Nanosized metal oxide, Titania, provides high surface area and specific affinity for the adsorption of heavy metals, including arsenic (As), which is posing a great threat to the world population due to its carcinogenic nature. In this study, As(III) adsorption was studied on pure and metal- (Ag- and Fe-) doped Titania nanoparticles. The nanoparticles were synthesized by liquid impregnation method with some modifications, with crystallite size in the range of 30 to 40 nm. Band gap analysis, using Kubelka-Munk function showed a shift of absorption band from UV to visible region for the metal-doped Titania. Effect of operational parameters like dose of nanoparticles, initial As(III) concentration, and pH was evaluated at 25°C. The data obtained gave a good fit with Langmuir and Freundlich isotherms and the adsorption was found to conform to pseudo-second-order kinetics. In batch studies, over 90% of arsenic removal was observed for both types of metal-doped Titania nanoparticles from a solution containing up to 2 ppm of the heavy metal. Fixed bed columns of nanoparticles, coated on glass beads, were used for As(III) removal under different operating conditions. Thomas and Yoon-Nelson models were applied to predict the breakthrough curves and to find the characteristic column parameters useful for process design. The columns were regenerated using 10% NaOH solution.
Naturally occurring elemental arsenic is ubiquitous and is present in both organic and inorganic forms. It is the 20th most abundant in earth’s crust, 14th in seawater, and 12th in the human body [
In natural water, arsenic exists as inorganic arsenate (As(V)) and arsenite (As(III)). As(III) is mainly found as arsenious acid (H3AsO3) and As(V), occurs as anionic species (
Humans are exposed to arsenic contamination mainly through ingestion, inhalation, or skin adsorption; however, among these ingestion is the most predominant form of arsenic intake. Large doses of arsenic can result in acute toxic effects like gastrointestinal symptoms like vomiting, diarrhea, poor appetite, disturbance of cardiovascular and nervous systems functions, or even death in severe cases [
High concentration of arsenic in drinking water has been reported in many parts of the world, including Argentina, Bangladesh, China, Chile, Canada, Hungary, India, Japan, Mexico, Poland, Taiwan, and USA [
Many technologies are available for the removal of arsenic from water. Among these, adsorption has been widely used because of its simplicity in operation, greater removal efficiency, cost-effectiveness, and adsorbent disposal qualities [
It has been reported that As(V) compared to As(III) is more readily adsorbed by various adsorbents, because As(III) is mainly present as nonionic H3AsO3 in natural water at near neutral pH [
TiO2 due to its chemical stability, nontoxicity, low cost, and resistance to corrosion [
In this study we have used pure and metal-doped titania nanoparticles, coated on glass beads for arsenic removal in a continuous flow system. The performance of titania coated glass beads evaluated through breakthrough studies by varying the column operation conditions, such as type of titania nanoparticles, bed depth, influent flow rate, and influent As(III) concentration. Thomas and Yoon-Nelson models were applied to column study. Column capacities were calculated.
All the solutions were prepared in deionized water (EC < 0.7
Arsenic (III) stock solution (100 ppm) was prepared by dissolving 420 mg of sodium arsenate in 1 L of deionized water. The pH of the solution was adjusted to 7.0 (±0.1), using 0.1 M nitric acid and 0.1 M NaOH. The solution was shaken well and then stored in dark. Further dilutions were made using this solution.
The quantitative determination of arsenic was done by atomic absorption spectrophotometer (AAS vario 6, analytic jena (Germany)) in hydride generation mode. A pH meter (CyberScan 500) was used to adjust the pH of the arsenic stock solution, with 0.1 M HNO3 and 0.1 M NaOH.
Pure and metal-doped titania nanoparticles were prepared using liquid impregnation method [
In 300 mL water in a beaker placed on a magnetic stirrer, 50 grams of GPR titania was added slowly and stirred for 24 hours. The titania suspension was then removed from the stirrer and allowed to stand for 24 hours, dried at 105°C, ground using mortar and pestle, calcinated at 400°C in furnace for 6 hours, and then allowed to cool down at 10°C per minute. The nanoparticles were then transferred to a plastic bottle and placed in dark.
The metal doped nanoparticles were prepared using the same procedure, with 1% molar ratio of the metal salt being added to the solution before adding titania.
S.E.M. analysis of pure and metal-doped titania nanoparticles was carried out with JEOL JSM 6460 scanning electron microscope to observe the metal distribution on the surface of TiO2 in doped species.
The crystal phase and crystalline size of the prepared titania nanoparticles were determined by powder X-ray diffraction analysis. XRD studies of pure and doped TiO2 were carried out using 3040/60 XPert Pro PANalytical X-ray diffractometer using CuK
The average grain size is determined using Scherer formula [
where
To find the wavelength of light needed for the excitation of TiO2 photocatalyst, it is very important to find the band gap [
When photons having energy greater than or equal to the band gap are absorbed by a semiconductor, an electron from the valance band jumps to the conduction band, and depending on the band gap energy there occurs a rise in the absorbency of the semiconductor. The relation between the absorption coefficient ( Direct transition—when the electron momentum is conserved. Indirect transition—when the momentum of electron is not conserved.
The electronic properties of the synthesized pure and metal doped TiO2 were analyzed using the remission function of Kubelka-Munk,
where
and
where
Therefore,
where
while for an indirect transition the equation takes the following form:
The plot of
The removal efficiencies of pure and metal-doped titania were calculated using the following formula:
where
For each sample 100 mL of 0.5 mg/L of As(III), solution was prepared from the stock and transferred into a 250 mL volumetric flask. An amount of 0.5 g of the respective pure or metal-doped titania nanoparticles was added to it and the flask was placed on an orbital shaker at 145 rpm for 60 minutes for the equilibrium to be reached, centrifuged at 4000 rpm, and finally analyzed.
The above process was repeated at different pH values (4, 7, and 10) and removal efficiency at each pH was calculated.
For adsorption studies, 100 mL of As(III) solution of different concentrations (0.1, 0.2, 0.4, 0.8, 1.5, 3, and 6 ppm) wastaken in a 250 mL volumetric flasks. An amount of 0.35 g of the concerned nanoparticles was added to it, placed on an orbital shaker at 145 rpm for 90 minutes, and centrifuged at 4000 rpm. They were then analyzed with atomic absorption spectrophotometer (AAS vario 6, Analytik Jena (Germany)).
The kinetic studies were performed using 1 liter volumetric flasks. An amount of 500 mL of 0.5 ppm As(III) was taken in it, 0.5 g of pure titania nanoparticles was added to the flask and then placed on an orbital shaker at 145 rpm, and 5 mL of solution was taken out from the flask after specific time (1, 3, 6, 9, 12, 15, 20, 30, 40, 50, 60, 70, 80, and 90 minutes), centrifuged at 4000 rpm, and finally analyzed.
The same process was repeated for the doped titania nanoparticles as well.
The amount of As(III) adsorbed (
where
Etching of the glass beads was done using HF. In this method, the glass beads were dipped in a 10% HF solution in a covered polyethylene bottle for 24 hours, and they were then removed from the solution and rinsed with water to remove any traces of HF left.
Glass beads were coated with nanoparticles using heat attachment method; 5 g of titania nanoparticles was dissolved in 100–200 mL water in a 250 mL titration flask and placed on a shaker at 150 rpm for 15 minutes; 200 grams of etched glass beads were weighed and transferred to the titania suspension, and this was kept on shaking for one hour. The glass beads were then transferred to a Petri dish and were dipped in the titania suspension and dried in oven at 105°C. The glass beads were then carefully removed from the Petri dish, transferred to a china dish, and placed in furnace at 600°C for about 2 hours. The coated glass beads were then cooled, washed with distilled water till none of the nanoparticles attached to the surface gets detached with water, dried in oven at 105°C, and finally transferred to a plastic bottle and kept in dark.
The metal-doped titania nanoparticles were also coated on the glass beads using the same procedure.
Ordinary 100 mL Burette was used for the preparation of columns, with a diameter of 1.5 cm and height of 2 feet. The bottoms of the columns were plugged in with folded aluminum foil just to make a support for the glass beads. Cotton wool or glass wool was not used due to possibility of clogging by the nanoparticles. The columns were packed with the desired nanoparticles coated glass beads to the height needed by weighing the glass beads and then packing them in the columns. The column was then operated in such-manner that a calculated amount of arsenic stock solution was constantly added to it through a graduated cylinder and allowing it to flow along gravity with a constant rate. The rate of flow was constantly checked by measuring the amount flowing per minute after every ten minutes. Once the process started, samples were collected at regular interval using plastic bottles. This process was continued till column exhaustion and the amount of arsenic in different samples was analyzed. From these results “Breakthrough curves” were drawn. Thomas and Yoon-Nelson models were applied to predict the breakthrough curves and to find the characteristic column parameters useful for process design.
When the column was fully exhausted after lengthy column runs, it was regenerated using 10% w/v NaOH solution. Sufficient amount of the regeneration (10 bed volumes) was passed through the column at a very slow flow rate (0.5 mL/min). The column was then rinsed thoroughly with mild warm deionized water at a flow rate of 2 mL/min for about 10 bed volumes.
X-ray Diffraction was used to determine the crystal phase composition and the crystallite size of pure and metal-doped titania nanoparticles. It is shown by the XRD analysis that the titania nanoparticles are 100% in anatase form. No rutile traces were seen in the XRD patterns. The crystalline sizes of the nanoparticles (as calculated from Shrerrer formula) were in the range of 30 to 40 nm (titania (33.83 nm), Ag-TiO2 (34.28 nm), and Fe-TiO2 (37.02 nm)). The XRD patterns are given in Figure
XRD patterns of pure and metal-doped titania nanoparticles.
Particle size and morphology were observed using S.E.M as shown in Figure
S.E.M. micrographs of pure and metal-doped titania nanoparticles. (a) TiO2, (b) Ag-TiO2, and (c) Fe-TiO2.
The band gaps for pure and metal-doped titania nanoparticles were determined by using (
The plot of
Diffused reflectance spectra for direct transition of pure and metal-doped titania nanoparticles.
However, for an indirect transition, (
The plot of
Diffused reflectance spectra for indirect transition of pure and metal-doped titania nanoparticles.
The band gap of TiO2 reported in the literature is 3.2 eV, which corresponds to a wavelength of 385 nm. For the synthesized pure titania nanoparticles, the direct transition (Figure
Direct and indirect band gap values of pure and metal-doped titania nanoparticles.
Nanoparticles | Band Gap | |
---|---|---|
Direct | Indirect | |
TiO2 |
|
|
Ag-TiO2 |
|
|
Fe-TiO2 |
|
|
The removal efficiencies of pure and metal-doped titania nanoparticles are shown in the Table
Removal efficiencies of different nanoparticles used.
Nanoparticles | TiO2 | Ag-TiO2 | Fe-TiO2 |
---|---|---|---|
Removal efficiency (%) |
|
|
|
The removal efficiencies were measured at three different pH values of 4, 7, and 10, as shown in Figure
Effect of pH on removal efficiency.
It is clear from the figure that as we approach the neutral pH the removal efficiency increases. The removal efficiency decreases on moving both to acidic and basic pH. The decrease in removal efficiency is more in basic region than the acidic.
Thus from this point forward, all the processes are carried out at neutral pH, that is, pH
Adsorption studies were carried out to determine the suitable conditions for maximum arsenic removal by the nanoparticles. The pH of the solutions was adjusted at 7. Langmuir and Freundlich models were applied for adsorption studies.
The Langmuir isotherm assumes monolayer adsorption at the adsorbent surface. The linear form of Langmuir adsorption isotherm equation is as follows [
where
Langmuir adsorption isotherms. (a) TiO2, (b) Ag-TiO2, and (c) Fe-TiO2.
The value of
where
The Freundlich isotherm equation is generally given by [
where
The plot of ln
Langmuir and Freundlich isotherm parameters of As(III) adsorption for pure and metal-doped titania.
Nanoparticles | Langmuir isotherm parameters | Freundlich isotherm parameters | |||||
---|---|---|---|---|---|---|---|
|
|
|
|
|
|
| |
TiO2 |
|
|
|
|
|
|
|
Ag-TiO2 |
|
|
|
|
|
|
|
Fe-TiO2 |
|
|
|
|
|
|
|
Freundlich adsorption isotherms. (a) TiO2, (b) Ag-TiO2, and (c) Fe-TiO2.
Several models are available to express the mechanism of adsorption of solute onto the sorbent. To investigate the mechanism of adsorption, both Lagergren and pseudo-second-order equation were used.
To apply the Lagergren equation, an initial As(III) solution of 0.5 ppm (500 mL) was taken in a 1 L flask and 0.5 g of pure titania nanoparticles were added to it and placed on an orbital shaker at 145 rpm, 5 mL of solution was taken out from the flask after specific time (1, 3, 6, 9, 12, 15, 20, 30, 40, 50, 60, 70, 80, 90, 120, and 150 minutes), centrifuged at 4000 rpm and finally analyzed. The kinetic data for this system is shown in Figure
Experimental curve for arsenic adsorption by pure and metal-doped titania nanoparticles.
The Lagergren rate equation [
where
Rearranging (
In the above equation,
The kinetic data in Figure
It is clear from the value of
Lagergren equation parameters for As(III) adsorption on pure and metal-doped titania nanoparticles.
Nanoparticles |
|
|
|
---|---|---|---|
TiO2 |
|
|
|
Ag-TiO2 |
|
|
|
Fe-TiO2 |
|
|
|
The pseudo-second-order kinetic equation is given as [
where
Rearranging this equation, we get
This gets the linear form as
where
Equation (
Thus plotting
Pseudo-second-order rate parameters for As(III) adsorption on pure and metal-doped titania nanoparticles.
Nanoparticles |
|
|
|
|
---|---|---|---|---|
TiO2 |
|
|
|
|
Ag-TiO2 |
|
|
|
|
Fe-TiO2 |
|
|
|
|
Pseudo-second-order kinetic model. (a) TiO2, (b) Ag-TiO2, and (c) Fe-TiO2.
In the beginning, all the As(III) was adsorbed by the column which resulted in zero effluent concentration. As more and more influent was passed through the column, a gradual rise in the effluent concentration was observed. In the down flow mode, when the As(III) bearing water was introduced at the top of the column, most of the As(III) adsorption first occurs in the first few centimeters of the column, called the
By plotting As(III) concentration (mg/L) in the effluent against time (hrs), a breakthrough curve is obtained. On the breakthrough curve, the point at which the effluent As(III) concentration reaches its maximum permissible limit (0.01 mg/L) is called the column breakthrough point and the corresponding time (hrs) as the breakthrough time. The point at which the effluent As(III) concentration reaches 90% of the influent concentration is known as column exhaustion point and the corresponding time (hrs) as exhaustion time.
Thomas and Yoon-Nelson models were applied for column design. Effects of different column parameters on the breakthrough and exhaustion time of column were found. The titania nanoparticles leaching from the 10 cm column during the adsorption studies were very small, 0.015 mg/L.
A successful design of column adsorption procedure needs prediction of the concentration-time profile or breakthrough curve for the effluent. The maximum adsorption capacity of an adsorbent is also required in design. The Thomas model generally serves the purpose well. Thomas model has the following form [
where
The kinetic coefficient
Plot of time (
The Thomas equation coefficients for As(III) adsorption by all the three types of nanoparticles coated glass beads are given in Table
Thomas model parameters for different nanoparticles coated glass beads.
Nanoparticles | TiO2 | Ag-TiO2 | Fe-TiO2 |
---|---|---|---|
|
|
|
|
|
|
|
|
The values of constants for titania nanoparticles coated glass beads were
Thomas model, comparison of experimental and predicted breakthrough curves.
As compared to other models the Yoon and Nelson model, is not only less complicated but also requires no detail related to the characteristics of adsorbate, adsorbent type, and the physical characteristics of the adsorption bed. This model is mainly based on the assumption that for each adsorbate molecule the rate of decrease in the probability of adsorption is proportional to the probability of adsorbate adsorption and the probability of adsorbate breakthrough on the adsorbent [
The linear form of Yoon and Nelson model is as follows:
where
Plot of
The model parameters obtained for the As(III) adsorption on all the three types of titania-based nanoparticles coated glass beads are given in Table
Yoon-Nelson model parameters for different nanoparticles coated glass beads.
Nanoparticles |
|
|
|
---|---|---|---|
TiO2 |
|
|
|
Ag-TiO2 |
|
|
|
Fe-TiO2 |
|
|
|
Alternatively, when
Yoon and Nelson model, comparison of predicted and experimental curves.
This equation develops the relation among the adsorption capacity (
Operational parameters such as flow rate, bed height, arsenic concentration, and variation of nanoparticles are important for column design. The influence of these parameters on the adsorption capacity of nanoparticles was studied for As(III) uptake. The effect of influent pH on adsorption was not studied here as our previous results revealed that optimum As(III) adsorption on our nanoparticles occurred at pH 7.
Effect of nanoparticles used on the column breakthrough and exhaustion time.
Nanoparticles coated on GB | Breakthrough time (hr) | Exhaustion time (hr) |
---|---|---|
TiO2 |
|
|
Ag-TiO2 |
|
|
Fe-TiO2 |
|
|
Effect of nanoparticles used on column parameters.
Effect of bed height on the column breakthrough and exhaustion time.
Nanoparticles coated on GB | Breakthrough time (hr) | Exhaustion time (hr) | ||||
---|---|---|---|---|---|---|
10 cm | 20 cm | 30 cm | 10 cm | 20 cm | 30 cm | |
TiO2 |
|
|
|
|
|
|
Ag-TiO2 |
|
|
|
|
|
|
Fe-TiO2 |
|
|
|
|
|
|
Effect of bed height on the column parameters.
From the data above it is clear that with an increase in bed height the arsenic removal efficiency of column increases, this is because of greater contact time plus more adsorbent for the adsorption of As(III).
Effect of flow rate on the column breakthrough and exhaustion time.
Nanoparticles coated on |
Breakthrough time (hr) | Exhaustion time (hr) | ||
---|---|---|---|---|
5 mL/min | 10 mL/min | 5 mL/min | 10 mL/min | |
TiO2 |
|
|
|
|
Ag-TiO2 |
|
|
|
|
Fe-TiO2 |
|
|
|
|
Effect of influent flow rate on column parameters.
It is evident from the results that an increase in flow rate causes a decrease in residence time, which in turn lowers the removal efficiency. With a lower flow rate, the removal efficiency increases due to the increase in residence time.
Effect of influent concentration on the column breakthrough and exhaustion time.
Nanoparticles coated on |
Breakthrough time (hr) |
Exhaustion time (hr) | ||||
---|---|---|---|---|---|---|
0.25 ppm | 0.5 ppm | 1 ppm | 0.25 ppm | 0.5 ppm | 1 ppm | |
TiO2 |
|
|
|
|
|
|
Ag-TiO2 |
|
|
|
|
|
|
Fe-TiO2 |
|
|
|
|
|
|
Effect of influent concentration on column parameters.
As expected with increase in influent arsenic concentration, the column breakthrough and column exhaustion times were decreased. This shows that the lower the influent arsenic concentration, the more the efficiency and life of the column.
For a viable sorption process, easy regeneration and reuse of the column media are necessary. The exhausted 20 cm bed volume columns (for all the three nanoparticles coated glass beads) after exhaustion were regenerated with 10% NaOH solution. During desorption, the arsenic recovery profile is shown in Figure
As(III) concentration profile during column regeneration.
After regeneration, the column was rinsed with mild warm deionized water to remove any traces of NaOH and lower the pH to normal, as in basic range the adsorption is less efficient. The column was then dried and was subjected to the next sorption cycle. Figure
Capacities of different nanoparticles for As(III) removal during two successive cycles.
This adsorption and column study of titania-based nanoparticles coated on glass beads is carried out to know the fixed bed column performance of these nanoparticles for As(III) removal. These nanoparticles (both in powder and coated form) were found to be efficient in removing As(III) from water at neutral pH (i.e., pH 7); any increase or decrease in pH resulted in a decrease in their efficiency. Metal doping shifted the absorption band of titania from UV into visible region, thus enhancing both its photocatalytic activity and efficiency to adsorb As(III). Among the nanoparticles used, Fe-doped titania gave the best results both in powder and coated form. Thomas and Yoon-Nelson models applied to evaluate column parameters gave results in clear agreement with the practical results. The column once used could be regenerated in a cost-effective way using 10% NaOH only within 10 bed volumes. Subsequent rinsing of the column with mild warm deionized water for 10–13 bed volumes made the column ready for second sorption cycle. The efficiency of the column was not much affected during the two consecutive cycles.