Determination of Fenitrothion in Water by Near Infrared Spectroscopy and Chemometric Analysis

Information extraction of UV-NIR spectral data in waste water based on Large Language Model

In recent years, with the rise of various machine learning methods, the Ultraviolet and Near Infrared (UV-NIR) spectral analysis has been impressive in the determination of intricate systems. However, the UV-NIR spectral analysis based on traditional machine learning requires independent training with tedious parameter tuning for different samples or tasks. As a result, training a high-quality model is often complicated and time-consuming. Large language model (LLM) is one of the cutting-edge achievements in deep learning, with the parameter size of the order of billion. LLM can extract abstract information from input and use it effectively. Even without any additional training, using only simple natural language prompts, LLM can accomplish tasks that have never been seen before in completely new domains. We look forward to utilizing this capability in spectral analysis to reduce the time-consuming and operational difficulties. In this study, we used UV-NIR spectral analysis to predict the concentration of Chemical Oxygen Demand (COD) in three different water samples, including a complex wastewater. By extracting the characteristic bands in the spectrum, we input them into LLM for concentration prediction. We compared the COD prediction results of different models on water samples and discussed the effects of different experiments setting on LLM. The results show that even with brief prompts, the prediction of LLM in wastewater achieved the best performance, with R2 and RMSE equal to 0.931 and 10.966, which exceed the best results of traditional models, where R2 and RMSE correspond to 0.920 and 11.854. This result indicates that LLM, with simpler operation and less time-consuming, has ability to approach or even surpass traditional machine learning models in UV-NIR spectral analysis. In conclusion, our study proposed a new method for the UV-NIR spectral analysis based on LLM and preliminary demonstrated the potential of LLM for application.

Detection of pesticide in water using two-dimensional fluorescence correlation spectroscopy and N-way partial least squares

To determine the concentration of carbaryl and chlorothalonil in water, the potential of two-dimensional (2D) fluorescence correlation spectra with N-way partial least squares (N-PLS) was investigated. A total of 40 mixture solutions of carbaryl and chlorothalonil were prepared and 27 of them were used to form a calibration set. The excitation-emission matrix (EEM) fluorescence spectra of all samples were measured. Under the excitation perturbation, 2D fluorescence correlation spectra of all samples were calculated and formed a 2D correlation spectral matrix. And N-PLS models for determination of carbaryl and chlorothalonil in water were built based on 2D correlation fluorescence spectral matrix and traditional EEM spectral matrix, respectively. Finally, the performances of N-PLS models using both methods were compared. For chlorothalonil, the root mean square error of calibration (RMSEC) were 3.43 × 10^-6 g L^-1 and 5.08 × 10^-6 g L^-1, the root mean square errors of prediction (RMSEP) were 5.86 × 10^-6 g L^-1 and 8.99 × 10^-6 g L^-1 for 2D correlation spectra and EEM spectra, respectively. For carbaryl, the RMSEC were 5.24 × 10^-7 g L^-1 and 6.18 × 10^-7, the RMSEP were 9.20 × 10^-7 g L^-1 and 9.63 × 10^-7 g L^-1 for 2D correlation spectra and EEM spectra, respectively. From the results of this study, it may conclude that 2D fluorescence correlation spectra is superior over the traditional EEM fluorescence spectra in terms of predictability and can be used as an alternative method for detection of organic pollutants in environment.

Quantification of pharmaceutical compounds in wastewater samples by near infrared spectroscopy (NIR)

The quantification of pollutants, as pharmaceuticals, in wastewater is an issue of special concern. Usually, typical methods to quantify these products are time and reagent consuming. This paper describes the development and validation of a Fourier transform near-infrared (FT-NIR) spectroscopy methodology for the quantification of pharmaceuticals in wastewaters. For this purpose, 276 samples obtained from an activated sludge wastewater treatment process were analysed in the range of 200 cm^-1 to 14,000 cm^-1, and further treated by chemometric techniques to develop and validate the quantification models. The obtained results were found adequate for the prediction of ibuprofen, sulfamethoxazole, 17β-estradiol and carbamazepine with coefficients of determination (R²) around 0.95 and residual prediction deviation (RPD) values above four, for the overall (training and validation) data points. These results are very promising and confirm that this technology can be seen as an alternative for the quantification of pharmaceuticals in wastewater.