Hot Water Freezes Faster Than Cold — And We STILL Don't Know Why

[00:00:00]All right, I want you to picture something. You are standing in your kitchen. You have two glasses of water.

[00:00:06]One glass is hot, not warm, hot, almost boiling. The other glass is cold. You put both glasses in the freezer at the same time. Same freezer, same shelf, same everything. Which one freezes first? You said the cold one. Of course, you said the cold one. That is the only possible answer. The cold water has less distance to travel. It is already closer to freezing. It has to lose fewer calories of heat. The hot water first has to cool down to the same temperature as the cold water and then it still has to do everything the cold water did after that. So the cold water has a head start. It wins every single time. Except it does not.

[00:00:47]Sometimes the hot water freezes first and nobody I mean nobody can fully explain why. This is not some internet rumor. This is not somebody's cousin in a bar telling you about something he read somewhere. Aristotle noticed this.

[00:01:02]The man who basically invented organized thinking about the natural world sat down somewhere around 350 BC and wrote that water which has been previously warmed cools sooner. He even described a practical application. People in the ancient Greek settlement of Pontis, who fished through holes in frozen lakes, would first set their water out in the sun before pouring it around the tent stakes at night because the warmed water froze faster and anchored the poles more quickly. These were not physicists.

[00:01:30]These were fishermen, and they had figured out something that modern science still cannot fully explain.

[00:01:36]Picture that. fisherman on a frozen lake 2,000 years before anyone wrote down a law of thermodynamics doing something that contradicts what any modern physics student would tell you. And it worked.

[00:01:48]It was practical knowledge. The kind of knowledge that does not come from theory. It comes from paying attention.

[00:01:54]Francis Bacon noticed it in the 17th century. He wrote about it. Daycart noticed it too and he tried to explain it. His explanation was that heating water drives off the particles that are least capable of being frozen. So what remains is purer and freezes more readily. That is actually not a terrible intuition for someone in the 1600 though it is not quite right. But at least Decart took the observation seriously enough to offer a mechanism. And then somehow the whole thing got swept under the rug. Scientists got more sophisticated. They built better thermometers. They wrote down Newton's law of cooling which describes how objects lose heat to their surroundings.

[00:02:32]And that law seemed so clean and so complete that nobody had reason to question it. The mathematics was settled. Hot things cool toward cold things. The bigger the temperature gap, the faster the cooling. But at no point does the hot thing leapfrog the cold thing. End a discussion. And that is where the story should have ended.

[00:02:52]Except for a 13-year-old boy in Tanzania. His name was Arasto Impemba.

[00:02:58]In 1963, he was a student at Magamba Secondary School in Tanganika, which is what Tanzania was called before independence. The boys at that school had a tradition. They would make ice cream. You boil some milk, you mix it with sugar, you let it cool down to about room temperature, and then you stick it in the freezing compartment of the refrigerator. Simple enough. But here is the thing about refrigerator space. In a school in East Africa in 1963, there was not much of it. Competition was fierce. So, one afternoon, the boys are all jockeying for the last spots in the freezer. One of Empa's classmates takes a shortcut. He skips the boiling step entirely. Just mixes cold milk with sugar and shoves it in. And Pemba goes the other way. He boils his milk, mixes in the sugar, but then he skips the cooling step. He cannot afford to wait.

[00:03:47]He takes the hot mixture and puts it straight into the freezer. About an hour and a half later, Empa comes back. His mixture is frozen solid into ice cream.

[00:03:58][clears throat] His classmates's cold mixture, still liquid, still slloshing around, not even close. Now, any reasonable person would say, "Well, something went wrong. The beers were different sizes. The freezer shelf was uneven. There was a draft. One container was deeper than the other. Maybe the cold mixture had more milk in it." Empa himself was not sure what happened, but he was curious. Not the kind of curious where you think about it for five minutes and then forget the kind of curious where where it nags at you. So he did what any good scientist does even if he did not think of himself as a scientist. He tried it again and he got the same result. He went to an ice cream seller in Tangga during a school holiday and asked the man how he made his product. The ice cream seller told him, "Oh, sure. You start with hot liquid. It freezes faster. Everyone in the business knows that that uh the guy had been doing this for years. It was just part of the trade, not a mystery, a fact.

[00:04:59]Emboldened, and Pemba went back to school and asked his physics teacher. He said, "Please, sir, why is it that when you put both hot milk and cold milk into a refrigerator at the same time, the hot milk freezes first?" And the teacher said, "And I absolutely love this exchange because it tells you everything about how authority can crush curiosity." The teacher said, "I do not think so, Empa." Impa pushed back. He said, "It is true, sir. I have done it myself." And the teacher's final answer was, "Well, all I can say is that that is Impemba's physics and not the universal physics." From that day on, whenever Empa made any kind of mistake in class, the teacher and the other students would say, "Oh, that is Impemba's mathematics. That is Impemba's physics." It became a joke, a put down, a way of saying, "You are not smart enough to question the textbook." But Impa did not quit. He kept the question alive in his mind. He sneaked into the biology laboratory when no teacher was around, filled two 50mm beers, one with cold tap water and one with hot water from a boiler, put them in the freezer, and came back an hour later. There was more ice in the beaker that had started hot. He was not dreaming. He was not confused. The result was repeatable. He brought three other boys the next day to witness the experiment.

[00:06:16]Um, they saw the same thing. Some of them started telling the rest of the school that Impea was right, but even they could hardly believe it. The head of the physics department said the experiment should not work and promised to try it himself, but apparently never followed through. Nobody with authority wanted to engage with the result because the result contradicted what they believed they already knew. Then something lucky happened. And Pembbea moved to Enqua High School, a professor of physics from the University College in Dar Salam. A man named Denise Osborne came to give a guest lecture. After the lecture, students were allowed to ask questions and Pemba stood up and asked his question. The other students laughed, but Osborne, and this is the crucial part of the story, Osborne did not laugh. He said something like, "I cannot think of a reason why that would happen, but I will try it when I get back to my laboratory." And when he tried it, he found that Empa was right.

[00:07:11]In 1969, Osborne and Empemba published a paper together in the journal Physics Education. The title of that paper was just one word, cool, with a question mark. That is the whole title. Osborne wrote in the introduction that the facts and Pembbea described might be familiar to many, though he himself had never heard of them, and he could find no prior references. He added pointedly that the story points to the danger of an authoritarian physics and is recorded in the hope that it will be of interest and encouragement to others. Interest and encouragement to others. That is a gentle way of saying if a student tells you something surprising, do not laugh at him. Try the experiment. Pembbember went on to study wildlife management and became a regional natural resources officer in Tanzania. He lived quietly, but his name traveled the world. In 2012, when the Royal Society of Chemistry held their competition, and Pembber himself was there to announce the winner, he had outlived the ridicule. The physics textbook had caught up with his kitchen. That paper launched decades of argument, hundreds of experiments, thousands of attempted explanations, and what I can only describe as one of the most delightful ongoing embarrassments in the history of thermodynamics.

[00:08:25]So, let us be detectives about this. Let us actually investigate. Let us figure out why the obvious answer might be wrong. The obvious answer relies on a very clean picture. You have two identical containers of water. One is at 90° C. The other is at 25°. You put them both in a freezer atus20°.

[00:08:45]The rate at which heat flows out of an object depends on the temperature difference between the object and its surroundings. That is Newton's law of cooling, a proportional relationship.

[00:08:55]So, the hot water does lose heat faster at first because the gap between 90 and minus 20 is enormous compared to the gap between 25 and minus 20. But here is the key assumption. As the hot water cools, it must pass through every temperature the cold water already sat at. When the hot water reaches 25°, it is caught up.

[00:09:16]And now the reasoning goes, it is in exactly the same situation the cold water was in at the start. Same temperature, same container, same freezer. So from that point on they should behave identically. The cold water started the race earlier. The cold water wins. That proof sounds airtight.

[00:09:33]If you wrote it on a chalkboard, most people in the room would nod and move on. No one would raise a hand. The logic seems watertight. It is the kind of argument where every step follows from the previous one and you feel clever for understanding it. But there is a hidden assumption in there buried so deep that almost nobody notices it. And the moment you drag it out into the light, the whole argument cracks open. The assumption is this, that a glass of water at 25°, which got there by cooling down from 90°, is the same in every relevant way as a glass of water that has been sitting at 25° the whole time.

[00:10:08]We are assuming that temperature tells us everything. That if two glasses of water are at the same temperature, they are in the same state. That the thermometer is the final word. What if it is not? What if something about the water changed while it was hot? What if the hot water on its way back down to 25° became a different kind of 25° water? Different in some subtle internal structural way that a thermometer cannot see. That is the crack in the logic.

[00:10:36]That is where everything gets interesting and that is where the investigation really begins. Let me walk through the suspects. We have at least five of them and not one of them has a perfect alibi. Suspect number one, evaporation. When you heat water, it evaporates faster. Steam comes off the surface. You can see it. So, by the time the hot water is cooled to the same temperature as the cold water, there is less of it. Less water means less mass to freeze. And less mass, all else being equal, freezes faster. Simple, elegant, satisfying. And it is real. It does happen. You can weigh the containers before and after and measure the mass loss. Some calculations have shown that evaporation alone, if you assume that is the only way the water loses heat, can account for the observed difference in freezing times. The math works out in certain idealized cases. But there are problems. First, the amount of mass lost to evaporation in typical experiments is small, a few percent. Is that enough to tip the balance? In some setups, maybe.

[00:11:38]In others, probably not. Second, and this is the big one, researchers have done this experiment in sealed containers where no water vapor can escape. No evaporation whatsoever. The lid is on, the mass stays constant. And they still see the empa effect.

[00:11:54]Evaporation contributes, but it cannot be the whole answer. If you plug the hole and the mystery persists, you need another suspect. Suspect number two, dissolved gases. Tap water is full of dissolved gases. carbon dioxide, oxygen, nitrogen, little bubbles of air, all mixed invisibly into the liquid. Hot water holds less dissolved gas than cold water. That is a basic principle of gas solubility. When you heat water, especially when you boil it, those gases bubble out. You can see them forming on the walls of the pot before it reaches a full boil. So, the water that started hot and then cooled back down has less dissolved gas in it than the water that was cold the whole time. And those dissolved gases might matter in ways that are not immediately obvious. Some researchers have speculated that dissolved gases interfere with convection currents, making it harder for cold water to circulate efficiently.

[00:12:49]Others have suggested that dissolved gas changes the specific heat capacity of the water slightly or affects the ease with which ice crystals can nucleate.

[00:12:57]There is even a fascinating theory involving microbubbles. When you heat water, especially vigorously, tiny gas bubbles form and persist in the liquid even after it cools. These microbubbles, too small to see with the naked eye, may act as little convective engines inside the water, shuttling heat from the warmer interior to the cooler surfaces faster than normal conduction would allow. One researcher estimated that the heat transfer enhancement for microbubbles could be proportional to the volume fraction of bubbles present, which itself depends on how hot the water was originally heated. This is plausible.

[00:13:30]Experiments have shown with high statistical certainty that dissolved gas content affects freezing behavior. But again, experiments with carefully degassed water, water that has been boiled and then sealed so no new gas can dissolve, still sometimes show the effect. So gases are a contributor, perhaps a significant one, but not the sole answer. Suspect number three, convection. This one is sneaky and beautiful. When hot water sits in a container inside a freezer, it does not cool uniformly. The water near the walls and the surface loses heat first. That cooler, denser water sinks. The hotter water in the center rises. You get these gorgeous circulation patterns. Little rivers within the glass. The hot water climbing. The cool water diving, carrying heat from the core to the surfaces where it can radiate away.

[00:14:17]These convection currents are vigorous in hot water. Moria much more vigorous than in lukewarm or cold water where the temperature is more uniform throughout the volume and the water is essentially sitting still. The temperature gradient, the difference between the hottest point and the coldest point inside the container is steeper in water that started hot. Steeper gradients drive faster heat transfer. The hot water in a sense stirs itself. There is a second piece to this. Cold water as it approaches freezing develops a peculiar behavior. water reaches its maximum density at about 4°. Below that, it actually gets less dense as it cools.

[00:14:56]So, near freezing water is lighter than slightly warmer water. That means the coldest water floats to the top where it freezes first, forming a lid of ice on the surface. And ice is a terrible conductor of heat. That ice lid traps the warmer water below it, insulating it, slowing the entire freezing process.

[00:15:13]Hot water with its strong convection currents resists forming that ice lid.

[00:15:18]The circulation keeps mixing the water, preventing the top from freezing into a solid sheet, while the interior stays liquid. The heat keeps escaping from an open liquid surface. When ice does eventually form, it often starts at the bottom and sides of the container rather than the top, leaving the surface exposed longer for efficient heat loss.

[00:15:36]Nicola Bregovich, a chemist from the University of Zagreb, who won the Royal Society of Chemistry's 2012 competition for the best explanation of the Empember effect, highlighted convection as essential. In his experiments, when he stirred the water continuously to eliminate convection effects, the Empember effect disappeared. That is a strong clue. But convection alone without the other suspects has never been demonstrated to fully account for the magnitude and consistency of the effect. It is a co-conspirator, not the ring leader.

[00:16:06]Let me pause here and point something out. We are three suspects and and each one of them has something real to offer.

[00:16:13]Each one is backed by evidence. Each one explains part of the phenomenon. But none of them alone is enough. And you might be wondering, well, if you add them all together, do they explain it?

[00:16:25]Maybe. But that is hard to test because each mechanism depends on the specific conditions of the experiment. And changing one condition might amplify one mechanism while suppressing another.

[00:16:35]This is why the impember effect is such a maddening problem. It is not that we have no explanations. We have too many.

[00:16:42]And sorting out which ones matter and how much and under what circumstances is like trying to figure out which ingredient in a stew is responsible for the flavor. The answer is probably all of them in some combination. But good luck proving that rigorously in a laboratory. Suspect number four. And this one shows just how tricky the problem is. The environment. Imagine your freezer has a thin layer of frost on the shelf. It many freezers do. You set a glass of cold water on that frost.

[00:17:11]The frost stays put. It sits between the glass and the metal shelf like a little insulating blanket. Frost is a lousy conductor. The heat from the glass has to fight through that layer to reach the cold metal underneath. Now you set a glass of very hot water on the same frost. What happens? The hot glass melts the frost. The ice layer disappears. Now the glass is sitting directly on the cold metal shelf with much better thermal contact. Heat flows out of the glass faster. The hot water has literally remodeled the environment it is sitting in. This is real. It absolutely happens. And Pemba himself, when he thought back on his original observation, considered this the most likely explanation for why his hot ice cream mix froze first. The school refrigerator almost certainly had frost buildup. There is also the thermostat effect. If you put a very hot container in a freezer, the temperature inside the freezer rises, the thermostat detects this spike and kicks the compressor into a harder cooling cycle. The freezer literally works harder. So, it is not that the hot water freezes faster on its own merits. It is that the hot water recruits the freezer to cool it more aggressively than the cold water did.

[00:18:22]Both of these are real mechanisms, but they are entirely dependent on the specific equipment. A frostfree freezer eliminates the frost explanation. A freezer without a responsive thermostat eliminates the compressor explanation.

[00:18:35]In a carefully controlled laboratory setup where the cooling bath is maintained at a constant temperature, regardless of what you put in it, these mechanisms vanish. So they explain some observations in some kitchens with some freezers. They do not explain the impember effect in general. Now suspect number five, the deep one. This is where the trail goes somewhere nobody expected. Water is strange. You hear people say that all the time, but I want you to understand what I mean. Water is genuinely, profoundly, almost unreasonably strange. Most liquids behave in predictable, boring ways. They get denser as they cool. They freeze at a single sharp temperature. Their properties vary smoothly and continuously with temperature. Water does almost none of this. Water reaches its maximum density at about 4° C, which is why ice floats. If ice sank, every lake and ocean on Earth would freeze solid from the bottom up, and most life would be impossible. Water has an abnormally high boiling point for a molecule its size. Water has an abnormally high heat capacity. Water has an abnormally high surface tension. The reason for all of these anomalies is the same, hydrogen bonds. A water molecule has one oxygen atom and two hydrogen atoms arranged in a Vshape. The oxygen atom is more electrogative. It pulls the shared electrons closer to itself, which makes the oxygen end of the molecule slightly negative and the hydrogen end slightly positive. So the hydrogen end of one water molecule is attracted to the oxygen end of the next. That attraction, that weak but persistent tug between molecules is the hydrogen bond.

[00:20:12]These bonds are about 10 to 20 times weaker than the covealent bonds that hold the atoms together inside each molecule. But they are far stronger than the Vanderwal's forces that hold most other liquids together. And in water, they form a dynamic, constantly shifting network. At any given instant, each water molecule is typically connected to about three or four of its neighbors by hydrogen bonds. These bonds break and reform billions of times per second, but the network as a whole persists. When you heat water, something happens at the molecular level that is subtle and, I think, beautiful. As the temperature rises, the molecules move faster. They jostle more and the hydrogen bonds stretch. The molecules move farther apart on average. But here is the counterintuitive part. As the hydrogen bonds stretch and the molecules separate, the coalent bonds within each individual molecule, the strong O bonds connecting each oxygen to its hydrogens, those bonds actually contract. They get shorter. They tighten. And in tightening, they store energy like a spring being compressed. This is the central insight from a line of research that came out of Nanyang Technological University in Singapore led by a physicist named Changqing Sunsoon around 2013. When you heat water, the covealent O bonds shorten and store energy. When you then cool the water, those bonds relax and release that stored energy.

[00:21:32]That release is equivalent to additional cooling. Extra heat being shed on top of the normal heat loss to the surroundings. Let me make an analogy.

[00:21:41]Imagine two identical rubber bands.

[00:21:44]One is just sitting on a table, relaxed.

[00:21:46]The other one has been stretched way out, held taut, and then allowed to snap back to its resting length. They are now both the same length. But the one that was stretched released energy when it snapped back. It did something. It was not the same as the one that just sat there the whole time. If you measured the length, they look identical. But the stretched one went through a process that the relaxed one did not. And the consequences of that process ripple outward. That is what is happening inside the water at the molecular level.

[00:22:16]The coalent bonds in the heated water were compressed. They are now relaxing and radiating energy as they do. So the water that was heated carries more of this stored energy than the water that was cold. So even when both samples reach the same average temperature, say 25°, the previously hot water has a larger reservoir of releasable energy locked in its covealent bonds. It continues to emit energy at a faster rate. The thermometer says 25° in both glasses, but the molecular reality is different.

[00:22:47]One glass is primed to cool faster. One glass has a loaded spring inside it. In 2017, Yungwin Tawa and colleagues took this further. They used vibrational spectroscopy and computational models to look at how the cluster structure of water changes with heating. Water molecules do not just pair off randomly.

[00:23:07]They form clusters, temporary aggregations of molecules linked by hydrogen bonds. In cold water, these clusters tend to be large and complex.

[00:23:16]They are stable and their structure is nothing like the hexagonal lattice of ice. When freezing begins, these clusters have to be dismantled and reorganized, which takes energy and time. They resist the transition to ice.

[00:23:29]When you heat water, you break up those clusters. The molecules separate. The big stubborn networks are shattered. And when the heated water cools back down, it does not immediately rebuild those large obstructive clusters. It spends some time in a more disordered, more fluid state, a state where the molecules are more free to reorganize into the ice pattern when the freezing temperature arrives. The heated water has been in a sense pre- loosened. This is the idea of hydrogen bond memory. The water remembers that it was hot not in any mystical or magical sense in a physical, measurable, structural sense. The arrangement of its bonds, the size of its clusters, the stored energy in its coalent connections all carry the fingerprint of the thermal history. And that fingerprint affects how the water behaves when it approaches freezing.

[00:24:20]There is a remarkable piece of evidence for this. If you take water at say 35° that has just been sitting around at room temperature and compare it to water at 35° that was heated to 70° and then cooled back down. They cool at different rates. The one that was previously heated cools to 0° about 40% faster than the one that was not. Same starting temperature, different history, different outcome.

[00:24:47]The difference is visible well above zero. It has nothing to do with ice formation. The water itself behaves differently before freezing even enters the picture. And that gets us to the last piece of the puzzle. Maybe the most chaotic, most unpredictable piece, super cooling. Water does not always freeze at 0°. In fact, if the water is pure and the container is smooth and there are no vibrations, water can remain liquid well below zero. This is called super cooling. The water is below its official freezing point, but it has not found a nucleation site, a tiny seed or impur or scratch where the first ice crystal can start to grow. Without that seed, the water just sits there cold and liquid, waiting for something to happen. Think about what that means. You have water at - 10°. Liquid water, colder than the inside of your freezer, perhaps, and still a liquid. It is in a metastable state, like a ball balanced on the top of a hill. It wants to roll down. and it wants to freeze, but it cannot find the trigger. When nucleation finally happens, it can be dramatic. Then the cost to one tiny crystal forms, and then the freezing cascades outward from that point like a wave. The water crystallizes in a burst and the temperature actually jumps up briefly as the latent heat of fusion is released.

[00:26:04]That sudden jump is called recallescence. If you have ever watched a super cooled bottle of water flash freeze the instant you tap it or drop an ice chip in, you have seen this happen.

[00:26:14]One moment it is liquid, the next moment in less than a second the whole thing is a slush of ice crystals. It is startling. The temperature at which nucleation kicks in varies wildly from one trial to the next. The same water in the same container might super cool to -6° one day and -14° the next. It depends on microscopic details that are nearly impossible to control. A speck of dust drifting in from the room. A scratch on the glass wall that was not there yesterday. A vibration from a truck driving past on the street outside. Super cooling is one of the most unpredictable phenomena in all of classical physics. James Brownidge, a radiation safety officer at the State University of New York, spent about a decade of his spare time running hundreds and hundreds of freezing experiments. Patient, meticulous work.

[00:27:05]What he found was that water which started hot tended to super cool to a higher temperature before freezing. In other words, the hot water would drop below zero but then begin to crystallize at say -6°.

[00:27:17]The cold water would drop below zero and keep going, not freezing until it reached -12 or -15°.

[00:27:23]The hot water found its nucleation site sooner. Why? Possibly because the hot water had lost its dissolved gases, which eliminated some of the factors that suppressed nucleation. Possibly because the hot water had smaller, less obstructive molecular clusters, making it easier for ice crystals to form.

[00:27:41]Possibly because the surface properties of the water had changed due to the thermal history. Whatever the exact cause, the result was clear. The hot water spent less time in the super cooled state, which meant it began the actual freezing process sooner. And this, I think, is the real twist in our detective story. The Impember effect is probably not one thing. It is probably five things or six or more. Evaporation reduces the mass. Degassing changes the water's internal behavior. Convection enhances heat transfer and prevents insulating ice lids. Environmental changes improve thermal contact.

[00:28:18]Hydrogen bond relaxation adds an extra cooling mechanism. And reduce super cooling means the freezing process starts sooner. Each one of these is modest on its own. Each one individually might not be enough to explain the full effect, but they all pull in the same direction. And when you stack them up, the combined push can be enough to let the hot water beat the cold water to the finish line. This is actually a very important idea in physics. And it is one that we do not talk about enough. We love clean single explanations. We love it when you can point at one equation and say there that is why, one cause, one effect. Case closed. But nature does not owe us that kind of simplicity.

[00:28:57]Sometimes the real explanation is a conspiracy of causes. Each one contributing its fraction. Each one insufficient by itself. All of them pulling together. And the reason nobody has been able to write down the explanation of the empa effect is that there is no single the there is a coalition. And here is what really drives the professionals crazy. The effect does not happen every time. You can set up two beers hot and cold. Put them in the same freezer under what you think are identical conditions. And sometimes the hot one freezes first.

[00:29:31]Sometimes the cold one wins and sometimes they are nearly tied. The variability is enormous. The result depends on the exact starting temperatures. The container shape, the container material, whether you stirred the water, whether you used tap water or distilled water, whether the freezer had frost, what temperature the freezer was set to, how often you open the door to check the ambient humidity, and probably six other things nobody has thought to measure. In 2016, Henry Buridge at Imperial College London and Paul Lindon at the University of Cambridge ran a very careful study and concluded that the Empa effect was, in their words, not observable in any meaningful way. They argued that once you control for all the variables properly, the effect vanishes into statistical noise. Previous reports of the effect, they said, were marginal at best. But in 2021, John Beckhoffer at Simon Fraser University in Canada described a protocol for reliably reproducing the effect using colloidal particles, tiny glass beads in water rather than pure water. He showed that under specific conditions, a hotter starting state really does reach equilibrium faster. And he could do it repeatedly. And in 2024, a team using a specialized Peltier cooling cell and a thermographic camera found that hot water drops consistently froze faster than cold ones across many trials with the effect becoming more pronounced for larger drops. So we have some people saying it does not exist, some people saying it does, and some people saying it exists sometimes under certain conditions. And the key is figuring out which conditions this is, I must be honest with you, the actual state of the science as of right now. You might find that unsatisfying. You came here wanting an answer, a clean, crisp, case closed answer. And I am telling you, we do not have one. But I want to suggest to you that this is actually the most interesting possible outcome. Think about what is happening here. Water is the most common substance on the surface of the earth. Every human being interacts with it every day. You drink it, bathe in it, cook with it, freeze it, boil it. You have done these things since you were a child. And we still do not fully understand how it freezes. We do not have a complete universally accepted theoretical model of something you do with ice cube trays in your kitchen. That is not embarrassing. That is thrilling. That is nature reminding us that it is still smarter than we are.

[00:32:03]You know what I think the real lesson of the effect is? It is not about water. It is about temperature. Or rather, it is about what temperature cannot tell you.

[00:32:13]Temperature is a number. A single number that describes the average kinetic energy of a collection of molecules.

[00:32:21]And we treat it as if it tells us everything we need to know about a systems thermal state. Two glasses of water, both at 25°, same temperature. Therefore, in our minds, same water, but temperature is an average. And averages are dangerous.

[00:32:39]They smooth over differences. They hide structure. They make complicated things look simple. Say you have two towns and both towns have an average income of $50,000 a year. In one town, every family makes 50,000. In the other, half the families make a h 100,000 and the other half make nothing. same average, completely different reality. You would not want to confuse those two towns. You would not want to make policy based on the average alone. Or think about it this way. Two orchestras both play at the same average volume, but one is playing a steady moderate drone. The other is alternating between crashing forimos and whispering pianisimos. Same average loudness, completely different music, completely different experience.

[00:33:27]If all you measured was the average, you would think they were the same. They are not. Water at 25° that was recently at 90° has a different internal structure, different cluster sizes, different dissolved gas content, different convection memory, and different covealent bond energies than water that has been sitting at 25° for an hour. The thermometer says they are the same. The molecules say they are not. Temperature, that single number we rely on so heavily, is not the whole story. It never was. We just assumed it was because nobody bothered to check. And that brings us back to Erasto Impemba.

[00:34:02]When he stood up in that classroom and asked his question, his teacher told him it was Impemba's physics and not the universal physics. The classmates laughed. The joke followed him from school to school. But Impea was right.

[00:34:15]Not because he had the answer. He did not have the answer. Nobody has the complete answer even now. But he was right because he had observed something real and he refused to let the authority of a textbook talk him out of what his own eyes and his own frozen ice cream had shown him. That is the first rule of doing science and it is a rule that practicing scientists sometimes forget.

[00:34:37]It does not matter how elegant your theory is. It does not matter how many equations support it. It does not matter how many professors agree with it. If your theory says something cannot happen and then you watch it happen, your theory is wrong. Not the observation, the theory. At least the theory is incomplete. It is missing something. And the proper response is not to dismiss the observation. It is to fix the theory. Newton's law of cooling is correct. Thermodynamics is correct.

[00:35:05]The first law is not violated. No energy appears from nowhere. No heat vanishes without accounting for it. But the assumptions we made about what happens inside the water, about whether two samples at the same temperature are truly in the same physical state, those assumptions were incomplete. And a 13-year-old boy in East Africa making ice cream in a school kitchen noticed the incompleteness that generations of professional physicists had walked right past. There is something I have always believed, and this story confirms it for me. The simple things are not simple.

[00:35:40]The things you think you understand so well that you never bother to look at them carefully. Those are precisely the things hiding the deepest surprises. A ball rolling down a ramp, light bouncing off a mirror, water freezing in a glass.

[00:35:53]We call these elementary. We cover them in the first week of a course and then move on to the fancy stuff, the quarks and the black holes and the quantum fields. But the physics inside a glass of freezing water is as rich and as surprising as anything in particle physics. Maybe more so because it is right there, right in front of you every single day and you walk past it without a second thought. Philip Ball, a science writer who reviewed the field for physics world in 2006, asked a question that I think is exactly the right question. He said, "Even if the emper effect is real, if hot water can sometimes freeze more quickly than cold, it is not clear whether the explanation would be trivial or illuminating." That is perfect because it captures the two possibilities.

[00:36:36]Either the effect is a boring artifact of messy experimental conditions, in which case it teaches us nothing, or it is a window into something deep about the nature of water, about hydrogen bonds, about what temperature really means, about the internal states of matter that our macroscopic instruments cannot see. I am betting on the second possibility, and so apparently are the physicists who keep studying it. In 2012, the Royal Society of Chemistry in Britain got so frustrated by the lack of a definitive answer that they held a public competition. They offered a,000 prize to whoever could provide the best explanation of the Empa effect. 22,000 people submitted entries. 22,000 for a question about freezing water. That is how deep this rabbit hole goes. The winner was a chemist named Nicola Brigovich from the University of Zagreb.

[00:37:25]And even he did not claim to have solved the problem. He was awarded the prize for approaching it carefully and honestly, summarizing the known mechanisms and demonstrating through experiments that convection and super cooling were essential ingredients.

[00:37:39]Arasto Impa himself, by then retired from a career as a wildlife officer in Tanzania, announced the winner at the ceremony. Imagine that, a boy who was mocked for his question, coming back 60 years later to judge the best scientific answer to it. In recent years, researchers have found empa-like effects not just in water, but in granular fluids, in magnetic alloys, in nanomechanical systems, and even in quantum systems. A team at Trinity College Dublin studying quantum thermodynamics demonstrated that in abstract computational models of spin systems, initially hotter states can reach thermal equilibrium faster than initially cooler ones. They found that certain initial conditions lead to a simultaneous increase in the thermalization rate and the free energy.

[00:38:22]The effect seems to transcend water. It may be telling us something general about how systems with complex internal structures respond to sudden changes in temperature. There is even an inverse empa effect. Cold systems can sometimes heat up faster than warmer ones. A team at Simon Frasier University demonstrated this experimentally using colloidal particles confined by laser tweezers.

[00:38:45]The symmetry is haunting. Whatever mechanism is at work, it operates in both directions. If all of this is true, then the empermber effect is not a curiosity. It is a clue. A clue to a deeper understanding of non-equilibrium thermodynamics. The physics of systems that are changing, evolving, relaxing toward equilibrium through pathways that depend on where they started. Most of thermodynamics deals with equilibrium with systems that have already settled down. Non-equilibrium thermodynamics is harder, messier, less understood, and far more important for the real world, which is never actually in equilibrium.

[00:39:20]So the question of why hot water sometimes freezes faster than cold water is, as far as we can tell, not one question. It is half a dozen questions tangled together. Each one touching a different corner of physics.

[00:39:33]Thermodynamics, fluid dynamics, molecular chemistry, surface science, nucleation theory, statistical mechanics. All of them meet in your freezer. And none of them alone can explain what happens when you put a hot glass next to a cold glass and watch.

[00:39:49]That is not a failure. That is an invitation. It means there is still work to do. It means the universe has not been fully figured out. Not even the parts you can find in your own kitchen.

[00:39:58]And it means the next person who notices something strange, something that should not happen according to the textbook, something their teacher says is impossible. Might be the one who finds the missing piece. It might be a school boy in Africa making ice cream with boiled milk because he could not get a spot in the freezer any other way. It might be a professor who takes the question seriously instead of laughing at it. It might be someone who looks at a glass of water and sees not a simple boring substance but a mystery wrapped in hydrogen bonds. Aristotle wondered about it. Daycart wondered about it.

[00:40:31]Impecca wondered about it. Physicists in Singapore and London and Canada and Spain are still wondering about it. If you ask me what the definitive answer is, I will tell you the truth plainly and without embarrassment. We do not fully know. We have good suspects. We have partial explanations. We have five or six mechanisms that each do part of the job. But the full picture is not in yet. And that, if you sit with it for a moment, is the most exciting sentence a physicist can say because it means the story is not over. It means there is room for you in it. I want to leave you with one more thought. We tend to think of the great discoveries in physics as dramatic revelations. Einstein sitting in his office imagining himself riding a beam of light. Rutherford firing particles at gold foil and discovering the nucleus. Newton watching an apple fall, though that story is probably more legend than history. We tell these stories as if discovery is a moment, uh, a flash, an epiphany. But the impember effect is a different kind of discovery.

[00:41:34]It is the kind that says, "Wait, we do not even understand the basics as well as we thought." It is not a flash of new knowledge. It is a slow recognition of old ignorance. And I think that kind of discovery is in many ways more important because it keeps us humble. It keeps us looking at the ordinary world with fresh eyes. It keeps us asking questions about things that everybody else takes for granted. So here's what I want you to do. Go to your kitchen. Go to your kitchen. Take two identical cups. Fill one with water as hot as your tap will go. Fill the other with cold water. Put them both in the freezer on the same shelf. Check on them every 30 minutes.

[00:42:14]Write down what you see. Does the hot water win? Does the cold water win? Is it close? And if the hot water wins, ask yourself, which of those five suspects do you think is most responsible? And what experiment would you design to find