Next Article in Journal
Overcoming Cancer Drug Resistance with Nanoparticle Strategies for Key Protein Inhibition
Previous Article in Journal
Differences in the Renal Accumulation of Radiogallium-Labeled (Glu)14 Peptides Containing Different Optical Isomers of Glutamic Acid
Previous Article in Special Issue
Numerical Modeling of Anisotropic Particle Diffusion through a Cylindrical Channel
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Molecules
Volume 29
Issue 17
10.3390/molecules29173991
molecules-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

How Doping Regulates As(III) Adsorption at TiO2 Surfaces: A DFT + U Study

by
Xiaoxiao Huang
,
Mengru Wu
,
Rongying Huang
and
Gang Yang
*
College of Resources and Environment, Southwest University, Chongqing 400715, China
*
Author to whom correspondence should be addressed.
Molecules 2024, 29(17), 3991; https://doi.org/10.3390/molecules29173991
Submission received: 7 July 2024 / Revised: 7 August 2024 / Accepted: 16 August 2024 / Published: 23 August 2024
(This article belongs to the Special Issue Feature Papers in Computational and Theoretical Chemistry)
Browse Figures
Versions Notes

Abstract

:
The efficient adsorption and removal of As(III), which is highly toxic, remains difficult. TiO2 shows promise in this field, though the process needs improvement. Herein, how doping regulates As(OH)3 adsorption over TiO2 surfaces is comprehensively investigated by means of the DFT + D3 approach. Doping creates the bidentate mononuclear (Ce doping at the Ti5c site), tridentate (N, S doping at the O2c site), and other new adsorption structures. The extent of structural perturbation correlates with the atomic radius when doping the Ti site (Ce >> Fe, Mn, V >> B), while it correlates with the likelihood of forming more bonds when doping the O site (N > S > F). Doping the O2c, O3c rather than the Ti5c site is more effective in enhancing As(OH)3 adsorption and also causes more structural perturbation and diversity. Similar to the scenario of pristine surfaces, the bidentate binuclear complexes with two Ti-OAs bonds are often the most preferred, except for B doping at the Ti5c site, S doping at the O2c site, and B doping at the O3c site of rutile (110) and Ce, B doping at the Ti5c site, N, S doping at the O2c site, and N, S, B doping at the O3c site of anatase (101). Doping significantly regulates the As(OH)3 adsorption efficacy, and the adsorption energies reach −4.17, −4.13, and −4.67 eV for Mn doping at the Ti5c site and N doping at the O2c and O3c sites of rutile (110) and −1.99, −2.29, and −2.24 eV for Ce doping at the Ti5c site and N doping at the O2c and O3c sites of anatase (101), respectively. As(OH)3 adsorption and removal are crystal-dependent and become apparently more efficient for rutile vs. anatase, whether doped at the Ti5c, O2c, or O3c site. The auto-oxidation of As(III) occurs when the As centers interact directly with the TiO2 surface, and this occurs more frequently for rutile rather than anatase. The multidentate adsorption of As(OH)3 causes electron back-donation and As(V) re-reduction to As(IV). The regulatory effects of doping during As(III) adsorption and the critical roles played by crystal control are further unraveled at the molecular level. Significant insights are provided for As(III) pollution management via the adsorption and rational design of efficient scavengers.

1. Introduction

Arsenic (As) is a ubiquitously distributed toxic metalloid. Its pollution has seriously endangered the ecological environment and human health and aroused global concern [1,2]. In order to remediate As-polluted sites, a number of techniques have been developed, such as adsorption, precipitation, flocculation, membrane separation, and reverse osmosis [3,4,5]. Among them, adsorption by minerals pronouncedly regulates arsenic migration, bioavailability, and fate and ranks to be the most widely used technique for pollution management [6,7,8]. Inorganic arsenic in natural environments predominates in the As(III) and As(V) forms. Although much more toxic, As(III) is highly mobile and more difficult for adsorption and removal because it exists predominantly as the uncharged As(OH)3 moiety [9,10,11]. It thus represents an imperative task to explore the usage efficient As(III) scavengers.
Metal (hydr)oxides [12,13], zero-valent iron [14,15], zeolites [7,16], clay minerals [17,18], and activated carbon [19,20] have been used for As(III) adsorption. However, they generally suffer from limited adsorption capacity and strength, and hence chemical modification is necessitated [21,22,23]. Mn doping of γ-Al2O3 enables the transfer of more electrons into a stable status, which further promotes As(III) adsorption [24]. As indicated by periodic density functional theory (DFT) calculations, doping of gibbsite results in strong M-3d and OAs-2p orbital interactions (M = Fe(III), Mn(III), Mn(IV)) that facilitate As(III) adsorption [25]. Owing to its superior stability, non-toxicity, and high activity, titanium dioxide (TiO2) has been applied extensively for environment-associated adsorption and catalysis [26,27,28,29,30]. Pena et al. [27] found that 0.5 mmol/g of As(III) is adsorbed by nanocrystalline TiO2, and the adsorption thermodynamics and kinetics can be well described by the Freundlich isotherm and pseudo-second-order model. Wu et al. [28] showed that due to adsorption by TiO2 nanoparticles, the accumulation of As in plants reduces by 40~90%. The controllable configuration of TiO2 nanoparticles was realized using 3D-printing technology, and after being used more than 10 times, the TiO2 nanoparticles remained effective for the treatment of raw-arsenic-polluted groundwater samples [29]. X-ray absorption spectra (XAS) revealed that As(III) at the TiO2 surface tends to adopt the bidentate binuclear configuration, and the Ti-As distance is approximately 3.32 Å [30], which is in good agreement with DFT calculated results [31,32,33]. A density of states (DOS) analysis [34] demonstrated that As(III) at the TiO2 surface generates anti-bonding orbitals above the Fermi level, and the Ti-OAs bonds are attributed mainly to the electron sharing between the O-2p and Ti-3d orbitals. Doping is widely documented to be capable of regulating the adsorption and catalytic performances of TiO2, e.g., Fe [35,36], Ce [37], B [38], V [39], and Mn [39] at the Ti site and N [40,41], F [42], and S [43] at the O site. Owing to the high efficiency of H2O2 utilization (99.1%), almost all As(III) adsorbed at the Ce-doped TiO2 surface is catalytically oxidized to As(V), and the activity of Ce-doped TiO2 remains after five cycles [37]. The incorporation of Fe into the TiO2 lattice improves the adsorption capacity of As by arresting the grain growth, which results in a higher affinity [44]. N doping into the TiO2 lattice effectively promotes the decolorization of double azo reactive black 5 (RB5) dye and shows significant bactericidal activity against Escherichia coli, with an inhibition rate of up to 92.47% [45]. F doping of TiO2 significantly enhances the adsorption capacity of Pb(II) [42], while S-doped TiO2 causes a redshift of the optical absorption edge, and hepatotoxin microcystin-LR is efficiently degraded under visible-light irradiation [43]. Despite their wide application, how these dopants within TiO2 affect As(III) adsorption and the associated structure–activity relationship remain enigmatic. DFT approaches have been testified to be well suited for structural engineering, mechanistic unraveling, and catalyst design due to their atomic-scale spatial resolution and absolute accuracy (<4.0 kJ/mol) [46,47]. In this study, DFT calculations were carried out, considering (1) two dominant polymorphs of TiO2 (rutile and anatase) to probe crystal dependence during As(III) adsorption; (2) all types of doping (at the Ti5c, O2c, and O3c sites, see Figure 1), to explore the most effective type of doping and site specificity for As(III) adsorption; and (3) a number of dopants, including those aforementioned, to establish the structure–activity relationship during As(III) adsorption. Then, the regulatory mechanisms of doping and crystal control during As(III) adsorption were unraveled at the molecular level. The results provide valuable insight into As(III) adsorption and removal by TiO2-based materials and feed back for the design of efficient scavengers for As(III) and other pollutants.

2. Results and Discussion

2.1. As(OH)3 Adsorption by Rutile (110) with the Ti Site Being Doped

Adsorption configurations of As(OH)3 at the rutile (110) surface with the Ti5c site being doped (DTi = Fe, Mn, V, Ce, B) are depicted in Figure 2 and Figure S2. They exhibit a marked difference from those of a pristine surface (Figure S3), wherein R1Pr, R2Pr, and R3Pr correspond to the physisorbed, monodentate, and bidentate binuclear complexes [31].
For Fe5c doping, R2Fe5c with the Ti1-O2 bond remains structurally similar as it appears at a pristine surface (R2Pr), while the other adsorption structures may be distinct: R1Fe5c with the Fe-As interaction vanishes, R4Fe5c with the Fe-O2 bond becomes monodentate, and R3Fe5c (Ti2-O1 + Fe-O2) and R5Fe5c (Fe-O2 + As-O4) are bidentate binuclear. R2Fe5c and R4Fe5c are further stabilized by the Fe-As (3.160 Å) and Ti2-As (3.047 Å) interactions, and R5Fe5c is featured by the As-OS bond that is absent at a pristine surface (OS refers to the surface O atom). Mn5c and V5c doping has As(OH)3 adsorption that resembles Fe5c doping, while Ce5c and B5c doping differs substantially, which originates from the divergence of their atomic radii: Ce (1.82 Å) >> Ti and V, Mn, Fe (1.17~1.32 Å) >> B (0.82 Å). Owing to the large atomic radius, Ce protrudes outside the plane of its bonded OS atoms (Figure S4) [48], and the interaction with As(OH)3 is promoted so that R6Ce5c of the bidentate mononuclear motif is generated. However, R6Ce5c is disfavored sterically by the hepta-coordination for Ce, and this can be reflected by the longer Ce-OAs bonds and a less negative Ead than in R3Ce5c of the bidentate binuclear motif (2.616, 2.571 vs. 2.428 Å, and −2.12 vs. −3.43 eV). B has the smallest atomic radius and forms only three B-OS bonds [49], which affects the coordination environment and reactivity of the adjacent surface atoms (Figure S5). B is not directly involved during adsorption, and hence R3B5c and R4B5c vanish while the B-As interaction lacks in R2B5c. R7B5c arises due to O4 transformation from O3c to O2c, and albeit with similar bonding interactions as R5B5c, it is less preferred due to a four-membered ring (Ti1O2AsO4): Ead = −3.20 vs. −5.15 eV.
The stability of different As(OH)3 adsorption configurations generally declines as bidentate (R3, R5, R6, R7) > monodentate (R2, R4) > physisorbed (R1). R3 of the bidentate binuclear motif is the most favorable for all dopants at a pristine surface (Figure 2 and Figure S2), which is consistent with EXAFS and ATR-FTIR observations [30,50,51]. The Ead amounts to −2.43, −2.73, −3.43, −3.79, and −4.17 eV for R3V5c, R3Fe5c, R3Ce5c, R3Pr, and R3Mn5c, respectively [50]. This agrees with the exergonic nature of As(III) adsorption onto Fe-doped TiO2(B) [52]. As(OH)3 adsorption in R3Mn5c is further enhanced by the auto-transfer of H1 and H2 to the OS atoms [31]. Hence, the adsorption efficacy of As(OH)3 is regulated pronouncedly by doping the Ti5c site and varies within a wide range. B5c doping is an exception, and R5B5c formation is strongly benefited: the reactivity of the Ti1 site is enhanced due to the elongation of Ti1-OS bonds (Figure S5) and the superior stability of the AsO4 tetrahedron. The O2AsO5 angle in R5B5c is closer to 109.5° for the regular tetrahedron than in the other adsorption configurations: 103.7° vs. 93.8°, 94.4°, 93.7°, 79.7° for R5B5c vs. R5Fe5c, R5Mn5c, R5V5c, and R5Ce5c. The deviation degree of the O2AsO5 angle also interprets the lowest relative stability of R5 vs. R3 for Ce5c doping (ΔEad = 1.49 eV).

2.2. As(OH)3 Adsorption by Rutile (110) with Doping the O Site

As(OH)3 adsorption over a rutile (110) surface with the O2c, O3c site being doped (DO = N, F, S, B), as shown in Figure 3 and Figure S6, is more perturbed than with the Ti5c site being doped. N doping creates the additional adsorption sites [40,41], and As(OH)3 bonding to N2c and N3c generates three and two new adsorption structures, respectively. R1, R2, and R3 remain similar to how they appear at a pristine surface, albeit with some differences: the Ti-As interaction is promoted in R1N2c (2.673 vs. 2.917 Å in R1Pr) and R2N3c (2.795 vs. 3.078 Å in R2Pr), while dual proton transfers occur in R3N2c and R3N3c instead of the single-proton transfer in R3Pr, and H3 in R3N2c bonds directly to N2c. The As-N bond is created in R4N2c and R4N3c, and a second bonding (Ti-O2) leads to bidentate R6N2c and R6N3c, while R6N3c is less favorable due to the formation of a four-membered ring (Ti1NAsO2) (see the Ead in Figure 3). It is surprising to find that R5N2c with only an As-O4 bond exists stably (Figure S6).
F doping alters As(OH)3 adsorption less than N doping, and a majority of adsorption structures (R1F2c, R2F2c, R3F2c, R2F3c, and R3F3c) remain similar to how they appear at a pristine surface. The As-OS bond is created in R5F2c and R7F3c, and the additional Ti1-O2 bond greatly stabilizes R7F3c (Ead = −2.38 vs. −0.25 eV, Figure 3 and Figure S6). Owing to the close chemo-properties of S and O, R1S2c, R2S2c, and R2S3c are structurally similar and have a comparable Ead as those at a pristine surface (Figure 3 and Figures S3 and S6). In addition to R5S2c and R5S3c with the As-OS bond, R4S2c with the As-S bond is produced and exists stably. The large atomic radius causes S to protrude outwards and facilitates the interaction with As(OH)3, which further leads to a tridentate complex: R8S2c (Ti1-O2 + Ti2-O1 + As-S). B2c doping is not considered due to its markedly lower stability compared to B3c doping (5.17 eV), which is distinct from other dopants (e.g., 0.15 eV for N2c vs. N3c doping). B3c doping causes the same surface rearrangement as B5c doping (Figure S7). In addition to R2B3c and R3B3c, which are structurally similar to how they appear at a pristine surface, R9B3c and R10B3c appear and the As-O4 bond forms at expense of the As-O2 bond rupture. R10B3c is further stabilized by the Ti3-O1 bonding.
The adsorption configurations of As(OH)3 at the rutile (110) surface have the following stability trend: tridentate (R8) > bidentate (R3, R6, R7) > monodentate (R2, R4, R5) > physisorbed (R1) (see the Ead in Figure 3 and Figure S6). R3 is generally the most preferred: −3.12, −2.93, −3.18, −3.00, −4.13, and −4.67 eV for R3F2c, R3F3c, R8S2c (transformed barrierlessly from R3S2c), R3S3c, R3N2c, and R3N3c, respectively. This also suggests that doping the O2c, O3c site may be more effective in enhancing As(OH)3 adsorption than doping the Ti5c site (Figure 2). For B3c doping, the preference of R10B3c over R3B3c (−4.72 vs. −2.59 eV) is ascribed to the structural reconstruction. This is confirmed by the superior stability of R9B3c (−4.65 eV) that is monodentate while it undergoes a similar reconstruction. Rutile (110) with doping of the O site is also efficient for As(OH)3 adsorption and removal, and the efficacy can be regulated within a wide range through the choice of dopants. The preferred adsorption with the O3c vs. O2c site being doped ranks as N > S > F, consistent with the likelihood to form more bonds: As(OH)3 forms a direct bond with N3c in R4N3c and R6N3c, creating the tetra-coordinated N site, and S2c in R4S2c and R8S2c, creating the tri-coordinated S site, whereas it forms no bond with F2c and F3c. R3N3c and R3F2c are the most favorable for N and F doping, while due to the tridentate motif, R8S2c is slightly preferred over R3S3c. This is corroborated by the stability trend of adsorption structures with the As-O/As-DO bond (N > S > F): for As-O4 bonding, R5N2c (−2.82 eV) > R5S2c, R5S3c (−1.14~−1.75 eV) > R5F2c (−0.25 eV), and for As-DO bonding, R4N2c (−3.32 eV) > R4S2c (−0.85 eV) > R4F2c (non-existent).

2.3. As(OH)3 Adsorption by Doped Anatase (101)

Figure 4, Figure 5 and Figures S8 and S9 depict the adsorption structures of As(OH)3 over the anatase (101) surface with the Ti5c, O2c, O3c site being doped (DTi = Fe, Mn, V, Ce, B; DO = N, F, S, B). Doping causes less alteration for As(OH)3 adsorption over anatase (101) vs. the rutile (110) surface, which can be deduced partially from the number of the disruption-prone physisorbed complexes (A1 and R1): nine vs. three remain after doping. Particularly, F2c, F3c, and S3c doping has a total of three adsorption structures (A1, A2, and A3) that are exactly identical to those at a pristine surface.
All new adsorption configurations are featured by the direct bonding of As(OH)3 with the doped sites, except B, while those with only the As-OS bond vanish. This confirms less alteration at the anatase (101) than at the rutile (110) surface: (1) As(OH)3 adsorption is similar for Fe5c, Mn5c, and V5c doping. The bidentate A5 (DTi-O2 + As-O4) is destabilized by the formation of a four-membered ring (DTiO2AsO4), while in A4, the ring strain is alleviated where the As center forms the interaction with O4 and O5 instead of direct bonding. (2) Ce5c and B5c doping behaves distinctly due to a structural disruption (Figures S10 and S11), as discussed for rutile (110). However, A6Ce5c with two Ce-OS bonds becomes preferred over A3Ce5c (Figure 4), probably due to the inferior reactivity of the Ti5c sites at the anatase (101) surface [53,54]. Ce5c doping enhances the reactivity, and the formation of more Ce-OAs bonds promotes adsorption (|Ead|: A6Ce5c > A3Ce5c > A3Pr), consistent with shorter Ce-OAs bonds in A6Ce5c vs. A3Ce5c (2.573, 2.594 vs. 2.640 Å). Although all of these have the As-O2c bond, A5B5c, A7B5c, and A8B5c have distinct chemical environments for O2c: Ti-O2c-Ti, O2c transformed from O3c, and B-O2c-Ti, respectively. (3) N2c doping leads to monodentate A4N2c (As-N) and tridentate A5N2c (As-N + Ti1-O1 + Ti2-O3), while N3c doping produces bidentate A6N3c (As-N + Ti1-O2). (4) S2c doping results in tridentate A5S2c (Ti1-O1 + Ti2-O3 + As-S), in addition to bidentate A6S2c (Ti1-O2 + As-S) that is disfavored by the large Ti-S distances (ca. 2.665 vs. 2.349 Å in A1S2c) (Figure 5 and Figure S9). (5) B3c doping causes the structural rearrangement producing three Ti-B bonds (Figure S12). A7B3c has a B-OAs bond, as further stabilized by the H2 transfer to O5, while A4B3c has an As-B bond and A8B3c has an As-OS bond, and a second bonding (Ti-OAs) produces A6B3c and A9B3c.
The stability of different adsorption configurations generally declines as tridentate (A5), bidentate (A3, A6, A9) > monodentate (A2, A4, A8) > physisorbed (A1) (see Figure 4, Figure 5 and Figures S8 and S9). The most favorable adsorption configurations may not be A3, and have Ead values of −1.24 eV for A3Fe5c, −1.50 eV for A3Mn5c, −1.45 eV for A3V5c, −1.99 eV for A6Ce5c, −2.05 eV for A7B5c, −2.29 eV for A4N2c, −2.24 eV for A6N3c, −1.54 eV for A3F2c, −1.83 eV for A3F3c, −1.45 eV for A5S2c, −1.10 eV for A2S3c, and −4.30 eV for A9B3c. These values agree with the literature reports available: the Ce-Ti hybrid oxide shows enhanced adsorption for As(III) compared to pure TiO2 [55], and Pb2+ adsorption by anatase increases due to F doping [42]. The choice of different dopants causes the As(OH)3 adsorption efficiency to vary within a wide range, and doping the O2c, O3c rather than the Ti5c site may exhibit larger promoting effects for As(OH)3 adsorption, which is in line with the results of the rutile (110) surface. However, the doped anatase (101) surface is apparently less efficient for As(OH)3 adsorption and removal than the doped rutile (110) surface, implying the critical role of crystal control played therein. Albeit both being bidentate, A9B3c is preferred over A6B3c due to the B-O2H bond formation, as verified by the superior stability of A8B3c that has B-O2H bonding, although it is monodentate. In order to form the tridentate motif, A5N2c has the stretched As-O1 and As-O3 bonds (ca. 1.880 Å), which lead to the lower stability compared to A4N2c. A7B5c has As-O2c bonding similar to A5B5c and A8B5c, while it is preferred due to the high reactivity of its O2c site that evolves from O3c. Albeit being the bidentate motif, A3S3c is destabilized by a serious structural distortion: the strong electrostatic repulsion with O1 and O3 causes the S atom to be pushed below the top surface, and the Ti1-Ti2 distances are significantly enlarged (4.387 vs. 3.815 Å in A3Pr).

2.4. Regulatory Mechanism of As(OH)3 Adsorption by Doping

The oxidation states of As in various As(OH)3 adsorption structures, as determined by the Bader charge and magnetic moments (Figures S13–S16), are listed in Figure 2, Figure 3, Figure 4, Figure 5 and Figures S2, S6, S8 and S9. As(III), As(IV), and As(V) correspond to Bader charges of 1.51~1.70, 1.78~2.06, and 2.24~2.55 |e|, respectively [31,56,57]. As(IV) differs from As(III) and As(V) by having the unpaired electrons and noticeable magnetic moments [6,58,59,60]. The assignment of oxidation states is further validated by reference compounds, and the Bader charge of the As centers in As2O3 and As2O5 amounts to 1.53 and 2.25 |e|, respectively. The As centers in the adsorption configurations are often in +III form, while they are automatically oxidized to As(IV) or As(V) when forming direct bonds with the highly electronegative surface O and N atoms. These are distinct from pristine surfaces, where only the As(III) form is detected (Figure S3) [31]. The doped rutile (110) rather than anatase (101) surface is more ready to execute As(III) oxidation to As(V): nine (R5B5c, R7B5c, R4N2c, R5N2c, R6N2c, R5S2c, R4N3c, R7F3c, and R5S3c) vs. two (A4N2c and A6N3c). The +IV species is prone to occur when As(OH)3 is multidentately adsorbed: R5Fe5c, R5Mn5c, R5V5c, R6N3c, R5F2c, R4S2c, R8S2c, A5Fe5c, A5Mn5c, A5B5c, A7B5c, A8B5c, A5N2c, and A6S2c. Owing to the strong interaction among the As center and two OS atoms, A4Fe5c and A4Mn5c have the As(IV) species. Instead, A4V5c has the apparently larger As-OS distances (ca. 2.90 Å), and hence the As center remains to be in the +III state. As(IV) may even appear when the As centers form the interaction with the Ti5c site: R2Mn5c, A1Fe5c, A1N3c, and A1B5c. All adsorption structures with As(V) have superior stability, and As(IV), rarely found in natural environments, is stabilized pronouncedly at the doped TiO2 surfaces; R8S2c and A7B5c become the most preferred for S2c and B5c doping.
Doping alters the electronic structures of TiO2 and further As(OH)3 adsorption [61] (see the charge density difference and spin density isosurfaces for Fe5c, N2c, and N3c doping in Figures S17 and S18). Fe doping is used to illustrate the change in charged states during doping of the TiO2 surfaces (see Figure S19). Charge transfer occurs from Fe to the adjacent O atoms, and the Bader charge of Fe shows some increase. Note that Fe(OH)3 is used as a reference for Fe in the +III state. The alteration of charged states due to Fe doping is a localized behavior, and one or several adjacent Fe-bonded O atoms fall between the O2− and O•− states, while the rest of the atoms of the TiO2 surfaces remain nearly intact. Similar changes can be found for other dopants. Accordingly, the doped sites and proximal-O atoms undertake the change in charged states due to doping [35,36,37,38,39,40,41,42,43]. Electron transfer occurs considerably between the As(OH)3 and TiO2 surface and is promoted by the direct interaction of the As center with the doped site that further leads to As(III) oxidation. In addition to the doped site, the adjacent Ti and O atoms participate closely during As(OH)3 oxidation (Figures S13–S16), e.g., O4 in R5Fe5c in O•− form and subsurface Ti3 in R5N2c in Ti(III)• form, consistent with the XPS spectra indicating that doping renders the Ti atoms to be more electronegative [61]. Electron redistribution and dispersion greatly stabilize the As species, especially As(IV), where more atoms may be involved, e.g., at least six O atoms in A4Fe5c have clear spin densities and fall between O2− and O•−. Electron back-donation from TiO2 to As(OH)3 is promoted due to the formation of the multidentate motif, and As(V) is likely to be re-reduced to As(IV), e.g., Ti1 in R4N3c is in Ti(III)• form, while it turns to Ti(IV) in R6N3c due to the Ti1-O2 bonding that causes As(V) re-reduction.
The PDOS peaks for atoms that are closely involved during As(OH)3 adsorption are significantly affected. In all scenarios (pristine and doped at the Ti5c, O2c, and O3c sites), only specific valence orbitals have considerable PDOS overlapping and play a central role during As(OH)3 adsorption, e.g., Ti-3d, O-2p, As-4p, Fe-3d, and N-2p instead of Ti-4s, O-2s, As-4s, Fe-4s, and N-2s (see Figure 6 and Figure S20). The major PDOS peaks fall at the right side of the Fermi level for Ti-3d (e.g., 0.2~5.9 eV in R3Pr, 1.7~6.2 eV in R3Fe5c, and 0.4~5.3 eV in R6N2c) and Fe-3d (e.g., 0.7~5.9 eV in R3Fe5c), whereas they fall at the left side for O-2p (e.g., −6.9~0.0 eV in R3Pr, −6.9~0.1 eV in R3Fe5c, and −9.5~−1.0 eV in R6N2c), As-4p (e.g., −6.9~−0.2 eV in R3Pr, −6.9~0.1 eV in R3Fe5c, and −9.2~−1.0 in R6N2c) and N-2p (e.g., −7.9~−0.6 eV in R6N2c). For the pristine rutile (110) surface, O-2p has the PDOS overlapping with the secondary peak of Ti-3d (e.g., −6.8~−0.1 eV in R3Pr), and this situation remains for Fe5c doping, e.g., overlapping with the secondary PDOS peaks of Ti-3d (−6.7~0.2 eV) and Fe-3d (−7.0~0.2 eV) in R3Fe5c. However, the major PDOS peaks of O-2p, As-4p, and N-2p have nearly identical PDOS domains that are beneficial for As(OH)3 adsorption, as evidenced by the As-N and As-OS bonding and large |Ead|. Although with similar bonding mechanisms of rutile (110) as As(OH)3, anatase (101) may have the obviously weaker PDOS peaks for Ti-3d (secondary), Fe-3d (tertiary), and N-2p (major) that disfavor adsorption, e.g., −5.4~−0.2 eV for Ti-3d and −5.4~−0.1 eV for Fe-3d in A3Fe5c. The PDOS overlapping of O1-2p with Ti1-3d and O3/O4-2p with Fe-3d shifts towards the higher-energy domains (−5.0~−0.2 vs. −5.7~−0.9 eV in A3), and neither of the two largest PDOS peaks of Fe-3d (1.8~6.3 eV; −7.9~−6.0 eV) participate during the interaction with those of O-2p. The major PDOS peak of N-2p in A4N2c is further divided into two sub-domains, −7.7~−4.4 and −4.4~−1.6 (main) eV, and the former dominates the overlapping with those of As-4p. This can be interpreted as that whether for pristine or for Fe-, N-doped forms, rutile (110) is more efficient than anatase (101) for As(OH)3 adsorption.

3. Computational Section

3.1. Models

The optimized lattice parameters of TiO2 polymorphs amount to a = b = 4.608 Å, c = 2.973 Å for rutile and a = b = 3.793 Å, c = 9.594 Å for anatase, and show good agreement with the literature reports [62,63]; (110) and (101) stand for the most frequently exposed and extensively studied facets for rutile and anatase [53,64], and their models are stoichiometric with the molecular formulas of Ti60O120 and Ti48O96, respectively (Figure 1). All models consist of six atomic layers [65,66,67], and the slabs were separated by 20.000 Å to avoid image interactions. The effects of water at TiO2 surfaces on As(OH)3 adsorption [68] were further investigated and found to be slight (see more details in Supplementary Materials S1: Effects of Adsorbed Water).
Figure 1 shows that the TiO2 surfaces have 2-fold (O2c) and 3-fold (O3c) coordinated O and 5-fold (Ti5c) and 6-fold (Ti6c) coordinated Ti atoms. The dopants investigated presently included DTi = Fe, Mn, V, Ce, B at the Ti5c site and DO = N, F, B, S at the O2c, O3c site [26,27,30,31,32,36,37,38,39,40,41,42,43,61]. The As(OH)3 adsorption configurations of rutile (110) and anatase (101) were nominated to be RnDTiDO and AnDTiDO, where R, A, DTi, and DO stand for rutile (110), anatase (101), and dopants at the Ti and O sites, and n refers to the number of specific adsorption configurations. For instance, R2Fe5c and A4N2c refer to the No. 2 adsorption configuration at the rutile (110) surface with Fe5c doping and the No. 4 adsorption configuration at the anatase (101) surface with N2c doping.

3.2. Methods

Spin-polarized DFT calculations were conducted using the Perdew–Burke–Ernzerhof (PBE) exchange–correlation functional at the generalized gradient approximation (GGA) level, as implemented within the Vienna ab initio simulation package (VASP) [69,70]. The projected-augmented wave (PAW) approach was employed to handle the electron–ion interactions, and the non-covalent interactions were described by the standard DFT-D3 (BJ) scheme [71,72]. DFT + U was adopted for the on-site Coulomb interaction of 3d and 4f electrons [73], and Hubbard correction was recommended for Ti-3d (Ueff = 4.2 eV) [74,75], Fe-3d (Ueff = 5.0 eV) [76], Mn-3d (Ueff = 4.5 eV) [77,78], V-3d (Ueff = 3.0 eV) [79,80], and Ce-4f (Ueff = 5.0 eV) [77,78]. The Brillouin zone was sampled using the 2 × 2 × 1 Monkhorst-Pack grid [31,56]. The cut-off energy was set to 400.0 eV and validated by higher values such as 450.0 and 500.0 eV. The structural optimizations were converged when the forces acting on all atoms were less than 0.05 eV/Å.
The adsorption energies (Ead) of As(OH)3 over the pristine and doped TiO2 surfaces were calculated similarly:
Ead(pristine) = EAs(OH)3/TiO2 − (ETiO2 + EAs(OH)3)
Ead(doped) = EAs(OH)3/D-TiO2 − (ED-TiO2 + EAs(OH)3)
where EAs(OH)3, ETiO2, and ED-TiO2 stand for the electronic energies of As(OH)3 and pristine and doped TiO2 surfaces, while EAs(OH)3/TiO2 and EAs(OH)3/D-TiO2 refer to the electronic energies of As(OH)3 adsorption structures corresponding to the pristine and doped TiO2 surfaces, respectively.
To further the understanding of interactions between As(OH)3 and TiO2 surfaces and effects of doping and facet control onto As(OH)3 adsorption, Bader charge and spin density calculations were conducted [81], and the oxidation states of the As centers in adsorption configurations were then determined. The isosurfaces of spin densities were visualized by means of the VESTA 3 software [82]. In addition, the PDOS was calculated to gain insight into the bonding mechanisms between As(OH)3 and TiO2 surfaces [83].

4. Conclusions

This study presents a comprehensive understanding of how doping regulates As(III) adsorption and removal by TiO2 and the critical roles of crystal control played therein. As(OH)3 adsorption is more altered at the rutile (110) rather than anatase (101) surface. (1) All dopants except B5c, F2c, and F3c form direct bonding with As(OH)3 that can be detected in a majority of new adsorption structures. (2) The atomic radius is critical for adsorption when doping the Ti5c site (Ce >> Ti, Fe, Mn, V >> B). Ce5c and B5c doping may have distinct As(OH)3 adsorption from Fe5c, Mn5c, and V5c doping: Ce5c doping leads to the bidentate mononuclear R6Ce5c and A6Ce5c, while B5c doping transforms O3c to O2c that further forms a direct bond with As(OH)3 in R7B5c and A7B5c. (3) For doping the O2c, O3c site, the perturbation extent of As(OH)3 adsorption is N > S > F and accords with the likelihood of forming more bonds. N doping stabilizes R5N2c with only the As-OS bonding and has the largest potential to form the As-N bond, with production of monodentate (R4N2c, R4N3c, and A4N2c), bidentate (R6N2c, R6N3c, and A6N3c) and tridentate (A5N2c) motifs. S doping facilitates the interaction with As(OH)3 and leads to tridentate R8S2c and A5S2c. (4) Doping the O2c, O3c rather than the Ti5c site causes more structural perturbation and diversity to As(OH)3 adsorption.
As(OH)3 adsorption structures generally have the following stability trend: tridentate, bidentate > monodentate > physisorbed. Similar to the scenario of pristine TiO2 surfaces, R3 and A3 with two Ti-OAs bonds often remain to be the most preferred adsorption configurations, while this is not the case for B5c (Ti-OAs + As-OTi), S2c (Ti-OAs + Ti-OAs + As-S) and B3c (As-OTi) doping of rutile (110), and Ce5c (Ce-OAs + Ce-OAs), B5c (As-OTi), N2c (As-N), S2c (Ti-OAs + Ti-OAs + As-S), N3c (Ti-OAs + As-N), S3c (Ti-OAs), and B3c (As-OTi) doping of anatase (101). Doping is crystal-dependent, and the doped rutile (110) rather than anatase (101) surface is apparently more efficient for As(III) adsorption. In addition, doping the O2c, O3c rather than the Ti5c site may lead to larger promoting effects for adsorption, and the Ead reaches −4.17, −4.13, and −4.67 eV for Mn5c, N2c, and N3c doping of rutile (110) and −1.99, −2.29, and −2.24 eV, for Ce5c, N2c, and N3c doping of anatase (101), respectively.
Distinct from pristine TiO2 surfaces, where the As centers are always in +III form, doping is prone to cause As(III) auto-oxidation when the As centers interact directly with the TiO2 surfaces. Rutile rather than anatase is more ready to trigger the auto-oxidation of As(III) to As(V). All adsorption structures with As(V) have superior stability, and doping also stabilizes the As(IV) species: R8S2c and A7B5c become the most preferred for S2c and B5c doping. The multidentate adsorption of As(OH)3 causes electron back-donation and may cause As(V) re-reduction to As(IV). The mechanisms regarding how doping regulates As(III) adsorption by TiO2 and the critical roles of crystal control played therein are further rationalized at the molecular level.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/molecules29173991/s1.

Author Contributions

Data curation, methodology, software, writing—original draft, X.H.; validation, writing—review and editing, M.W. and R.H.; supervision, conceptualization, writing—review and editing, G.Y. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the Innovative Research Groups of CQ, China, grant number CXQT19006, the Fundamental Research Funds for the Central Universities, grant number XDJK2019B038, and the Chongqing Scientific Innovation Project for Postgraduates, grant number CYB22122.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within this article and the Supplementary Materials.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Podgorski, J.; Berg, M. Global threat of arsenic in groundwater. Science 2020, 368, 845–850. [Google Scholar] [CrossRef] [PubMed]
  2. He, D.; Hu, H.; Jiao, F.; Zuo, W.; Liu, C.; Xie, H.; Dong, L.; Wang, X. Thermal separation of heavy metals from municipal solid waste incineration fly ash: A review. Chem. Eng. J. 2023, 467, 143344. [Google Scholar] [CrossRef]
  3. Han, Y.S.; Park, J.H.; Min, Y.; Lim, D.H. Competitive adsorption between phosphate and arsenic in soil containing iron sulfide: XAS experiment and DFT calculation approaches. Chem. Eng. J. 2020, 397, 125426. [Google Scholar] [CrossRef]
  4. Zhang, W.; Liu, C.; Zheng, T.; Ma, J.; Zhang, G.; Ren, G.; Wang, L.; Liu, Y. Efficient oxidation and sorption of arsenite using a novel titanium(IV)-manganese(IV) binary oxide sorbent. J. Hazard. Mater. 2018, 353, 410–420. [Google Scholar] [CrossRef]
  5. Luo, J.; Yu, D.; Hristovski, K.D.; Fu, K.; Shen, Y.; Westerhoff, P.; Crittenden, J.C. Critical review of advances in engineering nanomaterial adsorbents for metal removal and recovery from water: Mechanism identification and engineering design. Environ. Sci. Technol. 2021, 55, 4287–4304. [Google Scholar] [CrossRef]
  6. Yan, L.; Chan, T.; Jing, C. Arsenic adsorption on hematite facets: Spectroscopy and DFT study. Environ. Sci. Nano 2020, 7, 3927–3939. [Google Scholar] [CrossRef]
  7. Mo, Z.; Tai, D.; Zhang, H.; Shahab, A. A comprehensive review on the adsorption of heavy metals by zeolite imidazole framework (ZIF-8) based nanocomposite in water. Chem. Eng. J. 2022, 443, 136320. [Google Scholar] [CrossRef]
  8. Rosales, M.; Orive, J.; Espinoza-González, R.; de Luis, R.F.; Gauvin, R.; Brodusch, N.; Rodríguez, B.; Gracia, F.; García, A. Evaluating the bi-functional capacity for arsenic photo-oxidation and adsorption on anatase TiO2 nanostructures with tunable morphology. Chem. Eng. J. 2021, 415, 128906. [Google Scholar] [CrossRef]
  9. Liu, B.; Kim, K.H.; Kumar, V.; Kim, S. A review of functional sorbents for adsorptive removal of arsenic ions in aqueous systems. J. Hazard. Mater. 2020, 388, 121815. [Google Scholar] [CrossRef]
  10. Guan, X.; Du, J.; Meng, X.; Sun, Y.; Sun, B.; Hu, Q. Application of titanium dioxide in arsenic removal from water: A review. J. Hazard. Mater. 2012, 215, 1–16. [Google Scholar] [CrossRef]
  11. Sharma, V.K.; Sohn, M. Aquatic arsenic: Toxicity, speciation, transformations, and remediation. Environ. Int. 2009, 35, 743–759. [Google Scholar] [CrossRef] [PubMed]
  12. Liu, L.; Yang, Z.; Zhao, F.; Chai, Z.; Yang, W.; Xiang, H.; Liao, Q.; Si, M.; Lin, Z. Manganese doping of hematite enhancing oxidation and bidentate-binuclear complexation during As(III) remediation: Experiments and DFT calculation. Chem. Eng. J. 2023, 471, 144758. [Google Scholar] [CrossRef]
  13. Zhang, Y.; Zhao, B.; Wang, C.; Huang, Y.; Liu, X.; Wang, R.; Wang, C. Dual-functional effect encompassing adsorption and catalysis by Mn-modified iron-based sorbents for arsenic removal: Experimental and DFT study. J. Hazard. Mater. 2023, 459, 132079. [Google Scholar] [CrossRef]
  14. Cai, H.; Du, X.; Lin, Z.; Tao, X.; Zou, M.; Liu, J.; Zhang, L.; Dang, Z.; Lu, G. Enhanced arsenic (III) sequestration via sulfidated zero-valent iron in aerobic conditions: Adsorption and oxidation coupling processes. J. Hazard. Mater. 2023, 459, 132190. [Google Scholar] [CrossRef] [PubMed]
  15. Zou, Y.; Wang, X.; Khan, A.; Wang, P.; Liu, Y.; Alsaedi, A.; Hayat, T.; Wang, X. Environmental remediation and application of nanoscale zero-valent iron and its composites for the removal of heavy metal ions: A review. Environ. Sci. Technol. 2016, 50, 7290–7304. [Google Scholar] [CrossRef] [PubMed]
  16. Yang, D.; Wang, R.; Feng, X.; Chu, Z.; Li, J.; Wei, W.; Zheng, R.; Zhang, J.; Chen, H. Transferring waste red mud into ferric oxide decorated ANA-type zeolite for multiple heavy metals polluted soil remediation. J. Hazard. Mater. 2022, 424, 127244. [Google Scholar] [CrossRef] [PubMed]
  17. Li, Y.; Liu, J.R.; Jia, S.Y.; Guo, J.W.; Zhuo, J.; Na, P. TiO2 pillared montmorillonite as a photoactive adsorbent of arsenic under UV irradiation. Chem. Eng. J. 2012, 191, 66–74. [Google Scholar] [CrossRef]
  18. Uddin, M.K. A review on the adsorption of heavy metals by clay minerals, with special focus on the past decade. Chem. Eng. J. 2017, 308, 438–462. [Google Scholar] [CrossRef]
  19. Zou, C.; Liu, H.; Hu, H.; Huang, Y.; Wang, C.; Yao, H. Effect of CO2 on the As2O3 adsorption over carbonaceous surface: Experiment and quantum chemistry study. Chem. Eng. J. 2022, 446, 137156. [Google Scholar] [CrossRef]
  20. Wu, D.; Liu, J.; Yang, Y.; Zhao, Y.; Zheng, Y. The role of SO2 in arsenic removal by carbon-based sorbents: A DFT study. Chem. Eng. J. 2021, 410, 128439. [Google Scholar] [CrossRef]
  21. Asere, T.G.; Stevens, C.V.; Du Laing, G. Use of (modified) natural adsorbents for arsenic remediation: A review. Sci. Total Environ. 2019, 676, 706–720. [Google Scholar] [CrossRef] [PubMed]
  22. Kumar, R.; Patel, M.; Singh, P.; Bundschuh, J.; Pittman, C.U., Jr.; Trakal, L.; Mohan, D. Emerging technologies for arsenic removal from drinking water in rural and peri-urban areas: Methods, experience from, and options for Latin America. Sci. Total Environ. 2019, 694, 133427. [Google Scholar] [CrossRef]
  23. Du, T.; Wang, J.; Zhang, L.; Wang, S.; Yang, C.; Xie, L.; Liu, Z.; Ni, Y.; Xie, X.; Sun, J.; et al. Missing-linker engineering of Eu(III)-doped UiO-MOF for enhanced detection of heavy metal ions. Chem. Eng. J. 2022, 431, 134050. [Google Scholar] [CrossRef]
  24. Hu, P.; Wang, S.; Zhuo, Y. Research on As2O3 adsorption enhancement characteristics of Mn-modified γ-Al2O3. Chem. Eng. J. 2021, 426, 131660. [Google Scholar] [CrossRef]
  25. Huang, X.; Li, T.; Yang, G. Immobilization of As(III) by gibbsite and catalytic oxidation to As(V): Profound impacts of doping and unraveling of associated mechanisms. Chemosphere 2023, 313, 137583. [Google Scholar] [CrossRef]
  26. Chen, K.; Chen, C.; Ren, X.; Alsaedi, A.; Hayat, T. Interaction mechanism between different facet TiO2 and U(VI): Experimental and density-functional theory investigation. Chem. Eng. J. 2019, 359, 944–954. [Google Scholar] [CrossRef]
  27. Pena, M.E.; Korfiatis, G.P.; Patel, M.; Lippincott, L.; Meng, X. Adsorption of As(V) and As(III) by nanocrystalline titanium dioxide. Water Res. 2005, 39, 2327–2337. [Google Scholar] [CrossRef] [PubMed]
  28. Wu, X.; Hu, J.; Wu, F.; Zhang, X.; Wang, B.; Yang, Y.; Shen, G.; Liu, J.; Tao, S.; Wang, X. Application of TiO2 nanoparticles to reduce bioaccumulation of arsenic in rice seedlings (Oryza sativa L.): A mechanistic study. J. Hazard. Mater. 2021, 405, 124047. [Google Scholar] [CrossRef] [PubMed]
  29. Wang, D.; Zhi, T.; Liu, L.; Yan, L.; Yan, W.; Tang, Y.; He, B.; Jing, C.; Jiang, G. 3D printing of TiO2 nano particles containing macrostructures for As(III) removal in water. Sci. Total Environ. 2022, 815, 152754. [Google Scholar] [CrossRef]
  30. Zimmerman, A.J.; Gutierrez, D.G.; Shaghaghi, N.; Sharma, A.; Deonarine, A.; Landrot, G.; Weindorf, D.C.; Siebecker, M.G. Mobility and bioaccessibility of arsenic (As) bound to titanium dioxide (TiO2) water treatment residuals (WTRs). Environ. Pollut. 2023, 326, 121468. [Google Scholar] [CrossRef]
  31. Huang, X.; Yang, G. Enhanced immobilization of arsenic(III) and auto-oxidation to arsenic(V) by titanium oxide (TiO2), due to single-atom vacancies and oxyanion formation. J. Colloid Interf. Sci. 2023, 650, 1327–1338. [Google Scholar] [CrossRef] [PubMed]
  32. Umare, S.S.; Charanpahari, A.; Sasikala, R. Enhanced visible light photocatalytic activity of Ga, N and S codoped TiO2 for degradation of azo dye. Mater. Chem. Phys. 2013, 140, 529–534. [Google Scholar] [CrossRef]
  33. Wei, Z.; Liang, K.; Wu, Y.; Zou, Y.; Zuo, J.; Arriagada, D.C.; Pan, Z.; Hu, G. The effect of pH on the adsorption of arsenic (III) and arsenic (V) at the TiO2 anatase [101] surface. J. Colloid Interf. Sci. 2016, 462, 252–259. [Google Scholar] [CrossRef] [PubMed]
  34. Kumar, R.; Jing, C.; Yan, L. A critical review on arsenic and antimony adsorption and transformation on mineral facets. J. Environ. Sci. 2024, in press. [Google Scholar] [CrossRef]
  35. Ivan, A.; Dalibor, S.; Jelena, N.; Jugoslav, K.; Predrag, V.; Dragan, M.; Goran, R. Fe doped TiO2 prepared by microwave-assisted hydrothermal process for removal of As(III) and As(V) from water. Ind. Eng. Chem. Res. 2014, 53, 10841–10848. [Google Scholar]
  36. Song, G.; Gao, R.; Zhao, Z.; Zhang, Y.; Tan, H.; Li, H.; Wang, D.; Sun, Z. High-spin state Fe(III) doped TiO2 for electrocatalytic nitrogen fixation induced by surface F modification. Appl. Catal. B 2022, 301, 120809. [Google Scholar] [CrossRef]
  37. Shan, C.; Liu, H.; Hua, M.; Pan, B. Enhanced Fenton-like oxidation of As(III) over Ce-Ti binary oxide: A new strategy to tune catalytic activity via balancing bimolecular adsorption energies. Environ. Sci. Technol. 2020, 54, 5893–5901. [Google Scholar] [CrossRef]
  38. Cano-Casanova, L.; Ansón-Casaos, A.; Hernández-Ferrer, J.; Benito, A.M.; Maser, W.K.; Garro, N.; Lillo-Ródenas, M.A.; Román-Martínez, M.D.C. Surface-enriched boron-doped TiO2 nanoparticles as photocatalysts for propene oxidation. ACS Appl. Nano Mater. 2022, 5, 12527–12539. [Google Scholar] [CrossRef]
  39. García-Mota, M.; Vojvodic, A.; Abild-Pedersen, F.; Nørskov, J.K. Electronic origin of the surface reactivity of transition-metal-doped TiO2 (110). J. Phys. Chem. C 2013, 117, 460–465. [Google Scholar] [CrossRef]
  40. Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visible-light photocatalysis in nitrogen-doped titanium oxides. Science 2001, 293, 269–271. [Google Scholar] [CrossRef]
  41. Liu, J.; Liu, Q.; Fang, P.; Pan, C.; Xiao, W. First principles study of the adsorption of a NO molecule on N-doped anatase nanoparticles. Appl. Surf. Sci. 2012, 258, 8312–8318. [Google Scholar] [CrossRef]
  42. Zhang, J.; Liao, J.; Yang, F.; Xu, M.; Lin, S. Regulation of the electroanalytical performance of ultrathin titanium dioxide nanosheets toward lead ions by non-metal doping. Nanomaterials 2017, 7, 327. [Google Scholar] [CrossRef] [PubMed]
  43. Han, C.; Pelaez, M.; Likodimos, V.; Kontos, A.G.; Falaras, P.; O’Shea, K.; Dionysiou, D.D. Innovative visible light-activated sulfur doped TiO2 films for water treatment. Appl. Catal. B 2011, 107, 77–87. [Google Scholar] [CrossRef]
  44. Deedar, N.A.B.I.; Aslam, I. Evaluation of the adsorption potential of titanium dioxide nanoparticles for arsenic removal. J. Environ. Sci. 2009, 21, 402–408. [Google Scholar]
  45. Khan, M.S.; Shah, J.A.; Arshad, M.; Halim, S.A.; Khan, A.; Shaikh, A.J.; Riaz, N.; Khan, A.J.; Arfan, M.; Shahid, M.; et al. Photocatalytic decolorization and biocidal applications of nonmetal doped TiO2: Isotherm, kinetic modeling and In Silico molecular docking studies. Molecules 2020, 25, 4468. [Google Scholar] [CrossRef]
  46. Grajciar, L.; Heard, C.J.; Bondarenko, A.A.; Polynski, M.V.; Meeprasert, J.; Pidko, E.A.; Nachtigall, P. Towards operando computational modeling in heterogeneous catalysis. Chem. Soc. Rev. 2018, 47, 8307–8348. [Google Scholar] [CrossRef]
  47. Vogiatzis, K.D.; Polynski, M.V.; Kirkland, J.K.; Townsend, J.; Hashemi, A.; Liu, C.; Pidko, E.A. Computational approach to molecular catalysis by 3d transition metals: Challenges and opportunities. Chem. Rev. 2018, 119, 2453–2523. [Google Scholar] [CrossRef]
  48. Albuquerque, A.R.; Bruix, A.; dos Santos, I.M.; Sambrano, J.R.; Illas, F. DFT study on Ce-doped anatase TiO2: Nature of Ce3+ and Ti3+ centers triggered by oxygen vacancy formation. J. Phys. Chem. C 2014, 118, 9677–9689. [Google Scholar] [CrossRef]
  49. Finazzi, E.; Di Valentin, C.; Pacchioni, G. Boron-doped anatase TiO2: Pure and hybrid DFT calculations. J. Phys. Chem. C 2009, 113, 220–228. [Google Scholar] [CrossRef]
  50. Yan, L.; Du, J.; Jing, C. How TiO2 facets determine arsenic adsorption and photooxidation: Spectroscopic and DFT studies. Catal. Sci. Technol. 2016, 6, 2419–2426. [Google Scholar] [CrossRef]
  51. Yu, Y.; Zhou, Z.; Ding, Z.; Zuo, M.; Cheng, J.; Jing, C. Simultaneous arsenic and fluoride removal using {201} TiO2–ZrO2: Fabrication, characterization, and mechanism. J. Hazard. Mater. 2019, 377, 267–273. [Google Scholar] [CrossRef]
  52. Sarkar, A.; Paul, B. Synthesis, characterization of iron-doped TiO2(B) nanoribbons for the adsorption of As(III) from drinking water and evaluating the performance from the perspective of physical chemistry. J. Mol. Liq. 2021, 322, 114556. [Google Scholar] [CrossRef]
  53. Li, Y.F.; Aschauer, U.; Chen, J.; Selloni, A. Adsorption and reactions of O2 on anatase TiO2. Acc. Chem. Res. 2014, 47, 3361–3368. [Google Scholar] [CrossRef]
  54. Wang, Q.; Li, T.; Zhu, C.; Huang, X.; Yang, G. Molecular insights for uranium(VI) adsorption at nano-TiO2 surfaces and reduction by alcohols and biomass sugars. Chem. Eng. J. Adv. 2022, 10, 100264. [Google Scholar] [CrossRef]
  55. Li, Z.; Deng, S.; Yu, G.; Huang, J.; Lim, V.C. As(V) and As(III) removal from water by a Ce-Ti oxide adsorbent: Behavior and mechanism. Chem. Eng. J. 2010, 161, 106–113. [Google Scholar] [CrossRef]
  56. Huang, X.; Zhu, C.; Wang, Q.; Yang, G. Mechanisms for As(OH)3 and H3AsO4 adsorption at anhydrous and hydrated surfaces of gibbsite and possibility for anionic As(III) and As(V) formation. Appl. Surf. Sci. 2020, 525, 146494. [Google Scholar] [CrossRef]
  57. Jiang, S.; Li, X.; Yin, Y.; Luo, B.; Isah, A.G.; Zhang, Z.; Zhu, T. Extraordinary deactivation offset effect of zinc and arsenic on V2O5-WO3/TiO2 catalysts: Like cures like. J. Hazard. Mater. 2023, 441, 129894. [Google Scholar] [CrossRef] [PubMed]
  58. Lei, M.; Zhang, Y.; Wang, M.; Yang, W.; Gao, Z. Density functional theory investigation of As4, As2 and AsH3 adsorption on Ti-doped graphene. Chem. Eng. J. 2021, 421, 129747. [Google Scholar] [CrossRef]
  59. Chen, H.; Yan, L.; Jing, C. DFT study on TiO2 facet-dependent As(III) oxidation process: Importance of As(IV) species. Surf. Sci. 2023, 729, 122219. [Google Scholar] [CrossRef]
  60. Tobón, L.L.A.; Fuente, S.; Branda, M.M. Electronic and magnetic properties of the adsorption of As harmful species on zero-valent Fe surfaces, clusters and nanoparticules. Appl. Surf. Sci. 2019, 465, 715–723. [Google Scholar] [CrossRef]
  61. Wang, X.; Pehkonen, S.O.; Rämö, J.; Väänänen, M.; Highfield, J.G.; Laasonen, K. Experimental and computational studies of nitrogen doped Degussa P25 TiO2: Application to visible-light driven photo-oxidation of As(III). Catal. Sci. Technol. 2012, 2, 784–793. [Google Scholar] [CrossRef]
  62. Zhang, M.; He, G.; Pan, G. Binding mechanism of arsenate on rutile (110) and (001) planes studied using grazing-incidence EXAFS measurement and DFT calculation. Chemosphere 2015, 122, 199–205. [Google Scholar] [CrossRef] [PubMed]
  63. Howard, C.J.; Sabine, T.M.; Dickson, F. Structural and thermal parameters for rutile and anatase. Acta Crystallog. B 1991, 47, 462–468. [Google Scholar] [CrossRef]
  64. Pang, C.L.; Lindsay, R.; Thornton, G. Chemical reactions on rutile TiO2 (110). Chem. Soc. Rev. 2008, 37, 2328–2353. [Google Scholar] [CrossRef] [PubMed]
  65. Martsinovich, N.; Jones, D.R.; Troisi, A. Electronic structure of TiO2 surfaces and effect of molecular adsorbates using different DFT implementations. J. Phys. Chem. C 2010, 114, 22659–22670. [Google Scholar] [CrossRef]
  66. Kim, H.L.; Hidayat, R.; Khumaini, K.; Lee, W.J. A theoretical study on the surface reaction of tetrakis (dimethylamino) titanium on titanium oxide. Phys. Chem. Chem. Phys. 2023, 25, 22250–22257. [Google Scholar] [CrossRef]
  67. Di Liberto, G.; Tosoni, S.; Pacchioni, G. Role of surface termination in forming type-II photocatalyst heterojunctions: The case of TiO2/BiVO4. J. Phys. Condens. Matter 2020, 33, 075001. [Google Scholar] [CrossRef]
  68. Maleki, F.; Di Liberto, G.; Pacchioni, G. pH-and facet-dependent surface chemistry of TiO2 in aqueous environment from first principles. ACS Appl. Mater. Interfaces 2023, 15, 11216–11224. [Google Scholar] [CrossRef]
  69. Kresse, G.; Furthmüller, J. Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys. Rev. B 1996, 54, 11169. [Google Scholar] [CrossRef]
  70. Perdew, J.P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865. [Google Scholar] [CrossRef]
  71. Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B 1999, 59, 1758. [Google Scholar] [CrossRef]
  72. Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, 132, 154104. [Google Scholar] [CrossRef] [PubMed]
  73. Dudarev, S.L.; Botton, G.A.; Savrasov, S.Y.; Humphreys, C.J.; Sutton, A.P. Electron-energy-loss spectra and the structural stability of nickel oxide: An LSDA+ U study. Phys. Rev. B 1998, 57, 1505. [Google Scholar] [CrossRef]
  74. Wang, L.; Song, W.; Deng, J.; Zheng, H.; Liu, J.; Zhao, Z.; Gao, M.; Wei, Y. Facet-dependent photocatalytic decomposition of N2O on the anatase TiO2: A DFT study. Nanoscale 2018, 10, 6024–6038. [Google Scholar] [CrossRef]
  75. Zhang, J.; Peng, C.; Wang, H.; Hu, P. Identifying the role of photogenerated holes in photocatalytic methanol dissociation on rutile TiO2 (110). ACS Catal. 2017, 7, 2374–2380. [Google Scholar] [CrossRef]
  76. Xu, C.; Zhang, Y.; Chen, J.; Lin, J.; Zhang, X.; Wang, Z.; Zhou, J. Enhanced mechanism of the photo-thermochemical cycle based on effective Fe-doping TiO2 films and DFT calculations. Appl. Catal. B 2017, 204, 324–334. [Google Scholar] [CrossRef]
  77. Cen, W.; Liu, Y.; Wu, Z.; Wang, H.; Weng, X. A theoretic insight into the catalytic activity promotion of CeO2 surfaces by Mn doping. Phys. Chem. Chem. Phys. 2012, 14, 5769–5777. [Google Scholar] [CrossRef]
  78. D’Alessandro, O.; Pintos, D.G.; Juan, A.; Irigoyen, B.; Sambeth, J. A DFT study of phenol adsorption on a low doping Mn-Ce composite oxide model. Appl. Surf. Sci. 2015, 359, 14–20. [Google Scholar] [CrossRef]
  79. Ranea, V.A.; Quiña, P.L.D.; Yalet, N.M. General adsorption model for H2S, H2Se, H2Te, NH3, PH3, AsH3 and SbH3 on the V2O5 (001) surface including the van der Waals interaction. Chem. Phys. Lett. 2019, 720, 58–63. [Google Scholar] [CrossRef]
  80. Ranea, V.A.; Quiña, P.L.D. The structure of the bulk and the (001) surface of V2O5. A DFT+ U study. Mater. Res. Express 2016, 3, 085005. [Google Scholar] [CrossRef]
  81. Henkelman, G.; Arnaldsson, A.; Jónsson, H. A fast and robust algorithm for Bader decomposition of charge density. Comput. Mater. Sci. 2006, 36, 354–360. [Google Scholar] [CrossRef]
  82. Momma, K.; Izumi, F. VESTA3 for three-dimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallog. 2011, 44, 1272–1276. [Google Scholar] [CrossRef]
  83. van Santen, R.A.; Tranca, I. How molecular is the chemisorptive bond? Phys. Chem. Chem. Phys. 2016, 18, 20868–20894. [Google Scholar] [CrossRef] [PubMed]
Molecules 29 03991 g001
Figure 1. Periodic models for (a) rutile (110) and (b) anatase (101) surfaces, as well as corresponding top-surface structures. Color scheme: Ti (blue) and O (red). In (c,d), different types of surface atoms are marked: O2c (two-fold O atom, highlighted in green); O3c (three-fold O atom, highlighted in purple); Ti5c (five-fold Ti atom, highlighted in yellow); and Ti6c (six-fold Ti atom, highlighted in blue).
Figure 1. Periodic models for (a) rutile (110) and (b) anatase (101) surfaces, as well as corresponding top-surface structures. Color scheme: Ti (blue) and O (red). In (c,d), different types of surface atoms are marked: O2c (two-fold O atom, highlighted in green); O3c (three-fold O atom, highlighted in purple); Ti5c (five-fold Ti atom, highlighted in yellow); and Ti6c (six-fold Ti atom, highlighted in blue).
Molecules 29 03991 g001
Molecules 29 03991 g002
Figure 2. Adsorption configurations of As(OH)3 over rutile (110) surface with Ti5c site being doped (DTi = Fe, Mn, V, Ce, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), Fe (golden), Mn (purple), V (silvery), Ce (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Figure 2. Adsorption configurations of As(OH)3 over rutile (110) surface with Ti5c site being doped (DTi = Fe, Mn, V, Ce, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), Fe (golden), Mn (purple), V (silvery), Ce (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Molecules 29 03991 g002
Molecules 29 03991 g003
Figure 3. Adsorption configurations of As(OH)3 over rutile (110) surface with O2c, O3c site being doped (DO = N, F, S, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), N (gray), F (orange), S (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Figure 3. Adsorption configurations of As(OH)3 over rutile (110) surface with O2c, O3c site being doped (DO = N, F, S, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), N (gray), F (orange), S (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Molecules 29 03991 g003
Molecules 29 03991 g004
Figure 4. Adsorption configurations of As(OH)3 over the anatase (101) surface with the Ti5c site being doped (DTi = Fe, Mn, V, Ce, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), Fe (golden), Mn (purple), V (silvery), Ce (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Figure 4. Adsorption configurations of As(OH)3 over the anatase (101) surface with the Ti5c site being doped (DTi = Fe, Mn, V, Ce, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), Fe (golden), Mn (purple), V (silvery), Ce (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Molecules 29 03991 g004
Molecules 29 03991 g005
Figure 5. Adsorption configurations of As(OH)3 over the anatase (101) surface with the O2c, O3c site being doped (DO = N, F, S, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), N (gray), F (orange), S (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Figure 5. Adsorption configurations of As(OH)3 over the anatase (101) surface with the O2c, O3c site being doped (DO = N, F, S, B), together with adsorption energies (Ead) and oxidation states for the As centers (in parentheses). Color scheme: Ti (blue), O (red), As (green), H (white), N (gray), F (orange), S (yellow), and B (pink). Selected H-bonds are indicated by dashed gray lines. Distances are given in Å.
Molecules 29 03991 g005
Molecules 29 03991 g006
Figure 6. Projected density of states (PDOS) for As(OH)3 adsorption at rutile (110) and anatase (101) surfaces in pristine and doped forms (DTi = Fe; DO = N). The Fermi level (EF) is set to zero energy and highlighted by gray dotted line.
Figure 6. Projected density of states (PDOS) for As(OH)3 adsorption at rutile (110) and anatase (101) surfaces in pristine and doped forms (DTi = Fe; DO = N). The Fermi level (EF) is set to zero energy and highlighted by gray dotted line.
Molecules 29 03991 g006
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Huang, X.; Wu, M.; Huang, R.; Yang, G. How Doping Regulates As(III) Adsorption at TiO2 Surfaces: A DFT + U Study. Molecules 2024, 29, 3991. https://doi.org/10.3390/molecules29173991

AMA Style

Huang X, Wu M, Huang R, Yang G. How Doping Regulates As(III) Adsorption at TiO2 Surfaces: A DFT + U Study. Molecules. 2024; 29(17):3991. https://doi.org/10.3390/molecules29173991

Chicago/Turabian Style

Huang, Xiaoxiao, Mengru Wu, Rongying Huang, and Gang Yang. 2024. "How Doping Regulates As(III) Adsorption at TiO2 Surfaces: A DFT + U Study" Molecules 29, no. 17: 3991. https://doi.org/10.3390/molecules29173991

Article Metrics

Back to TopTop