CN108181262A - Method for rapidly determining content of sargassum horneri cellulose by utilizing near infrared spectrum technology - Google Patents

Abstract

A method for rapidly determining the cellulose content of copper algae using near-infrared spectroscopy technology comprises the following steps: the first step is to collect and pretreat copper algae samples; the second step is to determine the cellulose content of copper algae samples by using an improved sulfuric acid and potassium dichromate oxidation method; the third step is to obtain the near-infrared spectrum of the copper algae samples by scanning with a near-infrared spectrometer; the fourth step is to establish and evaluate a near-infrared spectroscopy quantitative analysis model; the fifth step is to apply the near-infrared spectroscopy quantitative analysis model. The method of the invention not only has the advantages of being fast, accurate, and environmentally friendly, but is also conducive to improving the quality control level of copper algae cellulose content, and can also be promoted and applied to the quality control of other seaweed biomass.

Description

A method for rapid determination of cellulose content in copper algae using near-infrared spectroscopy

technical field

The invention belongs to the field of cellulose content analysis, and in particular relates to a method for rapidly determining the cellulose content of copper algae based on near-infrared spectroscopy.

Background technique

Marine biomass resources represented by macroalgae have the advantage of not occupying land and freshwater resources, and have great development potential. Copper algae (Sargassum horneri (Turn.) Ag. commonly known as "Lilac House", belonging to the genus Sargassum (Sargassum), distributed in the coastal shallow sea area of China), hijiki and asparagus and other large seaweed plants are tall, with luxuriant branches and leaves. Known as the "sea forest", it is an ideal place for marine organisms to avoid enemies, feed, and lay eggs, and it is easy to scale artificial cultivation and environmental restoration (absorb nitrogen, phosphorus and heavy metals in the environment, release oxygen, adjust the pH of the water body, fix The ability of _CO2 is outstanding, and it is listed as one of the important species for rebuilding seabed algae farms and implementing marine ecological restoration. Copper algae (Sargassum horneri (Turn.) has fast growth speed, large output, single nature, tall plants, easy harvesting, and raw material collection and logistics costs are low, and can provide biomass raw materials on a large scale and stably. However, the taste of high-cellulose macroalgae is poor, and the added value of developing edible products is not high, which hinders the further expansion of its breeding scale; Growth and stay in the sea area decline, not only the problems such as eutrophication and heavy metal pollution in the sea area cannot be solved, but may even cause negative problems such as the large-scale proliferation of Enteromorpha. Therefore, it is of great importance to develop the high value-added conversion and utilization technology of copper algae biomass Practical significance.

Cellulose mainly exists in plant cell walls, and its content in different seaweed plants ranges from 1% to 40%. It is the most valuable component of biomass raw materials. The content of cellulose components in biomass feedstock has a significant impact on the production process of both biomass fuels and biomass-based chemicals. Therefore, quantitative analysis of the composition of biomass raw materials, especially the cellulose content, is of great significance for the precise proportioning of raw materials and the improvement of product yield and quality.

At present, the analysis of plant cellulose content is mainly based on textile standard GB/T5889-1986 and papermaking standard GB/T2677.10-1995. If the above-mentioned national standard is used to measure the cellulose content, not only the measurement takes up to 2 to 3 days, but also because the measurement object of the national standard method embodiment is ramie and wood, it is not necessarily applicable to the determination of seaweed cellulose. In the actual measurement process It is also found that the measurement results of seaweed cellulose content obtained by this method have great errors. Other available methods include Van Soest and its improved method and chromatographic method, wherein, Van Soest and its improved method can simultaneously measure the contents of cellulose content, moisture, hemicellulose, lignin and ash, but the determination steps are cumbersome and time-consuming. If gas chromatography, high performance liquid chromatography and other chromatographic methods are used, although the measurement results are relatively accurate, there are disadvantages of high cost and too long time consumption.

It can be seen that the traditional quantitative analysis methods of seaweed cellulose content generally have the disadvantages of cumbersome determination steps and long time-consuming, and the chromatography method has the disadvantage of high testing cost, which limits the promotion and application of these methods in the actual production process. Therefore, there is an urgent need to develop a quantitative analysis method for the cellulose content of seaweed biomass with simple analysis procedures, short time consumption and low cost.

Since the hydrogen-containing groups (C-H, N-H, O-H) in seaweed have obvious differences in the absorption of electromagnetic waves (near-infrared region) in the range of 700-2500nm in different chemical environments, the near-infrared spectrum contains hydrogen-containing organic substances. Rich structure and composition information. Near-infrared spectroscopy has disadvantages such as weak absorption intensity, spectral bandwidth and serious overlap, and weak characteristics. However, with the combination of near-infrared spectroscopy and chemometrics and the derivation of near-infrared spectroscopy analysis technology, it can be quickly and effectively established by establishing a mathematical model. Identifying the target information contained in the near-infrared spectrum has the advantages of non-destructive analysis, fast analysis, easy operation, accurate results and online analysis. In summary, near-infrared spectroscopy can be used to quickly determine the cellulose content of seaweed.

Contents of the invention

In view of the shortcomings of traditional methods for quantitative analysis of cellulose content in seaweed, the purpose of the present invention is to provide a method for quickly measuring the cellulose content of copper algae using near-infrared spectroscopy, so that the cellulose content in the copper algae biomass can be rapidly analyzed. , simple, cheap and accurate analysis.

The technical scheme step that the present invention adopts is as follows:

A kind of method utilizing near-infrared spectroscopic technology to measure the cellulose content of copper algae rapidly, comprises the following steps:

The first step is the collection and pretreatment of copper algae samples;

In the second step, the cellulose content of the copper algae sample was determined by the improved sulfuric acid and potassium dichromate oxidation method;

In the third step, the near-infrared spectrum of the copper algae sample is scanned by a near-infrared spectrometer;

The fourth step is the establishment and evaluation of the quantitative analysis model of near-infrared spectroscopy;

Import the cellulose content data obtained in the second step and the near-infrared spectrum data obtained in the third step into the numerical calculation software matlab 8.3, and adopt methods including abnormal sample elimination method, sample set division method, spectral preprocessing method, characteristic band selection method and A variety of chemometric methods, including the multivariate calibration method, were used to establish a quantitative analysis model for the cellulose content of copper algae, and the model performance was quantitatively evaluated using model evaluation parameters.

The fifth step is the application of the quantitative analysis model of near-infrared spectroscopy;

The established near-infrared spectroscopy quantitative analysis model was used to predict the cellulose content of unknown copper algae samples.

Further, in the first step, the copper algae sample pretreatment process includes: washing with water, air drying, drying, pulverizing with a pulverizer, and sieving with 60 mesh, and then putting it into a sealed transparent bag.

Further, in the second step, the step of improving sulfuric acid and potassium dichromate oxidation method to measure the cellulose content of each copper algae sample is:

2.1) put the copper algae sample into the Erlenmeyer flask containing the mixture of glacial acetic acid and nitric acid, and heat it in a boiling water bath;

2.2) After heating, cool to room temperature, filter, wash, and precipitate, place all the precipitates in an Erlenmeyer flask containing a mixture of sulfuric acid and potassium dichromate, and heat in a boiling water bath;

2.3) After heating, cool to room temperature, add potassium iodide solution and starch solution, titrate with sodium thiosulfate, and carry out blank control test simultaneously, calculate the cellulose content of copper algae sample according to the volume of consumed sodium thiosulfate solution.

Furthermore, in the third step, the near-infrared spectrum collection conditions are: the spectrum is collected in diffuse reflectance mode, the scanning wavenumber range of the spectrometer is 4000cm ^-1 ~ 12000cm ^-1 , and the resolution is 8cm ^-1 .

In the fourth step, the abnormal sample elimination method includes the Mahalanobis distance method, the t-test method, the spectral residual analysis method and the combination of the above methods.

In the fourth step, the sample set division method includes Kennard-Stone sample set division method, random sample selection method, SPXY method, elimination method and condensation method.

In the fourth step, the spectral preprocessing method includes a smoothing denoising algorithm, a derivative processing method, a standard normal variable transformation method, a multivariate scattering correction method, a Normalization normalization method, and a combination of the above methods.

In the fourth step, the characteristic band selection method includes interval partial least square method, moving window partial least square method, Monte Carlo method, correlation coefficient method, continuous projection method, genetic algorithm and the combination of the above methods.

In the fourth step, the multivariate correction method includes principal component regression and partial least squares method.

In the fourth step, the model evaluation parameters include calibration set leave-one-interaction verification standard deviation RMSECV, prediction correlation coefficient R, prediction standard deviation SEP and relative analysis error RPD.

The beneficial effects of the present invention are: the method for quickly measuring the cellulose content of copper algae by using near-infrared spectroscopy not only has the advantages of fastness, accuracy, and environmental protection, but also helps to improve the quality control level of the copper algae cellulose content, and can also be popularized It is applied to the quality control of other seaweed biomass.

Description of drawings

Fig. 1 is the cellulose content data distribution of 40 copper algae samples;

Fig. 2 is the original near-infrared spectrum of 40 copper algae samples;

Fig. 3 is the near-infrared spectrum residual analysis result of 40 copper algae samples;

Fig. 4 is the near-infrared spectrum characteristic band of the copper algae sample used to establish the quantitative analysis model (the original near-infrared spectrum is preprocessed by the Savitzky-Golay convolution second-order derivation algorithm);

Figure 5 is the prediction effect (validation set) of the near-infrared spectroscopy quantitative analysis model for the cellulose content of copper algae.

Specific implementation method

The implementation of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments.

Referring to Figures 1 to 5, a method for rapidly determining the cellulose content of copper algae using near-infrared spectroscopy comprises the following steps:

The first step is to collect samples of copper algae from different sources or batches, and perform necessary sample pretreatment.

The copper algae samples collected in the preferred embodiment of the present invention come from the Wenzhou sea area in southern Zhejiang, and a total of 40 copper algae samples were collected. After the sample was washed with water to remove the attached sediment and salt, it was placed in the sun to dry briefly, and then dried at 105°C. The washed and dried copper algae sample was crushed with a pulverizer and passed through a multi-layer sieve. , Take a 60 mesh sample and put it into a vacuum sealed transparent bag.

In the second step, the cellulose content of copper alga samples was determined by the improved oxidation method of sulfuric acid and potassium dichromate.

The detailed implementation steps of improving the oxidation method of sulfuric acid and potassium dichromate to determine the cellulose content of copper algae samples are as follows: crush the copper algae and pass through a 60-mesh sieve, weigh 0.2g (±0.0001g) copper algae particles and place them in a 100ml Erlenmeyer flask Add 5ml of a mixture of glacial acetic acid and nitric acid (volume ratio 1:1), put a glass stopper on it and place it in a boiling water bath to heat for 25 minutes with constant stirring; take it out and cool it to room temperature, then filter, discard the filtrate, and collect all The precipitate was washed 3 times with distilled water; the precipitate was placed in a 100ml Erlenmeyer flask, and 10ml of sulfuric acid solution with a mass fraction of 10% and 10ml of 0.1mol/L potassium dichromate solution were added to the precipitate, shaken and placed in a boiling Heat in a water bath for 10 minutes; add 10ml of distilled water, and after the solution is cooled to room temperature, add 5ml of potassium iodide solution with a mass fraction of 20% and 1ml of a starch solution with a mass fraction of 0.5%, and shake it well with 0.2mol/L sodium thiosulfate Titrate, and mix 10ml of 0.1mol/L potassium dichromate solution with 10ml of 10% sulfuric acid solution as a blank sample for titration. The formula for calculating the cellulose content of copper algae is:

In the formula, K represents the concentration of sodium thiosulfate solution, mol/L; a represents the volume of sodium thiosulfate solution consumed by blank titration, ml; b represents the volume of sodium thiosulfate solution consumed by the solution, ml, n represents the mass of copper algae particles, g.

The cellulose content distribution of 40 copper algae samples is shown in Figure 1, and the statistical analysis results of the overall data are shown in Table 1.

Table 1

In the third step, the near-infrared spectrum of the copper algae sample was scanned by a near-infrared spectrometer.

The near-infrared spectrum data of the copper algae sample was collected by a Nicolet iS10 Fourier transform near-infrared spectrometer (Thermo Fisher Scientific Corporation of the United States). spectrum. When scanning, the ambient temperature is kept constant (5-25°C), the humidity is lower than 25%, the copper algae sample is uniform in particle size and fully dry, and the instrument is set to automatically collect the background ^spectrum . The average spectrum is automatically saved and the background spectrum is deducted for the second time, with a resolution of 8cm ^-1 . Take an appropriate amount of copper algae sample into the rotating sample pool, fill it up and scrape it flat, and place it on the collection window of the integrating sphere to collect spectra. The copper algae sample was loaded and scanned 3 times, and the average value of the 3 scan spectra was taken as the original spectrum of the copper algae sample. The scanning results of 40 copper algae samples are shown in Fig. 2.

The fourth step is the establishment and evaluation of the quantitative analysis model of near-infrared spectroscopy.

Import the cellulose content data measured in the second step and the near-infrared spectral data obtained in the third step into the numerical calculation software matlab 8.3, and use the spectral residual analysis method to eliminate the abnormal data in the copper algae sample. The calculation formula of the spectral residual is: :

R = Y _pre - _Yylation

In the _formula , Y _and Y represent the predicted value matrix of the copper algae cellulose content in the prediction set and the wet chemical analysis data matrix of the copper algae cellulose content, R represents the residual matrix of the copper algae cellulose content in the prediction set, r _i Indicates the spectral residual value of the i-th sample in R, and f is the number of principal factors of the PLS prediction model. The results of abnormal data elimination are shown in Figure 3. From Figure 3, it can be seen that the data of two copper algae samples need to be eliminated.

From the 38 copper algae sample data after removing abnormal samples, 4 were randomly selected as unknown copper algae samples, and the remaining 34 copper algae samples were divided into a calibration set and a verification set by the Kennard-Stone sample set division method.

The specific implementation steps of the Kennard-Stone sample set division method are as follows:

(1) Calculate the Euclidean distance d _ij between all collected samples, and select the two samples with the largest Euclidean distance (ie sample No. 1 and sample No. 2) to enter the calibration set.

(2) Calculate the Euclidean distance between each of the remaining 34 samples and the two selected samples No. 1 and No. 2, and take the minimum value min(d _{i, No. 1} , d _{i, No. 2} ) , and then select sample No. 3 with the largest Euclidean distance value max(min(d _{i, No. 1} , d _{i, No. 2} )) to enter the calibration set.

(3) Calculate the Euclidean distance between each of the remaining 33 samples and the three selected samples No. 1, No. 2 and No. 3, and take the minimum value min(d _{i, No. 1} , d _{i, No. 2} , d _{i, No. 3} ), and then select sample No. 4 with the largest Euclidean distance value max(min(d _{i, No. 1} , d _{i, No. 2} , d _{i, No. 3} )) to enter the calibration set.

(4) Repeat the above process until 22 calibration samples are selected.

The statistical results of the calibration set and validation set are shown in Table 2.

Table 2

According to the measured value of cellulose content and near-infrared spectrum data of copper algae samples in the calibration set, the near-infrared spectral quantitative analysis model of copper algae cellulose content was established by using the multivariate calibration method. The number is 3.

The verification set is used for external verification of the near-infrared spectrum quantitative analysis model. The preferred spectral preprocessing method is the Savitzky-Golay convolution second-order derivation algorithm, the difference width is 5, and the polynomial fitting order is 3; the preferred feature band selection The method is the interval partial least squares method, and the selected characteristic bands are from 6883cm ^-1 to 10826cm ^-1 , and the selection results of the characteristic bands are shown in Fig. 4 .

For the verification set, the prediction effect of the infrared spectroscopy quantitative analysis model for the cellulose content of copper algae is shown in Figure 5.

Select calibration set leave-one-out interactive verification standard deviation RMSECV, prediction correlation coefficient R, prediction standard deviation SEP and relative analysis error RPD and other model evaluation parameters to evaluate the performance of the near-infrared spectroscopy quantitative analysis model. The specific calculation formula of each model evaluation parameter is shown below .

Leave-one-out cross-validation standard deviation (RMSECV):

In the formula, y _i,actual represents the chemical analysis value of the cellulose content of the i-th copper algae sample in the calibration set, y _i,predicted represents the model predicted value of the cellulose content of the i-th copper algae sample in the calibration set, and n represents the calibration The total number of samples in the set. The larger the standard deviation value, the greater the possibility of abnormal data in the calibration set.

Correlation coefficient (R):

In the formula, y _i,actual represents the chemical analysis value of the cellulose content of the i-th copper algae sample, Indicates the average value of the chemical analysis value of copper algae cellulose content, y _i,predicted indicates the predicted value of the cellulose content model of the i-th copper algae sample in the calibration set or validation set, and n indicates the total number of samples.

Standard Deviation of Prediction (SEP):

In the formula, y _i,actual represents the chemical analysis value of the cellulose content of the i-th copper algae sample, y _i,predicted represents the predicted value of the cellulose content model of the i-th copper algae sample in the verification set, and m represents the total number of samples in the calibration set . The closer the prediction standard deviation is to zero, the more accurate the model's predictions are.

Relative Analytical Error (RPD):

In the formula, SD _V represents the standard deviation of the cellulose content of all copper algae samples in the validation set. The wider and more uniform the property distribution of the validation set samples, the smaller the SEP and the larger the RPD value.

For the verification set, the model evaluation parameter results of the near-infrared spectroscopy quantitative analysis model for cellulose content in copper algae are shown in Table 3.

table 3

It can be seen from Table 3 that the standard deviation of leave-one-out interactive verification (RMSECV) is 1.0097, and the standard deviation of prediction (SEP) is 1.0288. The deviation from zero is relatively small, indicating that the model after removing abnormal samples not only has better stability, Moreover, for the verification set, the deviation between the predicted cellulose content of copper algae samples and the actual value is small, the correlation coefficient (R) is 0.9404, which is very close to 1, and the relative analysis error (RPD) is 2.94 greater than 2, indicating that the overall prediction effect of the model is relatively good. it is good.

The fifth step is the application of the quantitative analysis model of near-infrared spectroscopy.

Using the near-infrared spectroscopy quantitative analysis model of copper algae cellulose content established in the fourth step, the cellulose content of four copper algae samples was predicted, and the model evaluation parameter results were given. The prediction results of cellulose content of unknown copper algae samples are shown in Table 4, and the corresponding model evaluation parameter results are shown in Table 5. It can be seen from Table 4 and Table 5 that the practical application of the near-infrared spectroscopy quantitative analysis model for cellulose content in copper algae has been successful.

Table 4

table 5

Finally, it should be noted that the above preferred embodiments are only implementation modes of the present invention for easier understanding, rather than limiting the present invention. Although the present invention has been described in detail through the above-mentioned preferred embodiments, any person skilled in the art to which the present invention belongs should understand that any modification and change can be made in the form and details of implementation without departing from the present invention. The scope defined by the claims.

Claims (10)

1. a method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content, is characterized in that, described method comprises the following steps: The first step is the collection and pretreatment of copper algae samples; In the second step, the cellulose content of the copper algae sample was determined by the improved sulfuric acid and potassium dichromate oxidation method; In the third step, the near-infrared spectrum of the copper algae sample is scanned by a near-infrared spectrometer; The fourth step is the establishment and evaluation of the quantitative analysis model of near-infrared spectroscopy; Import the cellulose content data obtained in the second step and the near-infrared spectrum data obtained in the third step into the numerical calculation software matlab 8.3, and use various chemometric methods to establish a near-infrared spectrum quantitative analysis model for the cellulose content of copper algae, and Select the appropriate model evaluation parameters to evaluate the model; The fifth step is the application of the quantitative analysis model of near-infrared spectroscopy. The established near-infrared spectroscopy quantitative analysis model was used to predict the cellulose content of unknown copper algae samples. 2. a kind of method utilizing near-infrared spectroscopy to quickly measure copper algae cellulose content according to claim 1, is characterized in that, in the first step, described copper algae sample pretreatment process is: washing, Air-dried, oven-dried, pulverized by a pulverizer and sieved with a 60-mesh sieve, put into a sealed transparent bag. 3. according to claim 1 and 2, a kind of method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content is characterized in that, in the second step, the improved sulfuric acid and potassium dichromate oxidation The steps of measuring the cellulose content of each copper algae sample by method are: 2.1) put the copper algae sample into the Erlenmeyer flask containing the mixture of glacial acetic acid and nitric acid, and heat it in a boiling water bath; 2.2) After heating, cool to room temperature, filter, wash, and precipitate, place all the precipitates in an Erlenmeyer flask containing a mixture of sulfuric acid and potassium dichromate, and heat in a boiling water bath; 2.3) After heating, cool to room temperature, add potassium iodide solution and starch solution, titrate with sodium thiosulfate, and carry out blank control test simultaneously, calculate the cellulose content of copper algae sample according to the volume of consumed sodium thiosulfate solution. 4. according to claim 1 and 2 described a kind of method utilizing near-infrared spectrum technology to measure copper algae cellulose content rapidly, it is characterized in that, in the described 3rd step, described near-infrared spectrum acquisition condition is: in The spectrum was collected in the diffuse reflectance mode, and the scanning wavenumber of the spectrometer ranged from 4000cm ^-1 to 12000cm ^-1 , with a resolution of 8cm ^-1 . 5. according to claim 1 or 2, a kind of method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content is characterized in that, in the fourth step, the chemometric method includes abnormal sample removal method, sample set division method, spectral preprocessing method, characteristic band selection method and multivariate correction method. 6. according to claim 1 or 2, a kind of method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content is characterized in that, in the fourth step, the model evaluation parameters include correction set Cross Validated Standard Deviation RMSECV, Predicted Correlation Coefficient R, Predicted Standard Deviation SEP, and Relative Analytical Error RPD. 7. a kind of method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content according to claim 5, is characterized in that, described abnormal sample rejecting method comprises Mahalanobis distance method, t test method, spectral residual analysis methods and combinations of the above methods. 8. a kind of method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content according to claim 5, is characterized in that, described sample set division method comprises Kennard-Stone sample set division method, randomly selected sample method, SPXY method, elimination method and condensation method. 9. a kind of method utilizing near-infrared spectroscopy to quickly measure copper algal cellulose content according to claim 5, is characterized in that, described spectrum preprocessing method comprises smooth denoising algorithm, derivative processing method, standard normal variable Transformation method, multivariate scattering correction method, Normalization normalization method and a combination of the above methods. 10. a kind of method that utilizes near-infrared spectroscopy to quickly measure copper algal cellulose content according to claim 5, is characterized in that, described characteristic band selection method comprises interval partial least squares method, moving window partial least squares method, Monte Carlo method, correlation coefficient method, continuous projection method, genetic algorithm and combination of the above methods; said multivariate correction method includes principal component regression method and partial least squares method.