Highly Responsive Gate-Controlled p-GaN/AlGaN/GaN Ultraviolet Photodetectors with a High-Transmittance Indium Tin Oxide Gateâ€
, Xiangdong Li 1,2,*, Hongyue Wang 3, Yuebo Liu 3,*, Weitao Yang 4, Zesheng Lv 5, Meng Wang 1, Shuzhen You 1,2, Jincheng Zhang 1,2 and Yue Hao 1,2Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis letter presented a transistor-based UVPD, which is of great interest. The authors show the feasibility of fabricating high-performance UVPD based on the available e-mode technology. There is only few things that requres further elaboration.
1)The origin of the photocurrent should be well explained.
2) A comparison should be listed with other reports.
Author Response
Due to the modification of the design images, the comments were placed in the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors proposed and demonstrated a device structure of UV photodetector that achieves a high responsivity at the UV range with a small threshold voltage and low dark current compared to the reported data in the literature.
Detailed concerns to be addressed are listed below before they can be published on Micromachines -
Line 41 - references should be placed for avalanche photodiode and PMT for completeness.
Line 43 - a brief description of WHY HEMT with AlGaN/GaN is superior should be added.
Line 77-79 - please comment on if these steps are indeed necessary to deplete the 2DEG. Are any control experiments done?
Fig1. (a) - a descriptive word, eg UV, can be added to the UV light illustration.
Line 81 - while Ni/Au (10 nm/10 nm) is expected to have a smaller transmittance than ITO given its metallic nature. However, if it can made thinner, the transmittance can be larger and even comparable to ITO. Experiments or references should be addressed for this. Also, what is the trade-off of using ITO, eg., does the work function mismatch become worse compared to Ni/Au? Authors should emphasize this through experiments (eg, supporting data with the same structure but using Ni/Au) or references.
Fig1. (c) - Can authors comment on the ITO/glass transmittance? In the literature, ITO/glass usually has a higher transmittance than what authors are reporting in their results, maybe by ~20%. Is this ITO quality good, how thick is it? Together with this comment, authors should also comment on why high-temperature evaporation is better for the conductivity of ITO (line 88) - does this have any trade-off, maybe the transmittance? More comments are suggested to enhance the quality of the paper here.
Fig1. (b) - what is the layer on top of ITO gate plate? It has a corrugated boundary with the ITO, does this affect the reliability/repeatability of the device? Please comment.
Line 104 - 105, please comment on the reason that 30 s annealing 865 oC can achieve ohmic contact with supporting references.
Line 103 - please comment on the reason for the metal stack selection also their thicknesses. Some references might be sufficient.
Line 109 - ITO was mentioned to be evaporated in the aforementioned text. It is mentioned using sputtering. Which one is the actual way? And is the substrate temperature heated to 350 oC during your sputtering?
Line 116 - Reference for the criterion of Ids should be included.
Fig.2 Caption should specify what PDCR is for clarity.
Fig.2 - I suggest the responsivity vs. intensity figure can be plotted to check the linearity of the device, and therefore a detection limit can be estimated as well by extrapolation.
Eq 1 is redundant - it is too obvious to be shown.
Fig.3 a-b - it is understandable that a wavelength larger than 355 nm cannot provide the bandgap energy for carrier generation. However, it seems that the wavelength below 355 nm also shows similar performance. In other words, it seems to have a "resonant" wavelength at exactly 355 nm that has a very low Vth. Please clarify more.
Line 171 - what specially designed structure? Why it can achieve a fast response? Please clarify.
In the section explaining the mechanism, comments should be made for comparison to other contacts, for example, Ni/Au that ITO was set to be compared to. How those unselected contacts are evaluated in the context of this device structure? Other than ITO, IZO is also a material that has high transparency, how is it compared to ITO? These comments are crucial for landing a design rule for an ideal device strcuture that the results can imply.
Comments on the Quality of English Language
Overall okay.
Author Response
Due to the modification of the design images, the comments were placed in the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsITO gated p-GaN/AlGaN/GaN ultra-violet photodetectors are reported in this work exhibiting competitive device performance. The operation principle of the devices was well explained from energy band diagram point of view. The topic is of interest to the field of UV photodetectors and the paper is written in clear, concise language. However, the reviewer has the following questions that need to be addressed.
1. Since there have been reports using the similar structure without the ITO gate to achieve comparable or even better device performances according to Table. 1, it is unclear what the motivation is for adding the ITO gate on top. What advantages is the ITO gate expected to bring?
2. What is the underlying physics behind the plateaus in the transfer curves near threshold under light exposure, especially under high light power densities [Fig. 2(a)]?
3. It is understandable that at high wavelength the photon energy is smaller than GaN bandgap and the photo response should quench. However, why are the photo responses in Fig. 3 also very weak when the wavelength is smaller than 355 nm (photon energy larger than the bandgap)?
4. The colors of different curves in some of the figures are too close to distinguish. For example, the curves showing 280 nm, 370 nm and 450 nm have almost the same color in Fig. 3(a). Please adjust the color schemes of such figures.
5. What limits the speed of these UVPDs? How does the response time compare with similar detectors in literature?
Author Response
Due to the modification of the design images, the comments were placed in the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThanks for addressing the concerns. The paper has been improved.
Final loose ends following the revision -
Comment 7 - I suggest including the explanation in the figure caption for clarity.
Please include error bars in Fig3 and Fig4.
Author Response
Please download the attached attachment to read
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors have addressed most of my concerns and revised the manuscript to strengthen it. The manuscript would merit publication if the authors could address the remaining comment as follows:
1. The reviewer is not fully convinced that “a plateau period appears due to the saturation of the photocurrent in the light-controlled conduction process.” As seen in Fig. 2(c), the photocurrent does not seem to saturate before the device is turned on. The authors need to explain why the photo-induced drain current in Fig.2(c) has a two-step behavior under high illumination.
Author Response
The specific response content is designed with a chart, so please download the attached attachment to read.
Author Response File: 
Author Response.docx
