Carbon nanotubes (CNTs) are found to be promising materials in various applications
(Iijima, 1991). Many researches investigating the physical,
chemical and electrical properties CNTs have been conducted. Moreover, the manipulation
of CNTs with other polymers and functional groups became an attractive field
as it offers solutions for overcoming the difficulties of insolubility of CNTs
and obtaining new composites with unique properties (Balasubramanian
and Burghard, 2005). Carbon nanotubes have been manipulated for applications
in water treatment and the removal of heavy metals from water. It has been used
for many types of heavy metals and proved effective removal (Hsieh
and Horng, 2007).
Chitosan (CS) is a cationic polysaccharide with a dominant positive charge
and high molecular weight (=106 g mol-1). The common sources
of Chitosan are the shells of crabs and lobsters, from which chitin is obtained
and then deacetylated to produce chitosan (Okuyama et
al., 2000). The reactivity of Chitosan is due to the presence of the
amino group and primary and secondary hydroxyl groups at the positions C-2,
C-3 and C-6, respectively, which made chitosan an effective polymer for many
applications especially water treatment (Hong and Samuel,
2000; Divakaran and Pillai, 2002; Chi
and Cheng, 2006). Many works have been done to immobilize chitosan onto
carbon nanotubes for applications such as biosensors, gene and drug delivery
(Ke et al., 2006; Liu et
al., 2005; Wu et al., 2007). The attachment
of chitosan onto the surface of CNTs has been done through covalent and noncovalent
linkages. In this work we report the covalent immobilization of raw chitosan
onto CNTs and its first application for removal of lead metal from water. The
obtained nanocomposite was characterized by Fourier transform infrared spectroscopy
(FTIR) and Field Emission Scanning Electron Microscopy (FESEM) and the removal
efficiency was measured using Atomic Adsorption Spectrometry (AAS).
MATERIALS AND METHODS
Materials: Multiwall carbon nanotubes produced by chemical vapour deposition
(95% purity, 0.04-0.05 g cm-3 bulk density, with an outer diameter
of 10-20 nm and inner diameter of 5-10 nm and length 10-30 μm) was purchased
from Chinese science academy. Chitosan with 88% degree of deacetylation was
provided by department of biotechnology engineering, IIUM. Nitric acid (65%),
sulphuric acid (98%), NaOH pellets, 99% acetic acid, thionyl chloride (>99%),
anhydrous tetrahydrofuran THF and anhydrous dimethyl formamide DMF were all
purchased from MERK, Germany. 50 mg L-1 Lead stock solution from
HACH was used. pH adjustments were done using 0.1 M hydrochloride and/or 0.1
M sodium hydroxide.
Instruments: FESEM (JEOL, JSM-6700F) and FTIR (Bruker, TENSOR 27) were used for the characterization of the nanocomposite. Atomic Adsorption Spectroscopy (AAS) was used for measurement of the residual lead concentrations.
Preparation of the nanocomposite: CNTs were first functionalized
with carboxyl group by sonication at 40 °C for 2 h in acid mixture (65%
HNO3 and 98% H2SO4, ratio 1:3 (v/v)). The oxidized
carbon nanotubes (MWCNT-COOH) were suspended in a solution of thionyl chloride
in order to activate the carboxylic groups by conversion into acyl chloride
groups so that amidation can occur with the chitosan. The solution was stirred
for 24 h at 70°C in a necked flask with a reflux condenser and then separated
by membrane filtration and the obtained solid was washed with THF and then dried
in the oven to get acyl chloride functionalized nanotubes (MWCNT-COCl). Chitosan
was mixed with MWCNT-COCl in DMF. The mixture was stirred at 120 °C for
96 hours under nitrogen atmosphere. After reaction, the mixture was filtered
through a 0.22 μm polyether sulfone membrane. Finally, the product was
dried in the oven to get chitosan immobilized multiwall carbon nanotubes MWCNT-CS.
Moreover, the morphology of raw MWCNTs, carboxyl-functionalized MWCNTs, MWCNT-CS and the chitosan were examined using FTIR. In addition, FESEM along with Energy Dispersive X-ray (EDX) analysis were performed for raw, oxidized and chitosan immobilized MWCNTs in order to confirm the functionalization and attachments.
Adsorption experiments: Adsorption experiments were carried out at room temperature by shaking the series of bottles containing the desired dose of adsorbent (20-60 mg L-1) with predetermined concentration of lead (0.1-0.5 mg L-1). The pH (5-7) was adjusted using 0.1 M hydrochloric acid and/or 0.1 M sodium hydroxide. At the end of the run time (20-100 min) the samples where filtered and the residual concentrations were measured by Atomic Adsorption Spectroscopy (AAS).
RESULTS AND DISCUSSION
Immobilization: Figure 1 shows the FTIR show the spectra
of (a) the raw MWCNTs, (b) the MWCNT functionalized oxidation in acid mixture,
(c) the chitosan and (d) the MWCNT-CS nanocomposite. It can be observed that
very weak peaks appear in Fig. 1 (a) in the range 1850-2000
and 1750-1845 cm-1 corresponding to the C = O (Kim
et al., 2006). However, new peaks appear at 2970, 1738, 1434, 1365
and 1216 cm-1 in (b), the first peak corresponds to the OH
stretching, while the other peaks correspond to the COOH (Wu
et al., 2007). These results indicate that the oxidation resulted
in sufficient peaks for the immobilization to occur. Figure 1
also shows the FTIR spectrum of chitosan. The peak at 1645 cm-1 (-NH2)
and the small band at 1587 cm-1 assigned to N-H and amide groups
of chitosan were replaced by new bands at 1633 and the sharp band at 1538 cm-1,
suggesting that some amino groups were converted into amide groups. Moreover,
the bands at 1150, 1021 and 894 cm-1 (d), corresponding to the glucopyranose
rings of chitosan, implying the attachment of chitosan to the CNTs. Furthermore,
the small band at around 1335 cm-1 suggests that the primary hydroxyl
groups partially participated in the reaction. Also the wide O-H and N-H absorption
bands of CS at around 3287 cm-1 (c) shifted to 3328 cm-1
and became broader (d) suggesting the above conclusion as well (Ke
et al., 2006).
Figure 2 shows the FESEM of raw, oxidized and Chitosan immobilized
CNTs. The raw CNTs have a snake-like shape with very smooth surface (Fig.
2a). The oxidized MWCNT (Fig. 2b) appear shorter than
the raw MWCNT. Moreover, the surface of the raw MWCNT is smoother than the oxidized
one due to the damage to the surface caused by the acid treatment. In addition,
the oxidation resulted in reduction of the MWCNT diameter; the average diameter
of the pristine CNT is 50 nm while the oxidized one is around 30 nm. Figure
2c shows the chitosan immobilized MWCNT.
||FTIR of (a) raw MWCNT, (b) oxidized MWCNT, (c) chitosan and
(d) the chitosan-MWCNT nanocomposite
||FESEM of (a) Raw MWCNTs (b), MWCNT-COOH (c) and MWCNT-CS
Adsorption: Based on the ANOVA analysis of residual concentrations after the adsorption runs, a quadratic model was selected for the fitting of the data. The model in Eq. 1 describes the effect of each factor beside the interactions between the factors. The model is considered significant with p-value less than 0.05 for most of the parameters of the model. The corelation factor R2 value of the model is 78.45%.
The regression equation in terms of the actual factors:
The regression equation shows that as pH for example, increases the residual concentration of lead decreases, this would be due to the protonation of the -NH group of the chitosan that is immobilized in the surface on the MWCNTs which lead to better chelating of the metal ion and thus a reduction in the residual concentration.
From the analysis the optimal values are 0.3 mg L-1 for lead initial concentration, 40 mg L-1 for the CNT-CS dosage, pH 6 and 150 rpm agitation speed. Those values correspond to the center point and gave the maximum removal (98.4%).
A novel MWCNT-CS nanocomposite was prepared through the immobilization of the chitosan polymers onto the surface of the carboxyl oxidized MWCNTs. The attachment of chitosan to MWCNT-COOH was verified by FTIR and FESEM. The obtained nanocomposite was used as an adsorption material for lead removal from aqueous solutions to achieve water quality standard of 0.05 mg L-1 as stated by the Malaysian department of environment (DOE). The optimal conditions corresponding to the maximum removal percentage were those which correspond to the center point of the design values. In addition, the achieved residual concentration (0.008 mg L-1) is much less than the Malaysian DOE requirement for the waste water (0.05 mg L-1) and it is even less than the maximum allowable concentration for the drinking water (0.01 mg L-1). A quadratic empirical model showing the interaction of the parameters and their effect on the residual concentration has been obtained by analysis using design expert.
This project is gratefully funded by the Research Endowment Fund Type B (EDW B0703-47), Research Management Center, International Islamic University Malaysia (IIUM).