Investigation of Surface Defects in Optical Components Based on Reflection Mueller Matrix Spectroscopy

Round 1

Reviewer 1 Report

The authors present an interesting and detailed study about using the bidirectional reflection distribution functions (BRDF) and the Rayleigh-Rice scattering theory to identify different types of defects on the surface of optical elements. The theoretical simulation results are compared against experimental measurements. The findings of the paper support its conclusion. 

The paper could benefit from a number of structural and stylistic change, as well as the presentation of the results could be improved. 

 

My comments for the Figures are: 

 

Figures 1, 2, 3 and 4 could include information about parameters denoted in the diagram, such as n1, n2, thetha_s, thetha_i. For Figures 3 & 4, I suggest defining parameter "d" which refers to the distance of the defect from the surface. Similarly, I would recommend consistent use of parameter "a" to describe the radius of the defect, both for SSBD and dirt particle. For example, Eq (9) has parameter "a" in it which is not explained. 

 

Figures 5 & 6 and 7a): 

The suffixes ss, sp, ps, and pp probably refer to light polarisation and should be explained. The font on the axes scales and the colour legend should be bigger, they are hardly visible at the moment. Also perhaps one could consider changing the colour of the labels as black is not visible on dark blue. Figure 7a) uses sub-notation i), ii) etc. which could be implemented in Figures 5 & 6 for consistency. This would make it easier to describe the results in the caption also. 

 

Figure 7b) The legend uses an undefined parameter "alpha". This could be the same as parameter "a" which refers to the radius of the defect. If so, it needs to be corrected. There is a strange label of X and Y  in the middle of the graph, which does not appear to be relevant - I would suggest to remove it, or explain its significance. 

 

Table 1: There is a strange font for micrometers, I suggest adjusting it. 

 

Figures 8, 11, 12, 13: The axes need to be labelled. I guess the horizontal axes is the wavelength in nm of light wavelength. Many of these panels show flat lines, which is not informative. I would recommend adjusting the vertical axis so that the panels are more detailed (for example, as was done with Figure 8, panels M33 and M43). Please also move the legend outside the panels, as they apply to all panels. Figure 12 requires more detail to describe the difference between 12 a) and 12 b). The description for 12 a) and 12 b) in the caption is identical, but the panels are different. I would also suggest using a double-y axis plots for the panels themselves, where one y-axis contains the simulation and experimental results and the second y-axis shows the difference. This may allow a better presentation of the data. Figure 13 needs more explanation as well. Perhaps, dashed and dotted lines could be used for Figure 13, to avoid any comparison to results in Figure 12. Figure 13 legend should say: Dust (simulation), SSBD (simulation). I assume "dust" and "dirt" refers to the same defect, in which case it should be consistent throughout.  

 

 

Figure 9: 9a) indicate the x-y scale and size of the imaged area. What is the difference between area 1 and area 2? 9b) is a very unclear image. The scale is not visible, and the red line markings are too bunched together. I recommend zooming in into the shown trace to make the vertical variation more clear.  There are markings [1] and [2] in the figure which don't seem to have relevance. 

 

 

Figure 10: Define PSA and PSG in the caption. 

 

 

Figure 14: Explain in the caption the parameters of the simulation, such as the radius of the dust particle and its distance from the surface. Bigger font on the scale and axes labels and an explanation of the colour chart is needed also - does this refer to intensity? 

 

I would recommend the paper for acceptance, provided the issues I have raised are suitably addressed. 

For the text of the main paper: 

There are several locations in the text where the sentences are very long and hard to follow. I would recommend breaking them up in shorter sentences. Specifically, the following areas could be improved grammatically: 

page lines: 54-59

page line: 59-65

page line: 75-81

page line: 81-87 actually has a repeating fragment which does not make sense "Mueller matrix not only reflects the parameters of the target defects such as the surface" - this should be clarified

 

page line: 173 - use "of the surface are not the same" instead of "can't" 

 

page line: 186 - 191 could be split into two sentences

 

page line: 203 - 209 could be split into several sentences for clarity

 

page line: 220 a sentence can't be started with "Where"

 

page line: 224 - 230 could be split into two sentences

 

page line: 235 could start a new sentence "This study analyses..."

 

page line: 240 a sentence can't be started with "And"

 

page line: 250 "a particle size of a= 5 nm"

 

page line: 259 please describe the differences in the panels more clearly, avoid the use of the term "obvious". Explain the suffices ss, sp, ps and pp. 

 

page line: 342 - 343 "of this study" is repeated twice in one sentence and generally that sentence is hard to read. 

 

Use parameter "d" to describe the depth of the defect. 

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This manuscript proposes to use pBRDF for identifying the type of defect with the dimension smaller than the wavelength on the optical surface. A full simulation has been made and the result shows the effectiveness of the proposed concept. It would a good reference for the optical surface defect detection and this manuscript could be considered for the publication in Applied Science. A minor suggestion is given as follows for the authors’ reference.

 For the calculation of optimal test conditions, the BRDF of SSBD has been used for finding the best incident angle. Unlike the other cases where there are only specular reflection at the optical surface, the SSBD case has got the light penetrating into the optical surface and its BRDF could be largely different from its pBRDF. It is suggested to put more description on how the BRDF of SSBD case is made, and why the optimal test conditions from the simulation can be valid.

A few typos in the manuscript need to be corrected, and the most obvious one would be in line 82-83.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf