1. Introduction

Nano particles, colloid materials with average size range of 1-100nm, are very useful for fundamental and diversified applications [1]. They are used in several industrial applications including optical sensor devices, thermal conducting gel, body repairing, cancer therapy, high performance batteries, chromatography columns and magnetic fluids to mention but a few. In pure form they cannot be used because of their tendency to agglomerate. To avoid these problems, nano particles are embedded on a support material to form nano composites instead [2]. Preparation methods for nano composite particles depend on their intended use. Two most widely used methods for preparing nanoparticle/composite are chemical and physical procedures [3]. Chemical synthesis involves the use of chemicals reactions to produce nano units while physical methods involve breaking down of large material by mechanical grinding until the particles are in the nano range [4]. Furthermore, coal and Clay minerals are widely used in several industries and fields such as in polymers nano-composites adsorbents fields, in heavy metal ion adsorption, ceramics, and paper fillings [5-7] as well as in nano composites designed for specific applications. These include anion adsorbents [8,9], cation adsorbents [10,11,8] and in some cases adsorbents to aid purification of water by removing multiple contaminants simultaneously [12].

In this paper we present the preparation and characterisation of coal/clay nano composites. These composites are then used to investigate their possible use to purify water. This is done by studying the adsorption kinetics and thermodynamics of removal from water of methyl orange, an organic pollutant.

1.1. Adsorption Kinetics

1.1.1. Lagergren Pseudo 1^st Order Kinetics’Model

Lagergren pseudo first-order rate equation, the earliest known model describing the adsorption rate based on the adsorption capacity [9], is given below (equation 1):

(1)

Where is the pseudo first order rate constant in g/mg* min, q_e is the amount adsorbed at equilibrium, q_t is the amount adsorbed at any time t, in minutes. If the data obtained fits the model, a plot of against t should give a straight line and can be determined from the slope and from the intercept. The rate constant the theoretical and the equilibrium amount adsorbed q_e can be derived from this model plot.

1.1.2. The Lagergren Pseudo Second Order Model is Given by Equation 2

(2)

once again a plot of t/q_t against t should be linear and the second order rate constant k₂ can be determined from the intercept and q_e from the slope)

1.2. Adsorption Thermodynamics

Adsorption thermodynamics’ data is analysed using Langmuir, Freundlich and Temkin models, in this work the data did not fit the Langmuir isotherm hence it was not used.

The Freundlich adsorption model is given by equation 3 and 4

(3)

which linearizes to

(4)

a linear plot of lnq_e against ln C_e will give the constant K_f from the intercept and the constant n from the slope.

Temkin adsorption model is given by equation 5

(5)

and from a linear plot of the quantity adsorbed q_e against ln C_e the constants A (Temkin equilibrium binding constant) and B (a constant related to heat of adsorption in J/mol) can be determined from the slope and the intercept. In this case B=RT/b_T where b_T is Temkin isotherm constant.

A consideration of adsorption equilibrium results in equation 6 below;

(6)

Where K_o is the thermodynamic equilibrium constant is the surface concentration of dye on adsorbent (mol g^-1) is the concentration of dye in the equilibrium solution (mol L^-1), is the activity coefficient of the adsorbed dye and is the activity coefficient of dye in solution

For very dilute solutions hence the values of K_o is obtained from a plot of versus q_e. An extrapolation of q_e to zero [13] gives an intercept equal to ln K_o.

Thus the thermodynamics parameters can be calculated from equation 7 and 8 given below.

(7)

(8)

2. Experimental

2.1. Materials

2.1.1. Adsorbent

The coal-clay nano-composites were prepared using a method adopted from the literature [14,15] in which nano- composites with activated interfaces were prepared by mechanical grinding of magnesium hydrides used for storing hydrogen.

Mixtures of 1:1, 1:4 and 4:1 mass ratio of the coal (ESEM fig 1a) to clay (ESEM fig 1b [16]) were made and used to prepare the composites. The resulting powder was then soaked in deionised water at room temperature under ultrasonic stirring, to form a homogeneous mixture. The resulting mixture was subjected to mechanical grinding until the slurry mixture was formed in water [17]. The resulting mixture was centrifuged at 2000 rpm. Water was decanted and the top fine particles were dried in an oven. The resulting product was analysed with an Environmental Scanning Electron Microscope (ESEM figure 1c). This was repeated until particle reduction to nano level was achieved. These were used to investigate adsorption experiments.

Figure 1a. ESEM microgra of Morupule coal

Figure 1b. ESEM microgram of Lobatse clay [16]

Figure 1c. SEM microgram of a 1:1 clay: coal composites

2.1.2. Adsorbate Methyl Orange (MO)

Methyl orange (Anionic, water soluble azo dye) powder (C₁₄H₁₄N₃NaO₃S, figure 2a) supplied by SAARCHEM, South Africa was used without any further purification. A solution of the dye was made by dissolving a known weight into double distilled de-ionised water. Other required dye concentrations were prepared by serial dilution. Several of these concentrations were used to obtain a calibration curve of absorbance against concentration at a predetermined wavelength of maximum absorption λ=463 nm using a UV-Visible spectrophotometer spectronic 20. Concentrations of unknown samples were then determined from the calibration curve.

Figure 2. Methyl orange structure

2.2. Procedure

2.2.1. Studies of Adsorption Kinetics

0.05g of the nanocomposite was added to 20 ml of MO solution in a screw capped bottle contained in a water bath at constant temperature, and a stop clock was simultaneously started. Samples of the supernatant liquid were drawn off after predetermined time intervals, centrifuged at 2000 revolutions per minute for a minute or so to assist the separation of the nano particles from the supernatant, filtered using an appropriate filter paper and then their absorbance were taken using a UV/VIS spectrophotometer at a wavelength of 463 nm. The amount adsorbed was obtained by difference between the initial concentrations and the concentrations at the sampling time and plotted against time. The experiment was repeated for various MO concentrations, reaction temperature and pH.

2.2.2. Studies of Adsorption Thermodynamics

In this section, a batch technique was used. The clay-coal composite (0.05 g) was added to each of several 250 mL screw capped bottles containing 200ml of methyl orange of different initial concentrations. The bottles were kept in a thermostatic water bath shaker at a speed of 220 rpm for 4 hours at constant temperature. After the samples had reached equilibrium, the Solutions were centrifuged, filtered using 0.45 μm filter paper and the equilibrium concentrations of the filtrates were determined spectrophotometrically as before. The amount adsorbed at equilibrium, q_e (mol/g), was calculated by using equation 9 below:

(9)

Where C_o is the initial concentration of the dye and C_t is the concentration of the dye after time t (minutes) (mol/L), V (L) is the volume of methyl orange solution and m(g) is the weight of coal- clay composite. The equilibrium adsorption q_e values were used to plot an adsorption isotherm.

3. Results and Discussion

3.1. Analysis of Results from Adsorption Kinetics Studies

Figure 3a presents the adsorption kinetics of the dye onto the composites at 300K. For the three composites, the process reaches equilibrium within 30minutes. The composites show different adsorption capacities and different rates of adsorption at the various ratios. The nano composite with the highest coal content removes more of the dye compared to the other two. The nano composite with highest mass of coal and that with equivalent masses of coal and clay show the highest adsorption capacity and rate. The nano composite with the lowest coal content shows lowest adsorption capacity.

Figure 3a. Adsorption of methyl orange dye 3.14 × 10-5 M on to the composites at pH 7.18 and 300K

Figure 3b compares all the adsorbents studied in this work with commercial activated carbon. It is clear that the commercial adsorbent is still outperforming our prepared samples. However, the nano composites and acid activated clay are closest to the commercial activated carbon samples.

Figure 3b. Studies of methyl orange 3.14 x 10-5 M adsorption onto various adsorbents at 300K

3.1.1. Testing data on Kinetic Models

The chi (χ²) test was used to test the closeness of the calculated/theoretical data to the experimental data, by using equation 10.

(10)

where

is the experimentally determined equilibrium amount adsorbed

is the theoretical equilibrium amount adsorbed determined from the kinetics’ model equations

From this equation a small value of shows the model test fits the experimental data well.

Adsorption of methyl orange onto Morupule coal:Lobatse clay nano composite was investigated by fitting the raw data to the lagergren pseudo first and pseudo second order kinetic models and the resulting plots are given below (see fig 4, table 1).

Figure 4. Lagergren pseudo first order model plot for the adsorption of methyl orange onto the 1:1 composite at T= (298 and 300) K, 5.49 × 10-5 M, pH 7.18 and 10 g/L sorbent dosage

Table 1. Lagergren pseudo first order parameters

The sorption process was investigated by fitting the raw data onto the Lagergren pseudo first and second order kinetics models. The sorption process fitted the Lagergren pseudo second order kinetics model (fig (5) slightly better than the pseudo first order (see fig 4) confirming the continuation of simultaneous occurrence of two mechanisms or one mechanism in two different environment thought to be inner and outer surfaces, even on the nano composite. One would expect miniaturisation to eliminate delayed adsorption onto internal area originating from swelling. The results should hence fit the Lagergren pseudo first order model better for the nano-composite. This was not very clear as the results fitted both models. However, the fitting onto the pseudo second order was slightly better. This could be because of the clustering of the nano particles which creates temporary slit like pores out of spaces between particles. The sorption process was faster on the nano-composite than on the other two adsorbents (coal and clay fig 3b) as would be expected due to miniaturisation to eliminate or reduce internal areas.

Figure 5. Lagergren pseudo second order kinetics model for the adsorption of methyl orange onto the 1:1 coal: clay composite at T= (298 and 305) K, 5.49 × 10-5 M, pH 7.18 and 10 g/L sorbent dosage

3.2. Thermodynamics’ Data Analysis

Figure 6 shows a decrease in the adsorption capacity of the nano composites with an increase in pH indicating that adsorption is more favoured in acid conditions than in basic conditions.

Figure 6. Effect of pH change on the equilibrium adsorption of methyl orange onto the 1:1 composite at 3.14 × 10-5 M, 298 K and 10 g/L sorbent dosage

As pH is increased to more basic conditions, the double bond conjugation is lost and a proton is lost, and the MO molecule rearranges to form a negatively charged species (figure 7a). The positive charges have been neutralized. Due to excess anions in the solution, there will be repulsion of the dye by the negatively charged surfaces of the adsorbents, also an increase in pH leads to the modification of the pi (Π) system delocalization pattern.

In the literature, the uptake of methyl orange on several adsorbents was similarly found to be favoured at lower pH (acidic medium) [18]. Furthermore, at lower pH, the surface of the adsorbent was found to be activated by the acid in solution and the dye became protonated (see fig 7b, c &d) thus enhancing the adsorption process [9].

3.2.1. Effect of pH on the Structure of the Dye

Figure 7. Alkaline form of Methyl orange (a) and its mono-protonated zwitter ionic structures under acidic conditions(b, c and d)

3.2.2. Temperature Effect

Table 2

Studies of the thermodynamics of adsorption of methyl orange onto nano-composites by the batch technique were conducted and the data obtained was analysed using various thermodynamic models outlined earlier. The effects of solution pH and solution temperature were investigated. An decrease in temperature of the dye solution increased the capacity of the adsorbents (see fig 8) indicating a physisorption process. This is further confirmed by the low entropies and heats of adsorption obtained (see table 3).

Table 3. Thermodynamic properties for the adsorption of MO on 1:1 nano composite

Figure 8. Effect of change in temperature on the equilibrium adsorption of methyl orange onto the 1:1 composite at 3.14 × 10-5 M, 10 g/L sorbent dosage and pH 7

The process onto the 1:1 composite is favoured by lower pH (see fig 6). This is probably due to the resultant net positive charge on the dye molecule at lower pH (see fig 7) which weakly interacts with the negatively charged adsorbent surface.

The adsorption capacity of the clay, the 1:1 and 4:1 composite adsorbents compared reasonably well with that of activated carbon in spite of the much larger surface area of over 500 m²g^-1 for the commercial activated carbon (see fig 3b). The trend of the capacity was found to be: Activated carbon > 4:1 coal: clay composite ~ 1:1 coal: clay composite > 1:4 coal: clay composite. The changes in adsorption capacity in the nano composites appear to indicate the presence of some form of interaction in solid phase between the clay and coal nano-particles.

The Freundlich and Temkin isotherms models were found to describe the sorbent-sorbate interaction for this study best (see figs 9&10). The heats of adsorption were small and inconclusive in this case as for the coal and clay [17]. The data from adsorption onto the surface of the nano-composite fitted the Freundlich model better confirming that it was heterogeneous just like the coal and the clay. This correlated well with the scanning electron microscope micrograms; which showed varying sizes of nano agglomerates and irregular surfaces for coal and clay particles [19] (see fig 1a, 1b and 1c).

Figure 9. Freundlich isotherm for the adsorption of methyl orange onto 1:1 nano composite at 298 K, pH 7.18 and 10 g/L sorbent dosage

Figure 10. Temkin isotherm for the adsorption of methyl orange onto the 1:1 coal-clay composite at 298 K, pH 7.18 and 10 g/L sorbent dosage

Figure 11. Plot for the determination of Ko at various temperatures for the adsorption of methyl orange onto the 1:1 composite

Figure 12. Vant’Hoff plot for the adsorption of methyl orange onto the 1:1 nano composite

4. Conclusions

The possible use of the composites as water purifiers investigated by studying their interaction thermodynamics and kinetics at various temperatures and solution pH found that adsorption capacity, q_e, increases with increasing initial concentration (see fig 9), decreases with increasing pH (see fig 6) and decreases with increasing temperature (see fig 8). The enthalpies and entropies of adsorption were found to be -16.421 kJmol^-1 and 0.060 kJmol^-1 respectively (see table 3) for the 1:1 composite indicating that physisorption is still the mechanism in operation.

It is evident that miniaturisation to nano level greatly improves their capacity and rate for adsorbing methyl orange compared to that of the individual clay or coal samples. The fitting of the kinetics data on Lagergren pseudo second order was better than that on the pseudo first order. However the fitting on pseudo first order greatly improved from the cases of the individual coal and clay adsorbents observed in earlier work [21]. This indicates that as a result of miniaturisation, the internal areas originally accessed after swelling caused by initial adsorption which often lead to a second adsorption surge was now easily available at the start. There was therefore no adsorption initiated expansion of the material which was observed for both the clay and the coal individually [21,22]. However, its contribution to the overall available area is less that the contribution from steam activation since the commercial steam activated carbon still performed better indicating that steam activation is more effective in creating a super adsorbent than miniaturisation. This is therefore a cheap viable method of removing methyl orange comparable to many others [23-26].