CN102759509A - Detection method of cassiabarktree twig tuckahoe capsules - Google Patents
Detection method of cassiabarktree twig tuckahoe capsules Download PDFInfo
- Publication number
- CN102759509A CN102759509A CN2011101041005A CN201110104100A CN102759509A CN 102759509 A CN102759509 A CN 102759509A CN 2011101041005 A CN2011101041005 A CN 2011101041005A CN 201110104100 A CN201110104100 A CN 201110104100A CN 102759509 A CN102759509 A CN 102759509A
- Authority
- CN
- China
- Prior art keywords
- sample
- near infrared
- spectrum data
- infrared spectrum
- service
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a detection method of cassiabarktree twig tuckahoe capsules. The method comprises the steps that: solid powder in the cassiabarktree twig tuckahoe capsules is adopted as a sample requiring detection; the sample requiring detection is detected by using a near-infrared spectrometer, so that near infra-red spectrum data of the sample is obtained; according to correction models of gallic acid, peoniflorin, benzoic acid, benzoylpaeoniflorin, cinnamaldehyde, paeonol, and amygdalin, and the obtained near infra-red spectrum data, the contents of gallic acid, peoniflorin, benzoic acid, benzoylpaeoniflorin, cinnamaldehyde, paeonol, and amygdalin in the sample are obtained. According to the invention, the sample requiring detection does not need complicated pretreatments such as separation and extraction. The detection time is shortened, the detection efficiency is improved, and work load is reduced. Also, when the detection method provided by the invention is adopted in detections upon large batches of samples, the result is accurate, and the repeatability is good.
Description
Technical field
The invention belongs to Chinese medicine detection technique field, relate in particular to a kind of detection method of GUIZHI FULING JIAONANG.
Background technology
GUIZHI FULING JIAONANG is the modern formulation of ancient prescription guizhi fuling pill; Derive from Han dynasty Zhang Zhongjing " Synopsis Golden Chamber "; Mainly form by Chinese medicines such as cassia twig, Poria cocos, the root bark of tree peony, Chinese herbaceous peony, peach kernels; Effect with promoting blood circulation and removing blood stasis, the Li Shui that eliminates the phlegm, eliminating mass eliminate indigestion; Can be used for treating gynecological disease, like various disease conditions such as dysmenorrhoea, amenorrhoea, lochiorrhagia, pelvic infecton, pelvic lump, fibroid, uterus adenomyosis, ovarian cyst, the proliferation of mammary gland, infertility, vaginal bleeding after drug abortion, mullerianosises.
GUIZHI FULING JIAONANG generally prepares according to following steps: extract the effective constituent of Chinese crude drugs such as cassia twig, Poria cocos, the root bark of tree peony, Chinese herbaceous peony, peach kernel, after technologies such as concentrated, drying, prepare capsule.In GUIZHI FULING JIAONANG, the effective constituent of cassia twig is mainly cinnaldehydrum; The effective constituent of the root bark of tree peony is mainly Paeonol and Paeoniflorin; The effective constituent of peach kernel is mainly amarogentin; The effective constituent of Chinese herbaceous peony is mainly Paeoniflorin and benzoylpaeoniflorin; The effective constituent of Poria cocos is mainly gallic acid.Above-mentioned effective constituent had both contained volatile ingredient; Contain liposoluble constituent again; Also contain water soluble ingredient, each effective constituent nature difference is bigger, the difficult control of its content in the preparation process of GUIZHI FULING JIAONANG; Therefore, to the main effective constituent of GUIZHI FULING JIAONANG detect be guarantee GUIZHI FULING JIAONANG steady quality, high conformity important means it
General at present employing high performance liquid chromatography detects the main effective constituent of GUIZHI FULING JIAONANG; When adopting high performance liquid chromatography to detect; Need to prepare at first respectively the sample of volatile ingredient, liposoluble constituent and water soluble ingredient, just can detect, three kinds of method of sample preparation are following: in the powder of removing capsule shell, add water and ether; 90min refluxes in 75 ℃ of water-baths; Get ether layer after the cooling, the dissolving that adds diethyl ether after in 35 ℃ of water-baths, volatilizing obtains the sample of volatile ingredient; In the powder of removing capsule shell, add water; The situation refluxed 30min of boiling, the cooling back is centrifugal, the supernatant after centrifugal is obtained the sample of water soluble ingredient behind membrane filtration; With the boiling reflux 30min in the methyl alcohol that is deposited in after centrifugal; The cooling back is centrifugal, adds dissolve with methanol after the methanol solution that obtains is volatilized in 75 ℃ of water-baths, behind membrane filtration, obtains the sample of liposoluble constituent.Though adopt high performance liquid chromatography that the effective constituent of GUIZHI FULING JIAONANG is detected to have the separation efficiency height, advantage such as sensitivity height; But pre-treatment is comparatively loaded down with trivial details; Detection time is longer, when being unfavorable for producing in enormous quantities to the quality control of GUIZHI FULING JIAONANG.
Summary of the invention
In view of this, the technical matters that the present invention will solve is to provide a kind of detection method of GUIZHI FULING JIAONANG, and detection method provided by the invention need not to carry out the complicated sample pre-treatment, and detection time is short, testing result is accurate.
The invention provides a kind of detection method of GUIZHI FULING JIAONANG, may further comprise the steps:
With the pressed powder in the GUIZHI FULING JIAONANG is testing sample, utilizes near infrared spectrometer to detect said testing sample, obtains the near infrared spectrum data of said testing sample;
According to the calibration model and the said near infrared spectrum data of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin, obtain the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin in the said testing sample.
Preferably, when utilizing near infrared spectrometer to detect said testing sample, the thickness of said testing sample is 1mm～3mm, and scanning times is 500 times～700 times.
Preferably, the calibration model of said gallic acid is set up according to following method:
A11) GUIZHI FULING JIAONANG pressed powder sample is provided;
A12) utilize high performance liquid chromatograph to detect the content of gallic acid in the said sample, obtain the laboratory values of gallic acid in the said sample;
A13) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A14) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of gallic acid according to the result of said pre-service acquisition and the laboratory values of said gallic acid.
Preferably, said step a14) in, said pretreated wave band is 1300nm～2300nm.
Preferably, the calibration model of said Paeoniflorin is set up according to following method:
A21) GUIZHI FULING JIAONANG pressed powder sample is provided;
A22) utilize high performance liquid chromatograph to detect content of paeoniflorin in the said sample, obtain the laboratory values of Paeoniflorin in the said sample;
A23) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A24) adopt polynary scatter correction method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of Paeoniflorin according to the result of said pre-service acquisition and the laboratory values of said Paeoniflorin.
Preferably, said step a24) in, said pretreated wave band is 1100nm～2100nm.
Preferably, said benzoic calibration model is set up according to following method:
A31) GUIZHI FULING JIAONANG pressed powder sample is provided;
A32) utilize high performance liquid chromatograph to detect benzoic content in the said sample, obtain benzoic laboratory values in the said sample;
A33) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A34) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up benzoic quantitative correction model according to result and the said benzoic laboratory values that said pre-service obtains.
Preferably, said step a34) in, said pretreated wave band is 1100nm～2300nm.
Preferably, the calibration model of said benzoylpaeoniflorin is set up according to following method:
A41) GUIZHI FULING JIAONANG pressed powder sample is provided;
A42) utilize high performance liquid chromatograph to detect the content of benzoylpaeoniflorin in the said sample, obtain the laboratory values of benzoylpaeoniflorin in the said sample;
A43) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A44) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service; According to the result of said pre-service acquisition and the laboratory values of said benzoylpaeoniflorin, adopt PLS and cross validation method to set up the quantitative correction model of benzoylpaeoniflorin.
Preferably, said step a44) in, said pretreated wave band is 1100nm～1900nm.
Preferably, the calibration model of said cinnaldehydrum is set up according to following method:
A51) GUIZHI FULING JIAONANG pressed powder sample is provided;
A52) utilize high performance liquid chromatograph to detect the content of cinnaldehydrum in the said sample, obtain the laboratory values of cinnaldehydrum in the said sample;
A53) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A54) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of cinnaldehydrum according to the result of said pre-service acquisition and the laboratory values of said cinnaldehydrum.
Preferably, said step a54) in, said pretreated wave band is 1100nm～1900nm.
Preferably, the calibration model of said Paeonol is set up according to following method:
A61) GUIZHI FULING JIAONANG pressed powder sample is provided;
A62) utilize high performance liquid chromatograph to detect the content of Paeonol in the said sample, obtain the laboratory values of Paeonol in the said sample;
A63) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A64) adopt polynary scatter correction method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of Paeonol according to the result of said pre-service acquisition and the laboratory values of said Paeonol.
Preferably, said step a64) in, said pretreated wave band is 1300nm～2300nm.
Preferably, the calibration model of said amarogentin is set up according to following method:
A71) GUIZHI FULING JIAONANG pressed powder sample is provided;
A72) utilize high performance liquid chromatograph to detect the content of amarogentin in the said sample, obtain the laboratory values of amarogentin in the said sample;
A73) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A74) adopt the standard normalization method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of amarogentin according to the result of said pre-service acquisition and the laboratory values of said amarogentin.
Preferably, said step a74) in, said pretreated wave band is 1100nm～2300nm.
Compared with prior art; The present invention adopts NIR technology; In conjunction with Chemical Measurement and computer software technology; Realize detection through the calibration model of setting up gallic acid in the GUIZHI FULING JIAONANG, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin, thereby foundation is provided for the quality control of GUIZHI FULING JIAONANG GUIZHI FULING JIAONANG principal ingredient content.Detection method provided by the invention need not detection time has been shortened in the pre-treatment that testing sample separates, extraction etc. is complicated, has improved detection efficiency, has reduced workload.In addition, when detection method provided by the invention detected batch samples, the result was accurate, favorable reproducibility.
Experiment shows; Adopt method provided by the invention that the testing result of GUIZHI FULING JIAONANG is compared with the testing result that adopts high performance liquid chromatography; The relative average debiation of gallic acid is 1.627%; The relative average debiation of Paeoniflorin is 1.550%, benzoic relative average debiation is 1.012%, the relative average debiation of benzoylpaeoniflorin is 2.600%, the relative average debiation of cinnaldehydrum is 2.398%, the relative average debiation of Paeonol is 2.109%, the relative average debiation of amarogentin is 1.089%, and testing result is accurate, precision is good.
Description of drawings
The gallic acid quantitative correction model that Fig. 1 sets up for the embodiment of the invention;
The Paeoniflorin quantitative correction model that Fig. 2 sets up for the embodiment of the invention;
The benzoic acid quantitative correction model that Fig. 3 sets up for the embodiment of the invention;
The benzoylpaeoniflorin quantitative correction model that Fig. 4 sets up for the embodiment of the invention;
The cinnaldehydrum quantitative correction model that Fig. 5 sets up for the embodiment of the invention;
The Paeonol quantitative correction model that Fig. 6 sets up for the embodiment of the invention;
The amarogentin quantitative correction model that Fig. 7 sets up for the embodiment of the invention.
Embodiment
The invention provides a kind of detection method of GUIZHI FULING JIAONANG, may further comprise the steps:
With the pressed powder in the GUIZHI FULING JIAONANG is testing sample, utilizes near infrared spectrometer to detect said testing sample, obtains the near infrared spectrum data of said testing sample;
According to the calibration model and the said near infrared spectrum data of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin, obtain the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin in the said testing sample.
The present invention adopts NIR technology; In conjunction with Chemical Measurement and computer software technology; Realize detection through the calibration model of setting up gallic acid in the GUIZHI FULING JIAONANG, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin, thereby foundation is provided for the quality control of GUIZHI FULING JIAONANG GUIZHI FULING JIAONANG principal ingredient content.
In the present invention; Said GUIZHI FULING JIAONANG is for by cassia twig, Poria cocos, the root bark of tree peony, Chinese herbaceous peony and the peach kernel traditional Chinese medicine of the five flavours capsule through extracting, concentrate, making after the drying; Be preferably by cassia twig, the root of herbaceous peony, Poria cocos, peach kernel and the root bark of tree peony through extract, concentrate and drying after the capsule that makes, its effective constituent mainly comprises: gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin etc.
Before GUIZHI FULING JIAONANG is detected; At first set up each principal ingredient in the GUIZHI FULING JIAONANG; Like the calibration model of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin etc., obtain the content of each principal ingredient in the testing sample then according to the near infrared spectrum data analysis of testing sample.
In the present invention, said gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin calibration model are preferably set up according to following steps:
A) GUIZHI FULING JIAONANG pressed powder sample is provided;
B) utilize high performance liquid chromatograph to detect the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin in the said sample;
C) near infrared spectrum data of the said sample of collection;
D) said near infrared spectrum data is carried out pre-service, the content that obtains according to said pretreated result and step b) is set up the quantitative correction model of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin respectively.
For the present invention clearly is described; Below respectively the calibration modeling process of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin is described in detail; But; Those skilled in the art can be known; The calibration modeling process of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin is not independently to carry out fully, as GUIZHI FULING JIAONANG pressed powder sample is provided, utilizes steps such as high performance liquid chromatograph detection to accomplish simultaneously.
According to the present invention, the calibration model of said gallic acid is preferably set up according to following method:
A11) GUIZHI FULING JIAONANG pressed powder sample is provided;
A12) utilize high performance liquid chromatograph to detect the content of gallic acid in the said sample, obtain the laboratory values of gallic acid in the said sample;
A13) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A14) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of gallic acid according to the result of said pre-service acquisition and the laboratory values of said gallic acid.
GUIZHI FULING JIAONANG pressed powder sample (hereinafter referred sample) at first is provided.According to the present invention, said sample size is preferably more than 40, more preferably more than 50, most preferably is more than 60.After obtaining sample, at random said sample is divided into calibration set and checking collection, wherein, the sample in the calibration set is used to set up calibration model, and the sample that checking is concentrated is used to verify calibration model.
According to the present invention, need carry out efficient liquid phase chromatographic analysis and near-infrared analysis to said sample, so that obtain the laboratory values and the near infrared spectrum data of gallic acid in the said sample, thereby set up calibration model.
The present invention utilizes high performance liquid chromatograph to detect the content of gallic acid in the said sample, obtains the laboratory values of gallic acid in the said sample.The present invention is preferably according to " high-efficient liquid phase chromatogram condition of stipulating among an appendix VI of Chinese pharmacopoeia version in 2010 D detects the content of gallic acid in the said GUIZHI FULING JIAONANG.
When said sample is carried out near-infrared analysis, at first utilize near infrared spectrometer to gather the near infrared spectrum data of said sample.The present invention preferably under rate mode, the near infrared spectrum data that adopts static test sample mode and diffuse reflection mode to gather said sample.Carry out near infrared spectrum data when gathering, said sample thickness is preferably more than the 1cm, 1cm～3cm more preferably, and experiment shows, and thickness of sample is during less than 1cm, and the spectrum repeatability that collects is relatively poor.When carrying out the near infrared spectrum data collection, the scanning times of near infrared spectrometer is preferably 300 times～700 times, more preferably 500 times～700 times.
After obtaining the near infrared spectrum data of said sample, set up the quantitative correction model of gallic acid in the sample.
The present invention preferably adopts following method to set up the quantitative correction model of gallic acid:
Said near infrared spectrum data is carried out pre-service;
According to the result of said pre-service acquisition and the laboratory values of said gallic acid, adopt PLS and cross validation method to set up the quantitative correction model of gallic acid.
Before setting up the quantitative correction model of gallic acid, the present invention preferably rejects the chemical abnormality value (outlier) in the near infrared spectrum data of said sample through spectrum influence value (Leverage) and chemical score error (Residual) check.
When setting up the quantitative correction model of gallic acid; At first the near infrared spectrum data of said sample is carried out pre-service; Said pretreated method can be preferably 9 smoothing methods of first order differential for polynary scatter correction method, standard normalization method or 9 smoothing methods of first order differential.After adopting 9 smoothing methods of first order differential to carry out pre-service, when the gallic acid quantitative correction model that obtains is verified, its internal chiasma coefficient of determination (R
2) approach 1 most, internal chiasma checking mean square deviation (RMSEC) and checking root-mean-square-deviation (RMSEP) minimum.
After said near infrared spectrum data carried out pre-service, the result who obtains according to said pre-service and the laboratory values of said gallic acid adopted PLS and cross validation method to set up the quantitative correction model of gallic acid.
Adopting PLS to set up in the quantitative correction model process of gallic acid, though PLS can be handled full spectrum information, can comprise bulk redundancy information when wave band is wide, reduce the precision of prediction of calibration model.In order to improve the precision of prediction of imitating positive model, the wave band among the present invention is preferably 1300nm～2300nm, when wave band is in this scope, and when the gallic acid quantitative correction model that obtains is verified, its internal chiasma coefficient of determination (R
2) approach 1 most, internal chiasma checking mean square deviation (RMSEC) and checking root-mean-square-deviation (RMSEP) minimum.
Adopting PLS to set up in the quantitative correction model process of gallic acid, best number of principal components is 5.
Adopt PLS that the laboratory values of pretreated near infrared spectrum data of said process and said gallic acid is associated; Set up the quantitative correction model of gallic acid; And after adopting the cross validation method that the parameters of the quantitative correction model of said gallic acid is optimized, obtain the quantitative correction model of gallic acid.
According to the present invention, the calibration model of said Paeoniflorin is preferably set up according to following method:
A21) GUIZHI FULING JIAONANG pressed powder sample is provided;
A22) utilize high performance liquid chromatograph to detect content of paeoniflorin in the said sample, obtain the laboratory values of Paeoniflorin in the said sample;
A23) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A24) adopt polynary scatter correction method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of Paeoniflorin according to the result of said pre-service acquisition and the laboratory values of said Paeoniflorin.
According to the present invention, said benzoic calibration model is preferably set up according to following method:
A31) GUIZHI FULING JIAONANG pressed powder sample is provided;
A32) utilize high performance liquid chromatograph to detect benzoic content in the said sample, obtain benzoic laboratory values in the said sample;
A33) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A34) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up benzoic quantitative correction model according to result and the said benzoic laboratory values that said pre-service obtains.
According to the present invention, the calibration model of said benzoylpaeoniflorin is preferably set up according to following method:
A41) GUIZHI FULING JIAONANG pressed powder sample is provided;
A42) utilize high performance liquid chromatograph to detect the content of benzoylpaeoniflorin in the said sample, obtain the laboratory values of benzoylpaeoniflorin in the said sample;
A43) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A44) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service; According to the result of said pre-service acquisition and the laboratory values of said benzoylpaeoniflorin, adopt PLS and cross validation method to set up the quantitative correction model of benzoylpaeoniflorin.
According to the present invention, the calibration model of said cinnaldehydrum is preferably set up according to following method:
A51) GUIZHI FULING JIAONANG pressed powder sample is provided;
A52) utilize high performance liquid chromatograph to detect the content of cinnaldehydrum in the said sample, obtain the laboratory values of cinnaldehydrum in the said sample;
A53) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A54) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of cinnaldehydrum according to the result of said pre-service acquisition and the laboratory values of said cinnaldehydrum.
According to the present invention, the calibration model of said Paeonol is preferably set up according to following method:
A61) GUIZHI FULING JIAONANG pressed powder sample is provided;
A62) utilize high performance liquid chromatograph to detect the content of Paeonol in the said sample, obtain the laboratory values of Paeonol in the said sample;
A63) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A64) adopt polynary scatter correction method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of Paeonol according to the result of said pre-service acquisition and the laboratory values of said Paeonol.
According to the present invention, the calibration model of said amarogentin is preferably set up according to following method:
A71) GUIZHI FULING JIAONANG pressed powder sample is provided;
A72) utilize high performance liquid chromatograph to detect the content of amarogentin in the said sample, obtain the laboratory values of amarogentin in the said sample;
A73) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A74) adopt the standard normalization method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of amarogentin according to the result of said pre-service acquisition and the laboratory values of said amarogentin.
In the calibration modeling process of above-mentioned Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin, except that setting up the process of quantitative model, other processes are identical with the gallic acid calibration modeling process.
When setting up the quantitative correction model of Paeoniflorin, preferably adopt polynary scatter correction method that the near infrared spectrum data of said sample is carried out pre-service; Wave band when setting up model is preferably 1100nm～2100nm; Best number of principal components is 9.
When setting up benzoic quantitative correction model, preferably adopt 9 smoothing methods of first order differential that the near infrared spectrum data of said sample is carried out pre-service; Wave band when setting up model is preferably 1100nm～2300nm; Best number of principal components is 10.
When setting up the quantitative correction model of benzoylpaeoniflorin, preferably adopt 9 smoothing methods of first order differential that the near infrared spectrum data of said sample is carried out pre-service; Wave band when setting up model is preferably 1100nm～1900nm; Best number of principal components is 5.
When setting up the quantitative correction model of cinnaldehydrum, preferably adopt 9 smoothing methods of first order differential that the near infrared spectrum data of said sample is carried out pre-service; Wave band when setting up model is preferably 1100nm～1900nm; Best number of principal components is 6.
When setting up the quantitative correction model of Paeonol, preferably adopt polynary scatter correction method that the near infrared spectrum data of said sample is carried out pre-service; Wave band when setting up model is preferably 1300nm～2300nm; Best number of principal components is 11.
When setting up the quantitative correction model of amarogentin, preferably adopt the standard normalization method that the near infrared spectrum data of said sample is carried out pre-service; Wave band when setting up model is preferably 1100nm～2300nm; Best number of principal components is 6.
After setting up gallic acid calibration model, Paeoniflorin calibration model, benzoic acid calibration model, benzoylpaeoniflorin calibration model, cinnaldehydrum calibration model, Paeonol calibration model and amarogentin calibration model respectively; With each calibration model near-infrared analyzer of packing into; Can realize fast detecting, specifically may further comprise the steps GUIZHI FULING JIAONANG:
The near infrared spectrometer that above-mentioned each calibration model is equipped with in utilization detects GUIZHI FULING JIAONANG sample to be measured, obtains the near infrared spectrum data of said testing sample;
Above-mentioned each calibration model is analyzed according to said near infrared spectrum data, obtains the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin in the said GUIZHI FULING JIAONANG sample to be measured.
When detecting, said GUIZHI FULING JIAONANG sample thickness to be measured is preferably more than the 1cm, more preferably 1cm～3cm; The scanning times of said near infrared spectrometer is 300 times～700 times, more preferably 500 times～700 times.
Compared with prior art, detection method provided by the invention need not detection time has been shortened in the pre-treatment that testing sample separates, extraction etc. is complicated, has improved detection efficiency, has reduced workload.In addition, detection method provided by the invention is not affected by other factors, and when batch samples was detected, the result was accurate, favorable reproducibility.Experiment shows; Adopt method provided by the invention that the testing result of GUIZHI FULING JIAONANG is compared with the testing result that adopts high performance liquid chromatography; The relative average debiation of gallic acid is 1.627%; The relative average debiation of Paeoniflorin is 1.550%, benzoic relative average debiation is 1.012%, the relative average debiation of benzoylpaeoniflorin is 2.600%, the relative average debiation of cinnaldehydrum is 2.398%, the relative average debiation of Paeonol is 2.109%, the relative average debiation of amarogentin is 1.089%, and testing result is accurate, precision is good.
In order to further specify the present invention, the detection method of GUIZHI FULING JIAONANG provided by the invention is described in detail below in conjunction with embodiment.
Embodiment 1
59 GUIZHI FULING JIAONANG finished products that provide with Kangyuan Pharmaceutical Co., Ltd., Jiangsu Prov quality standard research center are sample, are divided into 53 calibration sets and 6 checking collection at random, are designated as correcting sample 1～53 respectively, verification sample 1～6; The portable AOTF near infrared spectrometer of Luminar5030 that adopts U.S. BRIMROSE company to produce detects, and detecting device is InGaAs; The spectra collection process software be Snap! Quantitative analysis software is TheUnscrambler; Testing conditions is following: wavelength coverage 1100nm～2300nm; Wavelength increment 2.0nm; Rate mode Ratio Mode.
1.1 high effective liquid chromatography for measuring GUIZHI FULING JIAONANG sample
Respectively according to " high-efficient liquid phase chromatogram condition of stipulating among an appendix VI of Chinese pharmacopoeia version in 2010 D detects the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin in the said GUIZHI FULING JIAONANG sample, obtains its laboratory values respectively; Wherein, the testing conditions of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum and Paeonol is following:
Chromatographic condition and system suitability test: with the octadecylsilane chemically bonded silica is filling agent (chromatographic column: Waters Symmetry C
18, 4.6 * 250mm, 5 μ m); To contain 0.02% trifluoroacetic acid aqueous solution is mobile phase A, is Mobile phase B with the acetonitrile, and the gradient elution program is as shown in table 1:
Table 1 gradient elution program
Flow velocity: 1mL/min;
The detection wavelength of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin and Paeonol is 230nm;
The detection wavelength of cinnaldehydrum is 275nm;
Number of theoretical plate is pressed the Paeoniflorin peak and is calculated, and should be not less than 4000.
Mix the preparation of reference substance solution: gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum and Paeonol reference substance decided in accurate title; Add 50% methyl alcohol and process the solution that every 1mL contains 100 μ g, 250 μ g, 15 μ g, 10 μ g, 30 μ g, 20 μ g, 150 μ g respectively, promptly get and mix reference substance solution.
The preparation of need testing solution: get about 0.25g GUIZHI FULING JIAONANG powder, the accurate title, decide, and places tool plug conical flask, accurate 50% methyl alcohol 25mL, the close plug of adding.Claim decide weight, ultrasonic (power is 250W, and frequency is 40KHz) handled 30min, puts coldly, weighs, and supplies the weight that subtracts mistake with 50% methyl alcohol, and subsequent filtrate is got in filtration, promptly gets need testing solution.
Determination method: accurate respectively the absorption mixed reference substance liquid and each 10 μ L of test sample liquid, injects liquid chromatograph, measures, and can obtain the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum and Paeonol.
The testing conditions of amarogentin is following:
Chromatographic condition and system suitability test: with the octadecylsilane chemically bonded silica is filling agent, and methanol-water (20: 80) is a moving phase;
Flow velocity: 1mL/min;
Detect wavelength 218nm;
Number of theoretical plate is pressed the amarogentin peak and is calculated, and should be not less than 2500.
The preparation of reference substance solution: it is an amount of to get the amarogentin reference substance, and accurate the title decides, and adds 50% ethanol and processes the solution that every 1mL contains 90 μ g, promptly gets reference substance solution.
The preparation of need testing solution: get about 0.2g GUIZHI FULING JIAONANG powder, the accurate title, decide, and places tool plug conical flask, the accurate 50% ethanol 25mL that adds; Claim to decide weight, ultrasonic (power is 250W, and frequency is 40KHz) handled 30min, puts cold; Weigh, supply the weight that subtracts mistake, filter with 50% ethanol; Discard filtrating just, get subsequent filtrate, promptly get need testing solution.
Determination method: accurate respectively reference substance solution and each 10 μ L of need testing solution of drawing, inject liquid chromatograph, measure, can obtain the content of amarogentin.
1.2 the near infrared spectrum data of collected specimens
Said sample is loaded in the sample cup, utilizes near infrared spectrometer, under rate mode, adopt the near infrared spectrum of diffuse reflection mode collected specimens, each sample repeats the scanning of dress appearance and averages the near infrared spectrum data of collected specimens three times.
1.2.1 confirming of thickness of sample
Make thickness of sample be respectively 0.5cm, 1cm, 1.5cm and 2.0cm, scan 600 times, obtain original spectrum respectively; And said original spectrum is carried out first order derivative handle; The result shows, when thickness of sample is 0.5cm, and original spectrum and first derivative spectrum repeated bad; Thickness of sample is 1mm when above, the repeatability of original spectrum and first derivative spectrum better, and the movable block standard deviation value of its first derivative spectrum (MBSD value) otherness is little.
1.2.2 confirming of scanning times
Making thickness of sample is 1cm, scans respectively 300 times and 600 times, obtains original spectrum respectively; And said original spectrum is carried out first order derivative handle; The result shows, the repeatability of the spectrum that scanning times obtains when being 300 times and 600 times is all better, still; When scanning times was 600 times, the MBSD value difference opposite sex of first derivative spectrum was littler.
In the present embodiment, making thickness of sample is under 1cm, 600 times the situation of scanning said sample to be analyzed, and obtains the near infrared spectrum data of each sample.
1.3 the foundation of quantitative correction model
1.3.1 the foundation of gallic acid quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The laboratory values of the gallic acid that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up gallic acid quantitative correction model with PLS and cross validation method.
1.3.1.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 2, table 2 is that preprocess method is to setting up the result that influences of gallic acid quantitative correction model.
Table 2 preprocess method is to setting up the result that influences of gallic acid quantitative correction model
Can know that by table 2 when adopting 9 smoothing methods of first order differential to carry out pre-service, the gallic acid quantitative correction model performance that obtains is better.
1.3.1.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up gallic acid quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 3, table 3 is that different-waveband is to setting up the result that influences of gallic acid quantitative correction model.
Table 3 different-waveband is to setting up the result that influences of gallic acid quantitative correction model
Can know that by table 3 when wave band was 1300nm～2300nm, the gallic acid quantitative correction model performance that obtains was better.
1.3.1.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up gallic acid quantitative correction model is 5.
1.3.1.4 the foundation of quantitative correction model
Adopt 9 smoothing methods of first order differential that 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample are carried out pre-service; Laboratory values according to the gallic acid that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up gallic acid quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1300nm～2300nm; Number of principal components is 5, and the result is referring to Fig. 1, and Fig. 1 is the gallic acid quantitative correction model of embodiment of the invention foundation.
1.3.1.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the gallic acid quantitative correction model that utilizes 1.3.1.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
1.3.2 the foundation of Paeoniflorin quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The laboratory values of the Paeoniflorin that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up Paeoniflorin quantitative correction model with PLS and cross validation method.
1.3.2.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 4, table 4 is that preprocess method is to setting up the result that influences of Paeoniflorin quantitative correction model.
Table 4 preprocess method is to setting up the result that influences of Paeoniflorin quantitative correction model
Can know that by table 4 when adopting polynary scatter correction method to carry out pre-service, the Paeoniflorin quantitative correction model performance that obtains is better.
1.3.2.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up Paeoniflorin quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 5, table 5 is that different-waveband is to setting up the result that influences of Paeoniflorin quantitative correction model.
Table 5 different-waveband is to setting up the result that influences of Paeoniflorin quantitative correction model
Can know that by table 5 when wave band was 1100nm～2100nm, the Paeoniflorin quantitative correction model performance that obtains was better.
1.3.2.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up Paeoniflorin quantitative correction model is 9.
1.3.2.4 the foundation of quantitative correction model
Adopt the multiple scattering correction method that 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample are carried out pre-service; Laboratory values according to the Paeoniflorin that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up Paeoniflorin quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1100nm～2100nm; Number of principal components is 9, and the result is referring to Fig. 2, and Fig. 2 is the Paeoniflorin quantitative correction model of embodiment of the invention foundation.
1.3.2.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the Paeoniflorin quantitative correction model that utilizes 1.3.2.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
1.3.3 the foundation of benzoic acid quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The benzoic laboratory values that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up benzoic acid quantitative correction model with PLS and cross validation method.
1.3.3.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 6, table 6 is that preprocess method is to setting up the result that influences of benzoic acid quantitative correction model.
Table 6 preprocess method is to setting up the result that influences of benzoic acid quantitative correction model
Can know that by table 6 when adopting 9 smoothing methods of first order differential to carry out pre-service, the benzoic acid quantitative correction model performance that obtains is better.
1.3.3.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up benzoic acid quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 7, table 7 is that different-waveband is to setting up the result that influences of benzoic acid quantitative correction model.
Table 7 different-waveband is to setting up the result that influences of benzoic acid quantitative correction model
Can know that by table 7 when wave band was 1100nm～2300nm, the benzoic acid quantitative correction model performance that obtains was better.
1.3.3.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up benzoic acid quantitative correction model is 10.
1.3.3.4 the foundation of quantitative correction model
Adopt 9 smoothing methods of first order differential that 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample are carried out pre-service; According to the benzoic laboratory values that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up benzoic acid quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1100nm～2300nm; Number of principal components is 10, and the result is referring to Fig. 3, and Fig. 3 is the benzoic acid quantitative correction model of embodiment of the invention foundation.
1.3.3.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the benzoic acid quantitative correction model that utilizes 1.3.3.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
1.3.4 the foundation of benzoylpaeoniflorin quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The laboratory values of the benzoylpaeoniflorin that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up benzoylpaeoniflorin quantitative correction model with PLS and cross validation method.
1.3.4.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 8, table 8 is that preprocess method is to setting up the result that influences of benzoylpaeoniflorin quantitative correction model.
Table 8 preprocess method is to setting up the result that influences of benzoylpaeoniflorin quantitative correction model
Can know that by table 8 when adopting 9 smoothing methods of first order differential to carry out pre-service, the benzoylpaeoniflorin quantitative correction model performance that obtains is better.
1.3.4.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up benzoylpaeoniflorin quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 9, table 9 is that different-waveband is to setting up the result that influences of benzoylpaeoniflorin quantitative correction model.
Table 9 different-waveband is to setting up the result that influences of benzoylpaeoniflorin quantitative correction model
Can know that by table 9 when wave band was 1100nm～1900nm, the benzoylpaeoniflorin quantitative correction model performance that obtains was better.
1.3.4.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up benzoylpaeoniflorin quantitative correction model is 5.
1.3.4.4 the foundation of quantitative correction model
Adopt 9 smoothing methods of first order differential that 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample are carried out pre-service; Laboratory values according to the benzoylpaeoniflorin that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up benzoylpaeoniflorin quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1100nm～1900nm; Number of principal components is 5, and the result is referring to Fig. 4, and Fig. 4 is the benzoylpaeoniflorin quantitative correction model of embodiment of the invention foundation.
1.3.4.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the benzoylpaeoniflorin quantitative correction model that utilizes 1.3.4.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
1.3.5 the foundation of cinnaldehydrum quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The laboratory values of the cinnaldehydrum that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up cinnaldehydrum quantitative correction model with PLS and cross validation method.
1.3.5.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 10, table 10 is that preprocess method is to setting up the result that influences of cinnaldehydrum quantitative correction model.
Table 10 preprocess method is to setting up the result that influences of cinnaldehydrum quantitative correction model
Can know that by table 10 when adopting 9 smoothing methods of first order differential to carry out pre-service, the cinnaldehydrum quantitative correction model performance that obtains is better.
1.3.5.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up cinnaldehydrum quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 11, table 11 is that different-waveband is to setting up the result that influences of cinnaldehydrum quantitative correction model.
Table 11 different-waveband is to setting up the result that influences of cinnaldehydrum quantitative correction model
Can know that by table 11 when wave band was 1100nm～1900nm, the cinnaldehydrum quantitative correction model performance that obtains was better.
1.3.5.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up cinnaldehydrum quantitative correction model is 6.
1.3.5.4 the foundation of quantitative correction model
Adopt 9 smoothing methods of first order differential that 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample are carried out pre-service; Laboratory values according to the cinnaldehydrum that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up cinnaldehydrum quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1100nm～1900nm; Number of principal components is 6, and the result is referring to Fig. 5, and Fig. 5 is the cinnaldehydrum quantitative correction model of embodiment of the invention foundation.
1.3.5.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the cinnaldehydrum quantitative correction model that utilizes 1.3.5.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
1.3.6 the foundation of Paeonol quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The laboratory values of the Paeonol that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up Paeonol quantitative correction model with PLS and cross validation method.
1.3.6.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 12, table 12 is that preprocess method is to setting up the result that influences of Paeonol quantitative correction model.
Table 12 preprocess method is to setting up the result that influences of Paeonol quantitative correction model
Can know that by table 12 when adopting polynary scatter correction method to carry out pre-service, the Paeonol quantitative correction model performance that obtains is better.
1.3.6.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up Paeonol quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 13, table 13 is that different-waveband is to setting up the result that influences of Paeonol quantitative correction model.
Table 13 different-waveband is to setting up the result that influences of Paeonol quantitative correction model
Can know that by table 13 when wave band was 1300nm～2300nm, the Paeonol quantitative correction model performance that obtains was better.
1.3.6.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up Paeonol quantitative correction model is 11.
1.3.6.4 the foundation of quantitative correction model
Adopt polynary scatter correction method that 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample are carried out pre-service; Laboratory values according to the Paeonol that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up Paeonol quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1300nm～2300nm; Number of principal components is 11, and the result is referring to Fig. 6, and Fig. 6 is the Paeonol quantitative correction model of embodiment of the invention foundation.
1.3.6.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the Paeonol quantitative correction model that utilizes 1.3.6.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
1.3.7 the foundation of amarogentin quantitative correction model
Use the TheUnscrambler quantitative analysis software; Near infrared spectrum data to the sample that obtains in 1.2 is carried out pre-service; The laboratory values of the amarogentin that obtains among the result and 1.1 who obtains according to this pre-service uses the TheUnscrambler quantitative analysis software to set up amarogentin quantitative correction model with PLS and cross validation method.
1.3.7.1 confirming of preprocess method
Use the TheUnscrambler quantitative analysis software; Adopt polynary scatter correction method, standard normalization method and 9 smoothing methods of first order differential that the correcting sample that obtains in 1.2 and the near infrared spectrum data of verification sample are carried out pre-service respectively, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best preprocess method, the result is referring to table 14, table 14 is that preprocess method is to setting up the result that influences of amarogentin quantitative correction model.
Table 14 preprocess method is to setting up the result that influences of amarogentin quantitative correction model
Can know that by table 14 when adopting the standard normalization method to carry out pre-service, the amarogentin quantitative correction model performance that obtains is better.
1.3.7.2 confirming of spectral range
Select 1100nm～2300nm, 1100nm～2100nm, 1100nm～1900nm and four wave bands of 1300nm～2300nm respectively, adopt minimum square law partially to set up amarogentin quantitative correction model, with the internal chiasma coefficient of determination (R
2), internal chiasma checking mean square deviation (RMSEC) is evaluation index with checking root-mean-square-deviation (RMSEP), confirms best band, the result is referring to table 15, table 15 is that different-waveband is to setting up the result that influences of amarogentin quantitative correction model.
Table 15 different-waveband is to setting up the result that influences of amarogentin quantitative correction model
Can know that by table 15 when wave band was 1100nm～2300nm, the amarogentin quantitative correction model performance that obtains was better.
1.3.7.3 confirming of number of principal components
The present invention is an evaluation index with internal chiasma checking mean square deviation error (RMSECV), confirms best number of principal components, and the result shows that the number of principal components of setting up amarogentin quantitative correction model is 6.
1.3.7.4 the foundation of quantitative correction model
Employing standard normalization method is carried out pre-service to 1.2 correcting samples that obtain and the near infrared spectrum data of verification sample; Laboratory values according to the amarogentin that obtains in said pre-service result and 1.1; Adopt PLS and cross validation method; Set up amarogentin quantitative correction model with the TheUnscrambler quantitative analysis software, wherein, the wave band during modeling is 1100nm～2300nm; Number of principal components is 6, and the result is referring to Fig. 7, and Fig. 7 is the amarogentin quantitative correction model of embodiment of the invention foundation.
1.3.7.5 quantitative correction verification of model
From 59 samples, randomly draw 6 samples, the amarogentin quantitative correction model that utilizes 1.3.7.4 to set up is analyzed, and the result is referring to table 16, and table 16 is the analysis result of each quantitative correction model to GUIZHI FULING JIAONANG.
Each quantitative correction model of table 16 is to the analysis result of GUIZHI FULING JIAONANG
Can know by table 16; The relative average debiation of gallic acid is 1.627%; The relative average debiation of Paeoniflorin is 1.550%, benzoic relative average debiation is 1.012%, the relative average debiation of benzoylpaeoniflorin is 2.600%, the relative average debiation of cinnaldehydrum is 2.398%, the relative average debiation of Paeonol is 2.109%, the relative average debiation of amarogentin is 1.089%; This shows; Above-mentioned each quantitative correction model can accurately detect the principal ingredient in the GUIZHI FULING JIAONANG, thereby realizes the accurate detection to GUIZHI FULING JIAONANG.
1.5 the detection of GUIZHI FULING JIAONANG
In 59 samples, select a sample at random; Each quantitative correction model with above-mentioned foundation detects it, and parallel detection 6 times is averaged as predicted value; The result is referring to table 17, and table 17 is the testing result of method provided by the invention to GUIZHI FULING JIAONANG.
Table 17 method provided by the invention is to the testing result of GUIZHI FULING JIAONANG
Can know that by table 17 method provided by the invention can realize the detection to GUIZHI FULING JIAONANG, and testing result is accurate, precision is good.
The above only is a preferred implementation of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; Can also make some improvement and retouching, these improvement and retouching also should be regarded as protection scope of the present invention.
Claims (16)
1. the detection method of a GUIZHI FULING JIAONANG may further comprise the steps:
With the pressed powder in the GUIZHI FULING JIAONANG is testing sample, utilizes near infrared spectrometer to detect said testing sample, obtains the near infrared spectrum data of said testing sample;
According to the calibration model and the said near infrared spectrum data of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin, obtain the content of gallic acid, Paeoniflorin, benzoic acid, benzoylpaeoniflorin, cinnaldehydrum, Paeonol and amarogentin in the said testing sample.
2. detection method according to claim 1 is characterized in that, when utilizing near infrared spectrometer to detect said testing sample, the thickness of said testing sample is 1mm～3mm, and scanning times is 500 times～700 times.
3. detection method according to claim 1 is characterized in that, the calibration model of said gallic acid is set up according to following method:
A11) GUIZHI FULING JIAONANG pressed powder sample is provided;
A12) utilize high performance liquid chromatograph to detect the content of gallic acid in the said sample, obtain the laboratory values of gallic acid in the said sample;
A13) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A14) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of gallic acid according to the result of said pre-service acquisition and the laboratory values of said gallic acid.
4. detection method according to claim 3 is characterized in that, said step a14) in, said pretreated wave band is 1300nm～2300nm.
5. detection method according to claim 1 is characterized in that, the calibration model of said Paeoniflorin is set up according to following method:
A21) GUIZHI FULING JIAONANG pressed powder sample is provided;
A22) utilize high performance liquid chromatograph to detect content of paeoniflorin in the said sample, obtain the laboratory values of Paeoniflorin in the said sample;
A23) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A24) adopt polynary scatter correction method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of Paeoniflorin according to the result of said pre-service acquisition and the laboratory values of said Paeoniflorin.
6. detection method according to claim 5 is characterized in that, said step a24) in, said pretreated wave band is 1100nm～2100nm.
7. detection method according to claim 1 is characterized in that, said benzoic calibration model is set up according to following method:
A31) GUIZHI FULING JIAONANG pressed powder sample is provided;
A32) utilize high performance liquid chromatograph to detect benzoic content in the said sample, obtain benzoic laboratory values in the said sample;
A33) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A34) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up benzoic quantitative correction model according to result and the said benzoic laboratory values that said pre-service obtains.
8. detection method according to claim 7 is characterized in that, said step a34) in, said pretreated wave band is 1100nm～2300nm.
9. detection method according to claim 1 is characterized in that, the calibration model of said benzoylpaeoniflorin is set up according to following method:
A41) GUIZHI FULING JIAONANG pressed powder sample is provided;
A42) utilize high performance liquid chromatograph to detect the content of benzoylpaeoniflorin in the said sample, obtain the laboratory values of benzoylpaeoniflorin in the said sample;
A43) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A44) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service; According to the result of said pre-service acquisition and the laboratory values of said benzoylpaeoniflorin, adopt PLS and cross validation method to set up the quantitative correction model of benzoylpaeoniflorin.
10. detection method according to claim 9 is characterized in that, said step a44) in, said pretreated wave band is 1100nm～1900nm.
11. detection method according to claim 1 is characterized in that, the calibration model of said cinnaldehydrum is set up according to following method:
A51) GUIZHI FULING JIAONANG pressed powder sample is provided;
A52) utilize high performance liquid chromatograph to detect the content of cinnaldehydrum in the said sample, obtain the laboratory values of cinnaldehydrum in the said sample;
A53) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A54) adopt 9 smoothing methods of first order differential that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of cinnaldehydrum according to the result of said pre-service acquisition and the laboratory values of said cinnaldehydrum.
12. detection method according to claim 11 is characterized in that, said step a54) in, said pretreated wave band is 1100nm～1900nm.
13. detection method according to claim 1 is characterized in that, the calibration model of said Paeonol is set up according to following method:
A61) GUIZHI FULING JIAONANG pressed powder sample is provided;
A62) utilize high performance liquid chromatograph to detect the content of Paeonol in the said sample, obtain the laboratory values of Paeonol in the said sample;
A63) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A64) adopt polynary scatter correction method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of Paeonol according to the result of said pre-service acquisition and the laboratory values of said Paeonol.
14. detection method according to claim 13 is characterized in that, said step a64) in, said pretreated wave band is 1300nm～2300nm.
15. detection method according to claim 1 is characterized in that, the calibration model of said amarogentin is set up according to following method:
A71) GUIZHI FULING JIAONANG pressed powder sample is provided;
A72) utilize high performance liquid chromatograph to detect the content of amarogentin in the said sample, obtain the laboratory values of amarogentin in the said sample;
A73) near infrared spectrum data that adopts the diffuse reflection mode to gather said sample;
A74) adopt the standard normalization method that said near infrared spectrum data is carried out pre-service,, adopt PLS and cross validation method to set up the quantitative correction model of amarogentin according to the result of said pre-service acquisition and the laboratory values of said amarogentin.
16. detection method according to claim 15 is characterized in that, said step a74) in, said pretreated wave band is 1100nm～2300nm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101041005A CN102759509A (en) | 2011-04-25 | 2011-04-25 | Detection method of cassiabarktree twig tuckahoe capsules |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101041005A CN102759509A (en) | 2011-04-25 | 2011-04-25 | Detection method of cassiabarktree twig tuckahoe capsules |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102759509A true CN102759509A (en) | 2012-10-31 |
Family
ID=47054039
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011101041005A Pending CN102759509A (en) | 2011-04-25 | 2011-04-25 | Detection method of cassiabarktree twig tuckahoe capsules |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102759509A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104132909A (en) * | 2014-08-05 | 2014-11-05 | 华南农业大学 | Near infrared rapid determination method of gallic acid content in terminalia chebula retz |
| CN106153803A (en) * | 2016-07-22 | 2016-11-23 | 江苏康缘药业股份有限公司 | A kind of based on the method for active component dissolution in Rotating shaker detection GUIZHI FULING JIAONANG |
| CN109060709A (en) * | 2018-07-03 | 2018-12-21 | 东北农业大学 | A method of cow's milk heat treatment degree is detected based on infrared spectrum technology |
| CN111707760A (en) * | 2020-08-07 | 2020-09-25 | 曹薇薇 | Method for measuring contents of paeoniflorin, cinnamaldehyde and paeonol in traditional Chinese medicine preparation |
| CN113130033A (en) * | 2019-12-30 | 2021-07-16 | 江苏康缘药业股份有限公司 | Application of substance in preparation of product for treating primary dysmenorrhea |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6026314A (en) * | 1997-09-05 | 2000-02-15 | Samsung Electronics Co., Ltd. | Method and device for noninvasive measurements of concentrations of blood components |
| CN1556404A (en) * | 2004-01-06 | 2004-12-22 | 江苏康缘药业股份有限公司 | Quality control method of Chinese medicinal preparation |
| WO2005099404A2 (en) * | 2004-04-08 | 2005-10-27 | Aradigm Corporation | Method for assaying chemicals deposited by a chemical delivery system |
| CN1843424A (en) * | 2006-02-13 | 2006-10-11 | 浙江大德药业集团有限公司 | Cassia twig tuckahoe effervescence tablet and preparation method and its quality control method |
| CN101059426A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method for non-destructive measurement for tea polyphenol content of tea based on near infrared spectrum technology |
| CN101846617A (en) * | 2009-12-29 | 2010-09-29 | 中国科学院地球化学研究所 | Sterile detection method of cane sugar content in culture media based on spectrum analysis |
-
2011
- 2011-04-25 CN CN2011101041005A patent/CN102759509A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6026314A (en) * | 1997-09-05 | 2000-02-15 | Samsung Electronics Co., Ltd. | Method and device for noninvasive measurements of concentrations of blood components |
| CN1556404A (en) * | 2004-01-06 | 2004-12-22 | 江苏康缘药业股份有限公司 | Quality control method of Chinese medicinal preparation |
| WO2005099404A2 (en) * | 2004-04-08 | 2005-10-27 | Aradigm Corporation | Method for assaying chemicals deposited by a chemical delivery system |
| CN1843424A (en) * | 2006-02-13 | 2006-10-11 | 浙江大德药业集团有限公司 | Cassia twig tuckahoe effervescence tablet and preparation method and its quality control method |
| CN101059426A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method for non-destructive measurement for tea polyphenol content of tea based on near infrared spectrum technology |
| CN101846617A (en) * | 2009-12-29 | 2010-09-29 | 中国科学院地球化学研究所 | Sterile detection method of cane sugar content in culture media based on spectrum analysis |
Non-Patent Citations (8)
| Title |
|---|
| YAN-WEN WU ETAL: "Fourier transform mid-infrared (MIR) and near-infrared (NIR)spectroscopy for rapid quality assessment of Chinese medicine preparation Honghua Oil", 《JOURNAL OF PHARMACEUTICAL AND BIOMEDICAL ANALYSIS 》 * |
| 夏俊芳 等: "不同预处理对近红外光谱检测脐橙VC含量的影响", 《农业机械学报》 * |
| 宫凯敏 等: "近红外光谱快速分析法在桂枝茯苓胶囊中的应用", 《中国中药杂志》 * |
| 尼珍 等: "近红外光谱分析中光谱预处理方法的作用及其发展", 《药物分析杂志》 * |
| 杨南林 等: "一种用于中药纯化过程的近红外光谱分析新方法", 《化学学报》 * |
| 胡钢亮 等: "近红外光谱在线检测技术在中药领域中的应用展望", 《中国中药杂志》 * |
| 褚小立 等: "近红外分析中光谱预处理及波长选择方法进展与应用", 《化学进展》 * |
| 马婷婷: "桂枝茯苓胶囊质量控制方法研究", 《中国知网硕士学位论文数据库》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104132909A (en) * | 2014-08-05 | 2014-11-05 | 华南农业大学 | Near infrared rapid determination method of gallic acid content in terminalia chebula retz |
| CN106153803A (en) * | 2016-07-22 | 2016-11-23 | 江苏康缘药业股份有限公司 | A kind of based on the method for active component dissolution in Rotating shaker detection GUIZHI FULING JIAONANG |
| CN109060709A (en) * | 2018-07-03 | 2018-12-21 | 东北农业大学 | A method of cow's milk heat treatment degree is detected based on infrared spectrum technology |
| CN113130033A (en) * | 2019-12-30 | 2021-07-16 | 江苏康缘药业股份有限公司 | Application of substance in preparation of product for treating primary dysmenorrhea |
| CN111707760A (en) * | 2020-08-07 | 2020-09-25 | 曹薇薇 | Method for measuring contents of paeoniflorin, cinnamaldehyde and paeonol in traditional Chinese medicine preparation |
| CN111707760B (en) * | 2020-08-07 | 2022-09-02 | 曹薇薇 | Method for determining contents of paeoniflorin, cinnamaldehyde and paeonol in traditional Chinese medicine preparation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101303294B (en) | Application method of near-infrared on-line test technology in Chinese medicine Yiqing granule production | |
| CN102106939B (en) | Quality control method for extract concentrated liquor of condensed pills of six ingredients with rehmannia root | |
| CN102759509A (en) | Detection method of cassiabarktree twig tuckahoe capsules | |
| CN106138131B (en) | A kind of bupleurum particles and its Chinese medicine preparation | |
| CN103235082B (en) | Method for detecting Jingwu capsule | |
| CN103033486A (en) | Method for near infrared spectrum monitoring of quality of pericarpium citri reticulatae and citrus chachiensis hortorum medicinal materials | |
| CN102048777A (en) | Panax root extract detection method | |
| CN104568822A (en) | Method for rapidly and simultaneously detecting multiple indexes of fructus forsythia medicinal material | |
| CN102778516B (en) | Rheum officinale and eupolyphaga capsule finger-print spectrum identification method | |
| CN108663337B (en) | Method for measuring tanshinone component and application thereof | |
| CN105987966A (en) | Quality control method for Callicarpa kwangtungensis based on the spectrum-activity relationship and method for establishing model thereof | |
| CN104316613B (en) | A kind of method for building up of dispelling wind detoxicating capsule finger printing | |
| CN102147361A (en) | Near-infrared online detection method used in Chinese magnoliavine fruit processing production | |
| CN102048796A (en) | Rhodiola rosea extract testing method | |
| CN106074695A (en) | A kind of DANSHEN KELI and Chinese medicine preparation thereof | |
| CN102147360B (en) | Detection method for drug combination | |
| CN108051396A (en) | A kind of rapid detection method of Xin Ke Shu ' tablet for treating coronary heart disease active constituent content | |
| CN101721475A (en) | Quality control method for Yaotongning capsules | |
| CN102175629B (en) | Biological activity detection-based evaluation method of quality of prepared radix rehmanniae | |
| CN102106950B (en) | Quality control method in NuJin capsule extraction and concentration process | |
| CN109342356A (en) | The construction method and detection method of near-infrared quantitative calibration models in zhenqi fuzheng granules production technology | |
| CN102175640A (en) | Near-infrared online detecting method in herba epimedii processing production | |
| CN105158352B (en) | A kind of method for building up of finger-print of perhexiline piece and its application | |
| CN106074742A (en) | A kind of Radix Paeoniae Alba granule and Chinese medicine preparation thereof | |
| CN110031564A (en) | The quality determining method of natural plants anticoccidial feed addictive based on HPLC finger-print |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20121031 |