Está en la página 1de 18

209

Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous


CeriumZirconium Oxide. Part I: Process Optimization by Response
Surface Methodology
C.A. Rivera Corredor1, M.. Gmez Garca1, I. Dobrosz-Gmez2,* - Grupo de Investigacin en
Procesos Reactivos Intensificados con Separacin y Materiales Avanzados (PRISMA). (1) Departamento de Ingeniera
Qumica, Facultad de Ingeniera y Arquitectura (2) Departamento de Fsica y Qumica, Facultad de Ciencias Exactas y
Naturales, Universidad Nacional de Colombia, Sede Manizales, Campus La Nubia, Km 9 va al Aeropuerto la Nubia,
Apartado Areo 127, Manizales, Caldas, Colombia.

(Received date: 16 September 2013; Accepted date: 8 February 2014)

ABSTRACT: Hydrous ceriumzirconium oxide (Ce0.25Zr0.75O2) was used for


Cr(VI) removal by adsorption. The response surface methodology (RSM) was
applied as a tool for the optimization of adsorption operational conditions
(temperature, pH and adsorbent dose) with the combined effects of the key
processing variables on the Cr removal efficiency (%RCr). The following
conditions were found to be optimal: pH = 2, temperature = 28 C, adsorbent
dose = 4 g l1. The %RCr obtained experimentally (92%) was in agreement with
the value predicted by RSM (88%). The adsorption process followed the
pseudo-second-order kinetic model and obeyed the Langmuir isotherm model.
Thermodynamic parameters (H, S and G) indicated the exothermic and
spontaneous nature of adsorption governed by physisorption interactions.
Hydrous Ce0.25Zr0.75O2 showed high adsorption capacity (25 mg g1), high
desorption efficiency (80%), reached adsorption equilibrium quickly
(30 minutes) and high chemical stability. The Cr(VI) adsorption on hydrous
Ce0.25Zr0.75O2 was not completely reversible. The reduction of Cr(VI) to Cr(III)
under acidic conditions was also responsible for Cr(VI) removal.

1. INTRODUCTION

Water pollution by chromium is of considerable concern owing to its widespread application in


various industries such as electroplating, leather tanning, mining, metal finishing, textile industries,
chromate preparation (Hua et al. 2012). In many developing countries, numerous industries are
operating in small and medium scales. These units can generate a considerable pollution load, which
is frequently discharged directly into the environment without any pre-treatment. Therefore, solving
the problem of water pollution by chromium is a challenging task. Chromium can exist in several
chemical forms displaying oxidation numbers from 0 to 6. However, its trivalent and hexavalent
forms are the most stable in the environment (Shriver et al. 1994). Trivalent chromium is an essential
element for human nutrition. Hexavalent chromium has been reported to be 500 times more toxic to
animals and humans than the trivalent one and it is a known carcinogenic and mutagenic agent (Sarin
and Pant 2006). The amount of Cr(VI) in industrial wastewaters ranges from 0.5 to 270,000 mg l1
(Patterson 1985). Over last few years in Colombia (South America), the average concentration of
Cr(VI) coming from the discharges of leather tanning and metal plating industry was approximately

*Author to whom all correspondence should be addressed. E-mail: idobrosz-gomez@unal.edu.co (I. Dobrosz-Gmez).
210 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

13 mg l1 and 62 mg l1, respectively (IDEAM 2013). According to the Colombian standard for
water quality (NTC-813-2010), the maximum permissible value for Cr(VI) in drinking water is 0.05
mg l1. Therefore, it is essential for the industries to treat their effluents to reduce the Cr(VI) to
acceptable levels before their disposal.
Different techniques have been proposed for the removal of heavy-metal ions from water and
wastewater, including but not limited to chemical precipitation, ion exchange, adsorption,
membrane filtration and electrochemical technologies (Hua et al. 2012). Among these methods,
adsorption offers flexibility in design and operation. The advantages of using adsorption are its
reversible nature, the possibility of regenerating the adsorbents used for multiple use (by
desorption) and economical recovery of Cr(VI). Therefore, selection of specific type of adsorbent,
which plays a dominant role in the adsorption process, is the key for developing proper adsorption
technology. Among the available adsorbents, nano-sized metal oxides (ferric, aluminium,
zirconium, titanium and cerium oxides) are classified as promising ones for heavy-metal removal
from aqueous systems (Alvares Rodrigues et al. 2010; Recillas et al. 2010; Hua et al. 2012). Their
large surface areas and high removal efficiency caused by the size-quantization effect are
considered as the most important properties for this kind of application.
Previously, we developed a systematic study on the effect of chemical composition of adsorbent
(Ce/Zr molar ratio), in the terms of its morphology, specific surface area, crystal structure, etc., on
Cr(VI) adsorption/desorption capacity (Vargas Ceballos et al. 2013). Hydrous Ce0.25Zr0.75O2
presented the highest Cr(VI) adsorption capacity from a series of Ce1xZrxO2 (x = 0, 0.25, 0.5,
0.75, 1) hydrous oxides and was characterized by physico-chemical stability, high exchange
capacity and good regeneration ability. It is well-known that the efficiency of adsorption process
depends on various physico-chemical parameters such as pH, temperature, adsorbent dose, initial
Cr(VI) concentration and adsorption time. However, only few authors have applied statistical
methods to investigate systematically the combination of parameters that provides optimal
adsorption conditions (Kalavathy et al. 2009; Dana and Sayari 2011; Han et al. 2013). As far as
we know, no similar study was performed for the optimization of Cr(VI) adsorption process on
hydrous metal oxide adsorbents.
Here, the response surface methodology (RSM) was applied as a tool for the optimization of
the operational conditions of Cr(VI) removal by adsorption on hydrous Ce0.25Zr0.75O2 with the
combined effects of the key processing variables (temperature, initial pH and adsorbent dose) on
the desired response (chromium removal efficiency, %RCr). The RSM is a statistical technique that
allows establishing the relationships between several independent variables and one or more
dependent ones. The optimization by the RSM involves the following steps: (i) the
implementation of the statistically designed experiments; (ii) the estimation of the coefficients of
a mathematical model using regression analysis technique; (iii) the prediction of the response and
(iv) the verification of the adequacy of the model. Among the available statistical design methods,
a multi-level BoxBehnken experimental design (BBD) was chosen for the purpose of response
optimization, as presented elsewhere (GilPavas et al. 2012). Moreover, kinetic and
thermodynamic studies of Cr(VI) adsorption on hydrous Ce0.25Zr0.75O2 were performed. Finally, the
elution test of Cr(VI) complemented the study.

2. EXPERIMENTAL ANALYSIS
2.1. Materials

The hydrous Ce0.25Zr0.75O2 was prepared by the co-precipitation method. The starting materials,
cerium(III) nitrate hexahydrate [Ce(NO3)3.6H2O, Merck, purity 99.9%] and zirconium(IV)
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 211

oxychloride octahydrate (ZrOCl2.8H2O, Merck, purity 99.9%), were dissolved in deionized water
(0.2M), and next their mixture was added dropwise to the excess of ammonium hydroxide
(approximately 28%, Sigma-Aldrich, purity 99.99%) solution (1:4) under constant stirring. After
ageing (12 hours), the precipitate was filtered, washed with deionized water several times until
complete removal of Cl and dried in an oven at 100 C until a constant weight was achieved.

2.2. Methods
2.2.1. Nitrogen Adsorption Measurements

Nitrogen adsorption/desorption isotherm at 196 C was measured using Sorptomatic 1900


apparatus (Carlo Erba). Before performing the measurement, the sample was degassed for 8 hours
at 90 C. The specific surface area, SBET, was calculated using the BET equation.

2.2.2. Scanning Electron MicroscopyEnergy-Dispersive Spectroscopy Studies

The morphology of hydrous Ce0.25Zr0.75O2 was determined using a scanning electron


microscopyenergy-dispersive spectroscopy (SEMEDS) technique. The SEM measurements were
performed by JSM-5910LV microscope (JEOL, Japan), using an acceleration voltage of 15 kV,
equipped with energy-dispersive spectrometer (Thermo Noran). Images were recorded at several
magnifications. The elemental analysis in the studied microarea of the oxide surface layer was
determined by the EDS method, based on the obtained characteristic X-ray spectra. The examined
sample was placed in a holder and before analyzing it was coated with a carbon monolayer, using
Cressington 208 HR system (Cressington Scientific Instruments Ltd., Watford, England), to reduce
the charge build-up on the sample.

2.2.3. X-Ray Diffraction Measurements

The crystalline structure of hydrous Ce0.25Zr0.75O2 was identified by X-ray diffraction (XRD).
Approximately 300 mg of sample, hand grounded in an agate mortar, was packed in the sample
holder. The XRD patterns were obtained at room temperature using a PANalyticals XPert PRO
MPD X-ray diffractometer (equipped with XCelerator) with nickel-filtered Cu-K radiation ( =
1.5406 ), operating at 40 kV and 30 mA. Data were collected in the 2 range of 2080, with
a step size of 0.0167 and step time of 10 seconds. Phase analysis was performed using XPert
HighScore Plus software and JCPDS-ICDD PDF-2 database (International Centre for Diffraction
Data-The Powder Diffraction File).

2.2.4. Determination of the Point of Zero Charge

The point of zero charge (ZPC) of the hydrous Ce0.25Zr0.75O2 was determined by mass titration
method (Di Paola et al. 2001), which involves finding the limiting pH value of the oxide/water
slurry as the oxide mass content is increased. Varying amounts of powders were added to water
and the resulting pH values were measured after 24 hours of equilibration. Typical values of
oxide/water by weight were 0.1%, 1%, 5%, 10% and 20%. The ZPC of hydrous Ce0.25Zr0.75O2
equals to 6.8 and indicates its amphoteric nature.

2.2.5. Batch Adsorption Studies

The stock solution containing 500 mg l1 of Cr(VI) was prepared by dissolving 141.1 mg of
potassium dichromate K2Cr2O7 (Panreac, purity 99.5%) in 100 ml of deionized water. Simulated
212 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

wastewaters with Cr(VI) concentration of 100 mg l1 were prepared by diluting the stock K2Cr2O7
standard solution with deionized water. Typically, batch adsorption experiments were carried out
in 50-ml conical flasks containing 10 ml of Cr(VI) solution and a given dose of hydrous
Ce0.25Zr0.75O2, which was placed in a thermostated shaker. The initial pH values of Cr(VI) solution
were adjusted to acidic or alkaline conditions by adding HCl (1M) or NaOH (1M). The pH was
measured using a pH meter. After the adsorption processes, the supernatant was separated from
the adsorbent by filtration. The concentration of Cr(VI) in the supernatant was estimated
spectrophotometrically (UVVIS) by the diphenylcarbazide method in acid solution (American
Public Health Association 2005), monitoring the absorbance at 540 nm. All the adsorption
experiments were carried out in duplicate. The relative deviations were in the range of 5.0%. The
amount of Cr(VI) adsorbed per unit mass of adsorbent at time t (minutes), qt (adsorption capacity,
mg g1) was evaluated using the following mass balance equation:

(C0 C t )V
qt = (1)
W

where C0 is the initial concentration of Cr(VI) in the solution (mg l1), Ct is the concentration of
Cr(VI) at time t of adsorption (mg l1), W is the mass of the adsorbent used (g) and V is the initial
volume of the Cr(VI) solution.
The Cr(VI) removal efficiency (%RCr) was determined using the following equation:

(C 0 C e )
%R cr = 100 (2)
C0

where C0 and Ce are the initial and equilibrium concentration of Cr(VI) (mg l1), respectively.

2.2.6. Experimental Design

The RSM was implemented to establish the effect of different operating factors on Cr(VI)
efficiency removal and to reveal its adsorption conditions on hydrous Ce0.25Zr0.75O2. Simulated
wastewaters containing Cr(VI) concentration of 100 mg l1 were used. A multi-factorial BBD was
defined to evaluate the interactive effects of adsorption variables and to optimize the adsorption
process. The following variables were selected for RSM: temperature, pH and adsorbent dose. The
experiments were programmed using Statgraphics 5.1 (Statistical Graphics Corp 19992004).
Thus, 15 tests with replica were randomly performed to avoid any systematic error. From
preliminary experiments (not presented here), three different levels (values) were chosen for each
of three variables. The independent variables and their levels, summarized in Table 1, were coded
according to equation (3).

TABLE 1. Process Variables and Their Levels for Experimental Design of Cr(VI) Adsorption on Hydrous
Ce0.25Zr0.75O2 Oxide
Symbol Factors Coded levels
1 0 1

A Temperature (C) 20 30 40
B pH 2 4.5 7
C Adsorbent dose (g l1) 2 3 4
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 213

(x i x pc )
Xi = (3)
x i

where Xi is the coded level, xi is the uncoded value, xpc corresponds to the uncoded value at the
central point and xi is the change value between (Montgomery 2005).
For the RSM, the experimental results were adjusted to a second-order multi-variable
polynomial [equation (4)]

3 3 3 3
Yi = 0 + i X i + ii Xii2 + ij Xi X j (4)
i =1 i =1 i =1 j =1

where Yi is the predicted response variable, 0 is the intercept coefficient, i is the linear term, ii
is the squared term, ij is the interaction term and Xi and Xj represent the coded independent
variables. The quality of this model and its prediction capacity were judged from the variation
coefficient, R2. Determination of the significant main and interaction effects of factors influencing
the removal efficiency was followed by analysis of variance (ANOVA). Therefore, the statistical
analysis was based on the ANOVA, Pareto diagram, BoxBehnken response surface plot and
variation coefficients.

2.2.7. Modelling of Adsorption Kinetics

For the kinetic studies, the experiments were performed under following conditions: initial
Cr(VI) concentration = 100 mg l1, pH = 2, hydrous Ce0.25Zr0.75O2 dose = 4 g l1, shaking time =
090 minutes, temperature = 20, 28 and 40 C.

2.2.8. Modelling of Adsorption Isotherms

To determine the adsorption isotherms, the following conditions of the adsorption experiments were
applied: hydrous Ce0.25Zr0.75O2 dose = 4 g l1; initial Cr(VI) concentrations = 25, 50 and 100 mg l1,
pH = 2, pre-determined shaking time = 60 minutes, temperature = 20, 28 and 40 C.

2.2.9. Determination of Thermodynamic Parameters

For the thermodynamic study, the experiments were performed under the following conditions:
hydrous Ce0.25Zr0.75O2 dose = 4 g l1, initial Cr(VI) concentration = 100 mg l1, pH = 2, pre-
determined shaking time = 60 minutes, temperature = 20, 28 and 40 C.

2.2.10. Batch Desorption Studies

To evaluate the reversibility of Cr(VI) adsorption and possibility of adsorbent regeneration, the
desorption studies were performed. The adsorbent used for the adsorption of Cr(VI) at optimized
conditions [Cr(VI) concentration = 100 mg l1, pH = 2, temperature = 28 C, pre-determined shaking
time = 60 minutes] was separated from the solution by filtration and washed gently with deionized
water to remove any unadsorbed Cr(VI). Next, the spent adsorbent was immersed in 10 ml of distilled
water at pH 5 and 12, which is adjusted by adding HCl (1M) or NaOH (1M), and agitated for a period
no longer than the equilibrium time (60 minutes). The amount of desorbed Cr(VI) was estimated as
before. The adsorption/desorption processes were repeated three times.
214 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

3. RESULTS AND DISCUSSION


3.1. Adsorbent Textural Properties

Nitrogen adsorptiondesorption isotherms together with corresponding textural properties (BET


surface area, pore size and pore volume) for hydrous Ce0.25Zr0.75O2 are presented in Figure 1(a). The
observed isotherm shows a fairly rapid rise in adsorbed quantity of N2 with increasing pressure up
to saturation and can be classified as Type I (Khalfaoui et al. 2003). It indicates a microporous
character of adsorbent at which adsorption is limited to the completion of a single monolayer of
adsorbate on its surface. The BET surface area and pore volume of the sample are 250 m2 g1 and
0.11 cm3 g1, respectively.
Figure 1(b) presents the XRD pattern of hydrous Ce0.25Zr0.75O2, which corresponds to a
tetragonal-type structure (JCPDS: 00-038-1437), with space group P42/nmc, if any. Hydrous
Ce0.25Zr0.75O2 is characterized by relatively low degree of crystallinity. However, no phase
segregation was detected suggesting that ZrO2 is incorporated into the CeO2 lattice to form a
solid solution.
The morphology of hydrous Ce0.25Zr0.75O2 is illustrated in Figure 1(c). The sample is highly
uniform. The agglomerations of regular and plated shape grains with dimension higher than 20 m
are found. The content of elements in the studied microarea of the oxide surface layer was
confirmed by the EDS method.

(a) (b)
100
XRD pattern
Volume of N2, cm3 .g-1 STP

90 JCPDS reference

80
Intensity, A. U.

70

Adsorption
60 Desorption

50 BET surface area = 250 m2. g-1


BJH pore size = 2.7 nm
Pore volume = 0.11 cm3.g-1
40
0.0 0.2 0.4 0.6 0.8 1.0 20 30 40 50 60 70 80
Relative pressure, P/P0 2q, degree
(c)
500x 1000x 2500x

Figure 1. Textural properties of hydrous Ce0.25Zr0.75O2: (a) N2 adsorptiondesorption isotherms; (b) XRD pattern; JCPDS:
00-038-1437 is included as reference, (c) secondary electron photomicrographs at different magnifications: 500, 1000
and 2500.
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 215

TABLE 2. Experimental Results of the %RCr for the Three Variables and Levels
Experiment Temperature, C pH value Adsorbent Cr(VI) removal
dose, g l1 efficiency (%RCr)

1 40 7 3 7.3
2 40 4, 5 4 34.2
3 40 4, 5 2 17.7
4 20 7 3 7.4
5 30 2 4 86.3
6 30 4, 5 3 28.4
7 20 4, 5 4 33.0
8 30 7 2 2.7
9 40 2 3 74.6
10 30 4, 5 3 28.4
11 20 4, 5 2 1.7
12 20 2 3 72.9
13 30 7 4 7.9
14 30 4, 5 3 28.4
15 30 2 2 59.4

3.2. Development of Regression Model Equation

To examine the combined effect of three independent process parameters on Cr(VI) efficiency
removal (%RCr), 15 experiments were performed. The experimental design is given in Table 2,
together with the corresponding experimental data.
Regression analysis was performed to fit the response function [Cr(VI) efficiency removal,
%RCr]. The developed second-order polynomial equation represents responses as functions of
temperature (A), pH (B) and adsorbent dose (C). An empirical relationship between the response
and the input test variables in coded units can be expressed by the following equation:

%RCr = 26.1985 + 3.0085A 27.639B + 55.5275C 0.026375A2


0.018AB 0.37AC + 2.366B2 2.17BC 4.1125C2 (5)

Equation (5) describes how Cr(VI) adsorption on hydrous Ce0.25Zr0.75O2 was affected by the
individual variables and/or their double interactions. The removal of Cr(VI) was linear with
respect to temperature, pH and adsorbent dose and also quadratic with respect to temperature, pH
and adsorbent dose. This indicates that there is a temperature-to-temperature, temperature-to-pH,
temperature-to-adsorbent dose, pH-to-pH, pH-to-adsorbent dose and adsorbent dose-to-adsorbent
dose interaction that can affect Cr(VI) removal.

3.3. ANOVA

The ANOVA was used to determine the significant main and interaction effects of factors
influencing the Cr(VI) adsorption capacity. The ANOVA results are presented in Table 3.
The ANOVA consists of classifying and cross-classifying statistical results. Fisher F test,
defined as the ratio of respective mean-square effect and mean-square error, was used to evaluate
the presence of a significant difference from control response and to calculate standard errors. The
216 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

TABLE 3. ANOVA Results of Response Surface Quadratic Model According to %RCr


Factor Sum of squares Degree of freedom Mean square F value P value

A 44.18 1 44.18 3.57 0.1175


B 8971.3 1 8971.3 724.75 0.0000
C 798.001 1 798.001 64.47 0.0005
AA 25.6852 1 25.6852 2.07 0.2093
AB 0.81 1 0.81 0.07 0.8083
AC 54.76 1 54.76 4.42 0.0894
BB 807.398 1 807.398 65.23 0.0005
BC 117.722 1 117.722 9.51 0.0273
CC 62.4467 1 62.4467 5.04 0.0746
Error total 61.8925 5 12.3785
Total (corr.) 11,001.6 14
Note: R2 = 0.9944; R2adj = 0.9842.

biggest the magnitude of F value, the most significant is the corresponding coefficient. The P
values were used to identify experimental parameters that have a statistically significant influence
on particular response. If P value is lower than 0.05, it is statistically significant with the 95%
confidence level (Montgomery 2005). According to ANOVA results, one can see that all terms in
the regression model are not equally important. Only four of them [pH (B), adsorbent dose (C)
and their interactions, namely, BB and BC] presented P values lower than 0.05 (Table 3), which
implies that they have a truthful effect on %RCr, with a confidence interval of 95% (Montgomery
2005). By contrast, the effect of the temperature on Cr(VI) removal efficiency is not as significant
(in the analyzed range). However, it can have some importance and should not be ignored.
The quality of the developed model was evaluated based on the variation coefficient (R2) and
standard deviation value. The closer the R2 value to unity and lower value of standard deviation,
more accurate the response could be predicted by the model. The R2 value for equation (5) was
found to be 0.9943, with standard deviation equalled to 3.51831, indicating that 99.43% of the
total variation in the Cr(VI) removal efficiency was attributed to the studied experimental
variables. Moreover, the value of predicted R2 (0.9944) is in very good agreement with the value
of adjusted R2 (0.9842). The R2 value close to unity indicates that there is a good agreement
between the experimental data and those predicted by the model adsorption efficiency. It is clear
from Figure 2 that the regression equation follows the experimental results with a good accuracy.
Therefore, the model can be considered as a good fit.

3.4. Pareto Analysis

The Pareto analysis was used to identify factors that have the greatest cumulative effect on Cr(VI)
removal efficiency, and thus to screen out the less significant ones. A Pareto diagram is a series of
bars whose heights reflect the frequency or impact of each factor. The bars are arranged in
descending order of heights from left to right. Therefore, the factors represented by the tall bars
are relatively more significant. Here, the Pareto analysis was also carried out to determine the
percentage effect of each factor according to the following equation (Zarei et al. 2010):

b2
Pi = i 2 100 (i 0) (6)
bi
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 217

100

Predicted value, % RCr 80

60

40

20

0
0 20 40 60 80 100
Observed value, % RCr

Figure 2. Comparison between the experimental and model-predicted Cr(VI) removal efficiency on hydrous
Ce0.25Zr0.75O2.

B: pH 82.76%

BB 7.27%

C: Adsorbent dose 7.27%

BC 1.16%

CC 0.45%

AC 0.45%

A: Temperature 0.37%

AA 0.26% +

AB 0.001% -

0 5 10 15 20 25 30
Standardized effect

Figure 3. Pareto diagram for Cr(VI) removal efficiency by adsorption on hydrous Ce0.25Zr0.75O2.

Thus, statistically important factors correspond to all those values that overpass the inner
vertical line (Figure 3). The vertical line corresponds to the t value in the t student distribution,
with a 95% confidence and for 14 degrees of freedom. Next, this value is compared with the
values of each effect and interaction of analyzed factor. The comparison defines the statistical
significance of each factor in the analyzed process. Therefore, factors that have an influence on
the adsorption process are pH (B), adsorbent dose (C) and their interactionsBB and BC. It
could be concluded that the %RCr is directly proportional (+) to the adsorbent dose (C) and BB
interaction and inversely proportional () to the pH value (B) and BC interaction. However, the
pH can be considered as the most dominating factor during Cr(VI) removal (Figure 3).
218 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

(a) (b)
Cr (VI) removal efficiency, %

Cr (VI) removal efficiency, %


90
90
80 80
70 70
60
50 60
40
50
30
20 40
10 20
0 7 25 4
6 3.5
20 5 30
25 4 3
30 3 pH 35 2.5
Temperature, C
35
40 2 Temperature, C 40 2 Adsorbent dose, g/l

(c)

(d)
Cr (VI) removal efficiency, %

7
100

80 6
%RCr= 20
60 5
pH

40 4 40

20 3 60
4
0 3.5 80
2 3 2
3 20 24 28 32 36 40
4 2.5
5 Adsorbent dose, g/l
pH 6
7 2 Temperature, C

Figure 4. Three-dimensional surface-response plots for the interactive effect of (a) temperature and pH (constant adsorbent
dose = 4 g l1); (b) adsorbent dose and temperature (constant pH = 2) and (c) pH and adsorbent dose (constant temperature
= 28 C) on the Cr(VI) removal efficiency by adsorption on hydrous Ce0.25Zr0.75O2. (d) Contour plot of Cr(VI) removal
efficiency. The + sign indicates the optimum values of temperature and pH (constant adsorbent dose = 4 g l1).

3.5. Estimation of Quantitative Effects of the Factors

Three-dimensional BoxBehnken response surface plots and contour plot were used to study the
effect of all the factors on the response variable (Figure 4). These type of plots summarize the effect
of two variables on the %RCr, considering that the third one was kept constant. One can see from
Figures 4(a and c) that the %RCr slightly increases with an increase in adsorption temperature (with
maximum at 28 C), suggesting that the adsorption process is an exothermic one. Moreover, a
significant effect of pH on the %RCr can be observed [Figure 4(a)]. The decrease in pH value from
7 to 2 resulted in an increase in %RCr from approximately 10% to 88%, at 28 C. It is well-known
that anion-exchange capacity is strongly related to both the pH of the solution and the surface
chemistry of the solid. Therefore, the observed effect of pH on %RCr can be directly related to both
the amphoteric nature of hydrous Ce0.25Zr0.75O2 (ZPC = 6.8) and possible different forms of
chromium in the solution.
According to the literature (Anirudhan et al. 2009; Yusof and Malek 2009; Alvares Rodrigues et al.
2010), Cr(VI) speciation is affected by the solution pH through the following equilibria:
HCrO4 CrO24 + H+ , pK a = 5.9 (7)
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 219

H 2 CrO 4 HCrO4 + H+ , pK a = 4.1 (8)

Cr2 O27 + H 2 O 2HCrO4 , pK a = 2.2 (9)

Considering the ZPC of hydrous Ce0.25Zr0.75O2, it can be stated that at pH lower than 6.8, the
adsorbent surface becomes charged positively and it can attract negatively charged monovalent
HCrO4 ions present in the solution. The strong electrostatic attraction of the dichromate anions
to the positively charge particles of adsorbent (H+ ions) can be the first step of the adsorption
process. If the solution has a pH much lower than the ZPC value, a chemisorption process can
also proceed through an anionic exchange reaction (Liu et al. 2008; Jain et al. 2009; Alvares
Rodrigues et al. 2010). By contrast, under increasing solution pH, the adsorbent surface becomes
negatively charged and therefore the %RCr decreases, as a result of electrostatic repulsion.
Consequently, a maximum %RCr (temperature = 28 C and adsorbent dose = 4 g l1) of 88% was
observed at pH equalled to 2. Nevertheless, it should be considered that under acidic conditions
and in the presence of oxygen vacancies, step edges (Campbell and Peden 2005;
Mukhopadhyaya et al. 2007) and small amounts of Ce3+ on the surface remaining from the
synthesis reaction, reduction of Cr(VI) to Cr(III) can contribute to its elimination (Recillas et al.
2010) [equation (10)].

Cr2 O27 + 14H+ + 6e 2Cr23+ + 7H 2 O, E 0 = 1.33V (10)

Figures 4(b and c) show that the increase in both adsorption temperature and adsorbent dose, in
the studied range, is favourable to the %RCr on hydrous Ce0.25Zr0.75O2. The slight increase in %RCr
from approximately 86% to 88% with an increase in temperature from 20 to 28 C was observed.
Any further increase in temperature up to 40 C resulted in a slight decrease in %RCr (84%). It can
be suggested that higher temperature provides a driving force to overcome mass-transfer
resistance of Cr(VI) between the aqueous and solid phases. By contrast, the increasing adsorbent
dose led to an increase in the number of collisions between Cr(VI) ions and adsorbent, enhancing
the adsorption process. Moreover, the adsorptive surface area increases and more active binding
sites on the surface of adsorbent appear. The increase in %RCr from approximately 56% to 88%
with an increase in adsorbent dose from 2 to 4 g l1 was noticed at pH 2. The summary of the effect
of temperature and pH on the Cr(VI) removal efficiency is presented in Figure 4(d). The + sign
indicates the optimum values of temperature and pH (adsorbent dose = 4 g l1).
In conclusion, the RSM optimization led to obtain the following optimal conditions for Cr(VI)
adsorption on hydrous Ce0.25Zr0.75O2: pH = 2, temperature = 28 C, adsorbent dose = 4 g l1 and the
predicted %RCr of 88%. Next, two batch experiments were performed, at optimized condition, to
check the developed mathematical model. The mean value of %RCr obtained experimentally was
approximately 92% and is considered to be very close to the predicted value, indicating the
adequacy of the proposed model of Cr(VI) adsorption.

3.6. Adsorption Kinetics

At the optimized adsorbent dose of 4 g l1, pH of 2, and at temperatures of 20, 28 and 40 C, a


kinetic analysis was developed by monitoring the evolution of Cr(VI) adsorption capacity, as a
function of time [Figure 5(a)].
220 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

(a) (b)
Cr (VI) adsorption capacity, mg g-1

25 4.5
4.0
20
3.5

t/q, min.g.mg g-1


3.0
15
2.5

10 2.0
1.5
5 40 C 1.0 40 C
28 C 28 C
20 C 0.5 20 C
0
0.0
0 20 40 60 80 100 0 20 40 60 80 100
t, min t, min

(c)
-2.0
-2.2 y = -7769,85x+23,0329
-2.4 R2 = 0,99812

-2.6
-2.8
Ln(K2)

-3.0
-3.2
-3.4
-3.6
-3.8
-4.0
0,00325 0,00330 0,00335 0,00340 0,00345 0,00350
1/T, K-1

Figure 5. Kinetics plots for Cr(VI) adsorption on hydrous Ce0.25Zr0.75O2: (a) variation of Cr(VI) adsorption capacity with
adsorption time, at various temperatures; (b) pseudo-second-order kinetics plots; (c) determination of the adsorption
activation energy (E).

Several kinetics models were used in the literature to explain the mechanism of adsorption
processes. In this study, the kinetics of Cr(VI) adsorption on hydrous Ce0.25Zr0.75O2 were analyzed
by applying the pseudo-first-order and pseudo-second-order kinetics models. A pseudo-first-order
kinetics is given by the Lagergren equation
K1
log(qe q) = logq e t (11)
2.303
where qe is the amount of solute adsorbed at equilibrium per unit weight of adsorbent (adsorption
capacity at equilibrium, mg g1), q is the amount of solute adsorbed at any time t (minutes) per
unit weight of adsorbent (adsorption capacity at any time t, mg g1) and K1 is the first-order
adsorption rate constant (minute1). The parameters K1 and qe can be calculated from the slope of
log(qe q) versus t (minutes).
By contrast, the pseudo-second-order equation based on equilibrium adsorption is expressed as
follows:

(12)
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 221

TABLE 4. Comparison of the Pseudo-First-Order and Pseudo-Second-Order Kinetics Adsorption Rate


Constants and Correlation Coefficients for the Adsorption of Cr(VI) on Hydrous Ce0.25Zr0.75O2 at Different
Temperatures
Kinetic model Parameter Temperature, C
20 28 40
2
Pseudo-first order R 0.9830 0.9134 0.9539
K1 (minute1) 0.0385 0.0405 0.0478
Pseudo-second order R2 0.9998 0.9986 0.9998
K2 (g mg1 minute1) 0.0156 0.0200 0.0303

where K2 is the pseudo-second-order rate constant (g mg1 minute1). The equation constants can
be determined plotting t/q (minute g mg1) versus t (minutes).
The convergence between the experimental data and those predicted by the model, expressed
as linear regression correlation coefficient value (R2), is a criterion of model selection. Based on
the value of R2, which value is closer to 1 (Table 4), one can see that the adsorption process
follows the pseudo-second-order kinetic model [Figure 5(b)].
Finally, the adsorption activation energy (E, kJ mol1), using the Arrhenius equation, [equation
(13)] was determined
K 2 = K 0 e E/RT (13)

where K0 is the temperature-independent factor (g mg1 minute1), R is the gas constant (8.314 J
mol1 K1) and T is Cr(VI) solution temperature (K).
Figure 5(c) shows the Arrhenius plot of the adsorption rate constants versus 1/temperature.
From the slope and intercept of the line, K0 is determined to be 1 1010 g mg1 minute1 and E/R
to be 7769.85 K.

3.7. Adsorption Isotherms

The relationship between adsorbate concentrations in the solid and aqueous phases at equilibrium
can be determined based on sorption isotherm models. The Langmuir and Freundlich models are
the most widely used to describe the equilibrium data of adsorption. According to the Langmuir
model, adsorption occurs uniformly on the active sites of the adsorbent, and once an adsorbate
occupies a site, no further adsorption can take place at this site (monolayer adsorption on the
active sites of the adsorbent with constant heat of adsorption for all sites). This model is given by
the following equation:

Ce 1 C
= + e (14)
qe b Q0 Q0

where Q0 (maximum sorption capacity corresponding to complete monolayer coverage, mg g1)


and b (equilibrium constant, l mg1) are the Langmuir model parameters, Ce is the equilibrium
solution concentration of Cr(VI) in the aqueous phase (mg l1) and qe is the equilibrium amount
of Cr(VI) adsorbed onto the adsorbent (adsorption capacity at equilibrium, mg g1). The values of
the Langmuir model parameters (Q0 and b) can be calculated from the slope and intercept of the
plot Ce/qe (g l1) versus Ce (mg l1).
222 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

By contrast, the Freundlich isotherm is an empirical model, based on the adsorption on


heterogonous surface, and is given by the following equation:

1
logq e = logK + logC e (15)
n

where K and n are Freundlich constants, which represent adsorption capacity (mg g1) and
adsorption intensity (g l1), respectively. It presents the relationship between equilibrium liquid
and solid-phase capacity based on the multi-layer adsorption (adsorption sites are distributed
exponentially with respect to the heat of adsorption). The values of K and n can be calculated from
the slope and intercept of the plot logqe versus logCe.
Table 5 presents constant parameters and correlation coefficients calculated for Cr(VI)
adsorption on hydrous Ce0.25Zr0.75O2 using different adsorption models, at different temperatures.
In this study, the Langmuir model fits (Figure 6) the adsorption data better than the Freundlich
model (Table 5, value of R2 was higher for the Langmuir isotherm). It indicates the applicability
of monolayer coverage of the hydrous Ce0.25Zr0.75O2 surface by Cr(VI), probably due to its high
specific surface area and homogeneous distribution of active sites on adsorbent surface. Moreover,

TABLE 5. Langmuir and Freundlich Isotherm Constants for Adsorption of Cr(VI) on Hydrous Ce0.25Zr0.75O2
at Different Temperatures
Adsorption model Parameters Temperature, C
20 28 40
2
Langmuir R 0.9999 0.999 0.9986
Q0 (mg g1) 22.52 25.19 25.91
b (l mg1) 1.58 1.31 1.25
Freundlich R2 0.9393 0.9772 0.9815
K (mg g1) 10.87 11.74 11.89
n (g l1) 3.38 2.65 2.56

0.6

0.5

0.4
Ce/qe, gl-1

0.3

0.2

40 C
28 C
0.1
20 C

0.0
0 2 4 6 8 10 12
Ce, mg l-1

Figure 6. Langmuir plots for the adsorption of Cr(VI) on hydrous Ce0.25Zr0.75O2 at different temperatures.
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 223

TABLE 6. Summary of Cr(VI) Adsorption Capacity (Q0) and Time Required to Reach Equilibrium, Derived
from Langmuir Equation, on Different Hydrous Adsorbent Materials
Hydrous adsorbent Q0, mg g1 Time required Initial Cr(VI) Reference
to reach concentration,
equilibrium, minutes mg l1

Ce0.25Zr0.75O2 25 30 100 This study


Al2O3 78 1440 150 Alvarez-Ayuso et al. (2007)
ZrO2 66 60 100 Alvares Rodrigues et al. (2010)
Fe2O3 36 90 50 Goswami et al. (2006)
-Fe2O3 19 30 100 Hu et al. (2006)
TiO2 14 40 100 Debnath and Ghosh (2008)
TiO2 30 30 10 Ghosh et al. (2003)
CeO2 7 300 50 Zhong et al. (2007)
SnO2 3 90 50 Goswami and Ghosh (2005)
Activated alumina 7 90 50 Mor et al. (2007)
Activated charcoal 13 120 50 Mor et al. (2007)
Activated carbon 8 180 50 Liu et al. (2007)
Carbon nanotubes 4 240 1 Hu et al. (2009)

the results show that the adsorption capacity increased from 22.52 to 25.91 mg g1 when the
temperature increased from 20 to 40 C (Table 5).
Table 6 presents the comparison of Cr(VI) adsorption capacity on hydrous Ce0.25Zr0.75O2, derived
from the Langmuir equation, with those reported in the open literature for different hydrous
adsorbent materials. One can see that of the 12 compared adsorbents, only two, Al2O3 and ZrO2,
present significantly higher Cr(VI) adsorption capacity than that of hydrous Ce0.25Zr0.75O2.
However, both of them are characterized by much longer equilibrium time necessary for Cr(VI)
adsorption. For example, adsorbents such as CeO2 (Zhong et al. 2007), activated carbon (Liu et
al. 2007) and carbon nanotubes (Hu et al. 2009) can reach equilibrium in more than 3 hours and
present Cr(VI) adsorption capacity significantly lower than that of hydrous Ce0.25Zr0.75O2.
Moreover, it should be noted that most of the studied hydrous oxides were applied for the
adsorption of Cr(VI) from aqueous solution with initial concentrations of Cr(VI) much lower than
that used in this study. In general, under increasing Cr(VI) initial concentration, the removal
percentage of Cr(VI) decreases. We believe that the obtained value of Cr(VI) removal by hydrous
Ce0.25Zr0.75O2 appears high enough to consider its application as an adsorbent for purification of
industrial wastewater from chromium.

3.8. Thermodynamic Study

Thermodynamic parameters such as Gibbs free energy (G), enthalpy (H) and entropy (S)
were calculated using the following equations:
qe
b= = Kc (16)
Ce

G = RTlnKc (17)

S0 H 0
lnK c = (18)
R RT
224 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

H0 = - 9,04 kJ mol-1
S0 = 0,03 kJ mol-1K-1
2
R = 0,9835

Figure 7. Graphical determination of H and S of Cr(VI) adsorption on hydrous Ce0.25Zr0.75O2.

where b is the equilibrium constant, Ce is the equilibrium solution concentration (mg l1) of Cr(VI) in
the aqueous phase and qe is the equilibrium amount of Cr(VI) adsorbed on the adsorbent (mg g1).
The values for H and S were determined from the slope and intercept of the Arrhenius plot of Kc
versus T (Figure 7).
The negative value of H confirms the exothermic nature of adsorption process and the
negative values of G indicate the feasibility of the process and its spontaneous nature.
Considering that the H value was lower than (-40) kJ mol1, it can be stated that Cr(VI) removal
on hydrous Ce0.25Zr0.75O2 occurs by physisorption process. The decrease in G with the increasing
temperature indicates more efficient adsorption at higher temperature, due to the increase in driving
force of Cr(VI) adsorption at these conditions. Finally, the positive entropy of adsorption (S)
reflects the affinity of adsorbent material towards Cr(VI) and suggests the increased randomness at
the solidliquid interface during the adsorption of Cr(VI) on hydrous Ce0.25Zr0.75O2 surface.

3.9. Desorption Studies

The results of desorption studies showed that the amount of Cr(VI) extracted from the hydrous
Ce0.25Zr0.75O2 strongly depends on pH of desorption solution. At pH 5, only 8% of Cr(VI) was
desorbed. However, the increase in pH value up to 12 resulted in 80% desorption of Cr(VI). This
strong effect of pH on desorption efficiency indicates that physisorption, ion-exchange and
chemisorption mechanisms are probably involved in the analyzed adsorption process. Therefore,
change in charge of the adsorbent surface (at pH = 12) resulted in desorption of Cr(VI) ions
adsorbed by physisorption and ion exchange. However, the chemisorbed ions were not removed
from the adsorbent at these conditions. According to the literature (Namasivayam and Sangeetha
2004), chemisorption involves zirconiumCr(VI) complex formation
Because hydrous Ce0.25Zr0.75O2 is characterized by the presence of high quantity of step edges
[Figure 1(c)] and probably small amounts of Ce3+ on the surface remaining from its synthesis
reaction, chromium elimination by its reduction should be also considered. Probably also because
Adsorptive Removal of Cr(VI) from Aqueous Solution on Hydrous Ce0.25Zr0.75O2 225

of this reason, its complete regeneration was not possible. The results also showed that hydrous
Ce0.25Zr0.75O2 can be recovered by Cr(VI) desorption with distilled water at pH = 12 with an
efficiency of 75% in the third cycle.

4. CONCLUSIONS

This study was focused on the adsorption of Cr(VI) on hydrous Ce0.25Zr0.75O2 from aqueous
solution. The RSM was applied as a tool for the optimization of the operational conditions such
as temperature, initial pH and adsorbent dose. The most significant results of the study can be
summarized as follows:
The following conditions were found to be optimal for Cr(VI) adsorption on hydrous
Ce0.25Zr0.75O2: pH = 2, temperature = 28 C, adsorbent dose = 4 g l1. The %RCr obtained
experimentally (92%) was in agreement with the value predicted by the model (88%).
Adsorption process followed the pseudo-second-order kinetic model and obeyed the
Langmuir isotherm model.
Thermodynamic parameters, H, S and G indicated exothermic and spontaneous nature
of the studied adsorption process, governed by physisorption interactions.
Hydrous Ce0.25Zr0.75O2 showed high adsorption (25 mg g1) capacity and desorption (80%)
efficiency, short time necessary to reach adsorption equilibrium (30 minutes) and chemical
stability.
Cr(VI) adsorption on hydrous Ce0.25Zr0.75O2 was not completely reversible. The reduction of
Cr(VI) to Cr(III), under acidic conditions, was also considered as supplier of Cr(VI) removal.
Hydrous Ce0.25Zr0.75O2 can find practical application in purification of industrial wastewater
from chromium due to its high adsorption capacity and desorption efficiency, short time
necessary to reach adsorption equilibrium and chemical stability.

ACKNOWLEDGEMENTS

The financial support of DIMA (Programa Para la Financiacin de Semilleros de Investigacin en


Pregrado en la Universidad Nacional de Colombia, Sede Manizales, 2012, Cdigo 15787) is
gratefully acknowledged. The authors are grateful to the staff of the Institute of General and
Ecological Chemistry of Lodz University of Technology (Poland), under the supervision of
Professor Jacek M. Rynkowski, for their help in the determination of adsorbent textural properties.

REFERENCES

Alvares Rodrigues, L., Maschio, L.J., Evangelista de Silva, R. and Caetano Pinto de Silva, M.L. (2010)
J. Hazard. Mater. 173, 630.
Alvarez-Ayuso, E., Garcia-Sanchez, A. and Querol, X., (2007) J. Hazard. Mater. 142, 191.
American Public Health Association (2005) Standard Methods for the Examination of Water and Wastewater,
21st Ed., Washington, DC, p. 367.
Anirudhan, T.S., Jalajamony, S. and Suchithra, P.S. (2009) Colloids Surf. A. 335, 107.
Campbell, C.T. and Peden, C.H.F. (2005) Science. 309, 752.
Dana, E. and Sayari, A. (2011) Chem. Eng. J. 167, 91.
Debnath, S. and Ghosh, U.C. (2008) J. Chem. Thermodyn. 40, 67.
226 C.A. Rivera Corredor et al./Adsorption Science & Technology Vol. 32 No. 2&3 2014

Di Paola, A., Ikeda, S., Marc, G., Ohtani, B. and Palmisano, L. (2001) Inter. J. Photoenergy. 3, 171.
Ghosh, U.C., Desgupta, M., Debnath, S. and Bhat, S.C. (2003) Water Air Soil Pollut. 143, 245.
GilPavas, E., Dobrosz-Gmez, I. and Gmez-Garca, M.. (2012) Water Sci. Technol. 65, 1795.
Goswami, S., Bhat, S.C. and Ghosh, U.C. (2006) Water. Environ. Res. 78, 986.
Goswami, S. and Ghosh, U.C. (2005) Water SA. 31, 597.
Han, C., Pu, H., Li, H., Deng, L., Huang, S., He, S. and Luo, Y. (2013) J. Hazard. Mater. 254255, 301.
Hu, J., Chen, G.H. and Lo, I.M.C. (2006) J. Environ. Eng. 132, 709.
Hu, J., Chen, C., Zhu, X. and Wang, X. (2009) J. Hazard. Mater. 162, 1542.
Hua, M., Zhang, S., Pan, B., Zhang, W., Lv, L. and Zhang, Q. (2012) J. Hazard. Mater. 211212, 317 (and
references therein).
IDEAM home page (2013). https://documentacion.ideam.gov.co/openbiblio/Bvirtual/publicaciones/
publicaciones2.html (Last accessed date: June 12, 2013).
Jain, M., Garga, V.K. and Kadirvelu, K. (2009) J. Hazard. Mater. 162, 365.
Kalavathy, H.M., Regupathi, I., Pillai, M.G. and Miranda, L.R. (2009) Colloids Surf. B 70, 35.
Khalfaoui, M., Knani, S., Hachicha, M. and Lamine, B. (2003) J. Colloid Interface Sci. 263, 350.
Liu, H., Sun, X., Yin, C. and Hu, C., (2008) J. Hazard. Mater. 151, 616.
Liu, S.X., Chen, X., Chen, X.Y., Liu, Z.F. and Wang, H.L. (2007) J. Hazard. Mater. 141, 315.
Montgomery, D. (2005) Design and Analysis of Experiments, 5th Ed., Wiley and Sons, Hoboken, NJ.
Mor, S., Ravindra, K. and Bishnoi, N.R. (2007) Bioresour. Technol. 98, 954.
Mukhopadhyaya, B., Sundq, J. and Schmitzc, R.J. (2007) J. Environ. Manage. 82, 66.
Namasivayam, C. and Sangeetha, D. (2004) J. Colloid Interface Sci. 280, 359.
Patterson, J.W. (1985) Industrial Wastewater Treatment Technology, 2nd Ed., Butterworth-Heinemann,
London, UK.
Recillas, S., Coln, J., Casals, E., Gonzles, E., Puntes, V., Snchez, A. and Font, X. (2010) J. Hazard. Mater.
184, 425.
Sarin, V. and Pant, K.K. (2006) Bioresour. Technol. 97, 15.
Shriver, D.F., Atkins, P.W. and Langford, C.H. (1994) Inorganic Chemistry, 2nd Ed., Oxford University
Press, Oxford, UK.
Tel, H., Altas, Y. and Taner, M.S. (2004) J. Hazard. Mater. B 112, 225.
Vargas Ceballos, A.D., Gmez Garca, M.. and Dobrosz-Gmez I. (2013) Proccedings of II Workshop on
Adsorption, Catalysis and Porous Materials, Bogota-Colombia, pp 132-133.
Yusof, Y.A.M. and Malek, N.A.N.N. (2009) J. Hazard. Mater. 162, 1019.
Zarei, M., Niaei, A., Salari, D. and Khataee, A. (2010) J. Hazard. Mater. 173, 544.
Zhong, L.-S., Hu, J.-S., Cao, A.-M., Liu, Q., Song, W.-G. and Wan, L.-J. (2007) Chem. Mater. 19, 1648.

También podría gustarte