pubs.acs.org/Langmuir © 2010 American Chemical Society
)
Size Matters: Incorporation of Poly(acrylic acid) and Small Molecules into Hierarchically Porous Metal Oxides Prepared with and without Templates Glenna L. Drisko,† Paolo Imperia,§ Massey de los Reyes,§ Vittorio Luca,*, and Rachel A. Caruso*,†,‡ †
)
Particulate Fluids Processing Centre, School of Chemistry, The University of Melbourne, Melbourne VIC 3010, Australia, ‡CSIRO Materials Science and Engineering, Private Bag 33, Clayton South VIC 3169, Australia, § The Bragg Institute and Institute of Materials Engineering, Australian Nuclear Science and Technology Organization, Lucas Heights NSW 2234, Australia, and Comisi on Nacional de Energı´a At omica, Gerencia Quimica, Avenida General Paz 1499, San Martin, Provincia de Buenos Aires 1650, Argentina Received April 9, 2010. Revised Manuscript Received June 9, 2010
Template synthesis of metal oxides can create materials with highly controlled and reproducible pore structures that can be optimized for particular applications. Zirconium titanium oxides (25:75 mol %) with three different pore structures were synthesized in order to relate polymer loading capacity to macropore architecture. Sol-gel chemistry was used to prepare the materials in conjunction with (i) agarose gel templating, (ii) no template, and (iii) stearic acid templating. The three materials possessed high surface areas (212-316 m2 g-1). Surface modification was performed postsynthetically using propionic acid (a monomer), glutaric acid (a dimer), and three molecular weights of poly(acrylic acid) (2000, 100 000, and 250 000 g mol-1). Higher loading (mg g-1) was observed for the polymers than for the small molecules. Following surface modification, a perceptible decrease in surface area and mesopore volume was noted, but both mesoporosity and macroporosity were retained. The pore architecture had a strong bearing on the quantity and rate of polymer incorporation into metal oxides. The templated pellet with hierarchical porosity outperformed the nontemplated powder and the mesoporous monolith (in both loading capacity and surface coverage). The materials were subjected to irradiation with 60Co γ-rays to determine the radiolytic stability of the inorganic support and the hybrid material containing the monomer, dimer, and polymer. The polymer and the metal oxide substrate demonstrated notable radiolytic stability.
Introduction The world is currently facing an energy crisis as population and energy consumption continue to grow while humans try to reduce their dependence on carbon-rich energy sources such as coal, oil, and gas. Many countries are increasing atomic energy production to lower carbon dioxide emissions, but this creates radioactive waste that presents storage and disposal problems. To address these concerns, porous hybrid materials are being developed for use as extraction agents with selectivity for particular radioisotopes present in radioactive waste and reprocessing streams.1 This application requires that the materials used be highly radiolytically and hydrolytically stable. Synroc is an example of a radiolytically stable material, composed of a variety of elements (primarily from the minerals hollandite BaAl2Ti6O16, zirconolite CaZrTi2O7, and perovskite CaTiO3).2 It has been extensively studied as a radioactive waste storage material but synroc has yet to be structured into an adsorbent. Adsorbents and many other materials (e.g., sensors, photocatalysts, biocatalysts, batteries, and fuel cells) could achieve higher performance if the internal as well as the external surface could be better utilized. The pore architecture can greatly affect the performance of adsorbents,3 but the optimum structure for adsorbents is yet to be determined. Hierarchical pore structures facilitate fluid diffusion through a material while maintaining a high active surface area.3,4 Fast fluid *Corresponding authors. E-mail:
[email protected], rcaruso@unimelb. edu.au. (1) Fryxell, G. E.; Lin, Y.; Fiskum, S.; Birnbaum, J. C.; Wu, H.; Kemner, K.; Kelly, S. Environ. Sci. Technol. 2005, 39, 1324. (2) Ringwood, T. Am. Sci. 1982, 70, 201. (3) Drisko, G. L.; Luca, V.; Sizgek, E.; Scales, N.; Caruso, R. A. Langmuir 2009, 25, 5286. (4) Antonietti, M.; Ozin, G. A. Chem.;Eur. J. 2004, 10, 28.
Langmuir 2010, 26(17), 14203–14209
flux rate is an important attribute for adsorbents. Macropore and mesopore structures can be created in a very simple, inexpensive, template-free synthesis.5-10 Although regular and complex pore architectures are produced through this method, the dimensions of the pores cannot be readily adjusted. Template synthesis can produce complex pore structures with a large degree of control over multiple length scales;11 however, this process often requires additional steps and/or reagents leading to higher production costs. Thus far, the sorption properties and capacity of nontemplated and templated materials have not been compared. Hybrid materials, consisting of a molecular mixture of inorganic and organic components,12 could enhance adsorbent properties by combining the mechanical stability of the inorganic with the selectivity of functionalized organic molecules. Some researchers have modified porous metal oxides through in situ polymerization to form an interpenetrating network within the pores13 or from the surface of the inorganic material.14,15 Although this can be an efficient route to surface modification, (5) Yu, J.; Su, Y.; Cheng, B. Adv. Funct. Mater. 2007, 17, 1984. (6) Zhu, J.; Wang, S.-H.; Bian, Z.-F.; Cai, C.-L.; Li, H.-X. Res. Chem. Intermed. 2009, 35, 769. (7) Collins, A.; Carriazo, D.; Davis, S. A.; Mann, S. Chem. Commun. 2004, 568. (8) Leonard, A.; Su, B.-L. Chem. Commun. 2004, 1674. (9) Deng, W.; Shanks, B. H. Chem. Mater. 2005, 17, 3092. (10) Yuan, Z.-Y.; Ren, T.-Z.; Azioune, A.; Pireaux, J.-J.; Su, B.-L. Chem. Mater. 2006, 18, 1753. (11) Drisko, G. L.; Zelcer, A.; Luca, V.; Caruso, R. A.; Soler-Illia, G. J. A. A. Chem. Mater.; DOI: 10.1021/cm100764e. (12) Sanchez, C.; Julian, B.; Belleville, P.; Popall, M. J. Mater. Chem. 2005, 15, 3559. (13) Strandwitz, N. C.; Nonoguchi, Y.; Boettcher, S. W.; Stucky, G. D. Langmuir 2010, 26, 5319. (14) Calvo, A.; Yameen, B.; Williams, F. J.; Azzaroni, O.; Soler-Illia, G. J. A. A. Chem. Commun. 2009, 2553. (15) Yameen, B.; Kaltbeitzel, A.; Glasser, G.; Langner, A.; M€uller, F.; G€osele, U.; Knoll, W.; Azzaroni, O. ACS Appl. Mater. Interfaces 2010, 2, 279.
Published on Web 07/27/2010
DOI: 10.1021/la101415c
14203
Article
it is difficult to control the properties of polymers that are grown within the support. In some cases, it is important to incorporate a well-defined polymer or a preformed macromolecule (e.g., enzymes and dendrimers), requiring postsynthetic incorporation of the surface functionality. Although the adsorption of macromolecules onto flat or external surfaces is well documented and well understood,16-19 the use of internal space has only recently been documented. A few studies have been undertaken on the incorporation of polymers into porous metal oxides20 and silica.21,22 Polyelectrolytes can be adsorbed onto the surfaces of inorganic materials through electrostatic attraction or hydrogen bonding with surface hydroxyl groups.20-22 For example, poly(acrylic acid) (PAA) has been adsorbed onto inorganic surfaces for a wide variety of applications: as an industrial dispersing or flocculating agent,23 as a compatibilizer for coated silica supports and quantum dots,24 and in the layer-by-layer assembly onto silica particles.25 The first study of polyelectrolyte adsorption within a porous inorganic material examined poly(N-methyl-4vinyl pyridine) incorporation into silica gel with a variety of pore sizes, where it was found that materials with small mesopore sizes had decreased adsorption and increased effects from ionic strength.21 Wang et al. have conducted a detailed study on the adsorption of PAA into confined monomodal mesoporous silica nanoparticles and found that larger mesopores could accommodate a larger range of polymer molecular weights.22 PAA has been incorporated onto porous silica beads,26 showing a change in the volume of porosity of the modified support with a change in pH. The change in porosity was attributed to chain extensions of the polymer, which are pH-dependent. The current article reports a comparison of the polymer adsorption behavior of materials with different macropore architectures by small molecules and PAA. Zirconium titanium oxide was chosen as the support because it is compositionally similar to synroc and thus could represent a highly radiation-resistant material, and because it is a simplified form, it is easier to synthesize into structured materials. Zirconium titanium oxide can be prepared using a wide range of Zr:Ti ratios; the ratio used here has previously been found to give the highest surface area materials.3 To enhance radionuclide adsorption by metal oxide supports, we have begun to investigate their surface modification using organic molecules and polymers. The incorporation of a monomer (propionic acid, PA), a dimer (glutaric acid, GA), and three molecular weights of polymer (PAA) was studied (Scheme 1). The sorption properties and loading capacity of two templated porous monolithic materials and a nontemplated porous powder were compared. Additionally, the metal oxide before and after surface modification with small molecules and PAA was irradiated with 60Co γ-rays to gain some knowledge regarding its (16) Borkovec, M.; Papastavrou, G. Curr. Opin. Colloid Interface Sci. 2008, 13, 429. (17) Netz, R. R.; Andelman, D. Phys. Rep. 2003, 380, 1. (18) Claesson, P. M.; Poptoshev, E.; Blomberg, E.; Dedinaite, A. Adv. Colloid Interface Sci. 2005, 114-115, 173. (19) Dobrynin, A. V.; Rubinstein, M. Prog. Polym. Sci. 2005, 30, 1049. (20) Drisko, G. L.; Cao, L.; Chee Kimling, M.; Harrisson, S.; Luca, V.; Caruso, R. A. Appl. Mater. Interfaces 2009, 1, 2893. (21) Mishael, Y. G.; Dubin, P. L.; de Vries, R.; Kayitmazer, A. B. Langmuir 2007, 23, 2510. (22) Wang, Y.; Angelatos, A. S.; Dunstan, D. E.; Caruso, F. Macromolecules 2007, 40, 7594. (23) Chibowski, S.; Wisniewska, M. Colloids Surf., A 2002, 208, 131. (24) Allen, C. N.; Lequeux, N.; Chassenieux, C.; Tessier, G.; Dubertret, B. Adv. Mater. 2007, 19, 4420. (25) Such, G. K.; Tijipto, E.; Postma, A.; Johnston, A. P. R.; Caruso, F. Nano Lett. 2007, 7, 1706. (26) Suzuki, K.; Siddiqui, S.; Chappell, C.; Siddiqui, J. A.; Ottenbrite, R. M. Polym. Adv. Technol. 2000, 11, 92.
14204 DOI: 10.1021/la101415c
Drisko et al. Scheme 1. Molecular Structures of PA (1), GA (2), and PAA (3)
radiolytic stability and thus the suitability of the material as a radioactive waste adsorbent.
Experimental Section Materials. Agarose powder (Scientifix), ethanol (Merck), hydrochloric acid (Scharlou), sodium hydroxide (Merck), ammonium hydroxide (Sigma-Aldrich), all molecular weights of PAA (Sigma-Aldrich), propionic acid (Sigma-Aldrich), glutaric acid (Sigma-Aldrich), Triton 100x (Sigma-Aldrich), stearic acid (Sigma-Aldrich), 97% titanium(IV) isopropoxide (SigmaAldrich), and 70% zirconium propoxide (Sigma-Aldrich) were all used as received. Templating of Agarose Gel to Form Metal Oxide Pellets. Agarose gel templating was performed as described in refs 3 and 26. Specifically, 2 wt % agarose gel was used as a template. A precursor solution was prepared by combining 97% titanium(IV) isopropoxide (37.5 g) with 70% zirconium(IV) propoxide (20 g) and isopropanol (11.2 g). The gels were soaked in 25:75 mol % zirconium(IV) propoxide:titanium(IV) isopropoxide for 18 h. Following saturation, the gels were transferred to a 1:1 volume isopropanol:water solution and gently agitated for 6 h. The gels were dried at room temperature for four days followed by one day at 60 C. The template was removed through calcination at 420 C for 5 h under air flow and a heating ramp of 2 C min-1. Synthesis of Nontemplated Powders. Zirconium propoxide (10.80 g) and titanium isopropoxide (19.72 g) were mixed and then poured into 1:1 isopropanol:water to form a 25:75 mol % Zr:Ti powder. The mixture was stirred for 1.5 h to ensure complete hydrolysis of the precursors and then drained. The powder was allowed to dry at room temperature for 4 days, followed by 1 day at 60 C, and was then calcined at 420 C for 5 h under air flow using a 2 C min-1 temperature ramp. Synthesis of Mesoporous Monoliths. A literature report was replicated here.28 Specifically, a 25:75 mol % Zr:Ti solution was made by combining 197.0 g of 97% titanium isopropoxide with 108.1 g of 70% zirconium propoxide. Stearic acid (85.8 g) was dissolved in the precursor solution (150.4 g). The sample was heated to 70 C for 1.5 h in order to dissolve the surfactant fully. Then the solution was aged by heating to 90 C for 16.5 h in a sealed Pyrex bottle. The solution was uncapped and placed in a humidity chamber at 35 C and ∼50% relative humidity for 2 days. At this point, the solution had decreased by 20 vol %. This solution was dripped into 500 mL of 3 M NH4OH containing 1 drop of Triton 100x, causing precipitation of the metal oxide and stearic acid. The solid was stirred in the ammonium hydroxide solution for 30-45 min, then drained and stirred in Milli-Q water for 30 min. The precipitate was washed five times with Milli-Q water and then allowed to dry at room temperature for 1 day. The solid was dried at 40 C for 36 h and calcined at 420 C for 5 h under air flow and a heating ramp of 2 C min-1.
Incorporation of Organic Molecules into Metal Oxides. Propionic acid, glutaric acid, and PAA were used to modify the surface of the metal oxides. In a typical experiment, the organic moiety was dissolved in Milli-Q water to produce a 10 wt % solution. The pH was adjusted using 1 and 0.1 M NaOH or HCl. The metal oxide (0.1 g) was added to 1.0 g of 10 wt % PA, GA, or (26) Suzuki, K.; Siddiqui, S.; Chappell, C.; Siddiqui, J. A.; Ottenbrite, R. M. Polym. Adv. Technol. 2000, 11, 92. (27) Zhou, J. F.; Zhou, M. F.; Caruso, R. A. Langmuir 2006, 22, 3332. (28) Sizgek, G. D.; Sizgek, E.; Griffith, C. S.; Luca, V. Langmuir 2008, 24, 12323.
Langmuir 2010, 26(17), 14203–14209
Drisko et al.
Article
PAA solution and swirled at 150 rpm for a given period of time. The solution was then drained, and the materials were washed with excess water, which was adjusted to the corresponding pH of the PA, GA, or PAA sorption solution. After three washes of 1 h each, the oxides were drained and dried at 60 C for a minimum of 8 h. For the molecular weight (Figure 3) and pH experiments (Figure 6), the metal oxides were stirred for 24 h in the 10 wt % solutions before draining and rinsing three times with an aqueous solution of the corresponding pH. For the molecular weight experiments, PAA of 2000 (PAA2K), 100 000 (PAA100K), and 250 000 g mol-1 (PAA250K) molecular weights were chosen, and 10 wt % solutions at pH 2.3 were prepared to investigate the effect that pore size had on PAA incorporation. To study the kinetics and the length of time needed to reach surface saturation, metal oxides were exposed to solutions of PAA2K and PAA100K at pH 2 for 10 min to 2 days (Figure 4). All experiments were performed in duplicate or triplicate, and the results were averaged. The nomenclature used is ZrTi-x-y, where x refers to the template, agarose gel (AG) and stearic acid (SA) used to structure the metal oxide and y refers to the molecule, monomer (PA), dimer (GA), and PAA of varying molecular weight, PAA2K to PAA250K. ZrTi-y indicates that no template was used. Modeling of Polymer Adsorption. To obtain information about the polymer adsorption mechanism, the data (10-480 min) was fit with the Fickian equation29-31 Mt ¼ Ktn M¥
ð1Þ
where Mt is the amount of polymer incorporated (mg m-2) at time t, M¥ is the adsorption capacity at equilibrium (mg m-2), and therefore Mt/M¥ is the normalization of polymer adsorbed at time t. K is proportional to the diffusion constant (min-n),30 and n indicates the adsorption mechanism. When n
