1. Introduction
Haloacetonotriles (HANs) are nitrogenous species of disinfection by-products (DBPs), which can be formed by reactions between chlorine, chloramine or bromine disinfectants and natural organic matter (NOM). HANs are of concern because they cause problems in wastewater and tap water. The disinfection of drinking water reduces microbial risks but increases chemical exposure to man. Increasing of human health risks are due to the formation of disinfection by-products (DBPs) in the organic and inorganic precursors. Monochloroacetronitrile (MCAN), dichloroacetonitrile (DCAN), trichloroacetonitrile (TCAN), bromochloroacetonitrile (BCAN), and dibromoacetonitrile (DBAN) are common HANs.
Several techniques can remove DBPs from water such as adsorption [1], ozonation [2] and membrane filtration [3]. Among these three processes, however, the adsorption process has been most popular for water treatment because of its low cost and simple application. Although different materials have been used to adsorb DBPs, a suitable one for use with tap water has not been identified. Consequently, a study on the removal HANs from tap water using modified adsorbents can be deemed useful.
Among porous materials, mesoporous silicas provide a number of advantages as adsorbents because of their high surface area and narrow pore size, which improves their adsorption capacity and adsorption selectivity. Moreover, hexagonal mesoporous silica (HMS) is one of many kinds of adsorbents that have been used. In the past, a polymer/silica composite has been studied as a new material that has the advantages of silica and organic polymer. Natural rubber (NR) is an organic material that many have attempted to modify because of its thermal stability and mechanical properties. A NR/HMS composite has been synthesized and characterized as an attractive prospective adsorbent.
Consequently, the objectives of this study are to investigate the adsorption efficiency of a thiol-functionalized mesoporous com posite made from natural rubber and hexagonal mesoporous silica (NR/HMS-SH) in haloacetonitrile (HANs) removal. In this study, dichloroacetonitrile (DCAN) was selected as the representative HAN because it is a chloro-DBP that is highly toxic and has a considerable presence in water. The modification of hexagonal mesoporous silica (HMS) with the thiol group (R-SH) and natural rubber was employed to investigate the effect of surface functional groups on the DCAN adsorption. The adsorption experiments were carried out in batch experiments. The adsorption kinetics and the adsorption isotherm were studied to investigate the adsorption mechanisms. Moreover, the ionic strength and pH of synthesized water was controlled so that it could adequately represent tap water in this study.
2. Materials and Methods
2.1. Preparation of NR/HMS-SH
2.1.1. Materials and reagents
TEOS (AR grade, >99%), MPTMS (AR grade, 95%), and DDA (AR grade, 98%) were purchased from Sigma-Aldrich (Germany), THF (AR grade, 99.5%), C2H5OH (AR grade, 99.9%), and H2SO4 (AR grade, 98%) were purchased from QRëC (New Zealand). The NR (commercial grade) was supplied by the Thai Hua Chumporn Natural Rubber Co., Ltd. (Thailand).
2.1.2. Synthesis of NR/HMS-SH
The thiol-functionalized mesoporous composite was made of natural rubber and hexagonal mesoporous silicate (NR/HMS-SH) according to the condition of NR/HMS-SO3H synthesis [4]. Firstly, 1 g of a natural rubber (NR) sheet was mixed with 30 mL of tetrahydrofuran (THF) at room temperature overnight. The NR sheet was completely dissolved in the THF to obtain a homogeneous solution. Secondly, 7.51 g of dodecylamine (DDA) was added in the solution and stirred. After 0.5 h, 21 g of tetraethoxysilane (TEOS) was added and stirred for another 0.5 h. Next, 106 g of H2O was added and stirred at 40°C for 0.5 h and then 4.96 g of 3-mercaptopropyltrimethoxysilane (MPTMS) was added into the mixture by stirring and then it was left to stand at 40°C for 1 h. The gel was aged at 40°C for 3 days, after which it was precipitated in 100 mL of ethanol. The solid product was vacuum dried at 60°C for 2 h. Finally, the template in the composite was removed with 0.05 M of H2SO4/EtOH that from the concentrated sulfuric acid was dissolved in the 99.9% ethanol at 70°C for 8 h and filtered; it was then vacuum dried with ethanol. The pH was checked at around 7 and dried at 80°C overnight.
2.2. Characterization of NR/HMS-SH
Structural information on the NR/HMS-SH composite was obtained using X-ray powder diffraction (XRD) analysis on X-ray power of 40 kV and 40 mA. The repeating distance between the pore centers of the hexagonal structure was calculated from the XRD data [5]. A Fourier-transform infrared (FT-IR) spectrometer equipped with a mercury cadmium telluride detector was applied for the identification of NR and the functional groups in the composite. A self-supporting disk (20 mm, 10–20 mg) was placed in the quartz cell attached to a conventional closed circulation system. All the IR spectra were recorded under evacuation at 25°C. A total of 64 scans over 400–4000/cm at a resolution of 4/cm was averaged for each spectrum.
The particle size of the adsorbent was taken by using a scanning electron microscope (SEM). The sample was fixed on an aluminum stub before it was observed by the SEM. The samples on the copper grids were observed without any metal coating.
2.3. Adsorption Study
Stock solutions of HANs were prepared in deionized water, and a phosphate buffer was used to adjust the pH of the solution to 7 and ionic strength (IS) to 10 mM. The kinetic study was prepared by varying the contact time from 0 to 36 h. DCAN was selected as the model HAN. The initial DCAN concentration was 1 mg/L at pH 7. After that, 0.025 g of the adsorbent and 50 mL of the DCAN solution were mixed in a 125 mL Erlenmayer flask and covered with aluminium foil. The mixture was stirred at 25oC and the supernatant solution was filtrated through a nylon syringe filter (pore size 0.45 mm). The concentration of DCAN was analyzed by gas chromatography with an electron capture detector (GC/ECD) according to EPA Method 551.1 [6]. The concentrations in the isotherm studies were varied from 0.05 to 2 mg/L in single solutions at pH 7, while the ionic strength, adjusted by a phosphate buffer, remain fixed at 10 mM. The contact time was obtained from the kinetic study. After reaching equilibrium, the mixture was separated using a nylon filter and the solution was analyzed by gas chromatography with an electron capture detector (GC/ECD).
3. Results and Discussion
3.1. Physiochemical Characteristics of the Synthesized Adsorbents
3.1.1. XRD analysis
Fig. 1 shows the XRD pattern of the NR/HMS-SH composite after the extraction of the template molecules. This material exhibited one diffraction peak at 2θ in the 2.2°, corresponding to the characteristics of the hexagonal porous structure.
3.1.2. FTIR spectroscopy
FTIR analysis was used to confirm the presence of NR and the SH functional group in the HMS structure of the NR/HMS-SH composites (Fig. 2). The stretching vibration of the silica framework (Si-O-Si) appeared between 1000 and 1300/cm. The broad band at around 3500/cm can be assigned to the free silanol group. The band related to the thiol group was observed at around 2550–2620/cm (S-H), but it was a very weak band.
3.1.3. Electron microscopy
The particle size of a NR/HMS-SH composite is shown in Fig. 3. The presence of NR/HMS-SH enhanced the agglomeration of HMS particles. It was observed that NR was homogenously dispersed throughout the particles of the NR/HMS composite.
3.2. Adsorption Kinetics
The kinetic curve for DCAN adsorption on the porous adsorbent is shown in Fig. 4. A large amount of DCAN was adsorbed over PAC and NR/HMS-SH, but with PAC equilibrium was reached after a short contact time (2 h). For NR/HMS-SH, the amount of adsorbed DCAN reached equilibrium at 12 h.
Kinetic modeling and pseudo-first-order and pseudo-second-order models were employed to investigate the adsorption mechanisms. Pseudo-first-order and pseudo-second-order equations can be defined as shown by Eqs. (1) and (2).
where qe is the amount of adsorbed contaminant at equilibrium (mg/g), qt is the amount of adsorbed contaminant at time t (mg/g), k1 is the rate constant of pseudo-first-order adsorption (min−1), and k2 is the rate constant of pseudo-second-order adsorption (g/mg·min).
Based on the pseudo-second-order model, the initial adsorption rate can be determined using Eq. (3).
In order to quantitatively compare the applicability of different kinetic models, a normalized standard deviation Δq (%) was used as shown in Eq. (4).
where N is the number of data points and qexp and qcal (mg/g) are the experimental and the calculated adsorption capacities. The best fit models should have the lowest Δq (%) values. The kinetic parameters were calculated and the results are shown in Table 1.
The data fitting curve for NR/HMS-SH is shown in Fig. 5. It displays three regimes; the first regime presents the external mass transfer in the boundary layer and the second presents the diffusion through the pores of the adsorbent. Moreover, the last step presents the DCAN adsorption on the internal site of the adsorbent. It occurred very quickly, so the rate limiting step of the adsorption must have happened during the first or the second step. Both steps together could thus be considered the controlling step.
3.3. Adsorption Isotherms
NR/HMS-SH is a functionalized silica-based material that had a comparable adsorption capacity to PAC. Moreover, the results indicate that increasing the specific surface area of the adsorbent cannot be used as an explanation for DCAN’s enhanced adsorption capacities when compared with the results of a previous study, which gave a similar specific surface area for HMS that functionalized with the thiol group without adding the natural rubber (SBET NR/HMS-SH of 880.4 m2/g). The results, however, can be used to explain the density of the functional group because NR/HMS-SH provided less density of sulfur than HMS in the previous study [7]. It can be argued that the natural rubber in the adsorbent may have affected the interaction that occurred between the functional group and HMS.
In order to model the adsorption mechanism, linear, Langmuir and Freundlich isotherm models were used to test the experimental adsorption process. The linear, Langmuir and Freundlich equations can be defined as shown in Eqs. (5), (6) and (7).
where qe is the amount of adsorbate adsorbed at equilibrium (mg/g), Kp is the linear constant (L/mg), qm is the maximum adsorption capacity (mg/g), and KL is the Langmuir constant (L/mg). KF and n are constants, Ce is the equilibrium concentration (mg/L), and Ce is the concentration of the adsorbate at equilibrium (mg/L).
4. Conclusions
A NR/HMS-SH composite with a high structure order and mesoporosity was successfully prepared. The formation of the NR/HMS-SH composite induced the coalescence of HMS nanoparticles, resulting in enhanced textural porosity and hydrophobicity. DCAN adsorption on the NR/HMS-SH adsorbent followed a pseudo-second order rate kinetic model. A linear model fit well with the DCAN adsorption isotherm at a low concentration level.