A Novel Catalytic System for the Oxidative Carboxylation of Stryene Based on a POM Embedded Cationic CTF Incorporated in Electro

Sander Bossier Development of a novel

Promotor: Prof. Dr. Karen Leus catalytic system for the Co-promotors: Prof. Dr. Peter Dubruel Prof. Dr Karen De Clerck Supervisors: Parviz Gohari Derakhshandeh oxidative carboxylation Dr. Tom Gheysens Dr. Eva Loccufier of styrene based on a

Academic Year 2018-2019 POM embedded cationic

CTF incorporated in A dissertation submitted to Ghent University in partial fulfilment of the requirements for the degree of Master of Science in Chemistry electrospun nanofibres

Acknowledgments

The realization of a thesis is never a solitary endeavour, without the people that I have met throughout this year, it could have never been done. I want to thank these people for their time, their support, their knowledge, and their help. Firstly, I would like to thank my promoter prof. dr. Karen Leus, and my co-promoters, prof. dr. Peter Dubruel and prof. dr. ir. Karen De Clerck, for allowing me to perform this research in their laboratories and use their equipment. Secondly, I would like to thank prof. dr. Pascal Van Der Voort for the use of his laboratory and equipment. Thirdly, my guides Parviz, Tom, and Eva, without you to show me the ropes, I do not know where I would be. And lastly of course, the COMOC, PBM, and CTSE groups, thank you for the good times. On top of this I would also like to thank Funda Ali¸cand Tom Planckaert for teaching me how to work with some of the equipment. Thank you dr. Anna Kaczmarek for supplying the heptammonium tridecavanadomanganate. I have to thank Katrien Haustraete too for the STEM-EDX images and Roseline Blanckaert for the ICP-AES analysis. And ﬁnally a shout-out to the Hellcats (thank you, Andreas and Geert for the great atmosphere), my fellow thesis students, my friends (time to go on another weekend with destination X) and everybody at KSA-Tielt for being a listening ear and the best friends I could have ever wished for.

iii A Novel Catalytic System for the Oxidative Carboxylation of Styrene based on a POM Embedded Cationic CTF Incorporated in Electrospun Nanofibres

S. Bossier, P. Gohari Derakhshandeha, T. Gheysensb, E. Loccufierc, A. Kazcmareka, P. Van Der Voorta, K. De Clerckc, P. Dubruelb, and K. Leusa

a COMOC, Department of Chemistry, Ghent University, Ghent b PBM, Department of Organic and Macromolecular Chemistry, Ghent University, Ghent c CTSE, Department of Materials, Textiles and Chemical Engineering, Ghent University, Ghent

The utilization of CO2 in the production of value-added chemicals is one way to reduce its emission. In this research, a dual-functional catalyst is envisaged to achieve the orthogonal tandem catalysis of the oxidative carboxylation of styrene to styrene carbonate (SC).1 This is achieved by incorporating a vanadium-based polyoxometalate (VPOM) in a cationic covalent triazine framework (cCTF) based on imidazolium moieties. More specifically, the embedded VPOM catalyses the epoxidation of styrene to styrene oxide with tert-butylhydroperoxide (TBHP) as the oxidant. Simultaneously, CO2 molecules are efficiently captured and activated by CO2-philic N-rich triazine groups of the CTF. Next, the bromide ion of the tetrabutylammonium bromide (TBAB) co- catalyst activates the epoxide via ring-opening. Subsequently the 2 activated CO2 is inserted to produce the desired styrene carbonate. The goal is to make this multistep reaction go to the desired product in one reactor without the need for separation, purification, and transfer of intermediates produced. The second part of the research is focused on embedding the VPOM@cCTF particles in a polyamide-4.6 (PA-4.6) nanofibre to increase ease of use and catalyst recovery. This is achieved by electrospinning a VPOM@cCTF and PA-4.6 solution from formic and acetic acid.3

Keywords Oxidative carboxylation; styrene; covalent organic framework; polyoxometalate; electrospinning; tandem catalysis; CO2 conversion; cyclic carbonate

Introduction 4 Nowadays, the total annual anthropogenic CO2 emission is about 36 Gt. Combine this with the fact that there is already an excess of more than 900 Gt of CO2 in the atmosphere, 5 nearing 1 Tt, means that there is a huge reservoir of untapped CO2 ripe for exploitation . Utilizing this CO2 as a C1 building block for the synthesis of chemicals with high molecular complexity, and thus high value, would create an economical incentive to further invest in now costly direct-air-capture technologies6. To that end a substantial amount of research is dedicated to efficiently produce such chemicals using CO2 as a starting product or as an intermediary. An interesting class of such complex chemicals, of which a large increase in production/demand is expected7, is the (cyclic) organic carbonates. These compounds could be used as solvents for aromatic compounds, polymers, and salt like compounds. Some other (potential) uses that are less to not industrially applied (but are being Ghent University, 2019 investigated) are e.g. as monomers to polymerize into a variety of homo- and copolymer polycarbonates with high molecular masses, extractants, plasticizers, electrolytes, intermediates in organic synthesis, spinning dopes for synthetic fibres, specific solvents for pharmaceutical and cosmetic preparations, or other niche usage applications.8 Tandem catalysis, in which multi-step chemical transformations are catalysed by multifunctional catalysts, has attracted increasing research attention. Selective multifunctional catalysts make a multi-step reaction go to the desired products in one reactor without the need for separation, purification, and transfer of intermediates produced in each step. Great efforts have been made to design heterogeneous catalysts for tandem reactions through the immobilization of metal complexes and nanoparticles on the surface of various porous supports such as polymers, zeolites, molecular sieves, and metal-organic frameworks (MOFs). Recently, covalent triazine frameworks (CTFs) emerged as a new type of porous materials. CTFs have similar characteristics in comparison to the well- known MOFs but exhibit enhanced chemical stability. Here, a novel approach is devised based on the incorporation of the vanadium-based POM (VPOM), heptammonium tridecavanadomanganate ([NH4]7[MnV13O38]), in a cCTF, based on a 1,3-bis(5-cyanopyridyl)-imidazolium bromide (bpim) cationic linker, as dual- functional materials to achieve a tandem catalyst for the efficient conversion of CO2 and 9 styrene into value-added styrene carbonate (SC) , see figure 1. More specifically, the CO2 molecules are efficiently activated by CO2-philic N-species of the cCTF. Concurrently to the activation of CO2, the POM catalyses the epoxidation of styrene. This is followed by the cycloaddition of the activated CO2 to the formed epoxide. CTFs exhibited great base 10 catalytic properties for the chemical conversion of CO2. High conversions and good recyclability would demonstrate the unique potential of this new multifunctional catalyst. Additionally, following the work of a previous student, an attempt will be made to embed the catalyst particles in a mesh of electrospun polyamide-4.6 (PA-4.6) nanofibres to increase the ease of catalyst recycling and prevent catalyst loss by filtration.11,12

Figure 1: Schematic representation of a) the envisaged structure of the POM@cCTF catalyst and b) the tandem reaction.

Ghent University, 2019

Experimental The following chemicals were used as received: acetic acid (Sigma-Aldrich, ≥99%), ACN (Fisher Scientific, HPLC grade), 2-bromo-5-cyanopyridine (TCI, >97.0%), CO2 (Air Liquide, ≥99.7%), chlorobenzene (Sigma-Aldrich, 99.8%), CHCl3 (Sigma-Aldrich, ≥99.5%), DCM (Sigma-Aldrich, 99%), 1,2-dichloroethane (ROMIL Chemicals, ≥99.8%), DMSO (Sigma-Aldrich, ≥99.5%), formic acid (Sigma-Aldrich, ≥95%), HCl (Carl Roth, 37wt.%), MeOH (Fisher Scientific, ≥99.5%), 1-Methyl-1H-imidazole (Sigma-Aldrich, 99%), polyamide-4.6 (Goodfellow, -), styrene (Sigma-Aldrich, ≥99%), tetrabutylammonium bromide (Sigma-Aldrich, ≥98.0%), THF (Sigma-Aldrich, ≥99.0%), tert- butylhydroperoxide (Sigma-Aldrich, 5.0-6.0 M in decane), and ZnCl2 (TCI, ≥98.0%).

Synthesis of the Cationic Dinitrile Linker; bpim The cCTF used in this research, from the work of Park et al.13, is synthesized from 1,3- bis(5-cyanopyridyl)-imidazolium bromide (bpim). This linker is prepared according to the recipe of Chen et al.14. 2-Bromo-5-cyanopyridine (2.176 g, 11.9 mmol, 2 eq.) and 1- methyl-1H-imidazole (472 µL, 5.9 mmol, 1 eq.) were charged into a Pyrex ampoule, which was degassed, flame sealed and heated up to 190°C at 1°C.min-1 in a muffle furnace and retained at that temperature for 18 h. After the reaction, the ampoule was cooled to room temperature (RT) and opened with a glass cutter. The collected crude black powder was dissolved in ca. 100 mL hot methanol (MeOH) to extract the pure compound. The linker dissolves poorly in MeOH and remains as a precipitate, which was filtered of and kept aside. The filtrate was allowed to cool to RT and put on the rotavapor until ca. 20mL MeOH remained, the formed crystals were filtered off. Important to note: if too much MeOH was evaporated, unreacted starting materials can precipitate. The obtained light brownish powders were added together, washed with chloroform, and dried for 24 h under vacuum at 120°C.

Synthesis of the bpim-based cCTF The synthesis method of the cCTF was adapted from the work by Park et al.13. Dried bpim (0.5 g, 1.42 mmol) and zinc chloride (ZnCl2, 0.95 g, 6.97 mmol) were loaded in a Pyrex ampoule and degassed under vacuum for 5 h to remove residual water. The ZnCl2 was dried beforehand under vacuum at 120°C overnight. The ampoule was flame sealed and placed in a muffle furnace and heated at a rate of 1°C.min-1 to 400, 500 or 600°C and kept at that temperature for 48 h. After 48 h, the ampoule was left to cool down to RT and opened with a glass cutter. The collected crude black powder was ground with a mortar and pestle in a minimal amount of water. The powder was added to a 250 mL round-bottom flask with 200 mL 1M HCl solution and a magnetic stirring bar. The mixture was refluxed at 110°C for 24 h under vigorous stirring to remove the ZnCl2. Subsequently, the mixture was filtered, and the powder was washed with water until the filtrate was neutral (ca. 3 x 100 mL water). Next, the powder was washed with acetone (3 x 100 mL) and left to dry in air for a little while. The powder was brought into a 250 mL round-bottom flask with 200 mL THF and stirred vigorously for 24 h. After filtration, the black cCTF powder was activated under vacuum at 150°C overnight.

Loading of the Vanadium-based POM The loading of the anionic VPOM in the cationic CTF is very straightforward, following the procedure from Ma et al.15. The VPOM was dissolved in 5 mL ACN in a 50 mL round-

Ghent University, 2019 bottom flask under magnetic stirring. A certain amount of cCTF was added and the mixture was left to vigorously stir for 48 h at RT. Afterwards, the mixture was filtered, washed with deionized water, and activated under vacuum at 120°C for 24 h. A series of VPOM@cCTFs with different loadings of the VPOM were prepared: 1. 11.2 wt.%: 12.62 mg VPOM in 100 mg cCTF. 2. 14.3 wt.%: 16.67 mg VPOM in 100 mg cCTF. 3. 20 wt.%: 25 mg VPOM in 100 mg cCTF. The initial catalytic tests were performed with the 14.3 wt.% VPOM@cCTF material and later with the 11.2 wt.% VPOM@cCTF. At 11.2 wt.%, 10mg of the catalyst should contain 1.12 mg or 7.72∙10-4 mmol VPOM given the VPOM structure formula of -1 [NH4]7[MnV13O38] (M = 1451 g.mol ). This corresponds to 0.01 mmol vanadium present in 10 mg 11.2 wt.% VPOM@cCTF provided the loading proceeds with a VPOM retention of 100 %. This means that the reaction would be catalysed using 1 mol% catalyst, assuming vanadium in the VPOM is the active catalytic site, a value that is preferably as low as possible.

Catalytic Set-Up The oxidative carboxylation of styrene was performed using Schlenk tubes. They were loaded with styrene (1 mmol, 112 µL), chlorobenzene as internal standard (1 mmol, 102 µL), ACN as solvent (3 mL), TBAB as co-catalyst (10 mol% to styrene, 0.1 mmol, 32.2 mg), TBHP (1.5 mmol, 300 µL of a 5M solution in decane), the VPOM@cCTF (10 mg), and are then finally purged with CO2 (3 bar). The reaction was proceeded at 75°C for 72 h. From the preliminary tests, it was concluded that the stepwise addition of TBHP to the reaction mixture increased the SC yield. For this reason, 30 µL TBHP solution was added at the start of the reaction in the morning. In the afternoon an additional 70 µL was added. This was repeated twice daily until a total amount of 300 µL TBHP solution was added. Samples for GC-MS analysis were taken at the beginning and at the end of the reaction to determine styrene conversion and product formation. Testing the base catalytic properties of the cCTF material in the cycloaddition of CO2 to styrene oxide (SO, 2.43 mmol, 278 µL) was done using the reaction conditions above. The only difference was that TBHP was not added and that the unloaded cCTF material was used. Samples for GC-MS analysis were taken at the beginning and at the end of the reaction to determine styrene conversion and product formation. The chromatographs were acquired on a Hewlett Packard 6890 GC system with the non-polar Agilent J&W Scientific DB-5ms column. The mass spectrometry detector was a Hewlett Packard 5973 Mass Selective Detector.

Electrospinning Set-Up The work of De Schoenmaker et al.3 serves the starting point for determination of the conditions for the electrospinning of PA-4.6. In this work PA-4.6 was successfully electrospun from a 15 wt.% solution consisting of 30 v.% of the non-solvent acetic acid and 70 v.% of the solvent formic acid. A WPI SP100iZ single-syringe infusion pump was utilized to pump the PA-4.6 solution out of an 18-gauge syringe in a horizontal position at a flux of 4 mL.h-1. The needle and metal collecting plate covered with Al-foil were placed 10 cm apart and a voltage of 15 kV was applied using a Glassman HV power supply. SEM, using a FEI Company Phenom Pro Desktop SEM, analysis of the electrospun fibres was used to analyse the morphology. The open source plug-in DiameterJ for the public domain image processing software ImageJ was used to determine the fibre diameters16.

Ghent University, 2019 Characterization and analysis N2-sorption experiments were performed on a Belsorp Mini at 77 K. Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) measurements were done using a ThermoScientific Nicolet 6700 FT-IR spectrometer. PXRD diffractograms were acquired on a ThermoScientific ARL X’Tra X-ray diffractometer using Cu Kα radiation of 40 kV and 30 mA. 1H and 13C NMR spectroscopy were performed on a Bruker 300 MHz AVANCE using non-deuterated DMSO as solvent and internal standard. The POM loading was determined using ICP-AES on an Optima 8000 spectrometer. The destruction was performed according to the WAC/III/B/001 standard procedure. The analysis was performed according to the WAC/III/B/010 standard procedure. Raman spectra were acquired on a Kaiser Optical Systems, Inc. RamanRxn Systems RXN1-532. TGA was performed on a Netzsch STA-449 F3 Jupiter-simultaneous TG-DSC analyser within a temperature range of 20-1000°C in an air atmosphere at a heating rate of 10C.min-1. BF- STEM-EDX was performed on a JEOL JEM-2200FS high-resolution scanning transmission electron microscope using a FEG operating at 200 kV.

Results & Discussion VPOM@cCTF synthesis and characterization Using 1H and 13C NMR spectroscopy it was confirmed that the linker synthesis was executed successfully, the reported and measured chemical shifts matched nicely. Using DRIFTS, the success of the trimerization reaction was confirmed. The nitrile stretching vibration at 2235 cm-1 is clearly present in the bpim spectrum, but not in the cCTF spectrum, indicating the success of the trimerization reaction. The expected in-plane triazine vibrations (at 1410 cm-1 and 1550 cm-1) on the other hand merged with the vibrations of the pyridine and imidazolium groups, resulting in broad peaks at 1650 cm-1 (C=C or C=N), 1365 cm-1 (C-N or C-C), and 800 cm-1 (heterocyclic aromatics). The broad peaks are the result of the harsh synthesis conditions resulted in a high degree of graphitization of the 17 framework and the formation of complex aromatic nitrogen-carbon moieties. N2-sorption of the cCTF materials synthesised at different temperatures revealed that the cCTF synthesized at 500°C had a high specific surface area (1833 m2.g-1) and pore volume (0.63 cm3.g-1), and was therefore chosen for further testing. All further references to cCTF thus refer to this material. The pore diameter of all three materials was around 2.30 nm, large enough for the POM to enter (diameter of 1 to 2 nm). Lastly, a 1:5 ratio of the monomer and ZnCl2 was used because it was determined to be the optimal ratio for the highest porosity, which is considered to be a significant contributing factor for high CO2 uptake, by Park et al. Using TGA it was confirmed that the material was thermally stable up to 400°C under an air atmosphere. The ionothermal synthesis method typically results in amorphous materials due to graphitization at the elevated temperatures used in this method. This is reflected in the PXRD diffractogram, that shows no sharp peaks, only a very broad one at 25.5°. ICP-AES analysis of the 14.3 wt.% VPOM@cCTF determined that the sample contained 4.03 wt.% V and 0.111 wt.% Mn. Calculating backwards from the wt.% of V gives a retained VPOM content of 8.83 wt.%. The 20 wt.% sample contained 6.37 wt.% V and 0.181 wt.% Mn corresponding 13.96 wt.% VPOM in the material. Analysis of the 11.2 wt.% VPOM@cCTF reported that the sample contained 4.52 wt.% V and 0.089 wt.% Mn. This corresponds to a VPOM loading of 9.9} wt.%. The loading can also be qualitatively assessed using the methods also used for the analysis of the linker and cCTF materials. For example, N2-sorption of the 11.2 wt.% loaded cCTF resulted in an expected type I isotherm,

Ghent University, 2019 but the maximum adsorbed N2 volume was lowered, see figure 2. Analysis of the data revealed a surface area decrease of ca. 300 m2.g-1 to 1542 m2.g-1 and a pore volume decrease to 0.53 cm3.g-1.

Figure 2: Overlay of the isotherms of cCTF and 11.2 wt.% VPOM@cCTF. Qualitative evidence that the retained materials is indeed the VPOM was obtained using PXRD. In the diffractogram of 20 wt.% VPOM@cCTF, the peaks of corresponding to the VPOM are superimposed on the cCTF spectra, clearly visible for the reflections at 28.25°, 32.0°, and 45.7°, see figure 3.

Figure 3: PXRD diffractograms of the cCTF, VPOM and 20 wt.% VPOM@cCTF.

Ghent University, 2019 Lastly, BF-STEM analysis of the 11.2 wt.% VPOM@cCTF with parallel EDX- mapping measurements allowed mapping of the N, V and Mn present in the cCTF powder to assess the distribution of these elements throughout the framework. An overlay of the mapping of the different elements is given in figure 5. The mapping shows clearly that the distribution of N and the VPOM is uniform throughout the VPOM@cCTF grains.

Figure 4: BF-STEM-EDX elemental overlay of 11.2 wt.% VPOM@cCTF. Catalytic results To assess the influence of each part of the catalytic system, a series of tests was performed in which each part of the catalytic system (VPOM, cCTF, TBAB, and VPOM@cCTF) and each possible combination of the parts were tested. The overall catalytic system was tested in triplet to assess the reproducibility of the results (at the 95% confidence level), giving a SC yield of 40.5±2.74%. Some interesting properties about the catalytic system can be concluded from this series: 1. The presence of the co-catalyst is paramount to the formation of SC, when it is not present, SC is not or barely formed. One could assume that any chloride ions present from the washing process during cCTF synthesis could activate the epoxide resulting in the formation of SC. Unfortunately, this is not the case, any chloride (or even bromide) ions present in the framework are not active. Our hypothesis is that a synergetic interaction between the electron-deficient ammonium N and the electron- rich epoxide oxygen activates the sterically free C-O bond resulting in efficient epoxide ring opening and subsequent cyclic carbonate formation. 2. The homogeneous addition of the VPOM has the effect of lowering benzaldehyde (BA) formation, the main side product. The heterogeneous addition of VPOM through VPOM@cCTF has the same effect. Additionally, the presence of VPOM in the framework resulted in a SC yield increase of ca. 7 %, compared to a system with only TBAB and cCTF. 3. The addition of cCTF to a reaction mixture with TBAB did not result in a significant increase in SC yield compared to the reaction with only TBAB. We hypothesize that as the VPOM is not present in the framework to catalyse the epoxidation, there is no local concentration increase of SO, meaning that the increased CO2 presence through interaction with the cCTF framework cannot be optimally utilized. However, the 7

Ghent University, 2019 presence of VPOM in the framework increases the SC yield with ca. 7 %, as mentioned in the point two. 4. Upon addition of homogeneous VPOM compared to an experiment with only TBAB, the SC yield drops with ca. 7 %, while the BA formation increases. One could assume that the increased oxidation rate and the low concentration of CO2 enhances BA formation at the expense of SC. ACN is chosen as the solvent for two reasons, (i) based on early tests it was observed that adding a solvent speeds up the reaction considerably, and (ii) PA-4.6 is stable in ACN. Determining the turnover number (TON) of the catalyst is at this stage of the research not possible. To calculate TON, the number of moles of catalytic sites must be known. In tandem catalysis systems this is difficult to define and depends on which reaction step is the rate limiting and the specific active sites of the catalyst. To test the base catalytic properties of the cCTF material as such in the cycloaddition reaction of CO2 to styrene oxide, four experiments were performed. A blank reaction, on with cCTF, one with TBAB, and one with both. From GC-MS analysis of the SO starting material was noticed that the bottle was severely contaminated with phenylacetaldehyde (PHAA), most likely formed by the isomerisation of the SO. The first two reactions did not result in the formation of any SC, as can be expected. From experiment three, it is clear that TBAB is required to initiate the cycloaddition through the activation of the epoxide ring. Surprisingly enough, both the SO and the PHAA seem to be converted in SC at about the same rate. This is interesting, because this seems to imply that an aldehyde group could be activated to produce a cyclic carbonate. Repeating this reaction with the addition of cCTF, results in a significant increase in SC yield, confirming our hypothesis that the local increased CO2 presence in the cCTF through interactions with the framework positively influence the reaction.

Result of the Electrospinning of PA-4.6 PA-4.6 was chosen for its thermal stability and excellent mechanical properties, its chemical stability was tested in the solvents ACN and 1,2-dichloroethane (1,2-DCE) under reaction conditions and as a reference under solvent free reaction conditions. From the SEM images in figure 6 can be seen that the nanofibres dissolved partly under solvent free (SF) conditions resulting in an elastic nanofibre clump. The opposite happened in 1,2-DCE, after drying, the nanofibre mat became very brittle and resulting in broken fibres. In ACN however, the fibres kept their morphology and did not dissolve or clump together. As an initial experiment to electrospin a [email protected], 10 mg of CTF-1 was dispersed in a 5 mL 15 wt.% PA-4.6 solution and electrospun at the above-mentioned conditions. In figure 7 SEM images are presented of the entangled CTF-1 particles in the nanofibres. To create a functional catalyst, a loading of ca. 10 mg the VPOM@cCTF in ca. 100 mg nanofibres (a 10 wt.% loading) must be achieved to create a catalytic carpet of appreciable dimensions. To that end, two solutions were made of 5 mg and 10 mg VPOM@cCTF in 1 mL 15 wt.% PA-4.6 solution. This would correspond to a loading of 3.88 wt.% and 7.75 wt.% respectively is the solution was fully electrospun. Although fibres were formed, the process was hindered by electrospraying of the solution resulting in black spots (the droplets contain considerably more black VPOM@cCTF) on the white fibre sheet as can be seen in figure 7.8a. Our hypothesis is that micrometre sized particles are too large to

Ghent University, 2019

Figure 6: Effect of reaction solvents on the stability of Figure 7: SEM images of the [email protected] proof- electrospun PA-4.6 mats as studied by SEM. of-concept. pass the Taylor cone uninterrupted, resulting in droplet formation at the needle tip and ultimately spraying of the solution, see figure 8b. A different strategy based on the effect of electrospraying was then attempted. The idea was to electrospin PA-4.6 on a rotating drum and electrospraying a VPOM@cCTF solution from a different angle. This requires the dispersion of the catalyst in a solvent that does not dissolve the nanofibres but has a high enough vapour pressure to quickly evaporate. THF was chosen and solutions of 10, 20, 30 and 40 mg VPOM@cCTF in 1mL THF were prepared. Due to unexpected interaction between the THF and the silicon tube connecting the syringe containing the solution and the needle, and an unexpected non-functional high voltage supply, this experiment unfortunately did not yield noticeable improvements, as can be seen in figure 7.8c. Given the above challenges for the development of an electrospun mat incorporating catalytic VPOM@cCTF particles, the obtained results are preliminary.

Figure 5: a) The result of electrospinning the 3.88 wt.% VPOM@cCTF@PA46 solution. b) The Taylor cone, crucial to ES, is repeatedly broken due to droplet formation. c) Although ES of PA-4.6 on the rotating drum proceeded as expected, the electro spraying of the VPOM@cCTF@THF solution proved problematic. 9

Ghent University, 2019 Conclusion & Future Work To the best of our knowledge, the POM@cCTF system for the oxidative carboxylation of styrene, or any alkene for that matter, presented here is the first of its kind. A cationic CTF (cCTF) material was successfully synthesised based on an imidazolium containing dinitrile linker using the well-known ionothermal synthesis method. This material was characterized using N2-sorption, PXRD, Raman, DRIFTS, TGA, and CHNS. Subsequently, the POM heptammonium tridecavanadomanganate (VPOM) was successfully retained inside the pores of cCTF through ion-ion interactions. The retention was qualitatively confirmed by N2-sorption, PXRD, Raman, DRITFS, and BF-STEM-EDX and quantitatively determined by ICP-AES. Using the mild conditions reported in this work, a styrene carbonate yield of ca. 40% can be achieved, while revealing some interesting properties about the reaction. Separate cycloaddition experiments of seemed to imply that aldehydes could be activated under these conditions to yield cyclic carbonates. Lastly, an attempt was made to embed this VPOM@cCTF in electrospun PA-4.6 nanofibres to create an easily recyclable catalytic system. Although the SC yield is not on par with the state-of- the-art, this material is the first of its kind, an attempt upon which can be built. For example, many more different POMs can be tested, different cationic linkers (e.g. a linear linker based on benzodiimidazolium18) or a more well-defined cCTF could be synthesized using recently developed mild synthesis methods for CTFs.19 The recyclability of the material is also a factor that still must be tested. And a first step to increase the electrospinning possibility could be to ball-mill the cCTF before POM loading (and test the impact of the reduced particle size on the catalytic reaction), using a needle with a larger gauge and/or electrospinning from a vertical position downwards to decrease the chance of cCTF build- up at the needle tip. Additionally, the synthesis of COFs using mild reaction methods typically results in nanoscale particles. Particles of this size, e.g. MOFs, have long been successfully electrospun.20 Reducing the size of CTF particles could be the first step to successfully embed them in electrospun nanofibres. A series of experiments with pure phenylacetaldehyde and different aldehydes, e.g. benzaldehyde or benzenepropanal, should also be performed to see if these substrates could indeed lead to cyclic carbonates.

Acknowledgments I would like to thank Prof. K. Leus (and the COMOC group), P. Dubruel (and the PBM group) and K. De Clerck (and the CTSE group) for their guidance and the use of their laboratories. Parviz, Tom and Eva for all their help in showing me the ropes. A. Kaczmarek for providing the VPOM, K. Haustraete for the STEM-EDX images, R. Blanckaert for ICP- AES analysis and F. Aliç and T. Planckaert for teaching me how to use the equipment. References 1 Fogg, D. E. et al. Coord. Chem. Rev. 248, 2365–2379 (2004); 2 Han, Q. et al. Nat. Commun. 6, 10007 (2015); 3 De Schoenmaker, B. et al. J. Nanomater. 2012, 1–9 (2012); 4 Olivier, J. G. J. et al. (2016); 5 Mikkelsen, M. et al. Energy Environ. Sci. 3, 43–81 (2010); 6 Aresta, M. et al. J. CO2 Util. 3–4, 65–73 (2013); 7 Working Group III of the Intergovernmental Panel on Climate Change. (Cambridge University Press, 2005); 8 Buysch, H.-J. in Ullmann’s Encyclopedia of Industrial Chemistry 7, 45–71 (Wiley-VCH Verlag GmbH & Co. KGaA, 2000); 9 Ma, H. et al. J. Am. Chem. Soc. 138, 5897–5903 (2016); 10 Roeser, J. et al. ChemSusChem 5, 1793–1799 (2012); 11 Verhoeven, L. (2017); 12 Leus, K. et al. J. Catal. 360, 81–88 (2018); 13 Park, K. et al. J. Mater. Chem. A 5, 8576–8582 (2017); 14 Chen, J. C. C. Organometallics 19, 5113–5121 (2000); 15 Ma, H. et al. J. Am. Chem. Soc. 138, 5897–5903 (2016); 16 Schneider, C. A. et al. Nat. Methods 9, 671–675 (2012); 17 Hug, S. et al. J. Mater. Chem. 22, 13956 (2012); 18 Du, J. et al. Microporous Mesoporous Mater. 276, 213–222 (2019); 19 Meier, C. B. et al. Polymer (Guildf). 126, 283–290 (2017); 20 Huang, Z. M. et al. Compos. Sci. Technol. 63, 2223–2253 (2003) 10

Contents

Page

Acknowledgments iii

Article iv

Contents xvi

List of Figures xvii

List of Tables xix

List of Abbreviations xxi

1 Introduction 1 1.1 Objectives ...... 2

2 Carbon Dioxide to Fine Chemicals 3 2.1 The Carbon Dioxide Problem ...... 3 2.2 Carbon Capture & Utilization (CCU) ...... 5 2.3 Carbon Dioxide Utilization in Industry ...... 8 2.4 Cyclic Carbonates ...... 9 2.4.1 Uses of Cyclic Carbonates in Industry & Niche Markets ...... 10 2.5 Oxidative Carboxylation of Styrene ...... 10 2.5.1 Tandem Reaction ...... 12

3 The Polyoxometalate 17 3.1 Introduction ...... 17 3.2 Structure, Synthesis & Properties ...... 18 3.3 POMs as Epoxidation Catalysts ...... 20

4 The Covalent Organic Framework 23 4.1 Introduction ...... 23

xv 4.2 The Covalent Triazine Framework ...... 25 4.2.1 The Cationic CTF ...... 26

5 Electrospinning 29 5.1 The Process of Electrospinning ...... 29 5.2 The Nanoﬁbre Composite ...... 30 5.3 Electrospinning of Polyamide-4.6 ...... 31

6 Materials & Methods 33 6.1 Synthesis of the Cationic Dinitrile Linker ...... 33 6.2 Synthesis of the bpim-based cCTF ...... 34 6.3 Loading of the Vanadium-based POM ...... 35 6.3.1 Heptammonium Tridecavanadomanganate ...... 36 6.4 Catalytic Set-Up ...... 37 6.5 Electrospinning Set-Up ...... 37

7 Results & Discussion 39 7.1 Analysis of the Cationic Dinitrile Linker ...... 39 7.2 Analysis of the bpim-based cCTF ...... 40 7.3 Analysis of the VPOM@cCTF ...... 43 7.4 Catalytic results ...... 47 7.5 Results of the Electrospinning of PA-4.6 ...... 51

8 Conclusion & Future Work 55

A Literature Study 57 A.1 From Carbon Dioxide to Fine Chemicals ...... 57 A.2 The Polyoxometalate ...... 59 A.3 The Covalent Organic Framework ...... 60 A.4 The Electrospinning ...... 61

B Materials & Methods 63 B.1 Used chemicals & equipment ...... 63 B.2 Analysis of the Cationic Dinitrile Linker ...... 66 B.3 Analysis of the VPOM@cCTF ...... 67

Bibliography 69

xvi List of Figures

1.1 Scheme of a) the POM@cCTF and b) tandem reaction...... 2

2.1 CO2 emissions by activity sector ...... 6 2.2 Structures of EC and PC...... 10 2.3 Oxidative carboxylation of styrene...... 11

3.1 a) PTA structure by Keggin, b) polyhedral representation of PTA. . . . . 18

4.1 Quaternary N-heterocyclic aromatic for cationic linkers...... 27 4.2 PTA doping of EB-COF:Br...... 28

5.1 ES set-up ...... 30 5.2 PA-4.6 Structure ...... 32

6.1 Schematic of the linker synthesis...... 34 6.2 Idealized cCTF synthesis reaction...... 35 6.3 Schematic representation of the VPOM loading...... 36

7.1 bpim linker structure with assigned NMR peaks...... 39 7.2 Isotherms of cCTF synthesized at diﬀerent temperatures...... 42 7.3 Overlay of the DRIFTS spectra of bpim and cCTF...... 43 7.4 TGA curve of the cCTF...... 44 7.5 Isotherms of cCTF and 11.2 wt.%...... 45 7.6 PXRD diﬀractograms overlay of cCTF, VPOM & VPOM@cCTF . . . . . 46 7.7 BF-SEM-EDX overlay of 11.2 wt.% VPOM@cCTF...... 46 7.8 SEM images of PA-4.6 before and after stability test...... 52 7.9 SEM images of [email protected]...... 53

xvii List of Figures

7.10 Overview of the spraying problem...... 54

A.1 The Keeling curve...... 57 A.2 Oxidative cleavage pathways to BA...... 58 A.3 Structure of silicotungstic acid as predicted by Pauling...... 59 A.4 Structure of heptammonium tridecavanadomanganate...... 59 A.5 COF linkages overview...... 60 A.6 CTF-1 synthesis...... 61 A.7 PVDF structure ...... 61

B.1 The 1H NMR spectrum of bpim...... 66 B.2 The 13C NMR spectrum of bpim...... 66 B.3 Overlay of the PXRD spectrum of VPOM and its predicted spectrum. . . . 67 B.4 Raman spectra overlay of cCTF, VPOM and 20 wt.% VPOM@cCTF . . . 67 B.5 IR spectra overlay of cCTF, VPOM and 11.2 wt.% VPOM@cCTF . . . . . 68 B.6 IR spectra overlay of cCTF, VPOM and 20 wt.% VPOM@cCTF ...... 68

xviii List of Tables

2.1 Industrial CO2 emisions in 2005 ...... 7

2.2 Six important CO2 transformations and the resulting products...... 8

2.3 Uses of CO2 and product lifetimes ...... 9 2.4 Catalytic oxidative carboxylation of styrene in literature...... 16

7.1 1H NMR spectroscopy of dinitrile linker...... 40 7.2 13C NMR spectroscopy of dinitrile linker...... 40 7.3 Porosity analysis of the cCTF materials ...... 41 7.4 Porosity analysis of the loaded cCTF materials ...... 44

7.5 Results of the cycloaddition of CO2 to styrene oxide ...... 49 7.6 Results of the catalytic oxidative carboxylation of styrene...... 50

A.1 The properties of EC, PC and other organic carbonates...... 58 A.2 The parameters inﬂuencing the ES process [221]...... 61

B.1 Used Chemicals ...... 65

xix

List of Abbreviations

BA Benzaldehyde BET theory Brunauer-Emmet-Teller theory BF-STEM Bright-ﬁeld scanning transmission electron microscopy bpim 1,3-Bis(5-cyanopyridyl)-imidazolium bromide

CCS Carbon Capture & Storage cCTF Cationic Covalent Triazine Framework CCU Carbon Capture & Utilization COF Covalent Organic Framework CTF Covalent Triazine Framework

DCM Dichloromethane, CH2Cl2 DMF Dimethylformamide

EDX Energy-Dispersive X-ray spectroscopy ES Electrospinning

HPA Heteropolyacid

IPCC Intergovernmental Panel on Climate Change

MOF Metal Organic Framework

NOAA National Oceanic and Atmospheric Administration

xxi POM Polyoxometalate PTA Phosphotungstic acid PTC Phase-transfer catalyst

RH Relative humidity; ratio of the partial pressure of water vapor to the equilibrium vapor pressure of water at a given temperature.

SC Styrene carbonate

+ – TBAB Tetrabutylammonium bromide, [CH3(CH2)3]4N Br + – TBAI Tetrabutylammonium iodide, [CH3(CH2)3]4N I TBHP Tert-butyl hydroperoxide, tBuOOH

xxii Chapter 1

Introduction

Tandem catalysis, in which multi-step chemical transformations are catalysed by multifunctional catalysts, has attracted increasing research attention. Selective multifunctional catalysts make a multi-step reaction go to the desired products in one reactor without the need for separation, puriﬁcation, and transfer of intermediates produced in each step. Great eﬀorts have been made to design heterogeneous catalysts for tandem reactions through the immobilization of metal complexes and nanoparticles on the surface of various supports such as polymers, zeolites, and molecular sieves. In this regard, metal-organic frameworks (MOFs) are at the top of the list of porous solid materials with the highest surface area and pore volume [1]. However, one of the main problems of MOFs in catalysis is their poor structural stability. Recently, covalent triazine frameworks (CTFs) emerged as a new type of porous materials. CTFs have similar characteristics in comparison to the well-known MOFs, but exhibit enhanced chemical stability. Here, a novel approach is devised based on the incorporation of polyoxometalates (POMs) in cationic CTFs as dual-functional materials to achieve a tandem catalyst for

the efficient conversion of CO2 and styrene into value-added styrene carbonate (SC), see figure 1.1 [2]. By embedding the anionic POM in our designed cationic CTF (POM@cCTF), this material can initiate the direct transformation of styrene to SC. More specifically, the

CO2 molecules are eﬃciently activated by CO2-philic N species of the cCTF. Concurrently

to the activation of CO2, the POM catalyses the epoxidation of styrene. This is followed by the cycloaddition of the activated CO2 to the formed epoxide. CTFs exhibited great base catalytic properties for the chemical conversion of CO2 [3]. High conversions and good recyclability would demonstrate the unique potential of this new multifunctional catalyst.

1 Chapter 1. Introduction

Additionally, following the work of a previous student, an attempt will be made to embed the catalyst particles in a mesh of electrospun nanoﬁbres to increase the ease of catalyst recycling and prevent catalyst loss by ﬁltration [4, 5].

Figure 1.1: Schematic representation of a) the envisaged structure of the POM@cCTF catalyst and b) the tandem reaction.

1.1 Objectives

1. Synthesize and characterize a N-rich cationic covalent triazine framework.

2. Load this cCTF with a POM, and analyse the yield of the loading process.

3. Optimize the POM@cCTF’s capabilities in the oxidative carboxylation of styrene.

4. Assess the catalysts recyclability and potential POM leaching.

5. Embed the catalyst in a mesh of electrospun nanoﬁbres.

6. Analyze the inﬂuence of the nanoﬁbres on the catalyst and its performance.

2 Chapter 2

Carbon Dioxide to Fine Chemicals

2.1 The Carbon Dioxide Problem

Mankind’s struggle with carbon dioxide and climate change has a long, complicated history [6]. But a good starting point is 1824, when Fourier calculated that the Earth would be far colder if it lacked an atmosphere [7]. In 1859 Tyndall, intrigued by this finding, tried to understand how gases could block heat radiation and thus affect the global climate. His work resulted in a series of papers and books demonstrating the influence of the concentration of these gases on the climate [8–12]. These results suggested to Arrhenius [13], Chamberlin [14–17], Callendar [18–23] and others that the Earth’s heat budget could be

controlled by changes in the CO2 concentration in the atmosphere. These theories were

further strengthened in 1956 by Plass, who calculated that the addition of CO2 into the atmosphere would have a signiﬁcant eﬀect on the radiation balance [24–26], and in 1957 by

Revelle, who found out that anthropogenic CO2 emissions would not be readily absorbed by the oceans [27]. The next major step was taken in 1960 by Keeling, who was the first to accurately measure the CO2 concentration in the atmosphere [28] and in 1961 already reported a rise. His work resulted in the famous Keeling curve, which is still being updated by the US National Oceanic and Atmospheric Administration (NOAA), see figure A.1. The pioneering 1967 computational work of Manabe and Wetherald predicted that if the amount ◦ of atmospheric CO2 would be doubled, the average global temperature would rise with 2 C [29]. In the view of many experts, this widely noted calculation gave the first reasonably solid evidence that greenhouse warming really could happen.

3 Chapter 2. Carbon Dioxide to Fine Chemicals

It is with the oﬃcial re-establishing of the NOAA in 1970 [30], and with the increased

available funds for exploring the relation between atmospheric CO2 and the climate, geology, etc. that the concept of global warming and climate change really took oﬀ. In the following years major events shaped the public opinion on and the science of climate change. Underneath, a brief overview is presented:

1972 The Club of Rome publishes The Limits to Growth. Although strongly criticized following its publication [31–37], this work has stood the test of time. Its “standard run” scenario, indicating that the limits to growth on earth would become evident by the 21th century leading to “sudden and uncontrollable decline in both population and industrial capacity”, has turned out to be surprisingly accurate [38–49].

1979 A US National Academy of Sciences (NAS) report demonstrated that it is highly ◦ credible that doubling the atmospheric CO2 concentrations will lead to 3±1.5 C of global warming [50].

1985 The Villach Conference declares consensus among experts that some global warming seems inevitable. They call on governments to consider international agreements to restrict emissions and increase research support [51].

1988 The Intergovernmental Panel on Climate Change (IPCC) is established. Its role is “to assess on a comprehensive, objective, open and transparent basis the scientiﬁc, technical and socio-economic information relevant to understanding the scientiﬁc basis of risk of human-induced climate change, its potential impacts and options for adaptation and mitigation” [52]. The increasingly harsh language of their reports is to urge international players to act on climate change [53–59].

2005 The Kyoto Protocol, signed in 1997 by 141 minor and major industrial nations except the US, was initiated [6, 60, 61]. Eﬀorts to reduce the emissions accelerated in Japan, Western Europe, US states, and by corporations.

2015 The Paris Agreement is signed, 196 nations pledge to set targets for their own greenhouse emission cuts and to report on their progress [62–64]. All while the mean global temperature is 14.8 ◦C [65], the highest in thousands of years, and the

level of CO2 in the atmosphere goes above 400 ppm (see ﬁgure A.1), the highest in millions of years.

4 Chapter 2. Carbon Dioxide to Fine Chemicals

2018 The Club of Rome launched a ten-point action plan at the European Parliament to “meet suitably ambitious reduction targets and ensure climate stability.” [66].

This small overview does not do justice to the work of countless people trying to understand, minimize, and solve global warming and climate change. The road ahead will be long and diﬃcult, while the solutions will have to be diverse, creative, and come from all facets of science and society. As material chemists, our focus is on developing new

materials to reduce CO2 production or to incorporate CO2 in new products to close the carbon loop and create a more sustainable way of life. Crucial in this last eﬀort is the development of materials and techniques that facilitate carbon capture on a large scale

and materials that allow this captured CO2 to be built into value added products to create an economic incentive for what has been denoted as Carbon Capture & Utilization (CCU).

2.2 Carbon Capture & Utilization (CCU)

Essentially three strategies exist to reduce the atmospheric CO2 emissions: (i) reduce the

CO2, (ii) use the CO2 and (iii) store the CO2 produced. The last two strategies are,

in practice, coupled to CO2 capturing techniques resulting in the names Carbon Cap- ture & Utilization (CCU) and Carbon Capture & Storage (CCS) respectively. Reducing

the amount of produced CO2 can be accomplished by increasing energy eﬃciency or by changing primary energy sources. On the other hand, CCU and CCS both involve the de-

velopment of new technologies for capturing and utilizing or sequestering CO2 respectively [67, 68].

CCU also brings on a new question: where to get high quality CO2? To answer this question a couple of points have to be kept in mind:

1. It’s essential that the recovered CO2 is from anthropogenic sources or directly from

the atmosphere. The target is to reduce net CO2 emissions in new synthetic technologies with respect to those that already exist, closing the C-loop.

2. The lower the concentration of CO2 in a gas mixture, the higher the energetic and economic cost will be of its separation.

3. The higher the concentration of gaseous pollutants in a gas mixture, the higher the energetic and economic cost will be of its separation.

5 Chapter 2. Carbon Dioxide to Fine Chemicals

4. For feeding an industrial application, a continuous point source is required.

The purpose of capturing CO2 is thus to produce a concentrated stream of high-pressure

CO2 that can be readily transported to a production facility.

Figure 2.1: Pie chart of the emission of CO2 by activity sector in 2013 [69].

Figure 2.1 shows the CO2 emission from various sectors, some of them are localized/s- tationary and continuous, such as the energy and industrial sectors, while others are de- localized/mobile and discontinuous, such as heating and transport. Keeping the previous points in mind, power stations or industrial plants seem to be the most interesting targets.

As the capture of CO2 from flue gases of coal-, oil- or gas-fired power plants is a mature technology which is commercially available, they are an attractive option to produce such a concentrated stream. There are three of these technologies available: (i) post-combustion capture, (ii) pre-combustion capture and (iii) oxygen-fired combustion.

As shown in table 2.1, ca. 3350 Mt of CO2 is annually emitted from industrial sectors that may represent a convenient CO2 source (values from 2005). As some of them, like ammonia production and fermentation are already exploited, there is over 3000 Mt/y of

CO2 now accessible for use (ca. 8.3 % of the global CO2 emissions), representing a large reservoir available for exploitation.

Expanding carbon capture technologies to atmospheric CO2 is an interesting option as it would decouple the use of CO2 from its source. The total annual anthropogenic

CO2 emission is about 36 Gt [72]. Combine this with the fact that there is already an excess of more than 900 Gt of CO2 in the atmosphere, nearing 1 Tt, means that there is a huge reservoir of untapped CO2 ripe for exploitation [73]. A 2011 analysis of this Direct $ Air Capture (DAC) technology estimated that the cost would be 600/tCO2 [74]. A recent $ study calculated that DAC cost has decreased to around 94–232 US /tCO2 , suggesting that

6 Chapter 2. Carbon Dioxide to Fine Chemicals

Table 2.1: Industrial emissions of CO2 in 2005 (excluding power stations) [70, 71].

Industrial sector MtCO2 /y produced

Ammonia 160 Cement >1000 Ethene and other petrochemical processes 155 Ethene oxide 10–15 Fermentation >200 Iron & steel ca. 900 LNG sweetening 25–30 Oil reﬁneries 850–900

the technology is inching closer towards commercial viability [75, 76]. In fact, in 2017 the Swiss company Climeworks build the world’s ﬁrst commercial-scale DAC plant [77].

Following its capture, the usage of CO2 can be divided into two groups: (i) those using its physical aspects (e.g. ﬁre extinguishers, in beverages, inert atmosphere, etc.) and (ii) those using its chemical aspects (i.e. as a reactant). Even though CO2 is an abundant and renewable carbon source only a few industrial processes utilize it as a raw material for its chemical aspects. The reason for this is that the C-atom in CO2 is in its most oxidized state and therefore relatively unreactive, requiring a large input of energy, typically thermal, to transform it into other chemicals wherein the C-atom has a lower oxidation number. There are however four ways of altering this:

1. Using high energy starting materials such as hydrogen or organometallics.

2. Choosing low energy synthetic targets (mainly chemicals in which C is still in its +IV oxidation state).

3. Removing a compound on the product side to drive the reaction to completion (i.e. Le Chˆatelier’sprinciple).

4. Supplying physical energy to the reaction, i.e. light or electrons.

This results in six important ways to transform CO2 in value added products, see table 2.2.

7 Chapter 2. Carbon Dioxide to Fine Chemicals

Table 2.2: Six important CO2 transformations and the resulting products.

Transformation Products

Chemical transformations Carbonates, carbamates, etc.

Photochemical reductions CO, HCO2H, CH4

(Electro)chemical reductions CO, HCO2H, MeOH

Biological conversion EtOH, Sugar, CH3CO2H

Reforming CO + H2

Inorganic transformations Carbonate salts (M2CO3)

2.3 Carbon Dioxide Utilization in Industry

Excluding enhanced oil recovery, ca. 200 Mt of CO2 is annually used in industry for diﬀerent chemical product classes and applications (see table 2.3). This use is only ca. 0.5 % of the total annual anthropogenic CO2 emission. As previously mentioned, CO2 is a stable ◦ −1 molecule (∆Gf = −396 kJ mol ), leading Aresta et al. to divide the reactions in which it is involved in two categories [78]:

1. Low energy processes, in which C maintains its +IV oxidation state or is lowered by only one unit. Typical products are: carboxylates, linear and cyclic esters, carbamates, organic carbonates, ureas, polycarbonates and polyurethanes, inorganic carbonates, and any compounds alike. As can be seen from table 2.3, ca. 192 Mt is

yearly used for these types of processes, representing the majority of industrial CO2 utilization.

2. High energy processes, in which C goes down at least two oxidation state units below +IV. Products are: formic acid, CO, formaldehyde, methanol, methane, hydrocar- bons, etc. In these cases, energy is required to drive the reaction. This energy can be in the form of electrons (electrochemical reduction), hydrogen (hydrogenation reactions), metals, radiation (photochemical conversion), and so on, or combinations thereof (e.g. photo-electrochemical reduction).

These products roughly correspond to chemicals and fuels respectively. Although the market for fuels is about 12–14 times larger than the one for chemicals, it is worth noting that chemicals may have a higher molecular complexity than fuels, increasing their

8 Chapter 2. Carbon Dioxide to Fine Chemicals

value [71]. The industrial use of CO2 emissions would thus be much higher if it was more eﬃciently incorporated into these products [70, 73, 79, 80]. To that end a substantial

amount of research is dedicated to eﬃciently produce these materials using CO2 as a starting product or as an intermediary. An interesting class of chemicals, of which a large increase in production/demand is expected (see table 2.3), is the (cyclic) organic carbonates.

Table 2.3: Present, short-term and future use of CO2 and the product lifetimes [70, 71, 73].

Amounts in Mt unless otherwise speciﬁed.

Chemical Actual pro- CO2 used Short-term CO2 needed Long-term Product life-

product class duction forecast CO2 use time or applica- forecast tion

Urea 155 114 180 132 – Six months Methanol 50 8 60 10 Gt Six months Organic car- 0.2 0.005 >2 0.5 100 Decades to bonates centuries Polycarbonates 4 0.01 5 1 – Decades to centuries Carbamates 5.3 0 >6 1 – Six months Polyurethanes >8 0 10 0.5 – Decades to centuries Formic acid 0.6 0 1 0.9 – Six months Inorganic car- 200 50 250 70 – Decades to bonates centuries Salicylic acid 0.06 0.02 – – 0.1 Six months Technological – 28 – 80 – Days to years

Total 200 296 Gt

2.4 Cyclic Carbonates

An organic carbonate, also known as a carbonic ester, carbonate ester or an organocar- bonate, is the ester of carbonic acid (HO(CO)OH). This functional group consists of a

carbonyl group ﬂanked by two alkoxy groups: R1O(C–O)OR2. In the case that the two R- groups are connected through an alk(en)yl-chain the structure is called a cyclic carbonate. The simplest of these are ethylene and propylene carbonate (EC, respectively PC), both consisting of a 5-membered ring (see ﬁg. 2.2). These are also the only cyclic carbonates produced at an industrial scale [81]. They are typically synthesized by reacting ethylene

9 Chapter 2. Carbon Dioxide to Fine Chemicals

or propylene oxide with CO2 under pressure using catalysts such as alkali metal iodides, ammonium and phosphonium salts, Lewis acids, heavy metal complexes, organometallic compounds, or ion exchangers with ammonium or phosphonium groups [82]. For the physical properties of EC, PC and other industrially relevant organic carbonates, see table A.1.

Figure 2.2: Structures of a) ethylene carbonate (EC) and b) propylene carbonate (PC).

2.4.1 Uses of Cyclic Carbonates in Industry & Niche Markets

An extensive list for the use of 5-Membered cyclic carbonates can be found in Ullmann’s Encyclopedia of Industrial Chemistry [82], with EC and PC being extensively employed as solvents for aromatic compounds, polymers, and salt like compounds. Some other (potential) uses that are less to not industrially applied (but are being investigated) are e.g. as monomers to polymerize into a variety of homo- and copolymer polycarbonates with high molecular masses, extractants, plasticizers, electrolytes, intermediates in organic synthesis, spinning dopes for synthetic ﬁbres, speciﬁc solvents for pharmaceutical and cosmetic preparations, or other niche usage applications.

2.5 Oxidative Carboxylation of Styrene

The formation of SC out of styrene, i.e. the oxidative carboxylation of styrene, is an orthogonal tandem reaction [83] composed of two steps, see ﬁg. 2.3:

1. The oxidation of styrene to styrene oxide by an oxidizing species, e.g. O2,H2O2 or tert-butylhydroperoxide (TBHP), i.e. the epoxidation of styrene.

2. The cycloaddition of CO2 to the epoxide ring forming the 5-membered cyclic carbonate, styrene carbonate.

10 Chapter 2. Carbon Dioxide to Fine Chemicals

Figure 2.3: The oxidative carboxylation of styrene to SC.

Styrene is an interesting substrate to be investigated due to its tendency to over- oxidize, forming benzaldehyde (BA). This over-oxidation is based on the oxidative cleavage of the C–C bond of the epoxide ring typically seen when using peroxo-based oxidants such as TBHP and to a lesser extent with H2O2 (see fig. A.2 for two proposed pathways of how this cleavage could be performed by TBHP) [84]. Other typically formed side-products are benzoic acid (formed through BA, although this is a slow reaction and therefore not frequently observed), phenylacetaldehyde (formed by isomerisation of styrene oxide, but produced to a lesser extent than BA). If there are (trace) amounts of water present during the epoxidation, it can possibly result in the hydrolysis of styrene oxide into 1-phenylethane-1,2-diol, especially when acids are present [85]. The difficulty of selective styrene epoxidation results in many systems sacrificing conversion over selectivity or vice versa. Limiting this over-oxidation process is an important step in the orthogonal tandem reaction. In their 2004 review, Fogg and dos Santos noted that “orthogonal reactions are characterized by their mutual independence. Orthogonal tandem catalysis, by analogy, involves two or more functionally distinct and (in principle) non-interfering catalysts or pre-catalysts, all of which are present from the outset of reaction.” [83]. The catalytic oxidation and cycloaddition reactions are mutually independent and thus orthogonal, with the non- interfering catalysts being a polyoxometalate (POM) and a covalent triazine framework (CTF) in the presented master thesis. The POM catalyses the epoxidation reaction and independently, an organic bromide salt (acting as a co-catalyst, opening the ring) and

covalent triazine framework (CTF, adsorbing CO2) catalyse the epoxide ring-opening fol-

11 Chapter 2. Carbon Dioxide to Fine Chemicals

lowed by the cycloaddition of CO2 and ring-closing. Both catalysts will be present in the same system to create a single catalytic material, the anionic POM will be anchored onto a cationic CTF through ion-ion interactions. Both independent reactions have been extensively investigated. The homogeneous epoxidation of olefins in a two-phase system, composed of an aqueous H2O2 solution and an organic solvent carrying the olefin, catalysed by a POM and a phase-transfer catalyst is known as the Ishii-Venturello chemistry [86–92]. Many heterogeneous systems for styrene epoxidation are also investigated, including, but not limited to polymer-supported Schiff base and Salen complexes [93–96], hybrid metal-organic frameworks [97–101], functionalized mesoporous silicas [102–105], zeolites [106, 107], pilar interlayered clay [108], etc. Additionally, the design of heterogeneous catalysts for the cycloaddition of CO2 to epoxides has been heavily investigated, resulting in many systems based on the above mentioned supports [109–119]. An alternative pathway for the oxidative carboxylation of olefins is via the oxy- bromination route first proposed by Eghbali and Li [120]. A terminal olefin reacts with a bromine source (e.g. N-bromosuccinimide (NBS)) and hydroxyl source (e.g. H2O/weak base or H2O2) forming a bromohydrin group. This is followed by a base-catalysed carboxylation step resulting in the olefin carbonate (e.g. 1,8-diazabicyclo[5.4.0]undecenc-7-ene (DBU) can be used as the base). This reaction pathway has been explored for the synthesis of SC in a flow set-up separating the two reactions [121], or as a one-pot synthesis [122–124].

2.5.1 Tandem Reaction

The orthogonal nature of these reactions allows the set-up of flow systems for the oxidative carboxylation of styrene. In 2017, Sathe et. al. [125] developed a “protocol utilizing packed bed flow reactors in series to couple rhenium catalysed olefin epoxidation and aluminum catalysed epoxide carboxylation in a single sequence.” While the yield of SC exceeded 98 %, recycling of the homogeneous rhenium catalyst was problematic. However, they believe that heterogeneous supported catalysts can be applied in this protocol, facilitating catalyst recovery and reuse. Separation processes are required in this type of two-step conversion of alkenes into cyclic carbonates. These processes have substantial energetic costs that could be avoided by exploiting the orthogonal nature of these reactions to run them in

12 Chapter 2. Carbon Dioxide to Fine Chemicals tandem as a “one-pot” synthesis. This is however not as straightforward as it sounds, as the epoxidation and cycloaddition reactions have diﬀerent optimal working conditions and possible side reactions. Many diverse catalytic systems have been investigated, based on:

1. One catalyst for both reactions.

2. Two catalysts, one for each step, added preliminary a.k.a. a single-step protocol.

3. Two catalysts, one for each step, added sequentially a.k.a. a multi-step protocol.

In which each catalyst can be either homogeneous or heterogeneous.

One catalyst

A few systems in this regard have been examined, however they did not demonstrate good SC yields. For example, in 1987, Aresta et al. carried out the reaction with a homogeneous

Rh(I)-complex and O2 as oxidant, resulting in a yield of 20.5 % to 30 % [126]. Following this in 2000, Aresta et al. investigated the activity of a whole series of heterogeneous metal- oxide catalysts, with Nb2O5 giving a yield of 4.5 % [127]. In 2002, upon addition of NbCl5 as a homogeneous co-catalyst, the yield increased to 11 %, still using O2 as the oxidant [128]. In 2004, a yield of 39 % was achieved by Sun et al. using molten tetra-butylammonium bromide (TBAB) as solvent and catalyst with TBHP as the oxidant [129].

Two catalysts in a single-step protocol

Increased yields are observed for systems using peroxides as the oxidant and diﬀerent catalysts for each reaction. In 2005, Sun et al. achieved a 42 % yield with a heterogeneous

Au/SiO2 epoxidation catalyst, combined with a homogeneous ZnBr2/TBAB cycloaddition catalyst using cumene hydroperoxide as the oxidant under mild conditions (80 ◦C, 10 bar

CO2 for 4 h) [130]. Four years later, Wang et al. reported a multi-component system consisting of a heterogeneous Au/Fe(OH)3 and a homogeneous ZnBr2/TBAB catalyst that ◦ yielded 53 % SC under mild conditions (80 C, 10 h, and 40 bar CO2) with TBHP as the oxidant. In this reaction, CO2 acted both as reagent and as solvent [131]. In 2010, Bai et al. reported on the aerobic oxidative carboxylation of styrene with a yield of 76 % using a homogeneous Ru-metalloporphyrin, [Ru(TPP)(O)2], catalyst and tetra-butylammonium iodide (TBAI) as co-catalyst [132]. Another attempt at the aerobic

13 Chapter 2. Carbon Dioxide to Fine Chemicals

oxidative carboxylation of styrene was reported by Carvalho Rocha et al. in 2015 [133]. A Mn(Salen)-complex is combined with choline chloride (a quaternary ammonium salt) as homogeneous catalysts for the direct conversion of styrene to SC in the presence of

O2 and CO2 using isobutyraldehyde (IBA) as a sacriﬁcial reductant. Unfortunately the direct transformation of styrene to SC using both catalysts was inadequate since no SC was detected. In the same year, Kumar et al. functionalized chitosan coated magnetic

nanoparticles with a [Co(acac)2]-complex and quaternary triphenylphosphonium bromide + – ([P Ph3Br ]). This material was used as a heterogeneous dual-functional catalyst with

CO2,O2, and IBA, yielding SC with 67 % [134]. As O2/CO2 mixtures are cheap and widely available, the direct aerobic oxidative carboxylation of alkenes deserves further research. In table 2.4, a more detailed overview is given of the most important systems for the oxidative carboxylation of styrene. Yields of 80 % and more are achieved, with the state- of-the-art system being developed by Han et al. in 2015 [1] . They achieved the enantioselective oxidative carboxylation of styrene using a POM/MOF hybrid material, a PO- 6– MOF. By incorporating Keggin-type anions ([W12O40] ), Zn(II) ions, NH2-functionalized

bridging links (NH2-bipyridine), and asymmetric organocatalytic groups (pyrrolidine-2-yl- imidazole) within one single MOF, they designed and synthesized two new enantiomorphic

POMOFs, ZnW–PYI1 and ZnW–PYI2. SC yields of 92 % with a 80 % enantiomeric excess were reported using 0.1 mol % catalyst and 1 mol % TBAB co-catalyst under mild ◦ conditions (solvent-free at 50 C, for 96 h, and 5 bar CO2). TBHP is used as the oxidant at a 2:1 ratio to styrene. Simultaneously Maksimchuk et al. reported on a mesoporous titanium-silicate, Ti-MMM-E, that catalysed the reaction with a 64 % yield, also under ◦ mild, solvent-free conditions (70 C, 48 h, and 8 bar CO2) using TBHP as the oxidant [135]. From an industrial viewpoint, the more benign an oxidant, the better, e.g. meta- chloroperoxybenzoic acid (m-CBPA) is typically used in the epoxidation of alkenes (the Prilezhaev reaction), but is too costly to ever be applied. Molecular oxygen on the other hand is the most benign/greenest oxidant known and has been used in this reaction, but it is also the least reactive and often requires a sacriﬁcial reductant, e.g. IBA. Such systems have been rarely investigated in recent years because of the following disadvantages:

(i) sacriﬁcial aldehydes negate the superiority of using O2, an environmentally and econom- ically benign oxidant, and (ii) the reactions undergo a free radical mechanism, tending to yield complicated by-products and thus creates tedious product separation problems [136].

H2O2 also has its limitations, mainly its hydrophilicity mixing badly with the hydrophobic

14 Chapter 2. Carbon Dioxide to Fine Chemicals styrene phase. TBHP, however not ideal, could be used to synthesize some of the niche cyclic carbonate chemicals. Tert-butanol that is released from TBHP during the reaction can be converted into commercially valuable chemicals, e.g. dehydrated to isobutene and etheriﬁed with methanol or ethanol for the production of octane boosters [137].

Two catalysts in a multi-step protocol

The same “two catalyst” idea, but with a twist, is using a multi-step protocol. Herein the second catalyst and CO2 are loaded into the reaction mixture upon ﬁnishing the epoxidation reaction. In 2013, Srivastava et al. succeeded in achieving a SC yield of 33 % using such a multi-step protocol with a Ti-MCM-41 epoxidation and N,N-dimethylaminopyridine (DMAP) cycloaddition co-catalyst [138]. Five years later, Dias et al. achieved a 70 % yield with a homogeneous Mn(III)-porphyrin epoxidation and a Cr(III)-porphyrin cycloaddition catalyst. The corresponding hybrid metalloporphyrin magnetic nanocomposites used as heterogeneous catalysts gave a yield of 52 % [139].

Unique systems

Other unique systems are also investigated. For example, Ono et al. examined an ionic liquid (IL) system in 2007. A SC yield of 83 % was obtained by coupling methyltriox- orhenium (MTO) for the epoxidation of styrene with an urea-H2O2 oxidant mixture and a Zn[EMIm]2Br4 IL for the cycloaddition reaction, while using [BMIm]BF4 as a solvent.

However, the instability of Zn[EMIm]2Br4 in the presence of H2O2 and the negative effect of CO2 on the epoxidation, resulted in the adoption of a multi-step protocol in which

Zn[EMIm]2Br4 and CO2 were added after epoxidation [140]. In 2013, Gao et al. presented an eﬃcient electrochemical route for the conversion of oleﬁns into cyclic carbonates. The synergistic action of I2 and NH3 electrochemically generated in situ from NH4I opened up an oxy-iodination (comparable to the oxy-bromination mechanism mentioned above) route. This system could smoothly transform styrene into SC under mild conditions with an isolated yield of 90 % [141].

15 Table 2.4: Overview of the catalytic oxidative carboxylation of styrene in literature.

Year Catalyst Styr (mmol) Oxidant Cat:Styr:Ox PCO2 (bar) Additive React. Cond. Yield (%) Ref.

1987 100 mg Rh(I)- 17.5 0.5 bar O2 1:100:- 0.5 - Multiple hours @ 20.5 to 30 [126] complex 40 ◦C in 30 mL THF (0.175 mmol Rh(I)) ◦ 2002 119 mg Nb2O5 17.5 1 bar O2 1:39:- 49 57 mg NbCl5 12 h @ 115 C in 11 [128] (0.45 mmol) (0.21 mmol) 10 mL DMF 2004 645 mg TBAB 17.3 34.6 mmol TBHP 20:173:346 10 - 6 h @ 80 ◦C in 39 [129] (2 mmol) molten TBAB ◦ 2010 37.5 mg 1.25 5 bar O2 1:25:- 11 0.1 mmol TBAI 60 h @ 30 C in 89 [132]

[Ru(TPP)(O)2] 5 mL EtOH (0.05 mmol) 2013 5 mg Cr-MIL-101 0.2 0.3 mmol TBHP 1:20:30 8 0.02 mmol TBAB 24 h @ 25 ◦C in 7 [142] (0.01 mmol Cr) 2 mL DCM 2015 300 mg Ti-MMM-E 4.8 7.2 mmol TBHP 1:52:77 8 0.48 mmol TBAB 48 h @ 70 ◦C 64 [135] (0.093 mmol Ti) 2015 37.9 mg ZnW-PYI 10 20 mmol TBHP 1:1000:2000 5 0.1 mmol TBAB 96 h @ 50 ◦C 92 (ee 80) [1] (0.01 mmol) 2018 190 mg MOF-892 1.31 2.48 mmol TBHP 1:17:32 1 2 mmol TBAB 9 h @ 80 ◦C 80 [143] (0.0786 mmol) Chapter 3

The Polyoxometalate

3.1 Introduction

Polyoxometalates (POMs) are composed of cations and polyanion clusters with extreme structural diversity [144]. These clusters consist of three or more oxometal polyhedra

(MOx ) linked together by shared oxygen atoms to form closed 3-D frameworks, under- standably also called polyoxoanions. The metal atoms are usually group 6 (e.g. Mo, W) or group 5 (e.g. V, Nb, Ta) transition metals in their highest oxidation states. The chemical behaviour of Mo(VI), W(VI), and V(V) in acidic aqueous solutions, dominated by the formation of these polyoxoanions, has been recognized since the 1820s. In 1826, Berzelius reported the formation of blue products from the reaction of molybdate with phosphate or arsenate, these were the first synthesized mixed-valence polyoxoanions, now called heteropolymolybdates [145]. Heteropolytungstates were discovered and characterized by Marignac in 1862 [146] and by the 1930s hundreds of such compounds were described in the Gmelin volumes [147, 148]. Around the same time, Keggin determined the first correct structure of a POM, namely phosphotungstic acid (PTA, [H3PW12O40] · 6 H2O), based on 32 powder XRD lines [149]. The progress in structural determination was slow until the 1980s when advances in X-ray crystallography, Raman spectroscopy, 1- and 2-D metal and 17O NMR, and the development of new analytical techniques allowed the establishment of important solid/solution structure links. With more than 70 different hetero-elements confirmed, with coordination numbers ranging from 3 to 12 and dozens of stoichiometries, the number of known POMs ranges in the hundreds. The (potential) applications of POMs reach from analytical and clinical chemistry (e.g. Mo-blue), homo- and heterogeneous

17 Chapter 3. The Polyoxometalate

(electro/photo)catalysis to solid-state applications (e.g. electronic/protonic conductors, nanomagnets, etc.), and medicine (as antitumoral, -viral, and -retroviral drugs) [150–158].

3.2 Structure, Synthesis & Properties

Although synthesized since 1826, the structure of POMs has long been the subject of speculation. In 1908, Miolati suggested structures for heteropolyacids (HPAs, the term HPAs is used for the hydrogen (or acid) forms of a POM) based on Werner’s coordination theory [159, 160]. Rosenheim further developed these proposed structures and applied them in the systematization of HPAs [161]. Pauling on the other hand argued that the Miolati-Rosenheim conception was far from satisfactory. It provided no explanation for the characteristic properties of these HPAs and their salts, and the single deﬁnite prediction it made, concerning the number of acidic hydrogens, disagreed with experimental results.

He proposed structures that were based on corner-sharing of MO6 octahedra (an example is given in figure A.3) [162]. When Keggin determined the actual structure of PTA, it was revealed that the more compact edge-shared polyhedral arrangements were more impor- n – tant, see figure 3.1 [149]. POMs with such general [XM12O40] structure are now called Keggin anions. The central X atom is a heteroelement in a tetrahedral configuration, e.g. 3– 4– 3– PO4 , SiO4 , AsO4 .

Figure 3.1: a) The structure of PTA as drawn by Keggin in 1933, including original description [149]. b) A modern polyhedral representation of PTA [163].

Since Keggins discovery many more POM structures have been determined, e.g. the

18 Chapter 3. The Polyoxometalate

2– n – Lindqvist structure ([M6O19] ) [164], the Wells-Dawson ([X2M18O62] ) [165, 166], the n – n – Anderson-Evans ([XM6O24] ) [167, 168], the Dexter-Silverton ([XM12O42] ) [169], the n – n – Weakley-Yamase ([XM10O36] ) [170, 171], the Allman-Waugh structures ([XM9O32] ) [172, 173], etc. From these discoveries, researchers learned that polyoxometalate structures are governed by two general principles [151]:

1. Each metal atom occupies an MOx coordination polyhedron (most commonly an octahedron or square pyramid) in which the metal atoms are displaced, as a result of the M–O π bonding, towards those polyhedral vertices that form the surface of the structure.

2. Structures with MOx octahedra that contain three (or more) free vertices are, in general, not observed (a.k.a. the Lipscomb restriction).

The typical building blocks of POMs are thus polyhedral units of 4 to 7-coordinated metal centres. These units share edges and/or vertices, or, less commonly, faces. It is not uncommon for the structures of some POMs to be derived from larger POM’s structures by removing one or more addenda atoms and their corresponding oxide atoms. These defective structures are called lacunary structures (from the word lacuna, meaning “a small opening; a small pit or depression”) [144]. Their long history has resulted in a complex nomenclature that has to consider both a name for the class as a whole and the systematic naming of individual species. Origi- nally a distinction was made between “ispoly” and “heteropoly” anions. Isopolyanions are 6– made up of a single type of metal-oxide polyhedra (e.g. the Keggin anion [(H2)W12O40] ), while heteropolyanions include polyhedra of heteroatom oxides (e.g. the Keggin anion of 3– PTA, [PW12O40] ). The distinction between the two groups is entirely artificial though and could lead to internal contradictions such as “mixed isopolyanion” to describe species 6– 4– as β-[Mo6V2O26] (isostructural with the isopolymolybdate β-[Mo8O26] ) or consider 6– [(H2)W12O40] (which is chemically and structurally equivalent to a heteropolyanion like 4– [SiW12O40] ) as an isopolyanion. For these reasons, less ambiguous names such as “polyoxoanions” and “polyoxometalates” are now more commonly used. Pope and Müller(re- spected names in the field) propose the term “metal-oxygen clusters”, as it can be used for the whole field and emphasizes the presence of metal centers [151]. The systematic nomenclature is the terrain of the IUPAC and has seen multiple iterations, the latest by Jeannin in 1998 [174].

19 Chapter 3. The Polyoxometalate

POMs are synthesized via a condensation process called “olation”. The oxides of group 5 or 6 metals, d0 metals, dissolve at high pH giving orthometalates. As the pH is lowered, the protonated orthometalates give oxide-hydroxide compounds. These condense via the loss of water to form M –O–M and M–M linkages. In chapter 3 of Inorganic Syntheses Volume 27, the synthesis of a whole series of these early transition metal POMs is described, all following this overall procedure starting from the orthometalate salts [175]. As POMs are clusters of highly oxidized transition metals with strong M–O bonds through d-orbital π bounding, they are known to possess excellent thermal and oxidative stability. They exhibit fast reversible multielectron redox transformations under mild conditions, making them highly efficient in oxidation reactions. This allows them to be used in gas-phase oxidation reactions conducted at high temperatures. These properties have resulted in their industrial application for the gas-phase oxidation of methacrolein or the direct oxidation of ethylene to acetic acid [176]. POMs in their HPA form also show strong and controllable Brönstedand Lewis acidity, making them interesting for acid catalysis. This unique behaviour makes them ideal candidates for the catalysis of Friedel-Crafts acy- lations, (trans)esterifications, reversible hydrolysis, etc. While POMs are best known for these two properties, their unique electronic structure, self-assembly capabilities and versatility have resulted in an research output exploring their possible catalytic properties (shortly mentioned above) [136]. In the interest of this thesis, the focus will be on the catalytic epoxidation of alkenes.

3.3 POMs as Epoxidation Catalysts

While POM chemistry has been around since the 1800s, initial research was focused on their synthesis and structural elucidation. It was only in the 1970s that, together with the advances in analytical techniques, POM catalytic research really took off. One of the earliest and most examined fields are selective oxidation reactions. This is not trivial, as selective oxidation affords various organic building blocks with versatile functional groups highly useful in industrial chemistry. As the chemical industry is moving away from the traditional wasteful oxidants (e.g. HNO3, HIO4, peroxy acids, etc.) towards greener oxidants

(H2O2 and O2) the need for environmentally benign catalysts to activate these species is increasing. POMs are ideal candidates due to their previously mentioned characteristic properties.

20 Chapter 3. The Polyoxometalate

Epoxides on the other hand are one of these versatile functional groups. They are usually provided by the initial oxidation of (terminal) alkenes, but are highly sensitive to hydrolysis in the presence of water. While this results in the formation of diols, also of great industrial importance, high selectivity for one or the other is preferred. Venturello-Ishii

systems are famous for the epoxidation of alkenes with POM catalysts and H2O2 oxidant [86, 87, 91, 177, 178]. In these bi-phasic systems, water-soluble HPAs were added to an

aqueous H2O2 solution and mixed with an organic phase carrying the alkene. Quaternary ammonium salts, modified by long-chain alkyl groups, were introduced into the systems as phase-transfer catalysts (PTCs). Under these conditions, the HPA reacts with the H2O2 forming active peroxo species that selectively oxidize the olefin to the epoxide. The olefin is brought into the aqueous phase, and the epoxide out of the aqueous phase, by the PTC. Optimized Venturello-Ishii systems under mild conditions show nearly 100 % yield. In these systems the PTA anion showed superior activity, so much research has been carried out since the 2000s to exploit its applicable value. For example, in 2001, Xi et al. successfully created a reaction-controlled, phase-transfer, H2O2-mediated system with the insoluble

cetylpyridinium salt of PTA. On reacting with H2O2, the salts peroxo species dissolves in

the organic phase allowing it to react with the olefin to form the epoxide. Once H2O2 is completerly consumed, the active peroxo species returns to its insoluble form for easy recycling [179]. Lambert et al. used the water soluble peroxo species of PTA to construct an efficient water-in-oil microemulsion system for alkene epoxidation. The system, coupled with ultrafiltration, shows potential for continuous epoxide production [180]. Another ingenious heterogeneous catalyst was made by Wang et al. by protonating and anion- exchanging amino-attached cations with PTA. The resulting ionic hybrid nanospheres had the advantages of simple preparation, flexible composition, convenient recovery, and steady reuse [181]. In another system, a PTA salt was trapped inside the pores of a mesoporous SBA-15 support. By grafting octyl groups at the pore entrances, POM leaching was prevented, allowing convenient recycling of the catalyst [182]. These are only a handful of examples from a field that is continuously expanding but they highlight the versatility of PTA in epoxidation reactions. While not explicitly mentioned many other POMs have been designed and tested for this and other oxidation reactions [136]. The focus on PTA is simply the result of the groundbreaking work of Venturello and Ishii, its simplicity and how well it works. The POM used in this thesis is the vanadium based heptammonium tridecavanado-

21 Chapter 3. The Polyoxometalate

manganate, [NH4]7[MnV13O38]. Its synthesis was ﬁrst reported in 1970 by Pope et al. [183] and in 1986 the structure of the polyoxoanion was determined by Nagai et al. using single-crystal XRD in collaboration with Pope [184], see ﬁgure A.4 for a ball and stick representation. The number of publications on this POM is however very low, but it has been investigated as a cathode material for lithium ion batteries [185], as a proton conductor [186], in the synthesis of 2-D lanthanide heteropolyvanadates [187] and in the synthesis of extended organic-inorganic hybrid frameworks [188–190].

22 Chapter 4

The Covalent Organic Framework

4.1 Introduction

Full control of atoms in the three dimensions of space is the ultimate goal of the synthetic chemist. In zero dimensions, organic chemists, utilizing the concepts of total, retro and rational synthesis, have mastered the craft of synthesizing specific molecules with precisely controlled functionality and geometry. Unfortunately these methods fall short in design- ing extended 2-D and 3-D solid-state crystalline structures. Not that extended solid-state structures are not designed and synthesized. Polymer chemists have exploited the methods of organic chemistry to create highly functionalized solid materials, with extraordinary properties, in one dimension a.k.a. chains. Learning from self-organizing systems in nature (e.g. DNA, enzymes, etc.), the supramolecular chemist studies non-covalent interactions to guide these elaborate molecules to assemble into more complex, functional systems. Al- though these are 3-D, such supramolecular systems are difficult to modify without changing or losing their structure. The polymer chemist also creates 3-D structures, e.g. the first synthetic polymer, bakelite, is a 3-D network of phenol and formaldehyde. Unfortu- nately, polymer synthesis is marred by the formation of amorphous, non-porous materials as capillary pressures and high surface energies tend to close small pores by deformation of the framework. To create porous polymers, templating techniques are frequently used [191, 192]. In order to afford ordered, crystalline 2- or 3-D materials, the synthesis must be carried out under thermodynamic conditions allowing for microscopic reversibility in the bond formation between rigid monomers. These bonds have to be strong as well, this would allow for the removal of residual solvent molecules from these organic solids resulting

23 Chapter 4. The Covalent Organic Framework in highly porous crystalline 2-D and 3-D frameworks [193]. This last concept, pioneered and named by Yaghi, is called reticular synthesis. In his 2003 Nature review article, he describes it as follows [194]: “In essence, reticular synthesis can be described as the process of assembling judiciously designed rigid molecular building blocks into predetermined ordered structures (networks), which are held together by strong bonding.” This process forms the basis of his groundbreaking work on metal-organic frameworks (MOFs) and covalent organic frameworks (COFs) [195, 196]. Conceptually this process involves ﬁve steps:

1. A chosen network topology is deconstructed into its fundamental geometric units [197].

2. These units are evaluated on their points of extension and geometry.

3. Geometrically equivalent rigid molecules are deployed as linkers. Often called secondary building units (SBUs), their size determines the materials porosity, density and internal surface area.

4. Forming strong bonds (covalent for COFs/ionic for MOFs) between the linkers pro- duces the chemical equivalent of the chosen network. By balancing the thermodynam- ics (self-correcting behaviour through reversible linkage) and the kinetics (appropriate reaction rate) of the bond-forming reaction, crystalline 2-D and 3-D frameworks can be made [198].

5. (Powder) XRD or electron diﬀraction techniques are used to evaluate the success of the synthesis by comparing the experimental pattern to the calculated pattern for the expected structure.

As an adjective reticular thus means “having the form of a periodic network” and just as crystals can be isostructural, MOFs and COFs can be isoreticular. This means that they are based on the same network, but that the length of the linkers differs resulting in larger or smaller pores, different density, etc. Varying MOFs and COFs structure can be easily achieved by changing the linker connectivity and size, and types of the linkages. COFs can be classified in terms of the linkages used to build up the network, an overview is given in figure A.5.

24 Chapter 4. The Covalent Organic Framework

By introducing functionalities in the COF backbone, the material properties can be tailored towards a speciﬁc application. It is the stability of the crystalline COF structure that allows functionalization without structural loss. There are two ways to achieve this:

Presynthetically In essence, functionalize the molecular building blocks. This method allows tailoring of the pore diameter and environment in COFs, while maintaining structural homogeneity and pore size uniformity [199].

Postsynthetically This entails modifying the preformed framework itself [200]. In many cases these modiﬁcations are based on click chemistries [201]- such as Cu(I)-catalysed alkyne-azide, Diels-Alder cycloadditions [202] or thiol-ene reactions [203]- these reactions are preferred as they give high reaction yields with well-known stereoselectivity at high rates.

The compatibility between the desired functional group and the COF-forming reaction determines the functionalization approach. Some functional groups may interfere in the linkage formation, preventing the COF-formation. Fortunately, the open, accessible pores of the crystalline organic backbone allow for a precise functionalization of the structures’ interior space when presynthetic modiﬁcation is not possible.

4.2 The Covalent Triazine Framework

Of all possible linkages, one of the most stable is based on the trimerization of nitrile groups to the aromatic triazine ring. This type of linkage was first reported in 2008 by Kuhn, Antonietti, and Thomas [204]. The trimerization of 1,4-dicyanobenzene (DCB) was ◦ carried out at 400 C in molten zinc chloride (ZnCl2) using a one-to-one ratio ZnCl2:DCB, ◦ see figure A.6. ZnCl2, with a melting point of 290 C [205], acts as both the linker solvent and the trimerization catalyst through strong Lewis acid-base interaction between the electron deficient Zn2+ ion and the free electron pair of the nitrile. Under these ionothermal conditions, the trimerization reaction seems to be sufficiently reversible to obtain crystalline porous polytriazines with a specific BET surface area of 791 m2 g=1 and a total pore volume of 0.40 cm3 g=1, values comparable to those of COF-1. These new COF materials were

denoted as covalent triazine frameworks a.k.a. CTFs. However, increasing the ZnCl2:DCB- ratio yields highly porous, but amorphous materials [206]. Overall, these material show

25 Chapter 4. The Covalent Organic Framework even higher speciﬁc surface areas, probably as a result of the lower overall density of amorphous compared to crystalline materials. While the ionothermal synthesis is harsh, the resulting CTF materials have high thermal, chemical, and mechanical stability, making CTFs very promising for potential applications in gas storage or as sensors, sorption materials, or catalyst supports [207]. Unfor- tunately this also results in the inevitable loss of linker functionality through uncontrolled C–C coupling. A thorough XPS study of the nitrogen species present in CTF materials, performed by Osadchii et al. [208], resulted in some interesting conclusions on this topic:

1. The decomposition of CTFs increases the speciﬁc surface area and porosity of the material (already noted by Kuhn et al.), but decreases the relative amount of accessible nitrogen species.

2. Increasing the reaction time, temperature, and use of oxidizing compounds as trimerization catalysts promote the decomposition reactions.

3. The use of monomers that form more stable Lewis acid-base complexes with the molten salt favours the CTF formation and reduces the rate of framework decomposition.

4.2.1 The Cationic CTF

The previously mentioned loss of functionality in the ionothermally synthesized CTFs makes postsynthetical modification difficult for obvious reasons. Chemical diversity in CTFs is thus mainly introduced via presynthetical linker differentiation. One example of these are the cationic CTFs (cCTFs) build up of stable cationic linkers. An easy way to introduce a cationic moiety in a dinitrile-linker is through N-heterocyclic aromatics, e.g. viologen [118], phenanthridinium [2, 209], and (benz)imidazolium [210–212] groups, see figure 4.1. These aromatic moieties are highly stable, easily introduced, and readily charged through quaternization. The pyridinium/imidazolium nitrogen species also enhance CO2 adsorption through dipole-quadruple interactions. The cCTF used in this research is based on the 2017 work of Park et al. [211]. Ionothermal trimerization of the imidazolium-based linker resulted in a N-rich, microporous, and thermally stable CTF. They argued that the introduction of charged N-species would increase the CO2 affinity, resulting in a material with an uptake capacity up to

26 Chapter 4. The Covalent Organic Framework

=1 2.77 mmol g at standard conditions. A 1:5 ratio of the monomer and ZnCl2 was used because it was determined to be the optimal ratio for the highest porosity, which was considered to be a signiﬁcant contributing factor for high CO2 uptake. This approach is not without merit [213], Huang et al. compared two imine COFs, one was charged using benzimidazolium-modiﬁed linkers, while the other one was a neutral analogue without the

imidazolium group. The ionic interface of the former resulted in a three-fold CO2 uptake capacity compared to the latter [210].

Figure 4.1: The a) viologen, b) phenanthridinium, c) imidazolium, and d) benzimidazolium structures. The R-groups can be easily modiﬁed to carry linkable groups, e.g. nitriles, amines, etc. For example, when for b) R = amine, R’ = ethyl, and R” = phenyl, the structure is an ethidium halide, used for the synthesis of an imine cCOF by Li et al. [209].

The affinity of CO2 for the charged N-species inspired Buyukcadir et al. [118] and Liu et al. [212] to employ these cCTFs as catalyst for the cycloaddition of CO2 to epoxides. The former system is based on a viologen-modified nitrile linker, while the latter is the cCTF first reported by Park et al. the previous year. The positively charged backbones of

the porous cCTFs could effectively adsorb CO2 increasing its availability and speed up the reaction. Furthermore, the positively charged moieties and the spatially confined nucle- ophilic chloride anions (resulting from the ZnCl2-based ionothermal synthesis and typical HCl-reflux washing steps) work in tandem on the substrate, leading to synergistically enhanced catalysis for the cycloaddition reaction. The chloride anion attacks the epoxide

ring forming a deprotonated chlorohydrin, that upon CO2 addition closes by expelling the anion to form the cyclic carbonate.

27 Chapter 4. The Covalent Organic Framework

The POM@cCTF

Cationic CTFs can also be used as a platform for anionic exchange, e.g. to trap anionic pollutants like charged organic dyes by Huang et al. [210], or to tune porosity and proton conduction by Ma et al. [2]. In the latter work, a β-ketoenamine linked cCOF was synthesized with ethidium bromide-based linkers (EB-COF:Br). Through anionic exchange with other halides, the porosity could be tuned, with decreasing pore size and speciﬁc surface area according to the increasing halide ion size; F–

Figure 4.2: Schematic representation of PTA doping of the ethidium bromide-based β- ketoenamine cCOF [2].

28 Chapter 5

Electrospinning

5.1 The Process of Electrospinning

Electrospinning (ES) has a longer history than one might assume. In 1887, Boys described what he called an “old, but now apparently little-known experiment of electrical spinning” in which, “If a small dish be insulated and connected with an electrical machine and filled with [. . . ] viscous material, the contents will, if they reach one edge of the dish, at once be shot out in the most extraordinary way in [. . . ] threads of extreme tenuity, travelling at a high speed along “lines of force.””, the lines of the electric field [214]. Not soon after, the first patents based on this phenomenon were granted [215–218] and by WWII the technique was commercially exploited to fabricate smoke filters for gas masks, known as “Petryanov filters” [219]. In the 1960s, Taylor reported the theoretical underpinning of the ES process and in the early 1990s, Reneker, Rutledge, and others demonstrated that many polymers could be electrospun in nanofibres. This sparked an increased interest in the technique and the number of publications about ES has risen exponentially since then [220]. Conceptually, ES is fairly simple, it requires only four components to fulfill the process: a high voltage supplier, a small diameter needle, a metal collecting screen, and a polymer solution pump, see figure 5.1 [221]. In the process, a high voltage is applied, 1 to 30 kV, between the needle and the collector, about 10 to 30 cm apart. The polarization of the electric field is typically such that it induces positive charges in the polymer solution. As the polymer solution is pushed through the needle, positive charges accumulate on its surface and Coulomb repulsion increases. A critical point is reached when the repulsive Coulomb forces overpower the contracting surface tension. A Taylor cone is formed at the tip of the

29 Chapter 5. Electrospinning needle, ejecting a charged jet of polymer solution towards the collector plate. This plate can either be oppositely charged or grounded. As the polymer jet accelerates towards the collector, a whipping process elongates the drying fibres, until they are deposited. This process, when well executed, leads to the formation of uniform fibres with nanoscale diameters [222]. Thus at first glance, ES gives the impression of being a very simple and, therefore, easily controlled technique for the production of fibres with dimensions down to the nanometre range. However, the shapes and dimensions of the fibres formed depend on a large set of parameters, an overview is given in table A.2.

Figure 5.1: Scheme of a basic laboratory electrospinning set-up [223].

The extremely small diameters of the fibres result in several amazing characteristics such as very large surface area-to-volume ratio, flexibility in surface functionalities, tunable porosity, and superior mechanical performance (e.g. stiffness and tensile strength) compared to other known forms of the material. Hybrid fibres composed of metals and ceramics are even attainable, as are nanofibres with a solid or liquid core and a solid shell, resulting in materials with a broad scope of applications, in fields as diverse as optoelectronics, sensor technology, catalysis, filtration, and medicine [222].

5.2 The Nanoﬁbre Composite

Despite the interesting catalytic possibilities of a POM@cCTF system, their use can be a nuisance. First oﬀ, as a powder, it must be dispersed into the solution, requiring ﬁltration to recuperate the catalyst after reaction. This limits their use to batch reactions as the

30 Chapter 5. Electrospinning powder must be contained, eventually resulting loss of catalyst. To mitigate this, the powder could be packed into pellets, an act typically done in industry, allowing the catalyst to be used in continuous flow reactors. However, the high pressures used during packeting could alter the structure of the material and induce porosity loss. This loss in porosity also increases the pressure drop over the reactor. An alternative approach could be to create a POM@cCTF/nanofibre composite by ES a POM@cCTF/polymer solution. The idea is that the 3-D networks of electrospun nanofibres are beneficial for the dispersion of COFs particles compared to pellets. This not only prevents porosity loss of COFs particles and the problem of high pressure drops, but also increase the contact surface area between COFs and substrate [224]. To the best of my knowledge, only two COF@nanofibre composites have been reported, both this year. Most recently, Wu et al. exfoliated a 2-D chiral COF based on imine linkers resulting in nanosheets of 200 to 300 nm diameter. These nanosheets, dispersed in DMF, were mixed with polyvinylidene fluoride (PVDF, structure see figure A.7) with loadings ranging from 5 to 10 wt.%. The solutions were successfully electrospun and tested for enantioselective terpene sensing through fluorescence quenching [225]. In the second example, Yan et al. synthesized an amide COF hydrothermally, resulting in nanoparticles smaller than 100 nm. These particles were successfully dispersed in a polyacrylonitrile (PAN, structure see figure A.7) solution and electrospun on a rotating drum and applied to pipette tip solid phase extraction of sulfonamides in meat samples [226].

5.3 Electrospinning of Polyamide-4.6

The oxidative carboxylation of styrene requires an oxidant, and as most research is done using TBHP, this oxidant will be chosen for easier performance comparison. TBHP is typically supplied as a 70 v.% solution in decane and in many cases a solvent speeds up a catalytic reaction. This means that the catalytic system needs to be able to withstand these compounds. POMs and CTFs are very stable in nearly all organic solvents, polymers on the other hand are not. As will be elaborated on in the experimental part, polyamide-4.6 (PA-4.6, structure see 5.2) was chosen as the polymer matrix. PA-4.6 was successfully electrospun in 1999 by Bergshoef et al. from a 10 wt.% solution in formic acid [227]. Of the commercially available polyamides, PA-4.6 is believed to have the highest mechanical performance and stability. This is credited to the high number of

31 Chapter 5. Electrospinning

Figure 5.2: The structure of polyamide-4.6.

amide units per given chain length enhancing the formation of hydrogen bonds. Together with the more symmetrical chain structure this gives the polymer a higher percentage of crystallinity in comparison to e.g. PA-6 and PA-6.6, resulting in a melting temperature of 295 ◦C. PA-4.6 also shows excellent creep resistance, toughness, and wear characteristics, making it able to withstand high loads and stresses at high temperatures and exposure to aggressive environments [228]. The high amide content also increases its water affinity resulting in a plasticizing effect during ES which thins the polymer jet, resulting in finer fibres with increasing RH. The absorbed water molecules not only act as plasticizer giving finer fibres, but also disturb the formation of stable crystals, causing a higher fraction of less stable crystals at higher RH [229]. This effect is counterbalanced by the high degree of fibre stretching during ES which induces crystallinity, resulting in nanofibres that have a higher thermal stability compared to the bulk material in certain cases [230]. Clearly many parameters influence the mechanical properties of nanofibre PA-4.6, luckily, the conditions for the oxidative carboxylation of styrene are very mild. Essentially, only a solvent must be chosen that does not dissolve PA-4.6. An in depth parameter study by De Schoenmaker et al. from 2012, serves as a starting point for the choice of practical conditions required to electrospin PA-4.6 [231].

32 Chapter 6

Materials & Methods

For an overview of the used chemicals and equipment, see chapter B.1.

6.1 Synthesis of the Cationic Dinitrile Linker

The cCTF used in this research, from the work of Park et al. [211], is synthesized from 1,3-bis(5-cyanopyridyl)-imidazolium bromide (bpim). This linker is prepared according to the recipe of Chen et al. [232]. 2-Bromo-5-cyanopyridine (2.176 g, 11.9 mmol, 2 eq.) and 1- methyl-1H -imidazole (472 µL, 5.9 mmol, 1 eq.) were charged into a Pyrex ampoule, which was degassed, flame sealed and heated up to 190 ◦C at 1 ◦C min=1 in a muffle furnace and retained at that temperature for 18 h. After the reaction, the ampoule was cooled to room temperature (RT) and opened with a glass cutter. The collected crude black powder was dissolved in ca. 100 mL hot methanol (MeOH) to extract the pure compound. The linker dissolves poorly in MeOH and remains as a precipitate, which was filtered of and kept aside. The filtrate was allowed to cool to RT and put on the rotavapor until ca. 20 mL MeOH remained, the formed crystals were filtered off. Important to note: if too much MeOH was evaporated, unreacted starting materials can precipitate. The obtained light brownish powders were added together, washed with chloroform, and dried for 24 h under vacuum at 120 ◦C. A proposed reaction pathway is given in figure 6.1.

33 Chapter 6. Materials & Methods

Figure 6.1: A schematic of the proposed reaction pathway of the linker synthesis based on 1H NMR spectroscopy of the quenched reaction by Chen et al. [232].

6.2 Synthesis of the bpim-based cCTF

The synthesis method of the cCTF was adapted from the work by Park et al. [211]. Dried bpim (0.5 g, 1.42 mmol) and ZnCl2 (0.95 g, 6.97 mmol) were loaded in a Pyrex ampoule and degassed under vacuum for 5 h to remove residual water. The ZnCl2 was dried beforehand under vacuum at 120 ◦C overnight. The ampoule was flame sealed and placed in a muffle furnace and heated at a rate of 1 ◦C min=1 to 400, 500 or 600 ◦C and kept at that temperature for 48 h. After 48 h, the ampoule was left to cool down to RT and opened with a glass cutter. The collected crude black powder was ground with a mortar and pestle in a minimal amount of water, and added to a 250 mL round-bottom flask with 200 mL 1M HCl solution and a magnetic stirring bar. The mixture was refluxed at 110 ◦C for

24 h under vigorous stirring to remove the ZnCl2. Subsequently, the mixture was filtered, and the powder was washed with water until the filtrate was neutral (ca. 3 × 100 mL water, then with acetone (3 × 100 mL), and left to dry in air for an hour. The powder was subsequently brought into a 250 mL round-bottom flask with 200 mL THF and stirred vigorously for 24 h. After filtration, the black cCTF powder was activated under vacuum at 150 ◦C overnight. A schematic representation of the reaction can be seen in figure 6.2.

34 Chapter 6. Materials & Methods

Figure 6.2: Idealized representation of the cCTF formation.

6.3 Loading of the Vanadium-based POM

The loading of the anionic POM in the cationic CTF is very straightforward, following the procedure from Ma et al. [233]. The VPOM was dissolved in 5 mL ACN in a 50 mL round-bottom flask under magnetic stirring. A certain amount of cCTF was added and the mixture was left to vigorously stir for 48 h at RT. Afterwards, the mixture was filtered, washed with deionized water, and activated under vacuum at 120 ◦C for 24 h. A schematic representation of the process is given in figure 6.3. A series of VPOM@cCTFs with different loadings of the VPOM were prepared:

11.2 wt.%: 12.62 mg VPOM in 100 mg cCTF.

14.3 wt.%: 16.67 mg VPOM in 100 mg cCTF.

20 wt.%: 25 mg VPOM in 100 mg cCTF.

The initial catalytic tests were performed with the 14.3 wt.% VPOM@cCTF material and later with the 11.2 wt.% VPOM@cCTF. At 11.2 wt.%, 10 mg of the catalyst should contain 1.12 mg or 7.72 × 10=4 mmol VPOM given the VPOM structure formula of

35 Chapter 6. Materials & Methods

=1 [NH4]7[MnV13O38] (M = 1451 g mol ). This corresponds to 0.01 mmol vanadium present in 10 mg 11.2 wt.% VPOM@cCTF provided the loading proceeds with a VPOM retention of 100 %. This means that the reaction would be catalysed using 1 mol % catalyst, assuming vanadium in the VPOM is the active catalytic site, a value that is preferably as low as possible.

Figure 6.3: Schematic representation of the VPOM loading.

6.3.1 Heptammonium Tridecavanadomanganate

Originally the POM phosphotungstic acid (PTA) was envisaged as the epoxidation catalyst given its excellent performance in the Ishii-Venturello systems. Unfortunately, BA was mainly formed during the catalytic tests. For this reason, a vanadium based POM was utilized, as V based catalysts are known to possess good oxidation reactivity [234, 235],

namely heptammonium tridecavanadomanganate, [NH4]7[MnV13O38]. The initial catalytic results using this POM were promising. The BA formation decreased while the selectivity for SO increased. For this reason, theis POM based epoxidation catalyst was used further on during this master thesis.

36 Chapter 6. Materials & Methods

6.4 Catalytic Set-Up

The oxidative carboxylation of styrene was performed using Schlenk tubes. They were loaded with styrene (1 mmol, 112 µL), chlorobenzene as internal standard (1 mmol, 102 µL), ACN as solvent (3 mL), TBAB as co-catalyst (10 mol % to styrene, 0.1 mmol, 32.2 mg), TBHP (1.5 mmol, 300 µL of a 5M solution in decane), the VPOM@cCTF (10 mg), and are ◦ then ﬁnally purged with CO2 (3 bar). The reaction was proceeded at 75 C for 72 h. From the preliminary tests, it was concluded that the stepwise addition of TBHP to the reaction mixture increased the SC yield. For this reason, 30 µL TBHP solution was added at the start of the reaction in the morning. In the afternoon an additional 70 µL was added. This was repeated twice daily until a total amount of 300 µL TBHP solution was added. Samples for GC-MS analysis were taken at the beginning and at the end of the reaction to determine styrene conversion and product formation.

Testing the base catalytic properties of the cCTF material in the cycloaddition of CO2 to styrene oxide (SO, 2.43 mmol, 278 µL) was done using the reaction conditions above. The only diﬀerence was that TBHP was not added and that the unloaded cCTF material was used. Samples for GC-MS analysis were taken at the beginning and at the end of the reaction to determine styrene conversion and product formation.

6.5 Electrospinning Set-Up

The work of De Schoenmaker et al. [231] serves the starting point for determination of the conditions for the electrospinning of PA-4.6. In this work PA-4.6 was successfully electrospun from a 15 wt.% solution consisting of 30 v.% of the non-solvent acetic acid and 70 v.% of the solvent formic acid. A WPI SP100iZ single-syringe infusion pump was utilized to pump the PA-4.6 solution out of an 18-gauge syringe in a horizontal position at a flux of 4 mL h=1. The needle and metal collecting plate covered with Al-foil were placed 10 cm apart and a voltage of 15 kV was applied using a Glassman HV power supply. SEM analysis of the electrospun fibres was used to analyse the morphology. The open source plug-in DiameterJ for the public domain image processing software ImageJ was used to determine the fibre diameters [236].

Chapter 7

Results & Discussion

7.1 Analysis of the Cationic Dinitrile Linker

To confirm the success of the synthesis of the cationic dinitrile linker, the chemical shifts in the 1H and 13C NMR spectra were compared to the reported values, see tables 7.1 & 7.2. The spectra were acquired at 300 MHz using non-deuterated DMSO as the solvent. The reported shifts were acquired at 400 MHz using deuterated chloroform as the solvent [211]. The close resemblance between the values confirms the success of the synthesis and washing procedure. The peak assignment to the linker molecule is presented in figure 7.1 and the NMR spectra are given in figures B.1 and B.2.

Figure 7.1: Structure of bpim with the atoms corresponding to the chemical shifts listed in tables 7.1 and 7.2.

39 Chapter 7. Results & Discussion

Table 7.1: Comparison of the experimental and reported 1H chemical shifts of the cationic dinitrile linker [211].

1H NMR Shift δ (ppm) Multiplet & Integration

Reported Measured Reported Measured Corresponding atoms

4 4 11.07 11.06 t, J HH = 1.66 Hz, 1H t, J HH = 1.61 Hz, 1H NCHN 5 5 9.24 9.25 dd, J HH = 0.77 Hz, dd, J HH = 0.75 Hz, CH-6py 4 4 J HH = 2.23 Hz, 2H J HH = 2.07 Hz, 1.93H 4 4 8.86 8.86 dd, J HH = 2.23 Hz, dd, J HH = 2.16 Hz, CH-4py 3 3 J HH = 8.70 Hz, 2H J HH = 8.55 Hz, 1.83H 4 4 8.84 8.84 d, J HH = 1.66 Hz, 2H d, J HH = 1.50 Hz, 2.28H NCHCHN 5 5 8.50 8.50 dd, J HH = 0.77 Hz, dd, J HH = 0.75 Hz, CH-3py 3 3 J HH = 8.70 Hz, 2H J HH = 8.67 Hz, 2.12H

Table 7.2: Comparison of the experimental and reported 13C chemical shifts of the cationic dinitrile linker[211].

13C NMR Shift δ (ppm) Multiplet

Reported Measured Reported Measured Corresponding atoms

153.3 152.8 s s CH-6py 148.5 148.0 s s CH-3py 145.1 144.6 s s CH-4py 136.3 135.8 s s NCHN 121.0 120.6 s s NCHCHN 116.5 116.0 s s C-5py 115.7 115.3 s s C-2py 111.0 110.5 s s CN

7.2 Analysis of the bpim-based cCTF

Multiple techniques were used to characterize the cCTF materials. Firstly, N2-sorption was employed to determine the porosity and specific surface area of the cCTFs synthesized at the different temperatures. An overlay of the resulting isotherms is presented in figure

40 Chapter 7. Results & Discussion

7.2. All three materials show a type I isotherm, also known as a Langmuir isotherm, without hysteresis. This is typical for CTFs and is the result of the microporous nature of these materials. Analysis of the isotherms using the BELMasterTM software gave the data summed up in table 7.3.

Table 7.3: Porosity analysis of the cCTF materials synthesized at diﬀerent temperatures utilizing the Langmuir and BET theory.

Temperature Speciﬁc surface area Pore volume Pore diameter ◦ 2 =1 2 =1 3 =1 T( C) As,Lang (m g ) As,BET (m g ) Vp (cm g ) dp (nm)

400 631 298 0.14 (p/p0 = 2.30 0.995)

500 1833 1067 0.63 (p/p0 = 2.36 0.990)

600 1975 1150 0.66 (p/p0 = 2.28 0.983)

The overall trend is that with increasing synthesis temperature, the specific surface area and pore volume of the material increase, while the average pore diameter remains fairly constant. This is a trend typical for the ionothermal synthesis method of CTFs and a result of the graphitization process that results in a gradual widening of the pores, accompanied by a shift of the pore size distribution toward the mesopore range without affecting the micropores [237]. The main assumption of the Langmuir theory is that the adsorbed gas forms a mono- layer exclusively, in reality this is not the case and multilayer formation occurs to a certain degree. This means that the Langmuir specific surface area is an overestimation of reality. The Brunauer-Emmett-Teller (BET theory) theory corrects for this multilayer formation phenomenon, but in the case of type I isotherms it is an overcorrection. This means that the BET specific surface area is an underestimation. The real value lays somewhere in- between. Another strength of the BET theory in comparison to the Langmuir theory is that it allows for the calculation of the average pore diameter, which is ca. 2.31 nm. These pores are large enough for a POM cluster, typically 1 to 2 nm in diameter, to enter [144]. The cCTF synthesized at 500 ◦C was chosen for further testing because of its high specific surface area (1833 m2 g=1) and pore volume (0.63 cm3 g=1), all further references to cCTF thus refer to this material.

41 Chapter 7. Results & Discussion

Figure 7.2: Overlay of the isotherms of the cCTFs synthesized at the diﬀerent temperatures.

DRIFTS is an elegant technique to follow the progress of the trimerization reaction. Both the nitrile stretching vibration at ca. 2230 cm=1 and the triazine in-plane vibrations at 1410 cm=1 and 1550 cm=1 are strong and should be easily detected. Looking at the spectra overlay presented in figure 7.3, we can very clearly see the nitrile stretching vibration at 2235 cm=1 present in the bpim spectrum, but not in the cCTF spectrum, which indicates the success of the trimerization reaction. The expected in-plane triazine vibrations on the other hand merged with the vibrations of the pyridine and imidazolium groups, resulting in broad peaks at 1650 cm=1 (C–C or C–N), 1365 cm=1 (C–N or C–C), and 800 cm=1 (heterocyclic aromatics). The broad peaks are the result of the harsh synthesis conditions resulted in a high degree of graphitization of the framework and the formation of complex aromatic nitrogen-carbon moieties [238]. CHNS elemental analysis supports this finding, it reported 3.14 wt.% hydrogen, 76.17 wt.% carbon and 20.79 wt.% nitrogen for the cCTF material. As the trimerization reaction proceeds with 100 % atom efficiency, the theoretical C/N-ratio of the framework is 2.14, the measured ratio on the other hand is 3.66. This higher C/N-ratio is another indication of the loss of nitrogen throughout the framework because of graphitization, which is the result of the harsh ionothermal synthesis method.

42 Chapter 7. Results & Discussion

Figure 7.3: Overlay of the DRIFTS spectra of bpim and the cCTF synthesised at 500 ◦C.

TGA analysis determined that the material is stable up to 400 ◦C under an air atmosphere, see figure 7.4. The ionothermal synthesis method typically results in amorphous materials due to graphitization at the elevated temperatures used in this method. This is reflected in the PXRD diffractogram presented in figure 7.6. The diffractogram of the cCTF shows no sharp peaks, just a very broad one at around 25.5°, corresponding most likely to the (001) reflection. The lattice distance corresponding to the (001) reflection in the diffraction pattern is probably the layer distance between the sheets in an idealized cCTF structure. A similar broad peak is seen at 24° for CTF-1 materials corresponding to the (001) reflection of the distance between the layers [204].

7.3 Analysis of the VPOM@cCTF

Analysis to determine the success of the VPOM retention was performed on samples loaded with 11.2, 14.3 and 20 wt.% VPOM. To determine the retention quantitatively, ICP-AES was employed as the technique of choice as it provides very precise data. Analysis of the 14.3 wt.% VPOM@cCTF determined that the sample contained 4.03 wt.% V and 0.111 wt.% Mn. Calculating backwards from the wt.% of V gives a retained VPOM content of 8.83 wt.%. The 20 wt.% sample contained 6.37 wt.% V and 0.181 wt.% Mn corresponding

43 Chapter 7. Results & Discussion

Figure 7.4: TGA of the cCTF material, indicating its thermal stability under air up to 400 ◦C, measured at a heating rate of 10 ◦C min=1 from 27 to 1000 ◦C.

13.96 wt.% VPOM in the material. Analysis of the 11.2 wt.% VPOM@cCTF reported that the sample contained 4.52 wt.% V and 0.089 wt.% Mn. This corresponds to a VPOM loading of 9.9 wt.%. The loading can also be qualitatively assessed using the methods also used for the analysis of the linker and cCTF materials. For example, N2-sorption of the 11.2 wt.% loaded cCTF resulted in an expected type I isotherm, but the maximum adsorbed N2 volume was lowered, see ﬁgure 7.5. Results of the analysis of the isotherm using Langmuir and BET theory are given in table 7.4. The lower speciﬁc surface area and pore volume indicate that a certain material was deposited on/inside the cCTF.

Table 7.4: Comparison of the porosity of unloaded and VPOM loaded cCTF utilizing Langmuir and BET theory.

Temperature Speciﬁc surface area Pore volume Pore diameter ◦ 2 =1 2 =1 3 =1 T( C) As,Lang (m g ) As,BET (m g ) Vp (cm g ) dp (nm)

cCTF 1833 1067 0.63 (p/p0 = 2.36 0.990)

11.2 wt.% 1542 909 0.53 (p/p0 = 2.32 VPOM@cCTF 0.990)

44 Chapter 7. Results & Discussion

Figure 7.5: Overlay of the isotherms of cCTF and 11.2 wt.% VPOM@cCTF.

Evidence that the retained materials is indeed the VPOM was obtained using PXRD, DRIFTS, and Raman spectroscopy. In the diffractogram of 20 wt.% VPOM@cCTF, the peaks of corresponding to the VPOM are superimposed on the cCTF spectra, clearly visible for the reflections at 28.25°, 32.0°, and 45.7°, see figure 7.6. The diffractogram of 11.2 wt.% VPOM@cCTF showed no discernable peaks, so is therefore omitted. A comparison between the predicted diffractogram of the VPOM with potassium as counter ion and the measured VPOM diffractogram is given in figure B.3. The difference in peak positions are most likely the result of the slight differences in crystal structures due to the size difference between the potassium and ammonium cation sizes, and the uncertainty inherent to the PXRD instrument and the prediction software built in the CCDC Mercury software suite. The same applies for the DRIFTS and Raman spectra, only the strongest VPOM vibrations are visible in the VPOM@cCTF spectra. The strongest IR-active vibrations are visible in the VPOM@cCTF spectra for both the loading of 11.2 wt.% and 20 wt.%, see figures B.5 and B.6. The Raman vibrations on the other hand are naturally much lower in intensity, only the strongest vibrations for the 20 wt.% loading are visible in the spectra of the VPOM@cCTF, see figure B.4. The 11.2 wt.% VPOM@cCTF spectra is omitted as it does not show VPOM Raman vibrations.

45 Chapter 7. Results & Discussion

Figure 7.6: PXRD diﬀractograms of the cCTF, VPOM and 20 wt.% VPOM@cCTF.

Figure 7.7: BF-STEM-EDX overlay of 11.2 wt.% VPOM@cCTF.

46 Chapter 7. Results & Discussion

BF-STEM analysis of the 11.2 wt.% VPOM@cCTF with parallel EDX-mapping measurements allowed mapping of the N, V and Mn present in the cCTF powder to assess the distribution of these elements throughout the framework. An overlay of the mapping of the diﬀerent elements is given in ﬁgure 7.7 above. The mapping shows clearly that the distribution of N and the VPOM is uniform throughout the VPOM@cCTF grains.

7.4 Catalytic results

To assess the inﬂuence of each part of the catalytic system, a series of tests was performed in which each part of the catalytic system (VPOM, cCTF, TBAB, and VPOM@cCTF) and each possible combination of the parts were tested. The overall catalytic system was tested in triplet (entry 9) to assess the reproducibility of the results (given at the 95 % conﬁdence level). An overview of the results is given in table 7.6. Some interesting properties about the catalytic system can be concluded from this series:

1. The presence of the co-catalyst is paramount to the formation of SC, when it is not present, SC is not or barely formed as can be seen from entries 1, 2, 4, 5, and 7. One could assume that any chloride ions present from the washing process during cCTF synthesis could activate the epoxide resulting in the formation of SC. Unfortunately as can be seen from entries 2, 4, 5, and 7, this is not the case, any chloride (or even bromide) ions present in the framework are not active. Our hypothesis is that a synergetic interaction between the electron-deﬁcient ammonium N and the electron-rich epoxide oxygen activates the sterically free C–O bond resulting in eﬃcient epoxide ring opening and subsequent cyclic carbonate formation.

2. The homogeneous addition of the VPOM has the eﬀect of lowering BA formation, as can be observed by comparing the blank reaction to entry 4. The heterogeneous addition of VPOM through VPOM@cCTF has the same eﬀect as can be seen by comparing the blank to entries 9 and 10. Additionally, the presence of VPOM in the framework resulted in a SC yield increase of ca. 7 %, as can be seen by comparing entry 6 to entries 9 and 10.

47 Chapter 7. Results & Discussion

3. The addition of cCTF to a reaction mixture with TBAB (entry 7) did not result in a signiﬁcant increase in SC yield compared to the reaction with only TBAB (entry 3). We hypothesize that as the VPOM is not present in the framework to catalyse the epoxidation, there is no local concentration increase of SO, meaning that the

increased CO2 presence through interaction with the cCTF framework cannot be optimally utilized. However, the presence of VPOM in the framework (entries 9 and 10) increases the SC yield with ca. 7 %, as mentioned in the point two.

4. When comparing the SC yield of entry 8, to the yield of entry 3, we notice something peculiar. Upon addition of homogeneous VPOM, the SC yield drops with ca. 7 %, while the BA formation increases. One could assume that the increased oxidation

rate and the low concentration of CO2 enhances BA formation at the expense of SC compared to e.g. entry 9.

ACN is chosen as the solvent for two reasons, (i) based on early tests it was observed that adding a solvent speeds up the reaction considerably, and (ii) PA-4.6 is stable in ACN. Determining the turnover number (TON) of the catalyst is at this stage of the research not possible. To calculate TON, the number of moles of catalytic sites has to be known. In tandem catalysis systems this is difficult to define and depends on which reaction step is the rate limiting and the specific active sites of the catalyst. To test the base catalytic properties of the cCTF material as such in the cycloaddition reaction of CO2 to styrene oxide, four experiments were performed. A blank reaction, on with cCTF, one with TBAB, and one with both. The results of the experiments are given in table 7.5 below. From GC-MS analysis of the SO starting material was noticed that the bottle was severely contaminated with phenylacetaldehyde (PHAA), most likely formed by the isomerisation of the SO. The first two reactions did not result in the formation of any SC, as can be expected. From entry 3, it is clear that TBAB is required to initiate the cycloaddition through the activation of the epoxide ring. Surprisingly enough, both the SO and the PHAA seem to be converted in SC at about the same rate. This is interesting, because this seems to imply that an aldehyde group could be activated to produce a cyclic carbonate. Repeating this reaction with the addition of cCTF, results in a significant increase in SC yield, confirming our hypothesis that the local increased CO2 presence in the cCTF through interactions with the framework positively influence the reaction.

48 Chapter 7. Results & Discussion

Table 7.5: Overview of the results of the cycloaddition of CO2 to styrene oxide (SO = styrene oxide, PHAA = phenylacetaldehyde, and SC = styrene carbonate).

Entry cCTF TBAB SO con- PHAA SC (%) Yield (%) (10 mg) (0.1 mmol) version conver- (%) sion (%)

1 (blank) 0 0 0 0 2 0 0 0 0 3 10 10 100 10 4 36 39 100 37.5

49 Table 7.6: Overview of the results of the catalytic oxidative carboxylation of styrene (ST = styrene, SC = styrene carbonate, SO = styrene oxide, BA = benzaldehyde, and PHAA = phenylacetaldehyde).

Entry cCTF TBAB VPOM VPOM- ST con- SC (%) SO (%) BA (%) PHAA Other % Yield (%) (10 mg) (0.1 mmol) (0.01 mmol) @cCTF version (%) (10 mg) (%)

1 (blank) 96 2 28 43 25 2 1.92 2 85 0 37 30 33 0 0 3 97 32 15 36 16 1 31 4 84 0 36 25 39 0 0 5 (11.2 74 0 22 41 30 8 0 wt.%) 6 91 36 9 50 6 0 33 7 90 1 31 35 32 1 0.9 8 97 25 5 57 9 3 24.2 9 (11.2 90 ± 3.7 45 ± 3.2 7 ± 1.3 31 ± 0.7 11 ± 1.0 6 ± 0.5 40.5 ± 2.74 wt.%) 10 (20 88 47 7 34 10 3 41.3 wt.%) Chapter 7. Results & Discussion

7.5 Results of the Electrospinning of PA-4.6

Based on the work by a previous thesis student [4, 5], an attempt was made to embed the VPOM@cCTF powder in a nanofibre mesh for easy catalyst recuperation. The first choice of polymers were poly--caprolactone (PCL) and polylactic acid (PLA), both are well studied in the field of ES and easy to spin. Unfortunately, PCL is not resistant against TBHP, and both dissolve in the reaction mixture. In the search for a more stable polymer, polyethylene (PE) seemed like an excellent choice. High density PE is known for its high chemical and solvent stability. However, dissolving PE poses a challenge. PE is typically electrospun from a heated solution in p-xylene [239, 240]. As equipment that would allow electrospinning from such a solution is not available, another polymer had to be found. Silica nanofibres were briefly considered, as they are easily electrospun from a tetraethyl orthosilicate solution. Unfortunately, they are very brittle and therefore would most likely not be stable under the reaction conditions purely due to mechanical break-down [241, 242]. Eventually PA-4.6, a polymer known for its thermal stability and excellent mechanical properties, was chosen. At this point in the research, different solvents were tested for their influence on the reaction. Therefore, the stability of PA-4.6 was tested in these solvents under reaction conditions. As a reference, the stability of PA-4.6 was also tested under solvent free reaction conditions. The solvents in question were ACN and 1,2-dichloroethane (1,2-DCE). SEM images of nanofibre PA-4.6 before and after the test are given in figure 7.8. From these images it clear that the nanofibres dissolved partly under solvent free (SF) conditions resulting in an elastic nanofibre clump. The opposite happened in 1,2-DCE, after drying, the nanofibre mat became very brittle and resulting in broken fibres as can be seen on the SEM image. In ACN however, the fibres kept their morphology and did not dissolve or clump together. The apparent decrease in pore size is most likely the result of the physical compressing of the nanofibre mesh under the influence of the magnetic stirrer. Diameter analysis of the pristine PA-4.6 nanofibres revealed that the fibre diameters are around 294±148 nm. After the stability test in ACN, the nanofibre diameters are shown to be about 240±94 nm. The large diameter deviations are the result of the thickness of the sample. Fibres that lay deeper in the SEM picture are seen as thinner by the algorithm although they could be the same thickness. These values as thus merely indicative that the nanofibres have retained their morphology. As an initial experiment to electrospin a PA-4.6 solution containing a CTF, 10 mg of

51 Chapter 7. Results & Discussion

Figure 7.8: Eﬀect of reaction solvents on the stability of electrospun PA-4.6 mats as studied by SEM.

CTF-1 was dispersed in a 5 mL 15 wt.% PA-4.6 solution and electrospun at the above- mentioned conditions. In figure 7.9 SEM images are presented of the entangled CTF-1 particles in the nanofibres. The average diameter of the fibres is 152±54 nm. As the CTF-1 particles are much larger than fibres, one could assume that its pores are not significantly blocked. To analyse the extent of pore blockage, one could perform a N2- sorption measurement. Unfortunately, the amount of CTF-1 present in the sample is to low to be measured. In order to create a functional catalyst, a loading of ca. 10 mg the VPOM@cCTF in ca. 100 mg nanofibres (a 10 wt.% loading) has to be achieved to create a catalytic

52 Chapter 7. Results & Discussion

Figure 7.9: SEM images of the [email protected] proof-of-concept.

carpet of appreciable dimensions. To that end, two solutions were made of 5 mg and 10 mg VPOM@cCTF in 1 mL 15 wt.% PA-4.6 solution. This would correspond to a VPOM@cCTF@PA46-nanofibre mat with a loading of 3.88 wt.% and 7.75 wt.% respectively is the solution was fully electrospun. Although fibres were formed, the process was hindered by electrospraying of the solution resulting in black spots (the droplets contain considerably more black VPOM@cCTF) on the white fibre sheet as can be seen in figure 7.10a. Our hypothesis is that micrometre sized particles are too large to pass the Taylor cone uninterrupted, resulting in droplet formation at the needle tip and ultimately spraying of the solution, see figure 7.10b. A different strategy based on the effect of electrospraying was then attempted. The

53 Chapter 7. Results & Discussion idea was to electrospin PA-4.6 on a rotating drum and electrospraying a VPOM@cCTF solution from a different angle. This requires the dispersion of the catalyst in a solvent that does not dissolve the nanofibres but has a high enough vapour pressure to quickly evaporate. THF was chosen and solutions of 10, 20, 30 and 40 mg VPOM@cCTF in 1 mL THF were prepared. Due to unexpected interaction between the THF and the silicon tube connecting the syringe containing the solution and the needle, and an unexpected non-functional high voltage supply, this experiment unfortunately did not yield noticeable improvements, as can be seen in figure 7.10c. Given the above challenges for the development of an electrospun mat incorporating catalytic VPOM@cCTF particles, the obtained results are preliminary.

Figure 7.10: a) The result of electrospinning the 3.88 wt.% VPOM@cCTF@PA46 solution. b) The Taylor cone, crucial to ES, is repeatedly broken due to droplet formation. c) Although ES of PA-4.6 on the rotating drum proceeded as expected, the electrospraying of the VPOM@cCTF@THF solution proved problematic.

54 Chapter 8

Conclusion & Future Work

To the best of our knowledge, the POM@cCTF system for the oxidative carboxylation of styrene, or any alkene for that matter, presented here is the first of its kind. A cationic CTF (cCTF) material was successfully synthesised based on an imidazolium containing dinitrile linker using the well-known ionothermal synthesis method. This material was characterized using N2-sorption, PXRD, Raman, DRIFTS, TGA, and CHNS. Subsequently, the POM heptammonium tridecavanadomanganate (VPOM) was successfully retained inside the pores of cCTF through ion-ion interactions. The retention was qualitatively confirmed by N2-sorption, PXRD, Raman, DRITFS, and BF-STEM-EDX and quantitatively determined by ICP-AES. Using the mild conditions reported in this work, a styrene carbonate yield of ca. 40 % can be achieved, while revealing some interesting properties about the reaction. Interestingly enough, separate cycloaddition experiments of seemed to imply that aldehydes could be activated under these conditions to yield cyclic carbonates. Lastly, an attempt was made to embed this VPOM@cCTF in electrospun PA-4.6 nanofibres to create an easily recyclable catalytic system. Although the SC yield is not on par with the state-of-the-art, this material is the first of its kind, an attempt upon which can be built. For example, many more different POMs can be tested, different cationic linkers (e.g. a linear linker based on benzodiimidazolium [243]) or a more well-defined cCTF could be synthesized using recently developed mild synthesis methods for CTFs [244, 245]. The recyclability of the material is also a factor that still has to be tested. And a first step to increase the electrospinning possibility could be to ball-mill the cCTF before POM loading (and test the impact of the reduced particle size on the catalytic reaction), using a needle with a larger gauge and/or electrospinning

55 Chapter 8. Conclusion & Future Work from a vertical position downwards to decrease the chance of cCTF build-up at the needle tip. Additionally, the synthesis of COFs using mild reaction methods typically results in nanoscale particles. Particles of this size, e.g. MOFs, have long been successfully electrospun [221, 225, 246]. Reducing the size of CTF particles could be the first step to successfully embed them in electrospun nanofibres. A series of experiments with pure phenylacetaldehyde and different aldehydes, e.g. benzaldehyde or benzenepropanal, should also be performed to see if these substrates could indeed lead to cyclic carbonates.

56 Appendix A

Literature Study

A.1 From Carbon Dioxide to Fine Chemicals

Figure A.1: The Keeling curve, frequently updated by the NOAA [247].

57 Appendix A. Literature Study

Table A.1: The properties of ethylene, propylene and other industrially important organic carbonates [82].

◦ ◦ =3 ◦ Carbonate mp ( C) bp ( C) ρ (g cm ) nD Flash point ( C)

Ethylene carbonate 39 248 1.3218 1.4158 150 Propylene carbonate =48.8 242 1.2069 1.4189 132 Dimethyl carbonate 4 90.2 1.073 1.3687 14 (closed cup) Diethyl carbonate =43 125.8 0.9764 1.3843 33 (closed cup) Diphenyl carbonate 78.8 302 1.1215 at 87 ◦C 168 (closed cup)

Figure A.2: Two possible pathways for the formation of BA through the oxidative cleavage of the C –C epoxide bond, proposed by Maurya [84].

58 Appendix A. Literature Study

A.2 The Polyoxometalate

Figure A.3: The structure of silicotungstic acid as predicted by Pauling, with original description. Although later to be determined a Keggin anion, the point-group symmetry was correctly predicted [162].

Figure A.4: Ball and stick representation of the structures of heptammonium tridecavanadomanganate. Green, orange, and gray represent V, Mn, and O atoms, respectively (the cations are omitted for clarity) [187].

59 Appendix A. Literature Study

A.3 The Covalent Organic Framework

Figure A.5: Linkages that have been used to reticulate building blocks into COFs [248].

60 Appendix A. Literature Study

Figure A.6: Idealized trimerization of DCB to trimers and oligomers and then to CTF-1 [206].

A.4 The Electrospinning

Table A.2: The parameters inﬂuencing the ES process [221].

Solution parameters Processing parameters Ambient parameters

Concentration Applied voltage Air ﬂow Conductivity Flow rate Relative Humidity Molecular weight (distribution) Collector shape and motion Temperature Solvent Polarity Injector-collector distance Solvent vapor pressure Needle gauge Surface tension Visco-elasticity

Figure A.7: a) The PVDF & b) PAN structures.

Appendix B

Materials & Methods

B.1 Used chemicals & equipment

A list of the used equipment and a table of the used chemicals.

BF-STEM Bright-field scanning transmission electron microscopy (BF-STEM) was performed on a JEOL JEM-2200FS high-resolution scanning transmission electron microscope with a spatial resolution of 0.13 nm, an image lens spherical aberration cor- rector, an electron energy loss spectrometer (filter), and a field emission gun (FEG) operating at 200 keV. The technique is combined in parallel with energy-dispersive X-ray spectroscopy (EDX), allowing for elemental mapping.

CHNS Elemental analysis (C, H, N, and O) were carried out on a ThermoScientiﬁc Flash 2000 CHNS-O analyzer equipped with a TCD detector.

DRIFTS Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) in the region of 4000-650 cm=1 was performed with a ThermoScientific Nicolet 6700 FT-IR spectrometer equipped with a liquid nitrogen-cooled MCT/A detector and a KBr beam splitter in autogain modus. The scans were taken at a 0.07 cm=1 resolution. The background correction was based on 256 scans. 256 scans were taken of each sample.

GC-MS Gas chromatography-mass spectrometry was used to determine the reaction yields. The chromatographs were acquired on a Hewlett Packard 6890 GC system

63 Appendix B. Materials & Methods

with the non-polar Agilent J&W Scientiﬁc DB-5ms column. The mass spectrometry detector was a Hewlett Packard 5973 Mass Selective Detector.

ICP-AES The POM loading was determined using inductively coupled plasma atomic emission spectroscopy (ICP-AES) on an Optima 8000 spectrometer. The destruction was performed according to the WAC/III/B/001 standard procedure. The analysis was performed according to the WAC/III/B/010 standard procedure.

Muﬄe furnace A Nabertherm P330.

NMR The 1H and 13C NMR analysis of the nitrile linker was performed on a Bruker AVANCE 300 MHz spectrometer using DMSO as internal standard.

N2-sorption Dinitrogen (N2) adsorption isotherms were obtained using a Belsorp Mini apparatus measured at 77 K.

PXRD Powder X-ray diffraction (PXRD) patterns were collected on a ThermoScientific ARL X’Tra diffractometer, operated at 40 kV, 30 mA using Cu Kα radiation (λ = 1.5406 A).¦

Raman Raman spectra were acquired on a Kaiser Optical Systems, Inc. RamanRxn Systems RXN1-532. The green laser had a wavelenghts of 532 nm. Spectra were taken from 4350 to 100 cm=1 for a total target exposure time of 3 min. Cosmic ray removal and intensity correction were enabled.

SEM SEM images were obtained using a FEI Company Phenom Pro Desktop SEM.

TGA Thermogravimetric analysis (TGA) was performed on a Netzsch STA-449 F3 Jupiter- simultaneous TG-DSC analyser within a temperature range of 20-1000 ◦C, under a ◦ =1 80/20 N2/O2 atmosphere at a heating rate of 10 C min .

64 Appendix B. Materials & Methods

Table B.1: Alphabetical list of used chemicals during the research.

All products were used as received.

Name CAS-number Supplier Purity

Acetic acid 64-19-7 Sigma-Aldrich ≥99 % Acetonitrile 75-05-8 Fisher Scientiﬁc HPLC grade 2-Bromo-5-cyanopyridine 139585-70-9 TCI >97.0 % Carbon dioxide 124-38-9 Air Liquide ≥99.7 % Chlorobenzene 108-90-7 Sigma-Aldrich 99.8 % Chloroform 67-66-3 Sigma-Aldrich ≥99.5 % Dichloromethane 75-09-2 Sigma-Aldrich 99 % 1,2-Dichloroethane 107-06-2 ROMIL Chemicals ≥99.8 % Dimethyl sulfoxide 67-68-5 Sigma-Aldrich ≥99.5 % Formic acid 64-18-6 Sigma-Aldrich ≥95 % Hydrochloric acid 7647-01-0 Carl Roth 37wt.% Methanol 67-56-1 Fisher Scientiﬁc ≥99.5 % 1-Methyl-1H -imidazole 616-47-7 Sigma-Aldrich 99 % Polyamide-4.6 50327-22-5 Goodfellow - Phosphotungstic acid 12501-23-4 Sigma-Aldrich Reagent grade Styrene 100-42-5 Sigma-Aldrich ≥99 % Styrene oxide 96-09-3 Sigma-Aldrich 97 % Tetrabutylammonium bromide 1643-19-2 Sigma-Aldrich ≥98.0 % Tetrahydrofuran 109-99-9 Sigma-Aldrich ≥99.0 % Tert-butylhydroperoxide 75-91-2 Sigma-Aldrich 5.0-6.0 M in decane Zinc chloride 7646-85-7 TCI ≥98.0 %

65 Appendix B. Materials & Methods

B.2 Analysis of the Cationic Dinitrile Linker

Figure B.1: The 1H NMR spectrum of bpim.

Figure B.2: The 13C NMR spectrum of bpim.

66 Appendix B. Materials & Methods

B.3 Analysis of the VPOM@cCTF

Figure B.3: Overlay of the PXRD spectrum of VPOM and its predicted spectrum.

Figure B.4: Raman spectra overlay of the cCTF, VPOM and 20 wt.% VPOM@cCTF.

67 Appendix B. Materials & Methods

Figure B.5: IR spectra overlay of the cCTF, VPOM and 11.2 wt.% VPOM@cCTF.

Figure B.6: IR spectra overlay of the cCTF, VPOM and 20 wt.% VPOM@cCTF.

68 Bibliography

[1] Qiuxia Han, Bo Qi, Weimin Ren, Cheng He, Jingyang Niu, and Chunying Duan. Polyoxometalate-based homochiral metal-organic frameworks for tandem asymmetric transformation of cyclic carbonates from oleﬁns. Nature Communications, 6(1):10007, dec 2015. ISSN 2041-1723. doi: 10.1038/ncomms10007. URL http://www.nature.com/articles/ncomms10007.

[2] Heping Ma, Bailing Liu, Bin Li, Liming Zhang, Yang-Guang Li, Hua-Qiao Tan, Hong-Ying Zang, and Guangshan Zhu. Cationic Covalent Organic Frameworks: A Simple Platform of Anionic Exchange for Porosity Tuning and Proton Conduction. Journal of the American Chemical Society, 138(18):5897–5903, may 2016. ISSN 0002-7863. doi: 10.1021/jacs.5b13490. URL http://pubs.acs.org/doi/10.1021/jacs. 5b13490.

[3] J´erˆomeRoeser, Kamalakannan Kailasam, and Arne Thomas. Covalent Triazine Frameworks as Heterogeneous Catalysts for the Synthesis of Cyclic and Linear Carbonates from Carbon Dioxide and Epoxides. ChemSusChem, 5(9):1793–1799, sep 2012. ISSN 18645631. doi: 10.1002/cssc.201200091. URL http://doi.wiley.com/10.1002/cssc.201200091.

[4] Luca Verhoeven. Metal organic framework loaded electrospun poly--caprolactone scaﬀolds as novel catalytic system, 2017. URL https: //lib.ugent.be/fulltxt/RUG01/002/349/805/RUG01-002349805{_}2017{_}0001{_}AC.pdf.

[5] Karen Leus, Chidharth Krishnaraj, Luca Verhoeven, V´eronique Cremers, Jolien Dendooven, Ranjith K. Ramachandran, Peter Dubruel, and Pascal Van Der Voort. Catalytic carpets: Pt@MIL-101@electrospun PCL, a surprisingly active and robust hydrogenation catalyst. Journal of Catalysis, 360:81–88, apr 2018. ISSN 00219517. doi: 10.1016/j.jcat.2018.01.018. URL https://doi.org/10.1016/j.jcat.2018.01. 018https://linkinghub.elsevier.com/retrieve/pii/S0021951718300344.

[6] Gale E. Christianson. Greenhouse: The 200-Year Story of Global Warming. Walker, New York City, 1999.

[7] Joseph Fourier. “Remarques Généralessur les TempératuresDu Globe Terrestre et des Espaces Planétaires.”Annales de Chemie et de Physique 27: 136-67. Translation by Ebeneser Burgess, “General Remarks on the Temperature of the Earth and Outer Space”. American Journal of Science, 32(1):1–20, 1837.

[8] John Tyndall. On the Absorption and Radiation of Heat by Gases and Vapours, and on the Physical Connection of Radiation, Absorption, and Conduction. Philosophical Magazine and Journal of Science, 22(146):169 – 194, 1861.

[9] John Tyndall. V. On the relation of radiant heat to aqueous vapour. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 26(172):30–44, jul 1863. ISSN 1941-5982. doi: 10.1080/14786446308643514. URL https://archive.org/details/ philtrans06503104https://www.tandfonline.com/doi/full/10.1080/14786446308643514.

[10] John Tyndall. On radiation through the earth’s atmosphere. Journal of the Franklin Institute, 77(6):413–418, jun 1864. ISSN 00160032. doi: 10.1016/0016-0032(64)90258-3. URL http://linkinghub.elsevier.com/retrieve/pii/0016003264902583.

[11] John Tyndall. Contributions to Molecular Physics in the Domain of Radiant Heat. Nature, 7(161):66–67, 1872. ISSN 0028-0836. doi: 10.1038/007066a0. URL http://www.tandfonline.com/doi/abs/10.1080/08912963.2014.890191http://www.nature.com/articles/007066a0.

[12] John Tyndall. Contributions to Molecular Physics in the Domain of Radiant Heat. D. Appleton and Company, 1873.

[13] Svante Arrhenius. XXXI. On the inﬂuence of carbonic acid in the air upon the temperature of the ground. The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 41(251):237–276, apr 1896. ISSN 1941-5982. doi: 10.1080/14786449608620846. URL https://www.tandfonline.com/doi/full/10.1080/14786449608620846.

[14] T. C. Chamberlin. A Group of Hypotheses Bearing on Climatic Changes. The Journal of Geology, 5(7):653–683, oct 1897. ISSN 0022-1376. doi: 10.1086/607921. URL https://www.journals.uchicago.edu/doi/10.1086/607921.

[15] T. C. Chamberlin. The Inﬂuence of Great Epochs of Limestone Formation upon the Constitution of the Atmosphere. The Journal of Geology, 6(6):609–621, sep 1898. ISSN 0022-1376. doi: 10.1086/608185. URL https://www.journals.uchicago.edu/doi/10.1086/608185.

69 Bibliography

[16] T. C. Chamberlin. An Attempt to Frame a Working Hypothesis of the Cause of Glacial Periods on an Atmospheric Basis. The Journal of Geology, 7(8):751–787, nov 1899. ISSN 0022-1376. doi: 10.1086/608524. URL https://www.journals.uchicago.edu/doi/10.1086/608524.

[17] T. C. Chamberlin. On a Possible Reversal of Deep-Sea Circulation and Its Inﬂuence on Geologic Climates. The Journal of Geology, 14 (5):363–373, jul 1906. ISSN 0022-1376. doi: 10.1086/621315. URL https://www.journals.uchicago.edu/doi/10.1086/621315.

[18] G. S. Callendar. The artiﬁcial production of carbon dioxide and its inﬂuence on temperature. Quarterly Journal of the Royal Meteorological Society, 64(275):223–240, sep 1938. ISSN 00359009. doi: 10.1002/qj.49706427503. URL http://doi.wiley.com/10.1002/qj.49706427503.

[19] G. S. Callendar. Variations of the amount of carbon dioxide in diﬀerent air currents. Quarterly Journal of the Royal Meteorological Society, 66(287):395–400, sep 1940. ISSN 00359009. doi: 10.1002/qj.49706628705. URL http://doi.wiley.com/10.1002/qj.49706628705.

[20] G. S. Callendar. Infra-red absorption by carbon dioxide, with special reference to atmospheric radiation. Quarterly Journal of the Royal Meteorological Society, 67(291):263–275, sep 1941. ISSN 00359009. doi: 10.1002/qj.49706729105. URL http://doi.wiley.com/10.1002/qj. 49706729105.

[21] G. S. Callendar. CAN CARBON DIOXIDE INFLUENCE CLIMATE? Weather, 4(10):310–314, oct 1949. ISSN 00431656. doi: 10.1002/ j.1477-8696.1949.tb00952.x. URL http://doi.wiley.com/10.1002/j.1477-8696.1949.tb00952.x.

[22] G. S. Callendar. On the Amount of Carbon Dioxide in the Atmosphere. Tellus, 10(2):243–248, jan 1958. ISSN 0040-2826. doi: 10.3402/ tellusa.v10i2.9231. URL https://www.tandfonline.com/doi/full/10.3402/tellusa.v10i2.9231.

[23] G. S. Callendar. Temperature ﬂuctuations and trends over the earth. Quarterly Journal of the Royal Meteorological Society, 87(371):1–12, jan 1961. ISSN 00359009. doi: 10.1002/qj.49708737102. URL http://doi.wiley.com/10.1002/qj.49708737102.

[24] Gilbert N. Plass. Eﬀect of Carbon Dioxide Variations on Climate. American Journal of Physics, 24(5):376–387, may 1956. ISSN 0002-9505. doi: 10.1119/1.1934233. URL http://aapt.scitation.org/doi/10.1119/1.1934233.

[25] Gilbert N. Plass. Carbon Dioxide and the Climate. American Scientist, 44(3):302–316, 1956. URL http://www.jstor.org/stable/27826805.

[26] Giblert N. Plass. The Carbon Dioxide Theory of Climatic Change. Tellus, 8(2):140–154, 1956. ISSN 0040-2826. doi: 10.3402/tellusa. v8i2.8969. URL http://tellusa.net/index.php/tellusa/article/view/8969.

[27] Roger Revelle and Hans E. Suess. Carbon Dioxide Exchange Between Atmosphere and Ocean and the Question of an Increase of Atmospheric CO2 during the Past Decades. Tellus, 9(1):18–27, jan 1957. ISSN 0040-2826. doi: 10.3402/tellusa.v9i1.9075. URL https: //www.tandfonline.com/doi/full/10.3402/tellusa.v9i1.9075.

[28] Charles D. Keeling. The Concentration and Isotopic Abundances of Carbon Dioxide in the Atmosphere. Tellus, 12(2):200–203, jan 1960. ISSN 0040-2826. doi: 10.3402/tellusa.v12i2.9366. URL http://tellusa.net/index.php/tellusa/article/view/9366.

[29] Syukuro Manabe and Richard T. Wetherald. Thermal Equilibrium of the Atmosphere with a Given Distribution of Relative Humidity. Journal of the Atmospheric Sciences, 24(3):241–259, may 1967. ISSN 0022-4928. doi: 10.1175/1520-0469(1967)024h0241:TEOTAWi2.0. CO;2. URL http://journals.ametsoc.org/doi/abs/10.1175/1520-0469{%}281967{%}29024{%}3C0241{%}3ATEOTAW{%}3E2.0.CO{%}3B2.

[30] Richard Nixon. Reorganization Plan No. 4 of 1970, 1970. ISSN 00333352. URL http://search.ebscohost.com/login.aspx?direct=true{&}db= bth{&}AN=4601724{&}site=ehost-live{&}custid=magn1307.

[31] Peter Passell, Marc Roberts, and Leonard Ross. The Limits to Growth, apr 1972. URL https://nyti.ms/1HczNl7.

[32] Carl Kaysen. The Computer That Printed out W*O*L*F*. Foreign Aﬀairs, 50(4):660, mar 1972. ISSN 00157120. doi: 10.2307/20037939. URL https://www.jstor.org/stable/10.2307/20037939?origin=crossref.

[33] H. S. D. Cole. Models of Doom: A Critique of The Limits to Growth. Universe Books, 1973. ISBN 0876639058.

[34] Robert M. Solow. Is the End of the World at Hand? Challenge, 16(1):39–50, mar 1973. ISSN 0577-5132. doi: 10.1080/05775132.1973. 11469961. URL https://www.tandfonline.com/doi/full/10.1080/05775132.1973.11469961.

[35] Julian L. Simon. The Ultimate Resource 2. Princeton University Press, 1996. ISBN 0-691-04269-1.

[36] Alan AtKisson. Believing Cassandra: How to be an Optimist in a Pessimist’s World. Routledge, 2011. ISBN 1849711720.

[37] Tim Worstall. The Club of Rome’s Limits to Growth Updated: Entirely Bizarre, apr 2012. URL https://www.forbes.com/sites/timworstall/ 2012/04/09/the-club-of-romes-limits-to-growth-updated-entirely-bizarre/{#}503d7b877394.

[38] Matthew R. Simmons. Revisiting The Limits to Growth : Could The Club of Rome Have Been Correct, After All?, 2000.

70 Bibliography

[39] Graham M. Turner. A comparison of The Limits to Growth with 30 years of reality. Global Environmental Change, 18:397–411, 2008. doi: 10.1016/j.gloenvcha.2008.05.001.

[40] Jeﬀ Hecht. Prophesy of economic collapse ‘coming true’. New Scientist, nov 2008. URL https://www.newscientist.com/article/ dn16058-prophesy-of-economic-collapse-coming-true/.

[41] Ugo Bardi. Cassandra’s curse: how ”The Limits to Growth” was demonized, 2008. URL http://europe.theoildrum.com/pdf/theoildrum{_}3551. pdf.

[42] Charles A. S. Hall and John W. Day. Revisiting the Limits to Growth After Peak Oil. American Scientist, 97(May-June):230–237, 2009.

[43] D.P. Van Vuuren and A. Faber. Growing within Limits. Technical report, Netherlands Environmental Assessment Agency (PBL), Bilthoven, 2009. URL http://www.rivm.nl/bibliotheek/rapporten/500201001.pdf.

[44] Jørgen Nørgaard, Krist´ınVala Ragnarsd´ottir,and John Peet. The History of The Limits to Growth. Solutions Journal, 1(2):59–63, 2010.

[45] Ugo Bardi. The Limits to Growth Revisited. Springer Science+Business Media, New York City, 2011. ISBN 978-1-4419-9415-8. doi: 10.1007/978-1-4419-9416-5.

[46] Chrisian Parenti. ‘The Limits to Growth’: A Book That Launched a Movement. The Nation, dec 2012. URL https://www.thenation.com/ article/limits-growth-book-launched-movement/.

[47] Graham M. Turner. Is Global Collapse Imminent? Technical report, Melbourne Sustainable Society Institute, Melbourne, 2014.

[48] Graham M. Turner and Cathy Alexander. Limits to Growth was right. New research shows we’re nearing collapse, sep 2014. URL https://www.theguardian.com/commentisfree/2014/sep/02/limits-to-growth-was-right-new-research-shows-were-nearing-collapse.

[49] Tim Jackson and Robin Webster. Limits Revisited: a review of the limits to growth debate. Technical Report April, All-Party Parlia- mentary Group (APPG), 2016. URL http://limits2growth.org.uk/revisited.

[50] National Research Council. Carbon Dioxide and Climate: A Scientiﬁc Assessment. The National Academies Press, Washington, D.C., 1979. ISBN 978-0-309-11910-8. doi: 10.17226/12181. URL http://nap.edu/12181.

[51] UNEP/WMO/ICSU. An Assessment of the Role of Carbon Dioxide and of Other Greenhouse Gases in Climate Variations and Assorted Impacts. Technical Information Circular 203. In Greenhouse Gases and Climate, page 8, Wellington, NZ, 1985.

[52] IPCC. Principles Governing IPCC Work. Technical Report October, Intergovernmental Panel on Climate Change, Batumi, 2013. URL https://www.ipcc.ch/site/assets/uploads/2018/09/ipcc-principles.pdf.

[53] IPCC. Climate Change The IPCC Scientiﬁc Assessment. Cambridge University Press, Cambridge, 1990. ISBN 0 521 40360 X. URL https://www.ipcc.ch/site/assets/uploads/2018/03/ipcc{_}far{_}wg{_}I{_}full{_}report.pdf.

[54] IPCC. Climate Chance The IPCC Impacts Assessment. Australian Government Publishing Service, Canberra, 1990. ISBN 0 644 13497 6. URL https://www.ipcc.ch/site/assets/uploads/2018/03/ipcc{_}far{_}wg{_}II{_}full{_}report.pdf.

[55] IPCC. Climate Change The IPCC Response Strategies. Intergovernmental Panel on Climate Change, 1990. URL https://www.ipcc.ch/site/ assets/uploads/2018/03/ipcc{_}far{_}wg{_}III{_}full{_}report.pdf.

[56] IPCC. IPCC Second Assessment Climate Change 1995. Intergovernmental Panel on Climate Change, 1995. doi: 10.1021/es405168b. URL https://www.ipcc.ch/site/assets/uploads/2018/05/2nd-assessment-en-1.pdf.

[57] IPCC. Climate Change 2001: Synthesis Report. A Contribution of Working Groups I, II, and III to the Third Assessment Report of the Integovernmental Panel on Climate Change. Cambridge University Press, Cambridge, 2001. ISBN 0 521 80770 0. URL https://www.ipcc. ch/site/assets/uploads/2018/05/SYR{_}TAR{_}full{_}report.pdf.

[58] IPCC. Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assess- ment Report of the Intergovernmental Panel on Climate Change. Intergovernmental Panel on Climate Change, Geneva, Switzerland, 2008. ISBN 92-9169-122- 4. URL https://www.ipcc.ch/site/assets/uploads/2018/02/ar4{_}syr{_}full{_}report.pdf.

[59] IPCC. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergov- ernmental Panel on Climate Change. IPCC, Geneva, Switzerland, 2014. ISBN 978-92-9169-143-2. URL https://www.ipcc.ch/site/assets/ uploads/2018/02/SYR{_}AR5{_}FINAL{_}full.pdf.

[60] Sebastian Oberth¨urand Hermann E. Ott. The Kyoto Protocol. Springer Berlin Heidelberg, Berlin, Heidelberg, 1999. ISBN 978-3-642- 08575-8. doi: 10.1007/978-3-662-03925-0. URL http://link.springer.com/10.1007/978-3-662-03925-0.

71 Bibliography

[61] William K. Stevens. The Change in the Weather: People, Weather, and the Science of Climate. Delacorte Press, New York City, 1999. ISBN 0385320078.

[62] UNFCCC. Paris Agreement on Climate, 2015. ISSN 1098-6596.

[63] Suzanne Goldenberg. How US negotiators ensured landmark Paris climate deal was Republican-proof, dec 2015. URL https://www. theguardian.com/us-news/2015/dec/13/climate-change-paris-deal-cop21-obama-administration-congress-republicans-environment.

[64] Andrew Restuccia. The one word that almost sank the climate talks, dec 2015. URL https://www.politico.eu/article/ one-word-almost-sunk-climate-talks-legally-binding-cop21-deal-global-warming/.

[65] NASA and GISS. Global temperature, 2018. URL http://climate.nasa.gov/vital-signs/global-temperature/.

[66] Members of the Club of Rome. The Club of Rome Climate Emergency Plan - A Collaborative Call for Climate Action. Technical report, The Club of Rome, 2018. URL http://www.clubofrome.org/.

[67] Xu Xiaoding and J. A. Moulijn. Mitigation of CO2 by Chemical Conversion: Plausible Chemical Reactions and Promising Products. Energy & Fuels, 10(2):305–325, mar 1996. ISSN 0887-0624. doi: 10.1021/ef9501511. URL http://pubs.acs.org/doi/10.1021/ef9501511.

[68] Hongqun Yang, Zhenghe Xu, Maohong Fan, Rajender Gupta, Rachid B. Slimane, Alan E. Bland, and Ian Wright. Progress in carbon dioxide separation and capture: A review. Journal of Environmental Sciences, 20(1):14–27, jan 2008. ISSN 10010742. doi: 10.1016/ S1001-0742(08)60002-9. URL https://linkinghub.elsevier.com/retrieve/pii/S1001074208600029.

[69] International Energy Agency. CO2 Emissions from Fuel Combustion 2015. OECD, nov 2015. ISBN 9789264245952. doi: 10.1787/ co2 fuel-2015-en. URL https://www.oecd-ilibrary.org/energy/co2-emissions-from-fuel-combustion-2015{_}co2{_}fuel-2015-en.

[70] Working Group III of the Intergovernmental Panel on Climate Change. Carbon Dioxide Capture and Storage. Cambridge University Press, 2005. ISBN 13 978-0-521-86643-9. URL http://www.ipcc.ch/pdf/special-reports/srccs/srccs{_}wholereport.pdf.

[71] Michele Aresta, Angela Dibenedetto, and Antonella Angelini. The changing paradigm in CO2 utilization. Journal of CO2 Utilization, 3-4: 65–73, dec 2013. ISSN 22129820. doi: 10.1016/j.jcou.2013.08.001. URL http://dx.doi.org/10.1016/j.jcou.2013.08.001https://linkinghub. elsevier.com/retrieve/pii/S2212982013000322.

[72] Jos G.J. Olivier, Greet Janssens-Maenhout, Marilena Muntean, and Jeroen A.H.W. Peters. Trends in Global CO2 emissions: 2016 report. Technical report, PBL Netherlands Environmental Assessment Agency, The Hague, 2016. URL http://edgar.jrc.ec.europa.eu/news{_}docs/ jrc-2016-trends-in-global-co2-emissions-2016-report-103425.pdf.

[73] Mette Mikkelsen, Mikkel Jørgensen, and Frederik C. Krebs. The teraton challenge. A review of ﬁxation and transformation of carbon dioxide. Energy Environ. Sci., 3(1):43–81, 2010. ISSN 1754-5692. doi: 10.1039/B912904A. URL http://xlink.rsc.org/?DOI=B912904A.

[74] Robert H. Socolow, Michael J. Desmond, Roger Aines, Jason J. Blackstock, Olav Bolland, Tina Kaarsberg, Nathan Lewis, Marco Mazzotti, Allen Pfeffer, Jeffrey J. Siirola, Karma Sawyer, Berend Smit, and Jennifer Wilcox. Direct Air Capture of CO2 with Chemicals. Technical report, APS Panel on Public Affairs, 2011. URL https://www.aps.org/policy/reports/assessments/upload/dac2011.pdf.

[75] David W Keith, Geoﬀrey Holmes, David St. Angelo, and Kenton Heidel. A Process for Capturing CO2 from the Atmosphere. Joule, 2(8):1573–1594, aug 2018. ISSN 25424351. doi: 10.1016/j.joule.2018.05.006. URL https://doi.org/10.1016/j.joule.2018.05.006https: //linkinghub.elsevier.com/retrieve/pii/S2542435118302253.

[76] Jeﬀ Tollefson. Sucking carbon dioxide from air is cheaper than scientists thought. Nature, 558(7709):173–173, jun 2018. ISSN 0028-0836. doi: 10.1038/d41586-018-05357-w. URL http://www.nature.com/articles/d41586-018-05357-w.

[77] Climeworks. Climeworks - About Us, 2019. URL http://www.climeworks.com/about/.

[78] M. Aresta, E. Quaranta, I. Tommasi, P. Giannocaro, and A. Ciccarese. Enzymatic Versus Chemical Carbon Dioxide Utilization. Gazzetta Chimica Italiana, 125(11):509–538, 1995.

[79] Michele Aresta, Angela Dibenedetto, and Antonella Angelini. Catalysis for the valorization of exhaust carbon: From CO2 to chemicals, materials, and fuels. technological use of CO2. Chemical Reviews, 114(3):1709–1742, 2014. ISSN 00092665. doi: 10.1021/cr4002758.

[80] The Group of Chief Scientiﬁc Advisors. Novel carbon capture and utilisation technologies. Publications Oﬃce of the European Union, 2018. ISBN 9789279820076. doi: 10.2777/01532.

[81] Toshiyasu Sakakura and Kazufumi Kohno. The synthesis of organic carbonates from carbon dioxide. Chemical Communications, 0(11): 1312–1330, 2009. ISSN 13597345. doi: 10.1039/b819997c.

72 Bibliography

[82] Hans-Josef Buysch. Carbonic Esters. In Ullmann’s Encyclopedia of Industrial Chemistry, volume 7, pages 45–71. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, jun 2000. ISBN 1073-449X. doi: 10.1002/14356007.a05 197. URL http://doi.wiley.com/10.1002/ 14356007.a05{_}197.

[83] Deryn E. Fogg and Eduardo N. dos Santos. Tandem catalysis: a taxonomy and illustrative review. Coordination Chemistry Reviews, 248(21-24):2365–2379, dec 2004. ISSN 00108545. doi: 10.1016/j.ccr.2004.05.012. URL https://linkinghub.elsevier.com/retrieve/pii/ S001085450400133X.

[84] Cristina I. Fernandes, Nuno U. Silva, Pedro D. Vaz, Teresa G. Nunes, and Carla D. Nunes. Bio-inspired Mo(II) complexes as active catalysts in homogeneous and heterogeneous oleﬁn epoxidation. Applied Catalysis A: General, 384(1-2):84–93, aug 2010. ISSN 0926860X. doi: 10.1016/j.apcata.2010.06.011. URL http://dx.doi.org/10.1016/j.apcata.2010.06.011http://linkinghub.elsevier.com/retrieve/ pii/S0926860X10004187.

[85] Mannar R. Maurya, Maneesh Kumar, and Sweta Sikarwar. Polymer-anchored oxoperoxo complexes of vanadium(V), molybdenum(VI) and tungsten(VI) as catalyst for the oxidation of phenol and styrene using hydrogen peroxide as oxidant. Reactive and Functional Polymers, 66(8):808–818, aug 2006. ISSN 13815148. doi: 10.1016/j.reactfunctpolym.2005.11.007. URL https://linkinghub.elsevier.com/retrieve/pii/ S1381514805002889.

[86] Carlo Venturello, Enzo Alneri, and Marco Ricci. A new, eﬀective catalytic system for epoxidation of oleﬁns by hydrogen peroxide under phase-transfer conditions. The Journal of Organic Chemistry, 48(21):3831–3833, oct 1983. ISSN 0022-3263. doi: 10.1021/jo00169a052. URL http://pubs.acs.org/doi/abs/10.1021/jo00169a052.

[87] Ishii Yasutaka, Kazumasa Yamawaki, Ura Toshikazu, Hiroshi Yamada, Tsutomu Yoshida, and Ogawa Masaya. Hydrogen Peroxide Oxida- tion Catalyzed by Heteropoly Acids Combined with Cetylpyridinium Chloride: Epoxidation of Oleﬁns and Allylic Alcohols, Ketonization of Alcohols and Diols, and Oxidative Cleavage of 1,2-Diols and Oleﬁns. Journal of Organic Chemistry, 53(15):3587–3593, 1988. ISSN 15206904. doi: 10.1021/jo00250a032.

[88] Masatoshi Hamamoto, Kouichi Nakayama, Yutaka Nishiyama, and Yasutaka Ishii. Oxidation of Organic Substrates by Molecu- lar Oxygen/Aldehyde/Heteropolyoxometalate System. Journal of Organic Chemistry, 58(23):6421–6425, 1993. ISSN 15206904. doi: 10.1021/jo00075a043.

[89] Dean C. Duncan, R. Carlisle Chambers, Eric Hecht, and Craig L. Hill. Mechanism and Dynamics in the H3[PW12O40]-Catalyzed Selective Epoxidation of Terminal Oleﬁns by H2O2. Formation, Reactivity, and Stability of {PO4[WO(O2)2]4}3-. Journal of the American Chemical Society, 117(2):681–691, 1995. ISSN 15205126. doi: 10.1021/ja00107a012.

[90] G. D. Yadav and D. V. Satoskar. Kinetics of epoxidation of alkyl esters of undecylenic acid: Comparison of traditional routes vs. ishii-venturello chemistry. JAOCS, Journal of the American Oil Chemists’ Society, 74(4):397–407, 1997. ISSN 0003021X. doi: 10.1007/ s11746-997-0097-x.

[91] G. D. Yadav and A. A. Pujari. Epoxidation of styrene to styrene oxide: Synergism of heteropoly acid and phase-transfer catalyst under Ishii-Venturello mechanism. Organic Process Research and Development, 4(2):88–93, 2000. ISSN 10836160. doi: 10.1021/op990055p.

[92] Nadya S. Antonova, Jorge J. Carb´o,Ulrich Kortz, Oxana A. Kholdeeva, and Josep M. Poblet. Mechanistic insights into alkene epoxidation with H2O2 by Ti- and other TM-containing polyoxometalates: Role of the metal nature and coordination environment. Journal of the American Chemical Society, 132(21):7488–7497, 2010. ISSN 00027863. doi: 10.1021/ja1023157.

[93] Filippo Minutolo, Dario Pini, and Piero Salvadori. Polymer-bound chiral (salen)Mn(III) complex as heterogeneous catalyst in rapid and clean enantioselective epoxidation of unfunctionalised oleﬁns. Tetrahedron Letters, 37(19):3375–3378, may 1996. ISSN 00404039. doi: 10.1016/0040-4039(96)00550-3. URL https://linkinghub.elsevier.com/retrieve/pii/0040403996005503.

[94] Mark D. Angelino and Paul E. Laibinis. Synthesis and Characterization of a Polymer-Supported Salen Ligand for Enantioselective Epoxidation. Macromolecules, 31(22):7581–7587, nov 1998. ISSN 0024-9297. doi: 10.1021/ma9804803. URL https://pubs.acs.org/doi/10. 1021/ma9804803.

[95] Biwei Gong, Xiangkai Fu, Junxian Chen, Yuedong Li, Xiaochuan Zou, Xiaobo Tu, Pingping Ding, and Liping Ma. Synthesis of a new type of immobilized chiral salen Mn(III) complex as eﬀective catalysts for asymmetric epoxidation of unfunctionalized oleﬁns. Journal of Catalysis, 262(1):9–17, feb 2009. ISSN 00219517. doi: 10.1016/j.jcat.2008.11.027. URL https://linkinghub.elsevier.com/retrieve/pii/ S002195170800434X.

[96] K.C. Gupta, Alekha Kumar Sutar, and Chu-Chieh Lin. Polymer-supported Schiﬀ base complexes in oxidation reactions. Coordination Chemistry Reviews, 253(13-14):1926–1946, jul 2009. ISSN 00108545. doi: 10.1016/j.ccr.2009.03.019. URL https://linkinghub.elsevier.com/ retrieve/pii/S0010854509000885.

[97] Kareen Brown, Santiago Zolezzi, Pedro Aguirre, Diego Venegas-Yazigi, Ver´onicaParedes-Garc´ıa,Ricardo Baggio, Miguel A. Novak, and Evgenia Spodine. [Cu(H2btec)(bipy)]: a novel metal organic framework (MOF) as heterogeneous catalyst for the oxidation of oleﬁns. Dalton Transactions, 0(8):1422, 2009. ISSN 1477-9226. doi: 10.1039/b810414j. URL http://xlink.rsc.org/?DOI=b810414j.

73 Bibliography

[98] Nataliya V. Maksimchuk, Konstantin A. Kovalenko, Sergey S. Arzumanov, Yurii A. Chesalov, Maxim S. Melgunov, Alexander G. Stepanov, Vladimir P. Fedin, and Oxana A. Kholdeeva. Hybrid Polyoxotungstate/MIL-101 Materials: Synthesis, Characterization, and Catalysis of H2O2-Based Alkene Epoxidation. Inorganic Chemistry, 49(6):2920–2930, mar 2010. ISSN 0020-1669. doi: 10.1021/ic902459f. URL https://pubs.acs.org/doi/10.1021/ic902459f.

[99] Samiran Bhattacharjee, Da-Ae Yang, and Wha-Seung Ahn. A new heterogeneous catalyst for epoxidation of alkenes via one-step post- functionalization of IRMOF-3 with a manganese(ii) acetylacetonate complex. Chemical Communications, 47(12):3637, 2011. ISSN 1359- 7345. doi: 10.1039/c1cc00069a. URL http://xlink.rsc.org/?DOI=c1cc00069a.

[100] Matthias J. Beier, Wolfgang Kleist, Michael T. Wharmby, Reinhard Kissner, Bertram Kimmerle, Paul A. Wright, Jan-Dierk Grunwaldt, and Alfons Baiker. Aerobic Epoxidation of Oleﬁns Catalyzed by the Cobalt-Based Metal-Organic Framework STA-12(Co). Chemistry - A European Journal, 18(3):887–898, jan 2012. ISSN 09476539. doi: 10.1002/chem.201101223. URL http://doi.wiley.com/10.1002/chem. 201101223.

[101] Hedayat Haddadi, Shahrbanou Moradpour Hafshejani, Mostafa Riahi Farsani, and Ali Kazemi Babahydari. Heterogeneous epoxidation of alkenes with H2O2 catalyzed by a recyclable organic-inorganic polyoxometalate-based framework catalyst. New Journal of Chemistry, 39(12):9879–9885, 2015. ISSN 1144-0546. doi: 10.1039/C5NJ01661D. URL http://dx.doi.org/10.1039/C5NJ01661Dhttp://xlink.rsc.org/?DOI= C5NJ01661D.

[102] Qinghong Zhang, Ye Wang, Satoko Itsuki, Tetsuya Shishido, and Katsuomi Takehira. Manganese-containing MCM-41 for epoxidation of styrene and stilbene. Journal of Molecular Catalysis A: Chemical, 188(1-2):189–200, sep 2002. ISSN 13811169. doi: 10.1016/S1381-1169(02) 00323-0. URL https://linkinghub.elsevier.com/retrieve/pii/S1381116902003230.

[103] Huidong Zhang, Song Xiang, Jianliang Xiao, and Can Li. Heterogeneous enantioselective epoxidation catalyzed by Mn(salen) complexes grafted onto mesoporous materials by phenoxy group. Journal of Molecular Catalysis A: Chemical, 238(1-2):175–184, sep 2005. ISSN 13811169. doi: 10.1016/j.molcata.2005.05.024. URL https://linkinghub.elsevier.com/retrieve/pii/S1381116905003328.

[104] R KURESHY, I AHMAD, N KHAN, S ABDI, K PATHAK, and R JASRA. Chiral Mn(III) salen complexes covalently bonded on modiﬁed MCM-41 and SBA-15 as eﬃcient catalysts for enantioselective epoxidation of nonfunctionalized alkenes. Journal of Catalysis, 238(1): 134–141, feb 2006. ISSN 00219517. doi: 10.1016/j.jcat.2005.11.042. URL https://linkinghub.elsevier.com/retrieve/pii/S0021951705004719.

[105] Ying Yang, Ying Zhang, Shijie Hao, and Qiubin Kan. Tethering of Cu(II), Co(II) and Fe(III) tetrahydro-salen and salen complexes onto amino-functionalized SBA-15: Eﬀects of salen ligand hydrogenation on catalytic performances for aerobic epoxidation of styrene. Chemical Engineering Journal, 171(3):1356–1366, jul 2011. ISSN 13858947. doi: 10.1016/j.cej.2011.05.047. URL https://linkinghub.elsevier.com/ retrieve/pii/S138589471100605X.

[106] Qinghu Tang, Ye Wang, Jun Liang, Ping Wang, Qinghong Zhang, and Huilin Wan. Co2+-Exchanged faujasite zeolites as eﬃcient heterogeneous catalysts for epoxidation of styrene with molecular oxygen. Chemical Communications, 0(4):440, 2004. ISSN 1359-7345. doi: 10.1039/b314864e. URL http://xlink.rsc.org/?DOI=b314864e.

[107] J SEBASTIAN, K JINKA, and R JASRA. Eﬀect of alkali and alkaline earth metal ions on the catalytic epoxidation of styrene with molecular oxygen using cobalt(II)-exchanged zeolite X. Journal of Catalysis, 244(2):208–218, dec 2006. ISSN 00219517. doi: 10.1016/j. jcat.2006.09.005. URL https://linkinghub.elsevier.com/retrieve/pii/S0021951706003150.

[108] R Kureshy. Dicationic chiral Mn(III) salen complex exchanged in the interlayers of montmorillonite clay: a heterogeneous enantioselective catalyst for epoxidation of nonfunctionalized alkenes. Journal of Catalysis, 221(1):234–240, jan 2004. ISSN 00219517. doi: 10.1016/j.jcat. 2003.07.012. URL https://linkinghub.elsevier.com/retrieve/pii/S0021951703003312.

[109] Takashi Yano, Hideo Matsui, Takahiro Koike, Hiroyasu Ishiguro, Hisashi Fujihara, Masakuni Yoshihara, and Toshihisa Maeshima. Mag- nesium oxide-catalysed reaction of carbon dioxide with an epoxide with retention of stereochemistry. Chemical Communications, 0(12): 1129–1130, 1997. ISSN 13597345. doi: 10.1039/a608102i. URL http://xlink.rsc.org/?DOI=a608102i.

[110] Fabian Jutz, Jan-Dierk Grunwaldt, and Alfons Baiker. Mn(III)(salen)-catalyzed synthesis of cyclic organic carbonates from propylene and styrene oxide in “supercritical” CO2. Journal of Molecular Catalysis A: Chemical, 279(1):94–103, jan 2008. ISSN 13811169. doi: 10. 1016/j.molcata.2007.10.010. URL https://linkinghub.elsevier.com/retrieve/pii/S0021951704004075https://linkinghub.elsevier.com/retrieve/ pii/S1381116907005742.

[111] Ya Du, Jin-Quan Wang, Jian-Yu Chen, Fei Cai, Jie-Sheng Tian, De-Lin Kong, and Liang-Nian He. A poly(ethylene glycol)-supported quaternary ammonium salt for highly eﬃcient and environmentally friendly chemical ﬁxation of CO2 with epoxides under supercritical conditions. Tetrahedron Letters, 47(8):1271–1275, feb 2006. ISSN 00404039. doi: 10.1016/j.tetlet.2005.12.077. URL https://linkinghub. elsevier.com/retrieve/pii/S0040403905027541.

[112] Fabian Jutz, Jan-Dierk Grunwaldt, and Alfons Baiker. Mn(III)(salen)-catalyzed synthesis of cyclic organic carbonates from propylene and styrene oxide in “supercritical” CO2. Journal of Molecular Catalysis A: Chemical, 279(1):94–103, jan 2008. ISSN 13811169. doi: 10.1016/j.molcata.2007.10.010. URL https://linkinghub.elsevier.com/retrieve/pii/S1381116907005742.

74 Bibliography

[113] Jinliang Song, Zhaofu Zhang, Suqin Hu, Tianbin Wu, Tao Jiang, and Buxing Han. MOF-5/n-Bu4NBr: An eﬃcient catalyst system for the synthesis of cyclic carbonates from epoxides and CO2 under mild conditions. Green Chemistry, 11(7):1031–1036, 2009. ISSN 14639262. doi: 10.1039/b902550b.

[114] William Clegg, Ross W. Harrington, Michael North, and Riccardo Pasquale. Cyclic Carbonate Synthesis Catalysed by Bimetallic Aluminium-Salen Complexes. Chemistry - A European Journal, 16(23):6828–6843, apr 2010. ISSN 09476539. doi: 10.1002/chem.201000030. URL http://doi.wiley.com/10.1002/chem.201000030.

[115] Rajendra Srivastava, D. Srinivas, and Paul Ratnasamy. Synthesis of cyclic carbonates from oleﬁns and CO2 over zeolite-based catalysts. Catalysis Letters, 89(1-2):81–85, 2003. ISSN 1011372X. doi: 10.1023/A:1024723627087.

[116] Da-Ae Yang, Hye-Young Cho, Jun Kim, Seung-Tae Yang, and Wha-Seung Ahn. CO2 capture and conversion using Mg-MOF-74 prepared by a sonochemical method. Energy Environ. Sci., 5(4):6465–6473, 2012. ISSN 1754-5692. doi: 10.1039/C1EE02234B. URL http: //xlink.rsc.org/?DOI=C1EE02234B.

[117] Jun Kim, Se-Na Kim, Hoi-Gu Jang, Gon Seo, and Wha-Seung Ahn. CO2 cycloaddition of styrene oxide over MOF catalysts. Applied Catalysis A: General, 453:175–180, feb 2013. ISSN 0926860X. doi: 10.1016/j.apcata.2012.12.018. URL https://linkinghub.elsevier.com/ retrieve/pii/S0926860X12007776.

[118] Onur Buyukcakir, Sang Hyun Je, Siddulu Naidu Talapaneni, Daeok Kim, and Ali Coskun. Charged Covalent Triazine Frameworks for CO2 Capture and Conversion. ACS Applied Materials & Interfaces, 9(8):7209–7216, mar 2017. ISSN 1944-8244. doi: 10.1021/acsami.6b16769. URL http://pubs.acs.org/doi/10.1021/acsami.6b16769.

[119] Jun Liang, Yuan Biao Huang, and Rong Cao. Metal-organic frameworks and porous organic polymers for sustainable ﬁxation of carbon dioxide into cyclic carbonates. Coordination Chemistry Reviews, 378:32–65, 2019. ISSN 00108545. doi: 10.1016/j.ccr.2017.11.013. URL https://doi.org/10.1016/j.ccr.2017.11.013.

[120] Nicolas Eghbali and Chao Jun Li. Conversion of carbon dioxide and oleﬁns into cyclic carbonates in water. Green Chemistry, 9(3):213–215, 2007. ISSN 14639262. doi: 10.1039/b615612f.

[121] Jie Wu, Jennifer A. Kozak, Fritz Simeon, T. Alan Hatton, and Timothy F. Jamison. Mechanism-guided design of ﬂow systems for multicomponent reactions: conversion of CO2 and oleﬁns to cyclic carbonates. Chemical Science, 5(3):1227, 2014. ISSN 2041-6520. doi: 10.1039/c3sc53422g. URL http://xlink.rsc.org/?DOI=c3sc53422g.

[122] Jing-Lun Wang, Jin-quan Wang, Liang-nian He, Xiao-yong Dou, and Fang Wu. A CO2/H2O2-tunable reaction: direct conversion of styrene into styrene carbonate catalyzed by sodium phosphotungstate/n-Bu4NBr. Green Chemistry, 10(11):1218, 2008. ISSN 1463-9262. doi: 10.1039/b807108j. URL http://xlink.rsc.org/?DOI=b807108j.

[123] Stephen G. Davies, Ai M. Fletcher, Wataru Kurosawa, James A. Lee, Giovanna Poce, Paul M. Roberts, James E. Thomson, and David M. Williamson. One-pot conversions of oleﬁns to cyclic carbonates and secondary allylic and homoallylic amines to cyclic carbamates. Journal of Organic Chemistry, 75(22):7745–7756, 2010. ISSN 00223263. doi: 10.1021/jo101614f.

[124] Jia Ning Xie, Zhen Feng Diao, Chang Qiao, Ran Ma, and Liang Nian He. One-pot stepwise synthesis of cyclic carbonates directly from oleﬁns with CO2 promoted by K2S2O8/NaBr. Journal of CO2 Utilization, 16:313–317, 2016. ISSN 22129820. doi: 10.1016/j.jcou.2016.08. 009. URL http://dx.doi.org/10.1016/j.jcou.2016.08.009.

[125] Ajay A. Sathe, Anirudh M.K. Nambiar, and Robert M. Rioux. Synthesis of cyclic organic carbonates via catalytic oxidative carboxylation of oleﬁns in ﬂow reactors. Catalysis Science and Technology, 7(1):84–89, 2017. ISSN 20444761. doi: 10.1039/c6cy01974a. URL http: //dx.doi.org/10.1039/C6CY01974A.

[126] M Aresta, E. Quaranta, and A. Ciccarese. Direct synthesis of 1,3-benzodioxol-2-one from styrene, dioxygen and carbon dioxide promoted by Rh(I). Journal of Molecular Catalysis, 41(3):355–359, aug 1987. ISSN 03045102. doi: 10.1016/0304-5102(87)80112-8. URL http: //linkinghub.elsevier.com/retrieve/pii/0304510287801128.

[127] Michele Aresta, Angela Dibenedetto, and Immacolata Tommasi. Direct synthesis of organic carbonates by oxidative carboxylation of oleﬁns catalyzed by metal oxides: developing green chemistry based on carbon dioxide. Applied Organometallic Chemistry, 14(12):799– 802, dec 2000. ISSN 0268-2605. doi: 10.1002/1099-0739(200012)14:12h799::AID-AOC82i3.0.CO;2-8. URL http://doi.wiley.com/10.1002/ 1099-0739{%}28200012{%}2914{%}3A12{%}3C799{%}3A{%}3AAID-AOC82{%}3E3.0.CO{%}3B2-8.

[128] Michele Aresta and Angela Dibenedetto. Carbon dioxide as building block for the synthesis of organic carbonates. Journal of Molecular Catalysis A: Chemical, 182-183:399–409, may 2002. ISSN 13811169. doi: 10.1016/S1381-1169(01)00514-3. URL https://linkinghub. elsevier.com/retrieve/pii/S1381116901005143.

[129] Jianmin Sun, Shin-ichiro Fujita, Bhalchandra M. Bhanage, and Masahiko Arai. One-pot synthesis of styrene carbonate from styrene in tetrabutylammonium bromide. Catalysis Today, 93-95:383–388, sep 2004. ISSN 09205861. doi: 10.1016/j.cattod.2004.06.105. URL http://linkinghub.elsevier.com/retrieve/pii/S0920586104002597.

75 Bibliography

[130] Jianmin Sun, Shin Ichiro Fujita, Fengyu Zhao, Masashi Hasegawa, and Masahiko Arai. A direct synthesis of styrene carbonate from styrene with the Au/SiO2-ZnBr2/Bu4NBr catalyst system. Journal of Catalysis, 230(2):398–405, 2005. ISSN 00219517. doi: 10.1016/j. jcat.2004.12.015.

[131] Yali Wang, Jianhong Sun, Dan Xiang, Lu Wang, Jianmin Sun, and Feng-Shou Xiao. A Facile, Direct Synthesis of Styrene Carbonate from Styrene and CO2 Catalyzed by Au/Fe(OH)3-ZnBr2/Bu4NBr System. Catalysis Letters, 129(3-4):437–443, may 2009. ISSN 1011-372X. doi: 10.1007/s10562-008-9820-y. URL http://link.springer.com/10.1007/s10562-008-9820-y.

[132] Dongsheng Bai and Huanwang Jing. Aerobic oxidative carboxylation of oleﬁns with metalloporphyrin catalysts. Green Chemistry, 12(1): 39–41, 2010. ISSN 14639262. doi: 10.1039/b916042f.

[133] Catarina Carvalho Rocha, Thomas Onfroy, and Franck Launay. Towards a combined use of Mn(Salen) and quaternary ammonium salts as catalysts for the direct conversion of styrene to styrene carbonate in the presence of dioxygen and carbon dioxide. Comptes Rendus Chimie, 18(3):270–276, 2015. ISSN 16310748. doi: 10.1016/j.crci.2014.12.003. URL http://dx.doi.org/10.1016/j.crci.2014.12.003.

[134] Subodh Kumar, Nikita Singhal, Raj K. Singh, Piyush Gupta, Raghuvir Singh, and Suman L. Jain. Dual catalysis with magnetic chitosan: Direct synthesis of cyclic carbonates from oleﬁns with carbon dioxide using isobutyraldehyde as the sacriﬁcial reductant. Dalton Transactions, 44(26):11860–11866, 2015. ISSN 14779234. doi: 10.1039/c5dt01012h. URL http://dx.doi.org/10.1039/c5dt01012h.

[135] Nataliya V. Maksimchuk, Irina D. Ivanchikova, Artem B. Ayupov, and Oxana A. Kholdeeva. One-step solvent-free synthesis of cyclic carbonates by oxidative carboxylation of styrenes over a recyclable Ti-containing catalyst. Applied Catalysis B: Environmental, 181:363– 370, feb 2016. ISSN 09263373. doi: 10.1016/j.apcatb.2015.08.010. URL http://dx.doi.org/10.1016/j.apcatb.2015.08.010https://linkinghub. elsevier.com/retrieve/pii/S0926337315300837.

[136] Sa Sa Wang and Guo Yu Yang. Recent Advances in Polyoxometalate-Catalyzed Reactions. Chemical Reviews, 115(11):4893–4962, 2015. ISSN 15206890. doi: 10.1021/cr500390v.

[137] Mario G. Clerici and Oxana A. Kholdeeva, editors. Liquid Phase Oxidation via Heterogeneous Catalysis. John Wiley & Sons, Inc., Hoboken, New Jersey, may 2013. ISBN 9781118356760. doi: 10.1002/9781118356760. URL http://doi.wiley.com/10.1002/9781118356760.

[138] R Srivastava, D. Srinivas, and P. Ratnasamy. Synthesis of Polycarbonate Precursors over Titanosilicate Molecular Sieves. Catalysis Letters, 91(1/2):133–139, nov 2003. ISSN 1011-372X. doi: 10.1023/B:CATL.0000006329.37210.fd. URL http://link.springer.com/10.1023/B: CATL.0000006329.37210.fd.

[139] Lucas D. Dias, Rui M.B. Carrilho, CésarA. Henriques, MárioJ.F. Calvete, Anna M. Masdeu-Bultó,Carmen Claver, Liane M. Rossi, and Mariette M. Pereira. Hybrid Metalloporphyrin Magnetic Nanoparticles as Catalysts for Sequential Transformation of Alkenes and CO2 into Cyclic Carbonates. ChemCatChem, 10(13):2792–2803, 2018. ISSN 18673899. doi: 10.1002/cctc.201800397.

[140] Fumitaka Ono, Kun Qiao, Daisuke Tomida, and Chiaki Yokoyama. Direct preparation of styrene carbonates from styrene using an ionic- liquid-based one-pot multistep synthetic process. Applied Catalysis A: General, 333(1):107–113, 2007. ISSN 0926860X. doi: 10.1016/j. apcata.2007.09.015.

[141] Xiaofang Gao, Gaoqing Yuan, Huoji Chen, Huanfeng Jiang, Yingwei Li, and Chaorong Qi. Eﬃcient conversion of CO2 with oleﬁns into cyclic carbonates via a synergistic action of I2 and base electrochemically generated in situ. Electrochemistry Communications, 34:242–245, sep 2013. ISSN 13882481. doi: 10.1016/j.elecom.2013.06.022. URL http://dx.doi.org/10.1016/j.elecom.2013.06.022https: //linkinghub.elsevier.com/retrieve/pii/S1388248113002543.

[142] Olga V. Zalomaeva, Nataliya V. Maksimchuk, Andrey M. Chibiryaev, Konstantin A. Kovalenko, Vladimir P. Fedin, and Bair S. Balzhin- imaev. Synthesis of cyclic carbonates from epoxides or oleﬁns and CO2 catalyzed by metal-organic frameworks and quaternary ammonium salts. Journal of Energy Chemistry, 22(1):130–135, jan 2013. ISSN 20954956. doi: 10.1016/S2095-4956(13)60017-0. URL http://dx.doi.org/10.1016/S2095-4956(13)60017-0https://linkinghub.elsevier.com/retrieve/pii/S2095495613600170.

[143] Phuong T. K. Nguyen, Huong T. D. Nguyen, Hung N. Nguyen, Christopher A. Trickett, Quang T. Ton, Enrique Guti´errez-Puebla, M. Angeles Monge, Kyle E. Cordova, and Felipe G´andara.New Metal-Organic Frameworks for Chemical Fixation of CO2. ACS Applied Materials & Interfaces, 10(1):733–744, jan 2018. ISSN 1944-8244. doi: 10.1021/acsami.7b16163. URL http://pubs.acs.org/doi/10.1021/ acsami.7b16163.

[144] Michael Thor Pope. Heteropoly and Isopoly Oxometalates, volume 8 of Inorganic Chemistry Concepts. Springer Berlin Heidelberg, Berlin, Heidelberg, sep 1983. ISBN 978-3-662-12006-4. doi: 10.1007/978-3-662-12004-0. URL http://link.springer.com/10.1007/978-3-662-12004-0.

[145] JönsJacob Berzelius. Beitrag zur näherenKenntniss des Molybdäns. Annalen der Physik und Chemie, 82(4):369–392, 1826. ISSN 00033804. doi: 10.1002/andp.18260820402. URL http://doi.wiley.com/10.1002/andp.18260820402.

[146] Jean Charles Galissard de Marignac. Recherches sur les acides silicotungstiques et note sur la constitution de l’acide tungstique. Gauthier- Villars, 1864.

76 Bibliography

[147] Leopold Gmelin. System No. 54 - Wolfram. In Gmelin Handbuch der Anorganischen Chemie. Verlag Chemie, Berlin, 1933.

[148] Leopold Gmelin. System No. 53 - Molybdan. In Gmelin Handbuch der Anorganischen Chemie. Verlag Chemie, Berlin, 1935.

[149] J. F. KEGGIN. Structure of the Molecule of 12-Phosphotungstic Acid. Nature, 131(3321):908–909, jun 1933. ISSN 0028-0836. doi: 10.1038/131908b0. URL http://www.nature.com/articles/131908b0.

[150] Makoto Misono and Naohiro Nojiri. Recent progress in catalytic technology in japan. Applied Catalysis, 64(C):1–30, sep 1990. ISSN 01669834. doi: 10.1016/S0166-9834(00)81550-X. URL https://linkinghub.elsevier.com/retrieve/pii/S016698340081550X.

[151] Michael T. Pope and Achim M¨uller.Polyoxometalate Chemistry: An Old Field with New Dimensions in Several Disciplines. Angewandte Chemie International Edition in English, 30(1):34–48, jan 1991. ISSN 0570-0833. doi: 10.1002/anie.199100341. URL http://doi.wiley.com/ 10.1002/anie.199100341.

[152] Craig L Hill and Christina M. Prosser-McCartha. Homogeneous catalysis by transition metal oxygen anion clusters. Coordination Chemistry Reviews, 143(0):407–455, sep 1995. ISSN 00108545. doi: 10.1016/0010-8545(95)01141-B. URL https://linkinghub.elsevier.com/retrieve/ pii/001085459501141B.

[153] C. L. Hill. Introduction: Polyoxometalates Multicomponent Molecular Vehicles To Probe Fundamental Issues and Practical Problems. Chemical Reviews, 98(1):1–2, feb 1998. ISSN 0009-2665. doi: 10.1021/cr960395y. URL https://pubs.acs.org/doi/10.1021/cr960395y.

[154] Masahiro Sadakane and Eberhard Steckhan. Electrochemical Properties of Polyoxometalates as Electrocatalysts. Chemical Reviews, 98 (1):219–238, feb 1998. ISSN 0009-2665. doi: 10.1021/cr960403a. URL https://pubs.acs.org/doi/10.1021/cr960403a.

[155] Achim M¨uller,Frank Peters, Michael T Pope, and Dante Gatteschi. Polyoxometalates: Very Large Clusters-Nanoscale Magnets. Chemical Reviews, 98(1):239–272, feb 1998. ISSN 0009-2665. doi: 10.1021/cr9603946. URL https://pubs.acs.org/doi/10.1021/cr9603946.

[156] De-liang Long, Eric Burkholder, and Leroy Cronin. Polyoxometalate clusters, nanostructures and materials: From self assembly to designer materials and devices. Chem. Soc. Rev., 36(1):105–121, 2007. ISSN 0306-0012. doi: 10.1039/B502666K. URL http://xlink.rsc. org/?DOI=B502666K.

[157] De-liang Long, Ryo Tsunashima, and Leroy Cronin. Polyoxometalates: Building Blocks for Functional Nanoscale Systems. Angewandte Chemie International Edition, 49(10):1736–1758, mar 2010. ISSN 14337851. doi: 10.1002/anie.200902483. URL http://doi.wiley.com/10. 1002/anie.200902483.

[158] Miao Sun, Jizhe Zhang, Piotr Putaj, Valerie Caps, Fr´ed´ericLefebvre, Jeremie Pelletier, and Jean-Marie Basset. Catalytic Oxidation of Light Alkanes (C1-C4) by Heteropoly Compounds. Chemical Reviews, 114(2):981–1019, jan 2014. ISSN 0009-2665. doi: 10.1021/cr300302b. URL http://pubs.acs.org/doi/10.1021/cr300302b.

[159] Alfred Werner. Neuere Anschauungen auf dem Gebiete der anorganischen Chemie. Friedr. Vieweg & Sohn, Braunschweig, 1920.

[160] A. Miolati and R. Pizzighelli. Zur Kenntnis der komplexen SäurenI. 1. Uber¨ die Leitfähigkeit von molybdänsäurehaltigenGemischen. Journal fürPraktische Chemie, 77(1):417–456, apr 1908. ISSN 00218383. doi: 10.1002/prac.19080770135. URL http://doi.wiley.com/10. 1002/prac.19080770135.

[161] Arthur Rosenheim and Johannes Jaenicke. Zur Kenntnis der Iso- und Heteropolysäuren.XV. Mitteilung Uber¨ Heteropolywolframate und einige Heteropolymolybdänate. Zeitschrift füranorganische und allgemeine Chemie, 101(1):235–275, nov 1917. ISSN 08631786. doi: 10.1002/zaac.19171010106. URL http://doi.wiley.com/10.1002/zaac.19171010106.

[162] Linus Pauling. THE MOLECULAR STRUCTURE OF THE TUNGSTOSILICATES AND RELATED COMPOUNDS. Journal of the American Chemical Society, 51(10):2868–2880, oct 1929. ISSN 0002-7863. doi: 10.1021/ja01385a002. URL http://pubs.acs.org/doi/abs/10. 1021/ja01385a002.

[163] Benjah-bmm27. Phosphotungstate-3D-polyhedra.png, 2007. URL https://commons.wikimedia.org/wiki/File:Phosphotungstate-3D-polyhedra. png.

[164] I. Lindqvist. On the structure of the paratungstate ion. Acta Crystallographica, 5(5):667–670, sep 1952. ISSN 0365110X. doi: 10.1107/ S0365110X52001817. URL http://scripts.iucr.org/cgi-bin/paper?S0365110X52001817.

[165] Alexander Frank Wells. Structural Inorganic Chemistry. Clarendon Press, Oxford, 1945.

[166] B. Dawson. The structure of the 9(18)-heteropoly anion in potassium 9(18)-tungstophosphate, K6(P2W18O62).14H2O. Acta Crys- tallographica, 6(2):113–126, feb 1953. ISSN 0365-110X. doi: 10.1107/S0365110X53000466. URL http://scripts.iucr.org/cgi-bin/paper? S0365110X53000466.

[167] J. STUART ANDERSON. Constitution of the Poly-acids. Nature, 140(3550):850–850, nov 1937. ISSN 0028-0836. doi: 10.1038/140850a0. URL http://www.nature.com/articles/140850a0.

77 Bibliography

[168] Howard T. Evans. THE CRYSTAL STRUCTURES OF AMMONIUM AND POTASSIUM MOLYBDOTELLURATES. Journal of the American Chemical Society, 70(3):1291–1292, mar 1948. ISSN 0002-7863. doi: 10.1021/ja01183a521. URL http://pubs.acs.org/doi/abs/10. 1021/ja01183a521.

[169] David D. Dexter and J. V. Silverton. A new structural type for heteropoly anions. The crystal structure of (NH4)2H6(CeMo12O42).12H2O. Journal of the American Chemical Society, 90(13):3589–3590, jun 1968. ISSN 0002-7863. doi: 10.1021/ja01015a067. URL http://pubs.acs. org/doi/abs/10.1021/ja01015a067http://xlink.rsc.org/?DOI=DT9740002021.

[170] John Iball, John N. Low, and Timothy J. R. Weakley. Heteropolytungstate Complexes of the Lanthanoid Elements. Part III. Crystal Structure of Sodium Decatungstocerate(IV)-water (1/30). J. Chem. Soc., Dalton Trans., 0(18):2021–2024, 1974. ISSN 0300-9246. doi: 10.1039/DT9740002021. URL http://xlink.rsc.org/?DOI=DT9740002021.

[171] T. Ozeki, M. Takahashi, and T. Yamase. Structure of K3Na4H2[TbW10O36].20H2O. Acta Crystallographica Section C Crystal Structure Communications, 48(8):1370–1374, aug 1992. ISSN 01082701. doi: 10.1107/S0108270192000155. URL http://scripts.iucr.org/cgi-bin/ paper?S0108270192000155.

[172] J. L. T. Waugh, D. P. Shoemaker, and L. Pauling. On the structure of the heteropoly anion in ammonium 9-molybdomanganate, (NH4)6MnMo9O32.82O. Acta Crystallographica, 7(5):438–441, may 1954. ISSN 0365110X. doi: 10.1524/zkri.1975.141.16.342. URL http://scripts.iucr.org/cgi-bin/paper?S0365110X54001302.

[173] R. ALLMANN and H. D’AMOUR. Strukturverfeinerung von Ammonium-9-molybdomanganat, (NH4)6[MnMo9O32]·nH2O (n =6-8). Zeitschrift f¨urKristallographie - Crystalline Materials, 141(1-6), jan 1975. ISSN 2196-7105. doi: 10.1524/zkri.1975.141.16.342. URL http://www.degruyter.com/view/j/zkri.1975.141.issue-1-6/zkri.1975.141.16.342/zkri.1975.141.16.342.xml.

[174] Yves P. Jeannin. The Nomenclature of Polyoxometalates: How To Connect a Name and a Structure. Chemical Reviews, 98(1):51–76, feb 1998. ISSN 0009-2665. doi: 10.1021/cr960397i. URL https://pubs.acs.org/doi/10.1021/cr960397i.

[175] Alvin P. Ginsberg, editor. Inorganic Syntheses Volume 27. Inorganic Syntheses. John Wiley & Sons, Inc., Hoboken, NJ, USA, jan 1990. ISBN 9780470132586. doi: 10.1002/9780470132586. URL http://doi.wiley.com/10.1002/9780470132586.

[176] Noritaka Mizuno and Makoto Misono. Heterogeneous Catalysis. Chemical Reviews, 98(1):199–218, feb 1998. ISSN 0009-2665. doi: 10.1021/cr960401q. URL https://pubs.acs.org/doi/10.1021/cr960401q.

[177] Carlo Venturello, Rino D’Aloisio, Jan C.J. Bart, and Marco Ricci. A New peroxotungsten heteropoly anion with special oxidizing properties: synthesis and structure of tetrahexylammonium tetra(diperoxotungsto)phosphate(3-). Journal of Molecular Catalysis, 32(1):107–110, aug 1985. ISSN 03045102. doi: 10.1016/0304-5102(85)85037-9. URL https://linkinghub.elsevier.com/retrieve/pii/0304510285850379.

[178] Carlo Venturello and Mario Gambaro. A Convenient Catalytic Method for the Dihydroxylation of Alkenes by Hydrogen Peroxide. Synthesis Communications, 1989(04):295–297, 1989. ISSN 0039-7881. doi: 10.1055/s-1989-27229. URL http://www.thieme-connect.de/DOI/DOI?10.1055/ s-1989-27229.

[179] Xi Zuwei. Reaction-Controlled Phase-Transfer Catalysis for Propylene Epoxidation to Propylene Oxide. Science, 292(5519):1139–1141, may 2001. ISSN 00368075. doi: 10.1126/science.292.5519.1139. URL http://www.sciencemag.org/cgi/doi/10.1126/science.292.5519.1139.

[180] Arnold Lambert, Pawel Plucinski, and Ivan V. Kozhevnikov. Polyoxometalate-catalysed epoxidation of 1-octene with hydrogen peroxide in microemulsions coupled with ultraﬁltration. Chemical Communications, 3(6):714–715, mar 2003. ISSN 13597345. doi: 10.1039/b300494p. URL http://xlink.rsc.org/?DOI=b300494p.

[181] Yan Leng, Jun Wang, Dunru Zhu, Mingjue Zhang, Pingping Zhao, Zhouyang Long, and Jun Huang. Polyoxometalate-based amino- functionalized ionic solid catalysts lead to highly eﬃcient heterogeneous epoxidation of alkenes with H2O2. Green Chemistry, 13(7): 1636–1639, 2011. ISSN 14639262. doi: 10.1039/c1gc15302a.

[182] Evelyne Poli, Rodolphe De Sousa, Fran¸coisJ´erome,Yannick Pouilloux, and Jean Marc Clacens. Catalytic epoxidation of styrene and methyl oleate over peroxophosphotungstate entrapped in mesoporous SBA-15. Catalysis Science and Technology, 2(5):910–914, 2012. ISSN 20444753. doi: 10.1039/c2cy20082a.

[183] C. M. Flynn and Michael T. Pope. 1:13 Heteropolyvanadates of Manganese (IV) and Nickel (IV). Journal of the American Chemical Society, 92(1):85–90, 1970. ISSN 15205126. doi: 10.1021/ja00704a014.

[184] Kenji Nagai, Hikaru Ichida, and Yukiyoshi Sasaki. The Structure of Heptapotassium Tridecavanadomanganate (IV) Octadecahydrate, K7[MnV13O38]·18H2O. Chemistry Letters, 15(8):1267–1270, aug 1986. ISSN 0366-7022. doi: 10.1246/cl.1986.1267. URL http://www. journal.csj.jp/doi/10.1246/cl.1986.1267.

[185] Wen-liang LI, Er-fu NI, Xin-hai LI, and Hua-jun GUO. Polyoxovanadate (NH4)7[MnV13O38] as cathode material for lithium ion battery and improved electrochemical performance. Transactions of Nonferrous Metals Society of China, 26(9):2372–2379, sep 2016. ISSN 10036326. doi: 10.1016/S1003-6326(16)64375-3. URL https://linkinghub.elsevier.com/retrieve/pii/S1003632616643753.

78 Bibliography

[186] Jia-Xuan Wang, Yi-Di Wang, Mei-Jie Wei, Hua-Qiao Tan, Yong-Hui Wang, Hong-Ying Zang, and Yang-Guang Li. Inorganic open framework based on lanthanide ions and polyoxometalates with high proton conductivity. Inorganic Chemistry Frontiers, 5(5):1213–1217, 2018. ISSN 2052-1553. doi: 10.1039/C8QI00108A. URL http://xlink.rsc.org/?DOI=C8QI00108A.

[187] Shuxia Liu, Dehui Li, Linhua Xie, Haiyan Cheng, Xinyu Zhao, and Zhongmin Su. Two-dimensional lanthanide heteropolyvanadates of manganese(IV) and nickel(IV) containing two types of heteropoly anions with 1:13 and 1:12 stoichiometry. Inorganic Chemistry, 45(20): 8036–8040, 2006. ISSN 00201669. doi: 10.1021/ic060623z.

[188] Dehui Li, Shuxia Liu, Chunyan Sun, Linhua Xie, Enbo Wang, Ninghai Hu, and Hengqing Jia. A novel 2D layered network based on 13-vanadomanganate(IV): K3(HABOB)4[MnV13O38]·9H2O (ABOB=N-amidino-4-morpholincarboxamidine). Inorganic Chemistry Com- munications, 8(5):433–436, may 2005. ISSN 13877003. doi: 10.1016/j.inoche.2005.02.002. URL https://linkinghub.elsevier.com/retrieve/ pii/S1387700305000481.

[189] Qing Lan, Huaqiao Tan, Ding Liu, and Enbo Wang. A series of inorganic aggregates composed of [MnV13O38]7- polyoxoanions and transition metal cations. Journal of Solid State Chemistry, 199:129–133, 2013. ISSN 00224596. doi: 10.1016/j.jssc.2012.12.005. URL http://dx.doi.org/10.1016/j.jssc.2012.12.005.

[190] Ding Liu, Ying Lu, Hua-Qiao Tan, Ting-Ting Wang, and En-Bo Wang. Series of Organic-Inorganic Hybrid Rare Earth Derivatives Based on [MnV13O38]7- Polyoxoanion: Syntheses, Structures, and Magnetic and Electrochemical Properties. Crystal Growth & Design, 15(1): 103–114, jan 2015. ISSN 1528-7483. doi: 10.1021/cg500955d. URL http://pubs.acs.org/doi/10.1021/cg500955d.

[191] Jens Weber, Markus Antonietti, and Arne Thomas. Mesoporous Poly(benzimidazole) Networks via Solvent Mediated Templating of Hard Spheres. Macromolecules, 40(4):1299–1304, feb 2007. ISSN 0024-9297. doi: 10.1021/ma062598m. URL https://pubs.acs.org/doi/10.1021/ ma062598m.

[192] Arne Thomas, Frederic Goettmann, and Markus Antonietti. Hard Templates for Soft Materials: Creating Nanostructured Organic Materials. Chemistry of Materials, 20(3):738–755, feb 2008. ISSN 0897-4756. doi: 10.1021/cm702126j. URL https://pubs.acs.org/doi/10. 1021/cm702126j.

[193] Christian S. Diercks and Omar M. Yaghi. The atom, the molecule, and the covalent organic framework. Science, 355(6328):eaal1585, mar 2017. ISSN 0036-8075. doi: 10.1126/science.aal1585. URL http://www.sciencemag.org/lookup/doi/10.1126/science.aal1585.

[194] Omar M. Yaghi, Michael O’Keeﬀe, Nathan W. Ockwig, Hee K. Chae, Mohamed Eddaoudi, and Jaheon Kim. Reticular synthesis and the design of new materials. Nature, 423(6941):705–714, 2003. ISSN 00280836. doi: 10.1038/nature01650.

[195] O. M. Yaghi, Guangming Li, and Hailian Li. Selective binding and removal of guests in a microporous metal-organic framework. Nature, 378(6558):703–706, dec 1995. ISSN 0028-0836. doi: 10.1038/378703a0. URL http://www.nature.com/articles/378703a0.

[196] Adrien P. Côté,Annabelle I. Benin, Nathan W. Ockwig, Michael O’Keeffe, Adam J. Matzger, and Omar M. Yaghi. Porous, Crystalline, Covalent Organic Frameworks. Science, 310(November):1166–1171, 2005.

[197] Michael O’Keeﬀe and Omar M. Yaghi. Deconstructing the Crystal Structures of Metal-Organic Frameworks and Related Materials into Their Underlying Nets. Chemical Reviews, 112(2):675–702, feb 2012. ISSN 0009-2665. doi: 10.1021/cr200205j. URL http://pubs.acs.org/ doi/10.1021/cr200205j.

[198] H. M. El-Kaderi, J. R. Hunt, J. L. Mendoza-Cortes, A. P. Cote, R. E. Taylor, M. O’Keeﬀe, and O. M. Yaghi. Designed Synthesis of 3D Covalent Organic Frameworks. Science, 316(5822):268–272, apr 2007. ISSN 0036-8075. doi: 10.1126/science.1139915. URL http://www.sciencemag.org/cgi/doi/10.1126/science.1139915.

[199] R. William Tilford, Sam J. Mugavero, Perry J. Pellechia, and John J. Lavigne. Tailoring Microporosity in Covalent Organic Frameworks. Advanced Materials, 20(14):2741–2746, jul 2008. ISSN 09359648. doi: 10.1002/adma.200800030. URL http://doi.wiley.com/10.1002/adma. 200800030.

[200] Atsushi Nagai, Zhaoqi Guo, Xiao Feng, Shangbin Jin, Xiong Chen, Xuesong Ding, and Donglin Jiang. Pore surface engineering in covalent organic frameworks. Nature Communications, 2(1):536, sep 2011. ISSN 2041-1723. doi: 10.1038/ncomms1542. URL http: //dx.doi.org/10.1038/ncomms1542http://www.nature.com/articles/ncomms1542.

[201] Hartmuth C. Kolb, M. G. Finn, and K. Barry Sharpless. Click Chemistry: Diverse Chemical Function from a Few Good Reactions. Angewandte Chemie International Edition, 40(11):2004–2021, jun 2001. ISSN 1433-7851. doi: 10.1002/1521-3773(20010601)40:11h2004:: AID-ANIE2004i3.0.CO;2-5. URL http://doi.wiley.com/10.1002/1521-3773{%}2820010601{%}2940{%}3A11{%}3C2004{%}3A{%}3AAID-ANIE2004{%}3E3. 0.CO{%}3B2-5.

[202] John E. Moses and Adam D. Moorhouse. The growing applications of click chemistry. Chem. Soc. Rev., 36(8):1249–1262, feb 2007. ISSN 0306-0012. doi: 10.1039/B613014N. URL http://xlink.rsc.org/?DOI=B613014N.

[203] Charles E. Hoyle and Christopher N. Bowman. Thiol-ene click chemistry. Angewandte Chemie - International Edition, 49(9):1540–1573, 2010. ISSN 14337851. doi: 10.1002/anie.200903924.

79 Bibliography

[204] Pierre Kuhn, Markus Antonietti, and Arne Thomas. Porous, Covalent Triazine-Based Frameworks Prepared by Ionothermal Synthesis. Angewandte Chemie International Edition, 47(18):3450–3453, apr 2008. ISSN 14337851. doi: 10.1002/anie.200705710. URL http://doi. wiley.com/10.1002/anie.200705710.

[205] Maryadele J O’Neil. The Merck index : an encyclopedia of chemicals, drugs, and biologicals. Whitehouse Station (N.J.) : Merck, 13th ed. edition, 2001. ISBN 0911910131. URL http://lib.ugent.be/catalog/rug01:000483506.

[206] Pierre Kuhn, Arne Thomas, and Markus Antonietti. Toward tailorable porous organic polymer networks: A high-temperature dynamic polymerization scheme based on aromatic nitriles. Macromolecules, 42(1):319–326, 2009. ISSN 00249297. doi: 10.1021/ma802322j.

[207] Wei Zhao, Lieyin Xia, and Xikui Liu. Covalent organic frameworks (COFs): perspectives of industrialization. CrystEngComm, 20(12):1613– 1634, 2018. ISSN 1466-8033. doi: 10.1039/C7CE02079A. URL http://dx.doi.org/10.1039/C7CE02079Ahttp://xlink.rsc.org/?DOI=C7CE02079A.

[208] Dmitrii Yu Osadchii, Alma I. Olivos-Suarez, Anastasiya V. Bavykina, and Jorge Gascon. Revisiting Nitrogen Species in Covalent Triazine Frameworks. Langmuir, 33(50):14278–14285, dec 2017. ISSN 0743-7463. doi: 10.1021/acs.langmuir.7b02929. URL http://pubs.acs.org/ doi/10.1021/acs.langmuir.7b02929.

[209] Zonglong Li, Hui Li, Xinyu Guan, Junjie Tang, Yusran Yusran, Zhan Li, Ming Xue, Qianrong Fang, Yushan Yan, Valentin Valtchev, and Shilun Qiu. Three-Dimensional Ionic Covalent Organic Frameworks for Rapid, Reversible, and Selective Ion Exchange. Journal of the American Chemical Society, 139(49):17771–17774, dec 2017. ISSN 0002-7863. doi: 10.1021/jacs.7b11283. URL http://pubs.acs.org/doi/10. 1021/jacs.7b11283.

[210] Ning Huang, Ping Wang, Matthew A. Addicoat, Thomas Heine, and Donglin Jiang. Ionic Covalent Organic Frameworks: Design of a Charged Interface Aligned on 1D Channel Walls and Its Unusual Electrostatic Functions. Angewandte Chemie International Edition, 56 (18):4982–4986, apr 2017. ISSN 14337851. doi: 10.1002/anie.201611542. URL http://doi.wiley.com/10.1002/anie.201611542.

[211] Kwangho Park, Kwangyeol Lee, Hyunuk Kim, Vinothkumar Ganesan, Kanghee Cho, Soon Kwan Jeong, and Sungho Yoon. Preparation of covalent triazine frameworks with imidazolium cations embedded in basic sites and their application for CO2 capture. Journal of Materials Chemistry A, 5(18):8576–8582, 2017. ISSN 20507496. doi: 10.1039/c6ta11226a.

[212] Tao Tao Liu, Rui Xu, Jun Dong Yi, Jun Liang, Xu Sheng Wang, Peng Chao Shi, Yuan Biao Huang, and Rong Cao. Imidazolium-Based Cationic Covalent Triazine Frameworks for Highly Eﬃcient Cycloaddition of Carbon Dioxide. ChemCatChem, 10(9):2036–2040, 2018. ISSN 18673899. doi: 10.1002/cctc.201800023.

[213] Manying Liu, Liping Guo, Shangbin Jin, and Bien Tan. Covalent triazine frameworks: synthesis and applications. Journal of Materials Chemistry A, 7(10):5153–5172, 2019. ISSN 2050-7488. doi: 10.1039/C8TA12442F. URL http://xlink.rsc.org/?DOI=C8TA12442F.

[214] C. V. Boys. On the Production, Properties, and some suggested Uses of the Finest Threads. Proceedings of the Physical Society of London, 9(1):8–19, apr 1887. ISSN 1478-7814. doi: 10.1088/1478-7814/9/1/303. URL http://stacks.iop.org/1478-7814/9/i=1/a=303?key=crossref. f728595d1eccb701fa6ae85b20e6f2d3.

[215] J. F. Cooley. US705691A - Apparatus for Electrically Dispersing Fluids., 1902.

[216] W. J. Morton. US705691A - Method of Dispersing Fluids., 1902.

[217] A Formhals. US1975504A - Process and Apparatus for Preparing Artiﬁcial Threads, 1934.

[218] A. Formhals. US2349950A - Method and Apparatus for Spinning, 1944.

[219] On the 100th anniversary of the birth of I.V. Petryanov-Sokolov -by colleagues and students-. Izvestiya, Atmospheric and Oceanic Physics, 43(3):395–395, jun 2007. ISSN 0001-4338. doi: 10.1134/S0001433807030164. URL http://link.springer.com/10.1134/S0001433807030164.

[220] Dan Li and Younan Xia. Electrospinning of Nanoﬁbers: Reinventing the Wheel? Advanced Materials, 16(14):1151–1170, jul 2004. ISSN 0935-9648. doi: 10.1002/adma.200400719. URL http://doi.wiley.com/10.1002/adma.200400719.

[221] Zheng Ming Huang, Y. Z. Zhang, M. Kotaki, and S. Ramakrishna. A review on polymer nanoﬁbers by electrospinning and their applications in nanocomposites. Composites Science and Technology, 63(15):2223–2253, 2003. ISSN 02663538. doi: 10.1016/S0266-3538(03)00178-7.

[222] Andreas Greiner and Joachim H. Wendorﬀ. Electrospinning: A Fascinating Method for the Preparation of Ultrathin Fibers. Angewandte Chemie International Edition, 46(30):5670–5703, jul 2007. ISSN 14337851. doi: 10.1002/anie.200604646. URL http://doi.wiley.com/10. 1002/anie.200604646.

[223] Daniel Crespy, Kathrin Friedemann, and Ana-maria Popa. Colloid-Electrospinning: Fabrication of Multicompartment Nanoﬁbers by the Electrospinning of Organic or/and Inorganic Dispersions and Emulsions. Macromolecular Rapid Communications, 33(23):1978–1995, dec 2012. ISSN 10221336. doi: 10.1002/marc.201200549. URL http://doi.wiley.com/10.1002/marc.201200549.

80 Bibliography

[224] Yi-nan Wu, Fengting Li, Huimin Liu, Wei Zhu, Minmin Teng, Yin Jiang, Weina Li, Dan Xu, Dehua He, Phillip Hannam, and Guangtao Li. Electrospun ﬁbrous mats as skeletons to produce free-standing MOF membranes. Journal of Materials Chemistry, 22(33):16971, 2012. ISSN 0959-9428. doi: 10.1039/c2jm32570e. URL http://xlink.rsc.org/?DOI=c2jm32570e.

[225] Xiaowei Wu, Xing Han, Qisong Xu, Yuhao Liu, Chen Yuan, Shi Yang, Yan Liu, Jianwen Jiang, and Yong Cui. Chiral BINOL-Based Covalent Organic Frameworks for Enantioselective Sensing. Journal of the American Chemical Society, 141(17):7081–7089, may 2019. ISSN 0002-7863. doi: 10.1021/jacs.9b02153. URL http://pubs.acs.org/doi/10.1021/jacs.9b02153.

[226] Zhiming Yan, Biqing Hu, Qianlian Li, Sunxian Zhang, Jie Pang, and Chunhua Wu. Facile synthesis of covalent organic framework incorporated electrospun nanoﬁber and application to pipette tip solid phase extraction of sulfonamides in meat samples. Journal of Chromatography A, 1584:33–41, jan 2019. ISSN 00219673. doi: 10.1016/j.chroma.2018.11.039. URL https://doi.org/10.1016/j.chroma.2018. 11.039https://linkinghub.elsevier.com/retrieve/pii/S0021967318314389.

[227] Michel M. Bergshoef and G. Julius Vancso. Transparent Nanocomposites with Ultrathin, Electrospun Nylon-4,6 Fiber Reinforcement. Advanced Materials, 11(16):1362–1365, nov 1999. ISSN 0935-9648. doi: 10.1002/(SICI)1521-4095(199911)11:16h1362::AID-ADMA1362i 3.0.CO;2-X. URL http://doi.wiley.com/10.1002/{%}28SICI{%}291521-4095{%}28199911{%}2911{%}3A16{%}3C1362{%}3A{%}3AAID-ADMA1362{%}3E3.0. CO{%}3B2-X.

[228] Herman F. Mark, Norbert M. Bikales, Charles G. Overberger, George Menges, and Jacqueline I. Kroschwitz, editors. Encyclopedia of Polymer Science and Engineering, Vol. 11. Wiley, New York City, 2nd edition, 1985. ISBN 978-0471865193.

[229] Bert De Schoenmaker, Lien Van der Schueren, Ruphino Zugle, Annelies Goethals, Philippe Westbroek, Paul Kiekens, Tebello Nyokong, and Karen De Clerck. Effect of the relative humidity on the fibre morphology of polyamide 4.6 and polyamide 6.9 nanofibres. Journal of Materials Science, 48(4):1746–1754, feb 2013. ISSN 0022-2461. doi: 10.1007/s10853-012-6934-9. URL http://link.springer.com/10.1007/ s10853-012-6934-9.

[230] Iline Steyaert, Marie-Paule Delplancke, Guy Van Assche, Hubert Rahier, and Karen De Clerck. Fast-scanning calorimetry of electrospun polyamide nanoﬁbres: Melting behaviour and crystal structure. Polymer, 54(25):6809–6817, nov 2013. ISSN 00323861. doi: 10.1016/j. polymer.2013.10.032. URL http://dx.doi.org/10.1016/j.polymer.2013.10.032https://linkinghub.elsevier.com/retrieve/pii/S0032386113009865.

[231] Bert De Schoenmaker, Lien Van der Schueren, Ozg¨urCeylan,¨ and Karen De Clerck. Electrospun Polyamide 4.6 Nanoﬁbrous Nonwovens: Parameter Study and Characterization. Journal of Nanomaterials, 2012:1–9, 2012. ISSN 1687-4110. doi: 10.1155/2012/860654. URL http://www.hindawi.com/journals/jnm/2012/860654/.

[232] Jack C C Chen. Palladium complexes containing a hemilabile pyridylcarbene ligand. Organometallics, 19(24):5113–5121, 2000. ISSN 02767333. doi: 10.1021/om000557n.

[233] Heping Ma, Bailing Liu, Bin Li, Liming Zhang, Yang Guang Li, Hua Qiao Tan, Hong Ying Zang, and Guangshan Zhu. Cationic Covalent Organic Frameworks: A Simple Platform of Anionic Exchange for Porosity Tuning and Proton Conduction - Supporting Information. Journal of the American Chemical Society, 138(18):5897–5903, may 2016. ISSN 15205126. doi: 10.1021/jacs.5b13490. URL http://pubs. acs.org/doi/10.1021/jacs.5b13490.

[234] Karen Leus, Ilke Muylaert, Matthias Vandichel, Guy B. Marin, Michel Waroquier, Veronique Van Speybroeck, and Pascal Van Der Voort. The remarkable catalytic activity of the saturated metal organic framework V-MIL-47 in the cyclohexene oxidation. Chemical Communications, 46(28):5085, 2010. ISSN 1359-7345. doi: 10.1039/c0cc01506g. URL http://xlink.rsc.org/?DOI=c0cc01506g.

[235] Karen Leus, Matthias Vandichel, Ying-Ya Liu, Ilke Muylaert, Jan Musschoot, Steven Pyl, Henk Vrielinck, Freddy Callens, Guy B. Marin, Christophe Detavernier, Paul V. Wiper, Yaroslav Z. Khimyak, Michel Waroquier, Veronique Van Speybroeck, and Pascal Van Der Voort. The coordinatively saturated vanadium MIL-47 as a low leaching heterogeneous catalyst in the oxidation of cyclohexene. Journal of Catalysis, 285(1):196–207, jan 2012. ISSN 00219517. doi: 10.1016/j.jcat.2011.09.014. URL http://dx.doi.org/10.1016/j.jcat.2011.09. 014http://linkinghub.elsevier.com/retrieve/pii/S0021951711002934.

[236] Caroline A Schneider, Wayne S Rasband, and Kevin W Eliceiri. NIH Image to ImageJ: 25 years of image analysis. Nature Methods, 9(7): 671–675, jul 2012. ISSN 1548-7091. doi: 10.1038/nmeth.2089. URL http://www.nature.com/articles/nmeth.2089.

[237] Pierre Kuhn, Aurelien Forget, Dangsheng Su, Arne Thomas, and Markus Antonietti. From Microporous Regular Frameworks to Meso- porous Materials with Ultrahigh Surface Area: Dynamic Reorganization of Porous Polymer Networks. Journal of the American Chemical Society, 130(40):13333–13337, oct 2008. ISSN 0002-7863. doi: 10.1021/ja803708s. URL https://pubs.acs.org/doi/10.1021/ja803708s.

[238] Stephan Hug, Michael E. Tauchert, Shen Li, Ursula E. Pachmayr, and Bettina V. Lotsch. A functional triazine framework based on N-heterocyclic building blocks. Journal of Materials Chemistry, 22(28):13956, 2012. ISSN 0959-9428. doi: 10.1039/c2jm31248d. URL http://xlink.rsc.org/?DOI=c2jm31248d.

[239] Steven R. Givens, Kenncorwin H. Gardner, John F. Rabolt, and D. Bruce Chase. High-Temperature Electrospinning of Polyethylene Microﬁbers from Solution. Macromolecules, 40(3):608–610, feb 2007. ISSN 0024-9297. doi: 10.1021/ma062398a. URL https://pubs.acs. org/doi/10.1021/ma062398a.

81 Bibliography

[240] D. M. Rein, L. Shavit-Hadar, R. L. Khalﬁn, Y. Cohen, K. Shuster, and E. Zussman. Electrospinning of ultrahigh-molecular-weight polyethylene nanoﬁbers. Journal of Polymer Science Part B: Polymer Physics, 45(7):766–773, apr 2007. ISSN 08876266. doi: 10.1002/ polb.21122. URL http://doi.wiley.com/10.1002/polb.21122.

[241] Sung-Seen Choi, Seung Goo Lee, Seung Soon Im, Seong Hun Kim, and Yong L. Joo. Silica nanoﬁbers from electrospinning/sol-gel process. Journal of Materials Science Letters, 22:891–893, 2003. doi: 10.1023/A:1024475022937.

[242] Abigail Freyer and Nancy Ortins Savage. Chapter 8 - Eﬀects of Sol Viscosity and Application to Thin Layer Chromatography. In Electrospun Silica Nanoﬁber Mats, pages 139–150. American Chemical Society, jan 2014. doi: 10.1021/bk-2014-1183.ch008. URL http: //pubs.acs.org/doi/abs/10.1021/bk-2014-1183.ch008.

[243] Jianfeng Du, Yuanzheng Cui, Yuchuan Liu, Rajamani Krishna, Yue Yu, Shun Wang, Chenghui Zhang, Xiaowei Song, and Zhiqiang Liang. Preparation of benzodiimidazole-containing covalent triazine frameworks for enhanced selective CO2 capture and separation. Microporous and Mesoporous Materials, 276(July 2018):213–222, mar 2019. ISSN 13871811. doi: 10.1016/j.micromeso.2018.10.001. URL https://doi.org/10.1016/j.micromeso.2018.10.001https://linkinghub.elsevier.com/retrieve/pii/S1387181118305262.

[244] Kewei Wang, Liming Yang, Xi Wang, Liping Guo, Guang Cheng, Chun Zhang, Shangbin Jin, Bien Tan, and Andrew Cooper. Covalent Triazine Frameworks via a Low-Temperature Polycondensation Approach. Angewandte Chemie International Edition, 56(45):14149–14153, nov 2017. ISSN 14337851. doi: 10.1002/anie.201708548. URL http://doi.wiley.com/10.1002/anie.201708548.

[245] Christian B Meier, Reiner Sebastian Sprick, Adriano Monti, Pierre Guiglion, Jet-Sing M. Lee, Martijn A Zwijnenburg, and Andrew I Cooper. Structure-property relationships for covalent triazine-based frameworks: The eﬀect of spacer length on photocatalytic hydrogen evolution from water. Polymer, 126:283–290, sep 2017. ISSN 00323861. doi: 10.1016/j.polymer.2017.04.017. URL http://dx.doi.org/10. 1016/j.polymer.2017.04.017https://linkinghub.elsevier.com/retrieve/pii/S0032386117303853.

[246] Jianwei Ren, Nicholas M. Musyoka, Perushini Annamalai, Henrietta W. Langmi, Brian C. North, and Mkhulu Mathe. Electrospun MOF nanoﬁbers as hydrogen storage media. International Journal of Hydrogen Energy, 40(30):9382–9387, 2015. ISSN 03603199. doi: 10.1016/j.ijhydene.2015.05.088. URL http://dx.doi.org/10.1016/j.ijhydene.2015.05.088.

[247] NOAA. Keeling Curve - May 2019, 2019. URL https://www.esrl.noaa.gov/gmd/ccgg/trends/full.html.

[248] Christian S. Diercks and Omar M. Yaghi. The atom, the molecule, and the covalent organic framework - Supplementary Materials. Science, 355(6328):eaal1585, mar 2017. ISSN 0036-8075. doi: 10.1126/science.aal1585. URL http://www.sciencemag.org/lookup/doi/10.1126/ science.aal1585.