«Inaugural-Dissertation towards the academic degree Doctor rerum naturalium (Dr. rer. nat.) Submitted to the Department of Biology, Chemistry and ...»
Inaugural-Dissertation towards the academic degree
Doctor rerum naturalium (Dr. rer. nat.)
Submitted to the Department of Biology, Chemistry and Pharmacy,
Freie Universität Berlin
Berit Annafrid Topolinski
The present work was carried out under the supervision of Prof. Dr. Dieter Lentz from April 2011
to January 2014 at the Institute of Chemistry and Biochemistry, at the Freie Universität Berlin.
First Referee: Prof. Dr. Dieter Lentz
Second Referee: Prof. Dr. Biprajit Sarkar Date of Defense: 26.03.2014 “Whatever it is you're seeking won't come in the form you're expecting.” Haruki Murakami (Kafka On The Shore) Acknowledgements Foremost I would like to express my gratitude to Professor Dr. Dieter Lentz for the possibility to carry out research in his group, for his supervision and advice. I thank Prof. Dr. Biprajit Sarkar for reviewing this thesis.
I am very grateful and in deepest debt to Professor Dr. Hidehiro Sakurai for his outstanding support, good advises, his trust and his hospitality and to Dr. Shuhei Higashibayashi for all the same reasons.
I am in debt to all members of Lentz group (former and present): Dr. Blazej Duda, Dr. Stefanie Fritz, Darina Heinrich, Dr. Thomas Huegle, Juliane Krüger, Dr. Moritz Kühnel, Annika Meyer, Dr.
Max Roemer, Dr. Bernd Schmidt and to all students who contributed to my project: Sebastian Czarnecki, Sergej Schwagerus and especially Michael Kathan and Antti Senf and much gratitude goes to all members and part-time members of Sakurai group who were always friendly and supportive: Dr. Nasir Baig, Dr. Raghu Nath Dhital, Setsiri Haesuwannakij, Yuka Ishida, Patcharin Kaewmati, Noriko Kai, Keita Kataoka, Jinyoung Koo, Dr. Yuki Morita, Sachiko Nakano, Yuki Okabe, Satoru Onogi, Prof. Dr. Gautam Panda, Dr. Patcharee Preedasuriyachai, Dr. Qitao Tan, Dr. Ryoji Tsuruoka, Dr. Tsuyuka Sugiishi and Dr. Sal Prima Yudha. Thanks to Dr. Mihoko Yamada, a former fellow corannulene-chemist.
Thanks to Prof. Dr. Sergej Troyanov (HU Berlin) for synchrotron measurements at BESSI.
Many thanks to Sarkar group for helping me with electrochemical problems and measurements, especially to Naina Deibl, Stephan Hohloch and Fritz Weißer.
I am very grateful to Doreen and Christiane Niether and Sandra Mierschink who suffered through over a decade of chemistry alongside with me.
Further, thank you all of my not-chemist-friends for just being there and all the good time we spent together and will hopefully keep spending.
I am very grateful to my family and of course to Bernd Schmidt who supported and encouraged me beyond all measure.
I thank the DFG (Deutsche Forschungsgemeinschaft) for funding in 2013/14.
Parts of this work have been published prior to submission of this manuscript:
B. Topolinski, B. M. Schmidt, S. Higashibayashi, S. Sakurai, D. Lentz, Sumanenylferrocenes and their Solid State Self-assembly, Dalton Transactions 2013, 42, 13809-13812.
B. Topolinski, B. M. Schmidt, M. Kathan, S. I. Troyanov, D. Lentz, Corannulenylferrocenes:
Towards a 1D, Non-covalent Metal-organic Nanowire. Chemical Communications 2012, 48, 6298-6300.
B. Topolinski, B. M. Schmidt, M. Kathan, H. Sakurai and D. Lentz “Ferrocenylated Buckybowls” 20th European Conference on Organometallic Chemistry, June 2013, St. Andrews (Scotland).
B. Topolinski, B. M. Schmidt, M. Kathan, H. Sakurai and D. Lentz “Ferrocenylated Buckybowls” 11th Ferrocene Colloqium, February 2013, Hannover (Germany).
8.3.14 1,1’’’-Dicorannulenylbiferrocene (39) and Biferrocenylcorannulene (40) 115 8.3.15 (R)-(+)-N,N-Dimethyl-1-ferrocenylethylamincorannulen (41) 117 8.3.16 1,6-Diferrocenyl-2,5-dimethylcorannulene (45) 118 8.3.17 1,6-Di(1’-neopentylferrocene)-2,5-dimethylcorannulene (46) 119 8.3.18 1,6-Bis(ethynylferrocenyl)-2,5-dimethylcorannunlene (47) 120 8.3.19 1,2-Bis(trifluoromethyl)-4,9-diferrocenylcorannulene (48) 121 8.3.20 1,2,5,6-Tetra(ferrocenyl)corannulene (51) 122 8.3.21 1,2,5,6-Tetra(1’-neopentylferrocenyl)corannulene (52) 124 8.3.22 1,2,5,6-Tetra(4-ferrocenylphenyl)corannulene (53) 125 8.3.23 1,2,5,6-Tetra(ferrocenylethynyl)corannulene (54) 126 8.3.24 1,2,5,6-Tetrakis(1’,2,2’,3,3’4,4’,5-octamethylferrocenyl) corannulene (55) 127 8.3.25 1,2,5,6-Tetrakis((E)-2-(ferrocenyl)ethenyl)corannulene (56) 128 8.3.26 Attempts to Synthesize 1,2,5,6-TetrakisR)-(+)-N,N-Dimethyl-1-ferrocenylethylamin)corannulene (57) 129 8.3.27 First Attempts to Synthesize sym-Pentaferrocenylated Corannulenes 130 8.3.28 1,3,5,7,9-Pentaferrocenylcorannulene (61) 133 8.3.29 1,3,5,7,9-Penta(1’-neopentylferrocenyl)corannulene (63) 135 8.3.30 Attempts to Synthesize Perferrocenylated Corannulenes 136
9. Abbreviations 138
1.1 Buckybowls Buckybowls are a subgroup of the large family of polycyclic aromatic hydrocarbons (PAHs).
In contrast to most PAHs these compounds are not planar or twisted but bowl-shaped and can be seen as subunits of the spherical, all-carbon fullerenes. The properties of these molecules are to some extent similar to those of fullerenes, but they offer more possibilities in modification of their CH-aromatic positions, superior solubility in common organic solvents and can be accessed by organic synthesis in solution or by flash-vacuum-pyrolysis.
1 Corannulene and sumanene, the two best known buckybowls, presented as subunit of the C60 Buckminster Fullerene.
Unlike planar aromatic systems, these non-planar hydrocarbons are not rigid in solution but undergo a bowl-to-bowl inversion via a planar (or S-shaped in case of larger derivatives) transition state, while the barrier of this inversion is strongly depending on the size and curvature of the buckybowl.
1 Figure 1.1.
2 Schematic bowl-to-bowl inversion of corannulene via a planar transition state depicted with chemdraw 3D.
Since the discovery of buckybowls in 1966 a variety of differently shaped and sized compounds was synthesized, testing the boundaries of synthetic methods. In 2011 the group of Scott created the so far deepest buckybowl, C50H10. This work focuses on the two smallest buckybowls, corannulene and sumanene.
1.2 Synthesis of Buckybowls
The C5v-symmetrical corannulene (C20H10) (Figure 1.2.1) was the first buckybowl to be [4,6] discovered, even before the discovery of fullerenes in 1985. Until today its synthesis has been improved considerably. The first synthetic route towards corannulene (C20H10) was reported by Barth and Lawton.[4, 7] A multistep organic synthesis, in 16 steps starting from 1,2dihydroacenaphthylene was applied and the desired compound was formed in a very low overall yield. A few decades later in 1991 the group of Scott presented a convenient new synthetic route. 2 Scheme 1.2.1 First published preparation of corannulene with a final flash-vacuum-pyrolysis step, as published by  Scott et al.
Corannulene was prepared starting from acenaphthoquinone which was reacted under Knoevenagel conditions, followed by a Diels-Alder and retro-Diels-Alder cycloaddition which provided a 7,10-fluoranthenedicarboxylic ester. The ester groups were reduced to diols and subsequently oxidized to the aldehyde. Applying Corey-Fuchs conditions aldehyde groups were converted to 2,2-dibromovinyl groups. The last step of synthesis was a pyrolysis at 1000 °C. After the purification of crude material a maximum of 10 % corannulene was obtained. The pyrolysis could also be applied to the tetrabromide substrate which provided higher yields of up to 40 % for the last step. This reaction was later modified by using a chlorovinyl pyrolysis precursor. Although yields improved, it was not possible to conduct the pyrolysis reaction on a large scale and the high temperature does not tolerate functional groups.
Parallel, the groups of Sygula, Rabideau and also Siegel worked on improving the milder liquid phase preparation of corannulene and in 2012 the group of Siegel finally reported a large scale, liquid phase synthesis for corannulene.
3 [13c] Scheme 1.2.2 Corannulene kilogram-synthesis according to Siegel et al. Yields are HPLC corrected yields.
In this synthesis dimethylacenaphtenechinone is reacted, similar to previous approaches, under Knovenagel and Diels-Alder conditions to construct 1,6,7,10-tetramethylfluoranthene, which then is brominated in its benzylic positions. The brominated fluroanthene derivative later undergoes a ring-closing reaction to yield 1,2,4,5-tetrabromocorannulene. After dehalogenation with palladium on carbon, corannulene can be obtained in up to a kilogram scale. This synthesis enabled corannulene to be prepared industrially and to be commercially available for the first time since its discovery 47 years ago.
1.3 Properties of Buckybowls Corannulene is the smallest existing buckbowl. It exhibits a rather low inversion barrier of 42.7 kJ*mol-1 at -64 °C in solution[2a] while showing a bowl depth of 0.87 Å. Corannulene’s solid state crystal structure is dominated by CH···π interactions but can strongly be influenced by the substitution of its CH-aromatic rim.[1b,17] Introduction of suitable substituents leads to a variety of packing motifs, including highly charge-conductive columnar structures dominated by π···π interactions.[17c] A noteworthy property of corannulene is its electron acceptor ability, similar to fullerenes. Corannulene can be chemically reduced in one-electron steps to the tetraanion using alkali metals, if the reduction is carried out in dry aprotic solvents and under inert conditions. Odd-numbered reduction states are paramagnetic species while even numbered states are diamagnetic. After prediction of the solid state structure of the tetraanion (C20H104-) as an octaanionic dimer in 1994, the group of Petrukhina presented the first crystal structure of a corannulene tetraanion with lithium cations in 2012, confirming the previously predicted sandwich-like structure. Electrochemically corannulene can be reduced reversibly up to three times under certain conditions. Besides being a fellow buckybowl, sumanene’s properties are drastically different from corannulene’s properties. Sumanene possesses three benzylic positions, a slightly larger surface area than corannulene, deeper bowl depth of 1.11 Å and a much higher inversion barrier of 82 kJ/mol at 140 °C. In contrast to corannulene the crystal structure of unsubstituted sumanene already shows a dense columnar arrangement.[23a,24] Sumanene displays a lower ability to exchange electrons than corannulene and only shows quasi-reversible (DMF) or irreversible (MeCN) reduction processes at very negative potentials.
51.4 Ferrocenylated Aromatics
Ferrocene was discovered in 1951 by Kelay and Pauson who tried to synthesize fulvalene by reaction of iron(III) chloride with cyclopentadiene magnesium bromide and was the first η5bond complex to be discovered. Today, already known for more than six decades, it is the most investigated sandwich complex. Because of its favorable properties like reversible redox behavior, its chemical modification possibilities and its stability under aerobic conditions it is not only used widely in catalysis and pharmaceutics but also frequently applied as a substituent in macromolecular assemblies, polymers and redox-systems. Ferrocene reliably undergoes oxidation to ferrocinium which is only modestly stable in aerobic solutions. Nevertheless, the ferrocene/ferrocinium redox couple is often used as a reference to compare the electronic properties of many different organic and organometallic compounds.
Star shaped oligoferrocenes are promising materials as multi-redox systems and in molecular electronics. Ridgid star-shaped perferrocenyl molecules with a benzene core have been reported and in 2006 hexaferrocenylbenzene, a molecule that previously had been regarded as impossible to synthesize, was presented. The compound was prepared by treatment of hexaiodobenzene with six equivalents of diferrocenylzinc. The resulting perferrocenylated molecule is highly crowded which leads to a distortion of the central benzene ring (as depicted in Figure 1.4.1).
1 Crystal structure of hexaferrocenylbenzene as a wireframe model visualized with Mercury. The  benzene ring is slightly distorted (left side).
6 The compound was reported to be sensitive to air, especially in solution. The ferrocene substituents show electronic communication which can be determined by the bathochromic shifts relative to ferrocene in the UV-spectrum and the cyclic voltammogram shows three separated redox waves (E1/2 = 163 mV (one-electron wave), 232 mV (two-electron wave), and 222 mV (three-electron wave) in dichloromethane versus ferrocene/ferrocinium). Further crowded ferrocenylated aromatic compounds were presented by the groups of Astruc and Lang. The group of Astruc created rigid redox stars, also prepared under Negishi-type reaction conditions.[31,33] The ferrocene substituents are connected by ethynyl spacers to the benzene core. Due to the low solubility of the compounds with ethynylferrocene substituents, methylated ethynylferrocenes were also prepared.
2 Hexa(ferrocenylethynyl)benzenes synthesized by Negishi-type C,C cross-coupling reactions of hexabromobenzene with the corresponding zincated ferrocenylethynylferrocene.
Electrochemically, the ethynyl spacers prevent significant electronical interactions between the ferrocene substituents, nevertheless frustration effects and electrostatic interactions were observed by using a fluorinated bulky non-coordinating conducting salt (n-NBu4BArF4), to prevent ion-pairing.[31,34] In contrast to hexaferrocenylbenzene, these ethynyl bridged compounds are stable in solution and under aerobic conditions. Chemical oxidation yields stable iron(III) compounds which can be reversibly reduced to iron(II) species. Within their studies the group of Astruc investigated the influence of the substitution pattern on the observation of electronic and electrostatic interaction on ethynyl substituted benzene compounds.
7 Scheme 1.4.