UC Berkeley

Employing synthetic modularity in organic molecules is an under-utilized tool for tuning the intricately connected reaction dynamics and/or characteristics of a chemical application of intended practice. Two applications in which the synthetic modularity of the respective substrates is ripe for exploitation are ring-opening alkene and alkyne polymerization (ROMP and ROAMP, respectively), and molecular recognition of carboxylates and nitro-compounds.

Living ROMP and ROAMP have the potential to create polyphenylenevinylenes (PPVs) and polyphenyleneethynylenes (PPEs) with well-defined, structurally predictable, defect-free properties–polymer attributes that are not as feasible via other polymerization methods. These types of polymers are useful for important electronic applications and mimicking the dynamics of biological polymers. However, in order for ROMP and ROAMP to become viable methods for accessing these polymers, a better understanding of their respective mechanisms and a greater substrate scope is required. Obtaining access to monomers with tunable electronic and structural properties opens this possibility of illuminating mechanistic characteristics and delivering a library of new conjugated polymers.

Synthesis of (2Z,5Z,8Z)-15,45,75-tribromo-1,4,7(1,3)-tribenzenacyclononaphane-2,5,8-triene (89), a template molecule for a library of ROMP and ROAMP monomers, was achieved in only 5 steps with 4 % overall yield. From this template molecule, two trienes, alkyl-substituted 107 and amide-substituted 108, were synthesized for ROMP and one triyne, alkyl-substituted 113, synthesized for ROAMP. ROMP of 107 and 108 with Grubb’s II (70) and Grubb’s III (71) catalysts resulted in both cyclic and linear PPVs of < n = 3. After long reaction times, the polymerizations with Grubb’s III resulted in a cyclic dimer that is also a potential precursor toward nanographenes.

ROAMP of 113 with molybdenum catalyst 78 was a much more selective polymerization resulting in linear conjugated PPEs. Kinetic studies revealed that thepropagating species emerging from catalyst 78 underwent some termination processes toward the end of the polymerization, but PDIs were still quite low (< 1.15) and in the absence of monomer, the molybdenum catalyst attached to the propagating polymer chain remained active and continued to incorporate equivalents of monomer added sequentially to the reaction mixture. The ring-opening alkyne metathesis polymerization with 78 has most of the characteristics of a living polymerization and enables, for the first time, a near-living polymerization toward conjugated PPEs. UV-Vis studies revealed that with the sequential addition of guest molecule, RDX, the PPE polymers underwent a secondary structural transition to a helical formation – a transition characteristic of biological polymers.

Enhancing the molecular recognition of carboxylates and nitro groups is vital for sensing of biological molecules and explosives, respectively. Current methods for recognition of explosives rely on π-π interactions or indirect methodologies that preclude the sensitive and specific sensing of non-aromatic explosives like RDX. Hydrogen-bond recognition of bidentate nitro-compounds has the potential to allow for specific sensing via a method that detects RDX or other non-aromatic explosives for their unique structural properties. Thus we designed a synthetically modular molecular receptor with tunable electronics, flexibility and hydrogen-bond donating ability in order to optimize binding to bidentate guests such as isophthalate and RDX.

Five sensors of varied flexibility, electronics and hydrogen-bond donors were synthesized. In order to optimize binding to RDX, binding studies were first conducted with isophthalate, a likely stronger binding partner than RDX, to probe the right set of characteristics for optimal sensing. Sensor R-EW-S exhibited the strongest binding, according to NMR titration experiments, to isophthalate (Ka ~ 106 M-1) – binding comparable to the best sensors for isophthalate. Comparison to the four other sensors, accessed from a modular synthetic route, revealed that this is likely due to its rigid preorganization, its thiourea hydrogen-bond donor, and electron-withdrawing nature. Binding experiments to RDX resulted in weaker binding to the array of sensors; NMR studies with F-EW-S demonstrated the strongest binding (~ 400 M-1) to RDX, perhaps due to its ability to create an induced fit to RDX (which is more structurally fluxional than isophthalate). However, fluorescence titration studies of sensor R-ED-O with RDX led to an estimate of Ka ~ 4000 M-1. This was the first example of a sensor exhibiting any appreciable binding via hydrogen-bonding to RDX.