On Saturday, June 1, 2002, the Ontario Heritage Foundation unveiled a provincial plaque commemorating the Nuclear Power Demonstration Reactor at Rolphton, near Chalk River. The bilingual provincial plaque reads as follows:



On June 4, 1962 the Nuclear Power Demonstration (NPD) Reactor 3 km east of Rolphton supplied the Ontario power grid with the first nuclear-generated electricity in Canada. A joint project of Atomic Energy of Canada Limited, Ontario Hydro and Canadian General Electric, NPD was the prototype and proving ground for research and development that led to commercial application of the CANDU system for generating electric power from a nuclear plant using natural uranium fuel, heavy water moderator and coolant in a pressure tube configuration with on-power refuelling. As a science and engineering research centre, NPD produced internationally significant knowledge and techniques. It was also a training centre for nuclear plant operators. NPD closed in 1987 after exceeding its operational goals.



Le 4 juin 1962, le réacteur nucléaire expérimental NPD (Nuclear Power Demonstration) situé à trois kilomètres à l'est de Rolphton, produit, pour la première fois au Canada, de l’électricité nucléaire. Construit par Énergie atomique du Canada Limitée, Ontario Hydro et Générale électrique du Canada, il est le prototype qui inspire la conception de son application commerciale, le système CANDU, réacteur à tubes de force qui utilise l’uranium naturel comme combustible, est ralenti et refroidi à l’eau lourde, et qui est ravitaillé en cours de fonctionnement. Le réacteur NPD concentre la recherche dans ce domaine et suscite des connaissances et des techniques capitales reconnues dans le monde entier. Après avoir été un centre de formation pour les techniciens de centrales nucléaires, il ferme en 1987, ayant dépassé ses objectifs opérationnels.



Historical Background:

On June 4, 1962, the Nuclear Power Demonstration (NPD) Reactor at Rolphton near Chalk River began to feed power into the Ontario power grid.  It was Canada’s first delivery of commercial nuclear electric power, the culmination of years of intensive research, and forerunner of the much larger, commercial-scale CANDU system.

The NPD reactor was never intended to be a commercial-scale generating station.  Instead, it was a working reactor test facility intended to demonstrate “the feasibility of producing electric power from a nuclear power plant using heavy water as the moderator and coolant and natural uranium as the fuel.”[1]


The convergence of an emerging technology and the need for more electric power

In the mid 20th century, Ontario was generating most of its electric power from water power.  However, by the early 1950s Ontario needed more electric power than its fabled “white coal” could provide.  It had two options: bring in fossil fuels to produce steam to drive turbines, or experiment with an emerging technology: nuclear power. 

            In the late 19th century, the discovery of radiation had led to the use of X-rays in research, industry and medicine.  Albert Einstein’s theoretical work suggested that a small amount of matter could be converted into a large amount of energy and by 1939 the atom had been split.  When researchers learned how to bombard uranium with neutrons to release one or more high-velocity neutrons, heat, and gamma rays the age of nuclear fission arrived.

            Scientists realized that if an emitted neutron split another uranium atom and its emitted neutrons went on to split more atoms, a chain reaction would start that would continue on its own.  The question was how this chain reaction could best be used.  One use was the atomic bomb, which was essentially a massive uncontrolled release of energy.  But could the energy released from nuclear reaction be harnessed for productive, rather than destructive, energy?

Although Canada had contributed to the development of the atomic bomb, Canadian scientists, engineers and politicians were committed to finding peaceful uses of nuclear energy.  Canadian experience with reactors such as ZEEP, NRU and NRX made Canada a leader in isotope production for medical and industrial uses.  Canada’s bomb was the so-called “cobalt bomb,” a radiotherapy unit used to treat cancer.

            Generating electric power from the heat of a controlled nuclear reaction was the next logical step in Canada’s progress as a major player in the field of peaceful nuclear energy.    


A three-way partnership


In 1955, after a year establishing the framework for the Canadian nuclear electric program, a three-way partnership was established to develop a Nuclear Power Demonstration (NPD) reactor.  The first partner, Atomic Energy of Canada Limited (AECL), would provide research data and pay for the nuclear portion of the power plant.  The second, Ontario Hydro, would design and pay for the actual building as well as the electric generating equipment, and then operate the plant when completed. The third, contractor Canadian General Electric of Peterborough, Ontario, would design and construct the Nuclear Steam Supply System (NSSS), which included the reactor itself.[2] Each partner stood to gain: Ontario Hydro needed the electricity, AECL wanted to demonstrate its research skills and the utility of nuclear research, and CGE hoped to gain a foothold in an emerging industry.[3] 

The next decision was the source of fuel. Canada had a good source of natural uranium in the Elliot Lake mines in Northern Ontario.  It was relatively inexpensive, but had one disadvantage.

Naturally occurring uranium is a mixture of two different forms or isotopes, U235 and U238.  U235 is useful in reactors because it is fissile – when it is hit by a neutron of any energy, it may split and release both energy and high-energy neutrons.  If the conditions are right, the released neutrons will split other U235 atoms, leading to a chain reaction, which can continue as a self-sustaining reaction as long as a fresh supply of U235 fuel is added.  Unfortunately, it is in short supply, only 0.7% of the uranium ore by weight.

Enriched uranium fuel contains more fissile U235 than naturally occurring uranium.  However, Canada did not have a plant to produce enriched uranium fuel, and had no nuclear weapons program that would justify the high cost of setting up an enrichment plant.  Canada needed a different approach.

The neutron is a type of projectile, but one that has more success in splitting its fissile target the slower it goes, not the faster.  Therefore, slowing neutrons down or “moderating” them increases the likelihood of getting a sustained chain reaction. 

The choice of moderator is important.  Some moderators slow the neutrons down but capture too many neutrons in the process.  Ordinary water (H2O) is a good neutron moderator, but it tends to capture too many neutrons.  But, like uranium, hydrogen also exists in different isotopes.  Hydrogen that contains an extra neutron in its nucleus is called deuterium and is heavier than conventional hydrogen.  Heavy water, D2O, is a better neutron moderator because it does not capture as many neutrons.  With the ZEEP and NRX reactors, Canada already had successful experience with a heavy water–moderated reactor.

Deuterium is scarce.  About 1 in 50,000,000 atomsmolecules of water are heavy water, and it is expensive to separate heavy from light water.  However, after initial capital costs of supplying the reactor with heavy water moderator and coolant, the continuing operating costs of the reactor are lower than with light water.

The heavy water had a secondary use: as a coolant to draw off heat and keep the reactor core at a reasonable temperature, after which it would travel to a heat exchange unit where it would heat conventional water to produce steam to drive a conventional turbine generator and produce electricity.

The project began with these assumptions: natural rather than enriched uranium fuel and heavy rather than light water moderator to control the rate of reaction and the amount of heat produced.  Now it was time to make things work.


A sudden change of plan


At that time, conventional nuclear technology required that the reactor core, fuel, moderator, and coolant be housed in a pressure vessel, a thick-walled cylindrical tank, which had to be very strong to deal with high temperatures and pressures.  The NPD’s vessel was being built in Scotland, because no one in Canada had the necessary expertise.

Ontario was in a hurry to start drawing on nuclear electric power, and planning had begun for Douglas Point, the first of the next generation of nuclear electric stations that would have ten times the 20 megawatts output of NPD.  But the Douglas Point engineers discovered that if the heavy water–moderated NPD reactor design were to be ten times as large, the pressure vessel would be too big, too expensive and possibly unsafe.

This was a crisis.  Research had shown the NPD reactor, which was intended to be an important development site for future reactors, risked being a technological dead-end.[4]  

Fortunately NPD was in the hands of people who had the courage to pay attention to research, call a halt to further work, and try a new approach.  This was not an easy decision to make.  It meant writing off a lot of expensive work, including the pressure vessel being built in Scotland, and striking out in a new, untried direction. 

In its 1957-58 Annual Report, AECL recounted:



In early 1957 it was decided to suspend work on the design and construction of the NPD reactor in order to incorporate certain new features which appeared very promising.  One of these features was the use of horizontal pressure tubes for the containment of the pressure system including the fuel elements, in place of a single pressure vessel which had previously been specified.  This substitution of pressure tubes for a pressure vessel removes one of the major problems in designing and manufacturing a nuclear power station of a much larger capacity.  A second feature was a new method of fuel loading and unloading, which would be a decided improvement in that a greater amount of energy could be obtained from the uranium fuel.[5]


The design changes allowed NPD to remain on track to fulfil its research, development and training mandate.  Construction re-started in early 1958.


Improved metallurgy to the rescue

One of the developments that made the new pressure system workable was a new metal alloy.  The inside of a nuclear reactor is a harsh environment for materials.  The pressure tubes and fuel rods must withstand high temperature and pressures.  Any alloy or metal used to make pressure tubes or sheath uranium oxide pellets must be physically strong and resist corrosion from the heavy water coolant which flows through the pressure tubes then over and around the fuel rods and bundles.  Moreover, the material chosen must not interfere with the nuclear activity, and the movement of either high-energy or moderated neutrons.

            Fortunately, at the time that scientists and engineers were searching for a way to solve the pressure vessel problem, new alloys were being developed.  Zirconium allowed free passage of neutrons, but it lacked the required strength and corrosion resistance.  By 1957 the addition of alloying agents, particularly niobium, led to the creation of Zircaloy.  Canadian research for the United States turned out to be critical.  “It was possible to envision the zirconium-alloy pressure tubes only because of the pioneering work undertaken [by AECL] for the United States’ naval reactor” for the submarine Nautilus.[6] 


The CANDU system

The redesigned NPD reactor was the prototype for what AECL dubbed “CANDU (Canadian Deuterium-Uranium)”. [7]  This internationally recognized “technology which Canada has pioneered” is characterized by the fact that “the reactor uses heavy water as the moderator and coolant and natural uranium oxide as the fuel.”[8]  More than a prototype, the NPD Reactor served for many years as a research and development facility for ideas and technology used in various CANDU system reactors in Canada and elsewhere.

A 1978 review article in Science noted, “To a physicist, the essence of a CANDU reactor is its use of heavy water as the moderator…. To an engineer, pressure tubes are the major distinguishing feature of a CANDU reactor.”[9]

            “The CANDU moderator, essentially unpressurized and cool, is contained in a reactor vessel (calandria) through which pass some hundreds of identical pressure tubes containing the fuel and coolant.  Each tube, made of a zirconium alloy, is about 6 meters long and 10 centimetres in diameter and has a wall 4 millimetres thick.”

            “The pressure-tube design facilitates on-power fueling, achieved by mobile machines which connect to opposite ends of the pressure tube to be fuelled….On-power fueling minimizes the inventory of neutron-absorbing fission products.  Heavy water, on-power fueling, and constant attention to neutron economy at all stages of design enable CANDU reactors to enjoy the economics of natural uranium fuel….”[10]

            The benefits of the system included:

        the low cost of natural uranium;

        a high net capacity because on-power fueling meant that there was no need to shut the reactor down for fuel changing;

        higher safety, because a leaking fuel assembly could be removed before it released significant radioactivity to the coolant and because the separate moderator constituted a large heat sink capable of absorbing any energy that might be released in the case of an accident;

        a simple design that could be manufactured in Canada.


The Achievements of NPD

By 1987 NPD could look back on 25 years of successful service.  It had far surpassed all of its original goals but by then there were other sources of research information, other training centers, and maintenance costs were rising.  NPD had outlived its economically useful life span and it was decided to shut it down.  Radioactive materials and components were removed, and some non-nuclear equipment was relocated or sold.  The project left no significant architectural remains.[11]  The site is now monitored remotely and inspected regularly. 

A demonstration project, NPD had contributed to knowledge about nuclear power and tested a specific design for a nuclear reactor.  It was not intended as a commercial venture.[12]  In particular, it demonstrated the reliability of the system and the economics of the fuel cycle; provided training for operators, and served as a prototype for a large nuclear generating station of the same type.  It was exhaustively tested and fine-tuned and the resulting modifications helped design the Douglas Point Station.

One of its most important features was “the concept of on-power fueling, which had not previously been attempted for a water-cooled [commercial power] reactor.”[13]  That kind of testing would only have been possible in a demonstration project where research, new knowledge, techniques and understanding were valued above all else for they were the key to the ultimate goal: viable commercial nuclear electric plants.

            NPD and CANDU reactors were known for good fuel performance.  This is particularly noteworthy, considering that natural uranium is a challenging fuel to work with.  The design of the fuel bundles of Zircaloy tubes containing uranium dioxide pellets was noted for its simplicity, an important engineering goal, because it contributes to reliability.  When combined with on-line, near-continuous refueling, the amount of energy produced per unit of fuel could be accurately measured.  Many other nuclear reactor designs did not include this feature and it was harder to gauge microlevel rather than macrolevel performance in these reactors.[14]

            The choice of expensive heavy water allowed the use of cheaper fuel and added engineering challenges which yielded important results.  Because heavy water is expensive to manufacture, standard equipment such as valves had to be redesigned to prevent leaks.  The modifications were so successful that they actually lowered maintenance requirements.[15]

            Another feature of the CANDU design was the fact that heavy water coolant was kept separate from heavy water moderator.  In some reactors, coolant and moderator are not separated.  However, coolant and moderator operate under different conditions and have different needs in terms of things such as optimal pH and additives to enhance performance or reduce damage to parts they come in contact with.  Separation of the coolant and moderator allowed more research to be carried on which allowed each to perform better than if moderator and coolant had been mixed.

            Finally, because much of the equipment could be manufactured in Canada, unlike the originally proposed pressure vessel, the CANDU reactor contributed to the country’s industrial capability.



The NPD Reactor (1962-1987) occupies a significant place in Ontario, Canadian, and international history.  It was the proving ground for research and development that led to the commercial application of the CANDU system.  It allowed engineers and scientists to discover solutions to problems that were applicable far beyond the borders of Canada. The NPD Reactor is an example of how Canadian engineers often provide elegant engineering solutions while tackling difficult problems in harsh environments.  Creating a pressure tube reactor using natural uranium fuel and heavy water moderator and coolant was not the easiest way to approach the need for generating nuclear electric power, but it was the way that led to a reliable, safe, economical nuclear electric generating system: CANDU.



For their patient help, interest and guidance I would like to thank the following nuclear scientists and engineers: Don Charlesworth, Tony Colenbrander, Lorne McConnell, Bob Morrow and Jeremy Whitlock.  I am also indebted to my research assistant Wendy Stocker and editor Philippa Campsie.


Prepared by Dr. Norman R. Ball, Director

Centre for Society, Technology and Values

Systems Design Engineering
University of Waterloo

[1] Atomic Energy of Canada Limited, Annual Report 1957-1958, p. 7

[2] Since 1955, the names of some of the partners involved in this project have changed.  Ontario Hydro is currently known as Ontario Power Generation and Canadian General Electric as GE Canada.

[3] Atomic Energy of Canada Limited, Canada Enters the Nuclear Age: A Technical History of Atomic Energy of Canada Limited. Montreal and Kingston: McGill-Queen’s University Press for Atomic Energy of Canada Limited, 1997, p. 395; Atomic Energy of Canada Limited, Annual Report 1957-1958, p. 5; Atomic Energy of Canada Limited, Annual Report 1959-1960, p. 7.

[4] Canada Enters the Nuclear Age, p. 17.

[5] Annual Report 1957-1958, pp. 5, 7. 

[6] Canada Enters the Nuclear Age, p. 235

[7] Annual Report 1957-1958, p. 7.

[8] Annual Report 1957-1958, p. 7.

[9] Robertson, “An Appropriate Technology,” pp. 657-658.

[10] Robertson, “An Appropriate Technology,” p. 658.

[11] After the site was decommissioned buildings used for functions such as training and administration were torn down. They were described by those who worked in them as typical nondescript industrial park type buildings. The concrete reactor building was noted for one innovation, namely the use of heavy concrete which used an iron aggregate from the Bancroft area. The iron aggregate was an effective radiation blocker which allowed walls to be thinner than they would have been with less specialized aggregates.

[12] Annual Report 1957-1958, p. 7.

[13] Robertson, “An Appropriate Technology,” p. 658

[14] Robertson, “An Appropriate Technology,” p. 659.

[15] Robertson, “An Appropriate Technology,” p. 660. For additional comments on the need to spend money wisely see Louis H. Roddis, Jr., “Nuclear power in the world today.” IEEE Spectrum, July 1965, pp, 90, 95-97 and in particular his comment “There is no magic in costs. The law of conservation of energy applies equally to the use of money. You never get more than you pay for” (p. 97) to which he might have added you sometimes get less if you are not careful.