@article {van:12 November 2000:1087-6596:606, author = "van der Tempel L.", author = "Melis G.P.", author = "Brandsma T.C.", title = "Thermal Conductivity of a Glass: I. Measurement by the Glass-Metal Contact", journal = "Glass Physics and Chemistry", volume = "26", year = "12 November 2000", abstract = "
The thermal (phonon) conductivity of glass has been measured by contacting the sample with a metal at a different uniform initial temperature. The subsequent temperature response in the metal is measured by a tiny thermocouple just underneath the (contact) surface. The coefficient of heat penetration 
c_p follows directly from the fitted asymptotic temperature jump or drop for long times. Division by the separately measured heat capacity
c_p yields the thermal conductivity
. The conductivity measurement reproducibility was
= 3%. The standard deviation between validation measurements and round robin test results on Pyrex glass was
= 5.8%, somewhat more than the accuracy
= 5.2% of the round robin test results. The measurement method is insensible for slight imperfections of the thermal contact and infrared radiation diffusion (photon conductivity) in a hot glass. The method has been used with minor modifications for solid and molten samples at temperatures of 50850°C and conductivities of 0.125 W/(m K). The thermal (phonon) conductivity of the investigated soda-lime silicate glasses increases slightly (
2730%) with temperature from ambient up to around the glass transition.