Atomic Layer Deposition of La2O3 Film with Precursor La(thd)3-DMEA
Tzia Ming Onn
Round 1
Reviewer 1 Report
The present paper deals with the synthesis and characterization of La precursors for La2O3 ALD. Three La complexes with mixed ligands are reported and their thermal properties are discussed. The authors investigated the effect of monoamine and diamine ligands on the volatility and thermal stability. The La(thd)3-DMEA complex is then used for ALD growing of La2O3 layer. The manuscript is clear, well-written and is suitable for publication in coating but some data are missing or not discussed in the manuscript. My comments are as follow.
-Why the crystallographic data of the 4th complex are not presented in table 1 and 2? Similarly, the TGA of complex 3 is not presented. It would be suited to add the missing data for complexes 3 and 4 at least in SI.
-The caption of table 1 does not correspond to the content.
-The SEM result in figure 5 is not discussed in the main text.
-Regarding the La2O3 ALD process, could the authors comment on the GPC increase above 270°C?
-After sputtering C and N impurities are still detected. I think they come from an incomplete removal on the N ligand and not only for air contamination. Could the authors comment on that? Why O3 was chosen as co-reactant?
Author Response
Point 1: Why the crystallographic data of the 4th complex are not presented in table 1 and 2? Similarly, the TGA of complex 3 is not presented. It would be suited to add the missing data for complexes 3 and 4 at least in SI.
Response 1: Although numerous attempts were made, a high-quality crystal of complex 4 could not be obtained, despite its easy solubility in organic solvents like n-hexane. The crystal structure of complex 3 supported the analysis of complex 4 because of the association between the two complexes. Based on the crystal structure and 1H NMR data, the structure of complex 3 appeared reasonable. Additionally, the 1H NMR data of complex 4 demonstrated that its structure conformed to the target product. The Thermogravimetric analysis of complex 3 has been described in the paper.
Point 2: The caption of table 1 does not correspond to the content.
Response 2: the caption of Table 1 was corrected in the paper.
Point 3: The SEM result in figure 5 is not discussed in the main text.
Response 3: The film thickness, which was used to calculate the growth per cycle, was determined by cross-sectional SEM imaging .As shown in Fig.5a. A 24 nm thick film, grown for 500 ALD cycles at 240 °C, was used to calculate the growth per cycle.
Point 4: Regarding the La2O3 ALD process, could the authors comment on the GPC increase above 270°C?
Response 4: Within the ALD window, the growth rate was maintained at a nearly constant value of proximately 0.4 Å/cycle, and the high growth rate at deposition temperature above 270 oC was attributed to the thermal decomposition of the precursor.
Point 5: After sputtering C and N impurities are still detected. I think they come from an incomplete removal on the N ligand and not only for air contamination. Could the authors comment on that? Why O3 was chosen as co-reactant?
Response 5: I agree with your viewpoint that the C and N impurities were still detected after Ar+ sputtering, indicating incomplete removal from the nitrogen-containing bidentate ligand and air atmosphere. According to literature research, ozone was the optimal co-reactant, which can prevent the formation of lanthanum hydroxide and improve the properties of the deposited films.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper explores a precursor La(thd)3-DMEA, which is an extension of the conventional La(thd)3 precursor used in ALD. The paper also provides schematic of the formation of the DMEA precursor and does a good job in characterizing the precursor material. This paper has a clear story and provides clear characterization of the complex. I would like the authors to address the following before recommending for publication:
1) Can the authors provide error bars for Figure 4? While I understand the growth rate is fairly smooth compared to many different thin film techniques, I think a sense of error bar lends credibility to the data. There is always some roughness associated with films and I appreciate the AFM image shown in Figure 5b. Can the authors provide a more zoomed in image of Figure 5a as a sub-panel? I am not certain with the claim that the author had about it being a smooth surface.
2) Can the authors refer to some of these papers for consideration for their use of La as a precursor and application in catalysis as well as other materials as oxides for dielectrics:
a. doi.org/10.1021/jacs.7b12900
b. doi.org/10.1021/jacs.2c09481
c. doi.org/10.1016/j.jcat.2019.11.040
3) I am trying to understand Figure 6 and Figure 7. What does the author mean by etching for 50s? Can the author provide a description about this? Is this referred to as the Ar+ sputtering? Please clarify. I don’t think Figure 7 is necessary for the main figure.
4) Would thermally heating the deposited film at elevated temperatures remove the amount of N in the deposited film?
5) While the authors did not comment on the thermal stability of the new precursor, can they provide some feedback or comparison between the thd vs. the thd-DMEA in terms of the stability or if any when exposed to air? What would happen if oxygen was used instead of ozone at the ALD window in this case- would it be enough to decompose the ligands, I understand ozone is supposedly better but I am wondering about the half-life of ozone and the distance it would travel in the entire ALD chamber to reach the sample?
6) Overall, I think the paper is clear in terms of what it is. A new precursor for La oxide ALD. Please address the comments above.
English is fine.
Author Response
Point 1: Can the authors provide error bars for Figure 4? While I understand the growth rate is fairly smooth compared to many different thin film techniques, I think a sense of error bar lends credibility to the data. There is always some roughness associated with films and I appreciate the AFM image shown in Figure 5b. Can the authors provide a more zoomed in image of Figure 5a as a sub-panel? I am not certain with the claim that the author had about it being a smooth surface.
Response 1: Thank you for your practical suggestion. I have modified Figures 4 and 5 in the paper accordingly.
Point 2: Can the authors refer to some of these papers for consideration for their use of La as a precursor and application in catalysis as well as other materials as oxides for dielectrics:
a.doi.org/10.1021/jacs.7b12900
b.doi.org/10.1021/jacs.2c09481
c.doi.org/10.1016/j.jcat.2019.11.040
Response 2: These articles are suitable for reference in this paper.
Point 3: I am trying to understand Figure 6 and Figure 7. What does the author mean by etching for 50s? Can the author provide a description about this? Is this referred to as the Ar+ sputtering? Please clarify. I don’t think Figure 7 is necessary for the main figure.
Response 3: I appreciate your reminder that the paper lacked unified terminology. I have made the necessary modifications. The purpose of Ar+ sputtering for 50 s to exclude the surface contamination prior to measurement. The purpose of figures 7a and 7b were to determine the composition of the film. Besides, the XPS spectrum of the film after sputtering was not analysis, because Ar+ may lead to change the composition of the film.
Point 4: Would thermally heating the deposited film at elevated temperatures remove the amount of N in the deposited film?
Response 4: After the annealing at 500 oC for 30 min in N2 ambient furnace, the content of the N element was dramatically decreased to 0.04 %. The decrease in the content of the N element with post-annealing may be attributed to densification of the films.
Point 5: While the authors did not comment on the thermal stability of the new precursor, can they provide some feedback or comparison between the thd vs. the thd-DMEA in terms of the stability or if any when exposed to air? What would happen if oxygen was used instead of ozone at the ALD window in this case- would it be enough to decompose the ligands, I understand ozone is supposedly better but I am wondering about the half-life of ozone and the distance it would travel in the entire ALD chamber to reach the sample?
Response 5: The complexes were synthesized in an air atmosphere, but the target products were not sensitive to air. Experimental verification showed that oxygen, as a co-reactant, could not completely decompose the ligands. Therefore, ozone, a strong oxidizer, was used to achieve self-limiting ALD-type growth due to the low basicity of the β-diketonate ligand. Generally, the half-life of ozone was 22.5 minutes. Ozone with a volume concentration of approximately 7% in O2 was generated from oxygen gas (99.999%) using an ozone generator.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
The current revision has addressed the issues raised.
