Sunteți pe pagina 1din 9

Journal of Colloid and Interface Science 442 (2015) 30–38

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science


www.elsevier.com/locate/jcis

Preparation and characterization of bifunctional Ti–Fe kaolinite


composite for Cr(VI) removal
Hussain Fida, Sheng Guo, Gaoke Zhang ⇑
School of Resources and Environmental Engineering, Wuhan University of Technology, Wuhan 430070, China

a r t i c l e i n f o a b s t r a c t

Article history: A novel bifunctional Ti–Fe kaolinite composite with excellent adsorption and photocatalytic properties
Received 2 October 2014 was synthesized by a simple precipitation method. X-ray diffraction analysis and high-resolution trans-
Accepted 11 November 2014 mission electron microscope analysis confirmed the existence of rutile phase TiO2 and amorphous iron in
Available online 18 November 2014
the composite. The specific surface area of the Ti–Fe kaolinite composite is 5.5 times higher than that of
the original kaolinite. The composite was used as an adsorbent as well as photocatalyst for Cr(VI)
Keywords: removal. The results indicate that the low pH is favorable to the Cr(VI) removal by the composite and
Adsorption
the removal rate of Cr(VI) reached 87% at pH 3.0. Visible light irradiation obviously increased the removal
Clay
Chromium
of Cr(VI) by the composite and greatly shortened reaction equilibrium time, which may be attributed to
Ti–Fe kaolinite composite the photocatalytic reduction of Cr(VI) to Cr(III) by TiO2 associated with simultaneous redox cycle of
Photocatalytic reduction Fe(III)/Fe(II). Various common co-existing ions did not show obvious effects on the removal of Cr(VI)
by the composite. The composite exhibited very high stability for the Cr(VI) removal. The adsorption
models and thermodynamics of Cr(VI) onto the composite were studied.
Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction biomass [11], zeolites and modified zeolites [12], tree fern [13],
bone char [14], clay and modified clay [15], have been used as
Chromium is widely distributed in the earth’s crust, and is adsorbent for chromium removal.
extensively used in many industries including paints, dyes, metal Clays and their modified forms have been employed extensively
plating, automobile spare parts, stainless steel, leather tanned for removal of chromium because of the low-cost and higher effi-
and petroleum refining. Hexavalent chromium Cr(VI) has been ciency [16]. Natural and Na-purified smectite and kaolinite were
reported to be toxic as well as carcinogenic [1,2]. Chronic exposure employed for Cr(III) removal, adsorption capacity was 117.5 and
to high level of chromium may cause serious health problems, 61.4 mg/g, respectively for natural and modified clays [17]. Lang-
including respiratory tract, stomach, small intestine, male repro- muir adsorption capacities for Cr(III) on montmorillonite modified
ductive system and developmental disorders in children [3]. Max- by poly(hydroxoaluminium) and cetyltrimethylammonium bro-
imum permissible intake level of chromium in drinking water set mide were 26.67 and 25.51 mg/g, respectively [18]. Wu et al.
by World Health Organization (WHO) is 0.05 mg/L [4]. Due to its [19] reported Ca-montmorillonite modified with humic acid for
high toxicity and a potential carcinogenic activity, removal of the adsorption of Cr(III). Turkish montmorillonite showed higher
Cr(VI) from aqueous media has attracted strong environmental adsorption capacity for Cr(VI) removal at pH 1.0 [20]. Many mod-
concerns. Many technologies such as chemical precipitation, phy- ified forms of kaolinite including acid-activated, polyoxozirconium,
toextraction, electrodialysis, reverse osmosis, ion exchange, mem- and tetrabutylammonium kaolinite, were also investigated as a
brane filtration and adsorption are widely used for removal of potential adsorbents for Cr(VI) in aqueous medium [21]. Modified
chromium(VI) from water and wastewater [5]. Among these, organo-bentonite was also found to be an efficient adsorbent for
adsorption seems to be more promising because of its cost effec- the removal of Cr(VI) [22]. However, to the best of our knowledge,
tiveness, high efficiency, ease of recovery and reusability [6]. Sev- there is no report regarding the preparation and application of the
eral conventional as well as nonconventional materials including Ti–Fe kaolinite composite for Cr(VI) removal to date.
red mud [7], coconut coirpith [8], sewage sludge [9], fly ash [10], In this work, we report the preparation of the novel bifunctional
Ti–Fe kaolinite composite by a facile precipitation method. For the
⇑ Corresponding author. Fax: +86 27 87887445. first time, this composite was used as adsorbent and photocatalyst
E-mail address: gkzhang@whut.edu.cn (G. Zhang). for the removal of Cr(VI). The effects of the Fe/Ti ratio in the

http://dx.doi.org/10.1016/j.jcis.2014.11.023
0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38 31

composite, adsorption conditions and visible light irradiation on solution. The effect of common coexisting ions including sodium,
the Cr(VI) removal by the composite were studied. A possible pho- potassium, iron, nitrate, chloride, sulfate and bicarbonate on the
tocatalytic mechanism for the reduction of Cr(VI) to Cr(III) sup- adsorption of Cr(VI) by the composite was also studied.
ported by XPS results was proposed.
2.4. Adsorption isotherms
2. Materials and methods
Adsorption isotherms are expressed in terms of the relationship
All reagents were analytical grade and used as received without between the concentration of adsorbate in the solution and the
any further purification. Deionized water was used throughout the amount adsorbed by the unit mass of the adsorbent at a constant
experiments. The original kaolinite sample used in this study was temperature. The isotherms models such as Langmuir, Freundlich,
obtained from Jiangsu Province (China) and was sieved through a Tempkin and Dubinin–Radushkevich equations were applied to
325-mesh prior to its use in the experiments. investigate the adsorption mechanism of Cr(VI) over the composite.
 
Ce 1 1
2.1. Preparation of the Ti–Fe kaolinite composite
Langmuir equation ¼ þ Ce ð1Þ
qe bQ m Qm

The Ti–Fe kaolinite composite was synthesized in two steps. In 1


the first step, 0.5 M NaHCO3 was added into the mixed solution of Freundlich isotherm log qe ¼ log K f þ log C e ð2Þ
n
TiCl4 and FeCl3 in different Ti4+/Fe3+ molar ratios (0.05:0.1, 0.1:0.1,
0.2:0.1, 0.3:0.1, 0.1:0.2, 0.1:0.3 and 0.1:0.4) with constant stirring. Tempkin equation qe ¼ BT ln AT þ BT ln C e ð3Þ
Then, the as-obtained mixed solution was aged for 3 h at 55 °C and
added drop wise to the suspension of kaolinite (2 wt%) under Dubinin—Radushkevich qe ¼ Q d
vigorous stirring followed by aging for 24 h. The precipitate was   2 !
collected by filtration and washed with deionized water for several 1
 exp BD RT ln 1 þ
times, until no chlorides were detected by AgNO3 test. The Ce
as-obtained product was dried in oven at 105 °C for 12 h, and ð4Þ
was ground to a fine powder, which was hereinafter named as the
Ti–Fe kaolinite composite.
2.5. Photocatalytic experiment
2.2. Characterization of the composite
The photocatalytic activity of the as prepared Ti–Fe kaolinite
The structure and crystallinity of the Ti–Fe kaolinite composite composite was evaluated by the Cr(VI) removal from aqueous solu-
were analyzed by powder X-ray diffraction (XRD) on a D/MAX-RB tion under the visible light irradiation (300 W Dy lamp with a
X-ray powder diffractometer (Rigaku, Japan) with Cu Ka radiation. 420 nm cutoff filter). In a typical experiment, 200 mg of the compos-
The morphology of the Ti–Fe kaolinite composite was analyzed by ite was dispersed in 100 mL of 50 mg/L Cr(VI). During irradiation,
scanning electron microscope (SEM) (JSM5610LV). High-resolution 2 mL of sample was taken at different time intervals and centrifuged
transmission electron microscopy (HRTEM) image was recorded immediately to remove the composite particles. The residual con-
with a HRTEM at an accelerating voltage of 200 kV. The nitrogen centration of Cr(VI) in the supernatant was measured by AAS.
adsorption/desorption isotherms were measured on an AUTO-
SORB-1 nitrogen adsorption apparatus (Quantachrome, USA). 2.6. Desorption and regeneration of the adsorbent
X-ray photoelectron spectroscopy (XPS) analysis was carried out
using an ESCALAB II XPS system with a monochromatic Mg Ka For the desorption study, 200 mg of the composite was con-
source and a charge neutralizer. All binding energies were referred tacted with 100 mL of 50 mg/L Cr(VI) solution. After the adsorption
to the C 1s peak at 284.5 eV of the surface adventitious carbon. experiment, the adsorbent was collected by filtration using
Peak deconvolution was performed using Gaussian–Lorentzian 0.45 lm millipore filter. Then it was transferred into 50 mL aque-
components after a Linear background subtraction. The FTIR spec- ous solution with deionized water, 0.1 M NaCl, 0.1 M HNO3 and
tra were collected by Nexus, Thermo Nicolet. Thermogravimetry 0.1 M HCl, respectively [24]. The mixtures were shaken at room
(TG) analysis of the samples was carried out on a Netzsch STA temperature overnight, and the filtrates were analyzed to deter-
449C Jupiter Aeolos (German) coupled to mass spectrometer with mine the concentration of Cr(VI) after desorption, the same exper-
a heating rate of 5 °C/min under N2. iment was repeated for six times.

2.3. Adsorption studies 3. Results and discussion

A definite amount 2 g/L (0.2 g) of the Ti–Fe kaolinite composite 3.1. Characterization of the Ti–Fe kaolinite composite
was added into a flask with 100 mL of 50 mg/L Cr(VI) aqueous solu-
tion for adsorption experiments. The same amount was maintained The Fe–Ti kaolinite composites with different Ti4+/Fe3+ molar
throughout the whole study unless otherwise specified. The flasks ratios (0.05:0.1, 0.1:0.1, 0.2:0.1, 0.3:0.1, 0.1:0.2, 0.1:0.3 and
were shaken on a shaker at 25 °C with speed of 190 rpm. After fil- 0.1:0.4) were prepared and their sorption capacity for Cr(VI) was
tration, the concentration of Cr(VI) was determined using atomic evaluated. As observed from Fig. 1, the increase of the Ti4+/Fe3+
absorption spectrophotometer (AAS, Analytik Jena ZEEnit 700). In molar ratio in the composite has no effect on the adsorption capac-
order to make a comparison, we took random samples from ity of the composite. However, the adsorption capacity increased
adsorption experiment and analyzed Cr(VI) in the samples by from 21.05 to 24.8 mg/g with increasing the Fe3+ molar ratio in
1,5-diphenyl carbazide method at a wavelength of 540 nm using the composite from 0.1 to 0.4. Meanwhile, the iron leaching of
visible spectrophotometer and AAS method separately. The as- the composite showed an increase with increasing the Fe3+ molar
obtained results were comparable within 5% [23]. The pH of the ratio. The composite with 1:1 molar ratio of Ti4+/Fe3+ was selected
Cr(VI) solution was adjusted by adding 0.1 M NaOH or 0.1 M HCl for further study because of its good adsorption capacity as well as
32 H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38

30 0.5

Ti-Fe kaolinite
Adsorption capacity (mg/g)

25 0.4

Leaching of iron (mg/L)

Transmittance
20 Adsorption capacity 0.3
Leaching of iron
Kaolinite

15 0.2
Ti-Fe kaolinite

Transmittance
Kaolinite

10 0.1
0.05:0.1 0.1:0.1 0.2:0.1 0.3:0.1 0.1:0.2 0.1:0.3 0.1:0.4 800 700 600 -1500 400
Wave number (cm )
4+ 3+
Ti /Fe molar ratio in Ti-Fe kaolinite composite 4000 3500 3000 2500 2000 1500 1000 500

Fig. 1. The Cr(VI) removal efficiency and iron leaching of the Ti–Fe kaolinite
Wavenumber (cm-1)
composite with different molar ratios of Ti4+/Fe3+ (conditions: adsorbent 2 g/L,
Fig. 3. FT-IR spectra of the original kaolinite and the Ti–Fe kaolinite composite.
Cr(VI) conc. 50 mg/L, pH 3.0, temperature 298 K).

acceptable iron leaching. The content of Ti and Fe in the Ti–Fe composite disappeared completely as compared to that of the ori-
kaolinite composite was 13.96 and 8.90 wt%, respectively, as ginal kaolinite. This indicates that the Ti–Fe kaolinite composite is
determined by ICP-OES. not a simple mixture of Ti oxide and Fe oxide, but a synergistic
The adsorption capacity of the Ti–Fe kaolinite composite was interaction between Fe and Ti occurred during the synthesis [27].
compared with that of the original kaolinite, Ti-kaolinite, Fe-kao-
linite and TiO2. The results in Fig. 2 indicate that adsorption rate 3.1.2. XRD patterns
of Cr(VI) by the Ti–Fe kaolinite composite reached 86.7%, which The XRD pattern of the original kaolinite exhibits the (0 0 1) and
is much higher that (all less than 56%) of the original kaolinite, (0 0 2) peaks of kaolinite at 2h/(planer distance) = 12.5°/7.19 and
Ti-kaolinite, Fe-kaolinite and TiO2. However, the adsorption effi- 24.9°/3.58 Å, respectively (PDF No.: 78-1996), as shown in Fig. 4
ciencies of the Ti–Fe and Ti–Fe kaolinite composite are comparable. [28]. The decreased intensity of the (0 0 1) and (0 0 2) peaks of kao-
The Cr(VI) removal efficiency was increased by 10% when kaolinite linite in the Ti–Fe kaolinite composite can be attributed to the dis-
was used as the support for Ti–Fe. Kaolinite is natural, abundant, tortion of the octahedral and tetrahedral structures in the kaolinite
environment friendly and cost-effective support for the Ti–Fe. in the composite [29]. The new peaks at 2h = 27.5°, 37.0° and 41.3°
Using kaolinite as a support makes Ti–Fe more efficient and eco- in the composite corresponding to the planar distances 3.18, 2.83
nomically feasible. and 2.17 Å can be assigned to (1 1 0), (1 0 1) and (1 1 1) of rutile
phase, respectively. The two new peaks at about 35.4°/2.53 and
62°/1.49 Å in the XRD pattern of the Ti–Fe kaolinite composite
3.1.1. FT-IR adsorption spectra
were observed, which could indicate the presence of poorly
The FT-IR spectra of the original kaolinite and the Ti–Fe kaolin-
ordered 2-line ferrihydrite [30].
ite composite are shown in Fig. 3. The spectrum of the original
kaolinite shows major absorption peaks at 3620–3695 cm1 which
3.1.3. SEM and HRTEM images
corresponds to –OH stretching vibrations [25]. The band at
The morphology of the original and Ti–Fe kaolinite composite
1050 cm1 could be ascribed to Si–O stretching vibrations. Strong
was studied by SEM analysis. The original kaolinite (Fig. 5a) exhib-
absorption bands at 902 and 540 cm1 indicate OH and Si–O bend-
its smooth surface and compose of the agglutinative flakes. Fig. 5b
ing frequencies [26]. The spectrum of the Ti–Fe kaolinite composite
shows similar absorption bands. However, the small peaks ranging
from 800 to 400 cm1 in the FTIR spectra of Ti–Fe kaolinite
K (001) K (002)

90 Ti-Fe Kaolinite

Ti-Fe
75 R- Rutile phase of TiO2

* Ferrihydrite
% removal of Cr (VI)

60 Ti-Kaolinite
Kaolinite
TiO2
45

R (110)
30 R (101)
R (111) Ti-Fe-Kaolinite
Kaolinite

15 Fe-Kaolinite * *
0 10 20 30 40 50 60 70
2 Theta (degree)
Fig. 2. The Cr(VI) removal efficiency of different materials (conditions: adsorbent
2 g/L, Cr(VI) conc. 50 mg/L, pH 3.0, temperature 298 K). Fig. 4. XRD patterns of the original kaolinite and the Ti–Fe kaolinite composite.
H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38 33

shows that the Ti–Fe kaolinite composite has coarse surface with
0.0020
many small particles. The phase structure of the Ti–Fe kaolinite 90
Kaolinite

dV/dlog (w) (cm g )


3 -1
0.0015
composite was further demonstrated by HRTEM image, as shown 80
Ti-Fe kaolinite composite
in Fig. 5d. The lattice fringe spacings of 0.24 and 0.32 nm corre- 0.0010

Volume adsorbed (cm g )


70

3 -1
spond to the (1 0 1) and (1 1 0) crystallographic planes of rutile 0.0005
TiO2 phase, respectively. These results are in well agreement with 60
0.0000
that of the XRD analysis.
50 0 50 100 150 200 250 300 350
Pore diameter (nm)
3.1.4. Nitrogen adsorption–desorption 40
Nitrogen adsorption–desorption isotherms and the pore size
30
distributions plot of the kaolinite and Ti–Fe kaolinite composite
are displayed in Fig. 6. The BET specific surface area of the original 20
kaolinite (Table 1) was found to be 19.47 m2/g, which is very close 10
to that reported in the literature [31,32]. The pore volume and pore
diameter were 0.0047 cc/g and 11.3 nm, respectively. The BET spe- 0
0.0 0.2 0.4 0.6 0.8 1.0
cific surface area of the Ti–Fe kaolinite composite reached
Relative pressure (p/p 0)
108.2 m2/g, which is in favor of the adsorption of Cr(VI). This result
very well interprets the Cr(VI) adsorption capacity of the original
Fig. 6. N2 adsorption–desorption isotherms and the pore size distributions plot of
kaolinite and the Ti–Fe kaolinite composite in Fig. 2. kaolinite and the Ti–Fe kaolinite composite.

3.1.5. Thermal analysis


Thermal behavior of the original kaolinite and Ti–Fe kaolinite in dehydration step. Similarly, the presence of Ti in the composite
composite was investigated by TG/DSC analysis, which is given in resulted in a decrease in the hydroxyl contents and a dehydroxyla-
Fig. 7. The thermogram of the original kaolinite showed a minor tion with smaller weight loss occurred at relatively low
weight loss below 100 °C and a major weight loss about 12% temperature.
appeared between 490 and 520 °C. The thermal behavior of the
Ti–Fe kaolinite composite is analogous, however the weight loss 3.2. Adsorption of Cr(VI) on the Ti–Fe kaolinite composite
associated with dehydration is larger (4%) and the dehydroxylation
process is smaller (6%) than that of the original kaolinite. This may 3.2.1. Effect of pH
be assigned to the presence of more water in the interlayer spaces The pH value of the medium is a significant factor in adsorption
of kaolinite because of the Fe introduction, which is easily removed process [33]. Fig. 8 reveals that the lower pH favors the adsorption

Fig. 5. SEM images of (a) the original kaolinite and (b) the Ti–Fe kaolinite composite; (c) TEM image of the Ti–Fe kaolinite composite; (d) HRTEM image of the Ti–Fe kaolinite
composite, the inset is HRTEM image of lattices of TiO2.
34 H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38

Table 1 Cr2 O2 


7 þ H2 O ! 2HCrO4 ð5Þ
BET specific surface area and pore size data of the original kaolinite and the Ti–Fe
kaolinite composite.
HCrO4 ! CrO2
4 þH
þ
ð6Þ
Sample BET specific surface Pore volume Pore diameter
area (m2/g) (cc/g) (nm)
Original kaolinite 19.47 0.047 11.3
HCrO4 þ 7Hþ þ 3e ! Cr3þ þ 4H2 O ð7Þ
Ti–Fe kaolinite composite 108.2 0.141 5.22
CrO2 þ 
4 þ 8H þ 3e ! Cr

þ 4H2 O ð8Þ

102
Cr2 O2 þ 
7 þ 14H þ 6e ! 2Cr

þ 7H2 O ð9Þ
Kaolinite
100
Ti-Fe-kaolinite
3.2.2. Effect of Cr(VI) concentration
98 The effect of initial Cr(VI) concentration on the adsorption is
96 indicated in Fig. 9. It is observed that the Cr(VI) adsorption rate
over the composite decreased from 99.06% to 75.47% and the cor-
TG (%)

94 responding adsorption capacity increased from 4.88 to 22.37 mg/g


92 with the increase of Cr(VI) concentration from 10 to 60 mg/L. This
may be attributed to the limited number of active sites on the
90 adsorbent surface. On the other hand, the increase in the adsorp-
tion capacity with the increase of the Cr(VI) concentration may
88
be due to the higher adsorption rate and the utilization of all the
86 active sites [37–39].
84
200 400 600 800 1000 3.2.3. Adsorption isotherms
O
Temperature ( C) 3.2.3.1. Langmuir isotherm. The obtained experimental data were
fitted by the Langmuir model, the plot of Ce/qe against Ce is shown
Fig. 7. TG curves of the original kaolinite and the Ti–Fe kaolinite composite. in Fig. 10. Langmuir constants qm and b were calculated from slope
and intercept of the plot. Monolayer adsorption capacity qm was
23.47 mg/g in the equilibrium at pH 3.0. The Langmuir dimension-
of Cr(VI) over the composite. The adsorption capacity decreased less equilibrium parameter RL value was equal to 1.0, which shows
with the increase of the solution pH. It is well known that, Cr(VI)
2
linear adsorption of Cr(VI) onto the Ti–Fe kaolinite composite. The
exists as HCrO4 ; Cr2 O7 in solution at optimum sorption pH. In Langmuir model was well fitted to the experimental data on the

acidic media, HCrO4 is the dominant form which results from the
basis of high R2 value (0.993). The data obtained from the adsorp-
hydrolysis of dichromate ion, the increase in pH would facilitate tion isotherms are summarized in Table 2.
the formation of Cr2 O2 
7 from HCrO4 [34,35]. Hence, it can be con-
cluded that Cr(VI) is adsorbed on the protonated active sites of the
3.2.3.2. Freundlich isotherm. The Freundlich constants Kf and n were
Ti–Fe kaolinite composite, with increasing the pH, the surface of
obtained from the slope and intercept of plot between log qe and
the adsorbent become negatively charged resulting repulsion of
log Ce which is shown in Fig. 11. Kf and n are 11.12 mg/g and
the chromate ions and decreases the adsorption of Cr(VI). Maxi-
0.324, respectively. The obtained R2 value is (0.991).
mum removal of Cr(VI) was observed at lower solution pH, the
adsorption capacity over the composite can reach 21.05 mg/g even
at pH 3.0, which is still higher than that of the adsorbent reported 3.2.3.3. Tempkin isotherm. The Tempkin isotherm constants AT and
in most of literatures at the same pH [36]. Further experiments bT can be calculated from plot of qe vs. ln Ce at constant tempera-
were carried out using pH 3.0. ture. The equilibrium binding constant AT is 42.00 L/mg; bT is
655.46 J/mol, which corresponds to the heat of adsorption of Cr(VI)

100
100

80
95 20
Adsorption capacity (mg/g)
Removal of Cr(VI) (%)

Removal of Cr(VI) (%)

60
90
15

40 85
Removal (%) 10
Adsorption capacity
20 80

5
0 75
2 4 6 8 10 10 20 30 40 50 60
pH Cr(VI) concentration (mg/L)

Fig. 8. Effect of pH on the removal of Cr(VI) by the Ti–Fe kaolinite composite Fig. 9. Effect of initial Cr(VI) concentration on the adsorption of Cr(VI) on the Ti–Fe
(conditions: adsorbent 2 g/L, Cr(VI) conc. 50 mg/L, time 240 min, temperature kaolinite composite (conditions: adsorbent 2 g/L, pH 3.0, time 240 min, tempera-
298 K). ture 298 K).
H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38 35

adsorption is supposed to happen for value of E between 1 and


0.7
16 kJ/mol while chemisorption prevails for E > 16 kJ/mol. The
0.6 obtained R2 is 0.845, which displays that the model does not fit
well with the equilibrium experimental data.
0.5
1
E ¼ pffiffiffiffiffiffi ð10Þ
0.4 BD
Ce/qe (g/L)

0.3
3.3. Effect of adsorbent dosage
0.2
The removal efficiency of Cr(VI) on the Ti–Fe kaolinite compos-
0.1 ite was also evaluated as function of the adsorbent dosage (Fig. 12).
The removal rate of Cr(VI) on the composite was increased from
0.0
66.19% to 97.11% while the adsorption capacity of the composite
-2 0 2 4 6 8 10 12 14 16 decreased from 32.1 to 11.78 mg/g with increasing the adsorbent
Ce (mg/L) amount from 1 to 4 g/L. The enhanced Cr(VI) removal efficiency
could be ascribed to the vast availability of surface area and
Fig. 10. Langmuir isotherm model for the adsorption of Cr(VI) onto the Ti–Fe adsorption sites [40]. The drop in the adsorption capacity may be
kaolinite composite. attributed to the sites remaining unsaturated during adsorption
process [41]. Higher adsorbent dosage may lead to particles aggre-
gation, resulting in the decrease of the exposed surface area as well
Table 2
as the increase of the diffusion path.
The constants and regression data of isotherms for the adsorption of Cr(VI) on the Ti–
Fe kaolinite composite (conditions: adsorbent 2 g/L, temperature 298 K, Cr(VI) conc. 3.4. Effect of temperature (thermodynamic studies)
10–50 mg/L, pH 3.0, time 240 min).

Adsorption isotherm Isotherm parameters R2 Temperature dependence of Cr(VI) adsorption on the Ti–Fe kao-
Langmuir Qm (mg g ) 1
23.47 0.993
linite composite was studied in the temperature range from 293 to
b (L mg1) 1.078 333 K. Significant increase in Cr(VI) removal has been observed
Freundlich isotherm Kf (mg11/n g1 L1/n) 11.12 0.991
with the increase of temperature. Higher temperature may facili-
n 0.209 tate Cr(VI) ions to overcome the adsorption activation energy bar-
Tempkin equation AT (L mg1) 42 0.981
rier and easily attach to the adsorbent surface [42,43].
BT (J mol1) 3.78 Thermodynamic parameters such as DH, DS and DG were calcu-
bT 588.46 lated from the following equations which are listed in Table 3.
Dubinin–Radushkevich Qm (mg g1) 17.07 0.874
E (kJ mol1) 3.745

100 35

3.2 95

Adsorption capacity (mg/g)


30
Removal of Cr(VI) (%)

3.0 90
Removal (%)
2.8 85 Adsorption capacity 25
2.6 80
ln[qe(mg/g)]

20
2.4
75
2.2
70 15
2.0
65
1.8 10
1 2 3 4
1.6 Adsorbent amount (g/L)
1.4 Fig. 12. Effect of adsorbent dosage on the adsorption of Cr(VI) onto the Ti–Fe
-3 -2 -1 0 1 2 kaolinite composite (conditions: Cr(VI) conc. 50 mg/L, pH 3.0, time 240 min,
ln[Ce(mg/L)] temperature 298 K).

Fig. 11. Freundlich isotherm model for the adsorption of Cr(VI) onto the Ti–Fe
kaolinite composite. Table 3
Thermodynamic data for the adsorption of Cr(VI) on the Ti–Fe kaolinite composite
(conditions: adsorbent 2 g/L, Cr(VI) conc. 50 mg/L, pH 3.0, time 240 min).
on the composite. The obtained R2 coefficient is 0.981, which is
Temperature (K) DG (kJ mol1) DH (kJ mol1) DS (J K1 mol1)
lower than that of the Langmuir and Freundlich isotherm.
293 4.61959 32.424 146.42
303 5.17362
3.2.3.4. Dubinin–Radushkevich (D–R) isotherm. A plot between ln qe 313 5.65172
323 6.57437
and E2 at constant temperature was used to calculate Qm and k of
333 7.1128
D–R isotherm. BD is related to the mean free energy (E), physical
36 H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38

Ce 2.2
Kc ¼ ð11Þ
Ci
2.0

DG ¼ RT ln K c ð12Þ 1.8

1.6
DS DH
ln K c ¼  ð13Þ

log K c
R RT 1.4

R is the gas constant (8.3143 J/mol K), Ce and Ci are the equilibrium 1.2
concentration (mg/L) of the Cr(VI) in the adsorbent and solution,
respectively and T is the solution temperature (K). DH and DS were 1.0
calculated from the slope and intercept of the plot of ln Kc vs. 1/T as
0.8
shown in Fig. 13. The negative values of DG (Table 3) illustrate the
spontaneous adsorption of Cr(VI), while positive value of DH indi- 0.6
cates that the process is endothermic [44]. The more negative value 0.0030 0.0031 0.0032 0.0033 0.0034
of DG indicates a greater driving force to sorption and the process 1/T (K)
becomes more spontaneous at higher temperature [45]. The value
of DH (Table 3) is high enough to ensure the strong interactions Fig. 13. Plot of ln Kc vs. 1/T for Cr(VI) adsorbed on the Ti–Fe kaolinite composite
(conditions: adsorbent 2 g/L, Cr(VI) conc. 50 mg/L, pH 3.0, time 240 min).
between Cr(VI) and the Ti–Fe kaolinite composite.

Table 4
3.5. Effect of contact time and kinetics of adsorption Calculated kinetic parameters for the adsorption of Cr(VI) on the Ti–Fe kaolinite
composite (conditions: adsorbent 2 g/L, temperature 298 K, Cr(VI) conc. 50 mg/L, pH
The adsorption of Cr(VI) as function of contact time was stud- 3.0).
ied. The results indicate that the adsorption capacity increased Parameters Values
with the increase of interaction time, till equilibrium was estab-
Experimental qe 21.05
lished in 240 min. Rapid adsorption was observed at low coverage, Pseudo first order K1  103 (min1) 22.4
but as the surface coverage increases, the number of available sur- qe (mg g1) 11.13
face sites drop down. Ultimately, the rate of adsorption decreases R2 0.912
until equilibrium is achieved. Different models such as pseudo first Pseudo second order K2  103 (g mg1 min1) 5.18
order, pseudo second order, Elovich equation and intra-particle dif- qe (mg g1) 20.96
R2 0.998
fusion were used to evaluate the kinetics of Cr(VI) adsorption on
the Ti–Fe kaolinite composite. The as-obtained results are summa- Elovich model b (mg g1 min1) 0.46
rized in Table 4. The pseudo first order kinetics model ln(qe  qt) vs. a (g mg1 min2) 92.16
2
R 0.985
time does not fit well because of the poor regression coefficient
Intraparticle diffusion ki (mg g1 min0.5) 0.609
(R2 = 0.897). The pseudo second order kinetic equation can be
R2 0.946
expressed as [46–48].

t 1 t
¼ þ ð14Þ
qt K 2 q2e qe 16
The pseudo second order rate constant qe was calculated from 14
slope and intercept of the plot of t/qt as function of time (Fig. 14
and Table 4). The value of qe obtained experimentally (21.05 mg/g) 12
is in good agreement with that calculated from pseudo second order
10
equation (20.96 mg/g). Furthermore, the value of R2 also confirms
t/qt

that the suitability of the model to the obtained data. Elovich model 8
was applied to explain the chemosorption on heterogeneous sur-
6
faces (Table 4). A plot of qt against ln t has good linearity
(R2 = 0.974). The value of coefficient a is 92.16 (g mg1 min2), indi- 4
cating the rapid initial intake of Cr(VI). To further evaluate the sorp-
2
tion process of Cr(VI) on the composite, the obtained data were
plotted using intraparticle diffusion equation (Table 4). A plot of 0
(qt vs. t0.5) for intraparticle diffusion yield a straight line
(R2 = 0.988) which confirms intraparticle diffusion sorption [49,50]. 0 50 100 150 200 250 300
Time (min)

Fig. 14. Pseudo second order kinetic model for the adsorption of Cr(VI) by the Ti–Fe
3.6. Effect of coexisting ions on Cr(VI) removal
kaolinite composite (conditions: adsorbent 2 g/L, Cr(VI) conc. 50 mg/L, pH 3.0,
temperature 298 K).
Some ions such as sodium, potassium, iron, chloride, sulfate,
nitrate and bicarbonate are usually found in polluted water. The
effect of these ions on the removal of Cr(VI) by the Ti–Fe kaolinite 3.7. Regeneration and stability of the adsorbent
composite was investigated. From Fig. 15, it is obvious that these
anions and cations have negligible effect on the sorption of Cr(VI) The desorption rate of Cr(VI) in the composite was 46.0% and
over the composite, indicating that the Ti–Fe kaolinite composite 68.87% by deionized water and 0.1 M NaCl, respectively. When
has great practical potential application value. 0.1 M HCl and 0.1 M HNO3 solutions were used, the desorption rate
H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38 37

80

80
Removal of Cr(VI) (%)

Removal of Cr(VI) (%)


60

40

60
20

Controlled Sodium Potassium Iron Chloride Sulfate Nitrate Bicarbonate


0
Fig. 15. Influence of coexisting ions on the adsorption of Cr(VI) onto the Ti–Fe 0 1 2 3 4 5 6
kaolinite composite (conditions: adsorbent 2 g/L Cr(VI) conc. 50 mg/L, pH 3.0, Regeneration times
temperature 298 K).
Fig. 16. Regeneration and reuse of the Ti–Fe kaolinite composite for the removal of
Cr(VI).
of Cr(VI) in the composite reached 78.08% and 95.55%, respectively.
The adsorbent regenerated with HNO3 was used for consecutive six
cycles. The adsorption capacity remains more than 80% of the fresh
adsorbent after fifth recycling (Fig. 16). In order to evaluate the risk 22
of iron leakage to the filtrate, the leached iron in the solution after

Cr(VI) adsorption capacity (mg/g)


adsorption was determined. The iron leaching was only 0.12 mg/L, 20
which shows that most of the iron is firmly attached to the kaolin-
ite substrate. This value is not enough to contribute excessive level 18
of iron in treated water. Hence, the Ti–Fe kaolinite composite pos-
sesses promising regeneration ability and high stability. 16
Without visible light irradiation
3.8. Photocatalytic activity 14
With visible light irradiation

The photocatalytic activity of the Ti–Fe kaolinite composite for


12
Cr(VI) removal was evaluated. The results in Fig. 17 confirm that, as
compared to the adsorption without visible light irradiation, the
10
removal rate of Cr(VI) by the Ti–Fe kaolinite composite increased
0 50 100 150 200 250 300
from 86.68% to 95.26% and the Cr(VI) uptake capacity of the com-
Time (min)
posite also increased from 21.05 to 22.12 mg/g under visible light
irradiation. The most important and interesting fact was that the Fig. 17. The removal of Cr(VI) by the Ti–Fe kaolinite composite with and without
equilibrium time of Cr(VI) adsorption over the composite under visible light irradiation.
visible light irradiation was only half the time of the adsorption
process without visible light irradiation. The enhanced adsorption
efficiency may be due to the fact that the following reactions take Cr(III)
place when the Ti–Fe kaolinite composite was illuminated with Cr2p3/2
visible light.
hm þ Cr(III)
TiO2 ! h þ e ð15Þ Cr(VI)
Cr2p1/2
Intensity (a.u)

Cr2 O2 þ 
7 þ 14H þ 6e ! 2Cr

þ 7H2 O ð16Þ

TiO2 hm
FeðIIIÞ þ e ! FeðIIÞ ð17Þ

TiO2 hm
3FeðIIÞ þ CrðVIÞ ! 3FeðIIIÞ þ CrðIIIÞ ð18Þ
The photogenerated electron reduces Cr(VI) to Cr(III), the possi-
ble mechanism is same (Eq. (16)) as reported in the literatures
[51,52]. In the absence of reducing agent, H2O accepts the valence 590 588 586 584 582 580 578 576 574 572
band holes and the following reaction takes place. The presence of Binding energy (eV)
Fe(III) in the Ti–Fe kaolinite composite may also contribute to the
Fig. 18. High resolution XPS spectra of the recovered Ti–Fe kaolinite composite.
enhanced removal capacity [52,53]. XPS high resolution spectra of
the recovered Ti–Fe kaolinite composite after visible light irradia-
tion in Cr(VI) solution have been given in Fig. 18. Significant bands
at binding energy 576 and 585.3 eV for Cr(III) were observed, which 579.8 eV corresponding to Cr(VI) indicates that most of the Cr(VI)
correspond to Cr2p3/2 and Cr2p1/2 of Cr(III), respectively [54]. has been reduced to Cr(III). The results obtained from XPS analysis
Whereas the band with lower intensity at binding energy at suggest the reduction of Cr(VI) to Cr(III).
38 H. Fida et al. / Journal of Colloid and Interface Science 442 (2015) 30–38

4. Conclusions [14] S. Dahbi, M. Azzi, N. Saib, M. de la Guardia, R. Faure, R. Durand, Anal. Bioanal.
Chem. 374 (2002) 540–546.
[15] J. Liu, L.G.K. Zhang, Phys. Chem. Chem. Phys. 16 (2014) 8178–8192.
The as-prepared Ti–Fe kaolinite composite showed interesting [16] E.C. Leonel, E. Henrique de Faria, R.C. Pimentel, E.J. Nassar, K.J. Ciuffi, M.J. dos
bifunctional properties for removing Cr(VI) from aqueous medium. Reis, P.S. Calefi, Quim. Nova 35 (2012) 1407–1411.
[17] I.G. Abid, A. Jrad, K. Nahdi, M. Trabelsi-Ayadi, Desalination 246 (2009) 595–
The surface area of the Ti–Fe kaolinite composite is 5.5 times
604.
higher than that of the original kaolinite. The low pH is favorable [18] B. Hu, H. Luo, H. Chen, T. Dong, Appl. Clay Sci. 51 (2011) 198–201.
to the Cr(VI) removal by the composite. Kinetics study suggests [19] P. Wu, Q. Zhang, Y. Dai, N. Zhu, Z. Dang, P. Li, J. Wu, X. Wang, Geoderma 164
(2011) 215–219.
that pseudo-second-order adsorption model was suitable to
[20] S.T. Akar, Y. Yetimoglu, T. Gedikbey, Desalination 244 (2009) 97–108.
describe the adsorption behavior of the Ti–Fe kaolinite composite. [21] K.G. Bhattacharyya, S.S. Gupta, Ind. Eng. Chem. Res. 45 (2006) 7232–7240.
Langmuir adsorption isotherm perfectly matched the data. Nega- [22] D.L. Guerra, H.C.P. Oliveira, P.C.C. da Costa, R.R. Viana, C. Airoldi, Catena 82
tive values of DG indicate spontaneous tendency while positive (2010) 35–44.
[23] G. Wang, Q. Chang, M. Zhang, X. Han, React. Funct. Polym. 73 (2013) 1439–
value of DH describes the endothermic nature of the reaction. 1446.
The removal efficiency of the Ti–Fe kaolinite composite markedly [24] W.B. Amorim, A.M. Hayashi, P.F. Pimentel, M.G.C. da Silva, Braz. J. Chem. Eng.
increased from 86.80% to 95.26% and the reaction time was halved 20 (2003) 3–9.
[25] S. Guo, G.K. Zhang, J.Q. Wang, J. Colloid Interface Sci. 433 (2014) 1–8.
when the reaction mixture was exposed to visible light irradiation. [26] K.G. Bhattacharyya, S.S. Gupta, Ind. J. Chem. Technol. 16 (2009) 457–470.
The enhanced activity may be attributed to the photocatalytic [27] L. Chen, B.Y. He, S. He, T.J. Wang, C.L. Su, Y. Jin, Powder Technol. 227 (2012) 3–
reduction of Cr(VI)–Cr(III) by TiO2 with simultaneously redox cycle 8.
[28] Y. Zhang, H. Gan, G. Zhang, Chem. Eng. J. 172 (2011) 936–943.
of Fe(III)/Fe(II). The Ti–Fe kaolinite composite possess enhanced [29] M. Topkafa, H.F. Ayyildiz, F.N. Arslan, S. Kucukkolbasi, F. Durmaz, S. Sen, H.
Cr(VI) removal and detoxification efficiency. The removal effi- Kara, Pol. J. Food Nutr. Sci. 63 (2013) 147–154.
ciency of Ti–Fe kaolinite composite is almost higher than most of [30] J. Lu, L. Dongfang, H. Jing, Z. Guowei, L. Bin, Chem. Eng. Res. Des. (2014), http://
dx.doi.org/10.1016/j.cherd.2014.05.001.
adsorbent [6,41]. However, the removal capacity is slightly lower
[31] Ahmet Sari, Omer Idak, Bull. Chem. Soc. Ethiop. 20 (2006) 8.
than [55,56] but the reaction time for Ti–Fe kaolinite composite [32] J.C. Miranda-Trevino, C.A. Coles, Appl. Clay Sci. 23 (2003) 133–139.
is three times faster. The negligible iron leaching, high desorption [33] J. Anwar, U. Shafique, Z. Waheed uz, Z. un Nisa, M.A. Munawar, N. Jamil, M.
Salman, A. Dar, R. Rehman, J. Saif, H. Gul, T. Iqbal, Int. J. Phytorem. 13 (2011)
rate, efficient uptake of regenerated adsorbent, minor effect of
410–420.
coexisting ions, high efficiency for the removal of Cr(VI) from water [34] Y. Li, Z. Jin, T. Li, S. Li, Water Sci. Technol. 63 (2011) 2781–2787.
indicate that the Ti–Fe kaolinite composite is a promising option [35] N. Ahalya1, R.D. Karamadi, T.V. Ramachandra, J. Environ. Sci. Res. Int. 1 (2008).
for treating waste water containing Cr(VI). [36] K. Pakshirajan, A.N. Worku, M.A. Acheampong, H.J. Lubberding, P.N. Lens, Appl.
Biochem. Biotechnol. 170 (2013) 498–513.
[37] S. Gupta, B.V. Babu, Chem. Eng. J. 150 (2009) 352–365.
Acknowledgments [38] T.S. Anirudhan, P.G. Radhakrishnan, J. Colloid Interface Sci. 316 (2007) 268–
276.
[39] X. Lv, X. Xue, G. Jiang, D. Wu, T. Sheng, H. Zhou, X. Xu, J. Colloid Interface Sci.
This work was supported by National Program on Key Basic 417 (2014) 51–59.
Research Project of China (973 Program) 2013CB632402, Program [40] L. Li, L. Fan, M. Sun, H. Qiu, X. Li, H. Duan, C. Luo, Colloids Surf. B. Biointerfaces
of Specialized Research Fund for the Doctoral Program of Higher 107 (2013) 76–83.
[41] Y. Zhao, S. Yang, D. Ding, J. Chen, Y. Yang, Z. Lei, C. Feng, Z. Zhang, J. Colloid
Education (20110143110015) and National Key Technology R&D Interface Sci. 395 (2013) 198–204.
Program of China (2012BAJ25B02-03). [42] B. Geng, Z. Jin, T. Li, X. Qi, Chemosphere 75 (2009) 825–830.
[43] X.S. Wang, Z.Z. Li, S.R. Tao, J. Environ. Manage. 90 (2009) 721–729.
[44] A.A. Rouff, J. Colloid Interface Sci. 392 (2013) 343–348.
References
[45] M. Waseem, S. Mustafa, K.H. Shah, A. Naeem, S.S. Shah, K. Mehmood, S.U. Din,
Sep. Purif. Technol. 107 (2013) 238–244.
[1] R. Aravindhan, K.J. Sreeram, J.R. Rao, B.U. Nair, J. Gen. Appl. Microbiol. 53 [46] M. Baikousi, A.B. Bourlinos, A. Douvalis, T. Bakas, D.F. Anagnostopoulos, J.
(2007) 71–79. Tucek, K. Safarova, R. Zboril, M.A. Karakassides, Langmuir 28 (2012) 3918–
[2] M.R. Vanengelen, B.M. Peyton, M.R. Mormile, H.C. Pinkart, Biodegradation 19 3930.
(2008) 841–850. [47] V.K. Gupta, A. Rastogi, A. Nayak, J. Colloid Interface Sci. 342 (2010) 135–141.
[3] V. Chantawong, N.W. Harvey, V.N. Bashkin, Water Air Soil Pollut. 148 (2003) [48] Y.J. Jiang, X.Y. Yu, T. Luo, Y. Jia, J.H. Liu, X.J. Huang, J. Chem. Eng. Data 58 (2013)
111–125. 3142–3149.
[4] W.H. Organization, Guidelines for Drinking Water Quality, World Health [49] X. Lv, J. Xu, G. Jiang, J. Tang, X. Xu, J. Colloid Interface Sci. 369 (2012) 460–469.
Organization, Geneva, 2011. pp. 668. [50] J. Zhu, S. Wei, H. Gu, S.B. Rapole, Q. Wang, Z. Luo, N. Haldolaarachchige, D.P.
[5] L.N. Shi, X. Zhang, Z.L. Chen, Water Res. 45 (2011) 886–892. Young, Z. Guo, Environ. Sci. Technol. 46 (2012) 977–985.
[6] T. Burks, M. Avila, F. Akhtar, M. Göthelid, P.C. Lansåker, M.S. Toprak, M. [51] N. Pirrone, M.F. Ghorab, R. Djellabi, R. Messadi, Photo-reduction of hexavalent
Muhammed, A. Uheida, J. Colloid Interface Sci. 425 (2014) 36–43. chromium in aqueous solution in the presence of TiO2 as semiconductor
[7] V.K. Gupta, M. Gupta, S. Sharma, Water Res. 35 (2001) 1125–1134. catalyst, in: e3s Web of Conferences, vol. 1, 2013, p. 25008.
[8] C. Namasivayam, M.V. Sureshkumar, Bioresour. Technol. 99 (2008) 2218– [52] G.K. Zhang, Y.Y. Gao, Y.L. Zhang, Y.D. Guo, Environ. Sci. Technol. 44 (2010)
2225. 6384–6389.
[9] N.S. Reigosa, S.P. Rodriguez, J.C.N. Munoz, M.A. Estevez, M.J.F. Sanjurjo, E.A. [53] X. Wang, S.O. Pehkonen, A.K. Ray, Ind. Eng. Chem. Res. 43 (2004) 1665–1672.
Rodriguez, A.N. Delgado, Environ. Sci. Pollut. Res. 20 (2013) 2670–2678. [54] J. Matusik, T. Bajda, J. Colloid Interface Sci. 398 (2013) 74–81.
[10] S. Rao, H.S. Lade, T.A. Kadam, T.V. Ramana, S.K. Krishnamacharyulu, S. [55] X. Lv, X. Xue, G. Jiang, D. Wua, T. Sheng, H. Zhou, X. Xu, J. Colloid Interface Sci.
Deshmukh, G. Gyananath, J. Environ. Sci. Eng. 49 (2007) 255–258. 417 (2014) 51–59.
[11] C. Zhao, Q. Yang, W. Chen, B. Teng, Can. J. Microbiol. 58 (2012) 23–28. [56] S.C. Xua, S.S. Pana, Y. Xua, Y.Y. Luoa, Y.X. Zhanga, G.H. Lia, J. Hazard. Mater. 283
[12] E.I. Basaldella, P.G. Vazquez, F. Iucolano, D. Caputo, J. Colloid Interface Sci. 313 (2015) 7–13.
(2007) 574–578.
[13] S.A. Cavaco, S. Fernandes, M.M. Quina, L.M. Ferreira, J. Hazard. Mater. 144
(2007) 634–638.