Environ. Sci. Technol. 2003, 37, 1448-1451
Removal of Natural Organic Polyelectrolytes by Adsorption onto Tobermorite S A T O S H I K A N E C O , * ,† K U M I K O I T O H , † HIDEYUKI KATSUMATA,† TOHRU SUZUKI,‡ KAZUAKI MASUYAMA,§ KUNIHIRO FUNASAKA,| KAZUYUKI HATANO,⊥ AND KIYOHISA OHTA† Department of Chemistry for Materials, Faculty of Engineering, Mie University, Tsu, Mie 514-8507, Japan, Environmental Preservation Center, Mie University, Tsu, Mie 514-8507, Japan, Mie Industrial Research Institute, Tsu, Mie 514-0819, Japan, Osaka City Institute of Public Health and Environment Sciences, Osaka, 543-0026, Japan, and Water Environment Department, CLION Co., Ltd., Owariasahi, Aichi 488-0052, Japan
Natural organic polyelectrolytes, such as humic and fulvic acids, were removed by adsorption onto silicate rocks. Tobermorite, zeolite, and molecular sieves 5A were used as the adsorbents. Tobermorite was more efficient by 4050% in the removal of fulvic acids, and by 30-50% for humic acid than zeolite or molecular sieves, respectively. Humic acid removal from the solution by adsorption onto silicate rocks took place more readily than fulvic acid removal. From the determined heat of adsorption, the adsorption process in the present study may be chemisorption (ligand exchange). Metal/humic acid complexes were effectively removed by adsorption onto tobermorite. Because tobermorite (a silicate rock) can be easily synthesized and obtained commercially, the adsorption method of removal of fulvic and humic acids is superior to their precipitation.
of water treatment processes that employ membranes or microporous adsorbents (4-6). Even when not specifically targeted for removal, macromolecular dissolved organic matter has been shown to compete with low-molecularweight synthetic organic chemicals, reducing their adsorption rates and equilibrium capacities (4-9). Therefore, the adsorption of humic substances has been widely investigated in order to optimize their removal from solution and to minimize their impacts on the adsorption of other compounds specifically targeted for removal. Published investigations on the removal of humic materials have included their adsorption onto activated carbon (10, 11), nanofiltration membranes (12), crystalline CaCO3 (13), ion-exchange resin (14), pyrolusite (β-MnO2) (15), colloidal alumina (γ-Al2O3) (16), aluminosillicates (17), hematite (R-Fe2O3) (18, 19), goethite (20), gibbsite (21), imogolite (21), kaolinite (19, 21), silanized SiO2 (22), polystyrene (22), mordenite (23), and montmorillonite (23). An adsorption study of humic substances by well-characterized oxide surfaces has provided information about the types of surface reactions in which humic substances may participate. These reactions include anion exchange (electrostatic interaction), ligand exchangesurface complexation, hydrophobic interaction, entropic effects, hydrogen bonding, and cation bridging (24). The dominant type of reaction in any given circumstance will depend on the character of humic substances, the chemistry of the surface, and such solution variables as pH, ionic strength, and the presence of divalent cations. Tobermorite, which is one of the silicate rocks, can be obtained commercially, as it is easily synthesized and is fairly uniform in its chemical and physical properties. Therefore, because tobermorite is an inexpensive treatment medium, it has been applied for removing heavy metals from wastewater (25). Despite demonstration of efficient adsorption for the heavy metals, there are few reports concerning the adsorption of organic compounds containing humic substances. Accordingly, the present study investigates the adsorption of polyelectrolytes such as humic and fulvic acids onto tobermorite.
Experimental Section Introduction Humic and fulvic acids are natural organic polyelectrolytes that comprise the greatest proportion of naturally occurring dissolved organic matter in aqueous systems (1-3). They are not well-defined substances, but generally can be subdivided into three fractions, namely: humin, which represents insoluble components in aqueous solutions at all pH values; humic acids, which are soluble in alkaline solution to weakly acidic solutions, but deposit at or below pH 2.0; and fulvic acids, which are soluble in aqueous solutions at all pH values. Humic materials may be specifically targeted for removal from potable water supplies because they can adversely affect appearance and taste, and they can react with chlorine to form potentially carcinogenic chlorinated organic compounds. Further, the presence of dissolved macromolecular organic matter may reduce the effectiveness * Corresponding author phone: (81)59231-9427; fax: (81)592319442, 9471, or 9427; e-mail:
[email protected]. † Department of Chemistry for Materials, Mie University. ‡ Environmental Preservation Center, Mie University. § Mie Industrial Research Institute. | Osaka City Institute of Public Health and Environment Sciences. ⊥ CLION Co., Ltd. 1448
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 7, 2003
Apparatus. The absorbance at 400 nm was measured on a JASCO UVIDE-610 spectrophotometer with 10- mm cells for the determination of humic and fulvic acids. A Hitachi S-4000 scanning electron microscope was used for monitoring the surface conditions of adsorbents. All removal procedures were conducted in a clean bench. Adsorbate. Humic Acid Solution (0.03 mg/mL). Humic acid (extracted from peat soil in Japan, Wako Pure Chemical Industries, Ltd.) was purified according to International Humic Substances Society (IHSS) protocol (26). Humic acid was dissolved in 0.1 mol/L KOH solution and filtered through 0.45-µm membrane filters (ADVANTEC Toyo Kaisha, Ltd.). The solution was acidified to pH 1 with 1 mol/L HCl to precipitate humic acid, which was then separated by centrifugation and dissolved in 0.1 mol/L KOH solution. The absorptivity at 400 nm was 10 and 12 cm2/mg at pH 7 and 13, respectively. The elemental composition (%) reported by the manufacturer was 30-35 (oxygen), 60-65 (carbon), and 4-5 (hydrogen). The terrestrial humic acid examined in this study possessed chemical properties (e.g., complexing ability, absorption spectra, and molecular weight distribution) similar to those of aquatic humic acid (27). Although a specific part of humic acid may be damaged under the alkaline treatment, the effect was assumed to be negligible because 10.1021/es020816v CCC: $25.00
2003 American Chemical Society Published on Web 02/21/2003
it was reported that the alkaline treatment causes no significant changes in the humic acid structure (28). Fulvic Acid Solution (0.02 mg/mL). Fulvic acid (extracted from brown forest soil based on the method of IHSS (29)) was dissolved in water. The absorptivity at 400 nm was 2.2 and 3.8 cm2/mg, at pH 7 and 13, respectively. The total acidity was 9.1 meq/g. The elemental composition (%) supplied by the Japanese Humic Substances Society was 48.1 (oxygen), 47.6 (carbon), 3.5 (hydrogen), and 0.8 (others). Solution of Metal/Humic Complexes (23). To 500 mL of humic acid solution (0.03 mg/mL, pH 7) was added 3.2 mg of copper(II) chloride or 0.74 mg of aluminum(III) chloride. To fulvic acid of 500 mL solution (0.02 mg/mL, pH 7) was added 8.0 mg of copper(II) chloride or 3.3 mg of aluminum(III) chloride. Then, the solutions were stirred magnetically for 1 h until the complete complexation. Adsorbents. Tobermorite (Ca5Si6O16(OH)2‚2H2O; TAX 3; Onoda A.L.C. Co., Ltd.), zeolite ((0.4 K + 0.6 Na)‚Al2O3‚2SiO2; Synthetic; A-3; Shot; 0.50-1.18 mm (14-30 mesh); Wako Pure Chemical Industries, Ltd.), and molecular sieves 5A (Na12[(AlO2)12(SiO2)12]‚27H2O; 1/16; φ1.6 mm; Nakarai Tesque Inc.) were used as adsorbents for the removal of humic substances. Zeolite and molecular sieves 5A were tested for the comparison. The surface areas of tobermorite, zeolite, and molecular sieves 5A were 19, 14, and 370 m2/g, while their pore sizes 50-2000, ∼3×10-4, and 5×10-4 µm, respectively. These silicate rocks were used without pretreatment and were directly added to the treatment column. Procedure. An adsorption column (φ16 × 100 mm) with a coarse filter was packed with 6 g of adsorbent. After the adjustment of pH to the desired value with HCl and KOH solutions, the sample solutions (10 mL) containing humic materials were passed through the adsorption column at a flow rate of about 3.3 mL/min. The flow rate was kept constant by controlling the degree of outlet reduced pressure. The removal process was performed at room temperature. The effluent was immediately analyzed by spectrophotometry. Gel Permeation Chromatography. Molecular weight distributions were measured by high-pressure gel permeation chromatography (HPGPC). The HPGPC system was a TRI ROTER-V JASCO solvent pump, an UVIDEC-100-VI JASCO variable-wavelength absorbance detector, a TU-100 JASCO thermostat, a Rheodyne rotary injector valve with a 20-µL sample loop, and a Waters Protein-Pak 50 HPGPC column. The column packing material (a polymerized diol covalently bonded to a silica support, exclusion limit 80 000, permeation limit 100) was selected on the basis of its low residual hydrophobicity and minimal ion-exchange capacity (20). The mobile phase consisted of 0.1 mol/L NaCl, 0.002 mol/L KH2PO4, and 0.002 mol/L Na2HPO4 solutions buffered to pH 6.8. The flow rate was 0.40 mL/min. The HPGPC system was calibrated using pullulan (Shodex standard P-82).
Results and Discussion Adsorption of Humic Substances onto Tobermorite. The adsorption characteristics of humic materials, such as fulvic and humic acids, were investigated by using tobermorite, zeolite, and molecular sieves 5A. The removal was performed in the range of pH 5-9, given that river water is generally close to neutrality. Results are shown in Figures 1 and 2. The removal efficiency by adsorption onto tobermorite was highest for fulvic and humic acids, being almost constant in the pH range tested. For the sorption of humic substances onto kaolinite and hematite (19), the removal efficiency increased with decreasing pH in response to positive-charge development on the sorbents. This pattern was consistent with a ligand-exchange mechanism, as previously suggested for humic substances by others (19, 30-33). However, only a weak effect of pH on the removal efficiency was observed in this study. The reason for the small effect is not clear. It
FIGURE 1. Removal of fulvic acid by adsorption onto silicate rocks: b, tobermorite; 1, zeolite; 4, molecular sieves 5a; fulvic acid solution, 0.02 mg/mL.
FIGURE 2. Removal of humic acid by adsorption onto silicate rocks: b, tobermorite; 1, zeolite; 4, molecular sieves 5a; humic acid solution, 0.03 mg/mL.
TABLE 1. Heat of Adsorption (Qa) for Humic Substance Adsorption onto Silicate Rocks at pH 7 heat of adsorption(kJ/mol) adsorbent
humic acid
fulvic acid
tobermorite zeolite molecular sieve 5A
>500 507 487
385 426 473
may be attributed to the differences of some factors, such as contact time and the condition of adsorbate/solute interaction. Although the surface area of tobermorite was approximately twenty times smaller than that for molecular sieves 5A, the adsorption of humic substances onto tobermorite occurred efficiently. It was remarkable that the removal efficiency of humic acid with tobermorite was almost quantitative. The usefulness of tobermorite for the removal of humic substances by adsorption may be due to its pore size and composition. The removal efficiency by adsorption of humic acid onto silicate rocks was larger than that of fulvic acid. These phenomena were expected from the gel permeation chromatography study described below. Therefore, it was found from the present study that humic acid removal from the solution by adsorption onto silicate rocks may take place more readily than for fulvic acid removal. This tendency has been reported by other researchers (34, 35). The heat of adsorption, Qa, on silicate rocks at pH 7 was measured for humic substances. Results are presented in Table 1. Whenever the heat of adsorption exceeds ∼20 kJ/ mol, the process is regarded as chemisorption, whereas physisorption occurs if Qa is less than 20 kJ/mol (24). Therefore, a high heat of adsorption in the present study could be considered as a result of chemisorption (or ligand exchange (36, 37)), as the contribution of heat from other processes, such as H-bonding, surface deprotonation during humic substance adsorption, transfer of humic substance VOL. 37, NO. 7, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1449
FIGURE 3. Removal of metal/fulvic acid complex by adsorption onto tobermorite: b, adsorption onto tobermorite; O, removal by precipitation; -----, fulvic acid (from Figure 1).
FIGURE 4. Removal of metal/humic acid complex by adsorption onto tobermorite: b, adsorption onto tobermorite; O, removal by precipitation; -----, humic acid (from Figure 2).
FIGURE 5. HPGP chromatograms of humic acid: (a) initial solution, (b) effluent from the adsorption column. from the bulk phase to the interfacial phase of silicate rocks, and the conformational changes of humic substance after adsorption, would be relatively small (36-39). Therefore, it may be understood that the adsorption is attributed to specific interactions between the reactive groups of humic acid (mainly carboxylic and phenolic) and the reactive groups of oxide surfaces and the electrostatic interactions (24, 40), as well.
Adsorption of Metal/Humic Complexes onto Tobermorite. Humic materials have been known to interact with many metal ions to form water-soluble humic complexes, which may reduce the toxicity level of simple hydrated ions and hydroxo complexes (e.g., for copper) (41, 42). This interaction may occur as chelation between a carboxyl group and a phenolic hydroxyl group, as chelation between two carboxyl groups, or as complexation with a carboxyl group (43). According to Hiraide et al. (27), nine trace metals in river water samples can be divided into three groups depending on their reactivity with humic substances: strong interaction - aluminum, iron, and copper; weak interaction - cobalt, nickel, and zinc; and no interaction - manganese, strontium, and barium. Furthermore, 30 to 70% of copper and aluminum contributed to the complexation with the humic substances. Therefore, the removal of Cu and Al/ humic complexes by adsorption with tobermorite was studied. First, the precipitation of metal/humic complexes was explored in the pH range of 3 to 10. Results are shown in Figures 3 and 4. Although aluminum/humic acid complex started to precipitate at pH values below 7, no other metal/ humic complexes could precipitate in the pH range of 5 to 9. Next, the removal efficiency for metal/humic complexes by adsorption with tobermorite was compared to that for humic substances. The efficiencies for Cu and Al/fulvic acid complexes were greater than those for fulvic acid, as shown in Figure 3. Metal/humic acid complexes could be effectively removed by adsorption onto tobermorite, as shown in Figure 4. Recently, Schulthess and Huang (23) investigated the humic and fulvic acid adsorption by silicon and aluminum oxide surfaces on clay minerals. According to this study, the adsorption efficiency of metal/humic complexes onto montmorillonite was greater than that for humic substances. This finding was related to essentially the promotion of a surfacemetal-humic substances bridging process. Therefore, the high removal efficiencies for metal/humic complexes may be attibuted to the acceleration of the bridging process. GPC Analysis. The molecular weights of humic materials before and after adsorption onto tobermorite were evaluated by HPGPC. The typical HPGP chromatograms of humic acid are shown in Figure 5. In the adsorption experiment, the amount of tobermorite used as adsorbents was lowered to 5 g in order to decrease the removal efficiency. One can see that a relatively large molecular weight range of humic acid was almost completely removed by adsorption, while a relatively small molecular weight range remained in the effluent. Therefore, the approximately 104-105 molecular weight range of humic acid appears to be preferentially
FIGURE 6. Scanning electron micrographs of the surface of tobermorite: (a) before the adsorption of humic acid, (b) after the adsorption. 1450
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 7, 2003
removed by adosrption onto tobermorite. These results were in agreement with the fact that the removal efficiency by the adsorption of humic acid onto silicate rocks was greater by 20-40% relative to that of fulvic acid. The result may be explained by the fact that humic substances with relatively large molecular weight ranges have a large number of reactive groups (e.g., carboxylic functions). Direct Observation of the Surface Conditions of Tobermorite by SEM. To the best of our knowledge, there is little information on the surface conditions of silicate rocks before and after adsorption of humic substances. Therefore, the surface conditions of tobermorite were analyzed by scanning electron microscopy (SEM). Typical results are shown in Figure 6. Before adsorption of humic acid, the fine cardhouse structure with needlelike crystals on the surface of tobermorite could be observed. However, after adsorption, the crystal line form disappeared from the SEM photograph. Hence, humic materials seem to chemisorb tobermorite with twine around its needlelike crystal. In conclusion, tobermorite is more efficient by 40-50% in the removal of fulvic acids, and by 30-50% for humic acid, than zeolite or molecular sieves, respectively. Moreover, tobermorite was an effective adsorbent for the removal of humic/metal complexes. The adsorption method for removal of fulvic and humic acids is superior to their precipitation. Because tobermorite can be easily and commercially synthesized, the water treatment system developed is rapid, simple, and inexpensive for the removal of natural organic polyelectrolytes, and it has potential for application in treatment plants for their removal, which could be adopted by developing nations.
Acknowledgments This work was partially supported by the Ministry of Education, Culture, Sports, Science, and Technology of Japan. A part of this work was performed at the Mie University Satellite Venture Business Laboratory (SBVL).
Literature Cited (1) Thurman, E. M.; Wershaw, R. L.; Malcolm, R. L.; Pinckney, D. J. Org. Geochem. 1982, 4, 27. (2) Morel, F. M. M.; Principles of Aquatic Chemistry; John Wiley and Sons: New York, 1983. (3) Buffle. J. Anal. Chem. Acta 1990, 232, 1. (4) Summers, R. S.; Haist, B.; Koehler, J.; Ritz, J.; Zimmer, G. J. Am. Water Works Assoc. 1989, 81, 66. (5) Speth, T. F. J. Environ. Eng. 1991, 117, 66. (6) Carter, M. C.; Weber, W. J., Jr.; Olmstead, K. P. J. Am. Water Works Assoc. 1992, 73, 81. (7) Pirbazari, M.; Weber, W. J., Jr. J. Environ. Eng. 1984, 110, 656. (8) Smith, E. H.; Tseng, S.; Weber, W. I., Jr. Environ. Prog. 1987, 6, 18. (9) Speth, T. F.; Miltner, R. J. J. Am. Water Works Assoc. 1989, 81, 141. (10) Summers, R. S.; Roberts P. V. J. Colloid Interface Sci. 1988, 122, 367.
(11) Ogino, K.; Yukihiro, K.; Minoura, T.; Agui, W.; Abe, M. J. Colloid Interface Sci. 1988, 121, 161. (12) Bouchard, C. R.; Jolicoeur, J.; Kouadio, P. Can. J. Chem. Eng. 1997, 75, 339. (13) Adam, U. S.; Robb, I. D. J. Chem. Soc. Faraday Trans. 1 1983, 79, 2745. (14) Fu, P. L. K.; Symons, J. M. J. Am. Water Works Assoc. 1990, 82, 70. (15) Bernand, S.; Chazal, Ph.; Mazet, M. Water Res. 1997, 31, 1216. (16) Gloor, R.; Leidner, K.; Wuhrmann, K.; Fleishmann, T. Water Res. 1981, 15, 457. (17) Elliott, H. A.; Huang, C. P. Water Res. 1981, 15, 849. (18) Davis, A. P.; Bhatnagar, V. Chemosphere 1995, 30, 243. (19) Murphy, E. M.; Zachara, J. M.; Smith, S. C. Environ. Sci. Technol. 1990, 24, 1507. (20) Wang, L.; Chin, Y.; Traina, S. J. Geochim. Cosmochim. Acta 1997, 24, 5313. (21) Parfitt, R. L.; Fraser, A. R.; Farmer, V. C. J. Soil Sci. 1977, 28, 289. (22) Avena, M. J.; Koopal, L. K. Environ. Sci. Technol. 1999, 33, 2739. (23) Schulthess, C. P.; Huang, C. P. Soil Sci. Soc. Am. J. 1991, 55, 34. (24) Gu, B.; Schmitt, J.; Chen, Z.; Liang, L.; McCarthy, J. F. Environ. Sci. Technol. 1994, 28, 38. (25) Kaneco, S.; Inomata, K.; Itoh, K.; Funasaka, K.; Masuyama, K.; Itoh, S.; Suzuki, T.; Ohta, K. J. Urban Living Health Assoc. 2000, 44, 211. (26) Swift, R. S. In Methods of Soil Analysis; Sparks, D. L., Ed.; Soil Science Society of America; Book Series 5; SSSA: Madison, WI, 1996; pp 1018-1020. (27) Hiraide, M.; Hiramatsu, S.; Kawaguchi, H. Fresenius’ J. Anal. Chem. 1994, 348, 758. (28) Sachs, S.; Bubner, M.; Schmeide, K.; Choppin, G. R.; Heise, K. H.; Bernhard, G. Talanta 2002, 57, 999. (29) Kuwatsuka, S.; Watanabe, A.; Itoh, K.; Arai, S. Soil Sci. Plant Nutr. 1992, 38, 23. (30) Kummert, R.; Stumm, W. J. Colloid Interface Sci. 1980, 75, 373. (31) Tripping, E. Environ. Sci. Technol. 1981, 15, 191. (32) Davis, J. A. Geochim. Cosmochim. Acta 1962, 46, 2381. (33) Tripping, E.; Cooke, D. Geochim. Cosmochim. Acta 1982, 46, 75. (34) Meier, M.; Namjesnik-Dejanovic, K.; Maurice, P. A.; Chin, Y.; Aiken, G. R. Chem. Geology 1999, 157, 275. (35) Specht, C. H.; Kumke, M. U.; Frimmel, F. H. Wat. Res. 2000, 34, 4063 (36) Afzal, M.; Khan M.; Ahmad, H. Colloid Polym. Sci. 1991, 269, 483. (37) Gatta, G. D. Thermochim. Acta 1985, 96, 349. (38) Zhang, Z. Z.; Low, P. F.; Cushman, J. H.; Roth, C. B. Soil Sci. Soc. Am. J. 1990, 54, 59. (39) De Keizer, A.; Fokkink, L. G. J.; Lyklema, J. Colloid Surf. 1990, 49, 149. (40) Vermeer, A. W. P.; van Riemsdijk, W. H.; Koopal, L. K. Langmuir 1998, 14, 2810. (41) Florence, T. M. Talanta 1982, 29, 345. (42) Nurnberg, H. W. Fresenius’ J. Anal. Chem. 1983, 316, 557. (43) Manahan, S. E. Environmental Chemistry; 5th ed.; Lewis Publishers: Chelsea, MI, 1991.
Received for review July 1, 2002. Revised manuscript received January 6, 2003. Accepted January 20, 2003. ES020816V
VOL. 37, NO. 7, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1451
