Micropore to Macropore Structure-Designed Silicas with Regulated

Jul 23, 2005 - Regulated Condensation of Silicic Acid Nanoparticles ... silicas were produced when carboxylic acids were applied; the formation of ...
0 downloads 0 Views 243KB Size
Download PDF
8042

Langmuir 2005, 21, 8042-8047

Micropore to Macropore Structure-Designed Silicas with Regulated Condensation of Silicic Acid Nanoparticles Hiroshi Isobe,*,†,‡ Shigenori Utsumi,‡ Kohzoh Yamamoto,† Hirofumi Kanoh,‡ and Katsumi Kaneko*,‡ Fuji Chemical Company, Ltd., 1683-1880 Nasubigawa, Nakatsugawa 509-9132, and Graduate School of Science and Technology, Chiba University, 1-33 Yayoi, Inage, Chiba 263-8522, Japan Received April 7, 2005. In Final Form: June 16, 2005 A new preparation method for porous silica particles was developed using activated silica sols which are called nano-silica solutions in this paper. Several kinds of organic and inorganic acids are employed to neutralize diluted sodium silicate solutions to form the nano-silica solutions: formic acid, acetic acid, propionic acid, oxalic acid, succinic acid, DL-malic acid, citric acid, and tricarballylic acid as carboxylic acids, and sulfuric acid and hydrochloric acid as inorganic acids. The effect of salts in the nano-silica solution is also studied. The products were investigated using a field emission scanning electron microscope, an X-ray diffractometer, the nitrogen adsorption technique, and a mercury porosimeter. Microporous silicas were produced when carboxylic acids were applied; the formation of micropores was influenced by the pH of the nano-silica solutions and molecular sizes of the carboxylic acids. Addition of a salt in a citric acid solution increased the mesopore volume. Macropores were formed when inorganic acids including salts were applied; the salt nanoparticles which were crystallized in silica spheres acted as templates. The anion types and salt concentrations in the nano-silica solutions affected the aggregation condition of silica nanoparticles, following the Schulze-Hardy rule.

1. Introduction 1

Silica has been widely applied in various industries. New porous silicas, such as MCM,2,3 FSM,4 and SBA5,6 of regular pore structures, have stimulated a wide range of fundamental studies on interface chemistry. The key to the preparation of new silicas with regular pore structures is to use the molecular assembly structure of surfactant molecules as templates for the pores. Although these new silicas have contributed to recent progress in interface science, they are not sufficiently applied to technology due to the limited pore region and for economic reasons. Therefore, if we can produce silicas with pores that can be controlled from micropores to macropores with a more concise technique using sodium silicate solutions, new and wide applications can be extended to porous silicas. We focus on the very early stage of polymerized silicic acids from sodium silicate solutions, which can be called nano-silica solutions because they include silicic acid nanoparticles as the intermediate between ions and particles. If aggregation of the silicic acid nanoparticles can be avoided during drying, porous silicas having specific * To whom correspondence should be addressed. (H.I.) Phone: +81-573-68-7222. Fax: +81-573-68-7228. E-mail: isobe@ fuji-chemical.jp. (K.K.) Phone: +81-43-290-2779. Fax: +81-43290-2788. E-mail: [email protected]. † Fuji Chemical Co., Ltd. ‡ Chiba University. (1) Iler, R. K. The Chemistry of Silica; Wiley: New York, 1979. (2) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710. (3) Beck, J. S.; Vartiuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.; Schmitt, K. D.; Chu, C. T.-W.; Olson, D. H.; Sheppard, E. W.; McCullen, S. B.; Higgins, J. B.; Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834. (4) Yanagisawa, T.; Shimizu, T.; Kuroda, K.; Kato, C. Bull. Chem. Soc. Jpn. 1990, 63, 988. (5) Monnier, A.; Schuth, F.; Huo, Q.; Kumar, D.; Margolese, D. I.; Maxwell, R. S.; Stucky, G. D.; Krisnamurty, M.; Petoff, P.; Firouzi, A.; Janicke, M.; Chemlka, B. F. Science 1993, 261, 1299. (6) Zhao, D.; Sun, J.; Li, Q.; Sturcky, G. D. Chem. Mater. 2000, 12, 275.

surface areas of more than 1000 m2/g can be obtained. One type of nano-silica solution is prepared from a reaction between a diluted sodium silicate solution and a mineral acid, such as sulfuric acid. This type of nano-silica solution had been used for water treatment and is called “activated silica sols”.7 Commercially available silica gels and precipitated silicas are usually produced from the reaction of the same kinds of solutions. Another type of nano-silica solution is prepared by means of sodium ion exchange to a proton using cation-exchange resins and is called “active silica” 1 or “active silicic acid”.8 Active silica is an important source for silica sol in industry. Active silica also consists of extremely small silica nanoparticles 1-2 nm in diameter.1,9 Hence, we can construct porous silicas using nanosilica solutions. However, we must suppress the rapid aggregation of silica nanoparticles in the nano-silica solution on drying. One applicable method is the supercritical drying method,10-12 which provides silica aerogels. Also other trials to control the aggregation were carried out.13-15 We have been trying to prepare silica particles using spray-drying methods; hollow silica spheres derived from sodium silicate solutions16,17 and porous silica particles derived from silicon tetrachloride18 have been obtained. (7) Henry, C. R. J.sAm. Water Works Assoc. 1958, 51, 61. (8) Tsai, M.-S. Mater. Sci. Eng., B 2004, 106, 52. (9) Birch, D. J. S.; Geddes, D. D. Chem. Phys. Lett. 2000, 320, 229. (10) Leventis, N.; Elder, I. A.; Long, G. J.; Rolison, D. R. Nano Lett. 2002, 2, 63. (11) Morris, C. A.; Rolison, D. R.; Swider-Lyons, K. E.; OsburnAtkinson, E. J.; Merzbacher, C. I. J. Non-Cryst. Solids 2001, 285, 29. (12) Hua, D. W.; Anderson, J.; Hæreid, S.; Smith, D. M.; Beaucage, G. Mater. Res. Soc. Symp. Proc. 1994, 346, 985. (13) Schwertfeger, F.; Frank, D.; Schmidt, M. J. Non-Cryst. Solids 1998, 225, 24. (14) Gerber, T. J. Sol-Gel Sci. Technol. 1998, 13, 323. (15) Lee, C. J.; Kim, G. S.; Hyun, S. H. J. Mater. Sci. 2002, 37, 2237. (16) Kaneko, K.; Isobe, H.; Katori, T.; Tokunaga, I.; Gouda, T.; Suzuki, T.; Ozeki, S.; Okuda, K. Colloids Surf., A 1993, 74, 47. (17) Isobe, H.; Tokunaga, I.; Noriyoshi, N.; Kaneko, K. J. Mater. Res. 1996, 11, 2908. (18) Isobe, H.; Kaneko, K. J. Colloid Interface Sci. 1999, 212, 234.

10.1021/la0509192 CCC: $30.25 © 2005 American Chemical Society Published on Web 07/23/2005

Micropore to Macropore Structure-Designed Silicas

Since the spray-drying method has been industrially established for a long time, a basic study using this method can extend to a large-scale plant easily. Also, spherical powders can be readily prepared using the spray-drying method. The preparation of silica spheres with uniform mesopores was reported in the literature,19 in which commercially available silica sols including uniform latex nanoparticles were used as the starting materials and the ultrasonic spray-drying method was employed for producing uniform particles. We selected activated silica sols with much smaller primary silica particles than those in commercially available silica sols. The problem we have to solve is to find a method to avoid the aggregation of silica nanoparticles. We have searched efficient aggregation inhibitors which intensify the suppression effect of the spray-drying method for activated silica sols. The salts produced by the neutralizing reaction between the base in the sodium silicate solution and the acid must act as an aggregation inhibitor. In this study, several kinds of organic and inorganic acids were selected as the reactant for the aggregation inhibitor. Spray-dried silicas were prepared from activated silica sols which were derived from reactions of sodium silicate solutions and several kinds of organic and inorganic acids, and then the pore structures of silicas produced under the various conditions were studied. 2. Experimental Section 2.1. Preparation of Nano-Silica Solutions. A sodium silicate solution (industrial grade, 3.2SiO2:1Na2O, 29 wt % silica content, Fuji Chemical Co.) was diluted to 10 wt % silica content using deionized water and kept at 293 ( 1 K. The concentration of NaOH as the base in the diluted sodium silicate solution is about 1 N. Since reactions should be operated on the acidic side to avoid gelation, acid solutions of 1-7 N were used. Sulfuric acid (2 N) and hydrochloric acid (2 N) as inorganic acids, formic acid (2-7 N), acetic acid (3 N), and propionic acid (3 N) as monocarboxylic acids, oxalic acid (1 N), succinic acid (2 N), and DL-malic acid (2 N) as dicarboxylic acids, and citric acid (2 N) (mentioned above, Wako Pure Chemical Co.) and tricarballylic acid (2 N) (Aldrich) as tricarboxylic acids were prepared using deionized water. Those acid solutions were kept at 293 ( 1 K. A diluted sodium silicate solution of 150 g was added to an acid solution of 150 g at a rate of 12-13 mL/min with moderate stirring. The obtained nano-silica solutions were stirred at 293 ( 1 K for 10 min. Succinic acid does not have enough solubility in water to form the transparent solution; adding the diluted sodium silicate solution induced a perfect dissolution of the produced succinic salt. To study the effects of inorganic salts, salt-rich types of nano-silica solutions were also prepared by acid solutions of dissolved salts of 10-30 wt % using sodium chloride in citric acid and hydrochloric acid solutions and sodium sulfate in sulfuric acid solutions. We prepared an active silica solution whose silica content is 5 wt % using cation-exchange resins to produce a reference silica without porosity. 2.2. Porous Silicas from Spray Drying. A laboratory-scale spray dryer (SD-1, Tokyo Rika Kikai Co.) was employed to prepare spray-dried silica samples at the conditions described below. The air was heated to 413 ( 10 K and introduced into the apparatus at a flow rate of 0.4-0.5 m3/min, and the nano-silica solutions were sprayed into a heating area of about 7 L using a nozzle at a rate of about 10 mL/min. The outlet gas temperatures before collection were 363 ( 10 K; the drying times were estimated to be about 1 s. The dried particles were collected using a cyclone. The obtained spray-dried samples were dispersed in water and stirred for more than 10 min followed by filtration and washing with enough water until a neutral pH was obtained for the filtrate. The washed samples were dried at 383 K for 3 h. Spray-dried samples except the reference silica were designated by using (19) Iskandar, F.; Mikrajuddin; Okuyama, K. Nano Lett. 2002, 2, 389.

Langmuir, Vol. 21, No. 17, 2005 8043 abbreviations indicating the acids, sulfuric acid (HS), hydrochloric acid (HC), formic acid (FA), acetic acid (AA), propionic acid (PA), oxalic acid (OA), succinic acid (SA), DL-malic acid (MA), citric acid (CA), and tricarballylic acid (TA), followed by a number indicating the concentration of the acid in normality, such as HS-2 and FA-2. Salt-added samples were designated by adding a number in parentheses indicating the salt concentration in the acid solution in weight percent; HS-2(10) indicates a spray-dried sample using sulfuric acid including 10 wt % sodium sulfate. The reference silica from active silica is noted as NP-Silica. Washed samples were designated by attaching the letter W after the spray-dried sample names. 2.3. Characterization. The morphology of the samples was observed with a field emission scanning electron microscope (Hitachi S-4100) with an acceleration voltage of 15 kV after addition of a Pt-Pd coating about 20 nm thick and a transmission electron microscope (JEOL 4000EX) with an acceleration voltage of 400 kV. The main elements in the samples were measured with an energy-dispersive spectroscope (EDAX DX-4) attached to the field emission scanning electron micrscope with an acceleration voltage of 20 kV without coating. X-ray diffraction patterns were obtained with an automatic X-ray diffractometer (MAC Science MXP18) using Cu KR radiation at 40 kV and 30 mA. The particle size distribution of NP-Silica was measured with a laser scattering particle size analyzer (HORIBA LA-910). The porosity from micropores to mesopores was examined with the nitrogen adsorption technique at 77 K using Autosorb-1 (Quantachrome) after preheating at 453 K and 1 mPa for 90 min. The porosity from mesopores to macropores was measured with a mercury porosimeter (CE Instruments Pascal 240) after drying at 453 K.

3. Results and Discussion 3.1. Silicas Prepared Using Organic Acids. The preparation conditions of nano-silica solutions are listed in Table 1, and the properties of the samples from carboxylic acids and the reference silica are listed in Table 2. According to X-ray diffraction (XRD), spray-dried samples prepared using monocarboxylic and dicarboxylic acids include crystalline materials, mainly sodium carboxylates. Unknown fine crystals are formed in MA-2. Spray-dried samples from tricarboxylic acids, TA-2 and CA-2, indicate amorphous phases. Typical scanning electron microscopy (SEM) images of washed samples prepared using 3 N monocarboxylic acids are shown in Figure 1 as representatives of all samples prepared using carboxylic acids. Samples using formic acid and acetic acid (FA-3W and AA-3W) are spherical silicas having smooth surfaces with partial dimples. SA-2W, MA-2W, TA-2W, and CA-2W have smooth spherical surfaces similar to those of FA-3W and AA-3W without dimples. Silica spheres from propionic acid (PA-3W) have rough surfaces originating from the gaps of primary silica particles. OA-1W has a surface structure similar to that of PA-3W. The analytical results by EDAX showed that most of the sodium carboxylates in the spray-dried samples were removed by washing; the Na content in the washed samples was in the range of 0.2-0.5 wt %. Nitrogen adsorption isotherms of samples from 3 N monocarboxylic acids are shown in Figure 2a. Adsorption isotherms of FA-3W and AA-3W are basically classified as type I, having a predominant uptake below P/P0 ) 0.01. These isotherms were analyzed by the Dubinin-Radushkevich (DR) equation; micropore volumes (W0) of FA-3W and AA-3W obtained by the DR plot were 0.11 and 0.30 mL/g, respectively. The size of the acid molecules affects the microporosity; the larger the molecular size, the greater the micropore volume. Strictly speaking, the isotherm of PA-3W is not classified as type I because of the development of mesopores. The micropore volume is 0.27 mL/g, which is smaller than that of AA-3W, and the

8044

Langmuir, Vol. 21, No. 17, 2005

Isobe et al.

Table 1. Preparation Conditions of Nano-Silica Solutions applied acid solution additive

sample name (spray-dried)

component

concn (N)

NP-Silica FA-2 FA-3 FA-7 AA-3 PA-3 OA-1 SA-2 MA-2 TA-2 CA-2 CA-2(10) HS-2 HS-2(10) HS-2(20) HS-2(30) HC-2 HC-2(10) HC-2(20)

none formic acid formic acid formic acid acetic acid propionic acid oxalic acid succinic acid DL-malic acid tricarballylic acid citric acid citric acid sulfuric acid sulfuric acid sulfuric acid sulfuric acid hydrochloric acid hydrochloric acid hydrochloric acid

0 2 3 7 3 3 1 2 2 2 2 2 2 2 2 2 2 2 2

formula

NaCl Na2SO4 Na2SO4 Na2SO4 NaCl NaCl NaCl

pH of nano-silica solutions

concn (wt %) 0 0 0 0 0 0 0 0 0 0 0 10 0 10 20 30 0 10 20

3.0 3.4 2.9 2.4 4.4 4.5 4.6 5.0 3.9 4.5 4.2 3.9 0.7

0.2

Table 2. Properties of Samples Prepared from Carboxylic Acids and the Reference Silica XRD results of spray-dried samples

pore structural parameters

sample name

main component [crystallite sizea (nm)]

sample name

aBETb (m2/g)

NP-Silica FA-2 FA-3 FA-7 AA-3 PA-3 OA-1 SA-2 MA-2 TA-2 CA-2 CA-2(10)

amorphous HCOONa HCOONa HCOONa CH3COONa‚3H2O C2H5COONa (COONa)2 CH2COOH‚CH2COONa unknown crystal amorphous amorphous NaCl [55]

NP-Silica FA-2W FA-3W FA-7W AA-3W PA-3W OA-1W SA-2W MA-2W TA-2W CA-2W CA-(10)W

0.6 335 195 30 720 715 205 455 725 810 715 860

Vpc (mL/g)

W0d (mL/g)

Vp - W0e (mL/g)

0.0008 0.20 0.11 0.02 0.43 0.50 0.25 0.26 0.42 0.47 0.41 0.71

0.0004 0.19 0.11 0.02 0.30 0.27 0.08 0.24 0.33 0.31 0.37 0.29

0.0004 0.01 0.00 0.00 0.13 0.23 0.17 0.02 0.09 0.16 0.04 0.42

a Crystallite sizes determined by the Scherrer method. b Specific surface areas obtained by the BET plot. c Total pore volumes calculated by the adsorption amount at P/P0 ) 0.98. d Micropore volumes obtained by the DR plot. e Approximate mesopore volumes calculated by Vp - W0.

Figure 1. Scanning electron micrographs of (a) FA-3W, (b) AA-3W, (c) PA-3W, and (d) CA-2(10)W.

approximate mesopore volume (VN - W0) obtained by subtracting the micropore volume (W0) from the total pore volume (VN) is 0.23 mL/g, being much larger than that of AA-3W, 0.13 mL/g. The hysteresis shape of PA-3W suggests the presence of cylindrical mesopores. The mesopores should stem from the gaps of primary silica particles as observed by SEM.

Figure 2b shows adsorption isotherms of silicas from dicarboxylic acids. The adsorption isotherm of OA-1W has characteristics of both types II and IV. OA-1W has an explicit adsorption hysteresis between P/P0 ) 0.4 and P/P0 ) 1.0 showing the presence of predominant mesopores. The adsorption isotherms of SA-2W and MA-2W are of type I. Both silicas have a considerably large micropore volume (SA-2W, 0.24 mL/g; MA-2W, 0.33 mL/g) according to DR analysis. The pore structural parameters are collected in Table 2. Adsorption isotherms of silicas from tricarboxylic acids are shown in Figure 2c. Both isotherms of TA-2W and CA-2W are of type I; micropore volumes of those samples are larger than 0.3 mL/g. 3.2. Control Factor of the Pore Structure of Silica with Organic Acids. The point of zero charge (PZC) of silica is in the range of 2.5-3.0 from the literature.20,21 The relationship between the PZC and pH of nano-silica solutions is an essential factor for aggregation of silica nanoparticles. When the solution pH is higher than the PZC, silica nanoparticles have negative charges of high concentration, interfering with compact aggregation of nanoparticles. Therefore, silica can have a predominant porosity with spray drying of a nano-silica solution with pH higher than the PZC. On the other hand, if the pH of (20) Parks, G. A. Chem. Rev. 1965, 65, 177. (21) De Bussetti, S. G.; Tschapek, M.; Helmy, A. K. J. Electroanal. Chem. 1972, 36, 507.

Micropore to Macropore Structure-Designed Silicas

Langmuir, Vol. 21, No. 17, 2005 8045

Figure 2. Nitrogen adsorption isotherms of samples at 77 K. The solid and open symbols indicate adsorption and desorption branches, respectively. Key: (a) FA-3W (b, O), AA-3W (2, 4), PA-3W (9, 0); (b) OA-1W (b, O), SA-2W (2, 4), MA-2W (9, 0); (c) TA-2W (b, O), CA-2W (2, 4), CA-2(10)W (9, 0).

the nano-silica solution is close to the PZC, spray drying tends to decrease the pore structure in silica. The pH of the nano-silica solution from 2-7 N formic acid is 2.43.4, being close to the PZC. Other organic acid solutions have pH higher than the PZC. Consequently, only the formic acid solution should not be fit for production of porous silica. The relationship between the pH of nano-silica solutions from formic acid and specific surface area obtained by the BET method shows that the specific surface area decreases with a decrease of the pH (see Figure A in the Supporting Information). The specific surface area can be used as a criterion for the aggregation degree of the primary silica nanoparticles. The above-mentioned adsorption measurements show that the porosity and the surface area of silica from formic acid (FA-7W) are the least of silicas prepared from other acids, as shown in Table 2. Thus, the pH of the nano-silica solution is a key factor in the construction of the pore structure. However, there is another factor that affects the pore structure. When sodium silicate solution is added to the carboxylic acid solution, fine crystallites of carboxylic salt are embedded in the silica particles on spray drying. The carboxylic salt crystallites can be removed by a washing procedure, leading to pores. Accordingly, the geometry of the salt crystallites determines the pore geometry. As spray drying suppresses the crystal growth of the salt of the carboxylic acid due to rapid drying, the crystallite formation rate of the salt should be important. Generally speaking, an organic acid of a larger molecular size can produce a more stable salt crystallite. Therefore, an organic acid of a larger molecular size can give a higher formation rate of the salt crystallite. A large quantity of the salt crystallites embedded in silica provides a larger pore volume by the washing procedure, which agrees with the experimental results in Table 2. Thus, the molecular size of the organic acid is another factor in determining the pore structures. Indeed, NP-Silica, which is prepared from the active silica solution of pH 3 without addition of organic acids using a procedure of ion exchange, is nonporous silica (see Figure Ba in the Supporting Information). The BET surface area is only 0.6 m2/g, indicating a silica particle size of 4.5 µm, which almost agrees with the average particle size from the particle size distribution measurement of NP-Silica (5.2 µm). As we understand that the crystallite geometry of the carboxylic salt determines the pore geometry, the effect

Figure 3. Transmission electron micrographs of (a) CA-2(10)W and (b) HC-2(20)W.

of the addition of inorganic salts on pore development was examined. CA-2(10)W was produced from a citric acid solution including sodium chloride. Figures 1d and 3a show the SEM and transmission electron microscopy (TEM) images of CA-2(10)W, respectively. The SEM and TEM images suggest the presence of mesopores. The nitrogen adsorption isotherm of CA-2(10)W above P/P0 ) 0.2 remarkably deviates upward from that of CA-2W, suggesting the predominant development of mesopores (Figure 2c). Then the coprecipitated sodium chloride crystallites on spray drying should give rise to mesopores. The X-ray diffraction of the sodium chloride embedded in the silica showed an average crystallite size of 55 nm from the Scherrer equation. Hence, further addition of inorganic salts can donate mesoporosity to the porous silica of a higher surface area. CA-2(10)W has a specific surface area of 860 m2/g, which is the largest in this experiment derived from 5 wt % SiO2 content nano-silica solutions. The aggregate particle size calculated from the surface area is 3.2 nm, which almost agrees with the early stage of the silica sol particle size as described by Iler.1 3.3. Macropore Development in Silicas with Inorganic Salts. As removal of coprecipitated organic salt crystallites in silica produces high porosity, a similar procedure using inorganic acids was applied to prepare porous silicas. SEM images of the spray-dried product using sulfuric acid (HS-2) and its washed sample (HS2W) show that both samples indicate similar spheres with surface roughness as seen in PA-3W. The effect of the addition of sodium sulfate is observed in the SEM images. An SEM image of HS-2(20) indicates that the surface is covered by crystalline materials. After washing, cracks

8046

Langmuir, Vol. 21, No. 17, 2005

Figure 4. XRD patterns of (a) HS-2(30), (b) HS-2(20), (c) HS2(10), (d) HS-2, and (e) HS-2W.

Figure 5. Pore size distributions by mercury porosimetry for (a) HS-2(30), (b) HS-2(20), (c) HS-2(10), (d) HS-2, and (e) NPSilica.

are produced on the silica surfaces of HS-2(20)W. The width of the cracks is in the macropore region (see Figure C in the Supporting Information). The properties of samples from inorganic acids are collected (see Table A in the Supporting Information). According to the XRD measurement as shown in Figure 4, HS-2 contains NaHSO4‚H2O, and HS-2(10), HS-2(20), and HS-2(30) contain Na3H(SO4)2 as the main component. The sulfates were removed by washing, indicating an amorphous phase by XRD. Additionally, analytical data support the XRD results. The EDAX results indicated that most of the sulfates in the spray-dried samples were removed by washing; Na and S contents in the washed samples were in the range of 0.2-0.5 and 0.1-0.3 wt %, respectively. Therefore, the sulfate crystals produced in the spray-dried silicas work as the template for macropores when the concentration of sodium sulfate in the nano-silica solutions is high enough to produce the appropriate amount of sulfate crystals. The macropore region was investigated by mercury porosimetry as shown in Figure 5. Usually, spaces between particles are measured as pores by mercury porosimetry. Since NP-Silica is nonporous silica, the mercury porosimetry result of NP-Silica can be used as a blank. The space sizes between the particles of NP-Silica are distributed around 1000-2000 nm, and its peak top is at 1150 nm. The peak tops of silicas prepared using sulfuric acid shift to the lower side and the pore size distribution

Isobe et al.

Figure 6. Scanning electron micrographs of (a) HC-2, (b) HC2W, (c) HC-2(20), and (d) HC-2(20)W.

becomes wider compared to those of NP-Silica. In HS-2W, mesopores are observed at 10-30 nm. In HS-2(10)W, a second peak slightly appears around 100-500 nm. Macropore volumes, VHg, are estimated from the region greater than 50 nm by subtracting the pore volume of NP-Silica by mercury porosimetry measurement (see Table A in the Supporting Information). The macropore volume increases with increasing concentration of salt; the macropore volume of HS-2(30)W, 2.8 mL/g, is about 8.5 times greater than that of HS-2W. Silicas prepared from hydrochloric acid show different morphologies. Typical SEM images of silicas from hydrochloric acid are shown in Figure 6. The spray-dried sample without the addition of salt, HC-2, its washed one, HC-2W, and the spray-dried sample with 20 wt % NaCl, HC-2(20), indicate similar spheres with rough surfaces. However, the washed sample with addition of 20 wt % NaCl, HC-2(20)W, has macropores whose shape is different from that of samples prepared using sulfuric acid (see Figure Cd in the Supporting Information). It is a uniform shape templated by a crystalline sodium chloride because the samples without washing include crystalline sodium chloride. A TEM image of HC-2(20)W is rather different from the SEM image as shown in Figure 3b. It looks like a crystal because of the low density and the shape created by NaCl crystals. The intensities of XRD patterns show that the crystallinity depends on the salt concentrations in the nano-silica solutions (see Figure D in the Supporting Information). The crystallite sizes were calculated using the Scherrer method (see Table A in the Supporting Information). The crystallite sizes increase with an increase of the concentration of NaCl. The value 180 nm for HC-2(20) agrees with the macropore size in HC-2(20)W observed by SEM, although the value of 125 nm for HC-2 does not. Analytical data by EDAX also showed that most of the sodium chloride in the spray-dried samples was removed by washing; Na and Cl contents in the washed samples were in the range of 0.1-0.4 and 0.00-0.03 wt %, respectively. The results of mercury porosimetry measurements for silicas from hydrochloric acid show that samples with addition of the salt have bimodal peaks in the macropore region (see Figure E in the Supporting Information) in contrast with the bimodal mesoporous silicas.22 The bimodal distribution is probably induced by the particle size distribution. Large silica particles have larger pores than small silica particles, as seen in the SEM image of HC-2(20)W (Figure 6d). Therefore, if silicas with a uniform

Micropore to Macropore Structure-Designed Silicas

particle size can be prepared by other methods such as ultrasonic spraying, the pore size distribution should indicate a single peak. Furthermore, those two peaks become closer with the salt concentration; the crystallite size distribution must become narrower with an increase of the salt concentration. The macropore volume of HC2(20)W, 2.1 mL/g, is about 5 times greater than that of HC-2W. 3.4. Effect on Microporosity of the Aggregation of Primary Silica Particles. The nitrogen adsorption isotherm of spray-dried silica prepared from sulfuric acid, HS-2, indicates the presence of a slight amount of micropores and mesopores. The nitrogen adsorption isotherm of the washed silica HS-2, HS-2W, indicates that micropores and mesopores remarkably increase in contrast with those of HS-2. The shape of the isotherm of HS-2W is quite analogous to that of HS-2, suggesting that a trace of pores is already present in HS-2. However, the micropore volume of HS-2W is not so large compared with those of silicas from carboxylic acids (see Figure Ba,b in the Supporting Information). The pH of the nano-silica solutions from inorganic acids in this research, 0.7 for sulfuric acid and 0.2 for hydrochloric acid, is below the PZC. Therefore, silica nanoparticles in the nano-silica solutions bear positive charges. According to the Schulze-Hardy rule,23 colloidal particles with positive charge are coagulated by the addition of anions, and the coagulation value of polyvalent anions is much lower than that of monovalent anions. Because a lower coagulation value indicates a stronger effect for aggregation, the sulfuric ion coagulates silica nanoparticles much more strongly than the chlorine ion. Also the concentration of anions affects the aggregation of silica nanoparticles; the higher the concentration of anions, the stronger the effect for aggregation. The aggregation of silica nanoparticles can be evaluated by the specific surface area measured by the nitrogen adsorption technique. The pore structural parameters of washed silica samples prepared from sulfuric acid and hydrochloric acid were calculated using their nitrogen adsorption isotherms (see Table A and Figure Bb,c in the Supporting Information). Addition of a salt affects the stability of the electric double layer around silica nanoparticles in nano-silica solutions. Especially anions coagulate positively charged silica nanoparticles. The nitrogen adsorption isotherms indicate the remarkable effect of the addition of sulfate salt. Furthermore, the plot of specific surface area versus salt concentration of the nanosilica solutions, in which the salt produced by the neutralization is also considered, shows a clear effect of salt addition as shown in Figure 7. The BET surface area of HS-2(10)W is almost similar to that of HS-2W, while (22) Sun, J.-H.; Shan, Z.; Maschmeyer, T.; Coppens, M.-O. Langmuir 2003, 19, 8395. (23) Verwey, E. J. W.; Overbeek, J. Th. G. Theory of the Stability of Lyophobic Colloids; Elsevier: Amsterdam, New York, 1948; pp 117119.

Langmuir, Vol. 21, No. 17, 2005 8047

Figure 7. Relationship between the salt concentration of nanosilica solutions from inorganic acids and the BET specific surface areas of porous silicas: b, sulfuric acid; 2, hydrochloric acid.

those of HS-2(20)W and HS-2(30)W decrease remarkably. Addition of sodium sulfate of concentration greater than 0.6 M in sulfuric acid solutions destroys the electric double layer in this system. On the other hand, in the case of using hydrochloric acid, the reduction of BET surface areas with salt concentration up to 2 M is not significant. Doubly charged sulfuric ions affect the aggregation of silica nanoparticles much more effectively than singly charged chloride ions, following the Schulze-Hardy rule. 4. Conclusions Porous silica particles with micropores to macropores are prepared from spray-dried nano-silica solutions. Porous silicas with high micropore volumes are produced using carboxylic acids. The formation of micropores is influenced by the pH of the nano-silica solutions and the molecular size of the carboxylic acids. The addition of a salt to a citric acid solution increases the mesopore volume remarkably. The maximum BET surface area was 860 m2/g, corresponding to an aggregate particle size of 3.2 nm, which agrees with the silica sol particle size. Porous silicas with macropores are produced using inorganic acids containing salts. Salt nanoparticles crystallized in the silica spheres work as a kind of template. The micropore volume of silicas prepared by adding sodium sulfate decreases with the salt concentration, while that of silicas prepared by adding sodium chloride does not. Acknowledgment. We thank Fuji-Sylisia Chemical Ltd. for conducting the mercury porosimetry measurements. Supporting Information Available: Properties of samples prepared from inorganic acids, BET specific surface area of porous silicas versus pH, nitrogen adsorption isotherms, SEM images of the spray-dried product, XRD patterns of spray-dried samples, and pore diameter distributions. This material is available free of charge via the Internet at http://pubs.acs.org. LA0509192