- #1
xavier_Ghz
- 3
- 0
Hi all,
I do an calorimeter experiment at temperatures around 40 mK. To get more grip on the time constants that are associated with the heat flows, I calculate the thermal resistance as well as the heat capitance of the materials in the set-up.
We assume that Newton's law of cooling is the only relevant process, as the compartiments of the cryostat are all in a vacuum (no convection) and the nearest plate from the dilution refrigerator (14 mK) has a temperature of 50 mK (so the importance of radiative heat transfer is negligible).
My question is: what is Cbrass @ mK temperatures? I can't find it anywhere.
Let's assume that I use the common 70/30 brass. I found somewhere (lost the link) that
Cbrass = 3*10^-2 J / (Kg*K) at T = 1 K.
This is somewhat comparable to that of copper, as
CCu = 1.3*10^-2 J / (Kg*K) at T = 1 K.
However, the experiments for copper are done for the submillikelvin temperatures, which yield
CCu = 6*10^-4 J / (Kg*K) at T = 40 mK, (Fig. 3.12 from 'Matter and Methods at Low Temperatures by Pobel).
There, on a log(C)-log(T) scale, almost all curves decrease in a linear way. Should I assume that this is also the case for Cbrass to extrapolate to T = 40 mK? With a slope that is comparable to that of the heat capacity of copper? I probably make a lot of mistakes then, but it would give me
Cbrass = 1.4*10^-3 J / (Kg*K) at T = 40 mK.
Your help is much appreciated!
Xavier
I do an calorimeter experiment at temperatures around 40 mK. To get more grip on the time constants that are associated with the heat flows, I calculate the thermal resistance as well as the heat capitance of the materials in the set-up.
We assume that Newton's law of cooling is the only relevant process, as the compartiments of the cryostat are all in a vacuum (no convection) and the nearest plate from the dilution refrigerator (14 mK) has a temperature of 50 mK (so the importance of radiative heat transfer is negligible).
My question is: what is Cbrass @ mK temperatures? I can't find it anywhere.
Let's assume that I use the common 70/30 brass. I found somewhere (lost the link) that
Cbrass = 3*10^-2 J / (Kg*K) at T = 1 K.
This is somewhat comparable to that of copper, as
CCu = 1.3*10^-2 J / (Kg*K) at T = 1 K.
However, the experiments for copper are done for the submillikelvin temperatures, which yield
CCu = 6*10^-4 J / (Kg*K) at T = 40 mK, (Fig. 3.12 from 'Matter and Methods at Low Temperatures by Pobel).
There, on a log(C)-log(T) scale, almost all curves decrease in a linear way. Should I assume that this is also the case for Cbrass to extrapolate to T = 40 mK? With a slope that is comparable to that of the heat capacity of copper? I probably make a lot of mistakes then, but it would give me
Cbrass = 1.4*10^-3 J / (Kg*K) at T = 40 mK.
Your help is much appreciated!
Xavier
Last edited by a moderator: