On Oct 12, 2009, at 7:01 AM, Jed Rothwell wrote:
In earlier reports I expressed concern about the calorimetry
employed by Kitamura et al. The flow rate of the cooling fluid is 6
ml per minute.
I wonder if there is a typo, i.e. it should be 6 ml/s.
This seems too low to me. Most people I know who use flow
calorimeters recommend at least 30 to 60 ml per minute. At 6 ml I
fear the fluid is not mixed and the temperature sensor may pick up
a streamline of warm or cold water.
I discussed this with Takahashi. He says there is a mechanical
mixer to ensure a uniform fluid temperature at the temperature
sensors. This is not shown in the schematic, but the schematic is
somewhat abstract.
Takahashi points out that the low power levels in this experiment,
a faster flow rate will reduce the sensitivity of the experiment
too much.
Other researchers use methods to ensure mixing that depend upon the
motion of the fluid, or the pressure. McKubre used a Venturi, and
Cravens has some in-line corkscrew shaped plastic stirrers. These
would not work at 6 ml/min. A mechanical stirrer should work fine.
- Jed
My 2 cents worth on that:
I built a small DC stirrer that ran at 1-2 V and about 0.3 W. It was
very effective for stirring a few hundred ml and worked by surface
tension on an axially rotating glass rod. That would add energy to
the stream at the rate of 0.3 J/s * 60 s = 18 J / min. That is 18 J
of heat added to 6 cm^3 of liquid, or 3 J/cm^3. The specific heat of
water is 4.186 joule/gram °C. Water has about 1 g/cm^3, so the
temperature of the water would thus be raised by a bit less than 1 °
C. A less energy demanding custom in-line stirrer could be made for
such a slow mass flow calorimetry stream if it is only of concern
that mixing occur in the mass flow stream tubing - although the
mixing problem, due to hot spots, may primarily exist in the
calorimeter envelope, not the flow tubing.
It is notable that such a slow a flow rate of 6 cm^3/min or 0.1 g/s
would limit the calorimeter to a maximum of (4.186 joule/gram °C)*
(0.1 g/s)*(70 °C) = 29 W total, assuming an input temperature of 30 °
C, and that would be pushing the calorimeter fluid to boiling, which
can cause errors.
It is also notable that at a more reasonable calorimeter wattage for
a 6 ml/min flow rate, say 10 W, that a poorly chosen stirrer, like
the typical magnetic lab kind, could introduce significant error.
Hopefully Kitamura was measuring a total of less than 20 W.
Best regards,
Horace Heffner
http://www.mtaonline.net/~hheffner/
