Pressure-Induced Reversible Local Structural Disorder in Superconducting AuAgTe₄

Round 1

Reviewer 1 Report

In this article, the authors studied the properties of phonons of AuAgTe4 under different pressures by Raman spectroscopy and the first-principle calculations. The research method combining theory and experiment is very systematic, but there are some problems that need to be addressed before publication.

 

1. On page 2, “The Raman experiment was carried out using natural AuAgTe4 crystals from the Kochbulak deposit, Kuraminsky Range, Uzbekistan.”, does the ratio of Au and Ag atoms strictly 1:1 in the samples? Or is there a certain error? If there is a certain error, I hope to see the specific error value measured in the article.

2. The author should add at least one schematic diagram of the crystal structure, which can help readers understand the structure of the material of this article most quickly and intuitively.

3. At the end of page 2, “The cut-off energy was chosen to be 500 eV and the mesh of 4 × 4 × 2 was used for integration over the Brillouin zone according to Monkhorst-Pack scheme [20].”, have the authors tested the convergence of the cutoff energy and k-mesh? The 4 × 4 × 2 grid seems to be small while the non-convergent parameters may cause the calculated results to be inaccurate.

4. At the end of page 4, “Unfortunately, the data on thermal expansion coefficient α(T) are unknown for the AuAgTe4 crystal.”, the thermal expansion coefficient can be estimated by using VASP, can it be used for more accurate calculations?

5. At the end of page 5, “Raman scattering experiments on three different samples under pressure have been performed up to 8 GPa, 25 GPa, and 9 GPa.”, this sentence is a bit confusing. Do the three pressures act on three samples respectively, or all three pressures have been acted on each sample? And why these three pressure values were chosen. (The order of the numbers is also a bit strange, why not list them from small to large, which is more intuitive.)

6. There are some mistakes in typing and format in the article. (1) Figure 3. “_el=150 cm−1 -dashed blue.”, do you mean “Γ_el” and “dashed red”? (2) Table 3. “aclculated” → “Calculated”? (3) At the end of page 4, “... according to the expression (3) [23]:”, which may be “expression (2)”? (4) It can’t be ruled out that there may be some undiscovered mistakes in the article, so the author should proofread the article carefully.

Author Response

1.Microanalysis of 10 small crystals AuAgTe4 showed maximum deviations of +_0.05 for all elements, the average composition turned out to be AuAgTe3.96

2.The figure was added

3.We verified the dependence of the total energy on k- grid. We have carried out several self-consistent calculations for the AuAgTe4 structure (0 and 10 GPa) using 4 × 4 × 2 and 8 × 8 × 4 grid, and obtained the following total energies: 37.79 and 37.79 eV respectively for structure at ambient pressure; 18.23 and 18.22 eV respectively for structure at 10 GPa. So, the energy saturated and the chosen k-grids (4 × 4 × 2) are sufficient for the present calculations.

4. Our main interest was to study the temperature behavior of the phonon linewidths, since electron induced part of them contributes to the electron-phonon coupling. Phonon frequencies clearly demonstrate anharmonic behavior, but in this context, the separation into the quasiharmonic and phonon-phonon mechanisms has not been the subject of study.

5.In the first experiment, after fixing the phase transition, we unexpectedly found a strong change in the spectrum in the region of 8 GPa, after which we unloaded the sample to its initial state. Then we repeated the experiment on another sample, significantly expanding the pressure range to 25 GPa, and then unloaded the sample again. The same experiment on a third sample up to 9 GPa was performed to ensure that the results obtained were reproducible.

6. You are quite right, we corrected found mistakes.

Reviewer 2 Report

 

 

The Authors present investigation of the phonon spectrum in mineral sylvanite in a wide range of temperatures and pressures by means of low-frequency Raman spectroscopy and supported by periodic DFT calculations. The motivation was to shed light on the mechanism of electron pairing in coinage metal tellurides in general and relate it to superconductivity observed for such tellurides at high pressure.

The manuscript contains interesting experimental observations, including Raman spectra in wide temperature range and at pressures up to 25 GPa and is well within the scope of this journal. Observation of possible displacive or order-disorder phase transition at above 8 GPa is certainly interesting and should inspire further study.

I recommend publication after a few of my concerns, listed below, have been carefully addressed.

 

Figure 2 : Represents clear trends and excellent fits to peak width for selected presumably experimental frequencies; are the trends similar for all tracked frequences?

Could they perhaps be added in the form of supporting information? The paper would certainly benefit from a thorough Supporting Information.

 

However, I find it hard to connect data from Figure 2 with Table 1. Oscillation mode Ag¹ in Figure 2 appears to be the first Ag mode in Table 1 (~47cm^-1 at 300K), but following the same logic, the second plot in Figure 2, marked Ag⁵, should correspond to the 5-th Ag mode in the Table 1, so to the Frequency #10, but here the experimental value at 300K is reported as 121 cm^-1, whereas the plot gives value distinctly above 122, and in the case of the third plot, assuming it represents frequency #14, the value is likewise higher in the plot than in the Table.

I would suggest assigning the exact same names (e.g.: names of the irreducible representations) for the frequencies in the Table 1 and in Figure 2 and double checking the values reported in that Table.

Table 1 would also clearly benefit from reporting the experimental frequencies with uncertainties. This would add the value to the discussion of how close the calculated frequencies actually are to the experimental ones.

 

I do not like the statement in line 116 : ‘the frequencies soften’; oscillation modes can soften and the experiementally observed result is the Raman peak broademing, but the frequency strictly speaking refers to the value at the maximum of such peak. Please rephrase to be more precise.

 

Table 2 states ‘Calculated and experimentally found at 300 K frequencies...’ but only one set of values and fit parameters are presented. The frequency values are presumably experimental, but do not fit exactly the values from Table 1. Please clarify and correct the problem.

 

Lines 159-160 ‘due to the presence scattering of carriers by impurities or phonons’ : the sentence is unintelligible, please clarify

 

Figure 3 : 3-rd line – what does the symbol _el stand for?

 

Table 3 Please correct the header for typos. Please add uncertainties to experimental values.

 

Similarly to the situation in Table 1, the ordering of the experimental and theoretical frequencies differ in Table 3, this time quite prominently. I assume the Authors assigned the Ag / Bg symmetry for experimental frequencies based on the polarization, but they state themselves that the actual orientation of the crystalline samples under pressure was unknown, so the assignment could be faulty. The Authors should comment on that.

 

Lines 224 – 225 could be illustrated by a graphical overlay of the predicted frequencies on Figure 6.

 

Paragraph 226 – 244 : disorder – yes, total amorphization - certainly not, sample once completely amorphized by non-hydrostatic medium would not revert to crystallinity upon decompresion; the Authors quatations concerning silica a.s.o are quite accurate; However, the strain exerted by non-hydrostatic meadium usually leads to much more peak broadening than expected from hydrostatic models, hence the frequency lines initially observed in 100 – 200 cm^-1 range are smeared to the point of being drowned in background.

 

Literature references are up to date.

 

Some typos should be removed from the References, e.g.: triple dot to be removed from line 318.

 

 

 

 

 

Author Response

Figure 2 : Represents clear trends and excellent fits to peak width for selected presumably experimental frequencies; are the trends similar for all tracked frequences?

Could they perhaps be added in the form of supporting information? The paper would certainly benefit from a thorough Supporting Information.

The trends shown in Fig. 2 were observed for all phonons that we could investigate in temperature measurements.

However, I find it hard to connect data from Figure 2 with Table 1. Oscillation mode Ag1 in Figure 2 appears to be the first Ag mode in Table 1 (~47cm-1 at 300K), but following the same logic, the second plot in Figure 2, marked Ag5, should correspond to the 5-th Ag mode in the Table 1, so to the Frequency #10, but here the experimental value at 300K is reported as 121 cm-1, whereas the plot gives value distinctly above 122, and in the case of the third plot, assuming it represents frequency #14, the value is likewise higher in the plot than in the Table.

I would suggest assigning the exact same names (e.g.: names of the irreducible representations) for the frequencies in the Table 1 and in Figure 2 and double checking the values reported in that Table.

Table 1 would also clearly benefit from reporting the experimental frequencies with uncertainties. This would add the value to the discussion of how close the calculated frequencies actually are to the experimental ones.

 We thank the referee for this suggestion and made changes to this table .

I do not like the statement in line 116 : ‘the frequencies soften’; oscillation modes can soften and the experiementally observed result is the Raman peak broademing, but the frequency strictly speaking refers to the value at the maximum of such peak. Please rephrase to be more precise.

Thank you much - we changed this sentence” With increasing temperature the frequencies of almost all lines in the spectra soften, and the linewidths increase.”

Table 2 states ‘Calculated and experimentally found at 300 K frequencies...’ but only one set of values and fit parameters are presented. The frequency values are presumably experimental, but do not fit exactly the values from Table 1. Please clarify and correct the problem.

 Changes have been made to table 2.

Lines 159-160 ‘due to the presence scattering of carriers by impurities or phonons’ : the sentence is unintelligible, please clarify

The meaning of the phrase is that the scattering of electrons by impurities or lattice vibrations induces the electronic transitions that can be observed in Raman scattering.

Figure 3 : 3-rd line – what does the symbol _el stand for?

Г and Г_el were used for phonon and electron damping

Table 3 Please correct the header for typos. Please add uncertainties to experimental values.

Changes have been made to table 3

Similarly to the situation in Table 1, the ordering of the experimental and theoretical frequencies differ in Table 3, this time quite prominently. I assume the Authors assigned the Ag / Bg symmetry for experimental frequencies based on the polarization, but they state themselves that the actual orientation of the crystalline samples under pressure was unknown, so the assignment could be faulty. The Authors should comment on that.

Changes have been made to tables 1-3. We know the phonon symmetries for both phases. The most intense Ag lines can be observed in both polarized and depolarized geometries in both phases, and all lines are identified for the LP phase. In the case of the HP phase, we were unable to perform measurements on different planes of the crystal, so we did not obtain an accurate identification of the observed lines. For example, the most high-frequency line of the LP phase at 160 cm-1 was observed in the polarized spectrum, but its intensity decreased significantly in the region of the structural transition, while it acquired a significant intensity in the HP phase. Possibly, in the HP phase at this frequency we already observe a phonon of Bg symmetry.

Lines 224 – 225 could be illustrated by a graphical overlay of the predicted frequencies on Figure 6.

The peak frequencies are now labeled in the figure.

Paragraph 226 – 244 : disorder – yes, total amorphization - certainly not, sample once completely amorphized by non-hydrostatic medium would not revert to crystallinity upon decompresion; the Authors quatations concerning silica a.s.o are quite accurate; However, the strain exerted by non-hydrostatic meadium usually leads to much more peak broadening than expected from hydrostatic models, hence the frequency lines initially observed in 100 – 200 cm-1 range are smeared to the point of being drowned in background.

We fully agree with the referee and believe that in our case there are signs of a structural disorder.

Literature references are up to date.

Some typos should be removed from the References, e.g.: triple dot to be removed from line 318.

Changed

Reviewer 3 Report

 

The manuscript describes lattice dynamics of sylvanite at high-pressure high-temperature conditions. The investigation was performed using Raman spectroscopy. The experimental results were compared to theoretical data obtained by first-principle calculations.

 

1.     It seems that for structures based on Krennerite, other Au:Ag ratios than 1:1 are possible (e.g., Ag0.1875 Au0.8125 Te2). Is AuAgTe4 a line phase or was a range of compositions detected? Could you please provide the results of composition and structure analysis of the investigated samples as Supporting Information?

2.      Is the arrangement of Au and Ag always ordered? Was a partial replacement Au/Ag either on Wyckoff site 2a or 2e or even on both sites observed? Will this result in different distortions of the coordination polyhedra around Au and Ag and in turn in an alteration of the Te sublattice?

3.      Please consider using another method (single crystal or powder X-ray diffraction, transmission electron microscopy) for the structural analysis of the samples. The specified analysis by SEM/EDXS and EPMA can only account for the composition and morphology, not for structural changes as you wrote. A thorough initial analysis of the samples is necessary to understand the occurring changes.

4.      Can you please add a table summarizing which peaks occurred at which pressure?

5.      Is KCl inert towards your samples, also at high pressures and high temperatures?

6.      Regarding the planes the measurements were carried out on: are those naturally occurring facets or cleavage planes of the crystals? Can you please specify this in the text? How was the orientation accounted for?

7.      How does the local structural disorder occurring at higher pressures compare to what was observed, e.g. in AuTe2 or other tellurides? Please include this in the discussion.

 

Author Response

1.It seems that for structures based on Krennerite, other Au:Ag ratios than 1:1 are possible (e.g., Ag0.1875 Au0.8125 Te2). Is AuAgTe4 a line phase or was a range of compositions detected? Could you please provide the results of composition and structure analysis of the investigated samples as Supporting Information?

Microanalysis of 10 small crystals AuAgTe4 showed maximum deviations of +_0.05 for all elements, the average composition turned out to be AuAgTe3.96. Detailed structural analysis of one AuAgTe4 grain was performed using the SEM-EBSD method at room temperature prior to the pressure experiments. Results were added to text.

2.Is the arrangement of Au and Ag always ordered? Was a partial replacement Au/Ag either on Wyckoff site 2a or 2e or even on both sites observed? Will this result in different distortions of the coordination polyhedra around Au and Ag and in turn in an alteration of the Te sublattice?

Structural investigations indicate that random distributed Ag on the Au sites suppresses the valence fluctuation of Au and, therefore, the structure modulations which consist mainly of displacements of Te atoms. (L. Bindi et al., American Mineralogist, Volume 94, pages 728–736, 2009). Such studies have not been carried out on our samples.

3.Please consider using another method (single crystal or powder X-ray diffraction, transmission electron microscopy) for the structural analysis of the samples. The specified analysis by SEM/EDXS and EPMA can only account for the composition and morphology, not for structural changes as you wrote. A thorough initial analysis of the samples is necessary to understand the occurring changes.

The authors (Y. Amiel et al., https://arxiv.org/abs/2301.08033 -ref.15) presented detailed X-ray diffraction experiment on a single crystal AuAgTe4 from the classical Transylvania locality. Additional diffraction measurements on our samples confirmed obtained structural results in the pressure range up to 10 GPa.

4. Can you please add a table summarizing which peaks occurred at which pressure?

In reality, we see the appearance of only two (113 and 142 cm-1) new peaks in the HP phase, so there is no point in making such a table. The remaining peaks of the HP phase inherit the peaks of the LP phase.

5.Is KCl inert towards your samples, also at high pressures and high temperatures?

The plane of the crystal on which the spectra were measured was in contact with the diamond window, and we did not see any changes both visually and in the measured spectra of both the inital sample and after the pressure was removed.

6.Regarding the planes the measurements were carried out on: are those naturally occurring facets or cleavage planes of the crystals? Can you please specify this in the text? How was the orientation accounted for?

The measurements were carried out on crystal cleavages, among which cleavages of both rhombic and rectangular surface morphology were found. It was natural to assume that cleavages of rhombic morphology contain an AC plane (see Fig.), the selection rules for measurements on which allow us to observe only Ag phonons both for parallel polarization and cross polarization of the incident and scattered light. The measurements confirmed this conclusion. Measurements on chips of rectangular morphology made it possible to observe Bg phonons in accordance with the selection rules. Thus, we were able to identify all crystal planes using only Raman data.

7.How does the local structural disorder occurring at higher pressures compare to what was observed, e.g. in AuTe2 or other tellurides? Please include this in the discussion.

The Raman spectra of AuTe2 under pressure are not presented in the literature. In As2Te3, authors  ( Dai, L. et al. J. Mater. Chem. C 2017) reported a structural phase transition, a crystalline-to-amorphous transformation and metallization by Raman spectroscopy, under both non-hydrostatic and hydrostatic conditions up to ~25 GPa. In addition, the amorphization and metallization of the arsenic telluride were irreversible under non-hydrostatic conditions but reversible under hydrostatic conditions.The effect of pressure environment up to ~25 GPa on the structural phase transition and metallization was found to be negligible for ZnTe ( Zhuang, Y. et al., Mod. Phys. Lett. B 2018 ).