Design of Diamond Power Devices: Application to Schottky Barrier Diodes

Round 1

Reviewer 1 Report

Diamond exhibits excellent optical, electronic, and thermal properties as the ultimate semiconductor, so that it expected to be a good candidate for high power, temperature application in future electronics. In this article, for the diamond Schottky barrier diodes, the authors investigate the tradeoff between On- and Off-state and  the optimization of design of floating metal rings to improve the effective breakdown voltage, using 1D analytical models and 2D numerical models, respectively, aiming to better figures of merit. Considering the above importance, I would recommend the manuscript to be published for Energies, although some corrections and explanations for queries as following should be made previously by the authors:

1) The authors should correct/modify the typo of unit, e.g., the bullet sign representing the product (I see no sigh or the period instead of the bullet sign for the applicable positions). In addition, the typo of PiN may be not general but PIN or Pin may be usual.　

2)  For usage of the 1D analytical model, the authors seem to refer to the references. Description of the analytical equations and solutions are helpful for the readers to understand the manuscript. 

3)  In fig. 13, how do we qualitatively interpret the optimization for the number of FMR of two in Design #2? This optimization is for the special case of the Design #2, where any explanation of effect of the thickness, the doping amount  on the number of FMR  is desirable.

Author Response

Thank you for your feedback. We have carefully taken your suggestions into account.

1) Bullet signs have been used throughout the text.

2) We agree that providing the models would increase further the archival value of our contribution. However, the paper is already quite long, with many new results and analyses. Adding the details of the models would make the key results more difficult to catch. Moreover, the models are quite common, with many references provided. We think that figures 2,3,4,5,6 and tables 2,3 provide more important information than recalling the classical models already published many times. Nonetheless, we have provided new equations (1) to (6) with key models related to P-type (boron doped) diamond: free hole concentration (1-3) and doping and temperature dependent mobility (4-6), and the values for the physical constants in the text.

3) Fig 13 shows that, with the lower doping levels ("Design #2"), only one FMR leads to a lower BV than 2 FMRs, and that 3 FMRs shows a lower BV than with 2 FMRs. The reason is that with just one FMR, the peak electric field at the edge of the sole FMR towards the cathode contact has been successfully reduced, but the peak electric field at the other edge of the sole FMR is actually larger. This leads to a premature breakdown. Please look at figure 9a), where one can see that the electric field can be higher at the FMR edge than at the cathode contact (e.g. black curve, Ws=0.15 µm, labeled 4.6MV/cm at the cathode edge, whereas the peak electric field at the edge of the FMR is larger than 6MV/cm). Consequently, adding a second FMR is efficient to reduce the peak electric field at the edge of the first FMR, leading to an improvement of the overall BV capability. A short qualitative analysis has been added after fig 13 (line 380-ish).

Reviewer 2 Report

The paper entitled “Design of diamond power devices: application to Schottky barrier diodes” by N. Rouger and A. Merechal is an interesting overview on the application of different designs diodes.

It deserves to be published on Energies after some revisions.

In particular:

-          Being the paper focussed in the power electronics field, the reviewer thinks that a comparison with other semiconductor materials (Si, 4H-SiC, GaN ) can help the reader to understand the benefits of diamond.

-          A summary of known dopants species both on p-type and n-type can be included in the manuscript.

-          A table with the literature Schottky barrier values obtained using different metals.

-          Please increase the font size in the figure labels.

Author Response

Thank you for your comments and suggestions.

(1) Indeed, this is an important remark. We do agree that it is important to further benchmark the performances of diamond vs. other materials. However, our contribution is already quite long and our aim is to show the best performances we could get from diamond Schottky diodes (design/optimization), with the specific tradeoff related to diamond. The reader can now take the quantitative information to further benchmark with other materials.
In this context, figure 2 already shows the performances of diamond (Figure of merit). Figure 14 is the result of a design optimization and clearly shows the possible performances of diamond Schottky diodes.
Nontheless, a short comparison has been added (line 128).

(2) A specific sentance has been added (Line 132)

(3) Line 195: Schottky barrier height ranges have been added

(4) FontSize has been increased.

Reviewer 3 Report

Dear Authors,

 The proposed paper is a good and well written report which I find interesting and straightforward to read. It discusses a up-to-date topic in power electronics. Since the requirments of power electronic devices continue to grow, it is necessary to develop better semiconductor materials. The paper consider four types of barrier diodes based on diamond. The wide introduction contains a disscusion about the trends and main drawbacks  in diode design and the necessity of trade-offs made for a successful and proper device design. A comprehensive simulation analysis has been conducted for  properties characterization of the investigated devices. An analysis of electric field distribution in the 2D structure, termination efficiency,  BV improvements and Ron resistance were carried out.

However there are minor issues which need to be properly addressed before so:

1)  table 1 - it is difficult to understand this table. This table should be reorganized.  Almost all parameter values are given at 300 K, therefore this information can be moved into the caption of the table.

2) Correct the references to the figures in the content of the paper.

Best regards,

The Reviewer.

Author Response

(1) -> Table 1: 300K  has been removed.
Indeed, it is quite difficult to compare different devices from different papers, where the measurement conditions are most of the time different (e.g. bias)
We tried to improve the table by highlighting the key information and modifying the layout.

(2) -> References to figures have been checked and modified.