After stopping cryocoolers in IXC, we observed 3 out gassing peaks at IXA position as a top graph of the attached file.
1)23.8K (note this peak actually could appear at smaller temperature if the mirror minimum temp was much cooler for this corresponding gas.)
2)30.1K,
3)36.4K.
The second graph shows the possible crossing point btw these temperature and several gas's saturated vapor pressure.
Assuming the pressure level inside the cryostat is smaller than that at IXA, the stars positions are possible.
It shows,
23.8K -> Ar. or O2 (these two curves are almost same with each other, so we cannot distinguish them)
30.1K -> CH4 ??
36.4K -> Kr ??.
The air components is (1975 International Model Air Components)
Gas Volume(%) ppm
N2 78.084 780,840
O2 20.9476 209,476
Ar 0.934 9,340
CO2 0.0390 390
Ne 0.001818 18.18
He 0.000524 5.24
CH4 0.000181 1.81
Kr 0.000114 1.14
SO2 0.0001> 1>
H2 0.00005 0.5
N2O 0.000032 0.32
Xe 0.0000087 0.087
O3 0.000007> 0.07>
NO2 0.000002> 0.02>
I2 0.000001> 0.01>
Kr and CH4 are not so negligible components.
According to Michimura-kun's summary(http://klog.icrr.u-tokyo.ac.jp/osl/?r=9731),
Ar/O2 could be the most possible source for frosting because the corresponding temp is btw 20K and 30K where we encountered the frosting ?
One peak was visible at 23.0 K.
No other peaks around 30k.
The data was missing at 36.4K.
N2(28), CH4(16), O2(32), Ar(20,40), H2O(18) and so on did exist at "IXC" (not EXC) cryostat, while Kr(36) was not so obvious.
Note that this was measurement at room temp area, not inside the cryostat.