Direct one-step synthesis of mesoporous ZnO-silicate matrix using a true liquid crystal method

2.1 Characterisation methods

A D8 single sample input was used for X-ray diffraction (XRD), having a copper target (k _α = 1.5418Å) 40 mA, and 40 kV, step size 0.01°, 659.00 steps at 0.50 s a step used for low angle scanning, and 7700.00 steps at 0.20 s a step for wide angle scanning, 0.05° diversion slits and 0.50° anti scattering slit. Micrometrics 2010 instrument using N₂ analysis gas and He back fill gas for the sorption test. BJH adsorption branches used to give the pore sizes. For SEM/EDX analysis, the sample was coated with carbon film prior to insertion into a Philips XL 30 ESEM instrument. EDX/EDS maps were collected using a TESCAN MIRA3 FEG-SEM fitted with an Oxford Instruments Aztec energy X-maxN 80 EDX detector. An Agilent Cary 5000 UV-Vis-NIR spectrophotometer was used to perform the UV-Vis spectroscopy, absorption 200–800 nm, scan rate of 600.000 nm min^–1, double beam mode, and baseline correction was switched on. TEM images were attained using a JEOL-JEM 3011 electron microscope fitted with a LaB6 filament and operated at 250 kV. The images were recorded using a Gatan 794 CCD camera. The JEOL-JEM 3011 TEM is equipped with a PGT Prism Digital Spectrometer EDX system. Thermo ESCALAB 250Xi spectrometer with a monochromator and an (1486.6 eV) Al-Kα source was used to perform the XPS analysis. Avantage data system was used to record the spectra. The C1 s binding energy (284.6 eV) of adventitious carbon was used as a reference line.

2.2 Synthesis

Mesoporous zinc oxide silicate was prepared using a tailored variation of Attard’s true liquid crystal templating method (TLCT) [29]. Firstly, zinc acetate (0.20 g) was dissolved in ethanol (3 cm³) at 58°C for 30 minutes under N₂, then cooled to 40°C. Cetyltrimethylammonium bromide (CTAB) (1.25 g, 0.0011 mol) was mixed in an acidic water solution (1.25 cm³, pH 2) by the addition of few drops of HCl and methanol (2.5 cm³) in a different round bottom flask. The mesophase was established from the CTAB and the solvent by heating at 40°C under reduced pressure by a rotary evaporator. Si(OCH₃)₄ (2.36 cm³, 0.016 mol) and the ZnO/ethanol precursor were then introduced to the slightly disordered mesophase using a rotary evaporator until the mesophase was re-constructed. The hydrolyzed precursors and the mesophase were left for 2 days at room temperature (aging), and then heated to 290 °C for 1 h under humid airflow (to convert zinc acetate to zinc oxide) then calcined to 600 °C in dry flowing air for 6 h at 3°C a minute heating rate. Reagents; zinc acetate dihydrate 98% (Aldrich), ethanol (Fisher Scientific), Cetyltrimethylammonium bromide (Aldrich), and methanol (Aldrich), Si(OCH₃)₄ 98% (Aldrich), all reagents used without further purifications.

3.1 Low angle XRD

Substitution of 6% wt/wt ZnO/SiO₂, which is equivalent to 11.4% Zn/Si atomic percent, was successfully reached by employing this method. Higher doping amount of ZnO to silicate adversely affects the mesoporous morphology leading to a structural collapse (nonporous structure). A low angle XRD pattern of the calcined 6% wt ZnO-SiO₂ sample (Fig. 1) shows a clear (100) Bragg’s peak indicative of mesoporous packing. However, the broadness of this peak may indicate the presence of a wider pore distribution than the typical mesoporous silicate pore distribution produced by the original TLCT method [29]. The other diffraction peaks shown in Fig. 1(a) were not easily determined due to their low intensity indicating that the pores did not have a long range order; sponge-like holes [28] type are present. Figure 1(b) represents a higher substitution of ZnO-SiO₂, 10–15% wt/wt, as can be observed, destroys the mesoporous structure completely. Therefore, the 6% wt/wt ZnO/SiO₂ substitution is considered the maximum limits of this procedure.

OPEN IN VIEWER

Fig.1

Low-angle XRD patterns of the calcined mesoporous product following 6% ZnO/SiO₂ (a), and 10% ZnO-SiO₂ substitution (b).

Energy dispersive X-ray spectroscopy (EDX) analysis was performed on the calcined 6% wt/wt ZnO-SiO₂ sample (the sample introduced in the solid-state) to determine the atomic and compound percentages present in the final product. The average percentage of ZnO found is 5.9%, based on the analysis of three different locations of the sample, Table 1.

The EDX spectra, Fig. 2, clearly show all the K- and L- edge peaks produced by the elements present in the sample; Zn, O and Si. Only carbon peaks were omitted because they originate from coating the sample with carbon film prior the EDX analysis. EDX spectra was collected at different sample locations as indicated by the SEM images. The corresponding quantifications of the elements and the compounds are presented in Table 1. The distribution of Zn within a SiO₂ aggregate was confirmed by the EDS maps as presented in Fig. 3. This EDS maps clearly show that Zn atoms are spread in the matrix but not as uniform as O or Si atoms (bearing in mind that the quantities of Zn atoms are less than O and Si, Table 1).

OPEN IN VIEWER

Fig.2

SEM/EDX performed at three different locations showing the K and L-edge peaks representing all detected elements in the 6% ZnO-SiO₂ sample.

OPEN IN VIEWER

Fig.3

Si, O and Zn EDS maps of a 6% ZnO-SiO₂ aggregate.

To determine the location of the ZnO particles further characterization were performed. The first employed was a sorption test, performed using nitrogen at 77 °K as the adsorbate, to confirm whether ZnO forms in the framework and to conclude if the material is porous. The isotherm, shown in Fig. 4, produced by the 6% ZnO-SiO₂ sample is a type IV isotherm typical of mesoporous materials [35]. This isotherm shows a region of mismatch (H1 type hysteresis loop), highlighted as ‘C’. This region between adsorption and desorption cycles (at relative pressures 0.4–0.6) indicates the existence of variety of pore volumes/sizes (inner capillary region) [35], which is consistent with the interpretation of the low-angle XRD pattern in Fig. 1a. In addition, the BJH pore distribution (Fig. 4 inset) peak-maximum confirms this observation: three maxima pore peaks are present, the first peak at 22Å, which is a typical pore diameter of mesoporous silicate produced by the TLCT method [29], the second and the third maxima at the new pore sizes of 28 and 38Å, likely due to the contribution of ZnO particles substituting the framework silicate (Zn-O bond length is larger than Si-O one). Furthermore, the sample possesses a large Brunauer-Emmett-Teller (BET) surface area of 530 m²g^–1, which is the highest surface area reported so far for this type of material. Considering, the highest surface area reported for porous ZnO siloxide matrix was 110 m²g^–1 [26]. This high surface area and large pore size is ideal for sorption and catalysis reactions.

OPEN IN VIEWER

Fig.4

N₂ sorption isotherm of 6% ZnO-SiO₂ synthesized in this work and inset the BJH pore size distribution in angstrom.

The UV/VIS spectrum of the solid-state sample, 6% ZnO-SiO₂, is shown in Fig. 5a. Two main zones of peaks are detected, the first at 300–360 nm (highlighted by no.1) which represents ZnO large particles of sizes ∼1.0–2.0 nm [2, 18]. These large particles likely represent non-framework clusters, Tkachenko et al. [31], which exist on the external surfaces of the mesoporous structure or inside its pores. The second zone of peaks (highlighted by no. 2) in Fig. 5a, observed at 280–300 nm likely represent ZnO clusters that are less than 1.0 nm [23, 36] or an interaction between the silicate framework with ZnO species likely as a result of the ZnO substitution [37]. To compare UV/VIS on pure mesoporous silica, containing no ZnO particles was used as a reference (Fig. 5b). It is noted that the UV region at 200–400 nm is almost flat indicating no ZnO particles are present and the silica does not interact with the UV energy. The drop in absorbance at 350 nm is indicates a switch between the visible and ultra violet source.

OPEN IN VIEWER

Fig.5

UV/VIS spectra of a) the ZnO/ Silica framework and b) mesoporous silica.

Further characterisation of the sample was performed using wide-angle XRD (Fig. 6). The XRD patterns indicate that it is not possible to clearly distinguish between the diffraction plane peaks of the ZnO particles from the noise created in the base line from the amorphous nature of the mesoporous silicon dioxide. This may indicate that the existence of very small ZnO particles and they did not exist at equal repeated distances from each other to be detected by wide-angle XRD. The expected diffraction ZnO planes locations are indicated by the arrows in Fig. 6.

OPEN IN VIEWER

Fig.6

Wide-angle XRD pattern of the 6% wt/wt ZnO/SiO₂ sample prepared in this work.

TEM (Fig. 7) was performed to further investigate this sample. Bright field images in a, b and c show the porous nature of the sample, confirming that the porosity is well distributed even after performing this test 10 years after the synthesis. This confirms that the material is durable and stable (long shelf life stability). The measured pore sizes fall between 2 to 4 nm (similar to the pore size distribution obtained from N₂ sorption results). Moreover these images did not show long range order of pores as predicted by the low-angle XRD. The dark field TEM image of the 6% wt/wt ZnO-SiO₂ sample in Fig. 7d shows only a few isolated large ZnO particles exist in the sample (indicated by red rectangle). Their 1.7 nm sizes are likely responsible for the UV peak at 300–360 nm [38]. The remaining particles vary in size within the range of 0.3–0.8 nm and are not uniformly distributed in the sample (confirm by EDS in Fig. 3). These ones likely have interacted with, and integrated into the framework by forming bonds with silicates during the heat treatment. This is deduced by the fact dark field TEM images show a combination of orientation contrast and material z-contrast so crystalline materials in this case the ZnO species will likely show a brighter contrast whereas amorphous silicates appear dark.

OPEN IN VIEWER

Fig.7

TEM images of mesoporous 6% wt/wt ZnO-SiO₂ (a, b, and c) are the bright field images (reveling the pores distribution), and (d) is the dark field image shows the ZnO particles distribution (bright spots); those outlined by the rectangle are the largest ZnO particles (of 1.7 nm) exist in the sample.

The oxidation states of Si and Zn as well as the bonds present in the samples were investigated using XPS. The Si XPS pattern, shown in Fig. 8a, contained Si-related peaks at 103, and 104 eV; these corresponded to Si⁴⁺ 2P_3/2 and 2p_1/2 orbital splitting respectively [39, 40], indicating the formation (Si-O) bonds of an SiO₂ framework. On the other hand, the O1 s XPS pattern, shown in Fig. 8b, contained a peak at 533 eV, which corresponds to framework O^2– atoms in a SiO₂ [39, 40]. This does not rule out the formation of hydroxide (OH^–) in a Zn-OH or hydride molecules (H₂O) adsorbed at the surface of ZnO/SiO₂ matrix [41–44], most probably occurred during the calcinations step under humid airflow (see experimental section). In addition to the formation of lower O^– state species in the ZnO to compensate for charge deficiencies at subsurface levels [41–44]. Finally, the Zn XPS spectrum (Fig. 8c) contained peaks at 1023 and 1046 eV, which correspond to Zn²⁺ 2p_3/2 and 2p_1/2 orbital splitting, respectively (Wei et al. 2007; Pesika et al. 2003; Tkachenko et al. 2003) [41–44], confirming the existence of ZnO particles in the SiO₂ matrix.

OPEN IN VIEWER

Fig.8

XPS spectrum Si2p, O1 s, and Zn2p for ZnO-SiO₂ matrix.

Conflict of Interest

The authors declare that they have no conflict of interest.