In this study a number of cage alkenes were synthesised and tested for activity towards ringopening metathesis polymerisation (ROMP) with the commercially available catalysts 55 (Grubbs-I) and 56 (Grubbs-II). The first group of monomers are derivatives of tetracyclo[6.3.0.04,1105,9]undec-2-en-6-one (1). The synthesis of these cage alkenes are summarised in Scheme 7.1. The cage alkene 126b was synthesised by a Diels-Alder reaction between 1 and hexachlorocyclopentadiene (9, Scheme 7.2). The geometry of 126b was determined from XRD data. Knowledge of the geometry of 126b also established the geometry of 127 since conformational changes during the conversion from 126b to 127 are unlikely. Synthesis of the cage alkene 125 by the cycloaddition of 9 to 118 failed. The cage alkene exo-11- hydroxy-4,5,6,7,16,16-hexachlorohexacyclo[7.6.1.03,8.02,13.010,14]hexa-dec-5-ene (124, Scheme 7.3) could therefore not be prepared. Synthesis of 125 by reduction of 126b with various reduction systems was not successful. Theoretical aspects of these reactions were investigated with molecular modelling. A possible explanation for the unreactive nature of 126b towards reduction is presented, but the lack of reactivity of 118 towards 9 eluded clear explanations. The synthesis of cage alkenes from 4-isopropylidenepentacyclo[5.4.0.02,6.03,10.05,9]-undecane-8,11- dione (23) did not meet with much success (Scheme 7.4). Numerous synthetic methods were investigated to affect the transformation from 134a/134b to 135 (Scheme 7.5). These attempts evolved into theoretical investigations to uncover the reasons for the observed reactivity. Possible explanations were established by considering the differences and similarities between the geometries and electronic structures of reactive and unreactive cage alcohols. ROMP of cage monomers based on 1 were mostly unsuccessful. Only the cage monomer 127 showed some reactivity. Endocyclic cage monomers with a tetracycloundecane (TCU) framework showed no reactivity. The results from NMR experiments verified the experimental results. Hexacyclo[8.4.0.02,9.03,13.04,7.04,12]tetradec-5-en-11,14-dione (3) exhibited notable ROMP reactivity. Examination of the orbitals of the cage alkenes used in this study suggested that the reactivity of 1 and 3 could possibly be enhanced by removal of the carbonyl groups. Decarbonylation of 1 and 3 yielded the cage hydrocarbons 159 and 175, respectively. ROMP tests revealed that 175 is an excellent monomer, but 159 was unreactive. The results obtained for the ROMP reactions in this study was rationalised by considering aspects such as ring strain, energy profiles, steric constraints, and frontier orbital theory. The concept of ring strain is less useful when describing the reactivity of cage alkenes towards ROMP and therefore the concepts of fractional ring strain and fractional ring strain energy (RSEf) were developed. A possible link between RSEf and the ROMP reactivity of cage alkenes was also established. The following criteria were put forth to predict the reactivity or explain the lack of reactivity of cage alkenes towards ROMP reactions with Grubbs-I and Grubbs-II. The criteria for ROMP of cage monomers: 1. Sufficient fractional ring strain energy (RSEf). 2. A reasonable energy profile when compared to a reference compound such as cyclopentene. 3. Ability to form a metallacyclobutane intermediate with reasonable distances between different parts of the cage fragment. 4. Sufficient ability of the polymer fragment to take on a conformation that exposes the catalytic site. 5. Sufficient size, shape, orientation and energy of HOMO and/or NHOMO at the alkene functionality of the cage monomer and of the LUMO at the catalytic site. / Thesis (Ph.D. (Chemistry))--North-West University, Potchefstroom Campus, 2012

Student Number : 0002306V - MSc dissertation - School of Chemistry - Faculty of Science / The first part of the dissertation involves the use of ring closing metathesis (RCM) and ruthenium mediated isomerisation-RCM tandem reactions to form a wide range of nitrogencontaining benzo-fused heterocycles. Those synthesized include the 6-membered isoquinolines, the 7-membered benzazepines and the 8-membered benzazocines. In order to put these compounds into perspective, a review of selected naturally occurring nitrogencontaining benzo-fused heterocycles is included along with some of their synthetic approaches. Of major significance is our utilization of the Wits methodology allowing one to access the 6-, 7- and 8-membered ring systems from a common synthetic intermediate. The 1,2,3,6-tetrahydro-2-benzazocines were all obtained after RCM in excellent yields (82-99%). We were also able to show that some ofthe protecting groups used were easily removed and that the ring could be hydrogenated after RCM to yield the 1,2,3,4,5,6-hexahydro-2- benzazocines. The isoquinolines were synthesized in 78% and 27% yield for the Ac- and Tsprotected compounds respectively, with no product isolated for the Boc- or SO2Bn-protected compounds. These poor results, caused a change to our strategy and we then used a “combinatorial-type” approach for the synthesis of the 2,5-dihydro-1H-2-benzazepines and the 2,3-dihydro-1H-2-benzazepines with yield of 9, 47, 58 and 82% and 8, 26, 39 and 82% obtained respectively for the RCM reaction Futhermore, we attempted the synthesis of the substituted 4-phenyl isoquinolines and 5-phenyl benzazepines, but we found that the systems would not undergo RCM even at high temperatures and with large amounts of Grubbs II metathesis catalyst. A short review is given in the second part of the dissertation concerning the naturally occurring and pharmaceutically useful indenols, indenones and indanones. It further highlights how our methodology was extended to include the synthesis of 4-isopropoxy-5- methoxy-1H-inden-1-ol (X), 4-isopropoxy-5-methoxy-1H-inden-1-one (X) and 4-isopropoxy- 5-methoxy-1H-indanone (X) through the use of ruthenium-mediated isomerisation and RCM from a similar common intermediate. We have shown the synthesis of 3-substituted indenols, indenones and indanones using the same synthetic procedure, but by changing the reaction temperature during RCM. This dissertation also answers many of the questions posed during the post-doctoral work of Coyanis. Namely, we were able to support our proposed mechanism that the conversion of the unsubstituted indenol to the indenone was occurring via a dehydrogenative-oxidation, through the use of 1H NMR studies that were coupled with an ICP-MS analysis. To the best of our knowledge, this is the first reported use of the Grubbs II catalyst (or its degradation products) in a tandem RCM-oxidation procedure by our group recently.

In this study a number of cage alkenes were synthesised and tested for activity towards ringopening metathesis polymerisation (ROMP) with the commercially available catalysts 55 (Grubbs-I) and 56 (Grubbs-II). The first group of monomers are derivatives of tetracyclo[6.3.0.04,1105,9]undec-2-en-6-one (1). The synthesis of these cage alkenes are summarised in Scheme 7.1. The cage alkene 126b was synthesised by a Diels-Alder reaction between 1 and hexachlorocyclopentadiene (9, Scheme 7.2). The geometry of 126b was determined from XRD data. Knowledge of the geometry of 126b also established the geometry of 127 since conformational changes during the conversion from 126b to 127 are unlikely. Synthesis of the cage alkene 125 by the cycloaddition of 9 to 118 failed. The cage alkene exo-11- hydroxy-4,5,6,7,16,16-hexachlorohexacyclo[7.6.1.03,8.02,13.010,14]hexa-dec-5-ene (124, Scheme 7.3) could therefore not be prepared. Synthesis of 125 by reduction of 126b with various reduction systems was not successful. Theoretical aspects of these reactions were investigated with molecular modelling. A possible explanation for the unreactive nature of 126b towards reduction is presented, but the lack of reactivity of 118 towards 9 eluded clear explanations. The synthesis of cage alkenes from 4-isopropylidenepentacyclo[5.4.0.02,6.03,10.05,9]-undecane-8,11- dione (23) did not meet with much success (Scheme 7.4). Numerous synthetic methods were investigated to affect the transformation from 134a/134b to 135 (Scheme 7.5). These attempts evolved into theoretical investigations to uncover the reasons for the observed reactivity. Possible explanations were established by considering the differences and similarities between the geometries and electronic structures of reactive and unreactive cage alcohols. ROMP of cage monomers based on 1 were mostly unsuccessful. Only the cage monomer 127 showed some reactivity. Endocyclic cage monomers with a tetracycloundecane (TCU) framework showed no reactivity. The results from NMR experiments verified the experimental results. Hexacyclo[8.4.0.02,9.03,13.04,7.04,12]tetradec-5-en-11,14-dione (3) exhibited notable ROMP reactivity. Examination of the orbitals of the cage alkenes used in this study suggested that the reactivity of 1 and 3 could possibly be enhanced by removal of the carbonyl groups. Decarbonylation of 1 and 3 yielded the cage hydrocarbons 159 and 175, respectively. ROMP tests revealed that 175 is an excellent monomer, but 159 was unreactive. The results obtained for the ROMP reactions in this study was rationalised by considering aspects such as ring strain, energy profiles, steric constraints, and frontier orbital theory. The concept of ring strain is less useful when describing the reactivity of cage alkenes towards ROMP and therefore the concepts of fractional ring strain and fractional ring strain energy (RSEf) were developed. A possible link between RSEf and the ROMP reactivity of cage alkenes was also established. The following criteria were put forth to predict the reactivity or explain the lack of reactivity of cage alkenes towards ROMP reactions with Grubbs-I and Grubbs-II. The criteria for ROMP of cage monomers: 1. Sufficient fractional ring strain energy (RSEf). 2. A reasonable energy profile when compared to a reference compound such as cyclopentene. 3. Ability to form a metallacyclobutane intermediate with reasonable distances between different parts of the cage fragment. 4. Sufficient ability of the polymer fragment to take on a conformation that exposes the catalytic site. 5. Sufficient size, shape, orientation and energy of HOMO and/or NHOMO at the alkene functionality of the cage monomer and of the LUMO at the catalytic site. / Thesis (Ph.D. (Chemistry))--North-West University, Potchefstroom Campus, 2012

Ring closing metathesis (RCM) is a chemical transformation that converts a bisalkene compound into a cycloalkene. It is catalyzed by transition metal complexes containing carbene ligands (that feature metal-carbon double bonds). The mechanism is well-understood, however, there are numerous details of the reaction that are less well understood, especially concerning catalyst activation and decomposition and formation of byproducts. This thesis takes a new approach to the study of RCM: analysis of the reaction using real-time mass spectrometric techniques. Electrospray ionization (ESI) mass spectrometry was employed in this study, and the real-time aspect was enabled by using pressurized sample infusion (PSI). Observation of the reactants and products was enabled using charge-tagged bis-alkenes of the general formula [Bu2N{(CH2)nCH=CH2}2]+ [PF6]–. These were synthesized in two steps using a generally applicable methodology to generate a wide range of ring sizes of the product, from 5- to 15-membered rings. Examination of their behavior under carefully optimized RCM conditions using Grubbs’ second-generation catalyst showed a wide variation in reaction rates and amount of byproducts, largely due to ring-strain effects (especially high for 5- and 9-membered rings). Byproducts always exhibited a 14 Da mass unit difference from starting materials or products, and Orbitrap MS analysis confirmed it was CH2. Isomerization was suspected to lead to byproducts. A pathway for byproducts via isomerization and cross metathesis was proposed. The source of actual isomerization catalyst was believed to be from the precatalyst itself as the evidence of precatalyst decomposition was observed. Finally, to prove our isomerization hypothesis, an authentic isomerization catalyst was deliberately added into a fast and clean reaction along with Grubbs’ second-generation catalyst, and it produced the expected byproducts. Only small amounts of oligomeric intermediates were observed, probably because of the low concentrations used. [ClPCy3]+ was a new short-lived decomposition product stemming from catalyst breakdown, along with already-known imidazolium and protonated phosphine decomposition products. Overall, the thesis provides deep new insights into the nature of RCM reactions, in particular revealing the importance of isomerization in RCM reactions that are slow due to ring strain effects and in uncovering a new decomposition pathway for important RCM catalysts. / Graduate