Крючков Фундаменталс оф Нуцлеар Материалс Пхысицал Протецтион 2011

Fuel fabrication processes

Uranium dioxide is the most extensively used and commercialized nuclear fuel for thermal and fast reactors.

Uranium dioxide advantages:

·high melting temperature (2780°С);

·chemical resistance to the main coolant types (light and heavy water, sodium, carbon dioxide);

·good compatibility with cladding materials (stainless steel, zirconium alloys) at operating temperatures;

·possibility of producing high-density pellets;

·acceptable radiation resistance at neutron fluxes of ~1014 n/cm2×s and fluences of ~1022 n/cm2, i.e. for 3 years;

·isotropic crystal lattice, which facilitates high-temperature sintering.

Uranium dioxide disadvantages:

·low heat conduction dropping abruptly as temperature increases. This accounts for sharp temperature differentials between pellet center and

periphery (DТ~1000–1500 оС);

·easy oxidation in air. It needs inert dry environment or vacuum, otherwise the pellet will get saturated with moisture and oxygen will be adsorbed by its surface layer. Moisture on the surface can cause hydrogenation of the cladding and failure of the fuel element;

·presence of oxygen moderates the neutron spectrum in a fast reactor and reduces the breeding ratio.

Steps in production of uranium dioxide pellets

1. Conversion of uranium hexafluoride into dioxide: a) “Wet” AUC process:

uranium hexafluoride is passed through aqueous solution of ammonium carbonate (NH4)2CO3, producing a solid insoluble precipitate of ammonium–uranyl–carbonate (AUC) – (NH 4)4UO2(CO3)3;

heat treatment of AUC at 550–650 оС, resulting in its thermal decomposition with formation of UO2 as fine powder;

b) “Dry” process:

uranium hexafluoride hydrolysis by steam at 150–300 оС to produce uranyl fluoride UO2F2:

41

UF6 + 2H2O → UO2F2 + 4HF;

UO2F2 pyrohydrolysis by steam and hydrogen at t 550 оC to form fine UO2 powder and hydrofluoric acid (HF):

UO2F2 + H2 → UO2 + 2HF.

The resulting fine powder of UO2 is made unfit for compaction by its very small particles (less than 0.6 μm). To obtain larger particles, the following operations are carried out:

2.Mixing with plasticizing agents;

3.Hydrocompaction: filling of a rubber mould; placement in a container with liquid, which is then pressurized (uniform compression), briquetting;

4.Granulation by briquette milling;

5.Annealing to remove plasticizers;

6.Cold molding to make pellets;

7.Pellet sintering;

8.Pellet quality control and sorting according to size, carbon content (plasticizers), and stoichiometry.

As regards the technology of producing mixed oxide (MOX) fuel, there

are three possible blends:

PuO2 + 238UO2, where Pu is taken from weapons materials;

PuO2 + 238UO2, where Pu is obtained by irradiated fuel reprocessing;

and

235UO2 + 238UO2, where 235U is taken from weapons materials.

MOX fuel production presupposes availability of two source materials: 238UO2 powder made from depleted or natural uranium; and

PuO2 coming from weapons or reactors, or 235UO2 from weapons uranium.

It is impossible to guarantee homogeneity of the (Pu, 238U) or (235U, 238U) mixture. Mixing of powder from different sources is the only stage to distinguish the production method of MOX fuel from that of uranium dioxide fuel. Mixture homogeneity is essential to reactor safety. As power rises, fissile isotopes will be the first to heat up. Doppler widening of capture and fission resonances takes place with the ensuing cumulative positive effect of reactivity. A fertile isotope takes more time to heat up, bringing about a negative Doppler effect of reactivity. This delay is the shorter, the higher is the mixture homogeneity. With poor mixing, the

42

positive effect of reactivity of fissile isotopes can cause power rise to an unacceptably high level, before the negative effect of reactivity of the fertile isotope can have its stabilizing action.

Fuel fabrication stages

1.Preparation of nuclear fuel (conversion of UF6 into UO2; powder production; pelletization and sintering);

2.Preparation of tubular claddings and end-pieces;

3.Preparation of fuel assembly components.

4.Fitting up of fuel rods: packing of tubular claddings with pellets; attachment of end-pieces; filling with helium; sealing of end-pieces; fuel rod quality control.

5.Putting together of fuel rods into assemblies; quality control; rig tests. Fabrication of fuel rods and assemblies is:

∙a precision process;

∙an automated quantity production process;

∙an important object of physical protection, accounting and control of nuclear materials.

It takes hundreds of thousands of components and millions of fuel pellets to build a reactor core.

Fuel use in nuclear reactors

A reactor is above-critical before its operation starts, but the reactivity margin is suppressed by special controls, such as absorber rods, boric acid in the coolant, burnable poison in fuel. Fuel burning and buildup of fission products (FP) cause the reactor to go subcritical (Кeff < 1). For the reactor to continue operating, Кeff should be raised above unity. The primary purpose of refueling is to restore the reactivity margin.

Another purpose is to flatten power density distribution for the highest possible energy generation and uniform fuel burnup.

These purposes may be attained by:

∙full or partial refueling;

∙rearrangement of fuel assemblies with different burnup in the core;

∙combining of the first two approaches.

Refueling of nuclear reactors can take the following forms:

Cyclic refueling, involving uniform fuel distribution and its complete replacement once the reactivity margin is exhausted.

The disadvantages of this approach include:

43

∙non-uniform power density distribution in the reactor;

∙quick burnup of the central core, with the reactivity potential of the peripheral fuel retained.

Partial cyclic refueling

It is only the fuel assemblies that have reached their burnup limit that will be unloaded from the reactor and replaced with fresh ones. The core is divided into a number of concentric zones. During each subsequent refueling, the burnt assemblies will be replaced with fresh ones in turns, one zone after another, from the center to the periphery.

The advantage of this approach lies in the same burnup of unloaded fuel.

A disadvantage of this approach is that fresh fuel will take time in moving towards the periphery. This results in impaired power density distribution, with its higher values found in the central zones.

Scattered refueling

The core is divided into groups of fuel assemblies, each including the same number of FAs, e.g., four. The assemblies replaced in the first refueling operation will all be itemized under No. 1. During the second refueling, this procedure will be repeated for assemblies No. 1 and 2, and so on. The distribution of fresh FAs will be uniform throughout the core, resulting in improved global uniformity of power density distribution.

Refueling “from the periphery to the center”

The core is divided into concentric zones containing the same number of fuel assemblies. The first unloading operation will involve the FAs of the central subzone with the highest burnup. They will be replaced by FAs of the second subzone, the place of the latter will be taken by assemblies of the third subzone, etc. The last subzone vacated will be filled with fresh fuel assemblies.

As a result, FAs concentrating in the center will have the highest burnup, i.e. they will be less reactive than the peripheral assemblies. The associated processes are depression of heat release in the center and reduction of the efficiency of reactor controls.

Modified scattered refueling

This method involves the following:

1.The peripheral ring of fuel assemblies is singled out, comprising, e.g., 1/5 of all the reactor’s FAs;

2.The central core area is divided into local groups, four FAs in each;

44

3. Assemblies No. 1 are removed from each group during the first refueling and are replaced by FAs from the outer ring. The vacated outer row is loaded with fresh FAs.

The advantage of this method lies in uniform power density distribution, without a surge in the center peculiar to “partial cyclic” refueling, or depression in the center characteristic of the “periphery-to center” process.

Refueling procedures

Refueling may take place:

∙after reactor shutdown and cooldown, with its vessel head removed;

∙after shutdown, without cooldown and vessel head removal;

∙at low or full power.

A full shutdown procedure is practiced at light water reactors. Once a

year, a reactor will be shut down for 4 to 6 weeks, its vessel head removed, the irradiated fuel assemblies unloaded, the remaining FAs reshuffled, and fresh fuel loaded. All these operations proceed under water.

In a fast reactor with liquid metal coolant, fuel is reloaded after reactor shutdown, with its vessel head left in place. Use is made of a rotary plug with a reloading mechanism. Two eccentric systems guide the mechanism to an appropriate FA so that it can grip the assembly by its top end-piece and move it into the in-pile storage at the core periphery. Fuel assemblies are kept for some time there to be later removed by a simpler mechanism during reactor operation.

Heavy water reactors, such as CANDU, can be refueled without shutting down the reactor or reducing its power. Fuel assemblies are placed in horizontal channels. The refueling process relies on the “sluice” principle. There is a reloading machine on either side of the reactor. Each machine has a sleeve, which is connected to either end of the fuel channel. Once the channel plug is removed, pressure in the machines and in the channel levels off, fuel is unloaded, and the channel is closed again. One of the machines inserts a fresh assembly at one end, while the other picks up the assembly pushed out at the opposite end.

RBMK reactors also feature on-load refueling. To this end, a reloading machine is used and the “sluice” principle is appli ed, as is the case with CANDU:

∙the reloading machine, filled with condensate, is joined to a channel;

∙pressure in the machine box becomes equal to that of the channel;

∙the depressurized channel receives cold condensate;

∙the spent FA is gripped and retrieved;

45

∙a mockup FA is inserted to check the channel for passability;

∙a fresh FA is loaded into the channel;

∙the channel is pressurized, pressure in the machine drops, and the machine is disconnected from the channel.

The reactors permitting refueling without power reduction and reactor depressurization, such as heavy-water CANDUs and water-graphite RBMKs, are the greatest proliferation hazards.

Discharged fuel assemblies are kept for 3 to 10 years in the on-site water pool to reduce their activity and residual heat.

The irradiated fuel pools are equipped with:

∙a water cooling system;

∙an ion exchange facility to remove radioactive substances and to clean the pool water;

∙a ventilation system to pass air through filters and to vent waste gas to the atmosphere.

Transportation of irradiated nuclear fuel

Irradiated fuel assemblies are carried in special casks by rail, motor and water transport. Such shipping casks weigh about 80–110 t, of which fuel accounts for merely 2–5 %. The rest is contributed by safety features.

A shipping cask looks as follows:

1.It is a large hollow thick-walled cylinder (with diameter of 2 m; height of 4–6 m, and wall thickness of 40 cm). It may be oriented either horizontally or vertically, and its structural material is steel, cast iron, or concrete.

2.Inside, it is lined with stainless steel for better corrosion resistance, which may have interlayers of neutron moderating material.

3.Its outer surface may be ribbed to increase the heat transfer surface area.

4.It contains metal racks for fuel assemblies. During transportation it is filled with coolant which will remove heat by natural convection or forced circulation.

5.Casks are made tight by strongly sealed lids.

6.Casks are fitted with an interior system for monitoring activity, heat release, temperature, and pressure and with an emergency decontamination system.

46

Technologies for reprocessing of irradiated nuclear fuel

The purposes of fuel reprocessing are:

1)to separate plutonium and uranium for recycling;

2)to separate fission products (FP) and transuranic elements as waste. Irradiated nuclear fuel is generated in the world at a rate of 7000 t per

year, while the existing facilities can reprocess annually about 5100 t of INF.

INF reprocessing methods

1.Aqueous (“wet”) methods:

∙extraction processes, with uranium and plutonium extracted from solutions by organic compounds;

∙precipitation processes, with poorly soluble uranium and plutonium compounds precipitating from solutions.

2.Non-aqueous (“dry”) methods:

∙pyrochemical processes, e.g., fluoride gas technology, with its principle lying in different volatility and sorption capacity of uranium and plutonium fluorides and FP;

∙pyrometallurgical processes, e.g., electrolytic refining, based on difference in transport of uranium, plutonium and FP in molten metals and salts.

Irradiated nuclear fuel is reprocessed using one of the above processes at a special radiochemical facility. Aqueous extraction processes are the best developed and proven technologies.

Main stages in the aqueous extraction technology (PUREX)

Disassembly and cutting.

1.Fuel assemblies are taken apart by cutting off end-pieces, cutting shrouds by disk saws, disassembling fuel lattices.

2.Fuel rods are cut, e.g., by guillotine shears or lasers in inert environment (nitrogen or argon).

Pre-oxidation (voloxidation).

Pre-oxidation of irradiated fuel takes place in oxygen at an elevated temperature. Uranium dioxide, UO2, transforms into uranium octaoxide,

U3O8:

3 UO2 + O2 → U3O8.

This gives rise to the following effects:

47

Fuel decompaction. The density of UO2 is tangibly different from that of U3O8: γ(UO2) = 11 g/cm3, γ(U3O8) = 8.3 g/cm3. Fuel increases its volume

by 30 %, becoming porous and loose; Change of crystalline lattice;

Intensive release of fission and tritium gas.

INF dissolution. Fuel is dissolved in boiling nitric acid, HNO3:

UO2 + HNO3 → UO2 (NO3)2 + NOX + H2O.

Zirconium and steel claddings will not dissolve. They are removed from solution and are treated as solid radioactive waste.

Preparation of INF solution for extraction proceeds as follows.

1.Clarification of solution:

∙filtration through cermet or porous polypropylene;

∙centrifugation with addition of coagulants.

2.Removal of volatile FP and fission gas from solutions:

∙air bubbling to remove iodine present as I– , IO3, which is then captured at silver nitrate (AgNO3) filters:

6 AgNO3 + 3 I2 + 3 H2O → 5 AgI + AgIO3 + 6 HNO3;

∙ozone bubbling to remove ruthenium, Ru4+:

Ru4+ + 2 O3 + 2 H2O → RuO4 + 2 O2 + 2 H2.

∙volatile ruthenium oxide (RuO4) is removed from gas in a reaction with NaOH;

∙removal of inert Kr and Xe by bubbling with gas sorption at zeolite or activated carbon at low temperatures.

Extraction is breakdown of a substance into two immiscible fractions: light organic fraction (TBP + diluent) and heavy aqueous fraction (acid solution of INF). A significant drawback of extraction is radiolysis of organic agents, i.e., decomposition under irradiation.

Separation of plutonium from uranium. Plutonium found in an INF

solution can be 3-, 4- or 6- valent. Separation of uranium and plutonium relies on the fact that U6+, Pu6+ and Pu4+ are easily dissolved both in the aqueous phase and in the organic phase, while Pu3+ is only slightly dissolved in the organic phase. When plutonium is restored to a condition in which it becomes 3-valent, it will completely pass into the aqueous

48

solution and will be removed from the process, whereas uranium will remain in the organic phase.

In separation of plutonium from uranium, hexavalent plutonium is first restored to a condition of tetravalency, and then to trivalency. Plutonium is converted from the hexavalent to tetravalent condition in a reaction with potassium nitrite, KNO2:

PuO2(NO3)2 + KNO2 → Pu(NO3)4 + KNO3 ,

i.e. Pu6+ is turned to Pu4+, which is then restored to Pu3+ in a reaction with Fe2+ compounds:

Pu4+ + Fe2+ → Pu3+ + Fe3+.

Iron gives up one valent electron to plutonium in a reaction with U4+:

Pu4+ + U4++ 2 H2O → Pu3+ + UO22+ + 4 H;

through electrochemical reduction of plutonium. An electric current is passed through the solution:

Pu4+ + e– → Pu3+;

UO22+ + 2e– + 4 H+ → U4+ + 2 H2O; and U4+ will act as another reducing agent.

When the organic phase is washed with a reducing solution, Pu3+ will go into the aqueous phase, while U will remain in the organic phase. In reextraction of the organic phase with a weak solution of nitric acid, uranium will pass into the aqueous phase (re-extract).

Stages of one extraction–re-extraction cycle

1.Fuel dissolution in nitric acid.

2.Extraction of uranium and plutonium compounds from solution using TBP. Uranium and plutonium pass into the organic phase.

3.Re-extraction with a reducing solution. Hexaand tetravalent plutonium turns into its trivalent variety and goes into the aqueous phase.

4.Uranium re-extraction from the organic phase with diluted nitric acid. Uranium passes into the aqueous phase.

49

Control for non-proliferation of nuclear materials at a reprocessing facility

A reprocessing facility is one of the most sensitive components of the nuclear fuel cycle in terms of proliferation resistance. The main problem here is control over plutonium, which is made even more complicated by a number of factors:

Large plutonium quantities. The existing reprocessing facilities are capable of treating about 1000 t of INF per year. One tonne of irradiated fuel from light water reactors contains 6–7 kg of p lutonium, which means that 6-7 t of plutonium can be put through the facility in one year.

High accuracy requirements. The Significant Quantity of plutonium, SQ(Pu), adopted by the IAEA is 8 kg. Suppose plutonium needs to be monitored at the reprocessing facility with an accuracy of 1 kg. With 7 t of plutonium passing though the reprocessing facility every year, the monitoring accuracy should be in the order of ~ 10–2 %. The realistic accuracy of Pu mass measurements is 0.1–1 %. The pe rmissible imbalance is about 0.1 %, which is close to the limit of measuring capabilities. As a result, inventory has to be taken several times a year, with the facility divided into material balance areas for location of a potential plutonium diversion point.

Plutonium is found in different phase states (solid, liquid, organic); it may be part of compounds with different valences, and is involved in periodic, continuous or semi-continuous processes.

The following factors are considered in assessing the extent to which plutonium compounds may be attractive for theft:

Density factor f1 defines the Pu content in its compounds. Factor f1 is treated as a function of NM volume containing 1 g of Pu. For Pu metal, this factor is assumed to be f1 = 1. The density of Pu metal is 19.8 g/cm3; i.e. its specific volume is I ~ 5×10–5 l/g. The datum point of function f1(Vуд) is 1, given Vsp=5×10–5 l/g. Other Pu–bearing materials have larger specif ic volumes and, hence, smaller values of factor f1.

Time factor f2 accounts for the time it takes for a group of specialists armed with modern equipment to convert a Pu–bearing material into the charge of a nuclear explosive device. It is assumed that plutonium metal can be turned into such a charge in a week’s time, i.e. the time factor for Pu metal is f2 = 1 with t = 7 days. For other Pu–bearing materials, the char ge production time is longer and the f2 value is smaller.

50