Review Reports - In Situ Metrology for Pad Surface Monitoring in CMP Using a Common-Path Phase-Shifting Interferometry: A Feasibility Study

Round 1

Reviewer 1 Report

Suggest authors review more of the recent research articles in this area to have an updated introduction. There is a lot of research done in this area in the past decade and needs to be presented here. For example, the confocal microscopy instrument by Nanofocus is capable of doing insitu measurements both in dry and wet state of the pads. There are manufacturers in the industry that offer similar capabilities.

For example, page 1, line 44: Authors have suggested reference 9 is a recent work, whereas this research work was published in 2011, that is 10 years back!

Also, authors need to present clearly how their work is different from the already established techniques for the CMP pad metrology and the benefits/limitations of their metrology.

Author Response

Dear Editor:

We would like to thank the reviewers for their constructive comments and helpful suggestions, upon which we have revised the paper in red color. After the revision, we believe that the quality of the paper is now significantly improved. Below, we address reviewer's comments point by point.

Regards,

E. S. Kim, and W. J. Choi

Tel: +82-10-9321-8730

Email: cecc78@cau.ac.kr

Reviewer Comments:

Reviewer 1

For example, page 1, line 44: Authors have suggested reference 9 is a recent work, whereas this research work was published in 2011, that is 10 years back!

Also, authors need to present clearly how their work is different from the already established techniques for the CMP pad metrology and the benefits/limitations of their metrology.

Thank you for your valuable comments and we apologize for overlooking the recent metrology techniques reviewer mentioned.
Hence, we have briefly described several confocal microscopy based three-dimensional (3-D) optical profilers including the Nanofocus product, capable of mapping the surface of the wet CMP pads in situ, which has been mentioned at the end of 2^nd paragraph of 1. Introduction (lines 51~57) in the revised manuscript as:
“More recently, advanced confocal technology based three-dimensional (3-D) surface profilers have been developed by several surface metrology companies [11-13]. By combining the metrologies with an appropriate immersion objective, the profilers are able to measure the wet pad on the polisher albeit being necessary for lateral beam scanning using either rotating multiple pinhole disk [11] or digital micromirror device (DMD) [12, 13]) and commonly vertical scanning of the objective with a piezo drive to generate a 3-D reconstruction of the pad surface.”
As mentioned in Introduction, the previous metrology techniques were ideal for static and dry pads, but not designed for pads under harsh manufacturing environment, where the pads are submerged and exposed to surrounding noises. Whereas, the proposed system is designed as an immersion metrology system, easily accessible to the wet pad. And its common-path interferometer is beneficial for yielding the interference immune to external noise, ensuring reliable 3-D reconstruction of the pad surface at the working place.
These benefits and limitations of our method have been properly mentioned at lines 253~270 in 4. Discussion of the revised manuscript as:
“The proposed system is designed for an immersion metrology, easily accessible to the submerged sample, and its common-path interferometry is beneficial for yielding the interference insensitive to external noises. Moreover, the phase-shifting manner requires no mechanical beam-scanning, providing the rapid reconstruction of surface topology. The result has shown that our approach was able to provide the areal topology map of wet or immersed CMP pad subject to external vibrations simulating the working place, which demonstrated one applicability of the proposed method for in situ inspection of the pad surface roughness during CMP.”
“Meanwhile, the present system has a few of technical issues to be addressed. (1) The phase-shifting manner using PZT motion might be instable to obtain the phase gain although the PZT used was driven by a piezo controller with positional feedback system. One method to achieve non-mechanical phase-shifting is to use polarization-base phase shifters consisting of quarter-wave plates and polarizers [21]. Therefore, the common-path interferometry equipped with the polarization optics would be able to allow for surface mapping of vibrating pads with a free of errors associated with the mechanical motion. (2) The current measurement area limited by the magnification of objective lens may not be enough to characterize the large surface area of the pad in working. Wide illumination through low N.A. optics could be a good choice for large area measurement with a compromise of lateral resolution.”

Author Response File: Author Response.pdf

Reviewer 2 Report

The article entitled “In situ metrology for pad surface monitoring in CMP using a 2 common-path phase-shifting interferometry: a feasibility study” presents an interesting study of measuring CMP pad surface based on a common-path phase-shifting interferometer.

The article presents a high scientific level and qualifies for publication, but there are a few places where the information could be clarified and the graphic side of the work could be improved.

The title is correctly and clearly formulated.

The abstract could mention the comparison of the presented method and the atomic force microscope (AFM) method.

The introduction is concise and logical, with sounds arguments. It introduces the readers to the main issues. The cited literature is not the latest, but in my opinion adequate to the discussed topic.

The innovative character of the work is well emphasised, but the aim of the work should be formulated more clearly.

In Figure 1, the values on the axes of the spectrum of the light source are not visible. The attached photo of the common-path of the system is too small and unclear. In addition, the captions make it impossible to identify the components. The coherence length marking is poorly visible. The references glass and pad layers should be marked on the enlargement of the measurement zone. The arrows showing the reflection points are hardly legible.

The Results section should provide the tapping and vibration parameters and locations.

The word ‘mechanical’ should be added to line 139 as follows: “method was unaffected by any MECHANICAL perturbations, creating noise-immune interferometry”.

Fig. 3 (f) - the obtained characteristics should be analysed and the measurement accuracy should be determined in nm.

Fig. 3 (g) - the captions under the axes are illegible.

Line 198 - the theme of tiny air bubbles trapped in the pad pores should be developed and the scale of the problem and its impact on the measurement should be presented more clearly.

Fig. 4(b) – Scale is missing.

Fig. 4(h) - Axis captions are illegible.

I recommend adding 2-3 newer entries from the last 5 years to the References section.

Comments for author File: Comments.docx

Author Response

Dear Editor:

Regards,

E. S. Kim, and W. J. Choi

Tel: +82-10-9321-8730

Email: cecc78@cau.ac.kr

Reviewer Comments:

Reviewer 2

The article presents a high scientific level and qualifies for publication, but there are a few places where the information could be clarified and the graphic side of the work could be improved.

The title is correctly and clearly formulated.

The abstract could mention the comparison of the presented method and the atomic force microscope (AFM) method.

Thank you for your comment. We have added the comparison in abstract of the revised manuscript as:
“The surface mapping performance of the proposed method was tested by retrieving a topology of a vibrating nanostructure in immersion, of which height profiles were consistent with the result from an atomic force microscopy (AFM).”

The innovative character of the work is well emphasised, but the aim of the work should be formulated more clearly.

We really appreciate for reviewer’s compliment and encouraging comments! We have made effort to polish the manuscript for clarification of the aim of the work in the revision.

As reviewer pointed out, Fig. 1 has been properly modified in the revised manuscript.

The Results section should provide the tapping and vibration parameters and locations.

Thank for your valuable point. In our experiment, high-frequency (>100 Hz) vibration was induced by a smart phone on the optical table and low frequency (few Hz) tapping was also applied to the corner of the optical table.
This point has been added at the beginning of section 3.1 (lines 136~140) in the revised manuscript as:
“We monitored temporal change in the interference signal subject to periodic high-frequency (>100 Hz) vibration induced by a mobile phone on the optical table as well as intermittent artificial disturbances by low-frequency (few Hz) tapping applied to the corner of the optical table to mimic the industrial noises.”

The word ‘mechanical’ should be added to line 139 as follows: “method was unaffected by any MECHANICAL perturbations, creating noise-immune interferometry”.

The word ‘mechanical’ has been inserted at line 160 in the revised manuscript.

Fig. 3 (f) - the obtained characteristics should be analysed and the measurement accuracy should be determined in nm.

Thank you for significant comment. We additionally repeated the measurements and then the measurement bias relative to the reference value (145.35 nm) by AFM was determined to be 0.37 nm on average, which can be considered as the measurement accuracy of our system.
We added this point at the end of section 3.2 (lines 190~192) in the revised manuscript as:
“From five-times repetitive measurements, the measurement error to the reference value (145.35 nm) was 0.37 nm on average, representing the measurement accuracy of our system.”

Fig. 3 (g) - the captions under the axes are illegible.

3 has been updated in the revised manuscript.

Line 198 - the theme of tiny air bubbles trapped in the pad pores should be developed and the scale of the problem and its impact on the measurement should be presented more clearly.

We have found the air bubbles at this pad experiment and it’s very careful to say the rationale of the bubble formation to current our knowledge. However, we speculate that when air-filled pad pores are immersed or poured by water, the air may be trapped accidentally on the immersed pores. It is evidenced from the fact that the gas bubbles at the liquid-solid surface interface tend to stay on the hydrophobic surface, especially, on the hydrophobic pits [Meng et al. “A degassing plate with hydrophobic bubble capture and distributed venting for microfluidic devices,” Micromech. Microeng. 16 (2006)]. Since the pad used is made of polyurethane, which is hydrophobic, naturally repelling water, accordingly, the air bubbles tend to preferentially attach to the pores.
This point has been properly mentioned at the 2^nd paragraph of section 3.3 (lines 225~233) in the revised manuscript as:
“The reason of bubble formation on the pores is carefully speculated that when the pad with the air-filled pores is immersed or poured by water, the air may be trapped accidentally on the immersed pores. It is evidenced from the fact that the gas bubbles in the liquid-solid surface interface tend to stay on the hydrophobic surface, especially, on the hydrophobic pit sites [19]. Since the pad used is made of hydrophobic polyurethane, naturally repelling water, accordingly, the bubbles may preferentially attach to the pores. However, the bubbles may not appear for the pad on CMP because of constant polishing and influx of chemical slurry particles into the pores.”
Furthermore, the presence of the strong reflective bubbles can prevent the generation of interference at the pores, leading to errors in the pad asperity inspection.
This has been added at the 2^nd paragraph of section 3.3 (lines 224~225) in the revised manuscript as:
“…otherwise, the high reflective liquid-air interface can prevent from generating interference at the pores, resulting in error in the pad asperity measurement.”

Fig. 4(b) – Scale is missing.

Scales have been added in the Fig. 4(b) in the revised manuscript.

Fig. 4(h) - Axis captions are illegible.

For readability, the axis captions have been enlarged in the Fig. 4(h) in the revised manuscript.

I recommend adding 2-3 newer entries from the last 5 years to the References section.

Four references describing recent metrology techniques have been newly added in the revised manuscript listed as:

[8] Zhao, T; Yuan, J; Deng, Q; Feng, K; Zhou, Z; Wang, X. Contrast experiments in dielectrophoresis polishing (DEPP)/chemical mechanical polishing (CMP) of sapphire substrate. Appl. Sci. 2019, 9, pp. 3704.

[11] http://onlinelibrary.wiley.com/doi/pdf/10.1002/opph.201500007.

[12] Matilla, A; Mariné, J; Pérez, J; Cadevall, C; Artigas, R. Three-dimensional measurements with a novel technique combination of confocal and focus variation with a simultaneous scan publication. In Optical Micro- and Nanometrology VI, Proceedings of SPIE Photonics Europe, Brussels, Belgium, 3–7 April 2016; Gorecki, C. Eds.; SPIE: Bellingham, WA, USA.

[13] https://www.sensofar.com/wp-content/uploads/2015/07/App-note-S-mart-CMP.pdf.

Author Response File: Author Response.pdf

Reviewer 3 Report

A) While the paper is good overall, some statements about CMP are not strictly correct or are objectionable in some way. Some examples:

Line 22: Moore's 1965 observation was that the number of transistors was doubling every two years, not every year.

Line 30: You may mean "... owing to a number of advantages in addition to global planarization ..."

Line 34: What do you mean by "to standardize wafer flatness"?

Line 36 says "Previously there have been several attempts at measuring the surface roughness of the CMP pads ..." These were not attempts but measurements that actually worked and served the intended purpose even though the measurements were static and mostly on dry samples. The authors may also not be aware that static contact area confocal microscopy measurements have also been performed on wet pad samples and on heated samples and that these are also useful. It is more effective to point out the flaws in other's methods and results if yours do not have some of the same shortcomings.

Line 57: What does "industrial spot" mean?

Most of these objections are probably just English expression issues, but they shouldn't remain in the paper.

B) While pad topography is important and it would be very helpful to monitor it in situ, pad-wafer contact area is also important since material removal occurs in the contacts.

C) The preliminary results in the paper are very promising, particularly the precision and stability of the measurements in the presence of noise. There is still a long way to go to apply this on a commercial CMP tool, though, since it is also necessary to contend with pad thickness non-uniformities on the order of +/- 20 microns, platen run-out of about the same magnitude, the fact that some slurries are nearly opaque, pad grooves, and other issues. Plus the method has to be fast to minimize throughput impact. It's a long-term effort.

Author Response

Dear Editor:

Regards,

E. S. Kim, and W. J. Choi

Tel: +82-10-9321-8730

Email: cecc78@cau.ac.kr

Reviewer Comments:

Reviewer 3

A) While the paper is good overall, some statements about CMP are not strictly correct or are objectionable in some way. Some examples:

Line 22: Moore's 1965 observation was that the number of transistors was doubling every two years, not every year.

Thank you for your point and we’re very sorry for the typo. We have modified “every year” to “every two years” in 1. Introduction of the revised manuscript.

Line 30: You may mean "... owing to a number of advantages in addition to global planarization ..."

Yes, you’re right! We have modified “…over…” to “…in addition to…” in 1. Introduction of the revised manuscript.

Line 34: What do you mean by "to standardize wafer flatness"?

We have changed the sentence as: “…during CMP to fulfil high quality of wafer flatness.” in 1. Introduction of the revised manuscript.

Thank you for your valuable points and we apologize for overlooking the recent metrology systems reviewer mentioned.
At first, we have changed “attempts” to “measurements” in 1. Introduction of the revised manuscript.
And further, as reviewer pointed out, a company, Sensofar Metrology has recently rolled out one of three-dimensional (3-D) optical sensors based scanning confocal microscopy using microdisplay approach. By combining this metrology with an appropriate immersion objective, it is able to measure the wet CMP pads on the polisher although being necessary for the beam scanning in both lateral (x-y) and axial (z) directions to generate a 3-D reconstruction of the surface, while our technique does not need such beam scan approach for 3-D reconstruction.
We have properly described the recent confocal metrology techniques including the Sensofar’s product at the end of 2^nd paragraph of 1. Introduction (lines 51~57) in the revised manuscript as:
“More recently, advanced confocal technology based three-dimensional (3-D) surface profilers have been developed by several surface metrology companies [11-13]. By combining the metrologies with an appropriate immersion objective, the profilers are able to measure the wet pad on the polisher albeit being necessary for the lateral beam scanning using either rotating multiple pinhole disk [11] or digital micromirror device (DMD) [12, 13]) and commonly vertical scanning of the objective with a piezo drive to generate a 3-D reconstruction of the pad surface.”

Line 57: What does "industrial spot" mean?

We’re sorry for confusing. We have properly modified the sentences involving the words at the beginning of 2. Materials and Methods (lines 65~66) in the revised manuscript as:
“In order to enable in situ pad surface monitoring, the proposed method adopted a common-path optical interferometer configuration [14] as shown in Fig. 1.’’

Most of these objections are probably just English expression issues, but they shouldn't remain in the paper.

B) While pad topography is important and it would be very helpful to monitor it in situ,pad-wafer contact area is also important since material removal occurs in the contacts.
C) The preliminary results in the paper are very promising, particularly the precision and stability of the measurements in the presence of noise. There is still a long way to go to apply this on a commercial CMP tool, though, since it is also necessary to contend with pad thickness non-uniformities on the order of +/- 20 microns, platen run-out of about the same magnitude, the fact that some slurries are nearly opaque, pad grooves, and other issues. Plus the method has to be fast to minimize throughput impact. It's a long-term effort.

For (B) and (C): Thank you for compliments and we totally agree with reviewer’s comments. As reviewer pointed out, in order to realize in situ monitoring, indeed, the proposed approach should be integrated into the production line and it should be capable of rapidly acquiring and automatically analyzing data to effectively monitor key pad characteristics such as material removal rate, pad thickness non-uniformities, platen run-out, and groove occlusion so on. It will be a long-term goal to achieve the capabilities although this preliminary study has successfully proved the proposed method as an alternative to prior interferometric technologies challenging to use in the manufacturing field.
We have properly added these points at the end of 4. Discussion (lines 271~277) in the revised manuscript as:
“Despite this preliminary work showed precise and stabilized surface mapping of the proposed technique, there is still a long way to go to apply this technique on commercial CMP tools. Much efforts will be necessary to integrate the system into the polishers in production line and to improve the system’s capabilities of rapidly acquiring and automatically analyzing data to effectively monitor key pad characteristics including the surface roughness such as material removal rate, pad thickness non-uniformities, platen run-out, and groove occlusion, which will be a long-term but achievable goal.”

Author Response File: Author Response.pdf

Reviewer 4 Report

In the actual CMP process, both the CMP pad and the wafer are rotating. Is the system still applicable to the situation to perform the "in situ" measurement?

Line 36; "in situ" should be italic.

Line 40; "no reports" can be an exaggeration. At least, you had already made one report in 2011 as shown in reference 9.
In 2015, a company, Sensofar, had also released a similar equipment to your work. https://www.sensofar.com/wp-content/uploads/2015/07/App-note-S-mart-CMP.pdf

Lin43; "wetted" can be changed to "wet"?

2. Materials and Methods; Could you kindly describe the information about the CMP pad you used? Manufacturer name and product name?

Line 58; "(CMP)" can be removed from the phrase "the polishing (CMP) pad"? Or can it be " the CMP pad"?

Line 182; I couldn't understand why you used the perforated anodized aluminum plate instead of an actual CMP pad.

Line 192; I couldn't understand why you needed to use the "fully processed (worn out)" CMP pad. Is there any reason that you cannot use a new CMP pad?

Figure 4 (e); Bubbles were still remained even though you removed the bubbles by micropipette?
Figure 4 (g); There was no description in the caption.

Line 223-235; I felt that the description can be moved to the Introduction part.

Author Response

Dear Editor:

Regards,

E. S. Kim, and W. J. Choi

Tel: +82-10-9321-8730

Email: cecc78@cau.ac.kr

Reviewer Comments:

Reviewer 4

In the actual CMP process, both the CMP pad and the wafer are rotating. Is the system still applicable to the situation to perform the "in situ" measurement?

Thank you for your comment! We think that the “in situ measurement” means not only the measurements during polishing, but also the measurements after polishing work on the polisher.

Line 36; "in situ" should be italic.

“In situ” has been italic in the revised manuscript.

Thank you for your valuable comment. We apologize for missing the recent confocal metrology techniques reviewer mentioned.
We have properly modified the sentences involving the words of “no reports” at the 2^nd paragraph of 1. Introduction in the revised manuscript as:
“In spite of promising results using these classical metrology technologies [6–9], the measurements of pad in situ have been scarce;”
Further, we have added a brief description of the recent confocal metrology techniques with related references at the end of 2^nd paragraph of 1. Introduction (lines 51~57) in the revised manuscript as:
“More recently, advanced confocal technology based three-dimensional (3-D) surface profilers have been developed by several surface metrology companies [11-13]. By combining the metrologies with an appropriate immersion objective, the profilers are able to measure the wet pad on the polisher albeit being necessary for the lateral beam scanning using either rotating multiple pinhole disk [11] or digital micromirror device (DMD) [12, 13]) and commonly vertical scanning of the objective with a piezo drive to generate a 3-D reconstruction of the pad surface.”

Lin43; "wetted" can be changed to "wet"?

“wetted” has been changed to “wet” at line 46 in the revised manuscript.

Materials and Methods; Could you kindly describe the information about the CMP pad you used? Manufacturer name and product name?

We’re very sorry about that. The pad used has been donated with no providing specification.

Line 58; "(CMP)" can be removed from the phrase "the polishing (CMP) pad"? Or can it be " the CMP pad"?

Yes, you’re right. (CMP) is unnecessary. We have modified the phrase at the beginning of 2. Materials and Methods (lines 65~66) in the revised manuscript as:
“In order to enable in situ pad surface monitoring, the proposed method adopted a common-path optical interferometer configuration [14] as shown in Fig. 1.”

Line 182; I couldn't understand why you used the perforated anodized aluminum plate instead of an actual CMP pad.

In addition to USAF resolution target imaging, we have tried to test our system performance with the aluminum anodized surface similar to the pad surface asperity before being offered the pads.

Line 192; I couldn't understand why you needed to use the "fully processed (worn out)" CMP pad. Is there any reason that you cannot use a new CMP pad?

Thank you for your point. Initially, it was not easy for us to secure commercially available new pad on CMP in pieces and thus the pad strip processed has been offered with a help of colleagues in other CMP labs. We are currently being in contact with a domestic wafer production company to receive the different states of pad samples in CMP for our next project to evaluate the pad characteristics.

Figure 4 (e); Bubbles were still remained even though you removed the bubbles by micropipette?

As mentioned in the original manuscript, Fig. 4(e) is the picture before removing the bubbles.

Figure 4 (g); There was no description in the caption.

We’re very sorry for missing the caption. The caption (g) has been added in Fig. 4 in the revised manuscript.

Line 223-235; I felt that the description can be moved to the Introduction part.

Thank you for your point. We think that it would be better to put the sentences following the system description. So, we have moved it to the end of 2. Materials and Methods (lines 120~133) of the revised manuscript as:
“However, the configuration of interferometer used is similar to Mirau interferometer that has been utilized in recent years for metrology of patterned wafer [17], thin film [18], and micro electro mechanical systems (MEMS) [19]. The most of Mirau interferometers commonly employ a commercial interferometry microscope objective involving a small reference mirror and a beam splitter in the body to generate an equal path length-induced interference for sample in air. For submerged samples such as pad in slurry, however, this arrangement is not suited to use because when the objective is immerged into the liquid, the sample and reference beam paths may not be identical due to refractive index mismatch between the surrounding media, not ensuring interference. Further, high reflective metallic mirror in the objective may degrade signal contrast in the interference for low reflectance materials to be inspected. On the other hand, the proposed interferometer has the individual immersion objective and the movable reference glass so that the index matching can be easy and the reference glass can be properly positioned within the coherence length, guaranteeing high-contrast interference on the submerged materials.”

Author Response File: Author Response.pdf