Competitive adsorption of molybdate, chromate, sulfate, selenate, and selenite on γ-Al2O3

(1)

A: Physicochemical and Engineering Aspects 166 (2000) 251 – 259

Competitive adsorption of molybdate, chromate, sulfate,

selenate, and selenite on

g-Al

₂

O

₃

Chung-Hsin Wu *, Shang-Lien Lo, Cheng-Fang Lin

Graduate Institute of En6ironmental Engineering, National Taiwan Uni6ersity,71Chou-Shan Road, Taipei106, Taiwan, ROC Received 8 September 1998; accepted 19 August 1999

Abstract

Competitive adsorption of molybdate, selenite, selenate, chromate, and sulfate ontog-Al2O3was investigated in the

present study. Binary-solute systems of MoO42 −/SeO32 −, CrO42 −/SO42 −, and CrO42 −/SeO42 − as well as single anion

systems were evaluated for the relative influence on competitive adsorption on oxide surface. As would be expected, the adsorption density of each anion in the binary-solute systems decreases, as compared to the respective density in a single anion system. Furthermore, MoO42 −inhibits SeO32 −adsorption in acidic condition and that SO42 −or SeO42 −

depresses CrO42 −. The order of the relative retainment of anions on oxide surface is molybdate\selenite\selenate

Keywords:Competitive adsorption; Molybdate; Chromate; Sulfate; Selenate; Selenite; TLM

www.elsevier.nl/locate/colsurfa

1. Introduction

Mobility of anions in aquatic environment is generally regulated by adsorption at the solid/wa-ter insolid/wa-terface as well as competition for surface binding sites among various anion species. In the past decades, several studies have been devoted for elucidating anion interactions with minerals and Fe/Al oxides [1 – 8]. In those studies, molyb-date, chromate, selenate, selenite, arsenate, arsen-ite, phosphate and sulfate are investigated because they are essential trace nutrients as well as toxic substances to plants and aquatic life. Anion

ad-sorption at the oxide/water interfaces is com-monly interpreted by the surface complexation mechanism in which anionic solute is binding to the surface reacting site to form either inner-sphere or ion-pair complex. Based on the effect of solution ionic strength and simulation by triple-layer modeling (TLM), adsorption of selenite, chromate, arsenate, and molybdate are regarded as specific coordination with surface hydroxyl groups [5,8,9]. Sorption of sulfate, however, has been typically recognized to form an outer-sphere complex [4,10]. With the use of extended X-ray absorption fine structure spectroscopy, surface configuration of the anion inner-sphere complex is further evident [11,12].

* Corresponding author. Tel.: + 886-2-2392-7653; fax: + 886-2-2362-7427.

(2)

Anion partitioning onto oxide surfaces is also greatly affected by the competitive adsorption of anionic cosolutes in the aqueous systems [2,13,14]. Anions with comparable adsorptive affinities for oxide reacting sites can exhibit a pronounced effect on competitive adsorption over a wide envi-ronmental condition and relative concentration [13]. For example, Roy et al. [13] indicated that the adsorption of arsenate and molybdate was significantly reduced by the competitive adsorp-tion of phosphate. Nevertheless, adsorpadsorp-tion of strongly binding anions can only be depressed if the weakly binding anions are present in a large excess amount and under low pH conditions [14]. Mesuere and Fish [6] showed that the sorption of chromate onto a-FeOOH in acidic environments was diminished by the competitive adsorption of organic oxalate. Sulfate has been shown to de-crease the adsorption of As(III) and As(V) in acidic systems [7,15]. Dynes and Huang [16] also indicated that the competitiveness of the organic acids with selenite for sorption sites increased with increasing organic acid concentration.

In most of the previous studies, influence of competitive adsorption on anion partitioning has been addressed in binary-solute systems contain-ing strongly bindcontain-ing anion with organic com-pound or sulfate [16,17]. A systematic investigation on the relative competition for sorp-tion onto oxides among various anions with dif-ferent binding affinity is rather limited. In practical situations, multi-component mixture competition might increase the complexity of an-ion interactan-ion at the water/oxide interfaces. Such complex systems frequently occur, and the magni-tude of these competitive interactions needs to be clarified in order to better predict contaminant behavior and to identify anions that are more likely to be affected in their transport in the presence of other anions. Consequently, this study focuses on the competitive adsorption of molyb-date, chromate, sulfate, selenate, and selenite onto g-Al2O3. The findings are of fundamental

signifi-cance in better understanding the influence of competitive adsorption on the fate of anions in the environment.

2. Materials and methods 2.1. Materials

All solutions were prepared with deionized wa-ter (Milli-Q) and reagent-grade chemicals. Stock anion solutions (10− 2 _{M) were prepared with}

Na2MoO4, Na2CrO4, Na2SO4, Na2SeO4, or

Na2SeO3. The background electrolyte NaNO3

concentrations were adjusted to 0.01, 0.05, and 0.1 M. The adsorbent g-Al2O3 obtained from

Aerosil Co. (Japan) was purified by electrodialysis (1200 V, 3 mA) before use in the sorption experi-ments. The impure substances which could easily affect adsorption results were removed by the purification procedures.

2.2. Sorption experiments

In batch equilibrium adsorption experiments, anions (5 × 10− 3 _{M) were equilibrated with}

g-Al2O3suspension (30 g/l) in the presence of 0.01,

0.05, or 0.1 M NaNO3. A small amount of HNO3

or NaOH was added to cover the pH range 4.5 – 9 for MoO42 −, pH 2 – 10 for CrO42 − and SO42 −, pH

2 – 9 for SeO42 −, and pH 5 – 10 for SeO32 −. After

the desired pH was reached, 10 ml of the suspen-sion was removed to a 15-ml polypropylene tube. All experiments were performed in a tightly capped 15-ml polypropylene tube under N₂ atmo-sphere at 2590.1°C for 24 h. At the end of the equilibrium period, the pH of each suspension was determined. The suspensions were centrifuged at 9500 rpm for 10 min, and the supernatant was then filtered through 0.2mm filter paper (Gelman Sciences) for later analyses of anion concentra-tion. In competitive adsorption experiments, same concentrations of adsorbate and adsorbent were used. Three binary-solute systems as pH 4.5 – 9 for MoO42 −+ SeO32 −, pH 2 – 10 for SO42 −+ CrO42 −

and SeO42 −+ CrO42 − were designated for

under-standing the relative binding strength between anions.

All anion concentrations were determined by ion chromatography (Dionex 2000i SP). The spe-cific surface area of g-A₂O₃ was calculated to be 100 m2_{/g based on the BET measurements.}

(3)

2.3. Modeling of adsorption data

The TLM developed by Davis and Leckie [18] and subsequently modified by Hayes and Leckie [19] was used for simulating the equilibrium parti-tioning of anion species and background elec-trolyte ions at the g-Al₂O₃/water interface. As described previously, model analogous of two dif-ferent types of surface complexes can be incorpo-rated to TLM, either as inner-sphere or outer-sphere ion-pair complexes. The former cor-responds to ions placed in the o-plane, the latter to ions placed in the b-plane. The TLM parame-ters and intrinsic acidity surface hydrolysis con-stants (Ka1intand Ka2int) and binding constants (KNaint+

and K_NO

3 −

int ₎ _of _the _background _electrolyte

(NaNO₃) with the surface are described in Table 1. Potentiometric titration experiments on 10 g/l g-Al2O3 suspensions with three different

back-ground electrolyte concentrations (0.1, 0.01 and 0.001 N) to calculate Ka1intand Ka2intand KNaint+and

K_NO

3 −

int _{follow the method of Davis et al. [22]. The}

acidity constants were determined by extrapolat-ing the titration data at lowest NaNO3

concentra-tion to zero fracconcentra-tional ionizaconcentra-tion. Similarly, the electrolyte binding constants were estimated by extrapolating the higher NaNO₃ concentrations results to zero fractional ionization. The log K_a1int_,

log Ka2int, log KNaint+ and log K_NO 3 −

int _{determined in}

this work were − 6.9, − 9.7, − 8.3 and 6.9, re-spectively. For comparison, the log Ka1int and

log Ka2int values reported by Hohl and Stumm [23]

were − 7.2 and − 9.5. The log K_Naint+ and

log K_NO

3 −

int _{values reported by Zhang and Sparks}

[4] and Hayes and Leckie [19] were − 9.1 and 8.7.

Table 2

Aqueous proton dissociated reaction constants used in this study

Reactions Constantsa Species

H2MoO4 HMoO4−= H++MoO42− pK2= 4.21 HCrO4−= H++CrO42−

H2CrO4 pK2= 6.49

H2SO4 HSO4−= H++SO42− pK2= 1.99 HSeO4−= H++SeO42−

H2SeO4 pK2= 1.92

H2SeO3 H2SeO3= H++HSeO3− pK1= 2.62 pK2= 8.32 HSeO3−= H++SeO32−

a_{From Perrin [24].}

Aqueous dissociated reaction constants are sum-marized in Table 2.

Ion-pair formation and surface coordination may occur at oxide/water interface. The reactions and equilibrium expressions in the TLM are sum-marized in Table 3, in which R is the gas constant;

T is the absolute temperature; F is the Faraday

constant; co is the surface potential on o-plane;

and c_b is the surface potential on the b-plane. Eqs. (1) and (2) in Table 3 describe protonation of reacting surface sites, and Eqs. (3) and (4) show the formation of complexes between the back-ground electrolyte ions and the surface. Ion-pairs are formed at the o-plane (Eqs. (5) and (6)), where the adsorption is nonspecific and the reac-tion product is an outer-sphere surface complex if anions react similarly to a background electrolyte with SOH. If adsorption of anions is visualized as a chemically specific reaction, the reaction can then be expressed as an inner-sphere surface coor-dination process (Eqs. (7) and (8)).

The parameters and equations in Tables 1 – 3 were used in the model analysis to determine the best-fit intrinsic equilibrium constants for the an-ion/g-Al2O3 reactions.

2.4. Kurbato6 plots for o6erall proton coefficient Differences in anion adsorption ong-Al2O3can

be examined from the slope of the Kurbatov plots for the adsorption data [3]. The anion adsorption reactions may be generalized by the following equation, in which the Kurbatov plot is based: SOH + nH+_{+ L}2 −_{= SOH}

n + 1L(n − 2) (9)

Table 1

TLM parameters forg-Al2O3 used in this study 100 (BET measurement) Specific surface area (m2_/g)

Site density (sites/nm2₎ _{8 [20]} Outer-layer capacitance 20 [21]

(C2,mF/cm2)

Inner-layer capacitance 80 [21] (C1,mF/cm

2₎

log Ka1int, log Ka2int −6.9, −9.7 (acid/base titration)

log K_Naint₊_{, log K} NO3−

int

−8.3, 6.9 (acid/base titration)

(4)

Ke=

[SOHn + 1L(n − 2)]

[SOH][H+_]n_[L2 −_] (10)

where K_e is an apparent equilibrium constant, SOH is proton-specific surface sites, and

SOHn + 1L(n − 2) is the total surface species, L2 − is

the total solution species, and n is the overall proton coefficient. The overall proton coefficient includes the effects of all surface and solution reactions involved in anion adsorption. The larger value of n implies a higher affinity for anion adsorbed onto oxide surface. Additionally, the proton coefficient is influenced by the nature of the adsorbent and by the sorbate-to-surface site concentration ratio. The distribution ratio (D) defined by

D =

%surface species %solution species

(11)

can be rewritten using Eqs. (9) and (10) so that log D is given by

log D = − npH + log Ke+ log[SOH]

log D = − npH + K (12) Assuming K (log K_e+ log[SOH]) is a constant, the overall proton coefficient (n) can, thus, be ob-tained by plotting log D versus pH.

3. Results and discussion

3.1. Effects of pH and background electrolyte The individual sorption envelopes of 5 × 10− 3

M anions (MoO₄2 −_{, CrO} 4 2 −_{, SO} 4 2 −_{, SeO} 4 2 −_{, and}

SeO32 −) onto 30 g/l g-Al2O3 in the presence of

three different NaNO3 electrolyte concentrations

(0.01, 0.05, and 0.1 M) are presented in Fig. 1. For all anions studied in this work, the sorption density increased with system pH shifting from alkaline to acidic conditions. Sorption of chro-mate (Fig. 1(b)), sulfate (Fig. 1(c)), and selenate (Fig. 1(d)) are affected by the NaNO3

concentra-tion, especially chromate in the low pH region. In contrast, sorption of molybdate and selenite does not show any noticeable differences with three NaNO₃ concentrations. As reported by other in-vestigators [4,5], the sorption envelopes of sulfate and selenate are dependent on the background

Table 3

Surface complexation reactions and equilibrium expressions for anion/g-Al2O3systemsa

Reactions Intrinsic equilibrium expressions SOH2+= SOH _K a1 int =[SOH][H +_] [SOH2+] exp

−c0F RT

(1) +H+ SOH = SO−_+H+ Ka2int= [SO−_][H+_] [SOH] exp

−c0F RT

(2) SOH+Na+ _K Na+ int = SO−_Na+ +H+ =[SO −_Na+_][H+_] [SOH][Na+_] exp

(c_b−c0)F RT

n

(3) K_NO 3 − int SOH+H+_+NO 3 − = SOH2+NO3− = [SO2 +_NO 3 −_] [SOH][H+_][NO 3 −_]exp

(c0−cb)F RT

n

(4) Outer-sphere complex K_L out 2− int SOH+H+_+L2− = SOH2+L2− = [SOH2 +_L2−_] [SOH][H+_][L2−_]exp

(c0−2cb)F RT

n

(5) K_HL out − int SOH+2H+ +L2− = SOH2+HL− = [SOH2 +_HL−_] [SOH][H+_][L2−_]exp

(c0−cb)F RT

n

(6) Inner-sphere complex SOH+H+_+L2− K_L in − int = [SL −_] [SOH][H+_][L2−_]exp

−c0F RT

n

= SL−_+H 2O (7) SOH+2H+_+L2− K_HL in − int ₌ [SHL] [SOH][H+_][L2−_] (8) = SHL+H2O a_{L denotes MoO} 4 2−_{, CrO} 4 2−_{, SO} 4 2−_{, SeO} 4 2−_{or SeO} 3 2−_.

(5)

Fig. 1. Adsorption of anions ontog-Al2O3as a function of pH under various background NaNO3concentrations. Symbols denote experiment data and line denotes the result of TLM simulation. (a) Simulation of SMoO4− using Eq. (7). (b) Simulation of SOH2 +_CrO 4 2 − + SOH2 +_HCrO 4

− _{using Eqs. (5) and (6). (c) Simulation of SOH} 2 +_SO

4

2 − _{using Eq. (5). (d) Simulation of} SOH2

+_SeO 4

2 − _{using Eq. (5). (e) Simulation of SSeO} 3 −

(6)

electrolyte strength. The background electrolyte concentration influences the double layer thick-ness and interface potential, thereby affecting the binding of the adsorbing species. Outer-sphere complexes are expected to be more susceptible to ionic-strength variations than inner-sphere com-plexes, since the background electrolyte ions are placed in the same plane as the outer-sphere complexes. Consequently, sorption of MoO42 −

and SeO32 − might imply the formation of

inner-sphere complexes. On the other hand, CrO42 −,

SO42 −, and SeO42 −are assumed to form the

outer-sphere complexes because their sorption is strongly dependent on background electrolytes. The complex forms of the anion/g-Al2O3 system

are verified by the following TLM simulation. The maximum adsorption densities of MoO₄2 − ,

CrO₄2 −_{, SO} 4 2 −_{, SeO} 4 2 −_{, and SeO} 3 2 − _{calculated in}

this study are 1.67 × 10− 6_{, 1.10 × 10}− 6_{, 1.42 ×}

10− 6_{, 1.32 × 10}− 6_{, and 1.50 × 10}− 6 _mol/m2_,

re-spectively. For comparison, the adsorption densities of these anions on goethite surface are higher, or 4.06 × 10− 6_{, 2 × 10}− 6_{, 2.46 × 10}− 6_,

1.59 × 10− 6_{, and 3.04 × 10}− 6 _mol/m2_,

respec-tively [4,5,8,9]. The variances in the adsorption densities for each anion might be due to differ-ences in laboratory conditions (e.g. adsorbate/ad-sorbent ratio) and adsorbate affinity for oxide surface.

During the course of a 24-h experiment as carried out by this study, g-Al₂O₃ may dissolve and recrystallize. Therefore, this study also car-ried out additional experiments to verify the dis-solution potential of g-Al₂O₃ under controlled conditions in each adsorption systems. The results indicate that the dissolved Al3 +_{is significant only}

in the CrO42 −/g-Al2O3 system. The ratio of

dis-solved Al3 +_{/add Al}3 +_{in MoO} 4 2 −_{, CrO} 4 2 −_{, SO} 4 2 −_,

SeO42 −, and SeO32 − systems is 0, 3.2, 0.15, 0.12,

and 0%, respectively. This result shows that the degree of dissolution of g-Al2O3 induced by the

adsorption of these anions is fairly insignificant. Stumm [25] illustrated that (complex-forming) lig-ands enhance the dissolution of Al2O3. Ligands

such as oxalate, salicylate, F−_{, EDTA, and NTA}

accelerate dissolution, while others, SO₄2 −_{, CrO} 4 2 −

, and benzoate, tend to inhibit dissolution. Krae-mer et al. [26] introduced d-Al₂O₃ dissolution in

the presence of arsenate and the organic ligand HQS, finding that HQS promotes oxide dissolu-tion and arsenate adsorbs to the d-Al₂O₃ surface but does not promote dissolution. Stumm et al. [27] showed that the most rapid dissolution rate is obtained when bidentate mononuclear surface chelates are formed and that monodentates are typically more inert with regard to dissolution reactions. Since the ligands in this study all form monodentate complexes and the focus of this study is not in dissolution, interactions between dissolution and adsorption were not performed.

3.2. TLM simulation

Species of MoO42 −, HCrO4−, CrO42 −, SO42 −,

SeO₄2 −_{, HSeO} 3

−_{, and SeO} 3

2 − _{are considered to}

react with g-Al₂O₃surface in the pH range evalu-ated according to the pK₂s listed in Table 2. Molybdate and selenite are hypothesized to form an inner-sphere complex. Outer-sphere complex formation of CrO42 −, SO42 −, and SeO42 − was

sim-ulated by Eq. (5) in Table 3. The results of TLM simulation for the anions/g-Al2O3systems are

pre-sented as the solid lines in Fig. 1. The fitness between the model prediction and experimental data indicates whether the assumption is accept-able or not. In this work, modeling sorption of molybdate, chromate, and selenite did not match the experimental data very well. The TLM was unable to coincide the molybdate sorption en-velop data in the entire pH range probably be-cause of the competition for reacting sites by OH−_.

In the CrO42 −/g-Al2O3 system, the model in

terms of the influence of ionic strength was unsat-isfactory in the acidic pHs (Fig. 1(b)). CrO42 −and

HCrO4− are both treated to react simultaneously

with the g-Al2O3surface under experimental

con-ditions. Model simulation using Eqs. (5) and (6) in Table 3 suggests that CrO42 − and HCrO4−form

outer-sphere complexes. The configuration of chromate on oxide surface seems to be the most controversy issue in the past studies. Mikami et al. [1] have reported that chromate complex is in outer-sphere configuration, whereas Grossl et al. [8] conclude that chromate sorption on goethite is an inner-sphere bidentate surface complex,

(7)

sug-gesting there may be a discrepancy in adsorption mechanism between an aluminum oxide and an iron oxide. The discrepancy on chromate surface complex can not be verified in this work; however, the results of the binary-solute experiments may infer that sorption of chromate onto g-Al₂O₃ is more likely to form outer-sphere complexes.

The selenite system is similar to chromate with two species (SeO32 −+ HSeO3−) reacting with the

g-Al2O3 surface under experimental conditions.

The TLM simulation for selenite is presented in Fig. 1(e). This result shows there is some dis-crepancy between the TLM simulation and exper-imental data. Since the background electrolyte concentration influences the sorption of SeO32 −

insignificantly, sorption of SeO32 −might imply the

formation of an inner-sphere complex. Hayes [11] and Zhang and Sparks [5], however, suggest that both HSeO₃− _{and SeO}

3

2 − _{form inner-sphere}

com-plexes on goethite surface.

3.3. Competiti6e adsorption

Competitive adsorption was examined for an-ions of similar affinity to the g-Al2O3 surface. As

mentioned above, adsorption can be categorized either as inner-sphere or outer-sphere adsorption; therefore, competitive adsorption experiments are examined for these two parts. The experiment considered not only the affinity to the oxide sur-face but also the similar adsorption pH edges for each anion. The systems of MoO₄2 −_{+ SeO}

3 2 −_, CrO₄2 −_{+ SO} 4 2 −_{, and CrO} 4 2 −_{+ SeO} 4 2 − _{were tested}

with results shown in Figs. 2 – 4, respectively. Competition between two anions for a given sur-face sorption site depends on the strength and type of the binding between the anion and the surface, and the binding rate on the surface.

In the binary-solute system of MoO42 − and

SeO32 −, the extent of anion sorption is both

de-pressed by each other in comparison to the single solute system. The decrease in sorption of selenite was more severer than that of molybdate at pHB 7. Sorption of selenite was, however, unaffected in the presence of co-anion in alkaline pHs. In sys-tems containing both strong adsorption anions, the competition seems to be dependent on the relative binding affinity of anions on oxide

sur-Fig. 2. Competitive adsorption of molybdate and selenite on g-Al2O3 as a function of pH (g-Al2O3= 30 g/l, [MoO42 −] = [SeO32 −] = 5 × 10− 3M).

face. At pHB6 where sorption of MoO₄2 − _is

preferred to that of SeO32 −, MoO42 − sorption is

able to out-compete SeO32 −. In the contrast, at

pH\6 where sorption of selenite is stronger than molybdate, molybdate sorption is somewhat in-hibited. The binary-anion systems of CrO42 −/

SO42 − and CrO42 −/SeO42 − exhibit similar results

on competitive adsorption. Chromate adsorption is significantly restrained in the presence of sulfate or selenate in acidic pHs where a 40% decrease in

Fig. 3. Competitive adsorption of chromate and sulfate on g-Al2O3 as a function of pH (g-Al2O3= 30 g/l, [CrO42 −] = [SO42 −] = 5 × 10− 3M).

(8)

Fig. 4. Competitive adsorption of chromate and selenate on g-Al2O3 as a function of pH (g-Al2O3= 30 g/l, [CrO42 −] = [SeO42 −] = 5 × 10− 3M).

CrO42 − retention in the subsurface environment is

suppressed in the presence of most anionic con-stituents in groundwater. This is similar to the results of this study for CrO₄2 −_{+ SO}

4 2 − _and

CrO₄2 −_{+ SeO} 4

2 − _{systems. From these}

experi-ments, the affinity or binding strength of MoO₄2 −

is stronger than SeO32 − and that SO42 − or SeO42 −

is stronger than CrO42 − at pHB6.

The overall proton coefficients (n) are calcu-lated from the slope of Kurbatov plots. The plots of log D versus pH for anion/g-Al2O3systems are

displayed in Fig. 5. The results showed that the steepest and flattest slopes are MoO42 − and SO42 −

/CrO42 −, respectively. The slopes for MoO42 −,

CrO42 −, SO42 −, SeO42 −, and SeO32 −are 0.68, 0.17,

0.17, 0.22, and 0.38, respectively. The order of overall proton coefficients given by the plots is MoO₄2 −_\SeO 3 2 −_\SeO 4 2 −_\CrO 4 2 −_SO 4 2 −

(Table 4). By comparing the overall proton coeffi-cients and the results of the competitive adsorp-tion experiments, it is suggested that a larger value of the overall proton coefficient corresponds to a higher affinity for the oxide surface.

4. Conclusions

Competitive adsorption of anions affects the partitioning and transport of anionic solutes in the subsurface and surface waters. Competition chromate adsorption is noticed. If one were to

consider the adsorption of chromate on oxide as an inner-sphere coordination, the out-competition by ion-pair sulfate or selenate sorption would be difficult to rationalize because of inner-sphere an-ion as a strong affinity adsorbate. By comparing the relative influence on competitive adsorption, it is suggested that sorption of chromate on oxide surface is more likely to form outer-sphere complexes.

The adsorption densities for single adsorbate and cosolutes are 1.67 × 10− 6_{(single) and 1.66 ×}

10− 6_mol/m2_(MoO 4 2 −_{+ SeO} 3 2 −_{) for MoO} 4 2 −_and

1.50 × 10− 6 _{(single) and 9.46 × 10}− 7 _mol/m2

(MoO₄2 −_{+ SeO} 3

2 −_{) for SeO} 3

2 −_{. As shown in Fig.}

3, the sorption densities are 1.10 × 10− 6 _(single)

and 4.85 × 10− 7 _mol/m2 _(CrO 4 2 −_{+ SO}

4 2 −_{) for}

CrO42 − and 1.42 × 10− 6 (single) and 1.14 × 10− 6

mol/m2_(CrO 4 2 −_{+ SO} 4 2 −_{) for SO} 4 2 −_{. In Fig. 4, the}

sorption densities for SeO42 −are 1.32 × 10− 6

(sin-gle) and 1.21 × 10− 6 _mol/m2 _(CrO 4

2 −_{+ SeO} 4 2 −₎

and for CrO42 − are 1.10 × 10− 6 (single) and

5.09 × 10− 7 _mol/m2 _(CrO 4

2 −_{+ SeO} 4

2 −_{). The}

re-sults clearly indicate that MoO42 − inhibits SeO32 −

adsorption and that SO42 − or SeO42 − depresses

CrO₄2 −_{. Zachara et al. [2] have indicated that}

inorganic SO₄2 − _{as well as CO}

2(g) and H4SiO4

distinctly reduces chromate adsorption and

(9)

Table 4

Surface complexation reactions, equilibrium constants, and overall proton coefficients for anion/g-Al2O3systemsa Anions Reactions in system log Keqint

SOH+H+_+MoO 4 2− _6.5 MoO42−(0.68) = SMoO4−+H2O SOH+H+_+CrO 4 2− CrO42−(0.17) 10.5 = SOH2+CrO42− SOH+2H+_+CrO 4 2− _14.5 = SOH2+HCrO4− SO42− (0.17) SOH+H++SO42− 10.4 = SOH2+SO42− SOH+H+_+SeO 4 2− SeO42−(0.22) 9.8 = SOH2+SeO42− SOH+H+_+SeO 3 2− SeO32−(0.38) 9 = SSeO3 − +H2O SOH+2H+ +SeO3 2− ₅ = SHSeO3+H2O

a_{The number in parentheses is the overall proton coefficient} and Keqintis the intrinsic equilibrium constant.

References

[1] N. Mikami, M. Sasaki, T. Klkuchi, T. Yasunaga, J. Phys. Chem. 87 (1993) 5245.

[2] J.M. Zachara, D.C. Girvin, R.L. Schmidt, C.T. Resch, Environ. Sci. Technol. 21 (1987) 589.

[3] C.C. Ainsworth, D.C. Girvin, J.M. Zachara, S.C. Smith, Soil Sci. Soc. Am. J. 53 (1989) 411.

[4] P.C. Zhang, D.L. Sparks, Soil Sci. Soc. Am. J. 54 (1990) 1266.

[5] P.C. Zhang, D.L. Sparks, Environ. Sci. Technol. 24 (1990) 1848.

[6] K. Mesuere, W. Fish, Environ. Sci. Technol. 26 (1992) 2365.

[7] J.A. Wilkie, J.G. Hering, Colloids Surf. A 107 (1996) 97. [8] P.R. Grossl, M. Eick, D.L. Sparks, S. Goldberg, C.C.

Ainsworth, Environ. Sci. Technol. 31 (1997) 321. [9] P.C. Zhang, D.L. Sparks, Soil Sci. Soc. Am. J. 53 (1989)

1028.

[10] L.M. He, L.W. Zelazny, V.C. Baligar, K.D. Ritchey, D.C. Martens, Soil Sci. Soc. Am. J. 61 (1997) 784.

[11] K.F. Hayes, Equilibrium, Spectroscopic, and Kinetic Stud-ies of Ion Adsorption at the Oxide/Aqueous Interface, Ph.D. Dissertation, Stanford University, Stanford, CA, 1987.

[12] S. Fendorf, M.J. Eick, P. Grossl, D.L. Sparks, Environ. Sci. Technol. 31 (1997) 315.

[13] W.R. Roy, J.J. Hassett, R.A. Griffin, Soil Sci. Soc. Am. J. 50 (1986) 1176.

[14] D. Hawke, P.D. Carpenter, K.A. Hunter, Environ. Sci. Technol. 23 (1989) 187.

[15] N.S. Bolan, J.K. Syers, M.E. Sumner, Soil Sci. Soc. Am. J. 57 (1993) 691.

[16] J.J. Dynes, P.M. Huang, Soil Sci. Soc. Am. J. 61 (1997) 772. [17] L. Chatelet, J.Y. Bottero, J. Yvon, A. Bouchelaghem,

Colloids Surf. A 111 (1996) 167.

[18] J.A. Davis, J.O. Leckie, J. Colloid Interface Sci. 67 (1978) 90.

[19] K.F. Hayes, J.O. Leckie, J. Colloid Interface Sci. 115 (1987) 564.

[20] J.B. Peri, J. Phys. Chem. 69 (1965) 211.

[21] K.F. Hayes, G. Redden, W. Ela, J.O. Leckie, J. Colloid Interface Sci. 142 (1991) 448.

[22] J.A. Davis, R.O. James, J.O. Leckie, J. Colloid Interface Sci. 63 (1978) 480.

[23] M. Hohl, W. Stumm, J. Colloid Interface Sci. 55 (1976) 281. [24] D.D. Perrin, Ionisation Constants of Inorganic Acids and Bases in Aqueous Solution, Pergaman Press, New York, 1982.

[25] W. Stumm, Chemistry of the Solid – Water Interface, Wiley, New York, 1992.

[26] S.M. Kraemer, V.Q. Chiu, J.G. Hering, Environ. Sci. Technol. 32 (1998) 2876.

[27] W. Stumm, G. Furrer, E. Wieland, B. Zinder, in: J.I. Drever (Ed.), The Chemistry of Weathering, Reidel, Dordrecht, The Netherlands, 1985, p. 55.

-between anions with strongly binding affinity to form inner-sphere complexes is largely dependent on the relative sorption capabilities in the specific conditions. Molybdate and selenite are both strongly binding adsorbates. At comparable so-lute concentration, molybdate depresses selenite sorption at acidic environment and selenite sup-presses molybdate sorption at alkaline conditions. For intermediately binding anions as chromate, selenate, and sulfate, adsorption of chromate is significantly inhibited in the presence of sulfate or selenate. The relative retainment of anions on oxide surface is molybdate\selenite\selenate sulfate\chromate.

Acknowledgements

The authors thank Professor Olive J. Hao, Uni-versity of Maryland, for the valuable discussions and the reviewers for their constructive com-ments. This work is partially funded by the Na-tional Science Council of the Republic of China under Grant No. NSC 83-0410-E-002-082.

Competitive adsorption of molybdate, chromate, sulfate, selenate, and selenite on γ-Al2O3