Membranes made of carbon nanotubes and cellulose acetate with polyacrylic acid were designed in order to study their properties and their applicability for chromium removal. The membranes were prepared by phase inversion method using cellulose acetate and polyacrylic acid. Carbon nanotubes were added to the membrane during their process of synthesis in proportions of 1% by weight. The pores in the material are formed in layers, giving the effect of depth and forming a network. Both the carbon nanotubes and membranes were characterized by IR, Raman, and SEM spectroscopy. In addition, the concentration of acidic and basic sites and the surface charge in the materials were determined. The concentration of acid sites for oxidized nanotubes was 4.0 meq/g. The removal of Cr(VI) was studied as a function of contact time and of initial concentration of Cr(VI). The removal of Cr(VI) (~90%) mainly occurs in a contact time from 32 to 64 h when the initial concentration of Cr(VI) is 1 mg/L.
Many investigations have been focused on the development of new polymeric membranes for different applications [
The addition of fillers in polymers is an attractive method in order to obtain materials with novel properties. Nowadays, carbon nanotubes are used as fillers to produce new materials. However, when new applications for carbon nanotubes are proposed, these materials need to be supported on other materials to obtain attractive composite structures with a better performance than the performance shown by pure initial components. The physicochemical characteristics of the membranes obtained from the polymer mixture can be changed if the membrane is prepared using different mixing ratios of the polymers [
In this work, synthetic polymer membranes made from polyacrylic acid, cellulose acetate, and graphitic materials were prepared and characterized in order to evaluate their properties, stability, and possible application in chromium removal. The use of polymers provides support to carbon nanotubes and permits their use in continuous processes for the removal of ions. This idea is based on previous studies showing that these materials have great potential in the adsorption of heavy metals [
Previous studies have demonstrated the technical feasibility of using carbon nanotubes (CNTs) for the removal of divalent ions (Ni2+, Cu2+, Pb2+, Cd2+, Zn2+, and Co2+) from aqueous solutions. These studies were based mainly on chemical interactions that occur between ions and surface groups of the CNTs. CNTs were regenerated and reused for several cycles of water treatment; this feature is ideal if you want to have low-cost processes [
Chromium is a heavy metal and has been identified as a contaminant of water and soil. Chromium pollution is produced from numerous industrial activities, such as the textile industry, the chemical industry, and metallurgy [
The polymer made from cellulose acetate (CA) and polyacrylic acid (PA) can be used as support of the graphitic materials. Cellulose acetate is one of the most commonly used materials for polymeric membrane fabrication [
In order to carry out this work, the following materials were used: cellulose acetate (Sigma Aldrich) with a molecular weight of 50,000 by gel permeation chromatography (GPC) and a degree of acetylation of 39.7% w and polyacrylic acid in aqueous solution (Sigma Aldrich) with a molecular weight of 30,000 g mol−1 and a percentage of 35% w. All commercial reagents were used without any further purification step. Polymer membranes were prepared according to a procedure previously reported [
Multilayer nanotubes (Cheaptubes Co.), with purity greater than 95%, 10 to 30 microns in length, and diameters between 20 and 30 nm, were oxidized through a chemical treatment based on a mixture of sulfuric acid (Jalmek, purity: 95–98%, MW = 98.08 g/Mol) and nitric acid (Fermont, purity: 69%, MW = 63.01 g/mol) in a 3 : 1 volume ratio. The oxidation was conducted at 85°C for 3 h. When the reaction finished, the sample was washed with distilled water and dried for 12 h at 40°C.
The characterization of polymer and carbon nanotubes was made by Fourier Transform Infrared spectroscopy with attenuated total reflectance (FTIR-ATR Vertex Model 70) in pressed KBr pellets (100 mg KBr and 1 mg of sample) of graphitic materials. For FTIR spectroscopy, the samples were dried at 333 K for 24 h. The characterization by FTIR was complemented with a Raman analysis (Renishaw Raman Microscope Invia Reflex, Wotton-under-Edge, UK). The thermal properties were studied using a TGA analyzer (TA-Q500, TA Instruments). The TGA measurements were performed using nitrogen in a temperature range of 25–700°C (10°C min−1). The morphology of cross-linked polymer was investigated with the aid of the scanning electron microscope (Jeol JSM-6610LV) operated in the high vacuum mode at an acceleration voltage of 20 kV and a pressure of 20 Pa; the materials were previously coated with gold. The SEM images for the rest of the materials were determined with an environmental scanning electron microscope (MEBA, Philips: Model XL30) operated in the high vacuum mode too. The effective surface area and pore size distribution of the graphite materials were determined using N2-BET (ASAP 2010 V5.03). About 0.2 g of sample was degassed in nitrogen at 120°C for 6-7 h before undergoing analysis. The pore size distributions with specific surface areas were measured by N2 adsorption/desorption according to the BET method.
The surface charge and point of zero charge of the materials were evaluated using a potentiometric titration method proposed by Loskutov and Kuzin. For the titration, a pH meter (Pinaracle, Model 540) was used. The experiments were conducted in a 50 mL flask. Initially, 100 mg of adsorbent material and 20 mL of 0.1 M NaCl were added to the vessel. Then, a sample between 0.05 and 5 mL of a titrant solution was added to obtain a curve over the range of pH. The solution with the adsorbent material was kept under constant stirring for 48 hours until it reached equilibrium. Then, the final pH value of titrant solution was measured. Thus, we obtained a curve that showed the variation of pH value of the solution as a function of volume of titrant solution with adsorbent. Another similar curve for the solution without adsorbent was obtained. These plots were used to determine the volumes of the different titrants solutions at the same final pH. The surface charge and point of zero charge of the materials were evaluated using the equations proposed by Loskutov and Kuzin. Finally, a curve that showed the variation of surface charge as a function of pH was obtained. The point of zero charge (PCC) was placed as the pH value where the surface charge is zero. The concentration of acidic and basic sites in the materials was determined using a method of acid-base titration proposed by Böehm. For the titration, a pH meter (Pinaracle, Model 540) was used. The experiments were conducted in a 50 mL flask. The acid sites were neutralized with basic standard solution (0.1 N NaOH) and the basic sites with an acid standard solution (0.1 N HCl). Initially, a neutralizing solution and 1 g of material were added to the vessel. The flask was partially immersed in water at 25°C and it was left in contact with water for 5 days to reach equilibrium. The flask was manually stirred two times a day. Then, a sample of 10 mL was taken and titrated with 0.1 N solution of HCl or NaOH. The titration was performed in triplicate for all cases. The concentrations of acidic and basic sites were calculated using the equations proposed by Böem.
All experiments were conducted in plastic tubes under ambient conditions by using batch technique. A sample of adsorbent material and a certain amount of synthetic solution of Cr(VI) were added to each tube. The tubes were kept in constant agitation. After the period of shaking time, the solid phase was separated from the solution. The contact time (adsorption kinetics), the dose of graphitic material, the effect of pH, the effect of the initial concentration of the synthetic solution, and finally the effect of temperature on the removal were studied. The data of the adsorption kinetics were simulated using pseudo-first-order models, pseudo-second-order models, Elovich model, model function of power, and kinetic rate law for three values of
The IR spectra of raw nanotubes and oxidized nanotubes are compared in Figure
Infrared spectra of the raw nanotubes and the oxidized nanotubes.
Molecular interactions between materials were studied by obtaining their FTIR spectra. The IR spectra of cellulose acetate, polyacrylic acid, and the polymer (AC-PA) are shown in Figure
Infrared spectra of the different polymers.
Different graphitic materials were added to the polymer (AC-PA). The IR spectra of polymer with carbon nanotubes are shown in Figure
Infrared spectra of the polymer (AC-PA) without and with raw and oxidized CNT.
The Raman spectrum of raw carbon nanotubes presents four characteristic peaks, Figure
Raman spectra of the raw CNTs and oxidized CNTs.
A Raman spectrum consists of bands which are caused by inelastic scattering from chemically bonded structures, as shown in Figures
Raman spectrum for cellulose acetate.
Raman spectrum for polyacrylic acid.
Raman spectrum for the polymer (CA-PA).
The spectrum for the polyacrylic acid is shown in Figure
We considered two reaction mechanisms associated with possible interactions between cellulose acetate and acrylic acid. The first mechanism is based on the idea that the oxygen present in the carbonyl group has weak basic properties as shown in Figure
Mechanism based on reaction between carboxylic groups and acetyl groups [
The second mechanism considers that hydrogen bonded to highly electronegative atoms can generate partial charges in the elements, which would transform into electrostatic attraction as shown in Figure
Mechanism based on reaction between carboxylic groups and acetyl groups [
Table
Thermal parameters of polymer (CA-PA) with raw and oxidized CNTs.
Samples | Graphitic material (%) | Thermogravimetric parameters | Residue at 700°C (%) | ||||
---|---|---|---|---|---|---|---|
Stage I | Stage II | ||||||
|
Interval (°C) |
|
Interval (°C) |
|
|||
Polymer (CA-PA) | 0 | 200 | — | — | 271–414 | 360 | 10.3 |
|
|||||||
Polymer (CA-PA) with raw CNTs | 1 | 200 | 190–262 | 239 | 271–412 | 358 | 13.5 |
|
|||||||
Polymer (CA-PA) with oxidized CNTs | 1 | 204 | 193–262 | 264 | 272–410 | 359 | 13.8 |
Thermogravimetric curves of polymer (CA-PA) with raw and oxidized carbon nanotubes (1% w).
The images obtained from electron microscopy are shown in Figure
Sequence of scanning electron microscope images for different polymeric materials.
The BET analysis was applied to carbon nanotubes to determine the effective surface area and the pore size distribution of the materials. The raw nanotubes presented a surface area of 149.72 m2/g and a pore size of 15.99 nm. The oxidized nanotubes showed a surface area of 91.99 m2/g and an average pore size of 24.77 nm. The adsorption-desorption isotherms obtained by BET analysis (Figures
Adsorption-desorption isotherm of raw nanotubes.
Adsorption-desorption isotherm of oxidized nanotubes.
Both the raw nanotubes and oxidized nanotubes show a pore diameter distribution between 10 and 100 nm and a very small portion of pores in a range of 3 to 4 nm, Figure
Pore size distribution in raw nanotubes and oxidized nanotubes.
Both acidic and basic sites were calculated in carbon nanotubes and polymeric materials using a method proposed by Böehm based on an acid-base titration. For the carbon nanotubes, concentration values for only the acid sites were obtained. The value obtained for oxidized carbon nanotubes was 4.0 meq/g. Both acidic and basic sites on the membranes without graphitic material were calculated using the method proposed by Böehm. The concentration of acidic sites in the polymeric material (4.9 meq/g) is 1.25 times higher than the concentration of base sites (3.9 meq/g). The basic sites in the polymeric material may be associated with unreacted sites on the cellulose, while the acid sites can be ascribed to sites vacated in the polyacrylic acid during the synthesis process of the copolymer. For the membranes with carbon nanotubes, no significant changes in the concentration values of the sites were observed.
The surface charge of the carbon nanotubes and reinforced membranes was placed using a method proposed by Loskutov Kuzin based on a potentiometric titration. The zero point load for the carbon nanotubes was determined in a pH value of 5.6. Thus, the material is positively charged at pH values lower than load point zero and negatively charged at pH values higher than load point zero, Figure
Distribution of the surface charge in the different materials: (a) oxidized carbon nanotubes, (b) polymer (CA-PA), and (c) polymer (CA-PA) with carbon nanotubes.
The removal of Cr(VI) from aqueous solutions using polymeric materials made from cellulose acetate and polyacrylic acid with carbon nanotubes was studied as a function of contact time at a pH value of 2.2. From the adsorption kinetics obtained, it can be seen that the removal of Cr(VI) increases when the contact time increases. The removal of Cr(VI) mainly occurs in a contact time from 32 to 64 h when the initial concentration of Cr(VI) is 1 mg/L and the charge of graphitic material in the membranes is 1% by weight, Figure
Chromium adsorption kinetics for polymeric membranes with carbon nanotubes (pH 2.2).
The data of the adsorption kinetics were simulated using pseudo-first-order models, pseudo-second-order models, Elovich model, model function of power, and kinetic rate law for three values of
Constants for the adsorption kinetics of Cr(VI) from polymer membranes with CNTs.
Model/parameters | Polymer-oxidized nanotubes |
---|---|
Pseudo-first-order | |
|
0.1034 |
|
0.9135 |
|
0.9984 |
|
|
Pseudo-second-order | |
|
0.9288 |
|
0.9212 |
|
0.9992 |
The effect of initial concentration of Cr(VI) on its removal was studied using polymeric materials with carbon nanotubes. The data were obtained at a pH value of 2.2 and different temperatures. For the materials studied, the percentage of Cr(VI) removal decreases when its initial concentration increases. When the dose of the graphitic material is constant, the availability of surface adsorption sites also remains fixed; in this way, the removal percentage decrease is due to electrostatic repulsion between ions. When the concentration increases, the competition between ions also increases, thus increasing the electrostatic repulsion.
The kinetic adsorption data were simulated with Langmuir and Freundlich Model, respectively. The results are listed in Table
Parameters of Langmuir Model and Freundlich Model.
Adsorbent | Temperature | Langmuir Model | Freundlich Model | ||||
---|---|---|---|---|---|---|---|
(°C) |
|
|
|
|
|
| |
Polymer (CA-PA) | 25 | 0.340 | 0.570 | 0.9606 | 1.613 | 1.707 | 0.9128 |
35 | 0.211 | 0.851 | 0.9795 | 1.850 | 1.391 | 0.9006 | |
|
|||||||
Polymer-oxidized nanotubes | 25 | 0.843 | 13.118 | 0.9982 | 3.909 | 1.560 | 0.8362 |
35 | 0.833 | 4.570 | 0.9978 | 2.662 | 1.913 | 0.8281 |
The results obtained show that it is possible to design polymers with carbon nanotubes whose pores are formed in layers, giving the effect of depth forming a network. It can be seen that the graphitic material is deposited on the outside of the polymeric material. The adsorption-desorption isotherms obtained by BET analysis showed that the adsorption occurs by a physical mechanism and that the analyzed samples have a hexagonal tubular capillary. Besides, the isotherms of adsorption/desorption obtained for graphite, graphite oxide, and graphene oxide showed characteristics similar to those of the carbon nanotubes. For the carbon nanotubes, concentration values for only the acid sites were obtained. These acid sites can be associated with the presence of carboxylic groups inserted during oxidation of the graphitic materials. The basic sites in the polymeric material may be associated with unreacted sites on the cellulose, while the acid sites can be ascribed to sites vacated in the polyacrylic acid during the synthesis process of the polymer (CA-PA). For the membranes with carbon nanotubes, no significant changes were observed in the concentration values of the sites. The carbon nanotubes are positively charged at pH values lower than load point zero and negatively charged at pH values higher than load point zero, whereas the surface of polymeric membranes is positively charged at pH values higher than the zero point load and negatively charged at pH values lower than the zero point load. Thus, the behavior of the surface charge of the membranes is opposite to the behavior shown by carbon nanotubes.
From these studies of removal of Cr(VI), we can establish the following conclusions: The removal of Cr(VI) using polymeric membranes with and without carbon nanotubes is strongly dependent on the pH values. Besides, the adsorption of Cr(VI) decreases with the increase of pH value. The adsorption of Cr(VI) using polymeric membranes with and without carbon nanotubes is fast in the beginning of the process and then becomes slow with increased contact time. The removal of Cr(VI) takes a considerable time when using polymeric membranes with and without carbon nanotubes as adsorbents. The kinetics of absorption of Cr(VI) can be represented by pseudo-second-order and pseudo-first-order models.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The authors are grateful to Universidad de Guanajuato and PRODEP for financial support. They also thank M. T. Carrillo for her help in revising the paper. J. A. Sánchez-Márquez thanks CONACYT for financial support during his Ph.D. studies, Grant no. 267260.