Environ. Sci. Technol. 2008, 42, 3207–3212
Sorption of Organic Contaminants by Carbon Nanotubes: Influence of Adsorbed Organic Matter X I L O N G W A N G , † J I A L O N G L U , †,‡ A N D B A O S H A N X I N G * ,† Department of Plant, Soil and Insect Sciences, University of Massachusetts, Amherst, Massachusetts 01003, and College of Resources and Environment, Northwest A&F University, Yangling, Shaanxi 712100, China
Received November 28, 2007. Revised manuscript received February 5, 2008. Accepted February 7, 2008.
Sorption of three types of dissolved organic matter (DOM; i.e., humic acid, peptone and R-phenylalanine) by a mutiwalled carbon nanotube (MWNT40) and sorption of phenanthrene (Phen), naphthalene (Naph), and 1-naphthol (1-Naph) by the original and DOM-coated MWNT40 were examined. Sorption data of Phen, Naph, and 1-Naph by all sorbents were fitted with Freundlich and Polanyi models. MWNT40 had nonlinear isotherms for all DOMs examined. Sorption of DOMs by MWNT40 followed the order peptone > humic acid > R-phenylalanine. The humic acid used in this study had much lower sorption for Phen, Naph, and 1-Naph than MWNT40, but its coating did not make striking changes on sorption of these compounds by MWNT40, suggesting that humic acid coating dramatically altered the physical form and surface properties of MWNT40. Peptone coating made the strongest suppression on sorption of Phen, Naph, and 1-Naph by MWNT40 among the three DOMs used, due to its highest sorption on MWNT40, thus causing a great reduction in accessibility of sorption sites. Polanyi modeling results showed that reduction in the maximum volume sorption capacity (Q0) of MWNT40 induced by DOM coating followed the order Phen < Naph humic acid > R-phenylalanine (Figure 1). Both humic acid and peptone are macromolecules, but peptone
b
Kf ((mg/kg)/(mg/L)n).
c
Sw, aqueous
had higher sorption on MWNT40. MWNT40 had BET surface area and microporosity of 87 m2/g and 0.036 cm3/g, respectively.PeptonesorptionsubstantiallyreducedMWNT40s surface area to 35 m2/g and microporosity to 0.016 cm3/g, due to its strongest sorption (Figure 1; Table S1). However, the surface area reduction of MWNT40 resulting from unit loading mass of DOMs ranged from 0.28 to 0.30 m2/mg, and that of microporosity was 0.11–0.20 mm3/mg; comparable reduction in both surface area and microporosity of MWNT40 among individual DOMs suggests that DOM sorption was of both surface-adsorption and micropore-blocking processes. Peptone, humic acid, and R-phenylalanine coatings increased the meso- and macropore volumes of MWNT40, which can be attributed to the coated DOM molecules and structural rearrangement of MWNT40 as induced by DOM coating. Sorption of HOCs. MWNT40 had nonlinear isotherms for Phen, Naph, and 1-Naph (Figure S5 (Supporting Information)). Sorption coefficient (Kd) of these compounds by MWNT40 followed the order Phen > Naph > 1-Naph, in line with the order of their log Kow values (Tables 2 and S2). Polar functionalities can provide sites for water clusters to form via H-bonding, thereby affecting the surface hydrophobicity of MWNT40. A slight increase in O-containing moieties would greatly enhance the water cluster formation. Peptone coating introduced substantial polar moieties to the surface of MWNT40 as indicated by the elevated O content (Table 1), which facilitated water sorption to hydrophilic sites, thus mitigating the hydrophobic interactions between MWNT40 and HOC molecules. The newly introduced polar functionalities made the peptone-coated MWNT40 energetically less favorable for HOC sorption in comparison with MWNT40. The newly introduced polar functionalities by peptone coating would also make the hydrophobic sites in MWNT40 less accessible to HOC molecules. Therefore, peptone coating strongly reduced sorption of Phen, Naph, and 1-Naph by MWNT40. Similarly, the surrounding polar moieties highly reduced the accessibility of HOC molecules to hydrophobic domains in natural sorbents, e.g., biopolymers (16). The VOL. 42, NO. 9, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3209
FIGURE 2. Polanyi model-based sorption isotherms of Phen, Naph, and 1-Naph by the original and DOM-coated MWNT40. MWNT40 ()); humic acid-coated MWNT40 (0); r-phenylalanine-coated MWNT40 (4), and peptone-coated MWNT40 (O). negative correlation between abundance of O-containing moieties and HOC sorption was also observed for soil organic matter (20). Likewise, sorption of 2-methylisoborneol by activated carbon decreased with increasing oxygen content of the sorbent (21). At low solute concentrations, sorption of HOCs by MWNT40 can be dominated by pore-filling mechanism because there are plenty of high-energy sites in micropores. Stronger suppression of HOC sorption by MWNT40 from peptone coating in comparison with R-phenylalanine-coating was due to its larger molecular size, so that it would be able to block more micropores thus reducing availability of sorption sites for HOC sorption (Table 1). At high solute concentrations, HOC sorption by MWNT40 would be dominated by surface-adsorption mechanism because there are limited high-energy sites in the pores for a given sorbent. Higher sorption of peptone to MWNT40 in comparison with R-phenylalanine caused a greater surface area reduction of MWNT40 (Tables 1 and S1), thereby reducing more sorption sites on the surface for Phen, Naph, and 1-Naph (Table 2). Humic acid had much lower sorption for Phen and 1-Naph than MWNT40 as shown by our previous study (12). However, humic acid coating only slightly reduced sorption of MWNT40 for these compounds (Figure S5; Table 2), suggesting that humic acid-coating dramatically altered the physical form and surface properties of MWNT40. Due to ionization of polar functionalities in humic acid, it is usually negatively charged. Its coating would reduce the aggregation of MWNT40 particles, thereby enhancing their repulsion and increasing the effective surface area and porosity for HOC sorption. Strong stabilizing capability of a standard Suwannee River DOM on MWNTs was observed, and microscopic analyses clearly showed that the suspension consisted mainly of individually dispersed MWNTs (8). Enhanced dispersion of MWNT40 by humic acid was visually observed in our DOM coating and sorption experiments, but not much by peptone and R-phenylalanine. It was possible that humic acid molecules had a disparate distribution of sorption sites with Phen, Naph, and 1-Naph on the MWNT40 due to their vast structural and composition difference. No striking changes in HOC sorption on MWNT40 by humic acid coating may also indicate that sorption reduction due to reduced accessibility from polar moieties was offset by the sorption increase due to the newly exposed sites from enhanced inter-MWNT40 particle repulsion, because humic acid coating introduced substantial Ocontaining moieties to MWNT40 (Table 1). Our observation demonstrating the weak effect of humic acid coating on sorption of Phen, Naph, and 1-Naph by MWNT40 was inconsistent with the previous findings showing that sorption of HOCs to activated carbon filters was reduced by “fouling” with DOM (22). Jonker et al. (23) also showed that sorption of PCBs to char decreased after addition of sediment. Hopman et al. (24) found that activated carbon fibers with only small micropores sorbed a little amount of DOM due to size exclusion, and DOM coating hardly reduced atrazine ad3210
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 9, 2008
sorption. In comparison, carbon fibers with larger micropores adsorbed a higher amount of DOM, hence decreasing its adsorption for atrazine. The different impacts of DOM treatment on HOC sorption by CNTs, activated carbon, and char could be a result of their structural differences. Humic acid is generally more hydrophobic than fulvic acid; however, fulvic acid was more effective at suppressing benzene sorption than humic acid by a wood char (25), which the authors interpreted that the fulvic acid molecules can penetrate more deeply into the pores thereby competing sorption sites. For our case, both humic acid and peptone coating introduced comparable O-containing moieties to MWNT40 (Table 1), but they affected HOC sorption differently, suggesting that the influence of adsorbed DOMs on HOC sorption by MWNT40 was dependent on their properties. Consistent with the results of Freundlich model fitting, Polanyi model results also showed that, humic acid coating exerted the weakest impact on sorption of Phen, Naph, and 1-Naph by MWNT40 (Figure 2). In comparison, peptone coating caused the greatest reduction in sorbed volume of these compounds. The same trend was observed after organic carbon normalization (Figure S6 (Supporting Information)). After the DOM loading normalization, peptone and R-phenylalanine coating resulted in comparable reduction of volume sorption capacity (Q0) of 1-Naph, much higher than that of humic acid. The peptone loading-normalized Q0 reduction of Naph was also much higher than that normalized with humic acid loadings. The strongest HOC sorption suppression by peptone coating was due to its highest sorption on MWNT40, thus the largest reduction in accessibility of sorption sites as stated above. Peptone coating caused comparable reduction in the sorbed volume of HOCs at both low and high solute concentrations (log scale of Figure 2). This suggests that peptone could have been proportionally sorbed to the surface and pore ends of MWNT40, thus blocking sorption sites. Humic acid and R-phenylalanine coating slightly decreased sorption and isotherm nonlinearity of Phen, Naph, and 1-Naph by MWNT40. In comparison, sorption and isotherm nonlinearity of these compounds by MWNT40 was greatly reduced after peptone coating (Table 2). Similar to competitive sorption systems, in the presence of cosolute, sorption of primary solute would be reduced and its isotherms would become more linear (26). The coated DOM molecules could have competed for sites with HOCs on the MWNT40. Except for sorption of Naph by R-phenylalanine-coated MWNT40, DOM sorption reduced the volume sorption capacity (Q0) of HOCs by MWNT40, following the order Phen (0–29%) < Naph (4–51%) < 1-Naph (12–75%) (Table 3). This order is in line with the ordinary competition thermodynamic theory showing that the more hydrophobic the primary solute is, the weaker the sorption suppression would be. The Polanyi model has three underlying assumptions (27): (1) for any molecule, the magnitude of its adsorption potential is a function of its proximity to the sorbent surface; (2) the adsorption potential is independent of temperature; and (3)
TABLE 3. Polanyi Model Parameters for Sorption of Phen, Naph, and 1-Naph by Original and DOM-Coated MWNT40 sorbents
log Q 0a
MWNT40 HA-MWNT40 PH-MWNT40 PE-MWNT40 MWNT40 HA-MWNT40 PH-MWNT40 PE-MWNT40 MWNT40 HA-MWNT40 PH-MWNT40 PE-MWNT40
4.319 ( 0.013 4.334 ( 0.019 4.235 ( 0.010 4.172 ( 0.019 4.336 ( 0.014 4.291 ( 0.011 4.319 ( 0.017 4.023 ( 0.023 5.286 ( 0.040 5.231 ( 0.027 5.154 ( 0.034 4.692 ( 0.027
compounds Phen
Naph
1-Naph
a
log Q0 (mm3/kg).
b
a (J/cm3)–b. c b, dimensionless.
ab d
d
bc
-0.001 ( 0.0006 -0.008 ( 0.003 -0.002 ( 0.001 -0.007 ( 0.002 -0.004 ( 0.001 -0.003 ( 0.001 -0.005 ( 0.001 -0.022 ( 0.003 -0.004 ( 0.003 -0.003 ( 0.001 -0.005 ( 0.002 -0.005 ( 0.001
Standard error of log Q0.
e
e
1.440& ( 0.123 0.988 ( 0.086 1.328 ( 0.102 1.199 ( 0.050 1.168 ( 0.034 1.221 ( 0.031 1.195 ( 0.041 0.873 ( 0.029 1.109 ( 0.011 1.234 ( 0.081 1.093 ( 0.091 1.183 ( 0.064
R2 f
0.987 0.991 0.990 0.996 0.998 0.999 0.997 0.998 0.986 0.994 0.991 0.996
Standard error of a. f Standard error of b.
FIGURE 3. Polanyi model-based sorbed volume of Phen, Naph, and 1-Naph by the original (A) and DOM-coated MWNT40 (B): Phen (0); Naph (4); 1-Naph (O). MWNT40 (black color); humic acid-coated MWNT40 (red color); peptone-coated MWNT40 (dark blue); r-phenylalanine-coated MWNT40 (light blue). the adsorption potential of any molecule is independent of its state of aggregation. On the basis of two additional assumptions, (1) all adsorption space is accessible to molecules, which means there is no molecular sieving effect and all sorbates should have a common limiting volume at zero adsorption potential, and (2) the molar volume is the best normalizing factor for adsorption potential, the correlation curves for different sorbates on a given sorbent would yield a single line (27). Xia and Ball (28) observed that the correlation curve of liquid solutes on a soil essentially overlapped, thus giving a common volume sorption capacity for all solutes tested. However, for solid sorbates, the correlation curve did not fall on a single line. Recently, it was reported that liquid solutes had a common correlation curve by chars, but that of solid solutes by the same chars was separated (29). For our case, Vm of 1-Naph was higher than those of Phen and Naph at individual RT ln(Sw/Ce)/Vs points for all sorbents, revealing that all sorbents had a higher affinity for 1-Naph (Figure 3). Such a trend was also observed after organic carbon normalization (Figure S7 of the Supporting Information). Specific interactions between the hydroxyl group on 1-Naph and the native and sorbed polar moieties on the sorbents would increase sorption affinity for 1-Naph, but such an interaction could not occur for Phen and Naph. The specific interactions between polar sorbates and sorbent surface would make their correlation curve separated from nonpolar ones (28). The normalizing factor was used to collapse the correlation curve of distinct compounds on a given sorbent into a single line. The previously proposed normalizing factors included molar volume, polarizability, and parachor. Basically these factors only accounted for van der Waals interaction forces
(30). In aqueous sorption systems, both van der Waals forces and hydrophobic interactions are important depending upon the sorbate and sorbent. A good normalizing factor should adequately consider all of the interaction forces that affect sorption. The overall sorption energy primarily consists of interaction energy between organic solute and sorbent, water and sorbent, intersorbed sorbate molecules, solute and water molecules, and interwater molecules (31). The correlation curve for sorption of aromatics, halogenated aliphatics, and aromatics by a coal-based activated carbon overlapped after the normalizing factor was adjusted from molar volume to LSER parameters which adequately covered the aforementioned interaction forces (31). On the basis of our observation and other studies (7, 29), to get a single correlation curve for sorption of various sorbates by a given sorbent, state of sorbate, and significance of specific interactions between sorbate and sorbent, as well as an improved normalizing factor, should be further considered. The Polanyi modeling results indicated that Q0 of 1-Naph by all sorbents was higher than those of Phen and Naph (Table 3). This can be attributed to their structural difference. The hydrophobic benzene rings of 1-Naph would mainly be sorbed close to surfaces of all sorbents through hydrophobic interactions, while leaving its -OH group facing the aqueous phase. The outward -OH of sorbed 1-Naph can form new “hydrogen bonding adsorption sites”, so a second adsorption layer for 1-Naph molecules would be created at the sorbent surface. Such a mechanism may facilitate 1-Naph sorption by forming multiple sorption layers, thus giving higher volume capacity. The same sorption mechanism was used to interpret the cooperative sorption of 1-Naph and 1-naphthylamine by nonpolar macroreticular adsorbents (32). Naph VOL. 42, NO. 9, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3211
was more hydrophobic, and it had smaller molecular volume than 1-Naph (Table S2), but it had lower Q0 than 1-Naph (Table 3), suggesting that pore filling could not be the dominant sorption mechanism of these compounds by all sorbents at high solute concentrations. Surface adsorption is another mechanism for CNTs, which is supported by our early report (7). Phen and Naph had similar structure but were different in hydrophobicity. They had comparable Q0 and correlation curves by all sorbents (Figure 3, Table 3), revealing that the correlation curve was applicable for sorption of aromatic compounds of similar structure by the original and DOM-coated CNTs. Environmental Implications. With increasing production of nanoscaled materials (e.g., CNTs), more and more of these materials would be introduced into the environment. DOM is ubiquitously present in the environment. Once they coexist, they would interact with each other. The interaction greatly altered the surface functionalities, surface area, and porosity of CNTs, thereby exerting strong impact on its sorption for HOCs and perhaps other chemicals as well. Peptone had the highest sorption on CNTs (i.e., MWNT40), consistently it exhibited the strongest suppression on Phen, Naph, and 1-Naph sorption by MWNT40, due to the greatest reduction in accessibility of its sorption sites. In addition, DOM may affect environmental behaviors of CNTs such as dispersion and transport, which would also alter the fate and transport of HOCs due to their association with CNTs.
(10)
(11) (12) (13) (14) (15) (16)
(17) (18) (19)
Acknowledgments We would like to thank Dr. Tamamura Shuji at Hokkaido University, Japan for surface area and porosity analysis. This project was supported in part by the Massachusetts Agricultural Experiment Station (Grants MAS90 and MAS860) and Massachusetts Water Resources Research Center.
Supporting Information Available Coated amount of DOMs on MWNT40 (Table S1), pysicochemical properties of Phen, Naph, and 1-Naph (Table S2), TEM images of all sorbents (Figure S1), sorption–desorption isotherms of N2 on all sorbents (Figure S2), DRIFT spectra (Figure S3), and pore size distribution (Figure S4) of the sorbents, sorption isotherms of Phen, Naph, and 1-Naph by all sorbents (Figure S5), Polanyi-model-based sorption isotherms (Figure S6), and sorbed volume (Figure S7) of Phen, Naph, and 1-Naph by sorbents on per carbon basis. This material is available free of charge via the Internet at http:// pubs.acs.org.
Literature Cited (1) Iijima, S. Helical microtubules of graphitic carbon. Nature 1991, 354, 56–58. (2) Dillon, A. C.; Jones, K. M.; Bekkedahl, T. A.; Kiang, C. H.; Bethune, D. S.; Heben, M. S. Storage of hydrogen in single-walled carbon nanotubes. Nature 1997, 386, 377–379. (3) Liu, C.; Cheng, H. M. Carbon nanotubes for clean energy applications. J. Phys. D: Appl. Phys. 2005, 38, R231–R252. (4) Cai, H.; Cao, X. N.; Jiang, Y.; He, P.; Fang, Y. Carbon nanotubeenhanced electrochemical DNA biosensor for DNA hybridization detection. Anal. Bioanal. Chem. 2003, 375, 287–293. (5) Peng, X.; Li, Y.; Luan, Z.; Di, Z.; Wang, H.; Tian, B.; Jia, Z. Adsorption of 1,2-dichlorobenzene from water to carbon nanotubes. Chem. Phys. Lett. 2003, 376, 154–158. (6) Long, R. Q.; Yang, R. T. Carbon nanotubes as superior sorbent for dioxin removal. J. Am. Chem. Soc. 2001, 123, 2058–2059. (7) Yang, K.; Zhu, L. Z.; Xing, B. S. Adsorption of polycyclic aromatic hydrocarbons by carbon nanomaterials. Environ. Sci. Technol. 2006, 40, 1855–1861. (8) Hyung, H.; Fortner, J. D.; Hughes, J. B.; Kim, J. H. Natural organic matter stabilizes carbon nanotubes in the aqueous phase. Environ. Sci. Technol. 2007, 41, 179–184. (9) Balavoine, F.; Schultz, P.; Richard, C.; Mallouh, V.; Ebbeson, T. W.; Mioskowski, C. Helical crystallization of proteins on
3212
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 9, 2008
(20) (21) (22) (23)
(24) (25)
(26) (27) (28) (29)
(30) (31)
(32)
carbon nanotubes: A first step towards the development of new biosensors. Angew. Chem., Int. Ed. 1999, 38, 1912–1915. Tsang, S. C.; Davis, J. J.; Green, M. L. H.; Hill, H. A. O.; Leung, Y. C.; Sadler, P. J. Immobilization of small proteins in carbon nanotubes: High-resolution transmission electron microscopy study and catalytic activity. J. Chem. Soc., Chem. Commun. 1995, 24, 2579–2579. Guo, Z. J.; Sadler, P. J.; Tsang, S. C. Immobilization and visualization of DNA and proteins on carbon nanotubes. Adv. Mater. 1998, 10, 701–703. Wang, X. L.; Xing, B. S. Roles of acetone-conditioning and lipid in sorption of organic contaminants. Environ. Sci. Technol. 2007, 41, 5731–5737. Kang, S.; Xing, B. S. Phenanthrene sorption to sequentially extracted soil humic acids and humins. Environ. Sci. Technol. 2005, 39, 134–140. Wang, X. L.; Xing, B. S. Sorption of organic contaminants by biopolymer-derived chars. Environ. Sci. Technol. 2007, 41, 8342– 8348. Ismail, I. M. K.; Rodgers, S. L. Comparisons between fullerene and forms of well-known carbons. Carbon 1992, 30, 229–239. Wang, X. L.; Yang, K.; Tao, S.; Xing, B. S. Sorption of aromatic organic contaminants by biopolymers: Effects of pH, copper(II) complexation, and cellulose coating. Environ. Sci. Technol. 2007, 41, 185–191. Lu, C.; Chung, Y. L.; Chang, K. F. Adsorption thermodynamic and kinetic studies of trihalomethanes on multiwalled carbon nanotubes. J. Hazard. Mater. 2006, B138, 304–310. Agnihotri, S.; Rostam-Abadi, M.; Rood, M. J. Temporal changes in nitrogen adsorption properties of single-walled carbon nanotubes. Carbon 2004, 42, 2699–2710. Braida, W. J.; Pignatello, J. J.; Lu, Y.; Ravikovitch, P. I.; Neimark, A. V.; Xing, B. Sorption hysteresis of benzene in charcoal particles. Environ. Sci. Technol. 2003, 37, 409–417. Wang, X. L.; Sato, T.; Xing, B. S. Sorption and displacement of pyrene in soils and sediments. Environ. Sci. Technol. 2005, 39, 8712–8718. Pendleton, P.; Wong, S. H.; Schumann, R.; Levay, G.; Denoyel, R.; Rouquero, J. Properties of activated carbon controlling 2-methylisoborneol adsorption. Carbon 1997, 35, 1141–1149. Carter, M. C.; Weber, W. J., Jr. Modeling adsorption of TCE by activated carbon preloaded by background organic matter. Environ. Sci. Technol. 1994, 28, 614–623. Jonker, M. T. O.; Hoenderboom, A.; Koelmans, A. A. Effects of sedimentary soot-like materials on bioaccumulation and sorption of polychlorinated biphenyls. Environ. Toxicol. Chem. 2004, 23, 2563–2570. Hopman, R.; Siegers, W. G.; Kruithof, J. C. Organic micropollutant removal by activated carbon fiber filtration. Water Supply 1995, 13, 257–261. Pignatello, J. J.; Kwon, S.; Lu, Y. Effect of natural organic substances on the surface and adsorptive properties of environmental black carbon (char): Attenuation of surface activity by humic and fulvic acids. Environ. Sci. Technol. 2006, 40, 7757– 7763. Wang, X. L.; Sato, T.; Xing, B. S. Competitive sorption of pyrene on wood chars. Environ. Sci. Technol. 2006, 40, 3267–3272. Manes, M. Activated Carbon Adsorption Fundamentals. In Encyclopedia of Environmental Analysis and Remediation; Meyers, R. A., Ed.; Wiley: New York, 1998; pp 26–68. Xia, G.; Ball, W. P. Adsorption-partitioning uptake of nine low polarity organic chemicals on a natural sorbent. Environ. Sci. Technol. 1999, 33, 262–269. Nguyen, T. H.; Cho, H. H.; Poster, D. L.; Ball, W. P. Evidence for a pore-filling mechanism in the adsorption of aromatic hydrocarbons to a natural wood char. Environ. Sci. Technol. 2007, 41, 1212–1217. Dubinin, M. M. The potential theory of adsorption of gases and vapors for adsorbents with energetically nonuniform surfaces. Chem. Rev. 1960, 60, 235–241. Crittenden, J. C.; Sanongraj, S.; Bulloch, J. L.; Hand, D. W.; Rogers, T. N.; Speth, T. F.; Ulmer, M. Correlation of aqueous-phase adsorption isotherms. Environ. Sci. Technol. 1999, 33, 2926– 2933. Zhang, W. M.; Chen, J. L.; Pan, B. C.; Zhang, Q. X. Cooperative adsorption behaviors of 1-naphthol and 1-naphthylamine onto nonpolar macroreticular adsorbents. React. Funct. Polym. 2006, 66, 485–493.
ES702971G
