2.2. Duckweed pyrolysis experiments
The experimental setup consists of a quartz tube reactor with the internal diameter of 13 mm heated by a tube furnace (‘‘Thermolyne”). Typically, a 2.27–2.53 g sample of duckweed (pre-dried at 120 C overnight) was placed inside the reactor which was connected to an argon (Ar) tank having a delivery pressure of 34.5 kPa. The flow rate of Ar sweep gas was varied in the range of 36–150 mL/min using a metering valve and a calibrated flow metering rotameter. A K-type thermocouple was inserted into the center of the heated zone for controlling the temperature via a PID controller (Eurotherm 2116). The reactor outlet was connected to a PyrexTM glass condenser which was cooled by ice. Prior to each run, the reactor and all connecting lines were purged with Ar. After the furnace temperature reached the desired level and stabilized, the section of the reactor containing the duckweed sample was inserted into the heated zone and kept there for 15 min. During the run, the pyrolysis liquid was condensed on the quartz tube walls outside the furnace and in the ice cooled condenser. The non-condensable gases were allowed to accumulate in a Teflon gas bag. The pyrolysis reaction was carried out at the temperature range of 400–700 C. At the end of the run, the tube furnace was switched off and the gas collected in the sampling bag was analyzed gas-chromatographically. The pyrolytic liquid product was collected by rinsing the reactor with dichloromethane (DCM) and gathering the liquid in the condenser (which already had some portion of pyrolysis oil condensed there during pyrolysis). DCM with dissolved liquid products of pyrolysis was left overnight to evaporate (small amount of water formed during pyrolysis was also evaporated). The resulting bio-oil was weighed and analyzed on a gas chromatography–mass spectrometer (GC–MS) system. The solid residue (char, or bio-char) remaining after each test was collected and weighed. The experimental data on the duckweed pyrolysis products yields were subjected to the one-way analysis of variance (ANOVA) as implemented in the GraphPad InStat 3 statistics platform. In order to ascertain that the observed variations in the yield of pyrolytic gas, bio-oil and bio-char as a result of changes in the residence time (expressed in terms of argon flow rate) and pyrolysis temperature are statistically significant, the probability (P) values were determined. If the P values were small (post-test following one-way ANOVA), then it is unlikely that the differences in the values of dependent variables (i.e., yield of gas, bio-oil and char) observed are coincidental and due to random sampling.