Proximity-Induced Magnetism in a Topological Insulator/Half-Metallic Ferromagnetic Thin Film Heterostructure

Round 1

Reviewer 1 Report

In this manuscript by Min Zhang et al., new thin film heterostructures based on topological insulator Bi2Se3 on ferromagnetic insulator La0.7Sr0.3MnO3 and LaAlO3 (100) single crystal substrates are reported. This work continues Ref 42, there such a TI/FMI heterostructure was reported recently.

In the present study, the authors synthesized a series of heterostructures with different deposition times which resulted in different magnetic, electrical and magnetotransport properties. The extra magnetic moment was found in Bi2Se3 films due to the magnetism of the La0.7Sr0.3MnO3 layers. The results are discussed and support the conclusions. This manuscript by Min Zhang et al. can be accepted to the Special Issue 'Thin Films of Electronic Materials' because it provides new insights into the magnetic properties of TI/FMIs. The manuscript can be published as it is in Coatings. 

 

Author Response

Thank you for your work and for the reviewers’ comments concerning our manuscript. We are so glad to receive your recognition for the work of coatings-1727148.

 

Reviewer 2 Report

A report of reviewing a manuscript, Coatings-1727148

This manuscript reports (i) the preparation, (ii) morphology including crystal quality, (iii) magnetotransport characteristics, and (iv) magnetization curves of the hetero structure of Bi₂Se₃ / La_0.7Sr_0.3MnO₃ thin films grown on LaAlO₃ (100) single crystal. These measurements were carried out on several Bi₂Se₃ thicknesses at various temperatures ranging from 10 to 300 K, in order to discuss how the transport mechanisms of Bi₂Se₃ are affected by the proximity effect of the ferromagnetism in La_0.7Sr_0.3MnO₃. The authors concludes that (i) the low-temperature (< 30 K) conductance of the hetero structure is mainly determined by Bi₂Se₃ whereas the high-temperature one is determined by La_0.7Sr_0.3MnO₃, (ii) the proximity effects are classified into the short-range and the long-range types, each of which are due to the chemical bonding characters of the Bi₂Se₃/La_0.7Sr_0.3MnO₃ interface and the spin injection mechanisms, respectively, (iii) the weak anti-localized (WAL) effect is suppressed by the magnetism induced by the proximity effects.

The introduction and experimental sections of the manuscript are satisfactorily prepared but several minor deficiencies must be corrected. The result and discussion section has materials, which should be improved/added before publication in Coatings.

The comments concerning the introduction and experimental sections are:

1. In L. 2, a colon “:” is incorrectly used in the title of this manuscript.
2. In L. 51, “hall” is must be “Hall”.
3. In L. 51, the sentence starting with “Open” is incomplete.
4. In 114 and 133, a term “core level” sounds strange because “core level” means f orbital; instead, “valance level” could be better for 3d orbital case.

The comments concerning the result and discussion section are:

1. Why the authors used a term of deposition time instead of sample thickness?
2. In L. 201 (caption of figure 5), the direction of magnetic field should be indicated.
3. In L. 206, an interpretation as to type of the spin injection could be useful for readers.
4. In L. 214, an argument as to mechanism of the diffusion of the spin-polarized quasi-particles could be useful for readers.
5. In L. 223 (caption of figure 6), “magnetoresistance (MR) ratio” should be used instead of “resistivity” and the direction of magnetic field should be indicated.
6. In L. 226, “magnetoresistance” should be changed to “magnetoresistance ratio (MR)”. The definition of MR is also necessary in text.
7. In L. 240(caption of figure 7), a clear indication as to sample type investigated is necessary because only “Bi₂Se₃ films” is indicated in this caption; two type of Bi₂Se₃ films are prepared in this study: the hetero structure of Bi₂Se₃ / La₇Sr_0.3MnO₃ thin films grown on LaAlO₃ (100) single crystal and (ii) a Bi₂Se₃ film grown on LaAlO₃ (100) single crystal.
8. Concerning Figure 7(a) and (b), the magnitude of 10 K-magnetoresistance ratio (MR) of t=3 min sample indicted in this figure are approximately 1 % at 1 T and approximately 2 % at 6 T. On the other hand, the 10 K-MR of t=3 min sample indicated in Figure 6 are approximately 45 % at 6 T. These MR magnitudes look to be inconsistent.
9. Concerning Figure 7(c), how is the magnetoconductivity σ is obtained from the present magnetotransport measurements? A clear description is necessary in text.

 

Author Response

Dear reviewers:

Thank you for your work and for the reviewers’ comments concerning our manuscript. The manuscript coatings-1727148 has been carefully revised. We appreciate the detailed and useful comments and suggestions from you and reviewers. The point-by-point answers to the comments and suggestions were listed as below.

 

Reviewer #2:          

 

This manuscript reports (i) the preparation, (ii) morphology including crystal quality, (iii) magnetotransport characteristics, and (iv) magnetization curves of the hetero structure of Bi₂Se₃ / La_0.7Sr_0.3MnO₃ thin films grown on LaAlO₃ (100) single crystal. These measurements were carried out on several Bi₂Se₃ thicknesses at various temperatures ranging from 10 to 300 K, in order to discuss how the transport mechanisms of Bi₂Se₃ are affected by the proximity effect of the ferromagnetism in La_0.7Sr_0.3MnO₃. The authors concludes that (i) the low-temperature (< 30 K) conductance of the hetero structure is mainly determined by Bi₂Se₃ whereas the high-temperature one is determined by La_0.7Sr_0.3MnO₃, (ii) the proximity effects are classified into the short-range and the long-range types, each of which are due to the chemical bonding characters of the Bi₂Se₃/La_0.7Sr_0.3MnO₃ interface and the spin injection mechanisms, respectively, (iii) the weak anti-localized (WAL) effect is suppressed by the magnetism induced by the proximity effects.

The introduction and experimental sections of the manuscript are satisfactorily prepared but several minor deficiencies must be corrected. The result and discussion section has materials, which should be improved/added before publication in Coatings:

The comments concerning the introduction and experimental sections are:

1. In L. 2, a colon “:” is incorrectly used in the title of this manuscript.

Answer: The colon “:” has been removed in L. 2.  

1. In L. 51, “hall” is must be “Hall”.

Answer: “hall” has been corrected to “Hall” in L. 51.

1. In L. 51, the sentence starting with “Open” is incomplete.

Answer: In L. 51, the sentence has been deleted.

1. In 114 and 133, a term “core level” sounds strange because “core level” means f orbital; instead, “valance level” could be better for 3d orbital case.

Answer: Thanks for your suggestion. The “core level” has been substituted “valance level” in 114 and 133.

The comments concerning the result and discussion section are:

1. Why the authors used a term of deposition time instead of sample thickness?

Answer: The film thickness corresponding to each deposition time has been measured, as shown in the table 1. We used a term of deposition time instead of sample thickness only because reading is more convenient.

2. In L. 201 (caption of figure 5), the direction of magnetic field should be indicated.

Answer: In figure 4 and figure 5, the results have been obtained with the B perpendicular to the ab plane. A sentence is added to illustrate the direction of magnetic field in L. 145-146. “The electrical current was applied in the ab plane，and B perpendicular to the ab plane.”

3. In L. 206, an interpretation as to type of the spin injection could be useful for readers.

Answer: Thank you for your comments.

4. In L. 214, an argument as to mechanism of the diffusion of the spin-polarized quasi-particles could be useful for readers.

Answer: Thank you for your comments.

5. In L. 223 (caption of figure 6), “magnetoresistance (MR) ratio” should be used instead of “resistivity” and the direction of magnetic field should be indicated.

Answer: Thanks a lot for your kind suggestion. “resistivity” is corrected to “magnetoresistance (MR) ratio” in L. 223.

6. In L. 226, “magnetoresistance” should be changed to “magnetoresistance ratio (MR)”. The definition of MR is also necessary in text.

Answer: Thank you for your comments. “magnetoresistance” is changed to “magnetoresistance ratio (MR)” in L. 226.

In L. 228-229, a sentence is added to illustrate the MR: “The resistivity changed caused by the extra magnetic field, and the rate of change is called magnetoresistance.”

7. In L. 240(caption of figure 7), a clear indication as to sample type investigated is necessary because only “Bi₂Se₃films” is indicated in this caption; two type of Bi₂Se₃ films are prepared in this study: the hetero structure of Bi₂Se₃ / La₇Sr₃MnO₃ thin films grown on LaAlO₃ (100) single crystal and (ii) a Bi₂Se₃ film grown on LaAlO₃ (100) single crystal.

Answer: Thank you for your comments. In L. 242, the sample type has been indicated.

8. Concerning Figure 7(a) and (b), the magnitude of 10 K-magnetoresistance ratio (MR) of t=3 min sample indicted in this figure are approximately 1 % at 1 T and approximately 2 % at 6 T. On the other hand, the 10 K-MR of t=3 min sample indicated in Figure 6 are approximately 45 % at 6 T. These MR magnitudes look to be inconsistent.

Answer: These MR magnitudes look to be inconsistent due to the different calculation formulas. In Figure 7, the normalized magnetoresistance (MR) = ρ/ρ(B=0). Here, ρ and ρ (B=0) is the resistivity with and without magnetic field respectively. In Figure 6, the MR=（ρ_H-ρ₀）/ρ₀×100%. Here, ρ₀ and ρ_H is the resistivity without and with an extra magnetic field respectively.

9. Concerning Figure 7(c), how is the magnetoconductivity σ is obtained from the present magnetotransport measurements? A clear description is necessary in text.

Answer: The magnetoconductivity σ is obtained by the resistivity, and the results of the resistivity shown in Figure 7(a) and (b). a sentence is added to illustrate the question in L. 262.

Author Response File: Author Response.docx

Reviewer 3 Report

The electron transport and magnetic properties of a two-layer heterostructure consisting of a topological insulator layer Bi 2 Se 3 deposited on a La0.7Sr0.3MnO3 layer with spontaneous magnetization. The layers are characterized by surprisingly high crystallinity. Analyzing the temperature and field behavior of the resistance, the authors observe manifestations of charge transfer and redistribution, orbital hybridization and exchange coupling between layers, and spin injection. The authors also observe excessive magnetization of the heterostructure. The latter is associated with the appearance of spontaneous magnetization in Bi 2 Se 3 due to the magnetic proximity to La0.7Sr0.3MnO3. The discovered effect can be used to control the electronic and magnetic state of such heterostructures, which has prospects for application in spintronics and thermoelectric devices. The article can be accepted, after taking into account the following comments.

 

1. The authors refer to La 0.7 Sr 0.3 MnO 3 as a ferromagnetic insulator, suggesting this as a well-known fact. However, this compound is referred to in the literature as half-metallic ferromagnetic La0.7Sr0.3MnO3 (LSMO) [ Vila-Fungueiriño, J. M. et al. sci. Technol. Adv. mater. 19, 702–710 (2018)] [Zhang, X. et al. J. Am. Ceram. soc. 104, 955–965 (2021)] and even as on metallic ferromagnet [Martin, M. C. et al. Phys. Rev. B 53, 14285–14290 (1996).]. Considering the behavior in Fig. 4d this layer is more a metal than an insulator. Here it is necessary to bring clarity.
2. The data from Figs. 8 for magnetization in units of emu/cm3 are used for discussion and the key statements. To estimate the magnetization in [emu/cm3], one uses a magnetic moment in units of [emu], then divide it by the volume of the sample. What volume is used to calculate the magnetization in Fig. 8c: is it the volume of the entire bilayer? It is important to accurately measure the thickness of each layer. What is this accuracy for each layer?
3. Part of the last sentence is missing in the abstract.
4. In line 273, “exponential attenuation relationship” should be replaced with “allometric behavior”. What is the error of the exponent of this dependence?

Author Response

Dear reviewers:

Thank you for your work and for the reviewers’ comments concerning our manuscript. The manuscript coatings-1727148 has been carefully revised. We appreciate the detailed and useful comments and suggestions from you and reviewers. The point-by-point answers to the comments and suggestions were listed as below.

 

Reviewer #3:          

 

The electron transport and magnetic properties of a two-layer heterostructure consisting of a topological insulator layer Bi 2 Se 3 deposited on a La0.7Sr0.3MnO3 layer with spontaneous magnetization. The layers are characterized by surprisingly high crystallinity. Analyzing the temperature and field behavior of the resistance, the authors observe manifestations of charge transfer and redistribution, orbital hybridization and exchange coupling between layers, and spin injection. The authors also observe excessive magnetization of the heterostructure. The latter is associated with the appearance of spontaneous magnetization in Bi 2 Se 3 due to the magnetic proximity to La0.7Sr0.3MnO3. The discovered effect can be used to control the electronic and magnetic state of such heterostructures, which has prospects for application in spintronics and thermoelectric devices. The article can be accepted, after taking into account the following comments.

1. The authors refer to La 0.7 Sr 0.3 MnO3 as a ferromagnetic insulator, suggesting this as a well-known fact. However, this compound is referred to in the literature as half-metallic ferromagnetic La0.7Sr0.3MnO3 (LSMO) [ Vila-Fungueiriño, J. M. et al. sci. Technol. Adv. mater. 19, 702–710 (2018)] [Zhang, X. et al. J. Am. Ceram. soc. 104, 955–965 (2021)] and even as on metallic ferromagnet [Martin, M. C. et al. Phys. Rev. B 53, 14285–14290 (1996).]. Considering the behavior in Fig. 4d this layer is more a metal than an insulator. Here it is necessary to bring clarity.

Answer: Thank you for your comments. We agree with the suggestion, and have changed “ferromagnetic insulator” to “half-metallic ferromagnetic”.

1. The data from Figs. 8 for magnetization in units of emu/cm3 are used for discussion and the key statements. To estimate the magnetization in [emu/cm3], one uses a magnetic moment in units of [emu], then divide it by the volume of the sample. What volume is used to calculate the magnetization in Fig. 8c: is it the volume of the entire bilayer? It is important to accurately measure the thickness of each layer. What is this accuracy for each layer?

Answer: Thank you for your comments. the sample volume containing the volume of Bi₂Se₃, LSMO and LAO, which can be obtained by the film thickness(d) multiplied by the film area. And the thickness of each film can be confirmed by SEM.

1. Part of the last sentence is missing in the abstract.

Answer: The sentence is deleted in L. 18.

1. In line 273, “exponential attenuation relationship” should be replaced with “allometric behavior”. What is the error of the exponent of this dependence？

Answer: Thank you for suggesting. The “exponential attenuation relationship” has been replaced by “allometric behavior”. The exponent equals 0.5, indicating that the material is a two-dimensional system. The exponent mainly depends on the coherence length l_ø, while the l_ø depends on the surface transport channel. If the exponential deviation error is too large, it becomes a three-dimensional system.

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

 

A report of reviewing a manuscript, Coatings-1727148-v2

(1)  Regarding previous comment #8 made for the result and discussion section, stating:

concerning Figure 7(a) and (b), the magnitude of 10 K-magnetoresistance ratio (MR) of t=3 min sample indicted in this figure are approximately 1 % at 1 T and approximately 2 % at 6 T. On the other hand, the 10 K-MR of t=3 min sample indicated in Figure 6 are approximately 45 % at 6 T. These MR magnitudes look to be inconsistent.  

The author’s answer is:

These MR magnitudes look to be inconsistent due to the different calculation formulas. In Figure 7, the normalized magnetoresistance (MR) = ρ/ρ(B=0). Here, ρ and ρ (B=0) is the resistivity with and without magnetic field respectively. In Figure 6, the MR=（ρH-ρ0）/ρ0×100%. Here, ρ0 and ρH is the resistivity without and with an extra magnetic field respectively

The author’s answer is not clear because the ρ/ρ(B=0) value, e.g., 1.02 at 6 T for t=3 min sample, indicated in Figure 7 gives the (ρ_H-ρ₀₎/ρ₀×100% values of approximately 2 %, which is inconsistent with the displayed MR values of 45 % in Figure 6.

(2)  Regarding previous comment #9 made for the result and discussion section, stating:

How is the magnetoconductivity σ is obtained from the present magnetotransport measurements? A clear description is necessary in text.

The author’s answer is:

The magnetoconductivity σ is obtained by the resistivity, and the results of the resistivity shown in Figure 7(a) and (b). a sentence is added to illustrate the question in L. 262.

 The author’s answer cannot be accepted for magneto-transport researchers, because magnetoconductivity σ_xx is given in terms of longitudinal resistivity ρ_xx and transvers resistivity (Hall resistivity) ρ_yx as

σ_xx = ρ_xx/[(ρ_xx)²+(ρ_yx)²].

 

 

 

Not only resistivity measurement but also Hall effect measurement are required for accurate assessment of magnetoconductivity. However, neither measurement nor argument on Hall effect on Bi₂Se₃ / La_0.7Sr_0.3MnO₃ thin films grown on LaAlO₃ (100) single crystal were made in the revised manuscript.

 

 

   Since the previous comments #8 and #9 made for the result and discussion section are strongly related to the author’s conclusions, i.e., (i) the low-temperature (< 30 K) conductance of the hetero structure is mainly determined by Bi₂Se₃ whereas the high-temperature one is determined by La_0.7Sr_0.3MnO₃, (ii) the proximity effects are classified into the short-range and the long-range types, each of which are due to the chemical bonding characters of the Bi₂Se₃/La_0.7Sr_0.3MnO₃ interface and the spin injection mechanisms, respectively, (iii) the weak anti-localized (WAL) effect is suppressed by the magnetism induced by the proximity effects. Therefore, in the present stage, it is difficult to publish the revised manuscript in Coatings.

 

Author Response

(1)  Regarding previous comment #8 made for the result and discussion section, stating:

concerning Figure 7(a) and (b), the magnitude of 10 K-magnetoresistance ratio (MR) of t=3 min sample indicted in this figure are approximately 1 % at 1 T and approximately 2 % at 6 T. On the other hand, the 10 K-MR of t=3 min sample indicated in Figure 6 are approximately 45 % at 6 T. These MR magnitudes look to be inconsistent.  

The author’s answer is:

These MR magnitudes look to be inconsistent due to the different calculation formulas. In Figure 7, the normalized magnetoresistance (MR) = ρ/ρ(B=0). Here, ρ and ρ (B=0) is the resistivity with and without magnetic field respectively. In Figure 6, the MR=（ρH-ρ0）/ρ0×100%. Here, ρ0 and ρH is the resistivity without and with an extra magnetic field respectively

The author’s answer is not clear because the ρ/ρ(B=0) value, e.g., 1.02 at 6 T for t=3 min sample, indicated in Figure 7 gives the (ρ_H-ρ₀₎/ρ₀×100% values of approximately 2 %, which is inconsistent with the displayed MR values of 45 % in Figure 6.

Answer: Figure 7 shows the normalized magnetoresistance (MR) =ρ / ρ ( B = 0 ), which is the ratio of resistivity with and without magnetic field. The measured result is the change of Hall resistivity with the magnetic field. In Figure 6, the MR=（ρH-ρ0）/ρ0×100%.  Perhaps the difference is due to the different measurement methods (measurement methods described in the next issue). We can 't answer this question clearly yet, and we need more experiments to verify our conjecture.

 (2)  Regarding previous comment #9 made for the result and discussion section, stating:

How is the magnetoconductivity σ is obtained from the present magnetotransport measurements? A clear description is necessary in text.

The author’s answer is:

The magnetoconductivity σ is obtained by the resistivity, and the results of the resistivity shown in Figure 7(a) and (b). a sentence is added to illustrate the question in L. 262.

 The author’s answer cannot be accepted for magneto-transport researchers, because magnetoconductivity σ_xx is given in terms of longitudinal resistivity ρ_xx and transvers resistivity (Hall resistivity) ρ_yx as

σ_xx = ρ_xx/[(ρ_xx)²+(ρ_yx)²].

 Not only resistivity measurement but also Hall effect measurement are required for accurate assessment of magnetoconductivity. However, neither measurement nor argument on Hall effect on Bi₂Se₃ / La_0.7Sr_0.3MnO₃ thin films grown on LaAlO₃ (100) single crystal were made in the revised manuscript.

Answer: We agree with the reviewers ' understanding of resistivity. σ_xx is given in terms of longitudinal resistivity ρ_xx and transvers resistivity (Hall resistivity) ρ_yx as σ_xx = ρ_xx/[(ρ_xx)²+(ρ_yx)²].  Four-probe magnetoresistance (MR) was measured in a 9 T Quantum Design PPMS system which has a base temperature of 10K. In MR measurements, we fixed the B field in the z direction and the Hall voltage probes in the y direction as shown in Fig. 7(a). In Fig. 7(b), we show the MR data of films at low T with B field applied perpendicular to the film plane, i.e.,θ= 90⁰， whereθ is the angle between B and the current I. Therefore, we can obtain the resistivity ρ_yx through this measurement method. The resistivity ρ_xx is shown in Figure 4. σ_xx is given by resistivity ρ_xx and Hall resistivity ρ_yx .

Some descriptions have been added in our manuscript