Plant moisture content determination method based on pXRF

Abstract

The invention provides a pXRF-based plant moisture content determination method, which comprises the following steps: carrying out intermittent drying and pXRF element content measurement on a plant sample, establishing a piecewise linear regression calibration curve of the plant moisture content and the pXRF element content, and determining the moisture content of the same kind of plant samples by using a calibration curve equation. According to the method, the water content of the plant can be rapidly, economically and accurately measured only by one-time standard sample treatment in a laboratory, calibration and establishment of a standard curve. The pXRF-based plant moisture content determination method is used for solving the problems of loss, complexity, high price, poor applicability and the like of the existing laboratory analysis method.

Description

Technical Field

[0001] The invention relates to the field of chemical analysis, and in particular to a method for determining plant moisture content based on pXRF. Background Art

[0002] In the field of chemical analysis technology, the current analysis of plant sample moisture content mainly relies on lossy methods such as drying, infrared moisture meter, microwave method, etc. based on sample heating and water loss, as well as non-destructive methods such as conductivity or capacitance method, near-infrared spectroscopy, etc. based on the electrical and optical physical properties of water. Existing non-destructive methods often require expensive equipment, complex technology, or have high requirements for the salt content in the sample, which greatly limits the efficiency of plant moisture content determination and affects the effective monitoring of plant growth status.

[0003] Portable X-ray fluorescence spectrometers have many advantages such as non-destructive, fast, economical and multi-element measurement, which can greatly reduce analysis time and reduce testing costs. Existing studies have obtained different moisture contents by grinding, adding water and other methods, and then studied the characteristics of reading changes or the relationship between readings and moisture content when pXRF measures samples such as soil and fertilizer. This method is easily limited by the uneven distribution of water in the sample, and it is impossible to achieve non-destructive analysis of plant samples. The inventors found in a large number of dehydration experiments that the dilution effect of water in plants will lead to linear changes in element content, which provides a theoretical basis for pXRF non-destructive detection of plant moisture content. However, traditional pXRF has not been applied to the determination of moisture content of plant samples, and there is no complete and feasible technical route and method flow. Summary of the invention

[0004] The purpose of the present invention is to provide a method for determining plant moisture content based on pXRF, so as to solve the problems of existing laboratory analysis methods such as loss, complexity, high cost and poor applicability.

[0005] To achieve the above object, the present invention provides a method for determining plant moisture content based on pXRF, comprising the following steps:

[0006] Step 1: Drying the plant standard samples: Dry the plant standard samples intermittently for a short time, measure and record the weight of the plant standard samples before and after drying, repeat this step until the weight no longer changes, calculate the moisture content of the plant standard samples based on the drying method, and obtain the plant standard sample moisture content set X i , where i represents different moisture contents;

[0007] Step 2: pXRF calibration of element content of plant standard samples: Use pXRF to measure the plant standard samples with different moisture contents in step 1, and obtain the pXRF reading value set Y of the element j content of the standard samples ij , j represents different types of elements;

[0008] Step 3: Use the piecewise linear regression analysis method to compare the pXRF reading value sets of different standard sample elements j with the water content sets to obtain the pXRF reading value set Y of the standard sample element j. ij and moisture content set X i The linear calibration curve equation between

[0009] For plant samples with small monolayer thickness:

[0010] Y ij (X i )=a 1i X i +b 1i ;

[0011] Among them, X i is the moisture content, Y ij is the pXRF measurement reading of element j at water content i, a 1i is the slope of the linear regression equation, b 1i is the intercept;

[0012] For plant samples with large monolayer thickness:

[0013] The turning point X of the piecewise linear regression equation is obtained by calculation i ',get

[0014]

[0015] Among them, a 2i and a 3i is the slope of the piecewise linear regression equation for moisture content i, b 2i and b 3i is the intercept, where X i 'Affected by the thickness of the sample;

[0016] Step 4: Element screening: Screen the element m with good fitting effect according to the fitting coefficient R2 of the piecewise linear regression equation;

[0017] Step 5: Use pXRF method to measure the plant sample to obtain the reading Y of element m content m , and quickly obtain the moisture content X of the plant sample to be tested through the piecewise linear calibration curve equation in step 3 m ;

[0018] Step 6: Obtain the mean moisture content of the element m screened in step 4 to obtain the mean moisture content equation of the plant sample to be tested.

[0019]

[0020] Among them, X mis the moisture content calculated for the screening element m, and W is the arithmetic mean of the moisture contents calculated for the screening element m.

[0021] Preferably, in step 1, the plant standard sample should be intermittently dried below 50 degrees Celsius until constant weight, and the drying time is gradually increased each time so that the moisture content of the standard sample is almost evenly reduced.

[0022] Preferably, in step 2, when performing pXRF calibration on the element content of the plant standard sample dried by the method of step 1, and when using pXRF to measure the plant sample to be tested in step 5, the sample thickness should be greater than or equal to 30 mm.

[0023] Preferably, in step 4, the element screening should select elements whose piecewise linear regression equation fitting coefficient R2 is simultaneously greater than 0.9.

[0024] Therefore, the present invention adopts the above-mentioned method for determining plant moisture content based on pXRF, and the technical effects are as follows:

[0025] (1) The pXRF-based plant moisture content determination method described in the present invention does not require pretreatment of plant samples, does not require large-scale instruments and equipment, is time-saving and low-cost, and can complete sample preparation and measurement in a simple environment.

[0026] (2) The pXRF-based plant moisture content determination method described in the present invention only requires one drying of a laboratory standard sample and pXRF measurement to obtain a standard calibration curve, which can be applied to moisture content detection of any identical plant sample with unknown moisture content.

[0027] (3) The pXRF-based plant moisture content determination method described in the present invention uses a method of taking the arithmetic mean of multiple measurements in the measurement of plant standard samples and / or plant test samples to reduce the accidental error of a single measurement and improve the accuracy of the pXRF measurement results.

[0028] (4) The pXRF-based plant moisture content determination method described in the present invention establishes a linear regression equation based on the dilution effect of water on the chemical composition of plants to achieve moisture content detection. This is a completely new method different from traditional moisture content measurement methods.

[0029] (5) The pXRF-based plant moisture content determination method described in the present invention has few restrictions on sample conditions and working environment, so the method has good universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A linear regression calibration curve diagram of a pXRF-based plant moisture content determination method according to K element determination provided in Example 1 of the present invention;

[0031] Figure 2 A linear regression calibration curve diagram of a pXRF-based plant moisture content determination method according to Ca element determination provided in Example 1 of the present invention;

[0032] Figure 3 A piecewise linear regression calibration curve diagram of a plant moisture content determination method based on pXRF provided in Example 2 of the present invention according to K element determination;

[0033] Figure 4 A piecewise linear regression calibration curve diagram of a plant moisture content determination method based on pXRF provided in Example 2 of the present invention according to Ca element determination. DETAILED DESCRIPTION

[0034] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0035] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0036] Embodiment 1

[0037] A method for measuring plant moisture content based on pXRF comprises the following steps:

[0038] Step 1, plant standard sample drying: cut the calla lily leaves (single layer thickness 0.3mm) of uniform thickness into segments of about 10cm in length and width, spread them out in a certain order, and perform intermittent short-term drying at a drying temperature of 50°C. The time interval is 2 minutes for the first time and then gradually increases to make the moisture content almost uniformly reduced. Measure and record the weight of the calla lily leaf samples before and after drying. Repeat this step until the weight does not change (initial weight and weight after 27 dryings, each weight is the average of three measurements: 56.1g, 54.7g, 53.3g, ... g, 49.8 g, 45.7 g, 44 g, 39.8 g, 36.8 g, 34.4 g, 32.2 g, 29.7 g, 27 g, 21.3 g, 19.9 g, 18.3 g, 17 g, 16.4 g, 15.3 g, 14.5 g, 13.8 g, 13.4 g, 12.8 g, 12 g, 11.8 g, 11.5 g, 11.5 g, 11.4 g, 11.1 g), based on the drying method, the moisture content of the calla lily leaf sample before and after each drying is calculated to obtain a moisture content set X i(80.0%, 79.6%, 79.0%, 77.5%, 75.5%, 74.6%, 71.9%, 69.6%, 67.5%, 65.3%, 62.4%, 58.6%, 47.4%, 43.9%, 38.8%, 34.1%, 31.7%, 27.0%, 22.7%, 19.1%, 16.6%, 12.6%, 6.4%, 5.2%, 2.7%, 2.6%, 1.7%, 0%);

[0039] Step 2: pXRF calibration of element content of calla lily leaf standard samples: Use pXRF to measure the calla lily leaf standard samples with different moisture contents in step 1, and obtain the pXRF reading value set Y of the element content of the standard samples. ij (Mean value of three pXRF measurements of element K: 5545.3μg / g, 6242.2μg / g, 5764.7μg / g, 6023.9μg / g, 7001.2μg / g, 6473.5μg / g, 9280.7μg / g, 13653.3μg / g, 16365.4μg / g, 19283.3μg / g, 20700.2μg / g, 25311.1μg / g, 32903.3μg / g, 31707.7μg / g, 41 530.4μg / g, 43062.2μg / g, 47138.6μg / g, 48714.8μg / g, 58256.4μg / g, 51690.8μg / g, 57371.3μg / g, 56596.2μ g / g, 63835.8μg / g, 67289.4μg / g, 63431.2μg / g, 69422.7μg / g, 68937.4μg / g and 67152.3μg / g; the average of three pXRF measurements of element Ca : 10285.8μg / g, 12159.3μg / g, 11142.1μg / g, 14663.2μg / g, 13951.5μg / g, 18111.7μg / g, 18807.3μg / g, 2728 4.8μg / g, 30989.3μg / g, 33689.5μg / g, 32861.6μg / g, 38130.8μg / g, 63766.2μg / g, 63718.5μg / g, 73995.3μg / g / g, 88666.6μg / g, 81528.3μg / g, 108543.1μg / g, 109001.4μg / g, 95450.7μg / g, 117017.3μg / g, 123015.6μg / g, 130063.7μg / g, 138036.7μg / g, 125081.2μg / g, 129395.9μg / g, 124131.5μg / g and 131107.7μg / g), j represents different element types;

[0040] When using pXRF to measure standard samples before or after drying, the same position of the standard sample should be measured each time to eliminate the potential impact of sample inhomogeneity.

[0041] When the element content pXRF calibration is performed on the plant standard sample dried by the method in step one, and when the plant sample to be tested is measured by pXRF in step five, the total thickness of the sample is 50 mm.

[0042] When using pXRF to measure standard samples or samples to be tested, multiple repeated measurements should be performed and the arithmetic mean should be calculated to obtain the accurate pXRF reading of the element content in the standard sample or sample to be tested.

[0043] Step 3: Use the linear regression analysis method to compare different pXRF reading value sets of standard sample element j with different water content sets, and obtain the linear regression equation between the pXRF reading value set of standard sample element j and the water content set:

[0044] Y ij (X i )=-80746X i +69952 (element K);

[0045] Y ij (X i )=-159219X i +136922 (element Ca);

[0046] Among them, X i is the set of moisture content i, Y ij is the pXRF measurement reading of element j at water content i.

[0047] like Figure 1 and Figure 2 As shown, the horizontal axis is the moisture content in %, and the vertical axis is the readings of element K and element Ca obtained by pXRF measurement in μg / g.

[0048] Step 4: Element screening: According to the fitting coefficient R of the piecewise linear regression equation 2 Elements with good fitting effects were screened, and K and Ca with fitting coefficients greater than 0.9 were obtained as predicted elements.

[0049] Step 5: pXRF was used to analyze the Alocasia leaf samples with unknown moisture content and intermittent drying, and the element K readings were 6899.3 μg / g, 15957.5 μg / g, 16438.4 μg / g, 29040.4 μg / g, 40884.8 μg / g and 54024.3 μg / g. The corresponding predicted moisture contents were 78.1%, 66.9%, 66.3%, 50.7%, 36.0% and 19.7% using the linear regression calibration curve equation obtained in step 3. The element Ca readings were 5047.3 μg / g, 19231.0 μg / g, 35833.6 μg / g, 53278.3 μg / g, 71239.6 μg / g and 106239.6 μg / g. The linear regression calibration curve equation obtained in step 3 gave the corresponding predicted moisture contents of 82.8%, 73.9%, 63.5%, 52.5%, 41.3%, and 19.3%. The average of the two element results gave the final moisture contents of 80.5%, 70.4%, 64.9%, 51.6%, 38.6%, and 19.5%.

[0050] The above steps 1 to 4 are the preparation of standard samples, measurement of moisture content, calibration of element content and establishment of calibration curves. Steps 1 and 2 propose a simple process for obtaining different moisture contents of samples and corresponding pXRF readings, with short sample preparation time and high measurement efficiency. Steps 3 and 4 are the establishment of linear regression calibration curves and prediction element screening. Each pXRF device only needs to go through a calibration curve establishment process once to achieve accurate determination of the moisture content of the calla lily leaf sample. As shown in Table 1, for the samples to be verified, the application of the linear regression calibration curve effectively predicts the moisture content. The calculation method of the relative error is: relative error = (predicted moisture content-actual moisture content) / actual moisture content. It can be seen from Table 1 that the relative errors of the moisture content predicted by K and Ca are less than 20%. In addition, the maximum error of the K predicted moisture content exceeds 15%, with large fluctuations, and the maximum error of the Ca predicted moisture content is about 11%, while the relative error fluctuation range of the mean of the two elements predicted moisture content is smaller and more robust.

[0051] Table 1 Predicted moisture content and relative error of calla lily leaf verification samples

[0052]

[0053] Embodiment 2

[0054] A method for measuring plant moisture content based on pXRF comprises the following steps:

[0055] Step 1, plant standard sample drying: cut the leaves of the imperial pineapple with uniform thickness into pieces of about 10 cm in length, spread them out in a certain order, and perform intermittent short-term drying at a drying temperature of 50°C. The time interval is 2 minutes for the first time, and then gradually increases to make the moisture content decrease evenly. Measure and record the weight of the imperial pineapple sample before and after drying. Repeat this step until the weight does not change (initial weight and weight after 44 dryings, each weight is the average of three measurements: 112.3g, 108.6g, 106.16g, 102.5g, 99.0g, 92.3g, 87.0g, 84.4g, 81.9g, 79.4g, 77.1g, 73.8g, 7 g, 31.6 g, 31.1 g, 30.6 g, and 30.5 g), and the moisture content of the imperial pineapple sample before and after each drying is calculated based on the drying method to obtain the moisture content set X i (72.8%, 71.8%, 71.2%, 70.2%, 69.1%, 66.9%, 64.9%, 63.7%, 62.7%, 61.5%, 60.3%, 58.6%, 57.3%, 56.0%, 54.3%, 53.2%, 52.0%, 50.9%, 49.3%, 47.5%, 46.2%, 44.1%) , 42.9%, 41.3%, 40.0%, 37.0%, 35.6%, 33.7%, 33.0%, 29.9%, 28.8%, 25.6%, 23.0%, 21.1%, 17.3%, 17.4%, 14.5%, 9.2%, 7.1%, 6.3%, 4.5%, 3.3%, 1.7%, 0.3%, 0.0%).

[0056] Step 2: pXRF calibration of element content of standard samples of royal pineapple: Use pXRF to measure the standard samples of royal pineapple with different moisture contents in step 1, and obtain the pXRF reading value set Y of the standard samples. ij(Mean values of three pXRF measurements of K element: 76.2 μg / g, 124.0 μg / g, 202.8 μg / g, 242.3 μg / g, 237.7 μg / g, 191.2 μg / g, 191.5 μg / g, 244.8 μg / g, 187.4 μg / g, 275.7 μg / g, 306.3 μg / g, 308.7 μg / g, 380.8 μg / g, 413.4 μg / g, 459.3 μg / g, 390.6 μg / g, 587.7 μg / g, 426.8 μg / g, 545.5 μg / g, 639.6 μg / g, 620.2 μg / g, 711.2 μg / g, 775.5 μg / g, 558.3 μg / g, 730.7 μg / g, 647.3 μg / g, 642.8 μg / g, 597.4 μg / g, 657.8 μg / g, 836.3 μg / g, 746.8 μg / g, 943.0 μg / g, 1255.3 μg / g, 1215.7 μg / g, 1042.5 μg / g, 1352.7 μg / g, 1540.8 μg / g, 2086.1 μg / g, 2112.0 μg / g, 2519.6 μg / g, 2817.3 μg / g, 3758.8 μg / g, 4835.4 μg / g, 5286.8 μg / g and 5589.8 μg / g; Mean values of three pXRF measurements of Ca element: 443.6 μg / g, 422.9 μg / g, 482.6 μg / g, 474.6 μg / g, 490.7 μg / g, 485.1 μg / g, 527.3 μg / g, 526.9 μg / g, 520.5 μg / g, 546.8 μg / g, 646.8 μg / g, 603.3 μg / g, 643.7 μg / g, 695 μg / g, 725.9 μg / g, 704.8 μg / g, 788.8 μg / g, 724.4 μg / g, 803.8 μg / g, 908.5 μg / g, 888.4 μg / g, 933.6 μg / g, 986.6 μg / g, 936.8 μg / g, 1418 μg / g, 1104.2 μg / g, 1159.3 μg / g, 1193 μg / g, 1227.5 μg / g, 1272.1 μg / g, 1387.4 μg / g, 1556.1 μg / g, 1773.1 μg / g, 2050.7 μg / g, 1793.1 μg / g, 2143.2 μg / g, 2227.4 μg / g, 2923.7 μg / g, 3203.6 μg / g, 3518.3 μg / g, 4479.3 μg / g, 4664.4 μg / g, 6786.8 μg / g, 7396 μg / g and 7574 μg / g), where j represents different element types;

[0057] When using pXRF to measure standard samples before or after drying, the same position of the standard sample should be measured each time to eliminate the potential impact of sample inhomogeneity.

[0058] When the element content pXRF calibration is performed on the plant standard sample dried by the method in step one, and when the plant sample to be tested is measured by pXRF in step five, the sample thickness is 50 mm.

[0059] When using pXRF to measure standard samples or samples to be tested, multiple repeated measurements should be performed and the arithmetic mean should be calculated to obtain the accurate pXRF reading of the element content in the standard sample or sample to be tested.

[0060] Step 3: Use the piecewise linear regression analysis method to compare the different pXRF reading value sets of standard sample elements j with the different water content sets, and calculate the turning point X of the piecewise linear regression equation. i ′=0.1 (10% moisture content), and finally the piecewise linear regression equation between the pXRF reading value set and the moisture content set of the standard sample element j is obtained:

[0061]

[0062] Among them, X i is the set of moisture content i, Y ij is the pXRF measurement reading of element j at water content i, X i '=0.1 (10% moisture content).

[0063] like Figure 3 and Figure 4 As shown, the horizontal axis is the moisture content in %, and the vertical axis is the readings of element K and element Ca obtained by pXRF measurement in μg / g.

[0064] Step 4: Element screening: According to the fitting coefficient R of the piecewise linear regression equation 2 Elements with good fitting effects were screened, and K and Ca with fitting coefficients greater than 0.9 were obtained as predicted elements.

[0065] Step 5: pXRF was used to analyze the samples of the royal pineapple with unknown moisture content and intermittent drying, and the element K readings were 328.1μg / g, 467.7μg / g, 1009.4μg / g, 1071.2μg / g, 2082.6μg / g, 2733.4μg / g, 3087.5μg / g and 3596.7μg / g. The corresponding predicted moisture contents were 59.7%, 52.7%, 25.1%, 22.0%, 6.9%, 5.5%, 4.8% and 3.8% using the piecewise linear regression calibration curve equation obtained in step 4. The Ca readings were 612.1 μg / g, 986.7 μg / g, 1508.9 μg / g, 1667.2 μg / g, 1853.7 μg / g, 4004.7 μg / g, 4873.2 μg / g, 4941.7 μg / g and 5846.2 μg / g. Using the piecewise linear regression calibration curve equation obtained in step 4, the corresponding predicted moisture contents were 60.2%, 46.9%, 28.3%, 22.7%, 8.7%, 5.4%, 4.1% and 3.9%.

[0066] The above steps 1 to 4 are the preparation of standard samples, moisture content measurement, element content calibration and the establishment of calibration curves. Steps 1 and 2 propose a simple process for obtaining different moisture contents of samples and corresponding pXRF readings, with short sample preparation time and high measurement efficiency. Steps 3 and 4 are the establishment of a piecewise linear regression calibration curve and the screening of predicted elements. Each pXRF device only needs to go through the calibration curve establishment process once to accurately determine the moisture content of the imperial pineapple sample. As shown in Table 2, for the samples to be verified, the application of the piecewise linear regression calibration curve effectively predicts the moisture content. The calculation method of relative error is: relative error = (predicted moisture content-actual moisture content) / actual moisture content. The predicted moisture content of the element K result is close to the turning point of the piecewise linear regression equation, and the predicted relative errors of the remaining moisture contents are all less than 10%, and the predicted moisture content of the element Ca is less than 10% except for two exceeding 11%. When the average predicted values ​​of K and Ca are used, the maximum relative error is about 10%, the overall fluctuation is smaller, and the prediction results are more robust.

[0067] Table 2 Predicted moisture content and relative error of the imperial pineapple verification samples

[0068]

[0069]

[0070] Therefore, the present invention adopts the above-mentioned pXRF-based plant moisture content determination method to solve the problems of existing laboratory analysis methods such as loss, complexity, costliness, and poor applicability.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for determining plant moisture content based on pXRF, characterized in that: The following steps are involved: Step 1: Drying the plant standard samples: Dry the plant standard samples intermittently for a short time, measure and record the weight of the plant standard samples before and after drying, repeat this step until the weight no longer changes, calculate the moisture content of the plant standard samples based on the drying method, and obtain the plant standard sample moisture content set X i , where i represents different moisture contents; Step 2: pXRF calibration of element content of plant standard samples: Use pXRF to measure the plant standard samples with different moisture contents in step 1, and obtain the pXRF reading value set Y of the element j content of the standard samples ij , j represents different types of elements; Step 3: Use the piecewise linear regression analysis method to compare the pXRF reading value sets of different standard sample elements j with the water content sets to obtain the pXRF reading value set Y of the standard sample element j. ij and moisture content set X i The linear calibration curve equation between For plant samples with small monolayer thickness: Y ij (X i )=a 1i X i +b 1i ; Among them, X i is the moisture content, Y ij is the pXRF measurement reading of element j at water content i, a 1i is the slope of the linear regression equation, b 1i is the intercept; For plant samples with large monolayer thickness: The turning point X of the piecewise linear regression equation is obtained by calculation i ',get Among them, a 2i and a 3i is the slope of the piecewise linear regression equation for moisture content i, b 2i and b 3i is the intercept, where X i 'Affected by the thickness of the sample; Step 4: Element screening: According to the piecewise linear regression equation fitting coefficient R 2 Filter the element m with good fitting effect; Step 5: Use pXRF to measure the plant sample to obtain the reading Y of the element m content m , and quickly obtain the moisture content X of the plant sample to be tested through the piecewise linear calibration curve equation in step 3 m ; Step 6: Obtain the mean moisture content of the element m screened in step 4 to obtain the mean moisture content equation of the plant sample to be tested. Among them, X m is the moisture content calculated for the screening element m, and W is the arithmetic mean of the moisture contents calculated for the screening element m.

2. A method for determining plant moisture content based on pXRF according to claim 1, characterized in that: In step 1, the plant standard sample should be placed below 50 degrees Celsius and intermittently dried until constant weight, and the drying time should be gradually increased each time so that the moisture content of the standard sample is almost evenly reduced.

3. The method for determining plant moisture content based on pXRF according to claim 1, characterized in that: In step 2, when the element content pXRF calibration is performed on the plant standard sample dried by the method in step 1, and when the plant sample to be tested is measured by pXRF in step 5, the sample thickness must be greater than or equal to 30 mm.

4. The method for determining plant moisture content based on pXRF according to claim 1, characterized in that: In step 4, the element screening should select the piecewise linear regression equation fitting coefficient R 2 At the same time, the elements that are greater than 0.9.

Images