General Comments

The study by Ran et al. outlines laboratory experiments aimed at studying the gaseous products emitted when the neonicotinoid pesticide nitenpyram (NPM) is photolyzed in aqueous solutions in the presence of iron(III). Analysis of the gas phase species NO and NO2 were performed using a chemiluminescence detector with a molybdenum converter, while the carbonate denuder method was used to measure HONO.  Aqueous phase reactive oxygen species such as superoxide and hydroxyl radical were found to be present using EPR. The main finding of this study is that in the presence of iron, organic nitrogen functional groups found in NPM are converted to inorganic species such as NOx and HONO, which outgas from solution and are detected in the gas phase. Some kinetic studies were carried out, which in principle can be used to assess the environmental (photochemical) fate of this pesticide. The authors also carried out a modeling exercise to determine the impact of NPM on global atmospheric HONO and NOx budgets. The novelty of the work is not so clear, although there appears to be some previous work done to show that neonicotinoids photolyze more rapidly in the presence of iron. Neonicotinoid pesticides are a major environmental hazard due to the massive quanitities they are applied to ag fields and their toxicity to pollinators. So, I would say understanding the fate of neonics in general is more important than its impact on HONO and NOx, which is likely quite minor.  The experimental methods are somewhat standard, although it appears that little attention was paid to controlling pH of the solution.  The most significant problem in the paper was how the modeling work done, the lack of detail provided, and what seem to be nonsensical results. In my opinion, the study requires significant work yet, especially on the modeling side.

Specific Comments

Introduction:  There needs to be a more thorough discussion of the previous work done on neonicotinoid pesticide photochemistry in the introduction section. There are a number of articles now on the topic, including showing how Fe3+ can catalyzed the photodegradation (some of these are cited later in the paper).

Abstract:  In justifying the importance of this paper in the abstract, the authors claim that neonicotinoid pesticides are a source of HONO and NOx during the daytime that can help “narrow the gap between field observations and model outcomes of HONO in the atmosphere…can be an important contribution to the global nitrogen cycle affecting the atmospheric oxidizing capacity as well as climate change.” As noted below, I believe the stated importance of neonicotinoid pesticides as a source of HONO and NOx may need to be reevaluated. Despite this, I think any research on the fate of neonicotinoid pesticides is useful in that these compounds are used extensively on crops and due to their deleterious impact on pollinator populations.  Thus, I think it may be wiser to redirect the focus of their paper to the fate of the neonicotinoid pesticides in general. In this context, HONO and NOx are simply photodegradation products and the experiments discussed are used to deduce the degradation mechanism. The quantum yields and photolysis rates calculated under actinic conditions can then be used to simply calculate a photolysis half-life. This would be much more useful than an erroneous global model where proper model inputs are not available or have huge errors associated with assumptions made.

Line 146:  The authors state, “The change of Fe3+ concentrations may alter the pKa of the acid-base balance…” This doesn’t seem to make chemical sense and needs revision/clarification. The pKa is a property of a weak acid that does not depend necessarily on the Fe3+ concentration unless the iron salts are changing the ionic strength considerably.  Rather it is the pH that may alter the speciation of the Fe3+. Perhaps the authors meant that the addition or presence of FeCl3 changes the pH of the system, which would affect the speciation of NPM?

 

Line 150:  I recommend being more quantitative here. i.e., estimate the proportion of weak acid and conjugate base present.

Section 3.2.  It is stated that iron is present in natural waters at concentrations ranging between 10-7 M and 1e-4 M.  What kind of natural waters? Aerosols, marine, rivers? Please explain why such high concentrations of iron were used for experiments. It is especially important for connecting to the modeling study. 

Line 198:  The authors mention they quantified quantum yields of HONO, NO2, and NO formation from NPM photolysis from the measured photolysis frequencies (J-values). However, they never report what they are, nor do they discuss them in the context of prior studies for related chemicals to evaluate whether they make sense or not. 

Line 278: The statement, “but the theoretical calculation predicted that the photolysis of NPM can generate NO2 rather than N2O. What is this referring to? This seems to refer to a theory study.  If this refers to the Aregahegn study, then clarify that this work also included calculations so it is clear.

Figure S4 suggests that nitrate is the major product in the absence of Fe3+. This is not explained clearly. Where does the nitrate come from?

 

Page 11: the authors attempt to evaluate the importance of NPM as a source of HONO and NOx in the atmosphere by using what appears to be a global model. This is the most problematic part of the paper due to the fact that the model is so poorly described. There seems to be little effort to thoroughly document the experimental and modeling methods and procedures to allow their work to be reproduced at a later date. For example, there are a total of 10 lines of text in the supplemental information file describing this model with very little detail that anyone can evaluate the validity of the approach. Moreover, there is no literature cited – nothing to aid in reproducing their modeling results besides the equations used to parameterize HONO and NOx emission rates in the model. Also, the equations are written in an ambiguous manner with no attention to defining terms or units.  It is not possible to evaluate whether the model was run appropriately. Any parameterization of photochemistry would need to account for wavelength-dependent molar absorption coefficients, quantum yields, and actinic flux. While there is mention of how the model models actinic flux, there is no information on which absorption coefficients or quantum yeid(s) are used (nothing is tabulated for someone who would like to reproduce the experiments). Are there any assumptions being made in their calculation and if so, what are they? How does the model account for fluctuations in iron content on a global scale?  What types of surfaces in the model are the sites for these reactions? Aerosols, soils, foliage, buildings? How are those surfaces represented? Such a model needs to account for mass loading of the pesticide on fields. This requires geospatial data of pesticide usage on a global scale. Yet, there is not discussion of the data set that was used for this.  Also, once applied to the fields, there will be a fraction of the pesticide that is absorbed by plants and soil and not available for photolysis. The way it is written, it seems the authors treat the entire globe as an aqueous reactor where the aqueous concentration of NPM is a uniform 50 microgram per liter. 

Figure 5:  Geospatial emissions of HONO and NOx are presented showing high emissions mostly over the oceans and in the mid latitudes. Numbers are written on the map as, “Global: 0.77; Land: 0.20; and Ocean 0.57). First of all, it is not clear what those numbers refer to as there are no units associated with them.  I am assuming they are in Tg N / y since the values somewhat resemble those stated in the text. These numbers are nonsensical for the fact that neonicotinoids are not applied to the World’s oceans.  They are applied to cropland and mostly in the northern hemisphere.  There is enough resolution on the map to allow one to identify the World’s major agricultural regions and it is puzzling why they show less emissions of HONO and NOx due to NPM photochemistry than do the oceans.  This indicates a major flaw in the model. For this reason, I do not trust the statement on line 294, which states that NM photochemistry represents 3.5% of the total anthropogenic emissions of NOx related to fossil fuel in the year 2017.   

Experimental section:  Experimental details are lacking in some areas. What is the volume of solution used? The pH? Was the reactor static or a flow reactor?  Was the system stirred?  Would be useful to provide a figure showing how the experiment was carried out. The authors appear to use a spectroradiometer to measure spectral irradiance. How was the irradiance measurement conducted? Does this account for light absorption through the walls/window of the reactor? Can the authors be sure that the light was not attenuated too much by the high concentrations of solutes in the reactor to the point that the reactor depth was not evenly irradiated? Why did the authors use a carbonate denuder to measure HONO? How did the authors account for breakthrough, which is a major problem with this method.  Also, how did the authors correct for the HONO that is converted to NO on the Mo catalyst and would lead to an erroneous NO2 signal?

Comment on title:  The title should be revised to read as:  “Formation of Reactive Nitrogen Species Promoted by Iron Ions through the Photochemistry of a Neonicotinoid Insecticide. Note the use of the indefinite article before Neonicotinoid since the authors only studied one insecticide, meaning it would not be correct to assume that the chemistry applies to all neonicotinoid insecticides without additional evidence.

Comment on short summary: The authors’ use of ‘eruptive’ is not appropriate in this context. I suggest revising the sentence to:  We report the formation of nitrous acid (HONO) and NOx (NO + NO2) during the photolysis of a neonicanoid insecticide in the presence of iron at the air-water interface.

Line 303: Regarding data availability, the authors should provide tables of wavelength-dependent absorption and quantum yields to allow readers to use this data in the future to either conduct their own research or to check over the published work.

1. HONO is indirectly measured as the difference between the NO₂ signal and the Na₂CO₃ tube. This could be subjected with some uncertainty. Other reactive nitrogen species might be included in this differential signal. The uncertainty discussions maybe added.
2. Sect. 2.5, I have quickly checked the unit of the equation 3 and 4, I can end up with the unit of 'min^{-1}', but it is better converted to 's^{-1}', the authors may further clarify it.
3. In the reference part, Wang Y et al., 2021 is duplicated.