Research Collection
Doctoral Thesis
Anthraphanes: a New Class of Potential Monomers for the Synthesis of Two-Dimensional Polymers
Author(s): Servalli, Marco
Publication Date: 2016
Permanent Link: https://doi.org/10.3929/ethz-a-010832536
Rights / License: In Copyright - Non-Commercial Use Permitted
This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.
ETH Library
DISS. ETH NO. 23866
Anthraphanes: a New Class of Potential Monomers for the Synthesis of Two-Dimensional Polymers
A thesis submitted to attain the degree of
DOCTOR OF SCIENCES of ETH ZURICH
(Dr. sc. ETH Zurich)
presented by
MARCO SERVALLI
MSc, ETH Zurich Chemistry
born on 10.07.1986
citizen of Bioggio (TI), Switzerland
accepted on the recommendation of
Prof. Dr. A. Dieter Schlüter, examiner Prof. Dr. Joost VandeVondele, co-examiner Prof. Dr. Peter Walde, co-examiner Dr. Michael Wörle, co-examiner
2016
ACKNOWLEDGEMENTS
Acknowledgements
Doing a PhD and writing a thesis is not a stand-alone matter and involves the collaboration of many people. I would like to start my acknowledgements from the scientific and work-related point of view:
At first, I would like to thank Prof. A. Dieter Schlüter, for offering me the chance to work on the stimulating, fascinating and challenging topic of two-dimensional polymers. Challenging projects deal specially with a lot of frustration and I therefore appreciate the continuous support I received from him during my PhD, in terms of scientific and non-scientific discussion, as well as the essential ongoing motivation to not give up and to keep pursuing my objectives. What I value the most is the large freedom and independence I was given during my work and his critical thinking and meticulous, endorsing attitude when reviewing scientific results. Also, with him I learned the critical importance of diplomacy and communication in academia and science.
Further thanks go to my former advisor Prof. Junji Sakamoto. During my PhD’s first year, I worked under his supervision and received constant support. The countless scientific discussions, inputs and clever ideas I received from him, helped my project to advance in the right direction and widened my knowledge and skills in organic chemistry.
Prof. Wolfgang Kinzelbach receives my deepest gratitude for helping me in having a smooth and trouble-free transition between my supervisors and for supporting one of my publications.
Recurring visits to our group from Prof. Gerhard Wegner and Prof. Benjamin T. King have always resulted in fruitful and enlightening discussions and additional motivation, especially when the project seemed to approach a dead-end. I am very grateful to have made their acquaintance and had the opportunity to witness and assimilate their extensive knowledge.
I am very grateful to Prof. Frank-Gerrit Klärner, for his offer to calculate countless electrostatic surface potentials for many of my compounds and sharing his knowledge in the field. I really appreciate his interest in my work and his support for my publications.
The main part of this thesis is based on crystallography and has therefore inevitably involved the collaboration of the exceptional crystallography team of the Small Molecule Crystallography Center (SMoCC) at ETH, composed by Dr. Michael Wörle, Dr. Nils Trapp and Michael Solar. I would like to thank Dr. Michael Wörle for his interest in the project, the discussions we had and for the easy access to the synchrotron in Grenoble that he provided; Dr. Nils Trapp solved countless crystal structures for me and struggled to make them publishable when the crystals were not diffracting
IV
ACKNOWLEDGEMENTS properly. I am also thankful for the countless discussions we had and for proof-reading my thesis and supporting my publications. Finally, I am really indebted with Michael Solar for mounting and measuring countless single crystals and for spending hours in the crystallography lab trying to find properly diffracting samples when needed; I thank him for being patient with me in the times I was a bit pushy. His help in preparing the set-up for the topochemical reactions is also greatly appreciated.
I would like to thank Dr. Thomas Weber of the Laboratory of Crystallography for his interest in my work, the discussions we had, for measuring some crystal structures and for the great chance of connecting me with Prof. Hans-Beat Bürgi, who I had the privilege to meet and with whom I could share my results in an enlightening discussion.
Many thanks go to Dr. Thomas Schweizer. During my PhD he provided me with a variety of LED- photoreactors and a heating plate with a PID controller for my crystallisation experiments; Also, I would like to thank him for his assistance for any technical-related problem: he repaired countless things and helped me to fix our recycling GPC. We are very lucky to have you in our group!
Many thanks go to Prof. Paul Smith for providing access to the light microscopes in his group and to Dr. Kirill Feldman for the kind introduction on the instruments.
I would like to thank Prof. Ralph Spolenak and Prof. André Studart for providing access to the AFM and SEM instruments respectively.
I am indebted to Dr. René Verel for his kind help with solid-state NMR spectroscopy measurements and any general NMR-related question or problem. Many thanks also go to Doris Sutter for measuring some of my samples on the 500 MHz spectrometer and for the “schoggi-stückeli”.
Many thanks go to Rolf Häfliger of the Mass Spectrometry Service at the LOC at ETH for measuring countless of my samples and putting effort into finding the molecular ions peaks when needed.
Many thanks also go to Dr. Sara Fornera and Sandra Luginbühl for their help with the fluorescence and UV/Vis absorption spectra in solution.
The help of Dr. Alessandro Lauria with the solid-state UV/Vis absorption measurements is greatly acknowledged.
I thank Dr. Jingyi Rao for her help with TGA and DSC measurements, Feng Shao for the confocal Raman spectroscopy measurements, Wenyang Dai for the AFM and SEM measurements and Vivian Müller for measuring the UV/Vis absorption and emission spectra of my Langmuir monolayers.
V
ACKNOWLEDGEMENTS
Special thanks go to Dr. Payam Payamyar for his help in anything related to the synthesis of two- dimensional polymers at the air/water interface; I’m really thankful for the introduction to the Langmuir-trough instruments, LEDs, in-situ fluorescence decay measurements both at the air water/interface and with the single crystals, TEM and SEM measurements and general scientific discussions.
For scientific discussions on my project I would like to also thank Gregor Hofer, Dr. Max Kory, Bernd Deffner and Prof. Yingjie Zhao.
Progress on my project was achieved thanks to the help of students that wished to carry out a semester project with me. In this regard I would like to thank Luzia Gyr, Andri Mani, Nadia Zuurbier and Livius Muff for their essential help in the lab. Special thanks go to my fantastic lab assistant Emmanuel Wirth, who helped a great deal with the synthesis of my key compounds in large scale.
Finally, I would like to thank the administrational machine behind our research group, Daniela Zehnder, for taking care of any organisational and administrative matter and Dr. Damir Bozic for all computer-related issues.
During my stay in the Schlüter’s group as an undergrad, I have worked in different labs, but I truly felt like home when I landed in G522 at the start of my PhD. The mysterious ability of this lab to gather, attract and keep the most chill and outgoing people resulted not only in a very fruitful working environment and constant mutual support but also in friendships and memorable funny moments with my lab mates:
Dr. Thomas Bauer, for all the parties we had together which resulted in cycling in the Limmat, exploring the grounds of the Stuz building, early morning BBQs with top-notch equipment and the memorable trip to Las Vegas, St. Patrick’s Day in NYNY, Area 51 and Spring Break in Lake Havasu with our pilot Matteo and logistic support from Eddie Lane. Other memorable mentions go to Jim, Tyler, Spider and the unbearable Noisy Lady on the plane in Philly. Also, without your foresight, we wouldn’t have taken the plane back to Zurich (but I still think we should have borrowed Jim’s teeth…);
Dr. Payam Payamyar, my good friend we had memorable parties and skiing trips together, especially in Boltigen and Gstaad: I’m sorry about that challenging black slope we had to take, but it was really the only way to go to the Ice Bar! On the bright side, the first 300 meters of slope before the cliff were indeed just a narrow flat path as I correctly predicted from the map. I am also thankful for the generous after-party hospitality and dinners at his place, and the countless lunch times we had together at WoKa although after 4 years I still haven’t tried the yellow curry! Ehrlich gesagt, so geht
VI
ACKNOWLEDGEMENTS das nicht! The time in Japan was epic when “we were feeling perfectly fine, no jet-lagged at all!” and I will never forget the karaoke sessions in Nara and the good food we enjoyed. I am also deeply thankful for the support during the late night fluorescence experiments at FIT.
Dr. Ming Li, for letting us visit and shoot an outstanding biographic documentary at his place at 4 am, but especially for his epic google search topics in the lab and the teaching of basic Chinese words;
Bernd Deffner, for being part of the aforementioned documentary starring as “The Creature” and afterwards helping me refreshing my ironing skills: now my shirts will always look flawless. Also, for the nice time at the Oktoberfest in Munich and dinners at his place; for enduring ten hours of Trololo in the lab and the banana incident. I hope your musical knowledge was also greatly widened: ”What what…”. Mediation with Master POS was also highly appreciated!
Chiara Gstrein, for the uncountable laughter and funny moments we had together, for the memorable ASOT parties in Holland and events at her place (especially the birthday party). Most of all I appreciate the nice conversations we had and the support during hard times. Thank you!;
I would also like to thank the offical adoptive members of our lab: Dr. Jingyi Rao, her cheerful and easy-going attitude was a bless during work; Dr. Simon Cerqua, who left our lab to live the royal life… the trip to Zermatt was unforgettable, as well as the countless parties we had together and our excursion to Amsterdam for King’s Day with Payam: now I know what being allowed to only drink champagne means, and what my cup of tea is “hahaha! *TAP*”. He was also a formidable adversary during the chemical wars in Ibis Hotel. I also thank him for the mutual support during the ancient sensei times; Dr. Xiaoyu Sun for the funniest conversations ever, and for his kind invitation to go to Shanghai; Dr. Radu Rusu, I think nobody can match the narrative of his various funny anecdotes. Also, for the BK and Nelson’s time and the subsequent three hours long, inter-cantonal, train trip at 5 am to go from Hauptbahnhof Zürich to Urdorf: I especially enjoyed Baden, Wintherthur and Zug.
Finally I would like to thank all the previous and current members of the Polymer Chemistry Group, especially Prof. Peter Walde, Dr. Baozhong Zhang, Dr. Julia Bättig, Dr. Sara Fornera, Dr. Benjamin Hohl, Dr. Animesh Saha, Dr. Andri Schütz, Dr. Tim Hungerland, Dr. Samuel Jakob, Dr. Anzar Khan, Dr. Patrick Kissel, Dr. Max Kory, Dr. Andreas Küchler, Sandra Luginbühl, Wenyang Dai and Marianna Marchini (especially for the tennis matches), Vivian Müller, Ralph Lange, Gregor Hofer, Stan van de Poll. The wide palette of peculiar personas and characters I came across at my stay at ETH will be treasured and has for sure prepared me for my future life.
Life during my PhD was not only in the lab and I want therefore to acknowledge all the people that were involved in my social life and supported me during stressful time; I am sure I will forget some
VII
ACKNOWLEDGEMENTS names but I would especially like to thank my Ticino Crew: Matteo Forni, Djamila Domeniconi, Michele “Bollicine” Bernasconi, Dr. Lorenza Ferretti, Pamela Ghilardi, Eliano Sonzogni and Joz for the countless events, dinners and parties we had, and the special NYE nights. Especially with Matteo Forni, I think it would be pointless to write all the trips and funny moments we had, as it would require a book; however I feel I should mention the ASOT parties, the 10 years holidays in Ibiza, the two Las Vegas trips with the flights on the Chessna and that unforgettable party in Marquee (and after-party). Gabriele Fenini, my good friend, science and party colleague, sharing the same passion for music! I think I would also need a book to write about all the adventures and crazy moments we had especially with our friend Mauro Pellanda and Gigi: memorable six years of ASOT and Tomorrowland, various EFs and ADEs, Street Parades and Ibiza nights: let’s hope to keep rocking it like this in the future! I’d like to thank Jovan Jancev, for the countless workouts at the gym we had together and the advice and consulting for supplements I received from you, the scientific discussions and parties we had. Jerry, good friend, with you the night-life in Zurich (and Milan) got an all new meaning: I will never forget the nice conversations, parties and after-parties we had. The same applies for Morten and Kristian, I had memorable moments with you guys. My sport’s and temporary room-mate Stephan, I really enjoyed this summer swimming, paddling and working out with you. I really wish you all the best in life!
Finally, I would like to express my deepest gratitude to my family, from which I received endless support: my mom Patrizia Servalli, my father Adriano Servalli, my brother Fabio Servalli and especially my grandfather Bruno Botta and grandmother Lenita Botta. Without them, I would not have been where I am standing now.
VIII
CONTENTS
Table of Contents
List of Abbreviations XII
Abstract XIV
1. Introduction 1
1.1 Two-Dimensional Polymers 3
1.2 Synthetic 2DPs 6
1.2.1 Single-Crystal-to-Single-Crystal (SCSC) Approach 7
1.2.2 Air/Water Interfacial Approach 12
1.2.3 Other Approaches 15
1.3 Post-Polymerisation Modifications 17
1.4 Potential Applications 18
1.5 The Importance of the Monomer’s Design 19
1.6 Aim of the Thesis 22
2. Anthraphane Monomer 23
2.1 Monomer’s Design 23
2.2 Synthetic Approach 28
2.3 Synthesis of Anthraphane 30
2.4 Single Crystal Structure of Precursor 11 36
2.5 Crystallisation of Anthraphane 38
2.5.1 First Insights into the Packing Behaviour 38
2.5.2 Interactions in Single Crystals of Anthraphane 40
2.5.3 Solvent Choice for Crystallisation 43
2.5.4 Solvent Screening 44
2.5.5 Crystallisation Procedure 49
2.5.6 Anthraphane Solvates 50
2.6 Packing of Anthraphane in the Single Crystal 51
IX
CONTENTS
2.6.1 Etf Packing 1 54
2.6.2 Etf Packing 2 64
2.6.3 Mixed etf/ftf Packing 1 67
2.6.4 Mixed etf/ftf Packing 2 82
2.6.5 Packing with no anthracene-anthracene Interactions 85
2.7 Considerations on the Packings 88
2.7.1 The “Geometrical” and the “Chemical” Crystal: etf/ftf Packing 1 vs etf Packing 1 88
2.7.2 Solvent-Induced Anomalies: etf Packing 2, etf/ftf Packing 2, no Anthracene-Anthracene Interactions Packing 93
2.7.3 Thoughts on Crystal Engineering and Crystal Structure Prediction 96
2.8 Towards the Right Packing 99
2.9 Topochemical SCSC Photodimerisations of Anthraphane 106
2.9.1 Dimers from the etf/ftf Packing 1 108
2.9.2 Dimers from the etf/ftf Packing 2 111
2.9.2 Thermal Stability of the Anthraphane Dimers 112
2.10 Conclusion and Outlook 115
2.11 Experimental 118
2.11.1 Synthesis 118
2.11.2 SC-XRD Analysis 124
3. Amphiphilic Anthraphanes 145
3.1 Monomer’s Design 145
3.2 Synthesis of Anthraphane-tri(OMe) 2a 147
3.3 Spreading of 2a at the Air/Water Interface 152
3.4 Synthesis of Anthraphane-tri(DEGME) 2b 153
3.5 Spreading of 2b at the Air/Water Interface 156
3.6 Photochemical Environment of 2b at the Air/Water Interface 159
3.6.1 Excimers Fluorescence Decay 163 X
CONTENTS
3.6.2 Packing at the Air/Water Interface 166
3.7 Polymer Sheet 168
3.8 Conclusion and Outlook 170
3.9 Experimental 172
3.9.1 Synthesis 172
3.9.2 Langmuir Monolayers at the Air/Water Interface 182
4. Diazaanthraphanes 187
4.1 Monomer’s Design 187
4.2 Synthetic Approach 190
4.3 Synthesis of Diazaanthraphanes 191
4.3.1 Synthesis of Methyl Ester Diazaanthraphane 3a 191
4.3.2 Synthesis of Ethyl and n-Propyl Ester Diazaanthraphane 3b-c 195
4.4 Crystallisation of Diazaanthraphane 3c and Packing in the Single Crystal 200
4.5 Preliminary Irradiation Studies 205
4.6 Conclusion and Outlook 209
4.7 Experimental 212
4.7.1 Synthesis 212
4.7.2 SC-XRD Analysis 231
5. Conclusions and Outlook 234
6. Appendix 239
References 246
Curriculum Vitae 257
XI
LIST OF ABBREVIATIONS
List of Abbreviations
1D One-dimensional 2D Two-dimensional 3D Three-dimensional 2DP Two-dimensional polymer 3-HPA 3-Hydroxypicolinic acid Ø Diameter A Acceptor AFM Atomic Force Microscope aq. Aqueous BAM Brewster’s Angle Microscopy bmim 1-Butyl-3-methylimidazolium calcd. Calculated CCD Charge-Coupled Device CCDC Cambridge Crystallographic Data Centre cn Centroid COF Covalent Organic Framework COSY Correlation Spectroscopy CP-MAS Cross Polarization Magic Angle Spinning Chemical shift D Donor DCM Dichloromethane DDQ 2,3-Dichloro-5,6-dicyano-1,4-benzoquinone DEGME Diethyleneglycol monomethyl ether DFT Density Functional Theory DIPA Diisopropylamine DIPEA N,N-Diisopropylethylamine DMAP 4-Dimethylaminopyridine DMF Dimethylformamide DMPU 1,3-Dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone DMSO Dimethyl sulfoxide DSC Differential Scanning Calorimetry eq Equivalent ESI Electrospray ionisation ESP Electrostatic Surface Potential EtOAc Ethyl acetate etf edge-to-face ftf face-to-face GBL Gamma-butyrolactone GISAXS Grazing Incidence Small Angle X-Ray Scattering h-BN hexagonal boron nitride HMBC Heteronuclear Multiple-Bond Correlation spectroscopy HRMS High-resolution Mass Spectrometry HSQC Heteronuclear Single Quantum Coherence HV High Vacuum ICR-FT Ion Cyclotron Resonance Fourier Transform IR Infrared IR-RAS Infrared Reflection Absorption Spectroscopy
XII
LIST OF ABBREVIATIONS
J Spin–spin coupling constant λ Wavelength LED Light-Emitting Diode M Molarity (mol/L) m meta m/z Mass-to-charge ratio MALDI-TOF Matrix-Assisted Laser Desorption/Ionization-Time Of Flight MeOH Methanol MEP Molecular Electrostatic Potential MMA Mean Molecular Area MOF Metal-organic Framework Mp Melting point NMP N-Methyl-2-pyrrolidone NMR Nuclear Magnetic Resonance o ortho ODCB ortho-Dichlorobenzene OM Optical Microscopy ORTEP Oak Ridge Thermal Ellipsoid Plot p para PAH Polycyclic Aromatic Hydrocarbons PID Proportional-Integrative-Derivative pln Plane PM3 Parameterized Model number 3 POM Polarised Optical Microscopy ppm parts per millions PPM Post-Polymerisation Modification PSM Post-Synthetic Modification Rf Retardation factor rGPC Recycling Gel Permeation Chromatography SCSC Single-Crystal-to-Single-Crystal SC-XRD Single Crystal X-Ray Diffraction SEM Scanning Electron Microscopy SP Surface Pressure SS-NMR Solid-State Nuclear Magnetic Resonance STM Scanning Tunneling Microscope TBAF Tetra-n-Butylammonium Fluoride TCE 1,1,2,2-Tetrachloroethane TEM Transmission Electron Microscopy TERS Tip-Enhanced Raman Spectroscopy Tf2O Triflic anhydride TGA Thermogravimetric Analysis THF Tetrahydrofurane TIPS Triisopropylsilyl TIPSA (Triisopropylsilyl)acetylene TLC Thin-Layer Chromatography TMS Trimethylsilyl TMSA Trimethylsilylacetylene UV Ultraviolet Vis Visible XPS X-ray Photoelectron Spectroscopy XIII
ABSTRACT
Abstract
Since the isolation of graphene in 2004, a single atom-thick molecular sheet of carbon, the research field on this revolutionary natural 2D material has literally exploded, revealing its peculiar and outstanding properties such as its enormous tensile strength and electrical conductivity. These interesting properties are the result of the molecular structure of graphene and particularly its confinement in two dimensions. Graphene is thus expected to gain a huge societal impact. Considerable interest has also arisen for other 2D materials such as hexagonal boron nitride (h-BN), with its structure analogous to graphene, and inorganic metal chalcogenides such as MoS2, WS2,
MoSe2 and NbSe2. However, the aforementioned 2D materials are mostly of inorganic nature and therefore lack the versatility that organic chemistry can offer, in terms of functionalities and chemical modification.
The synthesis of organic 2D polymers by mild recipes was first achieved in 2012 and since then, considerable interest in the field has developed. They are defined as free-standing, macromolecular sheets with one-monomer-unit thickness and a periodical internal structure. However, their synthesis can be very challenging, in fact, four years later, the reported cases of successfully synthesised 2D polymers can still be counted on the fingers of one hand. The biggest synthetic challenge relies in having a controlled growth reaction confined into two dimensions and a periodical polymeric structure. The two successful methods for synthesising 2D polymers rely into the pre- organisation and polymerisation of the monomers in layered single crystals and pre-organisation and polymerisation at the air/water interface.
As the field of 2DPs is still in its infancy, there is a need for new monomer and polymer systems, to widen the field and to better understand the potential properties and applications of these new materials. In particular, it is desirable to have a versatile monomer structure, which can be employed for both the single crystal and the air/water interface approach, so that a direct comparison of the two methods can be done in terms of feasibility, structural perfection of the polymer obtained and its characterisation. For this purpose, a novel anthracene-based monomer’s family was designed and synthesised: the “anthraphanes” 1, 2 and 3. These monomers all share the same basic skeletal structure, but according to their functionalities they can be in principle used for the single crystal approach and/or the air/water interface approach.
Chapter 1 provides a general introduction to 1D and 2D polymers and reviews the currently available synthetic 2D polymer systems, as well as the methods to prepare them. General criteria for designing monomers for the synthesis of 2D polymers will also be discussed.
XIV
ABSTRACT
Figure I. Anthraphane monomers used in this thesis. The photoreactive anthracene units are displayed in red colour.
Chapter 2 presents the anthraphane monomer 1, a trifunctional, anthracene-based, shape-persistent hydrocarbon macrocycle intended to be used for the single crystal approach. The anthraphane design represents the basic structure for this new family of monomers. The synthesis of the monomer will be presented along with its packing behaviour in the single crystals, with a thorough discussion on how to steer the crystal packing of anthraphane into the desired direction. Finally some SCSC reactions of anthraphane will be presented together with a study on the thermal stability of the anthraphane dimers.
Chapter 3 deals with the desymmetrisation of the anthraphane structure 1 to obtain amphiphiles suitable for the air/water interface approach. The synthesis of the novel monomers 2a and 2b will be presented and their spreadability at the air/water interface will be investigated. A polymerisation reaction of 2b forming a mechanically coherent two-dimensional covalent monolayer sheet will be shown, together with first preliminary insights into the internal structure of the polymer.
Chapter 4 presents an engineered structure of anthraphane 1 for the SCSC approach, based on substituted 1,8-diazaanthracene photoreactive units. With the experience gained with the crystallisation of the anthraphane monomer, diazaanthraphane 3 was designed to pack in the single crystals exclusively into the desired fashion, with the reactive units in close proximity. Synthesis of the monomer and first insights into its packing behaviour and photoreactivity will be presented.
This thesis not only provides access to new versatile monomers, but also highlights the challenges and problems associated with the synthesis of 2DPs and their relation with the monomer design.
XV
ABSTRACT
Riassunto
Sin dalla scoperta e isolazione del grafene nel 2004, uno strato monoatomico di atomi di carbonio, la ricerca e l’interesse per questo rivoluzionario materiale 2D sono letteralmente esplosi, elucidando le sue peculiari ed eccezionali proprietà come il suo enorme carico di rottura e conduttività elettrica. Queste interessanti proprietà derivano della struttura molecolare del grafene e in particolare il suo confinamento in due dimensioni. In futuro ci si può quindi aspettare che il grafene avrà un importante impatto sociale.
Interesse considerevole è anche stato dedicato ad altri materiali 2D come il boro nitruro esagonale
(h-BN), isostrutturale al grafene, e gli inorganici calcogenidi di metalli come MoS2, WS2, MoSe2 e
NbSe2. I materiali 2D sopracitati sono tuttavia principalmente di natura inorganica e mancano della versatilità che la chimica organica offre, in termini di funzionalità e modificazione chimica.
La sintesi di polimeri 2D organici tramite metodi dolci è stata ottenuta nel 2012 e da quel momento, si è sviluppato un interesse considerevole in questo campo. I polimeri 2D sono definiti come fogli molecolari autoportanti, aventi una struttura interna periodica e uno spessore corrispondente allo spessore dell’unità monomerica di cui sono composti. La sintesi di questi materiali può essere tuttavia molto ardua, infatti, quattro anni dopo, i casi di polimeri 2D riportati in letteratura si possono ancora contare sulle dita di una mano. La più grande sfida sintetica consiste nell’avere una polimerizzazione controllata e confinata in due dimensioni e nell’ottenere una struttura polimerica cristallina. I due metodi affermati per sintetizzare polimeri 2D contano sulla preorganizzazione e polimerizzazione dei monomeri in cristalli singoli stratificati e sulla preorganizzazione e polimerizzazione all’interfaccia aria/acqua.
Siccome i polimeri 2D sono ancora alla loro infanzia, occorre designare nuovi sistemi monomerici e polimerici, per ampliare questo campo di ricerca e meglio capire le proprietà e applicazioni potenziali di questi nuovi materiali. In particolare, è desiderabile avere delle strutture monomeriche versatili, che possano essere impiegate sia per l’approccio nei cristalli singoli, sia all’interfaccia aria/acqua, così che una comparazione diretta tra i due metodi può essere fatta in termini di fattibilità, perfezione strutturale del polimero ottenuto e la sua conseguente caratterizzazione. A tal fine, una nuova famiglia di monomeri basata sull’antracene è stata designata e sintetizzata : gli “antrafani” 1, 2 and 3. Questi monomeri condividono la stessa unità strutturale, ma in base ai loro gruppi funzionali possono essere utilizzati per l’approccio nei cristalli singoli e/o all’interfaccia aria/acqua.
XVI
ABSTRACT
Figura I. I monomeri antrafani utilizzati in questa tesi. Le unità fotoreattive sono mostrate in rosso.
Il Capitolo 1 fornisce un’introduzione generale sui polimeri 1D e 2D e recensisce i polimeri 2D e i metodi sintetici correntemente disponibili nella letteratura scientifica. Criteri generici per il design dei monomeri vengono anche discussi.
Il Capitolo 2 introduce il monomero antrafano 1, un macrociclo idrocarburo trifunzionale, basato sull’antracene e concepito per l’approccio in cristalli singoli. Il design dell’antrafano rappresenta la struttura basilare di questa nuova famiglia di monomeri. La sintesi di 1 viene presentata assieme ai suoi vari impaccamenti in cristalli singoli, seguita da una meticolosa discussione su come guidare l’impaccamento cristallino di 1 nella direzione desiderata. Infine verranno presentate delle reazioni cristallo-singolo-a-cristallo-singolo, seguite da uno studio sulla stabilità termica dei dimeri di antrafano.
Il Capitolo 3 tratta la desimmetrizzazione dell’antrafano 1 per ottenere un sistema anfifilico destinato all’approccio all’interfaccia aria/acqua. La sintesi dei monomeri 2a e 2b viene dunque presentata, seguita da uno studio sulla loro dispersione all’interfaccia aria/acqua. Il monomero 2b verrà dunque polimerizzato, formando un monostrato molecolare covalente e bidimensionale, caratterizzato da autoportanza. Studi preliminari sulla struttura interna del polimero verranno anche presentati.
Il Capitolo 4 presenta una nuova struttura derivata dall’antrafano 1, progettata appositamente per l’approccio in cristalli singoli: il diazaantrafano 3, basato sui fotoreattivi 1,8-diazaantraceni. Con l’esperienza accumulata nella cristallizzazione di 1, il diazaantrafano 3 è stato progettato per impaccarsi esclusivamente nel modo desiderato, con le unità fotoreattive in diretta prossimità. La sintesi del monomero e i primi studi nella sua cristallizzazione e fotoreattività saranno mostrati.
Questa tesi non solo fornisce accesso a nuovi monomeri versatili, ma evidenzia anche le difficoltà e i problemi associati alla sintesi di polimeri 2D, e la loro relazione con il design dei monomeri.
XVII
Author’s Declarations
Parts of the text presented in the abstract as well as chapters 2 and 4 are taken from the following two papers:
M. Servalli, N. Trapp, M. Wörle, F.-G. Klärner, “Anthraphane: An Anthracene-Based, Propeller-Shaped
D3h-Symmetric Hydrocarbon Cyclophane and Its Layered Single Crystal Structures”, J. Org. Chem. 2016, 81 (6), 2572–2580. DOI: 10.1021/acs.joc.6b00209.
and
M. Servalli, L. Gyr, J. Sakamoto, A. D. Schlüter, “Propeller-Shaped D3h-Symmetric Macrocycles with Three 1,8-Diazaanthracene Blades as Building Blocks for Photochemically Induced Growth Reactions”, Eur. J. Org. Chem. 2015, 4519–4523. DOI: 10.1002/ejoc.201500496.
XVIII
INTRODUCTION
1. Introduction
When thinking about how chemistry affects our daily lives, the first thought addresses molecules. Perhaps it goes in drugs, medicaments and biologically related compounds as they are directly linked to our health and well-being. Morphine, acetylsalicylic acid, paracetamol, penicillin and the more complex macrocyclic antibiotic erythromycin (Figure 1.1) are just a tiny fraction on the list of the essential medicines according to the World Health Organisation (WHO)[1]. This kind of compounds are commonly referred to as small molecules, conventionally having a molecular weight smaller than 900 Da[2].
Figure 1.1. Few representatives of small molecules, in this case essential medicines according to WHO.
There is however another class of compounds, which deeply impact our lives and which in fact are, the reason why life itself exists. These compounds can easily reach molecular weights up to 1 MDa and are called polymers or macromolecules. The terms were coined in the 1920s by Hermann Staudinger*, who in a pioneering life-time work, anticipated and later proved their existence, awarding him the Nobel Prize for Chemistry in 1953[3,4]. At the beginning of the 20th century, some polymer products were already present in the market, but they were prepared by trial-and-error procedures (mostly by chemically modifying natural polymeric materials[5]) since their molecular structure and architecture was unknown and poorly understood. It was believed that the properties of polymers were caused by the self-assembly of small molecules, which were held together by weak intermolecular forces rather than strong covalent bonds. After studying and experimenting with polyoxymethylene and natural rubber, Staudinger proposed his revolutionary concept that polymers were composed by long linear molecular chains, consisting of a large number of small repetitive base units (today called repeat units) linked together by covalent bonds. Biopolymers such as polynucleotides (DNA, RNA), polypeptides, proteins and polysaccharides have existed since the origin
* In reality the term polymer was coined by J. J. Berzelius in 1833, it was however referred to compounds different in molecular weight but sharing the same empirical formula. The modern concept of polymers was coined by H. Staudinger. 1
INTRODUCTION of life, but it was this understanding of the molecular architecture of polymers that opened the route to synthetic polymers (Figure 1.2).
Figure 1.2. The concept of polymer as macromolecule and two examples of synthetic polymers: polypropylene (PP) and polytetrafluoroethylene (PTFE).
Covalently linking small molecules (monomers) by using the toolbox of organic chemical reactions[6] has resulted in the most versatile and accessible class of materials ever: by using monomers with defined chemical functionalities combined with a controlled and selective synthesis, polymers can be made soft, rigid, conducting, insulating, transparent, opaque, permeable, impermeable, biodegradable etc. Such diverse properties can nowadays be easily tuned according to societal applications and needs, and the simplicity and cheap procedures by which polymers are made, have resulted into a gigantic global market, with a yearly production in the order of 311 million tonnes[7]. Nowadays polymers (commonly referred as plastics) are used in every aspect of our lives: polyethylene terephthalate (PET) bottles, plastic bags and containers, non-stick surfaces (Teflon) adhesives, automobile and airplane parts, elastomers, clothing, Styrofoam and insulating materials are just a fraction of the many applications of these materials.
From a topological point of few, despite the different conformations that polymers chains can achieve, they can be considered as being one-dimensional. The repeat units have a topological linear extension in space and when connected together by covalent bonds they reflect into topologically linear polymer chains. In the growth reaction, the monomers are linked together following a hypothetical line. The polymer chains can then entangle and/or be arranged in a crystalline fashion, but they still remain topologically 1D. The synthetic polymers discussed previously that have such a huge societal impact, are all polymerised one-dimensionally. The question that arises now is whether the polymerisation, the growth reaction, has to be inevitably restricted in 1D.
2
INTRODUCTION
1.1 Two-dimensional Polymers
In 2010, the Nobel Prize in Physics was awarded jointly to Andre Geim and Konstantin Novoselov for their study on graphene[8]. Their breakthrough paper was published in 2004, in which they could isolate and characterise the structure of graphene, a single atom-thick molecular sheet of carbon[9]. Since then, the research field on this revolutionary material has literally exploded, revealing its peculiar and outstanding properties such as its enormous tensile strength of 130 GPa and its Young modulus of 1 TPa[10] (compared to 0.50 MPa and 200 GPa respectively for steel[11]), optical transparency[12], electrical[13] and thermal conductivity[14]. These interesting properties are the results of the molecular structure of graphene and particularly its confinement in two dimensions. By looking in more detail into its structure, displayed in Figure 1.3, one notices the following characteristics: graphene has a periodical structure; it is composed of carbon atoms only, which are linked together by covalent bonds, giving mechanical coherence to the sheet, whose thickness corresponds to the thickness of one carbon atom only.
Figure 1.3. The molecular structure of graphene, a prototypical natural two-dimensional polymer.
Due to its periodicity, the structure of graphene can be deconstructed into its smallest structural element, or repeat unit (by using the terminology of 1D polymers), which in this case can be identified as a single sp2-hybridised carbon atom. At this point, the question of whether the topology of a polymer has to be restricted into 1D only starts to have an answer. In contrast to linear polymers, in graphene the repeat unit is topologically planar with three laterally extended bonding sites: attachment of the repeat units happens therefore on a plane rather than on a line. Graphene can consequently be regarded as the prototype two-dimensional polymer (2DP). In terms of dimensionality, the small molecules presented in Figure 1.1 and the monomers in 1.2 are zero- dimensional (can be considered as dot-like). Depending on how the monomers are connected to each other and how their binding sites are arranged in space, higher dimensionality topologies develop such as 1D, 2D and 3D. Figure 1.4 shows this concept with some ideal carbon-based
3
INTRODUCTION monomers and their corresponding polymeric products for each topological dimension: the carbon allotropes carbyne†, graphene and diamond.
Figure 1.4. Different topologies of repeat units reflect into different topologies of polymers. The topology of a repeat unit is defined by how its binding sites are arranged in space. A hypothetical linear acetylenic diradical repeat unit polymerises one-dimensionally into the elusive carbyne polymer; the planar sp2-hybridised carbon atom polymerises two-dimensionally into graphene; the tetrahedral sp3-hybridised carbon polymerises three-dimensionally into the diamond structure.
There are few other examples in nature of two-dimensional polymeric materials, also displaying interesting properties: hexagonal boron nitride (h-BN) with its analogous structure to
[15–17] graphene and inorganic metal chalcogenides such as MoS2, WS2, TaS2, MoSe2 and NbSe2 . In an analogy to 1D polymers’ history, these materials can be considered as the prototype natural polymeric rubber that Staudinger used to study and which ultimately led to the creation of synthetic polymers. The aforementioned 2D materials are mostly of inorganic nature and therefore lack the versatility that organic chemistry can offer, in terms of functional groups, reactions and mild synthetic conditions. Their isolation into monolayer sheets usually requires a top-down exfoliation
† Carbyne can also refer to the more general structure R-C 3•, an electrically neutral carbon with triple radical nature. The elusive 1D polymer is often referred to as linear acetylenic carbon. 4
INTRODUCTION from their bulk layered form[18–22]. Due to an increased interest into these materials however, bottom-up synthetic methods have also been developed that usually require harsh conditions such as thermolysis[23,24] or chemical vapour deposition[25–27]. Such methods would not be compatible with the functionalities offered by organic chemistry. In fact, for how exceptional graphene is, one could argue that the properties of this kind of material are limited by the lack of functional groups in its structure (although graphene can be to some extent functionalised[28–30]). It is therefore suggestive to think that if one could apply the same organic reaction toolbox that was used for synthetic 1D polymers, specifically designed synthetic organic 2D polymers with desired properties could be accessible. Monomers could be synthesised with the desired functionalities and linked together by the means of mild synthetic procedures, creating a wide palette of new 2D materials with different properties and applications (for potential applications of 2D polymers, see section 1.4).
Before venturing into the field of synthetic 2DPs, it is however important to define what a 2DP is and what characteristics should a macromolecule have to be classified as such. Based on the concepts of Staudinger’s 1D polymers and graphene, five criteria have been proposed by our group[31]:
1. A 2DP is a topologically planar molecular sheet and it is made up of topologically planar monomer units having typically 3, 4 or 6 binding sites; The repeat units can be seen as tiles, which when linked together can completely tessellate a plane. Monomers bearing 5 or 7 binding sites cannot tile cover a surface alone;
2. A 2DP has a periodical ordered structure and exhibits crystallinity in at least one of its conformations. Even if composed by structurally ordered units, a 2DP might relax into conformations which deviate from geometrical periodicity. In this regard, for detection of order by for instance x-ray techniques (which require geometrical periodicity), it is important that the polymer can assume a crystalline conformation within the time scale of the order detection experiment. A high number of degrees of freedom in the polymer could be imparted by monomers also having considerable degrees of freedom; the use of rigid and shape-persistent monomers could simplify detection of crystallinity in the 2DP;
3. The thickness of a 2DP corresponds to the thickness of its constitutional repeat unit;
4. The repeat units are linked together by robust covalent bonds;
5. The robust bonds between the repeat units ensure that a 2DP is free-standing and can withstand its own weight without collapsing when exposed to ambient gravitational conditions. Thus 2DPs can be isolated as single molecular monolayers.
5
INTRODUCTION
To summarise, a 2DP is a one-monomer unit thick, free-standing monolayer sheet with internal periodical structure, composed of covalently attached repeat units (Figure 1.5). Other definitions of 2D polymers have been proposed in the meantime which also encompass 2D entities that lack long-range order or persistent bonds between their constituent elements, and are not limited to single monolayer but also multilayers[32]. We however believe that such a holistic view of a 2DP can be reductive, as some of the properties of 2DP are expected to arise from their periodicity on the molecular scale. In this work we will therefore use the term “2D material” to refer to all 2D molecular entities independently from their periodical structure or monolayer nature, and consider a “2D polymer” a specific subclass of 2D material which meets the 5 criteria exposed previously.
Figure 1.5. Cartoon representation of a two-dimensional polymer with its five key features: it is composed by repeat units, it possesses a periodical structure, it is one monomer unit thick and the robust bonds between the monomers confer to the sheet mechanical coherence.
1.2 Synthetic 2DPs
The synthesis of organic 2D polymers by mild recipes is still very challenging, but it is not a synthetic fantasy anymore since 2012, when the first ever organic synthetic 2DP was made[33]. Since then, considerable interest in the field has developed and more scientific groups have started to work on the topic[32,34–37]. However, four years later, the reported cases of successfully synthesised 2DPs can still be counted on the fingers of a hand[33,38–41]. The biggest synthetic challenge relies in having a controlled growth reaction confined into two dimensions (without resulting into a cross-linked 3D network) and a periodical polymeric structure. These two main points can be addressed by pre- organising the monomers in two dimensions before polymerising them. The two most successful methods in this regard rely into the pre-organisation into layered assemblies and the pre- organisation at interfaces. In the next subchapters these two methods will be described and concrete 6
INTRODUCTION examples of synthesised 2DPs will be presented. Other potentially successful methods will be briefly addressed as well.
1.2.1 Single-Crystal-to-Single-Crystal (SCSC) Approach
This method is based on the topochemical polymerisation of monomers in single crystals. The field of topochemistry was developed by Schmidt and Cohen in the 60s and 70s[42,43]. Since then, topochemical reactions have become well established and many well studied examples can be found in the literature[44,45], among which the [2+2]-cycloaddition of trans-cinnamic acids is perhaps the most representative. A few years later, the concept was applied by Wegner and Hasegawa to difunctional molecules to create polymers via topochemical photopolymerisations. Notable examples are the polymerisations of diacetylenes[46] and diolefins[47].
Figure 1.6. Cartoon representation of the topochemical single-crystal-to-single-crystal (SCSC) approach for the synthesis of two-dimensional polymers.
The topochemical approach for the synthesis of 2DPs involves the crystallisation of a monomer into lamellar single crystals (Figure 1.6). In each layer, the monomers are regularly packed with their reacting sites in close proximity to their neighbours. The layers are held together by weak intermolecular forces and do not have the chance to chemically cross-link with each other: the directionality of the bonding sites of the monomers is 2D-confined in each layer, so that the growth reaction will also be confined in two dimensions. Thermal treatment or photoirradiation of the crystal then triggers the polymerisation reaction, with the formation of covalent bonds between the monomers, converting the monomer crystal into a 2D polymer crystal. Thermally- and photo-induced reactions are of particular advantage since they do not involve any mass transport associated with common reagents; the crystal packing is therefore minimally disturbed during the reaction, fulfilling
7
INTRODUCTION the topochemical principles. The 2D polymer crystal can then be chemically, thermally or mechanically exfoliated to yield single 2DP sheets.
Figure 1.7. The first 2DP ever synthesised. Top row: cup-shaped monomer structure (left) and its lamellar single crystal structure (center). The short distances between the acetylenic carbons and the 9,10 positions of the anthracene neighbour allowed for a topochemical [4+2]-cycloaddtion (right). Bottom row: AFM image of the single sheets obtained by exfoliation of the polymer crystals (left). TEM image of the polymer sheet (Wiener filter) showing the periodicity of the internal structure (center). Electron diffraction pattern of the polymer crystals showing crystallinity (right). Adapted by permission from Macmillan Publishers Ltd: [33], copyright 2012.
The first organic 2DP was prepared in 2012 with this method in our group by Kissel et al.[33], by synthesising in a 25 steps sequence an anthracene-based, cup-shaped rigid macrocycle with three- fold symmetry and crystallising it into lamellar hexagonal single crystals (Figure 1.7). Single crystal x- ray diffraction (SC-XRD) analysis of the monomer crystals revealed a crystal structure in which the anthracene units of the monomers were in close proximity with the alkyne functionalities of their neighbours, so that upon photoirradiation a [4+2]-cycloaddition reaction was triggered polymerising the monomers. Due to the reduced diffracting quality of the polymerised crystals, a structure of the polymer crystal could not be obtained by SC-XRD; however bond formation was confirmed by Raman and SS-NMR spectroscopy, and internal order by electron diffraction. The crystals could also be exfoliated into single monolayers, whose thickness was confirmed by atomic force microscopy (AFM). Shortly after, in 2013 another example of 2DP was provided by the work of Bhola et al.[38]: a three- fold symmetric anthracene monomer based on the trypticene motif, “antrip” was synthesised and crystallised into single crystals (Figure 1.8). This time the crystals exhibited a pseudo-lamellar 8
INTRODUCTION structure, however polymerisation nevertheless occurred by the photoinduced [4+4]-cycloaddition dimerisation between the neighbouring anthracene units. Bond formation was confirmed by IR spectroscopy and solid-state NMR (SS-NMR) and the crystals could be exfoliated into single sheets as confirmed by AFM. During polymerisation, the rearrangement of the molecules in the crystal from a monomeric pseudo-lamellar structure to a polymeric lamellar structure, resulted in drastic loss of crystallinity, so that a structural proof of the 2DP crystal could not be obtained by SC-XRD.
Figure 1.8. Chemical structure of the antrip monomer (left) and its packing in the single crystal showing a pseudo-lamellar arrangement (center). Proposed structure of the antrip polymer (right). Adapted with permission from reference [38]. Copyright 2013 American Chemical Society.
The first crystal structures of two 2DPs were reported one year later in two independent works from the King’s and our group. In the first work from Kissel et al.[39], a fluorinated version of the antrip monomer, “fantrip” was crystallised and polymerised in two consecutive steps by the photoinduced [4+4]-cycloaddition reaction between the fluorinated anthracene units (Figure 1.9). The reaction occurred in a single-crystal-to-single-crystal fashion, so that the monomer, dimer and polymer structures could be unequivocally obtained by SC-XRD. Exfoliation of the 2DP crystal resulted mostly into thin sheet stacks but also single polymer sheets.
9
INTRODUCTION
Figure 1.9. Top: chemical structure of the fluorinated fantrip, with its single crystals and lamellar crystal packing. Bottom: two-step SCSC photopolymerisation of fantrip. Adapted by permission from Macmillan Publishers Ltd: [39], copyright 2014.
The second example was provided by the trifunctional rotor-shaped anthracene-based monomer made by Kory et al.[40] (Figure 1.10). Compared to the previous examples, this monomer was easily accessible in terms of synthesis and could be produced in gram-scale without effort [48], rendering it interesting for potential large-scale applications. The monomer was crystallised and polymerised via [4+4]-cycloaddition in the single crystals and the structure of the 2DP characterised by SC-XRD; the thermally-induced reversibility of the polymerisation reaction was also demonstrated, showing that the single crystals could undergo polymerisation and depolymerisation without loss of crystallinity. However, the exfoliation procedure resulted challenging due to the very strong interactions between the layers, so that single layer entities could be isolated only in rare cases.
Figure 1.10. Top: chemical structure of the monomer, with its single crystals and lamellar crystal packing. Bottom: topochemical SCSC photopolymerisation. Adapted by permission from Macmillan Publishers Ltd: [40], copyright 2014. 10
INTRODUCTION
These four examples nicely highlight the strengths and weaknesses of this synthetic approach and already offer a prospect about the challenges of synthesising 2DPs. The SCSC approach clearly offers the possibility to easily characterise the polymer by SC-XRD and thus obtaining a single crystal structure of the polymer, which is the ultimate structural proof for periodicity. On the other hand, as shown by the early examples of Kissel and Bohla, topochemical reactions do not always result in SCSC transformations; the molecular movements and strains in the crystals associated with the reaction can result in phase segregations inside the crystal or even result in cracks, loss of crystallinity and disintegration of the crystal itself, effectively preventing SC-XRD analysis. While in some cases it is possible to steer a topochemical reaction towards a SCSC transformation by finely tuning the reaction parameters such as temperature, crystal size, wavelength and intensity of the light source, other spectroscopic methods complementary with x-ray diffraction might nevertheless be needed to prove bond formation and internal structural order. Another issue associated with this method is the size of the obtained 2DP; in fact the maximal lateral extension that can be reached by a sheet corresponds to the lateral extension of the single crystal itself. However, single crystals are usually far from ideal and can have defects: incorporation of impurities, dislocations, twinning, grain boundaries and mosaicity are not uncommon and can therefore reduce the size of the periodical domains in a crystal. In this regard, fine tuning of the crystallisation conditions might help improving single crystal quality and growing larger single crystals might yield larger 2DP sheets, which is desirable for applications purposes; contrastingly, the larger the single crystal, the less efficiently it can relax from the strain exerted by any mismatch between lattice metrics during a topochemical reaction, lowering the chance of a well-behaved SCSC transformation. Moreover, large single crystals also imply more defects. The exfoliation to single sheets can also become problematic if the intermolecular forces that keep the layers together in the crystals are too strong, as seen in the example of Kory. A poorly optimised exfoliation process could also be the cause of ill-shaped single sheets, which might rupture or fold during the process. Last but least, is the issue of the packing of the monomer in the single crystals, which can be extremely hard to predict and to steer into the desired direction, namely obtaining lamellar structures in which the monomers pack with their reactive units in close proximity to each other.
The SCSC approach is nevertheless a powerful method for synthesising 2DPs and we believe that a careful monomer design can efficiently help to tackle most of the shortcomings exposed in the previous paragraph, such as crystal packing and exfoliation process.
11
INTRODUCTION
1.2.2 Air/water Interfacial Approach
Another promising method for the synthesis of 2DPs is based on the pre-organisation of the monomers at an air/water interface. Studies on monolayers at the air/water interface were pioneered by Langmuir at the beginning of the 20th century with his experiments using amphiphilic fatty acid and eventually resulted in the invention of the so-called Langmuir trough[49,50].
Figure 1.11. Cartoon representation of a Langmuir trough used in the air/water interface approach for the synthesis of two-dimensional polymers.
The synthetic method towards 2DPs works as follows (Figure 1.11): amphiphilic monomers are spread at the air/water interface in a Langmuir trough and subsequently compressed with the help of movable barriers into an ideally crystalline state, in which structural periodicity starts to manifest. A monomer monolayer is then formed, which can undergo polymerisation by irradiation or diffusion of reagents from the sub-phase to the reactive sites of the monomers. With this method, the lateral extension of a 2DP is theoretically limited by the extension of the water surface, opening the possibility of having sheets of macroscopic size, a desirable feature for application purposes. There are several advantages for using water as the medium for pre-organisation: first of all water has a low surface-roughness of about 3 Å (calculated by the root mean-square method[51]), which ensures that the growth reaction will be in-plane; secondly, water helps to stabilise amphiphilic monomers (in terms of molecular orientation at the interface) but still allows for mobility, so that order formation can take place while defects can be healed; thirdly, the weak attractive forces between the monomers at the interface will result in a limited number of nucleation sites, allowing for larger ordered domains (typically in the micrometer range[52]). Apart from the virtually unlimited lateral extension of the molecular sheets that can be obtained, this method also has the advantage of directly yielding monolayers (without the need of exfoliation procedures from layered assemblies), whose thicknesses can be easily estimated by conventional methods such as AFM. However, if height analysis is trivial, proof of internal order is quite the opposite: compared to the SCSC approach, there is no fast and routine method such as SC-XRD. While there are monolayer-sensitive techniques (both in-situ or after transferring the monolayer on a substrate), for proof of bond formation such as infra- 12
INTRODUCTION red reflection absorption spectroscopy (IR-RAS), tip-enhanced Raman spectroscopy (TERS) and x-ray photoelectron spectroscopy (XPS), proof of crystallinity rely on techniques such as STM, TEM, high resolution AFM and x-ray techniques such as grazing-incidence small-angle scattering (GISAXS), which can require extensive skills and time for structural elucidation (see below). For these reasons, reports in the literature of covalent Langmuir monolayer sheets which still await proof of internal order are available, while reports of true 2DPs prepared by this method are still scarce. In the following paragraphs a few selected examples will be presented which also highlight the shortcomings of this approach to synthetic 2DPs.
The first examples regard 2D metal-organic frameworks (MOFs); organic trifunctional and hexafunctional terpyridine-based building blocks from Bauer and Zheng have shown to self-assemble at the air/water interface by metal complexation into mechanically coherent monolayers after addition of metal salts such as Fe2+ [53,54]. Another similar example from Sakamoto et al. involves trifunctional monomers with dipyrrin ligands[55] which also undergo complexation with Zn2+ salts to form monolayers (Figure 1.12). While these examples fulfil most of the five criteria for the classification of 2DP, there is no proof of structural periodicity available.
Figure 1.12. Terpyridine and dipyrrin based monomer for the synthesis of organometallic sheets at the air/water interface upon complexation with metal cations. Adapted from reference [55].
However in another case from Makiura[56], for a metallated palladium porphyrin equipped with four carboxylate ligands, formation of crystalline monolayer domains was confirmed and followed by in-situ grazing incidence x-ray diffraction after injection of copper salts into the water
13
INTRODUCTION sub-phase; proof of mechanical coherence of this metal-organic framework was however not reported (Figure 1.13).
Figure 1.13. Porphyrin based monomer for the synthesis of metal-organic frameworks sheets at the air/water interface. Adapted from reference [56].
The second examples regard the formation of organic Langmuir monolayer covalent sheets produced by photo-induced polymerisation. In the work of Payamyar et al.[57], an uncapped version with three exposed hydroxyl groups of the monomer made by Kissel[58], was photopolymerised at the air/water interface through [4+4]-cycloaddition between the anthracene units, producing free- standing monolayers. While bond formation was proved by fluorescence spectroscopy and TERS[59], periodicity could not be confirmed due to the structural flexibility of the monomer: DFT simulations showed that the degrees of freedom of the monomer reflected into a structurally flexible monolayer, which upon relaxation from the ideal periodical structure resulted into a seemingly amorphous network (Figure 1.14). Similar covalent sheets were also produced with a structurally related 1,8- diazaanthracene monomer[60] and the prospect of co-polymerisation between two different monomers was also demonstrated[61].
Figure 1.14. Amphiphilic monomer used for the synthesis of covalent monolayer sheets at the air/water interface. The hypothetical periodical polymer structure is shown to collapse into an amorphous network upon relaxation in vacuum. Adapted with permission from reference [57].
14
INTRODUCTION
In a final example by Murray et al.[41], an amphiphilic version of the shape-persistent antrip monomer used by Bhola[38], was successfully employed at the air/water interface for synthesising a covalent monolayer sheet, whose crystallinity was locally confirmed by STM analysis (Figure 1.15). This example established that it is in fact possible to synthesise 2DPs at the air/water interface.
Figure 1.15. Amphiphilic antrip monomer for the synthesis of 2DPs at the air/water interface. Periodicity of the sheets at the molecular level was confirmed by STM analysis. Adapted with permission from reference [41]. Copyright 2015 American Chemical Society.
As discussed previously, the main challenges of the air/water interface approach regard the characterisation of the obtained monolayers. However, like in the SCSC approach the potential problems with this method can be minimised by carefully designing the monomer; it has been shown for instance that when using shape-persistent monomers, the resulting polymers have higher chance to be investigated for internal order. Proof of bond formation can then be tackled by spectroscopic methods such as IR and Raman; however, when working with monolayers transferred on metallic surfaces, surface selection rules apply. It is therefore beneficial to have the dipoles moment associated with the newly formed bonds orthogonal to the surface (or at least not completely parallel).
1.2.3 Other Approaches
Apart from the SCSC and the air/water interface approach, there are other methods that have potential for the synthesis of 2DPs. They will be briefly addressed in the following paragraphs.
2D Covalent organic frameworks (COFs)
Two-dimensional covalent organic frameworks are layered porous crystalline structures whose organic building blocks are held together by covalent bonds. The first report by Yaghi dates back to 15
INTRODUCTION
2005[62] (Figure 1.16) and since then this research field (together with MOFs) has been very prolific due to the applications of these crystalline materials for hydrogen and methane storage[63,64]. A typical COF synthesis is very mild, scalable and basically consists in the mixing of the building blocks in the proper stoichiometry; it is therefore clearly attractive for the synthesis of 2DPs. However, in contrast to the SCSC and the air/water interface approaches, the growth reaction (polymerisation) and order formation (crystallisation) are not two separated processes, but happen instead at the same time during the reaction. The difficulty of separating and controlling these steps can result in considerable structural defects and microcrystalline products if there is a mismatch between the rate of polymerisation and the rate of crystallisation. In this regard, COF chemistry is based on dynamic covalent chemistry which employs reversible bonds (such as boronate esters or imines)[65–67], so that structural defects can in principle be healed. Weaker and more reversible bonds offer the chance for better healing but intrinsically impart chemical susceptibility to the COFs. Being stacked layered structures[68], COFs have to be exfoliated into stable single layers to be classified as 2DPs; single layers are expected to be more susceptible to hydrolysis with respect to their stacked form and could result in chemically unstable sheets at ambient conditions. So far, exfoliation of COFs to single layers has still to be achieved; however some progress has been made into exfoliating these materials down to few layers[69,70]. At the same time, obtaining single crystals of 2D COFs is still an open matter, although for 3D COFs it has been shown to be possible[71,72]. By solving these two crucial problems, it is foreseeable that 2DPs could be eventually synthesised by using COF chemistry.
Figure 1.16. Early example of COF-5 prepared by Yaghi through condensation reactions between alcohols and boronates. Adapted from reference [62]. Reprinted with permission from AAAS.
Polymerisation in smectic phases
Apart from the SCSC approach, pre-organisation of the monomers in layered assemblies can in principle happen in liquid crystals as well, providing a smectic mesophase is formed. Works in this
16
INTRODUCTION regard have been reported by Stupp et al.[73] in 1993, in which a rod-like mesogen underwent random crosslinking in a smectic phase. In principle by designing a mesogen having three, four or six binding sites, it is possible to obtain layered assemblies which can undergo in-plane photopolymerisation. The fluidity associated with liquid crystals could be however a double-edged sword: on one side it could help the reacting units coming close to each other, on the other side it could be the cause of in-plane disorder upon cross-linking.
1.3 Post-Polymerisation Modification (PPM)
In terms of applications and tuning of properties, chemical modification of 2DP after their synthesis, or post-polymerisation modification (PPM) is of particular importance. In 1D polymers it can happen on the unreacted binding sites at the extremities of a chain, the so-called end-groups, and/or on the functional groups attached to the repeat units. Similarly in 2DPs, there will be unreacted groups at the edges of the sheet, so-called edge groups, and/or functional groups on its surface and pores (Figure 1.17).
Figure 1.17. Functionalities available for post-polymerisation modification (PPM) in 1D polymers and 2D polymers.
In a recent report, Zhao et al.[74] showed the edge-group modification of 2DP single crystals of the Kory monomer, by reacting the anthracene edge-groups with different dienophiles via Diels-
17
INTRODUCTION
Alder reactions. A dienophile equipped with a fluorescent label, visually showed that the PPM indeed mostly occurred on the edges of the crystals (Figure 1.18).
Figure 1.18. Anthracenes edge-group modification of 2DP single crystals through Diels-Alder reaction with a dienophile equipped with a red fluorescent dye. After PPM, the edges of the crystals are mostly functionalised and fluoresce with a red colour. Adapted with permission from reference [74]. Copyright 2016 American Chemical Society.
While this methodology could not yet be applied to single sheets, it nevertheless opened a new intriguing prospect: PPM could help influence the exfoliation process and widen the range of attainable properties for these new materials. In MOFs, post-synthetic modification is a very powerful and convenient method to influence their properties, such as tuning of the surface area for gas adsorption or to introduce chelating ligands for catalytic purposes[75–77]. Although edge-groups will intrinsically be present on a 2DP sheet (if not quenched during the polymerisation), specific surface or pore functional groups have to be attached on the monomer structure; however, as trivial as it may sound, the functional groups must be carefully selected so that they do not inhibit the proper pre-organisation of the monomers and they should also not be affected by the polymerisation reaction.
1.4 Potential Applications
As 2DPs are intrinsically periodical porous structures, applications as ultra-thin membranes for molecular separation seem realistic[78,79]. The pore size depends on the monomer structure and its lateral extension, but through PPM it could be tuned according to the needs, so that with one particular monomer structure one could obtain a set of 2DPs with different pore sizes. In this regard, such thin membranes could be used for gas separation purposes, providing they operate in the Knudsen flow regime, where the pore size is equal to or smaller than the mean free path of the gas
18
INTRODUCTION molecules. With these conditions, the gas molecules will mostly interact with the pores’ walls rather than other gas molecules and separation will occur according to the molar mass of the molecules[80]. Functional groups inside the pores could in principle influence this process.
Another feature of a 2DP is its structural regularity, which would reflect into regularly distributed functional groups on the polymer surface. This makes a 2DP an ideal molecular scaffold, onto which different functionalisations could be precisely grafted by PPM. It would be of particular relevance for heterogeneous catalysis, where the main issue is usually the very low population and inhomogeneous distribution of catalytically active sites on the surface, which makes the study of the structure-activity relationship challenging[81]. A solution to this central problem relies in the creation of single-site catalysts, where all of the sites are structurally identical. 2DPs could offer the chance of designing well-defined catalysts with which predictable heterogeneous catalysis with clear structure- activity understanding could be developed.
1.5 The Importance of the Monomer’s Design
The successful synthesis of a two-dimensional polymer starts with the design of the monomer. At present, in the literature, despite the growing interest in the field, the cases of successfully synthesised 2DPs are still scarce[33,38–41]. At the time this present work was started in 2012, only a single case of 2DP was reported[33]. Therefore, when designing a monomer, there are some criteria that should be considered in order to reach the goal. These are useful guidelines that can apply regardless of the method chosen for the synthesis of a 2DP, be it the topochemical approach in single crystals or the air/water interface. a) In order to span a periodical two-dimensional network, tessellation with regular polygonal tiles is required. More precisely, in the context of our definition of 2DP, the monomer (tile) should have three, four or six bonding sites, arranged regularly in the geometry of the structure. b) The monomer structure should be as rigid as possible. Shape- persistence will limit the degrees of freedom of the monomer and consequently limit the number of conformations that its structure can assume. This has two main advantages: firstly, the reduced flexibility will facilitate pre-organisation of the monomer prior to polymerisation. This is especially true for the topochemical approach in single crystals, where too much flexibility in a molecule can complicate the goal of achieving the desired crystal packing (for instance through polymorphism) or in the worst case even inhibit crystallisation. Secondly, a shape-persistent monomer will reflect into a shape-persistent polymer. Structural rigidity is of particular importance when one tries to investigate the internal structure of a 2DP for proof of periodicity. This is especially true for the air/water interface approach, where the monomers are compressed mechanically into an ideal crystalline state
19
INTRODUCTION and then polymerised. After polymerisation and transfer onto substrates for analysis, in the absence of the external compression stimulus, a structure composed of flexible repeat-units will relax and possibly collapse into a disordered non-periodical network. c) The connection chemistry between the monomers should imply formation of robust bonds. A 2DP is a free-standing molecular entity, capable of spanning over micrometer-sized holes without collapsing under its own weight. This mechanical stability should be given by chemical bonds that are resistant at ambient conditions (light, moisture, heat) and possibly at higher temperatures. In fact, in the topochemical approach, after polymerisation, the layered two-dimensional polymer crystal has to be exfoliated into its monolayer components; this exfoliation process depends on the strength of the interactions between the layers and might involve a mechanical, chemical or thermal treatment (or a combination of the three), with temperatures way above ambient conditions. As such, strong covalent bonds or strong dative covalent bonds between the monomers are preferred. d) Choice of the reagents that will trigger the polymerisation between the monomers has to be considered carefully: if a reagent at the air/water interface can in principle diffuse to the reacting site from the sub-phase (for instance after being injected into it), in the topochemical approach, diffusion of a reagent homogeneously through a crystal lacking suitable pores can be problematic, if not impossible. Therefore, reagents that do not involve mass transport such as light or heat, should be preferred. e) There must be a minimum of attraction between the monomers, so that the reactive units can be brought close to each other for reaction to occur; ideally, if possible the strongest attractive intermolecular forces should be concentrated on the reactive units only, so that the chances for them to properly come together are increased. f) Bond formation between the monomers should be associated with the least possible molecular movement: going from the monomer to the polymer sheet should involve minimum compression or expansion of the sheet to prevent ruptures and defect formation. g) Last but not least, the monomers should be accessible in terms of synthesis and ideally be available on the gram scale.
This tentative recipe for the design of a monomer is not exhaustive and more criteria will be added during the course of this thesis, at the end of which a more complete picture will be presented.
A few words have to be spent as well on the connection chemistry between the monomers. As mentioned, light or thermal induced polymerisations are preferred and therefore suitable reactive units must be chosen. In this regard, the successful cases of synthesised 2DPs exposed in subchapter 1.2 all rely on monomers based on photoreactive anthracene units. Bond formation between anthracenes relies on a photoinduced [4+4]-cycloaddition reaction which results in an anthracene
20
INTRODUCTION dimer (Scheme 1.1). The reaction is reversible thermally or upon light irradiation at low wavelengths, typically below 300 nm).
Scheme 1.1. Anthracene dimerisation via photoinduced [4+4]-cycloaddition and thermally- or photo- induced back-reaction.
Upon dimerisation, bond formation usually happens at the 9 and 10 positions, where the highest π-electron density is concentrated. The reaction is well studied both in solution[82–84] and solid state[85], where in the latter case the p-orbitals of the 9 and 10 positions have to be aligned and overlap properly for reaction to occur; this usually is the case when the anthracenes are stacking in face-to-face (ftf) relationship at distances below 4.2 Å. In the context of 2DPs, 1,8-substituted anthracenes are particularly attractive as upon dimerisation/polymerisation, the amount of molecular movement in terms of expansion and shrinkage is minimal[31] (Figure 1.19). When dimerisation occurs between two anthracenes at the van der Waals distance of 3.5 Å, bond formation considerably draws near to one another the 9 and 10 positions; however the 1 and 8 positions only suffer a minimal expansion, making them the preferred positions for substituents. As such, for increasing the chances of a successful polymerisation (especially for what regards SCSC transformation), the anthracene reactive units should be attached to the skeleton of the monomer via the 1,8-positions. This is however just a guideline, as for the fantrip monomer presented in 1.2.1[39], a successful SCSC transformation was obtained with anthracenes substituted at the 2,3 position. What really counts in the end is the volume change of the elementary cells of monomer and polymer crystals, which has to be ideally as minimal as possible.
Figure 1.19. Structural changes between a ftf-stacked anthracene pair and its corresponding dimer. The atoms at the 1 and 8 positions undergo only minimal spatial shifts upon dimerisation.
21
AIM OF THE THESIS
1.6 Aim of the Thesis
As 2DPs are still in their infancy, there is a need for new monomer and polymer systems, to widen the field and to lay the foundations for a better understanding of how to access these new materials and their potential properties and applications. In particular, it is desirable to have a versatile monomer structure, which can be employed for both the single crystal and the air/water interface approach, so that a direct comparison of the two methods can be done in terms of feasibility, structural perfection of the polymer obtained and characterisation. The present thesis contributes in three main directions to this challenge:
- We want to design and establish the synthesis of a basic structural monomer and study its crystal packing behaviour, to see if it is suitable for the single crystal approach. In case the proper crystal packing cannot be obtained, we want to understand how it is possible to steer the packing into the desired direction.
- We want to investigate if upon modification, the same monomeric structure can be used for the synthesis of 2DPs at the air/water interface and see if the packing behaviour of the monomer at this interface differs from the one in the single crystal. Providing the proper packing is obtained, a 2D polymerisation at the interface will be performed, aiming at a novel 2DP.
- Finally, due to the uncertainty associated with obtaining the proper crystal packing for a topochemical 2D polymerisation to take place, it would be desirable to have other monomer structures for the single crystal approach. Optimally, such monomers would pack in the desired fashion without the need of tuning the crystallisation conditions such as solvent choice. It would also be desirable to have monomers which possess versatile functional groups, to pave the way for possible post-polymerisation modifications.
22
ANTHRAPHANE
2. Anthraphane Monomer
The following sections will be dealing with the „anthraphane“ monomer, which was conceived for the topochemical synthesis of 2DPs in single crystals. The design of the monomer will be addressed, followed by the synthetic approach employed. The crystallisation of the monomer will then be discussed along with a thorough analysis on its packing behaviour in the single crystal. Some thoughts on crystal engineering will be addressed, particularly the question on how one can steer the packing of the monomer into the desired direction. Finally, some single-crystal-to-single-crystal (SCSC) topochemical reactions of the anthraphane will be presented and discussed.
2.1 Monomer’s Design
The structure of the monomer 1 used in this work is depicted in Figure 2.1. During the design process, the considerations exposed in subchapter 1.5 were taken into account: the structure is very rigid and shape-persistent with three photoreactive binding sites, capable of forming C-C covalent bonds upon photoirradiation.
Figure 2.1. Chemical structure of the D3h-symmetric hydrocarbon propeller 1 synthesised in this work. The bonding sites, in this case photoreactive anthracene units, are marked in red colour.
The monomer is a hydrocarbon composed of three 1,8-anthryl units, orthogonally connected to two central benzene cores through acetylene spacers, giving it the shape of a concave trigonal
[86] prism possessing D3h symmetry. The molecule is formally a cyclophane and due to its structural features, the compound was conveniently named “anthraphane”: a portmanteau word of “anthracene” and “cyclophane”. The 1,3,5-substitution pattern on the benzene cores spatially separates the anthracene blades by 120° and therefore hampers the possibility of intramolecular dimerisation. The intrinsic geometry of the molecule also ensures exclusively anti-photodimers. Furthermore steric factors minimise the chance of intermolecular [2+2]-cycloaddition reactions
23
ANTHRAPHANE between triple bonds‡[87] and intermolecular [4+2]-cycloadditions between triple bonds and anthracene moieties[33,88,89] (see Figure 2.2).
120
Figure 2.2. a) Cartoon representation of anthraphane viewed along the C3-rotational axis: the photoreactive anthracene blades are spaced apart by 120°, preventing intramolecular reactions. b) Two anthraphane fragments approaching each other in anti-fashion, with the anthracene moieties stacking face-to-face (ftf). c) Further interpenetration of the fragments is sterically hindered, so that [2+2]-cycloaddition between the triple bonds is not possible and [4+2]-cycloaddition between triple bond and anthracenes is unlikely. To visualise the repulsive H-H interaction, the anthracenes were shifted in z-direction.
As a result, photoreactions of compound 1 are restricted to the intermolecular, anti-selective [4+4]-cycloaddition, providing the anthracenes are stacking in a face-to-face (ftf) arrangement. The monomer was designed in the hope of forming lamellar single crystals upon crystallisation, with the central parallel benzene cores acting as glue between the layers through π-stacking interactions.
Despite the rigid look, in reality the structure can distort itself to a certain degree from the ideal D3h symmetry as can be seen in Figure 2.3.
‡ [2+2]-Cycloadditions between acetylenic units are rare as the resulting cyclobutadienes are very unstable and can be generally detected only by using trapping reagents such as dienes through Diels-Alder reaction. 24
ANTHRAPHANE
Figure 2.3. Top view of anthrapahane 1 and possible distortions from the ideal D3h symmetry: triple- bond bending and torsion along the C3 symmetry axis.
Bending of triple bonds moieties from the ideal linear geometry is a very common phenomenon both predicted by theory and observed experimentally for a wide number of compounds, especially strained ones[90,91]. A minimum of distortion in this regard can therefore be expected for anthraphane. The same applies for the torsion along the C3-symmetry axis, which would imply a minimal compression of the structure with a shortening of distance between the benzene cores and a tilting of the anthracenes with respect to the ideal orthogonal geometry with the central cores. These distortions, even if encountered, would involve only a twisting or bending of few degrees, therefore the overall shape of the monomer would still be retained and its pre-organisation should not be negatively affected by it (for more details refer to subchapter 1.5).
As a final remark, it is important to point out why this structure was conceived best for the topochemical approach to 2DPs. In a conventional approach, i.e. a hypothetical solution polymerisation, when two monomers approach each other, each anthracene unit will always have two faces available for stacking and subsequent dimerisation. More precisely, each monomer has three anthracenes, each with two possible reacting sites. An important consequence of this regioselective dichotomy is the number of potential regioisomers that could arise after successive dimerisations. As the reacting fragment grows further into an oligomeric and polymeric entity, the regioselectivity of the dimerisations starts to have an important impact on the structural regularity of the polymer. As illustrated in Figure 2.6, a possible regioisomer of a fragment with 25 repeat units, displays significant defects in its structure. This results in chiral domains, unreacted sites and strong deformations which deviate from the desired hexagonal, honey-comb structure. In practice, a randomly cross-linked polymeric network would be created, lacking internal order. In this sense, a solution polymerisation with anthraphane in order to create a 2DP would probably not make sense. 25
ANTHRAPHANE
Figure 2.6. Hypothetical oligomeric fragment of anthraphane with 25 repeat units. As the polymerisation reaction continues, structural defects due to regiochemical mismatches can become important and compromise the internal order of the polymer. In this case, the deviation from the ideal hexagonal-like structure shown in the bottom left corner of the fragment is evident.
As already mentioned above, other systems were successfully developed for the topochemical synthesis of 2DP by exploiting controlled photo-induced growth confined in two dimensions in layered single crystals[33,38–40]. These successful synthetic strategies required the monomers to have at least three photoreactive units embedded in the proper geometry, allowing a two-dimensional growth, brought about by the well-studied photo-induced dimerisation of anthracenes via [4+4]-cycloaddition[83,84]. In particular, monomers 4 and 5, with their structures consisting of three anthracene blades fixed in a three-fold symmetric, shape-persistent arrangement, seemed to be helpful in attaining a packing useful for lateral polymerisation. Anthraphane, having similar structural features, appeared to be an ideal candidate as new monomer for the synthesis of 2DPs (Figure 2.7).
26
ANTHRAPHANE
Figure 2.7. Chemical structures of monomers 4 and 5 and anthraphane 1 presented in this study. All have a propeller geometry with three anthracene blades embedded in a three-fold symmetry (1,4: 1,8-anthryl; 5: 2,3-anthryl). Monomers 4 and 5, were successfully employed for the synthesis of 2DPs, making anthraphane a promising new potential monomer.
A schematic representation of the three essential steps for synthesising 2DPs by the topochemical approach is depicted in Figure 2.8: 1) the monomer, is expected to crystallise into layered single crystals. In each layer, the molecules should pack in a hexagonal honeycomb-like fashion with the photoreactive anthracene units stacking face-to-face at close distance (ideally below 4.2 Å). 2) Upon photoirradiation, a [4+4]-cycloaddition is triggered which polymerises the monomers in the layers topochemically; each layer is so converted into a 2DP and the monomer crystal becomes a 2DP crystal. 3) Exfoliation of the polymer crystal can then occur chemically, mechanically and/or thermally, affording 2DPs as molecular single layered entities.
Figure 2.8. Schematic representation of the three crucial steps for the topochemical synthesis of 2DPs in single crystals.
27
ANTHRAPHANE
2.2 Synthetic Approach
For the synthesis of anthraphane, the retrosynthetic approach depicted in Scheme 2.1 was considered. The monomer was disconnected into two main structural units: the anthracene moiety and the central benzene core. To achieve the target structure, only composed by sp2 and sp carbon atoms, the chemistry of transition-metal catalysed cross-coupling reactions for the formation of sp2- sp carbon-carbon bonds was chosen. Therefore a 1,8-disubstituted anthracene and 1,3,5- triethynylbenzene were selected as key building blocks and the well-known palladium mediated Sonogashira cross-coupling reaction as connection chemistry[92–95].
Scheme 2.1. Chemical structure and retrosynthetical analysis of anthraphane 1. As key building blocks, 1,3,5-triethynylbenzene and a 1,8-disubstituted anthracene were selected and as connection chemistry, the Sonogashira cross-coupling reaction.
With the cross-coupling partners selected, the most convenient synthetic pathway was conceived as follows: the monomer would be sequentially assembled by a first cross-coupling involving 1,3,5-triethynylbenzene with a large excess of the anthracene unit to promote a selective trisubstitution and obtain the flat monomer precursor 11, which upon a second Sonogashira cross- coupling with 1,3,5-triethynylbenzene would form the target anthraphane 1 (see Scheme 2.2 a).
28
ANTHRAPHANE
Scheme 2.2. a) Retrosynthetical strategy involving two consecutive Sonogashira cross-couplings from precursor to monomer. b) Precursor to monomer reaction: this complex step first involves an intermolecular cross-coupling with 1,3,5-triethynylbenzene, followed by two intramolecular cross- couplings.
Despite the relative simplicity of the retrosynthetic scheme, one should point out the complexity of the intended reaction from precursor to monomer (see Scheme 2.2 b). This is in fact a step which involves first an intermolecular coupling with 1,3,5-triethynylbenzene, followed by two intramolecular cross-couplings. In order to favour the intramolecular reaction, this step should be ideally performed under high dilution or pseudo-high dilution conditions[94,96–98].
29
ANTHRAPHANE
2.3 Synthesis of Anthraphane
As coupling partner for the Sonogashira reaction with the commercially available 1,3,5- triethynylbenzene 10, it was desirable to use a building block that could be easily synthesised in gram scale and with little effort.
Scheme 2.3. (a) Synthetic route for the novel synthesis of key building block 9. (b) Sequential Sonogashira cross-coupling reactions of 9 and 1,3,5-triethynylbenzene 10 to yield anthraphane 1. Despite the complexity of the final step, a difficult sequence of intermolecular- and intramolecular couplings, surprisingly high yields ranging from 28-40% were obtained.
The natural choice of using the halogenated 1,8-dibromoanthracene[99] or 1,8- diiodoanthracene[100] was discarded as their synthesis, starting from 1,8-dichloroanthraquinone, involves refluxing toxic nitrobenzene for the halogen exchange reaction, followed by two sequential reductions of the anthrone moieties with sodium borohydride proceeding with moderate yields. The efforts were therefore concentrated on the anthracene-1,8-ditriflate 9, as the triflate groups are generally introduced with high yields, confer solubility, and their reactivity towards oxidative addition lies between that of bromide and iodide[95]. However, this required the existing synthesis of 9 to be substantially improved[101]. The previously known route passed through a couple of tedious and time-consuming steps which rendered the synthesis a demanding procedure (see Scheme 2.4). More in detail, the old method involved the acetylation of 6, followed by a poorly selective and low yielding heterogeneous reduction with zinc powder, the major products of which, after a second acetylation, were 1,8-diacetoxyanthracene 13 and the overreduced 1,8-diacetoxy-9,10- dihydroanthracene, obtained as a mixture. In a final step, a rather tedious in-situ deprotection of 1,8-
30
ANTHRAPHANE diacetoxyanthracene with methylamine, followed by the actual triflation step, afforded 9. Due to the high instability of 1,8-hydroxyanthracene under basic conditions and in the presence of oxygen, the final deprotection and triflation steps had to be performed in strictly inert atmosphere and carefully degassed solvents. In addition, column chromatography was employed as purification method for every step, increasing the work up times considerably, especially for large-scale synthesis. This five step synthesis had an overall yield of 27% and required about a total of three weeks of work from starting material to the final anthracene-1,8-ditriflate 9.
Scheme 2.4. Old synthetic strategy for the synthesis of the key building block anthracene ditriflate 7 compared to the method developed in this work.
A new route was therefore devised (Scheme 2.3 a) which provided the desired intermediate 9 with a total yield over three steps of up to 71%, on the 15 g scale and within one week, accounting for reactions and work-up time. The new route starts from 1,8-dihydroxyanthraquinone 6, whose lithium aluminium hydride reduction leading to the stable 1,8-dihydroxy-9,10-dihydroanthracene 7 had already been reported[48]. By applying standard triflation conditions to 7 with triflic anhydride and pyridine in dichloromethane, the ditriflated compound 8 can be smoothly obtained in virtually quantitative yield on the 15 g scale within 2 h time. Purification involves standard workup, followed by treatment of the crude product with activated charcoal in boiling hexane and hot filtration over celite. This triflation step is an asset of the whole procedure as it is performed on 7, which is a phenol derivative and thus not sensitive under the basic conditions required for triflation, unlike the 1,8-
31
ANTHRAPHANE dihydroxyanthracene used in the traditional synthesis. In the final step, aromatisation to the target compound 9 is achieved in yields up to 93% by refluxing 8 with DDQ in dry dioxane for 5 h. Simple filtration over a silica plug affords the pure product (see Figure 2.9), whose elution can be followed conveniently under UV light at 366 nm. p-Chloranil was also tested as aromatisation reagent but found to be less effective than DDQ and was therefore not investigated further. Ditriflate 9 is well soluble in most organic solvents and can be stored in the dark for months. For further purification, 9 can be recrystallised from boiling hexane to afford pearly white needles. The advantage of this novel synthetic path lies in the fact that the work-ups are fast and simple, not involving tedious column chromatography.
1 Figure 2.9. H-NMR spectrum of pure compound 9 in CDCl3.
In the next stage, anthraphane precursor 11 was assembled (Scheme 2.3 b). A one-step reaction was performed by reacting an excess of ditriflate 9 (typically 5 – 6 eq) with the benzene core
10 in dioxane under standard Sonogashira conditions with Pd(PPh3)4 and CuI, as catalyst and co- catalyst respectively, and triethylamine as base. As expected, due to the rigidity of its structure, compound 11 precipitated during the reaction, but surprisingly once collected by filtration, the greenish solid was already pure according to NMR. Minor inorganic impurities responsible for the green colouration were removed by filtering the crude product over celite with chloroform, resulting
32
ANTHRAPHANE in the pure precursor as a bright yellow solid in yields up to 60% (see Figure 2.10). Precursor 11 was recrystallised from tetrachloroethane and its structure confirmed by SC-XRD (see section 2.4 for a more detailed discussion). It is noted that when working with relatively large amounts of 11 (above 1 g), due to its poor solubility large amounts of solvent have to be used for the filtration over celite and some material can get lost in this procedure.
1 Figure 2.10. H-NMR spectrum of precursor 11 in CD2Cl4.
Despite the appreciably high yield of 60% for this step and its relative simplicity, attempts to further optimise the precursor synthesis were performed: it was found that dioxane was the ideal solvent for the reaction, being a very good solvent for the starting material 9 and a very poor solvent for the product 11. Concentration did not play a big role (reactions performed in a range of 20 mM to 140 mM) and so did not the base employed, however it was found beneficial to use stoichiometric amounts of base rather than using it as co-solvent. This is most likely due to triethylamine being a poor solvent: when present in small amounts, the starting material and the various substituted intermediates are probably better solvated during the reaction. Attempts to slowly add 1,3,5- triethynylbenzene during the reaction did not result in any product. It is speculated that by using slow addition, during the course of the reaction, the precursor is formed at low concentrations at 33
ANTHRAPHANE which it remains soluble and can further react with more 1,3,5-triethynylbenzene and ditriflate forming high molecular mass entities. Another potential issue by using slow addition with a syringe pump, is the difficulty of keeping the solution inside the syringe strictly degassed; it is possible that despite using a gas-tight syringe, over a prolonged time, small amounts of oxygen could diffuse into the solution being added. It is in fact well-known that the presence of oxygen during a Sonogashira reaction, can trigger the competitive and unwanted Glaser-coupling between acetylenic moieties (in this case 1,3,5-triethynylbenzene), which can have detrimental consequences in terms of side- products formed and low yields of the reaction[92,102]. As a final remark, it is worth noting that the excess of ditriflate 9 could be recovered virtually quantitatively by subjecting the final reaction mixture to flash chromatography.
The anthraphane monomer 1 was then prepared by reacting 11 with an equimolar amount of
10, using Pd(PPh3)4 as catalyst in toluene. To minimise side reactions in this complex sequence of intermolecular- and intramolecular couplings, copper-free Sonogashira (Heck alkynylation) and high dilution (1-2 mM) conditions were employed. This afforded the target cyclophane 1 as a pale yellow solid in rather impressive yields ranging from 28-40%. Pseudo-dilution conditions to reduce the amount of solvent used were not considered as precursor 11 is only sparingly soluble in toluene; thus, slow addition of the reagents via syringe and control of the stoichiometry would not be possible. Despite the low solubility of the starting material, upon addition of the catalyst and heating, the reaction mixture from suspension slowly turned into a clear solution, indicating the onset of the reaction. First attempts to optimise this final step were performed: the catalyst amount was reduced from 50 to 20 mol-%, concentrations up to 2 mM were employed and 200 eq of base were found to be optimal. Reaction times of 4-5 days were employed. With these conditions yields of 40% could be achieved. Lower amounts of base such as 100 eq resulted in yields of 23-26%, consistent with the base concentration dependence on the reaction rate of copper-free sonogashira reactions[103]. Typical experiments were conducted with 700 mg of precursor and 300 mL of toluene as solvent and yielded approximately 180 mg of product. Worth of note is the simplicity of the work-up for this final step: the reaction mixture is simply filtered, concentrated and the residue washed with methanol. The obtained yellow solid can then be recrystallised from boiling TCE to obtain pure anthraphane as yellow needles. A control experiment using standard Sonogashira conditions with copper (5% mol Pd0 and 5% mol CuI) resulted in no product at all and the starting material was partially recovered from the reaction. This suggested the importance of working with copper-free conditions. From this experiment one can assume that due to the large amounts of solvent employed, thorough degassing of the reaction mixture can become difficult to achieve; as a consequence one can assume that consumption of the 1,3,5-triethynylbenzene core probably occurs by Glaser-coupling. Compound 1 was characterised by 1H-NMR spectroscopy (see Figure 2.11) and high-resolution mass spectrometry, 34
ANTHRAPHANE but its low solubility prevented analysis by 13C-NMR spectroscopy. However, the proposed structure for anthraphane was unambiguously confirmed by SC-XRD analysis, due to its tendency to readily form single crystals from a variety of solvents (see subchapter 2.5).
1 Figure 2.11. H-NMR spectrum of anthraphane 1 in CD2Cl4 measured at 80°C. Due to the insolubility of 1 a clear 13C-NMR spectrum could not be measured.
The solution UV/VIS and fluorescence spectra of 1 do not show signs of intramolecular cross- talk between the anthracene units and are virtually superimposable with the spectra obtained from ditriflate 9 and precursor 11 (see Figure 2.12). The vibronic structure attributed to the anthracene units is clearly visible, as well as the expected bathochromic shift of spectra for both anthraphane and its precursor due to an increased conjugation in their structures compared to the ditriflate. Anthraphane does not melt and decomposes at around 280°C, gradually tarnishing into a dark brown powder.
35
ANTHRAPHANE
Figure 2.12. UV/Vis absorption (solid lines) and fluorescence (dashed lines) spectra of the ditriflate 9, precursor 11 and anthraphane 1 measured in TCE. Concentration employed for absorption is 15 μM, concentration for emission is 2.5 μM. Excitation wavelength for emission: λ = 365 nm.
2.4 Single Crystal Structure of Precursor 11
Before moving into the crystallisation of anthraphane, precursor 11, also bearing three photoreactive anthracene units was considered as potential monomer for topochemical 2D-polymerisations as well. Rectangular yellow prisms grown by slow vapour diffusion of hexane into a 1,1,2,2-tetrachloroethane solution could be obtained and analysed by SC-XRD. The virtually flat structure of the precursor was confirmed (see Figure 2.13).
36
ANTHRAPHANE
Figure 2.13. ORTEP diagram of 11 (50% probability). A 70:30 disorder was found in all OTf ligands and the solvent molecule (shown in the image and fully modelled).
However, despite packing in tight layers (Figure 2.14), only one out of the three anthracenes was engaged in a poor parallel displaced π···π interaction, with distances between the 9 and 10 positions way above the Schmidt’s regime[104] (4.62 ŧ), as shown in Figure 2.15; similar results were also obtained with nitrobenzene and o-dichlorobenzene. Compared to cyclophane 1, in which the anthracenes in terms of π-π interactions can only interact among themselves and the acetylenic moiety, in the nearly flat precursor molecule the accessible π-surface is much larger, including the central benzene core and corresponding effectively to the entire carbon skeleton of the molecule itself. For this reason, the number of possible interactions and consequently crystal packings becomes even larger, making this molecule unattractive as monomer candidate.
Figure 2.14. Layered crystal structure of 11. The molecule assumes a nearly flat conformation and packs in tight layers separated by approximately 3.4 Å (solvent omitted for clarity).
§ For successful topochemical photoreactions above the Schmidt distance, see reference[193]. 37
ANTHRAPHANE
Figure 2.15. Details of the only π···π interaction among one of the anthracenes in 11. The parallel displaced interaction exhibits distances between the 9 and 10 positions of 4.618(4) Å.
2.5 Crystallisation of Anthraphane
2.5.1 First insights into the packing behaviour
Having solved the synthetic issue and with proper amounts of monomer in hand, the next major step, namely the crystallisation of anthraphane, was investigated. As expected, due to the high rigidity and lack of functional groups, the solubility of the molecule was very poor: in TCE, one of the best solvents for anthraphane, the solubility was found to be as low as 0.2 mg/mL at room temperature. Compound 1 could therefore be best crystallised by slow cooling of nearly saturated solutions in high boiling point solvents.
As a first test, a small amount (2-3 mg) of monomer was crystallised from boiling TCE: by cooling the solution to room temperature overnight in an oil bath, nice clear yellowish needles in sizes ranging from 100-150 µm were formed (see Figure 2.16). The crystals exhibited strong birefringence when observed between crossed polarisers and their quality was good enough for SC- XRD analysis.
38
ANTHRAPHANE
Figure 2.16. Optical micrographs of single crystals grown from boiling TCE.
SC-XRD analysis showed that anthraphane packed indeed in a layered fashion as required for the topochemical synthesis of 2DPs. As can be seen in Figure 2.17, the layers are tightly packed and appear wavy due to distortions in the triple bonds (anthracene units are displayed in red, solvent omitted for clarity).
Figure 2.17. Layered crystal structure of the TCE solvate of anthraphane. The layers appear wavy due to distorsions in the triple bonds (solvent and hydrogen atoms omitted for clarity).
However, when looking at how the molecules are arranged inside every layer, in particular how the anthracene units are interacting with each other, one sees that the criteria for a topochemical reaction are not fulfilled. In fact, as displayed in Figure 2.18, the anthracene units in the layers are all in an exclusive edge-to-face relationship (etf), interacting with each other by the
39
ANTHRAPHANE means of CH···π interactions. As expected, the voids between the anthraphanes are efficiently filled with TCE molecules.
Figure 2.18. Detailed view from top of a single layer in the crystal structure of the TCE solvate of anthraphane. The anthracene units are all arranged in an edge-to-face (etf) geometry.
This first crystal structure showed a propensity of anthraphane to avoid face-to-face interactions in its crystal packing, choosing instead an edge-to-face arrangement. In particular, each anthracene unit is engaged into a triplex of mutual etf CH···π interactions (shown in red in Figure 2.18). This result should come as no surprise, since etf interactions along with parallel-displaced ftf interactions are considered to be energetically more favorable than parallel ftf ones[105]. The element that perhaps is of more relevance here is that the solvent gets incorporated into the crystal structure forming a solvate; with this information in hand, we hoped that by varying the solvent for crystallisation, we would possibly be able to influence the packing of anthraphane.
2.5.2 Interactions in single crystals of anthraphane
When one thinks about the nature of an organic (molecular) crystal, “the supramolecule par excellence”[106], and how its solid construction comes about through the process of crystallisation, the archetype of a supramolecular reaction, one should think about how the molecules recognise and interact with each other and how they are being held together in the crystal. The glue that keeps the molecules packed together in these supramolecular assemblies can be the strong hydrogen bonding interactions and the weaker intermolecular van der Waals forces, namely the electrostatic 40
ANTHRAPHANE ion-induced dipole, ion-dipole and dipole-dipole forces (Keesom interactions), the permanent- induced dipole forces (induction or Debye forces), and the weaker dipole-induced dipole forces (London dispersion forces). When considering the intermolecular forces involved with anthraphane, a hydrocarbon, Keesom and Debye forces are not expected to play a major role, while London dispersion forces are clearly the dominant intermolecular interaction. Since anthraphane is exclusively composed by aromatic units and π-surfaces, the intermolecular interactions will involve mainly specific π-π and CH-π interactions. Regarding the still debated and controversial meaning of the term π-π interaction, it is not the scope of this work to elucidate the nature of the interactions involving stacked aromatics (ftf sandwich or parallel displaced), namely whether the dominant component of the interaction is of electrostatic (through the quadrupole moments of the aromatic unit) or dispersion nature. Detailed and insightful studies regarding the benzene dimer can be found in the literature[105,107–109]. We however accept that the shape of the quadrupolar moment of the aromatic units can influence the geometries of such π-π interactions. We also consider the study made by Grimme[110], in which he compared the interaction energies of the dimers of different polycyclic aromatics (PAHs) with the dimers of their saturated counterparts. The results indicate that for anthracene and higher PAHs, there are in fact special types of dispersion effects (attributed to electron correlations in p-orbitals), which could be attributed to a π-π stacking effect. In this work we will thus use the term π-π stacking and π-π interactions as a geometrical descriptor for the orientation of the aromatic units, namely when the anthracenes are ftf-stacked, and when analysing the intermolecular forces involved with anthraphane. For edge-to-face (etf) relationships between aromatics we will instead use the term CH-π interaction and consider it as a weak hydrogen bond. The hydrogen bonding characteristics of this interaction, especially in terms of its directionality and bond length has been demonstrated by different studies[111–115] and is commonly accepted in the literature**. As comparison, a CH-π interaction between benzene and ethane amounts to roughly 1.82 kcal/mol whereas the energy of one hydrogen bond in the water dimer is 5.00 kcal/mol and for the hydrogen fluoride dimer 39.0 kcal/mol[116,117]. However, despite having small energy when considered singularly, CH-π interactions are additive and can act cooperatively[112], becoming important in magnitude, especially in supramolecular entities.
It is perhaps the directionality and cooperativity of these interactions that can be used as rationale to explain why anthraphane formed a layered crystal structure: each anthracene unit on the monomer has three available protons for CH-π interactions on its long edge, which can form three parallel hydrogen bonds when coming into contact with the π-surface of a neighbouring
** There is some debate as to whether the term hydrogen bond can be applied to describe a CH-π interaction: the main argument relies in the fact that hydrogen bonds are of electrostatic nature, while CH-π interactions rely mainly on dispersion forces[194]. 41
ANTHRAPHANE anthracene (see Figure 2.19). In order to exploit all three hydrogen bonds, the interacting anthracenes will have to rest vertically more or less in the same plane (indicated in light blue colour in Figure 2.19), so that for each anthraphane, there is an in-plane molecular recognition with its neighbour through CH-π hydrogen bonds. As a consequence when crystallising, layers of anthraphane will be formed. The layers can then efficiently stack with each other by exploiting π-π interactions between the central benzene cores of the cyclophane.
Figure 2.19. To maximise the interactions through CH-π hydrogen bonds, the anthracene units are resting vertically more or less in the same plane. As a consequence of this directionality of the CH-π interactions, the anthraphane molecules will also be on the same plane when crystallising, forming layers in the crystal.
This kind of two-dimensionally restrained molecular recognition would be of particular interest for the synthesis of 2DPs in solutions, where during polymerization, growth should also be restricted in a plane, in order to avoid formation of cross-linked networks[31]. As a final remark, when analysed in the context of hydrogen bonding, it is noteworthy that each anthracene unit is involved in a two-body interaction, acting both as a hydrogen bond acceptor (through its π-surface) and donor (through its sigma C-H bonds on the long edge), with two anthracene neighbours, also acting as donors and acceptors and forming the observed triangular arrangement (or triplex) of mutual CH-π hydrogen bonds seen in the crystal structure (see Figure 2.20).
42
ANTHRAPHANE
Figure 2.20. Detailed view of the CH-π interactions in the TCE solvate. Each anthracene unit acts as hydrogen bond donor (C-H sigma bonds on the long edge) and acceptor (π surface), donating and receiving three hydrogen bonds from its two neighbours. The anthracene units thus associate into a triplex of mutual and in-plane CH-π hydrogen bonds.
Interestingly, this kind of triangular (or cyclic) motif, was also found to be the most stable configuration for the benzene trimer in the gas phase both experimentally[118] and computationally[119]. Having understood what kind of intermolecular interactions occur between anthraphane molecules, it can be now reasoned how these interactions can be influenced or disturbed if one varies the solvent for crystallisation.
2.5.3 Solvent choice for crystallisation
Crystal structure prediction is still a huge challenge nowadays and it becomes even more challenging if not impossible when one has to consider the inclusion of guest molecules, in this case solvent, which co-crystallises with the main molecule forming a solvate. Another factor that one has to consider is that during crystallisation there is also a packing factor: in a crystal, the molecules not only associate and recognize themselves supramolecularly (by the means of intermolecular interactions), but they also try to pack as tightly as possible, avoiding unnecessary voids in the crystal structure††. A systematic approach for the choice of the solvent was therefore devised, taking into account these supramolecular and geometrical factors.
†† For a detailed discussion into the packing behaviour of molecules, please refer to subchapter 2.7. 43
ANTHRAPHANE
Figure 2.21. Systematic approach for solvent screening. Steric and electronical factors of the solvent were considered.
Some of the factors that were considered are illustrated in Figure 2.21. For example, regarding sterics, how does the solvent influence the packing if it is bulky or small? How about the symmetry of the solvent molecule? And in terms of solvent-anthraphane interactions (electronical term) in the packing: how does a polar or apolar solvent influence the packing? How about aromaticity? In this latter case, how about if the solvent has donor or acceptor character? These dichotomies are general (one could in principle go into more detail when considering interactions of polar nature, i.e. dipole-dipole, H-bonding) but encompass all the factors that we wanted to take into account for this study. Moreover they have to be combined with each other, as molecules have both well-defined sterical and electronical properties, so that for instance a solvent can be bulky, symmetric, apolar and aromatic with a donor character. Among those combinations, in the case of anthraphane, it is perhaps natural for the reader to think that the most interesting crystal packings would be obtained by using aromatic solvents able to disturb or create new CH-π interactions, however aliphatic solvents can also contribute to such interactions by acting as CH-π hydrogen bonding donors[112].
2.5.4 Solvent screening
As mentioned earlier, the choice of the solvent was restricted by the solubility of anthraphane. We therefore started screening a variety of high-boiling point solvents meeting the characteristics
44
ANTHRAPHANE illustrated in the previous sub-chapter. The screening procedure was conducted as follows: a small amount of anthraphane (2-3 mg) was suspended in the desired solvent (1-2 mL) and the resulting mixture was heated under stirring close to the boiling point of the solvent (Tmax). No decomposition of anthraphane was observed for prolonged heating above 200°C. The results are summarised in Table 2.1, which shows the 69 solvents used for the solubility tests ordered by ascending boiling point with their dielectric constants (as a measure of polarity) and viscosities. The red, orange and green colour-codes mirror the solubility of anthraphane at Tmax: red means poor/insoluble, orange means OK soluble and green means good soluble with clear solutions of the cyclophane obtained. When looking at the table, the first aspect that becomes evident is the existence of a solubility threshold: below 170°C, anthraphane is generally quite insoluble except for some remarkably good solvents such as TCE, bromoform and 2,4,6-collidine, whereas at higher temperatures, it is generally well soluble in different solvents. One can therefore draw a solubility line at around 144°C, under which temperature it is virtually impossible to solubilise the molecule at concentrations practical for crystallisation purposes.
45
ANTHRAPHANE
Table 2.1. Solvents screened for crystallisation. The solvents are ordered by ascending boiling point. The term “good” means that a clear solution is obtained at Tmax.
[120] [121] Solvent Bp Tmax εr µ Solubility (20°C) (20°C) at Tmax [cP] 1 acetone 56°C 54°C 20.70 0.29 poor 2 hexafluoroisopropanol 58°C 55°C 16.70 1.65 poor 3 THF 66°C 64°C 7.60 0.48 poor 4 carbon tetrachloride 77°C 75°C 2.24 0.90 poor 5 chloroform 61°C 60°C 4.81 0.56 poor 6 ethyl acetate 77°C 75°C 6.02 0.43 poor 7 2-butanone 79°C 76°C 18.50 0.43 poor 8 benzene 80°C 80°C 2.28 0.61 poor 9 hexafluorobenzene 80°C 75°C 2.05 1.20 poor 10 cyclohexane 81°C 80°C 2.02 0.88 insoluble 11 acetonitrile 82°C 80°C 37.50 0.34 insoluble 12 1,4-difluorobenzene 88°C 80°C 2.26 - poor 13 dioxane 101°C 100°C 2.21 1.37 poor 14 2-methyl-3-butyn-2-ol 103°C 101°C - 3.43 poor 15 octafluorotoluene 104°C 100°C - - insoluble 16 piperidine (solution darkens) 106°C 104°C 5.90 1.57 poor 17 toluene 111°C 109°C 2.40 0.59 poor 18 pyridine 115°C 114°C 12.30 0.88 poor 19 1-butanol 117°C 115°C 17.80 2.57 poor 20 tetrachloroethylene 121°C 119°C 2.30 0.89 poor 21 cyclopentanone 131°C 128°C 14.45 1.13 OK 22 chlorobenzene 131°C 128°C 2.71 0.74 OK 23 triethyl orthoformate 143°C 141°C - - poor 24 o-xylene 144°C 142°C 2.57 0.81 OK 25 TCE 146°C 143°C 8.42 1.70 Good 26 bromoform (solution darkens) 146°C 130°C 4.40 1.86 Good 27 DMF 153°C 150°C 36.71 0.92 poor 28 α-pinene 155°C 153°C 2.64 1.29 insoluble 29 cyclohexanone 155°C 150°C 18.20 2.02 OK 30 pentafluorobenzaldehyde 164°C 162°C - - insoluble 31 mesitylene 165°C 162°C 2.40 0.66 poor 32 2,6-dimethyl-4-heptanone 169°C 167°C 9.91 1.00 poor 33 2,4,6-collidine 171°C 155°C 12.02 0.80 Good 34 limonene 176°C 170°C 2.36 0.92 insoluble 35 1,1,3,3-tetramethylurea 177°C 175°C 24.46 1.50 Good 36 diethylformamide 177°C 175°C 29.02 1.25 poor 37 eucalyptol 177°C 175°C 4.84 2.70 insoluble 38 o-dichlorobenzene 180°C 175°C 9.93 1.39 Good 39 butylbenzene 183°C 180°C 2.36 1.07 poor 46
ANTHRAPHANE
40 dimethyl sulfoxide 189°C 187°C 46.70 1.99 insoluble 41 t-butyl toluene 191°C 185°C - 1.25 poor 42 o-cresol 191°C 185°C 5.00 35.06 (45°C) Good 43 benzonitrile 191°C 180°C 26.00 1.27 Good 44 NMP 202°C 195°C 32.00 1.67 Good 45 hexachloroacetone 204°C 190°C 3.99 - Good 46 methyl cyanoacetate 204°C 200°C 28.80 2.61 Poor 47 GBL 204°C 195°C 39.10 1.70 Good 48 1,2-dimethoxybenzene 206°C 180°C 4.21 - Good 49 1,1,3,3-tetraethylurea 211°C 208°C 14.74 - Good 50 nitrobenzene 211°C 206°C 34.82 1.93 Good 51 2-cyanopyridine (solution darkens) 212°C 206°C 93.77 1.53 (40°C) Good 52 isophorone 213°C 190°C - 2.62 Good 53 1,2,4-trichlorobenzene 214°C 185°C 2.24 2.07 Good 54 1,3-dimethoxybenzene 217°C 180°C 5.48 - Good 55 2-morpholinoethanol 223°C 220°C - - Good 56 (R)-(-)-carvone 228°C 190°C 11.00 2.72 Good 57 quinoline (solution darkens) 237°C 185°C 9.00 3.47 Good 58 1-methylnaphtalene 240°C 238°C 2.92 Good 59 propylene carbonate 242°C 240°C 66.6 2.76 Poor 60 DMPU 246°C 180°C 36.12 1.94 Good 61 Triethylene glycol methyl ether 248°C 180°C - 6.20 Good 62 ε-caprolactone 253°C 205°C 39.43 - Good 63 diphenyl ether 265°C 230°C 3.90 2.60 (40°C) Good 64 ethyl 2-cyclohexanonecarboxylate >220°C 185°C - - Good 65 sulfolane 285°C 240°C 44.00 10.07 Poor 66 diethyltoluamide 290°C 170°C - 13.30 (30°C) Good 67 1-fluoro-2,4-dinitrobenzene 296°C 200°C - - Good 68 benzylbenzoate 323°C 185°C 4.80 8.29 Good 69 1,3-diphenylacetone 330°C 182°C - - Good
Figure 2.22 is a graphical representation of the table with the molecular structure of the 69 solvents divided into the aliphatic and aromatic categories, also ordered by ascending boiling point and with the same colour-code. By looking at the structures, one can perhaps better appreciate the solvent characteristics mentioned in 2.5.3: for aliphatic solvents there is a variety of polar, apolar, symmetric, asymmetric, small and bulky solvent molecules and the same pertains to the aromatic ones, which additionally can have a donor or acceptor character, according to their substituents. With these results, anthraphane was then crystallised from the good solvents hoping to grow single crystals suitable for SC-XRD.
47
ANTHRAPHANE
he dichotomy dichotomy he
Solvents screened for crystallisation ordered by ascending boiling point (from left to right, top to bottom) and divided in t in divided and bottom) to top right, to left (from point boiling ascending by ordered crystallisation for screened Solvents
. 2.22
Figure Figure aliphatic or aromatic. or aliphatic
48
ANTHRAPHANE
2.5.5 Crystallisation procedure
Of the 69 solvents used in the screening, the 30 good solvents were used for crystallisation. The crystallisation apparatus is depicted in Figure 2.23. In order to have controlled cooling rates during the crystallisation process, a PID controller coupled to a heating plate was used. A sand bath connected to a thermocouple was used as heating medium for the crystallisation vials. Screw caps lined with rubber were used to tightly seal the vials, preventing solvent loss and even allowing to work above the boiling point of the solvent with a slight overpressure if needed. The crystallisation apparatus was operated in a vibration-free environment in order to not disturb the crystallisation process.
Figure 2.23. Crystallisation apparatus used in this work. A PID controller coupled to a heating plate ensured controlled heating and cooling rates during crystallisation.
A typical crystallisation procedure was carried out as follow: 2-3 mg of anthraphane were put in a clean glass vial equipped with a magnetic stirring bar; 0.5-1.0 mL of solvent were added and the suspension was briefly purged with argon as a precautionary measure to avoid oxidation during the slow cooling process from the high temperatures employed. The vial was then sealed and put in the sand bath for heating at the desired temperature. Once a clear solution was obtained, the stirring was stopped and the solution was let cool down to room temperature without disturbances. Typical cooling rates employed were 24-36 h. Hot-filtration of the solutions prior to cooling did not result in
49
ANTHRAPHANE any evident benefit for crystal quality and size, and due the high temperatures, oxygen free conditions and small volumes of solvent involved, this challenging procedure was abandoned. Fairly good solvents (orange colour in Table 2.1) such as cyclopentanone, chlorobenzene, o-xylene and cyclohexanone were discarded for crystallisation for the same reason: with these solvents, only suspensions were obtained which would have needed to be hot filtered prior to cooling. If no crystals were present after the cooling process, the vials were let rest for additional days at room temperature and if needed stored at 4°C in the fridge to promote crystallisation.
2.5.6 Anthraphane solvates
To our delight, anthraphane easily formed yellow single crystals of different morphologies in the size range of 100-500 µm, suitable for SC-XRD. As depicted in Figure 2.24, of the 30 solvents used, only 21 crystal solvates could be obtained and analysed by SC-XRD (green colour).
Figure 2.24. Solvents used for crystallisation. From the 30 solvents, only 21 crystal solvates could be obtained (green colour). For the remaining 9 (gray colour), either bad quality crystals or no crystals at all were obtained.
For the remaining 9 solvents (gray colour), for different reasons proper crystals could not be obtained. More in detail: for tetramethylurea, tetraethylurea, 2-morpholinoethanol, ethyl 2- cyclohexanonecarboxylate, 2,4,6-collidine and 1-fluoro-2,4-dinitrobenzene, only poorly shaped and bad quality polycrystalline aggregates were obtained. For hexachloroacetone, due to its reactivity, only single crystals of its hydrated adduct could be obtained (see Figure 2.25), whereas for
50
ANTHRAPHANE diethyltoluamide and triethylene glycol methyl ether, even after storage in the freezer for months, no crystals at all could be obtained probably due to the high viscosity of these two media.
Figure 2.25. Hydrated adduct of hexachloroacetone obtained during the crystallisation process. The molecules form bilayers which are connected together via a network of hydrogen bonds by a layer of water molecules.
2.6 Packing of Anthraphane in the Single Crystal
This subchapter will be dealing with the SC-XRD structural analysis of the 21 solvates obtained and with a detailed discussion of the packing behaviour of anthraphane. As expected, the solvent can indeed influence the packing, but perhaps not surprisingly, different solvents can also give rise to the same packing. The different crystal morphologies observed are solvent-dependent and are not indicative of the internal packing: in other words, different solvents can result in identical packing but different crystal shape. A case of polymorphism was also observed with 2-cyanopyridine. For the analysis, the unit cells of the obtained crystals were probed using SC-XRD, but in some cases not all data sets were fully solved and refined due to time-consuming measurements and sample preparation resulting from tiny samples exhibiting high mosaicity and a high degree of disorder. In some cases with similar solvent and identical cell parameters (within the standard uncertainties), only one representative structure was solved and the others were assumed to be identical in terms of packing. The five main packing motifs obtained are summarised in Figure 2.26, together with the solvents from which the crystals were grown, the characteristic space groups and some representative optical micrographs of the single crystals.
51
ANTHRAPHANE
anthracene anthracene
-
noanthracene
2,e)
t, solvent t, molecules omitted are solvent for
packing
etf/ftf etf/ftf
, d) mixed d) ,
1
packing
etf/ftf etf/ftf
obtained, obtained, top view of a layer in the crystal structures, space group, optical
was
packing 2, c) mixed2, c) packing
etf
packing 1, b) packing
etf
: : solvents from which the packing
bottom
to to
top
packing motifs obtained: a) packing
five
The The
.
Society. Chemical American 2016 Copyright [195]. reference from permission with pted
Ada
.
interaction interaction packing. From micrograph of oneor more representative single layer crystals, arrangement in the crystal structure. In the layer arrangemen clarity 52 Figure2.26
ANTHRAPHANE
The packings were named according to the type of interactions between the anthracene units, namely edge-to-face (etf) or face-to-face (ftf). We could observe two different etf-packings, three different mixed etf/ftf packings and one packing with no anthracene-anthracene interactions. It is also noteworthy that in all cases, anthraphane always packed in a layered fashion. The packings and the interactions involved will be described individually in the following subchapters where a representative solvate will be used for the discussion of the structure on the basis of the geometric parameters displayed in Figure 2.27. These parameters were chosen in accordance with the literature[122] and are differentiated according to CH···π and π···π interactions. For the former, the distance between the H atoms and the centroids of the aromatic rings (dC-H···cn), the distance between the H atoms and the anthracene plane (dC-H···pln), the distance between the H atoms and the nearest C atoms (dC-H···C) and the angles between the ring planes (α) are considered. For π···π interactions, the distance between the rings’ centroids (dcn···cn) and the angle between the ring planes (α) are examined. The values were measured by using the software OLEX2.
Figure 2.27. Geometrical parameters to describe the various types of interactions. a) CH···π interactions: distance between H atoms and centroids of the aromatic rings dC-H···cn; b) distance between H atoms and anthracene plane dC-H···pln (red), distance between H atoms and nearest C atoms dC-H···C (green); c) angle between the ring planes α; d) π···π interactions: distances between rings’ centroids dcn ···cn.
53
ANTHRAPHANE
2.6.1 Etf packing 1
2.6.1.1 ODCB solvate
Figure 2.28. ORTEP diagram of anthraphane in the o-dichlorobenzene solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Note: one ODCB solvent molecule was too disordered to be modeled and was removed from the density using masking techniques inbuilt in the software OLEX2.
Crystals grow as clear light yellow rhombohedral plates from o-dichlorobenzene and belong to the triclinic crystal system (space group P-1). The packing exhibits exclusively etf interactions (Figure
2.29) with CH···π distances dC-H···cn ranging from 2.559(2) – 2.853(2) Å, dC-H···pln between 2.538(2) –
2.797(2) Å, dC-H···C between 2.763(5) – 2.948(5) Å and angles α ranging from 54.9(2) – 68.0(2)°. These distances are shorter than the usual van der Waals distances (< 2.97 Å[123]) which is a sign for strong CH···π interactions[124]. The voids between the monomers are filled with ODCB molecules (some of which are involved in mutual π···π interactions). No channels exist in the structure as the layers are arranged in a staggered fashion with no solvent molecules between layers (Figure 2.29). There are two different interlayer distances, i.e. the structure can be described as a sequence of tight bilayers with an internal layer distance of ~3.0 Å, which is separated from the next bilayer by ~4.7 Å (Figure 2.30). It is also interesting to note that the triple bonds are slightly distorted and not coplanar with
54
ANTHRAPHANE the central benzene cores, which in turn are not orthogonal to the anthracene moieties, adopting a slightly tilted conformation with angles varying from 1.4(2)° to 6.4(2)°.
Figure 2.29. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right) (Solvent omitted for clarity).
~4.7 Å
~3.0 Å bilayer
Figure 2.30. Detailed view of layered structure: there are two different interlayer distances which means that the structure can be described as a sequence of tight bilayers with an internal layer distance of ~3.0 Å, which are separated from the next bilayer by ~4.7 Å.
55
ANTHRAPHANE
2.6.1.2 1,3-Diphenylacetone solvate
Figure 2.31. ORTEP diagram of anthraphane in the 1,3-diphenylacetone solvate (50% probability) and and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right).
Crystals grow as clear light yellow rhombohedra from 1,3-diphenylacetone. The voids between the monomers are filled with two solvent molecules seemingly not interacting with each other by any pi- pi interaction. In this case however, the voids form small channels which run across the structure almost perpendicularly to the layers. This is possible due to a different stacking of the layers compared to the ODCB or TCE solvate (Figure 2.32).
56
ANTHRAPHANE
Figure 2.32. Detailed view from top of a single layer (left) and view in the space-filling model along the small channels (solvent omitted for clarity) (right). Every void in the layer is filled with two solvent molecules.
By looking at the structure in more detail, one notices that not all the anthracene units are properly resting in one plane as explained in Figure 2.19 in subchapter 2.5.2. In this solvate one anthracene unit does not exploit all three potential CH-π hydrogen bonds but only two of them (Figure 2.33). Moreover, the unexploited hydrogen atom does not seem to interact with anything in particular, neither with a neighbouring anthraphane nor with a solvent molecule. As a consequence, the structure is not properly layered and in the typical triangular motif of CH-π interactions, only two anthracenes are in the same planes (red colour in Figure 2.33, enclosed by dashed lines), while one is shifted by approximately one benzene ring (orange colour).
Figure 2.33. Detailed view from top of two layers (left) and view of the two triangular motif in them (center). In the triplex of CH-π interaction, two anthracene units (red colour) are resting in the same plane while one anthracene unit (orange colour) is shifted from the plane by one ring.
57
ANTHRAPHANE
This shift is compensated by a partial π-π overlap between anthracenes of the adjacent layers, marked in violet colour in Figure 2.34. In this structure, the triple bonds moieties are also particularly distorted, so that the cyclophane tends to be Y-shaped.
Figure 2.34. Distorted layered structure of the 1,3-diphenylacetone solvate. Interlayer interactions are due to a partial π-π overlap of the anthracene units (violet colour).
58
ANTHRAPHANE
2.6.1.3 DMPU solvate
Figure 2.35. ORTEP diagram of anthraphane in the DMPU solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Overall there is severe solvent disorder. 12 DMPU molecules in asymmetric unit, all but one are disordered. Solvents are modeled but some are massively restrained. Within the channel there are two groups/chains of solvents: one with 60:40 and one with 70:30 relative occupation.
Crystals grow as clear light yellow prisms from DMPU. The structure exhibits the usual etf interaction motif and is nicely layered, with the solvent molecules filling again the voids between the anthraphane (Figure 2.36). In this particular case, the stacking of the layers produces large channels with a diameter of approximately 11 Å (Figure 2.37). The channels are filled with solvent molecules and run almost perpendicularly to the layers across the structure.
59
ANTHRAPHANE
Figure 2.36. Detailed view from top of a single layer (right) and view along the channels filled with solvent (left).
Figure 2.37. Detailed view of layered structure (left): the interlayer distance is approximately 4.7 Å. Detailed view in the space-filling model (right) along the channels (solvent omitted for clarity). The channels diameter is approximately 11 Å (accounting van der Waals radii).
60
ANTHRAPHANE
2.6.1.4 Benzyl benzoate solvate
Figure 2.38. ORTEP diagram of anthraphane in the benzyl benzoate solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Some of the disorder (2 solvent sites) could be identified. There is more delocalized solvent in-between (1-2 molecules), which has been masked.
With benzyl benzoate, the yellow crystals grow with a variety of different morphologies: hexagonal platelets, prisms and fine needles. All the different morphologies were probed by SC-XRD but interestingly, they all corresponded to the same structure: the usual etf packing (Figure 2.39). Every void between the anthraphanes is filled by four solvent molecules, which interact by π···π stacking with the anthracene units. Similarly to the DMPU case, the stacking of the layers, produces channels of approximately 11 Å in diameter (Figure 2.40).
61
ANTHRAPHANE
Figure 2.39. Detailed view from top of a single layer (left) and view along the channels filled with solvent (right).
Figure 2.40. Detailed view of layered structure (left): the interlayer distance is approximately 4.4 Å. Detailed view in the space-filling model (right) along the channels (solvent omitted for clarity). Similarly to the DMPU case, the channels diameter is approximately 11 Å.
2.6.1.5 TCE, bromoform and 2-cyanopyridine (needles) solvates
The structure of the TCE solvate was already discussed in subchapter 2.5.1, as a preliminary investigation into the packing behaviour of anthraphane. The packing exhibited all etf interactions, with the layers tightly packed and no solvent-filled channel were found in the structure. The same result was observed with the bromoform solvate, whose crystals (clear yellow needles) were analysed by SC-XRD but the structure was only partially solved, due to time consuming measurement. For 2-cyanopyridine, two polymorphs were found, cylindrical platelets and needles. The cylindrical platelets will be discussed in subchapter 2.6.3, when the mixed etf/ftf packings will be analysed, whereas the needles exhibited the typical etf packing discussed so far. Once again due to poor crystal quality and time-consuming measurement times, the structure was only partially solved and will not be presented. The optical micrographs of the single crystals obtained from TCE, bromoform and 2-cyanopyridine are displayed in Figures 2.41, 2.42 and 2.43.
62
ANTHRAPHANE
Figure 2.41. Optical micrographs of the single crystals grown from TCE in bright field mode (left) and between crossed polarisers (right).
Figure 2.42. Optical micrographs of the single crystals grown from bromoform in bright field mode (left) and between crossed polarisers (right).
Figure 2.43. Optical micrographs of the single crystals grown from 2-cyanopyridine in bright field mode (left) and between crossed polarisers (right). There are two polymorphs present, needles and cylinders. The needles exhibit the etf packing discussed in this chapter.
63
ANTHRAPHANE
2.6.2 Etf packing 2
2.6.2.1 Diphenyl ether solvate
Figure 2.44. ORTEP diagram of anthraphane in diphenyl ether solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). One solvent molecule is severely disordered (completely modelled).
Crystals grow as clear light yellow convex needles from diphenyl ether and belong to the monoclinic system (space group P-1). The cyclophane packs in layers again, exclusively exhibiting etf interactions (Figure 2.45). However, in this case there are two kinds of interactions motifs. In the first, two out of three anthracenes form a quadruplex of CH···π interactions. This quadruplex is marked in red colour in Figure 2.45, with distances dC-H···cn ranging from 2.576(1) – 3.398(1) Å, dC-H···pln between 2.545(1) –
3.051(1) Å, dC-H···C between 2.762(3) – 3.281(3) Å and angles α ranging from 73.6(1) – 75.3(1)°. The other interaction involves the remaining anthracene, marked in blue colour, and seems to be a CH···π interaction with the p-orbitals of the external acetylenic carbon atom rather than the neighbouring anthracene carbon. This is supported by long distances dC-H···cn of 3.762(2) – 3.968(2) Å and short distances dC-H···C of 2.733(5) – 2.910(5) Å, which in this case correspond to the distance from the
64
ANTHRAPHANE external acetylenic carbon. The blue anthracene pair also interacts through π···π stacking with both sides of the aromatic unit of diphenyl ether.
Figure 2.45. The two types of interactions in the etf packing 2: the CH···π quadruplexes are displayed in red colour, whereas the CH···π interaction with one anthracene moiety and the acetylenic carbons is displayed in blue.
The voids between the cyclophanes are filled with either one or two solvent molecules (depicted in Figure 2.46 in green and orange colour respectively), the latter being involved in a mutual π···π interaction with one phenyl unit, whereas the second phenyl unit interpenetrates the cyclophane layers. The isolated solvent molecule (green colour) also stacks via parallel-displaced π···π stacking between the blue anthracene units, keeping them apart.
Figure 2.46. Voids between the cyclophanes are filled either by one solvent molecule (green colour) or by two solvent molecules (orange colour): in the latter case the aromatic units of diphenyl ether interact by parallel-displaced π···π stacking.
65
ANTHRAPHANE
Due to the highly distorted triple bonds the layers appear wavy and therefore the interlayer distance cannot be determined reliably, but it approximates to 3.4 Å. Worthy of note is also the distortion of the anthracene moieties with respect to the central benzene cores, two of which are tilted by 6.9(1)° and 6.4(1)° in one direction whereas one is tilted by 7.6(1)° in the other direction with respect to the ideal orthogonal geometry (see Figure 2.47).
Figure 2.47. Due to the highly distorted triple bonds the layers appear wavy and therefore the interlayer distance cannot be determined with precision (left). Worthy of note is also the distortion of the anthracene moieties with respect to the central benzene cores, two of which are tilted by 6.9(1)° and 6.4(1)° in one direction whereas one is tilted by 7.6(1)° in the other direction with respect to the ideal orthogonal geometry (right).
Similarly to the previous cases of the DMPU, 1,3-diphenylacetone and benzyl benzoate solvates, channels are also present in this structure, which are however much smaller in diameter, being approximately 6 Å (see Figure 2.48).
Figure 2.48. Detailed view along the channels without solvent molecules in the ball and stick model (left) and the space fill model (right). The pores diameter approximates to 6 Å.
66
ANTHRAPHANE
2.6.3 Mixed etf/ftf packing 1
2.6.3.1 Nitrobenzene solvate
Figure 2.49. ORTEP diagram of the nitrobenzene solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Both nitrobenzene molecules in the asymmetric unit are disordered about an inversion point.
Crystals grow as clear yellow cylindrical plates from nitrobenzene and belong to the triclinic crystal system (space group P-1). Similar to the etf packing 2, there is a quadruplex of CH···π interactions, in which two out of three anthracene moieties are involved. These are marked in red and pink colour in Figure 2.50. However, in this case the compound packs more densely, so that one pair of parallel anthracenes (red) interacts with the acetylene p-orbitals rather than those of the pink anthracenes, as supported by the distances dC-H···cn of 4.031(2) – 4.118(2) Å, dC-H···pln of 2.416(5) – 2.496(5) Å, dC-H···C of 2.701(3) – 2.751(3) Å (corresponding to the external acetylene carbon) and the angle α = 69.7(1)°.
In turn, the pink set of anthracenes interact with the red set through CH···π with dC-H···cn = 2.629(2) –
2.689(2) Å, dC-H···pln = 2.623(2) – 2.672(2) Å, dC-H···C = 2.802(3) – 2.906(3) Å and α = 69.7(1)°.
67
ANTHRAPHANE
Figure 2.50. The types of interactions in the mixed etf/ftf packing 1. The anthracenes involved in the CH···π quadruplexes are displayed in red and pink colour whereas the anthracenes involved in the ftf π···π stacking are displayed in blue. Solvent molecules π···π stack with the pink and blue sets of anthracenes, efficiently filling the voids.
Another interesting consequence of this dense packing is that the pink anthracenes in the quadruplex are interacting with each other through a parallel displaced π···π interaction with dcn···cn =
4.541(4) – 4.591(3) Å, dC-H···cn = 3.753(2) – 3.907(2) Å, dC-H···pln = 3.401(3) – 3.609(3) Å, dC-H···C = 3.478(4)
– 3.635(4) Å and α = 0° (Figure 2.51 a). The large dcn···cn is similar to the distance between the 9 and 10 positions of the anthracenes (4.538(6) Å), rendering a photodimerisation between this pair unlikely (for more details on topochemical reactions, see subchapter 2.9). The remaining blue-coloured anthracenes interact with the red set of the quadruplex according to dC-H···cn = 3.283(2) – 3.301(2) Å, dC-H···pln = 2.852(3) – 2.895(3) Å, dC-H···C = 2.962(4) – 2.981(3) Å (corresponding to the external acetylene carbon) and a small angle α = 42.9(1)°, the latter corresponding to a situation between etf CH···π and ftf π···π. More interestingly, the blue set of anthracenes are paired with each other by a slightly displaced ftf π···π interaction, with dcn···cn = 3.728(3) – 3.740(2) Å (Figure 2.51 b) and α = 0°. The hydrogens seemingly interact with the acetylene moieties with a dC-H···C = 3.513(4) – 3.630(4) Å.
68
ANTHRAPHANE
Figure 2.51. Detailed view of the ftf π···π interactions involved in the mixed etf/ftf packing 1 for the pink (top) and blue (bottom) anthracene pairs: side view of the anthracene pairs with distance from the 9 and 10 positions (left); distances between hydrogens and the closest carbon atom dC-H···C (center); top view showing the anthracene displacement (right). Distances are in Å.
Solvent molecules efficiently fill the voids between the monomers by stacking ftf with the pink and blue sets of anthracene moieties. The layers are staggered with an approximate interlayer distance of 4.9 Å and no channels are present in the structure (Figure 2.52).
Figure 2.52. View from top of multiple layers in the space-fill model showing the staggered arrangement. The flat acceptor-substituted nitrobenzene efficiently fills the voids between the cyclophanes. The layers are arranged so that there are no channels in the structure.
69
ANTHRAPHANE
The structural skeleton of anthraphane is deformed resulting in distorted triple bonds and anthracenes, one of which is tilted by 7.4(1)° in one direction. The remaining two are tilted by 1.9(1)° and 11.4(1)°, respectively, in the other direction from the ideal orthogonal geometry with respect to the central benzene core (Figure 2.53).
Figure 2.53. Due to the distorted triple bonds the layers appear wavy and therefore the interlayer distance cannot be determined with precision (left). The distortion of the anthracene moieties with respect to the central benzene cores is also shown (right): one anthracene is tilted by 7.4(1)° in one direction and the remaining two are tilted by 1.9(1)° and 11.4(1)° in the other direction with respect to the ideal orthogonal geometry with the central benzene core.
70
ANTHRAPHANE
2.6.3.2 Quinoline solvate
Figure 2.54. ORTEP diagram of the quinoline solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Two half solvents in the asymmetric unit, both disordered (one only disordered about the inversion (50:50), the other about the inversion and additionally split (4x 25% occupancy). Restraints applied for the solvents, basically only their orientation was refined.
Crystals grown from quinolone appear as clear yellow prisms. The structure of this solvate is very similar to the previously discussed nitrobenzene solvate. Quinoline molecules, despite being a bit more sterically demanding still fit in the pockets between anthraphanes and efficiently fill all the voids (Figure 2.55). The layers’ arrangement once again does not allow formation of channels in the structure.
71
ANTHRAPHANE
Figure 2.55. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right). The flat quinoline efficiently fills the voids between the cyclophanes. The layers are arranged so that there are no channels in the structure.
72
ANTHRAPHANE
2.6.3.3 o-Cresol solvate
Figure 2.56. ORTEP diagram of the o-cresol solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Two half solvents in the asymmetric unit, both disordered about inversion (as in the nitrobenzene case).
Crystals grown from o-cresol appear as clear yellow blocks. The packing (Figure 2.57) does not exhibit any noteworthy difference compared to the previous ones.
Figure 2.57. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right). The flat o-cresol efficiently fills the voids between the cyclophanes. The layers are arranged so that there are no channels in the structure.
73
ANTHRAPHANE
2.6.3.4 L-Carvone solvate
Figure 2.58. ORTEP diagram of the L-carvone solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). The solvent due to severe disorder could not be modeled and was therefore removed by masking techniques.
Crystals grown from L-carvone appear as clear yellow prisms. The packing is virtually the same as compared to the previous ones (Figure 2.59). The solvent could not be modeled due to severe disorder but it most likely fills the voids between the cyclophanes as seen previously.
Figure 2.59. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right). The layers are arranged so that there are no channels in the structure.
74
ANTHRAPHANE
2.6.3.5 1,3-Dimethoxybenzene solvate
Figure 2.60. ORTEP diagram of the 1,3-dimethoxybenzene solvate (50% probability) and optical micrographs (scale bar is 500 µm) of the single crystals in bright field mode (left) and between crossed polarisers (right). Two molecules and two solvents per asymmetric unit. Both solvents disordered, but only one had to be modelled (60:40 occupation). One cyclophane is disordered in one anthracene.
Crystals from 1,3-dimethoxybenzene grow as big clear yellow prisms with sizes up to 500 µm. Once again the solvent is stacking between the cyclophanes (Figure 2.61).
Figure 2.61. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right). The layers are arranged so that there are no channels in the structure. 75
ANTHRAPHANE
2.6.3.6 1,2-Dimethoxybenzene solvate
Figure 2.62. ORTEP diagram of the 1,2-dimethoxybenzene solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right).
Crystals grown from 1,2-dimethoxybenzene appear as clear yellow rectangular prisms. In this particular case however, the crystals belong to the monoclinic crystal system, with a C2/c space group. Despite this difference, overall the packing motif is basically the same as the ones observed previously (Figure 2.63) and was considered in the same category. Interestingly, in this structure, the typical solvent disorder (usually about an inversion point) was not observed. The solvent molecules are perfectly ordered and as usual fill the voids by stacking with the anthracene units.
76
ANTHRAPHANE
Figure 2.63. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right). The layers are arranged so that there are no channels in the structure. Despite the different space group, the packing motif is virtually the same as the ones seen previously.
Figure 2.64. Detailed view of the layered structure of the 1,2-dimethoxybenzene solvate.
77
ANTHRAPHANE
2.6.3.7 1-Methylnaphtalene solvate
Figure 2.65. ORTEP diagram of the 1-methylnaphtalene solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Two half solvents in the asymmetric unit, both disordered. One is only disordered about the inversion center (50:50), the other about the inversion and additionally split (4x 25% occupancy). Restraints applied for the solvents, basically only their orientation was refined.
Single crystals of anthraphane from 1-methylnaphtalene grow as clear yellow prisms and belong to the triclinical crystal system, with a P-1 space group. This is the only hydrocarbon solvate, composed only by carbon and hydrogen atoms. The packing motif is the usual with the solvent stacking between the cyclophanes (Figure 2.66).
78
ANTHRAPHANE
Figure 2.66. Detailed view from top of a single layer (left) and view from top of multiple layers in the space-fill model showing the staggered arrangement (right). The layers are arranged so that there are no channels in the structure.
2.6.3.8 1,2,4-trichlorobenzene, 2-cyanopyridine (cylinders), ε-caprolactone, GBL and benzonitrile solvates
As mentioned before, for the 1,2,4-trichlorobenzene, 2-cyanopyridine, ε-caprolactone, GBL and benzonitrile solvates, a complete data set was not collected due to long measurement times, instead only the unit cell was probed. The unit cell parameters confirmed the mixed etf/ftf packing discussed in this subchapter. It is expected that the solvent molecules are stacking in between the cyclophanes as observed before. With 1,2,4-trichlorobenzene, single crystals grow as clear yellow blocks, with 2- cyanopyridine as cylindrical platelets (the other polymorph, in needle form, has been discussed in 2.6.1.5), with ε-caprolactone as prisms, with GBL as oval-shaped needles and with benzonitrile as prismatic needles. The optical micrographs of the solvates can be found below.
79
ANTHRAPHANE
Figure 2.67. Optical micrographs of the single crystals grown from 1,2,4-trichlorobenzene in bright field mode (left) and between crossed polarisers (right).
Figure 2.68. Optical micrographs of the single crystals grown from 2-cyanopyridine in bright field mode (left) and between crossed polarisers (right). There are two polymorphs: the needles were discussed in chapter 2.6.1 and correspond to the etf packing 1, whereas the cylindrical platelets exhibit the mixed etf/ftf packing discussed in this chapter.
Figure 2.69. Optical micrographs of the single crystals grown from ε-caprolactone in bright field mode (left) and between crossed polarisers (right). 80
ANTHRAPHANE
Figure 2.70. Optical micrographs of the single crystals grown from GBL in bright field mode (left) and between crossed polarisers (right).
Figure 2.71. optical micrographs of the single crystals grown from benzonitrile in bright field mode (left) and between crossed polarisers (right).
81
ANTHRAPHANE
2.6.4 Mixed etf/ftf packing 2
2.6.4.1 Isophorone solvate
Figure 2.72. ORTEP diagram of the isophorone solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). Two and a half solvents in the asymmetric unit, one of which is disordered about the inversion.
Single crystals from isophorone grow as clear yellow prismatic needles and belong to the triclinic crystal system, with the P-1 space group. The molecules pack in a new type of motif, with both etf and ftf interactions between the anthracene units (see Figure 2.73).
82
ANTHRAPHANE
Figure 2.73. The types of interactions in the mixed etf/ftf packing 2. The anthracenes involved in the ftf π···π stacking are displayed in blue colour, while the other anthracenes are in red colour.
Similarly to the etf/ftf packing 1, there is a pair of anthracenes (blue colour) engaged in a displaced ftf π···π interaction with, with dcn···cn = 3.800(1) – 3.828(1) Å (Figure 2.74) and α = 0.3(1)°.
The hydrogen atoms interact with the acetylene moieties with a dC-H···C = 3.639(2) – 3.826(2) Å. This pair is slightly displaced but could in principle photodimerise topochemically as the distances between the 9 and 10 position of the anthracenes are in a suitable range being 3.827(2) Å.
Figure 2.74. Detailed view of the ftf π···π interactions involved in the mixed etf/ftf packing 2 for the blue anthracene pairs: side view of the anthracene pairs with distance from the 9 and 10 positions (left); distances between hydrogens and the closest carbon atom dC-H···C (center); top view showing the anthracene displacement (right).
83
ANTHRAPHANE
The most interesting feature of this packing is the role of the solvent, which packs and interacts with the cyclophane in two different ways. In the first one, as seen in the etf/ftf packing 1, one solvent molecule (disordered) is sandwiched between the blue anthracene pairs efficiently filling the voids between the cyclophanes. In the second case, the solvent molecules are organized in a quadruplex, which is encircled by the red anthracene units. Through the methylene and methyl moieties, the solvent molecules are interacting with the anthracenes by CH···π hydrogen bond, forming a network of interactions as depicted in Figure 2.75. Typical distances ranges are dC-H···cn =
2.637(1) – 3.634(1) Å, dC-H···pln = 2.631(2) – 3.020(2) Å and dC-H···C = 2.774(2) – 3.079(2) Å.
Figure 2.75. Detailed view of the isophorone quadruplex in the crystal structure. The solvent molecules are interacting via CH···π hydrogen bonds with the red anthracene units. The green dotted lines represent the distances to nearest carbon atom dC-H···C.
The structure is tightly arranged in layers with an approximate interlayer distance of 3.6 Å. The layers are staggered and there are no channels presents in the structure (see Figure 2.76).
Figure 2.76. Detailed view of layered structure (left): the interlayer distance is approximately 3.6 Å. Detailed view in the space-filling model (right) showing the dense packing and the absence of channels in the crystal structure. 84
ANTHRAPHANE
2.6.5 Packing with no anthracene-anthracene interactions
2.6.5.1 NMP solvate
Figure 2.77. ORTEP diagram of the NMP solvate (50% probability) and optical micrographs of the single crystals in bright field mode (left) and between crossed polarisers (right). One additional solvent molecule rests on the 3-fold axis and is too disordered to be clearly resolved. The corresponding density was removed from the data using masking techniques inbuilt in OLEX2.
Crystals grow as clear light yellow rhombohedra from NMP and belong to the trigonal crystal system, with a R-3c space group. In this very particular and fascinating case, the cyclophanes do not interact with each other within the layers neither through etf CH···π nor ftf π···π interactions (Figure 2.78). In fact, every face of the anthracenes is arranged in sandwich fashion between two NMP molecules, whose methyl groups are oriented alternatively upward and downward relative to the layer plane. Additionally, every edge of an anthracene is blocked by further two NMP molecules, so that each compound is tightly surrounded by 12 NMP molecules in total. This effectively prevents all interactions between anthracenes.
85
ANTHRAPHANE
Figure 2.78. Packing detail of the NMP solvate. In this particular case the cyclophanes do not interact directly with each other as in all the previous cases. Instead, the anthracene moieties are completely surrounded by solvent molecules.
Examining the details of the packing (Figure 2.79), CH···π interactions are found between the anthracenes and one hydrogen atom of the methyl group of NMP (dC-H···cn = 3.001(2) Å, dC-H···pln =
2.949(2) Å, dC-H···C = 3.073(2) Å and α = 43.8(6)°), one hydrogen of the β-methylene (dC-H···cn = 3.301(2)
Å, dC-H···pln = 2.872(3) Å, dC-H···C = 2.949(4) Å and α = 54.2(7)°) and, finally, one hydrogen of the λ- methylene (dC-H···cn = 2.629(1) Å, dC-H···pln = 2.601(1) Å, dC-H···C = 2.816(4) Å and α = 64.4(5)°). Such aliphatic CH···π interactions are the same observed in the etf/ftf packing 2, with isophorone as solvent.
86
ANTHRAPHANE
Figure 2.79. Detailed view of the CH···π interactions between the anthracene faces and the NMP molecules with displayed distances (without standard accuracy) between the hydrogens and the nearest carbon atom dC-H···C (left). De facto, every cyclophane is surrounded by a shell of twelve NMP molecules (in blue colour), completely isolating it from the neighbouring cyclophanes (right).
The layers are equidistant (~3.7 Å) and are arranged in a staggered fashion without any solvent molecule located between them, hence no channels are present in the structure (Figure 2.80). Finally, the triple bonds are only slightly distorted, with the anthracenes being almost orthogonal to the central benzene cores, showing a tilting angle of 3.1(1)°.
Figure 2.80. View from top of multiple layers in the space-fill model showing the staggered arrangement (right). Detailed view of layered structure: the layers are equidistant and spaced by approximately 3.7 Å (right).
87
ANTHRAPHANE
2.7 Considerations on the Packings
What can now be concluded from this collection of crystal structures? By considering the common features of the 21 crystal structures obtained, one sees that irrespective of the very different solvents employed, anthraphane always crystallises in layers, as hypothesised in subchapter 2.5.2. Between the layers, no embedded solvent molecules are found, and within the layers, the anthracene subunits are arranged virtually perpendicular to the layer’s plane; overall, the level of molecular distortions within the compounds is low, if at all existent. As a general view on intermolecular interactions, for compounds bearing anthracene units, the following cases of specific interactions could have been expected: etf, ftf, and mixed etf/ftf. Interestingly, while etf and mixed etf /ftf were actually found in several instances, the required all-ftf arrangements for 2D polymerization were absent. In addition to these specific interactions, situations in which the anthracenes do not show any particular interaction with each other can also occur. For this relatively rare case one example was discovered when using NMP as solvent. Regarding the systematic approach that we used in order to understand the solvent influence on the packings, it is difficult to identify clear trends in the obtained results. In this section we will anyway try to relate the different packings to each other and try to define and understand some general trends regarding the solvent- packing relationship.
2.7.1 The “geometrical” and the “chemical” crystal, etf/ftf packing 1 vs etf packing 1
The first aspect that becomes evident is that the two most common packings encountered are the etf packing 1 (with 7 solvates) and the etf/ftf packing 1 (with 12 solvates). Curiously, these two packings represent two old schools of thought which were employed to look at a crystal structure: the “geometrical approach” and the “chemical approach”‡‡. In the geometrical approach, as the name suggests, the molecules pack together in the crystal according to their geometry, recognizing themselves by size and shape and fitting together like puzzle pieces, resulting in the densest packing possible. The geometrical crystal is thus governed by close packing and the directionality of intermolecular interactions is not considered. This model was first proposed by A. I. Kitaigorodskii in 1960[125] and despite sounding outdated and incomplete, it can be still used today to explain some crystal structures[126,127]. In the chemical crystal instead, the molecules recognise each other chemically through directional interactions such as hydrogen bonds. This model was proposed by M.
‡‡ The terms „geometrical“ and „chemical“ were adapted from the literature[138]. Perhaps “chemical” could be substituted by “supramolecular”, as the interactions involved between molecules in a crystal are of weak nature. 88
ANTHRAPHANE
C. Etter in 1990 in an attempt to explain patterns in crystal structures in terms of hierarchy of conventional hydrogen bonds between heteroatoms[128]. Not surprisingly, for this more recent model, exceptions were found[129–131]. These two models do not necessarily exclude each other but are incomplete, as they neglect the role of weak intermolecular forces; in fact, nowadays it is understood that molecules pack by the means of dispersion and electrostatic interactions, following both geometrical and chemical factors; these are the same factors that we used for our systematic crystallisation approach in terms of sterical and electronical properties of the solvent. However by looking at the etf/ftf packing 1, one can see that the cyclophanes pack in the densest way possible (for a discussion on packing densities of objects with the same symmetry as anthraphane, please refer to the work of Murray[41] and its supplementary information), fitting together by minimising any unnecessary void space, which would correspond to the geometrical model. In the etf packing 1, instead, the cyclophanes prefer to assemble predominantly through CH···π interactions between the anthracene units, forming the triangular motif discussed in 2.5.2, and leaving considerable void space between the molecules. This packing would correspond to the chemical model§§. Figure 2.81 depicts the two packing motifs in the space-fill view together with the respective solvents used for crystallisation.
Figure 2.81. Etf/ftf packing 1 vs etf packing 1 together with their solvents used for crystallisation.
§§ It is very important to note that the two models only consider the packing of the main molecule, without considering any possible inclusion of other molecules such as solvent, which helps filling the voids. 89
ANTHRAPHANE
In order to better understand the electrostatic component of anthraphane, its electrostatic potential surface (EPS) was calculated with the semi-empirical PM3 method[132]***. Not surprisingly due to the lack of polarizing functional groups, we found a substantial negative charge on the faces of the anthracenes (Figure 2.82). At the central aromatic ring of the anthracene, the molecular electrostatic potential (MEP) amounts to -16.3 kcal/mol, giving the anthracene units a characteristic donor character[133].
Figure 2.82. EPS calculated by the semi-empirical PM3 method. For anthraphane the molecular electrostatic potential (MEP) at the marked position is -16.3 kcal/mol, classifying the compound as donor.
In terms of electrostatics, it seems reasonable to assume that it is exactly this negative potential and typical quadrupolar moment of the anthracenes that hampers the desired all-ftf packing. By looking at how the solvent is arranged in the etf/ftf packing 1, namely stacking between the anthracene units (see Figure 2.83), one could tentatively assume that this arrangement is particularly favoured when using aromatic solvents showing positive electrostatic surface potentials (acceptor character) such as nitrobenzene and 2-cyanopyridine, which form donor-acceptor (D-A) complexes with the anthracene units. Such donor-acceptor interactions are in fact often responsible for ftf stacking due to the favourable interactions of the quadrupolar moments[109]. However, this donor-acceptor relationship does not seem important here for two reasons: firstly, a very favourable D-A interaction would likely result in a crystal structure in which D and A units perfectly alternate themselves in a D-A-D-A-D fashion, something which is not observed here, where the sequence observed is D-D-A-D-D-A; secondly, the same packing is obtained also with milder acceptors such as quinoline and even donors such as dimethoxybenzene and 1-methylnaphtalene. Moreover, solvents lacking aromaticity such as GBL, ε-caprolactone and L-carvone also produce the etf/ftf packing 1. It
*** Semiempirical methods such as PM3 are well suited to represent the electrostatic surface potential of planar aromatic rings. DFT calculations were also carried out on smaller fragments of anthraphane and they agree in trend with the PM3 calculations[195]. 90
ANTHRAPHANE seems therefore that electronical factors such as polarity and aromaticity of the solvent do not play an important role for this kind of packing. Instead the important factor here is sterics, in terms of both size and shape of the solvent molecule: it seems that solvents which are flat or can easily assume a flat conformation and are small enough to fit into the cavities between the cyclophanes, have a preference for this packing. As a general observation, only one solvent molecule per cavity can be accommodated.
Figure 2.83. etf/ftf Packing 1 with solvent stacking between the anthracene units. For 2- cyanopyridine and nitrobenzene, it might seem that there is donor-acceptor relationship with anthraphane. However the same packing is also obtained with strong donors such as 1- methylnaphtalene.
Regarding the etf packing 1, it is again not easy to find clear patterns. It seems that the argument of flatness and size of the solvent holds, as there is a preference for this packing with non- flat aliphatic solvents, such as tetrachloroethane and bromoform, and non-flat and bulky solvents, such as 1,3-diphenylacetone and benzoyl benzoate. There are a few exceptions highlighted in blue in Figure 2.81, such as DMPU, ODCB which are flat and small and could in principle crystallise in the etf/ftf packing 1. Interestingly, 2-cyanopyridine produced both packings (indicating a very small energy difference between the two) in the form of two polymorphs: needles and cylindrical plates. This packing is the result of well-defined and directional CH···π interactions, one could therefore cautiously argue that if a solvent is not able to interfere with said interactions (such as the halogenated bromoform or tetrachloroethane), this packing would be the preferred one. In the etf packing 1, the voids between the cyclophanes are bigger and can therefore accommodate more than one solvent molecule. As a consequence, the most interesting feature of this packing are the channels that can be present in the structure according to the solvate. Figure 2.84 summarises the
91
ANTHRAPHANE results obtained in this regard. It seems in this case that the bulkier is the solvent, the higher is the chance to obtain channels in the structure. Bulkiness alone is not sufficient though, conformation seems also to play a role: in order to create channels, a solvent molecule able to interpenetrate the layers can be of advantage; in other words a solvent molecule has to be elongated enough to be present in at least two layers at the same time as can be seen in Figure 2.85 for benzoyl benzoate. The same phenomenon was observed with diphenyl ether in the etf packing 2 (subchapter 2.6.2.1, Figure 2.48), where a microporous structure with channels of approximately 6 Å in diameter could be observed.
Figure 2.84. Comparison of the solvates of the etf packing 1 in terms of porosity of their crystal structures. In the space-fill view, the solvent has been omitted for clarity to show the presence of the channels.
There is however an exception encountered with DMPU. Despite being a small molecule, the DMPU solvate contains channels as big as the ones observed with the benzoyl benzoate solvate. This implies that there must be very favourable interactions between the DMPU molecules, especially along the direction of the channels. The diameter of the biggest channels observed is approximately 11 Å, which is slightly bigger than the cavities in the prototype porous material MOF-5[134]. In contrast to MOF-5 however (and other MOFs), here the solvent molecules cannot be removed from the channels, by for instance vacuum[135], as the boiling point of the solvents is too high. Moreover, the intermolecular forces within and between the layers are only of weak nature (CH···π vs coordinate
92
ANTHRAPHANE covalent bond in MOFs), so that removal of the solvent from the crystals by for instance supercritical
[135] CO2 , would most likely result in the collapse of the structure.
Figure 2.85. Solvent molecules (here benzoyl benzoate, in light blue colour) that can interpenetrate the layers can help to create channels. Elongated conformations are of advantage.
2.7.2 Solvent-induced anomalies: etf packing 2, etf/ftf packing 2, no anthracene- anthracene interactions packing
Having ascertained the preference of anthraphane to pack in the etf packing 1 and the etf/ftf packing 1, what can be said for the other packings obtained, namely the etf packing 2, etf/ftf packing 2 and the special case where no anthracene-anthracene interactions occur at all? These packings have to be considered as solvent-induced anomalies, or better said: special variations of the main two packings due to specific solvent-anthraphane interactions. Let’s take for example the etf packing 2 obtained with diphenyl ether as solvent: despite the apparent complexity of the structure, it can be related to the usual etf/ftf packing 1. Figure 2.86 displays this relation, where the same colour code of the anthracene units has been retained in the two packings for better clarity: starting from the etf/ftf packing 1, the nitrobenzene molecules depicted in orange colour are substituted with diphenyl ether, more precisely, in each cavity a molecule of nitrobenzene is substituted with two molecules of diphenyl ether. The quadruplex of CH···π interactions of the pink and red coloured anthracenes is retained in both packings. To complete the transition, a diphenyl ether molecule (green colour) inserts itself and π···π stacks between the blue anthracene pairs at the site marked by the green dots, creating an array of ftf π···π interactions in the structure (for comparison and details on the interactions of the solvent molecules with the same colour code, please refer to Figure 2.46 in subchapter 2.6.2.1). Thus the etf packing 2 can be seen as a derivative of the etf/ftf packing 1.
93
ANTHRAPHANE
Figure 2.86. The transition from the etf/ftf packing 1 to the etf packing 2. The latter can be seen as a solvent-induced anomaly of the former.
Let’s now consider the etf/ftf packing 2 obtained with isophorone. This packing can also be seen as a derivative of the etf/ftf packing 1 and this time the relation is more straightforward as can be seen in Figure 2.87. By substituting the orange coloured nitrobenzene molecules with the quadruplex of isophorone molecules the transition between the packings occurs (for details, please refer to subchapter 2.6.4.1). The etf/ftf packing 2 is de facto an expanded version of the etf/ftf packing 1.
Figure 2.87. The transition from the etf/ftf packing 1 to the etf/ftf packing 2. The latter can be seen as an expanded version of the former.
94
ANTHRAPHANE
Finally, by analysing the NMP solvate, in which the anthracene units are not interacting with each other, one can see that there is a relation with the etf packing 1. In Figure 2.88 the etf packing 1 is displayed with its typical triangular motif of CH···π interactions (the solvent has been omitted for clarity, for more details please refer to subchapters 2.5.2 and 2.6.1.1). By inserting three NMP molecules at the site marked by the orange circle, the triangular network of CH···π interactions is disrupted, pulling apart and isolating the anthracene units from each other. The etf packing 1 is thus converted to the packing with no anthracene-anthracene interactions, which is basically an expanded version of the former.
Figure 2.88. Transition from the etf packing 1 to the packing without anthracene-anthracene interactions. The triangular motif of CH···π interactions is disrupted by the NMP molecules, which form themselves CH···π hydrogen bonds with the anthracene units.
One should perhaps spend a few words about the CH···π interactions that occur between aliphatic moieties of the solvent and the anthracene units: it is seemingly because of these interactions that two packing anomalies such as the NMP and isophorone solvates were discovered. In principle, the strength of a hydrogen bond increases with the acidity of the hydrogen atom involved, so that for a CH···π interaction, the strength should increase according to the hybridisation of the carbon atom so that sp3 < sp2 < sp; in the NMP and isophorone case, it seems that there is a
3 preference of the anthracene π surface to interact with the sp -hybridised CH3 and CH2 moieties of the solvents rather than the aromatic sp2 CH moieties of the neighbouring anthracenes as seen for instance in the etf packing 1 (see Figure 2.89). Nishio[112] justifies such kind of counterintuitive observations by introducing a favourable entropic term to the CH···π hydrogen bond: rotation of the
CH3 moieties and the general flexibility of the aliphatic units, along with the large π surface available
95
ANTHRAPHANE for interaction, offer the chance to interact in many different ways in terms of varieties of conformations and degrees of freedom.
Figure 2.89. CH···π interactions in the ODCB, NMP and isophorone solvates.
These considerations should be however taken with great care: GBL, ε-caprolactone and L- carvone also contain aliphatic moieties, but anomalous packing motifs could not be observed in their solvates. It might well be, that the observed short-contacts are simply the result of the best spatial arrangement of the solvent molecules in the packing rather than specific and directional CH···π hydrogen bonds.
2.7.3 Thoughts on crystal engineering and crystal structure prediction
The previous discussion on the packings uncovered the heart of the matter: it is not possible to predict how anthraphane will pack according to the solvent used for crystallisation. While it has been demonstrated that there might be a preference for the etf packing 1 and etf/ftf packing 1, an accurate packing prediction is not possible just by qualitatively assessing the potential interactions that the solvent might entertain with the cyclophane. Moreover, putting the role of the solvent aside for a moment, one realises that even just by looking at anthraphane itself, an accurate prediction of the packing is not possible. Desiraju introduced the concept of supramolecular synthon[136] as a tool to help predict crystal structures. Just like the term synthon introduced by Corey in 1967[137] in the context of organic chemistry represents a structural unit within a molecule that can be formed by known synthetic operations, the term supramolecular synthon in the context of crystal engineering represents structural units within supramolecules (crystals) which can be formed/assembled by known synthetic operations involving intermolecular interactions. One notices the parallel between
96
ANTHRAPHANE organic chemistry and crystal engineering: in organic chemistry, molecules are formed by covalently connecting atoms together through chemical reactions, while crystals (supramolecules) are built by connecting molecules with intermolecular interactions, through the process of crystallisation (supramolecular reaction). In this sense, polymorphs and different packings can be seen as different reaction products. The supramolecular synthon is a structural unit or pattern, which bests represents the entire crystal structure and encapsulates all the information in terms of molecular recognition, both chemical and geometrical. It can therefore be regarded as a common ground between the geometrical and chemical models discussed previously 2.7.1. Examples of supramolecular synthons, such as particular hydrogen bond patterns or even just the geometrical arrangement of alkanes are displayed in Figure 2.90.
Figure 2.90. Examples of supramolecular synthons.
However, if one tries to apply the concept of the supramolecular synthon to anthraphane, one sees that it is very arbitrary and does not help in predicting the packing. As displayed in Figure 2.91, one could choose for instance the triangular motif of CH···π interactions between the anthracene as synthon; similarly in the etf/ftf packing 1 one could choose the quadruplex of CH···π interactions. How about the other packings then? What would be the best synthon for the etf/ftf packing 2? There is no clear answer, as the supramolecular synthons are probabilistic by nature and can be arbitrarily chosen to justify the obtained results. Moreover, in reviews about crystal engineering, the role of the solvent in the crystal packing is very rarely if at all mentioned[138,139]! Perhaps conveniently, as if one would have to consider the interactions between solvent and solvate in addition to the solvate-solvate interactions, the variables in a qualitative structural prediction would increase dramatically.
97
ANTHRAPHANE
Figure 2.91. Selected supramolecular synthons depicted in blue colour in the anthraphane packings.
So far, the only reliable method for predicting a packing involves computational structural prediction[140]. In the past few years, several advances in the field have been made by Leusen, Kendrick and Neumann, whose algorithms were able to correctly predict four out of four crystal structures in a blind test organised by the Cambridge Crystallographic Data Center[141,142]. The structures included a flexible compound as well as a co-crystal. The prediction proceeds as follows: through a first molecular-mechanics step, the possible energies of a crystal are screened and ranked in order of stability; a tailor-made force field of the bond energies for the most stable configurations is then calculated, reducing the number of the most likely structures to around 100. Finally, the structures are re-ranked by being subjected to another molecular-mechanics calculation based on the lattice energy and stability of the molecule, including full treatment of dispersion effects, which are certainly structure-determining. This successful method requires costly powerful computing clusters and long computing times in the order of months, one can however imagine that in the future, due to the fast advances in computational technology, results will be obtained with a relatively low cost and computing time. It has to be stressed that this computational method ranks the crystal structures energetically and considers therefore the crystal as mere thermodynamic entity. Unfortunately, sometimes the experimentally observed structure is the kinetic crystal, which does not correspond to the energetic minimum. In this regard, a non-energetic criterion that could be used as a complement to this method could be the previously discussed supramolecular synthon, in fact, hydrogen bond patterns and motifs in crystal structures encode both kinetic and energetic information; therefore they could be used in the ranking of predicted crystals structure as a source of some kinetic information[142]. Kinetic crystals are however very difficult to control and to reproduce (polymorphs are good examples) due to the high susceptibility to external ambient conditions.
Computational crystal structure prediction could be in this sense used on anthraphane, to see if the desired all-ftf packing for topochemical 2D polymerisation would be energetically
98
ANTHRAPHANE reasonable. If not, one could try to vary the solvent to stabilise the structure until a feasible energetic landscape is obtained, and only then one would verify the computationally obtained results by crystallising the molecule. This approach, for how logical it might sound, could however take years and involve enormous costs in terms of computing time and power.
2.8 Towards the Right Packing
Is there a way to steer the packing into the desired direction from the empirical evidence gained in this study and without resorting to expensive computational methods? As stated at the beginning of this chapter, during the course of this work, compounds such as 4, were able to crystallise into an exclusive ftf-packing suitable for the creation of 2D polymers through lateral topochemical polymerisation[40]. The process to obtain the desired packing was the same as the one used in this work: a trial and error approach until the right solvate was obtained with 2-cyanopyridine. What is then the reason for compound 4 to crystallise in the desired packing? From a mere electrostatic point of view, if one compares the ESP of anthraphane and compound 4, one sees that both have a strong donor character on the anthracene units (Figure 2.92), a characteristic that should in principle favour etf interactions over ftf stacking. In fact, a preference for 4 to pack in an etf packing was observed for some solvates such as benzonitrile[48]. The packing exhibited the same features of the etf packing 1 of anthraphane with the classical triangular motif of CH···π interactions, as displayed in Figure 2.93.
99
ANTHRAPHANE
Figure 2.92. EPS of anthraphane and compound 4 calculated by the semi-empirical PM3 method. Both have a strong donor character on the anthracene units. For the central orthogonal connecting units there is however a difference: the benzene moieties have a strong negatively charged surface, whereas the triazine units have a typical electropositive surface.
Figure 2.93. Comparison of the etf packing 1 of anthraphane from ODCB and etf packing of compound 4 from benzonitrile (solvent omitted for clarity). The triangular motif of CH···π interactions is the common feature of the two packings. 100
ANTHRAPHANE
However, by crystallising 4 from 2-cyanopyridine, an all-ftf packing was obtained as displayed in Figure 2.94. This was possible due to 4 acting as a template for the packing: together with the solvent 2-cyanopyridine and by the means of parallel-displaced ftf interactions (green and orange dotted lines), a central spectating molecule (orange colour) forces the surrounding cyclophanes into exclusively parallel ftf arrangements (blue dotted lines), suitable for photodimerisation. This directing template effect could not be observed in the solvates of anthraphane. In the case of 4, triazine units which add additional weak interactions perpendicular to the layers are also present. Attempts to influence the interlayer interactions in anthraphane by using fluorinated aromatic solvents such as hexafluorobenzene were also performed, but due to the very poor solubility of anthraphane in these media, single crystals could not be grown (please refer to Table 2.1). Important structural differences between compound 4 and anthraphane are also present: compound 4 links its triazine units to the anthracenes by an oxygen linker whereas anthraphane does so by an acetylenic unit. As a consequence, lacking a triple bond, compound 4 is more rigid and compact, and the π surface available for interactions is restricted to the anthracenes only; moreover due to the kink induced by the oxygen linker in the structure, it is likely that only the external faces of the anthracenes are sterically available for a complete ftf stacking. This reduced flexibility and less π surface available for interactions might help to reduce the number of potential packings available for compound 4, simplifying the problem: thermodynamically, the entropy would be smaller. The oxygen atoms at the 1,8 positions of the anthracenes might also help to induce the antiparallel ftf stacking due to a favourable dipole orientation and compensation.
101
ANTHRAPHANE
Figure 2.94. All-ftf packing of compound 4. This packing is possible due to a templating effect. The orange molecule acts as a template and through parallel-displaced ftf interactions (orange dotted lines) forces the surrounding molecules into an ftf arrangement (blue dotted lines). The solvent 2- cyanopyridine fills the voids and is also involved into parallel-displaced ftf interactions (green dotted lines).
The question that arises now is the following: is it possible to induce such a template effect on anthraphane during its crystallisation? An idealised ftf packing of anthraphane (Figure 2.95) shows the typical honeycomb structure with voids of a diameter of approximately 1.4 nm. Since the cyclophane cannot self-template, it could be co-crystallised with a suitable guest molecule able to fill the voids. We considered C60, C70 and compound 4 as possible candidates for co-crystallisation due to their size and shape, with diameters ranging from 1.0 to 1.3 nm. Fullerenes are known to form a variety of inclusion compounds[143,144] and possess an electron acceptor character[145], which would be of advantage in terms of electrostatic interactions with the electron donor anthraphane. The smaller diameter of the fullerenes with respect to the void size could be compensated by the inclusion of solvent molecules in the structure. On the other hand, compound 4 is structurally related to anthraphane and would seem to geometrically perfectly fit in the voids. Figure 2.96 displays possible
102
ANTHRAPHANE arrangements in the hypothetical structure of the co-crystals, with solvent molecules filling the remaining voids.
Figure 2.95. Idealised all-ftf packing of anthraphane. The porous honeycomb structure has voids with diameters of approximately 1.4 nm, which could be filled during crystallisation by guest molecules such as C60, C70 and compound 4.
Figure 2.96. Hypothetical co-crystal with C60 and compound 4 as guest molecules. Solvent molecules (green colour) could fill the remaining void space.
We therefore started to investigate the co-crystallisation behaviour of anthraphane. It has to be noted that both fullerenes and compound 4 also suffer from low solubility like anhtraphane, therefore a common good solvent had to be found for the experiments. For the co-crystallisation
103
ANTHRAPHANE with C60, ODCB was chosen as best solvent[146] and the same crystallisation procedure described in subchapter 2.5.5 was employed. A mixture of C60 and anthraphane in the stoichiometric ratio 1 : 2 was dissolved in boiling ODCB and let cool to room temperature over 24 h. Among C60 crystals in the form of dark purple needles and anthraphane crystals in the form of yellow rhomboidal plates, promising reddish blocks were also obtained (Figure 2.97). Regrettably, after SC-XRD analysis the crystals resulted to be an ODCB solvate with the usual etf packing 1 previously seen in subchapter 2.6.1.
Figure 2.97. Optical micrographs of the reddish single crystals obtained from the co-crystallisation with C60 from ODCB. SC-XRD reveals the usual etf packing 1, without inclusion of C60.
Due to the unusual colour of the single crystals, it was possible that only a small amount of C60 was included in the structure which was not observed by SC-XRD analysis. The crystals were therefore analysed by confocal Raman spectroscopy. The results are summarised in Figure 2.98: the
[147] most intense peaks of C60 corresponding to the Ag vibrational modes were absent in the crystals, where only the anthracene vibrational modes and the triple bond stretch were present[33,59], consistent with the anthraphane crystal and confirming the results of the SC-XRD analysis. It is however conceivable that trace amounts of C60 which are not detectable by Raman spectroscopy might still be present in the crystal or on its surface and might induce the colour change observed by optical inspection; unfortunately, not enough material was available for more sensitive analytical
104
ANTHRAPHANE techniques such as UV/vis spectroscopy. Similarly, co-crystals with C70 could not be obtained as only the two homocrystals were observed.
Figure 2.98. Confocal Raman spectroscopy of the crystals obtained by co-crystallisation with C60 compared to pure C60 crystals.
Regarding the co-crystallisation of anthraphane with compound 4, three different solvents were employed: ODCB, nitrobenzene and 2-cyanopyridine, using the same crystallisation conditions as before. In all the three cases, both homocrystals (of poor quality) were found. Curiously with 2- cyanopyridine compound 4 crystallised exclusively as needles exhibiting the etf packing; hexagonal platelets with the all-ftf packing could not be found.
These preliminary results on co-crystallisation might not seem very promising, however, by fine-tuning the crystallisation conditions and finding the appropriate solvent, there is still a chance to obtain the desired co-crystal. Nevertheless, this procedure would involve a screening process and considerable work, which could last several months in order to produce positive results. Moreover, as the reader might have probably learned while reading this chapter, a good amount of luck is also needed when crystallisation is involved. As such, for the sake of this thesis, co-crystallisation was not explored further but will for sure be the object of future work. As a final remark, it has to be pointed out that alternative crystallisation methods such as sublimation were tried but unfortunately exposure of anthraphane for several days to high vacuum conditions at 2 x 10-6 mbar and temperatures up to 310°C, did not yield any sublimate; instead at temperatures above 280°C the
105
ANTHRAPHANE compound started decomposing. Crystallisation from exotic and bulky solvents such as ionic liquids[148,149] was also performed. However, in the two most common and apolar 1-butyl-3- methylimidazolium tetrafluoroborate ([bmim][BF4]) and 1-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF6]), anthraphane remained completely insoluble. Further combinations of cation and anion were not investigated due to the high cost of such media.
2.9 Topochemical Single-Crystal-to-Single-Crystal Photodimerisations of Anthraphane
Despite not having achieved the goal of an ftf packing suitable for a topochemical 2D polymerisation, the photoreactivity in the single crystals of anthraphane was nevertheless tested on the packings where at least some anthracene units were stacking ftf, namely the mixed etf/ftf packing 1 and the mixed etf/ftf packing 2. In particular, focus was put onto the irradiation conditions (wavelength, irradiation time, and temperature) and the thermal stability of the anthraphane dimers, in order to have some useful information in the event of obtaining an anthraphane 2D polymer crystal in the future. Care was taken to have topochemical single-crystal-to-single-crystal (SCSC) reactions so that the structure of the products could be unambiguously characterised by SC-XRD. A first step towards SCSC reactions involves the use of tail-end absorbed irradiation, a method pioneered by Enkelmann and co-workers in 1994[150]: irradiating a crystal with the lowest possible energy wavelength ensures a more homogeneous reaction throughout the entirety of the crystal. The solid-state UV/Vis absorption spectrum of anthraphane is displayed in Figure 2.99 and shows a tail-end absorption starting at around 450 nm; the irradiation wavelength was chosen to be 465 nm corresponding to visible blue light.
106
ANTHRAPHANE
Figure 2.99. Solid-state UV/Vis absorption spectrum of anthraphane. The chosen wavelength for tail- end irradiation is 465 nm.
An in-house built cylindrical photoreactor equipped with 16 high power LEDs was used as irradiation source and a cooler from an x-ray diffractometer was used to keep the temperature at the desired value during the photoreaction. A typical experimental setup is depicted in Figure 2.100: single crystals were selected and mounted on the pin of a goniometer head and their unit cell and diffraction quality was checked by collecting enough frames on the diffractometer. The pin was subsequently detached from the goniometer head and put under the cooler in the middle of the photoreactor where irradiation was performed at constant temperature for the desired amount of time. After irradiation, the crystal was again checked by SC-XRD and if diffracting properly, a complete dataset was collected for structure determination.
107
ANTHRAPHANE
Figure 2.100. Setup for the topochemical photoirradiation experiments, with cooler and LED reactor.
2.9.1 Dimers from the etf/ftf packing 1
Among the various solvates, single crystals grown from 1,3-dimethoxybenzene were chosen for the irradiation experiment due to their particular high quality diffraction, resistance to solvent loss and reasonable size. The monomer crystal was irradiated for 30 min at -10°C and as predicted the dimer crystal was obtained with full conversion (according to SC-XRD).
108
ANTHRAPHANE
Figure 2.101. SCSC dimerisation of the 1,3-dimethoxybenzene solvate. The reaction involves a considerable mismatch between the unit cell parameters of the monomer crystal and the dimer crystal. These result in cracks in the crystals upon photoirradiation, which however still retain their single crystal properties.
The results are summarised in Figure 2.101, with the same anthracene colour code discussed in subchapter 2.6.3.1. The striking feature of this topochemical dimerisation is the pronounced molecular movement and the deformation of the unit cell associated with it: the a-axis varies only slightly from 14.577(2) Å to 14.6127(3) Å (0.2% increase) whereas the b-axis lengthens considerably going from 15.067(2) Å to 17.9127(3) Å (5.3% increase) and the c-axis dramatically shortens from 24.253(4) to 21.1999(4) (12.6% decrease). The same happens with the angles of the unit cell which undergo variations ranging approximately from 3° to 9°. These considerable variations in the unit cell are reflected by cracks appearing on the irradiated single crystals, which however mostly retain birefringence, crystallinity and still diffract properly. The unit cell volume goes from 5124 to 5140 Å3 (0.3% expansion). It was found that keeping the temperature during irradiation between -20°C and - 10°C limits crack formation on the crystals, whereas irradiating near and above room temperature results in disintegration of the crystals. The change in unit cell parameters was used to investigate the rate of the reaction: crystals were irradiated for a determined amount of time and the unit cell was then determined. It was found that at -10°C the topochemical reaction is already completed after 2 min of irradiation. As a general remark, apart from tail-end irradiation, in order to favour SCSC reactions and prevent crystal disintegration, two other factors seem to be important: crystal size and 109
ANTHRAPHANE rate of reaction[151]. Smaller crystals are able to relax more efficiently and better dissipate the stress and strain resulting from the mismatch between lattice metrics, densities and intermolecular interactions exerted during the topochemical reaction. Likewise, rapid reactions could lead to rapid accumulation of strain leading to crystal cracking. Irradiation of smaller fragments of the single crystals still resulted in sporadic crack formation; it is therefore assumable that the problem might be the extremely rapid reaction rate. In this regard, using a weaker light source might result in a more well-behaved SCSC transformation. Another interesting feature of this topochemical reaction is that the solvent 1,3-dimethoxybenzene goes from a disordered state in the monomer crystal to a fully ordered one in the dimer crystal. It is not clear what is the role of this reorientation of the solvent molecules, namely whether it is a consequence of the dimerisation step (perhaps a readjustment of the dipole moment of the crystal) or if it is needed for the dimerisation to occur. In a preliminary experiment trying to find the optimal irradiation conditions, single crystals were irradiated at -100°C for several hours and were shown not to undergo any dimerisation; by irradiating the single crystals at -50°C for 4 h, a 50% conversion of the dimers was found (expressed as disorder), the solvent was however found to be still disordered. It is therefore suggested that reorientation might happen at the final stages of the reaction. A natural readjustment of the solvent orientation after crystallisation due to relaxation can be excluded as the monomer crystals were measured several weeks after being grown.
What is puzzling about this reaction, is the amount of molecular movement involved with it, which seems completely unnecessary since the reacting anthracene pairs are stacking nicely and within standard distances for dimerisation (Figure 2.51 in 2.6.3.1). It however nicely shows the temperature dependence of topochemical reactions, in which, kinetically speaking, temperature is not associated to the activation energy as in conventional reactions but to the molecular movement. In this case, low temperatures such as -100°C are not enough to allow molecular movement for the topochemical reaction to occur; at -50°C the reaction proceeds at a slow rate whereas at -20°C and higher temperatures the reaction rate becomes very fast. In future works, other solvates of the etf/ftf packing 1 should be dimerised and compared with the 1,3-dimethoxybenzene for a better understanding of this SCSC transformation; in fact, in the next subchapter, in the etf/ftf packing 2 of the isophorone solvate, it will be shown that the molecular movement associated with the topochemical reaction is minimal. As a final remark, it should be noted that attempts to dimerise in a SCSC reaction the pink coloured anthracene pair (Figure 2.101) in order to form 1D polymer chains were performed but without success. Prolonged irradiations at 80°C of the dimer crystals did not show any further dimerisation reaction and exposure of the crystals at temperatures above 100°C (with and without irradiation) resulted in the degradation of the crystals.
110
ANTHRAPHANE
2.9.2 Dimers from the etf/ftf packing 2
Single crystals grown from isophorone were irradiated following the same procedure described previously. The monomer crystals were irradiated for 30 min at -10°C and 25°C and in both cases the dimer crystals were obtained with full conversion (according to SC-XRD).
Figure 2.102. SCSC dimerisation of the isophorone solvate. In this case, the mismatch between the unit cell parameters of the monomer crystal and the dimer crystal is minimal and the single crystals withstand the reaction perfectly without cracking even at room temperature.
The results are summarised in Figure 2.102. Compared to the previous case, the SCSC reaction proceeds more smoothly without any visible cracks in the crystals, whose quality upon dimerisation even increases (reduced mosaicity, better diffraction). The molecular movement associated with this reaction is minimal which reflects a small mismatch between the unit cell parameters of the monomer crystal and the dimer crystal: the space group remains unchanged and the a-axis increases from 13.2429(9) Å to 13.4200(14) Å (1.3%), the b-axis increases from 14.1163(9) Å to 14.3444(17) Å (1.6%) and the c-axis decreases from 19.8453(14) Å to 19.785(2) (0.3%). The unit cell volume goes from 3210 to 3232 Å3 (0.7% expansion). The disordered solvent molecule next to the dimerising pair remains equally disordered in the dimer crystal. It can be assumed that due to the low degree of molecular motion involved, the temperature dependence on the rate of the reaction might be minimal compared to the previous case; a thorough investigation into this phenomenon could not be performed due to time restrictions and will be the subject of future works.
111
ANTHRAPHANE
2.9.3 Thermal stability of the anthraphane dimers
Due to the thermal reversibility of the anthracene dimerisation reaction, the stability of the anthraphane dimers was investigated. Unfortunately the back-reaction could not be followed topochemically by the means of SC-XRD, as the dimer crystals were found to fall apart at temperatures above 100°C. Therefore, 13C solid-state NMR was chosen as alternative analytical method. The experimental procedure was as follows: the dimer crystals were placed in a rotor, which was then heated in an oven at 180°C for different time intervals. After every heating cycle, a 13CP- MAS NMR spectrum was recorded and the peaks assigned to the bridge carbons were compared with the peaks corresponding to 1,3-dimethoxybenzene, which was used as internal standard. During the experiments, the parameters on the spectrometer were kept constant, in order to better quantify the conversion of the back-reaction. The results of the thermal stability of the anthraphane dimers are displayed in Figure 2.103 and 2.104:
Figure 2.103. Thermally induced back-reaction of the dimer crystals followed by 13CP-MAS NMR spectroscopy. The back-reaction was investigated at 180°C and conversion was estimated by the ratio of the integrals of the 1,3-dimethoxybenzene signals with the anthracene dimer bridge signals.
112
ANTHRAPHANE
Figure 2.104. Back-reaction of the dimer crystals at 180°C. After 72 h, the crystals still have approximately 20% of dimers.
The anthracene dimers were found to be thermally unstable at 180°C and started to progressively back-react over time: after 4 h, 20% of back-reaction was estimated and after 72 h, approximately 80% of conversion to the monomers was found. The anthraphane dimer thermal stability is consistent with the thermal polymer stability of the previously discussed compound 4[40].
The stability of the dimer crystals was also investigated in solution by 1H-NMR in order to understand if the dimers could be recrystallised into another packing without back-reacting. The dimer crystals were first powdered and dried over several days under high vacuum at 80°C in order to remove as much 1,3-dimethoxybenzene as possible; they were then suspended in an NMR tube in TCE-d4 (the best solvent available) and heated over several days at 150°C. Despite the extremely low solubility (lower than anthraphane) the suspension could be analysed at different time intervals with a resolution high enough to distinguish the main peaks. The results are reported in Figure 2.105. The dimers resulted to be stable over 4 days, without any noticeable formation of anthraphane due to the back-reaction: the 2:1 ratio between the bridge protons (yellow colour) and the 9 position protons (red colour) of the free anthracene units remained constant. The intensity of the 1,3- dimethoxybenzene peaks increases over time due to the partial dissolution of the crystals in TCE (promoted by stirring) at 150°C: the 1,3-dimethoxybenzene molecules are freed from the crystal packing and diffuse in the solution. The thermal stability is in good agreement with that of a 1,8- substituted anthracene dimer published in the literature[152]. With these information in hand, the recrystallization of the anthracene dimers was attempted at temperatures between 150°C-160°C, to
113
ANTHRAPHANE avoid significant back-reaction. Unfortunately, no solvent could be found to completely dissolve the dimers crystals at concentrations suitable for crystallisation.
Figure 2.105. Thermal stability followed by 1H-NMR spectroscopy of the anthraphane dimers in a TCE-d4 solution at 150°C (solvent residual peak is at 6 ppm). After 4 days, no noticeable back- reaction was observed. The ratio between the two bridge protons (yellow colour) and the four protons of the anthracene 9 positions (red colour) remained constant.
114
ANTHRAPHANE
As a final remark, it is important to point out that the selective synthesis of the anthraphane dimers can only be performed topochemically, as in solution uncontrolled oligomerisation would be expected with formation of a variety of regioisomers (as seen in 2.1).
2.10 Conclusion and Outlook
As stated at the beginning of the chapter, this work was primarily driven by the search for new monomers to be used for the creation of novel 2D polymers. “Anthraphane” 1, a hydrocarbon cyclophane with D3h-symmetry bearing photoreactive anthracene units was designed and synthesised as a new potential monomer for the topochemical synthesis of 2D polymers in single crystals. Anthraphane was obtained in only five steps and in rather impressive yields, using the key building block anthracene-1,8-ditriflate 9, whose synthesis was revisited and greatly simplified, allowing to obtain gram amounts of the monomer efficiently and in short times. Anthraphane was then systematically crystallised from 21 different solvents, aiming at an exclusive ftf-packing, suitable for topochemical photopolymerisation. Although the objective was not achieved, it was shown that anthraphane always crystallises in layers and that the solvents employed for crystallisation can influence the packing within the layers. Of the 21 solvates, only two main packing motifs (etf packing 1 and etf/ftf packing 1) and three solvent specific packings (etf packing 2, etf/ftf packing 2, no- anthracene interaction packing) were found, all characterized by different type of interaction motifs among the anthracene units of 1. In the etf/ftf packing 1, partial ftf π···π along with etf CH···π interactions were found. This packing, in which anthraphane packs in the densest way possible, was found to be the most observed when using solvents which are small and/or can easily assume a flat conformation, such as aromatics. In the etf packing 1 instead, the dominant interactions between anthraphane molecules were all etf CH···π; this packing seems to be preferred when using bulky, non-flat solvents. The etf packing 2, etf/ftf packing 2 and the packing without anthracene-anthracene interactions were found to be variations of the two main packings due to solvent-anthraphane specific interactions. From a mere electrostatic point of view, it was found that anthraphane seems predisposed to avoid ftf interactions due to its negatively charged electrostatic potential surface on the anthracene units, which prefer etf interactions due to the favourable orientation of their quadrupolar moments.
While the importance of the choice of the solvent for crystallisation was demonstrated, an empirical method to predict or influence the packing of anthraphane could not be developed; as a matter of fact, it was evidenced that crystal engineering can still be very speculative and that the only way to accurately predict the crystal packing of a molecule might rest into computational structure 115
ANTHRAPHANE prediction, which in 2016 still remains a huge challenge. For the structurally related compound 4, which was successfully employed for the topochemical synthesis of 2D polymers, the suitable ftf packing was found by luck, in the same trial and error method employed in this work. It is therefore possible that the suitable packing of anthraphane will be eventually found with an appropriate solvent and that an anthraphane based 2D polymer will be available in the future. In this regard, preliminary tests on its topochemical reactivity were performed by photoreacting the molecule in the etf/ftf packings 1 and 2: in both cases the monomer crystals were fully converted into dimer crystals in a fast SCSC transformation. The anthraphane dimers were then found to be thermally stable at temperatures up to 180°C by solid-state NMR analysis.
Concerning the tedious trial and error crystallisation procedure, co-crystallisation with a suitable guest molecule is proposed as an alternative method. Two scenarios which might enforce an ftf packing are therefore suggested: an ftf packing would involve the formation of large pores with a diameter of approximately 1.4 nm; in the first scenario, a molecule of roughly this size (preferably with an acceptor character, such as fullerenes) could efficiently fill the voids and perhaps even act as a template during co-crystallisation. In the second scenario, an acceptor-substituted anthraphane could be synthesised and co-crystallised in a 1:1 ratio with anthraphane, realizing a co-crystal with an all-ftf packing by the means of acceptor-donor interactions between the anthracene units as depicted in Figure 2.106. This second scenario is perhaps more appealing as it could result in the first ever synthesised 2D alternating copolymer, it however would require considerable synthetic work.
Figure 2.106. Idealised co-crystallisation scenario with an acceptor and a donor anthraphane. The co- crystal would result in an ftf packing due to the donor-acceptor interactions between the anthracene units.
116
ANTHRAPHANE
As a final remark, the robust synthetic strategy developed in this work opens the possibility for modification of the anthraphane structure, creating engineered derivatives which could be used for other synthetic approaches to 2DPs. As an example, desymmetrisation of the structure should give rise to substituted anthraphane derivatives with amphiphilic character as it would be needed in interfacial approaches to 2DPs.
117
ANTHRAPHANE
2.11 Experimental
2.11.1 Synthesis
Materials and methods
All reactions were carried out under nitrogen by using standard Schlenk techniques and dry solvents. DCM, dioxane and toluene were distilled by a solvent drying system from LC Technology Solutions
Inc. SP-105 under nitrogen atmosphere (H2O content < 5 ppm as determined by Karl-Fischer
[153] titration). Pd(PPh3)4 catalyst was freshly prepared following the literature procedure and stored in a glove-box in the dark under N2 at room temperature. Compound 5 was prepared according to literature procedures[48]. All reagents were purchased from Acros, Aldrich or TCI, and used without further purification. Column chromatography for purification of the products was performed by using Merck silica gel Si60 (particle size 40-63 μm).
NMR was recorded on a Bruker AVANCE (1H: 300 MHz, 13C: 75 MHz) at room temperature or 70°C. The signal from the solvents was used as internal standard for chemical shift (1H: δ = 7.26 ppm, 13C: δ = 77.16 ppm for chloroform, 1H: δ = 6.00 ppm, 13C: δ = 73.78 ppm for 1,1,2,2-tetrachloroethane, 1H: δ = 5.33 ppm, 13C: δ = 54 ppm for dichloromethane, 1H: δ = 2.50 ppm, 13C: δ = 39.52 ppm for dimethyl sulfoxide, 19F: δ = -164.9 ppm for hexafluorobenzene). For centrifugation, a Hermle Z 320 K table centrifuge was used at 25 °C. When possible, proton and carbon signal assignments were performed with the help of 2D-NMR experiments such as COSY, HSQC and HMBC (spectra not shown).
High resolution mass spectroscopy (HRMS) analyses were performed by the MS-service of the laboratory for organic chemistry at ETH Zurich with spectrometers (ESI- and MALDI-ICR-FTMS: IonSpec Ultima Instrument). Either 3-hydroxypicolinic acid (3-HPA) or trans-2-[3-(4-tert-butylphenyl)- 2-methyl-2-propenylidene]malononitrile (DCTB) were used as matrix.
UV/Vis absorption spectra were recorded with a JASCO V-670 UV-Vis-NIR spectrophotometer using a quartz cell with a path length of 1 cm. Emission spectra were recorded with a Spex Fluorolog 2 spectrophotometer from Jobin Yvon (United Kingdom) using a quartz cell with a path length of 1 cm by diluting by a factor of 30-60 (depending on the compound) the solutions employed for the UV/Vis absorption measurements.
SC-XRD analysis was performed by the Small Molecule Crystallography Center (SMoCC) at ETH Zurich.
118
ANTHRAPHANE
Synthetic procedures
1,8-dihydroxy-9,10-dihydroanthracene 7 Compound 7 was prepared following the already reported procedure[48]. Mp: 207-209°C.
1 H-NMR (300 MHz, DMSO-d6) δ/ppm: 9.39 (s, 2H), 6.97 (t, J = 7.7 Hz, 2H), 6.71 (d, J = 8.0 Hz, 2H); 6.68 (d, J = 8.2 Hz, 4H), 3.83 (s, 2H), 3.71 (s, 2H).
13 C-NMR (75.5 MHz DMSO-d6) δ/ppm: 154.2, 137.3, 126.2, 122.3, 118.1, 112.1, 34.9, 21.8.
+ HRMS (FT-MALDI): m/z calcd for C14H13O2 [M-H] : 213.0910; found: 213.0909.
119
ANTHRAPHANE
1,8-ditriflate-9,10-dihydroanthracene 8 1,8-dihydroxy-9,10-dihydroanthracene 7 (6.00 g, 28.0 mmol, 1 eq) was suspended in dry DCM (300 mL) and dry pyridine (7.00 mL, 84.0 mmol, 3 eq). The suspension was cooled to 0°C with an ice-bath and triflic anhydride (12 mL, 70.0 mmol, 2.5 eq) was slowly added via syringe under inert atmosphere. After addition, the resulting orange solution was stirred 15 min at 0°C followed by 2 h at room temperature. The reaction mixture was then concentrated in vacuo to approximately half of its volume and 150 mL diethyl ether were added. The solution was washed with 10% HClaq, followed by a saturated NaHCO3 solution and finally a saturated NaCl solution. The organic phase was then dried over MgSO4 and concentrated to dryness. The brown residue was dissolved in boiling hexane, treated with activated charcoal and stirred for 20 min. Hot filtration followed by concentration of the filtrate afforded 8 as a yellow oil that crystallised upon staying (12.8 g, 26.7 mmol, 96%). Mp: 76-78°C.
1 H-NMR (300 MHz, CD2Cl2) δ/ppm: 7.40 (dd, J = 7.5 Hz, 1.9 Hz, 2H); 7.36 (t, J = 7.5 Hz, 2H); 7.25 (dd, J = 7.5 Hz, 1.9 Hz, 2H); 4.11 (m, 4H).
19 F-NMR (282.5 MHz, CD2Cl2) δ/ppm: -76.17.
13 C-NMR (75.5 MHz, CD2Cl2) δ/ppm: 147.5, 139.6, 128.5, 128.0, 127.8, 119.9, 119.1 (q, JCF = 320.1 Hz), 36.0, 24.0.
+ HRMS (FT-MALDI): m/z calcd for C16H10F6NaO6S2 [M-Na] : 498.9715; found: 498.9714.
120
ANTHRAPHANE
Anthracene-1,8-ditriflate 9 Ditriflate 8 (8.70 g, 18.2 mmol, 1 eq) was dissolved in dry dioxane (100 mL). DDQ (5.37 g, 23.6 mmol, 1.3 eq) was added in one portion and the resulting suspension was refluxed at 125°C for 5h (conversion monitored by 1H-NMR) under inert atmosphere. The red reaction mixture was cooled to room temperature and then filtered to remove the hydroquinone. The solid was washed with DCM and the combined filtrates were concentrated to dryness. The residue was then eluted through a short silica plug (15 cm) with 20% DCM in hexane (elution followed by UV lamp at 366 nm, product appears as a blue band). Concentration of the eluate afforded pure 9 as pearly white needles (8.02 g, 16.9 mmol, 93%). The compound can be recrystallized from boiling hexane. Mp: 111°C.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.90 (s, 1H), 8.61 (s, 1H), 8.07 (d, J = 7.8 Hz, 2H); 7.62-7.50 (m, 4H).
19 13 F-NMR (282.5 MHz, CDCl3) δ/ppm: -76.36. C-NMR (75.5 MHz, CDCl3) δ/ppm: 145.7, 133.1, 128.9,
127.9, 125.6, 125.4, 118.9 (q, JCF = 320.3 Hz), 118.4, 114.1.
+ HRMS (FT-MALDI): m/z calcd for C16H8F6O6S2 [M] : 473.9661; found: 473.9661.
121
ANTHRAPHANE
Anthraphane precursor 11 (benzene-1,3,5-triyltris(ethyne-2,1-diyl))tris(anthracene-8,1-diyl) tris(trifluoromethanesulfonate) An excess amount of ditriflate 9 (2.00 g, 4.22 mmol, 5 eq) was placed under inert atmosphere in a 20 mL dry schlenk tube along with 1,3,5-triethynylbenzene 10 (126 mg, 0.84 mmol, 1 eq), Pd(PPh3)4 (49 mg, 0.04 mmol, 0.05 eq) and CuI (8 mg, 0.04 mmol, 0.05 eq). In a separate schlenk tube, a solution of dry dioxane (8 mL) and triethylamine (0.47 mL, 3.37 mmol, 4 eq) was degassed three times by freeze- pump-thaw cycles. The degassed solution was transferred via syringe into the reactant’s vessel and the obtained reaction mixture was stirred in the dark at 70°C for 36 h, during which time a greenish solid formed. After cooling, the reaction mixture was filtered and the obtained greenish solid was washed with 30 mL dioxane and 30 mL MeOH. The crude product was then suspended in warm chloroform and filtered over a pad of celite to obtain a bright yellow solution. The celite was then rinsed with copious amounts of chloroform in order to extract more product. The filtrate was concentrated to obtain 11 as a bright yellow solid (0.57 g, 0.51 mmol, 60%), which can be recrystallised from tetrachloroethane if needed. To recover the excess of starting material, the filtered reaction mixture is concentrated to dryness and the residue subjected to flash chromatography with 20% DCM in hexane as eluent (0.71 g, 1.50 mmol, 89%). Mp: decomposes above 270°C.
1 H-NMR (300 MHz, CD2Cl4) δ/ppm: 9.36 (s, 1H), 8.61 (s, 1H), 8.19 (s, 1H), 8.16-8.08 (m, 2H), 7.98 (d, J = 7.1 Hz, 1H), 7.63 (dd, J = 8.6 Hz, 7.0 Hz, 1H), 7.58-7.50 (m, 2H).
13 C-NMR (75.5 MHz, CD2Cl4) δ/ppm: 145.8, 134.5, 132.6, 131.8, 131.61, 131.57, 129.0, 128.6, 127.5,
126.0, 125.0, 124.3, 124.0, 121.3, 118.9, 118.7 (q, JCF = 321.0 Hz), 117.3, 94.0, 88.0.
19 F-NMR (282.5 MHz, CD2Cl4) δ/ppm: -76.65.
+ HRMS (FT-MALDI): m/z calcd for C57H27F9O9S3 [M] : 1122.0668; found: 1122.0673.
122
ANTHRAPHANE
Anthraphane 1 Precursor 11 (0.50 g, 0.44 mmol, 1.00 eq) was suspended in 300 mL dry toluene (1.48 mM) with 1,3,5-triethynylbenzene 10 (66.9 mg, 0.44 mmol, 1.00 eq) and dry triethylamine (12.0 mL, 89.0 mmol, 200 eq). The reaction mixture was degassed by cooling it to -80°C with an acetone-dry ice bath and then performing five cycles of vacuum (10 min) and nitrogen backfilling. Pd(PPh3)4 (103 mg, 0.09 mmol, 0.20 eq) was added with N2 counter-flow and the suspension was degassed twice again and backfilled with argon after the last cycle. After warming to room temperature, the reaction mixture was put in a preheated bath at 80°C and stirred in the dark under argon for 7 days. After cooling to room temperature, the reaction mixture was filtered: the filtrate was kept for further workup and the obtained brownish solid was washed with 50 mL MeOH. It was then suspended in hot chloroform and filtered through a celite pad to obtain a yellowish solution. The celite pad was washed with additional warm chloroform to extract as much crude product as possible and then the solution was concentrated to dryness. The filtrate obtained from the original reaction mixture was concentrated to dryness and the solid residue was washed with MeOH and separated by centrifugation. The washings and centrifugations were repeated until the methanolic phase resulted colourless. The obtained yellowish solid was combined with the solid obtained from the celite filtration and recrystallised from boiling tetrachloroethane to obtain pure 1 as a pale yellow crystalline solid (148 mg, 0.18 mmol, 40%). Mp: decomposes above 280°C.
1 H-NMR (300 MHz, CD2Cl4) δ/ppm: 9.54 (s, 1H), 8.53 (s, 1H), 8.09 (d, J = 8.6 Hz, 2H), 7.86-7.78 (m, 4H), 7.53 (dd, J = 8.6 Hz, 6.9 Hz, 2H).
+ HRMS (FT-MALDI): m/z calcd for C66H30 [M] : 822.2342; found: 822.2344.
Due to solubility problems, a resolved 13C-NMR spectrum could not be measured.
123
ANTHRAPHANE
UV/Vis spectroscopy UV/Vis absorption and fluorescence spectra were measured for ditriflate 9, precursor 11 and anthraphane 1 (λ = 365 nm for excitation wavelength). The measurements were performed at room temperature in tetrachloroethane with the following concentrations: 15 μM for absorption and 2.5 μM for emission.
Solid-state UV/Vis spectroscopy UV/Vis absorption spectra were measured on a JASCO V-660 UV-VIS-NIR Spectrophotometer (Jasco Inc., Tokyo, Japan) equipped with a 150 mm integrating sphere (ILN-725, Jasco Inc., Tokyo, Japan) using a powder holder against a white barium sulphate standard.
Solid-state NMR spectroscopy (SS-NMR) 13C-CP/MAS NMR spectra were recorded on Bruker AVANCE spectrometer operating at 176 MHz. The sample rotation frequency was 20 or 25 kHz and a 2.5 mm rotor was used. For the thermal back- reaction of the anthracene dimers, the rotor was filled with dimer crystals and a spectrum measured. The rotor was then heated in an oven at 180°C for the desired amount of time and another spectrum was measured without altering the spectrometer’s parameters, so to make a quantitative analysis possible. The process was repeated until the dimer crystals fully reverted back to the monomer crystals.
Optical (OM) and polarized microscope (POM) OM and POM was carried out with a Leica DMRX polarizing microscope equipped with a Leica DFC 480 Camera (Leica Microsystems, Heerbrugg, Switzerland) or with a Leica DM4000M optical microscope (Leica Microsystems GmbH, Wetzlar, Germany).
Electrostatic surface potential (ESP) calculations ESP calculations were performed with software Spartan (Wavefunction, Inc.). EPS and MEP values were calculated by the semi-empirical PM3 method for the fragments of cyclophane 1 and 4. DFT calculations were also performed and agree in trend with the PM3 method. Semi-empirical methods such as PM3 are well suited to represent the electrostatic potential surface of planar aromatic rings[132].
2.11.2 SC-XRD analysis General Crystals of compound 1 from ODCB, 1,3-diphenylacetone, DMPU, benzyl benzoate, diphenyl ether, 1,2-dimethoxybenzene, 1-methylnaphtalene, isophorone and NMP were measured on a Bruker
124
ANTHRAPHANE
Apex-II Duo diffractometer using a microfocus sealed-tube Cu-Kα source. Crystals of 1,1,1,3,3,3- hexachloropropane-2,2-diol, precursor 11 and crystals of compound 1 from nitrobenzene, quinoline, o-cresol, L-carvone, 1,3-dimethoxybenzene, were measured on the same instrument using graphite- monochromated sealed-tube Mo-Kα radiation (λ = 0.71073 Å). Unless otherwise indicated, crystals were kept at 100 K during measurements using an Oxford Cryosystems Cryostream 700 cooler. Data were integrated using SAINT and corrected for absorption effects using the multi-scan method (SADABS)[154]. The structures were solved using SHELXS or SHELXT[155] and refined by full-matrix least- squares analysis (SHELXL)[155] using the program package OLEX2[156]. Unless otherwise indicated, all non-hydrogen atoms were refined anisotropically. All hydrogen atom positions were constrained to ideal geometries and refined with fixed isotropic displacement parameters (in terms of a riding model). CCDC 1427487 (11), 1505883 (1,1,1,3,3,3-hexachloropropane-2,2-diol), 1427506 (1, ODCB), 1448719 (1, DMPU), 1448735 (1, 1,3-diphenylacetone), 1505884 (1, benzyl benzoate), 1427511 (1, diphenyl ether), 1427515 (1, nitrobenzene), 1448733 (1 quinoline), 1448734 (1, o-cresol), 1505885 (1, L-carvone), 1505886 (1, 1,2-dimethoxybenzene), 1505887 (1, 1,3-dimethoxybenzene), 1505888 (1, 1-methylnaphtalene), 1505889 (1, isophorone), 1427528 (1, NMP), contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44(1223)-336-033; e- mail: [email protected]), or via https://www.ccdc.cam.ac.uk/getstructures.
Crystals generally suffered from severe solvent disorder (in some cases with additional disorder in the cyclophane) and were small, both effects leading to very weak diffraction. Usually Cu radiation had to be used and long exposure times had to be chosen, raising additional problems with sample icing. With the exception of 1,1,1,3,3,3-hexachloropropane-2,2-diol, precursor 11 and the anthraphane solvates of 1,3-diphenylacetone, 1,2-dimethoxybenzene and isophorone, only substandard resolutions could be achieved and these structures should not be assessed applying the quality criteria of routine small molecule structures. Specifically, alerts rising due to low angular resolution and low completeness in the highest resolution shells are to be expected. We believe the structure quality is sufficient for the discussion of packing motifs, but certainly bond distances and angles will be biased. Many crystals were tested and the best one chosen for measurement. For anthraphane structures of ODCB and diphenyl ether, several full datasets were recorded independently (using different crystals) and the best results selected (in these cases the structures were identical except for statistical quality criteria and resolution).
125
ANTHRAPHANE
1,1,1,3,3,3-hexachloropropane-2,2-diol
Figure 2.107. ORTEP diagram of 1,1,1,3,3,3-hexachloropropane-2,2-diol (50% probability).
Sample and crystal data CCDC No. 1505883
Empirical formula C3H8Cl6O5 Formula weight 336.79 Temperature/K 100.0(2) Crystal system orthorhombic Space group Pbca a/Å 12.0067(17) b/Å 7.4963(14) c/Å 26.330(4) α/° 90 β/° 90 γ/° 90 Volume/Å3 2369.9(7) Z 8 3 ρcalcg/cm 1.888 μ/mm-1 1.441 F(000) 1344.0 Crystal size/mm3 0.18 × 0.16 × 0.015 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 4.592 to 55.376 Index ranges -15 ≤ h ≤ 15, -9 ≤ k ≤ 9, -34 ≤ l ≤ 34 Reflections collected 34009
Independent reflections 2753 [Rint = 0.0766, Rsigma = 0.0362] Data/restraints/parameters 2753/0/138 Goodness-of-fit on F2 1.052
Final R indexes [I>=2σ (I)] R1 = 0.0326, wR2 = 0.0633
Final R indexes [all data] R1 = 0.0496, wR2 = 0.0689 Largest diff. peak/hole / e Å-3 0.41/-0.54
126
ANTHRAPHANE
Compound 11
Figure 2.108. ORTEP diagram of 11 (50% probability). A 70:30 disorder was found in all OTf ligands and the solvent molecule (shown in the image and fully modelled).
Sample and crystal data CCDC No. 1427487
Empirical formula C59H29Cl4F9O9S3 Formula weight 1290.80 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.642(2) b/Å 14.778(3) c/Å 15.643(3) α/° 108.404(4) β/° 110.867(3) γ/° 98.818(4) Volume/Å3 2667.5(8) Z 2 3 ρcalcg/cm 1.607 μ/mm-1 0.432 F(000) 1304.0 Crystal size/mm3 0.23 × 0.16 × 0.06 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 3.044 to 55.168 Index ranges -17 ≤ h ≤ 17, -19 ≤ k ≤ 19, -20 ≤ l ≤ 20 Reflections collected 42329
Independent reflections 12131 [Rint = 0.0314, Rsigma = 0.0336] Data/restraints/parameters 12131/1335/929 Goodness-of-fit on F2 1.019
Final R indexes [I>=2σ (I)] R1 = 0.0568, wR2 = 0.1390
Final R indexes [all data] R1 = 0.0864, wR2 = 0.1589 Largest diff. peak/hole / e Å-3 0.69/-0.60
127
ANTHRAPHANE
ODCB solvate
Figure 2.109. ORTEP diagram of the ODCB solvate of 1 (50% probability). Note: one ODCB solvent molecule was too disordered to be modeled and was removed from the density using masking techniques inbuilt in OLEX2.
Sample and crystal data CCDC No. 1427506
Empirical formula C84Cl6H42 Formula weight 1263.87 Temperature/K 100 Crystal system triclinic Space group P-1 a/Å 15.231(4) b/Å 15.402(4) c/Å 18.805(6) α/° 92.352(11) β/° 110.492(11) γ/° 118.971(9) Volume/Å3 3493.7(17) Z 2 3 ρcalcmg/mm 1.201 m/mm-1 2.577 F(000) 1296.0 Crystal size/mm3 0.08 × 0.06 × 0.02 Radiation CuKα (λ = 1.54178) 2Θ range for data collection 5.2 to 118.26° Index ranges -16 ≤ h ≤ 16, -10 ≤ k ≤ 17, -20 ≤ l ≤ 20 Reflections collected 15034 Independent reflections 8693[R(int) = 0.0402] Data/restraints/parameters 8693/0/811 Goodness-of-fit on F2 1.026
Final R indexes [I>=2σ (I)] R1 = 0.0802, wR2 = 0.2218
Final R indexes [all data] R1 = 0.1039, wR2 = 0.2402 Largest diff. peak/hole / e Å-3 1.47/-1.10
128
ANTHRAPHANE
1,3-Diphenylacetone solvate
Figure 2.110. ORTEP diagram of anthraphane 1 in the 1,3-diphenylacetone solvate (50% probability).
Sample and crystal data CCDC No. 1448735
Empirical formula C96H58O2 Formula weight 1243.42 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 15.1984(6) b/Å 15.6148(6) c/Å 17.8003(7) α/° 75.887(3) β/° 67.008(3) γ/° 61.703(2) Volume/Å3 3415.1(3) Z 2 3 ρcalcg/cm 1.209 μ/mm-1 0.543 F(000) 1300.0 Crystal size/mm3 0.12 × 0.04 × 0.01 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 6.446 to 133.33 Index ranges -17 ≤ h ≤ 12, -18 ≤ k ≤ 17, -21 ≤ l ≤ 20 Reflections collected 39065
Independent reflections 11773 [Rint = 0.0514, Rsigma = 0.0485] Data/restraints/parameters 11773/114/883 Goodness-of-fit on F2 1.023
Final R indexes [I>=2σ (I)] R1 = 0.0619, wR2 = 0.1678
Final R indexes [all data] R1 = 0.0871, wR2 = 0.1880 Largest diff. peak/hole / e Å-3 0.61/-0.41
129
ANTHRAPHANE
DMPU solvate
Figure 2.111. ORTEP diagram of anthraphane 1 in the DMPU solvate (50% probability). Overall severe solvent disorder. 12 DMPU in asymmetric unit, all but one are disordered. Solvents are modeled but some are massively restrained. Within the channel there are two groups/chains of solvents: one with 60:40 and one with 70:30 relative occupation.
Sample and crystal data CCDC No. 1448719
Empirical formula C198H192N22O11 Formula weight 3055.72 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 15.7178(2) b/Å 24.0271(4) c/Å 24.3300(4) α/° 69.5590(10) β/° 77.4220(10) γ/° 75.9190(10) Volume/Å3 8261.4(2) Z 2 3 ρcalcg/cm 1.228 μ/mm-1 0.607 F(000) 3244.0 Crystal size/mm3 0.22 × 0.1 × 0.07 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 3.918 to 133.418 Index ranges -18 ≤ h ≤ 17, -27 ≤ k ≤ 28, -28 ≤ l ≤ 21 Reflections collected 101670
Independent reflections 28823 [Rint = 0.0350, Rsigma = 0.0314] Data/restraints/parameters 28823/3422/2704 Goodness-of-fit on F2 1.600
Final R indexes [I>=2σ (I)] R1 = 0.1034, wR2 = 0.3420
Final R indexes [all data] R1 = 0.1206, wR2 = 0.3662 Largest diff. peak/hole / e Å-3 1.23/-0.85 130
ANTHRAPHANE
Benzyl benzoate solvate
Figure 2.112. ORTEP diagram of anthraphane in the benzyl benzoate solvate (50% probability). Some of the disorder (2 solvent sites) could be identified. There is more delocalized solvent in-between (1- 2 molecules), which has been masked.
Sample and crystal data CCDC No. 1505884
Empirical formula C94H54O4 Formula weight 1247.37 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 15.5379(3) b/Å 15.6519(3) c/Å 19.0657(4) α/° 80.8370(10) β/° 85.9920(10) γ/° 60.2930(10) Volume/Å3 3975.72(14) Z 2 3 ρcalcg/cm 1.042 μ/mm-1 0.488 F(000) 1300.0 Crystal size/mm3 0.14 × 0.08 × 0.03 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 4.694 to 133.186 Index ranges -18 ≤ h ≤ 17, -18 ≤ k ≤ 18, -22 ≤ l ≤ 16 Reflections collected 41984
Independent reflections 13815 [Rint = 0.0684, Rsigma = 0.0737] Data/restraints/parameters 13815/452/878 Goodness-of-fit on F2 1.260
Final R indexes [I>=2σ (I)] R1 = 0.1218, wR2 = 0.3432
Final R indexes [all data] R1 = 0.1598, wR2 = 0.3748 Largest diff. peak/hole / e Å-3 1.54/-0.75
131
ANTHRAPHANE
Diphenyl ether solvate
Figure 2.113. ORTEP diagram of anthraphane 1 in the diphenyl ether solvate (50% probability). One solvent molecule is severely disordered (completely modelled).
Sample and crystal data CCDC No. 1427511
Empirical formula C90H50O2
Formula weight 1163.30
Temperature/K 100.0(2)
Crystal system monoclinic
Space group P21/c a/Å 16.5811(3) b/Å 15.2702(3) c/Å 24.5373(5)
α/° 90
β/° 100.3080(10)
γ/° 90
Volume/Å3 6112.5(2)
Z 4
3 ρcalcg/cm 1.264
μ/mm-1 0.571
F(000) 2424.0
Crystal size/mm3 0.04 × 0.02 × 0.01
Radiation CuKα (λ = 1.54178)
2Θ range for data collection/° 5.418 to 132.794
Index ranges -14 ≤ h ≤ 19, -17 ≤ k ≤ 17, -8 ≤ l ≤ 8
Reflections collected 35389
Independent reflections 4273 [Rint = 0.0601, Rsigma = 0.0345]
Data/restraints/parameters 4273/1238/857
Goodness-of-fit on F2 1.030
Final R indexes [I>=2σ (I)] R1 = 0.0381, wR2 = 0.0916
Final R indexes [all data] R1 = 0.0539, wR2 = 0.1016
Largest diff. peak/hole / e Å-3 0.28/-0.28
132
ANTHRAPHANE
Nitrobenzene solvate
Figure 2.114. ORTEP diagram of anthraphane 1 in the nitrobenzene solvate (50% probability). Both nitrobenzene molecules in the asymmetric unit are disordered about an inversion point.
Sample and crystal data CCDC No. 1427515
Empirical formula C72H35NO2
Formula weight 946.01
Temperature/K 100.0(2)
Crystal system triclinic
Space group P-1 a/Å 13.250(5) b/Å 14.375(5) c/Å 14.817(5)
α/° 91.436(7)
β/° 101.907(7)
γ/° 115.796(6)
Volume/Å3 2465.0(15)
Z 2
3 ρcalcmg/mm 1.275 m/mm-1 0.076
F(000) 980.0
Crystal size/mm3 0.23 × 0.19 × 0.02
Radiation MoKα (λ = 0.71073)
2Θ range for data collection 2.834 to 50.052°
Index ranges -17 ≤ h ≤ 17, -18 ≤ k ≤ 10, -19 ≤ l ≤ 18
Reflections collected 39472
Independent reflections 8354 [Rint = 0.0671, Rsigma = 0.0947]
Data/restraints/parameters 8354/870/679
Goodness-of-fit on F2 1.018
Final R indexes [I>=2σ (I)] R1 = 0.0653, wR2 = 0.1630
Final R indexes [all data] R1 = 0.1319, wR2 = 0.1984
Largest diff. peak/hole / e Å-3 0.88/-0.80
133
ANTHRAPHANE
Quinoline solvate
Figure 2.115. ORTEP diagram of anthraphane 1 in the quinoline solvate (50% probability). Two half solvents in the asymmetric unit, both disordered (one only disordered about the inversion (50:50), the other about the inversion and additionally split (4x 25% occ.). Restraints applied for the solvents, basically only their orientation was refined.
Sample and crystal data CCDC No. 1448733
Empirical formula C75H37N Formula weight 952.05 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.2486(11) b/Å 14.4507(12) c/Å 14.8798(12) α/° 91.989(2) β/° 102.018(2) γ/° 115.489(2) Volume/Å3 2490.1(4) Z 2 ρcalcg/cm3 1.270 μ/mm-1 0.073 F(000) 988.0 Crystal size/mm3 0.28 × 0.16 × 0.04 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 2.826 to 55.244 Index ranges -17 ≤ h ≤ 17, -17 ≤ k ≤ 18, -19 ≤ l ≤ 19 Reflections collected 41818
Independent reflections 11273 [Rint = 0.0301, Rsigma = 0.0390] Data/restraints/parameters 11273/1065/793 Goodness-of-fit on F2 1.041
Final R indexes [I>=2σ (I)] R1 = 0.0876, wR2 = 0.2620
Final R indexes [all data] R1 = 0.1239, wR2 = 0.2979 Largest diff. peak/hole / e Å-3 2.13/-1.58
134
ANTHRAPHANE
o-Cresol solvate
Figure 2.116. ORTEP diagram of anthraphane 1 in the o-cresol solvate (50% probability). Two half solvents in the asymmetric unit, both disordered about inversion.
Sample and crystal data CCDC No. 1448734
Empirical formula C73H38O Formula weight 931.03 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.250(6) b/Å 14.510(6) c/Å 14.908(6) α/° 91.851(17) β/° 101.701(17) γ/° 115.336(16) Volume/Å3 2513.6(19) Z 2 3 ρcalcg/cm 1.230 μ/mm-1 0.071 F(000) 968.0 Crystal size/mm3 0.36 × 0.24 × 0.02 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 2.816 to 54.962 Index ranges -17 ≤ h ≤ 17, -18 ≤ k ≤ 18, -19 ≤ l ≤ 18 Reflections collected 27808
Independent reflections 11167 [Rint = 0.0787, Rsigma = 0.1407] Data/restraints/parameters 11167/204/688 Goodness-of-fit on F2 1.041
Final R indexes [I>=2σ (I)] R1 = 0.1093, wR2 = 0.2911
Final R indexes [all data] R1 = 0.2222, wR2 = 0.3655 Largest diff. peak/hole / e Å-3 2.56/-0.80
135
ANTHRAPHANE
L-carvone solvate
Figure 2.117. ORTEP diagram of the L-carvone solvate (50% probability). The solvent due to severe disorder could not be modeled and was therefore removed by masking techniques.
Sample and crystal data CCDC No. 1505885
Empirical formula C66H30 Formula weight 822.90 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.1888(17) b/Å 14.4544(18) c/Å 15.4509(17) α/° 93.033(4) β/° 102.262(4) γ/° 114.623(4) Volume/Å3 2583.2(5) Z 2 3 ρcalcg/cm 1.058 μ/mm-1 0.060 F(000) 852.0 Crystal size/mm3 0.24 × 0.1 × 0.06 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 3.14 to 55.132 Index ranges -17 ≤ h ≤ 17, -18 ≤ k ≤ 18, -17 ≤ l ≤ 20 Reflections collected 40186
Independent reflections 11746 [Rint = 0.0422, Rsigma = 0.0524] Data/restraints/parameters 11746/0/595 Goodness-of-fit on F2 1.082
Final R indexes [I>=2σ (I)] R1 = 0.0789, wR2 = 0.2183
Final R indexes [all data] R1 = 0.1110, wR2 = 0.2312 Largest diff. peak/hole / e Å-3 0.31/-0.35
136
ANTHRAPHANE
1,2-dimethoxybenzene solvate
Figure 2.118. ORTEP diagram of the 1,2-dimethoxybenzene solvate (50% probability).
Sample and crystal data CCDC No. 1505886
Empirical formula C74H40O2 Formula weight 961.06 Temperature/K 100.0(2) Crystal system monoclinic Space group C2/c a/Å 20.4490(7) b/Å 21.2830(7) c/Å 24.6965(8) α/° 90 β/° 108.524(3) γ/° 90 Volume/Å3 10191.5(6) Z 8 3 ρcalcg/cm 1.253 μ/mm-1 0.572 F(000) 4000.0 Crystal size/mm3 0.12 × 0.05 × 0.04 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 6.166 to 133.182 Index ranges -24 ≤ h ≤ 8, -23 ≤ k ≤ 25, -25 ≤ l ≤ 29 Reflections collected 36572
Independent reflections 8897 [Rint = 0.0464, Rsigma = 0.0363] Data/restraints/parameters 8897/0/687 Goodness-of-fit on F2 1.014
Final R indexes [I>=2σ (I)] R1 = 0.0387, wR2 = 0.0967
Final R indexes [all data] R1 = 0.0551, wR2 = 0.1063 Largest diff. peak/hole / e Å-3 0.19/-0.19
137
ANTHRAPHANE
1,3-dimethoxybenzene solvate
Figure 2.119. ORTEP diagram of the 1,3-dimethoxybenzene solvate (50% probability). Two molecules and two solvents per asymmetric unit. Both solvents disordered, but only one had to be modelled (60:40 occupation). One cyclophane is disordered in one anthracene.
Sample and crystal data CCDC No. 1505887
Empirical formula C74H40O2 Formula weight 961.06 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 14.5278(17) b/Å 15.0150(18) c/Å 24.185(3) α/° 76.598(5) β/° 82.713(5) γ/° 88.583(5) Volume/Å3 5090.5(10) Z 4 3 ρcalcg/cm 1.254 μ/mm-1 0.074 F(000) 2000.0 Crystal size/mm3 0.16 × 0.08 × 0.06 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 2.788 to 34.472 Index ranges -12 ≤ h ≤ 12, -12 ≤ k ≤ 12, -20 ≤ l ≤ 20 Reflections collected 20240
Independent reflections 6152 [Rint = 0.0467, Rsigma = 0.0542] Data/restraints/parameters 6152/1956/1442 Goodness-of-fit on F2 1.231
Final R indexes [I>=2σ (I)] R1 = 0.0867, wR2 = 0.2533
Final R indexes [all data] R1 = 0.1253, wR2 = 0.3058 Largest diff. peak/hole / e Å-3 0.84/-0.49
138
ANTHRAPHANE
1-Methylnaphtalene solvate
Figure 2.120. ORTEP diagram of the 1-methylnaphtalene solvate (50% probability). Two half solvents in the asymmetric unit, both disordered. One is only disordered about the inversion center (50:50), the other about the inversion and additionally split (4x 25% occupancy). Restraints applied for the solvents, basically only their orientation was refined.
Sample and crystal data CCDC No. 1505888
Empirical formula C77H39 Formula weight 964.08 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.1858(4) b/Å 14.5908(5) c/Å 15.5997(6) α/° 77.023(3) β/° 75.331(3) γ/° 63.147(2) Volume/Å3 2569.15(17) Z 2 3 ρcalcg/cm 1.246 μ/mm-1 0.541 F(000) 1002.0 Crystal size/mm3 0.08 × 0.03 × 0.02 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 8.378 to 133.5 Index ranges -15 ≤ h ≤ 15, -17 ≤ k ≤ 16, -18 ≤ l ≤ 13 Reflections collected 31028
Independent reflections 8711 [Rint = 0.0621, Rsigma = 0.0577] Data/restraints/parameters 8711/173/705 Goodness-of-fit on F2 1.659
Final R indexes [I>=2σ (I)] R1 = 0.1247, wR2 = 0.3943
Final R indexes [all data] R1 = 0.1660, wR2 = 0.4347 Largest diff. peak/hole / e Å-3 1.62/-0.73
139
ANTHRAPHANE
Isophorone solvate
Figure 2.121. ORTEP diagram of the isophorone solvate (50% probability). Two and a half solvents in the asymmetric unit, one of which is disordered about the inversion.
Sample and crystal data CCDC No. 1505889
Empirical formula C177H130O5 Formula weight 2336.80 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.2429(9) b/Å 14.1163(9) c/Å 19.8453(14) α/° 102.531(4) β/° 94.499(4) γ/° 115.374(4) Volume/Å3 3209.8(4) Z 1 3 ρcalcg/cm 1.209 μ/mm-1 0.546 F(000) 1232.0 Crystal size/mm3 0.16 × 0.12 × 0.025 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 8.048 to 133.61 Index ranges -15 ≤ h ≤ 12, -15 ≤ k ≤ 16, -23 ≤ l ≤ 23 Reflections collected 40618
Independent reflections 11095 [Rint = 0.0432, Rsigma = 0.0384] Data/restraints/parameters 11095/103/874 Goodness-of-fit on F2 1.046
Final R indexes [I>=2σ (I)] R1 = 0.0445, wR2 = 0.1193
Final R indexes [all data] R1 = 0.0587, wR2 = 0.1299 Largest diff. peak/hole / e Å-3 0.30/-0.22
140
ANTHRAPHANE
NMP solvate
Figure 2.122. ORTEP diagram of anthraphane 1 in the NMP solvate (50% probability). Note: one additional solvent rests on the 3 axis and is too diffusely disordered to be clearly resolved. The corresponding density was removed from the data using masking techniques inbuilt in OLEX2.
Sample and crystal data CCDC No. 1427528
Empirical formula C96H84N6O6
Formula weight 1417.69
Temperature/K 100.0(2)
Crystal system trigonal
Space group R-3c a/Å 16.4148(8) b/Å 16.4148(8) c/Å 50.878(3)
α/° 90
β/° 90
γ/° 120
Volume/Å3 11872.1(13)
Z 6 3 ρcalcg/cm 1.190
μ/mm-1 0.584
F(000) 4500.0
Crystal size/mm3 0.12 × 0.08 × 0.015
Radiation CuKα (λ = 1.54178)
2Θ range for data collection/° 7.124 to 124.694
Index ranges -18 ≤ h ≤ 18, -18 ≤ k ≤ 18, -55 ≤ l ≤ 57
Reflections collected 21653
Independent reflections 2075 [Rint = 0.0383, Rsigma = 0.0196]
Data/restraints/parameters 2075/172/165
Goodness-of-fit on F2 2.165
Final R indexes [I>=2σ (I)] R1 = 0.0981, wR2 = 0.3669
Final R indexes [all data] R1 = 0.1076, wR2 = 0.3849
Largest diff. peak/hole / e Å-3 1.66/-0.56
141
ANTHRAPHANE
Topochemical single-crystal-to-single-crystal photodimerisations of anthraphane
An in-house built cylindrical photoreactor equipped with 16 x 2.8 W high power blue LEDs with λ =465 nm (from Seoul Semiconductors), each at 22 lm was used as irradiation source and a cooler from an x-ray diffractometer machine was used to keep the temperature at the desired value during the photoreaction. Single crystals were selected and mounted on the pin of a goniometer head and their unit cell and diffraction quality was checked by collecting enough frames on the diffractometer. The pin was subsequently detached from the goniometer head and put under the cooler in the middle of the photoreactor where irradiation was performed at constant temperature for the desired amount of time. After irradiation, the crystal was again checked by SC-XRD and if diffracting properly, a complete dataset was collected for structure determination.
142
ANTHRAPHANE
1,3-dimethoxybenzene solvate dimers
Figure 2.123. ORTEP diagram of the anthraphane dimers in the 1,3-dimethoxybenzene solvate (50% probability).
Sample and crystal data CCDC No. 1505891
Empirical formula C148H80O4 Formula weight 961.06 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 14.6027(11) b/Å 17.9119(15) c/Å 21.2636(18) α/° 72.935(5) β/° 76.088(4) γ/° 79.524(5) Volume/Å3 5123.7(7) Z 2 3 ρcalcg/cm 1.246 μ/mm-1 0.569 F(000) 2000.0 Crystal size/mm3 0.12 × 0.08 × 0.03 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 4.436 to 133.536 Index ranges -13 ≤ h ≤ 17, -21 ≤ k ≤ 20, -25 ≤ l ≤ 24 Reflections collected 41309
Independent reflections 17448 [Rint = 0.0381, Rsigma = 0.0487] Data/restraints/parameters 17448/0/1373 Goodness-of-fit on F2 1.028
Final R indexes [I>=2σ (I)] R1 = 0.0500, wR2 = 0.1288
Final R indexes [all data] R1 = 0.0702, wR2 = 0.1426 Largest diff. peak/hole / e Å-3 0.64/-0.40
143
ANTHRAPHANE
Isophorone solvate dimers
Figure 2.124. ORTEP diagram of the anthraphane dimers in the isophorone solvate (50% probability).
Sample and crystal data CCDC No. 1505890
Empirical formula C177H130O5 Formula weight 2336.80 Temperature/K 100.0(2) Crystal system triclinic Space group P-1 a/Å 13.4200(14) b/Å 14.3444(17) c/Å 19.785(2) α/° 103.867(4) β/° 91.639(4) γ/° 117.634(3) Volume/Å3 3232.2(6) Z 1 3 ρcalcg/cm 1.201 μ/mm-1 0.071 F(000) 1232.0 Crystal size/mm3 0.12 × 0.1 × 0.05 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 5.068 to 55.186 Index ranges -17 ≤ h ≤ 17, -18 ≤ k ≤ 18, -25 ≤ l ≤ 25 Reflections collected 34160
Independent reflections 14888 [Rint = 0.0519, Rsigma = 0.0905] Data/restraints/parameters 14888/75/874 Goodness-of-fit on F2 1.007
Final R indexes [I>=2σ (I)] R1 = 0.0600, wR2 = 0.1249
Final R indexes [all data] R1 = 0.1204, wR2 = 0.1499 Largest diff. peak/hole / e Å-3 0.59/-0.30
144
AMPHIPHILIC ANTHRAPHANES
3. Amphiphilic Anthraphanes
In the previous chapter, a new potential monomer for the synthesis of 2D polymers was developed, anthraphane 1. Considerable efforts were put into forcing the molecule to pack in single crystals with the anthracene units stacking ftf so to allow a photoinduced topochemical 2D polymerisation to take place (through anthracene dimerisation), but unfortunately the required all-ftf packing was not obtained. Having developed a robust and reliable synthesis of 1, questions arose if by chemically modifying the basic structure of anthraphane, the required all-ftf packing could be obtained by other means, such as the air/water interface. While not strictly necessary[53,54,56], amphiphilicity can help to stabilise the conformation/orientation of molecules at the air/water interface, therefore a synthetic approach towards amphiphilic anthraphanes was developed.
3.1 Monomer’s Design
Decoration of anthaphane with polar groups to turn it into an amphiphile can either take place on the anthracene units or the central benzene cores, or both (Figure 3.1).
Figure 3.1. Proposed structures for amphiphilic anthraphanes.
However, from a mere synthetic point of view, modification of 1,8-substituted anthracenes can be very challenging. Besides, due to its robustness, it was desirable to change the original synthesis of anthraphane as little as possible. Therefore the most reasonable amphiphilic structure was chosen to be the one that bears polar groups on the central benzene core only. The synthetic strategy depicted in Scheme 3.1 was chosen as the most feasible, with a polar triethynylbenzene core reacting with precursor 11 and yielding the desymmetrised amphiphilic monomer in the final step of the synthetic sequence.
145
AMPHIPHILIC ANTHRAPHANES
Scheme 3.1. Proposed synthetic scheme towards amphiphilic anthraphanes.
Compound 2 is expected to be oriented at the air/water interface with the polar moieties being in contact with the water subphase and the anthracenes standing vertically in the air. In the ideal case, the molecules are compressed (or spontaneously assembled) into a honeycomb-like crystalline state, in which the anthracenes units are stacking ftf, similar to the hexagonal packing required in the single crystal approach of Chapter 2. Upon photoirradiation, the monomer monolayer is then converted into a polymer monolayer (Figure 3.2).
Figure 3.2. Schematic representation of the synthesis of 2DPs at the air/water interface.
Because they are obtained in form of single layers at the air/water interface, their structure analysis is challenging. Compared to the single crystal approach, which provides up to gram quantities of material, single layers have extremely low mass. This low mass poses enormous difficulty regarding the sensitivity of practically all available analytical methods. As mentioned previously, for detection of order, it is important that during the time-scale of the experiment, the polymer structure remains shape-persistent. In this regard, the structural rigidity of anthraphane could reflect into a rigid and shape-persistent polymer, which is a very desirable feature for characterisation.
146
AMPHIPHILIC ANTHRAPHANES
3.2 Synthesis of Anthraphane-tri(OMe)
For the synthesis of monomer 2, the substituted 1,3,5-triethynylbenzene core 16 was synthesised first. The proposed synthetic strategy to the polar benzene core is reported in Scheme 3.2. It was of primary importance for the polar moieties to be compatible with the basic conditions of the Sonogashira and desylilation reaction steps: in this regard free alcohols and amines were discarded (strategies involving protection groups were not considered due to additional synthetic steps). Another important factor was to minimise the sterical bulkiness of the substituents, as performing a Sonogashira reaction on sterically demanding hexa-substituted benzenes can be challenging.
Scheme 3.2. Proposed synthetic strategy towards the polar 1,3,5-triethynylbenzene derivative 16.
Therefore, the nitrogens of triazines, aldehydes and acetyl groups were considered as possible polar functional groups for conferring amphiphilicity. The attractiveness of these versatile groups lies in the fact that they could be further functionalised once on the 2D polymer, opening the path to application-oriented post-polymerisation modifications (PPM).
Scheme 3.3. Synthesis of various substituted polar cores.
147
AMPHIPHILIC ANTHRAPHANES
The synthetic attempts towards the polar cores pursued in this thesis are shown in Scheme 3.3. They were unfortunately all unsuccessful due to different reasons that will be concisely addressed in the following paragraphs. For the synthesis of the triazine derivative 19, the standard procedure based on the Negishi coupling[157] described by Sonoda and Tobe[158] was applied to yield the TMS-substituted derivative 18. Desylilation of the compound by standard methods such as methanolate solutions (MeOH, K2CO3) or fluoride based reagents (i.e. TBAF or KF) did not yield the desired product. It is believed that the desylated product might be susceptible to nucleophilic attack on the terminal alkynes yielding substituted alkyl derivatives; other plausible side reactions could occur during the formation of the very reactive acetylide anion and lead to cross-linked carbon- nitrogen networks[159]. For the synthesis of the tricarbaldehyde core 24, the key building block 2,4,6- tribromobenzene-1,3,5-tricarboxaldehyde 22 was chosen, also known as Rubin’s aldehyde[160]. A two- step updated and efficient synthesis from Holst and Meier[161], starting from 1,3,5-tribromobenzene 20 and involving a Friedel-Crafts alkylation with chloroform, followed by a FeII-catalysed hydrolysis, yielded Rubin’s aldehyde 22 without problems in 70% yield. The following step, the Sonogashira cross-coupling to the TMS-substituted compound 23 worked only with 37% yield[162] and the subsequent desylilation did not yield the desired product. Again, mild conditions such as
MeOH/K2CO3 or more aggressive reagents such as TBAF were tried without success. Aldehyde groups would have been of particular interest for PPM due to their high versatility[163,164]. For the synthesis of the triacetyl-substituted core 28, standard bromination conditions[165] were applied on phloroglucinol 25 followed by acetylation[166], affording the key building block 27 in nearly quantitative yields. Sonogashira cross-coupling of 27 with TMSA did not yield product 28. In this regard, care was taken in performing the reaction in strictly anhydrous conditions and using non-nucleophilic bases such as 2,6-lutidine or DIPEA, to avoid cleavage of the acetyl groups. According to literature procedures the acetyl groups should be compatible with the reaction[167], it is however possible that in this case, the two acetyl groups being in α-position with respect to the bromide might coordinate to the catalyst and trigger unwanted side reactions. In fact, in all reaction batches, the reaction mixture turned into a tarry residue, which upon mass spectrometry analysis indicated the presence of high molecular masses but no presence of the desired product.
148
AMPHIPHILIC ANTHRAPHANES
In the context of this work, acetyl groups on the monomer were regarded as the perfect versatile groups for tuning its amphiphilicity, as they could be easily removed if needed, exposing the more hydrophilic hydroxyl groups, which in turn could be re-functionalised at will. As alternative, the attention was directed on the synthesis of the methoxy-substituted polar core 16a, as ethers are very robust functional groups. The major drawback of using ethers is however the reduced chance of functionalisation of the monomer and polymer respectively, as cleavage of ether groups (in this case a demethylation) usually involves harsh acidic conditions or electrophilic reagents such as BBr3, which would also attack the triple bonds of the monomer[168]. Despite these drawbacks, we nevertheless proposed a synthesis involving ether groups .
The synthesis of the trimethoxy-substituted benzene core 16a is depicted in Scheme 3.4 and was adapted and modified from an already published procedure from Hennrich[169,170].
Scheme 3.4. Synthesis of the the trimethoxy-substituted benzene core 16a.
In a first step, 1,3,5-trifluorobenzene 12 was quantitatively iodinated with periodic acid in sulfuric acid to give compound 13[171]. Triple nucleophilic aromatic substitution of the fluorides in anhydrous DMI with freshly prepared sodium methoxide yielded the methoxy substituted iodinated compound 14a in 85% yield. For the Sonogashira cross-coupling, the reaction conditions were slightly changed due to the particularly poor reactivity of 14a towards oxidative addition[172]. Instead of the
t [173] standard Pd(PPh3)4, the more aggressive and nucleophilic catalyst Pd(P Bu3)2 was used , which was
t generated in situ from Pd(OAc)2 and by deprotonation of HP Bu3BF4 with DIPA. For such an electron- rich substrate, the reaction worked very well with yields up to 65%. In a final step, desylilation of 15a with potassium carbonate in methanol afforded quantitatively the target compound 16a in high purity as displayed in Figure 3.3.
149
AMPHIPHILIC ANTHRAPHANES
1 Figure 3.3. H-NMR spectrum of compound 16a in CDCl3.
With the polar core in hand, the substituted anthraphane monomer 2a was then assembled by using the same conditions employed for the synthesis of anthraphane (Scheme 3.5). Conveniently, the final step worked as nicely as in the original synthesis and the work-up to isolate the monomer did not have to be changed: from 320 mg of precursor 11, 92 mg of 2a were obtained in 36% yield.
Scheme 3.5. Synthesis of the anthraphane-tri(OMe) monomer 2a.
150
AMPHIPHILIC ANTHRAPHANES
From a physical property point of view, 2a does not differ from its congener 1 in terms of thermal stability (around 280°C) or its UV/Vis spectrum. The general solubility of 2a was however a bit better due to its desymmetrised structure and the three methoxy groups. The structure of anthraphane-tri(OMe) was confirmed by 1H-NMR spectroscopy as shown in Figure 3.4, 13C-NMR spectroscopy and by SC-XRD. The monomer could be recrystallized from low boiling point solvents such as THF and exhibited the etf packing 1 discussed in Chapter 2. Crystallisation from nitrobenzene instead, yielded a packing that allowed for a topochemical 1D polymerisation: for more details into the packing and the polymerisation, please refer to the Appendix at the end of this thesis.
1 Figure 3.4. H-NMR spectrum of anthraphane-tri(OMe) 2a in CD2Cl4 at room temperature.
It has to be noted that substitutions on the central benzene cores do not affect the ESP on the anthracene units due to the orthogonality of the π systems: from a plain electrostatic point of view, anthracenes will tend to prefer etf orientations. In this sense, substituents such as methoxy groups would likely influence the crystal packing merely from a sterical point of view. Still, 2a could be an interesting candidate for the topochemical synthesis of 2DP in single crystals; however, a similar crystallisation screening procedure as the one used for anthraphane would be needed, to obtain the necessary all-ftf packing with the right solvate.
151
AMPHIPHILIC ANTHRAPHANES
3.3 Spreading of 2a at the Air/Water Interface
The interfacial behaviour of anthraphane-tri(OMe) was tested by spreading a dilute solution of the monomer in chloroform (0.5 mg/mL) at the air/water interface. During compression of the molecules, a surface pressure vs mean molecular area (SP vs MMA) isotherm was recorded and film formation was visually followed by Brewster angle microscopy (BAM). The results are summarised in Figures 3.5 and 3.6. Just by looking at the MMA axis of the isotherm, one sees immediately that the values are far below the expected ones: the curve slowly increases until 50 Å2 and then steeply grows until a SP of 60 mN/m is reached. Such MMA values are typical for molecules with a very small cross- section such as fully stretched aliphatic chain amphiphiles[49] (i.e. decanoic acid) and are incompatible with the structure of 2a. Such small MMAs are an indication of an inhomogeneous distribution of thickness due to low amphiphilicity: the molecules lack a preferred orientation upon compression and therefore continuously slide on top of each another without a visible transition. An alternative would be that the compound is pushed into the subphase. This is considered unlikely in view of the still considerable hydrophobicity of 2a.
Figure 3.5. SP vs MMA isotherm of anthraphane 2a at the air/water interface. The low MMA values indicate multilayer formation, which could be a consequence of the low amphiphilicity of the molecule.
Confirmation of such thickness distribution came from visual inspection of the film through BAM: in a series of BAM micrographs recorded at different SP values (Figure 3.6), the film always appeared inhomogeneous, with thicker domains clearly visible due to the higher contrast.
152
AMPHIPHILIC ANTHRAPHANES
Figure 3.6. BAM micrographs recorded at different surface pressures during the compression process. The film always looks very inhomogeneous independently from the surface pressure: thicker domains are clearly visible (higher contrast) and confirm multilayer formation.
From these results it was clear that the methoxy groups on the benzene core of 16a do not confer sufficient amphiphilicity. The monomer 2a was therefore discarded for the air/water interface approach and a new synthetic approach was devised in order to further improve the amphiphilicity.
3.4 Synthesis of Anthraphane-tri(DEGME)
To increase the amphiphilicity of the monomer, the three methoxy groups of 16a were replaced by the three diethylene glycol methyl ether (DEGME) groups in 16b. The synthesis of this new polar benzene core is depicted in Scheme 3.6.
Scheme 3.6. Synthesis of the diethylene glycol methyl ether-substituted benzene core 16b.
Overall the synthesis proceeded analogously to the previous one and yielded the product 16b without any particular problems. The DEGME chains conferred more solubility to the intermediates
153
AMPHIPHILIC ANTHRAPHANES and therefore column chromatography was employed as purification method. Gram amounts of pure 16b could be easily produced (see Figure 3.7). This successful synthetic strategy highlights the versatility of building block 13, which can be in principle substituted by a variety of nucleophiles if needed, giving access to more desymmetrised monomers for the air/water interface approach.
1 Figure 3.7. H-NMR spectrum of compound 16b in CDCl3.
The monomer anthraphane-tri(DEGME) was then assembled in the final copper-free Sonogashira reaction step similarly to 2a (Scheme 3.7). The slightly higher yield of 42% could be attributed to an increased solubility of the intermediates during the reaction. The molecule is not soluble in methanol and could be therefore isolated with the same precipitation workup employed for 1 and 2a. Subsequent recrystallisation from hot THF afforded yellow needles (see experimental part in subchapter 3.9.1 for details).
154
AMPHIPHILIC ANTHRAPHANES
Scheme 3.7. Synthesis of the anthraphane-tri(DEGME) monomer 2b.
The structure of anthraphane-tri(DEGME) was confirmed by 1H-NMR spectroscopy as shown in Figure 3.8 and 13C-NMR spectroscopy. As mentioned, single crystals of 2b could be grown from THF, however due to the high flexibility of the DEGME chains, only a disordered and poorly resolved crystal structure was obtained. The structure could however be sufficiently resolved to see that the monomer packed in the etf packing 1, with the DEGME chains folding back towards the anthracene units and no solvent molecules included in the structure (for details see the Appendix section).
1 Figure 3.8. H-NMR spectrum of anthraphane-tri(DEGME) 2b in CDCl3.
155
AMPHIPHILIC ANTHRAPHANES
3.5 Spreading of 2b at the Air/Water Interface
A dilute solution of anthraphane-tri(DEGME) 2b in chloroform was spread at the air/water interface and a SP vs MMA isotherm was recorded. This time a MMA of 215 Å2 per molecule was estimated, which resulted more consistent with the structure of 2b (see Figure 3.9). A more detailed discussion into the observed MMA values and possible packings of 2b at the interface will be presented in subchapter 3.7.2. The isotherm displayed good reproducibility and negligible hysteresis upon 5 cycles of compression/expansion, indicating good stability of the film (see Figure 3.10)†††.
Figure 3.9. SP vs MMA isotherm of anthraphane 2b at the air/water interface. At a surface pressure of 20 mN/m, the estimated MMA value is 215 A2.
††† MMA values are subjected to variations due to various reasons, such as concentration of the stock solution, cleanliness of the water subphase, and incorporation of impurities in the monomer solution. 156
AMPHIPHILIC ANTHRAPHANES
Figure 3.10. Hysteresis of 2b during 5 cycles of compression/expansion between 20 mN/m and 10mN/m.
Visual inspection of the film by BAM showed that the monomer forms islands, which upon further compression merge into a homogeneous lateral film with homogeneous thickness as shown by the monotonous grey contrast on the micrographs in Figure 3.11. Already at low surface pressures, such as 3 mN/m, only minor ripples were present, while at surface pressures above 10 mN/m the film looked macroscopically completely homogeneous.
Figure 3.11. BAM micrographs recorded at different surface pressures during the compression process. At low SP, islands are clearly visible, which upon compression coalesce into a lateral homogeneous film with homogeneous thickness.
157
AMPHIPHILIC ANTHRAPHANES
By comparing the results observed for 2b with the ones of monomer 2a, the importance of the amphiphilicity of a molecule becomes evident for a successful spreading at the air/water interface. In particular, it is important to have a versatile structure, whose amphiphilicity can be easily tuned if needed. The three polyether chains of 2b are expected to stabilise the orientation of the monomer at the air/water interface, behaving similarly to the aquatic organism Physalia physalis, better known as Blue Bottle jellyfish or Portuguese Man o’ War. As a matter of fact 2b and P. physalis are structurally similar as shown in Figure 3.12: the animal consists of a gas-filled bladder that allows it to float on the water surface, and from which, numerous venomous tentacles extend into the water[174]. Analogously, the main hydrocarbon body (bladder) of 2b is expected to float on the water surface, with the anthracene units exposed to the air, while the extending DEGME chains (tentacles) are expected to be immersed into the water subphase to maximise hydrogen bonding.
Figure 3.12. Structural similarities between 2b and the aquatic organism P. physalis.
To confirm the monolayer nature of the monomer film, height analysis by tapping mode AFM was performed on a monomer film transferred horizontally on a 285 nm SiO2-coated silicon substrate at 20 mN/m. Height profile analysis gave values of hAFM ≈ 0.9 - 1.0 nm which are in good agreement with the expected calculated value hCALC. for the monomer standing with the anthracenes perpendicular to the substrate (see Figure 3.13). Other orientations of the monomer on the substrate can be excluded as the expected hAFM would be much higher. The conformation of the polyether chains on the substrate is not known but it is reasonable to expect that they are flattened out on the silicon oxide surface, thus not contributing much to the overall thickness of the film. 158
AMPHIPHILIC ANTHRAPHANES
Figure 3.13. Tapping mode AFM height image and height profiles of the 2b monomer film transferred onto a silicon substrate. The thickness of the film is approximately 1.0 nm, which is in good agreement with the calculated height hcalc for the monomer standing with the anthracenes perpendicular to the substrate. Other orientations would give higher values of hAFM. DEGME chains are omitted but are supposed to flatten out on the hydrophilic surface of silicon oxide.
This AFM analysis confirmed the monolayer nature of the film and gave some insights into the orientation of the monomer at the interface, which seemed to correspond to the one required for lateral polymerisation to occur (i.e. anthracenes standing perpendicular to the interface). The next subchapter will be dealing with the photochemical environment of 2b at the interface and its photo-induced polymerisation.
3.6 Photochemical Environment of Anthraphane-tri(DEGME) at the Air/Water Interface
A striking characteristic of the anthraphane monomers is their exceptionally intense fluorescence (both in solution and solid/crystalline state) upon excitation with wavelengths such as λ = 254 nm. A monolayer of 2b can in fact be visualised by naked eye directly at the air/water interface by using a
159
AMPHIPHILIC ANTHRAPHANES standard laboratory UV lamp, giving an even more macroscopic view of the film with respect to BAM, but with enough detail to distinguish islands, homogeneity and even occasional multilayer formation (Figure 3.14). In particular the shape of the islands is quite irregular compared to the line tension- governed circular shape that one observes for instance with oil droplets floating on the surface of water. The irregular shape is the result of particular interactions between 2b and the subphase, but
Figure 3.14. Anthraphane 2b after being spread at the air/water interface (top) and after being compressed at 20 mN/m (bottom). The monolayer can be visualised by naked eye upon excitation at 254 nm with a standard UV lamp. 160
AMPHIPHILIC ANTHRAPHANES also intermolecular interactions between the monomers. This phenomenon could be a hint of crystallisation of 2b taking place on the surface of water through particularly strong anthracene- anthracene interactions. Fluorescence is not observed outside the barriers, which exclude diffusion of the monomer in the subphase.
The strong fluorescence emission could not be observed at the air/water interface with other monomers such as 29 used in previous studies aiming at 2DPs (Figure 3.15). This phenomenon could be related to fluorescence anisotropy due to the different transition dipole moment (S0 S1) orientation of the anthracene units with respect to the interface: it is possible that when the transition dipole moment (blue in Figure 3.15) on the short axis of the anthracenes is oriented parallel to the interface, as in the case of 2b, the emission will be enhanced compared to anthracenes whose short axis is perpendicular to the interface, such as in 29.
Figure 3.15. Comparison of the orientations at the air/water interface of the anthracenes units of anthraphane 2b and monomer 29. In the case of 2b the short axis of the anthracenes (blue arrow) is standing perpendicular to the interface whereas in 29 it is parallel. This orientation can have an impact into the strength of the emission.
For photodimerisation between the anthracene units to occur, which would then lead to photopolymerisation, there are two necessary conditions: a) the wavelength chosen for irradiation must be absorbed by the anthracene units and b) the anthracene units must be in close proximity and must form upon excitation an attractive pair in the form of an excimer (excited dimer). If these conditions are met the excimer can decay into an anthracene dimer upon [4+4]-cycloadditon. According to the UV/Vis absorption spectrum of the monomer monolayer (Figure 3.16), a wavelength of λ = 365 nm was selected for irradiation, a choice dictated by the availability of our LED equipment.
161
AMPHIPHILIC ANTHRAPHANES
Figure 3.16. UV/Vis absorption spectrum of the monomer monolayer transferred onto a quartz substrate.
Compared to the emission spectra of isolated anthracenes which exhibit a typical vibronic structure, excimers have instead a typical emission signature which is broad, unstructured and red- shifted[175]. Therefore, to see whether the anthracene units of 2b form excimers upon compression at the air/water interface, a fluorescence spectrum was measured in-situ at 20 mN/m using λ = 365 nm as excitation wavelength. The results are displayed in Figure 3.17: at the interface the emission consists of a broad peak centered at around 518 nm, considerably red-shifted compared to the emission spectrum of 2b measured in solution. The interfacial spectrum also lacks the vibronic structure typical for isolated anthracenes and could be therefore tentatively considered as pure excimer emission. It is however possible that isolated anthracenes may be present but not detected due to energy transfer phenomena: the emission of an isolated anthracene could be absorbed by an anthracene pair, which would then emit it as excimer emission, effectively disguising the emission of the isolated anthracene. While this scenario could be in principle possible, it would imply a very efficient energy transfer, as usually in a mixed photochemical environment, the emission of isolated anthracenes and excimers can be detected simultaneously[176,177]. This point will be addressed again in the next subchapter, where fluorescence decay experiments will be discussed.
162
AMPHIPHILIC ANTHRAPHANES
Figure 3.17. In-situ fluorescence emission spectrum of 2b at the air/water interface compared to a spectrum in a dilute chloroform solution (λ = 365 nm excitation wavelength). While in solution the typical vibronic structure of the isolated anthracene units is clearly distinguishable, at the air/water interface there is only pure excimer emission centered at 518 nm and no sign of isolated anthracenes. The spectrum at the interface was background-corrected to eliminate contributions from the LED, trough, subphase and its Raman scattering.
3.6.1 Excimers Fluorescence Decay
The presence of excimers in the monomer monolayer offers the opportunity to convert them into net points upon decaying into anthracene dimers through [4+4]-cycloaddition, which do not exhibit fluorescence anymore. Assuming the photodimerisation goes through a singlet excited state
[83,84]‡‡‡ S1 , the decay of the excimer into a photodimer is expected to follow first-order kinetics. An in- situ fluorescence decay experiment was therefore conducted by monitoring the decay of the excimer population during the irradiation time. The irradiation was performed at ambient conditions, by using λ = 365 nm irradiation wavelength. The same wavelength was also used as excitation when
‡‡‡ The photodimerisation of anthracenes is quenched by heavy atoms solvents, suggesting that the process does not go through a triplet state. The heavy atom is thought to quench dimerisation by intersystem crossing. 163
AMPHIPHILIC ANTHRAPHANES recording an emission spectrum. This experiment was also meant to give a feeling about the optimal irradiation time to achieve full conversion of the anthracene pairs. The results are summarised in Figure 3.18. As a remark, the fluorescence spectra measured at the interface were background- corrected to eliminate the contributions from the LED, the Langmuir trough, the water subphase and its Raman scattering.
Figure 3.18. In-situ fluorescence decay experiment of a compressed monolayer of 2b at the air/water interface under ambient conditions (λ = 365 nm excitation and irradiation wavelength) compared to a dilute solution emission spectrum of 2b in chloroform. The in-situ fluorescence spectra are baseline corrected.
After approximately 4 h, no residual intensity from the excimer emission was visible, indicating that the excited pairs were converted into anthracene dimers. More importantly, as the population of dimers decayed, no emission from unpaired anthracenes developed suggesting that energy transfer from isolated anthracenes to paired anthracenes is not a likely scenario. A possible reason for the absence of fluorescence of unpaired anthracenes could be due to their oxidation with atmospheric oxygen: endoperoxide formation by photooxidation of anthracenes is a well-known phenomenon[83,84,178–180] in dilute aerated solutions (< 1mM), but is not considered to be relevant at the air/water interface, where anthracene concentration is very high. Endoperoxide formation would simply destroy the anthracene units, rendering them unable to contribute to the formation of net points. If this reaction would be significantly competitive with the anthracene dimerisation, 164
AMPHIPHILIC ANTHRAPHANES considerable defects would be formed in the molecular network, having detrimental effects for its mechanical stability. This point will be discussed in more detail in subchapter 3.8. The decay of the excimers followed well-behaved first order kinetics as displayed in Figure 3.19 and had a half-life of t1/2 = 16.5 min.
Figure 3.19. Exponential decay of the excimer fluorescence upon irradiation at 365 nm. The decay follows first order kinetics and has a half-life of 16.5 min.
Consumption of the anthracenes also found confirmation visually as displayed in Figure 3.20: by irradiating with an LED a circular spot of monolayer (approx. 2 cm in diameter) for a few hours, the irradiated region did not display any fluorescence anymore upon inspection with a UV lamp. The presence of excimers offers the possibility of bond formation between the monomers presumably forming a two-dimensional covalent monolayer sheet; it however does not give any information about the internal structure and eventual periodicity of the polymer. In the next subchapter the possible packings of anthraphane-tri(DEGME) 2b at the air/water interface will be discussed with the experimental information gathered so far.
165
AMPHIPHILIC ANTHRAPHANES
Figure 3.20. Irradiation experiment on the monomer monolayer. In the irradiated spot, the fluorescent anthracene pairs get converted into non-fluorescent anthracene dimers; as a result by examining the monolayer with a UV lamp, the irradiated spot does not exhibit fluorescence anymore.
3.6.2 Packing at the Air/Water Interface
The study of the photochemical environment of 2b at the air/water interface can now be combined with the results obtained from the isotherm and MMA values in subchapter 3.6 to speculate about the packing of the monomer upon compression at the interface and the resulting internal structure of the polymer sheet after irradiation. Different packing scenarios and their respective calculated MMA values can be conveniently extracted from the studies made on the crystallisation of anthraphane 1 in Chapter 2 of this work (Figure 3.21, for details on the calculations, see subchapter 3.9.2).
Figure 3.21. Some of the likely packing scenarios of 2b at the air/water interface with their respective calculated MMA values. The packings are taken from the crystal structures of anthraphane 1 discussed in Chapter 2.
166
AMPHIPHILIC ANTHRAPHANES
For 2b an MMA value of 215 Å2 per molecule was estimated from its isotherm at a compression of 20 mN/m. This value is significant, as it excludes a dense packing at the interface which would correspond to the etf/ftf packing 1, with an MMA value of 138 Å2. The featureless isotherm without any noticeable phase transitions also seems to exclude a collapse of the monomer packing upon compression. At this point it is not easy to discriminate between an etf packing 1 or an ftf packing based on the similar MMA values, but according to the irradiation experiments performed in the previous subchapter, it would seem that isolated anthracenes are not present in the packing, neither before nor after irradiation. This would then exclude the etf packing 1 from the picture, where poor alignment of the π-orbitals of the anthracenes units would probably not result in any excimer formation due to their etf geometry. If by any chance some dimerisation would nevertheless occur, it would inevitably leave behind isolated anthracenes.
However this discussion has to be considered carefully, as it is based only on a combination of indirect and independent observations; without solid proof of the internal structure of the polymer by for instance STM, cryo TEM electron diffraction, high-resolution AFM or GISAXS, these assumptions only remain speculations. For the same reasons, the obtained polymer will be referred to as two-dimensional covalent monolayer sheet rather than 2DP. At this point however, assuming that an all-ftf packing was achieved, a natural question arises: what is the driving force that realises this packing at the air/water interface but not in the single crystals? A tentative answer can be found in the hydrophobic effect, combined with stacking interactions between anthracenes[181]. It can be speculated that when at the air/water interface, the anthracene units of 2b will be submerged to some extend into the subphase; the hydrophobic faces of the anthracene units will therefore try to minimise contact with water molecules and maximise contact with each other by stacking ftf (Figure 3.22).
167
AMPHIPHILIC ANTHRAPHANES
Figure 3.22. Possible orientation of 2b at the air/water interface (DEGME chains omitted for clarity). It is speculated that the anthracene units will be submerged to a certain degree into the water subphase (top); to minimise contact with the water molecules, the hydrophobic anthracene units will try to maximise contact with each other by stacking ftf (bottom). This ftf-inducing hydrophobic effect could be the reason for having an all-ftf packing at the air/water interface.
3.7 Polymer Sheet
As mentioned previously, formation of anthracene dimers in the monomer monolayer corresponds to the formation of net points in the polymer, which, if present above the rigidity percolation threshold will result into a mechanically coherent sheet. To qualitatively test the mechanical stability of the polymer sheet, it was horizontally transferred onto TEM grids with 20 μm × 20 μm holes and imaged by SEM. As a control experiment, the non-irradiated monomer monolayer was transferred and imaged in the same way (Figure 3.23).
Figure 3.23. Horizontal transfer onto TEM grids of the monomer monolayer and polymer.
168
AMPHIPHILIC ANTHRAPHANES
A self-sustaining network is expected to span the holes without collapsing under its own weight, while the molecules in the monomer monolayer, lacking in-plane bond formation, are expected to simply pass through the holes. SEM analysis confirmed this assumption as can be seen in Figure 3.24: the polymer film spanned well over the holes, with occasional ruptures attributed to the transfer procedure, while in the monomer case, no hole could be spanned. Each hole can potentially hold an impressive number of 2.04 x 108 repeat units (area divided by MMA).
This experiment supports that irradiation and subsequent anthracene dimerisation is the cause of the observed mechanical stability. While quantification of the conversion of the polymerisation cannot be directly calculated, it can be estimated to be above 65%, which is the bond percolation threshold for a hexagonal 63 honeycomb lattice, above which mechanical coherence is developed[182]. As discussed before, anthracene photooxidation cannot be completely excluded, but formation of anthracene peroxides (which subsequently decay into anthraquinones) would involve formation of defects in the network which would impact negatively its mechanical stability. As such, it can be concluded that photooxidation is not a major competitive reaction to the anthracene dimerisation.
Figure 3.24. SEM images or irradiated and non-irradiated monolayer transferred onto TEM grids with holes of 20 μm × 20 μm.
169
AMPHIPHILIC ANTHRAPHANES
Finally, the monolayer nature of the polymer was again confirmed by AFM height analysis after transferring the film on a silicon wafer (Figure 3.25). The film thickness was estimated to 1.1 nm, which is in excellent agreement with the molecular dimensions.
Figure 3.25. Tapping mode AFM height image and height profiles of the 2b polymer film transferred onto a silicon substrate. The thickness of the film is approximately 1.1 nm.
3.8 Conclusion and Outlook
In conclusion we have shown that the robust synthesis developed for anthraphane 1 can be modified in order to synthesise amphiphilic anthraphanes for the synthesis of covalent monolayer sheets at the air/water interface. Anthraphane-tri(OMe) 2a and anthraphane-tri(DEGME) 2b were successfully synthesised and spread at the air/water interface. The greater amphiphilicity of 2b proved to be essential for the formation of highly fluorescent, homogeneous, monomer monolayers. Irradiation of the monomer film resulted into a mechanically coherent two-dimensional molecular network, able to span over micrometer-sized holes without collapsing under its own weight. Mechanical stability is attributed to a high number of net points, which correspond to dimerised anthracenes. Combined molecular packing models and fluorescence spectroscopy tentatively suggest the formation of an all- ftf packing at the air/water interface, which would eventually result into a polymer with a hexagonal honeycomb 63 lattice. The conversion of the anthracene dimerisation is estimated to be above 65%, which is the expected bond percolation value for this kind of lattices, above which mechanical coherence is developed.
While the polymer obtained from 2b satisfies some criteria for the classification as 2DP (composed by covalently bound anthraphane repeat units, free-standing and one monomer unit thick molecular sheet), its internal structure still has to be elucidated. In this regard, the structural rigidity of the monomer is expected to result into a structurally rigid and shape-persistent polymer, whose periodical structure could be then characterised by x-ray techniques (GISAXS), electron
170
AMPHIPHILIC ANTHRAPHANES diffraction or STM. Structural rigidity is of particular importance: a shape-persistent anthracene- based monomer was in fact successfully employed for the synthesis of a 2DP at the air/water interface[41] and its internal structure confirmed by STM. For another trifunctional anthracene-based monomer however, due to its structural flexibility, internal order could not be detected[57]: it was shown by DFT simulations that after photopolymerisation the presumably regular structure relaxed and collapsed into an amorphous network with significant lateral disorder. Regarding the photoinduced [4+4]-cycloaddition dimerisation reaction, it would be desirable to have direct proof of bond formation by spectroscopical methods such as IR spectroscopy: ideally the C-H out-of-plane bend vibrational mode at around 897 cm-1 typical for anthracene would disappear upon dimerisation and be substituted by the appearance of the C-H bend at around 754 cm-1 typical for dimers. Similarly, a characteristic carbonyl stretch arising from the photooxidation of anthracene to anthraquinone would shed light in a possible detrimental role of oxygen during the polymerisation.
As a final remark, we wish to point out the advantage of synthesising a 2DP at the air/water interface, namely the achievable lateral extension of the sheet, which is virtually limited by size of the irradiated area. In the single crystal approach instead, the size of the molecular sheet is in the ideal case maximally limited to the dimensions of the crystal itself. Controlled transfer onto substrates, as well as mechanical stability of the sheet and its intrinsic porous nature potentially also open the possibility of using polymers based on 2b as ultra-thin membranes for gas permeation and separation[78,79].
The quenching of the fluorescence of the monolayer upon polymerisation as seen in Figure 3.20 and its potential recovery upon thermally induced back-reaction could also open the possibility of using monolayers of 2b as rewritable optical data storage material systems[180,183].
171
AMPHIPHILIC ANTHRAPHANES
3.9 Experimental
3.9.1 Synthesis
Materials and methods
All reactions were carried out under nitrogen by using standard Schlenk techniques and dry solvents unless otherwise noted. Dry diethyl ether, dry methanol, dry DMI and dry toluene were purchased from Acros and used directly. For large amounts of dry toluene, it was distilled by a solvent drying system from LC Technology Solutions Inc. SP-105 under nitrogen atmosphere (H2O content < 5 ppm as determined by Karl-Fischer titration). Diisopropyl amine was dried by passing it over a column of activated neutral aluminum oxide (Brockmann Activity I, Sigma-Aldrich) according to the
[184] [153] literature . Pd(PPh3)4 catalyst was freshly prepared following the literature procedure and stored in a glove-box in the dark under N2 at room temperature. All reagents were purchased from Acros, Aldrich or TCI, and used without further purification. Column chromatography for purification of the products was performed by using Merck silica gel Si60 (particle size 40-63 μm).
NMR was recorded on a Bruker AVANCE (1H: 300 MHz, 13C: 75 MHz) at room temperature. The signal from the solvents was used as internal standard for chemical shift (1H: δ = 7.26 ppm, 13C: δ = 77.16 ppm for chloroform, 1H: δ = 6.00 ppm, 13C: δ = 73.78 ppm for 1,1,2,2-tetrachloroethane, 19F: δ = - 164.9 ppm for hexafluorobenzene). When possible, proton and carbon signal assignments were performed with the help of 2D-NMR experiments such as COSY, HSQC and HMBC (spectra not shown).
High resolution mass spectroscopy (HRMS) analyses were performed by the MS-service of the Laboratory for Organic Chemistry at ETH Zurich with spectrometers (ESI- and MALDI-ICR-FTMS: IonSpec Ultima Instrument). Either 3-hydroxypicolinic acid (3-HPA) or trans-2-[3-(4-tert-butylphenyl)- 2-methyl-2-propenylidene]malononitrile (DCTB) were used as matrix.
UV/Vis absorption spectra were recorded with a JASCO V-670 UV-Vis-NIR spectrophotometer using a quartz cell with a path length of 1 cm. Emission spectra were recorded with a Spex Fluorolog 2 spectrophotometer from Jobin Yvon (United Kingdom) using a quartz cell with a path length of 1 cm by diluting by a factor of 30-60 (depending on the compound) the solutions employed for the UV/Vis absorption measurements.
172
AMPHIPHILIC ANTHRAPHANES
Synthetic procedures
1,3,5-Trifluoro-2,4,6-triiodobenzene 13 Periodic acid (5.13 g, 22.5 mmol, 1.50 eq) was suspended in 35 mL sulphuric acid at 0°C. Finely ground potassium iodide (11.2 g, 67.6 mmol, 4.50 eq) was added in portions over 5 min during which time iodine vapours evolved. 1,3,5-Trifluorobenzene (1.56 mL, 15.1 mmol, 1.00 eq) was then added by syringe over 5 min at 0°C and the reaction mixture was then heated to 70°C for 5 h, during which time additional sulphuric acid can be added for better stirring. After cooling to room temperature, the reaction mixture was poured into 350 g crushed ice. 200 mL diethyl ether were added and the layers separated. The organic layer was washed once with 150 mL of a 15% solution of sodium thiosulfate, once with 150 mL water, dried over MgSO4 and concentrated to dryness. The residue was purified by sublimation: the temperature was kept at 50°C (p = 0.04 mbar) for 1 h, during which time a brown layer coated the cold finger, then it was increased to 110°C to allow sublimation of the product. Compound 13 was obtained as a white crystalline solid (7.60 g, 14.9 mmol, 97%). Rf (hexane): 0.6, Mp: 156-158 °C.
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 162.4 (dt, J = 243.5, 7.8 Hz), 63.9 (ddd, J = 34.9, 34.1, 3.9 Hz).
19 F-NMR (282.5 MHz, CDCl3) δ/ppm: 68.83.
+ HRMS (FT-MALDI): m/z calcd for C14H13O2 [M-H] : 213.0910; found: 213.0909.
173
AMPHIPHILIC ANTHRAPHANES
1,3,5-Triiodo-2,4,6-trimethoxybenzene 14a Sodium hydride (0.94 g, 23.5 mmol, 6 eq, 60% dispersion in mineral oil) was suspended in 10 mL dry diethyl ether in a 100 mL Schlenk flask. MeOH (1.00 mL, 23.5 mmol, 6 eq) diluted in 10 mL dry diethyl ether was slowly added to the hydride under vigorous stirring. The reaction mixture was then stirred at room temperature for 1 h until no more hydrogen evolution was detectable. The volatiles were removed by a stream of nitrogen and the residue was dried on HV for 10 min. 20 mL dry DMI were then added to the residue and to the resulting suspension 1,3,5-trifluoro-2,4,6-triiodobenzene 13 (2.00 g, 3.92 mmol, 1 eq) was added in small portions over 30 min under vigorous stirring (caution: some foaming during this exothermic reaction is possible). The obtained orange suspension was stirred overnight at room temperature and then poured into 80 mL saturated NaHCO3 solution. The white precipitate was collected by filtration and washed with water until the pH of the filtrate resulted neutral. Recrystallisation from boiling methanol afforded compound 14a as white needles
(1.56 g, 2.86 mmol, 73%). Rf (30% EtOAc in hexane): 0.72, Mp: 176-178°C.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 3.86 (s, 9H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 161.5, 82.8, 60.9.
+ HRMS (FT-MALDI): m/z calcd for C9H9I3O3 [M-H] : 545.7680; found: 545.7680.
174
AMPHIPHILIC ANTHRAPHANES
((2,4,6-Trimethoxybenzene-1,3,5-triyl)tris(ethyne-2,1-diyl))tris(trimethylsilane) 15a 1,3,5-triiodo-2,4,6-trimethoxybenzene 14a (180 mg, 0.33 mmol, 1.00 eq) was placed in a dry 20 mL
Schlenk tube along with catalyst Pd(OAc)2 (4.00 mg, 0.02 mmol, 0.05 eq), co-catalyst CuI (3.00 mg, 0.02 mmol, 0.05 eq) and ligand tri-tert-butylphosphonium tetrafluoroborate (9.00 mg, 0.03 mmol, 0.10 eq). In a separate Schlenk tube, 10 mL dry diisopropylamine were degassed by four cycles of freeze-pump-thaw and then added by syringe to the reactants. Trimethylsilylacetylene (0.20 mL, 1.65 mmol, 5 eq) was added and the reaction mixture was sealed and heated to 65°C for 48 h under an argon atmosphere. Formation of a beige precipitate indicated the start of the reaction. After cooling to room temperature, the reaction mixture was filtered through a celite pad and concentrated to dryness. Separation by flash column chromatography (3% EtOAc in hexane) afforded the title compound 15a as a yellowish solid of purities high enough for the next step (98 mg, 0.21 mmol,
65%). Rf: 0.37, Mp: 82-84°C.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.02 (s, 9H), 0.24 (s, 27H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 163.5, 108.3, 103.5, 96.1, 59.3, 0.0.
+ HRMS (FT-MALDI): m/z calcd for C24H36O3Si3 [M] : 456.1972; found: 456.1970.
175
AMPHIPHILIC ANTHRAPHANES
1,3,5-Triethynyl-2,4,6-trimethoxybenzene 16a Compound 5a (0.20 g, 0.44 mmol, 1 eq) was dissolved in a solvent mixture of 5 mL THF and 2 mL
MeOH. Potassium carbonate (7.00 mg, 0.05 mmol, 0.12 eq) and 0.5 mL H2O were added and the reaction mixture was stirred at room temperature for 16 h. After removal of the solvents in vacuo, 5 mL H2O were added to the residue, which was then extracted with DCM (3 x 15 mL). The combined organic phases were dried over MgSO4, concentrated and subjected to flash column chromatography (10% EtOAc in hexane) to afford the title compound 16a as a white solid (97 mg, 0.40 mmol, 92%). The product must be stored in the fridge protected from light and under nitrogen (decomposition is characterised by a blue-coloration and insolubility in organic solvents). Rf: 0.24 , Mp: 125-127°C.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.06 (s, 9H), 3.46 (s, 3H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 165.7, 106.8, 85.8, 74.8, 61.7.
+ HRMS (FT-MALDI): m/z calcd for C15H13O3 [M-H] : 241.0859; found: 241.0870.
176
AMPHIPHILIC ANTHRAPHANES
Anthraphane-tri(OMe) 2a Precursor 11 (320 mg, 0.28 mmol, 1.00 eq) was suspended in 283 mL dry toluene (1.00 mM) with compound 16a (68.0 mg, 0.28 mmol, 1.00 eq) and dry triethylamine (7.90 mL, 56.6 mmol, 200 eq). The reaction mixture was degassed by cooling it to -80°C with an acetone-dry ice bath and then performing five cycles of vacuum (10 min) and nitrogen backfilling. Pd(PPh3)4 (98.0 mg, 0.08 mmol,
0.30 eq) was added with N2 counter-flow and the suspension was degassed twice again and backfilled with argon after the last cycle. After warming to room temperature, the reaction mixture was put in a preheated bath at 80°C and stirred in the dark under argon for 5 d. After cooling to room temperature, the reaction mixture was filtered and the filtrate was concentrated to dryness. The brownish residue was washed with MeOH to obtain a beige solid, which was collected by filtration and rinsed with more MeOH until the filtrate resulted colourless. The beige solid was then dissolved in tetrachloroethane and slowly precipitated with MeOH to obtain the pure product 2a as a pale yellow solid (92.0 mg, 0.10 mmol, 36%). The product can be recrystallised from boiling THF to obtain yellow needles. Rf (20%EtOAc in hexane): 0.27 (blue fluorescence with λ = 366 nm), Mp: decomposes above 280°C.
1 H-NMR (300 MHz, CD2Cl4) δ/ppm: 9.54 (s, 3H), 8.56 (s, 3H), 8.11 (d, J = 8.6 Hz, 6H), 7.90-7.80 (m, 6H), 7.77 (s, 3H), 7.60-7.50 (m, 6H), 4.14 (s, 9H).
13 C-NMR (75.5 MHz, CD2Cl4) δ/ppm: 164.2, 134.1, 131.7, 131.44, 131.41, 131.3, 130.2, 130.1, 129.3, 129.0, 127.3, 125.3, 125.1, 123.9, 123.8, 121.3, 120.6, 107.4, 94.6, 92.1, 88.3, 85.4, 62.0.
+ HRMS (FT-MALDI): m/z calcd for C69H36O3 [M] : 912.2659; found: 912.2657.
177
AMPHIPHILIC ANTHRAPHANES
1,3,5-Triiodo-2,4,6-tris(2-(2-methoxyethoxy)ethoxy)benzene 14b Sodium hydride (0.94 g, 23.5 mmol, 6 eq, 60% dispersion in mineral oil) was suspended in 8 mL dry diethyl ether in a 100 mL Schlenk flask. Dry 2-(2-methoxyethoxy)ethanol (2.77 mL, 23.5 mmol, 6 eq) diluted in 8 mL dry diethyl ether was slowly added to the hydride under vigorous stirring. The reaction mixture was then stirred at room temperature for 1 h until no more hydrogen evolution was detectable. The volatiles were removed by a stream of nitrogen and the residue was dried on HV for 20 min. 15 mL dry DMI were then added to the jelly-like residue and to the resulting suspension, 1,3,5-trifluoro-2,4,6-triiodobenzene 13 (2.00 g, 3.92 mmol, 1 eq) was added in small portions over 30 min under vigorous stirring. The obtained orange suspension was stirred overnight at room temperature and then poured into 80 mL saturated NaHCO3 solution. The white precipitate was collected by filtration and washed with water until the pH of the filtrate resulted neutral. Purification by flash column chromatography (70% EtOAc in hexane) afforded compound 14b as a white waxy solid (2.89 g, 3.57 mmol, 91%). Rf: 0.20, Mp: 57-60°C.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.17 (t, J = 5.2 Hz, 6H), 4.00 (t, J = 5.2 Hz, 6H), 3.85 – 3.75 (m, 6H), 3.65 – 3.56 (m, 6H), 3.41 (s, 9H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 160.3, 83.5, 72.4, 72.2, 71.0, 70.1, 59.3.
+ HRMS (FT-MALDI): m/z calcd for C21H33I3O9 [M] : 809.9253; found: 809.9253.
178
AMPHIPHILIC ANTHRAPHANES
((2,4,6-Tris(2-(2-methoxyethoxy)ethoxy)benzene-1,3,5-triyl)tris(ethyne-2,1- diyl))tris(trimethylsilane) 15b Compound 14b (1.00 g, 1.23 mmol, 1.00 eq) was placed in a dry 50 mL Schlenk tube along with catalyst Pd(OAc)2 (14.0 mg, 0.06 mmol, 0.05 eq), co-catalyst CuI (11.0 mg, 0.06 mmol, 0.05 eq) and ligand tri-tert-butylphosphonium tetrafluoroborate (36.0 mg, 0.12 mmol, 0.10 eq). In a separate Schlenk tube, 15 mL dry diisopropylamine were degassed by four cycles of freeze-pump-thaw and then added by syringe to the reactants. Trimethylsilylacetylene (0.90 mL, 6.17 mmol, 5 eq) was added and the reaction mixture was sealed and heated to 65°C for 48 h under an argon atmosphere. Formation of a beige precipitate indicated the start of the reaction. After cooling to room temperature, the reaction mixture was filtered through a celite pad and concentrated to dryness. Separation by column chromatography (30% EtOAc in hexane) afforded the title compound as yellow oil (641 mg, 0.89 mmol, 72%). Rf: 0.36.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.37 (t, J = 5.3 Hz, 6H), 3.85 (t, J = 5.3 Hz, 6H), 3.79 – 3.66 (m, 6H), 3.60 – 3.49 (m, 6H), 3.37 (s, 9H), 0.23 (s, 27H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 163.9, 108.0, 103.4, 96.3, 73.4, 72.1, 70.83, 70.76, 59.2, 0.0.
+ HRMS (FT-MALDI): m/z calcd for C36H60O9Si3 [M] : 720.3540; found: 720.3539.
179
AMPHIPHILIC ANTHRAPHANES
1,3,5-triethynyl-2,4,6-tris(2-(2-methoxyethoxy)ethoxy)benzene 16b Compound 15b (0.54 g, 0.75 mmol, 1 eq) was dissolved in a solvent mixture of 5 mL THF and 2 mL
MeOH. Potassium carbonate (13.0 mg, 0.09 mmol, 0.12 eq) and 0.5 mL H2O were added and the reaction mixture was stirred at room temperature for 16 h. 10 mL H2O were added to the reaction mixture and the organics were removed in vacuo. The aqueous residue was subsequently extracted with diethyl ether (3 x 20 mL) and the combined organic phases were dried over MgSO4 and concentrated to dryness. Purification by flash column chromatography (EtOAc) afforded the title compound 16b as a yellow oil (341 mg, 0.67 mmol, 90%). Rf: 0.36.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.41 (t, J = 5.1 Hz, 6H), 3.85 (t, J = 5.1 Hz, 6H), 3.77 – 3.67 (m, 6H), 3.59 – 3.50 (m, 6H), 3.43 (s, 3H), 3.38 (s, 9H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 164.5, 107.3, 86.1, 75.0, 73.5, 72.0, 70.7, 70.4, 59.1.
+ HRMS (FT-MALDI): m/z calcd for C27H36NaO9 [M-Na] : 527.2252; found: 527.2250.
180
AMPHIPHILIC ANTHRAPHANES
Anthraphane-tri(DEGME) 2b Precursor 11 (500 mg, 0.44 mmol, 1.00 eq) was suspended in 222 mL dry toluene (2.00 mM) with compound 16b (225 mg, 0.44 mmol, 1.00 eq) and dry triethylamine (12.4 mL, 89.0 mmol, 200 eq). The reaction mixture was degassed by cooling it to -80°C with an acetone-dry ice bath and then performing five cycles of vacuum (10 min) and nitrogen backfilling. Pd(PPh3)4 (154 mg, 0.13 mmol,
0.30 eq) was added with N2 counter-flow and the suspension was degassed twice again and backfilled with argon after the last cycle. After warming to room temperature, the reaction mixture was put in a preheated bath at 80°C and stirred in the dark under argon for 5 d. After cooling to room temperature, the reaction mixture was filtered and the filtrate was concentrated to dryness. The brownish residue was washed with MeOH to obtain a beige solid, which was collected by filtration and rinsed with more MeOH until the filtrate resulted colourless. The crude product was then purified by column chromatography (1% MeOH in DCM) and the obtained solid was recrystallized from boiling THF to obtain 2b as pale yellow needles (189 mg, 0.16 mmol, 36%). Rf: 0.67 (blue fluorescence with λ = 366 nm), Mp: decomposes above 280°C.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.49 (s, 3H), 8.50 (s, 3H), 8.06 (dd, J = 8.5, 3.9 Hz, 9H), 7.82-7.73 (m, 6H), 7.50 (dd, J = 8.4, 7.1 Hz, 6H), 4.54 (t, J = 5.1 Hz, 6H), 3.75 (t, J = 5.1 Hz, 6H), 3.26 (dd, J = 5.5, 4.1 Hz, 6H), 2.86 – 2.70 (m, 15H).
13 C-NMR (75.5 MHz, CD2Cl4) δ/ppm: 163.2, 134.1, 131.5, 131.25, 131.22, 131.7, 130.3, 130.2, 129.4, 129.2, 127.5, 125.3, 125.2, 123.7, 123.6, 120.8, 120.2, 107.9, 95.1, 91.9, 88.4, 85.4, 70.9, 69.8, 69.5, 58.2.
+ HRMS (FT-MALDI): m/z calcd for C81H60O9 [M] : 1176.4232; found: 1176.4224.
181
AMPHIPHILIC ANTHRAPHANES
3.9.2 Langmuir monolayers at the air/water interface
Langmuir trough and monolayer preparation A KSV 2000 System 2 Langmuir trough equipped with a platinum Wilhelmy plate and dipper was used for this study. The trough is made of Teflon® and the barriers are of hydrophilic Delrin®. As subphase, Millipore water was used. For cleaning, the trough was first rinsed with Millipore water followed by chloroform, ethanol and chloroform again by wiping it with dust-free paper to remove all traces of organic material. The barriers were cleaned and wiped with ethanol and Millipore water. The trough was then filled with Millipore water and its surface treated by vacuum-suction to remove residual particles of dirt (cleanliness of the water surface was checked by UV lamp for residual fluorescence and by Brewster’s angle microscopy). The Wilhelmy plate was rinsed with water and ethanol and then heated for a few seconds on a burner to incandescence.
For spreading of 2a, different stock solutions were prepared (0.5 mg/mL): pure chloroform, 1:1 chloroform/cyclohexane, 1:1 chloroform/THF and 1:1 chloroform/ethyl acetate; in none of the cases, homogeneous monolayers were formed. For spreading of 2b, a stock solution (0.25 mg/mL) of a 1:1 solvent mixture of chloroform/cyclohexane was prepared. The solution was spread dropwise at the interface with the use of an air-tight glass microsyringe. Typical applied volumes were 190 µL. The solution was stored in a glass vial in the dark at 4°C. Compression with the barriers was started after 30 min from the spreading to allow complete evaporation of chloroform. The compression rate was 3mm/min.
As a general remark, the values of MMA are prone to variations either due to the change in the concentration of the solutions kept over time or due to the varying types of impurities which might be incorporated into the solution.
Brewster’s angle microscopy To visualise the films by Brewster angle microscopy (BAM), a KSV MicroBAM operating by a 659 nm laser was used. To visualise the monolayers by naked eye, a standard laboratory UV lamp for TLC (Camag) was used at 254 nm.
AFM height analysis of the monolayers and film transfer Height analysis was performed by using a Nanoscope III Multimode system (Digital Instruments, Santa Barbara, Ca). OMCLAC160TS silicon tips (Olympus, Tokyo, Japan) were used for imaging with a resonance frequency in between 200 and 400 kHz and a spring constant of about 42 N/m. The monolayers were transferred at a constant surface pressure of 20 mN/m onto 285 nm-SiO2 coated silicon wafers, using a modified Langmuir-Schaefer-technique: a tilted stage with the substrate was attached to the dipper and immersed just below the water surface; after compressing and reaching
182
AMPHIPHILIC ANTHRAPHANES the target surface pressure, the stage was slowly pulled out of the water (transfer rate 0.5 mm/min) and the substrate left to dry overnight. For details on the transfer procedure see Figure 3.26. For visualisation of the monolayer on the substrate by optical microscopy, a Leica DM 4000M microscope was used in differential interference contrast (DIC) mode (see Figure 3.27).
Figure 3.26. Modified Langmuir-Schaefer technique for the horizontal transfer of monolayers onto 285 nm-SiO2 coated silicon wafers.
Figure 3.27. Optical micrographs in differential interference contrast (DIC) mode of a) monomer monolayer and b) polymer monolayer transferred onto 285 nm-SiO2 coated silicon wafers. The monolayers appear as dark purple. In the monomer monolayer occasional ruptures can be seen.
UV/Vis absorption and emission spectra in solution UV/Vis absorption spectra were recorded with a JASCO V-670 UV-Vis-NIR spectrophotometer using a quartz cell with a path length of 1 cm. Emission spectra were recorded with a Spex Fluorolog 2 spectrophotometer from Jobin Yvon (United Kingdom) using a quartz cell with a path length of 1 cm by diluting by a factor of 30-60 (depending on the compound) the solutions employed for the UV/Vis absorption measurements.
183
AMPHIPHILIC ANTHRAPHANES
In-situ fluorescence spectroscopy at the interface For recording the spectra from the interface, an Acton series spectrometer from Princeton Instruments (NJ, USA) was used. The spectrometer was equipped with an SP-2556 Acton Research 500 mm Imaging Spectrograph (Acton, MA, USA), FC fiber stage (modified in house to be 3 mm closer to the reflecting mirror) and three standard gratings with groove densities of 1200 mm-1, 600 mm-1 and 150 mm-1, with blaze size of 500 nm and 68 mm x 68 mm dimensional size. The CCD camera mounted on the spectrograph was a Princeton Instruments PIXIS 256E (NJ, USA). The read-out fiber was purchased from AFW Technologies Pty Ltd (Hallam, Australia) and had a core diameter of 50 m, numerical aperture of 0.12 and FC connectors on both ends. The light source used for excitation of the monomers on the interface was an LED with λ = 365 nm and with 250 W power, which was purchased from Omicron Laser (Rodgau, Germany). It was mounted on a lens tube along with a UV fused silica bi-convex lens with 40 mm focal length (Thorlabs, LB4030-UV, Newton, NJ, USA) as well as a band-pass filter transmitting at 370 nm with FWHM of 10 nm (Thorlabs, FB370-10, Newton, NJ, USA). The beam was focused at the interface to produce an illuminating spot with a diameter of about 2 mm. The read-out stage was aligned and focused so to collect as much emission signal as possible. The read-out fiber was fixed on a stage which was built in-house. The exposure time was 3 minutes for each spectrum while the LED was at full power during the acquisition time. The delay time between the spectra was 2 minutes. A binning of 4 was used to record the data. The CCD read- out was set to the region of interest (ROI), while the ROI set-up on the slit ran from 130 as the start point with a height of 23. Although the entire set-up of the LED for excitation, and the detector for collecting the fluorescence emission and interface had a fixed geometry for each experiment, small changes in this geometry were unavoidable between experiments. Therefore a direct comparison, e.g. of fluorescence intensity, is not possible. It is possible to draw quantitative conclusions (such as rate of fluorescence decay) from a single experiment only. The set-up for the experiment is shown in Figure 3.28.
Figure 3.28. Experimental set-up for in-situ fluorescence measurements at the air/water interface. 184
AMPHIPHILIC ANTHRAPHANES
Polymerisation and film transfer onto Cu TEM grids The monomer monolayers were polymerised by direct exposure to a fiber-coupled 365nm LED (Omicron Laser EDMOD365.250.OEM) at a constant surface pressure of 20 mN/m for 8 h. The irradiation spot was about 2 cm in diameter. After polymerisation, TEM grids made of copper with the mesh size of 1000 (PLANO, G2780C) were placed on the irradiated spot. A clean white piece of paper was then gently put on the grids so for them to adsorb on it. The paper along with the grids on it was then removed and dried overnight at room temperature.
Scanning electron microscopy (SEM) The TEM grids were placed on a PLANO G3662 holder and imaged with FEG-SEM, Zeiss LEO Gemini 1530, Germany, microscope with an in-lens detector.
MMA values calculations from crystal packings MMA values were calculated from the crystal packing structures obtained for anthraphane 1 in Chapter 2. The area of the unit cell was determined and divided by number of molecules in it.
Dense mixed etf/ftf packing 1
Figure 3.29. MMA calculation for the mixed etf/ftf packing 1: (21.099 Å x sin(81.490°) x 13.249 Å) / 2 ≈ 138 Å2.
185
AMPHIPHILIC ANTHRAPHANES
etf packing 1
Figure 3.30. MMA calculation for the etf packing 1: (15.555 Å x sin(58.940°) x 15.402 Å) ≈ 205 Å2.
ftf packing
Figure 3.31. MMA calculation for the ftf packing: (21.276 Å x sin(62.260°) x 20.846 Å) / 2 ≈ 196 Å2.
186
DIAZAANTHRAPHANES
4. Diazaanthraphanes
In the following chapter, the structure of anthraphane 1 will be engineered in order to create a new monomer useful for the topochemical synthesis of 2DPs in single crystals. The knowledge gained in Chapter 2 on the crystallisation behaviour of 1 and its different crystal structures will be exploited in the hope of obtaining a structure that packs exclusively in an all-ftf packing independently of the solvent used for crystallisation. This would avoid the tedious and luck-based solvent screening procedure that was used for 1 and increase the chances of successfully obtaining a new kind of 2DP.
4.1 Monomer’s Design
The main problem in achieving an all-ftf packing in single crystals with anthraphane 1 (see Chapter 2), was the predisposition of the anthracene units on the monomers to engage preferentially in etf interactions. In some instances, mixed etf and ftf interactions were observed, especially when using flat and small solvents; these partial ftf interactions were however the inevitable result of the monomer trying to pack in the densest way possible (for details please refer to subchapter 2.7). A possible rationale for the avoidance of ftf interactions was found in the considerably negative electrostatic surface potential on the faces of the anthracene units: from an electrostatic point of view, to avoid repulsion, the best orientation of the quadrupolar moments of the anthracenes was the etf geometry. We therefore designed the new anthraphane derivative 3 (Figure 4.1) specifically to avoid etf interactions, by substituting the anthracene units with 2,7-substituted 1,8- diazaanthracenes. This structure will be referred to as “diazaanthraphane”.
Figure 4.1 Chemical structure of the diazaanthraphane monomer 3 for the topochemical synthesis of 2DPs.
187
DIAZAANTHRAPHANES
Heterocyclic derivatives such as azaanthracenes and diazaanthracenes behave photochemically very similarly to anthracene, dimerising upon photoirradiation through [4+4]- cycloaddition and back-reacting upon proper thermal or photochemical stimulus (Scheme 4.1). Bond formation occurs at the 9 and 10 position, where addition of singlet oxygen is also possible through photooxidation under aerobic conditions[185].
Scheme 4.1 Photochemical dimerisation of 1,8-diazaanthracene.
Compared to anthracene, studies on the photoreactivity of diazaanthracenes are quite scarce in the literature, however, dimerisations of 1,8-diazaanthracenes have been reported in single crystals[185–187], in solution[186–188] and at the air/water interface[60], where the possibility of heterodimerisation between anthracenes and diazaanthracenes was also demonstrated[61]. By analysing the structure of 3, one can expect a reduced negative ESP on the faces of the 1,8- diazaanthracene units compared to anthraphane 1. The heterocyclic nature of the compound slightly depletes the electron density of the aromatic rings, making them electron-poor; this effect is strongly enhanced by the electron-withdrawing carbonyl groups in ortho position to the nitrogens, as can be seen in Figure 4.2.
Figure 4.2 ESP of anthraphane 1 and diazaanthraphane 3 (PM3 method). Compared to the anthracene units, the electron density on the carbonyl-substituted 1,8-diazaanthracene is strongly depleted resulting in a slightly positive MEP value. MEP values are measured at the marked position.
188
DIAZAANTHRAPHANES
In fact, the MEP value at the central ring of the substituted 1,8-diazaanthracenes is +0.7 kcal/mol, rendering this unit a weak acceptor in character, compared to the strongly negative value of -16.3 kcal/mol for the donor anthracene counterpart. With an almost neutral ESP on 3, one can expect a minimal repulsive electrostatic term between the 1,8-diazaanthracene units, increasing the chance for ftf stacking over etf interactions. ftf interactions in single crystals should also be preferred by the favourable dipole orientation: when the diazaanthracene stack in an anti-parallel fashion, the dipole moment induced by the aromatic nitrogens and the carbonyl groups are compensated (Figure 4.3).
Figure 4.3 Dipole moment (blue arrows) compensation is expected to help in promoting ftf stacking between the diazaanthracene units.
As a final remark, the methyl group at the 9 position, which is introduced for synthetic necessity (see subchapter 4.2), could also effectively prevent etf interactions; having a reduced electrostatic term on the aromatic surfaces, dispersion interactions are likely to govern the interactions among diazaanthracenes in the crystal structures of 3, promoting ftf stacking in order to maximise intermolecular contact. The ester groups, apart from influencing the ESP of 3, can also be seen as versatile units for post-polymerisation modification (for details refer to subchapter 1.3), in this case they would act as surface groups; similarly, in the ideal hexagonal honeycomb packing, the aromatic nitrogens would be located inside the pores, acting therefore as pore groups, available for protonation or alkylation (Figure 4.4). Moreover, the ester groups would point towards the layers in the expected lamellar crystal structure and could be therefore exploited to ease the exfoliation procedure of the 2DP crystal. In this regard, the 2DP crystals could be exposed to basic conditions in order to hydrolyse the esters; the formation of carboxylate groups could then effectively exfoliate the sheets by electrostatic repulsion, yielding negatively surface-charged 2DPs (Figure 4.5). Finally, we did not want the monomer to bear functional groups that could disturb the intended ftf stacking
189
DIAZAANTHRAPHANES by, for instance, creating additional interactions such as hydrogen bonds; therefore, carboxylic acids, amines, amides and alcohols were not considered as functional groups.
Figure 4.4 Example of post-polymerisation modification by exploiting the pore groups. The aromatic nitrogens could be protonated or alkylated to generate 2DP polyelectrolytes. Counter-anions are omitted for clarity.
Figure 4.5 Example of chemical exfoliation of 2DP crystals: the ester groups are hydrolysed exposing negatively charged carboxylate groups on the surface of the sheets (shown with a red cross-section), which upon electrostatic repulsion promote the exfoliation process. Counter-anions are omitted for clarity.
4.2 Synthetic Approach
The synthetic approach towards compound 3 was adapted from the synthesis of anthraphane 1, in the hope that the synthetic process would work similarly. Scheme 4.2 depicts the retrosynthetic strategy: the key building block is again a triflated species, compound 33, which upon Sonogashira
190
DIAZAANTHRAPHANES cross-coupling with 10 forms the precursor 34 and after a final copper-free Sonogashira reaction yields the diazaanthraphane 3. Regarding the nature of the ester groups, methoxy was chosen as starting point, since the synthesis of similar 1,8-diazaanthracenes derivatives was already reported in the literature[185,189].
Scheme 4.2 Retrosynthetical strategy towards diazaanthraphane 3. The ditriflate 33 was chosen as key building block.
4.3 Synthesis of Diazaanthraphanes
4.3.1 Synthesis of methyl ester diazaanthraphane 3a
The synthetic route towards the methyl ester diazaanthraphane 3a is depicted in Scheme 4.3; the first two steps of the sequence towards compounds 31a and 32a were already reported in the literature in 1983 by Molock and Boykin[189].
Scheme 4.3 Synthetic route for the synthesis of the methyl ester diazaanthraphane 3a.
191
DIAZAANTHRAPHANES
In a first step, commercially available dimethyl acetylenedicarboxylate (DMAD) 30a reacted with 2,6- diaminotoluene in an aza-Michael addition reaction to form compound 31a in quantitative yields, which could be isolated in pure by simple filtration. Double intramolecular cyclisation of 31a in boiling diphenyl ether afforded compound 32a without particular problems in 92% yield. For this step, care should be taken in working with degassed diphenyl ether and under diluted conditions (20mM) to maximise the yields. It is noted that the methyl group on the aromatic moiety of 31a selectively directs the cyclisation step towards the diazaanthracene derivatives 32a; without it, the product of this step would be a mixture of diazaanthracenes and diazaphenanthrenes. Triflation of
32a by treatment with Tf2O, 2,6-lutidine as base and DMAP as catalyst in dry DCM afforded quantitatively the diazaanthracene ditriflate 33a on a multi-gram scale (30 g). The product could be isolated by simple filtration and washing with methanol as a bright yellow crystalline solid, best soluble in chlorinated solvents such as chloroform or DCM. 33a turned out to be quite stable in the solid state under ambient conditions, but due to its photoreactive nature it was best stored in the dark; in fact, solutions of 33a exposed to direct sunlight visibly tarnished after few hours. In solution, compound 33a was found to be quite activated towards nucleophilic aromatic substitution through loss of the triflate moiety, as evidenced by its immediate reaction with secondary amines such as piperidine. 2,6-Lutidine being non-nucleophilic and sterically hindered, proved to be the optimal choice as mild base for working with this compound. It is noted that the key building block 33a was obtained in pure form over three steps with a total yield of 80% and without the need for column chromatography as purification method.
The synthesis of the precursor 34a was then attempted by using the same Sonogashira cross- coupling conditions used for the synthesis of anthraphane’s precursor 11, only using 2,6-lutidine as base instead of triethylamine. Unfortunately, in no instance, compound 34a could be isolated in pure form. The reaction worked analogously to the anthraphane case (see subchapter 2.3), starting as a suspension of ditriflate 33a in dioxane, which upon onset of the reaction became a clear solution and after 2 d a yellow suspension. Isolation of the yellow solid by filtration, followed by multiple washings with a variety of solvents and attempted recrystallisations did not yield the desired product. The NMR spectra of the yellow solid appeared as a mixture of different compounds impossible to distinguish but mass spectrometry could confirm the presence of the product in the mixture. The reaction was repeated without success by using different solvents such as DMF, toluene and THF and by employing higher temperatures (up to 100°C). It is speculated that the intermediates of this reaction suffer from insolubility and therefore precipitate out of the reaction mixture before reaching the precursor stage; in fact, perhaps counterintuitively, ditriflate 33a despite bearing different functional groups is much less soluble than its anthracene counterpart 9. This could however be a hint of the higher tendency to aggregate for 1,8-diazaanthracene-based compounds 192
DIAZAANTHRAPHANES with respect to anthracene ones, which could be a consequence of stronger π-π interactions. The mixture of intermediates was nevertheless employed for the final copper-free Sonogashira step to see if small amounts of 3a could at least be obtained, but unfortunately no product could be detected, nor by NMR spectroscopy nor by mass spectrometry.
Ditriflate 33a was therefore subjected to a series of test Sonogashira cross-couplings in order to exclude intrinsic reactivity problems with the compound. 33a was coupled with trimethylsilylacetylene (TMSA), trisisopropylsilylacetylene (TIPSA) and phenylacetylene by using the same conditions used for the synthesis of 34a (Scheme 4.4).
Scheme 4.4 Sonogashira cross-couplings of ditriflate 33a with different acetylenic moieties.
In all cases, the doubly substituted products 35, 36, and 37 were obtained in pure form without particular problems and in good yields. This experiment proved that ditriflate 33a is indeed reactive to cross-coupling under mild conditions and does not present any reactivity problem. Apart from the well soluble silylated compounds 35 and 36, compound 37 resulted completely insoluble in conventional solvents; clear solutions of 37 were achieved in boiling ODCB (180°C), from which upon cooling, single crystals in the form of yellow needles were obtained. SC-XRD analysis revealed a rather surprising packing, in which the 1,8-diazaanthracene units were engaged in a syn-parallel stacking rather than an anti-parallel stacking as hypothesised previously (Figure 4.6). This should however not concern the diazaanthraphane monomer 3, where syn-parallel stacking is intrinsically not possible due to its geometry. What is important to note with this structure, is the high tendency of the compound to aggregate and stack efficiently, which is also evident by the lack of solvent molecules incorporated in the crystal structure. If 37 can be considered as a model compound, then for the diazaanthraphane monomer 3, the chances of getting an all-ftf packing could be realistic.
193
DIAZAANTHRAPHANES
Figure 4.6 Crystal structure of compound 37. The molecule packs in arrays creating a herringbone motive. The 1,8-diazaanthracenes stack in an unusual ftf syn-parallel displaced fashion and are equidistant to each other. The distance between the 9 and 10 position of the two molecules is approximately 4.78 Å. The solvent ODCB does not co-crystallize with the molecule.
For the synthesis of precursor 34a, the brominated coupling partner 38 was synthesised as alternative to the ditriflate. The compound was smoothly obtained in nearly quantitative yields by treating 32a with phosphoryl bromide (Scheme 4.4). However, 38 resulted even more insoluble than its triflated counterpart and did not display any reactivity towards cross-coupling: even under forcing conditions such as microwave treatment at 150 °C in NMP, no coupling products could be obtained. This experiment nicely highlighted the superiority of triflate groups in Sonogashira cross-coupling reactions: not only have triflates a higher reactivity, but they can also be used as solubilising units.
Scheme 4.4 Synthesis of dibromide 38.
In order to make the synthesis of precursor 34 feasible, the solubility of the building blocks had to be enhanced: for this purpose, we decided to increase the length of the ester side-chains.
194
DIAZAANTHRAPHANES
4.3.2 Synthesis of Ethyl and n-Propyl Ester Diazaanthraphanes 3b-c
Attachment of the solubilising ester moieties was performed at the very first step of the synthetic sequence by esterifying the cheap and commercially available acetylenedicarboxylic acid 29; we chose ethyl- and n-propyl-esters as side chains: the choice was directed at slightly increasing the solubility of the synthetic intermediates, without compromising their crystallisability. Synthesis of higher esters such as n-hexyl and n-dodecyl was also explored in the hope of obtaining mesogenic diazaanthraphanes for the synthesis of 2DPs in smectic phases; for more informations please refer to the Appendix section. The synthesis of compounds 3b-c is depicted in Scheme 4.5:
Scheme 4.5 Synthesis of ethyl and n-propyl ester diazaanthraphanes 3b-c.
The sequence starts off with acetylenedicarboxylic acid 29. The acid was first esterified with the corresponding alcohol (b: ethanol, c: n-propanol) using TsOH as catalyst under azeotropic removal of water in benzene and afforded the diesters 30b-c without particular problems; the products were conveniently purified by distillation. Following the procedure for the methyl ester derivative 30a, the diesters were then subjected to double aza-Michael additions with 2,6- diaminotoluene as nucleophile in mixtures of CHCl3 and a protic solvent such as MeOH, to selectively ensure (Z)-isomer formation[190]. This resulted in fast and high-yielding formation of compounds 31b- c, which precipitated out of the reaction and could be recovered in pure form by simple filtration. Double intramolecular cyclization under dilute conditions in refluxing diphenyl ether afforded the pyridoquinolone derivatives 32b-c in good yields ranging from 89% to 95%. In particular, the slightly increased steric bulk provided by longer alkyl chains in 32c was beneficial for the scale-up of the
195
DIAZAANTHRAPHANES reactions. It prevented intermolecular reactions and allowed to use more concentrated solutions (e.g. 118 mM instead of 43 mM), though this effect was not fully investigated. The products were insoluble enough to conveniently be collected by filtration. Triflation of 32b-c by treatment with
Tf2O, 2,6-lutidine as base and DMAP as catalyst in dry DCM afforded quantitatively the ditriflates 33b-c on a multi-gram scale (40 g). In all cases the products could be obtained in high purity by simple filtration (Figure 4.7).
1 Figure 4.7 H-NMR spectra of ditriflates 33b-c in CDCl3.
In the next step, the direct target precursors were assembled. For this purpose the ditriflates 6b-c were coupled in excess with 1,3,5-triethynylbenzene 10 under standard Sonogashira conditions in dioxane, with Pd(PPh3)4 and CuI as catalyst and co-catalysts, respectively, and 2,6-lutidine as base. The reaction proceeded similarly to the attempted synthesis of 34a, starting as a suspension, clearing up at the onset of the reaction and ending up as yellow suspension. This time however, filtering of 196
DIAZAANTHRAPHANES the yellow solid afforded the monomer precursors 34b-c in yields ranging from 60-71% and on the gram scale. It needs to be mentioned that this synthetic step was fairly capricious and that the purity of the obtained precursors could vary between 75-95%; despite the longer ester side-chains, 34b-c were still poorly soluble which caused difficulty in purification, as it happened with compound 34a. If after multiple washings with dioxane, acetonitrile and methanol these two compounds could not be obtained in pure form, they were used as is. Column chromatography was not considered an option due to the six nitrogen functionalities, the insolubility of the precursors and the sensitivity of the triflate groups. Overall, the best results were always obtained with the propyl esters, probably due to the slightly better solubility. The excess of the ditriflates 6b-c employed for the reaction could be recovered virtually quantitatively by precipitation from the reaction mixture with MeOH and subsequent filtration on a silica plug. Figure 4.8 displays the NMR spectra of these two compounds in the highest purity achieved.
1 Figure 4.8 H-NMR spectra of precursors 34b-c in CDCl3 obtained in the highest purity.
For the final step to the diazaanthraphanes 3b-c, the precursors 34b-c were subjected to the required complex series of Sonogashira cross-couplings. As for anthraphane 1, copper-free 197
DIAZAANTHRAPHANES
Sonogashira conditions were applied under high-dilution (1-2 mM) with 10 as stoichiometric coupling partner. Pleasantly, the target monomers 3b-c could be obtained in 100 mg scale and, considering the complexity of the final step, yields ranging from 20-32% were quite satisfactory. Compounds 3b-c precipitated out during the reaction and could be purified by filtration and washing with toluene and methanol. Given the high symmetry of the compounds, analysis by NMR spectroscopy was simple; characterization by 1H-NMR (Figure 4.9) and high-resolution mass spectrometry (HRMS) confirmed the proposed structure of 3b-c. Well resolved 13C-NMR spectra could not be obtained because of insufficient solubility. It is noted that the poor solubility of these two compounds is a key element of their structure design as it will help to force these propellers into the crystalline state.
1 Figure 4.9 H-NMR spectra of diazaanthraphanes 3b-c in CD2Cl4.
The melting points were found to be above the decomposition temperature of the monomers, which was determined to be approximately 300 °C by thermogravimetric analysis. The UV/Vis absorption spectra of 3b-c are virtually superimposable and therefore the propyl series was chosen as representative for the characterisation. The fluorescence spectra were also measured for 198
DIAZAANTHRAPHANES the propyl representatives by diluting the same solutions used for the UV/Vis study. Figure 4.10 displays the absorption and emission spectra of the diazaanthraphanes 3c, precursor 34c, ditriflate 33c and the model compound 37. While in all cases the typical vibronic structure of 1,8- diazaanthracenes is clearly visible, for the rest, no particular noteworthy features are present.
Figure 4.10 UV/Vis absorption (solid lines) and fluorescence (dotted lines) spectra of the propyl ester series 33c, 34c, 3c and model compound 37 measured in CHCl3. Excitation wavelength for emission λ = 365 nm.
Overall, the synthetic sequence worked best for the propyl derivatives, probably due to slightly increased solubility of the intermediates at the late stages of the synthesis. The overall yields over four steps to the key building blocks ditriflate 33b and 33c were 49% and 72% respectively, whereas the overall yields over six steps to the monomers 3b and 3c were 6% and 16% respectively. With these new potential monomers in hand, their crystallisation behaviour was investigated.
199
DIAZAANTHRAPHANES
4.4 Crystallisation of Diazaanthraphane 3c and Packing in the Single Crystal
Similarly to anthraphane 1, compound 3c was recrystallised by slow cooling of nearly saturated solutions in high boiling point solvents. The crystallisations were carried out with the same apparatus described in Figure 2.23 and the same conditions for the crystallisation of 1 were used: 2-3 mg of diazaanthraphane 3c were put in a clean glass vial with 0.5-1.0 mL of solvent and heated until a clear solution was obtained. Prior to heating, the suspensions were purged with argon; typical cooling times were 24-36 h. The first solvent choice was o-dichlorobenzene (ODCB), from which yellow hexagonal platelets were obtained (Figure 4.11).
Figure 4.11 Optical micrographs of single crystals of 3c obtained from ODCB in bright field mode (left) and between crossed polarisers (right).
The hexagonal morphology of the crystals is already important information as it could mean that the monomer crystallises into the desired hexagonal honeycomb-like all-ftf packing. It is important to note that while the crystals looked promising by observation under the microscope in bright field mode, between crossed polarisers the birefringence revealed inhomogeneity throughout the crystals: as can be seen in Figure 4.11, domains are visible which result in colour variations on the same crystal. This inhomogeneity in crystallinity was confirmed by SC-XRD analysis, during which the crystals exhibited poor diffraction, so that information on the packing of the monomer could not be obtained. Multiple crystallisations from ODCB were performed by varying the concentration and cooling rate of the solutions but the crystal quality could not be improved. Longer cooling times (36- 96 h) produced larger crystals up to 700 µm in size but with considerable lower quality with respect to smaller crystals obtained with 24-36 h cooling. Due to the high density of ODCB, the crystals floated at the air/liquid interface and tended to stick on the glass walls of the vials, thus leaving the mother solution. Owing to this phenomenon, the poor crystal quality observed was attributed to a
200
DIAZAANTHRAPHANES potential solvent loss from the crystal. 3c was therefore crystallised from less dense solvents such as nitrobenzene and 1,3-dimethoxybenzene.
Despite the change in the crystallisation solvent, the obtained crystals still appeared as hexagonal platelets but retained the characteristics observed for the ODCB solvate. The optical micrographs of the nitrobenzene and 1,3-dimethoxybenzene solvates are displayed in Figure 4.12 and 4.13 respectively. Apart from the inhomogeneity in birefringence, some crystals appeared occasionally ill-shaped and had the tendency to stack on top of each other. Regardless of the low crystal quality, SC-XRD analysis was nevertheless tried but in both solvates diffraction was again poor and no reliable structure information could be extracted. To compensate for the low diffracting power, crystals from the 1,3-dimethoxybenzene solvate were subjected to SC-XRD with synchrotron radiation. Gratifyingly, a full data set could be recorded and an acceptable crystal structure for compound 3c could be obtained, finally shedding light into the packing behaviour of diazaanthraphane.
Figure 4.12 Optical micrographs of single crystals of 3c obtained from nitrobenzene in bright field mode (left) and between crossed polarisers (right).
Figure 4.13 Optical micrographs of single crystals of 3c obtained from 1,3-dimethoxybenzene in bright field mode (left) and between crossed polarisers (right). 201
DIAZAANTHRAPHANES
The first feature of the packing is its layered structure, as observed for every anthraphane derivative so far (Figure 4.14); this time however the layers are not as tightly packed as seen before, in fact the interlayer distance approximates to 1 nm. Unfortunately there is no information about the location of the solvent 1,3-dimethoxybenzene, but solvent molecules are presumably present between the layers and act as glue between them, as there is no evident interlayer interaction between the propyl ester chains of the monomers.
Figure 4.14 ORTEP diagram of compound 3c in the 1,3-dimethoxybenzene solvate (50% probability) (left). Layered crystal structure of 3c (right). The interlayer distance is roughly 1 nm, compared to 0.4 nm for a typical structure of anthraphane 1. No information about the location of solvent molecules is available, but their presence between the layers is hypothesised.
By analysing how the molecules are packed within the layers, a pleasant discovery was made: diazaanthraphane packs in the long sought honeycomb-like packing (Figure 4.15)! By looking in more detail, one sees that 3c adopts a pseudo-hexagonal packing in the layers, forming a typical porous structure, with pores being approximately 14 Å in diameter (accounting van der Waals radii). Only, two solvent molecules per pore could be located, but it is likely that the empty voids are filled with additional solvent. By looking at how the layers stack on top of each other, one sees that channels are present in the structure, similarly to the DMPU and benzyl benzoate solvates of anthraphane 1 (see subchapter 2.6.1.3 and 2.6.1.4).
202
DIAZAANTHRAPHANES
Figure 4.15 Detailed view of a layer in the crystal structure of 3c (left); the molecule packs in a honeycomb-like fashion with solvent molecules filling the pores. Stacking of the layers in the spacefill view (right); the layers are arranged so that channels with a diameter of 14 Å are present in the structure.
Figure 4.16 shows the packing in the layer and the stacking relationship with the red diazaanthracene units (side-chains omitted for clarity). The first aspect that becomes evident is that the packing is indeed all-ftf, possibly suitable for a topochemical 2D-polymerisation; there is however considerable displacement in the stacking between the photoreactive units as showed in Figure 4.17.
Figure 4.16 All-ftf packing of 3c. The photoreactive diazaanthracene units are displayed in red colour. Ester side-chains have been omitted for clarity.
203
DIAZAANTHRAPHANES
Figure 4.17 Stacking relationships between the photoreactive diazaanthracene pairs. The distances between the 9 and 10 positions of the violet, blue and green pairs are approximately 3.8 Å, 4.3 Å and 4.0 Å respectively. The displacement of the blue and green pairs is however considerable compared to the violet pair.
The three diazaanthracene pairs per monomer are depicted with violet, blue and green colour; the distances between the 9 and 10 positions of the violet, blue and green pairs are approximately 3.8 Å, 4.3 Å and 4.0 Å respectively. While the distances are in range for a typical topochemical reaction, the misalignment of the orbitals at the 9 and 10 positions of the blue and green pair could be problematic. The situation resembles the crystal structure of anthraphane 1 in the mixed etf/ftf packing 1 discussed in subchapter 2.6.3, where one anthracene pair (pink in Figure 2.50 and 2.51) could not topochemically dimerise due to the displacement of the anthracene moieties. Even irradiations at high temperature to allow more molecular movement in the crystals did not help in properly aligning the pair for a dimerisation to occur (see subchapter 2.9.1). While the crystal packing of 3c is different from the one of 1, a similar scenario cannot be totally excluded.
Regarding the quality of the crystals obtained, it is still not clear how it could be improved. As already mentioned, varying the crystallisation parameters such as temperature and cooling rate did not help. It seems however that the use of aromatic solvents is beneficial, as crystallisations carried out with TCE, DMPU, GBL or NMP only yielded low quality ill-shaped crystals. It might well be that the poor diffraction quality of crystals of 3c is related to the weak interactions between the layers; weakly interacting layers could result in misalignments in the crystal structure which would then result in disorder expressed as smeared bragg peaks. Aromatic solvents, being flat, are thought to better fit between the layers, hence the better crystal quality observed with respect to crystals grown from aliphatic solvents. It is however important to emphasize that a weak interaction between
204
DIAZAANTHRAPHANES the layers is likely to be beneficial for the exfoliation procedure after polymerisation. Shortening of the ester side chains could help in bringing the layers closer together, however solubility problems would start to manifest as seen for the synthesis of 3a. For ethyl ester diazaanthraphane 3b, the crystals obtained exhibited the same features as the ones of 3c and were poorly diffracting; unfortunately a comparative crystal structure of 3b could not be obtained.
As a final remark, we wish to point out that while full data-sets for the nitrobenzene and ODCB solvates of 3c could not be recorded due to the low diffracting power of the crystals, the unit cell parameters could be nevertheless determined and corresponded (within the standard uncertainties) to the unit cell parameters of the 1,3-dimethoxybenzene solvate. This is important information, as it means that the structure of diazaanthraphane 3c, as hypothesised in subchapter 4.1, has indeed a tendency to prefer ftf interactions irrespective of the solvent used for crystallisation. If it can be crystallised in the same all-ftf motif from different solvents, the frustrating and luck-based solvent screening for crystallisation employed for 1 can be completely avoided. This example nicely shows the importance of a careful monomer’s design; it is however possible that with other solvents, different packings for 3c could also be obtained.
4.5 Preliminary Irradiation Studies
In order to assess the best conditions for the topochemical 2D polymerisation of the crystals of 3c, their solid-state UV/Vis absorption spectrum was measured (Figure 4.18).
Figure 4.18 Solid-state UV/Vis absorption spectrum of anthraphane. The chosen wavelength for tail- end irradiation is 530 nm. 205
DIAZAANTHRAPHANES
Again, irradiating a crystal with the lowest possible energy wavelength ensures a more homogeneous reaction throughout the entirety of the crystal. According to the solid-state UV/Vis absorption spectrum of 3c, in order to employ tail-end irradiation, the wavelength was chosen to be 530 nm corresponding to visible green light. The choice of such a weak absorbing wavelength was dictated by the LEDs available in our lab. A similar circular LED reactor as the one shown in Figure 2.100 was used. It was found beneficial in order to not decrease the quality of the crystals, to not remove them from their mother solution; this could be attributed to possible solvent loss from the structure, due to the presence of channels. Irradiations were therefore performed directly on the glass vials containing the crystals in their mother solution.
Crystals obtained from 1,3-dimethoxybenzene were irradiated for 16 h at 4°C in the fridge and afterwards the crystals were analysed by optical microscopy. Figure 4.19 shows that upon irradiation the crystals lose their yellow coloration and become virtually transparent. After irradiation, birefringence is still partially retained and the hexagonal shape of the crystals stays intact.
Figure 4.19 Optical micrographs of crystals of 3c obtained from 1,3-dimethoxybenzene in bright field mode and between crossed polarisers before irradiation (left). Optical micrographs of the same crystals in bright field mode and between crossed polarisers after irradiation (right).
206
DIAZAANTHRAPHANES
Attempts to measure SC-XRD of the irradiated crystals were performed but did not lead to any useful result, as the crystals were non-diffracting. Also, the small amounts of crystals available prevented analysis by SS-NMR or IR for proof of bond formation between the diazaanthracene units. However, decolouration of the crystals could be a hint of dimerisation polymerisation taking place: as shown in Figure 4.20, similarly to the anthracene case, the highly fluorescent diazaanthracene pairs upon dimerisation become colourless and lose their typical fluorescence signature, meaning that with dimerised pairs the excimer signature should disappear completely.
Figure 4.20 Dimerisation of diazaanthracenes is accompanied by the loss of excimer fluorescence and the typical yellow colouration of the diazaanthracene molecules.
To strengthen our argument, few crystals were carefully separated from the mother solution, washed with fresh methanol and placed on a quartz slide inside a droplet of perfluoroalkylether in order to protect them from solvent loss. The fluorescence spectrum of the crystals was then measured, which revealed the typical broad peak centered at around 527 nm, characteristic for excimer emission. The crystals were subsequently irradiated at 530 nm and the fluorescence decay was measured. The results are displayed in figure 4.21: after 24 h, the excimer emission decayed by approximately 60%. To ensure full conversion, the crystals were therefore irradiated for additional 24 h, but the excimer population only decayed minimally. Even though the absorption of 3c at 530 nm is minimal, due to the high power of the LED photoreactor, full conversion should be expected maximally in a couple of hours (compare SCSC reactions of 1 in subchapter 2.9). This incomplete conversion could be the result of the misalignment of the diazaanthracene pairs as explained in Figure 4.17: the residual excimer signal observed, could be due to a minimal overlap in the orbitals of the misaligned pairs, which however cannot snap together to form a dimer due to restricted thermal motion. It has to be however noted, that the LED reactor generates some heat during irradiation so that the measurement is performed at temperature higher than ambient conditions. Unfortunately, due to the experimental set-up, a temperature control during the measurement is not possible.
207
DIAZAANTHRAPHANES
Another reason for the incomplete conversion could be the intrinsic sensitivity of the crystals: by taking them out of their mother solution and rinsing them with fresh solvent, their quality could considerably lower, which would result in more defects in their packing and possibly loss of crystallinity. This process is however necessary, as traces of mother solution on the crystals would result in additional unwanted emission signals. The heat generated by the photoreactor during the irradiation could also contribute to solvent loss from the crystals.
Figure 4.21 Fluorescence decay of crystals of 3c irradiated at 530 nm over 48 h.
To summarise: after irradiation the crystals completely lose their already weak diffracting power so that SC-XRD analysis is not possible; irradiation is probably accompanied to some degree by bond formation between the diazaanthracene pairs, as confirmed by the partial excimer decay and decolouration of the crystals (formation of endoperoxide cannot be completely excluded but it is unlikely[185]). As mentioned before, with a bigger amount of crystals available, proof of bond formation should be investigated by for instance SS-NMR; moreover, even if not quantitative, a partial degree of polymerization would result in mechanical stability within the sheets, so that the partially polymerised crystals could be nevertheless exfoliated in the hope of obtaining single sheets entities, even in the form of oligomers. This would be an additional confirmation that polymerisation happens between within the sheets. Unfortunately, due to time constraints, this investigation could not be carried out.
208
DIAZAANTHRAPHANES
4.6 Conclusions and Outlook
In conclusion, a new set of monomers derived from anthraphane 1, the diazaanthraphanes 3b-e were successfully synthesised. These diazaanthraphanes are based on 9-methyl-1,8- diazaanthracenes as photoreactive units; in contrast to anthracenes, this units were expected to pack in the single crystals by engaging exclusively in ftf interactions , due to a less pronounced electrostatic repulsive term between them. 3b-e were also equipped with ester side chains, in the view of easing the exfoliation process of the crystals after 2D-polymerisation and for possible post- polymerisation modifications. The synthesis, similarly to 1, passed through the key ditriflated building blocks 33b-e, which could be easily obtained in 40 g scale. The diazaanthraphanes monomers were then assembled and tested for pre-organisation: 3d-e, bearing n-hexyl and n-dodecyl ester side- chains could not be crystallised and did not display any tendency to organise in smectic phases; 3b-c were crystallised and for 3c, SC-XRD analysis revealed that the monomer arranged itself in the desired all-ftf packing, forming a porous hexagonal honeycomb structure. Moreover, it was demonstrated that the monomer achieves the same packing, by being crystallised from different solvents such as 1,3-dimethoxybenzene and ODCB (donors), and nitrobenzene (acceptor). This important finding showed that with the experience gathered in Chapter 2 regarding the crystallisation of 1, it was possible to design a structure that would pack exclusively in the desired fashion. This is valid independently of the solvent characteristics, so that the uncertainties and screening procedures associated with the single crystal approach towards 2DPs are eliminated. Unfortunately, the crystals did not diffract properly; a possible cause could be found in the disorder within the layers and between the layers induced by the side-chains. It is believed that the side chains might prevent the layers to come in close contact, resulting in a very weak interlayer interaction. This would likely result in poorly oriented and stacked layers. Due to the low quality and weak diffraction power of the single crystals obtained, analysis could be employed only with synchrotron radiation. In the crystal packing, the monomers also displayed a tendency to interpenetrate each other with their diazaanthracene units: this natural tendency of 3c to pack as tightly as possible resulted in considerable misalignment of the photoreactive units. Tail-end irradiation of the crystals with fluorescence decay measurements showed that upon prolonged irradiation, the excimer emission only decayed by approximately 60%. This partial photopolymerisation resulted into the decolouration of the crystals, which still retained their hexagonal shape but completely lost their diffractive power. As such, structural analysis of the irradiated crystals could not be performed by SC-XRD and due to the small amount of crystals available, bond formation could not be investigated by other methods such as SS-NMR.
209
DIAZAANTHRAPHANES
We however believe this system to be very promising for the synthesis of a novel 2DP; as such, the next step would be to exfoliate the partially polymerised crystals and see if single sheet entities can be obtained. The incomplete polymerisation, combined with the very weak interacting layers in the crystals, should result in a fast and easy exfoliation process; AFM analysis could then be employed to identify the monolayers. However, due to the intrinsic low quality of the crystals of 3b- c, unequivocal structural proof of the obtained polymers could be very difficult to obtain, even by using brighter x-ray sources.
To tackle the two main issues encountered in the crystal structures of 3b-c, namely the weakly interacting layers and the misalignment of the photoreactive units, we propose a further engineering of the diazaanthraphane structure, which does not imply the loss of functional groups available for PPM: the diazaanthraphane 3f displayed in Figure 4.22.
Figure 4.22 Further engineered diazaanthraphane 3f for simplifying SC-XRD analysis.
In 3f, the ester side-chains have been reduced to the minimal length possible by using methyl esters. The reduction of side-chains should hopefully result in an increased interaction between the layers in the crystals. It is worth mentioning again that the synthesis of a methyl ester monomer 3a could not be achieved due to severe solubility problems encountered at the precursor 34a stage. In this regard, the substituted central benzene cores would play two crucial roles: on one side they would confer solubility during the synthesis, on the other side they would sterically prevent the monomer interpenetration in the crystal packing, which usually results in misalignment of the photoreactive units. This phenomenon was observed in the crystal packing of monomer 2a in the nitrobenzene solvate: due to the methoxy groups on the central core, the anthracene units could not interpenetrate and were therefore perfectly stacking ftf, which resulted in an easy SCSC 1D- polymerisation of 2a (see Figure 6.2 and Figure 6.3 in the Appendix for details).
210
DIAZAANTHRAPHANES
As a final remark, we wish to point out that the synthesis of the diazaanthraphanes 3b-e opens a totally new prospect for the field of 2DPs: for instance, monomers 3d-e, due to their high solubility and low tendency to crystallise, could be employed for studying solution 2D polymerisations. Moreover due to their structural similarity, anthraphane 1 (donor) could be co- crystallised with diazaanthraphane 3c (weak acceptor) to realise novel alternating 2D copolymers (Figure 4.23). Unfortunately, time limitations associated with this thesis prevented further exploration in this direction.
Figure 4.23 Hypothetical 2D alternating copolymer obtained by the co-crystallisation of anthraphane 1 and diazaantraphane 3.
211
DIAZAANTHRAPHANES
4.7 Experimental
4.7.1 Synthesis
Materials and methods
All reactions were carried out under nitrogen by using standard Schlenk techniques and dry solvents. DCM, dioxane and toluene were distilled by a solvent drying system from LC Technology Solutions
Inc. SP-105 under nitrogen atmosphere (H2O content < 5 ppm as determined by Karl-Fischer
[153] titration). Pd(PPh3)4 catalyst was freshly prepared following the literature procedure and stored in a glove-box in the dark under N2 at room temperature. All reagents were purchased from Acros, Aldrich or TCI, and used without further purification. Column chromatography for purification of the products was performed by using Merck silica gel Si60 (particle size 40-63 μm).
NMR was recorded on a Bruker AVANCE (1H: 300 MHz, 13C: 75 MHz) at room temperature or 70°C. The signal from the solvents was used as internal standard for chemical shift (1H: δ = 7.26 ppm, 13C: δ = 77.16 ppm for chloroform, 1H: δ = 6.00 ppm, 13C: δ = 73.78 ppm for 1,1,2,2-tetrachloroethane, 1H: δ = 5.33 ppm, 13C: δ = 54 ppm for dichloromethane, 1H: δ = 2.50 ppm, 13C: δ = 39.52 ppm for dimethyl sulfoxide, 19F: δ = -164.9 ppm for hexafluorobenzene). For centrifugation, a Hermle Z 320 K table centrifuge was used at 25 °C. When possible, proton and carbon signal assignments were performed with the help of 2D-NMR experiments such as COSY, HSQC and HMBC (spectra not shown).
High resolution mass spectroscopy (HRMS) analyses were performed by the MS-service of the laboratory for organic chemistry at ETH Zurich with spectrometers (ESI- and MALDI-ICR-FTMS: IonSpec Ultima Instrument). Either 3-hydroxypicolinic acid (3-HPA) or trans-2-[3-(4-tert-butylphenyl)- 2-methyl-2-propenylidene]malononitrile (DCTB) were used as matrix.
UV/Vis absorption spectra were recorded with a JASCO V-670 UV-Vis-NIR spectrophotometer using a quartz cell with a path length of 1 cm. Emission spectra were recorded with a Spex Fluorolog 2 spectrophotometer from Jobin Yvon (United Kingdom) using a quartz cell with a path length of 1 cm by diluting by a factor of 30-60 (depending on the compound) the solutions employed for the UV/Vis absorption measurements.
SC-XRD analysis was performed by the Small Molecule Crystallography Center (SMoCC) at ETH Zurich and at the SNBL beamline BM01A at the ESRF, Grenoble.
212
DIAZAANTHRAPHANES
Synthetic procedures
(Z)-Tetramethyl 2,2'-((2-methyl-1,3-phenylene)bis(azanediyl))difumarate 31a Dimethyl acetylenedicarboxylate 30a (25.00 g, 175.92 mmol, 2.05 eq) was dissolved in a solvent mixture of CHCl3 (10 mL) and MeOH (30 mL). The reaction mixture was cooled to 0°C with an ice bath and a solution of 2,6-diaminotoluene (10.48 g, 85.81 mmol, 1 eq) in MeOH (100 mL) was added dropwise over 10 min under N2 atmosphere. After addition, the yellow suspension was warmed up to room temperature and stirred for additional 16 h. The reaction mixture was then filtered and the obtained yellow solid was washed with MeOH until the filtrate resulted colourless. Drying on high vacuum afforded 31a (33.04 g, 81.36 mmol, 95%) as a pale yellow solid.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.53 (s, 2H-N); 6.99 (t, J = 7.97 Hz, 1H); 6.58 (d, J = 7.97 Hz, 2H); 5.43 (s, 2H); 3.75 (s, 6H); 3.64 (s, 6H); 2.32 (s, 3H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 170.30; 164.75; 149.05; 140.21; 126.13; 124.16; 118.93; 93.37; 52.80; 51.37; 12.42.
+ HRMS (FT-MALDI): m/z calcd for C19H22N2O8 [M-H] : 406.1371; found: 406.1371.
213
DIAZAANTHRAPHANES
Dimethyl 10-methyl-4,6-dioxo-1,4,6,9-tetrahydropyrido[3,2-g]quinoline-2,8-dicarboxylate 32a 31a (5.50 g, 13.54 mmol) was dissolved in 700 mL (19 mM) degassed diphenylether and the resulting solution was refluxed at 270°C for 1.5 h under N2 atmosphere. Upon cooling, a yellow precipitate appeared. Hexane was added to complete precipitation and the reaction mixture filtered. The obtained solid was thoroughly washed with hexane until the distinctive smell of Ph2O was gone. Drying on HV afforded pure 32a (4.24 g, 12.39 mmol, 92%) as a beige powder.
1H-NMR (300 MHz, TFA-d) δ/ppm: 9.79 (s, 1H); 7.83 (s, 2H); 4.22 (s, 6H); 3.16 (s, 3H).
13C-NMR (75.5 MHz, TFA-d) δ/ppm: 180.71; 163.03; 147.23; 141.42; 127.43; 123.60; 120.88; 108.83; 57.86; 12.00.
+ HRMS (FT-MALDI): m/z calcd for C17H15N2O6 [M-H] : 343.0925; found: 343.0925.
214
DIAZAANTHRAPHANES
Dimethyl 10-methyl-4,6-bis(((trifluoromethyl)sulfonyl)oxy)pyrido[3,2-g]quinoline-2,8-dicarboxylate 33a 32a (10.00 g, 29.21 mmol, 1 eq) was suspended in 300 mL dry DCM along with DMAP (1.43 g, 11.69 mmol, 0.4 eq) and 2,6-lutidine (10.20 mL, 87.64 mmol, 3 eq). The reaction mixture was cooled to 0°C with an ice-bath and triflic anhydride (13.82 mL, 87.64 mmol, 3 eq) was slowly added via syringe under inert atmosphere. After the addition, the brown-red reaction mixture was stirred 1 h at 0°C and then 6 h at room temperature. The solvent was evaporated in vacuo and the residue washed with warm MeOH to afford a yellow solid. After filtration, the solid was rinsed with MeOH and dried on HV to afford pure 33a as a bright yellow fluffy solid (16.24 g, 26.78 mmol, 92%).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.79 (s, 1H); 8.26 (s, 2H); 4.15 (s, 6H); 3.61 (s, 3H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 164.37; 153.93; 149.92; 146.59; 144.02; 122.31; 119.50 (q, JCF = 321.1 Hz); 112.48; 111.84; 53.87; 13.75.
19 F-NMR (282.5 MHz, CDCl3) δ/ppm: -75.90.
+ HRMS (FT-MALDI): m/z calcd for C19H13F6N2O10S2 [M-H] : 606.9910; found: 606.9910.
215
DIAZAANTHRAPHANES
Dimethyl 10-methyl-4,6-bis((trimethylsilyl)ethynyl)pyrido[3,2-g]quinoline-2,8-dicarboxylate 35 Ditriflate 33a (1.00 g, 1.65 mmol, 1 eq) was suspended in 10 mL dry dioxane along with trimethylsilylacetylene (0.6 mL, 4.12 mmol, 2.5 eq) and 2,6-lutidine (0.6 mL, 4.95 mmol, 3 eq). The suspension was degassed three times by freeze-pump-thaw cycles and then Pd(PPh3)4 (57 mg, 0.05 mmol, 0.03 eq) and CuI (9 mg, 0.05 mmol, 0.03 eq) were added with positive nitrogen pressure. After heating at 65°C in the dark for 14 h, the reaction mixture was cooled to room temperature and filtered through a pad of celite. The filtrate was concentrated to dryness and dissolved in a minimum amount of DCM. Methanol was added and the formed yellow precipitate was separated by filtration. Recrystallization from acetonitrile afforded 35 as yellow needles (622 mg, 1.24 mmol, 75 %).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.14 (s, 1H); 8.35 (s, 2H); 4.11 (s, 6H); 3.54 (s, 3H); 0.40 (s, 18H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 165.63; 148.22; 144.65; 142.29; 132.01; 127.38; 125.22; 121.54; 107.17; 100.42; 53.41; 13.20; 0.15.
+ HRMS (FT-MALDI): m/z calcd for C27H31N2O4Si2 [M-H] : 503.1817; found: 503.1816.
216
DIAZAANTHRAPHANES
Dimethyl 10-methyl-4,6-bis((triisopropylsilyl)ethynyl)pyrido[3,2-g]quinoline-2,8-dicarboxylate 36 Ditriflate 33a (1.00 g, 1.65 mmol, 1 eq) was suspended in 10 mL dry dioxane along with trimethylsilylacetylene (0.9 mL, 4.12 mmol, 2.5 eq) and 2,6-lutidine (0.6 mL, 4.95 mmol, 3 eq). The suspension was degassed three times by freeze-pump-thaw cycles and then Pd(PPh3)4 (57 mg, 0.05 mmol, 0.03 eq) and CuI (9 mg, 0.05 mmol, 0.03 eq) were added with positive nitrogen pressure. After heating at 75°C in the dark for 24 h (TLC check to ensure complete di-substitution), the reaction mixture was cooled to room temperature and filtered through a pad of celite. The filtrate was concentrated to dryness and subjected to silica gel column chromatography (10% EtOAc in hexane) to afford 36 as a bright yellow solid (899 mg, 1.34 mmol, 81 %).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.11 (s, 1H); 8.35 (s, 2H); 4.12 (s, 6H); 3.56 (s, 3H); 1.33-1.10 (m, 42H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 165.66; 148.03; 144.69; 142.32; 132.38; 127.35; 126.12; 121.23; 104.73; 102.74; 53.38; 18.91; 13.25; 11.68.
+ HRMS (FT-MALDI): m/z calcd for C39H55N2O4Si2 [M-H] : 671.3695; found: 671.3695.
217
DIAZAANTHRAPHANES
Dimethyl 10-methyl-4,6-bis(phenylethynyl)pyrido[3,2-g]quinoline-2,8-dicarboxylate 37 Ditriflate 33a (500 mg, 0.82 mmol, 1 eq) was suspended in 20 mL dry dioxane along with phenylacetylene (0.2 mL, 1.81 mmol, 2.2 eq) and 2,6-lutidine (0.3 mL, 2.47 mmol, 3 eq). The suspension was degassed three times by freeze-pump-thaw cycles and then Pd(PPh3)4 (29 mg, 0.02 mmol, 0.03 eq) and CuI (5 mg, 0.02 mmol, 0.03 eq) were added with positive nitrogen pressure. The reaction mixture was heated under N2 to 65°C for 16 h, during which time a yellow precipitation appeared. After cooling to room temperature the solid was separated by filtration and washed with 40 mL MeOH. Recrystallization from boiling o-dichlorobenzene afforded 12 as yellow needles (343 mg, 0.67 mmol, 81%).
1 H-NMR (300 MHz, C2D2Cl4) δ/ppm: 9.39 (s, 1H); 8.37 (s, 2H); 7.64 (d, J = 7.1 Hz, 4H); 7.47 (t, J = 7.5 Hz, 2H); 7.30 (t, J = 7.6 Hz, 4H); 4.12 (s, 6H); 3.58 (s, 3H).
13C-NMR spectrum could not be measured even at high temperatures due to the insolubility of the compound.
+ HRMS (FT-MALDI): m/z calcd for C33H23N2O4 [M-H] : 511.1652; found: 511.1652.
218
DIAZAANTHRAPHANES
Dimethyl 4,6-dibromo-10-methylpyrido[3,2-g]quinoline-2,8-dicarboxylate 38 Pyridoquinolone 32a (2.20 g, 6.43 mmol, 1 eq) was suspended in 220 mL dry THF. Phosphoryl bromide (5.53 g, 19.28 mmol, 3 eq) was added in one portion and the reaction mixture was refluxed under N2 for 16 h, during which the colour changed from orange to yellow. After cooling the suspension was filtered and the solid washed with 100 mL MeOH. Drying on HV afforded 38 as a bright yellow solid (2.79 g, 5.95 mmol, 93%).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.09 (s, 1H); 8.56 (s, 2H); 4.12 (s, 6H); 3.56 (s, 3H).
13 C-NMR (75.5 MHz, C2D2Cl4) δ/ppm: 164.41; 148.29; 144.84; 142.94; 135.93; 127.81; 124.95; 123.70; 53.13; 13.48.
+ HRMS (FT-MALDI): m/z calcd for C17H13Br2N2O4 [M-H] : 466.9237; found: 466.9235.
219
DIAZAANTHRAPHANES
Diethyl acetylenedicarboxylate 30b Acetylenedicarboxylic acid 29 (25.00 g, 0.22 mol, 1 eq) was dissolved in a mixture of 150 mL ethanol and 150 mL benzene. After addition of p-toluenesulfonic acid monohydrate (4.17 g, 0.02 mol, 0.1 eq) the solution was refluxed for 16 h, with occasional emptying of the trap. After quantitative recovery of water (~7.9 mL), the reaction mixture was cooled to room temperature and the solvent removed in vacuo. The oily residue was subjected to vacuum distillation (p = 0.5 mbar, T = 70°C) to yield 30b (28.7 g, 0.74 mol, 77%) as a colorless oil.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.29 (q, J = 7.1 Hz, 4H); 1.32 (t, J = 7.1 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 151.93; 74.78; 63.13; 14.02.
+ HRMS (FT-MALDI): m/z calcd for C8H10NaO4 [M-Na] : 193.0471; found: 193.0473.
Dipropyl acetylenedicarboxylate 30c Acetylenedicarboxylic acid 29 (25.00 g, 0.22 mol, 1 eq) and 1-propanol (36 mL, 0.48 mol, 2.2 eq) were dissolved in 350 mL benzene. p-Toluenesulfonic acid monohydrate (4.17 g, 0.02 mol, 0.1 eq) was added and the solution was refluxed for 16 h until quantitative recovery of water (~7.9 mL), the reaction mixture was cooled to room temperature and the solvent removed in vacuo. The oily residue was subjected to vacuum distillation (p = 0.15 mbar, T = 73°C) to yield 30c (37.80 g, 0.19 mol, 87%) as a colorless oil.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 4.20 (t, J = 6.7 Hz, 4H); 1.72 (hex, J = 7.2 Hz, 4H); 0.97 (t, J = 7.4 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 152.01; 74.75; 68.54; 21.77; 10.27.
+ HRMS (FT-MALDI): m/z calcd for C10H15O4 [M-H] : 199.0965; found: 199.0965.
220
DIAZAANTHRAPHANES
(Z)-Tetraethyl 2,2'-((2-methyl-1,3-phenylene)bis(azanediyl))difumarate 31b
Diester 30b (25.00 g, 146.92 mmol, 2 eq) was dissolved in a solvent mixture of CHCl3 (10 mL) and MeOH (20 mL). The reaction mixture was cooled to 0°C with an ice bath and a solution of 2,6- diaminotoluene (8.97 g, 73.46 mmol, 1 eq) in MeOH (65 mL) was added dropwise over 10 min under
N2 atmosphere. After addition, the suspension was stirred for additional 16 h at room temperature. The reaction mixture was then filtered and the obtained solid was washed with MeOH until the filtrate resulted colourless. Drying on high vacuum afforded 31b (26.16 g, 56.56 mmol, 77%) as a pale yellow powder.
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.52 (s, 2H-N); 6.97 (t, J = 8.0 Hz, 1H); 6.61 (d, J = 7.9 Hz, 2H); 5.41 (s, 2H); 4.21 (q, J = 7.1 Hz, 4H); 4.10 (q, J = 7.1 Hz, 4H); 2.32 (s, 3H); 1.31 (t, J = 7.1 Hz, 6H); 1.07 (t, J = 7.1 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 169.88; 164.18; 149.37; 140.18; 125.88; 124.20; 119.10; 93.48; 61.95; 60.00; 14.43; 13.70; 12.34.
+ HRMS (FT-MALDI): m/z calcd for C23H31N2O8 [M-H] : 463.2075; found: 463.2075.
221
DIAZAANTHRAPHANES
(Z)-Tetrapropyl 2,2'-((2-methyl-1,3-phenylene)bis(azanediyl))difumarate 31c
Diester 30c (28.27 g, 142.60 mmol, 2 eq) was dissolved in a solvent mixture of CHCl3 (10 mL) and MeOH (20 mL). The reaction mixture was cooled to 0°C with an ice bath and a solution of 2,6- diaminotoluene (8.71 g, 71.30 mmol, 1 eq) in MeOH (60 mL) was added dropwise over 10 min under
N2 atmosphere. After addition, the suspension was stirred for additional 16 h at room temperature. The reaction mixture was then filtered and the obtained solid was washed with MeOH until the filtrate resulted colourless. Drying on high vacuum afforded 31c (33.28 g, 64.17 mmol, 90%) as a pale yellow powder, which if needed can be recrystallized from MeOH to form yellow square plates (95% recovery).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 9.53 (s, 2H), 6.96 (t, J = 8.0 Hz, 1H), 6.60 (d, J = 8.0 Hz, 2H), 5.42 (s, 2H), 4.11 (t, J = 6.7 Hz, 4H), 4.00 (t, J = 6.7 Hz, 4H), 2.33 (s, 3H), 1.70 (hex, J = 7.1 Hz, 4H), 1.47 (hex, J = 7.1 Hz, 4H), 0.97 (t, J = 7.4 Hz, 6H), 0.77 (t, J = 7.4 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 170.08; 164.45; 149.43; 140.33; 126.07; 123.99; 118.96; 93.64; 67.68; 65.78; 22.26; 21.67; 12.40; 10.58; 10.31.
+ HRMS (FT-MALDI): m/z calcd for C27H39N2O8 [M-H] : 519.2701; found: 519.2700.
222
DIAZAANTHRAPHANES
Diethyl 10-methyl-4,6-dioxo-1,4,6,9-tetrahydropyrido[3,2-g]quinoline-2,8-dicarboxylate 32b 31b (10.00 g, 21.62 mmol) was dissolved in 500 mL degassed diphenylether (43 mM) and the resulting solution was refluxed at 270°C for 1.5 h under N2 atmosphere. Upon cooling a yellow precipitate appears. Hexane was added to complete precipitation and the reaction mixture filtered.
The obtained solid was thoroughly washed with hexane until the distinctive smell of Ph2O was gone. Drying on HV afforded pure 32b (7.26 g, 19.60 mmol, 91%) as a beige powder.
1H-NMR (300 MHz, TFA-d) δ/ppm: 9.81 (s, 1H); 7.87 (s, 2H); 4.70 (q, J = 6.8 Hz, 4H); 3.18 (s, 3H); 1.52 (t, J = 7.2 Hz, 6H).
13C-NMR (75.5 MHz, TFA-d) δ/ppm: 180.29; 162.28; 147.23; 141.07; 127.10; 123.26; 120.55; 108.41; 69.04; 14.55; 11.66.
+ HRMS (FT-MALDI): m/z calcd for C19H19N2O6 [M-H] : 371.1238; found: 371.1236.
Dipropyl 10-methyl-4,6-dioxo-1,4,6,9-tetrahydropyrido[3,2-g]quinoline-2,8-dicarboxylate 32c 31c (12.21 g, 23.54 mmol) was dissolved in 200 mL degassed diphenylether (118 mM) and the resulting solution was refluxed at 270°C for 1.5 h under N2 atmosphere. Upon cooling a yellow precipitate appears. Hexane was added to complete precipitation and the reaction mixture filtered.
The obtained solid was thoroughly washed with hexane until the distinctive smell of Ph2O was gone. Drying on HV afforded pure 32c (8.44 g, 21.19 mmol, 90%) as a pale yellow solid. The compound can be recrystallized from i-PrOH if needed to afford yellow needles (92% recovery)
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.92 (s, 1H); 8.72 (s, 2H-N); 6.72 (s, 2H); 4.40 (t, J = 6.7 Hz, 4H); 2.54 (s, 3H); 1.84 (hex, J = 7.1 Hz, 4H); 1.06 (t, J = 7.4 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 179.63; 162.76; 139.16; 137.05; 124.16; 122.25; 110.43; 109.79; 69.09; 21.97; 10.43; 9.99.
+ HRMS (FT-MALDI): m/z calcd for C21H23N2O6 [M-H] : 399.1551; found: 399.1551.
223
DIAZAANTHRAPHANES
Diethyl 10-methyl-4,6-bis(((trifluoromethyl)sulfonyl)oxy)pyrido[3,2-g]quinoline-2,8-dicarboxylate 33b 32b (11.20 g, 30.24 mmol, 1 eq) was suspended in 200 mL dry DCM along with DMAP (1.48 g, 12.10 mmol, 0.4 eq) and 2,6-lutidine (10.50 mL, 90.72 mmol, 3 eq). The reaction mixture was cooled to 0°C with an ice-bath and triflic anhydride (15.20 mL, 90.72 mmol, 3 eq) was slowly added via syringe under inert atmosphere. After the addition, the red reaction mixture was stirred 1 h at 0°C and then 6 h at room temperature. The solvent was evaporated in vacuo and the residue washed with MeOH to afford a yellow solid. After filtration, the solid was rinsed with MeOH and dried on HV to afford pure 33b as a bright yellow fluffy solid (18.22 g, 28.71 mmol, 95%).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.78 (s, 1H); 8.25 (s, 2H); 4.61 (t, J = 7.1 Hz, 4H); 3.61 (s, 3H); 1.55 (t, J = 7.1 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 163.84; 153.89; 150.23; 146.60; 144.00; 122.23; 118.80 (q, JCF = 320.5 Hz); 112.45; 111.73; 63.17; 14.41; 13.70.
19 F-NMR (282.5 MHz, CDCl3) δ/ppm: -75.93.
+ HRMS (FT-MALDI): m/z calcd for C21H17F6N2O10S2 [M-H] : 635.0223; found: 635.0224.
224
DIAZAANTHRAPHANES
Dipropyl 10-methyl-4,6-bis(((trifluoromethyl)sulfonyl)oxy)pyrido[3,2-g]quinoline-2,8-dicarboxylate 33c 32c (17.60 g, 44.18 mmol, 1 eq) was suspended in 300 mL dry DCM along with DMAP (2.16 g, 17.67 mmol, 0.4 eq) and 2,6-lutidine (15.45 mL, 132.53 mmol, 3 eq). The reaction mixture was cooled to 0°C with an ice-bath and triflic anhydride (22.30 mL, 132.53 mmol, 3 eq) was slowly added via syringe under inert atmosphere. After the addition, the red reaction mixture was stirred 1 h at 0°C and then 6 h at room temperature. The solvent was evaporated in vacuo and the residue washed with warm MeOH to afford a yellow solid. After filtration, the solid was rinsed with MeOH and dried on HV to afford pure 33c as a bright yellow fluffy solid (28.68 g, 43.29 mmol, 98%).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.78 (s, 1H); 8.24 (s, 2H); 4.50 (t, J = 6.7 Hz, 4H); 3.60 (s, 3H); 1.94 (hex, J = 7.1 Hz, 4H); 1.13 (t, J = 7.4 Hz, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 163.88; 153.90; 150.22; 146.60; 143.99; 122.23; 118.80 (q, JCF = 320.6 Hz); 112.42; 111.71; 68.54; 22.21; 13.61; 10.57.
19 F-NMR (282.5 MHz, CDCl3) δ/ppm: -75.93.
+ HRMS (FT-MALDI): m/z calcd for C23H21F6N2O10S2 [M-H] : 663.0536; found: 663.0537.
225
DIAZAANTHRAPHANES
Ethyl ester Precursor 34b An excess amount of 33b (2.50 g, 3.94 mmol, 6 eq) was suspended in 25 mL dry dioxane with 1,3,5- triethynylbenzene (99 mg, 0.66 mmol, 1 eq) and 2,6-lutidine (0.46 mL, 3.94 mmol, 6 eq) and the reaction mixture was degassed three times by freeze-pump-thaw cycles. Pd(PPh3)4 (38 mg, 0.03 mmol, 0.05 eq) and CuI (6 mg, 0.03 mmol, 0.05 eq) were added, the reaction mixture was degassed two more times by freeze-pump-thaw cycles and then heated in the dark at 70°C for 2d. After cooling, the reacton mixture was filtered and the obtained yellow solid was washed with 25 mL dioxane and 25 mL acetonitrile to eliminate any excess of starting material (checked by TLC) and then MeOH until the filtrate resulted colourless. Drying on HV afforded 34b in purities high enough for the next step (632 mg, 0.39 mmol, 60%). To recover the excess of educt, MeOH was added to the filtrate and the obtained precipitate was dissolved in little DCM and passed through a short silica plug (eluent 20% ethyl acetate in hexanes) to obtain 33b (1.09 g, 1.72 mmol, 87 %).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.75 (s, 1H); 8.09 (s, 1H); 7.92 (s, 1H); 7.90 (s, 1H); 4.75-4.48 (m, 4H); 3.39 (s, 3H); 1.70-1.59 (m, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 163.81 (two carbonyl signals overlapping); 153.74; 149.00;
148.83; 145.41; 144.07; 143.44; 136.41; 130.71; 126.91; 123.56; 123.47; 121.17; 118.93 (q, JCF = 320.5 Hz); 115.69; 111.43; 98.78; 86.13; 62.93; 62.55; 14.50; 14.43; 13.10.
19 F-NMR (282.5 MHz, CDCl3) δ/ppm: -76.07.
+ HRMS (FT-MALDI): m/z calcd for C72H52F9N6O21S3 [M-H] : 1603.2198; found: 1603.2206.
226
DIAZAANTHRAPHANES
Propyl ester Precursor 34c An excess amount of 33c (5.00 g, 7.55 mmol, 6 eq) was suspended in 50 mL dry dioxane with 1,3,5- triethynylbenzene (189 mg, 1.26 mmol, 1 eq) and 2,6-lutidine (0.88 mL, 7.55 mmol, 6 eq) and the reaction mixture was degassed three times by freeze-pump-thaw cycles. Pd(PPh3)4 (73 mg, 0.06 mmol, 0.05 eq) and CuI (12 mg, 0.06 mmol, 0.05 eq) were added, the reaction mixture was degassed two more times by freeze-pump-thaw cycles and then heated in the dark at 70°C for 2d. After cooling, the reacton mixture was filtered and the obtained yellow solid was washed with 50 mL dioxane and 50 mL acetonitrile to eliminate any excess educt (checked by TLC) and then MeOH until the filtrate resulted colourless. Drying on HV afforded 34c in purities high enough for the next step (1.51 g, 0.89 mmol, 71%). To recover the excess of educt, MeOH was added to the filtrate and the obtained precipitate was dissolved in little DCM and passed through a short silica plug (eluent 20% ethyl acetate in hexanes) to obtain 33c (2.32 g, 3.51 mmol, 93 %).
1 H-NMR (300 MHz, CDCl3) δ/ppm: 8.90 (s, 1H); 8.15 (s, 1H); 8.05 (s, 1H); 7.99 (s, 1H); 4.59-4.40 (m, 4H); 2.09-1.92 (m, 4H); 1.30-1.10 (m, 6H).
13 C-NMR (75.5 MHz, CDCl3) δ/ppm: 163.79; 163.67; 153.71; 148.98; 148.55; 145.32; 143.87; 143.40;
136.40; 130.51; 126.73; 123.53; 123.21; 121.14; 118.93 (q, JCF = 320.8 Hz); 115.59; 111.43; 98.78;
86.09; 68.32; 67.90; 22.31; 22.22; 12.91; 10.70 (two aliphatic CH3 signals overlapping).
19 F-NMR (282.5 MHz, CDCl3) δ/ppm: -76.07.
+ HRMS (FT-MALDI): m/z calcd for C78H64F9N6O21S3 [M-H] : 1687.3137; found: 1687.3118.
227
DIAZAANTHRAPHANES
Ethyl ester Diazaanthraphane 3b 34b (535 mg, 0.33 mmol, 1 eq) was suspended in 330 mL dry toluene (1.00 mM) with 1,3,5- triethynylbenzene (50 mg, 0.33 mmol, 1 eq) and 2,6-lutidine (3.9 mL, 33.68 mmol, 100 eq). The reaction mixture was degassed by three freeze-pump-thaw cycles, then Pd(PPh3)4 (231 mg, 0.20 mmol, 0.6 eq) was added with positive N2 pressure. The reaction mixture was degassed again twice and heated in the dark at 80°C under argon for 7 days. After cooling, the reaction mixture was filtered and the obtained brown solid was washed with 20 mL toluene and then MeOH till the filtrate resulted colourless. The solid was then suspended in 100 mL warm chloroform and filtered through a pad of celite to obtain a bright yellow solution. The celite was then rinsed with copious amounts of chloroform in order to extract more product. The obtained solution was concentrated till saturation and put in the fridge to let the product precipitate as a bright yellow fluffy solid. Filtration afforded monomer 3b (87 mg, 0.07 mmol, 20%).
1 H-NMR (300 MHz, C2D2Cl4) δ/ppm: 9.23 (s, 1H); 8.36 (s, 2H); 7.95 (s, 2H); 4.60 (t, J = 7.1 Hz, 12H); 3.62 (s, 3H); 1.60 – 1.49 (m, 3H, superimposed with residual water peak).
13C-NMR spectrum could not be measured even at high temperatures due to the insolubility of the compound.
+ HRMS (FT-MALDI): m/z calcd for C81H54N6NaO12 [M-Na] : 1325.3692; found: 1325.3687.
228
DIAZAANTHRAPHANES
Propyl ester Diazaanthraphane 3c 34c (1.00 g, 0.59 mmol, 1 eq) was suspended in 390 mL dry toluene (1.50 mM) with 1,3,5- triethynylbenzene (89 mg, 0.59 mmol, 1 eq) and 2,6-lutidine (6.9 mL, 59.3 mmol, 100 eq). The reaction mixture was degassed by three freeze-pump-thaw cycles, then Pd(PPh3)4 (411 mg, 0.35 mmol, 0.6 eq) was added with positive N2 pressure. The reaction mixture was degassed again twice and heated in the dark at 80°C under argon for 7 days. After cooling, the reaction mixture was filtered and the obtained brown solid was washed with 20 mL toluene and then MeOH till the filtrate resulted colourless. The solid was then suspended in 100 mL warm chloroform and filtered through a pad of celite to obtain a bright yellow solution. The celite was then rinsed with copious amounts of chloroform in order to extract more product. The obtained solution was concentrated till saturation and put in the fridge to let the product precipitate as a bright yellow fluffy solid. Filtration afforded monomer 3c (263 mg, 0.19 mmol, 32%).
1 H-NMR (300 MHz, C2D2Cl4) δ/ppm: 9.20 (s, 1H); 8.35 (s, 2H); 7.94 (s, 2H); 4.47 (t, J = 6.6 Hz, 4H); 3.60 (s, 3H); 1.93 (hex, J = 7.0 Hz, 4H); 1.13 (t, J = 7.7 Hz, 6H).
13C-NMR spectrum could not be measured even at high temperatures due to the insolubility of the compound.
+ HRMS (FT-MALDI): m/z calcd for C87H66N6NaO12 [M-Na] : 1409.4631; found: 1409.4614.
229
DIAZAANTHRAPHANES
Solution UV/Vis spectroscopy UV/Vis absorption and fluorescence spectra were measured for ditriflate 33c, precursor 34c, model compound 37 and diazaanthraphane 3c (λ = 365 nm for excitation wavelength). The measurements were performed at room temperature in chloroform with the following concentrations: 15 μM for absorption and 0.5 μM for emission.
Solid-state UV/Vis spectroscopy UV/Vis absorption spectra were measured on a JASCO V-660 UV-VIS-NIR Spectrophotometer (Jasco Inc., Tokyo, Japan) equipped with a 150 mm integrating sphere (ILN-725, Jasco Inc., Tokyo, Japan) using a powder holder against a white barium sulphate standard.
Thermogravimetric analysis (TGA) Thermogravimetric analysis was performed with a TGA Q500 (TA Instrument, USA) under nitrogen using a platinum pan.
Optical (OM) and polarized microscope (POM) OM and POM was carried out with a Leica DMRX polarizing microscope equipped with a Leica DFC 480 Camera (Leica Microsystems, Heerbrugg, Switzerland) or with a Leica DM4000M optical microscope (Leica Microsystems GmbH, Wetzlar, Germany).
Electrostatic surface potential (ESP) calculations ESP calculations were performed with software Spartan (Wavefunction, Inc.). EPS and MEP values were calculated by the semi-empirical PM3 method for the fragments of cyclophane 1 and 4. DFT calculations were also performed and agree in trend with the PM3 method. Semi-empirical methods such as PM3 are well suited to represent the electrostatic potential surface of planar aromatic rings[132].
Photoirradiation An in-house built cylindrical photoreactor equipped with 16 x 2.8 W high power green LEDs with λ =530 nm (from Seoul Semiconductors) each at 22 lm was used as irradiation source. To prevent solvent loss from the crystals, they were irradiated directly in their mother solution.
230
DIAZAANTHRAPHANES
4.7.2 SC-XRD analysis
General Crystals of compound 37 were measured on a Bruker Apex-II Duo diffractometer using a graphite- monochromated sealed-tube Mo-Kα radiation (λ = 0.71073 Å). Crystals were kept at 100 K during measurements using an Oxford Cryosystems Cryostream 700 cooler. Data for crystals of compound 3c from 1,3-dimethoxybenzene were collected on the PILATUS@SNBL setup on SNBL beamline BM01A at the ESRF, Grenoble (http://www.esrf.eu/UsersAndScience/Experiments/CRG/BM01/bm01-a). Data were collected at 100K, using 0.6946Å radiation, on a Dectris Pilatus2M area detector in combination with a custom kappa stage diffractometer. The help of Dr. Phil Pattison is greatly acknowledged.
Data were integrated using SAINT and corrected for absorption effects using the multi-scan method (SADABS)[154]. The structures were solved using SHELXS or SHELXT[155] and refined by full-matrix least- squares analysis (SHELXL)[155] using the program package OLEX2[156]. In the case of compound 3c, peak indexing and integration was done using the CrysAlisPro software package[191], followed by absorption correction with SADABS. Unless otherwise indicated, all non-hydrogen atoms were refined anisotropically. All hydrogen atoms were constrained to ideal geometries and refined with fixed isotropic displacement parameters (in terms of a riding model). CCDC 1055088 (37) and 1505880 (3c, 1,3-dimethoxybenzene) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44(1223)-336-033; e-mail: [email protected]), or via https://www.ccdc.cam.ac.uk/getstructures.
Crystalline samples generally suffered from severe solvent disorder (in some cases with additional disorder in the cyclophane) and were small, both effects leading to very weak diffraction. Usually microfocus Cu sources had to be used and long exposure times had to be chosen, raising additional problems with sample icing. Only substandard resolutions could be achieved and these structures should not be assessed applying the quality criteria of routine small molecule structures. Specifically, alerts rising due to low angular resolution and low completeness in the highest resolution shells are to be expected. We believe the structure quality is sufficient for the discussion of packing motifs, but certainly bond distances and angles will be biased. Many crystals were tested and the best ones chosen for measurement.
231
DIAZAANTHRAPHANES
Compound 37
Figure 4.24. ORTEP diagrams of compound 37 (50% probability) from different views (a-axis, b-axis and c-axis from left to right).
Sample and crystal data CCDC depositary number 1055088
Empirical formula C33H22N2O4 Formula weight 510.52 Temperature/K 100.0(2) Color/shape Clear yellow rod Crystal system orthorhombic
Space group P21212 a/Å 19.2379(9) b/Å 9.4314(4) c/Å 13.7129(7) α/° 90 β/° 90 γ/° 90 Volume/Å3 2488.1(2) Z 4 3 ρcalcg/cm 1.363 μ/mm-1 0.090 F(000) 1064.0 Crystal size/mm3 0.1 × 0.09 × 0.02 Radiation MoKα (λ = 0.71073) 2Θ range for data collection/° 3.648 to 55.04 Index ranges -24 ≤ h ≤ 19, -12 ≤ k ≤ 12, -17 ≤ l ≤ 17 Reflections collected 20836
Independent reflections 5728 [Rint = 0.0586, Rsigma = 0.0632] Data/restraints/parameters 5728/0/359 Goodness-of-fit on F2 0.970
Final R indexes [I>=2σ (I)] R1 = 0.0447, wR2 = 0.0864
Final R indexes [all data] R1 = 0.0810, wR2 = 0.0983 Largest diff. peak/hole / e Å-3 0.22/-0.23 Flack parameter -0.7(6)
232
DIAZAANTHRAPHANES
Diazaanthaphane 3c
Figure 4.25. ORTEP diagrams of compound 3c in the 1,3-dimethoxybenzene solvate (50% n probability). Severe disorder in all CO2 Pr groups. One disordered solvent molecule could not be split into separate positions (hence the rather large ellipsoids), a second one could not be assigned at all and had to be masked from the density, using the procedure implemented in Olex2.
Sample and crystal data CCDC depositary number 1505880
Empirical formula C95H76N6O14 Formula weight 1525.61 Temperature/K 100 Color/shape Pale orange hexagonal platelet Crystal system triclinic Space group P-1 a/Å 17.153(2) b/Å 20.473(2) c/Å 21.3466(17) α/° 99.037(8) β/° 112.984(9) γ/° 113.747(11) Volume/Å3 5861.1(12) Z 2 3 ρcalcg/cm 0.864 μ/mm-1 0.056 F(000) 1600.0 Crystal size/mm3 0.08 × 0.08 × 0.02 Radiation Synchrotron (λ = 0.6946) 2Θ range for data collection/° 3.8 to 45.904 Index ranges -19 ≤ h ≤ 19, -22 ≤ k ≤ 22, -23 ≤ l ≤ 23 Reflections collected 36148
Independent reflections 13587 [Rint = 0.0614, Rsigma = 0.0936] Data/restraints/parameters 13587/1649/1236 Goodness-of-fit on F2 1.173
Final R indexes [I>=2σ (I)] R1 = 0.1391, wR2 = 0.3579
Final R indexes [all data] R1 = 0.2001, wR2 = 0.3910 Largest diff. peak/hole / e Å-3 0.59/-0.45
233
CONCLUSIONS AND OUTLOOK
5. Conclusions and Outlook
This thesis was meant to explore and find new potential monomers for the synthesis of 2DPs. In particular, it was desirable to have versatile monomeric systems that could be employed for different approaches towards 2DPs such as the single crystal and air/water interface approach. Synthesis of 2DPs by these two methods can still be very challenging and the key for success is usually intimately bound with the monomer design. This work was not only aimed at finding new successful systems, but also to highlight the challenges and difficulties that may arise during the synthesis of a 2DP, be it through the single crystal method, or through the air/water interfacial approach.
An entire new class of potential monomers, the anthraphanes 1, 2a-b and 3b-e, was therefore developed and their applicability to the synthetic methods towards 2DPs was assessed (Figure 5.1). It has to be stated that, at the beginning of this work, the focus was put on the single crystal approach, for which anthraphane 1 was designed and synthesised; the development of the other monomers, the amphiphilic anthraphanes 2a-b and the diazaanthraphanes 3b-e, is the product of a learning curve, an attempt to tackle the challenges encountered with 1. 2a-b and 3b-e are the result of a careful structural engineering of 1, to increase the chances of obtaining novel 2DPs.
Figure 5.1. The anthraphane class of monomers developed during this thesis. 234
CONCLUSIONS AND OUTLOOK
As stated previously, anthraphane 1, a hydrocarbon cyclophane with D3h-symmetry bearing photoreactive anthracene units was synthesised as potential monomer for the topochemical synthesis of 2D polymers in single crystals. This method requires the formation of lamellar single crystals, in whose layers, all the anthracene moieties must be engaged in mutual ftf stacking. However, forcing a molecule into a determined crystal packing can be very difficult and can require a screening procedure. Anthraphane was therefore systematically crystallised from 21 different solvents, aiming at an exclusive ftf-packing. Five main packing motifs could be identified, but unfortunately not the desired one: it was found that anthraphane seems predisposed to avoid ftf interactions due to its negatively charged electrostatic potential surface on the anthracene units, which prefer etf interactions due to the favourable orientation of their quadrupolar moments. This study highlighted one of the potential limitations of the single crystal approach: the inability of systematically forcing a molecule to pack in the desired manner, which can result in a tedious trial and error procedure.
The experience gathered with 1, arose questions if an all-ftf packing could be realised with a different method, such as at the air/water interface. The robust synthesis developed for 1 was therefore modified to produce the amphiphilic anthraphanes anthraphane-tri(OMe) 2a and anthraphane-tri(DEGME) 2b. It was highlighted how tuning the amphiphilicity of a monomer is crucial for the air/water interface approach: the low amphiphilicity of 2a, prevented its spreading at the interface, while the highly amphiphilic 2b easily formed monolayers. Photopolymerisation of these highly fluorescent monolayers of 2b resulted into a mechanically coherent two-dimensional molecular network, able to span over micrometer-sized holes without collapsing under its own weight. This covalent monolayer sheet satisfies some criteria for the classification as 2DP but its internal structure still has to be elucidated. Preliminary results tentatively suggest the formation of an all-ftf packing at the air/water interface, which would eventually result into a polymer with a hexagonal honeycomb lattice; however, only characterisation by x-ray techniques (GISAXS), electron diffraction or STM will eventually shed light on the internal structure of this polymer. Upon confirmation of crystallinity, polyanthraphane-tri(DEGME) will be regarded as a novel 2DP. Potential applications for this porous polymer include ultra-thin membranes for gas permeation and separation, and rewritable optical data storage material systems.
Further structural engineering of 1, resulted in a new set of monomers, the diazaanthraphanes 3b-e. These monomers were designed with the intention of addressing the main problem encountered with the single crystal approach, namely the issue of achieving the desired packing suitable for a topochemical 2D polymerisation, without resorting to a trial-and-error screening procedure. The monomers were also equipped with functional groups, to explore the
235
CONCLUSIONS AND OUTLOOK possibility of new synthetic methods towards 2DPs such as pre-organisation in smectic phases and to open the possibility of post-polymerisation modifications. Crystallisation of 3b and 3c revealed that the monomers indeed packed in the desired all-ftf packing, forming a porous hexagonal honeycomb structure; but the most important finding was that this kind of packing could be obtained independently of the solvent used for crystallisation. The biggest limitation of the single crystal approach was therefore eliminated by the careful monomer design. The ester side-chains introduced for PPM, prevented however a proper stacking of the layers in the single crystals, compromising their quality. Unfortunately, this phenomenon could not be anticipated. The crystals were nevertheless irradiated, whereupon, a partial degree of polymerisation was obtained according to fluorescence measurements. Structural analysis of the irradiated crystals could not be performed; however, by exfoliating them, one could expect to obtain single sheets entities. If structural analysis of such sheets by for instance electron diffraction or STM would confirm periodicity, then a novel polydiazaanthraphane 2D polymer would be obtained.
To conclude, we believe that this work offers an enormous potential for the future development of the field of 2DPs: first of all, new polymers have been prepared, which still await structural analysis but are nevertheless promising candidates as 2DPs; secondly, a collection of new monomers was synthesised, which allowed us to explore the challenges and limitations of the current approaches towards 2DPs. We feel that both these developments have set the foundations for a bright outlook, which will be discussed in the next paragraphs.
Anthraphane 1 - A suitable solvent for crystallisation that leads to an all-ftf packing of 1 is still to be found; It is however possible that by continuing the screening procedure, the right solvate will be obtained leading to a novel hydrocarbon polyanthraphane 2DP. In this regard, co-crystallisation with a suitable guest
molecule could help in achieving the packing. - The donor character of 1 opens the possibility of co- crystallisation with a structurally related acceptor, such as diazaanthraphanes 3b-c. This would result in the first ever 2D alternating copolymer.
236
CONCLUSIONS AND OUTLOOK
Anthraphane-tri(OMe) 2a - This monomer resulted insufficiently amphiphilic for the air/water interface approach. It could however be a possible candidate for the single crystal approach. In the nitrobenzene solvate, a SCSC photoreaction resulted in a 1D polymer (Figure 6.3). The proper solvent could result in an all-ftf packing, however a trial-and-error screening procedure would be needed. - Co-crystallisation with acceptors 3b-c to achieve 2D alternating copolymers could also be possible.
Anthraphane-tri(DEGME) 2b - This monomer was spread at the air/water interface and photopolymerised to produce a mechanically coherent covalent monolayer. The internal structure of the polymer will have to be investigated (by STM or GISAXS) to confirm its crystallinity. - 2b forms highly fluorescent monolayers; polymerisation results in the quenching of the fluorescence, which can be in turn recovered by a thermally induced depolymerisation. The ON/OFF switching of the fluorescence could make monolayers of 2b useful as rewritable optical data storage systems. - Polymer monolayers obtained from 2b are intrinsically porous (Ø approximately 1.4 nm) and could therefore be used as ultra-thin membranes for gas permeation and separation.
Diazaanthraphanes 3b-c (R = ethyl or n-propyl) - These monomers packed in the all-ftf packing, but the crystals were however of poor quality due to the weak interaction between the layers. Exfoliation of the irradiated crystals should be performed to see if single layers can be obtained, which could be structurally analysed by for instance x-ray diffraction. - It is possible that a suitable solvent for crystallisation might be able to help in keeping the layers in the crystals together.
- Co-crystallisation with donors 1 or 2a could be possible.
237
CONCLUSIONS AND OUTLOOK
Diazaanthraphanes 3d-e (R = n-hexyl or n-dodecyl) - These monomers, due to their high solubility and inability to crystallise, could be used as starting point for investigating 2D polymerisations in solutions.
Finally, we would like to propose other feasible structures based on the diazaantraphane motif (Figure 5.2), that could be of potential interest for the field and lead to the creation of new 2DPs. 3f and 3g are meant to tackle to problems encountered with 3b-c, namely the weak interactions between the layers in the crystals. By shortening the ester chains to the minimum or even eliminating them completely: an increased interlayer interaction should be achieved; the substituted central benzene cores would ensure better solubility and a better alignment of the photoreactive units in the crystal packing (for details, see subchapter 4.6). Structure 3h is instead a proposition for potential polymerisations in self-assembled bilayers or vescicles. As a final remark, a proper choice of the side groups in the structure of 3, could also result in mesogens useful for the novel synthesis of 2DPs in smectic phases.
Figure 5.2. Further proposed structures for the synthesis of 2DPs.
238
APPENDIX
6. Appendix
Chapter 3: Amphiphilic Anthraphanes
Anthraphane 2a could be crystallised from hot THF, whereupon single crystals in the form of yellow needles were formed. The crystals were subjected to SC-XRD analysis but the quality of the structure was very low and contained disorder in the cyclophane, it was however sufficient to see that 2a packed in the etf packing 1. Channels were also present in the structure, presumably filled with solvent molecules as in the DMPU solvate of 1. The preliminary structure is displayed below in Figure 6.1:
Figure 6.1. Preliminary structure of amphiphilic anthraphane-tri(OMe) 2a, crystallised from THF. The molecule packs in the typical etf packing 1. In the structures channels are present which are presumably filled with solvent molecules.
239
APPENDIX
Anthraphane 2a could also be crystallised from nitrobenzene, from which single crystals in the form of yellow blocks were obtained. SC-XRD revealed a new kind of packing (Figure 6.2), in which two out of three anthracenes were perfectly stacking ftf (blue coloured), so that in principle a topochemical
1D polymerisation could be performed so to obtain chains of poly1D anthraphane-tri(OMe); the preliminary crystal structure is shown below:
Figure 6.2. Preliminary crystal structure of amphiphilic anthraphane-tri(OMe) 2a, crystallised from nitrobenzene. In this new packing, two out of three anthracenes (blue coloured) per monomer are engaged in a perfect ftf stacking. It is noted how the methoxy groups on the central benzene core sterically hinder the interdigitation of the cyclophanes, so that a sandwich ftf stacking is obtained instead of a parallel-displaced ftf stacking.
240
APPENDIX
Single crystals were irradiated with the same conditions employed in subchapter 2.9.2 (465 nm, 30 min, -10°C) and SC-XRD analysis revealed the single-crystal-to-single-crystal 1D polymerisation of anthraphane-tri(OMe) (Figure 6.3).
Figure 6.3. Crystal structure of poly1D anthraphane-tri(OMe) obtained from the topochemical polymerisation of the nitrobenzene solvate of 2a.
241
APPENDIX
Figure 6.4. ORTEP diagram of poly1D anthraphane-tri(OMe) in the 1,3-dimethoxybenzene solvate (50% probability).
Sample and crystal data CCDC depositary number 1505881
Empirical formula C81H46N2O7 Formula weight 1159.20 Temperature/K 100.0(2) Crystal system monoclinic
Space group P21/n a/Å 10.9635(3) b/Å 27.2821(9) c/Å 22.5009(6) α/° 90 β/° 101.120(2) γ/° 90 Volume/Å3 6603.8(3) Z 4 3 ρcalcg/cm 1.166 μ/mm-1 0.593 F(000) 2408.0 Crystal size/mm3 0.19 × 0.12 × 0.01 Radiation CuKα (λ = 1.54178) 2Θ range for data collection/° 5.148 to 133.506 Index ranges -13 ≤ h ≤ 12, -31 ≤ k ≤ 31, -26 ≤ l ≤ 20 Reflections collected 78150
Independent reflections 11488 [Rint = 0.0905, Rsigma = 0.0710] Data/restraints/parameters 11488/657/928 Goodness-of-fit on F2 1.491
Final R indexes [I>=2σ (I)] R1 = 0.1420, wR2 = 0.3997
Final R indexes [all data] R1 = 0.1987, wR2 = 0.4339 Largest diff. peak/hole / e Å-3 1.42/-0.60
242
APPENDIX
Anthraphane 2b was also crystallised from THF and single crystals in the form of yellow needles were obtained. However, the flexible polyether DEGME chains of the molecule caused disorder in the crystal structure and prevented to obtain a fully resolved structure. The preliminary structure is shown in Figure 6.5: it is noticeable that the monomer packs in the etf packing 1. No solvent seems to be included in the structure; the voids seem to be filled instead by the DEGME chains, which fold back along the anthracene units.
Figure 6.5. Preliminary structure of amphiphilic anthraphane-tri(DEGME) 2b, crystallised from THF. The molecule packs in the typical etf packing 1.
243
APPENDIX
Chapter 4: Diazaanthraphanes
Synthesis of higher esters such as n-hexyl and n-dodecyl was also explored in the hope of obtaining mesogenic diazaanthraphanes for the synthesis of 2DPs in smectic phases but unfortunately, both monomers did not display any liquid crystalline feature. The compounds had a waxy consistency and could not be crystallised. The synthesis of monomers 3d and 3e is depicted in Scheme 6.1, for details on the synthetic procedures please refer to the published literature[192].
Scheme 6.1 Synthesis of n-hexyl and n-dodecyl ester diazaanthraphanes 3d-e.
244
APPENDIX
Figure 6.6 1H-NMR spectra of n-hexyl and n-dodecyl ester diazaanthraphanes 3d-e.
Figure 6.7. Thremogravimetric analysis (TGA) of diazaanthraphanes 3b-e performed under nitrogen (heating rate 10°C/min). Correlation between ester loss and mass loss is not conclusive. 245
REFERENCES
References
[1] “WHO | WHO Model Lists of Essential Medicines,” can be found under http://www.who.int/medicines/publications/essentialmedicines/EML_2015_FINAL_amended _NOV2015.pdf?ua=1, 2016.
[2] M. J. Macielag, Antibiotic Discovery and Development, Springer, 2012.
[3] H. Staudinger, Berichte der Dtsch. Chem. Gesellschaft A B Ser. 1920, 53, 1073–1085.
[4] “The Nobel Prize in Chemistry 1953,” can be found under http://www.nobelprize.org/nobel_prizes/chemistry/laureates/1953/
[5] R. Mülhaupt, Angew. Chemie - Int. Ed. 2004, 43, 1054–1063.
[6] G. Odian, in Princ. Polym., John Wiley & Sons, Inc., Hoboken, NJ, USA, 2004.
[7] PlasticsEurope, Plastics – the Facts 2015, 2015.
[8] “The Nobel Prize in Physics 2010,” can be found under http://www.nobelprize.org/nobel_prizes/physics/laureates/2010/
[9] D. J. K. S. Novoselov, A. K. Geim, S. V. Morozov, A. A. F. Y. Zhang, S. V. Dubonos, I. V. Grigorieva, Science 2004, 306, 666–669.
[10] C. Lee, X. Wei, J. W. Kysar, J. Hone, Science 2008, 321, 385–388.
[11] The Engineering ToolBox, “Modulus of Elasticity or Young’s Modulus - and Tensile Modulus for common Materials,” can be found under http://www.engineeringtoolbox.com/young- modulus-d_417.html
[12] A. B. Kuzmenko, E. Van Heumen, F. Carbone, D. Van Der Marel, Phys. Rev. Lett. 2008, 100, 2– 5.
[13] A. K. Geim, N. K. S., Nat. Mater. 2007, 6, 183–191.
[14] A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao, C. N. Lau, Nano Lett. 2008, 8, 902–907.
[15] P. Miró, M. Audiffred, T. Heine, Chem. Soc. Rev. 2014, 43, 6537.
[16] T. T. Tran, K. Bray, M. J. Ford, M. Toth, I. Aharonovich, Nat. Nanotechnol. 2015, 11, 1–12.
246
REFERENCES
[17] M. M. Ugeda, A. J. Bradley, S.-F. Shi, F. H. da Jornada, Y. Zhang, D. Y. Qiu, W. Ruan, S.-K. Mo, Z. Hussain, Z.-X. Shen, et al., Nat. Mater. 2014, 13, 1091–1095.
[18] K. S. Novoselov, D. Jiang, F. Schedin, T. J. Booth, V. V Khotkevich, S. V Morozov, A. K. Geim, Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 10451–10453.
[19] R. Mas-Ballesté, C. Gómez-Navarro, J. Gómez-Herrero, F. Zamora, Nanoscale 2011, 3, 20–30.
[20] G. Liu, N. Komatsu, ChemPhysChem 2016, 17, 1557–1567.
[21] C. Jin, F. Lin, K. Suenaga, S. Iijima, Phys. Rev. Lett. 2009, 102, 3–6.
[22] R. Ma, T. Sasaki, Adv. Mater. 2010, 22, 5082–5104.
[23] X. Li, H. Zhu, J. Mater. 2015, 1, 33–44.
[24] S. J. Kim, K. Choi, B. Lee, Y. Kim, B. H. Hong, Annu. Rev. Mater. Res. 2015, 45, 63–84.
[25] A. Reina, X. Jia, J. Ho, D. Nezich, H. Son, V. Bulovic, M. S. Dresselhaus, J. Kong, Nano Lett. 2009, 9, 30–35.
[26] Y. H. Lee, X. Q. Zhang, W. Zhang, M. T. Chang, C. Te Lin, K. Di Chang, Y. C. Yu, J. T. W. Wang, C. S. Chang, L. J. Li, et al., Adv. Mater. 2012, 24, 2320–2325.
[27] J. Yu, J. Li, W. Zhang, H. Chang, Chem. Sci. 2015, 6, 6705–6716.
[28] J. M. Englert, C. Dotzer, G. Yang, M. Schmid, C. Papp, J. M. Gottfried, H.-P. Steinrück, E. Spiecker, F. Hauke, A. Hirsch, Nat. Chem. 2011, 3, 279–286.
[29] T. Kuila, S. Bose, A. K. Mishra, P. Khanra, N. H. Kim, J. H. Lee, Prog. Mater. Sci. 2012, 57, 1061– 1105.
[30] K. C. Kemp, V. Georgakilas, M. Otyepka, A. B. Bourlinos, V. Chandra, N. Kim, K. C. Kemp, P. Hobza, R. Zboril, K. S. Kim, Chem. Rev. 2012, 112, 6156–6214.
[31] J. Sakamoto, J. Van Heijst, O. Lukin, A. D. Schlüter, Angew. Chem. Int. Ed. 2009, 48, 1030– 1069.
[32] J. W. Colson, W. R. Dichtel, Nat. Chem. 2013, 5, 453–65.
[33] P. Kissel, R. Erni, W. B. Schweizer, M. D. Rossell, B. T. King, T. Bauer, S. Götzinger, a. D. Schlüter, J. Sakamoto, Nat. Chem. 2012, 4, 287–291.
[34] Z. Xiang, D. Cao, L. Dai, Polym. Chem. 2015, 6, 1896–1911. 247
REFERENCES
[35] D. Rodríguez-San-Miguel, P. Amo-Ochoa, F. Zamora, Chem. Commun. 2016, 52, 4113–4127.
[36] X. Zhuang, Y. Mai, D. Wu, F. Zhang, X. Feng, Adv. Mater. 2015, 27, 403–427.
[37] C. E. Boott, A. Nazemi, I. Manners, Angew. Chem. Int. Ed. 2015, 54, 13876–13894.
[38] R. Bhola, P. Payamyar, D. J. Murray, B. Kumar, A. J. Teator, M. U. Schmidt, S. M. Hammer, A. Saha, J. Sakamoto, A. D. Schlüter, et al., J. Am. Chem. Soc. 2013, 135, 14134–14141.
[39] P. Kissel, D. J. Murray, W. J. Wulftange, V. J. Catalano, B. T. King, Nat. Chem. 2014, 6, 774–778.
[40] M. J. Kory, M. Wörle, T. Weber, P. Payamyar, S. W. van de Poll, J. Dshemuchadse, N. Trapp, A. D. Schlüter, Nat. Chem. 2014, 6, 779–784.
[41] D. J. Murray, D. D. Patterson, P. Payamyar, R. Bhola, W. Song, M. Lackinger, A. D. Schlüter, B. T. King, J. Am. Chem. Soc. 2015, 137, 3450–3453.
[42] B. M. D. Cohen, G. M. J. Schmidt, J. Chem. Soc. 1964, 1996–2000.
[43] G. M. J. Schmidt, Pure Appl. Chem. 1971, 27, 647–678.
[44] V. Ramamurthy, K. Venkatesan, Chem. Rev. 1987, 87, 433–481.
[45] K. Biradha, R. Santra, Chem. Soc. Rev. 2013, 42, 950–967.
[46] G. Wegner, Zeitschrift für Naturforsch. 1969, 24 B, 824–832.
[47] M. Hasegawa, Y. Suzuki, J. Polym. Sci. Part B Polym. Lett. 1967, 5, 813–815.
[48] M. J. Kory, M. Bergeler, M. Reiher, A. D. Schlüter, Chem. Eur. J. 2014, 20, 6934–6938.
[49] I. Langmuir, Trans. Faraday Soc. 1920, 15, 62–74.
[50] I. Langmuir, J. Am. Chem. Soc. 1917, 39, 1848–1906.
[51] A. Braslau, M. Deutsch, P. S. Pershan, A. H. Weiss, J. Als-Nielsen, J. Bohr, Phys. Rev. Lett. 1985, 54, 114–117.
[52] T. Kaindl, J. Oelke, A. Pasc, S. Kaufmann, O. V. Konovalov, S. S. Funari, U. Engel, A. Wixforth, M. Tanaka, J. Phys. Condens. Matter 2010, 22, 285102.
[53] T. Bauer, Z. Zheng, A. Renn, R. Enning, A. Stemmer, J. Sakamoto, A. D. Schlüter, Angew. Chem. Int. Ed. 2011, 50, 7879–7884.
[54] Z. Zheng, C. S. Ruiz-vargas, T. Bauer, A. Rossi, P. Payamyar, A. Schütz, A. Stemmer, J. 248
REFERENCES
Sakamoto, A. D. Schlüter, Macromol. Rapid Commun. 2013, 34, 1670–1680.
[55] R. Sakamoto, K. Hoshiko, Q. Liu, T. Yagi, T. Nagayama, S. Kusaka, M. Tsuchiya, Y. Kitagawa, W.- Y. Wong, H. Nishihara, Nat. Commun. 2015, 6, 6713.
[56] R. Makiura, O. Konovalov, Sci. Rep. 2013, 3, 1–8.
[57] P. Payamyar, K. Kaja, C. Ruiz-Vargas, A. Stemmer, D. J. Murray, C. J. Johnson, B. T. King, F. Schiffmann, J. Vandevondele, A. Renn, et al., Adv. Mater. 2014, 26, 2052–2058.
[58] P. Kissel, J. Van Heijst, R. Enning, A. Stemmer, A. D. Schlüter, J. Sakamoto, Org. Lett. 2010, 12, 2778–2781.
[59] L. Opilik, P. Payamyar, J. Szczerbinsk, A. P. Schutz, M. Servalli, T. Hungerland, A. D. Schluter, R. Zenobi, ACS Nano 2015, 9, 4252–4259.
[60] Y. Chen, M. Li, P. Payamyar, Z. Zheng, J. Sakamoto, A. D. Schlüter, ACS Macro Lett. 2014, 3, 153–158.
[61] P. Payamyar, M. Servalli, T. Hungerland, A. P. Schütz, Z. Zheng, A. Borgschulte, A. D. Schlüter, Macromol. Rapid Commun. 2015, 36, 151–158.
[62] N. W. Ockwig, A. P. Cote, M. O. Keeffe, A. J. Matzger, O. M. Yaghi, Science. 2005, 310, 1166– 1171.
[63] S. S. Han, H. Furukawa, O. M. Yaghi, W. A. Goddard III, J. Am. Chem. Soc. 2008, 130, 11580– 11581.
[64] S. S. Han, J. L. Mendoza-Cortes, W. A. Goddard III, Chem Soc Rev 2009, 38, 1460–1476.
[65] J.-M. Lehn, Chem. Eur. J. 1999, 5, 2455–2463.
[66] S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders, J. F. Stoddart, Angew. Chem. Int. Ed. Engl. 2002, 41, 898–952.
[67] Y. Jin, C. Yu, R. J. Denman, W. Zhang, Chem. Soc. Rev. 2013, 42, 6634–6654.
[68] B. T. Koo, W. R. Dichtel, P. Clancy, J. Mater. Chem. 2012, 22, 17460.
[69] I. Berlanga, M. L. Ruiz-González, J. M. González-Calbet, J. L. G. Fierro, R. Mas-Ballesté, F. Zamora, Small 2011, 7, 1207–1211.
[70] D. N. Bunck, W. R. Dichtel, J. Am. Chem. Soc. 2013, 135, 14952–14955.
249
REFERENCES
[71] Y. B. Zhang, J. Su, H. Furukawa, Y. Yun, F. Gándara, A. Duong, X. Zou, O. M. Yaghi, J. Am. Chem. Soc. 2013, 135, 16336–16339.
[72] D. Beaudoin, T. Maris, J. D. Wuest, Nat. Chem. 2013, 5, 830–4.
[73] S. I. Stupp, S. Son, L. S. Li, H. C. Lin, M. Keser, J. Am. Chem. Soc. 1995, 117, 5212–5227.
[74] Y. Zhao, R. H. M. Bernitzky, M. J. Kory, G. Hofer, J. Hofkens, A. D. Schlüter, J. Am. Chem. Soc. 2016, 138, 8976–8981.
[75] Z. Wang, S. M. Cohen, Chem. Soc. Rev. 2009, 38, 1315–1329.
[76] K. K. Tanabe, S. M. Cohen, Chem. Soc. Rev. 2011, 40, 498–519.
[77] H. Xu, J. Gao, Nat. Chem. 2015, 7, 905–912.
[78] K. Celebi, J. Buchheim, R. M. Wyss, A. Droudian, P. Gasser, I. Shorubalko, J.-I. J.-I. Kye, C. Lee, H. G. Park, Science 2014, 344, 289–292.
[79] T. Rodenas, I. Luz, G. Prieto, B. Seoane, H. Miro, A. Corma, F. Kapteijn, F. X. Llabrés i Xamena, J. Gascon, Nat. Mater. 2015, 14, 48–55.
[80] M. Vargas, S. Naris, D. Valougeorgis, S. Pantazis, K. Jousten, Vacuum 2014, 109, 385–396.
[81] J. D. A. Pelletier, J.-M. Basset, Acc. Chem. Res. 2016, 49, 664–677.
[82] H.-D. Becker, Chem. Rev. 1993, 93, 145–172.
[83] H. Bouas-Laurent, J.-P. Desvergne, A. Castellan, R. Lapouyade, Chem. Soc. Rev. 2000, 29, 43– 55.
[84] H. Bouas-Laurent, J.-P. Desvergne, A. Castellan, R. Lapouyade, Chem. Soc. Rev. 2001, 30, 248– 263.
[85] I. Zouev, D. Cao, T. V. Sreevidya, M. Telzhensky, M. Botoshansky, M. Kaftory, CrystEngComm 2011, 13, 4376–4381.
[86] R. Gleiter, H. Hopf, Modern Cyclophane Chemistry, Wiley-VCH, Weinheim, 2004.
[87] C. C. Lee, M. K. Leung, G. H. Lee, Y. H. Liu, S. M. Peng, J. Org. Chem. 2006, 71, 8417–8423.
[88] T. Tajima, A. Yamakawa, K. Fukuda, Y. Hayashi, M. Nakano, Y. Takaguchi, Chem. Lett. 2012, 41, 1622–1624.
250
REFERENCES
[89] A. P. Davis, G. H. Whitham, J. Chem. Soc., Chem. Commun. 1980, 639–640.
[90] H. Dodziuk, Strained Hydrocarbons : Beyond the Van’t Hoff and Le Bel Hypothesis, Wiley-VCH, 2009.
[91] A. Greenberg, J. F. Liebman, Strained Organic Molecules, Academic Press, 2013.
[92] R. Chinchilla, C. Najera, Chem. Soc. Rev. 2011, 40, 5084–5121.
[93] M. Bakherad, Appl. Organomet. Chem. 2013, 27, 125–140.
[94] F. Diederich, P. J. Stang, R. R. Tykwinski, Modern Supramolecular Chemistry : Strategies for Macrocycle Synthesis, Wiley-VCH, 2008.
[95] R. Chinchilla, C. Nájera, Chem. Rev. 2007, 107, 874–922.
[96] W. Zhang, J. S. Moore, Angew. Chem. Int. Ed. 2006, 45, 4416–4439.
[97] Y. Tobe, J. Kishi, I. Ohki, J. Org. Chem. 2003, 68, 3330–3332.
[98] J. Peter-Katalinic, F. Ebmeyer, C. Seel, F. Vögtle, Chem. Ber. 1989, 122, 2391–2393.
[99] M. Pérez-Trujillo, I. Maestre, C. Jaime, A. Alvarez-Larena, J. F. Piniella, A. Virgili, Tetrahedron Asymmetry 2005, 16, 3084–3093.
[100] M. Goichi, K. Segawa, S. Suzuki, S. Toyota, Synthesis 2005, 2116–2118.
[101] P. Kissel, F. Weibel, L. Federer, J. Sakamoto, A. D. Schlüter, Synlett 2008, 1793–1796.
[102] J. Jover, P. Spuhler, L. Zhao, C. Mcardle, F. Maseras, Catal. Sci. Technol. Catal. Sci. Technol 2014, 4, 4103–4442.
[103] T. Ljungdahl, T. Bennur, A. Dallas, H. Emtenäs, J. Mårtensson, Organometallics 2008, 27, 2490–2498.
[104] G. M. J. Schmidt, Solid State Photochemistry, Verlag Chemie, Weinheim, 1976.
[105] L. M. Salonen, M. Ellermann, F. Diederich, Angew. Chem. Int. Ed. 2011, 50, 4808–4842.
[106] J. D. Dunitz, Pure Appl. Chem. 1991, 63, 177–185.
[107] S. E. Wheeler, Acc. Chem. Res. 2012, 46, 1029–1038.
[108] M. O. Sinnokrot, C. D. Sherrill, J. Phys. Chem. A 2006, 110, 10656–10668.
251
REFERENCES
[109] C. R. Martinez, B. L. Iverson, Chem. Sci. 2012, 3, 2191–2201.
[110] S. Grimme, Angew. Chem. Int. Ed. 2008, 47, 3430–3434.
[111] O. Takahashi, Y. Kolno, S. Iwasaki, K. Saito, M. Iwaoka, S. Tomoda, Y. Umezawa, S. Tsuboyama, M. Nishio, Bull. Chem. Soc. Jpn. 2001, 74, 2421–2430.
[112] M. Nishio, CrystEngComm 2004, 6, 130–158.
[113] M. Nishio, Y. Umezawa, K. Honda, S. Tsuboyama, H. Suezawa, CrystEngComm 2009, 11, 1757– 1788.
[114] M. Nishio, Phys. Chem. Chem. Phys. 2011, 13, 13873–13900.
[115] M. J. Plevin, D. L. Bryce, J. Boisbouvier, Nat. Chem. 2010, 2, 466–471.
[116] J. W. Larson, T. B. McMahon, Inorg. Chem. 1984, 23, 2029–2033.
[117] J. Emsley, Chem. Soc. Rev. 1980, 9, 91.
[118] A. de Meijere, F. Huisken, J. Chem. Phys. 1990, 92, 5826–5834.
[119] T. P. Tauer, C. D. Sherrill, J. Phys. Chem. A 2005, 109, 10475–10478.
[120] C. Wohlfarth, Static Dielectric Constants of Pure Liquids and Binary Liquid Mixtures: Supplement to IV/6, Springer Science & Business Media, 2008.
[121] C. Wohlfarth, M. D. Lechner, Viscosity of Pure Organic Liquids and Binary Liquid Mixtures, Springer, 2008.
[122] H. Suezawa, S. Ishihara, Y. Umezawa, S. Tsuboyama, M. Nishio, Eur. J. Org. Chem. 2004, 4816– 4822.
[123] S. Alvarez, Dalt. Trans. 2013, 42, 8617–8636.
[124] Y. Umezawa, S. Tsuboyama, K. Honda, J. Uzawa, M. Nishio, Bull. Chem. Soc. Jpn. 1998, 71, 1208–1214.
[125] A. I. Kitaigorodskii, Molecular Crystals and Molecules, Academic Press, New York, 1973.
[126] J. D. Dunitz, A. Gavezzotti, Cryst. Growth Des. 2005, 5, 2180–2189.
[127] J. D. Dunitz, W. B. Schweizer, CrystEngComm 2007, 9, 266–269.
[128] M. C. Etter, Acc. Chem. Res. 1990, 23, 120–126. 252
REFERENCES
[129] P. W. Baures, J. R. Rush, A. V. Wiznycia, J. Desper, B. A. Helfrich, A. M. Beatty, Cryst. Growth Des. 2002, 2, 653–664.
[130] P. Vishweshwar, A. Nangia, V. M. Lynch, Cryst. Growth Des. 2003, 3, 783–790.
[131] J. F. Remenar, S. L. Morissette, M. L. Peterson, B. Moulton, J. M. MacPhee, H. R. Guzmán, Ö. Almarsson, J. Am. Chem. Soc. 2003, 125, 8456–8457.
[132] F.-G. Klärner, J. Panitzky, D. Preda, L. T. Scott, J. Mol. Model. 2000, 6, 318–327.
[133] J. S. (Jane S). . Murray, K. D. (Kali D.) Sen, Molecular Electrostatic Potentials : Concepts and Applications, Elsevier, 1996.
[134] N. L. Rosi, J. Eckert, M. Eddaoudi, D. T. Vodak, J. Kim, M. O’Keeffe, O. M. Yaghi, Science 2003, 300, 1127–1129.
[135] J. E. Mondloch, O. Karagiaridi, O. K. Farha, J. T. Hupp, CrystEngComm 2013, 15, 9258–9264.
[136] G. R. Desiraju, Angew. Chem. Int. Ed. 1995, 34, 2311–2327.
[137] E. J. J. Corey, Pure Appl. Chem. 1967, 14, 19–37.
[138] G. R. Desiraju, Angew. Chem. Int. Ed. 2007, 46, 8342–8356.
[139] G. R. Desiraju, J. Am. Chem. Soc. 2013, 135, 9952–9967.
[140] S. M. Woodley, R. Catlow, Nat. Mater. 2008, 7, 937–946.
[141] M. A. Neumann, F. J. J. Leusen, J. Kendrick, Angew. Chem. Int. Ed. 2008, 47, 2427–2430.
[142] G. M. Day, T. G. Cooper, A. J. Cruz-Cabeza, K. E. Hejczyk, H. L. Ammon, S. X. M. Boerrigter, J. S. Tan, R. G. Della Valle, E. Venuti, J. Jose, et al., Acta Crystallogr. Sect. B Struct. Sci. 2009, 65, 107–125.
[143] A. Sygula, F. R. Fronczek, R. Sygula, P. W. Rabideau, M. M. Olmstead, J. Am. Chem. Soc. 2007, 129, 3842–3843.
[144] N. Martín, L. Sánchez, B. Illescas, I. Pérez, Chem. Rev. 1998, 98, 2527–2548.
[145] R. C. Haddon, Phil. Trans. R. Soc. Lond. A 1993, 343, 53–62.
[146] R. M. R. S. Ruoff Doris S. Tse, D. C. Lorents, J. Phys. Chem. 1993, 97, 3379–3383.
[147] V. Schettino, M. Pagliai, L. Ciabini, G. Cardini, J. Phys. Chem. A 2001, 105, 11192–11196.
253
REFERENCES
[148] W. M. Reichert, J. D. Holbrey, K. B. Vigour, T. D. Morgan, G. a Broker, R. D. Rogers, Chem. Commun. 2006, 4767–4779.
[149] J. D. Holbrey, W. M. Reichert, M. Nieuwenhuyzen, O. Sheppard, C. Hardacre, R. D. Rogers, Chem. Commun. 2003, 476–477.
[150] V. Enkelmann, Mol. Cryst. Liq. Cryst. Sci. Technol. 1998, 313, 15–23.
[151] I. Halasz, Cryst. Growth Des. 2010, 10, 2817–2823.
[152] P. Kissel, A. D. Schlüter, J. Sakamoto, Chem. Eur. J. 2009, 15, 8955–8960.
[153] D. R. Coulson, L. C. Satek, S. O. Grim, Inorg. Synth. 1971, 13, 121–124.
[154] Bruker AXS Inc.
[155] G. M. Sheldrick, Acta Cryst. 2007, A64, 112–122.
[156] O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann, J. Appl. Crystallogr. 2009, 42, 339–341.
[157] A. O. King, E. Negishi, F. J. Villani, A. Silveira, J. Org. Chem. 1978, 43, 358–360.
[158] M. Sonoda, A. Inaba, K. Itahashi, Y. Tobe, Org. Lett. 2001, 3, 2419–2421.
[159] J. Kouvetakis, D. Grotjahn, P. Becker, S. Moore, R. Dupon, Chem. Mater. 1994, 6, 636–639.
[160] J. E. Anthony, S. I. Khan, Y. Rubin, Tetrahedron Lett. 1997, 38, 3499–3502.
[161] C. Holst, D. Schollmeyer, H. Meier, Z. Naturforsch. 2011, 66b, 935–938.
[162] D. Bruns, H. Miura, A. Stanger, K. P. C. Vollhardt, Org. Lett. 2003, 5, 549–552.
[163] A. D. Burrows, C. G. Frost, M. F. Mahon, C. Richardson, Angew. Chem. Int. Ed. 2008, 47, 8482– 8486.
[164] M. Servalli, M. Ranocchiari, J. a. Van Bokhoven, Chem. Commun. 2012, 48, 1904–1906.
[165] E. Kiehlmann, R. W. Lauener, Can. J. Chem. 1989, 67, 335–344.
[166] J. Song, B. K. Cho, Bull. Korean Chem. Soc. 2010, 31, 1835–1836.
[167] B. Lee, J. H. Kwak, S. W. Huang, J. Y. Jang, S. Lim, Y. S. Kwak, K. Lee, H. S. Kim, S. B. Han, J. T. Hong, et al., Bioorganic Med. Chem. 2012, 20, 2860–2868.
254
REFERENCES
[168] S. A. Weissman, D. Zewge, Tetrahedron 2005, 61, 7833–7863.
[169] G. Hennrich, Tetrahedron 2004, 60, 9871–9876.
[170] G. Hennrich, I. Asselberghs, K. Clays, A. Persoons, J. Org. Chem. 2004, 69, 5077–5081.
[171] H. H. Wenk, W. Sander, Eur. J. Org. Chem 2002, 3927–3935.
[172] M. Schilz, H. Plenio, J. Org. Chem. 2012, 77, 2798–2807.
[173] G. C. Fu, Acc. Chem. Res. 2008, 41, 1555–1564.
[174] J. R. Heard, Physalia physalis Portuguese man o' war, Marine Biological Association of the United Kingdom 2006.
[175] T. Hayashi, N. Mataga, Y. Sakata, S. Misumi, M. Morita, J. Tanaka, J. Am. Chem. Soc. 1976, 98, 5910–5913.
[176] B. Branchi, P. Ceroni, V. Balzani, M. C. Cartagena, F.-G. Klärner, T. Schrader, F. Vögtle, New J. Chem. 2009, 33, 397–407.
[177] J. C. Amicangelo, W. R. Leenstra, J. Am. Chem. Soc. 2003, 125, 14698–14699.
[178] W. Fudickar, A. Fery, T. Linker, J. Am. Chem. Soc. 2005, 127, 9386–9387.
[179] W. Fudickar, T. Linker, Chem. Commun. 2008, 1771–1773.
[180] W. Fudickar, T. Linker, Langmuir 2010, 26, 4421–4428.
[181] E. A. Kataev, T. A. Shumilova, B. Fiedler, T. Anacker, J. Friedrich, J. Org. Chem. 2016, 81, 6505– 6514.
[182] S. Feng, P. N. Sen, Phys. Rev. Lett. 1984, 52, 216–219.
[183] F. Li, J. Zhuang, G. Jiang, H. Tang, A. Xia, L. Jiang, Y. Song, Y. Li, D. Zhu, R. V August, Chem. Mater. 2008, 20, 1194–1196.
[184] D. B. G. Williams, M. Lawton, J. Org. Chem. 2010, 75, 8351–8354.
[185] E. Berni, C. Dolain, B. Kauffmann, J. M. Léger, C. Zhan, I. Huc, J. Org. Chem. 2008, 73, 2687– 2694.
[186] S. Yamada, C. Kawamura, Org. Lett. 2012, 14, 1572–1575.
[187] M. Li, A. D. Schlüter, J. Sakamoto, J. Am. Chem. Soc. 2012, 134, 11721–11725. 255
REFERENCES
[188] D. Jouvenot, E. C. Glazer, Y. Tor, Org. Lett. 2006, 8, 1987–1990.
[189] F. F. Molock, D. W. Boykin, J. Heterocycl. Chem. 1983, 20, 681–686.
[190] Q. Zhu, H. Jiang, J. Li, M. Zhang, X. Wang, C. Qi, Tetrahedron 2009, 65, 4604–4613.
[191] CrysAlisPro, Rigaku OD, 2015.
[192] M. Servalli, L. Gyr, J. Sakamoto, A. D. Schlüter, European J. Org. Chem. 2015, 4519–4523.
[193] V. Ramamurthy, Tetrahedron 1986, 42, 5753–5839.
[194] S. Tsuzuki, K. Honda, T. Uchimaru, M. Mikami, A. Fujii, J. Phys. Chem. A 2006, 110, 10163– 10168.
[195] M. Servalli, N. Trapp, M. Wörle, F. G. Klärner, J. Org. Chem. 2016, 81, 2572–2580.
256
CURRICULUM VITAE
Curriculum Vitae
Personal Data
Name Marco Natale Servalli
Date of Birth 10.07.1986
Place of Birth Bioggio (TI)
Nationality Swiss
Email [email protected]
Education
2011 - present Doctor of Philosophy, ETH Zürich, Switzerland Anthraphanes: a New Class of Potential Monomers for the Synthesis of Two-Dimensional Polymers Advisor: Prof. Dr. A. Dieter Schlüter
03.2011 - 09.2011 Post M.Sc. in Chemistry, Paul Scherrer Institute, Switzerland Post-synthetic modification of metal-organic frameworks and synthesis of novel phosphine-based metal-organic frameworks for asymmetric catalysis Advisors: Dr. Marco Ranocchiari and Prof. Dr. Jeroen van Bokhoven
2010 – 2011 M.Sc. in Chemistry, ETH Zürich, Switzerland Thesis: Towards 2D polymers: Synthesis of propeller shaped monomers bearing three 1,8‐anthrylene blades for UV induced solid‐state polymerisation Advisors: Dr. Junji Sakamoto and Prof. A. Dieter Schlüter
2006 – 2010 B.Sc. in Chemistry, ETH Zürich, Switzerland
2006 GEOS Queensland College of English, Gold Coast, Queensland, Australia
2001 – 2005 High School, Liceo Lugano 2, Savosa, Switzerland Scientific Matura: selective extraction of anthocyans from fermented Merlot grapes
257
CURRICULUM VITAE
Work and Teaching Experience
2013 Teaching Assistant at ETH Zürich, Switzerland Practical Laboratory Course of Chemistry for undergraduate students, Department of Materials
2009 – 2011 Teaching Assistant at ETH Zürich, Switzerland Teaching and Exercise in Inorganic Chemistry II: Structural and solid-state Chemistry
2007 – 2009 Substitute Teacher in Chemistry at High School, Liceo Lugano 2, Savosa, Switzerland Lectures and Lab Courses
2009 Substitute Teacher in Mathematics, Secondary School, Agno, Switzerland
2008 – 2010 Private Teacher in Chemistry
Additional Skills
Languages Italian (mother tongue), English (fluent), German (fluent), French (good), Spanish (basic)
IT Microsoft Office, Origin, Mercury, Gaussian, ChemBioOffice, OLEX2, TopSpin, MestReNova, SciFinder, Reaxys, Photoshop
Further Education General Management
Other Experiences and Activities
Since 2006 Civil Protection (Assistance)
Since 2007 Modelling
Leisure Activities and Interests
Fitness and Bodybuilding, Tennis, Jogging, Electronic Music
258
CURRICULUM VITAE
Publications
1. M. Servalli, M. Ranocchiari, J. A. van Bokhoven, “Fast and High Yield Post-Synthetic Modification of Metal-organic Frameworks by Vapor Diffusion”, Chem. Commun. 2012, 48, 1904-1906.
2. J. Vaclavik, M. Servalli, C. Lothschütz, J. Szlachetko, M. Ranocchiari, J. A. van Bokhoven, “AuI Catalysis on a Coordination Polymer: A Solid Porous Ligand with Free Phosphine Sites”, ChemCatChem 2013, 5, 692-696.
3. J. Sa, J. Szlachetko, E. Kleymenov, C. Lothschütz, M. Nachtegaal, M. Ranocchiari, O.V. Safonova, M. Servalli, G. Smolentsev, and Jeroen A. van Bokhoven, “Fine Tuning of Gold Electronic Structure by IRMOF Post-synthetic Modification”, RSC Adv. 2013, 3, 12043-12048.
4. Marco Servalli, Luzia Gyr, Junji Sakamoto, and A. Dieter Schlüter, “Propeller- Shaped D3h-Symmetric Macrocycles with Three 1,8-Diazaanthracene Blades as Building Blocks for Photochemically Induced Growth Reactions”, Eur. J. Org. Chem. 2015, 4519–4523.
5. P. Payamyar, M. Servalli, T. Hungerland, A.P. Schütz, Z. Zheng, A. Borgschulte, and D. Schlüter, “ Approaching Two-dimensional Copolymers. Photoirradiation of Anthracene- and Diaza-anthracene-bearing Monomers in Langmuir Monolayers”, Macromol. Rapid Commun. 2015, 36, 151−158.
6. L. Opilik, P. Payamyar, J. Szczerbinski, A. P. Schütz, M. Servalli, T. Hungerland, A. D. Schlüter, R. Zenobi, “Minimally Invasive Characterization of Covalent Monolayer Sheets Using Tip-enhanced Raman Spectroscopy”, ACS Nano 2015, 9, 4252–4259.
7. M. Servalli, N. Trapp, M. Woerle, and F.-G. Klaerner, “Anthraphane: An Anthracene-Based, Propeller-Shaped D3h-Symmetric Hydrocarbon Cyclophane and Its Layered Single Crystal Structures”, J. Org. Chem. 2016, 81, 2572–2580.
8. M. Servalli, N. Trapp, M. Solar, A. D. Schlüter, “Single-Crystal-to-Single-Crystal Photodimerisations of Anthraphane”, 2017, manuscript in preparation.
9. M. Servalli, P. Payamyar, A. D. Schlüter, “Amphiphilic Anthraphanes for the Synthesis of Covalent Monolayer Sheets at the Air/Water Interface”, 2017, manuscript in preparation.
10. M. Servalli, A. D Schlüter, “Synthetic Two-Dimensional Polymers“, Annu. Rev. Mater. Res. 2017, 47, in press.
259
CURRICULUM VITAE
Poster Presentations
Marco Servalli, A. Dieter Schlüter, Junji Sakamoto, “Propeller-shaped Monomers with Anthracene Blades for Topochemical Synthesis of 2D Polymers”, August 2012, 4th EuCheMS Chemistry Congress, Prague, Czech Republic.
Marco Servalli, A. Dieter Schlüter, Junji Sakamoto, “Photoreactive Propeller-shaped Monomers for Topochemical Synthesis of 2D Polymers”, September 2012, Swiss Chemical Society Fall Meeting, Zurich, Switzerland.
Marco Servalli, Junji Sakamoto, A. Dieter Schlüter, “Propeller-shaped Macrocycles with Three 1,8-Diazaanthracene and Three Anthracene Blades as Monomers for Topochemical 2D-Polymerisations”, July 2014, 1st International Symposium on Synthetic Two-Dimensional Polymers, Zurich, Switzerland
Marco Servalli, A. Dieter Schlüter, Junji Sakamoto, “Propeller-shaped Monomers with Three 1,8-Diazaanthracene and Three Anthracene Blades for Topochemical 2D- Polymerisations”, September 2014, Swiss Chemical Society Fall Meeting, Zurich, Switzerland.
Marco Servalli, Junji Sakamoto, A. Dieter Schlüter, “Rotor-shaped Macrocycles with Three 1,8-Diazaanthracene and Three Anthracene Blades as Monomers for Topochemical 2D-Polymerization”, July 2015, 16th International Symposium on Novel Aromatic Compounds (ISNA-16), Madrid, Spain.
Marco Servalli, A. Dieter Schlüter, “Anthraphanes as Potentially Versatile Monomers for the Synthesis of Two-Dimensional Polymers”, June 2016, 2nd International Symposium on Synthetic Two-Dimensional Polymers, Nara, Japan.
Oral Presentations
Marco Servalli, “Anthraphanes: versatile systems for the synthesis of 2D-polymers?”, June 2016, 2nd International Symposium on Synthetic Two-Dimensional Polymers, Nara, Japan.
Marco Servalli, “Anthraphanes: A new class of potential monomers for the synthesis of two-dimensional polymers”, November 2016, ETHZ-UTokyo Joint Symposium of Frontier Chemistry, Tokyo, Japan.
260