Efficient and simultaneous capture of iodine and methyl iodide achieved by a covalent organic framework

Radioactive molecular iodine (I2) and organic iodides, mainly methyl iodide (CH3I), coexist in the off-gas stream of nuclear power plants at low concentrations, whereas few adsorbents can effectively adsorb low-concentration I2 and CH3I simultaneously. Here we demonstrate that the I2 adsorption can occur on various adsorptive sites and be promoted through intermolecular interactions. The CH3I adsorption capacity is positively correlated with the content of strong binding sites but is unrelated to the textural properties of the adsorbent. These insights allow us to design a covalent organic framework to simultaneously capture I2 and CH3I at low concentrations. The developed material, COF-TAPT, combines high crystallinity, a large surface area, and abundant nucleophilic groups and exhibits a record-high static CH3I adsorption capacity (1.53 g·g−1 at 25 °C). In the dynamic mixed-gas adsorption with 150 ppm of I2 and 50 ppm of CH3I, COF-TAPT presents an excellent total iodine capture capacity (1.51 g·g−1), surpassing various benchmark adsorbents. This work deepens the understanding of I2/CH3I adsorption mechanisms, providing guidance for the development of novel adsorbents for related applications.


Reviewer #1
Comments: In this manuscript, two COFs (COF-TAPT and COF-TAPB) was synthesized for the simultaneous capture of iodine and organic iodides at low concentrations. I believe this work is suitable to be published in Nat. Commun. after addressing the following comments.

Response:
We are truly grateful to the reviewer for his/her appreciation of our work.
1. For the intro-part, the authors are supposed to encapsulate in detail about the different iodine/iodide species and isotopes in the off-gas to reinforce the scientific significance of this work.
Response: Following this suggestion, we elaborated in the revised manuscript the composition of the off-gas and its adverse effects on the environment and human health. "One of the major safety issues is the volatile radioactive waste produced during the reprocessing of spent nuclear fuels, which primarily consists of radionuclides, such as 129 I and 131 I in the form of molecular iodine (I2) or organic iodides (e.g., methyl iodide (CH3I) and ethyl iodide). [1][2][3][4][5] These compounds are harmful to the environment ( 129 I has an extremely long half-life of approximately 1.57 × 10 7 years) or severely affect human metabolism by damaging the thyroid gland, and must be removed before the off-gas is discharged. 6-8 " 2. The authors claim their material is mainly based on the filling of iodine molecules into the pores and the enhanced host-guest interaction. Please calculate the estimated adsorption capacity according to the pore volume of COF-TAPT and COF-TAPB as the previous works did (Jiang et al. Adv. Mater. 2018, 30, 1801991) and compare this value to the experimental value to further support their claim.

Response:
We thank the reviewer for this insightful comment. In many previous studies, I2 adsorption was performed in a static closed system with saturated I2 vapor at 75°C, and the adsorption capacity was determined based on the mass increase subsequently measured under ambient conditions. The I2 adsorption capacity determined in this way is often higher than the theoretical value calculated from the pore volume and the density of solid iodine (see Fig. R1 for summary data). A plausible explanation for this phenomenon is that under such measurement 2 conditions, there is a large amount of I2 adsorbed at the external surface of the adsorbent particles and condensed in the inter-particle pores. In this sense, the static I2 uptake value measured in this way is not a meaningful criterion for performance evaluation. We will discuss this issue shortly in a separation publication.
In the current study, we performed this measurement only for comparison purpose, whereas we highlighted the ability of COF-TAPT to simultaneously capture low-concentration I2 and CH3I under dynamic adsorption conditions.

Fig. R1
Static I2 adsorption capacities and pore volumes of various reported adsorbents. The straight line, plotted by the equation of qT = (pore volume×4.93) g·g -1 , represents theoretical adsorption capacities (qT) determined based on the pore volumes (cm 3 /g) and the density of solid iodine (4.93 g/cm 3 ). This summary indicates that many adsorbents (those above the straight line) exhibit static I2 adsorption capacities higher than theoretical values.
3. In the part of material regeneration, the I2-or CH3I-loaded COF was immersed into ethanol or acetone, and ultrasonicated for a certain period to obtain regenerated COF. Since the captured I2 can be dissociation from material due to the relatively weak interactions, I wonder why the CH3I-loaded material can be fully regenerated at such a mild condition after the methylation? Please specify it.

Response:
We thank the reviewer for raising this question. Yes, the captured CH3I can be released by the ultra-sonication-assisted extraction process using ethanol. Taking the imine N (the most favorable binding sites for CH3I in COF-TAPT according to DFT calculations (Fig. 6)) as an example, its methylation to form an iminium salt is a reversible reaction in protic solvents (e.g., H2O, alcohols). 9 In addition, ultra-sonication treatment, which is a common method for promoting the decomposition of organic molecules, [10][11][12][13] can effectively facilitate this reversible reaction.
In the revised manuscript, we discuss this point following the reviewer's suggestion.
"…and accelerated with the assistance of sonication. As a commonly used method for regenerating adsorbents after I2 adsorption, extraction with ethanol can also efficiently remove CH3I from COF-TAPT because the Nmethylation reaction is reversible in protic solvents. The regenerated COF-TAPT… " 4. From the XPS results before and after the adsorption (I2/CH3I), the peaks assigned to sp 3 -N are gradually shifted to higher binding energy, indicating the increase of binding energy from relatively weak interaction (between I2 and N) to stronger interaction (between CH3I and N). However, the specie of N (-C=N-) present another pattern, which may help reveal the affinity of different N atoms toward different species of iodine (Also, the authors listed the structure activity relationship between the N content and CH3I uptake). Please supplement quantitative and statistical analyses of the phenomenon for better understanding.

Response:
We thank the reviewer very much for his/her valuable insight and suggestion. Following the reviewer's suggestion, we briefly discuss in the revised manuscript the XPS results that imply the different affinity of I2 and CH3I at different N sites.
"…at various N sites through methylation reactions (Fig. 5a). In addition, after the adsorption of CH3I, the N 1s electron binding energies of imine/triazine N and sp 3 N in COF-TAPT increased by 1.3 and 3.7 eV, respectively, providing additional evidence for the binding of CH3I on N species (Fig. 5b). Compared with I2 adsorption, CH3I adsorption resulted in a less pronounced peak shift for imine/triazine N and a more pronounced peak shift for sp 3 N, implying differences in affinity between I2 and CH3I at different N sites. In FTIR…" To reinforce the established correlation between the CH3I uptake and the N content of the adsorbent, we include two additional COFs materials (i.e., COF-OH-0 14 and TPB-DMTP-COF 15 ) in the analysis. The results for seven adsorbents yield a good linear relationship between the CH3I uptake (at 75 °C) and the N content of the adsorbent with a reasonable coefficient of determination (R 2 =0.904). In the revised manuscript, these additional data and statistical analysis results are shown in Fig. S5, according to the reviewer's suggestion.

Reviewer #2
Comments: The authors synthesized two COFs to capture radioactive molecular I2 and CH3I. The design of the COFs is conventionally based on the consideration of introducing N sites, while the difference of the two COFs is the N content in the framework. Hence, the thought of structure design is not innovative enough. A merit of this work is the simultaneous adsorption of I2 and CH3I, the performance of the synthesized COFs is good but not excellent. The static I2 uptake ranks 4th and 5th compared with other 22 materials. The dynamic adsorption capacity of the synthesized COFs is also not the best. Moreover, the t80% is 18 h and 24 h for the synthesized COFs, which are longer than many of other materials, representing the low adsorption dynamic performance of the two COFs. Therefore, form both the viewpoints of the features of COF structure design and the adsorption performance, this work is not recommended to be published in nature communication.
Following questions are suggested to be addressed:

Response:
We thank the reviewer for the critical but insightful comments, and we would like to clarify the significance of this work from the following aspects.
As indicated in our manuscript, static I2 uptake measured in a closed system with saturated I2 vapor at 75°C is NOT a meaningful criterion for assessing the I2 capture performance, because these conditions are far from those in practical applications (co-existence of I2 and organic iodides at much lower concentrations). We measured static I2 uptakes in this study only for comparison purpose, and the ranking of our adsorbents on this metric is not important.
Adsorbing I2 at high pressures is not challenging due to the strong intermolecular interactions of I2.
In fact, many reported adsorbents exhibit static I2 uptakes higher than the theoretical values calculated from the pore volume and the density of iodine, because I2 can be easily adsorbed at the external surface of the adsorbent particles and condensed in the inter-particle pores. Please refer to What is really challenging and largely overlooked in the literature is the simultaneous capture of I2 and CH3I under low concentration and dynamic conditions, which is the focus of this study. Our adsorbent design is based on these considerations. The designed adsorbent, COF-TAPT, combines high crystallinity, a large surface area, and abundant nucleophilic groups, and exhibits a record-high static CH3I adsorption capacity (1.53 g·g -1 at 25 C). In the dynamic mixed-gas adsorption with 150 ppm of I2 and 50 ppm of CH3I (close to the practical application condition), COF-TAPT outperformed all tested adsorbents except an ionic COF in terms of total iodine capture. These results indicate that the iodine capture performance of COF-TAPT is not "good" but "excellent".
More importantly, the two COFs, namely COF-TAPB and COF-TAPT, exhibit the same crystal structure and textural properties, being different only in N content. Thus, they provide an unprecedented platform for systematic mechanistic studies on the different roles of textural property and N content in the adsorption of I2 and CH3I. Through these studies, we gained a series of new insights.
 I2 can be relatively easily adsorbed on a variety of electron-donor sites, including various N species and aromatic moieties, by forming charge-transfer complexes and polyiodides.
Therefore, the characteristics of binding sites and textural properties (e.g., surface area and pore volume) of the adsorbent both affect I2 uptake.
 The adsorption of CH3I occurs specifically on nucleophilic N sites with a one-to-one correspondence through N-methylation reactions to form salts and is unrelated to the textural properties of the adsorbent.
 Ionic groups can strongly promote the adsorption of I2 but have little promotional effect on the adsorption of CH3I.
 The CH3I binding energy at different N sites follows the order imine N > triazine N > sp 3 N.
These new insights are crucial for the understanding of I2/CH3I adsorption mechanisms, providing valuable guidance for the development of novel adsorbents for radioactive iodine capture and related applications.
We hope that the clarifications we offer here disentangle the true value of our work from the reviewer's concerns.
1. Figure 5d, the I2 transformed into I3 − and I5 − , what's the origin of the negative charge. If the COF framework takes one positive charge, will the COF be stable under this condition?

Response:
We thank the reviewer for raising this question. It has been well documented that I2 could be relatively easily adsorbed on a variety of electron-donor sites, including various N species and 7 aromatic moieties, by forming charge-transfer complexes and polyiodides. [15][16][17][18][19][20][21][22][23][24] That is, the negative charge of polyiodides originates from the functional groups of the adsorbent.
The framework charge does not affect the framework stability. Aluminosilicate zeolites have negatively charged framework interacting with extra-framework cations, while they are highly stable.
Likewise, COF-TAPT exhibits high stability after I2 adsorption, as evidenced by its ability to be reused without detectable changes in structure, porosity and I2 adsorption capacity (see Fig. S9). The reviewer is correct that the measured static I2 uptakes are higher than the theoretical values calculated from the pore volume and the density of iodine. This is a very common phenomenon (see Fig. R1 for summary data) and also why we emphasize that the static I2 uptake value is not a meaningful criterion for performance evaluation. A plausible explanation for this phenomenon is that under such measurement conditions, there is a large amount of I2 adsorbed at the external surface of the adsorbent particles and condensed in the inter-particle pores. We will discuss this issue shortly in a separation publication.
In the current study, we measured static I2 uptake following the widely used protocol for comparison purpose only, whereas our focus was the ability of COF-TAPT to simultaneously capture lowconcentration I2 and CH3I under dynamic adsorption conditions.
3. The I2 and CH3I are radioactive, stability of the COFs should be tested under radiation condition.

Response:
We thank the reviewer for raising this point. Following the reviewer's comments, we evaluated the stability of COF-TAPT under radiation conditions. Specifically, COF-TAPT was treated with high doses (100 kGy and 200 kGy) of β-irradiation using an electron accelerator (1.0 MeV, Wasik Associates Inc., USA). The PXRD results reveal that COF-TAPT well maintained its crystallinity after irradiation, confirming its excellent irradiation stability. These results are presented in the revised manuscript as Fig. S3. 4. The conclusion of "The CH3I adsorption capacity is positively correlated with the content of strong binding sites but is unrelated to the textural properties of the adsorbent." is one-sided as the number of COFs tested in this work is small.

Response:
We thank the reviewer for raising this point. Reviewer 1 also expressed the same concern, recommending "quantitative and statistical analyses".
In order to reinforce this conclusion, we include two additional COFs materials (i.e., COF-OH-0 14 and TPB-DMTP-COF 15 ) in the analysis. The results for seven adsorbents in total yield a good linear relationship between the CH3I uptake and the N content of the adsorbent with a reasonable coefficient of determination (R 2 =0.904). Moreover, the irrelevance between CH3I uptake and the BET surface area of the adsorbent is also further confirmed.
In the revised manuscript, these additional data and statistical analysis results are provided in Fig. S5, according to the reviewers' comments. 10 5. The content is similar to the previously published paper Angew. Chem. Int. Ed. 2021, 60, 2 -11.

Response:
We thank the review for raising this concern. However, we cannot find the literature according to the information provided by the reviewer.
If this is the case, please kindly note that we explicitly state in the manuscript that COFs with the same structures as COF-TAPT and COF-TAPB were reported before for CO2 adsorption (Chemical In the paper published in Angew. Chem., the COFs were used for photocatalytic hydrogen evolution.
Instead of claiming that the two COF materials are novel, we emphasize that we developed a new scope of application for them. The paper published in Angew. Chem. has been cited in the revised manuscript.

Reviewer #3:
Comments: Simultaneous capture of iodine and methyl iodide by porous materials through breakthrough test remains a challenge. As stated in the text, there is only one COF that covers this feature. In this work, Han et al prepared two comparable COFs but with the difference in nitrogen atom, thus, giving an insight into the relationship between I2, CH3I uptake and various nitrogen atoms.
Effective and simultaneous capture of iodine and methyl iodide were observed for the nitrogen-rich COFs. The results are very interesting. The preparation of this work was careful. But some points should be considered before publication.
Response: we are truly grateful to the reviewer for his/her appreciation of our work.
1. "To the best of our knowledge, there is only one COF material (SCU-COF-2) evaluated for both I2 and CH3I adsorption". In fact, Han and co-workers has also reported a MOF for such use. Thus, please revise this sentence.

Response:
We thank the reviewer for raising this point. The reviewer is correct that modified MOF MIL-101 has been tested for simultaneous capture of I2 and CH3I (Nature Communications, 2017, 8, 485). However, in our manuscript, the statement above was placed in the context of COF materials.
It is difficult for us to include a MOF-based adsorbent in this statement without affecting the narrative flow. Therefore, we prefer not to modify this statement.
2. Although the author has made deep research in both I2 and CH3I breakthrough test, however, as investigated in their previous work, simultaneous capture of iodine and methyl iodide at high temperature such as 423 K is suggested to be investigated, especially in the presence of HNO3.

Response:
We thank the reviewer for this insightful suggestion. We have measured the I2 and CH3I uptakes of COF-TAPT and COF-TAPB under dynamic condition at 75 °C. It turned out that for both materials, increasing the measurement temperature from 25 °C to 75 °C leads to significant decrease in I2/CH3I adsorption capacity (Fig. R3). These results indicate that COF-TAPT and COF-TAPB are unsuitable for iodine capture at elevated temperatures. Therefore, this study has been focused on the I2 and CH3I adsorption behaviors at room temperature.   Fig. S3. 5. The pore size is not given in Fig. 1.

Response:
As suggested by the reviewer, we have updated Fig. 1 to include the pore size distribution profiles for COF-TAPB and COF-TAPT, which are derived from the N2 sorption isotherms using the NLDFT method.
6. In Fig. 4, the uptake is based on weighing method, thus, error bar is suggested.
Response: Following this suggestion, Fig. 4 has been updated with error bars, which were determined based on the results of three parallel measurements. 14 taken from the mixture, separated with a 0.22 mm nylon membrane filter, and measured by UV-vis spectroscopy. The results indicate that CH3I@COF-TAPT can efficiently capture ~90% Cr(VI) within 30 min and ~95% Cr(VI) after 180 min, which is consistent with the apparent discoloration of the solution during the anion-exchange process. These results are shown in Fig. S13 in the revised manuscript to provide additional evidence that CH3I is adsorbed through methylation reactions.