Entropy Change During Phase Transitions

Phase transitions—the transformations of matter from solid to liquid or liquid to gas—are among the most common yet profound phenomena we observe. A familiar puzzle arises when boiling water: despite continuously adding heat, the temperature remains fixed at 100°C until the last drop becomes steam. This raises a fundamental question: where does that energy go? The answer lies in the concept of entropy, a measure of disorder and freedom at the microscopic level. This article demystifies the entropy changes that drive phase transitions, revealing a universal principle that connects everyday observations to the frontiers of science. In the following chapters, we will first explore the core thermodynamic and statistical principles governing these transformations, introducing foundational concepts like latent heat, Gibbs free energy, and the Clausius-Clapeyron equation. We will then journey across diverse disciplines to witness how these elegant rules explain the behavior of everything from planetary geology and quantum superconductors to living cells and smart materials.

Imagine you are boiling a pot of water. You turn the stove on, and the temperature of the water climbs steadily: 60°C, 80°C, 95°C… then, it hits 100°C and starts to bubble furiously. You keep the stove on, pumping energy into the pot, but a funny thing happens: the thermometer doesn't budge. It stays stuck at 100°C until the last drop of water has turned into steam. Where did all that energy go? It wasn't making the molecules move faster—their average kinetic energy, which is what temperature measures, was constant. This puzzle is the gateway to understanding phase transitions.

The energy you add during boiling isn't lost; it's being used to do something more profound than just speeding molecules up. It's being used to break them free from the cozy, jumbled attractions they feel for each other in the liquid state and launch them into the wide-open chaos of the gaseous state. This energy is called ​​latent heat​​ ($L$), "latent" because it seems to be hidden, not showing up as a change in temperature.

To account for where this energy goes, thermodynamics gives us a wonderfully powerful concept: ​​entropy​​ ($S$). During a phase transition at a constant temperature $T$, the change in entropy is simply the latent heat absorbed divided by the temperature at which it happens:

For a process at constant pressure, this latent heat is precisely the change in another thermodynamic quantity called ​​enthalpy​​ ($H$), which accounts for both the system's internal energy and the work done to make room for itself. So, for a transition like the vaporization of a refrigerant in a cooling system, the entropy increase is directly tied to measurable quantities like the change in internal energy and the work done by the expanding gas. This equation, $\Delta S = \Delta H / T$, is the macroscopic law of the land for phase transitions. But why does this rule hold? What is entropy, really?

To find the true meaning of entropy, we must shrink ourselves down to the molecular level. The great physicist Ludwig Boltzmann gave us the key with one of the most beautiful equations in all of science: $S = k_B \ln W$. Here, $k_B$ is a fundamental constant of nature (Boltzmann's constant), and $W$ is the magic ingredient: it’s the number of distinct microscopic arrangements—or ​​microstates​​—that are consistent with the macroscopic properties we observe. In simple terms, entropy is a measure of the system's microscopic "freedom" or the number of ways it can arrange its internal affairs.

Let's look at a substance as it goes from solid to liquid to gas:

• ​​Solid (Ice):​​ The water molecules are locked into a highly ordered, crystalline lattice. They are not still; they vibrate and jiggle in place. But they are caged, with very little freedom to move around. The number of possible arrangements, $W_{\text{solid}}$, is relatively small.

• ​​Liquid (Water):​​ As we add heat and the ice melts, the molecules gain enough energy to break free from their fixed lattice positions. They can now tumble, rotate, and slide past one another. They are still packed closely together, but their freedom has dramatically increased. The number of available microstates, $W_{\text{liquid}}$, is much larger than for the solid.

• ​​Gas (Steam):​​ When the water boils, the molecules break away from each other entirely. They fly about randomly, filling the entire volume of their container. They are no longer touching most of the time. The spatial freedom for each molecule explodes. Consequently, $W_{\text{gas}}$ is astronomically larger than $W_{\text{liquid}}$.

Since entropy is the logarithm of this number $W$, it's clear that both melting and boiling, which unlock new degrees of freedom for molecules, must result in a positive change in entropy, $\Delta S > 0$. The latent heat that seemed to "disappear" was actually paying the price for this newfound molecular freedom.

You might now wonder if the entropy increase is similar for melting and boiling. Experience and measurement tell us a resounding "no." The molar entropy of vaporization is almost always significantly larger than the molar entropy of fusion ($\Delta S_{\text{vap}} \gg \Delta S_{\text{fus}}$). Why this great disparity?

The answer lies in the dramatic difference in the type of freedom gained. When ice melts into water, the molecules are uncaged from a rigid lattice, but they remain in a condensed, crowded phase. The volume change is very small (in fact, for water, the volume actually decreases slightly). The primary gain is in local motional freedom.

The liquid-to-gas transition is a different story altogether. At standard pressure, when one mole of liquid water turns into steam, its volume expands by a factor of more than 1,600! This colossal increase in available space means the number of possible positions for each molecule increases to a staggering degree. The gain in this ​​configurational entropy​​ associated with volume completely dwarfs the gain from melting.

We can even put a number on this idea. Using the known latent heat of fusion for water, we can estimate the increase in molecular "options" upon melting. A simple calculation reveals that a single water molecule in the liquid state has roughly 14 times more available microscopic configurations than it had a moment before in the solid ice crystal. This gives a tangible feel for the "disorder" that entropy so elegantly quantifies.

We have a picture of systems tending towards higher entropy, towards more freedom. But there is a competing tendency: systems also tend towards lower energy, finding stability in strong, well-ordered bonds. The universe is a grand competition between these two drives: the quest for freedom (high entropy) and the quest for stability (low enthalpy).

The ultimate arbiter in this contest, for systems at constant temperature and pressure, is a quantity called the ​​Gibbs Free Energy​​, defined as $G = H - TS$. A system will always spontaneously evolve towards the state with the minimum possible Gibbs free energy.

Think of it as a balancing act. The $H$ term favors order and strong bonds. The $-TS$ term favors disorder and freedom. The temperature, $T$, acts as the referee, deciding how much weight to give to the entropy term.

• At low temperatures, the $TS$ term is small, and enthalpy ($H$) dominates. The system minimizes its energy by forming an ordered solid.
• At high temperatures, the $TS$ term is large and entropy dominates. The system minimizes its free energy by flying apart into a disordered gas.

A phase transition occurs at the precise temperature and pressure where the Gibbs free energies of two or more phases become equal—a perfect standoff. Along a coexistence curve, like the one separating liquid water and ice, $G_{\text{solid}} = G_{\text{liquid}}$. At the unique ​​triple point​​ of a substance, all three phases coexist in a beautiful equilibrium, and their Gibbs free energies are all equal: $G_{\text{solid}} = G_{\text{liquid}} = G_{\text{vapor}}$.

This condition of equal Gibbs energy, $G_1(P,T) = G_2(P,T)$, defines the boundary lines on a phase diagram. This simple equality contains a hidden gem. If we "walk" along this line, a tiny change in temperature $dT$ must be accompanied by a tiny change in pressure $dP$ in such a way that the Gibbs energies remain equal. A little calculus (essentially requiring that $dG_1 = dG_2$) reveals a powerful relationship governing the slope of that boundary. This is the famous ​​Clausius-Clapeyron equation​​:

This isn't just a formula; it's a profound statement about the inner workings of matter. It connects the macroscopic shape of a phase boundary ($dP/dT$) to the microscopic changes occurring during the transition: the change in entropy ($\Delta S$, or latent heat $L$) and the change in volume ($\Delta V$).

This single equation explains a host of everyday phenomena.

• ​​Why can you ice skate?​​ For the melting of ice, $\Delta V$ is negative (ice is famously less dense than liquid water). The equation tells us $dP/dT$ must be negative. Increasing the pressure lowers the melting point. The pressure from a skate blade can help melt the ice beneath it, creating a lubricating layer of water.
• ​​Why does a pressure cooker work?​​ For boiling, $\Delta V$ is large and positive. Thus, $dP/dT$ is positive. Increasing the pressure (by sealing the lid) forces the boiling point to a higher temperature, cooking food much faster.

The framework we've built is not only powerful but also beautifully self-consistent.

First, consider that entropy is a ​​state function​​—the change in entropy between two states doesn't depend on the path taken. This means the entropy change in going directly from solid to vapor (sublimation) must equal the sum of the changes for melting and then boiling. This directly implies that the latent heats must also add up: $L_{\text{sublimation}} = L_{\text{fusion}} + L_{\text{vaporization}}$. This is exactly what is observed experimentally, a reassuring check on our logic.

Second, the Clausius-Clapeyron equation makes a stunning prediction when combined with another deep law of physics: the ​​Third Law of Thermodynamics​​. The Third Law states that the entropy of a perfect crystal approaches zero as the temperature approaches absolute zero ($T \to 0$). This implies that for a transition between two ordered phases near $0$ K, the entropy change $\Delta S$ must also approach zero. Plugging this into the equation, we find that $\frac{dP}{dT}$ must go to zero. This means that all phase boundaries on a P-T diagram must become perfectly flat as they approach absolute zero! It's a profound and non-obvious prediction that demonstrates the deep unity of thermodynamic principles.

So far, all our transitions—melting, boiling, sublimating—have involved a jump in entropy and a non-zero latent heat. These are called ​​first-order phase transitions​​. But are there other kinds?

Indeed, there are. Consider the amazing phenomenon of ​​superconductivity​​. When certain metals are cooled below a critical temperature $T_c$, their electrical resistance vanishes completely. If you measure the thermodynamics of this transition, you find that there is zero latent heat! This means the entropy is continuous across the transition ($\Delta S = 0$). However, if you measure the specific heat capacity $C_P$, you see a sudden, finite jump right at $T_c$.

This is the signature of a ​​second-order phase transition​​. In the Ehrenfest classification, these are transitions where the Gibbs free energy and its first derivatives (like entropy and volume) are continuous, but a second derivative (like specific heat) is discontinuous. There is no latent heat "cost" to enter the superconducting state; instead, the ordering happens in a more subtle, continuous fashion, revealed only by the system's response to being heated.

We can now tie all these ideas together to see how scientists perform one of the great feats of thermodynamics: calculating the absolute entropy of a substance from scratch. Let's trace the journey of one mole of water from the coldest possible state to steam at 120°C.

1. ​​Start at 0 K:​​ By the Third Law, we define the entropy of a perfect ice crystal at absolute zero to be $S(0 \text{ K}) = 0$.

2. ​​Heat the Ice (0 K to 273.15 K):​​ We slowly add heat. For each tiny bit of heat $dQ$, the entropy increases by $dS = dQ/T$. We sum these contributions by integrating the specific heat: $\Delta S_1 = \int_0^{T_m} \frac{C_{p, \text{ice}}(T)}{T} dT$.

3. ​​Melt the Ice (at 273.15 K):​​ The temperature holds constant as the ice melts. A sudden, large jump in entropy occurs: $\Delta S_2 = \frac{\Delta H_{\text{fusion}}}{T_m}$.

4. ​​Heat the Water (273.15 K to 373.15 K):​​ Once all the ice has melted, the temperature of the liquid water begins to rise again. We add up the entropy changes as before: $\Delta S_3 = \int_{T_m}^{T_b} \frac{C_{p, \text{liquid}}(T)}{T} dT$.

5. ​​Boil the Water (at 373.15 K):​​ Another plateau. The temperature holds steady as the liquid turns to gas, producing a massive jump in entropy: $\Delta S_4 = \frac{\Delta H_{\text{vaporization}}}{T_b}$.

6. ​​Heat the Steam (373.15 K to 393.15 K):​​ Finally, we heat the steam to our target temperature of 120°C (393.15 K): $\Delta S_5 = \int_{T_b}^{T_f} \frac{C_{p, \text{gas}}(T)}{T} dT$.

The absolute entropy of the steam at the end is simply the sum of all these pieces: $S_{\text{final}} = \Delta S_1 + \Delta S_2 + \Delta S_3 + \Delta S_4 + \Delta S_5$. This grand tally, a journey of smooth climbs and abrupt jumps, beautifully illustrates how the continuous laws of heating and the discrete laws of phase transitions combine to define the thermodynamic state of all matter.

In our previous discussion, we laid bare the beautiful and surprisingly simple laws governing entropy change during phase transitions. We saw that the entropy change, $\Delta S$, when a substance melts or boils is simply the latent heat, $L$, absorbed during the process, divided by the temperature, $T$, at which it happens: $\Delta S = L/T$. One might be tempted to file this away as a neat but niche piece of thermodynamic trivia. Nothing could be further from the truth.

This single relationship is not a footnote in a dusty textbook; it is a master key, unlocking a breathtaking array of phenomena across science and engineering. It is the secret behind the formation of planets, the behavior of strange new materials, and even the very function of the cells in your body. It reveals, in a way few other principles can, the profound unity of nature. Let us now embark on a journey to see how this simple idea plays out on the grandest and most intimate of scales.

We can begin with something familiar and visceral: the heat of a furnace. In industrial processes like metallurgy or advanced additive manufacturing, pure elements are melted and reformed. When a laser melts iron powder to 3D-print a metal part, for instance, the total entropy of the iron increases in two distinct steps: first, as the solid powder heats up, and second, a significant jump as it melts into a liquid at a constant temperature. That jump, the entropy of fusion, quantifies the immense increase in molecular disorder as the rigid, crystalline lattice of solid iron gives way to the chaotic sloshing of atoms in the liquid state. This is thermodynamics in action, shaping the very materials of our modern world.

But what if we add another ingredient to the mix? What happens under the immense pressures found deep within the Earth? Here, the plot thickens. The relationship between phase transitions and entropy gives us the master tool for this inquiry: the Clausius-Clapeyron equation. This equation tells us how the melting or boiling temperature of a substance changes as we change the pressure. It predicts that if a substance expands upon melting (as most do), increasing the pressure on it will raise its melting point. It takes more thermal energy, more entropic drive, to overcome not only the bonds holding the solid together but also the immense external pressure trying to keep it compact.

This principle is at the heart of geochemistry. Consider the fascinating case of carbon. Under the familiar conditions of Earth's surface, graphite is the stable form. Diamond, its celebrated allotrope, is a breathtakingly ordered, crystalline structure—a state of much lower entropy than graphite. So how could the universe possibly favor the creation of this low-entropy gem from its more disordered cousin? The answer lies in pressure. The diamond structure is significantly denser than graphite. Deep within the Earth's mantle, the colossal pressures favor the denser arrangement. The Clausius-Clapeyron equation shows us that the transition from graphite to diamond becomes favorable under these extreme conditions, providing a beautiful example of how the interplay between energy, entropy, volume, and pressure dictates the very minerals that make up our world.

This drama of pressure and temperature plays out not just in static solid-state transformations but in the dynamic churning of the planet itself. When a chamber of molten magma begins to cool, crystals start to form. This act of crystallization is a phase transition from liquid to solid. As the crystals form, they release their latent heat of fusion into the surrounding rock. The magma, in crystallizing, becomes more ordered—its entropy decreases. But the heat it releases flows into the cooler surrounding rock, a process of irreversible heat transfer from a hot object to a cold one. The entropy increase of the surroundings is greater than the entropy decrease of the crystallizing magma, ensuring that the total entropy of the universe increases, as the Second Law of Thermodynamics demands. This process, known as fractional crystallization, is a primary driver of the diversity of rocks we see on Earth's surface, a planetary-scale sorting mechanism governed by the humble rules of entropy change.

The power of thermodynamics lies in its generality. The variables don't have to be pressure and volume. Any pair of "conjugate" variables—an intensive "force" (like pressure) and an extensive "displacement" (like volume)—will do. What if we replace pressure with something else, like a magnetic field?

We enter the strange, cold world of superconductivity. A type-I superconductor below a certain critical temperature, $T_c$, exhibits two astonishing properties: it conducts electricity with zero resistance, and it expels all magnetic fields from its interior (the Meissner effect). If you apply a strong enough external magnetic field, $H$, you can destroy the superconducting state and force the material back into its "normal," resistive state. This, too, is a phase transition.

Amazingly, the same logic we used for pressure and melting applies here. There exists a magnetic analogue to the Clausius-Clapeyron equation, which relates the slope of the critical field boundary, $dH_c/dT$, to the change in entropy and the change in magnetization, $\Delta M$, during the transition. The superconducting state is a highly ordered quantum state—a state of very low entropy. The normal state, with its vibrating atoms and scattered electrons, is far more disordered. Therefore, to transition from the ordered superconducting state to the disordered normal state, the system must absorb heat, just like melting ice. The entropy must increase. Our thermodynamic framework not only explains this but allows us to calculate precisely how much the entropy changes, based on purely magnetic measurements. That the same reasoning explains both the forging of a steel beam and the quantum behavior of a superconductor is a stunning testament to the unifying power of physics.

Let's bring our journey home, from the planetary and the quantum to the scale of molecules—the scale of life itself. Your own body contains trillions of cells, and each is enclosed by a membrane. This membrane is not a static wall; it is a fluid, dynamic interface, a "liquid crystal." Its ability to function depends critically on its physical state, which is governed by a phase transition.

At low temperatures, the long lipid molecules that form the membrane pack into a regular, rigid structure known as the "gel" phase. As the temperature rises, it undergoes a transition to a "liquid-disordered" state, where the lipid chains are free to move and wiggle—a state of much higher entropy. This is, in essence, a melting process. Life as we know it operates in this fluid state. This transition has a measurable latent heat and a corresponding entropy change that biophysicists can measure in the lab, a quantity that determines the membrane's fluidity, its permeability, and its ability to host the proteins that carry out the cell's business. The delicate dance of life happens right at the edge of this phase transition, controlled by the fundamental laws of entropy.

To truly engineer systems at this scale, whether they be biological or synthetic, we need models that capture the behavior of real substances. The ideal gas law is a wonderful simplification, but real molecules have size and they attract each other. More sophisticated models, like the van der Waals equation of state, account for these realities. What is so powerful is that our thermodynamic framework doesn't break down when faced with this complexity. By applying the very same principles—the Maxwell relations that stem from the nature of entropy as a state function—we can derive an expression for the latent heat of vaporization for a van der Waals fluid. We can see exactly how the parameters representing molecular size and attraction influence the entropy change during boiling. Thermodynamics provides the rigorous bridge from a microscopic picture of interacting molecules to the macroscopic properties we can measure.

Perhaps the most mind-bending application of these ideas lies in the field of modern materials, with so-called "smart" materials like shape-memory polymers. Imagine a piece of plastic that you can crumple into a ball, but when you gently heat it, it springs back to its original, complex shape. The "memory" is stored in entropy. These materials are made of long, coiled polymer chains tied together in a network. When you stretch the material (above a transition temperature), you are not primarily stretching atomic bonds; you are un-coiling the chains, forcing them into a more ordered, straightened, and statistically improbable low-entropy state. You can then "freeze" this low-entropy state in place by cooling the material. The desire of the material to return to its original shape is a powerful restoring force, driven not by the release of mechanical strain, but by the overwhelming statistical tendency of the polymer chains to return to their disordered, high-entropy coiled-up state. This is entropic elasticity, and it is the engine of shape recovery. The memory is not a stored force; it is a stored potential for disorder.

From the heart of a transforming planet, to the quantum perfection of a superconductor, to the fluid membrane of a living cell and the programmed "memory" of a smart polymer, the story is the same. Order gives way to disorder, and this change, this increase in entropy, can be harnessed, predicted, and understood. The simple relationship we began with is revealed not as a minor detail, but as a universal principle of profound beauty and astonishing reach.