Development Of Micro Near Infrared Spectrometer And Application Of Vegetable Oil Quality Detection

Near-infrared spectroscopy(NIRS)is a new spectroscopic analysis technique which has been rapidly developed in recent years.Compared with traditional disruptive detection technology,NIRS is widely used in agricultural,biological,petrochemical,food,medicine,textile and other fields with its fast and non-destructive advantages.Spectroscopy instrument is the main equipment of spectral analysis technology.With the continuous development of science and technology,large-scale laboratory nearinfrared spectrometer has been very well developed,while miniaturized and portable spectrometer is an important trend of development.Due to complexity of near-infrared spectroscopy,the development of large-scale instruments in China is good,and the development of miniaturized near-infrared spectroscopy is relatively backward compared with foreign countries.Therefore,independent development of miniature near-infrared spectroscopy has far-reaching significance to promote development of near-infrared spectroscopy technology in China.A complete near infrared spectroscopy system was developed using Hamamatsu’s MEMS-FPI spectral detector C13272-01,which included independent adjustable light source,spectral acquisition and Android handset client control.The performance of the spectral system was evaluated and the self-developed stoichiometric software NIRSA Ver 4.4.5 was used to establish the calibration model of three fatty acids(i.e.oleic acid,linoleic acid and linolenic acid)in edible vegetable oil to verify the practical effect of the spectral detection system.The main research achievements are as follows:1.Based on analysis to parameters of spectral detector C13272-01 including wavelength range,wavelength repeatability,dark current,and resolution,an overall design scheme was proposed,an independent adjustable light source was designed to solve the requirement of different light intensity,ATxmega128A3 U was used as the micro control chip,and USB and WIFI dual communication modes were employed for data transmission.2.The design includes an independent light source section and a spectral acquisition section.Voltage adjustment was used to achieve the adjustment of light source to meet the needs of different light intensity.Spectral acquisition part includes moving voltage control power supply,integral I-V amplifier circuit,A/D conversion circuit,power management and communication module.3.The software design includes the bottom driver for spectrum acquisition and the Android mobile client control program,which realizes the setting of basic parameters of the instrument,the acquisition of spectrum,the correction of wavelength,the preservation of data and the calling of models.4.The performance test of the spectrum detection system returned the test results on main technical parameters as follows.Spectral stability: the relative difference between the spectra was between 2% and 3% for 1 h continuous measurement.The standard deviation of the full spectrum of the 100%T baseline of the spectrum was between 0.0025 and 0.0045.Absorbance repeatability: the full range of standard deviations was between 0.0025 and 0.0035.5.In order to verify the practicability of the detection system,GC-MS was employed to determine the fatty acids in edible vegetable oil samples.The relative percentages of oleic acid,linoleic acid and linolenic acid were calculated by peak area normalization method.Trough correction model construction using PLSR method,the correlation coefficient(Rp)and the predicted standard deviation(RMSEP)of oleic acid prediction set were 0.8515 and 0.4681 respectively.The correlation coefficient and predicted standard deviation of linoleic acid prediction set were 0.8901 and 0.4772 respectively.The correlation coefficient and predicted standard deviation of linoleic acid prediction set were 0.8812 and 0.4610 respectively.The results showed that the correlation coefficients of the three fatty acids were higher than 0.85,which could meet the basic requirements of quality control in the production process.The performance of the designed near infrared spectrum detection system has met the design requirements.The results show that the model is stable and reliable,and it can be used in the detection of fatty acids in edible vegetable oil.It is of positive significance to promote the miniaturization of near infrared spectroscopy instrument and creates basic conditions for further cost reduction and development of the instrument for family use.