Respirable Fibre Detection from Light Scattering Patterns

Air-quality monitoring is essential in the reduction or prevention of potential respiratory problems and diseases. Many different instruments and techniques have been developed to detect and monitor different aerosol using light-scattering techniques in real time or close to it. The monitoring of respirable fibrous aerosol such as asbestos, however, is usually undertaken after filter collection taking around 24 hours for confirmation. Instruments have been developed for the detection of asbestos and other fibrous aerosol but perform poorly compared to filter analysis. Additionally, many of these instruments are costly or unavailable and mostly operate using ensemble scattering with additional methods other than the analysis of only scattered light. This project investigates the spatial light scattering from fibres and cylindrical objects using both modelled and captured data to develop single fibre measurement techniques using only forward scattered light. What was found is that the diameter, length, orientation, and refractive index of the scattering objects all contribute to changes in the forward scattered light that could be measured and used to determine the size. Different archetypical features of the scattering patterns were identified by analysing modelled data from varying input parameters, these features were then used as a reference for the measurement of other scattering patterns produced by fibrous aerosol. Due to the complexity of this problem, thorough investigations were limited to length and diameter measurements for fibres reasonably aligned with the incident radiation; orientation was investigated but not to the extent of the size measurements. Furthermore, fibre curvature was shown to affect the scattered light beyond the abilities of the developed measurement systems. The results indicated that a reasonably accurate measurement of the fibre could be made using only the forward scattered light. This was tested using both a simulated Monte-Carlo-based experiment and using data from captured scattering images and comparing the size distributions to microscopy data. Overall, both experiments showed positive results. The simulated Monte-Carlo-based experiment showed that most measurements produced a reasonable size, depending on refractive index: the lower the absorption coefficient, the further from the true size. Outliers were also identified as results of the reference measurement system. At the extreme limits of the reference data, the interpolation of the sizes resulted in obvious mismeasurements. The captured data and the microscopy data showed good visual comparisons in the size distributions. An alternative length measurement was required due the wild inaccuracy at first but showed a marked improvement when a new system was used. Additionally, some of the outliers in the modelled experiment reflected that in the aerosol experiment results. Inconsistencies were found in the size distribution comparisons for similar size ranges. This is an indication into the possibility of a low-cost monitoring system, more accessible and accurate than the current instruments, that could be developed using the research within this project.