3.3 Effect of independent variables on the γ-tocopherol purification
The mixed tocopherols were separated by thin-layer chromatography (TLC) first with a mixture of n-hexane, anhydrous ethanol, isopropyl alcohol, tetrahydrofuran and ethyl acetate as the developing agent. After a large number of tests, the mixed solution of n-hexane and ethyl acetate was finally chosen as the eluent for column chromatography tests.
Mixed tocopherols were separated and purified using column chromatography. As shown in Fig. 4a, we could know that the eluent ratio had a great influence on the purification of γ-tocopherol. With the increase of the volume ratio of n-hexane to ethyl acetate, the polarity of eluent decreased gradually, and the purity and yield of γ-tocopherol increased at first and then decreased. When the volume ratio of n-hexane to ethyl acetate increased from 88.5:11.5 to 94.5:5.5, the purity and recovery rate of γ-tocopherol increased from (76.22 ± 3.81) % and (43.08 ± 4.31) % to (90.06 ± 1.09) % and (82.98 ± 2.75) %, respectively. This may be due to the fact that when the eluent ratio is less than 94.5:5.5, the polarity of eluent was too strong to make γ-tocopherol be separated from other tocopherol monomers. When the volume ratio of n-hexane to ethyl acetate was more than 94.5:5.5, the polarity of eluent is too weak, and the time of separation process was too long, which caused the material diffused seriously in the column, so the separation effect is poor. For example, when the volume ratio of n-hexane to ethyl acetate was 97.5:2.5, the purity and recovery yield of γ-tocopherol were (76.06 ± 0.98) % and (64.16 ± 3.08) %, respectively. From what has been discussed above, the hexane/ethyl acetate eluent with a volume ratio of 94.5:5.5 was elected as the subsequent optimization tests.
The effects of loading amounts on the γ-tocopherol purification were shown in Fig. 4b. The purity and recovery rate of γ-tocopherol increased first and then decreased when the loading amounts of the sample increased from 0.25 g to 1.25 g. The maximum purity and yield of γ-tocopherol were (98.49 ± 0.29) % and (92.18 ± 0.34) %, respectively (loading amounts 0.5 g). Theoretically, when the other conditions of the column are certain, the maximum loading amounts is certain. Under the condition that the loading amounts was not overloaded, the larger the sample loading was, the better. But when the sample weight was 0.25 g, which did not exceed the maximum loading amounts of chromatography column, however, the purity and recovery rate of γ-tocopherol were only (56.97 ± 0.23) % and (54.06 ± 0.29) % respectively. This can be explained by the fact that much material were adsorbed on the silica gel column, and only a small portion of the target product and most of the impurities were cleared off (Duran et al. 2011). When loading amounts of sample was more than 0.5 g, the purity and recovery rate of γ-tocopherol were reduced gradually, for example, the purity and recovery rate of 1.25 g samples decreased by (10.82 ± 0.16) % and (41.09 ± 0.30) %, respectively, compared with that of 0.5 g samples. This may be due to the sample overload, the chromatographic band of each component overlapped in the separation process, which resulted in poor separation effect (Chen et al. 2017). Therefore, 0.5 g was selected as the best loading amounts of sample.
The ratio of column height to diameter was also an important factor, relative to the theoretical plate number of chromatography theory, therefore an appropriate ratio of column height to diameter(H/D) was really necessary for column chromatography. The effect of different ratios of column height to diameter (12:1, 14:1, 16:1, 18:1, 20:1) on the γ-tocopherol purification were shown in Fig. 4c. With the increase of the ratio of column height to diameter from 12:1 to 20:1, there was no significant change in the purity of γ-tocopherol (p>0.05), and the recovery rate of γ-tocopherol increased first and then decreased. When H/D value were 16:1 and 18:1, the purities and recovery yield of γ-tocopherol reached the maximum values, which were (98.89 ± 0.68) %, (93.23 ± 0.89) % and (98.49 ± 0.29) %, (92.18 ± 0.34) %, respectively. There was no significant difference in purity and recovery between the two (p > 0.05). When the H/D values were 12:1 and 14:1, the recoveries of γ-tocopherol were only (48.07 ± 3.06) % and (70.14 ± 1.88) %, the reason may be that when the ratio of column height to diameter was too small, there was not sufficient time to complete the adsorption resolution process for each component. When the H/D value was 20:1, the recovery rate of γ-tocopherol decreased, which may be due to the serious diffusion of the substance on the silica gel column. Therefore, a 16:1 ratio of column height to diameter was selected as the optimal ratio for the purification of γ-tocopherol.
An appropriate elution flow rate is crucial for attaining a good yield and elution capacity. The effects of elution flow rate on the γ-tocopherol purification were shown in Fig. 4d. The maximum purity (98.89 ± 0.68%) and recovery yield (93.23 ± 0.89%) of γ-tocopherol were obtained with the elution flow rate of 2.0 mL/min. When the elution flow rate was above 2.0 mL/min, the recovery yield of γ-tocopherol decreased significantly. The low separation efficiency can be explained by the fact that the flow was too fast to allow the components to complete adsorption or desorption during the purge process (Chen et al. 2017). Generally speaking, slow flow rate exerted a positive effect for adsorption process because adsorbate molecules had sufficient time to complete adsorption or desorption of the components (Liu et al. 2011). However, the separation time was much longer and resulted in diffusion of substances during the separation process with a slow flow rate (Wan et al. 2008).
The optimal purification conditions were determined as follows: elution solvents, n-hexane/ethyl acetate (94.5:5.5, v/v); loading amounts, 0.5 g; the ratio of column height to diameter, 16:1; and elution flow rate, 2.0 mL/min. Under these optimized conditions, the purity and recovery yield of γ-tocopherol were (98.89 ± 0.68) % and (93.23 ± 0.89) %, respectively.
3.4DynamicDesorption Tests
The dynamic desorption process was described at the optimum experimental condition (Fig. S1). Results showed that the impurities were eluted first, and then the four tocopherols were eluted in the following order: α-T, (β+γ) and δ-tocopherol. After six hours of elution, the impurities and α-tocopherol were completely eluted, at which point γ-tocopherol appeared in the eluent and δ-tocopherol appeared in the eluent 8.5 hours later. Therefore, the γ-tocopherol can be well separated by this method.