WO1991014268A1

WO1991014268A1 - Pressurized water reactor fuel

Info

Publication number: WO1991014268A1
Application number: PCT/US1990/001111
Authority: WO
Inventors: Nathan Fuhrman; Alf Ingemar Jonsson
Original assignee: Combustion Engineering, Inc.
Priority date: 1990-03-07
Filing date: 1990-03-07
Publication date: 1991-09-19
Also published as: FI923972A0; EP0518860A1; FI923972A; NO923373L; NO923373D0

Abstract

An improved fuel array including fuel rods of a mixture in the approximate range of from 0.2 to 1.5 percent by weight Er2O3 and the remainder UO2, for use in the core of a pressurized water nuclear reactor in which boron reduces beginning-of-cycle excess reactivity, and in which the fuel cycle of the reactor core is extended while maximizing a local power distribution uniformity, uniform burndown, thermal margin and available incore heat transfer area and ensuring a negative moderator temperature coefficient of reactivity.

Description

PRESSURIZED WATER REACTOR FUEL

BACKGROUND OF THE INVENTION

This is a continuation of copending application serial number 07/132,847 filed on December 14, 1987.

(1) Reactor Core, Generally

The core of a modern pressurized water nuclear reactor (PWR) contains slightly enriched uranium dioxide pellets enclosed in Zircaloy tubing to form fuel rods.

In present fuel designs, a support is assembled much in a way an egg crate is put together, and the assembled pieces are joined to form a rigid square grid. Several such grids are then supported at approximately one to two foot intervals to form a rigid fuel support. Fuel rods are then inserted individually into the grid positions thus creating a "fuel assembly". The grids have dimples or "tabs" which hold each rod firmly in place. Because the fuel rods are not physically bound to the grids, the rods are free to expand radially and axially and will not bow under differential expansion. This construction minimizes the fuel support structure present in the core. Since any structure is a parasitic absorber of neutrons which sustain the nuclear reactor, the smaller amount of structure in the fuel assemblies making up the core allows the use of uranium of lower enrichment in U-235 and results in a significantly reduced fuel cost.

These cores are termed "batch loaded cores". The core typically is divided into three regions, with fuel of different enrichment in U-235 being present in each region.

When the reactor core is refueled, the oldest fuel is discharged; the youngest region is moved to new locations; and a fresh charge of fuel is loaded into the outer region. The typical refueling operation is performed only approximately once per 12-24 months; the residence time of a fuel assembly in the reactor is typically 2 to 3 years. It would be desirable to be able to increase these time periods.

Reactivity control during reactor operation has been accomplished by means of mechanical control rods and by the use of a water soluble neutron absorber method called "chemical shim control". In chemical shim control, only part of the initial load of boric acid has been removed from the reactor before power operation begins. The remainder of the boron is removed -- a few parts per million per day -- over the lifetime of the core to compensate for changes in reactivity caused by burnup of fuel and by the accumulation of fission products. Rods containing burnable absorbers are also used for control of reactivity. These rods are fixed burnable absorber (BA) shims of solid material in the fuel assembly containing parasitic neutron absorbers having a concentration which permits most or all of it to be consumed during one or more cycles in the reactor. However, the presence of shim rods in the core still causes a distortion of the power distribution within the core. If this power maldistribution could be essentially eliminated, more energy could be extracted without exceeding safety and operational limitations.

It is a primary purpose of in-core fuel management to minimize the amount of U-235 or other fissile material required for a given energy output during a given cycle. This can be appreciated by the fact that for every 0.1 weight percent (wt%) increase in required core average enrichment, the increased cost of fuel for that cycle is in the millions of dollars.

Typical equilibrium cycle core average enrichments are about 3.3 wt% U-235. It can also be appreciated that the greatest savings in overall fuel costs will be achieved by minimizing the enrichment required for an equilibrium fuel management scheme. The major constraint on the flexibility of in-core fuel management is imposed by very strict peak power

distribution limitations required by safety considerations. For example, the predicted ratio of the powers produced in the hottest fuel rod to the core average fuel power is typically not permitted to exceed 1.60. This imposes correlative requirements on the ratio of power produced in a fuel assembly to the core average assembly power "Global Peak to Core

Average" (P/A)_G, and on the ratio of maximum rod power within an assembly to the average power in the assembly containing that rod "Local Peak to Assembly Average" (P/A)_L. In modern commercial PWRs, burnable absorber shims are frequently located in selected assemblies to control the power distribution and to replace chemical shim. These shims are strongly absorbent when the assembly is first placed in the core, and become weaker the longer they are exposed to the operating core environment.

Although the shims are useful for controlling the power distribution and other core characteristics such as the moderator temperature coefficient, the presence of residual shim poison at the end of a cycle presents an inherent reactivity penalty, and requires a greater initial U-235 enrichment and cost.

It is an object of this invention to eliminate or minimize the necessity of placing power distribution distorting shim rods in the fuel assembly arrays.

(2) Brief Description of the Prior Art Drawings

Fig. 1 is a schematic illustration of a quadrant with origin 0 of a typical prior art reactor core, showing a first cycle arrangement of fuel assemblies in the core.

Fig. 2 is a table of the characteristics of the prior art fuel assemblies of Fig. 1;

Fig. 3 is a schematic illustration of a typical fuel assembly square array showing by symbols three typical prior art fuel assembly types used in Fig. 1 and characterized in Fig. 2; Fig. 4 is a schematic illustration of the kind shown in Fig. 3 illustrating a fourth typical prior art fuel assembly type used in Fig. 1 and characterized in Fig. 2;

Fig. 5 is a key table to the symbols used in Figs. 3 and 4;

Fig. 6 is a schematic illustration of a typical prior art reactor core octant with origin 0 showing global planar average power distribution (P/A)_G at a core average burnup of 6000 MWD/T of the third cycle;

Fig. 7 is a schematic illustration of a typical prior art unshimmed fuel assembly octant with origin 0 showing local planar average power distribution (P/A)_L with a boron moderator concentration of 1470 parts per million (ppm);

Fig. 8 is a schematic illustration of a typical prior art fuel assembly with eight (8) shims of Gd₂O₃ (GD) with a boron moderator concentration of 1470 parts per milli

on (ppm);

Fig. 9 is a graph illustrating the percent of increased (P/A)_L of a typical fuel assembly with eight (8) Gd₂O₃ and one with eight (8) B₄C shim rods when compared to a typical fuel assembly with no burnable absorber (BA) shim rods. (3) Power Distribution Background

The reactor core consists of an arrangement of fuel assemblies, Figs. 1-5. The fuel assemblies, Figs. 3 and 4, consist of square arrays of rods having 16x16 rod positions, as one example of n x n rod positions when n = 14 to 20. In typical designs, these arrays are mostly fuel but some positions are reserved for control rods, burnable absorbers or for incore instrumentation.

The geometry and composition of the nuclear fuel determines the neutron flux and power distribution. Economical considerations as well as safety limits dictate that the power distribution across the core be maintained as flat, as close to 1.0 in ratio, as possible for most, if not all, of the time during each cycle. Because neutrons are lost radially by leakage across the perimeter of the core, the neutron flux and hence the power distribution cannot be perfectly flat even if the composition of the core were uniform. Furthermore, the reloading of fuel at the end of each cycle replaces only a portion of the core (typically 1/2 to 1/6), the exact portion being dictated by economic factors. Thus, the core will consist of fuel

assemblies of varying ages and compositions and this leads to a non-uniform flux and power distribution.

In one convenient terminology, batch A has the lowest enrichment, batch B a higher enrichment, and batch C the highest enrichment. At the end of the first cycle, batch A is removed from the reactor, batches B and C are rearranged, and a feed batch D of fresh fuel is placed in the reactor. This procedure is typical of three batch incore fuel management wherein an entire batch of fuel is removed and replaced with the same number of feed fuel assemblies every year for the life of the plant. It is usually desirable to achieve an

equilibrium incore fuel management scheme as early as possible in the plant lifetime, such that the feed assemblies will always have the same enrichment and will be placed in the same locations as the previous feed assemblies, and the once-burned and twice-burned assemblies that remain in the core will be shuffled to identical locations occupied by the previously once and twice-burned assemblies.

One important design objective is to minimize the power distribution peaks in the core. This is done by judicious placement of fuel assemblies of different types and to some extent by placing the fuel rods within assemblies in an optimal fashion.

The end of the operating cycle occurs when the soluble boron concentration is reduced to essentially zero and when the fuel fissile constituents of the fuel have been reduced (by fuel burnup) to the degree where neutron production from the fuel is no longer sufficient to compensate for neutron absorption by fission products, other materials in the reactor, and neutron leakage from the reactor. It is

established art that the length of the operating cycle can be increased (or, alternately, the fuel enrichment reduced) by reducing the neutron leakage from the core, and, further that this can be accomplished by locating more highly burned fuel at the core edge and placing fuel of the feed batch toward the interior of the reactor core ("low leakage" fuel management). Such placement of the feed batch generally requires the use of burnable absorbers to assist in the control of power production in such feed batch fuel; further, the depletion characteristics of such burnable absorbers must be such as to maintain power production control throughout the operating cycle.

Burnable absorbers (BA) have two functions, one being to control the moderator temperature coefficient of reactivity (MTC), and one being to assist in the control of the power distribution. A typical global power distribution (P/A)_G for the core is shown in Fig. 6. This provides a coarse

description of the total power distribution. Only the assembly average power values are shown. The detailed rod-by-rod power distribution within each fuel assembly is referred to as the local power distribution (P/A)_L. The global power distribution is primarily determined by the average compositions of each fuel assembly type and the local power is determined by the geometric and composition layout of the assembly itself.

Typical local power distributions are shown in Fig. 7, for an assembly without BA, and Fig. 8, with Gd₂O₃ (GD) as a burnable absorber. Because of its very high neutron cross-sections, the depletion of GD would be too rapid if applied uniformly in the fuel. This has been counteracted by concentrating all the GD in a few (8 in the example) rods so that the neutron flux in these rods becomes very low initially, i.e. the GD is

self-shielded and its effective absorption rate is considerably lower than it would be if uniformly applied.

Another example of the local power distribution could be shown for B₄C as a burnable absorber with a relatively low cross-section. In this case, one would desirably place the B₄C uniformly with the fuel. B₄C is not chemically or physically compatible with UO₂, however, and must therefore be placed in entirely separate rods. Thus, the geometry becomes again non-uniform and the (P/A)_G is driven up. Another disadvantage of this application of B₄C or Gd₂O₃ is that fuel is displaced so that some power capability or margin is lost.

The local (within an assembly) neutron flux distribution would be flat only if two conditions are

fulfilled:

a) the distribution of material is uniform; and b) the assembly is placed in an environment which does not impose external flux gradients.

The power distribution is determined by the product of flux and fission cross-section, the latter being determined by the distribution of fissionable material. It follows, that in order to minimize the local (P/A)_L power ratio, one should make the distribution of fuel and any other components as uniform as possible. Because space must be made available for control rods and for incore instrumentation, Figs. 3, 4, and 5, this is not possible, even when burnable absorbers are not part of the assembly. A typical power distribution in such a case is shown in Fig. 7.

When BA is introduced in this geometry, the (P/A)_L ratio increases. Figs. 8 and 9 show examples of this. For example, the initial increase in peaking may be 10% or more with GD (Fig. 9). The reason for this increase is that this BA is a strong absorber of neutrons so that it depresses the flux in its vicinity. Along with the flux, the local power drops in the vicinity of the BA rods and in order to maintain constant overall power, the local power must rise elsewhere. In other words, the (P/A)_L increases. This is the case with all BAs that must be placed in a few rods only. (GD, B₄C, borosilicate glass, etc.) The ideal BA is one that can be mixed

homogeneously with the UO₂ in a uniform (equally in all fuel rods in the assembly) fashion. If this were possible, the flux distribution and hence the power distribution would have essentially the same shape as without BA shim rods. If, in addition, as in one embodiment of this invention, the BA were such that it could be placed (in addition to within the fuel) in conjunction with other components (guide tubes) of the assembly with some suitable concentrations, it would be possible to bring the (P/A)_L ratio to values below the unshimmed assembly.

The difficulty in doing this lies in finding a BA material that does not deplete too fast (as do certain borides placed on the surface of the fuel). The effective neutron capture cross-section must be high enough but not too high and the BA should be physically compatible with UO₂ so that it can be given a degree of initial self-shielding.

SUMMARY OF THE INVENTION

The present invention demonstrates that a low (P/A). ratio is possible in the PWR neutron spectrum if Er₂O₃ (erbium oxide) is chosen as the BA material. It is physically and chemically compatible with UO₂ and can therefore be placed homogeneously in the UO₂ where the flux is lower than on the surface of the fuel rods (providing self-shielding and not too rapid depletion). Its cross sections are not too high so that uniform application is possible (low P/A results). The uniform application also permits the self-shielding not to become too high and hence the depletion rate is not too slow. Er₂O₃ affords an optimal depletion rate while not distorting the power distribution or displacing a significant amount of fuel. BRIEF DESCRIPTION OF THE INVENTION DRAWINGS

Fig. 10 is a schematic illustration of a proposed fuel assembly octant with origin 0 showing expected local planar average power distribution (P/A)_L with the Er₂O₃ uniformly applied to all rods and homogeneously mixed in the UO₂ with a typical concentration of 0.2-0.3% by weight;

Fig. 11 is an octant schematic illustration of a typical proposed fuel assembly with some of the fuel rods having Er₂O₃ homogeneously included in the UO₂ fuel in a typical concentration of 0.5% by weight and fuel rods having 100% UO₂;

Fig. 12 is a graph illustrating the typical percent of potential thermal margin improvement using Er₂O₃

homogeneously included in UO₂ fuel rod as a burnable absorber as opposed to B₄C discrete shim rods or Gd₂O₃ (GD) discrete shim rods used as a burnable absorber.

DESCRIPTION OF THE PREFERRED EMBODIMENT

The present invention relates to pressurized nuclear power reactors (PWR) and to burnable absorbers (BA) used in them for the purposes of:

a) Moderator Temperature Coefficient (MTC) of Reactivity Control.

b) Power Distribution Control.

The Er₂O₃ burnable absorber and UO₂ fuel mixture, which is the subject of the invention, is such that it achieves those two purposes with a minimum of (or even negligible) impact on the margin to safety limits.

A nuclear reactor of the PWR type is constructed with sufficient excess reactivity incorporated in the core to compensate for fuel burnup and increasing fission product neutron absorption during the fuel residence time in the core. The amount of reactivity designed into the core is sufficient to provide full power operation for one cycle (typically 12-, 18- or 24-month cycles). The release of this reactivity is gradual and typically achieved by diluting boric acid dissolved in the moderator. Boric acid is a strong neutron absorber. The longer the cycle, the more boric acid would be needed initially. As the concentration is increased, undesirable effects present themselves, however. The one of interest here is an increase in the moderator temperature coefficient of reactivity. In order to ensure a negative MTC at

beginning-of-cycle (BOC), burnable absorbers can be applied. These are neutron absorbers, usually with high cross sections to ensure a relatively rapid depletion, homogeneously mixed with the UO₂ fuel in a few fuel pins within a fuel assembly. They may also be separated from the fuel and placed in a few, discretely located absorber pins within the fuel assembly. In either case, their concentration does not change with moderator temperature so that the MTC is not greatly affected.

We thus see that BAs are conventionally located in a few discrete locations in a fuel assembly, Figs. 3 and 8.

Compared to a fuel assembly without BA, Fig. 7, one finds that the discrete mode of applying the BAs produces a distortion of the local (assembly) power distribution. A measure of the distortion is the peak-to-average local power ratio (P/A)_L. Fig. 9 shows typical (P/A)_L ratios for conventional BA's compared to the case without BA.

Increases of 5-10% in (P/A)_L can be seen. Since, frequently the core peak power is located in one of these fuel assemblies, an increase of 5-10% in local (P/A)_L will also mean a similar increase in core peaking (P/A)_G. This translates to a thermal margin loss of 5-10% which could seriously affect power capability. The invention thus involves a BA and fuel mixture of Er₂O₃ and UO₂ in rods arranged either alone or with pure UO₂ rods in fuel assemblies in a way which minimizes the increase in P/A ratio locally.

The present mixture of 0.2% to 1.50% by weight Er₂O₃ with the remainder UO₂ reduces MTC while maintaining a low (P/A)_L ratio close to or no different from that with no BA. (Compare Figs. 7 and 10) In addition, the invention permits a more efficient MTC control for a given initial neutron absorption rate. The latter feature translates to an economic advantage via less residual reactivity at end-of-cycle (EOC).

The invention replaces a conventional BA, such as B₄C or Gd₂O₃, with high neutron absorption cross section with one having moderately high cross-section but which is uniformly (rather than discretely) applied throughout the entire fuel assembly (Fig. 10) or almost the entire assembly, (Fig. 11). This permits the neutron flux to have a higher level in the BA and hence the depletion rate is acceptably fast so that little residual remains at EOC. The Er₂O₃ absorber used is compatible with the fuel. It is homogeneoously mixed with the UO₂ in all or some of the fuel rods. Specifically, erbium oxide (Er₂O₃), where the erbium (Er) is the active component, when applied in a PWR with a 24-month cycle, can provide margin gains of the size shown in Fig . 12, i .e. in excess of fi ve (5) percent rel ati ve to the use of discrete B₄C shim rods.

The Er₂O₃ and UO₂ mixture presents no chemical or physical problems in the fuel rods and is manufactured using the same technology as is used for all UO₂ PWR fuel rods.

Erbium oxide (Er₂O₃) has a high melting point (2355°C) so that when dispersed in UO₂ in small quantities (say, at about the 1 weight % or less level), it will have a minimal effect on reducing the melting point of the fuel. It is also physically stable at these high temperatures because of its relatively low vapor pressure. Also, physical properties, such as the thermal conductivity, of the resulting UO₂-Er₂O₃ pellets are not significantly different from 100% UO₂ for compositions at approximately the 1.5 weight % level or less.

Like other rare earth oxides (e.g., Gd₂O₃ and Dy₂O₃)

Er₂O₃ is chemically compatible with UO₂ tending to form a solid solution during fabrication of the dispersions of Er₂O₃ in UO₂ by the usual ceramic processes (e.g., sintering). U.S. Patent 3,257,285 suggests homogeneously combining metallic erbium, but not erbium oxide, and fuel for the single purpose of a "Triga" type, as opposed to a PWR, reactor fuel rod balancing fuel burnup and BA "burnout". Small additions of Er₂O₃ do not affect the sintering behavior of the UO₂ so that the normal fuel pellet fabrication process of compacting powder mixtures and sintering at high temperature (1650-1800°C) in a reducing atmosphere can be used.

The favorable chemical and physical stability of UO₂-Er₂O₃ mixtures is in sharp contrast to the behavior of additions of boron or any of its compounds to UO₂. They are either too volatile themselves (B₂O₃) or form volatile species by reaction with the UO₂ at sintering temperatures so that it is difficult to produce a target loading of burnable absorber. More important, at operating temperatures in-reactor, these volatile species can also form and as a result of concentration and temperature gradients will tend to migrate, thus changing the boron distribution. In summary, while boron would be otherwise an ideal neutron absorber to be dispersed in UO₂, it cannot be fabricated by conventional fabrication processes or operated in-reactor as a homogeneous mixture with UO₂ because of its adverse chemical compatibility and volatility

characteristics. Erbium added as the oxide is not volatile either during fabrication or in operation.

Claims

1. In a pressurized water nuclear reactor in which soluble boron reduces beginning-of-cycle excess reactivity, a reactor core for extending the fuel cycle while maximizing a local power distribution uniformity, uniform burndown, thermal margin and available incore heat transfer area and ensuring a negative moderator temperature coefficient of reactivity, which core includes fuel assemblies comprising:

a square array of n x n grid positions, where n = 14 to 20;

said array having a majority of grid positions occupied by fuel rods and other grid positions available for control rods and incore instrumentation;

at least some of said fuel rods including a homogeneous mixture of UO₂ and Er₂O₃, so that they, said fuel rods, act as a means for providing a low peak power to average power locally within said assemblies.

2. The fuel rods of mixed UO₂ and Er₂O₃ of claim 1 in which the mixture is in the approximate range of from 0.2% to 1.5% by weight of Er₂O₃ and the remainder is substantially all UO₂.

3. The fuel rods of mixed UO₂ and Er₂O₃ of claim 1 in which the mixture has been sintered at 1650ºC to 1800°C in a reducing atmosphere.

4. The fuel rods of mixed UO₂ and Er₂O₃ of claim 3 in which the mixture has been compacted before sintering.

5. The fuel assemblies of claim 1 in which the fuel rods of mixed UO₂ and Er₂O₃ are symmetrically arranged in grid positions.

6. The fuel assemblies of claim 5 in which diagonals of the square arrays are axes of symmetry.

7. The fuel assemblies of claim 5 in which water hole axes are axes of symmetry.

8. The fuel assemblies of claim 1 in which at least some of the fuel rods of mixed UO₂ and Er₂O₃ are in perimeter grid positions of the square arrays.

9. The fuel assemblies of claim 1 in which all of fuel rod grid positions are occupied by fuel rods of mixed UO₂ and Er₂O₃.