Research On Powder Sample Detection Method Based On UV Raman Spectroscopy

Quick and accurate analysis of powder sample properties is important to ensure public and food safety.Compared with traditional analysis methods,Raman spectroscopy has many advantages such as no sample preparation,multi-component simultaneous detection,and molecular specificity,providing an effective means for rapid analysis of powder samples.Therefore,it is of great significance to analyze powder samples using Raman spectroscopy.At present,there are still some key problems that need to be solved in powder samples detection using Raman spectroscopy,for example,a high-quality Raman spectrum signal cannot be obtained because of low detection sensitivity,the traditional lens-type UV Raman detection method cannot take into account the low sample risk and.high signal collection efficiency,and the difficulty of accurately identifying complex powder samples using single Raman spectroscopy.Therefore,the analysis of the interaction mechanism between light and powder samples is the starting point of this dissertation.Then,the UV Raman spectroscopy detection methods of powder samples are researched.It is expected to realize nondestructive,rapid,and accurate detection of powder samples.The main works of this dissertation are summarized as follows:(1)In view of the problem of weak Raman scattering signals of powder samples,the Raman spectroscopy enhancement methods of powder samples are studied.The near-infrared Raman spectroscopy detection method is studied,and the spectral characteristics of the powder samples are analyzed.Besides,the reasons for the low detection sensitivity are explored.Furthermore,a quartz tube-enhanced Raman spectroscopy detection method is proposed.The interaction process between light and powder in this method is discussed,and the influences of key parameters are analyzed.The results show that the signal intensity of the powder sample is 5.3 times that of the traditional near-infrared Raman spectroscopy.In addition,The UV Raman spectroscopy detection method of the powder sample is analyzed,and the features of the Raman UV spectra of the powder sample are studied.The results show that the UV Raman signal intensity of the powder sample is approximately 27.2 times that of the near-infrared Raman spectroscopy.Finally,the characteristics of the three Raman spectroscopy detection methods are compared from some aspects,such as detection sensitivity,operation complexity,and method applicability.(2)Given the problem of traditional lens-type UV Raman detection,such as the low sample ablation risk and high signal collection efficiency can’t be realized at the same time.A novel detection method based on double-off-axis parabolic mirrors is proposed to reduce the risk of sample ablation and improve the Raman spectroscopy signal collection efficiency.The excitation and collection optical paths are analyzed using optical simulation software.A set of UV Raman detection devices is designed and powder samples are detected.The experimental results show that the Raman signal intensity of the powder sample is 1.58 times that of lens UV Raman spectroscopy.Besides,the signal intensity of the baking soda powder sample is decreased to 54.6%when the excitation time is 50 s using the traditional UV Raman spectroscopy technology.However,the Raman signal intensity of the sample is almost unchanged using the proposed method in this dissertation,effectively reducing the risk of sample ablation.(3)In view of the problem of fluorescence interference in sample detection by UV Raman spectroscopy technology.The overlapping phenomenon of Raman and fluorescence spectrum is analyzed.Then,the separation method of spectral background and signal is studied.Based on the above analysis,a new algorithm that combines polynomial fitting and morphology is proposed.This method combines the advantages of the morphological algorithm to retain spectral features well and the polynomial fitting algorithm to be simple and effective.The problems of fluorescence interference and the complex background of UV Raman spectroscopy are solved.Moreover,the accurate separation of the UV Raman spectral signal is realized.Lastly,the effectiveness of this method is verified by analyzing the simulated and measured spectra.(4)The online measurement method of UV Raman spectroscopy is studied.To improve the recognition ability of powder sample properties,fluorescence spectroscopy technology is introduced under the condition of ensuring that the detection ability of Raman spectroscopy is not affected.The characteristics of UV Raman and fluorescence spectroscopy detection are analyzed,and the UV Raman spectroscopy online measurement optical path is designed.On this basis,the online measurement device is built,and the typical powder samples are detected.The experimental results show that the spectral information of typical powder samples is detected by the device.The problem that Raman spectroscopy technology is hard to accurately identify samples under the conditions of "different bodies with the same spectrum" and "Raman spectrum is complex and similar" is solved.Furthermore,the spectral data are analyzed by the recognition algorithm,which verifies that the online measurement method can improve the identification accuracy of the sample.(5)The research on the near-range(1-2m)UV Raman-Fluorescence detection method for powder samples is carried out.To improve the efficiency of signal collection,a high-precision displacement platform is used to assist signal collection,and focusing detection at different distances is realized.The features of near-range detection are analyzed.In addition,the influence of key factors such as detection distance and sample tilt angle on the detection ability is studied.Besides,the UV Raman-fluorescence detection optical path of near-range powder samples is optimized.Finally,the device is used to detect baking soda,1,2,4,5-tetratoluene,and amino acid powder with a distance of 1.5m and a content of about 4mg/cm2 on four different substrates(metal,cloth,paper,plastic).The ability of the device to detect trace powder samples on different substrates at the near range is verified.