Energy Storage Performance of Sandwich Structured Pb(Zr0.4Ti0.6)O3/BaZr0.2Ti0.8O3/Pb(Zr0.4Ti0.6)O3 Films
and Ping Qu 4Round 1
Reviewer 1 Report
This manuscript reports a study of relaxor ferroelectric sol-gel films that are promising for high-k values and large energy storage densities. Although this is manuscript is potentially interesting and publishable, I cannot recommend it for publication anywhere before several important issues are address in a new revision:
1) The manuscript must be proofread. For example, the authors used "tenability" multiple times while, I guess, they meant "tunability".
2) The manuscript must be fact checked for consistency. For example, in the abstract, the authors state that energy storage density of Ba(Zr0.2Ti0.8)O3 is 21.28 J*cm^-3. but in the Introduction on line 41 they cite the energy storage density of 30.4 for the same material! What is the difference? Different studies? Different preparation (sol-gel vs. epitaxy?)?
3) The preparation must be explained better. The authors claim in section 2.Experiment that they deposited a PZT/BZT/PZT solution (line 73), which totally confused me. Did they mix precursors together before depositions? Then, there will be no any sandwiched structure. If they made multiple depositions of alternating layers of PZT and BZT, the authors should say so and describe how many layers they deposited, and what was their starting layer. Also, I can only guess the authors had some electrodes on top of their films for electrical measurements. The electrode preparation methods and the electrode size has to be described in the paper.
4) A brief discussion of disadvantages of using Pb in the Introduction would be fine if the authors' main finding were that some lead-free films have superior properties. However, the main finding seems to be "more lead - better capacitors", which is in dissonance with the introduction/motivation.
In conclusion, I believe the authors have a good publishable research, but they need to read the manuscript several more times and clean it of linguistic, factual, and logical errors.
Author Response
Response letter to the Comments
Dear Editor,
Firstly, we would like to thank you for your kind letter and for reviewers’ constructive comments. These comments are all valuable and helpful for improving our paper. All the authors have discussed about all these comments. According to the reviewers’ comments, we have tried best to modify our manuscript to meet with the requirements. In this revised version, changes to our manuscript were all highlighted. Point-by-point responses to the reviewers are listed below this letter.
Reviewers' comments:
Reviewer 1:
The manuscript must be proofread. For example, the authors used "tenability" multiple times while, I guess, they meant "tunability".
We are very sorry for our incorrect writing. We have made correction according to the Reviewer’s suggestion and "tenability" has been replaced by "tunability" in the 5 parts of the paper.
The manuscript must be fact checked for consistency. For example, in the abstract, the authors state that energy storage density of Ba(Zr2Ti0.8)O3 is 21.28 J*cm^-3. but in the Introduction on line 41 they cite the energy storage density of 30.4 for the same material! What is the difference? Different studies? Different preparation (sol-gel vs. epitaxy?)?
Thank you for your valuable and thoughtful comments. The energy storage densities of Ba(Zr0.2Ti0.8)O3 are different because they are different studies. For the preparation of BaZr0.2Ti0.8O3 thin films, we used different preparation methods compared with the cited literature. Our experiments adopted sol-gel method, while Sun et. al prepared BaZr0.2Ti0.8O3 epitaxial thin films by using a radio-frequency magnetron sputtering system. Besides, the obtained energy storage densities correspond to different electric field intensities.
3) The preparation must be explained better. The authors claim in section 2.Experiment that they deposited a PZT/BZT/PZT solution (line 73), which totally confused me. Did they mix precursors together before depositions? Then, there will be no any sandwiched structure. If they made multiple depositions of alternating layers of PZT and BZT, the authors should say so and describe how many layers they deposited, and what was their starting layer. Also, I can only guess the authors had some electrodes on top of their films for electrical measurements. The electrode preparation methods and the electrode size has to be described in the paper.
Thank you for your instructive suggestions. Firstly, the preparation of sandwich structured films does not require the mixing of two precursor solutions. Secondly, the specific preparation process of the PZT/BZT/PZT films is as follows. The first layer of PZT film was spin-coated on the Pt(111)/Ti/SiO2/Si substrates by PZT solution, and the first layer was heated on the hot plate at 450 oC for 3 min before spinning the second layer. Next, BZT solution was used to prepare the second to seventh layers by spin-coating, and each layer was respectively heated on the hot plate at 450 oC for 3 min and then prepared the next layer. And the PZT solution was used again to prepare the top layer and the top PZT film was heated on the hot plate at 450 oC for 3 min. Finally, the prepared sandwich structured films annealed at 850 oC for 10 min. Combine the above discussion, we made the following changes: After aging 24 h, the BZT solutions and PZT solutions were deposited on the Pt(111)/Ti/SiO2/Si substrates by spin-coated process according PZT/BZT/PZT from the substrates. Finally, as for the preparation of electrode, we supplement it in this paper, and the supplementary content is as follows. Besides, for the electrical performance testing of the thin films, a top electrode with a diameter of 2 μm was prepared on the surface by using Pt to form the capacitor structure of the top electrode/ferroelectric thin film/bottom electrode.
A brief discussion of disadvantages of using Pb in the Introduction would be fine if the authors' main finding were that some lead-free films have superior properties. However, the main finding seems to be "more lead - better capacitors", which is in dissonance with the introduction/motivation.
Thank you for your careful reading of our manuscript. According to the opinions of the Reviewer, we delete this part after careful thinking. We think that the innovation of this paper is to make use of the advantages of the PZT thin film and combine it with BZT thin film to prepare sandwich structured films to achieve the purpose of optimizing the performance of the BZT thin film. Therefore, the introduction part is indeed improper with the discussion of PZT material.
Thank you again for your positive comments on our manuscript. If there are any other modifications we could make, we would like very much to modify them and we really appreciate your help.
Sincerely yours.
Qiu Sun
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The autors present in the introduction the toxicity of lead and the demand for alternativelead-free materials. However their material contains Pb. Please change the argumentation of the introduction.
The experiment part is well presented.
The background in XRD patterns can be removed. The XRD and SEM conclusions are well presented but not the dielectric characterizations.
It is said that the "Tc was about 25°C". First, I think that is not "was" but "is"... and secondly, the TC value in the figure 3a not seems to be 25°C. Please, evaluate correctly the TC value by plotting the inverse of the dielectric constant. For a comparison with the figure 6b, it will be better if the temperature reach 200°C.
The figure 3b is not correctly presented. For dielectric characteristics as a function of frequency, the frequency is generally in a log plot. I don't understand why the losses increase according to the frequence whereas the dielectric constant does not have the same evolution. It is not conform to Kramers Kronig relation.
The losses in the figure 3c seem strange. The material is in the ferroelectric state, the dielectric constant and the losses must describe a butterfly loop. However, for the losses, there is two waves. Can you explain it. For this figure, it will be better to plot the electric field and not the voltage for a comparison with the figure 5. The correct term is "tunability" and not "tenability". I don't understand why the values of the dielectric constant in figure 3c do not correspond to the values in the figure 3b...
Please, put the same scale in figure 4a and 4b.Why do you choose 30°C and 50°C ?
It will be interesting to have the PZT loop in figure 5a. I don't understand the stability of the polarization as a function of the temperature. The polarization and the dielectric constant are related and the dielectric constant is not stable (figure 3a) .
Why double slim loops are typical relaxer behavior and not soft ferroelectric behavior ?
I think that the relation (2) is not correct. Please verify in the references [24-26].
In the table 1, I think is not Jloss but the energy efficiency.
The figures 6 are not correct. We can’t have negative energy here. It will be better if Jrec and Jloss have the same scal to really compare the two energies and see directly the difference between the two curves which correpond to Jchg.
Author Response
Dear Editor,
Firstly, we would like to thank you for your kind letter and for reviewers’ constructive comments. These comments are all valuable and helpful for improving our paper. All the authors have discussed about all these comments. According to the reviewers’ comments, we have tried best to modify our manuscript to meet with the requirements. In this revised version, changes to our manuscript were all highlighted. Point-by-point responses to the reviewers are listed below this letter.
Reviewers' comments:
Reviewer 2:
The autors present in the introduction the toxicity of lead and the demand for alternativelead-free materials. However their material contains Pb. Please change the argumentation of the introduction.
Thank you for your careful reading of our manuscript and instructive suggestions. According to the opinions of the Reviewer, we delete this part after careful thinking. The innovation of this paper is to make use of the advantages of the PZT thin film and combine it with BZT thin film to prepare sandwich structured films to achieve the purpose of optimizing the performance of the BZT thin film. Therefore, the introduction part is indeed improper with the discussion of PZT material.
The experiment part is well presented.
Thanks a lot for your kind comments. We also appreciate your great efforts on our manuscript and the valuable suggestions and questions. To make it clear that the BZT solution and the PZT solution are not mixed, we made the following changes and supplemented the related content of the top electrode as follows: a) After aging 24 h, the BZT solutions and PZT solutions were deposited on the Pt(111)/Ti/SiO2/Si substrates by spin-coated process according PZT/BZT/PZT from the substrates. b) Besides, for the electrical performance testing of the thin films, a top electrode with a diameter of 2 μm was prepared on the surface by using Pt to form the capacitor structure of the top electrode/ferroelectric thin film/bottom electrode.
The background in XRD patterns can be removed. The XRD and SEM conclusions are well presented but not the dielectric characterizations.
According to the Reviewer’s advice, we have removed the background. In our study, XRD was mainly used to characterize the crystal structure of the sandwich films, besides SEM was used to confirm the micro-topography and the thickness of the films. The dielectric characterizations for the samples were measured by the impedance analyzer and the dielectric properties are mainly analyzed by dielectric temperature spectrum and dielectric frequency spectrum.
It is said that the "Tc was about 25oC". First, I think that is not "was" but "is"... and secondly, the TC value in the figure 3a not seems to be 25o Please, evaluate correctly the TC value by plotting the inverse of the dielectric constant. For a comparison with the figure 6b, it will be better if the temperature reach 200oC.
First of all, it is really true as Reviewer suggested that is not "was" but "is" and we have corrected it. Besides, we're sorry for our negligence about the TC value and we made the following corrections in the article: Tc is about 25 oC in BZT films, and the Curie temperature of the PZT/BZT/PZT film is about 40 oC, there is a slight rise in the Curie temperature. At last, for the use of materials, we aim for room temperature application so as to energy saving and easy operation. Besides, as the temperature increases, the energy loss increases and energy efficiency decreases. While the changes of energy loss and efficiency are not significant, it can be deduced to have a relatively good thermal stability in a certain temperature range between -140 oC to 200 oC.
The figure 3b is not correctly presented. For dielectric characteristics as a function of frequency, the frequency is generally in a log plot. I don't understand why the losses increase according to the frequence whereas the dielectric constant does not have the same evolution. It is not conform to Kramers Kronig relation.
At first, as Reviewer suggested that the frequency was generally in a log plot, we have changed it and the x coordinate has been represented in logarithmic form.
Secondly, in consideration of the questions of Reviewer, we consulted the relevant literature and some conclusions are drawn. Polarization of a dielectric material is the sum of the contributions of dipolar, electronic, ionic, and interfacial polarizations [1-2]. At low frequencies, all the polarizations respond easily to the electric field. But as the frequency of the electric field increases, different polarization contributions filter out. As a result, the net polarization of the material decreases, which leads to the decrease of εr. Further, the decrease of tanδ with the increase in frequency can be explained by Debye formula [3]. According to this formula, the tanδ at lower frequencies is inversely proportional to frequency, which explains the decrease of tanδ with frequency.
Thirdly, for our experimental results, the dielectric loss increased with the frequency increasing at the frequency up to 100 kHz. We speculate that the polarization value of the thin film cannot keep up with the change of electric field and the relaxation loss increases with the higher frequency. Besides, similar reports have been made in other literature [4]. We have added the above discussion on this issue in our revised manuscript.
Singh A.K., Goel T.C., Mendiratta R.G., et. al. Dielectric properties of Mn-substituted Ni-Zn ferrites. Journal of Applied Physics. 2002,91,6626-6629. DOI:1063/1.1470256. Liu B., Wang X., Zhang Y., et al. Synthesis and characteristics of KTa5Nb0.5O3 thin films. Ionics. 2014,20,1795-1799. DOI: 10.1007/s11581-014-1281-2. Kumar P., Palei P., Effect of sintering temperature on ferroelectric properties of 0.94(K5Na0.5)NbO3-0.06LNbO3system. Ceramics International. 2010,36,1725-1729. DOI: 10.1016/j.ceramint.2010.02.014. Chunli D., Hanxing L., Haiwu Z., et al. Enhanced energy storage properties of BaTiO3 thin films by Ba4Sr0.6TiO3, layers modulation. Journal of Alloys and Compounds. 2018,765, 362-368. DOI: 10.1016/j.jallcom.2018.06.199. The losses in the figure 3c seem strange. The material is in the ferroelectric state, the dielectric constant and the losses must describe a butterfly loop. However, for the losses, there is two waves. Can you explain it. For this figure, it will be better to plot the electric field and not the voltage for a comparison with the figure 5. The correct term is "tunability" and not "tenability". I don't understand why the values of the dielectric constant in figure 3c do not correspond to the values in the figure 3b...
Due to the selection of coordinate values, we adjusted it in figure 3c. Meanwhile, we have made correction according to the Reviewer’s comments about the use of "tenability" and marked the correct part in yellow. Secondly, the values of the dielectric constant in figure 3c are approximately equal to the values in figure 3a. Whereas the differences in numerical value of the three tests may be due to the different selection of measurement points. The film prepared by sol-gel method cannot achieve perfect homogeneity, there may be a small amount of error.
Please, put the same scale in figure 4a and 4b.Why do you choose 30°C and 50°C ?
As Reviewer suggested that we have put the same scale in figure 4a and 4b. Leakage current density has an important relationship with temperature[1-2]. The phase transition temperatures of the films we prepared are near room temperature, so we hope that the leakage current densities are small in room temperature range or a little higher temperature range, which is conducive to promoting the application of the film at room temperature.
Bouyssou E., Leduc P., Guégan g., et al. Leakage current conduction in IrO2/PZT/Pt structures.Journal of Physics Conference Series. 2005,10, 317-320. DOI: 10.1088/1742-6596/10/1/078. LiRen Z, GuangNing W, Yang L. Injection and Conduction Mechanisms of Charge Carriers in Insulators. Journal of Materials Science &Engineering. 2010,28,102-105. DOI: 10.14136/j.cnki.issn1673-2812.2010.01.034. It will be interesting to have the PZT loop in figure 5a. I don't understand the stability of the polarization as a function of the temperature. The polarization and the dielectric constant are related and the dielectric constant is not stable (figure 3a) .
Considering the Reviewer’s suggestion, we have summarized the following points through investigating literature and combining with some theoretical knowledge[1-3]. First, the influence of temperature on the dielectric constant of ferroelectrics is discussed. When the temperature of the medium increases, the thermal motion of each unit cell is intensified, and the direction uniformity of the molecular dipole moment in the domain is destroyed to some extent, and a certain degree of freedom is obtained. The molecular electric dipole will overcome the tendency of its orientation disorder due to thermal motion under the action of an external electric field, and turn to the direction consistent with the external electric field. Therefore, the dielectric constant of ferroelectrics increases with increasing temperature. When the temperature rises to the Curie temperature Tc, the self-generating domain in the medium completely disintegrates and the ferroelectricity disappears. And when the temperature continues to rise from Tc, there is no longer the effect of electric dipole released from the electric domain tending to the external field, and only the thermal motion makes the distribution disorder, so the dielectric constant decreases with the increase of temperature.
Second, ferroelectric character exists only in a certain temperature range. At lower temperatures, they have a less symmetric crystal structure, ferroelectricity and spontaneous polarization. When the temperature rises to the phase transition temperature, the crystal symmetry changes from a less symmetric structure to a more symmetric structure, the spontaneous polarization disappears and becomes a paraelectric phase. In this process of temperature rise, the spontaneous polarization gradually decreases to disappear. The corresponding saturation polarization value does not disappear but it decreases monotonously.
In conclusion, although the saturation polarization value and dielectric constant are both related to temperature, while they have a little difference.
Xi C, Xiaojun Q, Liaoyuan Z, et. al. Temperature dependence of ferroelectricity and domain switching behavior in Pb(Zr3Ti0.7)O3 ferroelectric thin films. Ceramics International. 2019,45,18030-18036. DOI: org/10.1016/j.ceramint.2019.06.022. Feng Z., Shi D., Zeng R., et. al. Large electrocaloric effect of highly (100)-oriented 0.68PbMg1/3Nb2/3O3-0.32PbTiO3 thin films with a Pb(Zr3Ti0.7)O3/PbOx buffer layer. Thin Solid Films. 2011,519, 5433-5436. DOI: 10.1016/j.tsf.2011.02.069. Kaddoussi H., Gagou Y., Lahmar A., et. al. Room temperature electro-caloric effect in lead-free Ba(Zr1Ti0.9)1-xSnxO3(x=0, x=0.075) ceramics. Solid State Communications. 2015,201,64-67. DOI: rg/10.1016/j.ssc.2014.10.003. Why double slim loops are typical relaxer behavior and not soft ferroelectric behavior ?
Thanks for the Reviewer’s professional guidance. By referring to relevant literatures [1-4], we have a deeper understanding of the concepts of soft ferroelectric and relaxation ferroelectric. It can be obtained from the P-E loop that BZT and PZT/BZT/PZT thin films both show the small residual polarization value, low coercive force field. So the samples show typical order-disorder ferroelectric characteristics, that is, the soft ferroelectric characteristic, as the reviewer mentioned. While relaxation properties are generally related to dielectric properties of the thin films rather than hysteresis loops, so we have made changes here that ‘relaxer’ is replaced by ‘soft’.
Morozov M. I., Damjanovic D.. Charge migration in Pb(Zr,Ti)O3 ceramics and its relation to ageing, hardening, and softening. Journal of Applied Physics. 2010,107, 034106. DOI:10.1063/1.3284954. Morozov M. I., Damjanovic D.. Hardening-softening transition in Fe-doped Pb(Zr,Ti)O3 ceramics and evolution of the third harmonic of the polarization response. Journal of Applied Physics. 2008, 104, 034107. DOI:10.1063/1.2963704. Cross L. E.. Relaxor ferroelectrics. Ferroelectrics. 1987,76, 241-267. DOI:10.1007/978-3-540-68683-5_5 Maiti T., Guo R., Bhalla A.S.. Electric field dependent dielectric properties and high tunability of BaZrxTi1-xO3 relaxor ferroelectrics. Applied Physics Letters. 2006,89, 3078. DOI:10.1063/1.2354438. I think that the relation (2) is not correct. Please verify in the references [24-26].
We have made correction according to the Reviewer's comments, and the correction as follows: ,.
In the table 1, I think is not Jloss but the energy efficiency.
We are very sorry for our negligence and we have corrected it in the article.
The figures 6 are not correct. We can’t have negative energy here. It will be better if Jrec and Jloss have the same scal to really compare the two energies and see directly the difference between the two curves which correpond to Jchg.
Thank you for pointing this out. Because there is a large difference between energy storage density and energy loss, in order to clearly show the trend of energy storage density and energy loss with the electric field increasing, a bi-coordinate approach is adopted here, in which the energy loss corresponds to the right coordinate.
Thank you again for your positive comments on our manuscript. If there are any other modifications we could make, we would like very much to modify them and we really appreciate your help.
Sincerely yours.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
After reviewing the last edition of the paper's draft, I found the manuscript has been improved, and I am satisfied with the authors' response to my earlier comments. I think that the article can be published now.
