Infrared spectroscopic study of hydrogen bonding topologies in the smallest ice cube

The water octamer with its cubic structure consisting of six four-membered rings presents an excellent cluster system for unraveling the cooperative interactions driven by subtle changes in the hydrogen-bonding topology. Despite prediction of many distinct structures, it has not been possible to extract the structural information encoded in their vibrational spectra because this requires size-selectivity of the neutral clusters with sufficient resolution to identify the contributions of the different isomeric forms. Here we report the size-specific infrared spectra of the isolated cold, neutral water octamer using a scheme based on threshold photoionization using a tunable vacuum ultraviolet free electron laser. A plethora of sharp vibrational bands features are observed. Theoretical analysis of these patterns reveals the coexistence of five cubic isomers, including two with chirality. The relative energies of these structures are found to reflect topology-dependent, delocalized multi-center hydrogen-bonding interactions. These results demonstrate that even with a common structural motif, the degree of cooperativity among the hydrogen-bonding network creates a hierarchy of distinct species. The implications of these results on possible metastable forms of ice are speculated.

8. Page 7. The authors state "Under the pulsed supersonic expansion condition in the present work, the presence of all five cubic isomers is quite surprising, indicating that our VUV-FEL spectroscopic technique is apt to explore low-lying neutral isomers unknown before". Why is this surprising that these higher energy isomers are seen? This is common in expansion studies of clusters. Also, these isomers were not unknown before. All of thenm are discussed in the paper of Tsai and Jordan. 9. Page 7, Last paragraph. The sentence about the population of higher lying isomers and the Boltzmann distribution, by itself is not useful. One also needs to know the temperature. For cold, equilibrium clusters one would not see these higher lying isomers. The authors must be seeing these because quenching in their experiment produces a non-equilibrium distribution.
10. Page 8. It makes no sense to report distributions up to 1000K at temperatures above about 300K, the clusters will evaporate on the time scale of the experiment. 11. Page 9. I believe that it is stretching things to call the bonding in these clusters aromatic.
12. Page 9. The authors state "The five water octamer isomers adopting pseudo-cubic structure is highly remarkable". I don't see what is remarkable about this. The cubic structures maximize the number of H bonds. Plus, as noted above, all five of these structures were predicted in theoretical studies.
13. Page 10. The authors again refer to the unexpected coexistence of several isomers. Again, I do not believe this is unexpected.
14. Page 11. The authors conclude with a reference to crystallization of ice. But they don't make a case that the octamer is relevant to such crystallization.
Reviewer #2 (Remarks to the Author): This paper reports an important investigation of a water cluster of central importance in untangling details of the solid to liquid transition in water. The use of a sophisticated new technology reveals a much higher degree of complexity in the energetically available isomer distribution than previous infrared spectroscopic studies of the octamer have shown. The accompanying theoretical calculations permit a reasonably credible assignment of the observed spectral features to particular isomers. This work is likely to stimulate further investigations into the remarkably complicated nature of the octamer.
Both the experimental and theoretical work seems to have been done at a high level, and the paper is written in an interesting fashion(although it could benefit from some preening of the English). I am thus pleased to recommend in favor of publishing this paper, essentially without change, except for the above comment(at the option of the authors), and the addition of two new references: An updated reference on the study of water clusters should appear as We are very grateful to the critical comments and constructive suggestions provided by the two reviewers, which have significantly helped us to improve the manuscript. Our manuscript has been revised accordingly, and the changes are highlighted by blue color in the text. The following lists our responses (in blue color) to the comments from the two reviewers.  Response: We agree with the referee that while water is a chemical complex, the water cluster is better not called a "complex". Therefore, the "complex" word has been deleted.

Page 3. Right after mentioning the octamer, the authors state "Experiments strongly suggest the presence of ice nanocrystals". What is the connection between ice nanocrystals and the octamer?
Response: Our statement was based on the following: The low-energy structures of the water octamer were predicted to be nominally cubic, with the eight tri-coordinated water molecules taking up positions at the corners of the cube. Such tri-coordinated water molecules have been identified at the surface of ice.
Given the referee's comment, we replaced "Experiments strongly suggest the presence of ice nanocrystals" by the more detailed description in the revised manuscript.

Page It is not clear what is meant by "diverse component"
Response: The "diverse component" has been changed to "disordered component". Response: We agree that the original statement is inaccurate. Therefore, the sentence "Multiple coexisting cubic octamers provide a coherent picture of structural diversity of bulk water and a cluster-scale precursor to the phase transition between solid and liquid water" has been removed from the manuscript.

MP2 or DFT calculations of vibrational frequencies do not assume an implicit description of ZPE.
3 Response: The referee is certainly correct; our sentence is confusing, and "implicit description of intermolecular zero-point motions" has now been changed to "neglection of intermolecular zeropoint motions". 6. Page 6, "Each structure of isomers I−V possesses" should be reworded as "Each of the isomers I−V possesses".

Response:
The "Each structure of isomers I−V possesses" has been changed to "Each of the isomers I−V possesses". Thanks!

Page 7 "analogous geometries" should be reworded "similar geometries"
Response: The "analogous geometries" has been changed to "similar geometries".

Page 7. The authors state "Under the pulsed supersonic expansion condition in the present work, the presence of all five cubic isomers is quite surprising, indicating that our VUV-FEL
spectroscopic technique is apt to explore low-lying neutral isomers unknown before". Why is this surprising that these higher energy isomers are seen? This is common in expansion studies of clusters. Also, these isomers were not unknown before. All of thenm are discussed in the paper of Tsai and Jordan.
Response: Sorry for missing noting the reference. Given this fact, the sentence "Under the pulsed supersonic expansion condition in the present work, the presence of all five cubic isomers is quite surprising, indicating that our VUV-FEL spectroscopic technique is apt to explore low-lying neutral isomers unknown before" has been removed from the manuscript.

Page 7, Last paragraph. The sentence about the population of higher lying isomers and the
Boltzmann distribution, by itself is not useful. One also needs to know the temperature. For cold, equilibrium clusters one would not see these higher lying isomers. The authors must be seeing these because quenching in their experiment produces a non-equilibrium distribution.
4 Response: We agree with the reviewer's comment. Quenching in our experiment produces most likely a non-equilibrium distribution, which benefits to the observation of all five cubic isomers.
Such discussion has been added into the manuscript.
10. Page 8. It makes no sense to report distributions up to 1000 K at temperatures above about 300K, the clusters will evaporate on the time scale of the experiment.
Response: Thanks for the comments. The temperature range has been changed to be 0−300 K in the text and figure. 11. Page 9. I believe that it is stretching things to call the bonding in these clusters aromatic.
Response: Thanks for the comments. We have removed this part about the bonding in these clusters being aromatic.
12. Page 9. The authors state "The five water octamer isomers adopting pseudo-cubic structure is highly remarkable". I don't see what is remarkable about this. The cubic structures maximize the number of H bonds. Plus, as noted above, all five of these structures were predicted in theoretical studies.
Response: Thanks for the comments. We intended to say that a non-planar cubic structure looks remarkable. Indeed it is a result of maximizing the number H-bonds, as is predicted theoretically.
We therefore revised this sentence.

Page 10. The authors again refer to the unexpected coexistence of several isomers. Again, I do not believe this is unexpected.
Response: Thanks for the comments. The "unexpected" word has been removed from the manuscript.
14. Page 11. The authors conclude with a reference to crystallization of ice. But they don't make a case that the octamer is relevant to such crystallization.
Response: Thanks for the comments. Nanometer-sized water clusters are engaged in a range of processes of cloud, aerosol, and ice formation, especially under rapid cooling . We agree that it is hard to make a clear case that the octamer is directly relevant to crystallization of ice.
Nevertheless, our observation of five water octamer cubes suggest that it cannot be fully excluded that there may exist a family of microcrystalline structures in ice formation that is yet to be fully