Pt2 clusters did not mediate the reaction.
+ + PtMCH2 + NH3 [Pt,M,C,H3,N] + H2
Scheme 7. C-N coupling at PtM+ clusters produced by laser desorption ionization and observed in an FT-ICR-MS.
1.7 Density Functional Theory as a Tool in Organometallic Chemistry
In the last two decades, density functional theory (DFT) has emerged as a practical and versatile tool to obtain information on molecular systems, which cannot be obtained from experiments. The popularity of DFT-based methods comes from its computational efficiency that allows acquiring, even for large molecules, thermochemical data, frequencies, ground state and transition state structures, along with spectroscopical predictions.156
DFT is derived from the Sommerfeld theory of the free electron gas and is designed to solve the Schrödinger equation
HΨ = EΨ eq 1.2
The general problem of these methods is the complex basic variable of an N-electron wave function Ψ(x1,x2,…,xn). This multibody problem was overcome in DFT by describing the energy as function of only one variable, the electron density ρ(r). The method is based on two theorems by Hohenberg, Kohn and Sham. The first theorem states that the energy of a molecule can be expressed as a functional of the electron density.157 The second states that the minimum value of the total energy functional is the ground state of the molecule only if the density ρ(r) is the exact ground state 158 density ρ0(r).
Caused by the continuous development of DFT methods and the progression in computer technology, a number of commercial programs have become available that can be used on desktop computers to perform DFT calculations, e.g. Spartan, Titan, and Gamess. One of the most popular is Gaussian, which was mainly used in this work.159,160 However, the computation time needed for calculations on full organometallic molecules is usually high and the molecules are therefore stripped down to model systems, e.g. in replacing the ligand Ar-N=CMeCMe=N-Ar by H-N=CHCH=N-H.58
14 1.7 Density Functional Theory as a Tool in Organometallic Chemistry
Due to the difficulties of investigating elementary steps in the Pt-mediated C-H activation (chapter 1.5), DFT-calculations have been frequently used in this field. Siegbahn and Crabtree modeled the solvent sphere in the Shilov reaction and found a methane-Pt bond energy of 10.5 kcal/mol.161 Other groups investigated the structures and activation barriers for isomerization of methane-Pt σ- complexes.162,163 A lot of work has been invested in the elucidation of the C-H activation step, which was found to proceed via oxidative addition at Pt-centers.164,165,166 Related to this are studies on the observation of deuterium isotopic scrambling.167 Goldberg and co-workers found that reductive elimination occurred from 5-coordinate intermediates for steric reasons.168 Periana investigated the thermodynamics of the Catalytica-process,80,169 while Ziegler examined the role of 170 SO3 as an oxidant.
DFT-calculations are particularly useful when properties of compounds are needed which are not consistently available or difficult to obtain by experimental methods. Armentrout and co-workers developed the Crunch-program for the evaluation of metal bond dissociation energies after CID– threshold data have been obtained. In addition to the experimental data the calculations require all normal modes of a molecule or complex of interest at the transition state, which have to be determined by computations.171 But also macroscopic properties of a molecule like its heat capacity may be obtained.
15 2 The Solvent in Pt-mediated C-H Activations
2 The Solvent in Pt-mediated C-H Activationsvi
2.1 Introduction
In 1999 Tilset, Ryan and Johansson published their results on the PtII-mediated C-H activation of benzene and methane (Scheme 5).56 For this purpose they used a PtII complex, which presented a new class of compounds in this research area. In addition, they where the first to use 2,2,2- trifluoroethanol (TFE) in reactions that mediate single reaction steps of the Shilov reaction. The precursor to the active species was produced from the dimethyl complex by protonolysis with HBF4 in TFE. The resulting cation was found to be present in equilibrium between the water adduct and a second species, which they identified as the TFE adduct. From the equilibrium between the two species the active species was obtained, which readily inserted into C-H bonds of benzene and methane, thereby producing methane and the phenyl or methyl complex, respectively. Questions remained, however, on the role of the solvent in the active species. As unambiguous results they found that TFE served as an inert and poorly coordinating solvent and that the PtII water adduct was the most abundant species in solution. More importantly, they observed C-H activation reactions "under the mildest conditions yet reported for such processes at cationic Pt complexes".
A year later they presented together with Bercaw's group a very detailed analysis of the kinetics of the C-H activation of benzene by a related PtII complex.31 They determined that when water was present in solution the water adduct of the cation was the most abundant species. They established a rate law (eq 1.1) and evoked the possibility of solvent assistance by TFE in the rate-determining step. Support for this hypothesis was gained from the negative entropy of activation (∆S≠ = - 16 cal/molK) for the overall reaction. Finally, an investigation on the C-H activation of toluene and p-xylene convinced them that TFE was associated with the rate-determining transition state.59
Ar N TFE H Pt N
H3C Ar
Figure 6. Five coordinate transition state in the C-H activation of arenes postulated by Tilset et al in ref. 59.
Given the potential economic significance of C-H activation reactions and the putative role of TFE in facilitating the reaction, i.e., solvent assistance, a mechanistic investigation to clarify this role is timely. Moreover, given that all of the complexes in the reaction are cationic, the reaction mechanism can be probed by gas-phase reactions of the complexes in a tandem mass spectrometer where the solvent molecule can be explicitly included or excluded from the reaction.
vi Parts of this chapter were published in ref. 147.
16 2.1 Introduction
2.1.1 Syntheses
In order to investigate the role of TFE in PtII-mediated C-H activations in the gas phase a number of compounds was synthesized. They are all related to the PtII complex A in Tilset's original report.56
The diimine ligands 1 – 4 were obtained in varying yields by the route of tom Dieck et al., which consists of a condensation of aromatic amines with 2,3-butanedione (Scheme 8).172 The Ligand 5 was previously prepared similarly by Christian Hinderling.173
R' R'' R
NH2 R' N O R R R HCOOH R + MeOH, r.t. N R' R' R' O R'' R R'' 1: R = R'' = H, R' = CF3; 20 % 2: R = R' = R'' = F; 6 % R' 3: R = F, R' = H, R'' = Br; 45 % 4: R = F, R' = H, R'' = F; 15 % 5: R = CH , R' = R'' = H 3 Scheme 8. Ligand synthesis by acid-catalyzed condensation.
When 6 was synthesized in the same way the yields were unacceptable (0.4 %) but sufficient for a first set of gas-phase experiments. The yield was improved by using toluene as a solvent, TFA as a catalyst and raising the reaction temperature to 110 °C (Scheme 9).
Cl
NH2 N O Cl Cl Cl TFA Cl + toluene, 110°C N O 31 %
Cl 6 Scheme 9. Improved ligand synthesis of 6.
It was found that electron withdrawing substituents on the aromatic ring reduced the condensation yield. The steric hindrance by o-substitution on the other hand seemed did not affect the conversion.
II Suitable Pt precursors were obtained in two steps from K2PtCl4 (Scheme 10). K2PtCl4 was reacted 174 with dimethyl sulfide to form [PtCl2(Me2S)2] (7) followed by reaction with methyl lithium to 175 176 give dimeric 8. For the synthesis of the deuterated analog 9, CD3Li was used, instead. Similarly, the phenyl substituted PtII dimer 10 was obtained by using phenyl lithium in the second step.177
17 2 The Solvent in Pt-mediated C-H Activations
Me 2 R' ' ' R' ' ' 8: R''' = CH 3 ; 69 % Me 2S R'''Lí S K2 PtCl2 [PtCl2 (Me 2 S)2 ] P t P t 9: R''' = CD 3 ; 70 % H O 2 ether R' ' ' S R' ' ' 10 : R''' = C H ; 60 % 7, 75 % Me 6 5 2 Scheme 10. PtII precursor synthesis.
The complexes 11 – 18 were obtained in varying yields by the reaction of the ligand with the respective PtII precursor in toluene at room temperature (Scheme 11).29
Ar N or + 8, 9 or 10 toluene, r.t. NN N Ar
R'' R' R' R R N Me N R''' Pt or Pt Me N R''' N R R 15; 99 % R' R' R''
11: R = R'' = H, R' = CF3; 95 % 12: R = CH3, R' = R'' = H, R''' = CH3; 77 % 13: R = R' = R'' = F, R''' = CH3; 8 % 14: R = F, R' = H, R'' = Br, R''' = CH3; 54 % 16: R = Cl, R' = R'' = H, R''' = CH3; 35 % 17: R = Cl, R' = R'' = H, R''' = CD3; 36 % 18: R = Cl, R' = R'' = H, R''' = C H ; 70 % 6 5 Scheme 11. General scheme for the synthesis of PtII complexes 11 – 18.
Again, electron withdrawing substituents on the ligand seemed to hamper the reaction. In order to increase the conversion the reactions were sometimes conducted under slight vacuum (200 mbar), thereby removing dimethyl sulfide from solution. The complexes are all strongly colored and exhibit a marked solvatochromism (Table 2). Although this behavior was not investigated in detail, the compounds were suitable for UV/Vis experiments. This behavior allowed for visual observation of the reaction progress, especially protonolysis (Scheme 12).
18 2.1 Introduction
Table 2. Solvatochromism of 16 at 298 K in various solvents.
Solvent εr color - - black plates
CH2Cl2 8.93 deep blue TFE 27.68 pink Benzene 2.28 green Cyclohexane 2.02 light green
CH3CN 36.64 purple Diethyl ether 4.27 green
CHCl3 4.81 light orange THF 7.52 green turquoise MeOH 33.00 purple
Complexes 11 – 18 dissolved satisfactorily in all solvents of Table 2. Treatment of these solutions with a strong acid resulted in the formation of the corresponding cationic PtII complexes 19 – 26, which in solution are stabilized by a solvent molecule. Protonolysis with acetic acid was not possible.
R'' + R' R' R + R N N R''' Me - - 11 - 18 + HX Pt X + Pt X solvent N solv N solv R R R' 21 R' R''
19: R = R'' = H, R' = CF 3 solv = water, 20: R = CH , R' = R'' = H, R''' = CH TFE, 3 3 acetonitrile, 22: R = R' = R'' = F, R''' = CH 3 benzene, 23: R = F, R' = H, R'' = Br, R''' = CH 3 acetic acid 24: R = Cl, R' = R'' = H, R''' = CH 3 - 25: R = Cl, R' = R'' = H, R''' = C 6H 5 X = [BF 4] - 26: R = Cl, R' = R'' = H, R''' = CD 3 [BArF]
Scheme 12. General scheme for the protonation of neutral PtII complexes. The resulting three coordinate 8 d -ML3 cations are ligated by a solvent molecule (solv).
19 2 The Solvent in Pt-mediated C-H Activations
2.1.2 X-ray Crystallographyvii
Crystals suitable for x-ray analysis of 2 and 6 were obtained by recrystallization from methanol. Ortep diagrams of their molecular structures are shown in Figure 7 and Figure 8.
Figure 7. Ortep diagram of 2. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: N3-C2 1.281(3), N3-C4 1.410(3), C4-C9 1.374(4), C4-C5 1.395(4). Selected bond and torsion angles [°]:N3-C2-C2' 115.2(2), C2-N3-C4 119.63(19), C2-N3-C4-C9 69.6(3), N3-C4-C9-F14 -4.7(4), N3-C4-C5-F10 3.7(3). The structure is C2 symmetric.
Figure 8. Ortep diagram of 6. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: C1-N3 1.272(2), N3-C4 1.410(2), C4-C5 1.392(3), C4-C9 1.396(3), Cl1-C5 1.736(2), Cl2-C9 1.739(2). Selected bond and torsion angles [°]: N3-C1-C1' 115.4(2), C1- N3-C4 121.41(18), C4-C5-Cl1 118.26(15), C4-C9-Cl2 118.46(15), C1-N3-C4-C5 -105.2(2), N3-C4-C5-Cl1 7.7(3), N3-C4-C9-Cl2 -6.2(3). The structure is C2 symmetric.
As the uncharged PtII compounds are perfectly air-stable, some were readily recrystallized under air. Crystals of 14 suitable for x-ray analysis were obtained from CH2Cl2/pentane, of 15 from toluene, of 16 from CH2Cl2/diethyl ether, and of 18 from CH2Cl2/pentane. Ortep diagrams of their molecular structures are shown in Figure 9 to Figure 12.
vii Details of x-ray crystallography are given in the experimental section in ch. 6.7.
20 2.1 Introduction
Figure 9. Ortep diagram of 14. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: Pt1-C10 2.01(2), Pt1-C11 1.93(3), Pt1-N2 2.085(15), Pt1-N3 2.087(14), N2-C12 1.31(2), N2-C16 1.459(19), C16-C17 1.34(3), N3-C13 1.31(2), N3-C22 1.44(2), C22-C27 1.42(3). Selected bond and torsion angles [°]: C10-Pt1-C11 88.1(9), N2-Pt1-N3 75.3(5), N2-C12-C13 113.8(15), C12-N2-C16 119.5(14), C12-N2-C16-C17 -96(2), C10-Pt1-N2-C12 -179.6(17), N3-C13-C12 115.0(15), C13-N3-C22 120.3(14), C13-N3-C22-C27 -93(2), C11-Pt1-N3-C13 -179.4(15).
Figure 10. Ortep diagram of 15. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: Pt1-C4 2.33(8), Pt1-C5 2.04(9), Pt1-N2 2.09(6), Pt1-N3 2.18(7). Selected bond and torsion angles [°]: N2-Pt1-N3 78(3), C4-Pt1-C5 89(3), C4-Pt1-N2-C17 -167(6), C5-Pt1-N3-C16 166(7).
Figure 11. Ortep diagram of 16. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: Pt1-C5 2.007(7), Pt1-C5' 2.007(7), Pt1-N2 2.076(5), Pt1-N2' 2.076(5), Pt1-Cl1 3.928, Pt1-Cl2 4.067, N2-C3 1.284(9), N2-C6 1.430(7), C6-C7 1.391(9), C6-C11 1.395(8), Cl1-C7 1.725(6), Cl2-C11 1.732(6). Selected bond and torsion angles [°]: C5-Pt1-C5' 87.9(5), N2-Pt1-N2' 74.9(3), C3'-C3-N2 113.8(3), C3-N2-C6 119.7(5), C7-C6-N2 120.5(5), C6-C7-Cl1 119.3(5), C6-C11-Cl2 118.0(5), C5-Pt1-N2-C3 -179.2(5), C3-N2-C6-C7 100.8(7), N2-C6-C7-Cl1 -6.3(9), N2-C6-C11-Cl2 4.1(9). The structure is C2 symmetric.
21 2 The Solvent in Pt-mediated C-H Activations
Figure 12. Ortep diagram of 18. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: Pt1-C24 2.003(3), Pt1-C30 2.004(4), Pt1-N2 2.103(3), Pt1-N5 2.095(3), Pt1-Cl14 3.943, Pt1-Cl15 4.097, Pt1-Cl22 3.916, Pt1-Cl23 4.097, N2-C3 1.292(4), N2-C8 1.419(4), C8-C9 1.369(7), C8-C13 1.411(7), C9-Cl15 1.735(6), C13-Cl14 1.722(6), N5-C16 1.432(4), C16-C17 1.389(6), C16-C21 1.386(5), C17-Cl23 1.735(4), C21-Cl22 1.725(4). Selected bond and torsion angles [°]: C24-Pt1-C30 90.27(14), N2-Pt1-N8 75.00(11), N2-C3-C4 114.6(3), C3-N2-C8 119.1(3), C8-C13-Cl14 119.0(3), C8-C9-Cl15 118.9(3), C30-Pt1-N2-C3 -176.9(3), C3-N2-C8-C9 92.3(5), N2-C8-C9-Cl15 -2.5(5), N2-C8-C13-Cl14 3.4(5), N5-C4-C3 115.0(3), C4-N5-C16 119.9(3), C16-C21-Cl22 118.6(3), C16-C17-Cl23 119.1(3), C24-Pt1-N5-C4 178.8, C4-N5-C16-C21 -98.7(4), N5-C16-C17-Cl23, -3.1(5), N5-C16-C21-Cl22 4.9(5).
Comparison of the various structures and the corresponding trends in coordination are beyond the scope of this work. However, the structure data provided a reference for quantum chemical calculations.
2.1.3 Quantum Chemical Methods
When the interpretation of experimental data was ambiguous use of quantum chemical calculations was made. For this purpose the Gaussian160, 159 package was used. For the calculation of physical data of organic molecules DFT was used with Becke's three parameter hybride exchange (B3)178 and the Lee-Yang-Parr (LYP)179 correlation functional. A scaling factor of 0.9614 was suggested for the fundamental vibrations of the molecules calculated with the 6-31* basis sets.180
In order to explore the structural manifold of 14 VE and 16 VE PtII and PtIV compounds the Perdue- Wang (PW91)181 correlation functional was used instead of LYP. Systematic surveys of the applications of various DFT functionals to calculations of cisplatin and carboplatin revealed that B3PW91 generally gives good structural correlation of charged and uncharged platinum complexes with their computed structures.182 In addition it was successfully employed in the investigation of an 14 VE rhodium complex.183 The Stuttgart-Dresden basis sets184 were used for platinum and atoms directly connected to the metal while for the remainder of atoms the 6-31G** basis sets185 were used. The results of a comparison of the structural data obtained from x-ray crystallography of 6, 16, and 18 with their computed molecular structures are given in Table 3.
22 2.2 Gas-Phase Experiments
Table 3. Mean deviation ∆mean and maximum deviation ∆max of computed structural parameters from parameters obtained by x-ray analysis.
Parameter ∆mean 6 ∆max 6 ∆mean 16 ∆max 16 ∆mean 18 ∆max 18 Bond distances [Å] 0.006 0.01 0.032 0.04 0.04 0.05 Bond angles [°] 0.60 2 1.31 3 1.32 2 Dihedral angles [°] 0.95 2 2.77 5 3.4 5
Calculations on intermediates of reaction coordinates were done similarly, however replacing the diimine ligand 6 with H-N=CHCH=N-H. The zero-point vibrational energies (ZPVEs) were scaled by 0.9772 as suggested for B3PW91/6-31G*.180
2.2 Gas-Phase Experiments
Tilset and co-workers showed that when 11 is protonated in TFE with a strong acid, the resulting cation [19]+ was ligated by residual water from the solvent or with TFE resulting in the formation of + + - [19(water)] and [19(TFE)] . When weakly coordinating anions such as [BF4] were employed the cation readily inserted into aromatic and aliphatic C-H bonds accompanied by the release of 56 + methane. Similarly, 12 was protonated by HBF4 to give [20(solv)] , which reacts smoothly with aromatic C-H bonds.31 The reactions were quenched by addition of acetonitrile. This observation suggests that acetonitrile coordinates to PtII center thereby replacing the more labile water and TFE. In order to identify the solvent adducts of [19(acetonitrile)]+ and [20(acetonitrile)]+ in the gas phase, -5 11 was dissolved in a 2:1 mixture of CH2Cl2/CH3CN (10 M, purple) and reacted with an excess of HBF4 (54 % in diethyl ether) resulting in a yellow, clear solution. This solution was introduced into a Finnigan MAT TSQ 700 through a regular electrospray source. + F3C
CF3
N Pt N NCCH3
CF3
F3C Intensity [19(acetonitril)]+ m/z = 759
650 700 750 800 m/z Figure 13. ESI-MS spectrum of the cationic acetonitrile adduct of [19(acetonitrile]+ with the corresponding calculated isotope pattern for C23H18F12N3Pt.
23 2 The Solvent in Pt-mediated C-H Activations
The spectrum resulting from this experiment is shown in Figure 13. In addition to the observed mass (m/z = 759) the distinct isotope pattern for C23H18F12N3Pt served to identify the acetonitrile adduct of [19(acetonitrile)]+. Analogous results were obtained for [20(acetonitrile)]+. Replacing CH2Cl2 bei TFE did not affect the results.
2.2.1 Intramolecular C-H, and C-F Activation in Tilset-like Platinum Complexes
When experiments were conducted in the absence of acetonitrile, coordination of water from residual moisture in TFE or from the atmospheric pressure region (API) of the instrument was observed. As the water adducts of [19(water)]+ and [20(water)]+ were found to be the most abundant species in solution in the C-H activation reaction of benzene, the adduct of [19]+ was tested in the reaction with benzene in the gas phase.56 This was accomplished by spraying a TFE/water solution of [19(water)]+ into the mass spectrometer and using benzene as a collision gas in the second octopole region (Q2). Disappointingly, only loss of water together with methane (m/z 702) was observed. Not only was C-H activation of benzene not observed but also, more importantly, intramolecular C-H activation occurred. The intermolecular ligand C-H activation reaction was considered as a side reaction in solution-phase experiments.56 Furthermore, intramolecular C-H activation of a ligand attached to an unsaturated transition metal center was extensively studied in the gas phase by Hinderling173 and Kim186 and was previously observed in gas-phase investigations on PdII-catalyzed polymerization reactions.187
The results from gas-phase investigations on the water adduct of [20]+ were even more discouraging. When a 10-5 M TFE solution of [20]+ is electrosprayed in a Finnigan TSQ Quantum mass spectrometer a number of signals is observed in the spectrum (Figure 14).
+ + + +
N N N N OH Pt Pt Pt Pt N OH2 TFE N N N
+ [20(water)] [20(TFE)]+ m/z = 486 m/z = 503 m/z = 519 m/z = 601
Intensity chloride containing species
450 500 550 600 m/z Figure 14. ESI-MS spectrum of cationic [20(water)]+ sprayed from TFE/water displaying intramolecular C-H activation.
24 2.2 Gas-Phase Experiments
Besides the adducts of water and TFE to the cation (m/z 519 and 601), the product of an intramolecular C-H activation (m/z 486), the replacement of the methyl group by a hydroxy group (m/z 503) and a chloride containing species were detected.
It was concluded that neither [19]+ nor [20]+ were suitable for further gas-phase studies. In order to reduce the flexibility of the ligand backbone and hence prevent intramolecular C-H activation phenanthroline was used as a ligand. Nevertheless, when 15 was protonated with HBF4, the resulting cation [21(solv)]+ lost methane in the gas phase.188 Second, o-fluoro substitution of the diimine ligand was tested in order to stabilize the cations in the gas phase. Cationic [22]+ and [23]+ were obtained by protonation from 13 and 14 and were tested in similar reactions.viii Unfortunately, upon reaction in the gas phase, all compounds displayed an unidentifiable loss of a fragment with mass 18. Water (18 Da) loss was ruled out as it was not present in most of the investigated compounds. That is why the attention was drawn towards stabilization by o-chloro substitution.
2.2.2 Collision-Induced Dissociation of Isolated Cationic PtII Solvento Complexes
+ Protonolysis of 16 with HBF4 (54 % in diethyl ether) yielded the cation [24] (m/z = 583), which, even without a stabilizing ligand like water or acetonitrile, gave very clean ESI-MS spectra (Figure 15).
+ +
Cl Cl Cl Cl N N Pt Pt OH N N 2 Cl Cl Cl Cl
+ [23]+ [23(water)] m/z = 583 m/z = 601 Intensity
m/z 300 400 500 600 700 800 900 Figure 15. ESI-MS spectrum of [24]+ electrosprayed from TFE in a Finnigan TSQ Quantum mass spectrometer. Inset: calculated isotope pattern for C17H15Cl4N2Pt in comparison with the observed isotope pattern.
viii Modifications on the diimine ligand are known to alter the rate of reaction within one order of magnitude. This relatively moderate change in reaction rate stems from the fact that the aryl rings on the ligands of 11 – 14 and 16 are twisted out-of-plane, so that the electronic influence of substituents on the rings at the metal center comes only through inductive effects. Furthermore, changes in the ligand shift the rate-determining step of the reaction. Nevertheless, the general reaction, C-H activation, remains intact.54
25 2 The Solvent in Pt-mediated C-H Activations
In order to get an indication of the energies involved in the subsequent ion-molecule reactions in the gas phase, the effective potential of [24]+ in the second multipole region (Q2) under standard conditions was determined. This was done by scanning the offset potential of Q2 while keeping the potential of Q0 constant (-3 V).189 In a first experiment m/z = 583 was selected in Q1 of a TSQ 700 instrument with the dau-mode and a width of 1 Da in the absence of a thermalizing gas in Q0. The dau-mode selects ions within a defined m/z range and permits the monitoring of fragments (daughter ions) after CID. Figure 16a shows the measured intensities of m/z = 583 when scanning Q2 from -15 to +15 V (laboratory frame) in 0.5 V intervals. This corresponds to a distribution of the kinetic energy of -12 to 18 eV around the mean kinetic energy in the center-of-mass frame.ix The resulting sigmoidal curve was differentiated numerically and fitted with a Gaussian curve to yield the effective potential of –0.89 V for [24]+ as well as its energy distribution in the laboratory frame (Figure 16b). a) b) 1.0
0.8 Intensity 0.6 ed ured Intensity ured Normaliz Meas 0.4
0.2
10 5 0 -5 -15 -10 -5 0 5 10 15 Offset of Q2 [V] Offset Q2 [V] Figure 16. a) Retarding curve of [24]+ by selection with dau-mode without thermalizing gas in Q0. b) Effective potential (-0.89 V, laboratory frame) and energy distribution (FWHM = 6.25 V) of [24]+ selected in the dau-mode without thermalization.
The difference of the effective potential of [24]+ in Q2 from its original potential in Q0 (-3 V) is influenced by the mass of the ion, by the resolution in Q1 and the various lens potentials around Q1. The smallest possible influence of the various potentials on the traveling ions is when Q1 is operated in the rfd-mode. In this mode Q1 is run in a RF-only mode and thereby acts as a high-pass filter, discriminating all ions under a selected m/z. But this mode is only useful if no ions with m/z higher than that of interest, e.g. solvent adducts, are present because they would also be detected in Q3. Additionally, a thermalizing gas in Q0 was supposed to reduce the width of the energy distribution. In a second experiment [24]+ was selected in Q1 with the rfd-mode and 0.01 torr of He were introduced in Q0. The resulting retarding data were treated similarly yielding an effective potential of –3.03 V (laboratory frame) with a FWHM of 1.49 V (Figure 17a).
ix The first multipole (Q0) is at –3V relative to the skimmer which is at ground.
26 2.2 Gas-Phase Experiments
a) 1.0 b) 1.0
0.8 0.8 tensity 0.6 0.6 lized Intensity lized In
Norma 0.4 Norma 0.4
0.2 0.2
0.0 0.0 -6 -4 -2 0 2 0102030 Offset Q2 [V] E [kJ/mol] cm Figure 17. a) Effective potential (-3.03 V, laboratory frame) and energy distribution (FWHM = 1.49 V) of + [24] selected in the rfd-mode with 0.01 torr He in Q0. b) Calculated maximum energy (Ecm) distribution of collisions of [24]+ with Ar.
The effective potential of [24]+ in the laboratory frame is transferred into a center-of-mass frame (Ecm) with the help of eq. 2.1 if perfect spheres and completely inelastic collisions are assumed: MN Ecm = • Elab eq. 2.1 MN + MI where MN is the mass of the resting neutral collision gas, MI is the mass of the ion and Elab is the kinetic energy in the laboratory frame of the ion.190 From eq. 2.1 the distribution of the maximum + energy Ecm,max available in a collision of [24] with Ar was calculated from its effective potential (Figure 17b). A mean energy +Ecm,max, of 18.8 kJ/mol for a single collision was thus determined. The mean energy +Ecoll, transferred in a random collision is 0.4 · +Ecm,max, due to the angular distribution of all collisions.190 Because in most experiments the ion of interest was accompanied by ions with higher and lower m/z in Q0 the dau-mode was chosen for all daughter ion experiments. From the energy distribution of [24]+ selected in the rfd-mode the collision energies of the present experiments can be estimated to be on the order of Ecoll = 10-50 kJ/mol, depending on the offset of Q2. Repetition of the experiments without a thermalizing gas or with Ar replacing He yielded within experimental error identical results.
+ - NMR experiments of [24(solv)] [BF4] in TFE-d3 showed two structures in equilibrium with each other. Tilset and co-workers attributed this equilibrium to the exchange of water in [24(solv)]+ for TFE.56 The gas-phase molecular structure of cationic [24]+ was investigated by DFT calculations. 8 II The resulting structure displays a T-shaped d -ML3 Pt fragment with an agostic stabilization by one of the o-chlorine substituents.191
27 2 The Solvent in Pt-mediated C-H Activations
Figure 18. Computed molecular structure of [24]+. Selected bond length [Å]: Pt1-C6 2.042, Pt1-N2 1.992, Pt1-N5 2.100, Pt1-Cl9 2.367, Pt1-Cl11 5.129, N2-C3 1.313, N2-C7 1.406, C7-C8 1.418, C7-C10 1.417, C7- Cl9 1.789, C10-Cl11 1.738. Selected bond and torsion angles [°]: N2-Pt1-C6 174.8, N2-Pt1-N5 79.3, C4-C3- N2 113.6, C3-N2-C7 130.2, N2-C7-C8 118.4, C7-C8-Cl9 120.5, C7-C10-Cl11 121.5, C6-Pt1-N5-C4 163.1, C3-N2-C7-C8 149.1, N2-C7-C8-Cl9 5.8, N2-C7-C10-Cl11 7.9.
Comparison of the computed structure of [24]+ with the x-ray structure of 16 reveals that only slight distortions, primarily a twist of the aromatic ring on the nitrogen cis to the empty coordination site, are necessary to compensate for the removal of one methyl group.
Finally, the stability of the PtII solvento species [24(solv)]+ was compared by means of CID- experiments in the gas phase.
1.0
0.8
0.6
0.4 Normalized Intensity Normalized
0.2 L = TFE
L = H2O
L = CH3CN 0.0 10 5 0 -5 -10-15-20 Offset of Q2 [V] Figure 19. Threshold curves of measured intensities of [24]+ in CID-experiments of [24(solv)]+ (solv = CH3CN, H2O and TFE) with benzene (Offset in laboratory frame).
Hence, [24]+ was electrosprayed from acetonitrile/TFE and water/TFE mixtures to produce the + + respective adducts [24(solv)] with solv = CH3CN and H2O. The TFE adduct [24(TFE)] was produced by spraying a TFE solution of [24]+ and adding 23 mtorr of TFE in Q0. In Q2 the mass-
28 2.2 Gas-Phase Experiments
selected ions were subsequently collided with benzene and the intensity of [24]+ was measured as a function of the offset of Q2 (Figure 19).
The gas phase results parallel the observation of the solution-phase reactivities perfectly. Experimentally a small bond enthalpy difference of 16.7 kJ/mol between the water and TFE adducts of [20]+ was found with the water adduct being the energetically favored species.31 Correspondingly, the threshold data for TFE and water loss from [24(solv)]+ are closely together with TFE being lost more easily (squares and circles in Figure 19). Furthermore, C-H activation reactions were found to be quenched by the presence of acetonitrile, which is reflected by the + 192 marginal loss of CH3CN from [24(solv)] (triangles). Related threshold measurements on a + 193 [CpIr(PMe3)Me] showed that quantitative BDEs were delicate to obtain with good reliability. Therefore, a quantitative analysis of the data was not encountered.
2.2.3 Coordination of Benzene to a Cationic PtII Center
After having established conditions for reproducible experiments, the actual C-H activation of benzene was investing in the gas phase. In a first set of experiments the forward reaction was studied, i.e. [24]+ was reacted in the gas phase with benzene to produce the phenyl-complex [25]+. Although, as no solvent molecules are present, this is not the best model for the complete solution- phase reaction, it was tantalizing to observe the reaction in the gas phase. In addition the C-H activation reaction in the gas phase was divided into single elementary reaction steps, which otherwise were not observable. Moreover elusive intermediates were characterized.
The first step of the reaction, coordination of the substrate, was thus tested for a possible kinetic 194 isotope effect. A TFE solution of 24 was prepared by protonolysis of 16 with HBF4 and electrosprayed in a Finnigan TSQ 700 mass spectrometer. The addition of 2.3 mtorr of a 1:1 + mixture of benzene/benzene-d6 in Q0 caused the formation of the benzene adduct [24(benzene)] + x and the benzene-d6 adduct [24(benzene-d6)] (Figure 20). The reaction was performed with conversions between 37 and 77 %. Subsequently one isotopic signal for each adduct overlapping the least with the other isotope pattern, m/z 664 and 670, respectively, was integrated with a width of 0.5 Da. Correction for the overlap and comparison of the integrals produced an adduct ratio of 1.01 ± 0.01.xi195
x The difference in enthalpy of vaporization, 33.9 kJ/mol for C6H6 and 34.2 kJ/mol for C6D6, was considered to be smaller than all other experimental errors and is included in the final error bounds.195 xi The very small error of 1 % stems only from the scatter of the data points. Experimental errors might add to that.
29 2 The Solvent in Pt-mediated C-H Activations
a) b)
100 integrated ed) t 80 alcula
Intensity 60 tensity (c tensity
40 Relative In
20
0 660 665 670 675 m/z 660 665 670 675 m/z Figure 20. a) Segment of the ESI-MS spectrum displaying the measured isotope pattern for [24(benzene)]+ + + and [24(benzene-d6)] . b) Calculated isotopic distribution for a 1:1 mixture of the [24(benzene)] (red) and + [24(benzene-d6)] (green).
Hence no kinetic isotope effect was observed for the coordination of benzene to [24]+ in the gas phase.
2.2.4 Reaction of a cationic PtII Complex with Benzene in the Gas-Phase
+ + + + The reaction of [24] , [24(H2O)] , and [24(TFE)] with benzene to give [25] and methane was investigated in a Finnigan TSQ 700 tandem mass spectrometer by ion-molecule reactions. A TFE solution of [24]+ was electrosprayed and [24]+ was mass-selected in the dau-mode in Q1 (∆m/z = 1). Subsequently, [24]+ was reacted with benzene in Q2 under various reaction conditions.
+ + + +
Cl Cl Cl Cl Cl Cl Cl Cl N N N Me Me N Ph Pt Pt Pt Pt Ph N N N C6H6 N C6H6 Cl Cl Cl Cl Cl Cl Cl Cl
+ [24]+ [25]+ [24(benzene)] [25(benzene)]+ Intensity m/z = 583 m/z = 645 m/z = 661 m/z = 723
550 600 650 700 m/z Figure 21. Daughter ion spectrum showing the C-H activation of benzene by collision of [24]+ with 2.3 mtorr of benzene in Q2 at an offset of –1.1 V (laboratory frame).
30 2.2 Gas-Phase Experiments
The resulting spectra (Figure 21) only contained signals for the educt [24]+ (m/z = 583), the product [25]+ (m/z = 645), the benzene adduct [24(benzene)]+ (m/z = 661), and the benzene adduct [25(benzene)]+ (m/z = 723). It should be noted that no structural claims are made for the benzene adducts, as they might comprise several interchanging structures.
+ + + Comparison of the gas-phase reactivities of [24] , [24(H2O)] , and [24(TFE)] towards benzene revealed that the solvent adducts displayed only very little reactivity towards benzene. Ion [24]+, on the other hand, readily formed [25]+ and the benzene adducts (Figure 22a). The efficiency of the reaction was determined by comparison of the ion peak intensities of the educt and the products under standardized conditions. The reaction proceeded most efficiently at low collision energies (more positive offsets of Q2). Under such conditions the collisions between ions and neutrals become more elastic because the excess energy of the reaction is reduced.196 However, fewer ions pass Q2 and the signal intensities decrease.
a) 0.6 b) 0.6
0.5 0.5
0.4 0.4
0.3 0.3 easured Conversion easured
0.2 Conversion easured M M 0.2
0.1 0.1 0.0
0.0 50-5-10 0-5-10 Offset of Q2 [V] Offset of Q2 [V] + + + Figure 22. a) Measured conversions of [24] (triangles), [24(H2O)] (circles), and [24(TFE)] (squares) with 1.95 mtorr of benzene in Q2 to give [24(benzene)]+, [25]+, and [25(benzene)]+ as a function of the collisional offset (laboratory frame). b) Measured conversions of [24]+ to products with 0.8 mtorr (open circles), 1.3 mtorr (circles), 1.8 mtorr (open squares), and 2.3 mtorr (squares) of benzene in Q2.
The reaction of [24]+ with benzene did not display saturation in the limit of the partial pressure of + benzene acceptable in Q2 (Figure 22b). When reactions of [24] were repeated with benzene-d6 instead of benzene consistently lower conversions were monitored and a kinetic isotope effect 1.18 ± 0.06 (95 % confidence, 30 experiments) was found.197 Again, ion peak intensities were used to measure the relative rates. No reactivity towards benzene other than loss of the solvent was + observed for [24(CH3CN)] in the gas phase.
31 2 The Solvent in Pt-mediated C-H Activations
2.2.5 Collision-Induced Dissociation of Cationic PtII Benzene Adducts with Argon, 2,2,2- Trifluoroethanol, and 1,1,1,2-Tetrafluoroethane
In independent experiments the reactivity of [24(benzene)]+ was investigated. Again, a TFE solution of [24]+ was electrosprayed in a Finnigan TSQ 700 tandem mass spectrometer. All ions were reacted with benzene at pressures below 10 mtorr in Q0 to give mainly [24(benzene)]+, which was mass-selected in the dau-mode in Q1. Subsequently, dissociation of the adduct ion was induced by Ar, TFE and 1,1,1,2-tetrafluoroethane (HFC 134a) in Q2 (CID). When 1.91 mtorr of Ar were used as collision gas, only the educt [24(benzene)]+ (m/z = 661) and the products of the elimination of methane ([25]+, m/z = 645) and of benzene ([24]+, m/z = 583) were detected in Q3 (Figure 23). The ratio of products/educt increased as a function of increasing collision energy (bringing down the offset of Q2). However, the ratio of [25]+/[24]+ remained unchanged for collision energies from -2 up to 14 eV (laboratory frame) with values of 84:16 (±3) (95 % confidence, 50 experiments).
+ + +
Cl Cl Cl Cl Cl N Cl N N Me Me Pt Pt Pt Ph C6H6 N N N Cl Cl Cl Cl Cl Cl
+ + + [24(benzene)] [24] [25] m/z = 661 m/z = 583 m/z = 645 Intensity
- CH4
- C6H6
600 650 m/z Figure 23. Daughter ion spectrum taken by selecting the [24(benzene)]+ ion with a width of 3 Da and performing collision-induced dissociation with 1.91 mtorr Ar with Q2 at –2 V offset (laboratory frame).
In similar experiments with TFE used as a collision gas in Q2 the branching ratio [25]+/[24]+ was found to be 83:17 (±2). No CID products containing TFE were found. Most importantly, when 1,1,1,2-tetrafluoroethane was used as a collision gas, the product ratio of [25]+/[24]+ was found to be identical with values of 85:15 (±1). These experiments were repeated with 6.8 mtorr of benzene-d6 in Q0 replacing benzene as a reagent gas. For the elimination of methane and of benzene from the benzene-d6 adduct, a kinetic isotope effect 1.18 ± 0.05 (95 % confidence, 95 experiments) was found along with a branching ratio of 81:19 (±1) for [25]+/[24]+. In addition, + deuterium isotopic scrambling in [24(benzene-d6)] was observed, which was deduced from the width of its signal in comparison with those for [24]+ and [25]+ (Figure 24).
32 2.2 Gas-Phase Experiments
+ 0.14 Cl 0.12 Cl N Me 0.10 Pt N C6D6 0.08 Cl Cl 0.06
+ 0.04 [24(benzene-d6)] m/z = 667 0.02
Intensity 0.00 580 590 650
- CHnD4-n
- C6HmD6-m
m/z 600 650 + Figure 24. Daughter ion spectrum taken by selecting the [24(benzene-d6)] ion with m/z = 3 and performing collision-induced dissociation with 1.91 mtorr 1,1,1,2-tetrafluoroethane with Q2 at –6 V offset (laboratory frame). An extraction of the isotope distribution found for partly deuterated [24]+ and [25]+ (green) is compared with a fully statistical distribution (red).
The insert in Figure 24 shows the calculated isotope distribution for partly deuterated [24]+ and + + [25] if deuterium isotopic scrambling in [24(benzene-d6)] was completely statistical (green) in comparison with the integrals of the experimental isotope signals (red). The observation of deuterium partitioning preferentially into the departing benzene vis-à-vis methane will be discussed in depth together with more detailed investigations in chapter 3.
2.2.6 Scaling a Pirani Gauge for Diverse Collision Gases
In the CID experiments of [24(benzene)]+ the gauge pressures of TFE and 1,1,1,2-tetrafluoroethane varied between 0.2 mtorr and 2.0 mtorr as read from the Pirani gauge attached directly to the housing of Q2. However, Pirani gauges measure thermal conductivity, which varies considerably among gases and vapors, rather than pressure.198 Because at low pressures (large mean free path λ) thermal conductivity is proportional to the product of heat capacity and pressure, the two gases should have the same actual pressure when the gauge readings are normalized by the ratio of their heat capacities.199 Only then a quantitative comparison of the results of the CID experiments conducted with TFE to those done with 1,1,1,2-tetrafluoroethane is possible.
Experimental heat capacities for TFE and 1,1,1,2-tetrafluoroethane are not available for low pressures. One reason is that at 1013 mbar and 343 K, manifold temperature of the instrument, TFE is a liquid and 1,1,1,2-tetrafluoroethane is a gas. The best value for the heat capacity of gaseous -1 -1 200 1,1,1,2-tetrafluoroethane, determined by increment calculations, is cp,298 = 85.1 J K mol . No comparable value for TFE was available. Therefore the heat capacities cV,343 of the two gases were determined from scaled harmonic frequencies of DFT calculations with B3LYP/6-31G* with Gaussian94. Calculations with different scaling factors revealed that the computed absolute heat
33 2 The Solvent in Pt-mediated C-H Activations
capacities for the two gases varied with 15 % (Table 4). TFE, having 3 normal modes more than 1,1,1,2-tetrafluoroethane, was found to possess a consistently higher heat capacity cV. However, the ratio of cV,TFE/cV,CF3CH2F remained almost constant with a value of 1.11. Furthermore this ratio is not altered by changes in the temperature (Table 5).
-1 - Table 4. Comparison of the computed heat capacities cV [J K Table 5. Comparison of the computed heat capacities cV [J K mol-1] of TFE and 1,1,1,2-tetrafluoroethane as function of the 1 mol-1] of TFE and 1,1,1,2-tetrafluoroethane at several scaling factor applied to the harmonic frequencies. temperatures.
Scaling F 3 C OH F3 C F cV ratio T [K] F 3C OH F3C F cV ratio 0.77 100.1 90.3 1.109 298.15 94.5 85.1 1.110 0.81 96.9 86.5 1.120 313.15 97.7 88.0 1.110 0.84 94.5 85.1 1.110 328.15 100.7 90.8 1.109 0.87 92.3 83.1 1.111 343.15 103.7 93.5 1.109 0.91 89.6 80.6 1.112 0.95 87.1 78.2 1.113
From the results of Table 4 a uniform normalization factor of 1.11 for the pressure readings of 1,1,1,2-tetrafluoroethane was deduced.
2.2.7 Quantitative Comparison of PtII-mediated C-H Activation Reactions in the Gas Phase
The reaction efficiency of the elimination of benzene and methane from [24(benzene)]+ by CID with TFE and 1,1,1,2-tetrafluoroethane was compared quantitatively. The offset of the collision cell (Q2) was varied between +2 V and –14 V. The collision gas pressure, as read from the Pirani gauge, was increased in steps of 0.2 mtorr between 0.2 mtorr and 1.2 mtorr. In 150 experiments, CID with 1,1,1,2-tetrafluoroethane always yielded higher conversions of [24(benzene)]+ to [24]+ and [25]+ than CID experiments with TFE. A difference in reaction efficiency of 1.13 ± 0.02 (95 % confidence) was found for all experiments. Table 6 gives the mean values for the relative reactivities observed for the various pressures.
Table 6. Conversions of [24(benzene)]+ to [24]+ and [25]+ depending on the pressure of the CID gas with the determined ratio of the conversions induced by 1,1,1,2-tetrafluoroethane and TFE.
Pressure F3C F F 3C OH Ratio 0.2 mtorr 0.17 0.15 1.107 0.4 mtorr 0.32 0.28 1.159 0.6 mtorr 0.46 0.40 1.164 0.8 mtorr 0.57 0.51 1.145 1.0 mtorr 0.63 0.57 1.105 1.2 mtorr 0.70 0.67 1.096
34 2.2 Gas-Phase Experiments
As stated above, the heat capacity of TFE is 1.11 times higher than that of 1,1,1,2-tetrafluoroethane. As a result, when the same pressure for the two gases is read, the actual pressure of 1,1,1,2- tetrafluoroethane was 1.11 times higher than that of TFE. When the results of CID with 1,1,1,2- tetrafluoroethane from Table 6 were scaled down by 1.11, the results from the two sets of experiments are undistinguishable. A superposition of the scaled conversions as a function of collision energy is shown in Figure 25. Throughout the investigations the branching ratio [25]+/[24]+ did not alter (see chapter 2.2.5).
0.7
0.6 onversion 0.5 Scaled C Scaled 0.4 1,1,1,2-tetrafluoroethane 2,2,2-trifluoroethanol
0.3
0.2 0-5-10 Offset of Q2 [V] Figure 25. Dependence of the normalized, combined yield of benzene and methane-loss products in CID of [24(benzene)]+ on the collision energy varied in steps of 1 eV (laboratory frame, lines connect the data points) for the two different collision gases at 0.60 mtorr pressure
2.2.8 Discussion
+ + The claim that the reaction of [19(H2O)] or [20(H2O)] with methane or arenes in TFE solution is unusually facile begs an explanation, especially in connection to the identification of the rate- determining step for the reaction. The long-term potential of C-H activation makes the identification of those structural or process parameters that confer high activity of considerable interest, especially given the much harsher reaction conditions that needed to be employed in the original Shilov chemistry.
+ + The mechanism of C-H activation of benzene by [19(H2O)] and [20(H2O)] can be drawn as indicated in Scheme 13, in which the various ligand substitutions connecting I-V as well as IX-XII are formulated as proceeding by associative mechanisms.xii
xii In Scheme 13, the structures should not be regarded to mean that a particular stereochemistry is assigned to the intermediates. The structures are presumed to undergo facile isomerization and agostic stabilizations from the diimine ligand must be envisioned. The present experiments make no prediction concerning the stereochemistry.
35 2 The Solvent in Pt-mediated C-H Activations
++ + + N Ar Ar Ar Ar N CH N CH3 CH N 3 3 Pt CH3 Pt N Pt N N Pt N OH TFE TFE Ar 2 Ar Ar Ar OH2 TFE I II III IV
+ + + + H N Ar Ar Ar Ar N CH N N CH3 3 CH Pt H Pt Pt CH3 3 N N N Pt N Ar Ar Ar Ar H
VIII VII VI V
+ + + + Ar Ar Ar N N H Ar N TFE TFE N OH2 Pt N Pt Pt N Pt CH3 N N Ar Ar Ar Ar OH2 TFE IX X XI XII Scheme 13. Schematic representation of the associative ligand exchanges and the scrambling process for II Pt -mediated C-H activations of benzene in TFE.
Isotopic scrambling,31 as well as equilibration between various isomeric aryl complexes,59 in solution-phase studies requires that interconversion between the PtIV intermediate VII, the two σ- complexes VI and VIII, and presumably also the π-complex V be much faster than elimination of benzene or methane, ruling out either the actual oxidative addition or reductive elimination for the rate-determining step. Therefore, it has been recently proposed that the rate-determining step in the C-H activation of arenes is the solvent-assisted associative ligand exchange reaction III to V, either as a single concerted step or as a combination of two steps, one of which being actually rate determining.31 Putting the structures III-V on a potential energy diagram in Scheme 14, one can argue from the Hammond postulate, applied to the two possible ligand dissociations from IV, that the net endothermic transformation III to V means that the rate-determining transition state must be the second step, i.e., the departure of TFE from the five-coordinate intermediate.
This proposition is directly tested in this investigation.
While solvation leads to large changes in the potential surfaces for gas-phase ion-molecule 201 reactions, such as proton exchange, SN2, or carbonyl additions, relative to their congeners in solution, reactions of organometallic complexes electrosprayed into the gas phase have proven surprisingly similar to the corresponding reactions in solution122 in other C-H activations126,127 we have examined, leading us to presume that the analogy in this case is also valid. Nevertheless, solution-phase experiments were executed to confirm the belief (vide infra).
36 2.2 Gas-Phase Experiments
+ N Ar
CH3 N Pt Ar TFE IV
forward reaction
Energy
Ar + Ar + N CH3 N CH3 Pt Pt N N TFE Ar Ar III V Scheme 14. Reaction coordinate with the rate-determining transition state for the C-H activation of benzene.
Four observations may be made from the ion-molecule reactions. First, we see the same gross reaction in solution and in the gas phase. Second, the overall kinetic isotope effects for the reaction in the two different settings are similar; kH/kD = 1.18 ± 0.06 in the gas phase versus 1.06 ± 0.05 in 31 solution. Third, when benzene-d6 is used instead of benzene, the distribution of deuterium between methane and benzene products shows nearly full equilibration. Last, in gas-phase experiments, [24(benzene)]+ gives branching ratios close to the reported value of 82:18 for reductive elimination of methane versus benzene observed when [(N-N)Pt(CH3)(C6H5)] is protonated in solution.31
Furthermore, the decisive observations in the gas phase, isotope partitioning and product branching ratio, do not stem from addition steps starting from unsaturated [24]+, which is nonexistent in solution, but rather from reactions after formation of [24(benzene)]+. The presumably isomerizing species [24(benzene)]+ in the gas phase represents exactly those species depicted as V, VI, VII, and VIII in solution (with the proviso that one cannot yet distinguish between π- and σ-complexes in the gas phase). These four observations serve to show that the gas phase chemistry does not differ grossly from that in solution. In particular, the significance of the addition reaction of [24]+ should be understood as primarily a gas-phase synthesis of [24(benzene)]+, whose subsequent reactions are shown below to make a relevant prediction for solution-phase mechanisms.
A rate-determining transition state with TFE bound at the metal is ruled out by the present gas- phase experiments: the threshold behaviors in Figure 25 for reductive elimination of benzene and methane from [24(benzene)]+ induced by low-energy collision of the ion with TFE or 1,1,1,2- tetrafluoroethane are identical. The collision of preselected [24(benzene)]+ with TFE corresponds to the reaction of V with TFE because both the gas-phase and solution-phase experiments show rapid equilibration of V-VIII prior to any other reaction. The two collision gases, TFE and 1,1,1,2- tetrafluoroethane, are isoelectronic and isostructural and have almost the same mass, making the
37 2 The Solvent in Pt-mediated C-H Activations
kinematics of the collisions similar. They have moreover almost identical dipole moments,xiii making even long-range electrostatic ion-molecule interactions very similar. They do differ, however, in that 1,1,1,2-tetrafluoroethane coordinates much more weakly to electrophilic centers than does TFE, if at all.xiv Moreover, the near-constant branching ratio in the elimination reactions of [24(benzene)]+ as the collision gas is varied from argon to TFE to 1,1,1,2-tetrafluoroethane suggests strongly that TFE participates in the elimination only as a vehicle for energy transfer to [24(benzene)]+. If coordination of TFE were important at the transition state, one would expect to see different thresholds, and perhaps different selectivities, for elimination of benzene or methane from [24(benzene)]+ when one changes the collision gas. By microscopic reversibility, the reaction in the other direction must also have no TFE bound at the transition state. Therefore, a putative solvent-assisted associative mechanism cannot be the explanation for the facile C-H activation reaction. The isotopic scrambling in the gas-phase experiment matches that seen in the analogous reaction in solution, which requires as a minimum conclusion rapid reversible interconversion between VI, VII, VIII, and presumably V, just as has been concluded for the solution-phase reaction.31 This information, combined with the experimental conclusion that TFE is not bound at the rate-determining transition state for either benzene or methane elimination, requires that there exists at least one kinetically significant intermediate between IV and V for benzene activation and at least one further intermediate between IX and VIII for methane activation. The rate-determining transition state for C-H activation of benzene by this class of cationic PtII complexes is necessarily the transition state in which this intermediate is transformed into V. Similarly, the rate-determining transition state for C-H activation of methane is the transition state in which the intermediate is transformed into VIII. The experiment does not specify what the structure of either of the new intermediates is, but their existence, and their crucial position immediately preceding the rate- determining step for the overall C-H activation reaction, is mandatory.
2.3 Solution-Phase Experiments
In order to verify the relevance of the gas-phase findings for the condensed phase, time-dependent UV/Vis experiments were performed.202 Two sets of experiments were carried out. In the first set, TFE was present in the reaction of [24(water)]+ with benzene in similar amounts to those described in literature.31 In the second series, TFE was excluded from the reactions. The general scheme of reaction was checked by comparison with the results for the reaction of [20(water)]+ with benzene. UV/Vis as analytical tool for this purpose was chosen for two reasons. First, all complexes, cations and their solvent adducts were strongly colored. Second, the reactions were known to proceed at rates which are in a range that is conveniently followed by this technique.
xiii From PM3 calculations on optimized geometries, one finds for TFE and 1,1,1,2-tetrafluoroethane dipole moments of 2.2 and 2.3 D, respectively. xiv DFT calculations at the B3PW91 level of theory on ensembles containing the model systems for [24(solv)]+ find that + + + [24(FCH2-CF3)] is 71 or 77 kJ/mol [24(FCF2-CH2F)] higher in energy than [24(HOCH2CF3)] , which is comparable + + to [24(CO2)] (73 kJ/mol). CO2 occasionally acts as an inert gas. The same calculations find that [24(benzene)] is only 9 to 26 kJ/mol higher in energy than [24(TFE)]+ depending on the conformation of the benzene relative to the coordination plane (see ch. 3.3). All ensembles contained the identical number of molecules in differing arrangements.
38 2.3 Solution-Phase Experiments
2.3.1 Benzene C-H Activation by PtII complexes in the Presence of 2,2,2-Trifluoroethanol
For the first set of experiments, the reaction of [24(water)]+ with benzene to give [25(water)]+ in the presence of TFE, purple solutions of 16 in TFE were prepared in UV/Vis cuvettes (1.4 ml) (Figure xv,203 + 26a). The addition of HBF4 (50 % in water) to the solution caused conversion to [24(water)] (Figure 26b). During the addition the gas formation (CH4) was observed and the solutions turned orange.
a) b) + 2.0 Cl Cl 2.0 Cl Cl N N Me Me Pt Pt 1.5 OH N Me N 2 1.5 Cl Cl Cl
Cl ce (A)
1.0 1.0 16 in TFE [24(water)]+ in TFE Absorban Absorbance (A) Absorbance
0.5 0.5
0.0 0.0 400 500 600 λ [nm] 400 500 λ [nm] Figure 26. a) UV/Vis spectrum of 16 in TFE with 0.16 M water at 25 °C. b) The same solution after the + addition of HBF4 50 % in water, i.e. [24(water)] . In order to ensure the applicability of Lambert-Beer's-law, the time-dependent experiments were performed at lower concentrations than displayed here.
The cuvettes were then thermalized to 25 °C in the UV/Vis instrument.xvi Subsequently, the reactions were initiated by the addition of measured volumes of benzene at the same temperature.xvii Benzene concentrations in these reactions were always high enough to ensure pseudo-first-order II conditions ([C6H6] > 10[Pt ]). In addition, water concentrations were chosen in a way that at least 90 % of the platinum species were present as the aqua adducts ([H2O] ≥ 0.04 M), so that the data can be treated in terms of a simple transformation of [24(water)]+ to [25(water)]+. Control experiments showed that the reaction rate was independent of the initial PtII concentration. Figure 27a shows the typical spectral change over time for the reaction between [24(water)]+ and benzene. Kinetic data were obtained by following the absorbance (A) at 310, 400, and 545 nm as a function of time (Figure 27b).
xv The TFE used for theses experiments was purified and dried according to literature procedures. Subsequently water was added to achieve a concentration of 0.16 M. xvi Independently from the temperature displayed by the heat regulator, the temperature of the reaction mixture was checked inside the cell. xvii In order to reduce the time delay between addition of the benzene and the start of the recording, custom-made caps for the UV/Vis cells were used. A PEEK cap with a 0.5 mm hole equipped with a septum permitted addition of benzene inside the instrument by a syringe while the recording was already running. A vigorous addition also assured rapid mixing of the reaction partners.
39 2 The Solvent in Pt-mediated C-H Activations
a) b) 0.75 0.8
0.70
0.6 ce (A) ce
bance (A) bance 0.65 Absor
Absorban 0.4 0.60
0.2 0.55
400 450 500 λ [nm] 020004000 6000 8000 Time [s] Figure 27. a) Stacked UV/Vis absorption spectra showing the time development of the spectrum of [24(water)]+ and benzene at 25 °C in TFE solution containing 3.75 M benzene and 0.14 M water. The displayed traces were recorded at 5 min intervals. Isosbestic points at 456 and 517 nm. b) Time dependence of the absorbance at 400 nm for the reaction ([benzene] = 0.43 M, [water] = 0.04 M). Data points were recorded at 3 s intervals. All data were used for a fit.
The observed rate constants kobs were obtained from a first-order three-parameter (kobs, A0, and A4) least-squares fit of the absorbance (A) versus time (t) data employing eq 2.2 (derivation see 7.1). A = A4 + (A0 - A4)exp(-kobst) eq 2.2 Thus five observed rate constants kobs were obtained for concentrations of benzene between 0.43 and 3.75 M and of water between 0.04 and 0.22 M. A plot of kobs as a function of [benzene]/[water] was fitted with a straight line through zero (Figure 28) indicating a rate law with first-order dependence in [benzene] and inverse first-order dependence in [water] (compare to eq 1.1). From -5 -1 the slope a calculated rate constant k298 = (4.93 ± 0.42) · 10 s was derived.
1.0x10-3
8.0x10-4 ] -1 -4 [s 6.0x10 obs k
4.0x10-4
2.0x10-4
0.0 051015 20 [benzene]/[water] Figure 28. Dependence of the pseudo-first-order rate constant of the reaction [24(water)]+ to [25(water)]+ at 25 °C in TFE solution containing benzene in concentrations between 0.43 M and 3.75 M and water between 0.04 and 0.22 M.
40 2.3 Solution-Phase Experiments
A number of control experiments were performed to exclude the possibility of major side reactions. The identity of [24(water)]+ was checked by recording a 1H NMR spectrum before and after the protonolysis of 16 with HBF4 in wet TFE-d3. The isosbestic points at 456 nm and 517 nm in the spectra (Figure 27a) revealed that no intermediates accumulated to a significant level. Furthermore, no ionic products other than [24(acetonitrile)]+ and [25(acetonitrile)]+ were detected by ESI-MS when a drop of the reaction mixture was added to an excess of CH3CN (Figure 29a). Finally, the spectrum at the end of a reaction was compared to a spectrum obtained under similar conditions by protonolysis of 18 and no differences were found. However, a slow background reaction was observed in the absence of benzene similar to what has been described for the reaction of [20(water)]+ in TFE.31 The kinetics of this reaction were investigated by leaving [24(water)]+ in TFE without benzene at 25 °C and following the absorbance at 400 nm. By applying a linear fit to -6 -1 xviii the data a rate constant kdecomp = (6.75 ± 0.01) · 10 s was extracted (Figure 29b). This background reaction is almost an order of magnitude slower than the reaction of [24(water)]+ and benzene. Furthermore, it seems to be suppressed in the presence of benzene as can be seen from Figure 27b were no deviation from pseudo-first-order behavior was observed. This decomposition reaction might be attributed to the formation of hydroxy-bridged PtII dimers, which were observed by Bercaw and co-workers for similar [(diimine)PtII (Me)(solv)]+ complexes when standing in TFE solution.54 A related decomposition reaction for cationic PtII methyl complexes ligated with bidentate phosphines was reported by Kubas and co-workers.204
a) + + b) 0.65 Cl Cl Cl Cl N N Me Ph Pt Pt 0.60 NCCH3 NCCH3 N N 0.55 Cl Cl Cl Cl 0.55 bance (A) bance Intensity 0.50 [24(acetonitrile)]+ [25(acetonitrile)]+ Absor 0.50
0.45 01000 0.45
0.40 400 500 600 700 800 m/z 0250050007500Time [s] Figure 29. a) ESI-MS spectrum of the reaction of [24(water)]+ and benzene quenched at 37 % conversion by acetonitrile. b) Time-dependent absorbance (A) at 400 nm of the side reaction of [24(water)]+ in TFE observed in the absence of benzene (lower curve) in comparison with the desired reaction (upper curve). The insert displays the expanded start of the reaction.
The observations described above strongly resembled the descriptions of the reaction [20(water)]+ and benzene by Tilset, Bercaw and co-workers.31 To confirm that this was not coincidentally and that the work was properly executed the UV/Vis experiments described in ref 31 were repeated. The same spectral behavior of [20(water)]+ was observed in the reaction with benzene (Figure 30a) along with the unidentified deviation from first-order behavior after 3-4 half-lives (Figure 30b).
xviii The data were not fitted to eq 2.2 because linear behavior was assumed for the background reaction for the time monitored (see chapter 7.2)
41 2 The Solvent in Pt-mediated C-H Activations
a) b) 1.5 1.00
1.0 0.95 ce (A) ce Absorban Absorbance (A) Absorbance 0.5 0.90
0.85 0.0 400 450 500 λ [nm] 025000 50000 Time [s] Figure 30. a) Stacked UV/Vis absorption spectra showing the time development of the spectrum of [20(water)] and benzene at 25 °C in TFE solution containing 1.88 M benzene and 0.17 M water. Isosbestic points at 436 and 485 nm. Pseudo-first-order rates were derived from the absorbance (A) at 375, 390, and 510 nm. The displayed traces were recorded at 60 min intervals. b) Absorbance (A) at 390 nm versus time for the same reaction. (~) illustrate the unidentified background reaction. These points were not used for a fit.
When the reaction was performed with varying benzene and water concentrations, a first-order dependence in [benzene] and inverse first-order dependence in [water] was also found (Figure xix -6 -1 31). A rate constant k298 = (9.24 ± 0.60) · 10 s was extracted, as described above, for the reaction of [20(water)]+ and benzene in TFE at 25 °C.
2.5x10-4
2.0x10-4
] -4
-1 1.5x10 [s
obs k
1.0x10-4
5.0x10-5
0.0 051015 20 25 [benzene]/[water] Figure 31. Dependence of the pseudo-first-order rate constant of the reaction [20(water)]+ and benzene at 25 °C in TFE solution containing benzene in concentrations between 0.57 M and 3.75 M and water between 0.14 and 0.22 M.
xix In ref. 30 the fit in Figure 2 does not go through zero, which might be one reason for the two different rate constants, found for the reaction of [20(water)]+ and benzene (see discussion 2.3.4). According to the rate-law a x and y intercept of zero is inevitable.
42 2.3 Solution-Phase Experiments
Throughout the studies described in this chapter it was observed that the reproducibility of results depended strongly on a number of factors. Thorough attention was paid to the purity of all employed compounds, especially solvents and water.203 The amount of acid had to be chosen in a way that complete conversion of 16 to [24(water)] and of 12 to [20(water)] was quickly guaranteed while a large excess of HBF4 led to turbidity of the reaction. A similar tarnish was observed at high benzene concentrations (o 3.75 M).
2.3.2 Benzene C-H Activation by PtII complexes in the Absence of 2,2,2-Trifluoroethanol
The putative influence of TFE on the reaction of [24(water)]+and [20(water)]+with benzene was investigated by performing the same experiments as in 2.3.1 in the absence of TFE. Hence, benzene was purified and dried according to literature procedures.203 Then it was mixed with water, heated to 50 °and cooled to 25 °C. Phase separation afforded benzene-containing water in concentration of 0.029 M.205 This solvent is referred to as wet benzene.xx Complex 16 dissolved readily in wet benzene, sometimes with the help of ultra-sound to produce a clear, green solution. Upon xxi + protonolysis with HBF4 in water/ether, [24(water)] instantly formed (yellow) and reacted quickly with benzene to produce [25(water)]+(orange) (Figure 32a). Again, kinetic data were obtained by following the absorbance (A) at 3 wavelengths, 330, 440, and 520 nm (Figure 32b).
a) b) 1.0 1.0
0.8 0.8
0.6 ce (A) 0.6 bance (A) bance Absor 0.4 Absorban 0.4
0.2 0.2
0.0 0.0 350 400 450 500 550 λ [nm] 50 100 150 200 250 Time [s] Figure 32. a) Stacked UV/Vis absorption spectra showing the time development of the spectrum [24(water)]+ and benzene at 25 °C in neat (11.25 M) benzene containing 0.029 M water. Isosbestic points at 339, 405, and 468 nm. The displayed traces were recorded at 20 s intervals. b) Time dependence of the UV/Vis absorbance at 330 (red), 440 (black), and 520 nm (green) for the reaction of [24(water)]+ and benzene at 25 °C in benzene (11.25 M) containing 0.029 M water. Data points were recorded at 0.5 s intervals. All data from three wavelengths were used for a fit.
As in the presence of TFE clean pseudo-first-order behavior was found for the time of monitoring. Performing a least-squares fit of the data to eq 2.2 resulted in an observed rate constant of kobs = (2.56± 0.64) · 10-2s-1, which by division by [benzene]/[water] yielded a first-order rate constant -5 -1 k298 = (6.61± 1.66) · 10 s . In addition to the control experiments described in 2.3.1, a linear relation between the absorbance at 520 nm and the absorbance at 330 nm was found in an xx Adding water to benzene in a well-defined manner is otherwise difficult to achieve, especially for small quantities because, the purified benzene might not be completely dry in the moment of addition or the temperatures of water and benzene might differ, thereby yielding unreliably volumes. xxi 50 volumes of ether were added to HBF4 50 % in water in order to expand the volume of the acid and was assumed to interfere only to a negligible extent with the reaction.
43 2 The Solvent in Pt-mediated C-H Activations
absorbance difference (AD) diagram (Mauser diagram) for the complete time of reaction (Figure 33).
0.3 = 520nm
λ Re ac tio n T im 0.2 e Absorbance (A)at
0.1
0.8 1.0 Absorbance (A) at λ = 330 nm Figure 33. AD diagram of the absorbance at 520 and 330 nm of the reaction of [24(water)]+ with benzene.
Once more coincidence of the results was ruled out by repeating the experiments with [20(water)]+ and benzene in the absence of TFE. 12 dissolved readily in wet benzene to yield a purple solution. + When HBF4 was added a yellow, clear solution of [20(water)] formed. The reaction with benzene was monitored by the time-dependent absorbance (A) at 390 nm. Figure 34a and b display the spectral change and the absorbance at 390 nm over time, respectively. a) b) 0.6 0.36
0.5 0.35
0.4 ce (A) 0.34
0.3 Absorbance (A) Absorbance Absorban 0.33 0.2
0.32 0.1
0.0 0.31 350 400 450 500 550 λ [nm] 200 400 600 800 1000 Time [s] Figure 34. a) Stacked UV/Vis absorption spectra showing the time development of the spectrum of [20(water)]+ and benzene at 25 °C in neat (11.25 M) benzene containing 0.029 M water. Isosbestic points at 363 and 425 nm. The displayed traces were recorded at 20 s intervals. b) Time dependence of the UV/Vis absorbance at 390 nm for the reaction of [20(water)]+ and benzene at 25 °C in neat (11.25 M) benzene containing 0.029 M water. Data points were recorded at 2 s intervals. All data were used for a fit to an exponential function.
44 2.3 Solution-Phase Experiments
A clean pseudo-first-order behavior was found for the time of reaction. A least-squares fit to eq 2.2 -3 -1 yielded an observed rate constant kobs = (2.91 ± 3.80)C10 s . When kobs was divided by -6 -1 [benzene]/[water] a first-order rate constant k298 = (7.50 ± 3.80)C10 s was calculated. The somewhat larger errors in the experiments without TFE are due to experimental inconveniences. As in the experiments in the presence of TFE, the reaction was initiated inside the UV/Vis instrument while already recording the absorbance. However, clean data were only obtained shortly after the cover slide was completely closed. In the wet benzene experiments this time was superior to 2 half- lives. Furthermore, no attempts were made to reduce the errors by omitting bad data. When the experiments in wet benzene were repeated with D2O replacing H2O the incorporation of deuterium into the phenyl residue of [25]+ was observed. This finding will be discussed in detail in chapter 4.2.
2.3.3 Activation Parameters for the C-H Activation of Benzene by a PtII complex
The temperature dependence of the C-H activation of benzene by [24(water)]+ was studied over the range between 5 and 50 °C in intervals of approximately 5 K. The observed absorbances at the various temperatures are shown in Figure 35. The activation parameters were calculated from the Eyring plot shown in Figure 35.
a) b) 0.78 -285 *T)] 0.72 b T [°] ce (A)
5.0 *h)/(k -290 obs
10.0 k 15.0
Absorban 20.0 0.66 25.0 R * ln[( 30.3 35.5 -295 40.0 45.0 50.0 0.60 0200400 600 800 1000 0.0031 0.0032 0.0033 0.0034 0.0035 0.0036 Time [s] 1/T [1/K] Figure 35. a) Measured absorbance (A) at 400 nm for the reaction of [24(water)]+ with benzene in TFE at temperatures between 5 °C (black) and 50 °C (purple). Experimental conditions: V = 1 ml, [PtII] = 0.03 mM, [H2O] = 0.01 M, [benzene] = 4.47 M, time interval: 0.5 s, maximum reaction time used for fit: 350 s. b) Eyring plot for the reaction of [24(water)] with benzene with the rates calculated from the data displayed in a).
An activation enthalpy of ∆H≠ = 13.7 ± 1.3 kJ/mol and an activation entropy of ∆S≠ = ≠ -243 ± 4 J/(K·mol) were determined resulting in an activation barrier at 298 °C of ∆G 298 = 86.3 ± 1.8 kJ/mol.xxii A similar value was obtained when the experiments were repeated in neat wet ≠ benzene between 15 and 45 °C at intervals of 5K, ∆G 298 = 93.2 ± 17.9 kJ/mol. The larger error stems from similar sources as described in 2.3.2.206
xxii For the sake of reproducibility the errors were computed from the scatter of the data points. Error calculations were performed with the symmetrized, weighted errors on the measured rate constants. Also estimates on the logarithmic errors were considered. Both procedures yielded similar values.
45 2 The Solvent in Pt-mediated C-H Activations
2.3.4 Discussion
The investigations of the elimination reactions from the benzene adduct [24(benzene)]+ to give [24]+ and [25]+ in the gas phase revealed that the reaction is not influenced by TFE. In order to confirm this result, solution-phase experiments of the microscopic reverse, C-H activation of benzene by [24(water)]+, were performed. Furthermore, the general validity of the results was corroborated by analogous experiments for a reaction described by Tilset, Bercaw and co-workers, the C-H activation of benzene by [20(water)]+.
The solution-phase kinetics of [19(water)]+ and [20(water)]+ were reported to be fairly similar, with rates differing by less than 1 order of magnitude.54 Therefore, a broadly similar chemistry for [24(water)]+ versus [19(water)]+ or [20(water)]+ may be assumed. The expectation is fully borne out by the solution-phase kinetics described in chapters 2.3.1 - 2.3.3. Not only is the chemical behavior for complex [24(water)]+ essentially identical to that for [20(water)]+, but even the absolute values -5 -1 -6 -1 of the rates are not too different, e.g., k298 = (4.93 ± 0.42) · 10 s versus (9.24 ± 0.60) · 10 s for [24(water)]+ and [20(water)]+, respectively. Given that very substantial data have been published for [19(water)]+ and [20(water)]+, it is gratifying to see that the rate constant measured in the present work for [20(water)]+ is in acceptable agreement with the value of (1.98 ± 0.03) · 10-5 s-1 reported for the same complex under the same conditions by Tilset, Bercaw, and co-workers.xxiii Moreover, the clean isosbestic points in the time-dependent UV/Vis spectra indicate that it is possible to find conditions under which all other side-reactions are much slower than the transformation [24(water)]+ [25(water)]+ or [20(water)]+ to the corresponding phenyl complex. These experiments are important to confirm that the solution-phase studies have been properly performed. This is further supported by the comparison of the activation parameters for the reaction [24(water)]+ and benzene in comparison to [20(water)]+ and benzene. Both reaction go through a highly ordered rate-determining transition state implied by ∆S≠ = -243 ± 4 J/(K·mol) for the + + ≠ reaction of [24(water)] versus -67 ± 4 J/(K·mol) for [20(water)] . Moreover, the ∆G 298 values of 86.3 ± 1.8 versus 100 ± 4 kJ/mol display the trend of the calculated rate constants, [24(water)]+ reacting faster with benzene than [20(water)]+.
To conclude, the solution-phase kinetics for [24(water)]+ and [20(water)]+ fully support the conclusions of the gas-phase study. Examination of Figure 28 and Figure 31 reveals the transformations [24(water)]+ [25(water)]+ and [24(water)]+ to the corresponding phenyl complex are each described by a single rate constant over a broad range of [benzene]/[water] in wet TFE solution. If one makes the decisive comparison of the derived rate constant for benzene activation in TFE solution to the corresponding rate constants in neat (wet) benzene, one finds that, for each of the complexes [24(water)]+ or [20(water)]+, the rate constant shows no dependence on TFE + concentration at all (Table 7). For the reaction [24(water)] and benzene, k298 = (4.93 ± 0.42) · 10-5 s-1 is found in TFE solution, and (6.61 ± 1.66) · 10-5 s-1 with no TFE at all. For [20(water)]+ and -6 -6 -1 benzene, the corresponding rate constants are k298 = (9.24 ± 0.60) · 10 and (7.50 ± 3.80) · 10 s . If the rate constant in the complete absence of TFE is experimentally indistinguishable from that when TFE is present as the solvent, i.e., the reaction is zero-order in TFE, it must be concluded that TFE cannot be involved in the rate-determining step for C-H activation. Arguments that complex [24(water)]+ is somehow unique in this regard are countered by the corresponding experiments with complex [20(water)]+, which displays identical behavior. One concludes that TFE serves only as an inert diluent. xxiii -5 -1 In their study Tilset and Bercaw report a second, differing rate constant k298 = (1.53 ± 0.03) · 10 s for the reaction of [20(water)] and benzene (Figure 3, ref. 31).
46 2.4 Conclusions
Table 7. Experimentally determined rate constants at 298 K [s-1] for the reaction of platinum complexes [24(water)]+ and [20(water)]+ and benzene in the presence and in the absence of TFE.
Reaction k298 in the presence of TFE k298 in the absence of TFE [24(water)]+ and benzene (4.93 ± 0.42) · 10-5 (6.61 ± 1.66) · 10-5 [20(water)]+ and benzene (9.24 ± 0.60) · 10-6 (7.50 ± 3.80) · 10-6
2.4 Conclusions
Gas-phase ion-molecule reactions of [(diimine)PtMe(solv)]+ (solv =water, TFE) with benzene show a remarkable similarity to the corresponding solution-phase reactions. Accordingly advantage is taken of electrospray ionization tandem mass spectrometry to isolate postulated intermediates and investigate the mechanism of C-H activation in the reaction of [(diimine)PtMe(solv)]+ with benzene. A rate-determining transition state in which TFE solvent is associated with the metal is ruled out by the results from the gas-phase experiments. This is confirmed by solution-phase kinetic studies. It is encouraging that an ordinary even hydroxylic solvent (solv') may by employed for C-H activation without slowing the reaction down, with the exclusive prerequisite that the solvent adduct [(diimine)PtMe(solv')]+ is not the energetically favored over the water adduct.
47 3 Scrambling in a Cationic PtII Complex in the Gas Phase
3 Scrambling in a Cationic PtII Complex in the Gas Phase
+ When the collision-induced dissociation (CID) of the benzene-d6 adduct [24(benzene-d6)] was investigated in the gas phase, incomplete deuterium isotopic scrambling around the phenyl and methyl ligands was observed (see ch. 2.2.5, p. 33, esp. Figure 24).
+
Cl Cl N CHnD3 - n Pt + N Cl
Cl - benzene-d3+n Cl Cl N CH 3 + Pt CID [24-d] D6 N + Cl Cl - methane-d1+m Cl Cl N C6HmD5 - m [24(benzene-d )]+ 6 Pt N Cl
Cl
[25-d]+ + + + Scheme 15. CID of [24(benzene-d6)] leads to four different products, methane, benzene, [24] , and [25] , and to isotopomers thereof. The different isotopomers can be only explained by deuterium isotopic + scrambling in [24(benzene-d6)] .
+ Similar observations were described in solution for the reaction of [(tmeda)Pt(Me)(Et2O)] and + 30,55 [(tmeda)Pt(Me)(NC5F5)] with deuterated hydrocarbons, for TpPt(Me)(H)2 in the presence of 32 + 56 methanol-d4, and for the reaction of [19(water)] with benzene-d6. Furthermore, Tilset and + Bercaw observed isotopic scrambling when [20(water)] was reacted with benzene-d6 and 31 1,3,5-benzene-d3 in TFE. Deuterium from benzene was incorporated into the departing isotopomers of methane almost fully statistically. However, the overall extent of deuterium incorporation into methane was smaller than expected from completely random statistics. They precluded ligand exchange and showed by 1H NMR experiments that the process was intramolecular once benzene was coordinated to the PtII center. Templeton and Goldberg ≠ 25 determined an activation barrier of ∆G 298 = 108 kJ/mol for isotopic scrambling in Tp'Pt(Me)2D. Again, the deuterium isotopic scrambling did not go to completion. Moreover, Templeton found the ≠ barrier for the exchange of a proton in Tp'Pt(H)(C6H6) to be ∆G 241 = 53 kJ/mol with a first-order -1 28 rate constant k241 = 17 s and a kinetic isotope effect of 3.0.
48 3.1 Deuterium Isotopic Scrambling in PtII Benzene Adducts in the Gas Phase
3.1 Deuterium Isotopic Scrambling in PtII Benzene Adducts in the Gas Phase
+ The rate of deuterium isotopic scrambling in [24(benzene-d6)] was tentatively assumed to occur with roughly comparable rates as observed in Tp'Pt(H)(C6H6) and Tp'Pt(D)(C6D6). Consequently, + the rate of deuterium isotopic scrambling in [24(benzene-d6)] was estimated on the basis of the data in ref. 28 for the conditions in Q0 of the modified TSQ 700 instrument. As a result an estimated -1 rate of deuterium isotopic scrambling on the order of k343 = 30000 s was obtained for the manifold temperature of 70 °C. A retention time on the order of 10 - 100 ms for ions in Q0 of the instrument prompted the expectation of complete statistical deuterium isotopic scrambling around the phenyl and methyl ligands.207,xxiv This was not observed, however, the low resolution of the Finnigan MAT TSQ 700 instrument at the experimental conditions precluded an accurate investigation of the effect.
Therefore similar experiments as in 2.2.5 were conducted in a Finnigan TSQ Quantum Tandem ESI-MS instrument. This instrument allowed higher mass resolution, especially in the dau-mode, along with an improved signal-to-noise ratio. However, addition of a reagent gas in Q0 was not possible. Therefore benzene-d6 was added in the encapsulated spray region of the instrument. When + [24] was electrosprayed from TFE solutions in the presence of benzene-d6 vapor, the adduct + formed efficiently (Figure 36a). Figure 36b shows the daughter ion spectrum of [24(benzene-d6)] , which clearly displays baseline separation of the individual isotopomers of the adduct.
+ [24(water)]+ [24(TFE)] a) m/z = 601 m/z = 683 b)
+ +
Cl Cl Cl Cl N N CH3 CH3 Pt Pt D6 N N Cl Cl Cl Cl Intensity
+ Intensity + [24] [24(benzene-d6)] m/z = 583 m/z = 667
580 600 620 640 660 680 m/z 664 666 668 670 672 m/z + Figure 36. a) ESI-MS spectrum of [24] electrosprayed from TFE with benzene-d6 vapor in the spray region recorded on a TSQ Quantum instrument with a resolution of 0.3 Da in Q1. b) Daughter ion spectrum of + [24(benzene-d6)] recorded with a resolution of 6.0 Da in Q1 and 0.2 Da in Q3 in comparison with the isotope pattern calculated for C23H15D6Cl4N2Pt.
When [24(benzene-d6)] was isolated in Q1 it was characterized in Q3 by its mass and its isotope pattern which both were in perfect agreement with the calculated values (Figure 36b). CID was subsequently accomplished by collision with Ar at 6 and 8 eV collision energy (laboratory frame). Deuterium isotopic scrambling was instantly recognized from the number of isotopomers of [24(benzene-d6)] isolated in Q1 (7) in comparison with the number of isotopomers detected in Q3 for the elimination products 24 and 25 (10) (Figure 37a). Integration of the individual isotope signals revealed that the observed isotope pattern deviated from the expected isotope pattern. The expected isotope pattern was simulated starting from the observed isotope pattern of [24(benzene-d6)] before CID and assuming a hypergeometric distribution of the 3 H and 6 D atoms. xxiv The retention time of the ions was calculated by taking into account the experimental conditions. After desolvation [24]+ enters Q0 (24-pole) which contains 7 – 9 mtorr of benzene.
49 3 Scrambling in a Cationic PtII Complex in the Gas Phase
From the comparison depicted in Figure 37b, it was deduced that deuterium accumulated preferentially in the departing benzene, i.e. the observed pattern for partly deuterated 24 is shifted to lower masses than expected for complete statistical scrambling (Figure 37b).
found a) + + + b) statistical 0.20 Cl Cl Cl Cl Cl Cl N N C H D CHnD3 - n 6 m 5 -m N CH3 Pt Pt Pt D N 6 N Cl Cl N Cl Cl 0.15 Cl Cl
[24-d]+ [25-d]+ + [24(benzene-d6)] m/z = 667 ed Intensity 0.10 Intensity Normaliz
0.05
0.00 580 590 650 660 670 m/z 580 590 650660670 m/z + Figure 37. a) Daughter ion spectrum of [24(benzene-d6)] recorded on a TSQ Quantum instrument after selection in Q1 (6 Da) and CID with 1.5 mtorr of Ar at 8 eV collision energy (laboratory frame, Q2). b) + + + Comparison of the observed isotope distributions of partly deuterated [24] , [25] , and [24(benzene-d6)] (red) in comparison with a fully statistical distribution (green).
To determine the origin of this observation a complementary experiment was executed. By protonolysis of 17 in TFE with HBF4 26 was obtained, which differs from 24 only in the deuteration of the PtII methyl group. When the solution was electrosprayed in the presence of 1,3,5- xxv trideuterobenzene vapor the adduct [26(benzene-d3)] was readily detected in Q1. The spectra recorded after CID under the conditions mentioned above also displayed extensive scrambling (Figure 38a).
found a) b) statistical + + + 0.20
Cl Cl Cl Cl Cl Cl N N C H D N CHnD3 - n 6 m 5 -m CD3 Pt Pt Pt D3 N Cl N Cl N Cl 0.15 Cl Cl Cl
+ + + [24-d] [25-d] [26(benzene-d3)] lized Intensity Intensity m/z = 667 0.10 Norma
0.05
0.00 580 590 650 660 670 m/z 580 590 650 660 670 m/z + Figure 38. a) Daughter ion spectrum of [26(benzene-d3)] recorded on a TSQ Quantum instrument after selection in Q1 (6 Da) and CID with 1.5 mtorr of Ar at 6 eV collision energy (laboratory frame, Q2). b) + + + Comparison of the observed isotope distributions of partly deuterated [24] , [25] , and [26(benzene-d3)] (red) in comparison with a fully statistical distribution (green).
xxv + + The observed isotope pattern of [26(benzene-d3)] differs from that of [24(benzene-d6)] because upon protonolysis with HBF4 the proton scrambled partly into the remaining Pt(II) methyl group. Consequently the potential statistical outcome of the experiment was not calculated on the basis of the isotope pattern of C23H15D6Cl4N2Pt. The hypergeometric distribution of D and H in the products was calculated from the observed isotope pattern of + [26(benzene-d 3)] without CID.
50 3.2 Installation of a Collision Cell in an Ion Trap
Again, the observed isotope distribution deviated from the isotope distribution expected for complete statistical scrambling. In this set of experiments an accumulation of deuterium in the departing methane was observed (Figure 38b).
3.2 Installation of a Collision Cell in an Ion Trap
+ With the goal to investigate the time dependence of the scrambling processes in [24(benzene-d6)] + and [26(benzene-d 3)] in the gas phase, a reaction chamber was installed in a Finnigan LCQ deca ion trap. The storage time of ions in such an instrument can be varied from 0.03 ms to 10000 ms. Two PEEK half shells (length: 18 mm, inner diameter: 9.7 mm) were placed around the ion guide in the first octopole region and joined together with Kapton film (Figure 39a). One half shell was chamfered on 5 mm while on the opposite end of the other half shell a Teflon tube was attached to be used as a gas inlet. This tube was connected to a ⅛ inch stainless steel tube, which was glued into a hole in the 4-pin feedthrough for the first octopole leads (part no. 97000-62004) in the top cover plate. Swagelok connectors were used for further connections to valves and to the forepump.
a) b)
Figure 39. a) Schematic representation of the multipole ion guides and the Paul-trap of a Finnigan LCQ deca ion trap. b) Block diagram of the vacuum regions of a LCQ ion trap with the corresponding pressures and pressure gauges. The first octopole region is colored in pink.
A split-flow turbomolecular pump provided the vacuum for the first octopole and the analyzer regions of the vacuum manifold (Figure 39b). The interstage port of the pump, which evacuated the first octopole region, was rated at 125 l · s-1. The high vacuum port of the pump, which evacuated the analyzer region, was rated at 200 l · s-1. Under normal operating conditions the pump provided a vacuum of approximately 10-3 torr in the first octopole region, and 2 · 10-5 torr in the analyzer region. A maximum pressure of 5 · 10-4 torr was tolerated in the first octopole region.
Complex [24]+ was electrosprayed by means of a nanospray sourcexxvi from a TFE solution. With -5 benzene-d6 present in the first octopole region (measured pressure 2.6 · 10 torr in the analyzer + region), [24(benzene-d6)] was formed in detectable amounts. One isotopomer (m/z = 667) was isolated and collided with He to result in the formation of partly deuterated [24]+ and [25]+ (Figure 40). Unfortunately, the partial pressure of benzene-d6 in the analyzer region was sufficiently high for the formation of new benzene-d6 adducts. No precise analysis of the reaction rate of deuterium isotopic scrambling was attempted because the exchange of benzene-d6 from the gas phase with xxvi A nanospray source replaced the regular source in order to reduce contamination by acetonitrile.
51 3 Scrambling in a Cationic PtII Complex in the Gas Phase
+ + [24(benzene-d6)] could not be precluded. In addition, the signal intensity of [24(benzene-d6)] was limited because the benzene-d6 deteriorated the mass resolution of the instrument. This can be seen + + + from the strongly varying width of the signals for [24] , [25] , [24(benzene-d6)] , and + [25(benzene-d6)] .
+ + + +
Cl Cl Cl Cl Cl Cl Cl Cl N CH D N N N n 3 -n C6HmD5 - m CH3 C6D5 Pt Pt Pt Pt D6 D6 N Cl N Cl N N Cl Cl Cl Cl Cl Cl
+ + d + + [24-d] [25-d] [24(benzene- 6)] [24-d (benzene-d6)] ensity Int
500 750 m/z + Figure 40. Daughter ion spectrum of [24(benzene-d6] , formed in the first ion guide of a LCQ ion trap, after mass selection of m/z = 667 and CID with He. Benzene-d6 is present in the analyzer region because the vacuum chambers for the first octopole region and of the analyzer region are not sufficiently separated.
3.3 Theoretical Study of the Deuterium Isotopic Scrambling
Density functional theory (DFT) provides a tool to investigate the intramolecular deuterium isotopic scrambling process in [24(benzene)]+ in more detail with modest computational cost, mainly with respect to the relative energies and the structures of intermediates involved. A similar investigation was conducted by Tilset and co-workers for the C-H activation of methane by [19(water)]+.58 From calculations, performed with ADF and Gaussian94 on the B3LYP level of theory employing LANL2DZ and 6-31G** basis sets they found that without major loss of structural accuracy the methyl and aryl groups of the diimine ligand could be replaced by protons. As a result of their study they proposed that C-H activation by PtII occurred through σ-methane complexes followed by an oxidative addition/reductive elimination mechanism. In addition they observed that all species were significantly stabilized by coordination of one solvent.
In order to determine the reliability of the subsequent computations, the investigations were started with a comparison of structural parameters of four model compounds, a and b calculated with Gaussian98,160 with the x-ray structures of 16 and the cation [(diimine)PtMe(acetonitrile)]+ 208 (diimine = ArN=CMeCMe=NAr, Ar = 2-(OSi(i-Pr3)-6-(CH3)-C6H3) D. The results are presented in Table 8. The calculations were performed with Gaussian98160 on the B3PW91178,181 level of theory employing LANL2DZ basis sets209 for Pt and N with the Los Alamos effective core potential
52 3.3 Theoretical Study of the Deuterium Isotopic Scrambling
for Pt (index L, column 3 and 6), or SDD basis sets184 for Pt and N with the corresponding effective core-potential for Pt (index S, column 4 and 7). For C and H the 6-31G** basis sets185 were used (see ch. 6.2). The transition states were located with the Gaussian98 implemented transition state optimization QST2, which requires only the optimized structures of educt and product that are connected by the transition state.210,211
a b Table 8. Comparison of the structures aL and bL, ,calculated with the LANL2DZ basis sets for Pt and N, and aS and bS, calculated with the SDD basis sets, with the data obtained from x-ray analyses. The deviations ∆ are calculated from all values for each basis set (14 bonds, 6 angles).
Parameter 16 aL aS D bL bS ∆L ∆S Bonds [Å] Pt1-N2 2.076(5) 2.382 2.093 1.955(4) 2.187 1.988 Pt1-N3 2.076(5) 2.382 2.093 2.098(3) 2.427 2.154 Pt1-C4 2.007(7) 2.111 2.037 2.034(5) 2.084 2.037 Pt1-L5 2.007(7) 2.111 2.037 1.985(4) 2.215 1.981 N2-C6 1.284(9) 1.284 1.306 1.285(6) 1.282 1.299 N3-C7 1.284(9) 1.284 1.306 1.266(6) 1.280 1.294 C6-C7 1.489(11) 1.475 1.445 1.452(7) 1.482 1.460 0.123 0.023 Angles [°] C4-Pt1-L5 87.9 87.1 90.8 88.9(2) 88.9 88.1 N3-Pt1-N2 74.9 69.7 75.2 78.2(1) 72.2 76.0 N2-C6-C7 113.8(3) 119.3 115.1 116.7(4) 119.5 115.2 3.8 1.5
Yielding results in better agreement with x-ray data, the SDD basis set was employed for the subsequent investigations (Pt, N). The mean deviations ∆S of all calculated structural parameters from the values from x-ray analysis were 0.023 Å (bonds) and 1.5° (angles). All energies were ZPE-corrected by applying a frequency correction of 0.9772180 and computed relative to the benzene adduct (c). Furthermore, all structures were cationic and all ground state structures displayed no imaginary frequency.
53 3 Scrambling in a Cationic PtII Complex in the Gas Phase
Benzene was found to coordinate approximately orthogonal with respect to a plane defined by N1, N2, and Pt in π-(C,C) fashion. A structure with the aromatic ring tilted towards the PtII methyl group (c) led directly to the transition state (d) for the formation of the PtIV hydride (e).xxvii The transition state displayed one imaginary frequency (521i cm-1), which corresponded to the motion of a benzene hydrogen towards the PtII center. Another transition state (f) connected e and the phenyl complex (g) with a η2-(C,H)-bound methane. Transition state f also displayed one imaginary frequency (540i cm-1), connected with the motion of the platinum-bound hydrogen towards the methyl group. Removal of benzene and methane from c and g, respectively, resulted in the three- coordinate structures h and i. The structures c – i are shown in Figure 41 with their corresponding energies.
150 144
i 100 h d f 102 93 93
nergy [kJ/mol]nergy 81 E
50 Relative Relative e 48
c g 0 0
Reaction ξ Figure 41. Computed energy surface of the scrambling process in [24(benzene)]+ based on model systems. The energies of h and i comprise a molecule of benzene and methane, respectively. All structures are -1 -1 ground state structures except for d and f that are transition states (νd = 521i cm , νf = 540i cm ).
It was found that a barrier of ∆G≠ = 93 kJ/mol separated the benzene adduct from the five coordinate PtIV hydride. A slightly lower barrier (0.1 kJ/mol) was determined for the transformation to form the methane adduct. Although this difference in ∆G≠ for the formation of the products c and g starting with the hydride e is smaller than observed in the experiments, the calculations predict the
xxvii A similar structure, 17 kJ/mol lower in energy, was found with the benzene ring tilted towards the N-H of the ligand. However this structure seemed to be an artifact caused by the absence of a substituent on the ligand nitrogen. Furthermore, this structure did not lead to a transition state in the hydride formation and was therefore discarded.
54 3.4 Discussion
correct relative position of the equilibrium of deuterium isotopic scrambling.xxviii Furthermore, high free dissociation enthalpies ∆G of 144 and 102 kJ/mol benzene and methane, respectively, to form 8 an unsaturated d -ML3 fragment evidence the fact that adducts of methane (g) are stable in the gas phase and posses a life-time of at least ms.
3.4 Discussion
A possible mechanism accounting for the observed deuterium isotopic scrambling in + [24(benzene-d6)] is depicted schematically in the upper line of Scheme 16. The DFT calculations illustrate that without additional stabilization, e.g. by an additional ligand L, elimination of a hydrocarbon from the coordination sphere of PtII is very unlikely in the gas phase.xxix Given the retention time of ions in Q0 on the order of ms and the calculated rate of deuterium isotopic scrambling, the scrambling process in the experiments described in ch. 3.1 was expected to go to statistical completion if no other reactions than those shown in the upper line of Scheme 16 interfered with the reaction. However, completion was not observed and potential reasons for this experimental observation will be discussed in the following section.
+ + + + + Ar Ar Me CH Ar Ar D Ar Me CH D N 3 Me Me Me N 2 Me Pt N CH3 N N CH2D Me Pt N Me Pt Me Pt CH3 Me Pt N Ar N N N Ar Ar Ar Ar D H H D6 D5 D5 D5 D5 I II III IV V
+ L + L + L + L + L
+ + + + + Ar L L Ar Me Ar Me N CH3 Me Ar Ar Me N CH2D Me Pt N CH3 Me N L N CH2D Me Pt N Me Pt Me Pt Me Pt N L Ar L N N N Ar Ar Ar Ar D H
D6 D5 D5 D5 H D5 + CH D + 3 +
+ Scheme 16. Equilibria involved in the scrambling of one deuterium (or hydrogen) of [24(benzene-d6)] . The scrambling process is quenched in the presence of an additional ligand L.
+ As can be seen from Scheme 16, a kinetically exact description of the reaction of [24(benzene-d6)] leading to the various isotopomers of [24]+ and [25]+ is extremely complex. The reaction leading to + [25-d3] , for instance, comprises at least 12 different equilibria of the type shown in Scheme 16 with 6 possible products differing only by their regiochemistry.xxx Furthermore, primary and secondary isotope effects render the situation even more complex. Nevertheless, some conclusions can be drawn.
xxviii The branching ratio of 81:19 at 343 K in favor for methane reductive elimination is equivalent to a ∆(∆G≠) = 4.1 kJ/mol. xxix The high barrier for elimination of the hydrocarbon also argues in favor of an associative ligand exchange mechanism and for the elimination being the rate-determining step of the reaction. xxx Each equilibrium depicted in Scheme 16 comprises oxidative addition and the microscopic reverse, reductive coupling. Elimination of hydrocarbons occurs in separate steps in the presence of an additional ligand L.
55 3 Scrambling in a Cationic PtII Complex in the Gas Phase
The influence of an isotope effect on the non-statistical course of the scrambling process is ruled out + + by the complementary experiments with [24(benzene-d6)] and [26(benzene-d3)] . If incompletion were due to isotope effects both experiments would lead to identical product distributions.
Ligation of one of the intermediates could potentially also influence deuterium isotopic scrambling + by shutting down the reaction irreversibly. The isomerizing adduct [24(benzene-d6)] in Scheme 16 + can be trapped by an additional ligand L at two different stages. While trapping [24(benzene-d6)] as an adduct (I, III, V) leads to four-coordinate PtII compounds with a free hydrocarbon, trapping at the hydride stage (II, IV) leads to six-coordinate PtIV species. Tilset and co-workers exploited this feature to determine the kinetic site of protonation of PtII complexes such as 12 by trapping irreversibly the intermediates corresponding to II and III with acetonitrile. When they varied the size of the o-substituent from hydrogen to iso-propyl, they found that the amount of trapped intermediate decreased from 100 to 90 %. No speculation was made about the remainder of the 2 2 + - protonated complexes. Norris and Templeton found that for [(κ -HTp')Pt(η -C6H6)(H)] [BArF] at temperatures below 220 K the solvent adduct of the PtIV hydride was not only preferred over the PtII benzene adduct but also over elimination.212 However, as no solvent molecules are present in the mass spectrometer and no unidentified adducts are detected under reaction conditions, trapping at either PtII or PtIV by an external ligand can be ruled out.
In ch. 2 the existence of one unidentified, kinetically significant intermediate in the C-H activation of benzene by [24(water)]+ was established. On the reaction coordinate this intermediate precedes the rate-determining transition state and has to occur between structures IV and V of Scheme 13. Furthermore it was established that this intermediate did not contain a solvent molecule.
A possible explanation that accounts for both findings, non-statistical deuterium isotopic scrambling and an unidentified kinetically significant intermediate, might be trapping of the unsaturated metal center by internal ligation. That is, the scrambling process competes with a process in which the ortho substituent of the Schiff base occupies the sixth site of the hydride structures II and IV. Once the chlorine coordinates to PtIV the reaction is terminated and hence deuterium isotopic scrambling is incomplete. This would also account for the observations described by Tilset and Templeton.xxxi Furthermore, sufficient activation energy to overcome the barrier for elimination to form 24, benzene, 25, and methane is only available after collision with Ar.
3.5 Conclusion
Deuterium isotopic scrambling was observed for partly deuterated PtII complexes in the gas phase. Although the scrambling process occurred at rates sufficient to achieve fully statistical scrambling during the isolation time of the ions in the mass spectrometer, no statistical distribution of H and D within the phenyl and methyl ligand was observed. This was interpreted by assuming two competing reactions, one being deuterium isotopic scrambling, the other internal trapping of the intermediate PtIV hydride by an o-chlorine atom from the ligand. The findings are explained by the assumption that Pt-Cl bond dissociation is much slower than deuterium isotopic scrambling.
xxxi In ref. 31 the appearance of a second complex in addition to [20(water)]+ in NMR experiments was attributed to + II + [20(TFE-d3)] because coordination of the anion to Pt was ruled out. However, [20(TFE-d3)] was not unambiguously identified and coordination of the ortho substituent of the ligand was not considered.
56 4.1 Syntheses and Materials
4 PtII-Acetato Complexes in Catalytic C-H Activationxxxii
In recent years, the mechanism of the catalytic conversion of methane to methanol by chloroplatinate salts, the so-called Shilov reaction11,213 originally reported in 1972,18 has been investigated using well-defined homogeneous PtII complexes as model systems.6,10,214 However, with a few known exceptions, 215,216,217,218 these well-defined PtII complexes do not mediate C-H activation leading to functionalized hydrocarbons. They are used primarily to model individual reaction steps. Several groups reported catalytic C-H activation and functionalization of methane under strongly oxidizing, electrophilic conditions with complexes of Pt and other metals.79,80,219,220,221,222,223,224 A comparable C-H activation process by which arenes can be deuterated (or tritiated) catalytically by heterogeneous Pt systems,15,60 or by homogeneous chloroplatinate salts,16,17,68 has received much less attention, despite the likelihood that all three processes are mechanistically related. In this process benzene is reacted with D2O in the presence of II acetic acid-d4 and Pt(0) or Pt to produce deuterated benzene.
4.1 Syntheses and Materials
II In order to study the role of acetic acid-d4 in the Pt -catalyzed C-H activation of benzene, a xxxiii + - precisely solid acid was needed. The acid [(Et2O)2H] [BArF] (BArF = tetrakis(3,5- bis(trifluoromethylphenyl)borate)) (27, HBArF) was synthesized by a procedure developed by Brookhart and independently by Taube (Scheme 17).225,xxxiv NaBArF was placed in a Schlenk-tube and dissolved in ether. Subsequently the tube was pressurized once with HCl. After filtration, the white, solid product was precipitated with hexane. - - F C CF F3C CF3 3 3 F C CF F3C CF3 3 3
+ HCl/Et 2O + Na B [(Et 2O)2)H] B - NaCl
F3C CF3 F3C CF3
F3CCF3 F3CCF3
27 Scheme 17. Synthesis of HBArF by reaction of NaBArF in ether with an excess of HCl.
Two compounds were needed for control experiments. The cationic PtII-acetato complex [28]+ was obtained by protonolysis of 16 with HBArF in TFE and subsequent reaction with excess acetic acid 1 + (Scheme 18). H-NMR spectra in TFE-d3 [28] indicated a symmetric structure, in which the acetato ligand is bound in an η2 mode.
xxxii Parts of this chapter have been published in ref. 148. xxxiii HBArF is a white powder, which can be conveniently purified by recrystallization. In addition it can be weighted very easily due to its high molecular weight, which results in more precise concentrations of the acid in the reaction mixture than commercially available acid solutions. xxxiv The starting material NaBArF was generously provided by Ruedi Hartmann. First samples of HBArF for control experiments were generously provided by Tillman J. Geldbach.
57 4 PtII-Acetato Complexes in Catalytic C-H Activation
+
Cl Cl Cl Cl N 1. HBArF, TFE N 2. acetic acid O - Pt Pt [BArF] N N O Cl Cl Cl Cl
16 28 Scheme 18. Synthesis of the cationic PtII-acetato complex [28]+.
A route leading to [28]+ in two steps (instead of four) was attempted. To this purpose, the cluster complex Pt4(OOCH3)8 (29) was synthesized from the reaction of PtCl4 and silver acetate in acetic acid (Scheme 19).226 However, the cluster 29 did not react with ligand 6 to yield [28]+[(OAc)]- under various reaction conditions, presumably because the 1,2-diimine ligand 6 was not capable of breaking up the tetranuclear cluster.227,228
acetic acid 6, CH 2Cl2 PtCl 4 + AgOOCH 3 [Pt(OAc) 2] 4 29 Scheme 19. Synthesis of the tetrameric PtII cluster 29.
Reproducible results of the C-H activation process were only obtained with benzene (OEKANAL xxxv grade) from Riedel de Haën and acetic acid-d4 in "100%" quality from Dr. Glaser AG. A mixture of cis- and trans-decalin was purified according to literature procedures203 and degassed immediately prior to use. All glassware and stir bars were washed with alcoholic KOH, water, acetone, and finally p.a. grade MeOH and dried in a 160 °C oven. The hot and dry glassware was then allowed to cool in a nitrogen-filled glovebox (M-Braun Labmaster 130). All chemical transformations were performed under exclusion of air.
4.2 Observation of Catalytic C-H Activation of Benzene by PtII
In order to establish whether or not [24]+, or the respective solvent adducts, performed catalytic deuterations similar to those described by Garnett and Hodges, two experiments were conducted. First, 16 was dissolved in dry benzene and protonated with HBArF to yield [25]+. Subsequently, heavy water was added and the temperature was raised to 45 °C. The reaction progress was followed by means of ESI-MS. Samples were taken immediately after addition of D2O, after 17 and 43 h and were quenched and diluted by addition of acetonitrile. These solutions were electrosprayed. Clean spectra of partly deuterated [25(acetonitrile)]+ were obtained, displaying progressive incorporation of deuterium into the phenyl residue (Figure 42a). Second, HBArF was + added to a deep blue solution of 18 in benzene-d6. The exchange of the phenyl residue of [25] with xxxv Acetic acid-d4 from Cambridge Isotopes contained an unspecified impurity, which could not be reliably removed by standard purification procedures.
58 4.3 Acetic Acid-d4 as Deuterium Source
benzene-d6 from solution was again monitored by electrospraying the acetonitrile adducts of the various isotopomers. The spectra shown in Figure 42b clearly proved that the undeuterated residue was partly exchanged with its deuterated analog after 6 h.
+ + Cl Cl N C6HnD5 - n Cl Pt a) b) Cl N NCCH3 N Cl C6HnD5 - n Cl Pt N NCCH3 Cl [25-d (acetonitrile)]+ Cl
[25-d (acetonitrile]+ Intensity Intensity
[25(acetonitrile)]+ m/z = 686
680 685 690 695 700 m/z 680 685 690 695 700 m/z + Figure 42. a) Reaction of [25] with D2O at 45 °C after mixing, after17, and 43 h (bottom to top). Products were analyzed as acetonitrile adducts by ESI-MS in a Finnigan TSQ Quantum instrument. b) Reaction of + [25] with benzene-d6 immediately after mixing (lower spectrum) and 6 h (upper spectrum). Products were analyzed as acetonitrile adducts as before.
However, even at temperatures of over 100 °C and reaction times of several days no deuterated benzene was detected by means of GC-MS in the headspace vapor of the reactions. Although one would normally expect an acceleration of the C-H activation upon raising the reaction temperature, appreciable amounts of partly deuterated benzene were still not detected. According to the rate law determined by Tilset and co-workers for the stoichiometric C-H activation of benzene by [20(water)]+, the reaction rate is inversely proportional to the concentration of water.31 Hence, the overall effects of heating, increased reaction rate and increased water concentration,205b seemed to cancel out and the reaction was probably too slow to produce macroscopic amounts of deuterated benzene.
4.3 Acetic Acid-d4 as Deuterium Source
The situation changed when D2O was replaced by acetic acid-d4. In a first experiment, 16, dissolved in benzene, was reacted with a slight excess of HBArF, yielding exclusively [25]+ within 10 minutes. Subsequent addition of acetic acid-d4 caused the orange solution to turn yellow within 5 minutes. When the reaction mixture was heated to 85 °C progressive deuteration of the bulk benzene was detected in the headspace vapor of the reaction by means of GC-MS (Figure 43).
59 4 PtII-Acetato Complexes in Catalytic C-H Activation
15
94 h 22 h 0 h
10 Intensity [%] Intensity
5
0 80 82 84 m/z Figure 43. Isotopomer intensities in the mass spectrum of benzene taken at various times during the + treatment with acetic acid-d4 and Pt complex [24] (8.0 mM) with concentration of [benzene] = 10.0 M and [acetic acid] = 2.0 M. The calculated TON based on the conversion acetic acid-d4 → acetic acid-d3 were 1044 (22 h) and 5199 (94 h).
A number of control experiments under identical experimental conditions but in the absence of platinum were performed. No deuteration of bulk benzene was observed, which ruled out the possibility that some type of Friedel-Crafts chemistry was responsible for the isotope exchange.
4.4 Kinetics of the Catalytic C-H Activation of Benzene by a PtII Complex
The effect of acetic acid-d4 on the catalytic deuteration of benzene was investigated in a kinetic study. To this purpose, a stock solution of [25]+ (0.00127 M) was prepared from 16, adding a slight excess of HBArF, and benzene as before. Aliquots were taken from the stock solution, from which + nine samples were prepared with various concentrations of [25] , benzene, and CD3COOD. In each sample, the final volume was brought to 1.4 ml by addition of decalin (a high-boiling solvent inert to C-H activation under these conditions) to produce samples that are less than 1 mM in [25]+. The nine sample vessels were sealed with septa and placed in a heated bath (85 ± 1.5 °C). From each vessel, five samples of the headspace vapor, taken periodically with a gastight syringe over a period of 23 h, were analyzed by GC-MS. The extent of deuteration was calculated from the mass spectrum of benzene by integration of the isotopic peaks and subtraction of the integrated intensities for unlabeled benzene with natural abundance 13C and 2H. The turnover number (TON) was calculated there from in terms of consumption of acetic acid-d4. Consumption of C6H6 was also determined. A further parameter calculated from the mass spectrometric intensities is the so-called Anderson and Kemball's M value,62 which is effectively a measure of the number of deuteriums incorporated into benzene in one "pass" through the catalytic cycle.
The formation of Pt clusters in the reaction mixtures was not detected by taking aliquots of the reaction mixture after 24 h, dilution with 2,2,2-trifluoroethanol (TFE), and analysis by electrospray. No signals for the protonated, free ligand were observed, which would appear if colloidal Pt were to
60 4.4 Kinetics of the Catalytic C-H Activation of Benzene by a PtII Complex
be formed. On the other hand, addition of excess ligand to these samples gave rise to strong signals of the protonated ligand, ligated Na+, and dimers thereof, while the original Pt signals remained unchanged, which indicates that free ligand would have been detected if present.
For the kinetic experiments in solutions diluted with decalin, the TON was run up to a maximum of ~1500 so that the kinetics would not be significantly affected by a build-up of deuterated benzene or undeuterated acetic acid.229 The nine separate sample runs can be organized into three sets in which, + for each set, the concentration of only one species, [25] , benzene, or acetic acid-d4, varies in that set. The results are plotted in Figure 44a, b, and Figure 45. From the data it becomes clear that the rate of deuteration is first-order in catalyst, first-order in benzene, and inverse first-order in acetic acid. From the slope in the plots of each of the nine data sets, the rate constant k358K in eq 4.1 can be -3 -1 calculated (Table 9). The average for the nine samples, k358 = (4.20 ± 1.02) · 10 s (2σ error bounds), contains all nine of the individual values within the stated bounds.
ν = k358 · [catalyst] · [benzene] / [acetic acid] (eq 4.1)
-5 a) 0.00 [benzene] = 3.21 M; [acetic acid] = 3.78 M; T = 85 °C b) 0.00 [cat] = 9.07 x 10 M; [acetic acid] = 3.78 M; T = 85 °C
-0.01
) -0.04 0 ) 0 /[AcOD] t
/[AcOD] -0.02 t
-0.08 ln ([AcOD]ln ln ([AcOD] -0.03 [catalyst] = 9.06 x 10-5 M [benzene] = 0.80 M [catalyst] = 1.81 x 10-4 M [benzene] = 1.61 M [benzene] = 3.21 M [catalyst] = 3.63 x 10-4 M -0.12 -0.04
025000 50000 75000 0 25000 50000 75000 Reaction Time [s] Reaction Time [s]
Figure 44. a) Semilogarithmic plot of the concentration of CD3COOD against time for three runs in which the catalyst concentration is varied as benzene and acetic acid concentrations are held constant. b) Semilogarithmic plot of the concentration of CD3COOD against time for three runs in which the benzene concentration is varied as catalyst and acetic acid concentrations are held constant.
The GC-MS data for benzene isotopomers give two rates, the ratio of which contains important mechanistic information. From the background-corrected intensities of the various C6H6-nDn species, one calculates the total rate of deuteration, i.e. the rate by which acetic acid-d4 is converted to acetic acid-d3, and the rate of consumption of C6H6. If only one deuterium were incorporated in a single pass through the catalytic cycle, the two rates would be identical. If, on the other hand, several deuteriums were to be incorporated in a single pass, the rate of deuteration would exceed the 62 rate of C6H6 consumption. This very ratio, termed M by Anderson and Kemball, is given in Table 10 for each measurement. The total of 45 measurements show consistent M values, with a mean value of M = 2.6 ± 0.3 (2σ error bounds).
61 4 PtII-Acetato Complexes in Catalytic C-H Activation
Table 9. Calculated rate constants at 358 K of the nine kinetic runs displayed in Figure 44a, b, and Figure 45.
[cat] = 9.07 x 10-5 M; [benzene] = 3.21 M; T = 85 °C -3 -1 0.00 Rate constants k358 [10 · s ] error
kcat1 2.75 0.89
kcat2 3.09 0.44 ) 0 -0.03 kcat3 3.99 0.22 /[AcOD] t kbenzene4 5.92 0.87
ln ([AcOD] -0.06 kbenzene5 5.45 0.45
[acetic acid-d ] = 5.04 M 4 kbenzene6 4.83 0.24 [acetic acid-d4] = 2.52 M
[acetic acid-d4] = 1.26 M kacid7 4.12 0.14
025000 50000 75000 kacid8 4.07 0.22 Reaction Time [s]
Figure 45. Semilogarithmic plot of the concentration of CD3COOD kacid9 3.53 0.42 against time for three runs in which the acetic acid concentration is varied as benzene and catalyst concentrations are held constant. final k358k 4.20 0.51
4.5 Gas-Phase Investigations of the Deuteration of Benzene catalyzed by PtII
With the intention to identify the Pt species involved in the C-H activation ESI-MS experiments were conducted with a Finnigan MAT LCQ deca ion trap and a Finnigan TSQ Quantum triple quad electrospray ionization mass spectrometer. To this purpose, 0.1 mg of 16 (1.7 · 10-7 mol) was dissolved in 1.5 ml of benzene to yield a homogeneous green solution. A slight excess of HBArF was added,xxxvi and after 10 minutes [25]+ had formed. Solutions thus prepared were stable and showed no detectable loss of [25]+ even after heating at 50 °C for more than 100 h. The clear orange + solution of [25] was then treated with 1.5 ml of CH3COOH (molar ratio Pt to benzene to acetic acid of 1:101 160:158 760) to produce the yellow solution of acetato complex [28]+ within 1-5 min at room temperature. The mixed benzene/acetic acid solution, approximately 56 µM in catalyst, was introduced directly into the ESI-MS spectrometer by introducing a sampling capillary directly into the reaction vessel to which an overpressure (1 bar) of nitrogen was applied.
xxxvi Solutions prepared with benzene purified according to ref. 203 and freshly prepared HBArF do not show the unspecified irreversible decay (over the course of a few hours) reported in ref. 147. Similarly stable solutions of the + cationic complex [25] can be prepared with HBF4 if the benzene solution is extracted several times with water immediately after activation is complete. Evidently, impurities, or perhaps excess acid, in the commercial HBF4 solution are responsible for decomposition of the cationic complexes.
62 4.5 Gas-Phase Investigations of the Deuteration of Benzene catalyzed by PtII
Table 10. Results from nine kinetic runs of the deuteration of benzene catalyzed by [24]+ (total volume:1.4 ml; reaction temperature 85 °C). Initial concentrations of catalyst, benzene, and acetic acid-d4 and are listed in in the first three columns. The extent of conversion of CD3COOD and C6H6 is given in columns 6 and 7.
[catalyst] [benzene] [acetic acid-d4] rxn time [s] TON conversion conversion benzene M [mol · l-1] [mol · l-1] [mol · l-1] acid [mol · l-1] [mol · l-1] 9.06E-05 3.21 3.78 8160 335 4.25E-05 1.82E-05 2.34 9.06E-05 3.21 3.78 13200 467 5.92E-05 2.17E-05 2.73 9.06E-05 3.21 3.78 18960 511 6.48E-05 2.44E-05 2.66 9.06E-05 3.21 3.78 27300 570 7.23E-05 2.56E-05 2.83 9.06E-05 3.21 3.78 80460 1011 1.28E-04 4.92E-05 2.61 1.81E-04 3.21 3.78 8400 335 8.50E-05 3.56E-05 2.39 1.81E-04 3.21 3.78 13500 456 1.16E-04 4.55E-05 2.54 1.81E-04 3.21 3.78 19200 557 1.41E-04 5.28E-05 2.68 1.81E-04 3.21 3.78 27540 702 1.78E-04 6.63E-05 2.68 1.81E-04 3.21 3.78 80700 1084 2.75E-04 1.07E-04 2.57 3.63E-04 3.21 3.78 8760 286 1.45E-04 5.83E-05 2.49 3.63E-04 3.21 3.78 13740 356 1.81E-04 6.86E-05 2.63 3.63E-04 3.21 3.78 19500 496 2.52E-04 9.32E-05 2.70 3.63E-04 3.21 3.78 27840 634 3.22E-04 1.22E-04 2.63 3.63E-04 3.21 3.78 81240 1173 5.95E-04 2.32E-04 2.56 9.07E-05 0.80 3.78 9360 113 1.44E-05 6.87E-06 2.09 9.07E-05 0.80 3.78 14160 199 2.53E-05 1.01E-05 2.50 9.07E-05 0.80 3.78 19860 203 2.58E-05 9.52E-06 2.71 9.07E-05 0.80 3.78 28080 301 3.82E-05 1.42E-05 2.69 9.07E-05 0.80 3.78 81420 487 6.18E-05 2.40E-05 2.57 9.07E-05 1.61 3.78 9600 237 3.01E-05 1.35E-05 2.23 9.07E-05 1.61 3.78 14400 364 4.62E-05 1.81E-05 2.55 9.07E-05 1.61 3.78 20100 395 5.01E-05 1.92E-05 2.61 9.07E-05 1.61 3.78 28380 508 6.44E-05 2.48E-05 2.59 9.07E-05 1.61 3.78 81600 906 1.15E-04 4.51E-05 2.55 9.07E-05 3.21 3.78 9840 483 6.13E-05 2.69E-05 2.28 9.07E-05 3.21 3.78 14640 538 6.84E-05 2.58E-05 2.65 9.07E-05 3.21 3.78 20520 653 8.29E-05 3.09E-05 2.69 9.07E-05 3.21 3.78 28800 715 9.08E-05 3.33E-05 2.73 9.07E-05 3.21 3.78 82020 1568 1.99E-04 7.84E-05 2.54 9.07E-05 2.41 1.26 10380 390 4.95E-05 1.89E-05 2.62 9.07E-05 2.41 1.26 15180 392 4.97E-05 1.71E-05 2.90 9.07E-05 2.41 1.26 20880 480 6.10E-05 2.35E-05 2.60 9.07E-05 2.41 1.26 28980 546 6.94E-05 2.50E-05 2.77 9.07E-05 2.41 1.26 82200 1049 1.33E-04 5.04E-05 2.65 9.07E-05 2.41 2.52 10620 336 4.27E-05 1.70E-05 2.51 9.07E-05 2.41 2.52 15420 380 4.83E-05 1.71E-05 2.83 9.07E-05 2.41 2.52 21060 492 6.25E-05 2.58E-05 2.42 9.07E-05 2.41 2.52 29280 551 6.99E-05 2.53E-05 2.76 9.07E-05 2.41 2.52 82380 1036 1.32E-04 5.11E-05 2.57 9.07E-05 2.41 5.04 10860 320 4.06E-05 1.79E-05 2.27 9.07E-05 2.41 5.04 15660 307 3.90E-05 1.65E-05 2.37 9.07E-05 2.41 5.04 21360 407 5.17E-05 1.95E-05 2.65 9.07E-05 2.41 5.04 29580 445 5.65E-05 2.02E-05 2.80 9.07E-05 2.41 5.04 82680 905 1.15E-04 4.41E-05 2.61
63 4 PtII-Acetato Complexes in Catalytic C-H Activation
The electrospray ionization mass spectrum of the reaction mixture directly introduced into the spectrometer, depicted in Figure 46, shows, aside from residual [25]+, principally two species in + + solution, which, by their m/z ratio, can be assigned to structures [28] and [28(CH3COOH)] . To support this assignment, [28]+[BArF]- was independently synthesized and characterized by 1H NMR, elemental analysis, and mass spectrometry. Collision-induced dissociation (CID) of one + + isotopomer of [28(CH3COOH)] yielded exclusively the corresponding isotopomer of [28] ; i.e., the only reaction is loss of acetic acid. It should be added though that the parent species, + [28(CH3COOH)] , appears in the ESI-MS spectra in appreciable amounts only when the desolvation conditions, e.g., tube lens potential, are set to be very mild.
+
Ar O m/z = 627 N O Pt N O Ar m/z = 687 OH
m/z = 687 CID
600 625 650 675 m/z + Ar + Ar N N O
Intensity (arb. units) Pt Pt N N O Ar Ar m/z = 627 m/z = 645
550 600 650 700 m/z Figure 46. Electrospray mass spectrum (ESI-MS) of the benzene-acetic acid solution containing 56 µM concentration of the PtII complex shortly after initiation of the reaction. The PtII phenyl complex disappears as the reaction reaches steady state. The inset shows the collision-induced dissociation (CID) spectrum of the mass-selected ion at m/z = 687. A single dominant loss of acetic acid is indicated, even under very gentle conditions of a (laboratory frame) offset of 0 V and 0.5 mtorr Ar in the octopole collision cell.
+ + The loss of acetic acid from [28(CH3COOH)] forming [28] is extremely facile. Studies using the Finnigan MAT LCQ ion trap mass spectrometer rather than the TSQ Quantum showed no amount of + [28(CH3COOH)] , which can be easily understood if one considers that the ion trap technology, based on scanning out stored ions, inherently stores ions under collisional conditions for a much longer time. This longer storage time makes the identification of labile species much more difficult. However, when [28]+ was isolated in the ion trap and collided with He it cleanly produced [24]+. The thermal decarboxylation of acetato complexes was described earlier.230 The observation of clean decarboxylation with no observable side-products instigates speculation about a possible reversal of this very reaction. Thus it is tempting to speculate about the utilization of the microscopic reverse of the decarboxylation that would make use of CO2 as a C1 building block. One possible application of this idea might be the conversion of the two most abundant greenhouse gases, CH4 and CO2, to acetic acid.
64 4.6 Theoretical Study of the Intermediates of the Catalytic Deuteration
4.6 Theoretical Study of the Intermediates of the Catalytic Deuteration
In order to gain more information on structural parameters of the species involved in the catalysis, DFT calculations were performed. Computations were performed with Gaussian98160 program package at the B3PW91 level of theory178,181 employing the Stuttgart-Dresden basis sets184 for Pt, N, and O, with the corresponding frozen-core potential for Pt, and 6-31G** basis sets185 for C and H. The transition state was located with the Gaussian98 implemented transition state optimization QST3, which requires the optimized structures of educt and product as well as a guess for the transition state structure.210,211 Again, as in the computations of deuterium isotopic scrambling (ch. 3.3, p. 52) the ligand 6 was replaced by the model H-N=CHCH=N-H. The computed structures displayed no imaginary frequencies, except for the transition state structure, which displayed exactly one imaginary frequency. All energies were ZPE corrected with a frequency-scaling factor of 0.9772.180
Such structures of model compounds for the adducts of [25]+ and [28]+ were computed. The structures together with the corresponding relative energies are depicted in Figure 47.
200
172 d 150 148
c
nergy [kJ/mol]nergy 100 E Relative Relative 50 b e 47 43
a 0 0
Reaction ξ Figure 47. Computed relative energies and structures of model compounds involved in the exchange of deuterium for hydrogen atoms in benzene catalyzed by [28]+. The energy of a and the energies of b – e comprise one molecule of benzene and acetic acid, respectively. All structures are local minima on the -1 energy hyper surface except for c that is a transition state (νc = 409i cm ).
The computations point, as well as the ESI-MS experiments described in the previous section, to + [28(CH3COOH)] (a) as the resting state of the reaction. All energies are given relative to a.
65 4 PtII-Acetato Complexes in Catalytic C-H Activation
Exchange of benzene for acetic acid yielded the benzene adduct b, which converted through IV -1 transition state (c) into the Pt hydride d. The imaginary frequency of c (νc = 409i cm ) was associated with the motion of a phenyl hydrogen to the PtII center. Finally, the acetic acid adduct + [25(CH3COOH)] e was computed. Attempts to locate structural analogs to a and e with the acetic acid coordinating by the oxygen of the hydroxy group were successful. However, the energies of the structures were both 57 kJ/mol higher. π-Coordination of benzene to [(diimine)Pt(Ph)]+ is less favorable by 44 kJ/mol as compared to the coordination of the carboxylic acid in e. Shifting the acetato ligand in d from η1 to η2 coordination raised the energy by 83 kJ/mol.xxxvii No attempts were made to locate the transition state for the intermolecular benzene for acetic acid exchange.
4.7 Discussion
There were reports both prior15,16 to, and contemporary68 with, those from Shilov17 in which heterogeneous, and later, homogeneous, Pt salts, chloroplatinates included, catalyzed the deuteration of benzene and other arenes with D2O as the deuterium source in the presence of acetic acid. The process, largely due to Garnett and co-workers,63 has been applied on the preparative scale for both deuteration73 and tritiation70-72 of arenes. The process is interesting from the mechanistic point of view because it is truly catalytic, with each turnover involving C-H activation of the arene. Moreover, the reaction occurs in the presence of coordinating solvents and reagents, which, in the work modeling Shilov chemistry, had been consistently considered to be deleterious. The present work is motivated by the desire to model the catalytic cycle of Garnett's reaction by way of well-defined PtII complexes for which the resting state and turnover-limiting step can then be determined. Renewed interest in the reaction, as seen in recent reports of H/D exchange catalyzed by rhodium224 and iridium81,229 complexes, makes the mechanistic study of well-defined model complexes timely.
One pass through the catalytic cycle is depicted in Scheme 20, starting from the resting state I. The overall catalytic cycle is split into two fused cycles in order to account for the two independent reactions, that is exchange of substrate and product (Cycle 1) and deuteration (Cycle 2). Each transformation in Scheme 20 may comprise several elementary steps. For example, the net ligand exchange reaction I → II may be either dissociative or associative, and the net oxidative addition reaction II → III may also include changes in π to σ coordination of benzene. Furthermore, no stereochemical claims are made; the structures are meant to denote stoichiometry and connectivity only. Nevertheless, Scheme 20 provides a general framework for a discussion of the mechanism and can accommodate the several possible mechanistic possibilities. It should be noted that, whether one considers the reaction V → VI to be an oxidative addition of the metal to an O-H bond or a protonation of the metal by a coordinated Brønsted acid, the result is the same. Furthermore, studies on related PtII complexes by Tilset show that the site of protonation is in fact the metal.192
xxxvii The large stabilization of the PtIV hydride by the η2 coordinated acetato ligand also argues in favor of the stabilization of the PtIV hydride by a chlorine from the ligand in the deuterium isotopic scrambling process (ch. 3.4).
66 4.7 Discussion
N OAc + Pt NO CD3 - C6H6-nDn DO I
+ AcOD Cycle 1 - AcOD + N C6H6-nDn Pt NOAc II
+ D H + N C6H5-nDn N C6H5-nDn Pt Pt VI III NO Cycle 2 NO O O
V IV + + N C6H5-nDn N C6H5-nDn Pt Pt NO NO CD3 CD3 DO HO
- AcOH + AcOD + Scheme 20. Proposed cycle for the deuteration of benzene by acetic acid-d4 catalyzed by [28(CH3COOH)] (I). The rate of II going through cycle 2 over the rate of II going through cycle 1 determines the Anderson Kemball M value. The index n denotes the number of deuterium atoms incorporated into the benzene molecule. It equals the number of passes of one benzene molecule through cycle 2.
The mass spectrometric results as well as the computations show clearly that the resting state of the catalytic cycle is a PtII-acetato complex with a molecule of acetic acid. The observed rate law indicates unambiguously that the turnover-limiting transition state contains one benzene unit more and one acetic acid unit fewer than does the resting state. It is in principle possible that there are one or more intermediates between the resting state and the turnover-limiting transition state; the transformations involving those intermediates would constitute preequilibria that do not affect the overall rate law.
After the resting state, the second crucial point in the catalytic cycle is the turnover-limiting transition state. The stoichiometry of the resting state and the observed rate law do not yet provide an unambiguous assignment of the turnover-limiting transition state. In addition, the M value, the mean number of deuteriums incorporated per pass through the catalytic cycle, with a single pass defined to be the reactions starting at the resting state and returning once to the same resting state,
67 4 PtII-Acetato Complexes in Catalytic C-H Activation
provides the remaining piece of information. From all runs in Table 10 a value of M = 2.6 ± 0.3 (2σ error bounds) was extracted. This means that between two and three deuteriums are incorporated in benzene per pass through the complete cycle shown in Scheme 20. The tangible consequence is visible in Figure 43 (p. 60), in which one sees qualitatively that, even at low conversion, multiply deuterated benzenes appear. Moreover, the peaks due to multiply deuterated benzene rise approximately parallel, rather than subsequent, to the rise of singly deuterated benzene. For M = 1, one would expect to see very little multiple deuteration until the amount of monodeuterated benzene had accumulated to a significant level in the bulk solvent. Hodges and Garnett68 report M ≈ 2-4 for 2- the deuteration of benzene by D2O catalyzed by PtCl4 in the presence of acetic acid and HCl, with the lower bound occurring for solutions with a high mole fraction of acetic acid, similar to the reaction conditions chosen for the experiments described in this chapter, and higher temperatures. It should be noted that these workers have excluded a simple electrophilic aromatic substitution mechanism for the deuteration, as well as reaction catalyzed by colloidal Pt(0). Considering the resting state and the larger-than-unity M value, one concludes that the turnover-limiting transition state necessarily lies between I and II, that is in Cycle 1 of Scheme 20, for the particular reaction conditions. The observed multiple deuteration per pass requires fast, reversible interconversion of all species in Cycle 2 of Scheme 20 through transition states that are not turnover-limiting. Changes in benzene and acetic acid concentration affect the various transition states unequally, so that it becomes possible that a different transition state can be rendered turnover-limiting for some other set of reaction conditions. However, no evidence for a change in the turnover-limiting step is evident in this study.231
There are several mechanistic possibilities for the ligand exchanges in Cycle 1 of Scheme 20. Normally, one would expect the exchange of acetic acid for benzene (I → II) and the exchange of benzene for acetic acid (II → I) in a square-planar d8 complex to proceed via an associative mechanism.51 This behavior has been reported for similar cationic PtII diimine complexes in other C-H activation studies.31,50 However, intermediates of a putative dissociative mechanism that are formally 14-electron PtII complexes have been prepared.44-49 Such complexes display a T-shaped structure and gain additional stabilization through intramolecular agostic interactions. The two possibilities, associative and dissociative ligand substitution,xxxviii are shown in Scheme 21. For each mechanism there are two possibilities for the turnover-limiting step, giving a total of four possible turnover-limiting transition states for each transformation. The stoichiometry of the resting state and the empirical rate law, first-order in benzene and inverse first-order in acetic acid, exclude three of the four in each case. The ligand substitutions I → II and II → I must occur via a dissociative mechanism with a turnover-limiting step located between the benzene adduct (II) and the η2 PtII- acetato complex. One presumes that the otherwise unusual dissociative mechanism for ligand substitution in a square-planar d8 complex becomes competitive because of the concomitant η2 coordination of the remaining acetate ligand. While no quantitative collision-induced dissociation (CID) threshold measurements were conducted, one can see from the CID experiment that the activation energy for loss of acetic acid from I is much lower than that for loss of coordinating solvent molecules, e.g., water, trifluoroethanol, and acetonitrile, from analogous complexes described in ch. 2.2.2.
xxxviii Dissociative in this context is meant only to imply that the coordinated acetic acid departs completely before coordination of incoming benzene. Whether the coordination of benzene occurs concurrent with, or after, the change from η2- to η1-acetato complex is neither determined nor specified in this work.
68 4.7 Discussion
+ N Ar OAc N Pt + benzene Ar - acetic acid O
+ OD + Ar Ar N OAc N OAc I Pt Pt II N O N Ar Ar DO - acetic acid + benzene
+ Ar N O Pt N O Ar
Scheme 21. Reaction scheme displaying the two possible mechanisms for the exchange of acetic acid for benzene. The upper mechanism depicts associative ligand exchange with a five coordinate intermediate. The lower pathway depicts the dissociative pathway.
The observation that ligand exchange accompanied with a haptotropic shift of the acetato ligand constitutes the turnover-limiting step is in itself not surprising. It is surprising, though, that this step is not so much slower than the other steps in the overall cycle. The M value of 2.6 indicates multiple deuteration per pass through the catalytic cycle, but it also indicates that the turnover-limiting ligand substitution is not drastically slower than the next slower step. If it were, then one would expect a value for M approaching 6. One would have ordinarily expected a Pt η2-carboxylato complex, the mutual intermediate of the two ligand exchanges in Cycle 1, to be a dead end in catalysis, or at least a structure whose stability would deleteriously affect turnover frequency. Such expectations had already been expressed in the literature.81 The present work indicates that ligand substitution via such an intermediate is not prohibitively slow. It is interesting to note that the earlier experiments by Garnett and co-workers were conducted with significant amounts of acetic acid present. Its role, however, was always assumed as co-solvent to keep the benzene/water mixture homogeneous. To the extent that the heterogeneous or homogeneous catalysts from Garnett are well modeled by the present system, one can surmise that acetato complexes and acetic acid regulate the turnover frequency in those systems as well.
+ The results suggest that complex [28(CH3COOH)] should be a competent precursor for catalytic + C-H activation of benzene. An attempted synthesis of [28(CH3COOH)] from the free ligand and [Pt(OAc)2]4 (29) failed. Nevertheless, ready access to organometallic complexes competent in C-H activation from coordination complexes lacking a metal-carbon bond should be possible, perhaps with sterically more bulky carboxylates. Such a procedure would be synthetically more convenient than the current preparative procedures starting from Pt alkyl complexes.
69 4 PtII-Acetato Complexes in Catalytic C-H Activation
4.8 Conclusion
It has been demonstrated that a well-defined cationic PtII-acetato complex is competent in at least one catalytic C-H activation reaction of benzene. The H/D exchange proceeds with TON in the thousands analogously to the H/D exchange reactions reported by Garnett and co-workers for in situ prepared heterogeneous Pt catalysts or chloroplatinate complexes. A combination of ESI-MS, solution-phase kinetics, and analyses of deuteration per turnover, supplemented by DFT calculations, unambiguously identifies the resting state in the cycle and finds the turnover-limiting dissociative ligand exchange step.
70 4.8 Conclusion
5 Pt/Cu-catalyzed Functionalization of Benzene
In the previous chapter a catalytic C-H activation of benzene with high TON was described. But the product of such an exchange of a hydrogen by a deuterium atom, partly deuterated benzene, is not so much more valuable than the reactant, regular benzene. The mechanistic investigations revealed that throughout the course of the reaction, PtII phenyl species were present in low concentrations (species IV and V in Scheme 20, p. 67, also compare [25]+ in Figure 46, p. 64). These species were reactive toward acetic acid. Whether or not their reactivity could be used in a reaction leading to functionalized benzene is a challenge that was addressed by the experiments described below.
The copper-mediated coupling of an aryl halide and a phenol, aniline or a arylthiol was first described by Fritz Ullmann. In 1903 he coupled o-chlorobenzoic acid and aniline in the presence of copper metal to give o-phenylaminobenzoic acid.88 Later on, it was found that an appropriate substrate aryl-X and a hydrogen-containing nucleophile could be coupled in the presence of a cooper salt to yield a wide variety of ethers, amines, and thioethers (Figure 1, p. 7).90 Evans and co- 102 workers reported the first catalytic version of the reaction in 1998. Using Cu(OAc)2 as a catalyst, they were able to couple phenylboronic acid and p-tBu-phenol with oxygen acting as the ultimate oxidant. Further improvements on the catalytic version of the Ullmann reaction were contributed by Collman and Zhong who used a dimeric, homogeneous CuII complex in the coupling of a number of arylboronic acids with imidazoles (Scheme 6).105 Lam and co-workers extended the substrate scope for these reactions significantly. They described catalytic reactions for vinylic-232 and N- heterocyclic95 boronic acids as well as for oxygen-containing nucleophiles.103 Related to that, phenyl acetate was recently reported to be a side-product in the Cu(OAc)2-catalyzed homocoupling of a number of arylboronic acids.233
Combining the concept of bimetallic catalysis84 with the precedent of acetoxylation of benzene76,82,83,87, heteroaromatics219c and saturated hydrocarbons77,78 offers a prospect for fusing the catalytic cycles depicted in Scheme 6 (p. 8) and Scheme 20 (p. 67). This concept is schematically outlined in Scheme 22 along with the corresponding net equation.
C-H Activation Coupling Cycle Cycle
Pt Ph Cu Ph Nu H2O + OH-
H
Nu H - + OH + 1/2 O 2 Pt Cu Ph
Transmetallation Nu
+ Nu - H + 1/2 O 2 + H 2O
Scheme 22. Fused catalytic cycles for the PtII-catalyzed C-H activation of benzene (left cycle) and of the CuII-catalyzed coupling of a phenyl residue with a nucleophile (right cycle). The lower line shows the resulting net reaction.
71 5 Pt/Cu-catalyzed Functionalization of Benzene
The two catalytic cycles have been studied in detail individually. However, the intersection of the two cycles, that is transmetallation of the phenyl group from platinum to copper, is less well understood while being decisive for the outcome of the reaction. In the present case at least the PtII complex [25]+ was positively charge, which rendered the reaction suitable for ESI-MS analyses. The influence of the ligands at platinum and copper on the transmetallation step was studied in the gas phase by the kinetic method introduced by Cooks in the early 1980s.234 The kinetic method is an alternative to equilibrium methods for determining thermochemical properties. It is based on the + rates of competitive dissociation of mass-selected ions (MA-X -MB) under collision-induced + + dissociation (CID) conditions to form the individual ionic monomers (MA-X and MB-X ). The ratio of the rates of the competitive dissociations of the dimer, expressed as the ratio of the abundance of the individual monomers, can be used to estimate the cation affinity difference between MA and MB. The kinetic method is sensitive to small thermochemical differences, and is applicable to polar and non-volatile samples even when they are not pure. It was originally introduced as a method to measure proton affinities of various compounds but its use was extended + 235 + 236 + 237 to the measurement of affinities to more complex ions such as NH4 , SiCl3 , OCNCO , to transition metals such as Cu+,238 Ni+, Co+,239 and to organometallic fragments such as CpFe+, CpCo+, and CpNi+ (Cp =cyclopentadienyl).240 Cooks241 and Erwin242 reviewed the kinetic method recently.
When the PtII/CuII heterobimetallic intermediate with different ligands on each side is fragmented, the ratio of the resulting PtII phenyl and CuII phenyl complexes will reflect their kinetic stability. On the assumption that the reverse barrier for both reaction is zero, entropic factors cancel out, and the energy distribution of the reactant ions can be approximated by a Boltzmann distribution; the observed kinetic stability will reflect the difference in thermodynamic stability as well. Thus, it is possible to establish an ordering of the relative stabilities of the different PtII phenyl and CuII phenyl complexes, depending on the electronic influence of the ligands.
5.1 Syntheses and X-Ray Crystallography
In order to investigate the feasibility of the envisioned coupling of the two catalytic cycles a number of Pt- and Cu-compounds was synthesized. With the intention to extend the range of PtII complexes, two ligands, 30 and 31, related to those presented in ch. 2.1.1, were synthesized analogously (Scheme 23).172
R
NH2 N O HCOOH + MeOH, r.t. N O R 30: R = OMe; 87 % R 31: R = NMe 2; 72 % Scheme 23. Synthesis of the ligands 30 and 31 by the route of tom Dieck.172
72 5.1 Syntheses and X-Ray Crystallography
In contrast to the ligand syntheses in ch. 2.1.1 where electron-poor anilines resulted in mediocre yields, p-methoxyaniline and p-dimethylaminoaniline led to good yields of the condensation products. Reaction of 30 and 31 with the PtII precursor 8 in toluene afforded the PtII complexes 32 29 II II and 33 (Scheme 24). Reactions of tmeda with Pt precursors 8 and 10 in CH2Cl2 afforded the Pt complexes 34 and 35.243
Ar N or + 8 or 10 N toluene or N CH Cl , r.t. N 2 2 Ar
R
N R' N Me Pt or Pt R' N Me N
34: R' = Me; 71 % 35: R' = Ph; 56 % R 32: R = OMe; 71 % 33: R = NMe 2; 56 % Scheme 24. General scheme for the synthesis of PtII complexes 32 – 35.
Protonolysis of 32 and 33 with HBArF in benzene at r.t. resulted in the formation of the cationic PtII phenyl complexes [36(solv)]+ and [37(solv)]+ within two hours (Scheme 25).xxxix The ligand solv denotes a solvent molecule that coordinates to the PtII center and stabilizes the cations. Treated in the same way, 34 and 35 yielded the PtII methyl cation [38(solv)]+ and the PtII phenyl cation [39(solv)]+.
xxxix The progress of the reaction was followed visually and by means of ESI-MS. It should be noted that the reaction with benzene of the Pt(II) methyl complexes resulting from protonolysis 32 and 33 proceeded much slower than the + reaction of [24(solv)] with benzene. The reaction times for the formation of the respective Pt(II) phenyl cations were larger by a factor of 5 – 10.
73 5 Pt/Cu-catalyzed Functionalization of Benzene
R +
+ N N R' Ph - - benzene [BArF] or [BArF] 32 - 35 + HBArF Pt Pt N solv N solv
38: R' = Me 39: R' = Ph solv = water, R TFE, 36: R = OMe acetonitrile, benzene, 37: R = NMe 2 acetic acid Scheme 25. General scheme for the protonation of neutral PtII alkyl and phenyl complexes. The resulting 8 three coordinate d -ML3 cations are ligated by a solvent molecule (solv).
The hydroxy-bridged CuII catalysts 40 and 41 were obtained from the reaction of CuCl and CuI triflate and water in methanol under an atmosphere of oxygen (Scheme 26) as described by de Jong.244 The reaction with CuOAc replacing CuCl or CuOTf yielded the CuII-acetato complex 42.
2+ H N O N - Cu Cu X 2 N O N H
40: X = Cl -; 75 % 41: X = OTf -; 12 % (22 %) N O2 N + CuX MeOH/H2O
N OAc or Cu * H2O N OAc 2
42: X = OAc -; 23 %
Scheme 26. Synthesis of the dimeric CuII complexes 40 –42. In the reaction CuI is oxidized to CuII by molecular oxygen.
It was not attempted to determine the structure of 42 because the compound was paramagnetic and could therefore not be determined by standard NMR methods. Closely related structures, however, II were described in literature. In the solid-state structure of Cu2(CH3COO)4 · 2H2O the two Cu centers are linked by µ-bridging acetato ligands.245 On the other hand, in the solid-state structure of II the nickel analogon to 42 [(tmeda)Ni(OAc)2]2 · H2O the Cu centers are µ-bridged with two acetato ligands and a molecule of water while the remaining acetato groups are ligated in η1-
74 5.2 Variations on the Ullmann Reaction
fashion.246Complexes 40 and 41 were isolated directly from the reaction mixture by precipitation while 42 was obtained by continuous extraction of the solids formed during the reaction with ether. The yield of 41 increased from 12 to 22 % when a route developed by Pfeiffer and Glaser was 247 employed. In this case Cu(OAc)2 was reacted with tmeda in water under air. Subsequently, the anion was exchanged in a one-pot reaction with an excess of NaOTf. When the reaction mixture was allowed to crystallize for seven days, dark blue cuboids were obtained. The crystals were suitable for x-ray analysis. The resulting structure is shown in Figure 48.
Figure 48. Ortep diagram of 41. Thermal ellipsoids are shown at 50 % probability. Hydrogen atoms are omitted for clarity. Selected bond lengths [Å]: Cu1-O11 1.9294(13), Cu1-O10 1.9566(14), Cu1-N2 2.0392(17), Cu1-N5 2.0488(17), Cu1-O14 2.4101(15), Cu1-O13 4.323, Cu1-Cu1' 2.9106(4), N2-C3 1.488(3), C4-N5 1.488(3), S12-O13 1.435(2), S12-O14 1.4378(14). Selected bond and torsion angles [°]: O11-Cu1-O10 80.94(7), N2-Cu1-N5 86.46(8), N2-Cu1-O10 173.13(7), O10-O11-N2-N5 1.7, N2-Cu1-O10- Cu1' –57.1. The structure contains a glide plane perpendicular to b. The structure contains one molecule of water, which is omitted for clarity.
II ++ The hydroxy-bridged Cu dimer [(tmeda)CuOH]2 had already been characterized by x-ray - 248 - 249 - 250 - 251 crystallography prior to this work, however the anions were BF4 , ClO4 , NO3 , , and Br .
5.2 Variations on the Ullmann Reaction
With the intention to elucidate whether the conditions found by Collman and Zhong for the catalytic coupling of phenylboronic acid and imidazole were also suitable for the coupling of a PtII phenyl intermediate and acetic acid, some test reactions were conducted.104 In these experiments the products were identified and quantified by means of GC-MS. First, under the original experimental conditions, that is 2 eq phenylboronic acid, 1 eq imidazole, 0.1 eq of 40 in CH2Cl2 under an atmosphere of oxygen, phenylboronic acid was coupled to imidazole to produce 1-phenylimidazole (Scheme 27).
75 5 Pt/Cu-catalyzed Functionalization of Benzene
B(OH)2 40, O2, r.t N + NNH N CH2Cl2
Scheme 27. Copper-catalyzed coupling of phenylboronic acid and imidazole under the condition described in ref. 104. The reaction was analyzed by GC-MS.
xl Next, the nucleophile, imidazole, was replaced by acetic acid-d4. The formation of phenyl acetate-d3 was observed after 12h. In a next step, the solvent, CH2Cl2, was replaced by acetic anhydride.xli This was done for two reasons. One was to exclude the chlorinated hydrocarbon from the reaction in order to prevent interference with the C-H activation in future reactions. A second reason was that previous reports stated that anhydrous conditions were favorable for the catalysis.95,102,104 Acetic anhydride was intended to act as a sacrificial drying reagent by removing water formed in the course of the reaction. In the two experiments, phenyl acetate and phenyl acetate-d3 were found in almost identical amounts in each experiment. However, fewer side- products were detected in the reaction mixture when CH2Cl2 was absent from the reaction. When acetic acid-d3 was excluded from the experiment, the yield of phenyl acetate determined by means of GC-MS increased slightly (Scheme 28). It was not investigated whether copper reacted with acetic anhydride itself or with acetic acid, the hydrolysis product of acetic anhydride with residual water. In one experiment, phenyl acetate was isolated by preparative TLC and identified by 1H NMR. From the isolated yield a TON of 4 based on CuII was calculated.
O B(OH)2 OO 40, O2, r.t O + O traces of H2O 21 %
Scheme 28. Catalytic coupling of phenylboronic acid with acetic anhydride and presumably water to produce phenyl acetate.
Because chloride was suspected to interfere with the future C-H activation catalysis, the CuII dimer 40 was replaced with 41, which contained triflate as a counterion. The results of these reactions were similar to the results from the previous experiments. When 42 was employed as a catalyst for the coupling reactions no product formation was observed. Finally, phenyl acetate was formed under similar conditions in the presence of benzene.xlii
5.3 Attempts to Fuse Catalytic Cycles
In a first experiment, 18, together with HBArF, were dissolved in benzene to produce [25(solv)]+. Subsequently, the reaction was purged with oxygen, and 41 and acetic anhydride were added to yield a deep blue solution (Scheme 29). xl Acetic acid-d4 in "100 %" quality from Glaser AG was chosen instead of acetic acid because it was known that the quality of the deuterated acid was sufficient to run catalytic cycles (compare to ch. 4.1). xli Although the acetic anhydride used was of puriss. p.a. quality, an impurity was found in the ESI-MS experiments (ch. 5.4). With a combination of GC-MS and HiRes ESI-FT-ICR-MS experiments, dehydroacetic acid (C8H8O4) was identified. For further ESI-MS experiments it was removed by purification according to literature procedures.203 xlii The GC-MS yield of phenyl acetate varied from one experiments to the next but under the reaction conditions described it was always detected to some amount.
76 5.3 Attempts to Fuse Catalytic Cycles
Cl Cl N OAc Ph 1. HBArF, benzene Pt 2. 41, Ac O N Ph 2 Cl Cl
18 Scheme 29. Attempted reaction of a PtII phenyl group with acetic anhydride through PtII/CuII bimetallic catalysis. After 17h at r.t., the reaction had turned turquoise but no phenyl acetate was found in the reaction mixture. Tiny amounts of biphenyl were detected by means of GC-MS. Comparable amounts of biphenyl were also found in experiments in the absence of copper. Biphenyl was found to be the product of PdII and PtII-catalyzed C-H activation reactions.219d In order to test the dependence of the reaction on a number of reaction parameters, a custom-made parallel reactor was used. The results from experiments varying the catalyst, the amount of acetic anhydride, of benzene, and of CH2Cl2 are presented in Table 11.
II II Table 11. Reaction conditions tested for the attempted Pt /Cu -catalyzed coupling of benzene and acetic acid.
Cu-Catalyst Ac2O [mmol] CH2Cl2 [ml] biphenyl phenyl acetate CuOAc 9.8 · 10-5 - yes - CuOAc 9.8 · 10-5 0.1 yes - CuOAc 3.2 · 10-4 - - - -5 Cu(OAc)2 9.8 · 10 - yes - -5 Cu(OAc)2 9.8 · 10 0.1 yes - -4 Cu(OAc)2 3.2 · 10 - - - 40 9.8 · 10-5 - yes - 40 9.8 · 10-5 0.1 yes - 40 3.2 · 10-4 - yes - 41 9.8 · 10-5 - yes - 41 9.8 · 10-5 0.1 yes - 41 3.2 · 10-4 - yes - 42 9.8 · 10-5 - yes - 42 9.8 · 10-5 0.1 yes - 42 3.2 · 10-4 - - - general reaction conditions: (4.0 ± 0.7) · 10-6 mol Cu-catalyst, 4.6 · 10-8 mol [25(solv)]+, 2.3 · 10-3 mol benzene, oxygen atmosphere, r.t., 20 h.
77 5 Pt/Cu-catalyzed Functionalization of Benzene
No phenyl acetate was produced under all the reaction conditions employed. Again, small amounts of biphenyl were detected in some instances.
5.4 Gas-Phase Experiments with PtII/CuII Heterobimetallic Clusters
The fact that no phenyl acetate was produced was attributed to a thermochemical feature. Normally, the bond strength of PtII phenyl bonds is superior to that of CuII phenyl bonds.51,252,253 Consequently, the transmetallation of the phenyl residue from platinum to copper might have been hampered (see Scheme 22, p. 71). Therefore, in a first step, a species in which transmetallation was supposed to happen was isolated and characterized in the gas phase. In a second step, it was attempted to loosen the PtII phenyl bond in order to facilitate the transmetallation.
All experiments were initiated by protonation of the PtII dimethyl complex with HBArF followed by reaction with benzene to yield the corresponding PtII phenyl complex.xliii The reaction depicted in Scheme 29 was repeated with the PtII dimethyl complex 16 instead of the PtII diphenyl complex 18. Subsequently, 41 and acetic anhydride were added. The reaction mixture was introduced directly from a Schlenk tube into a Finnigan TSQ Quantum triple quad ESI-MS instrument (compare ch. 4.5). The resulting mass spectrum is shown in Figure 49.
Cu
Cu2 Cu3
Pt Cu-Pt Intensity
400 500 600 700 800 900 1000 m/z Figure 49. ESI-MS spectrum of a reaction mixture containing [24(solv]+, [25(solv]+, 41, acetic anhydride, and benzene recorded on a TSQ Quantum instrument. A number of CuII, PtII, and PtII/CuII-mixed clusters can bee seen.
A number of singly and doubly charged compounds containing one, two or three copper atoms were detected.254 In addition, platinum-containing species were observed. The composition of all these compounds was deduced from their mass-to-charge ratio and their isotope pattern. However, no attempts were made to identify all of these homometallic species. Of more relevance in the present context, two prominent charged species could be identified containing platinum and copper. The xliii Residual [(diimine)Pt(Me)(solv)]+ does not interfere with the gas-phase experiments of other ions.
78 5.4 Gas-Phase Experiments with PtII/CuII Heterobimetallic Clusters
presence of both metals in one cluster is a prerequisite for transmetallation. From their mass-to- charge ratio and their isotope pattern it was deduced that both contained one each CuII, PtII, tmeda, 6, and two acetate moieties. In addition one of them (43, m/z = 878) contained a methyl, and the other ([44]+, m/z = 940) contained a phenyl group, respectively (Figure 50). At m/z =1052 another heterobimetallic species was observed which contained dehydroacetic acid. After purification of the acetic anhydride used in the reaction, this species was no longer observed.
calculated calculated + Cl Cl N OAc Ph N Cu Pt O O N N Cl Cl
[43]+ [44]+ m/z = 878 m/z = 940 Intensity
900 950 m/z Figure 50. Expanded spectrum of the high-mass region of the ESI-MS spectrum of a reaction mixture + described above. The inserts display the calculated isotope pattern for CuPtC27H37Cl4N4O4 (m/z =878, [43] ) + and CuPtC32H39Cl4N4O4 (m/z = 940, [44] ).
Collision-induced dissociation (CID) of mass-selected (dau-mode) [44]+ resulted in the formation of four compounds, which were identified by their mass-to-charge ratio and their isotope pattern (Figure 51). The major fragment signals, m/z = 238 and 824, were assigned to [(tmeda)Cu(OAc)]+ ([45]+) and [44 – tmeda]+, respectively. The signal at m/z = 625 was identified as the PtII acetato complex [28]+, which could stem from loss of benzene from [28(benzene)]+ (m/z = 703).xliv
xliv The formation of [28]+ by direct loss of (tmeda)Cu(Ph)(OAc) (46) from the heterobimetallic species [44]+ shown in Figure 50 is considered highly unlikely.
79 5 Pt/Cu-catalyzed Functionalization of Benzene
+
Cl + Cl N N O OAc Ph N Cu Cu Pt N O N O O N Cl Cl [45]+ [44]+ m/z = 238 + [44 - tmeda] m/z = 940 m/z = 824 Intensity
+ [28]+ [28(benzene)] m/z = 625 m/z = 703
200 400 600 800 m/z Figure 51. Daughter-ion spectrum of [44]+ (m/z = 940) recorded on a TSQ Quantum instrument with a resolution of 0.7 Da after selection in Q1 (10 Da) and CID with 1.5 mtorr of Ar at 14 eV collision energy (laboratory frame, Q2).
Having identified a species potentially capable of transmetallating the phenyl residue, a systematic study of the influence of the ligands on this process was started. The working hypothesis was to loosen the PtII phenyl bond by rendering the metal center more electron rich. In a first experiment, the PtII methyl complex [38]+ was reacted with benzene. Unfortunately, no PtII phenyl complex ([39]+) was detected after several hours.
MeO + Me 2N +
+ N N N N N OAc Me OAc Me OAc Me N Cu Pt Cu Pt Cu Pt O O O N N N O N N O O N
MeO Me 2N 47 49 51
MeO + Me 2N +
+ N N N N N OAc Ph OAc Ph OAc Ph N Cu Pt Cu Pt Cu Pt O O O N N N O N N O O N
MeO Me 2N 48 50 52
80 5.4 Gas-Phase Experiments with PtII/CuII Heterobimetallic Clusters
Therefore, [39]+ (see Scheme 25, p. 74) was tested in the reaction of benzene with 41 and acetic anhydride. By means of ESI-MS, similar heterobimetallic compounds containing a methyl ([47]+) or a phenyl group ([48]+) were detected (see chart). No indication for the formation of (tmeda)Cu(Ph)(OAc) (46) was found. Starting from the PtII dimethyl complex 32, the ESI-MS spectra of the reaction mixture displayed, among others, the heterobimetallic species [49]+ and [50]+ (see chart). Disappointingly, CID experiments with these ions did not indicate neutral loss of (tmeda)Cu(Ph)(OAc) (46).
Starting from the PtII dimethyl complex 33, the ESI-MS spectra of the reaction with 41, benzene and acetic anhydride displayed signals for heterobimetallic species: [51]+ (m/z = 828) and [52]+ (m/z = 890) (see chart and Figure 52a). The compound of interest ([52]+) while being present only in small amounts and accompanied by numerous homometallic CuII and PtII species, could be reproducibly detected (see isotope pattern in Figure 52a and expanded spectrum in Figure 52b)
Me2N + a) b) calculated [51]+ m/z = 828
Cu N OAc Ph N Cu Pt Cu2 O O N Cu3 N
Me2N [52]+ Pt m/z = 890 Intensity Cu-Pt Intensity calculated
300 400 500 600 700 800 900 m/z 800 850 900 m/z Figure 52. a) ESI-MS spectrum of a reaction mixture containing [37]+, 41, acetic anhydride, and benzene recorded on a TSQ Quantum instrument. A number of CuII, PtII, and PtII/CuII-mixed clusters can be seen. The heterobimetallic cluster [52]+ is only present in small amounts. b) Expanded spectrum of the high-mass region of the ESI-MS spectrum. The insets display the calculated isotope pattern for CuPtC31H51N6O4 + + (m/z =828, [51] ) and CuPtC36H53N6O4 (m/z = 890, [52] ).
Collision-induced dissociation of [52]+ resulted in formation of three ionic fragments (Figure 53). Loss of tmeda led to [52 – tmeda]+ (m/z = 774). In addition, two monometallic cations were observed: [(tmeda)Cu(OAc)]+ (m/z = 238, [45]+) and [(diimine)Pt(OAc)]+ (m/z = 575) (diimine = + ArN=CMeCMe=NAr, Ar = 4-NMe2-C6H4) ([53] ).
81 5 Pt/Cu-catalyzed Functionalization of Benzene
NMe2 + Me2N +
+ + [52 - tmeda] N O N O m/z = 774 N OAc Ph N Cu Pt Cu Pt N O N O N O O N
[45]+ + m/z = 238 [52] NMe 2 m/z = 890 Me2N [53]+ m/z = 575 Intensity
AcO N - Cu N
200 400 600 800 m/z Figure 53. Daughter-ion spectrum of [52]+ (m/z = 890) recorded on a TSQ Quantum instrument with a resolution of 0.9 Da after selection in Q1 (5 Da) and CID with 1.5 mtorr of Ar at 12 eV collision energy (laboratory frame, Q2). The insets display the expanded regions for the individual signals.
Having confirmed the influence of the ligand on PtII on the transmetallation of the phenyl residue from the platinum center to the CuII, attempts were made to apply similar ligand variations to the CuII center. Hence, to a mixture of the PtII methyl complex [37]+, benzene, and acetic anhydride was added a mixture containing [(bipy)Cu(OAc)]+ (bipy = 1,10-bipyridine). ESI-MS experiments of the resulting reaction mixture revealed the existence of a Pt/Cu-mixed species (m/z = 930, [54]+), which upon CID lost almost exclusively bipyridine.
5.5 Discussion
The acetoxylation of benzene, i.e. the coupling of benzene with acetic acid in the presence of a catalyst, was attempted in the present study. The reaction could provide a first step in a new route to obtain phenol from benzene.255 Phenol, the second largest volume chemical derived from benzene, currently consumes about 20 % of the total benzene production.256 However, the synthesis involves a three-step process starting from benzene (Hock process257) or a two-step process starting from toluene (Dow process258). Therefore, a two-step process for the production of phenol starting from benzene, acetoxylation of benzene and hydrolysis of the ester, is potentially attractive.xlv259
Platinum has proven to be a very efficient catalyst in the C-H activation of benzene and other hydrocarbons. However, no industrial process based on the conversion of the platinum-bound phenyl residue into a value-added product has so far been achieved. The coupling of the platinum- catalyzed C-H activation cycle to a copper-catalyzed coupling cycle is promising. In this way, a
xlvPhenyl acetate is also used as a starting material in the production of N-acetyl-p-aminophenol.259
82 5.6 Conclusion
direct oxidation of benzene could be avoided, which is problematic because the oxidation products of benzene are more reactive towards oxygen than benzene itself. In addition, the oxidation of benzene would occur at the copper center, which is conveniently reoxidized by molecular oxygen. However, the formation of phenyl acetate has not been achieved in this study. One problem lies in the transmetallation step of the phenyl residue from platinum to copper, which is depicted in Scheme 30. The reductive coupling to form the phenyl acetate would follow this step.
Pt Cu Pt Cu OO OO
Scheme 30. Schematic representation of the transmetallation in PtII/CuII heterobimetallic clusters, which was deduced from ESI-MS experiments in a Finnigan TSQ Quantum instrument.
The isolation of PtII/CuII heterobimetallic species in the gas phase proves the existence of compounds that are mediators for transmetallation. Isolation of these species in the gas phase made a systemic investigation of effects promoting or hampering this reaction step possible. Because the PtII phenyl bond is usually stronger than the CuII phenyl bond, transmetallation is thermochemically disfavored.xlvi CID experiments of [44]+, [48]+, and [50]+ gave no indication for the formation of a copper phenyl species. However, when the PtII center is more electron rich the situation changes. This can be seen from CID experiments of [52]+. While the formation of [(tmeda)Cu(OAc)]+ ([45]+) + corresponds to neutral loss of (diimine)Pt(Ph)(OAc) (53(C6H5)) from [52] , the formation of [(diimine)Pt(OAc)]+ ([53]+) compels loss of (tmeda)Cu(Ph)(OAc) (46) from [52]+.xlvii
A more detailed study of the electronic effects of the ligand is needed. First, quantitative data from CID threshold experiments have to be obtained. This should be followed by further modifications on the ligands on CuII and PtII, especially on the backbone of the diimine ligands instead of on the arenes.
5.6 Conclusion
A fusion of the PtII-catalyzed C-H activation with a CuII-catalyzed coupling with acetic acid was attempted. The formation of phenyl acetate was not observed. This was attributed to a endothermic transmetallation step of the phenyl group from platinum to copper. By means of ESI-MS, PtII/CuII heterobimetallic species, prerequisite for transmetallation, were characterized in the gas phase. The behavior of these species upon CID was studied systematically by variation of the ligand spheres of PtII and CuII. With the appropriate choice of ligands, transmetallation and formation of phenyl acetate might be feasible.
xlvi There is precedent for transmetallation of a phenyl residue from B, Si, Sn, Pb, and Bi to copper. (see ch. 1.4) xlvii Throughout the investigations on the Pt(II)/Cu(II) systems no indication for exchange of the ligands from Pt(II) to Cu(II) or vice versa was obtained in ESI-MS experiments. In addition no free ligand was detected in GC-MS analyses of the reaction mixtures.
83 6 Experimental Section
6 Experimental Section
6.1 Instruments
NMR spectroscopy. 1D-NMR experiments were recorded on Varian Mercury 300 and Varian Gemini 300 spectrometers. 2D-spectra were recorded on a Bruker DRX-400 spectrometer. 1H and 13C chemical shifts are reported in ppm relative to tetramethylsilane, with residual solvent proton 1 13 1 13 resonances as internal standard (CDCl3: H = 7.26, C = 77.0; CD2Cl2: H = 5.32, C = 53.8; 2,2,2- 1 1 13 1 13 trifluoroethanol-d3: H = 3.88; acetone-d6: H = 2.04, C = 29.8; benzene-d6: H = 7.15, C = 1 19 128.0; CD3OD: H = 3.30. F chemical shifts are reported in ppm relative to CCl3F as external standard. 13C NMR and 19F NMR spectra were proton broad-band-decoupled. The multiplicity of the signals are denoted by the following abbreviations: s: singlet, d: doublet, t: triplet, m: multiplet, dd: doublet of doublet, dt: doublet of triplet, br: broad.
UV/Vis spectroscopy. Spectra were recorded on Perkin Elmer Lambda 6 and Hitachi U-2010 UV/Vis spectrophotometers: The cuvette holders were temperature stabilized by a Haake F3 heat regulator.
IR spectroscopy. Spectra were taken in KBr pellets using a Perkin Elmer Paragon 1000 FT-IR spectrometer. Peak positions are given in cm-1 and intensities are reported: vs = very strong, s = strong, m = medium, w = weak, b = broad.
GC-MS measurements for the Pt-project were conducted with a Fisons MD 800 mass spectrometer coupled to a Fisons GC 8000 gas chromatograph with a 12 m DB-5MS column (0.25 mm inner diameter, 0.25 µm film). Helium was employed as carrier gas. The temperature program was 50 °C, 1 min, 50 °C min-1, 200 °C, 1 min.
For the Cu-project GC-MS measurements were conducted with a ThermoFinnigan TraceGC/TraceMS with a 60 m Zebron ZB-1 column (0.25 mm inner diameter, 0.25 µm film). Helium with a flow of 0.8 ml/min was used as a carrier gas. The temperature program was 60 °C, 2 min, 30 °C min-1, 200 °C, 6 min.
Melting Points were measured in open glass capillaries in a melting point apparatus from Büchi (Dr Tottoli), and are uncorrected.
Elemental analyses were carried out by the Mikrolabor of the Laboratorium für Organische Chemie der ETH Zürich.
X-Ray Crystallography. Crystals of Ligand and 2252 were measured on a Nonius CAD4 diffractometer with CuKα radiation (λ = 1.5418 Å). Crystals of 2077 were measured on a Bruker- Nonius Kappa-CCD with MoKα radiation (λ= 0.7107Å). The structures were solved by direct methods260 and refined by full-matrix least-squares analysis including an isotropic extinction correction. 261 All heavy atoms were refined anisotropically (H-atoms isotropic, whereby H- positions are based on stereochemical considerations).
Crystals of 213, 214, and 235 were measured on a Nonius CAD-4 Diffractometer with CuKα radiation (graphite monochromator, λ = 1.5418 Å). Part of the structure was solved by direct method with SIR97,262 the remaining non-H-atoms were found from a difference Fourier map. The
84 6.2 Chemicals and Experimental Procedures
non-H atoms were refined anisotropically with SHELXL-97.261 H-atoms were calculated at idealized positions and included in the structure factor calculation with fixed isotropic displacement parameters.
Thin Layer Chromatography was done on Merck TLC Plates, Silica Gel 60 F254.
Preparative TLC was carried out on Merck Pre-Coated PLC Plates, Silica Gel 60 F-254.
Computational Details. All calculations involving Pt were performed with the Gaussian98 package. All geometries were fully optimized using density functional theory (DFT) with Becke's three parameter hybrid exchange (B3) and the Perdue-Wang (PW91) correlation functional. The level of theory was checked in comparison with the x-ray structures of 2077 and 2252. Local minima were identified by the absence of any negative eigenvalues of the Hessian matrix in a vibrational analysis. Transition state structures were characterized by one negative eigenvalue. For Pt, N, and O the Stuttgart-Dresden basis sets were used. For Pt the corresponding effective core- potential was used replacing 60 core-electrons. The basis sets for C, H, F and Cl were 6-31G**. Heat capacities of TFE and 1,2,2,2-tetrafluoroethane were calculated similarly with the Gaussian94 package and the Lee-Yang-Parr (LYP) correlation functional. The basis set for all atoms were 6- 31G*. The scaling factor was 0.84.
Isotope Pattern were simulated with the Sheffield Chemputer (http://www.shef.ac.uk/chemistry/ chemputer/isotopes.html).xlviii
Exact masses were calculated with IsoPro 3.0.
6.2 Chemicals and Experimental Procedures
Reactions involving oxygen- and moisture-sensitive compounds were performed under the strict exclusion of air in nitrogen atmospheres, using Schlenk techniques or an mbraun Labmaster 130 glove box.
Syntheses developed together with André Müller or accomplished by Mr Müller are marked with André n.
Solvents for organic syntheses were purchased from Fluka or Aldrich in p.a. quality and used as received. All solvents used in syntheses involving platinum were purified according to literature procedures203and degassed prior to use.263 In the kinetics of PtII acetato complexes, benzene was purchased from Fluka in OEKANAL quality and used as received.
Deuterated solvents and chemicals were purchased from Dr. Glaser (NMR solvents, acetic acid-d4 in "100 %"-quality), Cambridge Isotope Laboratories (CD3I, TFE-d3, benzene-d6, D2O) or Aldrich (1,3,5-trideuterobenzene) and purified if necessary.203
Chemicals. Pentafluoroaniline, 4-bromo-2,5-difluoroaniline, 2,4,6-trifluoroaniline, 2,6- dichloroaniline, 1-chloropentane, copper(I) acetate, copperII acetate, copper(I) triflate toluene xlviii The Isotope Pattern Calculator v4.0 (http://www.geocities.com/junhuayan/pattern.htm) produced incorrect results when chlorine atoms were present. The Isotope Distribution Calculator and Mass Spec Plotter (http://www2.sisweb.com/mstools/isotope.htm) seemed to have a high-mass limit at roughly 1000 Da.
85 6 Experimental Section
complex were obtained from Aldrich. 2,2,2-trifluoroethanol, silver acetate, platinum(IV) chloride were obtained from ABCR. CH3Li in ether, HBF4 54 % in ether, (CH3)2S, N,N,N',N'- tetramethylethylenediamine, phenylboronic acid, cuprous chloride, sodium trifluoromethanesulfonate were obtained from Fluka. HBF4 50 % in water, 3,5-bis(trifluoromethyl)- aniline, (cis,cis-1,5-cyclooctadiene)-palladium(II)-chloride, CH3MgCl 22 % in THF were obtained from ACROS. K2PtCl4 was purchased from Merck. All chemicals were used as received.
6.3 Ligand Syntheses.
ArN=CMeCMe=NAr (Ar = 3,5-(CF3)2-C6H3) (1) (GG 171). 14.57 g (63.6 mmol) 3,5- bis(trifluoromethyl)-aniline and 2.76 g (31.8 mmol) 2,3-butanedione were dissolved in 6 ml of methanol and 0.6 ml of formic acid were added. After stirring for 96 h at r.t. the precipitate was isolated by filtration, washed 4 times with cold methanol and dried in vacuo. 3.24 g (6.38 mmol, 20 %) of a white powder (m.p. 198-199 °C) were obtained. 1H NMR (300 MHz, chloroform-d) δ 13 2.18 (s, 6H, CH3), 7.22 (s, 4H, ArHp). C (75 MHz, chloroform-d) δ 15.74 (CH3), 117.83 (ArCm), 1 2 118.94 (ArCo), 123.15 (q, J(F-C) = 272.3 Hz, CF3), 132.66 (q, J(F-C) = 33.4 Hz, ArCm), 151.62 19 (ArCipso), 169.75 (C=N). F NMR (282 MHz, chloroform-d) δ –63.00 (CF3). Anal. calcd. for C20H12F12N2: C, 47.26; H, 2.38; N, 5.51. Found: C, 47.41; H, 2.45; N, 5.48.
ArN=CMeCMe=NAr (Ar = C6F5) (2) (GG 213). 5.00 g (27.3 mmol) pentafluoroaniline and 1.18 g (16.7 mmol) 2,3-butanedione were dissolved in 2 ml of methanol, 0.1 ml of formic acid was added and the solution was stirred at r.t for 60 h. The precipitate was isolated by filtration, washed 3 times with methanol and dried in vacuo. 413 mg (0.99 mmol, 6 %) of a light yellow powder (m.p. 166- 1 19 167 °C) were obtained. H NMR (300 MHz, chloroform-d) δ 2.24 (s, 6H, CH3). F NMR (282 3 3 MHz, chloroform-d) δ -162.27 (t, J(F-F) = 22.4 Hz, Fm), -160.69 (t, J(F-F) = 22.5 Hz, Fp), - 4 3 150.67 (dd, J(F-F) = 5.7 Hz, J(F-F) = 22.4 Hz, Fo). Anal. calcd. for C16H6F10N2: C, 46.17; H, 1.45; N, 6.73. Found: C, 46.16; H, 1.53; N, 6.83.
ArN=CMeCMe=NAr (Ar = 4-Br-2,6-F2-C6H2) (3) (GG 223). 960 mg (4.57 mmol) 4-bromo-2,6- difluoroaniline and 207 mg (2.28 mmol) 2,3-butanedione were dissolved in 0.5 ml of methanol and 4 drops of formic acid were added. The reaction mixture started to solidify after one hour. After 26 h 0.7 ml of methanol were added and the reaction was filtered to yield thin, yellow needles which were washed 3 times with methanol and dried in vacuo. 479 mg (1.03 mmol, 45 %) of yellow 1 needles (m.p. 166-167 °C) were obtained. H NMR (300 MHz, chloroform-d) δ 2.20 (s, 6H, CH3), 3 13 4 7.16 (d, J(F-H) = 7.1 Hz, 4H, ArHm). C (75 MHz, chloroform-d) δ 16.93 (CH3), 115.66 (dd, J(F- 2 3 Cm) = 8.5 Hz, J(F-Cm) = 17.6 Hz ArCp), 116.13 (t, J(F-Cp) = 12.1 Hz, ArCp), 126.45 (ArCipso), 1 3 19 152.19 (q, J(F-C) = 249.8 Hz, J(F-C) = 6.8 Hz, Co), 172.83 (C=N). F NMR (282 MHz, 3 chloroform-d) δ -120.35 (d, J(H-F) = 6.9 Hz, Fo). Anal. calcd. for C16H10Br2F4N2: C, 41.23; H, 2.16; N, 6.01. Found: C, 41.22; H, 2.22; N, 6.01.
ArN=CMeCMe=NAr (Ar = 2,4,6-F3-C6H2) (4) (GG 226). 1.27 g (8.62 mmol) 2,4,6-trifluoroaniline and 371 mg (4.31 mmol) 2,3-butanedione were dissolved in 0.5 ml of methanol, and 3 drops of formic acid were added. Precipitation set in one hour. After 72 h the precipitate was isolated by filtration, washed two times with methanol and dried in vacuo. 223 mg (0.65 mmol, 15 %) of a light yellow powder (m.p. 139-141 °C) were obtained. 1H NMR (300 MHz, chloroform-d) δ 2.20 (t, 13 J(F-H) = 1.4 Hz, 6H, CH3), 6.77 (t, J(F-H) = 8.2 Hz, 6H, ArHm). C (75 MHz, chloroform-d) δ 3 4 4 16.81 (CH3), 100.58 (t, J(F-Cm) = 27.1 Hz, ArCm), 152.26 (ddd, J(F-Co) = 9.9 Hz, J(F-Co) =
86 6.4 Complex Syntheses.
1 4 1 14.6 Hz, J(F-Co) = 247.5 Hz, ArCo), 158.98 (td, J(F-Cp) = 14.3 Hz, J(F-Co) = 245.6 Hz, ArCp), 19 4 173.09 (C=N). F NMR (282 MHz, chloroform-d) δ -119.20 (d, J(F-F) = 7.0 Hz, 4F, Fo), -119.20 (m, 2F, Fp). Anal. calcd. for C16H10F6N2: C, 55.82; H, 2.93; N, 8.14. Found: C, 55.96; H, 3.11; N, 8.05.
ArN=CMeCMe=NAr (Ar = 2,6-Cl2-C6H3) (6) (André XV/1). In a two-necked reaction flask, equipped with a water separator, 2,6-dichloroaniline (20.1 g, 123 mmol) and 2,3-butanedione (5.95 ml, 61.9 mmol) were dissolved in 60 ml of toluene. 1.5 ml of trifluoro acetic acid were added and the solution was held at reflux for 3 h. Upon cooling to r.t. a brownish precipitate formed. It was purified by flash chromatography with petroleum ether containing 0.5 % acetonitrile over 80 g of silica. 7.20 g (19.2 mmol, 31 %) of the slightly yellow product were obtained (m.p. 192 °C). 1H 3 NMR (400 MHz, chloroform-d) δ 2.16 (s, 6 H, NCMeCMeN), 7.01 (t, J(Hm-Hp) = 8.4 Hz, 2 H, 3 13 ArHp), 7.36 (d, J(Hp-Hm) = 8.4 Hz, 4 H, ArHm). C NMR (90+135 DEPT, HSQC) (100 MHz. chloroform-d) δ 16.90 (NCMeMeCN), 123.84 (ArCo), 124.74 (ArCp), 128.21 (ArCm), 145.06 (ArCipso), 171.29 (NCMeMeCN). Anal. calcd. for C16H12Cl4N2: C, 51.37; H, 3.23; N, 7.49. Found: C, 51.44; H, 3.36; N, 7.35.
ArN=CMeCMe=NAr (Ar = 4-CH3O-C6H4) (30) (GG 4117). 2.45 g (19.9 mmol) 4-methoxyaniline were dissolved in 6 ml of MeOH and were acidified with 5 drops of formic acid. Drop wise 856 mg (9.9 mmol) of 2,3-butanedione were added. After 3 h the solids were isolated by filtration and were washed with 10 ml of MeOH, and 2 ml of ether. After drying in vacuo, 2.56 g (8.6 mmol, 87 %) 1 yellow powder (m.p.: 186 °C) were obtained. H NMR (300 MHz, dichloromethane-d2) δ 2.14 (s, 5 3 6H, CH3), 3.81 (s, 6H, OCH3), 6.75 (dd, J(H-H) = 2.5 Hz, J(H-H) = 6.8 Hz, 4 H, ArH), 6.91 (dd, 5 3 J(H-H) = 2.1 Hz, J(H-H) = 6.6 Hz, 4 H, ArH). Anal. calcd. for C18H20N2O2: C, 72.95; H, 6.80; N, 9.45. Found: C, 72.97; H, 6.87; N, 9.62.
ArN=CMeCMe=NAr (Ar = 4-(CH3)2N-C6H4) (31) (GG 4128). 1.73 g (12.7 mmol) 4- (dimethylamino)-aniline were dissolved in 6 ml of MeOH and were acidified with 8 drops of formic acid. Drop wise 546 mg (6.35 mmol) of 2,3-butanedione were added. After 20 h the solids were isolated by filtration and were washed with 20 ml of MeOH, and 3 ml ofether. After drying in vacuo, 1.47 g (4.55 mmol, 72 %) dark yellow powder (m.p.: 172-173 °C) were obtained. 1H NMR (300 MHz, dichloromethane-d2) δ 2.18 (s, 6H, CH3), 2.94 (s, 12H, NCH3), 6.76 (s, 8H, ArH). Anal. calcd. for C20H26N4: C, 74.50; H, 8.13; N, 17.38. Found: C, 74.52; H, 8.15; N, 17.61.
6.4 Complex Syntheses.
PtCl2((CH3)2S)2 (7) (André XII/16). A solution of 5.58 g (13.4 mmol) K2PtCl4 in 112 ml of water was cooled to 0 °C. During 30 minutes 2.3 ml (31.3 mmol) of (CH3)2S were added, the reaction stirred for 4 h at 0 °C and stored for 12 h at 4°C. The yellow solution was filtered off giving a pink- brownish precipitate (3.6 g). The powder was stirred with 150 ml of CH2Cl2 for 16 h, filtered and then the solvent was evaporated under reduced pressure to yield 2.48 g bright yellow powder. The water phase was evaporated to dryness and the residue was stirred with 150 ml of CH2Cl2. After filtration and evaporation of the solvent 1.48 g yellow powder were obtained. 3.96 g (10.1 mmol, 1 75 %) yellow powder (m.p. 156-159 °C). H NMR (300 MHz, dichloromethane-d2) δ 2.44 (s, 3 195 3 195 13 J( Pt-H) = 41.8 Hz, 2.5H, trans-CH3), 2.53 (s, J( Pt-H) = 49.3 Hz, 1H cis-CH3). C NMR 2 195 (75 MHz, dichloromethane-d2) δ 23.04 (s, J( Pt-C) = 17.6 Hz, cis-PtC), the trans isomer was not measured. Anal. calcd. for C4H12Cl2PtS2: C, 12.31; H, 3.10; Found: C, 12.42; H, 3.06.
87 6 Experimental Section
[Pt2(CH3)4(µ-S(CH3)2] (8) (André XIII/9). A suspension of 2.94 g (7.53 mmol) PtCl2((CH3)2S)2 in 120 ml of diethyl ether was cooled to 0 °C. Over 10 minutes 12 ml of a 1.6 M CH3Li solution in diethyl ether were added. After 3.5 h the reaction was quenched with saturated NH4Cl-solution. The phases were separated, the water phase extracted twice with diethyl ether and the combined organic phases were dried over MgSO4. After evaporation of the solvent under reduced pressure the residue was washed with pentane and dried in vacuo. 1.50 g (2.61 mmol, (69 %) white powder (decomp. 1 2 195 >102 °C). H NMR (300 MHz, dichloromethane-d2) δ 0.59 (s, J( Pt-H) = 85.8 Hz, 12H,PtCH3), 3 195 13 2.76 (s, J( Pt-H) = 20.4 Hz, 12H SCH3). C NMR (75 MHz, dichloromethane-d2) δ -7.90 (s, 1 195 J( Pt-C) = 388.2 Hz, PtCH3), 21.07 (s, SCH3). Anal. calcd. for C8H24Pt2S2: C, 16.72; H, 4.21. Found: C, 17.01; H, 4.11.
[Pt2(C6H5)4(µ-S(CH3)2] (10) (André XI/40). A suspension of 2.60 g (6.66 mmol) PtCl2((CH3)2S)2 in 110 ml of ether was cooled to 0 °C. Over 5 minutes, 9.2 ml of a 1.8 M C6H5Li solution in THF (16.6 mmol) were added. After 3 h the reaction was quenched with saturated NH4Cl-solution. After filtration, the phases were separated and the water phase extracted with diethyl ether. The combined organic phases were concentrated to 10 ml and stored at 4 °C for 16 h. The solution was filtered off the brown oily precipitate, which was washed twice with pentane and dried in vacuo. 1.26 (2.01 mmol, 60 %) brown powder (m.p. 146 °C). 1H NMR (300 MHz, chloroform-d) δ 2.47 (s, 3 4 3 J(Pt-Ho) = 41.9 Hz, 12H, CH3), 6.76 (tt, J(Ho-Hp) = 1.4 Hz, J(Hm-Hp) = 7.3 Hz, 2H, PtPhp), 6.92 3 4 4 3 ("t", J(H-H) = 7.4 Hz, J(Pt-Hm) = 43.9 Hz, 4H, PtPhm), 7.36 (dd, J(Hp-Ho) = 1.4 Hz, J(Hm-Ho) = 3 7.9 Hz, J(Pt-Ho) = 72.2 Hz, 4H, PtPho). Anal. calcd. for C28H32Pt2S2: C, 40.87; H, 3.92. Found: C, 40.96; H, 3.97.
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 3,5-(CF3)2-C6H3) (11) (GG 175). To a solution of 100 mg (0.17 mmol) [Pt2(CH3)4(µ-S(CH3)2] in 18 ml of toluene, 180 mg (0.35 mmol) ArN=CMeCMe=NAr (Ar = 3,5-(CF3)2-C6H3) were added. The reaction was stirred for 12 h at r.t. under the exclusion of light. The solvent was evaporated under reduced pressure to yield 266mg of dark purple crystals. The product was recrystallized at 4 °C from 9 ml of CH2Cl2/pentane 5:4, washed with pentane and dried in vacuo. 237 mg (0.32 mol, 95 %) purple crystals (decomp. >235 °C) were obtained. 1H 2 195 NMR (300 MHz, dichloromethane-d2) δ 1.14 (s, J( Pt-H) = 87.0 Hz, 6H, Pt-CH3), 1.36 (s, 6H, 19 NCMeCMeN), 7.57 (s, 4H, Ar-Ho), 7.89 (s, 2H, Ar-Hp). F NMR (282 MHz, dichloromethane-d2) δ -62.66 (s, 12F, CF3).Anal. calcd. for C22H18F12N2Pt: C, 36.02; H, 2.47; N, 3.82. Found: C, 36.12; H, 1.97; N, 3.79.
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 2,6-(CH3)2-C6H3) (12) (GG 2081). 199 mg (681 µmol) ArN=CMeCMe=NAr (Ar = 2,6-(CH3)2-C6H3) and 192 mg (340 µmol) [Pt2(CH3)4(µ-S(CH3)2] were placed in a Schlenk-tube and cooled to –20 °C. 10 ml of toluene were added and the yellow suspension was allowed to warm to r.t. After 48 h the solvent was filtered from the ruby colored precipitate, which was washed with pentane and dried in vacuo. Giving 273 mg (527 µmol, 77 %) 1 of ruby colored powder (m.p. >250 °C). H NMR (300 MHz, dichloromethane-d2) δ 0.80 (s, 2 195 J( Pt-H) = 86.4 Hz, 6H, Pt-CH3), 1.24 (s, 6H, NCMeCMeN), 2.17 (s, 12H, Ar-CH3), 7.11 (m, 2H, 3 Ar-Hp), 7.20 (d, J(Hp-Hm) = 7.5 Hz, 4H, Ar-Hm).Anal. calcd. for C22H30N2Pt: C, 51.05; H, 5.84; N, 5.41. Found: C, 50.30; H, 5.89; N, 5.26.
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = C6F5) (13) (GG 2080). 155 mg (372 µmol) ArN=CMeCMe=NAr (Ar = C6F5) and 128 mg (223 µmol) [Pt2(CH3)4(µ-S(CH3)2] were dissolved in 6.5 ml of 1,2-dichlorethane. The solution was held at 300 mbar and 39 °C for 6 h, and at r.t. for 12 h. The solvent was evaporated under reduced pressure and the green, vitreous residue was
88 6.4 Complex Syntheses.
recrystallized from 1.3 ml of CH2Cl2/pentane 8:5. Giving 25 mg (39 µmol, 8 %) of black, rhombic 1 2 195 crystals. H NMR (300 MHz, acetone-d6) δ 1.40 (s, J( Pt-H) = 55.3 Hz, 6H, Pt-CH3), 1.58 (s, 6H, 19 4 3 NCMeCMeN). F NMR (282 MHz, acetone-d6) δ -164.17 (dt, J(F-F) = 5.8 Hz, J(F-F) = 21.2 Hz, 4 3 4F, Ar-Fm), -163.19 (t, 21.4 Hz, 2F, Ar-Fp), -152.19 (dd, J(F-F) = 5.2 Hz, J(F-F) = 22.2 Hz, 4F, Ar-Fo).
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 4-Br-2,6-F2-C6H2) (14) (GG 235). To a solution of 20.0 mg (34.8 µmol) [Pt2(CH3)4(µ-S(CH3)2] in 3 ml of toluene, 32.4 mg (69.6 µmol) ArN=CMeCMe=NAr (Ar = 4-Br-2,6-F2-C6H2) were added. The reaction was stirred for 12 h at r.t. under the exclusion of light. The solvent was evaporated under reduced pressure to yield 37 mg of turquoise crystals. The product was recrystallized at –20 °C from 3.4 ml of CH2Cl2/pentane 3:0.4, washed with pentane and dried in vacuo. 26 mg (37.6 µmol, 54 %) of turquoise, rhombic crystals (m.p. >250 °C) were 1 2 195 obtained. H NMR (300 MHz, dichloromethane-d2) δ 1.23 (s, 6H, NCMeCMeN), 1.33 (s, J( Pt- 3 19 H) = 88.2 Hz, 6H, Pt-CH3), 7.36 (d, J(F-H) = 7.0 Hz, 4H, Ar-Hm). F NMR (282 MHz, 3 dichloromethane-d2) δ -118.03 (d, J(H-F) = 7.5 Hz, 4F, Ar-Fo).
(phenanthrolin)-dimethylplatinum (15) (GG 214). To the solution of 20.0 mg (34.8 µmol) [Pt2(CH3)4(µ-S(CH3)2] in 4.5 ml of toluene, 12.5 mg (69.6 µmol) 1,10-phenathrolin were added. After stirring the red solution for 14 h at r.t., the solvent was evaporated at reduced pressure. The red residue was washed with pentane, ether and again with pentane and dried in vacuo. 28 mg (69.1 µmol, 99 %) of a red powder (decomp. >225 °C) were obtained. Crystals suitable for x-ray 1 analysis were obtained by recrystallization from toluene. H NMR (300 MHz, dichloromethane-d2) 2 195 3 δ 1.14 (s, J( Pt-H) = 86.4 Hz, 6H, Pt-CH3), 7.85 (dd, J = 5.1 Hz, J = 8.4 Hz, 2H, Ar-H3), 7.95 (s, 2H, Ar-H5), 8.64 (dd, 4J = 1.4 Hz, 3J = 8.0 Hz, 2H, Ar-H4), 8.64 (dd, 4J = 1.4 Hz, 3J = 5.0 Hz, 3J(Pt- H) = 22.1 Hz 2H, Ar-H2). Anal. calcd. for C14H14N2Pt: C, 41.48; H, 3.48; N, 6.91. Found: C, 41.48; H, 3.39; N, 6.62.
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 2,6-Cl2-C6H3) (16) (GG 2252). 326 mg (871 µmol) ArN=CMeCMe=NAr (Ar = 2,6-Cl2-C6H3) and 300 mg (522 µmol) [Pt2(CH3)4(µ-S(CH3)2] were placed in a Schlenk-tube and were cooled to –20 °C. 20 ml of toluene were added, the pressure was reduced to 100 mbar, and the reaction was allowed to warm to r.t. and stirred for 36 h. After removal of the stir bar the reaction was cooled to 0 °C for 2 h and the dark green solvent was filtered off. The residue was washed twice with pentane and with diethyl ether and dried in vacuo. 184 mg (307 µmol, 35 %) black powder (m.p. >250 °C) were isolated. Crystals suitable for x-ray 1 analysis were obtained by recrystallization from CH2Cl2/Ether 1:5. H NMR (300 MHz, 2 195 dichloromethane-d2) δ 1.14 (s, J( Pt-H) = 87.0 Hz, 6H, PtMe), 1.16 (s, 6H, NCMeCMeN), 7.23 (t, 3 3 13 J(Hm-Hp) = 8.4 Hz, 2H, ArHp), 7.55 (d, J(Hp-Hm) = 8.4 Hz, 4 H, ArHm). C NMR (90, 135 13 C-DEPT) (75 MHz, dichloromethane-d2) δ -15.1 (Pt-CH3, J(Pt-C) not found), 21.3 (NCMeCMeN), 127.6 (Cortho), 127.7 (Cpara), 128.4 (Cmeta), 142.9 (Cipso), 172.6 (C=N). Anal. calcd. for C18H18Cl4N2Pt: C, 36.08, H, 3.03; N, 4.67. Found: C, 35.95; H, 2.77; N, 4.65.
(ArN=CMeCMe=NAr)Pt(C6H5)2 (Ar = 2,6-Cl2-C6H3) (18) (GG 2077). To a solution of 114 mg (138 µmol) [Pt2(C6H5)4(µ-S(CH3)2] in 3 ml of CH2Cl2, 96 mg (277 mmol) ArN=CMeCMe=NAr (Ar = 2,6-Cl2-C6H3) were added. After 4 h the solvent was evaporated and the black residue was dissolved in 3 ml of CH2Cl2. The procedure was repeated 3 times and finally the solvent was evaporated under reduced pressure. The black residue (194 mg) was dried in vacuo and recrystallized from 1.2 ml CH2Cl2/pentane 7:5. 70 mg (96.8 µmol, 70%) of black cuboids (m.p.
89 6 Experimental Section
1 >250 °C) were obtained. H NMR (300 MHz, dichloromethane-d2) δ 1.64 (s, 6H, CH3), 6.44 (tt, 4 3 3 J(Ho-Hp) = 1.4 Hz, J(Hm-Hp) = 7.0 Hz, 2H, PtPhp), 6.55 (m, 4H, PtPhm), 6.95 (d, J(Hm-Ho) = 3 3 6.5 Hz, J(Pt-Ho) = 69.9 Hz, 4H, PtPho), 6.97 (m, 2H, ArHp), 7.39 (d, J(Hp-Hm) = 8.2 Hz, 4H, 13 3 ArHm). C NMR (HSQC) (75 MHz, chloroform-d) δ 21.2(CH3), 121.4 (PtPhp), 125.6 ( J(Pt-C) = 2 81.9 Hz, PtPhm), 127.0, 127.7 (Arp), 128.1 (Arm), 136,9 ( J(Pt-C) = 32.9 Hz, PtPho), 142.0, 142.5, 174.6 (C=N). Anal. calcd. for C28H22Cl4N2Pt: C, 46.49, H, 3.07; N, 3.87. Found: C, 46.45; H, 3.11; N, 3.81.
+ - + - [(N-N)Pt(CH3)(H2O)] (BF4 ) ([24] [BF4] ) (GG 2090). To a suspension of 6.0 mg (10 µmol) (ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 2,6-Cl2-C6H3) in 0.6 ml of 2,2,2-trifluoroethanol-d3 was added 0.1 ml of a mixture of TFE-d3 (3.0 ml) and a 50 % solution of HBF4 (60 mg) in water. The 1 product dissolved and was not isolated from the orange solution. H NMR (300 MHz, TFE-d3) δ 0.81 (s, 2J(195Pt-H) = 72.6 Hz, 3 H, PtMe), 1.67 (s, 3 H, NCMeCMeN), 1.91 (s, 3 H, NCMeCMeN), 7.34 (m, 2 H, ArHp), 7.54 (m, 4 H, Hm).
2 + - [(N-N)Pt(η -O2CCH3)] [BArF] ( N-N = ArNCMeCMeNAr, Ar = 2,6-Cl2-C6H3; BArF = (3,5- + - (CF3)2-C6H3)4B ) ([28] [BArF] ) (GG 4166). Under an atmosphere of nitrogen, 22.7 mg + - (37.8 µmol) 1 and 38.3 mg (37.8 µmol) [(Et2O)2H] [BArF] were placed in a 50 ml Schlenk tube and 6 ml of TFE were added. The reaction was stirred for 10 minutes at r.t. and the solvent was removed under reduced pressure. The brown residue was dissolved in 2.5 ml of TFE and 2 ml of acetic acid were added. After 20 minutes the liquids were evaporated under reduced pressure and the brown residue was dissolved in 5 ml of acetic acid. After 80 minutes at r.t. the acid was evaporated under reduced pressure, the residue washed with pentane and dried in vacuo. The product was obtained quantitatively as a brown powder (decomp. >62 °C). 1H NMR (300 MHz, TFE-d3, 295 K) δ 1.30 (s, 3 H, O2CCH3), 2.04 (s, 6 H, NCMeCMeCN), 7.46 (m, 2 H, ArHp), 7.51 (m, 4 H, ArHm), 7.58 (br, 4 H, BArFHp), 7.77 (br, 8 H, BArFHo).Anal. calcd. for C50H27BCl4F24N2O2Pt: C, 40.27; H, 1.82; N, 1.88. Found: C, 40.50; H, 2.04; N, 1.78.
[Pt(O2CCH3)2]4 (29) (GG 3122).To 3.00 g (8.90 mmol) PtCl4 in 180 ml of acetic acid, 7.85 g (47.0 mmol) silver(II)-acetate were added. The mixture was refluxed for 90 min, cooled to 0 °C and filtered. The solvent was evaporated at 60 °C (100 mbar). The dark brown, solid residue (4.0 g) was extracted with dichloromethane for 3 h and dried in vacuo. 1.37 g (1.10 mmol, 12 %) purple powder 1 were obtained. H NMR (300 MHz, CD3OD, 333 K) δ 1.90 (s, 12 H, O2CCH3), 1.92 (s, 12 H, O2CCH3). Anal. calcd. for C16H24O16Pt4: C, 15.34; H, 1.93. Found: C, 15.28; H, 1.89.
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 4-CH3O-C6H4) (32) (GG 4118). 50.0 mg (87.0 µmol) [Pt2(CH3)4(µ-S(CH3)2] were dissolved in 10 ml of toluene with gentle heating. 51.6 mg (174 µmol) ArN=CMeCMe=NAr (Ar = 4-CH3O-C6H4) were added. After 15 h the volume was reduced to 8 ml and the solution was cooled to 4 °C. After 24 h the solution was filtered off, the red powder was washed with pentane, ether, and pentane and dried in vacuo. 53 mg (97 µmol, 71 %) claret powder 1 2 195 (decomp. >250 °C). H NMR (300 MHz, dichloromethane-d2) δ 1.09 (s, J( Pt-H) = 86.7 Hz, 6H, 3 PtMe), 1.45 (s, 6H, NCMeCMeN), 3.86 (s, 6H, OCH3), 6.94 (td, J(H-H) = 2.4 Hz, J(H-H) = 9.3 Hz, 4 H, ArH), 7.09 (td, J(H-H) = 2.4 Hz, 3J(H-H) = 9.0 Hz, 4 H, ArH). Anal. calcd. for C20H26N2O2Pt: C, 46.06, H, 5.02; N, 5.37. Found: C, 45.19; H, 5.07; N, 5.12.
(ArN=CMeCMe=NAr)Pt(CH3)2 (Ar = 4-(CH3)2N-C6H4) (33) (Andre XVI/10). 50.0 mg (87.0 µmol) [Pt2(CH3)4(µ-S(CH3)2] were dissolved in 7 ml of toluene. 56.1 mg (174 µmol) ArN=CMeCMe=NAr (Ar = 4-(CH3)2N-C6H4) were added. After 18 h the volume was reduced to
90 6.5 Deuterated compounds and HBArF
5 ml and the solution cooled to 4 °C. After 24 h the solution was filtered off, the red powder was washed with ether and dried in vacuo. 66 mg (97 µmol, 56 %) red powder (m.p. >250 °C). 1H NMR 2 195 (300 MHz, dichloromethane-d2) δ 0.91 (s, J( Pt-H) = 86.4 Hz, 6H, PtMe), 1.50 (s, 6H, 3 3 NCMeCMeN), 3.00 (s, 6H, NCH3), 6.80 (d, J(H-H) = 9.0 Hz, 4 H, ArH), 6.90 (d, J(H-H) = 9.0 Hz, 4 H, ArH). Anal. calcd. for C22H32N4Pt: C, 48.25, H, 5.89; N, 10.23. Found: C, 47.87; H, 6.15; N, 9.51.
(N,N,N',N'-Tetramethylethylenediamine)Pt(CH3)2 (34) (GG 4123). 51.0 mg (88.8 µmol) [Pt2(CH3)4(µ-S(CH3)2] were suspended in 0.5 ml of TMEDA and 0.5 ml of CH2Cl2 were added. After 2 h the volatiles were removed under reduced pressure. The white solid was dissolved in 0.4 ml of CH2Cl2 and 0.3 ml of TMEDA were added. The volatiles were removed under reduced pressure after 1 h. The solid was dissolved in CH2Cl2/Et2O/pentane (1:2:2), the solution was filtered and a light yellow solution was obtained. The solvents were evaporated and the residue dried in vacuo. 43 mg (126 µmol, 71 %) of a light yellow powder (decomp. >98 °C) were obtained. 1H 2 195 3 195 NMR (300 MHz, benzene-d6) δ 1.09 (s, J( Pt-H) = 88.3 Hz, 6H, PtMe), 1.63 (s, J( Pt-H) = 3 195 11.3 Hz, 4H, CH2), 2.18 (s, J( Pt-H) = 21.3 Hz, 12H, NCH3). Anal. calcd. for C8H22N2Pt: C, 28.15, H, 6.50; N, 8.21. Found: C, 28.15; H, 6.67; N, 6.98.
(N,N,N',N'-Tetramethylethylenediamine)Pt(C6H5)2 (35) (GG 4133). 75.0 mg (91.1 µmol) [Pt2(C6H5)4(µ-S(CH3)2] were dissolved in 0.15 ml CH2Cl2 and 0.6 ml of TMEDA were added. After 1 h the volatiles were removed under reduced pressure. This was repeated with 0.9 ml of CH2Cl2 and 0.5 ml of TMEDA. The beige powder was recrystallized from 0.7 ml of CH2Cl2/pentane 4:3 and dried in vacuo. 24 mg (51 µmol, 56 %) beige powder (decomp. >165 °C). 1H NMR (300 MHz, 3 195 3 195 dichloromethane-d2) δ 2.48 (s, J( Pt-H) = 22.4 Hz, 12H, NCH3), 2.67 (s, J( Pt-H) = 12.3 Hz, 4 3 3 4H, CH2), 6.67 (tt, J(Ho-Hp) = 1.4 Hz, J(Hm-Hp) = 7.3 Hz 2H, PtPhp), 6.79 (t, J(H-H) = 7.5 Hz 3 3 195 4H, PtPhm), 7.36 (d, J(Hm-Ho) = 6.8 Hz 4H, J( Pt-H) = 72.9 Hz, PtPho). Anal. calcd. for C18H26N2Pt: C, 46.44, H, 5.63; N, 6.02. Found: C, 44.12; H, 6.11; N, 5.71.
6.5 Deuterated compounds and HBArF
[Pt2(CD3)4(µ-S(CH3)2] (9) (André XIV/32). To 1.6 g Li in 100 ml of diethyl ether a solution of 10.5 g (72.5 mmol) CD3I in 30 ml of diethyl ether was added in a way that reflux was maintained. The reaction was kept at reflux for 2 h, cooled to r.t. and filtered. Over 15 minutes 9.7 ml of this 0.63 m CD3Li solution (6.11 mmol) were added to the suspension of 880 mg (2.25 mmol) PtCl2((CH3)2S)2 in 30 ml of diethyl ether at 0 °C. The suspension was stirred for 2 h at 0 °C, 0.5 h at r.t. and quenched with 50 ml of a saturated NH4Cl-solution. After phase separation the aqueous layer was extracted 3 times with diethyl ether, the combined organic phases dried over MgSO4 and the solution concentrated to 3 ml. After standing at 4 °C for 16 h a brown precipitate formed which was isolated and was washed 3 times with pentane and dried in vacuo. 460 mg (0.78 µmol, 70 %) light brown powder. Anal. calcd. for C8H12D12Pt2S2 (measured as: C8H24Pt2S2): C, 16.38; H, 2.06; D, 4.12 (H, 4.21). Found: C, 16.64; H, 4.01.
(ArN=CMeCMe=NAr)Pt(CD3)2 (Ar = 2,6-Cl2-C6H3) (17) (André XIII/48). 199 mg (0.53 mmol) ArN=CMeCMe=NAr (Ar = 2,6-Cl2-C6H3) and 187 mg (0.32 mmol) [Pt2(CD3)4(µ-S(CH3)2] were placed in a Schlenk-tube and cooled to –25 °C. 10 ml of toluene were added and the mixture was allowed to warm to r.t. After 24 h the reaction was left to crystallize at 0 °C for 2 h. The solution was filtered off and the residue was washed twice with pentane, ether, and again pentane. After
91 6 Experimental Section
drying in vacuo, 142 mg (0.23 mmol, 36 %) of a black powder (m.p. >250 °C) were obtained. 1H 2 195 NMR (300 MHz, dichloromethane-d2) δ 1.16 (s, J( Pt-H) = 87.0 Hz, 6H, PtMe), 7.25 (t, 3 3 J(Hm-Hp) = 8.1 Hz, 2H, ArHp), 7.55 (d, J(Hp-Hm) = 7.8 Hz, 4 H, ArHm). Anal. calcd. for C18H12D6Cl4N2Pt (measured as: C18H18Cl4N2Pt): C, 35.72; H, 2.00; D,2.00; N, 4.63; (H: 3.03). Found: C, 33.39; H 2.55; N, 4.18.
+ - [(Et2O)2H] [(3,5-(CF3)2-C6H3)4B] (27) (GG 3135). 716 mg (808µmol) [Na][(3,5-(CF3)2-C6H3)4B] were dissolved in 21 ml of diethyl ether. The Schlenk tube was pressurized with HCl and the reaction mixture was stirred for 30 min at r.t. After filtration the volume was reduced to 2 ml, 4 ml of hexane were added and the solution cooled to –20 °C. After 36 h the volume was reduced to 1 ml. This induced formation of a white precipitate. The solvent was filtered off and the solid residue was washed twice with pentane. 750 mg (741 µmol, 92 %) white powder (m.p. 58-68 °C). 1 3 3 H NMR (300 MHz, benzene-d6) δ 0.58 (t, J(CH2-CH3) = 7.2 Hz, 12 H, CH3), 2.92 (q, J(CH3- CH2) = 7.2 Hz, CH2), 7.61 (s, 4 H, ArHp), 8.04 (s, 8 H, ArHo), no signal for the OH was observed. Anal. calcd. for C40H33BF24O2: C, 47.45; H, 3.29. Found: C, 47.36; H, 3.56.
6.6 Copper compounds
[Cu(OH)(tmeda)]2Cl2 (40) (GG 4007). To 302 mg (3.05 mmol) CuCl in 1.5 ml of MeOH with 1 drop of H2O under an atmosphere of oxygen, 701 mg (6.03 mmol) of TMEDA were added. After 30 minutes the suspension was filtered, the residue washed with acetone and dried in vacuo. 528 mg (1.14 mmol, 75 %) fine, brown powder (m.p. 134 °C). IR(KBr): σ(cm-1) = 3346vs, 3007m, 2899m, 2820m, 2015w, 1468s, 1293w, 1135m, 1027s, 952s, 809s, 530s. Anal. calcd. for C12H34Cu2Cl2N4O2: C, 31.03; H, 7.38; N, 12.06. Found: C, 30.32; H, 7.18; N, 11.49.
[Cu(OH)(tmeda)]2OTf2 (41) (GG 4044). To a solution of 300 mg (1.65 mmol) Cu(OAc)2 in 4 ml of H2O, 596 mg (5.13 mmol) of TMEDA were added. After 45 minutes the solution was filtered and 1.26 g (7.32 mmol) NaOTf were added. The reaction vessel was placed in a disiccator with P2O5 for 7 days. Subsequently the mother liquor was removed by filtration, the dark blue quadratic prisms were washed with ether, ground up and were dried in vacuo. 255 mg (369 µmol, 22 %) blue powder (decomp. >165 °C). IR(KBr): σ(cm-1) = 3587m, 2907w, 2859w, 2818w, 1578w, 11469m, 1276vs, 1257vs, 1160s, 1033s, 954m, 810m, 641s. Anal. calcd. for C14H34Cu2F6N4O8S2: C, 24.31; H, 4.95; N, 8.10. Found: C, 24.20; H, 4.98; N, 7.91.
[Cu(OAc)2(tmeda)]2◦H2O (42) (GG 4032). To a solution of 380 mg (3.27 mmol) of TMEDA in 1 ml of MeOH and 50 µl of H2O under an atmosphere of oxygen, 200 mg (1.63 mmol) CuOAc were added. After 22 h the solvent was evaporated under reduced pressure. After washing the residue with ether and drying in vacuo, 326 mg of a turquoise powder were obtained. The crude product was extracted with 16 ml of diethyl ether for 48 h. After filtration and drying in vacuo, 117 mg (191 µmol, 23 %) of a light, blue powder (m.p. 137 °C) were obtained. IR(KBr): σ(cm-1) = 3415 b w, 2982w, 2905w, 2808w, 1631m, 1600s, 1582s, 1464m, 1388m, 1340m, 1310m, 810m, 682m. Anal. calcd. for C20H46Cu2N4O9: C, 39.14; H, 7.56; N, 9.13. Found: C, 39.38; H, 7.08; N, 9.46.
92 6.7 Crystallographic Data
6.7 Crystallographic Data 2 6 14 15
Empirical formula C16H6F10N2 C16H12N2Cl4 C18H16Br2F4N2Pt C14H14N2Pt Formula weight 416.22 374.08 691.24 405.36 Temperature 293(2) K 213(2) K 293(2) K 293(2) K
Radiation CuKα CuKα CuKα CuKa Wavelength 1.54184 Å 1.54180 Å 1.54184 Å 1.54184 Å Crystal system, space group Triclinic, P-1 Orthorhombic, Pbca Tetragonal, P43212 Monoclinic, P21/a Unit cell dimensions a = 6.555(2) Å α = 91.39(3) deg. a = 7.6333(11) Å a = 90 deg. a = 8.507(2) Å α = 90 deg. a = 7.694(3) Å α = 90 deg. b = 7.849(2) Å β = 108.20(3) deg. b = 14.320(2) Å β = 90 deg. b = 8.507(2) Å β = 90 deg. b = 17.519(6) Å β = 107.64(3)deg. c = 8.247(3) Å γ = 104.77(2) deg. c = 15.333(2) Å γ = 90 deg. c = 27.514(6) Å γ = 90 deg. c = 9.324(5) Å γ = 90 deg. Volume 387.3(2) Å3 1676.0(4) Å3 1991.2(8) Å3 1197.7(9) Å3 Z, Calculated density 1, 1.785 Mg/m3 4, 1.482 Mg/m3 4, 2.306 Mg/m3 4, 2.248 Mg/m3 Absorption coefficient 1.748 mm1 6.385 mm-1 18.274 mm-1 21.713 mm-1 F(000) 206 760 1288 760 Approximate crystal size 0.3 x 0.25 x 0.2 mm 0.16 x 0.12 x 0.10 mm 0.1 x 0.1 x 0.02 mm 0.10 x 0.02 x 0.01 mm Diffractometer CAD4, graphite monochromator CAD4, graphite monochromator CAD4, graphite monochromator CAD4, graphite monochromator Range for data collection 5.68 ≤ 2θ ≤ 66.94 deg. 5.77 ≤ 2θ ≤ 69.81 deg. 5.20 to 66.81 deg. 4.98 to 66.89 deg. Index ranges 0 ≤ h ≤ 7, -9 ≤ k ≤ 9, -9 ≤ l ≤ 8 0 ≤ h ≤ 8, 0 ≤ k ≤ 17, 0 ≤ l ≤ 18 0 ≤ h ≤ 10, 0 ≤k ≤ 10, -32 ≤ l ≤ 32 0 ≤ h ≤ 8, 0 ≤ k ≤ 20, -10 ≤ l ≤ 10 Reflections collected / unique 1378 / 1378 [R(int) = 0.0000] 1784 / 1524 [R(int) = 0.000] 4035 / 3057 [R(int) = 0.13] 2216 / 1985 [R(int) = 0.20] Completeness to 2θ = 69.81 0.831 0.829 0.838 0.411 Absorption correction no correction Semi-empirical based on psi-scan PSI PSI Max. and min. transmission not measured 0.99700 and 0.71300 0.99 and 0.53 0.8455 and 0.2200 Structure solution SIR97 SIR92 SIR97 SIR97 Refinement method Full-matrix least-squares on F2 Full-matrix least-squares on F2 Full-matrix least-squares on F2 Full-matrix least-squares on F2 Data / restraints / parameters 1147 / 0 / 140 1524 / 0 / 107 2781 / 1 / 248 876 / 0 / 74 Goodness-of-fit on F2, a 2.101 1.085 1.42 3.154 Final R indices [I>2σ(I)]b R1 = 0.0580, wR2 = 0.2063 R1 = 0.0298, wR2 = 0.0856 R1 = 0.0695, wR2 = 0.1617 R1 = 0.1699, wR2 = 0.3874 Extinction coefficient 0.044(11) 0.0051(3) -0.02(3) not measured Largest diff. peak and hole 0.391 and -0.548 eÅ-3 0.228 and -0.229 eÅ-3 1.486 and -4.459 eÅ-3 9.527 and -8.635 eÅ-3
6 Experimental Section
a 2 2 16 18 41 Goodness-of-fit = 1 / [σ (Fo ) + (0.0410P)2 + 0.8548P], P = 2 2 Empirical formula C18H18N2Cl4Pt C28H22N2Cl4Pt C34Cu2F6N4O8S2, 2(H2O) (Fo + 2Fc ) / 3 b Formula weight 599.23 723.37 727.68 R = Σ IIFoI - IFcII / Σ IFoI, Rw 2 = [(Σ w (IFoI – IFcI) / Σ w Temperature 213(2) K 295(2) K 200(2) K 2 1/2 Fo )] Radiation CuKα MoKα MoKα Wavelength 1.54180 Å 0.71073 Å 0.7107 Å Crystal system, space group Monoclinic, C2/c Monoclinic, P2(1)/c Monoclinic, P 21/m Unit cell dimensions a = 8.060(1) Å α = 90 deg. a = 11.1926(2) Å α = 90 deg. a = 7.7143(2) A α = 90 deg. b = 12.659(1) Å β = 92.19(1) deg. b = 14.0469(3) Å β = 94.24(1) deg. b = 13.6101(3) A β = 98.544(1) deg. c = 19.832(1) Å γ = 90 deg. c = 17.6129(3) Å γ= 90 deg. c = 14.2649(3) A γ = 90 deg. Volume 2022.0(3) Å3 2761.54(9) Å3 1481.08(6) Å 3 Z, Calculated density 4, 1.968 Mg/m3 4, 1.740 Mg/m3 2, 1.632 Mg/m3 Absorption coefficient 17.871 mm-1 5.488 mm-1 1.664 mm-1 F(000) 1144 1400 748 Approximate crystal size 0.15 x 0.10 x 0.03 mm 0.15 x 0.15 x 0.13 mm 0.18 x 0.16 x 0.16 mm Diffractometer CAD4, graphite monochromator Kappa CCD, graphite monochromator Kappa CCD, graphite monochromator Range for data collection 4.46 ≤ 2θ ≤ 64.89 deg. 1.86 ≤ 2θ ≤ 27.50 deg. 6.37 to 27.48 deg. Index ranges 0 ≤ h ≤ 9, 0 ≤ k ≤ 14, -23 ≤ l ≤ 23 -14 ≤h ≤ 14, -17 ≤ k ≤ 18, -22 ≤ l ≤ 22 -9 ≤ h ≤ 10, -16 ≤ k ≤ 17, -18 ≤ l ≤ 18 Reflections collected / unique 1806 / 1672 [R(int) = 0.024] 12065 / 6266 [R(int) = 0.0160] 6297 / 3447 [R(int) = 0.0162] Completeness to 2θ = 69.81 0.463 0.949 0.975 Absorption correction Semi-empirical based on psi-scan no correction None Max. and min. transmission 0.99900 and 0.27000 0.5356 and 0.4932 not measured Structure solution SIR92 SIR92 SIR97 2 Refinement method Full-matrix least-squares on F2 Full-matrix least-squares on F2 Full-matrix least-squares on F Data / restraints / parameters 1672 / 0 / 124 6266 / 0 / 339 3447 / 0 / 225 Goodness-of-fit on F2, a 1.111 1.175 1.014 Final R indices [I>2σ(I)]b R1 = 0.0367, wR2 = 0.1048 R1 = 0.0251, wR2 = 0.0567 R1 = 0.0325, wR2 = 0.1146 Extinction coefficient 0.00059(7) 0.00166(13) 0.020(4) Largest diff. peak and hole 2.014 and -1.699 eÅ-3 1.295 and -0.895 eÅ–3 0.686 and -0.406 eÅ -3
7.1 Derivation of eq. 2.2 and Lambert-Beer's law
7 Appendix
7.1 Derivation of eq. 2.2 and Lambert-Beer's law
In chapter 2.3 the reaction of [24(water)] and benzene to give [25(water)] and methane was investigated by means of UV/Vis spectroscopy. The absorbance at several wavelengths was fitted to eq. 2.2. Eq. 2.2 is derived in the following section.
Investigated was a reaction with the general stoichiometry A + B C + D. Pseudo-first-order conditions were guaranteed by a high concentration of benzene compared to [24(water)] ([benzene]/[Pt] > 14000). The reaction included no back-reaction, as the formation of methane was irreversible under the conditions of the experiments. This lead to the simpler apparent stoichiometry of E P. Thus the rate law for the concentration of E is
–d[E]/dt = kobs[E] eq. 1 Integration results in -kobs·t [E] = [E]0 · e eq. 2 Assuming the reaction goes to completion, denominated by 4,
[E]0 = [P]4 eq. 3 applies. The amount of a substance i is given by
ni = ni,0 - νiξ eq. 4 where ν is the stoichiometric number and ξ is the extent of reaction. Therefore,
nA = nA,0 - ξ eq. 5 and nP = nP,0 + ξ = ξ eq. 6 Division by the volume V yields the corresponding concentrations with x = ξ/V
[E] = [E]0 – x eq. 7 [P] = x eq. 8 and [E] = [E]0 – [P] eq. 9 Use of eq. 2 results in -kobs t -kobs·t [P] = [E]0 - [E]0 · e = [E]0 · (1 - e ) eq. 10 According to Lambert-Beer's law the absorbance of E and P is given byxlix
AE = εE · [E] · d eq. 11 AP = εP · [P] · d eq. 12 and Atot = A = AE + AP = (εE · [E] + εP · [P]) · d eq. 13 where εE and εE are the molar extinction coefficients of the educt and the product and d is the length of the cell. Using eq. 10 yields -kobs·t -kobs·t A = {εP · [E]0 · (1 - e ) + εE · [E]0 - εE · [E]0 · (1 - e )} · d eq. 14 -kobs·t and A = {(εP - εE) · [E]0 · (1 - e ) + εE · [E]0} · d eq. 15 Rewriting eq 15 yields -kobs·t -kobs·t A = εP [E]0 d - εE [E]0 d - εP [E]0 d e + εE [E]0 d e + εE [E]0 d eq. 16 -kobs·t and A = εP [E]0 d + (εE [E]0 d - εP [E]0 d) e eq. 17 applying eqs 3, 11, and 12 yields eq 4 xlix Working under dilute conditions ensured that Lambert-Beer's-law was applicable.
95 7 Appendix
-kobs·t A = A4 + (A0 - A4) · e eq. 2.2
For isosbestic points, where εE = εP, the time-dependent part of eq 15 becomes 0, and
A = const = εE · [E]0 eq. 18 The absorbance does not vary over time.
7.2 Principle of Linear Initial Rate
The rate of the background reaction in the absence of benzene (Figure 29b, p. 41) was determined by fitting the data to a linear equation of the form of y = a x + b
Eq. 19 is of the general form -kt y = const + a · e eq. 19 the slope of which is given by -kt dy/dt = - a · k · e eq. 20 which for t = 0 yields dy/dt = - ak eq. 21 displaying a linear correlation between the slope of the fit to the data in Figure 29b and the observed rate constant k for small ζ.
96 8 References
8 References
1 Gradassi, M.J.; Green, N.W. Fuel Proc. Technol. 1995, 42, 65. 2 Arndtsen, B.A.; Bergman, R.G.; Mobley, T.A.; Peterson, T.H. Acc. Chem. Res. 1995, 28, 154. 3 Reserves of natural gas rival those of liquid petroleum: Lunsford, J.H. Catal. Today 2000, 63, 165. 4 March, J. Advanced Organic Chemistry, 4th ed.; Wiley-Interscience: New York; 1992. 5 CRC Handbook of Chemistry and Physics, 82nd ed., Lide, D.R. ed.; CRC Press: Boca Raton, 2001. 6 Labinger, J.A.; Bercaw, J.E. Nature 2002, 417, 507. 7 Sen, A. in Applied Homogeneous Catalysis with Organometallic Compounds, Cornils, B.; Herrmann, W.A. eds.; Wiley-VCH: Weinheim, 2002, 1081. 8 Crabtree, R.H. J. Chem. Soc., Dalton Trans. 2001, 2437. 9 Seitz, O. Nachr. Chem. 2001, 49, 777. 10 Stahl, S.S.; Labinger, J.A.; Bercaw, J.E. Angew. Chem. Int. Ed. 1998, 37, 2180. 11 Shilov, A.E.; Shulpin, G.B. Chem. Rev. 1997, 97, 2879. 12 Halpern, J. Inorg. Chim. Acta 1985, 100, 41. 13 Strassner, T. in Applied Homogeneous Catalysis with Organometallic Compounds; Cornils, B.; Herrmann, W.A. eds.; Wiley-VCH: Weinheim, 2002, 737. 14 Shilov, A.E.; Shul'pin, G.B. Activation and catalytic Reactions of saturated Hydrocarbons in the Presence of Metal Complexes, Kluwer: Dordrecht, 2000. 15 Leitch, L.C. Can. J. Chem. 1954, 32, 813. 16 Garnett, J.L.; Hodges, R.J. J. Am. Chem. Soc. 1967, 89, 4546. 17 Gol'dshleger, N.F.; Tyabin, M.B.; Shilov, A.E.; Shteinman, A.A. Zh. Fiz. Khim. 1969, 43, 2174 (in Russian). Gol’dshleger, N.F.; Tyabin, M.B.; Shilov, A.E.; Shteinman, A.A. Russ. J. Phys. Chem. 1969, 43, 1222. 18 Gol’dshleger, N.F.; Es’kova, V.V.; Shilov, A.E.; Shteinman, A.A. Russ. J. Phys. Chem. 1972, 46, 785. 19 Kushch, L.A.; Lavrushko, V.V.; Misharin, Y.S.; Moravskii, A.P.; Shilov, A.E. Nouv. J. Chim. 1983, 7, 729. 20 Fekl, U.; Golberg, K.I. Adv. Inorg. Chem. 2003, 54, 259. 21 Stahl, S.S.; Labinger, J.A.; Bercaw, J.E. J. Am. Chem. Soc. 1996, 118, 5961. 22 Reinartz, S.; White, P.S.; Brookhart, M.; Templeton, J.L. Organometallics 2000, 19, 3854. 23 Reinartz, S.; White, P.S.; Brookhart, M.; Templeton, J.L. Organometallics 2001, 20, 1709. 24 Wick, D.D.; Goldberg, K.I. J. Am. Chem. Soc. 1997, 119, 10235. 25 Jensen, M.P.; Wick, D.D.; Reinartz, S.; White, P.S.; Templeton, J.L.; Goldberg, K.I. J. Am. Chem. Soc. 2003, 125, 8614. 26 a) Cotton, F.A.; Wilkinson, G.; Murillo, C.A.; Bochmann, M. Advanced Inorganic Chemistry, 6th ed.; Wiley-Interscience: New York, 1999, pp. 1080. b) Hollemann, A.F.; Wiberg, E.; Wiberg, N. Lehrbuch der Anorganischen Chemie; Walter de Gruyter: Berlin, 1985, pp. 1204. 27 Reinartz, S.; White, P.S.; Brookhart, M.; Templeton, J.L. J. Am. Chem. Soc. 2001, 123, 12724. 28 Norris, C.M.; Reinartz, S.; White, P.S.; Templeton, J.L. Organometallics 2002, 21, 5649.
97 8 References
29 Scollard, J.D.; Day, M.; Labinger, J.A.; Bercaw, J.E. Helv. Chim. Acta 2001, 84, 3247. 30 Holtcamp, M.W.; Henling, L.M.; Day, M.W.; Labinger, J.A.; Bercaw, J.E. Inorg. Chim. Acta 1998, 270, 467. 31 Johansson, L.; Tilset, M.; Labinger, J.A.; Bercaw, J.E. J. Am. Chem. Soc. 2000, 122, 10846. 32 Lo, H.C.; Haskel, A.; Kapon, M.; Keinan, E. J. Am. Chem. Soc. 2002, 124, 3226. 33 Crumpton, D.M.; Goldberg, K.I. J. Am. Chem. Soc. 2000, 122, 962. 34 Reinartz, S.; White, P.S.; Brookhart, M.; Templeton, J.L. J. Am. Chem. Soc. 2001, 123, 6425. 35 Fekl, U.; Kaminsky, W.; Goldberg, K.I. J. Am. Chem. Soc. 2001, 123, 6423. 36 Romeo, R.; Scolaro, L.M.; Plutino, M.R.; de Biani, F.F.; Bottari, G.; Romeo, A. Inorg. Chim. Acta 2003, 35 , 143. 37 Romeo, R.; Plutino, M.R.; Scolaro, L.M.; Stoccoro, S.; Minghetti, G. Inorg. Chem. 2000, 39, 4749. 38 Plutino, M.R.; Scolaro, L.M.; Romeo, R.; Grassi, A. Inorg. Chem. 2000, 39, 2712. 39 Romeo, R.; Scolaro, L.M.; Nastasi, N.; Arena, G. Inorg. Chem. 1996, 35, 5087. 40 Minniti, D.; Alibrandi, G.; Tobe, M.L.; Romeo, R. Inorg. Chem. 1987, 26, 3956. 41 Romeo, R.; Plutino, M.R.; Elding, L.I. Inorg. Chem. 1997, 36, 5909. 42 Romeo, R.; Alibrandi, G. Inorg.Chem. 1997, 36, 4822. 43 Romeo, R. Comments Inorg. Chem. 1990, 11, 21. 44 Baratta, W.; Stoccoro, S.; Doppiu, A.; Herdtweck, E.; Zucca, A.; Rigo, P. Angew. Chem. Int. Ed. 2003, 42, 105. 45 Carr, N.; Mole, L.; Orpen, A.G.; Spencer, J.L. J. Chem. Soc., Dalton Trans. 1992, 2653. 46 Mole, L.; Spencer, J. L.; Carr, N.; Orpen, A.G. Organometallics 1991, 10, 49. 47 Goel, R.M.; Srivastava, R.C. Can. J. Chem. 1983, 61, 1352. 48 Goel, R.G.; Srivastava, R.C. J. Organomet. Chem. 1981, 204, C13. 49 Scherer, O.J.; Jungmann, H. J. Organomet. Chem. 1982, 228, C61. 50 Johansson, L.; Tilset, M. J. Am. Chem. Soc. 2001, 123, 739. 51 Huheey, J.E.; Keiter E.A.; Keiter, R.L. Anorganische Chemie, 2nd ed., de Gruyter: Berlin, 1995. 52 Procelewska, J.; Zahl, A.; van Eldik, R.; Zhong, H.A.; Labinger, J.A.; Bercaw, J.E. Inorg. Chem. 2002, 41, 2808. 53 Fekl, U.; Zahl, A.; van Eldik, R. Organometallics 1999, 18, 4156. 54 Zhong, H.A.; Labinger, J.A.; Bercaw, J.E. J. Am. Chem. Soc. 2002, 124, 1378. 55 Holtcamp, M.W.; Labinger, J.A.; Bercaw, J.E. J. Am. Chem. Soc. 1997, 119, 848. 56 Johansson, L.; Ryan, O.B.; Tilset, M. J. Am. Chem. Soc. 1999, 121, 1974. 57 Johansson, L.; Ryan, O.B.; Rømming, C.; Tilset, M. Organometallics 1998, 17, 3957. 58 Heiberg, H.; Johansson, L.; Gropen, O.; Ryan, O.B.; Swang, O.; Tilset, M. J. Am. Chem. Soc. 2000, 122, 10831. 59 Johansson, L.; Ryan, O.B.; Rømming, C.; Tilset, M. J. Am. Chem. Soc. 2001, 123, 6579. 60 Brown, W.G.; Garnett, J.L. J. Am. Chem. Soc. 1958, 80, 5272. 61 Fraser, R.R.; Renaud, R.N. J. Am. Chem. Soc. 1966, 88, 4365.
98 8 References
62 Anderson, J.R.; Kemball, C. Adv. Catal. 1957, 9, 51. 63 Garnett, J.L. Cat. Rev. 1971, 5, 229. 64 Hodges, R.J.; Webster, D.E.; Wels, P.B. J. Chem. Soc. A 1971, 3230. 65 Blackett, L.; Gold, V.; Reuben, D.M.E. J. Chem. Soc., Perkin Trans. 2 1974, 1869. 66 Garnett, J.L; Hodges, R.J.; Kenyon, R.S.; Long, M.A. J. Chem. Soc., Perkin Trans. 2 1979, 885. 67 Garnett, J.L; West, J.C. Aust. J. Chem. 1974, 27, 129. 68 Hodges, R.J.; Garnett, J.L. J. Phys. Chem. 1968, 72, 1673. 69 See ref. 26a, ch. 21, pp. 1167. 70 Shevchenko, V.P.; Nagaev, I.Y.; Shevchenko, K.V.; Myasoedov, N.F. Radiochemistry 2002, 44, 593. 71 Myasoedov, N.F.; Sidorov, G.V. J. Labelled Comp. Radiopharm. 1998, 41, 993. 72 Williams, P.G.; Lukey, C.A.; Long, M.A.; Garnett, J.L. J. Labelled Comp. Radiopharm. 1991, 29, 175. 73 Kanski, R.; Kanski, M. J. Radioanal. Nucl. Chem. 2003, 257, 385 and references therein. 74 Kakiuchi, F.; Chatani, N. Adv. Synth. Catal. 2003, 345, 1077. 75 Jia, C.; Kitamura, T.; Fujiwara, Y. Acc. Chem. Res. 2001, 34, 633. 76 Goel, A.B. US Patent 4515983, 1985. 77 Gretz, E.; Oliver, T.F.; Sen, A. J. Am. Chem. Soc. 1987, 109, 8109. 78 Kao, L.C.; Hutson, A.C.; Sen, A. J. Am. Chem. Soc. 1991, 113, 700. 79 Periana, R.A.; Taube D.J.; Evitt, E.R.; Löffler, D.G.; Wentreck, P.R.; Voss, G.; Masuda, T. Science 1993, 259, 340. 80 a) Periana, R.A.; Taube D.J.; Gamble, S.; Taube, H.; Satoh, T.; Fujii, H. Science 1998, 280, 560. b) R.A. Periana, WO Patent 9850333, 1998. 81 Klei, S.R.; Golden, J.T.; Tilley, T.D.; Bergman, R.G. J. Am. Chem. Soc. 2002, 124, 2092. 82 Jintoku, T.; Taniguchi, H.; Fujiwara, Y. Chem. Lett. 1987, 1865. 83 Jintoku, T.; Takaki, K.; Fujiwara, Y.; Fuchita, Y.; Hiraki, K. Bull. Chem. Soc. Jpn. 1990, 63, 438. 84 a) Van den Beuken, E.K.; Feringa, B.L. Tetrahedron 1998, 54, 12985. b) for chiral analogs see: Steinhagen, H.; Helmchen, G. Angew. Chem. Int. Ed. Engl. 1996, 35, 2339. 85 Fujiwara, Y.; Takaki, K.; Taniguchi, J. Synlett 1996, 591. 86 Mukhopadhyay, S.; Rothenberg; G.; Lando, G.; Agbaria, K.; Kazanci, M.; Sasson, Y. Adv. Synth. Catal. 2001, 343, 455. 87 Okamoto, M.; Yamaji, T. Chem.Lett. 2001, 212. 88 Ullmann, F. Ber. Dtsch. Chem. Ges. 1903, 36, 2382. 89 Ullmann, F. Ber. Dtsch. Chem. Ges. 1904, 37, 853. 90 Ley, S.V.; Thomas, A.W. Angew. Chem. Int. Ed. 2003, 42, 5400. Correction: Ley, S.V.; Thomas, A.W. Angew. Chem. Int. Ed. 2004, 43, 1043. 91 Ma, D.; Cai, Q. Org. Lett. 2003, 5, 3799. 92 Kwong, F.Y.; Buchwald, S.L. Org. Lett. 2003, 5, 793. 93 Kwong, F.Y.; Klapars, A.; Buchwald, S.L. Org. Lett. 2002, 4, 581. 94 Chan, D.M.T.; Monaco, K.L.; Wang, R.P.; Winters, M. P. Tetrahedron Lett. 1998, 39, 2933.
99 8 References
95 Chan, D.M.T.; Monaco, K.L.; Li, R.; Bonne, D.; Clark, C.G.; Lam, P.Y.S. Tetrahedron Lett. 2003, 44, 3863. 96 Lam, P.Y.S.; Vincent, G.; Bonne, D.; Clark, C.G. Tetrahedron Lett. 2002, 43, 3091. 97 Lam, P.Y.S.; Deudon, S.; Hauptmann, E.; Clark, C.G. Tetrahedron Lett. 2001, 42, 2427. 98 Coles, S.J.; Costello, J.F.; Hursthouse, M.B.; Smith, S. J. Organomet. Chem. 2002, 662, 98. 99 Barton, D.H.R.; Finet, J.P.; Khamsi, J. Tetrahedron Lett. 1987, 28, 887. 100 Quach, T.D.; Batey, R.A. Org. Lett. 2003, 5, 1381. 101 Beringer, F.M.; Brierley, A.; Drexler, M.; Gindler, E.M.; Lumpkin, C.C. J. Am. Chem. Soc. 1953, 75, 2708. 102 Evans, D.A.; Katz, J.L.; West, T.R. Tetrahedron Lett. 1998, 39, 2937. 103 Lam, P.Y.S.; Vincent, G.; Clark, C.G.; Deudon, S.; Jadhav, P.K. Tetrahedron Lett. 2001, 42, 3415. 104 Collman, J.P.; Zhong, M. Org. Lett. 2000, 2, 1233. 105 Collman, J.P.; Zhong, M.; Zhang, C.; Costanzo, S. J. Org. Chem. 2001, 66, 7892. 106 Combes, S.; Finet, J.P. Tetrahedron Lett. 1999, 55, 3377. 107 Antilla, J.C.; Buchwald, S.L. Org. Lett. 2001, 3, 2077. 108 Electrospray Ionization Mass Spectrometry, Cole, R.D. ed; Wiley: New York, 1997. 109 Whitehouse, C.M.; Dreyer, R.N.; Yamashita, M.; Fenn, J.B. Anal. Chem. 1985, 57, 675. 110 Fenn, J.B. Angew. Chem. Int. Ed. 2003, 42, 3871. 111 Van Berkel, G.J.; McLuckey, S.A.; Glish, G.L. Anal. Chem. 1991, 63, 2064. 112 Van Berkel, G.J.; McLuckey, S.A.; Glish, G.L. Anal. Chem. 1991, 63, 1098. 113 Kane-Maguire, L.A.P.; Kanitz, R.; Sheil, M.M. Inorg. Chim. Acta 1996, 245, 209. 114 Aquilina, J.A.; Robinson, C.V. Biochem. J. 2003, 375, 323. 115 Gerlich, D. Adv. Chem. Phys. 1992, 82, 1. 116 Stafford, G. J. Am.Soc. Mass Spectrom. 2002, 13, 589. 117 Bakhtiar, R.; Nelson, R.W. Biochem. Pharm. 2000, 59, 891. 118 Przybylski, M.; Glocker, M.O. Angew. Chem. Int. Ed. Engl. 1996, 35, 806. 119 Wind, M.; Lehmann, W.D. J. Anal. At. Spectrom. 2004, 19, 20. 120 Plattner, D.A. Int. J. Mass Spectrom. 2001, 207, 125. 121 Plattner, D.A. Top. Curr. Chem. 2003, 225, 153. 122 Chen, P. Angew. Chem. Int. Ed. 2003, 42, 2832. 123 a) Noyori, R.; Ohkuma, T. Angew. Chem. Int. Ed. 2001, 40, 40. b) Kaminsky, W. Catal. Today 2000, 62, 23. 124 Katta, V.; Chowdhury, S.K; Chait, B.T. J. Am. Chem. Soc. 1990, 112, 5348. 125 Spence, T.G.; Burns, T.D.; Posey, L.A. J. Phys. Chem. A 1997, 101, 139. 126 Hinderling, C.; Plattner, D.A.; Chen, P. Angew. Chem. Int. Ed. Engl. 1997, 36, 243. 127 Hinderling, C.; Feichtinger, D.; Plattner, D.A.; Chen, P. J. Am. Chem. Soc. 1997, 119, 10793. 128 Feichtinger, D.; Plattner, D.A. Angew. Chem. Int. Ed. Engl. 1997, 36, 1718.
100 8 References
129 Plattner, D.A.; Feichtinger, D.; El-Bahraoui, J.; Wiest, O. Int. J. Mass Spectrom. 2000, 195-196, 351. 130 Feichtinger, D.; Plattner, D.A. J. Chem. Soc. Perkin Trans. 2 2000, 1023. 131 Feichtinger, D; Plattner, D.A. Chem. Eur. J. 2001, 7, 591. 132 Feichtinger, D.; Plattner, D.A.; Chen, P. J. Am. Chem. Soc. 1998, 120, 7125. 133 Hinderling, C.; Chen, P. Angew. Chem. Int. Ed. 1999, 38, 2253. 134 Hinderling, C.; Chen, P. Int. J. Mass Spectrom. 2000, 195-196, 377. 135 Hinderling, C.; Adlhart, C.; Chen, P. Chimia 2000, 54, 232. 136 Kim, Y.M.; Chen, P. Int. J. Mass Spectrom. 1999, 185, 871. 137 Kim, Y.M.; Chen, P. Int. J. Mass Spectrom. 2000, 202, 1. 138 Hartmann, R.; Chen, P. Angew. Chem. Int. Ed. 2001, 19, 3581. 139 Hinderling, C.; Adlhart, C.; Chen, P. Angew. Chem. Int. Ed. 1998, 37, 2685. 140 Adlhart, C.; Hinderling, C.; Baumann, H.; Chen, P. J. Am. Chem. Soc. 2000, 122, 8204. 141 Adlhart, C.; Volland, M.A.O.; Hofmann, P.; Chen, P. Helv. Chim. Acta 2000, 83, 3306. 142 Adlhart, C.; Chen, P. Helv. Chim. Acta 2000, 83, 2192. 143 Volland, M.A.O.; Adlhart, C.; Kiener, C.A.; Chen, P.; Hofmann, P. Chem. Eur. J. 2001, 7, 4621. 144 Adlhart, C.; Chen, P. Helv. Chim. Acta 2003, 86, 941. 145 Zhang, X.; Chen, P. Chem. Eur. J. 2003, 9, 1852. 146 Chen, X.; Zhang, X.; Chen, P. Angew. Chem. Int. Ed. 2003, 42, 3798. 147 Gerdes, G.; Chen, P. Organometallics 2003, 22, 2217. 148 Gerdes, G.; Chen, P. Organometallics 2004, 23, 3031. 149 Markert, C.; Pfaltz, A. Angew. Chem. Int. Ed. 2004, 43, 2498. 150 Rodgers, M.T.; Armentrout, P.B. in Comprehensive Coordination Chemistry II; McCleverty, J.A.; Meyer, T.J. eds.,; Elsevier Ltd.: Oxford, 2004, pp. 141. 151 Zhang, X.G.; Liyanage, R.; Armentrout, P.B. J. Am. Chem. Soc. 2001, 123, 5563. 152 Schwarz, H.; Schröder, D. Pure Appl. Chem. 2000, 72, 2319. 153 Brönstrup, M. Organometallic Ion Chemistry of Iron and Platinum in the Gas Phase; Shaker Verlag: Aachen, 2000, ch. 6 (Ph.D. thesis, Technische Universität Berlin, 1999). 154 Koszinowski, K.; Schröder, D.; Schwarz, H. Organometallics 2004, 23, 1132. 155 Koszinowski, K.; Schröder, D.; Schwarz, H. Angew. Chem. Int. Ed. 2004, 43, 121. 156 Koch, W.; Holthausen, M.C. A Chemist's Guide to Density Functional Theory; Wiley-VCH: Weinheim, 2000. 157 Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864. 158 Kohn, W.; Sham, L.J. Phys. Rev. 1965, 140, A1133. 159 Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T. A.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.; Al- Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.;
101 8 References
Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, revision C.3, Gaussian, Inc.: Pittsburgh, PA, 1995. 160 Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al- Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998. 161 Siegbahn, P.E.; Crabtree, R.H. J. Am. Chem. Soc. 1996, 11, 4442. 162 Hill, G.S.; Puddephatt, R.J. Organometallics 1998, 17, 1478. 163 Transition metal alkane complexes have been reviewed: Hall, C.; Perutz, R. Chem. Rev. 1996, 96, 3125. 164 Heiberg, H.; Swang, O.; Gropen, O J. Phys. Chem. A 1999, 103, 10004. 165 Sakaki, S.; Mizoe, N.; Musashi, Y.; Biswas, B.; Sugimoto, M. J. Phys. Chem. A 1998, 102, 8027. 166 Su, M.D.; Chu, S.Y. Organometallics 1997, 16,1621. 167 Iron, M.A.; Lo, H.C.; Martin, J.L.M.; Keinan, E. J. Am. Chem. Soc. 2002, 124, 7041. 168 Bartlett, K.L.; Goldberg, K.I.; Borden, W.T. J. Am. Chem. Soc. 2000, 122, 1456. 169 Kua, J.; Xu, X.; Periana, R.A.; Goddard III, W.A. Organometallics 2002, 21, 511. 170 Hristov, I.H.; Ziegler, T. Organometallics 2003, 22, 1668. 171 Rodgers, M.T.; Armentrout, P.B. Mass Spectrom. Rev. 2000, 19, 215. 172 tom Dieck, H.; Svoboda, M.; Greiser, T. Z. Naturforsch. 1981, 36b, 823. 173 Hinderling, C. Elektrospray-Tandem-Massenspektrometrie als vielseitiges Werkzeug zum mechanistischen Studium von organometallischen Reaktionen, PhD thesis Nr. 13158, ETH Zürich, 1999. 174 Cox, E.G.; Saenger, H.; Wardlaw, W. J. Chem Soc. 1934, 182. 175 Scott, J.D.; Puddephatt, R.J. Organometallics 1983, 2, 1643. 176 Ziegler, K.; Colonius, H. Justus Liebigs Ann. Chem. 1930, 479, 135. 177 Hadj-Bagheri, N.; Puddephatt, R.J. Polyhedron 1988, 7, 2695. 178 Becke, A.D. Phys. Rev. A 1988, 38, 3098. Becke, A.D. J. Chem. Phys. 1993, 98, 1372. Becke, A.D. J. Chem. Phys. 1993, 98, 5648. 179 Lee, C.; Yang, W., Parr, R.G. Phys. Rev. B 1988, 37, 785. 180 Scott, A.P.; Radom, L. J. Chem. Phys. 1996, 100, 16502. 181 Perdue, J.P.; Wang, Y. Phys. Rev. B 1992, 45, 13244. 182 Wysokinski, R.; Michalska, D. J. Comp. Chem. 2001, 22, 901. 183 Urtel, H.; Meier, C.; Eisenträger, F.; Rominger, F.; Joschek, J.P.; Hofmann, P. Angew. Chem. Int. Ed. 2001, 40, 781.
102 8 References
184 a) Bergner, A.; Dolg, M.; Kuechle, W.; Stoll, H.; Preuss, H. Mol. Phys. 1993, 80, 1431. b) Dolg, M.; Stoll, H.; Preuss, H.; Pitzer, R.M. J. Phys. Chem. 1993, 97, 5852. 185 Hariharan, P.C.; Pople, J.A. Theor. Chim. Acta 1973, 28, 213. 186 Kim, Y.M. A Study of Cationic Rhodium Complexes in the Gas Phase by Electrospray Ionization Mass Spectrometry – Towards a Model of the Diamond Surface, PhD thesis No. 13315, ETH Zürich, 1999. 187 Gerdes, G.; Chen, P. unpublished results. 188 When electron-rich 6-substituted 2,2'-bipyridines ligate Pt(II) related reactions are observed in solution: Minghetti, G.; Stoccoro, S.; Cinellu, M.A.; Soro, B.; Zucca, A. Organometallics 2003, 22, 4770 and references cited therein. 189 see chapter 4.3 in 173. 190 Feichtinger, D. Untersuchungen der intrinsischen Reaktivität von Metallorganischen Katalysatoren mittels Elektrospray Tandem Massenspektrometrie, PhD thesis Nr. 13595, ETH Zürich, 2000, chapter 6B. 191 Albright, T.A.; Burdett, J.K.; Whangbo, M.H. Orbital Interactions in Chemistry; Wiley- Interscience: New York, 1985. 192 Wik, R.J.; Lersch, M.; Tilset, M. J. Am. Chem. Soc. 2002, 124, 12116. 193 Chapter 5.6 in 173. 194 Coordination of benzene together with displacement of a solvent molecule from [(diimine)Pt(Me)(solv)]+ was found to be rate-determining depending on the ligand (see ref. 54). 195 a) James, C.S.; Acree, W.E. J. Phys. Chem. Ref. Data 2003, 32, 519. b) Konicek, J. Acta Chem. Scand. 1973, 27, 1496. c) Davis, R.T.; Schiessler, R.W. J. Phys. Chem. 1953, 57, 966. 196 Gioumousis, G.; Stevenson, D.P. J. Chem. Phys. 1958, 29, 294. 197 Comparable isotope effects were found by Tilset (ref. 56) and Bercaw (ref. ) for the C-H activation of benzene by related (diimine)Pt complexes in solution. Nevertheless it should be noted that the observed isotope effect in the gas-phase is not necessarily the same as the one for the C-H activation reaction in solution. 198 O'Hanlon, J.F. A User's Guide to Vacuum Technology; Wiley-Interscience: New York, 1980. 199 Atkins, P.W. Physikalische Chemie; VCH: Weinheim, 1990. 200 Andreevskii, D. N.; Antonova, Z. A. Russ. J. Phys. Chem. (Engl.Transl.) 1983, 57, 1340. Zh. Fiz. Khim. 1983, 57; 2202. 201 a) Wolf, J. F.; Harch, P. G.; Taft, R. W. J. Am. Chem. Soc. 1975, 97, 2904. b) Farneth, W. E.; Brauman, J. I. J. Am. Chem. Soc. 1976, 98, 7891. c) Aue, D. H.; Webb, H. M.; Bowers, M. T. J. Am. Chem. Soc. 1976, 98, 311. d) Olmstead, W. N.; Brauman, J. I. J. Am. Chem. Soc. 1977, 99, 4219. e) Staley, R. H.; Wieting, R. D.; Beauchamp, J. L. J. Am. Chem. Soc. 1977, 99, 5964. f) Asubiojo, O. I.; Brauman, J. I. J. Am. Chem. Soc. 1979, 101, 3715. g) Pellerite, M. J.; Brauman, J. I. J. Am. Chem. Soc. 1980, 102, 5993. h) Arnett, E. M.; Peinta, N. J. J. Am. Chem. Soc. 1980, 102, 3329. i) Sharma, S.; Kebarle, P. J. Am. Chem. Soc. 1982, 104, 19. j) Böhme, D. K.; MacKay, G. I. J. Am. Chem. Soc. 1981, 103, 978. k) Caldwell, G.; Magnera, T. F.; Kebarle, P. J. Am. Chem. Soc. 1984, 106, 959. l) McMahon, T. B.; Heinis, T.; Nicol, G.; Hovey, J. K.; Kebarle, P. J. Am. Chem. Soc. 1988, 110, 7591. 202 Perkampus, H.H.; Kaufmann, R. Kinetische Analyse mit Hilfe der UV-VIS-Spektrometrie; VCH: Weinheim, 1991.
103 8 References
203 Armarego, W.L.F.; Perrin, D.D. Purification of Laboratory Chemicals, 4th ed.; Butterworth- Heinemann: Oxford, 1997. 204 Konze, W.V.; Scott, B.L.; Kubas, G.J. J. Am. Chem. Soc. 2002, 124, 12550. 205 a) Becker, H. G. O.; Berger, W.; Domschke, G.; Fanghänel, E.; Faust, J.; Fischer, M.; Gentz, F.; Gewald, K.; Gluch, R.; Mayer, R.; Müller, K.; Pavel, D.; Schmidt, H.; Schollberg, K.; Schwetlick, K.; Seiler, E.; Zeppenfeld, G. Organikum. Organisch-chemisches Praktikum; Barth Verlaggesellschaft: Leipzig, 1993; p 662. b) Stephen, H.; Stephen, T., Eds.; Solubilities of Inorganic and Organic compounds; Pergamon Press: Oxford, 1963, Vol. 1, p. 433. 206 It is generally conceded that the standard deviations from an Eyring plot seriously underestimate the errors in ∆H≠ and ∆S≠, in first place because the introduction of systematic errors may result in tilting of an Eyring plot without seriously affecting its linearity. Sandström, J. Dynamic NMR Spectroscopy; Academic Press: London, 1982, p. 108. 207 a) Adlhart, C. Intrinsic Reactivity of Ruthenium Carbenes: A Combined Gas Phase and Computational Study, PhD thesis Nr. 15073, ETH Zürich, 2003, p. 20. b) Published in ref. 122. 208 Albietz, P.J.; Yang, K.; Eisenberg, R. Organometallics 1999, 18, 2747. 209 a) Dunning, T.H.; Hay, P.J. in Modern Theoretical Chemistry, Theory, Vol. 3, Schaefer III, H.F., ed.; Plenum Press: New York, 1976. b) Hay, P.J.; Wadt, W.R. J. Chem. Phys. 1985, 82, 270. b) Wadt, W.R.; Hay, P.J. J. Chem. Phys. 1985, 82, 284. d) Hay, P.J.; Wadt, W.R. J. Chem. Phys. 1985, 82, 299. 210 Peng, C.; Schlegel, H.B. Isr. J. Chem. 1994, 33, 449. 211 Peng, C.; Ayala, P.Y.; Schlegel, H.B.; Frisch, M.J. J. Comp. Chem. 1996, 17, 49. 212 Norris, C.M.; Templeton, J.L. Organometallics 2004, in press. 213 Shilov, A.E. Activation of Saturated Hydrocarbons by Transition Metal Complexes; Reidel: Boston, 1984. 214 Crabtree, R.H. Chem. Rev. 1995, 95, 987. 215 Heyduk, A.F.; Labinger, J.A.; Bercaw, J.E. J. Am. Chem. Soc. 2003, 125, 6366. 216 Dangel, B.D.; Johnson, J.A.; Sames, D. J. Am. Chem. Soc. 2001, 123, 8149. 217 Iverson, C.N.; Smith, M.R. J. Am. Chem. Soc. 1999, 121, 7096. 218 Chen, H.; Schlecht, S.; Semple, T.C.; Hartwig, J.F. Science 2000, 287, 1995. 219 For Pd-catalysis: a) Periana, R.A.; Mironov, O.; Taube, D.; Bhalla, G.; Jones, C.J. Science 2003, 301, 814. b) Muehlhofer, M.; Strassner, T.; Herrmann, W.A. Angew. Chem. Int. Ed. 2002, 41, 1745. c) Dick, A.R.; Hull, K.L.; Sanford, M.S. J. Am. Chem. Soc. 2004, 126, 2300. d) Ackerman, L.J.; Sadighi, J.P.; Kurtz, D.M., Labinger, J.A.; Bercaw, J.E. Organometallics 2003, 22, 3884. 220 For V-catalysis: Nizova, G.V.; Süss-Fink, G.; Stanislas, S.; Shul'pin, G. Chem. Commun. 1998, 1885. 221 For Ca-catalysis: Asadullah, M.; Kitamura, T.; Fujiwara,Y. Angew. Chem. Int. Ed. 2000, 39, 2475. 222 For I-catalysis: Periana, R.A.; Mirinov, O.; Taube, D.J.; Gamble, S. Chem. Commun. 2002, 2376. 223 For Ir-catalysis: a) Matsumoto, T.; Periana, R.A.; Taube, D.J.; Yoshida, H. J. Mol. Cat. A 2002, 180, 1. b) Periana, R.A.; Liu, X.Y.; Bhalla, G. Chem. Comm. 2002, 3000.
104 8 References
224 For Rh-catalysis: Matsumoto, T.; Periana, R.A.; Taube, D.J.; Yoshida, H. J. Catal. 2002, 206, 272. 225 a) Brookhart, M.; Grant, B ; Volpe, A.F. Organometallics 1992, 11, 3920. b) Brookhart, M.; Volpe, A.F.; DeSimone, J.M.; Lamanna, W.M. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 1991, 32, 461. c) Taube, R.; Wache, S. J. Organometal. Chem. 1992, 428, 431. 226 Wright, D. US Patent 3652613, 1972. 227 a) C.T. Carrondo, M.A.A.F. De; Skapski, A.C. J. Chem. Soc., Chem. Commun. 1976, 410. b) C.T. Carrondo, M.A.A.F. De; Skapski, A.C. Acta Cryst. B 1978, 34, 1857. 228 a) Yamaguchi, T.; Ueno, T.; Ito, T. Inorg. Chem. 1993, 32, 4996. b) Yamaguchi, T.; Shibata, A.; Ito, T. Chem. Eur. J. 2001, 7, 5409. 229 Golden, J.T.; Andersen, R.A.; Bergman, R.G. J. Am. Chem. Soc. 2001, 123, 5837. Kinetic analyses need to be done at low conversions (p. 13 in supporting information). 230 a) Powell, P. Principles of Organometallic Chemistry, 2nd ed.; Chapman and Hall: New York, 1988. b) Scherer, O.J.; Hussong, K.; Wolmershäuser, G. J. Organomet. Chem. 1985, 289, 215. 231 Compare the changes in rate-determining transition state upon subtle changes in the diimine ligand in ref. 54. 232 Lam, P.Y.S.; Vincent, G.; Bonne, D.; Clark, C.G. Tetrahedron Lett. 2003, 44, 4927. 233 Demir, A.S.; Reis, Ö.; Emrullahoglu, M. J. Org. Chem. 2003, 68, 10130. 234 a) Cooks, R.G.; Kruger, T.L. J. Am. Chem. Soc. 1977, 99, 1279. b) McLuckey, S.A.; Cameron, D.; Cooks, R.G. J. Am. Chem. Soc. 1981, 103, 1313. 235 a) Liou, C.C.; Brodbelt, J.S. J. Am. Chem. Soc. 1992, 114, 6761. b) Wu, H.F.; Brodbelt, J.S. J. Am. Soc. Mass Spectrom. 1993, 4, 718. 236 Sheng, S.Y.; Wong, P.; Shuguang, M.; Cooks, R.G. J. Am. Soc. Mass Spectrom. 1996, 7, 198. 237 Sheng, S.Y.; Chen, G.; Shuguang, M.; Cooks, R.G.; Gozzo, F.C.; Eberlin, M.N. J. Mass Spectrom. 1995, 30, 807. 238 Cerda, B.A.; Wesdemiotis, C. J. Am. Chem. Soc. 1995, 117, 9734. 239 Chen, L.Z.; Miller, J.M. J. Organomet. Chem. 1993, 448, 225. 240 Wong, P.S.H; Shuguang, M.; Feng, W.; Cooks, R.G. J. Organomet. Chem. 1997, 539, 131. 241 Cooks, R.G.; Wong, P.S.H. Acc. Chem. Res. 1998, 31, 379. 242 Erwin, K.M. Chem. Rev. 2001, 101, 391. 243 Clark, H.C.; Manzer, L.E. J. Org. Chem. 1973, 59, 411. 244 de Jonge, C.R.H.I. in Organic Synthesis by Oxidation with Metal Compounds; Mijs, W.J.; de Jonge, C.R.H.I., Eds.; Plenum Press Inc.: New York, 1986, pp 430. 245 Niekerk, J.N. van; Schoening, F.R.L. Acta Cryst. 1953, 6, 227. 246 Turpeinen, U.; Ahlgrén, M.; Hämäläinen, R. Finn. Chem. Lett. 1977, 246. 247 Pfeiffer, P.; Glaser, H. J. Prakt. Chem. 1938, 151, 1938. 248 Handley, D.A.; Hitchcock, P. B.; Lee, T.H.; Leigh, G.J. Inorg. Chim. Acta 2001, 316, 59. 249 a) Becker, M.; Heinemann, F.W.; Knoch, F.; Donaubauer, W.; Liehr, G.; Schindler, S.; Golub, G.; Cohen, H.; Meyerstein, D. Eur. J. Inorg. Chem. 2000, 4, 719. b) Arcus, C.; Fivizzani, K. P.; Pavkovic, S. F. J. Inorg. Nucl. Chem. 1977, 39, 285. 250 Nasakkala, M. Ann. Acad. Sci. Fenn., Ser. A2: Chemica 1977, 181, 72.
105 8 References
251 Mitchell, T.P.; Bernard, W.H.; Wasson, J.R. Acta Crystallogr. B 1970, 26, 2096. 252 Martinho Simõnes, J.A.; Beauchamp, J.L. Chem. Rev. 1990, 90, 629. 253 Ashcroft, S.J.; Mortimer, C.T. J. Chem. Soc. (A) 1967, 930. 254 Cu(II) compounds, especially when acetates are present, tend to form clusters. See ref. 26a, pp. 864. 255 Arpe, H.J.; Hörning, L. Erdöl Kohle Erdgas Petrochem. Brennst. Chem. 1970, 28, 79. 256 a) Weissermel, K.; Arpe, H.J. Industrial Organic Chemistry; Wiley-VCH: Weinheim, 2003. b) Ullmann's Encyclopedia of Industrial Chemistry; Wiley-VCH: Weinheim, 2003, Vol. 25, pp. 589. 257 Hock, H.; Lang, S. Ber. Dtsch. Chem. Ges. B 1944, 77, 257. 258 a) Kaeding, W.W.; Lindblom, O.; Temple, G. (Dow Chemical Co.) US 2727926, 1955. b) Taplin, W.H. (Dow Chemical Co.) US 2954407, 1960. c) Weaver, C.W. (Dow Chemical Co.) US 3235588, 1964. 259 Heidt, P.C.; Schad, R.C. in Acetic Acid and Its Derivatives, Agreda, V.H.; Zoeller, J.R., eds.; Marcel Dekker: New York, 1993, pp.323. 260 Altomare, A.; Cascarano, G.; Giacovazzo, C.; Guagliardi, A.; Burla, M.; Polidori, G.; Camalli, M. J. Appl. Crystallogr. 1994, 27, 435. 261 Sheldrick, G.M. SHELXL-97 Program for the Refinement of Crystal Structures, 1997, University of Göttingen, Germany. 262 Altomare, A.; Carrozzini, B.; Cascarano, G. L.; Giacovazzo, C.; Guagliardi, A.; Moliterni, A. G.; Rizzi, R. A package for crystal structure solution by direct methods and refinement, 1997, Bari, Italy. 263 Derome, A.E. Modern NMR Techniques for Chemistry Research in Organic Chemistry Series, Vol. 6, Baldwin, J.E. ed.; Pergamon Press: Oxford, 1997.
106 8 References
Résumé
Name Gerd Gerdes Date and place of birth July 3; 1973 in Friedrichshafen, D Citizenship German
Education
1980 – 1984 Grundschule Markdorf, D 1984 – 1993 Gymnasium Markdorf, D 1993 – 1994 Military service at the special sports boarding school Warendorf, D 1994 - 1999 Undergraduate studies in chemistry at the University of Heidelberg, D 1998 Visiting student at the Ecole Nationale Supérieure de Chimie de Montpellier 1999 Diploma thesis "Untersuchungen zum Mechanismus der σ-π-σ- Umlagerung an η1-Prop-2-enyl-[4,5-bis(diphenylphosphino)acridin]- palladium(II)-tetrafluoroborat und die erste direkte Synthese eines röntgenfähigenKupfer(II)-Phosphin-Komplexes" with Prof. Günter Helmchen at the university of Heidelberg, D 2000 – 2004 Doctoral thesis "Catalytic C-H Activation of Benzene by Pt(II): A Mechanistic Study" with Prof. Peter Chen at ETH Zürich, CH 2000 – 2002 Kékulé-stipend from the Fonds der Chemischen Industrie, D
107