[rhade (OP)]

Apparently, it is not the convention to refer to degrees Kelvin; one should only say "Kelvin." Why is this the case?

[Geezer]

The definition of the unit, the kelvin, makes the "degrees" redundant. Similarly, we just use V for volts, W for watts, J for joules etc.

[damocles]

With other temperature scales the "degrees" is needed because there is no sense in which a temperature of 20°C is "double" a temperature of 10°C. But with kelvin temperatures there are many ways that things do scale in that way -- for example an ideal monatomic gas has just double the kinetic energy at 400 K to what it has at 200K, or it exerts double the pressure if it is in a closed vessel. So it makes sense to say that a temperature of 400 K is double a temperature of 200K, and that makes the kelvin scale quite different to any scale that needs the "degrees".

There is a body called the International Union for Pure and Applied Chemistry that sets standards for notation and terminology in Chemistry at least: it makes rulings on how units and quantities are to be named, accepted values for constants, accepted names for chemicals, and the like. (It is often regarded as overly pedantic, and has been forced to back down on some of its rulings over the last several decades). There are presumably similar authorities in Physics and Engineering.

[Geezer]

Come to that, the "degrees" in "degrees Celsius" is just as redundant. Why don't we just say 20C?

[damocles]

Come to that, the "degrees" in "degrees Celsius" is just as redundant. Why don't we just say 20C?

No, it is not quite as redundant. We say "degrees" celsius because celsius is not some sort of amount of stuff. 2 metre is twice as much as 1 metre, 2 kilogram is twice as much as 1 kilogram, 2 kelvin is twice as much as 1 kelvin, but 2 degrees celsius is NOT twice as much as 1 degree celsius (unless we are talking about temperature difference rather than temperature as such!)

[Geezer]

Come to that, the "degrees" in "degrees Celsius" is just as redundant. Why don't we just say 20C?

No, it is not quite as redundant. We say "degrees" celsius because celsius is not some sort of amount of stuff. 2 metre is twice as much as 1 metre, 2 kilogram is twice as much as 1 kilogram, 2 kelvin is twice as much as 1 kelvin, but 2 degrees celsius is NOT twice as much as 1 degree celsius (unless we are talking about temperature difference rather than temperature as such!)

A change in temperature of 1 degree C is exactly the same as a change in temperature of 1degree K, and a doubling of temperature in K does not mean there is twice as much of any particular "stuff".

Inserting "degree" does not help. It's an anachronism.

[damocles]

A change in temperature of 1 degree C is exactly the same as a change in temperature of 1degree K, and a doubling of temperature in K does not mean there is twice as much of any particular "stuff".

Inserting "degree" does not help. It's an anachronism.

Geezer, 2K does mean twice as much "temperature" as 1K. 2°C does NOT mean twice as much temperature as 1°C. Temperature as "stuff" is expressed in such other quantities as heat energy, average kinetic energy, gas pressure, etc.

(and I had already pointed out that temperature change was quantifiable in °C in a way that temperature per se was not).

[Geezer]

A change in temperature of 1 degree C is exactly the same as a change in temperature of 1degree K, and a doubling of temperature in K does not mean there is twice as much of any particular "stuff".

Inserting "degree" does not help. It's an anachronism.

Geezer, 2K does mean twice as much "temperature" as 1K. 2°C does NOT mean twice as much temperature as 1°C. Temperature as "stuff" is expressed in such other quantities as heat energy, average kinetic energy, gas pressure, etc.

(and I had already pointed out that temperature change was quantifiable in °C in a way that temperature per se was not).

Yes, but that's all it means. There certainly is not twice as much heat or some other "stuff".

By the way, another thing that blows a hole in your argument is that the degree symbol is still used in Rankine temperatures  []

[damocles]

A change in temperature of 1 degree C is exactly the same as a change in temperature of 1degree K, and a doubling of temperature in K does not mean there is twice as much of any particular "stuff".

Inserting "degree" does not help. It's an anachronism.

Geezer, 2K does mean twice as much "temperature" as 1K. 2°C does NOT mean twice as much temperature as 1°C. Temperature as "stuff" is expressed in such other quantities as heat energy, average kinetic energy, gas pressure, etc.

(and I had already pointed out that temperature change was quantifiable in °C in a way that temperature per se was not).

Yes, but that's all it means. There certainly is not twice as much heat or some other "stuff".

By the way, another thing that blows a hole in your argument is that the degree symbol is still used in Rankine temperatures  []

IUPAC (youpack) is not particularly interested in rankine temperatures. I am fairly sure that the reasoning I have outlined is the basis of their deliberate decision to leave the degrees in celsius but delete it from kelvin.

As someone who taught chemistry for about 40 years I got very used to confused students who would claim that air sealed in a tin at 20°C would have a pressure of 2 atmosphere at 40°C, when they were being taught that air sealed in a tin at 200K and 1 atmosphere would have a pressure of 2 atmosphere at 400K.

[Geezer]

C is not an SI unit. K is.
 
Including a degree symbol with K would be completely inconsistent with the system, and because it's redundant, it was logical for SI to dump it. I'm sure IUPAC have their reasons for hanging on to the symbol in conjunction with C, but it's not because there is anything very special about K. If that were the case, the degree should have been dropped in R too.
 
I'd have no objection to abandoning the use of C (or F) completely, but I suspect I would be in the minority. I sympathize with your students' confusion, but I cannot see why including a degree symbol would help or hinder them.

[rhade (OP)]

As regards your latter point, Geezer, outside of science, very few people use Kelvin (if any) and I think celsius is a nice, easy system for people to get their heads around (freezing and boiling points of water, if you disregard the complications, like it depends on how clean the water is). I think only the Americans still use Fahrenheit, nationally anyway. As for the rest of this, I will let you guys argue it out. But thanks to both of you for responding!

[Geezer]

Kindly don't interrupt! We are having a pointless academic debate here.  []


EDIT: Of course Damocles is quite right about the Celsius scale. It really does confuse things a lot. We tend to think that an air temperature of 20C (I'd put in the degree symbol, but we seem to have lost the ability to do that - SI probably had something to do with that) is 100% hotter than a 10C air temperature.

In fact, it's only 3.5% "hotter". If we used K all the time, we would not be confused. With C it's a bit like saying a length of 20 meters is really zero meters, and anything shorter has a negative length.