Hatano Y., Katsumura Y., Mozumder A. (Eds.) Charged Particle and Photon Interactions with Matter - Recent Advances, Applications, and Interfaces

Подождите немного. Документ загружается.

980 Charged Particle and Photon Interactions with Matter

The corrosion potential of the stainless steel at the ECP sensor unit decreases with the decrease

in the dissolved oxygen concentration in the RPV bottom drain line. There is a unique relationship

between the corrosion potential and the oxygen concentration, as shown in Figure 34.13. The cor-

rosion potential decreased sharply in the range of the dissolved oxygen concentration below 10 ppb.

However, it should be noted that the corrosion potential was measured at the ECP sensor unit loca-

tion whereas the oxygen was measured at the end of the sampling line far apart from the sensor

location.

200

400

–200

EPRI guidelines –230 mV

SHE

1000100101

–600

–400

Dissolved oxygen in reactor water (ppb)

Corrosion potential (mV

SHE

)

Figure 34.13 Correlation between corrosion potential and dissolved oxygen in bottom drain under hydro-

gen

water chemistry.

400 14

200

Corrosion potential

in RPV bottom drain

Normalized average MSLRM reading

–200

EPRI guidelines –230 mV

SHE

–600

–400

Main steam line radiation

0.0 0.2 0.4 0.6 0.8 1.0

–1000

–800

Dissolved hydrogen in feedwater (ppm)

Corrosion potential (mV

SHE

)

Figure 34.12 Behavior of bottom drain line corrosion potential against hydrogen injection rate.

Radiation Chemistry inNuclear Engineering 981

The corrosion potential behavior has a strong dependency on the ow rate (Ichikawa etal., 1992,

1994). The relationship between the corrosion potential and the dissolved oxygen (in Figure 34.6)

was obtained with a ow rate of 3.1 m s

−1

. In the slower ow rate region, the corrosion potential

decreases

faster with the decrease of oxygen.

34.2.5.6

hydrogen

ater

Chemistry s

ummary

Through the long-term verication program, hydrogen water chemistry control has been veried

as an environmental countermeasure against potential IGSCC issues for BWRs. In order to evalu-

ate the effectiveness of hydrogen water chemistry control on the protection of the vessel bottom

region, a high ow and in situ corrosion potential measurement device was developed and installed

in a branched sampling line from the RPV bottom drain line. In this program, 0.8ppm hydrogen

in feedwater was necessary to achieve a corrosion potential of −230mV (SHE) at the vessel bottom

region. The ow rate effect on the corrosion potential should be taken into account to predict corro-

sion

potential distributions in the BWR primary system.

34.2.6 developMent of an ecp SenSor for bwr applicationS

An experimental approach to directly measure the corrosion potential at BWRs is important to

verify the modeling approach. In order to measure the corrosion potential, ECP sensors are used as

reference

electrodes and working electrodes.

34.2.6.1

reference

electrode

reference electrode is absolutely necessary for the ECP measurement. It shows its own potential,

which is not affected by water chemistry or ow changes. In BWR systems, a reference electrode,

such as silver/silver chloride (Ag/AgCl) or iron/iron oxide (Fe/Fe

), has been used in high-

temperature

water.

Under

hydrogen water chemistry control conditions, a platinum electrode is used as a reference

electrode because it works as a theoretical hydrogen electrode. It also has a longer lifetime com-

pared with the conventional reference electrodes. Therefore, it has long been accepted that both

conventional reference and platinum reference electrodes can be used.

34.2.6.2 working

electrode

Stainless

steel or nickel-based alloys are used as working electrodes that behave similar to plant

structural materials. Attention should be paid to the initial value before the electrode is prelmed.

The plant structural material itself can be regarded as a working electrode when the ground poten-

tial

from the plant is measured against the reference electrode.

34.2.7 radiation effectS on corroSion in nuclear reactor SySteMS

The radiation effects on material corrosion in nuclear reactor systems is typically classied as a direct

radiation effect on materials and an indirect effect through the change of the water chemical environ-

ment. The former is often recognized as radiation damage by neutrons or heavy ions, and the latter is

comprehended as enhanced corrosion in liquid underirradiation. In this chapter, the relationshipbetween

the corrosion behavior and water radiolysis is discussed based on laboratory experimental ndings.

34.2.7.1 radiolysis

ffects

on the w

ater

nvironment

34.2.7.1.1

Production of Oxidant Species by Radiation

Oxidizing radiolytic products enhance material corrosion due to their contributions to the cathodic

reactions.

Anodic reactions M M ne: = +

+ −n

(34.1)

982 Charged Particle and Photon Interactions with Matter

Cathodic reactions O H e H O

: + + =

+ −

4 4 2

(34.3)

H O H e H O

2 22

2 2 2+ + =

+ −

(34.4)

OH e OH+ =

− −

(34.31)

HO H e H O

22 2

+ + =

+ −

(34.32)

O H e H O

22 2

− + −

+ + =

(34.33)

Under irradiation conditions, various kinds of oxidizing radicals or species are produced,

which contribute to the cathodic reactions and enhance the corrosion reaction. The distribution

of the yield of each species is dependent on the quality of radiation known as linear energy

transfer (LET).

When reducing species like hydrogen coexist in the system, the radiation effect does not

always enhance the corrosion but sometimes reduces the oxidizing species concentrations due to

the recombination effect. As a result, the cathodic reactions are suppressed and so is the corro-

sion reaction.

34.2.7.1.2

Corrosion Behavior by Radiolytic Oxidant Species

34.2.7.1.2.1

Stainless Steel Corrosion behavior of materials in reactor systems is due to the

concentrations of the radiolytic oxidizing species such as oxygen, hydrogen peroxide, and radical

species. Those concentrations are inuenced by radiation. The oxidizing power of the system is

described by an ECP on the material surface, which is a function of not only the chemical species

concentrations

but also the oxide lm and hydrodynamic conditions.

the case of stainless steel corrosion in high-temperature water, both the general corrosion rate

and the SCC susceptibility are high at high corrosion potentials. Figure 34.14 shows the radiation

0.2

0.3

0.4

200 ppb H

31 ppb Cu

–0.1

0.0

0.1

200 ppb H

injection conditions

–0.4

–0.3

–0.2

corr

SHE

)

1 10 100 1000 10,000

–0.7

–0.6

–0.5

Dissolved oxygen (ppb)

Figure 34.14 The radiation effect on ECP of stainless steel.

Radiation Chemistry inNuclear Engineering 983

effect on the ECP of stainless steel (Andresen, 1992). Figure 34.15 also shows the gamma-ray effect

on the ECP of stainless steel, which increases under excess oxygen conditions due to the production

of hydrogen peroxide, and decreases under excess hydrogen conditions due to the recombination

reaction

of the oxidizing species (Ichikawa etal., 1987).

The

same tendency is shown in the case of SSRT (slow strain rate test) results to evaluate the

IGSCC susceptibility of sensitized Type 304 stainless steel, as shown in Figure 34.16 (Saito etal.,

1990). The IGSCC ratio on the fracture surface of the test specimen is reduced by gamma irradia-

tion

under hydrogenated conditions.

Figure

34.17 shows that the critical corrosion potential for IGSCC susceptibility is understood

in relation to ECP and it is almost the same regardless of the gamma irradiation (Saito etal., 1997).

However, for general corrosion, iron dissolution rates from stainless steel in high-temperature water

are accelerated by gamma-ray irradiation, as shown in Figure 34.18 (Ishigure etal., 1980). Figure

34.19 shows that the metal dissolution rate from stainless steel is also accelerated by hydrogen peroxide

addition

and gamma-ray irradiation (Hemmi etal., 1994).

500

Gamma-irradiation

Non-irradiation

corr

(mV

SHE

)

0 5 10 15 20 25 30 35 40

–500

Loop operating days (day)(a)

500

Gamma-irradiation

Non-irradiation

–500

0 5 10 15 20 25 30 35 40

Loop operating days (day)

(b)

corr

(mV

SHE

)

Figure 34.15 The gamma-rays effect on ECP of stainless steel. (a) Oxygen excess condition and (b) hydro-

gen

excess condition.

984 Charged Particle and Photon Interactions with Matter

34.2.7.1.2.2 Other Materials According to the above discussion, it seems acceptable to simulate

the irradiation effect by adding suitable amounts of hydrogen peroxide to the system based on the

nding that the oxidizing power generated by radiation is in most part due to the accumulated hydro-

gen peroxide concentration. The general corrosion rate of Alloy X750 (Ni-based alloy) and Stellite #6

(Co-based alloy) is reported to be enhanced by four times under the BWR core simulated condition

with the addition of hydrogen peroxide (Hemmi etal., 1994). Zircaloy is used as fuel cladding mate-

rial for both BWR and PWR reactors. It is reported that perhydroxy radical (HO

•

or O

•−

) produced

in the high radiation elds in the core plays a role in zircaloy corrosion (Hurst and Tyzack, 1974).

100

Gamma-irradiation

<1 0 0 0

IGSCC ratio (%)

0.1 0.32 0.56 0.1 0.32 0.56

Conductivity (μS cm

–1

)

(a) (b)

Non‐irradiation

Figure 34.16 SSRT (slow strain rate test) experiment to evaluate IGSCC susceptibility of sensitized Type

304 stainless steel, sensitized at 650°C for 3h. (a) BWR normal water chemistry condition at 250°C. (b) BWR

hydrogen

water chemistry condition at 250°C.

Gamma-irradiation

Non-irradiation

IGSCC ratio (%)

No IGSCC

–800 –600 –400 –200 0 200 400

corr

(mV

SHE

)

Figure 34.17 Critical corrosion potential of sensitized Type 304 stainless steel for IGSCC susceptibility at

the

SSRT speed of 4 × 10

−7

−1

. Sensitized condition is 24h at 620°C.

Radiation Chemistry inNuclear Engineering 985

34.2.8 radiation effect on MaterialS

It is widely known that direct neutron irradiation effects to the structural materials should be considered

in the core region. Austenitic stainless steel with solution heat treatment for an IGSCC countermea-

sure sometimes shows IGSCC under neutron irradiation. This is called irradiation-assisted intergranular

stress corrosion cracking (IASCC). Although the mechanism is not yet fully understood, it is speculated

that irradiation-assisted segregation in grain boundaries is a cause of corrosion property changes.

SCC susceptibility of austenitic stainless steel under neutron irradiation is shown in Figure

34.20 (Andresen etal., 1989). From the high-temperature water SSRT result with 32ppm of dis-

solved oxygen, the IASCC susceptibility increases with more than 5 × 10

−2

(E > 1 MeV) of

neutron uence and becomes signicant with more than 2 × 10

−2

(E > 1 MeV) of neutron

uence.

250

300

200

100

150

Gamma-irradiation

Non-irradiation

Crud concentration of Fe (ppb)

0 50 100 150 200 250 300

Time (h)

Figure 34.18 Acceleration of iron dissolution rate from Type 304 stainless steel in high-temperature water

250°C under 200

ppb

addition, gamma-irradiation

No H

, non-irradiation

0.1

4 6 8 10 12 14 16 18 20 22

0.01

Fe, Ni, Cr concentration (ppb)

Loop operating days (day)

Figure 34.19 Effect of gamma-rays and H

addition to the acceleration of iron dissolution rate from

Type

316 stainless steel in high-temperature water containing 400 ppb of O

and 60 ppb of H

at 270°C.

986 Charged Particle and Photon Interactions with Matter

The essential countermeasure to this level of radiation damage must be through material improve-

ment. However, research may eventually provide chemical control mitigations that will also work

increase corrosion resistance.

34.2.9 concluSion

The evaluation of the corrosion environment in BWRs is discussed from the viewpoint of both

modeling and experimental approaches. The comparison between measurement and calculation is

signicance for establishing the countermeasures for plant aging issues.

Corrosion

potential control is essential in reducing the IGSCC susceptibility of BWR primary

loop components. Two basic approaches for reducing the corrosion potential are reviewed. One is

cathodic reaction control and the other is anodic reaction control.

By applying the latest radiolysis and ECP models, normal water chemistry and hydrogen water

chemistry

control can be well evaluated.

34.3 radiation-induCed proCesses in the

spent

uClear

Fuel repro

Cessing

Increased human populations and industrialization are causing increased energy demands at the

same time that concerns about rising carbon dioxide concentrations in the atmosphere have made

fossil fuels less attractive. Opposition to nuclear power has decreased with a renewed interest being

expressed in many countries (Boullis, 2008). However, among the reasons cited by opponents against

nuclear energy are difculties with the disposition of nuclear waste. About 10,500t of spent fuel are

already discharged yearly from more than 400 nuclear reactors (IAEA, 2004). The rate of spent fuel

generation could reach 15,000t by 2050 (Laidler, 2008), and this spent fuel contains appreciable

quantities of actinides and ssion products, some with very long half-lives. For example, 1t of UO

fuel, after 30GW days of burnup, contains 950kg U, 9kg Pu, 75 g Np, 140g Am, 47 g Cm, and 31kg

ssion products (Benndict etal., 1981). With half-lives ranging from 433 years for

241

Am to 2.4 mil-

lion years for

237

Np, the alpha-emitting actinide nuclides determine the radiotoxicity of the waste

100

IASCC threshold

High stress

Low stress

(ppm)

γ (Rh

–1

)

32 —

0.2

0.2 —

0.005 —

1E19 1E20 1E21 1E22

Neutron ux E > 1 MeV (n cm

–2

)

Sensitivity of IASCC (IGSCC ratio (%))

2×10

Figure 34.20 SCC susceptibility of Type 304 stainless steel under neutron irradiation at 288°C.

Radiation Chemistry inNuclear Engineering 987

for periods of time on a geological scale. To address this, research programs in several countries are

investigating the partitioning of these radionuclides and their transmutation to short-lived isotopes

signicantly reduce the long-term hazard (OECD-NEA, 1999).

Aqueous

solvent extraction is the most mature of these partitioning strategies, beneting from

more than 60 years of research and experience at the industrial scale (Mathur etal., 2001). As cur-

rently performed, an alkane solution of tributyl phosphate (TBP) is used to complex and extract

U, Np, and/or Pu from nitric acid solutions of dissolved spent fuel. The process is called PUREX

(Plutonium Uranium Rening by EXtraction) and remains successful because it provides high

yields of the desired elements and high separation factors from the undesired elements, while pro-

ducing

only a small amount of secondary waste.

Looking

to the future, many countries are currently investigating solvent extraction processes

for the separation of the minor actinides Am and Cm from dissolved nuclear fuel. This separation

would both decrease the radiotoxicity of waste prior to deep geological repository disposal and

produce additional energy by burning these elements in a reactor. While none of the advanced pro-

cesses have yet been implemented, their development has reached demonstration tests at the labora-

tory scale (Nash etal., 2006). The eventual application of such a partitioning strategy will result in

natural uranium savings, uranium enrichment cost savings, simpler and safer waste disposal, and

provide

the potential for the recovery of other useful elements from spent fuel (Boullis, 2008).

The

new extraction schemes vary in their details, but following the traditional PUREX extraction

all would extract the minor actinides using hard donor (oxygen containing) ligands. Unfortunately,

undesirable short-lived lanthanide ssion products are also extracted and they are neutron poisons

that must then be separated from Am and Cm, prior to incorporating the actinides in fuel. This

second extraction employs soft donor ligands (usually nitrogen containing) capable of perform-

ing this difcult separation. Since these new complexing agents will be used in highly radioactive

applications, the effects of radiation chemistry on their performance must be understood prior to

their reliable application. Based on PUREX experience, the main effects may include decrease in

extraction efciency due to decomposition of the ligand, decrease in separation factors due to the

accumulation of radiolysis products that are also complexing agents, and deteriorated phase separa-

tion performance due to ligand or diluent radiolysis products (Pikaev etal., 1988). In addition, the

irradiation of aqueous nitric acid produces reactive species that can alter the oxidation states of the

metals to be complexed, affecting their extraction efciency (Pikaev etal., 1988). This chapter will

examine the conditions and reactive species to be expected in the irradiated biphasic organic/aque-

ous environment of the solvent extraction process, and review what is known about radiolysis effects

selected ligands and metals during solvent extraction for nuclear reprocessing applications.

34.3.1 radiolytically produced reactive SpecieS in the biphaSic SySteM

34.3.1.1 produced species in the aqueous phase

The direct radiolysis of ligands in a solvent extraction solution occurs in proportion to their abun-

dance as constituents of that solution. Since ligands are normally present in millimolar concentra-

tions, the diluent absorbs most of the radiation energy. Reactive species created by diluent radiolysis

are largely responsible for the extent and nature of the degradation of ligands and changes the physi-

cal characteristics of the solvent. The direct radiolysis of alkanes and aqueous nitric acid results in

the formation of electronically excited states, free radicals, and ions in spurs along the track of the

incident particle. These reactive species undergo recombination to create molecular species, or they

diffuse away from their point of origin to react with solutes in the bulk solution. The probability

of recombination or diffusion depends on the LET (eV nm

−1

) of the incident particle, which for x,

γ, and β

−

is much lower than for alpha particles. Reactive species that escape these spurs are the

source of the secondary radiolysis reactions that degrade ligands or react with metal ions, and this

phenomenon

begins to occur at about 100

after the initial event.

988 Charged Particle and Photon Interactions with Matter

For water irradiated by low LET particles such as x-rays, gamma-rays, or electrons (β

−

particles)

generated mainly by ssion product decay, the species in Equation 34.34 are produced with the

yields

(G-values in μmol J

−1

) shown in brackets (Buxton etal., 1988):

O [ . ] [ . ] [ . ] [ . ] [ . ] [ . ]0 28 0 27 0 06 0 07 0 27 0 05

2 2 3

i i

OH e H H O H O

+ + + + +

− +

(34.34)

The most reactive species produced are the oxidizing hydroxyl radical (

•

OH) and hydrogen peroxide

), and the reducing aqueous electron (e

−

) and hydrogen atom (H

•

). Massive, highly charged

particles such as alpha-particles (He

) generated mainly by actinide decay have short ranges and

high LET (156eV nm

−1

for 5MeV He

) (Pastina and LaVerne, 1999) and therefore deposit energy

in closely spaced overlapping spurs. This results in high localized concentrations of reactive spe-

cies, many of which undergo recombination before they can diffuse into the bulk solution. Thus,

higher yields of molecular species and lower yields of radicals are found. Using the data of Lefort

and Tarrago (1959) for 5.3MeV

210

Po alpha-particles, the water radiolysis equation may be rewritten:

O [ . ] [ ] [ . ] [ . ] [ ] [ . ]0 05 0 06 0 15 0 16

2 2 3 2

i i

OH x e H H O x H O H

+ + + + +

− +

(34.35)

The yields for the aqueous electron and hydronium ion are not given because the measurements

were made in acidic solution. It is seen that the yield of the molecular product hydrogen peroxide

due to alpha radiolysis of water is double that for gamma-radiation. Comparatively, few studies have

been

done using alpha-radiolysis, and this chapter will cover mainly low LET effects.

Despite

an initially equal production of oxidizing and reducing species, under the conditions of

nuclear solvent extraction, the aerated, nitric acid system will be predominantly oxidizing due to

fast

scavenger reactions. The aqueous electrons and hydrogen atoms are scavenged according to

e H H M s Buxton et al

− + − −

+ → = ×

k 2 3 10 1988

10 1 1

. ( ., )

(34.36)

e O O M s Buxton et al

− − − −

+ → = ×

2 2

9 1 1

1 9 10 1988

k . ( ., )

(34.37)

e NO NO M s El liot,

− − − − −

+ → = ×

3 3

2 9 1 1

9 7 10 1989

k . ( )

(34.38)

H O HO Gordon et al+ → = ×

2 2

2 1 10 1964

•

k . ( ., ) (34.39)

i i

HO H O p Bielski et al

a2 2

4 8 1985↔ + =

+ −

K . ( ., )

(34.40)

This leaves H

and the strongly oxidizing

•

OH radical as the most important species in solution.

The

•

OH radical reacts with organic solutes by hydrogen abstraction, addition to unsaturated carbon

bonds, or with organic and inorganic solutes by electron transfer reactions. Rate constants for

•

radical reactions have been tabulated (Dorfman and Adams, 1973; Buxton etal., 1988). The

•

radical produced in Equation 34.39 is also an oxidizing agent, although its reactions have not been

well

studied.

Additional

reactive species are created by nitric acid radiolysis. A thorough review of the radia-

tion chemistry of nitric acid is given by Katsumura (1998). Among the most important species, the

oxidizing

•

radical is produced indirectly by the reaction of the

•

OH radical with undissoci-

ated nitric acid and directly by radiolysis of the nitrate anion dissociation product of nitric acid

(Katsumura

etal., 1991):

Radiation Chemistry inNuclear Engineering 989

i i

OH HNO NO H O M s+ → + = ×

− −

3 3 2

7 1 1

5 3 10k .

(34.41)

NO NO

aq3 3

− −

(34.42)

The reactions of the

•

radical are similar to those of the

•

OH radical, including hydrogen atom

abstraction, addition, and electron transfer reactions, although with decreased electrophilic charac-

ter and lower reaction rates (Neta and Huie, 1986; Shastri and Huie, 1990). Rate constants for its

reaction

with numerous species have been compiled (Neta etal., 1988).

acidic solution, the NO

2−•

radical product of Equation 34.38 will protonate (Grätzel etal., 1970):

NO HNO H NO p a p a

3 2 3 1 2

4 8 7 5

− −

↔ ↔ = =

i i i

K K. , .

(34.43)

The

product H

•

decays to produce the

•

radical (Løgager and Sehested, 1993):

H NO NO H O s

2 3 2 2

5 1

7 10

i i

→ + = ×

−

(34.44)

This species is less reactive than the

•

radical, but has been shown to add to unsaturated carbon

bonds, or to carbon-centered radicals to produce nitrated derivatives of the original compound

(Eberhardt, 1975; Olah etal., 1978; Dzengel etal., 1999; Ershov etal., 2007). Perhaps more impor-

tantly, the addition product of these two nitrogen-centered radicals decays to produce nitronium ion,

powerful nitrating species (Mincher etal., 2009b):

i i

NO NO N O M s Katsumura et al

2 3 2 5

9 1 1

1 7 10 1991+ → = ×

− −

k . ( ., ) (34.45)

N O NO NO Sworski et al

2 5 2 3

1968→

+ −

+ ( ., ) (34.46)

Another important product of nitric acid radiolysis is nitrous acid. It is produced by direct and

indirect

effects:

HNO O HNO Nagaishi et al

3 2

1994

+ ( ., ) (34.47)

NO O NO Daniels,

3 2

1968

−

( ) (34.48)

NO H HNO p Lammel et al

a2 2

3 2 1990

− +

+ ↔ =K . ( ., )

(34.49)

i i

NO NO N O M s Gr tzel et al

f2 2 2 4

8 1 1

4 5 10 1969+ → = ×

− −

k . ( ., )ä (34.50f)

reverse

s Gr tzel et al= ×

−

6 0 10 1969

3 1

. ( ., )ä

(34.50r)

N O H O HNO HNO M s Olah et al

2 4 2 2 3

1 1

18 1978+ → + =

− −

k ( ., )

(34.51)

This species is important because of its affect on actinide oxidation states in irradiated aqueous

nitric acid solution. This topic is discussed in more detail in Section 34.3.4.1. Its maximum concen-

tration is limited by radiolytically produced H

, as shown (Bhattacharyya and Veeraraghavan,

1977;

Vione etal., 2003):