ML20087N298
| ML20087N298 | |
| Person / Time | |
|---|---|
| Site: | Brunswick  |
| Issue date: | 03/20/1984 |
| From: | Zimmerman S CAROLINA POWER & LIGHT CO. |
| To: | Vassallo D Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 8404030312 | |
| Download: ML20087N298 (28) | |
Text
.
s
.) 1 CD&L Carolina Power & Light Company MAR 2 01984 Director of Nuclear-Reactor Regulation
-- Attention:
Mr. D. B. Vassallo, Chief Operating Reactors Branch No. 2 Division of Licensing
~
United States Nuclear Regulatory Commission Washington, DC 20555.
BRUNSWICK' STEAM ELECTRIC PLANT, UNIT NOS. 1 AND 2 DOCKET NOS. 50-325 & 50-324/ LICENSE NOS. DPR-71 & DPR-62 REQUEST.FOR ADDITIONAL INFORMATION -
' POST ACCIDENT SAMPLING SYSTEM
Dear Mr. Vassallo:
As a result of a telephone conversation with your staff on January 20, 1984,
~ Carolina Power & Light Company (CP&L) was asked to clarify certain statements
.made concerning radiation effects on instrument accuracy in CP&L letter dateI December. 31,'1983 (LAP-83-579)..
Please find enclosed the information requested. Should you have any questions
-concerning this letter, do not hesitate to contact a member of our Licensing Staff.
Yours very truly, I
S. R. Zimmerman Manager Nuclear Licensing Section PPC/ccc (9616PPC)
Enclosures cc:'
Mr. D. O. Myers,(NRC-BSEP):
Mr. J. P. O'Reilly (NRC-RII)
Mr. M.-Grotenhuis ~(NRC) 8404030312 840320 PDR ADOCK 05000324 P.
PDR 411 Fayetteville Street
- P. O. Box 1551
- Raleigh, N. C. 27602 y
?
,s.
REQUEST.FOR ADDITIONAL INFORMATION POST ~ ACCIDENT SAMPLING NRC Comment Provide additional ~information demonstrating applicability of laboratory instrumentation in the. post-accident water chemistry radiation environment.
- CP&L Response Hhdrogenor'TotalGasAnalysis: Hydrogen or total gas analysis will be
- performed-by. gas chromatography (GC). The GOW-MAC, series 550 GC (or equivalent)'will be used. Adverse radiation effects on the GC components are
. not expected. The recording device used for the 19C ' analysis will be located at a. distance that will reduce the radiation levels well below these
~
enticipated.for-post-accident samples. Gas chromatography is the method recommended in NEDC-30088.' See attached reference - NEDC-30088, Page 3-3.
pH Measurement: The pH measurement (if required) will be made using a pH electrode,. Cole Palmer catalog no. C-5990-45 or equivalent. This electrode has been successfully demonstrated for the determination of pH in the presence
~ of gamma radiation through test and' analysis. See attached reference -
l NEDC-30088, Pages 6-16, 6-17, D-1, and D-6.
. Isotopic: Analysis:. Analysis by gamma spectroscopy will be conducted using extended shelves,~ sample dilutions and shielding as appropriate to reduce the
~ radiation levels present. See attached reference - NEDC-30088, Pages 6-11.
Boron and Chloride Analysis: Analysis for boron and chloride will be
-performed by ion chromatography.- The. ion chromatograph uses a fixed sample J yolume of approximately 100 pl.
Adverse radiation effects on the components in thex1C'are.not anticipated. -Litegaturescudies'showthatthecationresins
- begin
- to degrade at approximately 10 rads, that the electronic components are 5
. resistant to radiation exposure well'above 10 rada, and that elastomers of the type usgd to make 0-rings and silican rubber ' sealing compounds are stable to about: 10 rads. These values are well above those anticipated to be encountered by-the:IC during sample analysis. See the following attached references:
I.
NEDC-30088, Pages 6-4, 6-13, and 6-14.
1.
2.
Evaluation of the GE and SEC Chemical Procedures for Post-Accident Analysis of Reactor Coolant Samples - Prepared by Exxon Nuclear Idaho
- Company,;Inc., for the Nuclear Regulatory Commission, Pages 24, 25, and 26..
- 3. L Analysis L for ph, Chlorides, Dissolved Oxygen, Conductivity and Boron Under
" Post-Accident" Conditions - Prepared by NUS Corporation, Pages 5, 9, 10,
~
.and 11.-
f p _-
(
l'
REQUEST FOR ADDITIONAL INFORMATION POST ACCIDENT SAMPLING NRC Comment
_s
-Provide additional information demonstrating applicability of laboratory instrumentation in the post-accident water chemistry radiation environment.
CP&L Response Hydrogen or Total Gas Analysis: Hydrogen or total gas analysis will be performed by gas chromatography (GC). The GOW-MAC, series 550 GC (or equivalent) will be used. Adverse radiation effects on the GC components are not expected. The recording device used for the GC analysis will be located at a distance that will reduce the radiation levels well below these anticipated for post-accident samples. Gas chromatography is the method recommended in NEDC-30088. See attached reference - NEDC-30088, Page 3-3.
pH Measurement: The pH measurement (if required) will be made using a pH electrode, Cole Palmer catalog no. C-5990-45 or equivalent. This electrode has been successfully demonstrated for the determination of pH in the presence of gamma radiation through test and analysis. See attached reference -
NEDC-30088, Pages 6-16, 6-17, D-1, and D-6.
Isotopic Analysis: Analysis by gamma spectroscopy will be conducted using extended shelves, sample dilutions and shielding as appropriate to reduce the
. radiation levels present. See attached reference - NEDC-30088, Pages 6-11.
Boron and-Chloride Analysis: Analysis for boron and chloride will be performed by ion chromatography. The ion chromatograph uses a fixed sample volume of approximately 100 pl.
Adverse radiation effects on the components in the IC are not anticipa'ted. Literature studies show that the cation resins 8
begin to degrade at approximately 10 rads, that the electronic components are 5
resistant to radiation exposure well above 10 rads, and that elastomers of thetypeusgdtomake0-ringsandsilicanrubbersealingcompoundsarestable to about 10 rads. These values are well above those anticipated to be encountered by the IC during sample analysis. See the following attached references:
1.
NEDC-30088, Pages 6-4, 6-13, and 6-14.
l 2.
Evaluation of the GE and SEC Chemical Procedures for Post-Accident Analysis of Reactor Coolant Samples - Prepared by Exxon Nuclear Idaho Company, Inc., for the Nuclear Regulatory Commission, Pages 24, 25, and 26.
3.
Analysis for ph, Chlorides, Dissolved Oxygen, Conductivity and Boron Under
" Post-Accident" Conditions - Prepared by NUS Corporation, Pages 5, 9, 10, and 11.
W l
i
3
=
NEDC.30088 APRIL 1983
'd 4
,i. _
.x._,,
a RESPONSES TO NRC POST lMPLEMENTATION REVIEW CRITERIA FOR POST-ACCIDENT SAMPLING. SYSTEM D
.. A
, ~
>^
,j
[d yne
=,.
r e. c -
- a.,f.,<
,y pyp-
,..,. g.._
g, y3 -
y.
.,7, g,.
,.y,.7 e
- -. - - - ~ ~
,.. ~~; 4...
,. c l:.
? * '.,.; ~,
1 '.. :^. *
.r,,
i,.
,1
- ,a f.-
-4 GENERAL h ELECTRIC
g NEDC-30038 (This section is to be added to the PASS Manual) 4.5 0xygen Analysis - Residual Hydrogen Method I}
In order to accom.modate ALARA considerations, dissolved oxygen will be measured indirectly, whenever possible, using the residual hydrogen method.
Using this method, dissolved oxygen is verified to E
be <0.1 ppm by measurement of a positive hydrogen residual of >10 cc/kg.
Dissolved hydrogen concentration is determined by ga's ~
- [
chromatography as described in Section 7.
Gas chromatography'has 11 been successfully demonstrated for the determination of hydrogen in
[
the presence of gamma radiation through testing and analysis by l
Babcock and Wilcox on TMI-2 post-accident gas samples.
ii
?
i 1.
+
)
i i
k O
l
(
l 3-3 /3-4
,, jp_,
6.2 DISCUSSION 6.2.1 Review of PASS System Th'e General Electric Post Accident Sample Station (1) was developed by the General Electric Company and a group of BWR owners in response to the requirements of NUREG-0578 (2) and subsequently NUREG-0737 (3).
The system was designed to provide liquid and gaseous samples from one primary system, from the suppression pool, and from the primary containment atmosphere following a major reactor accident accompanied by a release of fission products as defined in Regulatory Guide 1.3 (4).
The development of the design specifications for this system involved a definite decision to avoid the use of inline analytical instrumentation, except for a conductivity monitor.
The inline instrumentation was considered to have uncertain reliability under accident conditions.
Instead, the emphasis'was placed on the ability to obtain grap samples for laboratory analysis.
Section 7 to Reference 1, includes suggested analytical procedures to meet the requirements of NUREG-0737 (3).
It was the specific intent that only the analysis of immediate need would be performed at the reactor site, and the balance would be performed at an offiste, commercial analytical fa,cility having the capability of processing multicurie samples.
In several cases, it was intended that the onsite analysis would be of a scoping nature with more accurate analyses subsequently being perfomred at the offsite facility.
This intent has been somewhat modified, depending on the particular reactor site and the various NRC clarification letters, with several of the sites having purchased ion chromatographs for boron and chloride analyses.
The PASS system was designed to obtain a variety of samples:
2 1.
A 0.1 milliliter primary coolant or suppression pool sample diluted to 10.0 milliliters by the processing of flushing the sample valve contents into the sample system bottle.
This sample is intended for radiochemical analysis and for initial site chemical analysis.
6 '
T-NEDC-30088 6,
b c
[
even under the best of conditions.
Uncertainties on the order of 5% to 10% are more in keeping with the precision for repetitive analysis of a l
sirigle sample.
If the volumetric errors in the small volume sampling *,
dilution, and equipment calibration are included, the overall analytical accuracies are more reasonably 10% to 20%.
In addition, a requirement to 9
j analyze to zero concentration is a poorly definad lisait anli ime.ediataly raises the issue as to how sensitive an analysis procedure raust be used.
The need for an error limit other than zero becomes more acute when sample volumes are very restr.ictive or samples are diluted to reduce exposure.
The error bands given in the clarification letter appear to be reasonable analysis limits; i.e., the boron analysis should be capable of detecting 50 ppm and the hydrogen method 5 cc/kg, etc.
Gross Activity, Gamma Spectra; There should be no problem in meeting the accuracy requirement of a factor of 2 for the major nuclides nor in analyzing the range of activities given in Table 1.
This capability is discussed in Section 6.2.2.4.
Providing the precautions described in Section 7.6 of Reference 1 are followed, the background at the multichannel analyzer should not be a limiting factor for measuring the principal gamma emitters. There will be an increased uncertainty introduced as a result of successive dilutions, but all things considered, the accuracy for analyzing the initial aliquot should be within 30 to 40%.
The major uncertainty lies in the representativeness of the initial sample.
This
~
subject is addressed in other tasks and is beyond the scope of the report. However, is should not be too significant for truly soluble nuclides. A major uncertainty may arise in analysis of the suppression pool water due to sample system contamination if the activity of the primary coolant is much higher than that of the suppression pool.
The effect of cross contamination, however, can be minimized by adequately.
flushing'the PASS between samples.
" Tests of the small volume samples (0.10 ml diluted to 10 ml) on the General Electric prototype PASS gave a sample and dilution precision of 15% with_a positive 6% bias in the 0.10 ml aliquot (i.e., 0.106 +0.005 milliliters).
6-11
. ~.
g NEDC-30088 interfere when present in concentrations greater than 25 ppm while fluoride causes negative interference due to the formation of fluoroborate
, iori.
These conditions are not expected to occur, however, in the BWR primary system.
Irradiation tests conducted by General Electric had reported in Appendix A, Chap. 7 Reference 1 showed that an energy absorption rate of 4.4 x 105 Rad /hr* during the color development phase of the analysis resulted in an error equivalent to 27 ugm of boron or 270 ppm for. a 0.1 ml coolant sample. Assuming the indication effect is proportional to dose, reducing the exposure to correspond to Reg. Guide 1.3 source terms and a one hour decay would result in only a 1 ppm error for the analysis cf 2 ml of a 100:1 diluted, 0.1 ml primary coolant sample.
Chloride Analysis:
Chloride analysis of highly radioactive solutions with chloride concentrations in the range of 0.05 to 20 ppm cannot be performed in a normal laboratory using conventional techniques.
Although a few of the BWRs using the General Electric PASS are planning to perform the analysis on site using an ion-chromatograph, most of the sites plan to i
send undiluted,10 m1 samples of the primary coolant to an offsite l
l analysis facility.
In the event of an-accident involving extesnive core l
damage, most BWRs would limit the chloride analysis at the site to a scoping analysis of samples of 0.1 ml primary coolant diluted at 10 milliliter volume.
The PASS manual, Section 7.4.3, proposed the turbidimetric method as the most reasonable method for this purpose when handling highly radioactive samples.
This method was taken from Reference 10 and was chosen as it did not involve boiling the sample
^1his cose rate is approximatel a factor of 50 larger than the 8 x 103 Rads /h calculated for a 0.1 ml primary coolant sample in a 25 ml analysis volum when using the Reg. Guide 1.3 source terms and a 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> decay.
The 8 x 102 Rads /h absorption rate is not significantly different from l
the 1 x 104 Rads /h in the criteria (10) matrix.
I 6-13
NEDC-30088 hl e
^
with attendant risk of evolving radioactive iodine gas.
The maximum sensitivity for this method is on the order of 0.1 ppm in the solution to be analyzed.
Since the analysis will be performed on a sample diluted by a factor of ;'0, the initial concentration must be greater than 10 ppm, and more realfstically, considering uncertainties in blank correction for chloride contamination in the diluent and the glassware, l
l t
greater than 20 ppm.
Consequently, the scoping procedure is barely f
capable of analyzing at the maximum specified concentration.
l Depending on the particular procedure, a chloride analysis is subject to several interferences in terms of the element matrix listed in the NRC clarification letter.
On an atom for atom basis, 40 ppm iodide is equivalent to 11 ppm chloride and will interfere in turbidimetric method by the formation of silver iodide.
In addition, iodide is also expected to significantly interfere with the measurement of chloride using a solid state, specific ion electrode.* In ion-chromatography, the boron peak is immediately adjacent to that of chloride.
Considering the high ratio of boron to chloride specified by the matrix and the chloride analysis requirements, there may also be serious boron interference in this method for chloride analysis.** Each utility planning to perform the chloride analysis on site should show that iodide or boron at the matrix specified concentrations will not effect the chloride anlaysis.
- 0rion instrument Co. stated they have developed a specific ion electrode method for measuring chloride which involves elimination of iodide interference by oxidation of iodide to iodine with sodium bromate.
This rethod is worthy of further consideration but will require special apparatus to avoid release of highly radioactive iodine vapor.
- Discussions with Dionex applications group revealed they are working l
on a method involving elution.with 0.002 fj sodium carbonate to separate chloride and borate.
An' application paper should.be available.in several weeks.
~.
g g
l 6-14 I
/Ab Cr would still be 2% hydrogen which is readily measurable by gas chromatography using argon carrier.
There is, however,- the potential problem that attempting to optiinize the analysis for low levels of oxyge'n may seriously effect the hydrogen sensitivity, pH Analysis:
1 Section 7.4.4 of the PASS manual (1) suggested the possibility of using jl pH indicator paper for estimation of pH.
This method, however, will not meet the Table accuracy requirements, and, as also described in Section 7.4.4, cannot be used at the required source term radiation levels due to total destruction of the indicator dye.
Section 7.4.4 also described the possible use of a micro pH probe.
Subsequent testing of this probe were' unsatisfactory even without radiation and it was dropped from further consideration.
Referenca 12 describes the results of irradiation tests of a typical combination pH electrode (Fisher Model 13-639-104).
The electrode performed satisfactorily at gamma flux levels up to 1.3 x 108 R/h in both buffered and unbuffered solutions at pH 4.0, 7.0, and 10.0.
Readings in the buffered and unbuffered acidic solutions were stable over irradiation periods up to an hour.
The unbuffered basic solution showed a decrease of approximately 0.1 pH units per minutes during a 5 minute test period.
This 'ype of electrode would meet the t
accuracy and stability requirements; however, the volume of sample required makes it impractical on the basis 'of personnel radiation exposure at the design basis source terms.
A smaller combination electrode which would use 0.1 to 0.3 milliliters of solution was suggested.
Two types of combination electrodes (semi micro and flat-surface), which would use as little as 0.1-0.2 ml of sample for measurement, were tested under high radiation conditions at the Vallecitos test facility.
The test results are given in Tables 6-2 and 6-3.
The flat-surface electrode performed satisfactorily in the ' highest radiation field (s1.3 x 10e R/h) exhibiting (0.3 pH unit drift in all cases.
For the O
6-16
..I
M y
- NEOC-300eg semi-micro electrode, the drift was less than 0.3 pH unit for a
~
radiation field,of 1 x 105 R/h.
(Note that the standard test matrix
_ recommended by the NRC suggests use of idi R/hr).
It was also found thit the pH readings were not very stable for $1 min. following'the sample' moving to or from the high flux Zone.
In some cases, the pH readings continued to deviate slowly for s30 minutes from the pre-exposure reading.. However, little or'no change in pH readings before or af ter' irradiation for s20 minute's was observed in any of the
' cases.
~
Appendix D provides a revised Subsection 4.4 to Section 7 of the PASS procedures manual and describes the procedure for use of these pH electrodes.
This write-up should replace the existing Section 4.4 discussion.
NUREG-073[ Criteria (5and2c) 6.2.2.3 6.2.2.3.a NUREG-0737 Criteria 5 "The time for a chloride analysis to be performed is dependent upon two factors: (a) if the plant's coolant water is seawater or brackish water and (b) if there is only a single barrier between primary containment systems and the cooling water.
Under both of the above conditions the licensee shall provide for a chloride analysis within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> of the sample being taken.
For all other cases, the licensee shall provide for the analysis to be completed within 4 days.
The chloride analysis does not have to be done on site."
6.2.2.3.b NRC Clarification Letter, Spring 1982
BWRs on sea or brackish water sites, and plants which use sea or brackish water in essential heat exchangers (e.g., shutdown cooling) that have only single barrier protection between the reactor coolant are required to analyze chloride within 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
All other plants have 96 c
i hours to perform a chloride analysis.
Samples diluted by up to a factor 1
i, 1
6-17
' /jb ~
o NEDC-30088 4.4 MEASUREMENT OF pH WITH pH ELECTRODE Because of the reliability, ease, an~d rapidity of measurement, as well as to be consistent with ALARA, two types.o'f combination electrodes (semi micro and flat surface) have been found to be adequai.e to measur_e the pH of primary coolant semples.
As little as 0.'l-0 2 al of undiluted sample is needed for the neasurement.
The flat surface electrode is preferred, a)
For the flat surface. combination e'lectiode, tee sample is taken directly into a regular 15 mi sam;ile vial. Only 0.1 ml of the undiluted sample solution collected at the bottom of the vial is needed.
For the measurement the electrode is inserted fully into the vial
- with the flat surface in contact with the liquid at the bottom of the vial.
b)
For the semi-micro (0.6 cm dia.) electrode; the sample can be taken at the small volume sampler by using an air filled syringe to blow approximately 0.1 ml aliquots two or three times from the sample valve into a centrifuge tube (approx. 1.0 cm dia., 4.7 cm long) contained in a regular 15 m1 sample vial **. Without completely removing the sample vial from the shielding cask, the vial cap is removed, and the electrode can.be inserted into the centrifuge tube in the vial.
The pH is measured directly in the undiluted solution by using a standard pH meter.
"The opening of the vial may be slightly too small for the electrode.
, Minor work may be needed to reduce the diameter of the plastic body of the electrode.
- It is noted that the centrifuge tube insert reduces the effective diameter of the sample vials which makes positioning of the sample vials very critical for insertion of the needles of the PASS.
It is possible to overcome this restriction by bending the two hypodermic needles inward toward the center, and/or by using a sample vial with a larger diameter opening.
l D-1
~
- - - ' ~
l Three types of combination e.lectrodes have been tested under high radiation j
fields.
The test results are described in Appendix A.
All three electrodes e
, performed satisfactorily at 10s R/h radiation field, exhibiting a drift j
t of'less than 0.3 pH units in all buffer solutions.
At 108 R/h radiation 3
field, one electrode showed a greater deviation and poor stability during y
.the test.
It is dif' ficult to identify the source of deviation and unstability.
The true pH in the solution may be affected under high radiation fields. "However, it is believed that with a small volume of j
sample, the radiation fields from the post accident coolant sample 3
(assumed at 104 R/gm) should be much lower than the radiation fieldsto
~
which the electrode was exposed under the test conditions (105-106 R/h).
f i
s 0
9 e- -
i e
a S
g 1
0-2 3
(l[/
nuuu-uvao APPENDIX A EFFECT OF GAMMA IRRADIATION ON PH MEASUREMENTS BY PH ELECTROD B
Three combination electrodesII) were tested with standard buffer soluti of pH 4.01, 7.0, and 10.0 in the Vallecitos Nuclear Center (VNC) gamma irradiation f acility at gamma flux levels up to 1.3x106 R/h.
The standard electrode was also tested with unbuf fered diluted H 50 (pH=4) and NaOH 2
(pH=10) solution.
The electrodes were tested first in the laboratory l
using the integral leads supplied, and then with a 25-ft long coaxial cable, to connect the electrode with a pH meter (Orion's Ionalyzer, Model 801). The long cable was necessary for obtaining readings while the electrode was lowered into the gamma flux in a water pool.
Although the long' lead and poor connections did make the emf sensitive to movement of the cable or electrode and produced unstable readings at times, fairly reproducible pH readings were obtained.
The test electrode was placed in a 5/8-in. I.D. test tube containing s3 al of test solutions.
The pH readings were taken at before irradiation, during irradiation at variable time intervals, and after irradiation.
The results of pH measurements for the buffer and unbuffered solutions are given in Table 1 and 2, respectively.
l (1) Three electrodes are:
1.
Standard polymer body, gelfilled combination electrode with Ag/AgCl reference (Fisher Cat. No. 13-639-104) 2.
Flat surface membrane combination electrode with Ag/AgC1 reference (Fisher Cat. No. 13-039-83) 3.
Epoxy-body, gelfilled combination electrode with Ag/AgCl reference, semi micro size (6 mm dia.) (Cole Parmer Cat. No.
C-5990-45) l D-3 t
^
.'NEDC-30088 hI
~
_. =..
1 e
i In tne buffer solutions, the standard and flat-surface electrodes performed 1
satisfactorily under the highest radiation field (s1.3x108 R/h), exhibiting
,(0.3 pH. unit drift in all cases.
For the semi-micro electrode, the I
maximum drift was near 1 pH unit under the same raditaion field.
- However,
]
the deviation was reduced to {0.3 pH unit when the electrode and solution were irradiated'at a' lower level of radiation field (*1x105 R/h).
It was also found that the pH readings were not very stable for si min following l
the sample moving t6 or from the high flux zone.'
In some cases, the pH f
readings continued to deviate slowly for s30 min from the pre-exposure g
reading.
However, little or no change in pH readings before or after '
~
irradiation for $20 min were observed in any of the cases.
1 In the unbuffered solutions, the basic solution in particular, the i
measured pH seemed to decrease with increasing total exposure in a 5-min test period.
However, such change was not observed in the acidic solution I
even after I hr of exposure.
1 1
f 5
e Y
j 3
e D-4..
W---- -
.. - auuaa TABLE 1 EFFECT ON GAMMA RADIATION ON pH MEASUREMENTS IN BUFFER SOLUTIONS
~
RADIATION FIELD IRRAD. TIME pH BUFFER SOLUTIONS
(
ELECTRODE (R/h)'
(MIN) 4.01 7.- 0 10.0 Standard 0 (pre exposure) 3.958 7.000 9.901
- 1. 3 x 102 -
1-2 3.955 6.950 9.896 1.3 x 104 1-2 3.963 6.922 9.895 l
1.3 x 105 1-2 3.96 6.945 9.889 1.3 x 106 1-2.
3.8 6.822 9.734 0 (post-exposure) 3.93 6.927 9.857 Flat Surface 0 (pre exposure)
S 4.271 7.301 10.149 4.000 7.000 9.994 1.2 x IOS 10 10.167 15 7.202 20 4.221 7.188 10.135 0 (post-exposure) 3.945 6.951 9.935 Semi-micro 0 (pre exposure)
(1)*4.017 7.000 10.000 (2) 3.996
- 7.000 9.997 1.2 x 108 5
(1) 5.132 6.930 10.932 (2) 4.998 10.744 10 (1) ---
6.966 10.917 4
(2) ---
7.813 15 (1) 5.074 6.985 10.904 (2) ---
20 (1) 5.013 6.990 10.885 (2) ---
10.724 0 (post exposure)
(1) 4.069 7.015 10.041 i }
(2) 3.970 7.019 10.041 Semi-micro O (pre-exposure) 3.996 7.000 9.997 1.0 x 105 5
4.217 7.075 10.050 10 4.165 7.145 10.073 0 (post exposure) 3.970 7.019 10.014 1
- (1) and (2) indicate Tests 1 and 2, respectively I
i l
D-5 L'
[/j)-
a f.
)
Table 2 4
Effect of Gamma Radiation on pH in Unbuffered Solutions j
Using the Standard Combination pH Electrode 1
t i
RADIATION FIELD pH IN SOLUTION (R/hr)
H,SO4 Na0H
{
I 0 (pre-exposure) 4.058 9.925 2
103 4.066 9.922
.10.4 4.070 9.923 l
105 4.070 9.915 108 (1 min) f 108 (3 min) 3.900 3.859 9.55
(
108 (5 min) f 9.42 108 (10 min) 3.846 l
108 (40 min) 3.948 0 (post-exposure) 4.107 9.475 J
^The reading was decreasing at s0.1 pH unit per minutes af ter 3 minutes, and the measurement was terminated at s5 min after exposed to the flux.
i
(
e e
W a.
e D L
y pg
~ ; < l-
~
EVALUATION OF GE AND SEC CHEMICAL PROCEDURES FOR POSTACCIDENT.
ANALYSIS 0.F REACTOR COOLANT SAMPLES November 1981 l
Prepared by Exxon Nuclear' Idaho Company, Inc.
l Idaho National Engineering Laboratory Idaho Falls, Id,aho 83401
^
for l. - -
The Nuclear Regulatory Cc. mission l
r l
e l
?
i
^ -
^
,e w <, -
y
(
z
- . y The advantage of the automatic mannitol titration procedure.is its relative simplicity, remote operational characteristics,
', uti1Ny[bnder routine and / accident, conditions, and wide mehsurement n / range.
~
The only apparent disadvantage of the procedure is yu
.the potential maintenance difficulty which might occur during replacement
~
p
~
~
af sensing ; elements under accident conditions.or rapid repair of the microprocessor. Howeve'r, as backup capabilities to analyze baron samples
"..lare"requ ired for inline sample methods, the Digichem analyzer should
- ~
~
2 meet a11 measurement requirements.
,p
~, -
s
- .~
s
.. g a'
The effects of high radiation fields have not been tested.- 'ENICO feels that the effects probably will not be significant;
. i.
- however, this should be confirmed.
n-
~
~
4.2.1.6 Ion Chromatocrachy (IC).. An ion chromatograph oper-o atis on the principle of selective retention and elution of ionic species on ' and.from-ion, exchange media.
It basically consists of a separator
~
column and eluent, a suppressor column, a conductimetric detector, and a readout _ device. f To perform an analysis for anions, such as borates or chlorides, the sample is first passed through the separator column - an I'
anion exchange medium which retains the anions and replaces them with
(
another anion from the exchange medium. -The retained anions are then selectiveiy removed froct the separator column with the eluent, normally m
l
~~
a dilute salt solution, and passed through the suppressor column. In the suppres'sor col'umn - a cation exchange medium - th'e anions are converted
- m..
.to their. acid forms which pass unretarded to the conductimetric detector.
.t The;. conductivity of these dilute ac.id solutions is a function of the j.p anion ; concentrations in the sample.
~
~
The time be' tween sample injection and the appearance
'i of conductivity peak for a'-particular anion depends on the sample size, the hhysical size of the columns, the types of exchange media,.and the types, concentrations,- and flow rate of the eluent.
As a result-differ-ent anions. in.a single sample can be separated and analyzed by proper 2
./
-selection of parameters.
. }
r
---m i.
.v r j -
1_.-w,,,,-
,,-,,,w,,,,-,
vv--w%--ve-,-c,
~+-.-y.-, - - -
-,y.m yw----.-e,
--ww--,
= - - - -
1,
.' g.:-
y)
-s
~
In the development of an icn chromatographic proce-dure for the analysis of boron and/or chloride; SEC/NUS studied various combinations of elvents, separator columns, suppressor columns,.and sam-ple injection loop sizes [
Initial testing resulted in a method which used a sodium tetraborate eluent and was applicable for chloride analysis of postaccident reactor coolant samples (cf Section 4.2.2.1).
- However, the analysis of boric acid solutions with the procedure showed inconsis-tent results.
[dditional develop::ent and testing by Dionex, the manufacturer of the ion chromotograph used, resulted in a procedure for
~
the simultaneous analysis of baron and chloride using a single sample.
In the. test program a modified Dionex Model 10 Ion Chromatograph was used. The modifications included two 4 x 250 m separa-tor columns, a 3 x 250 m suppressor column, a twenty em (0.043 ml) sam-pie -injection loop.. and a sodium carbonate / sodium hydroxide / mannitol elvent.
An additonal requirement identified was the need of a, cation pre-column to remove excess base and convert borates to boric acid prior to loading highly-basic samples into the, injection loop.
With a twenty.
five percent pump stroke, the necessary times for the boron and chloride peaks to appear following injection to the sampling loop are respective-
~
ly 5-6 and g-10 minutes.-
To consistently obtain satisf actory result's, peri-odic washing and/or regeneration of the suppressor and pre-columns is necessary.
The pre-column - requires regeneration after the analysis of every two to three samples.contairiing 'O.4 M sodium hydroxide.
The re-
^
~
quired frequency for washing, and regeneration of the supressor columns was not stated.
Howev,er.., based' on th'e frequency noted in the inital
~
chloride ana' lysis developm, erit work, estimated frequency for regeneration
~
is' every four hours of continuous.cperation.
The need for this 1s indi-cated by an erratic baselitie on the readout device.
The required fre-quency for washing the supp,ressor is once : daily or pri'er to each regeneration.,
e b
e e
e e
25 "
- 9::1-;
s If column washing and regeneration are not required, the analysis time is forty minutes.
If column washing and regeneration are required prior to analysis, the sample analysis time is approximately two hours.
Neither case ini:1udes system calibration time, which is fifteen minutes.
The IC procedere for simultaneous chloride and boren i
analysis has been laboratory tested using simulated posta'ccident reactor coolant samples, stable fission products, caustic, ecoling water impuri-ties, and normal reactor coolant chemical additives.
No sample matrix effects were observed within the specified measurement range.
The advantages of the procedure are its adaptability to remot'e operation, the large chloride measurement range, the simplicity of operation, small sample sizes, potentially short sample analysis time, and the lack.of chemical interferences.
The disadvantages of the procedure are the lack of a sufficient measurement range for boron, the need of a pre-column for basic samples, and the need for column washes and regeneration which might lead to long analysis times.
The effects of 'large irradiations associated with highly radioactive samples have not been evaluated.
However, based on a 1,iterature study of radi,ation effects on the ccmponents of the IC and on l
limited laboratory tests used to determine the effects of 0-200 ppm hy-drogen. peroxide in samples, no radiological effects are anticipated.
The literature showed that cation resins begin to degrade at approxi-8 mately 10 rads and that the electionic components are resistant to 5
exposure well
.above 10 rads.
B'oth levels are well above those anticipated to be encountered by the IC during analysis of samples.
4.2.1.7 Carminic Acid Secetrochotometry.
Two procedures were presented for boron analysis with carminic acid, one by SEC and one by GE.
The one presented by GE was detailed; it was developed by HACH Chemical Company and closely follows. an ASTM procedure.21 The 20
(
g..
%- f-y Y
smeseAi. cocas.
,; _)
=M P
..tANOh C As< TW3
~
CCFv ORATON 1990 CCCHAAN RCAO I
C
" :u"G" " C:a::
di3 *343* S E CO cvaus wm.msca civiS8CN ANALYSES FOR pH, CHLORIDES, DISSOLVED .. men, CONDUCTIVITY AND BORON UNDER " POST-TACCIDF # CONDITIONS r......-..
..;J -
BY W. NESTEL, COMMONWEALTH EDISON
{
W. LECHNICK, NUS CORPORATION R. C. RICE, NUS CORPORATION FEBRUARY 19S0 e
I e
t M*" N pe.
W*
pe w D
I e
n
_ fg h n-t
-n
.d ;.
3.? %
<3On e
- s gs-h
.E 2*a
=
w < {s <
'=
m g
{=R Ei=
8
. jos h
=
m.-
.u.
M>i fl1 2
_s t
q..
g e
{;;
I w*
l
- c3 l
-a to g
~
I A
i s
C l
h t__-________.,
c--- - - p..
_, _ -.-.L 5
g g
I I
I w
a I
g H
r-------8
-9 1
m I
a, I
e I
i I
e
.2 e
m I
f", L_
a
._L 3
a o:w Zw I
N e
O t
o I
odo
-do
_I m
I z>o m>o I
o g
~
w*c
>x e el ro w8 x
C,
~
a
~
J s
c o
C o
1 6
o y
g-c I
I m
.)
op I
A JL
- l "u U
3d 2
'x
'g o" l
oc l
e 2
o I
I ll 2
=
cc do R
a
-=
1 p
c g
- l 1 g g
o-u-
hl"g q
u I
I l
s
=
1 J
i I
c L_________.J e
3 g
ew c9 Q
w e
.- o.
<w c
O w yw c c z>
De o o
u W
os gE O w> 3 Q
ht J u<
m J <
22 M-o 8 b
Ai
=
.I
= o"
.[\\,w.)
wh 2
N wh t
w) a o
a D
~*
L' 23 w
w_S z
.o 0e ::
og
=
ee
-g s j
rD 2
1 2n 3
co 5
d3s o
0
$r "u"
{
V
{*3
$o15 St
-=
. "W ;~ta 1:
4
.=
C.
j W
G. 6 h
Sb U
o a
5.
.h..
6
.,,e S.
- 9. a=
C e
55
>><4
(.y) shielding requirements. The system as shown contains a Rexnorde oxygen probe which would not be used during post-accident sampling cor ditions. It is included as optional equipment at the request of CECO. Design capabil-Ities of the system are.x Table 3. All exposure times involved with
'o'peration of this system will be on the order ofseconds.
Design of the automated system provides for precalibration of the instru-mentation and final flush of the systems invo!ved to reduce radiation levels after the analyses are complete. 'Ihe system can be operated on a continuous.or intermittent basis. For continuous operation, calibration is required on a once-per-day basis. Calibration would be performed prior to use when operated intermittently. Provision has been made to direct sample line purge and flush flow back to a high purity waste system or to the radwaste system. Buffer solutions used for pH-conductivity calibration and all chloride analyzer solutions will be directed to the radwaste system.
Individual components within the system are described below:
r-u..
l A{.s n QklR G N $ SES )
The analyses will be performed by the use of ion chromatography. This is the only approach that can be used to analyze for chlorides that will
,not add significantly to radiation expo ~sure. Analyzing a sample by IC is fairly straight forward requiring about I ml of sample transferred via hard piping into the samp!!ng module to the IC. A fixed amoynt (400pl) of the sample is automatica!!y transferred to the sepo.ator column for analysis. Typical results achievdd under these conditions are shown in Figure 2.
Excess sample is discharged to drain. Most of the other availabfe methods for. chloride andysis involve mere exposure of the analyst to the radioactive solutions and for this reason were discarded as potential candidates fo'r this application.
'. e _..
- - ~..,,,, _..., _.,.
.F "Md 10 Gnh manufactur6d by.the Dionex cor oratioh'wilFb*e hids molicaticn._ ThisTiylic,m.....
~pe-v;~ nn~idM
..,,,::.... = =.: -
-n
_can be u gd. to, analyze.forMof e
4 e
.Qp-.
~
down ta o sensitivity cf thout 100 ppb without any pratreatm:nt of tha sample. However, this system cannot be used as is under accident conditions because of radiation exposure considerations.. This system must be modified to provide for hard line piping to the sampling port.
Also, shielding must be provided for that portion of the unit which will -
process the radioactive solution and secondary containment must be provided in the event of leakage.
B.
CONDUC.TIVITY MEASUREMENT This will be provided by an in-line probe contained in a redesigned probe holder.
The probe has a cell constant of 0.1.
Calibration of the conductivity probe will be performed with the buffer solution used for pH calibration.
C.
OXYGEN ANALYSES ~
These analyses will be performed by use of an in-line dissolved oxygen
{',
meter manufactured by the Yellow Springs Instrument Company. 'The,
probe holder has been redesigned to minimize fluid volume and reduce radiation exposure. Also, the probe holder is designed so that it can be flushed af ter it. is used i,n this application.
Design requirements includes provision for in-line calibration as is necessary to achieve accurate oxygen determinations. Calibration will be provided by use of a separate water source which will be saturated with respect to oxygen concentration. Normal calibration operations involving exposure to air cannot be used in this application because of radioact!ve contamination considerations.
D.
pH DETERMINATION An in.line probe will be used as is the case with the oxygen analysis system.
The probe holder has been redesigned to minimize fluid volume. The pH probe like the oxygen probe is calibrated in place since
,(~
- s j
9
- d4'- - l It cannot be removed from the system. The loop design provides for the addition of buffer solution to the pH probe and for flushing of this solution from the sistem after calibration is complete.
~
E.
OTHER DESIGN CONSIDERATIONS l.
Radiation Control
~
About six inches of lead will be used to provide for shielding of the tubing, probe assembly and valves which contain radioactive coel-ant. All components will be accessible from the back for ease of maintenance.
Secondary containment such as.a hood will be used to provide for control of radioactive gases and lodine leaking from valves and fittings.,
2.
Pressure and Temocrature Limitations The analytical system operating pressure limitation is 75 psi. This h based on an operating precure limitation of 100 psi for the pH probe. A margin.of safety was arbitrarily added for conservative
. reasons.
Optimum temperature range for operation of the analytical system is in the range of 75-90*F.- A small loss in accuracy for pH and conductivity ifeterminations will' result in. operating at a. higher temperature range. Maximum operating temperature is 125'F.
IV.
R ADIATION DEGR AD ATION No problem is anticipatt:d with radiation ~ damage to Instr'uments and/or 0-rings that may be used in the system based on results of radiation damage
~
effects reported in the literature. 'REMi='s dyMormehyWomaT}h u
- e
- (11)]-
Indicate-t.t-items such as.. semiconductor:r:icaWniP??C!!"ESUnA L.
~
yB
..mith;'etc;,--start-suf terinyredint!cy'ddinck ia tip_renhfrcy 10 FIN 5ihn3I9Ri-These3..yrJre imrJC;$[,cic5r,,s[.g Q++hi har-then-is-expect +fer-e
- "MqnchyJpmen@m IQ.,.
... c:c.d cce? "_p... -o:t-::cidi:it".r.caditto.n.d3 E~~~~~~?l'P p-?We ~. '
S-!$ ~?hV~5ffifh.MEr~.Is551'ng7...
((
'i.., v. m bER cr ?p?r n indicated by' datr. prc:entd/
- - ~
. =.~.=......
...a.:,;. a. w
..n 0;,.,
' > 1.. M 'c.;.:.ciiI;.;b.t..Q.-r,yg; _enr.,i!.icen_r,ubher._may_be uigd,,j
{tr."?f. EEmWoWWEr;TdspEtife U5veIs'.WilliS?Rf5iWe154~dhe' eld.b.lira rg01ijp3o 19;ur. radiation'.damcggd_
There is some possibility that the silver chloride in the pH probe will eventually be reduced to metallic silver by gamma radiation. Should this occur, it would be indicated by sluggish response during use of the pH instrumentation. Operating procedures require daily calibration when the pH instrumentation is in use. Corrective action would require. replacement of the probe assembly which is' easily achieved.
4
' Y.
BORON ANALYSES A colorimetric method utilizidg curcumin is currently recommended by NUS,
for boron determinations under " post-accident" conditions.
The concen-
,tration range for this method is 0.2 to 2 mg/l boron, thus the samples will typically be diluted by about a factor of 1000. One ml of the diluted sample is required for analyses. All analyticc! operations must be carried out in an operating fume hood to provide containment of activity.
The relative I
standard deviation for multiple analyses is about f.13% Total analyses time is two hours. The time frame does not meet NUREG-0573 requirements; however, it is considered justifiable from a safety analysis and personnel exposure viewpoint. The objective of the boren analyses is to determine that there is sufficient boron present to maintain the reactor in a suberitica!
condition during c'co!down. Time requirements for cooldown will be on the order of man'y hours for all reactor systems.
l [
.