University of Groningen Gold(I) Catalysis Taschinski, Svenja

Gold(I) Catalysis

Taschinski, Svenja

DOI:

10.33612/diss.126022756

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from

it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date:

2020

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Taschinski, S. (2020). Gold(I) Catalysis: Mechanistic Insights, Reactivity of Intermediates and its

Applications. University of Groningen. https://doi.org/10.33612/diss.126022756

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

Gold(I) Catalysis

Mechanistic Insights,

Reactivity of Intermediates and its Applications

Dissertation

Submitted to the

Combined Faculty of Natural Science and Mathematics

Heidelberg University, Germany

for the degree of

Doctor of Natural Sciences (Dr. rer. nat)

Oral examination:

Friday 5 June 2020

Presented by

Svenja Taschinski

born on 29 February 1988

in Ludwigshafen am Rhein, Germany

This thesis is handed in both to the Heidelberg University and the University of Groningen to achieve a Double Phd Degree.

The title pages of the thesis handed in differ according to the guidelines of each university but the content of both starting from page 5 are identical.

Gold(I) Catalysis

Mechanistic Insights,

Reactivity of Intermediates and its Applications

PhD Thesis

to obtain the degree of PhD at the

University of Groningen

on the authority of the

Rector Magnificus Prof. C. Wijmenga

and in accordance with

the decision by the College of Deans

and

submitted to the

Combined Faculty of Natural Science and Mathematics

Heidelberg University, Germany

for the degree of

Doctor of Natural Sciences (Dr. rer. nat)

Double PhD degree

This thesis will be defended in public on

Friday 5 June 2020 at 14.30 hours

by

Svenja Taschinski

born on 29 February 1988

in Ludwigshafen am Rhein, Germany

Supervisors

Prof. W.R. Browne Prof. A.S.K. Hashmi

Co-supervisor

Dr. J.E.M.N Klein

Assessment Committee

Prof. E. Otten

Prof. M.A. Gruden-Pavlović Prof. M. Tromp

Contents

Contents

5

Chapter 1 Introduction

19

1.1) Catalysis Involving Vinyl Gold(I) Complexes 20

1.2) References 27

Chapter 2 Mechanistic Investigation of the Divergence of a

Light-Mediated Gold-Catalyzed Reaction

29

2.1) Introduction 30

2.2) Results and Discussion 34

2.2.1) Control Experiments 36

2.2.2) Irradiation Experiments 39

2.2.3) Stoichiometric Experiments 40

2.2.4) DFT Calculations 43

2.2.5) Substrate Scope Substituted Benzofurans 46 2.2.6) Substrate Scope Substituted Azobenzofurans 48

2.3) Conclusions 51

2.4) Author Contributions and Acknowledgements 52

2.5) References 52

Chapter 3 Reactivity of Vinyl Gold(I) Complexes

55

3.1) Introduction 56

3.2) Results and Discussion 58

3.3) Conclusions 64

3.4) Author Contributions and Acknowledgements 65

6

Chapter 4 Synthesis of Substituted Benzo[b,b

’]difurans using

Gold(I) Catalysis

67

4.1) Introduction 68

4.2) Results and Discussion 71

4.3) Conclusions 77

4.4) Author Contributions and Acknowledgements 78

4.5) References 78

Chapter 5 Azobenzofurans - Kinetics of the Thermal Relaxation

Step

79

5.1) Introduction 80

5.2) Results and Discussion 84

5.3) Conclusions 94

5.4) Author Contributions and Acknowledgements 95

5.5) References 95

Chapter 6 Experimental Sections and Appendices

97

6.1) Experimental Sections, General Procedures 98 6.1.1) GP A: Sonogashira Coupling of Alkynes 98 6.1.2) GP B: Synthesis of Diazonium Salts 98 6.1.3) GP C: Synthesis of Azobenzofurans 99 6.1.4) GP D: Synthesis of Gold(I) Complexes 99 6.1.5) GP E: Kinetic Experiments using Vinyl Gold(I) Complexes 100 6.1.6) GP F: Synthesis of Diynes via Sonogashira Coupling 100 6.1.7) GP G: Deprotection of Diynes 101 6.1.8) GP H: Gold-Catalyzed Reactions of Diynes 101 6.2) Experimental Section, Chapter 2 102

6.2.1) Synthesis of Substrates 102

6.2.2) Synthesis of Diazonium Salts 106 6.2.3) Optimization of Gold-Catalyzed Reactions 109

6.2.4) Control Experiments 111

6.2.5) Irradiation Experiments 112

6.2.6) Reactions of Vinyl Gold(I) Complexes 114 6.2.7) Scope of Reaction - Substituted Benzofurans 115 6.2.8) Scope of Reaction - Substituted Azobenzofurans 118

7 6.3) Experimental Section, Chapter 3 127

6.3.1) Synthesis of Gold(I) Complexes 127

6.3.2) Kinetic Experiments 133

6.3.3) Gibbs Free Energy 145

6.3.4) Determination of Second Order Rate Constant and Gibbs Free

Energy 148

6.4) Experimental Section, Chapter 4 150

6.4.1) Synthesis of Substrates 150

6.4.2) Gold-Catalyzed Reactions 155

6.5) Experimental Section, Chapter 5 158 6.5.1) UV-VIS Absorption Spectra of Azobenzofurans 158

6.5.2) Kinetic Experiments 160

6.5.3) Eyring Plot in CDCl3 169

6.5.4) Eyring Plot in Toluene-d8 173

6.6) Appendix A, General Remarks 177 6.7) Appendix B, Crystallographic Data 179

6.8) References 193

6.9) Appendix C, Abbreviations 195

9

Abstract

Abstract

Vinyl gold(I) complexes are key intermediates in gold catalysis and so far only a limited number of complexes have been isolated. A few systematic studies on their reactivity towards electrophiles have been conducted, but understanding if and how the choice of catalyst or the reaction conditions influence the formed intermediate is still challenging. Especially how the mechanism can be controlled by the choice of ligand or the electronic nature of the substrate seems to be guided by intuition. Chapter one summarizes briefly the preliminary work on reactivity of gold(I) compounds as intermediates in catalytic cycles and their reactivity towards electrophiles to give a short introduction to the broad field of gold(I) catalysis.

Chapter two is dealing with a mechanistic study on the gold catalyzed reaction of o-alkynylphenols with diazonium salts that can be controlled by light. It could be demonstrated, that the reaction can either undergo C-C-bond formation to form substituted benzofurans with N2-extrusion (blue LED irradiation) or C-N-bond

formation to substituted azobenzofurans with N2-retention (no

irradiation). Stoichiometric experiments with an isolated vinyl gold(I) complex clearly demonstrate, that both reaction types can proceed through the same intermediate but lead to different products simply by irradiation of light. Both reactions can be carried out with the same starting materials under the same reaction conditions and do not require inert, water free conditions or the involvement of an additional photo(redox) catalyst. These results are in contrast to previously reported ones. Irradiation experiments of the gold(I) precursor Ph3PAuCl with base and diazonium salt show, that a

potentially involved gold(III) species is not formed before coordinating to the alkyne, but an involvement after the formation of the vinyl gold(I) species cannot be ruled out. However, the hypothesis of the formation of an electron-donor-acceptor-complex (EDAC) by the vinyl gold(I) complex and the diazonium salt is supported by DFT calculations that show a low lying excitation state for the EDAC on the triplet surface. After optimizing the reaction conditions, the scope for both transformations was explored. The products could be isolated with electronically varied substituents in moderate to high yields and the structure of some azobenzofurans were characterized by single-crystal X-ray crystallography.

In Chapter three, differently substituted vinyl gold(I) complexes with respect to the benzofuran core and the ligand were isolated in good to high yields and the non-substituted Ph3PAu- and

IMesAu-complexes could be characterized by single-crystal X-ray crystallography. The OMe-substituted Ph3PAu-complex was treated

with differently substituted phenols and the reactions were monitored by 1_{H NMR spectroscopy. When correlating the reaction rates with}

Hammett parameters of the phenols, a straight line with a positive slope is formed, demonstrating that the protodeauration is strongly dependent on the acidity of the phenol. When repeating the experiments with p-bromophenol as proton source and varying the substitution pattern of the benzofuran core, it turned out that the protodeauration is accelerated by electron-donating groups in

10

para-position. The reaction rates were correlated against their Hammett parameters for both ligands as a straight line with negative slope, respectively, that indicates the formation of a positive charge during the reaction. Both ligands studied show similar slopes within their error margins which demonstrates the negligible effect of the ligand involved in the protodeauration step in this case. The Gibbs free energy was explored by treating the para-F-substituted Ph3PAu-complex with different concentrations of p-bromophenol.

In Chapter four, substituted diynediols as starting materials for the gold-catalyzed reaction to substituted benzodifurans were synthesized and a possible transfer of the divergent mechanism explored in Chapter two was probed. It was possible to transfer the light mediated gold-catalyzed C-C-bond formation from benzofurans to substituted benzodifurans and the reaction can be performed without additional photo(redox) catalyst and under aerobic conditions, respectively. The F-substituted diynediols were treated with different diazonium salts and the isolation of three different substituted benzodifurans was possible. Adding a photo(redox) catalyst or performing the reaction under inert conditions had no beneficial effect to the reaction. Additionally, the tetrakisfluoro-BDF (benzodifuran) could be characterized by single-crystal X-ray crystallography. The formation of a substituted bisazo-BDF could not be clearly confirmed, but the isolation of a deep orange solid indicates the desired reactivity, which needs to be further explored in future.

In Chapter five, the spectroscopic characteristics of azobenzofurans (synthesized in Chapter two) were analyzed by UV-VIS and 1_{H NMR spectroscopy and these compounds were}

identified as molecular switches. UV-VIS measurements demonstrate that azobenzofurans can be switched by irradiation with blue LED light to a photostationary state (PSS). The photostationary distribution (PSD) for the OMe-substituted azobenzofuran at different temperatures was determined by monitoring the kinetics with 1_{H NMR spectroscopy. Furthermore,}

a systematic study to explore how the substituents might influence the switching ability was conducted by following the kinetics with 1_{H NMR spectroscopy and the reaction rates were}

correlated against their Hammett parameters. It can be clearly demonstrated that the electronic nature of the benzofuran core has a negligible effect on the switching ability. The major impact results in the electronic nature of the azo moiety which can be shown when correlating the reaction rates in a Hammett plot. Electron-donating groups at the azo unit show a straight line with a negative slope and electron-withdrawing groups a straight line with a positive slope to form a so called V-shaped Hammett plot. This V-shape indicates a change in switching mechanism which will be explored in future by DFT calculations. Additionally, the enthalpy, entropy and Gibbs free energy of activation were determined.

11

Samenvatting

Samenvatting

Vinyl goud(I)-complexen zijn essentiële intermediairen in goud-katalyse en er zijn tot nu toe slechts een beperkt aantal complexen geïsoleerd. Enkele systematische studies omtrent hun reactiviteit ten opzichte van elektrofielen zijn uitgevoerd, maar het is nog steeds een uitdaging om te begrijpen hoe de keuze van de katalysator of de reactieomstandigheden het gevormde intermediairen beïnvloeden. Met name het antwoord op de vraag hoe deze reactie kan worden gestuurd door aanpassingen in de structuur van de katalysator of substraat is vooralsnog grotendeels onbekend. In hoofdstuk één wordt kort ingegaan op het voorbereidende werk over de reactiviteit van goud(I)-verbindingen als tussenproducten in katalytische cycli en hun reactiviteit ten opzichte van elektrofielen om een korte inleiding te geven op het veelzijdige veld van goud(I)-katalyse.

Hoofdstuk twee beschrijft mechanistisch onderzoek naar de goud-gekatalyseerde reactie van ortho-alkynylfenolen met diazoniumzouten, die door licht kan worden beinvloed. Er kon worden aangetoond dat de reactie ofwel, C-C-bindingsvorming kan ondergaan om gesubstitueerde benzofuranen te vormen door middel van N2-extrusie (blauwe LED-bestraling), of

C-N-bindings-vorming ondergaat aan gesubstitueerde azobenzofuranen met N2-retentie (geen bestraling). Stoichiometrische experimenten met

een geïsoleerd vinyl goud(I)-complex tonen duidelijk aan dat beide reactietypes door hetzelfde tussenproduct kunnen gaan, maar tot verschillende producten leiden, simpelweg door de bestraling met licht. Beide reacties kunnen worden uitgevoerd met dezelfde startmaterialen onder dezelfde reactieomstandigheden en vereisen geen inerte, watervrije omstandigheden of de betrokkenheid van een extra foto(redox)-katalysator. Deze bevindingen zijn in tegenstelling tot eerder gerapporteerde resultaten. Bestralings-experimenten van de goud(I)-precursor Ph3PAuCl met base en

diazoniumzout tonen aan dat een potentieel betrokken goud(III)-verbinding niet wordt gevormd voordat deze wordt gecoördineerd met het alkyn, maar betrokkenheid na de vorming van de vinyl goud(I)-verbinding kan niet uitgesloten worden. De hypothese van de vorming van een elektron-donor-acceptor-complex (EDAC) door het vinyl goud(I)-elektron-donor-acceptor-complex en het diazoniumzout wordt echter ondersteund door DFT-berekeningen die een laag-liggende excitatietoestand voor de EDAC op het triplet-oppervlak laten zien. Na het optimaliseren van de reactieomstandigheden werden beide reacties onderzocht, de

producten konden worden geïsoleerd met verschillende substituenten in matige tot hoge opbrengsten en de structuur van sommige azobenzofuranen werd opgehelderd door eenkristaldiffractie.

In hoofdstuk drie werden verschillende gesubstitueerde vinyl goud(I)-complexen, met betrekking tot de benzofuran-kern en het ligand, met goede tot hoge opbrengsten geïsoleerd en konden de niet-gesubstitueerde Ph3PAu- en IMesAu-complexen worden gekarakteriseerd

door eenkristaldiffractie. Het OMe-gesubstitueerde Ph3PAu-complex werd gereageerd met

12

1_{H NMR-spectroscopie. Bij het correleren van de reactiesnelheden}

met de Hammett-parameters van de fenolen, wordt een rechte lijn met een positieve helling gevonden, die aantoont dat de protodeauratie sterk afhankelijk is van de zuurgraad van het fenol. Bij het herhalen van de experimenten met p-broomfenol als protonenbron en het variëren van het substitutiepatroon op de benzofuran-kern, bleek dat de protodeauratie wordt versneld door elektron-donerende groepen in parapositie. De reactiesnelheden waren gecorreleerd met hun Hammett-parameters voor beide liganden door een rechte lijn met een negatieve helling, die de vorming van een positieve lading tijdens de reactie aangeeft. Beide bestudeerde liganden vertonen vergelijkbare hellingen binnen hun foutmarges, hetgeen het verwaarloosbare effect aantoont van het ligand dat betrokken is bij de protodeauratie-stap. De Gibbs vrije energie van activering werd bepaald door het para F-gesubstitueerde Ph3PAu-complex te reageren met verschillende concentraties

p-broomfenol.

In hoofdstuk vier werden gesubstitueerde diyn-diolen als uitgangsmaterialen gebruikt voor de goud-gekatalyseerde vorming van gesubstitueerde benzodifuranen, en de overeenkomsten met het in hoofdstuk twee onderzochte mechanisme werden bekeken. Het was mogelijk om de door licht gemedieerde goud-gekatalyseerde C-C-bindingsvorming over te brengen van benzofuranen naar gesubstitueerde benzodifuranen, en de reactie kon worden uitgevoerd zonder extra foto(redox)-katalysator onder aerobe omstandigheden. De F-gesubstitueerde diyn-diolen werden behandeld met verschillende diazoniumzouten en de isolatie van drie verschillende gesubstitueerde benzodifuranen was mogelijk. Het toevoegen van een foto(redox)-katalysator, of het uitvoeren van de reactie onder inerte omstandigheden had geen gunstig effect op de reactie. Bovendien kon het product tetrakisfluoro-BDF (benzodifuran) worden gekarakteriseerd door eenkristaldiffractie. De vorming van een gesubstitueerde bisazo-BDF kon niet duidelijk worden bevestigd, maar de isolatie van een diep oranje vaste stof

geeft de gewenste reactiviteit aan, die in de toekomst verder moet worden onderzocht.

In Hoofdstuk vijf werden de spectroscopische eigenschappen van azobenzofuranen (gesynthetiseerd in Hoofdstuk twee) geanalyseerd met UV-VIS en 1_{H NMR-spectroscopie, deze}

verbindingen werden geïdentificeerd als moleculaire schakelaars. UV-VIS metingen tonen aan dat azobenzofuranen kunnen worden geschakeld door bestraling met blauw LED-licht naar een foto-stationaire toestand (PSS). De foto-stationaire verdeling (PSD) voor het OMe-gesubstitueerde azobenzofuran werd bepaald bij verschillende temperaturen door de kinetiek te volgen met 1_{H NMR-spectroscopie. Verder werd een systematische studie uitgevoerd om te}

onderzoeken hoe de substituenten het schakelen beïnvloeden door de kinetiek te volgen met

13

Samenvatting

Hammett-parameters. Het kon worden aangetoond dat de elektronische eigenschappen van de benzofuran-kern een verwaarloosbaar effect heeft op de schakel-eigenschappen. De belangrijkste impact is afkomstig van de elektronische aard van het azo-substituent, wat kan worden aangetoond bij het correleren van de reactiesnelheden in een Hammett-plot. Elektronen-donerende groepen op de azo-substituent vertonen een rechte lijn met een negatieve helling en elektronen-zuigende groepen een rechte lijn met een positieve helling om een zogenaamd V-vormig Hammett-plot te vormen. Deze V-vorm geeft een verandering in schakelmechanisme aan, die in de toekomst zal worden onderzocht met behulp van DFT-berekeningen. Bovendien werden de enthalpie, entropie en Gibbs vrije energie van activering bepaald.

15

Zusammenfassung

Zusammenfassung

Vinylgold(I)-Komplexe sind Schlüsselintermediate in der Goldkatalyse und bislang konnte nur eine limitierte Anzahl an Komplexen isoliert werden. Wenige systematische Studien in Hinblick auf deren Reaktivität gegenüber Elektrophilen wurden durchgeführt, dennoch ist es immernoch herausfordernd zu verstehen ob und wie die Wahl des Katalysators die gebildete Zwischenstufe beeinflusst. Speziell wie ein Mechanismus durch die Wahl des Liganden oder der elektronischen Natur des Substrates gesteuert werden kann wirken intuitiv. Das erste Kapitel fasst vorangehende Arbeiten an Vinylgold(I)-Verbindungen als Intermediate in Katalysezyklen und deren Reaktivität gegenüber Elektrophilen als kurze Einleitung in das mannigfaltige Feld der Gold(I)-Katalyse zusammen.

Das zweite Kapitel handelt von einer mechanistischen Studie der gold-katalysierten Reaktion von o-Alkinylphenolen mit Diazoniumsalzen, gesteuert durch Licht. Es wurde gezeigt, dass die Reaktion entweder unter C-C-Bindungsknüpfung zur Ausbildung

von substituierten Benzofuranen unter Abspaltung von N2

(Bestrahlung mit blauem LED-Licht) oder C-N-Bindungen unter Erhalt der N2-Gruppe (keine Bestrahlung) führt. Stöchiometrische

Experimente mit einem isolierten Vinylgold(I)-Komplex demonstrieren klar, dass beide Reaktionstypen über das gleiche Intermediat verlaufen, aber durch Bestrahlung zu unterschiedlichen Produkten führen können. Beide Reaktionen können ausgehend von den gleichen Edukten unter den gleichen Reaktionsbedingungen durchgeführt werden und benötigen keine inerten, wasserfreien Bedingungen oder die Beteiligung eines Photo(Redox)-Katalysators. Diese Ergebnisse sind konträr zu vorherigen postulierten. Bestrahlungsexperimente des Gold(I)-Präkatalysators Ph3PAuCl mit

Base und Diazoniumsalz zeigen, dass eine potentiell involvierte Gold(III)-Spezies nicht vor der Koordination an das Alkin gebildet wird, aber eine Beteiligung nach Bildung der Vinylgold(I)-Spezies nicht ausgeschlossen werden kann. Die Hypothese eines gebildeten Elektron-Donor-Akzeptor-Komplexes (EDAC) zwischen dem Vinylgold(I)-Komplex und dem Diazoniumsalz wird durch DFT-Rechnungen gestützt, die zeigen, dass es einen niedrig-liegenden angeregten Zustand auf der Triplettoberfläche des EDAC gibt. Nach Optimierung der Reaktionsbedingungen wurden beide Reaktionen kurz auf ihren Reaktionsumfang geprüft, die entstehenden Produkte mit elektronisch variierenden Substituenten in mäßiger bis hoher Ausbeute isoliert und die Struktur einiger Azobenzofurane durch Röntgeneinkristallstrukturanalyse charakterisiert.

16

In Kapitel drei wurden verschieden substituierte Vinylgold(I)-Komplexe in Bezug auf das Benzofuran-Gerüst und den Liganden in guten bis hohen Ausbeuten isoliert und der unsubstituierte Ph3PAu- und IMesAu-Komplex wurden durch

Röntgeneinkristallstrukturanalyse charakterisiert. Der OMe-substituierte Ph3PAu-Komplex wurde mit verschieden

substituierten Phenolen versetzt und die Reaktionen mit 1_{H NMR}

Spektroskopie überwacht. Werden Reaktionsraten mit den Hammett-Parametern der Phenole korreliert, wird eine Gerade mit positiver Steigung erhalten, die die starke Abhängigkeit der Protodesaurierung von der Azidität des Phenols zeigt. Die Wiederholung der Experimente mit p-Bromphenol als Protonenquelle und variieren der Substituenten am Benzofuran-Gerüst zeigt, dass die Protodesaurierung durch elektronenschiebende Gruppen in para-Position beschleunigt wird. Die Reaktionsraten für beide Liganden wurden als Gerade mit negativer Steigung gegen ihre Hammett-Parameter korreliert, was die Bildung einer positiven Ladung während der Reaktion indiziert. Beide getesteten Liganden zeigen innerhalb ihrer Fehlergrenzen ähnliche Steigungen, das in diesem Fall den vernachlässigbaren Effekt des Liganden, der in der Protodesaurierung involviert ist, zeigt. Die Gibbs-Energie wurde durch Versetzen des para-F-substituierten Ph3PAu-Komplexes mit verschiedenen Konzentrationen von

p-Bromphenol ermittelt.

In Kapitel vier wurden substituierte Diindiole als Edukte zur goldkatalysierten Darstellung von substituierten Benzodifuranen (BDF) hergestellt und ein möglicher Transfer des divergenten Mechanismus, untersucht in Kapitel zwei, getestet. Es war möglich die lichtvermittelte C-C-Bindungsknüpfung von Benzofuranen auf Benzodifurane zu übertragen und die Reaktion kann ebenso ohne zusätzlichen Photo(Redox)-Katalysator und unter aeroben Bedingungen durchgeführt werden. Die F-substituierten Diindiole wurden mit verschiedenen Diazoniumsalzen versetzt und die Isolation von drei verschieden substituierten Benzodifuranen war möglich. Der Zusatz eines Photo(Redox)-Katalysators oder die Reaktionsführung unter inerten Bedingungen hatte keinen vorteilhaften Effekt auf die Reaktion. Zusätzlich wurde das vierfach fluor-substituierte BDF (Benzodifuran) durch Röntgeneinkristallstrukturanalyse charakterisiert. Die Bildung eines doppel-azo BDFs wurde nicht eindeutig bestätigt, wenn auch die Isolation eines dunkelorange-farbenen Feststoffes auf die gewünschte Reaktivität hinweist, was ferner untersucht wird.

17

Zusammenfassung

In Kapitel fünf wurden die spektroskopischen Eigenschaften der Azobenzofurane (synthetisiert in Kapitel zwei) durch UV/VIS-Spektroskopie und 1_{H NMR-Spektroskopie analysiert}

und die Verbindungen als molekulare Schalter identifiziert. UV/VIS-Messungen demonstrieren, dass Azobenzofurane durch Bestrahlung mit blauem LED-Licht in einen photostationären Zustand (PSS) versetzt werden können. Die Verteilung des photostationären Zustandes (PSD) bei verschiedenen Temperaturen wurde für das OMe-substituierte Azobenzofuran über das 1_{H NMR-spektroskopische Verfolgen der Reaktionskinetik ermittelt. Weiterhin wurde}

eine systematische Studie durch Verfolgen der Reaktionskinetiken mittels

1_{H NMR-Spektroskopie durchgeführt, um zu ermitteln wie die Substituenten möglicherweise}

die Schaltfähigkeit beeinflussen und die Reaktionsraten wurden gegen die Hammett-Parameter korreliert. Es wurde klar demonstriert, dass die elektronische Natur des Benzofuran-Gerüsts einen vernachlässigbaren Effekt auf die Schaltfähigkeit hat. Der größte Einfluss resultiert in der elektronischen Natur der Azo-Einheit, was durch Korrelation der Reaktionsraten in einem Hammett-Plot gezeigt wurde. Elektronenschiebende Gruppen an der Azo-Einheit zeigen eine Gerade mit negativer Steigung und elektronenziehende Gruppen eine Gerade mit positiver Steigung und bilden einen sogenannten V-förmigen Hammett-Plot. Diese V-Form indiziert eine Änderung im Schaltmechanismus, der in Zukunft durch DFT-Rechnungen untersucht wird. Zusätzlich wurden die Enthalpie, Entropie und Gibbs-Energie der Aktivierung bestimmt.

Chapter 1

Introduction

Abstract: In this chapter, a general introduction to gold catalysis, the synthesis of catalysts

and vinyl gold(I) complexes as key intermediates are provided.

While many reactions involve intermediates that can hardly be synthesized, vinyl gold(I) complexes have been discovered as isolatable complexes a few years ago. Although some of them are air-stable and easy to isolate, only a few studies on their reactivity are published. Some recent examples are discussed as probes for the elucidation of possible pathways to deaurate vinyl gold(I) complexes with electrophiles, an elementary step that can proceed intra- or intermolecularly.

20

1.1)

Catalysis Involving Vinyl Gold(I) Complexes

Gold catalysis developed to a prominent research topic over the last decades.[1] _{One of the}

key reactions in the field of gold-catalyzed chemistry has been reported 1976 by Thomas and co-workers.[2] _{Without knowing that this might be one of the first published gold-catalyzed}

reactions that opens up a new research area, the authors reported a Markovnikov addition that transforms alkynes 1 into related ketones 2 as major products catalyzed by tetrachloroauric acid.[2] _{The mechanism was proposed to be a gold(III) oxidation similar to the}

known oxidation by mercury(II)[2] _{although the reaction reached almost six turnovers.}[1b]

Scheme 1: First experiments reported for the gold-catalyzed reaction of alkynes 1 to ketones 2, 3,

methyl vinyl ethers 4 and vinyl chlorides 5 by Thomas and co-workers.[2]

Despite some work reporting gold(III) in catalysis, for example reports by Hutchings (hydrochlorination of alkynes)[3] _{and Utimoto (hydration of propargyl ethers),}[4] _{gold seemed to}

be “catalytically dead” and known as a non-reactive noble metal.[5]

At the same time, the structural characterization of a L-Au(I) complex, which later on became one of the most used precursors for cationic complexes and catalytic reactions,[1b] _{has been}

reported by Soboroff and co-workers in 1976.[6] _{The group synthesized Ph}

3PAuCl 7Ph3P,Cl by

treating Ph3P 6 with tetrachloroauric acid (see Scheme 2) and characterized it by

single-crystal X-ray structure analysis.

Scheme 2: Left: Synthesis of Ph3PAuCl 7Ph3P,Cl reported by Soboroff and co-workers.[7] Right:

Single-crystal X-ray structure of Ph3PAuCl 7Ph3P,Cl (hydrogen atoms omitted for clarity; picture reprinted with permission from A. Fürstner, P. W. Davies, Angew. Chem. 2007, 119, 3478-3519; Copyright © 2007 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim).

The catalytic relevance of such cationic phosphine complexes has been demonstrated in 1986 by Ito and co-workers[8] _{and by Teles and co-workers in 1998.}[9] _{For the first time, gold(I)}

complexes were used for the conversion of alkynes,[9] _{namely a gold-catalyzed addition of}

methanol to propyne 8Me (see Scheme 3). With turnover frequencies up to 1500 depending

on the electronic nature of the ligand, this is the first study on ligand-depending reactivity reported.[9] _{The group explored that electron-poor ligands increase the activity of}

21

Introduction

Scheme 3: Selected turnover frequencies for the gold-catalyzed addition of methanol to

propyne 8Me reported by Teles and co-workers.[9]

Only two years later, the power of gold(I) in catalytic reactions has been demonstrated by Hashmi and co-workers.[10] _{A comprehensive study showed how useful gold catalysis can be}

compared with known reactivities of other transition metal catalysts, such as Ag(I) and Pd(II).[10] _{When treating the same starting materials with gold(III) complexes, reaction times}

could be decreased by the choice of catalyst from a week (AgNO3), over one hour

(PdCl2(MeCN)2) to one minute (AuCl3). Despite the shortening in reaction times, gold was able

to catalyze the transformation of propargylketones, such as 10, to furans 11 at room temperature (see Scheme 4).[10] _{This reaction could not be catalyzed by Ag(I)}[10] _{and required}

temperatures of 70-100°C when using a Pd(II)-catalyst.[11]

Scheme 4: Gold(III)-catalyzed C-O-bond formation reported by Hashmi and co-workers.[10]

Gold(III) catalysis was subsequently studied in detail due to its “special” reactivity allowing for mild conditions and the reactions reported earlier were re-explored. A number of different catalytic reactions were reported and the field of gold catalysis grew enormously[1] _{as well as}

the design of air-stable (pre-)catalysts,[9, 12] _{with phosphine-gold(I) or NHC-gold(I) catalysts}

being used most commonly.

The first isolated NHC-gold(I) complexes were reported in 1989 by Bovio and co-workers by transmetallating lithium imidazoles with Ph3PAuCl / Me2SAuCl at low temperatures

(-40°C).[12c] _{Their usage in catalytic reactions was first reported in 1998 by Teles and}

co-workers and 2003 by Herrmann and co-workers.[9, 13] _{A series of air-stable NHC-gold(I)}

complexes was introduced by Nolan and co-workers,[12a, 12b] _{who isolated and characterized}

Au-chlorides and their cationic analogues by transmetallation of Ag(I) complexes or through direct coupling of imidazol-2-ylidene carbenes with Me2SAuCl at room temperature (see Figure 1).

22

Figure 1: Structures (left) and single-crystal X-ray structures (right) of IPrAuCl 12IPr and

IMesAuCl 13IMes (hydrogen atoms omitted for clarity; picture reprinted with permission from P. de Frémont, N. M. Scott, E. D. Stevens, S. P. Nolan, Organometallics 2005, 24, 2411-2418; Copyright © 2005, American Chemical Society).[12b]

These types of complexes have a linear (180°, 13IMes) and almost linear (177°, 12IPr) geometry

at the C-Au-Cl angle and Au-C(NHC) distances of 13IMes = 1.998(5) Å and

12IPr = 1.979(3) Å.[12b] The values are comparable with the bond angle reported for

Ph3PAuCl 7Ph3P,Cl (P-Au-Cl = 179.68(8) Å), but the Au-P distance is longer (2.235(3) Å).[6]

Both complex chlorides can easily be transformed into their cationic analogues by ligand exchange with silver(I) salts.[9, 12a, 12b] _{With the possibility to generate the active, cationic}

complexes either in situ or isolate them first and subsequently add them to the reaction mixture, gold-catalyzed reactions are easy to apply and the influence of ligand or counter anion as well as solvent were studied systematically over time.[1a-d, 7, 14] _{A few theoretical}[14f, 15]

and experimental studies[14a, 14e, 14f, 16] _{explore how isolated}[16] _{or in situ-generated}[17] _gold(I)

complexes can be influenced by the electronic nature of the substituents, the choice of ligand or reaction conditions.

To date, gold(I) and gold(III) catalytic cycles to functionalize C-C-multiple bonds can be generally described by the coordination of the prior or in situ-activated catalyst to the -bond of the molecule as the start of the catalytic cycle. The Lewis-acidic gold catalyst 14 activates the multiple bond towards an inter- / intramolecular nucleophilic attack to form gold(I)/(III) intermediate 16.[18] _{Electrophilic inter- / intramolecular deauration closes the catalytic cycle}

and releases the active catalyst.[19]

Scheme 5: Prototypical catalytic cycle of gold-catalyzed reactions involving gold(I/III)

23

Introduction

To support the catalytic involvement of gold(I) in triple-bond conversions and to gain deeper understanding of its mechanism, well-characterized, air-stable vinyl gold(I) complexes were isolated by several groups[14a, 14f, 20] _{(see Scheme 6) with a coordination geometry that}

remains almost linear and Au-C(NHC) / Au-P-distances[20e-i] _{comparable to the related}

gold(I) chlorides.[6, 12b]

Scheme 6: Selected examples for isolated vinyl gold(I) complexes 18-21 and their single-crystal

X-ray structures (hydrogens are omitted for clarity, except top right; pictures reprinted with permission from J. A. Akana, K. X. Bhattacharyya, P. Müller, J. P. Sadighi, J. Am. Chem. Soc. 2007, 129, 7736-7737. Copyright © 2007, American Chemical Society; L.-P. Liu, B. Xu, M. S. Mashuta, G. B. Hammond, J. Am. Chem. Soc. 2008, 130, 17642-17643. Copyright © 2008, American Chemical Society; A. S. K. Hashmi, A. M. Schuster, F. Rominger, Angew. Chem. 2009, 121, 8396-8398; Angew. Chem. Int. Ed. 2009, 48, 8247-8249. Copyright © 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim and D. Weber, M. A. Tarselli, M. R. Gagné, Angew. Chem. Int. Ed. 2009, 48, 5733-5736. Copyright © 2009 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim).[20e-i]

The bigger the field grew, the more attention gold(I)-catalyzed reactions involving electrophilic deauration for C-C-bond or C-X-bond formation received. The majority of gold-catalyzed reactions are reported with a proton as electrophile that terminates the catalytic cycle.[14e]

Some approaches reported catalytic reactions that trap intermediates with different groups, such as carbon-,[20a, 21] _silicon-,[22] _{trifluoromethyl-,}[23] _halogen-,[20i, 24] _tin-,[25] _{or sulfonyl-}[26]

electrophiles, as well as some stoichiometric reactions involving gold complexes as intermediates, such as gold-alkynyl,[27] _-vinyl,[24c] _-allenyl,[20e] _-aryl[24c, 28] _{and -alkyl}[29]

complexes, respectively. Focussing on vinyl gold intermediates, there are three possible ways to deaurate a vinyl gold complex by an electrophile: (i) intermolecular with oxidative addition and reductive elimination; (ii) intramolecular by sigmatropic rearrangement and (iii) intermolecular by external substitution (see Scheme 7). The strategies will be elucidated by selected examples related to the benzofuran core that is key in this thesis.

24

Scheme 7: Possible strategies for electrophilic deauration of vinyl gold intermediates.

In the intermolecular strategy with an external electrophile and a metal redox-activation (see

Scheme 7, left), the electrophile is oxidatively added to the (pre)catalyst or intermediate and

reductive elimination results in C-R-bond formation. With this method, alkynylative arylation with iodoalkynes and [Ir(dF(CF3)ppy)2(dtbbpy)]PF6 as photo(redox) catalyst to substituted

benzofurans 26 was recently reported by Fensterbank and co-workers (see Scheme 8).[20a]

25

Introduction

The group proposes the formation of a vinyl gold(I) intermediate, that is oxidized by the Ir co-catalyst to a gold(III) intermediate and subsequent reductive elimination closes the catalytic cycle. Here, it shall be noted that stoichiometric attempts without the Ir co-catalyst gave the desired product in low to moderate yields and that the co-catalyst might only promote the excited state of the vinyl gold(I) intermediate and enhances the formation of 27.[20a]

As another selected example representing this strategy, the C-F-bond formation through a fluorination / hydration mechanism with selectfluor as oxidant to -fluorobenzofuranones 29 was reported by Shi and co-workers.[30] _{The vinyl gold(I) complex is oxidized by selectfluor to}

a vinyl gold(III) intermediate, that releases the fluorinated product 29 under reductive elimination and subsequent hydration (see Scheme 9).[30]

Scheme 9: C-F-bond formation through a gold(I/III) fluorination / hydration mechanism.[30]

The intramolecular sigmatropic rearrangement (see Scheme 7, middle) proceeds through a cationic vinyl gold(I) intermediate C, that is subsequently deaurated by the shifting group to form the substituted product 24. A structural analogue example to benzofurans for sigmatropic rearrangement was reported by Nakamura and co-workers, that used gold(I) catalysis to synthesize substituted benzothiophenes 31 starting form o-alkynylsulfides 30, respectively (see Scheme 10).[21c, 22b]

Scheme 10: Gold(I)-catalyzed sigmatropic rearrangement to substituted benzothiophenes 31 by

26

For the last strategy shown in Scheme 7 (right), an external electrophile deaurates the priorly formed vinyl gold(I) intermediate D to substituted benzofuran 24 and no change in oxidation state at the gold center is taking place. A selected example was given by Ackermann[31] _and

Döpp[32] _{in our group. The formation of substituted azobenzofurans 32 by external electrophilic}

deauration of vinyl gold(I) complex 46OMe,IPr with diazonium salt 34OMe was demonstrated (see

Scheme 11).

Scheme 11: Stoichiometric experiments for the formation of azobenzofuran 32OMe,OMe.[31]

The following thesis will give a deeper understanding in the deauration of vinyl gold(I) intermediates with external electrophiles and how the substitution pattern of both as well as the reaction conditions influence the reactivity.

27

References

1.2)

References

[1] a) A. Arcadi, Chem. Rev. 2008, 108, 3266-3325; b) A. S. K. Hashmi, Chem. Rev. 2007, 107, 3180-3211; c) A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118, 8064-8105; d) A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. Int. Ed. 2006, 45, 7896-7936; e) A. S. K. Hashmi, Angew. Chem. Int. Ed. 2005, 44, 6990-6993; f) A. S. K. Hashmi, Angew. Chem. 2005, 117, 7150-7154; g) A. M. Echavarren, C. Nevado, Chem. Soc. Rev. 2004, 33, 431-436; h) A. S. K. Hashmi, Gold Bulletin 2003, 36, 3-9; i) G. Dyker, Angew. Chem. Int. Ed. 2000, 39, 4237-4239; j) G. Dyker, Angew. Chem. 2000, 112, 4407-4409.

[2] R. O. C. Norman, W. J. E. Parr, C. B. Thomas, J. Chem. Soc., Perkin Trans. 1 1976, 1983-1987.

[3] G. J. Hutchings, J. Catal. 1985, 96, 292-295.

[4] a) Y. Fukuda, K. Utimoto, J. Org. Chem. 1991, 56, 3729-3731; b) Y. Fukuda, K. Utimoto, Synthesis 1991, 1991, 975-978; c) F. Yukitoshi, U. Kiitiro, Bull. Chem. Soc. Jpn. 1991, 64, 2013-2015.

[5] H. Schmidbaur, Naturwiss. Rundsch. 1995, 48, 443.

[6] N. C. Baenziger, W. E. Bennett, D. M. Soborofe, Acta Cryst. Section B 1976, 32, 962-963. [7] a) A. Fürstner, P. W. Davies, Angew. Chem. 2007, 119, 3478-3519; b) A. Fürstner, P. W.

Davies, Angew. Chem. Int. Ed. 2007, 46, 3410-3449.

[8] Y. Ito, M. Sawamura, T. Hayashi, J. Am. Chem. Soc. 1986, 108, 6405-6406.

[9] a) J. H. Teles, S. Brode, M. Chabanas, Angew. Chem. 1998, 110, 1475-1478; b) J. H. Teles, S. Brode, M. Chabanas, Angew. Chem. Int. Ed. 1998, 37, 1415-1418; c) J. H. S. Teles, M. (BASF AG), WO-A19721648,1997, Chem. Abstr. 1997, 127, 121499.

[10] a) A. S. K. Hashmi, L. Schwarz, J.-H. Choi, T. M. Frost, Angew. Chem. Int. Ed. 2000, 39, 2285-2288; b) A. S. K. Hashmi, L. Schwarz, J.-H. Choi, T. M. Frost, Angew. Chem. 2000, 112, 2382-2385.

[11] H. Sheng, S. Lin, Y. Huang, Synthesis 1987, 1987, 1022-1023.

[12] a) P. de Frémont, E. D. Stevens, M. R. Fructos, M. Mar Díaz-Requejo, P. J. Pérez, S. P. Nolan, Chem. Commun. 2006, 2045-2047; b) P. de Frémont, N. M. Scott, E. D. Stevens, S. P. Nolan, Organometallics 2005, 24, 2411-2418; c) F. Bonati, A. Burini, B. R. Pietroni, B. Bovio, J. Organomet. Chem. 1989, 375, 147-160.

[13] S. K. Schneider, W. A. Herrmann, E. Herdtweck, Z. anorg. allg. Chem. 2003, 629, 2363-2370. [14] a) Z. Lu, G. B. Hammond, B. Xu, Acc. Chem. Res. 2019, 52, 1275-1288; b) J. Schießl, P. M. Stein, J. Stirn, K. Emler, M. Rudolph, F. Rominger, A. S. K. Hashmi, Adv. Synth. Catal. 2019, 361, 725-738; c) J. Schießl, J. Schulmeister, A. Doppiu, E. Wörner, M. Rudolph, R. Karch, A. S. K. Hashmi, Adv. Synth. Catal. 2018, 360, 3949-3959; d) J. Schießl, J. Schulmeister, A. Doppiu, E. Wörner, M. Rudolph, R. Karch, A. S. K. Hashmi, Adv. Synth. Catal. 2018, 360, 2493-2502; e) W. Wang, G. B. Hammond, B. Xu, J. Am. Chem. Soc. 2012, 134, 5697-5705; f) R. Döpp, C. Lothschütz, T. Wurm, M. Pernpointner, S. Keller, F. Rominger, A. S. K. Hashmi, Organometallics 2011, 30, 5894-5903; g) M. Rudolph, A. S. K. Hashmi, Chem. Commun. 2011, 47, 6536-6544; h) A. S. K. Hashmi, M. Buehrle, Aldrichim. Acta 2010, 43, 27-33; i) Z. Li, C. Brouwer, C. He, Chem. Rev. 2008, 108, 3239-3265.

[15] a) L. Nunes dos Santos Comprido, J. E. M. N. Klein, G. Knizia, J. Kästner, A. S. K. Hashmi, Chem. Eur. J. 2017, 23, 10901-10905; b) C. A. Gaggioli, G. Ciancaleoni, D. Zuccaccia, G. Bistoni, L. Belpassi, F. Tarantelli, P. Belanzoni, Organometallics 2016, 35, 2275-2285; c) R. BabaAhmadi, P. Ghanbari, N. A. Rajabi, A. S. K. Hashmi, B. F. Yates, A. Ariafard, Organometallics 2015, 34, 3186-3195.

[16] K. E. Roth, S. A. Blum, Organometallics 2010, 29, 1712-1716. [17] A. Zhdanko, M. E. Maier, Chem. Eur. J. 2014, 20, 1918-1930.

[18] M. Pernpointner, A. S. K. Hashmi, J. Chem. Theory Comput. 2009, 5, 2717-2725.

[19] a) A. S. K. Hashmi, Angew. Chem. Int. Ed. 2010, 49, 5232-5241; b) A. S. K. Hashmi, Angew. Chem. 2010, 5360 – 5369.

[20] a) Z. Xia, V. Corcé, F. Zhao, C. Przybylski, A. Espagne, L. Jullien, T. Le Saux, Y. Gimbert, H. Dossmann, V. Mouriès-Mansuy, C. Ollivier, L. Fensterbank, Nat. Chem. 2019; b) L. Biasiolo, L. Belpassi, C. A. Gaggioli, A. Macchioni, F. Tarantelli, G. Ciancaleoni, D. Zuccaccia, Organometallics 2016, 35, 595-604; c) A. S. K. Hashmi, T. D. Ramamurthi, F. Rominger, Adv.

28

Synth. Catal. 2010, 352, 971-975; d) Y. Chen, D. Wang, J. L. Petersen, N. G. Akhmedov, X. Shi, Chem. Commun. 2010, 46, 6147-6149; e) D. Weber, M. A. Tarselli, M. R. Gagné, Angew. Chem. Int. Ed. 2009, 48, 5733-5736; f) A. S. K. Hashmi, A. M. Schuster, F. Rominger, Angew. Chem. 2009, 121, 8396-8398; g) A. S. K. Hashmi, A. M. Schuster, F. Rominger, Angew. Chem. Int. Ed. 2009, 48, 8247-8249; h) L.-P. Liu, B. Xu, M. S. Mashuta, G. B. Hammond, J. Am. Chem. Soc. 2008, 130, 17642-17643; i) J. A. Akana, K. X. Bhattacharyya, P. Müller, J. P. Sadighi, J. Am. Chem. Soc. 2007, 129, 7736-7737.

[21] a) M. Ackermann, J. Bucher, M. Rappold, K. Graf, F. Rominger, A. S. K. Hashmi, Chem. Asian J. 2013, 8, 1786-1794; b) P. Dubé, F. D. Toste, J. Am. Chem. Soc. 2006, 128, 12062; c) I. Nakamura, T. Sato, Y. Yamamoto, Angew. Chem. Int. Ed. 2006, 45, 4473-4475; d) L. Zhang, J. Am. Chem. Soc. 2005, 127, 16804-16805.

[22] a) P. McGee, G. Bellavance, I. Korobkov, A. Tarasewicz, L. Barriault, Chem. Eur. J. 2015, 21, 9662-9665; b) I. Nakamura, T. Sato, M. Terada, Y. Yamamoto, Org. Lett. 2007, 9, 4081-4083. [23] A. Boreux, G. H. Lonca, O. Riant, F. Gagosz, Org. Lett. 2016, 18, 5162-5165.

[24] a) D. H. Miles, M. Veguillas, F. D. Toste, Chem. Sci. 2013, 4, 3427-3431; b) A. S. K. Hashmi, T. D. Ramamurthi, M. H. Todd, A. S.-K. Tsang, K. Graf, Aust. J. Chem. 2010, 63, 1619-1626; c) A. Stephen K. Hashmi, T. Dondeti Ramamurthi, F. Rominger, J. Organomet. Chem. 2009, 694, 592-597.

[25] Y. Shi, S. D. Ramgren, S. A. Blum, Organometallics 2009, 28, 1275-1277.

[26] I. Nakamura, U. Yamagishi, D. Song, S. Konta, Y. Yamamoto, Angew. Chem. Int. Ed. 2007, 46, 2284-2287.

[27] a) R. Cai, M. Lu, E. Y. Aguilera, Y. Xi, N. G. Akhmedov, J. L. Petersen, H. Chen, X. Shi, Angew. Chem. Int. Ed. 2015, 54, 8772-8776; b) A. S. K. Hashmi, I. Braun, M. Rudolph, F. Rominger, Organometallics 2012, 31, 644-661; c) K. J. Kilpin, R. Horvath, G. B. Jameson, S. G. Telfer, K. C. Gordon, J. D. Crowley, Organometallics 2010, 29, 6186-6195.

[28] M. T. Johnson, J. Marthinus Janse van Rensburg, M. Axelsson, M. S. G. Ahlquist, O. F. Wendt, Chem. Sci. 2011, 2, 2373-2377.

[29] R. L. LaLonde, J. W. E. Brenzovich, D. Benitez, E. Tkatchouk, K. Kelley, I. I. I. W. A. Goddard, F. D. Toste, Chem. Sci. 2010, 1, 226-233.

[30] Q. Wang, Y. Jiang, R. Sun, X.-Y. Tang, M. Shi, Chem. Eur. J. 2016, 22, 14739-14745. [31] M. Ackermann, PhD thesis, "Goldkatalysierte Zyklisierungsreaktionen", Heidelberg University,

2011.

[32] R. Döpp, PhD thesis, "Methodische Untersuchung von palladium- und goldkatalysierten Kreuzkupplungsreaktionen", Heidelberg University, 2011.

Chapter 2

Mechanistic Investigation of the

Divergence of a Light-Mediated

Gold-Catalyzed Reaction

Abstract: In this chapter, the conditions for a gold-catalyzed light-mediated reaction are

explored. o-Alkynylphenols and diazonium salts can be converted into substituted benzofurans upon irradiation with blue LED under N2-extrusion or into substituted

azobenzofurans with N2-retention in the absence of light. With essentially the same reaction

conditions, starting materials react selectively either under C-C-bond formation or C-N-bond formation, respectively.

Furthermore, the mechanistic pathways are investigated by isolation of a vinyl gold(I) complex and conduction of stoichiometric experiments, which implicate this complex as a common intermediate in both pathways.

In contrast to previous reports in literature, arylative coupling is realized without the addition of a photo(redox) catalyst under ambient conditions. To explore how the photochemical activation is feasible, DFT calculations were performed proposing an Electron-Donor-Acceptor-Complex as a key intermediate, formed by a vinyl gold(I) complex and a diazonium salt.

Parts of the results are published in: Angew. Chem. Int. Ed. 2019, 58, 16988-16993; Angew. Chem. 2019, 131, 17144-17149; "Light-Induced Mechanistic Divergence in Gold(I) Catalysis: Revisiting the Reactivity of Diazonium Salts"

30

2.1)

Introduction

Photochemical transformations involving redox processes for C-C-bond formation have been demonstrated in various reactions.[1] _{Among the well-known photoreactions using transition}

metals, such as Ir, Ru, Rh for single-electron transfers (SETs),[1] _{many advantages have been}

realized in the use of dual-gold photoredox and gold-only (photoredox) catalysis for C-C-bond formation,[2] _{in which diazonium salts are used as aryldonors for coupling-reactions.}[3] _Arylative

coupling has been explored for different types of substrates, for example in cross-coupling reactions with allenoates by homolytical cleavage of the C-N-bond to provide radicals using thermal N2-extrusion (see Scheme 12).[3x]

Scheme 12: Gold-catalyzed arylative coupling under thermal conditions by Shin and co-workers.[3x]

Besides the thermal generation of the radical, a photochemical gold-only mechanism is reported via SETs through the oxidative addition of the diazonium salt to the gold-precursor, forming a gold(III) complex without an additional photo catalyst.[3j, 3v, 3w] _{Our group}

demonstrated a gold-catalyzed arylation to generate -aryl ketones 36 by light-mediated C-C-bond formation featuring diazonium salts (see Scheme 13).[3v, 3w] _{The catalytic cycle is}

proposed to start with an initial SET from gold(I) to the diazonium salt forming a gold(II) intermediate and an aryl diazo radical that both recombine to a gold(III) intermediate. Once the gold(III) species is formed (the formation might also be possible by direct addition of the diazonium salt to the gold(I) pre-catalyst), a light-mediated N2-extrusion forms the active

gold(III) species that coordinates to and activates the multiple bond. Subsequent back-side attack of the intermolecular nucleophile to the multiple bond and reductive elimination terminates the catalytic cycle under product formation 36 and recycles the pre-catalyst.[3v, 3w]

Scheme 13: Gold-catalyzed arylative coupling under photo catalyst-free conditions by Hashmi and

co-workers.[3v, 3w]

At the same time, Fensterbank and co-workers reported a dual photoredox gold-catalyzed approach for the synthesis of substituted benzofurans 37 using diazonium salts under blue LED irradiation (see Scheme 14).[3s] _{The oxidative arylation of the gold(I) pre-catalyst to}

the catalytically active gold(III) species is featured by two SETs from the co-catalyst [Ru(bpy)3](PF6)2.

31

Introduction

Scheme 14: Dual gold-catalyzed photo(redox) catalysis to substituted benzofurans by Fensterbank

and co-wokers.[3s]

The mechanism of the arylative coupling proposed by Fensterbank and co-workers[3s] _involves

a second catalytic cycle. The additional photo catalyst ([Ru(bpy)3](PF6)2) generates a phenyl

radical E by light-mediated SET which is reacting with the gold(I) precursor 38 and gold(II) species F is formed (see Scheme 15). By a second SET, the same photo catalyst is proposed to oxidize F to the gold(III) species G, which is afterwards coordinating to the triple bond, facilitating an intramolecular nucleophilic attack and vinyl gold(III) intermediate I is formed. Deprotonation by base and reductive elimination to product 37 closes the catalytic cycle.

Scheme 15: Proposed reaction mechanism for the dual-photo(redox) catalytic arylation involving

32

This type of reactivity has already been reported in several other studies[3b, 3f, 3ad, 4] _and

stoichiometric reactions for the oxidative addition of diazonium salts to gold complexes support the proposed mechanistic pathway through a gold(III) intermediate.[3v, 3w, 5]

A few years earlier, a reaction was reported by our group[6] _{using o-alkynylphenol 25}

OMe and

diazonium salt 34OMe to form a C-N-bond with N2-retention of the diazonium salt. The reaction

to substituted azobenzofuran 32OMe,OMe does not involve a change in oxidation state, but

instead a mechanism that proceeds through a vinyl gold(I) intermediate 46OMe,IPr which is

trapped electrophilically by the diazonium salt 34OMe (see Scheme 16).[6b]

Scheme 16: Stoichiometric experiments for the formation of azobenzofuran 32OMe,OMe.[6b]

Although first attempts have been successful in optimizing the reaction conditions, a very limited number of substrates were isolated in moderate to high yields when using two different catalysts and diazonium salts.[6]

Scheme 17: Isolated azobenzofurans 32 reported by Döpp.[6a]

Inspired by the work of Fensterbank and co-workers,[3s] _{who used diazonium salts for arylative}

C-C-bond formation (see Scheme 14) and previous work in our group[6] _{which made use of}

essentially the same starting materials to form C-N-bonds (see Scheme 16 and Scheme 17), the aim of this project was to explore if both reactions may have a common vinyl gold intermediate, such as other gold-catalyzed transformations previously reported,[4, 7] _{and where}

33

Introduction

Both reactions start with a substituted o-alkynylphenol 25 that reacts in the presence of a gold catalyst with diazonium salt 34 either to substituted benzofurans 37 or azobenzofurans 32 (see Scheme 18). The varied reaction conditions are proposed to lead to different intermediates in the reaction mechanisms, a vinyl gold(III) and a vinyl gold(I) intermediate, respectively (vide supra).

Scheme 18: Similarities and differences of the reaction reported by Fensterbank and

co-workers (top) and our goup (bottom).[3s, 6]

In the following chapter, the reaction conditions for a divergent gold-only catalysis using the same starting materials are explored to form C-C- and C-N-bonds, respectively. A mechanistic study considering a vinyl gold(I) intermediate is demonstrated and supported by DFT calculations. Both reactions are briefly optimized and probed to a limited extent in regard to scope.

34

2.2)

Results and Discussion

In the following section, mechanistic studies and catalytic experiments are described. The similarities and differences in reactivity are explored using the same starting materials under the same reaction conditions to form substituted arylated benzofurans 37 and azobenzofurans 32, respectively.

Scheme 19: Divergence in Au-catalyzed reactions leading to formation of arylated benzofurans 37

(N2-extrusion) without additional photo(redox) catalyst and substituted azobenzofurans 32 (N2-retention).

To probe whether or not there are similarities and differences in reactivity, the same starting materials were treated in the presence or absence of light with different catalysts varying the solvent, base and counter anion of the gold-catalyst (see Table 1). All experiments were carried out without an additional photo(redox) catalyst. The arylation of o-alkynylphenol 25Me

with diazonium salt 34H using a blue LED light source was feasible with catalytic amounts of

Ph3PAuCl in the absence of an additional photo catalyst providing the desired product 37Me,H

in 46% yield (entry 1) and small amounts of azobenzofuran 32Me,H were formed in the absence

of light using MeCN and CH2Cl2, respectively. Changing the anion of the gold complex to the

weakly coordinating NTf2- increased the yield of azobenzofuran 32Me,H as expected, but

surprisingly increased the yield of arylated benzofuran 37Me,H as well (entry 2). In an effort to

avoid protodeauration more efficient, the more soluble base 2,6-ditertbutylpyridine (DTBP) was used, however, it did not have a beneficial effect under these reaction conditions (entries 3 and 4).

These experiments clearly demonstrate that the reaction mechanism can be controlled using the same starting materials under the same reaction conditions by simply irradiating with light.

35

Results and Discussion

Table 1: Comparison of reactions conditions used to show the mechanistic divergence.

Entry Catalyst Base Yield 37Me,H [%] irradiation at 450 nm[a,c] Yield 32Me,H [%] No irradiation[a,b] MeCN CH2Cl2 MeCN 1 Ph3PAuCl NaHCO3 46[d,e] _{5 7} 2 Ph3PAuNTf2 NaHCO3 54[d,e] ₂₂[d] ₁₁ 3 Ph3PAuCl DTBP

4 not observed not observed

4 Ph3PAuNTf2 DTBP

11 19 23

[a] Averages of duplicate runs are given. [b] General conditions: 25Me (50 µmol), [Au] (5.00 mol-%), 34H (100 µmol), base (100 µmol), solvent (500 µL), r.t., 24 h, determined using 1_{H NMR spectroscopy with benzyl acetate as internal standard. [c] General conditions:}

25Me (100 µmol), [Au] (5.00 mol-%), 34H (200 µmol), base (200 µmol), solvent (1 mL), r.t.,

2 h, 450 nm light source, determined using GC MS with hexamethylbenzene as internal standard. [d] Full conversion of starting material. [e] With Au and a ruthenium photo(redox) catalyst present, similar yields were obtained with related substrates, see Ref. [3s].

Control experiments demonstrate the necessity of the gold catalyst for both cases and the results are shown in section 2.2.1). Stoichiometric experiments were performed to explore the involvement of a gold(I) intermediate and irradiation of the (pre-)catalyst in the presence of diazonium salt indicates, that the previous oxidation of the gold(I) precursor to a gold(III) species in the present case is not feasible (see paragraph 2.2.2), 2.2.3) and 6.2.5). DFT calculations support the hypothesis of an Electron-Donor-Acceptor-Complex, that is formed during the reaction between gold(I) complex and the diazonium salt as shown in paragraph 2.2.4). Further irradiation experiments demonstrate that the arylated product is consumed by the diazonium salt under irradiation explaining the moderate yields (see 2.2.5). The scope for both reactions is briefly probed as shown in section 2.2.5) and 2.2.6).

36

2.2.1) Control Experiments

To probe that the arylative coupling of o-alkynylphenols 25 and diazonium salts 34 to substituted benzofurans 37 is gold-catalyzed and light-mediated, several control experiments were conducted which are shown in Table 2. Further irradiation experiments have been accomplished to proof if the essential formation of a gold(III) intermediate prior to the activation of the triple bond of the substrate is possible. Moreover, a possible consumption of the arylated product 37Me,H was tested by treatment with an excess of diazonium salt. Additional control

experiments for the gold-catalyzed formation of azobenzofurans 37Me,H were carried out and

the results are shown in Table 3.

2.2.1.1) Arylated Benzofurans

As shown in Table 2, arylative coupling is observed in the absence of the gold catalyst by adding diazonium salt only (entry 1). The product was formed in trace amounts by adding base under the same reaction conditions (< 5%, entry 2). The possible formation of benzofuran 41Me as an intermediate was tested by exposure to diazonium salt only and small

amounts of product were observed (entry 3). Basic reaction conditions (entry 4) and additionally added Ph3PAuNTf2 yielded low amounts of arylated benzofuran 37Me,H. Neither

AgNTf2 (entry 7), nor HBF4 (entry 6) catalyzed the reaction efficiently

but yielded traces of product.

The irradiation of solely azobenzofuran 32Me,H as well as 32Me,H under

reaction conditions did not lead to the loss of nitrogen demonstrating that azobenzofurans can not be converted into substituted benzofurans 37Me,H (entries 8 and 9, see Figure 2).

Figure 2: Azobenzofuran 32Me,H solely in MeCN (left) and azobenzofuran 32Me,H under reaction

conditions (right) after irradiation with blue LED.

Table 2: Control experiments for the formation of arylated benzofurans.

Entry Starting

Material Catalyst Base

Yield 37Me,H [%] Irradiation 1 TS-696 25Me No No Not observed 2 TS-647 25Me No NaHCO3 < 5 3 TS-698 41Me No No 5 4 TS-679-B 41Me No NaHCO3 17

5 TS-679-A 41Me Ph3PAuNTf2 NaHCO3 15

6 TS-696 25Me HBF4(a) No < 5

7 TS-692 25Me AgNTf2(b) NaHCO3 < 5

8 TS-701-A 32Me,H(c) No No Not observed

9 TS-701-B 32Me,H(d) Ph3PAuNTf2 NaHCO3 Not observed

(a) 7.7 mol-% HBF4. (b) 6.2 mol-% AgNTf2. (c) 30 µmol in 200 µL MeCN. (d) 50 µmol in 500 µL MeCN, 100 µmol base, no diazonium salt added.

37

Results and Discussion

2.2.1.2) Azo compounds

The formation of azobenzofuran 32Me,H was not observed under the conditions shown in Table

3. Neither treatment of the starting material with the diazonium salt, nor adding additional base

yielded the desired product (entries 1 and 2). Product formation was not observed when treating benzofuran 41Me with diazonium salt solely or with catalyst under reaction conditions

(entries 3 and 4). AgNTf2 (entry 6) and HBF4 (entry 5) did not function as a catalyst.

Table 3: Control experiments for the formation of azobenzofuran 32Me,H.

Entry Starting

Material Catalyst Base

Yield 32Me,H [%]

No irradiation

1 TS-695 25Me No No Not observed

2 TS-663-KE 25Me No DTBP Not observed(a)

3 TS-678-C 41Me No DTBP Not observed

4 TS-678 41Me Ph3PAuNTf2 DTBP Not observed

5 TS-690 25Me HBF4(b,c) No Not observed

6 TS-689 25Me AgNTf2(b,d) DTBP Not observed

(a) Different substituted azo compound observed (see Scheme 20 and 6.2.4.3), (b) 4 eq of 34H. (c)15.3 mol-% HBF4. (d)6.7 mol-% AgNTf2.

A differently substituted azo compound 42 was formed under basic conditions in low yields after 19 days (entry 2, see Scheme 20 and 6.2.4.3). Without catalyst, the starting material 25Me is deprotonated prior to the reaction with the diazonium salt forming the quinoic

species K / L. The electron-rich diazonium salt 34Me reacts under electrophilic addition to the

deprotonated species K / L and forms the side product 42. The side product 42 can also be observed in traces with electron-poor diazonium salts under gold-catalyzed conditions.

38

Scheme 20: Variously substituted azo compound 42 observed under basic conditions

(see 6.2.4.3).

The THP-protected alkynol 45 was synthesized in high yield and treated with diazonium salt 34Me (see Scheme 21) to investigate, if a protecting group might affect the gold-catalyzed

formation of azobenzofurans. The THP-group is cleaved during the reaction, but the formation of the Au complex seems to be slower due to the protecting group at the oxygen atom explaining the low yield of the desired product 32H,Me.

Scheme 21: Synthesis of THP-protected alkyne 45 and its formation under reaction conditions to

azo compound 32H,Me.

The relevance of this experiment will be further explored later in Chapter 4 by investigation of the formation of (bisazo)benzo[b,b’]difurans.

39

Results and Discussion

2.2.2) Irradiation Experiments

The possibility that oxidative addition of the diazonium salt to the gold(I) (pre-)catalyst occurs first was evaluated through the treatment of insoluble Ph3PAuCl and soluble Ph3PAuNTf2 with

diazonium salt 34Me. Irradiation of the mixture with blue LED for more than two hours did not

alter the concentrations of the substrates in solution (i.e. over the same time during which the catalytic reactions were performed in Table 1; monitored with 1_{H NMR spectroscopy).}

Figure 3: Mixtures monitored by 1_{H NMR spectroscopy in CD}

3CN while under irradiaton at 405 nm (see 6.2.5). The pictures in the middle show the NMR tubes before irradiation.

Based on the irradiation experiments performed, oxidation of the gold(I) catalyst which might occur before coordinating to the triple bond of the alkyne in substrate 25can be excluded. However, activation of the gold(I) catalyst with a change in oxidation state after the coordination to the triple bond can not be ruled out.

40

2.2.3) Stoichiometric Experiments

Considering the earlier observations shown in Table 1 and previous work by our group,[6] _the

vinyl gold(I) complex 46Me,Ph3P was isolated (see Scheme 22)[7d] with the aim to probe if the

reactions shown in Scheme 19 involve a common intermediate.

Scheme 22: Isolation of vinyl gold(I) complex 46Me,Ph3P according to a modified procedure by

Hashmi and co-workers,[7d] _{and solid state X-ray structure of vinyl gold(I) complex 46}

Me,Ph3P. Thermal ellipsoids are shown at 50% probability, hydrogen atoms are omitted for clarity.

The air-stable complex 46Me,Ph3P could be isolated in high yield and was treated with diazonium

salt 34H with and without irradiation (see Scheme 23).

Azobenzofuran 32Me,H was formed in 46% yield without irradiation (determined by 1H NMR

spectroscopy, full conversion of starting material) and confirms the previous result determined by Ackermann,[6b] _{who treated 46}

OMe,IPr with 2.00 eq 34OMe and isolated

azobenzofuran 32OMe,OMe with 42% yield (see Scheme 16). Irradiating a similar sample with

blue LED yielded 65% of arylated benzofuran 37Me,H under N2-extrusion (determined by

GC MS, full conversion of starting material). A similar experiment, reported earlier by Shin and co-workers was conducted by treating an isolated vinyl gold(I) intermediate with a diazonium salt to achieve C-C-bond formation but still required the presence of a photo catalyst ([Ru(bpy)3](PF6)2).[3x]

Next to the consumption of the arylated species, which was explored in 2.2.2), a major side product observed for both reactions is benzofuran 41Me which is attributed to the

protodeauration of vinyl gold(I) complex 46Me,Ph3P. In a stoichiometric experiment, the

formation of 41Me was confirmed by treating vinyl gold(I) complex 46Me,Ph3P with phenol as a

proton source. In both reaction types with catalytic amount of the gold catalyst (see Table 1 and 2.2), a large excess of starting material 25Me is present which is identified as a possible