Dear Jorge,

Attached is one of our recent publications on the use of adsorbents in removing pollutants (batch and column experiments):

I have copied the experimental part for quick view:

2.2.5. Adsorption studies on micelle–clay complex and charcoal

2.2.5.1. Batch adsorption isotherms Equilibrium relationships between adsorbents (micelle–clay complex and activated charcoal) and adsorbate (amoxicillin trihydrate

and cefuroxime axetil) are described by adsorption isotherms which were obtained at adsorbate concentrations of 100, 200, 300, 400 and 500 ppm, prepared in distilled water at pH 8.2 (adjusted by 1M NaOH). The following procedure was applied: 100 mL from each solution was transferred to a 200 mL Erlenmeyer

flask; 0.500 g of the micelle–clay complex or activated charcoal was added to the flask. Then the flask was placed on the shaker for 180 min. Afterwards, each sample was centrifuged for 5 min, and filtered using a 0.45 μm filter. A study on the kinetics of adsorption was conducted by introducing 100 mL of 100 ppm amoxicillin trihydrate and cefuroxime axetil solutions into 250 mL flasks containing 0.500 g of either micelle–clay or charcoal and determining the concentration of amoxicillin trihydrate and cefuroxime axetil. The concentration of amoxicillin trihydrate and

cefuroxime axetil as a function of time was determined spectophotmetrically by recording the absorbance at λmax of 273 and 278 nm, respectively.

2.2.6. Analysis of adsorption isotherms

Equilibrium relationships between adsorbents (micelle–clay complex and charcoal) and adsorbate (i.e. amoxicillin trihydrate or cefuroxime axetil) were described by

the Langmuir adsorption isotherm which is considered the most widely used modelling for equilibrium data and determination of adsorption capacity.[27]

The linear form represented by Equation (1) was employed:

Ce/Qe = 1/(K Qmax) + Ce/Qmax, (1)

in which Ce is the equilibrium concentration of amoxicillin trihydrate or cefuroxime axetil (mgL−1), Qe the equilibrium mass of the adsorbed amoxicillin trihydrate or

cefuroxime axetil per gram of complex or activated carbon (mg g−1), K the Langmuir binding constant (L mg−1) and Qmax the maximum mass of amoxicillin trihydrate or cefuroxime axetil removed per gram of complex (mg g−1).

2.2.7. Column experiments

In the first experiment, a 25/1 (w/w) mixture of quartz sand and ODTMA-clay complex or granular activated carbon (GAC) (20 cm layer) was included in a column (25 × 5 cm). The bottom of the column was covered with a 3 cm layer of quartz sand. The quartz sand was thoroughly washed by distilled water and dried at 105°C for 24 h prior to its use. A wool layer of 2 cm was placed at the bottom

of the column. One thousand millilitres of 100 ppm amoxicillin trihydrate solution were passed through the column at a fixed flow rate of 2 mL min−1. For cefuroxime

axetil, 1000 mL of 50 ppm cefuroxime axetil solution were passed through the column at a fixed flow rate of 2 mL min−1. Eluted fractions of 100 mL (each) were collected, and the concentrations of amoxicillin trihydrate and cefuroxime axetil were determined spectrophotmetrically at λmax of 273 278 nm, respectively. All experiments described were conducted in triplicates. Additional filtration experiments employed the same columns, but with a 50/1 (w/w) mixture of quartz sand and ODTMA-clay complex or GAC (13 g) and the volume passed was several litres, at flow rates of 50 or 60 mL min–1.

2.2.8. Adsorption and convection in a column filter

The adsorption and convection are described by Equation (2) whose numerical solutions were executed by a FORTRAN program.[28] Briefly, a column of length L is filled with material, whose initial molar concentration of adsorbing sites is R0, whose concentration changes later to R(X,t). The beginning and end of the filter are at the coordinates X = 0 and X = L, respectively.We consider that the pollutant concentration at the inlet, C0 is constant, that is, C(X, t) = C0, X ≤ 0, where t denotes time:

dC(X , t)/dt = −v∂C/∂X − C1 ∗ C(X , t) ∗ R(X , t)

+ D1(R0 − R(X , t)) . . ., (2)

where C1 (M−1 min−1) represents rate constant of forward adsorption, D1 (min−1) the rate constant of desorption and v (cm min–1) the flow velocity.

The statistical criteria for the goodness of the fits were the closeness of R2 to unity, and RMSE, the root mean square error, which is given by RMSE = (yi,exp − yi,calc)2/(n − m) 0.5 , (3)

in which yi,exp and yi,calc are experimental and calculated values of per cent removal from water of the pollutant by the filter, n the number of data points and m the number of parameters. In our case, the parameters were R0, Ci and Di.

To view the full paper, please see attached file.

Hoping this will be helpful,

Rafik