Detection of heavy metal copper in vetiver grass roots based on Raman spectroscopy and resin adsorption technology

Abstract: Heavy metals are harmful to the vast majority of organisms including human. However, heavy metals can be gradually accumulated in plants and animals by the food chains. So the heavy metal pollution in the environment will threaten human health if we do not control the pollution. Heavy mental ions in plants always have chelation with the organic molecular groups in resin, and the complex has the Raman spectroscopy. Therefore heavy mental ions in plants can be indirectly detected by using Raman technique basing on the chelation. Vetiver grass can grow up in the soil with heavy metal pollution. It has the strong tolerance of heavy metals. Because of that, vetiver grass becomes one of the ideal plants for the conservation and phytoremediation of heavy metal pollution in soil and water. After preprocessing resin by HCl and NaOH, we compared copper adsorption rate for different types of resins. Based on the results of comparison, we chose ion exchange resin D113 to absorb copper. We then compared different oscillation time, solution pH, solution temperature for their impact on copper adsorption so that the best adsorption conditions could be determined. It showed that changing the temperature of solution had a little impact on resin adsorption rate. Therefore, experiments for copper was conducted at the conditions: at room temperature, pH value from 5 to 7, and the oscillation time 80 min. The adsorption rate can reach 99.54% in these conditions. An application of confocal microprobe Raman spectroscopy fast detecting heavy metal copper in vetiver grass roots was proposed, and partial least squares (PLS) regression combined with different data preprocessing methods (Savizky-golay smoothing, Baseline correction, first derivative, second derivative, Detrended fluctuation analysis) was used to develop quantitative models of heavy metal copper in vetiver grass roots. Calibration models were evaluated by an independent predictor of adaptive sample set. With the first derivative preprocessing, the best prediction model of heavy metal copper in vetiver grass roots was achieved with the external validation correlation coefficient of 0.78 and root mean square error of prediction (RMSEP) of 23.46% separately. The study showed that the fast detection of heavy metal copper in vetiver grass roots using Raman spectroscopy technique and the D113 resin adsorption technique was feasible. The calibration and validation statistics obtained in this study showed the potential of Raman spectroscopy to predict heavy metal in the root of vetiver grass. Although the accuracy of model with this method was lower than the routine analysis, Raman spectroscopy could be used as a fast and simple tool to diagnose the content of heavy metal in vetiver grass approximately. Overall, Raman spectroscopy can be used as a reference method for diagnosing the content of heavy metal in vetiver grass, but the accuracy needs to be improved.