Evaluation of the SBA-15 materials ability to accumulation of 4-chlorophenol on carbon paste electrode

The 4-chlorophenol (2,4-D, ≥99 %), triblock copolymer Pluronic P123 (M_n ~ 5800), tetraethoxysilane (TEOS, 98 %), aminopropyltriethoxysilane (APTES, ≥98 %) and graphite powder were from Sigma-Aldrich (St. Louis, USA). The ethanol (96 %), hydrochloric acid (35 %), sodium sulfate (≥99 %), potassium hexacyanoferrate(III) and potassium nitrate were obtained from Avantor Performance Materials (Gliwice, Poland). Three mesoporous silica materials were used in this studies - commercially available mesoporous silica received from Sigma-Aldrich (SBA-15-Sigma), and the two silicas (unmodified SBA-15 and amino-functionalized SBA-15-NH₂) synthesized according to the procedure described in Sect. 2.2. Stock standard solutions of 4-CP (0.5 mmol/dm³) were prepared in 0.1 mol/dm³ sodium sulfate for electrochemical studies and in distilled water for adsorption studies. The working solutions were prepared by dilution as needed.

The porous structure characteristics were obtained on the basis of low temperature nitrogen adsorption–desorption isotherms (Micromeritics ASAP 2020 volumetric adsorption analyser, Norcross, USA). The X-ray diffraction measurements were performed using an Empyrean X-ray diffractometer from PANalytical 2014 over the 2θ range from 0.3° to 5° with step size of 0.01°. Attenuated Total Reflectance (ATR-FTIR) spectra were recorded using an IR spectrometer TENSOR 27 (Brucker, 2014, Germany) equipped with diamond crystal. The spectra were recorded in the spectral range of 400–4000 cm⁻¹. Thermal analysis was carried out on a TG–DTA/DSC apparatus (STA 449 Jupiter F1, Netzsch, Germany). The samples were heated from 4 to 900 °C in Al₂O₃ crucibles under oxidative (synthetic air) atmosphere with flow rate 25 mL min⁻¹. The heating rate was 10 °C·min⁻¹. The analysis of structure ordering of the investigated samples was conducted by electron microscopy in high resolution transmission electron microscope S/TEM Titan3 G2 60-300 (FEI Company). The Elemental CHN analysis was performed by Perkin Elmer 2400 Elemental Analyzer (1997). Cary 3E UV–Vis spectrophotometer (Varian Inc., USA) was used for absorbance measurement. Cyclic voltammetry (CV) and DPV were performed in an analytical system (Autolab, Eco Chemie, Netherlands), potentiostat/galvanostat connected to a three electrode cell, linked with a computer and with software (Eco Chemie). The system was run on a PC using GPES 4.9 software. A conventional three-electrode cell assembly consisting of a platinum wire as an auxiliary electrode and the saturated calomel electrode as a reference electrode was used. The working electrode was either an unmodified carbon paste electrode or a carbon paste electrode modified with mesoporous material (SBA-15-Sigma, SBA-15 or SBA-15-NH₂).

Pure SBA-15 support was prepared by sol–gel technique (Zhao et al. 1998). In a typical preparation, 8.0 g of triblock copolymer Pluronic P123 as a structure directing agent was dissolved in 240 cm³ of HCl (2 mol/L) and 60 cm³ of H₂O. Mixture was stirred at 40 °C for 3 h. Afterwards 18.32 cm³ of tetraethoxysilane (TEOS) was added dropwise. The solution was stirred for 24 h at the same temperature. The precipitated mixture was aged at 80 °C for 48 h without stirring and next the solid product was washed with distilled water and organic components were removed by calcination at 550 °C for 4 h in muffle furnace under an air atmosphere.

The amino-functionalized SBA-15 was prepared by post-synthesis basic treatment by aminopropyltriethoxysilane (Wang et al. 2007). For post-synthesis modification. 2.0 g of as-synthesized SBA-15 powder without organic template was added into a mixed solution of APTES and ethanol (50 cm³ ethanol (96 %) and 3 cm³ of APTES). After stirring at 25 °C for 6 h, the powder (denoted as SBA-15-NH₂) was filtered, washed with distilled and gently dried at 35 °C for 24 h.

The adsorption experiments were carried out in Erlenmeyer flasks containing 0.15 g of silica materials and 10 cm³ of the solutions of a known concentration of 4-CP (from 50 to 500 μmol/dm³). The flasks were agitated for 6 h at a constant speed of 200 rpm at 25 °C. At the end of this period, the mixtures were filtered through a 0.2-μm pore size filter and analyzed by using UV–Vis spectrophotometry at 274 nm. The equilibrium amount of 4-CP adsorbed per unit mass of SBA-15, q_e (μmol/g), was calculated by the equation:where C ₀ and C _e are the initial and the equilibrium concentration of 4-chlorophenol (μmol/dm³), V is the volume of the solution (dm³) and m _a is the mass of the adsorbent (g). All the experiments were carried out in duplicate, and average values were used for further calculations.

A Teflon holder with a hole at one end (2.5 mm diameter and 3.5 mm deep) for the carbon paste filling served as the electrode body. Electrical contact was made with a stainless steel rod through the center of the holder. This rod can move up and down by screw movement to press the paste down when renewal of the electrode surface is needed. Modified CPEs were prepared by thoroughly mixing 5 or 10 mg of mesoporous material (SBA-15-Sigma, SBA-15 or SBA-15- NH2) with, respectively, 95 or 90 mg of graphite powder (diameter >45 µm) and, subsequently, adding 15 mg of mineral oil (13 % m/m). Homogenization was then achieved by careful mixing using agate pestle and mortar and afterwards rubbed by intensive pressing with the pestle. The mixture was kept at room temperature for 2 days. The ready prepared paste was then packed into the hole of the electrode body and the carbon paste was smoothed onto a paper until it had a shiny appearance. Unmodified carbon paste electrode was prepared in a similar manner, with the addition of only 100 mg of the same graphite powder and 15 mg of mineral oil (Nujol).

The physical properties of mesoporous silica materials were characterized using low temperature nitrogen adsorption–desorption isotherm, small-angle X-ray powder diffraction, ATR-FTIR infrared spectroscopy, thermogravimetry, SEM and TEM techniques.

Nitrogen adsorption/desorption isotherms for all silica mesoporous materials were determined at 77 K. Figure 1 shows the nitrogen adsorption–desorption isotherms and functions of pore size distribution (PSD) (Barrett, Joyner and Halenda (BJH) procedure 1951) obtained for the investigated samples. The shape of the sorption isotherms are typical for SBA-15 materials with ordered mesoporosity. All samples yield IV type of isotherm with H1 hysteresis loop in the p/p₀ range 0.55 ~ 0.75. This well-defined hysteresis loop of H1 type rendering to the IUPAC classification (Gregg and Sing 1982) shows parallel branches of adsorption and desorption. This type of hysteresis loops is related with capillary condensation of nitrogen in similar, uniformly ordered and opened cylindrical mesopores. The presence of such structures was confirmed by BJH pore size calculations. As a result narrow functions characterized by a maximum at 50 Å (SBA-15-NH2), 55 Å (SBA-15) and 70 Å (SBA-15-Sigma) what suggest presence of comparable mesopores, were obtained. After grafting of the aminopropyltriethoxysilane, the PSD decreased by 5 Å. The BET surface area as well as the total pore volume decreased for sample after modification. Consequently, the modification leads to a decrease of BET surface area by 210 m²/g and 157 m²/g for SBA-15-Sigma and SBA-15 respectively. Similarly, the total pore volume decreased by 0.26 cm³/g and 0.15 cm³/g. A decrease of textural parameters was observed as an effect of attaching the amino groups and agrees with the other literature references (Ziarani et al. 2012). The values of specific surface areas (S _BET ) of the materials, as well as the total pore (V _t ) and micropore (V _mic ) volumes calculated from the adsorption isotherms are presented in Table 1. In Table the mesopore diameter (w) calculated from adsorption branch of nitrogen isotherm is also given.

The ordering of porous structure of SBA-15 materials before and after modification by NH₂ groups were evaluated by X-ray diffraction measurements in range of small angles 2θ. The XRD patterns of SBA-15 materials are displayed in Fig. 2. In all cases the SBA-15 supports exhibit structural ordering according to hexagonal symmetry p6 mm characteristic for this type of materials. Three first and well-resolved reflections registered at small diffraction angles can be indexed as (100), (110) and (200) reflections of the hexagonal lattice and still observed after post synthesis modification. It can be observed that the three first reflections of SBA-15-Sigma sample (commercial SBA-15) are shifted slightly to lower angles. This indicates that the dimension of unit cell becomes higher than as-prepared mesoporous ordered silica. The unit cell parameters calculated from these three peaks were 10.1 nm, 10.3 nm and 11.0 nm in the case of SBA-15, SBA-15-NH₂ and SBA-15-Sigma, respectively.

The FTIR spectra of all silica materials are presented in Fig. 3. The ATR-FTIR spectra of investigated samples display the typical bands of mesoporous silica materials. In all cases the absorption peaks at wavenumbers of 1042, 800 and 445 cm⁻¹ correspond to the vibrations of Si–O-Si (SiO₄) (more specifically, Si–O–Si stretching vibrations, Si–O bending vibrations and Si–O out of plane deformation vibrations respectively). Moreover, on the IR spectrum of investigated samples small but characteristic bands appear at about 3745 cm⁻¹ which are assigned to isolated silanol groups on silica surface (Morrow et al. 1976). In general after modification of mesoporous silica by amine groups, reduction of the peak intensity at 1042 cm⁻¹ of Si–O–Si group and at 970 cm⁻¹ of stretching Si–OH group as well as the presence of typically hardly discernible bands of NH group at about 1510 cm⁻¹ (Chong and Zhao 2003) for SBA-15-NH₂ should indicate the successful modification of the silica surface. In the obtained ATR–FTIR spectra the intensity of the stretching peak of Si–O–Si as well as Si–O bending vibrations and stretching Si–OH groups of functionalized SBA-15 decreased significantly which correlates well with the mechanism of its reaction with aminopropyltrimethoxysilane. Unfortunately, in this case an additional peak directly related to the amine groups (located at about 1542 cm⁻¹ and describing the bending vibrations of –NH groups) is not clearly observed. In other words, some changes of the baseline were observed for modified sample, which may suggested presence of –NH bending vibrations, but the presence of a new, well-formed signal was far from obvious. It may be explained by the fact that the unequivocal characterization of this type of materials modified by amine groups using FTIR spectrometry sometimes is not so simple. Very often, these signals are very weak, even at significant content of surface localized functional groups. Regrettably, sometimes ATR-FTIR technique may be not sensitive enough to observe well-defined signals from modifiers.

Elemental analysis of carbon, hydrogen and nitrogen was performed to characterize the elemental composition of SBA-15-NH2 sample. Mass percentage of CHN elements was determined for this sample as 2.83, 1.06 and 1.14 %, respectively.

Figure 4 shows the results of thermogravimetric analysis of unmodified and amine functionalized SBA-15 silicas as the mass loss curves and the differential thermal analysis curves. The TGA and DTA analyses of investigated samples show no significant changes in the case of unmodified mesoporous silica (SBA-15-Sigma and SBA-15). SBA-15-Sigma and SBA-15 show a small weight loss whereas the SBA-15-NH₂ display significantly larger weight losses due to the incorporation of amine groups.

The observed total weight loss for SBA-15-Sigma and SBA-15 are small (4.22 and 1.99 % for SBA-15-Sigma and SBA-15 respectively). In this case the total weight loss can be associated with water molecules (physisorbed water and molecules eliminated by condensation of surface silanols). The observed total weight loss for SBA-15-NH₂ sample was significantly higher (6.9 %) and is related to a sum of physisorbed water and grafted ammine groups. All samples were compared in Fig. 4d.

The SEM and TEM images (Fig. 5) show that the prepared SBA-15 as well as SBA-15-NH₂ samples exhibit a well-ordered mesoporous channels arranged in 2D hexagonal structure according p6 mm symmetry. This structural ordering of mesopores is visible in the two directions to pore axis (parallel in Fig. 5b, f and perpendicular in Fig. 5c, d). The distances between two cylindrical mesopores in SBA-15 sample (Fig. 5b) estimated from the TEM images are about 10.2 nm. The diameter of mesopores and thickness of walls are about 6.2 and 4.0 nm respectively. In the case of SBA-15-NH₂ sample the distances between two mesopores is greater (10.5 nm) and the diameter and thickness of walls are identified as 7.3 and 3.2 nm respectively.

The experimental adsorption isotherms of 4-CP on the SBA-15-Sigma, SBA-15 and SBA-15-NH₂ are presented in Fig. 6. For their analysis the Generalized Langmuir (GL) isotherm equation (Marczewski and Jaroniec 1983) derived in terms of the general theory of adsorption on energetically heterogeneous solids was applied:

where q _e is the equilibrium adsorbed amount, q _m is the adsorption capacity, the heterogeneity parameters m and n characterize the shape (width) and asymmetry of adsorption energy distribution function (\( m,n \in (\left. {0,1} \right\rangle \)), and the equilibrium constant, K, describes the position of distribution function on energy axis. The GL equation is reduced to simpler isotherms for specific vales of heterogeneity parameters, e.g., for m = n it becomes the well-known Langmuir–Freundlich isotherm corresponding to the symmetrical energy distribution, and for m = n = 1 it is Langmuir isotherm.

The Generalized Langmuir equation was fitted to experimental data and in the case of all adsorption systems the simplest Langmuir model was found to be the best one (m = n = 1). The obtained isotherm parameters and the coefficients of determination R ² for the adsorption of the 4-CP onto the mesoporous silica materials are presented in Table 2. The high R ² values indicate that the equilibrium data obtained for all adsorption systems are well described by the Langmuir equation. Though typically adsorption isotherms of organics on porous solids show at least some effects of energetic heterogeneity, however, in our case due to low range of concentration, the adsorption corresponds to the high energy sites only and may be described by the Langmuir isotherm. Such a result may be also explained by a possible compensation (or even cancelling-out) of the effects of lateral interactions in adsorption space and the system nonhomogeneity (Marczewski et al. 1986, 2013). Moreover, it may be the additional effect of well-organized structure of adsorbent which diminishes the energetic and structural heterogeneity of adsorption system. The estimated adsorption capacities (q _m ) decrease in order: SBA-15-NH₂ > SBA-15-Sigma > SBA-15. The q _m values for 4-CP adsorption on unmodified SBA-15 and SBA-15-Sigma were lower in comparison to the adsorption capacity obtained for modified SBA-15-NH₂. The commercial SBA-15-Sigma was found to be slightly better adsorbent than SBA-15 which is probably due to its the greater specific surface area (517 vs. 464 m²/g). The higher adsorption capacity of SBA-15-NH₂ in comparison to the unmodified silicas is a result of a more favorable interaction between 4-CP and the adsorbent due to the presence of a free electron pair and the more alkaline properties of the mesoporous silica surface. These results are consistent with the results of other authors (Anbia and Amirmahmoodi 2011; Toufaily et al. 2013; Moritz 2015) who also found that amino-functionalized SBA-15 showed higher adsorption capacities for 4-CP than the untreated SBA-15.

The adsorption of 4-CP on mesoporous SBA-15 was previously studied by Anbia and Amirmahmoodi (2011), Toufaily et al. (2013) and Moritz (2015) but the obtained results vary significantly. The adsorption isotherms of 4-CP were better described by the Freundlich equation (Anbia and Amirmahmoodi 2011) and sometimes by the Langmuir equation (Toufaily et al. 2013; Moritz 2015). The Langmuir maximum adsorption capacity values obtained for 4-CP adsorption on unmodified SBA-15 were 57.5 mg/g (Anbia and Amirmahmoodi 2011), 0.3 mg/g (Toufaily et al. 2013) and 9.6 mg/g (Moritz 2015), respectively. These differences are very large, although the specific surface areas of the materials tested are comparable (812 m²/g (Anbia and Amirmahmoodi 2011), 678 m²/g (Toufaily et al. 2013), 828 m²/g (Moritz 2015)). Recently, Qin et al. (2012) described adsorption of 2-chlorophenol, 2,6-dichlorophenol and 2,4,6-trichlorophenol on mesoporous SBA-15 (S_BET = 744 m²/g). The q _m value for 2-CP adsorption was found to be 2.6 mg/g. The adsorption of 4-chlorophenol was not tested but, based on the other studies of the adsorption of monochlorophenols on the carbonaceous materials (Jung et al. 2001; Hamdaoui and Naffrechoux 2007; Liu et al. 2010; Kusmierek et al. 2014) the adsorption properties of 4-CP and 2-CP are very similar.

All this observations show that the mesoporous SBA-15 materials reveal varying adsorption capacity towards 4-chlorophenol. In addition, despite the high specific surface areas their adsorption capacity, compared to the activated carbon, is often lower for this and other chlorophenols. For comparison, the adsorption capacity of activated carbon (S_BET = 925 m²/g) for 4-CP was 255 mg/g (Kusmierek et al. 2015). Of course, for other adsorbates situation may be different (heavy metals for example).