A pressurized heavy-water reactor (PHWR) is a nuclear reactor that uses heavy water (deuterium oxide D₂O) as its coolant and neutron moderator. PHWRs frequently use natural uranium as fuel, but sometimes also use very low enriched uranium. The heavy water coolant is kept under pressure to avoid boiling, allowing it to reach higher temperature (mostly) without forming steam bubbles, exactly as for a pressurized water reactor (PWR). While heavy water is very expensive to isolate from ordinary water (often referred to as light water in contrast to heavy water), its low absorption of neutrons greatly increases the neutron economy of the reactor, avoiding the need for enriched fuel. The high cost of the heavy water is offset by the lowered cost of using natural uranium and/or alternative fuel cycles. As of 2025, 43 PHWRs were in operation, having a total capacity of 23.430 GW(e), representing roughly 11% by number and 6.5% by generating capacity of all current operating reactors. CANDU and IPHWR are the most common type of reactors in the PHWR family. Other designs include the German design PHWR KWU installed at Atucha Nuclear Power Plant in Argentina.

The key to maintaining a nuclear chain reaction within a nuclear reactor is to use, on average, exactly one of the neutrons released from each nuclear fission event to stimulate another nuclear fission event (in another fissionable nucleus). With careful design of the reactor's geometry, and careful control of the substances present so as to influence the reactivity, a self-sustaining chain reaction or "criticality" can be achieved and maintained.

Natural uranium consists of a mixture of various isotopes, primarily ²³⁸U and a much smaller amount (about 0.72% by weight) of ²³⁵U.

²³⁸U can only be fissioned by neutrons that are relatively energetic, about 1 MeV or above. No amount of ²³⁸U can be made "critical" since it will tend to parasitically absorb more neutrons than it releases by the fission process. ²³⁵U, on the other hand, can support a self-sustained chain reaction, but due to the low natural abundance of ²³⁵U, natural uranium cannot achieve criticality by itself.

In a heavy water reactor, the trick to achieving criticality using only natural or low enriched uranium, for which there is no "bare" critical mass, is to slow down the emitted neutrons (without absorbing them) to the point where enough of them may cause further nuclear fission in the small amount of ²³⁵U which is available. (²³⁸U which is the bulk of natural uranium is also fissionable with fast neutrons.) This requires the use of a neutron moderator, which absorbs virtually all of the neutrons' kinetic energy, slowing them down to the point that they reach thermal equilibrium with surrounding material. It has been found beneficial to the neutron economy to physically separate the neutron energy moderation process from the uranium fuel itself, as ²³⁸U has a high probability of absorbing neutrons with intermediate kinetic energy levels, a reaction known as "resonance" absorption. This is a fundamental reason for designing reactors with separate solid fuel segments, surrounded by the moderator, rather than any geometry that would give a homogeneous mix of fuel and moderator.

In a light water reactor, water makes an excellent moderator; the ordinary hydrogen or protium atoms in the water molecules are very close in mass to a single neutron, and so their collisions result in a very efficient transfer of momentum, similar conceptually to the collision of two billiard balls. However, as well as being a good moderator, ordinary water is also quite effective at absorbing neutrons. And so using ordinary water as a moderator will easily absorb so many neutrons that too few are left to sustain a chain reaction with the small isolated ²³⁵U nuclei in the fuel, thus precluding criticality in natural uranium. Because of this, a light-water reactor will require that the ²³⁵U isotope be concentrated in its uranium fuel, as enriched uranium, generally between 3% and 5% ²³⁵U by weight (the by-product from this process enrichment process is known as depleted uranium, and so consisting mainly of ²³⁸U, chemically pure). The degree of enrichment needed to achieve criticality with a light-water moderator depends on the exact geometry and other design parameters of the reactor.

One complication of this light water approach is the need for uranium enrichment facilities, which are generally expensive to build and operate. They also present a nuclear proliferation concern; the same systems used to enrich the ²³⁵U can also be used to produce much more "pure" weapons-grade material (90% or more ²³⁵U), suitable for producing a nuclear weapon. This is not a trivial exercise by any means, but feasible enough that enrichment facilities present a significant nuclear proliferation risk.

The solution to this problem, in a PHWR (pressurized heavy water reactor), is to use a moderator that does not absorb neutrons as readily as water. In this case potentially all of the neutrons being released can be moderated and used in reactions with the ²³⁵U, in which case there is enough ²³⁵U in natural uranium to sustain criticality. One such moderator is heavy water, or deuterium-oxide. Although it reacts dynamically with the neutrons in a fashion similar to light water (albeit with less energy transfer on average, given that heavy hydrogen, or deuterium, is about twice the mass of hydrogen), it already has the extra neutron that light water would normally tend to absorb.