Method for simultaneously determining content of menthol, WS3 and WS23 in electronic cigarette liquid based on Raman spectrum

Abstract

The invention discloses a method for simultaneously determining the content of menthol, WS3 and WS23 in electronic cigarette liquid based on Raman spectrum, and belongs to the technical field of analysis and detection. The method comprises the following steps: preparing a standard solution and a sample solution by using a mixed solvent, synchronously introducing a nano reinforcing material as a reinforcing medium into the standard solution and the sample solution, collecting a Raman spectrum, extracting a characteristic peak signal of a target component, and establishing a standard working curve to realize synchronous quantitative analysis of menthol, WS3 and WS23. The method has the advantages of simple sample pretreatment, fast detection speed, high sensitivity and accurate and stable determination result, and is suitable for rapid detection and quality control of a plurality of cooling agents and flavor components in the electronic cigarette liquid.

Description

Technical Field

[0001] This invention relates to the field of analytical detection technology, and in particular to a method for simultaneously determining the contents of menthol, WS3 and WS23 in e-cigarette liquid based on Raman spectroscopy. Background Technology

[0002] In recent years, e-cigarette products have been widely used due to their atomization inhalation method and diverse flavor formulations. E-cigarette liquids typically use propylene glycol and glycerin as the main solvent system, with added menthol and synthetic cooling agents to improve the cooling sensation and taste experience. Menthol, as a natural monoterpene alcohol compound, has a distinct cooling sensation and characteristic odor; synthetic cooling agents such as WS3 and WS23 are odorless or weakly odorous cooling additives, widely used in e-cigarette liquid formulations due to their high cooling intensity and good stability. The amount of these substances added directly affects the sensory quality and safety of the product; therefore, accurate quantitative detection of menthol, WS3, and WS23 in e-cigarette liquids is a crucial aspect of product quality control and regulatory testing.

[0003]

[0004] The structural formulas of menthol, WS3, and WS23

[0005] In existing technologies, the detection methods for menthol and cooling agents in e-cigarette liquids mainly employ gas chromatography (GC) or gas chromatography-mass spectrometry (GC-MS). While these methods offer high separation capabilities and qualitative accuracy, they typically require complex sample pretreatment steps, such as dilution, extraction, derivatization, or prolonged chromatographic separation. This results in long analysis cycles, high detection costs, and is not conducive to rapid, simultaneous detection of multiple components. In applications involving large-scale sample testing or rapid on-site screening, the efficiency and flexibility of these methods are somewhat limited.

[0006] Raman spectroscopy, an analytical technique based on molecular vibrational information, offers advantages such as simple sample pretreatment, fast detection speed, and the ability to simultaneously analyze multiple components. In recent years, it has been increasingly adopted for the qualitative and quantitative analysis of organic components in food, pharmaceuticals, and chemical products. However, e-cigarette liquids are typical multi-component organic systems with complex solvent backgrounds and significant fluorescence interference. Compounds such as menthol, WS3, and WS23 exhibit relatively weak scattering signals under conventional Raman conditions, making them susceptible to solvent Raman background and baseline drift. This results in insufficient signal-to-noise ratio and limited quantitative sensitivity, making it difficult to meet the requirements for accurate determination of low concentrations.

[0007] Furthermore, different solvent systems have a significant impact on the Raman spectral background and the signal intensity of the target component. Existing reports often use methanol, propylene glycol, or other organic solvents as detection media. However, under the complex e-cigarette liquid matrix conditions, a single solvent system often fails to simultaneously ensure the uniformity of the target component's dissolution and the Raman background suppression effect, easily leading to quantitative deviations and reduced repeatability.

[0008] Surface-enhanced Raman scattering (SERS) technology, by introducing noble metal nanomaterials to enhance the local electromagnetic field, can significantly improve the Raman signal intensity and has potential application value in the field of trace detection. However, existing SERS methods still have shortcomings in the stability, dispersibility, and compatibility with complex organic systems of the enhancing substrate, especially in high organic solvent systems such as e-cigarette liquids, where conventional metal nanoparticles are prone to aggregation or background interference, affecting the reliability and consistency of quantitative results.

[0009] Therefore, how to construct a Raman spectroscopy method suitable for the complex matrix of e-cigarette liquids and capable of rapid, sensitive and accurate determination of menthol, WS3 and WS23 simultaneously, while ensuring simple sample pretreatment, remains a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0010] Given that existing methods for detecting menthol and synthetic cooling agents in e-cigarette liquids largely rely on gas chromatography-mass spectrometry (GC-MS), which suffers from complex sample pretreatment, long analysis cycles, high equipment costs, and difficulty in achieving simultaneous and rapid quantification of multiple components; furthermore, existing Raman detection methods are susceptible to solvent background interference in complex organic matrices and suffer from insufficient sensitivity, resulting in high limits of quantification and limited precision and accuracy. The purpose of this invention is to provide a Raman spectroscopy-based analytical method that enables simultaneous, rapid, and accurate quantification of menthol, WS3, and WS23 in e-cigarette liquids without the need for complex separation steps.

[0011] To achieve the above objectives, this invention provides a method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy, comprising the following steps:

[0012] (1) Using the mixed solution as a solvent, a series of standard solutions of menthol, WS3 and WS23 were prepared to cover the actual content range of the components to be tested in the e-cigarette liquid;

[0013] (2) Take an e-cigarette liquid sample, quantitatively dilute it using the mixed solvent, and obtain a homogeneous and stable test solution by shaking and standing.

[0014] (3) Under consistent Raman testing conditions, Raman spectra were acquired for the standard solution and the sample solution. The obtained spectra were baseline corrected and smoothed. The characteristic Raman peak signals corresponding to menthol, WS3 and WS23 were extracted. The quantitative relationship between the response value and the mass fraction was established, and the content of each target component in the sample was calculated accordingly.

[0015] The mixed solution is a mixture of methanol and propylene glycol; preferably, the mixed solvent is prepared by mixing methanol and propylene glycol in a volume ratio of (1-3):(1-3);

[0016] Small methanol molecules enter the hydrogen-bonded network of propylene glycol, weakening multi-point association and reducing viscosity, making the sample easier to mix and defoam. Propylene glycol is compatible with the flue gas matrix, stabilizing the solvation layer and the dispersion of the nano-reinforced medium, and suppressing baseline drift. The synergistic effect of both reduces the noise term.

[0017] In a preferred embodiment, surface-enhanced nanomaterials are introduced into the sample and standard solutions as Raman signal enhancement media. These nanomaterials are silica-coated gold nanoparticles, or modified silica-coated gold nanoparticles further modified with organosilanes and cyclodextrins. By controlling the final concentration of the nanoparticles in the test solution, the Raman scattering signals of menthol, WS3, and WS23 are significantly enhanced without introducing additional background interference, thereby effectively reducing the detection and quantitation limits of the method and improving the sensitivity and repeatability of the assay.

[0018] A further preferred method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy includes the following steps:

[0019] Step 1: Preparation of standard solutions: Using the mixed solution as a solvent, prepare a series of standard solutions of menthol, WS3 and WS23 respectively.

[0020] Step 2: Sample preparation: Take 0.2-1g of e-cigarette liquid, add it to the mixed solvent described in Step 1 for dilution, so that the expected content of the analyte falls within the standard curve range; shake and mix at 100-300 r / min for 5-20 min, let stand for 1-10 min to defoam, take the supernatant, add the enhanced nanomaterial dispersion to the supernatant, so that the final concentration of the enhanced nanomaterial in the test solution is 0.01-1 mg / mL, vortex mix for 10-120s, let stand for 1-15 min, and then transfer to the Raman sample cell for testing;

[0021] Step 3: Raman spectroscopy acquisition: Raman spectra of standard solutions of various concentrations and sample solutions to be tested are acquired under the same conditions;

[0022] Step 4: Data Processing and Quantification: Perform consistent baseline correction and smoothing on the spectrum, extract the peak height or integral area of ​​the characteristic peaks of menthol, WS3 and WS23 as the response value Y, and establish a standard working curve with the corresponding mass fraction X; obtain the response value Y of the sample solution to be tested and substitute it into the regression equation to calculate the mass fraction of each component, and combine the sample weight and dilution factor to calculate the content of menthol, WS3 and WS23 in the original e-cigarette liquid.

[0023] Raman spectroscopy conditions:

[0024] The sample spectra were acquired using a 785nm laser Raman spectrometer with a laser power of 10-200mW and a spectral acquisition range of 200-2000 cm⁻¹. -1 Spectral resolution no higher than 4 cm -1 The integration time for each measurement was 0.5-60 s, and the average of 1-50 measurements was taken to obtain the single measurement spectrum. Each sample was measured 2-6 times at different locations in the sample cell, and the average spectrum was used as the final spectrum. A 520.7 cm⁻¹ silicon wafer was used before each batch of measurements. -1 Wavenumber calibration was performed, and blank background spectra corresponding to the composition of the test system were collected for background comparison and baseline correction. When no nano-reinforcing material was added to the test system, the blank background spectrum was the blank spectrum of methanol-propylene glycol mixed solvent. When nano-reinforcing material was added to the test system, the blank background spectrum was the blank spectrum of methanol-propylene glycol mixed solvent with the same type and final concentration of nano-reinforcing material as the standard solution and sample solution.

[0025] Preferably, menthol, WS3, and WS23 each have their own single-component Raman characteristic peaks; the positions of the characteristic peaks of each component are determined based on the Raman spectra obtained from the corresponding standards under the same test conditions. During simultaneous determination of the three components, peaks with minimal overlap with the solvent background peak, the nano-reinforcement material background peak, and other analyte peaks, exhibiting complete peak shapes, high signal-to-noise ratios, and good repeatability are selected as quantitative peaks, and other reproducible characteristic peaks are selected as auxiliary qualitative peaks. In a preferred embodiment, the peak height of the selected quantitative peak is used as the response value Y for data processing; in other embodiments, the peak area can also be used as the response value for quantitative analysis. The original Raman spectrum is first subjected to blank background subtraction, followed by baseline correction and smoothing; the baseline correction uses a fifth-order polynomial fitting baseline correction method, and the smoothing uses the Savitzky-Golay smoothing algorithm with smoothing parameters of a second-order polynomial and an 11-point window.

[0026] A standard working curve was established with the mass fraction X of each target component as the abscissa and the corresponding quantitative peak height Y as the ordinate, and linear regression was performed. The response value Y of the unknown sample was measured under the same conditions and then entered into the corresponding regression equation to calculate the mass fraction of each component in the sample solution. The contents of menthol, WS3 and WS23 in the original e-cigarette liquid were calculated by combining the sample weight and dilution factor.

[0027] The detection limit (LOD) is calculated as LOD = 3σ / k;

[0028] The limit of quantitation (LOQ) is calculated as LOQ = 10σ / k;

[0029] In the formula, k is the slope of the standard working curve of the corresponding component; σ is the standard deviation of the response value obtained at the corresponding quantitative peak position after repeated determination of the blank background solution under the same test conditions and after the same data processing.

[0030] Furthermore, by simultaneously introducing the same type and concentration of nano-reinforcing materials into the standard solution and the sample solution, and using the blank background spectrum of the corresponding solvent system for correction, the present invention effectively eliminates the influence of the nanomaterials themselves and the solvent system on the Raman background, ensuring the linear reliability and comparability of the quantitative model.

[0031] The preparation method of the enhanced nanomaterial dispersion is as follows:

[0032] S1. Add chloroauric acid to water and mix thoroughly to prepare a chloroauric acid aqueous solution with a concentration of 0.05-1 mmol / L. Take 50-300 mL of the 0.05-1 mmol / L chloroauric acid aqueous solution and place it in a round-bottom flask. Heat to boiling and keep stirring under reflux. Add 2-30 mL of 10-100 mmol / L sodium citrate aqueous solution and continue reflux for 5-60 min. Cool to room temperature and let stand for 15-60 min before use. Centrifuge at 5000-15000 rpm for 5-30 min, discard the supernatant, and redisperse the precipitate with anhydrous ethanol. Repeat the washing 1-3 times to change the system from the aqueous phase to the ethanol phase dispersion to obtain ethanol-dispersed AuNP.

[0033] S2. Take the above-mentioned ethanol-dispersed AuNP and add anhydrous ethanol to a total volume of 50-200 mL.

[0034] Add 0.2-2 mL of ammonia and 0.2-5 mL of water, stir at room temperature for 3-30 min, add 20-300 μL of tetraethyl orthosilicate dropwise while stirring, continue the reaction for 1-12 h, centrifuge at 6000-15000 rpm for 5-30 min to collect the precipitate, wash with ethanol 1-5 times, redisperse in ethanol, adjust the Au@SiO2 ethanol dispersion to a solid content of 1-5 mg / mL, add propyltriethoxysilane isocyanate to the dispersion to make the volume fraction of propyltriethoxysilane 0.2-3.3%, stir the reaction at 40-80℃ for 2-12 h, centrifuge and wash 1-5 times after the reaction to obtain modified nanoparticles;

[0035] S3. Add hydroxypropyl-β-cyclodextrin to anhydrous DMF to obtain a hydroxypropyl-β-cyclodextrin dispersion with a concentration of 15-50 mg / mL; add triethylamine to make its volume fraction 0.8-1.8%; add the modified nanoparticles to the above cyclodextrin solution to make the final concentration of nanoparticles 1-5 mg / mL, stir and react at 40-60℃ for 10-20 h. After the reaction is completed, centrifuge at 6000-15000 rpm for 10-30 min to collect the precipitate, wash it with DMF 1-3 times, ethanol 1-3 times, and water 1-3 times in sequence, and redisperse the obtained Au@SiO2-CD in ethanol to prepare a 5-15 mg / mL enhanced nanomaterial dispersion, and store it at 2-8℃ in the dark for later use.

[0036] The SiO2 shell isolates the gold nuclei from agglomeration, while the isocyanate silane introduces active sites and enhances organophilicity. Cyclodextrin anchors menthol and WS-type compounds through inclusion and hydrogen bonding, bringing them closer to the hotspot and ensuring consistent orientation. This enhances field stability, reduces baseline fluctuations, and thus improves signal intensity and repeatability, while lowering the detection or quantitation limits.

[0037] The beneficial effects of this invention are:

[0038] (1) This invention achieves simultaneous quantitative analysis of menthol, WS3 and WS23 based on Raman spectroscopy, without the need for chromatographic separation, simplifying sample pretreatment and significantly improving detection efficiency;

[0039] (2) By using a methanol-propylene glycol composite solvent system, the solubility uniformity of multiple components in e-cigarette liquid was effectively improved, and the influence of matrix effect on Raman quantification results was reduced.

[0040] (3) After introducing silica-coated gold nanoparticles, especially modified nanomaterials modified with cyclodextrin, the Raman signal intensity of the target component was significantly enhanced, the method detection limit and quantitation limit were significantly reduced, and the precision and spiked recovery rate were better than those of the unenhanced system.

[0041] (4) The present invention has the advantages of fast detection speed, simple operation, high sensitivity and good repeatability, which can meet the needs of rapid quality control and supervision of cooling agents and flavor substances in e-cigarette liquid. Attached Figure Description

[0042] Figure 1 Raman spectrum of WS23;

[0043] Figure 2 Raman spectrum of WS3;

[0044] Figure 3 This is the Raman spectrum of menthol. Detailed Implementation

[0045] The raw materials used in this embodiment and its comparative examples are all commercially available.

[0046] Example 1

[0047] A method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy includes the following steps: Step 1: Preparation of standard solutions: Using a mixed solution of methanol and propylene glycol at a volume ratio of 1:1 as the solvent, a series of standard solutions of menthol, WS3, and WS23 are prepared with mass fraction gradients of 0.7%, 1.3%, 3%, 6%, 11%, 24%, and 39%, respectively. The standard solutions are prepared according to a total mass M: Weigh the target substance mass m, make w = m / M × 100%, and then add the mixed solvent to the total mass M, mixing thoroughly; Step 2: Sample preparation: Take 0.5g of e-cigarette liquid, add the mixed solvent described in Step 1 to a total mass of 5g, ensuring the expected content of the analyte falls within the standard curve range; shake and mix at 180 r / min for 10 min, let stand for 2 min to defoam, collect the supernatant, and transfer it to the Raman sample cell for analysis;

[0048] Step 3: Raman spectroscopy acquisition: Raman spectra of standard solutions of various concentrations and sample solutions to be tested are acquired under the same conditions.

[0049] Raman spectroscopy conditions:

[0050] The sample spectra were acquired using a 785 nm laser Raman spectrometer with a laser power of 50 mW and a spectral acquisition range of 200–2000 cm⁻¹. -1 Spectral resolution no higher than 4 cm -1 The integration time was 0.5-60 s each time, and the average of 1-50 measurements was taken to obtain the single measurement spectrum. Each sample was measured three times at different positions in the sample cell, and the average spectrum was taken as the final spectrum. A 520.7 cm⁻¹ silicon wafer was used before each batch of measurements. -1Wavenumber calibration was performed, and a blank background spectrum of methanol-propylene glycol (1:1, v / v) was acquired for background control and baseline correction.

[0051] Raman spectra of menthol, WS23, and WS3 single-component standard solutions were acquired, and baseline correction and smoothing were performed to determine the characteristic peaks of each target component. The characteristic peak of menthol included a peak at approximately 547 cm⁻¹. -1 766cm -1 1036cm -1 and 1451 cm -1 The single-component characteristic peak of WS23 includes approximately 609 cm⁻¹. -1 769 cm -1 903 cm -1 926cm -1 1034 cm -1 1153 cm -1 and 1460 cm -1 The single-component characteristic peak of WS3 includes approximately 303 cm⁻¹. -1 478 cm -1 771 cm -1 916 cm -1 1036 cm -1 1140 cm -1 1174 cm -1 1452 cm -1 In this embodiment, menthol was selected at approximately 547 cm⁻¹ for simultaneous determination of the three components. -1 WS23 is approximately 1153 cm tall. -1 WS3 is approximately 1174 cm tall. -1 The characteristic peak at the specified location is used as the quantitative peak, while the remaining single-component characteristic peaks are used as auxiliary qualitative peaks. The above quantitative peaks have little overlap with the background peaks of the mixed solvent and other target component characteristic peaks, have complete peak shapes, and a high signal-to-noise ratio, making them suitable for establishing stable quantitative models.

[0052] The original Raman spectra obtained from the standard solution and the sample solution were first subjected to blank background subtraction, then baseline correction was performed using the fifth-order polynomial fitting baseline correction method, and finally smoothed using the Savitzky-Golay smoothing algorithm with smoothing parameters of second-order polynomial and 11-point window.

[0053] In this embodiment, the peak height of the quantitative peak is uniformly extracted as the response value Y. A standard working curve is established with the mass fraction X of each component as the abscissa and the corresponding peak height Y as the ordinate, and linear regression is performed. The response value Y of the unknown sample is measured in the same way and substituted into the corresponding regression equation to calculate the mass fraction of menthol, WS3 and WS23 in the sample solution. The corresponding content in the original e-cigarette liquid is calculated by combining the sample weight and dilution factor.

[0054] In this embodiment, the limit of detection (LOD) is calculated as LOD=3σ / k, and the limit of quantitation (LOQ) is calculated as LOQ=10σ / k; where k is the slope of the standard working curve of each component, and σ is the standard deviation of the response value obtained at the corresponding quantitative peak position after the blank background solution is repeatedly measured 10 times under the same test conditions and processed in the same way.

[0055] Table 1: Standard operating curves of the compounds in Example 1

[0056] compound Linear range (quality fraction, %) Regression equation (Y=aX+b) Correlation coefficient (R) Limit of detection (%) Limit of Quantification (%) Menthol 0.7-39 Y = 0.0258X + 0.0029 0.9997 0.08 0.26 WS3 0.7-39 Y = 0.0312X + 0.0024 0.9996 0.06 0.20 WS23 0.7-39 Y = 0.0289X + 0.0027 0.9996 0.07 0.23

[0057] The precision of the method was investigated by performing six parallel determinations on the same e-cigarette liquid sample (Table 2).

[0058] Table 2: Precision of Sample Measurement in Example 1 (n=6) (Unit: mass fraction, %)

[0059] compound average value RSD (%) Menthol 12.2 1.16 WS3 1.25 1.13 WS23 0.99 1.07

[0060] Using the same e-cigarette liquid sample as the matrix, menthol, WS3, and WS23 were spiked at three levels: low, medium, and high, at 1%, 6%, and 24% (mass fraction increment based on the original sample), respectively. The original sample content was determined and back-calculated according to the procedure in Example 1, and the spiked recovery rate was calculated. The results are shown in Table 3.

[0061] Table 3 Spike recoveries of Example 1

[0062] compound Low concentration (%) Medium concentration (%) High concentration (%) Average recovery rate (%) Menthol 99 101 100 100 WS3 98 99 101 99 WS23 97 100 102 100

[0063] Comparative Example 1

[0064] The process is basically the same as in Example 1, except that methanol is used as the solvent and standard solutions of menthol, WS3 and WS23 are prepared with the same concentrations as in Example 1.

[0065] Table 4: Standard operating curves of compounds in Comparative Example 1

[0066] compound Linear range (quality fraction, %) Regression equation (Y=aX+b) Correlation coefficient (R) Limit of detection (%) Limit of Quantification (%) Menthol 0.7-39 Y = 0.0270X + 0.0031 0.9995 0.1 0.30 WS3 0.7-39 Y = 0.0325X + 0.0027 0.9994 0.08 0.22 WS23 0.7-39 Y = 0.0300X + 0.0025 0.9993 0.09 0.25

[0067] The precision of the method was investigated by performing six parallel determinations on the same e-cigarette liquid sample (Table 5).

[0068] Table 5: Precision of Comparative Example 1 Samples (n=6) (Unit: mass fraction, %)

[0069] compound average value RSD (%) Menthol 12.5 1.32 WS3 1.22 1.20 WS23 1.05 1.08

[0070] Using the same e-cigarette liquid sample as the matrix, menthol, WS3, and WS23 were spiked at three levels: low, medium, and high, at 1%, 6%, and 24% (mass fraction increment based on the original sample), respectively. The contents of the original sample were determined and back-calculated according to the procedure of Comparative Example 1, and the spiked recovery rate was calculated. The results are shown in Table 6.

[0071] Table 6: Spike Recovery Rate of Comparative Example 1

[0072] compound Low concentration (%) Medium concentration (%) High concentration (%) Average recovery rate (%) Menthol 97 99 101 99.0 WS3 98 99 100 99.0 WS23 95 98 101 98.0

[0073] Comparative Example 2

[0074] The process is basically the same as in Example 1, except that propylene glycol is used as the solvent to prepare standard solutions of menthol, WS3 and WS23, and the concentrations are set the same as in Example 1.

[0075] Table 7: Standard operating curves of compounds in Comparative Example 2

[0076] compound Linear range (quality fraction, %) Regression equation (Y=aX+b) Correlation coefficient (R) Limit of detection (%) Limit of Quantification (%) Menthol 0.7-39 Y = 0.0205X + 0.0032 0.9994 0.10 0.35 WS3 0.7-39 Y = 0.0250X + 0.0030 0.9993 0.09 0.28 WS23 0.7-39 Y = 0.0225X + 0.0031 0.9994 0.08 0.30

[0077] The precision of the method was investigated by performing six parallel determinations on the same e-cigarette liquid sample (Table 8).

[0078] Table 8: Precision of Comparative Example 2 Samples (n=6) (Unit: mass fraction, %)

[0079] compound average value RSD (%) Menthol 11.8 1.45 WS3 1.12 1.30 WS23 0.98 1.21

[0080] Using the same e-cigarette liquid sample as the matrix, menthol, WS3, and WS23 were spiked at three levels: low, medium, and high, at 1%, 6%, and 24% (mass fraction increment based on the original sample), respectively. The contents of the original sample were determined and back-calculated according to the procedure of Comparative Example 2, and the spiked recovery rate was calculated. The results are shown in Table 9.

[0081] Table 9: Spike Recovery Rate of Comparative Example 2

[0082] compound Low concentration (%) Medium concentration (%) High concentration (%) Average recovery rate (%) Menthol 96 98 100 98 WS3 97 98 99 98 WS23 94 97 100 97

[0083] Comparative Example 3

[0084] The process is basically the same as in Example 1, except that acetic acid is used as the solvent to prepare standard solutions of menthol, WS3 and WS23, and the concentrations are set the same as in Example 1.

[0085] Table 10: Standard working curves of compounds in Comparative Example 3

[0086] compound Linear range (quality fraction, %) Regression equation (Y=aX+b) Correlation coefficient (R) Limit of detection (%) Limit of Quantification (%) Menthol 0.7-39 Y = 0.0185X + 0.0041 0.9992 0.12 0.32 WS3 0.7-39 Y = 0.0220X + 0.0035 0.9991 0.10 0.28 WS23 0.7-39 Y = 0.0200X + 0.0033 0.9992 0.11 0.30

[0087] The precision of the method was investigated by performing six parallel determinations on the same e-cigarette liquid sample (Table 11).

[0088] Table 11: Precision of Comparative Example 3 Samples (n=6) (Unit: mass fraction, %)

[0089] compound average value RSD (%) Menthol 10.9 1.52 WS3 1.08 1.40 WS23 0.92 1.34

[0090] Using the same e-cigarette liquid sample as the matrix, menthol, WS3, and WS23 were spiked at three levels: low, medium, and high, at 1%, 6%, and 24% (mass fraction increment based on the original sample), respectively. The contents of the original sample were determined and back-calculated according to the procedure of Comparative Example 3, and the spiked recovery rate was calculated. The results are shown in Table 12.

[0091] Table 12: Spike Recovery Rate of Comparative Example 3

[0092] compound Low concentration (%) Medium concentration (%) High concentration (%) Average recovery rate (%) Menthol 95 98 99 97.3 WS3 96 99 101 98.7 WS23 93 97 99 96.3

[0093] Example 1 utilizes a mixed solvent system of methanol and propylene glycol to maintain good solubility of menthol, WS3, and WS23 in the detection system and reduce the interference of matrix differences on the Raman signal, which is beneficial for effectively exposing the characteristic vibrational peaks of the target components. Compared with the single solvent or solvent systems with greater interference used in Comparative Examples 1-3, this mixed solvent can achieve a balance between polarity matching and background suppression, thereby improving signal stability and quantitative accuracy.

[0094] Example 2

[0095] A method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy includes the following steps: Step 1: Preparation of standard solutions: Using a mixed solution of methanol and propylene glycol at a volume ratio of 1:1 as the solvent, a series of standard solutions of menthol, WS3, and WS23 are prepared with mass fraction gradients of 0.7%, 1.3%, 3%, 6%, 11%, 24%, and 39%, respectively. The standard solutions are prepared according to a total mass M: Weigh the target substance mass m, make w = m / M × 100%, and then add the mixed solvent to the total mass M, mixing thoroughly. Step 2: Sample Preparation: Take 0.5g of e-cigarette liquid and add it to the mixed solvent described in Step 1 to a total mass of 5g, ensuring the expected content of the analyte falls within the standard curve range; oscillate at 180 r / min for 10 min, let stand for 2 min to defoam, and collect the supernatant. Add the enhanced nanomaterial dispersion to the supernatant to achieve a final concentration of nanoparticles of 0.1 mg / mL in the test solution. Vortex mix for 45 s, let stand for 6 min, and then transfer to a Raman sample cell for analysis. The nano-reinforcing material is modified SiO2-coated gold nanoparticles.

[0096] Step 3: Raman spectroscopy acquisition: Raman spectra of standard solutions and test sample solutions of various concentrations are acquired under the same conditions; wherein, before the determination, the standard solutions are also added with the same final concentration of nano-reinforcing material as in Step 2, and methanol-propylene glycol (1:1, v / v) solvent with the same final concentration of nanoparticles is used as blank background spectrum for background control and baseline correction.

[0097] Raman spectroscopy conditions:

[0098] The sample spectra were acquired using a 785 nm laser Raman spectrometer with a laser power of 50 mW and a spectral acquisition range of 200–2000 cm⁻¹. -1 Spectral resolution no higher than 4 cm -1 The integration time was 0.5-60 s each time, and the average of 1-50 measurements was taken to obtain the single measurement spectrum. Each sample was measured three times at different positions in the sample cell, and the average spectrum was taken as the final spectrum. A 520.7 cm⁻¹ silicon wafer was used before each batch of measurements. -1 Wavenumber calibration was performed, and a blank background spectrum was used as a background control and baseline correction, with methanol-propylene glycol (1:1, v / v) solvent containing the same final concentration of nanoparticles.

[0099] Raman spectra of menthol, WS23, and WS3 single-component standard solutions were acquired, and baseline correction and smoothing were performed to determine the characteristic peaks of each target component. The characteristic peak of menthol included a peak at approximately 547 cm⁻¹. -1 766cm -1 1036cm -1 and 1451 cm-1 The single-component characteristic peak of WS23 includes approximately 609 cm⁻¹. -1 769 cm -1 903 cm -1 926cm -1 1034 cm -1 1153 cm -1 and 1460 cm -1 The single-component characteristic peak of WS3 includes approximately 303 cm⁻¹. -1 478 cm -1 771 cm -1 916 cm -1 1036 cm -1 1140 cm -1 1174 cm -1 1452 cm -1 In this embodiment, menthol was selected at approximately 547 cm⁻¹ for simultaneous determination of the three components. -1 WS23 is approximately 1153 cm tall. -1 WS3 is approximately 1174 cm tall. -1 The characteristic peak at the specified location is used as the quantitative peak, while the remaining single-component characteristic peaks are used as auxiliary qualitative peaks. Since this embodiment introduces modified SiO2-coated gold nanoparticles as a reinforcing substrate, the aforementioned quantitative peaks have minimal overlap with the background peaks of the nano-reinforcing material in the reinforced system, exhibiting complete peak shapes and high signal-to-noise ratios, making them suitable for establishing stable quantitative models.

[0100] The original Raman spectra obtained from the standard solution and the sample solution were first subjected to blank background subtraction, then baseline correction was performed using the fifth-order polynomial fitting baseline correction method, and finally smoothed using the Savitzky-Golay smoothing algorithm with smoothing parameters of second-order polynomial and 11-point window.

[0101] In this embodiment, the peak height of the quantitative peak is uniformly extracted as the response value Y. A standard working curve is established with the mass fraction X of each component as the abscissa and the corresponding peak height Y as the ordinate, and linear regression is performed. The response value Y of the unknown sample is measured in the same way and substituted into the corresponding regression equation to calculate the mass fraction of menthol, WS3 and WS23 in the sample solution. The corresponding content in the original e-cigarette liquid is calculated by combining the sample weight and dilution factor.

[0102] In this embodiment, the limit of detection (LOD) is calculated as LOD=3σ / k, and the limit of quantitation (LOQ) is calculated as LOQ=10σ / k; where k is the slope of the standard working curve of each component, and σ is the standard deviation of the response value obtained at the corresponding quantitative peak position after the blank background solution is repeatedly measured 10 times under the same test conditions and processed in the same way.

[0103] The preparation method of the modified SiO2-coated gold nanoparticles is as follows:

[0104] S1. Dissolve chloroauric acid in water to prepare a 0.25 mmol / L chloroauric acid aqueous solution. Take 100 mL of the 0.25 mmol / L chloroauric acid aqueous solution and place it in a round-bottom flask. Heat to boiling and keep stirring under reflux. Add 10 mL of 38.8 mmol / L sodium citrate aqueous solution and continue reflux for 15 min. Cool to room temperature and let stand for 30 min. Centrifuge at 10,000 rpm for 15 min, discard the supernatant, and redisperse the precipitate with anhydrous ethanol. Repeat the washing process twice to change the system from the aqueous phase to the ethanol phase, obtaining ethanol-dispersed AuNP.

[0105] S2. Take the above-mentioned ethanol-dispersed AuNP and add anhydrous ethanol to a total volume of 100 mL.

[0106] Add 1 mL of ammonia and 1 mL of water, stir at room temperature for 10 min, add 110 μL of tetraethyl orthosilicate dropwise under stirring, continue the reaction for 5 h, collect the precipitate by centrifugation at 10000 rpm for 15 min, wash three times with ethanol, redisperse in ethanol, adjust the Au@SiO2 ethanol dispersion to a solid content of 3 mg / mL, add propyltriethoxysilane isocyanate to the dispersion to make the volume fraction of propyltriethoxysilane 1.1%, stir the reaction at 60 °C for 8 h, centrifuge and wash three times after the reaction to obtain modified nanoparticles;

[0107] S3. Hydroxypropyl-β-cyclodextrin was added to anhydrous DMF to obtain a cyclodextrin solution with a concentration of 30 mg / mL; triethylamine was added to make its volume fraction 1.25%; modified nanoparticles were added to the above cyclodextrin solution to make the final concentration of nanoparticles 3 mg / mL, and the mixture was stirred at 50℃ for 15 h. After the reaction was completed, the precipitate was collected by centrifugation at 10000 rpm for 15 min, and washed once with DMF, twice with ethanol, and twice with water. The obtained Au@SiO2-CD was redispersed in ethanol to prepare a 10 mg / mL enhanced nanomaterial dispersion, which was stored at 4 ℃ in the dark for later use.

[0108] Table 13: Standard operating curves of the compounds in Example 2

[0109] compound Linear range (mass fraction / %) Regression equation (Y=aX+b) Correlation coefficient (R) Limit of detection (%) Limit of Quantification (%) Menthol 0.7-39 Y = 0.0615X + 0.0031 0.9998 0.015 0.050 WS3 0.7-39 Y = 0.0742X + 0.0027 0.9998 0.012 0.040 WS23 0.7-39 Y = 0.0681X + 0.0029 0.9997 0.014 0.047

[0110] Table 14: Precision of Sample Measurement in Example 2 (n=6) (Unit: mass fraction, %)

[0111] compound average value RSD (%) Menthol 12.20 0.54 WS3 1.25 0.72 WS23 0.99 0.90

[0112] Table 15: Spike Recovery Rate (%) in Example 2

[0113] compound Low concentration (%) Medium concentration (%) High concentration (%) Average recovery rate (%) Menthol 99.4 101.2 100.3 100.3 WS3 98.7 100.5 101.0 100.1 WS23 99.1 100.2 102.0 100.4

[0114] Example 3

[0115] A method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy includes the following steps: Step 1: Preparation of standard solutions: Using a mixed solution of methanol and propylene glycol at a volume ratio of 1:1 as the solvent, a series of standard solutions of menthol, WS3, and WS23 are prepared with the same mixed solvent, and the mass fraction gradients are 0.7%, 1.3%, 3%, 6%, 11%, 24%, and 39%. The standard solution is prepared according to a total mass M: weigh the target substance mass m, make w = m / M × 100%, and then add the mixed solvent to the total mass M, and mix thoroughly; Step 2: Sample preparation: take 0.5g of e-cigarette liquid, add the mixed solvent mentioned in step 1 to the total mass 5g, so that the expected content of the analyte falls within the range of the standard curve; shake and mix at 180 r / min for 10min, let stand for 2min to defoam, take the supernatant, add the enhanced nanomaterial dispersion to the supernatant, so that the final concentration of nanoparticles in the test solution is 0.1 mg / mL, vortex mix for 45 s, let stand for 6min, and then transfer to the Raman sample cell for testing. The nano-enhancing material is SiO2-coated gold nanoparticles;

[0116] Step 3: Raman spectroscopy acquisition: Raman spectra of standard solutions and test sample solutions of various concentrations are acquired under the same conditions; wherein, before the determination, the standard solutions are also added with the same final concentration of nano-reinforcing material as in Step 2, and methanol-propylene glycol (1:1, v / v) solvent with the same final concentration of nanoparticles is used as blank background spectrum for background control and baseline correction.

[0117] Raman spectroscopy conditions:

[0118] The sample spectra were acquired using a 785 nm laser Raman spectrometer with a laser power of 50 mW and a spectral acquisition range of 200–2000 cm⁻¹. -1 Spectral resolution no higher than 4 cm -1 The integration time was 0.5-60 s each time, and the average of 1-50 measurements was taken to obtain the single measurement spectrum. Each sample was measured three times at different positions in the sample cell, and the average spectrum was taken as the final spectrum. A 520.7 cm⁻¹ silicon wafer was used before each batch of measurements. -1 Wavenumber calibration was performed, and a blank background spectrum was used as a background control and baseline correction, with methanol-propylene glycol (1:1, v / v) solvent containing the same final concentration of nanoparticles.

[0119] Raman spectra of menthol, WS23, and WS3 single-component standard solutions were acquired, and baseline correction and smoothing were performed to determine the characteristic peaks of each target component. The characteristic peak of menthol included a peak at approximately 547 cm⁻¹. -1 766cm -1 1036cm -1 and 1451 cm -1 The single-component characteristic peak of WS23 includes approximately 609 cm⁻¹. -1 769 cm -1 903 cm -1 926cm -1 1034 cm -1 1153 cm -1 and 1460 cm -1 The single-component characteristic peak of WS3 includes approximately 303 cm⁻¹. -1 478 cm -1 771 cm -1 916 cm -1 1036 cm -1 1140 cm -1 1174 cm -1 1452 cm -1 In this embodiment, menthol was selected at approximately 547 cm⁻¹ for simultaneous determination of the three components. -1 WS23 is approximately 1153 cm tall. -1 WS3 is approximately 1174 cm tall. -1 The characteristic peak at the specified location is used as the quantitative peak, while the remaining single-component characteristic peaks are used as auxiliary qualitative peaks. Because this embodiment introduces SiO2-coated gold nanoparticles as a reinforcing substrate, the aforementioned quantitative peaks have minimal overlap with the background peaks of the mixed solvent and the background peaks of the nano-reinforcing material. The peak shapes are complete and exhibit good repeatability, making it suitable for establishing a stable quantitative model.

[0120] The original Raman spectra obtained from the standard solution and the sample solution were first subjected to blank background subtraction, then baseline correction was performed using the fifth-order polynomial fitting baseline correction method, and finally smoothed using the Savitzky-Golay smoothing algorithm with smoothing parameters of second-order polynomial and 11-point window.

[0121] In this embodiment, the peak height of the quantitative peak is uniformly extracted as the response value Y. A standard working curve is established with the mass fraction X of each component as the abscissa and the corresponding peak height Y as the ordinate, and linear regression is performed. The response value Y of the unknown sample is measured in the same way and substituted into the corresponding regression equation to calculate the mass fraction of menthol, WS3 and WS23 in the sample solution. The corresponding content in the original e-cigarette liquid is calculated by combining the sample weight and dilution factor.

[0122] In this embodiment, the limit of detection (LOD) is calculated as LOD=3σ / k, and the limit of quantitation (LOQ) is calculated as LOQ=10σ / k; where k is the slope of the standard working curve of each component, and σ is the standard deviation of the response value obtained at the corresponding quantitative peak position after the blank background solution is repeatedly measured 10 times under the same test conditions and processed in the same way.

[0123] The method for preparing the SiO2-coated gold nanoparticles is as follows:

[0124] S1. Add chloroauric acid to water and mix thoroughly to prepare a 0.25 mmol / L chloroauric acid aqueous solution. Take 100 mL of the 0.25 mmol / L chloroauric acid aqueous solution and place it in a round-bottom flask. Heat to boiling and keep refluxing with stirring. Add 10 mL of 38.8 mmol / L sodium citrate aqueous solution and continue refluxing for 15 min. Cool to room temperature and let stand for 30 min before use. Centrifuge at 10000 rpm for 15 min, discard the supernatant, and redisperse the precipitate with anhydrous ethanol. Repeat the washing twice to change the system from the aqueous phase to the ethanol phase dispersion to obtain ethanol-dispersed AuNP.

[0125] S2. Take the AuNP dispersed in ethanol above, add anhydrous ethanol to a total volume of 100 mL, add 1 mL of ammonia water and 1 mL of water, and stir at room temperature for 10 min; under stirring conditions, add 110 μL of tetraethyl orthosilicate dropwise, continue the reaction for 5 h, centrifuge at 10000 rpm for 15 min to collect the precipitate, wash it 3 times with ethanol, redisperse it in ethanol, and adjust the Au@SiO2 ethanol dispersion to a solid content of 10 mg / mL.

[0126] Table 16: Standard operating curves of the compounds in Example 3

[0127] compound Linear range (mass fraction / %) Regression equation (Y=aX+b) Correlation coefficient (R) Limit of detection (%) Limit of Quantification (%) Menthol 0.7-39 Y = 0.0430X + 0.0030 0.9997 0.030 0.100 WS3 0.7-39 Y = 0.0520X + 0.0026 0.9997 0.025 0.083 WS23 0.7-39 Y = 0.0488X + 0.0028 0.9997 0.028 0.093

[0128] Table 17: Precision of Sample Measurement in Example 3 (n=6) (Unit: mass fraction, %)

[0129] compound average value RSD (%) Menthol 12.20 0.90 WS3 1.25 1.10 WS23 0.99 1.25

[0130] Table 18: Spike Recovery Rate (%) in Example 3

[0131] compound Low concentration (%) Medium concentration (%) High concentration (%) Average recovery rate (%) Menthol 97.8 99.6 100.8 99.4 WS3 97.2 99.1 101.3 99.2 WS23 97.5 99.4 100.9 99.3

[0132] Example 2 uses surface-modified and stably dispersed metal nanoparticles as the reinforcing substrate. The surface active sites are uniformly distributed, and the local electromagnetic field enhancement effect is stable, resulting in good consistency in the adsorption orientation and enhancement degree of menthol, WS3, and WS23, thus significantly improving signal intensity and repeatability. In contrast, the interface stability and hotspot consistency of the enhancement system in Example 3 are weaker, leading to increased signal fluctuations and lower overall detection sensitivity and precision compared to Example 2.

[0133] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy, characterized in that, Includes the following steps: (1) Preparation of standard solutions: Using the mixed solution as a solvent, prepare a series of standard solutions of menthol, WS3 and WS23 respectively; (2) Sample preparation: Take the e-cigarette liquid and add it to the mixed solution for dilution so that the expected content of the component to be tested falls within the range of the standard working curve corresponding to the series of standard solutions, and obtain the sample solution to be tested; (3) Raman spectroscopy acquisition: Raman spectra of standard solutions of various concentrations and the sample solution to be tested were acquired under the same conditions, and blank background spectra corresponding to the composition of the system to be tested were acquired for background comparison and baseline correction; (4) Data processing and quantification: The Raman spectra of the standard solution and the sample solution to be tested are subjected to consistent baseline correction and smoothing. The peak height or integral area of ​​the quantitative peak determined by screening in the characteristic peaks of menthol, WS3 and WS23 is extracted as the response value Y, and a standard working curve is established with the corresponding mass fraction X. The response value Y of the sample solution to be tested is obtained in the same way and substituted into the regression equation to calculate the mass fraction of each component. The contents of menthol, WS3 and WS23 in the original e-cigarette liquid are calculated back by combining the sample weight and dilution factor.

2. The method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy as described in claim 1, characterized in that: The mixed solution is a mixture of methanol and propylene glycol.

3. The method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy as described in claim 1, characterized in that: The mass fraction gradient of the series of standard solutions covers 0.1-50%.

4. The method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy as described in claim 1, characterized in that: The Raman spectroscopy was acquired using a 785 nm laser Raman spectrometer with a laser power of 10-200 mW and a spectral acquisition range of 200-2000 cm⁻¹. -1 .

5. The method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy as described in claim 1, characterized in that: Each integration time is 0.5-60s, and the number of accumulations is 1-50. The average of the results is taken to obtain the single measurement spectrum. Each sample is measured 2-6 times at different positions in the sample cell, and the average spectrum is taken as the final spectrum.

6. The method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy as described in claim 1, characterized in that: In the sample preparation step, a dispersion of nano-reinforcing material was added to the sample solution to make the final concentration of nanoparticles in the sample solution 0.01-1 mg / mL. The vortex mixing time was 10-120 s, and the standing time was 1-15 min before Raman spectroscopy was performed. The same final concentration of nano-reinforcing material was added before the standard solution was measured, and a methanol-propylene glycol mixed solvent with the same final concentration of nanoparticles was used as a blank background spectrum.

7. The method for simultaneously determining the contents of menthol, WS3, and WS23 in e-cigarette liquid based on Raman spectroscopy as described in claim 6, characterized in that: The nano-reinforcing material is modified SiO2-coated gold nanoparticles.

Images