Rapid industrialization in China has gradually led to long-term accumulated soil environmental problems. In addition, pollution caused by industrial and agricultural production processes constantly deteriorates soil environmental quality. Polycyclic aromatic hydrocarbons (PAHs) are a class of persistent organic pollutants that are mainly derived from man-made pollution and migrate globally with biogeochemical cycles. Residual PAHs in soil have a profound impact on environmental quality and human health. Therefore, it is of great practical significance to monitor organic pollutants in soil and timely grasp the pollution situation of the regional topsoil. At present, PAHs are detected using field sampling and laboratory instrument analysis. Trace analysis methods based on chromatographic separation have the advantages of low detection limit and high accuracy, but they usually require complex sample pretreatment, complicated operation, and long detection cycle. Thus, studies have developed a light-emitting diode (LED)-induced fluorescence methods using newly-invented luminescent materials and advanced production technologies. An LED can make up for the deficiencies of traditional excitation light sources, facilitating the production of small instruments. In this paper, the feasibility of using an LED-induced fluorescence spectroscopy technology for the rapid determination of PAHs in soil is discussed based on the fluorescence detection system of UV LED array excitation and a theoretical basis is provided for the application of LED-induced fluorescence in the rapid, accurate, and real-time detection of soil organic pollutants.

In this experiment, four types of PAHs were selected as the research objects. Two different types of soil, standard soil and actual soil, were selected to prepare soil samples with PAHs. An LED-induced fluorescence detection system mainly comprises an excitation light source, sample pool, optical fiber spectrometer, fluorescence signal acquisition device, and computer control unit. The UV LED array beam irradiates the soil sample, forming a moderately sized circular light spot on the surface of the sample. After filtering, the fluorescence signal generated via excitation was collected using the fluorescence collection device and transmitted to the spectrometer using the UV fiber for beam splitting and detection. Furthermore, the spectral data was stored and analyzed by the computer control unit, which is connected to the spectrometer. The Savitzky–Golay convolution smoothing method was used to preprocess the fluorescence spectrum of the detected PAHs to improve the accuracy of fluorescence signal extraction and effectively retain the useful information in the spectrum while filtering the noise.

In this study, a 255-nm UV LED-induced fluorescence detection system was designed to rapidly detect the fluorescence spectra of different PAHs in soil. The authenticity of the fluorescence spectra obtained by the experimental system was verified by comparing the LED-induced fluorescence spectra of PAHs in the soil with the three-dimensional fluorescence spectra of PAHs in the solution obtained by standard fluorescence instruments, which resulted in a reliable optical support system for the follow-up experiments. Our results show that the number of fluorescence bands and the position of characteristic peaks of PAHs in different soil types are basically the same, while the shape and intensity of the spectral peaks are slightly different. The fluorescence intensity and concentration of PAHs in two different standard soils show a good linear relationship within a certain concentration range, and the linear correlation coefficients are greater than 0.98. Under similar conditions, the system can effectively detect PAHs at lower concentrations in kaolin, and its detection limit is generally lower than PAH samples of loess. Thus, these findings indicate that soil microstructure characteristics and soil matrix complexity, and the combination of soil minerals and PAHs affect the detection of LED-induced fluorescence. The quantitative analysis of PAH soil samples under the actual complex soil background based on the fixed-point wavelength concentration inversion model reveal that the relative errors of the model for the concentration prediction of PAH soil samples in the test set are within the ideal range, except for some low concentration PAH samples. In addition, the average relative errors of different PAH samples are no more than 15.5%.

The LED-induced fluorescence method provides a new approach for detecting PAH organic pollutants in soil and makes up for the limitations of traditional chromatography and fluorescence spectroscopy. Based on the established UV LED-induced fluorescence detection system, this paper analyzed the LED-induced fluorescence characteristics of PAHs in different types of standard soil and further discussed the quantitative analysis of actual soil contaminated with PAHs. The experimental results verify the feasibility of the LED-induced fluorescence method for the rapid detection of PAHs in soil and its good applicability in different soil types. This also provides a technical reference for the rapid insitu detection of PAH organic pollutants in soil.