ML072980699

Response to Request for Additional Information - Technical Specification Change Request No. 337: Reactor Building Emergency Sump Ph Control System Buffer Change

ML072980699
Person / Time
Site:	Crane
Issue date:	10/19/2007
From:	Cowan P AmerGen Energy Co
To:	Document Control Desk, Office of Nuclear Reactor Regulation
References
5928-07-20202, TAC MD5963
Download: ML072980699 (21)

Text

AmerGeIISm AmerGen Energy Company, LLC www.exeloncorp.com An Exelon Company 200 Exelon Way Kennett Square, PA 19348 10 CFR 50.90 October 19, 2007 5928-07-20202 U.S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, DC 20555-0001 Three Mile Island, Unit 1 (TMI Unit 1)

Facility Operating License No. DPR-50 NRC Docket No. 50-289

Subject:

Response To Request For Additional Information - Technical Specification Change Request No. 337: Reactor Building Emergency Sump pH Control System Buffer Change (TAC No. MD5963)

References:

(1) USNRC letter to AmerGen Energy Company, LLC dated October 4, 2007, "Request for Additional Information Regarding Proposed Reactor Building Emergency Sump System Buffer Change (TAC No. MD5963)"

(2) AmerGen Energy Company, LLC letter to NRC dated June 29, 2007 (5928-07-20097), 'Technical Specification Change Request No. 337 -

Reactor Building Emergency Sump pH Control System Buffer Change" This letter provides additional information in response to the NRC request for additional information (RAI), issued October 4, 2007 (Reference 1), regarding TMI Unit 1 Technical Specification Change Request No. 337, submitted to NRC for review on June 29, 2007 (Reference 2). The additional information is provided in Enclosure 1.

Regulatory commitments established by this submittal are identified in Enclosure 3. If any additional information is needed, please contact David J. Distel at (610) 765-5517.

I declare under penalty of perjury that the foregoing is true and correct. Executed on the 19 'h day of October, 2007.

Respectfully, Pamela B. Cowan Director - Licensing & Regulatory Affairs AmerGen Energy Company, LLC

U.S. Nuclear Regulatory Commission October 19, 2007 Page 2

Enclosures:

1) 2)

3)

Response to Request for Additional Information Applicable Pages from C-1101-153-E410-040, Rev. 0, References 4.2 & 4.5 List of Commitments cc:

S. J. Collins, USNRC Administrator, Region I P. J. Bamford, USNRC Project Manager, TMI Unit 1 D. M. Kern, USNRC Senior Resident Inspector, TMI Unit 1 File No. 07029

ENCLOSURE 1 TMI UNIT 1 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION TECHNICAL SPECIFICATION CHANGE REQUEST No. 337 Reactor Building Emergency Sump pH Control System Buffer Change

5928-07-20202 Page 1 of 7 RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION (RAI)

TMI UNIT 1 TECHNICAL SPECIFICATION CHANGE REQUEST No. 337 Reactor Building Emergency Sump pH Control System Buffer Change

1. NRC Question In section 7.1, page 15 of Attachment 2 the licensee listed eight chemicals which will exist in the sump water after a loss of coolant accident (LOCA). In its analysis the licensee assumed that each of them will dissociate or speciate into positive or negative ionic species. By balancing the resulting positive and negative charges, to obtain neutrality of the solution, the licensee could determine the amount of trisodium phosphate (TSP) required for obtaining a given value of pH. Provide a list of the ionic species.

Response

The description of the methodology stated above is correct as utilized in the referenced calculation. The Boron Speciation can be found in section 7.7 of the previously submitted calculation (C-1101-1 53-E410-040, Rev. 0) (pages 22 to 24). The TSP Phosphate Speciation can be found in section 7.8 of the previously submitted Attachment 2 calculation (C-1 101-153-E410-040, Rev. 0) (pages 24 to 26).

2.

NRC Question Describe the procedure for determining speciation of boron and TSP.

Response

The equilibrium concentrations of the boron species are determined by using the method described in reference 4.5 of the previously submitted Attachment 2 calculation (C-1101-153-E410-040, Rev. 0). The applicable pages of this reference are provided in.

The equilibrium concentrations of the TSP species are determined by using the method described in reference 4.2 of the previously submitted Attachment 2 calculation (C-1101-153-E410-040, Rev. 0). The applicable page of this reference is provided in. This paper is written in Russian, with an English abstract on the last page.

The information in the abstract confirms the ionization constants as used in the above referenced pH calculation.

3.

NRC Question Explain the statement in section 2.6 on page 5 of Attachment 2 that "boron content will not be exceeded provided the power rate does not increase (2568 MWt) and the cycle duration (2 years) does not increase."

5928-07-20202 Page 2 of 7

Response

TMI Unit 1 is currently licensed to operate at 2568 MWt with a 24-month cycle and these current licensing basis parameters have been utilized in the analysis referenced above.

Any potential change to the TMI Unit 1 licensed power level or cycle duration would require additional evaluation of the impact on boron content. Potential changes to the licensed power level or operating cycle duration are not directly applicable to this license amendment application.

4.

NRC Question In section 6, on page 13 of Attachment 2, it is stated that water and boric acid in the sump is determined from a mass balance with two bounding conditions for water arid boric acid.

What are they and how were they determined?

Response

The two bounding conditions are Beginning of Cycle and End of Cycle chemistry configurations. Beginning of Cycle was chosen to maximize the boric acid concentration and mass. End of Cycle was chosen to minimize the boric acid concentration and mass.

The values such as Borated Water Storage Tank (BWST) concentration are based upon Technical Specification limits.

5.

NRC Question According to Regulatory Guide 1.183, if the sump pH is controlled at the minimum values of 7, the molecular iodine is prevented from formation and release to the containment.

However, during the engineered safety feature (ESF) recirculation phase, a higher pH is assumed to support iodine flashing fractions at 5 percent for the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and 2 percent for the following 29 days. In the submittal the licensee specified the required pH

>7.3. Discuss: (1) the methodology for determining this pH value and (2) was the sump water pH buffered at this pH?

Response

The methodology for determining a minimum pH requirement of 7.3 utilized an overall approach of applying the Standard Review Plan (SRP) guidance with differences only in the use of pH values and application of a model for the mass transfer of iodine across the boundary layer between the liquid surface and the bulk gas space in the auxiliary building, as described in Calculation No. C-1l101 -900-EOOO-087, Rev. 2, previously submitted in Reference 2. This approach identified at pH values below 7.3 the TMI Unit 1 UFSAR flashing fraction requirements of 5% for the first 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> and 2% for the following 30 days, could not be met. Therefore, a limiting, minimum pH value of 7.3 was established.

At a pH of 7.3 the sump water is buffered utilizing TSP. This pH value is indicative of the maximum allowable amount of boron and the minimum amount of TSP in the sump water, resulting in the lowest achievable pH value after buffering.

5928-07-20202 Page 3 of 7

6.

NRC Question Proposed insert C to technical specification (TS) page 4-2b describes the need to perform periodic solubility tests on the TSP to ensure adequate dissolution time. However, Insert D to the proposed TS surveillance requirements, TS page 4-10c, does not include solubility as one of the verifications performed on the TSP. Describe this apparent discrepancy between the TS bases and the TS surveillance requirement.

Response

Proposed insert D to the Technical Specification (TS) page 4-1Oc, describes the system surveillance requirements for the Reactor Building Emergency Sump pH Control System.

Surveillance Test "b" of proposed insert D states, "Verify that a sample from the TSP baskets provides adequate pH adjustment of borated water." This surveillance requirement, specifically the statement "adequate pH adjustment," is determined by performing a two-part test, which includes sequential solubility and buffering verification, as described in the proposed TS Bases, insert C. Therefore, the scope of proposed insert D TS surveillance requirement includes solubility verification testing, in addition to buffering verification. The elements and methodology of the proposed TS surveillance testing will be clearly delineated in station surveillance procedures (Reference commitment in Enclosure 3).

7.

NRC Question When verifying the TSP buffering capability, solubility, and total mass, how is the sample obtained? How does the sampling account for the potential that the TSP on the interior of the basket may exhibit different properties (hydration, density, etc.) than samples taken from the basket periphery? Discuss the potential impact that variations in material properties within each basket may have on surveillance tests to determine buffer capability, solubility, and total mass of TSP available.

Response

TMI Unit 1 will quantify the total mass of TSP buffer present during each refueling outage as described in the Technical Specification Change Request No. 337, submitted June 29, 2007. The TSP basket sample will be obtained by taking a total of approximately 500 grams of material from the basket. Procedures will ensure that samples are taken within the basket at both interior and peripheral locations to obtain a representative sample (Reference commitment in Enclosure 3). The mass of TSP in each basket will be determined each refueling outage to ensure the TSP mass is as required. Weighing each basket eliminates potential variations due to settling and changes in density that may occur. The TSP form to be used is TSP dodecahydrate (Na3PO4.12H20). The dodecahydrate form has the maximum water of hydration and will not absorb additional water and the water of hydration will remain constant during subsequent operating cycles.

Use of the maximum hydrated form will eliminate variability in hydration and density as it is fully hydrated. Determining the total mass and use of the maximum hydrated form will eliminate sample variability that may exist in different areas of the baskets. Measuring the total mass present eliminates the requirement to determine density of the material in the baskets and potential differences of material density in different sections of the baskets.

5928-07-20202 Page 4 of 7

8.

NRC Question Proposed insert C to TS page 4-2b describes the need to periodically determine the mass of TSP because leaking valves and components may dissolve some of the TSP from the baskets. If a leaking valve or component were to result in a portion of the TSP being dissolved, how would the dissolved TSP be accounted for when adding additional TSP to the basket. If there is no mechanism for removing the dissolved TSP then addition of more TSP may result in a total mass that exceeds the TS limits. Discuss how the loss of TSP through operational dissolution is accounted for when determining the total mass of TSP.

Response

If a leaking valve or component were to result in a portion of the TSP being dissolved, the solution would drain to the Reactor Building Sump by running across the floor or through floor drains. The sump water is periodically drained (i.e., lx per month and will be more frequent with the new sump design) to the Auxiliary Building sump for processing, thus removing the dissolved TSP from the Reactor Building. Any TSP residue remaining on the floor or sump would be insignificant. Initial TSP loading is approximately 27,000 Ibs, which provides significant margin to the maximum allowable TSP quantity of 28,840 lbs to accommodate potential TSP residue remaining on the Reactor Building floor. Any TSP that has dissolved and is lost to the floor or sump is accounted for by determining the total mass of TSP in each basket. If there is no loss by dissolution the mass will remain constant.

There is no mass lost from settling, only a volume change, which does not impact mass.

9.

NRC Question TMI-1 is proposing to switch to TSP as a buffering chemical. Testing indicates that TSP in the presence of dissolved calcium can result in rapid precipitation of calcium phosphate, which can create significant head loss across a sump strainer covered with a debris bed.

Provide a list of all potential sources and amounts of calcium within the TMI-1 containment and provide the calculated dissolved calcium concentration in a post-LOCA pool. Provide the relative chemical precipitate loading predicted by the WCAP 16530-NP, ["Evaluation of Post-Accident Chemical Effects in Containment Sump Fluids to Support GSI-191"], model for TMI-1 with TSP and at the maximum projected pH value (8) for the pool.

Response

TMI Unit 1 is evaluating chemical effects through alternate testing. However, use of the WCAP 16530-NP model is being considered, and if utilized, this issue will be addressed in the TMI Unit 1 Generic Letter 2004-02 response. The alternate testing program will estimate the contribution of chemical effects on strainer head loss performance. The key element of the program is small scale testing in a simulated post-LOCA containment environment in which the increase in head loss from chemical effects of materials (dissolution, leaching chemical compounds, and precipitate formation) is compared to head loss due to debris without chemical effects. The chemical effects head loss testing is intended to reproduce as close as possible the chemical interactions of materials (dissolution, leaching chemical compounds, precipitate and debris physical characteristics) in a post LOCA containment environment. This testing includes the effects of fibrous insulation installed at TMI Unit 1 and concrete.

5928-07-20202 Page 5 of 7 TMI Unit 1 does not utilize calcium silicate insulation, which is the significant contributor of calcium in the industry. The water sources that will be present in the Reactor Building sump during building spray actuation (BWST, RCS) are less than 1 PPB calcium. Other potential calcium sources at TMI Unit 1 are insignificant and would not impact the overall calcium loading.

10. NRC Question In Section 4.5 of Enclosure 1 to the proposed license amendment request, it is stated that the evaluation was based on utilizing and analyzing available industry and technical research data. The evaluation concluded that all environmentally qualified equipment located inside the Reactor Building (RB) are qualified for the revised RB spray chemical conditions resulting from the proposed change. Please provide details of the methodology, assumptions, regulatory provisions applied and analysis to support the above conclusion.

Response

Environmentally Qualified (EQ) equipment is currently qualified for Reactor Building (RB) spray with a pH range of 8.0 - 11.0. The proposed design alters the pH range to 4.0 - 10.0 over the initial 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, with a long-term pH of 7.3 -8.0 (essentially a neutral solution). To assess the EQ impact of the revised RB spray, EQ equipment located inside the RB was evaluated for the effects of an acidic RB spray (low pH extreme). The current EQ evaluation bounds the high extreme of the revised RB spray. This evaluation is described below:

Regulatory Provisions The EQ Program for TMI Unit 1 meets the requirements of 10 CFR 50.49. EQ equipment has been evaluated for compliance with either the Division of Operating Reactors (DOR)

Guidelines or 10 CFR 50.49, with guidance from Regulatory Guide 1.89.

TMI Unit 1 equipment qualified to the DOR Guidelines relies on partial test and analysis.

With regard to containment sprays, the DOR Guidelines, Section 5.3 states:

...[AJn item of Class 1E equipment may be shown to be qualified for a complete spectrum of service conditions even though it was only type tested for high temperature, pressure and steam. The qualification for service conditions such as radiation and chemical sprays may be demonstrated by analysis (evaluation).... Components enclosed entirely in corrosion resistant cases (e.g., stainless steel) may be shown to be qualified for a chemical environment by analysis of the effects of the particular chemicals on the particular enclosure materials. The effects of chemical sprays on the pressure integrity of any gaskets or seals present should be considered in the analysis.

TMI Unit 1 equipment qualified to 10 CFR 50.49 relies on sequential testing of a specimen that is identical or similar to installed equipment, with supplemental analysis. With regard to containment sprays, 10 CFR 50.49(e)(3) states:

5928-07-20202 Page 6 of 7 The composition of chemical used [for qualification] must be at least as severe as that resulting from the most limiting mode of plant operation (e.g., containment spray, emergency core cooling, or recirculation from containment sump). If the composition of the chemical spray can be affected by equipment malfunctions, the most severe chemical spray environment that results from a single failure in the spray system must be assumed.

With regard to qualification methods, 10 CFR 50.49(f)(4) identifies the following as an acceptable method:

Analysis in combination with partial type test data that supports the analytical assumptions and conclusions.

Qualification per 10 CFR 50.49 is based on type testing that includes chemical spray, though for different pH ranges.

As shown by these excerpts, supplemental analysis of the effects of the revised RB spray is permitted for equipment qualified to either the DOR Guidelines or 10 CFR 50.49.

Methodology The methodology for EQ evaluation of revised RB spray is to identify the age-sensitive and metallic materials wetted by the spray, and assess their resistance to chemical attack. The evaluation relied upon the original EQ test if it included a chemical spray that bounded the pH range of the revised RB spray. For cases not bounded by the EQ test, or in the absence of testing under spray conditions, evaluation relies upon available industry and technical research data, such as that published in Perry's Chemical Engineers' Handbook.

This handbook rates the resistance of materials to various chemical agents using general descriptors such as excellent, very good, fair, poor, etc.

Assumptions The only assumption in the EQ evaluation of revised RB spray is a conservative determination of TMI Unit 1 RB spray duration (52 hours6.018519e-4 days <br />0.0144 hours <br />8.597884e-5 weeks <br />1.9786e-5 months <br /> post-DBE). There are no assumptions regarding material identification since EQ Binders identify wetted parts and materials. Similarly, there is no assumption regarding chemical effects since EQ test data or published literature is available in all cases.

Evaluation The evaluation extends the current EQ basis with supplemental analysis of the resistance of materials to the effects of the revised RB spray. The low pH extreme (4.0) is addressed primarily by analysis. The high pH extreme (10.0) is bounded by the current qualification assessment and is justified by either analysis or EQ test data. The acceptability of the revised containment chemical spray pH range and duration, with respect to age sensitive materials and metal housings, is performed in accordance with the requirements of 10 CFR 50.49, "Environmental Qualification of Electric Equipment Important to Safety for Nuclear Power Plants." Age sensitive materials and metallic enclosures of the EQ equipment that will be exposed to chemical spray are identified. These materials are evaluated using available industry and technical/research data including chemical

5928-07-20202 Page 7 of 7 resistance of materials, effect of aging on materials exposed to spray, and the overall spray duration. The evaluation acceptance criteria are based on determination of age sensitive materials and metallic enclosures exposed to the chemical spray. The chemical resistance of the exposed age sensitive materials and metallic enclosures is determined utilizing Perry's Chemical Engineer's Handbook, Sixth Edition, and Harper's Handbook of Physics, Elastomers and Composites, Second Edition. The applicability of acceptance criteria to aged components and at elevated temperatures was evaluated and determined that the revised containment chemical spray at TMI Unit 1 does not exacerbate any significant adverse effect on aged components the acceptance criteria remains applicable to the EQ components located in the TMI Unit 1 containment.

Surface conductivity of sprays was evaluated for potential effects on conducting material/terminal blocks. At TMI Unit 1 this is not a concern since terminations are either sealed or not directly exposed to the chemical spray. Deposition of boron/foreign material, if any, will be washed away by the longer duration spray solution.

Cables installed in conduits, jacketed cables, splices, and terminations were evaluated.

The spray evaluations for these materials determined that there is no effect due to the revised spray condition.

Conclusion This evaluation demonstrated that the EQ equipment located inside containment at TMI Unit 1 remains qualified by the existing EQ files, as supplemented by the revised containment chemical conditions. No specific actions are required to physically protect equipment from spray contact.

11.

NRC Question Please confirm that the temperature profile inside the RB remains unchanged due to the proposed design change.

Response

The Reactor Building temperature profile remains unchanged due to the proposed buffer design change.

ENCLOSURE 2 Applicable Pages from C-1101-153-E410-040, Rev. 0, References 4.2 & 4.5

OCT 09 2007 15:04 FR CISTI ICIST 613 952 9303 TO 13122695932 P.08

ý3

,IIHTEPATYPA

1. N. B l err urm, A. Un m a c k. Kgl. Danske Videnskab. Seiskab., Mat.-fys. Medd.,

9, M& 1 (1929), 126, 132, 141 (LtXH. no :[25]).

2. L. F. Nims. J. Amer. Chem. Soc., 56,

? 65, 1110 (1934).

3. R. Q. Bates. J. of Research of the Nat. Bur. of Standards, 47, 3. 3, 127 (1951).

4. R. E. Mesiner, C. F. Bayes. J. of Solution Chemistry, 3, M 4, 307 (1974).

5. F. H. Sweton, R. E. Mesmer, C. F. Bayes Jr. J. of Solution Chemistry, 3, 191-214 (1974).

6. J. W. Wright, W. T. L i n d s a y, T. R. D r u g a. Westinghouse Atomic Power DI*

i1 "

vision Report. WAPD-TM-204. June 1961 (UIHT. no [25]).

7. A. A. Noyes, Y. Kato, R. R. Sosman. J. Amer. Chem. Sot., 32 (1910) 159 (IIT-. no [25]).

)H

8. H. H. MaKCHM0Ba, B. (D. IOmxeBnir..9neKTpoxtinms, 2, 577 (1962).

To

9. L. F. Nims. J. Amer. Chem. Soc., 55, X2 6, 1946 (1933).

rH

10. A. K. Grzybowski. J. Phys. Chem., 62, 555 (1958).

4e

11. R. G..Bates, S. F. Acree. J. of Research of the National Bureau of Standards, 30, N 2, 129-55 (1943).

12. F. Ender, W. Teltshik, K. SchAIer. Z. Elektrochem., 61, 775 (1957).

1

13. F. C uta, B. Po lef. Chem. listy, 49, Ne 4, 473 (1955).

0-

14. JI. A. IlaaBnm x, B. C. CMoa9ioB, n1.A. Kp0OKOa. H3B. CH6. OTA. AH CCCP, ja cep. xa m. Hayzc, N eo 7, Bbin. 3, 13 (1969).

H:

15. J1. A. IIaBjioK, B. C. CmOjiRKOB, HI.

A. KpboKos. H3B. CH6. oTA. AH CCCP, cep. xHM. HayK, X-. 7, Burr. 3, 3 (1972).

16..1. A. HIa BaXoK, I. A. Kp0oKoa.

Hs3.

CH6. oTA.

Al CCCP, cep. xIm.

nay*.,

HN 14, Bun. 6, 25 (1976).

• -17.

JI. A. Hn BA.i, B. C. C M o.jjn K o B. 113B. CT6. oTA. AH CCCP, cep.

Hm..nayr,.

10 Neo 12, Buwn. 5, 16 (1974)...-

18. 3I. A. 'laBloKc, B. C. C oq s*xos.

H1B. CuS. OTA. AH CCCP, cep. xHm. HayM,

)N 12, Bmiu. 5, 22 (1974).

19. M. Irving, J.J. Cox. Analyst., 83, 526 (1958).

20. B.0. r.

F

xe6panj, r.9. A e 1eabAen, r.A. BpafAT, H.1. ro ()am H. fpaK.

THqecKoe pyKOBOACTBO no HeopraHWIeCKOMy ana.rni3y.

M.,

rOCXH.VH3*aT,

1960,

c. 724, 744.

1*1

21. AnaJIma M*Hepanblioro cbipbs. rloA pea. 10. H. KHHnOBIq, 10. B. Mopatiencxoro.,

-FoCXMH3*aT, 1959, C. 986.

[11

22. B. C. C U 0 A q K 0 B. BHHHTH, N. 276-68.,Aen.

23. K. S. Pitzer. J. Amer. Chem. Soc., 59, 2365 (1937).

H-

24. R.W. G u r n e y. J. Chem. Phys.; 6, 499 (1938).

25. Stability Constants of metal-ion complexes. Special publication, N 17. London. The

.Chemical Society. Burlington House (1964).

2-L.A.Paviyuk, P. A. Kryukov SPECTROPHOTOMETRIC DETERMINATION OF THE FIRST AND OF THE SECOND IONIZATION CONSTANTS OF PHOSPHORIC ACID FROM 25TO. 175 0C 1-The first and the second ionization constants, K.,.and Kd2, of phosphoric acid have been determined at various temperatures between 25 and 175°C from spectrophotometric data with 2, 6-dinitrophenol and p-nitrophenol as indicators.

The results expressed as a function of absolute temperature are given by the i 2 equations:

pKia 583,01/T-2,715-0,.009801.

T and pK.2ý 12721*7A-1,154+0,01368

• T.

93 TOTAL PAGE. 08 **

Bomic AcID EQUILIBRIA TABLE I1 EXTENT OF REACTION:

QUENCHING V$. SPECTROPHOTOMATRVY

% complete-O'[1Sn('1)l],

10s[HtOt]e, 101[Cu(II)]j, Spectro-At if M

A uenching photometry 2.43 1.22 0.00 88 :1: 3 38 8- 0.5 1.17 2.34 0.00 71 :L 3 6.=- 1 1.17 1.17 0.00 57 :k 3 46 : 1 1.17 1.17 1.40 46 3

W50.L1 Conditions:

IHCIOl -

1.00 M, p -

2.00 (LiCIO,), 25',

255 nm.

Cahill and Taube' have demonstrated that isotope fractionation studies of hydrogen peroxide reactions can be used to gain information about the mechanism.

An initial l-equiv step should yield a value of 1.00 for the fractionation factor, as defined by Cahill and Taube, and a value of 0.940 for an initial 2-equiv step. The value observed by Cahill and Taube for the Sn(I1)-

H2O2 reaction in hydrochloric acid was 0.943 indicative of a 2-equiv step.

The results of some fractionation experiments are shown in Table III. The values of the fractionation TABLE III RzSULTS OF 1SOTOPIC FRACTIONATION EXPERIMENTS 10'[So(oI)].. 10' 1sO,],, 10'lCu(ul),. [Cucod.

[KCII, Fractionatlon M

Jf if Af At factor' 2.1 3.0 0.0 1.24 0.0 0.972 1.9 a.0 0.0 1.29 0.0 0.963 1.:

3.0 0.0 0.00 0.96 0.957 a As defined in ref S.

factors lie between those expected for a 1-equiv or a 2-equiv mechanism and suggest that both types of processes are operative.

Our value obtained from HCi solutions does not agree with that reported by Cahill Inorganic Chemistry, Vol. 11, No. 3, 1972 537 and Taube.

A brief series of kinetic experiments in-dicates that marked deviation from second-order be-havior occurs in HCI solutions. Adherence to a second-order rate law should have been observed if the mechanism was a single, 2-equiv step.

Higginson has used the induced reduction of KaCo-(C20 4)s as a test for the presence of Sn(III) as an inter-mediate in the oxidation of Sn(II).

In experiments where the initial concentration of the cobalt complex was about equal to the Sn(II) and HI:O concentrations, no induced reduction of K4Co(C:O4s could be detected.

Conclusions and Summary In perchlorate solutions the reaction between Sn(II) and H02 s is complicated.

Significant discrepancies in the apparent extent of reaction as determined by spectrophotometry compared to result% obtained via quenching and titration techniques suggest the existence of a non-steady-state intermediate which absorbs in the wavlength region used to follow the disappearance of tin(II). The addition of Cu(II) ion appears to elim-inate the reaction pathway giving rise to the inter-mediate, which probably is the result of a I-equiv reduction of H20 2.

However, even at high concen-trations of Cu(II) ion the reaction still takes place and then obeys a second-order rate law which is consistent with a 2-equiv mechanism.

It seems likely that the reaction takes place by parallel pathways involving 1-and 2-equivalent mechanisms. Isotope fractionation factors intermediate between those expected for the two mechanisms lend support to the dual nature of the mechanism.

Acknowledgment.--This work was supported by National Science Foundation Grant GP-13900.

CONTRIBIu*ON PROM OAK Ri=c1 NATIONAL LABo0ATOitV, OAX RIDGE, Tassasi 37830 Acidity Measurements at Elevated Temperatures.

VI.

Boric Acid Equilibria1 By R. E. MESMER,* C, F. BAES, JR., AND F. H. SWE'TON Received May 17, 1971 Borie acid-borate equilibria have been studied by means of a hydrogen electrode concentration cell from 50 to 290'.

The equilibrium in dilute boric acid solution, B(Ol)I

+ OH- - B(OH),-, was studied as a function of CCI concentration from 0.13 to 1.0m.

A small Ionic strength (1) dependence of the equilibrium quotient (Q,.,) was observed which decreases slightly with temperature.

The pressure dependence of the equilibrium quotient is very small up to 2000 psi.

Values of Q.l are given to better than 0.01 Ig unit by the expression log Q1.1 - 1

+ 28.6059 + 0.012078T -

13.2258 log 2T + (0.3250 -

0.000337')r -

o.0W121'/,

for values of! rfrom 0 to I m and T from 273 to 573"K. Results for more concentrated boric acid solutions (up to 0.6 m) in 1.0 m KCI have Indicated the formation of the polynuclear species %s(OH1 - and Bj(OH)1,- and minor sanounts of either B,(OH).t4-or B,(OlH),'-. The trimeric species is needed to fit the data at all temperatures studied (and previous data at 25'). The dimeric species, which has not been reported previously, is indicated quite strongly by the present results at 200' while the need for a fourth (tetrameric or pentameric) species is indicated by the results at W0°. The two alternative schemes--B(OH).,- Ba(OH),-, BP(OH)1G-, and B,(OH)14'- or Bs(OI)i----are both consistent with the data of Ingri at 25' and give as good a fit as the four-species scheme which he chose.

Thermodynamic parameters have been derived for all the species in I m KCC and for B(OH),-at Infinite dilution.

Introduction Boric acid-borate buffer mixtures serve as pH stan.

dards, occur in natural aqueous systems and in de-(1) Research oponsored by the U. S. Atomic Energy Cemmlsuion Under coutrset with Union Carbide Corp.

tergent solutions, and are used as burnable nuclear poisons in the coolants of pressurized water nuclear reactors. It is not surprising, therefore, that the equi-libria which occur in various aqueous solutions of boric acid and borate have been extensively studied,

$38 Inorganic Chemistry, Vol. 11, No. 3, 1972 although only at temperatures near room temperature.

From the results it is clear that, in addition to the mononuclear boric acid and orthoborate ion, which have been well established by infrared and Raman spectroscopy'-' to be trigonal (B(OH)3 ) and tetrahe-dral (B(OH)4-) species, a number of polyborate ions also are formed.

Manov, et at.,? and Owen5 have carried out careful measurements of the equilibrium B(OH)3 + OHR- *B(OH),

(1) at boron concentrations as low as 0.01 m where inter-ference from polyborates is insignificant. The two sets of measurements, both made over the tempera-ture range 0-60M in cells without liquid junction and containing hydrogen and Ag-AgCl electrodes, are in excellent agreement. They also show that borax (Na2-B401. 10H20) reacts quickly in dilute solution to pro-duce the mononuclear species in the equilibrium amounts.

The equilibria occurring in more concentrated boric acid-borate solutions have been carefully studied in several media by Ingri.1-'1 The species formed ap-pear to be as well established as can be expected con-sidering the number which evidently are formed and the relatively large medium change which must be introduced in order to study them.

Ingri found his extensive data to be most consistent with formation of the polymeric species Bs(OH)Io-, Bs(OH)u2-, and B4(OH)li,-. Ingri also has shown by means of a "self-medium" study12 that no species are formed with a hydroxide-to-boron ratio greater than 4 in the pres-ence of as much as 0.1 m hydroxide. This demon-strates convincingly that the equilibrium constants given by Konopik and Lebcrl" for dissociation of the second and third protons from boric acid (equivalent to the formation of B(OH).s-and B(OH)sG-) are much too high.

The recent work of SpessardU on the derivation of equilibrium quotients for several assumed species from data up to 900 unfortunately does not contribute to the identification of the actual species present since the total boron concentration was not varied and his technique provided data of very limited accuracy.

This latter point is illustrated by the poor agreement of his data with the more precise data of Owen and Kinge in 3 M NaCl and the irregular variation of log Q with temperature for several of his assumed species in several of the media studied.

Information on the rates of formation of some of the (2) (a) R. R. Serveos and H. M. Clark, J. Chem. Phyt., 26,.117TS (1937);

(b). 1. S. Goulden, S*#e ctothim. Agif,9, 637 (1959).

(3)

). E. Rethell and N. Sh1eppard, Trawu. Faraday Soa., I, 9 (1935).

(4) J. 0. Edwards, 0. C. Maorrison, V. F. Ross, and J. W, Schultz, J.

Amer. Chem. Soc., 77, 266 (1958).

(5) X. I¢rishnan. Pros.ladion Acad. S*i., Sect. A, S1, 103 (1963).

(6) 3. Ooubs*u and D. Hummel, Z. Pys. Chem., go, 13 (1959).

(7) G. 0. Manov. N. 3. DeLollis, and S. 7. Acres,.f. Rt,. Not. Bur. Stand..

SS, 28? (1944).

(8) B. D. Owen. A. Amer. Chem. Sac., 54, 1605 (1934); B. B. Owen and E

. King, ibid., TI, 1612 (1943).

(9) N. Ingrl, 0. Lagerstrom. M. Frydman, and L. 0. Si1len. Act* Chem.

Scand., 11, 1034 (1957).

(20) N. Uatl, ibid., 16, 439 (19*2).

(11) N. Ustr.ilid., IT, 881 (1963).

(12) N.

angf, ibid., 17, 573 (1083).

(13) X. Ingl. So. Kom. Mike., 78, 199 (19e3).

(14) N. Xonopik ard O. Leberl, Mon4~s*i..

0, 655 (1040).

(18) J. R. Speasard. J.I1nor. Aucr. C*em.. 32. 2507 (1970).

MESMER, BARS, AND SwEEToN polyborates has been obtained by Anderson, ef at.,'6 and 0sugi, el at.," using temperature-jump and pres-sure-jump techniques.

They are in agreement and assign a rate constant of about 3 X 10' M-1 sec-' for the reaction 2B(OH)i + B(OH).-

)BsOH) 1 9-(2)

Mfomii and Nachtrieblt found evidence for a slightly slower approach to equilibrium at concentrations of boron between 0.6 and 2.0 M at 250 in solutions made up from NaBbOs.

Two separate 111; resonance peaks with variable relative intensity were observed also by Onak, et a!.10 The former authors interpreted their data in terms of equilibria involving pentaborate ion BaO(OH)C-although no unique assignment could be made. Equilibria were attained in these studies dur-ing the time of mixing. All species in tetraborate so-lutions (M2BsO7) exchange rapidly giving a single nmr line. The surprising observation of apparent* slow changes in the infrared spectra of supersaturated boric acid-borate solutions reported by Valyashko and Vlasovau was not confirmed by Rannan spectra and pH measurements taken by us over a 10-week period at 25' for solutions containing 0.6-0.S m boron in boric acid-borate mixtures.

Because of the importance of borates in general and some of their uses which involve high temperatures, we have employed the high-temperature potentio-metric techniques developed recently in this labo-ratory"t to examine the hydrolysis o:: boric acid. The results of such a study in dilute borate solutions should be especially useful in extending the temperature range of pH buffer mixtures and in estimating the pH and other chemical properties of nuclear reactor coolants.

Results in more concentrated borate solutions at ele-vated temperatures should establish more firmly the identity of the polyborate species which are formed in aqueous solutions.

Experimental Section M-terial.-A stock solution of 3.3 m KCI prepared from J. T.

Baker analyzed reagent was purified by acidifying to pH 3.5 and purging with N, to remove COl. The fluoride content of the neutralized stock was 7 X 10- in as determined by the lanthanum fluoride electrode.

The protolytic impurities in a I m solution made from the stock were ca. 10-6 m as determined by titration between pH values of 5 and 0.

The KOH solutions were prepared from KOH pellets and a small stoichiometric amount of Ba(OH)2"811,O was added to pre-cipitate the C02-present. The base solutions were stored in paraffin-lined vessels under hydrogen.

The boric acid solutions were prepared from reagent boric acid which had been recrystallized from water.

The stock solution was standardized by titration in the presence of mannitol.

Ultrahigh-purity hydrogen (99.990%) from J. T. Baker was used throughout this study.

Potentlometric Apparatus and Techniqu s.-The pressure cell and electrcde assembly employed for this work were described in detail in a previous paper."1 At the beginning of each experi-ment the air was removed from the cell by successively pres-surizing with hydrogen to 500 psi and venting to the atmosphere.

This cycle was repeated three times before addition of hydrogen (15) J. L. Anderson, E. M. Hyrlng, and M. P. Whittaker, j. Phys. CAhem.,

$6, 1129 (1904).

(17).. Osuil. M. Sate, and T. Fujil,.Vipfos Kaahaa Zeishi. It, 582 (1968).

(181 R. 24aml and V,.schtrleb, Inarl. Chem.,., 1189 (1967).

(19) T. P. Onak, H. Landesnan. R. E. Williams, and 1. Shapiro, J. Phys.

Chim.. 6, 1653 (1939).

(20) M. G. Valyashko and E. V. VVtasov, Gtokhiniyo, T, 818 (1088).

(21) R. E. Mesmer, C. F. Bass. Yr., and F. H. Sweeton. J. PAy,. Chem.,

74, 1937 (1970).

Bo~ic ACID EQUILIBILIA for the equilibration. In the titration enperiments, the titrant was delivered from a Zircaloy vessel through platinum capillary tubing to the cell compartment by means of a pressure generator.

Corrections were applied to the data for the small amount of solvent in the gas phase at the higher temperatures.

The cor.

rection was limited to less than 1% at 2900 since the gas-to.

liquid volume ratio was small In both compartments.

From the data of Byrnus" on the distribution coefficient of boric acid between the aqueous and gas phases, the amount in the gas phasc under any of the conditions of the present measure.

ments is expected to be <0.1%.

This was verified by analysis of the gas phase over a 1 m boric acid solution in I m XCI at 248.

Potentiomsetric Measurements.-The cell representation for experiments conducted In dilute boric acid solutions (:5 0.02 m)

Is a m KCI a m XCI Pt, H -0.02a x 1B(OH),0 o

PtF, I (a)

-0.01a si KOH 1 "

The electrode compartment on the right contains the reference electrode.

In these experiments the concentration of KCI a was varied from 0.13 to I m, and the KOH concentration was gen-erally kept at about 1/,eth of the KCl concentration. The solution on the left contains boric acid and KOH In an approxi-mate ratio of 2:1.

The etm was observed as a function of tem-perature. The resulting values of the equilibrium quotient for reaction 1 (Q,.,) are summarized in Table I and Figure 1, wherein TABLE I DATA Pox Loc Qll Ask A FuNcTixo Ov MIMHSLPATURB AND IONIC SThRNOIII 3$

I XCI SOitflONSI 12-I I

0" q,.o

%1, sO.

I02 Medi00 0412) 3244

4. do 220 I*2 49a20 0.114 IU4 0 004 200.02 252 told" 0.430 16"?

000) 03,4 2A2 4*2,

9

".21 4.2111 0.00 30142 22 4*0"0 42 02

.0 1)013 111 2* 200) 111 3.201

.40 2 100 20 I001 U44 1.39 u0*6 49.24 051 I*I004 9.00 4 274 610" 29.60 si

.0.2 0*7 M 6.442 3202 Cold4 ISO. 3 i

6*

O 001 0.121 37?9 a

000.3312 M

• M0M4419 2200 I ke*

90.0~4 0.~:2 1M 4.94 434 000 99.60 at ON4 0.0*

4 3*21 8.04 2M1l 01 0l2064 8il2 1022 0.00 290.20 4.0t 402001 9.204 0.94 0001 994

=30 6 0002 02) 00 0

3)0.22 020 0,01 047O9lII 4174 0*00 344 1l 0*200 0043 1929 00st 264.0. 030 0*2299 0.27 I

3~

4001 21911 804 f

0i 007 salon 2 ai4 44

• 2.9 022 0*2203 4459 4324 0*06 O09.99 40 I002 f44 370 30 a4 5

300 0*200 Ol 01 4 Got 4£ 6A6.1 10.2 1

.2 a

ide 00 21 2t 4m

£30.02 8L26 0AIN02 @U3 1

0.$00 21)4 0271 03 l 009 4.224 2 4 0

.aM 3Z4OO 8220 0*20 U

0.n )

2.11 4 0A.m 09~39 8229 6.01044) 14)0l 830 464 SO.19 II3" 1.al" MeI AI#

_mI 99)0 8227 0*00I4.14 9225 20£ N00 2.44 80 2 9

)2$t 0.I43 27l0 a

.647

£931 4

8 l20 O064*91 00 0394 64*4

$11. 1820 0

142 32 1 0020 o Qo.i is the mo*ll eq*uilbrium quotient for reaction 1. 1I s the ionic strength, me is the boron molality, and U is the number of OH - ions bound per B(OH), group in solution.'

I is t.he ionic strength, ms i the total boron molslty, and ai is the ligand number, i.I.,

the number of a--

Ions bound per B(OH)a group in solution.

At higher concentrations of boron (0.02-0.8 in), where the ob1ective was to determine the species present, titration experi-ments were conducted in which the fCI concentration was kept at 1.0 m and the amount of bse added was limited arbitrarily to about 8% of the KCia concentration.

Some 1ediu change is inherent in such studies whereothe equilibria cdn be observed only at relatviely high concentrations.

Data were obtained from titrations at 50.3a 100.0, and 199.8w. These results are sum.

marLzed in Table II and Figure 2.

(22) D. 8. Byrne., "Some PhYsteohomlvical Studies of Boric Acld Solu.

tlons at HIgh Tasuptroturs,*' Rleport No. 3713, 'Wstinghouse Corp., Atomic Power Division, 1912.

Ir,*ganic Chemistry, Vol. 11, No. 3, 1972 539

,.52 + A 'r I

2.04 2.35 2.302120.

2.70 3.40

-I 4.30 4.20 3._ 1 0

0.2 0.4 0.6 0.18 1.0 1.2 Figure 1.-Log Qi,£ as a function of Ionic strength.

The curves were drawn using the equation in Table III. Thes dark rectangle at the intercept at 500 is taken from the data of ref 7.

TABLz ii Pocrxnouzacl DATA 0ON Boxic Acin SOLtTrbooo V1 1 M KCI*

I 22 1~

0

.8%

If 102 -a-to

.61(cm.-1 02 X

• 0so..

'a o aomi C.all 0.3612 1241 41804 9.603 0.JI39 1.144 9,3114 3.934 a.llow 0.041 0.2)11 7.249 122201 4464 0.1241 10*1 03234 1: 01

.4I20 4.203 I303 0.I 000 899 402 0.9231

.22 090011 4.4 819044 7.2 19 0*112 92 6,16042 9.990 849 0Mt 1..97 0.`31 0.00 UM0 I.644 0*29 3.022

$il4 I. got Vi.3l 1466 3.744 1211 13.31 1.706 3.327 I.7M 4S24 10.73 431`1 3.49V 0ist 1.202 0.723 0.47)

I.001

.Sid 41,114 kill 1.11,1 l,111 this 3.111 1.411 21) 7044 2.202 1361 JLCII It02 as

.60 0.023 06l1 Ilm 1 03 031 So

.7 All Al 9174,

.424 1.111

-inI 2.21 3.00

.424 2.4 V499 S-011 4474 4.144 3.913 3.341 2,741 3.340

&t24 1.013 2.7412 11A44 5.72" 1.342 4.917 4A44 4.411 4.243 4.139 I.600

$Igo 3.111 list 8523 1.772

.37?1 2.009 I'Ml

$.123 0.3 2) 3.004 4:32

4. 64 4.2to 4.037 6.1319 9.411 4.1111 5.444 4.016 9.119 f.ll) 0411 1310a 5.7I" 3,244 2.344 0.200 6.140 4.I00 6.205 0.23) 24101 1,164 0.042 0.$152 0.00)2 4,144 4.124 I I.I 3412 aid

.1164 9420 9.711S till 96449 2.214 IIIt 2.190 f*92 7,217 7471 0.371 9.$94

).720 0Jit title 8044 0.92 31441 11IIII 2.110 LI1M 4.111 Site 1.11$

L13;OL 0.003 114 0,0n 0.,42

.11"4 I.Im 10)4 1.)69 0.432 0ife

200A, 00433 9.0,4 22 01142 6.1411 0.3400 0.3274 0..114 11221 09001 0.9 79 0.0)0 0.9104 49ll3 0.14M7 0Itllit life 2912 2.941 I Vll, 1.077 2.779

)"IM 30141 7343 3307 3.47" adds 340f 21721 3,10 0*44 1406 3.44 2272 5.946

$404

$412 3103 V.30 1.314 1117 solo7 Satz 4.341 8007 2.701 I2.52 3.W1 Ltit 3,237 3.337 3,023 3A14 2.111 4fil LIS]2 4.723 1.III

?A]$1 1.173 1.1ol Lot 1 1.111 5.71J 0.272 20"1 0.331 3049 1,470 1.127 Left 2.776 2.02*

Last7 3.402 LI1 fil0l 4.167 4.2M

.3.70 2.323 4,267

1130 4,046

.19 7.1103 3.741 1.309 i3414 130,1 1.4*4 1aos 2*.2U 0.70 3.24) 3.4)91

$Adfa 1

.9 0.444 3.2)3 1.406 0

.)

0*72 0)*0 4.240 9.00 414' 0.210 4116 0141*

4.224 o.701 4.37 0.192 4.l13 1.014 fall IJ*S 3,4

.042 5.004 Coda0 3.044 9.Mff 6.144 0.301 4.1711 0.41 6.l30 11*11 4A71

@.142 4aid 1041 4.429

.i09 t150 2.231 411130 0.329 0.1331

.20421 9.lgll 4.3321 2.2223 412.190 8,1164

0. I22 1.0104111 10030 0""0 08009 0 00*2 0.99l 0.11" e 1449 00000 SIIA1" 0u44 1.942 l.s, 04274 I IIT I i.IM 02443 9.039 4.700

$6111 I'll 1344 00093 0.143 101M Seat0 Site3 a.St.

L.32 2.Sol 9637 3007

$1291 2.74 0

$III9 soil 4all 4.34 2 3.493 1202 M41l L+OU 3.64) 2.11}

I.0)

Live ti.ll 3.4111 Xi42 1.130 3.023

'040, 1116 0.002 tail 5.342 IA41 t1.841 SAO?

4, IfI 4.131 3.731 3.l44 0.342 2.23*

lilt 1.04 loss Lill

.4411 1259 374 1' 2447 U3412 4.720 llt 17111 16702 144n 1225 0,* Is the ligand number and ma is the boron mrolality.

540 Inorganic Chemistry, VOL 11, No. 3, 1972 Mzsmxit., BAES, AND SwaL-rot; 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

0 0.03, m 50.W/

100.0y 199.81 7

6 5

4 a

2

-log COHi)

Q.., were used to calculate a value of A for each data point by means of the expression R.- (ZYQ..vrtB(OH)sds1OHiV)/Ma (8) wherein the free boric acid concenttation w s obtained by reiters-tive solution of the material balance expretalon 113(OH~sI Zx.,A(H)][Oi (9)

Figure 2.-The effect of boron concentration and temperature on the hydrolysis of boric acid in I m KCL The approximate boron concentrations are listed above and the curves were cal-culated from the equilibrium quotients given by scheme II In TableVI.

Data Reduction and Analyses.-The potentials E for the cell (3) are given by the expression" E - (RT/F) In ([OH-]/[OH-j,) -

Doz([LOI1r -

[OH]) -

ZDiQi1, -

[1i) (4) where (OH] and (i) denote, respectively, the concentration of hydroxide ion and of each other ionic species in the solution. The subscript r refers to the reference solution.

The liquid junction potential is given by the terms on the right containing Dow and D, which arc calculated from the Hender-son equation.s The limiting equivalent conductances at high temperatures needed for the calculation of Dow and Di were ob-tained from Quist and Marshall.11 Also the approximation that the equivalent conductance of B(OH),- equals that of Cl-was used, with a large allowance (20%) for error from this assump-tion in the error analysis. The differences ([OH-1, -

EOH]) and (I], -

[i]) were all small compared to the ionic strength and as a result the maximum liquid junction potentials were about 1 mV.

The data were analyzed and interpreted according to the treatment presented previously for sinilar studies"s but here in terms of the average number of OH-ions, 9, bound to a central moiety (B(OH).).

This was derived from the data using eq 5 A -

([HW] + mcl - (OH-I)/ma (5) where mor and ma are the stoichiometric concentrations of base and boron in solution.

The OH-concentration in the solution was obtained by a reiterative solution of eq 4.

The H+ concen-tration, usually a small or negligible contribution to R, was then obtained from the estimated value of the dissociation quotient for water."

For the evaluation of Q1,1 from dilute boric acid experiments (Table I) the following relationship was used log Q1.1 - log [1-())OH The error Introduced by neglect of small amounts of polymers on this calculation of the value of log Q,,1 was less than the assigned experimental errors in Table I.

The analysis of the data from more concentrated boron solu-tions in terms of polymeric species was performed by the usual procedure:

(I) A set of equilibria xB (OH)a + yOH-B.(OH).,+,i-(7) was chosen, corresponding to a scheme of hydrolysis products.

(2) Trial values of the corresponding equilibrium quotients (23) A. a. Quist and W. L. Marshall. J. Phyi. Chem., 19, 2984 (1985).

(24) C. F. Baet, Jr.. N. 3.T Meyer. and C. E. Roberts. Ixorg. Chim., 4, 515 (1965): R. E. Mesmer and C. F. Data, Jr.. ibid., t, 1931 (1907).

(3) The Q.., values were adjusted until the best agreement was obtained between calculated and observed R values for all the data.

As will be described in the Discussion, a scheme was sought containing a minimum number of species BD(OH)s.a+'

which gives a satisfactory fit to the data.

Weighting of Data.-The data used in the least-squares analy-ses were weighted according to the assigned experimental errors.

Estimated errors were assigned to 17 Independently measured quantities Involved In solution makeup and in volume, potentlo-metric, and temperature measurements.

The effect of these errors on the quantity (ff - 0,) for each dita point was obtained by numerical differentiation.

Weights (VV) were then obtained for use in the least-squares procedure by summing all 17 vari-ances which could then be calculated to obtain the variance in (41 -- f) for each point.

The agreement factor defined in eq 10 was then used as the criterion for the schemes tested.

The weil:ht W is the reciprocal a (), iZGV(a - 1,) 2)/(oN. -

,1,OPI, (10) of the variance for the data point and (24 - Y,) Is the difference in the number of observations and the number of variables.

The a(A) should be unity when the weights are accurately set and the data fit the model exactly.

Discussion B(OH)r-B(OH)1-Equillbrlum.-Figure 1 shows the dependence of log Q1.1 for equilibrium 1 on I up to an ionic strength of unity at rounded temperatures.26 The equilibrium quotient has a small dependence on ionic strength and this dependence! decreases as the temperature increases.

A relatively small dependence of the equilibrium quo-tient on ionic strength was expectel since it depends only oil the ratio 'YOH-/yn(0o),--

For this ratio we have elected to use the two-term expression log (-to-/'cox,,-) -

+l

- bY'/,

(I1)

This simple relationship results when the usual Debye-Hilckel expression for this ratio is expanded as a power series in I.21 This expression is not to be construed as a uniquely significant form but simply an adequate description of the data.

A complete analysis of the temperature dependence of the log QI.i requires a knowledge of the pressure co-efficient of the equilibrium quotient since the total pressure in these experiments varies from about 500 to 1900 psi.

By experiment it was found that within the experimental error of the measurements (-'0.004 unit in log Q) there was no effect o0 a change of 1300 psi at 50° and 1400 psi at 200".

The upper limit for the absolute magnitude of AV for the reaction based on these data is 3-4 cmr/mol. The pressure coeffi-cient is therefore not a factor in setting the standard state for reaction 1 when expressing thermodynamic quantities at pressures below about 150 atm.

Table III shows the least-squares results of analyses (25) The small correction for temperature (less than 29) was made by tting the data at a given Ionia strength to the mxprisson log OQ.i - -A/F +

B + Cr -

D tog r and then using the derivative (d log Q/d(1/7)) to cal-culate the correction In log Q,*t.

(26) If the Ionic strength dependence for the a-tivlty coefficient of each Ion has the form log 7i -

--. Z1"/(l. + #if"e) 1." q from Debye-Edeekel theory and it 0 and a are different for the two Ions by M5 and As, then log (ton-/vaco:,-) - (a. - a6P)t + (PsOA +.-8.)'11 +..

BoRic ACID EQUILIpRIA TABLD III Frrs To *na Q1.i DATA OurATNZD WInt SZVIRAL AssumnroNS FOR mx Foxi or Tri TzimtsATruu DPBPNOBNCZ Agreement

Cam, AC, z

£a4tor 1

c (i, + i:,r) + sj'/s 2.62 2

cT (i

r

+

iT).+ i',

1.98 so C1 + CST (01 + ij)r + Qi1',

0.97 4

C, + C'T Uid "

'/'

1.*75

'Cue 8 gives log Qs.i - 1573.21/T + 28.6059 + 0.012078r'-

13.2258 log r + (0.3250 - 0.00033T)! - 0.00121P/.

according to several assumptions for the form of the temperature dependence AC, and for the temperature dependence of the ionic strength terms. The agree-ment factors (from eq 10) slhown are in terms of the estimated experimental error in log Q.

Case 3, wlhere the AC, is given linear dependence on temperature and one of the ionic strength terms is made temperature dependent, gives a satisfactory fit to the data. The other cases give poorer fits with pronounced systematic deviations.

Thermodynamilc Quantities at Infinite Dilutlon.-The thermodynamic quantities (Table *IV) for reaction 1 TAiLE IV THERMODYNAM[C PAXAMZTBRS AT INPiNrrI DILu'rzO DBRVIvD NOR Tim EgUIL=IDUM' B(OH)a + 011

ý 33(01`1)-

Ttmp,

.C 0

28 75 100 125 150 175 200 225 250 275 300

AMH, kcal MtooP°"

--10.25 -1: 0.13

--10.12 :E 0.08

-9.92 :6 0.05

-9.66 + 0.05

-9.81

E 0.05

-8.90 d: 0.06

-8.42 1h 0.06

-7.88 +b 0.08

-7.26 =- 0.08

-6.58 Z- 0.11

-5.82.I 0.15

-5.00 :L 0.22

-4.11 - 0.29 A3. Cal del-' toot'

-12.85 : 0.48

--12.18 4: 0.27

-11.54

I_ 0.18

-10.73

L 0.13

-9.79 :; 0.14

-8.73 + 0.18

-7.57 + 0.16

-6.32 : 0.18

-4.98

1 0.18

-3;57 d: 0.23

-2.09 + 0.32

-0.58 1 0.43 1.04 + 0.58 AC,*. Cal dq*'

mool-3.9 :k 2.6 6.7 +'2.1 9;4 :L 1.6 12.2

• 1.2 15.0 L- 0.8 17.7

• 0.7 20.5:4: 0.8 23.3 z" 1.2 20.0 -1 1.6 28.8 =- 2.1 31.5

• 2.6 34.3 8 3.0 37.7

• 3.5 Itorganic Ch.mistry, Vol. 11, No. 3, 1972 541 the much more negative value of the AS for the former reaction compared* to the latter (--12.2 vs. 19.3 cal deg-' mol-1).

The AC, values calculatel for reaction 1 using the expression from case 3, Table III, change from 3.9 6al deg-' mol-1 at 0( to 87.7 cal deg-' mol-h at3000.

Polyborate Equllibrla.-That polyborates do exist in aqueous solutions is clearly evident in Figure 2 from the dependence of the ft vs. -log

[OH"] curves on boron concentration above about 0.03 m at each of the three temperatures shown.

This was also" shown by the previous work at 250 and especially by the exten.

sive work of Ingri is a number of'different salt media.

Trends with increasing temperature can be seen (Fig-ure 2) in the shift of equilibria to higher hydroxide con-centrations and also in the decreasing spread of the curves With, boron concentration; these reflect a de-creasing amount of polymerization.

The principal species which have been invoked by Ingri in analysis of his data are B(OH)4-, BjO,(OH).-,

"BOO,(OH)al-, and BOh(OH),.-. In discussing such species we will use the notation (x,y) to represent the speclis according to the formula B,(OH)#,.,*-. Hence, thi polyborate species of Ingri become (3, 1)-Bj(OH) 10-,

(3,2)-B:(O.Hh t-,. and (4,2)-B4(OH)L-4.

Of coui'se, since no information can' be derived from potentio-metric studies alone regarding the hydration of such species, it is optional whether one writes BOs(OH) 4-or Bs(OH)hc-, etc.

Because the present data are of a precision compar-able to that of Ingri and, in particular, because they cover a wide temperature range, it should be possible to test rather sensitively for the best scheme of poly.

borate species.

By "best scheme" we maean the sim-plest scheme consistent with all the data.

If such a scheme involves a sufficiently small number of spe-cies, it is likely to be the correct one, since an incorrect scheme with the same small number of species is not likely to fit data covering a wide range of conditions.

In short, the simplest consistent scheme has the best chance of being the correct one.

The data of Ingri at 25* and our own data indicate that the maximum value of se is 1 and hence 'that no species exists which has a y/x ratio greater than unity.

In addition, Ingri's self-medium study at, very high concentrations of boron (2.5 M), show that at i > 0.9 the predominant polyborate species has an x -*y dif-ference of 2.0.

Applying' these constraints we have limited our ansaysis to the following array of 14 possible polymeric species:

(XY) -(2,0).

(21), (2,2); (3,0), (3,1), (3,2),

(3,3); (4,0), (4,1), (4,;2), (4,3);, (5,1), (5,2), (5,3).

In each icheme tested we have,' of course, included P(OH)4 -, (1,1), whidh has bien well ests.blished as thie only mononuclear borate species by the data in dilute solutions:

Least-squares analyses of the data atI 50, wherein we assumed that only one polymeric species was formed, not surprisingly failed in every case to give a satisfactoty fit, thus indicating agreement with.ngri that at least two polymeric species are formed.

Least-squares analyses were next performed for all the possible pairs from the above array with the data

&t each of the three temperatures.

The results as rep-resented by the agreement factors for the four best 6 The uncertalinties listed correspond to Sa, wherein, is based on the fit obtained with the expression in Table II1', derived from case 3.

were derived from the values of Q1,1 at I -

0 (i.e., the equilibrium constant K1,j) as given by the expression frbm case 3 in Table III. The uncertainties are three standard deviations as calculated, from the experi-mental, errors assigned to individual determinations of log Q1.:. This should allow for any systematic er-rors in the derived thermodynamic quantities intro-duced by the restriction'that AC, changes linearly with temperature.

The value of AHls,. in Table IV is -10.12

-* 0.08 kcal mol-1.

Although there have been no precision calorimetric measurements of this quantity, Harri*e*

has recently reported the nearly identical value of

- 10.2

• 0.2 kcal mol-t based on a thermometric titra-tion procedure.

As is general for weak acids this heat is less than the heat of neutralization of a strong acid

(-13.34 kcal mol-1).*

The formation of borate is opposed by the AS* and favored by the Al°. Since the addition of an OH-ion to boric acid does not lead to a charge neutralization whereas the neutralization of the hydronium ion does, this probably accounts for (27) R. I. X. Harr.a. Talasd.,

I. 1348 (1968).

(28) C. I. Vaudamtu and 3. A. Swat.., J3. Phys. Ck~m.,,

7. 260S (1963).

542 Inorganic Chemistry, Vol. 11, No. 3, 1972 TABLE V FIT OP THREE-SPECIDS MODILS EACH INCLUDING THE (1,1) AND (3,1) SPZCIZS ALONO WITH A Taiso Third Third specie

(,O spedes 6(,*)------

(z.Y) 50.3' 100.0o 299.8' (UY) 30.3" 100.0" 195.3" 2,1 2.66 1.60 1.44 4,2 2.a7 2.18 2.48 3.2 2.02 1.92 2.12 5,3 1.96 2.19 2.56 schemes are shown at each temperature in Table V.

The important outcome of this analysis is that the (3,1) species is comnmon to all the best schemes and that at the higher temperaturies the (2,1) gives a much better fit as the second polyimeric species than does any of the other three species ((4,2), (3,2), and (5,3)).

At the lowest temperature, '50.3", each of the last three species gives a significantly better fit than does the (2,1) species.

MEnMER, BAES, AND SWESTON of the present data should be in tesms of the scheme or the scheme (1.1), (2,1), (3,1), (4.2)

(1,1), (2,1), (3,1), (5,3)

(I)

(11) wherein we expect that the fourth spe-cies will be minor at the higher temperatures while the second species will be minor at the lower temperaturess.

This expectation is borne out by the least-squares analysis in terms of schemes I and II given in Table VI. The relative importance of each species in fitting the data may be judged'by the uncertainty generated for its stability cdnstant. It is noteworthy that the contributions of the (4,2) and the (5,3) species are so similar 'that replacement of one by the other has no TAsLn VI LoG Q,,, VALL-s FROMd DATA IN 1 M KCl

[I-.

.25*'

200.5' 109.80 1,1 2.1 3,1 4,2 1,1 2,1 3,1 5,3 r(u) 5.278 =L 0.004 5.95 d= 0.06 7.319 + 0.002 13.29 :- 0.07 6.279 - 0.003 6.02'+ 0.04 7.378 : 0.003 19.20 :f 0.08 Scheme I 4.301 - 0.004 3.8 : 0.4 5.08 - 0.01 10.78

• 0.11 2.38 Scheme 1I 4,296 -i 0.003 3.7 -4 0.4 5.997 + 0.008 16.438 0.07' 1.95 Analytical Expressions 3.428 + 0.002 3.46 ;- 0.06 4.707 z- 0.008 8.1

• 0.2 1.55 3.427 :L 0.002 3,43 =6 0.05 4.712 :L 0.008 12.9 :1 0.1 1.47 2.332 *: 0.002 2.47 L 0.04 2.98 -L 0.04 6.2

• 0.0 1.45 2.332 ! 0.002 2.47 -*- 0.04 2.98 E 0.04 8.6 -- 0.6 1.45 log Q,.11 -

1

+ 28.8397 + 0.011748T -

13.2258 log T 2756.1I log Q1,1d -

18.96 + 5.835 log V.

3389.5 log Qo,t 4 -

8.084 + 1.497 log T log Q&.

2 2 -

134.56 + 42.106 log T log Qm,, -

118.115 + 36.237 log r

• These values are derived from Ingri In 3 M NsClO,.

Because of the difference in medium, these quotients a'e not included in the derivation of the analytical expreslons.

6 Based on data from Tables I and II. 0 Based on scheme I. ' Based on scheme II.

The conclusion to be drawn from this analysis then is that while two polymeric species can account for the data at a single teniperature, quite clearly more than two are required to account for all the data.

Moreover, the identity of tihe correct species at the highest temperature seems quite 'clear in' view of the uniquely good fit given by the scheme (1,1), (2,1),

(3,1). At 50*, however, an additional polyhmeric spe.

cies evidently becomes important while the (2,1) spe-cies becomes less important. Since it is knqwn from Ingri's "selfrmedium" results tfiat k polymeric species is formed at high ft values in which the difference x --

y, equals 2.0, it is not surprising that improved fits to the lower temperature data are obtained whern the (4,2) or the (5;3) species replaces' the (2,1) species.

Thus it seems clear that the best overall interpretation significant effect on the formation quotients of the other species.

Schemes I and II were also used to fit Ingri's un.

weighted data in 3 M N*C1iO 4 at 2W°.

Both schemes I and II were found to give fits as good as Ingri's--(1,1),

(3,1), '(3,2);

and (4,2).

The values for log Q2.1, however, are somewhat higher than expected based on our results in the chloride medium.* Regarding.

the question of which one of 'these alternative four-species schemes is the correct one, we believe that scheme I or II is a better choice t.an Ingri's scheme becatise each is consistent with all the present data and Ingri's is not.

The analytical expression for the temperature de-pendence of the log Q,,, for each species is also given in Table VI. Thermodynamic parameters calculated

Bonrc ACID EQUU4ERL&

from these expressions at 298'K are given in Table VII (uncerýainties are la). The variations of the poly-mer distribution with concentration of boron and with temperature are shown in Figure 3 at 50 and 200*.

TAsLN VII THENRMODYNAXIC QUANTiTiB $OBo BORATE fEUILUJRa1A AT 2980 xN 1 KCI XB(OH), + YOH-'ý E.(OH),+,v-

ARM, AIRM.

fl kcal moI'-

Aslm, eu x,

kcul inol "-

4Sm, eu 1,1

-10.3 sh 0.2

-12.0 "k 0.6 4,2

-34 s, 6

-85 k 19 2,1

-9.2 +/-.

-9.1 3

5.3

-43 d-4

-SS m 14 8.i

-- 14.4 -k 0.3

-17.1 A. l Inorgqcni Chemistry, Vol. 11, No. 3, 1972 543 (ON),-

B(0O1H)o-by -OH-bridging, the positive AS associated with the first step (2.9 1 3 eu from Table VII) is in contrast to the negative AS associated with the combination of OH-ion and'B(OH)s (-12.0.* 0.9 ep)'and hence would seem more consistent with the formatioi d6 an bridge HO

)H B(OH),

+ B(0H),

Z=

B0 P

He~

+ HO 100 P0 I.0 0.050M BOMO~N 40 20 0.0 OPT BORON L.

,ICC 62 60 40 20 0

200*

0.050," BORON

.0 31 9

-1

~

9 7'

-log 10H2 since this is accompanied by the additional entropy contributed by the'molecule of water liberated. More-over, the tribbrate species is more stable than would be predicted from the stability of the dimer if 'a stepwise addition process were all that Was occurring. This sta-bility of the trinier led Ingri to suggest that it has the ring st.ruc'ture HOe (O.r <OH Such a structure is consistent with the following rules prbposed by Edwards and Ross!' for the structures of crystalline hydrated polyborates.

(i) Boron atoms exist in threefold aind fourfold coordinaticn; the former having a neutral charge and tlie latter* a 1

.charge.

(Hi) The basic' structure ofthe pblyborates is a six-membered ring with alternate boron and oxygen atoms.

(iii) To. be 'table the' ring must contain one or two tetrahedral boron atoms.

(iv) Other discrete anions may beform*d by the fusion of tw6"rin;s at a tetra-hedral boron atom.

(v) Long-chain po.yanlons may be fomned from the rings by repeated dehydration.

When applied to the polyborate ions which appear' to exist in aqueous solution, these rules lead t'the above structures for the dimtr and trimer and :to" the follow.

Ing stxuctures for the remaininig alteirnative7 species Figure 8.-Distribution of species calculated for solutions con-taining 0.05 and 0.60 m boron at 50 and 200'. The species are represented by the notation (x,y) r*o the formula B.(OH),..w-."

As the temperature increases, there Is a decrease in the amount of polyborates at a given boron concen-tration and A value, as was evidenced by the reduction in spread of the curves in Pigure 2.'

Stability of Polyborates.-The reality of the species B(0H)i-.can scarcely lie doubted in view of the ex-tensive data here, as well as elsewhere, which support equilibrium 1.' The reality of the (3,1) species, while somewhat less certain, is nonetheless highly'probable since it was found in all the better schemes at all tem-peratures in the present study and had previously been proposed by Ingri to interpret his data at 250.

The (2,1) species also seems highly probable in view of the fact that it, along ýwith the (1,1) and the (3,1) species, is uniquely sufficient to explain the present data above 100.' The identity of the fourth species (e.g., (4,2) or (5,3)) is relatively much less certain.

While the sequence of the first two polynuclear spe-cies might suggest the successive -addition of (0H),

groups to the orthoborate ion Finally, rule (iii) is consistent with the absence of a cyclic trinier of.boric. add (BsO0(Ofl),) in aqueous solu-tioli.

(29) 3. 0. Edwards &ad V. Ro, J'...*ar. NXxd. Chin., 15, 229 (1950).

ENCLOSURE 3 LIST OF COMMITMENTS

5928-07-20202 Page 1 of 1

SUMMARY

OF AMERGEN COMMITMENTS The following table identifies regulatory commitments made in this document by ArnerGen. (Any other actions discussed in the submittal represent intended or planned actions by AmerGen. They are described to the NRC for the NRC's information and are not regulatory commitments.)

COMMITMENT TYPE COMMITMENT COMMITTED DATE ONE-TIME OR "OUTAGE" ACTION PROGRAMMATIC (Yes/No)

(Yes/No)

The elements and methodology of the proposed TS Table 4.1-5, Item 2.b Upon implementation No Yes surveillance testing will be clearly of amendment.

delineated in station surveillance procedures.

Procedures will ensure that samples Upon implementation No Yes are taken within the basket at both of amendment.

interior and peripheral locations to obtain a representative sample.

Use of the WCAP 16530-NP model is Upon implementation Yes No being considered, and if utilized, this of amendment.

issue (ref Q.9) will be addressed in the TMI Unit 1 Generic Letter 2004-02

,response.