ML20134E585
| ML20134E585 | |
| Person / Time | |
|---|---|
| Issue date: | 12/06/1994 |
| From: | Clint Jones NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| To: | Grobe J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| Shared Package | |
| ML20134B601 | List: |
| References | |
| FOIA-96-444 NUDOCS 9702060366 | |
| Download: ML20134E585 (9) | |
Text
_
_..__ . , . . . - . _ . _ . . ._ . . _ . ._ ._ _.._._.___ _.._ __. . . _ _ . . _ _ _ . _ . . _m . . _ _ _ _
i e
e t I
l ,
From: Cynthia G. Jones (CGJ) N $ 3 l To: JAG Date: Tuesday, h /-o/([994gg/
cember 6 4:21 pm i
i
Subject:
CCleveleand Itr f
f Carl Has concurred in tha version last rec'd. Let us know when you get OGC ;
concurrence. i
. Cyn 6 i
i I !
1 t l
l l -i t
l l I i I !
, l l
l l
t i l
l l
9702060366 970203 l PDR FOIA i ENGLISH96-444 PDR l.- - . - . . - . _ . -. -. . . - . _ _ - _ _ .
US,HRC REGION III ID:708-515-1259 DEC 22'94 .1:08
- e. -
< TRANSMIT CONFIRMATION REPORT NO. : 001 RECEIVER : 301 504 5369 TRANSMITTER : US NRC REGION III DATE : DEC 22'94 10:08 DURATION : 09'34 MODE : STD PAGES : 17 RESULT : OK I
i l
I
o pwh a". 21 UNITED STATES
,g NUCLEAR REGULATORY COMMl3310N
[}/
g '
o, WASHINGTON, D.C. 206t&cotM
'% *,,,s s /
- ,o
/
f: '
g} [
MEMORANDUM T0: John Madera, Chief -
Materials Licensing Branch '
Division of Nuclear Materials Safety, RIII FROM: Josephine Piccone, Acting Chief Operations Branch Division of Industrial and Medical Nuclear Safety, NMSS
SUBJECT:
REVIEW OF THE PROPOSED WATER ANALYSIS PROCEDURES RECENTLY SUBMITTED BY ADVANCED HEDICAL SYSTEMS (AMS)
We have reviewed the water analysis procedures recently developed by AMS and sent to region III on June 11, 1996. These procedures included RSP-018, Operation of the Gama Spectrometer, and RSP-019, Assessment of Radioactivity in Water Samples. Both procedures were dated June 11, 1996.
Although the procedures are of better auality than the previous versions, they still leave some important questions unresolved. Detailed comments are ,
provided in the attachment to this memorandum, but the main points that '
require further work are the following:
- 1. The bases for the proposed MDAs of 70 pC1/1 for water samples and !
15 pCi/l for filters are not well developed and justified in the technical basis section of procedure RSP-019. We suggest that AMS provide a more defensible basis for their choice of MDA. In our letter to them, we had suggested that the MDA should be based on tne capability of a detection system that is state of the art for the application but not necessarily extraordinarily specialized or sophisticated. We recomend that AMS expand their technical basis i section to include a detailed description of their measurement systems and their testing of these systems to evaluate their measuremer.t capabilities.
- 2. The procedures make only brief mention of the quality assurance program to be used for their measuring systems. We suggest that AMS develop a better quality assurance program, and provide a detailed description of that program in the technical basis section. The quality assurance program should extend to any outside analytical laboratories that AMS may use to confirm their l results.
Contact:
Sami Sherbini (301) 415-7902 bO
l .
i John Madera j
- 3. The procedures do not fully develop the bases for determining
- whether a sample does or does not show activity. They are still l using the MOA as a criterion for this purpose, which is incorrect.
- They need to develop decision levels, independently of the MOA, to l allow this determination to be made. The development of this l
decision level should be described in detail in the technical basis section.
- 4. AMS has not resolved the discrepancies that appeared several times between their water analyses and those performed by NE0RSD. These discrepancies appeared to have been due to differences in sampling i protocols. AMS must either resolve these discrepancies or conduct tests to demonstrate that their proposed sampling method will produce representative samples.
- 5. We suggest that AMS retain someone to review their proposed i methods to ensure their technical soundness. The procedures contain several errors in the statistical manipulation of the data, most notably the error of using the MDA for making field l decisions, and the incorrect equations provided in the procedures for calculating the MDA. '
Please call Joe DeCicco at (301) 415-7833 if you have any questions regarding this matter; Sami Sherbini will be out of the office during the month of July.
Attachment:
As stated
i l
COMMENTS ON AMS PROCEDURES
)
! i Technical Racia for Water Discharge Cntena l 4 1. Page 7, Second point:
This point contradicts the first point. The first point stresses the regulatory 4
requirement that no insoluble Co-60 may be discharged to the sewer. The second point tries to estimate the amount of insoluble Co-60 that may be discharged to the
, sewer without causing the ash to exceed 8 pCi/g. The analysis in the second point also neglects to consider the possibility that Co-60 discharged to the sewer as soluble cobalt may still end up in the ash because of a number of reasons, such as precipitation of the " soluble" cobalt during waste treatment, or settling of the !
" soluble" cobalt that is, in fact, not soluble but very finely dispersed insoluble material. We suggest re-asse;sment of the second point and possibly deleting because -
it is not relevant. !
- 2. Page 7, Third point:
The drinking water standard is not relevar.: in this case, and only confuses the issue. l Suggest that it be removed because it does not contribute to the technical basis being !
developed. I 1 , l
- 3. Page 8, Second Point from the bottom:
l l
This statement concerning Information Notice 94-07, that "the standard does not l provide guidance on how much gross beta activity indicates an insoluble material," is incorrect. The standard states, on Page 4, that " activity in the suspended solids portion of the effluent greater than that found in similarly processed background water samples would indicate the presence of insoluble radioactive material." In other words, the Notice states that any activity that is statistically distinguishable from background indicates the presence of insoluble material. Background in this case is ;
the filter residue from water filtered in the same manner as the sample. The water used to produce the background filter is water obtained locally but that is not contaminated by the licensee's operations.
i Attachment
1 I
l
- 4. Page 8, Imt point:
1 This point is not relevant to the discussion and should be deleted.
l
- 5. Page 9, First point: l l
l This point is at variance with the data NRC has been getting from both commercial j
, laboratories that were used by AMS and NEORSD to analyze the water samples from the discharge tanks. The results from' these laboratories were routinely reported as ,
having been obtained using equipment capable of I-2 pCi/l using counting times as I low as one to two hours. How can these values be reconciled with the values indicated in this point? I In this connection, AMS has not described the system it intends to use for sample analyses. It is necessary to know the system to be used, type and size of detector, counting times, background levels in the counting laboratory, location of laboratory within the AMS facility, shielding for the detector, methods of spectral analysis to be used, type of blank samples, and source of water to serve as the background, and the quality assurance program for the system. Without such detail, it is not possible to determir;e if AMS is conducting orerations in accordance with current state of the art ,
methods. We suggest including inis in the technical basis section.
l We believe the technical basis confuses the issues involved. The issues are: 1) the limit on soluble Co-60 was set by court-mediated agreement to 100 pCi/1; 2) The limit on insoluble Co-60 is zero, based on regulatory requirements; 3) The method of implementing the zero release limit is described in IN 94-07. It involves filtering the water through a 0.45 m I filter and comparing the counts from that diter wid the counts from a filter used to filter an equal amount of " background" water if d,c counts are significantly different, the water is assumed to contain insoluble Co-60. The only open question is the sensitivity to be used in counting the filters. The NRC position is that this sensitivity should be that attainable by a system that is typical of state of the art equipment and methods, but not anything that is highly specialized or sophisticated. The licensee must address how all these requirements are to be met.
Procedure PJP-018: Operation of the Gamma Spectrometer Page 2,5.1.3:
It is not clear what is meant by "so that the two primary peaks fall in channels 155 and 176." Does that mean that the maxima of the peaks occur at these channel numbers? How critical is it to have the peaks at these locations?
l l
2 l
Procedure RSP-019: Asses; ment of Radioactivity in Water
- 1. Page 3, 5.1.2: It is not clear whether the analysis is to be performed in-house or by a commercial laboratory. What is to decide which it is? In either case, it should also be made clear that the laboratory performing the analyses must participate in a quality assurance / quality control program that is approved by AMS and periodically audited by a recognized group or organization outside the organization that operates the laboratory. Pre-qualification of the laboratory is only part of this ongoing program. A later section (5.4.3) states that samples may be forwarded to a commercial analytical laboratory for confirmatory analysis. Some indication should be provided as to when such an action may be necessary.
- 2. Page 3, 5.2.1: Sampling these tanks has a history that indicates that different results are obtained from samples taken from various locations within the tank.
The differences in the results have not been explained, giving the impression that the mixing used by AMS to date has not been effective, and that there r:mained stratification of the Co-60 even after prolonged mixing. It is therefore necessary for AMS to do one of two things:
either explain the differences in the results obtained in the past, or conduct tests to demonstrate that their proposed method does indeed produce representative samples. Without such data, the results will remain suspect because of unexplained past anomalies.
- 3. Page 3, 5.2.3: In the note, it is stated that samples may be collected from any location in the tank. In the interest of consistency, we suggest specifying a sampling method at this point.
- 4. Page 4, 5.4.2: It is not clear where the MDA of 70 pCi/l comes from. In our letter to AMS, we stated that AMS should establish a counting method that is considered typical of current and ordinary state of the art for such an ;
application. We have not seen any data to show that the licensee has done that. We therefore suggest that the licensee establish their well-shielded counting system, in a low background area, select a reasonably long counting time, and then establish the sensitivities achievable by such a system. We are confident that the MDA will be far lower than the proposed 70 pCi/1. Also, in the equation for MDA, if the time is to be in seconds, the procedure should clearly indicate that the count rate B, must also be in counts per second.
3
I The Note at the end of this section is not correct. The MDA will not ensure a 0.05 probability of a Type 11 error when comparing the sample result to the background result. Suggest removing or rewriting l to more accurately reflect the technical meanmg of the MDA.
The equation for MDA given in this section, and elsewhere in the l procedure, is incorrect. The equation in the form given contains the
! implicit assumption that the sample (or gross) counting time and the background counting time are equal. This is not the case, however, 1 because the background is counted for 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />, whereas the samples i I
are counted for times less than 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> (See procedure RSP-018 for sample counting times). The equation in the form given will underestimate the MDA.
- 5. Page 4, 5.5.1:
Although we understand the intent in this section, the logic is obscure and may be missed by, those who are not familiar with the background for this procedure. Therefore, it is necessary at this point to add a section that tells what to do if the sample contains more than 100 pCi/l of Co-60.
In addition, this step requires that all samples less than 100 pCi/l be drawn through a filter, even those that show no activity that is statistically different from background. Is this the intent?
Finally, how is the sample to be drawn throcgh the filter? Is it to be sent to the chemistry department, where people skilled in this type of j I
work will draw the sample? If not, how will the licensee ensure that the sample will be drawn correctly through the filter?
1 I
- 6. Page 5, 5.5.3:
I As in the above, where does the 15 pCi/l come from? The origin of this number is mentioned in the technical basis, but is not clear.
1 This step, or the remainder of the procedure, does not describe what to do with the results of the analyses on the filter. What criteria are to be ;
used to decide if the filter indicates insoluble activity?
i l
l i
4
- 7. Page 6, 5.7.4:
)
The condition given in 5.7.4.2 is not acceptable. The use of MDA as !
criterion for deciding if activity significantly different from background j was detected is incorrect. The licensee must establish a decision level, ;
independently of the MDA, that will be used to make this i determination. ,
1 l
l i
5
_