Direct investigation of the reorientational dynamics of A-site cations in 2D organic-inorganic hybrid perovskite by solid-state NMR

Limited methods are available for investigating the reorientational dynamics of A-site cations in two-dimensional organic–inorganic hybrid perovskites (2D OIHPs), which play a pivotal role in determining their physical properties. Here, we describe an approach to study the dynamics of A-site cations using solid-state NMR and stable isotope labelling. 2H NMR of 2D OIHPs incorporating methyl-d3-ammonium cations (d3-MA) reveals the existence of multiple modes of reorientational motions of MA. Rotational-echo double resonance (REDOR) NMR of 2D OIHPs incorporating 15N- and ¹³C-labeled methylammonium cations (13C,15N-MA) reflects the averaged dipolar coupling between the C and N nuclei undergoing different modes of motions. Our study reveals the interplay between the A-site cation dynamics and the structural rigidity of the organic spacers, so providing a molecular-level insight into the design of 2D OIHPs.


REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): Nat Comm Lin et al. investigate reorientational dynamics of methylammonium cations in 2D and 3D organicinorganic halide perovskites (OIHP) using a specialized NMR probe (Rotational echo double resonance, REDOR) to interrogate the A site dynamics. Related solid-state NMR methods have seen increased use in studies directed towards understanding how A site dynamics influence optoelectronic properties. Issues mentioned with related techniques concern the ability to resolve these dynamics due to overlapping transitions for organic spacers between perovskite octahedra. They also use an isotopic labeling approach to better resolve A-site dynamics. While I do not completely understand the REDOR technique it does appear that they are capable of reliably discriminating between A-site and organic spacer contributions to the NMR signals. The work appears well executed but the paper in its current form is difficult to follow because and it is not intuitive for non-specialists. Furthermore, it is difficult to understand the significance of the results and how they pertain to particular perovskite structures. I believe the work is publishable but needs some revisions to improve the readability for readers to appreciate the results. Comments appear below. Comments: -It would be extremely helpful if the authors included a cartoon of the structures they are discussing and the proposed dynamics being measured from experiment. While many understand the general perovskite structure it is harder to envisage lower dimensional systems and how they differ from conventional systems in both structure and properties.
-p. 7 lines 200-205: They state that "reorientational motion of MA is absent." in some systems but it is not clear why this is the case? -Several groups have investigated the effect of electric fields on 2D OIHP electronic properties but it's not clear to me how the measured values of MA dipole reorientation compare to those reported here. It would help to have a better contextual link here since it is not clear to me if this is due to the experimental probe influence or a natural property. Additionally, there is little connection to why the results are significant for a particular application or class of material. For example, how does the interplay between A-site and spacers potentially impact optoelectronic properties of 2D OIHP.
- Fig. 4 caption states the contents contain PL spectra but they only show temperature dependent XRD and NMR. Am I missing something? -p. 4 PEA was not defined earlier in the manuscript -p. 5 line 134: should "sorely" be "solely"?

Summary
The authors have studied the dynamics of the methylammonium ion in a number of perovskites, showing that the ion motion depends on nature of the solid phase and the constituents that make up the crystal. The NMR techniques employed required a lengthy synthetic effort. More sensitive results could have been obtained by using 2H NMR as specified in the detailed report and ref 11 of the manuscript with use of a partially deuterated compound. The authors also neglected to use available crystallographic data to good advantage. A good visual representation would have helped the reader in understanding the various structural features discussed in the paper.
General comments going through the text For obtaining information on the dynamics od the methylammonium ion the authors could have designed a far more effective approach. In Lines 93-107 the authors construct an argument to justify their experiments, saying that other techniques for studying the dynamics will fail because of spectral overlap. In fact the statement on lines 100-102 is nonsense -a 2H NMR lineshape study would tell them far more than their results. They could have looked at CD3NH3+ and CH3ND3+ (by D2O exchange) without all the effort and expense of producing 13C/N15 doubly labelled CH3NH3+. Simple exchange with D2O would also deuterate the NH3 groups of the spacer ions, but since the spacers are likely less dynamic than the MA ion their 2H lineshape will be considerably broader than that of MA. The authors the inadequacy of 2H and 14 N relaxation studies, but such studies would not be necessary. 2H NMr lineshapes as determined from quadrupolar interactions would give a Line 243: `… is spacer dependent`` --again i.e. structure dependent.
Line 254: The authors should mention the DSC result here, as it provides confirmation of the phase change and its temperature.
Line 259: conformational change -maybe, but might also be doubling of content of asymmetric unit in the crystal Line 270: How big is the shift?
Line 286-287: It isn't obvious how the author's results show that the structural deformation of the PbI6 layers -and again around line 327 .. There is kind of a circular argument here ---the single crystal structures will show variations in the similar, but not identical, PbI layers which affect the dynamics of the CH3NH3 ion enclosed within -so a change in MA dynamics indicates a change in its environment from one phase to another, but it does not tell you what that change in environment is. Figure 1(b): there does not appear to be an ``overlap problem`` between the MA line and the organic spacer lines for the PEA compound ! -but, granted, the isotopic labelling does greatly enhance the intensity. Figure 4: The caption mentions photoluminescence (PL) spectra but these are not in the figure -they have been moved to the supplemental section. I also note in Fig.4(d) that one of the 13C aromatic peaks of the PEA reduces quite dramatically as T increases -probably a sign of reorientation about the phenyl axis.
"XRD" results: The authors do not explain their "XRD" patterns (which ought to be labelled throughout as PXRD)the series of equally spaced peaks have been indexed and these reflections relate to the long axis of the unit cell. For both compounds they could at least give the derived axis parameter and compare it with the single crystal information as this would help confirm they have the right materials. They could also point out that these long axis parameters reflect the different spacings between the Pb2I7 layers (which they would have found out earlier if they had looked at the single crystal papers). From a rough measurement of the PXRD line spacings in figure S2 I calculate b=39.05 for the BA material and c=44.22 for the PEA material using their indexing numbers -note that the latter would correspond to a doubling of the unit cell given in the single crystal study (in line with the different Z values).
Supplementary data: Fig.S1: The PXRD patterns are not exactly the same as those shown in fig.4 of the main text -so are these at a different temperature, perhaps room temperature ? Do they have an explanation for the different intensity patterns between the natural abundance and labelled samples ?   The work by Lin et al. is a study of methylammonium dynamics in Ruddlesden-Popper lead halides based on phenylethylammonium (PEA) and butylammonium (BA). The authors use uniformly 15 N and 13 C labelled MA to investigate the dipolar coupling strength between the 15 N-13 C spin pair in three different materials and as a function of temperature. The authors record REDOR data which they use to conclude that the dynamics is qualitatively different in the PEA-and BA-based materials. The experimental data is of high quality. However, the analysis is relatively general and only focuses on qualitative differences between the MA dynamics in the different materials. That said, the authors have recorded quantitative REDOR data which should allow them to perform more in-depth analysis to obtain quantitative information on the spatial restriction and correlation time of the MA motion. This analysis will render the work substantially more prominent and of interest to the broad readership of Nature Commun. working in the fields of halide perovskite optoelectronics and solid-state NMR. There are also a number of highly relevant references which have been overlooked by the authors and should be acknowledged. This reviewer therefore recommends that the work should be considered for publication after the authors have addressed the following queries: "This result indicates that the MA in 2D (PEA)2([U-13C,U-15N]MA)Pb2I7 (n = 2) underwent a more restricted reorientational motion than the MA in 2D (BA)2([U-13C,U-15N]MA)Pb2I7 (n = 2), suggesting that the local environments for MA in these two materials were quite different." "The above results suggest that the reorientational motion of the A-site cations due to cooling in 2D OIHP crystals is spacer-dependent.
My main criticism regards the fact that the authors only draw qualitative conclusions regarding the MA dynamics (as illustrated by the above excerpts), despite having recorded high-quality data allowing them to perform quantitative analysis. The two key parameters characterizing dynamics are the order parameter and the correlation time, and the authors at present do not make an attempt to calculate and compare these parameters here for the three classes of materials under study.
In order for the dipolar coupling to be averaged out, the motion has to be isotropic (i.e. the order parameter has to vanish). When the authors refer to the motion as "more restricted" they imply that the order parameter is higher. Similarly, there is no attempt to calculate the correlation times. The discussion will benefit substantially from interpreting the results in the context of the model-free approach of Lipari and Szabo, see https://pubs.acs.org/doi/10.1021/ja00381a009 There are a number of experimental works using this framework for interpreting REDOR data (e.g. https://link.springer.com/article/10.1007/s10858-013-9787-x), which should greatly facilitate the task.
"Until now, only the dynamics of the organic spacers at the 2D OIHP crystals of (4NPEA)2PbI4 and (PEA)2PbI4 with n = 1, which consist of no A-site cations, have been analysed based on the temperature dependence of the ssNMR spectral line shapes " The authors should reference and acknowledge a study from 1996 which studied the dynamics of PEA in the n=1 material: Beyond dynamics, solid-state NMR studies of 2D OIHP have been rare so far with only a few reports so far. Acknowledging the above-mentioned works as well as the one below will provide a better representation of the current state of the art: https://pubs.acs.org/doi/abs/10.1021/acs.chemmater.0c04078 In addition, please define "4NPEA".
Regarding the assignment of the minor MA component in figure 1a: its chemical shift corresponds to unreacted MAI -could the author report the two compounds on the same scale and comment? This assignment seems more plausible than invoking structural defects. It is also consistent with the higher CP efficiency (MA in MAI is rigid).
Minor points: Please explain the notation: "U-" (for uniform labelling) Grammar/typos: line 134: "sorely" line 135: "the growth of 2D OIHP crystals consisting of the conventional naturally abundant MA as the A-site cation, were also synthesised"

Response letter for "Direct Investigation of the Reorientational Dynamics of A-site Cations in 2D Organic-Inorganic Hybrid Perovskite by Solid-State NMR"
Previous Manuscript ID : NCOMMS-21-03696 Below, we would like to address each of the reviewer's comments point-by-point. We prepared two versions of manuscripts, one clear version and the other with changes highlighted in red.

Comments:
Lin et al. investigate reorientational dynamics of methylammonium cations in 2D and 3D organic-inorganic halide perovskites (OIHP) using a specialized NMR probe (Rotational echo double resonance, REDOR) to interrogate the A site dynamics.
Related solid-state NMR methods have seen increased use in studies directed towards understanding how A site dynamics influence optoelectronic properties. Issues mentioned with related techniques concern the ability to resolve these dynamics due to overlapping transitions for organic spacers between perovskite octahedra. They also use an isotopic labeling approach to better resolve A-site dynamics. While I do not completely understand the REDOR technique it does appear that they are capable of reliably discriminating between A-site and organic spacer contributions to the NMR signals. The work appears well executed but the paper in its current form is difficult to follow because it is not intuitive for non-specialists. Furthermore, it is difficult to understand the significance of the results and how they pertain to particular perovskite structures. I believe the work is publishable but needs some revisions to improve the readability for readers to appreciate the results. Comments appear below.

Q1.
It would be extremely helpful if the authors included a cartoon of the structures they are discussing and the proposed dynamics being measured from experiment. While many understand the general perovskite structure it is harder to envisage lower dimensional systems and how they differ from conventional systems in both structure and properties.
Answer: Thanks for the reviewer's valuable suggestion. We have followed the reviewer's suggestion to add cartoon illustrations of 2D OIHP structural models with A-site cation undergoing reorientational motion. The cartoon illustration should be helpful for the readers to understand the structural differences between 2D layer perovskite and the conventional 3D perovskite.
Revision made: Fig. 1 has been modified in the revised manuscript.

Answer:
We have defined the PEA in introduction. This sentence "Ruddlesden-Popper perovskites are a typical example of layered 2D organic-inorganic hybrid perovskites having the generic chemical formula A′2An-1MnX3n+1. In this formula, A′ represents an organic spacer, such as long chain alkylammonium cation (e.g. 1-butylammonium, BA + ) or phenyl alkylammonium cation (e.g. 2-phenethylammonium, PEA + ), A is an organic cation, M is a metal, X is a halide, and n is the number of octahedral slabs per unit cell 25- 28 ." has defined PEA already.

Revision made:
The incorrect word "sorely" in line 134 has been replaced with the correct word "solely" in the revised manuscript.

Comments:
For obtaining information on the dynamics of the methylammonium ion the authors could have designed a far more effective approach. In Lines 93-107 the authors construct an argument to justify their experiments, saying that other techniques for studying the dynamics will fail because of spectral overlap. In fact, the statement on lines 100-102 is nonsense a 2H NMR lineshape study would tell them far more than their results. They could have looked at CD3NH3 + and CH3ND3 + (by D2O exchange) without all the effort and expense of producing 13 C/N 15 doubly labelled CH3NH3 + .
Simple exchange with D2O would also deuterate the NH 3 groups of the spacer ions, but since the spacers are likely less dynamic than the MA ion their 2 H lineshape will be considerably broader than that of MA. The authors the inadequacy of 2 H and 14 N relaxation studies, but such studies would not be necessary. 2 H NMR lineshapes as determined from quadrupolar interactions would give a quantitative description of the motion of the C-N axis. The authors mention that the dynamics change, but do not provide a quantitative description.

Answer:
We would like to thank the reviewer for the valuable comments and suggestions. REDOR NMR measures the dipolar coupling between the 13 C and 15 N nuclei of MA, which is modulated by the molecular reorientational motion of MA. Thus, we claimed that REDOR NMR measurement directly reflects this motional averaging process. We agree that the same molecular motion also averages the deuterium quadrupole coupling and, as a result, 2 H NMR lineshape also reflects the motion of the 2 H spin. In the case of using CD3NH3 + to replace the natural abundance MA in the synthesis of 2D OIHPs, the quadrupole splitting (νQ), measured cusp to cusp of the 2 H spectrum, would be reduced from the static value of ~120 kHz to ~40 kHz in the presence of rapid C3 rotation, and the existence of additional reorientational motion of the C3 axis, the C-N vector in this case, would result in narrower spectral lineshape.
Thus, even though it is not a direct observation of the motion of the C-N vector, 2 H NMR lineshape does allow one to learn valuable information about the dynamics of the C-N vector. In this regard, we thank the reviewer's suggestion and happily performed additional experiments to record 2 H NMR spectra of 2D (BA)2(d3-MA)Pb2I7 (n = 2) and 2D (PEA)2(d3-MA)Pb2I7 (n = 2) in our study, Fig. 4, where CD3NH3 + was used to replace the natural abundance MA in the synthesis of 2D OIHPs. We indeed learned quite a lot of valuable information from the additional experimental results and found that together with REDOR NMR and 2 H NMR, we can get more complete dynamics 2) and to study its dynamics. In addition, we observed a 4% change of 13 C{ 15 N}REDOR(ΔS/S0)2.4ms value of 2D (PEA)2( 13 C, 15 N-MA)Pb2I7 (n = 2), as shown in Fig. 3, when the temperature was cooled from 298 to 243K, while very little 2 H NMR lineshape change of 2D (PEA)2(d3-MA)Pb2I7 (n = 2) was observed in the same temperature range, as shown Fig. 4. Finally, we agree that the isotopically labelling material 13 C, 15 N-MA is not cheap. Thus, in this manuscript we would like to share with the community an inexpensive synthesis method for producing it.
Revision made: We thank the reviewer for reviewing our manuscript carefully and providing many valuable comments and suggestions. We took the reviewer's suggestion to include additional 2 H NMR spectra of 2D OIHPs prepared with CD3NH3 + , as shown in Fig. 4. Accordingly, we have included the description of the line shape analysis of the 2 H NMR spectra in the Results section, as well as in the Discussions section.
In addition, we have also revised the abstract as follows.
"Limited methods are available for investigating the reorientational dynamics of A-site cations in two-dimensional organic-inorganic hybrid perovskites (2D OIHPs), which play a pivotal role in determining their physical properties. Here we describe a novel approach to study the dynamics of A-site cations using solid state NMR and stable isotope labelling. 2 H NMR of 2D OIHPs incorporating methyl-d3-ammonium cations Answer: We answer this question together with the Q7 raised by the reviewer #2.
Thanks for the reviewer's valuable suggestion. We have followed the reviewer's suggestion to add cartoon illustrations of 2D OIHP structural models with A-site cation undergoing reorientational motion. The cartoon illustration should be helpful for the readers to understand the structural differences of 2D layer perovskite and the conventional 3D perovskite.
We also agree with the suggestion to cite the papers of the single crystal X-ray structure of 2D (BA)2MAPb2I7 and (PEA)2MAPb2I7 in the manuscript. These include references 32, 47-50 in the revised manuscript, and the more recent works (Chem.  Answer: We answer this question together with the Q11 raised by the reviewer #2 and the Q5 raised by the reviewer #3. We have plotted the 13 C NMR spectra of 2D Ideally, one should be able to construct a model to analyze both 2 H NMR line shapes and REDOR dephasing curve to extract the motional parameters quantitatively, provided the distribution of the C-N vector of MA relative to the PbI6 octahedral unit is known. The different orientation preferences of MA may be associated with the different motional modes. However, unambiguous information is not available, preventing us from doing reliable analyses to extract the motional correlation times.
The REDOR dephasing data at 2.4 ms can be correlated with the order parameters averaging over the different motional modes of the incorporated MA. The dynamics of the incorporated MA in BA-and PEA-based 2D OIHPs respond very differently to the temperature change. This main discovery is sufficiently supported with the provided evidence. We also like to thank the reviewer for pointing out some of the key references.
They are all properly cited in the revised manuscript.

Revision made:
We have included the suggested key references in the revised manuscript. These include the two references of the model-free approach 53 In addition to citing more references, we included the theoretical description of REDOR NMR in the presence of molecular motion to establish the relationship between the REDOR dephasing value and the order parameter, characterizing the reorientational motion. Moreover, we included the 2 H NMR experiments kindly Answer: We would like to express our gratitude to the reviewer for making valuable suggestions to include 2 H NMR experiments. As mentioned in our response to the major comment raised by the reviewer, we agree that the molecular reorientational motion also averages the deuterium quadrupole coupling and, as a result, 2 H NMR lineshape also reflects the motion of the 2 H spin. In the case of using CD3NH3 + to replace the natural abundance MA in the synthesis of 2D OIHPs, the quadrupole splitting (νQ), measured cusp to cusp of the 2 H spectrum, would be reduced from the static value of ~120 kHz to ~40 kHz in the presence of rapid C3 rotation, and the existence of additional reorientational motion of the C3 axis, the C-N vector in this case, would result in narrower spectral lineshape. Thus, even though it is not a direct observation of the motion of the C-N vector, 2 H NMR lineshape does allow one to learn valuable information about the dynamics of the C-N vector. In this regard, we thank the reviewer's suggestion and happily performed additional experiments to record 2 H NMR spectra of 2D (BA)2(d3-MA)Pb2I7 (n = 2) and 2D (PEA)2(d3-MA)Pb2I7 (n = 2) in our study, Fig. 4, where CD3NH3 + was used to replace the natural abundance MA in the synthesis of 2D OIHPs. We indeed learned quite a lot of valuable information from the additional experimental results and found that together with REDOR NMR and 2 H NMR, we can get more complete dynamics information of MA in 2D OIHPs. We have included the most valuable additional information learned from 2 H NMR lineshapes of 2D OIHPs, which is the existence of multiple motional modes of MA.

Revision made:
We took the reviewer's suggestion to include 2 H NMR spectra of 2D OIHPs prepared with CD3NH3 + . Accordingly, the last paragraph of the introduction has been revised as described in our response to the major comment of the reviewer. In addition, we included the description of the 2 H NMR results in the experimental section and discussed the results in the discussion section accordingly. Answer: Thanks for the reviewer's comment. We carefully examined the data reported in reference 11, where we also observed minor 2 H NMR lineshape change of CH3ND3PbI3 in the temperature range of 294 to 237 K, which is comparable with the temperature range used in our experiment from 308 to 243K. Moreover, we have included 2 H NMR lineshape studies (Fig. 4) of both 2D (BA)2(d3-MA)Pb2I7 (n = 2) and 2D (PEA)2(d3-MA)Pb2I7 (n = 2). We have discussed the comparison of these two ssNMR methods in our response to the major comment. Briefly, according to our results, these are two complementary ssNMR methods. In the case of 2D (PEA)2(d3-MA)Pb2I7 (n = 2), very little 2 H NMR lineshape change was observed when the temperature was cooled from 298 to 243K (Fig. 4), while 4% change of 13 C{ 15 N}REDOR(ΔS/S0)2.4ms value of 2D (PEA)2(d3-MA)Pb2I7 (n = 2) was measured during the cooling from 298 to 243K, as shown in Fig. 3. According to our results, the 2 H NMR lineshape analysis uniquely revealed the existence of the multiple motional modes. On the other hand, REDOR NMR measurements reflected the change of the averaged MA dynamics in response to the temperature change.
Revision made: The revision has been described in the response to the major comment of the reviewer, we have included the additional results of 2 H NMR line shape analysis and discussed them accordingly in the revised manuscript.

Q11.
Lines 217-219: Again this suggests that the minor component is due to an impurity phase.

Answer:
We have answered this question together with the Q5 raised by the reviewer #2 and the Q5 raised by the reviewer #3. We have plotted the 13 C NMR spectra of 2D  (BA)2( 13 C, 15 N-MA)Pb2I7 (n = 2) as shown in Fig. S2. Therefore, the minor component (27.2 ppm peak) might more likely be structural impurity (eg. one PbI6 octahedral unit missing or one spacer missing) according to recent work (Nat. Nanotechnol 15, 969 (2020)). The amount of the minor MA component was estimated to be only 3% of the total amount of MA and its existence does not influence the conclusion of this work.

Revision made:
The revision is described in the revision for the Q5 of the reviewer #2.
Q12. Line 227: as found in the crystal structure papers.
Answer: Thanks for the reviewer's suggestion. We followed the reviewer's suggestions. Consequently, the reorientational dynamics of the A-site cation were found to stay relatively unchanged in the PEA-containing 2D OIHPs." Q14. Line 243: `… is spacer dependent`` --again i.e. structure dependent.

Answer:
We totally agree with the reviewer's interpretation. Here, we answer this question together with Q13 above. Different types of organic spacers incorporated in 2D OIHPs may lead to different crystal structures, and subsequently cause the differences in molecular motion of the A-site cations.

Revision made:
The revision is described in the response to Q13.
Q15. Line 254: The authors should mention the DSC result here, as it provides confirmation of the phase change and its temperature.
Answer: Thanks for the reviewer's suggestion. We followed the reviewer's suggestions to include the DSC results, as shown in Fig. S6. The endothermic peak occurred around 280K in the DSC measurement of 2D BA2MAPb2I7 (n = 2) provides clear evidence of phase change.

Revision made:
We included the DSC results as shown in Fig. S6 and pointed out that phase change of 2D OIHP BA2MAPb2I7 (n = 2) occurred around 280K in the revised manuscript. In addition, the following description and the discussion of the DSC results have been included in the revised manuscript.
"Differential scanning calorimetry (DSC) and PXRD were further used to study the structural response of 2D (BA)2(MA)Pb2I7 (n = 2) and 2D (PEA)2(MA)Pb2I7 (n = 2) during the cooling process. The endothermic peak observed in the DSC measurement of 2D (BA)2(MA)Pb2I7 (n = 2) indicated a clear phase change occuring at approximately 280 K (Fig. S6). This echoes the finding in a recent study 32 , showing the associated phase change was from Cmcm to P-1 space group. In contrast, no evidence of phase change was found in the DSC measurement of (PEA)2(MA)Pb2I7 (n = 2). Moreover, based on the comparison of PXRD patterns recorded at 300K and 250K (Fig. 5(c) and 5(d)) a clear structural change for (BA)2(MA)Pb2I7 (n = 2) was observed with cooling from 300K to 250 K, while no change was observed for the (PEA)2(MA)Pb2I7 (n = 2)." Q16. Line 259: conformational change -maybe, but might also be doubling of the content of asymmetric unit in the crystal.
Answer: Thanks for the reviewer's comment. Our data interpretation is consistent with recent works (Chem. Mater 31, 5592-5607 (2019) and Chem. Mater 33, 3524-3533 (2021)). Regarding the responses of the materials to the temperature change, we agree that the BA-based 2D OIHP is going through a phase transition, as suggested by the reviewer, based on the NMR results together with the DSC and PXRD results. Regarding the interpretation of the dynamics change of MA, we have answered it in our response to Q6. Figure 1(b): there does not appear to be an ``overlap problem`` between the MA line and the organic spacer lines for the PEA compound! -but, granted, the isotopic labelling does greatly enhance the intensity.

Q19.
Answer: Here, we would like to emphasize the clear overlap problem in the 13 C NMR spectrum 2D OIHP perovskite with BA spacer. Without tackling this problem, we will not be able to compare the materials made by BA spacer with the one made by PEA spacer. Yes, the signal intensity is greatly enhanced by using the isotopic labelling as mentioned by the reviewer.

Revision made:
No additional revision made.
Q20. Figure 4: The caption mentions photoluminescence (PL) spectra but these are not in the figure -they have been moved to the supplemental section. I also note in Fig.4(d) that one of the 13C aromatic peaks of the PEA reduces quite dramatically as T increases -probably a sign of reorientation about the phenyl axis.
Answer: Thanks to the reviewer for pointing out the inconsistency between the caption and the figure. Fig. 4 in the previous version is now Fig. 5 in the revised version. We have revised the caption of Fig. 5 and included the PL spectra in Fig. S7. We also thank the reviewer for pointing out that the dramatic PEA signal reduction as the temperature increase may be due to the phenyl ring reorientation about the phenyl axis.

Revision made:
We have revised the caption of Fig. 5 and included the PL spectra in We have also included the following sentence in the revised manuscript.
"A signal reduction of a phenyl carbon resonance peak at 130 ppm with increasing temperature may be due to the flipping of the phenyl ring." Q21. "XRD" results: The authors do not explain their "XRD" patterns (which ought to be labelled throughout as PXRD)-the series of equally spaced peaks have been indexed and these reflections relate to the long axis of the unit cell. For both compounds they could at least give the derived axis parameter and compare it with the single crystal information as this would help confirm they have the right materials. They could also point out that these long axis parameters reflect the different spacings between the Pb2I7 layers (which they would have found out earlier if they had looked at the single crystal papers).  (PEA)2( 13 C, 15 N-MA)Pb2I7 (n = 2) recorded at various temperatures, ranging from 308 to 243K. The PXRD patterns of (c) 2D (BA)2MAPb2I7 (n = 2) and (d) 2D (PEA)2MAPb2I7 (n = 2) measured at 300K and 250K. Fig.S1: The PXRD patterns are not exactly the same as those shown in fig.4 of the main text -so are these at a different temperature, perhaps room temperature? Do they have an explanation for the different intensity patterns between the natural abundance and labelled samples? Answer: Thanks for the reviewer's comment. Because the instrument used for the temperature-dependent PXRD measurements is different from the one used for the room temperature PXRD measurements (as shown in Fig.S1). As a result, the peak intensities of the two PXRD spectra are not exactly the same. The temperaturedependent PXRD measurement using a Bruker D8 Discover X-ray diffraction system with Cu-Kα radiation in Bragg-Brentano geometry and equipped with a temperature control system (77-350K). Accordingly, we have revised the SM2 to include the details of the Materials Characterisations. Even though the intensities of the PXRD spectra of (BA)2( 13 C, 15 N-MA)Pb2I7 (n = 2) has shifted from 31.2 to 29.8 ppm. We suppose that the reviewer meant that the intensity of the resonance peak of the incorporated MA in 2D OIHP (BA)2( 13 C, 15 N-MA)Pb2I7 (n = 2) decreased as the temperature increased. All the experiments in the various temperatures were performed with the same crosspolarization parameters. We believe the signal decrease was due to the crosspolarization efficiency change due to the change in the reorientational motion.

Revision made:
The following sentence has been added in the caption of Fig. S5, which was Fig. S3 in the previously submitted version.
"The resonance peak of MA was found to have an upfield shift from 31.2 to 29.8 ppm with the temperature decreased from 308 to 243K." Q24. Fig. S4: The DSC (is this warming or cooling ?) very clearly shows a phase transition only for the BA compound, onset ~280K. This ought to be stated directly in the main text.
Answer: The DSC measurements in Fig. S6, which is the Fig. S4 in the version submitted previously, were recorded with the heating from 225 to 300 K. As the reviewer pointed out, obvious phase change is shown in the DSC heating curve of 2D OIHP BA2MAPb2I7 (n = 2) in Fig. S6. Accordingly, we have revised our manuscript to include this data interpretation.

Revision made:
We have included the experimental parameters of the DSC measurement in the SM2 as follows.
"The DSC heating curves were collected using a TA Q200 thermal analysis instrument at a scan rate of 5K min -1 and heated from 225 K to 300 K in sealed aluminum pans under ambient conditions." In addition, the following text has been included in the revised manuscript to point out the occurrence of phase transition in the 2D OIHP BA2MAPb2I7 (n = 2).
"The endothermic peak observed in the DSC measurement of 2D (BA)2(MA)Pb2I7 (n = 2) indicated a clear phase change occuring at approximately 280 K (Fig. S6). This echoes the finding in a recent study 32 , showing the associated phase change was from Cmcm to P-1 space group. In contrast, no evidence of phase change was found in the DSC measurement of (PEA)2(MA)Pb2I7 (n = 2)."

Comments:
The work by Lin et al. is a study of methylammonium dynamics in Ruddlesden-Popper lead halides based on phenylethylammonium (PEA) and butylammonium (BA). The authors use uniformly 15 N and 13 C labelled MA to investigate the dipolar coupling strength between the 15 N-13 C spin pair in three different materials and as a function of temperature. The authors record REDOR data which they use to conclude that the dynamics are qualitatively different in the PEA-and BA-based materials. The experimental data is of high quality. However, the analysis is relatively general and only focuses on qualitative differences between the MA dynamics in the different materials. That said, the authors have recorded quantitative REDOR data which should allow them to perform a more in-depth analysis to obtain quantitative information on the spatial restriction and correlation time of the MA motion. This analysis will render the work substantially more prominent and of interest to the broad readership of Nature Commun. working in the fields of halide perovskite optoelectronics and solid-state NMR. There are also a number of highly relevant references which have been overlooked by the authors and should be acknowledged. This reviewer therefore recommends that the work should be considered for publication after the authors have addressed the following queries: "This result indicates that the MA in 2D (PEA)2([U-13 C,U-15 N]MA)Pb2I7 (n = 2) underwent a more restricted reorientational motion than the MA in 2D (BA)2([U-13 C,U-15 N]MA)Pb2I7 (n = 2), suggesting that the local environments for MA in these two materials were quite different." "The above results suggest that the reorientational motion of the A-site cations due to cooling in 2D OIHP crystals is spacer-dependent.

Q1.
My main criticism regards the fact that the authors only draw qualitative conclusions regarding the MA dynamics (as illustrated by the above excerpts), despite having recorded high-quality data allowing them to perform quantitative analysis. The two key parameters characterizing dynamics are the order parameter and the correlation time, and the authors at present do not make an attempt to calculate and compare these parameters here for the three classes of materials under study.

Answer:
We thank the reviewer's comment and answer this question together with the Q8 raised by the reviewer #2. We took the suggestion to record the 2 H NMR spectra of 2D (BA)2(d3-MA)Pb2I7 (n = 2) and 2D (PEA)2(d3-MA)Pb2I7 (n = 2). The reorientational motion of MA provides motional averaging to the deuterium quadrupolar coupling of CD3NH3 + , as well as to the dipolar coupling between the 13 C and 15 N spins of 13 C, 15 N-MA. Thus, we should be able to extract the information of the reorientational motion from both the 2 H NMR lineshape analysis and REDOR NMR. Our results indicated that these are the two complementary methods. We observed the multiple double-horned quadrupolar splitting patterns in the 2 H NMR spectra, suggesting the existence of multiple motional modes of MA incorporated in 2D OIHPs. On the other hand, REDOR NMR, exhibiting the high resolution power of MAS, revealed the existence of the 3% of the minor MA component in 2D (BA)2( 13 C, 15 N-MA)Pb2I7 (n = 2). Ideally, one should be able to construct a model to analyze both 2 H NMR line shapes and REDOR dephasing curve to extract the motional parameters quantitatively, provided the distribution of the C-N vector of MA relative to the PbI6 octahedral unit is known. The different orientation preferences of MA may be associated with the different motional modes. However, unambiguous information is not available, preventing us from doing reliable analyses to extract the motional correlation times.
The REDOR dephasing data at 2.4 ms can be correlated with the order parameters averaging over the different motional modes of the incorporated MA. The dynamics of the incorporated MA in BA-and PEA-based 2D OIHPs respond very differently to the temperature change. This main discovery is sufficiently supported with the provided evidence. We also like to thank the reviewer for pointing out some of the key references.
They are all properly cited in the revised manuscript.

Revision made:
We have included the suggested key references in the revised manuscript. These include the two references of the model-free approach 53,54 , the REDOR derived order parameter, the previous studies of the spacer dynamics 28,41,42 and relevant solid-state NMR works 43 , listed below.
In addition to citing more references, we included the theoretical description of REDOR NMR in the presence of molecular motion to establish the relationship between the REDOR dephasing value and the order parameter, characterizing the reorientational motion. Moreover, we included the 2 H NMR experiments kindly suggested by the reviewer #2 for comparison. These changes are highlighted in our revised manuscript.