Optimal synthesis of amino-functionalized mesoporous silicas for the adsorption of heavy metal ions

Optimal synthesis of amino-functionalized mesoporous silicas for the adsorption of heavy metal ions

Microporous and Mesoporous Materials 236 (2016) 250e259 Contents lists available at ScienceDirect Microporous and Mesoporous Materials journal homep...

2MB Sizes 0 Downloads 61 Views

Microporous and Mesoporous Materials 236 (2016) 250e259

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

Optimal synthesis of amino-functionalized mesoporous silicas for the adsorption of heavy metal ions Shiyou Hao a, Antonio Verlotta b, Paolo Aprea b, Francesco Pepe c, *, Domenico Caputo b, Weidong Zhu a, ** a

Key Laboratory of the Ministry of Education for Advanced Catalysis Materials, Xingzi College & Institute of Physical Chemistry, Zhejiang Normal University, 321004 Jinhua, People's Republic of China ACLabseApplied Chemistry Laboratories, Department of Chemical, Materials and Production Engineering, University Federico II, P.le V. Tecchio 80, 80125 Naples, Italy c Department of Engineering, University of Sannio, Piazza Roma 21, 82100 Benevento, Italy b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 April 2016 Received in revised form 3 September 2016 Accepted 7 September 2016 Available online 13 September 2016

Amino-functionalized mesoporous silicas (AFMS) were synthesized by a neutralization route using the anionic surfactant dodecanoic acid (DAA) as structure-directing agent (SDA), aminopropyltrimethoxysilane (APTMS) as co-structure-directing agent (CSDA), and tetraethoxysilane (TEOS) as silicon source. The synthesis parameters, which affect the structural properties and the amino loadings of the resultant AFMS, were optimized. Various techniques, such as FT-IR, XRD, N2 adsorption-desorption, and TEM, were used to characterize the synthesized AFMS. The selective removal of Cu2þ, Pb2þ, Cd2þ, and Zn2þ from aqueous solutions in single-, binary-, ternary-, and quaternary-component systems by the synthesized AFMS was thoroughly investigated. The measured single-component adsorption isotherms of Cu2þ, Pb2þ, Cd2þ, and Zn2þ on the AFMS optimally synthesized can be well described by the Sips model, in which the extracted adsorption capacities are 2.34, 2.86, 1.71, and 1.36 mmol/g (0.149, 0.593, 0.192, and 0.089 g/g) for Cu2þ, Pb2þ, Cd2þ, and Zn2þ, respectively, higher than those on other adsorbents reported in the literature. Furthermore, Pb2þ and Cu2þ can be more selectively removed by the synthesized adsorbent, compared to Cd2þ and Zn2þ, confirmed by the results on the multi-component adsorption. © 2016 Elsevier Inc. All rights reserved.

Keywords: Amino-functionalized material Mesoporous silica Heavy metal Multi-component adsorption Selective adsorption

1. Introduction It is well known that heavy metals present high toxicity to the ecosystem and human beings, and tend to be accumulated in living tissues [1]. Heavy metals, such as Pb, Cu, Cd, and Zn, are the main toxic pollutants in industrial wastewater, and they have been found to be the major contaminants of surface and ground water, causing various diseases and disorders. Among heavy metals, Pb is one of the most toxic elements, even at rather low concentrations, causing mental disturbance, retardation, and semi-permanent brain damage [2]. Although Cu is an essential nutrient for humans, animals, and microorganisms, excess Cu may produce many toxic and

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (F. Pepe), [email protected] (W. Zhu). http://dx.doi.org/10.1016/j.micromeso.2016.09.008 1387-1811/© 2016 Elsevier Inc. All rights reserved.

harmful effects in living organisms [3]. Typically, itching and dermatitis, gastrointestinal diseases, alteration of liver and kidney functionalities, Wilson disease, hypoglycemia, and dyslexia can be caused by excess Cu [4]. Exposure to high levels of Cd has been related to respiratory, cardiovascular and renal problems, and furthermore to the notorious “itai itai” disease [5]. The accumulation of Zn in the human body can cause dehydration, electrolyte imbalance, stomach ache, nausea, dizziness and incoordination in muscles [6]. In general, the effluents from several industrial processes are characterized by the simultaneous presence of some heavy metal ions with significant concentrations. In particular, Pb2þ, Cu2þ, Cd2þ, and Zn2þ can be simultaneously present in wastewaters from petroleum refining, fertilizer production, foundries, and pulp and paper production [7]. Consequently, it is necessary to find efficient methods to remove these heavy metals from the environment, even when present at very low concentrations.

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

Compared with other technologies for the removal of heavy metal ions, such as ion exchange, precipitation, membrane separation, reverse osmosis, sedimentation, and electro-dialysis [8,9], the adsorption-based separation is a highly-efficient, cost effective method [10,11]. Up to now, many adsorbents, such as activated carbons [12], amino-functionalized materials [13], and biomass materials [14] etc., have been developed to remove heavy metal ions. Among these adsorbents, amino-functionalized mesoporous silicas (AFMS) appear to be particularly promising, due to their large adsorption capacities and high uptake rates for heavy metal ions [15]. Recently, the AFMS synthesized by an anionic surfactanttemplated method have attracted much attention due to the extremely regular arrangement of the amino groups introduced by the co-structure directing method [16,17]. Moreover, anionic surfactants are used in greater volume than any other surfactants because of their highly potent detergency and low cost of manufacture [18], which makes the anionic surfactant-templated method more advantage in full scale application. Currently, both neutralization and double decomposition routes are applied to the preparation of AFMS, but the neutralization route is simpler, because organic acids used as anionic surfactants can directly interact with the free amino groups in co-structure-directing agent (CSDA), thus avoiding addition of another acid (such as hydrochloric acid) to protonate the free amino groups in CSDA during the double decomposition process [18]. In general, the amino content, especially on the surface of mesopores, and the textural properties, such as pore size, specific surface area, and pore volume, play an important role in the adsorption of heavy metal ions on AFMS. Although the neutralization method has been used to prepare AFMS, the optimization of the synthesis parameters, which significantly affect the amino content and textural properties, has not been carried out in detail. To the best of our knowledge, although many studies have been focused on the single-component adsorption of Pb2þ and Cu2þ on AFMS, few have examined the binary adsorption of Pb2þ and Cu2þ as well as the effects of co-existing metal ions such as Zn2þ and Cd2þ on the adsorption of Pb2þ and/or Cu2þ [19e21]. In fact, the chemical composition of heavy metal ions in the polluted wastewater is rather complicated, because more than one contaminant are often detected at different concentrations. Therefore, it is important to characterize the adsorption behaviors of heavy metal ions on an adsorbent toward multi-component solutions. In the present study, a series of AFMS were prepared by the neutralization route. In order to obtain AFMS with high amino contents and desired textural properties, the synthesis parameters were optimized. Then we investigated the single-, binary-, ternary-, and quaternary-component adsorption of Pb2þ, Cu2þ, Zn2þ, and Cd2þ on the optimally synthesized AFMS sample as adsorbent. In particular, we focused on investigating the competitive adsorption effects on the removal efficiencies of the heavy metal ions from their aqueous solutions. 2. Experimental

251

solutions were prepared at room temperature using deionized water as solvent. 0.1 M single-component Cu2þ, Pb2þ, Zn2þ, and Cd2þ aqueous solutions were prepared from the precursors Cu(NO3)2$3H2O, Pb(NO3)2, Zn(NO3)2$3H2O, and Cd(NO3)2$4H2O, respectively, and the obtained solutions were used as atomic absorption spectrometry standards and as stock solutions. The singlecomponent solutions with different concentrations for adsorption experiments were then prepared by dilution of the stock solutions just prior to use, and their pH values were adjusted to the required ones using 0.1 M HCl or 0.1 M NaOH solution. AFMS were prepared by a neutralization method [18]. The typical synthesis process was carried out as follows: 2 mmol (0.4013 g) of DAA was dissolved into a mixture of 30 ml H2O and 30 ml alcohol (one among methanol, ethanol, propanol, and butanol) at 60  C. To the above-prepared solution, a mixture of 3.0 ml TEOS and x ml (0.2  x  3.0) APTMS was added drop by drop under vigorous stirring in a period of 30 min. In order to investigate the effects of the alcohol composition on the mesostructure of the resultant AFMS, different water/alcohol ratios were used while keeping the total volume of water and alcohol constant as 60 ml. The resulting mixture was then aged hydrothermally without agitation in a closed 100 ml reaction kettle placed in an oven at 80  C for 2 days. The product was filtered and the solid was washed with deionized water several times. The washed solid was then dried for 600 min at 80  C prior to a further analysis or use. The surfactant in the dried sample was removed by an extraction method. Typically, 1.0 g of the dried solid was dispersed in a solution containing 25.2 g of 3.3 M ethanolamine and 100 ml ethanol at room temperature under agitation. Afterwards, the mixture was refluxed at 90  C for 720 min. The solid was then recovered by filtration, washed with ethanol, and dried. The above extraction procedure was repeated twice. 2.2. Characterization The FT-IR spectra were recorded on a Nicole Nexus 670 spectrometer with a resolution of 4 cm1 using a KBr compression method. The powdered X-ray diffraction (XRD) patterns were performed on a PHILIPS PW3040/60 powder diffractometer using CuKa radiation (l ¼ 0.15406 nm). The morphology and particle sizes of the synthesized samples were examined with transmission electron microscopy (TEM) technique. The TEM images were obtained on a 2100 JEOL instrument working at 200 kV. The nitrogen contents of the synthesized AFMS samples were measured by a Vario ELIII elemental analyzer. The adsorption-desorption isotherms of N2 on AFMS at 196  C were performed on a Micromeritics ASAP 2020 apparatus, and the specific surface areas were then calculated using the multiple-point Brunauer-Emmett-Teller (BET) method in the relative pressure range of p/p0 ¼ 0.05e0.3. The pore size distribution (PSD) curves were computed using the Barrett-Joyner-Halenda (BJH) method and the average pore sizes were obtained from the peak positions of the PSD curves. Zeta potential analysis was conducted in ethanol solvent under a Malvern Zetasizer Nano ZS90 with MPT-2 autotitrator.

2.1. Chemicals and synthesis 2.3. Adsorption Dodecanoic acid (DAA), tetraethoxysilane (TEOS), and aminopropyltrimethoxysilane (APTMS) were purchased from Merck, Acros, and Aldrich, respectively. Cu(NO3)2$3H2O, Pb(NO3)2, Zn(NO3)2$3H2O, Cd(NO3)2$4H2O, methanol, ethanol, propanol, butanol, ethanol amine, NaOH, sodium ethylenediaminetetraacetic acid (EDTA), and 37% fuming hydrochloric acid were purchased from Sinopharm Chemical Reagent Co., Ltd., Shanghai. All aqueous

2.3.1. Single-component adsorption 0.02 g of the dried AFMS adsorbent was suspended into 20 ml of 2.0 mM Cu(NO3)2 solution, also applied to Pb(NO3)2, Zn(NO3)2, and Cd(NO3)2 solutions having a concentration of 2.0 mM, with an initial pH value of 5.07, under vigorous stirring at 25  C for different periods ranging from 2 to 120 min. The mixture was then quickly

252

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

filtered to remove the solid adsorbent and the solution was analyzed with a PerkinElmer optima 2100 DV atomic emission spectrometer (AES) to determine the concentration of the metal ion. The removal efficiency, R, of the metal ion on the adsorbent was calculated by the following equation:



C0  Ct  100% C0

(1)

where C0 (mM) is the initial concentration of the metal ion and Ct (mM) is the concentration at time t (min). The removal efficiency of the heavy metal ion as a function of adsorption temperature was carried out as follows: 0.01 g of the dried AFMS adsorbent was suspended in 20 ml of 3.0 mM Cu(NO3)2 solution, also applied to Pb(NO3)2, Zn(NO3)2, and Cd(NO3)2 solutions having a concentration of 3.0 mM, with an initial pH value of 5.07 at different adsorption temperatures ranging from 25 to 45  C. A circulating water bath was used to control the adsorption temperatures within ±0.1  C. The experimental procedure and the calculation of the removal efficiency were the same as those described above. The adsorption isotherms at 25  C were obtained by adding 0.02 g of the dried AFMS adsorbent into 20 ml of the metal ion solutions with the known initial concentrations ranging from 0.1 to 4 mM and the known pH values under vigorous stirring. After 120 min of the adsorption, previously proven to be sufficient to attain the equilibrium, 10 ml of the solution was withdrawn, filtered, and analyzed by AES.

2.4. Adsorbent regeneration The regeneration of the used adsorbent was carried out based on the method described in our previous work [17]. Typically, 0.05 g of the single-component heavy metal ion adsorbed AFMS was treated with 5 ml of 0.1 M EDTA solution under vigorous stirring at 50  C for 60 min, followed by washing with deionized water for several times and drying at 80  C for 120 min. The regenerated AFMS adsorbent was then used for the next adsorption. Six successive adsorption-regeneration cycles were performed to check the reusability of the developed AFMS adsorbent for capturing the heavy metal ions. 3. Results and discussion 3.1. Synthesis The AFMS were synthesized via the hydrolysis reactions of TEOS and APTMS, i.e., reactions (2) and (3), followed by their condensation onto the structure-directing agent (SDA) micelles, i.e., reactions (4) and (5). Since DAA, which acts as SDA in the synthesis, is scarcely soluble in water, some ethanol must be added in order to improve its solubility. On the other hand, ethanol itself is a product of reactions (2), (4), and (5) so that its amount in the reaction medium has to be optimized. H þ orOH 

SiðOC2 H5 Þ4 þ 4H2 Oƒƒƒƒƒƒƒƒƒ!SiðOHÞ4 þ 4C2 H5 OH

(2)

H þ orOH 

R  Si≡ðOCH3 Þ3 þ 3H2 Oƒƒƒƒƒƒƒƒƒ!R  Si≡ðOHÞ3 þ 3CH3 OH 2.3.2. Multi-component adsorption For the binary-component adsorption, the Cu2þ-Pb2þ, Cu2þZn2þ, Cu2þ-Cd2þ, Pb2þ-Zn2þ, Pb2þ-Cd2þ, and Cd2þ-Zn2þ solutions were obtained by diluting appropriate amounts of 0.1 M stock solutions and mixing them into a beaker. When investigating the effects of a given metal ion such as Pb2þ on the adsorption of another metal ion such as Cu2þ, 0.02 g of the dried AFMS adsorbent was suspended into a 100 ml beaker containing 20 ml of 1.0 mM Cu2þ and 0.25e1.0 mM Pb2þ solution under vigorous stirring. After 120 min of the adsorption, 10 ml of the solution was withdrawn and, after filtration, subjected to AES analysis. For the ternary-component adsorption, the ternary-component solutions had the following combinations: Cu2þ-Cd2þ-Zn2þ, Pb2þCd2þ-Zn2þ, Zn2þ-Cu2þ-Pb2þ, and Cd2þ-Cu2þ-Pb2þ. When studying the effects of two given ions, e.g., Cd2þ and Zn2þ, on the removal of the third one, e.g., Cu2þ, 0.02 g of the dried AFMS adsorbent was suspended into a 100 ml beaker containing 20 ml of a solution, in which the initial concentrations of the metal ions had two different sets as follows: a) 1.0 mM Cu2þ, 1.0 mM Cd2þ, and 0.25e0.75 mM Zn2þ (varied) and b) 1.0 mM Cu2þ, 1.0 mM Zn2þ, and 0.25e0.75 mM Cd2þ (varied). Also in this case, after 120 min of the adsorption, 10 ml of the solution was withdrawn, filtered, and analyzed by AES. For the quaternary-component adsorption, 0.04 g of the dried AFMS adsorbent was suspended into a 100 ml beaker containing 60 ml of a solution, in which the metal ions had three different sets of initial concentrations as follows: a) 1.5 mM of both Cu2þ and Pb2þ and 0.5 mM of both Cd2þ and Zn2þ, b) 1.5 mM of both Cd2þ and Zn2þand 0.5 mM of both Cu2þ and Pb2þ, and c) 1.0 mM of each Cd2þ, Zn2þ, Cu2þ, and Pb2þ, in which all the initial pH values were kept at 5.07. After 120 min of the adsorption, 10 ml of solution was withdrawn, filtered, and analyzed by AES. The measurement for the adsorption isotherm and the removal efficiency was repeated at least three times, and these measured experimental data shown in the figures later are taken as the descriptive statistic mean with an error bar.

(3) Hþ orOH 

≡Si  OH þ H5 C2 O  Si≡ƒƒƒƒƒƒƒƒƒ!≡Si  O  Si≡ þ C2 H5 OH (4) H þ orOH

R  Si≡ðOHÞ3 þ H5 C2 O  Si≡ƒƒƒƒƒƒƒƒƒ!≡Si  O  Si≡ þ C2 H5 OH

(5)

where R-represents the aminopropyl group, NH2(CH2)3e. In order to determine the optimal composition of the reaction medium, a series of experiments were carried out, in which the amounts of APTMS and TEOS were kept constant at 2.2 ml and 3.0 ml, respectively, while the amounts of water and ethanol were varied. The adsorption-desorption isotherms of N2 on the synthesized samples are shown in Fig. 1 and the corresponding results on the textural properties are summarized in Table 1. Under the synthesis conditions shown in Table 1 for experiments 1e3, no precipitate was formed, presumably due to the fact that ethanol is a product of the synthesis reactions, and when too much ethanol was added, the equilibria of the reactions were shifted to the left hand. However, under the synthesis conditions shown in Table 1 for experiments 4e6, the AFMS were formed, because adequate water/ ethanol ratios were used. Additionally, as shown in Table 1, the BET surface area and the pore size and volume are decreased, when the used amount of ethanol is reduced, probably due to the fact that if the ethanol in the synthesis solution is insufficient, DAA cannot properly exert its surfactant action. In terms of the textural properties of the resultant AFMS, the optimal volume ratio of ethanol to water in the synthesis solution seems to be 1/1. Besides ethanol, other alcohols, such as methanol, propanol, and butanol, were used as co-solvent. The adsorption-desorption isotherms of N2 on the synthesized samples are shown in Fig. 2 and the corresponding results on the textural properties and the

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

Fig. 1. Adsorption-desorption isotherms (196  C) of N2 on the resultant AFMS synthesized using 2.2 ml APTMS and 3.0 ml TEOS with different volumes of H2O and C2H5OH.

nitrogen contents of the AFMS are summarized in Table 2, indicating that the use of the shorter chain alcohols methanol and ethanol leads to the larger porosities and higher nitrogen contents of the synthesized samples, compared to those using the longer chain alcohols propanol and butanol. Methanol and ethanol mainly act as co-solvent [22] while propanol and butanol may additionally act as coesurfactant, leading to the formation of larger micelles, which in turn hinders the interactions between DAA and APTMS [23]. In terms of the amino contents and textural properties of the synthesized AFMS, ethanol appears to be an optimal co-solvent in the current synthesis. It is obvious that the initial concentration of APTMS in the reaction system can greatly affect the amino contents and textural properties of the resultant samples, which play an important role on the adsorption of heavy metal ions. Therefore, it is important to find the optimal APTMS content to prepare the desired adsorbents. Fig. 3 shows the nitrogen adsorption-desorption isotherms and PSD curves of the AFMS synthesized using different dosages of APTMS, at which 30 ml ethanol and 30 ml water were used as reaction media. The corresponding textural properties and nitrogen contents of the synthesized samples are summarized in Table 3. It can be seen from Fig. 3 that all the isotherms are of type IV as defined by IUPAC. The isotherms of samples a and b, synthesized using the lower amounts of APTMS (0.2 and 1.4 ml, respectively), show typical H1 hysteresis loops in the relative pressure range of p/ po ¼ 0.4e0.9, suggesting a cylindrical pore system, while those of samples c, d, and e, synthesized using the higher amounts of APTMS (2.2, 2.6, and 3.0 ml, respectively), show H3 hysteresis loops, indicating a cage type mesophase. This difference in structure is presumably due to the fact that an increase in the amount of APTMS leads to an increase in the initial pH of the solution, which

253

Fig. 2. Adsorption-desorption isotherms (196  C) of N2 on the AFMS synthesized using 2.2 ml APTMS and 3.0 ml TEOS with different alcohols.

apparently results in a change of surfactant ionization and thus in the different structures of the formed AFMS [24]. On the other hand, the aminopropyl groups in APTMS can perturb the selfassembly of surfactant micelles and silica precursor, due to the e formation of zwitterions (NHþ 3 — OSi) and the competition between protonated amine groups (-NH2Hþ) and positively charged silicate species (SiOHþ 2 ) for the surfactant DAA to form (-NH2Hþ)(XeSiOHþ 2 ) [25]. From Table 3, it can be seen that the amino content of samples a-e increases when the used amount of APTMS increases from 0.2 to 2.2 ml and then decreases slightly. In terms of the amino content, textural properties, and heavy metal removal efficiency (shown later), sample c was chosen to be the optimal candidate for investigation on the adsorption of the heavy metal ions. Fig. 4 shows the results characterized by the FT-IR, XRD, and TEM techniques for sample c. From the FT-IR spectrum, it can be concluded that the surfactant in the as-synthesized sample was effectively removed by extraction with ethanol, because the intensity of typical peaks around 2852 and 2932 cm1, corresponding to the symmetrical and asymmetrical stretching vibrations of the methylene groups of DAA and CSDA [26], is dramatically lower than that in the corresponding as-synthesized sample (not shown). The peak at about 955 cm1 can be attributed to the stretching of the framework SieOH group [27]. The vibrations of SieOeSi can be seen at around 796 cm1 (symmetric stretching), and 470 cm1 (bending) [28]. The presence of -N-H bending vibration around 692, 1489, and 1565 cm1 confirms the existence of amino groups [23,29]. The XRD pattern exhibits a distinguishable peak at approximately 2q ¼ 2.0 , indicating that an ordered mesostructure is formed. From the TEM image, it can be clearly seen that the centrosymmetric radial mesopores (emanating from the spherical

Table 1 Textural properties of the resultant AFMS synthesized using 2.2 ml APTMS and 3.0 ml TEOS with different dosages of H2O and C2H5OH. Experiment

H2O/ml

C2H5OH/ml

SBET/m2 g1

Average pore size/nm

Pore volume/cm3 g1

1 2 3 4 5 6

0 10 20 30 40 50

60 50 40 30 20 10

e e e 395 182 72.0

e e e 4.39 3.36 2.89

e e e 0.48 0.16 0.06

254

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

Table 2 Textural properties and nitrogen contents of the resultant AFMS synthesized using 2.2 ml APTMS and 3.0 ml TEOS with different alcohols. Solution composition 30 30 30 30

ml ml ml ml

methanol þ30 ml H2O ethanol þ30 ml H2O propanol þ30 ml H2O butanol þ30 ml H2O

N content/mmol g1

SBET/m2 g1

Average pore size/nm

Pore volume/cm3 g1

2.64 2.83 1.64 1.06

386 395 103 49.0

5.47 4.39 10.3 7.01

0.58 0.48 0.34 0.12

Fig. 3. Adsorption-desorption isotherms (196  C) of N2 on the AFMS prepared under the synthesis conditions shown in Table 3 with different dosages of APTMS. (a) VAPTMS ¼ 0.2 ml, (b) 1.4 ml, (c) 2.2 ml, (d) 2.6 ml, and (e) 3.0 ml (isotherms b-e vertically shifted for clarity).

center to the exterior surface) are formed, with a structure similar to those reported in our previous studies [17,30].

3.2. Adsorption 3.2.1. Single-component adsorption In order to examine the effects of the textural properties and nitrogen contents of the synthesized samples on the removal efficiency of the metal ions, 0.02 g of one of samples aee coded in Table 3 was added to 20 ml of 1.0 mM prepared metal nitrate solution with a pH of 5.07, and the adsorption was then carried out under vigorous stirring at 25  C for 120 min. The results on the removal efficiency are summarized in Fig. 5, showing that sample c exhibits the best performance in terms of the removal efficiency for Cu2þ (99.5%), Pb2þ (99.8%), Zn2þ (67%), and Cd2þ (82%). From Fig. 5, it can be also found that the removal efficiencies of Cu2þ and Pb2þ on each sample are much higher than those of Zn2þ and Cd2þ. This is due to the fact that the coordinative interactions between Cu2þ and amino groups are greater, compared to those between other

metal ions and amino groups, and Pb2þ possesses a stronger polarization because of its larger ionic radius [31]. Besides the equilibrium adsorption, the adsorption kinetics is crucial in the evaluation of an adsorbent for the removal efficiency of heavy metal ions. The removal efficiency as a function of time is shown in Fig. 6. It can be seen that the uptake rates of all the metal ions are very high in the first 5 min. Afterwards, the adsorption approaches the equilibrium, which is achieved within 40 min. The high adsorption capacities and fast uptake rates of the heavy metal ions are benefit from the high amino loading on the surface of the AFMS adsorbent because of the coordination between the nitrogen atoms in the amino groups and the heavy metal ions, in which lone pair electrons and unoccupied orbits are provided by the nitrogen atoms in the amino groups and the heavy metal ions, respectively. Therefore, a higher amino loading will result in a higher removal efficiency of the heavy metal ions. Generally, free amino groups are easy to interact with the heavy metal ions. However, the amino state is greatly affected by the pH of the solution, which in turn influences the removal efficiency of the heavy metal ions. In order to investigate the effect of pH on the state of the amino groups on AFMS, the Zeta potentials (z) of sample c at different pH values of the solutions were measured. It is found that all the measured z values are positive at the different pH values of the solutions: the z value decreases greatly from 28.64 to 21.25 mV, when the pH varies from 3 to 4, and then it keeps almost constant with a value of 21.20 mV, when the pH is increased up to 8. This is probably due to the fact that alkylamines have a pKb value of ca. 4 [17], indicating that at the pH values below 4 in the solution, the amino groups on the surface of the AFMS adsorbent are mostly in a protonated form, leading to a higher z value. When the pH values in the solution are above 4, the free amino groups on the surface of the AFMS adsorbent are available and z becomes constant. Consequently, an initial pH value of 5.07 in the solutions is suitable for the effective removal of the heavy metal ions by the AFMS from their aqueous solutions. The effects of adsorption temperature on the removal efficiencies of the single-component heavy metal ions by sample c are presented in Fig. 7. It can be seen that the removal efficiency of the heavy metal ion increases gradually with increasing adsorption temperature, in agreement with the observation by Da'na and Sayari [21]. Although the reaction of the amino groups with the metal ions is exothermic, the process for the activated diffusion of the heavy metal ions inside the adsorbent pores is endothermic and this positive enthalpy change might be larger than the negative

Table 3 Textural properties and nitrogen contents of the AFMS synthesized using 3.0 ml TEOS, 30 ml H2O, and 30 ml C2H5OH with different dosages of APTMS. Sample code

Volume of APTMS/ml

N content/mmol g1

SBET/m2 g1

Average pore size/nm

Pore volume/cm3 g1

a b c d e

0.2 1.4 2.2 2.6 3.0

1.12 2.14 2.83 2.76 2.62

122 334 395 338 95.0

6.40 5.33 4.39 6.75 7.10

0.25 0.54 0.48 0.58 0.18

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

255

Fig. 4. FT-IR spectrum (a), XRD pattern (b), and TEM image (c) of sample c prepared under the synthesis conditions indicated in Table 3.

enthalpy change due to the formation of complexes between amino groups and metal ions, leading to an overall positive enthalpy change [21]. Consequently, an increase in temperature is benefit for the adsorption of the heavy metal ions on sample c.

Fig. 5. Removal efficiencies of Cu2þ, Pb2þ, Zn2þ, and Cd2þ by samples a-e from 1.0 mM metal nitrate solutions with a pH of 5.07 at 25  C.

The measured adsorption isotherms are presented in Fig. 8. It can be seen that all the isotherms have a sharp increase in the adsorbed amount in a low metal-ion concentration range, indicating the strong interactions between metal ions and adsorbent.

Fig. 6. Single-component uptake rates of Pb2þ, Cu2þ, Cd2þ, and Zn2þ on sample c from 1.0 mM metal nitrate solutions with a pH of 5.07 at 25  C.

256

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

Fig. 7. Removal efficiencies of Cu2þ, Pb2þ, Zn2þ, and Cd2þ by sample c from 3.0 mM metal nitrate solutions with an initial pH value of 5.07 as a function of temperature.

Fig. 8. Single-component adsorption isotherms of Pb2þ, Cu2þ, Cd2þ, and Zn2þ on sample c at 25  C. Symbols are the experimental data and dashed lines are the Sips model correlations.

In order to describe the measured isotherm data, different isotherm models, such as the Langmuir, Freundlich, and Sips models [32], have been tried, but the Sips model describes the present case much better over the full range. 1=n

qe ¼

qm Ks Ce

(6)

1=n

1 þ Ks Ce

Table 4 Estimated parameter values and 95% confidence limits for the fitting of the adsorption data of Pb2þ, Cu2þ, Cd2þ, and Zn2þ on sample c at 25  C by the Sips model. Metal ion

Parameters qm (mmol/g)

Pb2þ Cu2þ Cd2þ Zn2þ

2.86 2.34 1.71 1.36

± ± ± ±

0.06 0.53 0.35 0.24

KS [(mM)1/n] 0.52 0.45 0.26 0.26

± ± ± ±

0.46 0.42 0.58 0.32

n 1.38 1.46 2.45 1.68

± ± ± ±

0.06 0.27 0.34 0.38

where qm (mmol/g) is the saturation adsorption capacity, KS [(mM)1/n] is the adsorption equilibrium constant, and n is the parameter characterizing the system heterogeneity (namely, the larger the parameter, the more heterogeneous the system). The estimated parameter values are listed in Table 4, and the corresponding fitted isotherms are shown in Fig. 8 as dotted lines. The adsorption affinity to the metal ions investigated is in the order of Pb2þ > Cu2þ > Cd2þ > Zn2þ, which follows the same order of the Pauling electronegativity for the corresponding metals. The extracted adsorption capacities are 2.86, 2.34, 1.71, and 1.36 mmol/g for Pb2þ, Cu2þ, Cd2þ, and Zn2þ, respectively, higher than those on other adsorbents reported in the literature, for example, adsorption capacities of 2.77 mmol/g for Pb2þ on [email protected] [33], 0.93 mmol/g for Cu2þ on APTS-SBA-15-AB [34], 0.20 mmol/g for Cd2þ on amino-functionalized [email protected] magnetic nanomaterial [35], 1.17 mmol/g for Zn2þ on EDTA functionalized silica [36], 0.50 mmol/g for Cd2þ on activated carbon [37], and 0.72 mmol/g for Cu2þ, 0.56 mmol/g for Pb2þ, and 0.21 mmol/g for Cd2þ on AAmAMPSNa/clay [38]. These higher adsorption capacities are probably ascribed to the fact that sample c possesses a higher amino content and better textural properties for the adsorption of the metal ions. 3.2.2. Multi-component adsorption Because the single-component adsorption experiments confirm that the optimally synthesized AFMS adsorbent has very good adsorption properties toward all the metal ions investigated, and in particular toward Pb2þ and Cu2þ, it appears useful to quantitatively evaluate the competitive adsorption effects deriving from the coexistence of the metal ions in solution. The results on the binary adsorption of the pairs from Cu2þ, Pb2þ, Zn2þ, and Cd2þ on sample c are presented in Fig. 9, indicating that competitive adsorption effects do exist and in general, the removal efficiency of one type of the metal ion is decreased in the presence of another type of the metal ion in the system. In particular, from Fig. 9c and d it appears that Cu2þ and Pb2þ significantly hinder the adsorption of Cd2þ and Zn2þ. For example, the removal efficiencies of Zn2þ and Cd2þ are decreased to about 55%, when 0.25 mM Pb2þ are present in the solution, and these values are dropped to around 40% when 0.25 mM Cu2þ are present in the solution. This is due to the fact that Pb2þ and Cu2þ have a higher affinity to the adsorbent than Cd2þ and Zn2þ, even though one would have expected a stronger inhibition by Pb2þ than by Cu2þ. On the other hand, the effect of Cd2þ on the adsorption of Cu2þ and Pb2þ is almost insignificant, as shown in Fig. 9a and b, for example, the removal efficiency of Cu2þ is decreased by about 10% while the adsorption of Pb2þ is almost unaffected. Zn2þ, however, has a somewhat stronger effect than Cd2þ, reducing the removal efficiency of Cu2þ from ca. 87% to ca. 75% and that of Pb2þ from ca. 90% to ca. 80%. Moreover, it is interesting to note that, within a certain concentration range, Pb2þ promotes the adsorption of Cu2þ and increases the removal efficiency from ca. 87% to almost 100% (Fig. 9a), while the reverse does not happen (Fig. 9b). Overall, it appears that the results on the binary adsorption can be interpreted in terms of the competitive adsorption of the coexisting metal ions on the adsorbent, in light of the affinity order mentioned above for the single-component adsorption. However, the reasons on some unexpected phenomena observed, i.e., the promotion effects of Pb2þ on the adsorption of Cu2þ, and Zn2þ having the stronger hindrance effects on the adsorption of Cu2þ and Pb2þ than Cd2þ, are still unclear. The effects of Zn2þ and Cd2þ on the adsorption of either Cu2þ or Pb2þ were investigated with different sets of the ternarycomponent adsorption experiments. These results are shown in Fig. 10. As shown in Fig. 10a and b, the presence of Cd2þ and Zn2þ in the solution has a limited effect on the adsorption of either Cu2þ or

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

257

Fig. 9. Effects of a co-existing metal ion with initial concentrations ranging from 0.25 to 1 mM in the solution on the adsorption of another metal ion with an initial concentration of 1.0 mM on sample c at 25  C.

Pb2þ, due to the fact that the adsorbent has a greater affinity to either Cu2þ or Pb2þ. On the other hand, the simultaneous presence of Cu2þ and Pb2þ significantly reduces the removal efficiency of either Cd2þ or Zn2þ, and Pb2þ can promote the adsorption of Cu2þ. Finally, the quaternary-component adsorption of Pb2þ, Cu2þ, Cd2þ, and Zn2þ on sample c was conducted, and the results are shown in Fig. 11. As expected, the removal efficiencies of Cu2þand Pb2þ are markedly higher than those of Cd2þ and Zn2þ. In addition, the removal efficiencies of Cd2þ and Zn2þ are increased with decreasing the concentrations of Cu2þand Pb2þ. However, when the concentrations of Cd2þ and Zn2þ are increased, the removal efficiency of Cu2þ is decreased while for Pb2þ it is unexpectedly increased, for which a further investigation is underway.

3.3. Adsorbent regeneration The reusability of an adsorbent is crucial in practical applications. Fig. 12 illustrates the results on 6 adsorption-regeneration cycles of the single-component heavy metal ions on sample c. It can be found that the removal efficiency is slightly decreased after 6 adsorption-regeneration cycles, indicating that the heavy metal ions are easily desorbed from the adsorbed AFMS using EDTA as

desorption effluent and the synthesized AFMS adsorbent is rather reusable.

4. Conclusions The amino-functionalized mesoporous silicas with high amino loading and desired textural properties were prepared via the optimization of the synthesis parameters, such as the type of solvent, the ratio of water to ethanol, and the content of APTMS. The optimized AFMS sample was used as adsorbent for the single- and multi-component adsorption of Pb2þ, Cu2þ, Cd2þ, and Zn2þ in aqueous solutions. The measured single-component adsorption isotherms can be well described by the Sips model, in which the extracted adsorption capacities are in the order of Pb2þ > Cu2þ >> Cd2þ > Zn2þ, which are higher than those on other adsorbents reported in the literature. The results on the single- and multicomponent adsorption show that the removal efficiencies of Pb2þ and Cu2þ are higher than those of Cd2þ and Zn2þ. In addition, the coexisting Pb2þ and Cu2þ can significantly reduce the removal efficiencies of Cd2þ and Zn2þ while the presence of Cd2þ and Zn2þ has a much less effect on the adsorption of Pb2þ and Cu2þ. The present study implies that the optimally synthesized AFMS sample could be

258

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

Fig. 10. Effects of two co-existing metal ions on the adsorption of another metal ion on sample c at 25  C. Initial concentrations of metal ions in the solutions: (a) Cu2þeZn2þeCd2þ system: A: 1 mM Cu2þ, 0.25 mM Zn2þ, 0.75 mM Cd2þ, B: 1 mM Cu2þ þ 0.5 mM Zn2þ þ 0.5 mM Cd2þ, and C: 1 mM Cu2þ þ 0.75 mM Zn2þ þ 0.25 mM Cd2þ; (b) Pb2þ-Zn2þ-Cd2þ system: A: 1 mM Pb2þ þ 0.25 mM Zn2þ þ 0.75 mM Cd2þ, B: 1 mM Pb2þ þ 0.5 mM Zn2þ þ 0.5 mM Cd2þ, and C: 1 mM Pb2þ þ 0.75 mM Zn2þ þ 0.25 mM Cd2þ; (c) Cd2þ-Cu2þ-Pb2þ system: A: 1 mM Cd2þ þ 0.25 mM Cu2þ þ 0.75 mM Pb2þ, B: 1 mM Cd2þ þ 0.5 mM Cu2þ þ 0.5 mM Pb2þ, and C: 1 mM Cd2þ þ 0.75 mM Cu2þ þ 0.25 mM Pb2þ; and (d) Zn2þeCu2þePb2þ system: A: 1 mM Zn2þ þ 0.25 mM Cu2þ þ 0.75 mM Pb2þ, B: 1 mM Zn2þ þ 0.5 mM Cu2þ þ 0.5 mM Pb2þ, and C: 1 mM Zn2þ þ 0.75 mM Cu2þ þ 0.25 mM Pb2þ.

Fig. 11. Quaternary-component adsorption of Pb2þ, Cu2þ, Cd2þ, and Zn2þ on sample c at 25  C. Initial concentrations of metal ions in the solutions: A: 1.5 mM Cu2þ1.5 mM Pb2þ-0.5 mM Cd2þ0.5 mM Zn2þ, B: 1.0 mM Cu2þ1.0 mM Pb2þ1.0 mM Cd2þ 1.0 mM Zn2þ, and C: 0.5 mM Cu2þ0.5 mM Pb2þ1.5 mM Cd2þ1.5 mM Zn2þ.

Fig. 12. Removal efficiencies of Cu2þ, Pb2þ, Zn2þ, and Cd2þ by sample c at 25  C from 1.0 mM metal nitrate solutions with an initial pH value of 5.07 as a function of the reused times of the adsorbent.

S. Hao et al. / Microporous and Mesoporous Materials 236 (2016) 250e259

used as an efficient adsorbent for the removal and separation of Pb2þ and Cu2þ from wastewater. Acknowledgments Financial supports by the Italian Ministry for Foreign Affairs in the context of the 2013e2015 Programme of Scientific and Technological Cooperation between People's Republic of China and Italy, the Natural Science Foundation of Zhejiang Province, China (LY14B070006, Y4110289), and the National Natural Science Foundation of China (21476214) are acknowledged. References [1] A. Garcia-Sanchez, A. Alastuey, X. Querol, Sci. Total Environ. 242 (1999) 179e188. [2] A.T. Paulino, L.B. Santos, J. Nozaki, React. Funct. Polym. 68 (2008) 634e642. [3] Z.O.K. Atakli, Y. Yurum, Chem. Eng. J. 225 (2013) 625e635. [4] N. Kawasaki, H. Tominaga, F. Ogata, K. Kakehi, Chem. Eng. J. 157 (2010) 249e253. [5] T. Inaba, E. Kobayashi, Y. Suwazono, M. Uetani, M. Oishi, H. Nakagawa, K. Nogawa, Toxicol. Lett. 159 (2005) 192e201. [6] C.K. Jain, D.C. Singhal, M.K. Sharma, J. Hazard. Mater. B 114 (2004) 231e239. € rstner, G.T.W. Wittmann, Metal Pollution in the Aquatic Environment, [7] U. Fo Springer-Verlag, Berlin, 1981. [8] F. Pepe, B. de Gennaro, P. Aprea, D. Caputo, Chem. Eng. J. 291 (2013) 37e42. rez-Quintanilla, A. Sa nchez, I. del Hierro, M. Fajardo, I. Sierra, J. Colloid [9] D. Pe Interf. Sci. 313 (2007) 551e562. [10] L. Semerjian, J. Hazard. Mater. 173 (2010) 236e242. [11] L. Wang, L. Yang, Y. Li, Y. Zhang, X. Ma, Z. Ye, Chem. Eng. J. 163 (2010) 364e372. [12] S. Oh, D.S. Kim, J. Environ. Sci. Heal. A 49 (2014) 710e719. [13] J.F. Madrid, G.M. Nuesca, L.V. Abad, Radiat. Phys. Chem. 97 (2014) 246e252. [14] D.H.K. Reddy, S.M. Li, K. Seshaiah, Water Air Soil Poll. 223 (2012) 5967e5982. [15] V. Manu, M.M. Haresh, C.B. Hari, V.J. Raksh, Ind. Eng. Chem. Res. 48 (2009) 8954e8960.

259

[16] T. Yokoi, H. Yoshitake, T. Yamada, Y. Kubota, T. Tatsumi, J. Mater. Chem. 16 (2006) 1125e1135. [17] S. Hao, Y. Zhong, F. Pepe, W. Zhu, Chem. Eng. J. 189e190 (2012) 160e167. [18] S. Che, A.E. Garcia-Bennett, T. Yokoi, K. Sakamoto, H. Kunieda, O. Terasaki, T. Tatsumi, Nat. Mater. 2 (2003) 801e806. [19] A.M. Showkat, Y.P. Zhang, M.S. Kim, A.I. Gopalan, K.R. Reddy, K.P. Lee, B. Kor, Chem. Soc. 28 (2007) 1985e1992. [20] A. Heidaria, H. Younesia, Z. Mehraban, Chem. Eng. J. 153 (2009) 70e79. [21] E. Da’na, A. Sayari, Chem. Eng. J. 166 (2011) 445e453. [22] C. Gao, H. Qiu, W. Zeng, Y. Sakamoto, O. Terasaki, K. Sakamoto, Q. Chen, S. Che, Chem. Mater. 18 (2006) 3904e3914. [23] X.G. Wang, K.S.K. Lin, J.C.C. Chan, S. Cheng, J. Phys. Chem. B 109 (2005) 1763e1769. [24] B. Tan, S.E. Rankin, J. Phys. Chem. B 108 (2004) 20122e20129. n, J.B. Rosenholm, R. Schwarzenbacher, M. Kriechbaum, [25] P. Ågren, M. Linde H. Amenitsch, P. Laggner, J. Blanchard, F. Schüth, J. Phys. Chem. B 103 (1999) 5943e5948. [26] H. Zheng, C. Gao, S. Che, Micropor. Mesopor. Mater. 116 (2008) 299e307. [27] M. Llusar, G. Monros, C. Roux, J.L. Pozzo, C. Sanchez, J. Mater. Chem. 13 (2003) 2505e2514. [28] A.S.M. Chong, X.S. Zhao, J. Phys. Chem. B 107 (2003) 12650e12657. [29] X. Yan, L. Zhang, Y. Zhang, G. Yang, Z. Yan, Ind. Eng. Chem. Res. 50 (2011) 3220e3226. [30] S. Hao, Q. Xiao, H. Yang, Y. Zhong, F. Pepe, W. Zhu, Micropor. Mesopor. Mater. 132 (2010) 552e558. [31] Y. Jiang, Q. Gao, H. Yu, Y. Chen, F. Deng, Micropor. Mesopor. Mater. 103 (2007) 316e324. [32] D.D. Do, Adsorption Analysis: Equilibrium and Kinetics, Imperial College Press, London, 1998. [33] S. Wang, K. Wang, C. Dai, H. Shi, J. Li, Chem. Eng. J. 262 (2015) 897e903. [34] E. Da’na, A. Sayari, Chem. Eng. J. 166 (2011) 445e453. [35] B. Luo, X.J. Song, F. Zhang, A. Xia, W.L. Yang, J.H. Hu, C.C. Wang, Langmuir 26 (2010) 1674e1679. [36] R. Kumar, M.A. Barakat, Y.A. Daza, H.L. Woodcock, J.N. Kuhn, J. Colloid Interf. Sci. 408 (2013) 200e205. [37] J. Wang, H. Liu, S. Yang, J. Zhang, C. Zhang, H. Wu, Appl. Surf. Sci. 316 (2014) 443e450. € z, A. Durmus, A. Kasgo €z, Polym. Adv. Technol. 19 (2008) 213e220. [38] H. Kasgo