Development of liquid crystal based sensors for the detection of toxic gases

Liquid crystal is an intermediate matter phase that between crystal solid and isotropic liquid phase. Liquid crystals are consisting of rod-shaped molecules. Liquid crystals are divided into two main groups as thermotropic, and lyotropic liquid crystals. They can be easily affected from external stimulants like heat, magnetic-electric field and solvents. Fast response rate of liquid crystals are enables them for using in sensor technologies. Especially in cholestric phase liquid crystals are known for fast optic output, because of their helical structures. In this context, it has the potential to be used for the detection of toxic volatile gases that are released uncontrolled by utilizing the optical properties of liquid crystal materials. In the fist half of this thesis work, glass coating processes has been done with polyvinyl chloride (PVC) polymer and dimethyloctadecyl[3-(trimethoxysilyl)propyl]ammonium chloride (DMOAP) coupling agent. For this surfaces, three different liquid crystal samples, thermotropic cholestric phase, blue phase and lyothropic cholesteric phase, has been prepared. Liquid crystal samples are injected in 20 µm thick glass cells that coated with PVC and DMOAP, which is prepared separately, and investigated under polarizing optical microscope (POM) for their structural characteristics, as a function of temperature. In the second half of this thesis work, prepared liquid crystals samples are exposed to phenol, toluene and 1,2 dichloropropane toxic volatile organic solvent vapors, isolated from outdoor environment with the isolated box, as the function of heat and time. In this process, wavelength changes in liquid crystals are observed through with a spectrophotometer. In the last stage of the thesis, the relationship between the inputs and outputs of the system was determined by factor analysis of the experimental data obtained with the fuzzy logic approach. The inference rules of the system were created by defining the membership function for each input and output. The developed fuzzy inference was made in Matlab program.