http://pubs.acs.org/doi/pdf/10.1021/acs.est.7b03871

Just as the comments from reviewers, the present paper is an interesting and worthy step in using QC approaches to examine on a microscopic level mechanisms of DBP formation. In view that quantum chemical modeling offers a great advantage of allowing to examine such reactions in silico and obtain such critically important aspects of these reactions such as their thermodynamic parameters, reaction barriers, structures of possibly formed intermediates, nature of reactive sites etc, we believe it would break a fresh ground of research in DBP formation.

Firstly, this paper verifies the three-step halogenation model by the use of E_HOMO calculations. Although the three-step halogenation model proposed by Prof. Cowman and Singer in 1990s is of potential to provide a quantitative explanation of the role and competition of chlorine, bromine and iodine incorporation into NOM to form THMs. The utilities are very limited and most of them hypothesized that the ratio of the kinetic preference coefficients for chlorine, bromine, and iodide incorporation in NOM at each step is the same. This is because that the complexity of reactions between NOM and halogen species makes it challenging to generate exhaustively detailed experimental data to verify the statistical result. If the model can not be ascertained, it would be more like a mathematical game, and the results are unreliable because there are no less than one groups of solution for equations. The present paper is an interesting and worthy step in using QC approaches to examine mechanisms of DBP formation on a microscopic level and verify the three-step halogenation model. (This had introduced in this paper in the introduction section and the reviewer 1 agreed with this point also.)

Secondly, this paper has found and established the correlation between △E_HOMO values and the kinetic preference coefficients for chlorine, bromine, and iodide incorporation into NOM. It is true that this paper is a very straight forward paper since it relies on available software to conduct HOMO-LUMO calculations and existing experimental data for the rate constant ratios, but we do not agree that we just present a simple correlation.

QC calculation is more and more widely used to interpret the mechanisms of chemical reaction. According to the transition state theory, the absolute value of the reaction constant can be determined only if the enthalpy and entropy values of reactants and intermediates are known. In the case of the system discussed in this paper, the structures of intermediates are difficult to determine, and similar or more pronounced limitations to determine the enthalpy and entropy values of these intermediates. In this manuscript, an alternative approach based on the Frontier Orbital Theory was introduced and we had deduced the equation (16) in paper to describe the log of the kinetic preference coefficients determined by the HOMO of product molecules. We have further examined the influence of the structure of intermediates on the data generated using in this study. It verified that the HOMO energy could reflect the stability of halogenated intermediates when the R- is sufficiently simple. The method used in this manuscript, especially the equation (16) provides a new way to interpret the mechanism of DBPs formation.

Thirdly, there are some interesting and useful findings in this manuscript, such as alternative reaction mechanisms about iodine and the difference of E_HOMO between two halogens and three halogens THMs. It is well known that the reactivity of I is higher than Cl and Br, and that more I-DBP formation at weak oxidation conditions (chloramintion) are because there is more the species of HOI formed at that conditions. While this study demonstrated that the reactivity of I is still significantly higher than expected (two outliers in both Figure 5 and 6). This manuscript demonstrates that the presence of iodine in the intermediates significantly increases their affinity toward further iodine incorporation. It implies that there is different mechanism of I-THMs formation that we have not known yet.

The latter will help us to reveal the reason that the ratio of two halogens DBPs is higher than three halogens DBPs in chloramination, compared with chlorination. This phenomenon had found in 1990s, but it is not well interpreted yet. Our recent study has further demonstrated the difference of two halogens and three halogens DBPs formation in chloramination and chlorination (Chemosphere 2017, 191, 477-484). We suspected that it is due to the higher HOMO energy of three-halogen DBPs (reaction barriers), NH₂Cl could not incorporation the third halogen onto two-halogenated DBPs, but Cl₂ could.

Due to the limitation of space of this paper, we have not expanded our discussion of these findings in details. We could not hope to solve all problems in one paper. But our research group are working further in this field and have gotten some interesting progresses recently, we hope could publish our further manuscript in near future.