Electrical and Hysteresis Characteristics of Top-Gate InGaZnO Thin-Film Transistors with Oxygen Plasma Treatment Prior to TEOS Oxide Gate Dielectrics

Round 1

Reviewer 1 Report

The present manuscript by Tien et.al. presents an investigation of the influence of Oxygen plasma on the IGZO semiconductor, prior to the dielectric deposition in a top-gate configuration. The authors have optimized the O2 plasma treatment by which they could achieve a reduced charge-carrier concentration of the IGZO back-channel. This in turn yields an overall improved thin-film transistor performance. The nature of the improvement has been well-investigated from a material interface standpoint by using AFM and XPS analysis to determine the surface morphology and change in chemical species pre and post O2 plasma treatment. The findings presented herein are certainly relevant to the readership of Coatings, especially to the oxide electronics community. However, prior to publication, the following areas should be addressed:

1. L-154-155: Can the authors explain why does the gate leakage rise one order of magnitude?, since the TEOS and the Top Gate are deposited after the O2 plasma treatment
2. Have the authors considered the impact of Hydrogen from the TEOS incorporated into IGZO during annealing? If it should be briefly mentioned for future considerations, since the choice of SiO2 precursor on the top insulator is also known to have H doping effects which could be detrimental or favourable depending on the nature of the chemical species of the IGZO.
3. In Fig. 7, the on current seems rather constant in the saturation regime. If this is due to a compliance cut-off to detect the low off current, please mention it briefly. This would help the reader understand better.
4. The schematic process flow chart has a very poor resolution and should be exchanged for a higher resolution graphic
5. 2 and L 133-135: Please discuss what causes the increased roughness of the IGZO films? This would help to clarify right at the start that the 60s treatment is the optimum balance for cleaning the residues and generating sufficient reactive sites on the surface of the IGZO back-channel for the initial TEOS reactions.
6. L-266. Please make it clear if the O2 annealing reduces the interfacial trap density or the bulk. The present statement seems rather general.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

In figure 3. O2 plasma 0s on current is higher than then O2 plasma 60s, but the O2 plasam 60s TFT field effect mobility is higher than the O2 plasma 0s. author need to  explain this point.  such as, estimated the field effect mobility max or estimated mobility at constant VGS or others. 

Author mention page 5 (line172), all devices show ohmic behavior indicating a good contact between a-IGZO and MO. but Figure 3 results O2 plasma 60s TFT are cut off behavior. it seem to be a high resistance in N+IGZO(IGZO/Mo contact area), due to the O2 plasma treatment. Even, output curve (Figure 4b) not a liner in low VDS (drain voltage). not easy to mention for the ohmic contact . Author need to explain this issue.  

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This work investigates the effect on top-gate IGZO TFTs of both an O2 plasma treatment (in the semiconductor/gate dielectric (GD) interface) and of an annealing in O2 atmosphere (after the deposition of the GD).

A few points should be made clearer:

1. The O2 plasma treatment seems to be effective in modifying the a-IGZO material, decreasing the oxygen vacancy concentration. This results in an interesting trade-off between the damaging of the IGZO layer and the decrease of carrier concentration (increase of Vth). In addition, there seems to be an optimal treatment time for S.S. also. The authors should comment on why S.S. decreases from no treatment to 60 s treatment time.

 2. It is not clear why does de O2 plasma treatment (before the gate dielectric deposition) affects the gate dielectric composition. Please clarify whether the analysis of the carbon and oxygen % (and their relation) is related to the dielectric/IGZO interface or if it is across all the dielectric.

3. The other metrics (besides hysteresis) from the non-annealed TFT could be included and compared with the annealed device.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

This is a well-done experimental work dealing with the effect of technological low pressure plasma treaetment on the properties of an electronic device. However, this work establishes only phenomenological correlations without adding new scientific insight. Also, it does not generalize the results by comparison with a large number of similar previous publications. On the other hand, there is some redundancy with Ref. 20 limiting novelty of this work. I think that there are some important problems that need to be clarified before I  can suggest this work for publication:
1) The authors should give more details on the used deposition equipment. In row 101 they mention an Oxford PECVD system, in the acknowledgement a RF magnetron sputter and ICP system. I guess a-IGZO were sputtered while plasma treatment and SiO₂ deposition was performed in a PECVD chamber. Were these different facilities and processing was performed interrupting vacuum or was it a multichamber deposition cluster (cf. rows 216 to 219) ? This is important for interface formation.
2) O₂ plasma treatment increases surface roughness sufficiently already after 60 s. On the other hand, a higher I_off and a deteriorated subthreshold swing were obtained only after 120 s plasma treatment. Moreover, the characteristics in table 2 show an inconsistent behaviour, e.g., V_th, S.S. and µ_FE are improved after 60 s plasma treatment. This needs an explanation. A model explaining the effect of surface roughness on transistor characteristics or a corresponding reference to literature should be given.
3) What means O/Si > 2 for SiO₂ ? Is this an artifact of SIMS ? Please give an explanation.
4) Conclusions should be more generalized and include that O₂ plasma treatment increases surface roughness, reduces carbon content, improves SiO₂ film stoichiometry.

- ...few literatures ... , a reference should be given at the end of this break in row 80.
- Figure 2 should be of better quality, so that the scales of the x,y and z-axes are visible.
- formate SiN_x in Ref. 7.
- english style needs some small improvement, cf. ... few literatures, fabricated at, etc.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have done a good job at providing the necessary clarification and additional inputs within the revised version of the manuscript. I am satisfied with the present quality of the manuscript and can recommend it for publication in Coatings.

Reviewer 4 Report

The manuscript was improved in part by considering my comments. However, the use of different deposition facilities (RF magnetron sputter and Oxford PECVD system) should be mandatory included into the manuscript text. Also, I recommend to assign a O/Si ratio of 2.64 as an appearant one. After this minor revision, I recommend the paper for publication.