Surface and Electrical Characterization of Non-Stoichiometric Semiconducting Thin-Film Coatings Based on Ti-Co Mixed Oxides Obtained by Gas Impulse Magnetron Sputtering

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors present a comprehensive investigation of gas impulse magnetron sputtering (GIMS) applied to (Ti, Co)Ox thin films, coupled with optical and electrical characterizations. The study demonstrates that meticulous control of the Co content allows for tuning the microstructure and electrical properties of the films, presenting promising applications in semiconductor technology. However, several aspects of the study remain unclear in the current manuscript. To enhance the manuscript for publication consideration, please address the following questions in the revision:

1.       Clarify the advantages of employing the GIMS technique for introducing oxygen to the sputtered thin films. Consider incorporating explanations or references to elucidate this technique.

2.       Discuss the anomaly observed in the morphology of the (Ti, Co)Ox thin film with 44 at.% Co, as it deviates from the general trend. Provide elaboration on this phenomenon in the manuscript.

3.       Simplify the context of the optical profiler measurements data and relocate repetitive results to a supporting file, as the data does not yield additional insights beyond AFM.

4.       Verify the alignment of the IV scan in Figure 10 and the resistivity data in Figure 11, ensuring accurate calculations and plots.

5.       Address the discrepancy in Figure 11, where the resistivity of (Ti, Co)Ox thin films exhibits differing trends compared to literature data. Offer sophisticated explanations for this variation.

6.       In the conclusion section, elaborate on the repeatedly mentioned potential applications of (Ti, Co)Ox thin films. Currently, none of these applications have been detailed in the manuscript. Provide further explanations for clarity.

Author Response

Answers to the report of Reviewers

on the manuscript entitled: Surface and electrical characterization of non-stoichiometric semiconducting thin-film coatings based on Ti and Co mixed oxides obtained by gas impulse magnetron sputtering

Authors: Patrycja Pokora, Damian Wojcieszak, Jarosław Domaradzki, Paulina Kapuścik

Authors :

We would like to express our gratitude for your remarks, which let us improve our manuscript. We have taken them into account in the revised version of my paper. Answering to the Reviewer’s remarks, We have introduced some revisions into the manuscript.

1. Reviewer 1:

The authors present a comprehensive investigation of gas impulse magnetron sputtering (GIMS) applied to (Ti, Co)Ox thin films, coupled with optical and electrical characterizations. The study demonstrates that meticulous control of the Co content allows for tuning the microstructure and electrical properties of the films, presenting promising applications in semiconductor technology. However, several aspects of the study remain unclear in the current manuscript. To enhance the manuscript for publication consideration, please address the following questions in the revision:

• Clarify the advantages of employing the GIMS technique for introducing oxygen to the sputtered thin films. Consider incorporating explanations or references to elucidate this technique.

Authors 1): Replacement of the previously used continuous gas dosing in standard technology with pulsed injection of working gas has led to an increase in the kinetic energy of plasma particles and the degree of plasma imbalance. It has also reduced the tendency for columnar material growth and the degree of defectiveness in the structure of the coating produced by the Gas Injection Magnetron Sputtering (GIMS) method. Pulsed gas dosing induces pulsed discharge in the interelectrode zone of the magnetron and pulse generation of vapor/plasma, from which the coating is synthesized. During the breaks between successive portions of injected gas, the pumping system removes the injected gas and any remaining material/target vapors to a pressure that prevents further discharges. Each new portion of pulsed injected gas initiates the plasma process again. Technological gases such as argon and oxygen can be introduced into the process chamber, with their presence and quantity (flow rate) determined by the requirements of the sputtering process (reactive and non-reactive processes). The flow rate of gases is determined by a mass flow meter (30 sccm for argon and 3 sccm for oxygen in our experiments) [1,2].

One of the paragraphs in Chapter One has been revised according to the following example:

One of the issues overlooked in the literature is the oxygen content in the manufactured nanomaterials, and as we know, the greater its amount, the greater the resistivity. In many processes, such as the sol-gel and chemical vapor deposition (CVD), the amount of oxygen added to materials is maximized, which is associated with the significant challenge of deposition, where one must precisely control the so-called oxygen deficit. Techniques like Gas Impulse Magnetron Sputtering (GIMS) allow for the controlled introduction of oxygen during the process, thereby enabling the fabrication of non-stoichiometric materials [42, 43]. The initiation of pulsed discharge in the interelectrode zone of the magnetron and the generation of vapor/plasma for coating synthesis are consequences of pulsatile gas dosing. In the interludes between consecutive gas injections, the pumping system evacuates the injected gas and any residual material/target vapors, maintaining a pressure that prevents further discharges. The introduction of each new pulse of gas reinitiates the plasma process. The process chamber allows for the introduction of technological gases like argon and oxygen, their presence and quantity (flow rate) being dictated by the demands of the sputtering process, encompassing both reactive and non-reactive procedures. Non-stoichiometric thin films based on metal oxides, including titanium oxide and cobalt oxide, are also relatively less explored and represent an innovative research material. Currently manufactured titanium and cobalt oxides are well oxidized, resulting in high transparency but unfortunately, high resistivity as well [44]. The dielectric nature of such materials poses a problem, limiting the potential applications of these materials, such as their use in gas sensing devices, transparent oxide semiconductors, and more.

• Discuss the anomaly observed in the morphology of the (Ti, Co)Ox thin film with 44 at.% Co, as it deviates from the general trend. Provide elaboration on this phenomenon in the manuscript.

Authors 2): In the case of a (Ti_0.64Co_0.44)Ox film, a morphological anomaly is observed compared to other films. Images obtained through atomic force microscopy clearly show grain sizes smaller than expected based on the growth trend. However, referencing our previous publication [3], which presented scanning electron microscopy (SEM) images of the surface of the examined film, it is evident that the film is highly homogeneous, although some sputtering spots are present on its surface. When measuring this film using scanning electron microscopy, it was challenging to find a region on its surface without visible sputtering spots. Therefore, based on SEM imaging, we posit that the smaller grains observed in the atomic force microscopy (AFM) images of the layer's surface (Ti_0.56Co0_.44)Ox are a result of a measurement error introduced by the AFM microscope during the layer measurement on sputtering spots visible in the SEM images.

In the manuscript, the topic of this anomaly has been elaborated, and a reference to the previous article has been indicated, as presented below:

The surface morphology of the TiO_x, (Ti,Co)Ox, and CoO_x thin films was examined using the AFM method, and the images of the deposited coatings are presented in Fig. 5 and 6. In all cases, the film surfaces were free of cracks, exhibited remarkable homogeneity, and consisted of small grains. The grain size increased with the higher cobalt content from 35 nm to 61 nm, except for the film containing 44 at. % of Co, which exhibited a larger grain size of 102 nm. These AFM measurements corroborate the results obtained by scanning electron microscopy, as previously reported in our publication [42]. It is evident that with an increase in the Co content, the maximum height of the surface profile increased significantly, ranging from 5.05 to 22.94 nm. Regarding the average surface height of the deposited coatings, the results revealed a Gaussian-like symmetrical distribution across all samples, indicating the excellent surface homogeneity (Fig. 7). The root mean square surface roughness (RMS) values gradually decreased (four times) with increasing Co content in the (Ti,Co)Ox films. In the cases of the TiO_x, (Ti_0.97Co_0.03)Ox, and (Ti_0.81Co_0.19)Ox films, the RMS value decreased only slightly from 2.09 to 1.87 nm. However, a further increase in cobalt content led to a significant alteration of the microstructure, resulting in a substantial reduction in the RMS value, falling within the range of 0.71 to 0.55 nm. Furthermore, the peak widths decrease fourfold with higher cobalt content, corroborating a common trend with the root mean square (RMS) roughness distribution (Figure 7).

• Simplify the context of the optical profiler measurements data and relocate repetitive results to a supporting file, as the data does not yield additional insights beyond AFM.

Authors 3): Measurements of surface roughness conducted using an optical profilometer and an atomic force microscope (AFM) yield similar parameters. The optical profilometer provides the S_q parameter, whereas the AFM allows for the calculation of the RMS parameter. The S_q parameter is the root mean square of height differences between points on the surface and the mean height. It is calculated by summing the squares of all height differences, then dividing by the number of measurements and taking the square root of this mean. The RMS parameter is the root mean square of the heights of points on the surface. It is calculated by summing the squares of all heights, dividing by the number of measurements, and taking the square root of this mean. Both parameters describe the dispersion of point heights on the surface but differ in their calculation methods. S_q considers height differences relative to the mean, while RMS takes into account the squares of the heights themselves.

We have decided to alter the presentation of results in the manuscript so as not to duplicate information but rather to compare it. The changes made are evident below:

The deposition rates of TiO_x, (Ti,Co)Ox, and CoO_x films were calculated, taking into account their thickness and sputtering time. The relationship between the Co content in (Ti,Co)Ox films and the sputtering rate during the sputtering process is illustrated in Fig. 4. The quantity of cobalt in the target influences the deposition rates of the films. It can be observed that as the amount of Co increases, the deposition efficiency of the coatings nearly doubles. In the case of a film containing 3 at. % of Co, the deposition rate is nearly on par with that of TiO_x alone. This indicate that targets with Co sputter more efficiently because particles in the magnetic field have a greater tendency to adhere to the substrate. However, the sputtering rate of a film containing only cobalt oxide was lower due to its magnetic properties.

Figure 4. The influence of the Co-content in the (Ti,Co)Ox thin films on the deposition rate and their roughness. Designations: S_q - root mean square surface height value.

The surface topography was also assessed using an optical profiler for the prepared samples, covering an area of approximately 800 μm × 800 μm (Fig. 5). The investigation results revealed that all films exhibited a crack-free, highly homogeneous, and smooth surface. The root mean square surface height (S_q) values of the TiO_x, (Ti_0.97Co_0.03)Ox, (Ti_0.81Co_0.19)Ox, (Ti_0.56Co_0.44)Ox, (Ti_0.40Co_0.60)Ox, and CoO_x thin films were measured at 1.92, 2.02, 1.86, 1.58, 1.39, and 1.19 nm, respectively (Fig. 4). Tests carried out using optical profiler revealed a decrease in the S_q parameter with an increasing amount of cobalt.

Figure 5. Three-dimensional surface profiles of TiO_x, (Ti,Co)Ox, and CoO_x thin films.

• Verify the alignment of the IV scan in Figure 10 and the resistivity data in Figure 11, ensuring accurate calculations and plots.

Authors 4): Both Figure 10 and Figure 11 have been re-examined by the authors, and no discrepancies were observed. Concerning the determination of current-voltage characteristics (Figure 10) for all investigated films on a single composite plot is feasible, but such a compilation does not provide a complete overview of the slope angles of the curves due to different voltage and current ranges applied during the experiments. An illustrative compilation of all curves is presented in Figure 2.

To determine the resistivity from the I-V characteristics (Figure 11), it is necessary to determine the slope of the line (calculate the slope). Using the formula below, the resistivity of each investigated film was individually determined:

where, a - the slope coefficient of the line, g – accounts for the geometry of contacts, d - the thickness of the film, 10^-7 – a factor considering all orders of magnitude.

• Address the discrepancy in Figure 11, where the resistivity of (Ti, Co)Ox thin films exhibits differing trends compared to literature data. Offer sophisticated explanations for this variation.

Authors 5): In the case of (Ti,Co)Ox-based materials, there is no available literature to directly compare the resistivity results of the layers produced by the authors. There are only two publications addressing the resistivity of nanomaterials in a specific manner, namely, M. Z. Musa [4] and S. Bhat [5]. However, both Musa and Bhat fabricated nanoparticles with a different structure than the layers, which is the first reason for the discrepancy between the resistivity values of the layers obtained by the authors and the literature. The second reason for the disparity in results is the use of a different fabrication method for these nanomaterials. Using the Gas Injection Magnetron Sputtering (GIMS) technique, we can selectively introduce oxygen into the material (non-stoichiometric materials). In contrast, the sol-gel technique does not allow us to control the amount of oxygen, resulting in always obtaining stoichiometric materials—this is the third reason for the disparity in results. Additionally, resistivity values in the literature have only been determined for (Ti,Co)Ox materials with Co content not exceeding 10 at%. The authors are unable to compare resistivity values for layers containing a higher amount of cobalt. Nevertheless, Figure 11 was intended to show that the resistivity values obtained by the authors are lower than in previous studies [6], and the layers still maintain a high level of transparency [2].

One section of Chapter 3.2 has been revised as shown below:

Combining titanium oxides with cobalt oxides, i.e., manufacturing them in the form of mixtures, can result in obtaining a material with properties and advantages that are a combination of both. Depending on the manufacturing method, composition, and form, the mixtures of these oxides exhibit various tendencies, and currently, the influence of cobalt on their resistivity cannot be unequivocally determined [20, 19]. It should be noted that while individual reports on the resistivity of films based on mixtures of titanium and cobalt oxides can be found, they typically concern materials with a high oxygen content. There is a lack of information regarding non-stoichiometric oxides. One noteworthy study for discussing the electrical properties of (Ti,Co)Ox is the work of M.Z. Musa and others [19]. It demonstrates that for thin films based on nanoparticles, the resistance decreases with an increase in the amount of cobalt, as indicated by the increasing slope in the I-V characteristic (fig. 10). In the study by Shreesha Bhat [20], it was shown that thin films based on TiO₂ with the addition of up to 2 at.% of Co exhibit an opposite tendency. With an increasing amount of cobalt, the I-V curve has a decreasing slope. Therefore, the nature of resistivity changes is also different, and it increases with the amount of Co in the film (fig.10). These differences likely stem from the different manufacturing methods employed for these nanomaterials but also to their form.

• In the conclusion section, elaborate on the repeatedly mentioned potential applications of (Ti, Co)Ox thin films. Currently, none of these applications have been detailed in the manuscript. Provide further explanations for clarity.

Authors 6): The detailed applications were not presented by the authors due to the review nature of the study. The coatings obtained are the result of numerous attempts to achieve specific properties of the produced layer. The authors aimed to produce coatings based on (Ti,Co)Ox with the highest possible level of transparency [3] and the lowest resistance value. Only such coating parameters provide opportunities for their utilization in the broad field of optoelectronics. As emphasized in the manuscript, non-stoichiometric coatings are poorly understood and explored, and their production is possible only through certain methods. Non-stoichiometric materials have an advantage over stoichiometric materials due to their significant potential for applications. Non-stoichiometric materials allow for the incorporation of a variable number of atoms or particles into their crystal structure, resulting in a wide range of properties for the obtained layers. Additionally, the production of (Ti,Co)Ox layers with varying cobalt content exceeding 10% at. Co, which is poorly documented in the literature, expands the potential scope of their applications. The literature suggests that layers based on (Ti,Co)Ox could be applied, for example, as materials for magnetoresistive random-access memory (MRAM) [7].

Literature:

[1] Zdunek K., Skowroński Ł., Chodun R., Nowakowska-Langier K., Grabowski A., Wachowiak W., Okrasa S., Wachowiak A., Strauss O., Wronkowski A., Domanowski P.; Novel GIMS technique for deposition of colored Ti/TiO_₂ coatings on industrial scale, Materials Science-Poland 2015, 34, 137; doi

[2] Wojcieszak D., Kapuścik P., Kijaszek P., Influence of Annealing on Gas-Sensing Properties of TiO_x Coatings Prepared by Gas Impulse Magnetron Sputtering with Various O₂ Content; Applied Sciences 2023, 13, 1724; doi

[3] Pokora P., Wojcieszak D., Mazur P., Kalisz M., Sikora M., Influence of Co-Content on the Optical and Structural Properties of TiO_x Thin Films Prepared by Gas Impulse Magnetron Sputtering; Coatings 2023, 13(5), 955; doi

[4] Musa M.Z., Ameran Z.F., Mamat M.H.; Malek M.F.; Rasheid B.A.; Noor U.M.; Rusop M. Effects of cobalt doping concentration on the structural, electrical, and optical properties of titanium dioxide thin films. 2011 International Conference on Electronic Devices, Systems and Applications (ICEDSA), Kuala Lumpur, Malaysia, 2011, 339–342. doi

[5] Bhat S., Sandeep K.M., Kumar P., Dharmaprakash S.M., Byrappa K.; Characterization of transparent semiconducting cobalt doped titanium dioxide thin films prepared by sol–gel process. J. Mater. Sci. Mater. Electron. 2018, 29, 1098–1110. doi

[6] Wojcieszak D., Mazur M., Pokora P., Wrona A., Bilewska K., Kijaszek W., Kotwica T., Posadowski W., Domaradzki J.; Properties of Metallic and Oxide Thin Films Based on Ti and Co Prepared by Magnetron Sputtering from Sintered Targets with Different Co-Content,  Materials. 2021, 14, 3797. doi

[7] Quiroz, H.P., Galíndez, E.F., Dussan, A.; Ferromagnetic-like behavior of Co doped TiO₂ flexible thin films fabricated via co-sputtering for spintronic applications. Heliyon 2020, 6, e03338. Quiroz). doi

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In the manuscript entitled "Surface and electrical characterization of non-stoichiometric semiconducting thin-film coatings based on Ti and Co mixed oxides obtained by gas impulse magnetron sputtering", the authors investigate the structural and (thermo-)electrical properties of sputtered TixCoyOz thin films. 

The overall manuscript and study is well conceptualized and the story is well written (see comment ton the English). However, several points which significantly weakens the manuscript have to be addressed before it can recommended for publication. 

P2: Fig 1 shows the resistivity data for TiO₂ prepared by PVD and LPD. The authors should also add some data for CVD/ALD grown thin films. This is especially in view of the fact that the authors also present a data set for CVD in Figure 2 and mention CVD of Titania in the text (e.g., p3 l78).

P2: Scale bar of Fig 1 and Fig 2 should be the same. 

P3 l71: „most frequently used for….“ ref needed 

P3 l85: „titanium and cobalt oxides are well oxidized“ The authors needs to explain what „well“ refers to. Which oxidation state is favorable? The thermodynamically most stable one?

P3 l86: Here the authors mention „high transparency“ for the first time. But what means high? To what do they compare it?

P4 l129-134: SI units instead of words should be used (This is a typical mistake done by AI-based paraphrasing tools).

P4 l138: „encompassed diverse imaging modes, namely tapping modes“ Tapping mode is one imaging mode; contact mode is another? Did the authors used different parameters for their tapping mode - or what do „diverse“ refers to?

P4 l144 „a attributes“ should be removed.

P l145 Which other deposition parameter than the time was used to calculate the deposition rate?

P4 l148 „current“ should have a small letter

P5 l158 SI unit needed

P5 l168-171 Here, the argumentation is not clear for me. Can the authors add some references which support their arguments.

P6 l176 How did the authors determine the grain size from the AFM images? A detailed description should be given, because typically grain sizes are determined by XRD using Debye-Scherrer equation. 

P6 l181 and following: „increase in the Co content, the maximum height of the surface profile is increased“ this statement is wrong and in contradiction to the images shown where the profile becomes more smooth with increasing Co content and also contradicts there own measurement with decreasing RMS with increasing Co content.

Fig 5: What is the difference between the left and the right column? Just the scanning area?

Fig 5 and Fig 6: What is the benefit of adding Fig 6 as well?Which additional information can be extracted, which are not given by Fig 5?

Fig 7 (inset) vs Fig 9: The rms trend shows a sigmoidal behavior whereas the Sq displays a more linear trend? What is the reason for the difference?

P9 l213: „To evaluate the impact of annealing temperature“ Where are the results shown for different annealing procedures? Or is this just a general motivation? 

Fig 10: It should be CoO_x instead of CoO. The latter one determines a specific stoichiometry; however, this hasn‘t be investigated at all in this study.

Fig 11: Hollow symbols are missing in the graph? They are just mentioned in the figure’s legend.

Major issue: The resistivity shown in Fig 12 is different to the resistivity shown in Fig 14. At the example of pure Co, Fig 12 shows a resistivity of around 1 Ohm*cm, whereas Fig 14 states 0.01 Ohm*cm. This is a difference of two orders of magnitude. The authors have to explain this difference.  Note, also for the other composition, a significant, though not so dramatic, deviation is obvious. 

Fig 13: The effect of annealing is mentioned, but it is not possible to predict the results of an annealing process of their samples based on annealing results by Bhat et al. The synthesis and the starting materials are totally different. I suggest, the authors add their annealing results as well or leave this figure and the discussion to it completely out.

Major Issue 2:  The maximum resistivity measured in Fig 11 is not reflected in the Seebeck measurements. Typically less conductive semiconducting materials reveal a higher Seebeck. Do the authors have an explanation for it? Furthermore, the Seebeck coefficient measurements reveal a second maximum with Co content of 44% which is not related to any of the other measured parameters. What is the origin of this, can the authors give at least a hypothesis substantiated by references?

Major Issue 3: P11  l271 (and following) „depends on temperature, as the numbers of charge carriers (and consequently, electrical resistance) increases exponentially with its rise.“ In fact, the number of charge carriers increases, but as a consequence the electrical resistance decreases. The word resistance should be replaced by conductance.

Still the same paragraph: „high activation energy = temperature-dependence = conductors“ vs „low activation energy = less temperature-dependence = semiconductor“ vs „zero activation energy = metal“. What is a conductor? Do they mean insulator? I would say Ea for a intrinsic semiconductor > doped semiconductor > metal. 

P11 l283: If I understood the authors correctly, they here refer to the work by Bhat et al. However, as mentioned above, this is not scientifically correct, since the starting material differs. Again, either they add their own results or keep the discussion out of this manuscript. 

Comments on the Quality of English Language

In my opinion (without having any proof, of course), I think that at least one AI-based paraphrasing and grammar tool was used. On the other hand, I also believe that the original manuscript was written by the authors themselves. Personally, I don't see a problem with the use of such software, but perhaps the authors should point this out in Acknowledgements or Author Contribution sections.   

Author Response

Answers to the report of Reviewers

on the manuscript entitled: Surface and electrical characterization of non-stoichiometric semiconducting thin-film coatings based on Ti and Co mixed oxides obtained by gas impulse magnetron sputtering

Authors: Patrycja Pokora, Damian Wojcieszak, Jarosław Domaradzki, Paulina Kapuścik

Authors :

We would like to express our gratitude for your remarks, which let us improve our manuscript. We have taken them into account in the revised version of my paper. Answering to the Reviewer’s remarks, We have introduced some revisions into the manuscript.

1. Reviewer 2:

In the manuscript entitled "Surface and electrical characterization of non-stoichiometric semiconducting thin-film coatings based on Ti and Co mixed oxides obtained by gas impulse magnetron sputtering", the authors investigate the structural and (thermo-)electrical properties of sputtered TixCoyOz thin films.

The overall manuscript and study is well conceptualized and the story is well written (see comment ton the English). However, several points which significantly weakens the manuscript have to be addressed before it can recommended for publication.

• P2: Fig 1 shows the resistivity data for TiO₂ prepared by PVD and LPD. The authors should also add some data for CVD/ALD grown thin films. This is especially in view of the fact that the authors also present a data set for CVD in Figure 2 and mention CVD of Titania in the text (e.g., p3 l78).

Authors 1): Figure 1 and figure 2 show the literature data for the most commonly used fabrication methods for TiO_x and CoO_x. The authors wanted to emphasize, the variety of methods used to produce the specified materials. For TiO_x-based materials, PVD and LPD methods are most commonly used. For cobalt oxides, SP and CVD techniques are often used in addition to the PVD method. The manuscript (p. 3 |78) highlights the statement that the CVD method as well as LPD are techniques in which it is difficult to control the oxygen deficit during deposition, which involves some difficulties during deposition - this applies to all materials, not just TiO_x or CoO_x.

• P2: Scale bar of Fig 1 and Fig 2 should be the same.

Authors 2): The scale bar has been improved as suggested by the Reviewer.

Figure 2. The resistivity of cobalt-based oxides depends on their manufacturing method and additional annealing [6-11, 16, 17]. Abbreviations: PVD - Physical Vapor Deposition; SP – Spray Pyrolysis; CVD – Chemical Vapor Deposition.

• P3 l71: „most frequently used for….“ ref needed

Authors 3): The reference on gas properties was added by the authors.

Of course, it should be emphasized that the type of conductivity in oxide materials can be modified through, for example, doping [28-40]. In the case of TiO_x, the most commonly used dopant to achieve hole-type conductivity is chromium [31-34]. Elements such as Mn [35], Al [36], and N [37] added to this matrix also allow the production of p-type materials. In contrast, multicomponent materials such as TiO_x:(Nb, Mo, W) exhibit electron conductivity and are most frequently used for the detection of harmful gases to humans [34, 38-40]. In the case of cobalt oxide, after introducing various dopants such as Ca [30], Cu [28], or Zn [29], p-type conductivity was achieved, and no changes in its type were observed. It is worth adding that there is a single publication regarding the use of cobalt as a dopant, but these ZnO:Co materials exhibit p-type conductivity [41].

[34] Zakrzewska K., Radecka M., Rekas M.; Effect of Nb, Cr, Sn additions on gas sensing properties of TiO₂ thin films. Thin Solid Films 1997, 310, 161-166. doi

[38] Yamada Y., Seno Y., Masuoka Y., Nakamra T., Yamashita K.; NO2 sensing characteristics of Nb doped TiO2 thin films and their electronic properties. Sensors and Actuators B 2000, 66, 164-166. doi

[39] Comini E., Sberveglieri G., Guidi V.; Ti–W–O sputtered thin film as n- or p-type gas sensors. Sensors and Actuators B: Chemical 2000, 70, 108-114. doi

[40] Comini E., Guidi V., Ferroni M., Sberveglieri G.; TiO2:Mo, MoO3:Ti, TiO+WO3 and TiO:W layer for landfill produced gases sensing. Sensors and Actuators B: Chemical 2004, 100, 41-46. doi

• P3 l85: „titanium and cobalt oxides are well oxidized“ The authors needs to explain what „well“ refers to. Which oxidation state is favorable? The thermodynamically most stable one?

Authors 4): The authors regarding the statement "well oxidized" meant titanium and cobalt oxides with their maximum degree of oxidation.

• P3 l86: Here the authors mention „high transparency“ for the first time. But what means high? To what do they compare it?

Authors 5): When stating "high transparency," the authors meant a light transmission rate that is above 75% [1]. This transmission value was compared to the layer that the authors presented in their previous publication [2].

[1] Wojcieszak D., Mazur M., Pokora P., Wrona A., Bilewska K., Kijaszek W., Kotwica T., Posadowski W., Domaradzki J.; Prop-erties of metallic and oxide thin films based on Ti and Co prepared by magnetron sputtering from sintered targets with different Co-content. Materials 2021, 14, 3797. doi

[2] Pokora P., Wojcieszak D., Mazur P., Kalisz M., Sikora M.; Influence of Co-content on the optical and structural properties of TiOx thin films prepared by gas impulse magnetron sputtering, Coatings 2023, 13, 955. doi

• P4 l129-134: SI units instead of words should be used (This is a typical mistake done by AI-based paraphrasing tools).

Authors 6): Si units were introduced in the manuscript according to the reviewer's comment.

• P4 l138: „encompassed diverse imaging modes, namely tapping modes“ Tapping mode is one imaging mode; contact mode is another? Did the authors used different parameters for their tapping mode - or what do „diverse“ refers to?

Authors 7): There is an inaccuracy in the sentence highlighted by the Reviewer, which has been corrected by the authors to the following statement: “The investigation employed contact mode imaging, utilizing force-modulated probes (with a spring constant of k = 0.2 N/m, WITec, Ulm, Germany). All experiments were conducted under uniform measurement conditions”,

• P4 l144 „a attributes“ should be removed.

Authors 8): There is an inaccuracy in the sentence highlighted by the Reviewer, which has been corrected by the authors to the following statement: “This equipment also facilitates the assessment of surface geometric, including surface flatness and thickness”.

• P4 l145 Which other deposition parameter than the time was used to calculate the deposition rate?

Authors 9): There is an inaccuracy in the sentence highlighted by the Reviewer, which has been corrected by the authors to the following statement: “The deposition rate was calculated based on the deposition time and the thickness of the resultant films”.

• P4 l148 „current“ should have a small letter

Authors 10): According to the reviewer's comment, the error has been corrected: “The current-voltage characteristics were obtained within a Faraday shield”.

• P5 l158 SI unit needed

Authors 11): According to the reviewer's comment, the error has been corrected: “To determine the Seebeck coefficient, a controlled temperature gradient (ΔT) was es-tablished between the 'hot' and 'cold' electrical contacts, ranging from 0 to 50 K, with the 'cold' contact being maintained at room temperature”.

• P5 l168-171 Here, the argumentation is not clear for me. Can the authors add some references which support their arguments.

Authors 12): In our prior publication [1], we thoroughly examined the fabrication processes employed for the films and elucidated the specific roles played by the Ti and Co targets in the procedure. It was observed that augmenting the quantity of cobalt in the target led to an elevation in its sputtering rate. This phenomenon is attributed to the interaction of the plasma with the Co target, given its magnetic nature. Consequently, particles propelled within the intensified magnetic field generated by cobalt are more readily and effectively sputtered onto the substrate.

[1] Wojcieszak D., Mazur M., Pokora P., Wrona A., Bilewska K., Kijaszek W., Kotwica T., Posadowski W., Domaradzki J.; Properties of Metallic and Oxide Thin Films Based on Ti and Co Prepared by Magnetron Sputtering from Sintered Targets with Different Co-Content,  Materials. 2021, 14, 3797. doi

• P6 l176 How did the authors determine the grain size from the AFM images? A detailed description should be given, because typically grain sizes are determined by XRD using Debye-Scherrer equation

Authors 13): The grain size for each film was determined by AFM measurements. The same method of determining grain size was used for all films. Namely, the size of the grains observed in the AFM images was measured using the WSxM software using the envelope method. For each film, 20 measurements of different grains were taken and averaged to a single value. An example of one grain measurement is provided in the figure below:

• P6 l181 and following: „increase in the Co content, the maximum height of the surface profile is increased“ this statement is wrong and in contradiction to the images shown where the profile becomes more smooth with increasing Co content and also contradicts there own measurement with decreasing RMS with increasing Co content.

Authors 14): The authors made an error when describing the AFM profiles. According to the Reviewer's comment, the height of the surface profile decreases (rather than increases) with greater amounts of cobalt. In the manuscript, the statement changed to: „It is evident that with an increase in the Co content, the maximum height of the sur-face profile decreased significantly, ranging from 22.94 to 5.05 nm”.

• Fig 5: What is the difference between the left and the right column? Just the scanning area?

Authors 15): Figure 15 presents Atomic Force Microscopy images. The left images depict a larger area, specifically 1x1 µm, whereas the right images are captured at a lower resolution of 600x600 nm. Adjacent to each AFM image, the X, Y, Z parameters corresponding to the measurement area are provided.

• Fig 5 and Fig 6: What is the benefit of adding Fig 6 as well? Which additional information can be extracted, which are not given by Fig 5?

Authors 16): Figure 5 depicts Atomic Force Microscopy images captured in three dimensions, while fig. 6 presents the same AFM images in two dimensions. The inclusion of both representations in the manuscript was chosen by the authors due to the 3D images providing information about the Z-coordinate, enabling visualization of the layer height. Conversely, the 2D images offer a clearer depiction of grain size and alignment.

• Fig 7 (inset) vs Fig 9: The rms trend shows a sigmoidal behaviour whereas the Sq displays a more linear trend? What is the reason for the difference?

Authors 17): Measurements taken with an optical profilometer are less accurate than those obtained using atomic force microscopy. The parameter S_q was employed by the authors as a complement to the investigations of surface roughness. In the case of the S_q parameter, a trend approximating linearity is indeed observed, while for RMS, sigmoidal behaviour becomes apparent. However, these values are very close to each other (changes averaging around 0.41 nm) and do not significantly affect the properties of the film. Additionally, the authors note the same trend of decreasing parameters describing surface roughness with respect to the amount of cobalt in the films in both cases. Moreover, it is worth mentioning that S_q and RMS parameters are determined slightly differently, so discrepancies may arise from different calculation methods. Sq represents the average of the squares of height differences between points on the surface and the mean height. RMS, on the other hand, is the square root of the average of the squares of height differences between points on the surface.

• P9 l213: „To evaluate the impact of annealing temperature“ Where are the results shown for different annealing procedures? Or is this just a general motivation?

Authors 18): There is an inaccuracy in the sentence highlighted by the Reviewer, which has been corrected by the authors to the following statement: “To evaluate the impact of Co-content on the electrical properties of the TiOx, (Ti,Co)Ox, and CoOx thin films, current-voltage (I-V) characteristics and Seebeck coefficient measurements were conducted”.

• Fig 10: It should be CoOx instead of CoO. The latter one determines a specific stoichiometry; however, this hasn‘t be investigated at all in this study.

Authors 19): Figure 10 was corrected by the authors to the following statement:

Figure 9. Current-voltage (I-V) characteristics (10 measurement cycles) of the prepared (Ti,Co)Ox films with varying cobalt concentrations.

• Fig 11: Hollow symbols are missing in the graph? They are just mentioned in the figure’s legend.

Authors 20):  Figure 11 was corrected by the authors to the following statement:

Figure 10. The effect of Co content on the resistivity (ρ) of our non-stoichiometric (Ti,Co)Ox thin films and exemplary films with immobilised nanoparticles in the TiO_x matrix and (Ti,Co)Ox nanoparticles, along with their current-voltage characteristics [19, 20].

• Major issue: The resistivity shown in Fig 12 is different to the resistivity shown in Fig 14. At the example of pure Co, Fig 12 shows a resistivity of around 1 Ohm*cm, whereas Fig 14 states 0.01 Ohm*cm. This is a difference of two orders of magnitude. The authors have to explain this difference. Note, also for the other composition, a significant, though not so dramatic, deviation is obvious.

Authors 21): The Authors have, in fact, committed a substantial error in the presentation of the results. It is presumed that the Reviewer intended to juxtapose the resistivity values found in Figure 11 (rather than Figure 12) with those in Figure 14. Upon scrutinizing the presented results, it has come to light that, in the case of current-voltage measurements conducted at room temperature, the Authors aggregated the results of measurements for the respective coatings immediately after their application. Conversely, the compilation of the results pertaining to the I-V characteristics measured under different temperatures was excessive for coatings that were manufactured one month prior. Therefore, the difference seen in the resistivity values is obvious from the fact that the films had oxygenated after a month, and thus had different electrical properties. The authors opted to incorporate the outcomes of both tests under identical measurement conditions. Consequently, both the I-V characteristics at room temperature and those at various temperatures are depicted for measurements conducted one month after deposition. The revised graphs are presented below (see question 19 and 20):

• Fig 13: The effect of annealing is mentioned, but it is not possible to predict the results of an annealing process of their samples based on annealing results by Bhat et al. The synthesis and the starting materials are totally different. I suggest, the authors add their annealing results as well or leave this figure and the discussion to it completely out.

Authors 22): Figure 13 was designed by the authors to illustrate the activation energy values of TiO₂, Co₃O₄, and (Ti,Co)Ox-based materials as reported in the literature. Due to the limited number of publications on this parameter, the authors opted to juxtapose their results with existing data. It is acknowledged that materials subjected to annealing undergo changes in their structural properties; hence, the decision was made to showcase the impact of annealing on this parameter. The primary focus of the authors' investigation was to examine the influence of the cobalt content in the (Ti,Co)Ox film rather than the annealing process itself. Additionally, a literature review did not yield information on (Ti,Co)Ox films that had not undergone additional thermal treatment, emphasizing a notable gap in information regarding these materials.

• Major Issue 2: The maximum resistivity measured in Fig 11 is not reflected in the Seebeck measurements. Typically less conductive semiconducting materials reveal a higher Seebeck. Do the authors have an explanation for it? Furthermore, the Seebeck coefficient measurements reveal a second maximum with Co content of 44% which is not related to any of the other measured parameters. What is the origin of this, can the authors give at least a hypothesis substantiated by references?

Authors 23): The Reviewer is absolutely correct in highlighting the correlation between resistivity values and the Seebeck coefficient. However, the materials outlined in the manuscript exhibit a notably feeble conductivity, even in the case of TiO_x. Consequently, establishing a connection with the resistivity values becomes challenging, especially considering the significance that a higher S-value would have brought – one that is at least two orders of magnitude greater. The primary intention was to underscore the weak conductivity inherent in the prepared films. In contrast, the Authors have previously acquired proficiency in dealing with materials characterized by a robust conductivity type [1], and indeed, a discernible relationship emerges in such cases. Specifically, less conductive films, signified by higher resistivity values, manifest a heightened Seebeck coefficient. Nevertheless, the Reviewer's insightful comments have prompted us to delve into a comprehensive study of these films post the annealing process. It is anticipated that elevated temperatures during annealing will likely induce a shift in conductivity type towards a more robust n-type or a diminished p-type. Moreover, the increased temperature is expected to initiate oxygenation within the layer, consequently altering its structural composition.

[1] Sieradzka K., Mazur M., Wojcieszak D., Domaradzki J., Kaczmarek D., Prociow E., P-type transparent Ti–V oxides semiconductor thin film as a prospective material for transparent electronics. Thin Solid Films 2012, 520, 3472-3476. doi

• Major Issue 3: P11 l271 (and following) „depends on temperature, as the numbers of charge carriers (and consequently, electrical resistance) increases exponentially with its rise.“ In fact, the number of charge carriers increases, but as a consequence the electrical resistance decreases. The word resistance should be replaced by conductance. Still the same paragraph: „high activation energy = temperature-dependence = conductors“ vs „low activation energy = less temperature-dependence = semiconductor“ vs „zero activation energy = metal“. What is a conductor? Do they mean insulator? I would say Ea for a intrinsic semiconductor > doped semiconductor > metal.

Authors 24): As suggested by the Reviewer, the statements indicated in the comment have been corrected: “The conductivity of materials strongly depends on temperature, as the number of charge carriers (and, consequently, electrical conductance) increases exponentially with its rise” and “For high activation energy values, the reaction rate is significantly temperature-dependent (intrinsic semiconductor). The smaller the activation energy (E_a), the less dependent the rate is on temperature (doped semiconductor). In the case of E_a=0 eV, the reaction rate is independent of temperature (metals)”.

• P11 l283: If I understood the authors correctly, they here refer to the work by Bhat et al. However, as mentioned above, this is not scientifically correct, since the starting material differs. Again, either they add their own results or keep the discussion out of this manuscript.

Authors 25): In the case of the E_a parameter, the authors did not aim to compare their study of this parameter to the existing work of Bhat et al. because it is a different material (with a different structure), manufacturing method and post-processing treatment. The Authors wanted to outline the lack of such studies on (Ti,Co)Ox materials. However, the Authors believe that it is important to refer to the existing results of other scientific researchers on this material in order to find potential relationships.

• In my opinion (without having any proof, of course), I think that at least one AI-based paraphrasing and grammar tool was used. On the other hand, I also believe that the original manuscript was written by the authors themselves. Personally, I don't see a problem with the use of such software, but perhaps the authors should point this out in Acknowledgements or Author Contribution sections.

Authors 26): During the writing of the manuscript, the Authors used several editing tools for improving the grammar and style of the text. The Authors emphasize that editorial tools were used to a small extent, especially in paragraphs where linguistic correctness needed improvement. However, the changes generated by the editing tools were meticulously analyzed by the Authors and applied or not. However, the Reviewer's suggestion is most appropriate and we will certainly take it into account.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Please see attached file.

Comments for author File: Comments.pdf

Author Response

Answers to the report of Reviewers

on the manuscript entitled: Surface and electrical characterization of non-stoichiometric semiconducting thin-film coatings based on Ti and Co mixed oxides obtained by gas impulse magnetron sputtering

Authors: Patrycja Pokora, Damian Wojcieszak, Jarosław Domaradzki, Paulina Kapuścik

Authors :

We would like to express our gratitude for your remarks, which let us improve our manuscript. We have taken them into account in the revised version of my paper. Answering to the Reviewer’s remarks, We have introduced some revisions into the manuscript.

1. Reviewer 2:

This manuscript from Patrycja et al. shows their detailed investigation of the structural, and electrical properties of non-stoichiometric (Ti,Co)Ox thin films prepared using the Gas Impulse Magnetron Sputtering technique. Authors characterize and reports the films surface topography and microstructure using Atomic Force Microscopy (AFM) and optical profiler techniques. The Ti-Co and Ti-Co-O systems are interesting material systems with promising applications in many fields such as chemical and optical industries. Therefore, these materials are well studies material, and it is known that properties may vary depending on the synthesis method.

The paper is well written, and the presentation is scholarly. I like the paper as the paper not only presents their own experimental results, but authors also extensively compared their results with existing literatures. I think the paper does not report any novel findings, however, their systematic and detailed characterisation along with fair comparison with existing literatures will provides a well-informed scientific report for the readers and/or application designers.

In summary, I suggest publishing this paper in the journal “Coatings” after minor revision below:

• In Figure 11, unfortunately, I do not see any data points for “nanoparticles with p-type conductivity [S. Bhat 2018]”. Not sure if the data points are hidden under other data points, legend or somehow missed, please check and fix.

Authors 1): In Figure 11, an error indeed occurred, which the Authors did not notice. The figure has been corrected as follows:

Figure 10. The effect of Co content on the resistivity (ρ) of our non-stoichiometric (Ti,Co)Ox thin films and exemplary films with immobilised nanoparticles in the TiO_x matrix and (Ti,Co)Ox nanoparticles, along with their current-voltage characteristics [19, 20].

• In Figure 14, I am wondering why the resistivity does not follow any trend with increasing Co doping at a given temperature, can you elaborate more on it.

Authors 2): In Figure 14, the dependence of resistivity on temperature is depicted within the range of 303 K to 353 K. The resistivity of the thin film with 3 at.% cobalt achieves its highest value with an increase in temperature, whereas films with a higher cobalt content exhibit very similar resistivity values with respect to temperature. The values of this parameter do not reveal any trend with respect to the amount of cobalt in the film but correspond to the values obtained from the current-voltage characteristics shown in Figure 10. The nearly negligible changes in resistivity with temperature suggest a very low activation energy for electrical conduction processes, which may be associated with the presence of various cobalt oxide ratios (CoO:Co₃O₄) in these thin layers [1].

• I guess the Title is too long. I will suggest reconsider and shorten it if possible.

Authors 3): In response to the Reviewer's suggestion, we have decided to abbreviate the name to the following: Surface and Electrical Characterization of Non-Stoichiometric Semiconducting Thin-film Coatings based on Ti-Co Mixed Oxides obtained by GIMS

Literature:

[1] Pokora P., Wojcieszak D., Mazur P., Kalisz M., Sikora M., Influence of Co-Content on the Optical and Structural Properties of TiO_x Thin Films Prepared by Gas Impulse Magnetron Sputtering; Coatings 2023, 13(5), 955; doi

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thanks for the authors' effort in editing the manuscript. The format and font size of the figures should be improved for better visualization before publication. 

Author Response

Answers to the report of Reviewers

on the manuscript entitled: Surface and Electrical Characterization of Non-Stoichiometric Semiconducting Thin-Film Coatings based on Ti-Co Mixed Oxides obtained by GIMS

Authors: Patrycja Pokora, Damian Wojcieszak, Jarosław Domaradzki, Paulina Kapuścik

Authors :

We wish to extend our sincere appreciation for your valuable feedback, which has greatly contributed to enhancing the quality of our manuscript. Your thoughtful remarks have been carefully considered and incorporated into the revised version of our paper. In response to the reviewer's comments, we have implemented several revisions to further refine and strengthen the content. Thank you for your insightful input, which has undoubtedly enriched the overall quality of our work.

1. Reviewer 1:

Thanks for the authors' effort in editing the manuscript. The format and font size of the figures should be improved for better visualization before publication.

Authors 1): In response to your feedback, we have diligently worked on enhancing the format and font size of the figures to ensure improved visualization before publication. We believe these adjustments will contribute to a more visually effective presentation of our research.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The revised version of the manuscript with a slightly altered title “Surface and Electrical Characterization of Non-Stoichiometric Semiconducting Thin-Film Coatings based on Ti-Co Mixed Oxides obtained by GIMS” has significantly improved compared to the first version.

In general, all comments have been addressed by the authors. After threee few minor additional revisions, I can recommend the manuscript for publication in MDPI Coatings.

Fig 1: I sill suggest adding titanium oxide in the resistivity table/figure. There is a bunch of publications on CVD/ALD-grown TiO2. The arguments given by the authors in their reply are somehow weak.

Grain-size analysis: It should be explicitly mentioned that the grain size has be determined by AFM image analysis.

P3 l83: For the reader of the manuscript it is still unclear what well oxidized mean. The should precisely define what “well” means, e.g. the thermodynamically stable phase or else. Also “high transparency” is not defined. The authors can give a number as they did in the answer to my initial comment.

Author Response

Answers to the report of Reviewers

on the manuscript entitled: Surface and Electrical Characterization of Non-Stoichiometric Semiconducting Thin-Film Coatings based on Ti-Co Mixed Oxides obtained by GIMS

Authors: Patrycja Pokora, Damian Wojcieszak, Jarosław Domaradzki, Paulina Kapuścik

Authors :

We wish to extend our sincere appreciation for your valuable feedback, which has greatly contributed to enhancing the quality of our manuscript. Your thoughtful remarks have been carefully considered and incorporated into the revised version of our paper. In response to the reviewer's comments, we have implemented several revisions to further refine and strengthen the content. Thank you for your insightful input, which has undoubtedly enriched the overall quality of our work.

1. Reviewer 2:

The revised version of the manuscript with a slightly altered title “Surface and Electrical Characterization of Non-Stoichiometric Semiconducting Thin-Film Coatings based on Ti-Co Mixed Oxides obtained by GIMS” has significantly improved compared to the first version. In general, all comments have been addressed by the authors.

After threee few minor additional revisions, I can recommend the manuscript for publication in MDPI Coatings.

• Fig 1: I sill suggest adding titanium oxide in the resistivity table/figure. There is a bunch of publications on CVD/ALD-grown TiO2. The arguments given by the authors in their reply are somehow weak.

Authors 1): Thank you for your suggestion. We have added the charts, and specific data points for publications on TiO₂ synthesized using the CVD method have been included in the resistivity figure 1.

Figure 1. The resistivity of titanium-based oxides depends on their manufacturing method and form [1-3, 12-18]. Abbreviations: PVD - Physical Vapor Deposition; LPD - Liquid Phase Deposition.

• Grain-size analysis: It should be explicitly mentioned that the grain size has be determined by AFM image analysis.

Authors 2): We acknowledge your valuable suggestion. In the revised version of the manuscript, we have explicitly stated that the grain size was determined through AFM image analysis.

An Atomic Force Microscope (AFM), specifically Nanosurf FlexAFM model from Liestal, Switzerland, was employed to assess the topographical alterations in the prepared thin films. The investigation employed contact mode imaging, utilizing force-modulated probes (with a spring constant of k = 0.2 N/m, WITec, Ulm, Germany). All experiments were conducted under uniform measurement conditions. The WSxM 5.0 Develop 10.2 software package [58] was utilized for the post-processing and analysis of the obtained data. For all prepared films the average size of grains was determined with the use of WSxM software and envelope method. These value were estimated based on 20 randomly selected grains at AFM image.

• P3 l83: For the reader of the manuscript it is still unclear what well oxidized mean. The should precisely define what “well” means, e.g. the thermodynamically stable phase or else. Also “high transparency” is not defined. The authors can give a number as they did in the answer to my initial comment.

Authors 3): Thank you for your insightful feedback on our manuscript. In the revised version of the manuscript, we have explicitly clarified that "well-oxidized" refers to achieving a specific thermodynamically stable phase. We believe that this modification will enhance the reader's understanding of the concept. In our revised manuscript, we have incorporated specific numerical values, as suggested, to define and measure the level of transparency.

Techniques like Gas Impulse Magnetron Sputtering (GIMS) allow for the controlled introduction of oxygen during the process, thereby enabling the fabrication of non-stoichiometric materials [46, 47]. The initiation of pulsed discharge in the interelectrode zone of the magnetron and the generation of vapor/plasma for coating synthesis are consequences of pulsatile gas dosing. In the interludes between consecutive gas injections, the pumping system evacuates the injected gas and any residual material/target vapors, maintaining a pressure that prevents further discharges. The introduction of each new pulse of gas reinitiates the plasma process. The process chamber allows for the introduction of technological gases like argon and oxygen, their presence and quantity (flow rate) being dictated by the demands of the sputtering process, encompassing both reactive and non-reactive procedures. Non-stoichiometric thin films based on metal oxides, including titanium oxide and cobalt oxide, are also relatively less explored and represent an innovative research material. Currently manufactured titanium and cobalt oxides are well oxidised (at Ti⁴⁺ and Co³⁺ state) what results in receiving TiO₂ or Co₃O₄ forms. The transmission coefficient (T_λ) of TiO₂ in a form of anatase or rutile is generally around 80%. Unfortunately, such high transparency occurs with high resistivity [48].

Author Response File: Author Response.pdf