Patent Application
20230075079
This application requires the priority of a Chinese invention patent application whose application date is Apr. 27, 2020, the application number is 202010345962.6, and the invention title is “A method for species identification and quality detection of edible oil based on nuclear magnetic resonance technology”.
This present invention relates to the technical field of magnetic resonance technology and liquid identification technology, specifically to a method of species identification and quality detection of liquid-like samples based on nuclear magnetic resonance relaxation technology.
Many foods and health products in daily life, such as edible oil, milk, jelly, donkey-hide gelatin, etc., are all in a liquid-like state. At present, species identification and quality detection of liquid-like foods and health products have become an important topic in food safety research. Take edible oil as an example. Edible oil is a necessity in life. Generally speaking, edible oil is composed of a variety of saturated and unsaturated fatty acids, with a unique flavor and rich nutrition. The unsaturated fatty acids (mainly oleic acid) can lower cholesterol, and essential fatty acids (such as linolenic acid, linoleic acid, arachidonic acid, etc.) can soften blood vessels, and thus lower blood lipids and blood pressure, promoting microcirculation. A reasonable intake of fatty acids can provide sufficient energy for the human body, meanwhile effectively preventing various cardiovascular diseases. Edible oil is widely used in daily cooking. China is a major producer and consumer of edible oil, but in order to make huge profits, some illegal traders in the market add lower price or different processing levels of edible oil to normal edible oils and sell the shabby edible oil as the good ones, causing serious damage to health and rights of consumers.
At present, the identification technology of edible oil used in the national standard in China mainly uses sensory experience and physical and chemical detection as preliminary measurement, and then uses gas chromatography to determine the composition of edible oil fatty acid and phytosterols as final identification. There are also some signature marker detection technologies and detection technologies based on a photoelectric sensor. However, the aforementioned methods have their own advantages and disadvantages, and thus it is difficult to meet the requirements of accuracy, stability, multiple measurement and operability at the same time. Meanwhile, the traditional chromatographic, mass spectrometry and optical spectroscopy often require sample pretreatment which destroys the sample during the detection, thus cannot realize non-destructive sample detection. More importantly, the identification technology of edible oil based on traditional chromatography, mass spectrometry, and optical spectroscopy usually rely on large instruments and cannot achieve rapid detection on site.
The existing patent CN108982570A relates to a method for identifying types of edible oil based on nuclear magnetic resonance technology, which is analyzed by the analysis of 1H nuclear magnetic resonance spectrum rather than relaxation analysis of edible oil. Accordingly, the method described in the patent CN108982570A has the detection principle which is completely different from that described in the present invention.
In the existing literature (Xin Wang et al., Food and Fermentation Industry 2011, 37, 177-181, DOI: 10.13995/j.cnki.11-1802/ts.2011.03.020; Xin Wang et al., Food Industry Science and Technology 2014, 12, 58-65; DOI: 10.13386/j.issn1002-0306.2014.12.003), comparison of different edible oils and adulterated identification are achieved by measuring 1H T2 relaxation properties of edible oils. The method can be used in the existing low-field and high-field magnetic resonance instruments. However, the method uses 1H T2 relaxation properties of the system under test obtained by the conventional method (i.e., CPMG sequence). From the experimental data, the 1H T2 relaxation properties of edible oil alone cannot be used to effectively distinguish different edible oils and the edible oil adulteration.
In addition, the adulteration of cow and goat milk, the mixing of different quality of donkey-hide gelatin, and the distinction of scorpion samples from different places of origin are also important issues related to foods and drugs safety, thus it is urgent to find a simple and fast method to distinguish similar samples.
In order to overcome the defects of current technologies, the present invention provides a method that utilizes the relaxation properties of liquid-like samples by amplifying the differences of relaxation properties to achieve species identification and quality detection of liquid-like samples. Firstly, edible oil is taken as an example to illustrate the content of the invention. The main chemical components in edible oils are various saturated and unsaturated fatty acids. The chemical structures of these saturated and unsaturated fatty acids are similar, but the types, contents, and relative proportions vary with different types of edible oils. Since different saturated and unsaturated fatty acids have different nuclear magnetic relaxation properties, this feature can be used for species identification and quality detection of edible oil. Similarly, the examples of the present invention show that the distinction of cow and goat milk, different donkey-hide gelatin or scorpions from different places of origin can also be achieved by using this method.
The present invention develops a method for species identification and quality detection of liquid-like samples based on nuclear magnetic resonance relaxation technology. The core design idea of this method is to amplify the differences in nuclear magnetic resonance relaxation properties of different liquid-like samples by designing a new nuclear magnetic resonance method to measure two-dimensional data set containing 1H T1 and T2 relaxation properties of liquid-like samples, thus realizing species identification and quality detection of liquid-like samples. Meanwhile, the present invention can realize non-destructive testing of samples without sample pretreatment. The method of the present invention can be implemented on a low-field magnetic resonance instrument which can achieve rapid on-site detection by moving on-board.
A more specific method comprises the following steps:
Step 1: Designing a pulse sequence that includes pulse block or composite pulses containing 1H spin-echo function and pulse block or composite pulses containing T1 filter function;
Step 2: Applying pulse sequence to the targeted liquid-like samples to obtain their 1H two-dimensional relaxation signals;
Step 3: Converting the obtained 1H two-dimensional relaxation signals into fingerprint spectra of the targeted liquid-like samples, so as to be used for species identification and quality detection of liquid-like samples.
Wherein the liquid-like samples refer to liquid and gel substances with a certain fluidity, including edible oil, cow and goat milk, donkey-hide gelatin, scorpion powder solution, yogurt, beverages, oils in general, etc.
In Step 1 of the present invention, the pulse sequence comprises the following designs and the sub-steps:
Step 1-1: Using a pulse block or composite pulses to excite 1H magnetic resonance signal of the system under test;
Step 1-2: Applying the pulse block or the composite pulses containing 1H spin-echo function to the system under test, and the pulse block or the composite pulses may contain one or more variables;
Step 1-3: Applying the pulse block or the composite pulses containing 1H T1 filter function to the system under test, and the pulse block or the composite pulses may contain one or more variables;
Step 1-4: Converting the 1H magnetic resonance signal of the targeted samples into a signal detectable by magnetic resonance instrument through the pulse block or the composite pulses, and then collecting the signals.
In the pulse sequence in the present invention:
The first step: Exciting the 1H magnetic resonance signal of the system under test with a 90° pulse with a phase of x;
The second step: Applying the composite pulse block [τ1−(180°)y−τ1]n, wherein n is the number of repetitions, on the system under test. The composite pulse block includes the time variable τ1 and the number repetition variable n;
The third step: Applying the composite pulse block [τ2−(90°)x−τ3] to the system under test, wherein τ2 is the time constant ranging from 10 μs to 20 μs, and τ3 is the time variable;
The fourth step: Converting the 1H magnetic resonance signal of the system under test into a signal detectable by magnetic resonance instrument with a 90° pulse with the phases of x, y, −x, −y, and then collecting the signals.
In Step 2 of the present invention, the 1H two-dimensional relaxation signal of the targeted sample can be obtained by controlling the variables in the pulse block or the composite pulses containing 1H spin-echo function and the variables in the pulse block or the composite pulses containing T1 filter function in the pulse sequence.
In Step 3 of the present invention, fn(x,y) is obtained by normalizing the signal intensity of the above-mentioned two-dimensional relaxation signal f(x,y); the fingerprint spectrum can be obtained by subtracting the reference function F(x,y) from fn(x,y); the said reference function F(x,y) is obtained by designing according to the 1H relaxation properties of the target sample, or performing surface fitting of fn(x,y), or averaging Fm(x, y), m=1, 2, . . . , i, which is acquired from surface fitting of the multiple two-dimensional relaxation signals.
In the present invention, when comparing the fingerprint spectrum for species identification and quality detection of liquid-like samples of the same type, the same reference function is used in the generation process of fingerprint spectrum in Step 3 for those belonging to the same sample type but in different qualities.
In the present invention, when using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of corn germ oil, the two-dimensional relaxation signal of corn germ oil, f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of corn germ oil is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of peanut oil, the two-dimensional relaxation signal of peanut oil f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of peanut oil is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of soybean oil, the two-dimensional relaxation signal of soybean oil f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of soybean oil is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of linseed oil, the two-dimensional relaxation signal of linseed oil f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of linseed oil is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of olive oil, the two-dimensional relaxation signal of olive oil f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values and n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of olive oil is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of the commercially available cow milk, the two-dimensional relaxation signal of the commercially available cow milk f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is a set of cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of the commercially available cow milk is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of the commercially available goat milk, the two-dimensional relaxation signal of the commercially available goat milk f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of the commercially available goat milk is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of pig hide gelatin, the two-dimensional relaxation signal of pig hide gelatin f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of pig hide gelatin is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of cow hide gelatin, the two-dimensional relaxation signal of cow hide gelatin f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of cow hide gelatin is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of Liaoning scorpion powder solution, the two-dimensional relaxation signal of Liaoning scorpion powder solution f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values and n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of Liaoning scorpion powder solution is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of Shanxi scorpion powder solution, the two-dimensional relaxation signal of Shanxi scorpion powder solution f(τ1, n) can be obtained by fixing τ3 and changing τ1 and n, wherein τ1 is a set of time values, n is cycle number, f(τ1, n) is the signal intensity corresponding to τ1 and n; the t2 distribution fingerprint spectrum of Shanxi scorpion powder solution is obtained by subtracting the selected reference function F(x, y) from the normalized f(τ1, n).
In the present invention, when using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of corn germ oil, the two-dimensional relaxation signal of corn germ oil fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of corn germ oil is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of peanut oil, the two-dimensional relaxation signal of peanut oil fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of peanut oil is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of soybean oil, the two-dimensional relaxation signal of soybean oil fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number. fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of soybean oil is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of linseed oil, the two-dimensional relaxation signal of linseed oil fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of linseed oil is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of olive oil, the two-dimensional relaxation signal of olive oil fa(τ3, n) can be obtained by fixing τ1 and meanwhile changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of olive oil is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% water by weight, the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% water by weight fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% water by weight is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% lard by weight, the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% lard by weight fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% lard by weight is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% beef tallow by weight, the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% beef tallow by weight fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% beef tallow by weight is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n); and/or,
When using the above-mentioned pulse sequence to acquire the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% butter by weight, the two-dimensional relaxation signal of the corn germ oil sample mixed with 1% butter by weight fa(τ3, n) can be obtained by fixing τ1 and changing τ3 and n, wherein τ3 is a set of time values, n is cycle number, fa(τ3, n) is the signal intensity corresponding to τ3 and n; the t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% butter by weight is obtained by subtracting the selected reference function Fa(x, y) from the normalized fa(τ3, n).
There are two important technical features in the present invention:
1. The above-mentioned pulse sequence and its data acquisition scheme—this is a newly designed pulse sequence and data acquisition scheme, which can be used to measure two-dimensional data containing 1H T1 and T2 relaxation properties of edible oils or other liquid-like samples, and amplify 1H T1 and T2 relaxation differences of different edible oils or other liquid-like samples;
2. Data processing method (subtracting the reference function from two-dimensional data to obtain the fingerprint spectrum and its variants)—this is a newly developed data processing method. The fingerprint spectrum obtained by this method can be used as a standard to distinguish different types of edible oils or other liquid-like samples, and its digital form is very suitable for constructing the big data of edible oils or other liquid-like samples and authenticity judgments based on artificial intelligence.
The present invention has innovative ideas which are different from previous inventions and documentary works:
1. Based on the feature that 1H T1 and T2 relaxation properties are different in edible oils or other liquid-like samples, the method in the present invention breaks through the limitation of using only 1H T2 relaxation to detect and identify edible oils or other liquid-like samples in previous works, meanwhile, innovatively proposes to amplify 1H T1 and T2 relaxation differences of different edible oils or other liquid-like samples by measuring two-dimensional 1H T1 and T2 relaxation data containing relaxation properties of edible oils or other liquid-like samples, thus realizing detection and identification of different edible oils or other liquid-like samples;
2. Based on the above-mentioned idea, a new pulse sequence, and the corresponding data acquisition method are developed to be used to amplify 1H T1 and T2 relaxation differences of different edible oils or other liquid-like samples;
3. A “fingerprint spectrum” containing 1H T1 and T2 relaxation properties of edible oil or other liquid-like samples is established. The fingerprint spectrum can be used as a standard to distinguish different types of edible oils or other liquid-like samples, meanwhile, its digital form is very suitable for constructing the big data of edible oils or other liquid-like samples and authenticity judgments based on artificial intelligence.
4. The method in the present invention can be used in high-resolution nuclear magnetic instruments and low-field magnetic resonance relaxometry, overcoming the dependence of patent CN108982570A on nuclear magnetic resonance signal resolution. At the same time, the measurement of 1H T1 and T2 relaxation properties overcomes low discrimination caused by only measuring 1H T2 relaxation measurement in traditional methods.
5. Compared with the traditional chromatographic, mass spectrometry and optical spectroscopy, the method in the present invention can realize a non-destructive sample testing without sample pretreatment. The method in the present invention can be implemented on a low-field magnetic resonance instrument which can realize rapid on-site detection by moving on board.
In the existing reports, there are two methods for detecting and identifying edible oil by using nuclear magnetic resonance. One is to detect and identify edible oils by using high-resolution magnetic resonance spectroscopy (the existing patent CN108982570A). The principle of this method is based on the recognition of molecular signals of edible oils in a high-resolution magnetic resonance spectrum, thus it relies on a high-resolution nuclear magnetic resonance spectrometer. Because a high-resolution nuclear magnetic resonance spectrometer is usually too bulky to move, the method described in the patent CN108982570A usually cannot achieve quick on-site detection of edible oils. Meanwhile, in the practical operation, the method described in the patent CN108982570A is not easy to effectively distinguish the different edible oils (
The following examples are given to further illustrate the specific solutions of the present invention. The implementation of the present invention process, including the conditions, experimental methods, etc., are common knowledge and known knowledge in the field. The present invention has no special limitations. Meanwhile, the embodiments are only used to illustrate the present invention and not to limit the scope of the present invention.
The invention discloses a method based on nuclear magnetic resonance technology that can carry out species identification and quality detection of liquid-like samples. The present invention applies a specially designed combined pulse sequence to liquid-like sample to obtain a two-dimensional relaxation signal containing the 1H T1 and T2 relaxation characteristics of the sample, which breaks through the weakness of the traditional methods that only uses 1H T2 and cannot effectively distinguish different liquid-like samples. From the two-dimensional relaxation signal, the fingerprint of the liquid-like sample can be established. The fingerprint spectrum is related to the essential characteristics of the liquid-like sample, and can be used as a standard for distinguishing a specific liquid-like sample from the others. Meanwhile, the digital form of the fingerprint spectrum obtained by the present invention is very suitable for constructing the big data of a sample and the quality detection and authenticity judgment of a sample based on artificial intelligence. This method has the characteristics of no need for pretreatment of the test sample, and non-destructive testing of the test object. It also has the advantages of convenience and quickness, strong operability, good stability and reproducibility, etc., and can be used for the identification of a variety of fluid samples. And quality inspection, has a wide range of application value.
The main steps of the implementation process are as follows:
Step 1: Designing a pulse sequence that includes pulse block or composite pulses containing 1H spin-echo function and pulse block or composite pulses containing T1 filter function. The scheme of pulse sequence is referred to
Step 2: Applying the pulse sequence obtained from Step 1 to the targeted liquid-like samples to obtain their 1H two-dimensional relaxation signals of the targeted liquid-like samples;
Step 3: Converting the obtained 1H two-dimensional relaxation signals from Step 2 into the fingerprint spectrum of the targeted liquid-like samples, so as to be used for species identification and quality detection of edible oil or the liquid-like samples.
Wherein, the said liquid-like samples refer to liquid and gel substances with a certain fluidity, including edible oil, cow and goat milk, donkey-hide gelatin, scorpion powder solution, yogurt, beverages, oils in general, etc.
In Step 1 of the present invention, the pulse sequence comprises the following designs and the sub-steps:
Step 1-1: Using a pulse block or composite pulses to excite 1H magnetic resonance signal of the system under test; Step 1-2: Applying the pulse block or the composite pulses containing 1H spin-echo function to the system under test, and the pulse block or the composite pulses may contain one or more variables; Step 1-3: Applying the pulse block or the composite pulses containing 1H T1 filter function to the system under test, and the pulse block or the composite pulses may contain one or more variables; Step 1-4: Converting the 1H magnetic resonance signal of the targeted samples into a signal detectable by the magnetic resonance instrument through the pulse block or the composite pulses, and then collecting the signals.
In Step 2 of the present invention, the 1H two-dimensional relaxation signal of the targeted sample can be obtained by controlling the variables in the pulse block or the composite pulses containing the 1H spin-echo function and the variables in the pulse block or the composite pulses containing T1 filter function in the pulse sequence. Through the design of the variables, different types of two-dimensional relaxation signal can be obtained.
In Step 3 of the present invention, fn(x,y) is obtained by normalizing the signal intensity of the above-mentioned two-dimensional relaxation signal f(x,y); the fingerprint spectrum can be obtained by subtracting the reference function F(x,y) from fn(x,y); the reference function F(x,y) is obtained by designing according to the 1H relaxation properties of the target sample, or performing surface fitting of fn(x,y), or averaging Fm(x, y), m=1, 2, . . . , i, which is acquired from surface fitting of the multiple two-dimensional relaxation signals.
In the present invention, when comparing the fingerprint spectrum for species identification and quality detection of liquid-like samples of the same type, the same reference function is used in the generation process of fingerprint spectrum in Step 3 for those belonging to the same sample type but in different quality.
There is some sample preparation processes in the examples. The methods and steps of the sample preparation processes are well-known in the field.
Sample: a commercially available corn germ oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing two-dimensional relaxation surface fn(τ1, n) of corn germ oil. The reference surface function is:
F(x,y)=18.54−17.11·x−6.667·y+6.41·x2+4.15·x·y+0.3606·y2−1.232·x3−1.024·x2·y−0.08823·x·y2+0.04467·y3+0.1212·x4+0.1138·x3·y+0.02533·x2·y2−0.02851·x·y3+0.003422·y4−0.00484·x5−0.004637·x4·y−0.002608·x3·y2+0.001928·x2·y3+0.001074·x·y4+0.0000372·y5
The t2 distribution fingerprint spectrum of the corn germ oil (
Sample: a commercially available peanut oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the peanut oil:
F(x,y)=18.54−17.11·x−6.667·y+6.41·x2+4.15·x·y+0.3606·y2−1.232·x3−1.024·x2·y−0.08823·x·y2+0.04467·y3+0.1212·x4+0.1138·x3·y+0.02533·x2·y2−0.02851·x·y3+0.003422·y4−0.00484·x5−0.004637·x4·y−0.002608·x3·y2+0.001928·x2·y3+0.001074·x·y4+0.0000372·y5
The t2 distribution fingerprint spectrum of the peanut oil (
Sample: a commercially available .soybean oil
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F(x,y)=18.54−17.11·x−6.667·y+6.41·x2+4.15·x·y+0.3606·y2−1.232·x3−1.024·x2·y−0.08823·x·y2+0.04467·y3+0.1212·x4+0.1138·x3·y+0.02533·x2·y2−0.02851·x·y3+0.003422·y4−0.00484·x5−0.004637·x4·y−0.002608·x3·y2+0.001928·x2·y3+0.001074·x·y4+0.0000372·y5
The t2 distribution fingerprint spectrum of the soybean oil (
Sample: a commercially available linseed oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F(x,y)=18.54−17.11·x−6.667·y+6.41·x2+4.15·x·y+0.3606·y2−1.232·x3−1.024·x2·y−0.08823·x·y2+0.04467·y3+0.1212·x4+0.1138·x3·y+0.02533·x2·y2−0.02851·x·y3+0.003422·y4−0.00484·x5−0.004637·x4·y−0.002608·x3·y2+0.001928·x2·y3+0.001074·x·y4+0.0000372·y5
The t2 distribution fingerprint spectrum of the linseed oil (
Sample: a commercially available olive oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F(x,y)=18.54−17.11·x−6.667·y+6.41·x2+4.15·x·y+0.3606·y2−1.232·x3−1.024·x2·y−0.08823·x·y2+0.04467·y3+0.1212·x4+0.1138·x3·y+0.02533·x2·y2−0.02851·x·y3+0.003422·y4−0.00484·x5−0.004637·x4·y−0.002608·x3·y2+0.001928·x2·y3+0.001074·x·y4+0.0000372·y5
The t2 distribution fingerprint spectrum of the olive oil (
Sample: a commercially available corn germ oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the two-dimensional relaxation surface fn(τ3, n) of the corn germ oil. The reference surface function is:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the corn germ oil (
Sample: a commercially available peanut oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the peanut oil (
Sample: a commercially available soybean oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the soybean oil (
Sample: a commercially available linseed oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the linseed oil (
Sample: a commercially available olive oil.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the olive oil (
Sample: a commercially available corn germ oil sample mixed with 1% water by weight.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% water by weight (
Sample: a commercially available corn germ oil sample mixed with 1% lard by weight.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% lard by weight (
Sample: a commercially available corn germ oil sample mixed with 1% beef tallow by weight.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was that of the corn germ oil:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% beef tallow by weight (
Sample: a commercially available corn germ oil sample mixed with 1% butter by weight.
NMR Instrument: Bruker AVANCE III 500 NMR spectrometer. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting the normalized two-dimensional relaxation surface fn(τ3, n) of corn germ oil. The reference surface function is:
F
a(x,y)=−2.281+2.317·x−0.4775·y−0.9071·x2−0.001151·x·y+0.2854·y2+0.1461·x3+0.09387·x2·y−0.1417·x·y2−0.007143·y3−0.00962·x4−0.01736·x3·y+0.01505·x2·y2+0.00902·x·y3−0.002398·y4+0.0001928·x5+0.0008389·x4·y−0.0003571·x3·y2−0.0006947·x2·y3+0.000005745·x·y4+0.00008974·y5
The t1-t2 correlation fingerprint spectrum of the corn germ oil sample mixed with 1% butter by weight (
Sample: a commercially available cow milk 1.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average value of two-dimensional relaxation surface fn(τ1, n) of the two commercially available cow milks and the goat milk. The reference surface function is:
F(x,y)=−0.4201−0.2412·x−0.7473·y−0.0085·x2−0.2879·x·y−0.4974·y2+0.0201·x3+0.0121·x2·y−0.0027·x·y2+0.0213·y3−0.0195·x4−0.0061·x3·y+0.0379·x2·y2+0.1243·x·y3+0.1041·y4−0.0102·x5−0.0052·x4·y−0.0047·x3·y2+0.0137·x2·y3+0.0411·x·y4+0.0214·y5
The t2 distribution fingerprint spectrum of the commercially available cow milk 1 (
Sample: a commercially available cow milk 2.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average value of two-dimensional relaxation surface fn(τ1, n) of the two commercially available cow milks and the goat milk. The reference surface function is:
F(x,y)=−0.4201−0.2412·x−0.7473·y−0.0085·x2−0.2879·x·y−0.4974·y2+0.0201·x3+0.0121·x2·y−0.0027·x·y2+0.0213·y3−0.0195·x4−0.0061·x3·y+0.0379·x2·y2+0.1243·x·y3+0.1041·y4−0.0102·x5−0.0052·x4·y−0.0047·x3·y2+0.0137·x2·y3+0.0411·x·y4+0.0214·y5
The t2 distribution fingerprint spectrum of the commercially available cow milk 2 (
Sample: a commercially available goat milk.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average value of two-dimensional relaxation surface fn(τ1, n) of the commercially available cow milks and goat milk. The reference surface function is:
F(x,y)=−0.4201−0.2412·x−0.7473·y−0.0085·x2−0.2879·x·y−0.4974·y2+0.0201·x3+0.0121·x2·y−0.0027·x·y2+0.0213·y3−0.0195·x4−0.0061·x3·y+0.0379·x2·y2+0.1243·x·y3+0.1041·y4−0.0102·x5−0.0052·x4·y−0.0047·x3·y2+0.0137·x2·y3+0.0411·x·y4+0.0214·y5
The t2 distribution fingerprint spectrum of the commercially available goat milk (
Sample: a pig hide gelatin.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average value of two-dimensional relaxation surface fn(τ1, n) of the pig hide gelatin and cow hide gelatin. The reference surface function is:
F(x,y)=−0.2675−0.1528·x−0.3691·y−0.0574·x2−0.2720·x·y−0.4243·y2−0.0024·x3−0.0270·x2·y−0.0552·x·y2−0.0768·y3+0.00054·x4+0.0133·x3·y+0.0302·x2·y2+0.1050·x·y3+0.0899·y4−0.00077·x5+0.0025·x4·y+0.0071·x3·y2+0.0280·x2·y3+0.0402·x·y4+0.0299·y5
The t2 distribution fingerprint spectrum of the pig hide gelatin sample (
Sample: a cow hide gelatin.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average of two-dimensional relaxation surface fn(τ1, n) of the pig hide gelatin and cow hide gelatin. The reference surface function is:
F(x,y)=−0.2675−0.1528·x−0.3691·y−0.0574·x2−0.2720·x·y−0.4243·y2−0.0024·x3−0.0270·x2·y−0.0552·x·y2−0.0768·y3+0.00054·x4+0.0133·x3·y+0.0302·x2·y2+0.1050·x·y3+0.0899·y4−0.00077·x5+0.0025·x4·y+0.0071·x3·y2+0.0280·x2·y3+0.0402·x·y4+0.0299·y5
The t2 distribution fingerprint spectrum of the cowhide gelatin sample (
Sample: Liaoning scorpion powder solution. The weight ratio between the scorpion powder and the water is 1:1.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average value of the two-dimensional relaxation surface fn(τ1, n) of the Liaoning scorpion powder solution and the Shanxi scorpion powder solution. The reference surface function is:
F(x,y)=−0.2675−0.1528·x−0.3691·y−0.0574·x2−0.2720·x·y−0.4243·y2−0.0024·x3−0.0270·x2·y−0.0552·x·y2−0.0768·y3+0.00054·x4+0.0133·x3·y+0.0302·x2·y2+0.1050·x·y3+0.0899·y4−0.00077·x5+0.0025·x4·y+0.0071·x3·y2+0.0280·x2·y3+0.0402·x·y4+0.0299·y5
The t2 distribution fingerprint of Liaoning scorpion powder solution (
Sample: Shanxi scorpion powder solution. The weight ratio between the scorpion powder and the water is 1:1.
NMR Instrument: VTMR20-010V-I relaxometry, Niumag Corp., Ltd., Shanghai, China. The B0 field is 0.5±0.05 T and the 1H Larmor frequency is 21.3 MHz. The experimental temperature is room temperature.
Method: The pulse sequence used in this experiment is shown in
In this example, the reference surface function was obtained by fitting and normalizing the average value of two-dimensional relaxation surface fn(τ1, n) of the Liaoning scorpion powder solution and the Shanxi scorpion powder solution. The reference surface function is:
F(x,y)=−0.2675−0.1528·x−0.3691·y−0.0574·x2−0.2720·x·y−0.4243·y2−0.0024·x3−0.0270·x2·y−0.0552·x·y2−0.0768·y3+0.00054·x4+0.0133·x3·y+0.0302·x2·y2+0.1050·x·y3+0.0899·y4−0.00077·x5+0.0025·x4·y+0.0071·x3·y2+0.0280·x2·y3+0.0402·x·y4+0.0299·y5
The t2 distribution fingerprint of Shanxi scorpion powder solution (
The present invention has the following features which are different from previous methods and technologies:
1. Based on the feature that the 1H T1 and T2 relaxation properties are different in different edible oils or other liquid-like samples, the method in the present invention breaks through the limitation of previous works using only 1H T2 relaxation to detect and identify edible oils or other liquid-like samples, innovatively proposes to amplify 1H T1 and T2 relaxation differences of edible oils or other liquid-like samples by measuring two-dimensional 1H T1 and T2 relaxation data containing relaxation properties of edible oils or other liquid-like samples, thus realize the detection and identification of edible oils or other liquid-like samples;
2. Based on the above idea, a new pulse sequence and the corresponding data acquisition method are developed to be used to amplify the 1H T1 and T2 relaxation differences of different edible oils or other liquid-like samples;
3. A ‘fingerprint spectrum’ containing the 1H T1 and T2 relaxation properties of edible oils or other liquid-like samples is constructed. The fingerprint can be used as a standard to distinguish different types of edible oil or other liquid-like samples, meanwhile, its digital form is very suitable for constructing the big data of edible oils or other liquid-like samples and the authenticity judgments based on artificial intelligence.
4. The method in the present invention can be used om high-resolution nuclear magnetic instruments and low-field magnetic resonance relaxometry, overcoming the dependence of the patent CN108982570A on the nuclear magnetic resonance signal resolution. At the same time, the measurement of 1H T1 and T2 relaxation properties overcomes the low discrimination of the traditional method caused by only measuring the individual 1H T2 relaxation.
5. Compared with the traditional chromatographic, mass spectrometry and the optical spectroscopy, the method in the present invention can realize a non-destructive sample testing without sample pretreatment. The method can be implemented on a low-field magnetic resonance instrument which can achieve a rapid on-site detection by moving on board.
It should be understood that the term “and/or” used in the present invention is only to describe a relationship between associated objects, which means that there can be three relationships, i.e A and/or B can mean three conditions: A alone, A and B, and B alone. In addition, the character “/” in the present invention generally indicates that the associated objects before and after being in an ‘or’ relationship.
The protection of the present invention is not limited to the following embodiments. Without departing from the spirit and scope of the idea of the invention, all changes and advantages that can be thought of by a person skilled in the field are included in the present invention and are protected by the attached claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 202010345962.6 | Apr 2020 | CN | national |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/CN2021/089234 | 4/23/2021 | WO |
