Understanding the properties of methanesulfinic acid at the air-water interface

Science of the Total Environment 668 (2019) 524–530

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Understanding the properties of methanesulfinic acid at the air-water interface Guochun Lv a, Heng Zhang b, Zehua Wang a, Ning Wang a, Xiaomin Sun a,⁎, Chenxi Zhang c, Mei Li d,e,⁎ a

Environment Research Institute, Shandong University, Jinan 250100, China School of Chemistry and Chemical Engineering, Shandong University, Jinan 250100, China College of Biological and Environmental Engineering, Binzhou University, Binzhou 256600, China d Institute of Mass Spectrometer and Atmospheric Environment, Jinan University, Guangzhou 510632, China e Guangdong Provincial Engineering Research Center for On-line Source Apportionment System of Air Pollution, Guangzhou 510632, China b c

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• The properties of MSIA at the air-water interface are studied using MD simulations. • The lowest free energy is located at the air-water interface. • Probability of MSIA at the interface is about 21%. • The tilted orientation at the interface is predominant. • The feature of hydration status at the interface is found.

a r t i c l e

i n f o

Article history: Received 17 January 2019 Received in revised form 27 February 2019 Accepted 3 March 2019 Available online 05 March 2019 Editor: Jianmin Chen Keywords: Methanesulfinic acid Air-water interface MD simulations Free energy Hydration status

a b s t r a c t Methanesulfinic acid (MSIA), an organic sulfur compound, is mainly produced in the oxidation process of dimethyl sulfide in the atmosphere. The properties of MSIA at the air-water interface were studied using molecular dynamics (MD) simulations. The result shows that the lowest system free energy is located at the interface. Because the free energy difference between the interface and water phase is 3.2 kJ mol−1, the MSIA molecule can easily get out of the free energy well and travel to water phase by the thermal motion, leading to only a 21% probability of its occurrence at the interface. The MSIA molecule tends to tilt at the interface with the sulfino group (-S (O)-OH) pointing toward the water phase. The feature of hydration status at the air-water interface may be favorable to the heterogeneous oxidation of MSIA. © 2019 Elsevier B.V. All rights reserved.

⁎ Corresponding authors. E-mail addresses: [email protected] (X. Sun), [email protected] (M. Li).

https://doi.org/10.1016/j.scitotenv.2019.03.032 0048-9697/© 2019 Elsevier B.V. All rights reserved.

1. Introduction

G. Lv et al. / Science of the Total Environment 668 (2019) 524–530

Sulfur-containing compounds emitted into the atmosphere have attracted tremendous interest of scientists for many years (Spiro et al., 1992; Andreae and Crutzen, 1997; Barnes et al., 2006; Faloona, 2009). Dimethyl sulfide (DMS) produced from phytoplankton in the oceans (Sunda et al., 2002; Li et al., 2014) is considered as the main source of natural sulfur emissions in the atmosphere (Barnes et al., 2006; Mardyukov and Schreiner, 2018). The oxidation of DMS is mainly initiated by OH radical during the day time (Barone et al., 1996; Barnes et al., 2006; Mardyukov and Schreiner, 2018) and NO3 radical in the night (Butkovskaya and LeBras, 1994; Allan et al., 2000; Barnes et al., 2006), and eventually leads to the formation of methanesulfonic acid (MSA) and sulfuric acid, both of which can contribute to the formation of non-sea salt sulfate (nss-SO42−) aerosol in the marine boundary layer (Lucas and Prinn, 2003; Hoffmann et al., 2016). Methanesulfinic acid (CH3S(O)OH: MSIA) is an important intermediate in the addition channel of DMS oxidation. It is mainly produced through the further oxidation of the significant intermediate (dimethyl sulfoxide, DMSO) of the addition channel. Urbanski et al. (1998) used the LFP-TDLAS technique to study the reaction of DMSO with OH radical. They found that MSIA is the main product in the reaction. The subsequent experiments also obtained the same results (Arsene et al., 2002; Kukui et al., 2003; Enami et al., 2016). The mechanism of OH + DMSO reaction has been studied widely using quantum chemistry calculations (Wang and Zhang, 2002; Resende et al., 2005; González-García et al., 2006; Baptista et al., 2008). These studies found that the channel involving the formation of MSIA is the dominant pathway, which is in agreement with the experimental result. The consumption of MSIA in the atmosphere mainly is its oxidation by OH radical and ozone in the gas phase, aqueous phase and at the air-water interface (Bardouki et al., 2002; Barnes et al., 2006; González-García et al., 2007; Tian et al., 2007; Hoffmann et al., 2016; Lv et al., 2019). The lifetimes of MSIA with respect to the reaction with OH radical are 3–5 h in the gas phase, 0.5 h in the aqueous phase, respectively (Barnes et al., 2006). The result implies that the aqueous oxidation of MSIA is important and the lost of MSIA in the atmosphere is fast. In addition, the large Henry's law constant indicates that MSIA can transport from air phase into the aqueous phase (Barnes et al., 2006). The partitioning of a gas molecule into the aqueous phase such as cloud droplets and aerosols needs firstly to be adsorbed on the surface before absorption into bulk solution (Davidovits et al., 1991). For some atmospheric species, the system free energy will increase when adsorbed molecules transform from the interface into the bulk liquid, leading to the surface preference of these species (Garrett et al., 2006; Hoehn et al., 2016; Sayou et al., 2017; Li et al., 2018; Wei et al., 2018). And the chemistry on cloud/aerosol water surfaces can play an important role in atmospheric chemical processes (Zhong et al., 2018). Thus, understanding the surface properties and surface preference of MSIA at the air-water interface can provide more information to scientists to further study heterogeneous oxidation of MSIA. Molecular dynamics (MD) simulation is a useful tool to study the surface properties on molecular level. The molecular details of the absorption of gas molecules, such as the free energy change and mass accommodation coefficient, have been widely studied (Julin et al., 2013; Julin et al., 2014; Ergin and Takahama, 2016; Hoehn et al., 2016; Sayou et al., 2017; Hirshberg et al., 2018; Li et al., 2018). The interaction, molecular orientation and structures at the air-water interface are also extensively probed using MD simulations (Ma et al., 2011; Blower et al., 2013; Habartová et al., 2014; Werner et al., 2014; Feng et al., 2016; Ekholm et al., 2018). However, to our knowledge, there are no researches on properties of MSIA at the air-water interface using MD simulations. In this work, we will use MD simulations to investigate the properties of MSIA at the air-water interface. The free energy change with the transformation of MISA from gas phase to aqueous phase is firstly discussed using the umbrella sampling method. To evaluate the surface preference of MSIA, we calculated its residence time at the air-water

525

interface. For MSIA, the interfacial orientation and hydrogen bonding interactions with water molecules are also been considered. 2. Computational methods The MD simulations were performed using GROMACS 2018 software package (Berendsen et al., 1995; Van Der Spoel et al., 2005; Abraham et al., 2015) with the general AMBER force field (GAFF) (Wang et al., 2004) in this work. The GAFF force field was developed with the aim of describing almost organic molecules made of C, N, O, H, S, P, F, Cl, Br and I atoms. The GAFF force field has been widely used in the studies involving the air-water interface (Warren and Patel, 2007; Habartová et al., 2014; Ergin and Takahama, 2016; Hirshberg et al., 2018; Li et al., 2018). The results from these studies have proven that the GAFF force field is suitable to predict the properties of the species at air-water interface. To obtain the force field parameters, electrostatic potential (ESP) calculations were carried out at HF/6-31G(d) level when geometry optimization was finished at MP2/6-31G(d) level. The geometry optimization and electrostatic potential (ESP) calculations were performed using Gaussian 16 suite of programs (Frisch et al., 2016). Based on ESP charges, the restrained electrostatic potential (RESP) charges were assigned to each atom using the Antechamber module of AmberTools 17 (Case et al., 2017). After the charges were determined, files including molecular coordinates and topologies with force field parameters were generated by tleap tool of AmberTools 17. ACPYPE (Sousa da Silva and Vranken, 2012) was used to convert these files to the GROMACS format. The force field parameters are put in Supplementary data (Fig. S1 and Table S1). In all MD simulations, the NVT ensemble at 298 K was used, and the temperature was controlled by velocity rescaling thermostat method (Bussi et al., 2007) with coupling time constant 0.1 ps. A time step of 2 fs was set and three-dimensional periodic boundary conditions were applied. The cutoff distances of van der Waals interactions, shortrange electrostatic interactions were set to 1.4 nm. Long-range electrostatic interactions were calculated using the particle-mesh Ewald (PME) summation method (Darden et al., 1993). The Verlet scheme was used as the cutoff scheme of neighbor searching (Páll and Hess, 2013). Bonding lengths were constrained using the LINCS algorithm (Hess et al., 1997). The extended simple point charge model (SPC/E) was employed to represent water molecule (Berendsen et al., 1987). The VMD software (Humphrey et al., 1996) was used to visualize molecular structures and configurations. 2.1. Free energy calculation To obtain free energy change when the MSIA molecule moves from gas phase into aqueous phase, the umbrella sampling method (Kästner, 2011) and weighted histogram analysis method (WHAM) (Kumar et al., 1992; Hub et al., 2010) were used. A cubic box of 5 nm side length with 4055 water molecules was firstly built. The water box was extended to 12 nm along the z axis direction. As shown in Fig. 1a, a single MSIA molecule was then put at the location with the coordinate (2.5 nm, 2.5 nm, 8.0 nm), which is the center of mass (COM) of the molecule. Thus, the COM of distance between the gas molecule and the water slab is 5.5 nm. Based on the ideal gas law, the average number of air molecules (N2 and O2) in the gas phase of the simulation box at 298 K is 3. Thus, the air phase in this work is approximatively treated as the void of air molecules, which is consistent with the description in the review (Garrett et al., 2006). For the new box, the equilibration with the restrained MSIA molecule at its initial location was carried out in NVT ensemble for 2 ns. After equilibration, to produce the umbrella windows, we need to pull the MSIA molecule from its initial location into the COM of water slab (Fig. 1a). The pull simulation was performed for 540 ps with pull force constant of 1000 kJ mol−1 nm−2 and the pull rate of 0.01 nm ps−1. In the pull simulation, the restraint of MSIA molecule at xy direction was used to eliminate the effect of its movement in xy direction on the reaction coordinate.

526

G. Lv et al. / Science of the Total Environment 668 (2019) 524–530

Fig. 1. Schematic representation of the umbrella sampling simulation (a) and the initial configuration of the 150 ns NVT simulation (b).

The position-restrained force constants in equilibration and pull simulation are 1000 kJ mol−1 nm−2. 50 umbrella windows were selected with the COM distance from 0 to 5 nm. For these 50 umbrellas windows, the brief NVT equilibration simulations (100 ps) were firstly executed, and then the data collection simulations were performed for 5 ns. 2.2. Properties of MSIA In this section, although the same water box of 5 × 5 × 5 nm3 obtained from above was elongated to 12 nm in z direction, we put the water slab in the middle of the box with the COM coordinate of (2.5 nm, 2.5 nm, 6.0 nm) and a MSIA molecule at (2.5 nm, 2.5 nm, 9.0 nm) (Fig. 1b). A 150 ns NVT simulation was performed. 3. Results and discussion 3.1. Free energy change of the system To understand the transformation process of MSIA molecule from gas phase into aqueous phase, free energy change of the system in this process was studied using the umbrella sampling method. The free energy profile is shown in Fig. 2. We also put water density in Fig. 2 so as to locate the air-water interface area, in which water density drops from

90% to 10% of its bulk value. To shows that the sampling is adequate, the histograms of the configurations within the umbrella sampling windows are given and put in Fig. S2. As we can see from Fig. 2, it is clear that the free energy do not change when the COM distance between MSIA and water slab is shortened from 5.0 nm to 3.5 nm (region c). Within this distance range, MSIA molecule is in gas phase because water density is 0 kg m−3, and is free due to no change of free energy, or in other words no interaction between it and water molecules. Starting from the COM distance of 3.5 nm (or 0.8 nm above the interface), the system free energy begins to decrease, indicating that interactions between MSIA molecule and water molecules occur from this distance. At the air-water interface, a free energy well can be found, and the minimum value of free energy is located at 2.3 nm (point b). As the MSIA molecule crosses the interface into bulk water, the system free energy increases. These phenomena imply the surface preference of MSIA. The solvation free energy (ΔGsolv) of a species is the difference of free energy in gas phase and in the solution. As shown Fig. 2, we use the free energy at point a (the COM of water slab) as the bulk value of free energy. Thus, for MSIA, the solvation free energy can be given by ΔGsolv = ΔGa − ΔGc ≈ −47.0 kJ mol−1, in which ΔGc refers to the mean value of free energy in region c, and ΔGa denotes the free energy of point a. The MD-calculated solvation free energy from free energy profile corresponds to the condition that the same concentration (c0 = 1 mol L−1) is used in gas and solution phases. Because the experimental value is always obtained under gas-phase standard state (p0 = 1 atm), the free energy correction for change of state in gas phase (1 mol L−1 → 1 atm) is considered so as to make this data more useful in experimental studies. The standard solvation free energy is given by the equation (Roeselová et al., 2004; Warren and Patel, 2007; Hoehn et al., 2016): ΔG0 = ΔGsolv + RTln(RTc0/p0) = ΔGsolv + 7.9 kJ mol−1 = −39.1 kJ mol−1, which is in agreement with the −38.3 kJ mol−1 in this work calculated using the DFT method (see Supplementary data for details). As the free energy profile is sensitive to the choice of the force field, thus it is usually used to validate the choice of the force field (Roeselová et al., 2004; Gladich et al., 2014; Hoehn et al., 2016). The result from the comparison between the MD-computed and DFTcomputed solvation free energy indicates that the selected GAFF force field is reliable in studying the interfacial properties of MSIA. The free energy difference from gas phase to the interface can be described as: ΔGc − ΔGb ≈ 50.2 kJ mol−1, where ΔGb is the free energy at point b (the minimum value). Crossing the free energy well into the water phase, MSIA needs to gain 3.2 kJ mol−1 of free energy (ΔGa − ΔGb), and no obvious free energy barrier is found. The result indicates that although there is the minimum free energy at the air-water interface, MSIA can easily transform from the interface into the bulk liquid by thermal motion. Thus, more discussion is required to assess the surface preference of MSIA.

3.2. Evaluation of surface preference

Fig. 2. The free energy profile and the density profile of water as the function of the COM distance between bulk water and the MSIA molecule.

To further evaluate whether MSIA molecule can get out of the free energy well and travel to water phase easily, a 150 ns NVT simulation is performed (Fig. 1b and Section 2.2). The time evolution of the position (z coordinate) of MSIA molecule (Fig. 3a) is monitored so as to observe whether the molecule stays at the air-water interface. The interface region (Fig. 3b) is also defined using the criterion mentioned above (10%– 90% of bulk water density). From Fig. 3a, we can see that MSIA can fully become solvated in water phase after staying for a while at the surface. As discussed above, the minimum value of the free energy located at the surface. Thus, the result indicates that MSIA can break the limitation of free energy well by thermal motion and move into water phase. As the system is inclined to the lowest free energy, the solvated molecule will go back to the interface. These two factors make the MSIA molecule shuttling back and forth between water phase and air-water interface.

G. Lv et al. / Science of the Total Environment 668 (2019) 524–530

527

Fig. 3. (a) The z coordinates of MSIA molecule as the function of simulation time; (b) the density profile of water; (c) the pie chart with the occurrence percentages of MSIA molecule at the air-water interface and in water phase.

The duration time for MSIA in every residence events at the airwater interface is different, and changes from hundreds of picoseconds to a few nanoseconds. In the 150 ns simulation, total duration time of MSIA at the interface approximately accounts for 21% of total simulation time (Fig. 3c). In other words, there is a 21% probability that MSIA molecule stays at the air-water interface after it is adsorbed at the interface. 3.3. Orientation and hydration environment Although the total duration time of MSIA molecule at the interface is shorter than that in water phase, the study on the interface properties of MSIA molecule is necessary due to the importance of interface chemistry. The interfacial orientation of MSIA is firstly considered. As shown in Fig. 4a, two angles are defined: one is the angle θ between the interface normal (z-axis) and the bisector of the included angle of two vectors (O1 → S and O2 → S); the other is the angle ϕ between the interface normal (z-axis) and the vector S → C pointing from the sulfur atom to the carbon atom. The probability distribution (Fig. 4b) of angle ϕ at different angle θ is obtained by averaging over the longest interface duration time of MSIA molecule (81.6 ns–84.6 ns). Although the MSIA molecule is not located at the interface in a fraction of the 3 ns duration time, it is near the defined interface, and is at the middle of the duration time (Fig. 3a). These points moved out the interface in the duration time are caused by the instantaneous fluctuations in the interface, and thus are also assigned to interface. From Fig. 4b, it can be seen that the two

angles both predominate in the range of 0°–90° and at three regions. These three regions are located around the points (30°, 58°), (54°, 36°) and (83°, 26°), respectively. The value of angle ϕ at these three representative points and the corresponding orientations are drawn in Fig. 5. It is clear that the methyl group of MSIA points toward gas phase while the sulfino group (-S(O)-OH) directs toward water phase. Although the methyl group is hydrophobic, the value angle ϕ at these predominant regions is not equal to zero, leading to the tilted orientation of MSIA molecule. The result indicates that there are some slight interactions between methyl group and water, which is in agreement with the literature (Hoehn et al., 2016). It can also be found from Fig. 4b that the different value of angle θ can be obtained when a value of angle ϕ is selected at these three regions. The result means that, when the tilted angle (ϕ) of MSIA molecule is identified, the rotation of the molecule around the axis composed with sulfur atom and the carbon atom is widespread. As shown in Fig. 5, through the rotation, the symmetry, asymmetry 1 and asymmetry 2 orientations can be found. Because there is only one symmetry orientation at the rotation process when the tilted angle (ϕ) is identified, the asymmetry orientations dominate the rotation orientation. In addition to the orientation of MSIA molecule, its hydration environment at the air-water interface is worthy to study. The spatial distribution functions (SDF) of water around the MSIA molecule at the interface (averaged from 81.6 ns to 84.6 ns) and in water phase (averaged from 85 ns to 88 ns) are considered so as to compare the different hydration status. As shown in Fig. 6a, when the MSIA molecule stay at

528

G. Lv et al. / Science of the Total Environment 668 (2019) 524–530

Fig. 4. (a) The schematic diagram about two angles (θ and ϕ); (b) the probability distribution of the two angles of MSIA molecule at the air-water interface.

the interface, water molecules are mainly located near the hydrogen and oxygen atom of sulfino group (-S(O)-OH), which indicates the existence of hydrogen bonding interaction between these atoms and water, and at the bottom of the MSIA molecule, which is the location of bulk

water. The SDF result at the interface means that the hydration shell only centers on the hydrogen and oxygen atoms of sulfino group. But the MSIA molecule in water phase is wrapped by hydration shell (Fig. 6b), indicating that it is fully solvated in water phase. As the sulfur

Fig. 5. Three representative values of the predominant regions of the angle ϕ and three different types of the rotation orientation.

G. Lv et al. / Science of the Total Environment 668 (2019) 524–530

529

Fig. 6. The spatial distribution functions (SDF) of water around the MSIA molecule at the air-water interface (a) and in water phase (b).

atom is not solvated at the interface and is the reactive site of the MSIA oxidation (González-García et al., 2007; Tian et al., 2007; Enami et al., 2016), the oxidants such as ozone and OH radical can easily approach MSIA and its reactive site without the obstruction of solvent, which may be favorable to the heterogeneous oxidation of MSIA at the airwater interface. The implication of the MD study in this work is that it provides a molecular level picture on the properties of MSIA at the air-water interface. The calculated free energy profile can give more insight into the uptake process of the gas molecule (MSIA) from gas phase into aqueous phase. These properties on molecular level are essential for further study on heterogeneous oxidation of MSIA. The information obtained from the MD study is an important complement to the experimental studies. 4. Conclusion The properties of methanesulfinic acid (MSIA) at the air-water interface have been studied using MD simulations. Based on umbrella sampling method, the free energy change with the movement of MSIA molecule from gas to water phase was obtained. The results show that the solvation free energy is −47.0 kJ mol−1 and the standard solvation free energy is −39.1 kJ mol−1. The shallow free energy well occurs at the interface. The further simulation shows that the MSIA molecule can get out of the well and moves to water phase easily by the thermal motion, leading to only a 21% probability at the interface. For the interfacial orientation of MSIA, the tilted orientations are mainly distributed near three values (58°, 36°, 26°) of angle ϕ. The asymmetry rotation orientations at each tilted orientation predominate. In addition, the hydration shell occurs only near the hydrogen and oxygen atoms of sulfino group (-S(O)-OH) when the MSIA stays at the interface. The characteristics may be favorable to the heterogeneous oxidation of MSIA at the air-water interface. Acknowledgements This work is supported by National Natural Science Foundation of China (21337001 and 21607011), Natural Science Foundation of Shandong Province (ZR2018MB043), and The Fundamental Research Funds of Shandong University (2018JC027). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2019.03.032. References Abraham, M.J., et al., 2015. GROMACS: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. SoftwareX 1-2, 19–25. Allan, B.J., et al., 2000. The nitrate radical in the remote marine boundary layer. J. Geophys. Res.-Atmos. 105, 24191–24204.

Andreae, M.O., Crutzen, P.J., 1997. Atmospheric aerosols: biogeochemical sources and role in atmospheric chemistry. Science 276, 1052–1058. Arsene, C., et al., 2002. Formation of methane sulfinic acid in the gas-phase OH-radical initiated oxidation of dimethyl sulfoxide. Environ. Sci. Technol. 36, 5155–5163. Baptista, L., et al., 2008. Theoretical investigation of the gas phase oxidation mechanism of dimethyl sulfoxide by OH radical. J. Mol. Struct-Theochem 851, 1–14. Bardouki, H., et al., 2002. Kinetics and mechanism of the oxidation of dimethylsulfoxide (DMSO) and methanesulfinate (MSI−) by OH radicals in aqueous medium. Atmos. Environ. 36, 4627–4634. Barnes, I., et al., 2006. Dimethyl sulfide and dimethyl sulfoxide and their oxidation in the atmosphere. Chem. Rev. (Washington, DC, U. S.) 106, 940–975. Barone, S.B., et al., 1996. Reaction of OH with dimethyl sulfide (DMS). 1. Equilibrium constant for OH + DMS reaction and the kinetics of the OH·DMS + O2 reaction. J. Phys. Chem. 100, 14694–14702. Berendsen, H.J.C., et al., 1987. The missing term in effective pair potentials. J. Phys. Chem. 91, 6269–6271. Berendsen, H.J.C., et al., 1995. GROMACS: a message-passing parallel molecular dynamics implementation. Comput. Phys. Commun. 91, 43–56. Blower, P.G., et al., 2013. Surface behavior of malonic acid adsorption at the air/water interface. J. Phys. Chem. A 117, 2529–2542. Bussi, G., et al., 2007. Canonical sampling through velocity rescaling. J. Chem. Phys. 126, 014101. Butkovskaya, N.I., LeBras, G., 1994. Mechanism of the NO3 + DMS reaction by discharge flow mass spectrometry. J. Phys. Chem. 98, 2582–2591. Case, D.A., et al., 2017. AMBER 2017. University of California, San Francisco. Darden, T., et al., 1993. Particle mesh Ewald: an N·log(N) method for Ewald sums in large systems. J. Chem. Phys. 98, 10089–10092. Davidovits, P., et al., 1991. Uptake of gas molecules by liquids: a model. J. Phys. Chem. 95, 6337–6340. Ekholm, V., et al., 2018. Strong enrichment of atmospherically relevant organic ions at the aqueous interface: the role of ion pairing and cooperative effects. Phys. Chem. Chem. Phys. 20, 27185–27191. Enami, S., et al., 2016. “Sizing” heterogeneous chemistry in the conversion of gaseous dimethyl sulfide to atmospheric particles. Environ. Sci. Technol. 50, 1834–1843. Ergin, G., Takahama, S., 2016. Carbon density is an indicator of mass accommodation coefficient of water on organic-coated water surface. J. Phys. Chem. A 120, 2885–2893. Faloona, I., 2009. Sulfur processing in the marine atmospheric boundary layer: a review and critical assessment of modeling uncertainties. Atmos. Environ. 43, 2841–2854. Feng, Y.-J., et al., 2016. π-Hydrogen bonding of aromatics on the surface of aerosols: insights from ab initio and molecular dynamics simulation. J. Phys. Chem. B 120, 6667–6673. Frisch, M.J., et al., 2016. Gaussian 16, Revision A.03, Wallingford CT. Garrett, B.C., et al., 2006. Molecular simulations of the transport of molecules across the liquid/vapor interface of water. Chem. Rev. (Washington, DC, U. S.) 106, 1355–1374. Gladich, I., et al., 2014. Adsorption, mobility, and self-association of naphthalene and 1methylnaphthalene at the water–vapor interface. J. Phys. Chem. A 118, 1052–1066. González-García, N., et al., 2006. Variational transition-state theory study of the dimethyl sulfoxide (DMSO) and OH reaction. J. Phys. Chem. A 110, 798–808. González-García, N., et al., 2007. Methanesulfinic acid reaction with OH: mechanism, rate constants, and atmospheric implications. J. Phys. Chem. A 111, 7825–7832. Habartová, A., et al., 2014. Partial hydration of n-alkyl halides at the water–vapor interface: a molecular simulation study with atmospheric implications. Theor. Chem. Accounts 133, 1455. Hess, B., et al., 1997. LINCS: a linear constraint solver for molecular simulations. J. Comput. Chem. 18, 1463–1472. Hirshberg, B., et al., 2018. N2O5 at water surfaces: binding forces, charge separation, energy accommodation and atmospheric implications. Phys. Chem. Chem. Phys. 20, 17961–17976. Hoehn, R.D., et al., 2016. Hydrogen bonding and orientation effects on the accommodation of methylamine at the air-water interface. J. Chem. Phys. 144, 214701. Hoffmann, E.H., et al., 2016. An advanced modeling study on the impacts and atmospheric implications of multiphase dimethyl sulfide chemistry. Proc. Natl. Acad. Sci. U. S. A. 113, 11776–11781. Hub, J.S., et al., 2010. g_wham—a free weighted histogram analysis implementation including robust error and autocorrelation estimates. J. Chem. Theory Comput. 6, 3713–3720.

530

G. Lv et al. / Science of the Total Environment 668 (2019) 524–530

Humphrey, W., et al., 1996. VMD: visual molecular dynamics. J. Mol. Graph. 14, 33–38. Julin, J., et al., 2013. Mass accommodation of water: bridging the gap between molecular dynamics simulations and kinetic condensation models. J. Phys. Chem. A 117, 410–420. Julin, J., et al., 2014. Near-unity mass accommodation coefficient of organic molecules of varying structure. Environ. Sci. Technol. 48, 12083–12089. Kästner, J., 2011. Umbrella sampling. Wiley Interdiscip. Rev. Comput. Mol. Sci. 1, 932–942. Kukui, A., et al., 2003. Gas-phase reactions of OH radicals with dimethyl sulfoxide and methane sulfinic acid using turbulent flow reactor and chemical ionization mass spectrometry. J. Phys. Chem. A 107, 5732–5742. Kumar, S., et al., 1992. The weighted histogram analysis method for free-energy calculations on biomolecules. I. The method. J. Comput. Chem. 13, 1011–1021. Li, C.-Y., et al., 2014. Molecular insight into bacterial cleavage of oceanic dimethylsulfoniopropionate into dimethyl sulfide. Proc. Natl. Acad. Sci. U. S. A. 111, 1026–1031. Li, W., et al., 2018. Free energy study of H2O, N2O5, SO2, and O3 gas sorption by water droplets/slabs. J. Chem. Phys. 148, 164706. Lucas, D.D., Prinn, R.G., 2003. Tropospheric distributions of sulfuric acid-water vapor aerosol nucleation rates from dimethylsulfide oxidation. Geophys. Res. Lett. 30. Lv, G., et al., 2019. Understanding the oxidation mechanism of methanesulfinic acid by ozone in the atmosphere. Sci. Rep. 9, 322. Ma, X., et al., 2011. Molecular dynamic simulation of dicarboxylic acid coated aqueous aerosol: structure and processing of water vapor. Phys. Chem. Chem. Phys. 13, 9374–9384. Mardyukov, A., Schreiner, P.R., 2018. Atmospherically relevant radicals derived from the oxidation of dimethyl sulfide. Acc. Chem. Res. 51, 475–483. Páll, S., Hess, B., 2013. A flexible algorithm for calculating pair interactions on SIMD architectures. Comput. Phys. Commun. 184, 2641–2650. Resende, S.M., et al., 2005. Theoretical study of the role of adducts in the atmospheric oxidation of dimethyl sulfoxide by OH, O2 and O3 and the kinetics of the reaction DMSO + OH. Chem. Phys. 309, 283–289. Roeselová, M., et al., 2004. Hydroxyl radical at the air–water interface. J. Am. Chem. Soc. 126, 16308–16309.

Sayou, M., et al., 2017. Energetic analysis of adsorption and absorption of small molecule to nanodroplet of water. J. Phys. Chem. B 121, 5995–6001. Sousa da Silva, A.W., Vranken, W.F., 2012. ACPYPE - AnteChamber PYthon Parser interfacE. BMC Res. Notes 5, 367. Spiro, P.A., et al., 1992. Global inventory of sulfur emissions with 1° × 1° resolution. J. Geophys. Res.-Atmos. 97, 6023–6036. Sunda, W., et al., 2002. An antioxidant function for DMSP and DMS in marine algae. Nature 418, 317. Tian, Y., et al., 2007. Ab initio study of the reaction of OH radical with methyl sulfinic acid (MSIA). Chem. Phys. 335, 133–140. Urbanski, S.P., et al., 1998. Mechanistic and kinetic study of the gas-phase reaction of hydroxyl radical with dimethyl sulfoxide. J. Phys. Chem. A 102, 10522–10529. Van Der Spoel, D., et al., 2005. GROMACS: fast, flexible, and free. J. Comput. Chem. 26, 1701–1718. Wang, L., Zhang, J., 2002. Ab initio study of reaction of dimethyl sulfoxide (DMSO) with OH radical. Chem. Phys. Lett. 356, 490–496. Wang, J., et al., 2004. Development and testing of a general amber force field. J. Comput. Chem. 25, 1157–1174. Warren, G.L., Patel, S., 2007. Hydration free energies of monovalent ions in transferable intermolecular potential four point fluctuating charge water: an assessment of simulation methodology and force field performance and transferability. J. Chem. Phys. 127, 064509. Wei, Q., et al., 2018. Molecular structure and adsorption of dimethyl sulfoxide at the air/ aqueous solution interface probed by non-resonant second harmonic generation. Phys. Chem. Chem. Phys. 20, 11758–11767. Werner, J., et al., 2014. Succinic acid in aqueous solution: connecting microscopic surface composition and macroscopic surface tension. Phys. Chem. Chem. Phys. 16, 21486–21495. Zhong, J., et al., 2018. Insight into chemistry on cloud/aerosol water surfaces. Acc. Chem. Res. 51, 1229–1237.