Determination of Multi-pesticides Residues in Jasmine Flower and Its Scented Tea

For minor crops such as jasmine, the lack of pesticide registration and maximum residue limits are important issues that need to be solved in order to facilitate trading and ensure food safety. Meanwhile, reliable and quick analytical methods for multi-pesticide residues in these commodities are few, but required by various stakeholders. In this study, a method for detecting twenty-five most frequently used pesticides in jasmine flower and its scented tea by multi-plug filtration cleanup and ultra-high-performance liquid chromatography-tandem mass spectrometry was developed and validated. The cleanup process was optimized and compared with the dispersive solid phase extraction procedure. The method was validated, showing that except for methomyl, recoveries of twenty-five pesticides were 64%–108%, with relative standard deviations (n = 5) of 0.33%–10%. The method was successfully applied to detect pesticide residues in marketed samples. The results showed that some flower and tea samples contained a combination of different pesticide residues.


Introduction
As a large agricultural country, China has a large variety of specialty minor crops of high economic and food value, and these occupy an important role in export trade and agricultural production. One such minor crop is Jasmine (Jasminum sambac (L.) Aiton), a shrub belonging to the genus Jasminum from the Oleaceae family. It was originally from India but is now widely distributed in southern China and around the world (Ray et al. 2014;Lavanya et al. 2016). Jasmine flower can not only be used as ingredients to make perfumes and essential oil because of its strong fragrance, but it can also be used to make floral teas for human consumption Khammee et al. 2021). Furthermore, as a traditional Chinese medicine, it contains a variety of nutrients and active substances with high medicinal and edible values. In particular, it has antispasmodic, antidepressant and anti-aging properties amongst others (Issa et al. 2020). However, pesticides are inevitably used to control plant diseases and insect pests during jasmine cultivation and as such, the plants might be exposed to various pesticides, resulting in the presence of excessive residues in Jasmine-derived products. As jasmine flowers are often used to make jasmine tea and are popular both at home and abroad, the presence of pesticide residues in these products also need to be seriously considered. Food safety in tea has always been a hot topic of concern in various countries. For instance, Japan, the European Union (EU) and the Codex Alimentarius Commission (CAC) have all introduced their own laws and policies to strictly control the pesticide residue limits in imported tea (Chen et al. 2016). As far as China is concerned, by the end of February 2022, there were 864 registered pesticide products for tea, of which insecticides were the main ones. China currently allows the use of some highly effective pesticides of low toxicity on tea. For example, in the GB 2763 − 2021 "National food safety standard-Maximum residue limits for pesticides in food", maximum residue limits (MRLs) were set for 106 pesticides used on tea and these consisted mainly of organophosphates, pyrethroids and carbamates. These pesticides tend to be heavily used in tea gardens and hence, they are frequently detected in tea (Sun et al. 2021).
Pesticide residues are one of the main problems that affect food safety and environmental pollution while negatively impacting food export . As a result, rapid and convenient methods for detecting pesticide residues are indispensable. In this context, the QuEChERS method (quick, easy, cheap, effective, rugged, and safe) provides a simple, rapid and low-cost approach for sample preparation (Anastassiades et al. 2003a, b). The purification step in the method is often referred to as dispersive solid-phase extraction (d-SPE) (Ajibola et al. 2021). However, there are limitations in the cleanup step to analyze the residues from complex matrices (Zhang et al. 2019). As a result, the method has been modified by researchers around the world (Zapata and Peñuela 2021;Zhuang et al. 2022). Previous studies have demonstrated the successful application of the QuEChERS method for determining multi-pesticide residues for floral tea like honeysuckle, calendula and chrysanthemum (Besil et al. 2017;Fang et al. 2021;Gong et al. 2020;Jiang et al. 2020). At the same time, there is only few research about the determination of pesticide residues in both flower and its scented tea. Therefore, it would be worthwhile to develop reliable and quick analytical methods for analyzing multi-pesticide residues in these commodities.
The multi-plug filtration cleanup (m-PFC) method is a new approach based on d-SPE which was developed by the current research group (Zhao et al. 2013(Zhao et al. , 2014. In the m-PFC column, the sorbents are packed tightly between two sieve plates in a short column like an SPE cartridge (Qin et al. 2016b(Qin et al. , 2017a. While pushing and pulling the piston, the extract interacts with the sorbents so that the interfering substances are adsorbed. In previous studies (Qin et al. 2016a(Qin et al. , 2017bSong et al. 2020), this method was successfully applied to determine pesticide residues from complex matrices. Compared with d-SPE procedure, the m-PFC method is shown to be more rapid without solvent evaporation, vortex and centrifugation steps (Qin et al. 2015;Zou et al. 2016;Pszczolinska and Kociolek 2022). So far, different studies have proven that the m-PFC method could be a fast, convenient and efficient method to determine pesticide residues.
In this study, with reference to the pesticides registered for tea and the global MRLs, twenty-five pesticides, commonly used in the field during the cultivation of jasmine and tea, were selected before comparing the purification efficiency of the m-PFC and d-SPE methods. This was followed by the development of an approach to determine twenty-five pesticide residues in jasmine flower and its scented tea based on m-PFC and ultra-high-performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS). This method was validated and successfully applied to determine the level of pesticide residues in marketed jasmine flower and scented tea samples.

Materials and Methods
Twenty-five pesticide standards (as in Table 1) were obtained from the Institute of the Control of Agrochemicals, Ministry of Agriculture People's Republic of China. The purity of the standard compounds was from 95% to 99%. Stock solutions (1000 mg L − 1 ) of the analyte mixtures were prepared in chromatographic grade acetonitrile and stored at − 20°C. The working solutions were then prepared from the stock solutions daily. Chromatographic grade acetonitrile, formic acid and methanol were purchased from Fisher Chemicals (Fair Lawn, New Jersey, USA) while analytical grade sodium chloride (NaCl), anhydrous magnesium sulfate (MgSO 4 ) and ammonium acetate were obtained from Sinopharm Chemical Reagent (Beijing, China). Ultra-pure water was obtained from a Milli-Q system (Bedford, MA, USA), with Multiwalled Carbon Nanotubes (MWCNTs) having a particle range of diameters 15-25 nm were provided by Tianjin Bonna-Agela (Tianjin, China). Finally, primary secondary amine (PSA), Octadecylsilane (C 18 ) and the m-PFC column were obtained by HAMAG instrument technology Co., Ltd. (Zhejiang, China).
The pesticide residues were analyzed by an Agilent Series 1290 ultra-high-performance liquid chromatography combined with an Agilent G6465B triple quadrupole mass spectrometer. The components were separated using a 2.1 × 50 mm × 3 μm Athena C18-WP analytical column from ANPEL Laboratory Technologies Inc. (Shanghai, China), with the column temperature, flow rate and injection volume set at 40°C, 0.25 mL min − 1 and 5 µL respectively. The mobile phase A consisted of an aqueous solution of ammonium acetate at a concentration of 5 mmol/L and containing 0.1% (V/V) of formic acid, while for the mobile phase B, methanol was used. The gradient elution was then performed as follows: 0-1 min at 2% B; 1-1.1 min at a linear gradient of 2%-50% B; 1.1-2 min at 50% B; 2-15 min, a linear gradient of 50%-98% B, held for 1 min; finally, returning to the initial ratio of composition and equilibrated for 2.5 min prior to the next injection. The total time taken to separate and stabilize the twenty-five analytes was 18.5 min. These compounds were subsequently analyzed using the multiple reaction monitoring (MRM) mode and the positive electrospray ionization (ESI) mode, with an Agilent Mass Hunter Workstation software subsequently used for data analysis. Table 1 shows the optimized MRM data acquisitions of UHPLC-MS/MS for the 25 pesticides, with a typical chromatogram provided in Fig.S1.
The jasmine flower and scented tea samples were obtained from a local supermarket and homogenized with a pulverizer at room temperature. The samples were stored at − 20°C until analysis.  Homogenized samples (2 ± 0.02 g) were weighed into a 50-mL centrifuge tube and after adding 4 mL of water, the tube was vortexed for 2 min. This was followed by the addition of 10 mL of acetonitrile and after shaking the tube vigorously for 5 min, 3 g of sodium chloride was added. The tube was again shaken for 3 min before centrifugation for 5 min at 3800 rpm. The supernatant was stand-by for further cleanup procedures.
To 2-mL centrifuge tubes containing combinations of different sorbents, 1 mL of supernatant was added. The tubes were then shaken for 3 min before being centrifuged for 1 min at 10,000 rpm. Finally, the supernatants were filtered through 0.22-µm filter membranes into autosampler vials for UHPLC-MS/MS analysis.
In order to evaluate the best combination and proportion of sorbents, the following combinations were chosen for the experiment: The m-PFC columns need to be packed before being used (Fig. 1). A polyethylene (PE) sieve plate, sorbent materials(combination F mentioned before) and another PE sieve plate were put into the column successively (Zhao et al. 2013(Zhao et al. , 2014. The selected sorbents were mixed homogeneously before being packed into the columns. One milliliter of supernatant was then introduced into the column. By pushing the piston, the liquid was passed through the sorbents into a 2-mL centrifuge tube. Using a pipette, the liquid in the tube was transferred to the m-PFC column before pushing the piston again. The supernatant was eventually filtered through a 0.22-µm filter membrane into an autosampler vial for UHPLC-MS/MS analysis. The optimized m-PFC method was validated to confirm the feasibility of the approach when using jasmine flower and scented tea for the pretreatment step. The analytical method for pesticide residues was validated referred to the European Union SANTE/11,312/2021 regulatory guidelines. The validated parameters were linear range, fortified recovery, precision, and limit of quantification (LOQ).
The matrix effect (ME) refers to the influence and interference of the impurities in the sample during the detection process. It was calculated as shown in Eq. (1): (1) ME = Slope ratio of matrix -matched standards Slope ratio of solvent standards

Results and Discussion
In a similar way to plants such as mulberry leaves and chrysanthemums (Wu et al. 2020;Wang et al. 2019), jasmine contains many active chemical components, including polysaccharides, flavonoids, pigments and volatile oil amongst others. Common sorbents such as PSA can remove some of the carbohydrates, fatty acids, polar pigments and polar organic acids . Similarly, C 18 has been used to remove sterols and non-polar interfering substance (Liu et al. 2017), while MWCNTs was shown to be effective in removing pigments, fatty acids and other matrix compounds (Zhao et al. 2014). In fact, a mixture of MWCNTs, PSA and C 18 could effectively remove the impurities in complex matrix . In order to identify the optimal sorbent combinations, the degree of purification of different combinations and proportions of PSA, C 18 and MWCNTs were compared at a level of 0.5 mg kg − 1 . The experiments involved transferring 1 mL of an acetonitrile supernatant of jasmine to 2-mL centrifugal tubes containing different sorbent mixtures (combinations A to F mentioned before), with the results regarding recoveries and relative standard deviations (RSDs) shown in Fig. 2. The results showed that the RSDs of combination F were lower than 10%, which were also lower than other combinations, and the recoveries of most pesticides were between 70% and 120%. Overall, there were more pesticides which met the requirements of pesticide residue analysis after purification with combination F. In addition, as shown in Fig.S2, jasmine samples purified by combination F appeared more colorless and hence, this combination was selected for jasmine. The addition of water to the dry matrix moistens the sample and can improve the extraction efficiency of pesticides. During the pretreatment of the tea matrix, which is similar to the jasmine matrix, a certain amount of water was added in order to obtain better recovery results (Hou et al. 2014;Cajka et al. 2012). In this study, the traditional QuEChERS method was used and the volume of added water was optimized using three levels (0, 4 and 8 mL). At the spiked level of 0.5 mg kg − 1 , the recoveries and the RSDs of the 25 pesticides at the three levels were compared. The results (Fig. 3a, b) indicated that for most pesticides, the recovery levels were between 70% and 120%. The RSDs ranged from 0.89% to 18% when no water was added while the addition of 4 mL and 8 mL of water changed the RSDs to 0.18%-13% and 0.56%-14% respectively. Based on the RSD results, when the volume of water was 4 mL, the recoveries of the 25 pesticides was more stable, thus allowing the water content to be optimized at 4 mL.
The m-PFC process is a purification step prior to chromatographic analysis, in which the times of cycle for the m-PFC column-based purification was optimized. As shown in Fig. 1, the first value was set when the piston was pushed to allow 1 mL of acetonitrile supernatant of jasmine to pass through the column. Then, the second value was set when the one-time purified solution was transferred into the column, and the piston was pushed again to make it pass through the column. Repeatedly as above, the two-time purified solution was transferred into and passed through the column, and the third value was set. Then the purifying effects of these three values were compared at the spiked level of 0.5 mg kg − 1 , which showed that the recoveries and RSDs seemed to be similar (Fig. 3c, d). For a two-time purification, recoveries of the pesticides showed a better uniformity because the RSDs were smaller. And based on previous research (Zou et al. 2016;Wu et al. 2020), interfering compounds would also more likely be desorbed from the sorbents as the times of purification increased. Subsequently, considering the laborious and time-consuming nature of the experiment, the second value was chosen in this research.
The d-SPE and the m-PFC procedures were compared at the spiked level of 0.5 mg kg − 1 by using jasmine flower. Though the d-SPE involved vortex and centrifugation which the m-PFC did not, the m-PFC would result in the adsorption and desorption of pesticides. Therefore, for the evaluation of these two procedures, the purification efficiency was compared as the basis in terms of the results of recovery, RSD and the matrix effect, which are listed in Table 2. The matrix effects showing little differences. However, the recoveries of the d-SPE and the m-PFC methods were 67%-133% and 65%-103% respectively. In fact, after the d-SPE procedure, the recoveries of three pesticides were above 120%, indicating that the pesticide residue requirements were not met. Furthermore, even though the RSDs were less than 15% for both methods, the m-PFC method still had significantly smaller RSD values, thus indicating more stable recoveries. The recoveries and RSDs of the two methods were further compared when applied to scented tea, with the results shown in Table S1. In this case, the recoveries of the d-SPE and m-PFC methods were 36%-119% and 70%-102% respectively, with their corresponding RSDs being 1.2%-12% and 0.33%-5.0%. The only exception was methamidophos for which the recovery was far less than 70% after d-SPE. Overall, the results of the m-PFC method met the requirements of pesticide residue analysis as the recoveries were much more consistent. Considering the simplicity and convenience of the m-PFC method, the results therefore support its application as the better approach.
Based on the above experimental results, the proposed method was validated, with the results summarized in Tables S2 and S3. Linearity was evaluated using a calibration curve with six concentrations (0.002, 0.01, 0.05, 0.1, 0.5 and 1 mg L − 1 ), while the correlation coefficients (R 2 ) of the 25 pesticides were all greater than 0.9962. Accuracy and precision were evaluated by recoveries and RSDs through recovery experiments which were performed at three spiked levels (0.01, 0.1 and 5 mg kg − 1 ). The recoveries of the 25 pesticides were in the range of 64%-106%, with RSDs between 0.58 and 10%. However, recoveries of methamidophos and carbaryl were slightly low at the spiked level of 0.1 mg kg − 1 but given that the data was stable, it was accepted that the method was proper for analyzation. The LOQ refers the lowest concentration or mass of the analyte that can be validated with acceptable accuracy by applying the complete analytical method and identification criteria. In this study, the LOQs were 0.01 mg kg − 1 . The method was then applied to scented tea and the feasibility of the method was validated. As Tables S4 and S5 show, the recoveries were in the range of 65%-108%, with RSDs between 0.33% and 8.8%, except for methomyl. The linearity was evaluated with a calibration curve using five concentrations (0.01, 0.05, 0.1, 0.5 and 1 mg L − 1 ), and the R 2 values of the 24 pesticides were greater than 0.9936. The LOQs were 0.05 mg kg − 1 .
The existence of the matrix may result in enhanced or weakened signal responses for the analytes (Zou et al. 2016;Qin et al. 2015). In this study, the ME of m-PFC and d-SPE method were compared by the slope ratio of matrix-matched and solvent standards. Generally, the ME can be ignored for values in the range of 0.9-1.1 (Wu et al. 2020). The ME is considered to be weakened when it is less than 0.9, and enhanced when it is greater than 1.1. As Tables S3 and S5 show, the matrix weakening effect (ME < 0.9) was present in the analytes as most of the ME values of target compounds in jasmine flowers and scented tea were far less than 0.9. It was therefore speculated that jasmine and scented tea matrices were complex mixtures, which strongly interfered with the analysis and detection of the target analytes. To address this issue, the matrix-matched standard solution was used for subsequent quantification.
The proposed method was applied to the analysis of pesticide residues in 22 real jasmine samples and 16 scented tea samples which were collected from the following 8 provinces: Fujian, Sichuan, Yunnan, Guangxi, Zhejiang, Guangdong, Anhui and Beijing. All the samples were found to contain studied pesticides, and a total of 20 pesticides were detected. These compounds had a detection range from 0.013 to 1.9 mg kg − 1 in the jasmine samples and 0.014 to 4.7 mg kg − 1 in the tea samples. Over 15 pesticides were detected in 5 jasmine samples and over 14 were detected in 4 tea samples, with the three main types detected being neonicotinoids, organophosphates and pyrethroids.
The results were compared with the MRLs of each pesticide on jasmine established by the EU, with the data shown in Table 3. In particular, the detection rate of fenpropathrin was up to 100%, with a range of 0.019-0.088 mg kg − 1 . Acetamiprid and chlorpyrifos also had high detection rate. Furthermore, the residues of 15 pesticides, such as thiamethoxam, acetamiprid and fenpropathrin amongst others were above MRLs. That probably because most of the EU values are at the lower limit of analytical determination. Countries such as China, the United States and Japan are yet to establish MRLs of pesticides for jasmine. Considering that jasmine is often used to make jasmine tea, the analytical results of the 19 detected pesticides were compared with the MRLs established by China and the CAC for each pesticide used on tea. Pesticide residues of carbofuran, fenvalerate and carbosulfan exceeded the MRLs developed by China, with all three pesticides also banned from Chinese tea gardens. Therefore, it is necessary to comprehensively assess the residue characteristics and risks of pesticides on jasmine flowers. For scented tea samples, the data are shown in Table S6. Methamidophos, imidacloprid, carbofuran, isocarbophos and chlorpyrifos had excessive pesticide residues compared with the MRLs established for tea in China while methamidophos, carbofuran and isocarbophos were detected despite being prohibited in tea plantations in China. The MRLs for pesticides used on tea, as established by the EU, are more stringent and the values are mostly at the lower limit of analytical determination. Therefore, the current situation regarding the level of pesticide residues in domestic scented tea may have an impact on the import and export trade.
As shown by the results, some pesticide residues were found in excess in both jasmine flower and its scented tea. As a unique tea consumed in China, jasmine and its scented tea are not only popular among consumers but are even sold abroad. Such situations may be harmful to the health of consumers and negatively impact the export of related products. Jasmine and scented tea are consumed in a similar way by brewing in water. For both jasmine flower and tea, the pesticides which were found to be in excess included those with low n-octanol-water partition coefficients (log K ow ). The log K ow values of the 25 pesticides studied are shown in Table 1. Transfer of pesticides to tea is negatively correlated with log K ow . Given that the values are relatively small, pesticides are therefore more likely to be transferred to the tea infusion, which may pose potential risks to the health of consumers. Pyrethroids have a relatively large log K ow of around 5-7 and although their detection rate was high, they may not cause too much harm. There is a lack of information on the registration of pesticides which are currently applied on jasmine and as such, the authorities would need to consider registering all types of pesticides applied on jasmine flowers. In addition, due to risks of ingestion, the use of pesticides which have small log K ow , are banned in tea gardens or have excessive residues such as methamidophos, imidacloprid, carbofuran, chlorpyrifos and fenvalerate, would need to be strictly controlled during cultivation. Overall, the use of pesticides during the cultivation of both jasmine and tea still needs to be improved. This study sought to develop a rapid and efficient method that could detect twenty-five pesticide residues in jasmine based on m-PFC and UHPLC-MS/MS. The method could also be applied to its scented tea. The method was validated in terms of its analytical range, precision, recovery and reproducibility. The results showed that the method met the requirements of pesticide analysis. Determining the pesticide residues in marketed jasmine and scented tea samples provided an overview of pesticide use, and it might be necessary to increase the focus on the use of water-soluble pesticides.