Chapter 8 · Thermodynamics

8.3 The Second Law and Reversible Processes

The first law says energy is conserved; the second law says which way it is willing to go. No engine can turn heat completely into work, heat never flows uphill by itself, and the best any engine working between $T_1$ and $T_2$ can do is the reversible Carnot efficiency $\eta=1-T_2/T_1$.

A petrol engine burns its fuel at well over $2000\ \text{K}$, yet only about a quarter to a third of the energy released ends up turning the wheels; the rest leaves through the radiator and the exhaust pipe. A coal or nuclear power station does a little better, but it still pours more than half of its heat into a river or a cooling tower. None of this is just poor engineering. The first law of thermodynamics would happily allow an engine that turned every joule of heat into work — and yet nobody has ever built one, and nobody ever will. Equally, the first law has no objection to a cup of tea getting hotter by taking heat from the cooler room around it; it simply never happens.

The rule that forbids these things is the second law of thermodynamics. In this section we describe the devices it is really about — heat engines, refrigerators and heat pumps — and state the law in its two classic forms (Kelvin–Planck and Clausius). We then separate reversible processes from the irreversible ones that happen in the real world, and use a reversible cycle, the Carnot cycle, to find the largest efficiency any engine can have between two temperatures: $\eta=1-T_2/T_1$. Throughout we use the sign convention of 8.2: $\Delta U=Q-W$, with $Q$ the heat given to the gas and $W$ the work done by the gas.

Heat engines, refrigerators and heat pumps

Turning work into heat is easy and can be done completely: rub your hands, stir a liquid, push a block across a rough floor. Turning heat into work continuously is harder. A gas expanding isothermally does convert all the heat it absorbs into work — but it ends up at a larger volume, and it cannot go on expanding for ever. To run for ever, a device must work in a cycle: its working substance returns to its starting state again and again. Over each complete cycle the internal energy of the working substance returns to its starting value, $\Delta U=0$, so the first law gives $Q_\text{net}=W_\text{net}$.

📘 Definition 8.9 — Heat reservoir

A heat reservoir (thermal reservoir) is a body with so large a heat capacity that it can give out or take in any amount of heat without any change in its temperature. The atmosphere, the sea, a river and a large furnace kept burning are good approximations. The hot reservoir an engine draws heat from is called the source (temperature $T_1$); the cold one it rejects heat to is the sink (temperature $T_2$, with $T_2\lt T_1$).

📘 Definition 8.10 — Heat engine and its efficiency

A heat engine is a device that, working in a cycle, takes in heat $Q_1$ from a source at $T_1$, converts part of it into work $W$, and rejects the remaining heat $Q_2$ to a sink at $T_2$. All three are magnitudes (positive numbers). Over one cycle $\Delta U=0$, so

$$W=Q_1-Q_2 .$$

The efficiency (thermal efficiency) of the engine is the fraction of the heat taken in that comes out as work:

$$\eta=\frac{W}{Q_1}=1-\frac{Q_2}{Q_1}.$$

$\eta$ is a pure number, often quoted as a percentage. Since $Q_2\ge0$, $\eta\le1$ by the first law alone; the second law will show that in fact $\eta\lt1$ always.

Every engine has the same four parts: a working substance (steam in a turbine, the air–fuel mixture in a car engine, an ideal gas in our models), a source, a sink, and a mechanism (piston, turbine) that delivers the work. In a car the "source" is the burning fuel inside the cylinder and the "sink" is the atmosphere; in a thermal power station the source is the boiler and the sink is the cooling water. Engines are usually drawn as an energy-flow diagram: a box (the engine) with an arrow $Q_1$ entering from the hot reservoir above, an arrow $Q_2$ leaving to the cold reservoir below and an arrow $W$ leaving sideways (Interactive 1).

Running it backwards. Reverse every arrow and you have a refrigerator: work $W$ is supplied (by an electric compressor), heat $Q_2$ is taken from a cold space at $T_2$ and heat $Q_1=Q_2+W$ is delivered to the warm surroundings at $T_1$. In a domestic refrigerator the working substance (the refrigerant) evaporates at low pressure in the coils inside the cabinet, absorbing heat, is compressed and then condenses in the black grid at the back, giving out heat to the kitchen. The same machine used to warm a house in winter — taking heat from the cold outdoor air or ground and delivering it indoors — is called a heat pump. An air conditioner is a refrigerator whose "cold space" is the room.

📘 Definition 8.11 — Refrigerator, heat pump and coefficient of performance

A refrigerator (or a heat pump) is a heat engine run in reverse: in each cycle work $W$ is done on the working substance, heat $Q_2$ is extracted from the cold reservoir at $T_2$ and heat $Q_1=Q_2+W$ is given to the hot reservoir at $T_1$. Its performance is measured by the coefficient of performance (COP) — what you want divided by what you pay for:

  • refrigerator (you want heat removed from the cold space): $\ \beta=\dfrac{Q_2}{W}=\dfrac{Q_2}{Q_1-Q_2}$;
  • heat pump (you want heat delivered to the warm space): $\ \beta_\text{hp}=\dfrac{Q_1}{W}=\dfrac{Q_1}{Q_1-Q_2}$.

Since $Q_1=Q_2+W$, for the same machine $\beta_\text{hp}=\beta+1$. Unlike an efficiency, a COP is not limited to 1: values of 3–6 are normal.

A COP bigger than 1 does not create energy. A heat pump with $\beta_\text{hp}=4$ delivers $4\ \text{kJ}$ of heat to the house for every $1\ \text{kJ}$ of electricity because the other $3\ \text{kJ}$ is moved from the cold outside air — energy that was there already. That is why heat pumps are four times cheaper to run than an electric heater, which delivers just $1\ \text{kJ}$ per kJ.

✏️ Example 1 — A car engine in numbers
In each cycle a petrol engine takes in $2.5\ \text{kJ}$ of heat from the burning fuel and rejects $1.8\ \text{kJ}$ to the surroundings. It completes $40$ cycles per second. (a) Find the work done per cycle and the efficiency. (b) Find the mechanical power output and the rate at which heat is rejected. (c) The fuel releases $4.5\times10^7\ \text{J}$ of heat per kilogram. How many kilograms of fuel does the engine burn per hour?
(a) $W=Q_1-Q_2=2.5-1.8=0.70\ \text{kJ}$; $\ \eta=\dfrac{W}{Q_1}=\dfrac{0.70}{2.5}=0.28=28\%$.
(b) Power $P=40\times700=2.8\times10^4\ \text{W}=28\ \text{kW}$. Heat rejected per second $40\times1800=7.2\times10^4\ \text{W}=72\ \text{kW}$.
(c) Heat taken in per second $40\times2500=1.0\times10^5\ \text{W}$; per hour $1.0\times10^5\times3600=3.6\times10^8\ \text{J}$. Fuel $=\dfrac{3.6\times10^8}{4.5\times10^7}=8.0\ \text{kg}$ per hour.
Check Power in $=100\ \text{kW}$ $=28\ \text{kW}$ (work) $+\ 72\ \text{kW}$ (heat out) ✓ — energy is conserved; the second law only decides how it splits.
✏️ Example 2 — A refrigerator and the same machine as a heat pump
A refrigerator's compressor draws $120\ \text{W}$ of electrical power and the refrigerator has a coefficient of performance of $4.0$. (a) At what rate is heat removed from the inside? (b) At what rate is heat delivered to the kitchen? (c) How long does it take to freeze $1.0\ \text{kg}$ of water already at $0^\circ\text{C}$ (latent heat of fusion $3.34\times10^5\ \text{J/kg}$)? (d) If the same machine, working between the same two temperatures, were used as a heat pump, what would its COP be?
Find the freezing time in (c) before you open the solution:
(a) $\beta=\dfrac{Q_2}{W}$, so per second $Q_2=\beta W=4.0\times120=480\ \text{W}$.
(b) $Q_1=Q_2+W=480+120=600\ \text{W}$ — the kitchen gains more heat than the cabinet loses.
(c) Heat to remove $mL=1.0\times3.34\times10^5=3.34\times10^5\ \text{J}$; $\ t=\dfrac{3.34\times10^5}{480}=696\ \text{s}\approx11.6\ \text{min}$.
(d) $\beta_\text{hp}=\dfrac{Q_1}{W}=\dfrac{600}{120}=5.0=\beta+1$.
⏱ Shortcut Write the three rates $Q_2:W:Q_1=\beta:1:(\beta+1)=4:1:5$ and scale by $120\ \text{W}$: $480$, $120$, $600\ \text{W}$.

An engine does $400\ \text{J}$ of work in each cycle and rejects $600\ \text{J}$ of heat to its sink. Its efficiency is

The second law of thermodynamics

Every engine ever built rejects some heat to a colder body, and in nature heat always flows by itself from hot to cold. These two observations, each generalised into a law, are the two classic statements of the second law. Notice what each statement forbids: not a process that merely includes some conversion or transfer, but one whose sole result is that conversion or transfer, with nothing else changed anywhere.

📐 Law 8.11 — Second law of thermodynamics

Kelvin–Planck statement. No process is possible whose sole result is the absorption of heat from a single reservoir and its complete conversion into work. Equivalently: no engine working in a cycle can have efficiency $\eta=1$; every heat engine must reject some heat to a colder reservoir.

Clausius statement. No process is possible whose sole result is the transfer of heat from a colder body to a hotter body. Equivalently: no refrigerator can work without an input of work; $\beta$ is always finite.

The words "sole result" matter. An ideal gas expanding isothermally does turn all the heat it absorbs into work, but that is not its sole result — the gas has ended up at a larger volume and lower pressure. A refrigerator does move heat from cold to hot, but not as its sole result — work has been supplied, and that energy has been dumped into the hot reservoir too. The second law says nothing against either.

📐 Result 8.12 — The two statements are equivalent

A device that violated the Kelvin–Planck statement could be used to violate the Clausius statement, and vice versa. So the two statements are two faces of one law.

Why (proof by combining devices). (i) Suppose a "perfect engine" existed, taking $Q$ from a hot reservoir and turning all of it into work $W=Q$. Use that work to drive an ordinary refrigerator, which takes $Q_2$ from a cold reservoir and delivers $Q_2+W$ to the hot one. The combined device needs no outside work, and its net effect is that the hot reservoir gains $(Q_2+W)-Q=Q_2$ while the cold one loses $Q_2$: heat has gone from cold to hot as the sole result — a violation of Clausius. (ii) Suppose instead a "perfect refrigerator" existed that moved heat from cold to hot with no work. Let an ordinary engine take $Q_1$, do $W=Q_1-Q_2$ and reject $Q_2$; let the perfect refrigerator carry that $Q_2$ straight back to the hot reservoir. The cold reservoir is unchanged, and the combination takes net heat $Q_1-Q_2$ from the hot reservoir alone and turns all of it into work — a violation of Kelvin–Planck.

The first law says you cannot win (get out more energy than you put in); the second law says you cannot even break even (turn heat entirely into work in a cycle). Machines that would break these laws are called perpetual-motion machines of the first and of the second kind. A ship that took heat from the sea and ran its propellers on it would not violate energy conservation at all — the oceans hold an enormous amount of internal energy — but it would be a perpetual-motion machine of the second kind, and it cannot be built.

✏️ Example 3 — Four inventions on trial
Each device below works in a cycle. Decide whether it obeys the first law, the second law, both, or neither. (a) A ship engine that takes heat from the sea water around it (a single reservoir at $290\ \text{K}$) and turns all of it into work at the propeller. (b) A box that cools a drink by moving heat out into the warmer room, with no battery or power supply. (c) An engine that takes $1200\ \text{J}$ from a hot source each cycle, does $500\ \text{J}$ of work and rejects $800\ \text{J}$ to a cold sink. (d) A refrigerator that removes $400\ \text{J}$ from its cabinet each cycle, uses $100\ \text{J}$ of electrical work and delivers $500\ \text{J}$ to the room.
(a) Energy balance is fine ($W=Q$), but it converts heat from a single reservoir completely into work in a cycle: it violates the Kelvin–Planck statement.
(b) Energy balance is fine (the drink loses what the room gains), but heat goes from cold to hot as the sole result: it violates the Clausius statement.
(c) Over a cycle $W$ must equal $Q_1-Q_2=1200-800=400\ \text{J}$, not $500\ \text{J}$: it creates $100\ \text{J}$ each cycle and violates the first law. (Whether it obeys the second law is then beside the point.)
(d) $Q_1=Q_2+W=400+100=500\ \text{J}$ ✓ (first law), and work is supplied, so Clausius is not violated. It is a legal refrigerator with $\beta=4.0$ — whether a COP of $4.0$ is achievable depends on the two temperatures, which is the question Carnot answers below.

In Interactive 1 choose refrigerator, set the hot side to $T_1=300\ \text{K}$, the cold side to $T_2=250\ \text{K}$ and the performance to $100\%$ of the Carnot limit: the COP reads $5.00$. Predict the COP when you lower the cold side to $T_2=200\ \text{K}$.

With $T_1=300\ \text{K}$ the readouts show a Carnot limit of $250/50=5.00$ at $T_2=250\ \text{K}$ and $200/100=2.00$ at $T_2=200\ \text{K}$; with $W=100\ \text{J}$ per cycle the heat drawn from the cold side drops from $500\ \text{J}$ to $200\ \text{J}$ (and $Q_1$ from $600\ \text{J}$ to $300\ \text{J}$). The best COP is $T_2/(T_1-T_2)$: pumping heat up a bigger temperature step costs more work per joule, which is why a deep freezer uses more electricity than a fridge. Now push the performance slider above $100\%$ and watch the verdict.
Interactive 1 — Energy-Flow Diagram: Engine, Refrigerator or Heat Pump

Choose a device and the two reservoir temperatures. The arrow widths show the energies per cycle (the engine takes $Q_1=1000\ \text{J}$; the refrigerator and heat pump use $W=100\ \text{J}$). The performance slider sets the device's efficiency or COP as a percentage of the Carnot limit — the best any device between these temperatures can do. Press Play to see the energy flow; go above $100\%$ and the second law objects.

500 K300 K60%
Carnot limit—
this device—
work $W$ per cycle—
$Q_1$ / $Q_2$ per cycle—

Which of the following would violate the Kelvin–Planck statement of the second law?

Reversible and irreversible processes

Film a ball bouncing on a hard floor and play it backwards: it looks wrong, because the real ball bounces lower each time, and in the reversed film it would bounce higher and higher, sucking heat out of the floor. Film a drop of ink spreading through water and play it backwards: the ink gathers itself into a drop again — something nobody has ever seen. Almost every real process has a direction; it cannot simply be "run backwards". Thermodynamics makes this precise.

📘 Definition 8.12 — Reversible and irreversible processes

A process that takes a system from state $i$ to state $f$ is reversible if it can be run in the reverse direction so that the system and its surroundings both return exactly to their initial states, with no change left anywhere in the universe. A process for which this is impossible is irreversible.

It is not enough that the system can be brought back — any system can be brought back by some route. A gas that has expanded can always be pushed back to its old volume. The question is whether the surroundings can be restored too. If restoring the system leaves some work converted into heat in the surroundings (or some heat moved from hot to cold), the process was irreversible.

📐 Result 8.13 — Conditions for a reversible process

A process is reversible only if both of the following hold:

  1. it is quasi-static: carried out infinitely slowly, so that the system passes through a continuous sequence of equilibrium states (uniform pressure and temperature throughout at every stage), with heat exchanged only across an infinitesimal temperature difference;
  2. there are no dissipative effects: no friction, viscosity, electrical resistance or inelastic deformation that turns work into internal energy.

Every real process breaks these conditions to some extent, so every real process is irreversible. Reversible processes are idealisations — limits that real processes can approach — but they set the boundaries on what is possible.

The usual sources of irreversibility are worth learning as a list, because JEE Main asks you to spot them:

By contrast, the following can be idealised as reversible: the slow, frictionless isothermal expansion or compression of a gas in contact with a reservoir at the gas's own temperature; the slow, frictionless adiabatic expansion or compression of a gas in an insulated cylinder; a slowly stretched spring or a slowly charged capacitor (with no resistance). On a $p$–$V$ diagram only a quasi-static process has a well-defined path, because only then does the gas have a single pressure at each instant; an irreversible process such as a free expansion cannot be drawn as a curve at all.

✏️ Example 4 — Free expansion: nothing seems to happen, yet it cannot be undone
Two moles of an ideal gas at $300\ \text{K}$ occupy one half ($5.0\ \text{L}$) of a rigid, thermally insulated vessel; the other half is a vacuum. The partition is broken and the gas fills the whole $10.0\ \text{L}$. (a) Find $Q$, $W$, $\Delta U$ and the final temperature. (b) The gas is now to be returned to its initial state by a reversible isothermal compression at $300\ \text{K}$. Find the work that must be done on it and the heat it gives out. (c) Explain why the free expansion is irreversible although $Q$, $W$ and $\Delta U$ for it are all zero.
(a) Insulated: $Q=0$. The gas pushes against nothing (a vacuum) and the walls are rigid: $W=0$. First law: $\Delta U=Q-W=0$. For an ideal gas $U$ depends only on $T$, so $T_f=300\ \text{K}$.
(b) $W_\text{on}=nRT\ln\dfrac{V_f}{V_i}=2.0\times8.314\times300\times\ln2=3.46\times10^3\ \text{J}$. At constant $T$, $\Delta U=0$, so the gas gives out heat $3.46\times10^3\ \text{J}$ to the reservoir.
(c) The gas is back where it started, but the surroundings are not: they have done $3.46\ \text{kJ}$ of work and received the same amount as heat. To finish the job they would have to turn that heat back into work completely, which Kelvin–Planck forbids. No route can restore both system and surroundings, so the free expansion is irreversible.
Note During the free expansion the gas is not in equilibrium (it rushes and swirls), so it has no single pressure and its path cannot be drawn on a $p$–$V$ diagram; only its end points can.

Which of the following is not a requirement for a process to be reversible?

The Carnot engine: the reversible ideal

If every real engine wastes some heat, how much must it waste? In 1824 the French engineer Sadi Carnot answered this by imagining an engine with no irreversibility at all. Between two reservoirs the only heat transfers that can be reversible are those at the reservoir temperature itself (an infinitesimal $\Delta T$), and the only reversible way of changing the temperature of the working substance between $T_1$ and $T_2$, without touching a reservoir, is an adiabatic process. So the reversible engine is built from two isotherms and two adiabats.

📘 Definition 8.13 — Carnot cycle and Carnot engine

A Carnot cycle is a reversible cycle of four quasi-static, frictionless steps, carried out by a working substance (here $n$ moles of an ideal gas) between a source at $T_1$ and a sink at $T_2$:

  1. 1 → 2, isothermal expansion at $T_1$ in contact with the source, from $V_1$ to $V_2$: the gas absorbs heat $Q_1$;
  2. 2 → 3, adiabatic expansion from $V_2$ to $V_3$, insulated, until its temperature falls to $T_2$;
  3. 3 → 4, isothermal compression at $T_2$ in contact with the sink, from $V_3$ to $V_4$: the gas rejects heat $Q_2$;
  4. 4 → 1, adiabatic compression from $V_4$ back to $V_1$, insulated, until its temperature rises to $T_1$.

An engine working on this cycle is a Carnot engine. Run backwards (1 → 4 → 3 → 2 → 1) it is a Carnot refrigerator or heat pump.

Finding the efficiency. For an ideal gas $U$ depends only on $T$, so on each isotherm $\Delta U=0$ and the heat equals the work, $nRT\ln(V_\text{final}/V_\text{initial})$ (from 8.2):

$$Q_1=nRT_1\ln\frac{V_2}{V_1},\qquad Q_2=nRT_2\ln\frac{V_3}{V_4}.$$

On the adiabats, $TV^{\gamma-1}$ is constant: $T_1V_2^{\gamma-1}=T_2V_3^{\gamma-1}$ for step 2 → 3 and $T_1V_1^{\gamma-1}=T_2V_4^{\gamma-1}$ for step 4 → 1. Dividing one by the other,

$$\left(\frac{V_2}{V_1}\right)^{\gamma-1}=\left(\frac{V_3}{V_4}\right)^{\gamma-1}\quad\Rightarrow\quad\frac{V_2}{V_1}=\frac{V_3}{V_4}.$$

The logarithms in $Q_1$ and $Q_2$ are therefore equal, and they cancel in the ratio: $\dfrac{Q_2}{Q_1}=\dfrac{T_2}{T_1}$. The two adiabatic steps do equal and opposite work, $nC_V(T_1-T_2)$ each, so they add nothing to the net work; $W=Q_1-Q_2$, the area enclosed by the cycle on the $p$–$V$ diagram (Interactive 2).

📐 Result 8.14 — Efficiency of a Carnot engine

For a Carnot engine working between a source at absolute temperature $T_1$ and a sink at $T_2$,

$$\frac{Q_2}{Q_1}=\frac{T_2}{T_1},\qquad\eta_\text{C}=1-\frac{Q_2}{Q_1}=1-\frac{T_2}{T_1}.$$

The temperatures must be in kelvin. $\eta_\text{C}$ depends only on the two reservoir temperatures — not on the working substance, the amount of it, or the volumes used. It reaches $1$ only if $T_2=0\ \text{K}$, which the Kelvin–Planck statement rules out: absolute zero cannot be used as a sink.

📐 Result 8.15 — Carnot's theorem

(a) No engine working between two given reservoirs can be more efficient than a reversible (Carnot) engine working between the same two reservoirs: $\eta\le1-T_2/T_1$. (b) All reversible engines working between the same two reservoirs have the same efficiency, whatever their working substance.

Why (proof). Suppose some engine X were more efficient than a Carnot engine R between the same reservoirs. Let X take $Q_1'$ from the source and deliver work $W$. Because R is reversible, we can run it backwards as a refrigerator, driven by exactly that work $W$; it then delivers $Q_1=W/\eta_\text{R}$ to the source. Since $\eta_\text{X}\gt\eta_\text{R}$, $\ W/\eta_\text{R}\gt W/\eta_\text{X}$, i.e. $Q_1\gt Q_1'$. The pair needs no outside work, yet the source gains $Q_1-Q_1'\gt0$ each cycle and, by the first law, the sink loses the same amount: heat goes from cold to hot as the sole result, against Clausius. So $\eta_\text{X}\le\eta_\text{R}$. If X is reversible too, the same argument with the roles swapped gives $\eta_\text{R}\le\eta_\text{X}$, so the two are equal — part (b).

Because the ratio $Q_2/Q_1$ of a reversible engine is the same for every working substance, it can be used to define temperature independently of any thermometric material: $T_2/T_1=Q_2/Q_1$. This thermodynamic (absolute) temperature scale, fixed by giving the triple point of water $273.16\ \text{K}$, is exactly the kelvin scale of the ideal-gas thermometer.

Running Carnot backwards. A Carnot refrigerator still has $Q_2/Q_1=T_2/T_1$, and the same reasoning shows that no refrigerator between two reservoirs can beat it.

📐 Result 8.16 — Best possible refrigerator and heat pump

For a Carnot (reversible) refrigerator or heat pump working between $T_1$ (hot) and $T_2$ (cold), and the upper limit for any real one between the same reservoirs:

$$\beta_\text{C}=\frac{Q_2}{W}=\frac{T_2}{T_1-T_2},\qquad\beta_\text{hp,C}=\frac{Q_1}{W}=\frac{T_1}{T_1-T_2}.$$

For a Carnot engine and a Carnot refrigerator between the same reservoirs, $\beta_\text{C}=\dfrac{1-\eta_\text{C}}{\eta_\text{C}}$. The closer the two temperatures, the larger the COP — and the smaller the efficiency of an engine.

🧭 Problem-solving strategy — Engines, refrigerators and the Carnot limit
  1. Convert every temperature to kelvin: $T=t\,(^\circ\text{C})+273$.
  2. Draw the energy-flow diagram and write the first law for the cycle: $Q_1=W+Q_2$ (all magnitudes).
  3. Use the definition that matches the device: $\eta=W/Q_1$ (engine), $\beta=Q_2/W$ (refrigerator), $\beta_\text{hp}=Q_1/W$ (heat pump).
  4. If the device is reversible (Carnot), add $Q_2/Q_1=T_2/T_1$. If it is real, the Carnot values are upper limits: any claim above them violates the second law.
  5. For two Carnot engines in series (the sink of the first is the source of the second, at $T$): equal work gives $T=\tfrac12(T_1+T_2)$; equal efficiency gives $T=\sqrt{T_1T_2}$.
✏️ Example 5 — One Carnot cycle, step by step
One mole of a monatomic ideal gas ($\gamma=5/3$) is taken round a Carnot cycle between a source at $600\ \text{K}$ and a sink at $300\ \text{K}$. During the isothermal expansion its volume doubles. Find (a) the heat $Q_1$ absorbed; (b) the ratio $V_3/V_2$ in the adiabatic expansion; (c) the heat $Q_2$ rejected; (d) the net work per cycle and the efficiency; (e) the work done in each adiabatic step. ($R=8.314\ \text{J mol}^{-1}\text{K}^{-1}$)
Find the net work per cycle in (d) before you open the solution:
(a) $Q_1=nRT_1\ln\dfrac{V_2}{V_1}=1\times8.314\times600\times\ln2=3458\ \text{J}=3.46\ \text{kJ}$.
(b) $T_1V_2^{\gamma-1}=T_2V_3^{\gamma-1}$ with $\gamma-1=\tfrac23$: $\dfrac{V_3}{V_2}=\left(\dfrac{T_1}{T_2}\right)^{3/2}=2^{3/2}=2.83$.
(c) $V_3/V_4=V_2/V_1=2$, so $Q_2=nRT_2\ln2=8.314\times300\times0.693=1729\ \text{J}$.
(d) $W=Q_1-Q_2=3458-1729=1729\ \text{J}$; $\ \eta=\dfrac{1729}{3458}=0.50=50\%$.
(e) $W_{2\to3}=nC_V(T_1-T_2)=1\times\tfrac32\times8.314\times300=3741\ \text{J}$ done by the gas; $W_{4\to1}=-3741\ \text{J}$ (done on it). They cancel.
Check $1-T_2/T_1=1-300/600=50\%$ ✓ — the volumes and $\gamma$ were needed for $Q_1$ and $V_3$, but not for $\eta$.
✏️ Example 6 — Hotter source or colder sink?
A Carnot engine has an efficiency of $40\%$ with its sink at $27^\circ\text{C}$. (a) Find the temperature of the source. (b) Keeping the sink fixed, by how much must the source temperature be raised to make the efficiency $50\%$? (c) Keeping the source at its original temperature, by how much must the sink be cooled instead? (d) Show that, in general, lowering the sink by a small $\Delta T$ raises the efficiency more than raising the source by the same $\Delta T$.
Find the new source temperature in (b) first:
(a) $T_2=300\ \text{K}$; $\ 1-\dfrac{300}{T_1}=0.40\Rightarrow T_1=\dfrac{300}{0.60}=500\ \text{K}$ ($227^\circ\text{C}$).
(b) $1-\dfrac{300}{T_1'}=0.50\Rightarrow T_1'=600\ \text{K}$: raise the source by $100\ \text{K}$.
(c) $1-\dfrac{T_2'}{500}=0.50\Rightarrow T_2'=250\ \text{K}$: cool the sink by only $50\ \text{K}$.
(d) Lowering the sink: $\Delta\eta=\dfrac{\Delta T}{T_1}$. Raising the source: $\Delta\eta=\dfrac{T_2}{T_1}-\dfrac{T_2}{T_1+\Delta T}=\dfrac{T_2\,\Delta T}{T_1(T_1+\Delta T)}$, smaller by the factor $\dfrac{T_2}{T_1+\Delta T}\lt1$. So a colder sink always wins kelvin for kelvin — but in practice the sink is the atmosphere or a river, which cannot be cooled cheaply, so engineers raise $T_1$ instead.
⏱ Shortcut $T_2/T_1=1-\eta$: from $0.6$ to $0.5$. Fix $T_2=300$: $T_1=300/0.5=600$. Fix $T_1=500$: $T_2=0.5\times500=250$.

In Interactive 2 choose the monatomic gas, $T_1=600\ \text{K}$, $T_2=300\ \text{K}$ and expansion ratio $V_2/V_1=2.0$: the readouts show $Q_1=3458\ \text{J}$, $W=1729\ \text{J}$ and $\eta=50.0\%$. What happens when you increase the expansion ratio to $4.0$?

At $V_2/V_1=4.0$ the readouts give $Q_1=nRT_1\ln4=6915\ \text{J}$, $Q_2=3458\ \text{J}$ and $W=3458\ \text{J}$ — the cycle on the $p$–$V$ diagram grows and so does its area, because $\ln4=2\ln2$. But $Q_2/Q_1$ is still $T_2/T_1$, so $\eta$ stays at $50.0\%$. Switch to the diatomic gas: the adiabats get less steep and the cycle changes shape, but $\eta$ still reads $50.0\%$. Only the temperature sliders move $\eta$.
Interactive 2 — The Carnot Cycle on a $p$–$V$ Diagram

One mole of ideal gas, starting at $V_1=10\ \text{L}$, runs a Carnot cycle: isothermal expansion at $T_1$ (red), adiabatic expansion (purple), isothermal compression at $T_2$ (blue), adiabatic compression. Press Play to follow the state point round the cycle; the shaded area is the net work $W$. Change the temperatures, the expansion ratio $V_2/V_1$ and the gas, and watch which readouts change.

600 K300 K2.0
$Q_1$ taken in at $T_1$—
$Q_2$ rejected at $T_2$—
net work $W$—
efficiency $\eta$—
⚠️ JEE Main traps

1. Celsius in a Carnot formula. $\eta=1-T_2/T_1$ and $\beta=T_2/(T_1-T_2)$ need kelvin. An engine between $100^\circ\text{C}$ and $0^\circ\text{C}$ has $\eta=1-273/373=27\%$, not $100\%$.

2. Refrigerator or heat pump? $\beta=Q_2/W$ (heat removed from the cold side) but $\beta_\text{hp}=Q_1/W$ (heat delivered to the warm side), and $\beta_\text{hp}=\beta+1$. A COP greater than 1 is normal and breaks no law.

3. The Carnot value is a ceiling, not a formula for every engine. For a real engine use $\eta=W/Q_1=1-Q_2/Q_1$; $1-T_2/T_1$ is its upper limit. A claimed efficiency above the Carnot limit violates the second law, even when the energy balance (first law) is perfect.

4. Same difference, different efficiency. $\eta$ depends on the ratio $T_2/T_1$, not on $T_1-T_2$: raising both temperatures by $100\ \text{K}$ lowers the efficiency. Lowering $T_2$ by $\Delta T$ helps more than raising $T_1$ by $\Delta T$.

5. "Slow" is not the same as "reversible". A slow process with friction is still irreversible. And a free expansion has $Q=W=\Delta U=0$ (so $T$ is unchanged for an ideal gas) yet is irreversible — zero heat and work do not mean "nothing happened".

A Carnot engine works between $400\ \text{K}$ and $300\ \text{K}$. If both reservoir temperatures are raised by $100\ \text{K}$, its efficiency becomes

Exercises

Take $R=8.314\ \text{J mol}^{-1}\text{K}^{-1}$ and $T\,(\text{K})=t\,(^\circ\text{C})+273$. "Carnot" means a reversible engine, refrigerator or heat pump working between two reservoirs. Heats $Q_1$, $Q_2$ and work $W$ are magnitudes per cycle unless stated otherwise.

I
Heat engines, refrigerators and heat pumps.
  • 1 An engine absorbs $5.0\ \text{kJ}$ of heat and rejects $3.5\ \text{kJ}$ in each cycle. Find the work per cycle and the efficiency (in %).
  • 2 An engine of efficiency $25\%$ delivers $2.0\ \text{kW}$ of mechanical power. Find the rate at which it takes in heat (in kW) and the rate at which it rejects heat.
  • 3 A power station delivers $600\ \text{MW}$ of electrical power with an overall efficiency of $38\%$. At what rate (in MW) must heat be removed by its cooling water?
  • 4 A refrigerator removes $300\ \text{J}$ from its cabinet in each cycle while $75\ \text{J}$ of work is done on it. Find its COP and the heat delivered to the room per cycle.
  • 5 A heat pump with $\beta_\text{hp}=3.5$ delivers $7.0\ \text{kW}$ of heat to a house. Find the electrical power it draws (in kW) and the rate at which it takes heat from the outside air.
  • 6 Prove that for the same machine working between the same reservoirs, $\beta_\text{hp}=\beta+1$.
  • 7 A car engine of efficiency $30\%$ burns $1.0\ \text{kg}$ of fuel that releases $4.5\times10^7\ \text{J/kg}$. Find the useful work obtained and the heat wasted.
  • 8 Even with no friction at all, a heat engine cannot have an efficiency of $100\%$. Which law says so, and what part of the engine makes it impossible?
1 $W=5.0-3.5=1.5\ \text{kJ}$; $\ \eta=1.5/5.0=0.30=30\%$.
2 Heat in $=P/\eta=2.0/0.25=8.0\ \text{kW}$; heat out $=8.0-2.0=6.0\ \text{kW}$.
3 Heat in $=600/0.38=1579\ \text{MW}$; rejected $=1579-600=979\ \text{MW}$ — more than one and a half times the electrical output.
4 $\beta=Q_2/W=300/75=4.0$; $\ Q_1=Q_2+W=375\ \text{J}$.
5 $W=Q_1/\beta_\text{hp}=7.0/3.5=2.0\ \text{kW}$; from outside $Q_2=7.0-2.0=5.0\ \text{kW}$.
6 $\beta_\text{hp}=\dfrac{Q_1}{W}=\dfrac{Q_2+W}{W}=\dfrac{Q_2}{W}+1=\beta+1$.
7 Heat in $4.5\times10^7\ \text{J}$; $W=0.30\times4.5\times10^7=1.35\times10^7\ \text{J}$; wasted $4.5\times10^7-1.35\times10^7=3.15\times10^7\ \text{J}$.
8 The Kelvin–Planck statement of the second law: a cyclic engine must reject some heat $Q_2\gt0$ to a sink, so $W=Q_1-Q_2\lt Q_1$. Without a sink at a lower temperature the engine cannot complete its cycle and run again.
II
The second law; reversible and irreversible processes.
  • 1 State the Kelvin–Planck and Clausius statements of the second law, and for each name a device it forbids.
  • 2 Each device works in a cycle. Which law, if any, does it break? (a) It absorbs $800\ \text{J}$ from a single reservoir and does $800\ \text{J}$ of work. (b) A refrigerator keeps its cabinet cold with no power supply. (c) An engine takes $1000\ \text{J}$, does $300\ \text{J}$ of work and rejects $600\ \text{J}$.
  • 3 An ice cube melts on a table in a warm room. Is the process reversible? Give the reason.
  • 4 One mole of an ideal gas at $300\ \text{K}$ expands freely from $10\ \text{L}$ to $30\ \text{L}$ in a rigid insulated vessel. Find $Q$, $W$, $\Delta U$ and the final temperature (in K).
  • 5 The same gas is taken between the same two states by a reversible isothermal expansion at $300\ \text{K}$. Find the work done by the gas (in J). Why is this different from part 4?
  • 6 A gas in a cylinder is compressed extremely slowly by a piston that rubs against the cylinder wall. Is the process reversible?
  • 7 A $2.0\ \text{kg}$ block sliding at $3.0\ \text{m/s}$ on a rough floor comes to rest. How much mechanical energy (in J) becomes internal energy? Can the block and floor ever cool down and set the block moving again by themselves? Which law answers this?
  • 8 Explain why a gas suddenly compressed by a fast-moving piston undergoes an irreversible process, even in an insulated, frictionless cylinder.
1 Kelvin–Planck: no process can have as its sole result the absorption of heat from one reservoir and its complete conversion into work — forbids a cyclic engine with $\eta=1$ (e.g. a ship run on heat from the sea). Clausius: no process can have as its sole result the transfer of heat from a colder to a hotter body — forbids a refrigerator that needs no work.
2 (a) Kelvin–Planck (first law obeyed). (b) Clausius. (c) First law: $Q_1-Q_2=400\ \text{J}\ne300\ \text{J}$ — it destroys $100\ \text{J}$ per cycle.
3 Irreversible: heat flows from the warm room to the ice across a finite temperature difference. The water will not refreeze by giving its heat back to the warmer room.
4 $Q=0$ (insulated), $W=0$ (expands into vacuum, rigid walls), $\Delta U=0$; $T_f=300\ \text{K}$.
5 $W=nRT\ln3=8.314\times300\times1.0986=2740\ \text{J}$ (and $Q=2740\ \text{J}$). The end states are the same but $W$ and $Q$ depend on the path; only $\Delta U$ ($=0$ in both) is fixed by the end states.
6 No. It is quasi-static, but friction turns some work into internal energy, which cannot be fully recovered when the process is reversed.
7 $\tfrac12mv^2=\tfrac12\times2.0\times3.0^2=9.0\ \text{J}$. No: that would turn heat from the surroundings completely into work as the sole result — forbidden by the second law (Kelvin–Planck), though it would obey the first law.
8 A fast piston compresses the gas next to it first, setting up pressure differences, waves and turbulence; the gas is not in equilibrium, so the process is not quasi-static. Pulling the piston back quickly does not retrace the same states, and the extra work done in the turbulence ends up as internal energy.
III
The Carnot engine, refrigerator and heat pump.
  • 1 Find the efficiency (in %) of a Carnot engine working between $527^\circ\text{C}$ and $127^\circ\text{C}$.
  • 2 A Carnot engine takes $2000\ \text{J}$ per cycle from a source at $500\ \text{K}$ and has its sink at $300\ \text{K}$. Find the work per cycle and the heat rejected.
  • 3 A Carnot engine rejects $900\ \text{J}$ per cycle to a sink at $27^\circ\text{C}$ while doing $600\ \text{J}$ of work. Find the source temperature (in K).
  • 4 A Carnot engine of efficiency $30\%$ has its sink at $280\ \text{K}$. Find the source temperature.
  • 5 A Carnot refrigerator works between $-13^\circ\text{C}$ and $27^\circ\text{C}$. Find its COP and the work needed to remove $1300\ \text{J}$ from the cold space.
  • 6 A Carnot heat pump keeps a house at $20^\circ\text{C}$ when it is $0^\circ\text{C}$ outside. Find its COP and the electrical power needed to supply $10\ \text{kW}$ of heat.
  • 7 Two moles of an ideal gas run a Carnot cycle between $500\ \text{K}$ and $300\ \text{K}$; in the isothermal expansion the volume triples. Find $Q_1$ (in kJ) and the net work per cycle.
  • 8 A monatomic ideal gas runs a Carnot cycle between $800\ \text{K}$ and $200\ \text{K}$. Find the ratio $V_3/V_2$ of the volumes at the end and the start of the adiabatic expansion.
  • 9 Explain why the work done in the two adiabatic steps of a Carnot cycle of an ideal gas adds up to zero.
1 $T_1=800\ \text{K}$, $T_2=400\ \text{K}$: $\eta=1-\tfrac{400}{800}=50\%$.
2 $\eta=1-\tfrac{300}{500}=0.40$; $W=0.40\times2000=800\ \text{J}$; $Q_2=1200\ \text{J}$.
3 $Q_1=W+Q_2=1500\ \text{J}$; $T_1=T_2\dfrac{Q_1}{Q_2}=300\times\dfrac{1500}{900}=500\ \text{K}$.
4 $\dfrac{T_2}{T_1}=0.70$: $T_1=\dfrac{280}{0.70}=400\ \text{K}$.
5 $T_2=260\ \text{K}$, $T_1=300\ \text{K}$: $\beta=\dfrac{260}{40}=6.5$; $W=\dfrac{1300}{6.5}=200\ \text{J}$.
6 $\beta_\text{hp}=\dfrac{293}{20}=14.65$ (about $15$); $P=\dfrac{10}{14.65}=0.683\ \text{kW}$. Real heat pumps reach perhaps a third of this.
7 $Q_1=nRT_1\ln3=2\times8.314\times500\times1.0986=9.13\times10^3\ \text{J}=9.13\ \text{kJ}$; $W=\left(1-\tfrac{300}{500}\right)Q_1=3.65\ \text{kJ}$.
8 $T_1V_2^{2/3}=T_2V_3^{2/3}$: $\dfrac{V_3}{V_2}=\left(\dfrac{800}{200}\right)^{3/2}=4^{3/2}=8$.
9 In an adiabatic step $Q=0$, so $W=-\Delta U=-nC_V\Delta T$. Step 2 → 3 has $\Delta T=T_2-T_1$ and step 4 → 1 has $\Delta T=T_1-T_2$: the works are $+nC_V(T_1-T_2)$ and $-nC_V(T_1-T_2)$, which cancel.
IV
Harder problems and applications.
  • 1 Two Carnot engines work in series between $1000\ \text{K}$ and $250\ \text{K}$: the first rejects its heat to an intermediate reservoir at $T$, from which the second takes it. Find $T$ (in K) if both engines do the same work per cycle.
  • 2 For the same arrangement, find $T$ if instead both engines have the same efficiency.
  • 3 A Carnot engine has efficiency $1/5$. When the sink temperature is lowered by $80\ \text{K}$, the efficiency doubles. Find the source temperature (in K) and the original sink temperature.
  • 4 A Carnot engine working between $600\ \text{K}$ and $300\ \text{K}$ drives a Carnot refrigerator working between $300\ \text{K}$ and $250\ \text{K}$. For every $1000\ \text{J}$ the engine takes from the $600\ \text{K}$ source, how much heat (in J) does the refrigerator remove from the $250\ \text{K}$ space?
  • 5 An inventor claims an engine that works between $450\ \text{K}$ and $300\ \text{K}$ with an efficiency of $40\%$. Find the Carnot limit (in %) and judge the claim.
  • 6 A freezer compartment is at $-18^\circ\text{C}$ in a kitchen at $27^\circ\text{C}$; the freezer's actual COP is $40\%$ of the Carnot value. Find the work needed to freeze $2.0\ \text{kg}$ of water already at $0^\circ\text{C}$ (latent heat $3.34\times10^5\ \text{J/kg}$), treating all the heat as removed at $-18^\circ\text{C}$.
  • 7 Show that a Carnot engine of efficiency $\eta$ becomes, when reversed, a refrigerator with $\beta=(1-\eta)/\eta$. Find $\beta$ for $\eta=20\%$.
1 $W_1=W_2$: $Q_1-Q=Q-Q_2$, and $Q_1:Q:Q_2=T_1:T:T_2$, so $T=\tfrac12(T_1+T_2)=\tfrac12(1000+250)=625\ \text{K}$.
2 $1-\dfrac{T}{T_1}=1-\dfrac{T_2}{T}$: $T=\sqrt{T_1T_2}=\sqrt{1000\times250}=500\ \text{K}$.
3 $T_2=0.80T_1$ and $T_2-80=0.60T_1$. Subtracting: $0.20T_1=80$, $T_1=400\ \text{K}$, $T_2=320\ \text{K}$.
4 Engine: $W=\left(1-\tfrac{300}{600}\right)\times1000=500\ \text{J}$. Refrigerator: $\beta=\dfrac{250}{50}=5$, so it removes $5\times500=2500\ \text{J}$ from the $250\ \text{K}$ space.
5 $\eta_\text{C}=1-\tfrac{300}{450}=33.3\%$. The claim of $40\%$ exceeds the Carnot limit, so it violates the second law (Carnot's theorem) — reject it, even though it may balance energy perfectly.
6 $\beta_\text{C}=\dfrac{255}{300-255}=5.67$; real $\beta=0.40\times5.67=2.27$. Heat to remove $2.0\times3.34\times10^5=6.68\times10^5\ \text{J}$; $W=\dfrac{6.68\times10^5}{2.27}=2.95\times10^5\ \text{J}$.
7 Reversed, the cycle has the same $Q_1$, $Q_2$, $W$ with arrows turned round: $\beta=\dfrac{Q_2}{W}=\dfrac{Q_1-W}{W}=\dfrac1\eta-1=\dfrac{1-\eta}{\eta}$. For $\eta=0.20$: $\beta=0.80/0.20=4$.
🎯 Exam corner JEE Main · MCQJEE Main · Numerical

E1. JEE Main · MCQ A Carnot engine has an efficiency of $25\%$. When the temperature of its sink is lowered by $100\ \text{K}$, keeping the source unchanged, its efficiency becomes $50\%$. The source and sink temperatures originally were (A) $300\ \text{K}$ and $225\ \text{K}$ (B) $400\ \text{K}$ and $300\ \text{K}$ (C) $500\ \text{K}$ and $375\ \text{K}$ (D) $600\ \text{K}$ and $450\ \text{K}$

Set up $T_2=0.75T_1$ and $T_2-100=0.50T_1$.
Solve Subtract: $0.25T_1=100$, so $T_1=400\ \text{K}$ and $T_2=300\ \text{K}$.
Answer (B).
⏱ Shortcut The 100 K drop changed $T_2/T_1$ by $0.75-0.50=0.25$, so $100/T_1=0.25$. All four options have $T_2/T_1=0.75$ — only the size matters.

E2. JEE Main · Numerical A Carnot refrigerator keeps a freezer at $-23^\circ\text{C}$ in a room at $27^\circ\text{C}$. Heat leaks into the freezer at $600\ \text{W}$, and the refrigerator removes it as fast as it arrives. Find the power (in W) the refrigerator must draw. Give the answer to the nearest integer.

COP $T_2=250\ \text{K}$, $T_1=300\ \text{K}$: $\beta=\dfrac{250}{300-250}=5$.
Power $P=\dfrac{Q_2/t}{\beta}=\dfrac{600}{5}=120\ \text{W}$.
Answer $120$.
⏱ Shortcut Using the heat-pump COP $300/50=6$ gives $100$ — the classic trap; and using $^\circ\text{C}$ gives nonsense.

E3. JEE Main · MCQ Which of the following processes can be treated as reversible? (A) A gas expanding freely into an evacuated vessel (B) Heat flowing from a hot iron block into cold water (C) A block sliding to rest on a rough horizontal floor (D) An ideal gas compressed extremely slowly by a frictionless piston while in contact with a reservoir at the gas's own temperature

Check (A) is not quasi-static (free expansion); (B) is heat flow across a finite temperature difference; (C) involves friction. All three are irreversible.
Reason (D) is quasi-static, frictionless, and exchanges heat across an infinitesimal temperature difference — a reversible isothermal compression.
Answer (D).

E4. JEE Main · Numerical Two Carnot engines A and B are operated in series. A takes heat from a source at $900\ \text{K}$ and rejects heat to a reservoir at temperature $T$; B takes the heat rejected by A from that reservoir and rejects heat to a sink at $400\ \text{K}$. If A and B have equal efficiencies, find $T$ in kelvin. Give the answer to the nearest integer.

Equal η $1-\dfrac{T}{900}=1-\dfrac{400}{T}\Rightarrow T^2=900\times400=3.6\times10^5$.
Solve $T=600\ \text{K}$ (each efficiency is $1/3$).
Answer $600$.
⏱ Shortcut Equal efficiency → geometric mean $\sqrt{T_1T_2}$; equal work → arithmetic mean $\tfrac12(T_1+T_2)=650\ \text{K}$ (the trap answer here).

E5. JEE Main · MCQ An inventor claims an engine that works between $127^\circ\text{C}$ and $27^\circ\text{C}$, taking in $1000\ \text{J}$ of heat and doing $300\ \text{J}$ of work per cycle while rejecting $700\ \text{J}$. The claim (A) is valid, because energy is conserved (B) violates the first law of thermodynamics (C) violates the second law, because the claimed efficiency exceeds the maximum possible $25\%$ (D) is valid provided the engine is reversible

First law $1000=300+700$ ✓, so (B) is wrong.
Second law $T_1=400\ \text{K}$, $T_2=300\ \text{K}$: $\eta_\text{max}=1-\tfrac{300}{400}=25\%$. Claimed $\eta=300/1000=30\%\gt25\%$. Even a reversible engine reaches only $25\%$, so (D) is wrong too.
Answer (C).
⏱ Shortcut Check the energy balance first, then compare $W/Q_1$ with $1-T_2/T_1$ in kelvin.

E6. JEE Main · Numerical A house is kept at $27^\circ\text{C}$ by an ideal (Carnot) heat pump that takes heat from the outside air at $7^\circ\text{C}$. The house loses heat at $3.0\ \text{kW}$. Find the electrical power (in W) the heat pump needs. Give the answer to the nearest integer.

COP $T_1=300\ \text{K}$, $T_2=280\ \text{K}$: $\beta_\text{hp}=\dfrac{T_1}{T_1-T_2}=\dfrac{300}{20}=15$.
Power The pump must deliver $Q_1/t=3000\ \text{W}$: $P=\dfrac{3000}{15}=200\ \text{W}$.
Answer $200$.
⏱ Shortcut $P=\dfrac{\dot Q_1\,(T_1-T_2)}{T_1}=3000\times\dfrac{20}{300}$. The refrigerator COP $280/20=14$ would give $214$ — the wrong one for heating.
🔬 Research–Teaching NexusResearch-tutored · optional, about 45 minutes plus an overnight run

Research question. How close does a real household refrigerator come to the Carnot limit $\beta_\text{C}=T_2/(T_1-T_2)$ — and is a modern air-source heat pump a better or worse "Carnot machine" than the fridge?

Why it matters. Refrigeration and air conditioning use a large share of the world's electricity, and the share is rising fast in hot countries such as India, Brazil and those of the Gulf. Countries across Europe and East Asia are replacing gas boilers with heat pumps to cut carbon emissions. Energy-efficiency labels (star ratings, seasonal COP figures) are all, in the end, statements about how far a machine falls short of Result 8.16. Knowing that limit tells you whether a manufacturer's figure is plausible.

What to do.

  1. Fill a plastic bottle with $1.0\ \text{kg}$ of water at a measured room temperature and place it in the refrigerator's freezer. Record its temperature every 15 minutes with a thermometer probe (or check it at intervals) until it has frozen.
  2. Plug the refrigerator into a plug-in energy meter and record the electrical energy used over the same period. Repeat with the freezer empty to estimate the "background" energy needed to hold the cabinet cold, and subtract it.
  3. Calculate the heat removed from the water: cooling to $0^\circ\text{C}$ ($mc\Delta T$), freezing ($mL$) and any cooling of the ice below $0^\circ\text{C}$.
  4. Find the actual COP, $\beta=Q_2/W$, and compare it with $\beta_\text{C}$, using the freezer temperature for $T_2$ and the temperature at the condenser grid at the back for $T_1$ (both in kelvin).
  5. Look up the rated COP of an air-source heat pump at an outdoor temperature of $7^\circ\text{C}$ and an indoor delivery temperature of $35^\circ\text{C}$, and find its fraction of the Carnot value in the same way.

Tools: A refrigerator with a freezer, a plug-in energy meter, a thermometer or temperature probe, a kitchen scale, a plastic bottle, a timer; manufacturers' data sheets.

Hand in: A table of the measured heats, energy used, actual and Carnot COPs; the fraction $\beta/\beta_\text{C}$ for the fridge and the heat pump; and a paragraph naming the irreversibilities (heat flow across finite temperature differences in the coils, friction in the compressor, throttling of the refrigerant) that cause the shortfall.

Going further. Repeat the measurement with the refrigerator thermostat set to two different temperatures. Does the actual COP change in the direction Result 8.16 predicts when $T_2$ is lowered?

Instructor note: typical results are an actual COP of about 1.5–3 for a domestic freezer, against a Carnot value of roughly 4–6 between about $-18^\circ\text{C}$ and a condenser at $35$–$45^\circ\text{C}$ — so 30–50% of Carnot, similar to good heat pumps. The background correction matters: insist on it, and on kelvin temperatures. The freezing step dominates the heat removed ($3.34\times10^5\ \text{J}$ against about $10^5\ \text{J}$ for cooling the water from room temperature).

🧪 Self-check quiz

0/5 correct

Q1. The Kelvin–Planck statement of the second law of thermodynamics says that

Q2. The maximum possible efficiency of an engine working between $100^\circ\text{C}$ and $0^\circ\text{C}$ is about

Q3. Which of the following processes is irreversible?

Q4. The coefficient of performance of a Carnot refrigerator working between absolute temperatures $T_1$ (hot) and $T_2$ (cold) is

Q5. A Carnot engine has an efficiency of $25\%$. If the absolute temperature of its sink is halved while the source is unchanged, its efficiency becomes