Dendrimer Sensors for Explosives

In the current world political climate, rapid and reliable detection of explosive compounds has become increasingly important. While there are many different types of explosive, nitroaromatic explosives are amongst the most common. 2,4,6-Trinitrotoluene (TNT) is arguably one of the most well known explosives in the world. It is one of the most important explosives for blasting charges of all weapons and is still used as a major component of anti-personnel landmines. Many different technologies exist for the detection of explosives including: trained animals, such as sniffer dogs; X-ray and neutron imaging; chromatographic separation and detection of mixtures of compounds; mass-based detection, such as ion mobility spectrometry (IMS); and various spectrometric and colorimetric techniques. However, while many of these technologies are highly sensitive in a laboratory setting they often lack the portability and robustness required to develop a hand-held device for the real-time detection of explosive vapours. Oxidative fluorescence quenching, in which the fluorescence of a sensing material is quenched by a high electron affinity explosive analyte, offers a potential route to such portable devices. Over recent years fluorescent sensors for explosives have become a popular area of research. However, there are still significant improvements to be made in order to allow the technology to be implemented in suitably portable and reliable real-world devices. This thesis investigates fluorescent conjugated dendrimers as new materials for the detection of trace quantities of explosives, and aims to provide an understanding of some of the interactions and structure-property relationships that are involved in fluorescent sensors for explosives. The convergent synthesis of two first-generation, fluorescent conjugated dendrimers is described. The dendrimers comprise bifluorene-based chromophores, biphenyl-based dendrons and 2-ethylhexyloxy surface groups. The first dendrimer (D1) is relatively planar with only one chromophore at its core, while the second (D2) has four of the bifluorene-based chromophores held in a three-dimensional arrangement around an adamantane centre. The dendrimers, along with the second-generation variant of D1 (D3) were tested as sensors for explosives in solution. Nitroaromatic, nitroaliphatic and non-nitrated aromatic control analytes were studied. Steady-state Stern-Volmer experiments showed that D2 had higher quenching constants with all of the analytes tested than the more planar dendrimers. However, time-resolved Stern-Volmer measurements showed that the mechanism of quenching (static or collisional) was invariant with the different dendrimers and instead depended on the structure of the analyte. The more aromatic rings present in the analyte, the higher the static component of the quenching. Finally, fluorescence depolarisation anisotropy measurements showed that the greater quenching ability of D2 was likely caused by an amplification effect of having multiple emissive chromophores per dendrimer. Interaction of one analyte with one chromophore could quench an exciton formed on any of the four chromophores. The dendrimers were all solution-processed into thin films, which were suitable for detection of explosive vapours. D1 and D2 were exposed to atmospheres saturated with analyte vapours and the fluorescence intensity monitored. In general D2 was quenched more rapidly than D1 with the desired analytes and had reduced signals from the control analytes. A pronounced difference was seen in the rates of fluorescence recovery after the analyte was removed with the fluorescence of D1 recovering intensity significantly faster than D2. The quenching rates for D1 and D2 were comparable to or better than those reported in the literature for the same analytes. Films of D1, D2 and D3 were then tested with the transient exposure quenching method, in which 1 second exposures of analytes, at sub-saturated concentrations in a carrier gas, were used to quench the fluorescence of the films. Under these conditions it was found that all three dendrimers underwent similar levels of quenching and D2 still showed the slowest recovery of fluorescence intensity. Importantly, a Stern-Volmer-like analysis of the results showed all three dendrimers were more sensitive to nitroaromatic analytes than the non-nitrated controls. For the three dendrimers, the detection limit for a 1 second exposure to 2,4-dinitrotoluene was less than 1 part per billion. Finally, neutron reflectometry (NR) was used to probe the distribution and concentration of the analytes within films of D1 and D2. It was found that, upon exposure to the analytes, films of both dendrimers swelled to incorporate the nitroaromatic analyte molecules, which were able to penetrate throughout the dendrimer films. Also, films of D2 sequestered approximately twice as much of these analytes as films of D1. The concentration of analyte molecules in the films was calculated and it was shown that one analyte was able to quench more than one chromophore. Comparison of the solid-state quenching and the NR results are consistent with the more open film morphology of D2 leading to reduced inter-dendrimer interactions. Consequently, ingress of the analytes disrupts the dendrimer-dendrimer pi-stacking in the films of D2 less than those of D1, hence the slower fluorescence recovery and rate of analyte removal for D2 compared to D1