MSc Chemistry
Energy and Sustainability
Master Thesis
Investigation Towards Zinc M12L24 Cages
By
Charilaos Asproulis
10629769
December 2020
60 EC
10 November 2016 – 31 July 2017
Supervisor/Examiner:
Examiner:
Prof. dr. J.N.H. Reek
Prof. dr. J. H. Van Maarseveen
Van ’t Hoff Institute for Molecular Sciences
Homogeneous, Supramolecular and Bio-Inspired CatalysisTo Zinc or not to Zinc?
Charilaos Asproulis – 10629769 Supervisor: Prof. dr. J.N.H. Reek Daily supervisor: Arnout Hartendorp MSc. Other supervisors: Valentinos Mouarrawis MSc.Nicolaas van Leest MSc.
Second reviewer: Prof. dr. J.H. van Maarseveen
Van ‘t Hoff Institute for Molecular Sciences University of Amsterdam
Table of Contents
List of Abbreviations ... 6
1. Abstract ... 7
2. Introduction ... 8
2.1 Background ... 8
2.2 M12L24 cages and catalysis ... 9
2.3 Zinc ... 12
2.4 Goal of the project ... 14
3. Results ... 16
3.1 Cage 1 ... 16
3.1.1 Synthesis of building block 1 ... 16
3.1.2 Cage 1 synthesis attempts ... 17
3.2 Cage 2 ... 20
3.2.1 Cage 2 synthesis attempts ... 20
3.3 Cage 3 ... 26
3.3.1 Synthesis of building blocks 3 and 4 ... 27
3.3.2 Cage 3 synthesis attempts ... 27
4. Conclusions ... 29
5. Future Prospects ... 30
6. Acknowledgements ... 31
7. Bibliography ... 32
8. Experimental ... 35
8.1 Materials and methods ... 35
8.2 Building blocks ... 35
8.3 Syntheses ... 36
8.4 1H-NMR spectra ... 42
List of Abbreviations
ACN Acetonitrile Au Gold B Boron Co Cobalt Cu Copper DCM Dichloromethane DMF Dimethylformamide DMSO Dimethyl sulfoxideDOSY Diffusion Ordered SpectroscopY
F Fluorine
L Ligand
LC-MC Liguid Chromatography-Mass Spectroscopy
M Metal
MeOH Methanol
NMR Nuclear Magnetic Resonance
Pd Palladium
Pt Platinum
Rh Rhodium
THF Tetrahydrofuran
TLC Thin layer chromatography TON Turnover number
TOF Turnover frequency
Si Silica
XRD X-ray crystallography
1. Abstract
The global energy demand is increasing rapidly and the scientific society needs now more than ever to invest in sustainability1. Catalysis is an energy-saving process, however, it is crucial to further improve the current catalytic efficiencies which will lead to more sustainable and cheaper syntheses of chemicals. In this direction, nature may serve as a source of inspiration on improving catalytic processes, as the working principles of enzymes have been thoroughly investigated, since their cavities can improve the catalytic efficiency. Enzymes perform catalytic transformations exceptionally, by utilizing the so-called second coordination sphere. There has been a development of abiological catalytic systems that mimic the molecular environment around the active centre in an enzyme; these systems are called (supra)-molecular cages and they introduce a second coordination sphere around the catalyst and create the desired confined space2. There is a lot of focus in a specific type of organic ligands, which can bind to metals such as Pt and Pd and form M12L24 cages 3. These cages have been successfully used in improving the catalytic process.4–6 Some examples involve the use of TEMPO or Au catalysts for tandem or hydro-alkoxylation reactions respectively. Present study attempts to investigate the synthesis of cage analogues using Zn as the metal instead of Pt and Pd. Zinc is more abundant than Pt and Pd and it is considered more sustainable7,8. Furthermore, it is less toxic,9 and therefore suitable for in vivo applications. Moreover, Zinc as a d10 metal is redox innocent10 and does not quench fluorescence.11 Finally, Zinc can adapt the desired geometry for the formation of the M12L24 type cages.12 The goal of the research project was to attempt and synthesize a M(2)12L24 cage based on Zinc.
2. Introduction
2.1 BackgroundThe rapid population increase has resulted in a major increase in energy consumption around the world1. Now more than ever, the need for sustainable processes is urgent in almost every industry. The chemical industry is no exception and a lot of research has been devoted to finding more sustainable methods to synthesize chemicals.1 Catalysis is a saving energy process as it minimizes the energy needed for a chemical reaction by lowering the activation energy and thus increasing the efficiency of the reaction (Figure 1). 13
Figure 1: Energy diagram showing the energy needed for the same reaction with (red line) and
without catalyst (black line) 13
Transition metal homogeneous catalysis plays an important role in the sustainability of chemical processes.14 Transition metal catalysts are used in the bulk and the fine chemical industry, and their main advantage is that they can be tuned by altering the ligands that are coordinated to the metal.2 This way, the efficiency for a wide range of reactions can be increased. 2 Homogeneous catalysis has come far and has provided valuable tools for improved catalytic efficiency,15 however, there is still much room for improvement, and many researchers investigate nature-inspired alternatives. 16,17
More specifically enzymes have been thoroughly studied due to their exceptional catalytic capabilities.18,19 Enzymes can perform catalysis in high selectivity and activity, and it has been suggested that the confined space created by them, plays a major role in the catalytic process.20 A lot of research focuses on creating artificial analogues.1 These analogues are synthesized by introducing a second coordination sphere, around the catalyst, which leads to the formation of a confined space. This confined space creates a different environment compared to the bulk, and this different environment comes with certain advantages. Two important advantages are the reduction of the energy barrier and the alternation of the pathway of a given reaction.2 An example of how a cage can affect a reaction is shown in figure 2.1
Figure 2: a) Comparison of energy barriers for a reaction within a cage (blue line) and in the bulk
(red dashed line) leading to the same product. b) Comparison of energy barriers for a reaction within a
cage (blue line) and in the bulk (red dashed line) leading to different products. 1
An artificial (supra)-molecular cage can affect a reaction for a variety of reasons. By creating a unique environment around the active centre 2 the substrates can enter the cavity due to lower entropy,21 hydrophobic forces,22 by metal coordination (possibly to the catalysts) or due to electrostatic effects.23 Moreover the confined space around the catalyst can pre-organize substrates by restricting their movement,21 so the reagents can have easier access to the catalysts’ active side.1 These confinement effects can result in different selectivity and activity compared to the bulk. Present report focuses on organometallic cages with M12L24 configuration.
2.2 M12L24 cages and catalysis
Present report focuses on metal-coordination cages that are based on metal coordination to an organic ligand.24–26 The cages of interest in the current report are those who result in the M12L24 configuration due to the angle of the 24 organic ligands (approximately 120o degrees) linked to the 12 metal centres.27,28 One of the reasons these cages are of great interest, is that they are self-assembled and simple modification of the ligands can result in endo- or exo-functionalization which opens a lot of possibilities (figures 3 and 4).3 These cages are the thermodynamic product, and during their self-assembly different intermediates can be formed. 29
The M12L24 cages have been successfully used in catalysis, and two examples are discussed in this section demonstrating the utility of the second coordination sphere in catalytic efficiency by increasing the catalyst concentration and by preventing catalyst de-activation. The first example is a gold catalysed process, which has gained a lot of attention the last decades30–33 due to the high efficiency of the electrophilic activation of alkynes.34–37 Gold catalysts demonstrate high activity (lower LUMO and poor back donation)38–40, however, their low turnover number is one of the reasons that the catalytic process is considered relatively expensive, and this is a major reason why it can be seen as challenging to use gold catalysts in large commercial scales41 Thus, the development of gold catalytic systems where
high TONs and TOFs can be obtained is of great utility. Joost Reek and co-workers successfully demonstrated the effect of M12L24 cages in gold catalysis.5 By utilizing mixtures of two different ligands for the formation of Pt12L24 cages (figure 5) they were able to achieve higher selectivity for reactions within the cage, compared to the bulk. This system proved to be a diverse catalytic platform where several gold-catalysed cyclization reactions were successfully performed (figure 6).4
Figure 3: Examples of endo- functionalized M12L24 self-assembly3
Figure 5: A the building block linked to the gold catalyst, B the empty building block, on the right is
an example of the cage5
More specifically they used 24 ligands endo-functionalized with gold chloride complex, resulting in the formation of a M12L24 self-assembled cage, with 24 gold complexes inside the cavity. The local concentration (1M) was estimated higher than a typical catalytic process-taking place in the bulk (10-6 to 10-3M). Also, they used the ligand without the gold complex and formed M12L24 spheres with different ratios between the ligand with the gold complex and the ligand without. The ratio varied from 0:24 to 24:0 with respect to the two types of ligands, giving the basis for the analysis of the effect of the concentration of the catalyst. Compared to the catalyst in the bulk where the gold-chloride complex is inactive42–44 the catalysts inside the cage (where the concentration of the catalyst was 0.27M or higher) showed conversion without the activation step. They demonstrated that, after a threshold of catalyst concentration, the Au-Au interactions form multinuclear complexes that are active and can initiate the catalysis (the step of de-halogenation is not needed). The higher concentration of the catalyst inside the cavity compared to the bulk, resulted in higher yields. Some general effects of higher selectivity were also found. However, absolute selectivity towards only one product was not achieved.
In the next example, prevention of catalyst deactivation was demonstrated by Fujita et al. More specifically they demonstrated how the M12L24 cages can be used in tandem catalysis. They were able to synthesize two M12L24 cages including two different catalysts (TEMPO and MacMIllan’s catalyst). By doing so they performed allylic oxidation followed by Diels-Alder cyclization (figure 7) due to the fact protective environment of the cages prevent the deactivation of the catalysts.6
Figure 7: Tandem catalysis (allylic oxidation and Diels-Alder cyclization) in two steps6
Fujita cages have made a positive impact on catalysis.4–6 These cages are based on Pt-pyridil and Pd-pyridil coordination. Even though their utility has been well established in catalysis, they come with certain limitations. Pt and Pd are not considered as sustainable and they are expensive, not abundant metals.7 Moreover, Pt and Pd are not ideal for in vivo application due to their high toxicity.45 Even though they are used in certain types of drug delivery including metallo-drugs for cancer, they result in high systemic toxicity45, which can be an issue for in vivo catalysis. The purpose of this report is to expand the design of the M12L24 cage in more sustainable and less toxic analogues.
2.3 Zinc
Zinc is a relatively abundant metal7 and as it is shown in figure 8. In figure 8 the scarcity of metals with respect to atoms per 106 atoms of silica is demonstrated. Platinum and Palladium, which are mostly used in the M12L24 cages, both have 10-3.5 atoms per 106 Si atoms whereas Zinc is much more abundant with 102 atoms per 106 Si atoms. Moreover, figure 9 demonstrates the material resource scarcity and the eco-cost metal scarcity.8 Palladium and Platinum are both higher than Zinc.
Figure 8: Abundance of elements (per 106 atoms of Si) to their atomic number7
Figure 9: Eco-cost metal scarcity8
For a sustainable future, the utilization of cheap and abundant metals is of great importance. Looking at the previous two figures one can say which metals should be considered relatively inexpensive. The abundance of Zinc is relatively high and as a result, it can be considered as a more sustainable metal compared to Platinum and Palladium. There are also other reasons, more than economic ones, to investigate Zinc. First, Zinc as a d10 metal is redox innocent. Moreover, Zinc does not quench fluorescence; its electron configuration does not allow Zinc to be present in either an electron or an energy transfer process.11,10 With that respect, Zinc is a relatively inert metal centre for the formation of a cage.
Another reason why Zinc is worth investigating is that it is less toxic, and it can be compatible with certain in vivo applications. Considering toxicity one should keep in mind that measurement of toxicity is not a simple thing. A lot of variables should be taken into account, however, Zinc is considered relatively harmless 9 and thus a good candidate for substituting Pt and Pd.
Zinc is convenient since its allowed geometries are suitable for a M12L24 configuration. As a d10 metal, Zinc’s most common geometry is tetrahedral.46 However the appropriate geometry for a M12L24 formation is that of an octahedral, and Zinc it has shown to adapt this coordination geometry (among others in figure 10).47 Moreover, to obtain the desired cage geometry, the respective pyridil building blocks should be replaced by their carboxylic acid analogues which leads to the formation of M(2)12L24. Finally, Zinc in a M(2)12L24 architecture is bound to two oxygen molecules, which may lead to increased stability of the cage.
Figure 10: Geometry formations of Zinc 47
2.4 Goal of the project
In the previous sub-section, the choice of Zinc was justified as a potential alternative for Pt and Pd in M12L24 self-assembled cages. The first step is to synthesize the acid building block 1 (figure 11), in which each carboxylic acid side will be able to form two bonds with Zinc centres, which will lead to cage 1; a M(2)12L24 self-assembled cage. Since the attempted cage is neutral, for the characterization and the confirmation of the formation of the cage, 2D NMR spectroscopy (1H-DOSY) will be used since not mass analysis is possible.
The second step is to synthesize M(2)12L24 cages. The building block 1 and a Spartan model of cage 1 are shown in figure 11.
The purpose of this report is to gain new insight about (supra)-molecular cages and more specifically to try to answer the research question about what role can Zinc play in forming more sustainable M(2)12L24 cages.
Figure 11: A common ligand used in the synthesis of M12L24 cages, building block 1 and the
Br Br + H
SiMe3
CuI, PdCl2(PhCN)2, Et3N, HP(t-Bu)3BF4 Dioxane, 46oC stirring overnight
Me3Si SiMe3
MeOH stirred at 25oC overnight H H
+ Br O O O O O O CuI, PdCl2(PPh3)4, Et3N THF, 45oC stirring overnight KOH THF, MeOH, H2O 60oC, 72h O OH O OH N N DBU Yield = 50% Yield = quantitatively Yield = 79% Yield = quantitatively
3. Results
3.1 Cage 13.1.1 Synthesis of building block 1
The synthesis of the building block was done in 4 steps (figure 12). First, a Sonogashira coupling, followed by deprotection, and a second Sonogashira coupling, and finally building block 1 was obtained after saponification. The formation of the desired building block was confirmed with 1H-NMR (figure 13).
Figure 13: H-HNMR of the aromatic region of the free building block in DMSO.
3.1.2 Cage 1 synthesis attempts
The self-assembly of cages can be achieved by mixing the building blocks with the metal precursor (at room or elevated temperatures). Cage attempts were done using Zinc acetate as a metal precursor, in 1:1 ratio with the building block 1 in DMSO and DMF at 700C. The 1H-NMR spectra of both attempts compared to the free building block are demonstrated in figure 14 and figure 15 respectively.
Figure 14: 1H-NMR comparison between the cage attempt (red on the top) in DMSO and the free
Figure 15: 1H-NMR comparison between the cage attempt (black line on the bottom) in DMF and
the free building block (red line on the top) indicating the formation of a new species.
The disappearance of the peak of carboxylic acid protons together with broadening and minor shift of the peaks of the aromatic region in the 1H-NMR indicate the formation of a new species. 1H-DOSY indicates a hydrodynamic radius of approximately 1.35nm, which is not in line with the expected radius based on the Spartan model and on previously reported similar cages48.
Figure 16: 1H-DOSY spectroscopy of the cage attempts in DMSO (blue). In red: DOSY of a sphere
with the appropriate size for comparison reasons. Calculated radius based on the diffusion constant in DMSO 1.35nm (2.7nm diameter).
Figure 17: 1H-DOSY spectroscopy of the cage attempts in DMF. Calculated radius based on the
diffusion constant in DMF 1.32nm (2.64nm diameter).
Since cage 1 was not formed based on 1H-DOSY NMR, a series of different strategies/attempts were pursued. The attempts for the cage formation included several strategies such as testing different concentrations, adding pyridine to labialize the Zinc bonds (in order to overcome a possible kinetic trap), the usage of different solvents and Zinc precursors and some attempts to use Rhodium (which forms stronger bonds).
In order to investigate the effect of concentration, different cage attempts in different concentrations were performed according to table 1 (section 8.4.5). 1H-NMR was taken for all the entries and it was compared with the spectrum of the building block in DMSO. 1H-NMR indicated that concentration did not affect the pathway of the experiments (section 8.4.5). Some attempts for the cage formation were performed via very slow addition of Zinc acetate. The total addition until the ratio of 1:1 with regards to the Zn(OAc)2 and the building block was done in a duration of total of 10 days. According to 1H-NMR this approach showed that there was not difference compared to the first attempts when the Zinc precursor and the building block were added in a ratio 1:1 into the solution at the same time.
The next attempt included the addition of pyridine in different concentrations. The reason for that was to overcome a kinetic trap, and form the thermodynamic product. Table 2 (section 8.4.6) summarizes the attempts of this approach. However, this strategy did not lead to the desired outcome. A series of experiments of different approaches were performed towards the formation of the desired assembly, including different Zinc precursors, in different solvents. More specifically Zn(OAc)2 and Zn(NO3)2 in THF, DMF, MeOH and DMSO were used as it is shown in table 3 (section 8.4.7).
In an attempt to check a different system, Zinc was substituted by Rhodium. Table 4 (section 8.4.8) summarizes the most representative attempts. In some attempts, slow addition over the course of 72 hours occurred. It is worth mentioning that Rhodium solutions were not colourless as Zinc’s, and a change of colour could be indicating the formation of a new
were observed based on 1H-NMR spectra. 1H-DOSY spectroscopy for two of the solutions indicated that even the largest molecule in the solution was not large enough compared to the size of cage 1. More specifically 1H-DOSY indicated a hydrodynamic radius of 1-1.3 nm thus a diameter 2-2.6, and based on the calculations in Spartan the expected value of diameter for cage 1 was around 5 nm. In the sub-section 3.2 an investigation of a system attempted to provide new insight on the coordination of Zinc to a shorter acid analogue, isophthalic acid.
3.2 Cage 2
After numerous attempts to form cage 1, an idea to form cage 2 emerged. According to literature in supra-molecular spheres, a determining factor of the geometry of the cage is the angle of the building block.27,28 The concept was to replace the building block 1 with isophthalic acid (building block 2) since both can coordinate to Zinc from the same angle (approximately 120o degrees) (figure 18).
Figure 18: building blocks 1,2 and cage 2
Previous research has shown that other transition metals such as Cobalt and Copper, have been successfully used for the formation of similar small frameworks.51. There are certain advantages of using building block 2. First, it is commercially available and inexpensive which provides the possibility for numerous cage attempts in a variety of conditions without any necessary synthetic steps. Moreover, it is more soluble than building block 1, which provides additional options regarding the solvent.
3.2.1 Cage 2 synthesis attempts
The following spectra (figure 19) demonstrate the obtained result for the formation of the cage 2. In the first one, the 1H-NMR shows some interesting shifts. First of all, singlet has shifted downfield which is an indication of its presence in an electron-denser environment
would be pointing inside). Secondly, the doublet and the triplet that would have pointed outside of the cage, they were expected to have shifted upfield, which was the case. Moreover, the doublet and the triplet are less sharp which is also promising since in literature the cages formed show broadening in the 1H-NMR peaks.48
Figure 19: 1H-NMR: comparison in the aromatic region between the free building block 2 (red) and
the product after the addition of Zinc acetate (black). The singlet shows a downfield shift, and the doublet and the triplet show an upfield shift.
Following the 1H-NMR spectroscopy, 2D spectroscopy also provided some promising results. Specifically, 1H-DOSY spectrum showed that the new species formed in the solution had a size significantly larger than the size of the free building block 2. This is demonstrated in the 1H-DOSY spectrum in figure 22, in which the peaks in red are referred to the free building block and the blue peaks on the product of the cage formation attempt. The D value extracted from the spectrum and the hydrodynamic radius and the diameter were calculated for the product. The Spartan calculations predicted diameter of 2.89nm and the diameter calculated by 1H-DOSY (figure 20) was 2.76nm (hydrodynamic radius = 1.38). Based on 1H-NMR and 1H-DOSY NMR the use of building block 2 resulted in the formation of the desired cage 2.
N N
Figure 20: 1H-DOSY comparison between the free building block 2 (red) and the product after the addition of Zinc acetate (blue)
To provide additional evidence an encapsulation of a guest molecule was attempted. The following 1,4-di(pyridine-4-yl)benzene (figure 21) seemed to be of the right size to be encapsulated inside the cage. First, as it is demonstrated by the Spartan model (figure 22) its length is appropriate, and second, it looks like there is enough space in the windows of the cage for the 1,4-di(pyridine-4-yl)benzene to go through them. However, the results indicated that encapsulation did not occur (figure 23,24).
Figure 21: 1,4-bis(4-pyridyl)benzene, the molecule used for the encapsulation
Figure 23: Coordination of the 1,4-bis(4-pyridyl)benzene, all peaks show downfield shifts.
Figure 24: 1H-NMR comparison of different equivalents of 1,4-bis(4-pyridyl)benzene per cage. Free
1,4-bis(4-pyridyl)benzene (red), 0.50 equivalent per cage (green), 1 equivalent per cage (blue), 2 equivalents per cage (purple). If encapsulation would have occurred then blue and purple should have been different. It seems that 1,4-bis(4-pyridyl)benzene was coordinated to Zinc from the outside of the cage.
In order to get a better understanding of what happened with the addition of 1,4-bis(4-pyridyl)benzene in the solution, and in search for additional proof that the cage was formed, the hypothesis that 1,4-bis(4-pyridyl)benzene coordinated to two cages from the outside (figure 25) was confirmed based on 1H-NMR spectroscopy (figures 26 and 27).
Figure 25: Visualisation of 1,4-bis(4-pyridyl)benzene coordinated to two cages based on Spartan
model.
Figure 26: 1H-NMR spectrum of the solution of cage 2 after the addition of the
1,4-bis(4-pyridyl)benzene, where a, b and c refer to the protons of 1,4-bis(4-pyridyl)benzene. Note (i) and (ii) are the metal Zinc centres of the cage, represented here with only one (instead of 4) ligands for each metal centre for simplicity.
A possible coordination of 1,4-bis(4-pyridyl)benzene to a M(2)12L24 cage, would transform a corner of a cage from (i) to (ii). Note that (i) and (ii) are visual parts of the cage, showing a
metal centre with only one ligand instead of four, for simplification. It can be observed that coordination of 1,4-bis(4-pyridyl)benzene does not affect protons pointing inside the cage (1), it causes a downfield shift on the triplet (3 ! 3’) and causes a split and a downfield shift in protons 2 and 4, which before the coordination were equivalent. Interpretation of the spectrum comes in line with the hypothesis that 1,4-bis(4-pyridyl)benzene coordinates from the outside of the cage, providing additional support of the cage formation.
Moreover, integration of the peaks (figure 27) indicates that one molecule of 1,4-bis(4-pyridyl)benzene affects 8 ligands (building block 2) and around 40 remain unaffected. In total a ratio of 1 molecule of pyridyl)benzene for 48 ligands show that 1,4-bis(4-pyridyl)benzene is coordinated to two M(2)12L24 cages.
Figure 27: 1H-NMR Integration of peaks from figure 26. Integration is set so that 4 represents 4
proton, 8 representing 4 protons etc. (Integration c = 4.01, b = 3.96, a = 4.09, 1+1' = 48.77, 2+4 = 81.84, 2' = 8.25, 4' = 8, 3 = 41.90, 3' = 8)
In search for further proof that the cage was formed, a series of crystallization attempts resulted in suitable crystals for X-ray analysis. However, the results showed a polymeric structure. Further process of the data shows in figure 28 the repetitive unit of the polymer. In the place where two building block 2 molecules are coordinated to two Zinc molecules, the one Zinc has an octahedral geometry configuration and the other a tetrahedral configuration. The molecule in the solution is not polymeric and thus polymerization probably occurred during the crystallization process.
Figure 28: Crystal structure of the repetitive unit obtained with XRD
3.3 Cage 3
A bonus exploratory experiment also took place mainly for fundamental research reasons. It has been shown in the literature that M2L4 self-assembled cages are promising for applications such as drug delivery, catalytic reactions, material science52. Furthermore, a Zinc M2L4 geometry has been reported, in which the Zn-Zn distance was 8.099 A, and where the Co-Co distance in the analogous cage was 8.075 A.52 In another research paper, Soo Lah and co-workers were able to synthesize an M(2)2L4 neutral metal-organic polyhedra using Cu.53 The research question for this bonus experiment was if the M(2)2L4 neutral Zinc cage could be synthesized, and more specifically when using building block 3 (figure 29).
The reason for using the building block 3 (figure 29) was twofold. First, the building blocks would provide a slightly larger M(2)2L4 cage (compared to other M2L4 cages), and secondly, its fluorescence properties could produce a fluorescent cage. This specific building block had not been used in a similar cage synthesis attempt before, and thus its pyridine analogous building block 4 (figure 29) was also synthesized to investigate if the formation of the cage would be possible in a more dynamic system using Palladium. According to a Spartan model a cage like the one in the figure 29 would be synthesized.
O O O O Pd(PPh3)4, CuI, Et3N THF, stirring overnight at 450C I I + O O H I2, HIO3, H2SO4, CHCl3
CH3COOH, stirring at 800C overnight
O OH HO O KOH THF, MeOH, H2O 60oC, 72h Yield = 92% Yield = 80 % Yield = quantitatively I I + H I2, HIO3, H2SO4, CHCl3
CH3COOH, stirring at 800C overnight
N Pd(PPh3)4, CuI, Et3N THF, stirring overnight at 450C N N Yield = 62% Yield = 92%
3.3.1 Synthesis of building blocks 3 and 4
The building block 3 was synthesized in three steps (for more details look at the experimental section). These synthesis steps are shown below. Confirmation of the synthesis of the building block was done by 1H-NMR (section 8.4.9).
Figure 30: Synthesis of the building block 2 in three steps.
The building block 4 was synthesized in two steps (one after the diiodo-fluorene synthesis). Characterization of the final product was done with 1H-NMR (section 8.4.11).
Figure 31: Synthesis of the building block 3 in two steps
3.3.2 Cage 3 synthesis attempts
Several attempts were performed for cage formations including different solvents, using different metals (even Rhodium), and different conditions. A small overview of the most representative attempts is shown in table 5. However, the results were conclusive, the formation of cage 3 was not succesful.
Table 5
Cage-synthesis attempts
Entry Metal
Precursor Building Block Solvent Temperature 1H-DOSY (diameter)
CA047 Rh(OAc)4 2 THF 800C 1.86nm
CA051 Zn(OAc)2 2 DMSO 800C 1.26nm
CA055 PdBF4 3 ACN 700C -
CA056 PdBF4 3 DMSO 700C -
CA057 PdBF4 3 THF 800C -
CA058 Rh(OAc)4 2 DMSO 1100C -
Note: In the column 1H-DOSY “-“ indicates multiple species impossible to identify (possibly polymers)
Figure 32: (1): 1H-NMR with palladium and building block 4, representative of all the attempts
leading to messy spectrum and probably the formation of multiple species and/or polymers;
The two 1H-DOSY spectra for Rhodium acetate with building block 3 in THF (2) and with Zinc
acetate with building block 3 (3) are representative of all the attempts that did not lead to molecules large enough to indicate the formation of the desired cage.
4. Conclusions
A variety of attempts to form cage 1 indicated that it is likely that the formation of the theoretically thermodynamic product is hardly achievable. The use of different Zinc precursors, in different concentrations and conditions in a variety of solvents, did not lead to the formation of cage 1. Moreover, the addition of pyridine to labialize the Zinc bonds did not lead to the desired result. In an attempt to investigate the analogue cage with Rhodium, led to the formation of multiple species and in some cases to polymeric structures. One reason that the cage was not formed, might be that our attempt led to a kinetic favourable product. Possibly kinetic intermediates towards M12L24 formation have been identified as M6L12, M8L16 and M9L28 geometries.54,29 Taking to account that the size of the complex in the solution had a diameter of around 2.7nm, it could have been a small M(2)6L12 but further analysis is needed to confirm that.
Unlike the cage 1, attempt to synthesize cage 2, showed promising results indicating that the cage was formed. 1H-NMR shifts indicated a successful cage formation. An attempt to encapsulate a guest molecule to provide additional information of the cage formation failed. However, further interpretation of the 1H-NMR spectrum showed that the guest molecule was exo-coordinated to two cages providing additional information that the cage is formed. Crystallization attempts were successful to produce appropriate crystals for XRD analysis; however, the crystal structure obtained was of a polymeric structure probably due to polymerization during the crystallization process.
Considering cage 3, results were not promising. Even though one species was formed, 1H-DOSY indicated a smaller radius compared to the one expected. Several attempts using Rhodium instead of Zinc, as well as the attempts using a pyridine analogue building block with Palladium indicated that the specific building block is not suitable for the formation of a M(2)2L4 cage. That is probably because of the angle, which the building block 3 was supposed to bond to the metal, and which could not lead to a M(2)2L4 cage geometry.
Regarding the research question and the role of Zinc in forming more sustainable and less toxic M(2)12L24 cages to improve catalytic efficiency, a partial answer was obtained. Even though Zinc was proved to be challenging as a Pt and Pd substitute, a synthesis of a M(2)12L24 was achieved. Thus, Zinc can be considered as plausible candidate for more sustainable less toxic cages. The actual toxicity of the cage, and tis role in improving catalytic efficiency still needs to be investigated.
5. Future Prospects
With regards to cage 1, there are some directions for future research. First of all, the use of a different precursor such as diethyl-zinc may produce different results. It has been shown in the literature that diethyl-zinc is much more robust, however, someone should have in mind that is a dangerous reactant, since it instantly reacts with oxygen in an exothermal way producing flames55 and thus people should be cautious when using it. Moreover, in order to investigate similar systems and provide new insight in the subject, future research could investigate metals such as Co and Cu, which might be difficult to analyze because of their paramagnetic properties; combined with the fact that the target-cage is neutral and mass analysis is challenging. However, it has been reported in literature that Co and Cu can form similar in geometry and formation of metal-organic polyhedra.51
Regarding cage 2, there are some implications. First of all, cage 2 is a novel Zinc-cage, which offers a cavity of approximately 2.8nm for catalysis. Its role in more sustainable and more efficient catalysis still needs to be investigated. Furthermore, its stability in different conditions needs to be investigated as well. This cage provided a relatively cheap end product since the building block 2 is not expensive, it is commercially available, and the synthesis of the cage is only one step. Furthermore, a cage 2, theoretically can be used in in vivo catalytic reactions, in which the redox and quenching fluorescence innocent Zinc provides some benefits. However, cage’s actual toxicity needs to be investigated as well.
Considering the characterization of cage 2, even though traditional mass spectroscopy cannot be used to provide additional evidence of the cage formation, future research can use advanced techniques for mass characterization of neutral molecules. 56 Alternatively addition of a charged moiety in the building block 2 could make the cage 2 appropriate for traditional mass analysis. Furthermore, encapsulation studies with different molecules than the one used in this project can provide additional evidence. Moreover, further crystallization attempts might result in a crystal structure providing exclusive information about cage 2 and its properties. Finally, substituting building block 2 with 1,1':3',1''-Terphenyl-4,4''-dicarboxylic acid (figure 33) might result in a bigger cage with a bigger cavity which will offer new potential for catalysis.
Figure 33: 1,1':3',1''-Terphenyl-4,4''-dicarboxylic acid as possible alternative of building block 2 since they can bind to Zinc from the same angle
6. Acknowledgements
After 9 months of research, I would like to thank several important people. First of all, I would like to thank Joost Reek for giving me the opportunity to be a part of the Homkat group and work on a very interesting and challenging project. I would also like to thank Bas de Bruin and Jarl-Ivar van der Vlugt for their useful input in the mini meetings. I would also like to thank Jan Maarseveen for being my second reviewer and for sharing with me his ideas for a project. Moreover, I would like to thank Arnout Hartendorp who was my daily supervisor; I would like to thank him for his guidance, patience and for the fact that he taught me to work efficiently and methodically in the lab. I also want to thank him for supporting, motivating and managing me, and for the way he helped me to adjust. Moreover, I would like to thank Valentino Marawi and Nicolaas van Leest for their input, they facilitated my fit to the group, they supervised me during weekends (both in the lab and in the drinks after), they supervised my written report Thesis and they helped me in numerous other ways; I literally have no words to thank these guys. I would also like to thank Simon Mathew for his useful advice and coaching when needed, his generosity of spirit and of course for the precious drunk moments shared (especially on Fridays). Moreover, I would like to thank Xavier and Eddie for their input as well as for being my dance partners, and thank them for being in the same level of crazy as I. I would also like to thank Lucas Charles, Pim and Bin for a creating a good office environment, Tessel, Yoeri and Marianne who are brilliant scientists and even though our discussions were primary social their input was important and unique. Furthermore, I would like to thank Eva for our cooperation in the lab and for being there for me when I needed help. Last but not least, I would like to thank the whole HomKat group for the amazing time I had while I was doing my internship, for the knowledge I gained, for the fun I had and for the transformation and growth I experienced both as a chemist and as a person. Without a doubt, these 9 months were and will be among the highlights of my life.
7. Bibliography
1. Koblenz, T. S., Wassenaar, J. & Reek, J. N. H. Reactivity within a confined self-assembled nanospace. Chem. Soc. Rev. 37, 247–262 (2008).
2. Leenders, S. H. A. M., Gramage-Doria, R., De Bruin, B. & Reek, J. N. H. Transition metal catalysis in confined spaces. Chem. Soc. Rev. 44, 433–448 (2015).
3. Harris, K., Fujita, D. & Fujita, M. Giant hollow MnL2n spherical complexes:
Structure, functionalisation and applications. Chem. Commun. 49, 6703–6712 (2013). 4. Leenders, S. H. A. M., Dürr, M., Ivanovic-Burmazovic, I. & Reek, J. N. H. Gold
Functionalized Platinum M12L24-Nanospheres and Their Application in Cyclization Reactions. Adv. Synth. Catal. 358, 1509–1518 (2016).
5. Gramage-Doria, R. et al. Gold(I) catalysis at extreme concentrations inside self-assembled nanospheres. Angew. Chemie - Int. Ed. 53, 13380–13384 (2014). 6. Ueda, Y., Ito, H., Fujita, D. & Fujita, M. Permeable Self-Assembled Molecular
Containers for Catalyst Isolation Enabling Two-Step Cascade Reactions. J. Am. Chem. Soc. 139, 6090–6093 (2017).
7. Survey, U. S. G. Rare Earth Elements — Critical Resources for High Technology. 1– 11 (2002).
8. Vogtländer, J., Peck, D. & Kurowicka, D. The eco-costs of material scarcity, a resource indicator for LCA, derived from a statistical analysis on excessive price peaks. Sustain. 11, (2019).
9. Plum, L. M., Rink, L. & Haase, H. The Essential Toxin: Impact of Zinc on Human Health. Int. J. Environ. Res. Public Health 7, 1342–1365 (2010).
10. Wu, Y. et al. Electroreduction of CO2 Catalyzed by a Heterogenized Zn-Porphyrin Complex with a Redox-Innocent Metal Center. ACS Cent. Sci. 3, 847–852 (2017). 11. Fabbrizzi, L. et al. Fluorescent Sensors for Transition Metals Based on
Electron-Transfer and Energy-Tranfer Mechanisms. Chem. - A Eur. J. (1996) doi:10.1002/chem.19960020114.
12. Crabtree, R. H. The organometallic chemistry of the transition metals. (2014). doi:10.1002/crat.2170200510.
13. File:Catalysis- Reaction progress.png - Wikimedia Commons.
https://commons.wikimedia.org/wiki/File:Catalysis-_Reaction_progress.png.
14. Cornils, B. & Herrmann W.A. Applied Homogeneous Catalysis with Organometallic Compounds. Applied Homogeneous Catalysis with Organometallic Compounds (Wiley-VCH Verlag GmbH, 2002). doi:10.1002/9783527618231.
15. Straub, B. F. Organotransition Metal Chemistry. From Bonding to Catalysis. Edited by John F. Hartwig. Angew. Chemie Int. Ed. 49, 7622–7622 (2010).
16. Vriezema, D. M. et al. Self-Assembled Nanoreactors. Chem. Rev. 105, 1445–1490 (2005).
17. Meeuwissen, J. & Reek, J. N. H. Supramolecular catalysis beyond enzyme mimics. Nat. Chem. 2, 615–621 (2010).
18. Breslow, R. Biomimetic Chemistry and Artificial Enzymes: Catalysis by Design. Acc. Chem. Res. 28, 146–153 (1995).
19. Wang, Z. J., Clary, K. N., Bergman, R. G., Raymond, K. N. & Toste, F. D. A
supramolecular approach to combining enzymatic and transition metal catalysis. Nat. Chem. 5, 100–103 (2013).
20. Zhang, X. & Houk, K. N. Why Enzymes Are Proficient Catalysts: Beyond the Pauling Paradigm. Acc. Chem. Res. 38, 379–385 (2005).
Am. Chem. Soc. 128, 10240–10252 (2006).
22. Dong, S. D. & Breslow, R. Bifunctional cyclodextrin metalloenzyme mimics. Tetrahedron Lett. 39, 9343–9346 (1998).
23. Jiao, J. et al. Design and self-assembly of hexahedral coordination cages for cascade reactions. Nat. Commun. 9, 1–8 (2018).
24. Liu, M., Liao, W., Hu, C., Du, S. & Zhang, H. Calixarene-Based Nanoscale Coordination Cages. Angew. Chemie Int. Ed. 51, 1585–1588 (2012).
25. Wiester, M. J., Ulmann, P. A. & Mirkin, C. A. Enzyme Mimics Based Upon
Supramolecular Coordination Chemistry. Angew. Chemie Int. Ed. 50, 114–137 (2011). 26. Dai, F.-R. & Wang, Z. Modular Assembly of Metal–Organic Supercontainers
Incorporating Sulfonylcalixarenes. J. Am. Chem. Soc. 134, 8002–8005 (2012). 27. Yokoyama, H., Ueda, Y., Fujita, D., Sato, S. & Fujita, M. Finely Resolved Threshold
for the Sharp M 12 L 24 /M 24 L 48 Structural Switch in Multi-Component M n L 2 n Polyhedral Assemblies: X-ray, MS, NMR, and Ultracentrifugation Analyses. Chem. - An Asian J. 10, 2292–2295 (2015).
28. Guillerm, V. et al. A supermolecular building approach for the design and construction of metal–organic frameworks. Chem. Soc. Rev. 43, 6141–6172 (2014).
29. Fujita, D., Yokoyama, H., Ueda, Y., Sato, S. & Fujita, M. Geometrically Restricted Intermediates in the Self-Assembly of an M 12 L 24 Cuboctahedral Complex. Angew. Chemie Int. Ed. 54, 155–158 (2015).
30. Liu, L. P. et al. Gold-catalyzed, intramolecular, oxygen-transfer reactions of 2-alkynyl-1,5-diketones or 2-alkynyl-5-ketoesters: Scope, expansion, and mechanistic investigations on a new [4+2] cycloaddition. Chem. - A Eur. J. 17, 10690–10699 (2011).
31. Liu, L. P., Malhotra, D., Paton, R. S., Houk, K. N. & Hammond, G. B. The [4+2], not [2+2], mechanism occurs in the gold-catalyzed intramolecular oxygen transfer reaction of 2-alkynyl-1,5-diketones. Angew. Chemie - Int. Ed. 49, 9132–9135 (2010).
32. Gorin, D. J., Sherry, B. D. & Toste, F. D. Ligand effects in homogeneous Au catalysis. Chem. Rev. 108, 3351–3378 (2008).
33. Brenzovich, W. E. Gold in total synthesis: Alkynes as carbonyl surrogates. Angew. Chemie - Int. Ed. 51, 8933–8935 (2012).
34. Garcia, P., Malacria, M., Aubert, C., Gandon, V. & Fensterbank, L. Gold-Catalyzed Cross-Couplings: New Opportunities for C-++C Bond Formation. ChemCatChem 2, 493–497 (2010).
35. Arcadi, A. Alternative synthetic methods through new developments in catalysis by gold. Chem. Rev. 108, 3266–3325 (2008).
36. Widenhoefer, R. A. Recent developments in enantioselective gold(I) catalysis. Chem. - A Eur. J. 14, 5382–5391 (2008).
37. Benitez, D. et al. A bonding model for gold(I) carbene complexes. Nat. Chem. 1, 482– 486 (2009).
38. Gorin, D. J. & Toste, F. D. Relativistic effects in homogeneous gold catalysis. Nature
446, 395–403 (2007).
39. Lein, M., Rudolph, M., Hashmi, S. K. & Schwerdtfeger, P. Homogeneous gold catalysis: Mechanism and relativistic effects of the addition of water to propyne. Organometallics 29, 2206–2210 (2010).
40. Pernpointner, M. & Hashmi, A. S. K. Fully relativistic, comparative investigation of gold and platinum alkyne complexes of relevance for the catalysis of nucleophilic additions to alkynes. J. Chem. Theory Comput. 5, 2717–2725 (2009).
41. Zahner, D., Rudolph, M., Stephen, A. & Hashmi, K. Homogenous gold catalysis. Gold Catalysis: Preparation, Characterization and Applications (2015).
doi:10.4032/9789814669290.
42. Belger, K. & Krause, N. Ammonium-salt-tagged IMesAuCl complexes and their application in gold-catalyzed cycloisomerization reactions in water. European J. Org. Chem. 2015, 220–225 (2015).
43. Kronig, S., Theuergarten, E., Daniliuc, C. G., Jones, P. G. & Tamm, M. Anionic N-heterocyclic carbenes that contain a weakly coordinating borate moiety. Angew. Chemie - Int. Ed. 51, 3240–3244 (2012).
44. Oliver-meseguer, A. J. et al. Room Temperature Linked references are available on JSTOR for this article : Small Gold Clusters Formed in Solution Give Reaction Turnover Numbers of 107 at Room Temperature in gold catalysis stems. 338, (2012). 45. Egorova, K. S. & Ananikov, V. P. Toxicity of Metal Compounds: Knowledge and
Myths. Organometallics 36, 4071–4090 (2017).
46. Crabtree, R. h. The organometallic chemistry of the transition metals. (John Wiley & Sons, 2014).
47. Patel, K., Kumar, A. & Durani, S. Analysis of the structural consensus of the zinc coordination centers of metalloprotein structures. Biochim. Biophys. Acta - Proteins Proteomics 1774, 1247–1253 (2007).
48. Bruns, C. J. et al. Emergent Ion-Gated Binding of Cationic Host–Guest Complexes within Cationic M 12 L 24 Molecular Flasks. J. Am. Chem. Soc. 136, 12027–12034 (2014).
49. Li, D., Kagan, G., Hopson, R. & Williard, P. G. Formula Weight Prediction by Internal Reference Diffusion-Ordered NMR Spectroscopy (DOSY). J. Am. Chem. Soc. 131, 5627–5634 (2009).
50. Cohen, Y., Avram, L. & Frish, L. Diffusion NMR Spectroscopy in Supramolecular and Combinatorial Chemistry: An Old Parameter?New Insights. Angew. Chemie Int. Ed. 44, 520–554 (2005).
51. Li, J.-R. & Zhou, H.-C. Bridging-ligand-substitution strategy for the preparation of metal–organic polyhedra. Nat. Chem. 2, 893–898 (2010).
52. Schmidt, A., Casini, A. & Kühn, F. E. Self-assembled M2L4 coordination cages: Synthesis and potential applications. Coord. Chem. Rev. 275, 19–36 (2014).
53. Jaya Prakash, M. et al. Edge-directed [(M2)2L4] tetragonal metal–organic polyhedra decorated using a square paddle-wheel secondary building unit. Chem. Commun. 46, 2049 (2010).
54. Yoneya, M., Tsuzuki, S., Yamaguchi, T., Sato, S. & Fujita, M. Coordination-Directed Self-Assembly of M 12 L 24 Nanocage: Effects of Kinetic Trapping on the Assembly Process. ACS Nano 8, 1290–1296 (2014).
55. Susan, B. The Merck index: an encyclopedia of chemicals, drugs, and biologicals. (Rahway, N.J., USA :Merck, 1989, 1989).
56. Sage, E. et al. Neutral particle mass spectrometry with nanomechanical systems. Nat. Commun. 6, 6482 (2015).
OH O HO
O
8. Experimental
8.1 Materials and methods
All reagents and solvents were commercially available. For the 1H-NMR spectra the solvents used were deuterated DMSO, deuterated DMF, deuterated THF and CDCl3 and the samples were analysed by Bruker ARX 400 and Bruker 300. 1H-DOSY spectra were obtained from a DRX 500 with the temperature and gradient calibration before the measurements. The temperature was at 298 K and the D values found by the measurement were used to calculate the radius of the molecule in the sample. D coefficient and radius are correlated.49 More specifically the mathematic formula used for that correlation is the Stokes-Einstein equation, D = K*T/(6*π*η*R)50 where D is the diffusion coefficient, K the Boltzmann constant, T the temperature in Kelvin, η the viscosity of the solvent and r is the radius of the molecular sphere 8.2 Building blocks Building block 1: Building block 2: Building block 3: Building block 4: O OH O OH
8.3 Syntheses
Building block 1 Synthesis of CA002
In 25mL schlenk, PdCl2(PhCN)2 (46mg, 0.12mmol, 0.06 eq), HP(t-Bu)3BF4 (70mg, 0.24mmol 0.12 eq) were added together with NEt3 (2.93mL, 21.02mmol, 10.51 eq). In another 25mL schlenk trimethyl-silyl-acetylene (0.73mL, 5.2mmol, 2.6 eq) and copper iodine (15mg, 0.08mmol, 0.04 eq) were added. 0.24mL of 1,3 di-bromobenzene with 12mL of dioxane were added to second schlenk. Then the first solution was added to the second one and the mixture left stirring overnight at 460C. The next day a crude 1H-NMR indicated that the reaction did not proceed fully (mono-substitution) so the amount of catalyst was doubled, 0.27mL of trimethyl-silyl-acetylene were added and the mixture was left stirring at 460C over the weekend. The reaction was cooled down; the mixture was diluted with 36mL of ethyl acetate and transferred to a 250mL separation funnel. 36mL of H2O were added. The water layer was extracted 3 times. The combined organic layers were dried over MgSO4 for 5 minutes. This was filtered and concentrated under vacuum. Column chromatography was followed with petroleum ether as eluent to obtain the final product (1H-NMR was good) in a 50% yield.
Synthesis of CA003
CA002 (27.5 mg 0.1mmol, 1 eq) together with DBU (0.03mL, 0.21mmol, 2.1eq) were added in a schlenk together with 1 mL MeOH and stirred in room temperature for four hours at room temperature. When the solvent was evaporated the yellow solution resulted in purple oil. De-protection succeeded in a quantitatively yield.
Synthesis of CA009
Pd(PPh3)4 (150mg, 0.03 eq), and NEt3 (7.5mL, 13 eq) and (bromobezoate) (1.95mL, 12mmol, 3eq) and copper iodine (25mg, 0.03 eq) and CA003 (0.52mL 3.9mmol, 1eq) were placed together in a schlenk with 35mL of THF in inert conditions under Nitrogen, at 60oC over the weekend. After LC-MS showed that the alkyne was not present in the solution, the solution was filtered with a small column in a separation flask with DCM, the solvent was evaporated and the residue was left in vacuum for thirty minutes. The product was purified with column chromatography with DCM-Hexane 4-1. The product was collected in a 79% yield.
OH O HO
O
Synthesis of building block 1 (CA021)
A solvent mixture of THF (30mL) MeOH (20ml) and H2O (10mL) was used. CA009 (50mg) in 20mL of the solvent mixture and KOH (250mg) in 40mL solvent mixture were mixed and stirred overnight at room temperature. LC-MS was used to check the reaction, when only one large peak in the UV spectrum, the rection was done. Yield= quantitatively.
1H NMR (400 MHz, DMF) δ 8.30 – 8.26 (m, 4H), 8.05 (d, J = 1.6 Hz, 1H), 7.98 – 7.94 (m, 4H), 7.91 (dd, J =
7.5, 1.7 Hz, 2H), 7.81 – 7.76 (m, 1H).
1H NMR (400 MHz, DMSO) δ 8.01 – 7.98 (m, 4H), 7.83 (d, J = 1.7 Hz, 1H), 7.73 – 7.69 (m, 4H), 7.68 (dd, J =
7.7, 1.7 Hz, 2H), 7.57 – 7.52 (m, 1H).
Building blocks 3 and 4
A solution of fluorene (0.03mol, 1 eq) Iodine (0.022mol, 0.74eq) Iodic acid (0.015mol, 0.5eq) concentrated sulfuric acid and chloroform in 40mL of acetic acid was heated to 80oC overnight. The resulting precipitate was filtered and dissolved in chloroform, the solution was washed with aqueous Na2S2O4 (0.5M) and dried over Na2SO4 and evaporated to dryness to a brown powder, which was recrystallized from EtOAc. The yield was 92%.
I I
I2, HIO3, H2SO4 CHCl3
CH3COOH, 800C overnight
Synthesis of building block 3
CA048: 2,7 di-Iodo-fluorene (39mg, 0.093mmol, 1eq) and (38mg, 0.237mmol, 2.5eq) were placed in a schlenk together with Pd(PPh3)4, CuI and Et3N (1mL) in 5mL of THF. The solution was heated at 50oC under Nitrogen overnight. The product was purified with column chromatography resulting in a yield of 28.9%.
I I + O OH H Pd(PPh3)4, CuI, Et3N THF, stirring at 450C, 72h O O O O
Synthesis of CA050
CA048 and KOH were added to the solvent mixture (3/2/1 THF/MeOH/H2O) and left stirring overnight at R.T
Synthesis of building block 4
2.7 di-iodo-fluorene 2,7 (500mg, 1.19mmol, 1eq) and (306mg, 2.975mmol, 2.5eq) together with the catalyst Pd(PPh3)4, and CuI were place in 250mL schlenk and dissolved in 10mL of Et3N and 72mL THF. The reaction occurred in inert conditions under nitrogen. The solution was left stirring at 50oC for 72h. The crude mixture was filtered and the solvent was evaporated. Plug-column with ethyl acetate was used to isolate the blue spot with was appeared in the TLC plate of the crude mixture. Then a new spot appeared in the TLC, which might be explained that it was in low concentration in the first TLC.
I I + N H N N Pd(PPh3)4, CuI, Et3N THF, stirring at 450C, 72h
Synthesis of the 1,4-bis(4-pyridyl)benzene
Synthesis according to literature, 1,4-dibromobenzene, and 4-pyridilboronic acid, and tetrakis(triphenylophosphine)-palladium were dissolved in solution of 1,4 dioxane and K2CO3 (2M) under inert conditions. The solution was heated to 1100C and stirred for 24 hours. Then the solution was left to cool down, solvent was evaporated and the residue was re-dissolved in DCM. The product was obtained after column chromatography.
8.4 1H-NMR spectra
8.4.1. Building block 1, H-NMR in DMSO: 1H NMR (400 MHz, DMSO) δ 8.01 – 7.98 (m, 4H), 7.83
(d, J = 1.7 Hz, 1H), 7.73 – 7.69 (m, 4H), 7.68 (dd, J = 7.7, 1.7 Hz, 2H), 7.57 – 7.52 (m, 1H).
8.4.2.a Building block 1, 1H-NMR in DMF: 1H NMR (400 MHz, DMF) δ 8.30 – 8.26 (m, 4H), 8.05 (d,
8.4.3 1H-NMR comparison between the cage attempt (red on the top) in DMSO and the free building block (black in the bottom) indicating the formation of a new species.
8.4.4 1H-NMR comparison between the cage attempt (black line on the bottom) in DMF and the free building block (red line on the top) indicating the formation of a new species.
8.4.5 Cage attempts with different concentrations (Zn acetate in DMSO)
Stock solutions were prepared, for the building block (9,9mg in 1 mL DMSO) and the Zinc acetate (20mg in 1mL DMSO)
Table 1
Cage attempts varying in Concentration
Entries C% Concentration VBuilding block V DMSO VZinc Acetate
A 80% 0.0180 0.330mL 0.340mL 0.330mL
B 60% 0.0140 0.259mL 0.482mL 0.259mL
C 40% 0.0090 0.160mL 0.680mL 0.160mL
D 20% 0.0046 0.080mL 0.840mL 0.080mL
Note: C% = percentages based on the concentrations of the initial attempt
Mixed and left stirring at 800C for 90 minutes
8.4.5 1H-NMR comparison between the building block (red in the bottom) and 4 cage formation attempts in different concentrations. . Note that in the spectrum concentration goes from the highest (yellow, A) to lowest (purple, D) with an upward direction
Another strategy for the cage formation was to add the Zn acetate very slowly. The idea was that if the Zn acetate was in low concentrations in the solution, maybe the formation of the corners (as they are shown in the Spartan figure below) will occur, and as the concentration increased cage 1 could possibly be formed. The total addition until the ratio of 1:1 with regards to the Zn(OAc)2 and the building block was done in a duration of total of 10 days. According to 1H-NMR this approach showed that there was not difference compared to the first attempts when the Zinc precursor and the building block were added in a ratio 1:1 into the solution at the same time.
8.4.6 Cage attempts with pyridine addition in different concentrations to labialize Zinc bonds (Zn acetate in DMSO)
S3,S4,S5 stock the stock solutions Table 2
Addition of pyridine
Entries eq pyridine n (pyridine) S3 (mL) S4 (mL) S5 (mL) (DMSO)
A 0.1 0.00226mmol 0.235 0.235 0.018 0.330mL
B 0.5 0.01130mmol 0.235 0.235 0.090 0.259mL
C 1 0.02260mmol 0.235 0.235 0.179 0.160mL
D 2 0.04520mmol 0.235 0.235 0.359 0.080mL
Note: S3 stock solution of the building block, S4 stock solution of the Zinc acetate,S5 = Stock solution of pyridine; eq with respect to the Zinc
8.4.6 1H-NMR comparison between the building block (red in the bottom) and 4 cage formation
attempts in different pyridine concentrations zoomed in the aromatic region. Note that in the spectrum concentration goes from the lowest (yellow, A) to highest (purple, D) with an upward direction (7.4.6). Note that in the spectrum concentration goes from the highest (yellow, A) to lowest (purple, D) with an upward direction
8.4.7 A variety of solvents and two different Zinc precursors Table 3.
Cage formation attempts Metal
Precursor
Equivalent Solvent Temperature Entry 1H-DOSY (radius)
Zn(OAc)2 1:1 DMSO 250C CA022 1.26nm
Zn(OAc)2 1:1 THF 250C CA024 0.5nm
Zn(OAc)2 1:1 DMF 250C CA027 1.32nm
Zn(OAc)2 1:1 DMF 900C CA027b 1.12nm
Zn(OAc)2 1:1 MeOH 250C - polymeric/cloud
Zn(OAc)2 1:1 MeOH 900C - polymeric/cloud
Zn(NO3)2 1:1 DMSO 1100C CA030 -
Zn(NO3)2 1:1 DMF 250C CA031a -
Zn(NO3)2 1:1 DMF 1000C CA031b -
Zn(NO3)2 1:1 DMF 1200C CA031c -
Note: “-“ indicates not 1H-DOSY was taken,
Example of 1H-NNMR of the attempts in the table above. Cage attempts with Zinc nitrate as a precursor in DMSO
8.4.7 1H-NMR comparison (zoomed in aromatic region) of the building block (red in the bottom), and
the addition of Zinc Nitrate after 8h at 1000C (yellow line), and 18h at 1000C (blue line), and after 48h
8.4.8 Rhodium analogue system
Table 4
Rhodium as a Metal instead of Zinc
Precursor Solvent Temperature t (hours)
1H-DOSY (radius)
Colour Entry
Rh2(OAc)4 DMSO 1100C 0 h - Red CA032a
Rh2(OAc)4 DMSO 1100C 18 h 1nm Yellow CA032b
Rh2(OAc)4 DMSO 1100C 48 h - Yellow +
Red precipitate CA032c Rh2(OAc)4 DMF 250C 0 h - Blue solution CA033a
Rh2(OAc)4 DMF 1200C 18 h 1.3nm Green CA033b
Rh2(OAc)4 DMF 1200C 48 h - Dark Brown CA033c Rh2(OAc)4 DMF 800C 0 h - Light Green CA052a
Rh2(OAc)4 DMF 800C 20 h - Dark brown
Rh2(OAc)4 DMF 800C 44 h - Dark
Brown
Rh2(OAc)4 DMF 800C 72 h - Brown Gel
Rh2(OAc)4 DMSO 800C 0 h - Red CA053
Rh2(OAc)4 DMSO 800C 1 h - Yellow
Rh2(OAc)4 DMSO 800C 5 h - Yellow
Rh2(OAc)4 DMSO 800C 0 h - Red CA054
Rh2(OAc)4 DMSO 800C 1 h - Yellow
Rh2(OAc)4 DMSO 700C 0 h - Red CA060
Rh2(OAc)4 DMSO 700C 240 h - Yellow
Note: For CA052 and CA053 building block was added in several steps until the 1:2 ratio For CA054 the Rhodium dimer was added in several steps until the ratio of 1:2
For CA060 the addition of the metal in the solution was slow and for the total 1:2 ratio 10 days in total passed
8.4.8a 1H-NMR comparison (zoomed in aromatic region) of the building block (red in the bottom),
and the addition of Rhodium acetate after 8h at 1000C (yellow line), and 18h at 1000C (blue line), and
after 48h at 1000C (purple line) in DMSO
8.4.8b 1H-NMR comparison (zoomed in aromatic region) of the building block (red in the bottom),
and the addition of Rhodium acetate after 8h at 1000C (yellow line), and 18h at 1000C (blue line), and
after 48h at 1000C (purple line) in DMF
8.4.9a 1H-NMR of building block 3 on THF zoomed in the aromatic region: 1H NMR (400 MHz, THF)
δ 6.38 (d, J = 1.8 Hz, 1H), 6.20 (dd, J = 7.8, 1.5 Hz, 1H), 5.99 (s, 1H), 5.95 – 5.88 (m, 3H), 5.67 (t, J = 7.7 Hz, 1H).
8.4.9b 1H-NMR of building block 3 on DMSO zoomed in the aromatic region: 1H NMR (400
8.4.10 1H-NMR comparison between the building block 3 in DMSO (green on top) and the solution where it was mixed with Zinc acetate (red in the bottom), showing the formation of a new species (zoomed in aromatic region). Below the comparison of their 1H-DOSY spectra is demonstrated.
8.4.11 1H-NMR of building block 4 on CDCl3 zoomed in the aromatic region: 1H NMR (400
MHz, CDCl3) δ 8.82 (s, 1H), 8.59 (d, J = 4.9 Hz, 1H), 7.87 (dt, J = 7.9, 1.9 Hz, 1H), 7.82 (d, J = 7.9 Hz, 1H),
7.77 (dd, J = 1.6, 0.8 Hz, 1H), 7.65 – 7.60 (m, 1H), 7.34 (dd, J = 7.9, 4.8 Hz, 1H).
8.4.12 1H-NMR cage attempt of M(2)L4 with platinum and building block 4 (zoomed in the aromatic
8.4.13 1H-NMR comparison between cage attempt of M(2)L4 with Rhodium acetate and building
block 3 (top blue) and the free building block 3 (red bottom) (zoomed in the aromatic region) No significant shifts are observed.
8.4.14 Summary of representative attempts for cage 3.
A variety of cage formations were attempted in different solvents, using different metals (even Rhodium since it is more robust than Zinc), and different conditions.
Table 5 Cage-synthesis attempts Entry Metal Precursor Building Block
Solvent Temperature 1H-DOSY
CA047 Rh(OAc)4 THF 800C 1.86nm
CA051 Zn(OAc)2 DMSO 800C 1.26nm
CA055 PdBF4 Acetonitrile 700C -
CA056 PdBF4 DMSO 700C -
CA057 PdBF4 THF 800C -
CA058 Rh(OAc)4 DMSO 1100C -
Note: In the column 1H-DOSY “-“ indicates multiple species impossible to identify (possibly polymers)
8.4.15 Cage 3
8.4.54a 1H-NMR comparison between cage attempt of M(2)12L24 with Zinc acetate and building
block 2 (bottom red) and the free building block 2 (green top) showing that the isophthalic acid peak disappeared
8.4.15b 1H-NMR comparison between cage attempt of M(2)12L24 with Zinc acetate and building
block 2 (bottom red) and the free building block 2 (green top). The singlet shows an uphill shift, and the doublet and the triplet show a downhill shift.
8.4.15c: 1H-NMR of 1,4-bis(4-pyridyl)benzene: 1H NMR (500 MHz, DMSO) δ 8.70 – 8.67 (m, 1H), 7.99
(s, 1H), 7.81 – 7.79 (m, 1H).
8.4.15d 1H-NMR comparison of different equivalents of 1,4-bis(4-pyridyl)benzene per cage. Free
1,4-bis(4-pyridyl)benzene (red), 0.50 equivalent per cage (green), 1 equivalent per cage (blue), 2 equivalents per cage (purple). If encapsulation would have occurred then blue and purple should have been different. It seems that 1,4-bis(4-pyridyl)benzene was coordinated to Zinc from the outside of
7.4 7.6 7.8 8.0 8.2 8.4 8.6 8.8 f1 (ppm) -200000 0 200000 400000 600000 800000 1000000 1200000 1400000 1600000 1800000 2000000 2200000 2400000 2600000 2800000 3000000 4 1 .9 0 8 .0 0 3 .9 6 4 .0 9 8 1 .8 4 8 .2 5 8 .0 0 4 8 .7 7 4 .0 1
8.4.15e Integration of peaks from figure 23d showing integrals very close to theoretical ratio of one
molecule of 1,4-bis(4-pyridyl)benzene affecting 8 ligands with 40 ligands remaining unaffected (thus coordinating to two M(2)12L24 cages)
8.5 1H-DOSY
Cage attempt, small M(2)12L24 compared with the free building block
Cage attempt, large M(2)12L24, Zn acetate in DMF, D value 9.706 radius 1.32, diameter 2.64nm (too small)
Cage attempt, large M(2)12L24, Rhodium acetate in DMSO, D value 9.938 radius around 1nm (too small)
Cage attempt, large M(2)12L24, Rhodium acetate in DMF, D value 9.687 radius around 1.3 nm (too small)
Cage attempt, M(2)2L4, Rhodium acetate in THF, diameter 1.86 nm (too small) (CA047)
