A HACH Spectrophotometer (Made in USA) with 1 cm quartz cell was used for the absorbance measurements. Residues left after extraction of water soluble Cr (VI) (extraction 1) and extraction of sparingly soluble Cr (VI) (extraction 2) were dried and ground. Infrared spectra of these powders were then recorded from KBr pellets in the range 4000-200 cm⁻¹ on a SHIMADZU FTIR 8400S spectrophotometer. EDX was performed with a JEOL Model JED – 2300. The cement sample (S1) and its residue left after extraction of insoluble Cr (VI) (extraction 3) was dried and ground. The fine powder, thus obtained, was used to take EDX spectrum.

All the chemicals used were of analytical / GR grade. Variamine blue dye solution (0.05%) was prepared by dissolving 0.05 g in 25 mL absolute alcohol and then the volume was made 100 mL using distilled water. A solution of potassium iodide (2.0%) was prepared by dissolving 2.0 g potassium iodide in water and diluting it up to 100 mL. A solution of sodium acetate (2M) was prepared by dissolving 16.407 g sodium acetate in distilled water and the volume made up to 100 mL in a volumetric flask. Sulfuric acid (1M) was prepared by diluting 6.95 mL of stock H₂SO₄ to the mark in a 250 mL volumetric flask with distilled water. Stock solution of Cr (VI) was prepared by dissolving 0.2829 g, K₂Cr₂O₇ in distilled water to make the volume 1L. Standard solutions of 0.05, 0.1, 0.2, 0.04, 0.08, 1 ppm were made by dilution of stock solution.

Cement samples used are fly ash based Portland Pozzolan Cement of different brands and the %age of pozzolanic material is not less than 10% and not more than 25% (Table 1). The study of chemical composition of PPC samples (Table 2) has been carried out at Shankar Laboratory, New Delhi. Due to the unavailability of hexavalent chromium certified reference cement, SRM 2701 Hexavalent Chromium in contaminated Soil from NIST (National Institute of Standards and Technology, USA) was used for revalidating the method accuracy. The certified concentration of Cr (VI) in SRM 2701 is 551.12 ± 34.5 mg/Kg.

The Cr (VI) from cement samples was extracted using five day extraction process (12, 13). In this method, 100 mL distilled water was added to 1.0 g cement sample. The contents were mixed thoroughly with a glass rod. This mixing was repeated for five days to ensure complete leaching of hexavalent chromium. After five days, filtration was done using Whatman 42 filter paper. Two washings were required. Insoluble part was kept for estimation of sparingly soluble and insoluble chromium and filtrate was used for detection of water soluble hexavalent chromium.

Sparingly soluble Cr (VI) is present in cement in the form of strontium chromate (14). For its detection, residue left after extraction 1 was transferred to a 250 mL glass beaker. To this was added, 40 mL of ammonium sulfate buffer (0.05 M ammonium sulfate and 0.05 M ammonium hydroxide) and contents were stirred for 30 min. Filtrate obtained after stirring was used to estimate sparingly soluble chromium and residue was kept for estimation of insoluble chromium (14).

To the residue obtained after extraction 2, was added 40 mL of carbonate buffer (3% (w/v) sodium carbonate and 2% (w/v) sodium hydroxide) in 250 mL glass beaker. The reaction mixture was stirred for 30 minutes and filtered (14). Filtrate was used to estimate the concentration of insoluble chromium.

To the filtrate of extraction 1, 1 mL of 2% potassium iodide was added. Addition of sulphuric acid (1.0 mol/L, 1 mL, the oxidation of iodide to iodine was effective in the pH range 1.0 to 1.5) to the reaction mixture gave appearance of yellow colour indicating the liberation of iodine. To this was added variamine blue (0.05%, 0.5 mL) and 2 mL of 2 mol/L sodium acetate solution (Buffer solution). The resulting solution was kept for 10–15 minutes to ensure the completion of reaction (appearance of violet colour). The absorbance at 556 nm, of the coloured solutions had a linear relationship to the Cr (VI) concentration. The similar procedure was used for filtrates obtained after extraction 2 and 3. All the solutions were also analyzed for chromium concentration by standard DPC method (15). The concentration of Cr (VI) can be calculated using the following formula: C_{(Cr (VI))} = C V₁/M V₂.

Where C_{(Cr (VI))} is the concentration of hexavalent chromium (ppm), C is the concentration of Cr (VI) in µg/ml, V₁ is the volume of water in which the original sample is suspended, V₂ is the volume of filtrate which is transferred to the volumetric flask, cm³. M is the mass of the cement sample.

The calibration graph is linear and can be explained by equation: y = 0.062x – 0.001, where y is integrated absorbance and x is analyte concentration (Figure 2). The regression value obtained for the calibration plot is 0.998. The detection limit (D_L= 3.3σ/S) and quantification limit (Q_L= 10 σ/S) of Cr (VI) were found to be 0.05 and 0.16 µg/ml, where σ is the standard deviation of blank samples (n=5) and S is slope of calibration curve. The experimental data to get calibration curve is given in supplementary material.

“SRM 2701 Hexavalent Cr in contaminated Soil” from NIST (National Institute of Standards and Technology, USA) was used for revalidating the results obtained by the proposed method (Table 3).

For complete leaching of water soluble hexavalent chromium, five days extraction procedure was followed. Long stirring in water ensured the extraction of some hexavalent chromium that was present in solid phases of cement. Sparingly soluble Cr (VI) was extracted by treatment of residue of extraction 1 with ammonium sulfate buffer (0.05 M (NH₄)₂SO₄–0.05 M NH₄OH). However, the insoluble chromium was extracted using carbonate buffer (3% (w/v) sodium carbonate and 2% (w/v) sodium hydroxide).

In the IR spectrum (16), bands for CO₃^2- ion appeared at 1425, 1497, 875 and 732 cm⁻¹ and for chromate ion in the range 850-950 cm⁻¹ (specially for ZnCrO₄ and CaCrO₄) (8, 17, 18). IR spectrum of cement sample S1 show band at 877 and 1430 cm⁻¹ (Figure 3). The intensity of these bands decreased after the extraction of water soluble and spirangly soluble Cr (VI) (residue of extraction 1 and 2 respectively). Decrease in intensity of band at 877 cm⁻¹ can either be due to leaching of hexavalent chromium in extraction process or reduction in carbonate concentration (due to formation of carboxyaluminates (19, 20). or replacement of carbonate by sulphate ion in Aft (alumina ferric oxide tri-sulfate) and AFm (alumina ferric oxide mono-sulfate) phases) (21, 22).

Since vibrational bands for carbonate and chromate band appeared in the range, 870–880 cm⁻¹, it is difficult to get a distinct information regarding decrease in concentration of chromate ion. Nevertheless, it can be used as supporting data to other techniques like EDX and UV-Vis spectrophotometer.

Extraction of hexavalent chromium was also confirmed by EDX spectra of sample S1 and the residues left after extraction 3. Generally chromium gave a peak in range of 5.0 to 6.0 KeV in EDX spectrum (23–26). In sample S1, this peak appeared at 5.2 KeV in its EDX spectrum (Figure 4), however this appeared as very weak peak after extraction 3. (Figure 5)

Variamine Blue method was applied to filtrates of extraction 1, 2 and 3 of six different samples of Portland pozzolan cement for estimation of water soluble, sparingly soluble and insoluble hexavalent chromium. To check the accuracy of proposed method, all the six cement samples were also tested for same using standard DPC methods.

Tables 4–6 gave a comparison of proposed method with standard method in terms of parametric tests (T and F test), percentage of relative standard deviation (RSD) and percentage of recovery. The percentage recoveries in six cement samples (from S1 to S6) were spiked with known concentration of hexavalent chromium. A null hypothesis indicating that both Cr (VI) determination methods adopted in this research relied on each other was evaluated by comparing both the means with parametric test (T and F test). These tests for the comparison of means were calculated. The calculated values were compared with the tabulated value at a proper degree of freedom (27) and results indicate reliability of these methods from 2 to 12 ppm concentration (28). At a 95% confidence level, the calculated T- and F-values do not exceed the theoretical values (Table 3–6) indicating no significant difference between the proposed and the reference method. Consequently, the developed method is as accurate and precise as that of reference method (Figure 6).

No insoluble Cr (VI) was detected by both the methods for samples, S1, S3 and S4. Results obtained by proposed method were in good agreement with standard method except for sample S3 and S4, where concentration of water soluble Cr (VI) by proposed method was found to be less than that of obtained by standard method. Concentration of water soluble and sparingly soluble Cr (VI) was found to be maximum in sample S4 and minimum in S5 by both the methods. It was observed that in Portland pozzolan cement samples, 54–55% of the total hexavalent chromium was water soluble. The percentage of sparingly soluble and insoluble Cr (VI) was found to be 30% and 10% respectively. Although, the industries and researchers are more concerned about water soluble hexavalent chromium, sparingly and insoluble Cr (VI) are also toxic (29). Thus detection of all type of Cr (VI) in cement sample is necessary for the safety of workers.