Robust ESD-Reliability Design of 300-V Power N-Channel LDMOSs with the Elliptical Cylinder Super-Junctions in the Drain Side

Round 1

Reviewer 1 Report

This paper explores the improvement of a LDMOS for ESD applications. This article is interesting, but there are too many explanations missing and some sentences are difficult to understand. Here is the detail of my review

L33: What means 'IC'?

What is the difference between 'lateral double-diffused MOSFET' and 'N-channel lateral diffused MOSFET'?

L36: UHV is explained in the keywords but should also be in text, as it is for LDMOS, ESD, etc. Everything should be consistent. Same remark for HBM.

L47: 'The CoolMOS™ registered by Infineon Technologies [51]' a verb or a word is missing, I don't understand the sentence.

L60-61: 'For improving the ESD capability of an UHV LDMOS, this study proposes a novel structure, which is added elliptical cylinder super-junctions (SJs) into the UHV circular N-channel LDMOS.' should be written as
'For improving the ESD capability of an UHV LDMOS, this study proposes a novel structure, where an elliptical cylinder SJs is added into the UHV circular N-channel LDMOS.

L61: SJ is already defined previously, no need to express it again. Same for nLDMOS.

L62-63: 'It is known that an N-channel LDMOS with these SJs in the drift region is a quite complicated structure and it will become quite different from a conventional UHV LDMOS.' I don't understand this statement. What will become different?

Figure 3: What mean STI, HVPB, HVNW and HVPW?

L81: What means 'RESURF'?

L90: Don't begin a sentence with 'and'.

L103: What means 'length_X'? I understand that it is the length of X, so you should state it like this. Or you should introduce this notation as the variable length of your elliptical cylinders.

L146: Why is there a 'T' in 'LDMOST'?

Figures 9 (a) and 9 (b) should be only one figure so we could clearly see (or not see) the small difference between the Ref and SJ DUTs.

Figure 10: What is the point of showing 5 identical bars? This figure should be reworked or removed.

L174: Don't begin with 'And'.

L187: What means 'OD'?

L187: 'This is caused by the OD distance from the outermost HVPW to the gate is too small, the heat dissipation cross-sectional area is small, and it will result in a decreasing in the HBM robustness'
should be
'This is caused by the OD distance from the outermost HVPW to the gate and the heat dissipation cross-sectional area BEING too small, resulting in a decreasing in the HBM robustness'

 

 

Author Response

Author's Response to Reviewers

(Modifications Report of the Paper #electronics-765576)

 

Response: Authors are grateful to the reviewers for the time and efforts in this paper review process. We have done with our best efforts to address all of the questions raised by the reviewers. A revision has been made by taking into account the reviewers’ comments so that it is now more readable to the reader. These updated parts were marked by some red words. Meanwhile, we have re-organized and updated this manuscript, so the page and line number of this new revision is somewhat different from the previous old version.

 

Response to the Comments of Reviewer:

(1). L33 (New L66): What means 'IC'?

Response:

Thanks the reviewer comment. IC means 'Integrated Circuit'. We have added this detailed description in the new revision marked by red words.

 

(2). What is the difference between 'lateral double-diffused MOSFET' and 'N-channel lateral diffused MOSFET'?

Response :

Thanks the reviewer comment. Generally, 'lateral double-diffused MOSFET' which includes 'N-channel lateral diffused MOSFET' and 'P-channel lateral diffused MOSFET'.

 

(3). L36 (New L69): UHV is explained in the keywords but should also be in text, as it is for LDMOS, ESD, etc. Everything should be consistent. Same remark for HBM.

Response:

Thanks the reviewer comment. 'UHV' is explained in the Abstract first. We have corrected these carelessness and added detailed descriptions in the new section 'nomenclature' marked by red words in the new revision.

 

(4). L47 (New L80): 'The CoolMOS™ registered by Infineon Technologies [51]' a verb or a word is missing, I don't understand the sentence.

Response:

Thanks the reviewer comment. We have fixed this error and marked by red words in the new revision.

 

(5). L60-61(New L93-94): 'For improving the ESD capability of an UHV LDMOS, this study proposes a novel structure, which is added elliptical cylinder super-junctions (SJs) into the UHV circular N-channel LDMOS.' should be written as
'For improving the ESD capability of an UHV LDMOS, this study proposes a novel structure, where an elliptical cylinder SJs is added into the UHV circular N-channel LDMOS.

Response:

Very thanks the reviewer comment. Now we have corrected this sentence and marked by red words in the new revision.

 

(6). L61 (New L94): SJ is already defined previously, no need to express it again. Same for nLDMOS.

Response:

Very thanks the reviewer comment. Now we have corrected this carelessness and marked by red words in the new revision.

 

(7). L62-63: (New L94-95) 'It is known that an N-channel LDMOS with these SJs in the drift region is a quite complicated structure and it will become quite different from a conventional UHV LDMOS.' I don't understand this statement. What will become different?

Response:

Thanks the reviewer comment. These SJs in the drain-side drift region will form a series of depletion regions, so this UHV nLDMOS-SJ is quite different from a conventional UHV LDMOS.

 

(8). Figure 3: What mean STI, HVPB, HVNW and HVPW?

Response:

Thanks the reviewer comment. We have added these detailed descriptions in the new section 'nomenclature' of the new revision and marked by red words.

 

(9). L81 [New L126]: What means 'RESURF'?

Response:

Thanks the reviewer comment. 'RESURF' means 'reduced surface field'. We have added this detailed description in the new section 'nomenclature' and the corresponding text of the new revision and marked by red words.

 

(10). L90 [New L135]: Don't begin a sentence with 'and'.

Response:

Thanks the reviewer comment. We have deleted this word in the new revision.

 

(11). L103 [New L149]: What means 'length_X'? I understand that it is the length of X, so you should state it like this. Or you should introduce this notation as the variable length of your elliptical cylinders.

Response:

Thanks the reviewer comment. We have added this detailed description in the new section 'nomenclature' and the corresponding text of the new revision and marked by red words.

 

(12). L146 [New L192]: Why is there a 'T' in 'LDMOST'?

Response:

Thanks the reviewer comment. For article consistent, we have corrected this word (delete 'T') and marked by red words in the new revision.

 

(13). Figures 9 (a) and 9 (b) (New Figures 8 (a) and 8 (b)) should be only one figure so we could clearly see (or not see) the small difference between the Ref and SJ DUTs.

Response:

Thanks the reviewer comment. Yes, we selected the nLDMOS-SJ device with the middle value of length_X 10mm and Ref. device under the strong inversion Vg = 5V for comparison.

 

(14). Figure 10 (New Figure 9): What is the point of showing 5 identical bars? This figure should be reworked or removed.

Response:

Thanks the reviewer comment. Yes, we have reworked and enlarged the y-axis of Fig. 9.

 

(15). L174 [New L223]: Don't begin with 'And'.

Response:

Thanks the reviewer comment. We have deleted this word in the new revision.

 

(16). L187 [New L246]: What means 'OD'?

Response:

Thanks the reviewer comment. We have added this detailed description in the new section 'nomenclature' and the corresponding text of the new revision and marked by red words.

 

(17). L187 [New L246-248]: 'This is caused by the OD distance from the outermost HVPW to the gate is too small, the heat dissipation cross-sectional area is small, and it will result in a decreasing in the HBM robustness'

should be

'This is caused by the OD distance from the outermost HVPW to the gate and the heat dissipation cross-sectional area BEING too small, resulting in a decreasing in the HBM robustness'

Response:

Very thanks the reviewer comment. We have corrected this sentence and marked by red words in the new revision.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper deals with improving the ESD immunity of 300-V Power N-channel LDMOS using the additional symmetric elliptical-cylinder SJs structure into the drain-side drift region and their impact to the HMB ESD capabilities. Authors present experimental samples for both case: symmetrical eight-zone elliptical cylinder SJs in the drain region in which the length of SJs are 5, 10, 15, 20 μm, respectively; and symmetrical four-zone elliptical cylinder SJs in the drain region. Those results could be useful for applications with the ESD-reliability demand. Overall quality of the paper is acceptable with a minor revision.

Comments:

• The theoretical part of the different MOSFET structures is omitted. In my opinion there is lack of mathematical description and physical meaning of presented structures. Those paper consist only from experimental results with a lot of references without explanation of the problem background.
• Why in the article are presented just these two structures: symmetrical eight-zone elliptical cylinder SJs in the drain region in which the length of SJs are 5, 10, 15, 20 μm, respectively; and symmetrical four-zone elliptical cylinder SJs in the drain region ?
• Please, could you explain in more details what the HBN capability is? What does it mean the OD distance?
• The sizes of figures are different, for example size of Fig. 3 and 4 is higher than of Fig.5a.

Author Response

Author's Response to Reviewers

(Modifications Report of the Paper #electronics-765576)

 

Response: Authors are grateful to the reviewers for the time and efforts in this paper review process. We have done with our best efforts to address all of the questions raised by the reviewers. A revision has been made by taking into account the reviewers’ comments so that it is now more readable to the reader. These updated parts were marked by some red words. Meanwhile, we have re-organized and updated this manuscript, so the page and line number of this new revision is somewhat different from the previous old version.

 

Response to the Comments of Reviewer:

(1). The theoretical part of the different MOSFET structures is omitted. In my opinion there is lack of mathematical description and physical meaning of presented structures. Those paper consist only from experimental results with a lot of references without explanation of the problem background.

Response:

Thanks the reviewer comment. Yes, we have added some theoretical explanations and a mathematical prediction model in the new revision and marked by red words shown in P.3 and P.8-9.

 

 (2). Why in the article are presented just these two structures: symmetrical eight-zone elliptical cylinder SJs in the drain region in which the length of SJs are 5, 10, 15, 20 μm, respectively; and symmetrical four-zone elliptical cylinder SJs in the drain region ?

Response:

Thanks the reviewer comment. Yes, in this article we have chosen two architectures (symmetrical eight-zone and symmetrical four-zone) with a high degree of symmetry and the layout is not too complicated (in order to the industrial practicality will be higher in the future).

 

(3). Please, could you explain in more details what the HBM capability is? What does it mean the OD distance?

Response:

Thanks the reviewer comment. The HBM capability is the HBM immunity level of a device as it suffered an HBM transient-noise bombardment. The 'OD distance' explanation is added in the new section 'nomenclature' and the corresponding text of the new revision and marked by red words.

 

(4). The sizes of figures are different, for example size of Fig. 3 and 4 is higher than of Fig.5a.

Response:

Thanks the reviewer comment. We have fixed sizes carelessness of Figs. 3~5 in the new revision.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

• The abstract is not giving a clear picture of the work. Too many details and abbreviations and no clear structure.
• 1 does not add any scientific value to this paper, it shall be removed and replaced with a graph e.g. on LDMOS reliability.
• Explain all abbreviation used in your text

-             Line 62 should provide a reference when stating “it is known”

-             Fig. 3: Abbreviations should be explained in the text

-             Fig. 5: If this is design M8 (corresponding to results in Fig. 11) this should be made more clear

 

• Line 121 Reference to the later discussed breakdown data on the influence of the different designs should be provided

-             Fig 7 does not add any scientific value to this paper, it shall be removed

• Explain the physics behind your observation

-             Line 155: The measured currents should be represented in a different way

-             Line 160: Is this typical data or average values? More statistics should be shown here. What about distribution of values across the wafer and some more statistics of these values would make the results more reliable.

-             Fig. 10: Should be replotted or removed, like this it does not add any value to the discussion

-             Line 183: This statement is not in agreement with Fig. 12 (M6 to M10)

-             Line 181-196 The discussion on the HBM characterization results should be more detailed. Explain the observations by using a physical model

Author Response

Author's Response to Reviewers

(Modifications Report of the Paper #electronics-765576)

 

Response: Authors are grateful to the reviewers for the time and efforts in this paper review process. We have done with our best efforts to address all of the questions raised by the reviewers. A revision has been made by taking into account the reviewers’ comments so that it is now more readable to the reader. These updated parts were marked by some red words. Meanwhile, we have re-organized and updated this manuscript, so the page and line number of this new revision is somewhat different from the previous old version.

 

Response to the Comments of Reviewer:

(1). The Abstract is not giving a clear picture of the work. Too many details and abbreviations and no clear structure.

Response:

Thanks the reviewer comment. We have added some more clearer descriptions in the Abstract and a new section 'nomenclature' of the new revision marked by red words.

 

(2). Do not add any scientific value to this paper, it shall be removed and replaced with a graph e.g. on LDMOS reliability.

Response:

Very thanks the reviewer comment. We have removed it (such the Fig. 7 in previous version) in the new revision.

 

(3). Explain all abbreviation used in your text

- Line 62 [New Line 94] should provide a reference when stating “it is known”

- Fig. 3: Abbreviations should be explained in the text

- Fig. 5: If this is design M8 (corresponding to results in Fig. 11 (New Fig. 10)) this should be made more clear

Response:

Thanks the reviewer comments. We have fixed these unclear statements and marked by red words in the new revision. 

- Line 94 has provided some references in this new revision.

- Fig. 3: Abbreviations have explained in the new section 'nomenclature' of the new revision.

- Yes, we have fixed these unclear statements about Fig. 5 and marked by red words.

 

(4). Line 121 [New Line 168] Reference to the later discussed breakdown data on the influence of the different designs should be provided

- Fig 7 does not add any scientific value to this paper, it shall be removed

Response:

Thanks the reviewer comments.

- Line 121 [New Line 168] We have added more clearer descriptions for the original unclear statements and marked by red words in the new revision.

- Yes, we have deleted Fig. 7 (old version).

 

(5). Explain the physics behind your observation

- Line 155 [New Line 201]: The measured currents should be represented in a different way

- Line 160 [New Line 205]: Is this typical data or average values? More statistics should be shown here. What about distribution of values across the wafer and some more statistics of these values would make the results more reliable.

- Fig. 10 [New Fig. 9]: Should be replotted or removed, like this it does not add any value to the discussion

- Line 183 [New Line 242]: This statement is not in agreement with Fig. 12 (M6 to M10)

- Lines 181-196 [New Lines 238-257]: The discussion on the HBM characterization results should be more detailed. Explain the observations by using a physical model

Response:

We appreciate the reviewer comments. We have fixed these unclear statements and marked by red words in the new revision.

- Line 155 [New Line 201]: We have fixed these unclear statements and marked by red words.

- Line 160 [New Line 205]: In this work, all of measured data were obtained by the average of three samples.

- Fig. 10 [New Fig. 9]: We have re-plotted and added testing data on the figure.

- Line 183 [New Line 242]: We have fixed this description and marked by a red word.

- Lines 181-196 [New Lines 238-257]: We have added some detailed descriptions marked by red words and adding a physical (empirical) formula (P.8-P.9) in the new revision.

 

 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The abstract needs still to be more focused on the physical content of the manuscript.

Fig 1 is not appropriate for such an article should be deleted

Equation 1 needs more explanation and discussion. Maybe a reference should be given.

The physics related to Fig 10 und Fig 11 needs to be explainede in more detail. What is the reason for the observed imrovement? What is the distribution and strength of the electric field in the device. How changes the field in the structure for each design? This part needs much more theoretical understanding and explanation. 

 

Conclusion needs to be focused on the new understanding developed in this study.

 

Author Response

Modifications Report of the Paper #electronics-765576

Response: Authors are grateful to the reviewers for the time and efforts in this paper review process. We have done with our best efforts to address all of the questions raised by the reviewers. A revision has been made by taking into account the reviewers’ comments so that it is now more readable to the reader. These updated parts were marked by some red words. Meanwhile, we have re-organized and updated this manuscript, so the page and line number of this new revision is somewhat different from the previous old version.

Response to the Comments of Reviewer :

(1). The abstract needs still to be more focused on the physical content of the manuscript.

Response:

Thanks the reviewer comment. We have added some more clearer physical descriptions in the section Abstract (Page 1) of the new revision marked by red words.

 

(2). Fig 1 is not appropriate for such an article should be deleted.

Response:  

Thanks the reviewer comment. Yes, we have deleted Fig. 1 (old version).

 

(3). Equation 1 needs more explanation and discussion. Maybe a reference should be given.

Response:

Thanks the reviewer comment. We have added some more clearer descriptions about the Equation (1) and a corresponding reference [52] is given (in P. 9) and marked by red words. And, the detailed physical equivalent-circuit model of an nLDMOS device embedded SJs in the drift region is added in Fig. 5 (Pages 5-6) in the new revision.

 

(4). The physics related to Fig. 10 and Fig. 12 (old version Fig. 11) needs to be explained in more detail. What is the reason for the observed improvement? What is the distribution and strength of the electric field in the device. How changes the field in the structure for each design? This part needs much more theoretical understanding and explanation. 

Response:

Very thanks the reviewer comment. We have added some detailed descriptions about the physics related to Fig. 10 and Fig. 12 (old version Fig. 11). Meanwhile, we added a new paragraph and Fig. 11 for the electric field reduction as inserting an SJ structure in an LDMOS device which are indicated in Lines 239-303 (Pages 9-11) and marked by red words.

 

(5). Conclusion needs to be focused on the new understanding developed in this study.

Response:

Thanks the reviewer comment. We have added some detailed descriptions about the new understanding developed in this study in the section of Conclusion (Page 11) marked by red words.

Author Response File: Author Response.pdf

Round 3

Reviewer 3 Report

After the addition of more explanations and reworking of some details the manuscript is now ready for publication.