Study On Quantitative Detection Method Of Apple Epidermal Acetamiprid Based On Enhanced Raman Spectroscopy

As an important pillar that supports social and economic stability,agriculture plays an indispensable role in China’s economic development.However,with the growth of population,the continuous advancement of agricultural technology,the expansion of the pesticide market,and the increasing popularity of pesticide use,the harm caused by pesticide residues to human health has gradually become apparent.In particular,the extensive use of new broad-spectrum insecticides,such as imidacloprid,in apple production has posed a serious challenge to pesticide safety.In order to meet market demand and ensure the quality of fruits and vegetables,it is necessary to strengthen pesticide residue detection and improve the accuracy and efficiency of detection.However,traditional detection methods currently have many shortcomings,such as poor universality,low sensitivity,and susceptibility to interference and errors.Therefore,the introduction of new detection methods and technologies to improve the accuracy and efficiency of pesticide residue detection in fruits and vegetables is of great significance for ensuring public health and promoting sustainable agricultural development.This article proposes a study on the quantitative detection of imidacloprid pesticide residues in apple peels using enhanced Raman spectroscopy technology,and has conducted research in four aspects:simulation,experiment,preprocessing methods,and quantitative analysis.（1）The configuration of the thiamethoxam molecule was optimized and its Raman spectrum was calculated using density functional theory.The results showed obvious Raman peaks at 436,564,628,828,1124,1316,1636,and 2292 cm^-1.Subsequently,a simulation study of the surface-enhanced Raman scattering（SERS）properties of thiamethoxam was conducted,and the enhancement effects of gold cluster Au3 and Au5on its Raman spectrum were analyzed.The study revealed that Au3 and Au5 could significantly enhance the intensity of Raman peaks,including those at 156,436,564,828,1060,1534,1631,2292,and 3052 cm^-1.As the size of the gold clusters increases,their surface reactivity increases,resulting in stronger enhancement effects.The atoms on the surface of the Au5 cluster have sufficient reactivity to form chemical bonds with the adsorbed molecules.In addition,the study found that the N₂₆ site of the thiamethoxam molecule had higher reactivity and affinity than the N₆ site.Furthermore,the atoms close to the surface of the gold clusters were affected,leading to a frequency shift of the peak at 2293 cm^-1 due to physical or chemical adsorption.（2）This study aimed to develop a new non-destructive method for detecting the maximum residue limit of clothianidin on the surface of apples,in accordance with the national standards requirements.To achieve this goal,the researchers converted the maximum residue limit of clothianidin per unit area of the apple surface and proposed a method for detecting clothianidin content on apple peels.This method involved the direct addition of an enhancer onto the apple peel,followed by the use of a portable Raman spectrometer.The method overcame the challenge of identifying low levels of clothianidin content in apple peel samples through SERS spectroscopy experiments.Through comparative experiments,the optimal ratio of clothianidin solution to gold sol was determined to be 5:2.The Raman peak at 1631cm-1 was identified as the surface-enhanced Raman spectroscopy characteristic peak of clothianidin in apple peels.The detection limit of the proposed method was found to be less than 0.3μg/cm²,which is lower than the maximum residue limit of clothianidin on apple surfaces set by the national standards.This research provides a practical and effective method for non-destructive detection of clothianidin residue in apples,which is expected to have significant applications in the food safety industry.（3）Based on the enhanced Raman spectrum data of apple skin detected by portable Raman spectrometer,the preprocessing algorithm of Raman spectrum signal was studied.Signal-to-noise ratio（SNR）and mean square error（MSE）were used as evaluation indices to analyze the performance of different wavelet bases under coefficient truncation and threshold filtering for denoising,as well as the filtering effects of Savitzky-Golay filter,local weighted regression,wavelet transformation filter,and FFT filter under different parameter settings.The Raman spectra of apple peel after denoising using the four different spectral preprocessing algorithms were comprehensively compared.It was found that the Savitzky-Golay filter with a fitting order of 2 and a window size of 21 had the best denoising effect.After processing with this filter algorithm,the SNR was 23.1483and the MSE was 0.0126.（4）The quantitative analysis method of cuprimeidine in apple based on enhanced Raman spectroscopy was studied.This study evaluated the performance of three quantitative regression models and found that machine learning-based models outperformed the traditional least-squares method in terms of prediction accuracy on both training and testing datasets.Among them,the backpropagation（BP）quantitative analysis model showed better prediction results.By optimizing the internal parameters of the BP and extreme learning machine（ELM）models using genetic algorithm（GA）and self-adaptive differential evolution artificial bee colony（SADEABC）algorithm,the results showed that both GA and SADEABC optimization algorithms can improve the prediction accuracy of the BP and ELM models,with SADEABC algorithm outperforming GA algorithm.In terms of model prediction accuracy improvement,SADEABC algorithm consistently outperformed GA algorithm.Among the four hybrid models,the SADEABC-ELM model showed the highest accuracy in predicting the imidacloprid residue in apple peels,with an R² of 0.9946 and a maximum deviation of0.1253%between the predicted and actual values.The proposed SADEABC-ELM model has higher stability and robustness,which can accurately predict the pesticide residues in apple peels,providing important guidance and reference for the effective,reliable,and accurate detection of pesticide residues in fruits and vegetables.