Figure 1 shows the absorption spectra of the diluted RhB (C_o = 12 mg·L⁻¹) aqueous solutions stirred with pure and hybrid V₂O₅ xerogels and measured at different time intervals. An aqueous solution of the pure xerogel shows a broad absorption spectrum in the visible region (Fig. 1(a)) that is due to the brown coloured ions and light scattering in the colloidal suspension.^[28] Moreover, two absorption bands in the ultraviolet region at 285 nm and 385 nm represent the electron charge transition from O → V⁵⁺ in the xerogel in the presence of excess water.^[10] The RhB aqueous solution that was not stirred with the xerogel shows a maximum absorption peak at 553 nm. It can be observed that the maximum absorption band of RhB shifted from 553 nm to 573 nm when stirred with the pure V₂O₅ xerogel solution. However, the rest of the absorption spectrum is similar to the absorption curve of the pure xerogel aqueous solution. In contrast, the intensity of these bands is negligible for the RhB aqueous solution stirred with the hybrid V₂O₅ xerogel, for which the main absorption band of RhB located at 553 nm was not shifted. This characteristic of the hybrid xerogel is beneficial for employing it as a matrix for the intercalation of the dye molecules.^[16] Moreover, the insoluble V₂O₅ hybrid xerogel did not form a brown coloured colloidal suspension, as shown in Fig. 2. After 30 min, ∼ 85% and 95% of the RhB were removed from the solutions stirred with the pure and hybrid V₂O₅ xerogels, as measured at λ_max = 573 nm and 553 nm, respectively. The RhB solution with the pure xerogel was stirred for another 100 min, but it did not show any further absorption of RhB. However, the intensity of the peaks in the UV region increased (Fig. 1(a)).

Fig. 1.

	Figure Option ViewDownloadNew Window
	Fig. 1. (color online) UV–Vis absorption spectra of RhB aqueous solutions with an initial concentration Co = 12 mg·L−1, stirred with pure (a) and V2O5 hybrid (b) xerogels, as a function of contact time.

Fig. 2.

	Figure Option ViewDownloadNew Window
	Fig. 2. (color online) Images of RhB aqueous solutions stirred with the pure V2O5 xerogel (a), and hybrid V2O5 xerogel (b) at successive time intervals.

When the RhB aqueous solution was mixed with the pure and hybrid xerogels, the pH values of the solutions were noted as 3.4 and 4, respectively. The point of zero charge (PZC) for the V₂O₅ xerogel is reported as approximately pH 2.^[8] At the PZC, a substance is considered to be neutral. At pH>PZC, a substance exhibits anionic behaviour. In the present study, the zeta potential of the pure and hybrid xerogels at their respective pH values was calculated to be in the range of −24 mV to −35 mV. This indicates that the xerogel particles have negatively charged surfaces in the pH range in question. In an aqueous solution, RhB can exist in three different ionic forms, i.e. protonated RBH⁺, neutral zwitterionic RhB^± and colourless lactone. The presence of these forms depends on the pH, concentration and temperature of the environmental solution (prototropic forms of RhB are shown in Fig. A2, Appendix A). At pH < 4, the RhB ions are in the protonated and monomeric molecular form. Thus, they can enter into the pore structure easily. At a pH > 4, the zwitterionic form of RhB exists in water, and causes the aggregation of RhB to form a larger molecular unit (dimer) which is not easily adsorbed by the sorbate.^[29] It is therefore speculated that at the pH values measured in the hybrid xerogel solution, RhB⁺ cations were attracted towards the negatively charged V₂O₅ layers via electrostatic forces, leading to dominating ion exchange interactions with the physically bonded water molecules and H₃O⁺ ions present in the xerogel.

The average pore diameter (d_por), specific surface area (S_BET) and average pore volume (V_por) of the pure xerogel were calculated by using the BET and BJH methods, and were found to be 33.8 nm, 1.25 m²·g⁻¹ and 4.17×10⁻⁵ m³·g⁻¹, respectively. The addition of TBO into the V₂O₅ xerogel induced a dramatic increase in the S_BET and V_por values to 27.57 m²·g⁻¹ and 3.72 × 10⁻⁴ m³·g⁻¹, respectively. Whereas the pore diameter for the hybrid xerogel was decreased to 7.25 nm, but the number of pores was increased as observed by increase in pore volume. The S_BET for the hybrid xerogel was more than 22 times larger than that of the pure xerogel. The significant increase of the surface area and pore volume of the hybrid xerogel compared to the pure xerogel indicates that the former possess a more hollow structure and provides more reaction sites for the target dye molecules. The pore diameter of xerogel is sufficient for the adsorption of monomers of RhB at the tested pH of aqueous solution. Moreover, The UV–Vis spectroscopy results confirm that the hybrid xerogel has more absorption capacity for the RhB dye than the pure xerogel. While the pore diameter of the hybrid xerogel was smaller at 7.25 nm, the number of pores was larger, as indicated by the increase in the total pore volume. Furthermore, the pore diameter of the xerogel is sufficient for the adsorption of RhB monomers at the measured pH of the aqueous solution. It is speculated that the absorption of RhB by the hybrid xerogel takes place in two steps: initially the RhB is adsorbed through surface diffusion over the V₂O₅ hybrid xerogel, and secondly, the dye monomers intercalate between the layers of the xerogel.

To investigate the effect of different initial concentrations and absorbent dosages on the RhB absorption capacity of the hybrid V₂O₅ xerogel at a fixed pH = 4, a series of experiments were performed at room temperature. RhB aqueous solutions with a volume of 50 mL and initial concentrations 36, 48, 60, 72, 84, and 120 mg·L⁻¹ were prepared. A fixed amount of the hybrid xerogel (30 mg) was added to each solution, which was then agitated in a shaker at a speed of 200 rpm for 24 h. Once the equilibrium state was reached, the amount of RhB absorbed by a unit mass of the absorbent q_e (mg·g⁻¹) was calculated by the following relation:

where V is the volume of the solution, M is the mass of absorbent, C_o is the initial concentration, and C_e is the concentration of the RhB at equilibrium time t.

The uptake of RhB increases linearly from 58 mg·g⁻¹ to 134 mg·g⁻¹ for initial dye concentrations in the 36 mg·L⁻¹ to 84 mg·L⁻¹ range, but starts to deviate from the linear trend when the dye concentration is increased further to 120 mg·L⁻¹, as shown in Fig. 3(a). Therefore, the above value indicates the saturation state of the absorbent corresponding to the maximum absorption of the dye.

Fig. 3.

	Figure Option ViewDownloadNew Window
	Fig. 3. (color online) The effect of initial dye concentration (a) and absorbent dose (b) on RhB uptake by the hybrid V2O5/TBO xerogel.

It must also be noted that the dose of absorbent in a RhB solution also affects its absorption capacity. For a fixed initial dye concentration of 120 mg·L⁻¹, the absorbent amount was varied from 10 mg to 60 mg. In this case, the uptake of RhB was observed to be in the range from 179 mg·g⁻¹ to 99 mg·g⁻¹, respectively (Fig. 3(b)). These results indicate a high RhB absorption capacity for the hybrid xerogel. From the graph in Fig. 3(b), it can be seen that the absorption capacity of the hybrid xerogel q_e has a somewhat inversely proportional relationship with the sorbate dose in the range 30 mg–60 mg, i.e., q_e decreases with increasing sorbate dose. This indicates that the sorbate has a much larger absorption surface available for the accommodation of the RhB dye molecules. For sorbate doses below 30 mg, the increase in the absorption capacity q_e of the xerogel is barely noticeable (e.g. 171 mg·g⁻¹ to 179 mg·g⁻¹ for sorbate doses from 20 mg to 10 mg, respectively). This result explains the non-linear trend observed in Fig. 3(a) at the saturation state for RhB dye absorption of the hybrid xerogel.

At the equilibrium time of 24 h, the hybrid V₂O₅ xerogel had absorbed 117 mg·g⁻¹ of RhB from the concentrated RhB solution (120 mg·L⁻¹). The RhB-loaded xerogel in the solution was allowed to settle for 2 h after absorption. The supernatant was carefully extracted using a set of transfer pipettes, and the precipitates were dried in an oven at 50 °C for 2 h. Two solvents, DI water and ethanol, were used in the desorption study. The solutions were gently stirred using a magnetic stirrer at 100 rpm for 24 h in a dark chamber. As shown in Fig. 4, it was observed that approximately 4 mg·g⁻¹ and 65 mg·g⁻¹ of absorbed RhB were desorbed from the dye loaded hybrid xerogel in DI water and ethanol, respectively. These values are equivalent to 3.41% and 55.5% of the total amount of the absorbed dye (117 mg·g⁻¹). The results indicate that after desorption of the dye, the xerogel can be reused for absorption. Furthermore, desorption studies provide insights into the absorption mechanism, i.e., whether it is physisorption or chemisorption. If desorption readily occurs in neutral water, then the absorbate is likely to be held on the absorbent by weak bonds. On the other hand, if the absorbate desorbs only in chemical reagents, it is an indication of stronger bonding between the absorbate and absorbent.^[30] Hence, the above results indicate that the absorption of RhB was driven by a strong binding force between RhB and the hybrid xerogel. Moreover, while the recovery of an organic cation may not be easily done using inorganic cations, one organic cation may be able to replace another.^[31]

Fig. 4.

	Figure Option ViewDownloadNew Window
	Fig. 4. (color online) Desorption of RhB from the V2O5/TBO hybrid xerogel in DI water and ethanol solvents.

The XRD patterns reflect the well-defined lamellar structure of the pure V₂O₅ xerogel in the c direction, as indicated by the strong (001) peak at 2θ = 7.47° and similar reflections of (003), (004) and (005) at higher angles (Fig. 5(a)). The peaks delineate the orthorhombic V₂O₅1.6H₂O hydrate according to the data in the JCP2-2CA database (pdf file No: 040-1296).

Fig. 5.

	Figure Option ViewDownloadNew Window
	Fig. 5. (color online) (a) XRD patterns of the pure V2O5 xerogel, V2O5/TBO hybrid xerogel and dye loaded hybrid xerogel. (b) Increase in the d spacing as measured by the shift in the (001) peak in the RhB loaded V2O5/TBO xerogel.

The substitution of titanium for vanadium in the V₂O₅/TBO hybrid xerogel destroys the stacking order of the lattice in the c direction, as shown by a decrease in the intensity of the 00l peaks.^[24] Moreover, compared to the pure xerogel, the interlayer spacing d decreased from 11.76 Å to 11.58 Å in the hybrid V₂O₅/TBO xerogel. This indicates that during dispersion of the V₂O₅ xerogel in the TBO/ethanol solution, loosely bonded water molecules were ejected from the interlayer regions of the xerogel, while Ti⁴⁺ ions were intercalated between the layers to form V–O–Ti layers.^[23] It is known that guest species do not always change the interlayer spacing or improve the stacking order. Instead, such changes depend on the chemical nature of the guest and host species, as well as the reactions taking place between them.^[32] A redox intercalation, which reduces V⁵⁺ to V⁴⁺ or forms VO²⁺ ions in the xerogel, can be suggested in this case. The change in the xerogel colour from brown to green is an indication of the reduction process, as shown in Fig. 2. It is speculated that Ti⁴⁺ ions are connected to oxygen atoms in the xerogel by strong covalent or ionic bonds, in contrast to the weak physical bonds of the water molecules. The strong bonding between the vanadium oxide layers due to the presence of Ti⁴⁺ ions stabilizes the V₂O₅ xerogel, leading to the production of an insoluble hybrid xerogel.

The XRD pattern in Fig. 5(b) shows a shift of the (001) peak of the V₂O₅/TBO hybrid xerogel from 2θ = 7.28° to 4.59° (d = 11.58 Å to 19.24 Å) after the absorption of the RhB⁺ cations. Taking into account the dimensions of the RhB⁺ ion (length: 18 Å, base: 7 Å),^[30] and the increase in the interlayer spacing d, we can reasonably speculate that either the intercalation of monomeric and dimeric forms of RhB⁺ or coexistence of both is the reason for the above observation.^[33,34] These results are further confirmed by FTIR spectroscopy, as discussed in the next section. The pure xerogel absorbs water molecules along with the RhB molecules from the aqueous solution due to its flexible layered structure, and forms a colloidal solution upon swelling. In contrast, the hybrid xerogel only accommodates the RhB molecules via exchange of hydronium ions present between the layers. The stable structure of the hybrid xerogel does not allow further absorption of water molecules.

The FTIR results of the pure vanadium oxide xerogel showed a broad stretching band υ_s at 3600–2500 cm⁻¹ due to O–H ions (Fig. 6(a)). A narrow band at 1610 cm⁻¹ describes the water molecules that are chemically bonded to the oxide network. Four bands can be observed in the wavenumber range 1000–400 cm⁻¹. The terminal oxygen symmetric stretching band υ_s of V=O, with the main peak at 980 cm⁻¹ and a shoulder band at 923 cm⁻¹, indicates the presence of mixed valance V⁴⁺ and V⁵⁺ ions. The asymmetric and symmetric stretching bands of the bridging oxygen atoms V– O–V are also shifted to lower frequencies of 719 cm⁻¹ and 427 cm⁻¹ from their typical band positions at 764 cm⁻¹ and 527 cm⁻¹, respectively.^[35] The reduced V⁴⁺ ions might have been produced during the slow heat treatment of the sol during the synthesis of the V₂O₅ xerogel. The presence of O–H vibration modes in the FTIR spectrum of the hybrid xerogel indicate that a large number of water molecules are still present in the interlayer space of the xerogel, even after treatment with TBO (Fig. 6(b)). The broadness of the V=O band in the hybrid xerogel, compared to that of the pure V₂O₅ xerogel, indicates weaker V=O bonding. FTIR confirmed that the Ti⁴⁺ ions were inserted in the interlayers of the V₂O₅ xerogel,^[24] and reduced the V⁵⁺ ions to V⁴⁺ by electron transfer in a manner similar to other metal cations such as Li, Na and Mg.^[36] The size of the reduced vanadium ion, V⁴⁺, is larger than the V⁵⁺ ion, and causes the expansion of the V=O bond. The polarity of the V–O–V bond is also affected by these structural changes, as observed by the shifting of the bands to lower wavenumbers, i.e. from 719 cm⁻¹ to 707 cm⁻¹ in the pure and hybrid xerogels, respectively. The IR spectrum of the dyeloaded hybrid xerogel clearly shows several bands belonging to RhB (Fig. 6(c)), which shows the intercalation of the RhB molecules within the layers of the xerogel. Moreover, a decrease in the intensity of bands corresponding to water molecules indicates replacement of the water molecules with the dye molecules. The bands belonging to the RhB molecules are also shifted slightly from their original positions (Fig. 6(d)). These band shifts are ascribed to the interaction of the RhB molecules with the vanadium oxide layers in the xerogel.

Fig. 6.

	Figure Option ViewDownloadNew Window
	Fig. 6. (color online) FTIR spectra of the pure V2O5 xerogel (a), V2O5/TBO hybrid xerogel (b), RhB loaded V2O5/TBO hybrid xerogel (c) and RhB in its dry form (d).