The Mpemba Effect Why hot water sometimes freezes faster than cold — and what this puzzle reveals about the nature of scientific knowledge he Mpemba effect — the observation that, under certain conditions, hot water freezes faster than cold water — is one of the most counterintuitive phenomena in everyday physics and one of the most contentious. Named after the Tanzanian student Erasto Mpemba, who in the 1960s made the observation while freezing ice cream and brought it to the attention of the British physicist Denis Osborne, it has been debated, disputed, confirmed, and reinterpreted by researchers across several decades without reaching a settled consensus. Whether the effect is real, under what conditions it occurs, and what mechanism explains it remain questions whose answers continue to be revised. Mpemba was a student at the Mkwawa Secondary School in Iringa when he noticed that his hot ice cream mixture froze more quickly than his classmates' cold mixture in the school freezer. His teachers were dismissive when he reported the observation. However, during a visit to the school, Osborne agreed to test the claim and confirmed that Mpemba's observation was reproducible. The two published their findings together in a 1969 paper that gave the phenomenon its name, though both acknowledged that a definitive explanation remained elusive. The paper attracted wide attention and inspired a generation of experimental and theoretical investigations. The Mpemba effect, if it is real, would be surprising on basic thermodynamic grounds. A body at a higher temperature has more thermal energy than one at a lower temperature and must lose more energy to reach any given lower temperature, including the freezing point. The rate of heat loss, according to Newton's Law of Cooling, is proportional to the temperature difference between the body and its surroundings: a hotter object cools faster initially, but whether it can 'catch up' with and overtake a cooler object depends on the specific conditions. Under idealised conditions with identical containers and cooling environments, the hotter sample should take longer to freeze — which is precisely why the Mpemba effect, when it occurs, requires explanation. Several mechanisms have been proposed to explain the effect. The most straightforward involves evaporation: as a hot water sample cools, it loses water through evaporation from its surface, reducing its mass. A smaller mass of water requires less energy to freeze, potentially allowing the initially hotter sample — now
slightly lighter — to freeze before the initially cooler one. This explanation is quantitatively plausible for some observed effects but may not account for them all. Other proposed mechanisms include differences in dissolved gases — hot water contains less dissolved oxygen and carbon dioxide, which affect its thermal properties — and convection currents created by temperature gradients within the cooling sample that may affect the rate of heat loss. A 2016 study published in the Journal of Chemical Physics proposed a molecular-level explanation based on the specific structure of hydrogen bonds in water. The researchers proposed that the energy stored in the stretching of hydrogen bonds — which are more extended in hot water — is released as the water cools, contributing energy to the cooling process and potentially accelerating freezing. This mechanism was controversial and prompted responses from researchers who disputed both its theoretical basis and its experimental support. The debate illustrated a general challenge in this field: the Mpemba effect, when observed, is highly sensitive to the specific conditions of the experiment — the shape of the container, the method of cooling, the presence of impurities — making reproducibility across different laboratories difficult. A rigorous meta-analysis of Mpemba effect experiments published in 2016 reached a sobering conclusion: across a large number of experimental studies, there was no consistent evidence for the effect under controlled conditions. The analysis found that many positive results could be attributed to experimental artefacts or reporting bias — the tendency to publish surprising positive findings while leaving negative or null results unpublished. This does not mean the Mpemba effect never occurs: in specific experimental conditions it has been reliably reproduced. It does suggest, however, that it is not a universal property of water under cooling but a phenomenon that depends critically on the specific details of the experimental setup. The Mpemba effect remains an instructive example of a scientific controversy in which a simple-sounding question has proved remarkably resistant to definitive resolution.