WO1997041565A1

WO1997041565A1 - Low coolant void reactivity fuel bundle

Info

Publication number: WO1997041565A1
Application number: PCT/CA1997/000279
Authority: WO
Inventors: Ardeshir R. Dastur; David B. Buss
Original assignee: Atomic Energy Of Canada Limited
Priority date: 1996-04-25
Filing date: 1997-04-23
Publication date: 1997-11-06
Also published as: CA2174983A1; AU2564297A

Abstract

A fuel bundle design for a heavy water moderated and cooled reactor and in particular, to a fuel bundle that provides control of the reactivity upon coolant voiding in a reactor such as a CANDU type nuclear reactor. The geometry of the fuel bundle reduces the coolant mean chord length. A centrally disposed neutron scattering material is surrounded by an annular region in which fuel pins are arranged. The scattering material can be amorphous carbon or reticulated silicon carbide. The fuel pins are arranged in concentric rings having equal numbers of pins with the pins in inner rings being smaller in diameter than, and nestled between adjacent pins in the adjacent outer ring.

Description

LOW COOLANT VOID REACTIVITY FUEL BUNDLE

HELD OF THE INVENTION

This invention relates to a fuel bundle design for a heavy water moderated and cooled reactor and in particular, to a fuel bundle that provides control of the reactivity upon coolant voiding in a reactor such as a CANDU type nuclear reactor.

BACKGROUND OF THE INVENTION

The reactor core of a heavy water moderated and cooled nuclear reactor such as a CANDU type reactor contains a plurality of fuel channels each of which contains a plurality of fuel bundles placed end-to-end inside a pressure tube. Each fuel bundle contains a set of fuel pins of fissionable material. When the reactor is in operation, heavy water coolant flows over the fuel bundles to cool the fuel and remove heat from the fission process. The coolant also acts to transfer the heat to a steam generator that drives a turbine to produce electrical energy. Because the coolant circuit is pressurized, the coolant does not boil significantly and the density, temperature and volume of the coolant in the fuel channels remain generally constant with time.

When there is a breach in the coolant circuit such as in a loss of coolant accident, the coolant starts to boil and coolant voiding occurs. In a heavy water cooled reactor, the rate of neutron multiplication increases rapidly when the coolant in the fuel channel is lost. This phenomenon is due to positive coolant void reactivity and is an undesirable occurrence. Reactivity is a measure of the ability of a reactor to multiply neutrons. Positive coolant void reactivity is the increase in reactivity due to voiding of the coolant.

The problem of positive coolant void reactivity is especially relevant in a major loss of coolant accident which results in a high rate of discharge of coolant from the reactor core and consequently a high rate of increase in the reactor power level. In anticipation of such a situation, it has been necessary to provide a shutdown system to detect abnormal situations in the reactor and respond by initiating steps to arrest the increased rate of neutron multiplication without damaging the fuel. The importance of such a shutdown system is accentuated by the fact that the coolant is no longer available to cool the fuel. Heavy water reactors such as CANDU type reactors have been equipped with two equally effective and totally redundant shutdown systems.

Although shutdown systems are an accepted means of ensuring safe operation of the reactor, an uncontrolled increase in the rate of neutron multiplication is an inherently unstable behaviour of the reactor. Furthermore, licensing authorities are becoming increasingly concerned about the potential problem of coolant void reactivity and the safety concerns which it generates. It has been suggested by some licensing authorities that the peak value of the positive coolant void reactivity be limited to the prompt critical reactivity which is the reactivity at which the cycle time of neutron multiplication starts to drop rapidly.

One possible means to reduce the value of the positive coolant void reactivity is to configure a fuel bundle specifically intended to minimize void reactivity on loss of coolant. A number of approaches have been considered in the prior art.

The first approach is to add burnable poisons to the fuel bundle. An example of this approach was disclosed in Canadian patent application No. 2,097,412 of the inventor Ardeshir R. Dastur filed May 31, 1993 and published December 1, 1994. In the fuel bundle described in the application, the creation of negative reactivity is achieved by the judicious distribution of burnable neutron poisons and of fissile material amongst the fuel pins. The design uses enriched uranium for producing an increase in neutron multiplication in the outer region of the fuel bundle (where the thermal neutron flux tends to decrease upon a decrease in coolant density) and depleted uranium in a central region of the fuel bundle (where the thermal neutron flux tends to increase upon a decrease in coolant density). Neutron absorbing material is mixed with the depleted uranium to absorb thermal neutrons in the central region. The rate of neutron absorption in the mixture of the fertile material and the absorber in the central region increases as the mixture redistributes the neutron flux across the bundle on coolant voiding and thereby produces negative reactivity when the coolant density decreases. One of the principal disadvantages of this approach is the requirement for the use of enriched fuel as opposed to natural uranium. Natural uranium fuel is the predominant fuel of CANDU type reactors. The increased need for U₂₃₅ enriched fuel represents a significant reduction in the resource utilisation advantage that CANDU type reactors have over light water reactors. In addition, U₂₃₅ fuel is not as accessible as natural uranium in many countries of the world and in particular, in many of the countries where CANDU reactors are currently used. This reduced accessibility is a disadvantage in and of itself and also results in significantly increased fuelling costs.

Another approach has been to replace the fuel material in the central pins with graphite. The primary disadvantage of this approach is that the reduction in positive coolant void reactivity is limited. In addition, the use of graphite in the fuel bundle raises the possibility of the rapid release of neutron kinetic energy that is stored in the graphite. Furthermore, this approach achieves a reduced fuel exit burnup and leads to a reduction in the rated power of the bundle.

Another approach has been to replace the central fuel pins with a core of non-displaceable scattering material. Such a concept is described by M. H. M. Roshd et al. in Transactions of the American Nuclear Society, 1977 Annual Meeting, June 16 - 17, 1977. In particular, calculations based on conventional 61 and 37 pin CANDU fuel bundles modified by removing the central pins and replacing them with air filled or D₂0 filled aluminum cans showed that a large reduction in coolant void effect can be achieved.

Some prior art fuel pin geometries adopt a relatively close packed arrangement of fuel pins in an effort to increase efficiency or improve energy output. For example, in U. S. Patent No. 3,179,571 Schabert et al. there is described a fuel unit having a geometric arrangement of differently sized fuel pins in an annular area about a central coolant filled tube. While this arrangement is stated to reduce the self-shielding effect and provide a uniform distribution of coolant, the design has not been optimized to reduce coolant void reactivity.

While some of the prior art designs have been successful in significantly reducing the coolant void reactivity without requiring the addition of either enriched fuel or burnable poisons as materials in the fuel pins, such designs generally suffer a penalty in fuel exit burnup and maximum linear element rating. Fuel exit burnup is a measure of the amount of fissile material consumed (or power produced) before the fuel element is removed from the reactor. Maximum linear power rating is a measure of the fuel centreline temperature at the bundle power output limit. Accordingly, while reductions in coolant void reactivity can be achieved with prior designs, there is a need for a fuel bundle which uses natural uranium fuel and which controls coolant void reactivity without incurring an unacceptable penalty in fuel exit burnup maximum linear power rating and without making unacceptable demands on the fuel handling system.

SUMMARY OF THE INVENTION

The present invention is based on the discovery that coolant void reactivity in heavy water cooled and moderated reactors is dependant on the mean chord length of the coolant in the subchannels between the fuel pins and that the reactivity can be decreased by reducing the mean chord length. In the present invention, the fuel bundle geometry provides reduced mean chord length and reduced coolant void reactivity.

The fuel bundle design contains a central area of neutron scattering material that does not void from the central area upon a loss of coolant accident. In an annular area about the central region, a plurality of fuel pins are disposed with coolant flow subchannels therebetween. The fuel pins can advantageously be disposed in concentric rings, each ring having an equal number of fuel pins with the diameter of the fuel pins being graduated from one ring to another such that the outermost ring has the largest fuel pins and the innermost ring has the smallest fuel pins.

In accordance with the present invention there is provided a fuel bundle for a heavy water cooled and moderated reactor comprising a central cylindrical region containing a neutron scattering material which dpes not void from said central region upon a loss of coolant accident and an annular region co-axially disposed about said central region containing a plurality of fuel pins of elongated cylindrical shape disposed in parallel spaced relation with coolant flow subchannels therebetween, said fuel pins being uniformly disposed in said annular region in equal numbers in a plurality of concentric rings, said fuel pins in each concentric ring being smaller in diameter than the fuel pins in the adjacent outer ring and disposed on radial lines midway between the radial lines on which adjacent fuel pins in adjacent rings are disposed.

In accordance with another aspect of the present invention, the difference in radii of adjacent rings is less than the sum of the radius of each fuel pin in one adjacent ring and the radius of each fuel pin in the other adjacent ring.

The above and other aspects of the present invention will become apparent from the following description taken with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

In drawings which illustrate embodiments of the invention,

FIG. 1 is a cross-sectional view of a 37-pin fuel bundle design of the prior art;

FIG.2 is a cross-sectional view of a 42-pin fuel bundle design of the present invention;

HG.3 is a portion of a cross-sectional view of the fuel bundle design of FIG. 2 showing the geometry of the subchannel;

FIG.4 is a cross-sectional view of a 36-pin fuel bundle design of the present invention; and FIG. 5 is a cross-sectional view of a 30-pin fuel bundle design of the prior art which includes a non-displaceable central scattering core of graphite.

DETAILED DESCRIPΗON OF THE PREFERRED EMBODIMENT

The fuel bundle design commonly used in CANDU reactors is the 37- pin design shown in FIG. 1. The fuel bundle, generally designated by reference numeral 8, contains a set of thirty-seven equally sized like fuel pins, one of which is designated by reference numeral 10, disposed in three concentric rings 12, 14, 16 about a central fuel pin 18. The fuel material in each of the fuel pins is generally natural uranium in the form of uranium dioxide pellets (not shown). The diameter of fuel pins 10 is 13.08 mm. A plurality of fuel bundles 8 arranged end-to-end in pressure tube 20 is encased within calandria tube 22. Gas, typically helium or carbon dioxide, is present within the annular space 24 between pressure tube 20 and calandria tube 22 to thermally insulate pressure tube 20 from calandria tube 22 and the heavy water moderator which flows in the space outside calandria tube 22. Heavy water coolant is contained within pressure tube 20 and fills the subchannels 26 between the fuel pins. When the reactor is in operation, the coolant flows over the fuel bundle to cool the fuel and remove heat from the fission process. Minimizing coolant void reactivity was not a predominant consideration in developing the prior art 37-pin fuel bundle design.

The fuel bundle of the present invention and the manner in which it minimizes coolant void reactivity can best be understood by first considering how coolant void reactivity arises in prior art fuel bundles such as the 37-pin design of Fig. 1. In a nuclear reactor, the relative rates of neutron production and neutron absorption determine the rate of neutron multiplication. Loss of coolant produces positive coolant void reactivity because it causes a loss of neutron scattering from the lattice and consequently an alteration in the speed of neutrons which has a direct effect on the rates of neutron production and neutron absorption.

Fast neutrons, such as fission neutrons which have not been slowed down by the moderator, can produce fission in all uranium and plutonium isotopes including U₂₃₈- The minimum neutron speed required to produce such fission corresponds to an energy of about 1 Mev. The natural uranium fuel used in CANDU type reactors is comprised of less than approximately 0.73 percent U₂₃₅ and PU₂₃₉ isotopes. Most of the remainder (approximately 99 percent) consists of U₂₃β- Because relatively few U₂₃₅ and Pu₂₃₉ nuclides are present compared with U_238/ the probability of a fast neutron colliding with these nuclides and producing fission before it collides with a nuclide of scattering material and slows down below an energy level of 1 Mev, is small. In contrast there is a much greater probability that a fast neutron will produce fission in U₂₃β because of the large number of U₂₃₈ nuclides present. Fission of U₂₃₈ by fast neutrons contributes almost 6 percent to energy production and 2.2 percent to neutron multiplication in CANDU type reactors.

There are two types of fissions by fast neutrons: (i) fissions of U₂₃₈ in a fuel pin that are produced by fast neutrons that were born in the same fuel pin, and (ii) fissions of U₂₃β in a fuel pin that are produced by fast neutrons that were born in other fuel pins. The rate of U₂₃β fission produced by external fast neutrons is relatively small because the fast neutron must travel through the heavy water coolant without collision to maintain sufficient speed to cause fission in U₂₃₈- Even a single collision of a neutron in heavy water is sufficient to slow a 1 Mev neutron to an energy that can no longer produce fission in U₂₃₈- When the coolant has voided, there is therefore an increased probability of a fast neutron reaching another pin without collision and accordingly, the rate of type (ii) U₂₃β fission referred to above increases on loss of coolant. The resulting increase in the rate of neutron multiplication accounts for approximately one third of the coolant void reactivity in CANDU type reactors.

The abundance of U₂₃₈ in natural uranium fuel is responsible for another major source of positive coolant void reactivity. In particular, the increase in reactivity is a result of the effect of the coolant on the rate of neutron absorption. Although neutrons of all speeds are absorbed by U₂₃β, the probability of neutron collision increases as the neutrons slow down and eventually reach the energy level of the heavy water moderator. The rate of slow (approximately 0.5 ev) neutron absorption in U₂₃β is orders of magnitude higher than the rate of fast (1 to 2 Mev) neutron absorption. Between these two limits, the relationship between absorption rate and neutron speed is monotonic except at a relatively narrow energy band of about 100 kev. At this so called "resonance energy band" the rate of neutron absorption by U₂₃₈ is disproportionately high. In fact, over 25 percent of the total neutron absorption occurs within this energy band. In the lattice of the reactor, fast neutrons are slowed down by collisions in the moderator and to some extent by collisions in the coolant. The collisions in the coolant cause a significant number of neutrons to be shifted into the resonance energy band which, in turn, promotes their absorption. Upon loss of coolant, this shift no longer occurs and the absorption rate of U₂₃₈ neutrons is therefore reduced. This reduction in the U₂₃β absorption rate accounts for another approximately one third of the positive coolant void reactivity in CANDU type reactors.

The remaining third of coolant void reactivity production in CANDU type reactors is a result of isotopes other than U₂₃₈, for instance U₂₃₅ and Pu₂₃₉, and the higher actinides and fission products.

It has been discovered that the amount of coolant void reactivity produced depends on the probability that a neutron will collide with a nuclide of the coolant when the neutron is travelling between fuel surfaces. This probability is determined by the number of nuclides of the coolant that a neutron will encounter between fuel surfaces. The latter parameter is called the coolant chord length. The mean chord length of the coolant is the mean distance that a neutron must travel through the coolant between fuel surfaces. It is directly proportional to the volume of coolant in the subchannel divided by the surface area of the three fuel pins in the subchannel. It depends on the shape of the subchannel, the volume of the coolant in the subchannel and on the angular direction in which the neutrons are travelling.

The fuel bundle design of the present invention exploits the above described behaviour of neutrons to minimize coolant void reactivity. In FIG. 2, a 42-pin fuel design of the present invention is shown. Pressure tube 32 contains a plurality of fuel bundles, one of which is designated by reference numeral 30. The fuel bundles are arranged end-to-end along the length of the fuel channel (not shown). Gas, typically helium or carbon dioxide, is present within the annular space 34 between pressure tube 32 and calandria tube 28. Calandria tube 28 is surrounded by moderator.

Fuel bundle 30 has two concentric rings of elongated cylindrically shaped fuel pins, two of which are designated by reference numeral 44 in FIG. 2. Fuel pins 44 are disposed in parallel spaced relationship in concentric outer ring 36 and inner ring 38 about a central solid cylinder 40 of neutron scattering mateπal. A zirconium alloy sheath 46 surrounds the perimeter of cylinder 40. Heavy water coolant flows through the subchannels 42 about fuel pins 44 between pressure tube 32 and central cylinder 40. Fuel pins 44 and central cylinder 40 are held together by endplates (not shown) at either end of the fuel bundle.

In the 42-pin design, outer ring 36 and inner ring 38 each contain twenty-one fuel pins. Outer ring 36 has a pitch circle radius of approximately 44.7 mm and inner ring 38 has a pitch circle radius of approximately 34.7 mm. The pitch circle radius is defined as the radius of the circle which passes through the centres of the fuel pins comprising the ring. Each of fuel pins 44 in outer ring 36 has a diameter of 11.40 mm and each of fuel pins 44 in inner ring 38 has a diameter of 9.40 mm. Fuel pins 44 are clad in a zirconium alloy sheath (not shown) having a thickness of approximately 0.3 mm to 0.4 mm. Central cylinder 40 has a diameter of 55.3 mm and sheath 46 has a thickness of at least 0.3 mm. Fuel pins 44 contain pellets of natural uranium dioxide as fuel material. Alternatively, slightly enriched uranium dioxide, plutonium dioxide mixed with uranium dioxide or unreprocessed spent light water reactor fuel (referred to as DUPIC) may alternatively be used.

The geometric arrangement of the pins is shown in detail in FIG. 3. The centres of outer ring fuel pins 44A and 44B are disposed along radial lines 48 and 50 respectively. The centre of inner ring fuel pin 52 is disposed along radial line 54 which is midway between radial lines 48 and 50. Consequently, the distance between pin 52 and pin 44A (which distance is designated reference numeral 56) is equal to the distance between pin 52 and pin 44B (which distance is designated reference numeral 58). Angle 60 is 17.143 degrees and angle 62 is half thereof.

In the fuel bundle design of the present invention the reduced size and the configuration of the subchannel between adjacent pins reduces the mean chord length and decreases the probability of collisions occurring in the coolant. Consequently, there is a decreased probability that during the slowing down process, neutrons will be scattered into the resonance energy band. Therefore, under normal operating conditions, the absorption rate of resonance energy neutrons in U₂₃₈ is low compared to the corresponding rate in standard fuel designs such as the 37-pin fuel bundle of FIG. 1. On loss of coolant, there is normally a reduction in the number of neutrons slowing down into the resonance energy band and hence a decrease in the rate of absorption which leads to increased neutron multiplication. However, in the fuel bundle design of the present invention, the reduced number of neutrons being scattered into the resonance energy band which occurs on loss of coolant produces significantly less decrease in the absorption rate of resonance neutrons in U₂₃₈ because the absorption rate of such neutrons in U₂₃₈ is already low under normal operating conditions. This minimal disparity in absorption rates before and after loss of coolant minimizes the increase in neutron multiplication and thus contributes to the reduction in coolant void reactivity.

The reduction in void reactivity of the design of the present invention is attributable to a number of factors. The presence of central cylinder 40 provides a centrally disposed scattering volume that does not void on loss of coolant minimizing the disparity in neutron multiplication before and after loss of coolant. In addition, in the present invention, the pins of the inner ring are nestled midway between the adjacent pins of the outer ring which advantageously reduces the size of the subchannels between the pins. As used herein, the term 'subchannel' refers to the area defined by a triangle whose apices are the centre of the fuel pin of one ring and the centres of the adjacent fuel pins of an adjacent ring less the area of that triangle occupied by the fuel pins. This close nestling of pins which results in reduced chord length and reduced subchannel volume is made possible by the use of an equal number of smaller pins in the inner ring as in the outer ring and the positioning of the inner pins on radial lines midway between the adjacent pins in the outer ring. As shown in FIG. 3, with the present invention, the difference in the pitch circle radius of adjacent rings can be equal to or less than the sum of the radius of pin 44A and the radius of pin 52 permitting a smaller subchannel volume than is typical in prior designs. With this arrangement, the radially outermost limit of the fuel pins in the inner ring can be equal to or greater than the radially innermost limit of the fuel pins in the outer ring which allows the pins to be nestled and the subchannel volume reduced while maintaining adequate spacing for thermohydraulic and fuel engineering considerations. With the particular design of FIG. 2, the maximum value of the minimum distance between any two adjacent fuel pins or between a fuel pin and the pressure tube or between a fuel pin and the central sheath should be limited to approximately 2.0 mm. The minimum value of the minimum distance is dictated by thermal hydraulic and other fuel engineering requirements and can be as low as approximately 0.5 mm.

In the present invention, the use of smaller pins in the inner ring is to be contrasted with other prior art designs such as Schabert et al. in which the fuel pins in the inner ring are larger in diameter and therefore have a larger surface area than the pins in outer rings. As noted above, mean chord length is dependent in part on the surface area of the pins in contact with the coolant. The reduced average pin size also decreases the probability of collisions of U₂₃β within a fuel pin and increases the probability of a fast neutron (such as an uncollided fission neutron) escaping a fuel pin without collision. The reduced pin size is also a factor which contributes to the size and shape of the subchannel between adjacent pins necessary to decrease the probability of collisions occurring in the coolant. The close nestling of the pins is also advantageous because it provides additional space in the centre of the fuel bundle for neutron scattering material. The placement of pins having a larger diameter in the outer ring is contrary to conventional thinking because fuel located near the outside of the fuel bundle is vulnerable to the highest operating temperature. Thus to minimize the temperature increase in the fuel, fuel bundles are conventionally designed with the smaller pins placed in the outer ring in order to accommodate a larger number of pins. In the present invention, the placement of the larger fuel pins in the outside ring does not impact unfavourably on fuel temperature because the neutron flux level in the two rings is closer with this type of design. In addition, the overall size of the fuel pins can advantageously be sufficiently small so as to not impact unfavourably on fuel temperature.

The central scattering volume of the present invention also reduces the coolant mean chord length of the subchannels by reducing the track length of the neutrons in the coolant that are scattering between fuel pins. Collision of fast neutrons in the central scattering volume reduces the probability, on loss of coolant, of fast neutrons (such as uncollided fission neutrons) escaping a first pin, reaching a second pin without collision, and producing fission of U₂₃β in the second pin. The magnitude of this effect can be controlled by appropriate selection of the central scattering material.

Central cylinder 40 preferably contains amorphous carbon as the neutron scattering material. Amorphous carbon is suitable for the neutron scattering material for several reasons. It is an effective neutron scattering material which does not absorb neutrons to any appreciable degree. It can also be specifically manufactured to have a density ranging from low to its theoretical density. In general, a higher density of neutron scattering material results in a greater reduction in coolant void reactivity. The ability to adjust the density allows a degree of control over the level of coolant void reactivity realized by the fuel bundle. Amorphous carbon is also suitable for the formation of the central cylinder because it is not chemically poisonous and because it prevents the explosive release of stored neutron kinetic energy during reactor operation. Reticulated silicon carbide is another suitable material for the formation of the central cylinder because it shares the characteristics of amorphous carbon outlined above.

Other scattering materials such as graphite and heavy water may also be used although due consideration must be given to engineering constraints. For example, while heavy water has scattering properties, provision would have to be made to ensure that it does not void from the central core upon a loss of coolant accident. In addition, the present invention may also include designs in which the neutron scattering in the central region is provided substantially by the effect of the structural central cylinder 46 and the interior space that it defines. The central cylinder may contain helium, carbon dioxide or other low density, non-corrosive and non-neutron absorbing materials. Alternatively, the central cylinder may be solid zirconium alloy.

While the present invention has been described as having 42 pins and the specific geometry referred to above, a person skilled in the art will appreciate that these details can be varied while remaining within the spirit and scope of the present invention. For example, FIG. 4 shows a cross- sectional view of an alternative 36-pin design. In this embodiment, there are eighteen pins per ring. The diameter of the pins in the outer ring 64 (one of which is designated reference numeral 66) is 11.4 mm and the diameter of the pins in inner ring 68 (one of which is designated reference numeral 70) is 11.0 mm. The pitch circle radii of outer ring 64 is 44.6 mm and the pitch circle radius of inner ring 68 is 34.5 mm. The thickness of the zirconium alloy sheath (not shown) surrounding the fuel pins is between about 0.3 mm and 0.4 mm. There is a central cylinder of amorphous carbon 72 or other suitable neutron scattering material with a diameter of 54.6 mm and a zirconium alloy sheath 74 having a thickness of at least 0.3 mm. Angle 73 is approximately 20 degrees and angle 75 is approximately 10 degrees. The arrangement of the pins is based on the same considerations as noted above with respect to the 42-pin design. The manner in which this design controls coolant void reactivity is based on the same principles as the 42-pin design. The 36-pin design provides a higher power output and a marginally lower average fuel exit burnup than the 37-pin design, especially when amorphous carbon is used.

Not only does the fuel bundle design of the present invention offer improved coolant void reactivity over prior designs, it does so without significant penalty to fuel burnup or maximum linear heat rating. For purposes of comparison, Fig. 5 represents the prior art 37-pin fuel bundle design shown in FIG. 1 which has been modified by the use of a non- displaceable central scattering volume 40, as suggested by Roshd et al. improve coolant void reactivity.

The fuel bundle designs described above were subjected to WIMS and MCNP simulations to evaluate coolant void reactivity, fuel exit burn-up and maximum linear rating. The WIMS simulation code is available from the Reactor Shielding Information Centre in Oakridge, Tennessee and is commonly used in conventional reactor design. MCNP simulation is described by J.F. Briesmeister in A General Monte Carlo Code For N -Particle Transport, LA-12625, 1988. The following fuel designs were evaluated:

(i) 37-pin fuel bundle design shown in FIG. 1;

(ii) 42-pin fuel bundle design shown in FIG. 2;

(iii) 36-pin fuel bundle design shown in FIG. 4;

(iv) 34-pin fuel bundle design shown in FIG. 5; and

(v) 30-pin fuel bundle design shown in FIG. 6

The results of the evaluation are shown in Table 1 below. To conduct the simulations, data was entered relating to the geometry, composition and operating temperature of the lattice. The values given in each case are for equivalent reactor operating conditions.

IASLR1

Void-Reactivity (mk) Fuel Exit Max. Linear

Beta WIMS MCNP Burnup Heat Rating

Design (mk) Fresh Equil Fresh Equil (MWd/Te) (kW/m)

42-Pin 6.1 8.7 5.6 8.3±.50 52 6208 58.4

36-Pin 5.9 10.1 65 9.4±.40 5.8 6527 602

37-Pin 5.9 17.2 13.8 16.51.36 13.1 6690 60.7

30-Pin 5.9 12.7 8.7 12.1±.47 8.1 6465 70.4

All designs included in Table 1 use natural uranium in the form of uranium dioxide as the fuel material. The scattering material in the central core for the 36-pin, 37-pin and 30-pin designs is amorphous carbon having a density of 1.6 g/cc. The values given for void reactivity are considered to be conservative because they are expected to drop by between 0.5 and 1.0 mk due to the increase in neutron leakage on coolant voiding which is not included in the calculations. The maximum linear heat rating values are for a bundle power of 950 kW. The MCNP equilibrium void reactivity values have been inferred from the drop in void reactivity with burnup as shown by WIMS.

The results in Table 1 show the clear superiority of the 42 and 36-pin designs of the present invention over the standard prior art 37-pin design and as modified with the inclusion of a central scattering core as represented by the 30-pin design. In the case of the standard 37-pin design, all void reactivity values are substantially in excess of the prompt critical value beta. While the 30-pin design which is a modification to the 37-pin design to replace the central 7 pins with a non-displaceable scattering core does improve the void reactivity numbers, it introduces a large penalty in the maximum linear heat rating value.

The 42 and 36 element designs of the present invention show extremely good void reactivity numbers, without penalty in maximum linear heat rating. For example, the 42-pin design has an equilibrium coolant void reactivity below prompt critical. This is also true of the 36-pin design of the present invention when one takes into account the further 0.5 to 1 mk drop due to increased reactor leakage on coolant voiding that is not reflected in Table 1. In contrast, neither the 37-pin or the 30-pin designs achieve below prompt critical void reactivity.

Moreover, in contrast to the 30-pin design that achieves some improvement in void reactivity at the expense of maximum linear heat rating, both the 42-pin and 36-pin designs have maximum linear heat ratings below that of the standard 37-pin design. In addition, there is only a slight fuel burnup penalty for the 42-pin and 36-pin designs as compared to the 37-pin design. In the case of the 36-pin design, the fuel burnup penalty is less than that of the 30-pin design. In fact, the 36-pin design is superior to the 30-pin design in void reactivity, burnup and heat rating.

The present invention provides a fuel bundle which acts to control the reactivity upon coolant voiding. As natural uranium may be used as the fuel material the need for enriched fuel is eliminated. The fuel bundle is also advantageous because there is little penalty of fuel burnup reduction or bundle power rating reduction. Furthermore, the fuel bundle adopts the overall geometry of current CANDU fuel bundle designs and can therefore be used in currently operating CANDU reactors.

While the invention has been described in connection with three specific embodiments thereof, numerous modifications and adaptations of those embodiments are contemplated by the inventors and will occur to those skilled in the art without departing from the spirit and scope of the invention as set forth in the appended claims. For example, although the 42-pin and 36-pin designs described here and shown in FIGS. 2 to 4 have two concentric rings of fuel pins, the present invention could include a design having fuel pins arranged in three or more concentric rings provided that each ring has an equal number of fuel pins and provided the diameter of the fuel pins gets progressively larger from the innermost ring to the outermost ring.

Claims

The embodiments of the invention in which an exclusive property or privileges is claimed are defined as follows:

1. A fuel bundle for a heavy water cooled and moderated reactor comprising a central cylindrical region containing a neutron scattering material which does not void from said central region upon a loss of coolant accident and an annular region co-axially disposed about said central region containing a plurality of fuel pins of elongated cylindrical shape disposed in parallel spaced relation with coolant flow subchannels therebetween, said fuel pins being uniformly disposed in said annular region in equal numbers in a plurality of concentric rings, said fuel pins in each concentric ring being smaller in diameter than the fuel pins in the adjacent outer ring and disposed on radial lines midway between the radial lines on which adjacent fuel pins in adjacent rings are disposed.

2. The fuel element of claim 1 wherein the difference in radii of adjacent rings is less than the sum of the radius of each fuel pin in one adjacent ring and the radius of each fuel pin in the other adjacent ring.

3. The fuel bundle of claim 2 wherein the number of concentric rings is two.

4. The fuel bundle of claim 1, 2 or 3 wherein said fuel pins contain a fuel material selected from the group consisting of natural uranium dioxide, enriched uranium dioxide, uranium dioxide mixed with plutonium dioxide and DUPIC.

5. The fuel bundle of claim 1, 2 or 3 wherein the neutron scattering material is selected from the group consisting of amorphous carbon, reticulated silicon carbide and graphite.

6. The fuel bundle of claim 5 wherein the neutron scattering material is amorphous carbon.

7. The fuel bundle of claim 5 wherein the neutron scattering material is reticulated silicon carbide.

8. The fuel bundle of claim 5 wherein the neutron scattering material is graphite.

9. The fuel bundle of claim 1, 2 or 3 wherein the central cylindrical region comprises a central void space surrounded by a zirconium tube.

10. The fuel bundle of claim 3 wherein said two concentric rings comprise an inner ring of a radius of 34.7 mm having twenty-one fuel pins of a diameter of 9.40 mm and an outer ring of a radius of 44.7 mm having twenty-one fuel pins of a diameter of 11.40 mm; and wherein said neutron scattering material is a solid cylinder of amorphous carbon having a diameter of 55.3 mm.

11. The fuel bundle of claim 3 wherein said two concentric rings comprise an inner ring of a radius of 34.5 mm having eighteen fuel pins of a diameter of 11.0 mm and an outer ring of a pitch circle radius of 44.6 mm having eighteen fuel pins of a diameter of 11.40 mm; and wherein said neutron scattering material is a central cylinder of amorphous carbon having a diameter of 54.6 mm.