Optical Simulations of Face Mask Materials using Forward Ray Tracing
Mangs, Oliver (2022)
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi-fe2022052037685
https://urn.fi/URN:NBN:fi-fe2022052037685
Tiivistelmä
In this thesis, I will study the feasibility of using UV light to decontaminate face masks. The process has been experimentally proven to be possible by Titta Kiiskinen. In her thesis, Kiiskinen explains that the mechanism of decontamination is the absorption of UV light in the microbe's DNA. The wavelength at which this absorption is most likely to happen is 280 nm.
The simulations were performed using forward ray tracing in software called TracePro. Ray tracing is a suitable option for optical characterisation of the face masks as it does not solve Maxwell's equations. It relies on classical optics theory, such as Fresnel's equations and the Beer-Lambert law to calculate the reflection and refraction at a surface.
To obtain a 3D model of the face mask structure, it was imaged using x-ray tomography. The smallest structure in a tomographic image is a single voxel which, in the case of the surgical face mask, has a width of approximately 3.8 micro m. There were not any issues with phenomena such as diffraction since the voxel is around 10 times larger than the wavelength of light used in this study.
The tomographic images were obtained in a TIF stack format. This format is comprised of several 2-dimensional images. To turn the images into a 3D model several software was used. ImageJ Fiji was used to convert from a TIF stack to a 3D surface mesh which was exported in STL format. Meshlab was used to smooth out rough surfaces caused by the voxel-based TIF stack. The final step was to convert from a surface mesh to a solid 3D object. This was done using Fusion 360 and several tools within it. After converting the surface mesh into a solid model, it was exported in ACIS format.
The ACIS file was loaded into TracePro and a simulation setup was built around it. The simulation setup consists of a mirror tube, light source, reflection and transmission boxes and the 3D model of the face mask. The mirror tube was placed such that the edges do not touch the model. The purpose of the mirror tube was to reflect any light escaping from the edges of the model back into the model. By doing this, the light has to travel through the model or reflect towards the light source. At each end of the mirror tube, there is a box. Each box has had the perfect absorber property applied to it. When the light has escaped through either end of the mirror tube, it is captured by the boxes. The absorbed flux can then be viewed to measure the reflectance and transmittance of the model.
The face masks were studied experimentally by measuring
reflectance- and transmittance spectra using OceanOptics2000+ and Cary5000 spectrophotometers. It was found that the outer layers of a surgical face mask contain whiteners and dyes which interfere with the measurements. The face mask was cut open and the measurements were completed only on the middle layer. The absorption was found to be zero above 300 nm which meant that the coefficient of extinction could be set to zero.
The simulations were first performed on the middle layer of the face mask at 450 nm with the coefficient of extinction set to zero. The refractive index was varied over several simulations. One can see that the results are highly independent of the refractive index and an unrealistically high refractive index would have been needed to meet the measured reflection and transmission values. Therefore, the value for the refractive index was chosen to be 1.9 as it was found in the literature. The same simulations were then performed at 280 nm with a fixed refractive index and a varying coefficient of extinction. The optimal value for the coefficient of extinction was found to be 4e-6.
The complete face mask structure was then imported into TracePro and the simulations were performed on it. The absorption profile was obtained using the volume flux tool. We found that absorption happens throughout the entire face mask and that there are three drops in the remaining light. Each of the drops corresponds to a layer in the face mask. Therefore, the light penetrates the entire face mask.
The simulations were performed using forward ray tracing in software called TracePro. Ray tracing is a suitable option for optical characterisation of the face masks as it does not solve Maxwell's equations. It relies on classical optics theory, such as Fresnel's equations and the Beer-Lambert law to calculate the reflection and refraction at a surface.
To obtain a 3D model of the face mask structure, it was imaged using x-ray tomography. The smallest structure in a tomographic image is a single voxel which, in the case of the surgical face mask, has a width of approximately 3.8 micro m. There were not any issues with phenomena such as diffraction since the voxel is around 10 times larger than the wavelength of light used in this study.
The tomographic images were obtained in a TIF stack format. This format is comprised of several 2-dimensional images. To turn the images into a 3D model several software was used. ImageJ Fiji was used to convert from a TIF stack to a 3D surface mesh which was exported in STL format. Meshlab was used to smooth out rough surfaces caused by the voxel-based TIF stack. The final step was to convert from a surface mesh to a solid 3D object. This was done using Fusion 360 and several tools within it. After converting the surface mesh into a solid model, it was exported in ACIS format.
The ACIS file was loaded into TracePro and a simulation setup was built around it. The simulation setup consists of a mirror tube, light source, reflection and transmission boxes and the 3D model of the face mask. The mirror tube was placed such that the edges do not touch the model. The purpose of the mirror tube was to reflect any light escaping from the edges of the model back into the model. By doing this, the light has to travel through the model or reflect towards the light source. At each end of the mirror tube, there is a box. Each box has had the perfect absorber property applied to it. When the light has escaped through either end of the mirror tube, it is captured by the boxes. The absorbed flux can then be viewed to measure the reflectance and transmittance of the model.
The face masks were studied experimentally by measuring
reflectance- and transmittance spectra using OceanOptics2000+ and Cary5000 spectrophotometers. It was found that the outer layers of a surgical face mask contain whiteners and dyes which interfere with the measurements. The face mask was cut open and the measurements were completed only on the middle layer. The absorption was found to be zero above 300 nm which meant that the coefficient of extinction could be set to zero.
The simulations were first performed on the middle layer of the face mask at 450 nm with the coefficient of extinction set to zero. The refractive index was varied over several simulations. One can see that the results are highly independent of the refractive index and an unrealistically high refractive index would have been needed to meet the measured reflection and transmission values. Therefore, the value for the refractive index was chosen to be 1.9 as it was found in the literature. The same simulations were then performed at 280 nm with a fixed refractive index and a varying coefficient of extinction. The optimal value for the coefficient of extinction was found to be 4e-6.
The complete face mask structure was then imported into TracePro and the simulations were performed on it. The absorption profile was obtained using the volume flux tool. We found that absorption happens throughout the entire face mask and that there are three drops in the remaining light. Each of the drops corresponds to a layer in the face mask. Therefore, the light penetrates the entire face mask.
Kokoelmat
- 114 Fysiikka [25]