Experiment 1 - homemade reactors A and B

This experiment series features a “biochar reactor”: from a steel pipe nipple with end caps and a 1/8" vent hole (replaceable later with a tube fitting) - Reactor A. And, an alternative reactor made from a one-gallon paint can, also with 1/8" vent hole - Reactor B

Experiment 1:
Measure the burner’s heat output by timing the boiling of a measured water volume.

The propane regulator is rated 0.6 m3/h. Net heating value of propane is 88341 kJ/m3; therefore the theoretical maximum regulator output is 53000 kJ/hr = 14700 W.

Heat capacity of water is 4.184 kJ/kg K. Time to heat 1L(1 kg) from 25°C to 100°C is 4.184*(100-25)/(53000 kJ/h) = less than 30 seconds if all heat could be transferred to the water.

The gas burner is equipped with a control valve and a mixing orifice - the mixing orifice appears to be the limiting feature of the burner output, though if the orifice were drilled larger, another aspect of the burner design could become limiting, such as the mixing tube, the flame orifices on the burner, the air inlet passage, or perhaps some material of construction might overheat.

Consider heat transfer coefficient and heating efficiency: overall heat transfer coefficient is Ux and Q (heat transfer) = Ux∆T – calculate U for each reactor based on actual time to boil water. On May 4, I observed temperatures on the surface of the empty reactor B at ¼ turn and ½ turn open gas flow and note that the heat transfer into the can is through the bottom primarily via conduction - see discussion below. The maximum attainable temperature on the inner surface of the can bottom is about 470 C at ¼ turn and 510 C at ½ turn. Take a value of 500 C for external temperature when calculating heat transfer coefficient.

Mass of empty reactors:

A: 1769 g + 193 g (support stand) = 1962 g

B: 337 g (including 52 g lid)

Cp(steel) = 0.49 kJ/kg °Cw

When heating reactor A, I observe that the support stand seems hotter than the reactor - heat transfer between the support and the reactor is poor. For experimental analysis of heat transfer effects, I will ignore the base mass. But when weighing the reactor, the support base will typically be used, so I will subtract its mass.

Experimental data and analyses are presented in the spreadsheet, tab = “Heat Xfer”

Benefits of reactor A: durable. reactor B: better heat transfer, greater volume, safety; lid can pop off. reactor B is lighter weight. reactor A end-caps are leak-prone. Therefore, focus on reactor B usage.

Figure 1, 2: Reactors A and B.

*The flame temperature of propane is reported to be 1980°C, but an unknown amount of ambient air rapidly mixes with the combustion gasses by convection around the burner. A pine splint (autoignition temperature near 500°C) will only ignite in the immediate vicinity of the open flame. The bucket, with larger area in proximity to the flame, is found to have the greater relative heat transfer coefficient. Heat transfer coefficients, while somewhat arbitrary for purposes of calculating heat transfer rate at consistent conditions, can be made more accurate when a measure of heating gas temperature is available. Observation of the temperature of the bottom of the empty bucket on the flame (below), and the location of ignition of the pine splint both support a temperature average near 500 C across the bottom of reactor B.

Conclusions from Experiment 1:

Reactor B has a heat transfer coefficient of approximately 2.4 W/°C; the expected delta T for pyrolysis will be less, corresponding with pyrolysis temperatures somewhat greater than prevailed while heating and boiling water.