Interaction of SBA-15 with Dilute Acetic Acid and Ammonia

Abstract

The SAXS pattern and pore volume of SBA-15 are not affected by aging SBA-15 with water, dilute acetic acid, and ammonia up for to 1 month, while the specific surface area is substantially depressed in interactions with basic solutions. The SEM images indicate pits on the side surfaces of SBA-15 particles in interactions with basic solutions. Aging of SBA-15 in ammonia solutions results in cavities formed by the collapse of walls between neighboring hexagonal channels in the SBA-15. This phenomenon is discussed with a special emphasis on its possible effect on the potentiometric titration curves. Especially, a standard procedure, in which the SSA measured before the titration is used to calculate the σ₀, is compared with a modified procedure, in which the SSAs measured before and after the titration are used to estimate the SSA for each data point separately.

1. Introduction

Nikola Kallay was especially interested in pH-dependent surface charging at the solid liquid interfaces. His excellent work was frequently cited by the senior author of this paper in his compilations of points of zero charge and isoelectric points PZC/IEP of various materials. Niki has studied several atypical systems, including the surfaces of metals [1] and electric double layers in nonaqueous and mixed-solvent systems [2]. He has shown that despite its disadvantages, the mass titration method can still provide useful information about the pH-dependent surface charging [3]. In this present paper, we discuss the changes in the porous structure of SBA-15 caused by the interaction of silica with aqueous solutions.

The surface charge density σ₀ is calculated from the following formula:

σ₀ = −c ΔV F/(m · SSA),

(1)

where c is the concentration of base used in the potentiometric titration; ΔV is the difference in the volume of base needed to bring it to a certain pH: the dispersion on the one hand and the same volume of supporting electrolytes w/o solid particles on the other (acid is counted as the base with a minus sign); F is the Faraday constant; m is the mass of dispersed solid; and SSA is its specific surface area. It is noteworthy that the volume of dispersion does not appear in Equation (1). The SSA is usually determined from the BET equation prior to the titration. The relevance of the SSA determined by nitrogen adsorption at its boiling point to the adsorption of ions from the solution at room temperature can be challenged, but this issue is beyond the scope of the present paper, and most authors use “the BET specific surface area” in their studies of surface charging. Equation (1) neglects the reaction of the base with solution monomers resulting from the dissolution of solid particles, but the contribution of solution monomers is negligible when m and SSA are high enough. Equation (1) implicitly assumes that the SSA does not change in the course of the titration; that is, the SSA determined for dry powder before the titration is relevant for all data points. This can very well be true for certain materials but not necessarily for all materials.

This study is devoted to mesoporous silica. Apparently, the SSA of mesoporous silica is substantially depressed upon contact with aqueous solutions, especially at basic pH [4,5]. These observations suggest that the SSA of mesoporous silica is also depressed in the course of potentiometric titration; thus, different values of the SSA should be used in Equation (1) for particular data points.

The surface charging of SBA-15 was chiefly studied by electrophoresis, and the studies by titration are rare [6,7]. The same SSA for all data points was used; that is, ref. [6,7] ignored the changes in the SSA in the course of the titration. In this study, we further investigate the interaction of SBA-15 with water, dilute acetic acid, and ammonia, and we present titration curves corrected for the depression in the SSA in basic media.

2. Materials and Methods

The chemicals used in the experiments are specified in our previous work [5]. Four lots of SBA-15 were obtained by means of a modified standard method [8].

In all four lots of SBA-15, a solution composed of P123, HCl, and water was slowly titrated with TEOS at 35 °C, and the dispersion was stirred at 35 °C for 1 day. Then, the dispersion was stored at 80 °C for another 1 day. Then, the particles were washed with water and dried at 60 °C in air for 1 day. Finally, they were calcined in air at 500 °C. In lots 1 and 2, heating at 35 °C was followed by heating at 80 °C and then by washing with water w/o delay. In lots 3 and 4, heating at 35 °C was followed by aging at room temperature for 2 weeks, and then heating at 80 °C was followed by aging at room temperature for the next 2 weeks before washing with water as described in more detail in [9]. Four separate lots were obtained, but only two different recipes were used. The details of these recipes are summarized in Table 1. Using the procedure described in our previous study [5], the specimens of the SBA-15 from lot 2, 0.5 g each, were aged for 28 d in the following:

• water;
• acetic acid solution, 10 μL of glacial acetic acid per 100 mL of water;
• ammonia solution, 10 μL of 25% ammonia per 100 mL of water.

The supernatant solution after seven selected aging times was analyzed for [Si]. A molybdenum blue method was used to study the Si concentration in the solution [10,11]. Ammonium molybdate produces a yellow silicomolybdate complex when added to an acidified solution containing silicic acid monomers. The yellow complex was reduced to molybdenum blue, and its concentration was determined spectrophotometrically. The chemistry of silicomolybdate complexes is complicated, and it was discussed in detail in [10]. The solid (SBA-15) was analyzed for SEM, SSA, and SAXS according to the procedures described in our previous study [5].

Table 1. The preparation and SSA of SBA-15.

Specimen	Mass of P123	Aging After Heating at 35 °C	Aging After Heating at 80 °C	SSA/m2/g
	g	Days	Days	Before Titration	After Titration
1	40	0	0	964	-
2	40	0	0	969	-
3	2	14	14	957	620
4	2	14	14	1151	485

Table 1. The preparation and SSA of SBA-15.

Specimen	Mass of P123	Aging After Heating at 35 °C	Aging After Heating at 80 °C	SSA/m2/g
	g	Days	Days	Before Titration	After Titration
1	40	0	0	964	-
2	40	0	0	969	-
3	2	14	14	957	620
4	2	14	14	1151	485

Specimens 3 and 4 were analyzed for electrokinetic potential and for surface charge density at 25 °C in 10⁻³ M NaCl. The electrokinetic measurements [12,13,14,15] were performed by means of a Malvern Zetasizer. The electrophoretic mobility was converted into ζ potential by means of a Smoluchowski equation. The potentiometric titration was performed in a nitrogen atmosphere to avoid atmospheric CO₂ [13,14,16,17,18]. After titration, the SBA-15 was quickly washed with pure water to remove the base and NaCl, dried at 80 °C in air for 1 day, and analyzed for SSA.

3. Results and Discussion

The effect of aging of SBA-15, specimens 1 and 2, in water (pH 5.5 and 6), dilute acetic acid (pH 3, 3.3, and 3.4), and dilute ammonia (pH 7.8, 8, and 9) on the concentration of silicon in solution is presented in Figure 1. Caution: several authors report [SiO₂] rather than [Si]. The results obtained at various pHs and with two specimens of SBA-15 are relatively consistent. In all specimens but one, the [Si] is about 100 ppm for sufficient equilibration time. There is one outstanding data point (pH 8, 28 days), which suggests a substantially lower [Si]. Unlike the [Si] after long equilibration, which was rather insensitive to the pH, the kinetics of dissolution was strongly pH-dependent. At the basic pH, the final [Si] was reached within 1 d, and further aging had a rather insignificant effect on [Si]. At an acidic pH, a systematic increase in [Si] was observed at least for the first week of aging.

Silica occurs mainly as an orthosilicic acid in its dilute aqueous solutions over a broad pH range, and anionic monomeric species prevail at pH > 9. Monomeric species at various degree of hydration were discussed in the literature. The aqueous chemistry of silica at concentrations >10⁻³ M is more complicated, and it involves species with various degrees of polymerization, hydration, and deprotonation. The aqueous chemistry of silica was discussed in detail in [10].

The effect of aging of SBA-15, specimens 1 and 2 in water (pH 5.5 and 6), dilute acetic acid (pH 3, 3.3, and 3.4) and dilute ammonia (pH 7.8, 8 and 9) on SSA is presented in Figure 2. The results obtained at various pH are very different. The specific surface area is rather insensitive to aging in acidic media, and aging in basic media (pH 7.8–9) depresses the SSA by a factor of 2. The behavior of SBA-15 in pH-neutral media is half-way between the behavior in acidic media on the one hand and basic media on the other. The SSA in basic media (pH 8 and 9) dropped by a factor 2 within the first day, and further aging had a rather insignificant effect on SSA. The SSA of specimens 3 and 4 before and after titration is presented in Table 1.

Table 1 indicates that the SSA of both specimens was substantially depressed in course of titration, which took about 3 h with specimen 3 and 5 h with specimen 4 (substantially shorter times than the times in Figure 2), and the pH increased from 4 to 8. The SSA of specimen 4 dropped more substantially than the SSA of specimen 3 as a result of titration. This is because the titration of specimen 4 was terminated at higher pH, and the contact between the SBA-15 and basic solution was longer.

The effect of aging on the pore volume (Figure 3) was insignificant irrespective of the pH, but the micropore volume (Figure 4) was substantially depressed in basic media. The micropore volume dropped within the first day in basic media, and further aging had rather insignificant effect on micropore volume.

The average pore diameter of SBA-15 (Figure 5) was rather insensitive to aging in the acidic pH range, and it increased by a factor of 2 in the basic pH range. The pore diameters presented in Figure 5 are a result of the depression in the SSA at constant pore volume.

The depression in SSA when the pore volume remains constant can be explained by cavities, which are formed as a result of the collapse of walls between neighboring hexagonal channels in the SBA-15 structure [5]. A cavity has a substantially lower surface area than the original channels, but it has the same volume as the sum of volumes of the original channels. The cavities are narrow enough to support the capillary condensation. This model was discussed in more detail elsewhere [5]. Similar phenomena (depression in SSA with constant pore volume) were observed by the others [4,19,20], although these authors studied aging of SBA-15 at different experimental conditions. Micropores in SBA-15 are narrow channels interconnecting the wider parallel channels (mesopores) [21]. The presence of micropores weakens the mechanical strength of the walls between parallel channels (mesopores), and these weaker walls are preferentially destroyed, rather than stronger walls w/o openings (micropores). This explains simultaneous decrease in SSA (Figure 2) and decrease in the volume of micropores (Figure 4) in certain specimens.

Figure 6 presents the surface charge density of SBA-15 calculated from Equation (1). The “uncorrected” curves were calculated from the original SSA of SBA-15 (measured before the titration). The corrected curves were calculated from the apparent SSA of SBA-15 depressed on contact with solution. The original SSA of SBA-15 from Table 1 was used for the first data point. The final SSA of SBA-15 from Table 1 was used for the last data point. For the other data points, the SSA was assumed to change proportionally to time t, that is,

SSA (t) = SSA (0) − [SSA (0) − SSA (final)] × t/t (total)

(2)

where t refers to the time elapsed between the first data point and the data point of interest. We do realize that Equation (2) is only a rough estimation, but direct measurement of the SSA for all data points in the course of a single titration is impossible. Perhaps verification of Equation (2) can be an object of a future study.

The shapes of the charging curves in Figure 6 and the values of σ₀ are similar to those obtained with other silicas in 10⁻³ M NaCl. The surface charging of silica is cation-sensitive (e.g., Na vs. K). The charging curves in Figure 6 are not useful for the determination of the PZC of silica because the σ₀ at a pH below 5 is nearly zero and almost pH-independent. The availability of charging curves of SBA-15 is limited, and we could not find titration data with 10⁻³ M NaCl in the literature. Salis et al. [6] observed a σ₀ of −0.04 C/m² in 0.05 M of NaCl. This figure is consistent with our results because the absolute value of σ₀ increases with the salt concentration. The σ₀ of SBA-15 is also reported in another paper from the same research group [7], and, most likely, the same results are reported there. Barczak et al. [22] studied the surface charging in 10⁻³ M NaCl. Unfortunately, they only report the results with SBA-15 modified with silanes, but not for the original SBA-15.

The uncorrected σ₀ of specimen 3 in Figure 6 is higher in absolute value than the uncorrected σ₀ of specimen 4. In contrast, the corrected σ₀ of both specimens are consistent. Apparently, our simplified model of the variation in SSA in the course of titration (Equation (2)) is quite successful.

The electrokinetic potentials of SBA-15 (specimens 3 and 4) are presented in Figure 7. These results are similar to our previous results [9] and to the results of electrokinetic studies of SBA-15 reported by the others. Apparently, the IEP falls at a pH of about 3, and this IEP is consistent with the IEP of SBA-15 reported elsewhere and with the IEP of other silicas.

The SEM images of the original SBA-15 and of the material from the same lot treated with dilute ammonia are presented in Figure 8 and Figure 9. The external surface of the particles is flat in the original material, and cavities are observed in the treated material. The cavities look alike in all particles of the treated material, and they are uniformly distributed over their surfaces. The microscopic images corroborate the aforementioned model explaining the depression in SSA at constant pore volume.

The cavities visible in Figure 9 (but not in Figure 8) are about 20 nm in diameter. The image in Figure 8 is representative of the original specimen and the samples aged in acetic acid solutions, and Figure 9 is representative of the samples aged in ammonia solutions (only a few selected samples were studied via SEM). An additional SEM image in a higher magnification is presented in the Appendix A (Figure A1).

The SAXS patterns of the original SBA-15 and the material after aging for various times in various solutions can be analyzed to detect minor changes in the interplanar spaces in the hexagonal structure of SBA-15. The positions of peaks in the SAXS pattern of SBA-15, specimen 2, aged at pH 3.4, 6, and 7.8 for various times are plotted in Figure 10. Complete SAXS data can be found in the Appendix A. The peaks are at almost constant values of 2 θ irrespective of the pH of the solution and the time of aging. They are also consistent with the peaks in the original SBA-15, specimen 2 (0.917°, 1.603°, and 1.855°, respectively), and with the peaks in specimen 1, original and aged [5]. This result indicates that the interplanar spacings are not affected by aging in the solutions studied in this paper (unless a wall between neighboring channels is destroyed). The SAXS patterns presented in Figure 10 further corroborate the model explaining the depression in the SSA with constant pore volume presented above. In other words, locally the structure of aged SBA-15 is unchanged, and this leads to an unchanged SAXS pattern, and in the cavities, the distance between the walls is too high to produce reflexes in the accessible range of 2θ (the intensities below 0.5° are very high and very irreproducible).

4. Conclusions

• The diameters of channels in the structure of SBA-15 are rather insensitive to aging in pH neutral, weakly basic, and weakly acidic solutions.
• The walls between neighboring channels break in weakly basic solutions.
• The walls with micropores are easier to break.
• Depression in SSA must be taken into account in interpretation of the surface-charging curves of SBA-15.