ML073040033

Public Meeting Slides for Meeting with Dominion Nuclear

ML073040033
Person / Time
Site:	Millstone, Surry, North Anna   (DPR-032, DPR-037, NPF-004, NPF-007, NPF-049)
Issue date:	11/01/2007
From:	Dominion
To:	Office of Nuclear Reactor Regulation
Jervey, Richard 301-415-2728
Shared Package
ML072831359	List: ML072890664 ML073040030 ML073040033
References
GSI-191
Download: ML073040033 (28)

Text

Gaomingm, GSI-1 91 Chemical Effects Testing Dominion Resources and AECL October 11, 2007 02003 Domnion Agenda m

w Introductions 0

Purpose 0

Chemical effects assessment and testing approach Conclusions 0

Q&A Other Issues 02003 Donirdon Dominion/NRC K. Basehore-Dominion D. Guzonas - AECL K. Basehore-Dominion Dominion/NRC K. Basehore-Dominion D'Dominion

.1

Purpose To provide the Dominion-AECL approach for chemical effects assessment and testing for resolving GSI-191 concerns for:

" Millstone Power Station Units 2 and 3

" North Anna Power Station Units 1 and 2

• Surry Power Station Units 1 and 2 0203 Doi*noon Dtminion.

.2

kECL Approach to Chemical Effects -

Assessment and Bench-top Testing Outline

" Review of Existing Data

" Aluminosilicate Formation

" Methodology for Release Calculations

• Precipitation

" Detailed Calculations 2

page 1

AECL Review of Existing Data 3

AECL Review of Existing Data US NRC sponsored five long term (30 day) generic ICET tests as well as thermodynamic modeling of chemical speciation - these were thoroughly reviewed PWR Owners Group conducted short term (24 h) single-effects tests under PWR sump water conditions Surveyed scientific'literature on aluminum corrosion and precipitation, formation of aluminosilicate species, the precipitation of calcium phosphate species, as well as literature related to dissolution of insulation materials 4

a ' /rNI A JEJ page 2

Concentrations of major species measured in solution in ICET Tests Data for [Na] divided by 100 to facilitate comparison 5

Weight Loss(-) or Gain of Al Coupons after 30 days Test Number Fibreglass CaI-siilFibreglass Coupon Location submerged unsubmerged I

1 p110 2

5 pH 8.5, borax pH 10,i noTSP pH7, TSP

-25%

-0.2%

0.48%

0.1%

3 4

pH 7, pH 10, TSP no TSP

-2.9%

U.1*%*

0%6 0.1%

0.1%________

0.15%____

6

'A.64 CL,, W L, ý page 3

identified during the ICET TESTS Test ID Deposit Formula tincalonite Na 2B4O.5H20 borax Na2BO,(OH)4-8H20 ICET 1 unknown Compound containing Al, B, Na, C032" unknown Compound containing Na, B, Al ICET2 calcium phosphate (hydroxyapatite?)

Cas(P0 4) 30H?

tobermorite Ca2.(Si 30 7.(OH), d(H 20) calcite CaCO3 ICET 3 sodium calcium hydrogen carbonate phosphate hydrate (CasH2(POJ 6.H20.NaHCO3.H20) lithium calcium hydrogen carbonate phosphate hydrate (CaH.2(PO4),.H20.Li2CO3.H 20) calcium phosphate (hydroxyapatite?)

Cas(PO 30H?

tobermorite Ca2 *(Si 3OTs(OH), s)(H20)

ICET 4 calcite CaCO3 ICET 5 unknown Compounds containing 0, Na, Al, C, Ca Mg and Si 7

Concentration of Selected Elements in Water Samples taken from the ICET Tests, Test I

Test Test 3

Test 4

Test 5

PH 10 TSR) 7 (TSP))

3TSP) 0.5 Mlaxinmum conrenitration in water samples (ml/L) 3800 1JtN

'51

( AI 0

• 2550 '

.1 0.

3390 N5)

(5.5)

Nir 1st -

Cu i

Zn NMg 0'

(1.8 0.2

0. 1.

3.5 Si Ca Na

( 15 I

5500 9(

I 9(0og

)oo 100 50 11500 12 34 1400 P-0.7 S=r 110.3 (0.3)

'5 (Y9 0.8 hd 2860 54

(

Ic) nlr a

MC W

page 4

hermodynamic Modeling (Jain et al.)

" Preliminary modeling, carded out before first ICET results were available, used input values from peer-reviewed literature (corrosion/dissolution rates) and ICET test plan (surfaces areas, water composition)

" Assumed system was in thermodynamic equilibrium

- no kinetic information was included

- Most oversaturated phase allowed to precipitate Reactions limited to those materials used in ICET tests and excluded the uptake of CO2 from air 9

• Thermodynamic Modeling (Jain et al.)

At pH 10, various amounts of silicate species were predicted

.to form over time;

- CaSi03,

- Ca 2Mg5Si8022(OH) 2

- Zn 2SiO4

- Si0 2

- NaAISi308 Fe3Si0 4010(OH)2

- Ca3FeSiO 3O12

- ZnFe 2O4 However, silicate phases not observed to form in ICET tests

- while these silicates are the thermodynamically stable phases, their formation is kinetically slow

- Simulations repeated with formation of silicates suppressed 10 page5 5W L _

L'

hermodynamic Modeling (Jainet al.)

Revised simulations for ICET Test 1 gave reasonable predictions of aluminum and calcium concentrations in solution for the first 720 h

- after this time the model overpredicted the concentrations

- attributed to the passivation of the surfaces Model also overpredicted the concentration of Si in solution at all times Model also predicted the formation of Fe(OH) 2 after 148 h and Zn(OH) 2 after 32 h 11 Major Results of Review

1. The ICET tests clearly show that, at the pH values studied, aluminum corrosion can give rise to the formation of an aluminum-bearing precipitate

a. The data show that:

1. Boric acid significantly increases the corrosion rate of aluminum

2. Aluminum corrosion may be inhibited by species present in the sump environment (e.g., phosphates, silicates)

3. The precipitate formed included boron, which affects the mass or flocculation properties of aluminum-bearing precipitate formed. In particular, boron increases the "solubility" of aluminum hydroxides

2.

For the surface areas of materials used in these tests, only low concentrations of iron,, nickel, magnesium and zinc dissolved into the simulated sump water, and these species did not lead to the formation of detectable amounts of precipitates 12 page 6 v

Major Results of Review

3.

Silicon and calcium can be released into the sump solutions from dissolution of fibreglass and cal-sil

a. If TSP is present, precipitates containing calcium and phosphate, or calcium, phosphate and carbonate, can form. In the absence of TSP, the calcium and silicon do not lead to the formation of detectable chemical precipitates

b. Release of calcium from fibreglass is much lower than that from cal-sil

4. Bare concrete is not a significant source of calcium in solution under these conditions

5. Thermodynamic modeling alone cannot properly predict the identity or quantities of precipitates formed under PWR sump conditions; kinetic factors are very important 13 Major Results of Review

6.

Based on AECL's assessment of available data, no evidence of direct chemical effects from paint debris However, radiolysis of organic compounds leaching from paints and coatings will lead to the formation of carbohate species, which will lower the pH of the sump solution

7.

While WCAP-16530 suggests that sodium aluminum silicate is a possible precipitate, a review of the literature on the thermodynamics and kinetics of aluminosilicate formation suggests that this is unlikely under PWR post-LOCA sump water conditions 14 A, I\\:ITCL WL

\\

page 7

Aluminosilicate Formation 15 Solubility of AI(OH) 3 w

I, o

0 2

z 0

0 6

5 4

3.

" Amphoteric Aluminium hydroxide precipitation favoured in the pH range 4 to 9

" Above pH 9 and low Al concentrations, dominant species in solution is Ai(OH)4-

• Wefers, K and Bell, G.M. [1972). Oaddes and Hydroaides of Aluminium. Technical Paper no. 19, Alcoa Research Laboratories 2

1 0 1 2 3 4 5 6 7 8 9 10 II 12 pH 16 Ar AEC'L' WL page 8

Solubility of Nepheline Glass, NaAISiO 4 450 400 E 350 C 3 00 w 250 200

.150 100" 500 a

-A Al Ar-Na

" In alkaline solutions, a high pH generally increases the solubility of aluminosilicates

" In acidic solutions, increasing acid concentration destroys the framework of aluminosilicates and increases the solubility 0

2 4

6 8

pH(25*C)

• Near neutral pH, nepheline has Athe lowest solubility; this behavior is also observed from 1 0 1 2 14 the solubility data for jadeite 10 1 14 and albite glasses [Hamilton, JR.P Brantley, S.L., Pantano, C.G., Criscenti, LA and Kubicks, JOD.

[2001]. Dissolution Of Nepheline, Jadeite, and MJite Glasses:

Toward Better Models for Ajuminosillcate Dissolution. (3eochimica et Casmcchimica Acta, 65, 3683).

17 Aluminosilicate Formation In the pH range 6.8 to 9.3 at 250C, the major dissolved silicon species is H4SiO4 When aluminium and silicon species are both present in aqueous solution, polymerization of aluminium and silicon similar to the polymerization of silicon species can occur to form various aluminosilicates Depending on Al/Si ratio and solution pH, aluminosilcates such as feldspars, nepheline and zeolites can be produced in which the molar ratio of Al/Si ranges from 0 to 1 18 page 9

'recipitation Zone of Sodium Aluminosilicate at 250C and 0.89 M Hydroxide

-1.5

-2.0 E -2.5 S-3.0

-3.5 precipitation no precipitation

" pH higher than ICET tests

" however, dependence of solubility of aluminosilicate is weak at pH<1M

" Relatively low [Si] and [Al]

and low weak alkali concentration,make sodium aluminosilicate precipitation unlikely 19

-4.0

-3.5

-3.0

-2.5

-2.0

-1.5

-1.0 log[Si]m olAg Park, H. and Englezos, P [2001]. Precipitation Conditions of Aluminosilicate Scales in the Recovery Cycle of Kraft Pulp Mills.

Pulp & Paper Canada, 102, 20.

luminosilicate Formation - Conclusions Based on a review of the effects of pH, temperature, and Al and Si concentrations on the solubility and precipitation of aluminosilicates, it is concluded that:

- Amorphous phases (which have much higher solubilities than crystalline phases) are generally the precursors in aluminosilicate precipitation

- Increasing temperature increases the solubilitybut decreases the induction time

- High pH increases the solubility but decreases the induction time to precipitate sodium aluminosilicates

- High supersaturation ratio decreases the induction time for precipitation

- The relatively low Si and Al concentrations and low alkali concentration means it is unlikely that sodium alum inosilicates would precipitate under sump water conditions 20 A-ML CL,-

page 10

Methodology for Release Calculations 21 Release Calculations

• Basic methodology outlined in WCAP-1 6530 used to calculate mass of aluminum released

• Rather than use WCAP-1 6530 release equation, data from WCAP-16530 and other sources used to develop release equations 22 22L page 11

. Solution Composition Temperature C) Corrosion rate (g1m 2 h)

Reference 0.28 M(3000ppm) B+0.15 M NaOH (3450 ppm Na) 55 0.35-0.61 Grless and Bacarella [19691 100 14.0-18.0 Not described 90 23.9 Nlyogl et al. (19821 pH --9.2 with NaOH, borated 90 1.45 PlIppo et al [1997]

pH =10 with NaOH. borated 0.012 Borated alkaline containment water at pH 10 60 0.986 90 1.89 Jain et al., [20061 110 2.21 pH 10 with NaOH 60 0.06 ASM [19991 ICET Test 1 Data 60 0.73 Dallman et al. L2005a1 WCAP-16530 data Lane et al. [20061 pH4 88 0.56 pH 8 88 2.68 pH 12 88 60.1 pH 4 130 5.4 pH8 130 23.7 pH 12 130 200 WCAP-7153A data 1

Bell et al. [19751 pH 7 99 0.078 pH8 99 2.2 pH9 99 12.96 pH 10 99 365 23

~Al Corrosion Data

• pH, temperature and time dependencies of the corrosion rates evaluated separately and then combined at the end Allows better comparison with existing literature data on aluminum corrosion 24 A' AL L page 12

pH Dependence Three data sets used to assess pH dependence:

- ICET data from Tests 1, 3 and 5

- data from WCAP-16530 at 88-C

- data from WCAP-16530 at 130-C Data normalized such that the largest corrosion rate in each data set was set to unity 25 pH Dependence 1.4

• WCAP0MD I

-c I.

-Expmwe~a Bw~ Fd 2 S0.6 0.4 Y=3E-078""

0.2 0.0 1 0

6 7

8 9

10 11 12 13 14 pH

.8 1.6 E

1.5

.4 Cr

.2 Also shown are data for the pH dependence of the aluminum corrosion rate in non-borated water [ASM, 19991 26 All., AW'L ECL page 13

)

pH Dependence

• All data show a similar trend of increasing corrosion as a function of pH

- similar to that observed in the absence of boric acid

• Increase in corrosion rate mirrors the increase in solubility of aluminum hydroxides 27 pH Dependence

• Exponential function used to fit the three data sets measured in borated water Exponential dependence of Al corrosion rate on pH reported in some studies Shatalov, A.Y. Dokiady Akad Naak, 86, 775,1952.

McKee, A.B.; Brown, R.H., Corrosion, 3, 595, 1948.

Vujicic, V.; Lovrecek, B., Surf. Technol. 25, 49,1985.

Function used to fit the data:'

Corrosion Rate (pH) = 3 x 10-7exp(1.3947-pH) 28 page 14

-MCL WL

(

.Temperature Dependence

• Similar approach used to develop expression for temperature dependence

• Rates of chemical reactions typically exhibit Arrhenius behavior, e.g.:

Corrosion Rate (T) = A-exp(-E/k-T)

- E = activation energy for the reaction

- A = frequency factor

- T = temperature in K

- k = Boltzmann's constant

• Plot of corrosion rate vs 1/T should show exponential behavior 29

• S*

Temperature Dependence Data sets with values at more than one temperature plotted as normalized (at 902C) corrosion rate vs 1T

- Reasonable fit observed Resulting temperature dependence:

Corrosion Rate (T) = 2x10 7. exp(-6301.1 T-1) scaled to pass through average corrosion rate at 90-C 30 page 15 A

ll

Temperature Dependence (0

IN OM2D 0,025 0D030 0.0035 1,7(K) 31 Final Scaling Corrosion Rate vs Temperature at pH 10 Fit scaled to pass through the average corrosion rate at 90-C.

Data used in fit plus average corrosion rate from ICET Test 1 100 o ICET I a VKJP.16530 0 Gdessand Bamaclla o Cal2atdRt(frmFigue4)

A 0

U-Uor(Oa~k Ridge)

E 0o is 0

0 0.11 26 40 60 80 100 120 140 160 Temperature CC) 32 page 16

Al Release Rate' Time dependence of Al release rate

- only the same as the Al corrosion rate if all the Al released by the corrosion process enters the solution 450 400 y= 74.8x - 17.5 o

350 R2 0,946 300 E

150 1001 0

1 2

3 4

5 6

Sqaure Root of Time (d 1

4

)

" Corrosion rates often

.exhibit a parabolic behavior Rate = a.t1"2

• Solid line is a linear least squares fit to the data 33 Release rate decreases with increasing exposure time

- Data reasonably well described by:

Aluminum Release =a-12

• Differentiating wrt time gives instantaneous Al release rate

- decreased by a factor of 20 over 30 days

• Using a single value for Al release rate obtained from short duration testing therefore excessively conservative when used to calculate aluminum release over long periods 34 A4 L

page 17

(

Inhibitory Effects

" Weight changes of Al coupons in ICET Tests 1 and 4 were significantly different

- both tests used NaOH to adjust the pH to the same target value (pH = 10); average pH in ICET Test 1 was only 0.4 pH units lower than that in ICET Test 4

" High [Al], which increased with experimental time, was measured in solution in Test 1

- only trace concentrations of Al present in ICET Test 4 solutions

• Attributed to the presence of silicate species from cal-sil

• Additional experiments by the PWROG confirmed the inhibitory effect of silicates and phosphates on aluminum corrosion 35 Al Release Calculations - Assumptions 1.

2.

3.

4.

6.

7.

Maximum temperatures of sump and spray water used Maximum pH values used during corrosion calculations No credit taken for inhibitory effects (silicate, phosphate, etc) on Al corrosion No credit for presence of pre-existing oxide films on Al surfaces All Al released by corrosion enters the solution No credit taken for effect of oxygen in the sump water.

No credit taken for decrease in corrosion rate as a factor of exposure time resulting from development of passive film Assumptions build a significant conservatism into Al release calculations Believed that Al release into sump water significantly overestimated (as much as 2-3 orders of magnitude) 36 All. MCL WL page 18

Precipitation 37 Solubility of Aluminium Hydroxide

• "Solubility" refers to the. mass of a substance dissolved in a given amount of solvent to form a saturated solution at a given temperature and pressure

• Major factors affect solubility of a substance

- Temperature - generally, as T increases, solubility increases

- Solution pH can increase or decrease the solubility

- Ionic strength increases the solubility

- Common ions decrease the solubility om An equilibrium'property - the kinetics of formation can be very important in determining the phases that form and the time required for their formation 38 A.

rlAlda FACL page 19

Solubility of Aluminum Hydroxides

" Aluminum can form various hydroxides in weak acidic and alkaline aqueous solutions: gibbsite, boehmite, bayerite and diaspore

- Solubilities of all of these species are similar

" Gibbsite is the thermodynamically most stable phase and the less stable phases can be transformed to gibbsite

" Amorphous aluminum hydroxide has much high solubility than crystalline aluminum hydroxides and can form in highly supersaturated aluminum solutions 39 Solubility of Gibbsite 1-E0.

U) 0)

o

-2

-3

-3.

Gibbsite

• PH 6.5 A, pH 7.5

• pH 8.5

• , pH 9.5

• pH 10.5 opH 11.0 000O

0.

0O000

• 0*

0 0 "0

oO o

O

• o.*

• o°.

. I 0 Cl1 D

AAA A

I A

A A

A4 000~~

ALA, A

• 2~

4 0

20 40 60 80 100 120 Temperature (o0)

• Solubility increases with increasing pH and T o Strong function of pH and weak function of T 40 A 'MCL FAACL page 20

Solubility of Calcium Phosphates

" Calcium and phosphate forms various low solubility salts:

- Hydroxyapatite,Ca,(PO) 30H

- Whitelockite, 13-Ca 3(PO.) 2

- Octacalcium phosphate, Ca8H2(PO4)6.5H20

- Monetite, CaHPO., and

- Brushite, CaHPO,.2H20

- Amorphous calcium phosphate

" Dissolution generally is not congruent (I.e., composition of the solution and solid are different) and accurate solubility data are hard to obtain

" Hydroxyapatite is the least soluble in water above pH 4 and thermodynamically most stable 41 Solubility of Hydroxyapatite

~,-1.5*

E -2.0 M -2.5 0

r -3.0

-3.5.

E a -4.0*

-o-4.5

• 0

.50~

0 50 100 150 200 250 temperature CC) 300 350 400 42 page 21 Al.A'CLFA

Detailed Calculations 43 Calculations Information provided by Dominion:

- Estimated Al surface areas

- Materials present in containment

- Post-LOCA temperature evolution of the sump water and containment air or spray water

- Minimum and maximum pH values for the sump water and spray water at each time interval.

• To be conservative, the maximum pH value was chosen for the calculation of the aluminum corrosion rates, since the aluminum corrosion rates are higher at higher pH values.

- Sump water volume 44 MC~L page 22

Calculations

• Al corrosion rate at each temperature and pH evaluated using the corrosion data presented earlier:

CORROSION RATE (T) = 9.2 x 2 x 107 x EXP(-6301.1 x (1'TEMPERATURE (K)))

and CORROSION RATE (pH) = 3 x 10"7x exp(1.3947 x pH)

Overall corrosion rate at given temperature and pH calculated using:

CORROSION RATE = CORROSION RATE (T) x CORROSION RATE (pH) 45 Calculations

" Average corrosion rate over time interval calculated assuming a'linear change in corrosion rate over interval

" Al release calculated using:

Al RELEASE OVER INTERVAL = CORROSION RATExlNTERVAL LENGTHxAI SURFACE AREA

" Total Al release over time segment and cumulative Al release since time = 0 also calculated

• 1 46 All MCL FC page 23

Bench Top Testing Purpose is to confirm the conservative nature of assumptions on precipitate formation Tests will be carried out to determine the concentrations at which aluminum hydroxide will precipitate under representative conditions (pH, temperature, ionic strength)

If required, bench top testing will also be' performed to determine the optimum method for preparation of precipitates for reduced scale testing

- Assess issUes related to storage, aging 47 Summary

" AECL's chemical effects calculations are based on a thorough assessment of the existing literature

" Based on this assessment, aluminum hydroxide and calcium ph hsphate were the only po-ssilrle precipitates considered in tis aa ysis

• While the spreadsheet from WCAP-1 6530 was not used for the calculations of the concentrations of species released into solution, the methodology developed by AECL is believed to be very conservative

- A number of conservative assumptions were made during the release calculations 48 A'

MCL W page 24

Conclusions AECL chemical effects assessment is consistent with WCAP-16530 but uses more realistic plant operating conditions Significant conservatism is maintained in the determination of chemical precipitants as well as in the overall chemical assessment approach Confirmatory bench-top testing will be performed to validate assessment results Based on bench-top testing results, head-loss testing will be performed if deemed necessary Should head-loss testing be required, it would not be completed prior to the 1st quarter of 2008 i~fr~

-t

,D4Nminion 02003 Dorinion Chemical Effects Assessment Questions?

-- -/

02003 Dwmirion il)Doiminion

.3

0 Dodron 2003 94