Synthesis and Characterization of Oligothiophene - Based Fully Π-Conjugated Macrocycles

Synthesis and Characterization of Oligothiophene - based Fully π-Conjugated Macrocycles

Dissertation

zur Erlangung des Doktorgrades Dr. rer. nat. der Fakultät der Naturwissenschaften der Universität Ulm

vorgelegt von

Gerda Laura Fuhrmann

aus Neustadt (Baia Mare) Rumänien

Ulm, 2006

Amtierender Dekan: Prof. Dr. Klaus-Dieter Spindler 1. Gutachter: Prof. Dr. Peter Bäuerle 2. Gutachter: Prof. Dr. Volkhard Austel Tag der Promotionsprüfung: 15.03.2006

This thesis was elaborated and written between November 1999 and February 2006 at the department of Organic Chemistry II, University of Ulm, Germany.

für meine Eltern

“The chemist does a mysterious thing when he wants to make a molecule. He sees that it has got a ring, so he mixes this and that, and he takes it, and he fiddles around. And at the end of a difficult process, he usually does succeed in synthesizing what he wants.”

There’s a Plenty of Room at the Bottom - R. Feynman

Ich möchte mich an dieser Stelle bei all denjenigen bedanken, die mich in den zurückliegenden Jahren begleitet und unterstützt haben und damit zum Gelingen dieser Arbeit beigetragen haben. Mein besonderer Dank gilt:

- Herrn Prof. Dr. Peter Bäuerle für die stete Unterstützung und Förderung bei der wissenschaftlichen Gestaltung dieser Arbeit, für sein Interesse und persönliches Engagement, sowie für die mir gewähren Freiräume bei der Bearbeitung des Themas - Frau Dr. Pinar Kilickiran für ihre moralische und inspirative Unterstützung, für ihre unermüdliche Hilfe, für die Durchsicht und sprachliche Überarbeitung des Manuskripts und für vieles mehr - Herrn Prof. Dr. V. Austel für die stets anregende wissenschaftliche Diskussionen und seine Bereitschaft diese Arbeit zu begutachten - Herrn Dr. G. Götz für seine unermüdliche Hilfsbereitschaft in Fragen der technischen Logistik und bezüglich elektrochemischer Messungen - Frau Dr. E. Mena-Osteritz für die ständige Diskussionsbereitschaft bezüglich physikalischer Fragestellungen rund um Makrocyclen - Herrn Prof. Dr. T. Debaerdemaeker (Sektion Röntgen- und Elektronenbeugung, Universität Ulm) für die kristallographische Untersuchungen und die anregende Diskussionnen - Herrn. Dr. M. Wunderlin, Dr. U. Werz und Dr. U. Ziegler für die Aufnahme zahlreicher Massen- und NMR Spektren - den Herren Dr. C. Schalley und A. Rang (Kekule Institut für Organische Chemie, Universität Bonn) für die Durchführung hochaufgelöster Massenspektren - Herrn Dr. E. Reinhold und Dr. W. Mästle für ihre stets freundlichen Hilfestellung in Fragen der technischen und praktischen Logistik - den Mitarbeitern der Service-Abteilungen der Universität Ulm für die gute Zusammenarbeit - Herrn Prof. Dr. F. Würthner und seinen Mitarbeitern Dr. A. Sautter, Dr. C. Thalacker, und Dr. R. Dobrawa für die viele fruchtbare Diskussionen und Anregungen und das hervorragende Verhältnis während der gemeinsamen Zeit im Labor N24/308 - meinen jetzigen MEG Kollegen von MSL, Sony Stuttgart für ihre Unterstützung - all meinen Freunden für den willkommenen und wichtigen Ausgleich zum Forscherindasein, insbesondere Dani, Kruschi und „my rainbow in the sky“ - allen Mitarbeitern der Abteilung Organische Chemie II für die gute Zusammenarbeit und das angenehme Arbeitsklima, insbesondere den Herren Dr. M.-S. Schiedel, Dr. J. Krömer, Dr. M. Ammann, Dr. L. Madrigal, Dr. J. Cremer, A. Kaiser, F. Nicklas, Frau Dr. D. Caras-Quintero, Frau S. Schmidt und zuletzt, aber allen voran meinen Eltern und Geschwistern für das in mich gesetzte Vertrauen und die langjährige finanzielle und moralische Unterstützung.

Table of contents

Numbering of the compounds V Abreviations IX

1 Introduction and aim of this work 1

2 Synthesis of π-conjugated oligothiophenes 7

2.1 π-Conjugated polymers and oligomers 8 2.1.1 From π-conjugated polymers to π-conjugated oligomers 8 2.1.2 General strategies for the synthesis of conjugated oligomers 9

2.2 π-Conjugated oligothiophenes 13 2.2.1 Introduction 13 2.2.2 General methods for the synthesis of conjugated oligothiophenes 13 2.2.2.1 Aryl-Aryl Bond Formation by Oxidative Coupling 14 2.2.2.2 Transition Metal Promoted Aryl-Aryl Bond Formation 19 2.2.2.3 Ring Closure Reaction from Acyclic Precursors 27

2.3 References 35

3 Shape-persistent nanosized macrocycles 41

3.1 Introduction 42

3.2 Properties of shape-persistent macrocycles 44 3.2.1 Introduction 44 3.2.2 Electronic and optical properties of shape-persistent macrocycles 46 3.2.3 Aggregation behaviour of shape-persistent macrocycles in solution 50 3.2.4 Solid state structures of shape-persistent macrocycles 53 II

3.2.5 Self-assembly of shape-persistent macrocycles at 55 the solid-liquid interface 3.2.6 Liquid crystalline mesophases of shape-persistent macrocycles 56 3.2.7 Host-guest interactions of shape-persistent macrocycles 57

3.3 General strategies for the synthesis of nanosized macrocycles 61 3.3.1 Introduction 61 3.3.2 The kinetic approach towards synthesis of macrocycles 61 3.3.3 The thermodynamic approach towards synthesis of macrocycles 66 3.3.4 The template approach in macrocycle chemistry 69

3.4 Alkynyl-alkynyl coupling reactions 72

3.5 Conclusion 76

3.6 References and notes 77

4. Macrocyclic oligothiophene-diacetylenes through 83 oxidative coupling reactions

4.1 Cyclo[n]thiophene target structure 84

4.2 Synthetic strategy towards cyclo[n]thiophenes 87

4.3 Synthesis of α,ω-diethynylated oligothiophenes 92 4.3.1 Synthesis of conjugated oligothiophenes 93 4.3.2 Synthesis of α,ω-diethynylated oligothiophenes as precursors for 102 cyclization

4.4 Macrocyclization by oxidative coupling reaction 104 4.4.1 Synthesis of macrocyclic terthiophene-diacetylenes C[3T-DA] 104 4.4.2 Synthesis of macrocyclic quinquethiophene-diacetylenes 119 C[5T-DA] 4.4.3 Intramolecular ring-closure reaction of α,ω-diethynyl- 128 undecithiophene III

4.5 Conclusion 134

4.6 Experimental section 135 4.6.1 Instrumentation and general experimental conditions 135 4.6.2 Synthesis and characterization of the compounds 137

4.7 Supplement – 169

MALDI TOF MS mass spectra of C[3T-DA]n and C[5T-DA]n macrocycles

4.8 References and notes 180

5. Conjugated macrocycles by metal template approach 183

5.1 Introduction 185

5.2 Basic concept of the metal template approach 185

5.3 Transition metal directed self-assembly 188

5.4 Transition metal σ-acetylide complexes 191

5.5 Carbon-carbon bond formation by reductive elimination from 199 a transition metal center

5.6 Results and discussion 216 5.6.1 Metal template approach - method development 216 5.6.1.1 Palladium mediated macrocyclization of diethynyl- 216 quinquethiophene 5.6.1.2 Synthesis of linear bis(oligothiophene alkynyl)-Pt(II) complexes 221 5.6.1.3 1,3-butadiynes-linked oligothiophene by reductive elimination 232 5.6.2 Macrocyclization by the metal template approach 234 5.6.2.1 Synthesis of bis(oligothiophene-diacetylene)diplatina macrocycles 235 5.6.2.2 Macrocyclic oligothiophene-diacetylenes by reductive elimination 249 5.6.3 Mechanistic investigations on the reductive elimination reactions 262 5.6.4 Synthesis of fully conjugated cyclo[n]thiophenes 277 IV

5.6.5 Optical characterization of platinum-linked ethynyl- 288 oligothiophenes 5.6.6 Optical characterization of the smallest conjugated macrocycles 45 292 and 47 5.6.7 Electrochemical characterization of the smallest conjugated 296 macrocycles 45 and 47 5.6.8 X-ray crystallography of cyclodimeric terthiophene-diacetylene 45 299

5.7 Conclusion 303

5.8 Experimental section 305 5.8.1 Instrumentation and general experimental conditions 305 5.8.2 Synthesis and characterization of the compounds 307 5.8.3 Qualitative and preparative electrolysis experiments 326

5.9 Supplement 327 5.9.1 MALDI-TOF MS spectra of the cyclo[n]thiophenes 47-50 327 5.9.2 X-ray crystallographic data of cyclodimeric terthiophene- 330 diacetylene 45

5.10 References and notes 336

Summary 341 Zusammenfassung 347

List of Publications Curriculum Vitae

Numbering of the compounds

1 3,4-Dibutylthiophene 2 2-Bromo-3,4-dibutylthiophene 3 2,5-Dibrom-3,4-dibutylthiophene 4 3,3’’,4,4’’-Tetrabutyl-2,2’:5’,2’’-terthiophene 5 3’,4’-Dibutyl-2,2’:5’,2’’-terthiophene 6 5,5’’-Dibromo-3’,4’-dibutyl-2,2’:5’,2’’-terthiophene 7 3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethiophene 8 5-Bromo-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene 9 5,5’’-Dibromo-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene 10 5-Bromo-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethiophene 11 5,5’’’’-Dibromo-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethiophene 12 3,3’’,4,4’’-Tetrabutyl-5-trimethylsilylethynyl-2,2’:5’,2’’-terthiophene 13 3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-5-trimethylsilylethynyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethiophene 14 3,4-Dibutyl-2-trimethylsilylethynyl-thiophene 15 1,4-Bis(3,4 –dibutylthien-2-yl)-1,3-butadiyne 16 1,4-Bis(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-2-yl)-1,3-butadiyne 17 1,4-Bis(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethien- 5-yl)-1,3-butadiyne 18 3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Octabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-septithiophene 19 3,3’’,3’’’’,3’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,4’’’’’’’’’’-Dodecabutyl-2,2’:5’,2’’:5’’, 20 2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-undecithiophene 21 5,5’’-Diiodo-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene 22 5,5’’’’-Diiodo-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethiophene VI

23 5,5’’’’’’-Diiodo-3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-octabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-septithiophene 24 5,5’’-Bis(trimethylsilylethinyl)-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene 25 5,5’’’’-Bis(trimethylsilylethynyl)-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethiophene 26 5,5’’’’’’-Bis(trimethylsilylethynyl)-3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-octabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-septithiophene

27a Cyclo{tris[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]}

27b Cyclo{tetrakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)- 1,1’-diyl]}

27c Cyclo{pentakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)- 1,1’-diyl]}

27d Cyclo{hexakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)- 1,1’-diyl]}

27e Cyclo{heptakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)- 1,1’-diyl]}

27f Cyclo{octakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]}

27g Cyclo{nonakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)- 1,1’-diyl]}

27h Cyclo{decakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)- 1,1’-diyl]}

29a Cyclo{bis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl]}

29b Cyclo{tris[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl]}

29c Cyclo{tetrakis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethien-5,5’’’’-diyl)-1,1’-diyl]}

29d Cyclo{pentakis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethien-5,5’’’’-diyl)-1,1’-diyl]} VII

29e Cyclo{hexakis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethien-5,5’’’’-diyl)-1,1’-diyl]}

30 5,5’’’’’’’’’’-Diiodo-3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,- 4’’’’’’’’,4’’’’’’’’’’-dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,- 2’’’’’’:5’’’’’’,2’’’’’’’:5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’- undecithiophene 31 5,5’’’’’’’’’’-Bis(trimethylsilylethynyl)- 3,3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,-4’’’’’’,4’’’’’’’’,4’’’’’’’’’’- dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’:5’’’’’’,- 2’’’’’’’:5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’-undecithiophene 33a Cyclo[diyne-2,2’-(3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,- 4’’’’’’’’,4’’’’’’’’’’-dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,- 2’’’’’’:5’’’’’’,2’’’’’’’:5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’- undecithien)-5,5’’’’-diyl)-1,1’-diyl]

33b Cyclo{bis[diyne-2,2’-(3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,- 4’’’’’’,4’’’’’’’’,4’’’’’’’’’’-dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,- 2’’’’’:5’’’’’,2’’’’’’:5’’’’’’,2’’’’’’’:5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,- 2’’’’’’’’’’’-undecithien)-5,5’’’’-diyl)-1,1’-diyl]

35 cis-Bis[3,4--dibutyl-2-(ethynyl-κC2)-thienyl][ propane -1,3-diyl-bis(diphenylphosphine-κP)]platinum(II)

37 cis-Bis[3,3’’,4,4’’-tetrabutyl-5-(ethynyl-κC2)-2,2’:5’,2’’-terthienyl][ propane -1,3- diyl-bis(diphenylphosphine-κP)]platinum(II)

38 trans-Bis[3,4--dibutyl-2-(ethynyl-κC2)-thienyl][propane -1,3-diyl-triphenylphosphine-κP)]platinum(II)

40 Cyclo[cis-bis[3,3’’,4,4’’-tetrabutyl-5,5’’-(diethynyl-κC2)-2,2’:5’,2’’- terthienyl][propane -1,3-diyl-bis(diphenylphosphine-κP)]diplatinum(II)]

41 Cyclo[cis-bis[3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-5,5’’’’-(diethynyl-κC2)- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethienyl][propane -1,3-diyl- bis(diphenylphosphine-κP)]-diplatinum(II)] 43 Cyclo[cis-bis[3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Octabutyl-5,5’’’’’’-(diethynyl- κC2)-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’’,2’’’’’’-septithienyl][propane -1,3-diyl- bis(diphenylphosphine-κP)]diplatinum(II)] 44 Cyclo[trans-bis[3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Octabutyl-5,5’’’’’’-(diethynyl- κC2)-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’’,2’’’’’’-septithienyl][propane -1,3-diyl- triphenylphosphine-κP)]diplatinum(II)] VIII

47 Cyclo[8]thiophene 48 Cyclo[12]thiophene 49 Cyclo[16]thiophene 50 Cyclo[18]thiophene

Abreviations abs. absolute b.p. boiling point calcd calculated corr. corrected CV cyclic voltammetry DCM dichloromethane DMF N,N-dimethylformamid DMSO dimethyl sulfoxide dppp 1,3-bis(diphenylphosphine)propane E0 standard potential EI electron impact em emission ESI electronspray-ionization FAB fast-atom bombardment Fc/Fc+ ferrocene / ferrocenium couple FT fourier transformation GC gas chromatography HOMO highest occupied molecular orbital HPLC high performance liquid chromatography IR infrared spectroscopy L.R. Lawesson’s reagent LUMO lowest unoccupied molecular orbital M metal MALDI matrix assisted laser desorption ionization max maximum m.p. melting point MS mass spectrometry NBS N-bromosuccinimide NMR nuclear magnetic resonance Ox oxidation Red reduction RT room temperature X

SCE saturated calomel electrode TBAAHFP tetra-n-butylammonium hexaflurophosphate THF tetrahydrofuran TLC thin layer chromatography TOF time of flight UV/Vis ultraviolet / visible light spectrum Chapter 1 Introduction and aim of this work 1

Chapter 1

Introduction and aim of this work

The research of shape-persistent macrocycles on the nanometer scale has become a topic of growing interest in the field of organic chemistry, of material science and more recently of nanotechnology.1 Apart from the intense synthetic activity towards developing versatile and efficient methods to prepare such macrocycles, special emphasis is given to the studies concerning their structural and physical properties. These include, among others, the binding of appropriate guest molecules, the pattern formation at interfaces, the organization of the molecules in the solid and in the liquid crystalline states, and consequently, their potential applicability as ion channels, liquid crystalline materials, microscopic reactors, porous molecular solids or artificial enzymes.2 Polythiophenes and their finite model oligomers, α-conjugated oligothiophenes, belong to the most thoroughly investigated π-conjugated systems in the field of material science.3 Due to their chemical stabilities in their various redox forms and their outstanding structural and electronic properties, they are potential candidates for applications in molecular electronic devices such as organic field-effect transistors,4 light-emitting devices,5 photovoltaic cells6 or even as molecular wires for information storage or transfer.7 The systematic investigations of well-defined oligothiophenes provide valuable information about the structure-property relationship of these compounds and by correlation of their corresponding polymers. Such correlations are essential prerequisites for a molecular design and engineering of material properties. In general, the physical properties of oligomers with a well-defined π-conjugated chain length are influenced by undesired perturbing end-effects.8 In this respect, much more appealing seems to be molecules with a well-defined cyclic structure. Over the usual linear conjugated oligomers and polymers, the cyclic compounds have the distinct advantage, to 2 Chapter 1 Introduction and aim of this work

ideally combine an infinite defect-free π-conjugated chain of an idealized polymer with the structural feature of a well-defined oligomer, but excluding perturbing end-effects. The aim of this thesis was to develop a method that enables the efficient synthesis of macrocyclic oligothiophenes (Figure 1.1). These fully π-conjugated macrocycles, which are designated as cyclo[n]thiophenes (n = number of thiophene rings), represent a novel class of shape-persistent nanosized macrocycles with particularly appealing characteristics and perspectives. Owing to their cyclic structures they are of special interest acting as ideal models for the corresponding polythiophenes. Moreover, the fascinating structural architecture and rigidity combined with fully π-conjugated periphery might enable these macrocycles to act as intringuing “molecular circuits” which additionally would include sites for selective recognition and selective complexation.

S S S

S S

S S S

Figure 1.1. Cyclo[n]thiophene target structure with n = number of thiophene rings.

A very promising synthetic strategy towards fully conjugated cyclo[n]thiophenes includes the formation of macrocyclic oligothiophene-diacetylenes C[mT-DA]n (m is number of thiophenes in one subunit; n is the number of diethynyl-oligothiophene subunits) by oxidative coupling reactions of appropriate diethynylated oligothiophenes (Figure 1.2). In a following reaction with sulfide nucleophiles, the butadiyne units of these oligothiophene- diacetylene macrocycles can be converted to thiophenes affording the targeted cyclo[n]thiophenes.9

Chapter 1 Introduction and aim of this work 3

C[n]T S S S

ring closure reaction with sulfide nucleophiles „ S2- “

C[mT-DA]n S S

sp-sp homocoupling of terminal acetylenes R

H S n S H H

Figure 1.2. Retrosynthesis towards cyclo[n]oligothiophenes.

The major macrocyclization reactions are based on oxidative coupling reactions. These are kinetically controlled processes in which oligomerization and cyclization take place at the same time. As a result, in general, product mixtures are obtained in low yields that critically depend on the structural type of the monomer building blocks and the feasibility to separate the obtained mixtures. Relatively new and more efficient concepts to prepare cyclic structures have been established in the field of supramolecular chemistry.10 In these approaches, the cyclization step is carried out by thermodynamically controlled reactions. These reactions are reversible and have the ability to self-heal by making and breaking of covalent bonds as an on-going process, enabling the formation of cyclic structures in high yields. In this respect, the development of a synthetic strategy towards cyclo[n]oligothiophenes based on a reaction that is thermodynamical controlled is envisioned in this study. The majority of the supramolecular chemistry research is based on macrocyles containing heteroatoms that enable directional coordinative bonds to transition metal centers. In this approach, transition metals are employed as the key structure templating units. Very recently, a few reports on synthesis of macrocycles having acetylide units directly linked to transition metals by σ-bonds have been published.11 Based on this concept, first, the preparation of metallacycles containing cis-orientated transition metal- bridged diethynylated oligothiophenes under thermodynamical control has been targeted. These novel cyclic compounds would represent a novel class of compounds in the area of 4 Chapter 1 Introduction and aim of this work

nananosized macrocycles. By a following reductive elimination reaction the transition metal corners of the oligothiophene-derived metallacycles should be expelled affording the macrocyclic oligothiophene-diacetylenes. Such reductive elimination processes from transition metal centers under simultaneous carbon-carbon bond formation of the organic partners is a well-known step in transition metal coupling reactions. However, this reaction was never tried in the synthesis of macrocyclic compounds. Thus, the particular challenge of this study is to investigate and develop conditions where the reductive elimination is applied in a controlled manner to yield the desired macrocycles. Subsequently, as previously mentioned, the macrocyclic oligothiophene-diacetylenes can be converted by reaction with sulfides to the targeted fully conjugated cyclo[n]oligothiophenes. This new strategy, which is designated as the “metal template approach”, is schematically illustrated in Figure 1.3.

nT nT

nT M M + M

nT nT

Figure 1.3. The strategy behind the metal template approach: M = transition metal of d8 group; nT = oligothiophene; = ethynyl-unit.

References

1 For recent reviews, see: a) D. Zhao, J.S. Moore, Chem. Commun. 2003, 807-818; b) C. Grave, A.D. Schlüter, Eur. J. Org. Chem. 2002, 3075-3098; c) S. Höger, J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 2685-2698; d) M.M. Haley, J.J. Pak, S.C. Brand, Top. Curr. Chem. 1999, 201, 81-130; e) A. de Meijere, S.I. Koszushkov, Top. Curr. Chem. 1999, 201, 1-42.

2 For a selection of the most recent examples: a) C. Grave, D. Lentz, A. Schäfer, P. Samori, J.P. Rabe, P. Franke, A.D. Schlüter, J. Am. Chem. Soc. 2003, 125, 6907-6918;.b) K. Campbell, R. Mc.Donald, M.J. Ferguson, R.R Tykwinski, Organometallics, 2003, 22, 1353-1355; c) M Srinivasan, S. Sankararaman, H. Hopf, B. Varghese, Eur. J. Org. Chem. 2003; 660-665; d) X. Shen, D.M. Ho, R.A. Jr. Pascal, Org. Lett. 2003, 5, 369-371; e) M. Higuchi, H. Kanazawa, K. Yamamoto, Org. Lett. 2003, 5, 345-347; f) P.N.W. Baxter, Chem. Eur. J. 2002, 8, 5250-5264; g) M. Schmittel, A. Ganz, D. Fenske, Org. Lett. 2002, 4, 2289- 2292. Chapter 1 References 5

3 a) P. Bäuerle in Oligothiophenes in Electronic Materials: The Oligomer Approach, Eds.: K. Müllen, G. Wegner, Wiley-VCH, Weinheim, Germany, 1998, pp. 105-197; b) Handbook of Oligo- and Polythiophenes, Eds.: D. Fichou, Wiley-VCH, Weinheim, Germany, 1999.

4 a) H. Sirringhaus, R.H. Friend, X.C. Li, S.C. Morati, A.B. Holmes N. Feeder, Appl. Phys.Lett. 1997, 71, 3871-3873. b) F. Garnier in Field- Effect Transistors Based on Conjugated Materials in Electronic Materials: The Oligomer Approach (Eds.: K. Müllen, G. Wegner,) Wiley-VCH, Weinheim, Germany, 1998, pp. 559-584; c) G. Horowitz, Adv, Mater. 1998, 10, 365-370; d) Z. Bao, Adv. Mater. 2000, 12, 227-230; e) H.E. Katz, Z. Bao, S.L. Gillat, Acc. Chem. Res. 2001, 34, 359-369; f) T. Otsubo, Y. Aso, K. Takimiya, Bull. Chem. Soc. Jpn. 2001, 74, 1789-1801; g) C.D. Dimitrakopoulus, P.R.L. Malenfant, Adv. Mater. 2002, 14, 99- 117.

5 a) U. Mitschke, P. Bäuerle, J. Mater. Chem. 2000, 10, 1471-1507; b) Y. Shirota, J. Mater. Chem. 2000, 10, 1-25.

6 a) N. Noma, T. Tsuzuki, Y. Shirota, Adv. Mater. 1995, 7, 647-648.

7 a) Molecular Electronics: Science and Technology, Eds.: A. Aviram, M. Ratner, New York, Academy of Sciences, New York, 1998; b) R.M. Metzger, Acc. Chem. Res. 1999, 32, 950-957; c) C. Joachim, J.K. Gimsewski, A. Aviram, Nature 2000, 408, 541-548; d) J.M. Tour, Acc. Chem. Res. 2000, 33, 791-804; e) T. Otsubo, Y. Aso, K. Takimiya, J. Mater. Chem. 2002, 12, 2565-2575.

8 P. Bäuerle, Adv. Mater. 1992, 4, 102-107.

9 J. Krömer, I. Rios.Carreras, G. Fuhrmann, C. Musch, M. Wunderlin, T. Debaerdemaeker, E. Mena-Osteritz, P.Bäuerle, Angew. Chem. Int. Ed. Engl. 2000, 39, 3481-3486.

10 a) J.-M. Lehn, Science 1985, 227, 849-856; b) M. Mascal, Contemp. Org. Synth. 1994, 1, 31-46; c) J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995; d) Comprehensive Supramolecular Chemistry, Eds.: J.L. Atwood, J.E.D. Davies, D.D. Macniol, F. Vögtle, Pergamon, Oxford, 1996; e) H.-J. Schneider, A. Yatsimirsky, Principle and Methods in Supramolecular Chemistry, Wiley, Chiester, 2000; f) G.R. Desiraju, Nature 2001, 412, 397-400.

11 a) R. Faust, F. Diederich, V. Gramlich, P. Seiler, Chem. Eur. J. 1995, 1, 111-117; b) J.D. Brandsaw, L. Guo, C.A. Tessier, W.J. Youngs, Organometallics 1996, 15, 2582-2584; c) D. Zhang, D.. McConville, C.A. Tessier, W.J. Youngs, Organometallics 1997, 16, 824-825; d) S.M. AlQaisi, K.J. Galat, M. Chai, D.G. Ray, P.L Rinaldi, C.A Tessier, W.J. Youngs, J. Am. Chem. Soc. 1998, 120, 12149-12150; e) J.J. Pak, T.J.R. Weakley, M.M. Haley, Organometallics 1997, 16, 4505-4507; f) E. Bosch, C.L. Barnes, Organometallics 2000, 19, 5522-5524; g) S.J. Lee, A. Hu, W. Lin, J. Am. Chem. Soc. 2002, 124, 12948-12949; h) K. Campbell, R. McDonald, M.J. Ferguson, R.R. Tykwinski, Organometallics 2003, 22, 1353-1354.

Chapter 2

Synthesis of π-conjugated oligothiophenes

Abstract

In this chapter, an overview on the research field of π-conjugated oligomers by focusing on new synthetic strategies towards conjugated oligothiophenes is provided. The linear π- conjugated oligothiophenes are the key building blocks in the synthesis of the targeted fully conjugated cyclo[n]thiophenes. 8 Chapter 2 Synthesis of π-conjugated oligothiophenes

2.1 π-Conjugated polymers and oligomers

2.1.1 From π-conjugated polymers to π-conjugated oligomers

Taking into consideration that the Nobel Prize 2000 for chemistry was given to studies on polyacetylene,1 it is certainly clear that today π-conjugated polymers are a very attractive and important research in the field of chemistry and physics. The study of these conducting and electroactive compounds started with the discovery of metallic electrical conductivity in oxidatively doped polyacetylene by Shirakawa, Heeger and MacDiarmid in 1977.2 Extensive theoretical and experimental studies were initially adressed towards the properties of trans-polyacetylene PA3 and then shifted towards environmentally more stable conjugated polymers such as poly-p-phenylene PPP,4 poly-p-phenylenevinylene PPV,5 polyanilin PAni,6 polypyrrole Py7 and polythiophene PT8 (Figure 2.1). These polymers overcame the limited stability of polyacetylene by stabilizing the polyene structure with heteroatoms or arene moieties.9

n n n PA PPP PPV

H N S N N N N n N n N S n S H H PAni PPy PT

Figure 2.1. Representative examples of conjugated polymers: trans-polyacetylene PA, poly-p-phenylene PPP, poly-p-phenylenevinylene PPV, polyanilin PAni, polypyrrole PPy, polythiophene PT.

As a result of enormous research efforts towards novel polymers with improved properties, a new class of materials which uniquely combine the electronic and optical properties of metals and semiconductors, with the processing advantages and properties of polymers, have been developed. In 1980s, a variety of bulk applications such as antistatic coatings, electromagnetic shielding and energy storage devices were targeted. With the first report on electroluminescence of PPV in 1990,10 a new era of these conjugated polymers as active components for electro-optical devices was opened. New applications as laser dyes,

2.1 π-Conjugated polymers and oligomers 9 scintillates, photoconductors, organic light-emitting and non-linear optical materials became possible.11 In spite of outstanding achievements in the field of device fabrication, a fully understanding of the intrinsic electronic and optical properties of the conjugated polymers is still far from being complete. This is partly due to the fact that most synthetic polymerization reactions are statistical processes generating polydisperse materials. Therefore, the influence of the structures on macroscopic properties is hard to determine. This makes the improvement of such materials difficult. Recognizing this situation, studies involving synthesis and investigations of monodisperse oligomers became a key feature in the area of polymer research. These oligomers with well-defined chain length lead to a more precise determination of structure-property relationships. Extrapolation of the physical properties toward infinite chain lengths should allow a description of the corresponding conjugated polymers enabling a prediction of the electronic, thermal, optical and morphological properties.12 Not only as excellent models for their corresponding polydisperse macromolecular analogues, but also as novel class of advanced nanoscale molecular materials in their own right, structurally well-defined conjugated oligomers have currently attracted much attention. Nowadays, they are promising active materials finding applications in electrical conductors, electroluminescent devices, field effect transistors, photovoltaic cells and nonlinear optics.12,13 Furthermore, they are expected to serve as molecular wires that will be indispensable to forthcoming single molecular electronic devices.14

2.1.2 General strategies for the synthesis of conjugated oligomers

In the last decades extensive effort was put in developing facile and reliable methods for the synthesis of well-defined conjugated macromolecular architectures. In the meanwhile, for all basic conducting polymers corresponding homologues series of oligomers have been synthesized. The major obstacle in studying defined oligomers and polymers lies in the difficulty of isolating them in pure form. The classical method for obtaining oligomers of discrete length involves random oligomerisation and subsequent fractionation. The yield of the desired product in this case is often low and an adequate separation technique is needed. For example, the copper-promoted oxidative coupling reaction of the bis-deprotected

10 Chapter 2 Synthesis of π-conjugated oligothiophenes alkyne depicted in Scheme 2.1 yielded in the presence of phenylacetylene as end-capping group a mixture of soluble oligomers which could subsequently be separated by chromatography.15 The authors suggest using these oligomers with a poly(triacetylene) backbone as models for intermediates between polyacetylene, polydiacetylene and carbyne. The remarkable stability of these nanometer-sized molecular rods could make them additionally useful as molecular wires in molecular electronics.

SiPr3 SiPr3

H CuCl / TMEDA / O2 Ph

H Ph PhC CH (2 equiv.) n

Pr3Si SiPr3 n = 1-5 [28% - 2%]

Scheme 2.1. Preparation of soluble carbon rods with a poly(triacetylene) backbone by random oligomerization.

Nowadays, more often controlled stepwise synthetic approaches based on modern organometallic coupling reactions are used for the preparation of defined oligomers. These controlled growth approaches can take several forms. The stepwise methods involve the repetitive addition of a monofunctionalized monomer or a longer unit to the growing chain. After the addition, the material is usually purified and the new end refunctionalized. Following, another unit is added to the end of the chain. The addition of a longer unit enables a more rapid chain growth. Besides, due to the size differences between the reacted and the unreacted segments, purification is simpler. However, to circumvent the time-consuming purification processes that are needed after each step a solid-phase support-based oligomer synthesis has been developed. This method has been pioneered by Merrifield for the preparation of oligopeptides.16 In the solid-phase synthesis the monomer units are bound via a covalent linkage to an insoluble polymer support. The free end of the polymer supported unit is then activated and coupled to a new unit that is dissolved in a suspension of the polymer supported material. After reactivation, new units are successively added to the ends of the previous units. By-products, such as homocoupling products, can be removed by filtration and washing of the solid phase. A distinct advantage of the solid-phase supported method is the possibility to use the coupling components in excess leading to improved yields of the coupling reactions. The

2.1 π-Conjugated polymers and oligomers 11 first effective application of the stepwise addition approach combined with the solid-phase technique has been reported by Young and Moore in 1994.17 The iterative synthetic sequence towards a series of conjugated oligo(phenylene ethynylene)s included first a deprotection of the silyl protected resin-bound phenylacetylene. For the chain grow in a following step the deprotected acetylene was reacted by a Sonogashira coupling reaction. Iteration of the sequence afforded after cleavage from the resin with iodomethane homologues up to the hexamers in a very good overall yields (Scheme 2.2).

SiMe3 N N 2 n CH2O R R

SiMe3 Pd2(dba)3 SiMe3 N CuI / PPh3 / NEt3 N 2 n+1 CH2O

R n = 1,2,3,4,5 R = H, t-Bu

Scheme 2.2. Synthesis of oligo(1,3-phenylene-ethylene)s by stepwise addition approach on solid support.

A much more rapid and elegant controlled growth approach is the iterative divergent/convergent strategy.18 As schematically illustrated in figure 2.2, in this case a monomer M with two orthogonal functional inactive groups X and Y is used. By one specific reaction, group X is activated to X´. In a second specific one Y is activated by conversion to Y´. When the two parts are reacted with each other, with the loss of X´Y´, a dimerization of the compound takes place. Since the same end groups X and Y are now present in the dimer, the procedure can be repeated with doubling of molecular length at each iteration.

X'MY X'MMY -X'Y' -X'Y' XMY XMMY XMMMMY etc.

XMY' XMMY'

Figure 2.2. Molecular length doubling by iterative divergent/convergent approach.

12 Chapter 2 Synthesis of π-conjugated oligothiophenes

Incomplete reactions result in unreacted material that is half the size of the desired product. Hence, purification at each step is far simpler. The high efficiency of this method to prepare structure defined long chain oligomeres by rapid molecular length grow was clearly demonstrated by Tour et al.19 In their approach to prepare soluble oligo(2,5- thiophene-ethynylene)s they used for the iterative synthetic sequence only three sets of reactions: an iodination and a protodesilylation of the monomer followed by a palladium- catalyzed coupling reaction between the two functionalized building blocks. At each stage of the iteration, the length of the framework is doubled. Thus, starting with the monomer they could prepare by this method the dimer, tetramer, octamer and the corresponding hexadecamer, the last one having a theoretical chain length of 10.0 nm (Scheme 2.3). These rigid rod conjugated oligo(thiophene-ethynylene)s were expected to act as molecular wires in molecular scale electronic devices.

C2H5

H S SiMe3 2n Pd(PPh3)2Cl2 n = 4

CuI / i-Pr2NH

C2H5

K2CO3 H S H 2n-1

C2H5 1. LDA I S 2. I2 SiMe3 2n-1

Scheme 2.3. Iterative divergent/convergent synthesis of 2,5-thiophene ethynylene hexadecamer.

2.2 π-Conjugated oligothiophenes 13

2.2 π-Conjugated oligothiophenes

2.2.1 Introduction

Poly- and oligothiophene based materials are particularly attractive as these compounds can be characterized in both, the neutral and oxidized state, showing excellent environmental and thermal stabilities. Due to their chemical stability, ease of functionalization, and intrinsic electronic properties, α-linked oligothiophenes are the most extensively investigated oligomers for conducting polymers in recent years.20 On account to their controllable and precisely defined structure, physical properties can be followed and correlated with the chain and conjugation length. With the implementation of these results to polymer design, conjugated oligothiophenes were recognized as a novel and independent class of materials in their own right. Moreover, suitable modifications on their molecular structures by functionalization at the terminal α or side β positions permit their application as molecular materials in organic field-effect transistors,21 light-emitting devices,22 photovoltaic cells23 or even as molecular wires for information storage or transfer.14,24

2.2.2 General methods for the synthesis of conjugated oligothiophenes

Several synthetic routes to linear conjugated oligothiophenes have been established, however, their synthesis by step-wise assembly of defined units are typically tedious with respect to (isomeric) purity and yield, particularly for longer oligomers.20 Generally, due to the rigid nature of their π-system unsubstituted oligothiophenes become sparingly soluble with the chain extension and no higher homologues than the octamer could have been prepared.25 Sufficient solubility is not only crucial for synthesis, purification and characterization but also for application purposes. This inherent disadvantage of oligomers can be overcome by incorporation of flexible alkyl groups. Introducing alkyl groups at the α-positions allowed an additional extention from octamer to decamer,26 but it still was not enough to further increase the solubility. Much more effective is substitution with alkyl groups at the β-positions of the thiophenes. Although branched alkyl groups can weaken the π-conjugation by inducing torsion around adjacent thiophene-rings or by inhibiting the

14 Chapter 2 Synthesis of π-conjugated oligothiophenes tight packing of molecules in the solid state, this approach has proven of great value and thus, in the last decade many extended derivatives were developed.27 It was also shown that such alkyl substituents, when they are properly arranged so as to spread in the lateral direction and not interact with one another, do not influence the inherent electronic structures and conductivities of oligothiophenes.28 In general, two different synthetic routes are preferred for the synthesis of α-linked conjugated oligothiophenes, either modern organometallic aryl-aryl coupling reactions of oligothiophene building blocks or ring closure reactions from acyclic precursors.

2.2.2.1 Aryl-aryl bond formation by oxidative coupling reaction

One of the useful methods to form σ-C-C-bonds is the metal-promoted coupling of organic halides. Already in the 1930’s a whole series of unsubstituted oligothiophenes up to the α- septithiophene have been synthesized, separated and characterized by Steinkopf et. al.29 They carried out an Ullmann analogues reaction of 2-iodothiophene with copper bronze as shown in Scheme 2.4.

Cu-bronze S S I S n S (n= 0-5)

Scheme 2.4. Synthesis of unsubstituted α-oligothiophenes by Steinkopf et al.

Considerably a more effective way to synthesize oligothiophenes is the oxidative coupling of active organometallic compounds, such as oligothienyl-magnesium or much better oligothienyl-lithium, with salts of transition metals, mostly copper(II) chloride.30,31 Lithiation of thiophenes can be achieved either by simple deprotonation or by halogen- metal exchange using n-butyllithium (n-BuLi) or lithiumisopropylamide (LDA) as base. This reaction, however, even upon using equimolar amounts of base always includes dilithiation as a side reaction. In general, the selectivity of this reaction decreases with the increasing chain length of the oligothiophene. In addition, due to the fact that the α-protons of the longer oligomers exhibit greater acidity than shorter ones, a following lithiation by

2.2 π-Conjugated oligothiophenes 15 deprotonation of the desired product and formation of its dimer as byproduct is also always found.31,32 Nevertheless, in 1996 following this approach Bäuerle et al. succeeded in the preparation of the longest oligothiophene at that time as a part of a homologues row of isomerically pure α-linked oligothiophenes substituted with solubilizing dodecyl alkyl chains (Scheme 27f 3.5). By oxidative coupling of lithiated tetramer 4T with CuCl2 the corresponding octithiophene 8T and duodecithiophene 12T were obtained in one step. Owing to their good solubility, the two compounds could be separated and pure isolated in 30% and 8% yields, respectively. The octamer was then coupled in a second step with the same reagents to give the corresponding hexadecamer 16T in 19% yield after chromatographic work-up. The authors studied their physical properties and obtained the first STM images on graphite of physisorbed two-dimensional monomolecular layers of the oligomers with submolecular resolution.

C12H25 S S S S

C12H25 4T

n-BuLi

CuCl2

C12H25 C12H25 S S S S H H H S S 2 + H S S 3

C12H25 C12H25 8T [30%] 12T [8%]

n-BuLi

CuCl2

C12H25

S S H H S S 4

C12H25 16T [19%]

Scheme 2.5. Homologues series of isomerically pure α-linked oligothiophenes reported by Bäuerle et al.

16 Chapter 2 Synthesis of π-conjugated oligothiophenes

While polythiophene itself is insoluble in any solvent and therefore not processable, in the neutral state, poly(3-alkylthiophenes) are much better soluble and exhibit comparable electrical and optical properties to the unsubstituted derivatives.33 In 1986, Sato et al.34 and Lemaire et al.35 reported the electrochemical polymerization of polythiophene. A few months later, Sugimoto et al.36 achieved the first chemical polymerization of 3- alkylthiophene utilizing 4 equivalents of iron(III)chloride as the oxidative reagent in chloroform. Using this method, chemical polymerization of various 3-alkylthiophenes has been performed. Studies concerning the dependence between the structural regularity and molecular weight of the obtained polymer and the overall performance of a polymer with regard to conductivity in combination with useful mechanical properties have been carried out.37 Due to head-to-head couplings between of thiophene rings, adjacent alkyl substituents cause sterical hindrance and diminution of the conjugation because of resulting non planar conformation. Since a much more effective synthesis of highly regioregular poly(3-alkylthiophenes) by 38 metal-catalyzed coupling reactions has been reported, FeCl3 is not used very often for the synthesis of polymers but is still finding applicability in the synthesis of well-defined oligomers. A remarkable study on a oligomerization of 3-(alkylthio)-thiophenes with iron(III)chloride was recently reported by Barbella et al.39 They showed that the formation of these regioregular conjugated oligothiophenes, from trimer to octamer, mostly depend on the length of the alkyl chain and the solvent used (Scheme 2.6).

2.2 π-Conjugated oligothiophenes 17

H3C CH3 S S CH Cl 2 2 S 62% S S

CH3

H3C CH3 S S CHCl 3 S S H3C S S S 52% S S

S H3C CH3 H C H C CH + 3 3 3 S S S 4 FeCl3 CH Cl 2 2 S S 36% S S S CH3NO2 S S

H3C CH3

H3C H3C CH3 S S S

CHCl3 S S S 5% S S S

S S S

H3C CH3 CH3

H3C S H3C H3C CH3 CH3 S S S S CHCl S 3 S S S S S S S S 41% + 4 FeCl S S S S 3 H C H C CH CH 3 3 3 3

Scheme 2.6. Regioselective oligomerization of 3-(alkylthio)-thiophenes with iron(III)chloride.

Very recently, following the same approach, but using iron(III) perchlorate as oxidative agent Otsubo et al. succeeded in the synthesis and characterization of extraordinarily long oligothiophenes up to a 96-mer, which has an estimated chain length of 37.2 nm.40 To provide sufficient solubility and to avoid regio-irregularity or undesirable reactions at ß- positions that become eminent for long oligothiophenes, both β-sites of the thiophenes were blocked by introducing 2,2-bis(butoxymethyl)-1,3-propanediyl groups. The oxidative homo-coupling of the hexamer 6T afforded a mixture of the 12-mer, 18-mer and trace amount of 24-mer which have been separated by gel-permeation liquid chromatography. Starting with the 12-mer the same reaction sequence led to the 24-mer, the 36-mer and the 48-mer. The repeated oxidation of 24-mer after separation of the oligomers successfully gave the 48-mer, the 72-mer and the 96-mer (Scheme 2.7).

18 Chapter 2 Synthesis of π-conjugated oligothiophenes

OBu OBu OBu OBu OBu OBu

S S S H H S S S n

OBu OBu OBu OBu OBu OBu

n=1 (6T), n=2 (12T), n=3 (18T), n=4 (24T)

n=6 (36T), n=48T), n=12 (72T), n=16 (96T)

Fe(ClO ) Fe(ClO ) 4 3 4 3 Fe(ClO4)3 6T 12T [60%] 24T [42%] 48T [18%] + + + 18T [3%] 36T [4%] 72T [9%] + + + 24T [0.3%] 48T [2%] 96T [7%]

Scheme 2.7: Series of β-blocked long oligothiophenes up to the 96-mer.

An important aspect in studying long conjugated oligothiophenes is certainly to establish a concept of the effective conjugation length. It is well known that oligothiophenes show a strong π-π* electron absorption band in the visible region which is progressively red- shifted with increasing chain length. In the series of short oligomers a good linear correlation between the transition energy and the inverse ring number (1/n) of the oligothiophenes can be observed, whereas a deviation from linearity appears for the long oligomers due to the occurrence of saturation. The effective conjugation length is defined as the point when saturation of the red-shift is reached. Supported by spectroscopic studies Wynberg et al. suggested that the effective conjugation length of polythiophene is not longer than 11 repeating units.41 Afterwards, it was observed that even for a longer 15- mer27g and 16-mer27f no saturation of the red-shift of the absorption has been reached. Later studies based on electronic absorption and emission spectra of a series of alkyl- substituted oligothiophenes showed that an effective conjugation length along the one dimensional chain extents to 20 repeating units.27a More interesting is that spectrocopic measurements on long, highly conjugated oligothiophenes series reported by Otsubo et al. evidently show no convergent limit for the conjugation up to the 96-mer. Furthermore, the longest wavelength absorption of the 96-mer (556.5 nm in solution) is more red-shifted

2.2 π-Conjugated oligothiophenes 19 than that reported for the regioregular poly(alkylthiophene)s (438-450 nm in solution and 500-526 nm in solid state).42 This indicates by neglecting the substituents effects that the conjugation length of a polythiophenes is shorter than that of the oligothiophene with 96 repeat units.

2.2.2.2 Transition metal promoted aryl-aryl bond formation

More recently, very effective cross-coupling reactions for the C-C bond formation have been developed allowing the effective preparation of oligothiophenes of different chain lengths. These cross-coupling reactions of an organometallic reagent R-M with an organic compound R´X (X is a leaving group) yielding R-R´ under C-C bond formation, are generally catalyzed by a complex of a low-valent transition metal, mainly, nickel or palladium (Scheme 2.8).43

Ni or Pd R' M + RX R R'

M = Li (Murahashi) M = Sn (Stille) Mg (Kumada, Corriu) Zr (Negishi) Zn (Negishi, Normant) Al (Negishi, Nozaki-Oshima) B (Suzuki) ......

Scheme 2.8. Transition metal catalyzed aryl-aryl coupling reactions.

From chronological point of view, the first example of a transition metal-catalyzed cross- coupling reaction employing a nickel catalyst was reported in 1972. In this case, as organometallic reactant a Grignard reagent was employed. Since then, many organometallic reagents of lithium, boron, aluminium, zirconium, zink, copper and tin were found to undergo with organic halides or tosylates this type of C-C bond formation. Further, palladium catalysts turned out to be more conveniently than nickel ones.

Kumada coupling reaction In 1972, Kumada44 and Corriu45 simultaneously described the rapid and efficient cross-

2 coupling of Grignard reagents with aryl and alkenyl Csp -halides catalyzed by complexes of

20 Chapter 2 Synthesis of π-conjugated oligothiophenes nickel (Scheme 2.9). These reactions allow the synthesis of a wide variety of unsaturated organic compounds from two different organic halides.

Mg L NiCl RX R MgX + R'X' 2 2 R R' + MgXX' Et O 2

Scheme 2.9. Kumada coupling reaction.

The group of Kumada further studied this reaction by using dihalo-bis(phosphine)-nickel complexes as the most effective catalyst. Later, he extended the scope of this nickel- catalyzed Grignard coupling to numerous examples of the arylation of halo-heterocyclic compounds. The reaction is selective, very efficient and applicable to various types of Grignard compounds (alkyl or aryl) and organic halides bearing a sp2-carbon (aryl or vinyl). The mechanism of the Kumada coupling reaction is depicted in Scheme 2.10.

L2NiX2 1 2 R-MgX 2 MgX2

L2NiR2

R'X' 2 R-R

5 R' L2Ni RMgX R-R' X' 3

MgXX'

R'X' R' L2Ni R' R L2Ni X'

4 R'X'

Scheme 2.10. Catalytic cycle of Kumada cross-coupling reaction.

2.2 π-Conjugated oligothiophenes 21

In a first step 1, the dihalonickel-phosphine L2NiX2 catalyst reacts with two equivalents of the Grignard reagent to form the intermediate L2NiR2. Reaction with an organic halide

R´X´ transforms L2NiR2 to its active form L2NiR´X´ under elimination of a homo-coupling product R-R (step 2). Due to this step the homo-coupling product can generally be found as undesired by-product of the reaction to an extent of around 0.5-1.5% depending on the amount of catalyst employed. In the following step 3, the halogen X´ is exchanged with the organic group R of another equivalent Grignard reagent leading to the nickel-complex

L2NiRR´ containing the two organic coupling components. Addition of a second organic halide R´X´ to this complex (step 4) leads to a five fold coordinated intermediate which subsequently eliminate the desired cross coupling product R-R´ under simultaneous formation of the active catalytic species L2NiR´X´ (step 5). The catalytic cycle is closed.

In the presence of catalytic amounts of Ni(dppp)Cl2 bromothiophenes effectively react with aryl or alkyl Grignard reagents to give cross-coupling products.46 A limitation of the applicability of the reaction occurs when the organic compound does not effectively add to the Grignard reagent or the halide does not give efficient halogen-magnesium exchange. These lead to a significant formation of undesired homo-coupling by-products. In oligothiophene chemistry, mostly, slow halogen-magnesium exchange arises when the haloaryl contains more than two thiophene rings. Since Grignard reagents from a bromoterthiophene is no longer possible to prepare, the synthesis of longer oligothiophenes by Kumada coupling is restricted to the stepwise approach in which the length of the oligomer chain increases by a mono- or bithiophene unit. However, a very impressive example in the application of this method was the synthesis of a complete series of end-capped oligothiophenes up to the heptamer by repeated Kumada coupling reaction (Scheme 2.11).47 The cyclohexene-caps were introduced, on the one hand to block the reactive α-positions of the oligothiophenes when they are transformed into cationic species and thus to allow a more precise characterization of the oligomers in various oxidation states, and on the other hand, to enhance solubility of the compounds. With this approach, excellent correlations between the spectroscopic and electrochemical data and chain lengths of the oligomers were obtained. The selective bromination with NBS of the key building block 4,5,6,7-tetrahydrobenzo-[b]thiophene EC1T led to the monobromo derivative which could then easily be converted to the Grignard reagent. The

Ni(dppp)Cl2-catalayzed coupling of the Grignard reagent with itself (1), dibromo derivatives of monothiophene (2), bithiophene (3) and terthiophene (4), resulted in the formation of EC2T, EC3T, EC4T, and EC5T, respectively. The higher homologues were

22 Chapter 2 Synthesis of π-conjugated oligothiophenes prepared starting with a mono-capped bithiophene which was obtained by cross-coupling of the monobromo derivative of EC1T with the Grignard of monobromothiophene. Following, the mono-capped bithiophene was then converted into the Grignard reagent and reacted under the same reaction sequence with the dibromo derivatives of monothiophene (5), bithiophene (6), terthiophene (7) to give EC5T, EC6T and EC7T, respectively.

S S S EC1T EC2T [47%] 1) NBS/DMF S 2) S S Br S 3) EC3T [64%]

4) S S MgBr S S S Ni(dppp)Cl2 EC4T [78%]

S S S S S S S EC5T NBS/DMF 5) [64%]

6) S S S S S S Br S S 7) EC6T [58%]

S S S S S S S EC7T [44%] Scheme 2.11. Synthesis of the end-capped oligothiophene series by Kumada coupling reaction: 1) a)

Mg/Et2O; b) Ni(dppp)Cl2; 2) a) Mg/Et2O; b) 2,5-dibromothiophene, Ni(dppp)Cl2; 3) Mg/Et2O; b) 2,5- dibromobithiophene, Ni(dppp)Cl2; 4) a) Mg/Et2O; b) 2,5-dibromoterthiophene, Ni(dppp)Cl2; 5) a)

Mg/Et2O/benzene; b) 2,5-dibromothiophene, Ni(dppp)Cl2; 6) a) Mg/Et2O/benzene; b) 2,5- dibromobithiophene, Ni(dppp)Cl2; 7) a) Mg/Et2O/benzene; b) 2,5-dibromoterthiophene, Ni(dppp)Cl2.

One of the critical limitations of the Kumada coupling reaction is its low tolerance to functional groups, such as amino-, nitrile-, ester- or carbonyl-groups. In order to overcome this problem and to be able to apply aryl-aryl coupling reactions to compounds with a wide variety of functional groups, less reactive and less nucleophilic metal-organyls from tin-, boron-, or zincorganyls, with appropiate palladium catalyst were used.

2.2 π-Conjugated oligothiophenes 23

A general palladium-promoted catalytic cycle for the cross-coupling reaction is depicted in Scheme 2.12.

Pd(II)

R'M

R'-R' + MX

Pd(0) RX R-R' 1 3 oxidative addition reductive elimination

R-Pd(II)-R' R-Pd(II)-X B A

2 transmetallation

MX R'M M= B, Sn, Si, Zn, Mg X= I, OTf, Br, Cl R= 1-alkenyl, 1-alkynyl, allyl, benzyl, aryl

Scheme 2.12. A general palladium-promoted catalytic cycle for cross-coupling reactions.

The mechanism is based on three reactions: oxidative addition (1), transmetallation (2) and reductive elimination (3). Although each step is individually influenced by various factors, such as ligand exchanges or addition of specific ligands, there is no doubt about the presence of the intermediates (A and B) which have been isolated and also characterized by spectroscopic analysis.48 A very wide range of palladium(0) catalysts or precursors can be used for the cross coupling reactions. Pd(PPh3)4 is the most commonly used catalysts, but Pd(Ph3)2Cl2, Pd(dppp)Cl2, Pd(dba)3, and Pd(OAc)2 in the presence of PPh3 or other phosphine ligands are also very efficient. They are air stable and readily reduced to the active species by the metal-organyl or additional phosphines. Palladium complexes that contain fewer than four phosphine ligands or bulky ligands are, in general, highly reactive due to the easy formation of coordinatively unsaturated palladium compounds. The oxidative addition leading to A is often the rate-determinating step in the catalytic cycle. The relative reactivity decreases in the order of I > OTf > Br > Cl. The reductive elimination of the coupling product from B reproduces the palladium(0) complex. The

24 Chapter 2 Synthesis of π-conjugated oligothiophenes elimination reaction takes directly place from the cis-conformer of B, while the trans- conformer must first isomerise to the corresponding cis-conformer prior to reductive elimination. The order of reactivity of B is diaryl- > alkyl-aryl- > dialkyl, suggesting participation of the π-orbitals by aryl groups during the bond formation.49 Although the mechanism of the oxidative addition and reductive elimination sequences are reasonably well understood, less is known about the transmetallation step because the mechanism is highly dependent on the organometallics or the reaction conditions used for the coupling reaction.

Stille coupling reaction The palladium-catalyzed coupling of organotin compounds, known as the Stille reaction, is nowadays in widespread use (Scheme 2.13).43,50 This is due to the growing availability of organostannanes, their stability towards moisture and air, and their compatibility with a large variety of functional groups.

R' SnR'' R'' SnX RX+ 3 R R' + 3

Scheme 2.13. Stille cross-coupling reaction

The organotin compounds R´SnR3´´ can be synthesized from the corresponding lithiated derivatives R´Li by a reaction with trialkylstannylchloride R3´´SnCl or by reacting organic halides R´X with hexaalkyldistannane R´´3Sn-SnR´´3. The organic component R´ can be alkynyl, alkenyl, aryl, allyl or benzyl and R´´ is, usually, methyl or butyl which are typically not transferred. In the coupling reaction as organic halides RX generally aryliodides, -triflates, -bromides, but in the meanwhile also –chlorides are used. The disadvantage of this reaction is the toxicity of the organotin reagents. The catalytic cycle for the Stille reaction is basically similar to that depicted in Scheme 2.12. With respect to oligothiophene synthesis the Stille reaction has mostly found application in the synthesis of mixed thiophene/heteroarene compounds51 or substituted oligothiophenes,52 especially, of 3,4-ethylendioxythiophenes (EDOT) as building blocks for poly-(EDOT)s. These polymers show excellent environmental and chemical stability combined with low oxidation potentials and moderate band gaps.53 Recently, Roncali et al. succeeded in the synthesis of a new series of stable and soluble oligo-EDOTs end-capped

2.2 π-Conjugated oligothiophenes 25 by n-hexyl groups, by carrying out Stille coupling reaction.54 The synthesis of the trimer and tetramer was achieved by coupling of the stannic derivative of the mono-EDOT and dibromo derivatives of EDOT and bis-EDOT as shown in Scheme 2.14. As expected, the association of the electronic and structural effects of the ethylendioxy groups leads to a self-rigidified π-conjugated system with lower energy band-gap than that for the corresponding oligothiophenes analogues.

O O O O BuLi H C 13 6 S Bu SnCl SnBu 3 H13C6 S 3

O O O O S Br Br S Br S Br O O Pd(PPh3)4 Pd(PPh3)4

O O O O O O O O

S S S C H C H H C 6 13 H13C6 S S 6 13 13 6 S S

O O [55%] [45%] O O O O

Scheme 2.14. Synthesis of n-hexyl end-capped oligo-EDOTs by Stille coupling reaction.

Suzuki coupling reaction In the early 1980s, Suzuki initiated a major improvement in the use of palladium-catalyzed aryl-aryl bond formation by using boronic acids as the nucleophilic coupling partner.55 This cross-coupling reaction which uses the less electropositive boron proceeds with good to excellent yields even in sterically demanding positions and also tolerates a wide range of functional groups. Another advantage is the low toxicity of the reagents especially compared to tin containing compounds. Moreover, even if they are not always easy to prepare, boronic acid derivatives are easy to handle and present much less manipulation risk than other metal-organyl derivatives. In the meanwhile, this coupling reaction is one of the most frequently used methods for C-C bond formation and the major improvement of the last two decades in this area have recently been reported in several reviews.43,48a

26 Chapter 2 Synthesis of π-conjugated oligothiophenes

High yields, fair selectivity and large scope of applications have made the Suzuki reaction a method of choice also for solid phase synthesis.43a,56 In the synthesis of oligothiophenes Kirschbaum et al. impressively demonstrated the distinct advantages that arise by using this coupling method in combination with the solid phase technique.27h A complete series of regioregular head-to-tail coupled oligo(3-alkylthiophene)s up a the dodecamer were synthesized on polymer support (Scheme 2.15). In this approach at first, a bithiophene carboxylic acid was anchored to chloromethylated polystyrene. For the oligomer chain elongation bithiophene building blocks were used. Iterative sequences of iodination and Suzuki cross-coupling reactions afforded in a few steps a series of the resin-bound oligomers in excellent yields and purities. The oligomers were cleaved from the resin by a transesterification reaction and then treated with iodomethane to afford α-carboxy- substituted oligomers. The dodecamer was obtained in 12% overall yield after 12 steps. The following saponification and decarboxylation of the oligomers yielded the non- functionalized regioregular oligo(3-alkylthiophenes).

H13C6

O S X S n O H13C6 C6H12 O S B X = H S Hg(OCOC5H11)2 O C H X = I I2 6 12 1. Bu NOH 4 Pd(PPh ) / CsF / THF 2. MeI 3 4

H C H C 13 6 1. Bu4NOH 13 6 2. HCl S H S H MeOOC S n H S n 3. Cu / quinoline

C6H12 C6H12

n = 1-6

Scheme 2.15. Synthesis of regioregular head-to-tail coupled oligo(3-alkylthiophenes) on solid support.

The use of zinc derivatives in palladium-catalyzed aryl-aryl coupling reactions has been studied by Negishi since 1982.57 Since organozinc derivatives are mostly used for the preparation of functionalized oligothiophenes or mixed thiophene/heteroarene compounds, such as furans, pyrrols, pyridins or pyrimidine, the reaction is not considered in detail in this work. 58

2.2 π-Conjugated oligothiophenes 27

2.2.2.3 Ring closure reactions from acyclic precursors

Beside organometallic aryl-aryl coupling reactions another synthetic strategy to prepare various oligothiophenes is the synthesis including ring closure reactions from acyclic precursors.59 1,4-Diketones can easily be converted to thiophene rings by reaction with sulfide-donor reagents. In general, hydrogen sulfide and hydrochloric acid, phosphorus(V) sulfide or Lawesson’s reagent is used (Scheme 2.16).60 Another effective method to prepare oligothiophenes from acyclic precursors is the ring closure reaction of 1,3- butadiynes with sulfide nucleophiles. In the following, this reaction is discussed in more detail.

H S/HCl or 2 S S S S S O O P2S5 or L.R.

S S MeO P P OMe S S

Lawesson's reagent L.R.

Scheme 2.16. Ring closure reactions of 1,4-diketones to thiophenes.

Ring closure reaction of 1,3-butadiynes This reaction was first studied by Bohlmann et al.61 in the field of natural diacetylenes and later successfully applied in the synthesis of oligothiophenes by Kagan et al.31 A homologues series of unsubstituted α-linked oligothiophenes were prepared by reacting the corresponding 1,3-butadiynes with sodium sulfide in boiling methanol (Scheme 2.17).

Na2S * 9H2O n H H S n S methanol S 2n+1

n = 1-3 [84%, 74%, 98%]

Scheme 2.17. Synthesis of oligothiophenes by sulfide addition a diacetylenes.

28 Chapter 2 Synthesis of π-conjugated oligothiophenes

The reaction can be carried out in different solvents and also by using other sulfide nucleophiles such as hydrogen sulfide or sodium hydrosulfide.27a,b,h,62 It is noteworthy, that if the thiophene rings are substituted in the positions next to the butadiyne units, higher reaction temperature is needed and therefore, the reaction has to be carried out in higher boiling solvents.62a This is probably due to the steric hindrance caused by the alkyl groups. No detailed mechanistical studies can be found in the literature for this reaction, but common for all reported examples is that basic reaction conditions are required. Generally, potassium hydroxide in an alcoholic solvent is employed as additive. The first crucial step in this approach is the synthesis of the precursor diacetylenes. Symmetrical diacetylenes are mostly synthesized by sp-sp carbon coupling processes. The long history of acetylenic coupling dates back to 1869 when Glaser observed that copper(I) phenylacetylide exposed to air underwent oxidative dimerization to 1,4-diphenyl-butadiyne (Scheme 2.18).63

2CuCl O 2 H Cu 2 NH OH/EtOH NH OH/EtOH 4 4

Scheme 2.18. First reported acetylenic coupling.

The reaction starts with the abstraction of the acetylenic proton by the base to form the copper(I) acetylide which then by intramolecular oxidative coupling yields the corresponding symmetrical butadiyne. In the following years, it was shown that dioxygen itself is not necessary but other oxidizing agents including potassium hexacyanoferrate, copper(II) salts, potassium permanganate and peroxides can also be used.64 The breakthrough for the wide applicability of this reaction was finally the discovery that the potentially explosive copper acetylides can also be generated in situ and no isolation is necessary.65 In the next years several studies concerning the influence of the reaction parameters (temperature, solvent, catalyst, pH) on the coupling of differentially substituted acetylides have been performed. A detailed summary review of the results and development in this field has been then published in 1963 by Eglinton and McCrae66 and six years later by Cadiot and Chodkiewitz.67 Nowadays, the Glaser coupling and several variations of it is the most used method for the preparation of linear or cyclic oligo- and polyynes with wide applications in all domains of chemistry.68

2.2 π-Conjugated oligothiophenes 29

One important modification of the Glaser coupling was reported by Eglinton in 1956.69 In this approach oxidative coupling is performed in a methanol/pyridine mixture in the presence of excess copper(II) salts, mostly copper(II) acetate. Following, this method turned out to be one of the most effective methods for the synthesis of various unsaturated or conjugated macrocycles, culminating with the application in the synthesis of annulenes by Sondheimer et al.70 This approach found also utility in the synthesis of different substituted linear di- or polyacetylenic compounds.

C8H17 S S H H S S 2

H17C8

C8H17 S S H H S S 2

H17C8

Cu(OAc)2 pyridine

C8H17 H17C8 S S S S H H S S 2 S S 2 n

H17C8 C8H17 n = 1-11

Scheme 2.19. Synthesis of oligo(octithienylene-diacetylene)s by random Eglinton coupling by Otsubo et al.

As illustrated in Scheme 2.19, Otsubo et al. succeded in the preparation of long oligomers of (oligothienyl-diacetylene)s among mono- and diethynyl-oligothiophenes by using a random Eglinton oxidative coupling.27a,b The mixtures obtained from such a random reaction were successfully separated into the individual homologues by preparative gel- permeation chromatography. By a similar protocol, very recently, they succeeded in the preparation of a series of oligo(octithienylene-diacetylene)s (Scheme 2.19). The largest compound isolated comprised twelve octithiophene- and eleven diacetylene units. The molecular length of this compound would reach 43 nm in a stretched conformation, which is the longest among single component conjugated nanomolecules.71

30 Chapter 2 Synthesis of π-conjugated oligothiophenes

A further and also very important modification of Glaser coupling which finds applications in the synthesis of linear di- or polyacetylenes was reported by Hay in 1962.72 In this approach, oxygen gas is passed through a solution of the alkyne and a catalytic amount of copper(I) salt in a complex forming solvent such as pyridine. Mostly, a catalytic amount of chelating ligand N,N,N’,N’-tetramethylenediamine (TMEDA) is also added with the purpose to increase the solubility of the active intermediate. Krömer et al. successfully used this method for the dimerization of ethynylated thiophenes to a homologues series of bis(oligothienyl)-1,3-butadiynes which were then reacted with sodium sulfide and transformed to the corresponding α-linked conjugated oligothiophenes up to the 11-mer (Scheme 2.20).27h

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu CuCl / TMEDA / O2 S S S n CH Cl n n S S H 2 2 S S S S n = 0-2 n = 0-2

Na2S x 9H2O methoxyethanol

Bu Bu Bu Bu Bu Bu Bu Bu

S S S S n S S S n

n = 0-2

Scheme 2.20. Synthesis of the homologues series of bis(oligothienyl)-1,3-butadiynes and the corresponding oligothiophenes

In further studies following the same Hay type coupling, but as a random process between different mono- and diethynyl-oligothiophenes, they could extend this linear homologous series of conjugated oligothiophenes in which the chain length only varies by bithiophene units, up to the 19-mer.27d Systematic studies concerning the relationship between the electronic and photonic properties of these β-substituted oligothiophenes and the chain and conjugation length have been performed and the theoretically predicted behaviour was also proven experimentally. The discovery of palladium-catalyzed coupling reactions three decades ago73 prompted a very fast development in the field of organometallic chemistry. In these days, it was observed that by coupling reaction of terminal alkynes and aryl- or vinyl halides symmetrical diacetylenes via dialkynyl-palladium intermediates were formed as

2.2 π-Conjugated oligothiophenes 31 byproducts.74 In 1985, Rossi et al. optimized this process to an effective homo-coupling reaction for aromatic terminal alkynes.75 Using palladium(0) and copper iodide as catalyst system, chloroacetone as oxidant to regenerate the palladium(II) catalyst from the palladium(0) formed in the catalytic cycle and triethylamine as the base, aromatic 1,4- diaryl-1,3-butadiyne have been obtained in very good yields under very mild reaction conditions (Scheme 2.21).

Pd(PPh3)4 / CuI / NEt3 Ar H Ar Ar chloroacetone

Ar = [94%]

Ar = [87%] S

Scheme 2.21. Palladium-catalyzed homo-coupling reaction of aromatic alkynes.

A modification of this reaction was reported by Liu and Burton.76 They used the more air stable catalyst Pd(PPh3)Cl2, diisopropyl amine instead of triethyl amine and iodine as oxidizing agent and extended this method to the synthesis of aliphatic diacetylenes. Generally, terminal aliphatic acetylenes are sluggish in undergoing Glaser dimerization due to the weaker acidity of the acetylinic proton. The proposed mechanism for the palladium-promoted alkynes homo-coupling reaction is illustrated in Scheme 2.22.

32 Chapter 2 Synthesis of π-conjugated oligothiophenes

2 R H 2 NR3HX 2 NR3 R CuX L Pd L2PdX2 2 1 R

2 R R

X L Pd 3 2 5 2

R X L Pd 2 L2Pd X R

CuX

2 R H 2 NR HX 3 2 NR 3

Scheme 2.22. Proposed reaction mechanism for the Pd-catalyzed oxidative homo-coupling of alkynes.

No intensive studies were carried out concerning the stepwise mechanism for the formation of dialkynes. Obviously, the key intermediate is a dialkynylpalladium complex which is derived in step 1 by a two fold transmetallation reaction from the terminal alkyne L2Pd(II) complex in the presence of copper iodide and amine. In the following step 2, under reductive elimination of the homo-coupling product R−≡−≡−R the active L2Pd(0) species is formed which then is regenerated by an oxidative addition of X2 to a palladium(II) complex (step 3). Recurrent formation of the dialkynylpalladium intermediate by a two fold transmetallation reaction (step 4) and subsequent reductive elimination (step 5) to the desired homo-coupling product R−≡−≡−R and active L2Pd(0) complex convert all the terminal alkynes to the corresponding dialkynes. It is important to note that as in the case of the palladium cross-coupling reactions, the reductive elimination takes directly place only from a cis-dialkynylpalladium intermediate. A trans-dialkynylpalladium complex before elimination first has to undergo an isomerization to the corresponding cis-complex. Although it was proven that this reaction represents a good alternative to the copper- promoted Glaser coupling, as it tolerates a variety of functional groups such as alkenyl,

2.2 π-Conjugated oligothiophenes 33 nitrile or hydroxyl, there are only a few reports on using this palladium sp-sp coupling reaction.68a,77 Nevertheless, this method was used also for the synthesis of long conjugated oligomers. Godt et al.78 reported the quantitative dimerization of monoethynylated oligo(p- phenylene-ethynylene)s by using Pd(PPh3)Cl2, CuI, piperidine in THF, while with the Hay method the reaction failed. Interestingly, in this case no oxidative reagent has been added because carrying out the reaction in air was enough (Scheme 2.23).

Hex

THPO H n Hex n = 1,2

Pd(PPh3)Cl2 / CuI piperidine / THF

Hex Hex

THPO OTHP n n Hex Hex

n = 1,2 [90%,96%]

Scheme 2.23. Synthesis of oligo(p-phenylene-ethynylene)s by palladium-catalyzed alkynyl dimerization

Unsymmetrical oligo- and polyynes are generally prepared by the Cadiot-Chodwietz reaction.67,79 In this heterocoupling reaction a terminal acetylene reacts with an acetylic halide in the presence of copper(I) salt and a suitable amine. Because this method tolerates numerous functionalized starting materials including alcohols, epoxides, amines, acetals, carboxyl derivatives, disulfides, etc. it has found broad applications for the synthesis of many polyunsaturated systems.80 In all above mentioned sp-sp coupling methods ethynyl-substituted oligothiophenes are necessary to prepare bis(oligothienyl)-1,3-butadiynes as precursors for conjugated oligothiophenes. Introduction of an ethinyl group to the oligothiophene core can be performed by various methods. In the Corey-Fuchs approach,81 first an aldehyde functionality of an oligothiophene is converted by a Wittig reaction to the 1,1- dibromoethylene which upon two successive hydrobromic acid elimination yields the desired ethynyl-oligothiophene (Scheme 2.24).

34 Chapter 2 Synthesis of π-conjugated oligothiophenes

Br CBr 4 n-BuLi H CHO Br H S n PPh H S n S n 3 H

Scheme 2.24. Synthesis of ethinyl-oligothiophenes by Corey-Fuchs approach.

A much more effective method to introduce ethynyl groups into the thiophene core is the cross-coupling reaction developed by Sonogashira74 in 1975. This palladium-catalyzed coupling reaction of aryl or alkenyl halides with alkynes proceeds smoothly under mild conditions in various amines as solvents and in the presence of a co-catalyst such as copper iodide. A broad functional group tolerance is characteristic for this method (Scheme 2.25).

Pd(PPh3)Cl2 / CuI RX + H R' R R' amine

R= aryl, alkenyl X= Cl, Br, I, OTf

Scheme 2.25. Sonogashira cross-coupling reaction.

As shown in Scheme 2.26, this protocol relies on the discovery of CuI transmetallation in 82 amine. As palladium source, Pd(PPh3)Cl2 in amine is commonly used which in step 1 reacts with two equivalents of a terminal acetylene forming a dialkenylpalladium complex. Upon reductive elimination (step 2) of a homo-coupling product from this complex the catalytically active and coordinatively unsaturated 14-electron complex, Pd(PPh3)2 is produced. The following steps are similar to the other palladium-catalyzed cross-coupling reactions. Through an oxidative addition (step 3) of the organic halide to the Pd(PPh3)2 active catalyst an organopalladium intermediate is formed. By a new transmetallation reaction (step 4) the diorganyl palladium containing the two coupling components is formed which then undergoes a spontaneous reductive elimination (step 5) to yield the desired coupling product and the active Pd(PPh3)2 catalyst.

2.2 π-Conjugated oligothiophenes 35

2 R' H 2 NR3HX 2 NR3 R'

CuX (Ph3P)2PdCl2 (Ph3P)2Pd 1 R' 2

R' R'

RX R R' 5 (Ph3P)2Pd

R R (Ph P) Pd Pd(Ph3)2 3 2 X R'

CuX

R' H NR HX 3 NR 3 Scheme 2.26. Palladium-catalyzed cross-coupling reaction of terminal acetylenes with C-sp2 halides.

Various conditions have been employed for this reaction, depending on the reactivity of the halide, the alkynes and the base used. Nowadays, this method is used widely for the construction of conjugated aryl-alkynes or enyne systems in diverse areas ranging from natural product chemistry to materials science.68b,c,80,83

2.3 References and note

1 Nobel Prize in Chemistry 2000 was awarded to A.J. Heeger, A.G. MacDiarmid, H. Shirakawa by The Royal Swedish Academy of Sciences.

2 C.K. Chiang, C.R. Fincher, Y.W. Park, H. Shirakawa, E.J. Louis, S.C. Gau, A.G. MacDiarmid, Phys. Rev. Lett. 1977, 39, 1098-1101.

3 a) H. Shirakawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang, A.J. Heeger, J. Chem. Soc., Chem. Commun. 1977, 578-580; b) C.K. Chiang, Y.W. Park, A.J. Heeger, H. Shirakawa, E.J. Louis, A.G. McDiarmid, J. Chem. Phys. 1978, 69, 5098-5104.

36 Chapter 2: Synthesis of π-conjugated oligothiophenes

4 D.M. Ivory, G.G. Miller, J.M. Sowa, L.W. Shacklette, R.R. Chance, R.H. Baughman, J. Chem. Phys. 1979, 71, 1506-1507.

5 G.E. Wnek, J.C.W. Chien, F.E. Karasz, C.P. Lillja, Polymer 1979, 20, 1441-1443.

6 A.F. Diaz, J.A. Logan, J. Electroanal. Chem. 1980, 111, 111-114.

7 A.F. Diaz, J.I. Castillo, J. Chem. Soc., Chem. Commun. 1980, 397-398.

8 G. Tourillon, F. Garnier, J. Electroanal. Chem. 1982, 135, 173-178.

9 Handbook of Conducting Polymers, 2nd ed., Eds.: T.A. Skotheim, R.L. Elsenbaumer, J.R. Reynolds, Marcel Dekker, New York, 1998.

10 J.H. Burroughes, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burn, A.B. Holmes, Nature, 1990, 347, 539-541.

11 a) H. Meier, Angew. Chem. Int. Ed. Engl., 1992, 31, 1399-1420, b) Organic Materials for Photonics, Eds.: G. Zerbi, Amsterdam, 1993; c) Introduction to Molecular Electronics, Eds.: M.C. Petty, M.R. Bryce, D. Bloor, Edward Arnold, London, 1995; d) A.J. Heeger, J. Jr. Long, Opt. Photonics News 1996, 7, 24-30; e) A Kraft, A.C. Grimsdale, A.B. Holmes, Angew. Chem. Int. Ed. Engl., 1998, 37, 402-428; f) R.W. Gymer, R.H. Friend, A.B. Holmes, E.G,J. Staring, R.N. Marks, C.Taliani, D.D.C. Bradley, D.A. dos Santos, J.L. Bredas, M. Lögdlund, W.R. Salaneck, Nature 1999, 397, 121-128; g) Semiconducting Polymers, Eds.: G. Hadziioannou, P.F. van Hutten, Wiley-VCH, Weinheim, Germany, 2000.

12 a) J.M. Tour, Chem. Rev. 1996, 96, 537-553; b) Handbook of Organic Conductive Molecules and Polymers, Eds.: H.S. Nalwa, John Wiley, Chichester, 1997; c) Electronic Materials: The Oligomer Approach, Eds.: K. Müllen, G. Wegner, Wiley-VCH, Weinheim, Germany, 1998; d) R.E. Martin, F. Diederich, Angew. Chem. Int. Ed. 1999, 38, 1350-1377; e) J. Roncali, Acc. Chem. Res. 2000, 33, 147-156.

13 a) G. Horowitz, J. Mater. Chem. 1999, 9, 239-241; b) J.L. Segura, N. Martin, J. Mater. Chem. 2000, 10, 2403-2435; c) J.L. Segura, N. Martin, Chem. Soc. Rev. 2000, 29, 13-25; d) D.M. Guldi, M. Prato, Acc. Chem. Res. 2000, 33, 695-703; e) D. Gust, T.A. Moore, A.L. Moore, Acc. Chem. Res. 2001, 34, 40-48; f) D. Holten, D.F. Bocian, J.S Lindsey, Acc. Chem. Res. 2002, 35, 57-69.

14 a) Molecular Electronics: Science and Technology, Eds.: A. Aviram, M. Ratner, New York, Academy of Sciences, New York, 1998; b) R.M. Metzger, Acc. Chem. Res. 1999, 32, 950-957; c) C. Joachim, J.K. Gimsewski, A. Aviram, Nature 2000, 408, 541-548; d) J.M. Tour, Acc. Chem. Res. 2000, 33, 791-804.

15 J. Anthony, C. Boudon, F. Diederich, J.-P. Gisselbrecht, V. Gramlich, M. Gross, M. Hobi, P. Seiler, Angew. Chem. Int. Ed. Engl. 1994, 33, 763-766.

16 R.B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149-2154.

17 J.K. Young, J.C. Nelson, J.S. Moore, J. Am. Chem. Soc. 1994, 116, 10841-10842.

18 G. Wegner, Thermoplastic Elastomers, A Comprehensive Review, Eds.: N.R Legge, G. Holden, H.E. Schroeder, New York, 1987.

19 D.L.Pearson, J.M. Tour, J. Org. Chem. 1997, 62, 1376-1387.

20 a) P. Bäuerle in Oligothiophenes in Electronic Materials: The Oligomer Approach, Eds.: K. Müllen, G. Wegner, Wiley-VCH, Weinheim, Germany, 1998, pp. 105-197; b) Handbook of Oligo- and Polythiophenes, Eds.: D. Fichou, Wiley-VCH, Weinheim, Germany, 1999.

21 a) H. Sirringhaus, R.H. Friend, X.C. Li, S.C. Morati, A.B. Holmes N. Feeder, Appl. Phys.Lett. 1997, 71, 3871-3873. b) F. Garnier in Field- Effect Transistors Based on Conjugated Materials in Electronic

2.2 References and notes 37

Materials: The Oligomer Approach (Eds.: K. Müllen, G. Wegner,) Wiley-VCH, Weinheim, Germany, 1998, pp. 559-584; c) G. Horowitz, Adv, Mater. 1998, 10, 365-370; d) Z. Bao, Adv. Mater. 2000, 12, 227-230; e) H.E. Katz, Z. Bao, S.L. Gillat, Acc. Chem. Res. 2001, 34, 359-369; f) T. Otsubo, Y. Aso, K. Takimiya, Bull. Chem. Soc. Jpn. 2001, 74, 1789-1801; g) C.D. Dimitrakopoulus, P.R.L. Malenfant, Adv. Mater. 2002, 14, 99- 117.

22 a) U. Mitschke, P. Bäuerle, J. Mater. Chem. 2000, 10, 1471-1507; b) Y. Shirota, J. Mater. Chem. 2000, 10, 1-25.

23 a) N. Noma, T. Tsuzuki, Y. Shirota, Adv. Mater. 1995, 7, 647-648.

24 T. Otsubo, Y. Aso, K. Takimiya, J. Mater. Chem. 2002, 12, 2565-2575.

25 a) J. Nakayama, T. Konishi, M. Hoshino, Heterocycles 1987, 26, 1793-1796; b) Z. Fu, D. Fichou, F. Garnier, J. Electroanal. Chem. 1989, 267, 339-342.

26 G. Zotti, G. Schiavon, A. Berlin, G. Pagani, Chem. Mater. 1993, 5, 430-436.

27 a) T. Otsubo, Y. Aso, K. Takimiya, Bull. Chem. Soc. Jpn. 2001, 74, 1789-1801 (72-mer and 48-mer); b) H. Nakanishi, N. Sumi, Y. Aso, T. Otsubo, J. Org. Chem. 1998, 63, 8632-8633 (27-mer); c) A.H. Mustafa, M.K. Shepherd, Chem. Commun. 1998, 2743-2744 (24-mer); d) J. Krömer, Dissertation 2000, University Ulm, Germany (19-mer); e) P.R.L. Malenfrant, L. Groenendaal, J.M. Frechet, J. Am. Chem. Soc. 1998, 120, 10990-10991 (17-mer), f) P. Bäuerle, T. Fischer, B. Bildlingmeier, A. Stabel, J.P. Rabe, Angew. Chem. Int. Ed. Engl. 1995, 34, 303-307 (16-mer); g) M. Sato, M. Hiroi, Polymer 1996, 37, 1685-1689 (15-mer); h) T. Kirschbaum, C.A. Briehn, P. Bäuerle, J. Chem. Soc., Perkin Trans. 1 2000, 1211-1216 (12-mer); i) A. Yassar, D. Delabouglise, M. Hmyene, B. Nessak, G. Horowitz, F. Garnier, Adv. Mater. 1992, 4, 490-494 (12- mer); j) D.M. de Leeuw, Synth. Met. 1993, 57, 3597-3602 (12-mer); k) W. ten Hoeve, H. Wynberg, E.E Havinga, E.W. Meijer, J. Am. Chem. Soc. 1991, 113, 5887-5889 (11-mer); l) J. Krömer, P. Bäuerle, Tetrahedron 2001, 57, 3785-3794 (11-mer).

28 a) T. Kirschbaum, R. Azumi, E. Mena-Osteritz, P. Bäuerle, New. J. Chem. 1999, 241-250; b) R. Azumi, G. Götz, T. Debaerdemaeker, P. Bäuerle, Chem. Eur. J. 2000, 6, 735-744.

29 a) W. Steinkopf, W. Köhler, Lieb. Ann. Chem. 1936, 522, 17-31; b) W. Steinkopf, H.-J v. Petersdorf, R. Gording, Lieb. Ann. Chem. 1937, 527, 272-274; c) W. Steinkopf, R. Leitsmann, K.-H Hofmann, Lieb. Ann. Chem. 1941, 546, 180-184.

30 a) W. Steinkopf, J. Roch, Lieb. Ann. Chem. 1930, 482, 251-264; b) S. Gronowitz, Ark. Kemi 1960, 17, 89; b) T. Kauffmann, Angew. Chem. Int. Ed. Engl. 1974, 13, 291-305.

31 J. Kagan, S.K. Arora, Heterocycles 1983, 20, 1937-1940.

32 a) S. Gronowitz, B. Cederlund, A.-B. Hörnfeldt, Chem. Scr. 1974, 5, 217-226; b) T. Kaufmann, H. Lexy, Chem. Ber. 1981, 114, 3674-3683.

33 a) M. Sato, S. Tanaka, K. Kaeriyama, J. Chem. Soc., Chem. Commun. 1986, 873-874; b) K. Y. Jen, G.G Miller, R.L Elsenbaumer, J. Chem. Soc., Chem. Commun. 1986, 1346-1347.

34 M. Sato, S. Tanaka, K. Kaeriyama, Synth. Met. 1986, 14, 279-280.

35 M. Lemaire, R. Garreau, J. Roncali, F. Garnier, E. Hannecart, French Patent 8,604,744, 1986.

36 R.-I. Sugimoto, S. Takeda, H.B. Gu, K. Yoshino, Chem. Express, 1986, 1, 635-636.

37 a) M. Leclerc, F.M. Diaz, G. Wegner, Makromol. Chem. 1989, 190, 3105-3116; b) W.A Goedel, N.S Somanathan, V. Enkelmann, G. Wegner, Makromol. Chem. 1992, 193, 1195-1206.

38 Chapter 2: Synthesis of π-conjugated oligothiophenes

38 a) R.D. McCullogh, R.D. Lowe, M. Jayaraman, P.C Ewbank, D.L. Anderson, S. Tristran-Nagle, Synth. Met. 1993, 55, 1198-1199; b) T.A. Chen, R.D. Rieke, J. Am. Chem. Soc. 1992, 114, 10087-10088.

39 G. Barbarella, M. Zambianchi, R. Di Toro, M. Jr. Colonna, D. Iarossi, F. Goldoni, A. Bonini, J. Org. Chem. 1996, 61, 8285-8292.

40 T. Izumi, S. Kobashi, K. Takimiya, Y. Aso, T. Otsubo, J. Am. Chem. Soc. 2003, 125, 5286-5287.

41 E. E Havinga, I. Rotte, E.W. Meijer, W. ten Hoeve, H. Wynberg, Synth. Met. 1991, 41, 473-478.

42 R.D. McCullogh, R.D. Lowe, M. Jayaraman, D.L. Anderson, J. Org. Chem. 1993, 58, 904-912.

43 For recent reviews, see a) G.C Fu, A.F. Littke, Angew. Chem. Int. Ed. Engl. 2003, 22, 4176-4211; b) J. Hassan, M. Sevignon, C. Gozzi, E. Schulz, M. Lemaire, Chem. Rev. 2002, 102, 1359-1469; c) Metal- Catalyzed Cross-Coupling Reactions, Eds.: F. Diederich, P.J. Stang, Wiley-VCH, New York, 1998.

44 K. Tamao, K. Sumitani, M. Kumada, J. Am. Chem. Soc. 1972, 94, 4374-4376.

45 R.J.P. Corriu, J.P. Masse, J. Chem. Soc., Chem. Commun. 1972, 144-145.

46 a) K. Tamao, S. Komada, I Nakajima, M. Kumada, Tetrahedron, 1982, 38, 3347-3354; b) M. Kumada, Pure Appl. Chem. 1980, 52, 669-679.

47 a) P. Bäuerle, Adv. Mater. 1992, 4, 102-107; b). P.Bäuerle, Habilitationsschrift Universität Stuttgart 1994, Germany.

48 a) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457-2483; b) J.K. Stille, K.S.Y. Lau, Acc. Chem. Res. 1977, 10, 434-442.

49 a) F. Ozawa, A. Yamamoto, J. Chem. Soc. Jpn. 1987, 773-784; b) P.J. Stang, M.H. Kowalski, J. Am. Chem. Soc. 1989, 8, 180-188.

50 a) J.K. Stille, Angew. Chem. Int. Ed. Engl. 1986, 25, 508-524; b) T.N. Mitchell in Metal-Catalyzed Cross- Coupling Reactions, Eds.: F. Diederich, P.J. Stang, Wiley-VCH, New York, 1998, Chapter 4.

51 a) R. Wu, J.S Schumm, D.L. Pearson, J.M. Tour, J. Org. Chem. 1996, 61, 6906-6921; b) S.S Zhu, T.M. Swager, J. Am. Chem. Soc. 1997, 119, 12568-12577; c) A. Hucke, M.P. Cava, J. Org. Chem. 1998, 63, 7413- 7417; d) U. Mitschke, E. Mena-Osteritz, T. Debaerdemaeker, M. Sokolowski, P. Bäuerle, Chem. Eur. J. 1998, 4, 2211-2224; e) G. Zotti, S. Zecchin, G. Schiavon, S. Berlin, M. Penso, Chem. Mater. 1999, 9, 2123- 2125.

52 a) L.L. Miller, Y. Yu, J. Org. Chem. 1995, 60, 6813-6819; b) A. Fazio, B. Gabriele, G. Salerno, S. Destri, Tetrahedron 1999, 55, 485-502; c) P.R.L. Malenfant, J.M.J. Frechet, M. J. Jean, Chem. Commun. 2000, 23, 2657-2658; d) C. Eder, J.M.J Frechet, Org. Lett. 2003, 5, 1879-1882.

53 L. Groenendaal, F. Jonas, D. Freitag, H. Pielartzik, J.R. Reynolds, Adv. Mater. 2000, 12, 481-494.

54 M. Turbiez, P. Frere, J. Roncali, J. Org. Chem. 2003, 68, 5357-5380.

55 N. Miyaura, T. Yanagi, A. Suzuki, Synth. Commun. 1981, 11, 513; b) A. Suzuki, Acc. Chem. Res. 1982, 15, 178-184; c) A. Suzuki in Metal-Catalyzed Cross-Coupling Reactions, Eds.: F. Diederich, P.J. Stang, Wiley-VCH, New York, 1998, Chapter 2.

56 B.A. Lorsbach, M.J. Kurth, Chem. Rev. 1999, 99, 1549-1582.

57 a) E. Negishi, F.T. Luo, R. Frisbee, H. Matsuchita, Heterocycles 1982, 18, 117-122; b) T. Frejd, T. Klingstedt, Synthesis 1987, 1, 40-42; c) W. Li, T. Maddux, L. Yu, Macromolecules 1997, 29, 7329-7330;

2.2 References and notes 39

58 a) W.M. Albers, G.W. Canters, J. Reedijk, Tetrahedron 1995, 51, 3895-3904; b) F. Effenberger, F. Würthner, F. Steybe, J. Org. Chem. 1995, 60, 2082-2091; c) R.D. Rieke, S.-H. Kim, X. Wu, J. Org. Chem. 1997, 62, 6921-6927; d) M.B. Goldfinger, K.B. Crawford, T.M. Swager, J. Am. Chem. Soc. 1997, 119, 4578- 4593.

59 a) R. Hakansson in The Chemistry of Heterocyclic Compounds, Volume 44, Part 5: Thiophene and Its Derivates, Eds. S. Gronowitz, John Wiley, 1992; b) J. Nakajama, T. Konishi, M. Hoshino, Heterocycles 1988, 27, 1731-1733.

60 a) H. Wynberg, J. Metselaar, Synth. Commun. 1984, 14, 1-9; b) T. Asano, S. Ito, N. Saito, K. Hatakeda, Heterocycles 1977, 6, 317-320; c) A. Merz, F. Ellinger, Synthesis 1991, 462-464; d) W.ten Hoeve, H. Wynberg, E.E. Havinga, E.W Meijer, J. Am. Chem. Soc. 1991, 113, 5887-5889; e) F. Freeman, D.S.H.L. Kim, E. Rodriguez, J. Org. Chem, 57, 1722-1727.

61 F. Bohlmann, T. Burkhardt, C. Zdero, Naturally Occuring Acetylenes, Academic Press London 1973.

62 a) J.-P. Beny, S.N. Dhawan, J. Kagan, J. Org. Chem. 1982, 47, 2201-2204; b) M.G. Voronkov, B.A. Trofimov, V.V. Krjuchkov, S.V. Amasova, Y.M. Skortsov, A.N. Volkov, A.G. Mal’kina, R.Y. Mushii, Chem. Heterocycl. Comp. 1981, 17, 1249-1251; c) A.D. Dzhuraev, K.M. Karimkulov, A.G. Makhusov, A.G. Amanov, Pharm. Chem. J. (Engl. Transl.) 1992, 26, 882-884; d) S. Kozhushhov, T. Haumann, K. Boese, B. Knierim, S. Scheib, P. Bäuerle, A. de Meijere, Angew. Chem. Int. Ed. Engl. 1995, 34, 781-783; e) P. Kilickiran, Dissertation 2001, TU Braunschweig, Germany.

63 a) C. Glaser, Ber. Dtsch. Chem. Ges. 1869, 2, 422-424; b) C. Glaser, Ann. Chem. Pharm. 1870, 154, 137- 171.

64 a) A. Baeyer, Ber. Dtsch. Chem. Ges. 1882, 15, 57-61; b) A. A. Noyes, C.W. Trucker, Am. Chem. J. 1897, 19, 123-128; c) F. Straus, L. Kollek, Ber. Dtsch. Chem. Ges. 1926, 59, 1664-1681; d) M. Nahawa, Proc. Jpn. Acad. 1950, 26, 38-42; e) H.H. Schlubach, V. Wolf, Justus Liebigs Ann. Chem. 1950, 568, 141-159; f) N.A Milas, O.L. Mageli, J. Am. Chem. Soc. 1953, 75, 5970-5971; g) N.A Milas, O.L. Mageli, J. Am. Chem. Soc. 1953, 75, 5970-5971. h) A. Vaitiekunas, F.F. Nord, J. Am. Chem. Soc. 1954, 76, 2733-2736; i) F.J. Brockmann, Can. J. Chem. 1955, 33, 507-510; j) M.D. Cameron, G.E. Benett, J. Org. Chem. 1957, 22, 557- 558.

65 W. Jeppe, Justus Liebigs Ann. Chem. 1955, 596, 1-224.

66 G. Eglinton, W. McCrae, Adv. Org. Chem. 1963, 4, 225-328.

67 P. Cadiot, W. Chodkiewitz in Chemistry of Acetylene, Eds.: H.-G. Viehe, Marcel Dekker, New York, 1969, 597-647.

68 a) P. Siemsen, R.C. Livingston, F. Diederich, Angew. Chem. Int. Ed. Engl. 2000, 39, 2632-2657; b) P.J. Stang, F. Diederich, Modern Acetylene Chemistry, VCH-Wiley, Weinheim, 1995; c) L. Brandsma, Preparative Acetylenic Chemistry, Elsevier, 1988.

69 G. Eglinton, A.R. Galbraith, Chem. Int. (London) 1956, 737-738.

70 F. Sondheimer, Pure Appl. Chem. 1963, 7, 363-388.

71 K. Inouchi, S. Kobashi, K. Takimiya, Y. Aso, T. Otsubo, Org. Lett. 2002, 4, 2533-2536.

72 A.S. Hay, J. Org. Chem. 1962, 27, 3320-3321.

73 a) R.F. Heck, J. Am. Chem. Soc. 1968, 90, 5526-5531; b) R.F. Heck, J. Am. Chem. Soc. 1968, 90, 5518- 5526; c) F. Heck, J. Am. Chem. Soc. 1969, 91, 6707-6714.

74 K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 4467-4470.

40 Chapter 2: Synthesis of π-conjugated oligothiophenes

75 R. Rossi, A. Carpita, C. Bigelli, Tetrahedron Lett. 1985, 26, 523-526.

76 Q. Liu, J. Burton, Tetrahedron Lett. 1997, 38, 4371-4374.

77 a) R.W. Wagner, T.E. Johnson, F. Li, J.S. Lindsey, J. Org. Chem. 1995, 60, 5266-5268; b) M. Vlassa, I. Ciocan-Tarta, F. Margineanu, I. Oprean, Tetrahedron 1996, 52, 1337-1342; c) M.E. Wright, M.J. Porsch, C. Buckley, B.B. Cochran, J. Am. Chem. Soc. 1997, 119, 8393-8394; d) X. Huang, J.-H Wang, Synth. Commun. 2000, 30(1), 9-14; e) A. Lei, M. Srivastava, X. Zhang, J. Org. Chem. 2002, 67, 1969-1971.

78 A Godt, C. Franzen, S Veit, V. Enkelmann, M. Pannier, G. Jeschke, J. Org. Chem. 2000, 65, 7575-7582.

79 a) W. Chodkiewitz, P. Cadiot, C. R. Hebd. Seances. Acad. Sci. (Paris) 1955, 241, 1055-1057; b) M. Alani, F. Ferri, Tetrahedron Lett. 1996, 37, 2763-2766.

80 K. Sonogashira in Comprehensive Organic Synthesis, Eds.: I. Fleming, B. Trost, Pergamon, Vol 3, New York, 1991,

81 a) E.J. Corey, P.L. Fuchs, Tetrahedron Lett. 1972, 13, 3769-3772; b) T.B. Patrick, J.L. Honegger, J. Org. Chem. 1974, 39, 3791-3793; b) J.-P. Beny, S.N. Dhawan, J. Kagan, S. Sundlass, J. Org. Chem. 1982, 47, 2201-2204.

82 K. Sonogashira, T. Yatake, Y. Tohda, S. Takahashi, N. Hagihara, J. Chem. Soc., Chem. Commun. 1977, 291-292.

83 a) K. Sonogashira in Metal-Catalyzed Cross-Coupling Reactions, Eds.: F. Diederich, P.J. Stang, Wiley- VCH, New York 1998, Chapter 5; b) L. Brandsma, S.F. Vasilevsky, H.D. Verkruijssse, Application of Transition Metal Catalysts in Organic Synthesis, Springer Verlag, Berlin, 1998, Chapter 10; c) T. Hundertmark, A.F. Littke, S.L. Buchwald, G.C. Fu, Org. Lett. 2000, 2, 1729-1731; d) E. Negishi, L. Anastasia, Chem. Rev. 2003, 103(5), 1979-2018.

Chapter 3

Shape-persistent nanosized macrocycles

Abstract

In this chapter, a detailed literature overview of research on nanosized macrocycles is provided. Strategies and recent developments in the synthesis of these molecules are discussed and the highly interesting inherent properties of these cyclic structures on the basis of some examples are illustrated.

42 Chapter 3 Shape-persistent nanosized macrocycles

3.1 Introduction

In the last decade, research on shape-persistent macrocycles on nanometer scale has become a topic of growing interest in the field of organic chemistry and material science.1 Shape-persistent macrocycles are built up, in contrast to flexible cycles such as cycloalkanes,2 by a rigid, non-collapsible backbone which have a tunable cavity in the nanometer regime well separated from the exterior. Apart from the intense synthetic activity towards developing versatile and efficient methods to prepare such macrocycles with varied structures, special emphasis is given to studies concerning the supramolecular aspects of these compounds such as the binding of appropriate guest molecules, their pattern formation at interfaces, and organization of the molecules in the solid and in the liquid crystalline states (Figure 3.1). Consequently, their potential applicability as ion channels, liquid crystalline materials, microscopic reactors, porous molecular solids or artificial enzymes have also received great attraction.Error! Bookmark not defined.

Host-guest interactions Aggregation in solution Self-assembly at interfaces and in solid-state

Figure 3.1. Properties of shape-persistent macrocycles.

Besides representatives such as cyclopeptides3 and cyclodextrines4 which are relatively rigid only if the diameter of the ring is not too large, and besides the class of metallacycles,5 the field of shape-persistent macrocycles is dominated by the structures consisting of aromatic backbones and sp- and/or sp2-hybridized groups.Error! Bookmark not defined.,6 Hence another and just as much important motivation behind the great fundamental and technological interest on these compounds arose, namely, the investigation of their electronic and optical properties. A common special feature among these macrocycles is that their molecular structures consist, like those of the conventional oligomers or polymers, of basic repeating units, but without end-groups. This renders them as interesting candidates for various future perspectives and applications in molecular

3.1 Introduction 43

electronics and optical devices. Particularly, well-defined π-conjugated macrocycles are of superior interest, not only because of their fascinating structural features. On the other hand, they may represent model systems which ideally combine an infinite defect-free π- conjugated chain of an idealized polymer with the advantage of a structurally well-defined oligomer, excluding perturbing end-effects. Due to their toroidal structures with fully conjugated periphery they could act as intriguing “molecular circuits” which would additionally include sites for selective recognition and selective complexation. Further, as ideal model systems for corresponding polymers their systematic investigation would provide specific information concerning correlations between structural parameters and physical behaviour that are necessary to design polymers with improved material performance.7 The low number of reports on cyclic structures with fully π-conjugated periphery clearly demonstrates the synthetic challenge to prepare such conjugated macrocycles. Impressive examples are the class of cyclic p-phenylenes8 or the fascinating [8.8]-p- cyclophaneoctayne reported by Haley (Figure 3.2).9

B A

Figure 3.2. π-conjugated macrocycles: A cyclic p-phenylene-ethynylene; B [8.8]-p-cyclophaneoctayne.

Fully π-conjugated p-phenylene-based acetylinic macrocycles have recently been reported by Kawase and Oda.10 The remarkable property of these circular systems to selectively recognize C60 will be presented in chapter 3.2.6. Noteworthy to mention is also Mayor’s very recently reported giant molecular ring whose electronic properties will be discussed in section 3.2.2.11 In the next chapter, firstly, the properties exhibited by shape-persistent macrocycles on the basis of some examples are discussed. General synthetic strategies and methods that lead to the successful preparation of rigid cyclic structures are discussed in the later sections.

44 Chapter 3 Shape-persistent nanosized macrocycles

3.2 Properties of shape-persistent macrocycles

3.2.1 Introduction

Motivated by the interesting novel properties and potential applications of these rigid macrocycles with non-collapsible backbones and defined interiors, intense efforts have been given to elucidate structural aspects on molecular levels including geometry, bonding, strain and dynamic aspects. Investigations on their supramolecular organizations have provided a better understanding of the non-covalent driving forces responsible for their association behaviour in solid and in solution states, as well as adsorption behaviours on solid phase.Error! Bookmark not defined. Due to their conformational rigid frames, in comparison to the linear analogues shape- persistent macrocycles exhibit a much lower solubility in common organic solvents. In order to improve the solubility of these compounds, introduction of flexible side chains is necessary. This matter is of crucial importance not only for the synthesis, purification, characterization or modification of the macrocycles, but also for application purposes. Furthermore, due to the substitution with flexible side chains a reduction of the high melting point of these rigid compounds below the decomposition temperature is achieved. Certainly, the cyclic backbone of these macrocycles can also be modified with peripheral side chains containing various functionalities. This allows a control over the chemical and physical behaviour of the targeted compounds and moreover, control over the interactions of the rings with other rings or with further components. Depending on the orientation and characteristic of the functionalities, new unique and interesting properties of these compounds arise, enabling their applications in biological as well as materials sciences. For instance, the perpendicular orientation of the functional side chains to the plane of the ring may promote stacking of the rings enabling the formation of tubular superstructures similarly to the cyclic oligopeptides reported by Ghadiri.12 As shown in Figure 3.3, the 24- membered macrocycle adopts a flat conformation and all amide functionalities lie approximately perpendicular to the plane of the structure.

3.2 Properties of shape-persistent macrocycles 45

H O H O N O H NO H R NN NN R O N H ON H H O H O H O H O O H O H N N R R N N N N O N H ON H H O H O H O H O N O H NO H R NN NN R O N H ON H H O H O H O H O O H O H N N R R N N N N O N H ON H H O H O

Figure 3.3. Tubular configuration of cyclopeptides by self-assembly.

In contrast, as demonstrated by Moore et al.,13 the outside orientation of functionalized anchor groups favours the formation of two-dimensional layered structures.

HO OH HO OH

OH HO OH O H

HO OH HO OH

HO OH O OH H

HO OH HO OH

Figure 3.4. Two-dimensional layered structure due to directive hydrogen bonding.

As illustrated in Figure 3.4, the rigid macrocycles having peripheral hydroxyl-groups attached tend to form two dimensional closest packet sheets due to the directive forces of

46 Chapter 3 Shape-persistent nanosized macrocycles

the hydrogen bonding. In this structure, two types of holes are existent: ones being the inner cavities of the macrocycles themselves and the other ones are formed between the macrocycles by the hydrogen bonds. Very interesting candidates are also rigid macrocycles with defined arrangement of side groups oriented to the inside of the ring.14 These systems, as host molecules, are able to bind guest molecules and due to their nanometer-size even induce chemical reaction inside their cavity.15

3.2.2 Electronic and optical properties of shape-persistent macrocycles

The isolation and characterization of [18]annulene in 1959 by Sondheimer et al.16 which provided experimental support for the Hückel theory of aromaticity, prompted a very extensive investigation towards a large number of macrocyclic π-electron perimeters. Beside [n]annulenes,17 hundreds of other annulenic structures, including analogous heteroatom bridged annulenes,18 dehydroannulenes and their benzofused derivatives containing acetylenic or diacetylenic bondsError! Bookmark not defined.c,19 have been synthesized and studied. In Figure 3.5 some of these cyclic annulenic structures are illustrated. One of the main impetuses driving these studies had been the fundamental question of whether planar examples of such ring systems were able to sustain induced ring currents, and thus exhibit a degree of aromatic character.20 Organic polymeric materials with extended π-conjugation continue to be the topic of wide spread current interest due to their novel and unique electrical, optical and structural properties. They can be designed and tailored to give materials with specialized properties for specific application towards the next generation of electronics and photonics.21 In this respect, fully conjugated shape-persistent macrocycles represent a new class of functional organic materials with superior properties. As mentioned above, the cyclic structure of these compounds represents an infinite well-defined π-conjugated chain.

3.2 Properties of shape-persistent macrocycles 47

Ar Ar N N X H H N HN S S N N n S S NH N Ar Y Y H H Ar N N

n = 1-2 A X = S, Se, O, NH, N-Me Y = S, Se C D B

R R R R R R

n R R n = 1-5 n R R G

R E R n = 0-2 F

Figure 3.5. Cyclic annulenic structures: A [18]annulene; B annulenic structures based on various heterocyclic rings; C sulphur bridged annulenes; D cyclo[8]pyrrole; E peretynylated aromatic dehydroannulene; F diyne-dehydrobenzoannulenes; G monoyne-dehydrobenzoannulenes.

Except for the class of expanded porphyrins, which in the meanwhile can be prepared in relatively good yields either through an acid-catalyzed condensation or oxidative coupling reactions of appropriate precursors,18f the synthesis of shape-persistent macrocycles by a direct oligomerization reaction is almost impossible. Due to the sterical strain inserted by the stiffness of the backbone, in general, additional synthetic steps after cyclic formation are required. Therefore, the synthetic strategies are usually based on modular approaches, such as acetylene scaffolding to give phenylene-ethynylene22 or phenylene-diethynylene macrocycles.23 A more detail discussion on the synthetic strategies applied for the preparation of shape-persistent macrocycles will be given in Chapter 3.3. Schlüter et al. succeeded in the synthesis of various large hexagons by applying the strategy of repetitive reaction sequences based on the Suzuki cross-coupling of kinked oligophenylene building blocks.24 After a remarkable synthetic effort that included many reaction steps, the largest homologue in these series, the cyclotetraicosaphenylene containing of 24 phenylene rings, could be isolated (Scheme 3.1).

48 Chapter 3 Shape-persistent nanosized macrocycles

R R

+ TMS* I

B(OH) R Br R n 2 n

[Pd]

R = C6H13 R

R R R R

Scheme 3.1. Cyclotetraicosaphenylene reported by Schlüter et al.

In general, the angular building blocks, such as the m-substituted phenylenes, are introduced in order to favour ring-formation. This is the reason why most macrocyles prepared up to now have a polygonic shape. The drawback of this strategy is that the angular tectons give rise to an electronic interruption and thus, conjugation along the chain. The UV-VIS spectra of these macrocycles are identical to the linear subunits in the ring. In the case of the cyclotetraicosaphenylene, due to the meta-substitution of each fifth repeating unit, electronically, only a penta-p-phenylene unit is active. More recently, Mayor et al. reported the synthesis of a giant conjugated macrocycle with an estimated diameter of about 12 nm.25

3.2 Properties of shape-persistent macrocycles 49

OC6H13 TMS OC6H13

TIPS OC6H13 TIPS OC6H13

Br S Br

C H O OC H 6 13 S 6 13

TIPS OC6H13 C6H13O TIPS

12 steps

C H O OC H 6 13 S 6 13

R OC H C H O R 6 13 6 13 16

Cu(OAc)2

C H O OC H 6 13 S 6 13

OC6H13 C6H13O 16

Scheme 3.2. Synthesis of a giant conjugated macrocycle by Mayor et al.

The giant molecule C1 comprising exclusively conjugation-active subunits, such as acetylene- and diacetylenes-bounded thiophene and p-benzene rings, was prepared by stepwise synthesis via oxidative acetylene-coupling strategies (Scheme 3.2). The longest wavelength absorption maxima of the corresponding deprotected open-chain oligomers showed, as expected, a continuous bathochromic shift with increasing chain length. The linear regression of these values plotted against the inverse chain length (1/n) gave a value of 462 nm, which represent the theoretical absorption maxima for an infinite conjugated

50 Chapter 3 Shape-persistent nanosized macrocycles

chain length. The measured value of 461 nm for the giant ring C1 corresponds with this theoretical value, and thus, confirms the existence of π-conjugation along the cyclic chain.

3.2.3. Aggregation behaviour of shape-persistent macrocycles in solution

Shape-persistent macrocycles having a rigid backbone with a large aromatic surface tend to stack in solution and in the condensed phase due to van der Waals and π-π interactions.Error! Bookmark not defined.a,26 In solution, the aggregation strongly depends on the solvent. A number of other factors such as surface area, and electronic characteristic of the attached substituents also influence the aggregation behaviour, and thus, the investigation and understanding of these complex interactions are rather difficult. However, within the last decade on the basis of Moore’s pioneered work22,27 significant developments have taken place in the elucidation of the aggregation behaviour of shape-persistent macrocycles. The key methods in these studies are NMR titration experiments complemented by vapour pressure osmometry (VPO) technique. The last enables to distinguish systems in which monomer-dimer equilibrium dominates from those where higher aggregates are formed. However, at lower concentration and in medium polar chlorohydrocarbon solvents such as chloroform, it was shown that the aggregation mostly affords dimers and the association to higher aggregates is negligible. In general, two different stacking motives, edge-to-face and offset face-to-face, are exhibited by associated aromatic compounds. The first one is found to be more favoured due to the maximum number of interacting units and minimal quadrupole-quadrupole interactions. As a model, in Figure 3.6 the two different aromatic interactions are schematically illustrated using simple benzene units.

a) b)

H H H

- + ++ - - - - ++ - - ++ - +

Figure 3.6. Geometries of aromatic interactions: a) edge-to-face; b) offset face-to-face.

3.2 Properties of shape-persistent macrocycles 51

The strong effect of the aromatic ring size on the aggregation strength has been demonstrated by Moore et al.22c The values of the association constants of the cyclic homologues series of phenylene-ethynylenes C2, C3 and C4 which are illustrated in Figure 3.7 were determined from the NMR data. The lower association constant of -1 cyclopentamer C2 (Kassoz = 11 M ) compared to its next higher homologue C3 (Kassoz = 60 M-1) is attributed to its smaller ring size. In contrast, the low association tendency of -1 macrocycle C4 (Kassoz = 16 M ) having a phenylene-ethynylene unit more than C3 is attributed to its non-planar geometry and conformational strain.

R R

n R R = CO nBu C2 n = 0 2 C3 n = 1 C4 n = 2

Figure 3.7. Phenylene-ethynylene macrocycles investigated by Moore et al.

As mentioned earlier, the π-stacking interaction is sensitive to the solvent. A very interesting study was reported by Höger et al.28 regarding the solvophobic nature of the association. The rigid macrocycle C5 (Figure 3.8) exhibits a good solubility in a variety of halogenated organic solvents, such as chloroform or dichloromethane, but no aggregation. An induced solvophobical π-π aggregation to dimers could be achieved by addition of a solvent which acts as a good solvent for the peripheric alkyl chains but as a poor solvent for the macrocyclic rigid core. As determined from the NMR chemical shift data, the -1 dimerization constant of macrocycle C5 was amplified from Kassoz = 130 ± 30 M to Kassoz -1 = 790 ± 180 M when the hexane composition in a CD2Cl2/hexane mixture was increased from 1:3 to 1:6.

52 Chapter 3 Shape-persistent nanosized macrocycles

CH3 R R O O

H C CH 3 O O 3

C3H7 C3H7

C3H7 C3H7 O O H3C CH3

O O R R

CH3 OC12H25

C5 R = OC12H25

OC H 12 25

Figure 3.8. Shape-persistent macrocycle investigated by Höger et al.

The strength of the association is also dramatically influenced by the electronic and steric characteristics of the substituents attached to the aromatic core. Detailed investigation on various phenylene-diethynylene macrocycles C6-C14 have been made by Tobe et al (Figure 3.9).29 The values of the association constants of C6-C14 are summarized in Table 1. By comparing the values of C6 and C7 with others, one can conclude that bulky, electron-donating groups such as tert-butyl prevent aggregation. In contrast, linear alkyl or triethyleneglycol-ester side chains strongly enhance the association of the rigid macrocycles. Furthermore, the results disclose that electron-withdrawing substituents enhance self-association and the side chain length has little influence. Very interesting is also the comparison between the association behaviour of the phenylene-diethynylene macrocycles C9, C11 and C14 to those of the corresponding phenylene-ethynylene macrocycles C2, C3, C4 of Moore22a,b having the same number of phenylene units and the same type of side chains (Figure 3.8). The in general higher association constants for the macrocycles of the former series were attributed to the electron-withdrawing ability of the butadiyne groups that lead to a more electron-deficient framework.

3.2 Properties of shape-persistent macrocycles 53

R R Table 1. Self-assoziation data from 1H

NMR (303 K in CDCl3) of the phenylene-diethynylene macrocycle C6-C14.

-1 Macrocycle Kassoz. [M ] C6 -- R R C7 -- n C8 27.9 ± 0.7 C9 28.7 ± 2.4 R C10 19.9 ± 1.3

C6 n = 0, R = tBu C10 n = 0, R = CO2(CH2CH2O)3CH3 C11 173 ± 17 C7 n = 2, R = tBu C11 n = 1, R = CO2-nC8H17 C12 150 ± 4 C8 n = 0, R = CO2-nC8H17 C12 n = 1, R = CO2-nC16H33 C9 n = 0, R = CO -nC H C13 n = 1, R = CO (CH CH O) CH 2 16 33 2 2 2 3 3 C13 42.7 ± 14 C14 n = 2, R = CO -nC H 2 8 17 C14 22.8 ± 0.9

Figure 3.9. Phenylene-diethynylene macrocyclereported by Tobe et al.

3.2.4 Solid state structures of shape-persistent macrocycles

The determination of single crystal structures of shape-persistent macrocycles is an inherently difficult task. The large cavities of the cycles are mostly filled with solvent molecules, the evaporation of which causing empty spaces. In many cases this results in a breakdown of the crystal lattice and in a total loss of X-ray diffraction power.30 In order to prevent such a loss of internal solvent the X-ray analysis of such cyclic compounds are generally carried out at low temperatures. Besides solvent molecules, a degree of disorder is caused also by the flexible side chains attached to the core. This is the reason why for macrocycles the R values of refinement that are the guide to the accuracy and precision between the X-ray diffraction data and the models of the molecular structures are often high.Error! Bookmark not defined. Due to their polygonal geometry, shape-persistent rigid macrocycles are expected to be basically flat on a time average, certainly, without excluding the attainment of various other conformations. Nevertheless, there are several examples reported where a distortion from planarity has been observed in the crystalline state. In most cases, this can easily be

54 Chapter 3 Shape-persistent nanosized macrocycles

explained by the geometric constraints of the framework. Large deviation from planarity in the single-crystal structure has also been reported for the large cyclotetraicosaphenylene that was shown in Scheme 3.1. The hexagonal macrocycle having a corner to corner and edge to edge distance of 3.1 nm and 2.7 nm, respectively, adopts in the crystal state a cyclohexane-like chair conformation.31 A major goal in material science is to use channels formed by crystals of these cyclic structures for transport processes. The formation of channels requires a packing in which the cyclic π-systems are layered one over another and the substituents point either an endo- or an exocyclic-fashion. X-Ray investigations on shape-persistent macrocycles show that these systems in the solid state preferentially form layered structures with a co-facial arrangement in which the centers are laterally offset. By this arrangement the repulsive interactions between the π-electrons are outweighted by attractive forces that arise due to the interaction between the positively charged σ-core and negatively charged π-backbone. As has been shown by different research groups, the slipping of the centers can be reduced either by diminishing the electron density within the π-systems or by additional directive forces, such as hydrogen bonds.32 A very interesting observation was recently reported by Höger et al.33 Amphiphilic macrocycle C16 illustrated in Figure 3.10 is functionalized by endo- and exocyclic polar and nonpolar side chains.

CH3

OH OH

H C CH 3 O O 3

C6H13 C6H13

C6H13 C6H13 O O H3C CH3

OH OH

CH3 C16

Figure 3.10. The amphiphilic macrocycle crystallises in two different forms depending on the solvent.

3.2 Properties of shape-persistent macrocycles 55

Due to the free rotation around the p-phenylene moieties the conformation of macrocycle C16 is adaptable to the polarity of the solvent resulting either in a hydrophilic or a hydrophobic cavity. Thus, the crystal structure obtained from pyridine solution shows a conformation state with the cavity filled with hydrophobic side chains. In contrast, the crystals obtained from THF, a less polar solvent, the hydroxyl groups were all found to be pointed to the inside. However, in both cases, the macrocycles stacked face-to-face, forming extended tubular channels with aligned void cavities. Schlüter et al.34 investigated the solid state association behaviour of macrocycles having bipyridine units in the backbones. The cycles illustrated in Figure 3.11 posses a planar conformation with the bipyridine units slightly tilted out of plane and their cavities are filled with solvent molecules and the side chains of adjacent cycles. Interestingly, although the three systems C17, C18 and C19 structurally have the same cyclic backbone, due to the different side chains, they show different packing behaviour in the solid state. These findings give further evidence for the strong effects caused by the nature of the side chains on the packing motives in the solid state of cyclic rigid compounds.

R1 R1 N

R2 R2

N R1 R1

C17 R1 = CH2OC6H13, R2 = H

C18 R1 = CH2OMOM, R2 = CH2OC6H13 C19 R = CH OTHP, R = CH OC H 1 2 2 2 6 13

Figure 3.11. Shape-persistent macrocycles containing bipyridine units investigated by Schlüter et al.

3.2.5 Self-assembly of shape-persistent macrocycles at the solid-liquid interface

Another considerable interest is nowadays directed toward investigations dealing with the arrangement of shape-persistent macrocycles into regular two-dimensional arrays. One key method to investigate self-assembly properties at the solid-liquid interface of molecules is

56 Chapter 3 Shape-persistent nanosized macrocycles

the scanning tunneling microscopy (STM).35 This technique provides valuable information on the molecular and packing parameters of compounds, factors that are crucial for improving material properties. Up to now, the STM technique has rarely been applied in the field of shape-persistent macrocycles.6a,b,28 The main reason for it is the difficulty to prepare a large variety of macrocycles with different substituents and substitutions patterns. Thus, systematic studies to determine the concrete structures which under adequate conditions would form two-dimensional arrays are limited.

3.2.6 Liquid crystalline mesophases of shape persistent macrocycles

The first discotic liquid crystals have been described by Chandrasekhar.36 Since then this area has attracted a lot of interest because of possible applications of these materials in display technology.37 More recently, discotic liquid crystals based on polycyclic aromatic cores have even been found to exhibit high mobilities for photoinduced charge carriers in the columnar phase,38 enabling the development of highly efficient photovoltaic elements based on such materials.39 In general, liquid crystalline materials are composed of disklike or macrocyclic cores and flexible side chains pointing to the outside. Thus, due to their non-collapsible framework, shape-persistent macrocycles bearing flexible side chains are promising candidates for such mesogens of columnar crystals. Besides all the studies that have been done on rigid macrocycles to elucidate their liquid crystalline properties,6b,40 a very interesting unique architecture of arylene-ethynylene macrocycles has been reported by Höger et al (Figure 3.12).41 Macrocycle C20 exhibits a thermotropic LC mesophase. This is not the case for macrocycle C21 which has a similar structure as C20 but the side chains are attached at the periphery of the backbone, at the corner units. The single crystal structure analysis of macrocycle C20 revealed that the long side chains were pointing into the cavity of the framework. As a result of this orientation that is enabled only in macrocycle C20 by the rotational freedom of the para-phenylene unit, the intermolecular entanglements and side chain-cavity interpenetrations are eliminated and the molecular mobility which is required for occurrence of mesophase is entailed.

3.2 Properties of shape-persistent macrocycles 57

CH3

C18H37 C18H37 O O

H C CH 3 O O 3

C3H7 C3H7

C3H7 C3H7 O O H3C CH3

O O

C18H37 C18H37

CH3 C 20

CH3

C3H7 C3H7 O O

C H O OC H 18 37 O O 18 37

C3H7 C3H7

C3H7 C3H7 O O C18H37O OC18H37

O O

C3H7 C3H7

CH3 C 21

Figure 3.12. Phenylene-ethynylene macrocycles investigated by Höger et al.

3.2.7 Host-guest interactions of shape-persistent macrocycles

Although significant developments have taken place within the last few years in the synthesis of shape-persistent macrocycles, the use of these structures having nanometer- sized cavities in host-guest chemistry has largely been unexplored.

58 Chapter 3 Shape-persistent nanosized macrocycles

Macrocycles with additional coordinative functionality integrated into the rigid framework of the macrocycles, such nitrogen in pyridine,42 bipyridines32 and terpyridines43 containing macrocycles (or as sulfur in the macrocyclic oligothiophenes investigated during this work) should considerably broaden the applicability of such systems with respect to host- guest chemistry. Depending on the orientation of the functionalities an exo- or an endocyclic complexation with metals or metal ions should be preferred. As schematically illustrated in Figure 3.13, the exocyclic orientation of the functionalities may favour the formation of rigid scaffolds with metal placed at defined distance. Metal-mediated one- dimensional, two-dimensional or even spatial (three-dimensional) supramolecular arrays would result. In turn, the endocyclic orientation should enable the specific complexation of guests inside the cavities of the macrocycles. These abilities combined with the unique inherent properties of shape-persistent macrocycles to self-associate in columnar channels open the way towards designing new materials with potential applications as porous solids or nanowires.44

a) b)

Figure 3.13. a) Exocyclicand b) endocyclical orientation of the functionalities.

The advantages that arise by making use of host-guest interactions in the field of macrocycles have been clearly pointed out by Sanders and Anderson.45 In their work, they showed that metallo-porphyrins, mostly comprising zinc, can be very effectively coupled in presence of pyridine derivatives acting as guests. As illustrated in Scheme 3.3, due to axial coordination to the metal a pre-arrangement of the precursors can be achieved. Later, Lindsey et al. extended this concept to the template-directed approach by using tripyridyl- units as guest molecules and succeeding in the synthesis of analogues shape-persistent cyclic hexameric porphyrin arrays.46

3.2 Properties of shape-persistent macrocycles 59

R R

N N Zn N N R

CuCl/CuCl2 [70%]

R R

N N Zn N N R R R R N N Zn N N R R Scheme 3.3. Oxidative coupling of metallo-porphyrins in presence of bipyridine acting as guest.

Remarkable examples in this field are contributions reported by Tykwinsky et al.47 They use the exocyclic metal coordination of pyridine-units of cross-conjugated macrocycles to generate well defined hybrid-metal structures and organized porous materials. Macrocyle C22 which comprises two pyridine rings assembles in the presence of Pt metal ions to the supramolecular species C23. In this structure two macrocycles are bound in pair in a face to-face orientation (Scheme 3.4.). The solid-state structure analysis revealed that C23 forms a three-dimensional network with bidirectional channels. Evidence for the selective uptake of small organic molecules by this porous solid was given by incorporation of 1,2- dichloroethane into the crystal lattice of C23.

Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph PEt Ph P 3 3 N N PPh N N TfO Pt PEt Pt 3 + 3 Ph P Pt 3 N PPh OTf N 3 Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph Ph C23 C22

Scheme 3.4. Metal-directed self-assembly of rigid macrocycle C23.

60 Chapter 3 Shape-persistent nanosized macrocycles

Impressive studies on host-guest interactions were recently reported by Oda and Kawase.10 As already mentioned, they succeeded in the synthesis and characterization of various fully π-conjugated macrocycles based on p-phenylene-ethynylenes. Some of these strained, but isolable novel carbon nanorings are depicted in Figure 3.14. The cyclic [6]-p-phenylene- ethynylene C24 has a smooth belt-shaped structure with a 1.31 nm diameter. Although the cavity size of the macrocyle is slightly smaller than that of C60 C25 was found to form very 48 stable inclusion complexes with fullerene C60 in solution as well as in the solid state. The explanation for the strong complexation is given by a driving force based on a novel concave-convex π-π interaction. The same interaction is suggested to play a role also in the formation of various layered carbon networks with closed curved structures, such as multiwalled carbon nanotubes,49 bucky onions50 and fullerene peapods.51 Furthermore, the slightly larger carbon nanoring C29 undergoes host-guest interactions with fullerene C70. Although the complexation is not very strong, it exhibits a high selectivity upon competitive complexation from a solution of a mixture of fullerenes. This nanoring can act as the best fluorescence sensor for fullerenes among all the known hosts to date.52

C24 n = 1 C26 n = 3 C29 C25 n = 2 C27 n = 4 C28

Figure 3.14. Carbon nanorings synthesized by Oda and Kawase capable to undergo complexation with fullerenes.

3.3 General synthetic strategies towards nanosized macrocycles 61

3.3 General synthetic strategies towards nanosized macrocycles

3.3.1 Introduction

Over the past decade intense efforts have been made to explore highly efficient methods to prepare such macrocycles of various structures providing diverse design requirements.Error! Bookmark not defined. For the synthesis of shape-persistent macrocycles generally two approaches are applicable: a kinetic or a thermodynamic approach. The classical way is to use kinetically controlled reactions, which result in an irreversible formation of the cyclic backbone. Because irreversible reactions do not have the ability to correct undesired bond formations (no self-healing behaviour), the product distribution formed during the cyclization step is kinetically determined. The determining factor in this case is the conformation probability of the linear precursor from which a ring-closure reaction can take place. A relatively new approach in this area is the preparation of macrocycles under thermodynamic control. Thermodynamically controlled reactions are reversible and have the ability of self-healing by making and breaking of covalent bonds as an on-going process in a chemical reaction. The product distributions in this case depend on the relative stabilities of the final products. Hence, the major product formed will be the most stable compound at the given reaction conditions. Very popular in macrocyclic chemistry is the use of a template. Depending on the strength of the template-substrate interaction and on the substrate-substrate bond that is subsequently formed, the role of a template can significantly vary. Classical templates bind non-covalently to their substrates, and assist in the formation of covalent bond. But closely related to this field are also reactions in which the bond between the template and the substrate is covalent.

3.3.2 The kinetic approach towards synthesis of macrocycles

In general, when a kinetic approach is used for the effective preparation of macrocycles, the largest effort is put in the development of synthetic strategies that lead to the desired product and avoid formation of linear oligomers or polymers. As illustrated in Figure 3.15, for the construction of the framework of shape-persistent macrocycles three major

62 Chapter 3 Shape-persistent nanosized macrocycles

strategies can be distinguished: 1) one step random oligomerization-cyclization; 2) intramolecular ring-closure reaction and 3) as a hybrid of the first two strategies, the intermolecular coupling between two or three ring fragments followed by unimolecular cyclization in one pot.

1) 2)

Figure 3.15. Schematic representation of macrocyclization strategies: 1) one step random oligomerization- cyclization; 2) intramolecular ring-closure reaction; 3) intermolecular coupling between two or more ring fragments followed by unimolecular cyclization in one pot, which is a hybrid of the first two strategies.

The contrast between the two extreme strategies, the one-pot oligomerization-cyclization route and the intramolecular macrocyclization, was pointed out in several contributions. Most of the macrocycles contained phenylene, phenylene-acetylene or phenylene- diacetylene units in the backbone. As a classical example, the synthesis of hexakis[m- phenylene-ethynylene] macrocycles C30 and C31 differing only in substituent R, is discussed. The two extreme strategies that led to the formation of this cyclic structure are depicted in Scheme 3.4. The one-pot oligomerization-cyclization was performed by Staab in 1974.53 Starting from a simple precursor containing a single repeating unit, namely copper m-iodophenyl acetylide, and employing a six-fold Stephens-Castro coupling the hexameric macrocycle C30 could be synthesized. The cyclic product was obtained in 4.6% yield after separation from a broad mixture of cyclic and linear oligomers of different lengths which were produced by this statistical reaction. The advantage of the one-pot oligomerization-cyclization strategy becomes evident; the starting material is easily accessible and the target cyclic molecule is generated in a single step. The drawback of this method is the inevitably low yield obtained for the desired cyclic structure. The ring- closure reaction is a unimolecular process that is intrinsically disfavoured by entropy due

3.3 General synthetic strategies towards nanosized macrocycles 63

to the loss of conformational freedom. In contrast, in order to afford oligomer precursor formation, in general such reactions must be run under conditions that favour bimolecular

a) R

R R

pyridine Cu

R R

C30 R = H R 4.6 %

b) R R R R TMS MeI Pd(0) / CuI TMS I + K2CO3 R Et2N3 Et2N3 H TMS

Et2N3

R R

R I I MeI R K CO + TMS 2 3 R R R several steps

R Et2N3 H

H Pd(0) / CuI R R

R R

C31 R = t-Bu R 75 %

Scheme 3.4. Synthesis of hexakis(m-phenylene-ethynylene) macrocycle by a) one-step oligomerization- cyclization and b) intramolecular cyclization strategies.

64 Chapter 3 Shape-persistent nanosized macrocycles

processes. A tedious undertaking is also the separation and purification of the target cyclic molecule from the complex mixtures containing polydisperse byproducts with similar molecular weights. The intramolecular approach that is the other extreme strategy towards synthesis of macrocycles was used by Moore et al. 20 years later. They reported the successful synthesis of various representatives of the same class of phenylene-acetylene based macrocycles.22b,54 In comparison to the one step method, besides yield improvement, the intramolecular ring-closure strategy discloses the remarkable advantage to provide a great versatility for absolute control over the size and functional group placement of the subsequently formed macrocycle. The main shortcoming of the method is evidently the time-consuming precursor synthesis prior to cyclization including an increased number of synthetic and purification steps. As shown in Scheme 3.4, the hexameric α-iodo-ω-ethynyl functionalized m-phenylene-ethynylene linear precursor was first prepared by using Pd-Cu catalyzed cross-coupling reactions and a pair of effective, orthogonal protecting groups. The subsequent intramolecular ring-closing Sonogashira reaction under pseudo high- dilution conditions generated the desired macrocycle C32 in 75% yield. In this final reaction, where bond formation of a difunctionalized precursor to a cyclic structure takes place, highly dilute conditions were applied. The use of high-dilution conditions in order to increase the yield of cyclic products over the linear oligomers was established by Ruggli already in 1912.55 The cyclization of a difunctionalized precursor is an intramolecular first order reaction. Its rate is proportional to the concentration of the precursor. The competitive intermolecular coupling, which affords linear or cyclic oligomers, is second order. Its rate is proportional to the square of the concentration of the precursor. Therefore, the application of high dilution kinetically favours the intramolecular reaction. In order to diminish the total amount of solvent and to avoid the use of very large equipment, the technique of pseudo high-dilution conditions is generally employed.56 Here, the reactants are introduced slowly with the aid of a syringe pump to the reaction vessel over a longer period of time. The rate of addition should be slower than the consumption of precursors in order to keep the stationary concentration low. The resulting long reaction time may sometimes pose problems, especially if the functional groups of the precursor are unstable or tend to undergo other side reactions which are detrimental for cycle formation. Even though some studies were undertaken to find or define the optimal balance between these two opposite effects,57 in general, the

3.3 General synthetic strategies towards nanosized macrocycles 65

influencing factors are individual for each system. Therefore, the dilution principle is often used empirically. The third strategy towards synthesis of macrocycles combines the advantages and circumvents the shortcomings of the two extreme methods. As was shown in Figure 3.15, in this case, a pair of oligomers with appropriate chain length and suitable terminal functionalities undergoes in a one-pot reaction a bimolecular coupling under cyclic formation. In general, this method generates macrocycles in lower yields as the intramolecular approach. The advantage is that fewer synthetic steps for the preparation of the precursors are required. Several research groups58,59 could show that, by comparing, the overall yield is enhanced and the time-consuming factor is reduced. Compared to the one- step cyclization approach, due to the enlarged precursor the number of possible oligomeric structures is reduced and thus, the product separation and purification simplified.

H C CH H3C 3 3

OTHP OTHP OC H H7C3O OTHP H7C3O 3 7

CuCl / CuCl2 CH H3C H3C 3 40-45 %

OTHP OTHP OC H H7C3O OTHP H7C3O 3 7

H C H C CH 3 3 C32 3

Scheme 3.5. Synthesis of shape-persistent macrocycle C32 using a bimolecular coupling/unimolecular cyclization strategy by Höger et al.

Höger et al. successfully applied this strategy, for example, towards the synthesis of a series of shape-persistent macrocyclic amphiphiles.60 According to GPC data, the cyclic dimer C32 whose synthesis is depicted in Scheme 3.5 was formed in 60-65% yield, along with a cyclic trimer and other cyclic or linear oligomers and polymers. As the cyclization step, an intermolecular Glaser coupling of rigid bisacetylenes (“half-rings”) under pseudo

66 Chapter 3 Shape-persistent nanosized macrocycles

high-dilution conditions was used. The pure macrocycle C32 could be isolated by simple recrystallization in 40-45% yield. They also reported that to simplify the synthesis by further breaking down the cyclization precursors into three pieces (”thirds of rings”) led to less desirable cyclization yields.

3.3.3 The thermodynamic approach towards synthesis of macrocycles

Since macrocyclizations carried out under kinetic control are generally low-yielding reactions, which result in the formation of a wide range of linear and cyclic oligomers and polymers, the prospect of preparing macrocycles under thermodynamic control is extremely attractive. The reversible nature of thermodynamic controlled processes introduces the prospects of “error checking” and “proof-reading”. The product formation occurs under thermodynamic control, and therefore the product distributions are dictated by the relative stabilities of the final products. If the desired macrocycle at the given reaction conditions is the most stable, it will be formed as the major product. This concept has already been widely applied in the field of supramolecular chemistry.61 Simple non- covalent synthesis or strict self-assembly has given access to a wide spectrum of metallacyclic polymers and polyhedra, while the supramolecular assistance to molecular synthesis, also referred to as self assembly followed by covalent modification, has lead to more complex systems, such as interlocked structures (catenanes, rotaxanes) as well as fascinating molecular capsules. A more detail discussion comprising the formation of cyclic structures based on non-covalent synthesis will be given in Chapter 5. However, there is an important difference between supramolecular and covalent chemistry. In general, equilibration processes in covalent systems are much slower than in non- covalent ones. Furthermore, most of the appropriate covalent bonds are so strong that the conditions which would initialize and/or equilibrate the system to the thermodynamically stable product are much too harsh to tolerate the presence of other functionalities in the molecule. As a result, the number of potential covalent functionalities enabling to carry out reactions of organic compounds under thermodynamic control is limited. Examples for such functionalities are esters and borate esters, acetals, disulfides, borazaaromatic anhydrides, hydrazones, imines, oximes and olefins (metathesis). Remarkable are the examples in which under use of thermodynamically controlled covalent bond forming processes certain structures of calixarenes can be obtained. As

3.3 General synthetic strategies towards nanosized macrocycles 67

illustrated in Scheme 3.6, the reaction of p-substituted phenols with formaldehyde in the presence of acid or base carried out under themodynamic control led to the cyclic tetramer calix[4]arene C33, In contrast, the same reaction yielded under kinetic control the cyclooctameric calix[8]arene C34.62 Furthermore, it was shown that at higher temperature calix[8]arene C35 can be converted via fragmentation and subsequent recombination of smaller oligomeric units to the thermodynamically more stable product, which in this case was the calix[4]arene C33.63

R R R

kinet. control therm. control + CH2O

8 4 OH OH OH

C34 C33 T

Scheme 3.6. Synthesis of calix[n]arenes from p-substituted phenols and formaldehyde.

Lindsey et al. have intensively investigated a number of porphyrin systems and the reversible characteristic of their formation.64 The synthesis of such porphyrins involves two steps: in a first thermodynamic controlled reaction of the benzaldehyde derivative with pyrrole yielding the porphyrinogen C35, which then is converted by a following oxidation reaction to the kinetically stable porphyrin macrocycle C36 (Scheme 3.7).

H O R R R R N N H + NH HN NH HN N H N N R1 R R R R

C35 C36

R = R1

Scheme 3.7. Synthesis of porphyrin macrocycles by reaction of benzaldehyde derivatives with pyrrole followed by oxidation of the resulting porphyrinogens system.

68 Chapter 3 Shape-persistent nanosized macrocycles

The ability to prepare selectively only one macrocyclic species under thermodynamic equilibrium conditions was impressively demonstrated by Tilley and his research group.65 By utilizing zirconocene coupling of silyl-substituted alkynes they could prepare a wide set of shape-persistent macrocycles with a variety of representatives in very high yields, mostly being over 90%. Due to the reversible nature of the carbon-carbon bond formation between silyl-substituted acetylenes, the coupling reaction resulted in almost exclusive formation of that cycle in which geometric factors of both the starting building block and the relevant intermediate are ideal. For example, the linear silyl-terminated diynes illustrated in Scheme 3.8 undergo zirconocene coupling with selective formation of cyclic trimers C37, which under acidic conditions can easily be converted to macrocycles of type C38. The length of the diyne-spacers could be increased from n= 1 to n=4 giving access to macrocycles with large nanoscale cavities.

Cp2 SiMe3 Zr Me Si SiMe Me3Si SiMe3 3 3

H+ II [Zr Cp2] n n n n SiMe n SiMe SiMe 3 SiMe3 3 3 Cp Zr 2 n ZrCp2 n

SiMe SiMe3 Me Si 3 Me3Si 3 SiMe3 n = 1-4 C37 n = 1-4 C38 n = 1-4

Scheme 3.8. Synthesis of cyclic trimers by zirconocene coupling reaction of linear diynes.

In comparison, by using various non-linear silyl-terminated diynes as building blocks in an analogue coupling reaction, cyclic dimers C39-C41 were selectively obtained as the thermodynamically stable products.65g As shown in Scheme 3.9, these zirconium- containing macrocycles can be converted in the corresponding metal-free cyclophanes C42-C44 by simple treatment with acids.

3.3 General synthetic strategies towards nanosized macrocycles 69

Me Si Zr SiMe 3 3 Me3Si SiMe3 SiMe3

II [Zr Cp2] H+

SiMe 3 Me Si SiMe 3 Zr 3 Me3Si SiMe3 C42 C39

SiMe 3 Me Si Zr SiMe 3 3 Me3Si SiMe3

II [Zr Cp2] H+

Me Si SiMe 3 Zr 3 Me3Si SiMe3 SiMe3 C43 C40

SiMe 3 Zr Me Si SiMe Me3Si SiMe3 3 3

II [Zr Cp2] H+ Me Si Me Si Me Si 2 Me2Si Me2Si 2 2

Me Si Me Si SiMe 3 Zr SiMe3 3 3 SiMe3 C41 C44

Scheme 3.8. Formation of cyclic dimers by zirconocene coupling reaction of “nonlinear” diynes.

3.3.4 The template approach in macrocycle chemistry

As mentioned earlier, an undoubtedly very elegant and effective method to increase the ratio of cyclic products versus linear ones is the template approach. The role of a template was described by Busch as follows: “a chemical template organises an assembly of atoms, with respect to one or more geometry loci, in order to achieve a particular linking of atoms.’’66 Hence, template processes are those in which a metal or another centre that with a definite stereochemistry and electronic state, serves as a mould, pattern or matrix to form reaction products that are difficult or even impossible to obtain by traditional methods.

70 Chapter 3 Shape-persistent nanosized macrocycles

With respect to macrocycle synthesis the main criterion for a template is its ability to pre- organize and activate initial precursors such that the local reactant concentration is considerably increased and thereby, the cyclization is facilitated. Molecular templates are ubiquitous in nature. The template effect is operative in the replication of the DNA, where each chain in the double helix structure serves as a template for the second chain.67 The DNA is a template for RNA, which in turn is a template for proteins. In the synthesis of macrocyclic compounds the first chemist who intentionally took advantage of the template process was Daryle Busch in the 1960s.68 Based on his work a renaissance in the area of macrocyclic chemistry took place, leading for example to the celebrated discovery of Pederson’s crown polyethers69 or the highly effective synthesis of catenanes and rotaxanes by Sauvage and Stoddart.70 The most common and simplest type of template-directed synthesis is the metal cation induced macrocyclization. For shape-persistent macrocycles the template effect based on simple metal coordination was first applied by Sanders et al.45 An example for the effective synthesis of the porphyrin-containing macrocycles in the presence of bipyridine is illustrated in Scheme 3.3. As previously discussed, due to the ability of the bipyridine to coordinate to the metals a pre-organization of the precursors is achieved and thus, an enhancement in the cyclic structure formation. The main shortcoming of the metal cation induced macrocyclization is the restriction that it can be applied only to the certain cases in which the requirement for coordination sites, such as nitrogen or oxygen donors in the cyclic systems is full filled. But macrocyclization templates have been shown not to be limited to metal ions. Depending on the nature of the template and precursors various other kinds of template bonds can be used, such as donor-acceptor,71 hydrogen bonding,72 ion-dipole,73 covalent74 or metal-carbon σ-bonds.75 The concept of a covalently attached template was introduced by Höger et al. and the advantages that arise by using this approach is clearly pointed out in their reports.Error! Bookmark not defined.c,60 A wide spectrum of phenyl-diethynylene macrocycles was obtained by applying the template approach. In their design, easily available bisacetylenic precursors were either attached to an appropriate template prior to cyclization or the bisacetylenic precursors were directly prepared at the template. As an example, in Scheme 3.9, the synthesis of macrocycle C45 where the latter approach was applied is illustrated. The Glaser coupling of the monomers which were covalently attached to 1,3,5-benzenetricarboxylic acid yielded the template-bound cyclic trimer in 95% as shown by GPC data. The same reaction without employing the templated synthesis

3.3 General synthetic strategies towards nanosized macrocycles 71

led to a mixture of cyclic or linear oligomers and polymers in which the cyclic trimer was formed only in 20-25% yield.

OC3H7

O O

O O O

C3H7O O

OC3H7 CuCl / CuCl2 [94%]

OC3H7

O O

O O O

C3H7O O

OC3H7 NaOH [95%]

OC3H7

O OH C3H7O O

OC3H7

C45 Scheme 3.9. Synthesis of phenyl-ethynylene macrocycle C45 by templated synthesis.

72 Chapter 3 Shape-persistent nanosized macrocycles

Although the template-directed methods have been shown to significantly improve the yields of the macrocyclization steps they still have not found a wide application yet. The reason for this is certainly the large demand in the structural and functional design of the template and the additional synthetic effort required for preparing the templates themselves. Not less important in this process is the release of the template after directing the formation of the product, in order to get the template-free product. Unfortunately in some cases the template becomes locked into the structure created and its removal is achieved only through the action of appropriate reagents.

3.4 Alkynyl-alkynyl coupling reactions

Shape-persistent macrocycles are built up by conformational rigid structures and comprise interior space ranging from slightly less than one up to several nanometers. Due to the requirement of shape-persistence, the field of rigid macrocycles is dominated by structures with aromatic or heteroaromatic backbones, mostly having all-carbon perimeters and containing only sp and sp2-hybridized carbon atoms. In 2002, a detailed summary review concerning the cycle formation reactions has been published by Schlüter et al.Error! Bookmark not defined.b So far, the most applied reactions for the synthesis of shape-persistent macrocycles are Sonogashira coupling6a,22,76 of halogenated aryls and terminal acetylenes and Glaser-type couplings11,29,77 of terminal acetylenes. Beside aryl/aryl couplings,24 the previous presented zirconocene couplings (referred here to as acetylene zirconacyclopentadienation)65 alkyne metathesis78 and McMurry olefination,79 are also used very frequently. The sp-sp oxidative coupling reaction has been introduced in Chapter 2.2. The wide applicability of this reaction with all its modifications is impressively demonstrated in diverse areas of organic chemistry including natural and pharmaceutical products, supramolecular architectures, carbon rich scaffolding and networking.80 The high competence of the acetylenic coupling can be accounted especially to its broad functional group tolerance and relatively simple starting materials requirement. Due to their structural linearity that do not allow cis-trans isomerization and the low sterical demand, 1,3- butadiyne linkages have proven to be valuable connecting units in the rational design of shape-persistent macrocycles. Very appealing for these structural units is furthermore the

3.4 Alkynyl-alkynyl coupling reactions 73

fact that due to the isotropic distribution along the π C-C axis they are capable to transmit electronic perturbation efficiently from one end of the conjugated system to the other, as has been already demonstrated in the field of molecular wires81 and conjugated dendrimers.82 The earliest mechanistic proposals of the copper-mediated oxidative coupling postulated the formation of acetylinic radicals which then recombine to afford the corresponding diynes.83 In a subsequent kinetic study, it was shown that copper(II) served as the direct oxidizing agent.84 Thus, the radical mechanism was discarded in favor of a heterolytic cleavage followed by an electron transfer of a single electron to the copper(II) salt. Studies by Bohlmann et al.85 could confirm these results. They found that a mixture of electronically different alkynes afford predominantly the homocoupled products. Such selectivity would not be expected in a reaction via a radical mechanism. They also noted that the rate of the coupling reaction depends on the nature of conjugated acetylenes. Under alkaline conditions more acidic acetylenes underwent more rapid dimerization, whereas under acidic conditions an inverse relationship was observed. In the last case, the addition of a copper(I) salt became necessary. These observations led to the assumption that a π-complex between copper(I) ions and the triple bond is formed resulting in an activation towards deprotonation of the alkyne as shown in Scheme 3.11.

- + R H R + H + + Cu Cu

Scheme 3.11. Acetylene activation toward deprotonation by π-complex formation.

Furthermore, due to the observed second-order dependence of the coupling rate on alkyne concentration, Bohlmann et al. suggested a dinuclear copper(II) acetylide complex as intermediate that directly affords the coupled product (Scheme 3.12). Such an intermediate also explains the dependence of the rate of the reaction from the substrate concentration. The intramolecular coupling reaction affording cyclic diynes follows a first-order kinetic. This mechanistic proposal was then also confirmed by other groups.86

74 Chapter 3 Shape-persistent nanosized macrocycles

2+ 2+ 2+ B B B B B B Cu Cu - Cu R - + R R + X X X R Cu Cu R Cu B B B B B B

+ R R + 2 Cu

Scheme 3.12. Proposed dinuclear copper(II) acetylide complex as intermediate in the oxidative coupling reaction; B = N ligand, X = Cl-, OAc-.

More recently, Temkin et al.87 proposed on the basis of kinetic studies, the involvement of three copper ions per alkyne unit, two copper(I) and one copper(II) ion, in the rate-limiting step of the coupling reaction. The existence of such a complex intermediate was strengthened by X-ray crystallography in the solid state, where an alkyne unit undergoes π coordination with two copper(I) ions and forms acetylide type bonds with two other copper(I) centers.87a However, despite the large number of mechanistic investigations on this field, apparently the exact understanding of the complex mechanism of this reaction still remains unsatisfactory. The high dependence of the mechanism on the experimental setup makes the systematic comparison of the results from various studies conducted under different conditions is not possible. For the synthesis of macrocycles the most important modification of the Glaser coupling was that reported by Eglinton.88 In this approach, the coupling is performed in the presence of copper(II) salts either in pure pyridine or by using a co-solvent, such as methanol, to prevent precipitation of the copper(I) derivates formed during the reaction.89 Later, it was shown that pyridine can even be entirely substituted by other solvents, such as acetonitrile90 or DMF.91 In order to increase the rate of the reaction which directly depend on the concentration of the catalyst, typically, a high excess of copper(II) salt is used. The oxidative coupling of diethynylated phenylene-butadiyne under Hay conditions

(CuCl/O2/TMEDA) did not generated any cyclic products. In contrast, the cyclization of the same precursor under Eglinton conditions gave the cyclic tetramer C6 and cyclic octamer C7 in 25 % and 17 % yield, respectively (Scheme 3.13).92 Interesting, the cyclic hexamer which would be the thermodynamically most favoured structure was not formed.

3.4 Alkynyl-alkynyl coupling reactions 75

H R R R

Cu(OAc)2 / CuCl

R R R H n

R = tBu R C6 n = 0 [25%] C7 n = 2 [13%] Scheme 3.13. Synthesis of shape-persistent phenylene-diacetylene macrocycles through Eglinton coupling.

In general, there is little information and mostly no predictability concerning the formation of cyclic versus linear products by oxidative coupling reactions. Furthermore, the effect of the different reaction conditions (catalyst, solvent, temperature) on the size distribution of cyclic oligomers is difficult to predict and the optimal conditions have to be determined empirically for each individual case. Although, best results in macrocycles synthesis were obtained by the Eglinton method, it is more advisable to try also other modifications in order to find out which works the best for a given application. As an example, the synthesis of functionalized perethynylated expanded radialenes reported by Diederich et al is illustrated in Scheme 3.14.93 The deprotected tetraethynylene precursor TEE 1 yielded a mixture of all-carbon core containing macrocycles C46a-c by Hay coupling under dilute conditions. In contrast, the macrocyclization of the similar precursor TEE 2 but containing (4’-cyanobiphenyl-4-yloxy)undecyl groups attached to the corresponding macrocycles C47a-c was found to lead to better results when reacted under Eglinton conditions.

76 Chapter 3 Shape-persistent nanosized macrocycles

R R

SiPr3

1) Bu4NF / THF 2) CuCl, TMEDA R R O2, CH2Cl2 n SiPr3

R = N(C H ) R 12 25 2 TEE1 R R C46a n = 1, [8%] C46b n = 2, [10%] C46c n = 3, [2%]

R R

SiPr3

1) Bu4NF / THF R 2) Cu(OAC)2 R pyridine/ benzene n SiPr3

R R R TEE2 C47a n = 1, [4%] C47b n = 2, [56%] C47c n = 3, [4%]

R= O(CH2)11O CN

Scheme 3.14. Synthesis of perethynylated expanded radialenes by different variations of the oxidative coupling reaction.

3.5 Conclusion

As illustrated in this chapter, within the last years significant developments have taken place in the synthesis of shape-persistent macrocycles. A set of versatile and efficient synthetic strategies and methods have provided the potential to prepare rigid macrocycles to meet diverse design requirements. The highly interesting inherent properties and applications of shape-persistent macrocycles clearly demonstrate the major scientific and

3.4 Alkynyl-alkynyl coupling reactions 77

technological importance of their investigations. However, equally clear is that the controlled synthesis on a practical scale of these cyclic structures has yet to be realized. The key to resolve this issue lies in the improvement of the synthesis and in the development of new strategies which in the end should be very fruitful for the entire area.

3.6 References and notes

2 See, for example: H. Keul, H. Höcker, Cycloalkanes and Related Oligomers and Polymers, In Large Ring Molecules, Ed.: J.A. Semylen, H. Wiley & Sons, Chichester, 1996, Chapter 10, pp. 375-406.

3 a) H.A. Sherega, Cyclic Peptides and Loops in Proteins, In Large Ring Molecules, Ed.: J.A. Semylen, H. Wiley & Sons, Chichester, 1996, Chapter 3, pp. 99-112; b) J.D. Hartgering, T.D. Clark, M.R. Ghadiri, Chem. Eur. J. 1998, 4, 1367-1372; c) J.M. Sanderson, S. Yazdani, Chem. Commun, 2002, 1154-1155.

4 a) J.L. Atwood, J.E.D. Davies, D.D. Macniol, Ed.: F. Vögtle, Cyclodextrines, In Comprehensive Supramolecular Chemistry, Vol. 3, Elsevier Science Ltd., Oxford 1996; b) A. Harada, Cyclodextrines, In Large Ring Molecules, Ed.: J.A. Semylen, H. Wiley & Sons, Chichester, 1996, Chapter 11, pp. 407-432; c) M. Wakao, K. Fukase, S. Kusumoto, J. Org. Chem. 2002, 67, 8182-8190.

5 a) M. Fujita, Chem. Soc. Rev. 1998, 27, 417-425; b) S. Leininger, B. Olenyuk, P.J. Stang, Chem. Rev. 2000, 100, 853-908; c) M. Schweiger, S.R. Seidel, A.M. Arif, P.J. Stang, Inorg. Chem. 2002, 41, 2556-2559; d) R. Takahashi, Y. Kobuke, J. Am. Chem. Soc. 2003, 125, 2372-2373.

6 For a selection of the most recent examples: a) C. Grave, D. Lentz, A. Schäfer, P. Samori, J.P. Rabe, P. Franke, A.D. Schlüter, J. Am. Chem. Soc. 2003, 125, 6907-6918. b) M Fischer, G. Lieser, A. Rapp, I. Schnell, W. Mamdouh, S. de Feyter, F.C. Schryver, S. Höger, J. Am. Chem. Soc. 2004,.214-222; c) K. Campbell, R. Mc.Donald, M.J. Ferguson, R.R Tykwinski, Organometallics, 2003, 22, 1353-1355; d) M Srinivasan, S. Sankararaman, H. Hopf, B. Varghese, Eur. J. Org. Chem. 2003; 660-665; e) X. Shen, D.M. Ho, R.A. Jr. Pascal, Org. Lett. 2003, 5, 369-371; f) M. Higuchi, H. Kanazawa, K. Yamamoto, Org. Lett. 2003, 5, 345-347; g) P. Bäuerle, E. Mena-Osteritz, G. Fuhrmann, A. Kaiser, M. Ammann, Polym. Mater. Sci. Eng. 2002, 86, 34-37; h) P.N.W. Baxter, Chem. Eur. J. 2002, 8, 5250-5264; i) M. Schmittel, A. Ganz, D. fenske, Org. Lett. 2002, 4, 2289-2292.

7 J.S. Moore, J. Zhang, Angew. Chem.. 1992, 104, 922-923.

8 R. Friederich, M. Nieger, F. Vögtle, Chem. Ber. 1993, 126, 1723-1732.

9 M.M. Haley, B.L. Langsdorf, Chem. Commun. 1997, 1121-1122.

78 Chapter 3 Shape-persistent nanosized macrocycles

10 T. Kawase, H.R. Darabi, M. Oda, Angew. Chem. Int. Ed. Engl. 1996, 35, 2664-2666; b) T. Kawase, N. Ueda, K. Tanaka, Y. Seirai, M. Oda, Tetrahedron Lett. 2001, 42, 5509-5511.

11 M. Mayor, C. Didschies, Angew. Chem. Int. Ed. Engl. 2003, 42, 3176-3179.

12 a) D.H. Cobden, Nature 2001, 409, 32-33; b) D.T. Bong, T.D. Clark, J.R. Granja, M.R. Ghadiri, Angew. Chem. Int. Ed. Engl. 2001, 40, 988-1011.

13 D. Venkataramen, S. Lee, J. Zhang, J.S. Moore, Nature, 1994, 371, 591-593.

14 For the concept of intraanular functional groups, see: E. Weber, F. Vögtle, Chem. Ber. 1976, 109, 1803- 1831.

15 a) C.J. Walter, H.L. Anderson, J.K.M. Sanders, J. Chem. Soc. Chem. Commun. 1993, 458-460; b) L.G. Mackay, R.S. Wylie, J.K.M. Sanders, J. Am. Chem. Soc. 1994, 116, 3141-3142.

16 a) F. Sondheimer, R. Wolovsky, Tetrahedron Lett. 1959, 3, 3-7; b) F. Sondheimer, R. Wolovsky, R. Amiel, J. Am. Chem. Soc. 1962, 84, 274-276.

17 F. Sondheimer, Acc. Chem. Res. 1972, 5, 81-91.

18 a) G.M. Badger, J.A. Elix, G.E. Lewis, Aust. J. Chem. 1965, 25, 70-89; b) T. Kauffman, Angew. Chem. 1979, 91, 1-19; c) L. Latos-Grazynski, J. Lisowski, M.M. Olmstead, K.L. Balch, J. Am. Chem. Soc. 1987, 109, 4428-4429; d) J.L. Sessler, M. Cyr, A.K. Burrel, Tetrahedron 1992, 48, 661-672; e) Z. Hu, J.L. Atwood, M.P. Cava, J. Org. Chem. 1994, 59, 8071-8075; f) J.L. Sessler, A. Gebauer, S.J. Weghorn, In The Porphirin Handbook, Vol.2, Eds.: K.M. Kadish, K.M. Smith, R. Guilard, Academic Press, San Diego 2000, pp. 50; g) V.G. Anand, S. Ventrakatman, H. Rath, T.K. Chandreskar, W. Teng, K. Ruhlandt-Senge, Chem. Eur. J. 2003, 9, 2282-2290.

19 a) H. Meier, Synthesis, 1972, 235-253; b) M. Nakagawa, Pure Appl. Chem. 1975, 44, 885-889; c) N. Huang, F. Sondheimer, Acc. Chem. Res. 1982, 15, 96; d) S. Akiyama, Anal. Sci, 1994, 10, 365; e) Q. Zhou, P.C. Swager, J. Org. Chem. 1994, 59, 1294-1301; f) F. Diederich, Nature 1994, 199-207; g) W.J. Youngs, C.A. Tessier, J.D. Brandsaw, Chem. Rev. 1999, 99, 3153-3180; h) M.M. Haley, J.J. Pak, S.C. Brand, Top. Curr. Chem. 1999, 201, 81-130.

20 a) P.J. Garatt, Aromaticity, Wiley-Interscience, New York 1986; b) A.T. Balaban, V. Ciorba, Annulenes, Benzo-, Hetero-, Homo-Derivatives, and their Valence Isomers, CRC Press, Boca Raton, B. ThulinFL, 1987.

21 a) Conjugated Polymers and Related Materials: The Interconnection of Chemical and Electronic Structure, Eds.: W.R. Salaneck, I. Lundström, B. Ranby, Oxford University Press, Oxford, U.K., 1993; b) Photonic and Optoelectronic Polymers, Eds.: S.A. Jenekhe, K.J. Wynne, American Chemical Society, Washington, DC, 1995; c) Electronic Materials, The Oligomer Approach, Eds.: K. Müllen, G. Wegner, Wiley-VCH, Weinheim, Germany, 1998; d) M.M. Haley, W.B. Wan, In Advances in Strained and Interesting Organic Molecules, Ed.: B. Halton, Vol. 8, JAI Press, Greenwich, 2000, pp. 1-41; d) V. Balzani, A. Credi, M. Venturi, Molecular Devices and Machines – A Journay into the Nanoworld, Wiley – VCH, Weinheim, 2003.

22 a) J. Zhang, J.S. Moore, J. Am. Chem. Soc. 1992, 114, 9071-9072; b) J. Zhang, D.J. Pesak, J.L. Ludwick, J.S. Moore, J. Am. Chem. Soc. 1994, 116, 4227-4239; c) A.S. Shetty, J. Zhang, J.S. Moore, J. Am. Chem. Soc. 1996, 118, 1019-1027

23 a) H.L. Anderson, J.K.M. Sanders, Angew. Chem. Int. Ed. Engl. 1990, 29, 1400-1403; b) A.M. Boldi, F. Diederich, Angew. Chem. Int. Ed. Engl. 1994, 33, 468-471; c) A. de Meijere, S. Kozhushkov, T: Haumann, R. Boese, C. Puls, M.J. Cooney, L.T. Scott, Chem. Eur. J. 1995, 1, 124-131; d) S. Höger, A.-D. Mechenstock, H. Pellen, J. Org. Chem. 1997, 62, 4556-4667; e) Y. Tobe, N. Utsumi, A. Nagano, K. Naemura, Angew. Chem. Int. Ed. Engl. 1998, 37, 1285-1287.

3.6 References and notes 79

24 a) V.Hensel, K. Lützow, J. Jakob, K. Gessler, W. Saenger, A.D. Schlüter, Angew. Chem. Int. Ed. Engl. 1997, 36, 2654-2656; b) V. Hensel, A.D. Schlüter, Chem. Eur. J. 1999, 5, 421-429.

25 M. Mayor, C. Didschies, Angew. Chem. Int. Ed. Engl. 2003, 42, 3176-3179.

26 a) C.A. Hunter, K.R. Lawson, J. Perkins, C.J. Urch, J. Chem. Soc., Perkin Trans. 2, 2001, 651; b) M. Pickholz, S. Strafstöm, Chem. Phys. 2001, 270, 245-251; c) C.A. Hunter, J.K.M. Sanders, J. Am. Chem. Soc., 1990, 112, 5773-5780.

27 S. Lahiri, J.S. Thompson, J.S. Moore, J. Am. Chem. Soc. 2000, 122, 11315-11319.

28 S. Höger, K. Bonrad, A. Mourran, U. Beginn, M. Möller, J. Am. Chem. Soc. 2001, 123, 5651-5659.

29 Y. Tobe, N. Utsumi, K. Kawabata, A. Nagano, K. Adachi, S. Araki, M. Sonoda, K. Hirose, K. Naemura, J. Am. Chem. Soc. 2002, 124, 5350-5364.

30 H. Plenio, Angew. Chem. Int. Ed. Engl. 1997, 36, 348-350.

31 P. Müller, I. Uson, V. Hensel, A.D. Schlüter, G.M. Scheldrick, Helvetica Chimica Acta 2001, 84, 778-785.

32 D. Venkataraman, S. Lee, J. Zhang, J.S. Moore, Nature, 1994, 371, 591-593; b) Y. Tobe, N. Utsumi, K. Kawabata, A. Nagano, K. Adachi, S. Araki, M. Sonoda, K. Hirose, K. Naemura, J. Am. Chem. Soc. 2002, 124, 5350-5364.

33 S. Höger, D.L. Morrison, V. Enkelmann, J. Am. Chem. Soc. 2002, 124, 6734-6736.

34 a) O. Henze, D. Lentz, A.D. Schlüter, Chem. Eur. J. 2000, 6, 2362-2367; b) O. Henze, D. Lentz, A. Schäfer, P. Franke, A.D. Schlüter, Chem. Eur. J. 2002, 8, 357365.

35 a) G. Binnig, H. Rohrer, C. Gerber, E. Weibe, Phys. Rev. Lett. 1982, 49, 57-61; b) G. Binnig, H. Rohrer, IBM J. Res. Dev. 1986, 30, 355-358; c) J.P. Rabe, S. Buchholz, Phys. Rev. Lett. 1991, 66, 2096-3000.

36 S. Chandrasekhar, B.K. Sadhashiva, K.A. Suresha, Pranama, 1977, 7, 471-480.

37 S. Kumar, S.K. Varshney, Angew. Chem. Int. Ed. Engl. 2000, 39, 3140-3142.

38 D. Adam, P. Schuhmacher, J. Simmerer, L. Haüssling, K. Siemensmeyer, K.H. Etzbach, H. Ringsdorf, D. Haarer, Nature 1994, 371, 141-143.

39 a) S. Ito, M. Wehmeier, J.D. Brand, C. Kübel, R. Epsch, J.P. Rabe, K. Müllen, Chem. Eur. J. 2000, 6, 4327-4342; b) L. Schmidt-Mende, A. Fectenkötter, K. Müllen, E. Moons, R.H. Friends, J.D. MacKenzie, Science 2001, 239, 1119-1122.

40 a) J. Zhang, J.S. Moore, J. Am. Chem. Soc. 1994, 116, 2655-2658; b) O.Y. Mindyuk, M.R. Stetzer, P.A. Heiney, J.C. Nelson, J.S. Moore, Adv. Mater. 1998, 10, 1363-1366.

41 S. Höger, V. Enkelmann, K. Borad, C. Tschierske, Angew. Chem. Int. Ed. Engl. 2000, 39, 2267-2270.

42 Y. Tobe, A. Nagano, K. Kawabata, M. Sonoda, K. Naemura, Org. Lett. 2000, 2, 3265-3268.

43 U. Lehmann, A.D. Schlüter, Chem. Eur. J. 2000, 3483-3487.

44 B.H. Hong, S.C. Bae, C.-W. Lee, S. Jeong, K.S. Kim, Science 2001, 294, 348-351; b) M.S. Gudiksen, L.J. Lauhorn, J. Wang, D.C. Smith, C.M. Lieber, Nature 2002, 415, 617-620.

45 a) J.K.M. Sanders in The Porphyrin Handbook, Eds.: K.M. Kadish, K.M. Smith, R. Guilard, Academic Press, San Diego, CA, 2000, Vol. 3, pp 348-368; b) H.L. Anderson, J.K.M. Sanders, J. Chem. Soc. Chem. Commun. 1989, 1714-1715; c) H.L. Anderson, J.K.M. Sanders, Angew. Chem. Int. Ed. Engl. 1990, 29, 1400- 1403; d) H.L. Anderson, J.K.M. Sanders, J. Chem. Soc. Chem. Commun. 1992, 946-947; e) S. Anderson,

80 Chapter 3 Shape-persistent nanosized macrocycles

H.L. Anderson, J.K.M. Sanders, Acc. Chem. Res. 1993, 469-475; f) S. Anderson, H.L. Anderson, A. Barshall, M. McPartlin, J.K.M. Sanders, Angew. Chem. Int. Ed. Engl. 1995, 34, 1096-1099.

46 a) J. Li, A. Ambroise, S.I. Yang, J.R. Diers, J. Seth, C.R. Wack, D.F. Bocian, D. Holden, J.S. Lindsey, J. Am. Chem. Soc. 1999, 121, 8927-8940; b) A. Ambroise, J. Li, L. Yu, J.S. Lindsey, Org. Lett., 2000, 2, 2563- 2566; c) K. Tomizaki, L. Yu, L. Wei, D. Bocian, J.S. Lindsey, J. Org. Chem. 2003, 68, 8199-8207.

47 a) K. Campbell, C.J. Kuehl, M.J. Ferguson, P.J. Stang, R.R. Tykwinski, J. Am. Chem. Soc. 2002, 124, 7266-7267; b) K. Campbell, R. McDonald, R.R. Tykwinsky, J. Org. Chem. 2002, 67, 1133-1140; c) K. Campbell, R. McDonald, N.R. Branda, R.R. Tykwinsky, Org. Lett. 2001, 3, 1045-1048.

48 a) T Kawase, Y. Seirai, H.R. Darabi, M. Oda, Y. Sarakai, K. Tashiro, Angew. Chem. Int. Ed. Engl. 2003, 42, 1621-1624; b) T Kawase, K. Tanaka, N. Fujiwara, H.R. Darabi, M. Oda, Angew. Chem. Int. Ed. Engl. 2003, 42, 1624-1628; c) T Kawase, K. Tanaka, Y Seinai, N Shiano, M. Oda, Angew. Chem. Int. Ed. Engl. 2003, 115, 5755-5758.

49 a) S. Iijima, Nature 1991, 354, 56-58; b) B.W. Smith, M. Monthioux, D.E. Luzzi, Nature 1998, 396, 323- 325.

50 a) H.W. Kroto, K.G. Mckay, Nature 1988, 331, 328-331; b) S. Iijima, J. Cryst. Growth 1989, 50, 675-683; c) D. Ugarte, Nature 1992, 359, 707-709.

51 a) B.W. Smith, M. Monthioux, D.E. Luzzi, Nature 1998, 396, 323-326; b) J. Sloan, A.I. Kirkland, J.L. Hutchison, M.L.H. Green, Chem. Commun. 2002, 1319-1332.

52 a) T. Haino, Y. Yamanaka, H. Araki, Y. Fukazawa, Chem. Commun. 2002, 402-403; b) S.R. Wilson, S. MacMahon, F.T. Tat, P.D. Jarowski, D.I. Schuster, Chem. Commun. 2003, 226-227.

53 H.A. Staab, K. Neunhoeffer, Synthesis 1974, 424.

54 J.S. Moore, J. Zhang, Angew. Chem. Int. Ed. Engl. 1992, 31, 922-924.

55 P. Ruggli, Liebigs Ann. Chem 1912., 392, 92-94.

56 a) K. Ziegler, Houben Weyl, Methoden der Organischen Chemie, Georg Thieme Verlag, Stuttgart, 1955, p. 729, 4th ed., vol. 4/2; b) P. Knorps, N. Sendhoff, H.-B Meckelburger, F. Vögtle, Top. Curr. Chem. 1991, 161, 1-36.

57 a) C. Galli, L. Mandolini, J. Chem. Soc. Chem. Commun. 1982, 251-253; b) G. Illuminati, L. Mandolini, Acc. Chem. Res. 1981, 14, 95-102; c) M.A. Winnik, Chem. Rev. 1981, 81, 491-499.

58 a) S. Höger, V. Enkelmann, Angew. Chem. Int. Ed. Engl. 1995, 34, 2713-2715; b) S. Höger, A.D. Meckenstock, S. Müller, Chem. Eur. J. 1998, 4, 2423-2434.

59 a) Y. Tobe, N. Utsumi, K. Kawabata, K. Naemura, Tetrahedron Lett. 1996, 37, 1285-1287; b) V. Hensel, K. Lützow, J. Jakob, K. Gessler, W. Saenger, A.D. Schlüter, Angew. Chem. Int. Ed. Engl. 1997, 36, 2654- 2656; c) D. Zhao, J.S. Moore, J. Org. Chem. 2002, 67, 3548-3434.

60 S. Höger, A.D. Meckenstock, H. Pellen, J. Org. Chem. 1997, 62, 4556-4562.

61 a) J.-M. Lehn, Science 1985, 227, 849-856; b) M. Mascal, Contemp. Org. Synth. 1994, 1, 31-46; c) J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995; d) Comprehensive Supramolecular Chemistry, Eds.: J.L. Atwood, J.E.D. Davies, D.D. Macniol, F. Vögtle, Pergamon, Oxford, 1996; e) H.-J. Schneider, A. Yatsimirsky, Principle and Methods in Supramolecular Chemistry, Wiley, Chiester, 2000; f) G.R. Desiraju, Nature 2001, 412, 397-400.

62 C.D. Gutsche in Calixarene Revisited, Eds.: J.F. Stoddart, Royal, Society of Chemistry, Cambridge, 1998.

63 C.D. Gutsche, D.E. Johnston, Jr. D.R. Stewart, J.Org. Chem. 1999, 64, 3747-3750.

3.6 References and notes 81

64 a) G.R. Geier III, J.S. Lindsey, J. Chem. Soc. Perkin Trans. 2 2001, 677-686; b) G.R. Geier III, J.S. Lindsey, J. Chem. Soc. Perkin Trans. 2 2001, 687-700; c) G.R. Geier III, B.J. Littler, J.S. Lindsey, J. Chem. Soc. Perkin Trans. 2 2001, 701-711; d) G.R. Geier III, B.J. Littler, J.S. Lindsey, J. Chem. Soc. Perkin Trans. 2 2001, 712-718.

65 a) T.D. Tilley, S.S.H. Mao, J. Am. Chem. Soc. 1995, 117, 7031-7032; b) T.D. Tilley, S.S.H. Mao, J. Am. Chem. Soc. 1995, 117, 5365-5366; c) J.R. Nitschke, T.D. Tilley, J. Org. Chem. 1998, 63, 3673-3676; d) J.R. Nitschke, T.D. Tilley, Angew. Chem. Int. Ed. Engl. 2001, 40, 2142-2145; e) L.L. Schafer, T.D. Tilley, J. Am. Chem. Soc. 2001, 123, 2683-2684; f) J.R. Nitschke, T.D. Tilley, J. Am. Chem. Soc. 2001, 123, 10183-10190; g) L.L. Schafer, J.R. Nitschke, S.S.H. Mao, F.-Q. Liu, G. Harder, M. Haufe, T.D. Tilley, Chem. Eur. J. 2002, 8, 74-83.

66 a) D.H. Busch, J. Inclusion Phenom 1992, 12, 389-395; b) D.H. Busch, N.A. Stephenson, Coord. Chem. Rev. 1990, 100, 119-154.

67 J.D. Watson, F.H.C. Crick, Nature 1953, 171, 737-738.

68 a) D.H. Busch, J.A. Burke Jr., D.C. Jischa, M.C. Thompson, M.L. Morris, Adv. Chem. Ser. 1962, 37, 125- 142; b) D.H. Busch, Adv. Chem. Ser. 1963, 37, 1-18; c) M.C. Thompson, D.H. Busch, J. Am. Chem. Soc 1964, 86, 3651-3656.

69 C.J. Pederson, Angew. Chem. Int. Ed. Engl. 1988, 27, 1021-1027.

70 J.P. Sauvage, Acc. Chem. Res. 1990, 23, 319-327; b) D.B. Amabilino, J.F. Stddart, Chem. Rev. 1995, 95, 2725-2828.

71 D.B. Amabilino, J.F. Stoddart, Chem. Rev. 1995, 95, 2725-2828.

72 T.R. Kelly, G.J. Bridger, . Zhao, J. Am. Chem. Soc. 1990, 112, 8024-8034.

73 G.W. Gokel, S.H. Korzenowski, Macrocyclic Polyether Syntheses, Springer Verlag, Berlin, 1982.

74 a) A. Schanzer, J. Libman, F. Frolow, Acc. Chem. Res. 1983, 16, 60-68; b) E. K. Haas, W. Ponikwar, H. Nöth, W. Beck, Angew. Chem. Int. Ed. Engl. 1998, 37, 1086-1089.

75 G. Gervasio, E. Sappa, A.M.M. Lanfredi, A. Tiripicchio, Inorg. Chim. Acta 1983, 68, 171.

76 a) K. Nakamura, H. okubo, M. Yamaguchi, Org. Lett. 2001, 3, 1097-1099; b) Y. Tobe, T. Fujii, H. Matsumoto, K. Tsumuraya, D. Nogushi, N. Nagawa, M. Sonoda, K. Naemura, Y. Achiba, T. Wakabayashi, J. Am. Chem. Soc. 2000, 122, 1762-1775; c) M. Schmittel, H. Ammon, Synlett 1999, 6, 750-752; d) S. Marujama, H. Hokari, T. Wada, H. Sasabe, Synthesis, 2001, 1794-1799; e) L. Yu, J.S. Lindsey, J. Org. Chem. 2001, 66, 7402-7419.

77 a) S.-S. Sun, Organometallics, 2001, 20, 2352-2358; b) M.B. Nielsen, M. Schreiber, Y.G. Baek, P. Seiler, S. Lecomte, C. Boudon, R.R. Tykwinsky, J.-P. Gisselbrect, V. Gramlich, P.J. Skinner, C. Bosshard, P. Günter, M. Gross, F. Diederich, Chem. Eur. J. 2001, 7, 3263-3280.

78 a) P.-H. Ge, W. Fu, A. Herrmann, E. Herdtweck, C. Campana, R.D. Adams, U.H.F. Bunz, Angew. Chem. Int. Ed. Engl. 2000, 39, 3607-3610; b) U.H.F. Bunz, Acc. Chem. Res. 2001, 34, 998-1010.

79 Y. Hosokawa, T. Kawase, M. Oda, Chem. Commun. 2001, 1948-1949.

80 a) P. Siemsen, R.C. Livingston, F. Diederich, Angew. Chem. Int. Ed. Engl. 2000, 39, 2632-2657; b) J. Breitenbach, J. Boosfeld, F. Vögtle in Comprehensive Supramolecular Chemistry, Vol. 2; Eds.: F. Vögtle, Pergamon, Oxford, 1996, Chapter 2, pp 29-67; c) P.J. Stang, F. Diederich, Modern Acetylene Chemistry, VCH, Weinheim, 1995; d) L. Brandsma, Preparative Acetylenic Chemistry, Elsevier, 1988.

82 Chapter 3 Shape-persistent nanosized macrocycles

81 a) R.E. Martin, F. Diederich, Angew. Chem. Int. Ed. Engl. 1999, 38, 1350-1377; b) J.M. Tour, Acc. Chem. Res. 2000, 33, 791-804; c) U.H.F. Bunz, Chem. Rev. 2000, 100, 1605-1644; d) D.T. McQuade, A.E. Pullen. T.M. Swager, Chem. Rev. 2000, 100, 2537-2574.

82 a) J.S. Moore, Acc. Chem. Res. 1997, 30, 402-413; b) S. Hecht, J.M.J. Frechet, Angew. Chem. Int. Ed. Engl. 2001, 40, 79-91; c) M.D. Watson, A. Frechtenkötter, K. Müllen, Chem. Rev. 2001, 101, 1267-1300.

83 Y.S. Zalkind, B.M. Fundyler, Ber. Dtsch. Chem. Ges. 1936, 69, 128-130.

84 A.L. Klebansky, I.V Grachev, O.M. Kuznetsova, J. Gen. Chem. USSR 1957, 27, 3008-3013.

85 F. Bohlmann, H. Schönowsky, E. Inhoffen, G. Grau, Chem. Ber. 1964, 97, 794-800.

86 a) L.G. Fedenok, V.M. Berdnikov, M.S. Shvartsberg, J. Org. Chem. USSR, 1974, 10, 934-936; b) H.J. Kevelam, K.P. de Jong, H.C. Meinders, G. Challa, Makromol. Chem. 1975, 176, 1369-1381; c) L.G. Fedenok, V.M. Berdnikov, M.S. Shvartsberg, J. Org. Chem. USSR, 1976, 12, 1385-1387; d) G. Challa, H.C. Meinders, J. Mol. Catal. 1977, 3, 185-190; e) L.G. Fedenok, V.M. Berdnikov, M.S. Shvartsberg, J. Org. Chem. USSR, 1978, 14, 1328-1333; f) J. C. Kennedy, J.R. MacCallum, D.H. MacKerron, Can. J. Chem. 1995, 73, 1914-1923.

87 a) H.M. Hoan, S.M. Brailovskii, O.N. Temkin, Kinet. Catal. (Engl. Transl.) 1994, 35, 242-246; b) ) H.M. Hoan, S.M. Brailovskii, O.N. Temkin, Kinet. Catal. (Engl. Transl.) 1994, 35, 431-435.

88 G. Eglinton, A.R. Galbraith, Chem. Int. (London) 1956, 737-738.

89 G. Eglinton, A.R. Galbraith, J. Chem. Soc. 1959, 889-896.

90 R. Berscheid, F. Vögtle, Synthesis, 1992, 58-62.

91 J.J. Pak, T.J.R. Weakly, M.M. Haley, J. Am. Chem. Soc. 1999, 121, 8182-8192.

92 Y. Tobe, N. Utsumi, A. Nagano, M. Sonoda, K. Naemura, Tetrahedron 2001, 57, 8075-8083.

93 J. Anthony, A.M. Boldi, C. Boudon, J.-P. Gisselbrecht, M. Gross, P. Seiler, C.B. Knobler, F. Diederich, Helv. Chim. Acta 1995, 78, 797-817.

Chapter 4

Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Abstract

In this chapter studies towards synthesis and characterization of fully conjugated macrocyclic oligothiophene-diacetylenes are disclosed. Firstly, the preparation and characterization of butyl-substituted conjugated oligothiophenes which are used as starting materials is described. The synthetic approach towards these oligomers is based on the application of various metal-mediated coupling reactions involving a novel palladium-mediated dimerization of ethynyl-oligothiophenes. In addition, the ring-closure reaction of butadiynes with sulfide nucleophiles to the corresponding thiophenes has been optimized and a proposed mechanism is presented. Finally, macrocyclization of diethynylated oligothiophenes by oxidative coupling reaction and detailed aspects of these reactions are discussed. Two homologues series of conjugated oligothiophene macrocycles have been synthesized and characterized. Furthermore, by using a terminal diethynylated undecithiophene as the monomeric precursor the intramolecular cyclic formation has been investigated.

84 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

4.1 Cyclo[n]thiophene target structure

As stated in Chapter 2, polythiophenes and their finite model oligomers, α-conjugated oligothiophenes, belong to the most thoroughly investigated π-conjugated systems in the field of material science.1 Due to their chemical stabilities in their various redox forms, their outstanding electronic properties and the widespread possibilities of functionalization, they are potential candidates for applications in molecular electronic devices, such as organic field-effect transistors,2 light-emitting devices,3 photovoltaic cells4 or even as molecular wires for information storage or transfer.5 Since oligomers can be prepared with a well-defined molecular length and structure, they are furthermore excellent models for their corresponding polymers, providing specific information that is necessary to design polymers with improved material performance. In numerous investigated series of monodisperse linear oligothiophenes the physical properties are well correlated to the structural parameters, such as chain and conjugation length of the oligomers, and thus, valuable information concerning structure-property relationships of these π-conjugated systems become accessible.Error! Bookmark not defined.a However, sometimes the physical properties of such well-defined oligomers are influenced by undesired perturbing end- effects of the conjugated chain.6 Model systems with much more appealing structural characteristics would be corresponding fully π-conjugated macrocycles. In comparison to usual linear conjugated oligomers and polymers, they would have the distinct advantage to ideally combine an ”infinite” defect-free π-conjugated chain of an idealized polymer with the advantages of a structurally well-defined oligomer, but excluding perturbing end- effects.

STM investigations undertaken on self-assembled monolayers of regioregular poly(3- alkylthiophene)s illustrated the occurrence of an intramolecular chain folding which is composed of seven to eight thiophenes in an all-cis conformation, forming a 180° semicircle.7 Motivated by these interesting intrinsic self-assembly properties of polythiophenes the challenging objective of this study was the development of an efficient synthetic route towards the synthesis of fully π-conjugated α-linked macrocyclic oligothiophenes, which are designated as cyclo[n]thiophenes (n = number of thiophene rings). The feasibility and importance of such macrocycles was further supported by theoretical calculations which have already been carried out in 1995 by Tol8 on cyclo[12]thiophene and of its corresponding charged species. For cyclo[12]thiophene the

4.1 Cyclo[n]thiophene target structure 85

calculations revealed a nearly non-strained and coplanar structure with an all-cis conformation of the thiophene rings. Moreover, as thoroughly discussed in Chapter 3, shape-persistent macrocycles are currently important synthetic targets with potential applications in the field of organic and material science.9 Typically, in order to favour their cyclic structure formation angular building blocks, such as meta-substituted aromatic corners, are introduced. As an undesired effect of these angular corner pieces, an electronic interruption on the overall conjugation results. Among all these macrocycles with rigid and non-collapsible backbones exhibiting highly interesting properties, cyclo[n]thiophenes are therefore expected to represent a completely novel class of cyclic derivatives with particularly appealing characteristics and perspectives. Due to their toroidal structure with fully π-conjugated periphery they could act as intriguing “molecular circuits” which additionally would include sites for selective recognition and selective complexation (Figure 4.1).

S S S

S S

S S S

Figure 4.1. Cyclo[12]thiophene as an example of a cyclo[n]thiophene target structure.

The synthesis of macrocycles containing only thiophene rings in their backbone was already reported in 1978 by Kauffmann et al.10 Cross-conjugated macrocycles C48-C51 were prepared by oxidative coupling reactions of lithiated bithiophene precursors (Scheme 4.1). However, due to the interruption of the electronic pathway which is caused by α,β- or β,β-linkages of the thiophene units, these macrocycles are, expectantly, only very weakly conjugated. For example, macrocycle C48 has an absorption maximum of λmax= 278 nm and thus it does not exhibit more conjugation than a bithiophene-unit (λmax= 302 nm for

86 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

2,2’-bithiophene; λmax= 263 nm for 3,3’-bithiophene). Furthermore, the macrocycles are non-planar with the thiophene-rings being twisted to each other as confirmed by X-ray structure analysis of cyclic C50.

S S S S S n-BuLi Br n-BuLi S S CuCl Br CuCl2 2 S S S S S S S

C48 [23%] C49 [28%]

S S S S S S Br n-BuLi Br n-BuLi S Br FeCl Br CuCl 3 S 2 S S S S S S C50 [26%] C51 [4%]

Scheme 4.1. Thiophene-based cross-conjugated macrocycles synthesized by Kauffmann et al.

[18]Annulene which was reported by Sondheimer et al.11 in 1959 has been regarded as a fundamental structure in aromatic chemistry. It complies with Hückel’s theory and by its 1H NMR data clearly demonstrates the existence of an aromatic ring current. A great deal of effort has then been made towards synthesis of structural variants of annulenes containing heteroatom bridges, as illustrated in Figure 4.2. Compared to the numerous studies on nitrogen bridged annulenes such as porphyrins, sulphur bridged analogues have been much less investigated.

NH N S

N HN S S

Figure 4.2. [18]Annulenes and corresponding nitrogen and sulphur analogues.

Several neutral and cationic thiophene-derived annulenes have been prepared by Cava et al. where they used McMurry coupling reactions as the key synthetic step.12 The partial replacement of nitrogen atoms in porphyrin systems by sulphur atoms was proven to result

4.1 Cyclo[n]thiophene target structure 87

in a longer wavelength absorption in the UV-spectrum. However, among all these sulphur- bridged annulene compounds only a few shows a peripheral aromatic ring current. Even though the thiophene analogue of [18]annulene is a [4n+2]π system, there is no significant ring current observable in its spectroscopic data (Figure 4.2). Its X-ray structure analysis confirmed the non-planarity of the cyclic structure with thiophene units being totally out of plane. Thus, this three sulphur-bridged annulene may be viewed as a combination of independent thiophene and vinylene units rather than as a conjugated annulene. In the same way, no antiaromatic ring current could be observed in its next higher homologue, [24]annulene tetrasulfide. The first neutral thiophene-derived annulene that showed peripheral aromatic ring current is the tetrathia[22]-annulene[2,1,2,1] C52. The synthesis of this 22 π system is depicted in Scheme 4.2. In the 1H NMR spectrum of the corresponding macrocyclic precursor of C52 no protons are observed below 6.8 ppm. In contrast, all protons of annulene C52 are greatly deshielded ranging from 10.9 to 12.3 ppm. The strong absorption maximum at 417 nm and several weaker absorptions at longer wavelengths (503, 540, 579, 771 nm) correspond to the typical absorption bands of porphyrins.

S S S S

TiCl4 / Zn DDQ; hydrazine S S [75%] [82%] OHC CHO S S S S

C52

Scheme 4.2. The first neutral aromatic thiophene-derived annulene reported by Cava et al.

4.2 Synthetic strategy towards cyclo[n]thiophenes

First attempts to prepare cyclo[n]thiophenes were undertaken by Fisher et al. by using metal-catalyzed cross-coupling reactions.13 A series of linear alkyl-substituted oligothiophenes have effectively been prepared by employing Ni(0)-mediated reductive coupling of brominated corresponding precursors. Fisher et al. applied the same method for the macrocyclization of a dibrominated octithiophene (Scheme 4.3). Although high dilution conditions were employed in order to favour macrocyclization, besides de-

88 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

brominated starting material, the reaction generated only insoluble polymeric material. An important aspect that has to be considered is certainly the high ring strain of the expected cyclo[8]thiophene which directed its formation rather difficult, even impossible.

H25C12 H25C12

Br S S S S S S S S Br

C12H25 C12H25

NiCl2 / Zn / PPh3

C12H25 H25C12 S S S S

S S S S

H C C12H25 25 12

Scheme 4.3. First attempt towards synthesis of cyclo[8]thiophene.

Another versatile and very convenient approach used for the synthesis of oligothiophenes is the ring closure reaction of oligothienyl-diacetylenes with sulfide nucleophiles. This reaction which was first applied for the preparation of homologous series of α-linked oligothiophenes by Kagan et al,14 has been presented and thoroughly discussed in Chapter 2. For the synthesis of shape-persistent macrocycles incorporating diacetylene-units in the framework, the most applied reaction is the oxidative coupling of diethynylated precursor building blocks.9 The scope and limitation of this method with respect to macrocycle synthesis have been illustrated in Chapter 3.4. Based on these considerations, the key starting materials for the successful synthesis of the targeted cyclo[n]thiophenes are the appropriate diethynyl-functionalized oligothiophenes. As illustrated in Figure 4.3, the oxidative coupling of these monomer building blocks should result in the formation of the conjugated oligothiophene-diacetylene macrocycles, which are referred to as C[mT-DA]n (m is number of thiophenes in one subunit; n is the number of diethynyl-oligothiophene subunits). By a subsequent reaction with sulfide it should be possible to convert them to the desired macrocyclic cyclo[n]thiophenes.

4.2 Synthetic strategy towards cyclo[n]thiophenes 89

C[n]T S S S

ring closure reaction with sulfide nucleophiles „ S2- “

C[mT-DA]n S S

sp-sp homocoupling

R of terminal acetylenes

H S n S H H

Figure 4.3. Retrosynthesis towards cyclo[n]thiophenes.

Initial synthetic efforts based on this approach focused on the oxidative coupling of the easily accessible building block, 3,4-dibutyl-2,5-diethynylthiophene. The oxidative coupling of this monomer has been carried out under Glaser, Hay and Eglinton conditions by applying high dilution conditions. In all cases a complex mixture of linear, cyclic and polymeric materials was obtained (Scheme 4.4).15 Even though 1H NMR spectroscopy and MALDI-TOF mass spectrometry analysis of the crude product has indicated the existence of the targeted cyclic compounds, all attempts to isolate a specific product from the complex mixtures failed.

Bu Bu Bu Bu

S n S CuCl / CuCl2 H H Bu Bu or CuCl, TMEDA, O2 or Cu(OAc) 2 + H S H high dilution Bu Bu

S n

Bu Bu Scheme 4.4. Macrocyclization of 3,4-dibutyl-2,5-diethynylthiophene by applying Glaser, Hay or Eglinton conditions.

90 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

S n S CuCl / CuCl 2 H H or CuCl, TMEDA, O2

or Cu(OAc)2 + S H H high dilution

S n

Scheme 4.5. Oxidative coupling of 3,4-diphenyl-2,5-diethynylthiophene by applying Glaser-, Hay-, or Eglinton-conditions.

As shown in Scheme 4.5, the same unsatisfactory results were obtained when 3,4- diphenyl-2,5-diethynylthiophene was used as precursor.16 Here, the phenyl groups as sterical demanding substituents were expected to favour the cyclization reaction against the formation of linear products improving yields of the cyclic products. However, oxidative coupling reactions carried out under various conditions (Glaser-, Hay-, Eglinton- conditions) generated again only complex mixtures of products including linear, cyclic and polymeric material. A separation of the product mixture proved to be elusive.

An important aspect in macrocyclization reactions is the number of precursor molecules that are involved in the cyclization step. According to statistics, the higher the number of units that have to react with each other forming a cyclic structure, the lower the yield will be. In general, an increase in the cyclization efficiency can be obtained by using enlarged precursor molecules. As a result, the number of possible oligomeric structures is reduced and the product purification should be simplified. In a subsequent project, the macrocyclization of diethynylated terthiophene by oxidative coupling reaction under pseudo high-dilution conditions has been investigated.17 The reaction in which Cu (I) and Cu(II) chlorides were used as catalysts in pyridine as solvent, resulted in the formation of a product mixture containing, besides linear oligomeric and polymeric material, also the

4.2 Synthetic strategy towards cyclo[n]thiophenes 91

cyclotrimeric terthiophene-diacetylene and the corresponding cyclotetramer in 4% and 6% yields, respectively (Scheme 4.6).

Bu Bu

S S Bu Bu S Bu Bu S Bu Bu Bu Bu

S CuCl2 / CuCl S S S H H pyridine / rt pseudo high dilution S Bu S Bu

Bu S S Bu n

Bu Bu

n = 1 [4%] n = 2 [6%] Scheme 4.6. Oxidative coupling reaction of diethynylalted terthiophene under pseudo high-dilution conditions.

Bu Bu Bu Bu Bu Bu

S S S S S H H

CuCl2 / CuCl pyridine / rt pseudo high dilution

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S n Bu Bu Bu Bu

n = 2 [1.8%] n = 3 [2.2%] Scheme 4.6. Oxidative coupling reaction of diethynyl-quinquethiophene under pseudo high-dilution conditions.

92 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

In an analogous procedure, the cyclization of a terminally diethynylated quinquethiophene afforded the formation of the trimeric and tetrameric quinquethiophene-diacetylene macrocycles in 1.8 % and 2.2 % yield, respectively (Scheme 4.6).18

In a subsequent reaction, oligothiophene-diacetylene macrocycles were treated with sodium sulfide to generate the corresponding fully α-conjugated cyclo[12]thiophene, cyclo[16]thiophene and cyclo[18]thiophene in 23%, 7% and 27 % yield, respectively (Scheme 4.7).19

Bu Bu Bu Bu

S Bu S Bu S Bu m S S Bu S Bu Bu Bu S S S Bu

Na2S x 9 H2O S S S

S Bu S Bu S S Bu Bu

m Bu S S Bu m S p S Bu S Bu n

Bu Bu Bu Bu

m = 1, n = 1,2 C[12]T p = 1 [23%] m = 2, n = 1 C[16]T p = 3 [7%] C[18]T p = 4 [27%] Scheme 4.7. Synthesis of fully conjugated cyclo[n]thiophenes with n = 12, 16, 18.

Motivated by the results presented above and with the aim to give access to larger quantities of oligothiophene-derived macrocycles, the oxidative coupling reaction of the appropriate diethynylated oligothiophenes has been further investigated in detail.

4.3 Synthesis of α,ω-diethynylated oligothiophenes

The precursor building blocks for the cyclization reactions are α,ω-diethynylated oligothiophenes. As mentioned earlier, the introduction of flexible alkyl groups in the core of oligothiophenes is necessary not only to provide sufficient solubility to the linear precursors, but also to the final macrocycles. An effective synthetic route for the

4.3 Synthesis of α,ω-diethynylated oligothiophenes 93

preparation of ß-butyl-substituted α-conjugated oligothiophenes has been reported by Krömer et al..20 The method involving a combination of aryl-aryl cross-coupling reactions, dimerization of ethynylated thiophenes and subsequent cyclization of the generated 1,3- butadiyne with sulfide was presented in Chapter 2 (Scheme 2.20). Homologues up to the 11-mer with members having the chain length varying by bithiophene units have been isolated and characterized (Scheme 4.8).

Bu Bu Bu Bu

S S S n

n = 1-5

Scheme 4.8. Homologues series of butylated α-conjugated oligothiophenes reported by Krömer et al.

The symmetrical substitution of each second thiophene ring with alkyl groups has been proven to be ideal to provide sufficient solubility of the compounds, and at the same time, to guarantee the formation of isomerically pure products. Furthermore, the sterical repulsion between individual thiophene rings that usually arise when ß-positions of adjacent the thiophene rings are substituted was minimized by choosing an alternating substitution pattern, as shown in Scheme 4.8. Thus, a distortion of the conjugated backbone that would result in reduced π-conjugation along the chain, increased redox potentials and increased optical band gaps was prevented.21 Based on the approach evaluated by Krömer et al.20 firstly, analogous butylated linear α- conjugated oligothiophenes have been prepared. In a following step, the acetylenic groups have been introduced to the cores by a palladium-promoted coupling reaction of acetylenes and halogenated oligothiophenes. The synthesis of the linear conjugated oligothiophenes and the subsequent preparation of the corresponding ethynylated oligothiophenes is presented in the following section.

4.3.1 Synthesis of conjugated oligothiophenes

The central building block for the synthesis of the oligothiophene series, 3,4- dibutylthiophene 1, was synthesized according to literature procedures (Scheme 4.9).22

94 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Bromination of thiophene led to tetrabromothiophene in 75 % yield which was further reacted with zinc in acetic acid to give 2,5-dibromothiophene in 62 % yield. Finally, 3,4- dibutylthiophene 1 was obtained in 72% yield by a nickel-catalyzed Kumada type cross- coupling reaction of 2,5-dibromothiophene with two equivalents of butyl magnesium bromide.

Br Br Br Br Bu Bu Br / CHCl 2 3 Zn / CH3COOH 2 BuMgBr / Et2O S Ni(dppp)Cl [75%] Br S Br [62%] S 2 S 1 [72%]

Scheme 4.9. Synthesis of the key building block 3,4-dibutylthiophene 1.

The oligothiophene series was synthesized following the same synthetic route reported by Krömer.20 In this approach, the first two homologues in the linear oligothiophene series, terthiophene 4 and quinquethiophene 7 were prepared by the nickel-catalyzed Kumada coupling reaction.23 Monobromination of 3,4-dibutylthiophene 1 was achieved by slow addition of 1.1 equivalents of NBS in DMF under exclusion of light to a solution of 1 at 0°C.24 After distillation under reduced pressure 2-bromo-3,4-dibutylthiophene 2 was isolated as a colorless oil in 82 % yield. The similar reaction of 1 with 2.4 equivalents of NBS gave the corresponding dibrominated thiophene 3 in an isolated yield of 90 % (Scheme 4.10).

Bu Bu Bu Bu Bu Bu 1.1 equiv.NBS 2.4 equiv. NBS DMF S Br S DMF Br S Br 2 [82%] 1 3 [90%]

Scheme 4.10. Selective bromination of thiophenes with NBS/DMF.

3,3’’,4,4’’-Tetrabutylterthiophene 4 was then prepared by a nickel-catalyzed Kumada-type cross-coupling reaction of the Grignard reagent of 2-bromothiophene 2 and commercially available 2,5-dibromothiophene. After purification by chromatography on silica gel, the product could be precipitated by adding methanol to a concentrated dichloromethane solution and was isolated as yellow solid in 84 % yield (Scheme 4.11).

4.3 Synthesis of α,ω-diethynylated oligothiophenes 95

Bu Bu Bu Bu Bu Bu 1. Mg / Et2O + S 2 Br S Br 2. Ni(dppp)Cl / Et O S Br 2 2 S S 2 4 [84%]

Scheme 4.11. Synthesis of tetrabutyl-terthiophene 4 by Kumada-type cross-coupling reaction.

Similarly, coupling of dibromothiophene 3 with an excess of Grignard reagent of commercially available 2-bromothiophene yielded 3’,4’-dibutylterthiophene 5 in 82 %.25 In the next step, 5 was reacted with 2.2 equivalents of NBS in DMF affording the corresponding dibrominated product 6 in 89 % yield (Scheme 4.12).

Bu Bu Bu Bu Bu Bu 1. Mg / Et O 2 NBS 2 + S S Br S S Br Br Br 2. Ni(dppp)Cl S S S Br 2 S DMF Et2O 3 5 [82%] 6 [89%]

Scheme 4.12. Synthesis of dibrominated dibutyl-terthiophene 6.

The subsequent nickel-catalyzed Kumada cross-coupling reaction of difunctionalized building block 6 and the Grignard reagent of bromothiophene 2 led to the formation of hexabutyl-quinquethiophene 7. After chromatography and subsequent precipitation of the product by adding methanol to a concentrated dichloromethane solution, the pure oligomer 7 was isolated in a very good yield of 87 % as orange microcrystalline solid (Scheme 4.13).

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu 1. Mg / Et2O + S S S S 2 Br Br S Br S 2. Ni(dppp)Cl2 / Et2O S S S 2 6 7 [87%]

Scheme 4.13. Synthesis of hexabutyl-quinquethiophene 7 by Kumada cross-coupling reaction.

As discussed in Chapter 2, the Kumada coupling reaction proved to be a less powerful method for the preparation of higher homologues.23 For longer metal organic coupling

96 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

components than bithiophene, the halogen-metal exchange step that leads to the formation of the organometallic reagent occurs less readily. An increased formation of undesired homo-coupling product results in purification problems and gradually decreased yields. Although the length of the oligothiophene chain can still be extended via a stepwise approach by adding mono- or bithiophene units per iteration to the growing oligomer, this approach remains less attractive since then the chain growth proceeds very slowly. A much more effective and rapid synthetic route for the preparation of longer oligothiophenes turned out to be a combined reaction sequence of sp-sp carbon coupling of mono-ethynylated oligothiophenes followed by ring closure of 1,3-butadiyne units with sulfide nucleophiles.20,26 Crucial at this point is the monofunctionalization of the oligothiophene building block. Despite the general difficulty to selectively monobrominate higher oligomers,27 terthiophene 4 was reacted under exclusion of light with 1.1 equivalents of NBS in DMF at 0°C. A slight excess of the reagent is necessary to obtain complete conversion of the educts. By adding a NBS solution to a dilute solution of the oligothiophene as slowly as possible, the formation of dibrominated product can be suppressed, but not completely excluded. However, the separation of the resulting product mixture could be achieved by chromatographic work-up. The desired bromoterthiophene 8 was obtained in 65 % and the corresponding dibrominated product 9 in 18 % isolated yields (Scheme 4.14).

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu 1.1 NBS S S + S S S DMF S S Br Br S S Br

4 8 [65%] 9 [18%]

Scheme 4.14. Bromination of tetrabutylterthiophene 4.

Similarly, monobromination of quinquethiophene 7 with the NBS/DMF system led, as expected, to a mixture of bromoquinquethiophene 10 and the corresponding dibrominated product 11 which were separated by repeated chromatography and isolated as yellow solids in 57 % and 25 % yields, respectively (Scheme 4.15).

4.3 Synthesis of α,ω-diethynylated oligothiophenes 97

Bu Bu Bu Bu Bu Bu

S S S S S Br Bu Bu Bu Bu Bu Bu 10 [57%] S S 1.1 NBS + S S S DMF

7 Bu Bu Bu Bu Bu Bu

S S Br S S S Br

11 [25%]

Scheme 4.15. Bromination of of hexabutyl-quinquethiophene 7.

In the following step, ethynyl groups were introduced to the oligothiophene cores. The reaction was performed by using the standard palladium-catalyzed Sonogashira-Hagihara coupling28 of brominated compounds with trimethylsilylacetylene. Due to the high volatility of the acetylenic reagent, the reaction had to be carried out in a thick-walled Schlenk flask. The conversion of bromoterthiophene 8 and quinquethiophene 10 with 1.2 equivalents of trimethylsilylacetylene led to the TMS-protected 5-ethynyloligothiophenes 12 and 13 which after purification by chromatography on silica gel were isolated in 94 % and 95 % yields, respectively (Scheme 4.16).

Bu Bu Bu Bu Bu Bu Bu Bu H Si(CH3)3 S S n Br n S S Pd(PPh3)Cl2 / PPh3 S S Si(CH3)3 CuI / pyridin / NEt3 8, 10 (n = 1,2) 12, 13 (n = 1,2) [94%, 95%]

Scheme 4.16. Sonogashira-Hagihara coupling reaction for the synthesis of monoethynyl-terthiophene 12 and –quinquethiophene 13.

In general, symmetrical 1,3-butadiynes are easily accessible through oxidative coupling reactions of acetylenes using copper salts as oxidizing agents.29 In the above mentioned work by Krömer, the effective dimerization of ethynylated oligothiophenes 12 and 13 to the corresponding bis(oligothienyl)-1,3-butadiynes was achieved by a Hay-type oxidative coupling reaction.20 The more recently reported palladium-catalyzed dimerization of acetylenes promises to be an attractive alternative to the Glaser coupling reaction due to the simplicity of the procedure, the mildness of the reaction conditions and high

98 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

compatibility with a large variety of functional groups.30 In this thesis, with the aim to prove and extend the utility of this method in the synthesis of oligothiophenes, especially in the presence of various functional groups, first the dimerization of the monoethynylated monothiophene 14 was investigated. Compound 14 was previously prepared from the corresponding 2-bromo3,4-dibutylthiophene 2 by a Sonogashira coupling reaction with trimethylsilylacetylene as depicted in Scheme 4.17. In the next step, the deprotection of the ethynyl group of compound 14 was accomplished under mild basic conditions. The reaction was monitored by TLC and completed after 3 h. Due to the inherent instability and the high tendency of unprotected acetylene derivatives to polymerize, deprotected 2- ethynylthiophene was immediately used for the subsequent reaction after aqueous work-up without further characterization. Stirring a solution of the deprotected 2-ethynyl-3,4- dibutylthiophene, catalytic amounts of Pd(PPh3)2Cl2, CuI and two equivalents of triethylamine in THF at room temperature for 12 h afforded the formation of bis(thien-2- yl)-1,3-butadiyne 15. The pure product was obtained after chromatography in excellent yield of 94 %, as a pale yellow solid.

Bu Bu Bu Bu Bu Bu 1. KOH / MeOH / THF Bu Bu H Si(CH3)3

S Br Pd(PPh )Cl / PPh S 2. Pd(dppp)Cl / CuI S S 3 2 3 Si(CH3)3 2 CuI / pyridin / NEt NEt / THF 15 [94%] 2 3 14 [82%] 3

Scheme 4.17. Palladium-promoted oxidative coupling reaction of 2-ethynyl-3,4-dibutylthiophene 14.

The standard protocol of the Hay-coupling reaction afforded the same compound in only 73% yield.20 Since this new protocol shows improvement in the yield of the coupling reaction similar conditions were applied for the dimerization of the higher homologues 12 and 13. Firstly, the TMS-protective group of the ethynyl units in 12 and 13 were removed by reacting with potassium hydroxide in THF and aqueous methanol. After aqueous work- up, the solution of the corresponding deprotected 5-ethynyl-oligothiophenes in the presence of Pd(dppp)Cl2, CuI, triethylamine in THF were stirred for 12 h at room temperature yielding corresponding bis(oligothienyl)-1,3-butadiynes 16 and 17 (Scheme 4.18). Compound 16 was isolated as orange solid in 91 % yield after chromatographic purification on silica gel. In the case of butadiyne 17, the yield was lower due to the formation of an untractable by-product. Nevertheless, the separation of the product was

4.3 Synthesis of α,ω-diethynylated oligothiophenes 99 achieved after repeated chromatographic work-up resulting in the isolation of analytically pure 17 as a red solid in 61 % yield.

Bu Bu Bu Bu 1. KOH / MeOH / THF S S S n 2. Pd(dppp)Cl2 / CuI / THF Si(CH3)3

12, 13 (n = 1,2) Bu Bu Bu Bu

Bu Bu n Bu S S S S Bu

n S S 16, 17 (n = 1,2) [91%, 61%]

Scheme 4.18. Palladium-promoted oxidative coupling reaction of 5-ethynyl-terthiophene 12 and 2-ethynyl- quinquethiophene 13.

When the two different variants of the coupling reaction are compared, it becomes obvious that the palladium-mediated reaction is superior to the copper-mediated method, as it gives higher yields under milder conditions and simpler work-up procedures.

In the next step, the conversion of the bis(oligothienyl)-1,3-butadiynes 16 and 17 to the corresponding oligothiophenes by reaction with sulfide anions has been investigated. Kagan et al. reported that while the reaction of bis(thien-2-yl)-1,3-butadiyne with sodium sulfide in boiling methanol works, no conversion could be observed for the reaction of the corresponding 3, 3’’-dimethylated derivative.14 Due to steric hindrance of the alkyl side chains on the adjacent diyne units the desired transformation could only be achieved by using the higher boiling 2-methoxyethanol instead of methanol and raising the reaction temperature. The same difficulties arose while converting bis(terthien-2-yl)-1,3-butadiyne 16 into the corresponding septithiophene.20 The reaction of 16 with sodium sulfide took place only in boiling 2-methoxyethanol and complete conversion was only achieved after stirring the reaction mixture for 12 h at 130°C. In order to find milder conditions and a more general applicability for this reaction, diyne 16 was further investigated with respect to the formation of the corresponding oligothiophene derivative. Interestingly, in some contributions it was reported that thiophene formation from 1,3-butadiynes proceeded in very high yields by using DMSO as solvent and potassium hydroxide as base, whereas the reactions in methanol did not lead to acceptable results.31 Under similar reaction

100 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

conditions, one equivalent of diyne 16 was stirred at 80°C with 10 equivalents of sodium hydrogensulfide hydrate and 5 equivalents of potassium hydroxide in DMSO. The reaction control by TLC showed that after already 3 hours a complete conversion was achieved. After chromatographic work-up followed by precipitation of the obtained product by adding methanol to a concentrated dichloromethane solution, octabutyl-septithiophene 18 was isolated in 77 % yield as red microcrystalline solid (Scheme 4.19). The same method was applied for the preparation of dodecabutyl-undecithiophene 19. The reaction of bis(quinquethien-5-yl)-1,3-butadiyne 17 with 10 equivalents of sodium hydrogensulfide hydrate and 5 equivalents of potassium hydroxide in DMSO yielded the product 19 after chromatographic purification in 72 %, as red solid.

Bu Bu Bu Bu NaHS hydrate Bu Bu n Bu S S S S Bu DMSO n S 16, 17 (n = 1,2) S

Bu Bu Bu Bu Bu Bu Bu Bu

S S S S S n S n S

18, 19 (n = 1,2) [77%, 72%]

Scheme 4.19. Synthesis of septithiophene 18 and undecithiophene 19 by ring closure reaction with sulfide.

Same yields have been previously reported for both compounds, but under much harsher reaction conditions.20 The use of DMSO instead of methanol or 2-methoxyethanol allows conversion under much milder conditions. This solvent effect on the reaction performance has not yet been studied. However, as mentioned in the previous chapter the addition of sulfide anions to conjugated triple bonds is a very common reaction in the field of natural diacetylenes and in the synthesis of 1,2-dithiin derivatives.32,33 The synthetic methods to access 1,2-dithiins involve usually an initial two-fold addition of a protected thiol (R´SH) to a diacetylene followed by a two step sequence involving deprotection of the R´S groups and oxidative ring closure of the resulting bis-thiolate through the S-S bond formation. Koreeda et al. reported the highly efficient formation of various 3,6-substitueted 1,2- dithiins when t-butyl mercaptan is employed as the protected thiol. As shown in Scheme 4.20, the stereo- and regiocontrolled bis-addition of t-BuSH to various 1.3-butadiynes was performed in DMF in the presence of a catalytic amount of KOH. Subsequently, the

4.3 Synthesis of α,ω-diethynylated oligothiophenes 101 corresponding 1,4-bis-thiol adducts could be converted into 1,2-dithiins via a one-step oxidative deprotection-cyclization sequence by treatment with an appropriate electrophile such as N-bromosuccinimide or iodine.33 When R’ is an acetyl-group the same transformation is reported to take place by simply employing oxidants such as oxygen or iodine under mild basic conditions.34

2 t-BuSH NBS or I2 R R R R R R KOH (cat) S S S S t-Bu t-Bu R = phenyl, TMS, CH OH 2

Scheme 4.20. Synthesis of 1,2-dithiins by a oxidative deprotection- cyclization sequence.

A number of natural 1,2-dithiins which primarily have been isolated from the leaves and roots of plants belonging to Astaraceae35 genus, as well as synthetic 1,2-dithiins, exhibit a wide spectrum of biological activities.36 These include antifugal, antibacterial, antiviral and antitumor activity and light-mediated activity against human immunodeficiency virus. A notable property of these compounds is their light sensitivity, which by brief exposure to visible or ultraviolet light give the corresponding thiophenes. This desulfurization process in 1,2-dithiins results in enhanced biological activity. The mechanistical route suggested by Towers et al.,37 after intensive photochemical studies in solution and under matrix isolation conditions of several 1,2-dithiins, is outlined in Scheme 4.21. Thus, a brief irradiation of the 1,2-dithiin at low temperature with visible light affords an intermediate which, after a electronic rearrangement, on warming or further exposure to light is converted to the corresponding thiophene under loss of sulfur.

hυ R R S S ''-S'' RRS

hυ hυ or ∆ ''-S''

S 4πa+2πa S R R R R S S

Scheme 4.21. Proposed mechanism for the desulfurization process of 1,2-dithiins.

102 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

By taking all these results in consideration, one can assume that the formation of thiophenes by the reaction of sulfide with 1,3-butadiynes follows a similar mechanistical route as illustrated in Scheme 4.22. In a first step, a bis-thiolate intermediate can be formed by addition of two equivalents of sodium hydrogensulfide to the triple bonds. Subsequently, under basic conditions in DMSO an oxidative ring closure takes place. The last step is certainly enhanced and probably occurs spontaneously in the presence of DMSO as oxidant. The desulfurization process of the 1,2-dithiin derivative to the corresponding thiophene follows the similar general route as described above for 1,2- dithiins.

oxidative 2 NaHS / KOH ring closure R R R R R R NaS SNa S S

hυ or ∆ hυ 4π + 2πa S ''-S'' S R a R R RR R S S S

Scheme 4.22. Proposed mechanism for the formation of thiophenes from 1,3-butadiynes by reaction with sulfide.

4.3.2 Synthesis of α,ω-diethynylated oligothiophenes as precursors for cyclization

The introduction of ethynyl groups to the oligothiophene cores has been carried out by the palladium-catalyzed Sonogashira-Hagihara coupling reaction.28 Therefore, in an initial step, oligothiophenes 4, 7, and 18 were selectively activated in the α-positions by iodination (Scheme 4.23). In each reaction, the oligothiophene was first treated with mercuric acetate and acetic acid to form an intermediate, α,ω-dimercurated oligothiophene.38 The addition of acetic acid was mostly necessary to maintain sufficient solubility of the reagents. Via a subsequent iododemercuration reaction, the addition of iodine afforded α,ω-diiodo-oligothiophenes 20-22 after chromatographic work-up in 91 %, 88 %, and 82 % yields, respectively.

4.3 Synthesis of α,ω-diethynylated oligothiophenes 103

Bu Bu Bu Bu Bu Bu Bu Bu 1. Hg(OAc2) / CH3COOH S S 2. I I S S n 2 I S S n

4, 7, 18 (n=1, 2, 3) 20- 22 (n=1, 2, 3) [91%, 88%, 82%]

Scheme 4.23. Iodination of the oligothiophenes 4, 7, 18 to the corresponding α,ω-diiodo-oligothiophenes 20- 22.

Since removal of TMS-group proceeds very easily and quantitatively under mild basic reaction conditions, TMS-acetylene was chosen as acetylenic coupling reagent. The coupling reaction of the diiodo-functionalized compounds 20-22 were carried out with a slight excess of TMS-acetylene in the presence of the catalytic system Pd(PPh3)2Cl2, CuI and PPh3 in pyridine and triethylamine. An increase in the coupling yields was achieved by performing the reaction in a closed thick-walled Schlenk flask with a teflon stopper avoiding the evaporation of the highly volatile acetylenic reagent. The protected α,ω- diethynylated oligothiophenes 23-25 could be successfully isolated after purification by chromatography followed by precipitation through addition of methanol to a concentrated dichloromethane solution in yields higher than 90% (Scheme 4.24).

Bu Bu Bu Bu Bu Bu Bu Bu H Si(CH3)3 S S I I S n Pd(PPh3)2Cl2 / PPh3 S S S (H3C)3Si n Si(CH3)3 CuI / pyridin / NEt3 20- 22 (n=1, 2, 3) 23- 25 (n=1, 2, 3) [92%, 92%, 90%]

Scheme 4.24. Sonogashira-Hagihara coupling reaction for the synthesis of α,ω-diethynyl-oligothiophenes 23-25.

Quinquethiophene 24 was also obtained in an analogous Sonogashira coupling reaction by using the dibromo-quinquethiophene 11 as starting material. 11 was previously isolated as by-product by the synthesis of the corresponding monobromo derivate 10 (see chapter 4.3.1). Typically, bromine compounds show lower reactivity in coupling reactions compared to the iodine analogues. However, in this case, the obtained yield of the pure product 24 dropped only slightly to 89% (Scheme 4.25).

104 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Bu Bu Bu Bu Bu Bu Bu Bu H Si(CH3)3 S S Br Br S S n Pd(PPh )Cl / PPh S S 3 2 3 (H3C)3Si n Si(CH3)3 CuI / pyridin / NEt3 11 (n=2) 24 (n=2) [89%]

Scheme 4.25. Palladium-catalyzed Sonogashira-Hagihara coupling reaction of dibromo-quinquethiophene 11 and trimethylsilylacetylene.

4.4 Macrocyclization by oxidative coupling reaction

4.4.1 Synthesis of macrocyclic terthiophene-diacetylenes C[3T-DA]n

Several variations of acetylenic couplings in the synthesis of a rich variety of macrocyclic systems have been reported. Among these, the procedure introduced by Breslow is remarkable.39 This method in which high excess of Cu(I) and Cu(II) chloride in pyridine is employed as catalyst system has proven to lead to cyclic products even in those cases in which cyclization under standard Glaser reaction conditions completely failed. Thus, macrocyclisation of the smallest building block, 5,5’’-diethynylated terthiophene 23 was first performed by using Breslow’s protocol. Firstly, desilylation of the TMS-protected ethynyl groups of diethynyl-terthiophene 23 was carried out under mild basic reaction conditions by using potassium hydroxide in THF and aqueous methanol. The reaction was monitored by TLC and complete conversion was obtained after 3 hours. Due to the known inherent instability and high tendency of unprotected acetylene derivatives to polymerize, the deprotected terthiophene 26 was immediately used after aqueous work-up for the subsequent cyclization step without further purification and characterization. As discussed previously, in order to favour cyclic product formation and to suppress linear oligomerization or polymerization during the reaction, in general, high-dilution40 or pseudo-high dilution conditions41 were employed. The latter procedure has the advantages of avoiding the use of extremely high amount of solvent and large-sized equipment. According to this method, a diluted pyridine solution of terthiophene 26 was slowly added at room temperature over 100 hours via a syringe pump to a slurry of Cu(I) and Cu(II) chloride in high excess in the same solvent (Scheme 4.26). The reaction mixture was additionally stirred for 168 hours at room temperature. After aqueous work-up, the generated cyclic or noncyclic polymeric materials and inorganic salts

4.4 Macrocyclization by oxidative coupling reaction 105

were removed by filtration through a short column of silica gel. A product mixture of various macrocycles C[3T-DA]n could finally be isolated after evaporation of the solvent as red microcrystalline solid in 8.1% yield (based on 23).

Bu Bu

S S Bu Bu S Bu Bu Bu Bu Bu Bu S S R S S R CuCl2 / CuCl S pyridine / RT 23 R = Si(CH3)3 KOH pseudo high dilution S MeOH / THF Bu Bu 26 R = H S Bu S S Bu n-2

Bu Bu

C[3T-DA]n 27a-h (n =3-10) total yield [8.1 %]

Scheme 4.26. Synthesis of macrocyclic terthiophene-diacetylenes C[3T-DA]n (n=3-10) by a modified Glaser coupling reaction under pseudo high-dilution conditions.

1 In the H NMR spectrum of the crude product mixture C[3T-DA]n no free acetylenic proton signals around 3.5 ppm are observed, clearly revealing the formation of cyclic products (Figure 4.4). Due to the high symmetry of the molecules the spectrum is very simple. In the aromatic region only one group of signals at around δ = 7.1 ppm is present which corresponds to the protons of the unsubstituted thiophene-rings H-3’,4’-Th. In the aliphatic region, three more groups of signals at δ = 2.7, δ = 1.5 and δ =1.0 ppm that belong to the butyl chains are apparent.

106 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

CH3-Bu γ H-3',4'-Th β α

H-3’,4’-Th S S Bu Bu S Bu Bu β,γ-CH2-Bu S

S α-CH -Bu 2 S Bu S Bu

Bu S S Bu n-2

Bu Bu

7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 C[3T-DA]n (ppm)

1 Figure 4.4. H NMR of the crude product mixture C[3T-DA]n obtained from a modified Glaser coupling reaction under pseudo high-dilution conditions.

The MALDI-TOF mass spectrum of the crude product mixture exhibited five main signals which correspond to the mass of macrocycles C[3T-DA]n containing 3 to 7 terthiophene units (n = 3-7) linked by diacetylenes-bridges (Figure 4.5).

27b 27c (n = 4) (n = 5) 100 2073 2591

60 27a 27d (n = 3) (n = 6) 40

I / a.u. 3110 1555 27e (n = 7) 20 3629

0 1000 1500 2000 2500 3000 3500 4000 m / z

Figure 4.5. MALDI-TOF MS of the crude product mixture C[3T-DA]n obtained from the oxidative coupling reaction of 26.

4.4 Macrocyclization by oxidative coupling reaction 107

The cyclic product mixture was further characterized by analytical HPLC. The measurements were accomplished on a nucleosil-modified silica column by using hexane/dichloromethane (86/14) as eluent at a flow rate of 1.3 mL/min. In the HPLC chromatogram of the cyclic mixture C[3T-DA]n, four main peaks along with two smaller ones at longer retention times are observed (Figure 4.6). Correlating these data with those obtained from the MALDI-TOF MS analysis, the first peak with a retention time tR = 7.8 min is attributed to the cyclotrimer C[3T-DA]3 27a, the second one to the cyclotetrameric terthiophene-diacetylene C[3T-DA]4 27b and the following three peaks to the next higher homologues 27c (n=5), 27d (n=6) and 27e (n=7). At tR = 18.1 min, an additional small peak is observed which is attributed to a cyclooctamer C[3T-DA]8 27f which was not detected by MALDI-TOF MS analysis of the product mixture.

λ = 410 nm, 20 nm 27b (n = 4) hexane / dichloromethane: 86/14

27a 27c (n = 3) (n = 5)

27d (n = 6) 27e 27f (n = 7) (n = 8)

0 5 10 15 20 minutes

Figure 4.6. HPLC chromatogram of the crude product mixture C[3T-DA]n obtained from the oxidative coupling reaction of 26.

According to analytical HPLC the main product formed by this oxidative coupling reaction was cyclotetramer C[3T-DA]4 27b followed by cyclotrimer C[3T-DA]3 27a and cyclopentamer C[3T-DA]5 27c. In general, due to their similar structure, compositions and solubilities, the separation of a specific macrocycle from such a product mixture is very difficult. The separation of the C[3T-DA]n macrocycles succeeded, even if tedious and associated with loss of material, by repeated preparative HPLC performed on a nucleosil- modified silica column by a flow rate of 80 mL/min. As eluent n-hexane/dichloromethane

108 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

mixture in ratio 88:12 was used. The polarity of the eluent was gradually increased during the run time of the separation by changing the solvent composition to 78:22 n- hexane/dichloromethane. Remarkably, the preparative HPLC separation of the product mixture resulted not only in the isolation of the pure macrocycles C[3T-DA]n with n = 3–7, but also in the isolation of the cyclooctamer C[3T-DA]8 27f, and very low amounts of cyclononamer C[3T-DA]9 27g and cyclodecamer C[3T-DA]10 27h. After preparative separation, each macrocycle was analyzed by analytical HPLC. The specific retention times of the macrocycles and the respective absorption maxima obtained by the UV- detector are outlined in Table 4.1. By precipitation upon addition of methanol to a concentrated dichloromethane solution each macrocycle 27a-h was purely isolated as red microcrystalline solid. The yields in which the macrocycles were obtained are given in Table 4.1.

Table 4.1. Analytical data and isolated yields of macrocyclic terthiophene-diacetylenes C[3T-DA]n (n=3-10) 27a-h.

Analytical data

[e] [a] [d] yields [%] C[3T-DA]n Molecular Mass HPLC

found calcd. tR[min] λmax [nm]

27a (n = 3) 1554.5 1554.6 6.2 422 1.0 27b (n = 4) 2073.0 2072.9 7.7 420 2.0 27c (n = 5) 2590.8 2591.1 8.8 431 1.6 27d (n = 6) 3109.1 3109.3 11.3 439 0.8 27e (n = 7) 3628.4 3627.5 5.8 443 0.7 27f (n = 8) 4145.6 4145.7 6.5 445 < 0.1 27g (n = 9) 4672.4[b] 4669.7[c] 8.5 445 < 0.1 27h (n = 10) 5190.9[b] 5188.5[c] 10.9 445 < 0.1

[a] All values, except for macrocycles 27g and 27h, represent the calculated and measured (operated in reflector modus) monoisotopical mass values of macrocycles 27a-f; [b] average mass measured (operated in [c] [d] linear modus) by MALDI-TOF mass spectrometry; calculated average mass; nucleosil column NO2, flow rate: 1.3 mL/min, eluent n-hexane/dichloromethane 85/15 for 27a-d and 82/18 for 27e-h; [e] total yield of cyclic products C[3T-DA]n prior to separation by HPLC: 8.1 %.

4.4 Macrocyclization by oxidative coupling reaction 109

All macrocyclic terthiophene-diacetylenes C[3T-DA]n (n = 3-10) are very well soluble, mainly in chlorinated organic solvents. The structure of the first four homologues 27a-d in 1 13 the series of C[3T-DA]n are unequivocally proven by H, C NMR and MALDI-TOF-MS analysis. The higher homologues in the macrocycles series C[3T-DA]n 27e-h were isolated only in small amounts, thus, a characterization by NMR analysis was not possible. Nevertheless, the molar mass of these compounds could be unambiguously proven from their MALDI-TOF mass spectra, which in each case exhibited solely one intense signal corresponding to the theoretical mass of the respective macrocycle. Further structural proof for the cyclic natures of compounds 27e-h was given by analytical HPLC measurements and UV-VIS spectroscopy (Table 4.1). The retention time values, tR, and absorption maxima, λmax, are parameters which are very diagnostic within a homologous series and significantly differ from those of corresponding linear compounds. Due to the high symmetry and similarity of the structures (the same oligothiophene units, only different ring sizes), NMR spectra show the same pattern of signals for each cyclic compound which can easily be assigned. In Figure 4.7, the 1H NMR spectrum for the cyclopentamer 27c is given as an example. Lack of a signal around δ = 3.5 ppm which is the characteristic resonance signal for a terminal acetylenic proton is the evidence for the proposed cyclic structure. In the aromatic region, there is only one signal at δ = 7.09 ppm which belongs to the protons of the unsubstituted thiophene-rings H-3’,4’-Th. In the aliphatic region, three more groups of signals at δ = 2.70, δ = 1.61-1.44 and δ =0.99 ppm that corresponds to the butyl chains are apparent. The 13C spectrum of macrocycle 27c (n = 5) shows four groups of signal between δ = 13-33 ppm corresponding to the carbons of the butyl chains (Figure 4.8). Very characteristic are two signals at δ = 77.9 and δ = 81.5 ppm for the acetylenic carbon atoms. Due to the high symmetry of the terthiophene-unit only six signals in the aromatic region of the 13C spectrum are visible.

110 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

7.0930 2.6937 1.6161 1.5467 1.4628 1.4479 0.9928

CH3-Bu

H-3’,4’-Th

β,γ-CH2-Bu α-CH2-Bu

2 8 16 12 7 6 5 4 3 2 1 0 (ppm)

1 Figure 4.7. H NMR spectrum of the cyclopentamer C[3T-DA]5 27c.

151.0892 138.7739 136.3321 133.3366 126.3334 116.8505 81.4893 77.9517 32.6223 28.6941 27.6969 22.9308 22.7109 13.8649 13.8460

β-Bu

γ-Bu CH3-Bu C-2,3-Th C-2‘‘,3‘‘-Th α-Bu C-4,4‘‘-Th C-3‘,4‘-Th C≡C C-2‘,5‘-Th C-5,5‘‘-Th

160 140 120 100 80 60 40 20 0 (ppm)

13 Figure 4.8. C NMR spectrum of the cyclopentamer C[3T-DA]5 27c.

As mentioned above, NMR spectra of the other cyclic homologues exhibit the same signal pattern as macrocycle 27c. The data obtained from the 1H and 13C NMR analysis of the

4.4 Macrocyclization by oxidative coupling reaction 111

macrocycles C[3T-DA]n 27a-d are summarized in Table 4.2. Furthermore, the relevant chemical shifts in 1H and 13C NMR spectra of the corresponding diethynyl-terthiophene precursor 26 and of the two linear terthiophene-diacetylenes 26 and L[(3T)3-(DA)2] whose structures are illustrated in Figure 4.9 are presented for comparison. The NMR data of the diethynylated terthiophene 16 and terthiophene-derivative L[(3T)3-(DA)2] were reported by others.18 Within the macrocycle series 27a-d no significant changes in the chemical shifts of protons and carbons belonging to the butyl-groups are apparent. The H-3’,4’ protons of the unsubstituted thiophene-rings are only minimally shifted and are basically similar to those of the linear compounds. Terthiophene-diacetylene macrocycles C[3T-

DA]n comprise exclusively conjugation-active subunits and thus, they respresent 4nπ- electron systems comparable to annulenes. However, in none of the 1H NMR spectra of the macrocycles C[3T-DA]n an antiaromatic ring current was apparent.

α''' 4' 2'' 3' 5'' S S 2 2' γ β S α 5'' 2'' S 2 5 5 S H S S H 4' 3' S 26 S S

S γ S n-2 β α α α' α'' 2 2 5 α''' 5 S S S S C[3T-DA]n 2'' 2'' 5'' S 3' 3' S 5'' 4' 16 4' 27a-d

γ β α' α α'' α'''

5 2 S 2'' S S 5'' 3' 4' S S S S

L[(3T)3-(DA)2] S S

Figure 4.9. Macrocyclic and linear terthiophene-derived acetylenes whose NMR data are given in Table 4.2.

112 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Table 4.2. NMR data of macrocyclic C[3T-DA]n 27a-d and of linear terthiophene-derived acetylenes 16, 26 and L[(3T)3-(DA)2].

Compound Macrocycle C[3T-DA]n Linear

27a 27b 27c 27d [a] L[(3T)3- 26 16 [a] n = 3 n = 4 n = 5 n = 6 (DA)2] H-3’,4’-Th 7.07 7.08 7.09 7.09 7.06 7.10[b] 7.13

α-CH2-Bu 2.66 2.69 2.70 2.71 2.67 2.68 2.74

ß,γ-CH2-Bu 1.62 1.62 1.61 1.61 1.50 1.51 1.57 H NMR 1 CH3-Bu 0.97 0.99 0.99 0.98 0.96 0.95 0.99 C-2’5’-Th 149.6 150.8 151.1 151.3 149.6 151.6 151.6 C-4,4’’-Th 138.3 138.7 138.7 138.8 138.5 138.5 138.5 C-2,2’’,3,3’’-Th 136.8 136.5 136.3 136.2 135.9 135.1 135.1 133.8 133.3 133.3 133.3 131.7 130.5 130.5 C-3’,4’-Th 124.9 126.0 126.3 126.5 126.4 126.5 126.5 C-5,5’’-Th 117.2 117.0 116.8 116.7 116.1 116.0 116.0 Th-C≡C 81.2 81.6 81.6 81.9 84.1[c] 81.2 81.2 C≡C-C 78.7 78.9 77.9 77.9 76.9[c] 77.2 77.6 ß-Bu 32.7 32.7 32.7 32.6 32.6 32.6 32.6 32.4 32.4 32.6 a-Bu 28.7 28.7 28.7 28.7 28.4 28.9 28.9 27.5 27.6 27.6 27.7 27.7 27.7 27.5 γ-Bu 22.9 22.9 22.9 23.0 22.7 22.7 22.9 22.7 22.7 22.7 22.7 22.7 C NMR C NMR

13 CH3-Bu 13.9 13.9 13.9 13.8 13.9 13.9 13.9 [a] 20 [b] 1 NMR data of 26 and L[(3T)3-(DA)2] taken from literature; due to different symmetry the H NMR spectrum of 16 shows for the H-3’,4’-Th protons two dublets at δ =7.10 and δ = 7.05 ppm; [c]the signals correspond to a Th-C≡C-H moiety.

Within the macrocycle homologous series 27a-d there are significant changes of the carbon signals which belong to the thiophene moieties. In Figure 4.10, the aromatic region of the 13C NMR spectra of cyclotrimeric 27a (above) and cyclohexamer 27d (below) are compared. In the spectrum of the smaller homologue 27a, C-2’,5’ and C-3’,4’-carbons of the middle-thiophene ring are approximately 1.6 ppm high-field shifted. The peaks appear at δ = 149.6 and δ = 124.9 ppm, whereas in the spectrum of cyclohexamer 27d they are at δ = 151.6 and δ = 126.5 ppm, respectively. The values obtained for the higher homologue

27d correspond very well to the value of the linear compounds 16 and L[(3T)3-(DA)2].

When the whole C[3T-DA]n series is compared, it is clearly seen that the signals of the C- 2’,5’ and C-3’,4’-carbons are gradually low-field shifted with increasing ring-size.

4.4 Macrocyclization by oxidative coupling reaction 113

Obviously, as the ring-size of the macrocycles gets smaller, ring strain gets larger. This has, as indicated by the changes in the chemical shifts, the most dramatic effect on the middle-thiophene in the terthiophene units. Remarkable are also the shifts of the acetylenic carbons. While the signal from the carbon atom which is directly connected to the thiophene (Th-C≡C) shows only slight changes, the inner carbon of the butadiyne units (C≡C-C) are shifted to higher fields with increasing ring-size within the homologues series

C[3T-DA]n. For the cyclotrimer 27a the above mentioned signal is observed at δ = 78.7 ppm, while for the cyclohexamer 27d at δ = 77.9 ppm. These values correspond also well with those of the linear terthiophene-diacetylenes 16 and L[(3T)3-(DA)2]. Apparently, the higher ring strain attributed to the smaller cyclic homologues has a significant influence on the butadiyne units, but not so on the thiophene-acetylide bonds.

Figure 4.10. 13C NMR spectra of cyclotrimeric macrocycle 27a (above) and cyclohexameric macrocycle 27d (below).

MALDI-TOF mass spectrometric analysis of each terthiophene-diacetylene macrocycle

C[3T-DA]n was performed using 1,8,9-trihydroxyanthracene (Dithranol) as the matrix material. The measured and calculated mass values of all macrocycles C[3T-DA]n are summarized in Table 4.1. The corresponding spectra are attached at the end of this chapter.

114 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Except for compounds 27g and 27h, all other compounds 27a-f were measured in reflector modus. The MALDI-TOF MS spectrum of each macrocycle 27a-f exhibited only one signal at the respective values given in Table 4.1. The measured monoisotopic mass signals correspond very well to the calculated monoisotopical molecular mass of the respective compounds. Further, the isotopic distributions of the signals for each macrocycle 27a-f were determined and were found to be in very good agreement with the calculated ones (see figures in supplement of this chapter).

As an example, the MALDI-TOF MS spectrum of cyclopentamer C[3T-DA]5 27c is depicted in Figure 4.11. The mass spectrum of 27c shows only one signal at m/z = 2591.3 which corresponds very well to the calculated molecular mass of the macrocycle 27c (m/z = 2591.1). As shown in Figure 4.12, the experimental isotopic distribution (curve below) is fully consistent with the calculated one (curve above).

2591.3 100

60 / a.u. I 40

2000 3000 4000 5000 m / z

Figure 4.11. MALDI-TOF MS spectrum of cyclopentameric macrocycle 27c.

4.4 Macrocyclization by oxidative coupling reaction 115

100 90 80 70 60 50 40 30 20 10 0 2.590 2.595 2.600

100

0 2588 2590 2592 2594 2596 2598 2600 2602 2604 m / z

Figure 4.12. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of cyclopentameric macrocycle 27c.

Due to their high molecular weights, the MALDI-TOF MS spectra of cyclononamer 27g and cyclodecamer 27h were recorded in linear modus. Thus, the spectra generally exhibited average mass signals without isotopic resolutions. In the mass spectrum of the cyclononamer 27g solely one signal at m/z = 4672.4 corresponding with the calculated average mass of the macrocycle (m/z = 4669.7) was observed. The retention time (tR = 8.5 min) obtained by HPLC analysis and the absorption maximum (λmax = 445 nm) of 27g correspond well to the expected values for a cyclononamer when correlated to those of the other cyclic homologues in the series (see Table 4.1). Thus, the difference of almost 2.7 Daltons between the measured and the calculated mass is likely due to unprecise calibration of the MALDI-TOF mass spectrometer. In general, MALDI-TOF MS measurements are characterized by an accuracy of ±1 Dalton. The MALDI-TOF MS spectrum of the largest homologue in the series of C[3T-DA]n, cyclodecamer 27h, is depicted in Figure 4.13. The spectrum shows a single peak at m/z = 5190.9 which, under these conditions, corresponds with a difference of 2.5 Daltons to the calculated mass of the macrocycle (m/z = 5188.5). The difference is again attributed to inaccurate calibration. The

116 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

retention time of tR = 10.9 min and λmax = 445 nm obtained by HPLC analysis for the compound fit well to the values within the macrocycle series (see Table 4.1).

5190.9 100

80 / a.u. I 60

0 4500 5000 5500 6000 6500 m / z

Figure 4.13 MALDI-TOF MS spectrum of cyclodecamer 27h.

As discussed in Chapter 3, the most frequently applied method for the synthesis of macrocyclic compounds is the Glaser-type coupling reaction reported by Eglinton in 1956.42 In this approach, typically, Cu(II) acetate is employed as oxidizing agent either in pure pyridine or in the presence of a co-solvent to provide sufficient solubility to the reactants during the reaction. A variation of this Eglinton protocol, where a mixture of Cu(II) acetate and Cu(I) chloride are employed in high excess was proven to be very effective in many cyclization reactions.43 Since the different modifications of the oxidative coupling reaction with the same system can lead to different results with respect to the ratio and the yields of the obtained macrocyclic mixture, the cyclization behaviour of 5,5’’-diethynylterthiophene 26 was investigated under the above mentioned variation of Eglinton coupling conditions (Scheme 4.21). Firstly, the TMS-protective groups of the acetylene units of terthiophene 23 were removed under basic conditions. The cyclization reaction was carried out at room temperature and again under pseudo-high-dilution conditions in order to favour macrocyclization. The long time that results from slow addition of the starting material 26

4.4 Macrocyclization by oxidative coupling reaction 117

under these conditions was considered as a factor that might lead to low yields due to polymerization of the unprotected acetylenes. For this reason, it was decided to shorten the reaction time by increasing the addition rate from 0.05 mL/min (applied in the synthesis of the first series) to 0.2 mL/min. Improved yields for diacetylene cyclization reactions with shorter reaction times were already reported by others.44 As catalyst system, a high excess of Cu(II) acetate and Cu(I) chloride in pyridine was used. A diluted pyridine/dichloromethane (4/1) solution of the deprotected diethynylterthiophene 26 was slowly added over 21 h to the catalyst mixture via a syringe pump. Dichloromethane was used as co-solvent in order to ensure sufficient solubility to the reactants. After complete addition, the reaction mixture was allowed to stir for an additional 72 hours at room temperature (Scheme 4.21). Aqueous work-up followed by removal of the inorganic and polymeric material by filtration through a short column of silica gel afforded a red microcrystalline product mixture in 12.7 % yield (based on 23). The analysis of the raw product by 1H NMR spectroscopy and MALDI-TOF mass spectrometry indicated the formation of a mixture of cyclic oligomers. In this case, cyclic homologues C[3T-DA]n containing 3 up to 9 terthiophene units linked by diacetylene bridges were detected.

Bu Bu

S S Bu Bu S Bu Bu Bu Bu Bu Bu S S S S Cu(OAc) x H O / CuCl R R 2 2 S pyridine / dichloromethane 23 R = Si(CH3)3 KOH pseudo high dilution S MeOH / THF Bu Bu 26 R = H S Bu S S Bu n-2

Bu Bu

C[3T-DA]n 27a-g (n =3-9) total yield [12.7 %]

Scheme 4.21 Synthesis of macrocyclic terthiophene-diacetylenes C[3T-DA]n (n=3-9) by a variation of Eglinton coupling reaction under pseudo high-dilution conditions.

The separation of each specific macrocycle from the resulting complex mixture was again achieved by preparative HPLC. Each macrocycle 27a-g could be isolated in analytically pure forms after precipitation upon addition of methanol to the concentrated

118 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

dichloromethane solutions of the HPLC-separated cyclooligomers. The respective yields are given in Table 4.3 and compared with those obtained by Glaser coupling. The C[3T-

DA]n macrocycles with n ranging from 3 to 6 were isolated as major products, while the larger derivatives with n = 7-9 were obtained only in minor quantities. It is interesting to note that with this method the highest homologue, the cyclodecamer C[3T-DA]10 was not formed. 1H-NMR, 13C NMR and MALDI-TOF MS analyses proved the cyclic structure of these compounds.

Table 4.3. Yields of macrocyclic terthiophene-diacetylenes C(3T-DA]n (n=3-9) 27a-h.

27a 27b 27c 27d 27e 27f 27g 27h Σ C[3T-DA] n n = 3 n = 4 n = 5 n = 6 n = 7 n = 8 n = 9 n = 10 [a] yield Glaser coupling 1.0 2.0 1.6 0.8 0.7 < 0.1 < 0.1 < 0.1 6.1 [%] Eglinton coupling 2.6 4.0 1.4 0.5 0.2 < 0.1 < 0.1 --[c] 8.7[b]

[a] [b] total yield of cyclic products C[3T-DA]n prior to separation by HPLC: 8.1 %; total yield of cyclic [c] products C[3T-DA]n prior to separation by HPLC: 12.7 %; not formed.

The obtained yields of the C[3T-DA]n macrocycles are in the range of what has been reported for similar cycles that have been prepared by such a random reaction involving oligomerization and cyclization in a single step.9 As already described by others, also in our case the cyclization results depends on the reaction conditions. Essentially by changing the catalyst system the total amount of cyclic products could be improved from 6.1 % to 8.7 % yield, the latter being obtained by using the modified Eglinton conditions. Both approaches afforded the preferential formation of the tetrameric species C[3T-DA]4, however, a change of the relative amounts of the macrocycles was observed. The modified Eglinton conditions favoured the formation of smaller macrocycles, especially this of the cyclotrimer and cyclotetramer, while the higher homologues were obtained only in minor quantities. Further attempts to improve the results of the macrocyclization reaction and to manipulate the relative ratio of the different macrocycles in the mixture by varying the reaction conditions (i.e. concentration, temperature, Cu-catalyst, solvent) resulted in no specific success, as no increase in the yields could be observed and a comparable mixture of products was always formed. Single crystals suitable for X-ray structure analysis were obtained for cyclotrimeric 27a and cyclotetrameric macrocycle 27b. The detailed structural characterizations of these two

4.4 Macrocyclization by oxidative coupling reaction 119

macrocycles have already been reported.19,45 A brief presentation of the structural characteristics is included in Chapter 5.6.8.

4.4.2 Synthesis of macrocyclic quinquethiophene-diacetylenes C[5T-DA]n

As discussed previously, one possible way to increase the low yield of a statistical intermolecular cyclization reaction is to reduce the number of possible oligomeric structures that can be formed by using an extended oligomer as precursor, and as a consequence the product separation and purification is also facilitated. In a previous project, the cyclization of a terminally diethynylated quinquethiophene had been investigated following the above presented modified Glaser protocol.18,19 The oxidative coupling using Cu (I) and Cu(II) chloride as catalysts resulted in the formation of a mixture of cyclic products, containing the cyclotrimer and cyclotetramer as main products. They were isolated and characterized along with traces of cyclodimeric species and higher homologues, which could only been detected and assigned by MALDI-TOF MS analysis. With the aim to prepare larger quantities for full characterization of all cyclic homologues in this quinquethiophene series and encouraged by the results obtained for the preparation of the terthiophene-diacetylene macrocycles C[3T-DA]n under Eglinton reaction conditions, in this study, 5,5’’’’-diethynylated quinquethiophene 24 was subjected to the identical conditions (Scheme 4.22). In close analogy to the above described preparation, firstly, the TMS-protective groups were removed under basic reaction conditions to generate the reactive diethynyl-quinquethiophene 28. Subsequently, cyclization was carried out by slow addition of a dilute solution of deprotected quinquethiophene 28 in pyridine/dichloromethane (4/1) over a period of 12 hours to a slurry containing high excess of Cu(II) acetate and Cu(I) chloride in pyridine. After stirring the reaction mixture for additional 72 hours at room temperature, the work-up was carried out. Filtration over silica gel removed the polymeric and inorganic materials and afforded the isolation of a red microcrystalline product mixture in 16.0 % yield (based on 24).

120 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Bu Bu Bu Bu Bu Bu

S S S S S R R

24 R = Si(CH ) KOH 3 3 MeOH / THF 28 R = H

Cu(OAc)2 x H2O / CuCl pyridine / dichloromethane pseudo high dilution

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S n-1 Bu Bu Bu Bu

29a-e (n = 2-6) total yield [16 %] C[5T-DA] n

Scheme 4.22 Synthesis of macrocyclic quiquethienyl-diacetylenes C[5T-DA]n (n=2-6) by a variation of the Eglinton coupling conditions and under pseudo high-dilution.

In the 1H NMR spectrum of the crude mixture a very small signal at around δ = 3.65 ppm indicated that besides cyclic products only a very low amount of linear oligomers were formed (Figure 4.14). The aromatic region of the spectrum shows only one group of signals at around δ = 7.1 ppm which corresponds to the protons of the unsubstituted thiophene-rings H-3’,4’,3’’’,4’’’. In the aliphatic region three more groups of signals at δ = 2.7, δ = 1.5 and δ =1.0 ppm that belong to the butyl-chains are present. According to the

MALDI-TOF MS spectrum, the mixture consisted of cyclic homologues C[5T-DA]n with n ranging from 2 to 5 (Figure 4.15). The signal corresponding to the cyclopentamer 27d is very weak. Remarkably, cyclodimer C[5T-DA]2 29a was formed as major product along with the corresponding cyclotrimer 29b and cyclotetramer 29c. In the aforementioned work,18 in which standard Glaser conditions have been used, the cyclodimer was formed only in traces and thus, could only be detected by MALDI TOF MS, but never be isolated.

4.4 Macrocyclization by oxidative coupling reaction 121

7.0887 2.7266 1.5879 1.5090 1.4720 1.4544 1.4370 1.0159 0.9976 0.9793 0.9638 0.9467 0.0000

γ β α 5 CH -Bu 2 3 3' S S 4' S S 2'' S S 5'' S S β,γ-CH2-Bu 3''' S S α-CH2-Bu 4''' H-3’,3‘‘‘,4‘,4‘‘‘-Th 5'''' n-1 α'''''

C[5T-DA]n 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 (p p m )

1 Figure 4.14. H NMR of the crude product mixture C[3T-DA]n obtained from a modified Eglinton coupling

29b (n=3) 2384 100 29a (n=2) 80 1589 60 29c (n=4) / a.u. I 40 3180 29d (n=5) 20 3973

0 1000 1500 2000 2500 3000 3500 4000 4500 m / z

Figure 4.15. MALDI-TOF MS of the crude product mixture C[5T-DA]n obtained from the oxidative coupling reaction of 28.

It was found that cyclodimer 29a is less soluble in common organic solvents than the higher homologues. This indicates that size and shape of the ring has a significant influence on the solubility. Nevertheless, several attempts to purely isolate macrocycle 29a by precipitation or recrystallization completely failed. Moreover, the restricted solubility of

122 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

cyclodimer 29a led to a partial precipitation during the chromatographic separation by preparative HPLC. The optimal composition of the eluent mixture that guarantees sufficient solubility and good separation had to be determined in preliminary studies. Firstly, the chromatographic behaviour of the compound mixture was evaluated by performing analytical separations on nucleosil-modified silica column. The measurements were carried out at flow rates of 1.3 mL/min by using different n-hexane/dichoromethane mixtures as eluent. A ratio of 83/17 has been found optimal from the separation point of view and to satisfy the solubility requirements. The HPLC chromatogram of the cyclic mixture C[5T-DA]n is shown in Figure 4.16. The four main peaks with retention times of tR

= 6.8 min, tR = 7.4 min, tR = 9.4 min and tR = 13.8 min are attributed to the macrocycles

C[5T-DA]n 29a-d with n = 2-5, respectively. The smaller peaks are attributed to linear oligomeric by-products.

29b λmax = 410 nm, 20 nm (n=3) n-hexane / dichloromethane : 83/17 29a 29c (n=2) (n=4)

29d

mAU (n=5)

minutes

Figure 4.16. HPLC chromatogram of the crude product mixture C[3T-DA]n obtained from the oxidative coupling reaction of 26.

Transferred on preparative scale, due to the moderate solubility of the mixture, the injection throughput was still limited to ca. 50 mg product mixture per 1 mL eluent. Thus, in order to obtain the HPLC-pure compounds, the separation had to be carried out by repeated runs. The applied flow rate was 80 mL/min. Similar to the previously discussed macrocyclic terthiophene-diacetylenes C[3T-DA]n, 27a-h in this case also the preparative separation resulted not only in the isolation of the pure macrocycles C[5T-DA]n with (n =

2-5) that were detected by MALDI-TOF MS. A cyclohexamer C[5T-DA]6 could be additionally isolated, however, in very low amounts. By precipitation upon addition of methanol to concentrated dichloromethane solutions, the individual macrocycles C[5T-

4.4 Macrocyclization by oxidative coupling reaction 123

DA]n 29a-e were purely isolated as red microcrystalline solids. The yields in which the macrocycles were respectively obtained, the specific retention times of the macrocycles and the absorption maxima obtained by analytical HPLC are given in Table 4.4. The structure of cyclic homologues 29a-c (n = 2-4) was unequivocally proven by 1H, 13C NMR and MALDI-TOF-MS analysis. Characterization of the higher homologues, cyclopentamer 29d (n = 5) and cyclohexamer 29e (n = 6) by NMR spectroscopy was hampered by the small amounts isolated. Nevertheless, the molecular mass of the two macrocycle 29d and 29e were proven by MALDI-TOF MS which, in each case, exhibits solely one intense signal corresponding to the theoretical mass of the respective macrocycle. Further structural evidence was obtained from their retention times tR and UV- VIS absorption maxima (see Table 4.4).

Table 4.4. Analytical data and isolated yields of macrocycles C(5T-DA]n 29a-e.

Analytical data yields [%]

Molecular Mass[a] HPLC[d] C[5T-DA]n total: 13.5[e] found calcd. tR[min] λmax [nm] 29a (n = 2) 1588.3 1588.6 6.8 421 3.0 29b ( n = 3) 2383.4 2382.9 7.4 425 8.8 29c ( n = 4) 3178.1 3177.3 9.4 436 1.5 29d ( n = 5) 3972.8 3971.6 13.8 436 0.2 29e ( n = 6) 4770.9[b] 4771.9[c] 12.5 436 < 0.1

[a] All values except for compound 29e represent the calculated and found (operated in reflector modus) monoisotopical mass values of the compounds 29a-d; [b] average mass found (operated in linear modus) by [c] [d] MALDI-TOF mass spectrometry; calculated average mass; nucleosil column NO2, flow rate: 1.3 mL/min, eluent n-hexane/dichloromethane 83/17 for 29a-d and 82/18 for 29e; [e] total yield of cyclic products

C[5T-DA]n prior to separation by HPLC 16.0 %.

The NMR spectra of quinquethiophene-diacetylene macrocycles C[5T-DA]n are very similar to those of the C[3T-DA]n macrocycles, exhibiting similar pattern of signals which can easily be assigned due to the high symmetry and similarity of the structures. In Figure 1 4.17, the H NMR spectrum of the cyclodimer C[5T-DA]2 29a is given as an example. Noteworthy is the lack of acetylenic proton signals in the range of 3.5 ppm clearly evidencing the cyclic structure of the compound. One signal corresponding to the

124 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

unsubstituted thiophene protons is seen at δ = 7.01 ppm. In the aliphatic region, three more groups of signals at δ = 2.63, δ = 1.55-1.36 ppm and δ =0.93 ppm belonging to the butyl- chains are present.

7.0120 2.6341 2.6158 1.5525 1.5304 1.5147 1.4162 1.3978 1.3795 1.3606 0.9414 0.9369 0.9237 0.9092 0.0000

H-3’,4’-Th CH3-Bu

β,γ-CH2-Bu

α-CH2-Bu

4 12 24 18 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 (ppm)

1 Figure 4.17. H NMR spectrum of cyclodimeric C[5T-DA]2 29a.

In the 13C spectrum of dimeric macrocycle 29a (n = 2) four groups of signals between δ = 13-33 ppm corresponding to the carbons of the butyl chains can be seen (Figure 4.18). The signals at δ = 78.3 and δ = 81.8 ppm belong to the butadiyne-carbons and that at δ = 117.2 ppm belongs to the carbon of the thiophenes directly attached to the butadiyne units. The other eight signals corresponding to the quinquethiophene-units appear in the aromatic region of the spectrum between δ =124.7 ppm and 150.2 ppm.

4.4 Macrocyclization by oxidative coupling reaction 125

150.1944 140.3285 138.3682 137.4923 136.1652 133.8751 130.2502 125.1732 124.6727 117.2031 81.8268 78.3536 32.6375 32.5806 32.4289 29.7065 28.7813 27.9130 27.6969 23.0256 22.9497 22.7147 13.8156 13.8004

CH3-Bu

γ-Bu

C-5,5‘‘‘‘-Th α-Bu C≡C-C β-Bu C-2‘‘‘,5‘‘‘-Th Th-C≡C

140 120 100 80 60 40 20 0 (ppm)

13 Figure 4.18. C NMR spectrum of cyclodimeric C[5T-DA]2 29a.

Due to the high symmetry and similarity of the structures, NMR spectra of the higher homologues in the macrocycle series C[5T-DA]n , cyclotrimer 29b and cyclotetramer 29c, show the same signal pattern as dimeric macrocycle 29a. As for C[3T-DA]n series 27a-d, the only significantly shifts within the series can be observed for carbons signals belonging to the unsubstituted thiophenes and those of the butadiyne units. The relevant chemical shifts in the 1H and 13C NMR spectra of the macrocycles 29a-c are summarized in Table 4.5. For comparison, the NMR data of the TMS-protected diethynyl-quinquethiophene 24 are included. The NMR spectrum of diethynyl-quinquethiophene 28 was not recorded due to the instability of the compound. However, as already observed for the C[3T-DA]n series, the signals for the 3’,4’,3’’’,4’’’-carbons of the inner-thiophene ring are low field shifted for macrocycles of larger ring size. The peaks for cyclodimer 29a appear at δ = 125.2 and δ = 124.7 ppm, whereas in the specta of next higher homologues 29b and 29c appear at δ = 126.2 and δ = 125.7 ppm, respectively. The changes in the chemical shifts of the two signals belonging to the acetylenic carbons Th-C≡C-C are not that pronounced within the

C[5T-DA]n series as was the case for the terthiophene-diacetylene macrocycles. For the larger macrocycles 29b and 29c both signals are only 0.4 ppm high field shifted compared to those of the cyclodimer 29a and appear at δ = 81.8 pm and δ = 78.4 ppm. A ring current due to cyclic structure formation is also not apparent in the series of quinquethiohene-

126 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

diacetylene macrocycles C[5T-DA]n. For all three macrocycle 29a, 29b and 29c the proton signals for the H-3’,4’,3’’’,4’’’ appear within a range of ∆δ = 0.1 ppm at δ= 7.0 ppm. Notable that these protons for cyclodimer 29a are seen as a singlet, whereas for the higher homologues 29b and 29c as two doublets with a coupling constant of J = 3.5 Hz.

Table 4.5. NMR data of macrocyclic C[5T-DA]n 29a-c and of TMS-protected diethynyl-quinquethiohene 24.

Compound Macrocycle C[5T-DA]n Linear

NMR 29a 29b 29c 24 n = 2 n = 3 n = 4 7.10 7.11 7.09 H-3’,4’,3’’’,4’’’-Th 7.01 7.08 7.09 7.08 125.2 126.2 126.2 129.8 C-3’,4’,3’’’,4’’’ 124.7 125.7 125.7 126.3 C-5,5’’’’-Th 117.2 116.6 116.6 117.3 Th-C≡C 81.8 81.4 81.4 101.9[a] C≡C-C 78.4 78.0 78.0 79.8[a] [a] the signals correspond to a Th-C≡C-Si(CH3)3 moiety.

MALDI-TOF mass spectra of quinquethiophene-diacetylene macrocycles 29a-d were recorded in reflector modus. The spectra for each macrocycle show characteristic signals with regard to both, the theoretical mass and the calculated isotope distribution. Due to its high molecular weight, MALDI-TOF MS analysis of cyclohexamer 29e was performed in linear modus. Thus, its spectrum exhibits solely one signal without isotopic resolutions at m/z = 4771.9 which corresponds well with the calculated average mass of the macrocycle

(m/z = 4770.9). All measured and calculated mass values of all macrocycles C[5T-DA]n are summarized in Table 4.6. The spectra are attached at the end of this chapter. As an example, the MALDI-TOF MS spectrum of cyclodimer C[5T-DA]2 29a is depicted in Figure 4.19. In the spectrum only one signal at m/z = 1588.3 is observed corresponding very well to the calculated molecular mass of the macrocycle 29a (m/z = 1588.6). In Figure 4.20, the experimental isotopic distribution (bottom) is displayed which is fully consistent with the calculated one (top).

4.4 Macrocyclization by oxidative coupling reaction 127

1588.3 100

60 / a.u. I 40

0 800 1200 1600 2000 2400 m / z

Figure 4.19. MALDI TOF MS spectrum of cyclodimeric quinquethiophene-diacetylene 29a (n=2).

g 100 90 80 70 60 50 40 30 20 10 0 1.586 1.588 1.590 1.592 1.594 1.596

100

0 1586 1588 1590 1592 1594 1596 1598 m / z

Figure 4.20. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of dimeric macrocycle 29a.

128 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

4.4.3 Intramolecular ring-closure reaction of α,ω-diethynyl-undecithiophene

In Chapter 3.3.2 various strategies towards preparation of macrocycles were presented. One of the extreme strategies is the intramolecular macrocyclization in which the final cyclization step usually yields only one product or at least one major product that easily can be isolated from the product mixture. The unavoidable pitfall of the intramolecular approach is the time-consuming stepwise synthesis of the corresponding precursor of defined length. However, the availability of well established synthetic methods for constructing longer oligothiophenes significantly impacted the challenge to investigate a more efficient cyclization reaction which incorporates only a one step intramolecular cyclization of an α,ω-difunctionalized oligothiophene. The smallest macrocycles formed during the previously presented intermolecular random reactions were the 48 chain members terthiophene-diacetylene cyclotrimer 27a and quinquethiophene-diacetylene cyclodimer 29a. Thus, the target structure of the precursor which by an intramolecular ring-closure reaction should generate a macrocycle is a diethynylated undecithiophene. The multi-step synthesis of butylated undecithiophene 19 was described in Chapter 4.3.1. The functionalisation of oligothiophene 19 was performed by methods established for the parent smaller homologues 23-25. In the first step undecithiophene 19 was selectively iodinated at the α-positions by elemental iodine and mercury(II) acetate to yield diiodoundecithiophene 30 in 65 % yield (Scheme 4.23). The introduction of the terminal acetylenic groups was achieved by palladium catalyzed Sonogashira coupling diiodoundecithiophene 30 and trimethylsilylacetylene. The TMS-protected undecithiophene 31 was isolated after chromatographic work-up in 95 % yield.

4.4 Macrocyclization by oxidative coupling reaction 129

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu

S S S S S S S S S S S

1. Hg(OAc2)

2. I2

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu

S S S S I S S S S S S S I

30 [65%]

H Si(CH3)3

Pd(PPh3)2Cl2 / PPh3

CuI / pyridin / NEt3

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu

S S S S S S S S S S (CH3)3Si S Si(CH3)3 31 [95%] Scheme 4.23. Synthesis of α,ω-diethynyl-undecithiophenes 31.

Intramolecular macrocyclisation of diethynylated undecithiophene 31 was carried out under the modified Eglinton reaction conditions described previously for the synthesis of cyclic terthiophene-diacetylenes C[3T-DA]n and quinquethiophene-diacetylenes C[5T-

DA]n. First, removal of the TMS-protecting groups was accomplished under mild basic conditions affording the active precursor diethynyl-undecithiophene 32 (Scheme 4.24). Cyclisation was carried out by adding at room temperature a dilute solution of deprotected of diethynyl-undecithiophen 32 pyridine/dichloromethane (4/1) with a syringe to a slurry containing high excess of Cu(II) acetate and Cu(I) chloride in pyridine. The work-up was carried out after stirring the reaction mixture for additional 36 hours at room temperature,. Filtration over short column of silica gel removed the polymeric and inorganic materials and afforded the isolation of a red microcrystalline product mixture in 60 % yield (based on 30). The product was analyzed by analytical HPLC on a nucleosil-modified silica column by using hexane/dichloromethane (82/18/) as eluent. In the chromatogram two main peaks at tR = 6.3 min and tR = 12.3 min in a 68:32 ratio were observed. The MALDI- TOF MS spectrum of the product mixture shows two isotope-resolved signals at m/z = 1622 and m/z = 3246 (Figure 4.21). This signals correspond to the theoretical mass of the

C[11T-DA]1 macrocycle 33a formed by a intramolecular ring-closure reaction and the

130 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

giant 96-membered C[11T-DA]2 dimer 33b resulting from an intermolecular cyclic formation.

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu

S S S S S R S S S S S S R

KOH 31 R = Si(CH3)3 MeOH / THF 32 R = H

Cu(OAc)2 x H2O / CuCl pyridine/dichloromethane pseudo high dilution Bu Bu Bu Bu Bu Bu S S S S Bu Bu Bu Bu S S S S Bu S Bu S S Bu Bu Bu S S Bu S S Bu Bu S S + S Bu Bu S Bu S S S Bu Bu Bu S Bu S S S Bu S S S Bu Bu Bu Bu S S S S 33a Bu Bu Bu Bu Bu Bu C[11T-DA]1 33b total yield [60 %]

C[11T-DA]2

Scheme 4.24. Macrocyclization of diethynyl-undecithiophene 31 by a modified Eglinton coupling reaction.

The two macrocycles 33a and 33b were separated by chromatography on silica gel and isolated in 31 % and 17.8 % yield, respectively. Each individual macrocycle 33a and 33b was then characterized by MALDI-TOF MS and the purity verified by analytical HPLC. Due to the low amounts of products isolated characterization by NMR analysis has not been accomplished. The MALDI-TOF MS spectrum of macrocycle 33a shows only one signal at m/z = 1622.4 which corresponds well with the calculated mass of the macrocycle (m/z = 1622.4) (Figure 4.22). In Figure 4.23, the experimental isotopic distribution (bottom) is displayed which is fully consistent with the calculated one (top). In Figure 4.24

4.4 Macrocyclization by oxidative coupling reaction 131

the MALDI-TOF mass spectrum of cyclodimeric 33b is depicted. The spectrum recorded in a dithranol matrix shows characteristic signal at m/z = 3245.3 which corresponds well to the theoretical mass calculated for the compound (m/z = 3245.2). The experimental isotropic distribution was compared to that calculated and was found to be in very good agreement (Figure 4.25).

33a 1622 100

60 / a.u. I

40 33b 3245 20

0 1500 2000 2500 3000 3500 4000 m / z

Figure 4.21. MALDI-TOF MS spectrum of the product mixture C[11T-DA]n obtained from macrocyclization of diethynyl-undecithiophene 31.

132 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

1622.4 100

60 / a.u. I 40

0 1200 1400 1600 1800 2000 m / z

Figure 4.22. MALDI-TOF MS spectrum of macrocycle 33a C[11T-DA]1.

g 100 90 80 70 60 50 40 30 20 10 0 1.620 1.622 1.624 1.626 1.628 1.630

100

0 1620 1622 1624 1626 1628 1630 1632 m / z

Figure 4.23. Calculated (top) and measured (bottom) mass isotropic distribution of macrocycle 33a.

4.4 Macrocyclization by oxidative coupling reaction 133

3245.3 100

60 / a.u. I 40

0 2000 2400 2800 3200 3600 4000 4400 m / z

Figure 4.24. MALDI-TOF MS spectrum of the giant macrocycle 33b C[11T-DA]2.

100 90 80 70 60 50 40 30 20 10 0 3.245 3.250 3.255

100

0 3245 3250 3255 m / z

Figure 4.25. Calculated (top) and measured (bottom) mass isotropic distribution of macrocycle 33b.

134 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

4.5 Conclusion

In this chapter the synthesis and characterization of two homologues series of conjugated macrocyclic oligothiophene-diacetylenes macrocycles has been described. As starting materials for the macrocyclization reaction terminally diethynylated oligothiophenes were employed. The synthesis of the homologue series of butylated oligothiophenes 23-25 was based on the effective combination of metal-mediated coupling reactions and ring-closure reaction of butadiynes precursors with sulfide nucleophiles. A novel palladium-mediated coupling reaction was developed in which ethynyl-oligothiophenes can be converted under mild conditions to the corresponding bis(oligothienyl)-butadiynes in high yields. The ring- closure reaction of butadiynes with sulfide nucleophiles to the correponding thiophenes has been optimized and a proposed mechanism was presented. By oxidative coupling reaction of diethynyl-terthiophene precursor 23 under pseudo-high dilution conditions a complete series of cyclic homologues C[3T-DA]n containing 3 up to 10 terthiophene units linked by diacetylenes bridges have been obtained. The variation of the reaction conditions, especially the change of the catalyst system from CuCl/CuCl2 to

Cu(OAc)2/CuCl, revealed an improvement in the cyclization results. Macrocyclization of diethynyl-quinquethiophene 24 under similar optimized reaction conditions resulted in the formation of cyclic homologues C[5T-DA]n containing 2 up to 6 quinquethiophene units linked by diacetylene bridges. Although the preparation of the necessary terminally diethynylated quinquethiophene precursor 24 required longer time due to more synthetic steps, macrocyclization of this longer oligomer has proven to be superior to the previous cyclization reaction which started with the corresponding terthiophene derivative 23. The final ring-closure step afforded the formation of cyclic products in higher total yield, as expected. Furthermore, formation of less macrocycles resulted in a better accessibility to specific macrocycles. Comparing the two homologues series, C[3T-DA]n and C[5T-DA]n, it is interesting to note that in the first one, the tetrameric cycle C[3T-DA]4 which comprises 64 chain members along the conjugation path was preferentially formed. In the

C[5T-DA]n series the major product obtained was the trimeric macrocycle C[5T-DA]3 with

72 chain members. The cyclotrimer C[3T-DA]3 and the cyclodimer C[5T-DA]2 both having the same number chain members (48), were obtained in lower amounts. Evidently, the product distribution is predominantly determined by the size, geometry and thus, the strain energy of the ring to be formed and not only by the statistics.

4.5 Conclusion 135

Another strategy that has been investigated towards high-yielding preparation of macrocyclic oligothiophene-diacetylenes was the intramolecular ring-closure approach. As appropriate monomeric precursor diethynyl-undecithiophene 31 has been chosen. The oxidative coupling reaction under pseudo-high dilution conditions led to an intramolecular cyclic formation affording macrocycle 33a as the main product in 31 % yield. By an intermolecular ring-closure reaction cyclodimer 33b was formed in 17.1 % yield. Although the macrocyclic structures could be isolated in higher yields, the method has its inherent limitation in the multi-step synthesis of the precursor. Beside the time-consuming factor, the overall yield of macrocycle 31, if one would consider ethynyl-quinquethiophene 13 as the starting material, is only 8 %.

4.6 Experimental section

4.6.1 Instrumentation and general experimental conditions

Solvents and reagents were purified and dried by usual methods prior to use.

Thin-layer Chromatography (TLC) was carried out on plastic plates Polygram SIL

G/UV254 from Macherey & Nagel. Developed plates were dried and examined under a UV lamp.

Preparative column chromatography was performed on glass columns of different sizes packed with silica gel 60 (particle size 0.040-0.063 mm), Merck.

Gas chromatography (GC) was executed with a Carlo-Erba Auto-HRGC MFC 500 equipped with PS086 glass capillary columns (Ø 0.32 mm, length 10 and 20 m). Helium 5.0 was used as carrying gas; eluted materials were detected by a Carlo-Erba EL 580 flame-ionization detector (FID). Chromatograms were recorded on a Spectra-Physics DP 700 integrator.

Gas chromatography – Mass Spectrometry (GC-MS) measurements were executed with a Varian 3800 equipped with CP4860 glass capillary columns (length 30 m). Helium 5.0 was used as carrying gas. Mass spectra were recorded on a Varian Saturn 2000. Ions were generated by electron impact (EI).

136 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Melting points were determined with a Büchi B-545 melting point apparatus and are uncor- rected.

FT-IR spectroscopy was performed on a Perkin-Elmer Spectrum 2000 spectrometer. Samples were prepared as KBr pellets. Band positions are reported in reciprocal centimetres. Relative intensities of single absorption bands are indicated by s = strong, m = middle, w = weak and br = broad.

NMR spectroscopy measurements were carried out on Bruker AMX 500 and DRX 400 spectrometers. Chemical shifts are expressed in parts per million (δ) downfield from the internal tetramethylsilane reference (δH = 0.00) or using residual solvent protons as internal standards (CDCl3: δH = 7.26, δC = 77.0). Spin multiplicities are indicated by the following symbols: s (singlet), d (doublet), t (triplet), m (multiplet), dd (doublet of doublets).

Mass spectra were recorded on a Varian MAT 711 spectrometer. Ions were generated by electron impact (EI) at 70 eV. Relative intensities of single peaks with respect to the base peak are given in parentheses. In some cases assignments of corresponding fragments are appended in square brackets. Matrix-assisted laser desorption ionization time-of-flight mass spectroscopy (MALDI-TOF MS) measurements were carried out on a Brucker Daltonik Reflex III spectrometer by using 1,8,9-trihydroxyanthracene (Dithranol) as matrix. Solutions of the compounds in tetrahydrofurane or dichloromethane were prepared. The calculated and experimental m/z values refer to the monoisotopic mass unless otherwise stated. High resolution fast bombardment mass spectra (HRMS, FAB) were recorded on a Finnigan MAT 95 spectrometer (University Stuttgart, Germany) by using nitrobenzylalcohol as matrix. ESI-FTICR MS measurements were carried out on a BrukerApex IV mass spectrometer (University Bonn, Germany). Solutions of the compounds in a mixture of tetrahydrofurane, water and acetic acid were prepared. The calculated m/z values correspond to the average masses and the experimental (found) m/z values correspond to the respective peak with the highest intensivity.

HPLC analysis was accomplished with a Shimadzu SCL-10AVP controller, LC-10AT pump, SPD-M10A VP detector using a Macherey-Nagel nucleosil column NO2 (4 mm x 250 mm, corn diameter 5µm), 1,3 mL/min flow rate.

Preparative HPLC was performed with a Shimadzu LC-8A pump, SPD-10A detector using a Macherey-Nagel nucleosil column NO2 (40 mm x 250 mm, corn diameter 100- 10µm), 80 mL/min flow rate.

4.6 Experimental section 137

UV/VIS/NIR spectra were recorded on a Perkin-Elmer Lambda 19 spectrometer in 1 cm cu- vettes.

Elemental analyses were performed on an Elementar Vario EL (University Ulm) and a Carlo Erba 1104 (University Stuttgart) (limit of experimental error: ± 0.3 %).

Nomenclature: All compounds were named according to IUPAC rules using ”ISIS-Draw 2.5 Plug-In Autonom-Standard” or ”IUPAC name Pro 6.0” programms.

Purchased starting materials and chemicals:

Acetic acid (Merck), [1,3-bis(diphenylphosphino)propane]nickel(II)-chloride (Aldrich), bis(triphenylphosphine)-palladium(II)-chloride (Merck), N-bromosuccinimide (Aldrich), 2- bromothiophene (Aldrich), butylbromide (Merck), chloroform-d3 with TMS (Merck), copper(II) acetate monohydrate (Merck), copper(I)-iodide (Aldrich), copper(I) chloride (Merck), copper(II)-chloride (Merck), 2,5-dibromothiophene (ABCR), N,N- Dimethylformamide (Aldrich), dimethylsulfoxide (Merck), iodine (Merck), magnesium (Merck), mercuric acetate (Merck), pyridine (Merck), sodium hydrogen carbonate (Merck), sodium hydrogensulfid hydrate (Fluka), sodium sulfate (Merck), sodium thiosulfate (Merck), thiophene (Merck), triethylamine (Merck), trimethylsilylacetylene (Merck), triphenylphosphine (Merck).

Starting materials prepared according to literature procedures: tetrabromothiophene,22 3,4-dibromothiophene,22 3,4-dibromothiophene.46

4.6.2 Synthesis and characterization of the compounds

General Procedures (GPs)

GP1: Bromination of thiophenes with NBS

Under exclusion of light, a solution of NBS in absolute DMF was added dropwise over a period of several hours to a solution of thiophene in absolute DMF at 0°C. The mixture was then allowed to warm to room temperature and stirred for several hours. After completion the reaction mixture was poured into ice and extracted with dichloromethane.

The organic phase was washed with saturated NaHCO3 solution and water and was dried

138 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

over Na2SO4. After evaporation of the solvent the crude product was purified either by distillation under reduced pressure or by column chromatography on silica gel (eluent).

GP2: Nickel-catalyzed cross-coupling (Kumada coupling) of halogenated thiophenes and metallated thiophenes

A solution of the bromothiophene in absolute diethyl ether was added dropwise to magnesium chips in boiling absolute diethyl ether. The resulting mixture was heated at reflux for several hours, allowed to cool to room temperature and with a syringe transferred to a dropping funnel of a second apparatus. The Grignard solution was then added dropwise to a solution of the dibrominated thiophene and 1,3- bis(diphenylphosphino)propane nickel(II) chloride in absolute diethyl ether. The resulting mixture was heated at reflux for several hours, cooled to 0°C, acidified with 1N HCl and extracted with dichloromethane. The organic phase was washed with saturated NaHCO3 solution and water and was dried over Na2SO4. After evaporation of the solvent the crude product was purified by column chromatography on silica gel (eluent).

GP3: Palladium-catalyzed cross-coupling (Sonogashira-Hagihara coupling) of haloga- nated thiophenes and trimethylsilylacetylene

Halogenothiophene, bis(triphenylphosphino)palladium(II)chloride, triphenylphosphine, copper(I) iodide, absolute pyridine and absolute triethylamine were loaded into a thick- walled Schlenk flask. The suspension was purged with argon for 20 minutes and heated to 60°C. After trimethylsilylacetylene was added by means of a syringe the apparatus was closed and the reaction mixture was stirred for several hours at 80°C. After completion, the reaction mixture was poured to ice cooled 1N HCl solution and extracted with dichloromethane. The organic phase was washed with saturated NaHCO3 solution and water and was dried over Na2SO4. After evaporation of the solvent the crude product was purified by column chromatography on silica gel (eluent).

4.6 Experimental section 139

GP4: Palladium-promoted dimerization of ethynylated oligothiophenes

A solution of potassium hydroxide in aqueous methanol was dropped to a solution of TMS- protected ethynyl-oligothiophene in tetrahydofuran. After stirring at ambient temperature for 2 hours ice was added to the reaction mixture which subsequently was extracted with dichloromethane. The organic layer was washed with 1N HCl, saturated NaHCO3 solution and water and was dried over Na2SO4. Under cooling the solution was concentrated to a small volume. The solution of deprotected ethynyloligothiophene was then added to a stirred solution of bis(triphenylphosphino) palladium(II) chloride, copper(I) iodide, triethylamine in tetrahydrofuran. The reaction mixture was stirred for 12 hours at ambient temperature, acidified with 1N HCl and then extracted with dichloromethane. The organic phase was washed with saturated NaHCO3 solution and water and was dried over Na2SO4. After evaporation of the solvent the crude product was purified by column chromatography on silica gel (eluent).

GP5: Ring closure of 1,3-butadiynes with sulfide anions to thiophenes

A solution of the butadiyne, sodium hydrogen sulfide hydrate, potassium hydroxide in dimethoxysulfoxide was stirred at 80°C for several hours. After removal of the solvent by vacuum distillation, water and dichloromethane were added. The organic layer was separated and then washed for several times with water and dried over Na2SO4. After evaporation of the solvent the crude product was purified by column chromatography on silica gel (eluent).

GP6: Iodination of thiophenes To a solution of thiophene in dichloromethane was added mercuric acetate and acetic acid. The reaction was stirred at ambient temperature for 8 hours. After addition of iodine at 0°C the reaction was allowed to warm to room temperature and stirred for 3 hours. The reaction mixture was quenched with saturated NaS2O3 solution and extracted with dichloromethane.

The organic layer was washed with saturated NaHCO3 solution and water and was dried over Na2SO4. After evaporation of the solvent the crude product was purified by column chromatography on silica gel (eluent).

140 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Preparation of the compounds

3,4-Dibutylthiophene (1) A solution of butylbromide (184 g, 1.36 mol) in absolute diethyl ether (300 mL) was added dropwise to magnesium chips (36.5 g, 1.50 mol) in boiling absolute diethyl ether (150 mL). The resulting mixture was heated at reflux for 3 h, allowed to cool to room temperature and under inert atmosphere transferred to a dropping funnel of a second apparatus. The Grignard solution was then added dropwise to a solution of the 3,4- dibromothiophene (132 g, 546 mmol) and 1,3-bis(diphenylphosphino)propane nickel(II) chloride (5.92 g, 10.9 mmol, 2 mol %) in 350 ml absolute diethyl ether. The resulting mixture was heated at reflux for 24 h, cooled to 0°C, acidified with 1N HCl and extracted with dichloromethane. The organic phase was washed with saturated NaHCO3 solution and water and was dried over Na2SO4. After evaporation of the solvent the crude product was purified by distillation and isolated as a colorless oil in 72 % yield (196.5 mmol, 77.2 g), bp 83°C/ 5x10-2 mbar.

1 3 H NMR (400MHz, CDCl3): δ= 6.89 (s, 2 H, H-2,5-Th), 2.53 (t, J(α,β-CH2-Bu) = 7.6 Hz, 4 H, 3 α-CH2-Bu), 1.58 (m, 8 H, β,γ-CH2-Bu), 0.97 (t, J(H-CH3, β-CH2-Bu) = 7.1 Hz, 6 H, CH3-Bu).

γ β α 3 4 2 5 S 1

13 C NMR (101 MHz, CDCl3): δ= 141.9 (C-3,4-Th), 119.8 (C-2,5-Th), 31.8, (β-CH2-Bu),

28.5 (α-CH2-Bu), 22.9 (γ-CH2-Bu), 13.9 (CH3-Bu). The analytical data are consistent with those described in literature.47

2-Bromo-3,4-dibutylthiophene (2)

The synthesis was carried out according to GP1, NBS (25.2 g, 142 mmol) in DMF (200 mL), dibutylthiophene 1 (25.3 g, 129 mmol) in 200 ml DMF, 8 h, 12 h, distillation, yield 82 % (29.1 g, 106 mmol), colorless oil, bp 74°C/ 10-3 mbar.

4.6 Experimental section 141

1 H NMR (400MHz, CDCl3): δ= 6.86 (s, 1 H, H-5-Th), 2.54 (m, 4 H, α-CH2-Bu), 1.52 (m,

8 H, β,γ-CH2-Bu), 0.97 (m, 6 H, CH3-Bu).

γ β α 4 2 5 S Br 2

13 C NMR (101 MHz, CDCl3): δ= 141.9 (C-4-Th), 141.9 (C-3-Th), 119.9 (C-5-Th), 109.2

(C-2-Th), 31.6, 31.5 (β-CH2-Bu), 29.3, 27.8 (α-CH2-Bu), 22.7, 22.5 (γ-CH2-Bu), 13.9

(CH3-Bu). The analytical data are consistent with those described in literature.20

2,5-Dibrom-3,4-dibutylthiophene (3)

The synthesis was carried out according to GP1, NBS (41.6 g, 233 mmol) in DMF (200 mL), dibutylthiophene 1 (19.1 g, 97.3 mmol) in 150 ml DMF, 4 h, 12 h, distillation, yield 90 % (31.0 g, 87.6 mmol), yellow oil, bp 112°C/ 10-3 mbar.

1 H NMR (400MHz, CDCl3): 2.45 (m, 4 H, α-CH2-Bu), 1.42 (m, 8 H, β,γ-CH2-Bu), 0.95 (t, 3 J(H-CH3, β-CH2-Bu) = 7.2 Hz, 6 H, CH3-Bu).

γ β α 4 5 2 Br S Br 3

13 C NMR (101 MHz, CDCl3): δ= 141.0 (C-3,4-Th), 108.0 (C-2,5-Th), 31.8, (β-CH2-Bu),

28.9 (α-CH2-Bu), 22.8 (γ-CH2-Bu), 14.1 (CH3-Bu). The analytical data are consistent with those described in literature.20

142 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

3,3’’,4,4’’-Tetrabutyl-2,2’:5’,2’’-terthiophene (4)

The synthesis was carried out according to GP2, bromothiophene 2 (27.5 g, 100 mmol), absolute diethyl ether (150 mL), magnesium (2.90 g, 120 mmol), 4 h; 2,5- dibromothiophene (11.0 g, 45.5 mmol), Ni(dppp)Cl2 (493 mg, 910 µmol, 2 mol %) in absolute diethyl ether (150 mL), 36 h; chromatographic work-up (petrol ether), yield 84 % (39.8 g, 84.1 mmol), yellow plates, m.p. 50°C.

2 2'' 5 S 5'' S S 3' 4' 4

The analytical data are consistent with those described in literature (m.p. 49-50°C).20

3’,4’-Dibutyl-2,2’:5’,2’’-terthiophene (5)

The synthesis was carried out according to GP3, 2-bromothiophene (30.7 g, 189 mmol), absolute diethyl ether (200 mL), magnesium (5.50 g, 226 mmol), 3 h; dibromothiophene 3

(22.3 g, 62.9 mmol), Ni(dppp)Cl2 (682 mg, 1.26 mmol, 2 mol %) in absolute diethyl ether (150 mL), 35 h; chromatographic work-up (petrol ether), yield 82 % (18.6 g, 51.6 mmol), yellow oil.

4' 3' S S 5'' 5 2'' S 2 5 The analytical data are consistent with those described in literature (yellow oil).25

4.6 Experimental section 143

5,5’’-Dibromo-3’,4’-dibutyl-2,2’:5’,2’’-terthiophene (6)

The synthesis was carried out according to GP1, NBS (13.3 g, 74.6 mmol) in DMF (100 mL), terthiophene 5 (12.2 g, 33.9 mmol) in DMF (100 mL), 5h, 8h, chromatographic work-up (cyclohexane/dichloromethane 20/1), yield 89 % (15.6 g, 30.2 mmol), yellow solid, m.p. 39°C.

4' 3' Br S S Br 5'' 2'' S 2 5 6

The analytical data are consistent with those described in literature (yellow solid, m.p. 39°C).20

3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethiophene (7)

The synthesis was carried out according to GP2, bromothiophene 2 (9.16 g, 33.3 mmol), absolute diethyl ether (20 mL), magnesium (0.97 g, 39.9 mmol), 4 h; dibromoterthiophene

6 (5.32 g, 11.1 mmol), Ni(dppp)Cl2 (120 mg, 222 µmol, 2 mol %) in absolute diethyl ether (100 mL), 72 h; chromatographic work-up (petrol ether/dichloromethane 15/1), yield 87% (7.21 g, 9.62 mmol), orange solid, m.p. 60°C.

2'''' S S 2 5'''' 5 S S 5' 2' S 7

The analytical data are consistent with those described in literature (orange crystals, m.p. 60-61°C).20

144 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

5-Bromo-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene (8)

The synthesis was carried out according to GP1, NBS (1.88 g, 10.6 mmol) in DMF (150 ml), terthiophene 4 (4.50 g, 9.62 mmol) in DMF (75 mL), 8 h, 24 h, chromatographic work-up (petrol ether), yield 65 % (3.45 g, 6.25 mmol), yellow oil.

2'' S 2 5 5'' Br S 2' S 4' 3' 8 The analytical data are consistent with those described in literature (yellow oil).20

5,5’’-Dibromo-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene (9)

This compound was isolated as byproduct from the synthesis of the corresponding monobromo derivative 8, yield 18 % (1.09 g, 1.73 mmol), yellow solid, m.p. 36°C.

5'' 2'' S 2 5 Br S S Br 4' 3' 9 The analytical data are consistent with those described in literature (yellow solid, m.p. 35- 36°C).20

5-Bromo-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethiophene (10)

The synthesis was carried out according to GP1, NBS (835 mg, 4.70 mmol) in DMF (150 mL), quinquethiophene 7 (2.93 g, 3.91 mmol) in DMF (100 mL), 8 h, 12 h,

4.6 Experimental section 145

chromatographic work-up (n-hexane/dichloromethane 12/1), yield 57 % (1.85 g, 2.22 mmol), yellow solid, m.p. 43°C.

2'''' S S 2 5 5'''' Br S S 5' 2' S

10 The analytical data are consistent with those described in literature (yellow solid, 42°C).20

5,5’’’’-Dibromo-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethiophene (11)

This compound was isolated as by-product from the synthesis of the corresponding monobromo derivative 10, yield 25 % (887 mg, 0.978 mmol), orange solid, m.p. 71°C.

5'''' 2'''' S S 2 5 Br Br S S 5' 2' S 11 The analytical data are consistent with those described in literature (orange solid, m.p. 70- 71°C).20

3,3’’,4,4’’-Tetrabutyl-5-trimethylsilylethynyl-2,2’:5’,2’’-terthiophene (12)

The synthesis was carried out according to GP3, bromothiophene 8 (2.40 g, 4.35 mmol),

Pd(PPh3)2Cl2 (153 mg, 218 µmol, 5 mol %), PPh3 (114 mg, 436 µmol, 10 mol %), CuI (82.8 mg, 436 µmol, 10 mol %), pyridine (25 mL), triethylamine (25 mL), trimethylsilylacetylene (512 mg, 5.22 mmol), 3 h; chromatographic work-up (n- hexane/dichloromethane 20/1), yield 94% (2.32 g, 4.07 mmol), yellow oil.

146 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

2'' S 2 5 5'' S 2' S Si(CH3)3 4' 3' 12 The analytical data are consistent with those described in literature (yellow oil).20

3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-5-trimethylsilylethynyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethiophene (13)

The synthesis was carried out according to GP3, monobromoquinquethiophene 10 (1.70 g,

2.05 mmol), Pd(PPh3)2Cl2 (71.9 mg, 103 µmol, 5 mol %), PPh3 (37.6 mg, 144 µmol, 10 mol %), CuI (27.4 mg, 144 µmol, 10 mol %), pyridine (15 mL), triethylamine (15 mL), trimethylsilylacetylene (242 mg, 2.46 mmol), 3 h; chromatographic work-up (n- hexane/dichloromethane 15/1), yield 95% (1.66 g, 1.96 mmol), orange solid, m.p. 61°C.

5'''' 2'''' S S 2 5 S S 5' 2' S Si(CH3)3 13

The analytical data are consistent with those described in literature (orange solid, m.p. 61°C).20

3,4-Dibutyl-2-trimethylsilylethynyl-thiophene (14)

The synthesis was carried out according to GP3, bromothiophene 2 (5.00 g, 18.2 mmol),

Pd(PPh3)2Cl2 (638 mg, 0.91 mmol, 5 mol %), PPh3 (333.6 mg, 1.27 mmol, 7 mol %), CuI (242.2 mg, 1.27 mmol, 7 mol %), pyridine (20 mL), triethylamine (20 mL),

4.6 Experimental section 147

trimethylsilylacetylene (2.67 g, 27.2 mmol), 6 h; chromatographic work-up (petrol ether), yield 82% (4.36 g, 14.9 mmol), yellow oil.

γ β α' α 2 5 S Si(CH3)3 14 The analytical data are consistent with those described in literature (yellow oil, b.p. 103°C/1x10-2 bar).20

1,4-Bis(3,4 –dibutylthien-2-yl)-1,3-butadiyne (15)

The synthesis was carried out according to GP4, KOH (956 mg, 17.1 mmol) in methanol (15 mL), protected ethynyloligothiophene 14 (0.50 g, 1,71 mmol) in THF (15 mL);

Pd(dppp)Cl2 (50.4 mg, 85.5 µmol, 5 mol %), CuI (32.6 mg, 171 µmol, 10 mol %), NEt3 (346 mg, 3.42 mmol) in THF (50 mL); chromatographic work-up (petrol ether), yield 94% (353 mg, 0.804 mmol), yellow solid, m.p. 60°C.

γ β α α α' α' 2 2 5 5 S S 15 The analytical data are consistent with those described in literature (yellow crystals, m.p. 60-61°C).20

1,4-Bis(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-2-yl)-1,3-butadiyne (16)

The synthesis was carried out according to GP4, KOH (1.77 g, 31.6 mmol) in methanol (20 mL), protected ethynyloligothiophene 12 (1.80 g, 3.16 mmol) in THF (20 mL);

Pd(dppp)Cl2 (93.2 mg, 158 µmol, 5 mol %), CuI (60.2 mg, 316 µmol, 10 mol %) in THF

148 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

(80 mL); chromatographic work-up (petrol ether/dichloromethane 15/1), yield 91% (1.43 g, 1.44 mmol), orange solid, m.p. 73°C.

γ β α α α' α'' 2 2 5 5 α''' S 2'' S S S 2'' 16 5'' S 3' 3' S 5'' 4' 4'

The analytical data are consistent with those described in literature (orange solid 72- 73°C).20

1,4-Bis(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethien-5-yl)- 1,3-butadiyne (17)

The synthesis was carried out according to GP4, KOH (975 mg, 17.4 mmol) in methanol (10 mL), protected ethynyloligothiophene 13 (1.47 g, 1.74 mmol) in THF (25 mL);

Pd(dppp)Cl2 (51.3 mg, 87.0 µmol, 5 mol %), CuI (23.2 mg, 121 µmol, 10 mol %) in THF (100 mL); chromatographic work-up (petrol ether/dichloromethane 10/1), yield 61% (0.82 g, 0.53 mmol), red solid, m.p. 85°C.

5 5 2 2 2'' S S S S 5'' 17 2'''' S S S S 2'''' 5'''' 5'''' S S

The analytical data are consistent with those described in the literature (red solid, m.p. 85- 87°C).20

4.6 Experimental section 149

3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Octabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,

2’’’’’:5’’’’’,2’’’’’’-septithiophene (18)

The synthesis was carried out according to GP5, 1,3-butadiyne 16 (1.95 g, 1.96 mmol), NaHS hydrate (1.10 g, 19.6 mmol), KOH (549 mg, 9.80 mmol) in DMSO (50 mL), 4 h; chromatographic work-up (petrol ether/dichloromethane 12/1), yield 77% (1.54 g, 1.50 mmol), red solid, m.p. 74°C.

2'''''' S S S 2 5'''''' 5 S S S 5' 2' S 4''''' 3''''' 4' 3' 18 The analytical data are consistent with those described in literature (red solid, 73-74°C).20

3,3’’,3’’’’,3’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,4’’’’’’’’’’-Dodecabutyl-2,2’:5’,2’’:5’’,

2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-undecithiophene (19)

The synthesis was carried out according to GP5, 1,3-butadiyne 17 (200 mg, 129 µmol), NaHS hydrate (119 g, 1.29 mmol), KOH (36.1 mg, 645 µmol) in DMSO (25 mL), 4 h; chromatographic work-up (petrol ether/dichloromethane 10/1), yield 72 % (146.6 mg, 92.8 µmol), red solid, m.p.83 °C.

2'''''''''' S S S S 5'' 2'' S 2 5'''''''''' 5 S S S S S S 4''''' 3''''' 19 The analytical data are consistent with those described in literature (red solid, m.p.83- 84°C).20

150 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

5,5’’-Diiodo-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene (20) The synthesis was carried out according to GP6, terthiophene 4 (10.0 g, 20.9 mmol) in dichloromethane (200 mL), mercuric acetate (14.7 g, 46.0 mmol), acetic acid (2.76 g, 46.0 mmol), iodine (11.7 g, 46.0 mmol); chromatographic work-up (petrol ether); yield 91 %

(13.8 g, 19.0 mmol), yellow solid, m.p 57°C, HPLC (n-hexane/dichloromethane 98/2): tR

(λmax) = 4.8 min (347 nm).

1 3 H NMR (400MHz, CDCl3): δ= 6.98 (s, 2 H, H-3’,4’-Th), 2.74 (t, J(α’,β’-CH2-Bu);(α’’,β’’-CH2-Bu) 3 = 7,8 Hz, 4 H, α’,α’’-CH2-Bu), 2.53 (t, J(α,β-CH2-Bu); α’’’,β’’’-CH2-Bu) = 7.6 Hz, 4 H, α,α’’’-CH2-

Bu), 1.48 (m, 16 H, β-β’’’,γ- γ’’’-CH2-Bu), 0.98 (m, 12 H, CH3-Bu).

β α''' α'' α' α

5'' 2'' S 2 5 I S S I 4' 3' 20

13 C NMR (101 MHz, CDCl3): δ= 147.5 (C-2’,5’-Th), 138.7 (C-4,4’’-Th), 135,8 (C-

2,2’’,3,3’’-Th), 126.2 (C-3’,4’-Th), 74.2 (C-5,5’’-Th), 33.0, 32.1, (β-β’’’-CH2-Bu), 31.0,

28.3 (α-α’’’-CH2-Bu), 22.9 (γ-γ’’’-CH2-Bu), 13.9, 13.8 (CH3-Bu). MS (EI, 70 eV): m/z (%) = 724 [M+].

Elemental analysis C28H38I2S3 (724.62) calcd. C 46.41 H 5.29 S 13.27 found C 46.36 H 5.33 S 13.43.

5,5’’’’-Diiodo-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinque thiophene (21) The synthesis was carried out according to GP6, quinquethiophene 7 (2.82 g, 3.76 mmol) in dichloromethane (100 mL), mercuric acetate (2.63 g, 8.27 mmol), acetic acid (497 mg, 8.27 mmol), iodine (2.10 g, 8.27 mmol); chromatographic work-up (petrol ether); yield 88 % (3.31 g, 3.30 mmol), yellow solid, m.p. 55°C, HPLC (n-hexane/dichloromethane 93/7): tR (λmax) = 4.3 min (390 nm).

4.6 Experimental section 151

1 3 H NMR (400MHz, CDCl3): δ= 7.07 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.7 Hz, 2 H, H-3’,4’’’- 3 Th), 7.00 (d, J(H-3’’’,4’’’-Th); (H-3’,4’-Th) = 3.7 Hz, 2 H, H-3’’’,4’-Th), 2.75 (m, 8 H, α’-α’’’-CH2-

Bu), 2.54 (m, 4 H, α, α’’’’-CH2-Bu), 1.47 (m, 24 H, β-β’’’,γ- γ’’’-CH2-Bu), 0.96 (m, 18 H,

CH3-Bu).

5'''' 2'''' S S 2 5 I I S 5''' S 2' S 4''' 21 3'

13 C NMR (126 MHz, CDCl3): δ= 147.5, 140.2, 138.6, 136.3, 136.0, 135.5 (C- 2,2’,2’’,2’’’,2’’’’,3,3’’,3’’’’,4,4’’,4’’’’,5,5’’,5’’’-Th), 126.3, 125.9, (C-3’,3’’’,4’’,4’’’-Th),

74.1 (C-5,5’’’’), 33.0, 32.9, 32.1 (β-β’’’’’-CH2-Bu), 31.0, 28.3, 28.0 (α-α’’’’’-CH2-Bu),

23.0, 22.9, (γ-γ’’’’’-CH2-Bu), 13.9, 13.8 (CH3-Bu).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C44H58I2S5 1000.1; found 1000.1 [M+].

Elemental analysis C44H58I2S5 (1000.1) calcd. C 52.79 H 5.84 S 16.01 found C 52.73 H 5.89 S 15.84.

5,5’’’’’’-Diiodo-3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-octabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’, 2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-septithiophene (22) The synthesis was carried out according to GP6, septithiophene 18 (1.35 g, 1.32 mmol) in dichloromethane (150 mL), mercuric acetate (923 mg, 2.90 mmol), acetic acid (174 mg, 2.90 mmol), iodine (735 mg, 2.90 mmol); chromatographic work-up (petrol ether/dichloromethane 10/1); yield 82 % (1.38 g, 1.08 mmol), red solid, m.p. 56°C, HPLC

(n-hexane/dichloromethane 92/8): tR (λmax) = 8.0 min (406 nm).

1 3 H NMR (400MHz, CDCl3): δ= 7.09 (s, 2 H, H-3’’’,4’’’-Th), 7.08 (d, J(H-3’,4’-Th); (H- 3 3’’’’’,4’’’’’-Th) = 3.6 Hz, 2 H, H-3’’’’’,4’-Th), 7.01 (d, J(H-3’,4’-Th, H-3’’’’’,4’’’’’-Th) = 3.6 Hz, 2 H, H- 3 3’,4’’’’’-Th), 2.76 (m, 16 H, α’-α’’’’’-CH2-Bu), 2,54 (t, J(α,β-CH2-Bu); (α’’’’’’,β’’’’’’-CH2-Bu) = 7,8

Hz, 16 H, α,α’’’’’’-CH2-Bu), 1.49 (m, 32 H, β-β’’’’’’,γ-γ’’’’’’-CH2-Bu), 0.97 (m, 24 H,

CH3-Bu).

152 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

5'''''' 2'''''' S 5''' S 2''' S 2 5 I I S S S 5' 2' S 4''''' 3''''' 4' 3' 22

13 C NMR (101 MHz, CDCl3): δ= 147.5 (C2’’’, C5’’’), 140.3, 140.2, 138.7, 136.4, 136.0, 135.5, 129.9, 129.7, 126.3, 126.1, 126.0 (C-2,2’’,2’’’’,2’’’’’,2’’’’’,3,3’’, 3’’’,3’’’’,3’’’’’,3’’’’’’,4,4’’, 4’’’,4’’’’,4’’’’’,4’’’’’’,-5’’,5’’’’,5’’’’’-Th), 74.1 (C-5,5’’’’’’-

Th), 33.0, 32.9, 32.0 (β-β’’’’’’-CH2-Bu), 31.0, 28.3, 28.0 (α-α’’’’’’-CH2-Bu), 23.0, 22.9 (γ-

γ’’’’’’-CH2-Bu), 13.9 (CH3-Bu).

MS (MALDI-TOF): m/z calcd. monoisotopic mass for C60H78I2S7 1276.2; found 1277.1 [M+].

Elemental analysis C60H78I2S7 (1277.5) calcd. C 56.41 H 6.15 S 17.57, found C 56.25 H 6.21 S 17.48.

5,5’’-Bis(trimethylsilylethinyl)-3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthiophene (23) The synthesis was carried out according to GP3, diiodoterthiophene 20 (5.00 g, 6.90 mmol), Pd(PPh3)2Cl2 (242 mg, 345 µmol), PPh3 (181 mg, 690 µmol), CuI (131 mg, 690 µmol), pyridine (50 mL), triethylamine (50 mL), trimethylsilylacetylene (1.49 g, 15.2 mmol), 3 h; chromatographic work-up (n-hexane/dichloromethane 24/1), yield 92 % (4.22 g, 6.35 mmol), yellow solid, m.p. 65°C, HPLC (n-hexane/dichloromethane 98/2): tR (λmax) = 4.6 min (375 nm).

1 H NMR (400MHz, CDCl3): δ= 7.04 (s, 2 H, H-3’,4’-Th), 2.64 (m, 8 H, α-α’’’-CH2-Bu),

1.50 (m, 16 H, β-β’’’,γ-γ’’’-CH2-Bu), 0.95 (m, 12 H, CH3-Bu), 0.25 (s, 18 H, CH3-

Si(CH3)3).

4.6 Experimental section 153

5'' 2'' S 2 5 S S (CH3)3Si Si(CH3)3 4' 3' 23

13 C NMR (101 MHz, CDCl3): δ= 149.6 (C-2’,5’-Th), 138.5 (C-4,4’’-Th), 136.0, 131.5 (C- 2,2’’,3,3’’-Th), 126.3 (C-3’,4’-Th), 117.3 (C-5,5’’-Th), 102.0, 97.7 (C≡C), 32.7, 32.3 (β-

β’’’-CH2-Bu), 28.4, 27.7 (α-α’’’-CH2-Bu), 22.9, 22.8 (γ-γ’’’-CH2-Bu), 13.9 (CH3-Bu), -

0.06 (C-Si(CH3)3). + HRMS (FAB): C38H56S3Si2 [M ] calcd. 664.3082. found 664.3076.

Elemental analysis C38H56S3Si2 (665.23) calcd. C 68.61 H 8.49 S 14.46. found C 68.63 H 8.39 S 14.56.

5,5’’’’-Bis(trimethylsilylethynyl)-3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’, 2’’’:5’’’,2’’’’-quinquethiophene (24) A) The synthesis was carried out according to GP3, diiodoquinquethiophene 21 (2.35 g,

2.31 mmol), Pd(PPh3)2Cl2 (81.0 mg, 116 µmol), PPh3 (60.6 mg, 231 µmol), CuI (44.0 mg, 231 µmol), pyridine (15 mL), triethylamine (15 mL), trimethylsilylacetylene (0.68 g, 6.93 mmol), 3 h; chromatographic work-up (n-hexane/dichloromethane 20/1), yield 92 % (2.00 g, 2.13 mmol). B) The synthesis was carried out according to GP2, dibromoquinquethiophene 11 (4.61 g,

5.08 mmol), Pd(PPh3)2Cl2 (178 mg, 254 µmol), PPh3 (133 mg, 508 µmol), CuI (96.7 mg, 508 µmol), pyridine (15 mL), triethylamine (15 mL), trimethylsilylacetylene (1.19 g, 12.2 mmol), 3 h; chromatographic work-up (n-hexane/dichloromethane 20/1), yield 89 % (4.04 g, 4.52 mmol); orange solid, m.p.113°C, HPLC (n-hexane/dichloromethane 95/5): tR (λmax) = 4.5 min (400 nm).

1 3 H NMR (400MHz, CDCl3): δ= 7.09 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.9 Hz, 2 H, H-3’,4’’’- 3 Th), 7.08 (d, J(H-3’’’,4’’’-Th); (H-3’,4’-Th) = 3.9 Hz, 2 H, H-3’’’,4’-Th), 2.73 (m, 8 H, α-α’’’’-CH2-

Bu), 1.52 (m, 24 H, β-β’’’,γ- γ’’’-CH2-Bu), 0.97 (m, 18 H, CH3-Bu), 0.27 (s, 18 H, CH3-

Si(CH3)3).

154 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

5'''' 2'''' S S 2 5 S S 2' S (CH3)3Si Si(CH3)3 24

13 C NMR (126 MHz, CDCl3): δ= 149.6, 140.3, 138.4, 136.2, 135.8, 131.7 (C- 2,2’,2’’,2’’’,2’’’’,3,3’’,3’’’’,4,4’’,4’’’’,5’,5’’,5’’’-Th), 129.8, 126.3, (C-3’,3’’’,4’,4’’’-Th),

117.3 (C-5,5’’’’), 101.9, 97.8 (C≡C), 32.9, 32.8, 32.3 (β-β’’’’’-CH2-Bu), 28.4, 28.0, 27.8

(α-α’’’’’-CH2-Bu), 23.0, 22.9, 22.8 (γ-γ’’’’’-CH2-Bu), 13.8 (CH3-Bu), -0.06 (C-Si(CH3)3).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C54H76S5Si2. 940.4; found 940.5 [M+].

Elemental analysis C54H76S5Si2 (941.7) calcd. C 68.88 H 8.13 S 17.02. found C 68.92 H 8.37 S 16.88.

5,5’’’’’’-Bis(trimethylsilylethynyl)-3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-octabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-septithiophene (25) The synthesis was carried out according to GP3, diiodoseptithiophene 22 (1.50 g, 1.17 mmol), Pd(PPh3)2Cl2 (34.5 mg, 58.5 µmol), PPh3 (21.5 mg, 81.9 µmol), CuI (15.6 mg, 81.9 µmol), pyridine (10 mL), triethylamine (10 mL), trimethylsilylacetylene (253 mg, 2.57 mmol), 3 h; chromatographic work-up (n-hexane/dichloromethane 10/1), yield 90 % (1.28 g, 1.05 mmol), red solid, m.p. 123°C, HPLC (n-hexane/dichloromethane 92/8): tR (λmax) = 5.3 min (415 nm).

1 3 H NMR (400MHz, CDCl3): δ= 7.10 (s, 2 H, H-3’’’,4’’’-Th), 7.09 (d, J(H-3’,4’-Th); (H- 3 3’’’’’,4’’’’’-Th) = 3.6 Hz, 2 H, H-3’’’’’,4’-Th), 7.07 (d, J(H-3’,4’-Th); (H-3’’’’’,4’’’’’-Th) = 3.6 Hz, 2 H, 3 H-3’,4’’’’’-Th), 2.74 (m, 16 H, α’-α’’’’’-CH2-Bu), 2.56 (t, J(α,β-CH2-Bu); (α’’’’’’,β’’’’’’-CH2-Bu) =

7.9 Hz, 4 H, α,α’’’’’’-CH2-Bu), 1.49 (m, 32 H, β-β’’’’’’,γ-γ’’’’’’-CH2-Bu), 0.97 (m, 24 H,

CH3-Bu), 0.25 (s, 18 H, CH3-Si(CH3)3).

4.6 Experimental section 155

5'''''' 2'''''' S 5''' S 2''' S 2 5 S S S 5' 2' S (CH3)3Si Si(CH3)3 4''''' 3''''' 4' 3' 26

13 C NMR (101 MHz, CDCl3): δ= 147.5 (C2’’’, C5’’’), 140.2, 138.4, 136.3, 135.5, 129.9, 126.3, 126.0, (C-2,2’’,2’’’’,2’’’’’,2’’’’’,3,3’’,3’’’’,3’’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’,4’’’’’’,5’’,- 5’’’’,5’’’’’-Th), 126.2 (C-3’’’,4’’’-Th), 117.2 (C-5,5’’’’’’-Th), 100.0, 94.3 (C≡C), 33.0,

32.8, 32.3 (β-β’’’’’’-CH2-Bu), 28.4, 28.0, 27.8 (α-α’’’’’’-CH2-Bu), 23.1 (γ-γ’’’’’’-CH2-

Bu), 13.9 (CH3-Bu), -0.01 (C-Si(CH3)3).

MS (MALDI-TOF): m/z calcd. monoisotopic mass. for C70H96S7Si2 1216.5; found 1216.4 [M+]

Elemental analysis C70H96S7Si2 (1218.1) calcd. C 69.02 H 7.94 S 18.43, found C 69.08 H 8.05 S 18.39.

Macrocyclisation of the diethynyl-terthiophene building block 23

Method A) Glaser coupling reaction TMS-protected diethynyl-terthiophene 23 (1.00 g, 1.51 mmol) was dissolved in THF (10 mL) and a solution of potassium hydroxide (845 mg, 15.1 mmol) in aqueous methanol (10 mL) was added. The reaction mixture was stirred at ambient temperature for 3 h and then poured onto ice and extracted with dichloromethane. The organic layer was washed with

1N HCl, saturated NaHCO3 solution and water, and dried over Na2SO4. Under cooling the mixture was concentrated to a small volume and then diluted by addition of pyridine (300 mL). This solution of the deprotected diethynyl-terthiophene 26 was slowly dropped by a syringe pump (rate = 0.05 mL/min) within 100 h to a stirred suspension of copper(II) chloride (2.11 g, 15.1 mmol) and copper(I) chloride (10.6 g, 106.9 mmol) in pyridine (200 mL) at room temperature. After complete addition the reaction mixture was allowed to stir for additionally 168 h. The solvent was removed under reduced pressure and the residue was dissolved in dichloromethane. The organic phase was washed with 1N HCl, saturated

NaHCO3 solution and water, and dried over Na2SO4. In order to remove the polymeric material and copper salts, the crude product was filtered through a short column of silica

156 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

gel with n-hexane/dichloromethane mixture (1:1) as eluent. After evaporation of the solvent a mixture of cyclic products 27a-h (63.5 mg, 8.1% based on 23) was isolated as red microcrystalline solid. The separation of the obtained mixture of macrocycles 27a-h was accomplished by repeated preparative HPLC with n-hexane/dichloromethane (85/15) as eluent.

Method B) Eglinton coupling reaction TMS-protected diethynyl-terthiophene 23 (910 mg, 1.37 mmol) was dissolved in THF (5 mL) and a solution of potassium hydroxide (770 mg, 13.7 mmol) in aqueous methanol (5 mL) was added. After stirring at ambient temperature for 3 h the reaction mixture was poured on ice and extracted with dichloromethane. The organic layer was washed with 1N

HCl, then saturated NaHCO3 solution and water, and dried over Na2SO4. Under cooling the mixture was concentrated to a small volume and then diluted by addition of pyridine/dichloromethane (4:1, 250 mL). This solution of the deprotected diethynyl- terthiophene 26 was then dropped by a syringe pump (rate= 0.2 mL/min) within 21 h to a stirred suspension of copper acetate monohydrate (6.64 g, 34.3 mmol) and copper chloride (2.71 g, 27.4 mmol) in pyridine (250 mL) at room temperature. After complete addition the reaction mixture was allowed to stir for additionally 72 h. The solvent was removed under reduced pressure and the residue dissolved in dichloromethane. The organic phase was washed with 1N HCl, then saturated NaHCO3 solution and water, and dried over Na2SO4. In order to remote the polymeric material and copper salts, the crude product was filtered through a short column of silica gel with n-hexanes/dichloromethane mixture (1:1) as eluent. After evaporation of the solvent a mixture of cyclic products 27a-g (90.6 mg, 12.7% based on 23) was isolated as red microcrystalline solid. The separation of the obtained mixture of macrocycles 27a-g was accomplished by repeated preparative HPLC using n-hexane/dichloromethane (85/15) as eluent.

4.6 Experimental section 157

α''' 4' 2'' 3' 5'' S S 2 2' γ β S α 5 S

S S

S S n-2

27a-h (n = 3-10) C[3T-DA] n

Cyclo{tris[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’-diyl]} (27a) The compound was isolated after separation HPLC by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield method A: 1.0 % (7.51 mg, 4.82 µmol); method B: 2.6 % (18.6 mg, 12.0 µmol); m.p. >

300°C; HPLC (n-hexane/dichloromethane 85/15): tR = 6.2 min.

1 H NMR (500 MHz, CDCl3): δ = 7.07 (s, 6 H, H-3’,4’-Th), 2.66 (m, 24 H, α-α’’’-CH2-

Bu), 1.62, 1.47 (m, 48 H, β-β’’’,γ-γ’’’-CH2-Bu), 0.97 (m, 36 H, CH3-Bu). 13 C NMR (126 MHz, CDCl3): δ= 149.6 (C-2’,5’-Th), 138.3 (C-4,4’’-Th), 136.8, 133.8 (C- 2,2’’,3,3’’-Th), 124.9 (C-3’,4’-Th), 117.2 (C-5,5’’-Th), 81.9, 78.7 (C≡C), 32.7, 32.4 (β-

β’’’-CH2-Bu), 29.7, 28.7, 27.6 (α-α’’’-CH2-Bu), 22.9, 22.7 (γ-γ’’’-CH2-Bu), 13.9 (CH3- Bu).

UV/VIS (dichloromethane): λmax (ε) = 422 nm (133500).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C96H114S9 1554.5; found 1554.8 [M+]. + HRMS (ESI-FTICR) C96H114S9 (M ) calcd. 1554.6401; found 1554.6373.

158 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Cyclo{tetrakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (27b) The compound was isolated after HPLC separation by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield method A: 2.0 % (15.5 mg; 7.47 µmol); method B: 4.0 % (28.3 mg, 13.6 µmol); m.p. >

300°C; HPLC (n-hexane/dichloromethane 85/15): tR = 7,7 min.

1 3 H-NMR (500 MHz, CDCl3): δ= 7.08 (s, 8 H, H-3’,4’-Th), 2.69 (t, J(α,β-CH2-Bu)- (α’’’,β’’’-CH2-

Bu) = 7.8 Hz, 32 H, α-α’’’-CH2-Bu), 1.62, 1.54, 1.44 (m, 64 H, β-β’’’,γ-γ’’’-CH2-Bu), 0.99

(m, 48 H, CH3-Bu). 13 C NMR (126 MHz, CDCl3): δ= 150.8 (C-2’,5’-Th), 138.7 (C-4,4’’-Th), 136.5, 133.3 (C- 2,2’’,3,3’’-Th), 126.0 (C-3’,4’-Th), 117.0 (C-5,5’’-Th), 81.6, 78.9 (C≡C), 32.6 (β-β’’’-

CH2-Bu), 28.7, 27.6 (α-α’’’-CH2-Bu), 22.9, 22.7 (γ-γ’’’-CH2-Bu), 13.9 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 420 nm (162800).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C128H152S12 2072.9; found 2073.0 [M+].

Cyclo{pentakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (27c) The compound was isolated after HPLC separation by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red solid; yield method A: 1.6 % (12.4 mg, 4.77 µmol); method B: 1.4 % (10.1 mg, 3.89 µmol); m.p. > 300°C; HPLC (n- hexane/dichloromethane 85/15): tR = 8.8 min.

1 3 H NMR (500 MHz, CDCl3): δ= 7.09 (s, 10 H, H-3’,4’-Th), 2.70 (t, J(α,β-CH2-Bu)- (α’’’,β’’’-CH2-

Bu) = 7.8 Hz, 32 H, α-α’’’-CH2-Bu), 1.61, 1.54, 1.45 (m, 80 H, β-β’’’,γ-γ’’’-CH2-Bu), 0.99

(m, 60 H, CH3-Bu). 13 C NMR (126 MHz, CDCl3): δ= 151.1 (C-2’,5’-Th), 138.7 (C-4,4’’-Th), 136.3, 133.3 (C- 2,2’’,3,3’’-Th), 126.3 (C-3’,4’-Th), 116.8 (C-5,5’’-Th), 81.5, 77.9 (C≡C), 32.6 (β-β’’’-

CH2-Bu), 28.7, 27.7 (α-α’’’-CH2-Bu), 22.7 (γ-γ’’’-CH2-Bu), 13.9 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 431 nm (211300).

4.6 Experimental section 159

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C160H190S15 2591.1; found 2591.3 [M+].

Cyclo{hexakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (27d) The compound was isolated after HPLC separation by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red solid; yield method A: 0.8 % (5.80 mg, 1.86 µmol); method B: 0.5 % (3.60 mg, 1.16 µmol); m.p. > 300°C; HPLC (n- hexane/dichloromethane 85/15): tR = 11.3 min.

1 H NMR (500 MHz, CDCl3): δ= 7.09 (s, 12 H, H-3’,4’-Th), 2.71 (m, 48 H, α-α’’’-CH2-

Bu), 1.61, 1.46 (m, 96 H, β-β’’’,γ-γ’’’-CH2-Bu), 0.98 (m, 72 H, CH3-Bu). 13 C NMR (126 MHz, CDCl3): δ= 151.3 (C-2’,5’-Th), 138.8 (C-4,4’’-Th), 136.2, 133.3 (C- 2,2’’,3,3’’-Th), 126.5 (C-3’,4’-Th), 116.7 (C-5,5’’-Th), 81.5, 77.9 (C≡C), 32.6 (β-β’’’-

CH2-Bu), 28.7, 27.7 (α-α’’’-CH2-Bu), 22.7 (γ-γ’’’-CH2-Bu), 13.8 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 439 nm (212900).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C198H228S18 3109.3; found 3109.1 [M+].

Cyclo{heptakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (27e) The compound was isolated after HPLC separation as a red solid; yield method A: 0.7 % (5.70 mg, 1.57 µmol); method B: 0.2 % (1,30 mg, 0.36 µmol); HPLC (n- hexane/dichloromethane 82/18): tR = 5.8 min.

UV/VIS (dichloromethane): λmax = 443 nm; ε was not determined due to low amount of compound isolated.

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C224H266S21 3627.5; found 3628.4 [M+].

160 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Cyclo{octakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (27f) The compound was isolated after HPLC separation in trace amounts as red solid; yield method A and B: < 0.1 %; HPLC (n-hexane/dichloromethane 82/18): tR = 6.5 min.

UV/VIS (dichloromethane): λmax = 445 nm; ε was not determined due to low amount of compound isolated.

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C256H304S24 4145.7; found 4145.6 [M+].

Cyclo{nonakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (29g) The compound was isolated after HPLC separation in trace amounts as a red solid; yield method A and B: < 0.1 %; HPLC (n-hexane/dichloromethane 82/18): tR = 8.5 min.

UV/VIS (dichloromethane): λmax = 445 nm; ε was not determined due to low amount of compound isolated.

MS (MALDI-TOF) (linear modus) m/z: calcd. average mass for C288H342S27 4669.7; found 4672.9 [M+].

Cyclo{decakis[diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’- diyl]} (27h) The compound was isolated after HPLC separation in trace amounts as a red solid; yield method A: < 0.1 %; HPLC (n-hexane/dichloromethane 82/18): tR = 10.9 min.

MS (MALDI-TOF) (linear modus) m/z: calcd. average mass for C320H380S30 5188.5; found 5190.9 [M+].

Macrocyclisation of the diethynyl-quinquethiophene 24

TMS-protected diethynyl-terthiophene 24 (710 mg, 754 µmol) was dissolved in THF (10 mL) and a solution of potassium hydroxide (422 mg, 7.54 mmol) in aqueous methanol (10

4.6 Experimental section 161

mL) was added. After stirring at ambient temperature for 3 h the reaction mixture was poured onto ice and extracted with dichloromethane. The organic layer was washed with

1N HCl, saturated NaHCO3 solution and water, and then dried over Na2SO4. Under cooling the mixture was concentrated to a small volume and then diluted by addition of pyridine/dichloromethane (4:1, 140 mL). This solution of the deprotected diethynyl- quinquethiophene 28 was then dropped by a syringe pump (rate = 0.2 mL/min) within 12 h to a stirred suspension of copper(II) acetate monohydrate (3.74 g, 18.6 mmol) and copper(I) chloride (1.48 g, 15.0 mmol) in pyridine (140 mL) at room temperature. After complete addition the reaction mixture was allowed to stir for additionally 72 h. The solvent was removed under reduced pressure and the residue was dissolved in dichloromethane. The organic phase was washed with 1N HCl, then saturated NaHCO3 solution and water, and dried over Na2SO4. In order to remove the polymeric material and copper salts, the crude product was filtered through a short column of silica gel with n- hexane/dichloromethane mixture (1:1) as eluent. After evaporation of the solvent a mixture of cyclic products 29a-e (95.9 mg, 16.0 % based on 24) were isolated as a red microcrystalline solid. The separation of the obtained mixture of macrocycles 29a-e was accomplished by repeated preparative HPLC using n-hexane/dichloromethane (83/17) as eluent.

γ β α 5 3' 2 S S 4' S S 2'' S S 5'' S S 3''' S S 4''' 5'''' n-1 α'''''

29a-e (n = 2-6) C[5T-DA] n

162 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Cyclo{bis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl]} (29a) The compound was isolated after separation by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield 3.0 % (17.9 mg, 11.2 µmol); m.p. > 300°C; HPLC (n-hexane/dichloromethane 83/17): tR = 6.8 min.

1 H NMR (500 MHz, CDCl3): δ= 7.01 (s, 8 H, H-3’,3’’’,4’,4’’’-Th), 2.63 (m, 24 H, α-

α’’’’’-CH2-Bu), 1.55-1.36 (m, 48 H, β-β’’’’’, γ-γ’’’’’-CH2-Bu), 0.94 (m, 36 H, CH3-Bu). 13 C NMR (126 MHz, CDCl3): δ= 150.2, 140.3, 138.4, 137.5, 136.2, 133.9, 130.3 (C- 2,2’,2’’,2’’’,2’’’’,3,3’’,3’’’’,4,4’’,4’’’’,5’5’’,5’’’-Th), 125.2, 124.7 (C-3’,3’’’,4’,4’’’-Th),

117.2 (C-5,5’’’’-Th), 81.8, 78.4 (C≡C), 32.7, 32.4, 31.6 (β-β’’’’’-CH2-Bu), 29.7, 27.9, 27.7

(α-α’’’’’-CH2-Bu), 23.0, 22.9, 22.7 (γ-γ’’’’’-CH2-Bu), 13.8 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 421 nm (118800).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C96H116S10 1588.6; found 1588.3 [M+]. + HRMS (ESI-FTICR) C96H116S10 (M ) calcd. 1588.6278; found 1588.6243.

Cyclo{tris[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl]} (29b) The compound was isolated by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield 8.8 % (52.7 mg, 22.1

µmol); m.p. > 300°C; HPLC (n-hexane/dichloromethane 83/17): tR = 7.4 min.

1 3 H NMR (400 MHz, CDCl3): δ= 7.10 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.49 Hz, 6 H, H- 3 3’,4’’’-Th), 7.08 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.49 Hz, 6 H, H-3’’’,4’- Th), 2.71 (m, 36 H,

α-α’’’’’-CH2-Bu), 1.61, 1.45 (m, 72 H, β-β’’’’’, γ-γ’’’’’-CH2-Bu), 0.98 (m, 54 H, CH3-Bu). 13 C NMR (101 MHz, CDCl3): δ= 151.0, 140.4, 138.5, 136.8, 135.7, 133.6, 130.0 (C- 2,2’,2’’,2’’’,2’’’’,3,3’’,3’’’’,4,4’’,4’’’’,5’,5’’,5’’’-Th), 126.2, 125.7 (C-3’,3’’’,4’,4’’’-Th),

116.6 (C-5,5’’’’-Th), 81.4, 78.0 (C≡C), 32.8, 32.6 (β-β’’’’’-CH2-Bu), 27.9, 27.7 (α-α’’’’’-

CH2-Bu), 23.0, 22.9, 22.7 (γ-γ’’’’’-CH2-Bu), 13.9 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 425 nm (205200).

4.6 Experimental section 163

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C144H174S15 2382.9; found 2383.4 [M+].

Cyclo{tetrakis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl]} (29c) The compound was isolated after separation by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield 1.5 % (9.10 mg, 12.0 µmol); m.p. > 300°C; HPLC (n-hexane/dichloromethane 83/17): tR = 9.4 min.

1 3 H NMR (400 MHz, CDCl3): δ=7.11 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.49 Hz, 8 H, H-3’,4’’’- 3 Th), 7.09 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.49 Hz, 8 H, H-3’’’,4’- Th), 2.74 (m, 48 H, α-

α’’’’’-CH2-Bu), 1.61 (m, 48 H, β-β’’’’’-CH2-Bu), 1.47 (m, 48 H, γ-γ’’’’’-CH2-Bu), 0.98

(m, 72 H, CH3-Bu). 13 C NMR (101 MHz, CDCl3): δ= 151.0, 140.4, 138.5, 136.8, 135.7, 133.6, 130.0 (C- 2,2’,2’’,2’’’,2’’’’,3,3’’,3’’’’,4,4’’,4’’’’,5’,5’’,5’’’-Th), 126.2, 125.7 (C-3’,3’’’,4’,4’’’-Th),

116.6 (C-5,5’’’’-Th), 81.4, 78.0 (C≡C), 32.8, 32.6 (β-β’’’’’-CH2-Bu), 27.9, 27.7 (α-α’’’’’-

CH2-Bu), 23.0, 22.9, 22.7 (γ-γ’’’’’-CH2-Bu), 13.9 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 436 nm (235200).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C192H232S20 3177.3; found 3178.1 [M+].

Cyclo{pentakis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethien-5,5’’’’-diyl)-1,1’-diyl]} (29d) The compound was isolated as a red solid; yield 0.2 % (1.20 mg, 0.30 µmol); HPLC (n- hexane/dichloromethane 83/17): tR = 13.8 min.

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C240H290S25 3971.6; found 3972.8 [M+].

164 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Cyclo{hexakis[diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl]} (29e) The compound was isolated in trace amounts as red solid; yield < 0.1 %; HPLC (n- hexane/dichloromethane 82/18): tR = 12.5 min.

MS (MALDI-TOF) (linear modus) m/z: calcd. average mass for C288H348S30 4771.9; found 4770.9 [M+].

Intramolecular cyclisation of diethynyl-undecithiophene 31

5,5’’’’’’’’’’-Diiodo-3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,4’’’’’’’’,4’’’’’’’’’’- dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’:5’’’’’’,2’’’’’’’: 5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’-undecithiophene (30)

The synthesis was carried out according to GP6, undecithiophene 19 (162 mg, 103 µmol) in 20 ml dichloromethane, mercuric acetate (93.1 mg , 216 µmol), iodine (54.8 mg, 216 µmol); chromatographic work-up (petrol ether/dichloromethane 10/1), yield 65 % (123 mg, 67.2 µmol), red solid, HPLC (n-hexane/dichloromethane 90/10): tR (λmax) = 9.0 min (417 nm). 1 H-NMR (400MHz, CDCl3): δ= 7.10 (m, 6 H, H-3’’’,3’’’’’,3’’’’’’’, 4’’’,4’’’’’,4’’’’’’’-Th), 3 3 7.08 (d, J(H-3’,4’-Th, H-3’’’’’’’’’,4’’’’’’’’’-Th) = 3.6 Hz, 2 H, H-3’’’’’’’’’,4’-Th), 7.07 (d, J(H-3’,4’-Th, H-

3’’’’’’’’’,4’’’’’’’’’-Th) = 3.6 Hz, 2 H, H-3’,4’’’’’’’’’-Th), 2.76 (m, 20 H, α’-α’’’’’’’’’-CH2-Bu), 3 2,54 (t, J(α,β-CH2-Bu); α’’’’’’’’’’,β’’’’’’’’’’-CH2-Bu) = 7,8 Hz, 4 H, α,α’’’’’’’’’’-CH2-Bu), 1.60-1.44 (m,

48 H, β-β’’’’’’’’’’,γ-γ’’’’’’’’’’-CH2-Bu), 0.97 (m, 36 H, CH3-Bu)

5'''''''''' S 2''''''''S 5'''''' 2'''''' S S 2''' S 2 5 I I S S S S S 5' 2' S 3''''''''' 4''''''''' 4''''' 3''''' 4' 3' 30

4.6 Experimental section 165

13 C-NMR (101 MHz, CDCl3): δ= 147.5, 140.3, 140.2, 138.7, 136.4, 136.0, 135.5, 129.9, 129.7, 126.0 (C-2-’’’’’’’’’’, 3-3’’’’’’’’’’,4-4’’’’’’’’’’,5’’-5’’’’’’’’’-Th), 74.1 (C-5,5’’’’’’’’’-

Th), 32.9, 32.1, 31.0 (β-β’’’’’’’’’’’-CH2-Bu), 29.7 (α-α’’’’’’’’’’-CH2-Bu), 23.0, 22.9 (γ-

γ’’’’’’’’’’-CH2-Bu), 13.9 (CH3-Bu)

MS (MALDI-TOF): m/z (monoisotopic mass) calcd. for C92H118I2S11 1828.42; found 1828.55 [M+]

5,5’’’’’’’’’’-Bis(trimethylsilylethynyl)-3,3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,- 4’’’’’’,4’’’’’’’’,4’’’’’’’’’’-dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’: 5’’’’’,2’’’’’’:5’’’’’’,2’’’’’’’:5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’- undecithiophene (31) The synthesis was carried out according to GP3, diiodoundecithiophene 30 (103 mg, 56.2

µmol), Pd(PPh3)2Cl2 (1.96 mg, 2.80 µmol), PPh3 (1.03 mg, 3.91 µmol), CuI (0.74 mg, 3.91 µmol), 9 ml pyridine, 9 ml triethylamine, trimethylsilylacetylene (13.3 mg, 134 µmol), 4 h; chromatographic work-up (n-hexane/dichloromethane 10/1), yield 95 % (94.6 mg, 53.4

µmol,), red solid, HPLC (n-hexane/dichloromethane 90/10): tR (λmax) = 4.6 min (420 nm).

1 H-NMR (400MHz, CDCl3): δ= 7.11 (m, 6 H, H-3’’’,3’’’’’,3’’’’’’’, 4’’’,4’’’’’,4’’’’’’’-Th), 3 3 7.08 (d, J(H-3’,4’-Th, H-3’’’’’’’’’,4’’’’’’’’’-Th) = 3.6 Hz, 2 H, H-3’’’’’’’’’,4’-Th), 7.06 (d, J(H-3’,4’-Th, H-

3’’’’’’’’’,4’’’’’’’’’-Th) = 3.6 Hz, 2 H, H-3’,4’’’’’’’’’-Th), 2.75 (m, 20 H, α’-α’’’’’’’’’-CH2-Bu), 3 2,55 (t, J(α,β-CH2-Bu); α’’’’’’’’’’,β’’’’’’’’’’-CH2-Bu) = 7,8 Hz, 4 H, α,α’’’’’’’’’’-CH2-Bu), 1.60-1.44 (m,

48 H, β-β’’’’’’’’’’,γ-γ’’’’’’’’’’-CH2-Bu), 0.97 (m, 36 H, CH3-Bu), 0.25 (s, 18 H, CH3-

Si(CH3)3).

5'''''''''' S 2''''''''S 5'''''' 2'''''' S S 2''' S 2 5 S S S S S 5' 2' S (CH3)3Si Si(CH3)3 3''''''''' 4''''''''' 4''''' 3''''' 4' 3' 31

13 C-NMR (101 MHz, CDCl3): δ= 147.5, 140.3, 140.2, 138.7, 136.4, 136.0, 135.5, 129.9, 129.7, 126.0 (C-2-’’’’’’’’’’, 3-3’’’’’’’’’’,4-4’’’’’’’’’’,5’’-5’’’’’’’’’-Th), 117.2 (C-

166 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

5,5’’’’’’’’’’-Th), 100.0, 94.3 (C≡C), 32.9, 32.1, 31.0 (β-β’’’’’’’’’’’-CH2-Bu), 29.7 (α-

α’’’’’’’’’’-CH2-Bu), 23.0, 22.9 (γ-γ’’’’’’’’’’-CH2-Bu), 13.9 (CH3-Bu), -0.02 (Si(CH3)3.

MS (MALDI-TOF): m/z (monoisotopic mass) calcd. for C102H136S11Si2 1768.71; found 1768.69 [M+]

Eglinton coupling reaction TMS-protected diethynyl-undecithiophene 31 (30 mg, 17 µmol) was dissolved in THF (3 mL) and a solution of potassium hydroxide (9.5 mg, 170 µmol) in aqueous methanol (3 mL) was added. After stirring at ambient temperature for 3 h the reaction mixture was poured on ice and extracted with dichloromethane. The organic layer was washed with 1N

HCl, then saturated NaHCO3 solution and water, and dried over Na2SO4. Under cooling the mixture was concentrated to a small volume and then diluted by addition of pyridine/dichloromethane (4:1, 5 mL). This solution of deprotected diethynyl- undecithiophene 32 was then dropped slowly with a syringe to a stirred suspension of copper acetate monohydrate (84.8 mg, 425 µmol) and copper chloride (33.4 g, 340 µmol) in pyridine (5 mL) at room temperature. After complete addition the reaction mixture was allowed to stir for additionally 36 h. The solvent was removed under reduced pressure and the residue dissolved in dichloromethane. The organic phase was washed with 1N HCl, then saturated NaHCO3 solution and water, and dried over Na2SO4. In order to remote the polymeric material and copper salts, the crude product was filtered through a short column of silica gel with n-hexanes/dichloromethane mixture (1:1) as eluent. After evaporation of the solvent a mixture of cyclic products containing 33a and 33b (16.5 mg, 60% based on 31) was isolated as red microcrystalline solid. The separation of the macrocycles 33a and 33b was accomplished by chromatography on silica gel with n-hexane/dichloromethane (80/20) as eluent.

Cyclo[diyne-2,2’-(3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,4’’’’’’’’,4’’’’’’’’’’- dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’:5’’’’’’,2’’’’’’’:5’’’’’’’, 2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’-undecithien)-5,5’’’’-diyl)-1,1’-diyl] (33a)

4.6 Experimental section 167

S S S

S S

33a

The compound was isolated as red solid; yield 31% (8.7 mg 5.36 µmol); HPLC (n- hexane/dichloromethane 82/18): tR (λmax) = 6.2 min (397 nm).

MS (MALDI-TOF) (linear modus) m/z: calcd. average mass for C96H118S11 1622.61; found 1622.42 [M+].

Cyclo{bis[diyne-2,2’-(3’’,3’’’’,3’’’’’’,3’’’’’’’’,3’’’’’’’’’’,4,4’’,4’’’’,4’’’’’’,4’’’’’’’’, 4’’’’’’’’’’-dodecabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’: 5’’’’’’,2’’’’’’’:5’’’’’’’,2’’’’’’’’:5’’’’’’’’,2’’’’’’’’’:5’’’’’’’’’’,2’’’’’’’’’’’-undecithien)- 5,5’’’’-diyl)-1,1’-diyl] (33b)

S S S S S S S S

S S

S S S S S S S S

33b

168 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

The compound was isolated as red solid; yield 17.8 % (4.9 mg 1.51 µmol); HPLC (n- hexane/dichloromethane 82/18): tR (λmax) = 12.3 min (420 nm)

MS (MALDI-TOF) (linear modus) m/z: calcd. average mass for C192H236S22 3245.23; found 3245.31 [M+].

4.7 Supplement 169

4.7 Supplement

MALDI TOF MS mass spectra of C[3T-DA]n and C[5T-DA]n macrocycles

1554.8 100

60 / a.u. I 40

0 1000 1500 2000 2500 3000 m / z

Figure 4.21. MALDI-TOF MS spectrum of cyclotrimer C[3T-DA]3 27a.

g 100 90 80 70 60 50 40 30 20 10 0 1.552 1.554 1.556 1.558 1.560 1.562

100

0 1552 1554 1556 1558 1560 1562 1564 m / z

Figure 4.22. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 27a.

170 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

2073.0 100

60 / a.u.

I 40

0 1000 1500 2000 2500 3000 m / z

Figure 4.23. MALDI-TOF MS spectrum of cyclotetramer C[3T-DA]4 27b.

g 100 90 80 70 60 50 40 30 20 10 0 2.070 2.075 2.080

100

2070 2072 2074 2076 2078 2080 2082 2084 m / z

Figure 4.24. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 27b.

4.7 Supplement 171

2591.3 100

60 / a.u. I 40

0 1500 2000 2500 3000 3500 4000 4500 m / z

Figure 4.25. MALDI-TOF MS spectrum of cyclopentamer C[3T-DA]5 27c.

100 90 80 70 60 50 40 30 20 10 0 2.590 2.595 2.600

100

0 2588 2590 2592 2594 2596 2598 2600 2602 2604 m / z

Figure 4.26. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 27c.

172 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

3109.1 100

60 / a.u. I 40

0 1500 2000 2500 3000 3500 4000 4500 m / z

Figure 4.27. MALDI-TOF MS spectrum of cyclohexamer C[3T-DA]6 27d.

100 90 80 70 60 50 40 30 20 10 0 3.105 3.110 3.115 3.120

100

0 3104 3108 3112 3116 3120 3124 m / z

Figure 4.28. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 27d.

4.7 Supplement 173

3628.4 100

60 / a.u. I 40

0 2000 2500 3000 3500 4000 4500 5000 5500 m / z

Figure 4.29. MALDI-TOF MS spectrum of cycloheptamer C[3T-DA]7 27e.

100 90 80 70 60 50 40 30 20 10 0 3.625 3.630 3.635 3.640

100

0 3625 3630 3635 3640 3645 m / z

Figure 4.30. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 27e.

174 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

4146.6 100

60 / a.u. I

0 3000 3500 4000 4500 5000 5500 m / z

Figure 4.31. MALDI-TOF MS spectrum of cyclooctamer C[3T-DA]8 27f.

100 90 80 70 60 50 40 30 20 10 0 4.140 4.145 4.150 4.155 4.160 100

0 4140 4145 4150 4155 4160 4165 m / z

Figure 4.32. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 27f.

4.7 Supplement 175

100 4672.4

60 / a.u. I 40

0 2000 3000 4000 5000 6000 7000 8000 m / z

Figure 4.33. MALDI-TOF MS spectrum of cyclononamer C[3T-DA]9 27g.

5190.9 100

60 / a.u. I 40

0 3000 4000 5000 6000 7000 8000 m / z

Figure 4.34. MALDI-TOF MS spectrum of cyclodecamer C[3T-DA]10 27h.

176 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

1588.3 100

60 / a.u. I 40

0 800 1200 1600 2000 2400 m / z

Figure 4.35. MALDI-TOF MS spectrum of cyclodimer C[5T-DA]2 29a.

g 100 90 80 70 60 50 40 30 20 10 0 1.586 1.588 1.590 1.592 1.594 1.596

100

0 1586 1588 1590 1592 1594 1596 1598 m / z

Figure 4.36. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 29a.

4.7 Supplement 177

2383.4 100

60 / a.u. / I 40

0 1000 1500 2000 2500 3000 3500 4000 m / z

Figure 4.37. MALDI-TOF MS spectrum of cyclotrimer C[5T-DA]3 29b.

g 100 90 80 70 60 50 40 30 20 10 0 2.380 2.385 2.390 2.395

100

0 2380 2384 2388 2392 2396 m / z

Figure 4.38. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 29b.

178 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

3178.1 100

60 / a.u. I 40

0 2000 2500 3000 3500 4000 4500 m / z

Figure 4.39. MALDI-TOF MS spectrum of cyclotetramer C[5T-DA]4 29c.

100 90 80 70 60 50 40 30 20 10 0 3.175 3.180 3.185 3.190

100

0 3175 3180 3185 3190 m / z

Figure 4.40. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of 29c.

4.7 Supplement 179

3972.8 100

60 / a.u. I 40

0 3000 3500 4000 4500 5000 m / z

Figure 4.41. MALDI-TOF MS spectrum of cyclopentamer C[5T-DA]5 29d.

4770.9 100

60 / a.u. / I 40

0 3000 4000 5000 6000 7000 m / z

Figure 4.42. MALDI-TOF MS spectrum of cyclohexamer C[5T-DA]6 29e

180 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

4.8 References and notes

1 a) P. Bäuerle in Oligothiophenes in Electronic Materials: The Oligomer Approach, Eds.: K. Müllen, G. Wegner, Wiley-VCH, Weinheim, Germany, 1998, pp. 105-197; b) Handbook of Oligo- and Polythiophenes, Eds.: D. Fichou, Wiley-VCH, Weinheim, Germany, 1999.

2 a) H. Sirringhaus, R.H. Friend, X.C. Li, S.C. Morati, A.B. Holmes N. Feeder, Appl. Phys.Lett. 1997, 71, 3871-3873. b) F. Garnier in Field- Effect Transistors Based on Conjugated Materials in Electronic Materials: The Oligomer Approach (Eds.: K. Müllen, G. Wegner,) Wiley-VCH, Weinheim, Germany, 1998, pp. 559-584; c) G. Horowitz, Adv, Mater. 1998, 10, 365-370; d) Z. Bao, Adv. Mater. 2000, 12, 227-230; e) H.E. Katz, Z. Bao, S.L. Gillat, Acc. Chem. Res. 2001, 34, 359-369; f) T. Otsubo, Y. Aso, K. Takimiya, Bull. Chem. Soc. Jpn. 2001, 74, 1789-1801; g) C.D. Dimitrakopoulus, P.R.L. Malenfant, Adv. Mater. 2002, 14, 99- 117.

3 a) U. Mitschke, P. Bäuerle, J. Mater. Chem. 2000, 10, 1471-1507; b) Y. Shirota, J. Mater. Chem. 2000, 10, 1-25.

4 a) N. Noma, T. Tsuzuki, Y. Shirota, Adv. Mater. 1995, 7, 647-648.

5 a) Molecular Electronics: Science and Technology, Eds.: A. Aviram, M. Ratner, New York, Academy of Sciences, New York, 1998; b) R.M. Metzger, Acc. Chem. Res. 1999, 32, 950-957; c) C. Joachim, J.K. Gimsewski, A. Aviram, Nature 2000, 408, 541-548; d) J.M. Tour, Acc. Chem. Res. 2000, 33, 791-804; e) T. Otsubo, Y. Aso, K. Takimiya, J. Mater. Chem. 2002, 12, 2565-2575.

6 P. Bäuerle, Adv. Mater. 1992, 4, 102-107.

7 E. Mena-Osteritz, A. Meyer, B.M.W. Langeveld-Voss, R.A.J. Janssen, E.W. Meijer, P.Bäuerle, Angew. Chem. Int. Ed. Engl. 2000, 39, 2679-2683.

8 A.J.W. Tol, Synth. Met. 1995, 74, 95-98.

9 For recent reviews, see: a) D. Zhao, J.S. Moore, Chem. Commun. 2003, 807-818; b) C. Grave, A.D. Schlüter, Eur. J. Org. Chem. 2002, 3075-3098; c) S. Höger, J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 2685-2698; d) M.M. Haley, J.J. Pak, S.C. Brand, Top. Curr. Chem. 1999, 201, 81-130; e) A. de Meijere, S.I. Koszushkov, Top. Curr. Chem. 1999, 201, 1-42.

10 a) T. Kauffmann, B. Greving, R. Kriegesmann, A. Mitschker, A. Woltermann, Chem. Ber. 1978, 111, 1330-1336; b) T. Kauffmann, H.P. Mackowiak, Chem. Ber. 1985, 118, 2343-2352.

11 a) F. Sondheimer, R. Wolovsky, R. Amiel, J. Am. Chem. Soc. 1962, 84, 274-284.

12 Z. Hu, J.L. Atwood, M.P. Cava, J. Org. Chem. 1994, 59, 8071-8075.

13 T. Fisher, Dissertation 1998, University of Würzburg, Germany.

14 J. Kagan, S.K. Arora, Heterocycles 1983, 20, 1937-1940.

15 C. Musch, Diploma Thesis 1995, University of Würzburg, Germany.

16 G. Fuhrmann, Diploma Thesis 1999, University of Ulm, Germany.

17 I. R Carreras, Diploma thesis 1999, University of Ulm, Germany.

18 J. Krömer, Dissertation 2000, University of Ulm, Germany.

19 J. Krömer, I. Rios.Carreras, G. Fuhrmann, C. Musch, M. Wunderlin, T. Debaerdemaeker, E. Mena- Osteritz, P.Bäuerle, Angew. Chem. Int. Ed. Engl. 2000, 39, 3481-3486.

4.8 References and notes 181

20 a) J. Krömer, P. Bäuerle, Tetrahedron 2001, 57, 3785-3794.

21 a) H, Matsuhashi, Y. Hatanaka, M. Kuroboshi, T. Hiyama, Tetrahedron Lett. 1995, 36, 1539-1540; b) K. Tamao, K. Kobayashi, Y. Ito, Tetrahedron Lett. 1989, 30, 6051-6054; c) G. Tourillon, F. Garnier, J. Electroanal. Chem. 1984, 161, 51-58.

22 C.W. Spangler, M. He, J. Chem. Soc. Perkin. Trans. I 1995, 715-720; b) J.M. Tour, R.Wu, Macromolecules, 1992, 25, 1901-1907.

23 a) K. Tamao, K. Sumitani, M. Kumada, J. Am. Chem. Soc. 1972, 94, 4374-4376; b) M. Kumada, Pure Appl. Chem. 1980, 52, 669-679.K; c) Tamao, S. Komada, I Nakajima, M. Kumada, Tetrahedron, 1982, 38, 3347-3354. 24 P. Bäuerle, F. Würthner, G. Götz, F. Effenberger, Synthesis 1993, 1099-1103.

25 C. Wang, M.E. Benz, E. LeGoff, J.L. Schindler, C.R. Kannenwurf, M.G. Kanadzidis, Polymer Preprints 1993, 34, 422-423.

26 a) T. Otsubo, Y. Aso, K. Takimiya, Bull. Chem. Soc. Jpn. 2001, 74, 1789-1801; b) H. Nakanishi, N. Sumi, Y. Aso, T. Otsubo, J. Org. Chem. 1998, 63, 8632-8633.

27 B. Nessakh, G. Horowitz, F. Garnier, F. Deloffre, P. Sristava, A. Yassar, J. Electroanal. 1995, 399, 97- 103.

28 a) K. Sonogashira, T. Yatake, Y. Tohda, S. Takahashi, N. Hagihara, J. Chem. Soc., Chem. Commun. 1977, 291-292; b) K. Sonogashira in Comprehensive Organic Synthesis, Eds.: I. Fleming, B. Trost, Pergamon, Vol 3, New York, 1991.

29 a) P. Siemsen, R.C. Livingston, F. Diederich, Angew. Chem. Int. Ed. Engl. 2000, 39, 2632-2657; b) P.J. Stang, F. Diederich, Modern Acetylene Chemistry, VCH-Wiley, Weinheim, 1995; c) L. Brandsma, Preparative Acetylenic Chemistry, Elsevier, 1988.

30 a) R. Rossi, A. Carpita, C. Bigelli, Tetrahedron Lett. 1985, 26, 523-526; b) R.W. Wagner, T.E. Johnson, F. Li, J.S. Lindsey, J. Org. Chem. 1995, 60, 5266-5268; c) M. Vlassa, I. Ciocan-Tarta, F. Margineanu, I. Oprean, Tetrahedron 1996, 52, 1337-1342; d) Q. Liu, J. Burton, Tetrahedron Lett. 1997, 38, 4371-4374; X. Huang, J.-H Wang, Synth. Commun. 2000, 30(1), 9-14; e) A Godt, C. Franzen, S Veit, V. Enkelmann, M. Pannier, G. Jeschke, J. Org. Chem. 2000, 65, 7575-7582; f) A. Lei, M. Srivastava, X. Zhang, J. Org. Chem. 2002, 67, 1969-1971.

31 a) A.D. Dzhuraev, K.M. Karimkulov, A.G. Makhusov, A.G. Amanov, Pharm. Chem. J. (Engl. Transl.) 1992, 26, 882-884; b) S. Kozhushhov, T. Haumann, K. Boese, B. Knierim, S. Scheib, P. Bäuerle, A. de Meijere, Angew. Chem. Int. Ed. Engl. 1995, 34, 781-783; c) P. Kilickiran, Dissertation 2001, TU Braunschweig, Germany.

32 a) W. Schroth, F. Billig, G. Reinhold, Z. Chem. 1965, 5, 352-353; b) W. Schroth, F. Billig, G. Reinhold, Angew. Chem. Int. Ed. Engl. 1967, 6, 698-699; c) M. Koreeda, W. Yang, Synlett 1994, 201-202; d) E. Block, C. Guo, M. Thiruvazhi, P.J. Toscano, J. Am. Chem. Soc. 1994, 116, 9403-9404; e) D.E. Bierer, J.M. Dener, L.G. Dubenko, R.E. Gerber, J. Litvak, S. Peterli, P. Peterli-Roth, T.V. Truong, G. Mao, B.E. Bauer, J. Med. Chem. 1995, 38, 2628-2648; f) E. Block, M. Birringer, C. He, Angew. Chem. Int. Ed. Engl. 1999, 38, 1604- 1607.

33 M. Koreeda, W. Yang, J. Org. Chem. 1997, 62, 446-447.

34 E. Block, M. Birringer, C. He, Angew. Chem. 1999, 111, 1710-1713. 35 a) J. T. Mortensen, J.S. Sorensen, N.E. Sorensen, Acta. Chem. Scand. 1964, 18, 2392-2394; b) F. Bohlmann, K.M. Kleine, Chem. Ber. 1965, 98, 3081-3086; c) F. Freeman, D.S.H.L. Kim, E. Rodriguez,

182 Chapter 4 Macrocyclic oligothiophene-diacetylenes through oxidative coupling reactions

Sulfur Rep. 1989, 9, 207-256; d) F. Freeman, M. Aregullin, E. Rodriguez, In Reviews of Heteroatom Chemistry, Eds.: S. Oae, Vol. 9, MYU Tokyo 1993, pp. 1-19.

36 a) J.B. Hudson, E.A. Graham, R. Fong, A.J. Finlayson, G.H.N. Towers, Planta Med. 1986, 52, 51-54; b) J.B. Hudson, E.A. Graham, R. Fong, A.J. Finlayson, G.H.N. Towers, Planta Med. 1986, 52, 453-457; c) C.P. Constabel, G.H.N. Towers, Planta. Med. 1989, 55, 35-37; d) J.B. Hudson, F. Balza, L. Harris, G.H.N. Towers, Photochem. Photobiol. 1993, 57, 675-680; e) G.H.N. Towers, R.C. Bruening, F. Balza, Z.A. Lopez- Bazzochi, I. US Patent 5.202.248, April 13, 1993.

37 E. Block, J. Page, J.P. Toscano, C.-X- Wang, X. Zhang, R. DeOrazio, C. Guo, R.S. Sheridan, G.H.N. Towers, J. Am. Chem. Soc. 1996, 118, 4719-4720.

38 C.A. Briehn, T. Kirschbaum, P. Bäuerle, J. Org. Chem. 2000, 65, 352-359.

39 D.O’Krongly, S.R. Denmeade, M.Y. Chiang, R. Breslow, J. Am. Chem. Soc. 1985, 107, 5544-5545.

40 P. Ruggli, Liebigs Ann. Chem 1912, 392, 92-94.

41 a) K. Ziegler, Houben Weyl, Methoden der Organischen Chemie, Georg Thieme Verlag, Stuttgart, 1955, p. 729, 4th ed., vol. 4/2; b) P. Knorps, N. Sendhoff, H.-B Meckelburger, F. Vögtle, Top. Curr. Chem. 1991, 161, 1.

42 G. Eglinton, W. McCrae, Adv. Org. Chem. 1963, 4, 225-328.

43 a) A. de Meijere, S. Kozhushkov, T. Haumann, R. Boese, C. Puls, M.J. Cooney, L.T. Scott, Chem. Eur. J. 1995, 1, 124-131; b) J.J. Pak, T.J.R. Weakly, M.M. Haley, J. Am. Chem. Soc. 1999, 121, 8182-8192.

44 a) Y. Rubin, M. Kahr, C.B. Knobler, F. Diederich, C.L. Wilkins, J. Am. Chem. Soc. 1991, 113, 495-499; b) Q. Zhou, P.J. Caroll, T.M. Swager, J. Org. Chem. 1994, 59, 1294-1301. 45 E. Mena-Osteritz, Adv. Mater. 2002, 14, 609-616. 46 S. Gronowitz, Acta Chem. Scand. 1959, 13, 1045-1046.

47 C.W. Spangler, M. He, J. Chem. Soc. Perkin. Trans I 1995, 715-720.

Chapter 5

Conjugated macrocycles by metal template approach

Abstract

In this chapter a novel method towards effective synthesis of (conjugated) macrocycles is presented. The strategy includes the high-yielding and facile formation of transition-metal σ-acetylide macrocycles under thermodynamic equilibrium conditions, and subsequent C- C bond formation by a reductive elimination at the metal center affording all-carbon perimeter macrocycles. The two topics that are of importance for the development of this method, the synthesis of transition metal containing cyclic structures, as well as the general principles and investigations on the elimination process from a transition metal center, are discussed in some detail. Preliminary studies for method development which have been carried out by using ethynylthiophenes as simple model compounds include the effective synthesis and characterization of novel acyclic bis(oligothienyl-ethynyl)platinum(II) complexes and the subsequent oxidatively induced eliminations which under C-C bond formation lead to the corresponding 1,3-butadiynes. The application of the newly developed metal template approach in the design and synthesis of cyclic structures is demonstrated. A number of platinum-bridged diethynylated 184 Chapter 5 Conjugated macrocycles by metal template approach

oligothiophene macrocyles which represent a novel topology and class of compounds in the area of nananosized macrocycles have been prepared in excellent yields. The subsequent elimination reactions of the transition metal units by means of an oxidant gave access to the targeted macrocyclic oligothiophene-diacetylenes. Furthermore, some investigations on the process of reductive elimination have been carried out and detailed aspects of these reactions and their mechanisms are addressed. This chapter includes furthermore, the synthesis of homologues series of α-linked macrocyclic oligothiophenes, which are designed as cyclo[n]thiophenes. These compounds represent a novel class of well-defined macrocycles with fully conjugated π-system periphery. The strategy is based on the reaction of macrocyclic oligothiophene- diacetylenes C[mT-DA]n with sulfide nucleophiles. An optimized protocol where much milder reaction conditions are employed in the thiophene-ring-closure reactions of butadiynes is applied. This method led to an improved yield in the preparation of the smallest homologue of the series, cyclo[8]thiophene, providing evidence for the efficiency of the synthetic methodology. Finally, the optical properties of the novel platinum-linked ethynyl-oligothiophenes have been investigated and compared with those of their metal-free precursors. The electronic properties of the two 32π-electron cyclic systems, cyclodimeric terthiophene-diacetylene 45 and cyclo[n]thiophene 47, were examined. For terthiophene-diacetylene macrocycle 45 a detailed structure characterization by X-ray analysis is presented.

5.1 Introduction 185

5.1 Introduction

Despite the remarkable potential utilities of macrocycles, their practical use is limited especially due to synthetic difficulties encountered in their preparation.1 In Chapter 4, the successful synthesis of two series of fully conjugated macrocycles, the terthiophene-diacetylenes C[3T-DA]n and quinquethiophene-diacetylenes C[5T-DA]n were described. As illustrated in Scheme 5.1, the strategy towards these cyclic architectures was based on the oxidative coupling of the corresponding diethynylated oligothiophene precursors under pseudo high-dilution conditions. Since this reaction involved oligomerization and cyclization at the same time by random irreversible processes, various cyclic products were generated as a mixture. As expected, the modest yields critically depended on the structural features of the monomer building blocks and it waas tedious to separate the obtained mixtures. Higher yield in the cyclization step was achieved when employing a diethynylated undecithiophene as precursor. The cyclization reaction yielded by an intramolecular ring-closure reaction the corresponding cyclic. However, the remarkable disadvantage of this approach is the tedious synthesis of the precursor with appropriate chain length. An exciting aspect and a challenging goal of this study was the development of a novel and more effective method towards the synthesis of such fully conjugated thiophene-derived macrocycles.

5.2 Basic concept of the metal template approach

In general, kinetically controlled macrocyclization reactions are low-yielding processes which lead to a mixture of cyclic products. As illustrated in Chapter 3, alternative and more promising approaches towards these fascinating structural motives include the use of templates, as well as cyclizations under thermodynamic control. Template-directed synthesis can be a powerful tool to prepare macrocyles in high yields, but mostly a lot of effort must be put to develope suitable templates with the required structural and functional properties. Moreover, the method is limited to certain cases where adequate functionalities in the core allow non-covalent or covalent interactions to the template.

186 Chapter 5 Conjugated macrocycles by metal template approach

Bu Bu a) S S Bu Bu S Bu Bu Bu Bu Bu Bu S S S S Cu(OAc) x H O / CuCl R R 2 2 S pyridine / dichloromethane 23 R = Si(CH3)3 KOH pseudo high dilution S MeOH / THF Bu Bu 26 R = H S Bu S S Bu n-2

Bu Bu

C[3T-DA]n 27a-g (n =3-9) total yield [12.7 %] b)

Bu Bu Bu Bu Bu Bu

S S S S S R R

24 R = Si(CH ) KOH 3 3 MeOH / THF 28 R = H

Cu(OAc)2 x H2O / CuCl pyridine / dichloromethane pseudo high dilution

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S n-1 Bu Bu Bu Bu

29a-e (n = 2-6) total yield [16 %] C[5T-DA] n

Scheme 5.1. Synthesis of fully conjugated macrocycles, the terthiophene-diacetylenes C[3T-DA]n and quinquethiophene-diacetylenes C[5T-DA]n by random oxidative coupling reaction. 5.3 Transition metal directed self-assembly 187

The advantages of performing macrocyclisation reactions under thermodynamic control have been clearly pointed out in the field of supramolecular chemistry.2 In the directional- bonding approach elaborated by Fujita and Stang,3 cis-complexes of transition metals are used to self-assemble with organic ligands to form well-ordered supramolecular cyclic structures under thermodynamic equilibrium. Such precursors having two accessible sites at a 90°C angle provide the formation of molecular squares when reacted with rigid organic ligands and of di-or trinuclear macrocycles when reacted with flexible organic ligands. A brief overview over the research in the field of transition metal-directed self- assembly under cyclic structure formation is comprised in Chapter 5.3. However, supramolecular chemistry is mostly focused on macrocyles containing heteroatoms which enable directional coordinative bonds to the metal centers. More recently, the synthesis and properties of macrocycles having acetylide units linked by a σ-bond to the transition metal have been reported.4 The preparation and properties of some transition metal σ-acetylide complexes will be discussed in Chapter 5.4. The most extensively used metal systems are cis-oriented square-planar tertiary-phosphine Pt(II) complexes, which under thermodynamic control effectively react with alkynes enabling the formation of a stable main cyclic structural feature. A great deal of interest is currently devoted to transition metal-mediated coupling reactions (see Chapter 2). An essential step in most of these reactions in which new carbon-carbon bonds are generated is the 1,1-reductive elimination from the metal center. Numerous investigations have been carried on this process, and particularly on those proceeding from transition metals of group 8. The reductive elimination from a square-planar transition metal complex has been shown to take place if the organic moieties are cis-oriented. The rate of the reaction is found to depend significantly on the metal used and on its electronic and coordinative environment. In Chapter 5.5 a detailed discussion of the reductive elimination from transition metal centers is given. In the light of the aforementioned considerations, a particularly promising approach towards the syntheses of macrocyclic oligothiophene-diacetylenes might involve a strategy which bases on a combination of two reactions in which transition metals play a key determining role. As schematically illustrated in Figure 5.1, in the first reaction, metallacycles containing cis-oriented transition metal-bridged diethynylated oligothiophenes should be formed under thermodynamic control. Subsequently, the generation of the targeted macrocyclic oligothiophene-diacetylenes should be achieved by reductive elimination of the organic moieties from the metal centers. 188 Chapter 5 Conjugated macrocycles by metal template approach

nT nT

nT M M + M

nT nT

M = transition metal of d8 group nT = oligothiophene = ethynyl unit

Figure 5.1. Strategy of the metal template approach towards fully conjugated thiophene-derived macrocycles by a metallacycle formation, and reductive elimination from the metal center.

Although in the true sense of the word, not the metals but the metallacycles serve as templates for our targeted fully conjugated thiophene-derived macrocycles, since in both reactions transition metals play the key determining role, this novel method will be designat as the “Metal Template Approach”.

5.3 Transition metal directed self-assembly

The field of supramolecular chemistry has witnessed tremendous growth over the past decade. This expansion has been driven by the growing interest in fascinating supramolecular complex systems exhibiting promising properties and functions with respect to future potential applications in the scientific fields of chemistry, biology, material science and nanotechnology.5 Most approaches giving access to discrete supramolecular and nanomolecular architectures are based on template and self-assembly strategies of relatively simple precursor molecules.6 Among them, one of the most effective and direct methods is the transition metal-directed self-assembly which was elaborated by Fujita and Stang.3 In this approach, well-ordered supramolecular cyclic structures are readily built up under thermodynamic equilibrium by the use of transition metals and their ability to coordinate to heteroatoms of organic ligands. Transition metals are typically introduced with directing or blocking ligands. The resulting metal coordination geometry, together with the orientation of the 5.3 Transition metal directed self-assembly 189

interaction sites in a given rigid organic ligand, define the main structural feature of the supramolecular compound that will be formed. In order to enable the formation of directional coordinative bonds to the metal mostly nitrogen-containing heteroarenes having donating donor character are used as organic ligands. The resulting metal-ligand bond energies of 10-30 kcal/mol are in between the range of stronger covalent bonds in classical macrocycles and weaker interactions which nature uses for the construction of various macromolecules (hydrogen bonding, π-π stacking, hydrophobic hydrophilic and electrostatic forces). The kinetic lability of the coordinative bonds allows self-healing processes of defect structures, which as a result of the equilibrium finally leads to the thermodynamically most stable structure under the given reaction conditions. Due to the higher number of energetically favourable interactions per building block unit with respect to enthalpy, cyclic structure formation is favoured over the formation of corresponding linear oligomers. Because of entropy, the unstrained macrocyclic structure with the minimum number of subunits will be preferentially formed. Taking advantage of this strategy a variety of supramolecular complex architectures become accessible in the meanwhile.3 As illustrated in Figure 5.2, the simple systematical combination of the proper transition metal complexes with organic building blocks coming from various families of molecules enabled the formation of supramolecular systems with different polygonal geometries, such as triangles, squares, rectangles, pentagons and hexagons.

Figure 5.2. Synthesis of cyclic molecular polygons by the systematic combination of metals as the shape determining units with difunctional conformationally rigid organic building blocks (directional-bonding approach).3b 190 Chapter 5 Conjugated macrocycles by metal template approach

Among all the accessible geometric shapes through the directional-bonding approach, molecular squares have been the most widely reported.7 The requirement for the formation of such a cyclic structure with 90° edges in the assembly is very simple. As can be seen in Figure 5.2, if the organic ligand is rigid, linear or pseudo-linear and additionally difunctional, then the transition-metal complexes must have at least two accessible coordination sites at 90°. By blocking two adjacent coordination sites of the metal with chelating ligands, whose dissociation from the metal is negligible under ordinary conditions, two coordination sites in a constrained cis-geometry remain free for the interaction with the organic building block. Therefore, transition-metal complexes, and in particular complexes of palladium(II) and platinum(II), with defined square planar geometry are predetermined for the formation of squares. Fujita et al. were the first who reported the quantitative formation of a molecular square by using this approach.7a As illustrated in Scheme 5.2, the reaction of the ethylendiamine complex of Pd(II) dinitrate with 4,4’-bipyridine at room temperature yielded molecular square C48 as the single and stable product. Corresponding Pt(II) analogue C49 was formed only after heating the reaction mixture to 100°C. In contrast, at room temperatures a mixture of oligomers was obtained by a kinetical process, which at higher temperature could be converted to the thermodynamically stable product, molecular square C49. This effect can be explained by the stronger bond of Pt(II)-N in comparison to that of Pd(II)-N bond.

NH2 H2N - 8 NO3 N M N H2N M N H2 N N

H2 N ONO2 water M + N N N ONO2 for M = Pd rt H2 for M = Pt 100°C N N N M N N M NH H2 2

NH2 H2 N

C48 M = Pd C49 M = Pt

Scheme 5.2. Molecular squares by the directional-bonding approach.7a

However, there are some inherent limitations of this approach. The initial requirement for the prediction of the geometry of the final self-assembled entity is the conformational 5.3 Transition metal directed self-assembly 191

rigidity of the subunits. Subtle changes in the protecting ligands at the transition-metal center or the use of conformationally more flexible organic ligands can lead to the formation of other molecular geometries than expected. Making use of the ligand flexibilities of the directional-bonding approach, various dinuclear macrocycles incorporating many different ligands with desirable physical properties have been synthesized and characterized.8 For example, as shown in Scheme 5.3, reaction of the same cis-protected enPd(NO3)2 complex with nonlinear or flexible organic ligands resulted in the formation of several dimetallic macrocycles of type C50 instead of nuclear squares.9

R 4+ H H2 H 2 N N 2 4 NO - N water N N 3 ONO2 M + N R N M M rt N N ONO2 N N N H H2 2 H2 R

C50 R = CH2 R = C=O R = CH2CH2 R = 1,4-CH (C F )CH R = C(OH)2 2 6 4 2 R = 1,4-CH (C H )CH R = C=CH2 2 6 4 2

Scheme 5.3. Synthesis of a series of dimetallic macrocycles incorporating flexible organic ligands by Fujita et al.

In a summary, one can conclude that complexation of cis-protected metals with rigid linear organic ligands result in macrocyclic tetranuclear complexes. In contrast, flexible organic ligands self-assemble with cis-protected metal complexes to entropically favoured di-or trinuclear macrocycles, even if the cycles are strained to some extent.

5.4 Transition metal σ-acetylide complexes

The challenge to develop new molecular materials with tailorable properties has paved the way to design cyclic molecules, but with the metal atom incorporated in the σ framework of the ring.4 Transition metal alkynyl σ-complexes are mostly prepared using a reaction developed by the group of Hagihara that involves Cu(I)-catalyzed dehydrohalogenation of metal halides in an amine solvent (Scheme 5.4).10 192 Chapter 5 Conjugated macrocycles by metal template approach

PR PR 3 CuI 3 X MX + 2 H R' R' M R' + 2 R''3NHX R'' N 3 PR PR3 3

M = Pd, Pt

Scheme 5.4. Synthesis of transition metal alkynes by Cu(I) dehydrohalogenation.

Reaction of terminal acetylenes with bis(tertiary phosphine)-metal(II) dichlorides, in the presence of Cu(I) iodide, results in the formation of bis(alkynyl)-metal(II) complexes in high yields. The products are square-planar and have a cis- or trans- conformation depending on the nature and geometry of the starting material and on the reaction conditions. The trans-product is thermodynamically favoured. Thus, at higher temperatures with non-chelating phosphine ligands isomerization from the cis-bis(phosphine)- platinum(II) dichloride species to the trans-complexes can take place prior to the alkynylation reaction. Further, cuprous halides are known to accelerate this isomerization process.11 Cis-acetylide isomers are accessible at lower temperatures or by using chelating ligands such as bis(diphenylphosphino)propane, enforcing a cis-confirmation around the metal center.

The first soluble transition metal σ-polyynes bearing [Pd(PBu3)2] and [Pt(PBu3)2] units in the main chain were already reported in 1975 by Hagihara and co-workers.12 In the meanwhile, transition-metal polyynes represent one of the most studied types of organometallic polymers, mainly as a result of the synthetic versatility of the reaction, but also because of their unique properties. A detailed review including the synthetic methods, the properties of metal-alkynyls and their applications has recently been published by Long et al.13 Introduction of a metal center into the conjugated chain gives rise to a range of properties, such as redox, magnetic, optical and electronic properties that differ from those of the organic counterparts. The stable polymers exhibit a degree of π-conjugation along the rigid-rod backbone and contain highly polarizable transition-metal units. A variety of platinum polyynes were prepared and their properties including nonlinear optical effects,14 luminescence and photoconductivity,15 electronic communication16 and liquid crystallinity17 were investigated by different research groups. In Scheme 5.5 some platinum polyynes with different aromatic spacers that were all prepared by the above mentioned Hagihara dehydrohalogenation method are illustrated.18 5.4 Transition metal σ-acetylide complexes 193

LBu3 Cl Pt Cl + H R' R R' H

LBu3

CuI amine

LBu3 LBu3

Pt R' R R' Pt n

LBu3 LBu3

L = P or As ; R = R' =

L = P ; R = ; R' = or S

L = P or As; R = ; R' =

Scheme 5.5. Synthesis of some platinum polyynes with different aromatic spacer groups.

The polyynes illustrated in Scheme 5.5 exhibit strong metal-to-ligand charge-transfer absorptions and have a much smaller band gap than the analogues dialkynyl monomers. With band gaps of approximately 3 eV, they can be classified as wide-band-gap semiconductors or insulaters in their undoped states. Recently however, a platinum polyyne P1 having a band gap of 1.77 eV has been reported (Scheme 5.6).19 Here, the concept of alternating electron-donor (tri-butyl-phosphane substituted Pt(II)-alkynyl) and acceptor (thieno[3,4-b]pyrazine) units were applied to obtain a soluble, deep-blue coloured polymer, which was formed in 43 % yield by the Hagihara reaction. Similarly, the donor acceptor system P2 exhibits the lowest band gap of 1.58 eV reported for a metal polyyne species.20

194 Chapter 5 Conjugated macrocycles by metal template approach

Ph Ph NC CN N N PBu 3 Bu P Pt 3 n S Bu3P Pt n

PBu3 P1 P2

Scheme 5.6. Examples of low band-gap Pt-polyynes.

The main limitation of the Hagihara dehydrohalogenation reaction turned out to be the use of amines as solvents. Many complexes of transition metals in group 10 are either insoluble in amines or, if soluble, decompose readily in such solvents. One method that avoids the use of amines is the metathesis reaction between trimethyltin reagents and metal halides that has been developed by Lewis and co-workers.21 The high potential of the method was demonstrated in the prepation of various monomeric, dimeric and polymeric alkynyl – transition metal complexes. Here, trimethylstannyl alkynyl reagents were employed in different ratios and reacted in the presence of Cu(I) iodide with metal chlorides in toluene or chlorinated solvents (Scheme 5.7).

H Y H

ML2Cl2 nBuLi Y ML Cl (2 equiv.) ClLnM n Me3SnCl

Me3Sn Y SnMe3

ML Cl ML2Cl2 2 2 Y MLn (0.5 equiv.) (1 equiv.) n

Y SnMe Me3Sn Y MLn 3

M = Ni, Pd, Pt L = PBu3 or AsBu3

Y = or

Scheme 5.7. Illustration of versatility of the Cu(I)-catalyzed reaction between trimethylstannyl alkynyl ligands and transition metal chlorides.

5.4 Transition metal σ-acetylide complexes 195

Transition metal complexes of acetylide-functionalized oligothiophenes are basically accessible through both methods. Younus et al., for example, reported the effective formation of dinuclear platinum complexes of a series of rigid-rod alkynes with extended π conjugation through oligothiophene linkage in the backbone using the Hagihara method (Scheme 5.8).22 The electronic spectra of the diplatinum complexes have been investigated and compared with those of the organic counterparts. It was shown that the absorption spectra are dominated by the π-π* transitions of the corresponding bridging oligothiophene units. A notable feature is that attachement of the platinum fragments to the organic precursors red-shifted the absorption maxima by a value of around 50 nm. This is attributed to the acceptor character of the platinum moieties.

PEt3 CuI PEt PEt 2 Ph Pt Cl + 3 S n 3 S n Ph Pt Pt PEt3 H H i-Pr2NH Ph PEt3 PEt3 n = 1-3

Scheme 5.8. Synthesis of dinuclear platinum complexes of oligothiophene acetylides.

A very effective one-pot procedure based on the Lewis method was developed by Altamura et al. for the synthesis of palladium-ethynylthiophene polymers.23 The straightforward insertion of ethynylthiophene and oligo(ethynylthiophene) spacers into the polymer backbone is illustrated in Scheme 5.9. In the first step, the diiodoaromatic compound is reacted with tributyl(ethynyl)tin in the presence of a Pd(0) catalyst. In this reaction the bisethynylaromatic compound is formed along with Bu3SnI as side product. Addition of LDA to this mixture affords deprotection of the terminal acetylenic functionalities and their recombination with the tin side product to form the corresponding tin acetylides. Following this transformation, a metal dihalogenid is added to the reaction affording the corresponding metalla-ethynylthiophene polymers. The advantage of this procedure over the standard Lewis method is that the recovery and recycling of the toxic tin reagents is fully integrated in the process. At the end of the overall process, the polymeric material can be isolated by simple precipitation. The Bu3SnI side product can be recovered from the filtrate, purified and reused.

196 Chapter 5 Conjugated macrocycles by metal template approach

I S S S n I

2 Bu3Sn H ''Pd(0)''

S 2 Bu Sn I S S n + 3 H H

LDA

S S S n Bu3Sn SnBu3

Pd(PBu3)2Cl2

PEt S 3 S S n Pt PEt 3 p n = 0,1

Scheme 5.9. Synthesis of metalla-ethynyloligothiophene polymers using a one-pot procedure developed by Altamura et al.

More recently, a lot of interest has been devoted to cyclic arrays comprising transition metal-alkyne units. By using the well-developed synthetic methods for metal-alkyne bond formation and applying strategies based on the knowledge of the metal coordination chemistry such as the directional-bonding approach, several transition-metal σ-acetylide macrocycles were prepared. Following, some examples are discussed in more detail. Haley et al. reported for example the synthesis and crystallographic characterization of a novel strained trans-platinadehydrobenzo[19]annulene C51. As illustrated in Scheme 5.10 the synthesis of C51 included first the formation of the organo tin derivative, which was subsequently reacted with the trans-platinum dichloride precursor.4e Due to the insertion of the transition metal within the annulenic cycle an increased nonlinear optical activity was expected. Again, as observed for the previously described linear metal-alkynyl compounds, the absorption maximum of platinacycle C51 is clearly red-shifted (λmax = 393 nm) compared to the absorption of the corresponding [19]dehydrobenzoannulene (λmax = 369 nm). This suggests an electronic delocalization throughout the entire platinacycle. 5.4 Transition metal σ-acetylide complexes 197

a) Bu4NF

b) Me3SnNMe2 Si(i-Pr)3 c) trans-PtCl2(PEt3)2, CuI

Pt(PEt3)2 Si(i-Pr)3

C51

Scheme 5.10. Synthesis of trans-platinadehydrobenzo[19]annulene C51.

A variety of neutral, dianionic, single pocket or double pocket platinum containing heterocyclynes have been reported by the groups of Tessier and Youngs.4b-d In Scheme 5.11, the synthesis of the dinuclear platinacyclyne C52 containing two cyclic enyne pockets and that of tetraethynylplatinum complex C53 having two tweezers and two pockets are shown as examples.

cis-Pt(PEt3)2Cl2

CuI / (i-Pr)2NH Pt(PEt ) (PEt3)2Pt 3 2

C52

b) 2- Cl Cl 2- a) n-BuLi Hg

b) PtCl2(tht)2 c) NBu Br HgCl2 + 4 + 2 NBu 2 NBu4 Pt 4 Pt

Cl Cl C53

Scheme 5.11. Synthesis of a) dinuclear platinacyclyne C52 and b) tetraethynylplatinum complex C53.

198 Chapter 5 Conjugated macrocycles by metal template approach

These planar metallacycles provide selective complexation behaviour towards various metallic species according to their size and coordination geometry. As an example, platinacyclyne C53 as an acetylide tweezer is able to complex a HgCl2 molecule. A family of novel chiral platinacyclophanes with potential application in enantioselective catalytic processes has been recently reported by Lin et al.24 The synthesis of these supramolecular architectures is based on the combination of rigid bridging ligands derived from enantiopure 2,2’-1,1’-binaphtol with appropriate transition metal units. As illustrated in Scheme 5.12, the Hagihara reaction of cis-Pt(PEt3)2Cl2 with enantiopure 6,6’- bis(alkynyl)-1,1’-binaphtalene gave the chiral dinuclear metallacyclophane C54 in high yield. In contrast, by using the analogue trans-Pt(II) complex a mixture of chiral polygons ranging from a triangle to an octagon was formed. The molecular polygons all bearing chiral functionalities could be separated by chromatography.

PEt Et3P 3 Pt

CuCl AcO AcO OAc + cis-Pt(PEt ) Cl AcO 3 2 2 AcO OAc HNEt2

Pt Et P 3 PEt3

C54 [59 %]

OAc OAc

PEt CuCl Pt 3 AcO Et3P Pt trans-Pt(PEt ) Cl PEt Et3P + 3 2 2 3 AcO HNEt2

PEt 3 OAc AcO Pt OAc AcO n PEt3

C55a-f (n=1-6)

[4-18%]

Scheme 5.12. Synthesis of chiral platinacyclophanes for application in asymmetric catalysis. 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 199

5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center

Carbon-carbon bond formation by reductive elimination from transition metal-organyl complexes is an important reaction in organometallic synthesis and catalysis but whose mechanistic determinants are yet poorly understood. 25 Responsible for both stochiometric and catalytic coupling reactions via transition metals, particularly those of group 10, is the combination of an oxidative addition and a 1,1- reductive elimination reaction. The mechanism of the Pd-promoted coupling reaction is schematically depicted in Figure 5.3. During the reductive elimination, which is the final product-releasing step, the formal oxidation state and the coordination number of the metal is reduced by two; metal-organyl bond breaking is accompanied by C-C bond making.

Pd(II)

R'M

R'-R' + MX

R-R' Pd(0) 1 RX

oxidative addition reductive elimination 3

R-Pd(II)-R' R-Pd(II)-X

transmetallation

MX R'M

Figure 5.3. Catalytic cycle of the palladium-promoted cross-coupling reactions.

Early mechanistic studies and construction of orbital correlation diagrams for four coordinated square-planar d8 complexes have shown that a concerted 1,1-reductive elimination at the metal center is symmetry allowed and can take place if, as shown in Scheme 5.13, the organic moieties occupy adjacent positions (cis-coordination) in the complex. From a trans-isomer, isomerization to a corresponding cis-complex must proceed 200 Chapter 5 Conjugated macrocycles by metal template approach

before reductive elimination. It was further demonstrated in a number of experiments that reductive elimination follows first order kinetics and is an intramolecular reaction.25

L R M RR+ 'ML ' R L 2 trans M = d8 transition metal L= ligands R = organic moieties L R M RR+ 'ML ' L R 2 cis

Scheme 5.13. Reductive elimination from square-planar d8 metal complexes.

Experimental evidence for these observations has been given, for example, in the studies carried out by Stille et al.26 In their investigations on cis-trans isomerization and mechanism of the 1,1-reductive elimination reaction, they examined the thermal behaviour of six different bis(phosphine)dimethyl-Pd(II) complexes in different solvents. As shown in Scheme 5.14, the palladium complexes possess cis- or trans conformations and chelating or non-chelating phosphine ligands.

Ph Ph P CH CH CH 3 Ph3P 3 MePh2P 3 Pd Pd Pd P Ph P MePh P CH3 3 CH 2 CH Ph 3 3 Ph

C56 C57 C58

Ph P MePh P CH3 3 CH3 2 Pd Pd H C H C CH 3 PPh 3 PPh Me Ph 3 3 2 Ph P Pd P C59 Ph C60 Ph H3C C61

Scheme 5.14. Bis(phosphine)-dimethylpalladium complexes C56-C58 in cis, C59-C61 in trans configuration. 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 201

In non-polar solvents such as in benzene or tetrachloroethane C57 and C60 were even at high temperature stable. Isomerization or reductive elimination from these complexes could not be observed. In more polar solvents such as THF or DMF, at 50°C, ethane was generated from both. The 1,1-reductive elimination from trans-isomer C60 was shown to occur after a trans to cis isomerization. By quenching a heated THF solution of C60 with pentane, only the corresponding cis-isomer C57 was formed. During this time ethane has not been produced.

benzene / 50°C MePh P CH CH 2 3 MePh2P 3 Pd Pd MePh P H C 2 CH 3 3 PPh2Me

C57 C60

CH THF/ 50°C MePh P CH MePh2P 3 2 3 Pd Pd H3C CH3 H C MePh P 3 2 CH PPh2Me 3

C60 C57

CH MePh2P 3 THF/ 50°C Pd H3C CH3 MePh2P CH3 C57

Scheme 5.15. Thermal behaviour of dimethyl-Pd(II) complexes C57 and C60.

In DMSO at room temperature, trans-complexes C59 and C60 were rapidly converted to the corresponding cis-isomers Above 45°C ethane was rapidly evolved from both complexes (Scheme 5.16).

Ph P CH CH 3 3 DMSO Ph3P 3 DMSO Pd Pd H3C CH3 H3C PPh3 rt Ph3P T>45°C CH3

C59 C56

Scheme 5.16. Isomerization of trans-complex C59 and subsequent reductive elimination from the corresponding cis-isomer C56 in DMSO.

202 Chapter 5 Conjugated macrocycles by metal template approach

The chelating ligand containing cis-complex C58 also underwent reductive elimination in DMSO at 80°C affording ethane (Scheme 5.17). In contrast, the thermolysis of trans- complex C61 in DMSO did not generate any ethane even at 100°C and after 10 h. Due to its geometry the complex can not isomerize to a cis-isomer and thus, elimination did not proceed.

Ph Ph DMSO / 100°C P CH3 DMSO / 80°C Pd H3C CH3 P CH Ph CH3 Ph 3 Ph Ph P Pd P Ph H C Ph C58 3 C61

Scheme 5.17. Thermolysis of chelate complexes C58 and C61.

The intramolecular character of the reductive elimination reaction from cis-configuration was evidenced by crossover experiments. For this reason, mixtures of equimolar amounts of cis-complexes C56, C57 and C58 and their respective perdeutoriomethyl analogues were thermolyzed in DMSO at 60°C for 6 h. As shown in Scheme 5.18 for C56, the elimination process is intramolecular affording only ethane and ethane-d6, and not mixed compounds.

Ph P CH Ph P CD 3 3 3 3 60°C / 6 h Pd Pd + H3C CH3 + D3CCD3 Ph3P Ph3P DMSO CH3 CD3 C56

Scheme 5.18. Intramolecular reductive elimination from C56 evidenced by crossover experiment.

Reductive eliminations with concomitant C-C bond formations are rare and proceed only when a three-center transition state has an intrinsic low activation barrier.27 In all other cases either a loss of an ancillary ligand or ligand association from the square-planar complex takes place prior to the reductive elimination step. The C-C bond is then either formed from a three-coordinated28 or from a five-coordinated intermediate.29 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 203

The addition of free phosphine ligand PPh2Me to a DMSO solution of C57 slowed the rate of reductive elimination. Additionally, in the 1H and 31P NMR spectra dissociations of coordinated phosphines from complexes C56 and C57 were detected in polar solvents. This was not the case in non-polar coordinating solvents in which reductive elimination do not proceed. The cis-chelating ligand dppe (Ph2PCH2CH2PPh2) did not dissociate in a detectable amount from the complex C58, even in strong polar solvents. At the same time, the rate of reductive elimination from C58 was found to be 50-100 times slower than from C57, a complex that is electronically and geometrically similar. Based on these results, the authors suggested that the reductive elimination of ethane from dimethyl-Pd(II) complexes occurs via a dissociative mechanism. As illustrated in Scheme 5.19, the function of the solvent was presumed either to aid in phosphine dissociation by solvation or by occupying the coordination site vacated by phosphine. However, once elimination took place, the dissociated ligands recoordinated to Pd(0). This was evidenced by NMR by the decreasing amount of free phosphine as the reaction proceeded.

R P CH3 R P CH 3 3 3 Pd Pd + Sol + PR3 R3P Sol CH3 CH3

CH # PR R P 3 3 R P CH 3 3 3 Pd Pd Pd + H3C CH3 CH CH Sol 3 Sol 3 Sol

PR 3 R3P

Pd + PR3 Pd Sol R3P Sol

Scheme 5.19. Proposed dissociative mechanism for reductive elimination from dimethylpalladium complexes.

These results and this mechanism are in direct contrast to those reported by Braterman et al. for the reductive elimination of biphenyls from cis-bis(phosphine)diaryl-Pt(II) complexes.29 In their studies, Braterman et al. examined the effect of tertiary phosphines on the thermal decomposition of several Pt(II) complexes as depicted in Scheme 5.20. Tertiary phosphines were chosen for their nucleophilic potency towards Pt(II)-substrates.

204 Chapter 5 Conjugated macrocycles by metal template approach

L Aryl T Pt Aryl Aryl L, L' = Ph2PCH2PPh2 (dppm) L Aryl Ph2PC2H4PPh2 (dppe)

PPh3

P(4-CH3-C6H4)3

L Aryl T Aryl = Ph, 4-CH -C H Pt + L' Aryl Aryl 3 6 4 L Aryl

Scheme 5.20. Cis-bis(phosphine)diaryl-platinum(II) complexes investigated by Braterman et al.

The pure Pt(II) complexes and intimate mixtures of them with free phosphines were thermolyzed in solution or in neat melts. Each system was analyzed by thermogravimetric analysis (t.g.a.), differntial thermal analysis (d.t.a.) and differential scanning calorimetry (d.s.c.) and product analysis was carried out by standard methods. The data obtained by pyrolysis are summarized in Table 5.1.

Table 5.1: Temperatures of the decomposition of cis-bis(phosphine)diaryl-Pt(II) complexes; dppe =

Ph2PCH2CH2PPh2, dppm = Ph2PCH2PPh2.

Temperature of onset Complex ∆T of decomposition [K]

PtPh2(dppm) 503 -105 PtPh2(dppm) + dppm 398

Pt(4-CH3-C6H4)(dppm) 516 -106 Pt(4-CH3-C6H4)(dppm) +dppm 410

PtPh2(dppe) 516 -35 PtPh2(dppe) +dppe 481

cis-PtPh2(PPh3)2 419 -39 cis-PtPh2(PPh3)2 + PPh3 380

cis-Pt(4-CH3-C6H4)( PPh3)2 380 -15 cis-Pt(4-CH3-C6H4)( PPh3)2 + PPh3 365

Two notable effects are apparent. Firstly, all complexes containing chelating phosphine ligands were in general more stable towards thermal decomposition. Secondly, upon free phosphine addition, all systems decomposed below the normal onset temperature of the 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 205

pure compounds. The decrease in the temperature of decomposition is illustrated by the ∆T values given in Table 5.1. This effect was attributed to the nucleophilic action of the added two-electron donors towards a Pt(II) substrate resulting in increased orbital occupancy at the reactive metal center, and therefore, facilitated reductive elimination. This is consistent with a mechanism involving a five-coordinate intermediate as depicted in Scheme 5.21.

L' L R + L' L R Pt Pt RR + PtL'L2 L R L R

Scheme 5.21: Reductive elimination from cis-[PtR2L2] complexes via a five-coordinated intermediate.

The intramolecular character of the reductive eliminatiton from the cis-bis(phosphine)- diaryl-Pt(II) complexes was evidenced by pyrolysis of a 1:1 mixture of Pt(4-CH3-

C6H4)2(dppm) and PtPh2(dppm) (Scheme 5.22). As products, only biphenyl and 4,4’- dimethyl-biphenyl were formed. The product 4-methylbiphenyl that would result by a binuclear elimination could not be detected.

Ph Ph + Ph Ph T P P Pt + Pt P P Ph Ph Ph Ph

Scheme 5.22: Pyrolysis of a 1:1 mixture of Pt(4-CH3-C6H4)2(dppm) with PtPh2(dppm); dppm=Ph2PCH2PPh2.

Thermolysis of cis-bis(phosphine)diaryl-Pt(II) was performed in toluene. The reductive elimination from cis-PtPh2(PPh3)2 and cis-Pt(4-CH3-C6H4)2( PPh3)2 was shown to be a concerted unimolecular first-order process, similar as in melt. At 80°C both complexes cleanly eliminated the expected coupling products biphenyl and 4,4’-dimethylphenyl, respectively. The kinetic consequences on the rate of elimination by added free phosphines have only been briefly investigated in solution. However, a slight acceleration of rate for the above mentioned two complexes was observed in the presence of a substancial excess 206 Chapter 5 Conjugated macrocycles by metal template approach

of PPh3. Interestingly, neither of the chelate complexes PtPh2(dppe), PtPh2(dppm) or Pt(4-

CH3-C6H4)2(dppm) (dppe = Ph2PCH2CH2PPh2, dppm = Ph2PCH2PPh2) gave any indication of thermal decomposition at 60°C. PtPh2(dppe) was still stable even in the presence of additional dppp, whereas the dppm containing complexes were observed to slowly eliminate biaryls when high excesses (10 mol equiv.) of dppm were introduced. The authors accounted the higher thermal stability of the chelate complexes to the lower flexibility and sterical demanding character of the chelating ligands rendering their requisite interaction more difficult. The tendency of bulky ligands such as PPh3 to stabilise low coordination numbers and oxidation states of metals and thus, enhancing reductive elimination was reported also by others.30

How easy the reductive elimination proceeds is also dependent on the metal, but also on its oxidation state and coordination environment. The stability of the group 10 transition metal complexes increases in the order Ni < Pd < Pt. Most 18-electron complexes of these transition metals are found to easily undergo reductive elimination, while 16-electron compounds are generally more stable.31,32,33 Ab initio calculations on the energies and geometries involved in the process of reductive elimination from Pd and Pt bis(phosphine) complexes have been extensively carried out by Low et al. in 1986 on simple model 34 systems. In Figure 5.4 the estimated energies for CH3-CH3 reductive coupling from

(PH3)2M(CH3)2 (M = Pd, Pt) and Cl2(PH3)2Pt(CH3)2 complexes on the basis of changes in electronic configuration of the metal at optimized geometries are given. The C-C coupling was found to be exothermic by 45.8 kcal/ mol for Pd(II), by 27.0 kcal/ mol for Pt(IV) and by 11.3 kcal/ mol for Pt(II) complexes. The activation energy barrier increases in the same order. These data support and explain the observations why reductive C-C couplings are facile for Pd(II) and more favourable for Pt(IV) than for Pt(II) systems. 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 207

II 0 II 0 (PH ) Pd(CH ) + H C CH (PH3)2Pt(CH3)2 Pt(PH3)2 + H3C CH3 3 2 3 2 Pd(PH3)2 3 3

∆ H# = 10.3 kcal / mol

∆ H# = 41.1 kcal / mol

∆ H298 = -45.8 kcal / mol

∆ H298 = -11.3 kcal / mol

IV II H C CH (PH3)2Pt(CH3)2Cl2 Pt(PH3)2Cl2 + 3 3

∆ H# = 34.2 kcal / mol

∆ H298 = -27 kcal / mol

Figure 5.4. Calulated energies at 298 K for the reductive eliminations from (PH3)2M(CH3)2 (M = Pd, Pt) and

Cl2(PH3)2Pt(CH3)2 complexes.

A notable trend for all transition metals is the lower stability of cis-diaryl complexes relative to cis-dialkyl complexes. For example, as shown in Scheme 5.23, bis(phosphine)diaryl-Pt complexes readily eliminate diaryls,29 while bis(phosphine)- dialkyl-Pt complexes are very stable.31

R3P Alkyl T Pt Alkyl Alkyl R3P Alkyl

Alkyl = CH3, C2H5

R P Aryl 3 T Aryl = ; Pt Aryl Aryl R P 3 Aryl

Scheme 5. 23. Thermal stability of bis(phosphine)Pt complexes.

208 Chapter 5 Conjugated macrocycles by metal template approach

The calculations made by Low showed that the hybridization of carbon atom bound to the metal turned out to have a significant effect on the mechanism and on the rate of reductive elimination.34 The increased s-character of the sp2-orbitals of aryl ligands causes this orbital to be less directional than the sp3-hybrids of alkyl ligands. Decreased directionality of an orbital leads to increased multicenter bonding in the transition state and lower activation energy for reductive elimination and thus for aryl-aryl coupling. Energies for an elimination from alkynyl-derived complexes have not been calculated. However, due to the sp-hybrids of the alkynyl ligands, Low proposed an increased lability for these complexes which would result in a faster reductive elimination under concomitant C-C bond formation. The ligands attached to the metal have a strong influence on the driving force of the elimination process. Negishi et al. have systematically investigated the effects of PR3 ligands and of various organic moieties on the reductive elimination from various ’ 35 complexes of type R 2Pd(PR3)2 (Scheme 5.24). Reaction of each complex Cl2Pd(PR3)2 was performed with a different organolithium compound R’Li and the yields of the C-C coupling products R’-R’ were determined after 1 hour and 3 hours. The results obtained are summarized in Table 5.3

THF ' 1 (PR ) Pd Cl + RLi [ (PR3)2PdR 2 ] R' R' 3 2 2 rt

PR3 = PPh3, PMePh2, PMe2Ph, PEt3

t-Bu R' = Ph , Me, t-Bu ,

Scheme 5. 24.Palladium complexes investigated by Negishi et al.

As indicated from these data, the ease of reductive elimination is inversely proportional to the basicity of PR3 (PEt3 < PPhMe2 < PPh2Me < PPh3). Especially noteworthy is the high contrast between PEt3 and PPh3. The reaction of Cl2Pd(PPh3)2 with the four different organolithium compounds gave the coupling products R’-R’ in > 95% yields within 1 h. By NMR experiments Negishi et al. could demonstrate that the instable intermediate ’ R 2Pd(PR3)2 is rapidly formed. The subsequent reductive elimination from these intermediates afforded the coupling products. In contrast, treatement of Cl2Pd(PEt3)2 with ’ the organolithium compounds yielded nearly quantitatively R 2Pd(PR3)2. The C-C coupling 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 209

products were detected, as can be seen in Table 5.3, only in very small amounts. The results obtained from this experiment provided also evidence for the higher stability of dialkyl-metal complexes compared to the diaryl-metal complexes. The reductive elimination of phenyl-groups from the metal center was faster with all phosphine ligands and afforded the C-C coupling products in higher amounts.

Table 5.3. Effects of the phosphines and organic moieties on the ease of the reductive elimination from ’ R 2Pd(PR3)2 affording C-C coupling products.

yield of R’-R’ [%] pKa of Cl2Pd(PR3)2 PhLi t-BuCH=CHLi t-BuC≡CLi MeLi PR3 1 h 3 h 1 h 3 h 1 h 3 h 1 h 3 h

Cl2Pd(PEt3)2 8.65 8 9 < 2 < 2 < 2 < 2 < 2 < 2

Cl2Pd(PMe2Ph)2 6.50 32 37 85 85 < 5 < 5 < 5 < 5

Cl2Pd(PMePh2)2 4.65 63 66 95 95 < 5 < 5 15 40 > > > Cl2Pd(PPh3)2 3.68 > 95 > 95 > 95 > 95 > 95 95 95 95

As previously mentioned, 16-electron compounds of Ni, Pd and Pt are more stable, whereas 18-electron complexes of these complexes readily undergo elimination reactions. The higher lability of the 18-electron species is demonstarted by examples shown in

Scheme 5.25. The activation energy for the composition of (bpy)NiEt2 (bpy = 2,2’- bipyridine) in the solid state to give butane is 275 kJ/mol. In the presence of olefins, however, this parameter is reduced to about 65 kJ/mol, and the 18-electron intermediate

(bipy)NiEt2(CH=CH2)X could be isolated. In many similar reactions that are based on such addition-elimination steps the 18-electron species are not detectable. As an example, in Scheme 5.24 the formation of methane from a platinum complex by the addition of HCl is depicted.36 210 Chapter 5 Conjugated macrocycles by metal template approach

C2H5 Ea = 275 kJ/mol (bpy)Ni bpy + Ni + H5C2 C2H5 C2H5

H2C CHX

H2C CH

C2H5 Ea = 65 kJ/mol

(bipy)Ni (bpy)Ni (H2CCH)X+ H5C2 C2H5

C2H5 X

H Et P Et P Cl Et P CH HCl 3 CH3 3 3 3 Pt CH Pt Pt Et P + 4 Et3P Cl 3 Cl Et3P Cl Cl

Scheme 5. 25. Enhanced reductive elimination from 18-electron transition metal complexes: a) formation of butane; b) direct formation of methane.

Labile 18-electron intermediates are believed to be involved in the Ni(II)-catalyzed Kumada cross-coupling reactions of organic halides and Grignard reagents (Scheme 5.26).

L2NiX2 1 2 R-MgX 2 MgX2

L2NiR2

R'X' 2 R-R

5 R' L2Ni RMgX R-R' X' 3

MgXX'

R' R' L2Ni R' R L2Ni R X

4 R'X'

Scheme 5.26. Catalytic cycle of the Kumada cross-coupling reaction. 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 211

The key steps which occur in both, the formation of the active catalyst and in the catalytic cycle (step 2 and 4), are the reactions of the organic halide with the diorganyl-Ni(II) complexes. They proceed most likely via oxidative addition of R’X to form 18-electron intermediates, followed by reductive elimination of a homo-coupling product R-R and R- R’, respectively.

Alkyl-Pt(IV) complexes are relatively stable compared to the other transition metal complexes and thus, are convenient substrates to study mechanisms. Numerous theoretical and experimental studies on reductive elimination reactions from such complexes have been carried out by Puddephatt37 or Goldberg.38 Reductive elimination from octahedral methyl-Pt(IV) complexes was proved to occur through a cationic five-coordinated intermediate, which is formed by preliminary ligand dissociation. For the complexes of the − general formula PtMe3XL2, (X = halide, L = tertiary phosphine ligand), either X + dissociates forming the five-coordinate species is [PtMe3L2] or ligand L dissociates to form intermediate [PtMe3XL]. From here reductive elimination rapidly takes place affording, as illustrated in Scheme 5.27, by route a the C-C coupling product. By route b, also methyl halogenide can be formed.

CH 3 CH + route a L 3 route b CH3 CH - Pt 3 L -X -L H C L 3 Pt CH CH Pt CH L 3 X 3 L 3

CH3 X

H C CH H C X + H3C CH3 3 3 3

Scheme 5.27 Mechanistic pathways of the reductive elimination from complexes of the type PtMe3XL2.

The production of ethane from Pt(IV) complex (PEt3)2PtMe3I has been reported to occur easily at 60 °C and from (dppe)PtMe3I (dppe = Ph2PCH2CH2PPh2) at a slightly higher temperature (79°C) (Scheme 5.28).37 Very easy reductive elimination was observed for complexes of the type [PtMe3XL] when X is a weakly coordinating ligand. The complex

[PtMe3(CF3SO3)(dppe)] which contains the highly labile triflate ligand, undergoes rapid reductive elimination even at room temperature affording ethane quantitatively. 212 Chapter 5 Conjugated macrocycles by metal template approach

CH3 R P CH 3 3 R P CH Pt 60°C 3 3 Pt H C CH R P CH + 3 3 3 3 R P I I 3

Ph Ph Ph CH3 Ph P CH3 79°C P CH Pt 3 Pt H C CH P CH + 3 3 3 P I Ph I Ph Ph Ph

Ph Ph Ph CH3 Ph CH P 3 rt P CH Pt 3 Pt H C CH P CH + 3 3 3 P O S-CF Ph 3 3 Ph O3S-CF3 Ph Ph

Scheme 5.28. Thermolysis of (PEt3)2PtMe3I (dppe)PtMe3I and [PtMe3(CF3SO3)(dppe)].

In comparison, complexes of the type [PtMe4L2] with chelating ligands (L2) and without anionic X groups which can easily dissociate were found to be more robust towards reductive elimination. Mechanistic investigations on the thermal reactivity of

(dppp)PtMe4 and (dppbz)PtMe4 (dppe = Ph2PCH2CH2PPh2, dppbz = o-Ph2PC6H4PPh2) 38 were intensively carried out by Goldberg et al. The thermolysis of (dppp)PtMe4 in THF required several days at 165°C to give quantitative formation of ethane and (dppe)PtMe2 (Scheme 5.29). The reaction exhibited first-order kinetic behaviour. Evidence for the intramolecular character of the reaction was provided through crossover experiments. The results from these experiments excluded a mechanistical pathway including radicals.

Ph Ph Ph CH3 Ph P CH3 THF / 165°C P CH Pt 3 Pt H C CH P CH [quant.] + 3 3 3 P CH Ph CH 3 Ph 3 Ph Ph

Ph Ph Ph Ph CH CH3 3 CH CH THF / 165°C P 3 P 3 H C CH Pt Pt + 3 3 [4 %] P CH P CH3 3 Ph CH Ph CH3 Ph 3 Ph

Scheme 5.29. Thermolysis of (dppp)PtMe4 and (dppbz)PtMe4. 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 213

However, thermolysis of (dppbz)PtMe4 at 165°C for the same period of time resulted only 4% conversion to ethane production. With a benzene backbone, dppbz is a more rigid ligand than dppe and thus, the formation of a five-coordinate intermediate by phosphine dissociation is believed to be inhibited. The reduced rate of ethane formation found for

(dppbz)PtMe4 provided strong support for an elimination reaction which proceeds, as shown in Scheme 5.30, through pre-dissociation of one end of the phosphine chelate to form a five-coordinate intermediate.

Ph Ph Ph Ph CH Ph CH 3 Ph 3 P P CH3 CH CH3 3 H C CH Pt P Pt + 3 3 Pt P CH P CH CH 3 3 P 3 Ph Ph CH CH Ph Ph 3 Ph 3 Ph

Scheme 5.30. Mechanistic pathway for reductive elimination from complexes of the formula [PtMe4L2].

Competitive to the reductive elimination, an intramolecular isomerization of the octahedral Pt(IV) complexes to a mixture of fac- and mer-isomers were shown to take place.37 Puddephatt et al. prepared a selectively deuteriated Pt(IV) complex by oxidative addition of CD3I to cis-Pt(CH3)2(PMe2Ph)2 (Scheme 5.31). Up on heating the product to 68°C in benzene, intramolecular scrambling of the alkyl groups took place. Increasing the temperature above 120°C led to decomposition, affording a mixture of CH3-CH3 and CD3-

CH3 in an expected ratio of 1:2. The proposed mechanism proceeds through a similar five- coordinated intermediate formed by phosphine dissociation, followed by a recombination reaction. The ligand dissociation and recombination steps, and thus the isomerization, are in general fast with respect to the elimination reaction. As a result, in many cases a mixture of products is obtained.

214 Chapter 5 Conjugated macrocycles by metal template approach

CH CD PhMe P 3 3 CH 2 CD I PhMe P 3 Pt 3 2 PhMe P Pt 2 CH trans-addition PhMe P 3 2 CH I 3

-PMe2Ph

CD3 CH3

PhMe2P Pt CH I 3

reductive intramolecular scrambling 68°C elimination 140°C of alkyl groups

+ PMe2Ph + PMe2Ph

CH3 CH CH PhMe2P 3 PhMe P 3 H C CD 2 Pt + H3C CH3 + 3 3 Pt PhMe P PhMe P 2 I 2 CD I 3

Scheme 5.31. Reductive elimination versus isomerization in an octahedral coordinated Pt(IV) complex.

As stated previously, transition metal complexes in which the orbital occupation is higher than usual for the respective metal are prone to undergo elimination reactions. Furthermore, any reaction or change in conditions which increases orbital occupation is likely to be followed by a counter-balancing elimination. Consequently, a widespread method to induce C-Pt bond cleavage in stable Pt(II)-substrates is the oxidatively induced reductive elimination. The oxidative activation can be achieved by the addition of electrophiles such as H+ or halogens to the electron rich metal centers. As previously mentioned, the addition of symmetrical or unsymmetrical reagents such as alkyl halides or halogens at unsaturated metal centers enhances elimination processes. In all cases, a high valent d6 platinum (IV) species, even if not isolable, is generated. Investigations where only one-electron transfer processes are used to stimulate elimination from organoplatinum complexes are limited.39 Recently, an interesting example of induced reductive elimination of alkynyl-aryls from a cis-(aryl)(alkynyl)Pt(II) complex was reported. The reaction was shown to be promoted by a ferrocenyl group which was attached to the alkynyl moiety. The Pt(II) complexes were reacted with one equivalent

DDQ or AgBF4 at room temperature. Although both reagents are mild oxidants and actually not sufficiently strong to oxidize Pt(II) the reaction yielded the coupling products quantitatively (Scheme 5.32).40 The mechanism of this reaction was investigated by 5.5 Carbon-carbon bond formation by reductive elimination from a transition metal center 215

cyclovoltammetry. It was shown that DDQ or AgBF4 first oxidize the ferrocenyl moiety to give the ferrocenium cation A. Via electron transfer from A to the Pt(II), species B is formed, in which the electron density on the metal center is diminished. Hence the rapid reductive elimination to the coupling product C becomes possible.

P DDQ or AgBF4 P Pt R Fe Fe C

R -e

R = H, Me, Cl, OMe, COOEt, COCH3 -e P P P Pt P Pt + Fe Fe*

R R A B

Scheme 5.32. Oxidatively induced reductive elimination by electron transfer. 216 Chapter 5 Conjugated macrocycles by metal template approach

5.6 Results and discussions

5.6.1 Metal template approach - method development

5.6.1.1 Palladium mediated macrocyclization of diethynyl-quinquethiophene

A new Pd(II)-mediated coupling reaction in which linear ethynylthiophenes are almost quantitatively converted to the corresponding thienyl-butadiynes has been developed and presented in Chapter 4.3.1. The proposed mechanism for this reaction is given in Scheme 5.33. An essential step is considered to be the formation of cis-bisalkynyl-Pd complexes which can spontaneously eliminate the corresponding alkyne homo-coupling products. Formation of only the coupling products 15 and 16 is attributed to the high reactivity of the intermediate bis(thiophene alkynyl)-Pd(II) complexes.

2 R H

2 NR3HX 2 NR3 R Bu Bu Bu Bu CuX L Pd S L2PdX2 2 R = 1 S S n R (n = 0,1) 2 R R

X L Pd 3 2 5 2

R X L Pd 2 L2Pd X R

CuX

2 R H 2 NR HX 3 2 NR 3

Scheme 5.33. Reaction mechanism for the Pd-catalyzed oxidative homo-coupling of alkynes. 5.6 Results and discussion 217

In order to prove if this assumption is valid when the reaction is performed under stochiometric conditions, ethynylthiophene 14 was reacted with one equivalent

Pd(dppp)Cl2 under the same reaction conditions (Scheme 5.34). A bis(thienylethynyl)- Pd(II) intermediate could not be detected, the reaction yielded the homo-coupling product 15 immediately.

Bu Bu Bu Bu Ph Ph CuI (10 mol%) P Cl Ph Ph NEt3 (2 equiv.) S Pd P 2 + S P Cl toluene, r.t., 24 h Pd R Ph Ph S P Ph Ph KOH/MeOH 14 R = Si(CH ) 3 3 Bu THF, 3h, r.t. 30 R = H Bu

Bu Bu Bu Bu

S S 15 [quant.]

Scheme 5.34. Reaction of 2-ethynylthiophene 14 with Pd(dppp)Cl2.

The concept of metal template approach towards the synthesis of fully conjugated macrocycles has been introduced in Chapter 5.2. The idea behind a template approach in the synthesis of macrocycles bases on the preorganization of the precursors in such a way that the local concentration of the reactive functionalities is considerably increased and thereby, the macrocyclization is facilitated. In the above presented homo-coupling of alkynes, the cis-arrangement of the reactive dialkynyl-Pd intermediates corresponds exactly to such a structure in which a high local concentration of the alkynyl units is realized. Based on this, it was anticipated that formation of cyclic structures should be enhanced when reacting a diethynylated oligothiophenes with cis-oriented Pd- complexes. Macrocycles having σ-bonded dialkynyl-Pd units in the backbone could form either as stable compounds or as cyclic intermediates, which subsequently by a reductive elimination would generate C-C homo-coupling products. Since the reductive elimination from a transition metal center is a concereted intamolecular process, the cyclic structure of the products should be preserved. 218 Chapter 5 Conjugated macrocycles by metal template approach

Following, the Pd-mediated oxidative coupling reaction was examined in the prospect of a macrocyclization of diethynyl-quinquethiophene 28. As shown in Scheme 5.35, firstly the TMS-groups of the protected diethynyl-quinquethiophene 24 were removed under basic reaction conditions. Subsequently, the diethynyl-quinquethiophene 28 was reacted with

Pd(dppp)Cl2 in toluene in the presence of triethylamine by using catalytic amounts of CuI. The reaction mixture was allowed to stir at room temperature for 24 h. During this time complete consumption of 28 occurred which was monitored by TLC. After work-up and removal of polymeric and inorganic material the product was characterized by NMR, analytical HPLC and MALDI-TOF MS.

Bu Bu Bu Bu Bu Bu Ph Ph P Cl S S Pd S S S + R R P Cl Ph Ph

KOH 24 R = Si(CH3)3 MeOH / THF 28 R = H

CuI (10 mol%)

NEt3 (2 equiv.) toluene, r.t., 24h

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S n-1 Bu Bu Bu Bu 29a-d

C[5T-DA] n (n = 2-6)

Scheme 5.35. Macrocyclization of diethynyl-quinquethiophene 28 by the metal template approach using a cis-Pd complex as precursor.

In the 1H NMR spectrum of the red coloured crude product, a lack of acetylenic proton at around 3.5 ppm indicates the formation of cyclic products. In Figure 5.5, the MALDI-TOF mass spectrum of the raw product is displayed. The signals correspond to the mass of the cyclic homologues C[5T-DA]n with n = 2 to n = 5. The total yield isolated was 13 % 5.6 Results and discussion 219

(based on 24). As a comparison, the macrocyclization of 28 by the statistical Glaser coupling afforded under optimized reaction conditions cyclic homologues C[5T-DA]n (n = 2-6) 29a-e in 16 % total yield (see Chapter 4.4.2.).

29b 29a (n=3) (n=2) 2384 100 1589

29c 60 (n=4) / a.u. I 40 3178 29d (n=5) 20 3973

0 1000 1500 2000 2500 3000 3500 4000 m / z

Figure 5.5 MALDI-TOF MS spectrum of the crude product C[5T-DA]n 29a-d obtained from the Pd- template-directed macrocyclization reaction.

In Figure 5.6, the HPLC chromatogram of the product mixture generated by the Pd template-directed macrocylization is displayed. Besides four main peaks which correspond to the homologues 29a-d, the analytical HPLC chromatogram reveals that the product mixture contains other by-products that were not detected by NMR or MALDI-TOF MS analysis. The small peaks at retention times between 2 and 5 min might be attributed to linear oligomers which, as known from previous works,41 have shorter retention times as the corresponding cyclic compounds. The peaks at 12.5 min and 16.1 min in the HPLC chromatogram indicate the existence of higher cyclic oligomers that were not be detected in the product mixture by MALDI-TOF MS. 220 Chapter 5 Conjugated macrocycles by metal template approach

(n=2) (n=3)

(n=4)

(n=5) (n=6)

t [min]

Figure 5.6. HPLC chromatogram of the product mixture C[5T-DA]n obtained by Pd template-directed macrocyclization.

Differences between the statistical and the template-directed macrocyclization reaction become apparent when one compares the ratios in which the macrocycles were formed. The statistical Glaser coupling of diethynylated quinquethiophene 28 led to the formation of the trimeric macrocycle C[5T-DA]3 29b as the major product in 8.8 % isolated yield.

The cyclodimer C[5T-DA]2 29a was formed in 3 % yield. In comparison, by the Pd- template directed method the ratio determined between the dimer and trimer changed significantly. The cyclodimer C[5T-DA]2 29a being formed in nearly similar amounts as the cyclotrimer C[5T-DA]3 29b. As determined by integration of the peak areas in the HPLC chromatogram, the ratio between the two macrocycles was 1.0 : 1.1. This is considered to be the result of the template effect. The low yields and mixture formation might be attributed to the low stability of the Pd-acetylide linked metallacycle making the reductive elimination from the presumed cyclic intermediate and the C-C bond formation faster than its formation. As previously mentioned, stable macrocycles having σ-bonded acetylide units to cis- oriented transition metal have been reported for platinum.4 With the aim to increase the stability of the metallacycle intermediates formed during the alkynyl-alkynyl coupling reaction, the use of cis-oriented platinum instead palladium complexes was decided.In order to explore the synthetic feasibility, this reaction has been firstly investigated by 5.6 Results and discussion 221

simple molecular systems, namely 3.4-dibutyl-2-ethynylthiophene 14 and 3,3’’,4,4’’- tetrabutyl-2-ethynylterthiophene 12. In the following sections, the detail synthesis of novel bis(oligothiophene alkynyl)-Pt(II) complexes and studies concerning the reductive elimination from these complexes under alkynyl-alkynyl bond formation are presented.

5.6.1.2 Synthesis of linear bis(oligothiophene alkynyl)-Pt(II) complexes

First attempts to prepare a bis(oligothiophene alkynyl)-Pt(II) complex was based on the common method developed by Hagihara et al.42 Transition metal σ-acetylide complexes were synthesized by a Cu(I)-catalyzed dehydrohalogenation reaction of metal halides in the presence of an amine as solvent. These reaction conditions are similar to those described in the previous chapter for the Pd-templated macrocyclization reactions. As starting material 3,4-dibutyl-2-ethynylthiophene 14 was employed. The synthesis of 14 was already described Chapter 4.3.1. In the long-term interest of generating products in which the organyl units are cis-bonded to the metal, cis-Pt(dppp)Cl2 was chosen as the starting metal complex. The chelating ligand dppp was expected to enforce the cis- geometry around the platinum center and to prevent a cis-trans isomerization. Firstly, the TMS-protected ethynylthiophene 14 was converted under mild basic conditions to the free acetylene 34. Then, reaction of 34 with 0.5 equivalent of Pt(dppp)Cl2 in the presence of catalytic amounts of copper iodide and two equivalents of triethylamine in toluene generated bis(thiophene alkynyl)-Pt(II) complex 35 as major product (Scheme 5.36). Additionally, the homocoupling product bisthienyl-1,3-butadiyne 15 was formed in traces. Due to the very often observed instability of transition metal complexes in halogenated solvents a mixture of petrol ether and THF was used as eluent for the chromatographic purification. Thiophene derived Pt-complex 35 was isolated in 71 % yield as a stable pale yellow solid that is soluble in organic solvents but shows slow decomposition in chlorinated solvents. Structural evidence of this new bis(acetylide)-Pt complex 35 was obtained by 1H and 31P NMR spectra, MALDI-TOF mass spectrometry, as well as elemental analysis.

222 Chapter 5 Conjugated macrocycles by metal template approach

Ph Ph P Cl Bu Pt Bu Bu Bu P Cl Ph Ph Ph Ph S P S R CuI (10 mol%) Pt NEt (2 equiv.) S P 3 Ph Ph toluene, r.t., 24h KOH/MeOH 14 R = Si(CH3)3 Bu THF, 3h, r.t. 34 R = H Bu [72 %] 35

Scheme 5.36. Synthesis of bis(thiophene alkynyl)-Pt(II) complex 35.

The 1H NMR spectrum of 35 is displayed in Figure 5.7. In the aromatic region, two multiplets at δ = 7.81 ppm and δ = 7.34 ppm that corresponds to the phenyl-groups of the dppp-moiety are apparent. Upon formation of the platinum-ethynylthiophene complex, for the H-5 proton of the thiophenes in 14 a clear high-field shift from δ = 6.80 ppm to δ =

6.43 ppm is observed. The protons of the ß-CH2-dppp moiety appear as broad signals at 2.43 ppm and 2.24 ppm, respectively. In the aliphatic region, further, signals assigned to the butyl-chains are present. It is noteworthy that the methyl-groups in 35 appear as two separated triplets at δ = 0.89 ppm and δ = 0.70 ppm, whereas in the spectrum of the precursor 14 they were seen as one multiplet at δ = 0.95 ppm. This is probably due to interactions of the butyl groups with the phenyl groups of the phospine ligand. The 31P NMR spectrum of 35 which is illustrated in Figure 5.8 shows only one signal at δ = -6.79 1 ppm with a set of platinum satellites ( JP-Pt = 2203 Hz). These values are consistent with typical shifts observed for the cis-platinum-acetylide complexes.4 Further structural evidence of 35 was obtained from its MALDI-TOF mass spectrum which showed one characteristic peak at m/z = 1044.5 (Figure 5.9). This value corresponds very well to the theoretical molecular mass of the compound (m/z = 1045.5). The experimental isotropic distribution was found to be in very good agreement with the calculated one (Figure 5.10).

5.6 Results and discussion 223

7.8141 7.3453 7.3305 6.4254 2.4390 2.3553 2.3380 2.3361 2.3165 2.2827 2.2638 2.2442 2.0390 1.4944 1.3271 1.3085 1.2368 1.0553 1.0361 0.9054 0.8871 0.8688 0.7169 0.6989 0.6806

CH3-Bu

H-aryl-dppp

β,γ-CH -Bu H-5-Th 2

ß-CH2-dppp α-CH2-Bu

α-CH2-dppp

8 12 2 428 4 8 4 6 6 8 7 6 5 4 3 2 1 0 (ppm)

Figure 5.7. 1H NMR spectrum of bis(thiophene alkynyl)-Pt(II) complex 35.

-6.7908

JPtP = 2203 Hz

20 10 0 -10 -20 -30 -40 (ppm)

Figure 5.8. 31P NMR spectrum of bis(thiophene alkynyl)-Pt(II) complex 35.

224 Chapter 5 Conjugated macrocycles by metal template approach

1044.5 100

60 / a.u. I 40

0 500 1000 1500 2000 2500 m / z

Figure 5.9. MALDI-TOF MS spectrum of bis(thiophene alkynyl)-Pt(II) complex 35.

g 100 90 80 70 60 50 40 30 20 10 0 1.040 1.045 1.050 1.055

100

0 1040 1045 1050 1055 m / z

Figure 5.10. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of bis(thiophene alkynyl)-Pt(II) complex 35. 5.6 Results and discussion 225

In order to demonstrate the general applicability of this coupling method in which an ethynylthiophene moiety is introduced as a η1 –ligand in a platinum complex, in the next reaction the higher homologue, namely ethynylated terthiophene 12, has been subjected to the identical reaction conditions. Firstly, TMS-protected ethynylterthiophene 12 was synthesized as previously described in Chapter 4. Selective monobromination of terthiophene 4 with NBS in DMF and a following Pd-mediated coupling reaction of the brominated terthiophene 8 with trimethylsilylacetylene afforded ethynylated product 12 in 94% yield (Scheme 5.37).

Bu Bu Bu Bu Bu Bu Bu Bu NBS S S S S DMF S S Br

4 8 [65%]

Bu Bu Bu Bu

H Si(CH3)3 S Pd(PPh )Cl / PPh S S 3 2 3 Si(CH3)3 CuI / pyridin / NEt3 12 [94%]

Scheme 5.37. Synthesis of protected ethynyl-terthiophene 12.

Cleavage of the TMS-group of protected ethynyl-terthiophene 12 was performed under mild basic conditions. Reaction of the resulting ethynyl-terthiophene 36 with Pt(dppp)Cl2 under identical conditions as described above afforded the cis-configurated bis(terthiophene alkynyl)-Pt(II) complex 37 (Scheme 5.38). After purification by column chromatography the product 37 could be isolated in 75 % as a stable yellow solid that is soluble in organic solvents but shows slow decomposition in chlorinated solvents. The Pt- complex 37 was characterized by 1H and 31P NMR spectroscopy, MALDI-TOF mass spectrometry, as well as elemental analysis.

226 Chapter 5 Conjugated macrocycles by metal template approach

Bu Bu Ph Ph Bu P Cl Bu Pt S S Bu Bu Bu Bu P Cl Ph Ph S S Ph Ph P S S Pt R CuI (10 mol%) S P NEt (2 equiv.) Ph Ph 12 R = Si(CH ) 3 KOH/MeOH 3 3 S S toluene, r.t., 24 h Bu THF, 3h, r.t. 36 R = H Bu Bu Bu 37 [75 %]

Scheme 5.38. Synthesis of bis(terthiophene alkynyl)-Pt(II) complex 37.

In the aromatic region of the 1H NMR spectrum of 37, the multiplet corresponding to the dppp-phenyl groups appear at δ = 7.81 ppm and δ = 7.38 ppm (Figure 5.11). The singulet at δ = 6.81 ppm belongs to the H-5’’ terminal protons of the terthiophenes. The ß-protons of the middle thiophene in the terthiophene unit appear as two doublets at δ = 6.91 ppm and δ = 6.82 ppm, which compared to those of the ethynyl-terthiophene 12 (d, 7.06 and d,

7.04) are strongly high field shifted. Very characteristic is the broad signals of the ß-CH2- dppp moiety at δ = 2.48 ppm. The signal of the α -CH2-dppp moiety appears at δ = 2.06 ppm being overlapped with the other signals belonging to the butyl-groups. The 31P NMR spectrum of 37 is illustrated in Figure 5.12. In accordance with the cis-configurated structure of Pt-complex 37, only one signal at δ = -7.13 ppm with a symmetrical set of 1 platinum satellites ( JP-Pt = 2198 Hz) is visible. These values are consistent with those found for the smaller homologue Pt-complex 35. In the MALDI-TOF MS spectrum of 37 a characteristic peak at m/z = 1597.6, very well corresponding to the theoretical molecular mass of this compound (m/z = 1597.0) was observed (Figure 5.13). As illustrated in Figure 5.14, the measured isotropic distribution is isdentical to the calculated one.

5.6 Results and discussion 227

7.8125 7.3825 7.3667 6.9702 6.9607 6.9140 6.9045 6.8218 2.7166 2.6971 2.6762 2.6055 2.5878 2.5664 2.5424 2.5234 2.5032 2.4818 2.3492 2.3302 2.3107 2.0556 1.6774 1.4261 1.1136 0.9867 0.8793 0.7707 0.7524 0.7341

CH3-Bu 2.7166 2.6971 2.6762 2.6055 2.5878 2.5664 2.5424 2.5234 2.5032 2.4818 2.3492 2.3302 2.3107

H-aryl-dppp α-CH2-dppp

H-3‘,4‘

β,γ-CH2-Bu 2.6 2.4 ß-CH -dppp (ppm) 2

H-5‘‘-Th α-CH2-Bu

8 12 4 2 4 12 4 8 20 22 6 8 7 6 5 4 3 2 1 0 (ppm)

Figure 5.11. 1H NMR spectrum of bis(terthiophene alkynyl)-Pt(II) complex 37.

-7.1372

JPtP = 2198 Hz

20 15 10 5 0 -5 -10 -15 -20 -25 -30 (ppm)

Figure 5.12. 31P NMR spectrum of bis(terthiophene alkynyl)-Pt(II) complex 37.

228 Chapter 5 Conjugated macrocycles by metal template approach

1597.0 100

60 / a.u. I 40

0 80010001200140016001800200022002400 m / z

Figure 5.13. MALDI-TOF MS spectrum of bis(terthiophene alkynyl)-Pt(II) complex 37.

g 100 90 80 70 60 50 40 30 20 10 0 1.595 1.600 1.605

100

0 1594 1596 1598 1600 1602 1604 1606 1608 1610 m / z

Figure 5.14. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of bis(terthiophene alkynyl)-(Pt)II complex 37. 5.6 Results and discussion 229

As expected, using the chelating ligand dppp,in the starting complex Pt(dppp)Cl2 only cis- products were formed. In the literature, there are only limited examples of stereospecific synthesis of cis-platinum diacetylides by using cis-platinum dichlorides with non-chelating ligands.42b Most reports indicate an isomerisation of the thermodynamically less stable cis- platinum diacetylides to the corresponding trans-platinum diacetylides. In general, this process is favoured by polar solvents and can be catalyzed in the presence of metal salts, including copper iodide.43 Another driving force for such an isomerization can be the strain caused by the steric repulsion which is dictated by the bite angle of the dialkyne ligands and the cone angle of the phosphine ligand. In a cis square planar complex, the phosphine ligands are coplanar with the organic ligand, while in the trans configuration the phosphine ligand are rotated out of plane and mutually perpendicular to the organic ligand, resulting in less strain. However, in order to gain further insight in the characteristics of this novel class of compounds in which ethynylated oligothiophenes act as η1-ligand at the platinum center, the process of isomerization has been investigated. This was accomplished by the use of trans-Pt(PPh3)2Cl2 and cis-Pt(PPh3)2Cl2 complexes, in reactions with ethynylthiophene 34. CuI-catalyzed coupling of the trans-Pt precursor with the deprotected ethynylthiophene 34 afforded only one product in 63 % yield, which could be assigned as the trans- bis(thiophene alkynyl)-Pt(II) complex 38 (Scheme 5.38).

Bu Bu Bu Bu PPh3 trans-Pt(PPh ) Cl S 3 2 2 Pt S S CuI (10 mol%) PPh Bu R 3 Bu NEt3 (2 equiv.) KOH/MeOH 14 R = Si(CH ) toluene, r.t., 24 h 3 3 38 [63 %] THF, 3h, r.t. 34 R = H

Scheme 5.38. Synthesis of trans-bis(thiophene alkynyl)-Pt(II) complex 38.

The value and the magnitude of the phosphorus signal and the coupling constant 1J (195Pt31P) are known to be very sensitive to the nature of the ligands in square-planar Pt(II) complexes. They can act as a useful probe for the geometry of any present isomer. In the 31P NMR spectrum of 38 only one typical signal at δ = 23.6 ppm with a coupling constant 1 of JPtP = 2652 Hz could be observed being consistent with other reported trans- bis(phosphino)-Pt derivatives (Figure 5.15).4 230 Chapter 5 Conjugated macrocycles by metal template approach

23.5748

JPtP = 2652 Hz

45 40 35 30 25 20 15 10 5 0 (ppm)

Figure 5.15. 31P NMR spectrum of trans-bis(thiophene alkynyl)-Pt(II) complex 38.

When the reaction of ethynylthiophene 34 was carried out with cis-Pt(PPh3)2Cl2 instead of trans- Pt(PPh3)2Cl2, two products were formed. These were identified as a mixture of the trans and the cis-platinum complex 38 and 39, respectively (Scheme 5.39).

Bu Bu Bu Bu Bu PPh Bu 3 S S PPh cis -Pt(PPh3)2Cl2 3 Pt + Pt S S CuI (10 mol%) PPh Bu PPh R 3 Bu S 3 NEt (2 equiv.) 3 38 KOH/MeOH 14 R = Si(CH ) 3 3 toluene, r.t., 24 h Bu THF, 3h, r.t. 34 R = H Bu 39

Scheme 5.39. Reaction of ethynylthiophene 34 with cis-Pt(PPh3)Cl2.

The proof of configurations was given by the different pattern and chemical shifts of the phosphor signals in the 31P NMR spectrum which was measured in DMSO-d6. Two signals 1 at δ = 23.3 and δ = 16.7 ppm with two sets of platinum satellites JP-Pt = 2652 Hz for the 1 trans and JP-Pt = 2245 Hz for the cis isomer were observed (Figure 5.16). 5.6 Results and discussion 231

23.2585 16.7280

JPtP = 2652 Hz JPtP = 2245 Hz

40 35 30 25 20 15 10 5 0 (ppm)

Figure 5.16. 31P NMR spectrum of the product mixture obtained by the reaction of ethynylthiophene 34 with cis-Pt(PPh3)Cl2.

Figure 5.17. 1H NMR spectrum of the product mixture obtained by the reaction of ethynylthiophene 34 with cis-Pt(PPh3)Cl2.

232 Chapter 5 Conjugated macrocycles by metal template approach

In general 1H NMR chemical shifts are also influenced by the cis and trans configurations of the complexes. In the spectrum of the product mixture two different signals at 6.58 ppm and 6.45 ppm for the H-5 protons of the thiophenes were observed in a 0.8 to 1.0 ratio (Figure 5.17). The products were not separated, however, the results obtained from this experiment strongly support the requirement of using metal complexes with chelating ligands in order to obtain isomerical pure cis products.

5.6.1.3 1,3-butadiyne-linked oligothiophenes from linear dialkynyl-platinum complexes

The next and most important step in the metal template approach, namely the process of reductive elimination, has been investigated by using bis(thiophene alkynyl)-Pt(II) complex 35 as starting material. As stated in Chapter 5.5, the reductive elimination at the metal center from a square-planar d8 Pt-complex is symmetry allowed. The reaction is a concerted intramolecular process when the organic moieties occupy adjacent positions in a cis-coordination. Diaryl-Pt(II) complexes have been observed to readily eliminate under C- C bond formation affording bisaryls. In contrast, dialkyl-Pt(II) complexes are stable and have never been observed to undergo reductive elimination. So far, no investigations on analogues dialkynyl-Pt(II) complexes have been reported. In general, reductive elimination from metal centers is enhanced at higher temperatures and in polar solvents. In order to investigate if a thermal activation will allow a reductive elimination process from oligothiophenes Pt(II) systems, in preliminary experiments, the thermolysis of compound 35 in polar solvents was examined (Scheme 5.40). For this purpose, a solution of complex 35 in THF or DMSO, respectively, was refluxed for 24 h. As monitored by TLC and proven by MALDI-TOF MS, under these conditions no elimination took place. As single compound, only starting material 35 was detected. The reductive elimination from diaryl-Pt(II) complexes with dppp chelating ligand has been found to be enhanced when nucleophilic phosphine ligands were added to the solution. This is due to the mechanism of the process which in the case of diaryl-Pt(II) complexes proceeds via a five coordinated intermediate.29 Thus, the effect of added phosphine investigated. Equivalent mixtures of complex 35 and PPh3 were refluxed in THF or DMSO (Scheme 5.40). In none of the cases any formation of the coupling product 15 could be observed. 5.6 Results and discussion 233

a) THF / 70°C Bu Bu b) DMSO / 160°C c) + 1equiv. PPh / THF / 70°C 3 Bu Bu Ph Ph S d) + 1equiv. PPh3 / DMSO / 160°C Bu Bu P Pt S S P S Ph Ph 15 Bu Bu 35

Scheme 5.40. Thermal activation for reductive elimination from bis(thiophene alkynyl)-Pt(II) complex 35.

As stated in Chapter 5.5, oxidatively induced reductive eliminations are commonly encountered because an increase in oxidation state of a metal center is apt to enhance reductive elimination or even, to make it possible. Selective cleavage of Pt-C bonds in the presence of iodine was mentioned as a footnote in a report of Whitesides, but none further comments or details of this reaction were given.44 However, iodine has been considered to be a suitable compound which can oxidatively add to the Pt(II) center and thus, promote a reductive elimination process. In order to probe this suggestion, Pt(II) complex 35 was stirred in THF with one equivalent of iodine at room temperature for 24 h (Scheme 5.41). The occurrence of a reaction was indicated by the immediate disappearance of the dark red colour of iodine upon addition to the solution of the Pt-complex. After chromatographic purification with n-hexane as eluent a pale yellow solid was isolated. The structural proof by NMR and MS analysis evidenced the formation of the coupling product 1,4-bis(3,4-dibutylthien-2-yl)-butadiyne 15 in quantitative yield.

As by-product a white solid was formed which was identified as Pt(dppp)I2.

Bu Bu Ph Ph Ph Bu Bu Ph Bu Bu S P I P THF Pt + Pt + I2 S S P r.t., 24h P I S Ph Ph Ph Ph 15 [quant.] Bu Bu 35

Scheme 5.41. Synthesis of 1,4-bis(3,4-dibutylthien-2-yl)-butadiyne 15 by reductive elimination from platinum center.

234 Chapter 5 Conjugated macrocycles by metal template approach

In close analogy, the reaction of bis(terthiophene alkynyl)-Pt(II) complex 37 under similar reaction conditions by using one equivalent of iodine led to the quantitative formation of the coupling product 1,4-bis(terthien-2-yl)-1,3-butadiyne 16, giving evidence for the general applicability of this reaction (Scheme 5.42).

Bu Bu Bu Bu Bu Bu Bu Bu Bu Bu S S Bu S S S S Bu Ph Ph 16 S S S P I / THF Pt 2 + P S r.t., 24h Ph Ph [quant.] S S Ph Bu Ph Bu P I Bu Pt Bu P I 37 Ph Ph

Scheme 5.42. Synthesis of bis(terthienyl)-butadiyne 16 by the reductive elimination from platinum center.

In this work some investigations concerning the mechanistical pathway by which the reductive elimination from bis(oligothienylethynyl)platinum(II) substrates proceeds have been carried out. A detail presentation of these experiments and the results is given in Chapter 5.6.3.

5.6.2 Macrocyclization by the metal template approach

5.6.2.1 Synthesis of dinuclear bis(oligothiophene alkynyl)-Pt(II) macrocycles

The method described in the previous chapter for mono-ethynylated oligothiophenes has been extended to the synthesis of platinum metallacycles composed of Pt-diacetylide bridged oligothiophene backbones. TMS-protected diethynyl-terthiophene 23 was prepared as described in Chapter 4.3. After removing the TMS groups under mild basic conditions, diethynylated terthiophene 26 was reacted with 1 equivalent of Pt(dppp)Cl2 in the presence of a catalytic amount of CuI and two equivalents of triethylamine in toluene (Scheme 43). The progress of the reaction was monitored by TLC and after stirring for 3 days at room temperature the reaction was completed as indicated by the disappearance of the starting material and slow formation of 5.6 Results and discussion 235

a yellow solid. Subsequently, the volume of the solvent was reduced to half and MeOH was added. The mixture was stirred for several hours at room temperature and the yellow coloured precipitate was filtered off. The crude material was washed several times with MeOH and after drying characterized by 1H, 31P NMR, IR, MALDI-TOF MS and ESI- FTIR MS analysis.

Bu Bu Ph Ph Bu Bu P Cl S Bu Bu S S Bu Bu Pt Ph Ph Ph Ph Cl S P P P S Ph Ph Pt R S R Pt P P CuI (10 mol%) Ph Ph Ph Ph KOH/MeOH NEt (2 equiv.) 23 R = SiMe3 3 S S THF, 3h, r.t. 26 R = H toluene, r.t., 72h S Bu Bu Bu Bu 40 [91 %]

Scheme 5.43. Synthesis of bis(terthiophene alkynyl)-diplatinum macrocycle 40.

First evidence for the formation of a cyclic structure came from 1H NMR analysis. Due to the moderate solubility of the product in common organic solvents, the NMR spectra of 40 6 had to be recorded at higher temperatures (323 K in CDCl3 or 363 K in DMSO-d ). In the 1H NMR spectrum of 40, signals corresponding to terminal acetylenic protons at around δ = 3.5 ppm were completely lacking. In the aliphatic region, protons belonging to the butyl chains and of the dppp moieties which were partially overlapped were observed. On the left side of Figure 5.18 the aromatic region of the 1H NMR spectrum of the starting material Pt(dppp)Cl2 (top) and of the metallacycle 40 (bottom) are displayed. The two sets of signals of the dppp-ligands are well separated and for platinacycle 40 slightly shifted to higher and lower field, respectively. Very characteristic are the H-3’,4’ protons of the central thiophene ring that appear in the spectrum of product 40 as singulet at δ = 7.01 ppm. The corresponding signal of precursor 26 was seen at 7.06 ppm. The 31P NMR spectrum of metallacycle 40 indicates a single ligand environment consistent with the formation of cyclic species (Figure 5.18, right). Compared to the phosphor signal of the precursor Pt(dppp)Cl2 (δ = -10.15 ppm) the signal is low-field shifted to δ = -5.59 ppm and shows a set of Pt-satellites with a coupling constant of JPtP = 2243 Hz supporting the cis configuration at the Pt-corners. These values are in good correlation with those found for the corresponding linear terthiophene-alkynyl Pt(II)-complex 37. Due to the limited 236 Chapter 5 Conjugated macrocycles by metal template approach

solubility of metallacycle 40, the signal/noise resolution was insufficient for 13C NMR specta.

-10.15 7.8161 7.4674

H-aryl-dppp

JPtP = 3409 Hz

20 10 0 -10 -20 -30 8.4 8.0 7.6 7.2 6.8 -5.59 7.0177 7.4047 7.9224

H-3‘,4‘

JPtP = 2243 Hz

8.4 8.0 7.6 7.2 6.8 20 10 0 -10 -20 -30 (ppm) (ppm)

Figure 5.18. The aromatic region of the 1H NMR spectrum (left) and 31P NMR spectrum (right) of the starting material Pt(dppp)Cl2 (top) and the diplatinum macrocycle 40 (bottom).

The MALDI-TOF mass spectrum of the product exhibited only one signal at m/z = 2250.4 which very well corresponds to the theoretical molecular mass of the dimeric metallacycle 40 (m/z = 2250.6). Remarkably, as evidenced by the mass spectrum of 40 depicted in Figure 5.19, none of higher cyclic oligomers was formed, the only product obtained was the dinuclear platinacycle 40 in 91 % yield. The experimental isotopic distribution was in very good agreement to the calculated one (Figure 5.20). The most convincing evidence for the formation of the cyclodimer 40 resulted from the ESI-FTIR MS analysis. The mass spectrum shows only one signal at m/z = 1125.31 that can be assigned to the dicationic species (z =2) of metallacycle 40 (calcd. m/z = 1125.33). In the IR spectrum of terthiophene-derived platinacycle 40 the typical C≡C−Pt streching bands are visible at ν = 2098 cm-1 and ν = 2362 cm-1 (Figure 5.21). These values are in full accordance with other recently published platinum σ-acetylide complexes.24 The lack of the C≡C−H stretching 5.6 Results and discussion 237

band which appears for precursor 26 at ν = 3308 cm-1 gives further prove for the cyclic structure of compound 40 (Figure 5.21).

2250.6 100

60 / a.u. I 40

0 1500 2000 2500 3000 3500 m / z

Figure 5.19. MALDI-TOF MS spectrum of terthiophene-derived diplatinum macrocycle 40.

g 100 90 80 70 60 50 40 30 20 10 0 2.245 2.250 2.255 2.260

100

0 2244 2248 2252 2256 2260 2264 m / z

Figure 5.20. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of diplatinum macrocycle 40. 238 Chapter 5 Conjugated macrocycles by metal template approach

100

95 2365 2098 1630 90 3053 85

80 2860 691

Transmittance [%] 1100 75 3438 1432 2931 2948 70 516

4000 3500 3000 2500 2000 1500 1000 500

Frequency [cm-1]

Figure 5.21. IR spectrum of terthiophene-derived diplatinum macrocycle 40.

The results obtained by this reaction have encouraged further investigations to prove the general applicability of the method for the synthesis of various oligothiophene-derived metallacycles. Thus, the method developed was applied to the preparation of metallacycles containing longer oligothiophenes in the core.

The first compound used for this purpose was diethynylated quinquethiophene 28. The preparation of the TMS-protected diethynyl-quinquethiophene 24 is described in Chapter 4.3. After removal of the TMS-protecting groups, diethynylated quinquethiophene 28 was reacted with 1 equivalent of Pt(dppp)Cl2 in the presence of a catalytic amount of CuI and two equivalents of triethylamine in toluene. The reaction mixture was allowed to stir at room temperature for 72 hours. During this time formation of a red-brownish precipitate could be observed. The volume of the solvent was reduced to half, MeOH was added and the mixture was stirred for several hours at room temperature. The dark-red coloured product 41 was isolated after filtration in 96% yield and characterized by 1H, 31P NMR, IR, MALDI-TOF MS and ESI-FTIR MS analysis (Scheme 5.44).

5.6 Results and discussion 239

Bu Bu Bu Bu Bu Bu

S S S S S R R

KOH 24 R = Si(CH3)3 MeOH / THF 28 R = H

Ph Ph CuI (10 mol%) P Cl NEt3 (2 equiv.) Pt toluene, r.t., 72h P Cl Ph Ph

Bu Bu Bu Bu Bu Bu S S Ph Ph Ph S Ph S S P P Pt Pt P S S P S Ph Ph Ph Ph S S Bu Bu Bu Bu Bu Bu 41 [96 %]

Scheme 5.44. Synthesis of bis(quinquethiophene alkynyl)-diplatinum macrocycle 41.

Most signals in the 1H NMR spectrum of quinquethiophene-derived platinacycle 41 are broadened and especially in the aliphatic region where the protons of the butyl groups and dppp-moieties are present, strongly overlaid (Figure 5.22). In the spectrum no signals at δ = 3.5 ppm attributable to terminal acetylenic protons are observed providing the cyclic structure. The characteristic signals of the phenyl groups of the dppp-moieties appear as two broad signals at δ = 7.80 ppm and δ = 7.28 ppm. Further, in the aromatic region of the spectrum three groups of overlaid signals at δ = 7.09 ppm, δ = 6.99 ppm and δ = 6.90 ppm corresponding to the H-3’,3’’’.4’,4’’’ protons of the unsubstituted thiophenes are visible. The 31P NMR spectrum of platinacycle 41 shows only one signal at δ = -6.58 ppm with a 1 symmetrical set of Pt-satellites ( JP-Pt = 2195 Hz) (Figure 5.23). These values are consistent with those observed for the terthiophene-derived platinacycles 40.

240 Chapter 5 Conjugated macrocycles by metal template approach

7.8048 7.3801 7.0937 6.9748 6.9094 2.7116 2.5920 2.4725 2.3461 2.0435 1.5597 1.4483 1.3542 0.9614 0.8474 0.8275 0.8238 0.8113 0.7590

β,γ-CH2-Bu

CH3-Bu H-aryl-dppp ß-CH2-dppp

α-CH2-Bu H-3‘3‘‘‘,4‘,4‘‘‘-Th α-CH2-dppp

8 7 6 5 4 3 2 1 0 (ppm)

Figure 5.22. 1H NMR spectrum of quinquethiophene-derived diplatinum macrocycle 41.

-6.5794

JPtP = 2195 Hz

4 2 0 -2 -4 -6 -8 -10 -12 -14 -16 (ppm)

Figure 5.23. 31P NMR spectrum of quinquethiophene-derived diplatinum macrocycle 41.

5.6 Results and discussion 241

2803.3 100

60 / a.u. I 40

0 1500 2000 2500 3000 3500 4000 m / z

Figure 5.24. MALDI-TOF MS spectrum of quinquethiophene-derived diplatinum macrocycle 41.

The MALDI-TOF MS spectrum of 41 which is displayed in Figure 5.24, exhibits only a strong signal at m/z = 2803.3. This value and the experimental isotropic distribution correspond very well to the theoretical molecular mass of dimeric platinacycle 41 (m/z = 2803.9) (Figure 5.25). Final structural proof of macrocycle 41 came from high resolution ESI-FTIR mass spectrometry which shows one signal at m/z = 2801.86 that can be assigned to the cationic species of 41 (m/z = 2801.83). As for macrocycle 40, the IR spectrum of platinacycle 41 shows the typical C≡C streching bands at ν = 2103 cm-1 and ν = 2362 cm-1 , whereas no C≡C−H stretching is visible (Figure 5.26).

242 Chapter 5 Conjugated macrocycles by metal template approach

g 100 90 80 70 60 50 40 30 20 10 0 2.800 2.805 2.810 2.815

100

0 2800 2805 2810 2815 m / z

Figure 5.25. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of diplatinum macrocycle 41.

100

2362 2103 90

1633 686 80 1106 1440 512

70 2860 3437 Transmittance [%]

60 2955 2927

4000 3500 3000 2500 2000 1500 1000 500 Frequency [cm-1]

Figure 5.26. IR spectrum of quinquethiophene-derived diplatinum macrocycle 41. 5.6 Results and discussion 243

The reaction was performed by using the next homologue in this series, namely the diethynylated septithiophene 42. The synthesis of TMS-protected oligothiophene 25 was described in Chapter 4.3.2. In close analogy to the above described preparation, firstly the TMS-protective groups were removed under basic reaction conditions to generate reactive diethynyl-septithiophene 42. Subsequently, reaction of 42 with 1 equivalent of Pt(dppp)Cl2 was carried out in the presence of catalytic amount of CuI and two equivalents of triethylamine (Scheme 5.45). After stirring the mixture for 72 h at room temperature, the reaction afforded the corresponding dimeric platinacycle 43 in 97 % yield. The structure of 43 was proven by 1H NMR, 31P NMR and MALDI-TOF mass spectrometry.

Bu Bu Bu Bu Bu Bu Bu Bu

S S S S S (CH ) Si S S 3 3 Si(CH3)3

25 R = SiMe3 KOH/ MeOH/ THF 42 R = H

Ph Ph CuI (10 mol%) P Cl

NEt3 (2 equiv.) Pt toluene, r.t., 72h P Cl Ph Ph

Bu Bu Bu Bu Bu Bu Bu Bu S S S Ph Ph S S Ph Ph S P S P Pt Pt P P S S Ph Ph Ph Ph S S S S Bu S Bu Bu Bu Bu Bu Bu Bu 43 [97 %]

Scheme 5.45. Synthesis of largest member 43 in the bisplatinum bis-σ-acetylide macrocycle series.

The cyclic structure of 43 was confirmed by 1H NMR. As can be seen from Figure 5.27, no signal at around 3.5 ppm attributable to terminal acetylenic protons is visible. The NMR spectrum of septithiophene-derived platinacycle 43 exhibits the same features as that of the quinquethiophene analogue 41. Easy to distinguish in the aromatic region are the broad signals of the phenyl groups of the dppp moieties at δ = 7.80 and 7.38 ppm. The protons 244 Chapter 5 Conjugated macrocycles by metal template approach

corresponding to the unsubstituted thiophene rings are visible as three sets of signals at δ = 7.06 ppm, 7.01 ppm and 6.93 ppm. Compared to the precursor 25 whose corresponding protons appear at δ = 7.11 ppm, 7.09 ppm and 7.08 ppm respectively, the signals are slightly high field shifted. In the aliphatic region the spectrum shows broad signals that correspond to the butyl-groups which are partially overlapped with the signals of the dppp moieties. The 31P NMR spectrum exhibit only one typical signal at δ = -7.17 ppm with a 1 set of platinum satellites ( JP-Pt = 2201 Hz) corresponding well with the values found for the smaller cyclic homologues 40 and 41 (Figure 5.28).

7.7975 7.3782 7.0695 7.0139 6.9262 2.7269 2.5880 2.4630 2.3355 2.0412 1.5734 1.4484 1.3783 1.3600 1.1396 0.9591 0.7558

CH3-Bu

H-aryl-dppp β,γ-CH2-Bu α-CH2-Bu ß-CH2-dppp α-CH2-dppp

H-Th

8 7 6 5 4 3 2 1 0 (ppm)

Figure 5.27. 1H NMR spectrum of septithiophene-derived diplatinum macrocycle 43.

5.6 Results and discussion 245

-7.1713

JPtP = 2201 Hz

4 2 0 -2 -4 -6 -8 -10 -12 -14 -16 (ppm)

Figure 5.28. 31P NMR spectrum of septithiophene-derived diplatinum macrocycle 43.

The MALDI-TOF MS spectrum of platinacycle 43 was recorded with the spectrometer operated in linear modus. Thus, only an average mass signal without isotopic resolution could be obtained. However, in the mass spectrum of 43 a characteristic peak at m/z = 3356.4 is visible. This value corresponds very well to the theoretical molecular mass of the cyclic compound (m/z = 3355.1) (Figure 5.29).

3356 100

60 / a.u. I 40

0 2000 2500 3000 3500 4000 4500 5000 m / z

Figure 5.29. MALDI-TOF MS spectrum of septithiophene-derived diplatinum macrocycle 43. 246 Chapter 5 Conjugated macrocycles by metal template approach

Fascinated by the fact that this synthetic strategy generates dimeric cycles as the thermodynamically most stable products, an exciting objective seemed to be the preparation of trans-coordinated dimeric Pt-macrocycles by applying analogous reaction conditions, but using a trans-platinum complex as starting material. It was obvious that the possibility to form trans-coordinated dimeric cycles with shorter oligothiophenes did not exist due to the limited flexibility of the thiophene chains. Thus, first attempts to synthesize metallacycles with Pt-moieties in a trans-conformation were undertaken by starting with the diethynylated-quinquethiophene 28. After removal of the TMS-protecting groups, diethynyl-quinquethiophene 28 was reacted with trans-Pt(PPh3)2Cl2 under similar conditions as described above. Unfortunately, as indicated by NMR and MALDI-TOF MS analysis of the crude product the reaction afforded an untractable mixture, which was not further purified or characterized. However, upon reacting the seven-membered derivative, diethynylated septithiophene 42, with Pt(PPh3)2Cl2 under the established conditions, trans- configurated dinuclear platinacycle 44 was isolated as the only product in 98 % yield (Scheme 5.46). Metallacycle 44 was characterized by 1H NMR, 31P NMR and also by MALDI-TOF spectrometry.

Bu Bu Bu Bu Bu Bu Bu Bu

S S S S S (CH ) Si S S 3 3 Si(CH3)3

25 R = SiMe3 KOH/ MeOH/ THF 42 R = H

CuI (10 mol%)

NEt3 (2 equiv.) trans-Pt(PPh3)Cl2 toluene, r.t., 72h

Bu Bu Bu Bu PPh3 Pt S S Bu PPh3 Bu S S Bu Bu S S

S S

S S Bu Bu S S Bu S PPh3 S Bu Pt

Bu PPh3 Bu Bu Bu 44 [98 %]

Scheme 5.46. Synthesis of the metallacycle 44 having trans-bis(acetylide)-platinum units in the framework. 5.6 Results and discussion 247

Conclusive evidence of the cyclic structure was derived from the 1H NMR spectrum of 44, where no signal at around 3.5 ppm attributable to terminal acetylenic protons was observed (Figure 5.30). The spectrum of trans-configurated platinacycle 44 shows similar features to that of the cis-analogue 43. The signals of the phenyl groups that belong to the triphenylphosphine moieties appear as broad signals at δ = 7.79 and 7.36 ppm. In the aliphatic region of the spectrum of 44, similar multiplet signals for the butyl groups like for 43 are observed. The 31P NMR spectrum exhibits a single peak at δ = 18.7 ppm, which is consistent with other reported trans-configurated dialkynyl-Pt(II) complexes45. The chemical shift correlates also well with the value found for the previously described acyclic model compound trans-bis(thiophene alkynyl)-Pt(II) complex 38 (δ = 23.6 ppm). 1 In Figure 5.31, two symmetrical P-Pt satellites with a coupling constant of JP-Pt 3273 Hz are observed. Further support for the formation of the dimeric macrocycle 44 comes from its MALDI-TOF mass spectrum that due to the high molecular mass of the compound has been recorded in linear modus. The mass spectrum shows a characteristic peak at m/z = 3579.1 which is identical with the calculated molecular monoisotopic mass of the cyclic dimer (Figure 5.32).

7.7912 7.3688 7.2709 7.2343 7.1806 7.0840 7.0228 6.8820 2.7345 2.5880 2.5097 1.5696 1.4585 1.3751 1.0683 0.9723 0.9092

CH3-Bu

H-aryl-PPh3 β,γ-CH2-Bu

H-Th α-CH2-Bu

8 7 6 5 4 3 2 1 0 (ppm)

Figure 5.30. 1H NMR spectrum of trans-conformated septithiophene-derived diplatinum macrocycle 44.

248 Chapter 5 Conjugated macrocycles by metal template approach

18.7287

JPtP = 3293 Hz

40 35 30 25 20 15 10 5 0 -5 (ppm)

Figure 5.31. 31P NMR spectrum of trans-conformated septithiophene-derived diplatinum macrocycle 44.

3579 100

60 / a.u. I 40

0 1500 2000 2500 3000 3500 4000 4500 5000 5500 m / z

Figure 5.32. MALDI-TOF MS spectrum of trans-conformated septithiophene-derived diplatinum macrocycle 44.

5.6 Results and discussion 249

All these new oligothiophene-derived platinacycles 40, 41, 43 and 44 are stable microcrystalline solids with moderate solubilities in organic solvents. With increasing chain length of the oligothiophene incorporated in the cyclic frame the solubility increases. In halogenated organic solvents, such as dichloromethane, after a while decomposition of the compounds occurs. This observation is in agreement with the other reported.4 In Table 5.4, the above discussed analytical data of the novel cyclic and acyclic bis(oligothiophene alkynyl)-Pt(II) complexes of general formula L2Pt(≡-nT)2 (L = phosphine ligand, ≡-nT = ethynyloligothiophene-unit with n= chain length of the oligothiophene) are summarized.

Table 5.4. Analytical data of the novel cyclic and acyclic bis(oligothienyl-ethynyl)Pt(II) complexes of general formula L2Pt(≡-nT)2 (L = phosphine ligand, ≡-nT = ethynyloligothiophene-unit with n= chain length of the oligothiophene). 31 [a] Compound P NMR Molecular Mass yield 31 1 (P-P)Pt(mT)n P [ppm] JP-Pt [Hz] found calcd. [%]

35 dpppPt(1T)2 -6.80 2203 1044.5 1045.4 72

37 dpppPt(3T)2 -7.13 2198 1597.0 1597.6 75 [d] 39 cis-(PPh3)2Pt(1T)2 16.7 2245 1597.0 1597.6 -

38 trans-(PPh3)2Pt(1T)2 +19.7 2656 1156.7 1157.4 63

40 cyclo[dpppPt(3T)]2 -5.59 2243 2250.4 2250.7 91

41 cyclo[dpppPt(5T)]2 -6.57 2195 2803.3 2802.9 96

43 cyclo[dpppPt(7T)]2 -7.17 2201 3356.4 3355.1 97 [b] [c] 44 cyclo[trans-(PPh3)2Pt(7T)]2 +18.7 3273 3579.1 3579.1 98 [a]All values, except for diplatina macrocycle 44, represent the calculated and found (spectrometer operated in reflector modus) monoisotopical mass values; [b] average mass measured (operated in linear modus) by MALDI-TOF mass spectrometry; [c] calculated average mass; [d] compound not isolated; data determined in DMSO-d6 from a mixture of 39 with 38.

5.6.2.2 Macrocyclic oligothiophene-diacetylenes by reductive elimination from platinum metallacycles

The newly developed method, namely the reductive elimination from platinum centers under simultaneous C-C bond formation by iodine was applied for the synthesis of macrocyclic diacetylene-oligothiophenes. For this purpose, terthiophene-derived dinuclear platinacycle 36 was reacted with two equivalents of iodine (one per Pt-corner) in THF. 100 % conversion of the starting material was obtained at an optimum temperature of 60°C 250 Chapter 5 Conjugated macrocycles by metal template approach

after 24 hours (Scheme 5.47). After chromatographic separation of the polymeric material and by-product Pt(dppp)I2 the desired cyclodimeric terthiophene-diacetylene 45 was isolated in 54 % yield as a stable red microcrystalline solid. Macrocycle 45 represents the smallest 26-membered macrocycle in the homologous series which, never could be detected in the previous random cyclooligomerization reaction (Chapter 4.4.1). As indicated by analytical HPLC, under these conditions none of the higher homologue of the macrocyclic terthiophenediacetylene series 27a-h was formed (Figure 5.33). The reaction afforded, besides polymeric material, under preservation of the cyclic structure only cyclodimer 45.

Bu Bu Bu Bu Bu Bu Bu Bu S S S S S S Ph Ph Ph Ph Ph Ph P I P P I2 (2 equiv.) Pt Pt + Pt P I P THF, 60°C, 24h P Ph Ph Ph Ph Ph Ph S S S S S S Bu Bu Bu Bu Bu Bu Bu Bu 40 45 [54 %]

Scheme 5.47. Synthesis of the cyclodimeric terthiophene-diacetylene 45.

λ= 410nm, 20nm 1250 hexane/dichloromethane: 83/17

1000

750 mAU 500

250

0 0 2 4 6 8 10 12 14 16 18 20 minutes

Figure 5.33. Analytical HPLC chromatogram of the product obtained by reductive elimination reaction from terthiophene-derived diplatinum macrocycle 40.

5.6 Results and discussion 251

Due to its highly symmetrical structure, 1H and 13C NMR spectra of 45 consist of only few signals which are directly comparable to those of the corresponding higher members in the 1 cyclic terthiophene-diacetylenes series C[3T-DA]n. The H NMR spectrum of 45 is depicted in Figure 5.34. The cyclic structure was first proven by the lack of signals which is characteristic for terminal acetylenes at around δ = 3.5 ppm. In the aromatic region of the spectrum, there is only one signal at δ = 6.84 ppm which belongs to the protons of the unsubstituted thiophene rings (H-3’,4’-Th). In the aliphatic region, two sets of triplets at around δ = 2.56 ppm and δ = 2.44 ppm that belong to the α-CH2 of the butyl chains are visible, while the other protons of the butyl chains appear as multiplets at δ = 1.54 ppm and δ = 0.96 ppm. The 13C spectrum of cyclodimer 45 (Figure 5.35) shows the same signal pattern than its higher cyclic homologues C[3T-DA]n that were presented in Chapter 4. In the aliphatic region four groups of signals between δ = 13-33 ppm corresponding to the C- carbons of butyl chains are visible. The signals of the acetylenic carbon atoms appear at δ = 85.9 and δ = 86.9 ppm. Due to the high symmetry of the molecule in the aromatic region of the 13C spectrum only six signals corresponding to the terthiophene moiety are present.

6.8412 2.5801 2.5621 2.5409 2.4642 2.4450 2.4254 1.5462 1.3830 1.3641 0.9676 0.9616 0.9433

CH3-Bu

H-3‘,4‘

β,γ-CH2-Bu

α-CH2-Bu

488 32 24 7 6 5 4 3 2 1 0 (ppm)

Figure 5.34. 1H NMR spectrum of cyclodimeric terthiophene-diacetylene 45. 252 Chapter 5 Conjugated macrocycles by metal template approach

145.1256 137.9011 137.5465 136.1393 123.5426 117.9855 86.8691 85.8920 32.3965 32.3022 28.8541 27.5638 22.9462 22.6444 13.9108 13.8655

CH3-Bu γ-Bu β-Bu C-4,4‘‘-Th α-Bu C-2‘,5‘-Th C-2,3-Th C-2‘‘,3‘‘-Th C-3‘,4‘-Th

C-5,5‘‘-Th C≡C

140 120 100 80 60 40 20 0 (ppm)

Figure 5.35. 13C NMR spectrum of cyclodimeric terthiophene-diacetylene 45.

A notable feature that can be observed by comparing the whole series of C[3T-DA]n macrocycles (Figure 5.36) is that the signal of the unsubstituted H-3’,4’ protons of the thiophene rings for the cyclodimer 45 is remarkably shifted to higher field. In Table 5.5, the relevant NMR data of the terthiophene-diacetylenes macrocycles C[3T-DA]n (n=2-6) are summarized.

Bu Bu 4' 3' 5'' S Bu S Bu

Bu S Bu 5 S

S Bu S Bu

S Bu Bu S n-2

Bu Bu

45 (n=2) C[3T-DA]n 27a-h (n = 3-10)

Figure 5.36. Homologues series of terthiophene-diacetylenes macrocycle. 5.6 Results and discussion 253

Table 5.5. NMR data of macrocyclic terthiophene-diacetylenes C(3T-DA]n 45 (n= 2), 27a-d (n = 3-6) and of the precursor diethynyl-terthiophene 26.

H-Th Th-C≡C C≡C-C C-2’5’-Th C-5,5’’-Th compound [ppm] [ppm] [ppm] [ppm] [ppm]

45 C[3T-DA]2 6.84 86.9 85.9 145.1 117.9

27a C[3T-DA]3 7.07 81.2 78.7 149.6 117.2

27b C[3T-DA]4 7.08 81.6 78.9 150.8 117.0

27c C[3T-DA]5 7.09 81.6 77.9 151.1 116.8

27d C[3T-DA]6 7.09 81.9 77.9 151.3 116.7 26[a] 7.06 84.1[b] 76.9[b] 149.6 116.1 [a]NMR data of 26 taken from literature41; [b]the signals correspond to a Th-C≡C-H moiety.

As can be seen, for the large homologues (n = 3-6) the above mentioned signal appears at δ= 7.07 -7.09 ppm. This value corresponds well with that of the diethynylated terthiophene precursor 26. In contrast, the signal of H-3’,4’-Th for the smallest 26-membered macrocycle 45 appears at δ = 6.84 ppm indicating a paratropic ring current in the macrocycle. This characteristic high-field shift indicates anti-aromaticity comparable to 4nπ-annulenes. In fact, the conjugation path in cyclodimeric terthiophenediacetylene 45 corresponds to a 32 π-electron system. By comparing the 13C NMR spectra of the cyclic terthiophene-diacetylenes 27a-d that were presented in Chapter 4 significant shifts were observed for the carbon signals belonging to the unsubstituted inner thiophenes and those of the butadiyne units. This was attributed to the ring strain that becomes larger as the ring- size of the macrocycles gets smaller. However, the changes in the chemical shifts were not as pronounced as is the case for cyclodimer 45 which apparently has the highest intrinstic ring strain. For comparison, the relevant chemical shifts in the 13C NMR spectra of the macrocycles 27a-d and 45 are summarized in Table 5.5. In the spectrum of the smallest homologue 45, the C-2’,5’ carbons of the inner thiophene ring are approximately 5 ppm shifted to higher field. The peak appears at δ = 145.1 ppm, whereas in the spectrum of the higher homologues 27a-d appears at around δ = 150 ppm. This value is also consistent with that observed for the C-2’,5’ carbons of precursor 26. Remarkable are also the shifts of the acetylenic carbon atoms. The signal from the Th-C≡C carbon atoms which are directly connected to the thiophenes are strongly shifted to lower field for the cyclodimer 45 (δ = 86.9 ppm). The corresponding signal for the higher homologues appears at around δ = 81 ppm. A similar strong shift, but to higher field, can be observed for the C≡C-C inner 254 Chapter 5 Conjugated macrocycles by metal template approach

carbon of the butadiyne units. For cyclodimer 45 the above mentioned signal is observed at δ = 85.9 ppm, whereas for the larger macrocycles 27a-d at δ = 78 ± 1 ppm. The MALDI-TOF mass spectrum of cyclodimeric 45 recorded in a dithranol matrix shows characteristic signal at m/z = 1036.6 which corresponds very well to the theoretical mass calculated for the compound (m/z = 1036.7). The mass spectrum of cyclodimer 45 is displayed in Figure 5.37. The experimental isotopic distribution was compared to that calculated and was found to be in very good agreement (Figure 5.38). In the high resolution ESI-FTIR mass spectrum a single signal at m/z = 1036.43 was visible corresponding exactly to the calculated average mass for 45 (m/z = 1036.43). Furthermore, a detailed structure characterization by X-ray structure analysis of single crystals of the dimeric cycle 45 will be presented following in Chapter 5.8.

1036 100

60 / a.u. I 40

0 1000 1250 1500 1750 2000 m / z

Figure 5.37. MALDI-TOF mass spectrum of cyclodimeric terthiophene-diacetylenes 45.

5.6 Results and discussion 255

100 90 80 70 60 50 40 30 20 10 0 1.034 1.036 1.038 1.040 1.042 1.044

100

0 1034 1036 1038 1040 1042 1044 1046 m / z

Figure 5.38. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of cyclodimeric macrocycle 45.

The next cyclic homologue, quinquethiophene-derived diplatina macrocycle 41, was subjected to similar reaction conditions as described above (Scheme 5.48). In contrast to analogue 40, the reaction of platinacycle 41 with two equivalents of iodine afforded, as indicated by NMR, MALDI-TOF MS and HPLC analysis the formation of a mixture of macrocycles C[5T-DA]n in 17 % total yield, including the cyclodimer (n =2) , cyclotrimer (n=3) and the cyclotetramer (n= 4).

256 Chapter 5 Conjugated macrocycles by metal template approach

Bu Bu Bu Bu Bu Bu S S Ph Ph Ph Ph S S S P P Pt Pt P P S S Ph Ph S Ph Ph S S Bu Bu Bu Bu Bu Bu 41

2 I2 THF, 60°C, 24h

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S

n-1 Bu Bu Bu Bu

29a-c (n = 2-4) total yield [17 %] C[5T-DA] n

Scheme 5.48. Macrocyclic quinquethiophene-diacetylenes C[5T-DA]n by reductive elimination reaction.

In Figure 5.39 the 1H NMR spectrum of the product mixture is displayed. The cyclic structure of the products is proven by the lack of the terminal acetylenic proton resonance signal around δ = 3.5 ppm. The spectrum shows in the aromatic region only one set of signals at around δ = 7.1 ppm which corresponds to the protons of the unsubstituted thiophene rings. The other signals in the aliphatic region belong to the butyl chains attached to the thiophenes.

5.6 Results and discussion 257

7.0966 2.7231 1.6046 1.4705 1.5879 1.5724 1.4534 0.9957

β,γ-CH2-Bu

CH3-Bu

H-3’,3‘‘‘,4‘,4‘‘‘-Th α-CH2-Bu

7 6 5 4 3 2 1 0 (ppm)

Figure 5.39. 1H NMR spectrum of the product mixture consisting of cyclic quinquethiophene-diacetylenes

C[5T-DA]n.

The MALDI-TOF mass spectrum of the crude product mixture exhibits three signals which correspond to the mass of macrocycles C[5T-DA]n containing 2 to 4 quinquethiophene units linked by diacetylene-bridges ( Figure 5.40).

1588 100 (n=2)

80 2383 (n=3) 60 / a.u. I 40

20 3178 (n=4) 0 1500 2000 2500 3000 m / z

Figure 5.40. MALDI-TOF MS spectrum of the cyclic mixture C[5T-DA]n 29a-c obtained by reductive elimination from diplatina macrocycle 41. 258 Chapter 5 Conjugated macrocycles by metal template approach

The cyclic product mixture was characterized by analytical HPLC that was accomplished on a nucleosil-modified silica column by using hexane/dichloromethane (85/15) as eluent at a flow rate of 1.3 mL/min. According to the chromatogram of the product mixture depicted in Figure 5.41, the products formed in this reaction were the cyclodimer 29a, the cyclotrimer 29b and the cyclotetramer 29c in a 4:2:1 ratio. The separation of the individual macrocyles 29a-c from the mixture was carried out by preparative HPLC on a nucleosil- modified silica column under previously established conditions. By using a mixture of n- hexane/dichoromethane (83/17) as eluent at a flow rate of 80 mL/min the cyclodimer

C[5T-DA]2 could be isolated in 12 % yield as the main product, the cyclotrimer C[5T-

DA]3 in 4 % and cyclotetramer C[5T-DA]4 in less than 1 % yield. Each macrocycle 29a-c was characterized by analytical HPLC, NMR and MALDI-TOF MS. The analytical data are consistent with those given for the respective compounds in Chapter 4.

Figure 5.41. HPLC chromatogram of the product mixture C[5T-DA]n (n = 2-4) obtained by reductive elimination reaction from diplatina macrocycle 41.

In Table 5.6 the results obtained by the three different methods for the macrocyclization of diethynyl-quinquethiophene precursor 28 are summarized for comparison. The statistical

Glaser coupling reaction afforded a mixture of cyclic homologues C[5T-DA]n with n ranging from 2 to 6 in a total yield of 16 %. The yields given in Table 5.6 represent for all methods the total yield of the cyclic mixture C[5T-DA]n based on TMS-protected quinquethiophene precursor 24. As can be seen, similar yields were obtained by both metal template approaches. However, the Pt-based approach is somewhat superior to the Pd mediated macrocyclization. The yield of 16 % for the Pt-template approach includes both steps: platinacycle formation 41 and the reductive elimination reaction. However, even if the metal template approach did not show the expected improvements regarding the total 5.6 Results and discussion 259

yield of cyclic products, a significant advantage of this method, especially of the Pt- template approach, can be pointed out. By this method the cyclodimeric structure is prefentially formed, and macrocycle C[5T-DA]2 is obtained in significantly higher yields. The separation of this individual macrocycle 29a from the cyclic mixture is therefore extremely facilitated.

Table 5.6. Yields of macrocycles 29a-e obtained by Pt-based metal template approach, Pd-based metal template approach and statistical Glaser coupling reaction, respectively.

Pt template Pd template Glaser d C[5T-DA]n Molecular Mass[ ] approach[b] approach[c] coupling

total yield[a] 16 % 13 % 16 % found calcd.

29a (n=2) 11.5 3.8 3.0 1588.3 1588.6 29b (n=3) 3.8 4.0 8.8 2383.4 2382.9 29c (n=4) 0.8 3.5 1.5 3178.1 3177.3 29d (n=5) -- 1.7 0.2 3972.8 3971.6 29e (n=6) -- -- < 0.1 4770.9[e] 4771.9[f]

[a] Yield of the mixture of macrocycles determined after work-up without separation by HPLC; [b] yield based on precursor 24 including diplatinacycle formation and reductive elimination; [c] individual macrocycle not isolated; yields calculated by integration of peak area in analytical HPLC chromatogram; [d] calculated and measured monoisotopical mass values of the compounds 29a-d; [e] average mass measured by MALDI-TOF MS; [f] calculated average mass.

In close analogy, reaction of the largest platinacycle 43 in the series with iodine lead to the formation of a similar mixture of macrocycles C[7T-DA]n in 22 % total yield (Scheme 5.49). Evidence for the cyclic structure of the products was given by lack of terminal acetylenic protons at δ = 3.5 ppm in the 1H NMR spectrum of the product mixture. The peaks in the MALDI-TOF MS spectrum illustrated in Figure 5.42 correspond to the mass of the dimeric macrocycle C[7T-DA]2 46a and the cyclotrimer C[7T-DA]3 46b. Analytical HPLC was carried out on a nucleosil-modified silica column by using hexane/dichloromethane (78/22) as eluent at a flow rate of 1.3 mL/min (Figure 5.43). The

HPLC chromatogram exhibits two main peaks at retention times tR = 5.5 min and tR = 6.7 min which by correlation with the MALDI-TOF MS data are attributed to the cyclodimer

C[7T-DA]2 46a and the cyclotrimer C[7T-DA]3 46b. According to the integrated area of the peaks the two macrocycle 46a and 46b were formed in a 2:1 ratio. In the HPLC chromatogram an additional peak at tR = 10.6 min is seen that is attributed to the 260 Chapter 5 Conjugated macrocycles by metal template approach

cyclotetrameric species 46c. However, due to the poor overall amount of material obtained from this reaction, no further efforts regarding the separation of the individual compounds from the mixture was undertaken.

Bu Bu Bu Bu Bu Bu Bu Bu S S S S S Ph Ph S S Ph Ph P P Pt Pt P P Ph Ph S S Ph Ph S S S S Bu S Bu Bu Bu Bu Bu Bu Bu

2 I2 THF, 60°C, 24h

Bu Bu Bu Bu

Bu S S Bu S S Bu Bu S S

S S

S S Bu Bu S S Bu S S Bu n-1 Bu Bu Bu Bu

46a-c (n=2-4) total yield [22 %]

C[7T-DA] n

Scheme 5.49. Macrocyclic septithiophene-diacetylenes C[7T-DA]n by reductive elimination reaction.

5.6 Results and discussion 261

2145 100 (n=2)

60 / a.u. I 40

20 3217 (n=3)

0 1000 1500 2000 2500 3000 3500 4000 m / z

Figure 5.42. MALDI-TOF MS spectrum of the cyclic mixture C[7T-DA]n 46a,b obtained by reductive elimination from diplatina macrocycle 45.

Figure 5.43. HPLC chromatogram of the product mixture C[7T-DA]n ( n = 2-4) obtained by reductive elimination reaction from diplatina macrocycle 45.

The reasons for the different results (lower yields and the less selectivity) obtained for the reductive elimination reactions from metallacycles having longer oligothiophenes in the framework could not been clarified in the frame of this work. Nevertheless, some 262 Chapter 5 Conjugated macrocycles by metal template approach

investigations in order to gain more insight to the mechanism of the reductive elimination process have been carried out. The findings resulting from these studies will be discussed in detail in the following chapter.

5.6.3 Mechanistic investigations on the reductive elimination

Reductive eliminations from organic transition-metal compounds under simultaneous C-C bond formation is a reaction sequence that plays a crucial role in many synthetic organic reactions catalyzed by transition metals, such as homo- and cross-coupling reactions and polymerization.25 Investigations concerning the thermal activation of these processes have been reported by many groups.27,34,37 As described earlier, attempts to induce a reductive elimination from the acyclic bis(thiophene alkynyl)Pt(II) complex 35 by thermal activation in solution completely failed. Nevertheless, the thermal behaviour of the novel oligothiophene-linked σ-alkynyl platinum complexes was of great interest. For the following studies, the above mentioned linear compound 35 and terthiophene-derived dinucler platinacycle 40 were chosen as representatives of this novel class of compounds (Figure 5.44).

Bu Bu Bu Bu Bu Bu S S Ph Ph Ph S Ph S Ph Ph P P P Pt Pt Pt P S P Ph P Ph Ph Ph Ph Ph S Bu S Bu S Bu Bu 35 Bu Bu 40

Figure 5.44. Bis(thiophene alkynyl)Pt(II) complex 35 and bis(terthiophene alkynyl)diplatina macrocycle 40.

The thermal decomposition of complexes 35 and 40 were examined by thermal gravimetry analysis (TGA) and differential scanning calorimetry (DSC) measurements. Both compounds exhibited an unexpected high thermal stability as evidenced by TGA thermal profiles. The decomposition of acyclic Pt(II) complex 35 starts at T = 303°C, while the platinacycle 40 decomposes 25° higher. In the DSC profile of 35, first an endothermic 5.6 Results and discussion 263

change at T = 127°C due to melting of the complexes was observed followed by an exothermic process at T = 303°C as probably elimination occurs. Metallacycle 40 does not melt at this temperature. Thus, the DSC profile of this compound shows only one exothermic peak that can be attributed to the decomposition process. In Figure 5.45 the TGA and DSC thermal profiles of metallacycle 40 are depicted as examples.

a) b) 110

100 T = 301°C onset 90 N - atmosphere 2 80 O - atmosphere 70 2 Heat flow Heat

60 T = 324°C decomp 50 50 100 150 200 250 300 350 Mass change [%] change Mass 40 T [°C] 30

20 0 100 200 300 400 500 600 700 800 T [°C]

Figure 5.45. a) TGA and b) DSC thermal profiles of platinacycle 40.

One reason for the stability of both compounds might be due to the often stated stabilizing effect of the dppp moiety, chelating ligands making a complex more inert toward reductive elimination. The addition of free phosphines has been found to enhance reductive elimination from Pt(II) complexes when the process proceeds via a five-coordinated intermediate. Thus, the effect of added free PPh3 as a two-electron donor has been examined. Therefore, intimate mixtures of the two Pt derivatives 35 and 40 and additional

PPh3 were thermolyzed under similar conditions as previously explained for the pure complexes. As shown in Table 5.7, the decomposition of the mixtures is facilitated in every case. The exothermic processes in the presence of free PPh3 were accomplished at temperatures at which the pure complexes are stable (∆T ca. 40°C). In Figure 5.46, the

DSC thermal profile of the mixture containing PPh3 and platinacycle 40 is illustrated. By comparing this with the DSC thermal profile of the pure compound (see Figure 5.45) a significant shift in the decomposition temperature to lower values is observed. An additional endothermic peak at T = 62°C is visible corresponding to the melting point of the PPh3 ligand.

264 Chapter 5 Conjugated macrocycles by metal template approach

Table 5.7. DSC data for the decomposition of 35 and 40.

Temperature of Temperature of decomposition ∆T decomposition by addition of PPh3

35 303°C 266°C 37°C

40 324°C 283°C 41°C

PPh 3 T = 267 °C onset Heat Flow Heat

T = 283 °C decomp

50 100 150 200 250 300 350 T [°C]

Figure 5.46. DSC thermal profile of the mixture containing platinacycle 40 and free PPh3.

These results strongly support a mechanism in which the elimination reaction proceeds via a five-coordinated intermediate. The addition of free ligand increases the electron occupation at the metal center and hence promotes the reductive elimination. Another notable effect that the added phosphine might have caused, is to promote the process by stabilizing the Pt(0) product subsequent to reductive elimination. An interesting remark at this point is that the bisaryl-Pt(II) complexes investigated by Braterman29 bearing similar chelating phosphine ligands show higher lability than bis(oligothiophene alkynyl)Pt(II) complexes 35 and 40. They decomposition occur at around T = 235°C. These results do not support the proposal of Low,34 who proposed a facilitated reductive elimination from the platinum center for complexes with alkynyl groups.

In an attempt to get further information about the reaction mechanism, NMR investigations were carried out. Therefore, the reaction of bis(thiophene alkynyl)Pt(II) complex 35 with 5.6 Results and discussion 265

iodine that quantitatively generated the coupling product butadiyne 15 has been monitored by 1H NMR spectroscopy (Scheme 5.50).

Ph Ph Bu Bu Ph Ph P P 5 Bu Bu Ph Pt THF-d P Bu Ph I I Bu + 2 + Pt I rt S S P Bu Ph S Bu 15 Ph 35 S

Scheme 5.50. Reaction investigated by 1H NMR spectroscopy.

The reaction has been carried out in THF-d5 at room temperature. The spectra recorded after certain time intervals are given in Figure 5.47. Reductive elimination from Pt(II) complexes 35 was accompanied by a partial precipitation of a white solid which has been 1 identified as the follow-up product Pt(dppp)I2. The H NMR spectrum at t = 0 min corresponds to the bis(thienyl-ethynyl)platinum complex 35 (Figure 5.46). As previously described, the two multiplets at δ = 7.81 ppm and δ = 7.34 ppm belong to the phenyl groups of the dppp moiety, while the singulet at δ = 6.40 ppm belongs to the H-5 protons of the thiophenes. The protons of the CH2-protons of the dppp moiety appear as broad signals at 2.56 ppm and 2.00 ppm, respectively. The other signals in the aliphatic region are assigned to the protons of the butyl chains. The terminal methyl groups of the butyl chains are seen as two triplets at δ = 0.90 ppm and δ = 0.72 ppm. After 4 minutes reaction time the spectrum is remarkably changed. The aromatic protons of the dppp moiety appear as broad signals at δ = 7.83 ppm and δ = 7.43 ppm. The signal at δ = 6.40 ppm disappeared and instead a new signal at δ = 7.06 and two small ones at δ = 6.93 ppm and δ = 6.57 ppm became apparent. All aliphatic protons corresponding to the butyl chains are shifted to lower field. Interestingly, the multiplet observed at δ = 2.56 for the protons of the α-CH2 of the butyl groups of the Pt(II) complex 35 is splited in two group of signals that are observed at δ = 2.71 ppm and δ = 2.54 ppm. In contrast, the two triplets observed for the terminal methyl groups of the butyl chains of 35 appear after 4 minutes reaction time as a set of signals at δ = 0.96 ppm. Striking is also the decrease of the signal intensity of the 1 aliphatic CH2-protons of the dppp moiety. The H NMR spectra recorded after 25 minutes and 90 minutes reaction time are almost similar indicating a rapid conversion of Pt(II) complex 35 to butadiyne 15. The spectrum recorded after 90 minutes can be fully assigned to the coupling product 15. In the aromatic region of both spectra, only the signal 266 Chapter 5 Conjugated macrocycles by metal template approach

corresponding to the H-5 proton of coupling product 15 and, since the follow-up product

PtdpppI2 is only slightly soluble in THF, to the aromatic protons of the dppp ligand of this complex are visible. Compared to the H-5 thiophene-proton signal of the Pt(II) complex 35, the signal in the spectrum of the coupling product 15 is strongly shifted to lower field

(from δ = 6.40 ppm to δ = 7.06 ppm). The α-CH2 protons of the butyl groups appear as two sets of triplets, while the CH3- protons as one triplet at δ = 0.96 ppm.

H-aryl-dppp

ß-CH2-dppp CH3-Bu α-CH2-dppp 0 min H-5-Th α-CH2-Bu β,γ-CH2-Bu

CH3-Bu

4 min

H-5-Th β,γ-CH2-Bu α-CH2-Bu

CH3-Bu H-aryl-dppp

25 min β,γ-CH2-Bu H-5-Th α-CH2-Bu

CH3-Bu

90 min

β,γ-CH2-Bu H-aryl-dppp α-CH2-Bu

8.0 7.5 7.0 6.5 2.0 1.0 0.0 (ppm)

Figure 5.47. 1H NMR investigation on the reductive elimination reaction from bis(thiophene alkynyl)-Pt(II) complex 35 with iodine. 5.6 Results and discussion 267

The reductive elimination reaction from Pt(II) complex 35 was also monitored by 31P NMR spectroscopy. The spectra recorded at room temperature in THF-d5 after the given reaction times are illustrated in Figure 5.48. Prior to elimination, in the spectrum of 35 (t = 0 min) only one phosphorus signal at δ = -7.63 ppm accompanied by two symmetrical 1 platinum satellites with a coupling constant of JPtP = 2188 Hz are visible. The spectrum recorded after 4 min reaction time contains, beside the above signals, one small signal at δ 2 = - 11. 7 ppm and two closely spaced doublets with a coupling constant of JPP = 28.7 Hz at around δ = -11.5 ppm and δ = -16.7 ppm (see also Figure 5.49). After 10 minutes the 31P NMR spectrum shows only a singlet 31P resonance at δ = -11.7 ppm with a coupling 1 constant of JPtP 3230 Hz. which is assigned to the follow-up product PtdpppI2. The spectrum after 25 minutes is similar to that recorded after 10 minutes supporting the assumption for a fast elimination process from the platinum center. Both, the 1H and the 31P NMR spectra recorded after 4 minutes reaction time indicate the formation of an unsymmetrical intermediate in course of the elimination reaction. At t = 4 minutes, in the 1H NMR spectrum, two signals that could not be assigned were observed at δ = 6.94 ppm and δ = 6.57 ppm. Analogously, in the 31P NMR spectrum at t = 4 minutes two doublets at δ = - 11.5 ppm and δ = -16.7 ppm became apparent. These highly deshielded resonances are indicative for a Pt(IV) complex structure in which the phosphorus atoms of the dppp ligand are not equivalent. In agreement with other reports,46 the doublet observed at higher field can be assigned to a phosphorus atom trans to iodide, whereas the other doublet at lower field is assigned trans to the organic moiety. The structure resulting from such an arrangement can be assigned to a Pt(IV) complex in which the organic groups and halides are mutually cis-oriented (see illustrated structure in Figure 5.50).

268 Chapter 5 Conjugated macrocycles by metal template approach

-7.63

JPtP = 2188 Hz 0 min -11.43 -11.60 -11.70 -7.63 -16.68 -16.85

JPtP = 2188 Hz 4 min -11.70

JPtP = 3230 Hz

10 min -11.70

JPtP = 3230 Hz

25 min

10 0 -10 -20 (ppm)

Figure 5.48. 31P NMR investigation on the iodine mediated reductive elimination reaction from bis(thiophene alkynyl)-Pt(II) complex 35 .

5.6 Results and discussion 269

-11.70 -11.43 -11.60 -16.69 -16.86

JPP = 28.7 Hz JPP = 28.7 Hz

-10.5 -12.0 -13.5 -15.0 -16.5 (ppm)

Figure 5.49. 31P NMR spectrum of the intermediate formed by reductive elimination from Pt(ÍI) complex 35 after 4 min reaction time.

Ph Ph I P I Pt P Ph Ph S

Bu Bu Bu S

Figure 5.50. Proposed structure for the Pt(IV) intermediate with organic groups and halides mutually cis- oriented.

Halogens are known to add stereospecifically trans to Pt(II) substrates, but on the above time scale no Pt(IV) intermediate resulting from trans-oxidative addition of iodine could be detected. Such an isomerization process is proven to occur via a five-coordinated intermediate by prior ligand dissociation. Consequently, for the elimination process under simultaneously C-C coupling a similar mechanism based on prior ligand dissociation is suggested.

On the basis of these results a detailed reaction mechanism for reductive elimination from acyclic Pt(II) complex 35 by reaction with iodine to butadiyne 15 is illustrated in Scheme 5.51. 270 Chapter 5 Conjugated macrocycles by metal template approach

Bu Bu Bu Bu Ph Ph Ph Ph I S P + I2 P S Pt Pt P P Ph S S Ph Ph Ph I Bu Bu Bu Bu 35 A trans-isomer

-I- + I -

Bu Bu Ph I Bu Ph Bu Ph Ph I P P S Pt Pt P + S P + Ph Ph Ph Ph S Bu B Bu Bu S D

Bu - + I -I - + I -

Ph Ph I Bu Bu P I Bu Bu Pt P Ph Ph S S S 15 Bu Bu Bu S + Bu Bu Bu

C cis-isomer I S E +

PtdpppI 2

Scheme 5.51. Proposed mechanism for the reductive elimination reaction of 35.

Firstly, iodine is oxidatively added in a trans fashion to the Pt(II) center of 35 to form the octahedral complex A. From complex A iodide dissociates under formation of the five- coordinated square-pyramidal intermediate B. The reductive elimination was shown to proceed from a trigonal-bipyramidal structure. As shown in Scheme 5.51 the trigonal- bipyramidal intermediate D, which readily eliminates affording butadiyne 15 and iodoethynylthiophene E, is easily accessible by fluctuation from the primarily formed square-pyramidal structure B. Competitive to reductive elimination reaction, iodine re- association to the five-coordinated intermediate B can occur generating the corresponding cis-isomer C. The dissociation-recombination processes of iodine and thus the isomerization are known to be fast in comparison to the reductive elimination. 5.6 Results and discussion 271

Furthermore, isomers in which the halogenides are mutually cis-oriented have been proven to be thermodynamically favoured over the corresponding trans-isomers. Consequently, the slower is the process of reductive elimination, the higher the ratio of the cis-isomer in equilibrium is. Taking a closer look at the possible pathways by which iodine can dissociate from cis- isomer C, it becomes obvious that the five-coordinated species B is not the only product that can result. As shown in Scheme 5.52, depending on which iodide dissociates, the axial or the equatorial one, two different square-pyramidal structures B1 and B2 can be formed. From B1 and B2, the reductive elimination to butadiyne 15 can occur via the trigonal- bipyramidal intermediate D1. However, square-pyramidal intermediate B2 can also be transformed to trigonal-bipyramidal D2 which is proposed even to be favoured due to less sterical repulsion of the trans-oriented alkyl-groups on the thiophene moieties. If reductive elimination occurs from the Pt-intermediate D2, only iodoethynylthiophene E would be formed, since elimination exclusively takes place from the cis-oriented positions. Alkynyl iodides are known to be reactive substrates in transition metal-catalyzed reactions. In the reductive elimination reaction described above as follow-up product an unsaturated and very reactive low-valent metal [Pt(II)] species is produced. According to the mechanism of most metal-promoted coupling reactions, this species can easily convert alkynyl iodides by means of oxidative addition/ reductive elimination processes to corresponding butadiynes. In case of Pt(II) complex 35, the butadiyne resulting from alkynyl iodide E is similar to the coupling product 15 which is obtained by the direct reductive elimination step. This is probably the reason why in the case of linear Pt(II) complexes the reaction with iodine led to the desired butadiynes as solely products. This is not the case for oligothiophene-derived dinuclear platinacycles that were investigated in this work. Since not only one, but two metal and thus two reaction centers are involved, the situation is more complicated. It is believed that the formation of higher cyclic homologues or polymeric material is attributable to the formation of oligothiophene alkynyl iodides as intermediates which promoted by unsaturated low-valent metal Pt(II) complexes subsequently can undergo oxidative addition/reductive elimination reactions.

272 Chapter 5 Conjugated macrocycles by metal template approach

Ph Ph I P I Pt P Ph Ph S

Bu Bu Bu S

Bu Ph Ph I + C P Ph Ph Pt P + I cis-isomer P Pt Ph Ph S P Ph Ph S Bu Bu Bu S Bu Bu Bu S Bu

Bu B2 B1 Bu

Ph I Bu Bu S Ph + Bu P Pt + P S Ph Ph Ph Ph P D2 Pt I P D1 Ph Ph Bu S

Bu Bu S

Bu Bu Bu Bu

S S Bu 15 Bu + I S Bu E Bu I S E

Scheme 5.52. Proposed decomposition pathway from the cis-isomer of an octahedral Pt(IV) complex.

The reductive elimination from terthiophene containing diplatinacycle 40 yielded the cyclic dimer 45 in 54 %. No higher cyclic homologue but polymeric material was formed. In contrast, reaction of both, quinquethiophene-derived platinacycle 41 and septithiophene- derived platinacycle 43 with iodine preferentially yielded the cyclodimer, but also higher cyclic homologues. The total yields of cyclic products were in case of the longer oligothiophenes lower than for the terthiophene containing macrocycle. The major difference between the three platinacycles 40, 41 and 43 is the chain length of the bridging oligothiophene unit. This has a direct impact on the ring strain energy of the respective 5.6 Results and discussion 273

targeted macrocycle. From this point of view the most difficult accessible macrocycle in the series is the terthiophene derived compound. Another factor that has to be taken in account when comparing the results obtained from the reductive elimination is the electron donor property of the corresponding bridging oligothiophene moiety. In a homologues series of oligothiophenes, the electron donor ability of the molecule increases progressively with the increasing chain length of the oligomer. As mentioned earlier, an increase in the electron density on the metal atom stabilizes the five-coordinated intermediate with respect to reductive elimination. Considering the higher donor strength of the quinquethiophene and septithiophene moieties compared to that of the terthiophene one, the Pt-C bond in macrocycles 41 and 43 are probably more resistant to cleavage than that in platinacycle 40. According to the proposed mechanism depicted in Scheme 5.51, the more difficult the elimination of ethynyl-moieties from the metal center is the higher is the ratio of the product formed by C-I coupling. High amount of oligothiophene alkynyl iodides leads in the case of the cyclic Pt(II) complexes to enhanced formation of polymeric material. This might be the explanation why the reductive elimination from platinacycles 41 and 43 containing longer oligothiophenes than terthiophene yielded more polymeric material and lower amounts of cyclic products. The higher conformational flexibility of longer linear oligothiophenes is certainly an additional factor that can be accounted to the increased formation of polymeric material.

Another aspect worth to discuss is the effect that substituents attached to the thiophene rings might have on the reductive elimination. Since the three platinacycles 40, 41 and 43 have the same substitution pattern, namely butyl groups attached to the outer positions of the terminal thiophene rings, the following considerations are valid for all of them. According to the proposed mechanism, iodine adds to the Pt(II)-center under formation of a octahedral complex A. The reductive elimination occurs from a trigonal-bipyramidal conformation which is formed by iodide dissociation to B and fluctuation to D. In an octahedral complex the angles between the substituents in plane are 90°, whereas in the trigonal-bipyramidal geometry it is 120°. Thus, if the oligothiophene such as the platinacycles investigated in this work, have bulky substituents in the positions next to the metal, the dissociation of iodine should be facilitated due to the sterical relaxation. At the same time, the recombination of iodine with the five-coordinated intermediate which generates cis-isomer C should be impeded due to sterical hindrance.

274 Chapter 5 Conjugated macrocycles by metal template approach

In order to investigate the ability of oligothiophene alkynyl Pt(II) complexes to undergo reductive elimination by simple electron transfer, cyclovoltammetric measurements were performed. The cyclic voltammograms were recorded in dichloromethane with 0.1 M + - Bu4N PF6 as the supporting electrolyte and the potentials were determined versus the ferrocene/ferrocenium couple (Fc/Fc+) at scan rates of 100 mV/s. The cyclic voltammogram of acyclic bis(thiophene alkynyl)-Pt(II) complex 35 is displayed p + in Figure 5.51, left. The first oxidation process proceeds at E Ox1 = 0.47 V vs Fc/Fc and leads to an irreversible wave indicating a follow-up reaction. A second irreversible p + oxidation wave is observed at E Ox2 = 0.93 V Fc/Fc . The cyclic voltammogram of p + butadiyne 15 shows that the compound is irreversibly oxidized at E Ox1 = 0.92 V vs Fc/Fc (Figure 5.51, right). This value coincides with the second oxidation peak of Pt(II) complex 35 indicating the formation of 15 as the follow-up product after oxidation of 35.

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

7.5 0.92 7.5

4.5 4.5 6.0 6.0

3.0 3.0 4.5 4.5 0.93

I [µA] 0.47 I [µA] 3.0 3.0 1.5 1.5 1.5 1.5

0.0 0.0 0.0 0.0

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 -1.5 -1.0 -0.5 0.0 0.5 1.0 E [V] vs. Fc/Fc+ E [V] vs. Fc/Fc+

Figure 5.51. Electrochemical characterization of acyclic bis(thiophene alkynyl)-Pt(II) complex 35 (left) and + - -3 -1 -1 butadiyne 15 (in dichloromethane, 0.1 M Bu4N PF6 , c = 1 x 10 mol l , v = 100 mV s ).

In order to prove the above assumption electrolysis of acyclic Pt(II) complex 35 has been carried out in dichloromethane solution (5 x 10-3 M) at a constant potential (Scheme 5.53).

The constant potential applied in the electrolysis experiment was Eappl = 0.95 V (vs Ag/AgCl). The consumption of the substrate was monitored by cyclic voltammetry showing complete conversion after 12 h. After work-up, the product was analyzed by GC- MS revealing the formation of butadiyne 15 as the solely product.

5.6 Results and discussion 275

Bu Bu Ph Ph Bu Bu Bu P S - Bu Pt -e P Ph S E = 0.95 V S S Ph appl 15 Bu Bu dichloromethane 35

Scheme 5.53. Electrolysis by constant potential Eappl = 0.95 V (vs. Ag/AgCl) of acyclic bis(thiophene alkynyl)-Pt(II) complex 35.

Similar electrochemical investigations were further performed on terthiophene-derived platinacycle 40. The cyclic voltamogramm recorded in dichloromethane is displayed in p + Figure 5.52, left. The first oxidation wave (E Ox1 = 0.06 V vs Fc/Fc ) is shifted to lower value compared to the monothiophene containing complex 35 and is irreversible. The p + p second oxidation wave at E Ox2 = 0.44 V vs Fc/Fc and the third wave at E Ox3 = 0.57 V vs + p Fc/Fc are reversible. A fourth irreversible oxidation wave is observed at E Ox4 = 0.72 V vs Fc/Fc+. The oxidation of terthiophene-diacetylene macrocycle 45 was similarly performed in dichloromethane. In the cyclic voltammogram (Figure 5.52, right) two reversible oxidation p + p + waves at E Ox1 = 0.39 V vs Fc/Fc s and E Ox2 = 0.59 V vs Fc/Fc are visible. These values correspond to the second and third oxidation waves of diplatinacycle 40 indicating the formation of macrocycle 45 as the follow-up product of oxidized platinacycle 40.

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

2.5 36 2.5 2.5 41 2.5 2.0 2.0 2.0 2.0 0.72 1.5 1.5 1.5 0.57 1.5 0.44 1.0 0.59 1.0

I [µA] 0.39 I [µA] 1.0 0.06 1.0 0.5 0.5 0.5 0.5 0.0 0.0

0.0 0.0 -0.5 -0.5

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 + E [V] vs. Fc/Fc+ E [V] vs. Fc/Fc

Figure 5.52. Electrochemical characterization of 40 and 45. Reduction and oxidation in dichloromethane, + - -4 -1 -3 -1 -1 Bu4N PF6 (0.1 M), c = 1 x 10 mol l for 40 and c = 1 x 10 mol l for 45, v = 100 mV s .

276 Chapter 5 Conjugated macrocycles by metal template approach

The first peak at 0.06 V should correspond to the oxidation of the Pt(II) metal fragment. The resulting diminished electron density on the metal center facilitates the occurance of a reductive elimination. Remarkably, the oxidatively induced elimination proceeds under preservation of the cyclic structure. Electrolysis experiments of terthiophene-derived platinacycle 40 in dichloromethane turned out to be somewhat difficult due to the moderate solubility of the compound. Benzonitrile was found to be a better solvent for this purpose. Firstly, the cyclic voltammogram of platinacycle was recorded in benzonitrile (Figure 5.53). Interestingly, in p + this case the first irreversible oxidation proceeds at E Ox1 = 0.19 V vs Fc/Fc followed by a p + second irreversible process at E Ox2 = 0.34 V vs Fc/Fc . The reversible oxidation waves that are attributed to the follow-up product, the cyclic terthiophene-diacetylene 45, are p + p + visible at E Ox3 = 0.45 V vs Fc/Fc and E Ox4 = 0.66 V vs Fc/Fc . The slight shift of the peaks to more positive values should be a solvent effect. At this stage, it is not clear why the cyclic voltammogram recorded in dichloromethane shows solely one irreversible oxidation wave of the Pt-centre, whereas that recorded in benzonitrile exhibits two irreversible waves.

0.00 0.25 0.50 0.75 1.00

3.5 40 0.66 3.5 0.45 3.0 3.0

2.5 0.34 2.5 2.0 2.0

0.19

I [µA] 1.5 1.5

1.0 1.0

0.5 0.5

0.0 0.0

0.00 0.25 0.50 0.75 1.00 E [V] vs. Fc/Fc+

Figure 5.53. Electrochemical characterization of diplatina macrocycle 40. Reduction and oxidation in + - -4 -1 -1 benzonitrile, Bu4N PF6 (0.1 M), c = 1 x 10 mol l , v = 100 mV s .

Following these studies, preparative electrolysis of terthiophene-derived platinacycle 40 -4 + was carried out at constant potential in benzonitrile (2 x 10 M) with 0.1 M Bu4N PF6 as supporting electrolyte. In order the improve the solubility of the compound the reaction was carried out at 80°C. A constant cell potential of Eappl = 0.85 V (vs. Ag/AgCl) was applied and the consumption of the Pt-complex was monitored by cyclic voltammetry. 5.6 Results and discussion 277

Complete conversion was indicated by the disappearance of the irreversible oxidation waves at 0.19 V and 0.34 in the cyclic voltammogram and was achieved after 48 h. After work-up and chromatographic purification cyclodimeric terthiophene-diacetylene 45 could be isolated in 6 % yield. The structural proof was obtained by 1H NMR and MALDI-TOF MS analysis. The analytical data were fully consistent with those described previously for 45. As indicated by analytical HPLC and MALDI-TOF MS analysis the electrolysis did not generate higher cyclic homologues.

Bu Bu Bu Bu Bu Bu Bu Bu S S S S S S Ph Ph Ph Ph P P -e- Pt Pt P P Eappl = 0.85 V Ph Ph Ph Ph S benzonitrile S S S S S Bu Bu Bu Bu Bu Bu Bu Bu 40 45

Scheme 5.54. Electrolysis of terthiophene-derived diplatina macrocycle 40 at a constant potential Eappl = 0.85 V (vs. Ag/AgCl).

The modest yield most likely reflects the poor solubility of the platinacycle 40, which led to precipitation of the substrate and coverage of electrode surface during electrolysis. Furthermore, since the experiment was carried out by using a small working microelectrode with an active surface area less than 1 mm2, the reaction rate is slow and the time required for the completion of the reaction becomes very long. Therefore, the use of a typical electrochemical cell for standard electrolysis with electrodes having larger active surface areas would certainly significantly improve the yield of the reaction.

5.6.4 Synthesis of fully conjugated cyclo[n]thiophenes

Cyclo[n]thiophenes (n= number of thiophene rings) having solely α-linked thiophene units in the backbone represent a novel class of well-defined macrocycles with fully conjugated π-system peripheries and thus with particularly appealing characteristics and perspectives.47 Owing to their fascinating structural architectures, they are of special 278 Chapter 5 Conjugated macrocycles by metal template approach

interest due to their potential to function as ideal models for the corresponding polymers. Moreover, the rigid cyclic structures with full π-conjugation might enable these macrocycles to act as intriguing “molecular circuits” which would additionally include sites for selective recognition and selective complexation. The cyclic precursors for the design of cyclo[n]thiophenes are, as discussed in Chapter 4.2, macrocyclic oligothiophenediacetylenes C[mT-DA]n where m is the number of thiophenes in one subunit and n is the number of diethynyl-oligothiophene subunits. Reactions of these conjugated macrocycles with sulfide nucleophiles should convert the diacetylene units to the corresponding thiophene rings and thus enable the successful synthesis of the fully conjugated cyclo[n]thiophenes (see Figure 4.2). The ring closure reaction of oligothiophenel-diacetylenes with sulfide nucleophiles is a common method used in the synthesis of oligothiophenes.48 As stated in Chapter 2, in general, for unsubstituted thiophenes the reaction can be carried out in methanol, while conversion of the corresponding derivatives with substituents in positions next to the butadiyne units requires the use of higher boiling solvents. This is the result of sterical hindrance of the alkyl side chains on the adjacent diyne units. A homologues series up to the 11-mer of parent butyl-substituted linear oligothiophenes was reported by Krömer et al.49 As illustrated in Scheme 5.55, the synthesis succeeded by reacting the corresponding bis(oligothienyl)-1,3-butadiynes with sodium sulfide hydrate in a mixture of xylene and 2-methoxyethanol. The addition of xylene was necessary in order to provide enough solubility to the oligothiophene-diacetylenes.

Bu Bu Bu Bu Bu Bu Bu Bu

S S S n S S S n n = 0-2

Na2S x 9H2O methoxyethanol / xylene

Bu Bu Bu Bu Bu Bu Bu Bu

S S S S n S S S n

n = 0-2

Scheme 5.55. Synthesis of a homologues series of butyl-substituted linear oligothiophenes by Krömer et al. 5.6 Results and discussion 279

As mentioned in Chapter 4, the same reaction had enabled the successful preparation of cyclo[n]thiophenes (n = 12, 16, 18) (Scheme 5.56).41,47,50

Bu Bu Bu Bu

S Bu S Bu S Bu m S S Bu S Bu Bu Bu S S S Bu

Na2S x 9 H2O S S S

S S Bu S Bu S Bu Bu

m Bu S S Bu m S p S Bu S Bu n

Bu Bu Bu Bu

m = 1, n =1,2 C[12]T p = 1 [23%] m = 2, n =1 C[16]T p = 3 [7%] C[18]T p = 4 [27%] Scheme 5.56. Reported synthesis of cyclo[n]thiophenes.

On this basis, the synthesis of the macrocyclic oligothiophenes, was first carried out under the same reaction conditions. The reaction of cyclodimeric terthiophene-diacetylene 45 with high excess of sodium sulfide nonahydrate in xylene/2-methoxyethanol at 140°C afforded cyclo[8]thiophene 47 as the smallest homologues in the series of fully conjugated cyclo[n]thiophenes (Scheme 5.57). Macrocycle 47 was isolated after chromatographic work-up as a red stable microcrystalline solid in 19% yield. The characterization of the product by analytical HPLC was accomplished on a nucleosil-modified silica column by using hexane/ dichloromethane (79/21) as eluent at a flow rate of 1.3 mL/min (Figure 5.54).

Bu Bu Bu Bu Bu Bu S Bu S Bu S S S S

Na2S x 9 H2O S S xylene, 2-methoxyethanol 140°C, 24h S S S S S Bu S Bu Bu Bu Bu Bu Bu Bu 45 47 [19 %]

Scheme 5.57. Synthesis of the fully conjugated cyclo[8]thiophene 47. 280 Chapter 5 Conjugated macrocycles by metal template approach

λ= 410nm, 20nm 100 hexane/dichloromethane: 79/21

mAU 40

0 0 2 4 6 8 10 12 14 16 18 20 minutes

Figure 5.54. Analytical HPLC chromatogram of cyclo[8]thiophene 47 (nucleosil-modified silica column, eluent: hexane/ dichloromethane (79/21) at a flow rate of 1.3 mL/min).

Structural proof of cyclo[8]thiophene 47 came from 1H and 13C NMR spectra which due to the highly symmetrical structure of the compound consist of only few signals. 1H NMR of macrocycle 47 which is illustrated in Figure 5.55 reveals one signal at δ = 6.88 ppm corresponding to the aromatic protons of the unsubstituted thiophene rings. The triplet at δ

= 2.63 ppm belongs to the α-CH2 protons of the butyl groups and at δ = 0.98 ppm to the terminal CH3-protons. In the aliphatic region of the spectrum, further two sets of signals belonging to the other protons of the butyl groups are apparent. The 13C spectrum of the fully conjugated macrocycle 47 is depicted in Figure 5.56. Only eight signals, four in the aromatic region at δ = 122-140 ppm belonging to the aromatic carbon atoms and four in the aliphatic range of the spectrum at δ = 13-32 ppm corresponding of the C-atoms butyl chains are visible. The MALDI-TOF mass spectrum of cyclo[8]thiophene 47 exhibits only one signal at m/z = 1104.5 which very well corresponds to theoretical calculated molecular mass (m/z = 1104.4) (Figure 5.57) as well as the experimental isotopic distribution (Figure 5.58).

5.6 Results and discussion 281

6.8849 2.6440 2.6285 2.6117 1.6124 1.5819 1.5676 1.5334 1.5016 1.4724 1.4574 1.4288 0.9992 0.9849 0.9700

CH3-Bu

H-3‘,4‘-Th

β,γ-CH2-Bu

α-CH2-Bu

2 4 86 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 (ppm)

Figure 5.55. 1H NMR spectrum of cyclo[8]thiophene 47.

139.2319 137.7921 131.4666 122.7993 32.8225 27.9449 23.0268 13.9511

CH3-Bu γ-Bu

C-3,4-Th α-Bu C-2,2‘-Th β-Bu C-3‘,4‘-Th

140 120 100 80 60 40 20 0 (ppm)

Figure 5.56. 13C NMR spectrum of cyclo[8]thiophene 47.

282 Chapter 5 Conjugated macrocycles by metal template approach

1104 100

60 / a.u. I 40

0 800 900 1000 1100 1200 1300 1400 1500 m / z

Figure 5.57. MALDI-TOF MS spectrum of cyclo[8]thiophene 47.

g 100 90 80 70 60 50 40 30 20 10 0 1.102 1.104 1.106 1.108 1.110 1.112

100

0 1102 1104 1106 1108 1110 1112 1114 m / z

Figure 5.58. Calculated (top) and measured (bottom) isotropic distribution of the mass signal for cyclo[8]thiophene 47. 5.6 Results and discussion 283

The same reaction with sodium sulfide in xylene/2-methoxyethanol at 140°C was then applied for oligothiophene-diacetylene macrocycles 29a, 27b, and 29b. The reaction of the cyclodimeric quinquethiophene-diacetylene 29a yielded after purification by chromatography corresponding α-linked cyclo[12]thiophene 48 in 31 % yield (Scheme 5.57). In previous reports, in which as precursor the cyclotrimer terthiophene-diacetylene was employed in the same reaction, 48 was formed in only 23 % yield.47a This lower yield can be explained by the higher number of diacetylenes units that had to be converted to thiophenes.

Bu Bu Bu Bu Bu Bu Bu S S S S S Bu Bu S S S S Bu Bu Bu Na2S x 9 H2O S S S S Bu Bu xylene, 2-methoxyethanol S S 140°C, 24h S Bu S S S Bu Bu S S Bu Bu S Bu Bu Bu

29a Bu Bu 48 [31%] Scheme 5.57. Synthesis of fully conjugated cyclo[12]thiophene 48.

Analogous reaction of cyclotetrameric terthiophene-diacetylene 27b with sodium sulfide led to the formation of the corresponding fully conjugated cyclo[16]thiophene 49 in a moderate yield of 8% (Scheme 5.58). This value is consistent with that reported previously for the same macrocycle.47a,50

284 Chapter 5 Conjugated macrocycles by metal template approach

Bu Bu Bu Bu

S S Bu S S Bu

Bu S S Bu

Bu S S Bu Bu S S Bu S S

Bu Bu Bu Bu 27b

Na2S x 9 H2O xylene, 2-methoxyethanol 140°C, 24h

Bu Bu Bu Bu

S S S S Bu S Bu

S Bu S Bu

S S

Bu Bu S S

Bu Bu S S S S S

Bu Bu Bu Bu

49 [8 %]

Scheme 5.58. Synthesis of fully conjugated cyclo[16]thiophene 45.

The higher homologue cyclo[18]thiophene 50 was obtained from the corresponding precursor, cyclotrimeric quinquethiophene-diacetylene 29b, under similar reaction conditions with sodium sulfide in 26 % yield (Scheme 5.59). The 1H, 13C NMR spectra and MALDI-TOF MS analysis of the cyclo[n]thiophenes with n = 12, 16 and 18 were consistent with those reported previously.47a The corresponding MALDI-TOF MS spectra of macrocycles 48-50 are attached at the end of this chapter.

5.6 Results and discussion 285

Bu Bu

Bu Bu Bu S S S Bu S S

Bu Bu S Bu S Bu S S

Bu S S Bu Bu S S Bu S S

Bu Bu Bu Bu 29b

Na2S x 9 H2O xylene, 2-methoxyethanol 140°C, 24h

Bu Bu

Bu Bu S Bu S S Bu S S

S S Bu Bu

S S

Bu S S Bu

Bu S S Bu S S S

Bu Bu Bu Bu 50 [23 %]

Scheme 5.59. Synthesis of fully conjugated cyclo[18]thiophene 50.

Due to the high symmetry and similarity of the structures (the same oligothiophene units, only different ring sizes), the NMR spectra of the fully conjugated macrocycles cyclo[n]thiophene (n=8, 12, 16, 18) are very similar showing the same signal pattern that easily can be assigned. In Table 5.9, the NMR data of macrocycles 47-50 are summarized.

286 Chapter 5 Conjugated macrocycles by metal template approach

Table 5.8. NMR data of the fully conjugated cyclo[n]thiophenes 47-50.

47 48 49 50 Cyclo[n]thiophene n = 8 n = 12 n = 16 n = 18 H-3’,4’-Th 6.87 7.06 7.09 7.09

α-CH2-Bu 2.61 2.70 2.69 2.74

ß,γ-CH2-Bu 1.53 1.52 1.53 1.54 H NMR 1 1.47 1.42 1.45 1.47

CH3-Bu 0.97 0.95 0.95 0.98 C-3,4-Th 139.2 140.3 140.3 140.1 C-2,2’-Th 137.8 136.6 136.1 136.1 131.5 130.1 130.1 130.0 C-3’,4’-Th 122.8 125.4 125.6 125.7 ß-Bu 32.8 32.8 32.8 32.8 C NMR C NMR 13 a-Bu 27.9 27.8 27.9 27.9 γ-Bu 23.0 23.0 23.0 23.1

CH3-Bu 13.9 13.9 13.9 13.9

Within the macrocyclic cyclo[n]thiophene homologues series, the high-field shift of the H- 3’,4’-Th protons of the unsubstituted thiophenes observed for the smallest homologue 47 is remarkable compared to those of the higher homologues. For the cyclo[8]thiophene 47 the H-3’,4’-Th signal appear at δ = 6.89 ppm, while for the higher homologues 48-50 they appear between δ = 7.06-7.09 ppm. Such a shielding indicates the existence of a paratropic ring current in the macrocycle. A similar behaviour was also observed in the case of the cyclodimeric terthiophene-diacetylene 45 which is the precursor of 47. Indeed, both macrocycles 45 and 47 are systems that exhibit a conjugation path corresponding to an anti-aromatic 32 π-electron system. By comparing the 13C NMR signals of the cyclo[n]thiophenes significant shift is observed only for the C-3’,4’ carbons of unsubstituted middle thiophenes. The signal is gradually low field shifted with increasing ring size. For comparison, the signal appears at δ = 122.8 for the smallest homologue 47, whereas for the largest 50 at δ = 125.7 ppm. Obviously, the ring strain that becomes higher as the ring size gets smaller has the most pronounced effect on the unsubstituted thiophenes. Similar observations were made for the series of the terthiophene-diacetylene macrocycles C[3T-DA]n (n = 3-10) that were presented in Chapter 4.

5.6 Results and discussion 287

All cyclo[n]thiophene macrocycles are stable red microcrystalline solids which are very well soluble in most organic solvents. The moderate yields obtained in the reactions with sodium sulfide, especially the conversion of cyclotetramer 27b to cyclo[16]thiophene 49, are attributed to the number of butadiyne units in one molecule that have to be transformed to a thiophene ring. In general, a 60 % yield per step is obtained for the parent, linear butyl-substituted oligothiophenes.49 An important aspect that has to be considered is the ring strain of these cyclic compounds. Certainly, both, the macrocycles used as starting materials and the cyclic products are likely to decompose at a reaction temperature of 140°C. The low yield of 19 % for cyclo[8]thiophene 47 is thus attributed to the partial decomposition of both macrocycles, the starting material 45 and product 47 with probably highest ring strains in the homologues series. In this thesis, a new protocol for the reaction of linear dithienylbutadiynes with sulfides to the corresponding linear oligothiophenes has been developed. A detail discussion of this reaction has been given in Chapter 4.3.1. It has been shown that by using DMSO as solvent in the presence of potassium hydroxide conversion takes place already at 80°C reaction temperature. In order to examine the efficiency of this new protocol, the synthesis of cyclo[8]thiophene 47 was carried out by using the same reaction conditions as above. Therefore, cyclodimeric terthiophene-diacetylene 45 was reacted with high excess of sodium hydrogensulfide hydrate and potassium hydroxide in DMSO at 80°C (Scheme 5.60). The reaction was monitored by TLC showing complete conversion already after 4 hours. After chromatographic purification the pure product, cyclo[8]thiophene 47, could be isolated in 36 % yield providing evidence for the efficiency of this method.

Bu Bu Bu Bu Bu Bu S Bu S Bu S S S S NaHS -hydrate S S DMSO / 80°C / 4h S S S S S Bu S Bu Bu Bu Bu Bu Bu Bu 45 47 [36 %]

Scheme 5.60. Synthesis of cyclo[8]thiophene under new optimized mild reaction conditions.

288 Chapter 5 Conjugated macrocycles by metal template approach

5.6.5 Optical characterization of platinum-linked ethynyl-oligothiophenes

Conjugated organometallic complexes and polymers have been subject of significant research because of their possible applications to many fields of material chemistry.51 By systematic investigations of their physical properties valuable information concerning structure-property relationships become accessible. Of great fundamental and technological interest is to study the effect caused by the insertion of the metal units which, in a conjugated system, may either promote or inhibit electron delocalization. In this work, a large number of acyclic and cyclic platinum-linked ethynyl-oligothiophenes have been prepared. The optical properties of these compounds that are illustrated in Figure 5.59 were investigated and compared with their precursors. The absorption and emission maxima and fluorescence quantum yields of the compounds are summarized in Table 5.10.

Bu Bu Bu Bu Bu Bu Bu Bu S S S n S S (H C) Si n Si(CH ) Ph Ph 3 3 3 3 S P 23, 24, 25 (n=1, 2, 3) Pt P S Ph Ph S S Bu Bu n Bu Bu PPh Bu Bu 3 S Pt 35, 37 (n = 0,1) S PPh Bu 3 Bu

38 Bu Bu Bu Bu PPh3 Pt Bu Bu S S Bu PPh3 Bu Bu Bu S S S Bu Bu S S S S Ph Ph Ph Ph P n P Pt Pt S S P P Ph Ph Ph Ph S S S S n Bu Bu S Bu Bu S S Bu S PPh3 S Bu Bu Bu Pt

Bu PPh3 Bu 40, 41, 43 (n= 1,2,3) Bu Bu 44

Figure 5.59. Acyclic and cyclic platinum-linked ethynyl-oligothiophenes prepared and their ethylnyl- oligothiophene precursors.

5.6 Results and discussion 289

All the electronic absorption spectra were measured in dichloromethane at a concentration of 1 x 10-5 mol/L. In general, the absorption spectra of the protected diethynyl- oligothiophenes 23-25 show typical absorption bands in the blue and visible region of the spectrum, which arise from the π− π∗ transition of the oligothienyl moieties. The broad π− π* transition bands reflect the non-coplanarity and rotational freedom of the individual thiophene units in the aromatic systems. As expected, an increase in the extent of π- conjugation with additional thienyl units results in a bathochromic shift and increase in the extinction coefficients.

Table 5.10. Optical properties of platinum-linked ethynyl-oligothiophenes and diethynyl-oligothiophenes.

abs [a] em [a] em [b, d] Compound λmax [nm] lg ε λmax [nm] φ295Κ

23 382 4.50 465, 492, 527 [c] 5

24 407 4.65 521, 553, 600 [c] 25

25 419 4.76 547, 583 19 precursors precursors

36 263, 286, 328, 344 4.33 - -

37 387 4.21 468, 490, 527 [c] < 1

38 266, 294, 369 4.19 - - linear

40 409 4.85 508, 541 < 1

41 423 4.98 555, 584 2

43 428 5.09 563, 595 10

44 436 5.15 564, 595 9 cyclic [a] Solvent dichloromethane, c = 1 x 10-5 mol/l (abs.) and c = 1 x 10-6 mol/l (em.); [b] external standard was 9,10-diphenylanthracene; [c] shoulder; [d] tolerance ±5.

The electronic spectra of the diplatina macrocycles 40, 41, 43 and 44 are dominated by broad π-π* transition bands of the corresponding oligothiophene moieties. Similar observations have been reported in the literature for other platinum-linked thiophenes.52,53 However, the transition energies are shifted to lower values and the absorption intensities are increased, indicating some enhancement in the degree of π-delocalization through the platinum system. In Figure 5.60, an example of the electronic absorption and corrected 290 Chapter 5 Conjugated macrocycles by metal template approach

emission spectra of protected diethynyl-quinquethiophene precursor 24 and the corresponding quinquethiophene-derived diplatinacycle 41 are shown.

100000

80000 ]

-1 60000 cm -1

40000 [L mol [L ε 20000

0 300 400 500 600 700 800

λ [nm]

Figure 5.60. Absorption and corrected emission spectra of protected diethynyl-quinquethiophene 24 () and the corresponding diplatinacycle 41 (⋅⋅-⋅⋅-) in dichloromethane, c = 1 x 10-5 mol/l (abs.) and c = 1 x 10-6 mol/l (em.).

When the absorption spectra is compared throughout the series of the diplatinacycles (40, 41, 43), a decrease in the transition energies and an increase in the extinction coefficient values with increasing π-conjugation through additional thienyl units can be observed. In Figure 5.61 the absorption spectra of metallacycles 40, 41 and 43 are illustrated. In this series of metallacycles, the absorptions are red-shifted and intensified with increasing chain length of the attached oligothiophenes. The red-shift is more pronounced for the shorter systems with an initial step of ∆λabs = 20 nm from compound 40 (terthiophene) to compound 41 (quinqethiophene), and ∆λabs = 8 nm from compound 41 to compound 43 (septithiophene). Similar trends are observed for all conjugated systems with comparable substitutional arrangements. These observations accounted for an increase in steric interaction which reduces the planarity of the π systems as the chain length increases.

5.6 Results and discussion 291

140000

120000

100000 ] -1 80000 cm -1

60000 [L mol

ε 40000

20000

0 300 400 500 600 λ [nm]

Figure 5.61. Absorption spectra of diplatinacycles 40 (⋅⋅⋅⋅⋅), 41 (⋅⋅-⋅⋅-), 43 () in dichloromethane, c = 1 x 10- 5 mol/l.

For the smallest linear homologues 35 and 38 the absorption bands of the phenyl- phosphine ligands at around λ = 265 and 290 nm are visible (Figure 5.62).

35000

30000

25000 ] -1 20000 cm -1

15000 [L mol

ε 10000

5000

0 250 300 350 400 450 λ [nm]

Figure 5.62. Absorption spectra of the linear cis(thienyl-ethynyl)Pt(II) complex 35 () and trans(thienyl- ethynyl)Pt(II) complex 38 (⋅⋅-⋅⋅-) in dichloromethane, c = 1 x 10-5 mol/l. 292 Chapter 5 Conjugated macrocycles by metal template approach

Noteworthy to mention is the effect of the configuration (cis- or trans) of the platinum moiety on the absorption. A comparison of the linear cis-isomer 35 with the corresponding trans-isomer 38 shows that the absorption maximum of the latter displays a significant bathochromic shift. This effect is also observed for the large cyclic compounds 43 and 44 and has been reported in literature also for other systems.54

The emission spectra of all compounds were measured in dichloromethane at a concentration of 1 x 10-6 mol/L. The fluorescence of the protected diethynyl- oligothiophenes is strongly quenched upon diplatinacycle formation. Weak emission of the complexes occurs when excited at λ = 400 nm with maxima given in Table 5.10. The emission maxima rise when the attached oligothiophenes become longer and exhibit the same trend as discussed above for the absorption maxima. A progressive decrease of ∆λem

= 47 nm from compound 40 to 41 to only ∆λem = 8 nm from compound 41 to 43 was found in the series. In most cases, the fluorescence spectra show very poor mirror symmetry with the lowest-energy absorption band. Weakly structured emission bands with a distinct shoulder in the red-region were found indicating a more planarized and stiffer structure in the excited states. The larger Stokes shifts for the longer systems imply that the oligomeric backbones become less rigid as the number of thienyl units is increased.

5.6.6 Optical properties of the smallest conjugated macrocycles 45 and 47

The optical and electrochemical properties of well-defined macrocycles with fully π- conjugated system periphery are of particular interest. Due to their appealing structural characteristics they represent ideal model compounds for the corresponding polymers and oligomers without suffering backdraws such as polydispersity and end-effects. In this work, the electronic properties of the smallest homologues in the series of conjugated macrocycles, the cyclodimeric terthiophene-diacetylene 45 and the cyclo[8]thiophene 47 (Figure 5.63), were investigated by means of optical and electrochemical analysis. The physical properties of all other conjugated macrocycles that were prepared in the frame of this thesis, the higher homologues in the series of cyclic terthiophene-diacetylenes 27a-h, the quinquethiophene-diacetylene macrocycles 29a-d, as well the cyclo[n]thiophenes (n = 12, 16, 18) have been investigated by others.41,55

5.6 Results and discussion 293

Bu Bu Bu Bu Bu Bu S Bu S Bu S S S S

S S

S S S S Bu S S Bu Bu Bu Bu Bu Bu Bu 47 45 Figure 5.63. Cyclodimeric terthiophene-diacetylene 45 and cyclo[8]thiophene 47.

The absorption and emission maxima of macrocycles 45 and 47 are given in Table 5.11 and the spectra are illustrated in Figure 5.64 and 5.65, respectively. The absorption spectra of both conjugated macrocycles 45 and 47 were measured in dichloromethane at a concentration of 3 x 10-5 M. Typically, parent linear butylated conjugated oligothiophenes and butadiyne-bridged oligothiophenes show broad and unstructured bands due to the π-π* transition reflecting the non-coplanarity and rotational freedom of the individual rings in the aromatic system. This is also observed in the absorption spectra of the macrocyles 45 and 47. The absorption spectrum of 45 shows a absorption band at 400 nm and a tail at longer wavelength with the shoulder centred at around 460 nm. The same feature in the absorption spectrum is also seen for the oligothiophene macrocycle 47 which has the absorption maximum at 396 nm.

Table 5.11. Optical properties of the fully conjugated cyclodimeric terthiophene-diacetylene 45 and cyclo[8]thiophene 47.

abs [a] em [a] λmax [nm] λmax [nm] em [b] Compound lg ε φ295Κ [%] [c] S0-S2 S0-S1 S2-S0 S1-S0

45 400 ∼490 5.13 545 584 < 1

47 396 460 4.71 565 600 < 1

[a] Solvent dichloromethane, c = 3 x 10-5 mol/l (abs.) and c = 3 x 10-6 mol/l (em.); [b] external standard was 9,10-diphenylanthracene; [c] shoulder.

294 Chapter 5 Conjugated macrocycles by metal template approach

160000 [a.u.] intensity fluorescence 1.0 140000

120000 0.8

] 100000 -1 0.6 cm 80000 -1

60000 0.4 [L mol

ε 40000 0.2 20000

0 0.0 300 400 500 600 700 800

λ [nm]

Figure 5.64. Absorption () and corrected emission spectra (⋅⋅-⋅⋅-) of cyclodimeric terthiophene-diacetylene 45 in dichloromethane, c = 3 x 10-5 mol/l (abs.) and c = 3 x 10-6 mol/l (em.).

1.0 50000 intensity fluorescence [a.u.] 0.8 40000 ] -1 0.6 cm 30000 -1

0.4 20000 [L mol ε

10000 0.2

0 0.0 300 400 500 600 700 800 λ [nm]

Figure 5.65. Absorption () and corrected emission spectra (⋅⋅-⋅⋅-) of cyclo[8]thiophene 47 in dichloromethane, c = 3 x 10-5 mol/l (abs.) and c = 3 x 10-6 mol/l (em.). In addition the excitation spectrum (--- ) for 47 at λem = 610 nm is depicted.

Within homologues series of π-conjugated systems, as the size and thus the conjugation length increases, in general, the absorption bands arising from π-π* transitions are shifted to the red. In Table 5.12, the absorption data of the cyclo[n]thiophenes (n = 8, 12, 16 and 18) are displayed for comparison. The optical properties of the cyclo[n]thiophenes with n = 5.6 Results and discussion 295

12, 16, 18 were investigated and reported by others.55 For comparison reasons, the absorption data of some parent butyl-substituted linear oligothiophenes nT (nT designate for the linear oligothiophene having n number of α-linked thiophenes in the chain) are given. As can be seen in the cyclo[n]thiophene series, the general tendency of the red shift with increasing conjugation length is realized for n = 12, 16 and 18. Compared to the related linear butylated oligothiophenes, these macrocycles exhibit absorption maxima values which correspond to those of the linear oligomers with half the number of thiophene rings. This is not the case for the extinction coefficients. For a example, the absorption maximum of cyclo[18]thiophene is almost similar to that of the nonathiophene 9T and much lower than that of the nonadecithiophene 19T. In contrast, the extinction coefficient of cyclo[18]thiophene is almost similar to that of the nonadecithiophene (19T). This characteristic behaviour of the cyclic oligothiophenes (n= 12, 16 and 18) is attributed to their geometry and highly symmetrical structure. As a result, the 0-1 transition is forbidden due to selection rules, while a 0-2 becomes more likely. The optical properties of the smallest homologue in the series of fully conjugated cyclothiophenes, namely the cyclo[8]thiophene, show rather unique features not observed abs for the higher homologues. The absorption maximum of macrocycle 47 is at λmax = 396 nm, which is a much higher value than expected. This value and that of the extinction coefficient correspond more likely to theoretical expected values of a linear em quaterthiophene than of an octithiophene. Remarkably, in the excitation spectrum (λmax = 610 nm) of cyclo[8]thiophene (Figure 5.63), an additional transition at longer abs wavelength (λ = 460 nm) is observed. The first value at λmax = 396 nm is attributed to the 0-2 transition, while that at λ = 460 nm to a 0-1 transition.

Table 5.12. Optical properties of cyclo[n]thiophenes with n= 8, 12, 16, 18 and parent butylated linear oligothiophenes.41

Cyclic oligothiophenes[a] Linear oligothiophenes[b] Compound C[8]T C[12]T C[16]T C[18]T 5T 7T 9T 15T 17T 19T

λ abs max 396 391 415 419 386 409 418 427 429 430 [nm]

lg ε 4.71 4.74 4.99 5.09 4.42 4.61 4.74 5.00 5.06 5.10

[a] C[n]T designate for cyclo[n]thiophene with corresponding n [b] nT designate for the linear oligothiophene having n number of α-linked thiophenes in the chain.

296 Chapter 5 Conjugated macrocycles by metal template approach

The emission spectra of macrocycles 45 and 47 were measured in dichloromethane at a concentration of 3 x 10-6 M. Both compounds exhibit a very low fluorescence (< 1 %). On account of vibronic couplings, more structured emission bands are found for both macrocycles indicating a more planarized and stiffer structure in the excited state. Remarkably, the intensity ratios of the 0-0 to the 0-1 transition states for the two macrocycles are different. For 45 the 0-0 transition state is the most intense, while for 47 both intensities are almost similar.

5.6.7 Electrochemical characterization of conjugated macrocycles 45 and 47

Although optical measurements can be correlated to the energy difference of the frontier orbitals, precise determination of oxidation and reduction potentials provide additional information on the relative energy levels of both HOMOs and LUMOs. Previous quantum mechanical calculations of a cyclo[12]thiophene showed intriguing properties with respect to charge delocalization in such cyclic molecules.56 It is well established that a single charge on a conjugated chain forms a polaron. This is a localized charge that induces a geometric distortion in a chain segment. When two charges reside on a conjugated chain, either two separated polarons or a bipolaron can be formed. By two separated polarons the coulombic interactions are minimized. By formation of a bipolaron the charges merge such that the geometric distortion is minimized. For the fully planar cyclo[12]thiophene two possible geometries of a dication were calculated. In one structure the macrocycle was treated as a bipolaron that is formed by removing two electrons from the HOMO. In the other structure, the dication is allowed to form two separated polarons. Although both geometry optimizations started from a completely delocalized electron distribution, in both models charge localization at the areas of quinoidal geometry resulted (Figure 5.66). In the bipolaron case, the quinoidal geometry included half of the ring on which the charge was located. In the polaronic case, there are two regions of quinoidal geometry comprising a terthiophene unit and bearing one charge. In such geometry, the charges are apart as far as possible. However, energy calculations revealed that, if counterions are not considered, the two-individual polaron geometry is more stable than a bipolaron.

5.6 Results and discussion 297

S S S

S S

S S -2e- -2e- S S S S S S S S S + * S S S S + S cyclo[12]thiophene S S + S

S S S S * + S S S S S S

bipolaron 2 polarons

Figure 5.66. Possible charge delocalization in cyclo[12]thiophene.

The cyclic volatammograms of the conjugated macrocycles 45 and 47 were recorded in dichloromethane. Values concerning the oxidation potentials of these compounds are listed in Table 5.13 and the voltammograms are displayed in Figure 5.67. ο Mmacrocycle 45 undergoes two reversible one-electron oxidation processes (Ε 1= 0.35 V ο + and Ε 2= 0.56 V vs. Fc/Fc ) and no reduction process. The difference between the anodic and cathodic peak for both waves amount to ∆E = 60 mV indicating two separate one- electron oxidations. The reversibility of the oxidations indicates the formation of a stable radical cation and dication: 45 → 45+• → 452+. The high stability is probably due to the lack of free α-positions of the thiophene rings included in a cyclic structure.

Table 5.13. Oxidation potentials of macrocyclic terthiophene-diacetylene 45 and cyclo[8]thiophene 47 determined in dichloromethane/ TBAHPF (0.1 M) vs Fc/Fc+ at 100 mV/s.

Compound 45 47

ο Ε 1 0.35 0.54

ο Ε 2 0.56 0.76

ο Ε 3 -- 1.45

298 Chapter 5 Conjugated macrocycles by metal template approach

-0.50 -0.25 0.00 0.25 0.50 0.75 -0.25 0.00 0.25 0.50 0.75 1.00 1.25 1.50

1.0 1.0 1.0 1.0

0.5 0.5 0.5 0.5

I [µA] I [µA]

0.0 0.0 0.0 0.0

-0.5 -0.5 -0.5 -0.50 -0.25 0.00 0.25 0.50 0.75 -0.25 0.00 0.25 0.50 0.75 1.00 1.25 1.50 E [V] vs. Fc/Fc+ E [V] vs. Fc/Fc+

Figure 5.67. Cyclic voltammograms of terthiophene-diacetylene 45 (left) and cyclo[8]thiophene 47 (right) measured in dichloromethane/ TBAHPF (0.1 M) vs Fc/Fc+ at a scan rate of 100 mV/s.

The cyclic voltammogram of cyclo[8]thiophene 47 exhibits three reversible one-electron ο ο ο + oxidation processes (Ε 1 = 0.54 V, Ε 2= 0.76 V and Ε 3= 1.45 vs. Fc/Fc ). Again no reduction process is observed. The reversibility of the oxidations indicates the formation of stable radical cation, dication and a stable radical trication. In Scheme 5.61, the proposed structure and charge distribution of these spezies are illustrated. Removal of one electron results in the formation of the radical cation 47+• with probably a fully quinoidal structure. For the dication 472+ a structure in which the charges are localized on the half of the ring with quinoidal geometry is seen. As mentioned previously, this corresponds to a bipolaronic structure. Different is the situation for the radical trication 473+•. Here the charges are as far apart as possible and the structure comprises two regions of quinoidal geometry. As for the cyclo[12]thiophene, the quinoidal areas include terthiophene units that bear either one or two charges.

5.6 Results and discussion 299

Bu Bu Bu Bu Bu Bu Bu Bu S Bu . + S Bu Bu + S Bu S S S S S S -e - -e - S S S S S S

S S S S S S S Bu Bu Bu S Bu Bu S + Bu

Bu Bu Bu Bu Bu Bu 2+ 47 47+. 47

-e

Bu Bu

Bu + S Bu . S S S S + S S Bu S + Bu

Bu Bu . 473+ Scheme 5.61. Proposed structure and charge distribution of stable radical cation 47+•, dication 472+ and radical trication 473+• of cyclo[8]thiophene.

5.6.8 X-ray crystallography of cyclodimeric terthiophene-diacetylene 45

X–ray data for nanosized macrocycles have been rare since the growth of single crystals of these compounds is very difficult. As discussed in Chapter 3, the large cavities of the cycles are mostly filled with solvent molecules, the loss of which, upon evaporation, causes destruction of the crystals.57 For solubility reasons, usually flexible side-chains are attached to the core of such macrocycles. This causes an additional intrinsic disorder and defects in the crystals, rendering the refinement values R often higher. Some examples reported in the literature including the solid-state structure characterization of shape- persistent macrocycles have been presented in Chapter 3.2.4. Crystallographic structure determinations of several α-linked linear oligothiophenes have been reported.58 Typically, the unsubstituted molecules adopt a coplanar all-anti conformation and they pack in a herring-bone pattern due to favourable edge-to-face aromatic interactions. Substitution at the β-positions of the thiophene rings lead to deviations from coplanarity resulting in twist of the thiophene rings and generally causes a 300 Chapter 5 Conjugated macrocycles by metal template approach

certain percentage of syn conformation. In these cases the molecules mostly tend to stack in parallel layers. X-ray structure characterizations of cyclic oligothiophene-diacetylenes have already been reported for cyclotrimeric 27a and cyclotetrameric 27b.47,59 27a was found to crystallizes in the triclinic space group P1 with two molecules in the unit cell (Z = 2), while the corresponding tetramer 27b in the monoclinic space group P-1 with one molecule in the unit cell (Z = 1).

Single crystals of cyclodimeric macrocycle 45 were obtained by slow evaporation from chloroform solution.60 Selected crystallographic data are given in Table 5.14. Complete crystallographic data and the refinement parameters have been listed as supplement at the end of this chapter.

Table 5.14. Selected crystallographic data for the cyclodimeric terthiophene-diacetylene 45.

crystal system monoclinic

space group P21/a

unit cell dimensions a = 9.0736(13) Å, b = 20.222(2) Å, c = 16.503(2) Å

α= 90°, β = 104.872(15)°, γ = 90°

volume 2926.6(6) Å3

Z 2

calculated density 1.177 g/cm3

Terthiophene-diacetylene macrocyle 45 crystallizes in the monoclinic space group P21/a with two molecules in the unit cell (Z = 2). In Figure 5.68 the side-view projection of the unit cell is displayed. In the unit cell two different layers can be seen where each of them composes of two molecules that are oriented parallel to each other. These two molecules, which are individually planar, form dimers in an edge-to-face fashion. The butyl chains interact with the side chains of the adjacent molecule and do not fill the cavity of the adjacent macrocycle. Similar formation of dimers along the molecular plane was observed also observed for cyclotetramer. In this case, the individual molecules were orientated parallel but slightly, laterally, displaced to each other. 5.6 Results and discussion 301

Figure 5.68. Side view packing along the b-axis of cyclodimeric macrocycle 45.

In Figure 5.69 the top view of an individual molecule is illustrated revealing the nearly prefect circular shape, and with the terthiophene moieties all in syn conformation. Due to the steric interactions of the butyl chains adjacent thiophene rings are distorted by 13.8° and -29.1°. Similar conformation was also reported for the higher homologues 27a and 27b. The dihedral angles between the syn arranged thiophene rings were 28.8° and 31.1° for the cyclotrimer 27a and between 27° and 40° for cyclotetramer 27b. The butadiyne moieties of macrocycle 45 that connect the terthiophene units are concavely bowed with bond angles between 160.3° and 167.7°, largely deviating from linearity. In contrast, the alternating carbon-carbon bond lengths are relatively unaffected and are consistent with those of the linear bisthienylbutadiyne 15.41 Obviously, the high ring strain that is imparted as a result of the planarity of the cyclic structure is the cause for the distortion of the butydiyne units. Analogues values for butadiyne bond angles have been reported for strained dehydroannulenes and cyclophanes.61 The distances inside the cavitiy of cyclodimer 45 are from S1 to S1’ 11.04 Å, and from S1’ to S3 8.07 Å. For comparison, the inner diameter of cyclotrimer was reported to be 9.60 Å and that of the cyclotetramer 27b with a more rectanular shape 19.4 Å and 13.4 Å.

302 Chapter 5 Conjugated macrocycles by metal template approach

C123

C19 C124 C1’ C17 C15 C122 C13 C121 C11 C12 S1‘ 8.06 Å C102 C104 S3 C10 C101 C9 C103 11.04Å C7

S2 C8

S1 C6 C3 C5

C1 C42 C44 C4 C2 C41 C43 C24 C21

C23 C22

Figure 5.69. Top view of an individual molecule of 45. Selected bond lengths [Å]: C11-C13 1.417, C13-C15 1.203, C15-C17 1.368, C17-C19 1.205, C1’-C19 1.416; bond angles C11-C13-C15 160.3°, C13-C15-C17 167.4°, C15-C17-C19 167.7°, C17-C19-C1’ 163.2° and torsion angles S1-C3-C5-S2 -29.9°, S2-C7-C9-S3 13.80°.

All three terthiophene-diacetylene macrocycles 45, 27a and 27b are practically planar molecules. Due to their different ring sizes, they have different shape and packing behaviours. The structures of all these three macrocycles may be regarded as composed layers. Each of these layers is defined by the macrocycle within the plane. Perpendicular to the plane of each macrocycle, structures with ’’channels’’ are formed. The packing of the smallest homologue 45 differs from that of the others by the fact that two molecules are oriented edge-to-face to each other. This gives rise to the formation of a second layer which is not parallel to the first one. The ring-strain increases as smaller the ring-size gets. This is mirrored by the angles of the butadiyne moieties of the three macrocycles. The butadiyne fragments of the larger homologues 27a and 27b deviate from linearity only slightly with bond angles between 5.6 Results and discussion 303

173°and 176°. In contrast, the butadiyne fragments of the smallest 32-memberd homologue 45 are severely bowed with angles of 160°-167°. Interestingly, despite this distortion, the 32π-electron system shows remarkable conjugation as is already discussed in Chapter 5.6.2.2.

5.7 Conclusion

In this study, a very efficient method for the synthesis of fully conjugated macrocycles has been developed. This method, which is referred to as the metal template approach includes the high yielding formation of diplatina macrocycles followed by subsequent C-C bond formation through elimination of the transition metals units by the help of an oxidant. The method development has been carried out by using ethynylthiophenes as simple model compounds. In this frame, novel acyclic bis(oligothienyl-ethynyl)platinum(II) complexes have been synthesized and characterized. The reductive elimination reaction leading to expellation of the metal units under simultaneous C-C coupling of the organic moieties has been investigated in detail. Quantitative reductive elimination from the cis- bis(oligothienyl-ethynyl)platinum(II) complexes to the corresponding butadiynes was achieved by equimolar addition of iodine. The application of the metal template approach towards the synthesis of cyclic structures resulted in the formation of various platinum-bridged diethynylated oligothiophene macrocycles in very high yields. These diplatina macrocycles represent a novel topology and class of compounds in the area of nananosized macrocycles. The subsequent elimination of the transition metal units from these diplatina macrocycles by treatement with iodine led, under preservation of the cyclic structure, to the formation of the targeted macrocyclic oligothiophene-diacetylenes. By this reaction the cyclodimeric spezies were favourable formed. Particularly in the case of the terthiophene-derived macrocycles, the reaction led to the formation of the strained cyclodimeric terthiophene-diacetylene macrocycle in a remarkable yield of 54%. This antiaromatic, 32 π-electron conjugated system is the smallest macrocycle within the series of conjugated oligothiophene- diacetylenes that was ever been detected. Following, a series of α-linked conjugated macrocyclic oligothiophenes, cyclo[n]thiophenes (n = 8, 12, 16, 18), has been prepared by reacting the corresponding macrocyclic oligothiophene-diacetylenes with sulfide anions. 304 Chapter 5 Conjugated macrocycles by metal template approach

The investigations on the optical properties of the novel platinum-linked ethynylated oligothiophenes clearly revealed that insertion of the metal units causes changes in the absorption and emission behaviours of conjugated systems. Even though the electronic spectra of the platinum complexes were still dominated by broad π-π* transition bands of the corresponding oligothiophene moieties, the transition energies were shifted to lower values and the absorption intensities were increased, indicating some enhancement in the degree of π-delocalization through the cyclic system. Investigations on the optical and electrochemical properties of the two smallest macrocycles, the cyclodimeric terthiophene-diacetylene and cyclo[8]thiophene, have been carried out. Both highly strained, 32π-electron cyclic systems exhibit unique and interesting electronic features which were not observed for their higher homologues. Typically, due to their highly symmetrical structure the 0-1 transition for cyclo[n]thiophenes is forbidden, while a 0-2 becomes more likely. This is the reason why these cyclic compounds absorb at values which correspond to those of the linear oligomers with half the number of thiophene rings. In this respect, it is remarkable that in the absorption spectra of the two smallest macrocycles tails at longer wavelength with the shoulder centred at values which are expected for a corresponding to a 0-1 transition appeared. The cyclic voltammetry studies showed that upon oxidation, both macrocycles form highly stable spezies. In the case of cyclodimeric terthiophene-diacetylene a stable radical cation and dication was observed, whereas in the case of the cyclo[8]thiophene even a stable radical trication was formed. By X-ray analysis a detailed investigation of the structural parameters was possible for the cyclodimeric terthiophene-diacetylene. In contrast to its next higher homologues, the cyclodimer shows a different packing behaviour by the fact that two molecules are oriented in and edge-to-face fashion to each other. In two dimensions, the backbone of the macrocycle shows similar conformation as its higher homologues forming almost planar sheets with the all-syn orientated thiophenes which are slightly tilted out of plane. The butadiyne fragments of the cyclotrimer and cyclotetramer deviates only slightly indicating ring-strain free structures. In contrast, the butadiyne fragments of the smallest 32-memberd homologue were severely bowed. Nevertheless, despite this distortion, a remarkable conjugation was found for the 32π-electron system.

5.8 Experimental section 305

5.7 Experimental section

5.7.1 Instrumentation and general experimental conditions

Solvents and reagents were purified and dried by usual methods prior to use. Thin-layer Chromatography (TLC) was carried out on plastic plates Polygram SIL

G/UV254 from Macherey & Nagel. Developed plates were dried and examined under a UV lamp. Preparative column chromatography was performed on glass columns of different sizes packed with silica gel 60 (particle size 0.040-0.063 mm), Merck. Gas chromatography (GC) was executed with a Carlo-Erba Auto-HRGC MFC 500 equipped with PS086 glass capillary columns (Ø 0.32 mm, length 10 and 20 m). Helium 5.0 was used as carrying gas; eluted materials were detected by a Carlo-Erba EL 580 flame-ionization detector (FID). Chromatograms were recorded on a Spectra-Physics DP 700 integrator. Gas chromatography – Mass Spectrometry (GC-MS) measurements were executed with a Varian 3800 equipped with CP4860 glass capillary columns (length 30 m). Helium 5.0 was used as carrying gas. Mass spectra were recorded on a Varian Saturn 2000. Ions were generated by electron impact (EI). Melting points were determined with a Büchi B-545 melting point apparatus and are uncor- rected. FT-IR spectroscopy was performed on a Perkin-Elmer Spectrum 2000 spectrometer. Samples were prepared as KBr pellets. Band positions are reported in reciprocal centimetres. Relative intensities of single absorption bands are indicated by s = strong, m = middle, w = weak and br = broad. NMR spectroscopy measurements were carried out on Bruker AMX 500 and DRX 400 spectrometers. Chemical shifts are expressed in parts per million (δ) downfield from the internal tetramethylsilane reference (δH = 0.00) or using residual solvent protons as internal standards (CDCl3: δH = 7.26, δC = 77.0). Spin multiplicities are indicated by the following symbols: s (singlet), d (doublet), t (triplet), m (multiplet), dd (doublet of doublets). Mass spectra were recorded on a Varian MAT 711 spectrometer. Ions were generated by electron impact (EI) at 70 eV. Relative intensities of single peaks with respect to the base peak are given in parentheses. In some cases assignments of corresponding fragments are appended in square brackets. Matrix-assisted laser desorption ionization time-of-flight 306 Chapter 5 Conjugated macrocycles by metal template approach

mass spectroscopy (MALDI-TOF MS) measurements were carried out on a Brucker Daltonik Reflex III spectrometer by using 1,8,9-trihydroxyanthracene (Dithranol) as matrix. Solutions of the compounds in tetrahydrofurane or dichloromethane were prepared. The calculated and experimental m/z values refer to the monoisotopic mass unless otherwise stated. High resolution fast bombardment mass spectra (HRMS, FAB) were recorded on a Finnigan MAT 95 spectrometer (University Stuttgart, Germany) by using nitrobenzylalcohol as matrix. ESI-FTICR MS measurements were carried out on a BrukerApex IV mass spectrometer (University Bonn, Germany). Solutions of the compounds in a mixture of tetrahydrofurane, water and acetic acid were prepared. The calculated m/z values correspond to the average masses and the experimental (found) m/z values correspond to the respective peak with the highest intensivity. HPLC analysis was accomplished with a Shimadzu SCL-10AVP controller, LC-10AT pump, SPD-M10A VP detector using a Macherey-Nagel nucleosil column NO2 (4 mm x 250 mm, corn diameter 5µm), 1,3 mL/min flow rate. Preparative HPLC was performed with a Shimadzu LC-8A pump, SPD-10A detector using a Macherey-Nagel nucleosil column NO2 (40 mm x 250 mm, corn diameter 100- 10µm), 80 mL/min flow rate. UV/VIS/NIR spectra were recorded on a Perkin-Elmer Lambda 19 spectrometer in 1 cm cu- vettes. Elemental analyses were performed on an Elementar Vario EL (University Ulm) and a Carlo Erba 1104 (University Stuttgart) (limit of experimental error: ± 0.3 %). Cyclic voltammetry and electrolysis experiments were performed with a computer- controlled EG&G PAR 273 potentiostat in a three-electrode single-compartment cell (5mL, 20 mL). The platinum working electrode consisted of a platinum wire sealed in a soft glass tube with a surface A = 0.785 mm2, which was polished down to 0.5 µm with Buehler polishing paste prior to use in order to obtain reproducible surfaces. The counter electrode consisted of a platinum wire. The reference was a Ag/AgCl secondary electrode. All potentials were internally referenced to the ferrocene/ ferrocenium couple. Argon 4.8 was used to purge all solutions before use. Routinely, a concentration of 10-4-10-3 mol/L of electroactive species was used. The electrolyte consisted of either dichloromethane or benzonitrile (Uvasol-Grade, Merck). The solvents were directly transferred by means of syringes to the electrochemical cell. The supporting salt was 0.1 M TBAHFP from Fluka which was recrystallized twice from ethanol/water and dried in high vacuum.

5.8 Experimental section 307

Nomenclature: All compounds were named according to IUPAC rules using ”ISIS-Draw 2.5 Plug-In Autonom-Standard” or ”IUPAC name Pro 6.0” programms.

Purchased starting materials and chemicals: Acetic acid (Merck), bis(triphenylphosphine) palladium(II) chloride (Merck), chloroform- d3 with TMS (Merck), copper(I) iodide (Aldrich), iodine (Merck), platinum(II)chloride (Strem), sodium hydrogen carbonate (Merck), sodium sulfate (Merck), sodium thiosulfate (Merck), tetrahydrofurane (Merck), triethylamine (Merck), triphenylphosphine (Merck).

Starting Materials Prepared According to Literature Procedures: [bis(diphenylphosphino)propane]palladium(II)chloride, [bis(diphenylphosphino)-propane] platinum(II)chloride, [bis(triphenylphosphine)]platinum(II) chloride.62

Materials Prepared During This Study and Presented in Chapter 4.3: 3,4-dibutyl-2-trimethylsilylethynyl-thiophene 14, 3,3’’,4,4’’-tetrabutyl-5-trimethylsilylethynyl-2,2’:5’,2’’-terthiophene 12, 5,5’’-bis(trimethylsilylethynyl)-3,3’’,4,4’’- tetrabutyl-2,2’:5’,2’’-terthiophene 24, 5,5’’’’-Bis(trimethylsilylethynyl)-3,3’’,3’’’’- ,4,4’’,4’’’’-hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’-quinquethiophene 25, 5,5’’’’’’- bis(trimethylsilylethynyl)-3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-octabutyl-2,2’:5’,2’’:5’’,- 2’’’:5’’’,2’’’’:5’’’’,2’’’’’:5’’’’’,2’’’’’’-septithiophene 26.

5.7.2 Synthesis and characterization of the compounds

General Procedures (GPs)

GP1: Preparation of linear bis(oligothienyl-2-ethynyl)-platinum(II) compounds A solution of potassium hydroxide in aqueous methanol was dropped to a solution of protected ethynyl-oligothiophene in THF. After stirring at room temperature for 3 hours ice was added to the reaction mixture which subsequently was then extracted with dichloromethane. The organic layer was washed with 1N HCl, saturated NaHCO3 solution and water and was dried over Na2SO4. The solution was concentrated by vacuum distillation under cooling to a small volume. This solution of the deprotected ethynyloligothiophene was then added through a syringe to a stirred suspension of 308 Chapter 5 Conjugated macrocycles by metal template approach

bis(phosphine) platinum(II) chloride, copper(I) iodide, triethylamine in toluene as solvent. The reaction mixture was stirred for 24 h at ambient temperature. In order to remove polymeric material and inorganic salts, after evaporation of the solvent the crude product was filtered through a short column of silica gel with petrol ether/THF (1:3) as eluent. Purification was accomplished by column chromatography on silica gel (eluent). Subsequent addition of methanol to the residue obtained after solvent evaporation yielded the desired product as microcrystalline solid.

GP2: Preparation of linear 1,4-bis(oligothien-3-yl)-1,3-butadiynes by oxidatively induced reductive elimination Iodine was added at room temperature to a solution of bis(oligothienyl-2- ethynyl)platinum(II) compound in THF. The reaction mixture was allowed to stir for 24 h. After filtration of the white precipitate the solution was poured into dichloromethane and water. The organic phase was separated, washed first with saturated NaS2O3 solution, then with water and dried over sodium sulfate. After removal of the solven, t the crude product was purified by column chromatography on silica gel (eluent). Precipitation upon addition of methanol to a concentrated dichloromethane solution yielded the desired product as microcrystalline solid.

GP3: Preparation of macrocyclic bis(oligothienyl-2-ethynyl)-diplatinum(II) compounds A solution of potassium hydroxide in aqueous methanol was dropped to a solution of protected diethynylated oligothiophene in THF. After stirring at room temperature for 3 h ice was added to the reaction mixture which subsequently was then extracted with dichloromethane. The organic layer was washed with 1N HCl, saturated NaHCO3 solution and water and was dried over Na2SO4. The solution was concentrated by vacuum distillation under cooling to a small volume. This solution of the deprotected diethynylated oligothiophene was then added through a syringe to a stirred suspension of bis(phosphino)platinum(II) dichloride, copper(I) iodide, triethylamine in toluene as solvent. The reaction mixture was stirred for 72 h at ambient temperature and then methanol was added. The precipitated solid was separated by filtration, washed with methanol and dried in vacuo affording the desired product. 5.8 Experimental section 309

GP4: Preparation of cyclic 1,4-bis(oligothien-3-yl)-1,3-butadiynes by oxidatively induced reductive elimination A solution of platinacycle in THF was heated to 60°C. Two equivalents of iodine were added and the reaction mixture stirred for 24 h. After filtration of the white precipitate the solution was poured into dichlormethane and water. The organic phase was separated, washed with saturated NaS2O3 solution and water, and dried over sodium sulfate. After removal of solvent the crude product was purified by column chromatography on silica gel (eluent).

GP5: Preparation of cyclo[n]thiophenes Oligothiophene-diacetylene macrocycle and sodium sulfide nonahydrate in a mixture of 2- methoxyethanol and xylene were stirred for 24 h at 140°C. After evaporation of the solvents in vacuo, dichloromethane and water were added. The organic phase was separated, washed several times with water and dried over Na2SO4. Purification was accomplished by column chromatography on silica gel with n-hexane/dichloromethane (4/1) as eluent. After evaporation of the solvent, the product was precipitated upon addition of methanol to a concentrated dichloromethane solution yielding the desired product as microcrystalline solid.

cis-Bis[3,4--dibutyl-2-(ethynyl-κC2)-thienyl][ propane -1,3-diyl-bis(diphenylphosphine-κP)]platinum(II) (35) The synthesis was carried out according to GP1, KOH (325 mg, 5.80 mmol) in methanol

(5 mL), protected ethynylthiophene 14 (170 mg, 580 µmol) in THF (5 mL); Pt(dppp)Cl2

(197 mg, 290 µmol), CuI (11.0 mg, 58 µmol), NEt3 (117 mg, 1.16 mmol) in toluene (25 mL); chromatographic work-up (petrol ether/THF 3/2), yield 72 % (218 mg, 209 µmol), pale yellow solid, m.p. 132°C.

1 H-NMR (400 MHz, CDCl3): δ= 7.81 (m, 8H; H-aryl-dppp), 7.33 (m, 12H; H-aryl-dppp) , 3 6.43 (s, 2H, H-5-Th), 2.44 (brs, 4H, α-CH2-dppp), 2.34 (t, J(α,β-CH2-Bu) = 7.6 Hz, 4H, α- 3 CH2-Bu), 2.28 (t, J(α’,β’-CH2-Bu) = 7.6 Hz, 4H, α’-CH2-Bu), 2.04 (brs, 4H, ß-CH2-dppp), 1.49

(m, 4H, ß-CH2-Bu), 1.32 (m, 4H, ß’-CH2-Bu), 1.23 (m, 4H; γ-CH2-Bu), 1.04 (m, 4H, γ’- 310 Chapter 5 Conjugated macrocycles by metal template approach

3 3 CH2-Bu), 0.89 (t, 6H, J(γ-CH2-Bu, CH3-Bu) = 7.3 Hz, CH3-Bu), 0.70 (t, 6H, J(γ-CH2-Bu, CH3-Bu) =

7.3 Hz, CH3-Bu).

γ β α' α 5 Ph 2 Ph S P β Pt P α S Ph Ph 5

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = -6.80 (s with Pt satellites JPt,P = 2203 Hz, 2P, P- dppp).

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C55H64P2PtS2 1045.4; found 1044.5.

C55H64P2PtS2 (1046.29) calcd. C 63.14 H 6.17 S 6.13. found C 63.03 H 6.20 S 6.24.

cis-Bis[3,3’’,4,4’’-tetrabutyl-5-(ethynyl-κC2)-2,2’:5’,2’’-terthienyl][ propane -1,3-diyl- bis(diphenylphosphine-κP)]platinum(II) (37) The synthesis was carried out according to GP1, KOH (325 mg, 5.80 mmol) in methanol

(8 mL), protected ethynylterthiophene 12 (189 mg, 332 µmol) in THF (8 mL); Pt(dppp)Cl2

(113 mg, 166 µmol), CuI (6.32 mg, 33.2 µmol), NEt3 (67.2 mg, 664 µmol) in toluene (17 mL); chromatographic work-up (petrol ether/THF 1/1), yield 75 % (199 mg, 125 µmol), yellow solid, m.p. 173°C.

1 H-NMR (400 MHz, CDCl3): δ= 7.81 (m, 8H; H-aryl-dppp), 7.38 (m, 12H; H-aryl-dppp), 3 3 6.97 (d, J(H-3’,4’-Th)= 3.8 Hz, 2 H, H-4’-Th), 6.91 (d, J(H-3’,4’-Th)= 3.8 Hz, 2 H, H-3’,-Th), 3 3 6.82 (s, 2 H, H-5’’-Th), 2.70 (t, J(α,β-CH2-Bu) = 7.6 Hz, 4H, α-CH2-Bu), 2.58 (t, J(α’,β’-CH2-Bu)

= 7.6 Hz, 4H, α’-CH2-Bu), 2.50 (m, overlapping, 8H, α’’-CH2-Bu, α-CH2-dppp), 2.33 (t, 3 J(α’,β’-CH2-Bu) = 7.6 Hz, 4H, α’’’-CH2-Bu), 2.06 (brs, 2H, ß-CH2-dppp), 1.65 (m, 4H, ß-CH2- 5.8 Experimental section 311

Bu), 1.55, 1.42, 1.34 (m, overlapping, 20H, ß’,β’’,β’’’,γ,γ’,-CH2-Bu), 1.15 (m, 4H, γ’’-CH2- 3 Bu), 0.96, 0.85 (m, overlaping, 22H, γ’’’-CH2-Bu, CH3-Bu), 0.75 (t, 6H, J(γ-CH2-Bu, CH3-Bu) =

7.3 Hz, CH3-Bu).

γ β 2'' 5'' α S S 2 5 Ph Ph S 4' P 3' β Pt α P S Ph Ph 5 S S 5''

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = -7.13 (s with Pt satellites JPt,P = 2198 Hz, 2P, P- dppp).

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C87H104P2PtS6 1597.6; found 1597.0.

C87H104P2PtS6 (1599.22) calcd. C 65.34 H 6.55 S 12.03. found C 65.09 H 6.58 S 12.01.

trans-Bis[3,4--dibutyl-2-(ethynyl-κC2)-thienyl][propane -1,3-diyl-triphenyl-phosphine- κP)]platinum(II) (38) The synthesis was carried out according to GP1, KOH (220 mg, 3.93 mmol) in methanol (5 mL), protected ethynylthiophene 14 (115 mg, 392 µmol) in toluene (5 mL); trans-

Pt(PPh3)2Cl2 (155 mg, 196 µmol), CuI (7.47 mg, 39.2 µmol), NEt3 (79.3 mg, 784 µmol) in toluene (35 mL); chromatographic work-up (petrol ether/THF 7/3), yield 63 % (143 mg, 123 µmol), brownish solid, m.p. 142°C.

1 H-NMR (400 MHz, CDCl3): δ= 7.52 (m, 12H; H-aryl-PPh3), 7.37 (m, 18H; H-aryl-PPh3) 3 3 , 6.42 (s, 2H, H-5-Th), 2.26 (t, J(α,β-CH2-Bu) = 7.6 Hz, 4H, α-CH2-Bu), 1.80 (t, J(α’,β’-CH2-Bu) = 312 Chapter 5 Conjugated macrocycles by metal template approach

7.6 Hz, 4H, α’-CH2-Bu), 1.43 (m, 4H, ß-CH2-Bu), 1.28 (m, 4H, ß’-CH2-Bu), 0.97, 0.87 (m, 3 overlapping, 14H, γ, γ’-CH2-Bu, CH3-Bu), 0.67 (t, 6H, J(γ-CH2-Bu, CH3-Bu) = 7.2 Hz, CH3-Bu).

PPh 3 S 5 2 Pt 5 S PPh3

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = 19.7 (s with Pt satellites JPt,P = 2652 Hz, 2P, P-

PPh3).

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C64H68P2PtS2 1157.4; found 1156.7.

C64H68P2PtS2 (1158.42) calcd. C 66.36 H 5.92 S 5.52. found C 66.17 H 5.95 S 5.35.

1,4-Bis(3,4 –dibutylthien-2-yl)-1,3-butadiyne (15) The synthesis was carried out according to GP2, iodine (7.23 mg, 28.6 µmol), platinum compound 35 (30.0 mg, 28.6 µmol) in THF (3 mL); chromatographic work-up (petrol ether), yield 99 % (12.4 mg, 28.3 µmol), yellow solid, m.p. 60°C.

1 3 H-NMR (400MHz, CDCl3): δ= 6.87 (s, 2 H, H-5-Th), 2.68 (t, J(α’,β’-CH2-Bu) = 7.6 Hz, 4 H, 3 α’ -CH2-Bu), 2.49 (t, J(α,β-CH2-Bu) = 7.4 Hz, 4 H, α -CH2-Bu), 1.49 (m, 16 H, β-β’’’’’’,γ-

γ’’’’’’-CH2-Bu), 0.95 (m, 12 H, CH3-Bu)

γ β α α α' α' 2 2 5 5 S S 15

5.8 Experimental section 313

13 C-NMR (101 MHz, CDCl3): δ= 149.8 (C-4-Th), 142.1 (C-3-Th), 122.9 (C-5-Th), 117.6

(C-2-Th), 97.7, 77.3 (C≡C), 32.3, 31.8 (β-β’-CH2-Bu), 28.6, 28.1 (α-α’-CH2-Bu), 22.6,

22.5 (γ-γ’-CH2-Bu), 13.9 (CH3-Bu). MS (EI), m/z (%): 438 (29), 438 (100), 353 (13). The analytical data are consistent with those described in the literature (yellow crystals, m.p. 60-61°C).Error! Bookmark not defined.

1,4-Bis(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-2-yl)-1,3-butadiyne (16) The synthesis was carried out according to GP2, iodine (7.92 mg, 31.3 µmol), platinum compound 37 (50.0 mg, 31.3 µmol) in THF (4 mL); chromatographic work-up (petrol ether/dichloromethane 15/1), yield 99 % (30.7 mg, 30.9 µmol), orange solid, m.p. 73°C.

1 3 H-NMR (400MHz, CDCl3): δ= 7.10 (d, J(3’,4’-CH2-Bu) = 3.7 Hz, 2 H, H-3’ or H-4’-Th), 3 7.05 (d, J(3’,4’-CH2-Bu) = 3.7 Hz, 2 H, H-3’ or H-4’-Th), 6.86 (s. 2H, H-5’’-Th), 2.68 (m, 12 3 H, α-α’’’-CH2-Bu), 2.52 (t, J(α’’’,β’’’-CH2-Bu) = 7.8 Hz, 4 H, α’’’-CH2-Bu), 1.51 (m, 32 H, β-

β’’’,γ-γ’’’-CH2-Bu), 0.95 (m, 24 H, CH3-Bu),

γ β α α α' α'' 2 2 5 5 α''' S 2'' S S S 2'' 16 5'' S 3' 3' S 5'' 4' 4'

13 C-NMR (101 MHz, CDCl3): δ= 151.6, 143.7 (C-2’,5’-Th), 139.1, 138.5 (C-4,4’’-Th), 137.4, 135.1, 133.6, 130.5( C-2,2’’,3,3’’-Th), 126.3, 125.9 (C-3’,4’-Th), 119.3 (C-5’’-Th),

116.0 (C-5-Th), 81.2, 77.2 (C≡C), 32.7, 32.6, 31.9 (β-β’’’-CH2-Bu), 28.9, 28.7, 27.8, 27.5

(α-α’’’-CH2-Bu), 23.0, 22.9, 22.7 (γ-γ’’’-CH2-Bu), 14.0, 13.9, 13.8 (CH3-Bu). MS (EI), m/z (%): 993 (22), 992 (47), 991 (63), 990 (100). The analytical data are consistent with those described in the literature (orange solid 72- 73°C).49

314 Chapter 5 Conjugated macrocycles by metal template approach

Cyclo[cis-bis[3,3’’,4,4’’-tetrabutyl-5,5’’-(diethynyl-κC2)-2,2’:5’,2’’- terthienyl][propane -1,3-diyl-bis(diphenylphosphine-κP)]diplatinum(II)] (40). The synthesis was carried out according to GP3, KOH (902 mg, 16.1 mmol), methanol (15 mL), protected diethynylterthiophene 23 (1.07 g, 1.61 mmol), tetrahydofuran (15 mL);

Pt(dppp)Cl2 (1.09 mg, 1.61 mmol), CuI (30.7 mg , 161 µmol), NEt3 (326 mg, 3.22 mmol), in toluene (200 mL); methanol (100 mL); yield 91 % (1.65 g, 73.3 µmol), yellow solid m.p. 255°C (decomp.).

1 H-NMR (500 MHz, CDCl3, 323 K): δ= 7.92 (brs, 16H; H-aryl-dppp), 7.40 (brs, 24H; H- 3 aryl-dppp) , 7.01 (s, 4H, H-3’,4’-Th), 2.57 (t, J(α,β-CH2-Bu) = 7.8 Hz, 8H, α,α’’’-CH2-Bu), 3 2.42 (brs, 8H, α-CH2-dppp), 2.07 (brs, 4H, ß-CH2-dppp), 1.97 (t, J(α’,β’-CH2-Bu) = 7.6 Hz,

8H, α’,α’’-CH2-Bu), 1.53 (m, 8H, ß,β’’’-CH2-Bu), 1.34 (m, 8H, ß’,β’’-CH2-Bu), 1.09 (m, 3 8H; γ,γ’’’-CH2-Bu), 0.88, (t, 12H, J(γ-CH2-Bu, CH3-Bu) = 7.3 Hz, CH3-Bu), 0.82 (m, 8H, γ’,γ’’- 3 CH2-Th), 0.56 (t, 12H, J(γ-CH2-Bu, CH3-Bu) = 7.3 Hz, CH3-Bu).

γ β α' α'' α α''' 2 2'' Ph 5 S 5'' Ph Ph Ph S S P P β Pt Pt P α P S S Ph Ph Ph S Ph

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = -5.59 (s with platinum satellites JPt,P = 2243 Hz, 4P, P-dppp). IR (KBr): 3438, 3053, 2948, 2931, 2860, 2365, 2098, 1630, 1432, 1100, 691, 516 cm-1.

MS (MALDI-TOF): calcd. monoisotopic mass for C118H128P4Pt2S6 m/z = 2250.7; found 2250.4. 2+ HRMS (ESI-FTICR) C118H128P4Pt2S6 (M ) calcd. 1125.3346; found 1125.3142.

5.8 Experimental section 315

Cyclo[cis-bis[3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-5,5’’’’-(diethynyl-κC2)-2,2’:5’,2’’:5’’, 2’’’:5’’’,2’’’’-quinquethienyl][propane -1,3-diyl-bis(diphenylphosphine-κP)]- diplatinum(II)] (41). The synthesis was carried out according to GP3, KOH (416 mg, 74.3 mmol), methanol (10 mL), protected diethynylquinquethiophene 24 (1.07 g, 743 µmol), tetrahydofuran (10 mL);

Pt(dppp)Cl2 (504 mg, 743 µmol), CuI (14.2 mg , 74.3 µmol), NEt3 (151 mg, 1.49 mmol), toluene (100 mL); methanol (50 mL); yield 96 % (1.01 g, 360 µmol), dark-red solid, m.p. >300°C (decomp.).

1 H-NMR (500 MHz, CDCl3, 323 K): δ= 7.80 (brs, 16H; H-aryl-dppp), 7.38 (brs, 24H; H- aryl-dppp) , 7.09, 6.99, 6.90 (overlaid brs, 8H, H-3’,3’’’,4’,4’’’-Th), 2.71, 2.59, 2.47, 2.35

(overlaid brs, 32H, α-α’’’’’-CH2-Bu, α-CH2-dppp), 2.04 (brs, 4H, ß-CH2-dppp), 1.56, 1.44,

0.96 (overlaid m, 48H, ß-β’’’’’,γ-γ’’’’’-CH2-Bu), 0.77, 0.74 (overlaid m, 36H, CH3-Bu).

γ γ''''' β α' α'' α''' α'''' α α''''' 2 2' S S 2'''' Ph 5 S 5'''' Ph Ph Ph S S P P α Pt Pt P P β S S Ph Ph S Ph Ph S S

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = -6.57 (s with platinum satellites JPt,P = 2195 Hz, 4P, P-dppp). IR (KBr): 3437, 3056, 2955, 2927, 2860, 2362, 2103, 1633, 1440, 1106, 686, 512 cm-1.

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C150H168P4Pt2S6 2802.9; found 2803.3. + HRMS (ESI-FTICR) C150H168P4Pt2S6 (M ) calcd. 2801.8634; found 2801.8342.

316 Chapter 5 Conjugated macrocycles by metal template approach

Cyclo[cis-bis[3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Octabutyl-5,5’’’’’’-(diethynyl-κC2)- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’’,2’’’’’’-septithienyl][propane -1,3-diyl- bis(diphenylphosphine-κP)]diplatinum(II)] (43). The synthesis was carried out according to GP3, KOH (62.2 mg, 1.11 mmol), methanol (3 mL), protected diethynylseptithiophene 25 (135 mg, 111 µmol), tetrahydofuran (3 mL);

Pt(dppp)Cl2 (75.5 mg, 111 µmol), CuI (2.1 mg , 11.1 µmol), NEt3 (22.4 mg, 222 µmol) toluene (15 mL); methanol (8 mL); yield 97 % (180 mg, 53.6 µmol); dark red solid, m.p. >300°C (decomp.).

1 H-NMR (500 MHz, CDCl3): δ= 7.80 (brs, 16H; H-aryl-dppp), 7.37 (brs, 24H; H-aryl- dppp) , 7.07, 7.01, 6.93 (overlaid brs, 12H, H-3’,3’’’,3’’’’’,4’,4’’’,4’’’’’-Th), 2.73, 2.59,

2.46, 2.34 (overlaid brs, 36H, α-α’’’’’’’-CH2-Bu, α-CH2-dppp), 2.04 (brs, 4H, ß-CH2- dppp), 1.57, 1.45, 1.34 (m, overlaid signals, 64H, ß-β’’’’’,γ-γ’’’’’-CH2-Bu), 0.96-0.76 (m, overlaid signals, 48H, CH3-Bu).

γ γ''''''' α'' α''' α'''' α''''' β α' α'''''' α S α''''''' 2 S S 2'''''' S 5 S 5'''''' Ph Ph Ph S 2' S Ph P P α Pt Pt P P β S S Ph Ph Ph Ph S S S S S

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = -7.17 (s with platinum satellites JPt,P = 2201 Hz, 4P, P-dppp).

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C182H208P4Pt2S14 3355.1; found 3356.4.

5.8 Experimental section 317

Cyclo[trans-bis[3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Octabutyl-5,5’’’’’’-(diethynyl-κC2)- 2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’:5’’’’’,2’’’’’’-septithienyl][propane -1,3-diyl- triphenylphosphine-κP)]diplatinum(II)] (44). The synthesis was carried out according to GP3, KOH (62.2 mg, 1.11 mmol), methanol (3 mL), protected diethynylseptithiophene 25 (135 mg, 111 µmol), tetrahydofuran (3 mL);

Pt(PPh3)2Cl2 (87.4 mg, 111 µmol), CuI (2.1 mg , 11.1 µmol), NEt3 (22.4 mg, 222 µmol), toluene (15 mL); methanol (8 mL); yield 98 % (195 mg, 54.3 µmol), dark red solid, m.p. >300°C (decomp.).

1 H-NMR (500 MHz, CDCl3): δ= 7.79 (brs, 24H; H-aryl-PPh3), 7.37 (brs, 36H; H-aryl-

PPh3), 7.27, 7.08, 7.02, 6.88 (overlaid brs, 12H, H-3’,3’’’,3’’’’’,4’,4’’’,4’’’’’-Th), 2.73,

2.58, 2.50 (overlaid brs, 32H, α-α’’’’’’’-CH2-Bu), 1.57-1.38 (m, overlaid signals, 64H, ß-

β’’’’’,γ-γ’’’’’-CH2-Bu), 1.07-0.91 (m, overlaid brs, 48H, CH3-Bu).

α'''' α'''

α'' α''''' S S S

α' S S α'''''' 2 2' 2'''''' α''''''' S β S γ''''''' γ α 5 5''''''

Ph3P Pt PPh3 Ph3P Pt PPh3

S S

S S S

31 1 P NMR (202 MHz, CDCl3, H3PO4): δ = 18.7 (s with platinum satellites JPt,P = 3273 Hz, 4P, P-PPh3).

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C200H216P4Pt2S14 3579.1; found 3579.1. 318 Chapter 5 Conjugated macrocycles by metal template approach

Cyclo[bis(diyne-2,2’-(3,3’’,4,4’’-tetrabutyl-2,2’:5’,2’’-terthien-5,5’’-diyl)-1,1’-diyl)] (45) The synthesis was carried out according to GP4, platinacycle 40 (200 mg, 88.7 µmol) in THF (25 mL), iodine (44.8 mg, 177 µmol); chromatographic work-up (n- hexane/dichloromethane 9/1); yield 54 % (49.7 mg, 47.9 µmol); red microcrystalline solid, m.p.= 262°C (decomp.); HPLC (n-hexane/dichloromethane 83/17): tR = 4.3 min.

1 3 H-NMR (400MHz, CDCl3): δ = 6.84 (s, 4H; H-3’,4’-Th), 2.56 (t, J(α,β-CH2-Bu) = 7.6 Hz, 3 8H, α,α’’’-CH2-Bu), 2.45 (t, J(α’,β’-CH2-Bu) = 7.6 Hz, 8H, α’,α’’-CH2-Bu), 1.54-1.36 (m,

32H; ß-ß’’’,γ- γ’’’-CH2-Bu), 0.97 (m, 24H, CH3-Bu).

3' γ β α' 4' α'' α 2 2'' S α''' 5 S S 5''

S S S

13 C-NMR (101 MHz, CDCl3): δ = 145.1, 137.9, 137.5, 136.1 (C-2,3,4,2’,5’,2’’,3’’,4’’),

123.5 (C-3’,4’-Th), 117.9 (C5,5’’-Th), 86.9, 85.9 (C≡C), 32.4, 32.3 (ß-CH2-Bu), 28.9, 27.6

(α-CH2-Bu), 22.9, 22.6, (γ-CH2-Bu),13.9, 13.8 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 401 nm (129000).

MS (MALDI-TOF): m/z = calcd. monoisotopic mass for C64H76S6 1036.4; found 1036.7. + HRMS (ESI-FTICR) C64H76S6 (M ) calcd. 1036.4266; found 1036.4269.

5.8 Experimental section 319

Cyclo[bis(diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl)] (29a) The synthesis was carried out according to GP4, platinacycle 41 (300 mg, 107 µmol) in THF (30 mL), iodine (54.0 mg, 214 µmol); chromatographic work-up (n- hexane/dichloromethane 8/2); yield 33 mg mixture of cyclic products. After separation of the mixture by preparative HPLC the cyclic dimer was obtained by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield 12.0 % (20.4 mg, 12.8 µmol); m.p. > 300°C; HPLC (n- hexane/dichloromethane 83/17): tR = 6.8 min.

1 H-NMR (500 MHz, CDCl3): δ= 7.08 (s, 8 H, H-3’,3’’’,4’,4’’’-Th), 2.70 (m, 24 H, α-

α’’’’’-CH2-Bu), 1.60 (m, 24 H, β-β’’’’’-CH2-Bu), 1.44 (m, 24 H, β-β’’’’’, γ-γ’’’’’-CH2-

Bu), 0.99 (m, 36 H, CH3-Bu).

γ β α 5 3' 2 S S 4' S S 2'' S S 5'' S S 3''' S S 4''' 5'''' α'''''

29a C[5T-DA] 2

13 C-NMR (126 MHz, CDCl3): δ= 150.2 (C-2’,2’’’,5’,5’’’-Th), 140.3, (C-4,4’’’’-Th), 138.4, 137.5, 136.2, 133.9, 130.3 (C-2,2’’,2’’’’,3,3’’,3’’’’,4’’,5’’-Th), 125.2, 124.9 (C-

3’,3’’’,4’,4’’’-Th), 117.2 (C-5,5’’’’-Th), 81.8, 78.5 (C≡C), 32.7, 32.4 (β-β’’’’’-CH2-Bu),

29.7, 28.7, 27.6 (α-α’’’’’-CH2-Bu), 22.9, 22.7 (γ-γ’’’’’-CH2-Bu), 13.9 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 421 nm (118800).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C96H116S10 1588.6; found 1588.3 [M+]. 320 Chapter 5 Conjugated macrocycles by metal template approach

+ HRMS (ESI-FTICR) C96H116S10 (M ) calcd. 1588.6278; found 1588.6244.

Cyclo[tris(diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl)] (29b) The compound was obtained as byproduct from the synthesis of the corresponding dimer 29a and was isolated after separation by preparative HPLC followed by precipitation upon addition of methanol to a concentrated dichloromethane solution as a red microcrystalline solid; yield 4.0 % (10.2 mg, 4.28 µmol); m.p. > 300°C; HPLC (n-hexane/dichloromethane

83/17): tR = 7.4 min.

1 3 H-NMR (400 MHz, CDCl3): δ= 7.10 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.49 Hz, 6 H, H- 3 3’,4’’’-Th), 7.08 (d, J(H-3’,4’-Th); (H-3’’’,4’’’-Th) = 3.49 Hz, 6 H, H-3’’’,4’- Th), 2.73 (m, 36 H,

α-α’’’’’-CH2-Bu), 1.61 (m, 36 H, β-β’’’’’-CH2-Bu), 1.45 (m, 72 H, γ-γ’’’’’-CH2-Bu), 0.98

(m, 54 H, CH3-Bu).

4' 3''' 3' 4''' γ β S 5'' α S 2'' S 2 α''''' S S 5'''' 5

S S

S S S S

29b C[5T-DA] 3

5.8 Experimental section 321

13 C-NMR (101 MHz, CDCl3): δ= 151.0 (C-2’,2’’’,5’,5’’’-Th), 140.4, (C-4,4’’’’-Th), 138.5, 136.8, 135.7, 133.6, 130.0 (C-2,2’’,2’’’’,3,3’’,3’’’’,4’’,5’’-Th), 126.2, 125.7 (C-

3’,3’’’,4’,4’’’-Th), 116.6 (C-5,5’’’’-Th), 81.4, 78.0 (C≡C), 32.8, 32.6 (β-β’’’’’-CH2-Bu),

27.9, 27.7 (α-α’’’’’-CH2-Bu), 23.0, 22.9, 22.7 (γ-γ’’’’’-CH2-Bu), 13.9 (CH3-Bu).

UV/VIS (dichloromethane): λmax (ε) = 425 nm (205200).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C144H174S15 2382.9; found 2383.4 [M+].

Cyclo[tetrakis(diyne-2,2’-(3,3’’,3’’’’,4,4’’,4’’’’-Hexabutyl-2,2’:5’,2’’:5’’,2’’’:5’’’,2’’’’- quinquethien-5,5’’’’-diyl)-1,1’-diyl)] (29c) The compound was obtained as byproduct from the synthesis of the corresponding dimer 29a and was isolated after separation by preparative HPLC as a red microcrystalline solid; yield < 0.4 % (1.36 mg, 0.43 µmol); m.p. > 300°C; HPLC (n-hexane/dichloromethane

83/17): tR = 9.4 min.

UV/VIS (dichloromethane): λmax (ε) = 436 nm (235200).

MS (MALDI-TOF) m/z: calcd. monoisotopic mass for C212H232S20 3177.3; found 3178.1 [M+].

Mixture of Cyclo[ (diyne-2,2’-(3,3’’,3’’’’,3’’’’’’,4,4’’,4’’’’,4’’’’’’-Hexabutyl-2,2’:5’,- 2’’:5’’,2’’’:5’’’,2’’’’ : 5’’’’’,2’’’’’’-septithien-5,5’’’’’’-diyl)-1,1’-diyl)] (46a-c) The synthesis was carried out according to GP4, platinacycle 43 (85.1 mg, 25.3 µmol) in THF (6 mL), iodine (12.8 mg, 50.6 µmol); chromatographic work-up (n- hexane/dichloromethane 7/3). After evaporation of the solvent a mixture of cyclic products 46a-c (11.5 mg, 22.0 % yield based on 43) was isolated as a red microcrystalline solid. The NMR, analytical HPLC and MALDI-TOF MS measurements proved the formation of the cyclic dimer and trimer in ratio of 2:1, and traces of the tetramer. No separation of the mixture was carried out.

1 H-NMR (400 MHz, CDCl3): no typical resonance signals for terminal acetylenic protons at ca. 3.5 ppm were detected.

HPLC (n-hexane/dichloromethane 80/20): 46a tR (λmax) = 5.5 min (416 nm); 46b tR (λmax)

= 6.75 (427 nm); 46c tR (λmax) = 10.5 (430 nm). 322 Chapter 5 Conjugated macrocycles by metal template approach

MS (MALDI-TOF) m/z: calcd. average mass for C128H154S14 (46a) 2143.6; found 2144.8; for C192H234S21 (46b) 3215.3; found 3216.8; no signal for 46c was detected.

Cyclo[8]thiophene (47) The synthesis was carried out by two different methods: A) according to GP5, macrocycle 45 (53 mg, 51 µmol) and sodium sulfide nonahydrate (228 mg, 1.02 mmol), 2-methoxyethanol (30 ml), xylene (10 ml); yield 19 % (10.1 mg, 9.14 µmol), red solid; m.p. 221°C (decomp).

B) A solution of macrocycle 45 (30 mg, 29 µmol), sodium hydrogen sulfide hydrate (90 mg), potassium hydroxide (22 mg, 290 µmol) in dimethoxysulfoxide (20 mL) was stirred at 80°C for 4 hours. After removal of the solvent by vacuum distillation, water and dichloromethane were added. Organic layer was separated and then washed several times with water and dried over Na2SO4. After evaporation of the solvent the crude product was purified by column chromatography on silica gel with n-hexane/dichloromethane (4/1) as eluent. After evaporation of the solvent the product was precipitated upon addition of methanol to a concentrated dichloromethane solution. The pure product 47 was isolated as red microcrystalline solid in 36 % yield (11.5 mg, 10.4 µmol); m.p. 221 (decomp); HPLC

(n-hexane/dichloromethane 79/21): tR [min] / λmax [nm] = 5.1 min / 396.

1 3 H-NMR (500 MHz, CDCl3, 306K) δ= 6.87 (s, 8H, H-3’,4’-Th), 2.61 (t, J(α,β-CH2-Bu) = 8.1

Hz, 16H, α -CH2-Bu), 1.53 (m, 16H, ß-CH2-Bu), 1.47 (m, 16H; γ,-CH2-Bu), 0.97, (t, 12H, 3 J(γ-CH2-Bu, CH3-Bu) = 7.3 Hz, CH3-Bu).

5.8 Experimental section 323

γ β α 3 4 3' 2' 2 S 5 4' S S 5'

S S

S S S

13 C-NMR (125 MHz, CDCl3, 306K): δ = 139.2 (C-3,4), 137.8, 131.5 (C2, C2’), 122.8 (C-

3’,4’), 32.8, (ß-CH2-Bu), 27.9 (α-CH2-Bu), 23.0 (γ-CH2-Bu),13.9 (CH3-Bu).

MS (MALDI-TOF): calcd. monoisotopic mass for C64H80S8 m/z = 1104.4; found 1104.5.

Cyclo[12]thiophene (48) The synthesis was carried out according to GP5, macrocycle 29a (30 mg, 18.8 µmol) and sodium sulfide nonahydrate (90.3 mg, 376 µmol), 2-methoxyethanol (15 ml), xylene (5 ml); yield 31 % (9.6 mg, 5.82 µmol), red solid; m.p. > 250°C; HPLC (n- hexane/dichloromethane 75/15): tR [min] / λmax [nm] = 4.9 min / 392.

1 3 H-NMR (400 MHz, CDCl3) δ= 7.06 (s, 12H, H-3’,4’-Th), 2.70 (t, J(α,β-CH2-Bu) = 8.0 Hz, 3 24H, α -CH2-Bu), 1.42 (m, 48H, ß,γ-CH2-Bu), 0.95, (t, J(γ-CH2-Bu, CH3-Bu) = 7.3 Hz, 36H,

CH3-Bu).

324 Chapter 5 Conjugated macrocycles by metal template approach

γ β α 3' 3 4 4' 2' 2 S 5 S 5' S S S

S S

S S S

13 C-NMR (101 MHz, CDCl3): δ = 140.3 (C-3,4), 136.6, 130.1 (C2, C2’), 125.4 (C-3’,4’),

32.8, (ß-CH2-Bu), 27.8 (α-CH2-Bu), 23.0 (γ-CH2-Bu),13.9 (CH3-Bu), m.p > 250°C.

MS (MALDI-TOF): calcd. monoisotopic mass for C96H120S12 m/z = 1656.6; found 1656.5. The analytical data are consistent with those given in the literature.47a,50

Cyclo[16]thiophene (49) The synthesis was carried out according to GP5, macrocycle 27b (20 mg, 9.64 µmol) and sodium sulfide nonahydrate (86.2 mg, 385 µmol), 2-methoxyethanol (10 ml), xylene (4 ml); yield 8 % (1.7 mg, 0.77 µmol), red solid; m.p. > 250°C. 1 3 H-NMR (400 MHz, CDCl3) δ= 7.09 (s, 12H, H-3’,4’-Th), 2.69 (t, J(α,β-CH2-Bu) = 7.9 Hz, 3 32H, α -CH2-Bu), 1.45 (m, 64H, ß,γ-CH2-Bu), 0.95, (t, J(γ-CH2-Bu, CH3-Bu) = 7.3 Hz, 48H,

CH3-Bu).

5.8 Experimental section 325

γ β α 3' 4' 3 4 2' 5' 2 S 5 S S S S

S S

S S S S S

13 C-NMR (101 MHz, CDCl3): δ = 140.3 (C-3,4), 136.1, 130.1 (C2, C2’), 125.6 (C-3’,4’),

32.8, (ß-CH2-Bu), 27.9 (α-CH2-Bu), 23.0 (γ-CH2-Bu),13.9 (CH3-Bu), m.p > 250°C.

MS (MALDI-TOF): calcd. monoisotopic mass for C128H160S16 m/z = 2208.8; found 2209.8.

UV/VIS (dichloromethane): λmax = 414 nm. The analytical data are consistent with those given in the literature.47a,50

Cyclo[18]thiophene (50) The synthesis was carried out according to GP5, macrocycle 29b (20 mg, 8.4 µmol) and sodium sulfide nonahydrate (56.4 mg, 252 µmol), 2-methoxyethanol (10 ml), xylene (3 ml); yield 26 % (5.43 mg, 2.18 µmol), red solid; m.p. 211°C. 1 3 H-NMR (400 MHz, CDCl3) δ= 7.09 (s, 18H, H-3’,4’-Th), 2.74 (t, J(α,β-CH2-Bu) = 7.9 Hz, 3 36H, α -CH2-Bu), 1.54 (m, 36H, ß-CH2-Bu), 1.47 (m, 36H, γ-CH2-Bu), 0.98, (t, J(γ-CH2-Bu,

CH3-Bu) = 7.3 Hz, 54H, CH3-Bu).

326 Chapter 5 Conjugated macrocycles by metal template approach

γ β α 3' 3 4 4' 2' 5' 2 S 5 S S S S S S

S S

S S S S S S S

13 C-NMR (101 MHz, CDCl3): δ = 140.1 (C-3,4), 136.1, 130.0 (C2, C2’), 125.7 (C-3’,4’),

32.8, (ß-CH2-Bu), 27.9 (α-CH2-Bu), 23.1 (γ-CH2-Bu),13.9 (CH3-Bu).

MS (MALDI-TOF): calcd. monoisotopic mass for C144H180S18 m/z = 2484.9; found 2485.2.

UV/VIS (dichloromethane): λmax (ε) = 421 nm (132000). The analytical data are consistent with those given in the literature.47a,50

5.7.3 Qualitative and Preparative Electrolysis Experiments

Small-scale electrolysis of acyclic Pt(II) complex 35 has been performed in a cyclic voltammetric apparatus. A dichloromethane solution of 35 (5 x 10-3 M) was exposed to a constant applied potential (Eappl = 0.95 V) for 12 h in presence of 0.1 M Bu4NPF6 as the supporting electrolyte. The consumption of the substrate was monitored by cyclic voltammetry. After work-up, the analysis of the crude product by GC-MS revealed the formation of the desired coupling product 15.

5.8 Experimental section 327

Preparative electrolysis of platinacycle 40 at constant working electrode potential has been performed in the same cell used in the cyclic voltammetry experiments. A solution of 40 (50 mg, 22.2 µmol) in benzonitrile (10 mL) was electrolyzed at a constant applied potential

(Eappl = 0.85 V) in presence of 0.1 M Bu4NPF6 (297 mg) as the supporting electrolyte. The consumption of the substrate was monitored by CV. After 48 h the solution was concentrated the solvent being removed in vacuo. The residue was extracted with dichloromethane. The extract was filtered, and the solvent removed. The cyclodimeric terthiophene-diacetylene 45 was obtained in 6 % yield (1.4 mg, 1.35 µmol) after purification by chromatographic work-up with hexane/dichloromethane 9/1 as eluent. The structure was unambiguously proven by 1H NMR and MALDI-TOF MS investigations and the analytical data are identical with those outlined above for this compound.

5.9 Supplement-

5.8.1 MALDI-TOF MS spectra of cyclo[n]thiophenes 48-50

1656 100

60 / a.u.

I 40

0 1000 1500 2000 2500 m / z

Figure 5.70. MALDI-TOF MS spectrum of cyclo[12]thiophene 48.

328 Chapter 5 Conjugated macrocycles by metal template approach

100 90 80 70 60 50 40 30 20 10 0 1.655 1.660 1.665

100

0 1654 1656 1658 1660 1662 1664 1666 1668 m / z

Figure 5.71. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of cyclo[12]thiophene 48.

2210 100

60 / a.u. I 40

0 1500 1750 2000 2250 2500 2750 3000 m / z

Figure 5.72. MALDI-TOF MS spectrum of cyclo[16]thiophene 49.

5.9 Supplement 329

100 90 80 70 60 50 40 30 20 10 0 2.210 2.215 2.220

100

0 2208 2212 2216 2220 2224 m / z

Figure 5.73. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of cyclo[16]thiophene 49.

2485 100

60 / a.u. I 40

0 2000 2200 2400 2600 m / z

Figure 5.74. MALDI-TOF MS spectrum of cyclo[18]thiophene 50. 330 Chapter 5 Conjugated macrocycles by metal template approach

g 100 90 80 70 60 50 40 30 20 10 0 2.485 2.490 2.495

100

0 2482 2484 2486 2488 2490 2492 2494 2496 2498 m / z

Figure 5.75. Calculated (top) and measured (bottom) isotropic distribution of the mass signal of cyclo[18]thiophene 50.

5.8.2. X-ray crystallographic data of cyclodimeric terthiophene-diacetylene 45

Diffraction data of macrocycle 45 were collected on a STOE-IPDS image-plate diffractometer (MOKα radiation, graphite monochromator) in the Φ-rotation mode. The structures were solved by direct methods with the XMY 93 program system63 and subjected to full-matrix refinement with SHELXL93.64 Non-hydrogen atoms were refined anisotropically, and hydrogen atoms were added in calculated positions.

Crystallographic data for the structure reported here have been deposited with the Cambridge Crystallographic Data Centre as supplementary publication no. CCDC 201462. Copies of the data can be obtained free of charge on application to CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44-1223/336-033; e-mail: [email protected]).

5.9 Supplement 331

Table 14. Crystal data and structure refinement for cyclodimeric terthiophene-diacetylene 45.

Empirical formula C64 H76 S6

Formula weight 1037.61

Temperature 223(2) K

Wavelength 0.71073 A

Crystal system, space group monoclinic, P21/a

Unit cell dimensions a = 9.0736(13) A α = 90 °. b = 20.222(2) A ß = 104.872(15)° c = 16.503(2) A γ = 90°

Volume 2926.6(6) A3

Z, Calculated density 2, 1.177 Mg/m3

Absorption coefficient 0.272 mm-1

F(000) 1112

Crystal size 0.15 x 0.23 x 0.38 mm

θ range for data collection 2.38 - 25.9°.

Limiting indices -11<=h<=11, -24<=k<=24, -20<=l<=20

Reflections collected / unique 22852 / 5347

[R(int) (F2) 0.0529

Completeness to θ = 25.97 93.2 %

Absorption correction None

Refinement method Full-matrix least-squares on F2

Data / restraints / parameters 5347 / 0 / 468

Goodness-of-fit on F2 0.945

Final R indices [I>2σ(I)] R1 = 0.0331, wR2 = 0.0755

R indices (all data) R1 = 0.0514, wR2 = 0.0798

Largest diff. peak and hole 0.539 and -0.186 eA-3

332 Chapter 5 Conjugated macrocycles by metal template approach

Table 5.15. Atomic coordinates ( x 104) and equivalent isotropic displacement parameters 2 3 (A x 10 ) for 45. U(eq) is defined as one third of the trace of the orthogonalized Uij tensor.

x y z U(eq)

S(1) 12354(1) 3890(1) 1903(1) 30(1) S(2) 9381(1) 3189(1) 834(1) 30(1) S(3) 7199(1) 3382(1) -899(1) 31(1) C(2) 13659(2) 3447(1) 3383(1) 26(1) C(7) 8631(2) 2491(1) 263(1) 26(1) C(12) 5529(2) 2462(1) -1749(1) 28(1) C(4) 12676(2) 2933(1) 2962(1) 25(1) C(3) 11893(2) 3102(1) 2161(1) 26(1) C(11) 5803(2) 3131(1) -1763(1) 28(1) C(17) 14672(2) 5189(1) 2934(1) 31(1) C(13) 5265(2) 3643(1) -2351(1) 30(1) C(9) 7450(2) 2585(1) -507(1) 26(1) C(102) 7470(2) 979(1) -1159(2) 34(1) C(21) 14573(2) 3413(1) 4274(1) 32(1) C(10) 6465(2) 2148(1) -1025(1) 26(1) C(1) 13592(2) 3998(1) 2878(1) 27(1) C(101) 6396(2) 1424(1) -824(1) 30(1) C(22) 13618(3) 3562(1) 4902(2) 44(1) C(5) 10764(2) 2738(1) 1535(1) 26(1) C(15) 5167(2) 4174(1) -2689(1) 32(1) C(19) 14306(2) 4625(1) 3016(1) 29(1) C(41) 12537(2) 2265(1) 3340(1) 28(1) C(43) 13362(3) 1061(1) 3382(2) 40(1) C(8) 9403(2) 1942(1) 632(1) 32(1) C(42) 13688(2) 1766(1) 3167(1) 31(1) C(103) 7479(3) 276(1) -840(2) 47(1) C(122) 4931(3) 2056(1) -3248(1) 44(1) C(6) 10587(2) 2080(1) 1344(1) 33(1) C(23) 12008(3) 4394(2) 5423(2) 58(1) C(124) 3224(3) 1095(1) -3883(2) 57(1) C(121) 4389(2) 2126(1) -2451(1) 34(1) C(123) 3713(3) 1789(1) -3994(2) 54(1) C(24) 13222(3) 4280(1) 4936(2) 50(1) 5.9 Supplement 333

C(44) 14408(3) 567(1) 3118(2) 52(1) C(104) 8627(4) -169(1) -1099(2) 64(1)

Table 5.16. Bond lengths [Å] and angles [°] for 45. ______

S(1)-C(1) 1.7235(19) S(1)-C(3) 1.7287(17) S(2)-C(5) 1.7316(18) S(2)-C(7) 1.7348(17) S(3)-C(11) 1.7227(19) S(3)-C(9) 1.7293(16) C(2)-C(1) 1.383(2) C(2)-C(4) 1.429(2) C(2)-C(21) 1.494(3) C(7)-C(8) 1.370(2) C(7)-C(9) 1.448(3) C(12)-C(11) 1.377(2) C(12)-C(10) 1.425(3) C(12)-C(121) 1.503(3) C(4)-C(3) 1.373(3) C(4)-C(41) 1.506(2) C(3)-C(5) 1.454(2) C(11)-C(13) 1.417(2) C(17)-C(19) 1.205(2) C(17)-C(15)#1 1.368(2) C(13)-C(15) 1.203(2) C(9)-C(10) 1.386(2) C(102)-C(103) 1.517(3) C(102)-C(101) 1.530(3) C(21)-C(22) 1.540(3) C(10)-C(101) 1.506(2) C(1)-C(19) 1.416(2) C(22)-C(24) 1.501(3) C(5)-C(6) 1.368(2) C(15)-C(17)#1 1.368(2) C(41)-C(42) 1.531(3) C(43)-C(44) 1.516(3) 334 Chapter 5 Conjugated macrocycles by metal template approach

C(43)-C(42) 1.517(3) C(8)-C(6) 1.402(3) C(103)-C(104) 1.518(3) C(122)-C(121) 1.524(3) C(122)-C(123) 1.526(3) C(23)-C(24) 1.538(3) C(124)-C(123) 1.497(3) C(1)-S(1)-C(3) 91.62(8) C(5)-S(2)-C(7) 92.96(8) C(11)-S(3)-C(9) 91.94(8) C(1)-C(2)-C(4) 111.58(16) C(1)-C(2)-C(21) 123.93(16) C(4)-C(2)-C(21) 124.38(16) C(8)-C(7)-C(9) 132.49(16) C(8)-C(7)-S(2) 109.41(14) C(9)-C(7)-S(2) 117.93(12) C(11)-C(12)-C(10) 112.36(15) C(11)-C(12)-C(121) 121.50(16) C(10)-C(12)-C(121) 126.11(15) C(3)-C(4)-C(2) 113.05(15) C(3)-C(4)-C(41) 122.93(16) C(2)-C(4)-C(41) 124.00(16) C(4)-C(3)-C(5) 130.89(15) C(4)-C(3)-S(1) 111.59(13) C(5)-C(3)-S(1) 117.51(13) C(12)-C(11)-C(13) 134.22(17) C(12)-C(11)-S(3) 111.91(13) C(13)-C(11)-S(3) 113.72(13) C(19)-C(17)-C(15)#1 167.7(2) C(15)-C(13)-C(11) 160.28(19) C(10)-C(9)-C(7) 132.24(15) C(10)-C(9)-S(3) 111.16(13) C(7)-C(9)-S(3) 116.58(12) C(103)-C(102)-C(101) 111.91(17) C(2)-C(21)-C(22) 112.78(16) C(9)-C(10)-C(12) 112.62(15) C(9)-C(10)-C(101) 122.76(17) C(12)-C(10)-C(101) 124.61(16) C(2)-C(1)-C(19) 132.61(17) 5.9 Supplement 335

C(2)-C(1)-S(1) 112.15(13) C(19)-C(1)-S(1) 115.21(13) C(10)-C(101)-C(102) 115.20(15) C(24)-C(22)-C(21) 113.30(18) C(6)-C(5)-C(3) 132.47(16) C(6)-C(5)-S(2) 109.64(14) C(3)-C(5)-S(2) 117.81(12) C(13)-C(15)-C(17)#1 167.4(2) C(17)-C(19)-C(1) 163.21(19) C(4)-C(41)-C(42) 112.38(15) C(44)-C(43)-C(42) 112.23(18) C(7)-C(8)-C(6) 114.00(16) C(43)-C(42)-C(41) 113.18(16) C(102)-C(103)-C(104) 113.6(2) C(121)-C(122)-C(123) 113.7(2) C(5)-C(6)-C(8) 113.91(16) C(12)-C(121)-C(122) 113.38(16) C(124)-C(123)-C(122) 114.1(2) C(22)-C(24)-C(23) 112.1(2) ______

Table 4. Anisotropic displacement parameters (A2 x 103) for 45. ______U11 U22 U33 U23 U13 U12 ______S(1) 36(1) 27(1) 24(1) 3(1) 4(1) -1(1) S(2) 34(1) 26(1) 26(1) -1(1) 2(1) 2(1) S(3) 34(1) 25(1) 29(1) 0(1) 2(1) -1(1) C(2) 24(1) 30(1) 24(1) 1(1) 7(1) 5(1) C(7) 28(1) 27(1) 25(1) -2(1) 9(1) -1(1) C(12) 26(1) 31(1) 28(1) -1(1) 9(1) -1(1) C(4) 25(1) 27(1) 24(1) 2(1) 8(1) 3(1) C(3) 27(1) 26(1) 26(1) 1(1) 9(1) 2(1) C(11) 28(1) 31(1) 25(1) 0(1) 5(1) -1(1) C(17) 33(1) 33(1) 25(1) 1(1) 1(1) 1(1) C(13) 30(1) 31(1) 26(1) -2(1) 3(1) -2(1) C(9) 27(1) 25(1) 26(1) 1(1) 8(1) 2(1) 336 Chapter 5 Conjugated macrocycles by metal template approach

C(102) 39(1) 31(1) 33(1) -2(1) 10(1) 0(1) C(21) 30(1) 37(1) 27(1) 2(1) 2(1) 4(1) C(10) 25(1) 27(1) 28(1) 0(1) 9(1) 1(1) C(1) 28(1) 29(1) 25(1) -1(1) 6(1) 2(1) C(101) 29(1) 27(1) 33(1) 1(1) 8(1) -3(1) C(22) 50(1) 55(1) 28(1) 2(1) 8(1) 8(1) C(5) 29(1) 30(1) 22(1) 2(1) 8(1) 2(1) C(15) 30(1) 33(1) 28(1) -2(1) 1(1) -2(1) C(19) 30(1) 33(1) 24(1) 0(1) 4(1) 2(1) C(41) 29(1) 31(1) 23(1) 5(1) 7(1) -1(1) C(43) 45(1) 31(1) 47(2) 8(1) 17(1) 2(1) C(8) 37(1) 25(1) 32(1) -2(1) 4(1) -1(1) C(42) 28(1) 31(1) 34(1) 7(1) 7(1) 0(1) C(103) 61(2) 31(1) 51(2) 3(1) 17(1) 7(1) C(122) 54(1) 44(1) 33(1) -1(1) 8(1) -14(1) C(6) 35(1) 29(1) 32(1) 3(1) 3(1) 5(1) C(23) 56(2) 79(2) 43(2) -11(1) 17(1) 14(1) C(124) 59(2) 51(1) 54(2) -12(1) 1(1) -12(1) C(121) 30(1) 35(1) 33(1) 0(1) 3(1) -4(1) C(123) 72(2) 46(1) 35(2) -3(1) -3(1) -4(1) C(24) 58(1) 49(1) 43(2) -6(1) 14(1) 0(1) C(44) 58(2) 38(1) 64(2) 9(1) 22(1) 12(1) C(104) 79(2) 43(1) 70(2) -2(1) 19(2) 22(1)

5.10. References and notes

1 For a selection of the most recent examples: a) C. Grave, D. Lentz, A. Schäfer, P. Samori, J.P. Rabe, P. Franke, A.D. Schlüter, J. Am. Chem. Soc. 2003, 125, 6907-6918; b) M Fischer, G. Lieser, A. Rapp, I. Schnell, W. Mamdouh, S. de Feyter, F.C. Schryver, S. Höger, J. Am. Chem. Soc. 2004, 126, 214-222.c) K. Campbell, R. Mc.Donald, M.J. Ferguson, R.R Tykwinski, Organometallics, 2003, 22, 1353-1355; d) M Srinivasan, S. Sankararaman, H. Hopf, B. Varghese, Eur. J. Org. Chem. 2003; 660-665; e) X. Shen, D.M. Ho, R.A. Jr. Pascal, Org. Lett. 2003, 5, 369-371; f) M. Higuchi, H. Kanazawa, K. Yamamoto, Org. Lett. 2003, 5, 345-347; g) P. Bäuerle, E. Mena-Osteritz, G. Fuhrmann, A. Kaiser, M. Ammann, Polym. Mater. Sci. Eng. 2002, 86, 34; h) P.N.W. Baxter, Chem. Eur. J. 2002, 8, 5250-5264; i) M. Schmittel, A. Ganz, D. Fenske, Org. Lett. 2002, 4, 2289-2292. 5.10 References and notes 337

2 a) J.-M. Lehn, Science 1985, 227, 849-856; b) M. Mascal, Contemp. Org. Synth. 1994, 1, 31-46; c) J.-M. Lehn, Supramolecular Chemistry, VCH, Weinheim, 1995; d) Comprehensive Supramolecular Chemistry, Eds.: J.L. Atwood, J.E.D. Davies, D.D. Macniol, F. Vögtle, Pergamon, Oxford, 1996; e) H.-J. Schneider, A. Yatsimirsky, Principle and Methods in Supramolecular Chemistry, Wiley, Chiester, 2000; f) G.R. Desiraju, Nature 2001, 412, 397-400.

3 For recent reviews, see for example: a) M Fujita, Chem. Soc. Rev. 1998, 27, 417-426; b) S. Leininger, B. Olenyuk, P.J. Stang, Chem. Rev. 2000, 100, 853-908; c) B.J. Holliday, C.A. Mirkin, Angew. Chem. Int. Ed. Engl. 2001, 40, 2023-2043.

4 a) R. Faust, F. Diederich, V. Gramlich, P. Seiler, Chem. Eur. J. 1995, 1, 111-117; b) J.D. Brandsaw, L. Guo, C.A. Tessier, W.J. Youngs, Organometallics 1996, 15, 2582-2584; c) D. Zhang, D.. McConville, C.A. Tessier, W.J. Youngs, Organometallics 1997, 16, 824-825; d) S.M. AlQaisi, K.J. Galat, M. Chai, D.G. Ray, P.L Rinaldi, C.A Tessier, W.J. Youngs, J. Am. Chem. Soc. 1998, 120, 12149-12150; e) J.J. Pak, T.J.R. Weakley, M.M. Haley, Organometallics 1997, 16, 4505-4507; f) E. Bosch, C.L. Barnes, Organometallics 2000, 19, 5522-5524; g) S.J. Lee, A. Hu, W. Lin, J. Am. Chem. Soc. 2002, 124, 12948-12949; h) K. Campbell, R. McDonald, M.J. Ferguson, R.R. Tykwinski, Organometallics 2003, 22, 1353-1355.

5 a) Inclusion, Phenomena, and Molecular Recognition (Eds.: J.L. Atwood), Plenum, New York, 1990; b) F. Vögtle, Supramolecular Chemistry, Wiley, Chichester, 1991; c) H.-J. Schneider, H. Dürr, Frontiers in Supramolecular Organic Chemistry and Photochemistry, VCH, New York, 1991; d) J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives, VCH, Weinheim, 1995; e) C.S. Lent, Science 2000, 288, 1597-1599.

6 a) J.-M. Lehn, Supramolecular Chemistry: Concepts and Perspectives, VCH, Weinheim, 1995, Chapter 9, p.139; b) Comprehensive, Supramolecular Chemistry (Eds.: J.L. Atwood, J.E.D. Davies, D.D. MacNicol, F. Vögtle, J.-M. Lehn) Pergamon, Oxford, 1996.

7 a) M. Fujita, J. Yazaki, K. Ogura, J. Am. Chem. Soc. 1990, 112, 5645-5647; b) P.J. Stang, D.H. Cao, J. Am. Chem. Soc. 1994, 116, 4981-4982; c) R.V. Slone, D.I. Yoon, R.M. Calhoun, J.T. Hupp, J. Am. Chem. Soc. 1996, 117, 11813-11814; d) M. Fujita, K. Ogura, Chem. Rev. 1996, 148, 249-264; e) S.B. Lee, S. Hwang, D.S. Chung, H. Yun, J.-I. Hong, Tetrahedron Lett. 1998, 39, 873-876; f) K.D. Benkstein, J.T. Hupp, C.L. Stern, J. Am. Chem. Soc. 1999, 120, 12982-12983; g) F. Würthner, A. Sautter, Chem. Commun. 2000, 445- 446; h) S.-S. Sun, A.J. Lees, J. Am. Chem. Soc. 2000, 122, 8956-8967.

8 a) C.M. Drain, J.-M. Lehn, J. Chem. Soc. Chem. Commun. 1994, 2313-2315; b) T. Habicher, J.-F. Nierengarten, V. Gramlich, F. Diederich, Angew. Chem. Int. Ed. Engl. 1998, 37, 1916-1919; c) M. Schmitz, S. Leininger, J. Fan, A.M. Arif, P.J. Stang, Organometallics, 1999, 18, 4817-4824; d) J. Fan, J.A. Whiteford, B. Olenyuk, M.D. Levin, P.J. Stang, E.B. Fleischer, J. Am. Chem. Soc. 1999, 121, 2741-2752.

9 M. Fujita, S. Nagao, M. Iida, K. Ogata, K. Ogura, J. Am. Chem. Soc. 1993, 115, 1574-1576; b) M. Fujita, F. Ibukuro, H. Hagihara, K. Ogura, Nature 1994, 367, 720-723; c) M. Fujita, F. Ibukuro, H. Seki, O. Kamo, M. Imanari, K. Ogura, J. Am. Chem. Soc. 1996, 118, 899-900; d) M. Fujita, M. Aoyagi, F. Ibukuro, K. Ogura, K. Yamaguchi, J. Am. Chem. Soc. 1998, 120, 611-612; e) M. Fujita, Acc. Chem. Res. 1999, 32, 53-61.

10 a) K. Sonogashira, T. Yatake, Y Tohda, S. Takahashi, N. Hagihara, J.Chem. Soc., Chem. Commun. 1977, 291-292; b) K. Sonogashira, Y Fujikura, T. Yatake, N. Toyoshima, S. Takahashi, N. Hagihara, J. Organomet. Chem. 1978, 145, 101-108.

11 K. Osaka, R. Sakata, T. Yamamoto, Organometallics 1997, 16, 5354-5364.

12 a) Y. Fujikura, K. Sonogashira, N. Hagihara, Chem. Lett. 1975, 1067-1070; b) K. Sonogashira, S. Takahashi, N. Hagihara, Macromolecules 1977, 10, 879-880. 338 Chapter 5 Conjugated macrocycles by metal template approach

13 N.J.Long, C. K. Williams, Angew. Chem. Int. Ed. 2003, 42, 2586-2617.

14 a) L.K: Myers, D.M. Ho, M.E. Thomson, C. Langhoff, Polyhedron 1995, 14, 57-67; b) P. Nguyen, G. Lesley, T.B. Marder, I. Ledoux, J. Zyss, J. Chem. Mater., 1997, 9, 406-408.

15 a) M.S. Khan, S.J. Davies, A. K. Kakkar, D. Schwartz, B. Lin, B.F.G. Johnson, J. Lewis, J. Organomet. Chem. 1992, 424, 87-97; b) K. Onitsuka, Y. Harada, F. Takei, S. Takahashi, Chem. Commun. 1998, 643-645.

16 F. Paul, C. Lapinte, Coord. Chem. Rev. 1998, 178-180.

17 S. Takahashi, H. Takati, K. Morimoto, K. Sonogashira, N. Hagihara, Mol. Cryst. Liq. Cryst. 1982, 32, 139- 142.

18 a) J. Lewis, M. S. Khan, A.K. Kakkar, B.F.G. Johnson, T.B. Marder, H.B. Fyfe, F. Wittmann, R.H. Friend, A.E. Dray, J. Organomet. Chem. 1992, 425, 165-176; b) N. Chawdurry, A. Köhler, R.H. Friend, M. Younus, N.J. Long, P.R. Raithby, J. Lewis, Macromolecules 1998, 31, 722-727; c) C.J. Adams, S.L. James, X. Liu, P.R. Raitby, L.J. Yellowlees, J. Chem. Soc. Dalton Trans. 2000, 63-67; d) N.J. Long, A.J.P. White, D.J. Williams, M. Younus, J. Organomet. Chem. 2002, 649, 94-99.

19 M. Karikoni, C. Kitamura, S. Tanaka, Y. Yamashita, J. Am. Chem. Soc. 1995, 117, 6791-6792

20 W.-Y. Wong, K.-H. Choi, G.-L. Lu, J.-X Shi, Macromol. Rapid Commun., 2001, 22, 461-465.

21 a) B.F.G. Johnson, A.K. Kakkar, M.S. Khan, J. Lewis, A.E. Dray, R.H. Friend, F. Wittmann, J. Mater. Chem. 1991, 1, 495-498; b) S.J. Davies, B.F.G. Johnson, M.S. Khan, J. Lewis, J. Chem. Soc. Chem.

Commun. 1991, 187-188; c) M.S. Khan, S.J. Davies, A. K. Kakkar, D. Schwartz, B. Lin, B.F.G. Johnson, J. Lewis, J. Organomet. Chem. 1992, 424, 87-97.

22 J. Lewis, N.J. Long, P.R. Raithby, G.P. Shields, W.-Y. Wong, M. Younus, J. Chem. Soc., Dalton Trans., 1997, 4283-4288.

23 P. Altamura, G. Giardina, C.L. Sterzo, M.V. Russo, Organometallics 2001, 20, 4360-4368.

24 a) H. Jiang, W. Lin, J. Am. Chem. Soc. 2003, 125, 8084-8085; b) H. Jiang, A.Hu, W. Lin, Chem. Commun. 2003, 96-97; c) J. Hua, W. Lin, Org. Lett. 2004, 6, 861-864.

25 a) J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles and Applications of Organotransition Metal Chemistry; University Science Books, Mill Valley, CA, 1987; b) J.D. Atwood, Inorganic and Organometallic Reaction Mechanisms, 2nd ed. Wiley-VCH, new York, 1997; c) R.H. Crabtree The Organometallic Chemistry of the Transition Metals, 3rd ed. John Wiley & Sons, New York 2001.

26 A. Gillie and J.K. Stille, J. Am. Chem. Soc. 1980, 102, 4933-4941.

27 a) F. Ozawa in Inorganic Chemistry, 3. Fundamentals of Molecular Catalysis; Eds.: H. Kurosawa, A. Yamamoto; Elsevier Science: Amsterdam, 2003; p. 479; b) S. Komija, Y. Abe, A. Yamamoto, T. Yamamoto, Organometallics 1983, 2, 1466-1468.

28 S. Komija, T.A. Albright, J.K. Kochi, J. Am. Chem. Soc. 1976, 98, 7255-7265; b) F. Ozawa, T. Ito, Y. Nakamura, A. Yamamoto, Bull. Chem. Soc. Jpn. 1981, 54, 1868-1880.

29 a) P.S. Braterman, R.J. Cross, G.B. Young, J. Chem. Soc., Dalton Trans. 1976, 1306-1309; b) J. Chem. Soc., Dalton Trans. 1976, 1310-1314; c) P.S. Braterman, R.J. Cross, G.B. Young J. Chem. Soc., Dalton Trans. 1977, 1892-1897.

30 a) B.L. Shaw, R.E. Stainbank, J. Chem. Soc., Dalton Trans., 1972, 223-226; b) A. Gillie, J.K. Stille, J. Am. Chem. Soc. 1980, 102, 4933-4941.

31 J. Chatt, B.L. Shaw, J. Chem. Soc. 1956, 705-716. 5.10 References and notes 339

32 K. Tatsumi, R. Hoffmann, A. Yamamoto, J.K. Stille, Bull. Chem. Soc. Jpn. 1981, 54, 1857-1867.

33 J.M. Brown, N.A. Cooley, Chem. Rev. 1988, 88, 1031-1046.

34 J.L. Low, W.G. Goddard, J. Am. Chem. Soc. 1986, 108, 6115-6128.

35 E. Negishi, T. Takahashi, K. Akiyoshi, J. Organomet. Chem. 1987, 334, 181-194.

36 T. Yamamoto, A, Yamamoto, S. Ikeda, J. Am. Chem. Soc. 1971, 93, 3350-3352.

37 a) M.P. Brown, R.J. Puddephatt, C.E.E. Upton, J. Chem. Soc., Dalton Trans., 1974, 2457-2460; b) S. Roy, R.J. Puddephatt, J.D. Scott, J. Chem. Soc., Dalton Trans., 1989, 2121-2125; c) G.S. Hill, R.J. Puddephatt, Organometallics 1997, 16, 4522-4524; d) L.M. Rendina, R.J. Puddephatt, Chem. Rev. 1997, 97, 1735-1754; e) G.S. Hill, G.P.A. Yap, R.J. Puddephatt, Organometallics 1999, 18, 1408-1418; f) R.J. Puddephatt, Angew. Chem. Int. Ed. Engl. 2002, 41, 261-263.

38 K.L. Bertlett, K.I. Goldberg, T. Borden J. Am. Chem. Soc. 2000, 122, 1462-1465. ) D.M. Crumpton, K.I. Goldberg, J. Am. Chem. Soc. 2000, 122, 962-963; e) D.M. Crumton-Bregel, K.I. Goldberg, J. Am. Chem. Soc. 2003, 125, 9442-9456.

39 a) J.Y. Chen, J.K. Kochi, J. Am. Chem. Soc. 1977, 99, 1450-1457; b) S, Fukuzumi, K. Ishikawa, T. Tanaka, J. Chem. Soc., Dalton Trans., 1985, 899-904; c) A. Pederson, M. Tilset, Organometallics 1993, 12, 56-64.

40 M. Sato, E. Mogi, Organometallics, 1995, 14, 3157-3159.

41 J. Krömer, Dissertation 2000, University of Ulm, Germany.

42 a) K. Sonogashira, T. Yatake, Y Tohda, S. Takahashi, N. Hagihara, J.Chem. Soc., Chem. Commun. 1977, 291; b) K. Sonogashira, Y Fujikura, T. Yatake, N. Toyoshima, S. Takahashi, N. Hagihara, J. Organomet. Chem. 1978, 145, 101.

43 a) R.J. Cross, M.F. Davidson, Inorg. Chim. Acta 1985, 97, L35; b) R.J. Cross, M.F. Davidson, J. Chem. Soc. Dalton Trans. 1986, 1987-1992; c) A. Harriman, M. Hissler, R. Ziessel, A. De Cain, J. Fisher, J. Chem. Soc. Dalton Trans. 1995, 4067-4080.

44 M. Hacket, G.M. Whitesides, J.Am. Chem. Soc. 1988, 110, 1449-1462.

45 K. Campbell, R. McDonald, M.J. Ferguson, R.R. Tykwinski, J. Organomet. Chem. 2003, 683, 379-387.

46 M.A. Bennett, S.K. Bhargava, M. Ke, A.C. Willis, J. Chem. Soc. Dalton Trans. 2000, 3537-3545.

47 a) J. Krömer, I. Rios.Carreras, G. Fuhrmann, C. Musch, M. Wunderlin, T. Debaerdemaeker, E. Mena- Osteritz, P.Bäuerle, Angew. Chem. Int. Ed. Engl. 2000, 39, 3481-3486; b) G. Fuhrmann, J. Krömer, P. Bäuerle, Synth. Met. 2001, 119, 125-126; c) E. Mena-Osterritz, P. Bäuerle, Adv. Mater. 2001, 13, 243-246; d) E. Mena-Osteritz, Adv. Mater. 2002, 14, 609-616; e) G. Fuhrmann, T. Debaerdemaeker, P. Bäuerle, Chem. Commun., 2003, 948 - 949.

48 J. Kagan, S.K. Arora, Heterocycles 1983, 20, 1937-1940.

49 J. Krömer, P. Bäuerle, Tetrahedron 2001, 57, 3785-3794.

50 I. R Carreras, Diploma thesis 1999, University of Ulm, Germany.

51 a) D.W. Bruce, D.O’Hare, Inorganic Materials, Wiley, London, 1992; b) M.S. Khan, S.J. Davies, A. K. Kakkar, D. Schwartz, B. Lin, B.F.G. Johnson, J. Lewis, J. Organomet. Chem. 1992, 424, 87; c) P. Nguyen, G. Lesley, T.B. Marder, I. Ledoux, J. Zyss, J. Chem. Mater., 1997, 9, 406;d) K. Onitsuka, Y. Harada, F. Takei, S. Takahashi, Chem. Commun. 1998, 643 340 Chapter 5 Conjugated macrocycles by metal template approach

52 a) N. Chawdhury, A. Köhler, R.H. Friend, W.-Y. Wong, J. Lewis, M. Younus, P.R. Raithby, T.C. Corcoran, M.R.A. Al-Mandhary, M.S. Khan, J. Chem. Phys., 1999, 110, 3963-4970.

53 a) W.-Y. Wong, G.-L. Lu, K.-F. Ng, K.-H. Choi, Z. Lin, J. Chem. Soc., Dalton Trans., 2001, 3250-3260; b) J. Hua, W. Lin, Org. Lett. 2004, 6, 861-864.

54 a) M. Hissler, R. Ziessel, J. Chem. Soc. Dalton Trans. 1995, 893-896; b) H. Jiang, W. Lin, J. Am. Chem. Soc. 2003, 125, 8084-8085.

55A. Kaiser, Dissertation, University Ulm, Germany.

56 A.J.W. Tol, Synth. Met. 1995, 74. 95-98.

57 H. Plenio, Angew. Chem. Int. Ed. Engl. 1997, 36, 348-350.

58 D. Fichou, C. Ziegler, Structure and Properties of Oligothiophenes in the Solid State: Single Crystals and Thin Films in Handbook of Oligo- and Polythiophenes; Eds.: D. Fichou, Wiley-VCH, Weinheim, 1999, 183- 282.

59 E. Mena-Osteritz, Adv. Mater. 2002, 14, 609-616.

60 G. Fuhrmann, T. Debaerdemaeker, P. Bäuerle, Chem. Commun., 2003, 948 - 949.

61 a) S.K. Collins, G.P.A. Yan, A.G. Fallis, Angew. Chem. Int. Ed. Engl. 2000, 36, 348-350; b) Q. Zhou, P.J. Caroll, T.M. Swager, J. Org. Chem. 1994, 59, 1294-1301.

62 T.G. Appleton, M.A. Bennett, I.B. Tomkins, J. Chem. Soc. Dalton. Trans 1976, 439-446.

63 T. Debaerdemaeker, Z. Kristallogr. 1993, 206, 173-182.

64 G.M. Sheldrick, SHELXL-93, University of Göttingen, 1993. 341

Summary

Shape-persistent macrocycles with cavities in the nanometer regime are of great fundamental and technological interest due to their unique properties and potential application, particularly in the field of novel organic materials. Among all the cyclic structures that have these characteristics, cyclo[n]thiophenes represent a particularly interesting new class of compounds since the core of these macrocyles is built solely of α- linked conjugated oligothiophenes. Therefore, these cyclic derivatives describe systems in which an infinite π-conjugated chain of an ideal polymer is perfectly combined with the advantages of a structural well-defined oligomer. The aim of the present thesis was the development of a new concept that enables the efficient synthesis of these fully conjugated macrocyclic oligothiophenes. The initial synthetic strategy was based on the oxidative coupling reaction of appropriate α,ω-diethynylated oligothiophenes to the corresponding oligothiophene-diacetylene macrocycles. Subsequent conversion of the butadiyne-units incorporated in the frame of these cyclic products to thiophenes by reaction with sulfide nucleophiles afforded the desired cyclo[n]thiophenes. As starting materials for the synthesis of the required diethynylated building blocks, a homologues series of linear π-conjugated oligothiophenes have been synthesized and characterized. The first two homologues in the series of these butylated oligothiophenes, terthiophene 4 and quinquethiophene 7, were prepared by nickel-catalyzed Kumada coupling reactions. The synthesis of the higher homologues, septithiophene 18 and undecithiophene 19, was based on in this thesis newly developed and very efficient palladium-mediated sp-sp carbon coupling reactions. The subsequent reaction of the resulting butadiyne-precursors with sulfide anions under mild conditions afforded the targeted α-linked oligothiophenes 18 and 19.

Bu Bu Bu Bu

S S S n

4,7,18,19 (n = 1,2,3,5)

The effective introduction of the acetylenic groups at the terminal positions of the oligothiophenes 4, 7 and 18 was accomplished by selective iodination and subsequent Sonogashira coupling reaction of the diiodothiophenes 20-22 with trimethylsilylacetylene.

342

These, in high yields accessible protected diethynylated oligothiophenes 23-25 represent the direct precursors in the synthesis of cyclic structures. Macrocyclization of the terminally diethynylated terthiophenes 26 and quinquethiophene 28 by oxidative coupling reaction under pseudo high-dilution conditions led to the formation of mixtures of corresponding macrocyclic oligothiophene-diacetylenes C[mT-

DA]n (27a-h, 29a-e) in expected modest total yields of 12.7 % and 16.0 %, respectively.

Bu Bu

S S Bu Bu S Bu Bu Bu Bu Bu Bu S S S S Cu(OAc) x H O / CuCl R R 2 2 S pyridine / dichloromethane 23 R = Si(CH3)3 KOH pseudo high dilution S MeOH / THF Bu Bu 26 R = H S Bu S S Bu n-2

Bu Bu

C[3T-DA]n 27a-g (n =3-9) total yield [12.7 %]

Bu Bu Bu Bu Bu Bu

S S S S S R R

24 R = Si(CH ) KOH 3 3 MeOH / THF 28 R = H

Cu(OAc)2 x H2O / CuCl pyridine / dichloromethane pseudo high dilution

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S n-1 Bu Bu Bu Bu

29a-e (n = 2-6) total yield [16 %] C[5T-DA] n

343

Separation of the mixtures of the two series was accomplished by preparative HPLC. Each individual conjugated macrocycle 27a-h and 29a-e was isolated as stable, red microcrystalline solid and characterized.

The main goal of this thesis was to develop an effective method for the synthesis of fully conjugated oligothiophene macrocycles. Towards this end, the metal template approach was developed. This approach is based on the reaction of appropriate diethynylated oligothiophene precursors with transition metal halides. This thermodynamically controlled reaction afforded stable metallacycles in excellent yields. Finally, an induced reductive elimination by the help of an oxidant resulted in the subsequent C-C bond formation, generating the targeted oligothiophene-diacetylene macrocycles. The development of the metal template approach was started with studies carried on ethynylthiophenes as model compounds. In this frame, novel acyclic bis(oligothiophene- alkynyl)-Pt(II) complexes 36, 37 and 38 were synthesized and characterized. Upon equimolar addition of iodine to the cis-platinum complexes, the metal units were expelled and by a simultaneous C-C coupling the corresponding butadiynes were formed in quantitative amounts.

Bu Bu Bu Bu S S n Bu Ph Ph Bu PPh3 S S P Pt Pt S PPh Bu P S 3 Bu Ph Ph S S Bu 38 Bu n Bu Bu 36, 37 (n = 0,1)

This novel approach was then applied to the synthesis of cyclic structures. By using

Pt(dppp)Cl2, a series of dimeric cis-platinum-bridged diethynylated oligothiophene macrocyles (40, 41, 43) were prepared in excellent yields. The preparation of a trans- coordinated dimeric Pt-macrocycle 44 succeeded by using a trans-platinum precursor only in the case of the highest homologue, the diethynyl-septithiophene 38. These oligothiophene-diplatina macrocycles 40, 41, 43 and 44 represent a novel topology and class of compounds in the field of nananosized macrocycles. Optical and electrochemical

344 studies on all these new acyclic and cyclic oligothiophene derived platinum complexes allow the determination of interesting features and provide a data set that is important for the valuation of properties of related compounds.

Bu Bu Bu Bu PPh3 Pt Bu Bu S S Bu PPh3 Bu S S Bu Bu S Bu Bu S S S S Ph Ph Ph Ph P n P S S Pt Pt P P Ph Ph Ph Ph S S Bu S S n Bu S S S Bu Bu Bu S PPh3 S Bu Bu Bu Pt

Bu PPh3 Bu 40, 41, 43 (n = 1, 2, 3) Bu Bu 44

The key step of the metal template approach, namely the reaction with iodine afforded the elimination of the two Pt-corners and under preservation of the cyclic structure the oligothiophene-diacetylenes macrocycles were formed. In the case of the terthiophene derived diplatina macrocycle 40, only the strained cyclodimeric 45 was formed in a remarkable yield of 54%. This antiaromatic 32 π-electron system represents the smallest macrocycle in the homologues series, which could never been detected in the previous random cyclooligomerization reactions. The next cyclic homologues, diplatina macrocycles 41 and 43 were subjected to similar reaction conditions. For both, the corresponding cyclodimeric species 29a and 46a were formed as the main products along with higher cyclic homologues as by-products. The total yields of cyclic products obtained in these cases were lower than that for the terthiophene derivative.

345

Bu Bu Bu Bu S S S

S S S Bu Bu Bu Bu 45

Bu Bu Bu Bu Bu Bu Bu Bu Bu S S Bu S S S S Bu Bu S S S S Bu Bu S S S S Bu Bu S S S S Bu Bu S S S S Bu S S Bu Bu Bu Bu Bu Bu Bu 29a Bu Bu 46a

Undertaken in this work are also some studies concerning the mechanism of the reductive elimination. Based on NMR studies a mechanism for the elimination reaction is proposed. In addition, other possible induction methods for the easy and controlled processing of the reductive elimination reaction from oligothiophene derived platinum complexes, such as thermal activation or electron-transfer, were examined. Finally, the reaction of oligothiophene-diacetylene macrocycles 27a, 29a-b, and 45 with sodium sulfide allowed the synthesis and characterization of a homologues series of fully conjugated cyclo[n]thiophenes 47-50 (n = 8, 12, 16, 18).

Bu Bu

Bu Bu S S S

S S

S S S Bu Bu p Bu Bu

47- 50 (p = 1, 3, 5, 6)

cyclo[n]thiophenes (n= 8,12,16,18)

Zusammenfassung 347

Zusammenfassung

Formstabile Makrocyclen mit Hohlräumen im Nanometerbereich sind aufgrund ihrer einzigartigen Eigenschaften und der potentiellen Anwendung, insbesondere im Bereich der neuen organischen Materialien, von großem fundamentalen und technologischen Interesse. Unter allen cyclischen Systemen, die diese Merkmale aufweisen, stellen Cyclo[n]thiophene eine besonders interessante neue Verbindungsklasse dar. Da das Rückgrat dieser cyclischen Derivate nur aus α-verknüpften konjugierten Oligothiophenen besteht, stellen diese Makrocyclen Systeme dar, in welchen eine unendliche π-konjugierte Kette eines idealisierten Polymers mit den Vorteilen eines strukturell gut definierten Oligomers, aber ohne störende Endgruppen-Effekte, perfekt vereint ist. Das Ziel dieser Arbeit war die Entwicklung eines neuen Konzepts, das die effiziente Herstellung vollständig konjugierter makrocyclischer Oligothiophene ermöglicht. Die ursprüngliche Synthesestrategie basierte auf der oxidativen Kupplung entsprechender α,ω-diethinylierter Oligothiophene zu cyclischen Oligothiophen-Diacetylenen. Die anschließende Umsetzung dieser cyclischen Produkte mit Sufidanionen führte zu den erwünschten Cyclo[n]thiophenen. Für die Synthese der benötigten diethinylierten Bausteine wurde eine homologe Reihe linearer π-conjugierter Oligothiophenen synthetisiert und charakterisiert. Die ersten zwei Homologe in dieser Serie von butylsubstituierten Oligothiophenen, das Terthiophen 4 und Quinquethiophen 7, wurden mittels Kumada-Kupplung hergestellt. Die Synthese der längeren Homologen, das Septithiophen 18 and Undecithiophen 19, basierte auf einer in dieser Arbeit neu entwickelten und sehr effizienten Palladium-katalysierten sp-sp-Kohlenstoff Kupplungsreaktion. Die anschließende Reaktion der entstehenden Butadiine unter milden Bedingungen mit Sulfidanionen führte zu den erwünschten α-verknüpften Oligothiophenen 18 and 19.

Bu Bu Bu Bu

S S S n

4, 7, 18, 19 (n = 1, 2, 3, 5)

Die effektive Einführung der Acetylengruppen in den endständigen Positionen der Oligothiophene 4, 7 and 18 wurde durch eine selektive Iodierung and eine anschließende Sonogashira-Kupplung der Diiodothiophene 20-22 mit Trimethylsilylacetylen erzielt.

348 Zusammenfassung

Diese in hoher Ausbeute zugänglichen TMS-geschützten Diethinylthiophene 23-25 sind die direkten Ausgangsverbindungen für die Synthese der Makrocyclen. Die oxidative Kupplung des entschützten Diethinylterthiophens 26 and Diethinyl- quinquethiophens 28 unter Pseudo-Hochverdünnungsbedingungen führte zu der Bildung entsprechender Mischungen aus macrocyclischen Oligothiophen-Diacetylenen C[mT-DA]n (27a-h, 29a-e) in, wie erwartet, moderater Ausbeute von 12.7 % sowie 16.0 %. Die Trennung der beiden Mischungen wurde durch präparative HPLC erzielt. Jeder einzelne konjugierte Makrocyclus 27a-h and 29a-e konnte als stabiler roter, mikrokristalliner Feststoff isoliert und charakterisiert werden.

Bu Bu

S S Bu Bu S Bu Bu Bu Bu Bu Bu S S S S Cu(OAc) x H O / CuCl R R 2 2 S pyridine / dichloromethane 23 R = Si(CH3)3 KOH pseudo high dilution S MeOH / THF Bu Bu 26 R = H S Bu S S Bu n-2

Bu Bu Bu Bu Bu Bu Bu Bu 27a-g (n =3-9) S S C[3T-DA]n S S S R R total yield [12.7 %]

24 R = Si(CH ) KOH 3 3 MeOH / THF 28 R = H

Cu(OAc)2 x H2O / CuCl pyridine / dichloromethane pseudo high dilution

Bu Bu Bu Bu

S S S S Bu Bu S S Bu Bu S S S S n-1 Bu Bu Bu Bu

29a-e (n = 2-6) total yield [16 %] C[5T-DA] n

Zusammenfassung 349

Das Hauptziel dieser Dissertation war die Entwicklung einen neue und effizienten Methode für die Synthese von vollständig konjugierten makrocyclischen Oligothiophenen. Zu diesem Zweck, wurde die Metall-Templatmethode ausgearbeitet. Diese Methode basiert auf der Reaktion von entsprechenden Diethinyl-Oligothiophenen mit Übergangsmetallen. Diese thermodynamisch kontrollierte Reaktion sollte zuerst zu stabilen Metallacyclen in hohen Ausbeute führen. Durch eine reduktive Eliminierung der Metallzentren unter gleichzeitiger C-C Kupplung sollten anschließend die makrocyclischen Oligothiophen-Diacetylenen resultieren. Die Metall-Templatsynthese wurde zunächst mit relativ einfache Ethinyl-thiophenen als Modellverbindungen entwickelt. Neue acyclische bis(oligothiophen alkinyl)-Pt(II) Komplexe (36, 37 and 38) wurden hierbei synthetisiert und charakterisiert. Durch die äquimolare Zugabe von Jod zu den cis-Platin-Komplexen, konnten anschließend die Metall-Einheiten eliminiert und die entsprechenden Butadiine durch eine gleichzeitige C-C Kupplung quantitativ erhalten werden.

Bu Bu Bu Bu S S n Bu Ph Ph Bu PPh3 S S P Pt Pt S PPh Bu P S 3 Bu Ph Ph S S 38 Bu Bu n Bu Bu 36, 37 (n = 0,1)

Diese neue Methode wurde dann für die Synthese cyclischer Strukturen angewendet. Unter

Verwendung von Pt(dppp)Cl2 wurde eine Reihe von cis-Platin-verknüpften Oligothiophen- Diacetylen Cyclodimeren (40, 41, 43) in exzellenter Ausbeute synthetisiert. Die Herstellung von dem trans-koordinierten-Platin Cyclodimer 44 gelang unter Anwendung einer trans-Platin Ausgangsverbindung nur im Falle des höchsten Homologen, das Diethinyl-septithiophen 38. Die cyclischen Platin-verknüpften Oligothiophene-Diacetylene 40, 41, 43, 44 stellen eine neue Struktur und Verbindungsklasse von Makrocyclen im Nanometerbereich dar. Optische und elektrochemische Untersuchungen an diesen neuen acyclischen und cyclischen Oligothiophen basierten Platin-Komplexen erlauben die Ermittlung interessanter Merkmale und stellen einen Satz von Daten zur Verfügung, die für die Voraussage von Eigenschaften ähnlicher Verbindungen wichtig ist.

350 Zusammenfassung

Bu Bu Bu Bu PPh3 Pt Bu Bu S S Bu PPh3 Bu S S Bu Bu S Bu Bu S S S S Ph Ph Ph Ph P n P S S Pt Pt P P Ph Ph Ph Ph S S Bu S S n Bu S S S Bu Bu Bu S PPh3 S Bu Bu Bu Pt

Bu PPh3 Bu 40,41, 43 (n= 1, 2, 3) Bu Bu 44

Die Reaktion der cyclischen Diplatin-Makrocyclen mit Jod, die die Schlüsselsequenz in der Metall-Templatmethode darstelltte, führte unter Eliminierung von zwei Pt-Ecken und unter Erhaltung der cyclischen Strukturen zu den erwünschten Oligothiophen-Diacetylen Makrocyclen. Im Fall des Terthiophenderivates wurde nur das gespannte Cyclodimer 45 in einer nennenswerten Ausbeute von 54% gebildet. Dieses antiaromatische 32 π- Elektronensystem ist der kleinste Makrocyclus in der homologen Reihe. Dieser konnte in den vorherigen statistischen Cyclooligomerisierungen niemals detektiert werden. Die nächsten cyclischen Homologen, die Diplatin-Makrocyclen 37 and 39, wurden gleicher Reaktionsbedingungen unterworfen. In beiden Fälle erhielt man als Hauptprodukt das entsprechende Cyclodimer 29a sowie 46a, neben weitere höhere cyclischen Homologen als Nebenprodukt. Die Gesamtausbeute an cyclischen Produkten war in diesen Fällen niedriger als für das makrocyclische Terthiophenderivat.

Zusammenfassung 351

Bu Bu Bu Bu S S S

S S S Bu Bu Bu Bu 45

Bu Bu Bu Bu Bu Bu Bu Bu Bu S S Bu S S S S Bu Bu S S S S Bu Bu S S S S Bu Bu S S S S Bu Bu S S S S Bu S S Bu Bu Bu Bu Bu Bu Bu 29a Bu Bu 46a

Durchgeführt wurden in dieser Arbeit auch Untersuchungen hinsichtlich des Mechanismus der reduktiven Eliminierung. Basierend auf NMR-Studien wurde ein Mechanismus für die Eliminierungsreaktion aufgestellt. Für den einfachen und kontrollierten Ablauf der reduktiven Eliminierung von Oligothiophen-basierenden Platin Komplexen wurden weiterhin auch andere Methoden, wie die thermische Aktivierung oder Elektrontransfer, untersucht. Schließlich erlaubte die Reaktion der makrocyclischen Oligothiophen-Diacetylenen 27a, 29a-b, and 45 mit Natriumsulfid die Synthese und Charakterisierung einer homologen Serie von völlig konjugierten Cyclo[n]thiophenen 47-50 (n = 8, 12, 16, 18).

Bu Bu

Bu Bu S S S

S S

S S S Bu Bu p Bu Bu

47- 50 (p = 1, 3, 5, 6)

Cyclo[n]thiophene (n= 8,12,16,18)

List of Publications

• J. Krömer, I. Rios-Carreras, G. Fuhrmann, C. Musch, M. Wunderlin, T. Debaerdemaeker, E. Mena-Osteritz, P. Bäuerle, Angew. Chem., Int. Ed. Engl. 2000, 39, 3481-3486 (VIP: very important paper): "Synthesis of the First Fully α- Conjugated Macrocyclic Oligothiophenes: Cyclo[n]thiophenes with Tunable Cavities in the Nanometer Regime’’. • G. Fuhrmann, J. Krömer, P. Bäuerle, Synth. Met. 2001, 119, 125-126: “Synthesis of Macrocyclic Oligothiophenes”. • G. Fuhrmann, T. Debaerdemaeker, P. Bäuerle, Chem. Commun. 2003, 948-949: "C-C bond formation through oxidatively induced elimination of platinum complexes – A novel approach towards conjugated macrocycles". • M. Amman, M. Enßle, G. Fuhrmann, A. Kaiser, P. Kilickiran, E. Mena-Osteritz, P. Bäuerle, Polymer Preprints 2003, 44, 379-380: "Functional π-Electron Materials for Nanoelectronics”. • M.C. Ruiz Delgado, J. Casado, V. Hernandez, J.T. Lopez Navarrete, G. Fuhrmann, P. Bäuerle, J. Phys. Chem. B. 2004, 108, 3158-3167: "Combined Raman and Computational Study of a Novel Series of Macrocyclic π-Conjugated Diacetylene- Bridged α-linked Oligothiophenes”. • J. Casado, V. Hernandez, R.P. Ortiz, M.C. Ruiz Delgado, J.T. Lopez Navarrete, G. Fuhrmann, P. Bäuerle, J. of Raman Spectroscopy 2004, 35, 592-599: “Application of Raman spectroscopy and quantum chemistry for featuring the structure of positively charged species in macrocyclic π-conjugated diacetylenes-bridged oligothiophenes”. • A. Barbieri, B. Ventura, L. Flamigni, F. Barigelleti, G. Fuhrmann, P. Bäuerle, S. Goeb, R. Ziessel, Inorg. Chemistry 2005, 44, 8033-8043: “Binuclear Wirelike Dimers Based on Ruthenium(II)-Bipyridine Units Linked by Ethynylene-Oligothiophene- Ethynylene Bridges”.

Poster presentations

• G. Fuhrmann, J. Krömer, P. Bäuerle, ”Synthesis of α-Cyclo[n]thiophenes” (Poster), ICSM 2000 Bad Gastein/ Österreich, 15-21.07.2000. • G. Fuhrmann, P. Bäuerle, ”Fully Conjugated Macrocycles by Metal Templated Synthesis” (Poster), Fifth International Symposium on Functional π Electron Systems, Ulm/ Deutschland, 30.05-04.06.2002. • G. Fuhrmann, P. Bäuerle, ”Fully Conjugated Macrocycles by Metal Templated Synthesis” (Poster), 225th ACS National Meeting, New Orleans, USA, 23- 27.03.2003.

Curriculum Vitae

Name : Gerda Laura Fuhrmann Date of birth : 01.16.1972 Place of birth : Baia Mare / Rumania Nationality : german

Education 11/99 – 02/06 Dissertation in Chemistry (corresponds to Ph.D. degree) University of Ulm, Germany Departament of Organic Chemistry II Supervisor: Prof. Dr. P. Bäuerle Thesis: " Synthesis and Characterization of Oligothiophene-based Fully π-Conjugated Macrocycles”

07/01 - 09/01 Internship Boehringer Ingelheim Pharmaceuticals, Ridgefield, CT, USA Research project in Department Medicinal Chemistry, R&D Supervisor: Dr. Y. Ward

10/92 – 07/99 Diplom in Chemistry (corresponds to M.Sc. degree) University of Ulm, Germany Thesis: "Synthesis of phenyl substituted α-Cyclo[n]oligothiophenes". Departament of Organic Chemistry II Supervisor: Prof. Dr. P. Bäuerle

16.07.1999 Diplomprüfung (diploma examination) 15.11.1994 Vordiplom (corresponds to a bachelor degree)

03/96 Internship Technical University Vienna, Austria Practical training in anorganic chemistry

Scholastic Education 09/89 – 06/92 Edith-Stein Gymnasium, Ravensburg, Germany 04.06.1992 Abitur (corresponds to A-levels) 01/87 – 08/89 Aufbaugymnasium in Saulgau, Germany