Simultaneously tuning interlayer spacing and termination of MXenes by Lewis-basic halides

The surface and interface chemistry are of significance on controlling the properties of two-dimensional transition metal carbides and nitrides (MXenes). Numerous efforts have been devoted to the regulation of Ti3C2Tx MXene, however, tuning interlayer spacing and surface halogen termination of other MXenes (besides Ti3C2Tx) is rarely reported while demanded. Here we propose a Lewis-basic halides treatment, which is capable of simultaneously engineering the interlayer spacing and surface termination of various MXenes. Benefited from the abundant desolvated halogen anions and cations in molten state Lewis-basic halides, the -F termination was substituted by nucleophilic reaction and the interlayer spacing was enlarged. Ti3C2Tx MXene treated by this method showed a high specific capacity of 229 mAh g−1 for Li+ storage, which is almost 2 times higher than pristine one. Considering the universality, our method provides an approach to regulating the properties of MXenes, which may expand their potential applications in energy storage, optoelectronics and beyond.


(Simultaneously tuning interlayer spacing and termination of MXenes by Lewisbasic halides)
We appreciate for the editor and reviewers' constructive suggestions to our manuscript.
All the suggestions are very helpful for us to improve our research. We have addressed all comments of the reviewers, and our responses to the reviewer's comments are listed below point-by-point. The revised sections have been highlighted in the revised manuscript.

Reviewer #1
The authors reported a new and effective strategy to regulating the terminations and interlayer spacing of MXenes simultaneously by Lewis-basic halides. This strategy can easily change terminations from F to Br and I with enlarged interlayer spacing at a relatively low temperature, which cannot be achieved in previous researches. The enlarged interlayer spacing and regulated termination species for MXenes are responsible for improved electrochemical performance. And I believe that this strategy can be broadened to more MXenes and applicable to other fields. Therefore, I recommend publication of this work after a minor revision.
Response: We appreciate the efforts the reviewer has spent on our manuscript and thanks for the reviewer's positive comments. The point-by-point response has been done as follows. Response: According to our experimental results, the increase in interlayer spacing of Ti3C2Tx, Nb4C3Tx and Mo2Ti2C3Tx is 3.4, 1.2 and 2.7 Å, respectively. In general, the intercalation process is largely controlled by terminations, as it affects the surface charges. We infer that the difference of interlayer spacing of the above MXenes is attributed to the different surface negative charges (Chem. Mater. 2022, 34, 678-693).
Comment 2. b value is an important factor for energy storage. Therefore, the b value of LB-Ti3C2Tx needs to be calculated to determine whether it is a battery/capacitive behavior.
Response: Thanks for the reviewer's suggestion. We have calculated b value in the revised manuscript. According to the power-law relationship i=av b (Nat. Mat. 2013, 12, 518-522), the correlation between the peak current (i) and the sweep rate (v) can be studied to distinguish the storage process of the charge, where a and b denote arbitrary coefficients. When b is close to 0.5, it shows diffusion-controlled charge storage. When it is close to 1.0, it shows capacitance-controlled charge storage. In our work, the calculated b value (0.87) is close to 1 (Fig. R1-1), indicating that the LB-Ti3C2Tx electrode is dominated by capacitance-controlled charge storage.  The yield of this method is about 92%. And it could be scaled up by using large volume stirrer.

Reviewer #2
The authors present a general method that is capable of simultaneously tuning the interlayer spacing and termination of MXenes by Lewis-basic halides. They claim that the abundant desolvated Na + and K + in Lewis-basic halides are responsible for the highly enlarged interlayer spacing through the efficient intercalation process, the desolvated halogen ions are involved in the termination substitution via nucleophilic reaction. This method paves the way for engineering the interfacial and surface chemistry of MXenes. However, there are some issues should be considered.
Response: We appreciate the efforts the reviewer has spent on our manuscript and thanks for the reviewer's positive comments. The point-by-point response has been done as follows. Comment 1. It is well known that MXene will self-oxidize at high temperature even in argon and it is relatively stable at low temperature. Through the treatment of eutectic molten salt, LB-Ti3C2Tx MXene shows more titanium dioxide species than pristine Ti3C2Tx in Fig. S6. Does this mean that this method has little effect on inhibiting MXene self-oxidation?
Response: According to XPS, the proportion of TiO2 increased a bit from 9.34% to 13.24% (Table R2-1). And there is no peak of TiO2 in the XRD pattern ( Fig. R2-1), indicating that the oxidation degree of LB-Ti3C2Tx is still relatively low. Although there is supposed to be no water/oxygen-containing molecules in molten salt, the O and OH terminations may self-oxidize MXene, which should be the reason for the small increase of TiO2 in MXene after treatment. Based on the above analysis, although only a low degree of oxidation was detected, we think our method couldn't completely inhibit MXene self-oxidation.    Further, we test the EDS of LB-Ti3C2Tx after cycling. As the charging and discharging cycles increase, the atomic content of Na + and K + decreased. (Table R2-2).
Therefore, we speculate that Na + and K + will migrate (replaced by Li + ) from the interlayer during cycling.   (Fig. S22), which means the interlayer structure of LB-Ti3C2Tx is pretty stable during cycling." Comment 4. We all know that Li + is much smaller than Na + and K + . Perhaps the storage of Na + or K + can better show the advantages of LB-Ti3C2Tx materials.
Response: Based on the reviewer's suggestion, we also tested the performance of Na in Na + battery. In contrast, the LB-Ti3C2Tx electrode achieves a much higher capacity of 122 mAh g −1 (Fig. R2-4c). The above results show that LB-Ti3C2Tx electrode has better storage of Na + than Ti3C2Tx, which is similar to the enhanced Li + storage behavior in LB-Ti3C2Tx. Voltage profiles of (b) Ti3C2Tx and (c) LB-Ti3C2Tx at various current densities.

Reviewer #3
This is paper that could possibly be published. But more information is needed (especially XPS) to shore up their conclusions. I would like the authors to address some of the following questions/comments.

Response:
We appreciate the efforts the reviewer has spent on our manuscript. The point-by-point response has been done as follows.

Nb4C3Tx and LB-Nb4C3Tx
In the high resolution XPS spectra of Nb4C3Tx and LB-Nb4C3Tx, the Nb 3d region

Mo2Ti2C3Tx and LB-Mo2Ti2C3Tx
In the high resolution XPS spectra of Mo2Ti2C3Tx and LB-Mo2Ti2C3Tx, the Mo 3d region ( Fig. R3-3a and 3d) could be fitted by three components (detailed peak information is shown in Table R3 In the high resolution XPS spectra of Ti3C2Tx and LB-Ti3C2Tx, the Ti 2p region (Fig. S6a and 6d) could be fitted by five components (detailed peak information is shown in Table S2). The largest fractions are attributed to Ti-C bond, and smaller fractions of TiO2 is also found. After treated by Lewis basic halides, LB-Ti3C2Tx exhibits slight increase of TiO2 and decrease of Ti-F bond, which could be induced by the substitution of F termination by Br termination. As shown in Fig     In the high resolution XPS spectra of Mo2Ti2C3Tx and LB-Mo2Ti2C3Tx, the Mo 3d region ( Fig. S14a and 14d) could be fitted by three components (detailed peak information is shown in Table S5). These components are mainly attributed to C-Mo and two small species (Mo +5 and Mo +6 ) belonging to mixed molybdenum oxides. The Ti 2p region ( Fig. S14b and 14e) could be fitted by four components, corresponding to Ti-C, Ti +2 -C, Ti +3 -C and Ti2O, respectively.  Response: We would like to thank for the reviewer's comments and totally understand the concern. We didn't intend to emphasize the expanded interlayer space, instead, we would like to show that our method could tune the termination and interlayer spacing simultaneously, especially for MXenes besides Ti3C2Tx. Following reviewer's suggestion, we revised the statement in the manuscript and further explained the reasons of enlarged interlayer spacing, which is attributed to the synergism of termination substitution and desolvated cations intercalation.
Action: We have modified the sentences in our revised manuscript, page 1 and 7: "This is in accordance with the variation trend of the amount of desolvated Na + and K + in AlBr3/NaBr/KBr molten salts as mentioned previously, demonstrating the increased interlayer spacing of Ti3C2Tx is dominantly attributed to the efficient intercalation of desolvated Na + and K + from AlBr3/NaBr/KBr." Response: We agree with reviewer that solvated ions readily intercalate at RT.
Following reviewer's suggestion, now we would like to emphasize that the desolvated ions require a lower activation energy than the solvated ions in the termination substitution process according to the Arrhenius equation. This is evidential that the substitution reaction of Br termination cannot be carried out in aqueous solution. At the same time, the desolvated cations could intercalate into the narrow interlayer space of MXenes (besides Ti3C2Tx) to efficiently increase the distance. This is the advantage of desolvated ions in our method.
Action: According to reviewer's comment, we have modified and marked the sentences in our revised manuscript, page 3: "If the above termination substitution process occurs in aqueous solution, the overall activation energy increases due to the energy consumption for the necessary step of ion de-solvation, which limits the reaction rate according to the Arrhenius equation. At the same time, the desolvated cations could intercalate into the narrow interlayer space of MXenes to efficiently increase the distance." Comment 4. The authors claim, " In this context, the molten salt system is of particular interest because "naked" ions could be directly involved in reaction, which is beneficial for largely increasing interlayer spacing and efficient termination substitution." I don't get logic here. If higher interlayer spacings are beneficial, then ideally one should prefer solvated ions over naked ions as the former have larger diameters than the latter.
Response: We think our expression is not appropriate here. The diameter of the solvated ions is indeed larger than that of the naked ions, and 16 Å spacing was achieved in Ti3C2Tx MXene at wet state. However, it is still difficult to enlarge the interlayer spacing of Nb4C3Tx and Mo2Ti2C3Tx via intercalation in salt solutions (NaCl, LiCl). By contrast, in our work, the "naked" ions are easier to intercalate into Nb4C3Tx and Mo2Ti2C3Tx. Therefore, we changed the statement to express that the "naked" anions are beneficial for termination substitution and "naked" cations could simultaneously intercalate MXenes.
Action: According to reviewer's comment, we have modified the sentences in our revised manuscript, page 3 and 7: "In this context, the molten salt system is of particular interest as "naked" anions are beneficial for termination substitution and "naked" cations could simultaneously intercalate MXenes" "For short conclusion, the desolvated anions (Br -) in AlBr3/NaBr/KBr play the important role on substituting termination and desolvated cations (Na + and K + ) could simultaneously enlarge the interlayer spacing." We highly appreciate the comments from reviewer and reconsider the reasons of  Table S1." "As shown in Fig. 1, the interlayer spacing of MXene could be enlarged due to the synergism of termination substitution and desolvated cations intercalation (Na + and K + ) in Lewis-basic halides (Fig. 1a)" "The desolvated halogen ions are involved in the termination substitution via nucleophilic reaction. On the other hand, the synergistic effect of termination substitution and cations (abundant desolvated Na + and K + ) intercalation are responsible for the enlarged interlayer spacing." "This is in accordance with the variation trend of the amount of desolvated Na + and K + in AlBr3/NaBr/KBr molten salts as mentioned previously, demonstrating the increased interlayer spacing of Ti3C2Tx is dominantly attributed to the efficient intercalation of desolvated Na + and K + from AlBr3/NaBr/KBr."  Response: We understand the concern from reviewer and really appreciate the suggestions. Actually, we didn't intend to overlook the previous work. We would like to show the advantage of our method: first, our method required a low temperature which can be obtained with a normal magnetic stirring heater easily. As a comparison, high temperature (450~750°C) is needed in previous works (Table R3-  "For Lewis acid molten salt method (above 450 ℃), termination is more controllable and tunable" "However, in molten salt system, the resulted MXenes showed a relatively small interlayer spacing (about 10.9 Å), in which the interlayer spacing could only be enlarged by organic reagents (difficult to be removed) intercalation in previous researches." Comment 7. If I look at Table S1, there is serious reduction in the Ti-C MXene signal.
That is worrisome. DO the authors have an explanation?
Response: We noted that, in Ti 2p region of XPS, the proportion of Ti-C decreased from 85.51% to 85.38% (almost no decrease), which showed that the Ti3C2Tx was not destroyed. However, the signal of Ti-C MXene was relatively decreased in C 1s.
Therefore, we infer that there is a small amount of organic solvent residue (anhydrous  Table S3?
Response: Based on the reviewer's suggestion, we have showed Nb results in Table   R3-2.
Action: We have added the Table S4 in Supplementary Information. Response: Dissolving AlBr3 in water will generate a lot of heat, and the reaction is violent. The solubility of AlBr3 by anhydrous diethyl ether is larger, and the exothermic heat is smaller. Next, the solubility of anhydrous diethyl ether in water is small (slightly soluble), and tetrahydrofuran is used to wash anhydrous diethyl ether (Nature 2017, 542, 328-331).

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): I appreciate the authors' careful responses to the other reviewers' and my comments. I believe the manuscript has been strengthened appreciably and is now suitable for publication.
Reviewer #2 (Remarks to the Author): The author mentioned what I care about. I think the current status is acceptable.
Reviewer #4 (Remarks to the Author): The authors have added XPS as requested and also made some other changes. But I still have some comments on the revised manuscript.
1) In the XPS discussion of -Br substituted Ti3C2 the authors say " It's worth noting that the strength of C-C and C-O bonds increases in C 1s and O 1s regions of LB-Ti3C2Tx". This is incorrect. An increase in peak area corresponds to an increase in the % content of the species, it does not imply, an increase in bond strength as the authors say.
2) XPS fittings need appropriate citations. How did the authors attribute the B.E's to specific components? give reference for each.
3) Why is there FWHM change for the same components before and after -Br substitution?
4) The authors need to fit the MXene contributions in XPS with asymmetric peaks, using symmetric peaks causes unwanted extra components. For ex: in the case of Mo 3d peak in Fig S14 and d, the contribution from oxide will be much lower than shown. See some recent reviews on XPS of MXenes, they discuss this exact problem. 5) Please quantify the chemical formulas based on XPS before and after salt treatment.
6) The authors say " At the same time, the desolvated cations could intercalate into the narrow interlayer space of MXenes to efficiently increase the distance". Ions always intercalate between MXene layers to balance charge. So there is nothing special per say here that the authors are emphasizing. De-solvated ions can be achieved in regular MXene by mere vacuum drying, and even in this case, will the desolvated ions not solvate as soon as they come in contact with the electrolyte in supercapacitors? Response: We appreciate the efforts the reviewer has spent on our manuscript. Thanks a lot for all the positive comments on our work.

Reviewer #2
The author mentioned what I care about. I think the current status is acceptable.
Response: We appreciate the efforts the reviewer has spent on our manuscript and thanks for the reviewer's positive comments.

Reviewer #4
The authors have added XPS as requested and also made some other changes. But I still have some comments on the revised manuscript.
Response: We appreciate the efforts the reviewer has spent on our manuscript and thanks for the reviewer's positive comments. We have further revised the manuscript per reviewer's suggestion. The point-by-point response has been done as follows.