Physics of Heat Transfer in Glass Windows

[Sherwood Botsford]

What is the physics behind heat transfer between two panes of glass?

Commonly windows now are filled with argon (some cost) or krypton (pricey) At a given temperature all gas molecules have the same energy per mode, so heavy ideal gas molecules move more slowly than light ones, so heat diffuses more slowly.

I know that it gets messier with molecules rather than atoms. There are more degrees of freedom, so energy goes into rotating and vibrating the molecule. I would expect that there would be 'impedance' issues in moving energy in and out of these various states. Can energy at one end of a molecule affect the energy transfer to the pane in the timeframe of a collision?

For a window fill, there are some pragmatics: The gas needs to have a low enough boiling point that it has a reasonable partial pressure at any temperature that the window encounters. Propane, for example liquefies at -40, making it an un-candidate in our winters that can dip into the -50s

It also needs to be reasonably cheap. The ideal spacing goes down as the molecular weight increases, but for argon it takes about 12 liters of gas per square meter of window (1/2" spacing)

So why aren't gasses such as SF6, freons, etc used for windows? You are going to get some additional heat transfer from vibrational modes, but the the overall greater mass would seem on the face of it to compensat

 

[Guy Madison]

Argon is probably used because it's so commonly available for welding operations and used for numerous other applications, why use anything else?

 

[sophiecentaur]

So why aren't gasses such as SF6, freons, etc used for windows?

Freons are being phased out because they are greenhouse gases (afaik). That would be a bad idea, bearing in mind that they would escape from every broken window. They are being phased out as refrigerants too, for the same reasons.
I thought that the reason for using a dense gas was that the mobility was lower so convectional heat transfer is much less. It's a compromise between cost and effectiveness.

 

[Sherwood Botsford]

@Guy As I understand it, diffusion is proportional to the speed of the molecule. Argon with a molecular weight of 40 is 33% heavier than air's average of about 29. So argon atoms move about 15% slower than air molecules. Argon is also a single atom, so has no rotation energy component. Diatomic adds rotation, which is why Ar is more than 15% better than air. (It's about 35% better)

SF6 has a molecular weight 146 -- about 5 times that of air. So molecular speed should be about 1/sqrt(5) that of air. So on a basis of molecular speed, it should be about 2.3 times as good an insulator.

I don't understand how the rotational modes transfer energy during the collision

@Sophie. Not all freons are equally evil. The problem with many of them is that they had chlorine in them which turned into ozone depletion. Some are potent green house gasses too, but in the quantities used, are not significant. More benign refrigerator gases are being substituted. In general, I would expect any good freezer gas to be a possible candidate as a window glazing fill gas. (SF6 is a potent greenhouse gas too, but it's presence it the atmosphere is measured in parts per trillion.)

 

[Guy Madison]

@Guy As I understand it, diffusion is proportional to the speed of the molecule. Argon with a molecular weight of 40 is 33% heavier than air's average of about 29. So argon atoms move about 15% slower than air molecules. Argon is also a single atom, so has no rotational energy component. Diatomic adds rotation, which is why Ar is more than 15% better than air. (It's about 35% better)

SF6 has a molecular weight 146 -- about 5 times that of air. So molecular speed should be about 1/sqrt(5) that of air. So on a basis of molecular speed, it should be about 2.3 times as good an insulator.

I honestly think the reason is cost and availability, I can buy a bottle of argon about 3 miles away from my house... there are other gases used for glass panes that are closer together like krypton but that's kind of expensive. There are better solutions but what is good enough? Think about volume manufacturing methods and the answer becomes quite obvious. It's readily available, it's cheap and there are no EPA issues with argon.

See https://windowguru.wordpress.com/2014/08/29/air-vs-argon-vs-kyrpton-vs-xenon/

"The ROI on Krypton and Xenon are probably not worth the upfront cost. Argon is an absolute no brainer."

FYI, I replaced my windows and it made little or no difference other than the noise. I had honeycomb blinds which added an extra layer of insulation. But I live in a fairly temperate climate.

 

[sophiecentaur]

OF course, there are also coatings available for reflecting IR and UV, to help both in summer and in winter. You can get some pretty useful U values for not too much money. (Though whether it repays the cost over the short term is debatable.)

 

[oz93666]

I believe the parameters we need to consider are the thermal conductivity of the gas , and how soon convection currents start in the the glazing gap

Argon has a conductivity of 0.0092--34% lower than that of air--and it is, by far, the most common low-conductivity gas for windows.
, Krypton has a conductivity of 0.0051, which is 63% lower than that of air... Xenon has a conductivity that's 79% lower.

SF6 is no good at all 11.627 mW/(m.K)... twice as conductive as Xenon 5.65mW/(m·K)..

The problem with krypton and xenon is that convection currents start up more easily , so the glazing gap has to be smaller to prevent this ...

Where there maybe room for optimisation is in the glazing gap , if you live in a mild climate then you can make this wider , greatly improving insulation ... the point at which convection starts within the gap, (and ruins insulation) , , is dependent on temp difference , and width of gap , for a given gas.

Some evidence to indicate banning of SF6 was politicians being influenced by corporations... the patent was about to expire loosing billions in revenue ...an excuse was needed to ban it ...ozone