SiO2

0 downloads 0 Views 2MB Size Report
Sep 18, 2017 - The adsorption capacity of the 20% MgO-SiO2 sample could be as high ..... of dyes by ACS prepared from waste tyre reinforcing fibre, effect of ...
Journal of Colloid and Interface Science 510 (2018) 111–117

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Regular Article

Ultra-high adsorption capacity of MgO/SiO2 composites with rough surfaces for Congo red removal from water Mengqing Hu, Xinlong Yan ⇑, Xiaoyan Hu, Jiajin Zhang, Rui Feng, Min Zhou School of Chemical Engineering & Technology, China University of Mining and Technology, XuZhou 221116, PR China

g r a p h i c a l a b s t r a c t

a r t i c l e

i n f o

Article history: Received 27 June 2017 Revised 13 September 2017 Accepted 14 September 2017 Available online 18 September 2017 Keywords: MgO-SiO2 composites Adsorption Congo red

a b s t r a c t Due to its high isoelectric point, relative safety and low environmental toxicity, magnesium oxide has attracted much attention for its role in the removal of toxic dyes from wastewater. Herein, MgO-SiO2 composites with rough surfaces were synthesized by a one-step method. The as-prepared composites were characterized for the adsorption of Congo red from water using adsorption kinetics and isotherms. The adsorption capacity of the 20% MgO-SiO2 sample could be as high as 4000 mg/g at 25 °C, which is the highest value reported to date. The adsorption process of Congo red on the as-synthesized samples obeyed the Langmuir adsorption model. The MgO-SiO2 composite sample could be regenerated by calcination, and the regeneration efficiency remained for up to 5 cycles of the regeneration. Ó 2017 Elsevier Inc. All rights reserved.

1. Introduction Excessive use of organic dyes in the textile, plastics, printing, and cosmetic industries has been a growing concern in recent years, especially because of the severe environmental problems, especially water pollution, caused by discharging of the dye containments into natural or engineered water bodies as industrial effluents [1,2]. Therefore, developing techniques for efficient dye removal from waste water is of great importance. Several commonly used techniques are photocatalysis, adsorption and coagulation. Among these, adsorption is the most ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (X. Yan). http://dx.doi.org/10.1016/j.jcis.2017.09.063 0021-9797/Ó 2017 Elsevier Inc. All rights reserved.

convenient method for wastewater treatment due to its simple and inexpensive operation, high efficiency and low energy consumption [3]. In the adsorption process, the identity of the adsorbent usually plays a key role. Thus, the development of efficient adsorbents is crucial to improve the adsorption method. So far, several kinds of adsorbents have been studied for the removal of dyes from waste water, such as clays [4,5], activated carbon [6,7], metal oxides, and hydroxides [8,9], however, several of these adsorbents have limitations such as low adsorption capability, high cost, and inefficient recycling. Metal oxides have been intensively studied because of their wide applications in areas such as adsorbents and catalysts [10–13]. One of the most attractive metal oxides is magnesium oxide (MgO), since it is low cost, stable, nontoxic and environmen-

112

M. Hu et al. / Journal of Colloid and Interface Science 510 (2018) 111–117

tally friendly [14–17]. Usually, MgO can be obtained by thermal decomposition of magnesium hydroxide or magnesium carbonate. Due to the high surface area to volume ratio of the particles, nanosized MgO often exhibits good performance; hence, by adjusting the preparation conditions, nano-MgO with different morphologies and particle sizes were synthesized and tested as potential adsorbents for removing pollutants from water [18–21]. In addition, to improve the adsorption performance of MgO particles for the removal of dye molecules like Congo red (CR), the development of internal pores or surface features on the adsorbents were also reported, which enhanced their adsorption capacity significantly [22–24]. On one hand, the development of the MgO particles’ external surface, except in the case of ultrafine nanoparticles, may be a good solution for increasing the adsorption capacity of CR, but it is difficult to do and rarely reported. Recently, silica particles with rough surfaces have been successfully synthesized and have attracted significant attention [25–26]. Thus, in this work, with the help of SiO2, MgO-SiO2 composites with rough surfaces were synthesized using a one-step method. The structures and properties of the composites were characterized by different techniques and the adsorption performances for CR were investigated. 2. Experimental 2.1. Materials All reagents used were of analytical grade and were purchased from Sinoreagent Company, including Mg(NO3)26H2O (>99%), tetraethoxysilane (TEOS, >28.4% SiO2), resorcinol (>99.5%), formaldehyde (>37%), aqueous ammonia (>28%), CR (>98%) and ethanol (>99.7%). 2.2. Synthesis Typically, 0.6 g of resorcinol, 0.84 ± 0.02 mL of formaldehyde and 12 ± 0.5 mL of aqueous ammonia were successively added to a solution of 140 mL ethanol and 20 mL deionized water. The mixture was stirred continuously for 6 h at room temperature followed by the addition of 2.4 mL of TEOS to the solution. After that, the solution was stirred for 8 min; then, 2.24 mL of formaldehyde and 1.6 g of resorcinol and Mg(NO3)26H2O (Mg(NO3)2/TEOS weight ratio: 0–30%) were added and the mixture continued to stir for another 2 h. Finally, the product was filtered, washed with ethanol, dried in air at 50 °C overnight, and then calcined at 550 °C for 5 h. 2.3. Characterization Powder XRD patterns of the samples were obtained using a Bruker D8 ADVANCE diffractometer with Cu Ka radiation (40 kV, 30 mA). X-ray photoelectron spectroscopy (XPS) data were recorded using a Thermo Scientific Escalab 250Xi spectrometer equipped with an Al Ka radiation source. The amount of Mg in the composites was determined by inductively coupled plasma optical emission spectrometry (ICP-OES, Varian 720-ES, USA). N2 adsorption-desorption isotherms were measured after sample degassing at 100 °C overnight in a Quantachrome IQ2 porosimeter. The specific surface area was calculated using the BrunauerEmmet-Teller (BET) method. Total pore volume was determined by the volume of liquid nitrogen adsorbed at a relative pressure of 0.99. The micropore volume (Vmicro) was estimated using the t-plot method [27], and mesopore volumes were obtained by subtracting Vmicro from the total pore volume. The surface morphology of the samples was characterized using a transmission electron

microscope (TEM) (FEI Tecnai G2 F20) and a scanning electron microscope (SEM) (FEI Quanta 400 FEG). FT-IR spectra were recorded with a Thermo Nicolet IS5 spectrometer in transmission mode in the range of 4000–400 cm1. CO2 temperatureprogrammed desorption (CO2-TPD) experiments were performed on a TP-5080 apparatus (XianQuan Company, Tianjin) equipped with a thermal conductivity detector (TCD). Thermogravimetric analysis (TGA, METTLER Toledo) was conducted under air atmosphere with a temperature increase rate of 10 °C/min. Particle size distribution of the samples were characterized by Zetasizer (Nano ZS, Malvern). The concentration of CR was detected with a UV–Vis spectrophotometer (TU-1810, Persee Co., China) based on the absorbance at 498 nm. 2.4. Adsorption experiments For the adsorption study, 10 mg of the adsorbent was added to 50 mL of solutions with different CR concentrations (100 mg/L– 1000 mg/L) and stirred continuously at 25 °C. At different adsorption times, a small amount of the mixture was collected and the adsorbent was separated using a syringe filter (PTFE, 0.22 µm); the filtrate was then collected for analysis. The effect of pH was analyzed by mixing 5 mg of adsorbent with 25 mL of CR solution (500 mg/L) at different pH levels for 24 h. The concentration of CR left in the solution was analyzed as above. The adsorption capacity of the MgO-SiO2 composite was calculated using the following equation:

qt ¼

ðC 0  C t ÞV m

in which qt is the adsorption capacity, C0 and Ct are the concentration of Congo red before and after adsorption, respectively, V is the solution volume (L), and m is the weight of used MgO composite (g). The recyclability of the adsorbents was tested by thermal treatment of the adsorbent after CR adsorption. 50 mg of the 20% MgOSiO2 sample was added to 250 mL CR solution with a concentration of 500 mg/L. After adsorption for 24 h, the solid sample was collected by filtration and then dried and calcined at 550 °C for 2 h. The recovered sample was reused for CR adsorption. 3. Results and discussion 3.1. Materials characterization The Mg content in the composites measured by ICP were 2.5 wt %, 5.9 wt%, 10.2 wt% and 11.4 wt% for 5% MgO-SiO2, 10% MgO-SiO2, 20% MgO-SiO2 and 30% MgO-SiO2 samples, respectively. Powder XRD patterns of all the samples with the variation of MgO wt% are shown in Fig. 1. A wide diffraction peak at 22° (2h) corresponding to amorphous silica is observed for all the samples [28]. No diffraction peaks indicative of magnesium oxide (2h = 36.92°, 42.90°) are observed [22], which suggests the MgO is highly dispersed or in insufficient amounts to be detected by this method. With the increase in MgO content in the samples, the intensity of the broad peak decreased slightly as compared to that of pure SiO2. This may be cause by the reduced quantity of SiO2 in the amorphous phase in the composite sample. Moreover, a slight up-ward shift in 2 theta degree could be observed, which may be caused by an increase in the densification of amorphous SiO2 in the composite [29]. To investigate the surface chemical composition, one of the samples (10% MgO-SiO2) was analyzed by XPS analysis (Fig. 2). From the region of Mg 1s, the spectra contained only one peak at the binding energies of 1304.83 eV, which is associated with the presence of magnesium in the form of Mg2+ (Fig. 2a) [30]. A peak

M. Hu et al. / Journal of Colloid and Interface Science 510 (2018) 111–117

Intensity (a.u.)

1500

SiO2 5% MgO-SiO 2

1000

10% MgO-SiO 2 20% MgO-SiO 2

500

30% MgO-SiO 2 0 10

20

30

40

50

60

70

2 theta (degree) Fig. 1. XRD patterns of the pure SiO2 sample and MgO-SiO2 samples.

centered at 103.52 eV in the region of the Si 2p area was observed (Fig. 2b), which can be interpreted as Si4+ [31]. Combined with the O 1s peak at 532.75 eV, it could be concluded that MgO and SiO2 were formed in the composites. Fig. 3 shows TEM images along with the energy dispersive X-ray spectroscopy (EDX) pattern of the MgO-silica nanoparticles. TEM images of all the samples are given in Fig. 3a–e. Without the addition of MgO, silica spheres covered with nanosized silica spikes can be observed in Fig. 3a. The nanospheres have an average diameter of 370 nm. Possibly because of scale-up preparation, the dimensions of the silica spheres in this study are larger than those of reported spheres [25], and solid spheres rather than hollow spheres were obtained in this study. By introducing 5% MgO to the silica, the 5% MgO-SiO2 sample exhibits a totally different morphology with the aggregation of nanoparticles. Moreover, the surfaces of the particles are covered by ‘‘sand-like” fine particles. When the weight ratio of Mg to Si precursor increased up to 10%, the aggregation of the particles still can be seen, but the particle size decreased. With further increases in the weight ratio, the size of the surface-covered particles continues to decrease (from 10 nm to 2–3 nm) (Fig. 3d– e). The spikes on the silica surface were formed by the cocondensation of resorcinol and formaldehyde together with silica species, growth of silica is restricted by the surrounding resorcinol-formaldehyde framework, resulting in the silica spikes [25]. With a small amount of Mg precursors added, the size of spheres decreased (5% and 10% MgO-SiO2 sample). With further increase of Mg content, the spheres break up into small pieces and aggregate together (20% and 30% MgO-SiO2 sample). From energy-dispersive X-ray spectroscopy (EDX) analyses (Fig. 3f), the presence of Mg, Si and O are apparent in the 10% MgO-SiO2 sample, proving the successful incorporation of MgO inside the SiO2.

113

Scanning electron microscopy (SEM) images (Fig. 4) further show the nanoscale surfaces of samples with different MgO contents. The pure SiO2 sample shows the sphere-like morphology with surfaces of a uniform roughness. By introducing Mg into the synthesis solution, the morphology of the resulting MgO-SiO2 composites changed. The aggregation of particles with different sizes could be seen for all the MgO-containing samples. With an increase in the MgO content, the size of the particles on its surface decreased; however, all the samples still exhibit rough surfaces. The particle size distribution of the samples was also analyzed using dynamic light scattering (DLS) (Fig. S1). Maybe caused by agglomeration of the nanoparticles, the particle sizes obtained from DLS are larger than that of calculated in the TEM images for all the samples. One of the samples (20% MgO-SiO2) was selected for EDX elemental mapping analysis. As expected, both the Si and Mg ions are well-dispersed in the composite sample. N2 adsorption/desorption isotherms of the MgO-SiO2 samples are shown in Fig. 5. Clearly, all isotherms exhibited the typical type IV adsorption curve with a hysteresis loop, which is characteristic of mesoporous materials. The isotherms shift to a larger amount adsorbed with increasing MgO incorporation, indicating an increase of porosity for the composite. The BET-specific surface areas and pore volumes of different samples are summarized in Table 1. The surface area of the pure silica sphere is only 46.5 m2/g, while the 5% MgO-SiO2 sample has a larger surface area of 161.9 m2/g, and this value continues to increase as the MgO content increases. Among the samples, 20% MgO-SiO2 exhibits the largest surface area and pore volume of 326.5 m2/g and 0.65 cm3/g, respectively.

3.2. Adsorption kinetics and isotherm The time profile of Congo red adsorption with 20% MgO-SiO2 and SiO2 spheres was investigated. As shown in Fig. 6, pure SiO2 can only adsorb a very small amount of CR (34 mg/g) and adsorption had reached its maximum within 15 min. The adsorption rate of the MgO-SiO2 sample without any pre-treatment is rather slow. During the first 250 min, the amount of CR adsorbed by the composite sample increased slightly. However, with a further increase in the adsorption time, the amount of CR adsorbed increased dramatically and the adsorbent had reached saturation after approximately 500 min. One reason why very little CR is adsorbed in the first 250 min may be because the nanoparticles of MgO-SiO2 are aggregated. To support this assumption, the 20% MgO-SiO2 sample was treated ultrasonic vibrations to obtain a more uniform dispersion before it was added to the CR solution. As expected, the adsorption rate of the composite sample was greatly improved after ultrasonic treatment. After treating the sample for 30 min, the time to reach adsorption equilibrium was shortened to approximately 300 min.

Fig. 2. High-resolution XPS of (a) Mg 1s, (b) Si 2p and (c) O 1s of the 10% MgO-SiO2 sample.

114

M. Hu et al. / Journal of Colloid and Interface Science 510 (2018) 111–117

Fig. 3. TEM images of (a) SiO2, (b) 5% MgO-SiO2, (c) 10% MgO-SiO2, (d) 20% MgO-SiO2, (e) 30% MgO-SiO2 and (f) EDX analysis of the 10% MgO-SiO2 sample.

Fig. 4. SEM images of (a) SiO2, (b) 5% MgO-SiO2, (c) 10% MgO-SiO2, (d) 20% MgO-SiO2, (e) 30% MgO-SiO2 and (f–j) elemental mapping images of the 20% MgO-SiO2 sample at 9223  magnification.

Further increasing the treatment time to 50 min reduced the equilibrium time to within 150 min. The adsorption equilibrium isotherms of CR to SiO2 and the SiO2-MgO composites are shown in Fig. 7. As the initial concentration of CR increased, the equilibrium adsorption capacity (i.e., qe) of the composites gradually increased up to a maximum value. The highest adsorption capacity of pure SiO2 for CR is 182 mg/g, with the MgO introduction, the uptake of CR enhanced significantly to 565 mg/g for the sample with 5% MgO-SiO2. With a further increase in the MgO content in the composites, the adsorption capacity gradually increased as well. Both the 20% MgO-SiO2 and the 30% MgO-SiO2 samples exhibit similar adsorption capacities and the values are the highest among these composites (4000 mg/g). This result was also confirmed using TGA analysis as shown in Fig. S2.

Two adsorption isotherm models, the Langmuir and Freundlich equations, were used to further describe the above adsorption equilibria. The Langmuir model [32–34] is used for monolayer sorption, and the Freundlich model is used to describe multilayer adsorption patterns [35,36] (Supporting Information). The fitting plots of these models are shown in Fig. S3. The Langmuir model gave higher correlation coefficients than those of the Freundlich equations as seen in Table S1. Therefore, it can be concluded that the Langmuir isotherm can be used to predict the adsorption of CR on to MgO-SiO2 adsorbent. The monolayer adsorption capacities calculated from the Langmuir model agree with the experimental adsorption capacities. Table 2 lists the maximum adsorption capacity of CR onto various MgO adsorbents. The 20% MgO-SiO2 sample tested in this work has the highest adsorption capacity (4000 mg/g).

115

M. Hu et al. / Journal of Colloid and Interface Science 510 (2018) 111–117

800

400

700

SiO2

20% MgO-SiO2

600

5% MgO-SiO2

qe (mg/g)

300

5% MgO-SiO2 SiO2

200

(A)

30% MgO-SiO2 10% MgO-SiO2

3

Amount adsorbed (cm /g, STP)

500

500 400 300 200

100

100

0 0.0

0

0.2

0.4

0.6

0.8

0

1.0

100

200

300

400

500

600

700

800

Ce (mg/L)

P/P 0 Fig. 5. Nitrogen adsorption-desorption isotherms of the pure SiO2 sample and MgO-SiO2 samples.

(B)

5000

Samples

SBET m2/g

Vt cm3/g

SiO2 5% MgO-SiO2 10% MgO-SiO2 20% MgO-SiO2 30% MgO-SiO2

46.5 161.9 294.5 326.5 320.6

0.16 0.26 0.41 0.65 0.45

qe (mg/g)

4000 Table 1 Surface area and porosity data of MgO-SiO2 composites.

3000 2000

10% MgO-SiO2

1000

20% MgO-SiO2 30% MgO-SiO2

0 0

50

100

150

200

250

300

Ce (mg/L) 2500

Fig. 7. Adsorption isotherm curves of CR adsorption on (a) SiO2 and 5% MgO-SiO2 and (b) 10% MgO-SiO2, 20% MgO-SiO2 and 30% MgO-SiO2. (Ce represents the equilibrium concentration and qe is the equilibrium adsorption capacity.)

qt (mg/g)

2000 1500

Table 2 Comparison of the adsorption capacities of different adsorbents for CR.

1000

Pure SiO2 Ultrasonic for 50 min Ultrasonic for 30 min Without Ultrasonic

500 0 0

100

200

300

400

500

600

700

Time (min) Fig. 6. Time profiles of CR adsorption on pure SiO2 and 20% MgO-SiO2 with and without ultrasonic treatment (Initial dye concentration: 500 ppm, adsorbent dosage: 0.2 g/L, 25 °C).

It is worth noting why the MgO-SiO2 samples have such a high adsorption capacity toward Congo red. According to the previous studies [22,41], the adsorption of ionic dyes or organic pollutants may be associated with electrostatic attractions and surface complexation. Congo red is a weak acid with a pKa value of 4.1, whereas MgO exhibits the properties of a strong base with an isoelectric point of approximately 12 [40]. The natural pH value (approximately 7.0) of the Congo red solution is much lower than the isoelectric point of MgO. Thus, the adsorption of Congo red on the composite samples can primarily be attributed to electrostatic attraction. As shown in Fig. S4, in the pH range of 4–10, the adsorption capacity of 20% MgO-SiO2 changes little with a change of pH. Unfortunately, further increasing pH to >10 or decreasing pH to