Eu(III) uptake on rectorite in the presence of humic acid: A macroscopic and spectroscopic study

Journal of Colloid and Interface Science 393 (2013) 249–256

Contents lists available at SciVerse ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

Eu(III) uptake on rectorite in the presence of humic acid: A macroscopic and spectroscopic study Changlun Chen ⇑, Xin Yang, Juan Wei, Xiaoli Tan, Xiangke Wang ⇑ Key Laboratory of Novel Thin Film Solar Cells, Institute of Plasma Physics, Chinese Academy of Sciences, P.O. Box 1126, Hefei 230031, PR China

a r t i c l e

i n f o

Article history: Received 16 July 2012 Accepted 12 October 2012 Available online 24 October 2012 Keywords: Eu(III) Humic acid Rectorite Modelling EXAFS

a b s t r a c t This work contributed to the comprehension of humic acid (HA) effect on Eu(III) uptake to Na-rectorite by batch sorption experiments, model fitting, scanning electron microscopy, powder X-ray diffraction, Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy, and extended X-ray absorption fine structure (EXAFS) spectroscopy. At low pH, the presence of HA enhanced Eu(III) sorption on Narectorite, while reduced Eu(III) sorption at high pH. The experimental data of Eu(III) sorption in the absence and presence of HA were simulated by the diffuse-layer model well with the aid of FITEQL 3.2 software. The basal spacing of rectorite became large after Eu(III) and HA sorption on Na-rectorite. Some of Eu(III) ions and HA might be intercalated into the interlayer space of Na-rectorite. EXAFS analysis showed that the REuAO (the bond distance of Eu and O in the first shell of Eu) and N values (coordination number) of Eu(III)–HA-rectorite system were smaller than those of Eu(III)-rectorite system. Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction With the development of nuclear energy and proceeding of various nuclear processes, a large amount of radionuclides were discharged into aquatic systems. The migration and transfer behavior of radionuclides in the environment depends on the prevailing geochemical conditions and processes influencing the speciation of radionuclides. Sorption on clay minerals strongly influences the fate and mobility of radioactive contaminants in the geosphere. Therefore, understanding the migration behavior of radioactive and non-radioactive toxic substances is essential for a reliable long-term safety assessment of potential nuclear waste disposal sites, and of subsurface dumps and sites with radioactive and/or heavy metal containing inventory. Clay minerals are among the most important sorbents for metal cations in soils and sediments due to the high abundance of clays, their large specific surface area, negative surface charge, and reactive surface hydroxyl groups [1,2]. Humic acid (HA) is a chemically heterogeneous compound having different types of functional groups at different proportions and configurations [3]. HA is organic matter that is defined as the fraction of humic substances soluble at pH values >1. HA solubility increases with increasing pH and decreases with increasing inert electrolyte concentration [4]. HA contains carboxyl (ACOOH), hydroxyl (AOH), amine (ANH2), and phenol (ArAOH) functional groups, and has negative charges in weakly acidic-to-basic media

⇑ Corresponding authors. Fax: +86 551 5591310. E-mail addresses: [email protected] (C. Chen), [email protected] (X. Wang). 0021-9797/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2012.10.032

due to deprotonation reactions [5]. It is very important to know the nature of organic functional groups of HA in the process of determining the stability of metal ion complexes. HA may enhance or reduce metal ion sorption, depending on the relative stabilities of metal–HA binary and metal–HA-clay surface ternary complexes as a function of pH [6]. Rectorite is an abundant clay mineral in China. Structure and characteristics of it are similar to those of montmorillonite [7]. It is a kind of regularly interstratified clay mineral with alternate pairs of dioctahedral mica-like layer (nonexpansible) and dioctahdral smectite-like layer (expansible) existing in 1:1 ratio. The cations of Na+, K+ and Ca2+ lie in the interlayer region between 2:1 mica-like layers and 2:1 smectite-like layers [8], while the exchangeable hydrated cations reside in the latter. The structure of rectorite can also cleave easily between smectite-like interlayers, forming monolithic rectorite layers (2 nm thick). Because of the presence of smectite layer in rectorite, it is anticipated that surface physicochemical properties of rectorite would be similar to that of smectite. Properties such as a strong sorption capacity, good stability in water and an easy regeneration has led to this material being tested as a sorbent of many cations and anions from the environment [9–13]. However, research on metal ion sorption to rectorite in the presence of HA, which represents a major fraction of dissolved organic compounds present in freshwater, is still scarce [12,13]. A better macroscopic and spectroscopic description of metal ion, HA, and rectorite interactions is still needed to improve the understanding of the behavior of metal ions and HA in the environment. Removal of long-lived radionuclides from nuclear waste solutions is an important environmental concern in nuclear waste management. In this work, the structures and species of surface

250

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

sorbed/complexed Eu(III) on rectorite in the absence and presence of HA were characterized by using powder X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), and extended X-ray absorption fine structure (EXAFS) spectroscopy. The effect of pH on the sorption of Eu(III) and HA to Na-rectorite was investigated, and the experimental data of Eu(III) sorption in the absence and presence of HA were simulated by the diffuse-layer model (DLM) with the aid of FITEQL 3.2 software. Europium was selected as both a fission product and a homologue of trivalent lanthanides and actinides [14]. Measurements were made, at low metal ion concentration, to get as close as possible to environmental concentration. This paper highlights the potential environmental application of rectorite in the removal of Eu(III) in the presence of HA in environmental pollution cleanup. 2. Experimental 2.1. Materials All reagents were of analytical reagent grade and used without further treatment. Milli-Q water was used in all experiments. Eu (III) stock solution at 0.1 mol/L was prepared from Eu2O3 (Purity, 99.99%) after dissolution, evaporation, and redissolution in 0.001 mol/L perchloric acid. The radiotracer 152+154Eu(III) was used in the batch sorption experiments. The Na-rectorite sample was derived from Zhongxiang county (Hubei, China). The characterization results of Na-rectorite were reported in earlier reports [13]. In brief, the N2-BET surface area was 11.9 m2/g, and the average particle size was 7.4 lm. The water content of 5.7% was taken into account in the calculation of the Na-rectorite content. HA was extracted from a soil from Gansu province (China) and was characterized by nuclear magnetic resonance and pyrolysiscapillary gas chromatography electron impact mass spectrometry. HA did not contain any inorganic constituents. The main constituents were the following: C, 60.44%; H, 3.53%; N, 4.22%; O, 31.31%; and S, 0.50% [15]. 2.2. Characterization The morphology of samples was obtained with scanning electron microscope (SEM) (JSM-6700F). Selected samples doped with Eu(III) were characterized using FTIR (Perkin Elmer spectrum 100, America) in pressed KBr pellets. For FTIR spectroscopy analysis, the samples were washed 3 times with ethanol for 5 min, filtered, and dried at 343 K for 24 h. XRD measurements were performed by D/Max-2400 Rigaku Xray powder diffractometer operated in the reflection mode with Cu Ka (k = 0.15418 nm) radiation. XPS data were obtained with a Thermo ESCALAB 250 electron spectrometer from VG Scientific using 150 W Al Ka radiation. EXAFS spectra were measured at beamline U7C of the National Synchrotron Radiation Laboratory (Hefei, China). After sorption experiments, Na-rectorite loading Eu(III) and/or HA was for SEM, FTIR, and XRD measurements. Samples preparation for XPS and EXAFS analysis were conducted using a 500 mL vessel with 0.2 g/L Na-rectorite, 0.01 mol/L NaClO4, and 0.2 lmol/L Eu(III) at pH = 5.0. For HA effect, Na-rectorite and NaClO4 were pre-equilibrated for 24 h, then HA was added to equilibrate for 24 h, and finally, Eu(III) stock solution was added. Detailed processes for the sample preparation for XPS and EXAFS analysis are shown in Supporting information SI 1. The acid–base titrations of Na-rectorite and HA suspension were carried out by potentiometric titration experiments [16– 18], which was conducted in a 100 mL Teflon vessel with a polyethylene lid. The vessel was surrounded with a glass jacket to

maintain a temperature of 25.0 ± 0.5 °C. A Teflon bar was used for stirring. All titrations were conducted using a computer controlled PC-titration system (DL50 Automatic Titrator, Mettler Toledo) with pH electrode (Delta 320). Argon was bubbled successively through NaOH, HClO4, and Milli-Q water to exclude CO2(g). Before beginning the titrations, Na-rectorite or HA suspension and background electrolyte were added to the vessel and purged with argon for at least 2 h. The pH was quickly lowered to approximately 3.0 by addition of 1.5721M HClO4. The high concentration of HClO4 could acidify the suspension quickly without obvious changing of solution volume. After 1 h of equilibrium, the suspensions were slowly back-titrated at a variable increment (0.008 up to 0.15 mL, which was automatically adjusted to keep a stable pH change value) with 0.04668M NaOH solution to pH 11. Each step was allowed to stabilize until the pH drift was less than 0.005 pH unit per minute. The low concentration of NaOH assured the gentle change of pH after each titration point. The data sets of pH versus the net consumption of H+ or OH in the absence of metal ions due to surface complexation were used to obtain intrinsic acidity constants with the aid of FITEQL 3.2 software. 2.3. Sorption experiments The sorption of Eu(III) on Na-rectorite was investigated by using batch sorption experiments in polyethylene centrifuge tubes sealed with a screw-cap under N2 condition. The stock suspension of Na-rectorite and NaClO4 solution were pre-equilibrated for 24 h, and then, Eu(III) stock solution was added to achieve the desired concentrations of the different components. The system was adjusted to the desired pH by adding negligible volumes of 0.01 or 0.1 mol/L HClO4 or NaOH. For HA adsorption at different pH values, Na-rectorite and NaClO4 were pre-equilibrated for 24 h, then HA was added, and the pH was adjusted. Samples were gently shaken for 24 h, and then centrifuged for 30 min at 18,000 rpm (Allegra 64R Centrifuge, Beckman Coulter) for the separation of solid phase from aqueous phase, pH was measured, and the concentrations of HA in the supernatants were determined using a UV–vis spectrophotometer (UV-2550, Shimadzu, Kyoto, Japan) at 259.6 nm (calibration at different pH values, and consideration of the contribution of ions leached out from Na-rectorite to the UV–vis signal). For the effect of HA on Eu(III) sorption, Na-rectorite and NaClO4 were pre-equilibrated for 24 h, then HA was added to equilibrate for 24 h, and finally, Eu(III) stock solution was added to achieve the desired concentrations of different components. Preliminary studies indicated that the sorption of Eu(III) on Na-rectorite achieved equilibrium in several hours. Samples were gently shaken for 24 h to obtain complete sorption equilibrium and then centrifuged for 30 min at 18,000 rpm for the separation of solid phase from aqueous phase. The efficiency of separation was checked by photo-correlation spectroscopy, which showed no colloidal particles in the supernatant solution. The counting of 152+154Eu(III) was analyzed by Liquid Scintillation counting using a Packard 3100 TR/AB Liquid Scintillation analyzer (PerkinElmer) with the scintillation cocktail (ULTIMA GOLD AB™, Packard). The sorption percentage R was calculated according to the equation of R(%) = 100  (1  AL/Atot); herein, Atot is the activity of the suspension and AL is that of the supernatant. 3. Results and discussion 3.1. Batch experiments Fig. 1A shows that HA adsorption on Na-rectorite decreases with increasing pH. HA adsorption is strongly dependent on

251

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

(A)

S+W

40

S+W

=S

No HA 10 mg/L HA

80

60

=S

100

Eu sorption %

HA Adsorption %

(B)

Experiment of adsorption Model of adsorption

80

OHHL1

OHHL3

60 40 20

20

0

0 2

3

4

5

6

7

8

9

2

10

3

4

5

0.8 Exp. data DLM

0.6

+

s

S OEuOH

0.4

s

2+

S OEu X3Eu

0.2

7

8

0.0

0.8

Eu(III) sorbed on rectorite µmol/g

(D)

1.0

Eu(III) sorbed on rectorite µmol/g

(C)

6

pH

pH

Exp. data DLM 0.6 +

S

S OL1Eu 0.4 S

2+

S OHL1Eu

+

S

S OL3Eu

S

S OL1EuOH

0.2 2+ SSOL EuOH 3

S

S OHL3Eu 0.0

3

4

5

6

7

pH

2

3

4

5

6

7

pH

Fig. 1. HA adsorption on Na-rectorite and fitting, and (B) Eu(III) sorption on Na-rectorite. The fitting of Eu(III) sorption in the absence of HA (C) and the presence of HA (D), as a functional of pH, at m/V (rectorite) = 0.2 g/L, C[Eu(III)]initial = 0.2 lmol/L, C(HA)initial = 10 mg/L, C(NaClO4) = 0.01 mol/L, T = 25 ± 1 °C.

electrostatic parameters such as sorbent surface charges, being dependent on pH. However, HA adsorbed on an oppositely charged surface experiences not only electrostatic attraction to the surfaces but also electrostatic repulsion within HA itself. At high pH values, the latter effect occurs when the ionization of carboxylic groups increases repulsion in an HA segment, which leads to an increase in the HA length. Because HA molecules are large, only a fraction of HA molecules may participate in the formation of complexes with surface sites. The high percentage adsorption of HA at low pH can be attributed to a highly ‘‘coiled’’ HA conformation as a result of low charge development [19,20]. This results in the adsorption of a large number of HA molecules because each molecule occupies a smaller area. At high pH, however, HA is more ‘‘expanded’’ [20] as a result of electrostatic repulsion, so the area occupied by each molecule will be higher. A lower adsorption percentage is therefore obtained at high pH values. From Fig. S1B, at pH < 4.2, the surfaces of Na-rectorite are positively charged, and at pH > 4.2, the surfaces of Na-rectorite are negatively charged. The point of zero change (pHpzc) of GO/Fe3O4 is 4.2. At low pH, Na-rectorite is protonated and becomes more positively charged. HA molecules are still negatively charged because the point of zero f potential is about pH 2.0 [21], and so Na-rectorite and HA can interact via electrostatic attraction, thereby enhancing the adsorption intensity. As the pH increases, the weakly acidic HA, which has carboxylic and phenol moieties, becomes more negatively charged. At high pH, Na-rectorite will be deprotonated and becomes more negatively charged. Thus, at higher pH, repulsion between HA and the adsorbent surfaces would be stronger, and thereby hindering the adsorption of HA on Na-rectorite. Fig. 1B shows Eu(III) sorption on Na-rectorite as a function of pH in the absence and presence of HA in I = 0.01 mol/L NaClO4 solu-

tion. Eu(III) sorption is strongly dependent on pH. Eu(III) sorption on Na-rectorite in the absence of HA increases very quickly from about 2% to 98% in the pH range from 2 to 6, remains high level as the pH increases. Changes in the solution pH can affect the distribution of Eu(III) species in solution and the surface charges of rectorite by dissociation of functional groups, promoting or suppressing the sorption of metal ions on sorbent surfaces. The hydrolysis constants (log b1 = 7.2, log b2 = 15.1, and log b3 = 26.2) (Fig. S3) [22] of Eu(III) suggest that Eu3+ ion is the only significant species at pH < 5, and Eu(OH)2+ begins to form at pH > 5. EuðOHÞþ 2 is present at about 5% and is less than 30% at pH = 7.5. The Eu(OH)3 species begins to form at pH > 6.5. Therefore, at the studied pH range from 2 to 7.5, EuðOHÞþ 2 and Eu(OH)3 species can be negligible. At pH < 5, the main species is Eu3+ and the removal of Eu3+ is mainly accomplished by ion exchange. At pH > 6.5, Eu(III) removal reaches a maximum and maintains a high level. The main species present in the pH range 6–7.5 is Eu(OH)2+, which can easily be sorbed on the negatively charged Na-rectorite surfaces. In the presence of HA, Eu(III) sorption shows very different trend. Eu(III) sorption on Na-rectorite is enhanced at pH < 5.5, and the maximum sorption edge is 1.5 unit lower (i.e., it shifts from 6.0 to 4.5) than that for Eu(III) sorption on Na-rectorite in the absence of HA. However, HA presence reduces Eu(III) sorption at pH > 5.5. HA has a macromolecular structure, so only a small fraction groups of the ‘‘adsorbed’’ HA is free to interact with Eu(III) ions [23]. After the adsorption of negatively charged HA on Narectorite, the surfaces of Na-rectorite become more negatively charged than bare Na-rectorite surfaces. This results in a more favorable electrostatic attraction for Eu(III) and enhances the formation of Eu(III)–HA-rectorite surface complexes. At low pH values, the negatively charged HA can be easily adsorbed on the

252

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

positively charged Na-rectorite, so the complexation of surfaceadsorbed HA and Eu(III) results in the enhancement of Eu(III) sorption on Na-rectorite. As shown in Fig. 1A, HA adsorption on Na-rectorite decreases with increasing pH. At high pH values, there are more free HA molecules available in the solution. The free HA molecules interacted with Eu(III) can form strong and soluble HA–Eu complexes [24] and can stabilize Eu(III) in solution, resulting in an overall decrease in Eu(III) sorption. The fact that fewer HA molecules are adsorbed on Na-rectorite (i.e., there are more HA molecules in solution) indicates that there is less complexation of Eu(III) with adsorbed HA on the surfaces of Na-rectorite. Eu(III) sorption therefore decreases with increasing pH in the presence of HA. The laser-introduced fluorescence spectroscopy analysis clearly showed that Cm(III) was adsorbed as Cm(III)-fulvate complex in the FA-montmorillonite hybrid at pH below 5 [25]. In our earlier work [13,26], at pH < 4, the presence of HA increased the sorption of Th(IV) on rectorite, and no drastic effect of HA on the sorption of Th(IV) at pH > 4 was observed. At pH < 8, the presence of FA/HA enhanced Ni(II) uptake to FA/HA-rectorite hybrids. At pH > 8, little effect of FA/HA on Ni(II) sorption was observed. Generally, the presence of humic substances enhances metal ion sorption at low pH values and decreases metal ion sorption at high pH values [27,28]. 3.2. Modelling A strong and weak site protolysis model with no electrostatic term was applied to describe the protonation and deprotonation of the amphoteric surface and hydroxyl groups (sites „SOH) situated at the platelet edges on Na-rectorite. The surface hydrolysis S W constants (K SðþÞ and K W ðþÞ , K ðÞ and K ðÞ , and KXH), the total densities of surface complexation sites ðT SBS OH and T W B Sn Þ, and exchange capacities (T„XNa/H) can be optimized to simulate the titration data with the aid of FITEQL 3.2 software. The relative surface parameters are listed in Table 1. Acid–base titration data of HA are shown in Fig. S2. The acidity constants pKa of HL1 (carboxyl groups), HL2 (hydroxyl groups), and HL3 (phenol groups) were 5.23, 7.24, and 9.57, respectively [13]. Fig. 1A shows that the adsorption of HA on Na-rectorite is well simulated by the constant capacitance model with the aid of FITQEL 3.2. The possible adsorption mechanism can be described as [13,29]:

BSOHsþw þ HL1 ! BSOHsþw HL1

ð1Þ

BSOHsþw þ HL3 ! BSOHsþw HL3

ð2Þ

In the pH range of our experiment, Eu(III) sorption data in the absence of HA was fitted using the DLM with the aid of FITEQL 3.2 software (Fig. 1C). In the modeling of Eu(III) sorption, since Eu (III) sorption was measured at the range of about trace concentration, Eu(III) sorption was modeled as occurring on strong type sites („SsOH) only, according to the reference [30]. According to analysis of the distribution of aqueous Eu(III) species mentioned above, at the studied pH range from 2 to 7.5, EuðOHÞþ 2 and Eu(OH)3 species can be negligible. At the same time, considering the distribution of the surface complex and exchange sites, at pH < 7.5, „SSO and „XNa/H exist on Na-rectorite surfaces at the studied pH range. Therefore, the complexes „SSOEu2+, „X3Eu, and „SSOEuOH+ are chosen. The obtained results are listed in Table 2, and the experimental data and fit curves and surface

Table 2 log K Values for Eu(III)-rectorite optimized in the absence and presence of HA. Sorption model

log K

WSOS/DF

0.95 6.51 13.46

0.3

BSs OHHL1 þ Eu3þ ! BSs OHL1 Eu2þ

15.34

0.8

BSs OHHL3 þ Eu3þ ! BSs OHL3 Eu2þ

16.62

BSs OL1 þ Eu3þ ! BSs OL1 Euþ

5.00

BSs OL3 þ Eu3þ ! BSs OL3 Euþ

6.34

BSs OL1 þ EuðOHÞ2þ ! BSs OL1 EuOH

6.82

BSs OL3 þ EuðOHÞ2þ ! BSs OL3 EuOH

8.11

In the absence of HA „ SsO + Eu3+ ? „ SsOEu2+ „ SsO + Eu3+ + H2O ? „ SsOEuOH+ + H+ 3BXNa þ Eu3þ ! BX3 Eu3þ þ 3Naþ In the presence of HA

specie distributions are shown in Fig. 1C. The distributions of the surface species „SSOEu3+, „X3Eu, and „SSOEuOH+ are in good agreement with those of the surface sites „XNa/H and „SSO as a function of pH. Bradbury et al. [22] reported that for Na/Camontmorillonite and Na-illite, the Eu(III) sorption edges could be quantitatively modeled in the pH range 3 to 10 using cation exchange reactions for Eu3+/Na+ and Eu3+/Ca2+ and three surface complexation reactions on the strong sorption sites forming „SSOEu3+, „SSOEuOH+, and BSS OEuðOHÞ02 inner-sphere complexes. In the presence of HA, it is difficult to fit metal ion sorption on clay minerals due to the complicate effect of HA. Presently, report on the sorption model of metal ions in the presence of HA is scarce. From Eu(III) species distribution in the presence of HA (Fig. S3), at pH < 4.0, Eu3+(HA) is the dominant species, at 4.0 < pH < 6.0, Eu3+ (HA) and Eu(OH)2+(HA) are the dominant species, and at pH > 6.0, Eu(OH)2+(HA) is the dominant species. Eu(III) sorption data in the presence of HA were also fitted using the DLM with the aid of FITEQL 3.2 software. The obtained results are listed in Table 2, and the experimental data and fit curves and surface specie distributions are shown in Fig. 1D. From Fig. 1D, at pH < 4.0, surface species „SSOHL1Eu2+, „SSOHL3Eu2+, „SSOL1Eu+, and „S S OL 3 Eu + are the dominant species. However, at 4.0 < pH < 6.0, surface species „SSOL1Eu+, and „SSOL3Eu+ are the dominant species. At pH > 6.0, surface species „SSOL1EuOH and „SSOL3EuOH are the dominant species. The fitting results are the combination of „SSOHL1Eu2+, „SSOHL3Eu2+, „SSOL1Eu+, „SSOL3Eu+, „SSOL1EuOH, and „SSOL3EuOH, which are in line with Eu(III) species distribution in the presence of HA (Fig. S3). 3.3. SEM images The SEM micrographs (Fig. 2) were used to probe the changes in morphological features of Na-rectorite before and after Eu(III) sorption in the absence and presence of HA. From Fig. 2A and B, the complexation of Eu(III) with functional groups of HA can easily make HA form aggregates. From Fig. 2C and D, there is no obvious morphological feature change of Na-rectorite observed before and after Eu(III) sorption due to less Eu(III) content. The microstructures of Eu(III)-rectorite (Fig. 2D), HA-rectorite (Fig. 2E), and Eu (III)–HA-rectorite (Fig. 2F) are quite different. HA and HA–Eu(III) are formed aggregates on the surfaces of Na-rectorite. The sizes of Eu(III)–HA-rectorite are smaller than those of HA-rectorite. HA adsorbed on the charged surfaces of rectorite experiences not only

Table 1 Parameters for modeling the acid–base surface chemistry of Na-rectorite. log K SðþÞ

log K SðÞ

log K W ðÞ

log K(XH)

Site density „SSOH (mol/g)

Site density „SWOH (mol/g)

Exchange capacity „XNa/H (mol/g)

3.88

5.69

0.97

3.76

1.23  104

2.65  104

0.95  104

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

253

Fig. 2. SEM images of (A) HA; (B) HA–Eu(III) (denoted as HA loading Eu(III)); (C) rectorite; (D) Eu(III)-rectorite (denoted as rectorite loading Eu(III)); (E) HA-rectorite (denoted as rectorite loading HA); (F) Eu(III)–HA-rectorite (denoted as rectorite loading HA and Eu(III)), at pH = 5.0.

electrostatic attraction to the surfaces but also electrostatic repulsion within HA itself. HA is more ‘‘expanded’’, as a result of electrostatic repulsion. HA is less ‘‘expanded’’ because the electrostatic repulsion is decreased after HA is bonded with Eu(III). Although SEM images do not show the structures at a large scale, the local structures of SEM images are still essential to observe the changes of the colloid particles. 3.4. XRD patterns The XRD patterns (Fig. 3) indicate that the Na-rectorite sample is composed of NaAl silicate hydroxide hydrate (NaAl4[Si, Al]8O20 (OH)4nH2O) (JCPDS card of 250781) [31] (line (a) in Fig. 3). The XRD peak intensities, peak shapes, and peak positions reflect the crystallinity extent of the sample. The patterns of Eu(III)-sorbed rectorite (lines (c and d) in Fig. 3) show poor crystallinity with broad, less intense peaks compared to Na-rectorite due to the sorption of Eu(III) and HA stacking on the surface, the edge or the interlayer of Na-rectorite. The basal spacing of each sample was calculated using Bragg’s law: 2d sin h = nk, where d is the basal spacing (Å), h is the angle of diffraction (°), k is the wavelength (nm), and n is the path differences between the reflected waves, which equal an integral number of wavelengths (k). The sizes of basal spacing are 23.9 Å for Na-rectorite, 24.5 Å for HA-rectorite hybrids, and 24.3 and 26.3 Å for Eu(III) sorption on Na-rectorite

in the absence and presence of HA, respectively. After Eu(III) sorption on Na-rectorite in the absence and presence of HA, the basal spacing of rectorite becomes large. Some of Eu(III) ions and HA may be intercalated into the interlayer space of Na-rectorite. After Ni(II) sorption on FA/HA bound Na-rectorite, the position of maximum peak moved further to the right, indicating that the increased distance between the tetrahedral sheets [26]. After Eu(III) sorption on Na-rectorite, no other reflection peak is observed in the XRD patterns of samples (b–d), and almost no change in the crystalline structure of the solid is detected. Eu(III) is expected to be able to penetrate the interlayer due to surface complexation and ion-exchange between Eu(III) ions and Na+ ions. However, this does not lead to the matrix of Na-rectorite undergoing severe structural change. Other researchers have indicated that the structures of kaolinite were not affected by Sr(II) sorption, and the interlayer positions of the kaolinite were virtually not involved in Sr(II) fixation due to the relatively stronger binding forces between the kaolinite sheets, where outer-layer sites were more effective in sorption than inter-layer sites [32]. 3.5. FTIR spectroscopy The FTIR spectra of the samples are shown in Fig. 4. The broad bands at 3642 and 3430 cm1 are due to the OAH stretching

254

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

2000

* 3.70

1500

*

3.65 7.86 28.52 17.57 19.97

(a)

* 3.64

1000

(b)

* 3.35 500

(c) (d)

0 10

20

30

40

2 Theta (degree) Fig. 3. XRD patterns of samples (a) rectorite, (b) HA-rectorite, (c) Eu(III)-rectorite, and (d) Eu(III)–HA-rectorite.

(d) (c)

4000

3500

3000

2500

2000

1500

700 1000

553 481

1030

927

3430

1640

1436

(a)

3642

Tansmittance

(b)

500

The XPS spectrum of Eu(III)-rectorite shows a C1s signal at 284.6 eV, indicating the occurrence of carbon in the sample. This signal is due to a sorbed Eu(III) carbonate species and minor amounts of carbon in the untreated rectorite sample. The peak of C 1s is enhanced obviously in the ternary Eu(III)–HA-rectorite system, suggesting that many functional groups are introduced to the surfaces of rectorite [33,34]. The C 1s spectrum of Eu(III)–HA-rectorite can be well-fitted by four Gaussian–Lorentzian functions, being assigned to CAC (charge referenced to 284.6), CxHy (285.5 eV), C@O (270.0 eV), and COO (289.2 eV) [35,36]. The last three carbon groups are due to the attachment of HA groups on the surfaces of rectorite. These carbon-containing functional groups are abundant on the surfaces of rectorite, provide numerous sorption sites, and thus increase Eu(III) sorption. Krepelová et al. [37] studied the HA sorption onto kaolinite by XPS. From analysis of C 1s/Al 2p intensity ratios, it was concluded that the HA-kaolinite sample surface contains only a small amount of carbon independent from the loading of the HA samples. Furthermore, it was concluded that the clay particles cannot be covered by a homogeneous layer of HA. A part of HA must be distributed between the kaolinite particles. The HA-rectorite sample surfaces contain approximately 7.5 at.% carbon. It is deduced that the surface of the rectorite particles cannot be covered by a homogeneous layer of HA. Part of the HA was distributed between the particles. This implies that significant parts of the rectorite surface that are not covered by HA are accessible for Eu(III) during its sorption. The binding energy of Eu3d5/2 increases from 1135.1 eV on rectorite to 1135.8 eV on HA–rectorite hybrids. The main peak of Eu3d5/2 changes about 0.5 eV, which indicates that HA has changed the mechanism or species of Eu(III) sorption on rectorite. The Eu3d5/2 core level region spectrum is fitted by deconvolution. In binary Eu(III)-rectorite system, two peaks at 1125.8 and 1135.3 eV are achieved; however, three peaks at 1126.0, 1135.8, and 1137.5 eV are used to fit the Eu3d5/2 spectrum well in ternary Eu(III)–HA-rectorite system. The peak at 1137.5 eV corresponds to O@CAOAEuAOA, which is due to the functional groups of HA [38,39].

-1

Wavenumber (cm ) 3.7. EXAFS analysis Fig. 4. FTIR spectra of samples (a) rectorite, (b) HA-rectorite, (c) Eu(III)-rectorite, and (d) Eu(III)–HA-rectorite.

vibration of the silanol (SiAOH) groups from the solid and HOAH vibration of the water molecules adsorbed on the surfaces of samples. The spectral bands at 1640 cm1 and 1436 cm1 reflects the stretching t (OH) and bending d (OH) vibration of water molecules being retained in the silica matrix. The strong band at 1030 cm1 represents the SiAOASi groups of the tetrahedral sheets. The spectral band at 927 cm1 reflects the stretching vibration of AlAOA(OH)AAl. The bands at 700, 553, and 481 cm1 are due to the deformation and bending modes of the SiAO bond [32,26]. No obvious changes of spectral band intensities, shapes, and positions were observed after Eu(III) and HA uptake on rectorite surfaces. The presence of HA may change the spectral band in the spectroscopic analysis, but its lower concentration leads to the signal being indiscernible. It is difficult to detect the vibrational changes of edge sites after the loading of Eu(III) onto rectorite. 3.6. XPS analysis XPS spectra demonstrate the sensitivity for identifying elements on the local surface. Fig. 5 shows the XPS spectra of Eu(III) sorbed on bare rectorite and HA-rectorite at pH 5.0 in 0.01 mol/L NaClO4. The sorbed Eu(III) can be identified by the Eu 3d XPS lines in the XPS survey spectra of Eu(III)-rectorite and Eu(III)–HA-rectorite.

The EXAFS technique has been applied to study the reactions between Eu and the environmental atoms, ions, molecules, or compounds [30]. The Eu–LIII EXAFS is sensitive to the first shell coordination environment and is also changeable with the changing of the chemical structure and valence state. Fig. 6A shows the Fourier transforms of the reference and sorption samples. The curve-fitting analysis are performed using the EXAFS equation [40] with the amplitude reduction factor S20 set at 1.0 to correctly reproduce the number of neighboring atoms [41]. From Fig. 6A, EXAFS spectra obtained for Eu(III) sorption samples look similar to each other, but different somewhat from those of the reference samples. The first Fourier transform peak near 1.8 Å in Eu(III)–HA, Eu(III)-rectorite, and Eu(III)–HA-rectorite samples arises from the single backscattering of the photoelectron on oxygen atoms in the first coordination sphere. The R (bond distance) values of the first peak in Eu(III)–HA, Eu(III)-rectorite, and Eu(III)–HA-rectorite samples are different from that of the first peak in reference samples. The bond distance difference between Eu and O may be attributed to the combination of Eu(III) with the carboxylate (or hydroxyl) groups on the surface of Na-rectorite. Fig. 6B shows the first-shell fit of the sorption and reference samples. The relative parameters are listed in Table 3. One can see that HA has slight effect on Eu(III) sorption to rectorite. From Table 3, the REuAO and N values of Eu(III)–HA-rectorite are smaller than those of Eu(III)–rectorite. The REuAO and N values of Eu(III)–

255

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

(A)

O1s

O1s

(B)

Na1s

Na1s Eu3d

Eu3d C1s

Intensity (a.u.)

C1s

200 300 400 500 600 700 800 900 1000 1100

200 300 400 500 600 700 800 900 1000 1100

C1s

C1s

C-C CXHy C=O

282

284

286

288

Eu3d

290

282

284

288 Eu3d5/2

290

satellite

satellite

1120

286

Eu3d

Eu3d5/2

O-C=O

1130

1140

1120

1130

1140

1150

Binding Energy (eV) Fig. 5. XPS spectra of the survey, C 1s and Eu 3d; (A) Eu(III)–rectorite and (B) Eu(III)–HA-rectorite. C 1s and Eu 3d spectra of the Eu(III)–HA-rectorite system are the results of a fit curve. m/V (rectorite) = 0.2 g/L, C[Eu(III)]initial = 0.2 lmol/L, C(NaClO4) = 0.01 mol/L, C(HA)initial = 10 mg/L, pH = 5.0 T = 25 ± 1 °C.

(A)

(B)

3+

Eu (aq)

3+

Eu (aq)

Eu2O3

Eu2O3

Eu(OH)3

k x (k)

FTs Magnitude

Eu(OH)3

2

Eu-HA-rectorite

Eu-HA-rectorite

Eu-HA(aq)

Eu-HA(aq)

Eu-rectorite

Eu-rectorite

2

4

6

2

o

4

6 O

8

-1

k (A )

R (A)

Fig. 6. The Fourier transforms (A) and first-shell fit of the EXAFS function of the reference and sorption samples (B). Experimental (open circles) and model (solid line) contribution for the EuAO shell.

HA are the smallest. The changes in R and N values point to changes in the near-neighbor surrounding of the complexing ion Eu(III) due to sorption and/or complexation processes. It is not surprising that surface complexes of Eu(III) sorbed onto rectorite show another complex surrounding compared to Eu(III) complexed with HA or surface complexes of Eu(III) sorbed onto rectorite in the presence of HA.

4. Conclusion A better macroscopic and spectroscopic description demonstrated the effect of HA on Eu(III) uptake to Na-rectorite. The presence of HA significantly modified the Eu(III) sorption in the aqueous environment. HA and HA–Eu(III) are formed aggregates on the surfaces of Na-rectorite. The sizes of Eu(III)–HA-rectorite

256

C. Chen et al. / Journal of Colloid and Interface Science 393 (2013) 249–256

Table 3 Quantitative analysis of the first EuAO EXAFS path.a Sample

Shell

REuAO (Å)

N

Eu(aq) Eu2O3 Eu(OH)3 Eu(III)-rectorite Eu(III)–HA Eu(III)–HA-rectorite

EuAO EuAO EuAO EuAO EuAO EuAO

2.422(3) 2.345(36) 2.417(2) 2.45(2) 2.39(1) 2.41(2)

8.33(23) 6.37(22) 8.14(10) 11(2) 8 (1) 10(2)

a Number in parentheses at the end of each value indicates the uncertainties. REuAO is the bond distance of Eu and O in the first shell of Eu. N is the coordination number.

are smaller than those of HA-rectorite. The Eu(III) uptake on rectorite in the absence and presence of HA was simulated well by the DLM model. HA has slight effect on the EXAFS structural parameters in the Eu(III)–HA-rectorite system. Some of Eu(III) ions and HA may be intercalated into the interlayer space of Na-rectorite. The findings of this work are of great environmental importance toward a molecular-level description of Eu(III) and related trivalent lanthanide and actinide uptake mechanisms at water–mineral interface in the absence and presence of humic substances. The information presented herein will allow scientists and engineers to develop better models to evaluate Eu(III) interaction with clay minerals. Acknowledgment Financial supports from NSFC (41273134, 91126020, 21071147, 21071107) are acknowledged. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2012.10.032. References [1] D.L. Sparks, Environmental Soil Chemistry, Academic Press, San Diego, 2003. [2] A. Krˇepelová, T. Reich, S. Sachs, J. Drebert, G. Bernhard, J. Colloid Interface Sci. 319 (2008) 40. [3] J.F. Boily, J.B. Fein, Chem. Geol. 168 (2000) 239. [4] H. Kipton, J. Powell, R.M. Town, Anal. Chim. Acta 267 (1992) 47. [5] N.C. Brady, The Nature and Properties of Soils, 10th ed., Macmillan, New York, 1990. p. 286.

[6] R. Apak, Encyclopedia of Surface & Colloid Science (Invited Review), Dekker, New York, 2002. [7] J.P. Olivier, M.L. Occelli, Microporous Mesoporous Mater. 57 (2003) 291. [8] H.L. Hong, L.Y. Tie, Q.J. Bian, Y. Zhou, J. Chin. Electron Microsc. Soc. 24 (2005) 124 (Chinese with English abstract). [9] G. Zhang, Y. Liu, Y. Xie, X. Yang, Appl. Clay Sci. 29 (2005) 15. [10] G. Zhang, X. Yang, Y. Liu, Y. Jia, G. Yu, S. Ouyang, J. Colloid Interface Sci. 278 (2004) 265. [11] T. Murakami, A. Inoue, B. Lanson, A. Meunier, D. Beaufort, Clay Clay Miners. 53 (2005) 440. [12] P.P. Chang, X.K. Wang, S.M. Yu, W.S. Wu, Colloid Surf. A 302 (2007) 75. [13] D. Xu, C.L. Chen, X.L. Tan, J. Hu, X.K. Wang, Appl. Geochem. 22 (2007) 2892. [14] Th. Rabung, Th. Stumpf, H. Geckeis, R. Klenze, J.I.A. Kim, Radiochim. Acta 88 (2000) 711. [15] J. Zhang, J.J. Zhai, F.Z. Zhao, Z.Y. Tao, Anal. Chim. Acta 378 (1999) 177. [16] W.X. Liu, Z.X. Sun, W. Forsling, H.X. Tang, J. Colloid Interface Sci. 219 (1999) 48. [17] Z.S. Chu, W.X. Liu, H.X. Tang, T.W. Qian, S.H. Li, Z.T. Li, G.B. Wu, J. Colloid Interface Sci. 252 (2002) 426. [18] C.L. Chen, J. Hu, D. Xu, X.L. Tan, Y.D. Meng, X.K. Wang, J. Colloid Interface Sci. 323 (2008) 33. [19] R.M. Floroiu, A.P. Davis, A. Torrents, Environ. Sci. Technol. 35 (2001) 348. [20] S.B. Yang, J. Hu, C.L. Chen, D.D. Shao, X.K. Wang, Environ. Sci. Technol. 45 (2011) 3621. [21] W.L. Yan, R. Bai, Water Res. 39 (2005) 688. [22] M.H. Bradbury, B. Baeyens, Geochim. Cosmochim. Acta 66 (2002) 2325. [23] T.J. Strathmann, S.C.B. Myneni, Environ. Sci. Technol. 39 (2005) 4027. [24] T.H. Yoon, H. Moon, Y.J. Park, K.K. Park, Environ. Sci. Technol. 28 (2009) 2139. [25] Y. Takahashi, T. Kimura, Y. Minai, Geochim. Cosmochim. Acta 66 (2002) 1. [26] X.L. Tan, C.L. Chen, S.M. Yu, X.K. Wang, Appl. Geochem. 23 (2008) 2767. [27] Y. Takahashi, Y. Minai, S. Ambe, Y. Makide, F. Ambe, Geochim. Cosmochim. Acta 63 (1999) 815. [28] G. Montavon, S. Markai, Y. Andres, B. Grambow, Environ. Sci. Technol. 36 (2002) 3303. [29] B.H. Gu, J. Schmitt, Z.H. Chen, L.Y. Liang, J.F. McCarthy, Environ. Sci. Technol. 28 (1994) 38. [30] P. Sukveeradachgul, W. Pijitrojana, Appl. Surf. Sci. 254 (2008) 7651. [31] Y.M. Korolev, Sov. Phys. Crystallogr. 16 (1971) 250. [32] T. Shahwan, H.N. Erten, Radiochim. Acta 93 (2005) 225. [33] F. Mercier, C. Alliot, L. Bion, N. Thromat, P. Toulhoat, J. Electron Spectrosc. 150 (2006) 21. [34] Y.Q. Wang, H. Viswanathan, A.A. Audi, P.M.A. Sherwood, Chem. Mater. 12 (2000) 1100. [35] D. Schild, C.M. Marquardt, Radiochim. Acta 88 (2000) 587. [36] X.L. Tan, Q.H. Fan, X.K. Wang, B. Grambow, Environ. Sci. Technol. 43 (2009) 3115. [37] A. Krepelová, T. Reich, S. Sachs, J. Drebert, G. Bernhard, J. Colloid Interface Sci. 319 (2008) 40. [38] J. Hizal, R. Apak, J. Colloid Interface Sci. 295 (2006) 1. [39] Q.H. Fan, X.L. Tan, J.X. Li, X.K. Wang, W.S. Wu, G. Montavon, Environ. Sci. Technol. 43 (2009) 5776. [40] W.J. Zhong, H. Bo, Z. Li, S.Q. Wei, J. Chin. Univ. Sci. Technol. 31 (2001) 328. [41] M.L. Schlegel, I. Pointeau, N. Coreau, P. Reiller, Environ. Sci. Technol. 38 (2004) 4423.