Constructing a Micro-Raman Spectrometer with Industrial-Grade CMOS Camera—High Resolution and Sensitivity at Low Cost

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Sir / Madam,

I have the following suggestions to improve the clarity and quality of your paper:

A. Regarding Introduction:

1. On page 2 of the introduction there is a paragraph in which the solution to the problem is presented by referencing to the supplementary material. After this paragraph the issue regarding high spectral resolution is further introduced.

2. At the end of page 2 you presented the decision to limit spectrometer range in visible.

3. After this paragraph you present the issues regarding lens.

4. The first paragraph on page 3 present the CMOS camera aspect.

5. After citing references 10, 14, 15 you present the primary motivation behind building the spectrometer.

It is recommended that before you start these paragraphs to introduce the reader to the purpose of the paper by replacing the paragraph discussed at point 1 with the Aim of the paper.

Or you can use them in corresponding sections.

6. The last two phrases which end the introduction are more suitable for conclusions due to the fact that the reader does not know the results of your work.

B. Section 2. Micro-Raman Spectrometer construction

1. It started by presenting spectrometer setup and the functionalities of the two blocks contained.


2. In the next paragraph is presented the motivation behind choosing the interchangeable cameras. It is advisable to present this issue in "The High resolution lens based spectrometer" section or in the introductory section as you have chosen. The motivation is that the two components of the spectrometer were not presented.

3. The subsections are not numbered.

B.1. Micro-Raman Probe

1. This paragraph start with laser presentation. It is recommendable to place the characterization of the laser from the supplementary material in the paper. In the interpretation of the results a single Gaussian function was used.

In the last parapgraph the reader is introduced to results obtained further in the paper which is not connected to the purpose of this subsection.

2. Micro-Raman Probeworking principle description must be referenced to figure 1

3. You have cited reference 16 and here their system can access wavenumbers in the ultralow-frequency (ULF) region, down to 10 cm⁻¹. Please comment how did you achieve wavenumbers larger or equal to 5 cm-1.


4. Regarding spot size and Wslt calculations it is recommended to place them in the paper.

B.2. High-resolution lens-based spectrometer

1. A comparison with the Raman Spectrometer setup from your previous paper (reference 22) should also be added.

C. 3. Calibration and characterization

C.1. Calibration of the Raman shift axis

C.2. Spectral resolution characterization

1. What is the difference between the spectra presented in the supplementary material and the spectra presented in the paper? Provide an explanation. If they are identical remove the spectra from supplementary material.

2. Display the remaining Two peaks of calcite as you have done in figure 2.c for 1085 cm-1 peak.

3. Provide a table with FWHM for all Raman peaks obtained from measurements and place it in the paper.

C.3. Signal-to-noise comparison with a commercial high resolution Raman spectrometer

1. You should present spectra acquisitions with increased exposure time because with your spectrometer setup you obtain the same SNR as Renishaw.

D. 3. Discussion

1. Indicate more clearly the inside table from figure 4 and reference it better in figure description.

2. Provide an explanation why the intensities of the Raman Spectra presented in figure 4 are comparable, while in the case of the spectra presented in figure 3 the Raman spectrum acquired with Renishaw has considerably higher intensity.

3. 100x objective is the closest to the sample. Provide an explanation of why you did not use other objectives that can be more distant from the sample. Describe also the standard CCl4 test sample you have used.

4. Provide the model of the other high-resolution Raman spectrometer you have used to obtain high spectral resolution.

5. A summary of the paper begins on page 13 not connected to its current section: Discussion.

6. Where are the examples of limitations for CCD camera's presented in the paper?

E. There is no clear conclusion section.

F. Supplementary material

1. Point 2: Measurement of the excitation laser line width with Optical Spectrum Analyzer

Correct Figure S2-1 intensity scale.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The development of a cheap, high resolution, high sensitivity and small size spectrometer for Raman spectroscopy is of great importance, mainly because this gives the opportunity to a wide range of user to work in this area, which in normal conditions is prohibited. 

I only have these points to observe:

A) The manuscript is well written and contains only a few typographical errors indicated in yellow in the manuscript.

Lines 52, 105, 292, 311

 

B) I also have some concerns about:

1. What gain has been used for the three cameras?
2. What was the slit width of the commercial spectrometer?
3. In lines 429-433 the authors say that the SNR is comparable between the three spectrometer, but this is not strictly true since they are using different integration times and the justification of integration time scaling is not clear why is so.

 

C) What is the difference between this work and the work reports in reference [22] of the same authors, apart from the application and different excitation wavelength? The two works have the same design and almost the same components. 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The work is devoted to the development of a micro Raman spectrometer with a relatively inexpensive detector of high sensitivity and high spectral resolution. The work made a good impression, I wanted to highlight the detailing of the description of the system elements, the description of the methods and approaches used in the development of the spectrometer and the assessment of its characteristics. I would also like to note a number of comments on the article.

1. From the title of the paper, it is not clear what kind of spectrometer the author is presenting. Is the system a Raman microscope, or a micro-Raman spectrometer using an output lens? How is the sample positioned, or can the spectrometer be attached to a microscope?
   
2. In addition to the previous comment, there are no photos of the spectrometer or any information about its appearance. Neither in this article, nor in any additional materials, nor in previous work by the authors. Therefore, readers may have questions about what a Raman spectrometer is and its specific functions. At the same time, one of the main achievements of this work is to reduce the size of the spectrometer. However, there are no specific numerical values provided and no comparison with similar spectral systems is given.
3. The study of the spectrometer's operation in the confocal mode may be insufficient, as it is not well described. Specifically, the characteristics of the confocal mode, such as depth resolution and other features, are not clearly explained. Perhaps a separate paper should be written to analyze these characteristics in more detail. While these characteristics are only mentioned in passing here, I would recommend providing more information to readers or not excluding them from this work altogether, but rather mentioning the possibility of a separate study.
4. Why are photographic lenses used in this project? The spectrometer presented in this project is made from inexpensive components. To what extent can these lenses be considered inexpensive? At the same time, the authors of the previous work used single aspherical lenses. Why were these not used in this project?
   
5. Why was a diffraction grating with a blaze wavelength of 750 nanometers chosen?
6. Line 257 contains an incorrect link. The reader may get the impression that this link leads to the attached information in this article, rather than to the previous one that the authors refer to. That previous article describes the calibration protocol.
   
7. Figure 2A - Why was the spectral resolution higher for the third point (1085 cm-1) for a detector with a large pixel size? This seems to be due to an error in the estimation method. Has this been tested on other samples, or is the width of the line likely to be estimated using a neon lamp generally?

8. Figure 3 - How was the noise assessed in the figures? The text states that a range of 420-480 cm-1 was selected. Was the range of values or the standard deviation used? For example, in Figure A, the signal is 520 cm-1 for a duration of 100 ms was 50 (60-10), with a noise level of about 5. The ratio is 10. However, the expected ratio should be 610, so the calculation appears to be incorrect. Similarly, in Figure B, the signal has 250 a.u. for duration of 50 ms and a noise level of 5. However, the ratio is given as 50 instead of the expected 880. In Figure C, the noise level is difficult to assess based on the data provided. If we assume a duration of 50 ms, the noise level would be 10 with, signal 1200, a ratio of 120. However, this does not match the expected value of 1300. It seems that the calculation of the signal-to-noise ratio in these figures is incorrect. Additionally, it is unclear why different exposure times are used for different devices.

9. Figure 4: It is unclear why we do not see individual peaks on the Renishaw spectrometer. If we look at the Raman spectra from the work with the Renishaw spectrometer (Figure 2), the line width for 460 cm-1 should be in the range of 3.0 - 3.8 cm-1, which is sufficient for resolving the bands in the tetrachloride spectrum. However, on the authors' spectrometer, the resolution is approximately 3.6 cm-1. This raises the question of how these two values agree with each other. Additionally, what is the declared spectral resolution of the Renishaw instrument? Can the merging of peaks be explained by other factors?
   

Author Response

Please see the attachment.

Author Response File: Author Response.pdf