0

I want to ask about infinitesimals and non-standard analysis. In physics we always use $\mathrm dx,~\mathrm dv,~\mathrm dt$ etc. as infinitesimal quantities. When we deduce equations in physics, when we set up integrals and in many other instances we use infinitesimals. For example in the first law of thermodynamics we have: $đQ=\mathrm dU+đW$.

Now when we ask teachers about what is $đQ$ here, they say it is an infinitesimal amount of thermal energy. So my question is, is it necessary to treat the whole non-standard analysis as an axiom for physics or do we have standard calculus explanation for physicist's use of $\mathrm dy/\mathrm dx$ as a quotient in all cases?

Qmechanic
  • 201,751
Nesar
  • 41
  • physics uses non-rigorous but intuitive math tools, which of course could be wrong. In most (or at least many cases though) they are later formalized by mathematicians to provide a rigorous foundation. –  Jun 04 '16 at 05:06
  • If you search Wikipedia for "forms" and "Relativity", you will see how the use and interpretation of infinitesimals is now used in modern physics. –  Jun 04 '16 at 05:15
  • 4
    Actually, we never really use infinitesimal quantities. We just pretend that we do. What we mean by those is that if someone decided to make a measurement, they could find a small enough, but finite, measurement interval on which the differential relationship is "good enough". Calculus is merely the mathematical approximation for these finite differences that is highly convenient to work with. In reality everything that starts with a $d$ in physics is really just a small enough $\Delta$. When you see mathematicians who are trying to formalize physics, you are seeing wasted effort. – CuriousOne Jun 04 '16 at 05:21
  • 3
    Related: https://physics.stackexchange.com/q/92925/2451 , https://physics.stackexchange.com/q/70376/2451 , https://physics.stackexchange.com/q/65724/2451 , https://physics.stackexchange.com/q/153791/2451 and links therein. – Qmechanic Jun 04 '16 at 06:09
  • @Nesar Just curious, what do you mean by "non-standard analysis"? Asking because the term is usually associated with Robinson's "non-standard analysis", https://en.wikipedia.org/wiki/Non-standard_analysis. – udrv Jun 04 '16 at 07:48
  • @udrv Yes, I am talking about Robinson's Non-standard analysis. – Nesar Jun 04 '16 at 12:20
  • In that case still no, it certainly isn't necessary, afak not even formally. Keep in mind that the differential form of the first law was set up way before (a century?) Robinson developed his framework and functions well provided the difference between đQ and dW etc is well understood. Perhaps this difference could be formalized deeper using non-standard concepts, I see the appeal, but not sure it would be of any help beyond the setup point. – udrv Jun 04 '16 at 15:39
  • @CuriousOne: if we set up an equation (describing a real world problem) in terms of real (finite) $\Delta y$ and $\Delta x$ (with a view of computing $y=f(x)$) and then apply the limit definition on $\lim_{\Delta x \to 0}\Delta y/\Delta x$ are we not really using infinitesimals? – Gert Jun 04 '16 at 15:40
  • @Gert: If we try to measure the velocity of a classical particle at ever smaller scales, let's say for particles of a few $\mu m$ size, we are encountering Brownian motion, and if we go down to the $nm$ scale, we are seeing quantum mechanics take over. Trying to extend Newtonian mechanics to ever smaller scales breaks at the level of physical phenomena, so it doesn't really make sense to say that the theory is valid for infinitesimal quantities. – CuriousOne Jun 04 '16 at 18:15

1 Answers1

1

Physicists do you the rules of calculus but can be sloppy in their notation.

The symbols $\delta, \Delta$ and $d$ often seem to be used interchangeably to mean a (small or infinitesimal) change in something or better still a final value minus an initial value.

So in your equation $dU$ is the change in internal energy of a system or final internal energy minus initial internal energy $U_f-U_i$.

However with this notation the use of $dW$ and $dQ$ is often frowned on as there is no initial amount of heat and final amount of heat it is just the amount of heat leaving or entering the system and the same is true of the work done.

To indicate this instead of using $d$ a new symbol $d$ with a line/bar through it is used (I do not known how to write this but it like $\hbar$). That being said if you look through many textbooks $\delta$ and $d$ are used but often with an explanation that they represent inexact differentials.

Farcher
  • 95,680