Incorporation of Inorganic Nanoparticles into Mesoporous Carbons

Jul 11, 2008 - Jung-Min Oh , Amar S. Kumbhar , Olt Geiculescu , and Stephen E. Creager. Langmuir 2012 28 (6), 3259-3270. Abstract | Full Text HTML | P...
1 downloads 0 Views 94KB Size
J. Phys. Chem. C 2008, 112, 11657–11660

11657

Incorporation of Inorganic Nanoparticles into Mesoporous Carbons Synthesized by Soft Templating Joanna Go´rka and Mietek Jaroniec* Department of Chemistry, Kent State UniVersity, Kent, Ohio, 44242 ReceiVed: April 18, 2008; ReVised Manuscript ReceiVed: May 21, 2008

Mesoporous carbon monoliths with embedded alumina and silica nanoparticles were synthesized using phloroglucinol and formaldehyde as carbon precursors and poly(ethylene oxide)-poly(propylene oxide)poly(ethylene oxide) block copolymer as a soft template. While the carbon mesostructure was not significantly affected by embedding silica nanoparticles (∼20 nm), this was not the case for the incorporation of alumina nanoparticles (∼50 nm). In the latter case a gradual reduction in the carbon mesoporosity was observed with increasing loading of alumina nanoparticles. Since various inorganic nanoparticles of different sizes and shapes are available commercially, the soft-templating synthesis route is a very promising way for the design of carbon-based materials with embedded nanoparticles of desired properties. Introduction The discovery of ordered mesoporous carbons (OMCs) in 19991 has opened a new chapter in the research of ordered mesoporous materials. The first OMCs were obtained by filling the pores of silica nanostructures (used as hard templates) with suitable carbon precursors followed by carbonization and dissolution of the template. The hard templating synthesis became a very popular way for the fabrication of OMCs, which in this case are inverse replicas of the silica templates used.2–7 Some disadvantages of this synthesis strategy relate to the need of preparing siliceous hard templates and dissolving them using hazardous hydrofluoric acid or sodium hydroxide solutions. The aforementioned environmental issue, multistep processing and high-cost make the hard templating synthesis unfeasible for a large scale production of OMCs. Recently, a simple and feasible way of preparing mesoporous carbons by self-assembly of appropriate polymerizing organics (carbon precursors) and triblock copolymers (soft template) has been reported.8 A controlled thermal treatment of the resulting polymeric nanocomposites is used to remove the triblock copolymer (soft template) leaving behind large and uniform mesopores and to carbonize the remaining polymer (carbon precursor). The development of soft-templated mesoporous carbons represents a major step in materials science because fewer steps are required for their preparation as well as the block copolymers used are commercially available and biodegradable, which makes this synthesis feasible from industrial viewpoint. The first successful work on the soft-templating synthesis of ordered mesoporous carbon films has been reported in 2004 by Liang et al.9 by using polystyrene-block-poly(4-vinylpyridine as a structure directing agent. Later, two independent groups10,11 proposed the use of poly(ethylene oxide)-poly(propylene oxide)poly(ethylene oxide) triblock copolymers (PEO-PPO-PEO) as structure forming agents (which are commonly employed in the synthesis of ordered mesoporous silicas such as SBA-15 and SBA-16) for the synthesis of OMCs. Besides that these groups developed two alternative synthesis routes leading to the ordered carbons not only in the form of thin films but also in the form * To whom correspondence should be addressed. Phone: 330-672 3790. E-mail: [email protected].

of fibers and monoliths. Zhao‘s strategy11 is based on prepolymerization of a carbon precursor called resol, which is a lowmolecular-weight copolymer of phenol and formaldehyde prepared in the presence of sodium hydroxide as a catalyst. Next, as-made resol after neutralization interacts with surfactant to assemble into a mesostructure. In contrast, the procedure proposed by Liang and Dai10 involves polymerization of phloroglucinol and formaldehyde as a carbon-yielding component in the presence of a block copolymer template under acidic conditions. An advantage of the latter synthesis route is the elimination of a prepolymerization step and pH adjustment required in the synthesis under basic conditions. Several modifications of the aforementioned synthesis routes have been proposed by using a mixture of resorcinol and phloroglucinol,12 special additives like triethyl orthoacetate13 or a mixed template consisting of different PEO-PPO-PEO copolymers such as F108 and P123.10,14 An interesting recipe combining block copolymer-resol selfassembly with hard-templating approach using silica colloidal crystals was employed to obtain hierarchically porous carbons.15 The silica nanoparticles were packed into colloidal crystal array to serve as a support for the carbon mesostructure formed in the voids of the silica monolith. From viewpoint of catalysis there is a great interest in the control of the pore size, pore volume and the specific surface area as well as in the introduction of different functionalities, heteroatoms and nanoparticles into mesoporous carbons. Among those, the latter issue, incorporation of nanoparticles into mesoporous carbons, has been accomplished by hard templating route only; 16–19 however, no reports are available on the softtemplating synthesis of mesoporous carbons in the presence of inorganic particles. Recently titania20 and iridium-containing21 mesoporous carbons prepared via soft templating have been reported. In both cases the appropriate inorganic salts were employed as precursors, which upon thermal treatment were transformed into inorganic nanoparticles embedded in the carbon matrix. It was shown previously that a direct addition of metal salts into synthesis mixture represents a convenient way for the formation of well-dispersed nanoparticles in the carbon matrix, but at the same time, it is difficult to control the size of these particles as

10.1021/jp803367p CCC: $40.75  2008 American Chemical Society Published on Web 07/11/2008

11658 J. Phys. Chem. C, Vol. 112, No. 31, 2008

Go´rka and Jaroniec

Figure 1. TG profiles in air for the alumina-containing carbons. Figure 3. Nitrogen adsorption isotherms and the corresponding pore size distributions (inset) for mesoporous carbons with embedded alumina nanoparticles.

Figure 2. TG profiles in air for the silica-containing carbons.

well as the formation of carbon mesostructure, which often becomes microporous.22,23 Thus, application of the latter type of carbons may be limited in catalysis due to the poor mass transfer in micropores. In spite of the aforementioned shortcomings the recently presented idea20,21 of using block copolymers as soft templates for the carbon structure formation in the presence of metal salts that undergo transformations into nanoparticles is a very promising approach for the design of nanostructured carbon-based catalysts. Therefore, it would be interesting to explore the possibility of a direct incorporation of commercially available nanoparticles in the form of colloidal solutions into mesoporous carbons. This approach, as it was shown in the case of hard templating,19 represents the simplest way for controlling the size of incorporated nanoparticles. Here we report the soft-templating synthesis of mesoporous carbons with embedded alumina and silica nanoparticles, which were directly incorporated into the resulting mesostructures. Furthermore these nanoparticles were found to be accessible and not protected by a carbon shell. An additional advantage of these carbons are large and uniform mesopores (∼10 nm) providing better mass transfer of reagents than carbons previously reported (usually around 6 nm). Also, a broad variety of commercially available nanoparticles of different sizes and shapes makes this synthesis route even more attractive for the design of carbon-based materials, especially catalysts, with desired properties. Experimental Section Chemicals. Poly(ethylene oxide)-poly(propylene oxide)poly(ethylene oxide) triblock copolymer (EO106PO70EO106; Pluronic F127) was provided by BASF Corp. Phloroglucinol (C6H3(OH)3; 98%) and formaldehyde (HCHO), were purchased from Acros Organics. HCl (35-38%) was acquired from Fischer and ethanol (95%) from Pharmco. Al2O3 nanoparticles suspension (50 nm) was provided by Nyacol Nanotechnologies Inc., whereas silica nanoparticles (20 nm) by Precision Colloids, LLC.

Materials. Mesoporous carbon samples were prepared according to a slightly modified recipe of Liang et al.10 In a typical synthesis, 1.25 g of phloroglucinol and 1.25 g of poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) triblock copolymer (Pluronic F127) were dissolved in 9.7 g of ethanol-water (10:9 wt ratio) solution and stirred vigorously at room temperature. After complete copolymer dissolution, an aqueous suspension of alumina nanoparticles was introduced to the synthesis mixture. The amount of nanoparticles used (according to manufacturer specification) was calculated to be 10 and 20 wt % with respect to the carbon precursors. Then, 0.08 mL of 37% HCl was added to the solution as a catalyst. The resulting solution was stirred for additional 30 min until a light pink color appeared. Subsequently, 1.25 mL of 37% formaldehyde was added to the synthesis mixture. The solution turned cloudy after 30 min and after additional 1-2 h separated into two layers. The polymer-rich bottom layer obtained after separation was kept under magnetic stirring overnight. Then the elastic, nonsticky monolith was transferred to an autoclave and treated at 100 °C for 24 h. Carbonization of the resulting monolith was performed in the tube furnace under nitrogen flow using a heating rate of 2 °C/min up to 180 °C, keeping the sample at this temperature for 5 h, resuming heating with 2 °C/min up to 400 °C and with 5 °C/min up to 850 °C, and finally keeping the sample at 850 °C for 2 h. The final samples were labeled C-Al2O3-x, where x indicates the weight percentage of alumina nanoparticles in the sample. For the purpose of comparison two mesoporous carbon monoliths with embedded SiO2 nanoparticles were prepared by employing the same synthesis route as that used for the C-Al2O3-x samples. The resulting samples were labeled C-SiO2x, where x denotes the weight percentage of silica in the sample. Measurements. Nitrogen adsorption isotherms were measured at -196 °C on ASAP 2010 and 2020 volumetric analyzers (Micromeritics, Inc., GA). Prior adsorption measurements all samples were outgassed at 200 °C for at least 2 h. Wide angle X-ray diffraction measurements were performed on a PANalytical X’Pert PRO MPD X-ray diffraction system using Cu KR radiation (40 kV, 40mA). All patterns were recorded using 0.02° step size and 4 s per step in the range of 5° e 2θ e 70°. Thermogravimertic analysis was made using a TA Instrument Hi-Res TGA 2950 thermogravimetric analyzer from 30 to 800 °C under air flow with a heating rate of 10 °C/min. Calculations. The BET specific surface area was calculated from nitrogen adsorption isotherms in the relative pressure range of 0.05-0.2.24 The total pore volume23 was estimated from the

Inorganic Nanoparticles in Mesoporous Carbons

J. Phys. Chem. C, Vol. 112, No. 31, 2008 11659

TABLE 1: Adsorption Parameters for the Carbon Samples Studieda sample

SBET m2/g

Vt cm3/g

Vmi cm3/g

Vme cm3/g

wKJS nm

C-Al2O3-10 C-Al2O3-20 C-SiO2-10 C-SiO2-20 C-SiO2-20-HF

438 360 441 416 500

0.65 0.36 0.71 0.69 0.86

0.04 0.05 0.03 0.02 0.03

0.56 0.26 0.65 0.63 0.79

10.6 9.2 10.6 10.9 10.6

R% 8 15 9 16 1.5

Dt °C 512 520 517 512 515

a Notation: SBET, BET specific surface area; Vt, single-point pore volume; Vmi, volume of fine pores (mainly micropores) obtained by integration of PSD up to 3.5 nm; Vme, volume of mesopores obtained by integration of PSD from 3.5 to 16 nm (in the case of SiO2-20-HF integration was done in the range of 3.5-20 nm); wKJS, mesopore diameter at the maximum of the PSD curve obtained by the improved KJS method;25 R, residue obtained from the TG curve recorded in air at 800 °C; Dt, decomposition temperature.

Figure 4. Nitrogen adsorption isotherms and the corresponding pore size distributions (inset) for carbon samples with embedded silica nanoparticles. The plot with open circles refers to the C-SiO2-20 sample treated with HF to remove the silica nanoparticles.

Figure 5. Wide angle XRD pattern for the alumina-containing carbons. The narrow and intense Al peaks originate from the sample holder used.

amount adsorbed at a relative pressure of ∼0.99. The pore size distributions were calculated from nitrogen adsorption isotherms at -196 °C using the improved KJS method25 for cylindrical mesopores with diameters up to 10nm. Results and Discussion Shown in Figure 1 are the TG profiles for the alumina-carbon nanocomposites revealing the total alumina content of each material after its thermal treatment in air. The TG residues were ∼8 and 15 wt % for the C-Al2O3-10 and C-Al2O3-20 samples, respectively. For comparison, the TG curves for the carbons containing silica nanoparticles were also recorded (Figure 2). These results reveal similar weight percentages of silica nanoparticles to those obtained for the Al2O3-carbon materials, i.e., ∼9 and 16 wt % for the C-SiO2-10 and C-SiO2-20 samples, respectively.

Figure 6. Wide angle XRD pattern for the silica-containing carbons. The narrow and intense Al peaks originate from the sample holder used.

All carbon samples exhibited similar thermal stability in air. Their oxidation occurred in the temperature range of 500-520 °C for the samples containing either Al2O3 or SiO2 nanoparticles. These results further indicate that the presence of alumina and silica nanoparticles did not affect the thermal stability of the samples studied. Nitrogen adsorption isotherms measured at -196 °C and the corresponding pore size distributions (PSDs) for the C-Al2O310 and C-Al2O3-20 samples are shown in Figure 3, while basic parameters evaluated by analysis of these isotherms are listed in Table 1. For all samples studied type IV nitrogen adsorption isotherms were recorded with steep condensation steps reflecting uniform mesopores. The condensation step for the C-Al2O3-10 sample is higher than that of C-Al2O3-20 reflecting its larger mesopore volume. While both samples exhibit similar microporosity (0.04-0.05 cm3/g), the mesopore volume of the C-Al2O3-10 sample is twice larger (see Table 1). Also, a decrease in the BET surface area is observed with increasing loading of nanoparticles, i.e., from 438 m2/g for C-Al2O3-10 to 360m2/g for C-Al2O3-20. Since the pore size did not change much with increasing loading of alumina nanoparticles (10.6 nm vs 9.2 nm; see PSDs in Figure 3 inset), the observed significant reduction in the volume of mesopores cannot be explained only by a decrease of the carbon amount in the sample due to the presence of nanoparticles. In addition, it seems that high loadings of alumina nanoparticles reduce the amount of mesopores, which are formed during self-assembly process. Another set of carbon samples was synthesized using silica nanoparticles instead of alumina. Nitrogen adsorption isotherms are presented in Figure 4. Regardless of the amount of nanoparticles introduced, both isotherms exhibited steep capillary condensation steps reflecting high uniformity of mesopores. In contrast to the alumina-containing carbons, an increase in the loading of silica

11660 J. Phys. Chem. C, Vol. 112, No. 31, 2008 nanoparticles from 9% to 16% (values obtained by TGA; Figure 2) did not cause significant changes in the nitrogen adsorption isotherms and the resulting adsorption parameters. The BET surface area was found to be 441 m2/g for C-SiO2-10 and slightly smallers416 m2/g for C-SiO2-20. Both silica-carbon samples had higher total pore volumes in comparison to those for the aluminacarbon samples. Pore size distributions for C-SiO2-10 and C-SiO220, shown in inset of Figure 4, are nearly identical. A higher loading of silica nanoparticles caused only a slight decrease in the mesopore volume, which differs from the results obtained for the alumina-containing carbons. Note that the silica surface is more hydrophilic than alumina and probably it does not disturb the mesostructure formation in hydrophilic domains of the block copolymer template. Therefore, the whole system can easier adapt to a higher loading of silica nanoparticles, causing only small structural changes. An experimental confirmation for the attraction of silica nanoparticles into hydrophilic domains of polymerizing carbon precursors is the easiness of block copolymer-templated self-assembly of the latter precursors in the presence of tetraethyl orthosilicate (TEOS), which leads to the silica-carbon mesoporous composites.26,27 Also, 2.5 times smaller size of silica nanoparticles in comparison to alumina particles facilitates silica incorporation into the carbon matrix without significant pore structure alteration. An additional plot shown in Figure 4 labeled as C-SiO2-20HF refers to the carbon sample obtained from C-SiO2-20 after removal of silica nanoparticles with 10% HF solution. Nitrogen adsorption isotherm for this sample exhibits a steep condensation step at the same relative pressure as that for the untreated C-SiO2-20 sample. The main difference between these two isotherms appears at the pressures close to the saturation pressure; namely, the isotherm curve for C-SiO2-20-HF shows another step, which reflects the spherical pores created after removal of silica nanoparticles. The HF etching increased the total pore volume from 0.69 to 0.86 m3/g, which is visible in Figure 4 inset showing a comparison of the PSD curves. In addition to a small peak reflecting fine pores below 3 nm and the main peak located around 10 nm that represents uniform mesopores, there is an extra peak on the PSD curve located between 15 and 20 nm (see inset in Figure 4). The latter peak represents new mesopores created after removal (HF etching) of silica nanoparticles. Also, the BET surface area increased from 416 m2/g for untreated sample to 500 m2/g for the HF treated silica-carbon composite. A successful etching of silica nanoparticles with HF, which is reflected the aforementioned extra peak on the PSD curve (Figure 4), proves that these nanoparticles are accessible for HF molecules. The powder XRD measurements for the thermally treated nanocomposites at 850 °C were used to identify the presence of crystalline domains of the embedded nanoparticles. The XRD patterns of the alumina-containing carbons are presented in Figure 5. The diffraction patterns reveal the presence of some aluminum oxide phase and graphitic carbon in both C-Al2O3-10 and C-Al2O320 samples. The diffraction peaks referring to the aluminum oxide phase are broad, which is characteristic for small crystallites. The latter feature of the XRD pattern may indicate that alumina nanoparticles have been well dispersed in the carbon matrix; otherwise, their agglomeration could lead to larger aluminum oxide crystals. Furthermore, a few reflections can be assigned to graphitic carbon. Similarly, the wide angle XRD patterns recorded for the carbons with embedded silica nanoparticles (Figure 6) reveal some reflections that may correspond to graphitic carbon; however, there is no evidence for crystalline silica phases.

Go´rka and Jaroniec Conclusions Mesoporous carbons with incorporated alumina and silica nanoparticles were successfully synthesized proving that the soft-templating synthesis represents a simple and effective way for the introduction of various inorganic nanoparticles into the carbon matrix. Regardless the type of nanoparticles, smaller loadings gave mesoporous carbons with comparable adsorption properties. In the case of carbons with larger loadings of nanoparticles, the aforementioned properties depend on the chemistry and size of these particles. It is shown that the embedded particles in mesoporous carbons are accessible for other molecules, which was proved by a successful removal of silica with HF solution. Since the soft-templating synthesis of mesoporous carbons is a one-pot process involving commercially available block copolymers, phenol derivatives and formaldehyde, it is especially well suited for addition of various inorganic nanoparticles. Acknowledgment. The authors thank BASF Co. for providing the triblock polymer, Nyacol Nanotechnologies Inc. for colloidal silica, and Precision Colloids, LLC for colloidal alumina solutions. References and Notes (1) Ryoo, R.; Joo, S. H.; Jun, S. J. Phys. Chem. B 1999, 103, 7743. (2) Ryoo, R.; Joo, S. H.; Kruk, M.; Jaroniec, M. AdV. Mater. 2001, 13, 677. (3) Lee, J.; Han, S.; Hyeon, T. J. Mater. Chem. 2004, 14, 478. (4) Yang, H. F.; Zhao, D. Y. J. Mater. Chem. 2005, 15, 1217. (5) Lu, A. H.; Schuth, F. AdV. Mater. 2006, 18, 1793. (6) Lee, J.; Kim, J.; Hyeon, T. AdV. Mater. 2006, 18, 2073. (7) Vinu, A.; Mori, T.; Ariga, K. Sci. Technol. AdV. Mater. 2006, 7, 753. (8) Wan, Y.; Shi, Y.; Zhao, D. Chem. Mater. 2008, 20, 932. (9) Liang, C.; Hong, K. L.; Guiochon, G. A.; Mays, J. W.; Dai, S. Angew. Chem., Int. Ed. 2004, 43, 5785. (10) Liang, C.; Dai, S. J. Am. Chem. Soc. 2006, 128, 5316. (11) Zhang, F. Q.; Meng, Y.; Gu, D.; Yan, Y.; Yu, C. Z.; Tu, B.; Zhao, D. Y. J. Am. Chem. Soc. 2005, 127, 13508. (12) Tanaka, S.; Katayama, Y.; Tad, M. P.; Hillhouse, H. W.; Miyake, Y. J. Mater. Chem. 2007, 17, 3639. (13) Tanaka, S.; Nishiyama, N.; Egashira, Y.; Ueyama, K. Chem. Commun. 2005, 2125. (14) Huang, Y.; Cai, H.; Yu, T.; Zhang, F.; Zhang, F.; Meng, Y.; Gu, D.; Wan, Y.; Sun, X.; Tu, B.; Zhao, D. Angew. Chem., Int. Ed. 2007, 46, 1089. (15) Deng, Y.; Liu, C.; Yu, T.; Liu, F.; Zhang, F.; Wan, Y.; Zhang, L.; Wang, C.; Tu, B.; Webley, P. A.; Wang, C.; Zhao, D. Chem. Mater. 2007, 19, 3271. (16) Chai, G. S.; Yoon, S. B.; Yu, J.-S.; Choi, J.-H.; Sung, Y.-E. J. Phys. Chem. B 2004, 108, 7074. (17) Liu, S.-H.; Lu, R.-F.; Huang, S.-J.; Lo, A.-Y.; Chien, S.-H.; Liu, S.-B. Chem. Commun. 2006, 32, 3435. (18) Park, I.-S.; Choi, M.; Kim, T.-W.; Ryoo, R. J. Mater. Chem. 2006, 16, 3409. (19) Jaroniec, M.; Choma, J.; Gorka, J.; Zawislak, A. Chem. Mater. 2008, 20, 1069. (20) Liu, R.; Ren, Y.; Shi, Y.; Zhang, F.; Zhang, L.; Tu, B.; Zhao, D. Chem. Mater. 2008, 20, 1140. (21) Gao, P.; Wang, A.; Zhang, T. Chem. Mater. 2008, 20, 1881. (22) Moreno-Castilla, C.; Maldonado-Ho´dar, F. J.; Rivera-Utrilla, J.; Rodriguez-Castellon, E. Appl. Catal., A 1999, 183, 345. (23) Rojas-Cervantes, M. L.; Alonso, L.; Diaz-Teran, J.; Lopez-Peinado, A. J.; Martin-Aranda, R. M.; Gomez-Serrano, V. Carbon 2004, 42, 1575. (24) Kruk, M.; Jaroniec, M. Chem. Mater. 2001, 13, 3169. (25) Jaroniec, M.; Solovyov, L. A. Langmuir 2006226757. Choma, J.; Gorka, J.; Jaroniec, M. Microporous Mesoporus Mater. 2008, 112, 573. (26) Liu, R.; Shi, Y.; Wan, Y.; Meng, Y.; Zhang, F.; Gu, D.; Chen, Z.; Tu, B; Zhao, D. J. Am. Chem. Soc. 2006, 128, 11652. (27) Lin, H. P.; Chang-Chien, C. Y.; Tang, C. Y.; Lin, C. Y. Microporous Mesoporous Meter. 2006, 93, 344.

JP803367P