ML20138E754

Discusses Summary of Proposed Analysis of Ref Facilities for Inclusion in Final Geis.Requests Review of cost-benefit Analyses by 950609

ML20138E754
Person / Time
Issue date:	05/31/1995
From:	Glenn J NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:	Astwood H NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS)
Shared Package
ML20137H228	List: ML20137H223 ML20137H274 ML20137H284 ML20138E754
References
FRN-59FR4300, FRN-59FR43200, RULE-PR-20, RULE-PR-30, RULE-PR-40, RULE-PR-50, RULE-PR-51, RULE-PR-70, RULE-PR-71, RULE-PR-71-97031, RULE-PR-72, RULE-PR-MISC SECY-97-046A-C, SECY-97-46A-C, NUDOCS 9507170197
Download: ML20138E754 (94)

Text

.-. - _ _ _ . -- - .- . . - - . _ - . -- . - _

I I

~

l l

l May 31, 1995 -

i NOTE T0: Those on Attached List Original Signou by

~ ~ " '

FROM: John E. Glenn, Chief -"

Radiation Protection and Health Effects Branch Division of Regulatory Applications, RES J

SUMMARY

OF PROPOSED ANALYSIS OF REFERENCE FACILITIES FOR INCLUSION IN FINAL GEIS I As you are aware, a number of comments were received on the draft Generic Environmental Impact Statement (GEIS) on the adequa:y of the reference facilities in modeling contamination, particularly soil contamination, :t nuclear facilities.

! As a result of the comments an effort has been made to review available "real world" data on the reference facilities. We have been assisted in this effort by staff from NMSS and NRR and by our contractors, PNL and SC&A. A summary of

- the results of the review and proposed method for analysis of the reference  :

facilities is contained in Attt:hment A; specific information on the each of  !

the reference facilities is contained in the draft Attachments 1-6.

In addition to the reference facilities, Attachment A (Section 2) discusses comments received on the general cost-benefit analysis approach and the status

, of our response to those comments. , i Before we direct our contractor to perform additional cost-benefit analyses of the reference facilities, we would like Core Group review of the Attachments.

We would appreciate your review of these dccuments by June 9. As necessary, a

Core Group meeting can be held the week of June 12 to discuss your input.

Thank you for your assistance.

. 4 Attachments: As stated Distribution: (G:\CARDILE\REF-FAC.526 CF Y N

% 88$p @ $ BMorris FCostanzi POR YN To recalwe e copy of this document. indcete in the boa: "C" = Copy without ettechment/ enclosure *E* = Copy with attachment / enclosure

• N" = No copy 0FFICE DRA:RPHEB a/f,- DRA:RPHES, l l \ l NAME FCardile F "h.

DATE 5/ 70/95 b 6D /#5""

V 0FFICIAL RECORD COPY (RES File Code) RES$7dI23 9 fo 7/ 70 / 9 7- k.9

May 31, 1995 H. Astwood et al 2 Addressees - Memorandum Dated / /

Hail Stoo

~-

T7-F27

~

Heather M. Astwood, NMSS Jack D. Parrott, NMSS T8-F37 Anthony M. Huffert, RES T9-C24 Christine Daily, RES T9-C24 Francis X. Cameron, OGC 015-818 Peter B. Erickson, NRR 011-820 Stephen P. Klementowicz, NRR 010-D4 William R. Lahs, NMSS T8-F37 Dennis M. Sollenberger, OSP 03-023 Michael F. Weber, NMSS T7-F27 l George E. Powers, RES T9-C24  !

Carl Feldman, RES T9-C24 I Brian Richter, RES T9-C24 -l l

-J l

I i

i r

t j

2

t

1. MCO yo. . & UNITED STATES'

'j NUCLEAR REGUL,. TORY COMMISSION

{ f WASHINGTON D.C. 20555-0001

+O ,

..... I May 31, 1995 i NOTE T0: Those on Attached List FROM: John E. Glenn, Chief f Radiation Protection d Health Effects Branch Division of Regulat Applications, RES

SUMMARY

OF PROPOSED ANALYSIS OF REFERENCE FACILITIES FOR INCLUSION IN FINAL GEIS As you are aware, a number of comments were received on the draft Generic Environmental Impact Statement (GEIS) on the adequacy of the reference facilities in modeling contamination, particularly soil contamination, at l nuclear facilities.

As a result of the comments an effort has been made to review available "real world" data on the reference facilities. We have been assisted in this effort by staff from NHSS and NRR and by our contractors, PNL and SC&A. A summary of I the results of the review and proposed method for analysis of the reference i facilities is contained in Attachment A; specific information on the each of i the reference facilities is contained in the draft Attachments 1-6. '

In addition to the reference facilities, Attachment A (Section 2) discusses i comments received on the general cost-benefit analysis approach and the status i of our response to those comments. ,

Before we direct our contractor to perform additional cost-benefit analyses of the reference facilities, we would like Core Group review of the Attachments.

We would appreciate your review of these documents by June 9. As necessary, a Core Group meeting can be held the week of June 12 to ' discuss your input.

Thank you for your assistance.

Attachments: As stated

~

i A,TTACHMENT A GEIS Technical Issues Raised in Comments - 5/23/95 A number of technical issues were raised by commenters on the DGEIS. These include:

1. The reference facilities used ia the analysis

2. Alternative cost-benefit approaches

3. The approach in calculating doses including:

a) The use of NUREG/CR-5512 scenarios and parameters b) The comparison of risks from nonradiological and radiological scenarios c) The time period for the analysis

4. Impacts on waste disposal capacity

5. Unit costs, including a) Costs of waste disposal b) Costs of surveys c) Costs of compliance THE FOLLOWING ISSUES ARE DISCUSSED IN THIS PAPER:

1

1. 1 The reference facilities used in the analysis - page 2

1. 2 Alternative cost-benefit approaches - page l l 1

1 1

, 1 1

I l

l l

l 1

i

1. REFERENCE FACILITIES

a. Public Comments on Soil Contamination The principal public comments received were on reactors and large fuel cycle uranium facilities and indicated that the model for soil contamination in the GEIS is not adequate for the following reasons: ,,

(a) soil volumes in the GEIS are too low (b) the models for distribution in soil in the GEIS are too simple, areal variability can be much more pronounced, contamination can be deeper than indicated and volumes requiring removal can increase rapidly with decreasing dose criteria. The physical reasons given as to why the l

model is not accurate include:

(1) Underground line breaks can occur between buildings, between >

outdoor tanks and buildings, or between tanks and discharge points; these leaks can be several feet underground i

(2) There can be large volumes resulting from seepage of slightly contaminated routine effluents from drain lines, from tank l overflows, and from natural buildup of permitted releases over  !

years of operation (3) Use of a 15 cm depth is only appropriate for farmland where the source of contamination is-dust or spreading of contaminated ,

topsoil. Under realistic conditions of mixing and burial, contamination will likely extend beyond 15 cm (4) The GEIS model results in primarily surface contamination. In real situations, contamination is more a function of site waste management practices, e.g., burials, deposition, soil mixing l through previous site activities, leaks in underground systems, etc.

(5) The effect of mixing that occurs during remediation The proposed response to the comments on the reference facilities is provided on the following pages.

2

b. Proposed Response to Public comments on Soil Contamination in response to the comments the staff has reviewed additional real world data.

A discussion of that review for the reference facilities is in attachments 1-

6. Based on the attachments, the following approach is planned for the Final i GEIS: l (1) Retain reference facilities as analyzed in DGEIS to provide base case analysis. Modify to include certain changes, for example, revised depth i profile models and revised waste d.isposal charges. j (2)~ Perform sensitivity analysis - a sensitivity analysis will be added to the GEIS which performs the following rdditional analyses for each facility:

A summary of the volumes for the DGEIS base case and the estimated volumes for the sensitivity analysis are found in Table A-1 of this attachment.

j a) P'ower reactors -

depth profile - use information from Humboldt Bay (HB) condensate storage tank leak which suggests that volume profile is approximately 4.5:1 in going from 100 to 15 mrem rather than 2.5:1 in DGEIS (see Att 1, Sect. A.2d and A.3) volume of soil - use three separate cases to model 3 potential types of contamination situations (1) 3000 ft2 area and HB depth profile for general site contamination (Att 1. Sect. A.2a(1) and A.3) l (2) 25000 ft3 of low activity (<.1-2 pCi/g) b'uried soil based on 20.302 burial information (Att.1, Sect. A.2a(2) and A.3)

(3) 3000-5000 ft3 volume and HB dhta profile data based on certain contaminating events, such as pipe leaks causing higher soil l activity (10-30 pCi/g)(Att. 1, Sect. A.2c and A.3) b) Uranium fab facilities and UF6 facilities i

l These two types of facilities are similar in the DGEIS except that the UF6 case has a larger contaminated soil area. For the sensitivity analysis, they will be combined as follows:

depth profile - use DOE facility data which suggests that volume profile is approximately 3:1 in going from 100 60 15 mrem rather than 1.2:1 in DGEIS (Att. 2, Sect. A.2b) volume of soil - use 660,000 ft3 at 30 pCi/gm based on total volume data from several facilities (Att. 2, Sect. A.2a and A.3) 3

4

.' l l

i l

(a commenter inquireo about evaluation of DU facilities with regard to l reference facilities and the overall decommissioning criteria; it is  !

expected that the sensitivity analysis and base case will boond this facility as well as other source material facilities) c) Research/ test reactors

~~

Specific comments were not received on these facilities. The data in ~'

Attachment 3 suggested that t.he reference cases in the DGEIS are not unreasonable (Att. 3, Sect A.2). Therefore the following will be the i

_ , only change

I depth profile - use Humboldt Bay (HB) condensate storage tank leak profile as developed for the power reactor 4

a 4

d) Broad R&D facilities Specific comments were not received on these facilities. Limited

' information is available for these facilities. They are generally not required to submit decommissioning plans or characterizations. The data in the attached (Att. 3 and 4, Sect. A.2 and A.3) suggests that the reference cases are not unreasonable. Therefore the sensitivity 4 I

analysis would be limited to modifying the depth profile as noted above.

i e) Sealed source /radiopharmaceutical manufacturer 4

The only specific comment received on these types of facilities was that

the selection of reference facilities does not include typical examples of radiopharmaceutical and radiochemical manufacturing. For reasons ,

indicated in attachment #4, this comment may have misinterpreted the

analysis. Limited information is available for these facilities. They 1 l

are generally not required to submit decommissioning plans or '

characterizations. Available information indicates that the reference

. case in the DGEIS is not unreasonable (Att. 4, Sect. A.2 and A.3). l Therefore the sensitivity analysis would be limited to modifying the  !

depth profile as noted above.

f) Uranium mills l depth profile - use volume profile suggested above under 2b.

volume of soil - use 6E6 ft3 based on total volume data from several facilities (Att. 5, Sect. A.2a and A.3) '

1 J

i 4

9 f) Rare earth facilities - ,

volume of soi1 - use 50,000 ft3 based on limited total volume data from several facilities; use modified depth profile based on .5-1 ft depth (Att. 6, Sect. A.2 and A.3)

c. Proposed responses on Building contamination A few comments were received on building contamination, mostly in Uie aiva tf-reactors. A commenter indicated that volumes of activated and contaminated equipment and structures" increase exponentially with decreasing dose rate below 500 mrem /yr. A comment was also received that for some reactor designs that significant volumes of concrete can become slightly contaminated (100,000 ft3).

Status -1) The reference facilities in the DGEIS appear adequate based on:

a) the limited nature of comments on building contamination, b) the staff review of the original data (most of it "real world" as gathered at 6 operating reactors by PNL) upon which the building contamination model is based, c) review of information in Section B of Attachments 1-6.

2) The basis for the comment that there is significant activated concrete volume is being reviewed (Att. 1, Sect. B)

3) Based on the review of data (Att. 4, Sect B), the sensitivity analysis would include a larger contaminated building area.

e J.D 5

l l

2. ALTERNATIVE COST-BENEFIT ANALYSIS APPROACH

a. The DGEIS uses a " knee-in-curve" approach. Commenters suggested that this is inappropriate.

Status - SC&A is providing tables of costs and impacts .(similar to Table 6.1 of the Reg Analysis), and curves, to allow NRC better understanding of costs and impacts on which to base de~cisions!-

b. The DGEIS did not have a criteria for making decisions regarding cost

~

vs. benefit other than the knee-in-curve. Some commenters suggested that a $250/personrem should be used.

Status - SC&A can provide tables which express the cost of risk reduction either as $/personrem or $/ mortality._ The value used as criterion will likely be $1000/personrem or $2M/ mortality.

c. The DGEIS analyzed costs and impacts for a cumbined site analysis, i.a structures and soil combined. Commenters suggested that this is inappropriate and that these should be analyzed separately. l Status - SC&A is providing both a separate analysis of structures and soils, as well as a combined analysis as done in the DGEIS, to allow NRC full understanding of costs and impacts on which to base decisions.

d. Other (1) DGEIS not based on analysis required by Executive Order 12886 to ensure regulation designed in most cost-effective manner

' (2) Don't base decisions on cost ,

(3) Use of conservative analysis not appropriate in cost-benefit analysis of EIS (4) Optimization should be done on a site specific basis below an upper limit of maximum acceptable dose rather than generic point of optimization across all industries Status - Responses will be developed 6

L.

j j i i

! Table A-1 l Comparison of DGEIS Soil Volumes with 5 Estimated Sensitivity Analysis Volumes' t Estimated  :

I CSEIS Sensitivity i Volume' . Vol uma*. .

~

Facility (ft3) (ft3) i l

} Power 460

-Reactor l l.

d Case 1 1350 l 1 Case 2 20,000'  !

1

! Case 3 3000-5000 Research 114 250 )

- Reactor Uranium fab 38,000 815,000 l

facility l l Sealed source 1170 2300 I l manuf i l 4 Rare earth 5000 50,000  :

i facility )

1 i Uranium mill 663,000 6.1E6 l

i 1- ,

i Notes: .

1) Volumes for DGEIS taken from NUREG-1493; Values' for sensitivity analysis are estimated fram Attachments 1 final sensitivity analysis values will be based on Core Group review and PNL fiaal analysis

2) Values are estimated to achieve 15 mrem /yr

3) Case 2 is for contaminated soil generally below 15 mrem /yr i

j l

l 7

i

h, O .

ATTACHM"1T 1 VERIFICATION /H00!FICATION OF REFERENCE POWER REACTOR Public comments have questioned the model used for the reference power reactor. These comments are summarized on page 2.

In response the staff has analyzed real world data to verify / modify the reference power reactor for both soil and building contamination. This analysis is contained on pages 3-20 as follows:

Soil contamination Page 2 - summary of public comments on soil centamination Page 3 - DGEIS reference power reactor soil contamination model Page 4 - analysis of real world data for soil Building contamination Page 17 - summary of public comments on building contamination Page 18 DGEIS reference power reactor building contamination model Page 19 - analysis of real world data for building contamination Results and preliminary conclusions Preliminary results of analysis of real world data for soil contamination are on page 12.

Preliminary results of analysis of real world data for building contamination are on page 21

1 l

A. Summary of Public comments on Soil contamination in Reference Power Reactors Soil Contamination - Commenters indicated that the model for soil contamination in the GEIS is not accurate for the following reasons:

~~

(a) soil volumes in the GEIS are too low (181,182,189, 216), _ l

)

(b) the models for distribution in soil in the GEIS are too simple (189, I 190,214), areal variability can be much more pronounced (214),

contamination can be deeper than indicated (214) and volumes requiring l removal can increase exponentially with decreasing dose criteria at levels below about 30 mrem /yr (87,190, 214).

The physical reasons given as to why the model is not accurate include: l (a) The GEIS ignores the pathway of deposition from airborne effluent release made over the life of the plant (155, 156, 157, 190)

(b) Underground line breaks can occur between buildings, between outdoor ,

tanks and buildings, or between tanks and discharge points; these leaks l l

can be several feet underground (181)

(c) There can be large volumes resulting from seepage of slightly contaminated routine effluents from drain lines, from tank overflows, and from natural buildup of permitted releases over years of operation (190)

(d) The effect of mixing that occurs during remediation (190)

Although one commenter did note that the model in the Gels may be appropriate for one means of contamination (e.g., where contamination results from discrete spills around perimeters of tank:) (190), they indicated that other sources given above would predominate.

2

VERIFICAT, ION / MODIFICATION OF REFERENCE F0WER REACTOR in response to the public comments, the staff reviewed the reference power reactor for the purpose of verifying and/or modifying the reference. cases.

The discussion below describes this review as follows: (1) The basis for the dGEIS reference facility is summarized; (2) The real world data considered is presented; (3) the basis for the verified and/or modified reference facility is given.

1. BASIS FOR DGEiS REFERENCE POWER REACTOR The reference power reactor is described in App. C of the DGEIS. Data on contamination levels is based on real world data from a 1986 PNL study of 6 reactors which indicated that contaminated soil volumes were on the order of 10s of m3 (100s of ft3). Little depth profile information was available so that a simple diffusion model for spread of contamination into the soil was used. Briefly the soil contamination model is as follows:

Site contamination:

Areal extent - 2000 ft2 Contamination level - Co60 Csl37 pCfg DCi/q 2-60 1-20 Depth profile - contamination with depth is based on diffusion into the soil such that soil must be removed as follows:

total Soil removal for: depth (ft) vol (ft3)-

100 mrem .1 196 30 mrem .2 406 15 mrem .23 460 3 mrem .38 750 Under building soil contamination Contamination is assumed to penetrate building cracks to a depth in soil of about 5 feet.

Depth profile - none estimated, i.e., it is assumed that all of this soil is removed, approx 24000 ft3 for each dose criteria 3

f i

2. ANALYSIS OF REAL WORLD DATA I 1

A principal means of this review was reviewing additional real world data. ,

The following general items were considered:

a) Information on General site contamination b) Information on contamination from airborne effluents ~~

c) Information on specific events causing contamination i j

These are each discussed below. l 1

a) Information on General site contamination j i

Information on general site contamination was obtained from the following two i sources: (1) facility decommissioning plans and (2) 20.302 onsite burial records. l (1) Decommissioning plans Decommissioning plans from the following facilities were reviewed to assess ,

whether real world data provided input to verify or modify the reference power reactor:

Rancho Seco l Yankee Rowe s Trojan Big Rock Point )

Shoreham-In general the data ennsisted of scoping studies and ground samples from gridded locations around the site. The following is a summary of the information available in those plans:

Big Rock Point' Most of the' site had fairly low contamination which would likely result in exposures of less than 15 mrem. The principal nuclides were C060, Cs137, Mn54, and Cs134 (pg 6-9). The areas on the site for which data were given are in Table 3.1.16:

Co60 Csl37 pC/q pCfg range .05 .21 .05-1.4 avg .16 .48 Table 3.1.16 also indicated the following specific areas were identified as having larger contamination:

Co60 Csl37 .

1 area no. pCi/g pC;/g .

3 30 l 5.5S/4.5E 5.5S/4.5E 1-3 11-24 l 4

)

l 5.5S/4.5E 8.6-13 22-100 '

~

6.5S/4.5E 2.9-3.6 13-32 8S/5E 10-48 12-70 avg 10 34 l The<n araas ware of a total area of about 300 m2 or about.3200 ft2; the depths sampled were 2-6". . -.

~

Yankee Rowe i Most of the site both inside and outside the radiation. controlled area had j fairly low' contamination which would likely result in exposures of less than 1 15 mrem. The areas on the site for which data from samples and in-situ' gamma spectrometry were given are in Tables 3.1.12 and 3.1.15 and the EA, Table 4:

Co60 Cs137 l' gC]3 gC]3  ;

samples range .3-1.6 .06-1.7  ;

avg .7 .5 i i

insitu gamma range <.1-3.9 i avg .12 j Table 3.1.12 also indicated the following specific areas were identified as l having larger contamination: l 1

Co60 Csl37 area pC1/q -Ci/q 46 19 8.4 60 13.5 11.3 61 3.6 9 avg 12 9

~

These. areas were of a total area of about 225 m2 or about 2500 ft2.

Other specific areas on the Yankee site were identified (pg 3.1-12, table 3.1.15) as follows:

area ft3 Co60 Csl37 gC]3 R13 C l' construction excavations 5400 .16 spent fuel b1dg 1700 2.5

' discharge sediment <.2 2-4 ,

(Fig 3.1-31, 32) l l

5

. . . __ . . _ . - _ _._ _... _ ._ _ . . . . _ _ _ .m. _ ._ _ _ _ . _ . . _ . _ _ _

r l

- t Rancho Seco , l The following areas on the site were identified as having either soil contamination which was removed or remaining contamination (Table 3-18, page 3-10,11):

location ft3 Co60 Csl37 l

~~

tank farm area removed 110 79 230 remaining .6 2 ,

retention basin removed 230 3.9(max) 4.9(max) remaining .8 1.4 ,

storm drains .2(max) . 9(max) ]

cooling tower sludge <.3 Clay Creek sediment .43(avg) 3.7 (avg) J radwaste sludge storm dr removed lft in 1.5(max) 11(max) drain ,

.15 48 remaining .

Soil Volume - App B, Table 4.3 estimates volume of soil requiring removal

- 350 ft3 (2) 20.302 Burial information.

A review'of 26 occurrences at 20 sites was conducted. Information examined included:

the volume of material buried the type of material buried .

the concentration of C060 and Csl37 in the materials the year in which the concentrations were measured i

l The concentrations were used to calculate the residual dose rate from the material in the year 1995 by decaying the concentration from the year of measurement to 1995 and applying the dose conversion factors developed for the residential scenario in NUREG/CR-5512.

The results are given in the Table 1. The data sets are ranked by estimated dose rate and divided into groups corresponding to the residual dose rate i

levels selected for analysis in the GEIS. The results are as follows:

Co60 Csl37 l no. of reactors pC/g no. of reac pC/g

- 17 0-1 21 0-4 l 4 3 1-2 3 4-15 i f

5 2-25 2 >15 1 >25 6

. . _ _ - . - . - .-.._- . -- - -..~._.-~.-.- -. .. - -- - -.

l a

i  !

i i

I t

1 d

i

! i i

j b) CONTAMINATION FROM EFFLUENTS An analysis of soil contamination resulting from airborne effluent deposition ,

i was done in NUREG/CR-0672 by PNL'. Annual curie release rates from the 1970s ]

(a time of high BWR fuel failures) was used. This analysis has been updated to use effluent data from NUREG/CR-2907 which contains effluents from PWP.S ?rd- ]

BWRS from 1990. i

a. .

j Table 2 shows a summary of the airborne release rates. ' Based on that  ;

j information, and ratioing the analysis in NUREG/CR-0672, the estimated values j of contamination are as follows:

$ Co60 Csl37 pC/g pC/g

<.01 <.01 l

i

! c) CONTAMINATION FROM EVENTS _

i Several specific events have occurred at reactors which have resulted in soil contamination. Much of these resulted in requests for burial. Specific

! events are noted below and are taken from the decommissioning plans and the

. burial records indicated above. These include the following:

j Bio Rock Point i

Big Rock Point decommissioning plan (pg 3-1,6-3) indicated that, in 1984,  ;

there was a 20000 gallon leak from an underground condensate line. The l j contamination was to a depth of several feet. '

J

. The following was the result: ,

i ft3 pCi/q l Contami:aLion excavated 60 2000

Contamination. remaining 5300 .14 i

Turkey Point letter requesting onsite burial indiceted that in 1982 there was a radioactive l

liquid spill resulting in soil contamination.

The following was the result:

, Co60 Cs137 ft3 pCLq p_Clg Contamination remaining 2500 180 5 i

Fermi

]

Letters requesting onsite burial indicated that in 1985 and 1986, there were 2 "

leaks from condensate storage tank of 50000 and 275000 gal leak into the soil.

.The contamination was into the top 1 5 feet of soil. This situation has been 1

7

4 ccrrected by addition of a liner. The following is the result:

ft3 Co60 -

p_C/9 contamination 30000 .12 Vermont Yankee ..

Letter requesfi69 onsite burial indicated that in 1991 there was a leak from

~

the sink drain through cracks in the chem lab floor. The following is the result:

Co60 ft3 pC/g pC/g avg range contamination 58500 .4 .09-1.1 Yankee Rowe Decommissioning plan indicated that there was a leak in the ion exchange pit in 1964 causing 74000 gal to be released to the area beneath the pit Dresden 1 The EA for Dresden 1 indicates that in 1975 there was a leak at the solidification building resulting in soil contamination. Soil was removed at that time. The remaining contamination is as follows:

l ft3 Cc60 Csl37

' contamination 8100 30 100 Rancho Seco f The decommissioning plan indicates (pg 3-10, and Table 3-18) that there was

' contamination in the vicinity of the borated water storage tank valves that was removed. The contamination was as follows:

ft3 Co60 Csl37 aCLg RCLa contamination initial removed 110 79 230 remaining .6 2.1 l

Palisades Letter requesting onsite burial indicated that in 1987 there was cooling tank overflow causing flooding of contaminated area in south radwaste building (overflow problem has been corrected). The following contamination occurred:

ft3 Co60 Csl37 contamination mci gC/g pC/g initial removed 1560 34 1.3 >500 remaining 6000 5.1 .05 30 8

d) DATA ON DISTRIBUTION OF CONTAMINAT.)N ,

Data on volumes and distribution of contamination is impor+ ant because it is useful in determining the amount of material (and cost and impact) which results from achieving alternate dose criteria. Generally this data is sparse which is why simple diffusion models were used in the DGEIS. Limited data on distribution is available as follows:

~ ~

Rancho Seco Table 3.1.12 indicates the following:

for areas of low contamination depth (in) Csl37 (pC/g) 2 .48 6 40 for areas of moderate contamination area. Co60(ratio) Cs137 2" 6" 2" 6" l .6 .05 .57 .14 2 24 11 3.1 1.1 3 13 32 2.9 3.6 4 70 12 4.8 1 5 2.4 5 3.1 .96 6 22 100 8.6 13 These data suggest that contamination goes deeper into the ground than diffusion model would suggest, although by themselves do not provide a very complete basis for modeling.

NUREG/CR-4289 - some limited data is avhilable from NUREG/CR-4289 for the following plants:

Humboldt Bay - Table C.2.9 presents data on contamination with depth. It is indicated that this is for areas that are close to background or from the fallout plume so the contamination levels are low.

Table C.2.ll presents depth data for a spill area around the condensate storage tank as follows:

Co60 Cs137 cm pC/g pC/g 0-2 306 88.9 2-6 67 21.4 6-10 15 4.9 10-14 7.9 2.7 14-18 8.3 2.4 22-26 3 1.2 26-30 .11 .03 26-30 2.3 .53 9

' I t

e 30-34 1.9 .3 34-38 2.1 .24 38-42 4.7 .48 j

~

42-46 8.5 .74 46-50 16 . 9 7.

l Table C.2.13 presents depth data for cooling canal sediments. This data  ;

4 appears to indicate that contamination does not decrease with depth in the sediments.

Dresden - Table C.3.6 presents a limited amount of data for an area near the radwaste building. This data appears to indicate that the contamination does i not decrease at a 12 inch depth. )

Turkey Point - Table C.6.10 presents contamination data for controlled and 1 outside controlled areas as follows: l

)

Co60 (pC/g) Cs137 (pC/g) no. 0-4 cm 4-8cm 0-4 cm 4-8 cm controlled area TPl 35.7 4.66 1.68 .3 TP2 26 2.48 3.63 .34 TP3 4 1.24 6.48 2.87 posted area TP8 14 45 4.12 11.8 outside controlled area TP4 .09 .016 .132 .086 TP5 .16 .019 .19 .09 TP6 .17 .02 .19 .1 TP7 .02 .02 .09 .09 For contamination events Palisades ft3 C360 Csl37 1560 1.3 >500 6000 .05 30 Big Rock 60 2000 l 5300 .14 l All of these data are fairly limited but suggest that contamination goes deeper into the ground than diffusion model predicts, especially in areas of ,

l higher contamination. Data above from Rancho Seco (area no.s 1,2,4,5), from

' Humboldt Bay condensate storage tank area, and from Turkey Point were plotted on Figure 1. Also plotted was depth profile from the DGEIS. It appears that the DGEIS. curve reasonably plots the profile for the early part of the curve, 10 l - .. -

but that the real data shows a bend and a deeper profile than the DGEIS.

Perhaps the FGEIS model could include this deeper penetration in some manner.

Because the Humboldt Bay curve seems to reasonably model the Turkay Point and Rancho plots, it is used in Figures 2 and 3 to estimate volumes and volume ratios.

The data from the Contamination Events and from the deeper renatraH on such as Dresden or Rancho, areas 3 and 6, need to be considered. Figure 4 centains a, suggestion ac to how the Contamination Event data and Humboldt data could be combined.

The'following table based on Figure 3 suggests a volume ratio which can be used based on the profile information in that Figure:

USE OF FIG 3 TO MODIFY PROFILE FOR SENSITIVITY ANALYSIS DGEIS DGEIS modified modified total incre. HBSC total incre.

Dose volume volume volume volume volume crit pC/g ft3 ft3 ratio ft3 ft3 100 20 196 60 12 296 100 1.65 317 121 30 6 406 110 2.5 500 183 15 3 459 53 4.5 900 400 10 2 511 52 3 .6 743 243 A

11

3. RESULTS OF USE OF REAL WORLD DATA - PRELIMINARY The following general results from the real world data available appear to be reasonable:

Based on review of general site data and specific site area information l 1.

i (Sect 2.a.1), most areas of site would appear to result in doses of <l5 mrem using dose conversion.factore..from NUREG/CR-5512.

2. Certain specific site areas appear to be more contaminated (Sect 2.a.1).

From Section 2.a.1, the levels.of contamination are comparable to those in the DGEIS (Co60 10-12 pC/g; Csl37 9-30 pC/g as compared to 2-60 and l

1-20 respectively in the DGEIS). The areal extent is also comparable to the DGEIS (2500-3000 ft2 as compared to 2000 ft2 in the DGEIS). l The depth of this contamination is not clear, although limited depth information is available from Rancho Seco. Other limited depth data is also available as noted in Section 2d above.

3. The large majority of information (Sect 2.a.2) from 20.302 burials appear to indicate that contamination is at a level which would result in doses of less than 15 mrem using dose conversion factors from NUREG/CR-55n and that a reasonable referece facility would have 4 contamination at doses less than 15 mrem.

4. Airborne effluents do not in general (Sect. 2.b) appear to add to the contamination levels from other causes such as spills

5. Information (Sect. 2c) on specific events that caused soil contamination appear to indicate that there may be instances which result in larger volumes of soil than estinted in the DGEIS.

6. There is limited information (Sect. 2d) on distribution of contamination with depth or volume. In particular the information from Vermont Yankee on distribution below the building is not clear.

RESULTS OF PRELIMINARY ANALYSIS - Based on the preceding items 1-5, it would appear reasonable to retain the reference facility from the DGEIS with regard to the analysis of contamination in excess of 15 mrem with three additional potential points of analysis:

1) provide an estimate of additional volumes of soil at levels below 15 mrem to take into account the various Nuices of contamination (e.g.,

buried waste, sediment). This analysis may be either quantitative or qual itative .

2) consider added data on soil profile (e.g., as noted in Section 2d) if available and reasonable. If unavailable, the current diffusion model would continue to be used.

3) based on information on contamination trom events, (underground leaks, below building contamination) consider an additional analysis of a larger soil volume. This might serve to indicate possible effect of 12

1 t

i e

i l

l larger soil volume. Use a real wor 1,d distribution if possible. ,

t t

l 1

+

1 .

I I i

l i

1

1 l

t I

I b

l 13 s....,.,._ _ _ _ _ - - _ . . - - - - - - - _ . . - - . _ _ _ _ . . . . _ . _ _ - - - - _ - - . . . .

i TABLE 1. ANALYSIS OF 10 CFR 20.302 LOW-LEVEL CS137(YR) 00SE(95)

WASTE SOURCE BURIALS VOLUME AT REACTOR AVE.VOL CUM. V01SITES V0tUME CDC0(YR)

PLANT YEAR TYPE of soil (ft3) (pCilgm) (pCilgml Myr i letter dete im3) (ft31 (ft31 '

measured 170.25919 5/3/82 70.0 2472 2472 2472 180 5

. Turkey Point 82 soil

~ 78 81.567168 91319 3 229.4 8100 26 Dresden1 76 soil 28.3 4550 7022 23 4 65.309169 8130l89 sew. sludge 1000 Vermont Y 90 i i

~ 37 42.887054 1 011 018 5 424.8 15000 22022 -t 15000 1.2 85 sand Ocon:e ~

0.42 15.2 ~1838772I 3I21191 118.2 .4173 26195

. soil 4173

• Palisades 87 *

' 0.5 12.795313 3/14185 380.0 9

Robinson 85

'82 soil soit 13420 17000 W 3,1 [8.1779764' 11/9152 481.4 i Dyster Creek 1.7 6.6204401 ' 513/8 3 1.5 53 4.6 83 sediment 1682.0 Robinson Dresden 1 88 soil soit 59400 162000 g 2 1 2.4 5.2272626 5.2191527.

91419 3 11/26186 4587.3 -)

Surry 86 4.4009452 1 211 619 3 934.5 Cook 82 sludge 33000 0.6 0.08 3.7 3.2 3.6768105 3I11/85 5776.6 69839 96034 k 85 2nd resin 204000 (Tl i Davis-Besse 0.7 2.0052946 8/24/90 382.3 13500 0.5 -

Big Rock 89 sediment 1.3638396 11118l91 1642.4

. 0.4 0 soi! 58000 Vermont Y 92 0.7 1.114037. . 513/8 2 1183.6 37767 133801 41800 0.5 Turkey Point 81 soil 0.5 0.13 0.8248295 12l11191 61164.4 sand 2160000 113.3 Bmnswick 85 0.15 . 0.5818009 12/16187 4000 0.23 ,

87 sew. sludge 300.0 i Point Beach 0 0.5 0.5085537 9/24l81 sand 10595  :

San Onefre 81 0.08 0.4899326 1 11818 4 1032.3 552763 686563 36455 0.28 85 sediment Robinson 0.12 0.2441659 41819 3 2237.0 79000 0.05 Pilgrim 88 soil 0.1631708 3/17187 850.0 0.12 0 i soil 30018 Formi 2 85 0 0.1192413 10/18184 2831.7 l 100000 0.1 McGuire 84 waste water 0.1045664 4/22186 86.0 0.1 0 i sew. sludge 3000  !

laSafie 83 0.02 0.0943831 9/1/94 424.8 15000 0.02 92 2nd sludge 150.0 -

Ktwaunee . 0.03 0.04 0.0793597 Sl8186 soil 5297 71994g i

Big Rsek 84 0.0620234 3/25187 - 39.2 33386 1384 0.04 01  !

Grand Gulf 86 filter residue

[

^@^A^A 3 ^E^C^D^Q I

966t( 2.

4 ci/mnht ci/mnht Co60 Sr$0 Cs137 MWH ce60 cs137 ANO1 6.00E-06 2.87E46 7.21E44 1.30E+07 4.685E-13 5.546E-11 ANO2 6.41E-06 2.03E-05 2.262+07 2.836E-13 8.982E-13 BV1 2.34E45 9.33E 06 2.02E+07 1.158E-12 4.619E-13 1.21E-06 1.43E+07 0 0 BV2 1.95E+07 2.569E-13 0 Brai&vood2 5.01E46 Byront 8.60E46 2.16E+07 0 3.981E-13 1.29E47 1.87E+07 0 0 Byron 2 2.45E+07 5.224E-14 0 Calowsy 1.28E46 3.25E48 Calvett1/2 1.38E45 1.05E44 4.34E+06 0 2.419E-11 Catanbat 1.04E45 3.67E.06 4.17E-10 2.08E+ 07 SE-13 2.005E-17 1 Catanba2 1.04E45 3.67E46 4.17E-10 1.93E+07 5.389E-13 2.161E-17 Cook - 1.18E44 5.07E 06 2.52E 02 3.58E+07 3.296E-12 7.039E-10 1.28E+07 2.414E-13 0 Crystal 3.09E46 1.11E-06 DB1 2.75E47 1.49E45 1.32E+07 2.083E-14 1.129E-12 Diablo1/2 6.46E.46 1.88E 06 5.11E+07 1.264E-13 3.679E-14 Ginna 5.29E-07 2.24E46 1.07E+07 4.944E-14 2.093E 13 Haddam 5.78E 05 2.55E-05 1.48E43 3.81E+06 1.517E-11 3.885E-10 2.05E+07 3.766E-12 0 Harris 7.72E-05 5.59E44 3.17E44 1.66E+07 3.367E-11 1.91E-11 ,

IP2 IP3 1.12E.45 1.57E+07 0 7.134E-13 Kew 7.51E46 9.69E47 1.24E+07 6.056E-13 7.815E-14 -

Maine Yan - 1.09E-02 1.00E43 1.50E+07 7.267E-10= 6.667E-11 i McGuirst 1.79E44 1.02E-06 9.09E-06 1.48E+07 1.209E-11 6.142E-13 McGuire2 1.79E44 1.02E46 9.09E46 1.93E+07 9.275E-12 4.71E-13 1.60E-08 2.21E46 2.52E+ 37 0 8.77E-14 l W3 3.97E-05 4.22E+07 3.934E-12 9.408E-13 l NorthAnt/2 1.66E-04 Ocont/2/3 1.67E45 1.49E-07 8.28E44 6.09E+07 2.742E-13 1.36E-11 Patsades 1.72E-05 2.88E-06 1.23E44 1.01E+07 1.763E-12 1.218E-11 PaloV1 4.31E44 1.86E45 1.45E+07 2.972E-11 1.283E-12 ,

0 l PaloV2 5.40E-05 1.91E+07 2.827E-12 5.90E48 2.92E+07 0 2.021E-15 l PaloV3 PointB1/2 3.56E46 1.91E44 2.25E+07 1.582E-13 8.489E-12 Prairie 1/2 5.01E46 1.19E45 2.45E+07 2.045E-13 4.857E-13 l Robinson 1.20E44 9.44E48 1.08E+07 1.111E-11.8.741E-15 1.90E+07 8.053E-13 0 Salemi 1.53E45 Salem 2 8.56E46 9.06E 07 1.69E+07 5.065E-13 5.361E-14 Son 9st 2.33E-07 2.43E.05 5.04E+06 4.623E-14 4.821E-12 Songs 2/3 7.17E45 3.48E45 4.59E+07 1.562E-12 7.582E-13 4.29E+07 1.219E-13 0 Sequ1/2 5.23E46 ,

SouthText 2.46E-45 3.13E44 1.90E+07 1.295E-12 1.647E-11 2.02E+07 5.545E-13 0 SouthTex2 1.12E-05 1.43E+07 8.881E-13 0 l Stluelet 1.27E45

- StLucle2 3.40E.45 1.70E+ 07 0 2E-12 Summer 1.69E-05 1.96E46 1.93E+07 8.756E-13 1.016E-13 Surry1/2 4.92E44 9.34E44 3.36E+07 1.464E-11 2.78E-11 .

TM1 5.66E-08 1.69E+ 07 0 3.349E-15 TM-dnm 9.13E-07 1.12E-05 1.90E+07 0 5.895E-13

/

• Tfs 181: E

.t c

WolfCr 7.53E-05 2.36E+07 3.191E-12 0 Yankee 1.99E44 1.04E43 2.94E+06 6.769E-11 3.537E-10 Zion 1/2 :i.o/t-@ ~ 1.72E45 2.25E+07 2.498E-12 7.644E-13 '

0.0002601 0.0006514 1.108E+09 2.133E-11 4.274F 11 , , _

4.65E44 9.32E44 ,

cl/mWht cl/mWht Co60 Sr90 Cs137 MWH Co60 Cs137 Big Rock 1.92E-04 3.39E46 2.19E44 1.40E+06 1.37 tE-10 1.564E-10 Brunst/2 4.47E43 4.30E46 9.35E45 2.67E+01 1.674E-10 3.502E-12 -

Cinton 2.01E44 1.15E+07 1.748E-11 0 Dres2/3 1.67E42 5.66E-06 2.74E44 3.02E+ 07 5.53E-10 9.073E-12 1 Arnold 2.49E43 1.23E47 6.67E-06 S.04E+06 2.583E-10 6.919E-13 .

Fermi 1.56E-04 6.01E-06 2.25E+07 6.933E-12 0 Fitz 4.53E44 1.38E46 3.93E-05 1.42E+07 3.19E-11 2.768E-12 '

GG1 6.96E-05 1.11E-07 6.30E47 2.43E+07 2.864E-12 2.593E-14 Hatcht/2 2.30E44 1.62E46 2.92E45 3.43E+07 6.706E-12 8.513E-13  ;

Hope 1.66E44 1.37E+07 1.212E-11 0 Lasalet/2 1.22E-03 4.55E+07 2.681E-11 0 Limerick 2 1.55E-04 2.29E+07 6.769E-12 0 hP1 2.51E44 1.38E.06 1.20E-04 1.56E+07 1.609E-11 7.692E-12 l 1

Monticeto 3.70E44 3.98E-06 1.76E44 1.40E+07 2.643E-11 1.257E-11 SP1 8.89E-04 8.14E-06 1.68E-04 4.07E+06 2.184E-10 4.128E-11 :l EP2 3.48E-04 1.34E+07 2.597E-11 0  !

OC1 3.22E-05 4.56E-06 1.36E+ 07 0 3.353E-13 PeachB2/3 2.19E-06 8.27E-05 9.05E-05 4.40E+07 4.888E-14 2.02E-12 Perry 6.75E-06 2.81E-11 2.00E+ 07 0 1.405E-18 Pilgrirn 1.35E-05 6.67E-06 1.51E-07 1.29E+07 1.047E-12 1.171E-14 QC1/2 1.01E-02 1.40E-05 2.73E44 3.08E+07 3.279E-10 8.864E-12 RiverBend 1.05E-04 2.66E-05 1.78E+07 5.899E-12 0 Susquet/2 1.55E-04 4.68E+07 3.312E-12 0 VY 5.62E-04 8.86E45 4.66E-04 1.13E+07 4.973E-11 4.124E-11 WNP2 5.19E-03 5.85E-05 1.20E-03 1.80E+07 2.883E-10 6.667E-11 9.524E-11 2.082E-11 2.08E-03 4.54E-04 uC/m2 pCl/gm

! C060 4.435E-05 0.0044352 CS137 3.395E-05 0.0033948 .

l l

I

1 I

t .

r f

R 6, /

1 vu e -

p 1 1 4

! l i

, ._ ,4_ _ _.,7 _ _ _.' ,_- q_ _.

_4 . .

-( _

- ;_ ._ _._p _ , _ . _ .

J , \}

\- -- --

- -i .

l  ;, , 4~\ ,-y -  :

4 .

ii \ _J g g \ J M i .i -

y, . .

\; -i ,

y ,

t4 -

t+.  ! - =usq ij -

.\:- , ..(.i. .

}

g .g . . . ;.\.- ._ # =2: i j w x\ _

x -

.'\

_. 1 .2 ..'."*1

.i .~...'.I .1._._ _ l. . . .. J._ . .Z _ 1. 17.~ ~~ ... '. . ' ' ' . " . ' _ ' ' . ~ . _

' .q.

m ..  ; .. g. , . . . . . . .

't '

MIEiEE7 EL. -

TES

-a = == EE*M =EE-= =s s=

. --- h - - --

Eska =

=as=-

=.=-.==-- -

-.-=.- :=--. - . -

\

o - - -

0 -s . -

m 5:

[ 4 %.-h= 341- . ... _.

I wb 30 .

y-- _x - v..p mm ,e ,n ,_. ,

,,,g .__,v r

g .

3> zip

. s. 1-g

.1 .

s_ x m m m.

'7 ' .\ \ t \ .V \. '.

~ '

D p . '- I \ \. \ N N.r T

• L 4-f ,

\ -\ \ ,

g , .--\ 54 Ti

\ -

7A .

2 N-

&.. 2 -

-43 ,

% p\ \ \= .A . v. q 7 \ \  %' N

? '

, \ \ \ N .

*l I; i  !\ \i N Nto'. 6 a y \, ..i x x. . . . . _ , .

3

\. .. . ... _ . _ _ _ ,

-s t, , \ iN* [s. , h~!N/ ,

, i ~

q  ;;;

j -!\\  !

3 s

,  :  ;  ! \ ~

f

. 4 -% . .p:. 1 Q .

< \ x

- Rt{0$-o\ w.

i _<oec -

Q .- .K. - ' m s g. s' .. .- .

g . . . . _. _ , .

g

my-- gm

~~+ v - n m np m -

nu-on w m3n yn;

. m

,w w

,r

r. r vg7q $m.w, , K 5 i -

g :: -n  ;

q g

---^ -

._ \ 1 ig

_1

,- .F_l =. .':

,. .' 1J ~ I ' ". i fi g,.

a. _.i. J. J .

e

\ . .g 't -- ;^ }h 5 *, .. j \- i . iS u r-jv=.hz

'~~'

.sp- e ~ n  : 4 t-+-\ 5.-i ". 7d #if

-t*-

} {: -

a -+=- --

u.

~.p -'

,94 } -W ICF 1.L .

y[g, ,

. .6- . ,- t.--- , .. i. . ' ,ir C-ij t : t- - 6 9 h" - -; }:; M -4M:: ' :- 9-DEE $h ---m----

g m - -

-- 9 - w :: :- .

q:gg

~

W l

-. 5 .

• ~

i l l .

~

1 d' ---j i I  :' i i s.

, v

.i .. .k i : . .

l t

-4

_ _ . _ 4 __ . __ _.__ L- . - _ . ___ .. __.' N,y

4 0

. Fi6. ?_.

l I

. l i l

_ __. . .,.m. .

..6 k.Lw. ..-+-w]. ,&.

, _-- __ _ __m

._..s, o_.

s n _ _ . , . w-.~

_ _ , _ - _, _9-.p . .- . _ . , , . . . .

____4.._..

6 ,, 1 9

, , , , 7 , y

__. '; _ _ .___ j..  : . _

- .l_ .

_ .__ e e . -

f; .

1 __ 1 11' -  !

y ,  ;- - .j- i -- i -

3 .

i

~

d a .i _ . . - l p ._

.l

.a _

_.. -.-,j l .

e r

i

. . . :- . -C

-}

__ r- I ~. .. . I :~ l

. t

_ _ , (=  :  :.;m s :.--- - rna ::: :::t=-. .r-- : .r .

. - rt =  : :- f.. : :::

-3_?~:.' -]*Si ~ ~.5.. i. 5. :. .T .^ .~. T 5 ' ^ ' -W~ ~ ~~~~~~

,~d.._ _ $. . h. . . . ~ ~ lH .

a . _ . . . .

.._ j . .j . .

. l 88 - ..w..,,s,,

8 j- ,y- ,

%p ;. ._ .

M.* ,

8 I I~ 'II r 1

i. - -

I I

h.

e 1 1 i t ' l o .

A S, }  !  !

to

,, 1 1 . .

< 1 l

% a

_ .-.  :. ~b ^

l 1

\D .

~

E 3 l.?00 E ,

• c 4g m na - e by .u gw .,__wm,n, ,

w w w. ~

m p+

wAw , rw

% .. i-(

\

l [, p ' \.

• i - ' "- I , ,

4 Q. i; i

s. _

l

• i e .

s .

,-- .1:m9 --~- k ,l-4 * :7 -- -Y- E --'

~ ' ~

i. o .c -

di .

t&%.. #@, = i .:

p  : ; N: .. ..t.

4 : .--

._- .- .r-.

.-wri: H05c ot,c.r.o l . ; =a u  :. .

5:}E=- = .b - m .. - d p= 5 E E lt: -

• l' i(

=c - ..

N

.1---. . :n :p - - - : : -.

3 y, i ,

j 2 y kh-  %

~

, 1.._.. ._ _ _ _ _. tx 2_ _ _. _

P9

9 e

L i.

e 1

i p._14

.x 4 H., + 4m. 4m. ~ . . . , . .

_m .

. - ., f,

.....p.~-.. _ . . . - e.r . _._ g_... 9 ..., _ _ . . _. ___p_,.

i,

_-.4.. _p~.. _ . _ _. .

_a ~ _. _4 . , , .

4 I I 1 1.

i e .I' i

, ~~.f--- -( - _ . _ ,

i* t ~ t*- -

l 5! 5  ![ .__ ..

'I I: . I '.

h it-' it ,

{g ,

l 6

W=

= . .

J.L . .

e

~~;.~ *; ~~ _. , ; . . - l

~ - -

-.: L ~. * . ~ ~ ~ ,. . . '.. "',"' .*^~.~* .

_" ~C - . . ." .': ...

bb-- -- kh kh :h .C - TC T -' ~C 7 ~~1 ~~

E

.. - ..... N~l _ _

I,

. 54.-

~

- , - lL.

t -

g...- _ _

x-' mVm g g 4 . .. , ,1 _-

, wt ' - , it i

Q g

. i

1. I z .

r 4 d ] 'i i i

-1 1  :

i 3 a

$ ',~ l .

9 , . . . - - - - - - ,

p 3 k  !

i I

i

~ ~

i i i  !  !

e. .,

O 6

sc -.of I5 -s y. .o _; .

...N... ' ...

. -N

., p4 g y '_ : t o - 3; g = 4g j]3 ,e ,p, gi pyg  ;, g 9

_ r

..+ ; .' . ,, ; -.

-+-

+-:.

hwg . ., .

_- Wv. -e- +q.=*A ;-

=. .. ,-

1  ; , .; :

,,;r-g .. .

y i. r k

JE e 'i .

-i' 4 -

. 6

' t '- -i '.t'.'~'

hi+.

g

_ _ _ _ f -

._ 4 5 =. -

.j --

Q j fi.1 rl.

-t b E-9 d 2.~- : - ." t ' .. -

~

g _-  :

4.- ..- Om : :

, g. -

- ~ '

_^ iri1l

) go a ,

. s:  !

,2 m

_ . .:.=.

2- ..

F, - - F.:.- _ - 9] l= 1-. g Q: _ til" pj .c.. H? M.i nit 54"; W-Q R

1 '

. g -

If 3 w

I _

M, .

da &44.af -+ 4h.4aee *A' adwsa .-r4b 34 i J 4 Sb #4JA- 5. S'<4- f-3-4.m 5-8E*h_ a* ~w.A 0.ha*6 A=M-i$4 e'---dJ -J$ D4Ed Et mkA6A2 **C-9*e=abaama s L.m.Ji.Jj.4,.. .M 4-.J.Jawm.

4. .Aa 4 -6*- 4h44.

i

• r t

5 i

r\ b ,

l'

.l .

I I

f

'l t

l j I l e

1 l

-.,-y  !.. ___{.__ _ ._,. ._; _. . t. . . .

._3 9 g

8

. j. _

-~.L.. _ _.-

, _.. q i

• b"*' ***-*4-

' ' - ' = .

-.n-+-

1

.e.. 6 I ' +-!  !,

ff. f' .' tI 'i-t I

• • SG #

N' S l'  !- l j IIt,'

I l- l t

. . - t 4

t cl  !

l u .

1 4 l a 'f -l l l 9 . .

t X

• t . . . .. .

l s

= .; 4 = . *-, : . 2 u .- .. . ;c -. :  ;.- ... . : _; . .. c- :;-  ;  ; . . -- .  :- *.: - ,

-! ?.**

• J 7 '. * !

.2* * * * *,,

- . L*., ". ; !

7.*

• .M*'*.*,

. ~':~^

• ,"J.$ " ? .-**

- - ~ ~ - - -

• ' ! . ;Or 3 ~~ ~

_-~.~-

t y

G. 1.

I

. , - 1 g 1 8

l-

- - 1 y . - , _ -. - , .. , _ _ _ _

-,- q 3

.i 1  :

r i

< i- i o

' l I 7 4

A : e . . -. -

)

3 e

.h h h N \ i h

M g(l i

.i .;

i \ i i

t Q a, .t I,

L e t. '

k= ,

i D $ 2 k

{

f '! )

S NP

+'d L I

! l Is - SQ t

%.i  : ,

l i

hofc .

i p

~ '

10 ' '

g T%iTI'

I ' Y'  !  !

I

'Y

\ 2

' 1 74 r 3 1 * ,, . s i i I.i N -. 's- i

'X-g

!_ *\ t'

\ i y

q (a 1'

!  !- \ I

1. .

, . 1 A . A._ %  !

ji s: %._% t- t-N -

T ) l[

p - - -

, , p 'i -

l . n',

l y, N -

t N -

j 4. r Na-g n-- ---

\s l  ;

.... e.r, . - -

. a =_ ,..  %. . .i . s g. ..4..... . m. g

%Q ~.

5\

n ., x b'b. ' N l l, f, ktm- I

\

. NN '

y l l

i , .

a- - . . -

t _

s _.

s -

Table 3 Summary of contamination events The following das the result:

ft3 Co60 Csl37 Big Rock CST leak 5900 1985 . (removed 60 2000 remains 5300 .14 Turkey Pt spill 2500 180 5 1982 Fermi CST leak 30000 .12 1985/6 VY Sink drain 58588 .09-1.1 1991 Yankee IX pit 1964 Dres 1 Solidif 8100 30 100 leak Rancho BWST valves removed 110 79 230 remains .6 2 Palisades Flooding removed 1560 1.3 >500 remains 6000 .05 30 h (No'P617)

b i l L

l B. Summary of Public comments on building contamination in Reference Power ,

l Reactors .

A commenter indicated that volumes of activated and contaminated equipment and i structures increase exponentially with decreasing dose rate below 500 mrem /yr (87). In particular commenters indicated that costs of activated concrete,  :-f- in particular for some reactor designs where significant volumes (100,000:

, -- ft3) of concrete become slightly activated, are not included in the GEIS (183,189) 1 VERIFICATION / MODIFICATION OF REFERENCE POWER REACTOR In response to the public comments, the staff reviewed the reference power reactor for the purpose of verifying and/or modifying the reference cases.

The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is presented; (3) the basis for the verified and/or modified reference facility '

is given, t

l I

i e

1' r-i

i

{

1. BASIS FOR DGEIS REFERENCE POWER REACTOR ,

The reference power reactor is described in App. C'of the DGEIS. Data on  !

contamination levels is based on real world data from a 1986 PNL study l I

(NUREG/CR-4289) of 6 reactors. Some depth profile information was available which was combined with a simple diffusion model to estimate spread of  !

coatominativa into the concrete. Briefly the building contamination model is as follows: t Structures contamination:

r General buildina contamination Areal extent - ft2 %contam Containment b1dg 20400 100 Aux b1dg' 43000 1-5 ,

Fuel bldg 53800 1-5 l Contamination level - Co60 Cs137  ;

dpm/100cm2 dom /100cm2 7.5E6 2.4E6 .

Depth profile - contamination with depth into the concrete (based on i

NUREG/CR-4289)_such that concrete must be removed as follows:

total '

Conc removal for: depth (in) vol (ft3) 100 mrem ~ .5 2276 30 mrem .5 2276  :

15 mrem .5 2276 '

3 mrem .63 n2780 ,,

Extended wet spot contamination -

Areas exposed to liquid contamination i'or extended periods of time is assumed to penetrate deeper, therefore it is assumed that the entire thickness of the floor is removed, a thickness of 6 in.

Depth profile - none estimated, i.e., it is assumed that the entire thickness of the floor is removed, approx 2500 ft3 for each dose criteria Activated concrete -

Activation of the bioshield is such that concrete would be removed as follows:

total Conc removal for: depth (ft) vol (ft3) 100 mrem 3.5 11000 30 mrem 3.9 12500 15 mrem 4.1 13000 3 mrem 4.6 15000 18

i .

l i

2. Real Warid Data Analyzed -

l 1

a) General building contamination

' Decommissionino plans i

Decontamination plans from the following facilities were reviewed to assess j whether real world data provided input to verify or modify the reference power reactor:

Rancho Seco Yankee Rowe  ;

Trojan Big Rock Point l Shoreham l I

In general the data consisted of scoping studies anJ samples from locations in different building areas. The following is a summary of the informatica l available in those plans:

Big Rock Point l i

Table 6.2.1 and 3.1-1 and 3.1-8 indicate the following contamination level range:

1000 - 2.4E6 dpm/100 cm3 1E 1E-7 uti/cm3 i

Rancho Seco Table 3-5 indicates the following contamination level range ft2 dpm/100 cm2 reactor bldg 83000 lE3 - 3E5 aux b1dg 80000 <1E3 - SE5 fuel bldg 12000 Turb bldg 68000 <1E3 Table 3-16 indicates the following reactor bldg 0060 - 600 pC/g; Csl37 - 1570 pC/g Page 3-9 indicates that the majority of the contamination is in the top cm

except for cracks. Also it is indicated that the floors and wall activity is l l

l the same as for DAW.

The total volume of concrete expected to be removed is as follows (from Table 4.2 of App B):

ft3 fuel storage bldg 2160 aux bldg 2760 RB 11600 19 l

i Yanktre Rowe Table 3.1-6 and EA, pg 10 indicate the following:

dpm/100 cm2 depth (in) j general <1E3 - SE4 .2 to .6 --

l contamination  !

spent fuel pit lE4 4 IX pit SES 6  ;

fuel chute 2E4 4 l

The total volume of concrete expected to be removed is as follows (SER table 7 and EA table 1) is 9000 ft3.

NUREG/CR-0672 Add. 4 l

ibis document explored differences between the PNI. model used as the basis for the DGEIS and the model used by TLG Engineering. TLG is a recognized provider of cost estimates for power reactors, and as such maintains cognizance of  ;

reasonable contamination models. l 1

Table 5.9 of NUREG/CR-0'672 provides comparisons of radwaste volumes for decommissioning. It is indicate there that the TLG estimate for concrete  ;

volumes is approximately 2/3 of the PNL reference reactor estimate. l NUREG/CR-5884. Revl. App. M This document provides updates on the cost of decommissioning a reference power reactor. Appendix M contains comments received by TLG Engineering on the draft document. TLG is a recognized provider of cost estimates for power i reactors, and as such maintains cognizance of reasonable contamination models.

Comment 003-30 indicates agreement with the data of NUREG/CR-4289 thtt concrete contamination rarely penetrated more than 1 cm depth into the concrete. Also comment 003-30 does not indicate disagreement with the building surface areas indicated as requiring removal.

Comment 003-31 does not indicate disagreement with the volumes of biological shield concrete used in the analysis. l 20

?

e i

.e .

I I

I

3. RESULTS OF USE OF REAL WORLD DATA - PRELIMINARY The following general results from the real world data available appear to be reasonable: i

1. Based on review of concentration-levels, the levels used in the DGEIS I appear to agree with the levels experienced. l

2. Based on review of waste volumes, the volumes used in.the DGEIS appear to agree with the volumes experienced.

3. Comments received from technical experts on earlier PNL analyses of reference facilities appear to indicate general agreement with.tne ,

extent of contamination.

PRELIMINARY ANALYSIS PROPOSAL - Based on the preceding items 1-3, it would appear reasonable to retain the reference facility from the DGEIS with regard to the analysis of contamination with one additional potential points of analysis:

1) the basis for the comment that there are significant activated concrete volumes in excess of the DGEIS should be investigated i

21

s .

e-ATTACHMENT 2 l VERIFICATION / MODIFICATION OF URANIUM FABRICATION PLANT, -

Public comments have questioned the model used for the reference uranium ,

fabrication plant. These comments are summarized on page 2.

In response the staff has analyzed real world data to verify / modify the ,

reibrence ufab plant for both soil and building contamination. This analysis

~

is contained on pages 3-20.as follows:  ;

Soil contamination ,

Page 2 - summary of public comments on soil contamination Page 3 - DGEIS reference ufab plant soil contamination model  ;

Page 4 - analysis of real world data for soil Building contamination Page 8 - summary of public comments on building contamination l Page 9 - DGEIS reference ufab plant building contamination model ,

Page 9 - analysis of real world data for building contamination Results and preliminary conclusions Preliminary results of analysis of real world data' for soil contamination are on page 7  !

Preliminary results of analysis of real world data for building I contam'ination are on page 11 l i

i l

4 A. Summary of Public comments on Soil contamination in Reference Ufab plant Soil Contamination - Commenters indicated that the model for soil contamination in the GEIS is not accurate for the following reasons:

(a) the volume of contaminated soil in the GEIS.was too low (128,171,172, 173)

(b) the.model for distribution of contamination in soil was too simple (128, 172) and that therefore contamination would be spread over significant volumes of soil; thus the volumes requiring removal would increase rapidly (even geometrically) with cecreasing dose criteria (128,172, 173).

The physical reasons given as to why the models are not accurate are as follows:

(a) Use of a 15 cm depth is only appropriate for farmland where the source of contamination is dust or spreading of contaminated topsoil. Under realistic conditions of mixing and burial, contamination will likely extend beyond 15 cm (128)

(b) The GEIS model rasults in primarily surface contamination. In real situations, contamination is more a function of site waste management practices, e.g., burials, deposition, soil mixing through previous site activities, leaks in underground systems, etc. (128, 172)

(c) the vertical variebility in the GEIS is too simple which leads to contamination depths in the CEIS of < 15 cm even though experience indica.es depths of contamination considerably deeper (214).

2

i l

I .

l

. I l

VERIFICATION / MODIFICATION OF REFERENCE UFAB PLANT in response to the public comments, the staff reviewed the reference Ufab ,

plant for the purpose of verifying and/or modifying the reference cases.

l The. discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is s="@d; (2) The real world. data considered is i

< presented; (3) the basis for the verified and/or modified reference facility 0 is given.

1. BASIS FOR DGEIS REFERENCE UFAB PL' ANT The reference Ufab plant is described in App. C of the DGEIS. The reference

case was modeled in the following manner:

1. Contamination levels are based on limited real-world data which indicated that at the Apollo site the contamination levels had ranged

- from 20 to in excess of 500 pC/g.

2. The areal extent of contamination for a r=ference case was based on engineering judgement.

3. Because little depth profile information was available, a simple diffusion model for spread of contamination into the soil was used.

Briefly the soil contamination model is as follows:

3 Site contamination:

Areal extent 50000 ft2

. Contamination level - Unat 1000 Depth profile - contamination with dapth'is based on diffusion into the soil such that soil must be removed as follows:

total l Soil removal for: depth (ft) vol (ft3) 100 mrem .6 30183 30 mrem .7 34777 i

.75 37729 (

15 mrem i 3 mrem .84 41830 Under building soil contamination Contamination is assumed to penetrate Suilding cracks to a depth in soil  !

of about 5 feet.

Depth profile - none estimated, i.e., it is assumed that all of this l

soil is removed, approx 24000 ft3 for each dose criteria l l

3 l

1

2. ANALYSIS OF REAL WORLD DATA l 1

Information on site contamination was obtained principally from facility  ;

decommissioning plans and other general reports on decommissioning of uranium '

l facilities. These are discussed below.

(a) Extent of contamination Decommisst e ing plans and other documentation from the following facilities were reviewed to assess whether real world data provided input to verify or modify the reference ufab plant:

Apollo i Cimarron I UNC Wood River Jct UNC Montville The following is a summary of the information available for those facilities: f Apollo The decommissioning plan provides estimates of total volumes of soil contamination removed. It is indicated that remediation and disposal of soils l containing > 30 pC/g is planned.  ;

1 I

No information on distribution of contamination with volume is indicated.

Based on site sampling, total volumes potentially > 30 pC/g are indicated in Table 2-10 are indicated below.

area ft3 Beneath main b1dg 373,000 North sewer 52,880 South sewer / riverbank - 433,774 Middle sewer 37,712 Parking lot 241,429 It is noted in Table 2-10 that some portion of the material may be found to be

< 30 pC/g. Information given elsewhere indicates that soil volumes from all of the areas total approximately 500,000 ft3. An additional data source (Paper by Carlson et al, " Decommissioning' of B&Ws Fuel Conversion Plant",

l- presented to NAS, June 15, 1994) indicates that 665,000 ft3 of contaminated

! soil at > 30 pC/g was disposed of offsite at Envirocare. No information is l given in the paper as to the breakdown of the above areas, the following is an estimate:

I area ft3 Beneath main bldg 90,000 f North sewer 4,000 South sewer / riverbank 295,000 4

l 1

I

Middle rewer 24,000 Parking lot' area 233,000 As noted above, several areas were contaminated. Information in Figure 2-4 '

regarding contamination is combined with the site layout of Fig 2-1 to estimate the areas noted here:

area ft2 depth (ft) general site areas 90,000 2 (some deeper) south sewer 26,000 5-25 middle sewer 5,000 4-20 below bldg 1/,500 2-15 Cimarron The information available for Cimarron indicates the following with regard to volumes of soil:

ft3 pCi/a 500,000 between 30-100; avg 70 150.000 >100 (enriched U)

UNC Wood River Junction The information available for UNC Wood River Junction indicates that there was lagoon sludge in the following volumes which was cleaned to the level indicated: y ,

ft3 gCfg 300,000 30 .

i UNC Montville The information available for UNC Montville indicates that appoximately 20,000 ft3 of~ soil was removed. UNC Montville produced Navy fuel and is in general different than typical Ofab plants in that it is making the end product and the material is encapsulated when received.

l Texas Instruments Partial information on uncompleted work at Texas lostruments indicates that approximately 200,000 ft3 of soil would be removed to reach 30 pCi/g.

(b) DATA ON DISTRIBUTION OF CONTAMINATION 5

i Data on volumes and distribution of contamination is important because it is l useful in aetermining the amount of material (and cost and impact) which I results from achieving alternate dose criteria. i Generally this data is sparse which is why simple diffusion models were used in the DGEIS. In the review of data, there was limited data on distribution of contamination in the decommi.ssionino..njans.

Documentation concerning distribution profiles is contained in a letter from DOE to EPA on 2/24/94 which contained draft graphs of cleanup criteria versus waste volume for 3 DOE uranium facilities, Elza Gate, Ventron, and Aliquippa Forge. These plots are included here as Figures 1, 2, 3. The ratios of the volumes in Figures 1-3 are shown here:

pC/g DGEIS Ventron Aliquippa Elza  !

130 1 1 1 1 80 1.02 1.23 2.3 1.03 39 1.15 1.6 4 1.04 30 1.2 2.27 1.25  ;

2.44 l 20 1.25 2.76 10 13 1.26 3.38 14 4.4 4 1.39 4.5 As can be seen, the data, while tending to vary and being fairly limited, suggest that the contamination goes deeper into the ground than diffusion model predicts. The DOE provides a discussion of the assumptions regarding ,

the Elza site. The Aliquippa site contains lower volumes. Thus use of the )

Ventron as representing the real cases is illustrated below in Section 3.

By way of comparison, Commenter #63 submitted a figure which was stated as using actual site data indicating that the soil volume multiplier versus pC/g ratio from 100/15 mrem was approximately 3-4. The data supporting the figure was not submitted. Also Commenter #96 submitted a figure based on Cushing information which indicated that the ratio was approximately 6.s The concentration data supporting the figure was not given.

I 6

. t I

3. RESULTS OF USE OF REAL WORLD DATA - PRELIMINARY ,  ;

i The following general results from the real world data available appear to be j reasonable: ..

1. - The total volume of contamination for the cases reviewed is larger than  !

the DGEIS case, the total square footage is larger, and the depths from l Apollo teble.zW the DOE sites suggest deeper penetration with depth. i Specifically: "~

a) The general site area contamination is about 2x the area and 4x l the depth as the DGEIS - thus the general site area volume is  !

about 8x the DGEIS  !

b) Certain specific events cause soil contamination such as sewer  !

leaks. These volumes were not accounted for in the DGEIS.

c) Below building contamination is about 3x deeper than the DGEIS -  ;

volume is about 3-4 x the DGEIS ,

i

2. There is limited information on distribution of contamination with depth  !

or volume. The DOE data suggests the contwination is deeper than the DGEIS. 3 i

RESULTS OF PRELIMINARY ANALYSIS - Based on the preceding items 1-2, it would appear reasonable to modify or conduct a sensitivity analysis on the reference i facility in the DGEIS as for example as indicated in the following simplified ,

calculation of contamination using real world data:  ;

1) Assume 150,000 ft2 of contamination, and total volume is 665,000 ft3 at 30 pC/g (a level 2 analysis).

2) Resultant volumes are as follows:

l pC/g DGEIS distr Ventron distr )

130 533,000 290,000 1 8r 545,000 360,000 39 640,000 470,000 ,

30 665,000 665,000 1 20 690,000 815,000 13 700,000 1,000,000 4 770,000 1,300,000

3) Alternate approach would have 2 areas of site; one general site area of about 100,000 ft2 with Ventron distribution and one smaller area with deeper contamination. This is a level 3 analysis.

7 l

_ _ _ . _ _ - . _ . - . _ . . _ . ._ . _ _ _ . _ . . . . _ . . _ _ _ _ . . _ _ . . _ . - ~ . _ _ _ _ _ . . _ _ _ _ _ _ . _ ..

l i I

L

~

l

8. Summary of Public comments on building contamination in UFab plants  !

Specific pohlic comments were not received en the model used for the reference ,

UFab plant. Nevertheless comments were received on reactors indicating that the model for concrete contsmination was not accurate.  ;

VERIFICATION / MODIFICATION 0F Ufab plant . . -.

In response to the public comments, the staff . reviewed the reference Ufab plant for the purpose of verifying and/or modifying the reference cases.

• The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is ,

presented; (3) the basis for the verified and/or modified reference facility  :

is given. '

i L

8

I r' l .

l

! i l

l

1. SASIS FOR DGEIS REFERENCE UFAB PLANT The reference Ufab plant is described in App. C of the OGEIS. The reference case was modeled after the Wilmington facility as follows: .

Structures contamination:

General buildino contamination , __

l Areal extent - ft2 %contam Fuel manuf bldg 208,000 50 Chem lab 8,300 40 U scrap room 3,700 90 Warehouse 8,700 30 Incinerator area 2,400 100 Waste treatment area 2,500 100*

Contamination level - Unat dom /100cm2 18,000 Depth profile - contamination with depth into the concrete such that concrete must be removed as follows:

total Conc removal for: depth (in) vol (ft3) 100 mrem .15 2400 30 mrem .15 2400 l 15 mrem .15 2400 l 3 mrem .15 2400 Extended wet spot contamination -

Areas exposed to liquid contamination for extended periods of time is assumed to penetrate deeper; therefore it is assumed that the entire thickness of the floor is removed, a thickness of 6 in.

Depth profile - none estimated, i.e., it is assumed that the entire  : '

thickness of the floor is removed, approx 2400 ft3 for each dose criteria

2. Real World Data Analyzed Decontamination plans and other documents from the following facilities were reviewed to assess whether real world data provided input to verify or modify the reference power reactor:

1 Apollo -

Cimarron UNC Wood River Jct UNC Montville 9

The following is a summary of the information availcble for those facilities:

Apollo it is noted that uranium contamination of the interior of the main building occurred as a result of occasional process liquid leaks and spills. the concrete information from Apollo is complicated by the nature of the operation

~

at Apollo, specifically thit a number of overlaying pours were made to form the floor. Some floors were over 44 inci,es Liiitic and were made in 3 or 4 successive pours with each pour covering contaminated surfaces (Carlson paper, pg4) The majority of wall contamination was <30 pC/g, with some selective areas to 2000 pC/g. Floor contamination in some areas exceeded 2000 pC/g.

Specifically the decommissioning plan provided the following:

1. From Fig 2-10, the following are estimates of contaminated areas:

contam (pC/g) ft2 0-30 10,000 30-250 16,000 250-2000 5,000

>2000 8,000

2. It is noted that the buildings are deconstructed using jackhammers, etc.

The volumes of floors and walls from Tables 2-7 and 2-8 are as follows:

area thickness (ft) ft3 ,

floors: l LEU 3 34,100 ground floor .5 15,600 ,

tiEU .2 .5 5,500 i Total floors 63,500 l Valls 53,000 l i

10

k b

3. RESULTS Of USE OF REAL WORLD DATA - PRELIMINARY , .

The Apollo data is difficult to use in a comparison with the DGEIS as is .

noted, it is based on deconstructing a facility which has areas of deep  !

contamination below several poured layers. Other facilities did not have building data readily available.

de e l

n

[

l l

1 11

.- - ~ ~ - - _ . - . - . - . _ - .

, t l

i i

ATTAL.. MENT 3 ,

VERIFICATION / MODIFICATION Of REFERENCE RESEARCH REACTOR {

Specific public comments were not received on the model used for the reference l research reactor. Nevertheless, comments were received in general questioning l- the'model_of contamination in the DGEIS. Therefore, the staff has analyzed  !

' real world data to verify / modify the reference research reactor for both soil .

and building contamination. This analysis is contained on pages 3-14 as - l follows: l i

Soil contamination i Page 2 - summary of public comments on soil model ,

i Page 3 -

DGEIS reference research reactor soil contamination model '

Page 4 - analysis of real world data for soil j

Buildina contamination Page 8 - summary of public comments on building model l Page 9 DGEIS reference research reactor building contamination ,

model Page 10 - analysis of real world data for building contamination l Results and preliminary conclusions Preliminary results of analysis of real world data for soil contamination are on page 7.

Preliminary results of analysis of real world data for building contamin' tion are on page 14.

l l

l-i-

i i

i

?  !

_ _ _ _ _ u

A. Summary of Public comments on Soil contamination Soil Contamination - Specific comment were not received on research and test reactors. Nevertheless comments were received in general indicating that model for soil contamination in the GEIS is not accurate for the following reasons:

(a) soil volumes in the GEIS are too low (181,182,189, 216),

(b) the models for distribution in soil in the GEIS are too simple (189, 190, 214), areal variability can be much more pronounced (214),

containination can be deeper than indicated (214) and volumes requiring removal can increase exponentially with decreasing dose criteria at levels below about 30 mrem /yr (87,190, 214).

The physical reasons given as to why the model is not accurate include:

(a) The GEIS ignores the pathway of deposition from airborne effluent release made over the life of the plant (155, 156, 157, 190)

(b) Underground line breaks can occur between buildings, between outdoor tanks and buildings, or between tanks and discharge points; these leaks can be several feet underground (181)

(c) There can be large volumes resulting from seepage of slightly contaminated routine effluents from drain lines, from tank overflows, and from natural buildup of permitted releases over years of operation (190) 1 (d) The effect of mixing that occurs during remediation (190) j I

2

i t 4

" f

. I l-  !

VERIFICATION / MODIFICATION OF REFERENCE dFSEARCH REACTOR  !

The staff reviewed the reference research reactor for the purpose of verifying l

and/or modifying the reference cases. '

i. /

i l The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is __ '

presented; (3) the basis for the veriff ' and/cr modified reference facility ~ --

is given.

_l. BASIS FOR DGEIS REFERENCE RESEARCH REACTOR The reference research reactor is described in App. C of the DGEIS. D:0 e contamination levels is based on real world data from a 1986 PNL study of e i power reactors which indicated that contaminated soil volumes were on tne order of 10s of m3 (100s of ft3). Little depth profile information was >

available so that a simple diffusion model for spread of contamination into the soil was used. The reference research reactor was derived-from that information. Briefly the soil contamination model is as follows- ,

Site contamination:

500 ft2 I 6 "al extent -

Contamination level - Co60 Cs137 J

pCLg ' pCi/q 2-60 1 I Depth profile - contamination with depth is based on diffusion into the soil such that soil must be removed as follows:

total l Soil removal for: depth (ft) vol (ft3) 100 mrem .1 49 30 mrem .2 102 15 mrem .23 114 3 mrem .38 189 Under buildinq soil contamination Contamination is assumed to penetrate building cracks to a depth in soil of about 5 feet.

Depth profile - none estimated, i.e., it is assumed that all of this soil is removed, approx 3460 ft3 for each dose criteria 3

1 l

{

l' l

f 2. ANALYSIS OF REAL WORLD DATA _

l i There is a wide variety of research reactors in the U.S. The reference l reactor used was the 1 MWt Oregon State Triga reactor. The volumes and j contamination levels in structures and soils for this facility should be l representative and even limiting for most real world situations based on the l

following: . . _

1) NUREG/CR-1756, Table 3.1-1 indicates the fcAlowing:

power level no. of reactors

<1 57 1.5 2.

2 4 5 3 10 1

2) NUREG-14576, Table G.3-3 indicates the following:

type no. of reactors university 42 government 5 industry 7 i

Thus in reviewing real world data, cases considered as reference would in general be those of a cniversity type of puwer level less than 1 MWt. It is also of use to obtain real world data from a larger, industry type facility to observe sensitivity of variation of facility type. l Information on site contamination was obtained from the following two sources:

(1) general reports on facility decommissioning, and (2) facility l decommissioning plans.

(1) General reports on decommissioning NUREG/CR-1756, Add.1 presents information on case histories of 5 research reactors.

For one of the facilities, the Ames Laboratory Research Reactor (a 5 MW tank f type reactor), it is indicated that soil samples were taken from 65 sites around the reactor including areas inside and outside the reactor fence, and from 5 control sites. At the time NUREG/CR-1756 was issued, all of the samples had not yet been analyzed, however it was indicated that all samples ,

i analyzed to that time showed no differences between the Cs137 rehetor site samples and the control samples.

For the other 4 facilities, there is no indication that there was soil l

contamination. l t

(2) Decommissioning plans Decommissioning plans and final status reports from the following facilities 4

. _ . _ _ . . . ~ _ _ _ _ _ _ _. - - . _ _ . _ - . _ _ . _ _ . _

k 5

4 were reviewed to assess whether real world data provided input to verify or '

l modify the reference research reactor:

l University of Kansas University of Texas >

University of California l 4

Northrup TRIGA {

.l-Watertown The following is a summary of the information available i.; thox plans and ,

i t reports: I j University of Kansas .25 MWt

• The final report (pg 5) indicated that there had been no release from the 3 -facility. No information on soil contamination removal was indicated.

University of Texas .25 MWt L -

The decommissioning plan (pg 1-10) indicated that minor spills were cleaned up promptly. It was indicated that <50 ft3 of soil volume cleanup was estimated

as a contingency (Table 5-1), and that only surface contamination was expected

(12/2/85 letter pg 7).

The Final decommissioning plan indicated that one area of soil contamination

l beneath the reactor shield room from an area of pool overflow (pg 3,6) at a f concentration of Co60 of 1.2 pC/g was removed of a volume of <10 ft3. There l- was no other detectable activity. Soil samples on the site indicated no l

{ readings distinguishable from background (pg 6) i 4 University of California - 1 MWt TRIGA

! The' final plan (pg 8) indicated that there were no releases likely to cause contamination and did not indicate any soil removal activities.

i Surveys of all effluent release points showed no activity > background (pg 50); one area of a storm sewer in an alleyway showed an area of < 2 pC/g (pg

51). _ Information reported from outside areas indicated Csl37 levels ranging

, from .2 .6 pC/g (letter pg 6), which were likely due to fallout.

Northrup - 1 MWt TRIGA The final decommissioning plan indicated no environmental sampling above background (pg 2-15, Table 4.3.5, Fig 4.3.1). No removal of soil .

contamination was indicated. l l

! Watertown Arsenal - 5 MWt  !

The reactor shut down in 1970; therefore C060 and Csl37 levels could be decay j corrected.

I' 1

5

-.m- w- 7. 9 -y- -- - -.- - . -._--.---,--e m . - . - . ,,y

I-  !

i,  !,

The final plan indicated that soil samples taken indicated the following (pg j 5-5, Table 5.2): '

] avg max  :

! pC/g pC/g '

Cc60 .4 .6 l l- Csl37 .3 .5 I 4

~

l The Co60 contamination was from one particular area; subsequent sampling of i

.that area did not indicate Co60 contaminat;oi, (vg-3-0) .i

i

! Appe'ndix B provided further information.on soil contamination. Cs137 j contamination ranged,from .1 .57 pC/g.

i 4 )

3 i

i i

i i

o 1

I i .

h i

! !i i.

a l l I

i i

1

~

l l

l l 6

i

I

3. RESULTS OF USE OF REAL WORLD DATA - PREllMINARY The following general results from tne real world data available appear to be reasonable:

1. Based on review of the information from the decommissioning plans and final site reports, the level of contamination at research reactors appears to be low and localized. __

j

2. Most areas of site would appea;-is resJit in doses of <l5 mrem using I

dose conversion factors from NUREG/CR-5512.

RESULTS OF PRELIMINARY ANALYSIS - Based on tne preceding items 1-2, it would appear reasonable to retain the reference facility from the DGEIS with modification of the depth profile as suggested in Attachment 1.

i i

3 i

a l

f I

k 4

l 4

7

I l

B. Summary r F Public comments on building contamination Specific public comments were not received on the model used for the reference l research reactor. Nevertheless comments were received in general indicating i that the model for concrete contamination was not accurate for the following reasons. A commented indicated that volumes of activated and contaminated equipment and structures increece axpnaentially with decreasing dose rate below 500 mrem /yr (87). In particular commenters indicated that costs of activated concre'te, in particular for some reactor designs where significant volumes (100,000s of ft3) of concrete become slightly activated, are not included in the GEIS (183,189)

VERIFICATION / MODIFICATION OF REFERENCE RESEARCH REACTOR In response to the public comments, the staff reviewed the reference research reactor for the purpose of verifying and/or modifying the reference cases.

The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is presented; (3) the basis for the verified and/or modified reference facility is given.

I l

3 8

l

1. BASIS FOR DGEIS REFERENCE RESEARCH htACTOR The reference research reactor is described in App. C of the DGEIS. Data on ,

contamination levels is based on real world data from a 1986 PNL study (NUREG/CR-4289) of 6 power reactors. Some depth profile information was available which was combined with a simple diffusion model to estimate spread of contamination into the concrete. The reference research reactor was derived based on that information. Briefly the building contamination model is~as toil 5w~s:

Structures contamination:

General building contamination Areal extent - ft2 %contam Reactor bldg 1155 20-100 Annex b1dg 735 10-100 HX bldg 740 10-100 Pump house 805 10-100 Waste storage 570 10-100 Rad center bldg 30000 4 Total 35005 10 Contamination level -

Co60 Cs137 dom /100cm2 pC/cm3 dom /100cm2 pC/cm3 1.02E5 460 3.3E4 150 Depth profile - contamination with depth into the concrete (based on NUREG/CR-4289) such that concrete must be removed as follows:

total Conc . emoval for: depth (in) vot (ft3) 100 mrem .25 240 30 mrem .375 307 15 mrem .375 307 3 mrem .375 307 Extended wet spot contamination -

Areas exposed to liquid contamination for extended periods of time is assumed to penetrate deeper; therefore it is assumed that the entire thickness of the floor is removed, a thickness of 6 in.

Depth profile - none estimated, i.e., it is assumed that the entire thickness of the floor is removed, approx 350 ft3 for each dose criteria 9

I I

2. ANALYSIS OF REAL WORLD DATA l

There is a wide variety of research reactors in the U.S. Tne reference ,

reactor used was the 1 MWt Oregon State Triga reactor. The volumes and contami' nation levels in structures ano soils for this facility should be i representative and even limiting for most real world situations based on the l following:

1) NUREG/CR-1756, Table 3.1-1 indicates the following:

1 power level no. of reactors

<1 57 1.5 2 l 4 l 2

l 5 3 10 1

2) NUREG-1496, Table G.3-3 indicates the following:

type no. of reactors university 42 government 5 industry 7 Thus in reviewing real world data, cases considered as reference would in general be those of a university type of power level less than IMWt. It is also of use to obtain real world data from a larcjer, industry type facility to observe sensitivity of variation of facility type.

Information on building contamination was obtained from the following two sources: (1) general reports on facility decommissioning, and (2) facility decommissioning plans and final status reports.

(1) General reports on decommissioning NUREG/CR-1756, Add.1 presents information on case histories of 5 research reactors.

Information on contamination is as follows:

Diamond Ordnance Radiation Facility - .25 MWt TRIGA Contamination range - Table A.2-6 indicates the range of contamination in the facility is 4 pC/g - 120 pC/g it was indicated on page A.12 that there was activated concrete requiring removal however no volume was given.

It was indicated on page A.14 that the tota' radwaste volume (including concrete, wood, aluminum, steel, etc.) was 1470 f t3.

Ames Laboratory Research Reactor - 5 MWt tank type 10

l l

It was indicated on page A.25 that demolition of the biological shield was the most expensive activity. C060 levels in the concrete in the shield were given in Fig A.2.ll and A.2.12 as lE5 pC/g and IE4 pC/g. .

I It was indicated on page A.28 that primary coolant leaks had resulted in ~

j tritium contamination of the concrete in the reactor room. As a result, it  ;

was decided that unrestricted use of the reactor room was not immediately 1

attainable. Its further use as a controlled laboratory area was being

! considered. l l

Lynchburo Pool Reactor - ISw pool-type It was indicated that this facility may be released for controlled use. l l .

It was indicated on page A.43 that the total waste volume was 570 ft3; and  :

f that this was more than anticipated due to the amount of concrete that had to j

be removed from the pool area.

l North Carolina State Reactor 3 .01 MWt pool type It was indicated on page A.52 + hat the greatest amount of activation products are found in the concrete under the reactor core. No volumes or activities were given.

(2) Decommissioning plans Decommissioning plans and final status reports from the following facilities were reviewed to assess whether real world data provided input to verify or modify the reference research reactor: l University of Kansas University of Texas University of California Northrup TRIGA Watertown In general, the decommissioning plans noted that criteria of Reg Guide 1.86 and 5 uR/hr were used.

The following is a summary of the information available in those plans and reports:

University of Kansas .25 MWt General building areas:

The decommissioning plan indicated (pg 3-1, SER pg 2) that there were no spills within the facility; that there was no contamination except for the biological shield (pg 3-1, final plan, pg 5, SER pg 2,3) .

Activated materials:

11 l

l

It was indicated that activated concrete was removed from the facility; the volume of concrete was 120 ft3 (final plan pg 14,15)

University of Texas .25 MWt l

General building areas. -

~

The decommissioning plan (pg 1-10) indicated that minor spills were promptly cleaned up and major contamination did not exist. The 12/2/85 letter (pg 5),

the final status report (pg 1), and the 3/22/93 letter (pg 3, 5) indicated that areas of the facility outside the pool meet RG 1.86, except for a few localized spots, such as the radwaste storage area, hood, glove box, some l components. Routine surveys indicate levels <200 dpm/100 cm2. Only a few l sources are >200 dpm/100 cm2 and no smears indicate levels >5000 dpm/100 cm2 i outside pool area. l l

Activated materials: 1 The major decommissioning activities (Table 1.3) included removal of the entire thickness of concrete from the activated portion of the reactor pit floor. The neutron activation analysis is contained in the decommissioning i plan (pg 1-13, Table 1-5). The volume of activated concrete requiring removal from the reactor pit wall and floor was estimated in the decommissioning plan, pg 5-3, Table 5.1) to be 280 ft3. The final status report indicated (pg 1) that 88 ft3 of reactor core structural components were removed and 250 ft3 i consisting of neutron activated parts of the aluminum pool 'iner, concrete shield structure, and miscellaaeous parts. The 3/22/93 letter indicated that removal of activated concrete was to a depth of 12 inches. the total volume of the aluminum liner nr.d the biological shield wall was 520 ft3.

University of California - 1 MWt TRIGA

. General building areas:

The final status report (pg 5) indicated that there were no accidents, occurrences, or releases, and that there was minor surface contamination (pg 6). The two letters (1/31/89) indicated that building areas were clean. No concrete removal was indicated in the report.

i Activated materials:

Activation of concrete resulted in removal of the pool wall (pg 27, 28). It was indicated that removal of .5 .8 in from 5 surfaces would be necessary resulting in 2140 ft3 of concrete.

Northrup - 1 MWt TRIGA General building areas:

The ER indicated that there were not uncontrolled releases or contamination of areas due spills. General building sources requiring concrete removal were not found (i. il plan pg 2-29). The ER (Table XIV) indicated measurements 12

j 1

i l

l range from 1-170 dpm/100 cm2, with most areas < 1 dpm/100 cm2. No removal of concrete other than activated was indicated. No contamination in hot cell found (pg 2-29). Rad chemistry lab contamination was minor and was on hood, i benches, storage wells, drain pipes - this was decontaminated or packaged and  ;

shipped.

Neutron activated materials: .

The final plan (pg 2-27, Table 4.4.8) indicated the following neutron -

l activation of the exposure room:

depth Co60  !

in. pC/g 0-3 796 17 2.92 I 20 2.74 24 2.5 29 .36

Decommissioning tasks (final plan pg 2-26) included removal of 24 in of exposure room activated concrete for a mass of 130 MT. This reduccd dose to 10 uR/hr. To meet 5 uR/hr, wall and ceiling surfaces of exposure room l activated concrete were removed (final plan pg 2-29) for an additional mass of l 4 MT. The total volume of activated concrete removed was indicated (Table 1 5.8.1) as 1624 ft3 from the exposure room (93%) and beam ports and walls (7%). I Watertown Arsenal - 5 MWt  !

l

The reactor shut down in 1970; therefore Co60 and Cs137 levels could be decay corrected.

l The decommissioning plan (pg 1-21, and final plan pg 3-1) indicated that there were leaks early in the operations which were eliminated by design fixes.

These resulted in some concrete contamination in the reactor area. The

. decommissioning plan also indicated that there was an unplanned leak from the cistern (pg 1-22).

The following contamination information was given (final plan Table 5-1):

area dpm/100 cm2 basement general <200 smear 34 290 operating floor <200 reactor annulus floor 200-700 cistern <200 The estimated total volume of concrete removed is 6750 ft3. (final plan) 13

_ _ _ __ ~ _ _ - .

\

l I

3. RESULTS OF USE OF REAL WORLD DATA - PRELIMINARY The following general results from the real world data available appear to be reasonable:

1. Based on review of concentration levels, the levels used in the DGEIS appear to be consistent with the levels experienced. l

2. Based on review of waste volumes, the volumes used in the DGEIS appear -

to be consistent with the volumes experienced; however the volumes ~ot activated concrete do not appear to be accounted for in the DGEIS. l PRELIMINARY ANALYSIS PROPOSAL - Based on the preceding items 1-2, it would  !

appear reasonable to retain the reference facility from the DGEIS. It should be noted that the volumes of activated concrete indicated in the decommissioning plans were assumed in the reference analysis to be removed as a part of the reactor removal procedure and do not affect the analysis in the DGEIS.

1 14

' s d

ATTACHMENT a ,

VERIFICATION / MODIFICATION OF REFERENCE SEALED SOURCE MANUF/ BROAD R&D f

Only one public comment was received on the model used for the reference sealed source manufacturer (SSM) and none on Broad R&D. Nevertheless,

! comments were received in general questioning the model of contamination in the DGEIS. Therefore, the staff is analyzing real world data to verify / modify the reference SSM/Rau tor 60tn soil and building contamination. This analysis

- ~

is contained on pages 3-14 as_ follows:

Soil contamination Page 2 - summary of public comments on soil model Page 3 -

DGEIS reference SSM/R&D soil contamination model Page 4 - analysis of real world data for soil Building contamination Page 7 - summary of public comments on building model ,

Page 8 DGEIS reference SSM/R&D building contamination model  !

Page 9 - analysis of real world data for building contamination Results and preliminary conclusions Preliminary results of analysis of real world data for soil contamination are on page 6.

Preliminary results of analysis of real world data for building l contamination are on page 11.

l I

i l

I l

e 4

1

); ,

I 4  !

A. Summary of Public comments on Soil contamination Soil Contamination - Only the following specific comment was received on the ,

l j reference SSM; none were received on the Broad R&O:

In addition, it is noted that the selection of reference facilities does not include typical examples of radiochemical and radiopharmaceutical manufacturing. The impact of decommissioning on these facilities to the levbf of the proposed standard would be significant, yet was not j

considered as part of the. impact analys,is (from CORAR). ,

1 Other comments were received in general, although not on SSM and R&D indicating that model for soil contamination in the GEIS is not accurate for the following reasons:

j (a) soil volumes in the GEIS are too low

(b) the models for distribution in soil in the GEIS are too simple, areal variability can be much more pennnunced, contamination can be deeper than indicated and volumes requiring ramoval can increase exponentially l

with decreasing dose criteria at leveis below about 30 mrem /yr l

The physical reasons given as to why the model is not accurate include: 1 1 (a) The GEIS ignores the pathway of deposition from airborne effluent i

release made over the life of the plant  ;

(b) Underground line breaks can occur between buildings, between outdoor i i

tanks and buildings, or between tanks and discharge points; these leaks

! can be several- feet underground i (c) There can be large volumes resulting from seepage of slightly 4

contaminated routine effluents from drain lines, from tank overflows, and from natural buildup of permitted releases over years of operation

l 4

4 VERIFICATION / MODIFICATION OF REFERENCE SEALED SOURCE MANUFACTURER The staff reviewed the reference SSM/R&D for the purpose of verifying and/or modifying the reference cases. ,

The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) .The.real world data considered is presented; (3) the basis for the verified and/or modified reference facility

" - 'i s given.

1. BASIS FOR DGEIS REFERENCE SSM/R&D The reference SSM/R&D is described in App. C of the DGEIS. Data on '

contamination levels is based on limited real world information from two facilities regarding areal extent (f t2) of contamination. Contamination levels (pci/g) are estimated from those two facilities and also derived from a ,

1986 PNL study of 6 power reactors. Little depth profile information was available so that a simple diffusion model for spread of contamination into the soil was used. Briefly the soil contamination model is as follows:

Site contamination:

5000 ft2 Areal extent -

Contamination level - Co60 Csl37 gCIg pCi/q 2-60 1-20 i Depth profile - contamination with depth is based on diffusion into the soil such that soil must be removed as follows:

total Soil removal for: depth (ft) vol (ft3) 100 mrem .1 500 30 mrem .2 920 15 mrem .23 1170 3 mrem .38 .1400 Under building soil contamination Contamination is assumed to penetrate building cracks to a depth in soil l

of about 5 feet.

Depth profile - none estimated, i.c , it is assumed that all of this soil is removed, approx 600 ft3 for each dose criteria 4

3

i I

l 1

2. ANALYSIS OF REAL WORLD DATA There is a wide variety of sealed source manufacturers and broad R&D facilities in the U.S. Based on Table 4-2 of the Regulatory Analysis (attached), there are about:

a) 295 sealed source manufacterarc Licensed by NRC, of which 142 are byproduct manufacturing and distribution. ,

b) 11.00 broad R&D of which over 800 are academic, medical, private _

I organization laboratory type facilities. Academic and medical broad l

licenses are issued to educational and medical institutions for the possession and use of radionuclides for teaching, training, and research.

Thus, for broad R&D facilities, in reviewing real world data, cases considered j as reference would in general be those of university / medical / organization l laboratory type facilities noted above. A generic single building was used as l the reference broad R&D facility. Larger facilities could be scaled up.

For the sealed source manufacturer, based on the above, the areal dimensions of a facility (AMS) which manufactures sealed sources was used for the reference SSM.

The staff attempted to obtain information on site contamination from the i l

reports on facility decommissioning, either general or specific. l Information from reports on facility decommissioning NUREG/CR-1754 and NUREG/CR-1754, Addendum 1, present information on decommissioning at non-fuel-cycle facilities. The information presented is for the decommissioning of the various components and floors and walls of  !

laboratory facilities. l 4

i These documents present a review of decommissioning experience at laboratory facilities including that at New England Nuclear Corporation, Mound Laboratory, and Lawrence Livermore Laboratory. Significant soil contamination r was not indicated. Information on contamination at SSM/R&D facilities I indicate that it is commonly found in localized areas such as in fume hoods, glove boxes, hot cells, workbenches, sinks and drains, ductwork, and to some extent, room surfaces. Past experience indicates that these facilities can generally be readily decontaminated to le/els comparable to Reg Guide 1.86.

Limited information is generally available for these facilities. They are not generally required to submit decommissioning plans or characterizations of rite contamination prior to cleanup. NUREG-1444 (the Site Decommissioning Management Plan) was reviewed with regard to possible sites with more extensive soil contamination. Four SSM/ Broad R&D facilities are in NUREG-1444. These include AMS, Budd Co., Permagrain, and Safety Light. The following information is available in NUREG-1444 for this group:

4

i

- i l

5 1

TABLE 4-2 ESTIMATEDNUMBERS PROGIW.i CODE OF NL~ (NL AGREEMENTS i C.=,wsise sw Lkwe NU ~dhkd hd*ar of Me*r E NRC PengNa NBC Powywe Code Dacrteden Ca sser*ee' NRC LAs====

Code

$ssa/ Tesal tid 1(1 H

M 38 8800 ACAMMaC TYPE A 42 me ACADEMIC TYPE 8 46 40 f 19 _

ACADEMIC TYPE C 230 337 11 2 101 13 tile MEDICAL 9effff. ta0AD 9 4

18G0 VET peon 4fUMAJs D3 Md _.

j los 139 l 34:0 es.Vrino TTritas 343 j the SAD SA0ADTYPE A M 17 M10

18 Mit SAD 880AD TYPE B 9

19 3_ l 9

Mit SAD BAQAD TYPE C 4

3 MI3 RAD MULTHffEAtEGION litt 16M l Sit SAD OTHER 4 34 2 *se tjern e hr sui 0 e

182M MILJTARY SOURCE 64ATER.tAL D(DOOR O l I Nm I w by Stats.

MtLJT AAY SOURCE MATERIAL OUTDOOR 3 4I 221 1 2

• 21133 DECOMMti$10ND8G 21340 4 i

'lle URAN 1UM FVE3. RAD 11 3

j 8 NO $T ATE CATEGORY CtJTICAL M A15 M ATIALAL 16 21310 le 3 %0 ST ATE CATECORY 2t T33 C M M UC. CN thvN. M 38

ho IT ATE CATECORY 22tto UNSEALID plt /TONfUM < 208 M 38 32 No gT ATE C ATEGORY 22:18 UN5EALAD U D5 c 330s. U.D3 ( 200g IIT 774 g  :

81 NO IT ATE CATEGORY i 1

1 221]D $NM PLiff0NfUM < Elog l ZW l Mil ties l l l

f f TOTAL RAD F ACult1E3 83 l f, 1 NO ST ATE CATECORY l 4113) l LAAE EAATM EXIILACT10h M 38 1211 MANUf ACJDLTTRS A 12 is j(  ;

1 9

nis MANUFACsotsTun a  ;

6 .

h 3

3:11 MANUF ACOtSTRS C D8 h 1 78 h

ist MANLT . OtTT M ANUF AC/DisTRS OTMEA l$

12's

$ TOTAL 10 l NUCLEAR t.AUNDAY 4 f

• h 1218

• NO STATE CAff' GOR Y 3219 DECONT AMINAft0N 5EAY',CU NO TT ATE CATEGogy 39 51 _f 18 5OURCE M ATY111AL < t% 1 88 III 11300 se bO ff ATE CATEGORY SOURCE M ATDjAL > 110ka ID I E)

^ 18300

$1 P ACEM AK.fjt $kM MED IN$T 4 6 22s40 t j LNDUST RIAL 22t62 P ACtha AKE8 MANUF AC.0tSTKl8 eM l '10 j 3 2e) l l l l TOT AL 5EALED SOURCE M ANUf ie)

AC F ActLtilu l f Yrs' k 64 3 l {

f CRAND l TOTAL

• • .# w v su J ras wa

• ik u,im 0 5 ewr' sa ==== === * - =e ==w a==a == *e e

• 6@!.31 tb(& a ! IC 1

8'.e de b.Eruwe 'l 4 IIh del Ihr maydw t of % f ( t.nzrese 1 al 4 6#"E'8 IN 4+5kui enWissWS &#4 tuhand en 42s stue of als emanhet

, . - .en s : n:: m3 w i ., , .4 a u s ,

l T w, i . e. . .

v.

I '

t

AMS l

AMS previously_ manufactured sources for distribution. Contamination was found in various building-areas and ductworl, and some contamination was found in ~

4 sediments, soil and vegetation in the Nothern portion of the AMS. property.

NUREG-1444 indicates that contaminated material at AMS consist of equipment j

.and concrete contaminated with Co60 as well as C060 in sludge in sewer piping. ,

i' i Budd Budd manufactured sealed Ir-192 and Co-60 sources in a hot cell facility. The i i

! principal reason for Budd's inclusion on the_ SDMP was contamination in the. hot -

cell. It is indicated in NUREG-1444 that activity is largely confined to the building. -

I Old Vic  ;

j This facility produced electronic tubes containing Ra-226 and Ni63. NUREG- l 1 1444 indicates that these nuclides are found on building structures such as walls and floors.

permagrain l

i

! The facility originally housed a research reactor and hot cells; use was

, subsequently made of the pool as an underwater irradiator. It is indicated in l i NUREG-1444 that contamination is in inactive portions of the building and  !

f equipment. There is not an indication of soil contamination.

Safety Light This facility originally manufactured materials containing Ra-226 (not regulated by AEC/NRC) and other nuclides including H-3, Sr90, Am241, and Csl37. NUREG-1444 indicates that there are a number of contaminated areas on the site with average / peak concentrations .n soil.for Csl37 of 20/630 pCi/g and Sr?0 of 3.5/15.4. However characterization is not complete and volumes are not estimated.

1 i

t 5

j

l l

I l 3. RESULTS OF USE OF REAL WORLD DATA - PRELIMINARY The following ge,neral results from the real world data available appear to be reasonable:

i i 1. Based on review of the nature of facilities involved, i.e., laboratory facilities using a variety of equipment, most contamination should be confined to localized areas for mu3c of the facilities, i.e., the l reference facility.

2. Based on the limited information on soil contamina~ tion, th'e areas and volumes of contamination used in the DGEIS (5000 ft2 and 1100 ft3) does not appear to be an unreasonable estimate for the majority of facilities. However use of a wider contamination area and a deeper soil penetration based on the analysis in Attachment I would suggest a greater volume by a factor of 2-3.

3. As noted above, limited information is available. More extensive checking through the NUDOCS/michrofiche system could be done (with concommittment use of staff resources), however for the majority of facilities of this type it is considered that the reference facilities are not unreasonable.

RESULTS OF PRELIMINARY ANALYSIS - Based on the preceding items 1-2, it would appear reasonable to retain the reference facility from the DGEIS, however a sensitivity analysis could be performed using a wider area and deeper soil penetration model suggested in Attachment 1.

6

i

8. Summary of Public comments on building dontamination in Reference SSM/R&D ,

.One' general public comment was received on the model used for the reference i SSM as noted above; no comments were received on the reference broad R&D.

1 Other comments were received on other facility types on concrete 1 contamination.

t

).

j VERIFICATION / MODIFICATION OF REFERENCE SSM/R&D-

. In response to the public comments, the staff reviewed the reference SSM/R&D for the purpose of verifying and/or modifying the reference cases.

l j The discussion below describes this review as follows: (1) The basis for the

DGEIS reference facility is summarized; (2) The real world data considered is i l

~

presented; (3) the basis for the verified and/or modified reference facility is given.

?

L

,1 h

4 4

i i

l 7

l 1

4 f )

i l'. BASIS FOR OGEIS REFERENCE SSM/R&D

' The reference case The reference SSM/R&D is described in App. C of the DGEIS. l

. is modeled in the following manner:  ;

~ \

! -1. .The cost of decontamination / removal / disposal of fume hoods, hot cells,  ;

3 benches, sink and drain lines, ductwork, and other components are NOT l 1 included in the DGEIS analysis of costs. Only the analysis of removal .;

i of building structures is included based on the scope of the rulemakirg ' --

)

l- .and-the modeling in NUREG/CR-5512. Because.information from NUREG/CR- {

d 1754 indicates the' large traction of the cost of decommissioning of a l

,SSM/R&D facility is related to disposal of the above components, this '

may be part of the reason why. the analysis and costs in the DGEIS may

appear. lower than the commenter noted in Section A.1 feels is reasonable.  ;

j 2. Data on contamination levels is based on limited real world data a::

! noted above. Some depth profile information for concrete penetration at reactor facilities was available which was combined with a simple diffusion model to estimate spread of contamination into the concrete.

Briefly, the building contamination model is as follows: i Structures contamination:

General building contamination Areal extent - ft2 %contam l l

Floors 6000 10 Walls 4600 5 l Contamination level -

Co60 Cs137 dom /100cm2 dom /100cm2 1.02E5 3.3E4 l Depth profile - contamination with depth into the concrete (based on NUREG/CR-4289) such that concrete must be removed as follows:

total .

f Conc removal for: depth (in) vol (ft3) 100 mrem .25 70 30 mrem .375 80 15 mrem .375 80 3 mrem .375 100 Extended wet spot contamination -

Areas exposed to liquid contamination for extended periods of time is assumed to penetrate deeper; therefore it is assumed that the entire thickness of the floor is removed, a thickness of 6 in.

Depth profile - none estimated, i.e., it is assumed that the entire thickness of the floor is removed, approx 70 ft3 for each dose criteria 8

O

2. ANALYSIS OF REAL WORLD DATA There is a wide variety of sealed source manufacturers and broad R&D facilities in the U.S. Based on Table 4-2 of the Regulatory Analysis (attached), there are about:

a) 295 sealed source manufacturers licensed by Nl ' which 142 are . -

byproduct manufacturing and distribution. __

b) 1100 broad R&D of which over 800 are academic, medical, private organization laboratory type facilities. Academic and medical broad

, licenses are issued to educational and medical institutions for the possession and use of radionuclides for teaching, training, and research.

Thus, for broad R&D facilities, in reviewing real world data, cases considered as reference would in general be those of university / medical / organization laboratory type facilities noted above. A genetic single building was used as the reference broad R&D facility. Larger f acilities could be scaled up.

For the sealed source manufacturer, based on the above, the areal dimensions of a facility which manufactures sealed sources was used for the reference SSM.

The staff attempted to obtain information on site contamination from the reports on facility decommissioning, either general or specific.

General reports on facility decommissioning NUREG/CR-1754 and NUREG/CR-1754, Addendum 1, present information on decommissioning at non-fuel-cycle facilitias. The information presented is for the decommissioning of the various companents and floors and walls of laboratory facilities.

These documents present a review of decommissioning experience at laboratory facilities including that at New England Nuclear Corporation, Mound Laboratory, and Lawrence Livermore Laboratory. Information available in the above documents notes that for NEN the major building areas of contamination were floors underneath lab benches and cabinets, areas not accessible for routine cleaning. Contamination levels on floors and walls from NEN indicated that levels ranged from lE3 to SE4 dpm/100 cm2 for Cs137. Floors may be concrete or wood covered with tiling surfaces thus limited contamination in concrete to cracks or unlined areas. Information from Mound laboratory included information that in one lab there was 439 linear feet of glove boxes, 236 of which were contaminated.

Other sources of information on contamination at SSM/R&D facilities indicates that contamination is commonly found in localized areas such as in fume hoods, glove boxes, hot cells, workbenches, inks and drains, ductwork, and to some extent, room surfaces. Past experience indicates that these facilities can generally be readily decontaminated to levels comparable to Reg Guide 1.86.

9 i

_ . _ _ _ . _ __ _ - . _ _ _ _ ._. _ _ . _ _ . .___m... _ _ . _ _ _ _

s t

l 1

Limited information is generally available for these facilities. They are not -

4 generally required to submit decommissioning plans or characterizations of site contamination prior to cleanup. NUREG-1444 (the Site Decommissioning

Management Plan) was reviewed with regard to possible sites with more extensive building contamination. Four SSM/ Broad R&D facilities are in NUREG-1444. These include AMS, Budd Co., Permagrain, and Safety Light. The following information is available in NUREG-1444 for this group

AMS 3

AMS previously manufactured sources for distribution. Contamination was found in various building areas and ductwork. An ORISE survey indicated contamination up to IE6 dpm/100 cm2. NUREG-1444 indicates that contaminated

material at AMS consist of equipment and concrete contaminated with Co60 but volumes are not indicated.

l Budd i

Budd manufactured sealed Ir-192and Co-60 sourras in a hot cell facility. The j principal reason for Budd's inclusion on the SDMP was contamination in the hot cell. It is indicated in NUREG-1444 that at the time of shutdown that quantity of Co60 in the hot cell was less than 5 Ci; at the time of remediation it was about 1 Ci. Decommissioning produced about 1200 ft3 or radioactive waste but it is not indicated if this were components or concrete.

! Old Vic This facility produced electronic tubes containing Ra-226 and Ni63. NUREG-j.

1444 indicates that these nuclides are found on building structures such as 4 walls and floors.  !

i Permagrain The facility originally housed a research reactor and hot cells; use was subsequently made of the pool as an underwater irradiator. It is indicated in NUREG-1444 that contamination is in inactive portions of the building and  ;

equipment.

Safety Light This f acility originally manufactured materials containing Ra-226 (not regulated by AEC/NRC) and other nuclides including H-3, Sr90, Am241, and Csl37. NUREG-1444 indicates that there are a number of contaminated areas on the site. However characterization is not complete and volumes are not estimated.

10

t j

t l

3. RESULTS OF USE OF REAL WORLD DATA - IRELIMINARY j The following general results from the real world data available appear to be reasonable:

l

1. Based on review of the nature of facilities involved, i.e., laboratory facilities using a variety of equipment, most contamination should be l

confined to localized areas for most of the facilities, i.e., the i referance facility. - -- [

l 2. Based on the limited 'information on building contamination, the areas and volumes of contamination used in the DGEIS (6000 ft2 and 10%

contamination) does not appear to be an unreasonable estimate for the majority of facilities. However use of a larger % of area (50-100) contaminated may be reasonable for a sensitivity analysis based on

~

limited information for SDMP and other-sites.

3. As noted above limited information is available. More extensive checking through the NUDOCS/michrofiche system could be done (with l concommittment use of staff resources), however for the majority of i facilities'of this type it is considered that the reference facilities are not unreasonable.

RESULTS OF PRELIMINARY ANALYSIS - Based on the preceding items 1-2, it would  !

appear reasonable to retain the reference facility from the DGEIS, however a sensitivity analysis could be performed using a larger building contaminated area.

l l

l l

l l

11

l l

1

'e ATTACHMENT 5 ,

VERIFICATION / MODIFICATION OF URANIUM MILL Public comments have questioned the model used for the reference uranium mill.

These comments are summarized on page 2.

~

In re'sponse the staff has analyzed real world data to verify / modify the reference umill for both soil and building contamination. This andlysis is contained on pages 3-20 as follows:

Soil contamination Page 2 - summary of public comments on soil contamination Page 3 - DGEIS reference umill soil contamination model Page 4 - analysis of real world data for soil Buildino contamination Page 8 - summary of public comments on building contamination Page 9 - DGEIS reference umill building contamination model Page 9 - analysis of real world data for building contamination Results and preliminary conclusions Preliminary results of analysis' of real world data for soil contamination are on page 7 Preliminary results of analysis of real world data for building contamination are on page 11 l

l l

1

{

l l

l

L .

1

.f,

. (

I 4

i 1  !

L ,

l A. Summary of Public comments on Soil contamination in Reference Umill l l

i 3 Soil Contamination - Commenters indicated that the model for soil i contamination in the GEIS is not accurate for the following_ reasons: l

-i

) (a) the volume of.cnntaminated soil in the GEIS was too low

~ ~

(b) the model for ' distribution of contamination in soif was too simple and

' that.therefore contamination would be spread over significant vglumes of ,

soil; thus the volumes requiring removal would increase rapidly (even

j. geometrically) with decreasing dose criteria.  ;

}

The physical reasons given as to why the models are not accurate are as 1

follows- /

l l (a) Use of a 15 cm depth is only appropriate for farmland where the source of contamination is dust or spreading of contaminated topsoil. Under l realistic conditions of mixing and burial, contamination will likely  !

i extend beyond 15 cm

\

. (b) The GEIS model results in primarily surface contamination. In real  :

i situations, contamination is more a function of site waste management i practices, e.g. . burials, deposition, soil mixing throinh previous site i j activities, leaks in underground systems, etc.

l f

l

} (c) the vertical variability in the GEIS is too simple which leads to

' contamination depths in the GEIS of < 15 cm even though experience l indicates depths of contamination considerably deeper l

! l 1

1 l

I J  !

1 a

i 4

i

2 I I

i

1

VERiflCATION/ MODIFICATION OF REFERENCE UMILL 2

In response to the public comments, the staff reviewed the reference umill for the purpose of verifying and/or modifying the reference cases.

The discussion below describes this review as follows: (1) The basis for the i OGEIS reference facility is summarized; (2) The real world data considered is., .

presented

(3) the basis for the verified and/or modified reference facility is given. - --

j 1. BASIS FOR DGEIS REFERENCE UMILL 4

The reference umill is described in App. C of the DGEIS. The reference case j was modeled after the model mill in NUREG-0706 in the following manner:

1. Contamination levels are based on limited real world data at uranium facilities 9

t 2. The areal extent of contamination for the reference case is 20 acres a which is approximately the area of the umill facility and the ore pad area.

3. Because little depth profile information was available, a simple i diffusion model for spread of contamination into the soil was used.

Briefly the soil contamination model is as follows:

Site contamination:

Areal extent - 880,000 ft2 (20 acres)

Contamination level - U RCLq 30-1000 Depth profile - contamination with depth is based on diffusion into the soil such that soil must be removed as follows:

total Soil removal for: depth (ft) vol (ft3) 100 mrem .6 531,100 30 mrem .7 611,960 15 mrem .75 663,900 3 mrem .84 736,100 Under building soil contamination Contamination is assumed to penetrate building cracks to a depth in soil of about 5 feet.

Depth profile - none estimated, i.e., i'. is assumed that all of this soil is removed, approx 10000 ft3 for each dose criteria 3

f

2. ANALYSIS OF REAL WORLD DATA .

< Information on site contamination was obtained principally from facility decommissioning plans and other general reports on decommissioning of uranium facilities. These are discussed below.

(a) Evteat af rantamination (1) Decommissioning /reclamatiin pians Limited information is available from specific facil'i ties. Decommissioning plans and other documentation from the follov.ing facilities were reviewed to  :

I assess whether real world data provided input to verify or modify the reference umill:

Homestake Mining Company - Grants Corp.

Atlas Minerals - Moab Mill l The following is a summary of the information available for those facilities:

Grants The reclamation plan provides e:timates of total volumes of soil contamination to be removed. It is indicated that contaminated soils will be removed from he restricted area and all adjacent areas as needed to reduce soil radium

, levels to not more than background + 5 pCi/g RA-226. This soil will excavated

using scrapers which will place the soil in the tailings pile.

It is indicated in the plan that the soil survey conducted in 1990 has been used to estimate the volume of contaminated soil to be cleaned up during reclamation, and that it is anticipated that excavation of soil will be necessary as follows:

Area Depth Volume location ft2 acres ft ft3 Vicinity of 712,000 16.5 275 1.78E6 Ore pad Remainder of 9.8E6 226 0.5 4.9E6 Site No information on distribution of contamination with volume is indicated.

Moab Mill The decommissioning plan indicates that it includes the excavation and transport of contaminated soils to the tailings pile for disposal. The primary area of concern is the general mill area to the east and north of the tailings pile. A second area is located to the southeast of the pile.

The volumes are estimated in the decommissioning plan to range from the following 4

. . - . . , _ ~ . . - . - , - . - _ - - .

. . - . - . - _ ~ . - . - _ _

_ . . . . . . . - . . . - . . ~ . . _

r i

, amounts: ,

t 4

Volume i i Location ft3 I Within mill' l.9E6 to 4E6 area 1 Area SE of. 0.675E6

" Pile

' No information on areas, depths, or_ distribution of contamination wit 6 volume is indicated; Also information is not given as to the basis.for the-j 8

radiological criteria on which the above volumes are based.

(2)'Other Reports on Decommissioning -

00E-EIA-0592 was published by DOE in February 1995. This document, entitled i~

" Decommissioning of U.S. Uranium Production Facilities," presents information on the approach of decommissioning at uranium mills and a summary of decommissioning costs for a number of mills (Table 3 of DOE-EIA-0595). -

Information on volumes of contamination, specific contamination levels, or .

specific cost components 1is not given, however Table 3 provides a comparison j of the range of costs. From Table 3, the average mill dismantling cost (including soil disposal) is:

Cost Facility $K Mills Average 912 Range 500-2500 -

l Grants 1650  ;

I NUREG-0706 presents information on the decommissioning of a model mill. l NUREG-0706 estimates that site soil contamination would be such that it would j require that soil be removed as follows:

Area Depth Volume location ft2 acres ~ft ft3 Vicinity of 864,000 20 3 2.6E6 Ore pad Remainder of 13E6 300 0.5 6.5E6 Site (b) DATA ON DISTRIBUTION OF CONTAMINATION Data on volumes and distribution of contamination is important because it is useful in determining the amount of material (and cost and impact) which  ;

results from achieving alternate dose criteria. ]

5 i

l- .

l i

Generally this data is sparse which is why simple diffusion models were used  ;

in the DGEIS. In the review of data, there was limited data on distribution i of contamination in the decommissioning plans.

• I r The following discussion-is taken from that presented in Attachment 2.

Documentation concerning distribution profiles is contained in a letter fr~n --

i 00F tp FPA on 2/24/94 which contained draft graphs of cleanup criteria versus  !

waste volume for 3 DOE uranium facilities, El;:a Gate, Ventron, and Aliquippa  ;

Forge. These plots are incl.uded here as Figures 1, 2,.3. The ratios of the  !

volumes in Figures 1-3 are shown here: i pC/g DGEIS Ventron Aliquippa Elza 130 1 1 1 1 80 1.02 1.23 2.3 1.03  :'

39 1.15 1.6 4 1.04 30 1.2 2.27 1.25 20 1.25 2.76 10 2.44 r 13 1.26 3.38 14 4.4 4 1.39 4.5 As can be seen, the data, while tending to vary and being fairly limited,  !

suggest that the contamination goes deeper into the ground than diffusion model predicts. The DOE provides a discussion of the assumptions regarding the Elza site. The Aliquippa site contains lower volumes. Thus use of the i Ventron as representing the real cases is illustrated below in Section 3.

By way of comparison, Commenter #63 submitted a figure which was stated as .

using actual site data indicating that the soil volume' multiplier versus pC/g i ratio from 100/15 mrem was approximacely 3-4. The data supporting the figure was not submitted. Also Commenter #96 submitted a figure based on Cushing information which indicated that the ratio was approximately 6.s The concentration d .ta supporting the figure was not g ven.

b 1 6

i 5

d

)

3. RESULTS OF USE 6C REAL WOR 80 DATA - PRELIMINARY The following general results from the real world data available appear to be reasonable: '

1. The amount of data on soil contamination is limited. The area and volume estimates for the Grants Mill described above are indicated as being based on a 1990 survey, however they ara camn=r=ble to the model

, ~~

mill in NUREG-0706 (prepared in 1980). Nevertheless, the total volume of contamination for the cases reviewed suggests that contamination can ,

be larger than the DGEIS case, especially if the total site is )

considered.

2. There is liniited infor. nation on distribution of contamination with depth or volume. The DOE data suggests the contamination is deeper than the j DGEIS.

RESULTS OF PRELIMINARY ANALYSIS - Based on the preceding items 1-2, it wound appear reasonable to modify or conduct a sensitivity analysis on the reference facility in the DGEIS as for example as inaicated in the following simplified calculation of contamination using real world data-l

1) For the 20 acres located near the facility and the ore pad: I I

pC/g GEIS Grants GEIS sensitivity ("

130 528,000 1.78E6 1.78E6 l' 20 660,000 NA(" SE6

2) For the remainder of the site GEIS sensitivity ("

pC/9 GEIS Grants 130 NA 4.9E6 4.9E6 ,

20 NA NA 6.lE6  !

3) Total site i

pC/g GEIS sensitivity"'

130 4.8E6 20 6.lE6 Notes (1) Based on a combination of the Grants volume information and the DOE distribution profile l (2) NA - not calculated (3) Based on use of the Grants volume infr<mation and the GEIS distribution profile in this region where leaks, etc, shouldn't be driving contamination.

4 (4) Based on Items #1 and #3, ano co:recting Grants to the average mill cost 7

. . - . _ _ . _ . _ . _ . _ _ . _ . . _ . . _ _ _ _ . . . . _ . _ . _ . _ _ m _ . _ . . _ _ _ _ _ _ _ . . . . _ _ _ _ _ . . , _ _ _ _ . _ .

l

. I i

8. Summary of Public comments'on buildino contamination in Umi]_ls  ;

i Specific public comments were not received on the model used for the reference Umill. Nevertheless comments were received on reactors indicating that the ,

model for concrete contamination was not accurate. ' '

"~

VERIFICATION / MODIFICATION OF Umill In response to the public comments, the staff reviewed the reference Umill for j the purpose 'ot verifying and/or modifying the reference cases. l The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is presented; (3) the basis for the verified and/or modified reference facility is given.

i l

i  !

I 8

i l _ -, - -- , i

_ _ . _ . _ ~ .- ._ _ _ _ _ . . _ _ _ . _ _ _ _

l l

l l

1. BASIS FOR DGEIS REFERENCE UMILL l l

The reference Umill is described jn App, C of the DGEIS. The reference case I was modeled after the model mill described in NUREG-0706 as follows: j Structures contamination:

General building contamination Areal extent - '

ft2 %contam Floors 100,00'u ' 100 Walls 130,000 100 Contamination level - U dom /100cm2 1.lE6 Depth profile - contamination with depth into the concrete such that concrete can be scabbled and removed as follows:

total Conc removal for: depth (in) vol (ft3) 100 mrem .15 1000 30 mrem .15 1000 15 crem .15 1000 3 mrem .15 1000 Extended wet spot contamination -

Areas exposed to liquid contamination for extended periods of time is assumed to penetrate deeper; therefore it is assumed that the entire thickness of the floor is removed, a thickness of 6 in.

Depth profile - none estimated, i.e.,'it is assumed that the entire l thickness of the floor is removed, approx 1000 ft3 for each dose  !

criteria j

2. Real World Data Analyzed ,

Limited information is available for the facilities described above. The plans indicate that mill buildings will be demolished and the mill process and miscellaneous structures will be buried inplace or in on-site pits, or placed

< in the large tailings pile. As such, estimates of scabbled concrete to reach

! certain dose criteria are not generally provided.

l I 3. RESULTS OF USE OF REAL WORLD DATA - PREllMINARY l l Data does not appear to be available to alter at this time the estimates in t

the DGEIS. ,

1 9

. . . . - - . . - . - - . . .. . . . - . . . . - . . - - . _ _ . - - _ . _ . _ . - ~

e-e e

." l f>

i

, ATTACliMENT 6  ;

l

, VERIFICATION / MODIFICATION OF RARE METAL FACILITY  !

l \

! - Public comments have questioned the model used for the reference rare metal

! facility. These comments are summarized on page 2.

! l In response the staff has analyzed real world data to verify / modify the i i reference rare metal facility for both soil and building contamination. This i

! analysis is contained on pages 3-20 as follows:

Soil contamination Page 2 - summary of public comments on soil contamination Page 3 - DGEIS reference rare metal facility soil contamination model Page 4 - analysis of real world data for soil Building contamination ,

Page 8 - summary of public comments on building contamination Page 9 - DGEIS reference rare metal facility building contamination model Page 9 - analysis of real world data for building contamination Results and preliminary conclusions Preliminary result: of analysis of real world data for soil l contamination are on page 7 l Preliminary results of analysis of real world data for building contamination are on page 11 I

I i

1

A.. Summary of Public comments on Soil contamination Soil Contamination - One' comment was received which indicated that the soil removal volumes and costs were underestimated, and that there would be a large

- increase wi'th soil removal volumes at low dose criteria. Other commenters, commenting on other facilities, indicated that the model for soil contamination in the GEIS is not accurate for the following reasons:

(a) the volume of contaminated soil in the GEIS was too low (b) the model for distribution of contamination in soil was too simple and that therefore contamination would be spread over significant volumes of soil; thus the volumes requiring removal would increase rapidly (even geometrically) with decreasing dose criteria The physical reasons given as to why the models are not accurate are as follows:

(a) Use of a 15 cm depth is only appropriate for farmland where the source of contamination is dust or spreading of contaminated topsoil. Under realistic conditions of mixing .and burial, contamination will likely extend beyond 15 cm (b) The GEIS model results in primarily surface contamination. In real situations, contamination is more a function of site waste management practices, e.g., burials, deposition, soil mixing through previous site activities, leaks in underground systems, etc.

(c) the vertical variability in the GEIS is too simple which leads to contamination depths in the GEIS of < 15 cm even though expeiience indicates depths of contamination considerably deeper 2

________.7

(

2 VERIFICATION / MODIFICATION OF REFERENCE RARE METAL FACILITY In response'to the public comments, the staff reviewed the reference rare metal facility for the purpose of verifying and/or modifying the reference l

. cases.

The discussion belon dem-i bc 2 this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is presented; (3) the basis for the verified and/or modified reference facility  ;

is given.

1. BASIS FOR DGEIS REFERENCE RARE METAL FACILITY ,

The reference rare metal facility is described in App. C of the DGEIS. The reference case was modeled in the following manner:

1. Slag / tailings pile volumes are NOT included in the DGEIS analysis. Only contamination of soil separate from the pile is included in the .

analysis. This is because it is assumed that this slag has fairly high i levels of contamination and would have.to be either:

a) removed in its entirety - in this case, the soil analysis

considers removal to reach unrestricted use, or; i

b) left onsite in a stabilized mode - in this case, the soil analysis is not a factor because the license would not necessarily be -

terminated putting this outside the scope of this rulemaking .

i

! 2. Contamination levels in soil are based on limited real world data are a rare metal f acility which indicated that contamination levels had ranged from 100 to 200 pC/g over most of the site.

i

( 3. The areai extent of contamination in soil was t;ased on engineering l

i judgement based on the size of the site and the type of operations.

4. Because little information was available on profile of contamination into the soil, a simple diffusion model for spread of contamination into the soil was used.

Briefly the soil contamination model is as follows:

Site contamination:

Areal extent - 100,000 ft2 l

Contamination level - Th-232 1 pCL9 30-200

Depth profile - contamination with depth is based on diffusion of thorium into the soil such that soil must be removed as noted as follows (note that this is only soil contamination; the volume of the tailings 3

pile is not included): .

total Soil removal for: depth (ft) vol (ft3) 100 mrem .03 2600 30 mrem .04 4000 15 mrem .05 5000 3 mrem .07 7000

Under building soil contamination

- Contamination is assumed to penetrate building cracks to a depth in soil of about 5 feet.

Depth profile - none estimated, i.e., it is assumed that all of this soil is removed, approx 14600 ft3 for each dose criteria 4

4

R. ANALYSIS OF REAL WORLD DATA There are about 10 facilities of.this type. Most of these facilities are listed in NUREG-1444, "The Site Decommissioning Management-Plan (SDMP)." The facilities listed on the SDMP are considered to have specific contamination / decontamination problems which require that they be addressed in a site-specific manner. One of the primary reasons that these facilities are placed on the SDMP are the large tailings piles and volumes of material ~ ~~

present at these sites.

~

Thus in reviewing real world data, cases' considered as reference would in general be those contained.on the SDMP.

Information on site contamination was thus o' otained principally from NUREG-1444 and from facility decommissioning plans and other general reports on a decommissioning of rare metal facilities. The:e are discussed below. l l

(a) Extent of slaa contamination Table 1 presents the information from NUREG-1444 on slag information.

(b) Extent of soil contamination  ;

1 The descriptions in NUREG-1444 principally present slag pile information. l Little information is presented regarding the extent of contamination in soil. l This is likely.because of the following:

a)'the principal problem at these sites is the high volume of slag contamination 4 b) several of the discussions indicate tSat the slag retains the thorium.

Specifically, the following is noted:

Cabot - the thorium / uranium is contained in insoluble slag l Dow - in 1978 a leaching study of the slag material indicated that even under aggressive conditions the waste would leach at very low rates S' alloy - a leachability test in 1991 indicated that diffusive leach of thorium from the rock-like slag material is insignificant l

Whitaker leaching studies conducted by ORISE indicate that under conditions encountered in nature that the slags will not leach to any significant degree

'Thus soil contamination at these sites beyond the slag piles is likely due to specific' site activities unrelated to the slag piles. The following limited information is available in NUREG-1444 regarding soil contamination at these sites:

Cabot-Readinq Areas of contamination were found in numerous isolated spots outside the processing building at levels of 13-51 pC/g for Th; 13-78 pC/g for uranium 5

e i

1 F

Fansteel-The only~cantamination outside the settling ponds is Unat and thorium in low concentration's in the soil. Volumes are not indicated. ,

l Heritage ,

The site contains approximately 20,000 ft3 of monazite rich sands at levils of i

4000 pC/g of thorium .

Molycorp - Wasshington. \

Thorium bearing slag was used as fill over portions of the site. Thus there l is thorium spread in low concentrations over most of the site with a range of  :

100-200 pC/g and. ranging between 10 - 2600 pC/g. This procedure of using slag as. fill is not seen at'the other rare metal facilities. .j Molycorp - York i

There are low levels of thorium in soil throughout the site. In 1987 soil ',

containing up to 700 pC/g was sent offsite. Another 4000 ft3 os soil containing up to 70 pC/g was being excavated in 1993. It is not indicated if  ;

there is additional contamination.

Shieldalloy Newfield Soils around.the piles and at numerous locations around the main yard of the j site and foundry building are contaminated. Average soil levels are 28 pC/g 4 of Th232 and 10 pC/g U238. Soil contaminants appear to be limited to the 1 upper 1 - 2 feet of soil. The soil contamination may derive from the baghouse dust piles (c) DATA ON DISTRIBUTION OF CONTAMINATION IN S0JL Data on volumes and distribution of contamination is important because it is useful in determining the amount of material (and cost and impact) which results from achieving alternate dose criteria.

Generally this data is sparse which is why simple diffusion models were used

-in the DGEIS. As noted above, there is limited data on distribution of contamination in the decommissioning plans.

6 i

~

l

- J l

i l

l TABLE 1 -

J SLAG CONTAMINATION AT RARE METAL FACILITIES l Volume pC/g pC/g

• Facility ft3 ft2 ava range' location Cabot-Boyertown 190,000 slag in conc vaults Cabot-Reading 20,000

~ 11ag on embankment-I l Dow - Bay City 1,080,000 188 2-7000 slag piles  ;

Dow - Midland 320,000 29 low-2000 slag piles Fansteel 400,000 solid residue in settling ponds; Heritage 20,000 4000 monazite rich ~ sand' Mag-Elektron 1,600,000 settling ponds Molycorp-Wash 3,200,000 5-1000 slag-used as soil fill Molycorp-York 5,000,70-700 soil Shieldalloy-Cambr

- west pile 6,000,000 350,000 1.4-42 slag / soil

- east pile 1,200,000 112,000 4 slag Shieldalloy-Newf l

l standard pila 600,000 516 slag pile

- high pile 35,000 366 slag pile

- dust pile 530,000 55 lime dust Whittaker 1,000,000 0-7000 slag pile i

i I

(

i 7

\- .

1

3. RESULTS OF USE OF REAL WORLD DATA - PREL'MINARY ,

The following general results from the real-world data available appear to be

. reasonable:

1. there is large volume contamination at these facilities but the data i available appears to indicate that it is principally in the slag piles.

j As noted above in Section A.1, this is not included in this analysis.

l

2. Limited information on soil contaminacidn'does not suggest volumes of i I

soil similar to the volumes of Attachment 2, however the volumes may be larger than that in the DGEIS based on the size of the yard area and the 1 depths noted. )

l l

i l

RESULTS OF PRELIMINARY ANALYSIS - Based on the preceding items 1-2, it would j appear reasonable to conduct a sensitivity analysis on the reference facility i in-the DGEIS as for example as indicated in the following simplified calculation of contamination using real world data:

1) 100,000 ft2 of contamination based on yard areas. I

2) total volume of 50,000 ft3 at depths of approximately 0.5 - 1 foot depth. .

b e

O 8

I B. Summary of Public comments on building contamination Specific public comments were not received on the model used for the reference '

rare metal facilities. Nevertheless the reference facility model was  !

reconsidered based on comments received on other facilities. j VERIFICATION / MODIFICATION OF RARE METAL FACILITY In response to the public comments, the staff reviewed the reference rare ~ .

metal facility for the purpose of verifying and/or modifying the reference t cases. l

~!

The discussion below describes this review as follows: (1) The basis for the DGEIS reference facility is summarized; (2) The real world data considered is presented; (3) the basis for the verified and/or modified reference facility '

is given.

i l

i ,

1 l

l

(

l t

i 4

l J

l-I i

9 j

I e

i t

1. BASIS FOR DGEIS REFERENCE RARE METAL FACILITY ,

The reference rare metal facility is described in App. C of the DGEIS. The '

reference case was modeled.after the structures at the Holycorp facility as follows-

• Structures contamination: ._ j General building coritamination f Areal extent - ft2 %contam I Floors 150,000 40  :

Walls 180,000 10 j Contamination level - Th232  :

dom /100cm2' '!

18,000 i Depth profile - contamination with depth into the concrete such that l concrete must be removed as follows:  !

total  !

Conc removal for: depth (in) vol (ft3) '

100 mrem .15 1500 30 mrem .15 1500 i 15 mrem .15 1500 1 3 mrem .15 1500 I 1

Extended wet spot contamination - l Areas exposed to liquid contamination'for extended periods o# time is assumed to penetrate deeper; therefore it is assumed that the entire thickness of the floor is removed, a thickness of 6 in.

. Depth profile - none estimated, i.e., it is assumed that tne entire thickness of the floor is removed, approx 1500 ft3 for each dose criteria I

I

?

l I

10 l l L _ _

i

4

2. Real World Data Analyzed NUREG-1444 was reviewed to assess whether real world data provided input to .

4 verify or modify the reference rare metal facility

4. The following is a summary of the information available for those facilities:

Fansteel A single process building and liquid waste treatment facility are contaminated with small concentrations of natural uranium"aha thorium, however most of the contamination is found in settling ponds. The levels in the building were an 4

average of 100 pC/g (800 mrem /yr).

Heritaqe i

> Exposure rates in the dry mill building were about 13 pC/g (100 mrem /yr); in the area of the dry mill feed about 70 pC/g (600 mrem /yr): in the area of the i dry mill tailings discharge about 60 pC/g (500 mrem /yr).

Molycoro-Washington

. Radioactivity levels in building 34 has #ixed alpha contamination up to 92

dpm/100 cm2 and direct radiation levels up to 40 pC/g (300 mrem /yr) 1 11

l t

i

'e l

l 1

3. RESULTS OF USE OF REAL WORLD DATA - PRELIMINARY l l

Limited information is available for the facilities described above. ,Most of l the concern and information at these facilities relates to the slag piles. 1 The data available does not alter at this time the estimates in the DGEIS. l 1

6 l

l l

l 1

l l

I l l l \

l I l 1

l i

i I

i a

e 12  !

, <