The aim of this work is to determine paraquat adsorption capacity of zeolite NaX and Al-MCM-41. All adsorbents were synthesized by hydrothermal method using rice husk silica. For Al-MCM-41, aluminum (Al) was added to the synthesis gel of MCM-41 with Al content of 10, 15, 20 and 25 wt%. The faujasite framework type of NaX and mesoporous characteristic of Al-MCM-41 were confirmed by X-ray diffraction. Surface area of all adsorbents determined by N2 adsorption–desorption analysis was higher than 650 m2/g. Al content and geometry were determined by X-ray fluorescence and 27Al nuclear magnetic resonance, respectively. Morphology of Al-MCM-41 were studied by transmission electron microscopy; macropores and defects were observed. The paraquat adsorption experiments were conducted using a concentration range of 80–720 mg/L for NaX and 80–560 mg/L for Al-MCM-41. The paraquat adsorption isotherms from all adsorbents fit well with the Langmuir model. The adsorption capacity of NaX was 120 mg/g-adsorbent. Regarding Al-MCM-41, the 10% Al-MCM-41 exhibited the lowest capacity of 52 mg/g-adsorbent while the other samples had adsorption capacity of 66 mg/g-adsorbent.

INTRODUCTION

Paraquat (1,1′-dimethyl-4,4′-dipyridinium dichloride) or methyl viologen is a herbicide which controls broadleaf weeds and grasses. It is widely used in Thailand including Nakhon Ratchasima province which has the largest area of cassava cultivation. However, paraquat is extremely toxic to humans and animals when contacted via ingestion, through skin contact, or splashed into the eye (Suntres 2002). Paraquat may also damage the kidneys, liver and esophagus. There have been several reports that paraquat was misused as a suicide agent, which caused death when swallowed (National Institutes of Health 2014). With high solubility, paraquat can easily contaminate water. Therefore, it is necessary to investigate paraquat removal from aqueous solution. A convenient method is adsorption which is efficient, fast and inexpensive.

Previously, paraquat removal by adsorption on rice husk silica (RHS) and porous materials synthesized from RHS including zeolite NaY, NaBEA and MCM-41 was reported (Rongchapo et al. 2013). The adsorption capacity increased with aluminum (Al) content in the following order: NaY > NaBEA > MCM-41 > RHS. The proposed mechanism was ion exchange. To further improve the adsorption capacity, a zeolite with higher Al content was investigated in this work. Zeolite NaX was selected because it has the same faujasite (FAU) framework type as NaY and higher Al content. Moreover, improvement in paraquat adsorption on MCM-41 was investigated. Although MCM-41 had a lower adsorption capacity than zeolites, it was the best adsorbent for blue dye in the commercial paraquat (Rongchapo et al. 2013). Thus, it is still interesting to improve its adsorption capacity for paraquat. This could be achieved by adding Al into the framework of MCM-41.

EXPERIMENTAL

Synthesis of adsorbents

Zeolite NaX was synthesized by the hydrothermal method using sodium silicate and alumina trihydrate solutions with a method modified from the work of Robson (2001) and Khemthong et al. (2007). Sodium silicate solution was prepared by slowly adding 13.7 g of RHS into 50 mL of 2.7 M NaOH solution under magnetic stirring. An Al source was prepared by adding 24.4 g of alumina trihydrate into 25 mL of 25 M NaOH solution under magnetic stirring at 100 °C until dissolved. The solution was cooled to 25 °C and mixed with 50 g of distilled water. The sodium silicate solution (55 g) and alumina trihydrate solution (25 g) were separately mixed with 153 mL of 2.5 M NaOH solution. Then, both solutions were quickly mixed together, stirred for 30 minutes, transferred into a polypropylene bottle, capped and sealed with paraffin film. Crystallization was carried out at 90 °C for 14 hours; the sample was cooled down to room temperature and washed with distilled water and dried at 100 °C overnight.

Al-MCM-41 with various Al contents was synthesized using sodium silicate from RHS, hexadecyltrimethylammonium bromide (CTAB, ≥ 99%w/w, Acros) and sodium aluminate with a method modified from the literature (Chen et al. 1997; Preethi et al. 2008). A template solution was prepared by dissolving 4 g of CTAB in 30 mL of distilled water. The Al-MCM-41 was synthesized by the hydrothermal method using a gel containing 10, 15, 20 and 25 wt% of Al. The gel was prepared by adding the CTAB solution to the sodium silicate solution under stirring for 30 minutes followed by addition of sodium aluminate solution, also under stirring until a homogeneous mixture was obtained. The mixture pH was adjusted to 9 by 2 M H2SO4. The mixture was transferred into a Teflon-lined stainless steel autoclave and crystallized at 110 °C for 72 hours. After cooling down to room temperature, the solid was washed with distilled water, separated by centrifugation, dried at 100 °C overnight and calcined at 600 °C for 6 hours.

Table 1

Element content and surface area of all Al-MCM-41 and NaX

Sample Gel Si/Al mole ratio Si/Al mole ratio from XRF Gel Na/Al mole ratio Na/Al mole ratio from XRF BET surface area (m2/g) 
10%Al-MCM41 7.78 10.92 8.84 2.78 847 
15%Al-MCM41 5.18 8.77 6.47 1.65 805 
20%Al-MCM41 3.88 7.48 5.02 1.09 832 
25%Al-MCM41 3.11 5.90 4.15 0.53 654 
NaX 1.24 1.27 15.6 1.37 735 
Sample Gel Si/Al mole ratio Si/Al mole ratio from XRF Gel Na/Al mole ratio Na/Al mole ratio from XRF BET surface area (m2/g) 
10%Al-MCM41 7.78 10.92 8.84 2.78 847 
15%Al-MCM41 5.18 8.77 6.47 1.65 805 
20%Al-MCM41 3.88 7.48 5.02 1.09 832 
25%Al-MCM41 3.11 5.90 4.15 0.53 654 
NaX 1.24 1.27 15.6 1.37 735 

Characterization of adsorbents

Powder XRD analysis of the adsorbents NaX and Al-MCM-41 was carried out by using a Bruker D8 ADVANCE with Cu Kα radiation at 40 kV and 40 mA. Si, Al and Na content in the Al-MCM-41 was determined using energy dispersive X-ray fluorescence (EDXRF) (Oxford ED2000). The N2 adsorption–desorption isotherms were obtained from a Micromeritics ASAP 2010 at liquid N2 temperature. The samples were degassed at 300 °C under vacuum for 8 hours before the measurement. Surface area was calculated with the Brunauer–Emmett–Teller (BET) method. Morphology of Al-MCM-41 samples was studied by transmission electron microscopy (TEM) (Tecnai G20 LaB6, FEI-2012) with an accelerating voltage of 200 kV. The sample was suspended in absolute ethanol, dropped on a carbon-coated copper grid and brought to dryness. Moreover, the Al-MCM-41 samples were analyzed by aluminum-27 nuclear magnetic resonance (27Al-NMR) (Bruker AVANCE III 500 MHz).

Paraquat adsorption

Adsorption experiments and calculations were conducted with a procedure described by Rongchapo et al. (2013). In a 125 mL polypropylene bottle, 0.05 g of adsorbent was added into 20 mL of paraquat solution in the concentration range 80–720 mg/L for NaX and to 80–560 mg/L for Al-MCM-41. The mixture was stirred for a given time at room temperature. After 60 minutes, the solution was separated by filtration using a 0.45 μm nylon syringe filter. The filtrate was analyzed for paraquat using a UV–visible light spectrophotometer (Varian Cary 1E) at 257 nm. The Langmuir and Freundlich isotherms were used to explain paraquat adsorption on NaX and Al-MCM-41.

The Langmuir adsorption isotherm is defined by Equation (1) 
formula
1
where KA is the Langmuir adsorption equilibrium constant, XE (mg/L) is the concentration of the adsorbate in a solution at equilibrium, q (mg/g) is the amount of adsorbate adsorbed per gram of the adsorbent, and qm (mg/g) is the amount of the adsorbate adsorbed to form a monolayer coverage. The data can be fit in a linear form by Equation (2) 
formula
2
The Freundlich adsorption isotherm can be defined by Equation (3) 
formula
3
where both Kf and n are Freundlich constants. The adsorption data can be fit in a linear form by Equation (4) 
formula
4

RESULTS AND DISCUSSION

Properties of NaX and Al-MCM-41

The X-ray diffraction (XRD) patterns of NaX and Al-MCM-41 are shown in Figure 1(a) and 1(b), respectively. Characteristic XRD peaks of NaX (Figure 1(a)) were sharp and strong indicating that the obtained NaX had high crystallinity. The hkl of strong peaks are assigned by comparing with a simulated XRD powder pattern of hydrated NaX (Treacy & Higgins 2001). The XRD patterns of Al-MCM-41 (Figure 1(b)) were similar to those in the literature (Chen et al. 1997; Preethi et al. 2008). The peaks of alumina were not observed, indicating a good dispersion of Al, probably in the framework of MCM-41. As the Al content increased, the peak intensities were lower possibly from either a less uniform structure or secondary scattering from oxides of Al. Moreover, the main peaks shifted slightly to higher values with Al loading, indicating the expansion of d spacing. A similar observation was reported by Preethi et al. (2008) for Al-MCM-41 with Si/Al mole ratio of 25, 50, 75 and 100 and Chen et al. (1997) for Al-MCM-41 with Si/Al mole ratio of 10, 50, 100 and ∞. This result indicated the drawback of Al addition to the formation of the mesoporous structure.

Figure 1

(a) XRD pattern of NaX showing hkl of the strong peaks and (b) Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%. The peaks in the pattern of Al-MCM-41 correspond to the plane 100, 110, 200 and 210.

Figure 1

(a) XRD pattern of NaX showing hkl of the strong peaks and (b) Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%. The peaks in the pattern of Al-MCM-41 correspond to the plane 100, 110, 200 and 210.

N2 adsorption–desorption isotherms of NaX and Al-MCM-41 are shown in Figure 2(a) and 2(b), respectively. The isotherm of NaX was type I, which is a characteristic of microporous material such as zeolites. The adsorbed volume increased rapidly at the beginning due to adsorption mainly in micropores and on external surface area. Then, the amount adsorbed became nearly constant indicating a complete monolayer coverage. The isotherms of all Al-MCM-41 samples (Figure 2(b)) were type IV. The rapid uptake at low relative pressure indicated monolayer coverage, and the gradual increase afterward indicated multilayer adsorption. An increase at the relative pressure between 0.25 and 0.45 corresponded to condensation in mesopores (Rouquerol et al. 1999). A hysteresis loop at the relative pressure of 0.5–1.0 could be classified as type H4 in the IUPAC classification, contributing to slit-shaped pores (Rouquerol et al. 1999). This hysteresis loop was not observed in Al-free MCM-41 (Rongchapo et al. 2013). The addition of Al to the synthesis gel could cause defects in the Al-MCM-41. The hysteresis loop in a high pressure range was also reported in Al-MCM-41 synthesized from alkaline-treated ZSM-5 (Zhang & Yan 2013). In addition, the work by González et al. (2009) showed a similar adsorption isotherm from Al-MCM-41 with Si/Al mole ratio of 20, 30 and 50, but a hysteresis loop was not reported. They also reported that the wall thickness decreased with the higher Al content.

Figure 2

(a) N2 adsorption–desorption isotherm of NaX and (b) Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%. The filled symbols are data from adsorption and the open symbols are data from desorption.

Figure 2

(a) N2 adsorption–desorption isotherm of NaX and (b) Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%. The filled symbols are data from adsorption and the open symbols are data from desorption.

The BET surface area of NaX and Al-MCM-41 samples are presented in Table 1. Their surface area are in the same range as that of NaY, which was 870 m2/g (Rongchapo et al. 2013). The surface area of 25% Al-MCM-41 was the lowest, suggesting the least uniform.

The Si/Al and Na/Al mole ratio of the gel and solid products from elemental analysis by EDXRF are shown in Table 1. The Si/Al values in the solid products are higher than those in the gel, indicating that only some amount of Al was incorporated in to the structure. The Na/Al values from the gel were higher than that from the solid although the solid products were washed thoroughly with water after the synthesis. When an Al atom is incorporated into the structure of MCM-41 (namely, by replacing a tetrahedral [SiO4] unit with [AlO4]), a negative charge is generated and a charge balancing cation is needed. Thus, the ideal Na/Al mole ratio is one. However, the only sample that gave Na/Al mole ratio of one was 20%Al-MCM-41. In the other samples, the additional Na cation probably bonded to the surface.

Figure 3 displays 27Al-MAS-NMR spectra of the Al-MCM-41. The main peak was observed around 55 ppm corresponding to tetrahedral coordinated Al (Chen et al. 1997; González et al. 2009). The intensity increased with Al content, consistent with the XRF results.

Figure 3

27Al-MAS NMR spectra of the Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%.

Figure 3

27Al-MAS NMR spectra of the Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%.

TEM images of Al-MCM-41 are shown in Figure 4. Ordered hexagonal mesostructure, defects and macropores were observed in all samples, confirming that loading with Al lowers the sample uniformity.

Figure 4

TEM images of Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%.

Figure 4

TEM images of Al-MCM-41 with Al amount of 10, 15, 20 and 25 wt%.

Paraquat adsorption on NaX and Al-MCM-41

Paraquat adsorption isotherms of NaX and Al-MCM-41 are shown in Figure 5. The adsorption capacity of NaX was higher than that of all Al-MCM-41 samples in all concentrations. Because NaX had higher Al content, it had higher ion exchange capacity than Al-MCM-41. For Al-MCM-41, the adsorption capacities were higher than that for MCM-41 (Rongchapo et al. 2013). When Al is incorporated into the framework of MCM-41, a negative charge is generated and a charge balancing cation is required. The charge balancing cation can exchange with paraquat, which is a dication. Ten percent Al-MCM-41 had the lowest adsorption capacity due to the lowest Al content. The adsorption capacities of 15% Al-MCM-41 and 20% Al-MCM-41 were similar probably because their surface area and Al content from XRF were not much different. In the case of 25% Al-MCM-41, it had the highest Al content but the lowest surface area. Thus, its adsorption capacity was not different from those of other Al-MCM-41 samples.

Figure 5

Isothermal adsorption of paraquat (analytical reagent grade) on NaX and Al-MCM-41 at room temperature. Each experiment was done in triplicate with 0.05 g of adsorbent and 20 mL of 80–560 ppm paraquat solutions. All isotherms fit well with the Langmuir model.

Figure 5

Isothermal adsorption of paraquat (analytical reagent grade) on NaX and Al-MCM-41 at room temperature. Each experiment was done in triplicate with 0.05 g of adsorbent and 20 mL of 80–560 ppm paraquat solutions. All isotherms fit well with the Langmuir model.

When the adsorption data from each sample were fitted with Langmuir and Freundlich isotherms, the better fit was obtained from the Langmuir model with R2 coefficient more than 0.99 (Table 2). The maximum capacities calculated from the Langmuir equation are included in Table 2. Compared to the previous work (Rongchapo et al. 2013), the adsorption capacities from NaX and Al-MCM-41 were still lower than that from NaY. Although both NaX and NaY synthesized with RHS have the same FAU framework type and cavity size, they were different in Si/Al mole ratio and surface area. The Si/Al mole ratio of NaX and NaY was 1.27 and 2.2 (Rongchapo et al. 2013), respectively; the surface area of NaX and NaY was 735 m2/g and 870 m2/g (Rongchapo et al. 2013), respectively. The lower surface area may be the reason for the lower adsorption capacity.

Table 2

Correlation coefficient (R2) from fitting paraquat adsorption data of MCM-41 and Al-MCM-41 to the Freundlich and Langmuir isotherms; and the maximum adsorption capacity calculated from the Langmuir model

    Langmuir
 
  
  Freundlich   Maximum capacity   
Adsorbents R2 R2 (mg/g-adsorbent) Reference 
NaY 0.8343 0.9997 185.2 Rongchapo et al. (2013)  
NaX 0.9912 0.9940 120.3 This work 
MCM-41 0.8110 0.9859 21.3 Rongchapo et al. (2013)  
10%Al-MCM41 0.8853 0.9986 51.6 This work 
15%Al-MCM41 0.9220 0.9982 66.4 This work 
20%Al-MCM41 0.8753 0.9991 66.1 This work 
25%Al-MCM41 0.8923 0.9997 65.6 This work 
MCM-41 (from mixed templates) N/A 1.00 77.4 Brigante & Avena (2014)  
    Langmuir
 
  
  Freundlich   Maximum capacity   
Adsorbents R2 R2 (mg/g-adsorbent) Reference 
NaY 0.8343 0.9997 185.2 Rongchapo et al. (2013)  
NaX 0.9912 0.9940 120.3 This work 
MCM-41 0.8110 0.9859 21.3 Rongchapo et al. (2013)  
10%Al-MCM41 0.8853 0.9986 51.6 This work 
15%Al-MCM41 0.9220 0.9982 66.4 This work 
20%Al-MCM41 0.8753 0.9991 66.1 This work 
25%Al-MCM41 0.8923 0.9997 65.6 This work 
MCM-41 (from mixed templates) N/A 1.00 77.4 Brigante & Avena (2014)  

N/A: Not available.

For Al-MCM-41, the addition of Al into the framework of MCM-41 increased the adsorption sites but the Al content was still much lower than that in NaX and resulted in the lower adsorption capacity. Moreover, high Al content was not suitable because it could lead to more defects and lower surface area as seen in 25% Al-MCM-41. Another strategy to improve the adsorption of MCM-41 was reported by Brigante & Avera (2014), using mixed surfactant to produce MCM-41 with uniform particle size (around 1.5 μm).

CONCLUSIONS

Zeolite NaX was synthesized by the hydrothermal method using RHS. Its structure was confirmed by XRD; its surface area determined from N2 adsorption–desorption was 735 m2/g; and the Si/Al mole ratio determined by XRF was 1.27. The paraquat adsorption capacity of NaX was 120.3 mg/g-adsorbent.

Al-MCM-41 samples with various Si/Al mole ratio were synthesized by the hydrothermal method by adding an Al source to the synthesis gel. The mesoporous structure was confirmed by XRD, N2 adsorption–desorption and TEM; the Si/Al mole ratio was determined by XRF; and the tetrahedral coordination of Al was confirmed by 27Al NMR. Moreover, macropores and defects were observed in the TEM images. The paraquat adsorption capacity from all Al-MCM-41 samples was lower than zeolite NaX. The paraquat adsorption isotherms of NaX and Al-MCM-41 fit well with the Langmuir model.

ACKNOWLEDGEMENTS

The scholarship for W. Rongchapo is from the Royal Golden Jubilee PhD Program (RGJ) from the Thailand Research Fund and Suranaree University of Technology (contract number PHD/0163/2552).

REFERENCES

REFERENCES
Chen
X.
Huang
L.
Ding
G.
Li
Q.
1997
Characterization and catalytic performance of mesoporous molecular sieves Al-MCM-41 materials
.
Catal.
Lett.
44
,
123
128
.
González
F.
Pesquera
C.
Perdigón
A.
Blanco
C.
2009
Synthesis, characterization and catalytic performance of Al-MCM-41 mesoporous materials
.
Appl. Surf. Sci.
255
,
7825
7830
.
Khemthong
P.
Wittayakun
J.
Prayoonpokharach
S.
2007
Synthesis and characterization of zeolite LSX from rice husk silica
.
Suranaree J. Sci. Tech.
14
,
367
379
.
National Institutes of Health
2014
.
Preethi
M. E. L.
Revathi
S.
Sivakumar
T.
Manikandan
D.
Divakar
D.
Rupa
A. V.
Palanichami
M.
2008
Phenol hydroxylation using Fe/Al-MCM-41 catalysts
.
Catal. Lett.
120
,
56
64
.
Robson
H.
(ed.)
2001
Verified Syntheses of Zeolitic Materials
,
2nd revised edn.
Elsevier Science, Amsterdam, The Netherlands
.
Rongchapo
W.
Sophiphun
O.
Rintramee
K.
Prayoonpokarach
S.
Wittayakun
J.
2013
Paraquat adsorption on porous materials synthesized from rice husk silica
.
Water Sci. Technol.
68
,
863
869
.
Rouquerol
F.
Rouquerol
J.
Sing
K.
1999
Adsorption by Powders and Porous Solids
.
Academic Press
,
London
.
Treacy
M. M. J.
Higgins
J. B.
(eds)
2001
Collection of Simulated XRD Powder Patterns for Zeolites
,
Elsevier, Amsterdam, The Netherlands
.