Grasas y Aceites 72 (3)
July-September 2021, e426
ISSN-L: 0017-3495
https://doi.org/10.3989/gya.0662201

# Quality detection of tea oil by 19F NMR and 1H NMR

## Determinación de la calidad del aceite de té mediante19F RMN y1H RMN

T. Liu

School of Environmental and Chemical Engineering, Shanghai University, 351, Nanchen Road, Shanghai, 200444, People’s Republic of China

https://orcid.org/0000-0002-8975-4989

T.M. Olajide

School of Environmental and Chemical Engineering, Shanghai University, 351, Nanchen Road, Shanghai, 200444, People’s Republic of China

https://orcid.org/0000-0003-4751-8715

W. Wang

Anhui Xueyan Tea Oil Co., Ltd, Standardized Factory Building, Jintongtong Pioneer Park, Jinzhai County Modern Industrial Park, Lu’an City, Anhui Province, 237000, People’s Republic of China

https://orcid.org/0000-0002-0617-8307

Z. Cheng

Xuhui District Donger Primary School, 2, Wanping South Road, Shanghai, 200030, People’s Republic of China

https://orcid.org/0000-0003-2059-7396

Q. Cheng

Anhui Xueyan Tea Oil Co., Ltd, Standardized Factory Building, Jintongtong Pioneer Park, Jinzhai County Modern Industrial Park, Lu’an City, Anhui Province, 237000, People’s Republic of China

https://orcid.org/0000-0001-9234-6962

X.C. Weng

School of Environmental and Chemical Engineering, Shanghai University, 351, Nanchen Road, Shanghai, 200444, People’s Republic of China
School of Life Sciences, Shanghai University, 351, Nanchen Road, Shanghai, 200444, People’s Republic of China

https://orcid.org/0000-0003-2047-1654

SUMMARY

The nuclear magnetic resonance (NMR) technique was applied to monitor the quality of tea oil herein. The adulteration of virgin tea oil was monitored by 19F NMR and 1H NMR. The 19F NMR technique was used as a new method to detect the changes in quality and hydroperoxide value of tea oil. The research demonstrates that 19F NMR and 1H NMR can quickly detect adulteration in tea oil. High temperature caused a decrease in the ratio D and increase in the total diglyceride content. Some new peaks belonging to the derivatives of hydroperoxides appeared at δ-108.21 and δ-109.05 ppm on the 19F NMR spectrum when the oil was autoxidized and became larger when the hydroperoxide value increased. These results have great significance in monitoring the moisture content, freshness and oxidation status of oils and in detecting adulteration in high priced edible oils by mixing with cheap oils.

KEYWORDS:
1H NMR; 19F NMR; Hydroperoxides; Quality detection; Tea oil
RESUMEN

En este trabajo se utiliza la técnica de resonancia magnética nuclear (RMN) para controlar la calidad del aceite de té. La adulteración del aceite de té virgen se controló mediante las técnicas de 19F RMN y 1H RMN. La técnica de 19F RMN se utilizó como un nuevo método para detectar los cambios en la calidad y el índice de hidroperóxido del aceite de té. La investigación demuestra que las técnicas 19F RMN y 1H RMN pueden detectar rápidamente la adulteración del aceite de té. La alta temperatura provoca una disminución en la proporción D y un aumento en el contenido total de diglicéridos. Algunos picos nuevos, pertenecientes a derivados de hidroperóxidos, aparecieron a δ-108,21 y δ-109,05 ppm en el espectro de 19F RMN cuando el aceite se autoxidaba e incrementaban cuando aumentaba el índice de hidroperóxido. Estos resultados tienen gran importancia en el seguimiento del contenido de humedad, de la frescura y del estado de oxidación de los aceites y en la detección de la adulteración de aceites comestibles de alto valor con aceites baratos mediante el uso de 19F RMN y 1H RMN.

PALABRAS CLAVE:
1H RMN; 19F RMN; Aceite de té; Detección de calidad; Hidroperóxidos

Submitted: 06  June  2020; Accepted: 26  August  2020; Published online: 24 September 2021

Citation/Cómo citar este artículo: Liu T, Olajide TM, Wang WJ, Cheng ZJ, Cheng Q, Weng XC. 2021. Quality detection of tea oil by 19F NMR and 1H NMR. Grasas y Aceites 72 (3), e426. https://doi.org/10.3989/gya.0662201

CONTENT

### 1. INTRODUCTION

Camellia oleifera, commonly called oil tea tree, belongs to the Theaceae family and has been widely cultivated in China for a long time (). Camellia oil is widely referred to as tea oil and is commonly utilized for cooking in China (). It is rich in unsaturated fatty acids such as linolenic, linoleic and oleic acids (). Its oleic acid content is similar to that found in olive oil, reaching up to 70%. In addition, there are also some unsaponifiable compounds such as tocopherols, squalene, phytosterols, and flavonoids present in tea oil (; ). These nutrients can easily be digested and absorbed by the human body, and are beneficial to lowering cholesterol, preventing and treating hypertension, and cardiovascular diseases (; ). Tea oil has been reported to exhibit antioxidant activity (), and is also known as “Oriental Olive Oil”, which has been very much preferred by consumers in China. Among plant oils, the low-temperature and cold-pressed tea oil has a higher price than others because it retains nutrients as much as possible. Thus, some unscrupulous merchants adulterate tea oil with other low-price plant oils in order to make higher profits.

The adulteration of tea oil by other low-cost oils damages consumer interest. There are some certain analytical methods for detecting the adulteration and quality of oils, such as GC-MS, ultraviolet spectroscopy, infrared spectroscopy and nuclear magnetic resonance (NMR) spectroscopy (; ; ). However, the NMR technique, especially 1H NMR (), has become a favorable choice (), owing to its fast and effective approach over the traditional methods like GC-MS. analyzed the degree of unsaturation of combined and free fatty acids in several plant oils (soybean, corn, sunflower, canola, linseed, cottonseed and jatropha) using 1H NMR, which was found to be satisfactory when compared to other conventional methods. used 1H NMR as a fast method to determine soybean oil deterioration during deep frying and discovered that it is similar to the conventional gas chromatography method for analyzing secondary oxidation products. 1H NMR can be also used to determine the chemometric characteristics of extra-virgin olive oil (EVOO) in order to identify the specific compounds responsible for olive oil characteristics (Ingallina et al., 2019). used 1H NMR combined with chemometrics for the rapid detection of adulteration in tea oil, and also confirmed the efficacy of this method in terms of speed and accuracy.

Our laboratory has previously established a new technique for monitoring the quality and adulteration of olive oil by 19F NMR (). used this method combined with 1H NMR to detect EVOO adulteration successfully. As far as we know, no study based on a 19F NMR approach for the detection of the quality of tea oil according to temperature and time changes has been reported. In this work, the quality and adulteration of tea oil using 19F NMR was studied and specifically, the determination of moisture content and the detection of oxidation with temperature were conducted in order to expand the application of NMR techniques for the assessment of plant oil quality.

### 2. MATERIALS AND METHODS

#### 2.1. Chemicals

All solvents were of reagent or analytical grade. Hexafluorobenzene (99%), 4-tert-butylphenol, pyridine and chloroform-d were purchased from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China). The deriving fluorine reagent (4-fluorobenzoyl chloride, purity: 98%) was purchased from Sigma-Aldrich (Shanghai, China). The rest of the reagents used in the experiment were from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China).

#### 2.2. Samples

Two tea oils were prepared in our lab: one was extracted with petroleum ether (bp range 60-90 °C) at room (or low) temperature (TOL) and the other was extracted by the Soxhlet method with petroleum ether (TOS) and then the solvent was removed by a vacuum rotary evaporator in a water bath at 30 °C. Other commercial plant oils were purchased from the local supermarket. The samples were stored at room temperature away from light.

5, 10, 15, 20, 25, 30, 35, 40 and 45% refined tea seed oil were separately added to TOL.

#### 2.3. Sample preparation for NMR analysis

0.1 mL hexafluorobenzene in a stock solution of pyridine and CDCl3 mixed in a ratio of 1:1.5 (v/v) was mixed with 0.5 g 4-tert-butylphenol. Hexafluorobenzene was used as reference material because the chemical shift was observed at δ164.90 ppm n the 19F NMR analysis. And 4-tert-butylphenol was as a quantitative standard in the 19F NMR analysis. 0.1 Gram oil sample was put in a 4 mL centrifuge tube mixed with 0.4 mL stock solution. The resulting solution was transferred to a 5 mm NMR tube and 30 μL of deriving reagent were added. Then the reaction mixture was left to react for 0.5 h in the tube at room temperature away from light. After completion, the 19F NMR spectra of the samples were determined immediately.

20 μL oil sample were dissolved in 0.4 mL CDCl3 with 0.03% trimethylsilane (TMS). The resulting solution was transferred to a NMR tube and then the 1H NMR spectra were recorded.

#### 2.4. NMR spectroscopy experiments

All NMR recordings were conducted on a Bruker AVANCE III HD 600MHz spectrometer, operating at 564 and 600 MHz for the 19F and proton nucleus, respectively. Typical spectral parameters for this 19F NMR experiment were as follows: 90° pulse width = 19.3 μs, sweep width = 100 kHz, relaxation delay = 1 s, memory size = 64 K. 32T transients were accumulated for each spectrum. For all FIDs, line broadening of 0.3 Hz was applied and drift correction was performed prior to Fourier transformation.

Typical spectral parameters for 1H NMR experiment were shown as: 16 scans and 4 dummy scans for each free induction decay, 32K for time domain points with a spectral width of 12.0 ppm, 90° pulse width of 9.0 μs, acquisition time of 2.7 s and relaxation delay of 1.0 s.

#### 2.5. Effect of temperature on tea oil

20 Grams of TOL were divided into two equal parts, one of which was spiked with 0.02% tert-butylhydroquinone (TBHQ) and the other without any treatment. The samples were placed on a Rancimat instrument for heating and oil samples were taken every 4 h at 80, 100, 120, 140 °C for a total of 24 h,

#### 2.6. Oven test, reducing autoxidized tea oil, determination of POV

40 Grams of tea oil were placed in a clean beaker, and the beaker was placed in a 63 ± 1 °C oven. The peroxide value was measured every 5 days, and the peak between δ-108 ~ δ-110 ppm on a 19F NMR spectrum was measured and calculated.

Two grams of tea oil were dissolved at 63 °C in the oven after the 35 days in 30 mL solvent (acetic acid: CHCl3 = 3:2) and reduced by KI for 3 min and then washed with 100 mL water 3 times to remove the acetic acid. Anhydrous sodium sulfate was used to dry moisture, while CHCl3 in the solution was dried on a rotary evaporator.

The method of peroxide value measured by titration () was used according to the following formula:

Where V is the volume of sodium thiosulfate standard solution consumed by the sample, V0 is the volume of sodium thiosulfate standard solution consumed by blank sample, c is the concentration of sodium thiosulfate standard solution, m is the weight of the oil. The formula for the hydroperoxide value between δ-108 and δ-110 ppm measured by 19F NMR is

$n(mmol/100g)=((A/A1)×N)/m×100$

Where: A is the area of the peak between δ-108 and δ-110 ppm, Ai is the area of the peak of 4-tert-butylphenol, N is the millimolar amount of 4-tert-butylphenol, m is the weight of oil.

#### 2.7. GC-MS experiments

The methyl esterification method for samples before GC-MS was carried out. 50 Milligrams of oil sample dissolved in 4 mL n-hexane (chromatographic grade) were put in a tube with stopper. Then 200 μL of 2 M KOH-CH3OH were added and the tube was shaken vigorously for 1 minute for methyl esterification. Then it was left to stand for 5 min to allow solid-liquid separation. One gram of sodium hydrogen sulfate monohydrate was added to the solution to neutralize potassium hydroxide, followed by immediate analysis of the supernatant by GC-MS (, ).

The GC-MS analyses were performed on a Shimadzu GC2010A (Kyoto, Japan) gas chromatography. A Rtx®-Wax capillary column (30 m length, 0.25 mm i.d., and 0.25 μm film consisting of cross-bond polyethylene glycol (Restek)) was used. The conditions of the GC-MS analysis were as follow: column temperature = 140 to 250 °C, rate = 4 °C/min, injection temperature = 220 °C, carrier gas = nitrogen, column flow (nitrogen flow rate) = 1.36 mL/min, injection volume = 0.5 μL, split ratio = 30:1.

Mass spectroscopy conditions: ion source temperature = 230 °C, interface temperature = 280 °C, ionization voltage = 0.2 kv.

The peak area normalization method was used to calculate the relative content.

#### 2.8. Statistical analyses

All statistical analyses were determined using IBM SPSS 22.0. The experiments were performed in duplicate, and values were expressed as mean ± standard deviation (SD).

### 3. RESULTS AND DISCUSSION

#### 3.1. 19F NMR analysis

Our lab previously found a novel method to detect the quality and adulteration of olive oil using the 19F NMR technique (). The main principle of the method is based on the derivatization of the active hydroxy groups like diglycerides (DGs) and water with 4-fluorobenzoyl chloride, and the integration of the appropriate peaks in the 19F NMR spectrum in the MestReNova. The deriving reagent and the intermediate products between 4-fluorobenzoyl chloride and the stock solution peaks were observed at δ-105.05 and δ-103.60 ppm, respectively. The peak at δ-107.45 ppm was attributed to 4-tert-butylphenol, the internal standards in this experiment. The peaks at δ-107.95 and δ-107.86 ppm belong to the α- and β-hydroxyl groups of 1,2-DG and 1,3-DG, respectively. The water peak was attributed to δ-110.42 ppm. The quality of tea oil was analyzed by calculating the characteristic peak appearing on the 19F NMR spectrum.

#### 3.2. Diglyceride content

The DG content is an important indicator of the quality of oil, and it can be detected by 19F NMR. The contents of 1,2-DGs, 1,3-DGs, total diglycerides (TDGs) and the ratio D (1,2-DGs to TDGs) in tea oil and other plant oils are summarized in . After comparing five different kinds of tea oils, it was found that the fresh TOL had a high D ratio (0.77) and lower TDGs (1.45%). The content of TDGs (2.14%) and D ratio (0.27) of refined tea oil showed the opposite trend. Clearly, the D ratio is a significant indicator for judging the quality of oils because it usually occurs from the isomerization of 1,2-DGs to 1,3-DGs during oil storage and refining. A study by Vigli () showed that the D ratio of extra virgin olive oil (EVOO) freshly extracted from normal mature olives should be close to 1. Although the D ratio in all oil samples decreases with storage time, the closer the D ratio is to 1, the fresher the oil. When two kinds of tea oil based on extraction temperature were compared, TOL had a higher D ratio, indicating it has better quality and freshness. Usually, commercial tea oil is inevitably affected by temperature during the production process, especially the refining process. shows the D ratio of TOL (0.77) > TOS (0.57) > tea oil from the supermarket (0.29, 0.32) > refined tea oil (0.27). This phenomenon is supported by the present results, that is, the quality of tea oil is affected by temperature and freshness. EVOO is a high quality cold-pressed plant oil because of its nutritional value and health benefits. The D ratio of TOL in our lab is much higher than all EVOO samples detected in our previous papers (; ) because all EVOO samples were imported from Spain and Italy to Shanghai at least 1.5 years after being prepared. In addition, the 1,3-DG content of all the other plant oils was higher than 1,2-DGs, and the D ratio D was about 0.3 which is in agreement with the study by . This indicates that the extraction temperature has a great influence on the quality and freshness of plant oil. The lower the extraction temperature is, the better the quality of the oil.

TABLE 1.  Compositional parameters of tea oil and some other plant oils determined by 19F NMR Spectroscopy.
Sample1,3-DGs1,2-DGsTotal DGsD
TOL0.33 ± 0.011.12 ± 0.101.45 ± 0.110.77 ± 0.01
TOS0.63 ± 0.060.85 ± 0.151.49 ± 0.210.57 ± 0.02
tea oil from the supermarket 11.31 ± 0.070.54 ± 0.011.86 ± 0.060.29 ± 0.02
tea oil from the supermarket 21.30 ± 0.090.62 ± 0.021.92 ± 0.110.32 ± 0.01
Refined tea oil1.57 ± 0.060.57 ± 0.012.14 ± 0.070.27 ± 0.01
Extra virgin olive oil0.58 ± 0.100.68 ± 0.001.26 ± 0.100.54 ± 0.04
Soybean oil0.24 ± 0.060.10 ± 0.020.34 ± 0.080.31 ± 0.01
Refined soybean oil0.67 ± 0.030.34 ± 0.021.01 ± 0.010.34 ± 0.02
Rapeseed oil2.02 ± 0.220.80 ± 0.132.82 ± 0.340.28 ± 0.01
Palm oil3.01 ± 0.141.09 ± 0.074.10 ± 0.100.27 ± 0.03
Corn oil2.16 ± 0.051.22 ± 0.113.38 ± 0.140.36 ± 0.02

a Data are expressed as mean ± standard deviation (n=2).
b TOL, tea oils extracted at low (room) temperature; TOS tea oils extracted by Soxhlet method.

#### 3.3. Moisture content

Moisture content is also a significant indicator for judging the quality of oils (). In order to study the moisture content of tea oil by 19F NMR, tea oil was placed in the oven at 63 °C until its mass was constant, then cooled to room temperature. 0, 0.01, 0.02, 0.04, 0.06, 0.08 and 0.1% distilled water was added to the tea oil, and shaken vigorously. 19F NMR was subsequently used to detect the moisture content. The moisture contents in the tea oil were detected at 0, 0.008 ± 0.002, 0.015 ± 0.003, 0.042 ± 0.005, 0.055 ± 0.002, 0.060 ± 0.004, 0.059 ± 0.005%, respectively by 19F NMR. The water contents detected by 19F NMR agree well with those added. The detection of moisture content is usually done together with volatile matters in oils according to the official methods of AOAC (1997, method Cd 8b-90). But in some cases, volatile matters are desired flavors for some edible oils, such as EVOO, or Chinese traditional ground sesames oil. So 19F NMR can directly detect the specific moisture contents in oils. As expected, the solubility of moisture in oil is very low. When the content of added moisture reaches the saturated state of oil dissolution, the excess water will layer from the oil quickly and easily. According to the results of the moisture content detected by 19F NMR, when the content of added water reaches 0.06%, the moisture content becomes saturated in the oil. When over 0.06% water is added to oil, high deviation happens.

#### 3.4. Adulteration of tea oil with refined tea oil

Refined tea oil has the same fatty acid composition of tea oil prepared at a low temperature, but it is much cheaper, so some unscrupulous merchants add refined tea oil to tea oil to make more profit. In addition, there are some minor nutritional components in tea oil that are lost during refining. Some researchers have suggested that the D ratio can be utilized as an index to distinguish different grades of the same kind of oil such as olive oil (). Hence, using 19F NMR to detect adulteration in cold-extracted tea oil with refined tea oil is feasible.

In , the contents of 1,3-DGs, TDGs increased and D ratio decreased with the adulteration level. The D ratio and the adulteration level showed good correlation (r = 0.9653). It can be seen that it is feasible to use 19F NMR to detect the incorporation of refined tea oil into TOL. Therefore, the D ratio is a key parameter for determining whether TOL is adulterated with refined tea seed oil or not. The higher the D ratio is, the fresher the tea oil.

FIGURE 1.  1,2-DGs (%), 1,3-DGs (%), TDGs content (%), D ratio of the adulteration of tea oil with refined tea oil determined by 19F NMR. Values are mean ± standard deviation (n=2).

#### 3.5. Adulteration of tea oil with other plant oils

##### 3.5.1. Determination of fatty acid composition

1H NMR has been demonstrated as a method for determining the fatty acid composition of oil (). The contents in unsaturated fatty acids (oleic, linoleic and linolenic acids), saturated fatty acids (SFAs) and squalene in plant oils can be obtained by calculating the integral peak area on the 1H NMR spectrum (). The assignment of fatty acid signals has been established by . According to the various signal intensities appearing in the 1H NMR spectra, the peak at δ1.68 ppm in the 1H NMR spectrum of tea oils belongs to squalene. In comparison with another study on the 1H NMR spectrum of squalene, the peak was identified as methyl protons belonging to the CH3-17 and CH3-29 of squalene (; ).

FIGURE 2.  600 MHz 1H NMR spectrum of Camellia Oil.

According to , all spectra of different plant oils have similar shape but different peak intensities. So according to the diversity of the fatty acid composition of plant oils, the detection of adulteration in tea oil can be carried out. Then the 1H NMR technique was used to detect the adulteration of tea oil with other plant oils, and GC-MS served as a standard method.

##### 3.5.2. Adulteration of tea oil with soybean oil

Soybean oil, which is much cheaper than TOL and may be used to adulterate TOL, has a different fatty acid composition compared to tea oil. As shown in , in addition to SFAs, the other three parameters, linolenic, linoleic, and oleic acids are all in good agreement with the adulteration level. With an increase in the adulteration level, the contents of linolenic and linoleic acids also increased, whereas the content of oleic acid decreased. GC-MS is a traditional method for detecting oleic acid content in adulterated TOL. The sensitivity and veracity of 1H NMR in determining fatty acid composition can be compared to GC-MS. When the adulteration level reached 45%, the linolenic, linoleic and oleic acid contents according to 1H NMR were 2.63, 28.12 and 52.17%. The content of linolenic, linoleic and oleic acid by GC-MS were 2.53, 29.10 and 54.85%, respectively. The differences in linolenic, linoleic and oleic acid values were 0.1, 0.98 and 2.68%, which showed about 3.95, 3.37 and 4.89% deviation from the data measured by GC-MS. It can be seen that the fatty acid contents measured by 1H NMR were consistent with GC-MS. So 1H NMR can accurately detect the adulteration in tea oil with soybean oil more rapidly than GC-MS.

FIGURE 3.  Four parameters (oleic, linoleic, linolenic acids and SFAs) in the adulteration of tea oil with other plant oils determined by 1H NMR. (a) adulteration of tea oil with soybean oil, b) adulteration of tea oil with rapeseed oil) (■ linolenic acid, ▲ linoleic acid, ★ oleic acid). Values are mean ± standard deviation (n=2).

Comparing the linear equation of linolenic, linoleic and oleic acids by 1H NMR and GC-MS, it can be found that the contents of linolenic and linoleic acid show the best relationship (R=0.9889) with the adulteration level. According to the linear equation (y = 0.0537x + 0.3111) for linoleic acid, the adulteration level can be accurately calculated.

##### 3.5.3. Adulteration of tea oil with rapeseed oil

Therefore, there is no significant difference between the two methods used for detecting the adulteration of tea oil. This phenomenon may be applied to determining the content of 18-chain fatty acids in soybean oil and rapeseed oil. Rapeseed oil has more 18 carbon chain fatty acids than soybean oil. The more 18 carbon chain fatty acid content present in oils, the more sensitive the 1H NMR detection method is ().

#### 3.6. Effect of temperature on the quality of tea oil

The temperature during extraction, transportation and storage of tea oil has a certain impact on its quality. High temperatures can lead to a decline in the quality of tea oil, which often leads to the loss of some nutrients and speedy autoxidation, isomerization. High temperature can also cause an increase in peroxides in oils. All these influences can be detected by 19F NMR.

##### 3.6.1. The content of TDGs

shows the changes in DG contents and D ratio at 80, 100, 120 and 140 °C within 24 hours in the 19F NMR spectrum when the S/N ratio of the peaks is 8. The D ratio steadily decreased for 16 h and then became stable as heating time increased under 80 °C. Whereas, the D ratio decreased sharply in the first 4 h and then remained stable as temperature increased. The results show that the isomerization of 1,2-DGs to 1,3-DGs will accelerate when temperature rises. But antioxidants do not affect the speed of the isomerization.

FIGURE 4.  The tea oil content of total diglycerides (TDGs) (Columns) and D ratio (Curves) in tea oil and tea oil+tert-butylhydroquinone (TBHQ) heated at 80, 100, 120 and 140 ºC within 24 hours. (■: tea oil; ★: tea oil+TBHQ)
##### 3.6.2. The hydroperoxides

Characteristic peaks at δ-108.21 and δ-109.05 ppm were observed on the 19F NMR spectrum. These peaks are reasoned to belong to the hydroperoxides due to four factors. Firstly, the peak areas of compounds at δ-108.21 and δ-109.05 ppm on the 19F NMR spectrum increased markedly when the heating time increased to 140 °C, as shown in . Secondly, the addition of TBHQ to the oil as a positive control greatly inhibited the increase in the peak areas at δ-108.21 and δ-109.05 ppm during 16 to 24 h because TBHQ is a very strong antioxidant and can effectively retard autoxidation of oils, and inhibit the increase in hydroperoxides. But oils commonly contain some natural antioxidants themselves, so oils with and without the addition of antioxidants can retard autoxidation during the first 12 h. But after 12 h, oil without antioxidant addition consumed its own natural antioxidants, and the amount of hydroperoxides increased more rapidly than the oil with the addition of antioxidants. All natural antioxidants, including the ones added, were consumed completely and the hydroperoxides reached maximum levels, as shown in . Thirdly, POV (including hydroperoxides and other peroxides), as determined by classic titration is very closely correlated (R2 = 0.943), as shown in . Fourthly, it seems that two peaks appear at -109.05 ppm () after tea oil oxidized for 35 days. But after it was reduced by KI in the acetic acid solution, two peaks at -109.05 ppm disappeared completely and two new peaks appeared at -109.78 ppm () and were much more distinguishable than those at -109.05 ppm. This means the hydroperoxides were reduced to related alcohols. The peaks at δ-108.21 ppm showed the same change, disappeared completely and a new peak appeared at δ-109.00 ppm. may explain why the 19F NMR spectrum of autoxidized tea oil appeared in two peaks because the conjugated di-double bonds had a stronger de-shielding effect than the mono-double bond. Also, it is easily explained that two 19F NMR peaks of alcohols are more distinguishable than hydroperoxide ones because the carbon chains of alcohols affect fluorine more than those of hydroperoxides because of one more oxygen atom between fluorine and carbon chains in hydroperoxides.

FIGURE 5.  600 MHz 19F NMR spectrum of tea oil heated at 140 °C for 24 h.
FIGURE 6.  The difference between tea oil and tea oil+tert-butylhydroquinone (TBHQ) as detected by 19F NMR during heating at 140 °C for 24 h. (brown: the spectrum of tea oil; blue: the spectrum of tea oil+TBHQ)
FIGURE 7.  The relationship between the peroxide values (POVs) measured by titration and hydroperoxide values by the 19F NMR technique. Values are mean ± standard deviation (n=2).
FIGURE 8.  The 19F NMR spectra of tea oil stored in the oven at 63 °C for 35 days. (a. Before reduction; b. After reduction by KI in acetic acid solution.)
FIGURE 9.  Autoxidation of unsaturated fatty acids and the reaction of hydroperoxides of unsaturated fatty acids with 4-fluorobenzoyl chloride.

According to the , the peroxide value for tea oil must be lower than 0.25 g/100g. On the basis of the linear equation y = 7.2077x + 0.1244 for peroxide value measured by titration and 19F NMR, the limit of detection for the hydroperoxide value level was 1.93 mmol/100g, according to the 19F NMR method. Over this hydroperoxide value (1.93 mmol/100g), the peroxide value for tea oil exceeds regulations and the quality of tea oil is not up to standard.

### 4. CONCLUSIONS

This is the first time that 19F NMR is used to determine the quality of tea oil, especially the changes in the quality and autoxidation of tea oil. The results from this study demonstrated that 19F NMR and 1H NMR can effectively detect the adulteration of low-temperature extracted tea oil with refined tea oil and other low-price edible oils. The characteristic peaks appearing at δ-108.21 and δ-109.05 ppm on the 19F NMR spectrum belonged to hydroperoxides, and they can be used as an indicator for the determination of the quality and oxidation of tea oil. This phenomenon has great significance for the quality determination of tea oil. Meanwhile, it was found that high temperatures affected the TGDs content and D ratio in tea oil in a short time, and then remained at a certain level within 24 h. In addition, 19F NMR technology can detect long-term dynamic changes of the quality of tea oil. This method is a new, faster and more comprehensive method to determine the quality of tea oil.

### ACKNOWLEDGMENTS

The Authors wish to thank Dr. Yan-Hong Song and Dr. Hong-Mei Deng at the Instrumental Analysis and Research Center of Shanghai University for NMR spectra recording and technical assistance, as well as Anhui Xueyan Tea Oil Co., Ltd. for providing research funds and the tea oil samples.

### REFERENCES

Andrade DF, Mazzei JL, Kaiser CR. 2012. Assessment of Different Measurement Methods Using 1H-NMR Data for the Analysis of the Transesterification of Vegetable Oils. J. Am. Oil Chem. Soc. 89, 619-630. https://doi.org/10.1007/s11746-011-1951-4

Castejón D, Mateos-Aparicio I, Molero MD, Cambero MI, Herrera A. 2014. Evaluation and optimization of the analysis of fatty acid types in edible oils by 1H NMR. Food Anal. Methods 7, 1285-1297. https://doi.org/10.1007/s12161-013-9747-9

Clement H, Renner R. 1977. Studies of the Utilization of High and Low Erucic Acid Rapeseed Oils by the Chick. J. Nutr. 107, 251-260. https://doi.org/10.1093/jn/107.2.251

GB (national test standard). 2018. Oil-tea camellia seed oil. GB/T11765-2018. Int. Organ. Stand., China.

Gurdeniz G, Ozen B. 2009. Detection of adulteration of extra-virgin olive oil by chemometric analysis of mid-infrared spectral data. Food Chem. 116, 519-525. https://doi.org/10.1016/j.foodchem.2009.02.068

Hu L, Toyoda K, Ihara I. 2008. Dielectric properties of edible oils and fatty acids as a function of frequency, temperature, moisture and composition. J. Food Eng. 88, 151-158. https://doi.org/10.1016/j.jfoodeng.2007.12.035

ISO (International Organization for Standarization). 1990. Animal and vegetable fats and oils-Analysis by gas chromatography of methyl esters of fatty acids. ISO 5508:1990 (IDT.). Int. Organ. Stand., Geneva, Switzerland.

ISO (International Organization for Standarization). 2000. Animal and vegetable fats and oils-Preparation of methyl esters of fatty acids. EN ISO 5509:2000 (E). Int. Organ. Stand., Geneva, Switzerland.

Jiang X, Huang R, Wu S. 2018a. Correlations between 1H NMR and conventional methods for evaluating soybean oil deterioration during deep frying. J. Food Meas Charact. 12, 1420-1426. https://doi.org/10.1007/s11694-018-9757-9

Jiang XY, Li C, Chen QQ, Weng XC. 2018b. Comparison of 19F and 1H NMR spectroscopy with conventional methods for the detection of extra virgin olive oil adulteration. Grasas Aceites 69, 249. https://doi.org/10.3989/gya.1221172

Knothe G, Bagby MO, Weisleder D. 1996. Evaluation of the olefinic proton signals in the 1H-NMR spectra of allylic hydroxy groups in long-chain compounds. Chem. Phys. Lipids. 82, 33-37. https://doi.org/10.1016/0009-3084(96)02559-5

Lambelet P, Grandgirard A, Gregoire S. 2003. Formation of modified fatty acids and oxyphytosterols during refining of low erucic acid rapeseed oil. J. Agric. Food Chem. 51, 4284. https://doi.org/10.1021/jf030091u

Lee CP, Yen GC. 2006. Antioxidant activity and bioactive compounds of tea seed (Camellia oleifera Abel.) oil. J. Agric. Food. Chem. 54, 779. https://doi.org/10.1021/jf052325a

Li X, Kong W, Shi W, Shen Q. 2016. A combination of chemometric methods and gc-ms for the classification of edible vegetable oils. Chemometr Intell. Lab. 155, 145-150. https://doi.org/10.1016/j.chemolab.2016.03.028

Mannina L, D’Imperio M, Capitani D, Rezzi S, Aparicio R. 2009. 1H NMR-Based Protocol for the Detection of Adulterations of Refined Olive Oil with Refined Hazelnut Oil. J. Agric. Food Chem. 57, 11550-11556. https://doi.org/10.1021/jf902426b

Memon A. 2011. Phenolic compounds and seed oil characterization of Ziziphus Mauritiana L. fruit grown in Pakistan. FASEB. J. 25, 3515-3524. https://doi.org/10.1002/jcb.23318

Qin S, Rong J, Zhang W, Chen J. 2018. Cultivation history of Camellia oleifera and genetic resources in the Yangtze River Basin. Bio. Sci. https://doi.org/10.17520/biods.2017254

Sacchi R, Addeo F, Paolillo L. 1997. 1H and 13C NMR of virgin olive oil. An overview. Magn. Reson. Chemi. 35, S133-S145. https://doi.org/10.1002/(SICI)1097-458X(199712)35:13<S133::AID-OMR213>3.0.CO;2-K

Santos JS, Escher GB, Marcos DSPJ. 2018. 1H NMR combined with chemometrics tools for rapid characterization of edible oils and their biological properties. Ind. Crop. Prod. 116, 191-200. https://doi.org/10.1016/j.indcrop.2018.02.063

Shi T, Zhu M, Chen Y, Yan X, Chen Q, Wu X, Lin J, Xie M. 2018. 1H NMR combined with chemometrics for the rapid detection of adulteration in camellia oils. Food Chem. 242, 308-315. https://doi.org/10.1016/j.foodchem.2017.09.061

Shi T, Zhu M, Zhou X, Huo X, Long Y, Zeng X, Chen Y. 2019. 1H NMR combined with PLS for the rapid determination of squalene and sterols in vegetable oils. Food Chem. 287, 46-54. https://doi.org/10.1016/j.foodchem.2019.02.072

Stuffins CB, Weatherall H. 1945. Determination of the peroxide value of oils and fats. Analyst. 70, 403-409. https://doi.org/10.1039/AN9457000403

Tu PS, Tung YT, Lee WT, Yen GC. 2017. Protective Effect of Camellia Oil (Camellia oleifera Abel.) against Ethanol-induced Acute Oxidative Injury of the Gastric Mucosa in Mice. J. Agric. Food Chem. 65, 4932-4941. https://doi.org/10.1021/acs.jafc.7b01135

Vigli G, Philippidis A, Spyros A, Dais P. 2003. Classification of Edible Oils by Employing 31P and 1H NMR Spectroscopy in Combination with Multivariate Statistical Analysis. A Proposal for the Detection of Seed Oil A. J. Agric. Food Chem. 51, 5715-5722. https://doi.org/10.1021/jf030100z

Wang Y, Huang S, Shao S. 2012. Studies on bioactivities of tea (Camellia sinensis L.) fruit peel extracts: Antioxidant activity and inhibitory potential against α-glucosidase and α-amylase in vitro. Ind. Crop. Prod. 37, 520-526. https://doi.org/10.1016/j.indcrop.2011.07.031

Weng X, Yun Z, Zhang C. 2018. Comparison of the Characteristics of Two Kinds of Tea Seed Oils: Oil-tea Seed Oil and Green-Tea Seed Oil. J. Food Stud. 7, 56. https://doi.org/10.5296/jfs.v7i1.12289

Xiao H, Yao Z, Peng Q, Ni F, Sun Y, Zhang CX, Zhong ZX. 2016. Extraction of squalene from camellia oil by silver ion complexation. Sep. Purif. Technol. 169, 196-201. https://doi.org/10.1016/j.seppur.2016.05.041

Zhou LL, Li C, Weng XC, Fang XM, Gu ZH. 2015. 19F NMR method for the determination of quality of virgin olive oil. Grasas Aceites 66, e106. https://doi.org/10.3989/gya.0242151

Zhou X, Xu L, Feng S, Jing L, Zhou L, Yang R, Ding C. 2018. Antioxidant effect of hawk tea extracts on camellia oil oxidation during microwave heating. J. Consum Prot. Food S. 13, 1-8. https://doi.org/10.1007/s00003-018-1167-8