Environ. Sci. Technol. 2005, 39, 6246-6250
Bisphenol A Removal from Water by Activated Carbon. Effects of Carbon Characteristics and Solution Chemistry I. BAUTISTA-TOLEDO, M . A . F E R R O - G A R C IÄ A , J. RIVERA-UTRILLA,* C. MORENO-CASTILLA, AND F . J . V E G A S F E R N AÄ N D E Z Departamento de Quı´mica Inorga´nica, Facultad de Ciencias, Universidad de Granada, 18071 Granada, Spain
The present study aimed to analyze the behavior of different activated carbons in the adsorption and removal of bisphenol A (2-2-bis-4-hydroxypheniyl propane) from aqueous solutions in order to identify the parameters that determine this process. Two commercial activated carbons and one prepared in our laboratory from almond shells were used; they were texturally and chemically characterized, obtaining the surface area, pore size distribution, mineral matter content, elemental analysis, oxygen surface groups, and pH of the point of zero charge (pHPZC), among other parameters. Adsorption isotherms of bisphenol A and adsorption capacities were obtained. The capacity of the carbons to remove bisphenol A was related to their characteristics. Thus, the adsorption of bisphenol A on activated carbon fundamentally depends on the chemical nature of the carbon surface and the pH of the solution. The most favorable experimental conditions for this process are those in which the net charge density of the carbon is zero and the bisphenol A is in molecular form. Under these conditions, the adsorbent-adsorbate interactions that govern the adsorption mechanism are enhanced. Influences of the mineral matter present in the carbon samples and the solution chemistry (pH and ionic strength) were also analyzed. The presence of mineral matter in carbons reduces their adsorption capacity because of the hydrophilic nature of the matter. The presence of electrolytes in the solution favor the adsorption process because of the screening effect produced between the positively charged carbon surface and the bisphenol A molecules, with a resulting increase in adsorbent-adsorbate interactions.
1. Introduction The adsorption of organic contaminants by activated carbons is one of the most effective and widely used methods to purify waters (1). However, most research in this field has ignored the influence of the surface chemistry of the carbon on the extraction of contaminants from water. Thus, little is known about the different mechanisms that control these processes. In-depth study of the parameters that influence a given * Corresponding author phone:34-958-248523; fax: 34-958-248526; e-mail:
[email protected]. 6246
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
adsorption process could contribute to enhancing the effectiveness of the extraction of a given contaminant (2-4). It is suspected that some chemicals function as hormones in the living body and may pose risks to human health. These chemicals, designated endocrine-disrupting chemicals, have recently become a social issue. Bisphenol A comes under this category because of its weak oestrogen-like effect (5, 6). It is a contaminant of particular importance because of its wide use in the production of polycarbonates, epoxy resins, and other plastics. Furthermore, it is antioxidant, nonbiodegradable, and highly resistant to chemical degradation so that high concentrations of bisphenol A are found in surface waters as well as industrial wastewaters (7, 8). The objective of the present study was to study the behavior of different activated carbons in the removal of bisphenol A from water, analyzing the influence of the characteristics of the carbon (surface area, porous texture, surface chemistry, and mineral matter content) and the chemistry of the solution (pH and ionic strength) on this process. Thus, prior to the use of the activated carbons as adsorbents of bisphenol A, they were texturally and chemically characterized.
2. Experimental Section 2.1. Activated Carbons. Three activated carbons were used: two were commercial charcoal-based carbons from SorboNorit, 3-A-7472 (sample S), and Merck, K27350518015 (sample M), and one (sample A) was prepared in our laboratory from almond shells. The almond shells were washed, carbonized in N2 at 1273 K for 1 h, and then steam-activated at 1123 K for 5 h. The percentage burnoff obtained (%BO) was 42.0%. Part of the inorganic matter in carbons S and M was removed by treatment with HCl following a method described elsewhere (9). The demineralized S and M carbon samples were designated S-HCl and M-HCl, respectively. 2.2. Characterization of the Activated Carbon Samples. All activated carbon samples were texturally characterized by N2 adsorption at 77 K. Prior to adsorption experiments, the carbon samples were outgassed at 378 K under pressure less than 10-4 Pa for at least 15 h. The BET equation was applied to the N2 adsorption isotherms to determine specific surface area (SBET). Allocation of pore size distribution involved subdividing the N2 adsorbed amount, from the adsorption isotherms, in the relative pressure ranges 0-0.01, 0.01-0.40, and 0.40-0.95, corresponding to adsorption in primary micropores, secondary micropores, and mesopores, respectively. Micropore volume (Vmic) and the external specific surface area (Sext) were calculated by applying the Rs-method to the N2 adsorption data (10, 11). Mercury porosimetry data were obtained using Autoscan 60 (Quantachrome) equipment and were used to determine both macropore volume and particle density (FHg) of the activated carbons. The particle size of all activated carbon samples was 0.5-0.8 mm. Elemental analysis of the carbons was carried out using a Fisons Carlo Erba 1108 analyzer, which measures the C, H, and N contents; the O content was calculated by difference. The mineral matter was obtained by burning the carbon in oxygen plasma using low-temperature ashing equipment (LTA-302). The main metals present in the mineral matter of these carbons were identified using a Perkin-Elmer model 5100 ZL Zeeman atomic absorption spectrophotometer. The procedure described by Boehm (12) was used to determine the acid and basic groups on the carbon surface. The pH of the point of zero charge (pHPZC) was established using pH shift analysis (13, 14). 10.1021/es0481169 CCC: $30.25
2005 American Chemical Society Published on Web 07/09/2005
TABLE 1. Elemental Analysis and Ash Content of the Activated Carbon Samples (daf, %)
FIGURE 1. Bisphenol A formula developed.
sample
C
H
N
O (by diff.)
ash
S M A S-HCl M-HCl
89.7 90.3 96.2 82.7 91.0
0.3 1.1 0.2 0.5 1.0
0.2 1.6 0.6 0.3 1.7
9.8 6.9 3.0 16.6 6.3
6.1 1.7 0.1 4.7 0.4
TABLE 2. Metals Detected in the Mineral Matter of Carbons (wt %)
FIGURE 2. Species distribution diagram for bisphenol A. 2.3. Bisphenol A Characterization. Bisphenol A was supplied by Sigma. Molecular modeling study of the tetrahedral molecule of bisphenol A (Figure 1) established the following characteristics: largest distance (between the hydroxyl groups of both aromatic rings), 0.94 nm; height, 0.53 nm; benzenic ring width, 0.43 nm; volume, 0.70 nm3; surface area, 4.32 nm2; and electrostatic potential, from -45.6 to 34 eV. Molecular modeling data were obtained by using the SPARTAN 4.0 program (15) running on a Silicon Graphics O2 workstation. The protonation equilibria of bisphenol A (H2BPA) were studied potentiometrically by titrating acidified aqueous solutions (HCl) of bisphenol A (concentrations from 1.0 × 10-3 to 2.5 × 10-3 M and ionic strength 0.1 M KCl) with 0.1 M KOH. The protonation constants were obtained from 100 experimental points in the pH range 2.0-12.5. The potentiometric assembly was controlled with a radiometer VIT90 video titrator unit connected to an ABU91 autoburet with a 25 mL exchanging unit (precision of (0.005 mL). A calomel K4040 electrode and a glass G2040B electrode were used for the emf and pH measurements. The sample solutions were titrated in a double-walled vessel at 25.0 ( 0.1 °C under continuous flow of nitrogen, previously bubbled through ascarite. The species distribution diagram (Figure 2) shows that the first deprotonation of bisphenol A started at around pH 8.0 and the second one at around pH 9.0. The values obtained for the first and second protonation constants (log β1 ) log K1; log β2 ) log K1+ log K2) were
log β1 ) 10.4288;
BPA2- + H+ f BPAH-
(1)
log β2 ) 20.0562;
BPA2- + 2H+ f BPAH2
(2)
2.4. Bisphenol A Adsorption. Adsorption isotherms of bisphenol A on the carbon samples were obtained by adding 0.1 g of carbon (particle size 0.5-0.8 mm) to flasks containing 100 cm3 of bisphenol A solution at different concentrations (50-350 mg/L). Prior to these experiments, the activated carbons were washed with distilled water and then dried at 383 K until constant weight. The flasks containing both carbon and bisphenol A were kept in a thermostatic bath at 298 K with constant agitation; after 7 days, the concentration of bisphenol A in each flask was measured. From previous kinetic studies, it was deduced that 7 days is more than
carbon
Fe
Ca
Mg
Na
S M
2.1 nil
16.0 2.8
4.4 0.5
0.5 0.6
adequate to attain adsorption equilibrium. Adsorption isotherms were determined at a pH of 6.5-7.0, which was obtained by adding the activated carbons to the bisphenol A solution. The adsorption isotherms of Bisphenol A were also determined at different ionic strengths in carbon samples S, M, and A by adding NaCl at concentrations of 0.1 or 0.01 M. Adsorption experiments were also conducted at different pH values, obtained by adding HCl or NaOH to the solution. Concentrations of bisphenol A were determined by spectrophotometry at 275.5 nm.
3. Results and Discussion 3.1. Activated Carbons. Table 1 exhibits the results of the elemental and ash analyses of the carbon samples; carbon S presented the highest ash content (6.1%) and carbon A the lowest (0.1%). A low ash content is an important feature of almond shell-derived carbon (carbon A), especially in relation to its use in water treatments. After HCl treatment, the ash content of carbon S only reduced to 4.7% (sample S-HCl), compared with 0.4% for carbon M (sample M-HCl). Table 2 shows some metals detected in the ashes of carbons S and M. It is important to note the high Ca and Mg contents of carbon S: 16.0% and 4.4%, respectively. Figure 3 shows the N2 adsorption isotherms of the three activated carbon samples and Figure 4 shows those corresponding to the demineralized activated carbons. All of these isotherms correspond to Type I isotherms, indicating that these carbon samples are mainly microporous; the fact that the corresponding plateaus are not completely horizontal implies that this microporosity is associated with some mesoporosity (10). The textural parameters of these carbon samples were determined by applying the calculation method described in the Experimental Section.
FIGURE 3. N2 adsorption isotherms on activated carbon. S (]), M (4), A (O). VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6247
FIGURE 4. N2 adsorption isotherms on activated carbon. S-HCl ([), M-HCl (2).
TABLE 3. Textural Characteristics of the Carbon Samples sample
GHga (g cm-3)
SBETb (m2 g-1)
Sextc (m2 g-1)
Vmicd (cm3 g-1)
S M A S-HCl M-HCl
0.661 0.705 0.743 0.695 0.730
1225 1084 1216 1277 1158
46.9 56.9 48.6 49.6 54.5
0.563 0.496 0.505 0.564 0.565
a Particle density by mercury porosimetry. b Specific surface area by the BET method. c External specific surface area by the Rs-method. d Micropore specific volume by the R -method. s
FIGURE 5. Adsorption isotherms of bisphenol A on activated carbon. S (]), M (4), A (O).
TABLE 5. Chemical Characteristics of the Activated Carbons carboxylic lactone phenol basic groups groups groups sites carbon (µequiv g-1) (µequiv g-1) (µequiv g-1) (µequiv g-1) pHPZC S M A
sample
secondary micropores,b 0.8-2 nm (cm3 g-1)
mesoporesc 2-50 nm (cm3 g-1)
macroporesd >50 nm (cm3 g-1)
S M A S-HCl M-HCl
0.392 0.319 0.404 0.404 0.325
0.190 0.197 0.119 0.179 0.249
0.044 0.034 0.065 0.051 0.049
0.481 0.259 0.155 0.448 0.244
a-c Calculated from the N adsorbed at different relative pressure 2 ranges. d By mercury porosimetry.
Table 3 lists the values of some textural characteristics of the carbon samples. All three original carbons had large specific surface areas of more than 1080 m2 g-1 and the largest was presented by carbon S (1225 m2 g-1). The order by micropore specific volume was the same as by specific surface area, i.e., M < A < S, although the values of these parameters were similar in the three activated carbons used. Comparing SBET with Sext values, it can be indicated, as deduced from the type of the corresponding isotherms stated above, that most of the surface area of these activated carbons is contained in the micropores, and only a small surface fraction, ranging from 46.9 m2 g-1 (carbon S) to 56.9 m2 g-1 (carbon M), corresponds to mesopores and macropores. Table 4 shows the pore volume distributions of the carbon samples. The sum of both the primary and secondary micropore volumes for each sample was similar to the micropore specific volume (Vmic) obtained by the Rs-method, with differences of less than 3.5% between the micropore volumes estimated by the different methods. Micropore size distribution differs between one activated carbon and another. Thus, the percentages of primary micropore volumes ranged from 77% (carbon A) to 62% (carbon M), and therefore, those of the secondary micropores ranged from 23% (carbon A) to 38% (carbon M). The highest 6248
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
111 252 63
160 145 66
1285 399 559
12.1 7.5 10.6
TABLE 6. Results Obtained by Applying the Langmuir Equation to Bisphenol A Isotherms on Original and Demineralized Carbons
TABLE 4. Pore Volume Distributions of the Carbon Samples primary micropores,a A . M. VOL. 39, NO. 16, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6249
TABLE 7. Results Obtained by Applying the Langmuir Equation to Bisphenol A Isotherms on Original Carbons in the Presence of Ionic Strength 0.01 M NaCl
0.1 M NaCl
carbon
Xm (mg g-1)
∆X m (mg g-1)
BXm (L g-1)
Xm (mg g-1)
∆ Xm (mg g-1)
BXm (L g-1)
S M A
227.3 263.6 238.1
97.7 0.5 49.2
38.0 158.7 96.2
238.1 270.3 263.2
108.5 7.2 74.3
56.2 476.2 88.5
All of these results are explained by the positive net charge of the carbons and the molecular form of the bisphenol A under the conditions of the adsorption experiments (pH ) 7). Therefore, the ions from NaCl are placed between the bisphenol A molecules and the carbon surface. This produces a screening effect of the surface charge (19) that favors adsorbate-adsorbent dispersion interactions (see above), thereby enhancing the adsorption of bisphenol A. This effect is more marked with increases in the positive charge density of the carbon and in the ionic strength of the solution. According to the pHPZC values of these carbons (Table 5), the positive charge increases in the order M < A < S, the same order found for the increase in Xm due to the presence of ionic strength in the solution. The presence of NaCl in solution also causes a salting-out effect, decreasing the solubility of bisphenol A and enhancing, therefore, its adsorption on the activated carbons. This fact could contribute, in part, to the increase of the adsorption capacity of the activated carbons observed when NaCl is present.
Acknowledgments The authors are grateful to the “Grupo de Modelizacio´n y Disen ˜ o Molecular” (University of Granada) for its contribution to the Molecular Modeling study, and to MEyC and FEDER (Project: CTQ2004-07783-C02-01/PPQ) for financial support.
Literature Cited (1) Mattson, J. S.; Mark, H. B., Jr. Activated carbon: surface chemistry and adsorption from solution; Marcel Dekker: New York, 1971. (2) Radovic, L. R.; Moreno-Castilla, C.; Rivera-Utrilla, J. Carbon materials as adsorbents in aqueous solutions. In Chemistry and Physics of Carbon, Vol. 27; Radovic, L. R., Ed.; Marcel Dekker: New York, 2001; pp 227-405. (3) Wu, S. H.; Pendleton, P. Adsorption of anionic surfactant by activated carbon: Effect of surface chemistry, ionic strength, and hydrophobicity. J. Colloid Interface Sci. 2001, 243, 306315.
6250
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 16, 2005
(4) Pendleton, P.; Wong, S. H.; Schumann, R.; Levay, G.; Denoyel, R.; Rouquerol, J. Properties of activated carbon controlling 2-methylisoborneol adsorption. Carbon 1997, 35, 1141-1149. (5) Steinmetz, R.; Brown, N. G.; Allen, D. L.; Bigsby, R. M.; BenJonathan, N. The environmental estrogen bisphenol A stimulates prolactin release in vitro and in vivo. Endocrinology 1997, 138, 153-160. (6) Pe´rez, P.; Pulgar, R.; Olea-Serrano, F.; Villalobos, M.; Rivas, A.; Metzler, M.; Pedraza, V.; Olea, N. The estrogenicity of bisphenol A related diphenylalkanes with various substituents at the central carbon and the hydroxyl groups. Environ. Health Perspect. 1998, 106, 167-174. (7) Lee, H. B.; Peart, T. E. Bisphenol A contamination in Canadian municipal and industrial wastewater and sludge sample. Water Qual. Res. J. Can. 2000, 35, 283-298. (8) Staples, C. A.; Dorn, P. B.; Klecka, G. M.; O’Block, S. T.; Branson, D.; Harris, L. R. Bisphenol A concentrations in receiving waters near US manufacturing and processing facilities. Chemosphere 2000, 40, 521-525. (9) Lo´pez-Ramon, M. V.; Moreno-Castilla, C.; Rivera-Utrilla, J.; Hidalgo-Alvarez, R. Activate carbons from a subbituminous coal: pore textural and electrokinetic properties. Carbon 1993, 31, 815-819. (10) Gregg, S. J.; Sing, S. W. Adsorption, surface area and porosity; Academic Press: London, 1982. (11) Rodrı´guez-Reinoso, F.; Martı´n-Martı´nez, J. M.; Prado-Burguete, C.; McEnaney, B. J. A standard adsorption isotherm for the characterization of activated carbons. J. Phys. Chem. 1987, 91, 515-516. (12) Boehm, H. P. Chemical identification of surface groups. Adv. Catal. 1996, 16, 179-274. (13) Newcombe, G.; Hayes, R.; Drikas, M. Granular activated carbon: importance of surface properties in the adsorption of naturally occurring organics. Colloids Surf. A 1993, 78, 65-71. (14) Ferro-Garcı´a, M. A.; Rivera-Utrilla, J.; Bautista-Toledo, I.; Moreno-Castilla, C. Adsorption of humic substances on activated carbon from aqueous solutions and their effect on the removal of Cr (III) ions. Langmuir 1998, 14, 1880-1886. (15) SPARTAN version 4.0, 18401 Von Karman Ave. # 370 Irvine, CA 92715 USA. 1995 Wave Function. (16) Couglhin, R. W.; Ezra, F. S. Role of surface acidity in the adsorption of organic pollutants on the surface of carbon. Environ. Sci. Technol. 1968, 2, 291-298. (17) Rivera-Utrilla, J.; Sa´nchez-Polo, M. The role of dispersive and electrostatic interactions in the aqueous phase adsorption of naphthalenesulphonic acids on ozone-treated activated carbons. Carbon 2002, 40, 2685-2691. (18) Julien, F.; Baudu, M.; Mazet, M. Relationship between chemical and physical surface properties of activated carbon. Water. Res. 1998, 32, 3414-3424. (19) Stuart, M. A.; Fleer, G. J.; Lyklema, J.; Norde, W.; Scheutjens, J. M. H. M. Adsorption of ions, polyelectrolytes and proteins. Adv. Colloid Interface Sci. 1991, 34, 477-535.
Received for review November 30, 2004. Revised manuscript received May 25, 2005. Accepted June 8, 2005. ES0481169
