WORK, PV DIAGRAMS, AND REVERSIBILITY

In 1824 a young French engineer named published Réflexions sur la puissance motrice du feu — Reflections on the Motive Power of Fire — the founding work of thermodynamics. It was almost entirely prose. Carnot reasoned about heat and engines in words and analogies, with scarcely a formula, and the book sank without trace.

Ten years later picked it up and did one decisive thing: he drew it. He put pressure on the vertical axis and volume on the horizontal, and suddenly every state of a gas was a point on a plane, every process a curve, and every engine cycle a closed loop. The abstract became visible. Engineers who would never follow Carnot's verbal reasoning could now read a heat engine off a chart. That chart is the PV diagram, and it is still how physicists think about gases two centuries later.

When a gas pushes a piston outward by a small volume $dV$, it does work against the pressure holding the piston in: a force $P\,A$ over a distance $dx$, and since $A\,dx = dV$, the work is $P\,dV$. Add up the increments along the whole expansion and you get the Thermodynamic work:

$W = \int_{V_1}^{V_2} P\,dV$

In words: the work done by a gas is the integral of its pressure over the change in its volume — geometrically, the area between the process curve and the volume axis. For the special case of constant pressure this collapses to the rectangle $W = P\,\Delta V$. The sign follows the physics convention from the first law: expansion ($\Delta V > 0$) is positive work out, compression is negative.

This is why the PV diagram is not just a bookkeeping device but a calculating one. You do not need calculus to read off the work — you measure an area.

Four idealised paths recur so often that they have names, and each is a distinctive shape on the plane.

An isobaric process holds pressure constant: a horizontal line, work $P\,\Delta V$. An isochoric process holds volume constant: a vertical line, zero work — the gas cannot push what does not move. An isothermal process holds temperature constant, so by the ideal gas law $PV = nRT$ is fixed and the path is a hyperbola $P = nRT/V$. An adiabatic process exchanges no heat at all; its curve $PV^{\gamma} = \text{const}$ is a steeper hyperbola than the isotherm, because an expanding gas with no heat coming in must cool, losing pressure faster.

Every curve drawn so far assumes something subtle: that the gas has a single, well-defined pressure at each instant. Squeeze a gas suddenly and it does not — it bunches up near the piston, swirls, and has no one pressure to plot. A path on the PV plane only exists if the process is quasi-static: carried out so slowly that the gas is in equilibrium at every step.

A Quasi-static process that is also frictionless can be run backwards through the very same states — that is a Reversible process. No real process is reversible; it is a limiting ideal, like a frictionless plane. But it is an indispensable one, because the maximum work a process can deliver, and the entire Second Law, are stated in terms of reversible paths.

The work gap between the slow and sudden expansions is not a rounding error. It is energy that became disordered and can never be fully recovered — the first hint of why the the second law forbids perfect engines.

Here is the deepest lesson the diagram teaches. Take a gas from state A to state B by two different routes — say, expand first then drop the pressure, versus drop the pressure first then expand. The two paths enclose different areas, so they do different amounts of work. By the first law $\Delta U = Q - W$, and since the internal energy $U$ is a state function that depends only on the endpoints, $\Delta U$ is the same for both routes. Therefore the heat $Q$ must also differ between them, by exactly as much as the work does.

$\Delta U = U_B - U_A \quad\text{(same for every path)}, \qquad W, Q \quad\text{(path-dependent)}$

In words: where you end up fixes the change in internal energy, but how you travel fixes how much of that change was paid in work and how much in heat. This is why you cannot speak of the "work in" or "heat in" a gas — only of work and heat that crossed its boundary along a particular path.

Close the path into a loop — return the gas to exactly the state it started in — and something clean happens. Over one full cycle the internal energy comes back to its initial value, so $\Delta U = 0$, and the first law reduces to

$\oint dU = 0 \quad\Longrightarrow\quad Q_{\text{net}} = W_{\text{net}} = \text{area enclosed by the loop}$

In words: over a cycle, the net heat absorbed equals the net work done, and both equal the area inside the loop on the PV diagram. The direction matters. A clockwise loop encloses positive work — it takes in heat and delivers net work, a heat engine. Run the same loop counter-clockwise and the work flips sign: now net work goes in and heat is pumped from cold to hot — a refrigerator.

Every real engine — steam, petrol, diesel, the Stirling cycle, the refrigerator in your kitchen — is some loop on a PV diagram, and its useful output is an area. Clapeyron's picture turned the design of engines into the geometry of curves.

We can now read work straight off a chart, distinguish the four canonical processes by their shapes, and see why a cycle's output is the area it encloses. Two of those processes deserve a closer look, because they sit at opposite extremes of how a gas trades heat for work: the isothermal path, pinned to a temperature by a slow exchange with its surroundings, and the adiabatic path, sealed off from heat entirely.

That is the subject of isothermal and adiabatic processes — why a bicycle pump scalds when you work it fast and barely warms when you work it slow, and where the steep adiabatic curve comes from. From there, the loops on this plane become real engines, and the question that consumed Carnot returns: of all the cycles you could draw, which one does the most work?