Identification of heavy metal by testing microalgae using confocal Raman microspectroscopy technology

Area of Science:

• Environmental Science
• Biotechnology
• Analytical Chemistry

Background:

• Copper is a common environmental pollutant with potential toxicity to aquatic organisms.
• Microalgae like C. pyrenoidosa are sensitive indicators of water quality and can accumulate heavy metals.
• Accurate and rapid detection methods are crucial for monitoring copper contamination in aquatic ecosystems.

Purpose of the Study:

• To investigate the physiological responses of C. pyrenoidosa to varying copper concentrations.
• To develop and validate a spectral analysis model for identifying copper stress levels in microalgae.
• To assess the efficacy of machine learning algorithms in detecting copper contamination.

Main Methods:

• C. pyrenoidosa cultures were exposed to five different copper concentrations (0-4 mg/l) for five days.
• Biomass, chlorophyll, and carotenoid content were measured.
• Raman spectroscopy (mapping and single-point) was employed to acquire spectral data, followed by chemometric analysis including Principal Component-Linear Discriminant Analysis (PC-LDA) and Back Propagation-Artificial Neural Network (BP-ANN) modeling.

Main Results:

• Copper exposure induced measurable changes in microalgal biomass and pigment content.
• Spectral data from Raman spectroscopy showed distinct patterns corresponding to different copper concentrations.
• The BP-ANN model demonstrated superior performance in identifying copper concentrations, achieving a prediction accuracy of 92% by day 4.

Conclusions:

• Raman spectroscopy combined with BP-ANN offers a promising, non-destructive method for detecting copper stress in C. pyrenoidosa.
• This approach can serve as an effective tool for early warning systems in aquatic environmental monitoring.
• The study highlights the potential of spectral analysis and machine learning for rapid ecotoxicological assessments.