Evaluation of Presence or Absence of α-Tocopherol and α-Tocopheryl Acetate in Various E-Liquids

α-Tocopherol, α-tocopheryl acetate, 2,5-bis(5-tert-butyl-benzoxazol-2-yl)thiophene (Uvitex-OB), and 2,6-di-tert-butyl-4-methylphenol (BHT) as well as acetonitrile, formic acid, iso-propanol, and methanol were purchased from Sigma-Aldrich (St. Louis, MO, USA). E-liquids were commercially available and were purchased from the U.S. market in the period from 2018–2020. The GC vials were 2-mL with screw top caps with PTFE lined septa.

Two HPLC systems were used for the analysis. A1260–1290 HPLC system from Agilent (Wilmington, DE, USA) was used for UV detection. This HPLC system consisted of a binary pump, autosampler, column thermostatted compartment, and UV detector (diode array). The peak integration was performed with OpenLab ChemStation Edition. The other system was an Agilent 1200 HPLC binary system that consisted of a binary pump, an autosampler with cooling capability, and a column thermostatted compartment coupled with an API-5000 triple quadrupole mass spectrometer (AB Sciex, Framingham, MA, USA) controlled using Analyst 1.6.2 software.

The peak integration was performed with MultiQuant 3.0.1 software. Both HPLC systems were equipped with a Zorbax Eclipse XDB-C18 column, 4.6 × 250 mm, 5 μm, also from Agilent. An ultrasonic bath 5510 from Branson (Danbury, CT, USA) from Thermo Fisher Scientific was also used.

Prior the analysis the e-liquids were kept in a freezer at −20 °C. For the analysis, 100 mg (±10 mg) of e-liquid was precisely weighed in a 2-mL GC vial. 900 μL of methanol containing 560.4 μg/mL BHT, and 20 μL internal standard (I.S.) was added to the vial. BHT was added as an antioxidant. The stock concentration of the Uvitex-OB that was used as I.S. was 523.5 μg/mL in methanol. The sample was submitted for analysis by HPLC-UV and if a peak was detected at the corresponding retention time for α-tocopherol or of α-tocopheryl acetate, the same sample was submitted for analysis by HPLC-MS/MS.

The HPLC separation was isocratic using as solvent a mixture of 15% isopropanol in acetonitrile. The injection volume for UV detection was 20 μL. The detection by UV absorption was performed at 210 nm. The flow rate was 1.5 mL/min. For the MS/MS detection electrospray ionisation (ESI) technique has been used with multiple reaction monitoring (MRM) detection in positive mode. The mobile phase for the MS/MS analysis was the same as for the UV analysis but the acetonitrile contained 0.1% formic acid, and because the flow rate of 1.5 mL/min was too high for MS/MS system the effluent from the chromatographic column was split and only 0.5 mL/min were introduced into the mass spectrometer. Also, the injection volume was 10 μL. The transitions utilized for the MS/MS detection were as follows: for α-tocopherol 431.4 → 165.1, for α-tocopheryl acetate 473.4 → 165.1, and for Uvitex-OB (I.S.) 431.5 → 415.3. The experimental conditions for the MS/MS detection were the same as described in reference (16).

Quantitation has been performed using the same procedure as described in reference (16), using a set of 10 standards with concentrations between 98.63 μg/mL and 0.19 μg/mL for α-tocopherol and between 96.47 μg/mL and 0.19 μg/mL for α-tocopheryl acetate. The calibration was linear for both UV detection and for MS/MS detection, with equations of the form: [1] Y=a X+b {\rm{Y}} = {\rm{aX}} + {\rm{b}}

For the UV detection, the values for Y were the concentrations of the analyte while the values for X were the peak areas necessary for the calculation of the concentrations of the unknown samples. The peak areas of the internal standard were used in case of UV detection only to monitor the reproducibility of each analysis. The coefficients a and b for the calibrations in case of UV detection are indicated in Table 1.

For the quantitation using the MS/MS data, an equation of the same form as given by expression [1] was used, but in this case Y remained the concentrations of the analyte while the values for X were the peak areas of the analyte normalized by the area of the internal standard. The stability of the MS/MS detector is lower than that of the UV detector, and the normalization generated more precise values. The coefficients a and b for the calibrations in case of MS/MS detection are indicated in Table 2.

Both LC-UV and LC-MS/MS methods for α-tocopherol and α-tocopheryl acetate analysis can be considered as having good specificity. For the UV detection, retention time is basically the only criterion for selectivity, but for the MS/MS detection, besides the retention time the masses of the ion for Q1 (the parent ion) and the ion for Q3 (the daughter ion) are also important.

Since the e-liquids are miscible with methanol, no analyte recovery was performed in this study. The limits of quantitation (LOQ) for both α-tocopherol and α-tocopheryl acetate was taken as equal to the concentration of the lowest calibration standard, and because the samples are prepared using a 1/10 dilution in methanol, the LOQ per g of sample was 10 times lower (LOQ = 1.9 μg/g sample). The solutions of both α-tocopherol and α-tocopheryl acetate in the presence of BHT are stable up to two weeks when stored at −5 °C.

A variety of e-liquids containing nicotine were evaluated for the potential presence of α-tocopherol and/or α-tocopheryl acetate. The list of these e-liquids is given in Table 3. The table also indicates (when available) the nicotine level and content of glycerin and propylene glycol (PG). None of the e-liquids listed in Table 3 were found to contain either α-tocopherol or α-tocopheryl acetate at levels above limits of detection (LOD) of 0.6 μg/g.

In addition to the e-liquids containing nicotine, three e-liquids containing cannabidiol (CBD) were also evaluated. The list of these three e-liquids is given in Table 4.

The samples “Korent” and “Kore Organic” did not contain any α-tocopherol or α-tocopheryl acetate. However, in the LC-UV trace of the “Quanta” sample, a peak eluted at 11.904 min, a retention time very close to the retention time of α-tocopherol. No peak corresponded to the retention time of α-tocopheryl acetate. The HPLC chromatogram with UV detection for the “Quanta” sample is shown in Figure 1.

The retention time of the suspected α-tocopherol peak in the “Quanta” sample does not match exactly the retention time for α-tocopherol, and the peak shape is far from Gaussian. For this reason, it was suspected that an interference is responsible for the peak at 11.73 min in Figure 1.

Figure 1

However, the sample was also analyzed by LC-MS/MS procedure. The extracted ion trace for α-tocopherol in the LC-MS/MS chromatogram for the “Quanta” sample is shown in Figure 2.

Figure 2

The result from Figure 2 indicated that α-tocopherol is indeed present in the “Quanta” sample. The quantitation based on the LC-MS/MS trace indicated that the level of α-tocopherol in the sample is 10.2 μg/g (RSD% = 3.71%). The calculation based on the area count from the UV trace indicated a different, significantly higher value, indicating that the peak in the UV trace has an interference.