Environ. Sci. Technol. 2002, 36, 2515-2518
Porous Polysilsesquioxanes for the Adsorption of Phenols MARK C. BURLEIGH, MICHAEL A. MARKOWITZ,* MARK S. SPECTOR, AND BRUCE P. GABER Laboratory for Molecular Interfacial Interactions, Code 6930, Center for Bio/Molecular Science and Engineering, Naval Research Laboratory, Washington, DC 20375
Arylene- and ethylene-bridged polysilsesquioxane materials have been synthesized by the hydrolysis and condensation of alkoxysilyl precursors under basic conditions. Cetyltrimethylammonium chloride was used to increase the porosity and surface areas of these materials via the surfactant template approach. Structural characterization of these materials was carried out by nitrogen gas sorption and X-ray diffraction. The adsorption of three phenolic compounds (4-nitrophenol, 4-chlorophenol, 4-methylphenol) has been investigated by both batch and column testing. The arylene-bridged material exhibited a much greater affinity for all three phenols. The efficient removal of adsorbed phenols by a simple ethanol wash led to sorbent regeneration and separation of the aromatic species.
Introduction The introduction of toxic pollutants can have a severe impact on the many organisms that live in aquatic ecosystems. Phenols are some of the most common hazardous chemicals introduced by man into such environments. Nitrophenols are introduced by agricultural runoff and effluent waste from the pesticide, explosives, and dye industries (1). Chlorinated phenols are used as decoloring agents in pulp and paper manufacture, as wood preservatives, and are a decomposition product of many herbicides (2, 3). Of major concern is the stability of these compounds once they enter the environment. Phenols are persistent pollutants that have been found in water, soil, and air samples around the globe as well as in animal tissues (4, 5). They have been designated as priority pollutants by the U.S. Environmental Protection Agency (6). A variety of techniques have been implemented to purify water contaminated with phenols. Ozonolysis (7), photolysis (8), and photocatalytic decomposition (9) have all been used with limited success. The adsorption of phenols onto solid supports such as activated carbons allows for their removal from water without the addition of chemicals or UV radiation. Activated carbons have a large adsorption capacity for a variety of organic pollutants but are expensive due to their difficult regeneration (10, 11) and high disposal costs. There is a clear need for new adsorbent technologies for the removal of phenols from wastewater and the remediation of contaminated sites. For any sorbent to be feasible, it must combine high adsorption capacity and fast adsorption kinetics with inexpensive regeneration. Functionalized mesoporous molecular sieves may be designed to satisfy all three criteria. * Corresponding author phone: (202)404-6072; fax: (202)767-9594; e-mail:
[email protected]. 10.1021/es011115l CCC: $22.00 Published on Web 04/25/2002
2002 American Chemical Society
FIGURE 1. Polysilsesquioxane precursors: (a) 1,4-bis(trimethoxysilylethyl)benzene and (b) 1,2-bis(triethoxysilyl)ethane. Mesoporous molecular sieves are synthesized using a surfactant template approach that creates large surface areas and internal pore volumes (12, 13). These characteristics have led to their application as catalytic and adsorbent materials (14, 15). Although it has been used primarily to synthesize metal oxides (16, 17), the surfactant template approach is an attractive route for engineering the porosity of a wide variety of new materials. Polysilsesquioxanes are hybrid organicinorganic polymers that contain organic moieties covalently bonded to siloxane groups. Typically, they have been synthesized by the hydrolysis and condensation of organic poly(alkoxysilyl) monomers (18, 19). Researchers have used the surfactant template approach in order to create polysilsesquioxanes with ordered, mesoporous structures (2023). A wide variety of organic bridging groups can been used to give these materials various properties. This ability to design polymeric materials that exhibit specific combinations of organic and inorganic properties gives polysilsesquioxanes the potential for many applications. Herein, we describe the synthesis and characterization of mesoporous polsilsesquioxane materials and their application as sorbent materials for the removal of phenols from aqueous solutions. Supramolecular assemblies of cetyltrimethylammonium chloride (CTAC) are used to engineer the porosity in these materials by surfactant templating (24). The organosilane monomers 1,2-bis(triethoxysilyl)ethane and 1,4-bis(trimethoxysilylethyl)benzene (Figure 1) are polymerized by base-catalyzed hydrolysis and condensation reactions to form organic-inorganic materials with ethylene and diethylbenzene organic bridging groups, respectively. The CTAC is then removed by extraction in acidified ethanol, giving the porous polysilsesquioxane. These materials have been characterized by nitrogen gas sorption and X-ray diffraction (XRD). The ability of these materials to adsorb 4-nitrophenol (4-NP), 4-chlorophenol (4-CP), and 4-methylphenol (4-MP), has been investigated by both batch and column testing. Regeneration of the arylene-bridged sorbent by washing in absolute ethanol has been investigated.
Materials and Methods Chemicals. 1,2-Bis(triethoxysilyl)ethane and 1,4-bis(trimethoxysilylethyl)benzene were obtained from Gelest, Inc. NaOH, HCl, CTAC, and EtOH were obtained from Aldrich. All chemicals were used as received. Water used in all synthetic procedures was deionized to 18.0 MΩ cm. Synthesis. We prepared the polysilsesquioxanes by basecatalyzed hydrolysis and condensation of the alkoxysilyl precursors around supramolecular assemblies of CTAC. In a typical synthesis, 6 mL of CTAC (25 wt %) and 2.6 mL of 25% NaOH were added under stirring to 100 mL of deionized water. To this mixture was added 9.4 mL of 1,4-bis(trimethoxysilylethyl)benzene. The reaction flask was then VOL. 36, NO. 11, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2515
TABLE 1. Nitrogen Gas Adsorption Data of Porous Sorbents polysilsesquioxane
BET surface area (m2/g)
total pore volume (cm3/g)
BJH pore diameter (Å)
arylene-bridged ethylene-bridged
550 1300
0.40 1.10
30 33
TABLE 2. Percent Uptake of 4-Nitrophenol on Polysilsesquioxane Sorbentsa
FIGURE 2. X-ray diffraction pattern of ethylene-bridged polysilsesquioxane. covered, and the mixture was stirred at room temperature until gelation (about 2 h). Postsynthetic Treatment. The synthesis gels were heated at 70 °C for 48 h. The resulting as-synthesized materials were then placed in excess (350 mL/g) acidified ethanol (1 M HCl) and refluxed for 6 h to extract the surfactant templates. The products were recovered by filtration and washed with absolute ethanol. The extraction procedure was then repeated. The samples were then dried under vacuum at 60 °C for 10 h. Characterization. X-ray diffraction measurements were made on an Enraf-Nonius FR591 rotating-anode using a bent graphite monochromator that selected Cu KR radiation and provided in-plane resolution of 0.014 Å-1 full-width at halfmaximum. Powder samples were placed in 1.0 mm quartz capillary tubes. Surface area experiments were performed using a Micromeritics ASAP 2010 gas sorption analyzer. Nitrogen gas was used as the adsorbate at 77 K. Batch tests were conducted with the porous sorbents and aqueous solutions of phenolic compounds. The quantity adsorbed was calculated by the difference in concentration between the filtrate and the initial standard solutions as measured by UV/vis spectroscopy. UV-visible spectra were measured with a Cary 2000 scanning spectrophotometer. Typical batch tests were conducted using 100 mg sorbent and 10 mL standard solutions for a given contact time. Column tests were performed by mixing 1 g of sorbent with 19 g of sea sand (1 bed volume ) 15 mL), placing it in a 13 mm column, and eluting with 10-3 M 4-NP.
Results and Discussion X-ray Diffraction. Powder X-ray diffraction analyses were performed on both of the extracted polysilsesquioxane materials. The ethylene-bridged sample exhibits a prominent peak in the diffraction pattern at approximately 2θ ) 2 degrees (Figure 2). This peak corresponds to the d100 reflection in materials with short-range hexagonal order. Similar diffraction patterns have been reported for MCM-41 and other ordered mesoporous materials such as HMS (25). No peaks were resolved in the diffraction pattern of the arylene-bridged material. Nitrogen Sorption. The porous polysilsesquioxanes were characterized by N2 sorption at 77 K. The resulting adsorption/desorption experiments yielded type IV isotherms (Figure 3). The adsorption branch of the arylene-bridged material exhibits a linear region from P/Po ∼0.2-0.7, characteristic of a broad pore size distribution in the mesoporous range. The structural parameters calculated from nitrogen gas sorption measurements of the two polysilsesquioxanes are shown in Table 1. The ethylene-bridged material exhibits more than twice the surface area and porosity than that of the arylene-bridged material. The X-ray diffraction and nitrogen gas sorption experiments clearly 2516
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 11, 2002
initial phenol concn (M)
arylene
ethylene
1.0 × 10-4 2.5 × 10-4 5.0 × 10-4 7.5 × 10-4 1.0 × 10-3
99.4 99.1 98.7 98.2 96.9
16.5 16.3 15.8 15.3 15.2
a Uptakes are an average of three separate batch tests, ( 0.3%, contact time 15 min.
TABLE 3. Percent Uptake of Phenols on Arylene-Bridged Polysilsesquioxanea initial phenol concn (M)
adsorbed 4-NP (%)
adsorbed 4-CP (%)
adsorbed 4-MP (%)
1.0 × 10-4 2.5 × 10-4 5.0 × 10-4 7.5 × 10-4 1.0 × 10-3
99.4 99.1 98.7 98.2 96.9
98.6 98.3 97.5 95.3 90.0
97.9 97.2 96.3 95.2 92.4
a Uptakes are an average of three separate batch tests, ( 0.3%, contact time 15 min.
indicate that the arylene-bridged material forms a more disordered, less porous matrix than that of the corresponding ethylene-bridged material.
Adsorption of Phenols Batch Testing. We performed batch tests using both the diethylbenzene and ethylene-bridged polysilsesquioxane materials to study their ability to adsorb 4-NP. A direct comparison (Table 2) shows that while 0.10 g of the arylenebridged sorbent removed greater than 99% of the 4-nitrophenol from 10 mL of a 10-4 M solution in 15 min, the ethylene-bridged sorbent removed only 16.5%. This, despite the fact that nitrogen gas adsorption experiments show that the ethylene-bridged material has more than twice the surface area and pore volume of the arylene-bridged sorbent. Based on surface area, the arylene-bridged material adsorbed 14 times more 4-NP than the ethylene-bridged sorbent. The arylene-bridged polysilsesquioxane contains siloxane, ethylene, and arylene functional groups. The high affinity of this material for phenolic compounds must then be due to interactions between one or more of these groups and the phenolic species. The ethylene-bridged polysilsesquioxane contains both siloxane and ethylene functionalities but does not exibit such a strong affinity for the phenols. Based on these observations, we feel that π-π interactions between the aromatic rings (26) of the arylene-bridged sorbent and the phenolic compounds play an important role in the binding between them. The efficient separation of 4-NP from aqueous solutions by our porous arylene-bridged polysilsesquioxane led us to study the ability of this material to adsorb the similar phenolic species, 4-chlorophenol (4-CP) and 4-methylphenol (4-MP). Table 3 compares the uptake of the 4-CP and 4-MP with that of 4-NP from solutions ranging from 10-4-10-3 M. The arylene-bridged sorbent exhibits efficient adsorption of all
TABLE 4. Percent Uptake of Phenols on Activated Carbon and Arylene-Bridged Sorbentsa initial phenol concn (M)
adsorbed 4-NP (%)
adsorbed 4-CP (%)
adsorbed 4-MP (%)
activated carbon arylene-bridged
99.9 98.7
96.4 98.1
98.6 95.5
a Uptakes are an average of three separate batch tests, ( 0.3%, contact time 60 min.
FIGURE 5. Breakthrough curves of 4-NP on arylene-bridged polysilsesquioxane and activated carbon columns.
FIGURE 3. Nitrogen adsorption/desorption isotherms of the (a) ethylene- and (b) arylene-bridged materials.
FIGURE 4. Adsorption kinetics of 4-NP on arylene-bridged polysilsesquioxane. three of these compounds within the 15 min contact time. These results indicate that this material may provide utility in the separation of a wide variety of monocyclic aromatic hydrocarbons from wastewater. Batch testing was also performed with a commercially available activated carbon (27) and the three phenolic compounds at a concentration of 10-3 M and a 60 min contact time. A comparison of the uptakes of the arylene-bridged sorbent with the activated carbon is shown in Table 4. The performance of the porous organosilica is comparable with that of the activated carbon. The adsorption kinetics of 4-NP by the arylene-bridged sorbent were investigated by performing batch procedures for specific contact times. Figure 4 illustrates the overall trend. Greater than 96% of the 4-NP was adsorbed during the first
minute. A 15 min contact time resulted in the removal of more than 99% of the adsorbate. Sorbent Regeneration. Important goals in the development of sorbent materials include simple regeneration and adsorbate isolation. Regeneration allows for the repeated use of the sorbent material, decreasing costs, and solid waste production. Isolation of adsorbate molecules can allow for their reuse or efficient disposal. We have accomplished both of the above goals by washing the loaded sorbent in ethanol. The high solubility of the phenols in ethanol leads to desorption back into the liquid phase. Desorption of the adsorbate molecules from the loaded polysilsesquioxane using alcohol was rapid. For example, a 0.10 g sample of the sorbent that had adsorbed greater than 99% of the 4-NP from a 10 mL aliquot of aqueous 10-4 M solution in 15 min was then placed in 10 mL of absolute ethanol for 15 min. The regenerated sorbent was then separated by filtration. Analysis of the ethanol by UV/vis spectroscopy showed greater than 98% of the adsorbate was removed from the sorbent. The ethanol was then recovered by distillation, leaving residual 4-NP. The activated carbon sample also adsorbed >99% of the 4-NP from an aqueous 10-4 M solution, but a similar alcohol wash was ineffective. Only 1-2% of the total 4-NP was removed. Column Testing. Column tests were also performed with both the arylene-bridged sorbent and activated carbon, using 10-3 M 4-NP as the eluent. The effluent was monitored for the presence of 4-NP. The breakthrough curves are shown in Figure 5. The arylene-bridged sorbent exhibited a sharp breakthrough at 24 bed volumes (BV). The activated carbon did not start to break through until 60 BV, and breakthrough was not complete until 160 BV. The distinctly different shapes of these curves may be related to the homogeneity of the adsorption sites in these materials. A distribution of adsorption sites of varying affinities on the activated carbon could explain the strong adsorption from 0 to 60 BV, followed by much weaker adsorption from 60 to 160 BV. Column Regeneration. The column containing the arylene-bridged sorbent was regenerated by washing with ethanol. Following breakthrough, all eluent was allowed to pass through the column, and 1 BV of absolute ethanol was added. The ethanol was collected as a single aliquot. This was followed by the addition of 1 BV of deionized water to the column to rinse out the alcohol. The column was then reused. Ten successive regeneration cycles were performed. Breakthrough occurred after 22-24 BV for all cycles. Characterization of the alcohol washings indicated that >95% of the 4-NP was removed. No decrease in overall column performance was exhibited. In fact, the breakthrough curves for the first and tenth cycles are hard to distinguish (Figure 6). Mesoporous arylene and ethylene-bridged polysilsesquioxane materials have been synthesized from alkoxysilyl VOL. 36, NO. 11, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2517
Literature Cited
FIGURE 6. Breakthrough curves of 4-NP on arylene-bridged polysilsesquioxane column for first and tenth cycles. precursors using a surfactant template approach. The arylene-bridged material has been shown to be an effective adsorbent for phenols. Greater than 95% of 4-NP was removed from a loaded sample by washing with ethanol. This easy regeneration is a potential advantage in the use of the porous polysilsesquioxanes over conventional activated carbons for the removal of phenolic compounds from aqueous solutions. This material exhibits fast adsorption kinetics as it adsorbs greater than 96% of 4-NP during the first minute of a typical batch test. Column testing with a 10-3 M 4-NP solution indicates a high adsorption capacity and facile regeneration with ethanol. Structural characterization by XRD and nitrogen gas sorption indicate that the ethylene-bridged polysilsesquioxane has a much larger surface area and porosity but fails to efficiently remove phenols from aqueous solutions. This seems to indicate that π-π interactions between the aromatic rings of the phenols and the arylene-bridged sorbent have an important role in the binding between them. In conclusion, high adsorption capacities, fast adsorption kinetics, and easy regeneration make this porous arylenebridged polysilsesquioxane a candidate for the removal of phenols from wastewater and contaminated sites. Further work needs to be done to better understand the mechanism of adsorption of phenols on these materials. Future research will investigate the ability of the arylene-bridged polysilsesquioxane to adsorb other monocyclic aromatic pollutants (i.e. BTEX), more detailed desorption/regeneration studies, and the development of selective sorbent materials.
Acknowledgments Mark C. Burleigh is a NRC postdoctoral research fellow. This project was funded by the Office of Naval Research through a Naval Research Laboratory Accelerated Research Initiative.
2518
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 11, 2002
(1) List of Worldwide Hazardous Chemicals and Pollutants, The Forum for Scientific Excellence; J. B. Lippincot Co.: New York, 1990. (2) Chapman, P. M.; Romberg, G. P.; Vigers, C. A. J. Wat. Poll. Control Fed. 1982, 54, 292. (3) Exon, J. H. Vet. Hum. Toxicol. 1984, 26, 508. (4) Crosby, D. G. IUPAC Reports Pesticides 1981, 14, 1051. (5) Paasivirta, J.; Heinola, K.; Karjalainen, A.; Knuutinen, J.; Mantykoski, K.; Paukku, R.; Piilola, T.; Surma-Aho, K.; Tarhanen, J.; Weling, L.; Vihonen, H.; Sarka, J. Chemosphere 1985, 14, 469. (6) U.S. Environmental Protection Agency. 4-nitrophenol, Health and Environmental Effects Profile No. 135, Washington, DC, 1980. (7) Beltran, F. J.; Gomez-Serrano, V.; Duran, A. Water Res. 1992, 26, 9. (8) Chow, Y. L. Photochemistry of Nitro and Nitroso Compounds. In The Chemistry of Functional Groups; Patai, S., Ed.; J. Wiley & Sons: Chichester, 1982; Supplement F, p 181. (9) Balcioglu, I. A.; Inel, Y. Environ. Sci. Health 1996, A(31), 123. (10) Bercic, G.; Pintar, A. Ind. Eng. Chem. Res. 1996, 35, 4619. (11) Matatov, Y.; Sheintuch, M. Ind. Eng. Chem. Res. 2000, 39, 18. (12) Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.; Schmitt, K. D.; Chu, C. T.; Olsen, D. H.; Sheppard, E. W.; McCullen, S. B.; Higgens, J. B.; Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834. (13) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710. (14) Pinnavaia, T. J.; Tanev, P. In Access in Nanoporous Materials; Pinnavaia, T. J., Thorpe, M. F., Eds.; 1995; p 55. (15) Corma, A. Chem. Rev. 1997, 97, 2373. (16) Bagshaw, S. A.; Prouzet, E.; Pinnavaia, T. J. Science 1995, 269, 1242. (17) Huo, Q.; Margolese, D. I.; Stucky, G. D. Chem. Mater. 1996, 8, 1147. (18) Loy, D. A.; Shea, K. J. Chem. Rev. 1995, 95, 1431. (19) Cerveau, G.; Corriu, R. J. P. Coord. Chem. Rev. 1998, 178-80, 1051. (20) Inagaki, S.; Guan, S.; Fukushima, Y.; Ohsuna, T.; Terasaki, O. J. Am. Chem. Soc. 1999, 121, 9611. (21) Melde, B. J.; Holland, B. T.; Blanford, C. F.; Stein, A. Chem. Mater. 1999, 11, 3302. (22) Asefa, T.; MacLachlan, M. J.; Coombs, N.; Ozin, G. A. Nature 1999, 402, 867. (23) Chiaki, Y. I.; Tewodros, T.; Coombs, N.; MacLachlan, M. J.; Ozin, G. A. Chem. Commun. 1999, 2539. (24) Stein, A.; Melde, B. J.; Schroden, R. C. Adv. Mater. 2000, 12, 1403. (25) Zhang, W.; Pauly, T. R.; Pinnavaia, T. J. Chem. Mater. 1997, 9, 2491. (26) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525. (27) Activated carbon: 100 mesh powder, acid washed, steam activated, BET surface area 480 m2/g, was purchased from J. T. Baker.
Received for review July 5, 2001. Revised manuscript received March 5, 2002. Accepted March 5, 2002. ES011115L
