Mechanistic insights into I₂O₅ heterogeneous hydrolysis and its role in iodine aerosol growth in pristine and polluted atmospheres

Higher-order iodine oxides are intricately linked to marine aerosol formation; however, the underlying physicochemical mechanisms remain poorly constrained, particularly for I₂O₅, which is stable yet conspicuously absent in the atmosphere. While reactivity with water has been implicated, the direct hydrolysis of I₂O₅ (I₂O₅+ H₂O → 2HIO₃) fails to account for this discrepancy due to its high activation barrier (21.8 kcal mol⁻¹). Herein, we have probed heterogeneous hydrolysis of I₂O₅ mediated by prevalent chemicals over oceans through Born-Oppenheimer molecular dynamics and well-tempered metadynamics simulations. Our results demonstrate that self-catalyzed pathways involving I₂O₅ and its hydrolysis product HIO₃ substantially reduce the reaction barrier, thereby accelerating the conversion of I₂O₅ to HIO₃ in pristine marine environments. In polluted regions, interfacial hydrolysis of I₂O₅ mediated by acidic or basic pollutants (e.g., H₂SO₄ or amines) proceeds with even greater efficiency, characterized by remarkably low barriers (≤ 1.3 kcal mol⁻¹). Collectively, these proposed heterogeneous reactions of I₂O₅ are relatively effective, acting as a hitherto unrecognized sink for I₂O₅ and a source of HIO₃ – processes that facilitate marine aerosol growth and rationalize the high iodate abundances detected in aerosols. These findings provide mechanistic insight into the elusive I₂O₅-to-HIO₃ conversion, offering an unheeded step toward improving the representation of iodine chemistry and marine aerosol formation in atmospheric models, with implications for climate prediction and environmental impact assessment.

1 Introduction

Given the vast coverage (71 %) of oceans on Earth's surface, marine aerosols exert substantial impact on the global atmosphere, altering cloud microphysical properties, radiative balance, and climate change (Kaloshin, 2021; Mahowald et al., 2018). Hence, understanding how aerosols form is fundamental for the climate system, especially ruinous extreme weather, as it introduces the greatest uncertainty in climate forcing (Gettelman and Kahn, 2025). Earlier studies have confirmed a solid association between marine aerosols and sulfur chemicals derived from the oxidation of dimethyl sulfide (DMS) (Barnes et al., 2006; Chen et al., 2018; Li et al., 2024; Ning and Zhang, 2022; Shen et al., 2020; Zhang et al., 2022a, 2024a). While the recent evidence has shown that iodine-bearing precursors of oceanic origin such as iodine oxides (I₂O₂₋₅) and iodine oxyacids (HIO₂₋₃) significantly contribute to marine aerosol formation, owing to their high efficiency and rising levels (Allan et al., 2015; Carpenter et al., 2021; Droste et al., 2021; Gómez Martín et al., 2021, 2022b; Huang et al., 2022; Li et al., 2022; McFiggans et al., 2010; O'Dowd et al., 2002; O'Dowd and De Leeuw, 2007; Roscoe et al., 2015; Saiz-Lopez et al., 2012, 2014). Despite the experimental and theoretical studies uncovering the role of HIO₂₋₃ in aerosol formation (He et al., 2021, 2023; Liu et al., 2023; Zhang et al., 2022b, 2024b; Zu et al., 2024), the atmospheric fate and impacts of I₂O₂₋₅ are yet to be fully established (Finkenzeller et al., 2023; Gómez Martín et al., 2013, 2020; Leroy and Bosland, 2023; Lewis et al., 2020; Liang et al., 2022), limiting the accuracy of atmospheric models in simulating iodine-driven climate effects.

As one of the highest iodine oxides, iodine pentoxide (I₂O₅) serves as the anhydride of iodic acid (HIO₃), playing a vital role in bridging iodine oxides and iodic acid in iodine chemical cycle, however, its true fate is still elusive, with the conflicting findings reported. Specifically, I₂O₅ is highly thermally stable (Kaltsoyannis and Plane, 2008), yet it is not evidently present in the gas phase, implying its unknown sinks. Sipilä et al. (2016) hypothesized that I₂O₅ may participate in the formation of marine aerosol particles through the dehydration reaction

$$\begin{array}{}\text{(R1)}& \mathrm{2}{\mathrm{HIO}}_{\mathrm{3}}\to {\mathrm{I}}_{\mathrm{2}}{\mathrm{O}}_{\mathrm{5}}+{\mathrm{H}}_{\mathrm{2}}\mathrm{O}\end{array}$$

based on the identified oxygen-to-iodine ratio of 2.4 in field-collected particles. Nevertheless, no direct detection of I₂O₅, such as mass spectral signals, was provided. And the theoretical study has shown that Reaction (R1) hardly occurs due to its high energy barrier (27.8 kcal mol⁻¹) and endothermic nature (Khanniche et al., 2016), suggesting that this sink pathway seems not well supported. A most recent experimental evidence indicates that I₂O₅ can indeed participate in aerosol particle formation, but under the unrealistically low-humidity conditions (Rörup et al., 2024). Upon the addition of water, I₂O₅ is markedly depleted, suggesting its reaction with water, accompanied by the formation of HIO₃ (Gómez Martín et al., 2022a). This finding seems to support well the absence of I₂O₅ in high-humidity marine atmospheres. While the theoretical study of Xia et al. (2020) found that the direct reaction between I₂O₅ and H₂O

$$\begin{array}{}\text{(R2)}& {\mathrm{I}}_{\mathrm{2}}{\mathrm{O}}_{\mathrm{5}}+{\mathrm{H}}_{\mathrm{2}}\mathrm{O}\to \mathrm{2}{\mathrm{HIO}}_{\mathrm{3}}\end{array}$$

in gas phase also needs to cross a high energy barrier of 21.8 kcal mol⁻¹, suggesting that this process is unlikely to occur effectively, even with another H₂O or I₂O₅ acting as a catalyst (Kumar et al., 2018). As a result, this gas-phase mechanism alone appears insufficient to explain the observed absence of I₂O₅ in experimental or clean marine environments, with the I₂O₅ sink remaining unrevealed. Furthermore, Kumar et al. (2018) proposed the heterogeneous reactions of I₂O₅ at the air-water interface, which is ubiquitous over oceans like aerosol surface (Zhong et al., 2018). Unfortunately, this interfacial mechanism still fails to account for the observed well-established link between I₂O₅ depletion and HIO₃ formation, as it points to the formation of other novel product H₂I₂O₆. Despite its dynamic stability within the simulation scale (34 ps) (Kumar et al., 2018), H₂I₂O₆ has not been detected in the prior iodine-related experimental and field studies (Gómez Martín et al., 2020, 2022a; He et al., 2021, 2023; Rörup et al., 2024; Sipilä et al., 2016), highlighting the poor understanding of the reaction mechanism of I₂O₅ under pristine conditions.

The mystery of I₂O₅ chemistry likely deepens in polluted coastal regions, e.g., Zhejiang, China (Yu et al., 2019), where the evident burst of iodine aerosols has been also observed. Similarly, I₂O₅ is also evidently absent but HIO₃ (detected as IO${}_{\mathrm{3}}^{-}$) is abundant in the field-collected aerosols, indicating I₂O₅ may undergo an I₂O₅-to-HIO₃ conversion. Yet the underlying mechanisms in polluted regions remain unclear, which is likely more complex than that in pristine regions due to the pollutant-mediated impacts from typical amine or sulfuric acid that can facilitate the heterogeneous reactions of other iodine oxides (Ning et al., 2023, 2024). Taken together, these unresolved discrepancies and mechanistic uncertainties render the fate of I₂O₅ elusive in both pristine and polluted marine environments, which limits our understanding of its true role in atmospheric iodine chemistry and aerosol formation.

In this study, we have performed the Born-Oppenheimer molecular dynamics (BOMD) simulations to elucidate the potential hydrolysis mechanism of I₂O₅ at the air-water interface, as represented by aqueous aerosol surfaces, mediated by different kinds of prevalent chemicals. We first investigate the heterogeneous hydrolysis of I₂O₅ mediated by I₂O₅ and HIO₃, which potentially occur in pristine marine regions; while the reactions of I₂O₅ mediated by pollutants such as acid (sulfuric acid, SA) and alkaline (methylamine (MA), dimethylamine (DMA), trimethylamine (TMA)) were also probed to comprehend the loss path of I₂O₅ in polluted marine environments. Alongside mechanistic insights, we further quantified the reaction barriers by metadynamics simulations (MetaD) and block error analysis. By comparing the resulting energy barriers, the dominant mechanisms of I₂O₅ hydrolysis under different environments were identified. Furthermore, we performed wavefunction analysis to examine the reactive sites of I₂O₅ hydrolysis products at the air-water interface, which may facilitate the condensation of gaseous precursors to promote aerosol growth. This work provides mechanistic insights into heterogeneous hydrolysis of I₂O₅ that potentially occur under marine conditions, enhancing our understanding of iodine chemistry and marine aerosol formation.

2 Methods

2.1 Quantum Chemistry Calculations

Structural optimizations for gas-phase molecular geometries were performed using the Gaussian 16 software package (Frisch et al., 2016) with tight convergence criteria. The I₂O₅ molecule with lowest Gibbs free energy has been selected from isomers (Kaltsoyannis and Plane, 2008; Khanniche et al., 2016; Kim and Yoo, 2016), and details of the structures and coordinates are provided in the Supplement (Fig. S1 and Table S3). During the geometry optimization, the M06-2X functional was selected for gas-phase molecules due to its excellent performance in calculating main group elements (Zhao and Truhlar, 2008), and the adopted basis set was aug-cc-pVTZ (Kendall et al., 1992) (for C, H, O, N, and S atoms) + aug-cc-pVTZ-PP with ECP28MDF (for I atom) (Peterson et al., 2003). The frequency analysis was performed at the same level of theory to ensure that the optimized structures have no imaginary frequencies. The geometries and coordinates of gas-phase molecules (i.e. I₂O₅, HIO₃, H₂SO₄, MA, DMA, and TMA) are provided in Fig. S2 and Table S2 in the Supplement, respectively.

2.2 Classical MD Simulations

Classical molecular dynamics (MD) simulations were carried out by the GROMACS 2022 package (Van Der Spoel et al., 2005) to probe the interfacial propensity of I₂O₅ at the air-water interface. Regarding the initial configurations for MD simulations, we used the PACKMOL code (Martínez et al., 2009) to build a cubic water slab with a 3.3 nm edge, the simulation box, consisting of 1000 water molecules with a tolerance of 2.0 Å (minimum interatomic distance). And the simulation box was expanded to 3.3 × 3.3 × 16 nm³ along the z-axis to build the vacuum layers, forming the air-water interface. Umbrella sampling was employed, and a harmonic bias of 5000 kJ mol⁻¹ nm⁻² was applied to I₂O₅ molecule along the z-axis to investigate its free energy profile crossing the air-water interface from bulk water into gas phase. Furthermore, the generalized AMBER force field (GAFF) (Wang et al., 2004) with parameters obtained from the Sobtop 1.0 code (Lu, 2024) was combined with the TIP4P water model (Jorgensen et al., 1983) to describe the driving forces in the MD simulations. The restrained electrostatic potential (RESP) method (Bayly et al., 1993) was used to fit the atomic charges using the Multiwfn 3.7 code (Lu and Chen, 2012). Prior to MD simulations, the energy minimization of the system was performed by the steepest descent algorithm. The MD simulations were carried out in the constant volume and temperature (NVT) ensemble, with the temperature controlled at 300 K by the Nosé–Hoover thermostat (Evans and Holian, 1985). A time step of 2.0 fs was employed for all MD simulations. Lennard–Jones (LJ) and Coulomb potentials were used to simulate nonbonded interactions, with the electrostatic interactions calculated using the particle mesh Ewald (PME) summation method (Darden et al., 1993) and a 10 Å real-space cutoff for nonbonded interactions. The hydrogen-involved bonds were treated via the LINCS algorithm (Hess et al., 1997). In each umbrella sampling window, the system was equilibrated for 5 ns. The free-energy profile was calculated by the weighted histogram analysis method (WHAM) (Kumar et al., 1992).

2.3 BOMD Simulations

To explore the interfacial reaction of I₂O₅, the underlying mechanism was studied by BOMD and the stepwise multi-subphase space metadynamics (SMS-MetaD) (Fang et al., 2022) simulations using the CP2K 2022 program (Kühne et al., 2020) combined with PLUMED software (Bussi and Tribello, 2019). The SMS-MetaD method is well-suited for effectively modeling chemical systems with complex potential energy surface, especially at the air-water interface (Fang et al., 2022), which has already been successfully employed in the studies of heterogeneous reactions (Fang et al., 2024a, b; Tang et al., 2024; Wan et al., 2023). The QUICKSTEP module was applied with the Gaussian and plane wave (GPW) method, and the electronic exchange–correlation term was described by the BLYP-D3 functional (Perdew and Wang, 1992). The DZVP-MOLOPT-SR-GTH basis set and Goedecker–Teter–Hutter (GTH) pseudopotentials (Goedecker et al., 1996; Hartwigsen et al., 1998) were adopted here. The plane-wave cutoff was set to 280 Ry, and that for Gaussian was 40 Ry. All BOMD simulations were performed in the NVT ensemble, with the temperature controlled at 300 K by the Nosé–Hoover thermostat (Evans and Holian, 1985), and the time step was set to 1.0 fs. The dimensions of 3-D periodic cell are x=18 Å, y=18 Å, and z=30 Å, in which the water slab containing 128 H₂O was fully relaxed to reach a statistical equilibrium for temperature and potential energy (see Fig. S4). To probe the reaction barriers of I₂O₅ hydrolysis mediated by different chemicals, a series of SMS-MetaD simulations were carried out. In the SMS-MetaD simulations, collective variable (CV), as the key parameter, was set to effectively differentiate distinct states (Figs. S3 and S8–S9), with its upper and lower limits referring to the interfacial behavior observed in unbiased simulations (see more details in the Supplement). Gaussian hills with adaptive heights and sigma widths of 0.1 Å were deposited every 50 steps to efficiently explore the free energy landscape and accelerate the convergence (Figs. S10–S13). And block average analysis was performed to assess the error associated with the calculated reaction barriers (Fig. S14). The wave function analysis and visualization of key structures from the BOMD simulations were carried out using VMD 1.9.3 (Humphrey et al., 1996) and Multiwfn 3.7 software (Lu and Chen, 2012).

3 Results and Discussion

3.1 Surface Preference

The interfacial preference of I₂O₅ influences whether its heterogeneous reactions can effectively occur. Herein, umbrella sampling was applied to obtain the free energy profile of I₂O₅ from the bulk water to the air along the Z direction, where I₂O₅ was initially placed in the center of water layer. As shown in Fig. 1, the air-water interface is defined as the region with 10 %–90 % of the bulk density, delineated by the two dashed lines. As I₂O₅ moves from the bulk water into air across the interface, the free energy first decreases and then increases, reaching the lowest values near the Gibbs dividing surface (GDS, the blue dash line). The results show that I₂O₅ has an evident preference for aqueous surface, suggesting its potential to react with other chemicals at the air-water interface.

Figure 1 Interfacial propensity of I₂O₅. Mass density (g dm⁻³) of the water (black line); the free energy of I₂O₅ (purple line) along the z axis (Å). The vertical dashed line indicates the Gibbs dividing surface (GDS), which corresponds to a location with half of the bulk density. The shaded region represents the air-water interface (10 %–90 % of bulk density).

3.2 H₂O-Mediated Interfacial Hydrolysis of I₂O₅

Experimental evidence (Gómez Martín et al., 2022a) has confirmed a high reactivity of I₂O₅ in the presence of H₂O, which stands in contrast to theoretical predictions based on gas-phase reactions (Khanniche et al., 2016; Xia et al., 2020). To address this gap, we first probed unexplored heterogeneous hydrolysis of I₂O₅ at the air-water interface by the BOMD simulations. The results indicate no apparent tendency for I₂O₅ to react with interfacial H₂O to form HIO₃ in the unbiased BOMD simulation within 50 ps (Fig. S7), implying that this process may involve an energy barrier. Accordingly, the enhanced sampling (i.e., SMS-MetaD) simulations were further performed to examine the interfacial reactions of I₂O₅ and to quantify the associated reaction barrier (see CV settings in the Supplement). Figure 2a–b illustrates the process of I₂O₅ hydrolysis, where the reaction initiates with proton (H⁺) transfer from H₂O to I₂O₅ facilitated by another H₂O as a “bridge”. Then I₂O₅ binds with H⁺ to form the intermediate HI₂O${}_{\mathrm{5}}^{+}$, with the system existing in a quasi-transition state (TS). Subsequently the I–O bond in the HI₂O${}_{\mathrm{5}}^{+}$ breaks, and the OIO⁺ motif quickly combines with the OH⁻ to form a HIO₃, leaving another HIO₃. Figure 2c presents the free energy profile for the interfacial hydrolysis of I₂O₅, showing a reaction barrier of ∼ 9.4 kcal mol⁻¹ upon convergence of the results (Fig. S10), with ±0.1 kcal mol⁻¹ error as indicated by the block analysis (see more details in the Supplement), which is lower than that of the reported corresponding gas-phase reaction (13.6 kcal mol⁻¹) (Xia et al., 2020). However, this barrier remains considerable, limiting the direct heterogeneous hydrolysis from proceeding rapidly. The previous study (Kumar et al., 2018) reported the formation of H₂I₂O₆ from the reaction of I₂O₅ with interfacial water and indicated its dynamic stability over a 35 ps simulation period. However, H₂I₂O₆ was unfortunately not observed in the reported iodine-related experimental and field studies (Gómez Martín et al., 2020, 2022a; He et al., 2021, 2023; Rörup et al., 2024; Sipilä et al., 2016), nor was it observed in our unbiased BOMD simulations. Thus, the H₂O-mediated heterogeneous mechanism still faces challenges in representing the identified I₂O₅-to-HIO₃ conversion, pointing to the faster pathways mediated by other chemicals responsible for the absence of I₂O₅ in the experimental or field observations.

Figure 2 Details of H₂O-mediated hydrolysis of I₂O₅. (a) Snapshot structures of SMS-MetaD simulation representing reactant complex (RC), transition state (TS), and product complex (PC). The pink, red and white spheres represent I, O, and H atoms, respectively (the same applies in Figs. 3–6 below). (b) Reaction mechanism for the H₂O-mediated hydrolysis of I₂O₅. Blue molecular indicates the water involved in hydrolysis, black one indicates the I₂O₅, red one indicates the catalyst. The arrows indicate the direction of atom transfer (the same below). (c) Free energy profile for I₂O₅ hydrolysis at the air-water interface mediated by H₂O (black line) with the error band (pink region).

3.3 Self-Mediated Hydrolysis of I₂O₅

Given that H₂O-mediated hydrolysis is inefficient, we next explored whether leftover I₂O₅ can self-catalyze a faster reaction – a critical question for explaining its atmospheric depletion. During the unbiased BOMD simulation, two I₂O₅ molecules combined together to form iodine oxide complex (I₂O₅⋯I₂O₅) via intermolecular halogen bond (XB), without undergoing hydrolysis within 50 ps (Fig. S7). Further SMS-MetaD simulations, shown in Fig. 3a–b, present the mechanistic details of this hydrolysis. The I₂O₅⋯I₂O₅ complex initially binds a H₂O molecule via hydrogen bond (HB) and XB, forming a three-membered ring. Next, the H₂O molecule dissociates into OH⁻ and H⁺, each approaching a distinct I₂O₅. Meanwhile, the H₂O-involved HB (O–H⋯O) and XB (O–I⋯O) generally converted into new H–O and I–O covalent bonds, resulting in the formation of two HIO₃ molecules. And the residual OIO and IO₃ of the two I₂O₅ molecules combined to form a new I₂O₅, finalizing the self-catalyzed reaction. As shown in Fig. 3c, the resulting reaction barrier is 3.5 kcal mol⁻¹, which is 5.9 kcal mol⁻¹ lower than that of H₂O-catalyzed mechanism. The results suggest that I₂O₅ facilitates its own hydrolysis more pronouncedly than H₂O. Thus, the I₂O₅-mediated mechanism efficiently drives I₂O₅-to-HIO₃ conversion, resolving the earlier paradox of I₂O₅'s stability yet absence in the atmosphere – and explaining the high HIO₃ levels in marine aerosols.

Figure 3 Details of I₂O₅-mediated hydrolysis of I₂O₅. (a) Snapshot structures of SMS-MetaD simulation representing RC, TS, and PC. (b) Reaction mechanism for the I₂O₅-mediated hydrolysis of I₂O₅. (c) Free energy profile for I₂O₅ hydrolysis at the air-water interface mediated by I₂O₅ (black line) with the error band (pink region).

3.4 HIO₃-Mediated Interfacial Hydrolysis of I₂O₅

We further performed SMS-MetaD simulations to explore the interfacial reaction mediated by the product HIO₃, due to its widespread presence in gas and aerosol phase. And HIO₃ possesses both proton donor and acceptor sites, potentially facilitating proton transfer, which is a key step in I₂O₅ hydrolysis. As shown in Fig. 4a–b, H⁺ transfers from H₂O to HIO₃, then HIO₃ donates H⁺ to I₂O₅, finally yielding two new HIO₃. The resulting energy barrier of this process is ∼ 4.3 kcal mol⁻¹ (Fig. 4c), slightly higher than that mediated by I₂O₅ (3.5 kcal mol⁻¹). However, given the high concentration and widespread distribution of HIO₃ over oceans, its role in mediating hydrolysis of I₂O₅ should be fully considered. Moreover, as this reaction proceeds, the product HIO₃ accumulates to further catalyze the reaction, establishing positive feedback. Thus, in the pristine iodine-rich region such as Mace Head, with HIO₃ concentrations as high as 10⁸ molec. cm⁻³ (Sipilä et al., 2016) the HIO₃-catalyzed heterogeneous hydrolysis of I₂O₅ may represent its primary loss pathway. This may help explain why I₂O₅, though hypothesized to be involved in aerosol formation, is not observed in mass spectra, where iodate signals appear instead.

Figure 4 Details of HIO₃-mediated hydrolysis of I₂O₅. (a) Snapshot structures of SMS-MetaD simulation representing RC, TS, and PC. (b) Reaction mechanism for the HIO₃-mediated hydrolysis of I₂O₅. (c) Free energy profile for I₂O₅ hydrolysis at the air-water interface mediated by HIO₃ (black line) with the error band (pink region).

3.5 Sulfuric Acid-Mediated Interfacial Hydrolysis of I₂O₅

In fact, I₂O₅ seems not only absent over pristine oceans but also in polluted marine environments. For example, along the eastern coast of China (e.g., Zhejiang), bursts of iodine-bearing aerosols were observed, featuring high levels of IO${}_{\mathrm{3}}^{-}$, without I₂O₅ signal detected (Yu et al., 2019). Despite the proposed conversion of hydrated I₂O₅ to HIO₃ (detected as IO${}_{\mathrm{3}}^{-}$) (Gómez Martín et al., 2022a), the underlying mechanism remains unclear, especially under the impacts of pollutants. Notably, HSO${}_{\mathrm{4}}^{-}$ (ionization of typical acidic pollutant H₂SO₄) was also detected at considerable concentrations within iodine aerosols. Accordingly, we explored the H₂SO₄-mediated hydrolysis of I₂O₅ at the molecular level. As a strong acid (pK${}_{\mathrm{a}\mathrm{1}}=-\mathrm{3.0}$), H₂SO₄ can rapidly lose a proton to form HSO${}_{\mathrm{4}}^{-}$ at the air-water interface, which has a surface preference (Hua et al., 2015). At this stage, HSO${}_{\mathrm{4}}^{-}$ loses strong acidity and shows dissociation equilibrium, thus serving as a proton-transfer bridge. As shown in Fig. 5b, H⁺ initially moves from HSO${}_{\mathrm{4}}^{-}$ to I₂O₅, meanwhile, the H atom of H₂O is attracted by SO${}_{\mathrm{4}}^{\mathrm{2}-}$ and remains the OH⁻, leading to I–O bond breaking, forming two HIO₃ molecules. This process couples the self-dissociation of HSO${}_{\mathrm{4}}^{-}$ and the hydrolysis of I₂O₅, expected to occur efficiently on aqueous surface, owing to its lower barrier (∼ 1.3 kcal mol⁻¹) than that mediated by I₂O₅ (3.5 kcal mol⁻¹). The findings indicate that HSO${}_{\mathrm{4}}^{-}$-mediated pathway may play an important role in I₂O₅ depletion under polluted conditions, thereby providing mechanistic insights into the absence of I₂O₅ and the coexistence of HSO${}_{\mathrm{4}}^{-}$ and IO${}_{\mathrm{3}}^{-}$ in field-collected aerosols (Yu et al., 2019). To further quantify the efficiency of these reactions, we calculated the reaction rate constants through Transition State Theory (Table S1). Notably, the hydrolysis of I₂O₅ mediated by I₂O₅ or HIO₃ proceeds 4–5 orders of magnitude faster than that mediated by H₂O, but still more slowly (1–2 orders of magnitude) than that mediated by H₂SO₄. These results highlight the significance of intercomponent coupling in enhancing heterogeneous iodine chemistry in marine atmospheres, from pristine to polluted.

Figure 5 Details of SA-mediated hydrolysis of I₂O₅. (a) Snapshot structures of SMS-MetaD simulation representing RC, TS, and PC. Yellow atoms represent S. (b) Reaction mechanism for the SA-mediated hydrolysis of I₂O₅. (c) Free energy profile for I₂O₅ hydrolysis at the air-water interface mediated by SA (black line) with the error band (pink region).

3.6 Amine-Mediated Interfacial Hydrolysis of I₂O₅

Aside from acid pollutants, atmospheric bases may also affect interfacial chemistry under polluted conditions. Amines, as the typical bases, exhibit evident surface preference and the potential to affect hydrolysis of iodine oxides at the air-water interface (Ning et al., 2024). Accordingly, we investigated the amine-mediated hydrolysis of I₂O₅. Figure 6 shows the DMA-involved key structural snapshots and time-dependent evolution of key bond distances. Initially, DMA is positioned in the gas phase, without notable interaction with aqueous surface. Over time, DMA gradually approaches interfacial water to form HB (O3–H1⋯N). At 1.12 ps, interfacial water bridges I₂O₅ and DMA via HB (O3–H1⋯N) and XB (O2–I2⋯O3), forming a pre-reaction complex. By 1.19 ps, the covalent bond between the H1 and O3 elongated, forming a TS-like structure, then proton H1 transfers from H₂O to DMA. The deprotonated H₂O molecule, generating OH⁻, immediately approaches the I2 atom of I₂O₅. The I–O bond in I₂O₅ gradually changes from covalent bond into XB, ultimately yielding HIO₃ and IO${}_{\mathrm{3}}^{-}$ (Fig. S18). By 1.23 ps, the reaction ends to form HIO₃, IO${}_{\mathrm{3}}^{-}$, and alkylammonium salts (DMAH⁺). The BOMD results indicate that the DMA-mediated hydrolysis of I₂O₅ occurs rapidly on a picosecond timescale, thus attesting to a barrierless pathway. Also, we studied the MA- and TMA-mediated hydrolysis of I₂O₅, which shows similar capabilities of other amines like MA and TMA to accelerate this process (see Figs. S5 and S6). Accordingly, in polluted environments, the role of pollutants in activating iodine chemistry should not be overlooked, as their mediation of rapid heterogeneous processes likely affects aerosol composition and growth dynamics.

Figure 6 Details of DMA-mediated hydrolysis of I₂O₅. (a) Snapshot structures of BOMD simulation, which illustrate the mechanism for this process. The dashed lines indicate the intermolecular interactions (hydrogen or halogen bonding). Blue and cyan atoms represent N and C atoms, respectively. (b) Time evolution of key bond distances (O3–I2, H1–N, O2–I2, and O3–H1). The arrows indicate the direction of atom transfer.

In addition to the above analysis of the reaction mechanism and energy barriers, we further explored how interfacial hydrolysis of I₂O₅ affects aerosol growth (Scheme 1) through analysis of dynamic trajectories and wave function of key structures. Here, the DMA-mediated process is taken as an example (see details in the Supplement). The resulting products (HIO₃, IO${}_{\mathrm{3}}^{-}$, and DMAH⁺) are partially solvated, as they are surrounded by interfacial water (Figs. S15 and S16) via HBs and XBs but still exposed outside to air with unoccupied HB or XB sites (Fig. S17). These HB and XB interactions serve the following roles: (i) stabilizing the product complex by preventing its decomposition via intermolecular interactions within complex; (ii) inhibiting evaporation of product via interactions with interfacial water; (iii) promoting aerosol growth by unoccupied HB or XB sites for uptake of gaseous species, in particular, the hydrophilic ionic products (e.g., MAH⁺, DMAH⁺, and TMAH⁺) accelerate water condensation. Thus, the chemisorption of I₂O₅ contributes to aerosol growth not only directly, by yielding low-volatility products, but also indirectly by enhancing hygroscopic growth. Understanding the impacts of these heterogeneous reactions and the resulting products may advance our knowledge of how iodine oxides influence iodate aerosol formation.

Scheme 1 Illustration of aerosol growth driven by I₂O₅ hydrolysis at the air-water interface, highlighting the potential reaction pathways and resulting products in pristine and polluted environments.

4 Conclusions

Iodine chemicals are closely linked to marine aerosol formation; however, the underlying physicochemical process, especially involving higher-order iodine oxides (e.g., I₂O_x, x≥2), remains highly uncertain. HIO₃ is widely detected in marine aerosols, while I₂O₅ is absent, pointing to a potential I₂O₅-to-HIO₃ conversion, although the mechanism is still elusive. To address this, we have investigated interfacial hydrolysis of I₂O₅, the typical I₂O_x, on aqueous aerosol surfaces at the molecular level by BOMD and SMS-MetaD simulations. The results show that I₂O₅ uptakes directly onto the pure water surface, where the interfacial hydrolysis proceeds slowly (I₂O₅+ H₂O $\stackrel{{\mathrm{H}}_{\mathrm{2}}\mathrm{O}}{⟶}$ HIO₃ in Fig. 2). And the self-catalyzed pathways involving I₂O₅ (reactant) and HIO₃ (product) notably lower the energy barrier and facilitate the reaction. Thus, in pristine marine environments, the proposed I₂O₅- and HIO₃-mediated heterogeneous mechanisms (Figs. 3 and 4) potentially serve as both an important sink for I₂O₅ and a source of HIO₃, which helps clarify the chemical speciation of field-collected iodine aerosols. In contrast, in polluted regions, pollutant-mediated pathways – with barriers as low as 1.3 kcal mol⁻¹ (H₂SO₄-mediated) or even barrierless (amine-mediated) – dominate over self-catalyzed processes, driving faster I₂O₅ interfacial hydrolysis and yielding HIO₃, which helps to elucidate the rapid formation of iodate aerosols observed in polluted eastern coast of China (Yu et al., 2019). It also highlights the role of air pollutants in activating iodine chemistry. Accordingly, these heterogeneous mechanisms can advance our understanding of the roles of I₂O₅ in iodine chemistry, which may help explain the well-established but mechanistically elusive I₂O₅-to-HIO₃ conversion. Notably, the hydrolysis products (i.e. HIO₃ and ionic species) enhance aerosol growth through their low volatility, unoccupied HB/XB sites, and hygroscopicity – linking I₂O₅ hydrolysis directly to aerosol growth. These insights deepen our understanding of how iodine oxide affects aerosol formation – both directly and indirectly – in both pristine and polluted marine environments.

Altogether, this study highlights the critical role of heterogeneous hydrolysis of I₂O₅ mediated by some important chemicals in iodine chemistry and aerosol growth. Here, we find that intercomponent coupling – i.e., interactions of iodine oxides with themselves, iodic acids, and atmospheric pollutants – is a key, yet frequently overlooked factor that can reshape heterogeneous iodine chemistry across environments, where the reactive pollutant-mediated processes outcompete self-catalyzed pathways. These findings can help explain the observed but poorly understood I₂O₅ and HIO₃ distribution, as well as their chemical conversion, which sheds light on iodine aerosol growth. The iodine-driven impacts are likely global and expected to grow, due to rapidly increasing oceanic iodine emissions. Accordingly, integrating these proposed mechanisms into atmospheric models is expected to reduce uncertainties (unrecognized sink or source for iodine oxides and iodine oxyacids) arising from iodine oxides, which can refine the representation of iodine cycling and further improve the assessment of globally environmental and climatic effects (like aerosol-driven radiative forcing) driven by atmospheric iodine aerosols. The real atmosphere is chemically complex, including iodine species (e.g. HOI, HIO₂, HIO₃, I₂O₃, I₂O₄, and I₂O₅) and atmospheric pollutants (e.g. H₂SO₄, HNO₃, organic acids, and ammonia), which are likely to influence the heterogeneous hydrolysis of I₂O₅. In future work, we intend to confirm the impacts from other atmospheric components.