Sol–gel derived mesoporous GaAlPO₄ glass for heavy metal ion sequestration

Abstract

Mesoporous phosphate materials with chemical diversity and a high surface area are essential for their practical applications as heavy metal ion adsorbent materials. Herein, novel gallium aluminum phosphate mesoporous glasses are prepared through a simple aqueous sol–gel route free of any surfactant or template. The glass structure and its evolution from the gel to the final glass were monitored using solid-state NMR techniques, providing detailed insight into the reaction mechanism. During the formation of the mesoporous GaAlPO₄ glass, the gallium entered into the PO₄ tetrahedron via Ga–O–P bonds. The resulting mesoporous glass samples were tested as adsorbents to remove Pb(II) anions from solution and the ternary oxide glasses show high adsorption capacities (>36.63 mg g⁻¹, that is 0.18 mmol g⁻¹). Owing to their integrated features (including their variable compositions, high specific surface area, tunable pore size and shape flexibility) as well as specific acid–base surface properties, such mesoporous gallium aluminum phosphate materials are also expected to have potential either as catalysts or catalyst supports for industrial applications.

---

1. Introduction

The increase of heavy metal pollution in water resources is a major environmental health risk to human beings.¹ Heavy metal ions in water, such as Pb(II), Cd(II), Hg(II) and so on, can seriously threaten local residents’ health.² One of the most efficient routes for the removal of heavy metals is adsorption-based processes, the fundamental challenge of which is developing highly efficient adsorbents.^3,4 Mesoporous materials have attracted tremendous research interest in the exploration of ideal adsorbent materials.^5,6 They possess an intrinsic ability for accommodating capacious guest species as a result of high specific surface areas, tunable pore sizes, large pore volumes, as well as stable and interconnected frameworks. Since their discovery, mesoporous materials have shown great potential as adsorbents for heavy metal ion sequestration.⁷ To date, the most developed mesoporous materials employed for heavy metal ion sequestration are based on silica and carbon-based systems, the neutral SiO₄ and C skeleton of which lack binding properties for heavy metal ions.^6,8 The generation of specific active sites on the pore surfaces is based on post modification through attaching functional groups. However, the incorporation of organic groups would lead to an inhomogeneous distribution of active sites on the pore surface, which would limit its high adsorption efficiency and the organic active sites hinder application in harsh water environments (high temperature, strong acidic or basic). Therefore, the aim of the present task is synthesis of mesoporous materials that can be directly used as adsorbents without any further surface modification, through careful selection of the active component for binding heavy metal ions.

Phosphates have been shown to be high performance adsorbents in the field of water treatment to remove heavy metal ions.^9,10 The phosphate group is an active component in the adsorption of heavy metal ions due to its strong binding capacity for heavy metal ions.¹¹ Sol–gel synthesis has become a predominant approach for the preparation of mesoporous materials, because of the flexibility in shaping that allows these materials to be made into films, monoliths or coatings, which could be applied as separation membranes, sensors, and low-dielectric-constant (low-k) insulators.¹² Although great efforts have been made during the past two decades, progress has been limited in the development of sol–gel derived mesoporous phosphate materials with chemical diversity and high surface areas,^13–15 which are considered to be critical to the adsorption properties.

In recent years, our group has contributed research on exploiting novel aqueous sol–gel routes based on an unusual chelation of the metal precursors to prepare a series of phosphate multi-component mesoporous glasses with attractive high surface areas (>500 m² g⁻¹).^16–19 Both aluminum phosphate and gallium phosphate materials with open frameworks have attracted continued interest in the area of porous materials.^16,20 However, studies concerned with the incorporation of gallium into aluminum-based mesoporous materials have so far been poorly developed and not well understood. Here, we report the preparation of mesoporous GaAlPO₄ glasses using a facile aqueous sol–gel technique, free of any surfactant or template, for the first time. The compositions of the ternary oxide glasses varied from 0 to 25% Ga, resulting in high surface areas up to 515 m² g⁻¹. In contrast, Ga-rich samples tend to phase-separate (>30% Ga) and crystallize (>50% Ga) during the calcination process. The resultant GaAlPO₄ (GAP) mesoporous glasses were employed as adsorbents for the adsorption of the heavy metal ion Pb(II) from solution. The ternary oxide glasses show a fast adsorption process and high adsorption capacities, with a maximum adsorption capacity of 36.63 mg g⁻¹ (0.18 mmol g⁻¹) achieved using the mesoporous Ga_0.1Al_0.9PO₄ glasses, indicating their potential for application in water treatment. This facile and reproducible sol–gel approach is an important step toward the synthesis of multi-component mesoporous phosphates for environmental protection and industrial catalysis applications.

2. Experimental

2.1 Materials and methods

Gallium aluminum phosphate samples were prepared via an aqueous sol–gel route, using aluminum lactate, gallium nitrate and H₃PO₄ (1 M) as the precursors. A H₃PO₄ (1 M) solution was prepared by dissolving solid H₃PO₄ (98%, Fluka) in distilled water. The pH of the precursor solutions was adjusted with ammonia solution (1 M) and controlled to within 0.01 U using a pH meter (Mettler Toledo S20, Switzerland). Gel annealing was performed in a Nabertherm muffle furnace. A more detailed synthesis procedure can be found in the ESI.†

A stock solution containing 1000 mg L⁻¹ of Pb(II) was prepared by dissolving 0.8041 g of Pb(NO₃)₂ in 500 mL of deionized water. Simulated wastewater samples with different Pb(II) concentrations (50.0, 100.0, 200.0 mg L⁻¹) were prepared through dilution of the stock Pb(NO₃)₂ standard solution with deionized water. Typically, the Pb(II) adsorption experiments were carried out in a 20 mL glass bottle containing 10 mL of simulated wastewater at room temperature (25 °C) and without further pH adjustment. After the adsorption process, the samples were filtered through a 0.45 μm membrane, and the filtrates were analyzed using ICP-AES (iCAP 6300, Thermo Scientific). All the adsorption experiments were carried out in duplicate. The relative standard deviations were in the range of ±5%. The amount of Pb(II) adsorbed per unit mass of the adsorbent was evaluated using a mass balance equation (eqn (1))

q_t = (C₀ − C_t)V/W (1)

where q_t (mg g⁻¹) is the amount adsorbed per gram of adsorbent at time t (min), C₀ is the initial concentration of Pb(II) in the solution (mg L⁻¹), C_t is the concentration of Pb(II) at time t of the adsorption process (mg L⁻¹), W is the mass of the adsorbent used (g), and V (L) is the initial volume of the Pb(II) solution.

2.2 Characterization

The crystallinity of the prepared samples was checked via X-ray powder diffraction using the Guinier method, with CuKα1 radiation and using α-quartz (a: 5491.30 pm, c: 5540.46 pm) as an internal standard. Differential thermal analysis (DTA) and thermogravimetric analysis (TGA) were carried out with an EXSTER instrument (EXSTAR, TG/DTA 7300 Japan) using a heating rate of 10 K min⁻¹. Measurements for the BET surface area were obtained using a Micro-meritics ASAP 2010 volumetric adsorption analyzer with N₂ as the adsorbate at 77 K. Prior to the analysis, samples weighing between 0.1 and 0.3 g were outgassed at 200 °C for at least 6 h under vacuum until a residual pressure of ≤6 μm Hg was reached. The BET specific surface area was calculated according to the BET equation, using nitrogen adsorption data in the relative adsorption range from 0.06 to 0.2. The total pore volume, V_p, was obtained from the amount adsorbed at a P/P₀ of about 0.99. Mesopore size distributions were obtained using the Barrett–Joyner–Halenda (BJH) method and a cylindrical pore model.

The NMR measurements were carried out at ambient temperature with a Bruker AVANCE-400 spectrometer, using 4.0 mm MAS NMR probes operated at spinning speeds between 9 and 12 kHz. At a field strength of 9.4 T, the resonance frequencies were 104.3 MHz for ²⁷Al, and 162.4 MHz for ³¹P, respectively. The typical acquisition parameters were a pulse length of 5.0 μs (90°) for ³¹P and 1.0 μs (30°) for ²⁷Al, and a recycle delay of 60 s (³¹P) and 1 s (²⁷Al), respectively. The ²⁷Al and ³¹P NMR chemical shifts were referenced to a 1 M aluminium nitrate aqueous solution and 85% H₃PO₄. XPS measurements were performed using a Thermo Fisher Scientific K-Alpha spectrometer using non-monochromatic Al K radiation with a constant pass energy of 20.0 eV. All the binding energies (BE ± 0.2 eV) were referenced to the C-1s signal (284.6 eV). The morphology of the sample surface was determined using a Zeiss scanning electron microscope (SEM, JSM-6360LA).

3. Results and discussion

3.1 Composition and porous properties

Ga₂O₃–Al₂O₃–P₂O₅ system glasses have not been reported previously, while a wide compositional range of amorphous materials can be achieved using our sol–gel route (see Fig. S1†). Grey colored Ga₂O₃–Al₂O₃–P₂O₅ glasses were obtained when P/(Al + Ga) = 1.5 and 2. The grey color suggests the entrapment of small amounts of residual free carbon in the gel-derived samples as a result of the low specific surface area for the samples of this composition line. In particular, homogeneous transparent colorless glasses with a high surface area were obtained from the composition line of Ga_xAl_1−xPO₄. These glasses remained transparent and colorless until the amount of Ga was increased up to 25%. Increase of the amount of Ga results in phase separations (separated amorphous phases) and crystallization. The XRD results of Ga_0.3Al_0.7PO₄ confirm its amorphous structure after phase separation (Fig. S2†). The positions of the XRD peaks of Ga_0.5Al_0.5PO₄ match well with those of the AlPO₄ and Ga₂O₃ crystallization phases (Fig. S2†). Fig. S3† summarizes the thermal analysis data. The TGA trace of the xerogel with Ga : Al = 1 : 9 indicates that the volatile components (e.g. organics and water) are not completely removed below 400 °C, and a final weight loss near 700 °C was found. Xerogels with different Ga : Al ratios show similar curves. Fig. S3b† illustrates the TGA and DTA curves for the Ga_0.1Al_0.9PO₄ glass, and the minor weight loss observed below 200 °C is most likely due to an endothermic event arising from desorption and removal of physically adsorbed water in the porous material. The real compositions, and the glass transition temperatures (T_g) and crystallization temperatures (T_x) determined from the DTA traces are summarized in Table S1.† The ICP results of the samples from the Ga_xAl_1−xPO₄ composition line prove that the real compositions are quite consistent with the ratio of precursors that was used for the sol–gel route.

The specific surface area and porosity of selected GAPs were investigated using N₂ adsorption and desorption isotherms as shown in Fig. 1. A material is generally classified as mesoporous if it has a pore size between 2 and 50 nm, according to the International Union of Pure and Applied Chemistry (IUPAC). All the isotherms of the samples reveal stepwise adsorption and desorption branches of type IV curves, showing the typical pattern of a mesoporous material that has a three dimensional (3D) intersection according to the IUPAC classification.²¹ The hysteresis loop with stepwise adsorption and desorption branches was observed over a wide pressure range (P/P₀), and the surface area values for the Ga_0.05Al_0.95PO₄ and Ga_0.1Al_0.9PO₄ obtained at 600 °C are 388 m² g⁻¹ and 328 m² g⁻¹, respectively. Fig. 1b and d illustrate the pore size distributions of the two representative samples. After calcination at 600 °C, the average pore diameter of Ga_0.05Al_0.95PO₄ is 8.2 ± 0.1 nm. Table 1 further summarizes the main pore texture characteristics, BET surface area (S_BET), pore volume (V_p) and average pore diameter (d_p) of some representative glasses with different compositions. As indicated in Table 1, when 1% gallium was incorporated into the AlPO₄ network, the pore texture characteristics were similar to those for gallium-free AlPO₄ glasses prepared via a similar sol–gel route. As the amount of gallium increases, the average pore diameter (d_p) increases to 8–12 nm, which suggests that the incorporation of gallium into AlPO₄ glasses serves to tune the pore size of the mesoporous structure. With continuous increase of the gallium, the surface area decreases significantly. This behavior was attributed to the fact that the SiO₂-like three-dimensional network was destroyed through the incorporation of gallium, as indicated by the NMR results. To the best of our knowledge, the present contribution is the first successful preparation of Ga₂O₃–Al₂O₃–P₂O₅ glassy materials, exhibiting a high surface area and mesoporous structure.

Fig. 1 (a) Adsorption–desorption isotherms and (b) pore size (diameter) distribution (calculated using a BJH method) of the Ga_0.05Al_0.95PO₄ glass obtained after calcination at 600 °C for 6 h. (c) Adsorption–desorption isotherms and (d) pore size (diameter) distribution of the Ga_0.1Al_0.9PO₄ glass obtained after calcination at 600 °C for 6 h.

Table 1 The BET surface area, S_BET, total pore volume, V_p, and average pore diameter, d_p, of the samples with different compositions annealed at 600 °C for 6 h

Sample	SBET ± 5/m^2 g^−1	dp ± 0.1/nm	Vp ± 0.05/cm^3 g^−1
Ga0.01Al0.99PO4	415	4.1	0.42
Ga0.02Al0.98PO4	515	8.6	1.11
Ga0.05Al0.95PO4	388	8.2	0.71
Ga0.1Al0.9PO4	328	8.2	0.71
Ga0.2Al0.8PO4	132	12.1	0.40
Ga0.25Al0.75PO4	152	8.2	0.31

---

3.2 Gel to glass transition

To monitor the structural evolution during the gel → glass transition and reveal the local structure and site distribution of various atoms in the sol–gel samples, ³¹P and ²⁷Al NMR spectra were recorded for some representative samples after various processing steps. Fig. 2 (left) shows the ²⁷Al MAS-NMR spectra of samples with a composition of Ga_0.2Al_0.8PO₄ annealed at different temperatures. First of all, Fig. 2a and b illustrate the ²⁷Al MAS-NMR spectra of the Ga_0.2Al_0.8PO₄ xerogels annealed at 50 °C and 100 °C, respectively. Peak assignments can be made in relation to corresponding gallium-free aluminum phosphate xerogels.²² The two ²⁷Al peaks can be assigned to [Al(lact)₂(PO₃)₂]⁻ (∼5 ppm), and [Al(lact)₁(PO₃)₄]⁻ (∼−4 ppm). The ²⁷Al spectra indicate that lactate ligands (bound to aluminum) tend to be increasingly replaced by phosphorous ligands as the processing temperature increases. Dramatic structural changes occurred when the gels were heated from 100 °C to 200 °C. The signal attributed to [Al(lact)₂(PO₃)₂]⁻ was completely suppressed, whereas three new lactate-free Al sites were predominant, for which the peaks were attributed to Al(OP)₄ (near 38 ppm), Al(OP)₅ (near 4 ppm) and Al(OP)₆ (near −12 ppm). After calcination at 300 °C or above, no lactate-linked species can be observed and the aluminum was completely converted into three lactate-free units, which can be assigned as four-, five- and six-coordinated (Al(OP)₄, Al(OP)₅ and Al(OP)₆) aluminum sites. Subsequent heating at higher temperatures results in more subtle spectral changes for the aluminum sites. The Al(OP)₄ sites were the major Al-species, the proportion of which was increased somewhat at the expense of the Al(OP)₆ units. The final gallium aluminum phosphate glass structure is established after heating above 600 °C. Further evidence for the truly amorphous rather than nanocrystalline nature of the prepared materials is the five-coordinate aluminum species in the final structure.²³

Fig. 2 104.3 MHz ²⁷Al (left) and 162.4 MHz ³¹P (right) MAS-NMR spectra of Ga_0.2Al_0.8PO₄ samples at various processing stages during the gel to glass conversion.

Fig. 2 (right) shows ³¹P MAS-NMR spectra during the gel to glass conversion. The Qⁿ(mAl) notation was used to assign these peaks to specific phosphorus species. Here n is the number of P next-nearest-neighbors (NNN) per P tetrahedron and m is the number of Al NNN per P. The center of gravity shifts monotonically towards a lower frequency with increase of the annealing temperature, reflecting an increase in the average degree of P/Al connectivity. Q⁰(3Al), at −17.2 ppm, is already an overwhelmingly dominant species, while no Q⁰(0Al) sites can be detected, for the xerogel heated at 100 °C (Fig. 2i). For the gel heated at 200 °C (Fig. 2j), the Q⁰(3Al) species is significantly diminished again (only present as a shoulder), and the main ³¹P spectral component occurs at −22.7 ppm. This peak can be assigned to polymerized phosphate species of type Q⁰(4Al), based on the same trend of these chemical shift with that of gallium-free Al₂O₃–P₂O₅ glass.²² Fig. 2k and l indicate that this Al environment dominates in the samples calcined at 400 °C or higher temperatures, which is consistent with the conclusions drawn from the ²⁷Al MAS NMR spectra (Fig. 2 left). Compared with gallium-free Al₂O₃–P₂O₅ glass, we have noted that all the Q⁰(mAl) peaks in the ³¹P MAS-NMR spectra are slightly shifted (∼2–4 ppm) towards a lower frequency, the structural evolution induced by the introduction of gallium is discussed in detail in the ESI.†

3.3 Capacity and behavior for Pb(II) adsorption

Due to the high specific surface area and the high affinity of the phosphate group for metals, the resulting mesoporous GAPs were expected to show promise in the uptake of heavy metal ions for water purification applications.

Lead was chosen to illustrate the adsorption capabilities of the resulting GAPs. In this study, the Ga_0.1Al_0.9PO₄ glass, as a representative sample, was tested for adsorptive removal of Pb(II) from simulated wastewater. Fig. 3 shows the time profile of Pb(II) removal using 5 g L⁻¹ of the Ga_0.1Al_0.9PO₄ mesoporous glass with different initial Pb(II) concentrations and at neutral pH. The adsorption rates within the first 5 min were surprisingly fast for all the concentrations, and above 88% of the adsorption process occurred in the first 5 min. In a material with accessible active sites, the adsorption kinetics are determined by surface chemistry.²⁴ However, diffusion takes precedence over surface chemistry in monomodal mesoporous materials.²⁵ Therefore, Pb(II) adsorption occurred at a significantly fast rate with the mesoporous GAPs. The adsorption process was almost finished within 80 min, and there was no significant change from 80–240 min. The removal efficiencies were found to be 100, 96.9%, and 91.6% when the initial Pb(II) concentration was 50, 100, and 200 mg L⁻¹, respectively. Therefore, the maximum Pb(II) adsorption capacity of the Ga_0.1Al_0.9PO₄ mesoporous glass was found to be 36.63 mg g⁻¹ with an initial Pb(II) concentration of 200 mg L⁻¹. This reveals a high efficiency for the adsorption of Pb(II) onto the mesoporous GAPs in aqueous solution. It was also observed that the adsorption amount of Pb(II) increases with the initial Pb(II) concentration. This is because more Pb(II) results in a higher driving force for the ions to relocate from the solution onto the mesoporous GAPs and results in more collisions between the Pb(II) ions and active sites of the mesoporous GAPs. Similar phenomena have also been reported in the literature.^26,27 The excellent heavy metal adsorption capacity of phosphates is derived mainly from the ion-exchange properties of H_xPO₄⁻ groups, which can exchange with heavy metal ions.^28,29 The H_xPO₄⁻ ions required for ion-exchange with Pb²⁺ are provided from surface hydrolysis of the GAPs. It has been shown that the primary mechanism of metal ion removal by phosphates under neutral and weakly acidic environments is governed by surface hydrolysis, followed by subsequent ion-exchange.³⁰ Thus, the conceived mechanism for heavy metal adsorption by the GAPs can be described as follows:

Ga_0.1Al_0.9PO₄ + xH₂O ↔ Ga_0.1Al_0.9(OH)_x(H_xPO₄)

Ga_0.1Al_0.9(OH)_x(H_xPO₄) + xM²⁺ ↔ Ga_0.1Al_0.9(OH)_x(MH_x−2PO₄) + 2xH⁺

where M represents the corresponding heavy metal element. In the above hypothesis, where P in the form of H_xPO₄⁻ helps to adsorb the metal ions in aqueous solutions, H⁺ ions within H_xPO₄⁻ are released in a continuous and stepwise manner to allow for exchange with heavy metal ions.

Fig. 3 (a) Time profile of Pb(II) removal with Ga_0.1Al_0.9PO₄ mesoporous glass. The concentration of the GAP was 5.0 g L⁻¹; the initial Pb(II) concentration ranged from 50 to 200 mg L⁻¹. (b) The adsorption isotherm for Pb(II) on Ga_0.1Al_0.9PO₄ mesoporous glass, (c) Langmuir and (d) Freundlich adsorption isotherms for Pb(II) on Ga_0.1Al_0.9PO₄ mesoporous glass.

Adsorption isotherms were used to describe the adsorption behavior for Pb(II) of the mesoporous GAPs, among which Langmuir and Freundlich isotherms are the two most common models presenting linear equations as follows:³¹

Langmuir isotherm:

C_e/Q_e = 1/(Q_mK_L) + C_e/Q_m

Freundlich isotherm:

log(Q_e) = log(K_F) + (1/n)log(C_e)

The assumption for application of the Langmuir isotherm is that there is a homogeneity of the adsorption surface with all the adsorption sites having an equal adsorption surface and equal adsorbate affinity, while the Freundlich isotherm model is often applicable for heterogeneous adsorption surfaces, where Q_e (mg g⁻¹) and C_e (mg L⁻¹) are the amount of Pb(II) adsorbed and the Pb(II) concentration at equilibrium, respectively; Q_m (mg g⁻¹) and K_L (L g⁻¹) are Langmuir constants indicating the adsorption capacity and energy of adsorption, respectively; K_F (mg g⁻¹) and 1/n are Freundlich constants, related to the capacity and intensity of the adsorption, respectively. Here, both of the models are used to assess the adsorption data (Fig. 3c and d) and the adsorption isotherm parameters along with the correlation coefficients R² are presented in Table 2. The values of R² (close to unity) indicate the high suitability of these two adsorption isotherms, but the Langmuir isotherm seems to fit better according to its higher R² value.

Table 2 Adsorption isotherm parameters using Langmuir and Freundlich models for Pb(II) on Ga_0.1Al_0.9PO₄ mesoporous glass

Langmuir isotherm model			Freundlich isotherm model
Qm (mg g^−1)	KL (L g^−1)	R^2	KF (mg g^−1)	1/n	R^2
43.78	0.254	0.994	12.539	0.380	0.984

---

The kinetics of Pb(II) removal with the mesoporous GAPs were further investigated using different initial Pb(II) concentrations in the presence of 5.0 g L⁻¹ of the GAP. Pseudo-second-order rate constants (k₂) and the amount of Pb(II) adsorbed at equilibrium (q_e) were calculated from the slope and intercept of the plots of t/q_t versus t, according to eqn (2):²⁶

t/q_t = 1/k₂q_e² + t/q_e (2)

where k₂ (g mg⁻¹ g⁻¹) is the pseudo-second-order rate constant, q_e is the adsorption amount of Pb(II) (mg g⁻¹) at equilibrium, and q_t is the amount adsorbed (mg g⁻¹) at any time t. The kinetics of the removal process are illustrated in Fig. 4. Table 3 summarizes the theoretical and calculated q_e values, pseudo-second-order rate constants, k₂, and correlation coefficient values (R²). The calculated q_e value was obtained from the slope and intercept of the plots of t/q_t versus t according to eqn (2). The calculated q_e values matched well with the theoretical ones, showing a quite good linearity with R² above 0.999. Therefore, a chemisorption process was revealed as the adsorption kinetics follow a pseudo-second-order model. Moreover, the leaching behavior of the GAPs manifests a high chemical durability from pH3 to pH9 in aqueous solution, which is beneficial for application in a harsh environment (Fig. S5†).

Fig. 4 Removal kinetics for 50, 100, 200 mg L⁻¹ of Pb(II) in the presence of 5.0 g L⁻¹ of the mesoporous GAP. The contact time was 4 h.

Table 3 Theoretical and calculated q_e values, pseudo-second-order rate constants, k₂, and correlations. The contact time was 4 h

Theoretical qe (mg g^−1)	Calculated qe (mg g^−1)	k2 (g mg^−1 min^−1)	R^2
10	9.3658	4.1144	0.9999
20	19.8972	2.2568	0.9998
40	39.2773	1.0121	0.9998

---

3.4 Mechanisms of Pb(II) adsorption

N₂ adsorption isotherms for as-prepared and Pb-adsorbed Ga_0.1Al_0.9PO₄ samples (Fig. S6a†) revealed a decrease in the BET surface area from 328 m² g⁻¹ (as-prepared) to 75 m² g⁻¹ (Pb-adsorbed), and a corresponding reduction in the pore diameter from 12.1 nm to 9.1 nm (Fig. S6b†). These observations are consistent with the presence of a significant amount of adsorbed heavy metal ions attached to the framework walls of the GAPs, causing constriction of the channels. The SEM images in Fig. S7† show significant morphological differences for the surfaces and sections of the as-prepared and Pb-adsorbed samples. Fig. S7† demonstrates clearly that the Ga_0.1Al_0.9PO₄ has a mesoporous morphology with regular continuous wormlike mesopores of ca. 10.0 nm in diameter, which is consistent with the BET results. A narrow size distribution of the formed mesopores and no large pores were observed in the image, indicating the homogeneity of the GAP matrix and no phase separation during the sol–gel process of the GAP formation. The homogeneous distribution of the pores in the matrix provides a basis for uniform distribution of the active sites in the mesoporous framework, an essential prerequisite for highly efficient adsorption of heavy metal ions. The surface of the Pb-adsorbed samples turns rough and a layer of sediment was formed on the surface.

X-ray photo-electron spectroscopy was applied to further investigate the surface states of the as-prepared and Pb(II)-adsorbed GAP samples. In addition to Ga, Al, P, and O, XPS peaks for Pb²⁺ ions (Fig. 5a) are evident for the Pb(II)-adsorbed sample. Furthermore, the high resolution XPS spectrum of the Pb-4f region clearly demonstrates the presence of Pb(II) (Fig. 5c). The two broad peaks were assigned to Pb-4f_5/2 and Pb-4f_7/2, respectively. Changes in the center of gravity and full width at half maximum (FWHM) for the O-1s, Al-2p, P-2p and Ga-3d spectra are summarized in Table 4. The O-1s peak after Pb(II) adsorption was broader and had shifted to a lower binding energy as a result of the appearance of additional types of oxygen species, which overlap with the original O-1s signals. It is shown that the O-1s peak of the Pb-adsorbed sample (Fig. 5b) can be deconvoluted into four Gaussian peaks at about 530.8, 531.5, 532.0, and 532.6 eV, which were attributed to oxygen atoms in H–O–P, Pb–O–P, Ga–O–P and Al–O–P groups, respectively.^32–34 The formation of H–O–P and Pb–O–P after Pb adsorption suggests that hydroxyl groups played a vital role in the adsorption. The average P-2p binding energy shifts to lower values after Pb(II) adsorption, presumably reflecting the continuous alteration from P–O–Al and P–O–Ga toward P–O–H and P–O–Pb on the surface. Consistent with this difference, the average Al-2p and Ga-3d binding energies tend to decrease after Pb(II) adsorption. This trend is due to the higher ionicity of P–O–Pb and P–O–H compared with P–O–Al and P–O–Ga bonds. The XPS spectra give an impression of how the surface chemistry state of the GaAlPO₄ glass is modified through adsorption of Pb(II) in water. The ion-exchange mechanism induced the formation of more highly polarized P–O–Pb and P–O–H bonds on the surface.

Fig. 5 (a) Survey, (b) O-1s, (c) P-2p and Pb-4f, (d) Al-2p, and (e) Ga-2p high resolution XPS spectra of the as-prepared and Pb(II)-adsorbed Ga_0.1Al_0.9PO₄ mesoporous glass.

Table 4 Average binding energies and FWHM values measured from the O-1s, P-2p, Al-2p and Ga-2p XPS spectra of the as-prepared and Pb-adsorbed Ga_0.1Al_0.9PO₄ glass

Sample ID	O-1s	P-2p	Al-2p	Ga-2p
Ga0.1Al0.9PO4	532.4(2.9)	135.6(4.4)	76.1(2.5)	1119.8(4.1)	1146.6(3.5)
Pb-Adsorbed	531.8(3.2)	134.6(4.6)	75.5(2.6)	1118.9(4.0)	1146.1(3.5)

---

4. Conclusion

Mesoporous GaAlPO₄ glasses with varying compositions (Ga content from 0 to 25%), high specific surface areas (up to 515 m² g⁻¹) and uniform worm-hole like mesopores (∼8–12 nm) were fabricated via a facile sol–gel route free of any surfactant or template, for the first time, based on an unusual chelation of the metal precursors in aqueous solution. The conversion from the gel to the final glass was monitored using advanced solid-state NMR techniques, elucidating the molecular mechanism and the nature of the gallium incorporation into the aluminum phosphate materials. The mesoporous ternary oxide glasses had a high adsorption ability (>36.63 mg g⁻¹, that is 0.18 mmol g⁻¹) for Pb(II) anions in aqueous solution, showing a high potential for heavy metal ion sequestration applications. Given that the precursors are all common and non-toxic (metal salts and acid), the present synthetic strategy can be further extended to the development of cheap phosphate mesoporous materials with chemical diversity and a high surface area for large-scale applications, such as the fabrication of sensors, separation membranes, supercapacitors, and catalysts.

Acknowledgements

We thank Prof. H. Eckert for helpful discussions on the glass structure. We are grateful to Dr M. Xu and Prof. J. Ren for NMR characterization, Mr J. Xie for the XRD analysis, Dr T. Ye for help with the BET measurements and Mr Y. Xu for the ICP analysis. This work was financially supported by the National Natural Science Foundation of China (Grant No. 61275208).

References

1. H. Zhao, B. Xia, C. Fan, P. Zhao and S. Shen, Human health risk from soil heavy metal contamination under different land uses near Dabaoshan Mine, Southern China, Sci. Total Environ., 2012, 417, 45–54.
2. Z. Li, Z. Ma, T. J. van der Kuijp, Z. Yuan and L. Huang, A review of soil heavy metal pollution from mines in China: pollution and health risk assessment, Sci. Total Environ., 2014, 468, 843–853.
3. A. Demirbas, Heavy metal adsorption onto agro-based waste materials: a review, J. Hazard. Mater., 2008, 157, 220–229.
4. R. Celis, M. C. HermosÍn and J. Cornejo, Heavy Metal Adsorption by Functionalized Clays, Environ. Sci. Technol., 2000, 34, 4593–4599.
5. A. Sayari, S. Hamoudi and Y. Yang, Applications of Pore-Expanded Mesoporous Silica. 1. Removal of Heavy Metal Cations and Organic Pollutants from Wastewater, Chem. Mater., 2005, 17, 212–216.
6. Z. Wu and D. Zhao, Ordered mesoporous materials as adsorbents, Chem. Commun., 2011, 47, 3332–3338.
7. L. Mercier and T. J. Pinnavaia, Access in mesoporous materials: advantages of a uniform pore structure in the design of a heavy metal ion adsorbent for environmental remediation, Adv. Mater., 1997, 9, 500–503.
8. R. I. Nooney, M. Kalyanaraman, G. Kennedy and E. J. Maginn, Heavy metal remediation using functionalized mesoporous silicas with controlled macrostructure, Langmuir, 2001, 17, 528–533.
9. A. Aklil, M. Mouflih and S. Sebti, Removal of heavy metal ions from water by using calcined phosphate as a new adsorbent, J. Hazard. Mater., 2004, 112, 183–190.
10. Z. Elouear, J. Bouzid, N. Boujelben, M. Feki, F. Jamoussi and A. Montiel, Heavy metal removal from aqueous solutions by activated phosphate rock, J. Hazard. Mater., 2008, 156, 412–420.
11. B. Volesky and Z. Holan, Biosorption of heavy metals, Biotechnol. Prog., 1995, 11, 235–250.
12. C. J. Brinker, Evaporation-induced self-assembly: functional nanostructures made easy, MRS Bull., 2004, 29, 631–640.
13. B. Tian, X. Liu, B. Tu, C. Yu, J. Fan, L. Wang, S. Xie, G. D. Stucky and D. Zhao, Self-adjusted synthesis of ordered stable mesoporous minerals by acid–base pairs, Nat. Mater., 2003, 2, 159–163.
14. I. Jiménez-Morales, A. Teckchandani-Ortiz, J. Santamaría-González, P. Maireles-Torres and A. Jiménez-López, Selective dehydration of glucose to 5-hydroxymethylfurfural on acidic mesoporous tantalum phosphate, Appl. Catal., B, 2014, 144, 22–28.
15. C. Sui, Y. Lu, H.-L. Gao, L. Dong, Y. Zhao, L. Ouali, D. Benczédi, H. Jerri and S.-H. Yu, Synthesis of Mesoporous Calcium Phosphate Microspheres by Chemical Transformation Process: Their Stability and Encapsulation of Carboxymethyl Chitosan, Cryst. Growth Des., 2013, 13, 3201–3207.
16. L. Zhang, A. Bögershausen and H. Eckert, Mesoporous AlPO₄ glass from a simple aqueous sol–gel route, J. Am. Ceram. Soc., 2005, 88, 897–902.
17. C. C. de Araujo, L. Zhang and H. Eckert, Sol–gel preparation of AlPO₄–SiO₂ glasses with high surface mesoporous structure, J. Mater. Chem., 2006, 16, 1323–1331.
18. L. Zhang, C. C. de Araujo and H. Eckert, Aluminum lactate-an attractive precursor for sol–gel synthesis of alumina-based glasses, J. Non-Cryst. Solids, 2007, 353, 1255–1260.
19. L. Zhang, C. C. de Araujo and H. Eckert, Structural role of fluoride in aluminophosphate sol–gel glasses: high-resolution double-resonance NMR studies, J. Phys. Chem. B, 2007, 111, 10402–10412.
20. K. J. Balkus, S. J. Kim and A. M. Sargeson, in Synthesis and Characterization of GaPO₄ Molecular Sieves Using Metal Complexes as Templates, MRS Proceedings, Cam. Univ. Press, 1996, p. 217.
21. S. Lowell and J. E. Shields, Powder surface area and porosity, Springer Science & Business Media, 1991, vol. 2.
22. L. Zhang and H. Eckert, Sol–gel synthesis of Al₂O₃–P₂O₅ glasses: mechanistic studies by solution and solid state NMR, J. Mater. Chem., 2004, 14, 1605–1615.
23. C. A. Fyfe, J. L. Bretherton and L. Y. Lam, Solid-State NMR Detection, Characterization, and Quantification of the Multiple Aluminum Environments in US-Y Catalysts by ²⁷Al MAS and MQMAS Experiments at Very High Field, J. Am. Chem. Soc., 2001, 123, 5285–5291.
24. V. R. Stamenkovic, B. Fowler, B. S. Mun, G. Wang, P. N. Ross, C. A. Lucas and N. M. Marković, Improved oxygen reduction activity on Pt₃Ni (111) via increased surface site availability, Science, 2007, 315, 493–497.
25. J. Fan, C. Yu, F. Gao, J. Lei, B. Tian, L. Wang, Q. Luo, B. Tu, W. Zhou and D. Zhao, Cubic mesoporous silica with large controllable entrance sizes and advanced adsorption properties, Angew. Chem., Int. Ed., 2003, 115, 3254–3258.
26. H. Xiao, Z. Ai and L. Zhang, Nonaqueous sol–gel synthesized hierarchical CeO₂ nanocrystal microspheres as novel adsorbents for wastewater treatment, J. Phys. Chem. C, 2009, 113, 16625–16630.
27. J. S. Hu, L. S. Zhong, W. G. Song and L. J. Wan, Synthesis of hierarchically structured metal oxides and their application in heavy metal ion removal, Adv. Mater., 2008, 20, 2977–2982.
28. X. Wang, X. Yang, J. Cai, T. Miao, L. Li, G. Li, D. Deng, L. Jiang and C. Wang, Novel flower-like titanium phosphate microstructures and their application in lead ion removal from drinking water, J. Mater. Chem. A, 2014, 2, 6718–6722.
29. S. Saxena and S. D’Souza, Heavy metal pollution abatement using rock phosphate mineral, Environ. Int., 2006, 32, 199–202.
30. S. Mustafa, S. Murtaza, A. Naeem and K. Farina, Ion exchange sorption of Pb²⁺ ions on CrPO₄, Environ. Technol., 2005, 26, 353–360.
31. Z. Reddad, C. Gerente, Y. Andres and P. Le Cloirec, Adsorption of several metal ions onto a low-cost biosorbent: kinetic and equilibrium studies, Environ. Sci. Technol., 2002, 36, 2067–2073.
32. Z. Wang, Y. Feng, X. Hao, W. Huang and X. Feng, A novel potential-responsive ion exchange film system for heavy metal removal, J. Mater. Chem. A, 2014, 26, 10263–10272.
33. R. K. Brow, R. J. Kirkpatrick and G. L. Turner, Local Structure of xAl₂O₃(1 − x)NaPO₃ Glasses: An NMR and XPS Study, J. Am. Ceram. Soc., 1990, 73, 2293–2300.
34. C. Viornery, Y. Chevolot, D. Léonard, B.-O. Aronsson, P. Péchy, H. J. Mathieu, P. Descouts and M. Grätzel, Surface Modification of Titanium with Phosphonic Acid To Improve Bone Bonding: Characterization by XPS and ToF-SIMS, Langmuir, 2002, 18, 2582–2589.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra17766b

---