Ordered cubic mesoporous silica KIT-5 functionalized with carboxylic acid groups for dye removal

Abstract

Cubic cage-type KIT-5 mesoporous silica functionalized with carboxylic acid (–COOH) groups, ranging from 0 to 45 mol% based on silica, were synthesized via co-condensation of tetraethyl orthosilicate (TEOS) and carboxyethylsilanetriol sodium salt (CES) in the presence of Pluronic F127 in acidic media. The materials thus obtained were characterized by powder X-ray diffraction, nitrogen sorption, transmission electron microscopy, FTIR, thermogravimetric analysis, acid–base titration, and solid-state NMR measurements. The –COOH functionalized KIT-5 was used to adsorb different types of dyes such as phenosafranine (PF), methylene blue (MB), orange II (OII) and rhodamine B (RhB). The –COOH functionalized KIT-5 materials exhibited excellent adsorption capacities for cationic PF and MB, and amphoteric RhB in comparison to pure silica KIT-5, but the opposite for anionic OII. The adsorption was highly dependent on the electrostatic interaction between the adsorbent and the dye molecules. External mass transfer and intra-particle diffusion collectively control the adsorption process.

---

1 Introduction

Ordered mesoporous silica materials with high surface areas, large pore volumes, and uniform and tunable pore size distributions are promising candidates for use in the fields of adsorption, catalysis, sensors, semiconductors and separation.^1–3 Since mesoporous silicas such as MCM-41 and SBA-15 with hexagonal pore structures were reported, there have been considerable efforts devoted to the synthesis of different types of mesoporous silicas, especially for mesoporous silicas with cubic structures since they can avoid problems such as diffusion limitation and pore blocking. Among various cubic mesoporous silicas, KIT-5 is interesting due to its well-ordered face-centered cage-type cubic structure. Each cage is connected with 12 neighboring nanocages through small pore entrances.⁴ KIT-5 and FDU-12 have the same Fm3m symmetry and they are both templated by using Pluronic F127 in acidic media. In contrast to the synthesis of FDU-12, KIT-5 was synthesized in a weak acid medium without the use of swelling agent (trimethylbenzene, TMB) and inorganic salt (KCl). The synergetic roles of TMB and KCl are critical for the synthesis of FDU-12.⁵ Remarkable discrepancy is observed in their XRD patterns, indicating there are some differences in their pore configuration. Although KIT-5 exhibits excellent structural features, it suffers from poor acidity and hydrothermal stability, which severely hinders its potential applications in catalysis, separation, and adsorption. These demerits can be overcome by functionalization of KIT-5 with organic groups.⁶ However, incorporation of organic groups into mesoporous silicas might change their textural and morphological properties tremendously. Synthesis of highly organic group loaded mesoporous silicas is important as it determines many key properties of the materials.^7,8 The pore size distribution of the functionalized mesoporous silicas is affected by the synthesis conditions such as temperature, composition and reaction time.^5,9

Incorporation of organic groups in mesoporous silica materials can be achieved either by post-grafting or by the direct synthesis approach. The co-condensation (or direct synthesis) method is often preferred over post-grafting because the organic groups can be incorporated into mesoporous silicas with uniform distributions.^10–12 This one-pot synthesis route was used by Lu et al. to produce amine-functionalized FDU-12.¹³ Yang et al. has synthesized amine functionalized SBA-16 for controlled growth of metal nanoparticles.¹⁴ Cubic mesoporous silicas have been functionalized with thiol groups and used as supports to synthesize gold nanoparticles for aerobic oxidation of trans-stilbene.¹⁵ Although KIT-5 has an interesting cubic mesostructure, very limited literature was reported on the functionalization of KIT-5 with organic groups. Yang et al. has successfully synthesized aminopropyl-functionalized KIT-5 with an amino loading up to 20% by co-condensation of tetraethyl orthosilicate (TEOS) and 3-aminopropyltrimethoxysilane (APTMS).¹⁶ It still remains a challenge to develop a simple and direct technique for the synthesis of large pore KIT-5 functionalized with a high content of organic functional groups. The acidic group loaded mesoporous silica is thought to be more beneficial for its use in various acid-catalyzed reactions. Although there are a few reports available on the acidic functionalization of mesoporous silicas, most of them are related to only hexagonally structured mesoporous silicas.^6,17 Acid functionality in mesoporous silicas can be created either by modifying with trivalent metal ions or by loading acidic functional groups onto the surface of the pore walls. For example, Vinu et al. have reported the preparation of aluminum-supported KIT-5 with excellent textural characteristics.¹⁸ However, the acidity of the material was low. The surface of KIT-5 has been modified with a superacid like CF₃SO₃H to increase the acidity of the material.¹⁹ Recently, our group has synthesized highly ordered FDU-12 functionalized with high loadings of carboxylic acid groups (–COOH).²⁰ In this work, we report the one-pot synthesis of –COOH functionalized KIT-5 via co-condensation of TEOS and carboxyethylsilanetriol sodium salt (CES), an organosilane containing carboxylate moiety, using triblock copolymer Pluronic F127 as a structure directing agent under weak acidic conditions. There is a concern that condensation of TEOS and CES under different synthesis conditions may induce some unexpected phase transformation of mesostructures. Earlier, Garcia-Bennett et al. has reported the phase transformation from Pm3̄n to p6mm during the synthesis of amine functionalized mesoporous silicas.²¹ Since the synthesis conditions are different for FDU-12 and KIT-5, functionalization of them with the same organic functional groups could have different situations. There is still a need to explore this possibility for the case of the –COOH functionalized KIT-5. Since there are only few reports on the use of cubic mesoporous silicas as adsorbents, more investigation is required for their possible use in adsorption and other fields, especially for dye adsorption.

A large quantity of dyestuffs is produced in the world each year and about 1–20% of the total world production of dyes lost during dyeing process which appears in wastewater.^22,23 Removal of dyes from wastewater is very important as most of them are toxic and have adverse effects on aquatic as well as in human lives.^24,25 Dyes are generally removed from wastewater by adsorption,²⁶ coagulation,²⁷ oxidation,²⁸ reduction,²⁹ filtration,³⁰ and biological treatment.³¹ Adsorption has some advantages such as simplicity in operation, low cost in comparison to other separation methods and no sludge formation. Among the various types of adsorbents, activated carbon is the most widely used adsorbent for the removal of dyes from aqueous solutions.^32,33 However, the microporous structure of activated carbon may hamper its use for adsorption of bulky molecules due to its slow adsorption kinetics and low adsorption capacity. Alternately, mesoporous silicas are good candidates for use as adsorbents for wastewater treatment due to its pronounced advantages such as good periodicity, high surface area, large pore volume and uniform pore size distribution.^34–36 However, the high production cost and stability of mesoporous silicas are often thought as the main obstacles for their widespread industrial applications. Although further cost reduction is needed, mesoporous silicas can be greatly needed for many specific cases where removal of one or two problematic effluents from wastewater can save the cost of the treatment significantly.

The –COOH functionalized KIT-5 materials synthesized in this work were used as adsorbents for the adsorption of various dyes such as cationic phenosafranine (PF) and methylene blue (MB), anionic orange II (OII) and amphoteric rhodamine B (RhB) from aqueous solutions. The structures of the dye molecules investigated in this study are shown in Scheme 1. The presence of the –COOH functional group in KIT-5, along with its cage size, is expected to facilitate the interactions with the adsorbates. Therefore, the –COOH functionalized KIT-5 materials are of great interest for their use as adsorbents. Since the pore size is one of the key factors for efficient adsorption, this work also provides a chance to evaluate the factors governing the adsorption capacities between KIT-5 and FDU-12 with the same symmetry.

Scheme 1 Structures of (a) PF, (b) MB, (c) OII and (d) RhB.

2 Experimental

2.1 Materials

TEOS, F127, and dyes were received from Sigma-Aldrich. CES (25 wt% in water) was purchased from Gelest. All chemicals were used as received without any further purification.

2.2 Preparation of –COOH functionalized KIT-5

For the synthesis of –COOH functionalized KIT-5, the surfactant F127 (0.625 g) was first dissolved in a mixture of HCl (1.313 g, 12 M) and H₂O (30 mL) and the solution was stirred at 45 °C for 4 h. TEOS and CES was first pre-mixed and then added to the mixture to form a solution, which was stirred continuously for another 24 h at 45 °C. The reaction mixture was treated hydrothermally at 100 °C for 24 h. The resultant white precipitate was filtered, washed and air dried. To incorporate various contents of –COOH group, the molar composition of the reaction mixture was varied as 1−(x/100) TEOS: x/100 CES: 0.0035 F127: 0.88 HCl: 119 H₂O. The template removal was achieved by adding H₂SO₄ (100 mL, 48 wt%) and heating the mixture into the as-synthesized materials (0.3 g) at 95 °C for 24 h under continuous stirring. The product was filtered and washed with acetone and water, and then dried at 70 °C for 24 h. The final product was denoted as CK-x, where x is the molar percentage ratio of CES/(TEOS + CES), i.e., the functionalization level.

2.3 Characterization methods

Powder X-ray diffraction (XRD) patterns were collected on Wiggler-A beamline (λ = 0.133367 nm) at the National Synchrotron Radiation Research Center in Taiwan. N₂ adsorption–desorption isotherms were measured at 77 K on a Micromeritics ASAP 2020 analyzer. The cage size was determined from the adsorption branch of the N₂ isotherm using the nonlocal density functional theory (NLDFT) model for spherical pores.³⁷ Pore volumes were obtained from the volumes of N₂ adsorbed at P/P₀ = 0.95 or in the vicinity. Thermogravimetric analysis (TGA) was performed using a Perkin-Elmer TGA 7 apparatus at a heating rate of 10 °C min⁻¹ under a nitrogen flow. Transmission electron microscopy (TEM) was carried out on a JEOL JEM2100 microscope operating at 160 kV. Solid-state ¹³C and ²⁹Si NMR spectra were recorded on a Varian Infinityplus-500 NMR spectrometer with Larmor frequencies for ¹³C and ²⁹Si nuclei of 125.7 and 99.3 MHz, respectively. ¹³C CPMAS NMR spectra were acquired at a spinning speed of 9 kHz and at a contact time of 1 ms. ²⁹Si MAS (magic angle spinning) NMR spectra were acquired with a recycle delay of 200 s. Both ¹³C and ²⁹Si chemical shifts were externally referenced to tetramethylsilane (TMS) at 0 ppm.

2.4 Adsorption kinetics

The adsorption kinetics of the various dyes onto CK-x were investigated at a dye concentration of 100 mg L⁻¹, pH 9 and at 25 °C in order to explore the effect of contact time on adsorption and to determine the kinetic parameters. The solution was shaken for different interval of time (0 to 600 min) and the amount of dye adsorbed (q_t) on adsorbent at time t was measured as.where C₀ is the initial concentration of dye and C_t is the concentration of the dye at time t, V is the volume (L), and m is the mass of the adsorbent (g).

2.5 Adsorption studies

The batch adsorption method was used to study the adsorption behavior. A series of aqueous dye solutions (10 to 300 mg L⁻¹) were prepared by dissolving the dye in double distilled water. In each adsorption experiment, 5 mg adsorbent was added in 5 mL dye solution and shaken at 25 °C in 300 rpm using a rotating oscillator incubator. As the density of siliceous materials is around 2.2 g cm⁻³, the mesoporous silica samples in the powder form were immersed at the bottom of the sample container, which confirmed good wettability of CK-x. The solution was then centrifuged to separate the adsorbed dye and the residual amount of dye in the solution was measured using UV-Vis spectrometer (T 90+). The amount of dye adsorbed (q_e) was calculated as:where C₀ and C_e are the initial and equilibrium concentrations of the dye solution (mg L⁻¹). The influence of pH in the adsorption process was studied by carrying out the experiments at pH 3 and pH 9. The pH of the dye solution was adjusted by using 0.01 M NaOH and 0.01 M H₂SO₄.

3 Results and discussion

3.1 Structural ordering of CK-x as a function of CES content

The small angle powder XRD patterns of the as-synthesized and template-extracted CK-x (x = 0–50) are displayed in Fig. 1. The patterns showed three diffractions peaks in the region of 2θ = 1–3°, which can be indexed to the (111), (200) and (220) reflections of a face centered cubic structure of Fm3m symmetry, similar to the XRD patterns of KIT-5 reported in the literature.¹⁶

Fig. 1 XRD patterns of (A) as-synthesized and (B) template-extracted CK-x, where x = (a) 0, (b) 10, (c) 20, (d) 30, (e) 40, (f) 45, and (g) 50.

As shown in Fig. 1, the structural ordering was decreased with increasing the CES contents in the reaction mixture. This structural degradation can be attributed to the different hydrolysis and condensation rates of the silica precursors used, i.e., TEOS and CES. The most dominant (111) peak disappeared when the CES content exceeds 45% in the reaction mixture. No phase transformation between different mesostructures was observed. The XRD peak positions were slightly shifted after template removal, indicating little shrinkage of the cell dimensions of the materials. The structural ordering of the template-extracted sample was slightly less than the as-synthesized samples, which could be attributed to the use of highly corrosive sulphuric acid for template removal. In general, an HCl–ethanol mixed solution is used for template removal after the synthesis of mesoporous silicas functionalized with organic groups. In the case of –COOH functionalized mesoporous silicas, however, this method will result in the formation of undesirable carboxylic ester (–COOC₂H₅) due to the reaction of carboxylic acid and ethanol. Therefore, sulphuric acid is employed to remove the template F127 via direct dissociation of F127 without forming esters.²⁰ The unit cell parameter a₀ was determined from the d₁₁₁ spacing and the results are summarized in Table 1.

Table 1 Structural and textural properties of CK-x with various CES/(TEOS + CES) ratios^a

X	d111 (nm)	a0 (nm)	ABET (m^2 g^−1)	DNLDFT (nm)	Vmi (cm^3 g^−1)	Vt (cm^3 g^−1)
a a0: cell parameter; ABET: surface area; Vt: total pore volume; DNLDFT: cage size estimated by NLDFT; Vmi: micro/mesopores volume. The numbers in parentheses are obtained from the as-synthesized samples.
0	12.3 (12.3)	21.3 (21.3)	660	8.7	0.02	0.74
10	11.9 (12.3)	20.6 (21.3)	606	8.3	0.01	0.56
20	11.9 (12.3)	20.6 (21.3)	612	7.5	0.01	0.48
30	11.9 (12.3)	20.6 (21.3)	612	7.5	0.01	0.57
40	11.6 (11.6)	20.1 (20.1)	687	6.4	0.02	0.53

---

It has been found that the order of mesoporous materials prepared with the co-condensation method is highly dependent on the concentration of the precursors. In the present case, the triol sites of CES can co-condense with TEOS and assembled to form the ordered silica network up to the CES/(CES + TEOS) ratio of 45% during the synthesis. With the increase in the CES contents over 45%, the large amount of negatively charged carboxylate groups from CES causes the difference in the charge density on the wall surface, which significantly affects the interaction between the surfactant micelle and silicate, and thus interrupts the pore formation at the end.

3.2 Textural and structural studies

The N₂ adsorption–desorption isotherms of the template-extracted CK-x (x = 0–45) samples are shown in Fig. 2. All the isotherms are of type IV with a sharp capillary condensation step at high relative pressures and a broad H2 hysteresis loop. The width and height of the hysteresis loops decreased with the increase in the CES contents, suggesting the reduction of the cage size of CK-x with the increase in functional group loading. The shape of the N₂ adsorption–desorption isotherm changed when the contents of –COOH groups was increased due to the partial blockage of the pores, particularly at the pore entrance.

Fig. 2 N₂ adsorption–desorption isotherms of template-extracted CK-x, where x = (a) 0, (b) 10, (c) 20, (d) 30, (e) 40, and (f) 45.

The hysteresis loops closed sharply at a relative pressure of ∼0.40, corresponding to the lower limit of the adsorption hysteresis, which indicated that desorption occurred via cavitation and entrance sizes were smaller than 5 nm.^37,38 The cage size was estimated from the pore size distribution of the adsorption branch of the N₂ isotherm using the NLDFT model for spherical pores³⁷ and the results are presented in Table 1. The decrease in the cage size with the increase in the CES contents in the reaction mixture can be attributed to the occupation of the mesopores by the –COOH groups incorporated. As seen in Table 1, introduction of the –COOH group leads to a decrease in surface area and cage size, which gives support that the organic groups incorporated are located within the mesopores. However, the surface area and pore volume appeared to be relatively high for the –COOH loading up to 40%, but substantially decreased thereafter. The highly ordered –COOH functionalized KIT-5 materials are anticipated to serve as excellent adsorbents for adsorption applications.

The 3D cubic mesostructures of template-extracted CK-x (x = 30 and 40) were further confirmed by TEM. The TEM images along [100], [211] and [110] directions (Fig. 3) showed that the –COOH functionalized KIT-5 materials have a highly ordered cubic mesophase with Fm3m symmetry.

Fig. 3 TEM images of (A) CK-30 recorded along (a) [211] and (b) [100] directions and (B) CK-40 recorded along (a) [100] and (b) [110] directions (Fourier diffractograms shown as insets).

3.3 FTIR and TGA

FTIR was employed for the qualitative determination of the –COOH groups incorporated into KIT-5. The band at 1720 cm⁻¹ due to the C=O stretching vibrations of carboxylic species, as shown in Fig. 4, gives clear evidence of the presence of the –COOH groups. The intensity of this peak increases with the increase in the CES content, suggesting that the –COOH groups were incorporated into CK-x in a quantitative manner. The bands at 1080 cm⁻¹ and 802 cm⁻¹ correspond to the asymmetric stretching and symmetric stretching modes of the siloxanes. The other bands observed at 3460 and 1643 cm⁻¹ were due to the vibrational modes of water molecules. The weak peak at 960 cm⁻¹ was associated with the non-condensed Si–OH groups present in the samples.

Fig. 4 FTIR spectra of template-extracted CK-x, where x = (a) 0, (b) 10, (c) 20, (d) 30, (e) 40, and (f) 45.

The thermal stability of CK-x was investigated by TGA. The TGA and DTA curves of the as-synthesized CK-20 and template-extracted CK-x (x = 20–40) are shown in Fig. S1 (ESI†). A weight loss below 180 °C was due to the evaporation of physically adsorbed water. Another weight loss in the temperature range of 180–350 °C was mainly due to the decomposition of the surfactant F127. A major weight loss of about 20 wt% was evidenced in the temperature range of 300–450 °C for the decomposition of the –COOH groups. The weight loss due to the decomposition of carboxylic acid group was more pronounced with increasing the CES contents, as evidenced from the shift in the peak position from 495 to 400 °C in the DTA profiles of the template-extracted CK-x. The exact quantification of –COOH groups in CK-x by TGA was difficult due to severe overlapping of the weight losses between the surfactant F127 and –COOH groups. A further weight loss observed at temperatures higher than 500 °C could be mainly due to condensation of the silica walls. The TGA results manifested that the CK-x materials functionalized with the –COOH groups exhibited high thermal stability up to 400 °C.

3.4 ¹³C and ²⁹Si solid-state NMR

The successful incorporation of the –COOH groups in CK-x as well as the efficiency of the surfactant F127 removal were confirmed by solid state NMR spectroscopy. The ¹³C CPMAS NMR spectra of CK-x, shown in Fig. 5(A), revealed three resonances at 7, 27, and 178 ppm. The peak appeared at 7 ppm was assigned to the –CH₂ group (C3 carbon) linked directly to the silicon atom of CES. The resonances at 27 and 178 ppm were assigned to the –CH₂ (C2 carbon) and –C=O of the –COOH group, respectively, which confirmed the presence of the –COOH group in CK-x. The ¹³C cross-polarization magic angle spinning (CPMAS) NMR spectra for all the samples were similar as the signals were from CES only. The results of elemental analysis indicated that the carbon contents of CK-0, CK-10, CK-20, CK-30 and CK-40 were 0.02, 0.31, 0.49, 0.53 and 0.56 mmol mg⁻¹, respectively, which come from the residual F127 and carboxylic acid moiety. No peak due to the surfactant F127 was observed, indicative of complete removal of surfactant.

Fig. 5 (A) ¹³C CPMAS and (B) ²⁹Si MAS NMR spectra of template-extracted CK-x, where x = (a) 10, (b) 20, (c) 30, and (d) 40.

²⁹Si MAS NMR spectroscopy is a more direct and quantitative tool than TGA for measuring the degree of organic functionalization in CK-x. The ²⁹Si NMR spectra presented in Fig. 5(B) depict three peaks at −96, −103, −113 ppm, which can be assigned to the Q², Q³ and Q⁴ species of the silica framework (i.e., Qⁿ = Si(OSi)_n(OH)_4–n, n = 2–4). Additional peaks observed around −53, –61 and −70 ppm confirmed the presence of T¹[–CSi(OH)₂(OSi)], T²[–CSi(OH)(OSi)₂] and T³[(–CSi(OSi)₃)] species, respectively. The peak intensities of the T species increased with the increase in the CES content. Meanwhile, the relative intensities of the Q species decreased, suggesting further condensation of silanol groups in the pore wall due to the presence of large amounts of CES. Each ²⁹Si MAS NMR spectrum was deconvoluted to obtain the quantitative functionalization level in CK-x, and the results are listed in Table S1 (ESI†). The molar ratio of T^m/(T^m + Qⁿ) thus obtained agreed relatively well with the composition used in the initial mixture. The amount of incorporated –COOH groups estimated by pH titration (Fig. S2, ESI†) was 10.0 mol% in CK-10, 20.0 mol% in CK-20, 22.7 mol% in CK-30 and 33.6 mol% in CK-40. The lower amounts of –COOH groups obtained by the pH titration method suggested that some –COOH groups might be buried inside the silica matrix.

3.5 CK-x as dye adsorbent

3.5.1 Effects of contact time and pH. For an adsorption process, the equilibrium time between the dye and the adsorbent is important in wastewater treatment. The contact time for reaching equilibrium is dependent on the initial dye concentration. In addition, the adsorption capacity also increases with the initial concentration.³⁹ It has been proposed that the adsorption rate should be proportional to a driving force times surface area.⁴⁰ An efficient adsorbent means a rapid uptake of the dyes and a short time to reach the equilibrium when it is used for wastewater treatment. Due to the highly ordered structure, large pore volume, large cage size, CK-30 was chosen to study the adsorption behaviour. The effects of contact time on the adsorption of PF, MB, OII and RhB onto CK-30 are illustrated in Fig. 6. It is clear that a rapid dye uptake was observed in the initial 5 min, then proceeds at a slower rate within 10–150 min, and finally reached equilibrium at the end of 300 min. The initial rapid uptake can be attributed to the chemisorptions processes, and the slow rate thereafter was probably due to the longer time period required by the dye molecules to access all the reactive sites within the cubic structure of CK-x. The mechanism of dye adsorption onto the adsorbent can be analyzed by several kinetic models, for example, pseudo-first order and pseudo-second order kinetic models. These models were tested to identify the step possibly controlling the adsorption of the dyes onto CK-x in this work.

Fig. 6 Effects of contact time and pH on the adsorption of (a) PF, (b) MB, (c) OII, and (d) RhB by CK-0 and CK-30.

A linear form of the pseudo-first order kinetics can be expressed as follows,⁴¹

ln(q_e − q_t) = ln(q_e) − k₁t (3)

where q_e and q_t are the adsorption capacities at equilibrium and at time t, respectively, and k₁ is the rate constant of pseudo-first order adsorption. The pseudo-second order equation, on the other hand, can be represented as eqn (4),⁴²where k₂ is the pseudo-second order constant. As revealed by eqn (4), the equilibrium adsorption capacity q_e can be determined from the slope and intercept of t/q_t versus t plot. Table 2 summarizes the measured rate constants and the equilibrium adsorption capacities determined from the kinetic models. Clearly, the pseudo-second order kinetic model delivered a better correlation for the adsorption of dyes on CK-30, as compared to the pseudo-first order kinetic model. In the adsorption of adsorbate by an adsorbent, several steps are involved, including transport of the adsorbate molecules from the aqueous phase to the adsorbent surface and diffusion of the adsorbate molecules into the interior of the pores. Since the kinetic results are better fitted to a chemisorption model, the intra-particle diffusion model was employed in order to understand the influence of mass transfer resistance on the binding of the dye molecules.

Table 2 Kinetics parameter for adsorption of PF, MB, OII and RB onto CK-30

Dye	Pseudo-first order model			Pseudo-second order model			Intra-particle diffusion model
	k1 (min^−1)	qe,cal (mg g^−1)	R^2	k2 (g mg^−1 min^−1)	qe (mg g^−1)	R^2	kid (mg g^−1 min^−1/2)	R^2	C
PF	0.001	1.51	0.232	4.6 × 10^−5	500	0.962	8.9	0.979	134
MB	0.002	1.57	0.231	1.2 × 10^−4	454	0.999	16.0	0.972	205
OII	0.001	1.43	0.227	1.4 × 10^−4	333	0.993	8.2	0.989	122
RhB	0.001	1.44	0.227	1.6 × 10^−4	270	0.992	14.8	0.996	70

---

According to Weber and Morris, the intra-particle diffusion rate constant (k_id) is given by the following equation,⁴³

q_t = k_idt^0.5 + C (5)

The intra-particle diffusion rate constant can be determined from the slope of the q_t versus t^0.5 plot. Fig. S3 (ESI†) shows the amount of dye adsorbed versus the square root of adsorption time for intra-particle transport of the four dyes using CK-30 as the adsorbent. As noticed, the plots present a multilinearity, which indicated that three steps occurred in the process simultaneously. The first, sharper portion (A) was attributed to the diffusion of adsorbate through the solution to the external surface of adsorbent or the boundary layer diffusion of dye molecules. The second portion (B) described the gradual adsorption stage, in which intra-particle diffusion was the rate limiting step. The third portion (C) was attributed to the final equilibrium stage where intra-particle diffusion started to slow down due to low dye concentrations in the solution. The intra-particle diffusion (k_id) determined from the q_t versus t^0.5 plots are presented in Table 2. The intra-particle rate constant k_id was measured to be lower than the film diffusion (k_fd) for all the four dyes, indicating that the intra-particle diffusion is the rate limiting step. The C values, obtained from the intercepts of the plots, indicate that the line did not pass through the origin, and hence intra-particle diffusion is not the only rate limiting mechanism. It can be concluded that both the external mass transfer and intra-particle diffusion may control the overall adsorption process, but intra-particle diffusion played a predominant role in controlling the adsorption process.

The surface charge of the adsorbent, along with the degree of ionization of the materials present in the solution is affected by the solution pH. Therefore, adsorption of dye molecules onto an adsorbent is highly pH dependent. Increasing the solution pH will favor the dissociation of the –COOH functional groups on the surface active sites of the adsorbent CK-x, which in turn affects the adsorption process. To study the effect of pH on the adsorption capacity of CK-x, the dye solutions were prepared at two different pH values of 3 and 9. As seen in Fig. 6(a and b), PF and MB uptakes increased as the pH value was increased from 3 to 9. At a higher pH, the –COOH groups in CK-30 were present in the form of –COO⁻, and hence exhibited higher adsorption capacity due to the strong electrostatic attraction between the cationic dyes (PF and MB) and the negatively charged surface of CK-30. The enhanced adsorption capacity with the increase in pH was also due to the competition between the cationic dyes and the excess OH⁻/H⁺ ions in the solution. On the other hand, the dye uptake for the anionic OII dye adsorption was the highest on the adsorbent CK-0 at pH 3, as can seen from part (c) of Fig. 6. The higher adsorption of anionic dye by CK-0 in comparison to CK-30 at lower pH could be well explained by the protonation properties of the adsorbent. Since there is no organic functionalization in CK-0, more silanol groups (Si–OH) on the mesopore of CK-0 are expected. The higher hydrogen ion concentration at low pH created more protonated Si–OH groups (i.e., Si–OH₂⁺) on the mesopore surface of CK-0. As a result, the surface provides active sites with positive charges for adsorption of anionic dye OII. Moreover, the higher surface area and pore volume of CK-0, in comparison to CK-30, are also beneficial for adsorption of anionic dye at low pH. However, the RhB dye uptake reached the maximum when CK-30 was used as the adsorbent and the pH value was 3, as shown in part (d) of Fig. 6. RhB is an aromatic amino acid with amphoteric characteristics due to the presence of both amino (–NHR₂) and –COOH groups, and hence the charge state of RhB is dependent on the pH of the solution. The pK_a value for the –COOH groups present on RhB molecule is about 4. When the solution pH was 3, the RhB ion took on a positive charge on one of the nitrogen atoms, while the –COOH group was unionized. The electrostatic attraction between the positively charged RhB ion and negatively charged –COO⁻ groups on CK-30 lead to the increase in adsorption capacity at pH 3. When the solution pH was 9, the ionization of the –COOH group in RhB took place and led to the formation of the zwitterions of RhB.⁴⁴ The zwitterions of RhB in water may increase the dimerization of RhB, which made the RhB dimer molecule too large to enter the pores of CK-30. As a result, the adsorption capacity of CK-x toward RhB was lower at pH 9. The highest adsorption capacities of 428 and 462 mg g⁻¹ were obtained for PF and MB adsorbed onto CK-30 at pH 9, respectively, 290 mg g⁻¹ for OII onto CK-0 at pH 3, and 270 mg g⁻¹ for RhB onto CK-30 at pH 3.

3.5.2 Adsorption isotherm studies. Adsorption studies were performed on MB to determine the equilibrium adsorption capacity of CK-x. The equilibrium adsorption isotherms (Fig. 7) were recorded under equilibrium conditions for six different concentrations of MB. The adsorption was increased dramatically with the increase in the dye concentration because more adsorbate molecules can compete for the available binding sites on the surface of the adsorbent at high dye concentrations. It was also observed that CK-x exhibited much higher saturated adsorption capacities than the conventional adsorbents such as activated carbon (80–300 mg g⁻¹)⁴⁵ and the previously reported carboxylic-acid-functionalized mesoporous silica (159 mg g⁻¹),⁴⁶ which can be attributed to the presence of active –COOH groups, 3D cubic pore connectivity, and well-defined cage structure. The 3D cubic pore connectivity allows the transport and diffusion of dye molecules from the aqueous phase to the binding sites of CK-x more homogenous, and thus facilitates the accessibility of the dye into the cage. Most importantly, the adsorption capacity for adsorbate molecules depended on the surface chemistry of the absorbent CK-x, as the presence of –COOH significantly enhanced the dye adsorption as compared to the sample without the –COOH group (i.e., CK-0). The favorable interfacial interaction between the dye and the adsorbent was believed to be the major reason for the adsorption enhancement. The adsorption efficiency of each active site (–COOH) in the samples is evaluated and the results are presented in Table S2 (ESI†). Although CK-40 exhibited the maximum amount of –COOH groups, its less ordered structure with poor textural properties inhibited the effective adsorption. It can be concluded that adsorption efficiency depends not only on the contents of –COOH groups, but also on the structural properties such as surface area, since CK-0 (without –COOH groups) also exhibits a significant adsorption capacity. The MB adsorption capacity of CK-30 was higher than our recently reported MB adsorption capacity of –COOH functionalized FDU-12 (i.e., FTC-30).²⁰ The enhanced dye adsorption capacity of CK-x can be attributed to its higher surface area (611 m² g⁻¹), as compared to FTC-30 (377 m² g⁻¹), which provided a larger number of active sites for interaction between dye and adsorbent.

Fig. 7 Effect of initial concentration of methylene blue onto (a) CK-0, (b) CK-10, (c) CK-20, (d) CK-30, and (e) CK-40.

Both Langmuir⁴⁷ and Freundlich⁴⁸ isotherm models were employed to analyze the experimental equilibrium adsorption data. In principle, the Langmuir isotherm model assumes that the surface of the adsorbent material is homogeneous and contains a number of active sites with the same activation energy. The Langmuir equation can be expressed as:

where q_e is the equilibrium amount of dye adsorbed on adsorbent (mg g⁻¹), C_e is the concentration of adsorbate under the equilibrium condition (mg L⁻¹), q_max is the maximum adsorption capacity (mg g⁻¹) and K_L is the adsorption equilibrium constant (L mg⁻¹). The corresponding linear Langmuir transforms of the isotherms C_e/q_e vs. C_e are shown in part A of Fig. S4 (ESI†). The maximum adsorption capacity q_max and Langmuir constant were determined from the slope and intercept of the plot and the calculated parameters are given in Table S3 (ESI†).

The Freundlich isotherm, on the other hand, is an empirical equation based on a heterogeneous surface. The Freundlich isotherm is expressed by the following equation:

q_e = K_FC^1/n_e (7)

where K_F (mg^(1−1/n) L^1/n g⁻¹) and n are the Freundlich constants, which are related to the adsorptive capacity and the adsorption intensity, respectively. The magnitude of n is an indication of system suitability for different adsorption conditions. For instance, n > 1 corresponds to favorable adsorption; n = 1, linear adsorption, and n < 1 represents unfavorable adsorption. The Freundlich parameters K_F and n can be determined (Table S3, ESI†) from the plot of log C_e versus log q_e (Fig. S4, ESI†). As seen in Table S3 (ESI†), the parameter n with values greater than unity (n = 2–7) indicates a favorable adsorption condition.

As the Langmuir model exhibits higher correlation coefficients than the Freundlich model, it suggests that the former model can predict the MB adsorption onto CK-x more accurately. The experimental maximum adsorption capacity was close to the theoretical maximum adsorption capacity determined by using the Langmuir equation, which confirmed the fact that MB adsorption onto CK-x takes place via a monolayer adsorption process.

Methylene blue adsorbed CK-30 (0.2 g) was washed with 2 M HCl to assess the possibility of regeneration and reuse of the synthesized materials. The study showed more than 99% removal efficiency after use for 3 times (Fig. S5, ESI†). The high MB removal efficiency of regenerated CK-30 was due to the easy protonation of functionalized –COOH groups at a lower pH value and probably similar numbers of active sites were available in the subsequent cycle for the dye adsorption. The regeneration capacity of CK-30 was remarkably higher than the conventionally used activated carbons.

4 Conclusions

Synthesis of highly carboxylic acid functionalized, cage-type mesoporous silica KIT-5 with cubic Fm3m symmetry and its application as adsorbent for the removal of various dyes is presented. The –COOH functionalized KIT-5 was synthesized by co-condensation of TEOS and CES using Pluronic F127 as the structure directing agent in acid media. The structural ordering of the –COOH functionalized KIT-5 decreased with increasing the CES content in the synthesis mixture. The adsorption study showed that the –COOH functionalized KIT-5 materials could serve as highly efficient adsorbents for PF, MB, and RhB. The electrostatic interaction between the cationic dyes PF and MB and –COO⁻ groups, which were de-protonated at higher pH, made the adsorption more favorable. On the other hand, the protonation of –COOH groups at a lower pH exhibited an adverse effect for the adsorption for the anionic dye OII. The prepared materials exhibited a higher RhB adsorption capacity at a lower pH due to the amphoteric nature of the dye. Therefore, it can be concluded that the adsorption capacity depends not only on the surface area of the adsorbent, but also on the specific interaction between the functional groups and the adsorbates. The adsorption kinetic results showed that the overall dye adsorption process was jointly controlled by the external mass transfer and intra-particle diffusion.

Notes and references

1. C. T. Kresge, M. E. Leonowicz, W. J. Roth, J. C. Vartuli and J. S. Beck, Nature, 1992, 359, 710–712.
2. D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka and G. D. Stucky, Science, 1998, 279, 548–552.
3. Y. Deng, D. Qi, C. Deng, X. Zhang and D. Zhao, J. Am. Chem. Soc., 2008, 130, 28–29.
4. F. Kleitz, D. Liu, G. M. Anilkumar, I. -S. Park, L. A. Solovyov, A. N. Shmakov and R. Ryoo, J. Phys. Chem. B, 2003, 107, 14296–14300.
5. J. Fan, C. Z. Yu, J. Lei, Q. Zhang, T. C. Li, B. Tu, W. Z. Zhou and D. Y. Zhao, J. Am. Chem. Soc., 2005, 127, 10794–10795.
6. J. A. Melero, R. Van Grieken and G. Morales, Chem. Rev., 2006, 106, 3790–3812.
7. A. Stein, B. J. Melde and R. C. Schroden, Adv. Mater., 2000, 12, 1403–1419.
8. A. Sayari and S. Hamoudi, Chem. Mater., 2001, 13, 3151–3168.
9. Q. Wei, Z.-R. Nie, Y.-L. Hao, L. Liu, Z.-X. Chen and J.-X. Zou, J. Sol-Gel Sci. Technol., 2006, 39, 103–109.
10. M. E. Davis, Nature, 2002, 417, 813–821.
11. F. Schüth and W. Schmidt, Adv. Mater., 2002, 14, 629–638.
12. F. Hoffmann, M. Cornelius, J. Morell and M. Fröba, Angew. Chem., Int. Ed., 2006, 45, 3216–3251.
13. S. B. Hartono, S. Z. Qiao, K. Jack, B. P. Ladewig, Z. Hao and G. Q. Max Lu, Langmuir, 2009, 25, 6413–6424.
14. Y. Hao, Y. Chong, S. Li and H. Yang, J. Phys. Chem. C, 2012, 116, 6512–6519.
15. A. Boullanger, S. Clément, V. Mendez, S. Daniele, C. Thieuleux and A. Mehdi, RSC Adv., 2013, 3, 725–728.
16. Y.-T. Hsu, W.-L. Chen and C.-M. Yang, J. Phys. Chem. C, 2009, 113, 2777–2783.
17. Y. Gu, A. Karam, F. Jerome and J. Barrault, Org. Lett., 2007, 9, 3145–3148.
18. P. Srinivasu, S. Alam, V. V. Balasubramanian, S. Velmathi, D. P. Sawant, W. Bohlmann, S. P. Mirajkar, K. Ariga, S. B. Halligudi and A. Vinu, Adv. Funct. Mater., 2008, 18, 640–651.
19. P. Kalita, B. Sathyaseelan, A. Mano, S. M. Javaid Zaidi, M. A. Chari and A. Vinu, Chem.–Eur. J., 2010, 16, 2843–2851.
20. W.-C. Chang, J. R. Deka, H.-Y. Wu, F.-K. Shieh, S.-Y. Huang and H.-M. Kao, Appl. Catal., B, 2013, 142–143, 817–827.
21. R. Atluri, Y. Sakamoto and A. E. Garcia-Bennett, Langmuir, 2009, 25, 3189–3195.
22. J.-W. Lee, S.-P. Choi, R. Thiruvenkatachari, W.-G. Shim and H. Moon, Dyes Pigm., 2006, 69, 196–203.
23. B. C. Oei, S. Ibrahim, S. Wang and H. M. Ang, Bioresour. Technol., 2009, 100, 4292–4295.
24. G. McKay, M. S. Otterburn and D. A. Aga, Water, Air, Soil Pollut., 1985, 24, 307–322.
25. A. R. Gregory, S. Elliot and P. Kluge, J. Appl. Toxicol., 1991, 1, 308–313.
26. K. C. Maheria and U. V. Chudasama, Ind. Eng. Chem. Res., 2007, 46, 6852–6857.
27. B. Shi, G. Li, D. Wang, C. Feng and H. Tang, J. Hazard. Mater., 2007, 143, 567–574.
28. F. L. Emmert III, J. Thomas, B. Hon and A. J. Gengenbach, Inorg. Chim. Acta, 2008, 361, 2243–2251.
29. H.-Y. Shu, M.-C. Chang, H.-H. Yu and W.-H. Chen, J. Colloid Interface Sci., 2007, 314, 89–97.
30. G. Capar, U. Yetis and L. Yilmaz, J. Hazard. Mater., 2006, 135, 423–430.
31. D. Georgiou and A. Aivasidis, J. Hazard. Mater., 2006, 135, 372–377.
32. M. Ghaedi, B. Sadeghian, A. A. Pebdani, R. Sahraei, A. Daneshfar and C. Duran, Chem. Eng. J., 2012, 187, 133–141.
33. S. Altenor, B. Carene, E. Emmanuel, J. Lambert, J. J. Ehrhardt and S. Gaspard, J. Hazard. Mater., 2009, 165, 1029–1039.
34. L.-C. Juang, C.-C. Wang and C.-K. Lee, Chemosphere, 2006, 64, 1920–1928.
35. W. Teng, Z. Wu, D. Feng, J. Fan, J. Wang, H. Wei, M. Song and D. Zhao, Environ. Sci. Technol., 2013, 47, 8633–8641.
36. T. X. Bui and H. Choi, J. Hazard. Mater., 2009, 168, 602–608.
37. P. I. Ravikovitch and A. V. Neimark, Langmuir, 2002, 18, 1550–1560.
38. J. C. Groen, L. A. A. Peffer and J. Perez-Ramirez, Microporous Mesoporous Mater., 2003, 60, 1–17.
39. M. Doğan, M. Alkan, Ö. Demirbas, Y. Özdemir and C. Özmetin, Chem. Eng. J., 2006, 124, 89–101.
40. Y. S. Ho, T. H. Chiang and Y. M. Hsueh, Process Biochem., 2005, 40, 119–124.
41. S. Lagergren, K. Sven. Vetenskapsakad. Handl., 1898, 24, 1–39.
42. Y. S. Ho and G. McKay, Process Biochem., 1999, 34, 451–465.
43. W. J. Weber Jr and J. C. Morris, J. Sanit. Eng. Div., Am. Soc. Civ. Eng., 1963, 89, 31–59.
44. A. V. Deshpande and U. Kumar, J. Non-Cryst. Solids, 2002, 306, 149–159.
45. K. Y. Ho, G. McKay and K. L. Yeung, Langmuir, 2003, 19, 3019–3024.
46. Z. Yan, S. Tao, J. Yin and G. Li, J. Mater. Chem., 2006, 16, 2347–2353.
47. I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361–1403.
48. H. M. F. Freundlich, Z. Phys. Chem., 1906, 57A, 385–470.

---

Footnote

† Electronic supplementary information (ESI) available: TGA and DTA curves, acid–base titration curves, intraparticle diffusion plot, Langmuir and Freundlich isotherm plots, removal efficiency plot, tables of deconvolution results of ²⁹Si MAS NMR and isotherm parameters. See DOI: 10.1039/c4ra08819k

---