Molecular Dynamics Simulations of Water Molecule-Bridges in Polar

Aug 24, 2011 - ... of Environmental and Soil Chemistry, Universität Koblenz-Landau, ... Drazen Petrov , Daniel Tunega , Martin H. Gerzabek , and Chri...
0 downloads 0 Views 4MB Size
ARTICLE pubs.acs.org/est

Molecular Dynamics Simulations of Water Molecule-Bridges in Polar Domains of Humic Acids Adelia J. A. Aquino,*,†,‡ Daniel Tunega,† Hasan Pasalic,‡ Gabriele E. Schaumann,§ Georg Haberhauer,† Martin H. Gerzabek,† and Hans Lischka‡ †

Institute of Soil Research, University of Natural Resources and Applied Life Sciences Vienna, Peter-Jordan-Strasse 82, A-1190 Vienna, Austria ‡ Institute for Theoretical Chemistry, University of Vienna, W€ahringer Strasse 17, A-1090 Vienna, Austria § Institute for Environmental Sciences, Department of Environmental and Soil Chemistry, Universit€at Koblenz-Landau, Fortstrasse 7, 76829 Landau, Germany

bS Supporting Information ABSTRACT: The stabilizing effect of water molecule bridges on polar regions in humic substances (HSs) has been investigated by means of molecular dynamics (MD) simulations. The purpose of these investigations was to show the effect of water molecular bridges (WAMB) for cross-linking distant locations of hydrophilic groups. For this purpose, a tetramer of undecanoid fatty acids connected to a network of water molecules has been constructed, which serve as a model for spatially fixed aliphatic chains in HSs terminated by a polar (carboxyl) group. The effect of environmental polarity has been investigated by using solvents of low and medium polarity in force-field MD. A nonpolar environment simulated by n-hexane was chosen to mimic the stability of WAMB in a hydrophilic hotspot surrounded by a nonpolar environment, while the more polar acetonitrile environment was chosen to simulate a more even distribution of polarity around the carboxylic groups and the water molecules. The dynamics simulations show that the rigidity of the oligomer chains is significantly enhanced as soon as the water cluster is large enough to comprise all four carboxyl groups. Increasing the temperature leads to evaporization processes which destabilize the rigidity of the tetramer-water cluster. Embedding it into the nonpolar environment introduces a pronounced cage effect which significantly impedes removal of water molecules from the cluster region. On the other hand, a polar environment facilitates their diffusion from the polar region. One important consequence of these simulations is that although the local water network is the stabilizing factor for the organic matter matrix, the degree of stabilization is additionally affected by the presence of nonpolar surroundings.

’ INTRODUCTION Humic substances (HSs) constitute natural components of soils and sediments, being largely heterogeneous, control the fate of environmental pollutants, and the biogeochemistry of organic carbon in the global ecosystem.1 Partitioning models2,3 as well as polymer and polymer distribution models410 have been developed to characterize the molecular structure of HSs. Increasing evidence emphasizes the relevance of noncovalent intermolecular interactions (e.g., ππ, CH-π, van der Waals, charge-transfer, and hydrogen-bonding) between relatively small organic moieties11 and indicates a micellar structure of HSs protecting hydrophobic interiors from contact with neighboring water molecules.1214 A recent review on natural nanomaterials in soils15 reported that HSs are able to rearrange and restructure in response to environmental changes such as pH, ionic strength, and moisture. Within the supramolecular picture of HSs many studies have shown that individual compounds in soil organic material (SOM) can be bridged via water molecule bridges (WAMB)1618 and/or cation bridges (CAB).8,1921 In previous theoretical r 2011 American Chemical Society

investigations we have studied supramolecular interactions in HSs taking into account pH effects on the carboxylic groups and the effect of a water network interacting through hydrogen bonding with carboxylic groups has been demonstrated.16,19,22 Schulten et al.23 showed that in humic acids most hydrophilic functional groups are too far from each other to form direct intermolecular cross-links. They therefore require WAMB for intermolecular cross-linking. Swelling and wetting processes can drastically affect the sorption properties of SOM and the intrusion of contaminants and nutrients into soils.18 Differential scanning calorimetry (DSC) experiments24 on swelling peat samples showed an antiplasticizing effect of water on the organic matrix which increased with equilibration time and which decreased significantly after sudden changes of moisture conditions but Received: June 6, 2011 Accepted: August 24, 2011 Revised: August 18, 2011 Published: August 24, 2011 8411

dx.doi.org/10.1021/es201831g | Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology

Figure 1. Top and side view of the tetramer-water cluster.

reincreased after some time of storage.18,25 These phenomena were rationalized by formation and disruption of WAMBs17 and may further be responsible for the reduced dissolution rate of dissolved organic matter (DOM) from premoistened soil material.26 In the same way, it could prevent organic molecules from entering or leaving certain microregions in SOM17 unless those molecules disrupt the WAMB via competition for hydrogen bonds. It is, therefore, important to obtain advanced understanding of interactions between water, small organic molecules, and organic material forming intermolecular cross-links to support the above-mentioned models. While the former studies demonstrated the relevance of WAMBs for intermolecular crosslinking under various spatial and pH conditions, it is unknown to which extent the nanospatial heterogeneity in the organic matter affects the relevance and stability of the WAMB. Furthermore, the effect of the polarity of the nanoenvironment on the stability of the WAMB still remains unresolved, and the direct link to experimental data, e.g., the DSC step transition temperature T*,17,24,27 is still open. Modeling of water-OM interactions in humic substances requires simulation of situations of limited mobility and large distances between functional groups, for example by using a reduced molecular structure and at the same time fixing certain points in space as done in the two preceding studies.19,28 This concept reduces the complexity of the molecule to the characteristics of interest without having to model complete polymeric structures. As prototype for modeling polar interactions

ARTICLE

the carboxyl group represents a good choice since it is found in humic substances in a high content. For example, it has been found that the Swannee River humic substance contains 9.6 mol kg1 of carboxyl carbon.29 In this context classical molecular dynamics simulations have been performed, and interactions with water have been studied for fulvic acid model.30 It also should be noted that carboxyl groups are responsible for forming negatively charged sites in HSs already at relatively low pH values (dissociation starts at pH > 4.4).29 The concept of limited mobility and focus on the carboxyl group is also used in this work, where a region of a polar center interacting with a water cluster has been constructed by using a set of four medium-sized fatty acid segments which serve as models for spatially fixed aliphatic chains in HSs terminated by a polar (carboxyl) group. The arrangement of these four chains is shown in Figure 1. Each chain has a great torsional flexibility due to a long aliphatic tail. The nonpolar ends of the chains were kept fixed in order to simulate the situation of a stiff backbone such that the carboxyl groups cannot approach each other to undergo direct interactions but have some spatial mobility. Only this fixation of the nonpolar ends of the fatty acid segments allows for mimicking the crucial situation in humic substances this study aims for. The hydrophilic carboxyl heads were immersed into a cluster of water molecules to investigate local wetting and drying effects on the structural and energetic stability of the system (Figure 1). Environmental effects were included by means of embedding this aggregate into solvents (n-hexane and acetonitrile) presenting different polarity to test the stability of the WAMB toward an excess of small organic molecules. The nonpolar environment simulated by n-hexane was chosen to mimic the stability of WAMB in a hydrophilic hotspot surrounded by a nonpolar environment, while the more polar acetonitrile environment was chosen to simulate a more even distribution of polarity around the carboxylic groups and the water molecules. Several temperatures ranging from 300 K to 420 K were used in the dynamics simulations in order to investigate temperature stability on the rigidity of the WAMB network and to obtain first links with the experimental WAMB transition temperature T*.17,24,27 In addition to modeling the interactions with neutral carboxyl groups, selected cases of deprotonation (carboxylate) were studied as well in order to account for the importance of negatively charged sites mentioned above. In continuation of our previous philosophy, molecular dynamics (MD) simulations are performed, primarily at quantum mechanical level using the semiempirical density functional based tight-binding31,32 (DFTB) method using the self-consistent charge (SCC) approach.33 The use of an ab initio approach is too expensive for such big systems especially in view of the dynamics simulations. However, it has been shown16,19 that the DFTB method represents a very good compromise in terms of accuracy for describing hydrogen-bonded interactions and computational efficiency for performing MD simulations. Explicit inclusion of solvent molecules would lead to MD simulation times, which even at the DFTB level were not practical. Therefore, in this case the molecular mechanics (MM) method based on empirical interatomic force field (FF) has been used.

’ METHODS The structural model used to describe the interactions of HS segments (Figure 1) is composed of four fatty acid chains 8412

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology (tetramer Tm) using the undecanoid acid, CH3(CH2)9COOH. The positions of the terminal methyl carbon atoms were kept fixed during the simulations at a distance of 20 Å. The carboxylic groups represent a hydrophilic center which is immersed into a cluster consisting of 20 water molecules (Tm-(H2O)20 model). Wetting and drying processes were simulated by removing two water molecules from the initial Tm-(H2O)20 model in steps of two creating models Tm-(H2O)n, where n = 18...2. Two water molecules were taken arbitrarily from ten different positions in the model Tm-(H2O)20. In this way ten different structures Tm-(H2O)18 were created, and MD simulations were performed by starting from them. Based on the average energy computed for each MD the five most stable cases were selected. Two water molecules were randomly taken from different positions of the final structures of each of the aforementioned five selected MD simulations constructing new ten Tm-(H2O)16 models. In this way we continued in the construction of Tm-(H2O)n down to n = 2 always using ten independent MD simulations. A single deprotonation was performed at each of the four carboxyl groups individually. Test calculations showed that in such a case the SCC iterations in the DFTB approach did not converge due to the negative overall charge. To achieve charge neutrality, the carbon atom of the terminal methyl group in the chain containing the carboxylate group was replaced by N+ in the sense of an internal counterion, leading to overall charge neutrality. The SCC calculations converged, and this technique was used successfully in the DFTB MD calculations. In the final analysis an average over all four deprotonation cases was performed. The MD simulations were carried out at the DFTB level in the canonical (NVT) ensemble at T = 300 K using the Andersen thermostat.34 A time step of 1 fs was used in the Velocity Verlet integration algorithm35 for a simulation time of 50 ps for each case. The analysis of the MD data was performed for the last 30 ps of the simulations. Each two hundredth configuration was selected for this purpose. In order to represent differences in nanospatial heterogeneity, we decided to simulate environments of differing polarity around the WAMB. Calculations in two solvents were performed using acetonitrile and n-hexane as representatives of nonpolar (relative dielectric constants εr = 2.02) and polar environment (εr = 37.7). The simulated system consisted of the Tm-(H2O)20 structure embedded in 357 acetonitrile and 143 n-hexane molecules, respectively, arranged in a cubic box with length of 32 Å in order to reproduce experimental densities36 of the solvents at T = 300 K. Periodic boundary conditions combined with the minimum image convention37 were applied in all three dimensions. Classical force field MD (FF-MD) simulations in the canonical (NVT) ensemble at T = 300 K were performed in this case. A time step of 1 fs and the Nose-Hoover38,39 thermostat were used. Both the intra- and intermolecular interactions were described by the all atom optimized parameters for liquid simulations (OPLSAA) force field.40 The total simulation time was 1 ns. To investigate the changes in energetic stability with respect to temperature as observed e.g. in differential scanning calorimetry (DSC) experiments,18 MD calculations were performed both for the isolated cluster (DFTB-MD) and in solvent (FF-MD) for temperatures of 300, 340, 380, and 420 K, respectively. The simulation time was 200 ps in the former case and 1 ns in the latter. Since the simulation including an environment aimed at the representation of a rigid matrix structure of varying polarity, the density of the solvent and the size of the unit cell were kept fixed at the values used for T = 300 K.

ARTICLE

Figure 2. z-density profiles for carbon atoms belonging to the carboxyl group and of all oxygen atoms for selected tetramer-water clusters.

The structural analysis comprises normalized one-dimensional atomic density profiles calculated along the z direction as indicated in Figure 1, the distribution of interchain CC distances of carbon atoms belonging to neighboring carboxyl groups, the average number of hydrogen bonds, and the incremental energy calculations with respect to the number of added/ removed water molecules. The atomic density profiles are normalized with respect to the total number of atoms in the cluster and the number of configurations. The DFTB calculations were carried out using the DFTB+ code.41 For the FF-MD calculations with the solvents n-hexane and acetonitrile the program TINKER42 was used.

’ RESULTS AND DISCUSSION Density Profiles. In this section the average spatial distribution of the water molecules and carboxyl groups obtained from MD simulations is examined by means of z density profiles and the distribution of interchain CC distances of the carboxyl carbon atoms. Atomic density profiles were calculated for the oxygen atoms belonging both to the water molecules and the carboxyl groups as well as profiles of the carbon atoms of the carboxylic groups. The major results are presented in Figures 2-8; more details can be found in the Figures 1S-6S of the Supporting Information. 8413

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology

Figure 3. Histograms of interchain CC distances of carbon atoms belonging to neighboring carboxyl groups for selected tetramer-water clusters.

The Tm-(H2O)n Complexes. Starting the discussion with the Tm-(H2O)4 cluster (Figure 2a) one can see that the distribution of the carbon atoms is structured into two peaks, covering a region of about 4 Å. The distribution of oxygen atoms is somewhat broader than that for the C atoms. The carboxyl CC distances (Figure 3a) extend over a broad range with major contributions between 10 and 18 Å characterizing the large flexibility in the motion of the individual chains. Additional peaks at about 4 Å indicate specific, direct interactions between chain pairs. For n = 14 the water cluster is large enough to connect the carboxyl groups of all four chains. Both z-profiles and CC distance histograms (Figures 1Sb and 2Sb) exhibit stronger localization of the carboxyl groups illustrating nicely the stabilizing capability of the water clusters on restricting the motion of these groups. This trend is preserved in the 18-water complex (Figures 2b and 3b). In contrast to the structural changes observed for the carboxyl groups, the water molecules remain spread out with increasing number but always exhibit a strong overlap with the distribution of the carbon atoms (Figure 2 and Figure 1S). Especially for the 14- and 18-water complexes the

ARTICLE

Figure 4. DFTB simulations on singly deprotonated carboxyl groups for a 20 water cluster: (a) z-density profiles for carbon atoms belonging to the carboxyl/carboxylate groups and of all oxygen atoms and (b) histograms of interchain CC distances of carbon atoms belonging to neighboring carboxyl/carboxylate groups.

oxygen distribution is significantly wider than the carbon distribution because of the high fluctuation of molecules in clusters with larger water content. An analysis of the analogous dynamics of deprotonated carboxyl groups in contact with a cluster of 20 water molecules is presented in Figure 4. Comparison of the z-profiles shown in Figure 4a with those of Figure 2b indicates similar localization of the carboxyl/carboxylate groups in the neutral and deprotonated cases. The distribution obtained in the latter case is slightly more extended. More significant differences are found for the water distributions. For the carboxylate case the water distribution appears smoother but also shows a significant overlap with the distribution of the carboxyl/carboxylate groups. Finally, the CC distance histograms (Figure 4b) display the same compactness of the carboxyl/carboxylate groups due to the water network as in the case of the neutral carboxyl groups (Figure 3b). Thus, the simulations show that also in the single deprotonation case the surrounding water network is capable of keeping the polar groups together. 8414

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology

Figure 5. Temperature dependence of z-density profiles for carbon atoms belonging to the carboxyl group and of all oxygen atoms for the Tm-(H2O)20 cluster.

Temperature Effects. Figures 5 and 3S display the oxygen and carbon z-profiles derived from the dynamics of the Tm-(H2O)20 cluster computed at different temperatures. These calculations should set a first simple step toward the interpretation of the DSC experiments on samples of soil organic matter.18 The z-profiles for the oxygen atoms display an increasing fluctuation and evaporation of the water molecules with increasing temperature within the MD simulation period. With decreasing number of water molecules available for linking the carboxyl groups, the still unvaporized water molecules remain located in the bridging region, covering, however, only two to three carboxyl groups. Effect of Temperature and Environment. The density profiles for Tm-(H2O)20 in n-hexane (Figures 6, 4S, 7 and 5S) and acetonitrile (Figure 8) were computed at the FF level for different temperatures. The CC distance profiles for the isolated T-(H2O)18 complex (Figure 3b) and that for n-hexane solution (Figure 7a (T = 298 K)) display a similar spatial compactness of the carboxyl groups. This range of 412 Å remains almost unchanged up to 380 K (Figures 7 and 5S). Only at 420 K a distinct additional peak appears at around 20 Å because of the starting separation of one chain from the others. Visual inspection of the evolution of the cluster geometries and

ARTICLE

Figure 6. Temperature dependence of z-density profiles for carbon atoms belonging to the carboxyl group and of all oxygen atoms for the Tm-(H2O)20 cluster in n-hexane.

the fact that the maximum of the z-profile for the carboxyl carbon atoms is located at about zero in the FF dynamics (Figure 6) demonstrate that the whole carboxyl-water cluster undergoes an inversion motion along the z axis, a process which is not found in the DFTB dynamics (c.f. e.g. Figure 2b). However, as the above discussion shows, the major effect of the spatial confinement of the carboxyl groups by the water cluster, which is the focus of this work, is very similar in both methods. Comparison of the z-profiles for the oxygen atoms computed for the isolated cluster (Figure 2b) and for n-hexane solution (Figures 6 and 4S) manifests a remarkable cage effect of the hydrophobic environment of n-hexane on the water cluster. Even though the cases of the free cluster (DFTB) and the n-hexane solution (FF) are not exactly comparable since different methods are used, one can see for the free cluster that the distribution of the water molecules is significantly wider than that of the carboxyl groups (much broader oxygen distribution in comparison to the carbon distribution), whereas in n-hexane the water distribution is very similar to the one of the carboxyl groups (similar extension of the distributions). This fact indicates that in the latter case the water molecules stay always closer to the 8415

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology

ARTICLE

Figure 8. Temperature dependence of z-density profiles for carbon atoms belonging to the carboxyl group and of all oxygen atoms for the Tm-(H2O)20 cluster in acetonitrile.

Figure 7. Temperature dependence of histograms of interchain CC distances of carbon atoms belonging to neighboring carboxyl groups for the Tm-(H2O)20 cluster in n-hexane.

carboxyl groups due to the confining influence of the hydrophobic environment. The z-profiles obtained from the MD simulations of Tm-(H2O)20 in acetonitrile (Figures 8 and 6S) display completely different characteristics. The higher polarity of the solvent enhances the mobility of the water molecules substantially and, thereby, strongly weakens the stabilizing water network responsible for the compactness of the chain tetramer. The z-profiles are very broad and extend practically over the whole unit cell. Beginning with T = 340 K the oxygen atoms are almost uniformly distributed over a large range of z values. A corresponding behavior is observed for the carboxyl carbon atoms. Hydrogen Bond Analysis. Hydrogen bonding is the major interaction motif for the systems investigated in this work. Taken this fact into account, a detailed structural analysis of the hydrogen bonding between the carboxyl groups and water molecules is presented in this section. A hydrogen bond (HB) was considered to exist if the distance between an oxygen atom of the carboxyl group and the water oxygen was less than 3.5 Å and the HO 3 3 3 O angle was less than 30°.43 The

number of hydrogen bonds formed by all four carboxyl groups and averaged over the simulation time for the Tm-(H2O)n aggregates is plotted in Figure 9a. The number of HB increases with the number of waters molecules and reaches a certain limit between 7 and 8 bonds for n > 10. Starting with this threshold, a complete water network comprising all four carboxyl groups is being formed. The water network stabilization is also observed in terms of energetics as will be demonstrated in the next section. For n > 10 the number of hydrogen bonds displays some oscillation representing a statistical noise in the hydrogen bond analysis. To verify both the effect of the temperature and environment on the number of hydrogen bonds, an analysis of the isolated and solvated Tm-(H2O)20 clusters was performed (Figure 9b). As expected, the number of hydrogen bonds decreases with increasing temperature. In agreement with the cage effect of n-hexane discussed above, the average number of hydrogen bonds is somewhat higher with respect to the isolated complex. Obviously the water molecules cannot leave the region of the carboxyl/ water network in n-hexane when the temperature increases. In the case of acetonitrile the average number of hydrogen bonds between the carboxyl groups and the water molecules has decreased drastically. Increasing the temperature enhances this effect further. Energy Profiles. In this section a profile of incremental energy with respect to removing or adding water molecules in the Tm-(H2O)n complex is discussed. The incremental energy ΔEincr(n) was calculated according to ΔE̅ incr ðnÞ ¼ E̅ tot ½Tm-ðH2 OÞn2  þ E̅ tot ½ðH2 OÞ2   E̅ tot ½Tm-ðH2 OÞn , n ¼ 2:::18

ð1Þ

with Etot [(H2O)2] being the averaged total energy of two water molecules, and Etot [Tm-(H2O)n] and Etot [Tm-(H2O)n2] being the averaged energies of Tm-water clusters with n and n2 water molecules, respectively, as computed from the dynamics simulation. Figure 7S in Supporting Information depicts the incremental energy profile computed from MD simulations of the isolated clusters at room temperature. The trend displayed in Figure 7S follows approximately the one of the number of 8416

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology

Figure 9. (a) Average number of hydrogen bonds between the carboxyl groups and water molecules in Tm-(H2O)n clusters at 298 K and (b) temperature dependence of the average number of hydrogen bonds in the isolated Tm-(H2O)20 cluster and in n-hexane and acetonitrile.

hydrogen bonds given in Figure 9a. Removal of water molecules from the clusters with n > 10 requires more energy than for the clusters with smaller number of water molecules where the water network is not fully established. The distinction is, however, overall not very pronounced. Consequences. The dynamics simulations manifest that the compactness of the polar region of the aliphatic chains is significantly enhanced as soon as the water cluster is large enough to be able to connect carboxyl groups via a hydrogen bond network. In our simulations we used a concrete situation of the spatially distributed fatty acid chains immersed in water clusters of different size. Observable changes happen at around 1014 water molecules. For the isolated cluster a continuous increase of the evaporation of water molecules with increasing temperature is found leading to a decrease in the stability of the carboxyl/ water network and enhancing the mobility of the polar tails of the aliphatic chains. Embedding the whole water/tetramer cluster

ARTICLE

into different environments leads to interesting effects. In case of a nonpolar environment (n-hexane) the water cluster is tightly kept together due to the hydrophobicity of the surroundings, even at elevated temperatures. On the contrary, embedding into the polar environment (acetonitrile) leads to a dissociation of the water network and to a strong mobility of the aliphatic chains. The simulations have been performed mostly for chains containing neutral carboxyl groups. However, the investigations on singly deprotonated carboxylate/carboxyl systems interacting with a water cluster show that also in this case the binding power of the water network. One important consequence of these simulations is that although the local water network is the stabilizing factor for the organic matter matrix, the degree of stabilization is additionally affected by the presence of nonpolar surroundings. The study thus predicts only low stabilizing effects of WAMB in more homogeneous organic matter like weakly decomposed peat or cellulose-rich organic matter with high concentration and even distribution of hydrophilic functional groups. The stabilizing effect is expected to manifest itself best in organic matter with strong nanospatial heterogeneity in polarity as expected in stronger decomposed and degraded soil organic matter. As a consequence, the expression of WAMB and their role in stabilizing organic matter thus will increase with ongoing soil formation. This is in accordance with the increase in matrix rigidity, expressed as the DSC step transition temperature T*, in profiles of a podzol, two cambisols, and a luvisol under spruce,27 with increasing soil depth and thus with increasing degree of degradation as well as the increase in the T* of mineral organic associations with increasing soil age in a chronosequence in Hawaii.44 Our simulations suggest a central role of WAMB for stabilization of organic matter in soils: Organic matter stabilized by WAMB in a more nonpolar environment will be protected from further microbial degradation via two mechanisms: (i) exclusion of water from the outside and (ii) stabilization of the supramolecular cluster from the inside. In these terms, the increased protection of labeled organic carbon from mineralization with increasing hydrophobic character of humic matter observed by Piccolo et al.45,46 could be at least partly due to the WAMB stabilization in hydrophobic nanoenvironments.

’ ASSOCIATED CONTENT

bS

Supporting Information. Figures 1S-7S for z-density profiles and interchain CC distances of different water clusters, temperature dependence of z-profiles for the isolated Tm-(H2O)20 cluster and in n-hexane and acetonitrile, and variation of the incremental energy with respect to water content. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT We are grateful for the financial support from the Austrian Sciences Fund (project P20893-N19), and the German Research Foundation, the priority program SPP 1315, project nos. GE 1676/1-1 and SCHA849/8-1 and 8-2. The technical support and 8417

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology computer time at the Vienna Scientific Cluster (project 70055) is gratefully acknowledged.

’ REFERENCES (1) Aiken, G. R.; McKnight, D. M.; Warshaw, R. L.; MacCarthy, P. Humic substances in soil, sediment, and water; Wiley: New York, 1985. (2) Niederer, C.; Goss, K.-U.; Schwarzenbach, R. P. Sorption Equilibrium of a Wide Spectrum of Organic Vapors in Leonardite Humic Acid: Experimental Setup and Experimental Data. Environ. Sci. Technol. 2006, 40 (17), 5368–5373. (3) Goss, K. U. The air/surface adsorption equilibrium of organic compounds under ambient conditions. Crit. Rev. Environ. Sci. Technol. 2004, 34 (4), 339–389. (4) Leboeuf, E. J.; Weber, W. J. A distributed reactivity model for sorption by soils and sediments 0.8. Sorbent organic domains: Discovery of a humic acid glass transition and an argument for a polymer-based model. Environ. Sci. Technol. 1997, 31 (6), 1697–1702. (5) Xing, B. Sorption of naphthalene and phenanthrene by soil humic acids. Environ. Pollut. 2001, 111 (2), 303–309. (6) Kopinke, F. D.; Porschmann, J.; Stottmeister, U. Sorption of Organic Pollutants on Anthropogenic Humic Matter. Environ. Sci. Technol. 1995, 29 (4), 941–950. (7) Lu, Y. F.; Pignatello, J. J. Demonstration of the “Conditioning effect” in soil organic matter in support of a pore deformation mechanism for sorption hysteresis. Environ. Sci. Technol. 2002, 36 (21), 4553–4561. (8) Schaumann, G. E. Soil organic matter beyond molecular structure Part I: Macromolecular and supramolecular characteristics. J. Plant Nutr. Soil Sci. 2006, 169 (2), 145–156. (9) Schaumann, G. E. Soil organic matter beyond molecular structure Part II: Amorphous nature and physical aging. J. Plant Nutr. Soil Sci. 2006, 169 (2), 157–167. (10) Sutton, R.; Sposito, G. Molecular structure in soil humic substances: The new view. Environ. Sci. Technol. 2005, 39 (23), 9009–9015. (11) Simpson, A. J.; Kingery, W. L.; Hayes, M. H. B.; Spraul, M.; Humpfer, E.; Dvortsak, P.; Kerssebaum, R.; Godejohann, M.; Hofmann, M. Molecular structures and associations of humic substances in the terrestrial environment. Naturwissenschaften 2002, 89 (2), 84–88. (12) Piccolo, A. The supramolecular structure of humic substances. Soil Sci. 2001, 166 (11), 810–832. (13) Wershaw, R. L. Molecular aggregation of humic substances. Soil Sci. 1999, 164 (11), 803–813. (14) Schnitzer, M. A lifetime perspective on the chemistry of soil organic matter. Adv. Agron. 2000, 68, 1-58. (15) Wilson, M. A.; Tran, N. H.; Milev, A. S.; Kannangara, G. S. K.; Volk, H.; Lu, G. Q. M. Nanomaterials in soils. Geoderma 2008, 146 (12), 291–302. (16) Aquino, A. J. A.; Tunega, D.; Schaumann, G. E.; Haberhauer, G.; Gerzabek, M. H.; Lischka, H. Stabilizing Capacity of Water Bridges in Nanopore Segments of Humic Substances: A Theoretical Investigation. J. Phys. Chem. C 2009, 113 (37), 16468–16475. (17) Schaumann, G. E.; Bertmer, M. Do water molecules bridge soil organic matter molecule segments?. Eur. J. Soil Sci. 2008, 59 (3), 423–429. (18) Schaumann, G. E.; Leboeuf, E. J. Glass transitions in peat: Their relevance and the impact of water. Environ. Sci. Technol. 2005, 39 (3), 800–806. (19) Aquino, A. J. A.; Tunega, D.; Pasalic, H.; Haberhauer, G.; Gerzabek, M. H.; Lischka, H. The thermodynamic stability of hydrogen bonded and cation bridged complexes of humic acid models - A theoretical study. Chem. Phys. 2008, 349 (13), 69–76. (20) Aquino A. J. A.; Tunega D.; Schaumann G. E.; Haberhauer G.; Gerzabek, M. H.; H., L. The Functionality of Cation Bridges for Binding Polar Groups in Soil Aggregates. Int. J. Quantum Chem. 2011, 111 (78), 15311542. (21) Iskrenova-Tchoukova, E.; Kalinichev, A. G.; Kirkpatrick, R. J. Metal Cation Complexation with Natural Organic Matter in Aqueous Solutions: Molecular Dynamics Simulations and Potentials of Mean Force. Langmuir 2010, 26 (20), 15909–15919.

ARTICLE

(22) Aquino, A. J. A.; Tunega, D.; Pasalic, H.; Schaumann, G. E.; Haberhauer, G.; Gerzabek, M. H.; Lischka, H. Study of solvent effect on the stability of water bridge-linked carboxyl groups in humic acid models. Geoderma doi:10.1016/j.geoderma.2010.12.006. (23) Schulten, H.-R.; Leinweber, P.; Schnitzer, M. Environmental particles: structure and surface reactions of soil particle. In Analytical pyrolysis and computer modelling of humic and soil particles; Huang, P. M., Senesi, N., Buffle, J., Eds.; Wiley: Chichester, 1998. (24) Schaumann, G. E. Matrix relaxation and change of water state during hydration of peat. Colloids Surf., A 2005, 265 (13), 163–170. (25) Hurrass, J.; Schaumann, G. E. Influence of the sample history and the moisture status on the thermal behavior of soil organic matter. Geochim. Cosmochim. Acta 2007, 71 (3), 691–702. (26) Schaumann, G. E.; Siewert, C.; Marschner, B. Kinetics of the release of dissolved organic matter (DOM) from air-dried and premoistened soil material. J. Plant Nutr. Soil Sci. 2000, 163 (1), 1–5. (27) Hurrass, J.; Schaumann, G. E. Is Glassiness a Common Characteristic of Soil Organic Matter?. Environ. Sci. Technol. 2005, 39 (24), 9534–9540. (28) Aquino, A. J. A.; Tunega, D.; Schaumann, G. E.; Haberhauer, G.; Gerzabek, M. H.; Lischka, H. The Functionality of Cation Bridges for Binding Polar Groups in Soil Aggregates. Int. J. Quantum Chem. 2011, 111 (78), 1531–1542. (29) Baalousha, M.; Motelica-Heino, M.; Le Coustumer, P. Conformation and size of humic substances: Effects of major cation concentration and type, pH, salinity, and residence time. Colloids Surf., A 2006, 272 (12), 48–55. (30) Porquet, A.; Bianchi, L.; Stoll, S. Molecular dynamic simulations of fulvic acid clusters in water. Colloids Surf., A 2003, 217 (13), 49–54. (31) Elstner, M.; Porezag, D.; Jungnickel, G.; Elsner, J.; Haugk, M.; Frauenheim, T.; Suhai, S.; Seifert, G. Self-consistent-charge densityfunctional tight-binding method for simulations of complex materials properties. Phys. Rev. B 1998, 58 (11), 7260–7268. (32) Seifert, G.; Porezag, D.; Frauenheim, T. Calculations of molecules, clusters, and solids with a simplified LCAO-DFT-LDA scheme. Int. J. Quantum Chem. 1996, 58 (2), 185–192. (33) Porezag, D.; Frauenheim, T.; Kohler, T.; Seifert, G.; Kaschner, R. Construction of Tight-Binding-Like Potentials on the Basis of Density-Functional Theory - Application to Carbon. Phys. Rev. B 1995, 51 (19), 12947–12957. (34) Andersen, H. C. Molecular-Dynamics Simulations at Constant Pressure and-or Temperature. J. Chem. Phys. 1980, 72 (4), 2384–2393. (35) Ferrario, M.; Ryckaert, J. P. Constant Pressure-Constant Temperature Molecular-Dynamics for Rigid and Partially Rigid MolecularSystems. Mol. Phys. 1985, 54 (3), 587–603. (36) D’Ans, L. Taschenbuch f€ur Chemiker und Physiker, 3rd ed.; Springer Verlag: Berlin, Heidelberg, NY, 1967; Vol. 1. (37) Allen, M. P.; Tildesley, D. J. Computer Simulation of Liquids; Oxford University Press: New York: 1997. (38) Nose, S. A Unified Formulation of the Constant Temperature Molecular-Dynamics Methods. J. Chem. Phys. 1984, 81 (1), 511–519. (39) Hoover, W. G. Canonical Dynamics - Equilibrium Phase-Space Distributions. Phys. Rev. A 1985, 31 (3), 1695–1697. (40) Jorgensen, W. L.; Maxwell, D. S.; TiradoRives, J. Development and testing of the OPLS all-atom force field on conformational energetics and properties of organic liquids. J. Am. Chem. Soc. 1996, 118 (45), 11225–11236. (41) Frauenheim, T.; Seifert, G.; Elstner, M.; Niehaus, T.; Kohler, C.; Amkreutz, M.; Sternberg, M.; Hajnal, Z.; Di Carlo, A.; Suhai, S. Atomistic simulations of complex materials: ground-state and excitedstate properties. J. Phys.: Condens. Matter 2002, 14 (11), 3015–3047. (42) Ponder, J. W.; Richards, F. M. An Efficient Newton-Like Method for Molecular Mechanics Energy Minimization of Large Molecules. J. Comput. Chem. 1987, 8 (7), 1016–1024. (43) Luzar, A.; Chandler, D. Effect of environment on hydrogen bond dynamics in liquid water. Phys. Rev. Lett. 1996, 76 (6), 928–931. (44) Mikutta, R.; Schaumann, G. E.; Gildemeister, D.; Bonneville, S.; Kramer, M. G.; Chorover, J.; Chadwick, O. A.; Guggenberger, G. 8418

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419

Environmental Science & Technology

ARTICLE

Biogeochemistry of mineral-organic associations across a long-term mineralogical soil gradient (0.34100 kyr), Hawaiian Islands. Geochim. Cosmochim. Acta 2009, 73 (7), 2034–2060. (45) Piccolo, A.; Spaccini, R.; Haberhauer, G.; Gerzabek, M. H. Increased sequestration of organic carbon in soil by hydrophobic protection. Naturwissenschaften 1999, 86 (10), 496–499. (46) Spaccini, R.; Piccolo, A.; Conte, P.; Haberhauer, G.; Gerzabek, M. H. Increased soil organic carbon sequestration through hydrophobic protection by humic substances. Soil Biol. Biochem. 2002, 34 (12), 1839–1851.

8419

dx.doi.org/10.1021/es201831g |Environ. Sci. Technol. 2011, 45, 8411–8419