As a big family of layered silicates, phyllosilicates include various natural clays, such as montmorillonite, saponite, laponite, attapulgite, and many synthetic metal phyllosilicates^[1]. Because of their specific structural features and ion exchange properties, phyllosilicates have large application potential in the field of heterogeneous catalysis, electrocatalysis, waste water treatment, etc. In the past years, plenty of research works have been carried out by using natural clays as catalysts supports or adsorbents for dye removal after inorganic or organic modification^[2,3,4,5,6]. In recent years, the synthesis and application of metal phyllosilicates, such as Ni^[7,8,9], Co^[10], Cu^[11,12], Mg^[13] containing phyllosilicates, have aroused more and more attention.

For nickel phyllosilicates, the difference in the ratio of tetrahedral and octahedral sheets generates two types of structures: (1 : 1 and 2 : 1 type nickel phyllosilicates) with the formula of Ni₃Si₂O₅(OH)₄ and Ni₃Si₄O₁₀(OH)₂^[14,15]. The tetrahedral sheets consist of silicon-oxygen tetrahedra, which are linked by two-dimensional corner-sharing three of every four oxygen atoms so as to form sheets of indefinite extent, in which the ratio of silicon to oxygen is 2 : 5. The octahedral sheets consist of two planes of closely packed O^2-, OH^- anions of octahedra with Ni²⁺ as the central cations. Based on this fundamental structure, nickel phyllosilicates can be assembled into different two-dimensional (2D) or three-dimensional (3D) architecture with enhanced application performances by using unique preparation methods. For instance, nanotubular Ni₃Si₂O₅(OH)₄ can be synthesized by hydrothermal methods in strongly alkaline environment and applied as lithium battery anode support materials^[15,16,17]. Several efforts have been made to synthesize Ni₃Si₂O₅(OH)₄ with hollow sphere or core-shell structure by using SiO₂ microspheres as templates^[18,19]. However, this preparation route often requires hard-controlled conditions and subsequent complicated procedures for the removal of templates. Efficient preparation of Ni₃Si₂O₅(OH)₄ with 3D hierarchical structure is very meaningful for expanding its applications.

Nickel phyllosilicates have so far been well applied in the preparation of Ni-based catalysts. Considering their well-formed layered structure, nickel phyllosilicates are also expected to be very effective adsorbents for dye removal from waste water. However, to our best knowledge, no report focused on this issue is available up to now. In this study, Ni₃Si₂O₅(OH)₄ hierarchical porous microspheres were synthesized via a simple hydrothermal method using NiCl₂, tetraethoxysilane (TEOS) as the raw materials, and then were used as the adsorbent for the removal of basic fuchsin (BF) from simulated waste water.

All chemicals were analytical grade and obtained from Sinopharm Chemical Reagent Beijing Co., Ltd., Beijing, China. In a typical procedure for the synthesis of Ni₃Si₂O₅(OH)₄, 0.4420 g (2.12 mmol) TEOS and 64 mL deionized (DI) water were added to a conical flask, then 0.5043 g (2.12 mmol) NiCl₂·6H₂O was added to the conical flask and dissolved completely. 0.8 g (13.3 mmol) urea was added to the solution followed by vigorous stirring for 0.5 h. The resultant solution was transferred to a Teflon-lined stainless steel autoclave with a capacity of 100 mL. The autoclave was sealed and heated to 210 ℃ and kept in an isothermal state for 12 h, then cooled down to room temperature. After washed with DI water, the solid product was collected and finally dried at 75 ℃ for 12 h.

X-ray diffraction (XRD) patterns were recorded on a MiniFlex600 Powder X-ray powder diffractometer (Rigaku, Japan). Scanning electron microscopy (SEM) measurements were carried out on a JSM 6700F field emission scanning electron microscope (JEOL, Japan). Transmission electron microscopy (TEM) measurements were carried out on a JEM-2100PLUS high resolution transmission electron microscope (JEOL, Japan). Nitrogen adsorption-desorption isotherms were measured using a Kubo-X1000 surface area and porosity analyzer (Builder Electronic Technology, China), and the specific surface areas were calculated using the Brunauer- Emmett-Teller (BET) equation. Zeta potential of the sample was tested on a ZEN3700 Zeta potential analyzer (Malvern, UK).

For a typical adsorption removal of BF from simulated waste water, 20 mg Ni₃Si₂O₅(OH)₄ microspheres were mixed with 50 mL basic fuchsin solution (original concentration: 50 mg·L^-1). At preset period, 2 mL suspension was taken from the solution, and the liquid was monitored by a UV756CRT UV-Vis spectroscope (Yoke Instrument, China) after solid-liquid separation. The adsorption capacity of the adsorbents for BF was evaluated by the following Eq. (1):

where q_t (mg·g^-1) is the adsorption capacity at time t, c₀ and c_t (mg·L^-1) the original and real-time concentrations of BF solution, respectively, V (L) is the volume of the employed BF solution, and m (g) is the mass of the employed adsorbent.

NiCl₂ and TEOS were used as the raw materials to synthesize nickel silicates through the urea-assisted hydrothermal method. The effect of Ni/Si molar ratio on the phase composition of the products is shown in Fig. 1. In previous study it was found that the 2 : 1 type nickel phyllosilicate Ni₃Si₄O₁₀(OH)₂ can be synthesized with a low Ni/Si molar ratio (0.25 : 1)^[20]. Fig. 1 shows that the 1 : 1 type Ni₃Si₂O₅(OH)₄ is easier to form when the Ni/Si molar ratio is set up to or higher than 0.5 : 1, and the diffraction peak at 2θ=12.2° which is associated with the (002) plane of Ni₃Si₂O₅(OH)₄, is obviously increased with the increase of Ni/Si molar ratio from 0.5 : 1 to 1.5 : 1. Besides this characteristic peak for the layered structure of Ni₃Si₂O₅(OH)₄, there is another small diffraction peak appearing at 2θ=8.3°, which represents a larger layer spacing (1.067 nm) than the layer spacing of Ni₃Si₂O₅(OH)₄ (0.0736 nm), but smaller than that of Ni₃Si₄O₁₀(OH)₂ (1.510 nm). It can be considered an intermediate state transforming from 2 : 1 type to 1 : 1 type layered structure. As shown in Fig. 1, single phase Ni₃Si₂O₅(OH)₄ was obtained with a Ni/Si molar ratio of 1 : 1. And the further increase of Ni/Si molar ratio resulted in the formation of NiOOH (JCPDF 06-0141).

Fig. 2 shows the SEM and TEM images of the Ni₃Si₂O₅(OH)₄ hydrothermally synthesized with the Ni/Si molar ratio of 1 : 1 in the raw materials. As shown in Fig. 2(a, b), the Ni₃Si₂O₅(OH)₄ microspheres present a microsphere-shaped morphology with an average diameter of ca. 2.5 μm. The magnified SEM image indicates that these microspheres are porous and assembled by nanosheets. TEM images further confirm that the porous microspheres are composed of nanosheets (Fig. 2(c, d)). In Fig. 2(d), layered structure can be found at the edge of the nanosheets, which corresponds to the (002) plane of Ni₃Si₂O₅(OH)₄.

The N₂ adsorption/desorption isotherms and the pore size distribution (PSD) curves of the hydrothermal products with different Ni/Si molar ratios in the raw materials are illustrated in Fig. 3 and Fig. 4, respectively. The experimental data for specific surface areas, porous volumes and the average pore sizes are summarized in Table S1. All of the samples exhibit similar adsorption- desorption isotherms which can be assigned to type IV adsorption isotherms with type H2 hysteresis loops, implying the presence of ink-bottle-like pores. As shown in Table S1, the BET surface areas were significantly affected by Ni/Si molar ratio in the raw materials. With the increase of Ni/Si ratio from 0.5 : 1 to 1.5 : 1, the BET surface areas decreased successively from 139.4 m²·g^-1 to 95.5 m²·g^-1.

For the materials with small mesopores (pore diameter 2-10 nm), the tensile strength effect (TSE) during the adsorption measurements could significantly affect the adsorption isotherm, and therefore leading to incorrect assessment of PSD when the BJH model was applied to the desorption branch of the isotherm. Instead, the adsorption branch is hardly affected by the TSE phenomenon and is preferred for pore size calculations as pointed out in the literature^[21]. In this study, the desorption branch and the adsorption branch were both adopted to analyze the PSD as shown in Fig. 4. The desorption branch indicates that all products have dominant mesopores centered at about 4.1 nm, which is completely different as compared to the results obtained from the adsorption branch. This large difference should be attributed to the TSE phenomenon, which leads to the misinterpretation of the PSD derived from the desorption branch. In fact, the prominent peak observed around 4.1 nm is not the real porous property. Wide PSD, especially in the range from 20 to 100 nm, is observed for all products according to the adsorption branch. This is in good agreement with the SEM and TEM images, in which piled pores produced by the assembly of nanosheets can be found.

To better understand the effect of urea on the formation of the Ni₃Si₂O₅(OH)₄ microspheres, various alkali source and different adding amount of urea were employed for the hydrothermal synthesis. XRD patterns and SEM images of the products are shown in Fig. 5. Without alkali addition, there was no solid product obtained after the hydrothermal reaction. The addition of ammonia, sodium hydroxide and urea can render the formation of Ni₃Si₂O₅(OH)₄. The reaction is considered as the following Eq. (2):

Thus, alkaline condition is necessary for the formation of the product. Fig. 5 shows that the morphologies of the products vary dramatically with different alkali sources. When ammonia was used, the product was mainly composed of porous near-spherical particles with the diameter less than 1 μm. In comparison, the use of sodium hydroxide resulted in the formation of dense irregular blocks. It shows that mild alkali environment is conducive to the formation of porous structure. The presence of urea in the hydrothermal process can also provide mild alkali environment as the situation in ammonia, because its decomposition in aqueous solution leads to the release of CO₂ and NH₃ into the system. NH₃ can easily dissolve into water, generating NH₄⁺ and OH^- ions. In this basic aqueous solution, TEOS reacts with Ni²⁺ and OH^- ions to form Ni₃Si₂O₅(OH)₄. Yet there are two differences compared with that using ammonia directly. Firstly, OH^- ions were gradually released with the decomposition of urea. It did not cause the rapid formation of abundant crystal nucleus, and larger particles would be gradually formed by the subsequent crystal growth. On the other hand, the presence of CO₂ bubbles in the solution can be heterogeneous nucleation centers according to the previous reports^[22,23]. Ni₃Si₂O₅(OH)₄ tends to aggregate together around the CO₂ bubbles and self-assemble into larger particles, resulting in the porous structure of the final products. It was found that SiO₂ particles formed on the surface of the microspheres under the conditions of low temperature or short time hydrothermal treatment, as revealed in the synthesis of Ni₃Si₄O₁₀(OH)₂^[20]. It is considered that besides the reaction shown in Eq. (2), base-catalyzed hydrolysis of TEOS takes place with the gradually decomposition of urea, leading to the formation of SiO₂. As the reaction approaches completion, the concentration of OH^- ions is thereby increased. Then the SiO₂ formed through the hydrolysis of TEOS is etched under this elevated alkaline condition. Based on the above analysis, the possible formation mechanism is illustrated in Fig. 6.

As shown in Fig. 5(f), the hydrothermal product turns to be dense irregular blocks with the amount of urea increasing to 20.0 mmol. Besides, its composition is different from the other samples. The disappearance of the diffraction peak at 2θ=12.2° can be detected in Fig. 5(a₅), and it obviously shifts to low diffraction angle. This sample is more likely composed of Ni₃Si₄O₁₀(OH)₂, indicating that excessive urea makes more TEOS participate in the reaction.

The as-synthesized Ni₃Si₂O₅(OH)₄ microspheres were applied as adsorbent for the removal of BF from simulated waste water. BF is a kind of cationic dye with the molecular structure as shown in Fig. 7(a). Ni₃Si₂O₅(OH)₄ microspheres are negatively charged according to the Zeta potential measurements as shown in Fig. 7(b). Thus, charge-charge interaction contribute to the adsorption of BF on Ni₃Si₂O₅(OH)₄ microspheres. At different pH, the variations of the adsorption rate and adsorption capacity as a function of contact time with the Ni₃Si₂O₅(OH)₄ microspheres dosage of 20 mg in a 50 mg·L^-1 solution (50 mL) are displayed in Fig. 7(c). Obviously, the amount of adsorbed BF increased with the contact time increasing and then reached equilibrium. In the first 5 min, the removal efficiency ((1-c_t/c₀)×100%) was up to 63.2%, 69.0% and 76.8% at pH=4, 7 and 9, respectively, representing a quite rapid adsorption rate at the beginning of the contact between BF and the adsorbent. The increase of adsorption rate with the increase of pH can be attributed to the higher negative charge. After the first 5 min, the adsorption was significantly getting slow and gradually reached equilibrium. The ultimate adsorption capacity within 180 min was calculated to be 98.3, 120.7 and 111.5 mg·g^-1 at pH=4, 7 and 9, respectively. The fast adsorption at the initial stage might cause pore structure block, resulted in the decrease of adsorption capacity at pH=9.

The adsorption performances of the samples with different Ni/Si molar ratios are shown in Fig. 7(d). As aforementioned, the product was composed of Ni₃Si₄O₁₀(OH)₂ when the Ni/Si molar ratio was 0.25 : 1, and its ultimate adsorption capacity within 180 min was 100.8 mg·g^-1. With the increase of Ni/Si molar ratio, the adsorption capacity was greatly increased higher than 120 mg·g^-1 for the samples (Ni/Si molar ratio of 0.5 and 0.75), which was a very similar performance as on the sample (Ni/Si molar ratio of 1 : 1) (Fig. 7(c)). It shows that Ni₃Si₂O₅(OH)₄ possesses better adsorption property than Ni₃Si₄O₁₀(OH)₂. With the further increase of Ni/Si molar ratio to 1.25 and 1.5, the adsorption capacity was significantly decreased, which might be ascribed to the change in phase composition and the decrease of surface area as found by the XRD and BET measurement.

The adsorption kinetic for BF on Ni₃Si₂O₅(OH)₄ microspheres was also investigated. Two typical kinetic models, the pseudo-first-order (PFO) and pseudo-second- order (PSO) were applied to fit the adsorption data obtained from the time-dependent experiments. The linear fitting results are shown in Fig. S1, and the kinetic parameters as well as the determination coefficients (R²) obtained by linear regression are listed in Table S2. The R² was as high as 0.9979, and meanwhile the calculated q_e (118.5 mg·g^-1) was very close to the experimental value (120.7 mg·g^-1), indicating that the PSO model was suitable to describe the adsorption kinetic of BF on Ni₃Si₂O₅(OH)₄ microspheres. Therefore, it can be considered a chemical adsorption process.

To give better insight into the removal efficiency of BF on Ni₃Si₂O₅(OH)₄ microspheres, comparison with the results in the literature using other adsorbents is made and the results are shown in Table S3. Various materials have been used as the adsorbents for the removal of BF from solution, including modified clay, molecular sieves, polymers, oxides/hydroxide, and many of them are composite materials. Table S3 reveals that Ni₃Si₂O₅(OH)₄ microspheres have a higher adsorption capacity than many adsorbents reported in recent years. Besides, they are also competitive in view of the simple preparation process. As comparison, Ni₃Si₂O₅(OH)₄ was also prepared by using SiO₂ templates according to the literature^[19], and applied as adsorbent for BF removal. Under the same condition, its adsorption capacity was 101.0 mg·g^-1, less than that on the Ni₃Si₂O₅(OH)₄ microspheres in this study. As layered silicates, Ni₃Si₂O₅(OH)₄ microspheres have a much higher adsorption capacity than layered clay derived materials, such as alkali-activated diatomite, hydroxy-aluminum pillared bentonite, iron-manganese oxide coated kaolinite. A negative charge on the surface of the adsorbents is very beneficial for the adsorption of cationic dyes. And the pore structure and surface area also have significant effects on the adsorption capacity. In natural clays, the high negative charged framework is mainly balanced by alkali and alkali-earth cations, which limits the adsorption of cationic dyes. In hydroxy- aluminum pillared bentonite, the alkali and alkali-earth cations were replaced by inorganic hydroxy-metal polycations, bringing about lower surface potential and therefore contributing to the increase in adsorption capacity. Ni₃Si₂O₅(OH)₄ microspheres exhibited more negative Zeta potential comparing with hydroxy- aluminum pillared bentonite and iron-manganese oxide coated kaolinite, indicating a stronger charge-charge interaction with BF. Furthermore, the hierarchical porous structure of Ni₃Si₂O₅(OH)₄ microspheres rendered a much larger pore volume than hydroxy-aluminum pillared bentonite and alkali-activated diatomite.

The adsorption capacity of BF on Ni₃Si₂O₅(OH)₄ microspheres as function of the BF concentration in the solution at equilibrium conditions was investigated, and the Langmuir and Freundlich isotherm models were used to analyze the experimental data for the understanding of the adsorption mechanism. The experimental data of C_e and q_e are plotted in Fig. S2 and fitted according to the Langmuir and Freundlich isotherm models. The fitting parameters are listed in Table S4. As shown, when fitted with Freundlich model, the determination coefficient R² was 0.9919, much higher than that fitted with the Langmuir model(R² =0.7920), which revealed that the adsorption of BF on Ni₃Si₂O₅(OH)₄ microspheres was more consistent with the theory of multilayer adsorption. The 1/n value was 0.1678, indicating that the surface was heterogeneous^[24] and the adsorption strength was strong.

When NiCl₂ and TEOS are used as the raw materials to prepare nickel phyllosilicate microspheres by the urea-assisted hydrothermal method, the Ni/Si molar ratio has a significant impact on the composition of the products. The suitable adding amount of TEOS is in excess of 50% than the stoichiometry of Ni₃Si₂O₅(OH)₄ to obtain pure-phase product. Compared with other alkali source, urea contributes to the assembly of the hierarchical porous microspheres because of the gradual release of OH^- ions and the CO₂ bubbles as soft-template. Ni₃Si₂O₅(OH)₄ microspheres have been proved to be an efficient adsorbent for removal of BF. The highly negative Zeta potential and the large pore volume are believed to be the significant factors for the strong adsorption strength and high adsorption capacity.

Supporting materials related to this article can be found on the https://doi.org/10.15541/jim20210063.

Synthesis of Hierarchical Porous Nickel Phyllosilicate Microspheres as Efficient Adsorbents for Removal of Basic Fuchsin

WANG Tingting, SHI Shumei, LIU Chenyuan, ZHU Wancheng, ZHANG Heng

(School of Chemistry and Chemical Engineering, Qufu Normal University, Qufu 273165, China)

BF solution: 50 mg·L^-1, 50 mL; Ni₃Si₂O₅(OH)₄ microspheres (Ni/Si molar ratio of 1 : 1) : 20 mg. The linear regression equations for PFO and PSO models are expressed as Eqs. (1) and (2), respectively. In the equations, q_t (mg·g^-1) is the adsorption capacity at the corresponding time t (min) obtained from the experiments. And q_e (mg·g^-1), k₁ and k₂ (g·mg^-1·min^-1) are the fitting parameters, among which q_e refers to the calculated equilibrium adsorption capacity

The Langmuir and Freundlich isotherm models are expressed as Eqs. (3) and (4), respectively. In the equations, C_e (mg·L^-1) and q_e (mg·g^-1) are the concentration of the adsorbate in solution and its amount on the adsorbent at equilibrium, respectively. For Langmuir isotherm model, q_m (mg·g^-1) is the theoretical maximum adsorption capacity corresponding to entire monolayer adsorption, and b (L·mg^-1) is the equilibrium constant connected with the energy of adsorption. For Freundlich isotherm model, which indicates the nonideal multilayer adsorption on active sites with nonuniform distribution of adsorption heat and affinities throughout the heterogeneous surface, K_f and 1/n are the isotherm constants, representing the adsorption capacity at unit concentration and the adsorption strength, respectively.

References:

[1] ZHAO YH, GENG JT, CAI JC, et al. Adsorption performance of basic fuchsin on alkali-activated diatomite. Adsorption Science & Technology, 2020, 38(5/6): 151-167.

[2] IBRAHIM A G, SAYED A Z, ABD EL-WAHAB H, et al. Synthesis of a hydrogel by grafting of acrylamide-co-sodium methacrylate onto chitosan for effective adsorption of Fuchsin basic dye. International Journal of Biological Macromolecules, 2020, 159: 422-432.

[3] YUAN J, QIU F, LI P. Synthesis and characterization of β-cyclodextrin-carboxymethyl cellulose-graphene oxide composite materials and its application for removal of basic fuchsin. Journal of the Iranian Chemical Society, 2017, 14(9): 1827-1837.

[4] HAO Y, YAN L, YU H, et al. Comparative study on adsorption of basic and acid dyes by hydroxy-aluminum pillared bentonite. Journal of Molecular Liquids, 2014, 199: 202-207.

[5] KHAN T A, KHAN E A, SHAHJAHAN. Removal of basic dyes from aqueous solution by adsorption onto binary iron-manganese oxide coated kaolinite: non-linear isotherm and kinetics modeling. Applied Clay Science, 2015, 107: 70-77.

[6] NISTOR M A, MUNTEAN S G, MARANESCU B, et al. Phosphonate metal-organic frameworks used as dye removal materials from wastewaters. Applied Organometallic Chemistry, 2020, 34(11): e5939.

[7] BA MOHAMMED B, HSINI A, ABDELLAOUI Y, et al. Fe-ZSM-5 zeolite for efficient removal of basic Fuchsin dye from aqueous solutions: synthesis, characterization and adsorption process optimization using BBD-RSM modeling. Journal of Environmental Chemical Engineering, 2020, 8(5): 104419.

[8] LI X, XIE L, YANG X, et al. Adsorption behavior and mechanism of β-cyclodextrin-styrene-based polymer for cationic dyes. RSC Advances, 2018, 8(70): 40321-40329.

[9] SAKTI S C W, LAILY R N, ALIYAH S, et al. Re-collectable and recyclable epichlorohydrin-crosslinked humic acid with spinel cobalt ferrite core for simple magnetic removal of cationic triarylmethane dyes in polluted water. Journal of Environmental Chemical Engineering, 2020, 8(4): 104004.

[10] LI L, FAN L, LUO C, et al. Study of fuchsine adsorption on magnetic chitosan/graphene oxide. RSC Advances, 2014, 4(47): 24679-24685.

[11] AI L, JIANG J. Fast removal of organic dyes from aqueous solutions by AC/ferrospinel composite. Desalination, 2010, 262(1): 134-140.

[12] GUAN Y, WANG S, WANG X, et al. Preparation of mesoporous Al-MCM-41 from natural palygorskite and its adsorption performance for hazardous aniline dye-basic fuchsin. Microporous and Mesoporous Materials, 2018, 265: 266-274.

[13] MIAO J, ZHAO X, LI Y, et al. Facial preparation of hierarchical porous Ba(B₂Si₂O₈) microsphere by sacrificial-template method and its highly efficient selective adsorption of triphenylmethane dyes. Colloids and Surfaces A, 2020, 602: 124883.

[14] HUSSAIN D, SIDDIQUI M F, KHAN T A. Preparation of NiFe₂O₄/polythiophene nanocomposite and its enhanced adsorptive uptake of Janus green B and Fuchsin basic from aqueous solution: Isotherm and kinetics studies. Environmental Progress & Sustainable Energy, 2020, 39(3): e13371.