NIST Standard Reference Database 30
Last Update to Data Content: 2002
"Thermal Conductivity of Magnesium Oxide from Absolute, Steady-State Measurements," A.J. Slifka, B.J. Filla, and J.M. Phelps, Journal of Research of the National Institute of Standards and Technology, Vol. 103 [4], pp. 357-363 (1998), published by National Institute of Standards and Technology.Language: English
"The specimens are commercial, sintered disks of polycrystalline magnesium oxide. They are 69.75 mm in diameter and have thicknesses of 2.59 mm, 5.04 mm, and 7.64 mm. The surface finish of the specimens varies from specimen to specimen and ranges from 0.2 µm to 0.5 µm centerline average roughness. ... The purity of a representative specimen was checked using energy dispersive spectrometry (EDS) with both scanning electron microscope (SEM) and analytical electron microscope (AEM) samples. ... Characteristic x-ray maps showed that the impurities are distributed randomly throughout the sample. ... we estimate the total mass fraction of impurities to be around 1%."
"The specimens have 93% of the theoretical density, based on measurements of mass and dimensions, using 3.581 g · cm-3 as the theoretical density of magnesium oxide."Optical microscopy
"The average grain size is 25 µm, measured by optical microscopy of a fracture surface." No additional measurement details were noted.Steady-state heat flow
"The tests were done using a one-sided guarded hot plate, which is a modified version of the ASTM C 177 specification. ... The specimen rests between two sensor plates and experiences an upward, one-dimensional heat flow because the lower part of the measurement stack is an isothermal hot "cup". Since the apparatus operates at high temperature, up to 1300 K, a thermal grease cannot be used between the specimen and sensor plates. A pliable metal like indium cannot be used due to its low melting temperature, and metals that can handle the high temperature are too stiff to give intimate thermal contact over the large surface area of the 69.75 mm diameter specimen. Additionally, a metal foil to provide intimate thermal contact between specimen and sensor plates is not feasible because type-s thermocouples embedded in the surface of the sensor plates would short if a metal foil were used. We are left having to measure the thermal resistance between our specimen and sensor plates. The Fourier conduction equation in one dimension modified to include this additional thermal resistance term is ΔT · A/Q = Δx/k + 2·RT where ΔT is the temperature difference, A is the cross-sectional area that heat flows through, Q is the heat flow rate, Δx is the length over which the temperature difference is measured, k is the thermal conductivity, and RT is the specific interfacial thermal resistance. By measuring two specimens of different thickness, we can solve for the two unknowns k and RT. ... An uncertainty analysis of our system gives a 5% relative standard uncertainty for our experiments."
| Relative Density ( % ) | Density ( g cm-3 ) |
|---|---|
| 93 | 3.33 |
| Grain Size ( µm ) |
|---|
| 25 |
| Temperature ( K ) | Thermal Conductivity ( W m-1 K-1 ) |
|---|---|
| 400 | 29.4 |
| 450 | 26.6 |
| 500 | 23.8 |
| 550 | 21.6 |
| 600 | 19.7 |
| 650 | 18.3 |
| 700 | 16.6 |
| 750 | 15.3 |
| 800 | 14.3 |
| 850 | 13.1 |
| 900 | 12.3 |
| 950 | 11.7 |
| 1000 | 11.1 |
| 1050 | 10.5 |
| 1100 | 10.0 |
| 1150 | 9.5 |
| 1200 | 8.9 |
| 1250 | 8.3 |
| 1300 | 8.1 |