Rapid detection of moisture content in intermediate Suhuang granules of Suhuang Zhike capsules by near infrared spectroscopy
Objective To determine the moisture content of Suhuang granules.Methods The Antaris Ⅱ fourier transform near infrared spectrometer and Avantes portable near infrared spectrometer were used to collect near infrared spectra of Suhuang granules with different moisture absorption degrees.Meantime,the oven drying was used to measure the moisture content of samples as the reference value.A quantitative analysis model of moisture content of Suhuang granules based on near-infrared spectrum was established with first derivation(1 stD),standard normal variate transformation,multiple scattering correction and partial least squares.Passing-Bablok regression was used to compare the ability of two near infrared devices to predict the moisture content of Suhuang granules.Results The prediction ability of the synthesized models was excellent after the pretreatment with 1stD.R2(c)of FT-NIR model and P-NIR model was both bigger than 0.9897,RMSEC was less than 0.0019,R2(p)was bigger than 0.9716,and RMSEP was less than 0.0029.The R2 of the FT-NIR batch validation model was consistently higher than 0.997,so was the R2 of the P-NIR batch validation model at 0.935,indicating that the models for both instruments had excellent predictive capabilities for unknown batch samples.Conclusion Although the P-NIR method exhibits poorer accuracy than that of FT-NIR,it still demonstrates good stability and precision in detecting the moisture content of Suhuang granules.Both methods can be used in mass production and sample testing.
portable near infrared spectroscopynear-infrared spectroscopySuhuang granuleSuhuang Zhike capsulemoisture
NETL
NSTL
万方数据
