Molecular Insights into Glyphosate Adsorption to ... - ACS Publications

Article Cite This: Environ. Sci. Technol. 2018, 52, 1946−1953

pubs.acs.org/est

Molecular Insights into Glyphosate Adsorption to Goethite Gained from ATR-FTIR, Two-Dimensional Correlation Spectroscopy, and DFT Study Wei Yan† and Chuanyong Jing*,†,‡ †

Environ. Sci. Technol. 2018.52:1946-1953. Downloaded from pubs.acs.org by UNIV OF TEXAS AT EL PASO on 10/29/18. For personal use only.

State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China ‡ University of Chinese Academy of Sciences, Beijing 100049, China S Supporting Information *

ABSTRACT: Glyphosate (PMG) complexation on iron (hydr)oxides impacts its fate and transport in the environment. To decipher the molecular-level interfacial configuration and reaction mechanism of PMG on iron (hydr)oxides, the PMG protonation process, which influences the chemical and physical properties of PMG, was first determined using ATR-FTIR spectroscopy. The FTIR results reveal that the deprotonation occurs at carboxylate oxygen when pKa1< pH < pKa2, at phosphonate oxygen when pKa2< pH < pKa3, and at amino nitrogen when pH > pKa3. PMG complexation on goethite was investigated using in situ flow-cell ATR-FTIR, two-dimensional correlation spectroscopy (2D-COS), and density functional theory (DFT) calculations. The results indicate that the phosphonate group on PMG interacts with goethite to form inner-sphere complexes with multiple configurations depending on pH: binuclear bidentate (BB) and mononuclear bidentate (MB) without proton under acidic conditions (pH 5), mononuclear monodentate (MM) with proton and BB without proton at pH 6−8, and MM without proton under alkaline conditions (pH 9). Phosphate competition significantly impacted the PMG adsorption capacity and its interfacial configurations. As a result, the stability of the adsorbed PMG was impaired, as evidenced by its elevated leachability. These results improve our understanding of PMG-mineral interactions at the molecular level and have significant implications for risk assessment for PMG and structural analog pollutants.

■

INTRODUCTION

The disagreement in PMG deprotonation seriously hinders the molecular-level understanding of PMG adsorption on minerals including goethite,11−13 and divergent phosphonate group involved mechanisms have been proposed on PMG surface complexation. For example, Sheals et al.6 speculated that PMG was adsorbed on goethite surfaces via a primary monodentate complex and a secondary bidentate complex at neutral pH. Tribe et al.13 found that PMG only formed monodentate surface complexes in all the pH ranges. In contrast, Barja et al.14 identified two predominant surface species coexisting on goethite, namely, monodentate protonated and bridging bidentate complexes. Recent studies suggest that adsorbed ions on goethite may exhibit a variety of surface configurations depending on the exposed crystal faces, adding complexity to the structural identification.15−17

Glyphosate (N-(phosphonomethyl)glycine, PMG) is a nonselective, postemergence herbicide for weed control.1 Due to the worldwide application of PMG, its degradation, bioavailability, and transport in the environment are of great concern, and these environmental processes are mainly regulated by its chemistry at mineral/aqueous interfaces.2−5 PMG interacts with mineral surfaces through its three polar functional groups, namely, phosphonate, amine, and carboxylate (Figure S1, in the Supporting Information (SI)). These functional groups are readily deprotonated and/or dissociated depending on pH. The dissociation constants have been accurately determined, but the deprotonation sequence is still a subject of controversy. Some earlier studies reported that the deprotonation of PMG occurred at the carboxyl group when pKa1 < pH < pKa2, at the phosphonate group when pKa2 < pH < pKa3, and at the amine group when pH > pKa3 (SI Figure S2a).6−8 Recent DFT9 and NMR10 studies suggested a different deprotonation sequence which is carboxyl, amine, and phosphonate group as pH increase (Figure S2b). © 2018 American Chemical Society

Received: Revised: Accepted: Published: 1946

November 4, 2017 January 21, 2018 January 21, 2018 January 22, 2018 DOI: 10.1021/acs.est.7b05643 Environ. Sci. Technol. 2018, 52, 1946−1953

Article

Environmental Science & Technology

Figure 1. ATR-FTIR spectra of 1000 mg L−1 dissolved PMG (a) and sarcosine (b) in 0.1 M NaCl for pH values 1.5, 4.2, 9.0, and 12.5.

used as solvent in the FTIR experiments. The IR spectra were collected using a horizontal attenuated total reflectance (HATR) cell (PIKE Tech) with a 45°ZnSe (pH/pD 5−9) or Ge (beyond pH/pD range of 5−9) ATR crystal in 0.01−0.2 M NaCl (SI Figure S4). A total of 1000 and 256 scans were recorded for static and in situ flow cell ATR-FTIR spectrum, respectively, over 1800−800 cm−1 at a resolution of 4 cm−1.22 The details of static and in situ flow cell ATR-FTIR measurements are provided in the SI. For better identification of peaks, the peak fittings of all IR spectra were carried out with Peakfit v.4.12 software using the second derivative fitting algorithm. The goodness of fit R factor in all of the fitted spectra was greater than 0.999. The details are provided in the SI Figure S5. Frequency Calculations. The IR frequency of the PMG molecule and PMG surface complexes with different configurations were calculated using the Gaussian 09 program with the B3LYP hybrid DFT method.23 The cluster models of goethite, PMG, and PMG surface complexes were constructed and the details are provided in the SI. The cluster model results could only be used as a qualitative guide in IR peak analysis rather than as an absolute energy value. Because PMG surface complexes are not dispersion-dominated and no interaction energies are involved, the dispersion corrections to standard DFT functions are not applied. 2D-COS Analysis. In order to determine the origin of the IR bands in the obtained IR spectra, 2D-COS analysis was performed using 2Dshige software (Shigeaki Morita, Japan). Compared to conventional FTIR analysis, 2D-COS has the advantage of being able to distinguish one band from another by its origins, which has been successfully applied to probe complicated surface interaction processes in different systems.20−22 A detailed description of 2D-COS analysis is provided in the SI. Batch Competitive Adsorption Isotherms. To explore the impact of phosphate on PMG adsorption and the corresponding structures of PMG complexes, competitive adsorption experiments of PMG and phosphate on goethite surfaces were conducted at pH 7. The details of batch adsorption experiments and the sample analysis, including HPLC-ESI-MS/MS and colorimetric assay of PMG and phosphate, are provided in the SI.

The ambiguities in deprotonation sequence and surface molecular structure may largely be attributed to the lack of effective approach for the structural determination. As a widely used technique to probe the solid−liquid interface reactions, attenuated total reflectance (ATR) Fourier transform infrared (FTIR) spectroscopy often results in equivocal peak assignment due to its intrinsic low resolution to resolve convoluted peaks,18 which may impair the accurate identification of surface complexes. Thus, complementary techniques such as twodimensional correlation spectroscopy (2D-COS) analysis, and DFT calculations are central to re-examine the PMG deprotonation process and the surface complexation. The application of PMG is usually accompanied by phosphorus fertilization. The abundant phosphate in the soil can compete with PMG for the available surface sites on minerals, leading to the leaching and runoff of PMG to deeper layers of the soil.19 Nevertheless, the molecular-level impacts of phosphate on the PMG interaction with surfaces and the corresponding structures of PMG surface complexes are far from well understood. The objective of this study was therefore to explore the molecular-scale interaction of PMG on Fe(III) (hydr)oxide surfaces using multiple complementary techniques, including in situ flow-cell ATR-FTIR spectroscopy (in H2O and D2O), 2DCOS analysis, and DFT calculations. The insights gained from this study improve our knowledge in predicting the environmental fate of PMG and structural analog pollutants.

■

MATERIAL AND METHODS

Materials. All chemicals were analytical reagent grade or higher and were used without further purification. Milli-Q water (18.2 MΩ) for ATR-FTIR analysis was boiled for 60 min and cooled with N2 gas purging to remove CO2. Goethite was selected as a typical iron (hydr)oxide, considering its ubiquity in the environment and highly reactive surfaces. Goethite was synthesized via iron nitrate hydrolysis as detailed in our previous report (SI Figure S3).20,21 The BET surface area (84.7 m2/g) was determined with an ASAP2000 instrument (Micromeritics Instrument Corp.). The point of zero charge (PZC) of goethite was determined to be 8.9 with a Zetasizer Nano ZS (Malvern Instruments, U.K.). ATR-FTIR Measurements. ATR-FTIR measurements were performed using a Thermo-Nicolet Nexus 6700 FTIR spectrometer equipped with a liquid-nitrogen cooled MCT detector. Both H2O and D2O (99.9 atom % D, Sigma) were 1947

DOI: 10.1021/acs.est.7b05643 Environ. Sci. Technol. 2018, 52, 1946−1953

Article


■

(980 cm−1), E (1094 cm−1)) and a P−C vibration at about 750 cm−1.26 The third deprotonation occurs at the amine group to yield PMG3− (pKa3 = 11.0, SI Figure S1), evidenced by the diminishing intensity of δ(NH2) at 1567 cm−1 (Figure 1a and SI Figure S6).24 Concurrently, υas(COO−) on PMG and sarcosine shifted from 1605 to 1614 cm−1 to 1570 cm−1 (pH 12.5, Figure 1). The downward shift of υas(COO−) is mainly attributed to the loss of intramolecular hydrogen bonding as a consequence of R2NH2+ deprotonation.6 Similarly, the loss of intramolecular hydrogen bonding also results in the downward shift of υ(PO3) from 1094 to 1067 cm−1 (Figure 1). The above peak assignments could be further corroborated by the spectra of dissolved PMG in D2O (SI Figure S7). In general, due to the atomic-mass effect, the deuterium exchange could result in the red shifts and/or lowering the peak intensity, suggesting the presence of protons.27 On the other hand, blue or no identifiable frequency shift indicates the existence of nonprotonated groups or complexes.28 Compared with the spectra in H2O, a downward shift of υ(CO) in COOH and υ(P−OH) was observed at pD 1.5 by 9 and 20 cm−1, respectively (SI Figure S7a), indicating the presence of protons in both carboxylate and phosphonate groups of PMG. At pD 4.2 (SI Figure S7b), the carboxylate group was deprotonated, as evidenced by a upward shift of υas(COO−) at 1621 cm−1 with enhanced intensity due to decoupling from the bending modes of the protonated amine group, δ(NH2), and water molecules, δ(H2O). As pD increased to 9.0 and 12.5 (SI Figure S7c, d), the phosphonate group was deprotonated because the associated peaks (e.g., 1094 and 980 cm−1 at pD 9.0, 1067, and 980 cm−1 at pD 12.5) were at approximately the same frequencies as those in H2O. These results in D2O are in agreement with our determination of the PMG protonation process in H2O. Our peak assignment is also justified by the high consistency (r2 > 0.99) between the results of DFT calculations and deconvoluted ATR-FTIR spectra (SI Figure S6, S8, S9, Table S3−S6). Specifically, the DFT results for the PMG2− species are aligned with the deprotonated phosphonate group rather than the deprotonated amine group (slope: 0.99 vs 0.96, SD: 30 vs 59, SI Figure S10), supporting our FTIR results that the second deprotonation of PMG occurs at the phosphonate oxygen rather than the amino nitrogen. Surface Complexation of PMG on Goethite. The protonation process of PMG has a substantial impact on its solid−water interfacial interactions in the environment. Since PMG and PMG3− exist, respectively, under extremely acidic and alkaline conditions, only environmentally relevant PMG species, i.e. PMG− and PMG2− at pH 5−9, were investigated regarding their surface complexation. Though PMG is a tridentate chelating ligand, mounting evidence suggests that the carboxylate and amine groups do not directly interact with goethite.6,13,14 As shown in Figure 2 and SI Figure S6, adsorption did not shift the peak of υas(COO−) at 1400 cm−1, indicating either no or weak outer-sphere carboxylate surface interactions.6 In addition, the XPS analysis by Sheals et al. showed no evidence of direct interaction between the amine group and goethite.6 Conversely, the IR vibrations of the phosphonate group (1200−900 cm−1) exhibited dramatic changes upon adsorption at different pH values (Figure 2), indicating the formation of inner-sphere surface complexes on goethite via the phosphonate group.

RESULTS AND DISCUSSION PMG Deprotonation Process. Because of its central importance, the PMG deprotonation process was determined first by analyzing the IR spectra for various dissolved PMG species (Figure 1a). According to the dissociation constants (pKa1 = 2.3, pKa2 = 6.0, pKa3 = 11.0),7 the predominant species of PMG at pH 1.5, 4.2, 9.0, and 12.5 are neutral (PMG), monoanion (PMG − ), dianion (PMG 2− ), and trianion (PMG3−), respectively (SI Figure S1). It worth note that the phosphonate group can protonate to form a positively charged species at pH < 0.8 (SI Figure S2),7 but this pH condition is extremely rare in the environment. Therefore, neutral and anionic PMG species were the focus of our study. The IR spectra of sarcosine, the structural analog to PMG without a phosphonate group, were also collected as a control (Figure 1b). Comparison of the IR spectra between dissolved PMG and sarcosine reveals that the characteristic band region of the PMG phosphonate group is 1200−800 cm−1, while that of the amine and carboxylate groups is 1800−1200 cm−1 (Figure 1). For the neutral PMG species (pH 1.5, Figure 1a), the CO stretching (υ(CO)) in the carboxylate group was observed at 1740 cm−1,18 and the coupled C−(OH) stretching and C−O− H bending motions (υ(C−OH)/γ(C−O−H)) were at 1244 cm−1.24 These characteristic bands of carboxylate groups were also present in the spectrum of sarcosine, with a slight shift (pH 1.5, Figure 1b). The bands at 1188, 1080, and 915 cm−1, which were not observed in the spectra of sarcosine, were attributed to υas(PO2) (A″), υs(PO2) (A′), υ(P−OH) (A′), respectively.14,18,24 The υ(P−C) (A′) band should be lower than 800 cm−1 because the strength of a P−C bond is weaker than that of a P−O bond. Indeed, this υ(P−C) band was reported to be around 750 cm−1.25 Because this υ(P−C) peak is rather weak, the P−O vibrations were the focus of this study. The number of peaks and peak assignments for the phosphonate group conform to the C2v symmetry and group theory.26 PMG− is formed by removing a proton from the neutral PMG species, and this does not occur on the phosphonate group. As shown in Figure 1a, the above three phosphonate vibrations at pH 4.2 were almost the same as those at pH 1.5, indicating that the symmetry of the phosphonate group remained unchanged when pH increased from 1.5 to 4.2. By contrast, significant changes in the carboxylate region were observed, including the disappearance of υ(CO) of COOH at 1740 cm−1 and υ(C−OH)/γ(C−O−H) at 1244 cm−1, as well as the enhancement of the bands of υas(COO−) at 1610 cm−1 and υs(COO−) at 1400 cm−1. This observation indicated that the first deprotonation of PMG occurs at the carboxylic acid (pKa1 = 2.3, SI Figure S1). The second deprotonation (pKa2 = 6.0) has been the focus of controversy as to whether the phosphonate or amine group releases the proton,6−10 and our FTIR results implied that it occurred on the phosphonate group. The observed change in peak number and position of P−O(H) agreed well with the symmetry analysis for the shifting of PMG− (C−P(OH)O2−, Cs) at pH 4.2 to PMG2− (C−PO32−, C3υ) at pH 9.0 (Figure 1a). According to the correlation SI Table S2,26 the A′ (1080 cm−1) and A″ (1188 cm−1) modes of the P−O stretching vibration under Cs symmetry merged as a doubly degenerate E (1094 cm−1) mode under C3υ symmetry. Therefore, three IR active peaks were observed, namely, two P−O vibrations (A1 1948


Article


two facts: (1) more peaks appear in the calculated spectra than in the experimental spectra due to the asymmetry of cluster models and (2) the DFT-calculated spectrum only corresponds to one specific surface configuration, whereas more than one PMG surface configurations may coexist at certain pH. To solve these problems, we first categorized the observed IR peaks into different groups by 2D-COS analysis due to its merits in distinguishing one band from another by its origins upon external perturbations.20 The 2D-COS analysis results in Figure 4 contoured the synchronous and asynchronous

Figure 2. Deconvoluted ATR-FTIR spectra of adsorbed PMG on goethite surfaces at pH 5, 6, 7, 8, and 9 at equilibrium.

For an in-depth understanding of the PMG-goethite interaction mechanism, DFT calculations were performed to provide the theoretical IR bands of various PMG surface complexes. The possible PMG surface configurations included mononuclear monodentate (MM), mononuclear bidentate (MB), and binuclear bidentate (BB) structures. In addition, each structure considered two protonation states of phosphonate moiety: protonated (Type I) and deprotonated (Type II) species, leading to an overall of six PMG complexes (Figure 3). The comparison of DFT-calculated and experimental IR spectra showed that none of the single configuration match the experimental IR data. The disagreement can be attributed to

Figure 4. Synchronous (a, c, e) and asynchronous (b, d, f) correlation contour maps of dynamic IR spectra for the adsorption of PMG on goethite surface at pH 5, 7, and 9. The blue (red) regions were defined as negative (positive) correlation intensities.

correlation maps for PMG-goethite complexes, which were obtained from the dynamic IR spectra of PMG on goethite at pH 5, 7, and 9 (SI Figure S11). According to the detailed discussion in the SI, the IR peaks can be classified into the following five groups at pH 5−9: (a) 1142, 1018, 987 cm−1 and (b) 1122, 1052, 956 cm−1 at pH 5; (c) 1025, 987, 938 cm−1 and (d) 1128, 1094, 1060 cm−1 at pH 7; whereas only (e) 1102, 1004, and 980 cm−1 at pH 9. Compared to the DFT-calculated spectra, group (a) and (c) belong to BB II; group (b), (d), and (e) belong to MB II, MM I, and MM II, respectively (Figure 3, SI Figure S12). This 2D-COS result highlights the diversity of PMG configurations on goethite surfaces. To confirm the existence of multiple PMG surface complexes, ATR-FTIR spectra in D2O were collected and compared with those in H2O. As shown in Figure 5, most IR peaks at pD 5 and 9 in D2O had the same positions to those in H2O. This observation implied that the phosphonate group of the PMG surface complexes at pH 5 and 9 were without proton.28 At pH/pD 7, however, the group d as classified by the above 2D-COS analysis, namely, 1128, 1094, and 1060 cm−1, had a significant shift to lower frequencies, whereas the peaks in the group c (1025, 987, and 938 cm−1) remained approximately the same positions upon the deuterium exchange. This result strongly suggested that two PMG surface complexes coexisted at pH 7. Integrating IR spectra, 2D-COS analysis, and DFT results concluded that BB II and MB II complexes were predominant under a slightly acidic condition (pH 5), MM I and BB II under neutral condition (pH 6−8), whereas only MM II under alkaline condition (pH 9). The comparison of the calculated and experimental IR vibrational frequencies and the

Figure 3. Phosphonate group involved PMG complexes with goethite surfaces. The possible interfacial configurations, including mononuclear monodentate (MM), mononuclear bidentate (MB), and binuclear bidentate (MB), were optimized. Each configuration has two types, namely, the phosphonate group with (I) and without (II) proton. The explicit H2O molecules were not shown for clarity. Fe: cyan; C: gray; O: red; N: blue; P: orange; H: white. 1949


Article


Figure 5. IR spectra of adsorbed PMG in H2O and D2O for pH/pD 5 (left), pH/pD 7 (middle), and pH/pD 9 (right).

Table 1. Calculated IR Vibrational Frequencies (cm−1) and Assignment of Six Possible Phosphonate Group involved PMG Complexes with Goethite Surfacesa solvent = H2O assignment δ (PO−H) of phosphonate group υ (PO) of phosphonate group υas (PO2) of uncomplexed P−O υs (PO2) of uncomplexed P−O υas (P−(OFe)2) of bidentate complexes υs (P−(OFe)2) of bidentate complexes υ (P−OFe) of monodentate complexes

MM Ib

MM IIb

1208 1129(1128)c

MB II

1181

BB I

BB II

1194 1126 (1122)

1148(1142)

1108(1102) 1037 1099 1002 1092(1094), 1054(1060)

1048 (1052) 957 (956)

1105 1048

1014(1018), 984(987), 935(936)

BB I

BB II

1008(1004), 972(980) solvent = D2O

MM I δ (PO−H) of phosphonate group υ (PO) of phosphonate group υas (PO2) of uncomplexed P−O υs (PO2) of uncomplexed P−O υas (P−(OFe)2) of bidentate complexes υs (P−(OFe)2) of bidentate complexes υ (P−OFe) of monodentate complexes

MB I

MM II

1186 1118(1105)

MB I

MB II

1160

1178 1130 (1118)

1146(1137)

1118(1102) 1030 1074 997 1078(1073), 1044(1041)

1044 (1052) 957 (956)

1096 1040

1008(1014), 981(984), 938(931)

1011(1005), 981(980)

a

MM: mononuclear monodentate; MB: mononuclear bidentate; BB: binuclear bidentate. All proposed interfacial configurations are shown in Figure 3. b“I” and “II” represent protonated and deprotonated type, respectively. cValues in the parentheses are experimental IR vibrational frequencies.

of PMG and phosphate should be of great environmental significance considering the coapplication of phosphate fertilizers and the herbicide PMG.19 Both PMG and phosphate adsorption conformed to the Langmuir isotherm, and the adsorption capacity of PMG (42.3 mg/g or 0.25 mM/g, Figure 6a) was much lower than that of phosphate (159.3 mg/g or 1.021 mM/g, Figure 6b). In the competitive system, phosphate dramatically reduced PMG adsorption by up to about 50% when the molar ratio of phosphate to PMG increased to 5:1 (Figure 6a). In contrast, PMG had negligible effect on phosphate adsorption, even when the molar ratio was as high as 10:1 (Figure 6b). The difference in the macroscopic competitive adsorption behaviors of PMG and phosphate can be explained by the FTIR results (Figure 6c). The spectrum of the PMG and phosphate coexisting system (molar ratio = 1:1, bottom of Figure 6c) was almost identical to that of phosphate adsorption alone (middle of Figure 6c), both of which were in stark contrast to that of PMG adsorption alone (top of Figure 6c). This observation indicates that phosphate has a stronger affinity for goethite

assignment of possible PMG complexes was summarized in Table 1. The pH dependence of PMG surface complexes can be attributed to the ligand-exchange mechanism. Under acidic conditions, the ligand, phosphonate group on PMG, is concentrated to the goethite surface (pHpzc 8.9) by the electrostatic attraction, which increases the energetic favorability to replace the surface hydroxyl group to form bidentate structures, including BB and MB configurations. As the pH increases to neutral, the competition of OH− hinders the ligand exchange reaction of PMG with surfaces, resulting in a successive disappearance of the MB configuration. MB is a four-membered ring structure which has a relatively unfavorable energy compared with the BB configuration.14 Under alkaline conditions, the MM complex without proton becomes the predominant configuration due to the strong competition with OH−. Competition of Phosphate with PMG for Surface Complexation. Because PMG forms surface complexes through the phosphonate group, the competitive adsorption 1950


Article


Figure 6. Adsorption isotherms of PMG (a) and phosphate (b) in the competitive systems at pH 7. Symbols represent the experimental results, and the solid lines are Langmuir adsorption model simulations. (c) Deconvoluted ATR-FTIR interfacial spectra of PMG (top), phosphate (middle), and the competition of phosphate with PMG (bottom) in 0.1 M NaCl solution at pH 7. The red values in the parentheses are relative peak areas, which are normalized to the most intense peak located at 1050 cm−1.

Notably, phosphate surface complexes may also consist of multiple species.17,29 Because phosphate dramatically reduced PMG adsorption whereas PMG had a negligible effect on phosphate adsorption (Figure 6a, b), it is unlikely that PMG has a great impact on phosphate surface structure. Therefore, the configurations of phosphate complexes were relatively stable in the presence of PMG and the uneven increase in peak areas should be mainly attributed to the change in the relative proportion of PMG surface species. The impact of surface structures on the mobility of adsorbed PMG and phosphate motivated our further study using extraction experiments. The release of PMG and phosphate followed the pseudo-second-order kinetics model (SI Figure S14, Table S8−S9). The leaching results show that 19.1−29.5% of PMG was released in the ternary PMG-phosphate-goethite system, appreciably higher than that in the binary PMGgoethite system (9.6−18.6%, SI Table S8). This result clearly demonstrates that phosphate competition altered the PMG surface complex to a less stable configuration. Conversely, PMG resulted in an insignificant impact on phosphate desorption (3.8−4.0% in ternary system vs 6.1−6.3% in binary system, SI Table S9), in agreement with their adsorption behaviors (Figure 6). Environmental Significance. PMG has become an agricultural panacea due to its high efficiency and broadspectrum properties, while its environmental behaviors is still unclear. Our results here provide new insights into the molecular-level surface complexation of glyphosate on goethite. The phosphonate group plays a key role in the formation of inner-sphere complexes on goethite surfaces, whereas the carboxyl group may only take part in outer-sphere surface interactions via electrostatic attraction. Notably, multiple PMG surface configurations may coexist and transform depending on

surfaces. The unchanged profile and peak positions at 1590, 1400, and 1324 cm−1 in the spectrum of PMG and phosphate coadsorption (SI Figure S13) indicate that the carboxyl group of PMG may not directly interact with goethite. Further quantitative analysis on the area of deconvoluted interfacial peaks can provide insights into the structural information. Compared to the phosphate-only system (middle of Figure 6c), the relative peak areas of several peaks increased in the competitive system, especially the peak at 1124 cm−1 (28% increase relative to 1050 cm−1, bottom of Figure 6c), the position of which is close to the strongest peak (1128 cm−1) in the PMG-only system (top of Figure 6c). This result indicated that a small amount of PMG complexes, though less than phosphate, contributed to the inner-sphere complexation in the competitive system. Interestingly, phosphate preferentially occupied the goethite sites that originally were prone to form the BB II configuration in the PMG-only system. For example, for the peak at 980 and 942 cm−1 in the competitive system, corresponding to 987 and 938 cm−1 of BB II in the PMG-only system (top of Figure 6c), the relative peak areas only increased 8% and 1%, respectively (bottom of Figure 6c). By comparison, for the peaks at 1124 and 1088 cm−1 in the competitive system, which are close to the peaks at 1128 and 1094 cm−1 of MM I in the PMG-only system, the relative peak areas appreciably increased by 28% and 21%, respectively. The uneven increase in peak areas showed that phosphate not only displayed a stronger affinity than PMG for goethite surfaces, but also preferentially occupied the limited surface sites which originally formed the stable PMG surface complex, namely, BB II. As a result of phosphate competition, a great proportion of PMG molecules were compelled to form relatively unfavorable MM I. 1951


Article


(7) Li, W.; Wang, Y.-J.; Zhu, M.; Fan, T.-T.; Zhou, D.-M.; Phillips, B. L.; Sparks, D. L. Inhibition mechanisms of Zn precipitation on aluminum oxide by glyphosate: A 31P NMR and Zn EXAFS study. Environ. Sci. Technol. 2013, 47, 4211−4219. (8) Harris, W. R.; Sammons, R. D.; Grabiak, R. C.; Mehrsheikh, A.; Bleeke, M. S. Computer simulation of the interactions of glyphosate with metal ions in phloem. J. Agric. Food Chem. 2012, 60, 6077−6087. (9) Peixoto, M. M.; Bauerfeldt, G. F.; Herbst, M. H.; Pereira, M. S.; da Silva, C. O. Study of the stepwise deprotonation reactions of glyphosate and the corresponding pKa values in aqueous solution. J. Phys. Chem. A 2015, 119, 5241−5249. (10) Liu, B.; Dong, L.; Yu, Q.; Li, X.; Wu, F.; Tan, Z.; Luo, S. Thermodynamic study on the protonation reactions of glyphosate in aqueous solution: potentiometry, calorimetry and NMR spectroscopy. J. Phys. Chem. B 2016, 120, 2132−2137. (11) Gimsing, A. L.; Borggaard, O. K.; Sestoft, P. Modeling the kinetics of the competitive adsorption and desorption of glyphosate and phosphate on goethite and gibbsite and in soils. Environ. Sci. Technol. 2004, 38, 1718−1722. (12) Mazzei, P.; Piccolo, A. Quantitative evaluation of noncovalent interactions between glyphosate and dissolved humic substances by NMR spectroscopy. Environ. Sci. Technol. 2012, 46, 5939−5946. (13) Tribe, L.; Kwon, K. D.; Trout, C. C.; Kubicki, J. D. Molecular orbital theory study on surface complex structures of glyphosate on goethite: Calculation of vibrational frequencies. Environ. Sci. Technol. 2006, 40, 3836−3841. (14) Barja, B. C.; dos Santos Afonso, M. Aminomethylphosphonic acid and glyphosate adsorption onto goethite: A comparative study. Environ. Sci. Technol. 2005, 39, 585−592. (15) Villalobos, M.; Perez-Gallegos, A. Goethite surface reactivity: A macroscopic investigation unifying proton, chromate, carbonate, and lead(II) adsorption. J. Colloid Interface Sci. 2008, 326, 307−323. (16) Villalobos, M.; Cheney, M. A.; Alcaraz-Cienfuegos, J. Goethite surface reactivity: II. A microscopic site-density model that describes its surface area-normalized variability. J. Colloid Interface Sci. 2009, 336, 412−422. (17) Kubicki, J. D.; Paul, K. W.; Kabalan, L.; Zhu, Q.; Mrozik, M. K.; Aryanpour, M.; Pierre-Louis, A. M.; Strongin, D. R. ATR-FTIR and density functinal theory study of the structures, energetics, and vibrational spectra of phosphate adsorbed onto goethite. Langmuir 2012, 28, 14573−14587. (18) Barja, B. C.; dos Santos Afonso, M. An ATR−FTIR study of glyphosate and its Fe(III) complex in aqueous solution. Environ. Sci. Technol. 1998, 32, 3331−3335. (19) Sasal, M. C.; Demonte, L.; Cislaghi, A.; Gabioud, E. A.; Oszust, J. D.; Wilson, M. G.; Michlig, N.; Beldoménico, H. R.; Repetti, M. R. Glyphosate loss by runoff and its relationship with phosphorus fertilization. J. Agric. Food Chem. 2015, 63, 4444−4448. (20) Yang, Y.; Yan, W.; Jing, C. Dynamic adsorption of catechol at the goethite/aqueous solution interface: A molecular-scale study. Langmuir 2012, 28, 14588−14597. (21) Yan, W.; Wang, H.; Jing, C. Adhesion of Shewanella oneidensis MR-1 to goethite: A two-dimensional correlation spectroscopic study. Environ. Sci. Technol. 2016, 50, 4343−4349. (22) Yan, W.; Zhang, J.; Jing, C. Adsorption of enrofloxacin on montmorillonite: Two-dimensional correlation ATR/FTIR spectroscopy study. J. Colloid Interface Sci. 2013, 390, 196−203. (23) Frisch, M. J.; T, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;

pH. Considering that the BB structure is more stable than MB and MM, we should pay particular attention to the pH factor in evaluating the behavior, bioavailability, and fate of PMG in the environment. Our work also demonstrates that the presence of phosphate inhibits PMG adsorption by competing for surface sites. Moreover, phosphate competition impairs the affinity of PMG toward goethite surfaces by restricting PMG to forming an energetically unfavorable MM rather than a more stable BB structure, resulting in higher mobility for adsorbed PMG. Given the widespread phosphorus fertilization and its strong affinity to soil components, phosphate, the structural analog to the active moiety of PMG, should be carefully considered in future studies on PMG risk assessment.

■

ASSOCIATED CONTENT

S Supporting Information *

Details of in situ flow cell ATR-FTIR measurements; Peak fitting procedure; 2D-COS analysis; PMG and phosphate Determination; analytical procedure; dynamic spectra of PMG on goethite at pH 5, 7, and 9; and additional figures. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.7b05643. (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone: +86 10 6284 9523; fax: +86 10 6284 9523; e-mail: [email protected]. ORCID

Chuanyong Jing: 0000-0002-4475-7027 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge the financial support of the National Key Basic Research Program of China (2015CB932003, 2014CB441102), the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB14020302), the National Natural Science Foundation of China (21477144, 41425016, and 21321004).

■

REFERENCES

(1) Sammons, R. D.; Gaines, T. A. Glyphosate resistance: State of knowledge. Pest Manage. Sci. 2014, 70, 1367−1377. (2) Doublet, J.; Mamy, L.; Barriuso, E. Delayed degradation in soil of foliar herbicides glyphosate and sulcotrione previously absorbed by plants: Consequences on herbicide fate and risk assessment. Chemosphere 2009, 77, 582−589. (3) Zablotowicz, R. M.; Accinelli, C.; Krutz, L. J.; Reddy, K. N. Soil depth and tillage effects on glyphosate degradation. J. Agric. Food Chem. 2009, 57, 4867−4871. (4) Dousset, S.; Jacobson, A. R.; Dessogne, J.-B.; Guichard, N.; Baveye, P. C.; Andreux, F. Facilitated transport of diuron and glyphosate in high copper vineyard soils. Environ. Sci. Technol. 2007, 41, 8056−8061. (5) Jonsson, C. M.; Persson, P.; Sjöberg, S.; Loring, J. S. Adsorption of glyphosate on goethite (α-FeOOH): Surface complexation modeling combining spectroscopic and adsorption data. Environ. Sci. Technol. 2008, 42, 2464−2469. (6) Sheals, J.; Sjoberg, S.; Persson, P. Adsorption of glyphosate on goethite: Molecular characterization of surface complexes. Environ. Sci. Technol. 2002, 36, 3090−3095. 1952


Article

Environmental Science & Technology Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision A.01; Gaussian, Inc.: Wallingford, CT. 2009. (24) Sheals, J.; Persson, P.; Hedman, B. IR and EXAFS spectroscopic studies of glyphosate protonation and copper(II) complexes of glyphosate in aqueous solution. Inorg. Chem. 2001, 40, 4302−4309. (25) Barja, B. C.; Tejedor-Tejedor, M. I.; Anderson, M. A. Complexation of methylphosphonic acid with the surface of goethite particles in aqueous solution. Langmuir 1999, 15, 2316−2321. (26) Nakamoto, K. Infraed and raman spectra of inorganic and coordiantion compounds. Part A. In Theory and Applications in Inorganic Chemistry, 5th ed.; John Wiley & Sons, Inc., 1997; pp 189− 209. (27) Chabot, M.; Hoang, T.; Al-Abadleh, H. A. ATR-FTIR studies on the nature of surface complexes and desorption efficiency of p-arsanilic acid on iron (oxyhydr)oxides. Environ. Sci. Technol. 2009, 43, 3142− 3147. (28) Johnston, Chad P.; Chrysochoou, M. Mechanisms of chromate adsorption on hematite. Geochim. Cosmochim. Acta 2014, 138, 146− 157. (29) Elzinga, E. J.; Kretzschmar, R. In situ ATR-FTIR spectroscopic analysis of the co-adsorption of orthophosphate and Cd(II) onto hematite. Geochim. Cosmochim. Acta 2013, 117, 53−64.

1953


Molecular Insights into Glyphosate Adsorption to ... - ACS Publications

Recommend Documents