Results and Discussion

96

collected in the gasbag were analyzed with gas chromatography (GC), and H2, CO, CO2, and C1−C4 hydrocarbons were quantified.

97

mechanism of HO internal recycling (in next section) since the structure of char was difficult to make clear.

4.3.2. The Comparison of Product Distribution.

In contrast to the process with HO internal recycling, higher liquid yield was obtained from the process without HO recycling, but the water content was as low as 35 wt %. As listed in Table 4-1, no significant difference was found to the gas yields from both processes, which was in good agreement with the result from Section 4.3.1. If the change of gas yield was ignored, the decrease in the oil yield was considered to be equivalent to the increases in the char and water yields due to the HO internal recycling. However, this did not necessarily mean that the oil was only converted to char and water. There were two possibilities for the oil change:

one was that the HO was only converted to char and water, while the light oil was left and had no change; the other was that the HO was converted to not only char and water, but also light oil. The detail would be discussed in Section 4.3.3, which supported the latter. In any cases, reactions with dehydration polymerization were considered to play an important role during the HO charring.

4.3.3. The Comparison of Liquid Products between BO1 and BO2.

Compared with BO1, BO2 was a single phase liquid, which was probably caused by a relatively lower water content. Volatility is an important property for a liquid fuel. TGA was employed to investigate the volatility of BO2, as demonstrated in Figure 4-4. In contrast to Figure 3-9, a significant difference found in this figure was that BO2 left around 3 wt % of residue even heating up to 900 °C, which was undesirable in many applications and downstream treatments, such as turbine fuel¹⁴ and catalytic upgrading¹⁵. Such residue-forming portion in BO1 was completely removed because of the internal recycling of HO.

98

The detailed qualitative and quantitative GC/MS analyses were carried out to BO1 and BO2 with external standard method. Some compounds were quantified using the response factors of other compounds with similar structures because of their non-commercial availability. Although the details were not shown, two different concentrations of standard solutions and their peak areas together with origin of coordinates were employed to draw a calibration curve. The peak areas of compounds in BO1 and BO2 were adjusted by methanol dilution to be less than their maximum peak areas in standard solutions. The total ion chromatograms of BO1 and BO2 were exhibited in Figure 4-5, and the respective compounds were listed in Table 4-2. The result from Table 4-2 showed that there were 4 kinds of changes among the compositions of BO1 and BO2 because of the HO internal recycling: new formed compounds of Numbers 1 to 20; increased contents in Numbers 21 to 45; reduced contents in Numbers 46 to 54; removed compounds of Numbers 55 to 71. All of benzene and its homologues were either new formed or increased in their contents due to the HO internal recycling, while most of the compounds with relative macromolecular weights and nitrogen-containing compounds were either removed or reduced in their contents.

The most significant change caused by HO internal recycling was that levoglucosan, the most abundant compound in BO2 with a content of as high as 0.18 g/g-tar, was completely converted. Levoglucosan acted either as a product of cellulose pyrolysis or as an intermediate for the formation of other products.^16,17 Kawamoto et al.^18,19,20 reported levoglucosan could degrade to lower-molecular-weight products or polymerize into polysaccharides, which would be carbonized to form char; there was competition between the formation of lower-molecular-weight products and solid char. Self-charring of HO might include such

99

polymerization of levoglucosan to char. Furthermore, the pyrolysis vapors from cellulose and lignin induced the secondary decomposition of levoglucosan²¹, which indicated that co-pyrolysis of HO and biomass affected the char yield to a more or less extent. In addition to this, chemical interactions between vapors and pyrolyzing solid, such as chemical vapor deposition (CVD), were commonly taken place in a moving bed reactor and increased char yield significantly. Levoglucosan included hydroxyl groups that were easily interacted with char active sites (such as hydroxyl groups). After a condensation reaction between two hydroxyl groups, a polymer of char-levoglucosan and water were produced, and the polymer could be further carbonized to char. That was why the yields of char and water increased simultaneously, which was described in Section 4.3.2. The increase in the char yield resulted from HO internal recycling was thus considered to be mainly caused by reactions of levoglucosan, of which the extent was decided by the cellulose content in the feedstock if the conditions of pyrolysis were kept same. The same result was drawn from Table 3-5 in Chapter 3. In comparison to the char yields of nc = 1 and 6 in Table 3-5, a similar increase in the char yields was found for the woody biomasses (pine, mallee and cherry), whose cellulose contents accounted for 40–50 wt %²², while a less char yield growth took place to rice husk with a cellulose content of around 30 wt %²³, which also suggested that the cellulose content of the parent feedstock was an important factor to influence the increase in the char yield because of the HO internal recycling.

On the other hand, as shown in Table 4-2, some furans and phenols, such as 5-methylfurfural, furfural, phenol, 4-methylphenol, 3-methylphenol, 2-methylphenol, 2,5-dimethylfuran, and so on, became rich by 2–4 times with the help of the HO internal recycling. Acetic acid, the most abundant compound in BO1, was also

100

improved in its content by 1.68 times. The increases in the contents of those compounds were considered to mainly result from the direct decomposition of the HO. Although the composition of HO was not clear, some precursors of acetic acid, furans, and phenols were expected to be present in the HO. Re-pyrolysis of those precursors in the recycled HO released acetic acid, furans, and phenols. Therefore, HO was converted to not only char and water but also light oil during the internal recycling, which was mentioned in Section 4.3.2.