Investigation on the Lignin Removal From Tobacco Stem by a Solvothermal Method Using Ethylene Glycol as a Solvent

Tobacco, an annually harvested plant, is one of the most significant crops in the world in terms of profitability. It is cultivated extensively in China, Brazil, India and the USA (1). However, while most of the harvested tobacco is manufactured for production of cigarettes, using its high-quality leaves, a considerably quantity of tobacco stem, as byproduct, is wasted (2). Therefore, for the last four decades, the utilization of tobacco stem has been widely investigated, especially with intent of producing reconstituted tobacco sheet (3, 4, 5, 6, 7), which has been successfully industrialized. The advantages of such an approach mainly contribute to a reduction of the cost of cigarette manufacture as well as of the tumorigenic activity of mainstream smoke condensates and “tar” content. Meanwhile, reconstituted tobacco sheet could play a significant role in improving the combustion properties and also enhance the filling capacity of cigarettes. Regrettably, higher lignin content is noticed in tobacco stem than that detected in tobacco leaves, resulting in more harmful phenolic derivatives (8) and higher concentrations of carbon monoxides (9) released during combustion. Even when tobacco stem is treated at lower temperatures (e.g., 180 °C – 540 °C) in lieu of a combustion procedure (e.g., 600 °C – 900 °C), most of the lignin still undergoes a pyrolysis process that will generate undesired formaldehyde and propanal (10). Furthermore, higher lignin content leads to miscellaneous biological acitivities and a bitter smell during tobacco smoking and inhaling (11). Therefore, it is of great importance to lower lignin content in tobacco stem in order to improve the quality of reconstituted tobacco sheet and to facilitate its utilization.

There have been a few investigations on reducing the lignin content in tobacco stem, including the process utilizing ionic liquid (12), enzymes (2, 13, 14), microorganisms (15), acid/alkali involved treatment (16, 17). Unfortunately, none of above approaches has been industrialized due to their own drawbacks. In detail, ionic liquid and enzymes are rather expensive and less productive; efficiency of using microorganism is always an issue; acidic and alkaline taste cannot be avoided during the inhaling of smoke by applying acid/alkali involved treatment. On the other hand, alcohols have been reported for isolation of lignin from lignocellulosic biomass (18, 19), which is both efficient and low-cost without potentially affecting the quality of smoke sensory. Interestingly, to the best of this authors’ knowledge, to date there is no other investigation on tobacco stem using alcohols.

Therefore, in this work, reduction of lignin content in tobacco stem was investigated by using ethylene glycol. This was set in comparison to lignin removal from wheat straw and corncob under mild conditions. In order to understand the reason for differences in performance of ethylene glycol on tobacco stem and on other typical lignocellulosic biomass (e.g., wheat straw), the effect of tobacco-derived nicotine and solanesol was investigated on wheat straw, and n-hexane was applied to eliminate the effects of solanesol. Finally, the solanesol-free tobacco stem was treated with ethylene glycol to achieve a decrease in lignin content.

For comparison, wheat straw, corncob and tobacco stem were treated in ethylene glycol. The lignin content for wheat straw, corncob and tobacco stem before and after treatment is listed in Table 2 and the removal efficiency of lignin is demonstrated in Figure 1. As can be seen in Table 2, the highest lignin content of 18.50 wt/wt was observed from untreated wheat straw, while the lowest lignin content of 5.15 wt/wt was detected in the raw tobacco stem. Furthermore, it is obvious that lignin could be removed from all three samples after experiments, indicating the potential application of this approach. This can be rationalized in terms that strong hydrogen bond interactions between ethylene glycol and the free hydroxyl groups present in lignin will contribute to the lignin dissolution, which was also reported by Sun et al. (19). However, as observed from Figure 1, an evident decrease for removal efficiency of lignin was observed when tobacco stem was treated in comparison to that observed from corncob (e.g., 13.9% vs. 44.1%). It is therefore deemed that some tobacco-specific compounds might be the key to retarding the lignin dissolution in ethylene glycol from tobacco stem.

Figure 1

In order to understand the reason for different performances of ethylene glycol on tobacco stem and on other lignocellulosic biomass, wheat straw was chosen for further investigation by adding tobacco-derived nicotine and solanesol during the experiment. The efficiency of lignin decrease for wheat straw with addition of nicotine and solanesol is illustrated in Figure 2. It is clear that, by adding nicotine in the reaction system, similar removal efficiency of lignin is achieved in comparison to that observed from pure wheat straw. Interestingly, with addition of solanesol, the lignin deduction efficiency was drastically dropping from 39.3% to 0.8%. This implies that solanesol, as a typical compound derived from tobacco, plays a key role in hindering the lignin elimination from tobacco stem with ethylene glycol. Since the formed hydrogen bonds between ethylene glycol and lignin are of high significance to the removal of lignin, it is speculated that hydrogen bonds are pre-formed between ethylene glycol and solanesol, which further prevent the dissolution of lignin.

Figure 2

Based on the former understanding, to improve the efficiency of lignin removal from tobacco stem, n-hexane was applied to extract solanesol from tobacco stem. The solanesol content before and after n-hexane extraction is given in Table 3. As observed from Table 3, complete removal of solanesol from tobacco stem was obtained after n-hexane treatment. In addition, although the content of solanesol was not that high in tobacco stem, it is deemed that solanesol in tobacco stem could be dissolving out to react with ethylene glycol for the formation of hydrogen bond under such reaction conditions. This also explains the aforementioned theory about solanesol affecting lignin removal with ethylene glycol treatment.

The solanesol-free tobacco stem was then treated in ethylene glycol under the same reaction conditions and the efficiency of lignin removal is shown in Figure 3. From Figure 3, as expected, a significant improvement of lignin dissolution was observed when solanesol-free tobacco stem was used in lieu of raw tobacco stem without n-hexane pre-treatment, i.e., 40.5% vs. 13.9%. This is mainly due to the absence of solanesol, which further proves that solanesol is the main reason for preventing the removal of lignin from tobacco stem using ethylene glycol as a solvent.

Furthermore, the tobacco stem before and after experiments was characterized via FT-IR spectroscopy and scanning electron microscope (SEM). Figure 4 shows the development of absorbance intensity of the main functional groups that are identified from their characteristic bands, of which detailed information is given in Table 4. A broad band at around 3405 cm⁻¹ was observed for both treated and untreated tobacco stem, which can be attributed to the hydroxyl groups in phenolic and aliphatic structures. The peak centered at 2931cm⁻¹ was predominantly related to C–H stretching in aromatic methoxyl groups and in methyl and methylene groups of side chains. As to the bands detected at 1626 cm⁻¹, 1515 cm⁻¹ and 1428 cm−1, originate C–C stretching from aromatic skeleton in lignin (22). Furthermore, two bands arising from C–O stretching were shown at 1322 cm⁻¹ and 1248 cm⁻¹, attributing to syringyl and guaiacyl in lignin, respectively (22). It is important to mention that the strength of absorbance bands related to lignin after the experiment is obviously weaker than that observed from fresh tobacco stem sample. This implies that part of the lignin is successfully removed from tobacco stem after the treatment.

Figure 3

Figure 4

In addition, SEM images of native and treated tobacco stem are illustrated in Figure 5. As can be seen, the fresh sample has a compact and smooth structure. In comparison, wrinkles and fractures were observed from treated tobacco stem, demonstrating that part of lignin was eliminated. Textural properties of tobacco stem before and after treatment with solanesol and ethylene glycol are listed in Table 5. It can be observed that specific surface area (S_BET) of tobacco stem sample was increased from 0.8047 m² g⁻¹ to 3.2843 m² g⁻¹ after the treatment with solanesol and ethylene glycol, while the pore volume was raised from 0.0004 cm³ g⁻¹ to 0.0038 cm³ g⁻¹. This can also be attributed to the partial dissolution of lignin.

Figure 5

A solvothermal method using ethylene glycol was developed for the removal of lignin from tobacco stem. It was found that the removal efficiency of ethylene glycol on tobacco stem is much lower than that achieved on wheat straw and corncob, i.e., 13.9% for tobacco stem vs. 44.1% for corncob and 39.3% for wheat straw. This can be rationalized in terms of the presence of solanesol in tobacco stem, as confirmed by the results of lignin removal from wheat straw dropping from 39.3% to 0.8% with addition of solanesol. This could be explained in terms that solanesol retards the strong hydrogen bond interactions between ethylene glycol and the free hydroxyl groups present in lignin. On the other hand, when solanesol was eliminated by applying an n-hexane extraction procedure, 40.5% of lignin was successfully dissolved from tobacco stem. Furthermore, the successful removal of lignin was also confirmed by the characterization results of SEM, FTIR and N₂-sorption. The developed method provides a promising way to lessen the negative effect of lignin on utilization of the wasted tobacco stem for production of reconstituted tobacco sheet.