Extending Libraries of Extremely Localized Molecular Orbitals to Metal Organic Frameworks: A Preliminary Investigation

Round 1

Reviewer 1 Report

In this paper the authors exploit ELMOdb sofware to test the viability of a QM/ELMO description of a well-known MOF structure.

The introduction is thorough. The methology proposed is fairly simple. Obtain localized orbitals for the linker, then use QM/ELMO to calculate the MOF unit, and then transfer thre results to a crystal structure.

In my opinion there are some points the authors are not discussing in proper detail. The authors compare in some detail the full DFT and QM/ELMO results for the HKUST-1 unit. In particular, NCI, topological parameter on some critical points fo the density, delocalization indices and electrostatic potential for three diferent electronic states of the Cu-Cu core. Why not energies? Probably, DFT and QM/ELMO energies are not comparable, but certainly, energy differences between the three states can and should be reported and discussed.

In addition, there are very significant differences in the DI for Cu-Cu and to some extent Cu-O in the spin-coupled (singlet state). These can not be adscribed merely to "delocalization effects". It is clear from the results that that QM/ELMO does not permit charge-transfer between the fragments, but why such large differences in the QM region for Cu-Cu? On the other hand, DI accounts for a covalent bond order, therefore the more polarization of the bond the smaller the bond order.  In the case of Cu-O, the expected trend is boserved but again this can not explain the huge effect on Cu-Cu, which is in the QM part.

Another minor point, why using 6-311G(2d,2p) basis set when, as stated in the paper,  there is a databank of many fragments and fucntional groups that could be used for other five basis set, not including this particular one?

There is something else concerning Figure 1 the authors should clarify. It appears, fromt he Figure, that the fragmentation of the molecule is not exhaustive, meaning that the "sum of fragments" is not the total molecule, e. g. setting a particular atom as fragment, and then a bond where this same atom is contributing as another fragment. Is that possible within the ELMO-SCF approach? 

After these poitns are properly addressed, the paper should be publishable.

There are a number of typos:

anti-ferromagenric, calcualtions, calaculations,...

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

This review is a nice research of the possibility of using the ELMOs trans- 14

ferability in order to obtain wavefunctions, electron densities, and electrostatic potentials of 15

three-dimensional coordination polymers such as metal organic frameworks (MOFs).

 

Reccomendation: ACCEPT IN CURRENT form

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

It can be read from this paper that the usefulness of a method for rapidly acquiring the wave function, electron density, and electrostatic potential of a three-dimensional coordination polymer using the ELMOs transferability is very interesting and promising. Choosing the metal-organic framework HKUST-1 for benchmark testing was also appropriate. I will not comment about detailed calculations, etc., as my specialty is different.

However, in order to judge the value of this paper from, insufficient comparison with the results of previously published papers on HKUST-1. According to Ref. 108 and 109, since the two Cu sites are crystallographically equivalent, the details of their electronic states are unknown only by diffraction experiments. From this point, the calculation assuming three magnetic states between two Cu sites is a very interesting result. It will be necessary to compare these calculated results with the information obtained from the actual HKUST-1 experimental results, such as spectroscopy or magnetic susceptibility, ESR and so on. In particular, since it is difficult to obtain precise experimental information on the electronic state of MOF containing molecules in the cavity, how well the electronic state of the hollow cavity is reproduced determines the effectiveness of this calculation method. The results of multiple computational conditions help us understand the experimental results.

Finally, there are some typos in the treatise, so I recommend proofreading.

Author Response

Please see the attachment

Author Response File: Author Response.pdf