All primary chemicals used were of analytical reagent grade. Potassium chromate (K₂CrO₄) and chromium (III) chloride hexahydrate (CrCl₃・6H₂O) were purchased from Riedel De-Haen. Sodium hydroxide, hydrochloric, sulfuric, and nitric acids were purchased from Paneac. 1,5 di-phenyl carbazide, and acetone were obtained from Sigma Aldrich_. The wool was washed with water and detergent for two days. The adsorbent used in the study was sheep wool, freshly trimmed from sheep in the Sharjah animal market. It was carefully cleaned, riffled, separated from impurities then air-dried. It was finally sized into 2-cm long fibers and placed in a sealed container labelled “original wool”. A portion of this wool was freeze-dried in liquid nitrogen for 10 min and then milled. Powder below 0.60 mm (mesh 30) was labelled “powdered wool”.

Total chromium was determined using Varian Liberty axial sequential inductively coupled plasma-atomic emission spectrometer (ICP-AES, Varian, Australia). Cr (VI) concentrations were determined spectrophotometrically using Cary a 50 spectrophotometer (Varian, Australia). IR spectra were recorded using Spectrum One FTIR spectrometer (Perkin Elmer, USA). pH was measured by Thermo-Orion 210A+ pH meter (USA) using a combined glass electrode calibrated with buffers of pH 4.0, 7.0, and 10.0. Solutions were shaken using Edmund Buhler KS-15/TH-15 shaker (Germany) and their temperature was maintained at 25.0˚ ± 0.1˚. Scanning electron microscopy (SEM) was performed on a Tescan VEGA III LMU (Czech Republic). Electron dispersive x-ray spectroscopy (EDS) was recorded on Oxford Instruments INCA XMAX (UK).

Stock solutions containing 500 ppm Cr(VI) were prepared by dissolving potassium dichromate in deionized water. Standard solutions of 10, 20, 50, 100, 150 & 200 ppm were prepared by dilution. Stock ligand solutions were prepared by dissolving 1,5 diphenyl carbazide in acetone.

Batch adsorption studies were carried out in flasks containing 100 ml of test solutions at the desired initial chromium concentrations. pH was adjusted using either 0.10 M HCl or 0.10 M NaOH. Wool was then placed in the flasks and the contents shaken for 3-h at 25.0˚C and 150 rpm. At known equilibration times, aliquots were withdrawn and analyzed for total chromium and Cr(VI).

Spectrophotometry was used to determine Cr(VI) concentrations using 1,5 di-phenyl carbazide. A 0.10 ml aliquot was collected and mixed with 0.20 ml of this ligand and 2 drops of 6.0 M sulfuric acid and the volume adjusted to 10 ml. The Cr(VI)-ligand complex was allowed to equilibrate for 10 min before measuring its absorbance at 540 nm [22] . ICP was used to determine total chromium in samples and standards using the method provided by the manufacturer.

Figure 1 shows the effect of contact time in the short term mode (0 - 3 h). The data reveal no dicsernible effect of particle size (original vs. powdered wool) on Cr(VI) removal efficiency. Furthermore, there is no apparent distinction between Cr(VI) and total chromium concentrations, indicating that Cr(VI) is virtually the only species present The maximum removal efficiency is ca 88%, in conformity with our previous findings [5] [6] .

Figure 2 shows the effect of contact time on Cr(VI) removal in the long term mode (0 - 165 h). For both wool forms Cr(VI) was almost totally removed from solution after 100 h. However, the total chromium concentration at pH 1.5 decreased by 90% up to 120 h equilibration before rising at higher times. The increase in Cr(III) concentration with time is indicative of removal of Cr(VI) by wool, followed by reduction to Cr(III) and subsequent release of Cr(III) into solution. The sequence is further supported by FTIR data (see below). Control experiments in absence of wool indicate that Cr(VI) is relatively stable in both time modes.

The effect of pH on chromium removal efficiency is shown in Table 1 for the long term mode for the pH range 1.5 - 6.0, from which it can be deduced that removal is favored by high acidity. This can be seen in Figure 2 which shows results at pH 1.5 and pH 2.0.

On the other hand, short term mode studies (2-h contact time) showed no difference in removal at pH 1.5 and 2.0. Table 1 also shows that the percent removal of Cr(VI) is much larger than that of total chromium, indicating speciation between Cr(VI) and Cr(III) [23] [24] . The acid catalyzed reduction of Cr(VI) to Cr(III) on wool may be represented by

This mechanism is similar to that proposed in previous reports [20] [24] [25] for removal of Cr(VI) by other natural adsorbents. Table 1 shows the optimum pH for the removal of Cr(VI) by original wool is 1.5.

Figure 3 summarizes the effect of wool dosage on removal of Cr(VI) and total chromium at pH = 2.0, for a 5-day contact time. The data show that the optimum dosage of wool is 8.0 g/L. Figure 3 also reveals that the extent of reduction of Cr(VI) in the long term mode is enhanced by higher dosages. Whereas Cr(VI) removal at dosages higher than 8.0 g/L remains constant, total chromium removal is hindered by high dosages. This result may be explained by positing that high dosages serve to increase the rate of reduction of Cr(VI) to Cr(III), followed by release of Cr(III) into solution.

Adsorption data were fitted to the Langmuir (Equation (2)) and Freundlich (Equation (3)) isotherms [3] .

where C_e is the equilibrium concentration of total chromium (mg/L), q_e is the amount adsorbed at equilibrium, in mg/g adsorbent. Q (mg/g) and b (L/mg) are the Langmuir constants, representing adsorption capacity and energy, respectively. K_f and n are the Freundlich constants.

Figure 4 gives the variation of q with the equilibrium concentration of Cr(VI) in the long term contact time mode. The data reveal that q follows the same pattern at pH 1.5 and 2.0, the asymptotic trend indicating a monolayer. The Langmuir isotherm gives a good fit for the data (Figure 5). However, the dependence of the term C_e/q_e on equilibrium concentration in the long term mode does not obey this isotherm, indicating that calculations based on total chromium in solution cannot be attributed to Cr(VI) alone. Thus speciation of chromium takes place, with release of Cr(III) into solution following reduction of adsorbed Cr(VI). This finding demonstrates the efficacy of species resolved techniques in the study of adsorption isotherms for this type of complex system. By contrast, an earlier study using short term equilibration (2-h contact time) revealed that adsorption data for total chromium obeyed the Langmuir isotherm [3] . For this mode both Cr(VI) and total chromium fit this isotherm (Figure 6). These results may be explained by a two-step mechanism. The first is adsorption on wool accompanied by rapid equilibrium, and the second is a slow reduction of Cr(VI) to Cr(III), followed by release of Cr(III) into solution. Results for the short term mode reflected primarily the fast first step.

Figure 5 and Figure 6 yield the Langmuir parameters for short and long term modes. Table 2 summarizes the results for total chromium and Cr(VI) and shows that Q is larger for Cr(VI) in the long term mode and is also larger than that for total chromium in both modes. This indicates that in long term mode, more sites are available for Cr(VI) than in short term mode. This finding may be explained by postulating that some sites occupied by Cr(VI) undergo simultaneous reduction and release of Cr(III) to the solution.

Table 3 summarizes the fit of the data to the Freundlich isotherm Inspection of this table reveals that the correlation coefficient for both Cr(VI) and total chromium in the short term mode is smaller than that for Langmuir isotherm (Table 2). By contrast, for the long term mode, total chromium data obey the Freundlich, but not the Langmuir, isotherm. This indicates the non-ideality of total chromium adsorption which in turn may be attributed to the proposed mechanism above. Upon comparing the maximum adsorption capacity (Q) obtained under long term equilibration with values previously reported for low cost adsorbents [21] [22] [26] -[30] , the value obtained in this work is among the highest 5%.

Previously reported data on the elution profile of Cr(VI) from columns packed with wool [19] had been based solely on ICP, which measures total chromium. The authors had assumed that all Cr(VI) is retained after passing through the column, and had not considered the possibility of reduction to Cr(III). Reduction may indeed not

occur when the contact time between wool and the eluent is very short. Under such conditions, all forms of chromium are essentially removed. However, when solutions containing Cr(VI) were left in contact with wool in the column for longer periods, the first fraction contained higher concentrations of total chromium, and this can now be attributed to the presence of Cr(III). To verify this conclusion, 100 ppm Cr(VI) solution was applied to a column packed with original wool and eluted at a constant rate of 2.0 ml/min. 100 ml fractions of eluents, representing short term equilibration were collected for four successive overnight periods, with the elution stopped at the end of each day, thereby allowing long term equilibration of Cr(VI) on wool. The data for total chromium and Cr(VI) are presented in Figure 7. Virtually, the only form of chromium eluted overnight is Cr(III). This corroborates the finding that upon prolonged exposure on wool, Cr(VI) undergoes reduction to Cr(III).

Figure 8 displays FTIR spectra of wool before and after adsorption of Cr(VI). The very weak peak at 939 cm⁻¹ in free wool becomes much more pronounced in wool loaded with chromium for long term contact, due to the oxidation of wool by Cr(VI). This peak can be attributed to S=O bonds that form when cystine in wool is oxidized. A similar assignment has been reported for wool oxidized by UV irradiation [31] . This finding supports the proposed mechanism for removal of Cr(VI) by wool from aqueous solution (Equation (1)).

Scanning electron microscopy (SEM) was employed to study the surface of wool before and after loading with chromium. In order to obtain the elemental analysis of the surface, EDS spectra were recorded for both samples (Figure 9). Inspection of this figure reveals that chromium is retained on wool at the end of the long term contact time, with the appearance of a chloride peak from HCl.