Identification of Buffer and Surface Traps in Fe-Doped AlGaN/GaN HEMTs Using Y₂₁ Frequency Dispersion Properties

Round 1

Reviewer 1 Report

The reviewed manuscript presents interesting studies on buffer and
surface traps in GaN/AlGaN/GaN high electron mobility transistor. The
authors reproduced the electrical characteristics of  GaN/AlGaN/GaN 
HEMTs (output and transfer I-V curves) using the TCAD Modeling and on
this basis, they determined the surface/buffer trap parameters.  The reviewed manuscript can be considered for publication in Electronics after taking into account the following issues.

The physical model of the GaN/insulator interface employed by the authors in the TCAD simulations is not realistic. This model included only one discrete donor-like state with a fixed charge, which compensates the negative polarization charge. On the other hand, it is well known that the interface states at various GaN/insulator interfaces including the SiN insulator exhibit a broad continuum distribution in the band gap. In particular, near the conduction band, there is a continuum of acceptor-like states with or without a discrete donor-like state (mostly nitrogen vacancy, VN). It should be noted that the nature of the donor-like state, which compensates the negative polarization charge is still under debate. The authors in the revised manuscript should include the continuous interface states together which discrete one (which compensates the negative polarization). If the authors cannot apply a model of continuous interface states, they should provide a discussion about such states including comment on how their influence the I-V characteristics, for example. More details about the discrete and continuous interface states the authors can find in the paper by M. Ťapajna et al. [Journal of Applied Physics 116, 104501 (2014)] and by M. Matys et al. [Journal of Applied Physics 120, 225305 (2016)]. Please, provide the discussion as mentioned above, about the continuous interface states based on these two papers.

Author Response

Prelude : At the outset, we thank the reviewers for their useful comments. We have revised the manuscript based on the reviewer’s comments. Specific responses to the reviewer’s comments are attached herewith .

Please see the attachment

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors have previous work on characterizing trapping phenomena in GaN HEMTs. They also have a published report on identifying GaN buffer traps analyzing the LF-Y ₂₂ parameters. This manuscript presents a thorough follow-up research to their previous work, in which they utilize Y₂₁ Frequency Dispersion Properties to characterize the buffer as well as surface traps in AlGaN/GaN HEMTs. They have provided in-depth analysis and detailed study of their experimental procedure, results, and findings. The authors have also included simulation results to substantiate their findings. The introduction is thorough and discusses the previous relevant works in this field while also explaining the significance of this work. The conclusions drawn from their findings are well supported. While I think this manuscript can be accepted as is, I have a few questions and the answers to those would greatly help my understanding. These questions are listed below.

1. Can a design guideline be developed from these studies to improve device performance?
2. It would be better if the authors include a few lines explaining or justifying their use of dopants (Fe in this case). How the use of other dopants would change the characterization of the traps, if at all?
3. The authors have included temperature dependent study. Is there any dependence on aging and on the presence of light for the trapping behaviors? Can a time dependent study be modeled here or in a future repot?
4. The manuscript can be revised for minor spelling and grammar checks.

Author Response

Prelude: At the outset, we thank the reviewers for their useful comments. We have revised the manuscript based on the reviewer’s comments. Specific responses to the reviewer’s comments are attached herewith.

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This is an interesting work for those focusing on GaN transistors.  This work will be more complete, if the authors provided the device process details and buffer GaN doping levels. This is the key for this paper.  

1. Reviewer understand the restrictions of Intellectual property protection. But the current version of the paper is not accomplished with the completeness without the epi-layer properties and device processing details.  Authors can think of providing the Fe-doping profile (its concentration) of the used transistors.
2. Authors claims that Ec-0.45 eV is an acceptor-like buffer traps and Ec-0.2 eV is a surface donor states.  If it is a acceptor-like buffer traps, it supposed to be Ev+X.XX eV.  How the authors are claiming this trap is related to acceptor-like buffer traps.  This needs to be clarified in the revised manuscript. 
3. What is the physical origin of the acceptor-like buffer traps and surface donor states.  This has to be explained in the revised manuscript.
4. Authors are also required to provide the current collapse characteristics of these devices having both surface and buffer related traps.  Which one is dominating for the observation of current collapse?
5. Ref #14, (Benvegnù et al.) explained the details of the measurement concepts.  This paper explains the identification of surface and buffer traps.  In the Ref #14, they have given 0.64 eV as a buffer traps (Other research group has also observed the activation closer to 0.64 eV, Refer the papers DOIs: (i) 10.1063/1.1571655 & (ii) 10.1063/1.1512820 with the trap nearer to the valence band +0.61 eV).  However, authors have got the activation energy of Ec-0.45 eV which is much different from the prior reports. Authors are required to clarify this point in the revised manuscript.

6. Authors claims that EC – 0.2 eV (D1) is originated from the surface of HEMT structure.  Is it from passivation layer or barrier AlGaN layer.  This has to be clarified in the revised manuscript. 

Author Response

Prelude: At the outset, we thank the reviewers for their useful comments. We have revised the manuscript based on the reviewer’s comments. Specific responses to the reviewer’s comments are attached herewith.

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Authors have covered most of the reviewer's comments.  The revised manuscript can be acceptable for publication.