Metal Ions’ Dynamic Effect on Metal-Assisted Catalyzed Etching of Silicon in Acid Solution
, Yichen Wang and Ya Hu *Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper is of good quality. The topic is relevant, the approach quite sound and English is more than acceptable. However, before final acceptance some points need to be addressed.
1. Authors keep using the word nanofilm to indicate the nanoparticles assembled and covering the surface. This is clearly a misnomer. A film is a continuous layer while particles sitting close to each other but leaving room for MACE to generate nanopillars cannot be continuous. Please change it.
2. (p. 2, line 75) Ag is surely not stable in the presence of oxidants. I would not say the same for Au. Please explain better what is meant.
3. (p. 2, line 100) 'illustrated' may not be the right word. Do you mean 'explored'?
4. (p. 2, line 109, sub-section title) I suppose it is Fe3+, not Fe2+
5. (p. 4, line 163) I think that you mean that Au3+ is reduced, not oxidized. Higher oxidation states are uncommon and anyway if it were oxidized over the process it would not inject hole into Si.
6. (p. 5, lines 199-206) You might like to rephrase the statement, which is quote confusing.
7. (p. 9, line 311) That "cations cannot be generally used as an oxidant" is quite contrasting with the widespread use of Ag+ in MACE. Please remove the statement.
When these points will have been cleared the paper could be considered for publication.
Author Response
Comments1: Authors keep using the word nanofilm to indicate the nanoparticles assembled and covering the surface. This is clearly a misnomer. A film is a continuous layer while particles sitting close to each other but leaving room for MACE to generate nanopillars cannot be continuous. Please change it.
Response 1：Thank you for your suggestion. After reviewing the literature, we agree that the term "nanofilm" may not accurately describe the structure composed of nanoparticles that are closely arranged but not formed into a continuous layer, so that the nanostructure can be generated by the MACE process. We have revised the manuscript and replaced "nanofilm" with "nanoparticle layer" to emphasize the discrete nature of the nanoparticles and their arrangement on the surface.
Comments 2. (p. 2, line 75) Ag is surely not stable in the presence of oxidants. I would not say the same for Au. Please explain better what is meant.
Response 2：Thank you very much for your suggestion. Ag is not stable under the action of oxidants, but Au is. We have made modifications to the paper. -p. 2, line 76
Comments 3. (p. 2, line 100) 'illustrated' may not be the right word. Do you mean 'explored'?
Response 3：Thank you for your valuable suggestions. We have revised the manuscript and used "stated" instead of "illustrated". - p. 3, line 101
Comments 4. (p. 2, line 109, sub-section title) I suppose it is Fe3+, not Fe2+.
Response 4：Thank you for your suggestion. We have made revisions to the manuscript. - p. 3, line 110
Comments 5. (p. 4, line 163) I think that you mean that Au3+ is reduced, not oxidized. Higher oxidation states are uncommon and anyway if it were oxidized over the process it would not inject hole into Si. 5.
Response 5：Thank you for your suggestion. During the etching process, Au3+ is strongly oxidizing, injecting holes into the adjacent silicon wafer to obtain electrons and reduce to Au. We have made modifications to the manuscript. - p. 4, line 163
Comments 6. (p. 5, lines 199-206) You might like to rephrase the statement, which is quote confusing.
Response 6：Thank you for your suggestion, we have made changes to the manuscript. p. 5, line 195-204
Comments 7. (p. 9, line 311) That "cations cannot be generally used as an oxidant" is quite contrasting with the widespread use of Ag+ in MACE. Please remove the statement.
Response 7：Thank you for your suggestion, we have made changes to the manuscript. p. 9, line 305-307
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsOverall, the article is interesting to read and is devoted to the important problem of generating porous silicon, which is increasingly used. An important advantage of the article is that the task is to study a number of nitrates of metals of the 4th period of the periodic table. At the same time, although this study is described in the abstract and in the text of the article, no comparison of the results for nitrates of these metals is carried out, which certainly reduces the value of the article. Another drawback is the extremely unsuccessful wording in some sentences and errors in the English language. Overall, I believe the article deserves to be published after a thorough revision of the text.
I provide a list of some comments below, but please note that it is far from complete, since there are a lot of similar errors.
1. Lines 27-29 - silicon nanowires and porous silicon are mentioned, but no connection is shown between them. Of course there is one, but the authors should specify in what sense they understand this connection.
2. Line 31. The authors talk about high pressure in the method discussed in [7,8], but a careful reading of these papers does not provide any information about pressure.
3. Lines 38-39, I doubt the correctness of the use of the terms "more orderly" and "flow law". I think it is worth checking that they are used this way.
4. Line 43. The term "nano-silicon" appears here, how is it related to porous silicon and nanowires? See remark 1.
5. Line 44. I am not sure that it is correct to say "liquid system"
6. Line 44 mentions "cleaned silicon wafer". What was the cleaning procedure? or does it mean something else?
7. The formula C3H6O appears twice in the article, but it can mean a dozen different chemical substances: https://en.wikipedia.org/wiki/C3H6O
8. I am completely perplexed by the use of the word hole in the work. In some places, it obviously means a positive quasiparticle (absence of an electron) in silicon, in some places it is a physical void, and in some places it is something else. I would like the authors to clearly specify what they mean in each case. Here are a few examples: "Ag+ is oxidizing, can inject hole", "The silicon injected into holes is oxidized", "the position of the silicon atom at 68the bottom of the silver nanoparticle becomes a small hole". I would like to remind the authors that the absence of a silicon atom in a crystal lattice site is usually called a vacancy.
9. line 63. "neat nanowire array" - you should specify what do you mean
10. line 71. "The silicon atoms are continuously oxidized and 71dissolved." - the meaning is not very clear. If soluble silicon compounds are formed, then that's how it should be written.
11. line 96 - "fume hood" - the term is unclear
12. line 96. "Then took out the silicon wafers 96and cleaned it with the DI water," - there's something wrong with the English language
13. In line 98, you mention an autoclave. What about the pressure in it? Does the process depend on the pressure?
14. line 104 - extra "in"
15. line 113. What is "general doping"
16. line 113. "the crystal orientation of the SiNW arrays are determined by the crystal orientation of the substrate." Did you experiment with different crystal orientations? How was the crystal orientation in the nanowire investigated?
17. When discussing the X-ray diffraction results, firstly, an explicit reference to the X-ray diffraction files should be given (not just 41-1402). Secondly, the angle of incidence of the radiation should be indicated. In Figure 1f, some peaks are labeled as Ag peaks, while others are labeled as Si. Why are the other peaks not listed? Do they refer to silicon, Ag or silver chloride? The absence of silicon peaks in the first graph and their presence in the second should be explained.
18. Line 154 - "plating solution" - where is its composition indicated?
19. The results in Figure 4 should be discussed in particular. Firstly, why were such different concentrations chosen? How can these results be compared? Is it possible to obtain some numerical estimates - for example, the height of silicon needles, the characteristic distance between them (easily determined by the minimum autocorrelation function, which can be calculated when processing the SEM image). Probably a good idea to summarise numbers for different solutions in a table. The results, the analysis and the conclusion of the article must necessarily contain conclusions about these results. At the same concentration, which solution produces faster etching? How does the etching rate depend on the concentration of the solution?
20. line 267. "Divalent and tetravalent dissolution" are mentioned first and last time in the paper. What is the mechanism and what is an idea of it? It should be described in the introduction. "The cathode is the reduction of the oxidant" - i don't understand
21. Line 266. "According to our previuos work" - please add a reference
22. Line 229 - "particulate matter attached" - i don't understand
23. line 280. I think K2SiF6 is slightly soluble in water (not insoluble). Please check and provide a reference
24. Reactions in lines 284-299 look different for K and Fe-based solutions. Could you provide such equations for other metals in the paper? How do the chemical properties of these metals relate to the reactions? How do they relate to the resulting characteristics of the silicon structures?
25. Please add doi to ref. 7: 10.1016/S0375-9601(96)00777-3
26. Conclusions are not informative. Please update it to be more in line with the abstract. It is not good when you mentioned 7 metals nitrates of 4th period in the abstract and pay no attention to them in conclusions.
27. Line 323. "Please add:" should be omitted.
Comments on the Quality of English Language
I think the idea is good, but there a lot to be improved in the paper.
Author Response
Comments 1. Lines 27-29 - silicon nanowires and porous silicon are mentioned, but no connection is shown between them. Of course there is one, but the authors should specify in what sense they understand this connection.
Response 1：Thank you for your suggestion. Silicon nanowires are one-dimensional nanostructures with a high aspect ratio and high specific surface area, which makes them have potential application value in the fields of photocatalysis and optoelectronics. Porous silicon is a porous structure whose pore structure can provide more active sites and show good performance in fields such as photodetectors and sensors. The two have their characteristics in structure and performance. The connection between them is mainly reflected in the fact that they are both nanostructures based on silicon materials and have similar application prospects in the fields of photocatalysis, optoelectronics and energy storage.
Comments 2. Line 31. The authors talk about high pressure in the method discussed in [7,8], but a careful reading of these papers does not provide any information about pressure.
Response 2：Thank you for your suggestion. The purpose of citing this part of the work in this article is to introduce the development of silicon nanowire preparation methods, from the initial preparation of porous silicon at high temperature and high pressure to the preparation of silicon nanowires by MACE technology at room temperature and pressure.
Comments 3. Lines 38-39, I doubt the correctness of the use of the terms "more orderly" and "flow law". I think it is worth checking that they are used this way.
Response 3：Thank you for your suggestion. “More orderly”: In the literature, K.Q. Peng did mention that silicon nanowires prepared by MACE technology have better order. This statement is based on the results of comparing with other preparation methods, emphasizing the advantage of MACE in structural orderliness. Regarding the use of “flow law”, we refer to the relevant literature [Hu Y, Peng K Q, et al. Scientific Reports, 2014, 4(1): 3667; Han H, et al. Nano Today, 2014, 9(3): 271-304.], confirming that this term is appropriate to describe the flow behaviour of electrons and holes during redox etching, especially in the discussion based on band bending theory.
Comments 4. Line 43. The term "nano-silicon" appears here, how is it related to porous silicon and nanowires? See remark 1.
Response 4：Thank you for your suggestion. Nano-silicon, porous silicon and nanowires are all nanostructures based on silicon, but they have different morphologies and properties. Their respective characteristics and advantages make them play an important role in different technical fields. In addition, silicon nanowires and porous silicon can be used as part of nano-silicon structures.
Comments 5. Line 44. I am not sure that it is correct to say "liquid system"
Response 5：Thank you for your suggestion. The MACE technology in the whole paper is carried out in a liquid environment, so it is very appropriate to use a liquid system to describe the etching environment here.
Comments 6. Line 44 mentions "cleaned silicon wafer". What was the cleaning procedure? or does it mean something else?
Response 6：Thank you for your suggestion. The procedure for cleaning silicon wafers in the paper is: cut the silicon wafer into 2*2 cm2 size, ultrasonically clean it in deionized water, CH3COCH3 and C2H5OH for 10min, rinse the silicon wafer three times with deionized water and soak it in 80℃ H2SO4-H2O2 solution for 30min, and finally use enough deionized water to clean the residual acid on the silicon wafer. This is described in detail in the Materials and Methods section.
Comments 7. The formula C3H6O appears twice in the article, but it can mean a dozen different chemical substances: https://en.wikipedia.org/wiki/C3H6O
Response 7：Thank you for your suggestion. We have made revisions to the manuscript. -p. 2, line 91、93
Comments 8. I am completely perplexed by the use of the word hole in the work. In some places, it obviously means a positive quasiparticle (absence of an electron) in silicon, in some places it is a physical void, and in some places it is something else. I would like the authors to clearly specify what they mean in each case. Here are a few examples: "Ag+ is oxidizing, can inject hole", "The silicon injected into holes is oxidized", "the position of the silicon atom at 68the bottom of the silver nanoparticle becomes a small hole". I would like to remind the authors that the absence of a silicon atom in a crystal lattice site is usually called a vacancy.
Response 8：Thank you for your suggestion, we have made revisions to the manuscript. In the process of preparing silicon nanowires by metal catalytic etching (MACE), "holes" refer to positive charge carriers in semiconductor silicon. Taking HF-AgNO3-H2O2 as an example, in the MACE process, due to the difference in electronegativity between silver ions and silicon, a local galvanic cell is formed, and a redox reaction occurs. Silver ions, as oxidants, obtain electrons and are reduced to silver nanoparticles, which are deposited on the surface of the silicon wafer. At the same time, holes are injected into the adjacent silicon, and silicon atoms lose electrons and are oxidized to form SiO2. SiO2 cannot exist stably in HF solution. After dissolution, holes will gradually form at the bottom of the silver nanoparticles, and finally form silicon nanowires. These holes participate in the oxidation process of silicon atoms and are one of the key steps in the formation of silicon nanowires.
Comments 9. line 63. "neat nanowire array" - you should specify what do you mean
Response 9：Thank you for your suggestion. We have made revisions to the manuscript. -p. 2, line 64
Comments 10. line 71. "The silicon atoms are continuously oxidized and 71dissolved." - the meaning is not very clear. If soluble silicon compounds are formed, then that's how it should be written.
Response 10: Thank you for your suggestion, we have made revisions to the manuscript. As the reaction proceeds, more Fe3+ in the solution injects holes into silicon atoms through silver particles. Silicon atoms are continuously oxidised to SiO2, which reacts with HF to generate silicofluoride and dissolves in the solution. -p. 2, line 72、73
Comments 11. line 96 - "fume hood" - the term is unclear
Response 11: Thank you for your suggestion, we have changed "fume hood" to "fume cupboard" in the manuscript. -p. 3, line 97
Comments 12. line 96. "Then took out the silicon wafers 96and cleaned it with the DI water," - there's something wrong with the English language
Response 12：Thank you for your suggestion. We have made revisions to the manuscript. -p. 3, line 97、98
Comments 13. In line 98, you mention an autoclave. What about the pressure in it? Does the process depend on the pressure?
Response 13：Thank you for your question. In our experiments, the Teflon reactor was sealed and placed in a 50℃ incubator for one hour. The reaction was carried out at normal pressure because the incubator was operated under normal atmospheric conditions and did not change the pressure. The metal-catalyzed etching of silicon nanowires is generally independent of pressure and is mainly driven by the electronegativity difference between metal nanoparticles and silicon. Pressure changes have little effect on it within the typical atmospheric pressure range. The Teflon reactor was chosen for the study because of its excellent chemical corrosion resistance and its ability to operate over a wide temperature range without the need for high pressure.
Comments 14. line 104 - extra "in"
Response 14：Thank you for your suggestion. We have made revisions to the manuscript. -p. 3, line 105
Comments 15. line 113. What is "general doping"
Response 15：Thank you for your question. In the field of semiconductor materials, "doping" refers to the intentional introduction of impurities into a material to change its conductive properties. "General doping levels" refer to the typical concentrations of dopants in silicon wafers used to make p-type or n-type silicon. P-type silicon is produced by doping with elements such as boron, aluminium, or gallium, which have fewer valence electrons than silicon and therefore create "holes" that facilitate the conduction of positive charge carriers. On the other hand, n-type silicon is made by doping with elements such as phosphorus, arsenic, or antimony, which have more valence electrons than silicon, resulting in an excess of electrons that favours the conduction of negative charge carriers. In our experiments, the silicon wafers used had typical doping levels that are commonly used in device manufacturing in the semiconductor industry. -p. 3, line 114
Comments 16. line 113. "the crystal orientation of the SiNW arrays are determined by the crystal orientation of the substrate." Did you experiment with different crystal orientations? How was the crystal orientation in the nanowire investigated?
Response 16：Thank you for your suggestion. We have carried out metal-catalyzed etching of silicon nanowires with different crystal orientations. For details, please refer to the literature [Hu Y, Peng K Q, et al. Scientific reports, 2014, 4(1): 3667.] and other related literature [Li X. Current Opinion in Solid State and Materials Science, 2012, 16(2): 71-81; Leonardi A A, Nanomaterials, 2021, 11(2): 383.]. In the process of metal-catalyzed etching of silicon nanowires, the crystal orientation of the nanowires can be analyzed by testing methods such as TEM, XRD and Raman scattering spectroscopy. -p. 3, line 116
Comments 17. When discussing the X-ray diffraction results, firstly, an explicit reference to the X-ray diffraction files should be given (not just 41-1402). Secondly, the angle of incidence of the radiation should be indicated. In Figure 1f, some peaks are labeled as Ag peaks, while others are labeled as Si. Why are the other peaks not listed? Do they refer to silicon, Ag or silver chloride? The absence of silicon peaks in the first graph and their presence in the second should be explained.
Response 17：Thank you for your suggestion. We have made changes to the manuscript. In Figure 1f, we have listed and marked all the peaks that appear. Si (200) refers to silicon, Ag (111) and Ag (200) refer to silver, and the rest are AgCl. In addition, the explanation is also given in the manuscript. The absence of a silicon peak in the first figure may be because the surface of the silicon wafer is covered with silver nanoparticles. The size and distribution of silver nanoparticles may affect the diffraction signal of XRD, resulting in the silicon peak being unclear or undetected. Silver nanoparticles may have a certain shielding effect on the crystal structure of the silicon wafer in the XRD test due to their surface plasmon resonance (SPR) characteristics, thereby affecting the detection of silicon peaks. The surface plasmon resonance characteristics of silver nanoparticles may affect their interaction with the substrate [Wen J B. Preparation, characterization and antibacterial properties of silver nanoparticles [D]. Central South University, 2024]. In the second figure, silver chloride particles are generated on the surface of the silicon wafer, which enhances the crystal structure of the silicon wafer surface, making the silicon peak obvious in XRD detection. -p. 3, line 138、141
Comments 18. Line 154 - "plating solution" - where is its composition indicated?
Response 18：Thank you for your suggestion. The composition of the plating solution is AuCl4H-HF-H2O.
Comments 19. The results in Figure 4 should be discussed in particular. Firstly, why were such different concentrations chosen? How can these results be compared? Is it possible to obtain some numerical estimates - for example, the height of silicon needles, the characteristic distance between them (easily determined by the minimum autocorrelation function, which can be calculated when processing the SEM image). Probably a good idea to summarise numbers for different solutions in a table. The results, the analysis and the conclusion of the article must necessarily contain conclusions about these results. At the same concentration, which solution produces faster etching? How does the etching rate depend on the concentration of the solution?
Response 19：Thank you for your suggestion. The concentration was chosen to ensure that the NO3- concentration was 0.48M because a better etching effect could be achieved when the NO3-concentration was 0.48M. In the experiment to discuss the effect of different cation types on the morphology and etching rate of SiNWs arrays, in the HF-M(NO3)x-H2O system with a NO3- concentration of 0.48M, the SiNWs formed by the silver-coated silicon wafer reacting at a treatment temperature of 50℃ for 1h had obvious length differences, which represented different etching rates. At the same concentration, the length of the SiNW array prepared in Ni(NO3)2、Co(NO3)3 and Ca(NO3)3etching solutions was about 5-6 μm, and the etching rate was faster. The factor that has the greatest impact on the etching rate is the different cation types. The cation type determines the MACE etching rate and thus affects the morphology of the SiNWs array, which has been shown in Figures 4 and 5 of the paper. In different M(NO3)x systems, when the types of metal ions are different, there are obvious differences in the SiNWs formed. For example, in the HF-KNO3-H2O system, as the concentration of KNO3 in the silver-coated silicon wafer increases, a layer of K2SiF6 will form on the surface of the silicon wafer, which will hinder the further occurrence of the etching reaction and affect the etching rate of SiNWs, thereby affecting the morphology of the SiNWs array.
Comments 20. line 267. "Divalent and tetravalent dissolution" are mentioned first and last time in the paper. What is the mechanism and what is an idea of it? It should be described in the introduction. "The cathode is the reduction of the oxidant" - i don't understand
Response 20：Thank you for your suggestion. We have revised the paper. In the 1950s, Professor Uhlirs of Bell Labs observed the generation of bubbles when preparing porous silicon micro-nanostructures in hydrofluoric acid solution by applying voltage [M.J. Sailor. Porous Silicon in Practice: Preparation, Characterization and Application[M]. 2011. 1-2.]. He used the traditional drainage method to collect bubbles in a test tube and ignited them with a spark, which made a crisp "bang" sound. Professor Uhlirs initially determined that the bubbles produced were hydrogen, and proposed that silicon would be oxidized and dissolved in two different ways under different voltage conditions: two-electron reaction and four-electron reaction, as shown in the figure below. When the applied voltage is low, the corrosion is slow to produce porous silicon, a two-electron reaction occurs, silicon loses two electrons, and the hydrogen ions in the solution are also reduced to become silicofluoride and produce hydrogen. When the applied voltage is very high, the corrosion is severe, the surface of the silicon wafer is polished, a four-electron reaction occurs, and silicon loses four electrons and directly becomes silicofluoride. In addition, according to previous research by our research group, the generation of hydrogen was detected by gas chromatography. At the same time, it was found that the water film on the surface of the silver-coated silicon wafer accumulated more and more as the reaction time increased, but no water film was found on the surface of the blank silicon wafer. Based on the above facts, we believe that the MACE process of silicon in Au or Ag is a mixed process of 2 electrons and 4 electrons [Y. Hu, et al. Nano. Letters 14 (2014) 8]. In addition, as explained in Figure 6, the reduction reaction of NO3- and Fe3+ at the cathode, which itself acts as an oxidant, has been demonstrated in the manuscript.
Figure 1 General curve relationship between voltage and current density of the electrochemical reaction of silicon in hydrofluoric acid
Comments 21. Line 266. "According to our previuos work" - please add a reference
Response 21：Thank you for your suggestion. We have made revisions to the manuscript. -p. 7, line 260
Comments 22. Line 229 - "particulate matter attached" - i don't understand
Response 22：Thank you for your suggestion. We have made revisions to the manuscript.
Comments 23. line 280. I think K2SiF6 is slightly soluble in water (not insoluble). Please check and provide a reference
Response 23：Thank you for your advice. By looking up relevant information, K2SiF6 is almost insoluble in cold water, but soluble in hot water, and will decompose into KF, HF and H2SiO3 after dissolving.
1.https://www.chembk.com/cn/chem/%E6%B0%9F%E7%A1%85%E9%85%B8%E9%92%BE.
Comments 24. Reactions in lines 284-299 look different for K and Fe-based solutions. Could you provide such equations for other metals in the paper? How do the chemical properties of these metals relate to the reactions? How do they relate to the resulting characteristics of the silicon structures?
Response 24：Thank you for your suggestion. In our paper, we introduced in detail the etching reaction mechanism equations in the HF-Fe(NO3)3-H2O and HF-KNO3-H2O systems. The reaction equations in other metal solution systems in the paper are the focus of our next study. The MACE process is mainly based on the galvanic cell reaction mechanism, so for the metal ions in the solution, we focus on the effect of the redox property of the metal ions on the etching rate. This paper mainly studies the MACE of silicon wafers pre-deposited with metal nanofilms in HF-M(NO3)x-H2O etching solution (where M(NO3)x is nitrate of an element in the fourth period of the periodic table), discusses the dynamic effects of different types of metal cations on the etching process, and proves that the type of cations affects the morphology and etching rate of the SiNW array, as shown in Figures 4 and 5.
Comments 25. Please add doi to ref. 7: 10.1016/S0375-9601(96)00777-3
Response 25：Thank you for your suggestion. We have made revisions to the manuscript. -p. 10, line 340
Comments 26. Conclusions are not informative. Please update it to be more in line with the abstract. It is not good when you mentioned 7 metals nitrates of 4th period in the abstract and pay no attention to them in conclusions.
Response 26：Thank you for your suggestion. We have made revisions to the manuscript. -p. 9, line 114
Comments 27. Line 323. "Please add:" should be omitted.
Response 27：Thank you for your suggestion. We have made revisions to the manuscript. -p. 9, line 301、301
Author Response File: Author Response.pdf
