Amphoteric Doping Effect of Ho³⁺ on the Performance of Medium-Temperature-Sintered PLZT Energy Storage Ceramics

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

General Comments

As a reviewer, I find this manuscript to be a solid contribution to the field of dielectric energy storage ceramics, particularly in exploring the role of rare-earth doping (Ho³⁺) in PLZT systems combined with medium-temperature sintering. The work demonstrates novelty in investigating the amphoteric behavior of Ho³⁺ and its impact on phase transitions, microstructure, and energy storage performance. The experimental methods are comprehensive, covering structural (XRD, Raman, SEM), ferroelectric (P-E loops), charge-discharge, and nanoscale (PFM) characterizations. The results are well-presented, and the conclusions align with the data, highlighting improved recoverable energy density (W_rec = 0.91 J/cm³), efficiency (η = 87%), power density (P_D = 7.4 MW/cm³), and discharge time (t_{0.9} = 69.2 ns) for the H4 sample.

However, the manuscript has some limitations that warrant revision for clarity, completeness, and scientific rigor. Major issues include insufficient details on the sintering aid composition, limited testing on higher-doped samples (e.g., charge-discharge and PFM only on H1), and a need for more quantitative evidence supporting the amphoteric doping mechanism. Comparisons with state-of-the-art PLZT-based ceramics could strengthen the claims of superiority. The writing is generally clear but could benefit from improved English grammar and flow in places (e.g., abstract and introduction). I recommend major revisions before acceptance, as the core science is sound but requires bolstering to meet journal standards.

Major Concerns and Questions

1. Sintering Aid Details: The experimental section mentions incorporating Mg_{0.8}Mo_{0.2}O_{1.4} as a sintering aid to enable medium-temperature sintering (1030°C), but the amount added (e.g., wt% or mol%) is not specified. This is critical for reproducibility. How was the sintering aid prepared or added? Did it affect the phase purity or defect chemistry? Provide quantitative details and discuss any potential interactions with Ho³⁺ doping.
2. Amphoteric Doping Mechanism: The paper claims Ho³⁺ exhibits amphoteric behavior (A-site donor and B-site acceptor), supported by no systematic XRD peak shifts and Raman asymmetry in B-O modes. However, this evidence is indirect. Why not use techniques like XPS, EPR, or DFT simulations to confirm site occupancy? The ionic radius argument (Ho³⁺ = 0.894 Å vs. Pb²⁺ = 1.20 Å for A-site; vs. Zr⁴⁺/Ti⁴⁺ ≈ 0.72/0.605 Å for B-site) is mentioned, but calculations of lattice parameters or tolerance factors for each composition would strengthen this. Question: What fraction of Ho³⁺ is estimated to occupy B-sites, and how does this evolve with x (0.01–0.04)?
3. Limited Testing on Samples: Charge-discharge performance and PFM imaging are only reported for the H1 sample (lowest Ho³⁺ content), despite H4 showing the best W_rec and breakdown strength (E_b = 216 kV/cm). This is a significant gap, as the paper emphasizes Ho³⁺-induced improvements. Why were these tests not extended to H2–H4? Provide data for at least H4 to validate the ultrafast discharge claims across compositions. Similarly, dielectric properties (e.g., ε_r vs. temperature/frequency) are missing, which are essential for confirming relaxor behavior.
4. Energy Storage Performance Comparison: The achieved W_rec (0.91 J/cm³) and η (87%) are good, but how do they compare to other medium-temperature-sintered PLZT or similar systems (e.g., Pb-free alternatives or higher-doped rare-earth PLZTs)? References [2–4,6–7] are cited, but a table summarizing key metrics (W_rec, η, E_b, P_D, t_{0.9}) from literature vs. this work would highlight novelty. Is 0.91 J/cm³ truly "excellent" for PLZT at this thickness (0.1 mm)? Discuss limitations, such as why W_d (0.18 J/cm³) is lower than W_rec due to high-frequency losses.
5. Microstructure and Grain Refinement: Grain sizes decrease from 0.625 µm (H1) to 0.549 µm (H4), attributed to Ho³⁺ pinning and medium-temperature sintering. However, no density measurements (e.g., Archimedes method) are provided to confirm "dense" microstructures. Question: What are the relative densities (>95%?) and how do they correlate with E_b? SEM images show homogeneity, but EDS mapping could confirm Ho distribution and rule out segregation.
6. Raman Deconvolution: Only H1's Raman spectrum is deconvoluted (Fig. 3b). Extend this to H4 for quantitative analysis of mode broadening/dispersion as evidence of relaxor transition. The disappearance of the 145 cm⁻¹ peak is well-noted, but discuss implications for defect dipoles (e.g., Ho_A^{3+} - V_O^{••} or Ho_B^{3+} - V_Pb^{''}).
7. Ferroelectric and Discharge Data: P-E loops are slim, indicating relaxor behavior, but frequency/temperature dependence of loops would confirm this. For discharge, underdamped curves show oscillations—discuss circuit parameters (e.g., inductance) affecting this. The efficiency drop from quasi-static (87%) to dynamic (75%) is attributed to dielectric losses, but quantify loss tangent (tan δ) or impedance spectroscopy data.

Minor Comments and Suggestions

1. Abstract: The phrase "medium-temperature sintering effectively mitigated Pb volatilization" is repeated in the introduction; consolidate for conciseness. Specify "medium-temperature" as 950–1100°C earlier.
2. Introduction: Expand on why PLZT is chosen over other perovskites (e.g., BT, BNT). Reference [9] is from 2002—update with recent works on Ho-doping in perovskites.
3. Experimental Procedures: Sample thickness is ~0.1 mm after polishing—clarify if electrodes (e.g., Ag, Pt) were applied and their area for P-E testing. Raman spectrometer details (laser wavelength?) are missing.
4. Results – XRD (Fig. 1): Label the superlattice peaks more clearly (e.g., arrows). Discuss Rietveld refinement results for lattice parameters to support no-shift claim.
5. Results – SEM (Fig. 2): Insets show grain size distributions; provide standard deviations and number of grains measured (>100?).
6. Results – Raman (Fig. 3): Y-axis label as "Intensity (a.u.)" consistently.
7. Results – P-E Loops (Fig. 4): Why unipolar instead of bipolar? Report P_r and E_c values explicitly in text.
8. Results – Charge-Discharge (Fig. 5): Inset in Fig. 5c is small; enlarge for readability. Discuss why only 100 kV/cm max—limited by setup?
9. Results – PFM (Fig. 6): Topography scale bars vary; standardize. Phase image shows domains—estimate domain size (~nm?) and discuss evolution with Ho content.
10. Conclusion: Avoid repeating results; focus on broader implications (e.g., for MLCCs with Ni/Cu electrodes).
11. References: Some are outdated (e.g., [8] from 2007); add recent reviews on relaxor ferroelectrics. Check formatting (e.g., inconsistent journal abbreviations).
12. Typos and Clarity:
    • Page 1, line 20: "7.4 MW/cm3" → "7.4 MW/cm³" (superscript).
    • Page 3, line 111: "Pb2+ (1.20 Å )" → "Pb²⁺ (1.20 Å)".
    • Page 5, line 159: "8.4 μC/cm2" → "8.4 µC/cm²".
    • Improve sentences like "This work not only offers a viable strategy... but also elucidates..." for better flow.

Recommendations

• Revise and Resubmit: Address major concerns with additional data (e.g., dielectric spectra, extended testing on H4).
• Figures/Tables: Add a table comparing performance metrics with literature.
• Length: The manuscript is concise; expansions won't exceed limits. This paper has strong potential for publication in Coatings after revisions.

Author Response

First, we would like to express our thanks to the Reviewers for their instructive comments concerning our manuscript entitled ‘Amphoteric Doping Effect of Ho³⁺ on the Performance of Medium-Temperature-Sintered PLZT Energy Storage Ceramics’(coatings-3847716). We have studied the comments carefully, and then revised the whole manuscript. The revisions/explanation corresponding to comment are shown in the following (Reviewers' comments are in italic font):

 

1. Sintering Aid Details: The experimental section mentions incorporating Mg₈Mo_0.2O_1.4 as a sintering aid to enable medium-temperature sintering (1030°C), but the amount added (e.g., wt% or mol%) is not specified. This is critical for reproducibility. How was the sintering aid prepared or added? Did it affect the phase purity or defect chemistry? Provide quantitative details and discuss any potential interactions with Ho³⁺ doping.

A: Thank you for your valuable comments regarding the sintering aid. The addition amount of Mg_0.8Mo_0.2O_1.4 was 1.0 wt%. This sintering aid was developed in our group based on long-term experimental studies, and both its composition and dosage were determined from our previous research; therefore, it was not the focus of investigation in this work. XRD analysis did not reveal any additional secondary phases, indicating that the sintering aid was either fully dissolved into the lattice or formed an amorphous phase at the grain boundaries, without adversely affecting the purity of the main phase. The primary role of the sintering aid is to lower the densification temperature and suppress abnormal grain growth, and it does not undergo direct chemical reactions with Ho³⁺ doping. The synergy between the two enables dense microstructures and stable electrical properties to be achieved under medium-temperature sintering conditions.

In addition, no Mg/Mo-related secondary phase peaks were observed in the XRD patterns. Furthermore, EDS mapping performed in both grain interiors and grain boundary regions revealed no Mg or Mo enrichment or continuous phases, with their signals being close to the noise level. Therefore, it can be reasonably concluded that Mg_0.8Mo_0.2O_1.4 acted merely as a flux during the sintering process and did not form any detectable residual phase in the final ceramics. The corresponding EDS mapping images of Mg and Mo are provided below:

2. Amphoteric Doping Mechanism: The paper claims Ho³⁺ exhibits amphoteric behavior (A-site donor and B-site acceptor), supported by no systematic XRD peak shifts and Raman asymmetry in B-O modes. However, this evidence is indirect. Why not use techniques like XPS, EPR, or DFT simulations to confirm site occupancy? The ionic radius argument (Ho³⁺ = 0.894 Å vs. Pb²⁺ = 1.20 Å for A-site; vs. Zr⁴⁺/Ti⁴⁺ ≈72/0.605 Å for B-site) is mentioned, but calculations of lattice parameters or tolerance factors for each composition would strengthen this. Question: What fraction of Ho³⁺ is estimated to occupy B-sites, and how does this evolve with x (0.01–0.04)?

A: We sincerely appreciate your valuable comments regarding the amphoteric doping mechanism. We fully agree that the evidence provided by XRD and Raman spectroscopy is indeed indirect. To address this, we promptly carried out XPS measurements on H1 and H4 to confirm the site occupancy, and we have added a new section in the main text presenting the XPS analysis. The detailed additions are as follows:

Page 7. Line 180-203 have been added:

 

Figure 5. XPS spectra of (a-c) H1 and (d-f) H4 ceramics.

The X-ray photoelectron spectroscopy (XPS) results of H1 and H4 are presented in Figure 5. As shown in Figure 5(a) and 5(d), the Ho 4d spectra of both H1 and H4 were deconvoluted into two spin-orbit components, Ho 4d_5/2 near 161 eV and Ho 4d_3/2 near 166 eV. These binding energies are in good agreement with the reference values of Ho³⁺ in Ho₂O₃, confirming that the incorporated holmium ions predominantly exist in the ceramics in a stable +3 oxidation state.[22,23] In addition, the Pb 4f spectra of H1 and H4 (Figure 5(b) and 5(e)) exhibit the characteristic Pb²⁺ spin-orbit doublet (Pb 4f_7/2 ≈ 138 eV, Pb 4f_5/2 ≈ 143 eV), without noticeable chemical shifts between H1 and H4, suggesting that the oxidation state and chemical environment of the A-site cations remain unchanged.

The O 1s spectra (Figure 5(c) and 5(f)) can be fitted into three components. The low-binding-energy peak (~529.5 eV) is assigned to lattice oxygen (O^2-) in the perovskite framework, the medium-binding-energy peak (~531.0 eV) is associated with oxygen atoms adjacent to oxygen vacancies (O_vac), and the high-binding-energy peak (~532.2 eV) is attributed to adsorbed oxygen (O_ads) at grain boundaries.[24] Importantly, the relative fraction of the O_vac peak increases from 66.12% in H1 to 70.31% in H4, indicating a significant rise in oxygen vacancy concentration with increasing Ho³⁺ content. According to defect chemistry, substitution of Ho³⁺ for A-site Pb²⁺ (donor doping) is primarily compensated by lead vacancies ( ) and should not generate substantial oxygen vacancies. In contrast, substitution of Ho³⁺ for B-site Zr⁴⁺/Ti⁴⁺ (acceptor doping) requires charge compensation through the formation of oxygen vacancies ( ). Therefore, the increased oxygen vacancy fraction in highly doped samples provides strong evidence that a portion of the Ho³⁺ ions occupy the B-site.

The calculation of the tolerance factor effectively supports our proposed amphoteric doping mechanism. According to the Goldschmidt formula, the oxygen ionic radius is r_O = 1.4 Å. For H1, assuming Ho³⁺ occupies the A-site, we have r_A = 1.1927 Å and r_B = 0.71 Å, yielding a tolerance factor t = 0.8689. For H2, still assuming Ho³⁺ occupies only the A-site, r_A = 1.1896 Å and r_B = 0.71 Å, resulting in t = 0.8678. These results indicate that if Ho³⁺ exclusively occupies the A-site, the tolerance factor t would decrease with increasing doping content, implying an enhancement of lattice distortion. However, this is completely inconsistent with our XRD observations, which show a reduction of distortion and a transition toward a higher-symmetry pseudo-cubic phase. Only when Ho³⁺ is also considered to occupy the B-site does the increase in the average B-site radius lead to a larger t value, thereby driving the structural evolution toward a higher-symmetry phase. Our XPS analysis confirms that the occupancy of Ho³⁺ at the B-site is non-negligible. Although we cannot provide an exact percentage, the trend is clear: the fraction of Ho³⁺ at the B-site increases with the total doping level.

3. Limited Testing on Samples: Charge-discharge performance and PFM imaging are only reported for the H1 sample (lowest Ho³⁺ content), despite H4 showing the best W_rec and breakdown strength (E_b = 216 kV/cm). This is a significant gap, as the paper emphasizes Ho³⁺-induced improvements. Why were these tests not extended to H2–H4? Provide data for at least H4 to validate the ultrafast discharge claims across compositions. Similarly, dielectric properties (e.g., ε_r vs. temperature/frequency) are missing, which are essential for confirming relaxor behavior.

A: We sincerely appreciate your valuable comments. Since PFM measurements were only performed on H1, the initial reporting of charge–discharge tests included only the performance of H1 for consistency in sample analysis. We have now included the data of H4 in the main text. In addition, thanks to your careful guidance, we noticed that the inset in Figure 5(c) was relatively small; it has now been enlarged for better readability, and the t₀.₉ value obtained from the inset has been corrected accordingly. The detailed modifications are as follows:

Page 9. Line 238-272 have been modified:

 

Figure 8. (a) The underdamped discharge current curves of H1. (b) The overdamped discharge current curves of H1. (c) The variation of discharge energy density of H1 with time. (d) The C_D and P_D of H1 under different electric fields. (e-h) The corresponding pulsed discharge performances of H4.

To assess the suitability of PLZTH ceramics sintered at intermediate temperatures for practical pulsed power applications, pulsed charge-discharge measurements were performed on H1 and H4. In the underdamped mode (Figure 8(a) and 8(e)), both samples exhibited a linear increase in the first peak current density with the applied field, and the discharge process was completed after 2-4 oscillations. Such oscillatory behavior originates from the intrinsic RLC characteristics of the testing circuit, where the oscillation period follows . Here, L represents the equivalent inductance of the circuit and C is the capacitance of the sample; since all measurements were carried out on the same platform, L can be regarded as constant. The similarity of oscillation profiles at different fields indicates that the capacitance of the ceramics remains stable during nanosecond-scale discharge. Beyond W_rec, current density (C_D) and P_D are also critical metrics for evaluating pulsed capacitor dielectrics.[34,35] As shown in Figure 8(d) and 8(h), both C_D and P_D increase monotonically with the applied field, reaching 148.1 A/cm² and 7.4 MW/cm³ for H1 at 100 kV/cm and 320.3 A/cm² and 28.8 MW/cm³ for H4 at 180 kV/cm.

In the overdamped mode with a load resistance of 205 Ω (Figure 8(b) and 8(f)), all discharge current traces exhibited a rapid rise followed by exponential decay. The peak discharge currents increased progressively with the applied field, reaching 6.9 A for H1 and 8.5 A for H4. Integration of the current-time profiles yielded the discharge energy density (W_d), as shown in Figure 8(c) and 8(g). Wd increased linearly with field strength, attaining 0.18 J/cm³ for H1 at 100 kV/cm and 0.43 J/cm³ for H4 at 180 kV/cm. These values correspond to 75% and 73% of the recoverable energy densities obtained from the quasi-static P-E loops (H1: 0.24 J/cm³; H4: 0.59 J/cm³). The slight energy loss is attributed to intrinsic dielectric dissipation under high-frequency conditions during nanosecond-scale discharge, where polar nanoregions cannot fully follow abrupt field reversals, and conduction and interfacial polarization losses convert part of the stored energy into heat.[36–38]

Throughout the entire test range, t_0.9 remained below 20 ns, with values of 18.85 ns for H1 at 100 kV/cm and as low as 4.97 ns for H4 at 180 kV/cm, demonstrating an ultrafast discharge response. The enhanced charge transport and energy release efficiency induced by Ho³⁺ doping further highlights the promise of medium-temperature sintered PLZTH ceramics as advanced dielectric candidates for pulsed power capacitors.

Meanwhile, we promptly performed dielectric measurements, and the detailed results are as follows:

We conducted temperature-dependent dielectric characterization on the H4 ceramic, as shown in the figure above. The relative permittivity (ε_r) exhibits typical relaxor behavior with a broad and diffuse dielectric peak, maintaining a plateau over a wide temperature range, without the sharp Curie peak characteristic of normal ferroelectrics. This indicates that the system undergoes a diffuse phase transition. More importantly, ε_r shows a strong frequency dispersion: at the same temperature, ε_r decreases significantly with increasing measurement frequency. In addition, the peak temperature of the dielectric maximum (T_m) shifts noticeably toward higher temperatures, especially evident in the 100 Hz curve. The presence of this diffuse dielectric peak and pronounced frequency dispersion provides clear evidence of relaxor behavior.

4. Energy Storage Performance Comparison: The achieved W_rec (0.91 J/cm³) and η (87%) are good, but how do they compare to other medium-temperature-sintered PLZT or similar systems (e.g., Pb-free alternatives or higher-doped rare-earth PLZTs)? References [2–4,6–7] are cited, but a table summarizing key metrics (W_rec, η, E_b, P_D, t_{0.9}) from literature vs. this work would highlight novelty. Is 0.91 J/cm³ truly "excellent" for PLZT at this thickness (0.1 mm)? Discuss limitations, such as why W_d (0.18 J/cm³) is lower than W_rec due to high-frequency losses.

A: We sincerely appreciate your valuable comments regarding performance comparison and in-depth discussion. We fully agree that placing our work in the context of existing studies through a comparative analysis is crucial for highlighting its novelty and advantages. We reviewed the performance parameters of recently reported ceramics and compiled them into a performance comparison figure. A new section analyzing this comparison has been added to the main text. The detailed additions are as follows:

Page 8. Line 219-236 have been added:

Figure 7. A comparison of energy densities and efficiencies of recently reported (anti)ferroelectric-based energy storage dielectric material.

Figure 7 summarizes the reported energy storage performances of representative Pb-based (anti)ferroelectric ceramics in recent years.[25–33] As can be seen, most systems exhibit a trade-off between W_rec and η. For instance, conventional PLZT and PLBZST ceramics typically show η values in the range of 60-75%, and although some compositions achieve W_rec values approaching 1 J/cm3, their energy conversion efficiency remains limited. Conversely, systems such as PSZT and PMN-PT demonstrate relatively higher efficiencies, but their W_rec usually does not exceed 0.6 J/cm³.

In contrast, the PLZTH ceramics developed in this work achieve a W_rec of 0.91 J/cm³ and an η of 87% at a thickness of 0.2-0.3 mm, thereby striking a more favorable balance between energy density and efficiency. Compared with other reported PLZT-based ceramics, PLZTH maintains a moderate energy storage density while exhibiting a remarkable advantage in energy conversion efficiency. This indicates that Ho³⁺ doping not only sustains a relatively high W_rec but also effectively suppresses energy dissipation during polarization switching, leading to reduced energy loss under fast charge-discharge conditions.

The reason why W_d is lower than W_rec has also been specifically explained in the manuscript, with relevant references cited. The details are as follows:

Page 9. Line 263-267 have been added:

The slight energy loss is attributed to intrinsic dielectric dissipation under high-frequency conditions during nanosecond-scale discharge, where polar nanoregions cannot fully follow abrupt field reversals, and conduction and interfacial polarization losses convert part of the stored energy into heat.[36–38]

5. Microstructure and Grain Refinement: Grain sizes decrease from 0.625 µm (H1) to 0.549 µm (H4), attributed to Ho³⁺ pinning and medium-temperature sintering. However, no density measurements (e.g., Archimedes method) are provided to confirm "dense" microstructures. Question: What are the relative densities (>95%?) and how do they correlate with E_b? SEM images show homogeneity, but EDS mapping could confirm Ho distribution and rule out segregation.

A: We sincerely appreciate your valuable comments regarding density and elemental distribution. Using the Archimedes method, the densities of H1–H4 ceramics were measured to be 7.30 g/cm³, 7.43 g/cm³, 6.85 g/cm³, and 6.62 g/cm³, respectively. Based on the theoretical density (~7.9-8.0 g/cm³), the relative densities of these samples range from 84% to 94%. We acknowledge that the densification, particularly for the high-doping samples, is not optimal.

However, an interesting observation is that although the H4 sample has the lowest relative density, it exhibits the highest breakdown field. This indicates that in this system, the grain refinement effect induced by Ho³⁺ doping contributes far more to the enhancement of breakdown strength than the relative density. We have revised the SEM analysis section accordingly. The detailed modifications are as follows:

Page 5. Line 130-142 have been modified:

 

Figure 2. The SEM images of H1-H4 ceramics. The insets show the grain size distribution.

The cross-sectional SEM images and grain size distributions of H1-H4 ceramics are presented in Figure 2(a-d). All samples consist of densely packed grains with clear boundaries, indicating high sintering density. The average grain sizes of H1, H2, H3, and H4 are 0.625 μm, 0.573 μm, 0.566 μm, and 0.549 μm, respectively, showing a systematic decrease with increasing Ho³⁺ content. Such grain refinement is attributed to the effective suppression of abnormal grain growth by the medium-temperature sintering process, together with the grain-boundary migration pinning effect induced by Ho³⁺ doping.[17] The reduced grain size lowers the probability of large-scale defect formation and restricts the mean free path of charge carriers, thereby suppressing avalanche breakdown under high electric fields. Consequently, the breakdown strength (E_b) is enhanced, enabling the ceramics to sustain higher operating fields and resulting in an increased W_rec.[18,19]

In addition, we performed energy-dispersive X-ray spectroscopy (EDS) surface scanning on the H4 ceramic, and the results have also been added to the main text. The detailed content is as follows:

Page 5. Line 143-155 have been added:

 

Figure 3. (a) SEM image of a selected area on the surface of H4 ceramic, and (b-f) the corresponding energy-dispersive X-ray spectroscopy (EDS) elemental mapping images for Pb, La, Zr, Ti, and Ho, respectively.

An energy-dispersive X-ray spectroscopy (EDS) mapping analysis was performed on H4 to evaluate the chemical homogeneity and the spatial distribution of the doped elements (Figure 3). The SEM image of the analyzed region is presented in Figure 3(a). The corresponding elemental mapping results (Figure 3(b-e)) reveal that the major constituent elements, including Pb, La, Zr, and Ti, are uniformly distributed within the microstructure. As shown in Figure 3(f), although Ho is present only as a trace dopant, its signal remains detectable and exhibits a homogeneous distribution similar to that of the primary elements. Importantly, no evidence of Ho enrichment or local segregation was found across the entire scanned area.

6. Raman Deconvolution: Only H1's Raman spectrum is deconvoluted (Fig. 3b). Extend this to H4 for quantitative analysis of mode broadening/dispersion as evidence of relaxor transition. The disappearance of the 145 cm⁻¹ peak is well-noted, but discuss implications for defect dipoles (e.g., Ho_A^{3+} - V_O^{••} or Ho_B^{3+} - V_Pb^{''}).

A: We sincerely appreciate your valuable comments regarding the Raman spectroscopy analysis. We fully agree that such an analysis can greatly strengthen our argument on the relaxor phase transition. Following your suggestion, we have performed peak deconvolution on the spectra of all H1–H4 ceramics and integrated the results into the new Figure 3(a). We have also discussed the related effects of defect dipoles. The detailed content is as follows:

Page 6. Line 157 and 170-178 have been modified:

Fig 3. Raman spectra of H1-H4 ceramics.

Amphoteric doping of Ho³⁺ inevitably introduces point defects to maintain charge neutrality. As shown in Figure 4(a), the gradual disappearance of the 145 cm^-1 peak with increasing Ho³⁺ content indicates the collapse of long-range translational symmetry, which may be associated with weakened A-site ordering caused by  defect dipoles. In addition, the formation of defect dipoles generates local and random electric and stress fields in the lattice, which strongly scatter phonons and shorten their lifetimes. This results in a pronounced broadening of all Raman peaks, a phenomenon closely related to the incorporation of  dipoles.

7. Ferroelectric and Discharge Data: P-E loops are slim, indicating relaxor behavior, but frequency/temperature dependence of loops would confirm this. For discharge, underdamped curves show oscillations—discuss circuit parameters (e.g., inductance) affecting this. The efficiency drop from quasi-static (87%) to dynamic (75%) is attributed to dielectric losses, but quantify loss tangent (tan δ) or impedance spectroscopy data.

A: We sincerely appreciate your insightful comments and valuable suggestions regarding our ferroelectric and discharge data. These issues address the core of the material’s electrical performance and are extremely helpful for improving our work. Variable-frequency or variable-temperature P–E loop measurements are classical methods for confirming relaxor behavior. However, due to current experimental limitations, we were unable to perform these measurements. We fully recognize this as an area for further improvement and plan to systematically investigate the temperature- and frequency-dependent dielectric spectra of this material in future work to more comprehensively characterize its relaxor behavior. Nevertheless, we have actively conducted dielectric measurements, which confirm the relaxor behavior.

In addition, we have also discussed the circuit parameters. The detailed content is as follows:

Page 9. Line 246-250 have been added:

Such oscillatory behavior originates from the intrinsic RLC characteristics of the testing circuit, where the oscillation period follows . Here, L represents the equivalent inductance of the circuit and C is the capacitance of the sample; since all measurements were carried out on the same platform, L can be regarded as constant.

8. Abstract: The phrase "medium-temperature sintering effectively mitigated Pb volatilization" is repeated in the introduction; consolidate for conciseness. Specify "medium-temperature" as 950–1100°C earlier.

A: We sincerely appreciate your valuable comments. In the revised abstract, we have consolidated the description of the medium-temperature sintering to avoid redundancy and have explicitly specified its temperature range (950–1100 °C) at the beginning to enhance clarity. The revised abstract content is as follows:

The development of dielectric capacitors with high energy-storage density and ultrafast discharge capability is essential for next-generation pulsed power systems. In this work, (Pb, La, Ho, Zr, Ti)O₃ (PLZTH) ceramics were fabricated via medium-temperature sintering (950-1100℃) combined with Ho³⁺ doping to systematically tailor their energy-storage properties. This processing strategy not only mitigates Pb volatilization but also enhances compatibility with base-metal electrodes such as Ni and Cu. In addition, Ho³⁺ ions exhibit amphoteric doping behavior, which contributes to the enhancement of relaxor characteristics and grain refinement. H4 ceramic delivers an outstanding re-coverable energy-storage density (W_rec) of 0.91 J/cm3 and a high energy efficiency (η) of 87% under 216 kV/cm, along with a power density (P_D) of 28.8 MW/cm³ and an ultrafast discharge time (t_0.9) of only 4.97 ns at 180 kV/cm. This study not only proposes a viable route toward high-performance medium-temperature-sintered PLZT ceramics but also elucidates the effective mechanism of Ho³⁺ amphoteric doping in modulating the relaxor state and properties of perovskite-based ceramics.

9. Introduction: Expand on why PLZT is chosen over other perovskites (e.g., BT, BNT). Reference [9] is from 2002—update with recent works on Ho-doping in perovskites.

A: We sincerely appreciate your valuable suggestions on improving the Introduction. We fully agree that clearly explaining the rationale for selecting the PLZT system and citing the most recent relevant literature is crucial for enhancing the manuscript’s persuasiveness and relevance. Accordingly, we have expanded the Introduction by comparing PLZT with other perovskite systems such as BT and BNT, providing a more detailed justification for choosing PLZT as the research platform and highlighting its unique advantages in terms of high polarization and tunable phase transitions. The detailed content is as follows:

Page 1. Line 29-41 have been modified:

Within the broad family of perovskite oxides, lead-free ceramics such as barium titanate (BT), sodium bismuth titanate (BNT), silver niobate (AgNbO₃), and sodium niobate (NaNbO₃) have been intensively investigated owing to their environmental benignity. However, their practical application is generally constrained by intrinsically low breakdown strengths and relatively broad P-E loops, which in turn limit both the re-coverable energy density and storage efficiency.[3–5] By contrast, lanthanum-modified lead zirconate titanate (PLZT) ceramics combine outstanding ferroelectric/antiferroelectric properties with a high dielectric permittivity, and their phase transition characteristics can be flexibly tailored through rare-earth substitution or Zr/Ti ratio adjustment. Such tunability enables the realization of slim P-E loops together with high energy conversion efficiencies, rendering PLZT one of the most competitive and scalable candidates for advanced dielectric energy-storage applications.[6,7]

In addition, we have also updated the references. One of the new references, published in 2022, discusses defect engineering in rare-earth-doped BaTiO₃ ceramics, including the doping behavior of Ho³⁺. The details of the updated reference are as follows:

Page 11. Line 343-345 have been modified:

[12] Meng, Y.; Liu, K.; Zhang, X.; Lei, X.; Chen, J.; Yang, Z.; Peng, B.; Long, C.; Liu, L.; Li, C. Defect Engineering in Rare-earth-doped BaTiO₃ Ceramics: Route to High‐temperature Stability of Colossal Permittivity. J Am Ceram Soc. 2022, 105, 5725-5737, doi:10.1111/jace.18512.

10. Experimental Procedures: Sample thickness is ~0.1 mm after polishing—clarify if electrodes (e.g., Ag, Pt) were applied and their area for P-E testing. Raman spectrometer details (laser wavelength?) are missing.

A: We sincerely appreciate your comments regarding the insufficient detail in the Experimental Procedures section. For the charge–discharge measurements, the sample thickness was 0.14 mm, and silver electrodes with a diameter of 3 mm were used. For the P-E loop measurements, the sample thickness ranged from 0.2 to 0.3 mm, with small silver electrodes of 3 mm diameter and large electrodes of 6 mm diameter. In addition, the laser wavelength of the Raman spectrometer was 532 nm. The detailed content is as follows:

Page 2. Line 85-97 have been modified:

Raman spectra were acquired using a micro-Raman spectrometer (LabRAM HR Evolu-tion, Horiba Scientific, Japan) with the excitation source of 532 nm. P-E hysteresis loops were characterized employing a ferroelectric analyzer (TF2000, aixACCT, Germany). For this purpose, ceramic disks with a thickness of 0.2–0.3 mm were prepared, in which a top electrode of silver paste with a diameter of 3 mm was deposited, while the bottom side was coated with a larger 6 mm electrode to act as the ground contact. The practical charge-discharge performances of the ceramics were assessed on a commercial testing platform (CFD-001, Gogo Instruments Technology, Shanghai, China). In this case, thinner specimens with a thickness of ~0.14 mm were employed, and symmetrical silver electrodes with a diameter of 3 mm were applied to both surfaces. X-ray photoelectron spectroscopy (XPS) measurements were performed using a spectrometer(Thermo Fisher, Thermo Scientific K-Alpha, USA).

11. Results – XRD (Fig. 1): Label the superlattice peaks more clearly (e.g., arrows). Discuss Rietveld refinement results for lattice parameters to support no-shift claim.

A: We sincerely appreciate your valuable comments. In the revised manuscript, we have used prominent arrows in Figures 1(c) and 1(d) to more clearly indicate the positions of the superlattice diffraction peaks. In addition, we performed Rietveld refinement, and the evolution of the obtained lattice parameters has been summarized in the newly added Figure 1(f). The detailed content is as follows:

 

Figure 1. (a) XRD patterns of H1-H4 ceramics. Enlarged patterns of H1-H4 ceramics in the regions of (b) 29.5°-31.5°, (c) 37°-38.5° and (d) 43°-44.5°. (e) Rietveld-refined XRD patterns of H1. (f) The lattice constants a, c and the tetragonal ratio c/a as a function of Ho3+ content.

Figure 1(e) displays the representative Rietveld refinement of H1, and the refined lattice parameters are summarized in Figure 1(f). The tetragonality (c/a ratio) decreases pro-gressively with increasing Ho3+ content.

12. Results – SEM (Fig. 2): Insets show grain size distributions; provide standard deviations and number of grains measured (>100?).

A: We sincerely appreciate your comments regarding our SEM images. The standard deviations for H1 to H4 are 0.254 µm, 0.185 µm, 0.229 µm, and 0.214 µm, respectively, and the corresponding numbers of measured grains are 266, 278, 324, and 302. We have also supplemented the insets with the standard deviation data. The detailed content is as follows:

 

13. Results – Raman (Fig. 3): Y-axis label as "Intensity (a.u.)" consistently.

A:We sincerely appreciate your comments regarding our Raman spectra figures. We have revised the original figures so that the y-axis label consistently reads “Intensity (a.u.)”. The detailed content is as follows:

 

14. Results – P-E Loops (Fig. 4): Why unipolar instead of bipolar? Report P_r and E_c values explicitly in text.

A: We sincerely appreciate your comments regarding the serious typographical error in the description of Figure 4. You are absolutely correct that the tested and presented loops are indeed bipolar P–E loops. We sincerely apologize for this oversight. In the revised manuscript, all instances of “unipolar” in the figure caption of Figure 4 and the corresponding text have been corrected to “bipolar.” Additionally, we have explicitly included the values of remanent polarization (P_r) and coercive field (E_c) extracted from the P-E loops in the main text. The detailed content is as follows:

Page 8. Line 206-211 have been modified:

Figure 6. (a) Bipolar P-E loop of H1-H4 ceramics. (b) The calculated Wrec and η of H1-H4 ceramics.

At the breakdown electric field and a frequency of 10 Hz, the bipolar P-E loops of H1-H4 ceramics are shown in Figure 6(a). Unlike the double hysteresis loops or broad hysteretic loops typically observed in long-range-ordered materials, all samples exhibit extremely slim P-E loops with low remanent polarization (Pr) and coercive field (E_c)(H4: P_r = 0.4 μC/cm², E_c = 7.9 kV/cm).

15. Results – Charge-Discharge (Fig. 5): Inset in Fig. 5c is small; enlarge for readability. Discuss why only 100 kV/cm max—limited by setup?

A: We sincerely appreciate your valuable suggestions regarding the presentation of figures and experimental details. The inset in Figure 5(c) of the original manuscript was indeed too small, which affected the clarity of the data. Based on your advice, we have redesigned Figure 5 and included the data for H4. The detailed content is as follows:

 

Based on the P–E loop measurements, the breakdown testing for H1 was conducted up to 100 kV/cm, whereas for H4 it was conducted up to 180 kV/cm.

16. Results – PFM (Fig. 6): Topography scale bars vary; standardize. Phase image shows domains—estimate domain size (~nm?) and discuss evolution with Ho content.

A: We sincerely appreciate your suggestions regarding our PFM images. We have remade Figure 6 to ensure that all scale bars are standardized. Measurements indicate that the domain widths are approximately in the range of 50-200 nm. Although PFM characterization was performed only on the H1 ceramic, based on the observed macroscopic relaxation trend from H1 to H4, it is reasonable to infer that with increasing Ho³⁺ content, the introduced chemical disorder and random fields severely disrupt the long-range polarization correlations observed in H1. Consequently, the sizes of these nanodomains further decrease, their structures become more fragmented, the domain walls become less distinct, and the overall piezoelectric response is significantly reduced, ultimately evolving into the dynamic polar nanoregions characteristic of typical relaxor behavior.

17. Conclusion: Avoid repeating results; focus on broader implications (e.g., for MLCCs with Ni/Cu electrodes).

A: We sincerely appreciate your comments regarding the conclusion section. In the revised manuscript, we have avoided repeating the results and instead focused on the broader implications. The detailed content is as follows:

In this work, a series of Ho³⁺-doped PLZT ceramics (H1-H4) were fabricated via medium-temperature solid-state sintering, and their structural evolution and energy-storage properties were systematically studied. Through the relaxor phase transition induced by Ho³⁺ amphoteric doping, H4 delivered outstanding energy-storage performance (W_rec = 0.91 J/cm3, η = 87%), along with a power density of 28.8 MW/cm³ and an ultrafast t_0.9 of 4.97 ns at 180 kV/cm. These results demonstrate that the synergistic strategy of medium-temperature sintering and Ho³⁺ amphoteric doping not only effectively mitigates Pb volatilization but also enhances processing compatibility with low-cost internal electrodes such as Ni and Cu, thereby offering a promising material platform for high-performance multilayer ceramic capacitors (MLCCs) and advanced pulsed power systems.

18. References: Some are outdated (e.g., [8] from 2007); add recent reviews on relaxor ferroelectrics. Check formatting (e.g., inconsistent journal abbreviations).

A: We sincerely appreciate your valuable suggestions regarding the references. We fully agree that citing the latest relevant studies and maintaining a consistent formatting style are fundamental requirements of academic writing. We have updated the references and standardized their formatting. The detailed content is as follows:

Page 11. Line 340-342 have been modified:

[11] Jain, A.; Wang, Y.G.; Guo, H. Emergence of Relaxor Behavior along with Enhancement in Energy Storage Performance in Light Rare-Earth Doped Ba_0.90Ca0.10Ti_0.90Zr_0.10O₃ Ceramics. Ceramics International 2021, 47, 10590–10602, doi:10.1016/j.ceramint.2020.12.171.

19. Typos and Clarity:

Page 1, line 20: "7.4 MW/cm3" → "7.4 MW/cm³" (superscript).

Page 3, line 111: "Pb2+ (1.20 Å )" → "Pb²⁺ (1.20 Å)".

Page 5, line 159: "8.4 µC/cm2" → "8.4 µC/cm²".

Improve sentences like "This work not only offers a viable strategy... but also elucidates..." for better flow.

A: We sincerely appreciate your thorough review and your corrections regarding spelling and language issues in our manuscript. We have carefully proofread and polished the entire text.

In this revising chance, we revised the manuscript over and over again, besides all revisions shown above, there are other modifications, which are shown as below:

1. We confirmed the presence of Ho element using X-ray fluorescence spectroscopy (XRF), the detailed results are as follows:

 

 

The spectra and corresponding peak identification are shown in the figure above. All the designed constituent elements, including Pb, Zr, Ti, La, Mg, Mo, as well as the critical dopant element Ho, can be clearly identified by their characteristic fluorescence peaks. In particular, the presence of Ho-LA peaks located at 53° and 57° directly confirms the successful incorporation of Ho into the ceramic matrix. The XRF results provide a solid experimental basis for our subsequent discussion based on chemical stoichiometry.

2. We have promptly corrected the y-axis labeling in the SEM inset, as shown below:

 

We really hope that the revisions and responses are made the paper better now, and we will not hesitate to revise the paper if there are any farther comments later.

Thanks again for the reviewer’s precious comments and careful corrections.

Sincerely yours,

Qingwei Liao

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript «Amphoteric Doping Effect of Ho3+ on the Performance of Medium-Temperature-Sintered PLZT Energy Storage Ceramics» is devoted to the creation of novel (Pb, La, Ho, Zr, Ti)O3 (PLZTH) ceramics with enhanced physical properties, such as recoverable energy storage density, energy efficiency, power density and discharge time. This theme may be of interest to a wide range of Coatings readers. Nevertheless, in my opinion, the experimental data and discussion presented do not fully prove the achievement of the stated results and should be strongly revised (a major revision).

• There are no results of Energy-dispersive X-ray spectroscopy (EDX) confirming the composition, intended ratio of elements in the products, as well as the presence of holmium in them.
• The experimental section mentions the use of Mg0.8Mo0.2O1.4 composite oxide in the sintering process. Does its presence affect the purity of the resulting product? How did the authors confirm this?
• The authors did not provide convincing evidence of holmium inclusion in the PLZT structure. The presented discussion of the X-ray phase analysis results does not prove the incorporation of holmium into the perovskite structure. As a minimum, a full-profile analysis of the diffraction patterns by the Rietveld method is necessary to clarify the occupancy of the positions. It is also confusing that the perovskite structure card number 89-1277 indicated by the authors does not exist in the PDF database.
• The sentence (line 119) “All samples exhibit a dense and homogeneous microstructure without discernible pores or secondary phase precipitates” is not entirely correct. The absence of a second phase according to scanning electron microscopy data can be concluded from images obtained only in the backscattered electron mode. It should also be noted that the numbering on the y-axis in the particle size distribution graphs (figure 2) does not correspond to percentages as stated in the axis caption.
• The authors have obtained a sample H4 with good physical characteristics. Should we expect their growth with a further increase in the degree of holmium doping (x> 0.04)? How do the obtained absolute values of recoverable energy storage density, high energy efficiency, power density and discharge time compare with the literature data? How is the obtained sample H4 inferior or superior to its analogs among perovskite ceramics?

Author Response

First, we would like to express our thanks to the Reviewers for their instructive comments concerning our manuscript entitled ‘Amphoteric Doping Effect of Ho³⁺ on the Performance of Medium-Temperature-Sintered PLZT Energy Storage Ceramics’(coatings-3847716). We have studied the comments carefully, and then revised the whole manuscript. The revisions/explanation corresponding to comment are shown in the following (Reviewers' comments are in italic font):

Reviewer #2:

1. There are no results of Energy-dispersive X-ray spectroscopy (EDX) confirming the composition, intended ratio of elements in the products, as well as the presence of holmium in them.

A: We sincerely appreciate your valuable suggestion regarding the confirmation of chemical composition. We fully agree that directly verifying the chemical composition of our samples and the successful incorporation of Ho³⁺ is of great importance. Although EDX measurements were not performed, we employed another characterization technique with similar functionality and, in certain aspects, higher sensitivity—X-ray fluorescence spectroscopy (XRF)—to address the concern you raised. The detailed results are as follows:

 

 

The spectra and corresponding peak identification are shown in the figure above. All the designed constituent elements, including Pb, Zr, Ti, La, Mg, Mo, as well as the critical dopant element Ho, can be clearly identified by their characteristic fluorescence peaks. In particular, the presence of Ho-LA peaks located at 53° and 57° directly confirms the successful incorporation of Ho into the ceramic matrix. The XRF results provide a solid experimental basis for our subsequent discussion based on chemical stoichiometry.

In addition, our XPS analysis and EDS mapping images further confirm the presence of Ho³⁺ in the ceramics.

2. The experimental section mentions the use of Mg₈Mo_0.2O_1.4 composite oxide in the sintering process. Does its presence affect the purity of the resulting product? How did the authors confirm this?

A: Thank you for raising the issue of possible impurities caused by sintering aids, no Mg/Mo-related secondary phase peaks were observed in the XRD patterns. Furthermore, EDS mapping performed in both grain interiors and grain boundary regions revealed no Mg or Mo enrichment or continuous phases, with their signals being close to the noise level. Therefore, it can be reasonably concluded that Mg_0.8Mo_0.2O_1.4 acted merely as a flux during the sintering process and did not form any detectable residual phase in the final ceramics. The corresponding EDS mapping images of Mg and Mo are provided below:

3. The authors did not provide convincing evidence of holmium inclusion in the PLZT structure. The presented discussion of the X-ray phase analysis results does not prove the incorporation of holmium into the perovskite structure. As a minimum, a full-profile analysis of the diffraction patterns by the Rietveld method is necessary to clarify the occupancy of the positions. It is also confusing that the perovskite structure card number 89-1277 indicated by the authors does not exist in the PDF database.

A: We sincerely appreciate your valuable comments, and we have added EDS mapping analysis. The results confirm that the Ho element is uniformly distributed throughout the ceramic microstructure, with no evidence of segregation or formation of Ho-rich secondary phases. This provides the most direct evidence for the successful solid solution of Ho³⁺, as detailed below:

Page 5. Line 143-155 have been added:

 

Figure 3. (a) SEM image of a selected area on the surface of H4 ceramic, and (b-f) the corresponding energy-dispersive X-ray spectroscopy (EDS) elemental mapping images for Pb, La, Zr, Ti, and Ho, respectively.

An energy-dispersive X-ray spectroscopy (EDS) mapping analysis was performed on H4 to evaluate the chemical homogeneity and the spatial distribution of the doped elements (Figure 3). The SEM image of the analyzed region is presented in Figure 3(a). The corresponding elemental mapping results (Figure 3(b-e)) reveal that the major constituent elements, including Pb, La, Zr, and Ti, are uniformly distributed within the microstructure. As shown in Figure 3(f), although Ho is present only as a trace dopant, its signal remains detectable and exhibits a homogeneous distribution similar to that of the primary elements. Importantly, no evidence of Ho enrichment or local segregation was found across the entire scanned area.

In addition, we have completely removed the incorrect card number and performed XRD Rietveld refinement, which revealed a decrease in the tetragonality (c/a ratio), as detailed below:

Page 4. Line 103-106 and 126-128 have been added:

 

Figure 1. (a) XRD patterns of H1-H4 ceramics. Enlarged patterns of H1-H4 ceramics in the regions of (b) 29.5°-31.5°, (c) 37°-38.5° and (d) 43°-44.5°. (e) Rietveld-refined XRD patterns of H1. (f) The lattice constants a, c and the tetragonal ratio c/a as a function of Ho³⁺ content.

Figure 1(e) displays the representative Rietveld refinement of H1, and the refined lattice parameters are summarized in Figure 1(f). The tetragonality (c/a ratio) decreases pro-gressively with increasing Ho³⁺ content.

Meanwhile, our XPS analysis provided a comprehensive investigation of the samples and clarified the site occupancy, as detailed below:

Page 7. Line 180-203 have been added:

 

Figure 5. XPS spectra of (a-c) H1 and (d-f) H4 ceramics.

The X-ray photoelectron spectroscopy (XPS) results of H1 and H4 are presented in Figure 5. As shown in Figure 5(a) and 5(d), the Ho 4d spectra of both H1 and H4 were deconvoluted into two spin-orbit components, Ho 4d_5/2 near 161 eV and Ho 4d_3/2 near 166 eV. These binding energies are in good agreement with the reference values of Ho³⁺ in Ho₂O₃, confirming that the incorporated holmium ions predominantly exist in the ceramics in a stable +3 oxidation state.[22,23] In addition, the Pb 4f spectra of H1 and H4 (Figure 5(b) and 5(e)) exhibit the characteristic Pb²⁺ spin-orbit doublet (Pb 4f_7/2 ≈ 138 eV, Pb 4f_5/2 ≈ 143 eV), without noticeable chemical shifts between H1 and H4, suggesting that the oxidation state and chemical environment of the A-site cations remain unchanged.

The O 1s spectra (Figure 5(c) and 5(f)) can be fitted into three components. The low-binding-energy peak (~529.5 eV) is assigned to lattice oxygen (O^2-) in the perovskite framework, the medium-binding-energy peak (~531.0 eV) is associated with oxygen atoms adjacent to oxygen vacancies (O_vac), and the high-binding-energy peak (~532.2 eV) is attributed to adsorbed oxygen (O_ads) at grain boundaries.[24] Importantly, the relative fraction of the O_vac peak increases from 66.12% in H1 to 70.31% in H4, indicating a significant rise in oxygen vacancy concentration with increasing Ho³⁺ content. According to defect chemistry, substitution of Ho³⁺ for A-site Pb²⁺ (donor doping) is primarily compensated by lead vacancies ( ) and should not generate substantial oxygen vacancies. In contrast, substitution of Ho³⁺ for B-site Zr⁴⁺/Ti⁴⁺ (acceptor doping) requires charge compensation through the formation of oxygen vacancies ( ). Therefore, the increased oxygen vacancy fraction in highly doped samples provides strong evidence that a portion of the Ho³⁺ ions occupy the B-site.

4. The sentence (line 119) “All samples exhibit a dense and homogeneous microstructure without discernible pores or secondary phase precipitates” is not entirely correct. The absence of a second phase according to scanning electron microscopy data can be concluded from images obtained only in the backscattered electron mode. It should also be noted that the numbering on the y-axis in the particle size distribution graphs (figure 2) does not correspond to percentages as stated in the axis caption.

A: We sincerely appreciate your valuable suggestion regarding the confirmation of composition. We fully agree with your point that SEM is mainly used to characterize the micro-morphology and densification, and indeed cannot provide a definitive conclusion about the presence of secondary phases. The backscattered electron (BSE) mode is the ideal tool for evaluating compositional contrast. We apologize for the lack of rigor in our initial description. We have revised the statements related to microstructural analysis in the manuscript. In addition, we also thank you for pointing out an error in the description of Figure 2, and we have promptly corrected the y-axis labeling. The detailed content is as follows:

Page 5. Line 130-142 have been modified:

 

Figure 2. The SEM images of H1-H4 ceramics. The insets show the grain size distribution.

The cross-sectional SEM images and grain size distributions of H1-H4 ceramics are presented in Figure 2(a-d). All samples consist of densely packed grains with clear boundaries, indicating high sintering density. The average grain sizes of H1, H2, H3, and H4 are 0.625 μm, 0.573 μm, 0.566 μm, and 0.549 μm, respectively, showing a systematic decrease with increasing Ho³⁺ content. Such grain refinement is attributed to the effective suppression of abnormal grain growth by the medium-temperature sintering process, together with the grain-boundary migration pinning effect induced by Ho³⁺ doping.[17] The reduced grain size lowers the probability of large-scale defect formation and restricts the mean free path of charge carriers, thereby suppressing avalanche breakdown under high electric fields. Consequently, the breakdown strength (E_b) is enhanced, enabling the ceramics to sustain higher operating fields and resulting in an increased W_rec.[18,19]

5. The authors have obtained a sample H4 with good physical characteristics. Should we expect their growth with a further increase in the degree of holmium doping (x> 0.04)? How do the obtained absolute values of recoverable energy storage density, high energy efficiency, power density and discharge time compare with the literature data? How is the obtained sample H4 inferior or superior to its analogs among perovskite ceramics?

A: We sincerely appreciate your valuable comments. We have now included the data of H4 directly in the main text. In addition, we noticed that the inset in the original Figure 5(c) was relatively small; it has now been enlarged for better readability, and the value of t_0.9 obtained from the inset has been corrected accordingly. The detailed modifications are as follows:

Page 9. Line 238-272 have been modified:

 

Figure 8. (a) The underdamped discharge current curves of H1. (b) The overdamped discharge current curves of H1. (c) The variation of discharge energy density of H1 with time. (d) The C_D and P_D of H1 under different electric fields. (e-h) The corresponding pulsed discharge performances of H4.

To assess the suitability of PLZTH ceramics sintered at intermediate temperatures for practical pulsed power applications, pulsed charge-discharge measurements were performed on H1 and H4. In the underdamped mode (Figure 8(a) and 8(e)), both samples exhibited a linear increase in the first peak current density with the applied field, and the discharge process was completed after 2-4 oscillations. Such oscillatory behavior originates from the intrinsic RLC characteristics of the testing circuit, where the oscillation period follows . Here, L represents the equivalent inductance of the circuit and C is the capacitance of the sample; since all measurements were carried out on the same platform, L can be regarded as constant. The similarity of oscillation profiles at different fields indicates that the capacitance of the ceramics remains stable during nanosecond-scale discharge. Beyond W_rec, current density (C_D) and P_D are also critical metrics for evaluating pulsed capacitor dielectrics.[34,35] As shown in Figure 8(d) and 8(h), both C_D and P_D increase monotonically with the applied field, reaching 148.1 A/cm² and 7.4 MW/cm³ for H1 at 100 kV/cm and 320.3 A/cm² and 28.8 MW/cm³ for H4 at 180 kV/cm.

In the overdamped mode with a load resistance of 205 Ω (Figure 8(b) and 8(f)), all discharge current traces exhibited a rapid rise followed by exponential decay. The peak discharge currents increased progressively with the applied field, reaching 6.9 A for H1 and 8.5 A for H4. Integration of the current-time profiles yielded the discharge energy density (W_d), as shown in Figure 8(c) and 8(g). Wd increased linearly with field strength, attaining 0.18 J/cm³ for H1 at 100 kV/cm and 0.43 J/cm³ for H4 at 180 kV/cm. These values correspond to 75% and 73% of the recoverable energy densities obtained from the quasi-static P-E loops (H1: 0.24 J/cm³; H4: 0.59 J/cm³). The slight energy loss is attributed to intrinsic dielectric dissipation under high-frequency conditions during nanosecond-scale discharge, where polar nanoregions cannot fully follow abrupt field reversals, and conduction and interfacial polarization losses convert part of the stored energy into heat.[36–38]

Throughout the entire test range, t_0.9 remained below 20 ns, with values of 18.85 ns for H1 at 100 kV/cm and as low as 4.97 ns for H4 at 180 kV/cm, demonstrating an ultrafast discharge response. The enhanced charge transport and energy release efficiency induced by Ho³⁺ doping further highlights the promise of medium-temperature sintered PLZTH ceramics as advanced dielectric candidates for pulsed power capacitors.

Furthermore, we reviewed the performance parameters of recently reported ceramics and compiled them into a performance comparison figure. A new section analyzing this comparison has been added to the main text. The detailed additions are as follows:

Page 8. Line 219-236 have been added:

Figure 7. A comparison of energy densities and efficiencies of recently reported (anti)ferroelectric-based energy storage dielectric material.

Figure 7 summarizes the reported energy storage performances of representative Pb-based (anti)ferroelectric ceramics in recent years.[25–33] As can be seen, most systems exhibit a trade-off between W_rec and η. For instance, conventional PLZT and PLBZST ceramics typically show η values in the range of 60-75%, and although some compositions achieve W_rec values approaching 1 J/cm3, their energy conversion efficiency remains limited. Conversely, systems such as PSZT and PMN-PT demonstrate relatively higher efficiencies, but their W_rec usually does not exceed 0.6 J/cm³.

In contrast, the PLZTH ceramics developed in this work achieve a W_rec of 0.91 J/cm³ and an η of 87% at a thickness of 0.2-0.3 mm, thereby striking a more favorable balance between energy density and efficiency. Compared with other reported PLZT-based ceramics, PLZTH maintains a moderate energy storage density while exhibiting a remarkable advantage in energy conversion efficiency. This indicates that Ho³⁺ doping not only sustains a relatively high W_rec but also effectively suppresses energy dissipation during polarization switching, leading to reduced energy loss under fast charge-discharge conditions.

In this revising chance, we revised the manuscript over and over again, besides all revisions shown above, there are other modifications, which are shown as below:

1. We have refined the description of the experimental procedure, and the revised content is as follows:

Page 2. Line 85-97 have been modified:

Raman spectra were acquired using a micro-Raman spectrometer (LabRAM HR Evolu-tion, Horiba Scientific, Japan) with the excitation source of 532 nm. P-E hysteresis loops were characterized employing a ferroelectric analyzer (TF2000, aixACCT, Germany). For this purpose, ceramic disks with a thickness of 0.2–0.3 mm were prepared, in which a top electrode of silver paste with a diameter of 3 mm was deposited, while the bottom side was coated with a larger 6 mm electrode to act as the ground contact. The practical charge-discharge performances of the ceramics were assessed on a commercial testing platform (CFD-001, Gogo Instruments Technology, Shanghai, China). In this case, thinner specimens with a thickness of ~0.14 mm were employed, and symmetrical silver electrodes with a diameter of 3 mm were applied to both surfaces. X-ray photoelectron spectroscopy (XPS) measurements were performed using a spectrometer(Thermo Fisher, Thermo Scientific K-Alpha, USA).

2. We have expanded the Introduction by comparing PLZT with other perovskite systems such as BT and BNT, providing a more detailed justification for choosing PLZT as the research platform and highlighting its unique advantages in terms of high polarization and tunable phase transitions. The detailed content is as follows:

Page 1. Line 29-41 have been modified:

Within the broad family of perovskite oxides, lead-free ceramics such as barium titanate (BT), sodium bismuth titanate (BNT), silver niobate (AgNbO₃), and sodium niobate (NaNbO₃) have been intensively investigated owing to their environmental benignity. However, their practical application is generally constrained by intrinsically low breakdown strengths and relatively broad P-E loops, which in turn limit both the re-coverable energy density and storage efficiency.[3–5] By contrast, lanthanum-modified lead zirconate titanate (PLZT) ceramics combine outstanding ferroelectric/antiferroelectric properties with a high dielectric permittivity, and their phase transition characteristics can be flexibly tailored through rare-earth substitution or Zr/Ti ratio adjustment. Such tunability enables the realization of slim P-E loops together with high energy conversion efficiencies, rendering PLZT one of the most competitive and scalable candidates for advanced dielectric energy-storage applications.[6,7]

3. In the revised manuscript, we have avoided repeating the results and instead focused on the broader implications. The detailed content is as follows:

In this work, a series of Ho³⁺-doped PLZT ceramics (H1-H4) were fabricated via medium-temperature solid-state sintering, and their structural evolution and energy-storage properties were systematically studied. Through the relaxor phase transition induced by Ho³⁺ amphoteric doping, H4 delivered outstanding energy-storage performance (W_rec = 0.91 J/cm3, η = 87%), along with a power density of 28.8 MW/cm³ and an ultrafast t_0.9 of 4.97 ns at 180 kV/cm. These results demonstrate that the synergistic strategy of medium-temperature sintering and Ho³⁺ amphoteric doping not only effectively mitigates Pb volatilization but also enhances processing compatibility with low-cost internal electrodes such as Ni and Cu, thereby offering a promising material platform for high-performance multilayer ceramic capacitors (MLCCs) and advanced pulsed power systems.

4. In the revised manuscript, all instances of “unipolar” in the figure caption of Figure 6 and the corresponding text have been corrected to “bipolar.” Additionally, we have explicitly included the values of remanent polarization (P_r) and coercive field (E_c) extracted from the P-E loops in the main text. The detailed content is as follows:

Page 8. Line 206-211 have been modified:

Figure 6. (a) Bipolar P-E loop of H1-H4 ceramics. (b) The calculated Wrec and η of H1-H4 ceramics.

At the breakdown electric field and a frequency of 10 Hz, the bipolar P-E loops of H1-H4 ceramics are shown in Figure 6(a). Unlike the double hysteresis loops or broad hysteretic loops typically observed in long-range-ordered materials, all samples exhibit extremely slim P-E loops with low remanent polarization (Pr) and coercive field (E_c)(H4: P_r = 0.4 μC/cm², E_c = 7.9 kV/cm).

5. We have updated the references and standardized their formatting. An example is shown below:

Page 11. Line 340-342 have been modified:

[11] Jain, A.; Wang, Y.G.; Guo, H. Emergence of Relaxor Behavior along with Enhancement in Energy Storage Performance in Light Rare-Earth Doped Ba_0.90Ca0.10Ti_0.90Zr_0.10O₃ Ceramics. Ceramics International 2021, 47, 10590–10602, doi:10.1016/j.ceramint.2020.12.171.

6. We have also discussed the circuit parameters. The detailed content is as follows:

Page 9. Line 246-250 have been added:

Such oscillatory behavior originates from the intrinsic RLC characteristics of the testing circuit, where the oscillation period follows . Here, L represents the equivalent inductance of the circuit and C is the capacitance of the sample; since all measurements were carried out on the same platform, L can be regarded as constant.

We really hope that the revisions and responses are made the paper better now, and we will not hesitate to revise the paper if there are any farther comments later.

Thanks again for the reviewer’s precious comments and careful corrections.

Sincerely yours,

Qingwei Liao

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I recommend the publication of this article in its current form

Author Response

Thanks for your positive comment.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have provided a detailed response to the comments, however, some of the conclusions presented were not convincing to me. The EDS mapping clearly demonstrates the presence of a magnesium-rich phase in the sample, which contradicts the authors' conclusions. It is necessary to provide R-factors for the results of the full-profile analysis of all samples, as well as the errors in the values ​​of the unit cell parameters. It seems that the authors' description of the differences in the diffraction patterns may be due not to structural ordering, but to the presence of an impurity. It is unclear which structure (card number) was used for the Rietveld refinement. It is necessary to provide the results of the Rietveld refinement of all samples (confirmation of the occupancy of positions and the coordinates of the atoms).

Author Response

Dear Editor,

First, we would like to express our thanks to the Reviewers for their instructive comments concerning our manuscript entitled ‘Amphoteric Doping Effect of Ho³⁺ on the Performance of Medium-Temperature-Sintered PLZT Energy Storage Ceramics’(coatings-3847716). We have studied the comments carefully, and then revised the whole manuscript. The revisions/explanation corresponding to comment are shown in the attached file.

Best Wishes,

Qingwei Liao

Author Response File: Author Response.pdf