Study On In-situ Monitoring Of Furfural Dissolved In Transformer Oil Based On Silver Nanoparticles Surface Enhanced Raman Scattering

The concentration of furfural dissolved in transformer oil is regarded as an important mark for the degradation diagnose in oil-paper insulation.How to accurately detect furfural concentration in transformer oil is the key to obtain diagnostic results,which is an important technical support to ensure the safe and reliable operation of power transformer.At present,the major methods of detecting the concentration of furfural dissolved in transformer oil are high performance liquid chromatography,spectrophotometry and electrochemical methods.However,these methods need to preprocess transformer oil samples,such as extraction.Besides,they need professional operators,good testing environment,complex operation process,and time-consume.So,these methods can only be done in the laboratory.Compared to traditional methods,Surface enhanced Raman spectroscopy(SERS)employs a single frequency laser to complete the rapid and non-destructive in-situ detection without complex pretreatment process,which has good application prospects in the field of detecting the aging materials dissolved in transformer oil.The application of SERS in the detection of furfural concentration in transformer oil has been investigated in this paper.The main research contents and achievements are as follows:1)Based on the density functional theory(DFT),the furfural molecule model was constructed by Gaussian 09 W software.The Raman vibration characteristics of furfural molecule was analyzed by optimizing the parameters of the basis sets.And the vibration modes of 24 Raman peaks were identified completely.Then,the complex models of furfural molecule-silver cluster with different adsorption positions were established,and the enhanced Raman vibration characteristics of furfural molecule were analyzed.Besides,the assignments of vibrational modes of SERS peaks were identified.All of these results can provide theoretical reference for studying the adsorption characteristics of furfural molecules on the surface of silver substrates.2)In order to meet the requirement for in-situ detection of trace furfural dissolved in transformer oil,the liquid SERS detection platform was designed and set up.The optimal detection parameters were selected,and the detailed in-situ detection process was developed,which provided the hardware support for the experimental study of SERS application.3)The silver nano-plates and nano-bulks were prepared by the modified displacement method on the copper sheet directly.The surface electric field distribution of silver nanoparticles was simulated and analyzed.It is found that the surface enhanced “hot spots” are mainly on the surface of the particles with higher curvature and the gaps between the particles.Copper-based silver nanoparticles were used as the SERS substrates,and transformer oil samples with different furfural concentrations were detected.The Raman peaks at 756 and 1571 cm-1 were selected as the characteristic SERS peaks,and their detection limits were up to 0.50 and 0.42 mg/L,respectively.Based on the least squares method,the quantitative analysis model was constructed by selecting the optimal internal standard.The detection performance of two kinds of SERS substrates were analyzed,such as the enhancement effect,repeatability and stability.The feasibility of the SERS method for detecting furfural concentration in transformer oil was verified.4)In order to obtain better surface topography and higher repeatability,the mask-less reactive ion etching combinated with the magnetron sputtering method was used for the preparation of silicon-based silver nanoparticles.The surface electric field distribution was simulated and analyzed.It is found that the surface enhanced “hot spots” are mainly in the gaps between the columns and on the surface of the particles with higher curvature.Several transformer oil samples with different furfural concentrations were in-situ detected by using silicon-based silver nanoparticles as the SERS substrates.When the Raman peak at 929 cm-1 was selected as the characteristic SERS peak,the detection limit reached 1.2 mg/L.Thus,its enhancement effect was slightly worse than those of copper-based silver nanoparticles.Based on the least squares method,the quantitative analysis model for detecting furfural concentrations in transformer oil was built,where the Raman spectral peak ratio I929/I1005 was used as the furfural function.The uniformity and repeatability of silicon-based silver nanoparticles are better than that of copper-based silver nanoparticles.However,the silicon-based silver nanoparticles are more easily oxidized.In this paper,the SERS vibration characteristics of furfural molecule were calculated and analyzed based on the density functional theory;the liquid SERS detection platform was designed and set up for the in-situ detection of furfural dissolved in transformer oil;silver nanoparticles with different morphologies were prepared by chemical and physical methods,and their in-situ detection performances for furfural dissolved in transformer oil were compared and analyzed.The experimental results verified the feasibility of the SERS method for in-situ detecting the concentration of furfural dissolved in transformer oil,which lays the foundation for the in-situ detection of furfural dissolved in transformer oil by Raman spectroscopy method.