I found this to be a very nice paper by Gomez et al. comparing the efficiency of SQ and ZQ symmetry-based polarization transfer sequences at moderate and high MAS rates. The search for optimal sequences was thorough, which adds confidence in the results. I have a few minor corrections and comments that I will outline below.

1) Line 99 – Although the sentence is correct, it sounds as though the author is trying to say that the rf power increasing with homo-nuclear dipolar interactions makes it unsuitable for high MAS rates. I think it should be split into two sentences. We have changed this.

2) Line 112 – Why define R’? The phase phi is defined and R with a phase of -phi is sufficient. This has been done.

3) Line 137 – Please include a reference for dipolar truncation. This has been done. N.C. Nielsen, H. Bildsoe, H.J. Jakobsen, M.H. Levitt, J. Chem. Phys., 101 (1994) 1805; C. Martineau, B. Bouchevreau, F. Taulelle, J. Trébosc, O. lafon, J.-P. Amoureux, Phys. Chem. Chem. Phys., 14 (2012) 7112-7119.

4) Line 164. The reason given for screening the phase of the sequence to those near 90° is that it is what is used, but the authors could be more specific that these sequences are better compensated for rf inhomogeneity due to the 180° phase difference between the two pulses. This has been done.

5) Line 264 – This statement may be misleading.  R12⁵₉ has a non-zero scaling factor when the rf field is greater than 2n_R. Indicated.

6) Figs.2e and f are distorted. They have been changed.

7) My last comment has to do with the generality of the results being presented and perhaps the authors might want to look into this point more deeply. As the authors have demonstrated in an earlier JACS paper, the INEPT sequences can outclass PRESTO when the dipolar couplings are weak, for instance with low-g nuclei, since the ZQ sequences can sustain longer recoupling times. We have also noticed that PRESTO-II far surpasses INEPT when the dipolar couplings are very strong (since it enables for very short non-synchronized recoupling periods, presumably), and so there might be some cross-over point in dipolar coupling strength when INEPT surpasses PRESTO. ²⁷Al is likely near this cross-over point. So likely, the best RINEPT sequences will remain the best, and the same for the PRESTO sequences, but the comparison of the two types of sequences may be very dependent on the dipolar coupling strength and the time required to recouple the interaction. We disagree to this point. Indeed, we have published a RINEPT-II version, which has been shown to be more efficient than PRESTO-II for large dipolar interactions (e.g. OH groups). Ref: Nagashima et. al, Magnetic Resonance in Chemistry, In press, https://doi.org/10.1002/mrc.5121.

There has been an increasing interest in CPMAS from ¹H to half-integer quadrupolar nuclei in order to materialize the high ¹H DNP enhancement through efficient spin diffusion for detecting quadrupolar nuclei. This work is practically useful to this effort in addition to the proximity information by optimizing the pulse sequences. The authors have chosen g-alumina as the sample system, which has a fairly large shift dispersion at 800 MHz field. The proton homo-nuclear coupling is relatively weak as compared to rigid organic molecules. They have also chosen 20 and 62.5 kHz spinning speeds and typical rf-fields that 3.2 and 1.3 mm probes can achieve. From previous publications, the authors also have added adiabatic pulses to improve the offset performance and some modifications to the blank rotor period to reduce the ¹H T₂’ loss. There is no doubt that they have improved the performances, but I am not sure if this optimization is generally applicable to other samples and spin systems. My particular concern is the strong ¹H homo-nuclear coupling. It is correct that the first-order ¹H homo-nuclear term is not 'recoupled' for the selected hetero-nuclear recoupling sequences. What about the high-order homo-nuclear terms that cause the short T₂’, especially at not-so-high spinning speeds? My observation is that ¹H T₂’ under the recoupling is often very short, and thus can have a strong effect onto the performances.

8) I wonder if the authors could provide some ¹H T₂’ information under the recoupling sequences, say as compared to a regular spin-echo T₂’ decay. We fully agree with Zhehong on this point. We have introduced the T₂’ results of these experiments at 62.5 kHz MAS. These values are very interesting, and they highlight many previous experimental limitations we have observed.  

9) In addition, is it possible to program the experiment to avoid the blank rotor period to rotor during the CT p-pulse by simultaneous pulses.

The p-pulses have two purposes: eliminate the ¹H CSA and keep the ¹H-S dipolar coupling. In 2007, we have proposed with Zhehong several methods in JMR, to avoid these p-pulses. However, it was for D-HMQC or RINEPT sequences with R³ recoupling, and ‘inverting the phase of R³ pulses also reverse the sign of the average Hamiltonians’. This led to sequences in Fig.1d and f of this article. Here, we recouple with SR4²₁ which is an I_zS_z operator, and thus inverting the phases should not change anything. On the contrary, this change of phase is used with PRESTO, which has a transverse operator.    

1. Trébosc, B. Hu, JP. Amoureux, Z. Gan, Through-space R³-HETCOR experiments between spin-1/2 and half-integer quadrupolar nuclei in solid-state NMR, J. Magn. Reson, 186 (2007) 220-227.

We have also shown in Nagashima et. al, 2021, that it is impossible to avoid the blanks close to the ¹H p/2 and p-pulses, except with very large rf-fields that are not accessible with conventional MAS probes.

10) I would be interested in the overall transfer efficiency. It seems very low based on the S/N. Of course, only a small portion of g-alumina has ¹H in its vicinity. It would help to know where the main loss comes from, for future improvement. We have not quantified this efficiency, because it is uneasy to define. However, we remind here that we are using a 1.3 mm rotor with ca. only 2.5 mL of active volume, and we show now in Fig.9 a 2D HETCOR spectrum with high S/N recorded in only 72 min.

11) The first citation on PRESTO, p2 line-66, should include the paper from the original paper from Levitt's group. Xin Zhao, Wilfried Hoffbauer, Jörn Schmedt auf der Günne, Malcolm Levitt, Solid State Nucl. Magn. Reson., 26(2) (2004) 57-64, doi: 10.1016/j.ssnmr.2003.11.001 Done.

I would like to recommend this work on the through description of symmetry based recoupling and search for recoupling sequences that are optimal for CPMAS with practically feasible spinning speed and rf field for 3.2 and 1.3 mm probes.