Why would the Kelvin temperature scale not just be based on the freezing point of water, which is only .01 degrees Celsius off from it? Is this just a more accurate way of measuring and quantifying the value?

  • 1
    $\begingroup$ It's independent of pressure, otherwise someone could ask if melting in, say, one bar or one atmosphere. $\endgroup$
    – Mithoron
    Sep 12 '18 at 22:32
  • 1
    $\begingroup$ I'm voting to close this question as off-topic because it's not really about chemistry. $\endgroup$
    – Mithoron
    Sep 12 '18 at 22:33
  • 4
    $\begingroup$ The triple point is unique, the melting temperature isn’t. $\endgroup$
    – Jon Custer
    Sep 13 '18 at 0:09
  • 5
    $\begingroup$ I don't see how the selection of a standard is not about chemistry. $\endgroup$
    – A.K.
    Sep 13 '18 at 3:05

In reality all the "primary" scientific values have changed over time as experimental techniques have improved. Science has evolved to the point where a set of seven "SI base units" (ampere, kelvin, second, meter, kilogram, candela, and mole) have been selected and defined so that better measurements won't cause disruptive changes in scientific data.

So the gist is that the "temperature of ice water" depends on the atmospheric pressure. So it is necessary to define that 0 °C is the temperature of ice water at "standard atmospheric pressure." That means that if the "standard atmospheric pressure" changes, then so does the "temperature of ice water at standard pressure."

However the triple point of water doesn't change and has a unique value. So by defining the triple point of water as exactly 0.01 °C in 1954, then the Celsius scale is anchored as much as the Kelvin scale. By also defining absolute zero as 0 K and −273.15 °C then the temperature scale itself won't change if better ways to measure temperature are devised.

Thus by the time the whole notion of the primary base units was sorted out there was a massive amount of scientific data that had been accumulated using the old definition of the Celsius scale that was based on ice water at a standard atmosphere. Letting the triple point be exactly 0.01 °C kept all of that "old" temperature data "reasonably accurate" since 0.01° is only a 0.01/273.15=+0.0037% change.

  • $\begingroup$ It's less than 0.0037% change. The old melting point standard was 0°C=273.15 K at one atmosphere, whereas the triple point is a little higher than the one atmosphere melting point. You guessed it -- that difference is almost exactly 0.01°C or K. $\endgroup$ Sep 13 '18 at 12:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.