User Tools
docs:become
Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
|
docs:become [2020/04/22 22:08] pklapetek |
docs:become [2020/04/22 22:58] pklapetek |
||
|---|---|---|---|
| Line 99: | Line 99: | ||
| even if they are not infinitely periodic. | even if they are not infinitely periodic. | ||
| A bigger problem is that the diffracted intensity is about 0.147 of the incident intensity, | A bigger problem is that the diffracted intensity is about 0.147 of the incident intensity, | ||
| - | which is less than expected (the expected value is 0.178). Something has to be wrong. | + | which is less than expected (the expected value is 0.186). Something has to be wrong. |
| The first suspect is the refractive index. | The first suspect is the refractive index. | ||
| Line 111: | Line 111: | ||
| * default metal setting: 6 0.89583 0 13.8737e15 0.0207332e15 1.3735 -0.504659 7.59914e15 4.28431e15 0.304478 -1.48944 6.15009e15 0.659262e15 means n=(0.036 + 4.147i) and leads to first order diffraction intensity of 0.150. | * default metal setting: 6 0.89583 0 13.8737e15 0.0207332e15 1.3735 -0.504659 7.59914e15 4.28431e15 0.304478 -1.48944 6.15009e15 0.659262e15 means n=(0.036 + 4.147i) and leads to first order diffraction intensity of 0.150. | ||
| - | Another suspect is the voxel spacing. However, at least for 5 nm and 10 nm voxel spacing the resulting diffraction maximum was nearly same (less than 0.5 percent difference). | + | The easiest way how to check the metal properties is to calculate the reflectance of a bulk. |
| + | This uses very similar geometry to our calculation (TSF source, periodic and CPML boundaries), | ||
| + | so we can expect that there are not many additional potential error sources when comparing to our calculation. | ||
| + | |||
| + | First of all, error in the reference value (PEC reflection) is below 0.001 percent. This certainly | ||
| + | cannot affect our results. | ||
| + | Reflectance of the default metal for this particular voxel spacing and other settings is 0.983 | ||
| + | Reflectanceof the fitted metal model is 0.9587. | ||
| + | Result from the Filmmetrics reflectance calculator is 0.9678, which is something in between. | ||
| + | This is probably also not the source of problem (however might be compared to reflectance | ||
| + | coming from FEM). | ||
| + | |||
| + | Another suspect is the voxel spacing. In the first tests for 5 nm and 10 nm voxel spacing the resulting diffraction maximum was nearly same when studied on PEC (less than 0.5 percent difference). In past we have however observed that the voxel spacing has an impact on the reflectance on real metals (not on PEC, which is ideal). | ||
| + | Too coarse mesh cannot treat the metal correctly, probably due to its small skin depth. | ||
| + | Here, in a quick check, for the fitted model we get reflectances of 0.959 for 10 nm voxel size, 0.962 for 5 nm voxel size and 0.958 for 2 nm voxel size, when roughly evaluated from the local fields. This means differences up to 1 percent. However, the difference between the expected value and the simulated value is about 20 percent. So, reflectance itself is probably also not the source of troubles. | ||
| + | |||
| + | The voxel size could impact the overall performance in more general way than only to wrong | ||
| + | reflectance. The very preliminary tests (with less accurate boundaries and other imperfections) however did not show large dependence on the voxel size, at least for reasonable voxel sizes, in the range of 3-12 nm. However, this could be still studied with the present setup. | ||
docs/become.txt · Last modified: 2020/04/24 12:27 by pklapetek
Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Noncommercial 4.0 International
