0

With thermodynamics, systems are studied using macroscopic variables (pressure, temperature, volume, etc.) which do not need a mechanical explanation, which is what statistical mechanics deals with. So I wonder if it is possible to do thermodynamics without atomic theory (as positivists like Mach would have done. To follow this program it would be necessary first of all to do chemistry without atomic theory. So is it possible to do chemistry entirely with equivalents? I tried. The concept of equivalent is complex, but a definition can be given for each type of reaction. Despite this, however, I was not able to completely eliminate the mole. for example the mole (or the number of molecules) already takes place in the equation of state of ideal gases. In addition, the first equation of state for real gases is that of Van der Waals. It was obtained with molecular hypotheses. However, it is true that the equation of state for real gases can be obtained with a series development of the gas compression factor. But only statistical mechanics tells me if this series is convergent. In addition I am not sure if we can eliminate the concept of atom, molecule and subatomic particles (electrons):

  1. in chemical thermodynamic systems that deals with the equilibrium of reactions,
  2. in thermodynamic that deal with electric and magnetic systems (however, it is true that in this case electromagnetism uses continuous charge distributions)
Roger V.
  • 58,522
user273366
  • 11
  • 1
  • 2
    Thermodynamics and statistical physics are two different ways of looking at the same thing. The former indeed does not require atomic theory, see, e.g., [here](https://physics.stackexchange.com/questions/714220/what-actually-are-microscopic-and-macroscopic-viewpoints-in-thermodynamics/714229#714229 and here – Roger V. Jun 24 '22 at 06:27
  • 1
    @RogerVadim, why not post your comment as an answer? -NN – niels nielsen Jun 24 '22 at 07:26
  • Well, e.g. temperature dependence of molar heat capacity of multiatomic gases (progressive involving of quantized rotational and later vibrational modes) cannot be addressed without molecular quantum approach. Or generally, any macroscopic behavior which is result of behaviour or properties of atoms or molecules. – Poutnik Jun 24 '22 at 07:28
  • 1
    @Poutnik it cannot be explained without using microscopic approach, but it can be measured and used for practical purposes, as did the Gibb's contemporaries throughout the industrial revolution in XIXth century (while the development of statistical mechanics took place in the beginning of the XXth). – Roger V. Jun 24 '22 at 07:31
  • @RogerVadim Well, classical motion of objects cannot be explained without Newtonian mechanics or the equivalent, but it can be measured and used for practical purposes without it. One could have e.g. an empirical table for the given object, how long it takes to fall from a particular height, without any notion of speed nor acceleration. – Poutnik Jun 24 '22 at 07:33
  • 1
    @Poutnik indeed, and people built palaces and sailing ships long before Newton. It ultimately boils down to the difference between thermodynamics and statistical mechanics, a phenomenological and a microscopic theory - I won't pursue this discussion, since I explained it in the answers linked. – Roger V. Jun 24 '22 at 07:37

1 Answers1

2

Thermodynamics and statistical physics are two different ways of looking at the same thing. The former indeed does not require atomic theory, see, e.g., here and here.

Also relevant: Timeline of thermodynamics and statistical mechanics

Roger V.
  • 58,522
  • It does not require it, but there will be left mysteries for more advanced courses. BTW, I remember having in (deep) past an undergraduate course Statistical thermodynamics. – Poutnik Jun 24 '22 at 07:40
  • @Poutnik well, even Landau&Livshits do not discriminate the two in their text, which leads to lots of confusion. Yet, many texts do it in parallel, and the two certainly existed independently for some time. I recommend Prigogine's book as a modern example of doing thermodynamcis without statistical physics. – Roger V. Jun 24 '22 at 07:44
  • 1
    The name trascription is Lifshitz (Russian: Евге́ний Миха́йлович Ли́фшиц ). Thanks for all your responses. – Poutnik Jun 24 '22 at 07:49
  • @Poutnik yes, you are right. – Roger V. Jun 24 '22 at 07:50