Electrical Characterization of RF Reactive Sputtered p–Mg-In_xGa_1−xN/n–Si Hetero-Junction Diodes without Using Buffer Layer

Round 1

Reviewer 1 Report

The authors reported about the electrical characterization of p-Mg-InGaN/n-Si hetero-junction diodes prepared by RF reactive sputtering. The paper is well written and potentially interesting for p-GaN designer and manufacturer community. However, the paper consist several formal and technical weaknesses:
• The authors should highlight the novelty in the manuscript especially in the abstract, introduction and conclusion. The title should by aligned by the main contribution of the paper. Is the main contribution in electrical characterization or design and manufacturing?
• Please explain in the paper why you used thermo emission for characterization of your hetero-junction PN diodes. Why drift-diffusion is not taken in to account?
• Φb is usually used for Schottky barrier height (Metal-Semiconductor). Please indicate the Φb in the energy band diagram (Fig. 2). It would be better to depict the band diagram proportionately (the best way from exact calculated values from doping concentration). Did you take in to account the Eg lowering by increasing In mole fraction (x) in In(x)Ga(1-x)N? (please discus and refer in the manuscript e.g.: doi: 10.1557/S1092578300002301or similar)
• Please explain why Richardson constant is used for p-GaN and not for n-Si.
• The high ideality factor ~5 indicates that the tunneling is main contribution to the current (please discus and refer e.g.: doi:10.1063/1.3560919 in the paper Please discus the accuracy of the extracted Φb by neglecting the tunneling. How can it be clarified that the different currents are caused by different Φb and not by different tunneling contribution? It would be more accurate to extract the Φb by optical method (please discus and refer e.g.: DOI: 10.1007/s11664-012-2184-5)
• Fig 3a and b seem be redundant. It would be better to add comparison of the characteristics in logarithmic current scale and voltage range about from -4V to +4V where the thermo emission (or potential tunneling) is dominant.
• Please explain all abbreviations when first used (e.g. RT).

Author Response

Dear Reviewer,

Thank you for your comments, I will answer with a point-by-point response to your comments

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present the analysis of Schottky diodes which they fabricated using RF sputtering.

In the methods section, the authors mention that they carried out Hall measurements to determine hole concentration, electrical conductivity and carrier mobility. However, these measurements require four terminal devices, yet Figure 1 implies the devices are two terminal.  The authors need to detail how these measurements were carried out and also present the relevant figures in the results section. Also in the methods section, the authors detail SEM and EDS measurements that were carried out to investigate the surface morphology and compositional analysis. These figures should also be included in the manuscript to substantiate the claims made in the results section. In Figure 3, the authors show the IV characteristics on a linear scale. It would be beneficial to the reader to also show the data on a semi-logarithmic scale to highlight the saturation current under reverse bias for these devices at room temperature. On page 6, line 146, the authors have calculated effective Richardson constants based on effective charge carrier masses. The authors should include the formula that was used for determining these values. On page 6, line 156, the authors state that “the barrier height values were found to increase from 0.54 eV to 0.69 eV (device-A) and 0.50 eV to 0.62 eV (device-B). As the Schottky barrier using thermionic emission is extracted from a Richardson plot of ln(I₀/T²) vs. 1/T, as shown from Equation (2), are the authors claiming that the barrier is changing with temperature, and if so, how have the authors extracted this information? The quality of the writing is extremely poor, making it difficult for the reader to understand the work carried out and the arguments presented in the manuscript. This needs to be greatly improved before any decision on the acceptability of the manuscript can be made.  The authors should consider employing the services of a language editing service.

Author Response

Dear Reviewer #2,

Thank you for your suggestion, I will answer with a point-by-point response to your comments

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript can be accepted after the improvement of the English for the entire manuscript.

Author Response

Thank you for your comments .We also edit and check carefully English all manuscript

Author Response File: Author Response.docx

Reviewer 4 Report

The study of the sputter-deposition-made diodes based on p-Mg-InxGa1-xN/n-Si hetero-junctions looks sound enough for publication in Materials. There are multiple errors in English grammar, and the manuscript should be thoroughly proofread before publication. In the referee's opinion, the phrase in the conclusion "the obtained electrical parameters from our devices were shown very impressive" is not supported by comparison with other options and/or results of other studies. It should be elaborated what data specifically allow one to make such a conclusion. 

Author Response

Thank you for your comments .We also edit and check carefully English all manuscript

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have given a detailed response to the relevant issues raised by myself and the other reviewer. However, several explanations remain still unclear and unsubstantiated.
• Φb is the key parameter extracted for the characterized diodes. Φb should be indicated in the energy band diagram (Fig. 2) to be exact which parameter is really extracted. It is recommended to illustrate the band diagram proportionately. Please show Eg of InAlN about 3 times higher than Eg of Si. Moreover, calculate values of Fermi levels from doping concentration.
• The high ideality factor ~5 and author’s comments indicate that the thermoemission is not main current contribution. The used methodology for Φb extraction (eq.1-4) is widely used for its simplicity. However, lot of works showed that it is not accurate method and can cause wrong results mainly when thermoemission is not dominant transport. The authors should confirm the calculated Φb by other novel characterization methods.
• It is recommended to add comparison of the I-V diode characteristics in logarithmic current scale and voltage range about from -4V to +4V where the thermoemission (or potential tunneling/trap-assisted generation-recombination…) is dominant.
• If the authors will illustrate Φb to the energy band diagram it would be seen that the barrier height can decrease with the temperature only about the temperature dependence of Eg (~ -1eV/°C). The apparent rise of barrier height with temperature (Tab. 1) is most likely caused by increasing thermoemission current contribution (eq 2: I0~T^2) compared to tunneling or trap-assisted generation-recombination (I~T). It confirms the decreasing ideality factor.

Author Response

Thank you for your comments, we will answer with a point-by-point response to your comments

Author Response File: Author Response.docx

Reviewer 2 Report

The authors have answered some of the points raised previously, though there are still points that need to be suitably addressed before any decision on the acceptability of the manuscript can be made.

Within the methods section of the manuscript the authors continue to imply that the Hall measurements were carried out in this work (“In this work, our diodes were designed at the low sputtered-temperature and high pressure. The holes concentration, electrical conductivities, and mobilities of p–Mg-InxGa1-xN films and n–Si wafer were measured by Hall measurement”). The authors should not include the measurement details in this section if they are referring to previously published work. As above, if the authors are referring to previously published work, the inclusion of the measurement details in the methods section is confusing to the reader as it implies that it was carried out as part of this work. The authors have now included the semi-log plots of the IV characteristics making it clearer for the readers to appreciate changes in the current under reverse bias. The authors have now included the details of how they estimated the effective Richardson constant. Using a calculated effective Richardson constant to extract the Schottky barrier height provides an inaccurate result due to discrepancies between the theoretical and experimental A^** It is widely considered to be best practice to minimise errors on the Schottky barrier height to use a Richardson plot where the barrier height is extracted from the slope of the linear fit and the effective Richardson constant is extracted from the intercept (Missous and Rhoderick, Journal of Applied Physics 69, 7142 (1991), Guo, et al. Appl. Phys. Lett. 67, 2657 (1995)).  The quality of the writing is still poor with numerous spelling mistakes (e.g. “dioes” on page 2, line 66) and poor grammer (e.g. “In recent years, the success of Mg doped-InxGa1-xN films has been greatly benefits for developing LEDs, electric devices, and photo detector” line 39, page 1), making it difficult for the reader to understand the novelty of the work carried out and the work presented in the manuscript. This still needs to be greatly improved before any decision on the acceptability of the manuscript can be made.  The authors should consider employing the services of a language editing service.

Author Response

Thanks you for your comment

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

Not even second time the authors implemented all recommendations raised in previous reviews. In particular the results haven’t been sufficiently enhanced about additional analysis and interpretations even after two rounds of revisions. In my opinion this is a borderline paper. At least define the key parameter Φb in the energy band diagram (Fig. 2).

Author Response

Thanks you for your comment

Author Response File: Author Response.docx