Method for analyzing the blend composition of cigarette samples based on the thermal properties of tobacco leaves

Abstract

This invention discloses a method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves. The method includes (1) preparing a cigarette sample to be analyzed and a single-grade tobacco sample; (2) collecting thermal analysis spectra of the cigarette sample to be analyzed and the single-grade tobacco sample; and (3) studying the thermal analysis spectrum algorithm to obtain the composition and proportion of tobacco leaves in the cigarette sample to be analyzed. The method of this invention can complete the compositional analysis of an unknown cigarette within minutes and obtain a clear composition and proportion of components, and is objective, efficient, versatile, highly reproducible, and highly sensitive, offering unique advantages in cigarette analysis in the tobacco industry.

Description

Technical Field

[0001] The present invention relates to the technical field of tobacco, and specifically to a method for analyzing the blend composition of cigarette samples based on the thermal properties of leaf tobacco.

Background Art

[0002] The quality and flavor characteristics of cigarettes are mainly formed by product designers combining different types of leaf tobacco from various origins, varieties, and grades. Usually, based on leaf group experience and sensory evaluation, it is necessary to manually select 10 - 20 types of leaf tobacco from hundreds of types of leaf tobacco raw materials in stock and perform component design at different ratios. Therefore, the blend composition of cigarettes becomes very complex, and it is impossible to manually analyze the blend composition of unknown cigarettes. If it is possible to analyze the blend composition of cigarettes using objective data and scientific and technological means through instrumental measurement methods, it will have important significance for the analysis of competing cigarettes and leaf group design.

[0003] Thermogravimetric analysis (TG / DTA) can provide stable reaction conditions under programmed heating conditions and is the most ideal measurement method in the thermal decomposition research of tobacco. The derivative thermogravimetric analysis derived from thermogravimetric analysis, also called differential thermogravimetry, is a technique that gives the first derivative of the TG curve with respect to temperature or time. The result obtained by measurement is the differential thermogravimetric curve, that is, the DTG curve. The characteristics of the DTG curve can accurately reflect the reaction start temperature, maximum reaction rate temperature, and reaction end temperature of each weight loss stage. The area of each peak on the DTG curve is proportional to the corresponding sample weight loss on the TG curve. When the step appearing in a specific heating process on the TG curve is not clear, the DTG curve can be used to clearly distinguish them. The main feature of thermogravimetric analysis is its high quantification ability, and it can accurately measure the mass change and change rate of substances. From this feature, it can be said that when the mass changes when heating a substance, it can be studied by thermogravimetric analysis.

[0004] The current analysis of cigarette blend composition often involves a combination of methods, including chemical composition analysis of loose tobacco, chemical composition analysis of smoke, and sensory evaluation. However, this method relies heavily on manual labor, is highly subjective, and yields ambiguous and unreliable results.

[0005] To solve the above problems, we propose the present invention. [Overview of the Initiative]

[0006] Because DTG curves effectively represent the quality information of shredded tobacco / loose tobacco, by measuring and calculating the differences in DTG curves, it is possible to simulate and evaluate the components and analyze the agreement between the composition and the actual blend components. To improve the versatility of cigarette blend analysis and the workability of analysts, and to analyze and represent the quality information of shredded tobacco in cigarettes using thermal analysis spectra, the present invention has designed an optimization algorithm that combines a DTG difference correlation model and component analysis, automatically retrieves the proportion of loose tobacco in the blend components of cigarettes, and analyzes the blend composition of cigarettes using objective data, thereby clearly obtaining the composition and proportion of blend components, which has important implications for the analysis of competing tobaccos and the design of blend components.

[0007] This invention provides a method for analyzing the blend composition of cigarette samples based on the thermal properties of tobacco leaves. Specifically, it analyzes competing cigarettes (the cigarettes to be analyzed) based on conventional cigarettes to obtain the specific tobacco leaf composition and component ratios.

[0008] The technical means of the present invention are as follows:

[0009] A method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves includes the steps of (1) preparing a cigarette sample to be analyzed and a single-grade tobacco sample, (2) collecting thermal analysis spectra of the cigarette sample to be analyzed and the single-grade tobacco sample, and (3) studying an algorithm for thermal analysis spectra to obtain the composition and proportion of tobacco leaves in the cigarette sample to be analyzed.

[0010] Preferably, in step (1), there is one cigarette sample to be analyzed, and there are 50 or more single-grade tobacco leaf samples, and each sample is equilibrated for 48 hours or more in a constant temperature and humidity environment of (22±1)℃ and relative humidity of (60±2)%. Typically, the cigarette sample to be analyzed has no added flavorings or additives, weighs 5g or more, and has a grinding mesh count of 100 mesh or more. More than 50 typical single-grade tobacco leaf samples are selected, and the information of the tobacco leaf samples needs to cover different grades, different origins, and different parts, as there are significant differences in the taste of typical single-grade tobacco leaf samples. The tobacco leaf samples weigh 5g or more and have a grinding mesh count of 100 mesh or more.

[0011] Preferably, the step of collecting the thermal analysis spectrum in step (2) involves weighing (5.00 ± 0.05) mg of sample, placing it in an alumina crucible for thermogravimetric analysis and heating it. The program is set to start at 50°C, heating rate at 10°C / min, and end at 850°C. The sample is kept at 850°C for 5 minutes, with both the protective gas and reaction gas being nitrogen gas, and the measurement is performed at a flow rate of 20 mL / min. The measurement results are plotted with temperature (°C) on the X axis and mass (%) on the Y axis, and the derived data is used as the TG result data. Before sample analysis, impurities are emptied from the body of the alumina crucible for thermogravimetric analysis to be used, and the thermogravimetric analyzer is set to be held at 900°C for 10 minutes to use the empty crucible as a reference. The thermogravimetric analyzer's balance has a sensitivity of 0.1 μg or higher and a curve resolution of 50 million resolution points or higher.

[0012] Preferably, the specific steps of studying the thermal analysis spectrum algorithm of step (3) and obtaining the composition and proportion of tobacco leaves in the cigarette being analyzed are: (A) The obtained TG result data is differentiated first with respect to time to obtain the differential weight loss DTG curve. The DTG matrix of the analyzed cigarette sample is Y, and the DTG matrix of the single-grade tobacco sample is X = [x1 x2...x n ](n is the number of single-grade tobacco leaf samples), (B) Temperature range selection: Within the temperature range of 50°C to 850°C, select the DTG curve corresponding to a specific temperature range as the actual analysis curve. (C) Determination of the objective function: Select an appropriate objective function for the purpose of optimization. (D) Determination of limiting conditions: In order to ensure computational accuracy and prevent the waste of computational resources, appropriate limiting conditions are set according to the actual fitting situation. (E) Determination of optimization method: Establish linear constraints, optimize the objective function, and obtain the combination coefficients and proportions of single-grade tobacco leaves that are most similar to the cigarette sample under analysis, i.e., the composition and proportions of tobacco leaves in the cigarette under analysis.

[0013] Preferably, the temperature range in step (B) is 150 to 380°C.

[0014] Preferably, the objective function of step (C) is: (a)F=1 / corr((c * X), Y) (where corr is the calculation of the correlation coefficient and c is the combination coefficient of single-grade tobacco leaves), or, (b)F=sum(sqrt(((c * XY). / Y).^2))(where sqrt is the calculation of the root mean square, sum is the calculation of the sum, and c is the combination coefficient of single-grade tobacco leaves), or, (c)F=sqrt(sum((c *It is either XY).^2) / sum(Y.^2)) (where sqrt is the calculation of the root mean square, sum is the calculation of the sum, and c is the combination coefficient of single-grade tobacco leaves).

[0015] Preferably, the constraint in step (D) is: TIFF0007872362000001.tif9170, c∈[0,1], c i >p... (where z is a natural number other than 0, and p is a specific value). Each of these indicates that the sum of the combination coefficients of cigarettes is 1, the combination ratio of cigarettes is a multiple of 0.025, the combination ratio of cigarettes is between 0 and 1, and the number of cigarette combinations in a given carton is greater than or equal to p....

[0016] Preferably, the optimization method for step (E) is one or more of a global optimization algorithm, gradient descent, or genetic algorithm.

[0017] The present invention has the following beneficial effects.

[0018] 1. The method of the present invention provides thermal analysis spectra of finished cigarettes and single-grade tobacco leaves to be analyzed, and through dynamic studies of thermogravimetric analysis program heating, selects an appropriate objective function, sets appropriate limiting conditions and optimization methods to obtain the composition and proportion of single-grade tobacco leaves that are similar to, and even identical to, the sensory quality of the cigarettes to be analyzed. The method of the present invention can complete the compositional analysis of commercially available finished cigarettes to be analyzed within minutes, significantly reducing workload and the number of tests, and is objective, efficient, versatile, highly reproducible, and highly sensitive, offering unique advantages in the analysis of finished cigarettes in the tobacco industry.

[0019] 2. The method of the present invention does not require a large number of wet chemical operations such as chemical composition analysis of cut tobacco and chemical composition analysis of smoke in the blend analysis of conventional cigarettes. Instead, it uses dry chemical operations, which are easy to operate, require an extremely small sample usage of within 10 mg, are non-toxic and harmless, do not cause harm to operators, and do not cause environmental pollution.

[0020] 3. The method of the present invention provides specific component design goals, rich data support, and digital programming technology for the development of cigarette products, realizes automatic search and objective evaluation of the component design plan of cigarette products, and can sufficiently prevent the influence of subjective factors and the expression of differences caused by the experience and sensory evaluation of conventional experts.

Embodiment for Carrying out the Invention

[0021] Hereinafter, the present invention will be described in detail by way of examples, but the present invention is not limited to these examples. Experimental methods for which specific conditions are not specified in the examples usually follow the conditions described in conventional conditions and manuals or the conditions proposed by the manufacturer. General devices, materials, reagents, etc. used can be obtained commercially unless otherwise specified. All raw materials required in the following examples and comparative examples are commercially available.

[0022] Example The method for analyzing the composition and ratio of the blend of a finished cigarette sample (the cigarette to be analyzed) of a well-known brand in China includes the following steps.

[0023] (1) Prepare one finished cigarette sample (the cigarette to be analyzed) of a well-known brand in China and 5 grams each of 50 single-grade leaf tobacco samples (single-grade leaf tobacco) from different origins, different parts, and different grades. Sieve the cigarette to be analyzed and the single-grade leaf tobacco through a 100-mesh sieve, and perform equilibration treatment for 48 hours in a constant temperature and humidity environment of (22 ± 1)°C and relative humidity (60 ± 2)%.

[0024] (2) Before sample analysis, the thermogravimetric analyzer is set to maintain a temperature of 900°C for 10 minutes to discharge impurities from the main unit and to use an empty crucible as a reference. A sample of (5.00 ± 0.05) mg is weighed and placed in a platinum crucible for thermogravimetric analysis. The heating program is set to a starting temperature of 50°C, a heating rate of 10°C / min, and an ending temperature of 850°C. The analyzer is kept at 850°C for 5 minutes, and both the protective gas and reaction gas are nitrogen gas, with a flow rate of 20 mL / min. The measurement results are shown on the X axis in temperature (°C) and the Y axis in mass (%), and the derived data is used as the TG result data.

[0025] (3) Differential weight loss curve data (DTG matrix) is obtained by taking the first derivative of the weight data with respect to time. As shown in Tables 1 and 2 below, the DTG matrix for the cigarettes analyzed is Y, and the DTG matrix for single-grade tobacco is X = [x1 x2...x 50 ]

[0026] TIFF0007872362000002.tif40170

[0027] TIFF0007872362000003.tif63170

[0028] (4) Select DTG data corresponding to the temperature range of 100~200℃ as X_New and Y_New.

[0029] TIFF0007872362000004.tif36170

[0030] TIFF0007872362000005.tif63170

[0031] (5) The objective function is F = 1 / corr((c * X), Y), constraint parameters Set to TIFF0007872362000006.tif13170, establish parameter calculation based on an overall optimization algorithm, and the Python code is as follows: from scipy import stats import numpy as np #Objective function based on correlation coefficient def fitness(c): return 1 / stats.pearsonr(np.dot(Data_New, c), Target_New).statistic #Restrictions m,n=Data_New.shape x0 = np.random.random(n) # Initializes with a random number between 0 and 1 bnds=[(0,1)] * n# Define the range of values ​​for coefficients. cons=({'type':'eq','fun':lambdax:sum(x)-1}) # The sum of the coefficients will be 1 #Basinhopping algorithm-based global search from scipy.optimize import basinhopping minimizer_kwargs={”method”:”SLSQP”, "constraints":cons, "bounds": bnds,} ret=basinhopping(fitness,x0,minimizer_kwargs=minimizer_kwargs, (niter=200)

[0032] The calculation results are shown in Table 3 below.

[0033] TIFF0007872362000007.tif111170

[0034] Verification Experiment: Samples of shredded tobacco for cigarettes were prepared by mixing 14 types of single-grade tobacco leaves and their component ratios as shown in Table 3 above. Nine sensory evaluation experts were assembled to perform a sensory evaluation using a scoring method to assess the differences in sensory quality between the above-mentioned mixed shredded tobacco samples and the sample of shredded tobacco from the target cigarettes. As a result, no differences were observed in the sensory evaluation results.

[0035] The above examples illustrate only a few embodiments of the present invention, and while the description is more specific and detailed, it should not be understood as limiting the scope of the invention. Those skilled in the art will note that various modifications and improvements can be made without departing from the spirit of the invention, and all of these fall within the scope of protection of the invention. Accordingly, the scope of protection of the invention shall be defined in the appended claims.

Claims

1. A method for analyzing the blend composition of cigarette samples based on the thermal properties of tobacco leaves, The method includes (1) the step of preparing a sample of cigarettes to be analyzed and a sample of single-grade tobacco leaves (1), (2) the step of collecting thermal analysis spectra of the sample of cigarettes to be analyzed and the sample of single-grade tobacco leaves (2), and (3) the step of studying an algorithm for thermal analysis spectra and obtaining the composition and proportion of tobacco leaves in the cigarettes to be analyzed (3). The specific steps for studying the thermal analysis spectrum algorithm in step (3) above and obtaining the composition and proportion of tobacco leaves in the cigarette being analyzed are as follows: Step (A) The obtained TG result data is differentiated first with respect to time to obtain the differential weight loss DTG curve. The DTG matrix for the cigarette sample to be analyzed is Y, and the DTG matrix for the single-grade tobacco sample is X = [x １ x ２ ...x ｎ ] (where n is the number of single-grade tobacco leaf samples), Step (B) Temperature Range Selection: Within the temperature range of 50°C to 850°C, select the DTG curve corresponding to the specific temperature range as the actual analysis curve. Step (C) Determination of the objective function: Select an appropriate objective function for the purpose of optimization. Step (D) Determining the limiting conditions: Set appropriate limiting conditions according to the actual fitting situation. Step (E) Determination of optimization method: A method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves, characterized by establishing linear constraints, optimizing the objective function, and obtaining the combination coefficients and proportions of single-grade tobacco leaves that are most similar to the cigarette sample under analysis, i.e., the composition and proportions of tobacco leaves in the cigarette sample under analysis.

2. The method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves, according to Claim 1, characterized in that in step (1), there is one cigarette sample to be analyzed, there are 50 or more single-grade tobacco leaf samples, and each sample is equilibrated for 48 hours or more in a constant temperature and humidity environment of (22±1)°C and relative humidity of (60±2)%.

3. The step of collecting the thermal analysis spectrum in step (2) is characterized in that the samples are each placed in a thermogravimetric crucible and heated, the program is set to start at 50°C, heating rate at 10°C / min, end at 850°C, constant temperature at 850°C for 5 minutes, both protective gas and reaction gas are nitrogen gas, and measurement is performed at a flow rate of 20 mL / min, the measurement results are shown with temperature (°C) on the X axis and mass (%) on the Y axis, and the derived data is used as TG result data, the method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves as described in Claim 1.

4. The method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves, according to Claim 1, characterized in that the temperature range of step (B) is 150 to 380°C.

5. The objective function of step (C) is: (a) F=1 / corr((c ＊ X), Y) (where corr is the calculation of the correlation coefficient and c is the combination coefficient of single-grade tobacco leaves), or, (b) F=sum(sqrt(((c ＊ X - Y) / Y) ^2)) (where sqrt is the calculation of the root mean square, sum is the calculation of the sum, and c is the combination coefficient of single-grade tobacco leaves), or, (c) F=sqrt(sum((c ＊ A method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves, as described in claim 1, characterized in that it is one of the following: X - Y) ^2) / sum(Y ^2)) (where sqrt is the calculation of the root mean square, sum is the calculation of the sum, and c is the combination coefficient of single grade tobacco leaves).

6. The limiting condition of step (D) is: , c∈[0,1], c ｉ A method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves, as described in claim 1, characterized by including > p... (where z is a natural number other than 0, and p is a specific value).

7. The method for analyzing the blend composition of a cigarette sample based on the thermal properties of tobacco leaves, according to Claim 1, characterized in that the optimization method in step (E) is one or more of a global optimization algorithm, a gradient descent method, or a genetic algorithm.

Images