Merging the Sol–Gel Technique with the Pulsed Microplasma Cluster Source Deposition to Improve Control over the Memristive Response of TiO₂ Thin Films

Round 1

Reviewer 1 Report

The authors present the merging the sol-gel and pulsed microplasma cluster source deposition techniques to fabricate TiO2 thin films. The authors expected a better quality film via hybrid approach to have a better control over the memristive response of TiO2 films. There are many issues that need serious attention for publication. Hence, I want to reconsider the paper publication based on the major revision of the following comments.

1. There are many structural and grammatical errors. The manuscript need to be checked with the linguistic expert or MDPI English editing service for clarity.

We carefully checked the English language quality in the paper and, according to the Reviewer’s requirement, we extensively revised the whole manuscript with major revisions of several sentences in the text.

2. The authors have sufficiently explained the aims and goal of the study. However, the abstract is void of the most important experimental results grabbed from the hybrid approach.

We agree with the Reviewerabout the lack of description of achieved results in the abstract. We thank her/him for this comment and changed the abstract (lines 17-26) to improve visibility of achieved results, as follows: “In this paper, we report on an innovative approach by combining the sol-gel and the pulsed microplasma cluster source (PMCS) methods, exploiting the low temperature and low cost of the former process and precise control over nanocristallinity of the latter. We show that this approach overcomes the reported limitations that each technique shows in fabricating memristive devices when independently used. A side-by-side comparison of the TiO2thin films produced by the PMCS, sol-gel, and PMCS/sol-gel hybrid methods (HM) demonstrates an improvement of the memristive properties and a reduction of the electrical shorts in the TiO2based devices.”

3. Introduction line 34-40: How does the authors differentiate between physical and chemical methods? I believe every growth method involves some chemical reaction for film deposition? Clarify the intended meaning by readers' understanding.

We agree with the Reviewer, the distinction betweenchemical and physical methods can bemisleading. The first methodleadsto the synthesis of materials starting from specific precursors, such as wet chemistry approaches in air (sol-gel) as well as in vacuum (ALD, metal organic chemical vapor deposition). Differently, typical physical methodsuse physical vapor of the studied material, such as evaporation/sublimation, sputtering occurringin vacuum,but that can also take advantage of chemical reactions in the presence of specific gases. To remove all possible misunderstanding, we changed the text from line 38to 45as follows: “Driven by this interest, a variety of thin films with a typical thicknessofaround tens of nm was fabricated by methods where synthesis is achieved by chemical reactions with specific precursors or exploiting direct physical deposition of material in vacuum. In particular, TiO2thin films were successfully fabricated by atomic layer deposition (ALD) [10,11], reactive magnetron [12], ion beam [13], radio frequency (R.F.)sputtering [14] and laser ablation [15]. In addition to these methods, TiO2thin films were fabricated by the sol-gel approach [16] and metal organic chemical vapor deposition [17].”

4. I want the authors to strategically plan the introduction section. Follow the MDPI instructions to divide the section into well pronounced paragraphs. Provide the novelty of this study and the most important results in the last paragraph of introduction section.

We agree with the Reviewer;the original form of the introduction did not properly evidence the aims and conclusions. Accordingly, we introduced new paragraphbreaks at lines 41 and 57 and modified the last paragraph to better evidencethenovelty of our studyas follows: “In this work, we propose merging the PMCS and the sol-gel techniques to profit from the methods’ advantages and overcome the specific limitations. We present a proof of concept and feasibility study showing that this hybrid method is a versatile tool for developing the target TiO2thin films with a memristive response. A side-by-side comparison of the TiO2thin films produced bythe PMCS, the sol-gel, and the PMCS/sol-gel hybrid method (HM) shows that the new approach leads to materials with memristive switching, a large reduction of electrical shorts probability, and improved device electrical performances."

5. Define all the abbreviations at the first instance of their usage in the manuscript.

We thank the Reviewerfor this comment. We controlled the definition of all abbreviations, also removing unnecessary repetitions,introducingtext changes in lines39, 57, 71, 113, 117, 197, 214.

6. It is difficult to distinguish which scale bar is 1 microns and which one in 500 nm. It is better if the authors provide the exact length of the scale bar within the SEM figure itself for clarity.

We agree with the Reviewer, so weintroduced the scale number in each image in Figure 2.

7. Did the authors measure the film thickness via cross sectional SEM images? I think it is inevitable to measure so for effective memristor device study.

We measured film thickness using a TENCOR P6 mechanical profilometer, as reported in line 104, Materials and Methods section. Concerning the HM film’sthickness, we evaluated the PMCS first layer thickness and the final two layers thickness, as reported in Table 1. We did not measure via cross sectional SEM images, but we agree with the Reviewerthis could add more information. Nevertheless, we based our work on well established TiO2film growth methods, and materials have been extensively studied by different techniques (see Refs. 16, 18-22). On these bases and considering all presented results, we believe that weclearly and reliably demonstratedthe feasibility of the proposed hybrid approach. 

8. The whole idea to use the hybrid method was to control the material stoichiometry. How did the authors conclude the stoichiometry ratio in the grown materials? I recommend the authors to perform XRD, TEM and EDS analysis for better understanding of the crystal structure of the TiO2 films fabricated via hybrid method.

We thank the Reviewerfor this valuable comment. The stoichiometry control is one of the main features of the PMCS method, together with the ability to grow nanocrystalline film at room temperature. Thiswork aims at verifyingthe feasibilityand validityof our new hybrid method combining sol-gel and PMCSto exploit both approaches towards reducing intrinsic weaknesses whileincreasingtheperformances of memristive devices.From the results of TiO2thin films produced by sol-gel and PMCS, the main issues wereimprovingthe memristive properties and reducing theelectrical shorts, respectively. We focused on these criticalpoints, consideringthat control of stoichiometry given by PMCS would be an adjoint value but thatit was mandatory,first of all,to verify the reliability of the hybrid method. Thus,we decided to use only PMCS TiO2film with onestoichiometry, the nominal one, leaving the intriguing and promising use of films with different oxygen atomic percentages to further development, as claimed in the Conclusions (lines 319-323). It is useful to remember thatwe based our work on well established TiO2film growth methods. We extensively studied materials using different techniques (see Refs. 16, 18-22) to analyze 
structure, morphology, stoichiometry, achieving a high control and reproducibility of the material properties. Thus, several analyses have not been repeated, as TiO2H is just the superposition of two already studied TiO2films. Moreover, some techniques such as XRD and EDS would give averaged results of the sol-gel and PMCS films. However, as underlined adequatelyby the Reviewer, some phrases in the paper can lead to the idea that stoichiometry will be explored.In particular,in the Abstract, line 17-23, that we changed as previously described in response toPoint 2,removing any claim about stoichiometry to avoid any possible misunderstanding due to misleading statement.

9. The authors need to improve the overall quality of all the figures. The graphs are very blur and least magnified. There is an unwanted line beneath all the graphical figures. 

We agree with the Reviewer for this suggestion. Accordingly, we increased resolutionof Figures 3, 4 and 5as well as of Figure 2, as discussed in response to Point 6.

Reviewer 2 Report

The review article "Merging the sol-gel technique with the pulsed microplasma cluster source deposition to improve control over the memristive response of TiO2 thin films” by Valentina Prusakova, Giovanni Giusti, Cristian Collini, Giancarlo Pepponi, Mario Barozzi, Leandro Lorenzelli, Salvatore Iannotta, Roberto Verucchi, and Sandra Dirè describes a combination of the sol-gel and the pulsed micro plasma cluster source (PMCS) methods, exploiting the low temperature and low cost of the former process and a precise control over stoichiometry of the latter. The aim of their work was to overcome the limitations of both approaches in the development of the TiO2 films with efficient memristive performance. The authors reported a side-by-side comparison of the TiO2 thin films produced by the PMCS, sol-gel and PMCS/sol-gel hybrid method (HM). They concluded that these combined results reported by them prove that they were able to combine and exploit the best properties of both uniform types of TiO2, or strengthen their weaknesses using the proposed hybrid method approach to build memristive devices with improved performances. In my opinion the work presents a scientific value and I recommend to publish it after minor corrections of text editing.

My comments are following:

1. Lines 173: It is “Figure 4, 5, 6” it ought to be “Figure 3, 4, 5”

We are sorry for this issue and thank the Reviewerfor her/his suggestion, we changed text in line 173.

2. Figure 3 should be marked as Figures 3a, 3b and 3c. The same remark applies to Figures 4 and 5.

We agree with the Reviewer, the suggested figure marking improves clarity and ease of reading. Moreover, we changed text in lines 193, 194, 198, 204, 205, 206, 207, 210, 211, 215, 227, 228, 274.

3. Reference number 14 lacks the name of the first author.

We thank the Reviewerfor her/his important comment. We used Mendeley, a bibliography software package,to managereferencesand avoid issues.Nevertheless,some missing information or wrong citation occurred. The case of Ref.14 has been appropriately corrected (see lines 375-378). Moreover, we checked all references and introduced missing data in Ref.13 (lines 373-374), Ref.15 (lines 380-381), Ref.21 (line 399) and Ref.26 (lines 411-413).

Reviewer 3 Report

The authors proposed a hybrid method combining sol-gel method with pulsed microplasma cluster source deposition to improve control over the memristive response of TiO₂ thin films, and the structure and electrical properties of the TiO₂ thin films produced by the PMCS, sol-gel and PMCS/sol-gel hybrid method were compared. The manuscript shows some interesting results. But I don’t recommend the publication in coatings. Some suggestions are shown below.

1.All of the figures are not clear in the text, especially Figure 2d. The pictures are too fuzzy to show good grain effect.

We thank the Reviewerfor her/his suggestion, we improved graphical quality of all figures(Figure 3, 4, 5), including SEM images (Figure 2) where contrast has been enhanced to show the presence of grainsbetter.

2. In the Figure 3, Figure 4 and Figure 5, the horizontal and vertical coordinates of the data graphs of the three comparative experiments are different, and the pictures are underlined, and part of the I-V graph seems to be cut off.

We thank the Reviewerfor thesecomments;accordingly,we improved graphical quality of figures 3, 4 and 5, removing all artifacts. Regarding the different vertical and horizontal coordinates, we agree with the Reviewer and for R-V and device life curves in Figures 3, 4, 5 b and c we modified vertical coordinates using the same scale. However, for the I-V curves(Figures 3, 4, 5 a), due to the different currentandvoltage rangesfor the three types of film are different (as can be seen from Tab.1),we adoptedscales to emphasize the curve’s ease of interpretation rather thanacomparison between the films.

3. Regarding the electrical performance characterization of the three kinds of TiO₂ thin films, there are too few data, and more characterization is needed to highlight the advantages of the hybrid method.

This work aimedto mainly verifyingthe feasibility of our new approach, i.e., to combine and exploit the best properties of TiO2deposited by the two different methods, reducing intrinsic weaknesses to improve performances of memristive devices. Our experiments’bottom electrode/top electrode scheme was adopted with this aimand fitted our aims very well. To achieve extensive electrical characterization,a cross-bar/cross-point array is mandatory to investigate in detail reproducibility, cycle-to-cycle current variation(see for instance Coatings 2020, 10, 765; doi:10.3390/coatings10080765). However, such a complex architecture requires information about the feasibility of the hybrid approach thatcan be better and more reliably achieved with the used electrode architecture. Moreover, considering that one of the main problemsin TiO2films by the PMCS method was the presence of electrical shorts, a complex cross-bar architecture would be practically useless. The resultsacquired in this work arethe key pointto promote the use of the hybrid approach, as also claimed by the Reviewer, and will be more extensively studied in the future.To better emphasize our main goal, we modified the last paragraph of the introduction as follows: “In this work, we propose merging the PMCS and the sol-gel techniques to profit from the methods’ advantages and overcome the specific limitations. We present a proof of concept and feasibility study showing that this hybrid method is a versatile tool for developing the target TiO2thin films with a memristive response. A side-by-side comparison of the TiO2thin films produced by the PMCS, the sol-gel, and the PMCS/sol-gel hybrid method (HM) 
shows that the new approach leads to materials with memristive switching, a large reduction of electrical shorts probability, and improved device electrical performances.”

4. There seems to be something wrong with the layout of the chart in the text.

If the Reviewer refers to description of Figure 1,we modified caption text as follows to improve clarity: “A schematic representation of the three memristive cells (top) with the TiO2layers produced by (from left to right) the PMCS (TiO2P), the hybrid (TiO2H), and the sol-gel (TiO2S) methods, together with the images of the different type of devices (bottom).”

Round 2

Reviewer 1 Report

The authors have provided a detailed point by point response to the review comments. The manuscript quality has been improved a lot and the structure is better. Most of the comments in review round one was on minor nature, asking for reshuffling, checking grammar and text editing. Only two comments demanded major revision including the SEM cross-section and justifying the nanocrystalline stoichiometry via certain characterization techniques. I am happy that authors have addressed all the minor comments, however, the major comments are not addressed and neglected. As stated previously, the stoichiometry of the formed nanocrystals is one of the most important factors in studying any application. The memristive and electrical characteristics all depend on the inside chemistry, defect states and stoichiometry of the materials. Simply removing the word stoichiometry from the manuscript does not serve the purpose. I want the authors to study the crystalline quality and stoichiometry of the, at least, formed TiO2H nanocrystals and compare them with the references of the previously published papers on TiO2S and TiO2P. In this backdrop, if the authors believe that EDS and XRD provide just a basic level information on the crystal structure and stoichiometry, I recommend the authors to examine the samples with TEM, high magnification SAED and XPS analyses.

First of all,we want to apologize if the Reviewer could have interpreted the removal of some phrases about the stoichiometry issue from the Abstract and Introduction sections as disrespectful. This was not our aim.On the contrary,the Reviewer’s comments helped remove possible misunderstanding about the study of stoichiometry in our work, whichwas not our goal.FollowingtheReviewer’s suggestions, we performed an XPS analysis oftheTiO2Hsurface. We found chemical properties typical of a stoichiometric titanium dioxide, without oxygen defects and presence of only Ti+4species(no other oxides are present), in agreementwith reference data (see new ref. 24). Moreover,results are also in agreement withprevious XPSanalysis ofTiO2Ssurface(see ref. 23), suggesting a good reproducibility of material properties and the synthesis technique’s control. We think these results satisfy the fundamental request about stoichiometry study. However, we are sorryto inform theReviewer that itwas impossible to perform otheranalyses, such as XRD or TEM, due to the limited time availablebefore submission and, most of all, to the limited availability of laboratory access due to the Covid pandemic.We hope the results we present will convince the Reviewer ofthe reliability of our approach. Accordingly,we introduced the following changes:
• In Materials and Methods section, lines 103-109: “X-Ray Photoelectron Spectroscopy (XPS) has been carried out in an Ultra-High-Vacuum (UHV) system, equipped with a non-monochromatized X-ray source (Mg Kα photon at 1253.6 eV) and a VSW HA100 hemispherical analyzer, with a total energy resolution of 0.86 eV [22]. The binding energy (BE) scale of XPS spectra was calibrated by using the Au 4f peakat 84.0 eV as a reference. The core level analysis has been performed by Voigt line-shape deconvolution after background subtraction of a Shirley function. Stoichiometry was evaluated using sensitivity factors corrected by analyzer electron transmission.”
• In Results section, lines 161-170: “We previously studied the chemical properties of TiO2S and TiO2P [18-20, 24]. Thus, we analyzed only the TiO2H surface by XPS. Figure 3ashows a wide range spectrum, put-ting in evidence the presence of O1s, Ti2p, and C1s core levels. Ti2p feature is composed of a 3/2-1/2 doublet, located at 458.56 and 464.30 eV (Figure 3b), the typical BEs of TiO2 [18-20,24]. No other components are present, suggesting the absence of other types of tita-nium oxides. O1s core level (Figure 3c) is characterized by a main peak at 529.83 eV, re-lated to oxygen in TiO2, and a second feature at 531.37 eV, due to CO groups in agreement with the presence of the C1s core level. The energy difference between Ti2p 3/2 peak and O1s component 
related to Ti-O bond is 5.72 eV, while O/Ti stoichiometry ratio is 2.03. All data are in good agreement with the expected values for TiO2 [25].”
• In Discussion section, lines 276-280: “TiO2H surface has stoichiometric titanium dioxide chemical properties, with no ox-ygen defects and absence of other oxides fingerprints. XPS data agree with those of TiO2S [24], being XPS, a technique with high surface sensitivity and sampled depth of about 4-8 nm for this material, i.e., the topmost TiO2 layer realized by sol-gel. This confirms we achieved high reproducibility and control of the sol-gel approach.”
• In Results section, one new figure (Figure 3, with three spectra) and caption about XPS analysis: “Figure 3. XPS analysis of TiO2H surface. Wide range spectrum (a), Ti2p (b) and O1s core levels are shown, with details about peak components.”
• Two new references about the technique (ref.22) and TiO2 analysis (ref. 24).

Reviewer 3 Report

The author revised all the questions, can be accepted now.