Carbon dioxide and methane adsorption over metal modified mesoporous SBA-15 silica

Synthesis of materials

Mesoporous silica SBA-15 was synthesized by conventional procedure using triblock copolymer Pluronic P123 ((EO)₂₀(PO)₇₀(EO)₂₀)) and tetraethylortosilicate (Si(OC₂H₅)₄, TEOS) in acidic conditions (HCl) [3]. A typical gel composition in terms of molar ration was TEOS/HCl/H₂O/P123=1/5.9/193/0.017. A portion of 4.0 g of Pluronic P-123 was dissolved with stirring in a mixture of 30 g of water and 120 g of 2 M HCl at 35°C and 8.5 g of TEOS was added after dissolution. The resulting mixture was stirred at 35°C for 20 h and then was aged at 80°C for 24 h. The as-synthesized sample was recovered by filtration and freely dried. The organic template was removed by calcination in air at 500°C for 7 h.

For the preparation of metal doped samples a modified procedure reported in [11] was applied. The titanium(IV)-isopropoxide Ti(OC₃H₇)₄, aluminium(III)-isopropoxide Al(OC₃H₇)₃ and zirconium-tert-butoxide Zr(OC(CH₃)₃)₄ were used as metal sources in the synthesis. In a typical synthesis, 8.5 g of TEOS, the required amount of the respective metal alkoxide to maintain Si/M molar ratio=25 (M=Al, Ti or Zr) and 2g of water were mixed and stirred for 1 h at 20°C. After this time, the mixture was added to the HCl solution of Pluronic P123. The final synthesis gel had the molar ratio Si/M/HCl/H₂O/P123=1/0.04/5.9/193/0.017 (M=Al, Ti or Zr). The resulting mixture was stirred for 24 h at 30°C, then transferred to the teflon lined steel autoclave and kept at 80°C for 24 h. The solid product was recovered by filtration, washed with water several times, and dried overnight. Finally, the product was calcined at 500°C to remove the template.

Characterization

The small angle X-ray scattering (SAXS) experiments were carried out at B1 Hasylab/Desy beamline with the beam energy 12 keV (λ=1.03 Å) using PILATUS detector. The HRTEM micrographs were taken with a JEOL JEM 3010 (LaB₆ cathode) microscope operated at 300 kV. Copper grid coated with a holey carbon support film was used to prepare samples for the TEM observation. A powdered sample was dispersed in ethanol and the suspension was treated in an ultrasonic bath for 10 min. Scanning electron micrographs were measured on Tescan Vega instrument. Quantitative estimation of Si/M ratio in the samples was performed using inductively coupled plasma mass spectrometer (Agilent). The samples were digested by acid treatment with H₂SO₄ and HF. The porosity and specific surface area of materials were determined at 77 K by nitrogen adsorption/desorption using Quantachrome NOVA 1200e surface area and pore size analyser. Prior to the experiments, the samples were outgassed at 423 K for 24 h. To confirm reproductivity of measurements and suitability of selected temperature of activation, the control measurements for new batch of the sample and higher temperature of activation (473 K) were performed. Original and control measurements were similar, confirming homogenous dispersion of metal ions in the samples and sufficiency of temperature 423 K selected for degassing. The specific surface area, S_BET, was estimated using the Brunauer-Emmett-Teller (BET) equation in a pressure range 0.05–0.30. Pore size distribution was calculated using the NLDFT method as well as Barrett, Joyner and Halenda (BJH) method. Carbon dioxide and methane adsorption isotherms were measured at 303 K by commercial gravimetric equipment IGA001 (Hiden Isochema) using sample weight 50 mg. Prior the measurements the sample were degassed under vacuum at 250°C for 8 h.

TEM, SEM characterisation and elemental analysis

The periodic structure of the samples was confirmed by HRTEM measurements. The HRTEM micrographs of the M-SBA-15 samples (M=Al, Ti, Zr) (see Fig. 1) show on periodic hexagonal well-organized mesostructures, formed by alternating pores and silica walls. The unit cell parameter of the hexagonal mesostructures, as estimated from HRTEM micrographs, is about 8 nm for SBA-15, 9 nm for Al-SBA-15 and Ti-SBA-15 and 10 nm for Zr-SBA-15. These values are in a good agreement with the values calculated from SAXS and nitrogen adsorption data (see below). The ICP-MS measurements confirmed experimental Si/M molar ratio 23.5 (M=Al), 26.3 (M=Ti) and 26.7 (M=Zr).

Fig. 1:

HRTEM micrographs of the samples SBA-15 (a), Al-SBA15 (b), Ti-SBA15 (c), Zr-SBA15 (d).

The scanning electron micrographs (SEM) of the prepared materials show, that morphology of all prepared samples is quite similar and samples consist of uniformly sized particles of glomerular appearance (see Fig. 2). The average particle size determined from SEM micrographs is about 720×180 nm.

Fig. 2:

SEM images of SBA-15 (a), Al-SBA15 (b), Ti-SBA15 (c), Zr-SBA15 (d).

SAXS characterisation

Figure 3a shows the SAXS patterns of pure siliceous SBA-15 and samples modified with Al³⁺, Ti⁴⁺ and Zr⁴⁺ ions. For all materials a diffraction peaks indexed as (10), (11), and (20) in the hexagonal P6mm symmetry can be observed. It is obvious that (10) diffraction peak shifts to lower q values after modification of SBA-15 silica by metals and the respective q values were observed at 0.0658 Å⁻¹ for SBA-15, 0.0645 Å⁻¹ for Al-SBA-15, 0.0620 Å⁻¹ for Ti-SBA-15 and 0.0591 Å⁻¹ for Zr-SBA-15. These values correspond to the d-spacings with the values d=95.5 Å for SBA-15, d=97.3 Å for Al-SBA-15, d=101.3 Å for Ti-SBA-15 and d=106.3 Å for Zr-SBA-15. The unit cell parameter, calculated from the (10) diffraction peak using the equation a=2*d10/3 was 110 Å for pure SBA-15 silica sample, 112 Å for sample Al-SBA-15, 117 Å for the sample Ti-SBA-15 and 123 Å for the sample Zr-SBA-15. The change in the unit cell parameter reflects the size of the heteroatom incorporated into SBA-15. The different sizes of [MO₄] tetrahedral units, caused by increasing M–O bond lengths in order Si<Al<Ti<Zr, led to the increase of unit cell parameter of hexagonal mesoporous structure. Figure 3b shows change of unit cell parameter with change of ionic radius of the metal ion. It is obvious, that nearly linear increase of unit cell parameter was observed, which confirms, that metal ions were successfully incorporated into the mesoporous structure.

Fig. 3:

(a) SAXS patterns of the SBA-15 silica and the samples M-SBA-15 (M=Al, Ti, Zr). (b) Dependence of unit cell parameter on ionic radii of atom M (M=Si, Al, Ti, Zr).

Nitrogen adsorption/desorption

Nitrogen adsorption/desorption isotherms of studied samples are present in Fig. 4a. The isotherms are of IV type with H1 hysteresis loop according to IUPAC classification [21]. The initial part of the adsorption isotherms corresponds to adsorption in the micropores and formation of the multilayer film on the pore walls. The isotherms shows a well-defined adsorption/desorption step between partial pressures P/P₀ of 0.66–0.78 for the sample SBA-15, P/P₀=0.70–0.83 for the sample Al-SBA-15, P/P₀=0.71–0.85 for the sample Ti-SBA-15 and P/P₀=0.73–0.88 for the sample Zr-SBA-15, indicative of the filling of mesopores. The shift of the mesopore filling step to higher pressures from the sample SBA-15 to the sample Zr-SBA-15 indicates increase of the pore size.

Fig. 4:

(a) Nitrogen adsorption/desorption isotherms, (b) t-plot curves for the studied samples.

Desorption from the samples is delayed in comparison with the adsorption and the evaporation step occurs in the range of P/P₀=0.69–0.55 for the sample SBA-15, P/P₀=0.72–0.60 for the sample Al-SBA-15, P/P₀=0.74–0.60 for the sample Ti-SBA-15 and P/P₀=0.75–0.61 for the sample Zr-SBA-15.

From the adsorption–desorption curves, t-plots (the statistical film thickness) were estimated. Figure 4b shows the t-plots obtained from the nitrogen sorption measurements in the region from t=3, 5–8 Å. The intercept of the fitted t-plots reflects the amount of microporosity in the studied materials. The value of the intercept is lowest for Zr-SBA-15 silica and highest for pure siliceous SBA-15. The results from the t-plot agree very well with micropore surface area and pore size distribution of the samples. The values are summarized in Table 1.

The specific surface area, S_BET, was estimated using the Brunauer-Emmett-Teller (BET) equation in a pressure range 0.05–0.30, pore size distribution was calculated using the NLDFT as well as BJH method. The textural properties are summarized in Table 1. It can be seen, that modification of SBA-15 with metal ions led to the decrease of the surface area from 947 m²/g for SBA-15 to 732 m²/g for Al-SBA-15, 807 m²/g for Ti-SBA-15 and 657 m²/g for Zr-SBA-15. On the other hand, the modification led to the increase of pore size.

In Table 1, the pore sizes of the samples determined by NLFDT and BJH methods are presented. The pore sizes, as determined by both methods, differ by 4–17 Å. It was well documented, that the Kelvin equation based procedures, such as the BJH method, significantly underestimate (~20–30%) the pore size for mesopores with <~10 nm [21] and NLDFT, implemented in commercially available software, provides a reasonably reliable assessment of the pore size distribution for both, micro- and mesopores. Thus, the values in Table 1 determined by NLDFT method should be decisive.

Carbon dioxide and methane adsorption

Single-gas carbon dioxide and methane adsorption isotherms measured at 303 K and up to 20 bar are presented in Fig. 5. The CO₂ adsorption isotherms are virtually Henry-type for all considered samples and thus consistent with moderate gas–solid interactions. At low pressures, below 1 bar, the heteroatoms give only a minor contribution to carbon dioxide adsorption and no significant difference in adsorption between the studied samples was observed. The adsorption capacity at 1 bar was in the range 0.74–0.78 mmol/g. At higher pressures the difference between the samples was observed. The highest carbon dioxide adsorption capacity was observed for the sample Al-SBA-15 (5.91 mmol·g⁻¹), followed by the sample Ti-SBA-15 (5.52 mmol·g⁻¹) and sample Zr-SBA-15 (5.07 mmol·g⁻¹). The lowest adsorption capacity was observed for pure siliceous sample SBA-15 (4.34 mmol·g⁻¹). At the maximal experimental pressure no of the measured isotherms reached plateau, what indicates, that the mesopores are still not filled at the maximal pressures. Moreover, the samples differ in the surface area and we have to have in mind, that the textural properties of the samples play relevant role at the adsorption process. Therefore when we normalized adsorption capacity to surface area, the highest adsorption capacity per square meter shows Zr-SBA-15 silica (8.16×10⁻³ mmol·m⁻²) followed by Al-SBA-15 silica (7.83×10⁻³ mmol·m⁻²), Ti-SBA-15 silica (6.13×10⁻³ mmol·m⁻²) and purely siliceous sample SBA-15 silica (4.46×10⁻³ mmol·m⁻²). These data well documented the influence of metal ions on the adsorption capacity.

Fig. 5:

Carbon dioxide and methane adsorption over studied metal doped samples.

In addition to carbon dioxide, we have also studied the adsorption of methane. Carbon dioxide adsorption capacity is markedly higher than methane and depends on samples composition and textural characteristics. One factor influencing the lower CH₄ adsorption is its zero quadruple moment [22] in comparison to permanent electric quadrupole moment of carbon dioxide. The interaction of the carbon dioxide with the metal centers in the silica structure is stronger, in comparison with methane, leading to the higher adsorption. Moreover, the preferential carbon dioxide adsorption at high pressures can be explained by the fact, that adsorption of carbon dioxide at 303 K is about at its critical temperature (T_C ), while methane is far above. At 303 K carbon dioxide condensation inside adsorbent pores is possible while methane not. Studies describing carbon dioxide multilayer adsorption at temperatures around critical (even slightly higher) and high pressures confirm this hypothesis [23], [24], [25].