CN102243172A - Method for detecting limy materials in flour by intermediate infrared spectroscopy - Google Patents

Abstract

The invention discloses a method for detecting limy materials in flour by an intermediate infrared spectroscopy, which relates to the filed of food safety. The method comprises the following steps of: acquiring intermediate infrared spectrum of a flour sample to be detected; determining whether the flour sample to be detected contains the limy materials according to the intermediate infrared spectrum characteristics of the limy materials and the pure flour; acquiring the intermediate infrared spectrum of a standard series of flour sample; establishing a quantitative calibration model according to the infrared spectrum of the standard series of flour sample; and determining the content of the limy materials in the flour sample to be detected. The method for detecting the limy materials in the flour by the intermediate infrared spectroscopy is simple in steps and pollution-free, and can quickly detect whether the flour contains the limy materials. Moreover, various limy materials in the flour can be quickly and quantitatively determined at the same time through the quantitative calibration model according to the intermediate infrared spectrum of the standard series of flour sample.

Description

The method of lime class material in the mid-infrared light spectrometry detection faces powder

Technical field

The present invention relates to the food security technical field, the method for lime class material in particularly a kind of mid-infrared light spectrometry detection faces powder.

Background technology

Flour as large agricultural product is one of main source of the daily food of people.Whether not only the quality safety of flour direct relation people's diet health, and is the significant problem that involves the interests of the state and the people.Yet recent years, the flour quality safety problem enjoys the extensive concern of various circles of society.Except that the problem flour that whitening agent causes, add excessive lime class material and also be exposed serious harm people's safe diet and healthy when causing very big negative effect with the problem flour that obtains effects such as weightening finish.

Based on red, orange, green, blue, yellow (ROGBY), its defective is to separate and complicated pretreatment detected sample to the traditional detection mode of adjuvant in the flour, detects complex steps, length consuming time, and use chemical reagent in the testing process, cause environmental pollution, therefore restricted it and applied widely.

(Mid-Infrared Spectroscopy MIR) belongs to molecular spectrum to middle infrared spectrum, and its mechanism of production is a material centering infrared range absorption of electromagnetic wave.The absorption great majority in this spectrum district are that the fundamental frequency of functional group or chemical bond absorbs, and also comprise a spot of frequency multiplication and absorb.The middle infrared spectrum of pure material has that absorption intensity height, spectrum peak separate obviously, spectrum peak implication is easier to characteristics such as explanation.Advantages such as that the mid-infrared light Zymography has is highly sensitive, analysis speed fast, can realize nondestructive analysis, environmentally friendly, not only be widely used in the attributional analysis of agricultural product and food, but also be widely used in numerous areas such as commercial production, criminal investigation, legal medical expert's evaluation, synthetic compound sign.

The condition that produces according to middle infrared spectrum as can be known, some inorganicss as lime class material, have stronger characteristic absorption at middle infrared range.The characteristic absorption peak of calcium oxide, calcium hydroxide and lime carbonate has that spectrum peak position separates, characteristic obviously, be easier to characteristics such as explanation.This is that the mid-infrared light spectrometry is qualitative to this kind inorganics, the spectroscopy foundation of quantitative test, that is the mid-infrared light spectrometry has feasibility to the detection of inorganics.In addition, the application of Technique of Attenuated Total Reflectance makes mid-infrared spectral gatherer process become very simple, for hardware foundation has been established in the popularization of this technology.

Summary of the invention

(1) technical matters that will solve

The technical problem to be solved in the present invention is: the method how lime class material in simple, the free of contamination mid-infrared light spectrometry of a kind of step detection faces powder is provided.

(2) technical scheme

For solving the problems of the technologies described above, the invention provides the method for lime class material in a kind of mid-infrared light spectrometry detection faces powder, comprise step:

S100: the middle infrared spectrum of gathering flour sample to be measured;

S200:, judge in the described flour sample to be measured whether contain lime class material according to the mid-infrared spectral behavior of lime class material and pure flour.

Preferably, if contain lime class material in the described flour sample to be measured of the judgment result displays of described step S200, also comprise step after the then described step S200:

S300: the middle infrared spectrum of gathering standard series flour sample;

S400:, set up the quantitative correction model according to the middle infrared spectrum of described standard series flour sample;

S500:, measure the content of lime class material in the described flour sample to be measured according to described quantitative correction model.

Preferably, described lime class material comprises calcium oxide, calcium hydroxide and lime carbonate, and described pure flour is not for being mixed with the flour of any additives.

Preferably, standard series flour sample comprises a plurality of known flour samples among the described step S300, described known flour sample is the flour that is added with calcium oxide, calcium hydroxide and lime carbonate, and the content of described calcium oxide, calcium hydroxide and lime carbonate is mutually independent, the calcium oxide content of described a plurality of known flour samples evenly distributes in certain interval range, the calcium hydroxide content of described a plurality of known flour samples evenly distributes in certain interval range, and the calcium carbonate content of described a plurality of known flour samples evenly distributes in certain interval range.

Preferably, the sample size of described standard series flour sample is no less than 20.

Preferably, described step S400 comprises:

S401: adopt data normalization, data smoothing, data normalization or baseline correction algorithm that the middle infrared spectrum data of described standard series flour sample are carried out pre-service;

S402: adopt the offset minimum binary algorithm to set up described quantitative correction model in conjunction with full validation-cross algorithm.

Preferably, utilize the mid-infrared light spectrometer to gather the middle infrared spectrum of described flour sample to be measured and described standard series flour sample, and the signal to noise ratio (S/N ratio) of mid-infrared light spectrometer is not less than 10000.

(3) beneficial effect

The method of lime class material in the mid-infrared light spectrometry detection faces powder of the present invention, step is simple, pollution-free, can carry out the fast qualitative detection to whether containing lime class material in the flour.And the middle infrared spectrum and the quantitative correction model of combined standard series flour sample can carry out quantitative determination simultaneously to multiple lime class material in the flour.

Description of drawings

Fig. 1 is the method flow diagram of lime class material in the described mid-infrared light spectrometry of the embodiments of the invention detection faces powder;

Fig. 2 is the mid-infrared light spectrogram of the embodiment of the invention 1 described flour sample to be measured;

Fig. 3 is the predicted value and the true value corresponding relation figure of the described quantitative correction model based on calcium oxide content of the embodiment of the invention 2;

Fig. 4 is the predicted value and the true value corresponding relation figure of the described quantitative correction model based on calcium hydroxide content of the embodiment of the invention 2;

Fig. 5 is the predicted value and the true value corresponding relation figure of the described quantitative correction model based on calcium carbonate content of the embodiment of the invention 2.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples are used to illustrate the present invention, but are not used for limiting the scope of the invention.

Fig. 1 is the method flow diagram of lime class material in the described mid-infrared light spectrometry of the embodiments of the invention detection faces powder.As shown in Figure 1, the method for lime class material in the mid-infrared light spectrometry detection faces powder comprises step:

S100: gather the middle infrared spectrum of flour sample to be measured, the signal to noise ratio (S/N ratio) of the used mid-infrared light spectrometer of gatherer process is not less than 10000.

S200:, judge in the described flour sample to be measured whether contain lime class material according to the mid-infrared spectral behavior of lime class material and pure flour.Described lime class material comprises calcium oxide, calcium hydroxide and lime carbonate, and described pure flour is not for being mixed with the flour of any additives.

If contain lime class material in the described flour sample to be measured of the judgment result displays of described step S200, also comprise step after the described step S200:

S300: the middle infrared spectrum of gathering standard series flour sample.Described standard series flour sample comprises and is no less than 20 known flour samples, in described known flour sample, be added with calcium oxide, calcium hydroxide and lime carbonate, and described calcium oxide, there is not correlativity between the content of calcium hydroxide and lime carbonate, promptly the content of three kinds of lime class materials is mutually independent, the calcium oxide content of described a plurality of known flour samples evenly distributes in certain interval range, the calcium hydroxide content of described a plurality of known flour samples evenly distributes in certain interval range, and the calcium carbonate content of described a plurality of known flour samples evenly distributes in certain interval range.The signal to noise ratio (S/N ratio) of gathering the used mid-infrared light spectrometer of middle infrared spectrum of standard series flour sample is not less than 10000.

S400: adopt preprocessing algorithms such as data normalization, data smoothing, data normalization or baseline correction that the middle infrared spectrum data of described standard series flour sample are carried out pre-service, adopt the offset minimum binary algorithm to set up the quantitative correction model in conjunction with full validation-cross algorithm.After described quantitative correction modelling is finished, the quality and the estimated performance thereof of this model are estimated, evaluation procedure is as follows:

Adopt coefficient of determination R ², correction error root mean square RMSEC, validation-cross error mean square root RMSECV be as the parameter of evaluation model quality.The model evaluation standard is: R ²As far as possible near 100%, RMESC and the RMSECV as far as possible little and RMSECV/RMSEC of numerical value separately are no more than 5.0.Wherein, R ²Calculating as shown in Equation (1).

$R^2=(1-\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left({\mathrm{Differ}}_i\right)}^2}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(y_i-y_m)}^2})\×100\%---\left(1\right)$

Wherein: Differ _iBe illustrated in and set up in the quantitative correction model process the poor of the predicted value of lime class material and true value in i sample of standard series, y _iBe the true value of lime class material in i sample of standard series, y _mBe the mean value of lime class material true value in each sample of standard series, n is the sample size of standard series.Wherein, true value is meant the real content (percentage by weight, down together) of certain sample target lime class material (for example calcium hydroxide); Predicted value is meant the percentage composition value through the target lime class material (for example calcium hydroxide) in this sample that draws behind the model calculation.

RMSEC computing formula (2) is as follows:

$\mathrm{DMSEC}=\sqrt{\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left({\mathrm{Differ}}_i\right)}^2}{n}}---\left(2\right)$

RMSECV computing formula (3) is as follows:

$\mathrm{RMSECV}=\sqrt{\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left({\mathrm{Diff}}_i\right)}^2}{n}}---\left(3\right)$

Wherein, Diff _iBe illustrated in the validation-cross process the poor of the predicted value of lime class material and true value in i sample of standard series.

With prediction standard deviation SEP and the parameter of relative estimated performance RPD as the evaluation model estimated performance.

The computing formula (4) of prediction standard deviation SEP is as follows:

$\mathrm{SEP}=\sqrt{\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\left({\mathrm{Diff}}_i\right)}^2}{n-1}}---\left(4\right)$

Relatively the computing formula (5) of estimated performance RPD is as follows, and being not less than 3 with the RPD value, to be considered as model accuracy very high.

$\mathrm{RPD}=\frac{S_D}{\mathrm{SEP}}---\left(5\right)$

Wherein, S _DIt is the sample standard deviation of described standard series flour sample.

S500:, measure the content of lime class material in the described flour sample to be measured according to described quantitative correction model.

The method of lime class material in the mid-infrared light spectrometry detection faces powder of the present invention, step is simple, pollution-free, can carry out the fast qualitative detection to whether containing lime class material in the flour.And the middle infrared spectrum and the quantitative correction model of combined standard series flour sample can carry out quantitative determination simultaneously to the multiple lime class material in the flour.

Embodiment 1

Carry out qualitative detection to whether containing lime class material in a series of flour samples to be detected, its step is as follows:

S100: use signal to noise ratio (S/N ratio) to be not less than the middle infrared spectrum that 10000 mid-infrared light spectrometer is gathered flour sample to be measured.Fig. 2 is the mid-infrared light spectrogram of the embodiment of the invention 1 described flour sample to be measured.The horizontal ordinate of Fig. 2 is a wave number, and ordinate is a transmitance.As shown in Figure 2, the wave-number range of this flour sample is 4000cm ^-1～3000cm ^-1

S200:, judge whether contain lime class material in the described flour sample according to the mid-infrared spectral behavior of lime class material and pure flour.Wave number is at 3640cm among Fig. 2 ^-1Near absorption peak (indicating part in the frame of broken lines) is the characteristic absorption of lime class material, can qualitatively judge in view of the above in this series flour sample and contain lime class material.

Embodiment 2

It is as follows that calcium oxide in the flour sample, calcium hydroxide and lime carbonate are carried out the quantitative measurement process simultaneously.

S300: the middle infrared spectrum of gathering standard series flour sample.Calcium oxide content (percentage by weight in the included flour sample of described standard series flour sample, in the 0.3%-8.0% scope, evenly distribute down together), calcium hydroxide content evenly distributes in the 0.3%-9.0% scope, calcium carbonate content evenly distributes in the 0.3%-8.0% scope, and the sample size of standard series is 20.Wherein, do not have correlativity between the content of calcium oxide, calcium hydroxide and lime carbonate, promptly three's content is independently of one another.The content of calcium oxide, calcium hydroxide and lime carbonate sees table 1 for details.

The content of calcium oxide, calcium hydroxide and lime carbonate in table 1 standard series

Numbering	Calcium oxide (%)	Calcium hydroxide (%)	Lime carbonate (%)
				1	0.60	0.36	0.37
2	0.61	0.47	0.93
				3	1.06	0.67	1.83
4	0.39	1.41	0.58
				5	0.75	1.85	3.00
6	1.44	2.51	1.88
				7	2.12	2.84	3.45
8	3.44	2.92	3.62
				9	6.84	5.86	6.24
10	2.85	3.48	2.49
				11	5.44	5.33	6.63
12	0.43	1.70	1.50
				13	2.21	0.90	0.42
14	2.97	3.92	1.49
				15	3.54	3.77	0.59
16	0.95	5.89	1.65
				17	0.72	5.52	4.60
18	3.59	8.06	5.28
				19	5.16	8.99	1.58
20	7.16	5.56	7.77

The signal to noise ratio (S/N ratio) of gathering the required mid-infrared light spectrometer of middle infrared spectrum of standard series flour sample is not less than 10000.The spectra collection scope is 4000cm ^-1-400cm ^-1, scanning times is 20 times, resolution is 4cm ^-1

S400: at the quantitative correction model of calcium oxide, the middle infrared spectrum data preprocessing method that adopts is: the data normalization method is that data centerization, baseline correction method are 5 derivatives of single order, adopts the offset minimum binary algorithm in conjunction with the quantitative correction model (call calcium oxide model) of full validation-cross algorithm foundation based on calcium oxide content; Quantitative correction model at calcium hydroxide, the middle infrared spectrum data preprocessing method that adopts is: the data normalization method is that data centerization, baseline correction method are 5 derivatives of single order, adopts the offset minimum binary algorithm in conjunction with the quantitative correction model (call calcium hydroxide model) of full validation-cross algorithm foundation based on calcium hydroxide content; Quantitative correction model at lime carbonate, the middle infrared spectrum data preprocessing method that adopts is: the data normalization method is that data normalization, data smoothing method are that the SG-19 point is level and smooth, and the full validation-cross algorithm of employing offset minimum binary algorithm combination is set up the quantitative correction model (calling the lime carbonate model in the following text) based on calcium carbonate content.

The coefficient of determination R of calcium oxide model ²=99.67%, correction error root mean square RMSEC=0.12, validation-cross error mean square root RMSECV=0.47, prediction standard deviation SEP=0.48, estimated performance RPD=4.49 relatively; The coefficient of determination R of calcium hydroxide model ²=99.60%, correction error root mean square RMSEC=0.15, validation-cross error mean square root RMSECV=0.42, prediction standard deviation SEP=0.43, estimated performance RPD=5.78 relatively; The coefficient of determination R of lime carbonate model ²=99.00%, correction error root mean square RMSEC=0.21, validation-cross error mean square root RMSECV=0.32, prediction standard deviation SEP=0.33, estimated performance RPD=6.74 relatively.Fig. 3 is the predicted value and the true value corresponding relation figure of the described quantitative correction model based on calcium oxide content of the embodiment of the invention 2; Fig. 4 is the predicted value and the true value corresponding relation figure of the described quantitative correction model based on calcium hydroxide content of the embodiment of the invention 2; Fig. 5 is the predicted value and the true value corresponding relation figure of the described quantitative correction model based on calcium carbonate content of the embodiment of the invention 2.As Fig. 3, Fig. 4 and shown in Figure 5, the predicted value of calcium oxide, calcium hydroxide and calcium carbonate content and true value have very high degree of correlation, this shows that calcium oxide model, calcium hydroxide model and lime carbonate model have sufficiently high precision, can be used for the quantitative measurement to flour sample lime class material to be measured.

S500: as flour sample to be measured,, measure the content of lime class material in the external certificate collection flour sample according to described calcium oxide model, calcium hydroxide model and lime carbonate model with the external certificate collection.In the external certificate collection sample content of lime class material when it prepares by the analytical balance accurate recording.With external certificate collection flour sample collection middle infrared spectrum, in spectroscopic data difference substitution calcium oxide model, calcium hydroxide model and lime carbonate model, calculate the predicted value of calcium oxide, calcium hydroxide, lime carbonate respectively.The contrast of the actual value of external certificate collection sample (the weighing value by analytical balance obtains), predicted value sees table 2 for details, and wherein, deviation is meant that predicted value deducts the poor of actual value.

Actual value, predicted value and the deviation of table 2 external certificate collection sample calcium oxide, calcium hydroxide, lime carbonate

The data presentation of table 2, adopt the difference (deviation) of the resulting predicted value of calcium oxide, calcium hydroxide and calcium carbonate content in calcium oxide model, calcium hydroxide model, the lime carbonate model prediction external certificate collection sample and its actual value separately very little respectively, illustrate that the model prediction result has very high precision, can be used as the final measured value of calcium oxide in the unknown sample, calcium hydroxide, lime carbonate.

The described detection method of the embodiment of the invention also is applicable to fast qualitative, the detection by quantitative of the lime class material in the agricultural product such as ground rice, tapioca starch, corn flour.

Above embodiment only is used to illustrate the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; under the situation that does not break away from the spirit and scope of the present invention; can also make various variations and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.

Claims (7)

1. the method for lime class material in the mid-infrared light spectrometry detection faces powder is characterized in that, comprises step:

S100: the middle infrared spectrum of gathering flour sample to be measured;

S200:, judge in the described flour sample to be measured whether contain lime class material according to the mid-infrared spectral behavior of lime class material and pure flour.

2. the method for claim 1 is characterized in that, if contain lime class material in the described flour sample to be measured of the judgment result displays of described step S200, also comprises step after the then described step S200:

S300: the middle infrared spectrum of gathering standard series flour sample;

S400:, set up the quantitative correction model according to the middle infrared spectrum of described standard series flour sample;

S500:, measure the content of lime class material in the described flour sample to be measured according to described quantitative correction model.

3. method as claimed in claim 2 is characterized in that, described lime class material comprises calcium oxide, calcium hydroxide and lime carbonate, and described pure flour is not for being mixed with the flour of any additives.

4. method as claimed in claim 3, it is characterized in that, standard series flour sample comprises a plurality of known flour samples among the described step S300, described known flour sample is for being added with calcium oxide, the flour of calcium hydroxide and lime carbonate, and described calcium oxide, the content of calcium hydroxide and lime carbonate is mutually independent, the calcium oxide content of described a plurality of known flour samples evenly distributes in certain interval range, the calcium hydroxide content of described a plurality of known flour samples evenly distributes in certain interval range, and the calcium carbonate content of described a plurality of known flour samples evenly distributes in certain interval range.

5. method as claimed in claim 4 is characterized in that, the sample size of described standard series flour sample is no less than 20.

6. method as claimed in claim 2 is characterized in that, described step S400 comprises:

S401: adopt data normalization, data smoothing, data normalization or baseline correction algorithm that the middle infrared spectrum data of described standard series flour sample are carried out pre-service;

S402: adopt the offset minimum binary algorithm to set up described quantitative correction model in conjunction with full validation-cross algorithm.

7. as each described method in the claim 2～6, it is characterized in that, utilize the mid-infrared light spectrometer to gather the middle infrared spectrum of described flour sample to be measured and described standard series flour sample, and the signal to noise ratio (S/N ratio) of mid-infrared light spectrometer is not less than 10000.

Images