First Law of Thermodynamics is expressed as[1]

The internal energy of an isolated system is constant.

If $w$ denotes the work done on a system, $q$ for the energy transferred as heat to the system, and $\Delta U$ for the resulting change in internal energy, then it follows that $$\Delta U = q + w \label{1}\tag{1}$$

$$\mathrm{d}U = \mathrm{d}q + \mathrm{d}w \label{2}\tag{2}$$

What I don't understand is how does $\eqref{2}$ follow mathematically from $\eqref{1}$. Taking the limit $\Delta U \rightarrow 0$ gives the left-hand side but how does $q$ change into $\mathrm{d}q$ and $w$ into $\mathrm{d}w$? I understand the interpretation of both the equations physically. Is this related to the fact that $U$ is a state function whereas $q$ and $w$ are not?


[1] Atkins, P.; de Paula, J. Physical Chemistry, 10th ed; Oxford UP: Oxford, U.K., 2014.

  • 4
    $\begingroup$ This is the reason why some authors do not write dq and dw, but use a sort of crossed symbol of d, that I am unable to represent here. It looks like a d with a superimposed / bar. $\endgroup$
    – Maurice
    Feb 9, 2020 at 10:47
  • $\begingroup$ It's just the authors playing fast and loose (inventing their own version) with the mathematics. $\endgroup$ Feb 9, 2020 at 12:53
  • 4
    $\begingroup$ Mathematically, neither dq neither dw are total differentials, but their sum is. $\endgroup$
    – Poutnik
    Feb 9, 2020 at 12:56
  • $\begingroup$ Some texts use $\delta$ to indicate inexact differentials. Not to be confused with $\partial$. $\endgroup$
    – Andrew
    Feb 28, 2020 at 19:35

1 Answer 1


The first law of thermodynamics is a statement of energy conservation thus and defines the internal energy E as an extensive state function. In an infinitesimal transformation, the first then law reduces to $$\mathrm{d}E = \mathrm{d}Q + \mathrm{d}W$$

where $\mathrm{d}E$ is a total (exact) differential for infinitesimal transformation.

However, $\mathrm{d}Q$ and $\mathrm{d}W$ are not exact ($Q$ and $W$ are not state functions); $Q$ and $W$ in a thermodynamics transformation are process-dependent. All of these are properties of functions of more than one variables


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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