Metrology for Measuring Bumps in a Protection Layer Based on Phase Shifting Fringe Projection

Round 1

Reviewer 1 Report

Summary:

The authors impressively present a novel approach for 3D measurement of bumps within ball grid arrays using fringe protection. An additional reflectometry sensor helps to improve occurring height offsets due to a transparent protection layer.  The paper is very well written, scientifically and technically highly interesting.

Review:

The manuscript is clear and relevant for the field and presented in a well-structured manner. Cited references are relevant and I don't see abnormal number of self-citations. The manuscript is scientifically well prepared and the experimental design appropriate to test the hypothesis. The manuscript’s results are reproducible based on the details given in the methods section. However, some more technical information could be given (will be addressed within the specific comments). The figures/tables/images/schemes are appropriate - they properly show the relevant data and are easy to interpret and understand. The data is interpreted appropriately and consistently throughout the manuscript. The conclusions are consistent with the evidence and arguments presented.

Specific comments:

1. Line 24: The authors should think about adding multiwavelength digital holography as a possible measuring technology for bump height measurements. The following paper might be of interest: 10.37188/lam.2021.015
2. Line 120ff.: This paragraph should include more information about the measuring speed of the system. Measuring speed is one of the key parameters (besides measuring quality) in production.
3. Line 120ff.: Please add clearer information about the lateral sampling with the different magnifications.
4. Line 123: Which objective lens is used here? You specifically name the other components afterwards, but not this one?
5. Line 137: Which wavelengths have been used for the latter measurements?
6. Line 155ff.: You could add that measurements will be shown later and only introduced here. It helps the reader to capture the structure of the paper.
7. Line 222: Indices of n_PL e. g. and later are somehow not subscripted correctly
8. Line 239ff.: Which magnification did you use for these measurements, what's the lateral sampling here?
9. Line 274: What's the radius of the gaussian filter used here?
10. Line 276: The intensity has been introduced as I(x,y), maybe I_{0}, I_{2pi/3} is clearer?
11. Line 317: Why is the math part elevated above the fonts surrounding it? Typesetting may be improved.

Author Response

The authors would like to express sincerely thanks to the reviewer's valuable comments.

Author Response File: Author Response.docx

Reviewer 2 Report

This is a very good paper. Surface inspection at high speed for moving surfaces coated by thin films is a real challenge for semiconductor industry. The authors successfully introduce a new robust fringe projection technique to measure microstructures features on semiconductor wafers at high speed. The system is combined with a reflectometry sensor to correct for errors produced by coating layers. A mathematical description was provided for phase and height calculation, showing the analysis to correct for errors produced by coating layers. Other measurement standard techniques such as white light interferometry and SEM were used to verify the results. I would highly recommend publishing this paper after correcting the following minor issues:

 

1. Page 2, line 46. It has been mentioned that a “three-row line scan camera” was used to simultaneously acquires three fringe images.

My comment: Did the measurement consider a single exposure time to acquire the three images? There is no explanation how this was achieved. Is this a phase masked camera? Please give more explanation.

2. Page 3, line 127. It has been mentioned that camera was set with an active area 4096(H)x50(V).

My comment: It was mentioned earlier that this is a three-row line scan camera. This is confusing. Please link your description about the ROI with the three-row concept to avoid confusion for the readers.

3. Page 6, line 189. It has been mentioned that the ratio of d_D/d_c was selected to be 1/3.

My comment: Is this selection based on Nyquist criteria? Please specify.

4. Page 6, line 194. It’s been written “This displacement in the lateral direction ….”.

My comment: Using ‘this’ word might confuse the reader. Please rephrase this sentence such as “The displacement caused by the shift in the fringe position can be given by”

5. Page 219, line 219. It’s been stated that the reflectometry “is fast enough”.

What is the speed of the measurement to provide the correct factor? Please specify. The paper has been mentioned that the frame rate of the camera can be up to 10,000 fps. Can the reflectometry cope with this speed?

6. Page 9, line 269. It has been mentioned that “The phase can then be unwrapped to obtain the continuous phase image intensity”.

My comment: This paper didn’t discuss the unwrapping algorithm. The computing time for phase unwarping is usually slow if robust algorithm was chosen to avoid error from defects. Did the authors consider a 1D or 2D phase unwrapping algorithm? What is the computing time for this?

7. Figure 7 b, c, and d.

My comment: Did the system acquire the data at a single exposure time (see my comment in point 1 above)? Please state the acquisition time.

8. Page 11, line 323. It has been mentioned that “Because the oxide–silicon surface acts like a mirror, a stronger signal at the corresponding OPD 2”.

My comment: Does this mean that the given analysis has ignore the influence of (A&B interference)? Please state if this interference can be ignored and won’t introduce noticeable error.

 

 

9. Page 11, line 332. It has been mentioned that “the roughness of the 332 PI–oxide and oxide–silicon interface also influenced the amplitude of the reflectance”.

My comment: What is the resolution of the reflectometry at full spectrum? If the authors can specify the resolution, so the readers can judge if the variation in film thickness due to roughness can be detected and influenced the amplitude of the reflectance.

10. Finally, this research suggests using fringe projection and three-phase shift algorithm to measure moving surfaces. However, I couldn’t find any specification about the time. I have highlighted this in my comments above. However, it is important the authors the specify the overall achieved measurement time including the data acquisition time, phase and height calculation, and phase unwrapping time? What is the speed limit for this technique?

 

 

 

Comments for author File: Comments.pdf

Author Response

The authors would like to express our sincere thanks to the reviewer's valuable comments.

Author Response File: Author Response.docx