A Novel System for Crystal Polymorph Discovery via Selective-Wavelength Infrared Irradiation Using Metamaterials
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors present their study of how the crystallization of ritonavir is dependent upon selective IR irradiation of functional groups. While the physics of the MIM emitter are not in my area of expertise, the data presented in Figure 4 appears to validate their conclusion that specific areas of the infrared region can be targeted using the developed MIM selective radiation plate. It is indeed very interesting that form II can be selectively crystallized upon irradiation of either the NH/OH or C=O functional groups that would be expected to have an impact upon hydrogen bonding. As the authors suggest, one would expect the form with overall weaker extent of hydrogen bonding (form I) would be especially impacted by the irradiative disruption. It would certainly be helpful to see at least one additional success story for this method in order to definitively say that irradiation at a certain functional group area in the IR spectrum will have an anticipated effect, but I suppose this manuscript can serve as a “communication” of sorts. I would suggest considering a few changes:
1. The introductory paragraph has many broad statements without any references to back up claims. I would suggest adding references as appropriate.
2. Technical grammar point: line 61 should be “continuous irradiation with infrared radiation” and not “continuous irradiation of infrared radiation”.
3. Figure 1: the supplied figures were too small to see the actual apparatus clearly. I would suggest that perhaps larger images be supplied either in the manuscript or in the supporting information.
4. Figure 2: I am confused as to the supplied measurements: one indication is that the metal “dot” is one micrometer in diameter, while the other area says 2 micrometers. Are these measuring the same thing? Also, one of the titles is partially obscured by the image in the middle.
5. Figure 5: the key is missing as to which trace is the MIM output and which is the ceramic output.
6. Table 2: While the text clearly states that the solutions are all at 50 °C, the Peltier temperatures listed in the table confused me as I initially thought the samples were crystallizing at two different temperatures. Perhaps a footnote to the table that says something along the lines of “Peltier temperature required to maintain sample at 50 °C” would be helpful.
7. Related to the discussion of lines 214-228: This would greatly benefit from providing figures that illustrate the differences in hydrogen bonding patterns involved between the two forms. Also, providing estimated H-bonding energies for the various hydrogen bonding partners (if available) would strengthen the arguments made.
8. Finally, the authors argue that the reason Form II is preferentially formed upon irradiation at 3.3 and 5.8 micrometers is because crystallization of Form I is inhibited. This is not an unreasonable explanation. However, reference 11 suggests that one reason that Form II is not predominant is that the concentration of the conformation leading to form II is of very low concentration in solution (they suggest a 99:1 ratio of the two conformers leading to form I and form II, respectively). Perhaps irradiation of the solution at the two wavelengths results in a buildup of the conformer concentration in solution that leads to form II, which then increases the opportunity for it to crystallize as form II. So, rather than inhibiting the formation of form I (as the authors argue, which again, is certainly possible), an alternative reason might be the increase in concentration of the conformation leading to form II. It might be interesting to analyze the 1H NMR spectrum of a solution before and after irradiation to see if this is true, if an NMR instrument is available.
The results reported here are very interesting and I certainly support publication with attention to the above comments. I also look forward to future studies that will provide additional information as to the molecular processes leading to the observed outcomes.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsReport on the manuscript:
A Novel System for Crystal Polymorph Discovery via Selective-Wavelength Infrared Irradiation Using Metamaterials
This contribution is a promising start for some more investigations with this method. The investigated compound is very complex and I would have been expected that the authors validate their method also for one or two simple organic compounds that are know for many polymorphs (e.g. theophylline) or a compound which has exactly two polymorphs (e.g. 2,5-diaminopyridine).
The reference list is relatively short. I would have expected that there are more reports on similar experiments.
Perhaps the authors should check some articles on single crystal growing on a diffrctometer. Also in such cases IR radiation is often used to transfer some energy for crystal and/or polymorph formation (for example Roland Böse, Crystal strauctures of n-alkanes; Angew. Chem.)
Generally this technical paper is of interest to researchers in the field of polymorphism of organic materials.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThis paper explores utilizing selective infrared irradiation to influence polymorphism in crystal growth. Specific wavelengths of radiation are utilized during evaporative crystallization of a test substrate, Ritanovir, which is an excellent choice for this investigation. The results obtained by this study indicate that each Ritanovir polymorph can be obtained by tuning the desired infrared radiation utilizing their selective wavelength MIM emitter. Overall, this was a very interesting paper.
General comments:
The authors used the term "chapters." "Sections" would be more appropriate.
The text in both Figure 1 and Figure 2 were spaced too far vertically. It was challenging to decipher the labels for each of the arrows.
The green arrows in Figure 1 were very difficult to see.
The text appeared on the image in Figure 2 and was very difficult to read.
It would have been helpful to have the chemical differences between Form I and Form II of Ritanovir explained, which would justify why irradiating different hydrogen bonding sites might give different crystal forms. This would give the discussion in lines 214-228 more support.
Were the crystals that were grown analyzed in any way other than FTIR? XRD? Raman?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 4 Report
Comments and Suggestions for AuthorsThis is an interesting paper and should be published eventually. As I understand, it describes illuminating an evaporation of a pharmaceutical solution with selected narrow band IR to prevent crystallization of an undesirable polymorph. A few remarks to consider:
1) Lines 77-81 Contain the premise of the described work & support my understanding. Figure 2 shows the MIM emitter used for the IR source. It is confusing; the print labels are not all well placed, and it doesn't represent the device that is actually used. The metal pattern shown is a square array of 1µm circular elements, but the description (lines 97-98) and Figure 3 seems to describe a square array of different (0.6-1.7µm) square emitters with a different configuration of layers. The Figure 1 drawing should be more consistent with the actual experimental apparatus.
2) line 130-131 refers to emission measurements at 5 points about the MIM surface but doesn't show any resulting comparison. Were they all the same? Figure 4 shows 3 plots; only the 3.3 emitter shows more than 1 peak, but I can't tell what is being shown.
3) Figure 5 lacks text for the legend marking which curve is which. The axis titles are very small - the technical editor will complain.
4) line 151 & 166. I presume the ZnSe window is that pictured in Fig. 1. That should be in Fig. 1 as well - makes it much clearer. Any idea of how much IR power is projected onto the sample? NB: the actual sample temperature is perhaps more important than the Peltier set point particularly as the sample is illuminated by an IR light.
5) line 174 (& elsewhere). This isn't really a "screening" method. It's IR directed crystallization (or some such). Calling it "screening" could be confusing to the reader as it implies a method used to sort samples. It also isn't a "discovery" method (as in title), because the MIM devices aren't in operando tunable. The method relies on having IR spectra from the suite of polymorphs and then selecting/designing a MIM device to suppress crystallization of undesirable ones. This is after true discovery has found the polymorphs via solvent/temperature/etc. survey experiments. This very evident from the description in lines 181-187 for Ritonavir.
6) Are the crystal structures of Ritonavir I & II known (reference?)? Which one is more effective? Or was stability (to what, may I ask) the issue. Figure 7 shows IR spectra - text again too small & not well placed. Unfortunately, the IR spectra of the respective solids isn't really indicative of what is going on here. Otherwise, the two very similar (for 5.8 & 6.7) ought not to have different polymorphs appear. In fact, the Peltier temperatures were different (25°C & 53°C). Worse yet, the results in Table 2 suggest that temperature is the controlling factor not IR wavelength, i.e. 53°C gives phase I & 25°C gives phase II.
In conclusion, this is a very interesting concept, but I'm not sure authors have proved it. Experiments at one fixed temperature should be done to sort this out.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 4 Report
Comments and Suggestions for AuthorsI believe the authors have addressed my concerns in their revision. So, the paper should be published.
