THERMODYNAMICS • SECOND LAW AND ENTROPY

Entropy Definition & Changes — Define entropy and interpret entropy changes

Discover how entropy quantifies irreversibility and governs the direction of every natural process.

Historical Context & Motivation

The concept of entropy arose from a deceptively practical question: why can't a heat engine convert all of its absorbed heat into useful work? By the mid-nineteenth century, engineers and physicists recognized that something fundamental limited the efficiency of steam engines—something beyond mere friction or mechanical imperfection. The search for this missing quantity led to one of the most profound ideas in all of science, a concept that not only reshaped thermodynamics but ultimately connected macroscopic energy transformations to the microscopic behavior of matter.

Before entropy was formalized, Sadi Carnot laid essential groundwork by analyzing idealized heat engines. He showed that the maximum efficiency of any engine depends only on the temperatures of its hot and cold reservoirs—a startling result that implied a universal limit independent of the working substance. Rudolf Clausius later distilled Carnot's insights into a rigorous mathematical framework, coining the term Entropie from the Greek word tropē (transformation) to denote a new state function that tracks the irreversibility inherent in all real processes.

1824
Carnot's Ideal Engine
Sadi Carnot publishes Réflexions sur la puissance motrice du feu, establishing that engine efficiency depends only on reservoir temperatures—foreshadowing the entropy concept.
1854
Clausius Inequality
Rudolf Clausius introduces the inequality ∮ δQ/T ≤ 0 for cyclic processes, showing that heat divided by temperature tracks irreversibility and defining entropy as a state function.
1865
Entropy Named
Clausius formally names the quantity Entropie and declares: 'The entropy of the universe tends to a maximum,' articulating the Second Law in its most concise form.
1877
Boltzmann's Statistical Interpretation
Ludwig Boltzmann connects entropy to the number of microstates Ω via S = kB ln Ω, providing a molecular-level explanation for why entropy increases.
1948
Information Entropy
Claude Shannon introduces information entropy in his mathematical theory of communication, revealing deep structural parallels between thermodynamic disorder and uncertainty in data transmission.

The central question that entropy answers is deceptively simple: in which direction will a process spontaneously proceed, and how far from ideal is a given transformation? Heat flows from hot to cold, gases expand into vacuums, and ice melts in warm rooms—none of these processes violate energy conservation if run in reverse, yet we never observe them doing so. Entropy provides the missing criterion that the First Law alone cannot supply.

Core Principles & Definitions

Entropy is a thermodynamic state function, meaning its value depends only on the current equilibrium state of a system—not on the path taken to reach that state. This path-independence is what allows us to calculate entropy changes for irreversible processes by constructing any convenient reversible path between the same initial and final states. At the macroscopic level, entropy quantifies the fraction of a system's internal energy that is unavailable for doing work at a given temperature; at the microscopic level, it measures the number of ways energy can be distributed among the particles of a system.

1

State Function Property

Entropy S depends only on the thermodynamic state (P, V, T, composition). The change ΔS between two equilibrium states is path-independent, enabling calculation via any reversible path connecting those states.
2

Clausius Definition

For a reversible process, the infinitesimal entropy change is dS = δQrev / T. This ratio of reversible heat transfer to absolute temperature is the thermodynamic definition of entropy.
3

Second Law Statement

For any process in an isolated system, ΔStotal ≥ 0. Equality holds only for reversible processes; all real (irreversible) processes produce a net entropy increase.
4

Boltzmann's Statistical View

S = kB ln Ω, where Ω is the number of accessible microstates. Systems evolve toward macrostates with the greatest number of microstates, which is the statistical basis for the Second Law.
5

Entropy Generation

The quantity Sgen = ΔStotal = ΔSsys + ΔSsurr is always ≥ 0 and measures the degree of irreversibility of a process.
KEY TAKEAWAY
Think of entropy like shuffling a well-ordered deck of cards. There is only one arrangement that is perfectly sorted (ace through king in each suit), but there are roughly 8 × 1067 possible arrangements. Each shuffle is overwhelmingly likely to move you toward one of the enormous number of disordered states, not back to the single ordered one. Entropy formalizes this asymmetry: systems naturally evolve toward macrostates that correspond to the largest number of microstates, not because ordered states are forbidden, but because they are astronomically improbable.

Visual Explanation — Entropy in a Carnot Cycle

A temperature–entropy (T–S) diagram provides one of the clearest windows into how entropy behaves during thermodynamic processes. In such a diagram, reversible heat transfer appears as the area under the process curve, isothermal steps are horizontal lines, and adiabatic reversible (isentropic) steps are vertical lines. The Carnot cycle—the benchmark of maximum efficiency—forms a simple rectangle on the T–S plane, making it an ideal starting point for building visual intuition about entropy changes.

The Carnot cycle appears as a rectangle on the T–S diagram. The top edge (A → B) represents isothermal heat absorption at TH, where the system's entropy increases from S₁ to S₂. The right edge (B → C) is an isentropic expansion—entropy stays constant while the temperature drops. The bottom edge (C → D) is isothermal heat rejection at TC, and the left edge (D → A) completes the cycle with an isentropic compression. The enclosed area equals the net work output of the cycle.

Several critical observations emerge from this diagram. First, during the isothermal expansion A → B the system absorbs heat QH = TH(S₂ − S₁), which is just the area under that horizontal line. During isothermal compression C → D, heat QC = TC(S₂ − S₁) is rejected to the cold reservoir. The net entropy change of the working fluid over one complete cycle is zero because entropy is a state function and the system returns to its initial state. For the Carnot cycle, the entropy gained by the cold reservoir exactly equals the entropy lost by the hot reservoir, so the total entropy change of the universe is also zero—confirming that this idealized cycle is fully reversible.

Mathematical Framework

The mathematical formulation of entropy begins with Clausius's definition and extends through the entropy balance equation used in engineering analysis. Understanding these expressions and their conditions of applicability is essential for computing entropy changes in both idealized and real processes.

CLAUSIUS DEFINITION
dS = δQ_rev / T
dS = infinitesimal entropy change (J/K), δQrev = infinitesimal heat transfer along a reversible path (J), T = absolute temperature (K). The δ indicates an inexact differential—heat is path-dependent—but the ratio δQrev/T is exact.
FINITE ENTROPY CHANGE
ΔS = ∫₁² (δQ_rev / T)
For a process taking the system from state 1 to state 2, integrate along any reversible path connecting those states. Because S is a state function, the result is independent of the specific reversible path chosen.
IDEAL GAS ENTROPY CHANGE
ΔS = nC_v ln(T₂/T₁) + nR ln(V₂/V₁)
n = number of moles, Cv = molar heat capacity at constant volume (J/(mol·K)), R = universal gas constant (8.314 J/(mol·K)), T = absolute temperature (K), V = volume (m³). An equivalent form uses Cp and pressure: ΔS = nCp ln(T₂/T₁) − nR ln(P₂/P₁).
ENTROPY BALANCE (CLOSED SYSTEM)
ΔS_sys = ∫(δQ/T)_boundary + S_gen (S_gen ≥ 0)
ΔSsys = entropy change of the system, ∫(δQ/T)boundary = entropy transfer accompanying heat transfer across the system boundary, Sgen = entropy generated due to irreversibilities within the system (always ≥ 0). For a reversible process, Sgen = 0.
Important Distinction
Entropy can decrease for a system (e.g., when heat flows out), but the total entropy of the system plus its surroundings can never decrease. When you calculate a negative ΔSsys, always check that ΔSsys + ΔSsurr ≥ 0 to verify consistency with the Second Law.

Entropy Changes for Common Processes

To develop practical fluency with entropy, it is helpful to catalogue entropy changes for the standard idealized processes encountered in thermodynamics courses. The diagram below maps out how entropy changes during four fundamental processes for an ideal gas, and the accompanying table summarizes the key formulas.

On a T–S diagram, an isothermal process is horizontal (constant T, changing S), while an isentropic process is vertical (constant S, changing T). Constant-volume and constant-pressure heating curves are concave upward, with the constant-pressure curve having a steeper slope because Cp > Cv.
Summary of entropy change expressions for common thermodynamic processes.
ProcessConstraintΔS Expression (Ideal Gas)Sign of ΔS
IsothermalT = constnR ln(V₂/V₁)> 0 if expansion, < 0 if compression
Isentropic (Adiabatic Rev.)Q = 0, reversible0= 0
Constant VolumeV = constnCv ln(T₂/T₁)> 0 if heated, < 0 if cooled
Constant PressureP = constnCp ln(T₂/T₁)> 0 if heated, < 0 if cooled
Phase ChangeT, P = constQphase / T> 0 for melting/boiling, < 0 for freezing/condensation
Free ExpansionQ = 0, W = 0, ΔU = 0nR ln(V₂/V₁)> 0 (irreversible)

Notice that the free expansion has the same ΔS formula as the isothermal expansion despite being an entirely different physical process (no work done, no heat transferred). This is precisely because entropy is a state function: the entropy change depends only on the initial and final states (same T, different V), not on the process path. To compute ΔS for the irreversible free expansion, we replace it with a hypothetical reversible isothermal expansion connecting the same endpoints, evaluate the integral, and obtain the answer. This 'reversible-path trick' is one of the most powerful techniques in entropy calculations.

Worked Example — Entropy Change for Heat Transfer Between Two Blocks

A 2.00 kg copper block at 500 K is placed in thermal contact with a 3.00 kg copper block at 300 K inside an insulated enclosure. The specific heat capacity of copper is c = 385 J/(kg·K). Find the final equilibrium temperature and the total entropy change of the system.

Entropy Change for Irreversible Heat Transfer
1
Step 1 — Find the Equilibrium TemperatureBecause the enclosure is insulated, energy conservation gives m₁c(Tf − T₁) + m₂c(Tf − T₂) = 0. Since the material is the same, c cancels: Tf = (m₁T₁ + m₂T₂) / (m₁ + m₂) = (2.00 × 500 + 3.00 × 300) / (2.00 + 3.00).
Tf = 1900 / 5.00 = 380 K
2
Step 2 — Entropy Change of the Hot BlockFor a solid with constant specific heat undergoing a temperature change: ΔShot = m₁c ln(Tf/T₁) = 2.00 × 385 × ln(380/500) = 770 × ln(0.760) = 770 × (−0.2744).
ΔShot = −211.3 J/K
3
Step 3 — Entropy Change of the Cold BlockΔScold = m₂c ln(Tf/T₂) = 3.00 × 385 × ln(380/300) = 1155 × ln(1.267) = 1155 × 0.2364.
ΔScold = +273.0 J/K
4
Step 4 — Total Entropy ChangeΔStotal = ΔShot + ΔScold = −211.3 + 273.0.
ΔStotal = +61.7 J/K
5
Step 5 — InterpretationThe positive total entropy change confirms that heat transfer across a finite temperature difference is an irreversible process. The entropy gained by the cold block exceeds the entropy lost by the hot block. The 61.7 J/K represents the entropy generated—energy that has been degraded in quality and can never be fully converted to work.

Reversible vs. Irreversible Processes — Entropy Perspective

The distinction between reversible and irreversible processes is the conceptual backbone of entropy analysis. Every real process—friction, heat conduction across a temperature gradient, unresisted expansion, mixing of different gases—generates entropy and is therefore irreversible. A reversible process is an idealization in which the system passes through a continuous sequence of equilibrium states and can be exactly reversed without leaving any trace on the surroundings. Understanding the contrast between these two categories clarifies when ΔStotal equals zero versus when it must be positive.

Comparison of reversible and irreversible processes from an entropy standpoint.
FeatureReversible ProcessIrreversible Process
S_gen= 0> 0
ΔS_total (universe)= 0> 0
EquilibriumSystem is in equilibrium at every instant (quasi-static)System passes through non-equilibrium states
Driving forceInfinitesimal (ΔT → 0, ΔP → 0)Finite (ΔT, ΔP, friction, etc.)
Work outputMaximum possible for given state changeLess than maximum (some energy dissipated)
Exists in practice?No — it is a theoretical benchmarkYes — every real process
ΔS_sys calculationDirectly from ∫ δQ/T along actual pathMust use a hypothetical reversible path between same endpoints
KEY TAKEAWAY
A reversible process is like a perfectly balanced seesaw: the slightest push in either direction tips it the other way with no energy wasted. An irreversible process is like dragging a box across sandpaper—you can push it forward, but you can never recover the heat generated by friction to push it back to exactly where it started. In thermodynamics, the 'sandpaper' is any finite gradient (temperature, pressure, concentration) that drives a spontaneous change, and the entropy generated measures how much 'useful work potential' was permanently lost.

Connection to Statistical Mechanics and Free Energy

The Clausius definition of entropy is purely macroscopic—it says nothing about molecules. Boltzmann's statistical definition bridges this gap by relating entropy to the number of microstates Ω consistent with a given macrostate: S = kB ln Ω. Here kB = 1.381 × 10⁻²³ J/K is the Boltzmann constant. A macrostate with more accessible microstates has higher entropy. Since systems naturally explore all available microstates with equal probability, they spontaneously evolve toward macrostates of higher Ω—this is the statistical basis of the Second Law.

Comparing macroscopic, statistical, and free-energy perspectives on entropy.
AspectClausius (Macroscopic)Boltzmann (Statistical)Gibbs/Helmholtz (Free Energy)
DefinitiondS = δQ_rev / TS = k_B ln ΩG = H − TS; spontaneity when ΔG < 0 at constant T, P
What it measuresRatio of heat to temperature along reversible pathLogarithm of the number of microstatesCombines entropy and enthalpy into a single spontaneity criterion
ScopeAny system in thermal equilibriumSystems amenable to microstate counting (ideal gas, lattice models)Constant T and P processes (chemistry, biology)
StrengthsNo molecular model required; universalProvides molecular insight; explains fluctuationsDirectly predicts reaction spontaneity and equilibrium
LimitationGives no molecular-level understandingCounting Ω can be intractable for complex systemsRestricted to specific constraints (const T, P or const T, V)

In more advanced courses, you will encounter the Gibbs free energy G = H − TS and the Helmholtz free energy A = U − TS. These constructions fold the entropy of the surroundings into a single system-level quantity, yielding the criterion ΔG < 0 for spontaneity at constant T and P—arguably the most widely used criterion in chemistry and biochemistry. The connection is straightforward: at constant T and P, ΔG = ΔH − TΔS, and the condition ΔG < 0 is mathematically equivalent to ΔStotal > 0. Understanding entropy deeply will make these powerful tools feel natural rather than mysterious.

Practice Problems

PROBLEM 1CONCEPTUAL
An ideal gas undergoes a free expansion into a vacuum (Q = 0, W = 0). The gas temperature remains unchanged. Does the entropy of the gas increase, decrease, or stay the same? Does the entropy of the surroundings change? Is this process reversible or irreversible? Justify each answer.
PROBLEM 2BASIC CALCULATION
Calculate the entropy change when 2.00 mol of an ideal gas expand isothermally and reversibly from 10.0 L to 30.0 L at 350 K. Use R = 8.314 J/(mol·K).
PROBLEM 3INTERMEDIATE
A 5.00 kg iron block (c = 449 J/(kg·K)) at 600 K is quenched by dropping it into a large lake at 290 K. The lake is so large that its temperature remains essentially constant at 290 K. Calculate ΔS of the iron block, ΔS of the lake, and ΔStotal. Is the process reversible?
PROBLEM 4APPLIED
In a Carnot refrigerator operating between a cold space at 255 K and the ambient environment at 305 K, the compressor delivers 800 J of work per cycle. Determine the entropy changes of the cold reservoir, the hot reservoir, and the total universe per cycle.
PROBLEM 5CRITICAL THINKING
Consider two processes that both take one mole of an ideal monatomic gas from state A (300 K, 10 L) to state B (600 K, 20 L). Process I is a reversible path consisting of a constant-volume heating followed by an isothermal expansion. Process II is an irreversible path. Show that ΔS is the same for both processes, compute its numerical value, and explain conceptually why the entropy change must be path-independent even though the heat transferred is not.

Lesson Summary

Entropy is a thermodynamic state function defined macroscopically by dS = δQ_rev / T (Clausius) and microscopically by S = k_B ln Ω (Boltzmann). It quantifies the degree of energy dispersal in a system and, at the molecular level, the number of accessible microstates. The Second Law dictates that the total entropy of an isolated system (or the universe) can never decrease: ΔStotal ≥ 0, with equality holding only for reversible processes.

To calculate entropy changes, leverage the path-independence of ΔS by constructing any convenient reversible path between the two states—this is the 'reversible-path trick' central to all entropy problems. For ideal gases, the formula ΔS = nCv ln(T₂/T₁) + nR ln(V₂/V₁) covers all processes. For solids and liquids with constant specific heat, ΔS = mc ln(T₂/T₁). For phase changes at constant T and P, ΔS = Qphase/T. Always verify that ΔStotal ≥ 0 to confirm consistency with the Second Law, and remember that a positive Sgen signals irreversibility and lost work potential.

Varsity Tutors • Thermodynamics • Entropy Definition & Changes