Hello Faiyaz,

Can you, please, describe how you are measuring the thermal conductivity of the metal powders? It would help to know something about your experimental setup, i.e., a photograph with some accompanying text. For example, how do you apply mechanical pressure to the powder, how do you create a temperature gradient, how do you measure the heat flux, what is the size of your metal particles, etc.

Your observations of thermal conductivity versus temperature might be a measurement artifact, but it is hard to tell without more information. After all, we are not there looking over your shoulder.

Regards,

Tom Cuff

Faiyaz Ahsan

Knowing the temperature range that you're examining this property at will also help - along with the experimental conditions (in vacuo?)

Hello James Garry and Thomas Cuff for your comments. I am using laser flash method to conduct the experiments, where laser irradiates the sample from the top. The time taken for the bottom of the sample to heat up is tracked and used to calculate the thermal diffusivity. The average size of the powder particles was 75 micron ( varies between 45-110 um).

I measured the diffusivity at a temperature range of 50-800 degree celsius. Compressed nitrogen was used to avoid any oxidation. I saw the non-monotonic behavior with all the alloys I used ( Inconel, TI6al4v, SS 316L etc.)

If you could shed some lights for this particular behavior, it will be of great help to me!

Faiyaz Ahsan

Mmm.

I can imagine that radiative coupling will become an increasingly important transfer mode - so I would expect a rise in effective thermal conductivity.

The observed fall is interesting. That's a puzzle.

Hello Faiyaz Ahsan

Thank you for your description of your experimental setup.

Let me make an observation. You are employing a laser flash diffusivity measurement setup. Such a setup only measures diffusivity directly, see my RG page for a complete derivation, https://www.researchgate.net/project/Laser-Flash-Diffusivity-Measurment , based on the equations presented in the book: R. P. Tye (editor); Thermal Conductivity, Volume 2; Academic Press; 1969; pp. 190-196. Because the experimental setup measures transient, as opposed to steady state, heating, you cannot measure the thermal conductivity directly. Thus, you are inferring the value of the thermal conductivity by solving the expression D=k/(rho*Cp), where D is the measured thermal diffusivity, k is the unknown thermal conductivity, rho is the mass density, and Cp is the heat capacity at constant pressure.

You did not explain to us, how you obtained the values of rho and Cp. If the sample of inconel you had used was a bulk solid, cast, for example, then you simply could have looked up the parameters, rho and Cp, in the literature and substituted them back into the equation for D. How did you obtain the values for these two parameters? This is a crucial question since your sample is a pressed powder and not a bulk solid. For a pressed powder, one would expect that both rho and Cp would be significantly different from their corresponding bulk solid values.

If you look in any standard textbooks on heat transfer, you will sometimes find data on the thermal conductivity of a type of inconel, though not necessarily the type you are using, as a function of temperature. For example, the following textbook has this type of data: Frank P. Incropera, David P. DeWitt; Fundamentals of Heat and Mass Transfer, Second Edition; John Wiley & Sons; 1985; p. 757 (Appendix A). Their data is for inconel X-750 (73% Ni, 15% Cr, & 6.7 % Fe - not sure why the percentages don't add up to 100%, perhaps a typo), and their thermal conductivity data goes from 100K to 1,500K and monotonically increases with increasing temperature in this temperature range.

I am glad James Garry asked you about the temperature range of your measurements. I assume that the sample holder for your sample is located inside a cylindrical furnace so the you can set the ambient temperature to whatever you like, and that the laser beam is shot down the center of the long axis of this furnace via a window.

Have you the ability and facilties to allow a direct measurement of the thermal conductivity employing a steady state thermal procedure? The advantage of a steady state procedure is that you need not worry about rho and Cp for the sample itself. There is also the thermal conductivity comparator (steady state) method, if you have this equipment available.

Note, you did not explain exactly how you used the nitrogen gas to press the inconel powder, and how you compensated for the nonzero linear coefficient of thermal expansion of your powder throughout the temperature range you employed.

I am still not sure what exactly is going on with your measurement, but there seems to be plenty of room for measurement artifacts. Did you perform a literature search to see if anyone else has tried to do similar measurements?

Regards,

Tom Cuff

Hello Faiyaz Ahsan and James Garry ,

I checked a reference volume for diffusivity versus temperature for alloys such as nickel + chromium alloys (Inconel) and iron + nickel + chromium alloys (SS 3161). And over the range of temperatures you (Faiyaz) cited (50-800C), the diffusivity usually increased with increasing temperatures. These results were, of course, for bulk solid samples, not powders. The reference book was: Y. S. Touloukian et al.; Thermophysical Properties of Matter, Volume 10 - Thermal Diffusivity; Plenum Publishing Corp.; 1974; pp. 298-301 (Ni + Cr) & 344-352 (Fe + Ni + Cr). Note, the graphes of diffusivity versus temperature are found on pp. 298 and 344, and they cover a very wide temperature range >1,200K. This reference book is very good and it discusses all the measurement modalities used by researchers whose data is presented.

The most non-monotonic behavior was seen well below -100C for both Inconel and SS 3161, see Fig. 119 on p. 298 (Ni + Cr) and Fig. 136 on p. 344 (Fe + Ni + Cr).

You (Faiyaz) mentioned Tl6Al4V, which I assume is an intermetallic compound, but I did not find it listed in the Grouping of Materials List on pp. xix-xxvi.

Regards,

Tom Cuff