Characterization of Titanized-silica Chromatographic Supports for High Performance Liquid Chromatography (HPLC)
Anízio M. de Faria, Daniel R. Magalhães, Carol H. Collins
LabCrom - Instituto de Química - Universidade Estadual de Campinas
Caixa Postal 6154 - Campus “Zeferino Vaz” - CEP 13083-970 - Campinas, SP, Brasil
No intuito de unir as propriedades cromatográficas desejáveis da sílica e a maior estabilidade hidrolítica oferecida pela titânia, procurou-se neste trabalho promover a síntese do suporte de sílica-titanizada, bem como caracterizar o material obtido para elucidação da disposição do óxido de titânio na superfície da sílica.
Para alcançar os objetivos propostos, vários métodos físicos e químicos de caracterização foram empregados neste trabalho, tais como fluorescência de raios X, emissão fotoeletrônica de raios X (XPS), espectroscopia no infravermelho (IR), microscopia eletrônica de varredura (SEM), etc.
Análises de área superficial específica deste material mostraram que a incorporação de titânio na sílica não afeta drasticamente as propriedades da sílica pura. A área superficial da sílica diminuiu em apenas 20%. Micrografias eletrônicas de varredura mostraram que, apesar das condições drásticas aplicadas durante a síntese, as partículas permaneceram intactas e uniformes. A possibilidade de formação de ligação química entre os óxidos inorgânicos, dando maior estabilidade ao suporte cromatográfico, parece ser remota. Análises por XPS, DR-UV-vis e FTIR fornecem resultados similares a aqueles obtidos para interações fracas entre a sílica e a titânia, como as forças de Van der Waals. Porém, a obtenção deste material é satisfatória para finalidades cromatográficas e se apresenta como alternativa em potencial para suprimir os problemas relacionados à sílica.
Palavras chaves: Sílica-titanizada, suportes cromatográficos, caracterização.
With the intention to combine the desirable chromatographic properties of silica and the greater hydrolytic stability offered by titania, this work investigated the synthesis of a titanized-silica support, characterizing the material obtained to elucidate the disposition of titanium on the silica.
To reach these objectives, several physical and chemical characterization methods were used, such as X-ray fluorescence (XRF), infra-red (IR) and X-ray photoelectron emission (XPS) spectroscopies, scanning electron microscopy (SEM), etc.
Analyses of the specific surface area of this material show that titanium incorporation onto silica does not drastically affect the properties of the pure silica, as the silica surface area is decreased by only 20%. Scanning electron micrographs showed that, despite the drastic conditions applied during the synthesis, the particles remained unbroken and uniform. The possibility of formation of a chemical bond between titania and silica, giving even greater stability to the chromatographic support, seems to be less probable. Analyses by XPS, DR-UV-vis and FTIR spectroscopies suggest only weak interactions between silica and titania, such as from Van der Waals forces. However, the attainment of this material is satisfactory for chromatographic purposes and it presents a potential alternative to suppress the stability problems related to silica.
Keywords: Titanized-silica, chromatographic supports, characterization.
High performance liquid chromatography in the reversed phase mode (RP-HPLC or RPLC) is the principal technique used for analysis of environmental compounds; in the agro-industrial; chemical, pharmaceutical and petrochemical industries; and in many other fields of science. This is due the availability of numerous stationary phases that allow obtaining the desired selectivity for analysis of such compounds [1,2]. In RPLC, the stationary phase is less polar than the mobile phase. Usually, the stationary phase is constituted of a rigid and porous solid support covered with a stationary liquid, often retained through chemical bonds [2,3].
Among chromatographic supports, silica (SiO2) is the most used, with over 90% of the applications already published in the RPLC field . Its widespread use is due mainly to its highly favorable properties for application in the chromatographic process, such as: mechanical resistance to high pressures, allowing conditions of rapid mass transfer during chromatographic separations; the ease of incorporation of groups having different polarities (stationary liquids) onto its surface, due to the presence of highly reactive silanol groups (SiOH); commercial availability of different silica types, with a wide variety of particle diameters, shapes and pore sizes; etc. [5-10].
In spite of these numerous advantages, stationary phases based on silica supports present two important limitations with respect to use in RPLC. The first is related to the presence of the residual silanol groups that did not react during the modification of the silica surface with the stationary liquid [4,7]. The residual silanols affect the chromatographic parameters, because the silanols may interact with the analytes, introducing normal phase separation mechanisms : absorption (interaction with the stationary liquid) and adsorption (interaction with the residual silanol groups), hindering the repeatability of the chromatographic runs. Besides, basic compounds interact strongly with residual silanols and are strongly adsorbed, resulting in separations with low chromatographic performances . The other important limitation is the hydrolytic instability of silica and of the silica-based stationary phases when acidic or basic mobile phases are used, because silica possesses stability only in the range of pH 2-8 [12-14]. Below pH 2, siloxane linkages undergo hydrolytic attack and the liquid bonded onto the silica is slowly removed from the surface, leading to reduction of chromatographic resolution. Above pH 8, the silica is slowly solubilized, leading to a collapse of the column bed [5,6].
Seeking to reduce these limitations and to produce more stable stationary phases, different alternatives have been proposed. Silica has been replaced by alternative supports, both organic [1,4] and inorganic [15,16]. Polystyrene-divinylbenzene  is an example of an organic support employed for analyses in HPLC. In spite of its greater hydrolytic stability, these materials are inconvenient for use in RPLC due to their low mechanical strength, since HPLC uses high pressures to force the passage of the mobile phase though the column . Inorganic supports, such as titania [15,18], zirconia [15,19] and alumina  have been used but, in spite of their greater stability over a larger pH range, they do not perform with the same efficiency as silica. Different strategies have been used to increase the stability of stationary phases on silica and to reduce the activity of the residual silanol groups. These include end capping , sterically protected phases , horizontal polymerization , embedded polar group stationary phases  and monolithic phases . Another strategy, initially proposed by Gushikem and co-workers [26,27], later having its chromatographic value demonstrated by workers from LabCrom [28-31], is the synthesis of silica-based supports covered with metallic oxides, such as zirconia and titania. These supports have the objective of uniting the desirable chromatographic properties of silica and the stability offered by the inorganic materials.
Due to the good perspectives for these new stationary phases, this work aimed at promoting the synthesis of a silica support covered with titania and its physical and chemical characterization, to elucidate the structure of this new support material.
Spherical Kromasil silica (Akzo Nobel) having a mean particle diameter of 5 mm, 0.90 mL g-1 specific pore volume and 340 m2 g-1 specific surface area was used to prepare the chromatographic support. Titanium tetrabutoxide, TiBuO4, was obtained from Aldrich. Water was distilled and then purified in a Milli-Q system from Millipore. Analytical reagent grade or HPLC grade solvents were obtained from Tedia (toluene, butanol).
2.2 Synthesis of Titanized-Silica (Si-Ti)
Titanized silica was synthesized by an adaptation of the method of Fonseca et al.  by reaction of silica with titanium (IV) tetrabutoxide in toluene (or butanol). The synthesis was carried out according to the 24-1 fractional factorial design shown in Table 1. 1.5 g of silica were dissolved in previously dried toluene (or butanol) and 3.0 g (or 3.8 g) of titanium tetrabutoxide was added. The mixture was placed in a thermostated bath at 25 °C (or 50 °C) for 3 h (or 5 h). After this period, the solution was centrifuged for 15 min, the supernatant was discarded and the resulting solid was washed with dried solvent and again centrifuged. This step was repeated five times. The material was then hydrolyzed with 10-3 mol L-1 nitric acid, washed four times with water to remove all organic residues from the surface of the support and finally dried at 120 °C for 24 h before characterization.
Table 1: Fractional factorial design for the synthesis of the titanized-silica.
2.3 Characterization of the Titanized-Silica Support
X-ray Fluorescence (XRF). Titanium incorporation onto silica was quantified by X-ray fluorescence. The titanized-silica samples were analysed using a Shimadzu EDX 700 model instrument.
Thermogravimetric Analysis (TGA). The thermal stability of the titanized silica support was studied using samples of approximately 5 mg, with a heating rate of 10 °C min-1 in an air atmosphere, with a TA model TGA-2050 instrument.
FTIR Spectroscopy. The FTIR spectra were obtained using a Perkin-Elmer model FT-IR 1600 to evaluate the presence of residual silanols and possible alterations caused by incorporation of the titania.
Scanning Electron Microscopy (SEM). A morphological study of the titanized silica support was made through micrographs of the particles with x2000 magnification. The samples were sputter-coated with gold and then examined with a JEOL model JSM-6360LV scanning electron microscope at 20 kV.
X-ray Photoemission Spectroscopy (XPS). The XPS measurements were done using characteristic K radiation from an Al anode to excite the samples and a 100 mm mean radius hemispherical analyzer operated with constant pass energy of 44 eV, which results in a 1.6 eV FWH for the Au 4f line. A small quantity of each sample was pressed between two stainless steel plates to form a thin conglomerate that was fixed to the sample holder with double faced conducting tape. The analyses were done at a base pressure of 5x10-9 mbar and charging effects were corrected by shifting the spectra so that the C 1s line was at 284.6 eV. Prior to fitting the data, using Gaussian, a Shirley background was subtracted (Origin 5 peak fitting module).
Specific Surface Area (sA). Specific surface areas were measured by adsorption of nitrogen at 77 K following the BET method, using a Micromeritcs Flowsorb II 2300 model instrument.
Diffuse Reflectance UV-vis Spectrophotometry (DR-UV-vis). The DR-UV-vis measurements were carried out with a Varian CARY 5G model UV-vis-NEAR spectrophotometer, with diffuse reflectance accessories, to study the characteristic bands relevant to bonding between titanium and silica.
3 Results and discussion
The syntheses of the titanized-silica were carried out according to the 24-1 fractional factorial design shown in Table 1. Eight experiments were carried out. The changes (reaction temperature and time, titanium tetrabutoxide amount and solvent) provided different titanium concentrations on the silica that were quantified by X-ray fluorescence (Table 2).
Table 2. Titanium content incorporated onto the silica surface, quantified by X-ray fluorescence.
The calculation of the main contrasts and interactions between two variables for the factorial design are presented in Table 3. The results indicate that the solvent is the variable that presented the largest contrast. In other words, the variable that most affects titanium incorporation onto silica surface is the reaction solvent. Employing toluene as solvent for the reaction, the largest titanium incorporation onto the silica surface is obtained.
On the other hand, the titanium tetrabutoxide amount, within the levels studied, does not affect titanium incorporation within the confidence interval, 95%, of the t-test. The other variables affect the titanium incorporation process as shown in Table 3.
Table 3: Effects of the factors and interactions obtained by the fractional factorial design.
Equations (1) and (2) present possible reactions involved during the synthesis process of the titanized-silica support.
3.1 Scanning electron microscopy (SEM)
The Kromasil silica used in this work has a spherical form, with a mean particle diameter of 5 mm and a uniform particle size distribution. It is classified as type B, having a high degree of purity and low acidity. During the synthesis process, the silica was submitted to drastic treatments that could damage the particles structure and modify some of its desirable chromatographic properties. Analyses of the titanized-silica samples by scanning electron microscopy, as shown in Figure 1, do not present changes in the shape and particle size distribution, proving the high degree of rigidity of the silica.
Moreover, we show that the synthesis process for titanized-silica was successful. The micrographs show isolated particles, without cluster formation, without visible impurities being present or broken particles. It is important to also point out that the different percentages of titanium incorporated onto the silica surface did not affect the physical structure. Figure 1 shows unreacted silica and the titanized-silica obtained in experiment Si-Ti 02, the other micrographs presented similar appearance and, thus, are not shown.
Figure 1: Electron micrographs of the pure silica (a) and titanized-silica (b) particles.
3.2 Specific surface area (sA)
Another important characteristic of silica, its specific surface area, also was measured to evaluate the effect of the titanium presence on its surface. The surface area of the pure silica samples was, on average, equal to 309 m2 g-1. The titanized-silica samples show lower areas but none were so low as to compromise the sample acceptance capacity of the new materials. The results of the measurements by the BET method are presented in Table 4. These values are, at most, 25% lower compared to pure silica. It can also be observed that the relation between titanium percent and the specific surface area of the titanized-silica samples presents a random behavior. Independent of the titanium content, about 20% of the silica surface area, was lost in the titanized-silica samples.
Table 4: Surface area measurements, by BET method, for pure silica and titanized-silica.
This reduction of the surface area occurs mainly due to a titanium bond, in its elementary or oxide form, at the reactive sites of the silica (silanol groups). As more than 95% of these reactive sites are located inside of the pores, there is the possibly of a partial filling of the biggest pores or total filling of micropores, diminishing the surface area of the titanized-silica.
3.3 Thermogravimetric analysis (TGA)
The thermogravimetric analyses of the titanized-silica samples (Figure 2) show a low percentile weight loss of the new materials, less than 7% of the total weight, in general. This weight loss is mainly caused by the evaporation of the weakly adsorbed water (up to 150 °C), and physically adsorbed water (200-450°C) or condensation surface hydroxyls groups of the silica or titania (> 600°C). However, the material presents good thermal resistance and is appropriate for use in RPLC, as chromatographic analyses normally are carried out at room temperature, or on special occasions, up to, at most, 90°C. The titanium percent incorporated onto the silica surface seems not to have an effect on the thermal stability of the support. Therefore, the different titanium concentrations on the silica showed similar weight losses.
Through the thermograms of the titanized-silica samples it was also possible to evaluate its purity. No accentuated weight losses were observed which would be due to combustion of residual tetrabutoxides. Thus, the hydrolysis and washing of the material was sufficient to remove all the residues.
As all thermograms presented similar results. Thus only two are shown in Figure 2, one that presented the greatest (Figure 2a) and the other (Figure 2b) the least thermal resistance.
FFigure 2: Thermograms of the titanized-silica samples. (a) Si-Ti 03 and (b) Si-Ti 04.
3.4 Fourier transform infra-red spectroscopy (FTIR)
The analysis of the samples by Fourier transform infra-red spectroscopy (FTIR) allow evaluations of changes in the properties of the silica support, such as in stretching of the free silanols (ºSiOH) and geminal (ºSiOH2) groups, the reactive sites of the silica that make possible the incorporation of other materials. The IR spectra of pure silica (Figure 3a) and titanized-silica (3b) present the same bands, implying in a similarity of the properties of these materials. The main bands of the spectra can be attributed to the stretching (n) of hydroxyl groups of the physically adsorbed and hydrogen bonded water and the geminal hydroxyl groups (3500 cm-1), siloxane groups (ºSi-O-Si) (~1100 cm-1) and free silanol groups (~975 cm-1).
With the increase of the titanium percent in the silica there is a reduction of the intensity, mainly of the band relative to the stretching of the silanol groups (975 cm-1). This band does not disappear completely even in the samples with larger titanium concentrations, as only 50% of the total silanol groups are available for reaction.
Two possible types of interaction between TiO2 and SiO2 exist; they can be physically mixed with interaction forces governed by the weak Van der Waals forces; or chemically bonded, where the formation of the Si-O-Ti linkages occurs. When chemically bonded, the chromatographic support presents greater stability compared to the support formed by weak interactions between silica and titania. From the FTIR it should be possible to determine the existence of these chemical bonds. A band in the region of 930-960 cm-1 is accepted as the characteristic vibration due to the formation of Si-O-Ti bonds. However, the IR spectra (Figure 3) do not present indications of the formation of this bond. Others techniques, such as DR-UV-vis and XPS, should give further information on the type of bonds between the two oxides.
3.5 Diffuse Reflectance UV-vis Spectroscopy (DR-UV-vis)
Another form to show the presence of chemical bonds between titanium and silicon oxides is the analysis of the diffuse reflectance UV-Vis spectra. The samples were submitted to DR-UV-vis, to establish the coordination geometry of Ti atoms in the silica matrix. The bands that make it possible to identify these geometries and, consequently, to infer the bond type, are located at 222-245 nm (TiO2-SiO2 chemically bonded - tetrahedral coordination) and at 208-256 nm (TiO2/SiO2 only supported - octahedral coordination) . The titanized-silica samples did not presented the former bands (Figure 4). The DR-UV-vis spectra shown in Figure 4 were corrected for a standard, that consisted of a mechanic mixture of silica and titania, suggesting that the bonding between these oxides is governed only by weak Van der Waals forces, independent of the titanium concentration.
Figure 3: Infra-red spectra of pure silica (a) and titanized-silica (Si-Ti 01-06) (b) samples.
Figure 4: DR-UV-vis spectra of the titanized-silica samples.
3.6 X-ray Photoelectron Spectroscopy (XPS)
XPS was used to determine the atomic splitting of the elements present on the support surface and their binding energies. Table 5 shows the results for two titanized-silica samples and a SiO2/TiO2 mechanical mixture. The XPS spectra presented peaks in 459.0 eV and 459.6 eV (samples 4 and 5, respectively), that were attributed to the 2p3/2 electron of titanium. This binding energy is similar to the mechanical mixture, 459.4 eV. The spectra also shows peaks at 532.3 and 531.0 eV (samples 4 and 5, respectively) attributed to oxygen, while that for the mechanical mixture is 529.6 eV. However, when the oxygen peak is deconvoluted by the Gaussian method, it results in two peaks, seen for the samples and for the mechanical mixture. These results are also shown in Table 5. The peak of higher energy was attributed to the oxygen bonded to the silicon atom and of lesser energy to the oxygen bonded to the titanium atom.
Table 5: Binding energies for titanized-silica and mechanic mixture samples determined by XPS.
The reacted oxide showed two oxygen Auger parameters, whose values are closer to each other than the values of the mechanical mixture. This fact is correlated with the energy splitting between two peaks associated with the O 1s photoemission signal, as shown in Table 5. This splitting is 2.8 eV for the physical mixture and 1.9 eV for the reacted oxides. Thus, the samples submitted for analyses presented a lower splitting than that for the mechanical mixture, 2.6 eV. However it is less than expected for oxide bonding, indicating only weak interactions between the oxides. The charge of the Ti atom of titanized-silica changes due to the interface formation of TiO2/SiO2. The formation of interface Ti-O-Si linkages decreases the positive charge of the Ti atoms at the interface, resulting in a lower binding energy of the 2p electron of Ti in TiO2/SiO2 , which was confirmed by XPS. The binding energy of the 2p-electron of Ti of pure titania and TiO2/SiO2 was 458.6 eV and 459.4 eV, respectively. Figure 5 shows the spectra for the titanized-silica 05 sample, for Ti 2p (a); Si 2p (b) and O 1s (c). Other samples, including the mechanical mixture, were similar.
Figure 5: Binding energies for titanized-silica sample 05, (a) Ti 2p, (b) Si 2p and (c) O 1s.
Through the characterization studies it is concluded that approximately 7% of titanium were incorporated onto silica using the most favorable reaction conditions, without affecting the structure of the silica particles. Other properties of the material were observed through different techniques. Despite not having evidence for the formation of a strong chemical bond between silicon and titanium, the weak interactions that join these oxides provide stability sufficient for chromatographic requirements.
The authors thank FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) and CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico) for fellowships and financial support and Prof. Richard Landers (Instituto de Física da Unicamp) for performing the XPS analyses.
 STELLA, C.; RUDAZ, S.; VEUTHEY, J. L.; TCHAPLA, A. Silica and Other Materials as Support in Liquid Chromatography. Chromatographic Tests and their Importance for Evaluating these Supports. Part I. Chromatographia, v.53, S113 - S131, 2001.
 SNYDER, L. R.; KIRKLAND, J. J. Introduction to Modern Liquid Chromatography. 2nd ed. New York: John Wiley and Sons, 1979.
 COLLINS, C. H.; BRAGA, G. L.; BONATO, P. S. Introdução a Métodos Cromatográficos. 4th ed. Campinas: Editora da Unicamp, 1990.
 TONHI, E.; COLLINS, K. E.; JARDIM, I. C. S . F.; COLLINS, C. H. Fases Estacionárias para Cromatografia Líquida de Alta Eficiência em Fase Reversa (CLAE-FR) Baseadas em Superfícies de Óxidos Inorgânicos Funcionalizados. Quim. Nova, v.25, 616-623, 2002.
 BERTHOD, A. Silica: Backbone Material of Liquid Chromatography Columns Packings. J. Chromatogr., v.549, 1-28, 1991.
 UNGER, K. K. Porous Silica, Its Properties and Use as Support in Column Liquid Chromatography. 3rd ed. New York: Elsevier Science Publishing Company Inc., 1986.
 NAWROCKI, J. The Silanol Group and Its Role in Liquid Chromatography. J. Chromatogr. A, v.779, 29-71, 1997.
 BUCHMEISER, M. R. New Synthetic Ways for the Preparation of High-Performance Liquid Chromatography Supports. J. Chromatogr. A, v.918, 233-266, 2001.
 JARONIEC, C. P.; JARONIEC, M.; KRUK, M. Comparative Studies of Structural and Surface Properties of Porous Inorganic Oxides Used in Liquid Chromatography. J. Chromatogr. A, v.797, 93-102, 1998.
 ZHURAVLEV, L. T. The Surface Chemistry of Amorphous Silica. Zhuravlev Model. Colloids and Surfaces A, v.173, 1-38, 2000.
 Nahum, A.; Horvath, C. Surface Silanols in Silica-Bonded Hydrocarbonaceous Stationary Phases: I. Dual Retention Mechanism in Reversed-Phase Chromatography. J. Chromatogr., v.203, 53-63, 1981.
 KIRKLAND, J. J.; VAN STRATEN, M. A.; CLAESSENS, H. A. High pH Mobile Phase Effects on Silica-Based Reversed-Phase High-Performance Liquid Chromatographic Columns. J. Chromatogr. A, v.691, 3-19, 1995.
 KIRKLAND, J. J.; HENDERSON, J. W.; DE STEFANO, J. J.; VAN STRATEN, M. A.; CLAESSENS, H. A. Stability of Silica-Based, Endcapped Columns with pH 7 and 11 Mobile Phases for Reversed-Phase High-Performance Liquid Chromatography. J. Chromatogr. A, v.762, 97-112, 1997.
 KIRKLAND, J. J. Stability of Silica-Based, Monofunctional C-18 Bonded-Phase Column Packing for HPLC at High pH. J. Chromatogr. Sci., v.34, 309-313, 1996.
 NAWROCKI, J.; DUNLAP, C.; LI, J.; ZHAO, J.; MCNEFF, C. V.; MCCORMICK, A.; CARR, P. W. Part I. Chromatography Using Ultra-Stable Metal Oxide-Based Stationary Phases for HPLC. J. Chromatogr. A, v.1028, 1-31, 2004.
 KURGANOV, A.; DAVANKOV, V.; ISAJEVA, T.; UNGER, K. K.; EISENBEISS, F. Characterization of Covalently Bonded and Adsorbed Polymer Coatings on Silica, Alumina and Zirconia by Means of Physico-Chemical and Chromatographic Methods. J. Chromatogr. A, v.660, 97-111, 1994.
 MAJORS, R. E. Advances in the Design of HPLC Packings. LC-GC, v.18, 586-598, 2000.
 MURAYAMA, K.; NAKAMURA, H.; TAKAHASHI, K.; YOSHIDA, A. Reversed-Phase Separation of Basic Solutes with Alkaline Eluents on an Octadecyl Titania Column. Anal. Sci., v.10, 815-816, 1994.
 Li, J.W.; Carr, P.W. A Study of the Efficiency of Polybutadiene-Coated Zirconia as a Reversed-Phase Chromatographic Support. Anal. Chem., v.69, 2193-2201, 1997.
 PESEK, J. J.; SANDOVAL, J. E.; SU, M. New Alumina-Based Stationary Phases for High-Performance Liquid Chromatography: Synthesis by Olefin Hydrosilation on a Silicon Hydride-Modified Alumina Intermediate. J. Chromatogr., v.630, 95-103, 1993.
 SUDO, A.; WADA, T. Characteristics of Octadecylsilylated Silica Gels End-Capped by High-Temperature Silylation. J. Chromatogr. A, v.813, 239-246, 1998.
 KIRKLAND, J. J.; GLAJCH, J. L.; FARLEE, R. D. Synthesis and Characterization of Highly Stable Bonded Phases for High-Performance Liquid-Chromatography Column Packings. Anal. Chem., v.61, 2-11, 1989.
 WIRTH, M. J.; FATUNMBI, H. O. Horizontal Polymerization of Mixed Trifunctional Silanes on Silica - a Potential Chromatographic Stationary Phase. Anal. Chem., v.64, 2783-2786, 1992.
 CZAJKOWSKA, T.; JARONIEC, M. Selectivity of Alkylamide Bonded-Phases with Respect to Organic Acids under Reversed-Phase Conditions. J. Chromatogr. A, v.762, 147-158, 1997.
 MEYERS, J. J.; LIAPIS, A. I. Network Modeling of the Convective Flow and Diffusion of Molecules Adsorbing in Monoliths and in Porous Particles Packed in a Chromatographic Column. J. Chromatogr. A, v.852, 3-23, 1999.
 KUBOTA, L. T.; GUSHIKEM, Y; MOREIRA, J. C. Adsorption of Chromium(VI) by Titanium(IV) Oxide Coated on a Silica-Gel Surface. Analyst, v.116, 281-283, 1991.
 PEIXOTO, C. R. M.; GUSHIKEM, Y.; BACCAN, N. Selective Spectrophotometric Determination of Trace Amounts of Chromium(VI) Using a Flow-Injection System with a Microcolumn of Zirconium(IV) Oxide Modified Silica-Gel. Analyst, v.117, 1029-1032, 1992.
 MELO, L. F. C.; COLLINS, C. H.; COLLINS, K. E.; JARDIM. I. C. S. F. Stability of High-Performance Liquid Chromatography Columns Packed with Poly(Methyloctylsiloxane) Sorbed and Radiation-Immobilized onto Porous Silica and Zirconized Silica. J. Chromatogr. A, v.869, 129-135, 2000.
 SILVA, R. B.; COLLINS, K. E.; COLLINS, C. H. Effects in high-performance liquid chromatography of a high pH in the mobile phase on poly(methyloctylsiloxane) immobilized by g-radiation on titanium-grafted silica. J. Chromatogr. A, v.869, 137-, 2000.
 FONSECA; D. A.; COLLINS, K. E.; COLLINS, C. H. Titanized silica-based stationary phases prepared with thermally and microwave-immobilized poly(methyloctylsiloxane). J. Chromatogr. A, v.1030, 209-215, 2004.
 MELO, L. F. C.; JARDIM, I. C. S. F. Development of C8 Stationary Phases Immobilized by g-Radiation on Zirconized Silica for High-Performance Liquid Chromatographic Applications. J. Chromatogr. A, v.845, 423-431, 1999.
 GAO, X.; WACHS, I. E. Titania-Silica as Catalysts: Molecular Structural Characteristics and Physico-Chemical Properties. Catal. Today, v.51, 233-254, 1999.
 RYU, D. H.; KIM, S. C.; KOO, S. M. Deposition of Titania Nanoparticles on Spherical Silica. J. Sol-Gel Sci. Techn., v.26, 489–493, 2003.