Optimization of the Production Method for Mesoporous Silica

OPTIMIZATION OF THE PRODUCTION METHOD FOR MESOPOROUS SILICA

Viktor A. Valtsifer, Vladimir N. Strelnicov, Natalia B. Kondrashova,

Ekaterina V. Saenko, Valery I. Karmanov

Institute of Technical Chemistry, Ural Division of the RAS, Perm, Russia

Abstract

The influence of synthesis conditions, of the ratio between reaction components (CTAB/TEOS, NaOH/TEOS, water/TEOS), of the use of organic additives on textural and structural characteristics of the finished material during production of mesoporous silica by the sol-gel method has been studied. Optimization of technological procedures has enabled to produce the material with high specific surface (1400 m2/g), total pore volume over 1 cm3/g and average pore diameter approximately 4 nm.

1. Introduction

Mesoporous mesophase materials (MMM) with well-expressed porous hexagonal structures with dimension range 2-10 nm are currently of a significant interest. Such materials are promising for the use in chemical and chemical-biological processes as inorganic sorbents, catalysts and catalyst carriers. Therefore, the production of mesoporous synthetic materials with controllable pore dimensions is of current importance.

One of the most effective methods for the production of nano-scale materials with predictable properties of a finished product is the chemical liquid-phase condensation method based on formation of poorly soluble compounds in the course of chemical interactions. Alterations of sedimentation conditions temperature, pH, ratio/concentration of components, introduction of surfactants and of organic modifiers allow the wide-ranging control of phase composition, size and shape of resulting particles 1. Literary sources on the matter are inconsistent, thereforethe optimization of technological procedures for the production of mesoporous silicaas a basis of metal oxide catalysts appeared necessary.

1. Experimental part

Tetraethoxysilane (TEOS) was used as a source of silica. Being hydrolyzed, TEOS produces SiO2 as per the following reaction:

Si(OC2H5)4 + 4H2O  Si(OH)4 + 4C2H5OH

n Si(OН)4 (SiO2)n + 2n H2O[1].

At that, production of silica is accompanied by various physical processes: phase formation, isothermal over-condensation, coagulation (sol formation) and gel formation. Process kinetics, gel properties and structure of finished materials are in many respects determined by initial conditions of synthesis.

Many methods describing sol synthesis with a narrow pore size distribution curve are based on ideas of phase formation in supersaturated solutions 2-5.

The formation of silica particles grains takes place during induction period and their concentration does not change on subsequent stages of particles growth which is a necessary condition for production of sols with the narrow pore size distribution curve 2. The induction period diminishes as the total concentration of silica increases. In a system with more high initial concentration of silicic acid when a lot of silica particles’ grains are generated, i.e. the inter-phase surface is bigger, the rate of the process is higher 5.

The sol-to-gel transition is related to the formation of spatial chains which penetrate the whole volume of sol. As the silica concentration increases, the strength of gel increases also but at the same time its degradation rate increases as well 2.

The pH value has also its influence. Starting with a certain low over-saturation, polycondensation proceeds only as isothermal over-condensation of particles, and its rate is dependent on the pH value of the system while increasing in the range from 3 to 10 6. At that, the higher is pH value the lower is the strength of the gel structure and the earlier takes place its spontaneous destruction. Structure and properties of gel-like layers of the surface of a silica particle are also dependent on the quantity of alkali in the system. Interaction between alkali hydroxides and silica particles is accompanied by destruction of siloxane bonds, the same factors lead to the advent of steric repulsion forces between particles. In such gels, sediment separation of the system in two phases takes place. Thus, the formation of sedimentally unstable dispersions – not of gels – becomes possible 7,8.

The use of surfactants as peculiar matrixes plays enormous significance during the formation of porous structures. Cationic hexadecyltrimethylammonium bromide (CTAB) – CH3(CH2)15N(CH3)3+ Br- – was used as a surfactant. On the dissolution of the surfactant as a result of micelle aggregation, the so-called hexagonal tubes are generated surrounded by groups of silicates followed by the formation of inorganic structures reflecting the structure of hexagonal micellar matrixes. The length of carbonic residue of the surfactant plays also a considerable role – it determines sizes of micelles and, subsequently, dimensions of pores. The removal of the surfactant (by baking or by means of a solvent) leads to the formation of a mesoporous material (Fig.1) 9.

Fig.1.The formation pattern of hexagonal porous structures.

1 – CTAB micelle; 2 – micellar rod; 3 – hexagonal matrixes; 4 – inorganic structures; 5 – MCM-41

The formation of gel structure was fulfilled by hydrothermal treatment at 80°C for 3 hours; the removal of the surfactant – by the baking of the material. In order to more completely remove the surfactant, temperature and time of baking were specified on the stage of xerogel formation by means of thermogravimetric analysis. Thermal-oxidative degradation of CTAB was carried out on Paulic-Paulic Erdey derivatograph (MOM Company, model Q 1500). The programmed heating was carried out at the rate 10°C/min up to 1000 °C.

Fig.2 (temperature curve) shows that practically complete degradation of CTAB takes place at approximately 650°C. This temperature was taken as the baking temperature on the last stage of synthesis of mesoporous silica.

Methods for production of mesoporous silica with various ratios of initial components are described in the literature. For example, in items 10-12 mol ratios of components were as follows: 1TEOS:0.035STAB:0.0175NaOH:692.5H2O;1SiO2:0.1CTAB:0.28NaOH:25H2O;1SiO2:0.33Na2O:0.5CTAB:74H2O accordingly.

Fig 2. Thermal destruction of CTAB.

Therefore, in order to produce a material with more high specific surface, low dimensions of pores and distribution in a narrow range, three series of experiments with subsequently changing mol ratios were fulfilled:

• Quantity of surfactant (CTAB) to the source of silica (tetraethoxysilane TEOS);
• Quantity of NaOH to TEOS;
• Quantity of liquid phase (water) to TEOS

Textural parameters of samples were determined by nitrogen sorption at -196°C on the Micromeritics ASAP 2020 device after the degassing of the material in vacuum at 350°C for 3 hours. Specific surface of samples was determined as per the BET method:

where:a – absorption value at gas pressure p, am – capacity of adsorption monolayer per mass unit of adsorbent, ps – pressure of saturated vapor of adsorbed substance at given temperature, c = exp(Q-)/RT. The pore size distribution was determined as per desorption isotherms while using the BJH model.

Results of experiments are shown in Tables 1,2 and 3; adsorption/desorption isotherms of gaseous nitrogen and pore size distribution obtained upon analysis of samples are shown on Fig. 3,4, 5.

Table 1. Alterations of textural characteristics of the material depending on the CTAB/TEOS ratio during the synthesis of mesoporous SiO2

Nr. / CTAB/TEOS ratio / Specific surface, m2/g / Total volume of pores, cm3/g / Volume of micropores, cm3/g / Fraction of pores with diameter
1.5 – 5 nm, % / Fraction of pores with diameter
 20 nm, % / Average diameter of pores, nm
1 / 0.11:1 / 892.6 / 1.73 / 0.09 / 5.55 / 82.60 / 21.5
2 / 0.22:1 / 1077.8 / 0.43 / 0.14 / 67.45 / 12.61 / 5.2
3 / 0.33:1 / 1207.1 / 0.49 / 0.45 / 77.20 / 5.95 / 4.5
4 / 0.44:1 / 1270.4 / 1.43 / 0.35 / 78.02 / 11.81 / 3.8
5 / 0.55:1 / 1230.6 / 0.90 / 0.31 / 58.64 / 25.03 / 3.8

Fig.3.On the left: nitrogen adsorption/desorption isotherms; on the right: pore size distribution in samples with various CTAB/TEOS ratios used during the synthesis of mesoporous SiO2.

1 –CTAB/TEOS ratio = 0.11; 2 –CTAB/TEOS ratio = 0.22;3 –CTAB/TEOS ratio = 0.33; 4 –CTAB/TEOS ratio = 0.44; 5 –CTAB/TEOS ratio = 0.55

Table 2. Alterations of textural characteristics of the material depending on the NaOH/TEOS ratio during the synthesis of mesoporous SiO2

Nr. / NaOH/TEOS ratio / Specific surface, m2/g / Total volume of pores, cm3/g / Volume of micropores, cm3/g / Fraction of pores with diameter
1.5 – 5 nm, % / Fraction of pores with diameter
 20 nm, % / Average diameter of pores, nm
1 / 0.1 / 738.6 / 0.17 / 0.07 / 31.06 / 39.28 / 5.6
2 / 0.2 / 1091.4 / 0.55 / 0.10 / 37.48 / 25.03 / 5.3
3 / 0.3 / 1119.1 / 1.45 / 0.18 / 58.65 / 8.64 / 4.7
4 / 0.4 / 1270.2 / 1.43 / 0.31 / 89.96 / 2.80 / 3.8
5 / 0.5 / 1243.0 / 1.28 / 0.27 / 90.54 / 3.46 / 4.1

Fig.4. On the left: nitrogen adsorption/desorption isotherms; on the right: pore size distributionin samples with various NaOH/TEOS ratios used during the synthesis of mesoporous SiO2.

1 – NaOH/TEOS ratio = 0.1; 2 –NaOH/TEOS ratio = 0.2; 3 – NaOH/TEOS ratio = 0.3;

4 –NaOH/TEOS ratio = 0.4; 5 – NaOH/TEOS ratio = 0.5

Table 3. Alterations of textural characteristics of the material depending on the liquid phase (water)/TEOS ratio during the synthesis of mesoporous SiO2

Nr. / Water/TEOS ratio / Specific surface, m2/g / Total volume of pores, cm3/g / Volume of micropores, cm3/g / Fraction of pores with diameter
1.5 – 5 nm, % / Fraction of pores with diameter
 20 nm, % / Average diameter of pores, nm
1 / 80 / 1188 / 0.85 / 0.47 / 83.85 / 5.70 / 2.9
2 / 100 / 1224 / 1.07 / 0.48 / 88.24 / 3.33 / 3.4
3 / 200 / 1242 / 1.40 / 0.27 / 93.97 / 2.15 / 4.5
4 / 300 / 1256 / 1.46 / 0.35 / 88.72 / 3.40 / 4.6
5 / 400 / 1232 / 1.26 / 0.43 / 81.38 / 9.32 / 4.1
6 / 500 / 1170 / 0.83 / 0.45 / 32.38 / 51.10 / 3.8

Fig.5. On the left: nitrogen adsorption/desorption isotherms; on the right: pore size distribution in samples with various water/TEOS ratios used during the synthesis of mesoporous SiO2.

1 –water/TEOS ratio = 80; 2 – water/TEOS ratio = 100; 3 – water/TEOS ratio = 200; 4 – water/TEOS ratio = 300; 5 –water/TEOS ratio = 400; 6 – water/TEOS ratio = 500

Comparison of results obtained from various series leads to a conclusion that the shape of isotherms (Fig. 3,4,5) is typical for silicate MMMs with hexagonal structure. Initial area of isotherms corresponds to adsorption on the entire surface, area of abrupt upsurge – to the filling of mesopores on mesophase, further the adsorption proceeds on the remaining surface. Exclusion is the nitrogen adsorption/desorption isotherm for the NaOH/TEOS ratio = 0.1 (Fig.3); obviously this is related to the availability of pores with big diameter.

The highest values of specific surface, of total volume of porous formations with diameter 3-4 nm in synthesized materials were observed in samples with CTAB/TEOS ratios equal to 0.55:1, 0.44:1 (Table 1); with NaOH/TEOS ratios equal to 0.5:1, 0.4:1; with water/TEOS ratios equal to 100-300 (Table 3).

The totality of obtained experimental data enabled to select optimal components ratios and conditions of synthesis. However, in order to improve textural properties of mesoporous silicia, the introduction of organic additives (2-cyanoethyltriethoxysilane 13, 2-aminopropyltriethoxysilane 14 and for the first time tris-trimethylsiloxysilane) during the synthesis was proposed. The additives/TEOS mol ratio equaled to 1:10.

The influence of additives on textural characteristics of silica is shown in Table 4 and on Fig.6. It can be seen that the introduction of additives proposed by us significantly improves specific surface of the material.

The influence of organic additives on properties of mesostructural silicae is related, apparently, to the redistribution of the charge density as the condensation of SiO2 goes on. The introduction of 2-cyanoethyltriethoxysilane, owing to electron-seeking properties of CN-, leads to the decrease in the charge density on the framework of silica; the introduction of the amine group, owing to its electron-releasing properties, leads to contrary results. The influence of tris-trimethylsiloxysilane during the synthesis is, most likely, due to the fact that apart from the decrease in the charge density on generated particles, concentration of condensation grains of the new phase increases as a result of hydrolysis of this reagent. Introduction of organic modifiers, finally, changes interaction between silica which is generated as the condensation goes on, and surface-active substance (cetyltrimethylammonium bromide), and this interaction affects the mesoporous structure and surface properties of synthesized materials.

Structural characteristics of mesoporous materials were explored by means of the Raman spectroscopy, of Fourier-transform infrared spectroscopy and of X-ray phase analysis. The Raman spectroscopy was fulfilled on the multifunctional SENTERRA spectrometer (Bruker, Germany), the Fourier-transform infrared spectroscopy – on the IFS-66 spectrometer (Bruker, Germany).

The use of traditional Fourier-transform infrared spectroscopy for the exploration of nanostructural samples proved to be ineffective, i.e. only absorption bands of the basing material – silica – were observed. Infrared spectra of produced mesoporous materials are shown on Fig.7,8.

In the spectra of silica with specific surface of approximately 1400 m2/g (Fig.7), modes of isolated tetrahedron SiO4 – 979 cm-1, 494 cm-1; of distorted tetrahedron – 830 cm-1 and 443 cm-1 and of a SiO3 fragment – 604 cm-1 as well as weak bands corresponding to SiO2vibrations – 1380, 1050 and 1200 cm-1 (condensed phase) were observed 9. The spectrum of the material with specific surface of approximately 1200 m2/g has resembling appearance with lower intensity of peaks.

Table 4. Textural characteristics of mesoporous silica synthesized with organic additives

Textural characteristics of samples / Addition of 3-aminopropyltriethoxysilane / Addition of 2-cyanoethyltriethoxysilane / Addition of tris- trimethylsiloxysilane / Without organic additives
Specific surface of samples, m2/g / 1004 / 1302 / 1401 / 1270
Total volume of samples pores, cm3/g / 0.9 / 0.8 / 1.4 / 1.4
Average diameter of pores in samples, nm / 3.9 / 3.0 / 3.1 / 3.8

Fig. 6. On the left: nitrogen adsorption/desorption isotherms; on the right: pore size distribution in samples of mesoporous silica synthesized with organic additives.

1 – 3-aminopropyltriethoxysilane; 2 – 2-cyanoethyltriethoxysilane; 3 – tris- trimethylsiloxysilane

Fig. 7. Data of the Raman spectroscopy of mesoporous silica with specific surface 1400 m2/g.

Fig. 8. Data of the Raman spectroscopy of mesoporous silica with specific surface 1270 m2/g.

The X-ray phase analysis of samples was carried out on the XRD-7000 diffractometer (Shimadzu, Japan) using the CuKa-radiation in the angular interval 1-10° 2Θ. According to results of analysis, synthesized samples exhibit reflexes (100) and more weak reflexes corresponding to hexagonal (cellular) structure P6mm typical for mesophases like MCM-41 (Fig.9.) Introduction of various organic additives during the synthesis procedure preserves the structural order.

Fig. 9. Patterns of X-ray phase analysis of mesoporous silica synthesized with various organic additives.

1 – tris-trimethylsiloxysilane; 2 – 2-aminopropyltriethoxysilane; 3 – 2-cyanoethyltriethoxysilane; 4 – sample without additives

Conclusions

Thus, synthesis modes of mesostructural silica have been optimized; comparative estimation of textural and structural properties of materials produced with various ratios of components has been fulfilled. Procedures for the improving of textural characteristics by means of introduction of organic substances during the synthesis have been proposed and experimentally checked. All this has enabled to produce materials with high specific surface of approximately 1400 m2/g, with total pore volume over 1 cm3/g and with pore diameter of approximately 4 nm.

References

1. Shabanova N.A., Popov V.V., Sarkisov P.D. Chemistry and technology of nanodispersive oxides. Moscow, Akademkniga, 2007, 309 p.
2. La Mer V.K., Dinegar R.H. Theory, production and mechanism of formation of monodispersed hydrosols // J. Amer. Chem. Soc. 1950. V.72, № 11. P. 4847 – 4854.
3. Frolov Yu.G. Course of colloid chemistry: surface phenomena and dispersive systems, 2nd edition. Moscow, Chemistry, 1989, 464 p.
4. Shukina E.L., Pertsov A.V., Amelina, E.A. Colloid chemistry. MoscowStateUniversity Publications, 1982, 352 p.
5. Shabanova N.A., Sarkisov P.D. Grounds of the sol-gel technology of nanodispersive silica. Moscow, Akademkniga, 2004, 208 p.
6. Shabanova N.A., Popov V.V., Frolov Yu.G. Kinetics of polycondensation and coagulation in silica hydrosols. Colloid J., 1984, vol.46, № 5, pp.986-993.
7. Hurd Ch. Studies on silicate gels. J. Phys. Chem. 1954. V.58, №. 10. pp. 914-915.
8. Shabanova N.A., Silos I.V. The sol – gel transition under conditions of electrolytic coagulation of colloid silica. Colloid J., 1996, vol.58, № 2, pp.266-271.
9. Krylov O.V. Heterogeneous catalysis. Moscow, Akademkniga, 2004, 679 p.
10. Wei Zeng, Xue-Feng Qian, Yan-Bo Zhang, Jie Yin, Zi-Kang Zhu. Materials Research Bulletin, 2005, № 40, pp. 766-772.
11. Liu X., Hui Sun H., Yang Y. Rapid synthesis of highly orderedSi-MCM-41, Journal of Colloid and Interface Science (2007), doi: 10.1016/j.jcis.2007.11.025.
12. Achyut P. Bhatt, Kavita Pathak, Ralksh V. Jasra, Rukhsana I. Kureshy, Noor-ul H. Khan, Sayed H.R. Abdi. Chiral lanthanum-lithium-binaphthol complex covalently bonded to silica and MCM-41 for enantioselective nitroaldol (Henry) reaction. Journal of Molecular Catalysis A: Chemical, 2005, 244, pp. 110-117.
13. Doyle A.M., Ahmed E., Hodnett B.K.The evolution of phases during the synthesis of organically modified catalyst supportМСМ-48. Catalysis Today, 2006.02.083.
14. Samoilova E.A.; Golovnev N.N.; Kozlova, S.A.; Kirik, S.D. The functionalized mesostructural silicates MCM-41 and SBA-15 and their sorption properties. Herald of KrasnoyarskStateUniversity, 2006, № 2, pp.44-52.

1