Entropy, Irreversibility, and the Arrow of Time
~8 min read · Tier toggle changes depth (T1 plain · T2 standard · T3 rigour)
You should be comfortable with the first law of thermodynamics and the definition of internal energy. You must also understand the distinction between reversible and irreversible processes, as established in the previous lecture.
Why this matters
Entropy provides the mathematical criterion that distinguishes the past from the future in physical systems.
While the fundamental laws of mechanics are time-reversal symmetric, macroscopic phenomena exhibit a clear directionality. A cup of coffee cools down but never spontaneously heats up; ink diffuses in water but never spontaneously un-diffuses. This "arrow of time" is not derived from the microscopic equations of motion directly, but emerges from the statistical behaviour of large numbers of particles. Entropy quantifies this irreversibility, providing a rigorous bound on the direction of spontaneous change. In this lecture, we move beyond the definition of entropy as a state function to explore its profound implications for the dynamics of isolated systems and the fundamental nature of time in thermodynamics.
The Second Law in Plain Language
Imagine a sealed box divided into two halves by a partition. One half contains gas, the other is a vacuum. If you remove the partition, the gas will expand to fill the entire box. It will never spontaneously compress back into one half. This observation is the essence of the second law of thermodynamics. Unlike the first law, which conserves energy, the second law dictates the direction of processes. It tells us that while energy is conserved, the "quality" of energy degrades over time. Entropy is the measure of this degradation. In an isolated system, entropy can only stay the same or increase. It never decreases. This simple rule explains why we remember the past but not the future, why heat flows from hot to cold, and why perpetual motion machines of the second kind are impossible. The universe is not merely running down; it is evolving towards states of higher probability, which we quantify as higher entropy.
The Entropy Inequality
The rigorous statement of the second law is encapsulated in the Clausius inequality. For any cyclic process, the integral of the heat transfer divided by the temperature of the reservoir is less than or equal to zero. This inequality is the mathematical backbone of thermodynamics. It implies that for any real process, the change in entropy of the system is greater than or equal to the heat transferred divided by the temperature. This leads to the crucial result that for an isolated system, where no heat is exchanged with the surroundings, the entropy change must be non-negative. This is the formal statement of the second law: the entropy of an isolated system never decreases.
Equality holds only for reversible processes, which are idealizations that occur infinitely slowly. All real processes are irreversible, meaning they generate entropy. This entropy generation is the measure of the irreversibility of the process. The more irreversible the process, the greater the entropy produced. This concept allows us to quantify the "cost" of irreversibility in terms of lost work potential, a topic we will explore in the context of thermodynamic potentials.
Statistical Interpretation and Microstates
To understand why entropy increases, we must look at the microscopic level. Entropy is related to the number of microstates \(\Omega\) accessible to a system in a given macrostate. The Boltzmann formula, \(S = k_B \ln \Omega\), connects the macroscopic quantity entropy to the microscopic count of configurations. A macrostate with a higher entropy corresponds to a vastly larger number of microstates. When a system evolves spontaneously, it moves from a macrostate with fewer microstates (lower probability) to one with more microstates (higher probability). This is not a force, but a statistical inevitability. For a system with \(10^{23}\) particles, the probability of all particles spontaneously gathering in one corner is so infinitesimally small that it is effectively zero. Thus, the arrow of time is a statistical arrow, pointing towards the most probable configuration.
This statistical view resolves the apparent paradox of time-reversal symmetry in mechanics. While individual particle collisions are reversible, the collective behaviour of a huge number of particles is overwhelmingly likely to evolve towards equilibrium. The "past" is the state with lower entropy, and the "future" is the state with higher entropy. This perspective bridges the gap between the deterministic laws of classical mechanics and the probabilistic nature of thermodynamics.
Irreversibility and Dissipation
Irreversibility is closely linked to the concept of dissipation. In any real process, some energy is "dissipated" into random molecular motion, increasing the entropy of the system and its surroundings. This dissipated energy is no longer available to do useful work. The second law imposes a limit on the efficiency of heat engines and refrigerators. No engine can convert all heat into work; some heat must always be rejected to a cold reservoir. This is because the entropy of the working substance must return to its initial state after a cycle, requiring the rejection of heat to maintain the entropy balance. The concept of dissipation is central to understanding the limitations of energy conversion technologies and the fundamental constraints on physical processes.
The total entropy change of the universe (system plus surroundings) is always positive for real processes. This global increase in entropy is the ultimate measure of irreversibility. It tells us that while local decreases in entropy are possible (e.g., freezing water), they are always accompanied by a larger increase in entropy elsewhere (e.g., the release of latent heat to the environment). This global perspective is essential for analyzing complex systems and understanding the direction of spontaneous change in non-isolated systems.
The derivation
The Clausius Inequality and Entropy as a State Function
For any cyclic process a system undergoes while exchanging heat with reservoirs at absolute temperature T, the Clausius integral obeys ∮ dQ/T ≤ 0, with equality if and only if the cycle is reversible. The vanishing of ∮ dQrev/T makes ∫ dQrev/T path-independent, so dS = dQrev/T defines a state function, the entropy S; and for any real process dS ≥ dQ/T, whence the entropy of an isolated system never decreases.
A common misconception is that entropy is a measure of "disorder." While this heuristic is often useful, it is imprecise. Entropy is a measure of the number of microstates compatible with a macrostate. A system can become more "ordered" (e.g., crystal formation) while its total entropy increases, provided the entropy of the surroundings increases by a larger amount. Always rely on the rigorous definition \(S = k_B \ln \Omega\) and the inequality \(dS \geq \delta Q/T\) rather than intuitive notions of disorder. C
Worked example
Problem. An ideal gas with \(n = 2.0\) moles is initially at \(T_1 = 300\ \mathrm{K}\) and \(V_1 = 10\ \mathrm{L}\). It undergoes an irreversible free expansion into a vacuum until its volume is \(V_2 = 20\ \mathrm{L}\). Calculate the change in entropy of the gas. Assume the gas is ideal and the process is adiabatic.
- Identify the process: Free expansion into a vacuum is an irreversible adiabatic process. No work is done (\(W = 0\)) and no heat is exchanged (\(Q = 0\)).
- Since the gas is ideal and the process is adiabatic with no work, the internal energy change is zero (\(\Delta U = 0\)). For an ideal gas, internal energy depends only on temperature, so the final temperature \(T_2\) is equal to the initial temperature \(T_1\). Thus, \(T_2 = 300\ \mathrm{K}\).
- Entropy is a state function. We can calculate the change in entropy by considering a reversible path between the same initial and final states. A convenient reversible path is an isothermal expansion from \(V_1\) to \(V_2\) at temperature \(T\).
- For a reversible isothermal expansion of an ideal gas, the change in entropy is given by \(\Delta S = nR \ln(V_2/V_1)\).
- Substitute the values: \(n = 2.0\ \mathrm{mol}\), \(R = 8.314\ \mathrm{J/(mol \cdot K)}\), \(V_1 = 10\ \mathrm{L}\), \(V_2 = 20\ \mathrm{L}\).
- Calculate: \(\Delta S = 2.0 \times 8.314 \times \ln(20/10) = 16.628 \times \ln(2)\).
- \(\ln(2) \approx 0.6931\). Therefore, \(\Delta S \approx 16.628 \times 0.6931 \approx 11.52\ \mathrm{J/K}\).
Units check. The unit of \(nR\) is \(\mathrm{J/(mol \cdot K)} \cdot \mathrm{mol} = \mathrm{J/K}\). The logarithm is dimensionless. The final unit is \(\mathrm{J/K}\), which is correct for entropy.
Three quick tests
- Can entropy decrease in an isolated system?
Answer
No. The second law states that the entropy of an isolated system never decreases (\(\Delta S_{\text{isolated}} \geq 0\)). - What is the relationship between entropy and the number of microstates?
Answer
Entropy is proportional to the natural logarithm of the number of microstates: \(S = k_B \ln \Omega\). - Why is free expansion irreversible?
Answer
Free expansion generates entropy because the gas moves from a state of lower probability (fewer microstates) to a state of higher probability (more microstates) without any compensating decrease in entropy elsewhere.
Problem set for PU-203 →
Each derivation in this unit ends with five graded problems with full solutions.
Lecture notes checked line by line by the editor; every number in the worked example was recomputed.