ML20154F876
| ML20154F876 | |
| Person / Time | |
|---|---|
| Issue date: | 09/07/1988 |
| From: | Gnugnoli G NRC OFFICE OF NUCLEAR MATERIAL SAFETY & SAFEGUARDS (NMSS) |
| To: | Feraday M INTERNATIONAL ATOMIC ENERGY AGENCY |
| References | |
| REF-WM-3 NUDOCS 8809200163 | |
| Download: ML20154F876 (7) | |
Text
.-
4 i GG 8/22 8EP o 7 g M. A. Feraday Department of Nuclear Energy and Safety Division of Nuclear Fuel Cycle International Atomic Energy Agency Wagramerstrasse 5 P. O. Box 100 A-1400 Vienna, Austria Ref: 642-T2-TC-668
Dear Mr. Feraday:
In response to your letter of July 14, 1988, I am providing suggested text revisions to the report: "The Design of Impoundment and Disposal Facilities for Uranium Mill Tailings and the Rehabilitation of Tailings Piles." The first enclosure consists of responses to 3our questions in Annex A of your letter.
The second enclosure consists of an annotated copy of the above report. Should you have any questions about the enclosures, please telephone after 1600 during any work day.
The overall organization and approach outlined in your letter is acceptable.
Specifically, your intent to shorten the report is a necessary step to bring tie guidance into a managecble framework. My annotations of the text will address the questions in your cover letter more specifically. However, I am listing a few recommendations, which I hope will be helpful:
o Sections 10 and 11 are quite long. This is one area where referencing other IAEA sources may be more appropriate.
Specific deletions will be made in the text annotations
- enclosed, o The revision I sent to you is necessary as a companion to the new write-u) in Appendix B-3 on radon; it, or a reasonable facsimile siculd )e retained in the text at the end of Section 5.
If you need to delete or reduce text in this area, I would reconinend deleting or reducing 5.3.2.2 and Appendices B-1 and B-2, since these are dated approaches, as far as the U. S. is concerned.
Some guidance in radon attenuation should be included in the report.
o I agree with your recouraendations on shortening Sections 10.2, 11.2, and 11.3 along with the referencing.
I think that Appendix A has its uses and doesn't take up much room. However, sirplification of the diffusion mechanism of radon in soil can lead to some drastic underestimations of the actual release of radon into the environment. The U. S. experience has resulted in a departure from the use of graphical representations and nonnalized flux conversion f actors.
Perhaps the most flexible
@ 92 g G G80907 p,/
WM-3 pg I
31WL A UL
l
- 1. -
GG 8/22
-2 path is to reproduce a radon attenuat'on computer code, as listed in USHRC Regulatory Guide 3.64 (I left a copy with you).
o With regard to the French approach, perhaps it would be better to wait to see whether it comes up at the Technical Comittee Meeting.
l I wish you the best of luck with this report, and do not hesitate to contact me should you need any assistance in bringing the report revision to a successful conclusion.
Sincerely, l
S Giorgio N. Gnugnoli Division of low-level Waste Management l
and Decomissioning Office of Nuclear fiaterial Safety and Safeguards
Enclosures:
- 1. Responses to Annex A
- 2. Annotated Report TRS-209 cc: R. Dale Smith, URF0 w/ enc 1.
/
V 0FG :L U NAME:
1 /dfw :
l DATE:
/7/88 l
OFFICIAL RECORD COPY 1
GG 8/22 ENCLOSURE ENCLOSURE 1 RESPONSES TO ANNEX A 1.
This sounds like one of Roy's contributions.
I do not have this portion of the text included in the copy I brought back with me. However, an NRC researchreportpublishedin1985(NUREG/CR-3906UraniumHillTailings Neutralization: ContaminantComplexationandTailingsLeachingStudies) indicated a significant reduction of Th-230 in solution when pH values were altered to 7.2 with addition of calcium hydroxide. However, it should also be noted that carbonate complexation (resulting fron neutralization strategies) and increased pH values can lead to remobilized concentrations of other constituents (Co, Mo) and radionuclides (U). The experience in Canada might be dependent on initial application of lime to result in precipitation, and capture of the precipitates in the deeper reduced environments of the lakes.
You are right; as written the last sentence of the paragraph sounds "iffy."
I have no problem with deletion.
2.
The full title is:
"Sumary of the Waste Management Prograns at Uranium Recotery Facilities as They Relate to the 40 CFR Part 192 Standards.
November 1985." NUREG/CR-4403.
I would steer away from value judgements on which constituent was the most "serious."
I suggest dropping the asterisks.
By the way, this is one of the documents I lef t with you.
I hadn't intended going much deeper into this subject at this point of the docuinent; especially if you want to cut down on the length of the report.
3.
The emanation rates corresponding to radon releases of 2.6 E+14 to 1.1 E+15 Bq/a are 16.5 to 69.7 Bq/sq. m./sec. by my calculations. This appears to be a mistake in the original TkS-209. Assuming that 2 E-7 s/cu
- m. is a "representative" effective dispersion factor (NRC calculated one uf 2.06 E-6 s/cu. m. for the model mill in NUREG-0706), then the 2.6 E+14 to 1.1E+15 Bq/a release rates would correspond to a downwind concentration (at 2 km) of 1.65 to 6.98 Bq/cu. m.
It appears that there were either some calculational mistakes, or else some other complexity was invoked without informing the reader.
As far as the 37 Bq/cu. m. natural background radon concentration goes, this seems to be consistent with the generally accepted values of 1 pCi/L estimate used here in the U.S.
Refer to NCRP Report No. 77 Exposures From the Uranium Series With Emphasis on Radon ar.d its Daughters,(3.7 Bq/cu. m) issued March 15, 1984 This report gives values of 100 pC1/cu. m.
to 150 pCi/cu. m. (5.6 Bq/cu. u); even up to 1000 pCi/cu. m (37 Bq/cu.
m.).
The estimate of 37 Bq/cu. m. is on the high end of the average range, but not unrealistic. Also, see page I.J of reference 19A of the report for more examples of background flux values.
1
i O
l-1 GG 8/22 ENCLOSURE 4.
[27]USEnvironmentalProtectionAgency.
Final Environmental Impact Statement for Remedial Action Standards for Inactive Uranium Processing Sites (40 CFR Part 192). EPA 520/4-82-013-1.
October 1982. Section 3.4.
[23]USDepartmentofEnergy.
Project Schedule and Cost Estimate Report. UMTRA-00E/AL-400127.0166. March 1988.
l 5.
I think what should be said is that heterogeneity, which is typical in subsoil conditions, greatly complicates the modeling of the novement of contaminants. Homogeneity, if anything, would allow simpler modeling strategies. However,1 do agree that the transition is not very clear.
6.
Sorry, but you might check with Roy John.
l l
7.
Permeability would tend to increase as the subsurface becomes saturated; it is also considered to be a constant in the zone of saturation.
But in this case, the unsaturated zone itself is becoming saturated, the l
statement may be referring to the formerly unsaturated zone after the decrease or loss of hydraulic head. More likely the use of the word "may" is a result of unwillingness to admit to an absolute statement.
8.
Third and fourth paragraphs. The point of this discussion is that a low-permeability liner is not necessarily the ultimate goal.
In cases, where precipitation levels are not effectively balanced by liner in conjunction with a higher evapotranspiration,alow-permeabilit{ bathtub"phenomenon.
permeability cover can result in the As you point out, this caused problems at LLW repositories.
It is up to the national authority to balance the concerns of ground-water protection and those of prolonged drying / dewatering operations (and the continued release of radon and other effluents) f rco a tailings impoundrent, when the decision is made to select ona disposal strateg) over another.
In the U. S. the EPA's Clean Air Act may lead to the use of smaller impoundments or phased or continuous disposal during the operational life of the facility.
The last two paragraphs are not irrelevant.
9.
In this case I would use "ray" instead of "will."
What if the ground-water quality is putrid to begin with?
- 10. The more gentle the slopes, the more resistant a cover will be to wind and water erosion. That is an engineering fact. Perhaps the sentence could be i
qualified by adding:
"..whenever possible or practicable...." Low is on the order of 1-3% slopes. See page 68 (top).
l 11.
It does sound funny, but that is the term used to describe radon uptake l
through the plant from the root system and released to the atmosphere.
It 1
c GG 8/22 ENCLOSURE is distinguished from radon release through the channels left by either shrinking or desiccated plant root systems.
Strange as it sounds, the proper term is radon exhalation.
- 12. NUREG-0706, Volume II, Appendix B, Section 3.3 discusses nonconventional leachates. Most of the list is referred to in this section on p. B-13.
The NUREG lists two references:
l E. Landa, "Isolation of Uranium Mill Tailings and their Component Radionuclides From the Biosphm--Soine Earth Science Perspectives."
Geological Survey Circular 814, 1980.
F. M. Scheitlin and W. D. Bond, "Removal of Hazardous Radionuclides from Uranium Ore and/or Mill Tailings:
Progress Report for the l
Period October 1, 1978, to September 30, 1979," ORNL/TM-7065.
I will photocopy the page from NUREG-0706 for your use, and place it following p. 48 of the annotated report.
- 13. OK.
14.
I grant you that the sentence somewhat belies the extent of differences.
Other differences include the French comm1trent to active post-clusure maintenance in place of an intentional long-term isolation design.
I am
(
not sure of the point you are trying to rake.
I suppose that the l
restriction of access may be one of many differences, but I am not sure l
that I would classify it as the primary one. Someone from IAEA would be a better judge of this.
- 15. This one is confusing. You refer to [55] in Annex A, but page 55 of the reportlists[45].
I presume [55] is a typo and that you are referring to
[45].
In any case, USNRC Regulatory Guide 3.11 is the correct reference.
Guidance for erbankment design, as well as discouragement of using tailings for the embankments are discussed.
I do not know what the contentofreference[51]is.
[ figure 12, someone began the process of converting to metric, but didn't 53]isthecorrectreferenceforfigure12.
However, if you nctice in 16.
l fini:h.
- 17. The high pyrite content could result in enough generation of sulfuric acid that the temperature would be increased. This, in turn, could induce layer mixing in the lake. This is not desirable.
- 18. Unfortunately, the USNRC technical position is still unsublished.
However, the following NUREG contractor leport states tie saine conclusion:
i l
GG 8/22 ENCLOSURE l US Nuclear Regulatory Comission.
"Design Considerations for Long-Term Stabilization of Uranium Mill Tailings Impoundments,"
NUREG/CR-3397. October,1983.
19.
I agree; it should say preferential paths.
20.
I don't know the Holder reference.
i
- 21. One of N problems with an underdrainage system, or any drainage system at uran u mill tailings ponds, is keeping the slimes away from the else clogging will occur. Since the discharge strategy system, en concentrates the slime fraction with the standing water, this would present operational difficulties with clogging of the drainage system with the fines. Placing the drains within coarser tailings fraction would reduce clogging, but would be counterproductive. By the way, you had this coment from Annex A referring to sage 78, as opposed to page 79; Roy might be having some trouble with t11s too.
j
- 22. Since few facilities are presently operating, it would be difficult to say that this is "widely" used. A number of the mills use drum filters, but 1 have only seen belt-filtration proposals for one mill, which was never built. The reference al!udes to residual tailings moistures dcwn to 18 to 23%, where belt-filtration was used.
23.
R. John's response.
j
- 24. There is nothing very exotic about this, and this is why there doesn't seem to be a lot said. However, the idea is to consider using settlement and displacement plate monitors and erosion markers to "observe" the long-term progress of the construction.
Other devices that can be used include inclinometers, which can monitor the stresses placed on the slopes. The best bet is to design an encapsulation for the tailings, which to the greatest degrte "fits" into the surrounding environment.
If existing slopes do not exceed 3%, it would be inadvisable to design slopes of 20% without some protective l
strategy; e.g., diversion ditches, rock-armorad sicpes, etc....
The kinds of remedial actions would include fi ig in gullies, regrading covers, filling more cover and rock in fissures and depressions created by settlement, and so on.
Strategies for repair and closure design should be considered within the national authority's philosophy in waste disposal.
One country that relies on institutional controls and recoverability of the waste does not )ursue the same approach of the country that puts little stock in eitier.
I don't think you need to put much else in this cection until you get feedback from the technical comittee participants in November.
GG 8/22 DISTRIBUTION:.
central file - LL)#4 412.1:s' 104SS rf LLOB rf GGnugnoli, LLOB HFliegel, LLOB PLohaus, LLOB JSurmeier, LLTB HBell, LLRB JGreeves, LLWH MKnapp, LLk11 DSollenberger, LLOB KDragonette, LLRB Riahy,GPAilP RHauber, GPA/IP
'l EHawkins, URF0 HPettengill, URF0
ENO2,CSURE INTERNATIONAL ATOMIC ENERGY AG6NCY AGi t'CE INTERNATION ALE DE L'ENERGIE t T0klQUE ME*.lyH APollHOE ATEHTCTB0 Il0 ATOMHOR 3HF4H ORGANISMO INTERNACIONAL DE ENERGIA ATOMICA WACRAMIRSTR AS$t 3. P O. SOX IM A 1400 VIENNA. At!$TRIA TELIX.1 12H3, CAalt. IN ATOM VIENN A. F AC31 MILE. #' 222 230184 TELIPHONE i222123%
'eais'a'sNaN![s'a'[$Ns as<a co.poisa seasEtIIt@M[NYa'Ns$
1988-07-14
Dear Giorgio,
First let se thank you for your hard work and extensive input into the revision of T15-209 (Curren l'ractices and Options for Confinesent of Uraniun Mill Tailings) during the consultants' seating (16-20 May). Enclosed are a revised versios of the draft report and Annex A a list of questions.
The reviaicas I have made are, to a large extent, editorial rather than technical. I added short sections on "Mining and Milling of Uranica Ores" Section 3 and."Types of Wastes" Section 4.1.
In Section 4, the major factors which affect the management and disposal of mill tailings have been gathered together in three main subsections 2 types of wastes; characteristics of tailings which affect disposall and basic principles for the management of radioactive and toxic wastes.
Section 5 deals with the release of pollutants and their transport to san including taportant release mechanisas; pathways to humans; and facters which control the releases.
Section 6 covers means of reducing environmental impact by mill process selection.
In Section 7, which covers tapoundment siting and design options. I included part of the old Section 9.4 as a new section 7.2.2 on types of covers. The selection of a combination of these cover materials to sake an engineered cap for a particular type of f acility is covered in Section 9.6 (Selection of Engineered Covers). I thought the individual covers should so into Section 7.2 along with the other components of a total facility, i.e.
impoundment type, liner types, cover types and management.
Mr.G.Gauhli Operatiens 3 ranch Division of Low Level Vaste & Decommissioning US Nuclear Regulatory Commission Washington, DC i
20555 USA 7
ago2noodr ut107T4
. -- "'I C R WASTE wn 'J PDC
_. v-f
' ection 8 on treatment of liquid effluents and monitoring has not changed Section 10 is now quite long. Also I added some further data in Section 11 to make the generic safety assessment ans modelling analyses consistent with other IAEA waste management documents on these topics. Now Sections 'O and 11 are quite long and I may have to cut these during the next revision.
I would like your opinion on that, especially what to cut out if we do some cutting.
I have added a few figures in various places as well as a new Section 5.3.3 (see page 43) received from G. Causneli by sail after the meeting.
'iould that go better as an introductiun to Appendix B-37 i
The report is getting quite long now and I would like to shorten it by 10 to 20 pages if possible. Please give re your consents on the following as possible places to cut or suggest other placess (1) Shorten 10.2 and refer to the code of practice.
(2) Mocten 11.2 and 11.3 and refer to the Safety Series document.
(3) Im Appendix A of such value?
(4) Shorten Appendix B-3 and refer to the reports. I have a bit of a probles trying to decide if all the equations in the text relating to radon ini Appendix B-3 and 5.3.2.2 are of value in this type of summary report or are graphical representations auch as the attached figure and table along staplified analytical 3]thods better. I would guess that the readers for this report would be 50% free developing countries.
One other approach that we could use would be to remove all or parts of Sections 10 and 11 and nake them into a new report with Annex D.
In the report we don't give auch prominence to the French approach to tailings handling, i.e. controlled seepage, use of tailings for outer esbankments, etc. Do you think we should say more or wait and see if it comes up at the TCMP In reviewing the report. I would appreciate it if you would tick eff in red those references that you can verify as being the correct ones.
I would appreciate receiving your coseents as soon as possible and no later than 15 September so I can incorporate them into the revised version to be sent to the participants of the TCM which meets on ?-11 November.
Thanks again for your as.1 stance. Best regards.
Yours sincerely.
+, $.2]
M.A. Feraday Division of Nuclear Fuel Cycle Enes.
p Annex A Cuestions on the text (1) Page 13 Reference 10 required.
The last statenant ssess to nake the hypothesis pretty iffy?
(2) Page 14A -
Details of NUREG-CR-4403 required (reference 12).
Check to see if nost serious pollutants are marked with a *.
(3) Page 21 Onpage20,theesinationratesfromthepilewere1.5to74 Sq.a". s *1 At 2 km the atmospheric concentration is 1.5 to 74 3q.a-3, is that correct? I don't understand how it was calculated.
The natural concentration given seems very high according to Table 2.3 (attached) 'res reference 19A.
(4) Page 23 Reference 27 and 28 required.
(5) Page 30 Last paragraph (new) in Section 5.3.1.1.
The first sentence says that the description implies heterogeneity in the soil. I don't understand where it says this. I think it implies homogeneity.
(6) Section Do you have any idea which reference the equations came 5.3.1.1 froe?
(7) Page 31 Third paragraph, second sentence. If the unsaturated abne becomes saturated the pe' mability will increase not say.
Is that correct?
)
l (8) Page 32 Third paragraph. It say be desirable to have a low seepage l
rate but I can't see that it is desirable to have a porous l
cap. In the second sentence, the cover is described as "stabill ed*, is this not after resedial actions after closure? If so, then the last two paragraphe are irrelevant.
I thought the biggest probles with a stab 111:ed cover having a higher permeability than the liner would be that the l
topoundsent would eventually fill up with water and the pathway to man would be direct. This is the type of probles that occurzed with LLW repositories.
(9) Page 33 End of first paragraph. See change.
(10) Page 35 Last sentence in 5.3.2.1.
That la not true in all cases.
Should this sentence be deleted? How low is low?
(11) Page 41 Iten 1.
Increased erhalation?
(12) Page 47 Are refsrences available for some or all of suggestions 1 to 97 (13) Page A8 Last paragraph 6.2.1.
Deleted. Plocation of non-radioactive pollutants discussed in 6.2.2.
s
" (14) Page 50 Line 19.
Is not the prise national difference between France and North America. The French allow controlled seepage whereas in North America the objective is to, minimize seepage?
(15) Page 55 Reference $$ (27 in original text) is RC-3.11.
I den't think that is correct. Is reference 51 a suitable reitrence?
(16) Page 64 Is reference 53 (Scarano) correct for Figure 12.
In TRS-209 it shows Scarano (reference 39) in the text and reference 38 on Figure 77 (17) Page 66 Bottoa. If free orygen is not available why should there he conce rns ? Or are we saying that free oxygen will in fact be present?
(18) Pa.
68 Top. Reference required. Is there any case except in arid areas where this is true, i.e. a slope of 1 to 3%?
(19) Page 70 Mid page. Preferentisi paths??
l l
(20) Page 73 Mid page. Reference required - Holder and reference 66 (Golder).
(21) Page 78 Bottoa. Why is little underdrainage used with this sys,tes?
(22) Page 83 Mid page. Is belt filtration still receiving attention? Is it noisture reduction of 18 to 23% or to 18 to 23 ?
(23) Page 84 Insert at top.
R. John please elaborate on the note you I
left with me (attached). Are you saying that 5 to 10 s of I
water over the ta111 ass is required to prevent pyrite oxidation?
(24) Page 91 Section 9.3 is pretty thin. Any suggested improvements such ass what should the slope usually bei kinds of remedial actions, etc.
A l
.r. Wm e
t 1.4 1.2 Om ORY TAILINGS
~
iO M
.G
)a 0.8
=
NN y
0.6 MOIST TAILINGS N'
i O.4 a
WET TAILINGS 1
0.2 O
I I
0 5
10 15
=-
OEPTH (f t) rigure 2.2-1 Graphic Representation of Radon Emanation from Wet, Moist, and Dry Tailings as a function of Ra-226 Concentration and Depth of Tailings (From reference 111 etAlL~ T m f1S'l Radon Emission Per Cent Moisture Content Frca Tallings of Cover (pci'/m2 s) 6 8
10 12 3
100 1.7 1.3 1.0 0.7 Soo 3.4 2.6 2.0 1.5 1000 4.1 3.2 2.4 1.8 Table 2.2-1 Estimated Cover Thickness (metres) to Reduce Radon Emission Prem gncovered Tailings Piles to 20 ;<i'/m *s M)
No e n - /9 2.
- t pCi = 3 7 a Sq
p
' 'I O i E, p (.3y 2.%
sect 7.2.: 4 /u t-saw Tk4 cdhd A AWtdd
$ o) x 5f,.p o scu%
z (P khudde. (le 0 5 Q,%3 t)
V r
So f
%M o
'antir IJ a ue (q raz q.
(om1 %
ochidwata"
/
\\
{0n US D
M 1
l-1m, e
l
I
-y, 26 T ABLE : 3 5e.ettes %taccr Averste 86 cast:wns Rac:n Ocn:enirst.ere n A,r Csncenirsi.cn 3
t e. m p c.. t.
La<st sn w.. e i,,1 O ct 3 : cts
$s tn Ps e 3 37 3 :C2 Ins;sn ocena 3 37 3 002 Seatn Pacar.
32 3 OC4 North At:nntic 2
3 358 Japan 3
0 ct0 G errn any 3
QW London 9
0.173 Rwas,a 3
0 30 France I
U ? ? te:
No. Yots C.ty 6
3 170 a
3 210 New Yor Rural 3
3 240 New Monice 18 2 40) 0%e W est een U.na.um M 11 e n 1.t etI i
Monticeito. Utah 13 3 343 j
I 56 4 Lane C.sy. L'tstt 1.
3 360 19 3 313 D ursa ge, Coloride Grand Junction. Colorsde 31 3 840 l
4 0 100
$lerttem Mee steere Estimate IRepeeted La V $ Congrese,1643 bhoster and $dl,194%
tRepersed.a Nations ceanea en R.di.iie. Prote< tion and usuoremenie uts are provided in Appendbt D. Typically,in conventional mills, %
95 percent cf the uranium a extracted, but nearly all of the other radionuclides articularly the 32*Th and 33*Ra. are included ir.
the tulings. Chapter 4 diwunes process alternativ s that may make it possible to remove these long. lived auchdes ecenomicr!!r in the milling process.
o O
The Design of impoundment and Disposal Facilities for Uranius Will Tallings and the Rehabilitation of Tailings piles Table of Contents 1.
Introduction 2.
S cie 3.
Mining and milling uranium cres 4.
Factors in the management of uranium mining /aniling westes 4.1 Types of westos 4.1.1 Mining westes 4.1.2 milling westes 4.2 Major characteristics of tailings which effect disposal technology 4.2.1 Uranius ore grade 4.2.2 Radioactivity 4.2.3 Acid generation potential 4.2.4 Non-radiological conta-inants 4.2.5 Chemical precipitates 4.3 Basic principles for tailings management 5.
Release af pollutents from uranium mill tailings to humans 5.1 Important release mechanisms 5.1.1 Spille during transport of tailings to the Lagoundment 5.1.2 Erosion 5.1.3 Radon amenation end transport 5.1.4 Wind transport of particles 5.1.5 Structural failure of tailings embankments 3.1.6 Unauthorised removal and use of ta!!ings for building and fill f.1.7 Centre 11ed releases of sentaminated water 5.1.8 Uncontrolled water releases 5.2 General pathways to humans 5.2.1 Atmospheric pathways 5.2.2 Atmospheric and terrestrial pathways 3.2.3 Aquatic pathways 5.3 Facters teatro111ag release of pollutsats 5.3.1 Sub-surface transport of waterhorne pollutants 3.3.1.1 metardaties by soil ten enchange 5.3.1.2 Asepage sentret 5.3.1.3 Leseking eestrel 5.3.2 Airkerne transport of pollutants 5.3.2.1 Centrolling wind transport of particulates 5.3.2.2 centrolling reden releases 6.
havironmental considerations in mill process selection 6.1 process options 6.2 Reducing envireemental impacts 6.2.1 Radioisetspes 6.2.2 Ben-redieettive pollutants l
4
' 7.
Impoundment siting and design options 1.1 Information collection and site selection 7.1.1 Information collection 7.1.2 Site selection 7.2 Tailings impoundments 7.2.1 physical confinement 7.2.1.1 Valley das impoundments 7.2.1.2 ting dyke impoundsents 7.2.1.3 Eine-pit impoundments F.2.1.4 Specially dug pit layoundsents 7.2.1.5 Underground mine impoundments 7.2.1.6 Deep lake impoundments 7.2.2 Types of ecvar materials 1
7.2.2.1 Cley covers j
7.2.2.2 Wa%1v4 soil tovers 7.2.2.3 11 prep covers 7.2.2.4 Other evers 7.2.2.5 Surface vegetation 7.2.3 Seepage cor.tvol 7.2.3.1 Liners 7.2.3.2 Control of the hydraulic gradient 7.2.4 Tallings management systaans 7.2.4.1 Saturated asnagement 7.2.4.2 Wet management 7.2.4.3 semi-dry management 7.2.4.4 Dry senagement 8.
Treatment of liquid eftluents and monitoring 8.1 pH control 8.2 Radium removal 8.3 monitoring 8.3.1 layoundment facility monitoring 8.3.2 anvironmental monitoring 9.
Stabilisation and rehabilitation of tapounements 9.1 General preparations 9.2 Demetering 9.3 Stabilising sentrol strweturee 9.4 Diversten ditsbes and spillways I
9.3 Berriers to seepage 9.6 aslecties of cagineered severs 9.4.1 control of redom emissions from impoundment surfaces l
9.6.2 control of games rediation at the ungoundment surface 9.6.3 Breslea centrol 10.
Application of radio 10sical protection principles 10.1 Engineering practice and dose limitations 10.2 optimisation of radiation preteet.ien 10.2.1 Differential cost-benefit analysis 10.2.2 malti-attributs analysis
O 11.
Assessing the long-tern safety of closeout options 11.1 Introduction 11.2 overview of safety assessw nt 11.2.1 Methodological approach 11.2.2 3cenario analysis 11.2.3 Consequence analysis 11.2.4 Evaluati n and application of t*vrults 11.3 modelling and analysis tecimiques 11.3.1 Probabilistic analysis techntques 11.3.2 Deterministic analysis techniques 12.
Sussnary and conclusions
- 13. References Appendix A Abbetriated check list of considerations for the management of alli tailings Appendix 3 Approziasting the redon source strength of a tailings impoundment Appendix C Example of a simplified technique for assessing site and tailings
- annagement systems Apg andLx D An example of the use of methematital andeis 70 A 3 i(3 I 74 iskg Ttlt A SA(GT$ e t.'
A rt,,s p ts,pygf cg ogggy g-gyng sv.f I
I t
]
l l
l
-~
g i
.a.
0622y 1988-07-06 1.
Introduction m mining and milling of ores to produce the uranium required to fuel the a17 nuctsar power reactors currently in operation produce large quantities of solid and liquid westes which must be dieposed of.
For exaay1s in Canada over 160 million tonnes of urentum mill tailings are stored at various sites and it is expected that these westas will inersese et the rate of 10 million tonnes per annus. At least 18 countries are reported to have uranius mining and milling capability and eight of them have a produ4 tion capacity of at least 1000 tonnes of uranium per annus.
Urentum mill tailings are the solid residues and associated 11gulds remaining af ter uranive has been extracted from an ore. The tailings initially contain about 861 of the radioactivity originally present in the i
equilibrium eres and up to 9M of the anos of the original ere. The tailings may also contain varying amounts of chemicals employed in the artraction process, other toxic chemicals indigenous to certain ores and gypsum which forms when acide in the tailing sisrry are neutralised with ilme.
l l
Although the radioactivity and some of the chemical pollutants in the ta111nss are natural in origin, the mining and milling processes bring these pollutanta to the surf ace and change the chemical and physical form. These changes increase mobi11ty and consequently enhance the possittlity for
[
I disperstem in the environment and detrimental Lupact to teamans. Also because of the presense of residual urenius (which has seen very long-11ved r
radielsetepes) and its decay daughters, the tallings will remain slightly 4
radioettive virtually in perpetuity. Fortunately the specific activity of r
b uranium mill tailtags isT ow.
i i
Permanent isolation of tallings from the envirereent cannot be eseured but disposal practices should attempt to ensure that basic r+distion protection principles are applied and that release rates to the environment new and in the future will comply with current authorized LLaits.
Persistence of the radioactivity in the tailings requires that the future period of concern be defined in terms of thousands of years. Won-radiological contaminants persist indefinitely in tailings; so tailings can be a perpetual l
l
O source of arsenic, molybdenum and other hasardous substances. On euch a time span significent sociological. citaatological. geomrThelogical and technological changes are expected to occur. The nature and trend of these changes cannot be reliably predicted today. Responsibility to future generations suggests that long-term stabLlisation of mill tailings be assured j
to the best current tectusical and economic ability.
Engineering designs for tallings Layoundments that are discussed its this rescrt involve solutions used or available for tailings retention. Depending on their design bases. many of these impoundment concepts can be expected to l
provide for secure retention of W tailings materials and to asintain rolesses of radionuclides to the environment within design limitJ for a period of the order of hundreds of ye'ers or more. Throughout this report this is referred to as the ' design-life'. It should be noted. hoemver, that i
engineered systems any well, and of ten do last considerably longer than their design life.
It is recognised that engineering solutions for tallings impoundmente should provide for more perunnent retention of taL11 age material, bearing in mind W longevity of the radioactivity contained within them. The most advanced engineering design practice can, and of ten does. Lacorporate welitative consLderation of the effect of predictable geomorphological and c1Lantelegical processes en W integrity of the impoundment system, during ohich tLas the W r '
.t remains substantially unaffected by these i
pescessa. This seule cover periode up to 10 000 years but is typically about 1000 years. This period is referred to es the ' Lees-tors' period in this report. W 1argely te sesseryhological processee within such periods, even I
the meet oe.wl, e.stmeered==mt eye
-t he -eted to -re semplete retention of the taillags material. Ikesever, in sortaLa sirouestances asturel changes any produce an incrosse in m confinement of the westos rather then a decrease.
ch.
(tLe I V
.ey
. time seale of wareds of me., the difficulty of,rwicung p g. k,- turai ch.nges in the envi - t, ch.nges e,susht b, e..anctag
)
e dL and the diff.eeuw in sniat stmetura anb. lo g-tem predictiou fp uu no.ev.c if th. b.ste,rtact,tu of th. uteu. tavotvw ar. seit J h Aue. ruta.
r.tood.
m di.,osal t-w tog, is
..d on souw. a u e,t w engineering nothods, then short-tern predictions can be accepted with u > @n.
ru sou ble confidence. sven long-term predtettons will be useful. If the
.vA Jvy lh u /r
/
i 04227 i
~_
e 6-
} 3 p k Q predictions show that, eventually the releases will decrease, se dit:olution
,(
end decay reduce the inventory, then the Lapact of a serious failure will be
/ acceptably low. This, however, takes into account only the on-going 6) processes. Catastrophic natural events can be factored into the prediction 4
and the probability of occurence may indeed be low, but should such an event M
4 take pisco the result could be a problem in certain areas. Isore difficult p M L4e 4 still is to incottocate social changes. In particular, misuse of the takings vg mast always be a concern. g However, because of the low specific radioactivity of tailings, the consequences of branching the confinement systen and the resultant dispersal of tailings would not lead to catastrophic or necessarily to significant radiological impact, since leas and rustained exposure to radioactivity in tailings would be required to produce significant adverse results. Even though the potential consequences of breaching a tailings confinement system 1
over the long tem are orpected to be small, our responsibility to future
[
generations suggests that the aim of mill tailings management should be to.
i C.
1 stabilise and confine the tai 1 Lass for an indefinite period, using the best L
W practical means avallable.
D
~%
Cs
,N In seen countries such as the United States, the competent authorities require a continued presence at W site. Such a presence is expected to Lc perform surveillance and, where appropriate, monitoring, theuld any action or i
7 Cy repair need to be effected, the custodial agency (state or federal government)
$g would take the required action.
I"y thiring the recent poet, significant efforts have been nede to understand A t
.the ymoos which are Lupertant La wrenim mill taL1 Lass management.
m
-. pre.....n.
,,.ng,n r
L.,_
n,..c m,,..,,
to 51 In e.dition, effective ways of sleeing off these Layoundments for long E
taru disposal and for deins remedial actions on elder tailings piles have been developed.& This report is an update of an earlier IAEA report on the management of mill taillngs (6) and instudes en overview of the most recent work.
j 2.
Scope i
l This report prssents en overview of the current practices used in the design siting, construction and closecut of Lepoundment f acilities for uranius mill tailings. The objective of the report is to present an i
_._..___--__,._m
- ~ _ _ _..., _ _
o intedrated overview of the technological, safety and radiation protection aspects of these topics to ensure Wt the potential radiological and non-rediological risks associated with the management of uranium mill tailings are minimised now and in the future.
I The reports j
(a) identifies the nature and source of radioactive and non-radioactive
- pollutants in urentum mill t4111ngs (b) identifies the important mechanisms by which pollutants can be released from the tailings impoundment and the parameters that control these anchenisms j g g,, 4 y g g
N' g
(c) describes the pathways by which the pollutants may eseeh A
I-(d) describes some of the site selection and design options that may be l
i
'\\
t=1ement.d to ttatt the est.nt of re1=s.s f re. the tapoundmentg
/
D (e) reviews options and considerations for final stabilisation and i
g rehabuit.ti of uutnas tu,.on.monts (f) describes the ways ta assess sleeure strategies.
{
L Secause of the templexity of the pollutant release nochaniens and the site-specific nature of the design and annagement conteels that can be j
implemented, it is not practical for a report of this nature to be either I
/
/
exhaustive er applicable in detail to all esses. The methods by which i
/
confinement controls are employed for any particular tailings impoundment will depend on the sountry and its site-specifie perfernance eriteria which must be defined by the relevant regulatery authoritiae [5]
U' (br
(
Both operating and post-operating senditions are censidered. After y
shutdown of the mill plant and stabilisation of N taittage, sentinuing sj suree m e
int co e - 4, e.ri.us1,.
- i. ore....,,,e - to.
to some estaat. Bogardless of N 1ses-term care stretegy, future stability of the eenfinement elements should be given semaidorable emphasis, j
The report does not omanine scenaries which, in the long-tors, may result in partial er even substantial damese to the Layoundment eyeten, nor does it detail the redielegical consequences reeutting free tailings dispersal l
through such events. unthodologies for further====ination of these questions j
are presented elsewhere (?).
l l
06331
=.... -
-s-i l
3.
Mining and milling uranium ores Before uranium ore can be extracted from en open pit or underground aine, a considerable amount of weste rock and low-grade uranium ore has to be removed to petuit access to the economic grade ore. The ores are assayed by various means to segregate the weste rock and low-grade ore from the mill i
w ality ore. The uranium ere is estracted from the mines and then transported to the mill where it is processed to produce a concentrate, called yellow cake, containing at least 65% uran'Lua. Although each allt is designed specifically for the characteristics of the particular ore, the processes used by many mills are stallar to those shown in Figure 1 and have the following basic steps:
crushing and grinding of the ore into a consistency of fine sand to emyces surfaces to the leaching chemicals leaching of the uranium from the finely-ground ore using suly k ric ac'L4 or en alkaline carbonate solution depending on the ore. Acid leaching is the most widely used process solid-lip id separation and washing:
solvent astraction or ton eschenge separation of urentum from the leach lip id precipitation, drying and packaging of the yellow-cake concentrate, wrentum diurenate.
la these processes, significent wantities of reagents are used: for example vp to 100 kg of sulphric acid are used la the acid process per tonne of ore silled.
Quite eften, the urentum La the low-grade ore is removed by heap leaching or in-situ teaching. In heap leaching, en acid solution is injected into a pile of low-grade ore to leech the uranius. Tb8 pregnant liwor is collected in a plastic liner under the pile and sent to the mill.
i 04227
o EA RUN OF MINE ORE
--O y) CRUSHING AND GRINDING V
i g
(
)
Acid 2
^
Omident r
R ecycle 2
Solu tIO n LEACHING t
a L
1
- _l_ d Te 9z3_
i
% r-3 Oa I o bM e<<
V I
dC3 SOLVENT l
J f D "* V o
k
^'
'o o
oA EXTRACTION l
z Owm4 o
l l
RaHinate to i
'7 W
0
- recycle and M" %,,
weste disoosat
,, Stripoing o g o
o Water agent I
I f
j c_N"b_,
SOLVENT u
TXTRACTION i
N H, i
o[j o
l to weste disposal PRECIPITATION.
PACKAGING AND ORYING 4
i
[
)
Yellow.cak e product i
FIG..
I Uranium ore processing flowsheet.
i
~
e
.g.
4.
Factors in the management of uranium mining / milling westes 4.1 Types of wastes Lacte volumes of weste rock and water arise from ths mining of uraniva i
eres and large vol m es of tailings result free the nilling process. This l
..cuen,rie,1,...cri,es
. w.st...
4.1.1 Itining westes Westes arising from the mining of uranium cres can be classified as exploration and operational westes, the latter having by far the largest volume.
I Ruploration westes could senstst of trench weste, drilling stuere and/or core semples. Itatorials free emploration trenches for open pit mines should be segregated as unsch as possible into everburden, weste rock, low grade oro asw high stede ore. )Aten the say1 oration is finished, the trench should be ref Lited with h highest activity asterial at N bettee sad overburden at the top. Since drilling studge can betone contaminated with radionuclides, staat care should be taken to ensure that sludge and drilling water are contained. Also sore samples taken during emploretion should be disposed of in an approved menner when they are me longer required.
The operational westes from uranim alming senatet of weste rock, low-grede ers apud water. The weste reek and low-grade ore have to be reonoved before the mill-quality ore een be aimed. The velues of weste rock and leep-grade ore will depend on amay fastere such ass W ore stede, depth of W mine and whe b e the ore is in ses11 pockets or asesive deposits. At certain mines, the volume of weste rock any esteed N volume of the oro extracted.
The inactive oeste rock saa be dieposed of or used in the following ways once its minerelegy, redleestivity cad therical activity has been assessed as being acceptable for the intended uset to refill the mines especially opee pitet to construct embeniments, impoundments, roads, water diversion channels, etc. on the site.
A499w
.. Low-grade ores een be used as a source of uranium which can be recovered by heap leaching. The teached rock would then have to disposed of in a suitable menner possibly with the alli tallings.
Large quantitles of water result from the mining operations either froe its use in drilling and dust sentrol or from seepage. This water becomes contaminated with radioactive meterials, non-radioactive 'oxic miner is and with enemonia and nitrate free blasting agents. Allminewatek u treated in the weste management system before discharge to the environment. Generally this water is used in the milling operation where dissolved uranium and other contaalt.
s are removed.
4.1.2 ELiling westes Uranium mill taillass are the finely ground solid residues and associated liquide roeulting from the processing of ore for the cocovery of, uranium. The tailings typically consist of (a) s11mse: W lighter, f W e particles La the tat 11nss, including those in W micron /sub-micron rangs. mode up of the elsys, silts and other very fine porttelest (b) sands: W beavier, seerser partleteel
~
(c) chemical neiduos free W alli peesees:
)
(d) a variety of heavy metal sentaminants (section 4.2.4) and other subetences such as gypsum which fores when residual seide are neu'.rs1Lsed with lias.
l These ta111 ass sonstitute a particular kind of Low Level radioactive weste eheretterised by rotatively large volumes and law concentrations of 1ses-lived naturally escurring radienue1 Lees.
l l
l These residuos, in N form of a slurry containing 25 to 40% solids, are I
neutentised with time to a pH of about 10 La N mill and discharged to the l
tailings basin. In N tailtage basin. N ee11de settle out and, because of the alkallnity of the water. W dissolved heavy mew 1s.
Th.
Fb and 226 auch of the sa are precipitated out. The esistively clean liquid decant
)
overflows to the precipitation pond where it is treated with berius chloride l
to proeipitate to the dissolved sa as a Re-ga sulphate sludge. Water flowing from the precipitation pond to the stream or lake is required to meet 0622y
, regulatory water standards for example less than 0.37 Sq/1 (10 PCL/1) for 226 R4.
The settling process in the precipitation pond can be quite slow and retention times of 2 to 20 days are required depending on the design and chemistry. ylocculants such as ferric chloride are sometimes used to speed up the process. As en alternative, the radiva precipitate een be removed by filtration in sand bed e tallings basin is the largest of the two steas scatay wouis escupy 40 hectares compared with to hectares for. the 4 p h ce. V 4v<sd*/
vZee.
(
precipitation pond, } % g.f. C y/y wmalouf-lu.
pc.q' lea. Sr. l.2.
When the mine and mill cease production, the facilities are decosetissioned and the mill tailings storage areas are modified to meet the long ter:: disposal requirements layed down by the regulatory authoritles. The best methods of saf ely closing out tailings piles and treatment ponds to minimise the long-tors radiological impact on aan is still under discussion (see Section 7).
The main emphesis in this report relates to the managrant and disposet of the milling westes which are generally more toxic than urentum mining
. y; f keh n M
M,. a, o au uh A. L !
Thevolumeoftailingsgeneratedeasyyearataparticulermilideponed on the mill throughput, the(oncentreties of utentum La the eS end the g
secondary wested, such as gypous, which are produced during the process.
Table 1 Lists seus data en a fear of the operating mills and the eres they process.
4.2 major sharetteristas of tatitags which effect diepeest tectetelegy The asia radiological, ehemical end minerategical characteristics of ta111 ass which effect disposal technetegy are: uranium ere grade, radioactivity, amid generation potential, men-rediological contaminants and chemical precipitates. DA h h % c d h,e w. ( f m b 4*J I-l 4M.M h 4.2.1 Uranium ere grade M 4 etcbdd $ %
The v 0, cent t of ure iu. er.s curreatir betag mined rang.s f ro.
3 less than 0.11 to over 40% with typical food grades to the mill in each of about 0.1 and 2% respectively. The low-grade cres tre.is produce 20 ti.nes wre tailings per unit of uranium produced then the high-grade cres. However.
I 0422y
o
. the high-grade eres have a proportlenally higher specific activity of radionuclides. Since the tallings from high-grade cres are of greater redlological concern, they may require the use of more sophisticated disposal facilities. An evaluation of the relative costs of tailings disposal for low, medium-and high-grade ores in one country showed that for generally accepted disposal tectutology, the disposal cost per kg of uranium recovered
[
were slailer. Newever, disposal costs per tonne of tallings increased by a facter of 18 from low-to high-grade eres (5).
4.2.2 Radioactivity l
The radionuclides found in urentum ta11tnas are the some as those found in all mill tallings. However, the concentrations of these radionuclides in urentum tailings are nerasily higher then in other types of mill residues.
[
Undisturbed uranium ores contain low concentratiras of the decay daughters of urania and thorium (rigure 2) in equilibrium with the parent.
asterials es a result of natural radioactive disintegration of U ond Th.
yer each 0.1% uranium present, the virgin ere has about 12.4 Sq (335 pCL) of each t
230 member of the 0 series per gram of ore er about 174 Bq (4700 pCL) of total activity per gram. As a result, one tennne of ere at 0.2% uranium would centsin about 23 S q (0.7 act) for each member of the U series er a total
(
of streut 350 mg of radioactivity.
Radium-226 is considered to be the most layectant towie decay produt* in f
the urentum desay series. Det only does it have a "very high radiotoxicity" l
indes (8), it else produces Sm. a radioactive Laert ges, uhese decay f
i products saa eause lung saater. Redinan abeerytten from the gestrointestinal l
treet late blood and oef t tissue is signittaant with eventual deposition in l
bene marrow to),
i la addities to wronism and its decay daughters, near eres contain significant amounts of Th and its decay daughters.
The total radioactivity remaining in the mill tallings depends on the ere grade, the percentage of urentum recovered and the radioactive daughter products presamt. Deras117 shout 91 to 93% of the uranium a; O f the total activity la the ore remains in the concentrate, tailings initially Ht* h d.
Ek CJ hh W u k o+
c[d y
04221
e a
lib
.., w,,
e.. w m.es m.se swes mes og awa.
w hoe e
jg
) i g ogs g j.a
. 4 e '. .
emy g g,, ogs,
i l
egen i.
'gte
..e 1
6.e 1
i i, s
s=
. ii.
.., i t.
A i
i l
q; is e,
go.
s ea.
>:I 1
l w
is m.
I 1
1 u.e.
p i..
.;p.
i 1
', A
's
'='.
g I
i 1
w ua.
p g.,
p
-.J.,
.L i --
1.
I
".P.
yg} {
4 i
Ta qq
- f~
f m
m t
.;p, ee -
i.
i
- g.,
ses a e;.i u.=
1p n.
t
';p.
so.
. en I ammim gs.
is..
1,wi so.-
ppg 4, pasumat sedemew amusuunse a eu suse auf W l
- -- - -- - --1
o contain about 86% of the radioactivity originally present in the equilibrium ore but this rapidly decreases to FM as some of the short-lived daughters, such as Th. decay. Although the proportions vary from one mill to another, over S W of the radioactivl'.y (including ta) in the tailings is in the solids. W remainder is in solution. Approximately 75% nf the activity in the solid phase is associated with the fine fractions which make i
up about one-third of the total asse. It is therefore very important that the i
fine particles be effectively settled before the sluery water can be safely discharged to the environment.
The Es concentration in th3 tailin' solution may vary from about
{
10 to 200 Sq/1 (250 to 5000 pCL/1). Although other radionuclides such as uranius.
Pt and to are also present in the solutitcas, ta is the l
radionuclide of most concern. In acidic solutions, mere than SM of the Th in the ore any be dissolved and could be of equal radiolosteel L ortance,e.
se 22 % is a decay dau.hter..eutralisation of the 4
tailings 11guld will reduce the # Th concentration significantly. Becausp 2M 7. of biological wptake factors, whe W e Th is ersonically or inorganically 9
i b
bound, will aske a big difference to its contribution to dese, organically 2M g kM5-bound Th is of far greater concern. leewever, predictions using 1phM 2M et-Latim models strgly indiute Wt W Lurganic complans of Th l
1
(,
will dominate W dose consequences. Although field emperiments tend to ou,,eri m s kr,et (sis, the.* c.nclesien. a only t.nt.tive h..use of analytical and semple uncertalaties (ftt) and difficulties in validating (t) j models (10).
)
Although h specifis setivity el aset uranium atti tailings is relatively low. W radiological hasard will virtually last La perpetuity due, to the very tems half-Lives of the rediensclidee involved.
l The consentrations of radioactive daughters in taL11 age will r_ise 3* m $' U '
.h.r,ty whe. etcher.r are si.ed. ror -1.. tailtngs f(n ur.nig i
i 226 A
hfg erse would cont 4La about 370 Sq/s (10.000 pCL/g) of Sa. 16ewever W volume of tailiass per unit of urentum produced from the richer eres will be p
markedlyQess. One deposit at Cigar Lake in Saskatchemen La Canada contains s
M enium emide concentraIItus ranging up to 63 weight 1 with a nominathe W,[
g M
l c e l J J lu.
% UN6 4 #) U 0. Y 4 yak y
f j
- clung.
=
)
Os22y
- l
_m
.-,.-s_
..-y-,,
4.2.3 Acid generation potential The acid generation potential of certain ores in Canada, south Africa and some other countries is quite high because outphide minerals such as pyrrhotite and pyrite are present. These resetive minerals are susceptible to oxidLastion in the presence of asisture and owygen to form suphric acid. The production of acid een result ins 4
elevated concentrations of tsuit heavy metals and some radionuclides in seepage a reduction in the pH of adjacent water systems and the destruction of l
i
[
equati g ies:
6 quantitiu of sludge if the acidic tu chate is tru %
&a & w c,$4 s+c { h f r*Asdd l %'h V*Ur 5 tenedial asarures to prevent acid formation include removal of the pyrite 4.,,/
durirg processing, sell covers or flooding of the tailings ta s'inLaine emidation of N pyrite.
4.2.4 son-rediological contaminants I
There are a large raent>er of men-redlelogical contmainents in urantum tat 1 Lass which san be mobillsed under acidic senditions and appear La seepass i
including heavy metals, rete earths, salts and nutrients. Elements and campounds common to many uranin taillass PLlos are shown La Table 2.
Tte
.igestt-of the to ic.he sule Late the enviro.e e fre. t.utngs,ue. -st f
, M M red,d e s.- b
[A besentre151".Or'M 4 &~f L.,, L.d l
we me a n J
?
- .2.s Natut pe=tyt 4t= fL Ay,,gg,, & ( f $4y Pro ues ehemisals seed La N mill and precipitates from the subsequent treeteent of teL1 Lass saa sentribute both to the wlume and weight of weste genereted. For emanyle large volumes of types (calcium sulphatO can be
' formed when sulphurts acid solutions are neutralised with time timestone, f
Precipitates of S N r elements such se iron and aluminium can aise contribute j
to Lacrossed weight of tallines. These precipitates hinder consolidation of the tailings and are difficult to deweter. Sulphates can degrade concretes j
and aske clays more perusable and less plastis and esuld N eefore effect containment stevetures. The treatment of liquid offluents from the mill can result in fairly large volumes of Be(ta)30, precipitate which has a l
relatively high concentratLon of 226,,
g I
l
ifA Tania 2 List of non-radioactive, potentially toxic substances in uranivs 3111 tallings 111, 12)
I Depending en the uranium eroe and the milling processes used, the mill tailings seu14 sentain nony er all of the felteering potentia 11-, comte sube tenc es. - L.. -. _ _ f '
- L = r m, a _.. -.. __ _ F M ' -* = _
l Arsents moreury Serium Molybdee m Seren Wiehel caemim ultrate
-m a
1 Copper Selenium Cyantee silver tren Thorium Lead Drenia tensenese Tened1m Di (2 ethytesyl) 70, h
t eoopherts sold Pyrite 3
Pyvestite 3
3 00 toedesanet 3
1 M
Mi h tyl M W e 3
30 Tertiary Astees 4
06457
. e.3 Basic principles of tallings asnagement since the non-radiological pollutants from uranium mill tallings can be i
as great a hasard to the environment as the radiological pollutants, both sust be factored into the safe management of these westes.
The basic radiological protection principles for uranium tallings weste asnagement are hosed on the Basic Safety standards for Radiation protection (13) which apply to all sources or practices involving emporure to ionizing radiation subject to contret by the competent authority. These standards are based on the system of dose limitation reconmended by the ICRP 114), which I
comprises three principles: justification, optimisst un and individual dose l
1LaitatLon. These principtos are expressed in the sesic safety standards (13) as follows:
t JUSTIp! CATION OF A ptACTICE: To prevent unnecessary esposure, no practice invokving exposure to lentsing radiation shall be authorized by the relevent competent authoritles unless the introduction of tha l
practice produces a positive met benefit.
0FTIN11AT105 0F RADIAT105 pt0TSCT105: The desLan, plan and eubequent use end operetten of eeurces and practless shall be performed in a manner to ensure that espesures ete as low as ressenably achievable.
economis and social fetters beLus taken Lato e< count.
I IWO!VIDUAL DOSE LIMITATION: De individual shall be espesed, as a result of sentre11ed sources and prestises. La escoes of the receamended 1Laits.
In semeldering met benef 6t. the diepeest of uranius tallings should not be seasidered on its own. The met benefit from the whole practice for which the uranius ore is being mined, that is the generstlen of electrical power using uranius fuelled reacters, snaat be seneidered. Although erp11 cation of the latter two principtos suet take Lato secount the total fuel cycle as well.
these two principles can and should sloe be applied within the management of mill t411 Lass alone.
The Code of proctice and Guide to the Code on tha safe Kanagement of Weste from the Mining and utiling of Uranium and thorius Ores gives detailed
(
0422y
l
! guidance on the application of the Basic Safety standards to tailings disposal ll). the Guide deals with the control of hecad e free the radioactive constituent in the tailings and is concemed with the protection of ammbers of the public from the westes t gives detalls of two methods of optimising 3 I
hgr. -g ifadiation protections test-benefit analysis and an altemative qualitative
,ma.U appt sch, multi-attribute analysis.
(See slee section 10.21.
s
}
6
, it t.
q nte.4 w t ner. ar. difficutst.e in th. ap,lication of th.
h p4 y
teste Safety St eds due to the characteristics of the westes from mining P, g
.no oilling.f rutoa ve re..
rw..e arise fre. ao large a l -. of a.
lwestes,potentialwidedi raion and the long half-lives of some of the l
- b '"""k/
cadtuctin constituente.
4asic safety standards were prepared mainly
/
j Ad
.ith eye.u,e,,o
-,e conventio
,u istion..,ce, a aiu, i.e. they
)
d Nwf.
typly particularly to the centrol of eures ehich are certain to occur i
within a fairly short time frame. In the of mining and eilling westes,
(
aere is al.o a need to c..ider eye-c. 4%y or -y not ecor, for
- x saaeyte espesure due to been inttveien into the we es at some future time,.
a In these respectd the problems are similar to the-1 ans tered in developing a radiation protection philosophy appropriate to other type f solid i
radioactive weste disposal. Recent publications by 514 11 en ICSP 116)
.Il sedr.n as.. t.eu... tut fursh.c guidant. sad clartitution er. 11
..d.d
- as,rinet,1 eu ae.. Le -kieve ano,s..le rediation,retotion in i
l relatten to mining and milling westes, purther discuesten en the subject is /
(require $ at natitinal and intomattenal levels.
)
l Begulatory authorities la a sountry should ensure that these principles l
are applied La the derivettee of release limits aseeciated with the dieposal I
of ureate mill tailings. Emmaples of suen limite are thou set for 222,,
)
amenation rete from a tailings dieposal area er the Be concentration in 226 j
ester roleseed to public strooms. auch rolesee limits should be complied with j
by designers and operators.
In addition to radiological criteria, the following non-radiological I
f actors should be considered La the management of uranium allt tailings:
I I
)
(a) the na. ural envirorument mast sise be protected from teric a
men-rediological haserts which see have a mach larger of fect on tne j
environment then the rediological pollutants. yor people, one of the i
l i
0422y i
l 1
. seet obvious environnen ef fects seen to date from uranium mill tailings is the destruction !f equatic species due to the release of pyrite generated acids released to adjacent water systems; (b) the disposal of tailings should, where possible, not interf ere with the emploitation of resourtos by future generations; i
(t) consistent with safety, the land areas used in the storage and disposal i
of weentum tailings should be minialsed:
(d) reliance on attive future control and eurveillance should be kept as low l
as reasonebay schhvable when tensidering the stabilisation designt however, Ce; tempetoot authority pheu14 require sees level of I
post-operational surveillance, monitoring and care, to ensure that the i
stabilisation defign is perforving adequately.
)
j I
j Long-tors management of urentum tailinge under nerinal conditions is cons',dered a thronic te W e then en acute concern. Icelation of taL1Lngs from
. h environment cannet be sucreatoed for the long time span involved, but thp l
disposal practice mast attempt to ensure that release rates now and in the,
i futare will toeply with appropriste su W rised limite.
1 i
l J
j The general application of these rodielegical and neo-rediological j
principles to m siting, design and sleeure of Lepoundsent t'esilities for the safe disposal of urentum mill tailings is diceveeed Le the following sections.
j
!= applying thee* priatiples, nemy senateerations met be included in j
W planning and Laplementatise of mill tailings feat 11 ties. Appendia A gives
- e. ame,iated shook-list e< es.e., mee ee ideretto i-ludt.g Wee j
- ith are baye.d the so m er the - t d - t.
1 S.
heteeee et pollutants from uteutum mill taL1Lage to tesmana
\\
l A variety of anchenisms are avethble by With w.cium L4111 ass and/or the pollutanta La these tailings could be released to the environment. Once j
the pollutants have been roleesed, they can rseth hasmans by a variety of atasepherit and wetar Settersys.
I 1
-l This section briefly describes the Laportant rolesse anchanians, the general pathways by dish the roleesed pollutants can reach h eens and factors af fecting such releases. The sitias and engineering design options available j
to control these releases are discussed in Settien 7.
I i
l 0422y i
.n,
.. 5.1 Ingortant release mechanisms The potential mechanisms for release of pollutants from the tailings and/or the tailings thanselves to the environment include:
spills during the transport of 'milings to the Lar wndment; erosion of the cover or embankments; redon ennenstion; wind transport of partfiles; stm etural fath re of imilings embankments;
-s M
unauthorised removal of tallings; controlled release of contadinated water; gf uncontrollea r.lasse of water.
GLG N C)C.g. f u k (*N W
MS "1
yf
~
h
- '5.1.1 Spills during transport of tailings to the impoundment During the operational phase of the mill, slurried tallings containirig from 40 to 70% water are transported to the impoundment area through piping systems Wich are of ten many kilometess in length. For "dry tailings" which contain 15 to 351 water, transport is by truck or conveyor. Regardless of the means of transport, tailing epills may occur during the life of the alli l
cousing contaminatioie of nearby areas.
Thrwgh a well-designed surveillance programme, f ailure can be identified and mitigation sessures *.sken quickly. The impact of ev:h a i
failure can be limited by minimising the route length and choosing the tallings conveyance path of least sensitivity, and through incorporating safety margias and redundancy in structurst design and monitoring, some ways to reduce this problem eres (c) loeste mill and impoundment se c1eee together es possible and minimize pumpias heads and pressures (b) avoid taL11nss transport routes that cross t'Jter courses, unstable slopes and poor foundation conditions (c) use concentric pipelines and/or specik11y desigr.ed pipeline corridors in i
sensitive locations, and constevet interceptor channels to lead spills to catch pits (d) use automatic shut-down surveillance.
l l
01221
1 1
FICURE 3 Erosion Processes ($)
usueuw TAa.sec8 anPvWDesDff anomon I
sm4 t99 WafsR MECELLANsotA saouon smomon saouen Manas innman y
e Seeseen e
m n=
sunsacs
.4rw=.
asomon Pined ouLLY h and a-smomon a4 s 4
Nome: w somme er enemme in.,,.== wmn.
- Sheet erosion - Erosion which occurs as a result of impact of raindrops loosening surface soil which is then transported by wind and/or by water flowing in small, short-lived tills (seall streams). Or sloping ground, sinor undulations can rapidly lead to the development of rill channels and to sully formation in low resistance nat uial.
N esO 6
, 5.1.2 Erosion yigure 3 shows the potentisi erosion processes which could lead to the failure of covers or dans and to the release of pollutants from a ur6niun tailings site. The principal processes are wind and water erosion. Once the impoundsent cover has been stabi1Laod, wind erosion tends to be less of a problem than water erosion except in arid areas where both mechanisms are importsat (5).
Surface water erosion is one of the more likely mechanisms for the disruption of uranium tailings impoundments over the long-tors. Water can erode by surf ace runoff or by subsurf ace seepage or piping. yor example if diversion and spillway structures are designed for 1 in 200 year flood events, the as-designed flow will probably be exceeded several times during the period for which tailings impoundsents are expected to be effective, for example 1000 re, unless larger and more stable diversion and spillway years. Ttw '
structures, x incorporated into designs, some degree of flood erosion wili oc cur.
To minimise the r91 ease of contaminants from sheet erosion ty rainfall or amiting snow during mill operetion, all (except well-covered) embankments should be constructed of uncontaminated materials, and all water f.owing over tailings should be collected and managed, yor stabilization, use o. e coarse gravel covering antarial, vegetation er other oresion protection'such as a protective layer of rock, and contouring to gentle slopes would minimize erosion. Site specific conditions will define what strategies are needed to stabiliseapileagainsttheravaggotereeton. In some cases well-protected r{
(arneured) diverr. ion abonnels, aprons and other construetten featurse may be needed to further improve the durebility of a stabilisation design. It should be pelated out, bewever, that although use of vegetation is en accepted method to minimise erosion, more research is needed to understand the possible remobilisation by vegetation of sortain contaminants such as redium[
Wind erosson in arid regions, such as occur in the south wast part of the USA and parts of Australia, can be a serious problem if covers are not properly designed. Howe with proper design of ver end control structure., wind erostA
- WA%Nte.',; m=.
4 a
06227
. 5.1.3 Badon emanation and transport Diffusion of En the decay product of Ra, from tailings into indoor air is one of the main environmental oathways. Breathing redon, an inert gas, and its short half-life decay products, which attauh to tiny dust particles, exposes the lungs to alpha radiation (principally frote Po and The exporures may be large for persons who have tailings in or p
7 around their houses or who live very close to arid tailings piles, people
\\D 0
with bouses built in areas having high' natural concentrations of Ra are subject to similar dangers. The best method of asasuring exposure due to redon from such sources in or around buildings is from direct measurements of the redon decay product indoors.
Because of its gaseous nature En is continuously released from 22e Ea.
^ft;; 2^ i.. et aill.r...iive
!;t.-a 4 tailings material containing as
-.u ta. m y =16.,
_-1, D.., mu....r e.7 -.,ning.. o 9,
{@
W y
ma',a a _. _. u_ _. t.
,.. gy o..
..u....
,a
. m
( "AW * /
* - !! ! :. Only a small fraction of'this radioactivity is rolessed into the hird staosphere, the value of which is detenmined essentially by the physical and y
j V
minerslogical characteristics of the tailings particles and the moisture W
G 222 H$
content and thicimess of the tailings. The short half-life of an is an i
V h.lN important f actor, since diffusion times within the tailings are usually j hMj' sufficient to permit a significant decay of gaseous an to its solid decay 2
P products which remain with tae tailints. 5 t_=t=
- = t i,,
[. [,,,}4
..,--- y..n_,i....a_u__ %,,.-_ _- _.._, _m yevW NedA4W
...._f
)t
.g a.,,
w s.-
h e
- 1E '- 1.1 te_t" M h Aar_-i-- "M^
' = = ar i f
Qj7^'~(y f- : w.....s u s-u + -x 5,np gg pg i, atys) l c:i o - -
m - - t: ie :. :___ w... - z,;. = w - to 1
\\
p&
,k
& Aum,w a,m. w u &
o Badeo-222 emanation rates will be reduced if the tailings are saturated AdvY l
..g )W k* g g@g g
or nearly caturated with water er by the addition of cover material over the o M'
J pile. For le. if pore ces are saturated, the corresponding annus1 71e 2:_f " " @9 !-
u stad.
mw t tg f t).
1
- 1013 *. A maximan 11 alt f
etlease _..
M r. -
for 2n amenation free stabilised tailinse of 0.74 Sq.a-2,,-1 A N 22 W
(20 pCL.a.s' ) has been established in the U,3.A. [17].
aMM :
p~
14 15
-1 b C,
f)
With redon emissions in the range 2.6 1 10 to 1.1 I 10 sq.a the atmospheric concentration tef an at distance of 2 km from the source d;
I5 ruce..
l I
e i
t
-___.____._,_.,_.,___,~.-,,,..___.,---.-_____,..,,__m___.___.,___..,__.,_.,___,,__.
Y
( )(LA f4 Qg. CW y
-M d (W 5
)
N g
3 and sing e ical ' dilution factor' of 2 1 10 a.m woul be in the
}, range of 1.5 to 74 Sq.a This is of t same order as W natural 222 concentration of Rn (of approximately 37
.a~ ) resulting from an
'?7 i
exhalation rate of 0.015 Sq.a".s (18] from the Ra concentration
~
naturally present in the top roil layers of areas not containing uranium ore (approximately 0.03F Sq.g' ).
These values are in good agreement with recently measured concentrations taken throughout the Elliot Lake district.
which range from 5 to 12 Sq.a".
The USME has reported air concentrations and flux for redon at 24 locations sa part of the characterisation process in the UtfTRA Project cleanup. The results (Table 3) showed wide variacion of both concentration and flux from site to sita as wall as wide differences in concentration and flux et some cites, for example at spook the flux varied from 7 to 104 Sq.m'.s Although uraniva a1111ng sites are characterised by a single value twpresenting the redon release, the redon concentrations in air vary not only with distance from the pile (Figure 4] but also with the timp of year (Figure 3).
Airborne transport of redon amenating from the Latiirus can result '.n large variations in the concentrations directly over the pils and in thes vici. '.ty.
Because of the above variability in redon, the United States has relied on a standard relating the redon flux to design, not actual performance.
UWRte t. O f* TMf*
Factors affecting the4 release of redon from tallings are discussed further la section 3.3.2.2.
5.1.4 Wind transport of particles Airborne transport met only giv*e rise to the spread of radioactive gaseous pollutants but also of particulate pollutants.
Wired erosion of dry and mcovered tailings piles een coeult in the suspension of radioettive partistes la the atm q here. The amount of tailings dispersed by the wind is a fumatisa of the ptysical characteristits of the tailings (particle else, asisture esotent, surface texture, a14 ) and meterotonical canditions.
B 04227
boa TASLE 3 Dans en Raden Enuas.e= flates and Coatentru.eae at lautsee Sites Prior te Remedial Act.en Flus fregi Ceaeontest.ea over Ta.hase.
Teil.nes.
Other ceau ntruiene.
Leut.ea Be (pcilm.e Se/m'(PCi/L)
Se/m8 (>C./L)
I Basse 3I(54) 0 074 (0 Oct) u 3 am smr.eid I f tool (mas )
Be man Il(ell (man)
Derpse 1 3 18 8 (30 310) rone C.s,8 0.l.3 e (s.fel 70 (1 el seatsrevad reeertei u sf (1) smeeeweed as 0 $ am CreadJunegaaI O S.34 (fl.464) 3300 (59) (maa poparted) se (3 4) as 0 $ am Green Rever*
8.3 4 e(33 83C) e4 (3 3) es 4 4 tm. 3310 9) si 3 3 tm besserewad llIi 1)
I*
138 (8 0) (mes, resorted) 8t (10) as I te e tm Cenassea 5 e 33.6 36 3 (480 780) sise M (0 4 se e f), eistanos nei Laaeo.ewA secessed Leem 8 8 8 5(40 180)
Besseround averades 44 (131 Wspteu 3e.87(fl.se)
See (le 4)
Seekeround as a<en u L10 (3) eeing te large amewat of eee in reg on Wea.ua lias 0684(141000) 833e.30,4e0 (St.ast) f 4 (3) na 0 e am, oneagre.ne 33 (00)e43nm Weasewei V apey 0 8.l 1(14 39)
See (e e)
Basegmad 33 (0 8) at 0 6 km Nuentad 38.e4 (f 0 1840)
Masemwm SG4 (all as 0 3 km delante. Duegrowed el te (3) u 8 km New R.Ae 3 4 58(f9.l(40) lose (le)
Besagrowne u i e am I
e 18 4(4 94e)
Norte C tineet 044
? 6-46 (3to.130s) laes(les)
Bestseowad 66 i e im Pasabes/Ua. sed I 5 83(es.aes) nees (10) a4 0 3 km. Dastgrowed cooectee a ll0. loc (3 ie 8) i flieertee8 194e(ll.41) 844 (4.4) (mes.)
Besseromad el (1.5) u I 4 tm I
sa>< Lese C i,
)
(v.we pad) 8 f.Se (100 000) f 4.tes (6 318 4) i Seatgrowed 33 (0 s) at 0 t am 3 0 8 e(13 187) of 4 (13 4) 31(1) u sete uovadary Shipress ls4ea f 0 10e(les.3000) ese (17) (mas )
Ses&eromed na a tm Toba C.sp 0 4 88(ll.40s) ele (33) 74 (3) st 0 3 tm. bungrowed 36 (8 f) es 5 km i
I O 3 0 9(4 84) 140(88)u43km Ca.ea Care.de 1
$0URCE rrown dua peseemoed in U.S M - : et Beerey (1983).
Acesered with earw e th.:tasemme of sneaarial.
Ecevered..la me to 4.a af monaried, eerises stesse of seemse, s
feele[* todteestfee letll'ese $ssersted by the IIsel Will8 Secticeleiss. sithe bM, 583 et.FM.
heden. 2 Saillien lese D.F38 h45 ft.lM tt/
AM hetlet 808 8toege pod 8 L87 9.8F 4M 0fe h
tut eemsktse ese prestiege Age 6.5, 4.5 3Dtteemehe Wytag 6 poetaffagd let 8.n 0.16 negitgttle SHitsee Ptle 4.7 188 W
eeII Steespeed see 4 teillegs8 p pg ow.4
- fte hetet for seems 4tg collegese em visse te4 pose #4r$4. All essert ese cessmed to tue 6tytfleest eigtts.
I hets that honeses of tee teert tolf.ltfe (3.3 g) of poes.M.
little of the et e Cl if test restessette releases e,etag a yese,113 en greems te nas cessent et see ese of tant sese. getite,%e esmess se,eie of coes,eiesse e(n,eepl, e,s stjg,) em Its een of rettesettie esedt is estaelsease voletteely estaaty, fu fter. selease et a essetert vote seleelet to 8000 (t/pr will poesnt to tal; M El of bM eettting ausste tes emerie 44 eer tSee.
- fetal mess celeased fece all es s esseettme is estleted to to 4.3 6tMur, eith a sessiff t estteits 3.4 ttees ans seeeegr seestfis esttette et ene see.
' fetal ets relected free sellemene eseestles le settested to to 1.4 kg gettjent, if eseceps esse att der /reer.
ey, ese isitis, e,ogg ggis,,eag e,, tog es,,gi;ees,9,,qee ge e ess,,e,y sesses of h.rt!
V Q',
N.1 S I es messies. fte f tpee glose fee ets neares et Cl/pr. Is test talmletse se i J
result eftee il pean eserette.
.. ~.
i
..___-..-,,---,.-,,,,--._,,,,-,,--,,,__.,.,,,_,,,-.,__,_,,-,,.-.,g---,,
_c__n n.m.,e-,.,,
ny,nm.n._,.n-_c
-0, d -
t4s is oi -
0 y'
e aseeacae,c w...,e.
- s....,. e.
- e. c es.
3
~
ts4 74 42 01 Ix i
3 2
l e
y }f ti ciL e
e 5
I:
i 18 510 $# -
0 t60006 t12000n (18000s le 37 5$
Cist &NCE 8 80w V A ANivM Wu. em s'esti gg 9
Fict.RE as Wristion of atmosphere raden concentransa.ici 4.sianseg
$hitley Basin Mine/ Mill Cornples.( fi. IT A) so -
.s
<a t#=
0.
<-, eanw g
ix 5s l 20
- 9. i.
1
/
\\
y n
4 s
8 s
.0 E'"N h~
~
~
~~
$~.
s v
tLammeaan CAR 411
''''''''''''''I ''''''''''''I 0 FW A M J J A SO N O J P M A M J J AS O N O J P M AM J J A S O N O J F M %
Ite l
t981 l
1982 l 1943 i
) noune & u inir,.u.....,.r.... c i.4..iew=.,i. D 1 A. w]
. tatisation of wind erosion and suspension of tailings into the atmosphere is similar to that of estimating re-suspension of other substances from soil. TtrJs, research and siodelling, which have been carried out in the area of suspension of other materials, any be applicable to uranLua tailings.
For example, on-going research into the relationship between the physical charseteristics of farm soils, wind conditions and suspension of the material is applicable to tailings. This work also shows saltation of larger than expected particles (21). Air-dispersion, plume-depletion, and redon-daughter-ingrowth models have been developed for use in determining radiological impacts fece this source, for example references 22 and 23.
Wind erosion and resuspension can be estimated by using a soil-losa equation as developed by a number of researchers 124, 25). While these equations are generally applicable and serve to illustrate the relative importance of the various factors influencing wind erosion, the accuracy of the calculated values as applicable to a specific esse for uranium mill tailings must be confirmed experimentally.
i The predominant dose from airborne particias is to the bones from asting foods contaminsted by Th,
- Ra and O
O pb and at is snell. Some esposure by inhalation of airborne respirable particles is possible.
Table 4 shows the results of recent caleviations relating to the radioactive emissions from the tailings pile of a model mill in the U.s processing 1500 teemos of (0.1% U 0, equMient) on in en add y,
3 (11). Releases aristat from other parts of the mill operation (ore storage y ped, erwehint, grindias, wreatum product drytat sad packaging) are auch smaller them these from tai 1 Lass piles for all radionuclides exceyt U and 2H,,
5.1.3 Steveturst failure of tailings embankments 8ttvetutel failure of a tailings den seuld result in the release of large quantities of tailiass solution free the impoundament. Where tailings are saturated, lorso discharges of solide may occur af ter this type of breach. Such an incident occurred in Church Beck, New Mexico in 1979.
Approxiastely 357 million litres of tailings liquid and an estiasted 990 06221
. i tonnes of tailings and erlids were released into an adjacent arroyo. The liquids floered through a network of arroyos and eventually septied into the nearby Mio puerco river 126).
The probability of embanianent failure can be minimized by using appropriate design and constmetion, and using materials that take into account the specific foundation conditions, seismicity and characteristics of the impunded materials. Improved stability can be obtained by limiting the des haight, using gentle side slopes and underdrainage. This is one area where an effective inspection schedule by competent personnel can significantly reduce the~ potential for das failures. Inspectors trained to distinguish settlement cracks from those indicating a shearing process can provide W mill operator with early woming in case of stmetural failure.
purthermore, vigorous application of quality control in the retention das constmetion may lead to a more rollable barrier to ensure weste retention.
The consequences of das failure my be severe if the tailings being i
impounded are saturated and of toer deessity. These characterl N r.d t'..r-* s e '
the potential for the tallings to floor as a liquid and which would reeMi in large discharges that might travel a considerable distance, as in the case of the church Rock das failure. The potential for a das to fail by liquifaction during an earthquake are reeced if the tailings are sufficiently dry or dense.
3.1.6 Unauthorised removal and use of tailings for buildings or fill f
suy3rficia11y. the seerser fraction of mill tat 11 ass looks exactly like f
clean, good quality sand. In the post there have been cases where tailings g
were removed from a disposal area and used as material for land fill and road bh or buildias esastevetica 127). The U.S. Dspartment of anergy (USD03) j
{(
estiantes that $134 million (US) will be required to cleanup approximately l
ik 5000 Ladividual leastions la the Usa where tailings had been used for Nd sonstavetises and '.eekfill (24). His practies has resulted in etmetures sentaining redor. levels that esseed aeseptable 1Laits. Corrective actions have been taken la some eases to limit deoes to the public resulting from this practice. It is tapertant that steps should be taken to prohibit such use of tailings.
1 04221 l
i l
5.1.7 Controlled release of contaminated water yor mill sites where evaporation appreciably exceeds precipitation, such as many of those in the western United States of America. Niger, southern Africa or Australia, aill circuits may be designed for seco surf ace water discharge.
P~ ** natural water evaporation into the atmosphere.
In areas having loer not evaporation, the natural water balance may be such that a controlled release to the environment or discharge to deep
' geological media and/or artifiest evaporation of decent solution from the tailings impoundment usast be practised. If such relaase water is not treated to remove dissolved pollutants, it could become an taportant release mechanism
\\
for such pollutants. However, discharge water can be treated using current i
technology to reduce concentrations of 34, other radionuclides of importance, such as Pb and non-radioact!.ve pollutants to meet regulatory standards. The controlled water release volume can be minimized by recyclin's decent solution and other process waters to W mill, and by proper site' selection and engineering to control the inflow of fresh water to W mine and mill and W inflow of fresh water to W impoundment area.
\\
It should be noted Wt use of water treatment systems such as ion exchange, reversa osmosis, or electro dialysis produce cleaner water and a by-product of concentrated radioactive sludge. In sees countries this sludge would cause radiological as well as non-radiological concerns for the regulatory bodies. This any end up caublag a greater administrative problem than the technical one associated with contaminated weste stesans.
S.1.3 Unsostrolled water releases 4
C 4 m ing of the tailings den after excessive water inflow could result la erosion and breachlag of the -h = % t.
such a breach could cause the release of a large amount of tailings and contaminated water. since water is usually impounded emir during the operating phase of the alli plant, this is only of primary concern during this period. Inflow volume free natural precipitation is directly related to the catchment areal thus, ideally this should be minLaised by careful selection of the site and/or use of diversion channels to route the water from the catchment area sway from the 04227
25 -
impound emmt. Maximra probable floods or precipitation should be considered in determining adequate freeboard for the dam. The design should also meet the requirement that heavy rainfalls would not cause erosion around the tailings impoundment. The inclusion of emergency spl11 ways from the tailings Layoundsent and/or stability of the Layoundment can provide further protection against f ailure of the impoundment in the event of stors conditions exceeding the design basis stora. Disposal of tailings in a below-grade cell or pit or in e pertially below-grade Lapoundsent will lessen the potential for das or retention well failues. However, such disposal may increase the infiltration pot ~ential after stabilisation, unless other mechanisms are used to reduce water inflow.
g dh 5.2 General pathways to humans The generalised pathways by which radionuclides released from tallings
~
impoundsents can give rise to human exposures (Figure g) can be divided into those where the radionuclides are transported through the atmosphere and la,4 or altr,rnatively through aquatic systems. Exposures of teemans: to radiation nar t'm follow from inhalation of contaminated air or ingestion of contaminated water, or less directly through ingestion of contaminated foodstuf fs.
External irradiation, either directly from the impoundment area de f t,ma deposited asterials released from the impoundment, is also a consideration. Exposure pathways may be generalised as follows:
Atmosph stic pathways Inhalation of redon and its daughers Inhalation of airbeme radioactive particulates External irradiation Atasepheris and terrestrial pathways Ingesties of contaminated foodstuffs External irradiation Aquatic pathways Ingestion of contaminated water Ingestion of foods produced using irrigation, fish and other aquatic
] biota External irradiation.
04227
____..___,_____-mr
__,,_y,_
.,----.,._._-__.,,-,--..__y.
JSA is
~
l n.
4.
==o, c
~
p n.
b I
is m
w om e ese.ne.
l w-l W
=-
I i
i l
i i
i i
-f I
8 1
I I
i
s=
ne me=
Lo
,.=
Figure (,
Generauzad Source, Easonmental Transfer and Dose Model ( s) 1
_-~
,n.-
ADD NEW SECTION( 5.1.9 1 @ ON PAGE 25.
p b I* Y 5.1.9 Releases from accidents.
Althoughthelargestreleasesofthistypewouldbeexpectedtoj)bein the form of a tailing slurry release (dam or embankment failure' or the result of a tornado occurrence, smaller releases can result from fires in the milling circuit. The use of kerosene as a colvent to extract uranium and of propane in the yellowcake drying process has resulted in seven known fires at uranium mills in the United States.
Offsite c'ntamination has resulted from dam / embankment failures (Ref.
o
[11] Chapter 7). No significant levels of offsite radioactivity has been detected as a result of fires at uranium mills.(Ref. USNRC, A Regulatory Report on Emergency Preparedness for Fuel Cycle and Other Radioactive Material Licensees, NUREG-1140, June 1985.)
5./1.10 F1 oding of tail as nds Q) e py ecb y nic come c,;.Ac ca i
e i
9
, The critical pathways (i.e. those that produce the highest dose to man) at any particular site are dependent on the local environment and habits, and have to be determined for the specific site af ter a thorough rurvey of local conditions. The inhalation pathway is generally more ig ortant in dry whereas the pathway through surf ace or groundwater will -laa M4CsC4,
- :ily el to in wet climates. The significance of the different pathways may, however, change rith time.
S.2.f Atmospheric pathways The radionuclides of most concern for atmospheric pathways are gaseous En end its daughters, endt airborne particulates of Th, Ra and pb.
Health impacts from an result from inhalation of its products of radioactive decay and ingestion of its ground-deposited daughter pb.
Because En is a gas, it may be transported over considerable distances and subsequently expose large populations, albeit to extremely small radiat%on increases above background. The levels are normally not detectable from variations in natural levels short distances from the source.
O O
Exposures due to the radionuclides Th.
Ra and po of the same order of magnitude as the IctP dose limits have been reported as occurring in the immediate vicinity of an inactive tailings impoundment [29).
Airborne concentrations of tailings materials significantly above background at distances of about 1 km of more have been reported (30). Although the I
greater part o* the windblown particulate releases from the uranium milling operations tand to deposit near the site, the fallout consists mostly of the coereer particulates, such as tailias sends. piner, more easily respirable particlates are transported longer distances, but it is estimated that most of the perticulate releases should deposit within en 90 lua radius of the mill 111). Appreciable espesures are, however, genernity limited to within 0.5 km of the tellings tapeundsents. Because of depletion of particulates in the air by deposition, radiation does rates decrease rapidly with distance from the seures.
5.2.2 Atmoepheric and terrestrial pattways Deposition et released meterial from tailings any also land to enhanced 0422y
~.
. ambient radiation fields in the downwind environment. This contamination is noenstly related to wind transport of esposed dry or freeze-dried tailings.
The depositad material any be resuspended and give rise to inhalation exposure, or to enhanced concentrations in the surf ace-water if the asterial lands in water.
The gasuaa-radiation dose at 1 m above an uncovered impoundment containing tailings from the milling of a 0.2% U 0, equ helent in ore 3
would be in the order of 1 ared/h according to the formata [31]
Camma dose este (wt/h) 0.093 C jg = 2.5 C PCi/O
=
ta where C is the Ra contration in taili in units of becquerels per gram, and is the concentration in picoeuries per fg geme. Strict contral of public access to, and prevention of the use of, the material for building or filling should therefore be exercised.
5.2.) Aguatic pathn ys Radiation exposures to the public due to Ra in surface waters are mainly through drinking water, ingestion of fish or other aquatic biota or through the ingestion of animals that have consumed contaminated aquatic biota or contaminated water. In areas where fish is a major source of food, the fish and water pathways can dominate legipuussgWury. Similarly, other aquatic X
organiens such as shellfish and plants can be eaten and contribute significantly. Anlasis, such as moose, which feed on aquatic plants do not usually asLi a mejor contribution. This is beesuse b y are mobile and only feed in W fffected area for a proportion of the year.
The radium concentration la fish flesh can be about 4 times and in fish bones 100 times the respective concentrations in water 1321. However, because of low uptake of bones by humans, the cot.tribution of radiation dose free this motorial is M.Mv4eue117 unimportant. Because of W retardation by soils of radium misestion gg through groundwater, radiation exposure via this pathway is also usually
-.&in u liq3* %eli A4 4~ W LiL U8~
W YtN NbU Y k Vh. Sca. /
}q f
5.3 Factors controlli release of pollutants 1
04221 I
~v
-_n--,---------,------,--.n.r,-----,,---------n,,n
.-.n,,,a-
.,..,.,n,---,
...,e--,.nn n-,
,,,,w,,m,,,.
~,--n
s e
1 5.3.1 Sub-surface transport of waterborne pollutants Tailings liquid and water added by precipitation can infiltrate the ao11 underlying the impoundment. This seepage may contain pollutants originally present in W tailings, for example radionuclides notably Es. some chemicals introduced in the mill process, and o W e non-radioactive pollutants originally contaLned in the ore. Also for some sites, which for a time were abandoned without any controls, other contaminated asterial may have been introduced by unauthorised dumping. These radionuclides and other contaminants may migrate into N sub-soil, and be transported by ground water and may eventually discharge to the surf ace. Seepage flow is determined by the hydrogeological properties of the sub-strets, including gradients, perusabilities, porosities, fracture densities, etc. After final stabilisation of the tailings impoundment area, the amount of water likely to continue to percolate through the tailings depends on W atmospheric l
conditions, structure and geographical location of the area, the infiltration i
potential of the cover placed over the tailings and the long-tern stability of GIJo k M JM ga be the cover.
uany sub-soils are capable of sorption of pollutants dissolved in seepage. The geochemical properties of soils, namely Wir neutralising and ion-exchange capabilities for N radionuclides as well as other pollutants, should be considered when selecting a site for W tailings impoundsent area.
When stabL11 sing a tailings pond for the post closure period, the natural subeolls' attenuation or buffering c4PebL11 ties should be considered in the decision to place a liner er to dispense with one. The ultimate criterion should be whether groundwater resources are threatened by W stabilised tailings yfle.
5.3.1.1 metardation by sell ion exchange The eschenge capacity of a geological medium for a given ten is cosmonly described by W distribution coefficient, ee E factor. It is W ratio of g
the number of ions per unit of weight of the solid sorption mediuss to the number of ions per unit of volume of the water.
For certain enions such as bromide. W re is no absorytion on the solid phaseg in such cases W transport of the enton relates only to the water movement and melocular 4Lffusion through the water.
0422y
s Civen a horiscatal barrier of thickness H, with a vertical seepage velocity of the water U, two characteristic transport times T fot ions having Me EtW eMA to v4 wrH !, oc.
esso exchange should be considered
/'
4 j
~
T, convection time
=
2 T
=
. diffusion time d
- 2 where K is the effective vertical diffusion coefficient of the ions through g
the liquid in the pores. The two times are equivalent if U = K /H; if U<K /H, dif fusion is the predominant process. This means that even with no g
seepage flow there will be transport by diffusion. If UzK /M convection is 1
proJoninant.
In the case of ions havingCe -
Q lon-exchange capacity with soil, K
Lon transport N ' '
retardation can be expressed as X
l l
l A =
1+
(
- )
pEj l
1 where e is the effective porosity, K4.s the distribution coefficient of the ion, and p is the specific gravity of the solid phase. The above characteristic times then become s
[
T
= A g
\\%>
g J
Th. E, for 22'ne = r = stir r
.h n i of 0.1 to 10'.'.s.
2 in J
aany soils (A being in the range of 100 to 1000), with retardation becoming M'
highir efficient. Values for other nuclides, such sa Th U. Pt and Po are QM.ven high.r.
Th. E, for v.cr 1.w =1= tty =r me t.sen as an.us 0.01 M 0 " *M4 kJ e 6( 2.j(%f N D ia Q DC M>
& d,t. x Os227 kie J. hlA
% % av,.A,lf4.d/LUcd b A Ra -n.Ja 4A a UDa
' O.03 a, which is the ion diffusion coefficient D divided by the tortuosity factor where the diffusion path is not a straight line.
It can be shown that the time taken for radium to be transportad through
[
a geological barrier of material having a high exchange property such as clays or bentonite could be several torndred tLaos greater than that for water or ions that are not adsorbed. When transport by diffusion is faster than by convection, i.e. for M e (E /U)A, the time of penetration of an ion is g
proportional to the square of the thickness of the berrier, T = (AN)2j For water movement by convection (Dercy velocity, i.e. for M = (E /U)A, g
T = AN/U), the time needed for the concentration to increase from 25 to 75% of 8
the equilibrium value is in the order of (&/U)(E /U)
For a 100-ce g
thick barrier of sand aimed with 2% bentonLte (K e 10 a.ks" ) in
~
wetconditions,andaseepagevelocityof1a.a'f,thepenetrationtLasis in the order of 700 years.
y The above description iny11es a heterogeneity in the subsoll surf ace, conditions whLeh creates signLficent complications in modeling the movement of contaminants in groundwater. The presence of multiple types of materials, each with its com properties, such as permeabL11ty, tendency for fracture, etc., makes the accurate prodletion of groundwater transport extremely i
uncertaiti. Eeny countries are supporting efforts in computer mode 1 Jing of I
weste repositories, such as mill tatlings. Existing models can be used for envirotuonntal asseseaset. Newever, me one esaputer andel is at present considered sufficiently capeble of secuestely' predicting th*8* processes foe the required safety assessment. The United states is relying on monitoring em p ir-ete.
tgu=t--- such as liners and aquifer restoration
[
strategies to sentrei release from mill tatilngs sites h LH]l)SEPA Hou:hPad m2. SLhh W Wd S "
k'**y I
a %.
w e ponere p,. # gN uw-e S Isa,M n. 0 n % 4 Dr.
Two methods (341 any be ed to limit the less of 1Lguid free a
$d h tap ings i n ' --t by seepage b firotfWyes contatraent of the liquid l
in the impoundsent by weLas a low-petissabL1Lty berrier i.e. ene having a value 5/L, (transmissivity) as leer as practical. A barrier with low transelssivity y
any be provided by constrvetLag linero of synthetic or natural materials, or
(,
by loesting the impoundment on a natural stratum with low perissability and L
(
S adw {'t? 'l, '
(G Q
0C22y
_m
y.-------_ww
_y-m,,,,w,
,m,muyww,,
. including a low-permeability core in the dam. Under these conditions, tallings any be layounded in a wet state with resulting low seepage rates so long as the berriers remain intact.
The options for seepage control using this approach are apparent from the general equation for flow rate q, along a seepage path given by D
i e
(
4>)
where k is the effective permeabilityylong the path, h is the hydraulic head dissipated along the path, L is the path length, and a is the effective cross sectional area of the path.
Forneability in nature varies from greater t.han 1 ca.s for coarse gravoit to less then 1 1 10' em.s for compact line clays. In view of the direct relationship between seepage rate and permeability, this is a pripe factor in control of seepage. The average path penneability any be controlled or modified by site selection and use of low-permeability liners.
The unsaturated zone encountered by seepage water as it percolates down to the water table has en oppreciably lower permeability than is encountered in W saturated sone 133) $. If the seepage release rate and volume are 4
1arge enough to result in saturaties of b norme11y unsaturated sone, its permeability agn increase. With low seepage rates and volanes. W reduced
@).
l permeability theracteristic of the unsaturated some any considerably increase the time period required for seepage to reach the water table and begin LLs lateral migration evey free h sita. If the asterial La the partiv saturated sees is below its seisture-belding sapecity then a flatte volume of water will be retained. Reese, for at least eene period of time, seepage may be obsorbed withia W peres of the soil.
Rydraulis heads directly affect the seepage rates. geduction of W water level height la b b r '-.t reduces the seepage less rete.
(,
0,.
Nydraulis heedy resulting free both the regional flow pettom and flow from WA3k Y the tapounement, influence the lateral migration rates of the pollutants.
kN w 'g Nyde-ti. he.d. any he it.ited by do,e.iti= dry t.ilngs.c b.e,ing the.eter M
1e m.. ciated. m m t. m se m. 1 ve1ue.
[ pen -
Q 0422y
32 -
Seepage rate is inversely proportional to seepage path length. The 4 41 47 letter is theg)./d 7"M....... :1 -. the flow path from the impoundment to the point X
of discharge to the surface (e.g. via a spring, well etc.).
seepage rates can therefore be reduced by increasing the seepage path length by suitable site selection and design.
I The soepage velocity U may be determined from th* general equation 9
U =
as where e is the effective porosity and q and a are as defined for the flow-rate equation. Where soepage velocity is low and flow path lengths are large, the time taken to complete the flow path any be long (e.g. thousands of years).
One potential concern in using a low pecneability liner is the possible "bath tub" effect. This results when the stabilised tailings cover is signLficantly more perunable than the 1Lner beneath the tailings. Depending on N environmental conditions, the tailings water recharga rate may be
.gr M ignificantly greater than the seepage rete. M*f:_;t 'tir '; -;;__;
b p :2m,_.m.. _
..., 7.. _, _. =:: ::
- : u ^. '.., _... L... ^......, -.. - ^
--. it 'ney prove to be a significant }
p&
obot..le to ions t.. st.buisauen offort. at m u.e of.iosure..ro nov.
g equipment onto tallings '7-x ' ts. N tailings have to be sufficiently I
tenso11 dated to bear W weight of the equipeont. as well as the weight of the t
cover meterials without undergolas estensive differential settlesments. In these seees in site dewetering een be appliedt e.g. dewetaring trenches.
insta11atten of under drains, electro-eemosis (34).
there the site sendittens are suitable and the regulatory euhrity is in agreement, systems any be used where the permeability of the liner is greater then that of the sever. In this situation, the resharge rate can be controlled to be less than N saepage rete from N liner. The liner can
]
aise be plased in sembinetten with other fumational layers, such as a neutralisation susliner. er a drainage er perselation/ hydraulic break layer near the aevoc. Such toebination technologies can be used to mitigate, to I
some extent. W impacts te valuable groundwater resources.
1 04221
_.--__-,m.-e-----_
._nm-,--
m m -,-w nm y e,
_,w.w.,,r,---
. l l
A second approach to reduce seepage in acid zones is to limit the water volume that can soep from the Layoundment by placing tailings in en essentially dry state. In the absence or near absence of soepage, the contaminants transport rate will be limited. Since the soils through which 1
the contaminanta migrate have a limited absorption and neutrali ing capacity
[
137). W 1ess contaminant discharged to soil 11er will be the sone of A
sub-soil effected by contamination. Lastly, if tallings are in an essentially dry state when the impoundment is stabilised, the need for long-term r.onitoring for liner performance or tailings-solution discharges may be decreased. This approach any not be as effective if fluctuations of the groundwater table bring the natural groundwater flow into the tailinas s
Psut ntm wa mo tLL 1
impoundmentlevelj The long-term impacts from suchp ri'
- y " 1 "_
/
significantly affect groundweter quality even though the tailings are placed in a "dry" state. Liners er relocation of the tallings may need to be considered, but monLtoring should still play an Laportant role,in these instances.
5either of W above concepts is absolutely effective in ellainating or delaying transport of radionuclides from the tailings Layoundment. All liners possess a finite permeability and ' dry' taL11 ass contain some solsture. The effective lifetime of synthetic liners is 1Laited. seepage is dependent on W attenuation characteristics of W matarials through which seepass occurs (34, 35. 38), and W 4Llution and dispersion accountered along W flow path 137). The amount of seepage lesses that are aseettable are highly site-specifie. Time sensitivity of the demnstroen envirosusent to releases of contaminents any Lafluence permissable releases.
5.3.1.3 Leseking ac2 trol Leechlag La the process of aebilising sentasunants sentained in the tallings by water aseestated with the taL11 ass er freet water contacting W tai 11 ass from natural ocurces. Leechlag saa be reduced by either limiting the water percoletten through the t er by chemically fixing W contaminents. Inflow of fresh water tan be minimised by locating the i=pach.t in a smell catchment arse and/or above the fluctuating water table and by sentouring W stabilised slopes to prevent ponds forming. Anothe r 04227
. anthod of minimising percolation is to place a cap with low permeability over the tailings. Mobilisation of some contaminants in the tailings any fregaontly be inhibited chemically by adjusting the pH to be alkaline, in the case of seid-leach tailings (Table 6).
5.3.2 Airborne transport of pollutants Airborne transport gives rise to pollution free radionuclides both in 1
particulat,e and gaseous form (sections 3.1.3 and 5.1.4).
As the surface of the tailings pile dries out, salid tailings are subject to airborne movement, resulting in dispersal of particulate radioactive contamination beyond the site boundaries. Wind erosion and mechanical dusting are mechanisms by which particulate matter is lif ted for wind dispersal, neden and other radioactive sesos released from W tailinas surface can also be carried by the wind for significant distances.
3.3.2.1 Centro 111ng wind transport of particulates Three major conditions contributing to wind erosion and dispersion of tailings are: 1eeee. finely divided. dry tailings or cover materials; a smooth, bere ta111 ass er oeil surface; and strong wind 139).
)
There are various fasters te seasider la sentrolling wind dispersion of tailings during the siting. operating and poet-stabilisation phases.
selection of the i ;r"
.t site is e anjer eensideration to minimise the wind 4ispersten potenttal met only esting the *peratins 11fe ef the mill but.
{
p
.2 W importantly. in the longer teraN The peinelpel features for n
seasideration co prede inent wind diresttens and veleelsty,.nd shots. ring provided by topogrepMeal features. The frequener diste h tien of high wind
.G tosity is more Lupertant them the everage wind veloetty. aurias W h
ting phase of W mill, en the other hand, a windy eres is a favourable l
characteristic that taproves seter evaporation, which any be an Laportent sensideration in some W r ' st systems.
The particle else distributtee of taL11 ass is en important factor in W l
potential for wind dispersten during operation of W mill. For emangle, a hoop leech operation Wt results in tailings of a rotatively large particle 06227 l
t
C TA&LE M 3 emmary of I,abetstof7 Studice of M t'ot'en Pet'at'*l ( @
L ei, Immonitised at Largely Ness.Newsre! pH Immetilised VWwee (4 8 8 l) dement 65pH VWwee bwt Mobile el Insuf.
er between 3 i Les pH Ysivw f.sient Constituent and 88 (pH below 4 l)
Mobile D at a Cl X
.N O X
A1 X
X
$0, v
X Cr X
Mn X
re X
Ce X
be X
Cu X
Zn X
A4 X
s.
X Me X
Ag X
Cd X
86 X
I Pb X
Ra X
Th X
U X
i I
i I
I 1
Il I<
$4<
m l
.g...
02<
0 0
0 2
3 4
5 m e w es sw eg 4
% e w eve w=== senn =%
e,areee m amm= ww 1
fl0J Jel(<enftaramont forter of a neaknee seremotament.
l
l
- l b
7 P
i J
i 1
4 l
I f
u I
i l
4 l
I e
l i
1 i
Recommendat(cn i
i Starting with the mark on page 36 with (a) Emanation factor, the l
l discussion could be deleted to section 6 on page 40. I would suggest j
refereticing this discussion, but I don' t know where it came f rom. It t
appears to be technically correct. albeit someone archaic in notation.
i 1
My discussion in Appendi:< B-3 covers the topic f rom the point of view 1
of designing a cover to attenuate the release of radon from tallings.
1 Whether we need to educate the reader on how radon forms in the pore j
spaces, and so on
, may not be all that valuable. This might be j
something the technical committee can consider.
e p
l l
)
J I
, size will not have as great a tendency for wind dispersion as conventional mill tailings impounded at the same site. Methods to minimize wind dispersion during mill operation include keeping the tallings wet or covered, or adding a chemical ctveting agent or earth to the tailings surf ace. Consideration should be given in planning tallings disposal programmes to using methods which provide immediate covering of tallings Lapoundments, since these techniques will help in controlling airborne emissions during elli operation.
The U.S. Nuclear Reguistory Consission suppor* e 3 r* earch s.ich i
prLmerily investigated strategies for controlling particulate releases from tailings impcundment and other milling storage or staging areas. Aside from actually covering the tallings with a soil cover, most thenleal stabilisers were ef fective to some degree but only for short duration times (on the order of one year). Care needs to be taken in evaluating whether frequent applications of these stabiltsees may affect the chealcal inventory relating, to contaminant mobility with respect to groundwater protection. The USWRC has published c final regulatory guide on source ters estLaatio 5, which also gives guidance on reducing particulate releases (40). Other supportive work relating to restrictLng wind blown releases during operational and drying phases of the sites can be found elsewhere, for example reference 41.
i After mill operations, wind dispersion can be prevented by adding a cap Q
r cover to the impoundment area. A revegetation programme, or the addition of riprep in arid areas, provides protection for the cover material. t.ow effective slopes should be used when designing the final contours of the 3gAf*
La, nde.nt area.
h 3
N S.3.2.2 Centrefling raden releases Rodeo-222, the gaseous decay daughter of Ra, will escapa from the uy solid taillags perticles into the pero spaces between the particles p The gas is them free to diffuse to the w rface and eecepe to the staesthere (section 5.1.3).
The redon half-life is suificiently ahert (appersimetely 3.3 days) that 218 many of the gaseous reden atoes any decay to se114 pe before reaching the surface. The self-confinement f acter of a tallings plie is a measure of 0422y
36 -
efficiency for retarding the diffusion of Rn through the impoundinent pore spaces. The greater the retardation factor, the greater will be the trapping of the radon gas in the pile as it decays to solid
'po.
l In the diffusion cessjthe solid phase of the medius may m -
4 4#3a4 adaampA 4.-
'l-rad on 1 f.:
1,.f l....
2.-. i -- - -........ wi th i
g a resultant decrease in the rate of release.
f leoisture in W tailings plie is known to retard the diffusion process.
Under totally saturated conditions no air remains in the pore spaces and 222an diffusion is through the water medium only. The molecular diffusion coefficient for an atoms in water is approximately 4 orders of magnitude lower than that for air, thereby reaulting in a marked decrease in redon emanation.
Changes in atmospheric pressure can cause temporary perturbations in redon release. Generally, if the pressure increases a temporary decrease in redon emanation will occur. Conversely, a drop in atmospheric pressure will cause a temporary rise in the redon flux.
Most of the discussions about redon releases Wt follow in this section pertain to homogeneous taillags media. Mowever, in practice, a layering effect occurs which in certala cases can affect W diffusion process so as to reduce the :tmospheric radon releases below Wee sapected from homogeneous tailings.
In the followims sections.the general relationships (42) that affect j
redon rele6ese rates from homogeneous taillage piles are described. An example in the use of seme of the principles discussed below to settaste redon release retas is given in Appendix 3.
(a) amenation faster The amenation facter t is defined as that fraction of the redon atoms produced that escapes free the oe;td phase to the pero spaces. Radioactive decay of gaseous redon to its solid daughter products acceents for the i
fraction that f ails to enter the pere space.
i a
i 0422y v-
-- - - - - - - -...v-,
w
---m e--
.,-y..--,e
. The maaber of En atoms produced per unit time in a volume V of medium is A Y atos/s where A is the radium radioactivity level per unit voluse of medium (34.m"3).
Therefore, the number of an atoms per unit time entering the 222 pore spaces acjacent to this volume V is t(Vatoas/s M
Given the effective porosity e of the medius, the solid phase in V l h then occupies a volume (1 - e)V at.d the pore space a volume of eV.
From
}0 Mhis an estimate of the emnation f actor can be made. For moet uranium p
h tailings g
u d % e f n m, e is in the order of 0.03 to 0.3.m meML h
N (b)
Self-confinement factor MW XD A
yphk T -
The self-confinement factor s is defined as that fraction of'the endon q )+8 atoms entering the pore spaces that eventually reaches the atmosphtte without decaying within the tailings pile. With a tailings layer of height (thickness) H. and eurface area 3. the rate at which an atoms enter the pore spaces is given as tA M.S. stoes/s (see sub-section 5.3.2.2a)
If the thicknees of the ta111 ass layer is small compared with the diffusion enlamatten lensth (defined later), then the rete of En reteaee fros the surfaea will ateo be ta M.s. stoes/s h
If the thickness of tat 11 ass is relatively large, meny stars w\\11 decay to 218 solid pe and attach to the solid phase of tailings dur'.ng dif fusion.
Incesasing the thickness of the tailings therefore provit es increasing aelf-confinement of the redon atone, and the net release to the atmosphere will be 0622y
l stA,H.S.
stons/s g
The self-confinement factor s is a function of the tailings thickness H, and of the effective diffusion relaxation length H,gg, for ne pc m s d ua.
As derived in Ref. [42) one has s(H. N,gg)
(H,gg/H) $ hyperbolic tangent (H/H,gg)
[
s =
=
The dependence of s on H/N is shown in Fig. 7.
gg With a dry solid phase and with no adsorption of redon, the effective dif fusion relaxation length is given by the formale Og off g
222 where D is the molecular diffusion of the am atoms in air in stuera g
- tres,er s on (o,= 1. 2 x 1.-5 2.s-%, is the tortuosity factor, accounting for the diffusion path not being a straight line but rather one which passes around the solid particles (thus incrossing the effective length): and 1 is the decay constant of an(t = 2.~. i 1.
s
).
i
~
The tortuosity f actor id usually la the orter of 1.3 for send but may be much inrger according to the train-size distribution and shape factor.
M for e dry medium is tyyissily in the order of 1.5 to 2 m.
If the g
thicauws of the pile is large coupered with H gg, then the samber of atome
]
rsleased t7 the pile is given by i
tA N I *t**'#8 h eff I
os22y i
4 l
l
l l
l 1 (c)
Adsorption The solid phase of the tallings pile may haie naturally or artificially introduced adsorption properties for redon. This implies that the atoms of redon will distributa themselve. in the gaseous phase and in the solid medium.
l If N and 3, en W cecontnum of ph aW in We air and in the solid phase, respectively, then in the equilibrium situation
'S " ad A
.here.g is the adsorption coeF(icient. In this esce the relaxation length beeones A
- ett
,d>
.u.
Therefore, edsory W on can have W sffect of reducLas the relaxation length l
end hence reducing the Rn release from W tallings pile. For dry chartoel, for instance, M,gg any b in the order of 0.1 m.
l (d) lecisture effects The aristence of noisture in W form of a wetor flia around W particios and bridges bet. mn them will reduce W emanation of moisture for the followins reasons:
[*bW(i)
The emanation faster s will be Jecreased, since the redon e.caping the
<b y
solid phas' stu have to be desorbed from W.ater before entering the i
h air thsee. This asen. th t more reden will decay in tha solid and 1
W
.J Y
A ttwid phas.. mn uld.etur in der t tunes p)s.16)gui m tort
.it, f.eto, T wiu incre..e, uaur u mre e 1,t, as isYy,Au.ualinm.edtu,1 i
iM urb.or,uon end d..or,uon,res s.. of r on et,u a.
n.e 1e.s.et.c turat.d - s le.s - r.ious d
- 3. u e em.n m u,uu
(
..gM A th. g.s phas.s wiu r uts in a reducuon in the.ff.cuv. duru. ion 4-coetticient. If k is defined a. the estio of the redou e contration in
}N p yO'O D "
a F
og,,,
. the 11w id phase to that in the air phase (k has a value of 3 at 17'C), and the saturation factor s by Volume of water Volume of pore space then the effective molecular diffusion coefficient for an atoms in the medium is given )y (1-s)D + ksD
- tt e
1 - s + ks where D and D are the molecular diffusion coefficient for an g
y atoes in water and air, respectively. W effective relation length is then given by l
l 1 - s)Dg + kmDy j
n (1 - s + ks)
If caturation is templete (e =1), no air is avellable in the pore spaces and only the melocular diffusion in water has to be considered as H,gg reduces to W
"off g
Sisse D se D I 10, M,gg as % ed h h eNer of 1 es.
y g
Values fee k at various temperatures are given below T(*C)
G S
10 15 20 25 3C 35 40 ed, K
/
4.54 0.41 0.34 0.20 0.25 0.22 0.19 0.17 0.16 4
g f., }N 3 M taperimmetal resuite Ladicate that for a saturstion coeffielent s Q d tween 0.5 and 1. W relaxation leaath is in N erder of a few tenths of a 8
ne thicine.s.f
.t s.iu.es,ue. L..co.s.r th.n x,,,i thus st e j,
._ essa.
ty;)
[g yedon at escap.: free m 114 phase will decay beto n reaching the free atasephere and t1 mas within N 1apoundment. only the radon
- o. O emanated by a surface layer of 6 tent of a metre thich is released to it p%
m ata..., _ in s uusuon..ere
.ois,u - i.
in,ai..d.
1,. 43 04227
7 TABLE W. ATTENUATION FACTORS FOR COVER MAliRI ALS 08 1
2 3
4 5
6 i
H,
\\
l Os
- 0. $6 0.47 0.18 0 07 0 025 0.009 0 005 0 001 1
0.41 4 37 0.13 0 05 0.02 0 007 0 002 0 00I 2
432 0.16 0 09 0 03 0.01 0 004 0 002 0 001 pc. g m.
is s== P.aos u,
,w.~ /
/M,i/[4j'"'
/
- s%%%
/
I
]
f
,, _., c, c.
l l
f,,,c%.a. se,
caem oene.<see t
i s
_Growed St. (2 3% G'.eet w
(
m F1 CURE S4** *th EI*8" 'I
- TIW
\\
l 1
1 fw.
.u-w) f yMW-Q f./
()
L, I).
a
. (e)
Atmospheric pressure ef fects
[p, 6.
Changes in staospheric pressure can result in tersporary changes in e on exhalation rates. Usually the duration of atmopheric pressure changes y
6 f is much less than the half-Life of redon (3.8 days). In this case, the effect of a tyclic change of atmospheric pressure of strength 1 Ap is to release j
(all the radon emanated in the surf ace layers o' tw stoc L
The release due to the pressure change is therefore M (AP/p).n.
.p f r-w
,A,s.n.
atos/s g
- l g
snd e total release may be satinated using the for-ula
(, 0 V
}'
f p# '
, i,,.. x.-
x at
,s i
. x,,,/
s A
P P
(f) maltilayering of tailiras The above discussions relate to homogeneous tallings media. Hoere ve r, as the liquid and solid effluents are discharged at the surf ace of the pile, a segregation any occur in deposits according to the grain-sise distribution, the teareer partLeles having the highest sett1Las velocity. A multilayer ettveture is ersated, with a veritical variation of seisture and porosity.
This seuld in sortain seses roeuce the relaxation length to the order of the thiennese of eesh layer. M th2 M* b CNuf ] b N/"a#
Mdw3 PMr*.LN '.
~
(s) herface. 4 etation atracts Use of surf ace vegetation is en acce9ted practice to stabilise the surface cover of a tailings pile. xowever, the vegetative cover may have some disadvantages (43, 44) ro*ating to redon rettese, namely:
(1) increased (??) exhalation or transport of reden from the soil to the (ll) '
b l
got atmos,here by the,egetation h (2) reduction of soll moisture, with a reeuttant change in the reden diffusion rate, as a w it of noisture transpiratior,by vegetation 06227
. (3) increased permeability of the soil as a result of plant roots.
Further research is requited to understand and quantify these mechanisms.
(h) surf ace cover ef fects
& cover of clayey soil any be used above the tailings to reduce the exhalation of reden. The amount of reduction depends very such on the moisture within the cover, as discussed in sub-saction 5.3.2.2d.
A complete i
water cowar, for instance, may serve as en ef fective barrier. A reduction of several orders of magnitude of redon emanation due to snow and ice cover har been reported (45). Where climated perm 1$ leaving the t-ilings pile in a X
situation where permanent frost can form may be an effective way to EI l
redon flur escaptng to the atmosphere and reduce releasas of liquid effluents.
i The main parameter govem*ng the effectiveness of surf ace cover is t'he rette of thickness of the cover h to its own diffusion reti stion length i
y M.
Other parameters are the thickness of the taillage pile h and its 2
relaxation length N.g If F, is the flux of the pile without cover and F is the flux it the j
pile is coverM, a confinement facter G is defined by F
s.
'e If g is las'5e campered with IL, G ney be approxiasted by g
,i e e es, (-ta,/W )
]
Yelves for the attenuation facter for cover meterials derived by the method shown in sub-secties 3.3.2.2b are shown la Table 7 for various values e
of h,/M, with M /M eque,1 to 0.5. I and 23 and for a thickness of the j
y 2
pile h terrs compared with M (s Ngg/M ).
If for instance the g
lMS' W.
flux is to be reduced by a factor of 50 (C = 0.02) one needs h #N2 " 4' F
2 g
g For a dry cover, this requires 6-g a thickness. If the saturation coefficient j
g g
is =e,t high.nough, cover of.n!, about 1 - 2 m is required.
i g$f> gle at q Lc w kn + %)$
S ~" * &
- w M
A.s.. &
d, a:Aw
- 0422
s In the case of a cover made of succo'esive layers of various materials (1) and thicknesses (h ), the total confinement factor is the product of the g
confinnent factor of each layer g\\
[
h
' tot j
(
"i /
IHset&
7 6.
snvironmental considerations in mill process selection The process used for recovery of urentum from the ore is dictated to a large extent by mineralogy (ore type), geology and economics. However, since large quantities of chemicals are required and these chemicals may have a marked Lapact upon surface er groundwater quality, selection of the overs 11 stil process should not be made solely er..tydrometallurgical grounds.
Subeitutions for the u.are toxic reagents should be ande where possible. Aleo.
some modification er addition to the mill process scheme any be feasible which would reduce either the inmediate er long-tera enviovemental impact of the a
taillnes.
I 6.1 process options The t.,,rece...
st used for di. 1 ring uraniv. f-the ere are 1
acid er alkaline leeching. These are followed by a variety of concentration
{
and purification teslutiques to peedue, a uren'ma car.eentrate product of desirable quality.
The said leseking process, which uses sulphuric acid as a teachant.
has the advantages of high urentim rwevery, easy process tantret, genere11y modest reagent requirements. and ieer power censusption. Its asjor i
disadvantage is that it dissolves risnificant amounts of the undesirable constituents in ere. This than requires sophiaticated leech 11guer treatment by ion exchange (II) selvent eartreettai (51), or both, to upgrade the purity J
of the uranium in solutim and to alleer precipitation of a uranium concertrete of acceptable quality. The sulpb ria oc1G 1eech procese also frequently has 4 high requirement for fresh water. In an acid c11aste this additional water is j
l 0622y I
...---,,_,-.--.n._
_,__,-_ n --,,,, _ _, - ~ - -, _. -----_
n,
-n-
f 41A TN b:
U $. hb
~
Ab pce G.
Au M ALg.
k v N N""
G /Qd-3 G3 (an,eb.es.*brsWT t
fi A
su P V.?
)
I 5.3.3 Alternative Nethod for Radea Releases used in the United States of America.
i Appendix B-3 describes the methoc ology used in the USA for predicting the resultant tailings flux from covered tailings. The methodology is i
the result of research efforts by the USNRC and USDOE during the late 1970's and early 1980's (Ref. 1.2 ). These research efforta produced computer codes.which can be used with relative ease to predict the j
necessary thickness of multiple 1 yers of earthen materials in the design of stabilized uranium mill ailings disposal areas. These codes !
a
(}
are identified as either RAECOM ( ee Ref. 3 NOREG/CR-3533] or RADON 1
See Ref 4 USNRC Regulatory Guide 3.64).
l This methodology reflects the sensitivity of the radon diffusion I
process to certain factors such as the long-tors moisture content of the earthen meterials used. Furthermore, the RAECOM code allows the j
user to factor in cost / benefit considerations in the design of multi-layer stretegies. Appendix B-3. subsection B-3.2 allows a user j
to approximate the exact solution to determine the required thicknass j
needed to meet any redon flux performance criterion, in the case where i an operator is using a single uniform layer of earthen cover material. ;
5.3.3.1 Factors affecting the reliability of the performance of a
l earthen covers.
The deeign of earthen covers for radon attenuation depends on the values of a variety of parameters which characterise the tailings and j
cover satorials. These parameters include the thickness of the cover.
as well as of the tailings: the overall radium content and emanation rate of the tellingst also the density, compaction porosity and moisture content of the ta111 ass and the cover. Some of these characteristics are factored into specific parameters, such as the radon diffusion coefficient (See Appendix B-3).
Of the above characteristics of the cover material itself, the long-term moisture content of the cover is the most significant in characterising an earthen cover material's capacity to attenuate radon gas. This can be a measured value, or it can be estimated from other soil characteris';ics (See Appendix B-3.2.1). It is important to nate that the condit'.ons under which a cover material may initially be place do not necessarily represett the long-tera performance of a cover materia). The cover may at through drying and freese/ thaw I
cy:les, soil drainage and evapo-transpiration, which over time can change the coadition of the cover with respect to radon attenuation N Ret 3. USNRC Regulatory Guido 3.64). These processes can lead to a j
reduction in the effectiveness of a radon barrier. For this reason the i
-... ~. _ -...
f.
+ 2 at M
t wilting point (that soil asisture tension which prevents say twrther yield of soil moisture to plant root systems) is often selected as the 1evel to which the noisture centent is tied. Appendix B-3.2.2 1
l describes an espirical formula to estimate this long-tors moisture value.
Other values for the parameters to be used in predicting the redon cover's performance can be approximated from either field or laborata y
--a urements, or can in turn be estimated from some basic soil condition Ref. 3). This approach assumes that the performance is based on a yearly basis. Advoctive factors, such as diurnal / nocturnal barometric variations and seasonal thermal effects, are not explicitly considered, because these effects are negligible in comparison to diffusion on an annual, averaged basis.
5.3.3.2 Regulatory framework for radon releases from etabil; zed uranium mill tailings piles.
The uranium mill tailings piles are separated into two regulated classes. The first consists of 24 abandoned mill sites and are treated as waste disposal sites to be cleaned up. The second class of mills are those considered to be commercially active and are regulated as operational sites with active controls fo. radon releases.
l For the abaadoned sites, referred to as UNTRAF sites, the l
stabilisation performance criteria for the earthen cover consists of a l
redon flux limit 20 pCi/a**2-s (0.74 3/m**2-s) or alternatively an I
off-site redon concentration limit above background: 0.5 pCi/L (0.02 5/L). The methodology for modelling the releases and transport from i
the stabilised piles is not limited to any mods 1 approach, although there are models available for such a calculation (MILDOS. See
(,2 h N URIG/CR-2011).
For the ooemercial sites, only the radoa flux of 20 pC1/ ass 2-s ( 0.7<
3/ase2-s) above background applies. In both cases the flux limit applies to redon levels above normal ambient levels charectoristic of the local area. In addition, the criterion is to be applied in such a manner, so as to insure that there is reasonable assurance that the design meets this numerical limit. This is interpreted to mean that the 20 pC1/ ass 2-s criterion is not to be a "best estimate;" that it la just as likely that the flux is above 20 pC1/a**2-s, as it may be below. The flux in actuality may exceed the limit in some areas and f all below it in others, but tne annual, surface-area average must not exceed the 20 pCi/a**2-s limit. It should be noted that this is a 1Of' design standard, so there must be a quantitative rationale that this UL limit is achieved by the design in question.
3
- \\.
ResearchSponsoredbytheNRCatPacificNorthwestLaboratoryfroa)/
Footnote 1. Ba telle Pacific Northwest Laboratory. Uranium Recove y 1980-1985. Final Program Review, January 1985.
_,,/
VF C I
l i
Footnote 2. Technological Suasary of the UMTRA Project. Technology Development Program (1980-1984). UMTRA-DOE /AL 200125.0000. Jar.uary 1985.
8 f
I I
g gdm L 3.
)
i L
t 1'
i s
)
I j
.i i
t e
I l
- l l
l J
i I
i i
4
,-,,,,,,,_.--,-,,,_m,,_
,n,,_,
_--y
._,-,g,n,_n
--,._,-,_-,,.,__y.
4
. readily evaporated from the mill tailings impour.dment. In a wet climate recycling of water from the tailings impoundment to the mill may be required to reduce the water consumption, but discharge of excess tsilings water to a surface water tourse after appropriate treatment is generally necessary.
Isetals leached with the ursnius may include vanadium, arsenic, selenium, nickel. tron, copper and a wide variety of others depending the mineralogy of the ore. The radioactive elements associated with ursnium such as radium, thorium lead and polonius are also dissolved to some degree. In the acid leaching process about 11 of the radium remains dissolved (46), and frequently about 50% of the thorium (47). Data on lead and polonius are aarse, but these are expected to be somewhat soluble.
L l
The major chemicals used in the acid teach process cou-only include sulphuric acid and sodiva chlorate or manganese dioxide in the leaching circuit; svunonium nitrate, ammonium sulphate or sodium chloride and amines, alcohols and kerosene in the purification steps and ammonia or magnesia for concentrate precipitation. The choice of chemicals used will have difierent effects on the environment.
fgjs The alkaline teach process uses a alxture of sodium carbonate (or
- ,' " )-
i lus ce b A te for in situ leaching) and sodium bicarbonate solution with t
s 0
oxidising agent to dissolve uranSun. selectivity for uranium is high, and A
Iq further purification, as is required after sulphuric acid leaching, is l
pcu+* d seestimes unnecessary. The urentie soncentrate is either precipitated d
f.,directly using sodium hydroxide or af ter purificatiou by en ion exchange or solvent autraction process. Radium dissolution (commonly 3 to 51 of the total MM present) is Senere11y higher then La the acid teach pro;ess, but the
- gaW W
j diseeletion of other radioactive elements is auch lower at the high pH of the leseheat. The alkaline leech process is usually acre surensive to operate and
- J u hi 1 or e.. ry of. vat 1. t. ureniu. th.n th..cid 1.aeh proc.ss.
i y,.;
.t.
Thererere. it t. eta-etected =nt== the cia e-stoa *r the are -ves h
reagent consumption for tha scid leach process excessive C,
b va kw 4 d.'d i
p, g4 A e 4 ~/,/ u u -b g Td' M Seth the "id and alkaline 1each processes generally require that the g
ore is crushed 5 a finely ground. In the acid leech process a particle size y%=* Y(4,. of minus 500 we is generally required. and quite of ten firer.
For t.he wre/* M tgN.
04221
k*
6 M W,
5[qg, Q
&gg W f& f i
y, h
6-
& gj &
f&
- alkallne leech the 0ere is generally ground to about minus 0.2 uma (501. less than 75 gas). This grinding largely detetuines the physical charactecistics of tailings.
One procedure that is becoming of increasing importance as lower grade ore bodies are developed is the heap leaching tactinique. With this technique the ore is crushed to less than about 2 es in size and piled in a heap or bed Ftsent t) en a,drai(n ped Uraniur. Le dissolved by spraying and draining a solution g-g containing sulphuric acid through the heap or bed. The reeuttant urentua-containing solution is collected from the pad and concentrated and purified in a similar f ashion to that used in a conventional sulphuric acid leach process. This heap leech results in auch coarser tailings with an appreciably lower vetor content then in a conventional mill, with attendant advantages for tailings disposal. As for the other leaching processes, the teached ore remaining will contain many of the radioisotopes originally
'Aa nontent, for example, will be largely present in the oce. The unchanged. However the diffusion path for thc redon in the lors*r particles will be longer and less redon will escape from a p rticle. Tallings can be managed in a menner generally stallar to those free c6nventional leaching proteases, but perhaps with esmewhat simple.' teclutiques that are poulble because of the lower dispersibility of the taL1Lage due to
- heir larger particle else.
6.2 Reducing environmental impacts The selecties of an acid or alkatine leech mitt process makes relatively little difference la terne of lens-tors radiolosteal inyect to W
[ en esent. Tottiass free ot h e pree cent 4La sLaller amounts of 226,,
and thus monate similar quantLttee of an. In centrast to the alkaline presees. h 11guld phase esswinted with N acid teach tallings contains more wreatus, brium, leed and polonium, which are avaL1able for possible tiensport outside the ta111 ass impounement by either seepsge or through discharge to surf ace watercourses. Sm.trolisattom of the acid tailings with line or some stallar suitable subetences before discherse will significantly i
decrease N armts of these materials in the liquid phase. The mobility of moet undesirable metal ions La seid teach tailings that have been neutralized at a pH of greater than 8.5 is met significantly different from that of those same ions in tailings from alkaline teaching, although the presence of carbonate in tallings from alkaline leachir-tends to dissolve some metals.
0422r
. Increased uranium recovery in the milling process will decrease the residual content of uranius isotopes with long half-lives, and the amount of ore to be processed for a given enount of uranium product.
') " UL l' The practice of heap leaching has the advantage that the resulting 1
p}
tallings are less susceptible to dusting than conventional tailings. Also.
less water is associated with the pile. and because further water entry could
-g M, be minimised as operation proceeds, the potential contamination of ground and surface waters is significantly reduced.
py 2.1 Radioisotopes Tallings solutions from an acid leach process contains dissolved j
i uranium, radium, thorium. Lead and polonium. In an area where precipitation (W*\\
exceeds avaporatLon. the liquid phase of tallings can eften be released to j
surf ace watercourses af ter cypropriate treatment. Where such a release is '
possible, current practice is to adjust the pH of the tallings solution to a rense of 4.5-8.5 to precipitate many of the dissolved metals. This pH i
adjustment is best done La the mill plant to allow for seed process control to maximise precipitation and rentention of radielsetspos within the. taL1Lngs impoundsent. The major constituents in water discharges af ter euch treatment are non-radioactive process chemicals.
l I
In caser where evaporation and/or seepage lesses account for removal i
of eacess tailis.gs water (i.e. overflow of tailings water is not practised).
pH edjustment is generally not practised. To reduce the environmental effects of seepage, the tailings W=t at can be placed on clays or soils with high edsorptive especity to retard the anvenant of radionuclides to groundweter.
h Radium tends to concentrat in the fines er silas (sims 10 wn) portions of t4111 age se11ds. A saad/ slims separatLos would present en opportunity to give special treatamat to the slime fraction which is normally leas then one-thire of the total weight of taL11 ass. Eixing 6.he thickened 1
s11 ass with dry silicate asterials or cement could result in a stabilised f
weste with low tendency for dusting er leaching of radium. This technique, however is largely watested ant. still leaves the bulk of the tr.11Lngs for I
I 0622y
41 -
separate disposal. In addition, the separation of redtonuelides into the slime fraction is not effective enough to eliminate the waste controls needed for the sand fraction.
Many alternatives to the basic acid and alkaline laach process have been proposed to reduce environmental ingsets. These were meant to provide a more benign type of tallings that could be easily disposed of.
These A:ggestions include:
1) radionuclide floatation: About 65% removal for radium and thorium was achieved with tailings from the sulphuric acid process.
3 M
2) thorium recovery: Thorium can be stripped from uranium solvent-extraction raffinate by an additional solvent extraction
(
step. This process was run cosamercially until the market for thorius
[
declined. Recoveries of 90% are possible.
3) pressure leach: This process was examined to improve uranium recovety for high sulphide eres, W sulphur is converted to sulphuric acid,'
thus reducing acid rogstrements, and eliminating a anjor source of 4
envireemental concern. D OSMe" WM#s 'J WI8P M I" '# #Y " *'I y{
ma m e ns ra casaa e+ Cr 3 a) radium entraction: Various entractants, salt and EDTA for example,
\\j%
f have been suggested for the dissolution of radium in acid-loached P,
ta111 ass. Salt leeching is Lahibited by low concentrations of I (e sulphate, and rectifying this causes new problems. EDTA will extract redium and will aloe eartreet other estals, but met therium.
$)
hydrochloeic askd Bedium is readily soluble La the chieride fotia and J
g sulphate interfirence can be overcome by a two stage extraction. The 7
radium is precipitated with bortuna sad then has te ha disposed of.
J Large amounts of chieride mast be etenemically prevented from leaving j
O the plant.
6) sklerination proceses & someehat similar process to the hydrochloric ocid leach.
i 7) mitric acid lasch: Bemoval of radium and therium at the 98% 1evel have been reported. The process is complex and results in high l
attrata discharge.
g) proconcentrations & variety of processas 1the flotation and
)
i l
radiometric sorting have been investigated.
9) ferric chloride, fteric rulphate, Caro's acid and persulphuric acid:
l 06221 l
1
- es -
These are all variants of previously discussed processes. Caro's and persulphuric replace sodium chlorate and thus lower chloride in the effluent.
Without exception, these processes have higher overall costs and i
produce concentrates which present an equivalent or greater hazard.
One ther
- ~ proc s addi on that a 16 be sideret is frot flotation o depyr 1.z e th talling ) hey to conc trate e be use for acid oducti e/
I
/
fi
' tac pisulting/silings gli still(contaigh redl/nuclides). ithou some
( l radionu ides wil float the ite /,
pyrite y nt ces us will proba y be di sed of the t lings e a but it 11 cou less l
env oneental anace.
J t
6.2.2 son-cadioactive pollutants The potential Lapact attributed to non-redioactive constituents is highly dependent on mineralogy, c1Laste and the process chemicals used.
i An important source of environmental concern from tailings in some areas is the ewidation of pyrite which results in the production and release et selpturic acid and leeched metals to the environment. At some mills in Metw South Africa pyrita is recovered first by, flotation and then the concentrate is used for production of sulphuric acid es a product. In view of the unexpectedly severe problems that oceur as a direct result of the ewidation of pyrite, its control requires careful consideration en a site-specific basis.
(
rs ern eal G 8eme eres a mus mer h*ksme11 quantities of manganese, copper.
/t selenium, aslybdesus, venedium, michel, ereenic and sine minerale, iron evides, monasite and rare earth minerals. Only in rare cases de metals such as these exist is uranim eres in economically attrective concentrations, even though their concentratiosa can be significant with regard t4 toxicity. Thus, they are usually met recovered and and up s m in the tailings. However >
JC M some of these elements tend to remain diseelved under conditions which precipitate other undesirable asterials. Some mills in the United States of America recovered venadium from their eres by solvent extraction and j
in canada the recovery of nickel at one mill and ge10 at anotoer is being l
contemplated 1481 j
i i
0422y l
l
)
4*13
- 3. 7. 6 Monradiological (missions Our anius ellling, small quantitles of a numeer of airborne cheetcal (noeradioactive) contaminan also released to the environment, fee products of caecustion free the burn of fuel (e.g., na u ) in the process and heating boilers release C0. nitrogen
- s. and 3
water vapor. Sulfur 4 o nd sulfuric acid fumes are releasert rene tne teach tant vent systee. genera very low concer trations. Vaport gan6c solvents, mostly kerosene, are released in varyi unts free the sol straction ventilation systee.
Data f ree several environeental impact statese niue mills " D 4 11,ing y pical emission ng 1800 MT ( R i of are per say. Esission of dg(L cates have been adjusted fu a etil or, s f ree 0.5 to 16 kg (1 to I er day. fysical *elease organic solvents (925 kercsene lp#
- ates of sulfur dioside
.uding sul' uric acid fumes) from leach temas from 0.2 to 1.5 kg (0.5 to er day. The eurning of fuel oil Instead of natural gaa c wit in the emission and We at rates of around 22 kg (50 lb) per day and 5 kg (10 to) per v:ly. The eetssion of H 0 and CO: as concustion products are of sinor importance.
.y M
- 3. 3 Monconventional teachates k
Landa (23) summarized the results of several studies on the leachability of radium f rom urantus
/
ores and all1 tallings. The follo ing generalizations resultad:
g[k 1.
Radlun in uranium are is only slightly soluole in H:50. but highly soluble in HCL and mo.
s and distilled 2.
Radium in acid processed tallin s is leachable with (OTA HCL. HNO3 water. The solucility in disti led water is highly dependent upon the liquid to solid ratio used in the test, suggesting a limiting soluellity product or suspention pel I
e f f ect. Data presented suggest that uranius e111 tailings in the environment may constitute a long tere source for radtve contastnation of contacting surface and i
groundwaters. Tailings which have been exposed to weathering forces for several years ".
hari apparently shown evidence of radius depletion.
3.
Variable results were obtained when entraction of radlw free tallings was attempted by use of various salt solutions.
4.
Soes of the cheetcal properties of grounduater that appear to f avor the transport of thorium (high concentration of sulf ate, low concentration of calcium. and low lonic strength) may inhibit the grounosater transport of radium.
5.
At lower pH's, thortus becomes more solutle. Acid-leach et tling may dissolve free 30 to 90 percent of the thortue in the ore. The solubited thortue can De subsequently precipitated if the acidic effluent is neutrallied eit h r by contact with natural media, or by process additions of line and/or limestone h ?.he waste solutions.
6.
Sells and related eaterials have been shown to torn significant quantities of urantua.
The sorption of uranyl ions by such natural media appears to be reversible. Uranius oust to reduced to U'* Dy a soit related substrate or a mettle phase, such as H 5.
secent studies perforeed by Oak Itage National Laboretoryse have indicated the following:
1.
It becomes more difficult to leach radium from ore with H90s as the teaching proceeos.
Very little further leeching of radi m occurs after aoout the 10 pCl/g level is reached.
Three eslar HMo was an ef fective leachant for tallings. =hile 0.5 solar HNO:.as not.
2.
s 3.
Ilitric acid *1eeched ore whose radium content has been decreased to less than 10 pct /g will not leech significant amounts of radlue with water.
a.
Progressive teaching of tallings from the sulfuric acid leach process with Hae0s and mater will require further stuey.
5.
It appears that thorius and uranium can to entracted from leach solutions by tri-yeutyt phosphate (TBP), and tPat f actus can to carried on 8a50..
It also appears that chelating agents such as (OTA and QTP4 are effective teachants for reeoving radi w from urani a tailings.
Leaching of cre or tallings to reeove radium and thorius can be beneficial for environmentally sound disposal of tallings. % ever, reeoval of these radioactive eaterials by leaching with nitric acid or other leachants say te only a partial solution to tee procles, availaDie studies do not indicate if leachirw) to remove radioactive esterials (such as radle or thorim) free ore or tallings will also reeove other pollutants norselly associated with tallings. Alle radiue 4 thoria cause such of the environmental hatarti associated with tallinos, further treateent may be necessary to remove other estals, sulfates. ersenic. And other pollutants to ensure er.virormentally safe afsposal of tailings.
a w
--9
.c,
8 4
a
~
', > D*** p se we P &8 k
MW be.e >
Amus se M GEEm**e8 mt+M Am same saa ne W
e' e' esew 4 etyg eq yg
- W sw me a
er -e
. N A:-=x
,.e/
/
e e ee
~~*
c-,
0 0
gg gticp. u feeVAE lC m Trowelexy dr o sneeungment n
e 4
M,M F ese W
, ame
~~
"* h
..v
. ~. ~. -
".* :::::::..n :::::::::::
%*=r
-: ~.-
.=22 1
W WW M Camose *was use s en eerme nef {
^ meep p _ _ _. 64 31
^
W bammeare sese,ame-..g s
_W s
poseese ayure
\\ \\
e e em
,o.
ua, 7
w e,se m pass
-*aspoaeee=ew' m a
see M $sernea throma resatu ebernenre emerno smoomageentsM { 3 0],
n psygg ll
. Embaniments are usually designed using conventional das design principlet involving interTsal soning (e.g. sones with low path '8!'ty, sones with filter materials, sones with permeable rockfill and riprep), and constmetod frwa selected htTow meterials or mine weste rock. The dans may also be constmeted from the coarser tallings fraction, althaugh this fraction usually contains adficient amounts of radioactive elaments to aske their use undesirable.
Confinement is achieved by W selection of appropriate low-permeability earthen materials for the core of the embanianents, eithough when this is not ave 11%1e artiftetal liners any be used during the mill operation phase. Appropriate confinement over the base of the impoundment say be achieved by the selection of a site with a low-permeability bastr.. When the base is relatively permaabte (cogared with the embaniment), or is rendered permeable by the presence of joints, f aults or other weaknesses, a i
blenket of material with low perusability any be used to achieve suitable confinement. Earthen meterials used in various senes la en embaninnent met be selected se that substantial transport of asterial from a low-perweabilty zone to one of higher perweability will not oseur. Appropriate filter mies met be satisfied for the greding of meterials la sentiguous senes, including between the foundation and foundation seatest senes. Grouting any be used to sentrei near-surfees oespage looses.
The following festere and sheracteristias should be suwidered in assessing the leastion and use of a ring dybe layoundments (1) ting dytes are weed la flatter areas then valler dans and thus have a greater flexibility la location.
(2) la flatter armes, ring dybes have a lower maximum embankment height for a given volume of tailings (particularly when compared wth valley damas near the band of catahments), thereby reducing N riska of den instability and the sensequences of larger hydraulic gradients.
(3) yor a gives volume of taillags, the anximum embankment height any be reduced by constneting it in a single stage, using constmetion moterleis borrocod free within the storage area. The use of internal borrow areas minimises environmental impact.
0422y i
r
, (4)
Increased embanionent lengths may tesult in an increased probability of l
embeninnent failure and/or soepage losses through the well and the foundation.
I
~
(5)
The relatively low maximum embsalonent beight reduces problems associated with the re-establishment of drainage when tailings are eevered during rehabl11tation.
(4) ging dytes een ceamonly be located close to N mine, thereby allowt
)
' for easy use of mine weste materials to be prograsened into
[ tonattvetion and used for tai'ipgs rehabl1Ltatio [ The ratio of h
I catchment area to storage area in a ring dyke can be minimised and the b[0 risk of overtopping due to flooding can be reduced.
(7)
When located close to the mine, ingoundments any be used for W disposal of weste rock as well as for evaporation ponds should they be required.
t (8) ging dytes may be located away free streams, thereby reducing the risk of ereston er damage due to flooding.
(9)
Seepage from the M c '..t is soumonly met released directly into active strooms. The pattersys to potential sentamination of surf ace waters is therefore generally longer than for voller dams, and a greate-time period any be avettable to tap 0taant remedial maaeures if needed.
(10) h regular ehepe and mis.Laisattee of the satchment area allswo cone.rel ever water depth, taittags ee11de butte-up, liner insta11stien, and, by malat4Latas a water sever ever the taL1Lage, seats.1 of redse eehalatten.
(11)
DF ensevettas a deep internal borrow area a large propertion' of the j
o111 ass een be held below h level of W matural ground surf ace, t
) Q epas
- i i-th.rek, r.d i ri. of ero.i af t.r. t of N.m.
1 f*u
~-
1 q.i.\\.H l
J (12)
Ring dytes are commonly located near the crests of ridges where the depth to the water table is greatest. This tends to increase the time required for rolessed contaminants to reset the water table.
(13)
The sonstruction af ring dytes requires senalderable quantities of earthf111 and/er rockfill.
(14)
Additional tai 11nss storage cepecity ten be readily achieved by constructing additional stages er additional dytes.
04221
. (15)
In regions of high rainfall, the location of impoundments on the cr'est of ridges exposes tallings to more favourable conditions for evaporation of tailings water. However such locations often increase the probabi1Lty of windblown particles.
7.2.1.3 Itine-plt impoundments berked-out mine pits any be used for tallings impoundment.
The procedure for backfilling the pit with tailings varies considerably depending on the c1Laste, the depth to and variability of groundwater, the proximity to strooms, the susceptibility of W pit to flooding, W permeability of W well rocks in the plt. the mining programme adopted, whether wet or dry tellings senagement is practised, and that characteristics of the tallings.
In arid regions, ruch as in the western parts of the United States of America where evaporation considerably exceed precipitation and the depth to
- gro3 @ tar la " $ 17 great, it is considered feasible to backft11J he varied-out mine plt te at least 7 a shove W water table using general backfil1[ hen to install e uner, ans iinasty so yieza saiti@
p t.
4 g.O The effects on the liner Lategrity of settlemunt of the fill and hence the b
'l liner's offsetiveness in preventins sentaminated seepage from LetL the 5W active groundwater system are met as yet fully detemined, s tierly, the friskoflinerdamageduetoitsplacementegeLastthefracturedrockoft, t.in.,tt
.t b. c.nete.r.d b.f.co.de, stas tsis t.cwn!,u. f a,.estbl. erst.
W of backf111Las in arid areas is shown in Fig. 11(a).
g6 la high relafall areas such as merthern Australia, where the water table reeshes the gremed surface during W aamuel wet season, it La met possible to store tallings steve W water table and it is consleered appropriate for tallings to be stored at a minimum of 7 m below W active groundwater system. Consideration is given to using yneumatically applied concrete, grouting er filter layers to restrict the migration of soepage from the Lapoundment.
The precedures that would be used for backfilling the mine pit with taL11ngs is cieerly site-specific and must be engineered according to the prevelling conditions.
f
{ ghw;.g M ks-MI
., a, w a p +tt" T h a l x
4 A.. & V a
A new technique iJ being applied at one s[ta in Canada [&O y thug e meest l
g pit wells, fracturedsasKebaseofthepitisfittedwith e; A p
underdrains. Current mill westes are put into the pit and the liquid ef fluent h,h '
is pumped from the underdrains to the surf ace for treatment. When the pit is 4
fM full, the surf ace will be capped. contoured and sealed. precipitation will
.g MI flow to the pit edges and percolate thrauch na fracturaA none and into_the l
}
seemdwater system. Thus the tailings mass will be bypassed and the opportunity for groundwater contamination will be limited.
Wh 03 y
lp 2,h f t
* 4 g'I, g ltJf' In some locations, mine-pit ingoundment any be carried out in conjunction with mining although the protection required against radon g
exhalation and the safety of W operation aust be considered in detall, rigure 11(b) illustrates e possible technique for this type of operation. For j
this systes a liner and a compacted dyke are required. The feasibility of such an operation is clearly site-epecific because in many applications it will not be possible *.o place compacted clay liners on pit slopes that are steeper than 3 hertsontal to 1 vertical.
The following f actors and characteristics should be considered in assessing the use and construction of mine-pit iWts:
(1) hiccessful egylication depends to a high degree on the mining programme adopted. It any be used where W orebody can be mined gf h5 M coupletely before milling operations sesumonce er where Wee is en g g.d 6 A2.
existing worked-out mine pit, ehrwise there would be the need to g$f Mg) construct a tamporary boldLas pond for tallings untL1 W mine pit l
l b.c.no.,311.ht..
p (2
m ore a low everburden-to-ere rette exista, tallings in a satursted A
sendittee any be too voluminous to be sentained in m pit.
O 3
Effective rehabilitation of the impoundment any not be practical Lt 4,3l g).etureted t.ilings h.re been do,esited in no,it.
In - a c.ses it
,4e m.y.aly b.,..s tbl. t.,ta.e. s.11 e.,or.n to,.f a. to titas. and l
this any not provide adequate etabllisation.
(4)
Wees tallings are placed deep below W water t4ble and a great depth of water is left in W pit, the mont11ty of this water may represent an environmental hatard depending on the chemical properties of the water. This could be a perticular problen if sulphide minerals are present.
b h.(.
ob y
f et 3
~:1r a re M Ts, Am."
Lou LadUede d %sd ubit. (5g y.yJ a (k. M Ta&.,
pc. Ff. Q,p. r,a
O s
e, M.
td, dN*M %$
l c
h g 1*J g{
M qp hM#
3%&
~
(5)
The greater depth of burial available can provide greater assurance of post-operation confinenent depending on the potential for erosion of the pit cover.
(6)
Visual and air pollution ingaets are low.
(7)
Increasing the pit depth usually minimises the surface area of tallings and hence the enount of covering asterial required.
(8)
Dykes are not usually required and hence risk of f ailure from this source is eliminated. Instability of the pit well sey affect the integrity of any liner.
(9)
Where 1Lning of the pit is necessary, flatter excavation of the pit slopes may be required to facilitate liner installation. The roeuttant requirement for a wider pit could have increas3d economic and envirofreental consequences.
(10)
Where a liner is used or required to retard seepage from the pit it would also retard water drainage into the pit through the walls, leading to possible instability and/or breach of the pit wells. This problem is evercome where tallings een be placed above the weer table, g chv7
- {
(11)
A thorough knowledge of the 41Laste, groundweter fluettstions, rock g
se structure end permeability, and transmissivity is necessary y
before can.id.rtas a in.-,tt ta,eund.ent.
g#
(12)
Liners any be subject to defernetten end tvyture due to dif ferential settling by the backf111.
(13)
Urentum-enriched ere sense mer be unevenly scattered through the generally mineralised redies. State the planned pit boundaries are frequently defined by mining econeelts and not necessarily by ore reserves, the tailings mer cover future ere reserves. Re-mining of the ama any be rendered 4Lf fleult, particularly if saturated er wet disposal techniques are used along with a containment 11.nor.
M r_- ' Eta of dry or semi-dry tallings wouls be more tsadily re-eined.
(14)
The use of mined-out pita may cosy 1Leste future mining operations near the pit, particularly for underground mining methods.
7.2.1.4 specially dug plt impounements gatavation may be undertaken specifically for the purpose of providing in-pit ingoundsent. The tactinique le related to the rLas dyke Layoundment system in that the meteriale excavated to form the plt may be used to form the
e e
D* Ca t it. L **Jhlf eVC'*0*e 3
toe go.6,e t e t g 8 ' l' %4 A'"
gasces gdstext C%i E*0 *3I**Il
/
- x\\
. sggsw
^ t ecavat:14
~
8v'st.4 t< ve.t e N140 90 es'60 Cam acce tio%at I sca va ticas mea tt nia6 toett A acta asacceto446 w=fmtSaCefoku seg, tag e a * ?= ** t l'af.L + ta ' ?%
- 1 a 8 *tt 00a00
{
Ik'*C34'0V'O la stig g,3,,,g,,,,gg
/.
N w,u a. so.n::i,:0.,uy m.y no.i s.=f t.,t.t t<,. f e
,.,,,,,0,,0.
3easo. taw e
t arttaug om tves act was 10 e= 900 los6 C0 vie
/
sgt.*'s't'sC0vta /
m ito.agamovcaw.
m N Svetad tat und e
, * ** TO Caoysicana tt e DWAt EL M $00taedy M N despeant speoen & il ft
- S 060 K JC ab o
e 6
=
. surrounding retaining embankments. The principei difference between the two techniques is that in the specially dug pit impoundment the entire tailings may be placed below the original ground surface. whereas for the ring dyke imgoundment tha only portions of the tailings stored below the surf aco are those contained within the intamal borrow area. With incrSasing excavation depth the procedures becomes essentially sLallar to the mine-pit layoundsent discussed in sub-section 7.2.1.3.
A possible form of specially dug pit impoundsent suitable for use in an arid region is shoen diagransnatically in Fig.12 IM1.
53 The procedure in siting and constructing such an Lapoundment consists essentially of:
l (1) sele ting a geologically stable site with favourable roll and strength conditions to peruLt deep excavation (2)
Maktr.g an excavation of suf fleient volume to cent 4Ln W anticipated quantLties of ta111 ass i
(3)
Forming the surround embankments with meterial excavated from the pit (4)
Depositing W tatlings and free-tallings water sufficiently below the natural grade to ensure that seepage from the pit does not occur through near-surf ace petsnable seit horisons, sad Wt a minlaus depth of about 3 a is available for taL11 ass coverage with a stable inert reterial at the coupletion of operations, g (5)
After completion of W mill plant operation, covering the tailings N
with surrdhkment meterials and contouring W surf ace to A
ensure that reinf all is shed from the wrf ace and does not inflitrate h depeelt. eredients for evn-off and eroeien protection should be la escordance with appropriata engineering practice.
(6)
Constructing other layoundments, which may proceed at the same tima es. er subsequent to, the initial epecial-pit escavation. Progressive rehabilitation will minimise W effects of redon omhalation from J
tailings.
The follt.rlas facters and characteristics should be considered in assessing the location and use of specially dus pit impovedmonts:
(a)
The pit should be in a location that is conven16nt to a mill and has favourable topography, hence favourable drainage. Also W geological l
A499w
)
&[
u,u. A l 4 hey y w N w f/ k~' S ya M 6f,,
Y*
Y eWb -
W I
and hydrogo/ds dIti a sh Id ha the impoundwnt is e
positioned both above the groundwater table and the relatively ingemeable horizons, and at locations free from erosion or flooding (b)
This system is applicable principally to arid areas because of the high erosion potential of the excavated material for the pit while the pit is being iLiled with taillags (c)
Af ter rehabilitation the wind end air pollution Lapacts are minimited (d)
The system involves considerable excavation when saturated or wet tailings management are involved (e) excavation costs in shallow bedrock are uruelly excessive in comparison with costs for other impoundment systems (f)
Providing the site is carefully selected, impoundment is relatLvely free from the risk of erosion, flooding or fatture (g)
If wet management La practised, tallings any be dif ficult to I
rehabilitate because post-operation tailings esttling may lead to a
failure of the cover.
7.2.1.5 Underground mine Lapoundment Underground mine Laroundment neraelly involves separation of the coere= tat 1 Lags fraction free W stimes fraction, and using W fetwee as backetti within W mine. In autrent practice, taL11ngs used as beckf L11 are placed in underground mine areas to serve as floors for mining equipment as part of W miains operation, as well as fee piller support. The proportion of taL11 mas ht may be returned te the mine is a functies of the ere mineralogy end u1111mg operation which detetuine the propertion of coarse tattiass to s1 Laos. It is possible for 50 to 69% of the taLitags to be rotermed to the mine in this momer. The remainder of W tailings suet be diepseed of La ether ways er la ether mines.
The following factors and characteristics should be considered when assessing the use of underground mine Layoundment of tallingt:
l (1) stabL11:stion of W placed tailings requires that they have a high perinnability to permit rapid drainage of excess water during placement.
(3)
Basses water should be confined within the everett mine-mill water managemeent system and not be released to the environment without treatment, to reduce the radioactivity to acceptable levels.
l 1
i
. (3)
A significant fraction of the radium, thorius and other contaminants are contained within the slines fraction (approximately 75% of the redive stays in the s11aes): tN s, underground disposal of the coarse tallings fraction does not appreciably reduce the overall radioactivity which has to be contended with in the remaining weste.
(4) surf ace management of W slines by themselves is made complex because t, hey behave essentially as dense liquids one do not consolidate to any appreciable degree.
(5) suitable techniques for stabi11 stion of W slimes fraction in the surf ace impoundsent have not yet been demonstrated; surface slimes impoundment thus remains vulnerable to erosion and instability.
(6)
Tailings placed votorground will contrNte to endon and redon daugher levels in the mine. For active mines, this may require additional ventilation and special precautions for handling excess water seeping free W tailings. Conversely, backfilled tellings reduce the mine volume, which any have a beneficial influence en ventilation requirements.
(7)
Possible re-mining at a later date, te recover lower grade ore, may be I
complicated by the presence of beckfilled tailings.
(s) the attention must be given to the more fractured condition of the l
rock as a reau t af W mining operat on (for what purposettf).
us4 &
w.u. h M.1 4 v H e.c4u 4 M I A k.t Q&
q.bb. w= rih 7.2.1.6 Deep
- S pte c
c ts 4g l
l v e of d.op 1. bee for the 41.p.eal of urentum satitass has m.en consid. red fee statas e,.retion. La ca ede altmo.gs th.ee are n. ureniu.
1 mining operations eversetty ustaC this method. The sensept is ta place j
tallines suffistently deep La a lake se W t free sayson will not be available I
y i
to reest shemically with W taips and hence keep W radionuclide dissolutism rete lows thus it is intended Wt tat 11.se will remain dorment on the lake betten. With time a natural sediment layer is espected ta form en the tat 11nss surf ace (or altamatively, a sediment layer mer be formed artificially). the providing further protection against dissolution or roeuspension of the teilnse by take water. This method is very site-specific, and considerable investigation must be carried out before the method could be shown to be acceptable. The limited evidence available euggests Wt this methodmightworkforwesteswithlawpyrite,buttherearestrongercon o.or its elve for ugh,yrit. -tariate n,n. h p,hl%yw
.A O.;t)q q.
rl..
l 0422y
l 7.2.2 Types of cover materials A wide variety of cover materials are avallable to use as a cap to tiese out impoundsent f acilities for uranius tailings. The design selected (Section 9.6) wi11 depend on many factors, & esebination of sever asterials usually provides better protection then any one asterial. The advantages and i
et..dv.nte.e. of ee.e.o en cow.c
- t. rials ar. dt.cus..d S. low.
7.2.2.1 Clay covers g g g f-4 d
h6 Clay.
y ifh. has the potential of substantially redu p
r don ar.L:sions(. Generally, clay used as a cover should be protected fros l
1 rect exposure to surf ace drying and erosion processes since it is not euttable for direct espesure to atmospheric influences. Clay is more a
i susceptible to wind and water ereelon then is sell because of its fine and i
re W e uniform particle else and leek of ergenic binders. Clay readily fleeures from water ereeton. Caspection soute tenathan W weeful 1Lfe of the clay sever. Other dealan soebinations een include pit tva rock layers to j
enhance drainase swer from the surf ace of the star sever layer, and this would i
\\
i reduce W infiltretion of water through N slay layer 154). As long as j
there is eene astet clay present, reden emissions would be reduced.
]
7.2.2.2 sative seit sever Use of motive seti to sever W taillage would be desirable fres W viewpelat of establiotment of Indigenous pleet species. Betive soil sever, l
j opplied in several layere ef ter settling has oesurred. has provided adequate l
resistense te wind and water eresten in se.e sesos. Generally vegetative growth will enhance erseten resistence. Even if eseyseted. however. native enti in W western United States of Amerise is usually estatively inetf estive pg in slag rodea emissions emines suffleteet seil depth is weed, yer 1 ' stern considerations, native soil mer not provide the durenlity needed by g
g performance criteria tapeeed by the se p tent authority. la N United States, the regulatory design criterion Li 1000 years, with a mLatam of 200 years depending on the site tirovnetences. If the native soit sever and its I
slopes are protected by a riprey cover (see Section 7.2.2.3), then relatively 0422y i
h N'
8~
toep slopes can provide sufficient control. However, a simple soil covee may require a very gradual slope of 1 - 3% in some instances (Usnc 1980 unpublished technia.a1 posittonj.
Steeper slopes can be used in a "sacrificial manner". This is Jane by extendig the cover outward and in thickness so that when the cover erodes to the slopes prevalent in the area surrounding the dispossi alte, the citver will still be sufficiently thtak to provide adequate controls.
7.2.2.3 titrap covers tipray is rock or stone which can be applied over tie tailings to control soll eresion. It is cosamonly used in applications tuch as highway embankments, duas or (1 cod control thennels, and it could b e employed to stabi11:e tailings onwnkennte and overcarder. cover. While applications of riprep to date have been to handle shorter ters erosion cent erns than these f aced in taillage disposal, the guidance developed for short-tors applications as presented in various engineering and design taats can be of snee use. In the United States, riprep has been used to reclata abandoned uranium mill taillags sites La Pennsylvania and Celerede 1541. The Uspos teek into consideration the peat flooding history, the le<ation of the disposal site, the size. quality. durability and *Lacement of the rock to provide some level of confidence that the stabilised pile could survive la the ions tern.
Reports such as reference 35 have been published relating gootechn! tal desksn principles to the stanttisation of uranium mitt tat 1 Lass. Since reteated i
freese-thew syttes and other weathering phenomena tan reduce peer quality rock to an insuf fittent stse La a mutter of a fear years, not only the eise of estk riprey but the reek gaelity must be acceptable.
Good quality, agil-staed rock which is properly placed (gaps in placement can easily taltiate svily eroeien) can provide reasonable protar clon A
from significant retaf all and floodtag eventa s
.31ae+
sot only will this previde a level of protection against edgeMimaastny Ming and rainf all events, a properly designed riprep cover providse excellent protection aginst the sentinuou. sequence of lesser events, e.g.100 X
years of 2-year floods. Unprotected earthek eed to be designed to deel with the cueulative offset of lesser reinf all and flooding events.
04227
9/., A /,sd, M
M
[ AY
)
M e M m..M_, \\
m' 69
- f i
giprep. in addition to providing an 'arisouring' of tailings cover against erosion, may enhance vegetation growth. It may provide protection for the collection of wind-blown soil particles which will form a favourable habitat for vegetatier. to grow between rocks. It will also tend Lt reduce intneten by burrowing entasis. However, the use of inappropriate sises and senunts of riprep could. in some cases, fester burrowing by entaals.
I
-d A airture of soil and rock (or riprep) is a favourable asterial to use in sever construction. The combination is ref:istant to air and water erosion
(
and it will retain moisture better than riprep alone end hence will better 1
i l
sustain vegetacion.
I 7.2.2.4 Other covers other types of meterials have been considered as condidates for tallings covers. These include synthetic asterials such as pleottes (hypelon)
)
and asphalt, as well ne sixtures of soil and senereto (rollerete). The United
)
j Statsa of America and Canada have inveettgated these materials for use in
)
i toden attentuetten and in oreeton sentret. In meet *.aees they have been rejected for t.se se a sendidate sever meterial because cf seet, peer durability and longevity or defittency la some other meesesary performance
(
)
facter 156) a*I
\\
V Y@w' " M 8 i
)
~
7.2.2.5 surf ace vegetation CY ',.' wl !
+
AY g
rf - e eget.ti.
he - off tive - t., rete. ting te m.,
'l e t. m e e r fre. est.c.r
. er.eien. re.mre ohl.,in,1u..e t,.
ef fectiveness of surface revegetation en ign'.ts een be breedly classed i
into sliastelegical and agrebiologisas fastere.
i the theise of plant species suet take assount their adaptability to the porticular stiastic esaditisme in the tailtage ersa. We fatters include the amount and Matribution of reinfall, entresse of temperature. Length of daylight and growing,tvs :. wind tenditisme and the presence of perinef rost.
i i
I Concerning agrobiological f actors, the nature of the ore used and the mill processes will largely determine the characteristics of the tailings fres 4
l j
~..
1 !
l l
the point of view of their potentist for sustatning srowth. With.
- <c teent or covering, tallings are usually not able to sustain significant 3:
Considerable efforts to correct adterse characteristics, such att
.c high pH and lack of plant nutrient content, will usually be required betore tallings can s'Jetain growth.
r taportant considerar, ions in this respect concern particle sise. aggregation of = articles, and water retention capacity.
i Because of the wide variation in conditions from site to site it is j
not meaningful to describe here the vegetation practices followed in various I
countries. Significant success has been achieved in revegetsting tailings covers at various sites. The detailed procedures and types W vegetations currently being used at urenLum and other mineral mill tailings ettes are described in Ref 157-61).
Recent tests in Canada (621 show no discernable diffence in redon release between a vegetated site and an unvogetated site, and moleture is l
thought to be h controlling factor. It would also explain why previous work l
l in more acid cliastca has reached a 41tforent conclusion. presumabl W
vegetation driM up the shrunken root systems provide prefseential tN(ff) [I 9) and thus increase the e.a lation rate. This does no', happen in Canada as 1
there is always an aburdant supply of water. Thus direct revegetation of tallings surfaces can be considersJ provided that annual precipitation will maintain the water table ano that other agrobiological factors ace resolved.
Vegetated settfaces raise other concerns. Directly sed ing onto the tailings sustate will only produce viable, eustaining vegetation if the surface has been mcdified adequately. Geneve11y tailings are sterile and needed added fertiliser. The seed six shooen must include nitrogen-fixing as well as nitrogen-using p1sats if a natural cycle is empocted. Additionally.
If b pile is in a forested region, tree growth will be inevitable. If a soil sever has been use1 to promote ths growth of grass and e h r eas11 plants, t)pa tree seedlings aan be emptocted in wetter 411 metes within the first few years. The reets of these trees will penetrate until b y reach the ehemically stressed region below the aeL1 sever. More the roots will travel horisontally. sve.atselly W tree will become lege enough to toeple, pullin'.
vp W shallow reet systen and breaching the sever integrity.
0622y
l '
In dry Periods, grass (or forest) fires destroy the veget.ation cover and result in a loss of protective value for wind and dust erosion. If natural or artificial re-seeding is not done rapidly, serious damage can occur.
Good quality vegetation will also attract animals, especially in areas which era otherwise barren.
. Trees have rooted in weste rock ) des and riprap areas also. some persistant species (like bireh) can be expected than to 1stablish themselves in rock cover. There is less likelihood of eventual damage to the cover integrity in this case.
7.2.3 seepage control
, Seepage control is effected primarily by ensuring that water barriers,
of low transmissivity are la place in all directions over which a hydraulie, head or drivinC force is in action, to drive water away from the impoundent.
The term ' liner' is used in this report in the generic sense and refers to a wide range of natural or synthetic barriers with relativel
'""-';"'7;wa+ ** g&"" #
}s Other forms of seepage control include using a erials, largely
)
61M e e M CAess{ "tufrai/W f 577w JM'7 IW 0 F natural, having geochemical properties -
p radionuclide or chemical contaminants in N seepage r
'['_
4
^ "
to ensure that natural dilution and dispersion have sufficient time to reduce the concentrations to ecceptable levels before they are dissharged to the environment. A similar effect is accomplished where the eeth lengths to the point of discharge into tAa environment are long.
Nydraulic barriers are somstumes used to sootrol seepage. W concept being to reverse the hydrau11e 4 riving force such that seepage flow is into the b _ ^- t area. Seepose any aise be kept low by reducing h hydraulic heed set-up in the tailings impoundownt. In other cases, a significant aarmt of soepage acy be permitted to occur over short distances, provided the seepage is ecliected before it impacts mac's environ.nont, and subsequently mana6c4 as required.
04227
f
. l
(
Site selectiot. is fundamental to control and management of seepage, and it can have a very significant influence on the reliabilities and costs of seepage control methods.
Specific attention is given in the folloering sections to the use of liners for seepage controls other methods are briefly discussed in genere1 terms.
7.2.3.1 Liners
/ # A TJr'a% A f 4 7 4 I TD d' f j i
Liners are discussed in some detalig ny:, ".;;. _.; ;Wiim...
1.
I fEr53 ; only a few design coucopts and alternative solutions are considered i
here. Whils a nualer of different liner options exist there are generally three basic groups that are applics' ole f
(a)
Geological liners (formed by natural geo'ogical formations) i (b)
Clay and other compacted soil liners l
(c)
Synthetic liners, t.hich include 1
(i) synthetic membranes (ii) pneumatically applied sectar and concrete l
(iii) asphaltic concrete, or spreys.
7 Amoredetailedlistingoflining'typesisgiveninTable/
M Where natu?al soological formations, such as luw-permeability soil or rock strata, exist en the tailings C& nt site by are usually effective in providing en eseremical seat. Becadre of h ir usual thickness and stability under site asaditions. W ir long-term effectiveness is generally reliable. la their natural fers they usually seal only part of the i====
- it basin, such as the floor and portions of W contairument ws11s.
and are therefore frs1uently need la seemination with ehr liner types.
Their integrity, i.e. the absence of substantial possemble faults er. joint systems, mast be established or such features accounted for in W engineering design.
1 As a result of the cation exchanga and neutre1Leing capacities of 4 geological liners. Wy also serve to ret.ard migration.50"~r & e
.s or the endionuclides, i
g es well as some of the non-radioactive contaminants. Clay, either fros 04227
"7 63 TABLE 9 UNING CLA551FICATIONS ( After Kays [M p.3)
FletiWe Land Escellutous Mastics Guaate Bestenate clays Eastemers Comrete Gem 4 cal trearmen u Aspast pancts Steel taiwtorse usetments Compacted sods Aaptait seemte Combaastiou
&sdesseet
. Law permeatdary Nash permeab4; Masswas Compected wds Bastomers Cusate Asphalt Steele Ceedrete Steel Aaphalt esecrete Sod concet katoeste clays Oesucal treatments w tert >eret tJeesments a
Caetaneous Noe4eetasucus Mestace Coi.apected nous Eastoswrs Guane Aspaalt paseos Coeerste Steel Aarheit osecrgte Seal seneet Bestende days Cheesesi treatmosa Wetortcree Wesets
(
d site-derived c. / or Laported materia 1x (e.g. bentonite), have long been used to construct effective liners. perasabilities of 10 ' to 10 m/.2 138) and
~
lower are attainable with these materials. However, W se low permeabilites may be adversely affected by seepages of high cation content or low pH, which can induce an exchange of the sodium ions on the c16y mineral. Sodium tripolyphosphate dispersing agent is helpful in esterding this change 163). A number of proprietar*..mntonite-based products have been developed which are described es contastaation resistant hentonites (34). The cation exchange capacity of clay 1 bars enables them to retain er at least retard some contaminants. Long-tors retention of the very low permeability characteristics of these natural materials is a point of concern, but it is still rueosnLaed that W natural liner will not disin egrate entirely.
{
N Y EbH E*
O it.2,
The conuson effective thickness.of a clay liner is 0.5 to 1.0 m.
For clayey soil liners, the typical equivalent effective thickness is in the order f 5 to 10 m.
Material with a high proportion of clay is difficult to place and compact for liner use, particularly on slopes. soils with a substanital,
fraction of antsetal coarser than & clay fraction can, when adequately cosyaeted at the optisans moisture content, form eI*fective linera with perimabilities between 10' and 10 m/s. Lower permeabilities may still
~
be obtained by addir.s bentonite.
Synthetic membranes and their application to uranium tallings is cresidered in eene detail by Williams 134) and Molder (?). perusabilities of
~
10 to 10 m/s are reported as applicable to the various types. It has long been recognised that imperfections of assuafecture, site seasning and installation result la at least some leakage [64). It is believed W t offective overall permeability values of 10"I' to 10"iia /s could be assumed fet' eesperisen purposes. At these permeabilities. W seepage rate would be about 10E of W t of en unlined basin, such liners any be subject to chemisal atteek and other Jegredation from verteus seuroes. Mypalon is considered one of the aset desirable synhtic membrane linet-s for urenium tailings [34, 66). synthetis membrar.es are necuelly eartremely thin (0.5 to 1.5 sus) and con provide good eenfimt for the sher't ters. Newever, their long-tors stability, over thousands of years, whec subject to the chemical and physical envirorument of the tailings 8 r'
.t sust be questioned seriously. Since & membrane is of uniform thickness it may be anticipated that when f ailure occurs it could effectively fail in a 1erse area of the 0622y
\\
b
1 membrane within a short time sp. n.
The pollution consequences may be significent if up to that stage the membrane had effectively contained a wet deposit.
Although synthetic liners have the limitations noted above, thele efftectiveness in seepage control has be n demonstrated in numerous instances, yor seepage control during the operating life of a uranium alli plant, appropriate synthetic liners are en acceptable option. They can also be used, in addition te liners of natural notet Lals, to further enhance control of seepage and to further improve longer term reliability.
Additional factors and characteristics that should be considered in assessing the use of the various liners includes (1) synthetic membranes are subject to leakage through Laperfect seams and small undetected tvytures made during installation. Leak detection before fLiling is difficult unisse the defect is quite large or obvious. However, a leak which will pass as such as 4 litres per minute at 6 m head any be easily missed, even when the location of the leak is known in advance.
(2)
Where synthetic membranes have holes and are laid on loose soll or sand beds, the seepage pathway can follow the relsuvely pervious bedding in all directions.
(3)
Clay linets any be subject te shrinkage and cracking if allowed to dry out before filling. Such creeks should close on wetting, but any filling by more permeable seit while the liner is erecked any result la narroer but relatively high-permeability eeepage paths.
(4)
Compacted seil liners having a lost stay content exhibit less shrinkage and erecking then shoem for liners with a high clay sentent but will slee have a higher permeabL11ty.
(5)
Jointing e? faulting any increase the permeabL11ty of natural geological liners. Det all joints and faults are more permeable than the rock to which they escur. Often highly jeLated ro k strata still dLeylay earttemely leer permeability (lees then 10 m/s). Joints and faults with a high poramability any be Lajected with suitable asterials, such as censat, bentonite, special chemicals, to reduce the permeabL11 ties near the igeundment. Cyt-off grouting ur. der the core of a den is often used. permeabilities of 10' a/s can sencr611y be obtained in grouted areas.
1 natty
4 p. 6A<-. Au u 9 4
%~s%. g.s2 7
ES b-L.
V Qw.wa %L M a kJ y
P w
u, w % 9 a." u s a d s.- bm (6)
Large total stresses can exist under substantial depths of saturated tailings. Defomations of the strata supporting the liners will occur kn proportion to their compressibility, and variations in meterial and p
loading any result in differential settlement. Liners sust have h
g sufficient floritMity to acconmodate these deformations without h W L
rupturing. Rigid liners, such as pneumatically applied nortar, are h #
1 only compatible with rigid support, such as sound open-pit walls in
~
hard rock. Grout any be subject to shrinkage and cracking, and to attack by chemical agents.
(7) tihere step-type (or shear displacement) differentisi settling occurs.
Liners have a high tendency to shear. any remaining effectiveness of h*
the liner at W shear leestion will depend on the thickness and h
strength of W 18ner reistive to W shear displacement. Liners p
gem he 4nnot tolerate appreciable horizontal extension for cracking) in the c
b Y f,our.44 tion strata, as mer be associated with considerable differencial,
^! das/#
settling. Large differential settling can be sapected to be b
associated with substantial depths of unconsolidated backfill or impoundments constructed over subetantial c11uvip1 depths in valleys.
Where steep-welled open pits are used, this differential settling is accentuated by pit geometry, and extremely large shear displacements can be anticipated in W vislaity of the wells.
(s)
Liners applied to steep-sided Lepoundment wells (>30' from horizontal) any be subject to large shear stresses as the tailings settle within the impoundment. The danger exists that variations of the shaar strength on the tailings-liner intafface er liner-wall interface sould induce shear slippage in some areas on the one l
interface and other areas en the other laterface. Between W se areas ruptweee aan eseur la the liner. These concerns any limit lined slopes ta about 3 horisontal to 1 vertical.
l 7.2.3.2 Centest of the hydraulic gradient Limiting the water content in the tallings impoundment will reduce the word hydesulic drivias forces emos tende to push water out free N tailings.
X selection of a site or !
ci
--t sharacterised by a minimum fresh water inflow and the discharge of tailings La a dry or semi-dry condition can limit water influu to the twoundment, resulting in lower hydraulic gradients.
Os227
(
k Underdrainage (i.e. drainage of water from near the bottom of the pile) above the liner is sometimes used to reduce hydraulic pressure heads and, thus, the seepage rate shrough the liner. Underdrainage placed be'ow the liner may be used during the mill operating phase to collect seepage from the liner for further annagement as required. Drainage may also be employed to de-water the tailings end Mee, to reduce W total volume of water available for seepage I
and thereby the driving pressure. This will norina11y lead to further consolidation, and thus stabilisation of W tailings.
l l
Underdrainage systems may be either continuous blankets or interconnected drainpipes. auch drains would normally comprise a slotted or open-jointed pipe laid in gravel and protected by a filter to prevent clogging by the inflow of fines. yllter fabrics any be used to advantage, but suitable design precautions are required to ensure that clogging does not occur and that the fabric is not chealcally affected by seepage liquid. The long-life integrity of filter fabrics is of concern. Chemical precipitation and l
bacterist activity may also affect the efficiency of underdrainage. Special
- precautions ar, be required when the tailings contain pyrites.
Artifical hydraulic barriers may be used to reverse W direction of the pressure head and, hence, to force the seepage back into the impoundment.
This is normally done using pumping and/or natural water flow control, and therefore is considered a suitable approach only during oper'stion of the mill.
7.2.4 Tailings annagement systems Tailings annagement systems are concerned with the treatment and handlias of tailings. They include the systems for tailings management during mill operetion, includies W various pumps, pipelines and tailings distribution systems, and the menner in which these are operated to handle taL11nss from the time they leave the mill until the time they are emplaced in the impoundment arms.
Tallings management is closely related to water senagement for the mine and mill since the volume el emesse water stored in the tailings impoundment influences decisions regarding the recirculation of tailings water back to W mill, which in turn influences other aspects of W water management system.
6422v
-n-Selection of a tailings management system is very site-specific-In arid regions, for example, the tailings surf ace may t,a allowed to dry out,'
resulting in redon exhalation and wind erosion that can distribute contaminantsbeyondthelimitsoftheimpoundmentandintotheenvironmente.[n regions with' wetter climates suffici6nt excess water will usually be ava8.lable to ensure that a water cover of required miniara depth is maintained at all times. Where h water management system permits, a water cover over the tailingsfis desirtble to limit W redon and particulate emissions.
DwtW 4 C /dK hT106/ &
N*\\
Tailings management system Jesign will normally include the following j
asisture control features:
(1)
'Jistribution of tallings within the impoundsent system. Where distribution is effected beneath h water surface, by discharge from a i G" barge, considerable control een be exercio;4 on h location and k h4U t.h distributtaa at hath the coarse and fina Penettaaa.
Whera M*) Op distribution is from the sides of the embankment, control can be M., M 4 '**W exercised over the location and geometry of W beach and the location
/
and size of the pond. For the layered tailings concept, a spigot p g fd l W
M g system is used [67).
44 h s (2)
/d l 6A4 Control of excess waterm in W impoundmew during N mill
/C operating period. W ere a water cover is maintained over the tailings this water adds star.ificantly to the hydraulic sesdient and hence to the potential seepage lesses. After rehabilitation of tailings impoundments, pended water will ne longer exist er.d W hydraulic gradient will be reduced considerably, as will N rete of seepage less from the system.
(3)
Centrol of freeboard in the i-- r' it.
Provisions should be anda in CP( h the water management gystem to provide for recirculation of water to the mill should W freeboard en the embankment be reduced to a level b3 seitical to the cafety of the embankment.
(4) centrol of density, sempreesibility, permeability and shear strength of N tailings withia w rious impoundment sones. Where tailings are discharged belcw h water --faae from a pipeline attached to a g
berse, it is possible to enapea * * * ' '*- re r m : 'r--M on are m-intetsized, thereby developing a deposit which is of uniform perasability and compressibility. Where tailinen are discharged from j
e point source on W beach, segregation of W fine and coarse I
0622y l
l
. fractiong occurs and results in relatively dense perineable deposits in the discharge area and a highly compressible deposit of low pomeability in the s11aec ares. This segregation any render the task of stabilisation and tailings rehabilitation more difficult.
Tailings asnagement systems mast be designed for a specific site and mill system and give consideration to important parameters such as W -
phy11 cal state and noisture content of W ' tailings and h method of tallings implacement. Of particular importance is the moisture content which can lead M
to saturated wet, semi dry and dry management variati s.
$ p' fdAH%Wk')gelUA d N ': ^ ' M}*
7.2.4.1 Satursted management i V
(
' Saturated management' to used here to indicate placement methods by which the tailings are transported, distributed and maintained at all times in,
a saturated state.
In this system the tailings are tesworted to W impoundment area in a slurry commonly thickened to about 40t solids by eight, and are discharged l
beneath the water surface from a pipeline attached to a berge. The barge is manoeuvred by winch ropes attached to W shore: this allows for build-up of h tailings deposit in essentially horisontal layers. The discharge is not permitted to build up a beach and W roby expose the tailings to wind erosion. This method of management results in an even distribution of the coarse and fine frec une throughout the deposit sad develops a low-permeability deposit that can significantly restrict seepage. Hence the potential remobilisation of contaminants is reduced. as will be W possible release of sentaminents to the environment.
I h
y
.or 7,
yf ter st.but.auen the taui.e de,esit sh.uld be s@
soupletely on...m.
Initial diffleutties is, moving equipment ever the pile can be oversome by larias a send cover ever the deposit. The uniform distribution of the variog jrections of the deposit will allow unifem settling to oceur, thus SiiiiffBeffn} h risk of differential settling and A
subsequent tvyture of W covers.
niith this system and its inherent water pressure heed, seepage losses of tailings liquid can be espected to occur through the embannment and the 03227
. floor of the impoundment. Seepage through the embanlosent and the near-surf ace sones of the floor region any be collected by seepage collector systems et the dotmstream and of the embanlonent.
7.2.4.2 Wet management
' Wet managessent', as used here, indicates placement methods resulting in tailings that have substantial saturated zones, but a total tailings impoundment that is not saturated.
In this system, the tailings slurry is pumped to tt,e impoundment, af ter being thickened to abwt 405 solids by we.ight, and discharged by either a point or line discharge (see yig. 9) at the periphery of the impoundment area, frequently along the upstream fare of the dam. A beach is fotined which slopes etiny from the discharge point. Coarse solids settle close to the discharge point, and fines (slines) and water run into a pond fot9aed by the beach. h slines then settle, and the partially clarified surf ace water evaporates or is available for decanting either by gravity flow or by a barge and pump system. Using a single point or line discharge allows some control over the size and location of the pond and beach. If the tailings das is designed for water storage, the pond any be large.
The eres below the pond surface elevation remains wet and saturated and has low sheer strength. For this reason, discharge in the vicinity of the den well is ceumon. If large beach areas are maintained to limit the pond size and water pressure head to reduce seepage, the beaches may dry out and be subject to wind ereelon. Only a limited cepability exists for keeping beaches wet by additional discharges free the discherse points. N pond and soft slurries are loested adjacent to the natural besin sides remote from the dam.
1.arge armes of sof t, p.rarnently wet slines result, and these may be difficult to effectively cover using nochenical equipment at the time of tailings pile stabilisation. As the pond elevaties rises and water is distributed over new areas, seepage losses can increase. Little underdrainage__is used with this h9*
system. With the comparatively larte size of the pond, which is typical for this systaa, large eroes of the das are wet and saturated for long time periods.
When it is ready for stabilisation, the tailings deposit is 04227
1 1
- pplLLN glO substantially saturated. The deposit consists of ine lately consolidated tallings with a high water table, and it has large : es of under-consolidated slines with very low strength. The slimes may have old volumes greater than the volumes of the solids. The high water table provides high pressure heads that tend to induce seepage through the liner. Significant excess pore pressures exist in the slimes, which furWe inersase pressure heads and l
seepage (44).
The excess pressure head is dependent on W permeability (which is dependent on W grain size of the material in which it occurs). The time required for dissipation of this excess pressure head af ter shut-down of the mill will range from nearly seco for W very coarse beach area to hundreds of years for the very fine sihe sones. In time, the excess pore pressure will dissipate, resulting in en effective stress increase within the tallings l
depcelt. Similarly, as the water table in the impoundment slowly reduces to the base of the dam, the effective stress in all the tailings incrassf Theap X
stress increases can be large (34) and result in large subsidence. Lesser i
subsidence occurs in the sone with coarser-grained tailings. The larga differences in sot ling rates and enounts in the various areas tJ1ngs pilesn
^
j---
f integrity of the cap and the con oured drainagepatternhtheimpoundment. f I
e l
The sein characteristics of a wet disposal system that distinguish it from other systems are Pvtosv 44~~
(1) high water table and pressure heads, hence high soepage /s at the liners l
(2) 1arge ve hmee of contaminated water available for seepage loss (3) apprestable areas of eeft stimas with low sheer strength N t are diffleult to sever (4) high tailings sempressibility senhimed with high effective stress changes due to drainage, reeuttias in high surface settling and differential settlinal ggy(
(S) high water content and low ta111 ass density essansegliquef action potential for the tailings N, increasing risk est consequent dam
[X failures AL (6)
Irwer operatlog costs and the need for only a single tailings and X
solution impoundment.
L 7.2.4.3.
Seel-dry management l
l Use of controlled ring discharge allows beaches to b* developed from the periphery inwards. h pond is then located at the centre of ti.e iagoundment as illustrated in Fig. 10.
The size of this pond can be minimized by decanting the maximum amount of water possible and by preventing the I
co11ection of precipitation water from a catchment area in the impoundment.
Decented,weter can be stored in a separate impoundment of conventional design. In this case, as the tailings pond size is snall, the beach ara is marimized. The beach segmnts can dry out between successive deposits and wind erosion and redon exhalation could become a concern.
By using a thickened di; charge and short tallings deposition periods for each beech segment, individual layers of tailings, in the order of 75 sus thick, can be built over the beach area. When cascharge into a particular segment is stopped, solids settle and some of the water ' bleeds' to the top arid runs down the beacn slope to the pond. In this settling process some classification of the tallings solids occurs in the vertical direction, with the coarse fraction concentrating at the bottom and the fines near the top of the layer. Some classification also occurs down the length of the beach, with a concentration of coarse material near the diccharge point and the finer material near the peal. Studies of the classification in South African gold / uranium tailings show that a greater degree rf classification occurs due to gravitational sorting as a 100 mm layer of tailings slurry settles vertically, compared with that elseg a 600 m wide beach or that achieved by conventional sogaration of particulates in cyclones feel. Figure 13 illustrates eces of the grading variations measured in the investigations.
Seesessive layers of tallings deposited on the beaches form a highly stratified depeelt with moderate permeability perallel to the layer interfaces and the low pstnesbility normal to them. This low vertical pommability prevents the re-entry of subetontial quantities of water during subsequent deposits and sensequently a partially satursted (' semi-dry') deposit results.
The deposittenal sequence and the permissible rate of depeelt rise depend on j
local climatic conditions.
l l
In this system underdrainate is required to ensure that there is not a build-up of a surf ace that contains free water in the das from the 04227
4.s see l
sr y a 1
,o se
- d AU l
l
,e l
rr e
p R
ff 1
y l.1 1"
- I N
/c 3,,
y
\\
5
,d sossetscasion cessseen semes.as as ie se se se i.e in.
,<r 2
f at Jar 3ssl 60m e
y A.l te
--P
- 1 Kfonseriosnoe, e e<aa o
e saw>es ese see o n.
reaseCse $ste aw o 33
'4
\\
es e f Cassa.f eCateous twa rim 4 F=, I * ~ I w I ca I
e-le 1(= I gl
>=H sa==.--
i ne I
ee
- 'a o,
,\\
f<>e en oa j
nan a s. r ne nsig w
es j
EGG A
"y p
se e.
1
-4 k.
l s~
.F
,I as se see see see ese sese pas h.
/
l r
a =we sore e e=wetore e s
co saven tosses soa ss es. petssuee sneen e a a en 4
I 1 et l
5
.,I li,tti uf. z
_-3
' on w, esc,re a _
~ lQ
./ roa seena*some
,,n u,....,..,
e r..e
.e e.
I" a
/
_ anwesore a n
owenosom fgg g jy s
- l, m
,a M
f
,d y as 7
,/--.. ) pl-se g! l::-
se.am p
.< r
.ene e se e se se
.eo e <anumerasso canistas sein e -e r., Ie. 1 I c-s.
I sa.--
I <-=
s-sw need Comepareeen elpersute senes an sen8 saga depensa lej tercule asse eersteresfee f f) eeressnee r4 see 4egges of a s.agle shmees Jan med /JJ.erassaee frene sease se meant fee i2 d*fferent danos. ibi separeemre hp g*e**seteenalsere.ng and ees 8 nee separatene.
. continuously wet pond area. The design and construction of ef fective underdrainage systems are cruttal to the effectiveness of the semi-dry management system. Idell-controlled and specific tallings depositionil techniques are required initially to prevent underdrainage clogging er X
c w M 4 ~ 4 n % la al' m.
a-(j bk. $'ve C h h G &
During sun-drying of beach areas, negative pore water pressures are induced that produce high effective stresses in the beach areas. These stresses tend to consolidate and reduce the void ratio of tailings. In this manner changes in void ratio, i.e. from greater than 1.5 to less than 0.0, are effected. Figure 14 illustrates some typical void ratiou versus consolidation pressure curves applicable to normally consolidated slurry and sun-dried samples. The lower void ratio implies:
(1)
Lower moisture content at saturation (2)
Considerably lower settl eent potential under increases in effective. stress (3)
Lower total tailings volume (4)
Considerably increased sheer strength (5)
Low liquefection potential Fines tend to be concentrated near the upper surf ace of each successive icyer and on drying tend to form a slightly cohesive delicate crust. This crust resists wind erosion until damaged and is renewed with each successive layer placed. If dusting does occur, beaches can be dampened by discharging water or more tailings free the ring discharge or by applying a chemical crvstias agent.
Results from the South African gold / uranium tailings impoundments using.remi-dry management techniques indicate that free anisture contents of about 10E can be achieved in the bet *h areas above W water tables [34). The minimum voisture contacts that are achievabia are dependent on W grain-size distribution of the taillags.
)dith a suitably managed tailings diaCarte proterine the locat.Lon of the decent pool can be slowly shifted about the layoundment interior as the layoundment rises. This
- halts the depth of s1 Lees and yet,alts top and bottom water drainage from the older s1 Lass deposita. The extent and depth of any 04321
l 1
[
4II f
i se a s,,aecue a m,-.s or o, i.e. acts r.,itive, u tent unae.ns \\
- . Optta, s. E.. m. E. Dodson and R. J. 5 erne. 1984. Laborato asamLme unce, retmary s.a. Im.
Cessa%iaal ngiwari, no rwe a i= set, a.Paro
- e. c, tomb seat, f=ie.,etty, re. corrio, cdom,s,ciet!
i reseae:aa of a toestone and time heatraltrattaa of AH3 c Uraatus v
Mirmisses sif.eles: Pr.gress me rt.
saiK U C D 449 pa[ uH). u.5. m.ci.ar aes latory camtistoa. Washlagton. 0.C.
~
SEEPAGE AND SEe6 PACE CoffrAdmamt g? TNg g.1 TAIL 1stGS AleEA 5 erne, s. J., 5. 2. Petersona and G. W. Gee. 1983. L aborat ory measurements of Costastaaet Attenesattaa of Uraalass MIII~Taillegs t enhet es Sediments and Clay liners. lasRUff-Hf4 George L. stof fman and Thumas Wung TV -
. tTh=ciear nes.iatory Commisst a. washlagtoa. 0.C.
anSTea rn nemonocrion Jena.2. E. A.
1968.
- Controls on see Fe Ce St. Cu aaJ la Concentraticas la Solli and Water: Ihe$tgmIficantdeleof unetconineralsCorgeration(unetcad1,ganornatingtheA-9 pit Nydrous Sta sad Fe Caldes." Adv. Chem. Se[g, 73:337-387.
railings osapuman Area at can mitts, wyoming in s=cesher 1979. This was the first app 1& cation of 1,elow grade disposal for uranius mill Fey, n. V.
and J. 3. Olsen. 1981. "Synthesis and Properties of tailings la a sais.ed out pit (Paris, hekomeryer, and Wung, 1980).
Poorly Crystalline Mydrated Alustaus Geothites.* Clays and Clay palacrals_29(2):91 100.
Through its groundwater munitoring program, umeteo detected the paesence of meepage at nume close prosimity cuerd teolls. mydro-Engan-
- . Environmental Protectica Agency. 40 Cft 143. *mattomal Seccadarf serig was entiated to detesmine the sete and entent of groundwater Ortaklag Water segulation." 44:42-98 Fed. Ret. (July 19. 19I9)*
contamination ena to rec==.no mitigative action. mir asuvember 19e2 stiaay indicated that to A-9 Tallings area was seeping at a rate of 20 gge, that seepage had unty asigrated almut 154 feel from thm pit, and e.
Scha, M. L., J. ncaseal and G. A. O'Connor. 1979. 5081 Cheetstry, p. 154. John Wiley and Sons, lac.. New York.
that a seepage collection sfaten would tm effective in containing con-g i t,,ig,.gio.,
- s. ttssay. W. L.
1919. Chemical Equilibria in 5o815. John Wiley w cong.cgion,ysee. Initi.ggy ca asind og o,,* ta= ping wull.
An and $ang, stew York, hans Verk.
acid drip system as.4 a secusas collection well weese-installe.d an april 1984.
t.
Lestas. D. M.
- t. K. Ayas and K. P. Strong. 1978.
- Leaching of h acid drip system eletrinated the form.at son of precapatate, sadless fram Ursat sm Ialltags." Paper presented at Seetcar on and time secord collec tion well 411uwod t!.e pumpang of s.ut t iciernt volume
{
staa4gement. Staktlif atloa and Environmental Impact of Urantum to reverse the hydraulic gradsent du.sgaaJaent of ttw collection system.
selli Talltags. July 1918. Albergeserque. lesw Isaalco.
la reversing the hydraulic gradaent, the collection system is eftec-l eg,,gy ca.g.g.i,ca g i
.g.39,,gio,
Shearer. 5. O., and G. F. Lee. 1964. *teachab esty of gadium 226 from drastum Mill Solids and Alwer 5ediments., Health caouesD-esAfta terLiectacT PhyitC*, Vol. 10. p. 217-221.
h A-9 pit esists in tt.e outcrop of the win 1 aiver formation. m 8.
Landa. E. R.
1960. "Isolatfoe of Urantum Mtil fallia9s and Wand sitwer has !=en dif ferentiated into sapper and lower sani's.
W tv-Their Compnent RaJteauClldes from the Stosphere - Some [arth per vand niver manit consista or sandstone, conglomerate, and m61 stone.
Science Perspecttwes.* U.S. G8 W 804I 18F "1 CII'"I*I III*
lower Wlad niver 1. moraally faaer rained maternal that consasts of siltatosie, mudstoce, and fine-graimd sands % on massive conglom-Arlington, Virgiata.
erste.
In the area of the A-9 pet the uger and Iameur Wasm3 kiver sanits are separated 1,y about 40 feet of m 61stons. The flon>r of the A-9 pat prior to tallings disicaal was in the uggen wind haver format aan, whitta estende another 30 to 50 feet 1,elow the pit floor..
1 7;eorge L. asof fman, Nydsologist Mydro-Engineering, tamper, wymang.
Thomas Wong, Environmental coordinator, unu*cc Mis ea41s Curgerat aan.
t;as Mills, Wyoming.
umetco is a wholl; omseed subsidaary of thison Carbade Coricsatiun.
that11 april 19U4, it was the Metals Divismun of unaven CasLade Corgo-Er '.lon.
m
41)
D e etterfere=== the t*** of the tycer ulnd miver arget for wee dew-eleged to d=fis* are*e of g referentist f low in the ther wind River h8tLt - 1300
- h OLD T4?t.tNGh - 1200' [
4:e e,.,er
- i,.n.
T,e e,r.t.r.
e owe t t a *e,.r c a ei i.
t e mer.. it eiwer f r. mat.c. e.iot o.ic
~ wester. and ther.
g g
ed,,e of the a-* pie.
The w ewe channen. pt..e.
ca.e s.se t m e,ti we,t of t re..
cit.
Th..e i m ser et.or.1 are...r. is, rta.t i.e-g
,:7.,;f.;.
couee neee are are.e of 5.eatest.e.r.t ed nach e.e and ther.for. have g
g:pm,.
t e,re.te t p eentsai to -oever w.eer.
E
.tra cx-e ere.
r,e e te t-h.-e heea m dweted the twer =i e niver
"="
JJ4#49npg, t
.,eifer.. the a-,.re. inn
- i,en. a t.e.seie switr of ioo,al/d.y/rt I
e
. g :.
- ,Y,7,,.:
so tenessered terresentat t sw cf the ther wind p!wer argulfer in the g
ne a-ore,e h dr==ine condwet t-
- ~ :. :.
che ei.re. some>we.t of the a-, r a t.
1 est, ere me.sinity: of nie - area se ro.,hi, root rer dar. A g
m e,.cific yie:4 et o.it for the ther wi.4 n!wer aq.ifer we. deter-i.e4 e
freun e owlt i-well gesag* test in this aree.
l
- 39 er e
se,et of the depthe to water in the Urf*r used alwes engwifer are in I
T JJ N p,,
the reage of 90 to !!O feet.
Water teeele in the ther hind p!wer ogst-44 gfc f3 fer near the n.9 pit have essen sprMter thee tem feet etace the start cf this f acilit y.
Croiand water f t,we f r en the northeast to the south-g meet i:.thee eres et a ea'e at 50 fee *. ger year in the more permeable
\\
[@N channel arese.
A rate of I gre is est imated me tSe asument of ground g
c=
w.ter t t.ee f iowis,9..h.,We, i d.l.e, e ifer 3 t so.u seet of the A-9 pit for a width of ScioO feet prier to collectioso.
l
- gidle0#4ff0E tS A-9 sawro.se cearent reticon mage were eased t-a define the 1941 ground-I
- MNOh
=,,,
ester goality conditiene.
F.gare 1 shows the 1982 chloride concentra-g, h
tiene for W tWer Wind giver angulfer. Chloride wee selected toscanoe q
I st le cemeervat swe me it errres witen tailinge seegwege through the typer g
e3 gggggr ggg 74fggg sind Clwer a<gelfer.
poet other major ces.stiteesta are elightly retarded
- .:.:.:O I
??
pg34 as they migrate threwgts flee formatica. Figure 1 shows that elevated g
chievide ccourentrattens esisted very meer the A-9 pit. The concentra-g 7
t.as.e m r - i t.f ~ h.e,se m. m rt,,.r.
11, tt,..
et. ~
e PW(4 U W 4 -"* "}y '
the=ght to 2. e eed by the seepage frem the inactive tastings. The eEPW2 gpw 3 su i
eeepage fr wt in the (184er wised Rie r aquifer wee within two Inundred b eamosessessmen g OS feet of W 8 9 teiltage in Igst 2.
g g9 Scales fo SOti
.u Crescentretiese plete for chloride and pst are showse for ese11 Ot3 in g
Fspere 2.
The p85 started 19 decresee while the chloride increased le 30026 PWS PW G 19#I. Asidicating that see-.ege had reached well ce). The changee 1" these 'wo parametere ese were cosictuelve that seepage hee reached this oC G
9 82 pertien of the aquifer. The 'Jg per wised River agaf fer meterally contalme l
l water with some low pg vetees, rigsre 3 shows the pet and chloride con-for well EPW7. wen ch thcess comelstently low get and c9tloride contratione e
M wolves since annitects=g startad in 1980. The pst eatwee of water froue j
teseoes raeser goercoatsoas emoef arv t est well E?w2 are meteral Specawee the ChIoride ConceptretICHe are Tery Iott.
E e M e EeB e M e euW e Eel e me e e o e em e ese -
Some decreseee 1, pat he,e orcerred in weter free ther wind River wolls without a correefending increase is chloride. These decreases in pst f
shoul3 be considered natural. nie data shows that conclestone about tsslinge seepage esenste act to mede free reestte of only one parameter.
FIGfpE 1. (DChT1058 ItAP Afe C381DetIDE C088CFNTRATIOft OF TNE UFPEP WIND RIVER AQUIFER, IN org/1.
The retardatiese coef fielent (0.0 el/ess) for redleen at thle ette la wery low, pretebly beceeve er.et ad===rptieve sites are already occupied try f
4
415 41%
radium that naturally magrated late the E M E Mind River aquifer as a ado ' a
- y result of uranium deposition. The reterhtion coefficient for radiuss is approatmately one-half the value for sulfate showiag that radium to more mobile than sulfate at this mate.
ReStum (S0 pct /1) and uraniusa a
o gas (1000 g --- gg go 548/1) concentrations are naturally elevated in the Upper Wind
=
C1 [
River aquifer since it le the uranium cae Imering formation, and there-fore are not very good parameters for detecting meepage at this site.
s2
,, g g
e h ement retarJetion anJ high natural levels of radium are not typical as a
for most uranium tallings sites.
'I
\\.'
- a o
i w
SEEPAGE FRCat A-9 Pl?
as u 20u g
Tuc, methods were used to antitante the seepage rate froen the A-9 o
tailings. One estimate was made by C*Japeting a teater balance for the he A-9 Pit.
This water % stance indicated that 18 gem of water is seeping
,g fram the A-9 pit. h mecond method was based on a calculation of the
.O lou
[**-----
- 4 volume of seepage which went into storage in ground water and lato the
~*~
partially saturated sone. The maall tranemitting st>ttity of the upper e
wind River equifer has caused most of the seepage from the A-9 pit to go lato ground-weter storage, volume of wates level rise asmi the spec-ific yield (0.11) of the equirer were used to otAain an estimate of 0
l-- -
'" ""* 9'
- * *#*9""
gg, gm 3961 IM2 I"I I 4 ct;Asin tFIJ. G81 rate is thought to be too large because acune of the a gulfer contains mudstones within the water level rise sone and p.whably stores water FI M 2-mzgr App gas CreCE3sTRATims5 rta isE1J. CW) at a storage value of such less than O.11.
Seepage probably flows in a partially saturated flow condition from the bottom of the A-t pit to the top of the t+per telad alver e+sifer (approskaately 40 feet).
Assum-ing that five p. scent of this values stores sewped water, then an addi-I tional 4.2 eps of meepage has gcne into storage in the partially satu-rated scae.
The quantaty of see,page water which has gone into storage
-5 indicates that the aJerage seepage rate to the fall of 1982 was about so 22 gym. The average of these two estimatec (20 ggwi) la probably slightly higher than the actual eeepage rate because both of these esti-mataa are conservative, inst it le nur best estimate of the average seep-i a
D 1 48 o
o
"~O age sate fross the A-9 pit between December 1979 to August 1982.
8
~
G E
-7 Isheen the original applicatiose was autaitted, the estimate indicated o
i o
that a seepage rate of 15.5 we w>ald be espected to occur during the i
N I
first 2% years of operatten. This estimate compenes f avorably with the U
, g observed seepage rate during the same time frame.
E st2 Pact rraaCT80es 5F5700 h
A co11 action system was designed to latercclt a!! neopage in the U
20
., 2 Upper blind River agulier on the dossegradient (southwest) side of the A-g pit. The criterion for interceptir.g the scopage is the reversal of the ground-water gaad4ent downgradioat et the collection system ao ao
^-
g ground water movee beyond the rowersed rene.
The Thets equation was g,
used to predict the dr - h ttut would result from pimping at differ-
- "^
eat rates. Dreudoesna at 400 and GO teet fri.ma the collection weit were g
lyst 1982 19'a n 1984 evaluated to determine what collect 1.ee rate is needed to revwrse the gradient at these two distanees. Thcee t w distances were used t=cause gg g n m gag Asa.g2 CDes M dAt10sts M N W a reversal at this distance from the A-9 pit woesto stevalop a reversal some greater than the width of the
-9 pet.
I I
48%
417 n, ~ rr,i l,,,the,ee,e,- een m i.e. r,. tee c. ut.d of -
REVERSAL WELL NOe I h.,h w.u.
. wnh a m.s.:,o. et. ore m e -. u was d. 1, d to
- a, ie +. ~ -o n.co.e.,et.e....e. wo n for. - = >
REVERSRL WE,t,L NOe 2 w
to.
~ ring m. um. ai ci,u.t. wee fer.is.s. in m wou.
..ns w e em.r,e it e.
.. t.te wnd., r eca f: s fr-WRTER ELE,VATIONS
%.r. et 2 -,se,
,s
. ma u.,e.
.a s..ma m f
1,ae.it u e per..et.d w.u etue - d.c....e m.. a, rect,i-_
l t.u.m in ut. w. n c.e,s.,a t u.e.,e
.r.
,a
, i,0,t,
.u, m.:
~ w. n - e re,iac a wi u, o,>.,. a e dif ferent r=1supletiews tectestiese.e wee used with the 9 mope tleet this might E~
elinanote RB.e 3 secipitate pret les.
This did retard the lose of well ef f setesw y. Qaseepo cununes s to feit_ and precipitate cesit tmood to to l
e protileo.
41tasc%g% a OB was remptag a* reduced flow, the hydraulle l
gre4icat downstreann et or3e was flattened diaring this period.
m.5 -
timetce received gvreano ton te AnetalI a necesid pemyback we11 of 3C, and a selferic acid drip eystem for weet pusspeech well in search 1944.
Os)C wee necessary teenwee one gensipha-k eelt could anot resume esionsgh wolume to reverse the hydraelic gra.taent.
The secosed gmszpback esa!! be-E~
I gam everating at the enJ cf searcse, ased W bydraulic gradient esse re-F f
verood within 3 days.
"he et*1clesicy of OP33 did est improve with tame I'*
j addition of acid, so it will Le replered wath a new well. h increased gasering rate f rrun the revised see;==ge cus31ection system will provide aa e4titlenet margin of estety f.sr casetalaiseg contaminant migratiose.
WES -
e H
De acid drip was twarg used e.a pe wwent the formation of precipitate E
J and to increase tane efGreesecy of Orla. N acid drip eyetee wee b
started in early April 1944 pri ping mutturic acid into two welle ed-J EE~
ge jacent to the coIIection well was eelorted a= the todiselease to prevent s e,*
l {8 t6ee aluso preelpit t_lan f roms forelseo me the focustian ased in the esell.*
s i
I A test ese the abas g re:1petete indicated that a pas of 3.5 woesto need to
{
be maintained *ee 3 sesent the formissy of t3ne pa ecipitate. Die indicatej g
- g a
that 9.0 galtcas/ des of one enorant sulfuric acid needs to be mixed with gg$.
T,#
'y, e *,,,
groisput irater collected at a rate est to y to redere the pas to 3.5.
De I
s acid drip system does as1= stain a kgo pu and gsrevente W formation of I
3*recipitate, but it kae armi a,a effective for dissolviseg precipitate
'g end increasing well efficaeacy.
~
4
- se The reversal of *.3a gradient eloesngradient of the collectices system 31 roves that the cettectiese se is tercepting all seepage in the 44ter arged giver aquifer. W criter!can fee the collectlen system to be oper-ating adespeately Wrefore is a reverned gredleet at welle sert and ser2.
WW.5 -
l F.Tura 4 presente the water-level elevettene in these two wolle werews time.
De gradient prie*r to collection wee frum esell part to Iser2, ishich 13 Recated ocethwest of piel.
rtie vales of the grMient wee decreased after the start of W firw* collect ness well in Pearch of 1983 best the EW s s a u a a a s s u a a e a u
i
,redt t w not new.r e.
N ga me:.et wee re reed shortir.ft.r m J r M M M J J R 5 0 N D J F M a M J J n' S 0 N O l
addittee of the necemnt cellection ese13 and Sna, meistalmed the reversat
{M IW ca.ee ap.n of i,... s.e s.mim..m e. of the reversal shows that th.
present collectione rate of approeirately nine ggma is adequate to inter-UtETCO MDERALS GAS HILLS OPERHTION eept the seepage te, the tester tr ert River aquifer.
a ricuns 4.
oestyyro tutTEn-1mvtr. parvatioses fogt PEvtRSAI. WEILT WI A8*
i l
l l
l l
419 A paes*=ets k m88* is enw i61 6as etefinang the zone the collection g
OLD TAttf4G,S - 12GO' /
(}z....
system as antesE*54 ang the f aow am the ugt.a 8f and pswer aquifer. tag-getg,. 83 DO e.,,
t o t m. s i,.... t e r-iece....t aos. -, e op, e j '[.*.,# **
gg..J aaver eq.na ter amJ mhows t sut flow times ese each side of the A-9 w
taalangs ese comweggag to t e to!Iectioon wolle. It also shows that I
l f
- (I"***
tLe ga sattent has tarea rowersed cia bassu1 red feet soutt.want of the collec-a v
.',-1,*..
t aos system at ell pas 2.
This mar shows that the 3-roment collection l
l E
- ;.h} *;'
sates ese amieq ately e steamans all of the A-9 eaepage that is in the I
-l."
- 1;r,...;,,
o,,e,..
.a.or m irer. c m e.tratso
- i. the og.ra wi.
.i.er ix.
I I
ecirc
.., i,.4.. e.a.
au t = -e.a,.4 5 13 2.
=ier,t i t, - itor-
[
j M* "
w@45rygg L_
O g.g an ee sa.ows that meepave to t'.e tsig.or 'ola.d stiver a gus ter as a)t cost-M s a a.1 to..g..te.Si-agradaeet.
l g
, #T*O, -
- f.. : f..
-. ::3 uw ra c<=. cam aa g
I O
se.pege tse the A-9 Pat Ta.li.ags Dingesel trea has occurred at a sete aimaler to 18me desigs.e4 est &= ate.
A sas'g>1.s collect &ost system has At 39 agr E
e a.staan.d ti.at i. interc.ptis.,att.,r ti
.e age 6. the opper wind yug h
ca.or p.ater.
Tsa. evaluation de.wistrates ti t a cettection system to g
ggg,,3
- s;
,.. ae s
o, me....s.,see,go.
l l
l S
ainsaxxs f
bWI NTD00, 142, *GrM-Wtes Elgdrologw Itear the A-9 Pit Below crede p
20 r,M Tant &%s.* consult ang separt by 'sytro-casineering for smiaan carbide EN
.::EldI'04M/OM:
ef
- s,,, N il A-S
- l:-
r q::::...:M #*::
cm po..t...
I roso
_7
!I m an.orter, o.wie s. and or-A a, bestor, 1979, w.gs calculattoes e*
ros twaasian tan t as es oise. sat Alternat&we vit, can mills o ratica, PE.2 - ;
s Qoser sq m,iass us.io. ca >ade corpo..t sos..
ca.s itig menere ro, unio. c.riado c"P*'-
I I
I l
pw 3 GWS tas a s, u.c., swhomeyer, P.c., and naaseg, T., 19dao a gulatory Issues, g
=
f agameer ang t=sigst, and "hts.ect iosa Coat s for Disgm1 of Urael:ssa still l\\
k_
O Tastamos an a maned-out rat " Tv ind synge iism osi usanium statt i 212ags
- EPW2 gGW 3
\\-
g
>" ' 'h' * = === s a..p, 3 ew4 \\
G wJ -
-s n,.. eat, c nora.hs st.t. t a.e sie.
r
- siw s 9
Scoles 1 500* I Rw2 srns g
srse s I
l
ern -
I 30026 PW$
PW6 i
j I
i l
l *==h 9 Low cuyciem I
i u
tarea ies miq g,s e emn*e=aan e es e an s an,s s ans a sen.so,e.. e t.,,,
u i
me eam oam,-
FICURE 5. efATER-LEVEL 4.'1E/AT10sss tyr THg g.ppgg yg,gg ggyg, ggg,y7g,.
IN FT-pCo., ApagL, g e,;sg g,
. soft slimes deposit at the tbas of stabilisation are therefore considerably less than with wet management. Hence, considerably less differential settlement can occur following stabL11:stion, with the associated reduced risk of cap failure.
During operation of a semi-dry tailings impoundment system, considerably less water is available for seepage, and hydraulle heads above the liners are considerably lower that for saturated er wet tailings iagoundment. Liner characteristics which are margins 1 or unsuitable f'or wet disposal systems may possibly be effective with semi-dry systems. After stabl11:stion, the total volume of water that can be released from tk impoundment is easil and a slow deterioration of the liner any result in low votumes of role.ses to men's -vies. %,4.gw,.A a s,g..us4!
(
7.2.4.4.
Dry annagement Dry tallings can be produced either by evaporative drying in layers and retransport to the eventual tallings impoundment, or by mechanical drying in the mill plant. Tailings ' drying' by belt (Litration in the milling proes s[O) receiving considerable attention in Worth America.
Tests made on it a r-various ore samples from Worth
?ican uranium tallings show li9) that f reu;-
moisture reductions betvaen 18 275 can be achieved in some cases. The Q?
handling of tailings at this low moisture content is dif ficult and presents
/,
some problees. Mills currently using belt filters sometimes prefer to re-wet hQl and purp the ta11 Lass to the layoundment la thick slurry form. This procedure
~* l can be modified to provide en swee11ent means for obtaining a thickened tailings discharge for the semi-dry managemer t systan, k f4Nc,4 l dd Altesmative methoda are sloe being investigated with the object of n 4
j adding further fluid before placing Os tallings. It is possible that by centrolled placing and spreadinge meisture centents staller to these obtained by the semi-dry placement teetwiiques seu14 be achieved. The slimes pond would be entirely eliminated and dust sentret may have to be effected using epray-on stabilising agents.
In Sweden, coeroe tallings (80s > 0.8 un dia.) with a l's w asisture content of approminately IF% are teveked to the impoundsent ae. coepscted in layers. Using this tecletique, tallings settisment is minimum, tailings 04227
. {
W I
temeability is reduced to approximately 10 m/s, and redon esmnation is
~
kept low. Further, it is possible to saturate the tailings af ter compacting l
h and in so doing minimise pyrite oxidation and further reduce radon emanation.
l
'Tkt r/17s cs4 L d X1 fev fa,teturmoy ?% A PM /11~ OptpAT/w /S l
h e/44t o m A7t1.Y
$" */o Of 0% s.estesV QfA T/oAs A 'T $*Tt/AMT7ed )J. C. $"
q /) TO A A W Y M7itC of Ws* 1EM., itM ADlwl O 70 fa6*1sovy op.pp y w/
8.
Treatment of liquid effluents and monitoring 8F M' T-
/*' W T-6 4 w TPIkwc.1, Q _,' ?st l
t%M r A.!
1 l
Where cliantic conditions require the controlled release of tailings water to surface or groundwaters, treatment for pH adjustment and precipitation of dissolved radium and other metals are nomally required.
Releases from the tailings impoundment may be as high as 5 tonnes of water for each tonne of ore processed 170), and, without treatment, may have ta concentrations of up to 0.19 Sq/1 (50C0 PCi/1).
s.1.
PH control
% y,3.)
Control of the p of the tailings water diseberge to about neutral (pH
- 7) usually ensures t precipitation of radioisotopes. Where warranted I
additional neutral ing agent is added to allow for in situ buffering of acids l
produced from p te.
When the precipitat.es settle the only potentially significant re ological hasard associated with the liquid discharge will be that of In Canada, la is accepted as a good indicator for tailings impoundment discharges as long as the pH is controlled above pH 6.0 h
(70). Such a pH adjustment will also precipitate most non-radioactive metals g'suchthattheconcentrationsofthesematerialsintheliquidarebelow n,kwd g,
acq=ptable levels for release to the environment.
g
& (kklk,' T% & a 4,e &.%s. (.
99 8.2 Redim removal Tailings water discharges, even af ter pH adjustment, contain dissolved 22634 concentrations which any be e Mers of anguitude above recommended criterion for release to surf ace or groundwater. Treatment commonly consist.s of adding narium chloride to the tailings discherse and running it into a system of settling ponds or lagoons. The radium is co-precipitated with berium sulphate by the reaction.
2 + Ea** + 30
. (Ra, Be)30 + 2C1"
~~
DeC1 4
0485y
l l
- 3$ -
A barium chloride concentration of 15-150 as/1 of weter being treated is notsally used. Under properly designed and operating conditions this method has been found to remove over 9p% of W dissolved sa from the liquid
[71). However, the precipitate is very finely colloidal, and the settling rate is very slow. With time, crystal growth and agglomeration result in sufficiently large particles that setslu to the bottom of the precipitation pond. Retention times being used at Canadian facLlittee range fros'obout two days to several weeks, although the effective retention times may be considerably less as a result of short-circuiting. It is believed that four days retention tLas is required to consistently produce liquids containing no 26 more than 0.1 Rg/l (3 pC1/1) of dissolved ta radioactivity (72).
However the use of barium chloride for the co-precipitation of dissolved 226
,,, (3,. Re) m 1phate asy result in treated effluents that contain,
3 suspended particles contaLning Ra resulting in much higher Ra concentrations than dissolved values. At one property in 311104 Lake.
Ontario. the total radium radioactivity is reported to be about 14 times that of the dissolved radLum 173). Removal of suspended radium requires either use of very large precipitation basins. additional chenilcal coagulant aids, or both, yor example, ferric chloride added at approximately 10 mg/l has been found to be effective. Ytmo disadvantage in adding flocculents is they result in a substaatial increase in the volume of W precipitates settled in the ponds. Filters, for removal of suspended Es, have been used in some
- places, sludges accumisting La W bottoes of the precipitation ponds can contain "Ra radioettivity levels of up to 220 Rg/s (60 000 pCi/s). In the stabilisation of tailings facilities. recovery and seperste dieposal of W ee redium-containing precipitates may be required er it any be possible to dispose of them La the tailings anas. One Canadian mill has a backflush on j
its filters that weshes W precipitate back to the taL11nss line.
The limitations of the berium chloride treatment for removal of 226 euepended Ba and production of sludge in W botton of large precipitation ponds has prouytod searches for alternatives for radium l
removal. Ice-eschange is capable of producing effluenta very low in
- ' ta 170). but ef ten the large volume of tailings water requiring treatment makes 0485y
w(5hf %W 1
[7 y]
en, ',
6 I
this approach impractical. A fluidised bed reactor used in canada fa + mn) improves efficiency but stiL1 has a waste product needing disposal.
8.3 Monitoring Monitoring during W development and operational phases of the tallints l
La,oundment is vaertaberto det.minec l
l l
(1) the integrity of the confinement systems and the available safety factors (2) cometience with authorized criteria and release limits (3) the validity of the models being used to predict the future confinement performance of the facility so that the stabilisation methods to be undertaken at the close of alli operation can be planned and implemented.
To achieve Wee objectives, monitoring of the impoundment f acility and on,the l
nearby environment must be done. Monitoring af ter shut-down of the mill plant and stabilisation of the tellings is aise, required to ensure extended protection of the environment.
l 4.3.1.
Impoundment facility monitoring Where applicable, monitoring of W confinement strveture. controlled l
releases, seepage and the general facility should be carried out.
1 Confinement monitoring l
Quality assurance on das sonstruction asterials P and W M
methods and scheduling of (as cenattvetion are required to ensure that odequate fasters of safety are amintained. Safety fasters determined at the
(.eaisa stage are beoed on the physiasi properties of the anticipated construction asterials. If the proyecties of the actual construction meterial differ from these anticipated. W re any be a need to re-evaluate the design.
The saturated surf ace through the layoundment embankments should be monitored and N results used for re-eveluating the stability of the embankaant. These data can aise be used to estimate seepage flows through the embankment. The performance of liners and drainage systems should be monitored to ensure that W y are functioning as designed.
068Sy
_I n s e r_t into disqussign_in sectign_8..;
1 Some studies have been performed which indicate as much as 94% removal rate by using concentrated inorganic acids or organic chelating agents (OM HNQ or 1.5-3.0M HCL).(19AJ As with the use of barium chloride, there is a precipitated residue disposal problem. A further concern is that removal of radium new does not, in any way, treat radon releases in the future, since radium will grow in from thorium. In 1000 years the thorium activity present today will result in 06% percent of that activity in radium. Some of the extraction processes (HCL) also remove thorium to some degree as well.Ci9A]
)
b
[
/ 4k- /Ws b/7N) f l
l 6
p
. Dans should be visually inspected.c a routine basis and observations of anomalies carefully recorded and evaluated by the engineering staff.
i Monitoring of controlled liquid releases to the environment 5 d bb>
hh tihere effluents are released on a controlled basis, routine monitoring
=hauta 6 daa tar 'ta =ad eaae atratt aa at: "'na '"u '"r$ i '"i solids content; environmentally igertant metals; nitrates, asenonia, etc.
y p
{ wO' wwwwy c, )
age monitoring P*
Monitoring is required to determine the existence and extent of seepage from the facility. This may include sempting, as appropriate, via seepage collection devices that are located under the tailings impoundment area, and/or using monitoring wells (loested selectively around and near the tailings Lapoundment) that penetrate geological formations which are likely seepage pathways.
g gh Qg
[,
[,c @ ft l
General facility monitoring Visual inspections of water diversion systems should be carried out at j
least on a seasonal basis to identify as soon as possible problems that may influence their effectiveness. Te111 ass pipelines should N inspected a
frequently for ses11 leaks er weer and, en a seasonal bests, to determine the stability of the foundations supporting the lines. Care should be taken that snow or plant growth does met eboeure a potential problem.
8.3.J.
Davironmental monitoring l
A progreene of enviremmental monitoring is required primacL1r to determine the impact the tallings fosility is having en the health and safety of the public and on ties environment. If adverse impacts are found, the requirements for remedial actions een be quickly detetuined and the appropriate actions Laplemented promptly and offsetively.
To evaluate the changes in environmental levels during and efter operation of a uranium mill, it is necessary to make pre-operational I
068Sy
.. e.....
- es -
asasurements of the radionuclide levels in the environement. of particular laportance are asasurements of En concentrations in air and ta levels in surface waters and sediments in the area. laonitoring of ground <ater at the tailings disposal site befo o the mill plant or tailing impoundment begins operations shew 14 also be performed te establist a bat eline for operational monit.oring. The prograssee should include analyses for U, Th.
Ra and other toxic metal and chemical eyeci espected to be present in
( Sy 4, sof A...
tauings sotuu.s.
n the event of an accidental release of tailings materials to the environment the tapact or potential impact will require monitoring. This type g pd of contir.4ency monitoring should be plar.ned in advance, keeping in mind that
/D its main objective should be to obtain information necessary to assess the (6
situation and to decide on the need d spesific actions for intervent on.
a C4c4 of ab.IC-L N M85/ *b y
"Id g
where2s MM overs 1 obi.'ctive of tailings dieposal is to emplof 5
disposal tectatiques that do not require human surveillance on a long-ters basis, it can not yet be assured that such an objective can be reached with the tectrtology currently available. Thus the tM exists for periodic ig/i% "
wurvel11snee and enviroraental monitoring af ter decommaissioning of the mill
\\W V< V an6 stabilisation of the tallings have been completed to ensure that kg o u.
successful rehabilitation has been achieved.
/
h jf.g kt4 5 h Ch 4 hp8.$OfJ kg (r The objective and design of environmental son oring progressess for t
radioactive contaminenta are given in references
, 75, 16 and 77.
Typ** af I
i samples taken, sampling and analytical methods, and the location and freM4 o
of samp11ms theuld be as approved by the appropriate regulatory bodies tidl.,
i 9.
Stabilisation and rehabilitation of impoundments Atter teralmetten of the active operations at a mill and its associated tailings impoundment f acility, a stabilisation programme should be undettaken to provide assurance that the future performance of the closed wt f acility will continue to meet the requirements of public health, safety and environmental protection.
i Typically, the stabilisation programme would include t'w following j
actions:
as s e.,
- e. -
s==*
omenawG SS-16190 f U C If 0 e
f
.m
~
coas s seer..a imx
, ~.
%'-= h
" *~.
r.
- a
^~o auseu 1na.m, x
a
%.f////
~
c nm c ar c 9
'Mmi//, x//Ab,ib, lf/
WL
/ - u n
*($ enee sco re.ea, g
/
co.
r..-
i.,
iu
.-<i i.i<
im o.- i~
r-.-
i ',
t'" c; i - hi i
.,m.
i:
F
\\mes.m. t==e (r
,I.
pi e,.a.
- c. anne-e e.
e.,
iIL 3 ? -C '
M:
e
]NT!.'Q W % l i
, e a.s e e l
.i-V. -
o
. ?m. ~ m i'
fl4U$6 i$~ C.
- r: - i v.c_ n-_
y
-.- p. ;,o e, 4
.e Q Me%ee!904W W O F*.
\\
4.s '
e -
.f. -
Atsom: suarart
.n w.
g a. w. v.,,.,
/~h
/
w oroe on a ras i
fea e -l k
yr
~~
/
s
.s
~~ % ti,
? g,.
~ m m.
gtg-,.-
=
r.
.,a
##/##,.#,/###/#E4'7##/#.../#..#/##/
/
r,S ou er o f,,,,E
,y c
~
am a ine n.- amn er surraan s-sua~s was-r r s n rnes (s) 6 g Asty F/* '1tK.
l'TK A C.G A W.A TC g<,
f.,a. /.,,t.'w -
u r-(Q C Lffiel CA A Ut
.ne 3 ~ ~ 1 :,
g(.
.~
,=
D)
(g)isc p,A. h/(
, r.
,..*[, ' ~ *
- 04'
~
m,..
G < #1f/d 4 :
- 1.** *.ticI-
~
. y g",b j
g
. g sa e fa) (3) a)
~.
F#47EM I~M 4 C
=
N.
,f t yg.,
) (3) j
'59,-
we
-a.e a g j f_/
.=
,, ~
r Q r15't.eoe.f//
dt: f r%eoo.f/4 I
g 8
. 'o a) general preparations such as consolid ng miscellaneous westes into the mein tailings pilo; b) dowatering the plie; 1_
c) stabilising the control strvetures if the shape or guality is not as k
required d) ensuring that diversion ditches flood control, etc. are up to standard; e) selecting en engineered cover or liner to control seepage, redon ensissions, surf ace gamma r sy dose and erosion control.
=
d 4
Ideally, the plans to stabilise and rehatilitate a tailings "acility wul' be made before 3 P,nt construction so N wek and cost would be kept to p
a minissen, while obtaining maxiwa environmental and human protwettua. This has not generally been done in the past so stab 11uing and sahabilitation measurse had to be adapted to site specific problems.
k o
Whatever the hirtory, the objectives of the ramedial work are tot M $ sk (a) ensure stability of the tatning structures and tallings mass 8 (b) castrict liquid, solid or gaseous releases to the minism ar,eepted by the appropriate resuleting esenc.v (c) ensure Nt the safeguards will seriet ever the long-tors with as little numen intervention as possible:
(d) restrict acc6ms to the site such tS the pots,at.ist for mieuse of tailings (as bulf ding asterial, etc.) is minimised:
(e) ensure tant the undesirable inyecta of the facility decline with ttne to a state of minor cor.cern.
f MA YnVVvw. Ck*N t,I) m & d u c.> & su t
9.1 General preparations Setore W,ta'ains any physical wek to atabilize and rehabilitate W sailings every effort shenld be made to acmamulate wastas from the operating
{
- y.,
- Dispread t.allings from assidentel liam "uoakates, windblown material.
i area.
1 g,,i
\\
das breaches.%pilasy failures sad so on should be brought to the '.Aln kt k
.C
- tallings area.
V c operations us'.ng s. r. tium removal system auch as barium chloride coprecipitation or the new fluidised bed-barium c.1.. tide injection, disposal 0485y
_m 90 -
of the precipitated material would have to be arranged [79).
Th' current proposal for radium barium sludge disposal is to bury it 1.4 the tailings plie. Generally the objective for the tailims is to keep them as anaerobic as possibic, to prevent oxidiation and mobilisation of pollutants in the
/
I tailings. Recent research has shown that if sufficient carbon based nutrient is present, bacteria 11y-mediated reduction and redissolution of the radium will occur. Typical.ly, tailings are nutrient poor (unless added for the vegetation cover) and avon if dissolution occurs, the radium will be severely retarded in its movement through the tallings.
i 9.2 Dewatering l
Tallings are usually delivered to the disposal area as a slurry. The l
moisture content varies widely from mill to mili and depends on the ore I
minerology (principally the clay content) and the technology used. Vacuus l
dises, drums or belt filters and similar techniques can reduce the moisture content to 20% or less for low clay ores.
In the Lapoundment e.he tailings will cons 311date, expelling water to the surlace where it drained away. Euporation will also contribute to the drying out process. Consolidation by self-weight een be enhanced by adding addittoaal cover (111 to induce settling.
i 4
Internal drainage syst es installed prior te deposition will enable the
[
operator to accelerate W removal of pers water at eleceout. Underdrain
(
gravity-flow systems Lastalled before depositten have the highest probability
[
of remaining offactive for lens time periode eM grevide en econeetcal means of deweterias (34). For existias or Laattive tar
--to where underdraina have met been installed, horisontal dreLa pipes een be placed through the confining embankment and into the tatitags. The offestiveness of the he-issatel dretas ai.wedt on the depth at etich by can be placed and the length of drain that can be drilled into W tallings. Other techniques such as atagle ese11, wo11 peLat syetems. vortital deains, evopetrenspirstion and I
electre-eemesis can aise be used whert underdrains were not built er if they have bloc 144 up.
f f
I r
9.3 Stabilising control strvetures
'Y
/~/6 eAf pf fgens2 $ 0 tst f T Wif t4 L V "0 0 0
.nsy,~~u w,c-n wa s um on u.a um~ v ncn no ou a snrw (s); ons oe.aa or pruonne sewer &h co as Asm a r now A A rars ptw. 94 Aous (o) ;
A&9 A sTLt.
h*D D AM S YTTEM (E).fwe /cyanweg-y, l
e-.. - -..,.-
-v-,..,,,----,n----.,,,...---,,_,.._,_.-_n,_,,-_,,_-.,,w,--n
,.-a.,__n,..mnm,n-w-n,.-r.
PRECIPITATlON:
EMPola ANSPIR ATION:
34.1 IN/YR 25.6 HIYR FINAL ORADE j
p VEGETATIVE f [ N, kCOVER tC
\\
Pu'b e.. Afy%.:?.T
'if'..i
.,.s
- e -
t<
aw is m a
TMiCR
' d. -
,,,ct RUNOy*.
c.
1" l
Sql gy g
A 8
- O E-is m p
T,.ic a y
'9
- v. cot.aTc=
pen
," *". n
. % j...
n
},0 ovo so v it.s = " a ao o
, 4,y,S=
in m
' 4 ;,,
' 'N
' / /,,*
([f
/
c
,5En l
/
"~'"*"cw l
curt sue j ! y og
/
o
..... '. [.
/
W<W
/,
,.e o u vio= m a ovoa g$,
E covie i.: w va
- . *
- s..'. *l *
- W I.
T n ic = n e e. ! '.'. * @.**,**.
"."t. ato e=.,,,,..
,. 'l2. l
' *f,..'..I,*
g2 yames,
,....lAsus6 w
...,i co W.
moes vaa,,
\\
5
}
.,., d.,.:::.:,-
\\
,g 3
bg TMtC K tw<
\\
m Rg
/m
/
j j
e
/
/
$ 8'd
,/ /
' '~ M::.:.... Wing.:4[$.,,:
\\
" "%. e W
,. g
{
aera.e g x
i ACo(ATcw esto amisTwo sous.: wva
{
l FH3URE W /(
PROFILE OF ENCAPSULATION COVER AND LINL9 WITH WUER BUDGET RESULTS - CANONSBURG SITE l
- il -
t after the general preparations for closeout have been coupleted, the ceredial planners should ensure that the main control structures were built to ensure tous-tem physical and radiologiew1 stability. Factors which een lead 7,
to unstability in the long tem include: goonstry (slope) cf the retaining wells, material properties, t reatLe surf ace and earthquake loading.
h t
k.
If after analyses of the structures potential weaknesses are discovered.
remedial action to correct.these weaknesses wm:14 he required to ensure b
long-tem sbdility.
gwc '.
f 9.4 Diversion ditches and spl11 ways OJ.
/
Diversion ditet.As are usually required to carry
-of f frw the catelseant area, and reduce infiltration thereby 1----
' 7 s*??) the teachate volume. In designing ditet.as and spillways the 1Lhely flow velocities should be fast enough to prevent sedimentation und slow enough not to cause erosion *.
Ditches should be wide enough to allow for partist bloctase by ice or debrio.
sventually ditches will fall as they become plugged with debris. Every attempt should be ande to minialso the diversion requLeements and tNs avoid long-term probleme (30).
9.5 Barriers to peopage Barriers to oeopage fall into three anjer categories. These which haie minimise soeposa of water free the tailim6s (1tr. ors) esa are normally put in place before the facility starts its operation. Various sci.uses for Pht lisuses have been used (Session 7.2.3).
Were conditions eemit, i
it is seesttet.Le not to have eny liner. Some facilities have used natural materials, sucu as slay er bestanite-modified soil. Others have used cynthetic meterials ist. 32).
The second type of barrier is the esp which prevents infiltration and thus lowers the potential seepage volume. Staller materials to tinose used to line the basin are used. The desir!'.if a esp (Section 9.6) will ir:1ude other considerations such as reden reduction, erosion control and doethetics.
l V*gstation may or any not effect the inflitration rates depending on the local I
I conditions.
l 0485y
l
..e
. The third type of barrier is the interceptor barrier composed of some material that will alter the characteristics of a groundwster plumes. The barrier is placed in the pluume flow and, for exaraple, will consist of limestone to modify the pH.
Other materials will produce other changes.
9.6 Selection of engineered covers h} G, M
The final stop in the stabilisation of a uranium *.allinss Layoundment is the selee the engineered cover which will control redon smanation, i
seems radietion and erosion of the mill tallings to acceptable levels.
p.3 k*
In section 7.2.2 the types of cover asterial which have been used in
&[
different countries were reviesed. In this section the types oi sovers required to tasure that the releases from the tailings pile are within limit,s and the long-term stability is controlled are examined. The actual design selected for safe closure will depend on f actors such est c11asticcond'iti$ns, available materials location of the pile, economics and population distet% tion.
./g, f
TnTi In t1m U.S.A..peandidate cover asterials &
g has been performed and authorizationha been given for ursnium mill tailings Layoundment covering 4
comprised of layera of clay, native sei{ sad riprep. Usually the strategy saderlying the choice of sever asterials addresses not only erosion protection, but slee rede attenue* 4.on. sectasimical stabilit/ and ground-6eter protesties. Envir m al fasters play a large role in the selection of the oever materials, as we'.1 as the ultiente sever configuration, flieA t li $Meeer &CL652 Sgtprsu 89 1;99 Cev A M*O L/N fit.,
WSO AT TH CAnlaut derm t.*TV~ sW Tks' V M [53)
N 9.6.1 Osetrol of reden esdesiens from the t r ^--
.t surface l
Various manufactured Luperanable asterials, such as plastic, seil/coment mixturse, and sephalt. have been sus pated as a herrier to prevent redon amenating from a tailings Layoundment arSt. Newever, the intag-ity and durability of these materials ever the long term is questionable, purthermore. the cost of installation in mest esses is at least as high as for natural earthen sever asterials. Natural esi! sever will reduce the radon omanatisa rate and can b. relatively stable over the lens tars, utfferent soils will provide variable redon flux attenuation depending for the rest part 04ssy
, on thele effective diffusion coefficient for redon. The most significant parameter in evaluating the ef fectiveness of en earthen cap as a redon attenuatLon barrier is the noisture cretent of the candidate soil (83). This is reflected in the significant dependance of the ef fective dif fusion coefficient's formula on the moisture saturation fraction of the soil matrix.
The ability to maintala optime condLtions may easily erode over the long tem. many comp
- tent su h rities require a "reasonable assurance" warranty on W espected performance of such cover' systems. [. $ h I.}
I f
If N desired perfomance criterion is numerical in nature, such as a redon flux or a redon coceentration limit, the congetent authority may elect to use a fraction of N resulatory Limit as a criterion for redon cover l
I design. For exemple, if the regulatory limit is 0.74 Sq.m'.e (20 PCL.m-2,,-1). then a safety factor of 50% would determine a soit cover configuration (thatknees type of sell, etc.) whith should reduce the flux to i
0.37 Sq.u-2,-1 as a best estLaste. but with reasonable assurance Nt the 0.n
.m
.e
- i. achi..e..
. t..or fr.cti. is.eleu e.. it s W 14
-2
-2 reflect the range of possible senditions to be espected at any reclataed taL1 Lass site. This fraction sould even te set en a site-epecific basis by l
the competent authority, depending en the level of sentidence in characteristas presens and empseted processoa and conditions at such a site.
i
)
On the other hand the tempotent authority any decide to focus on one or j
l more of the co'.1=.*d.1 factere used La predictims N performance of the soil l
cover strat4y La light of the sammerical stenderd. For instance, the moisture i
l saturation fresties of the soil sever asterial eeu14 be senservatively banded i
in a resseeable aanmer by requirtas the wiltias point moisture content frestaan to detersiae the thicknees of a seit asterial sever. The witting pelat is a site-speelfic value at which native plants een t e longer derive
)
asisture from the surroundint ***
- Ach a choice for the estature fraction would provide a reasonable bound
.atermine how dry sendLtions could becoes La predictLas the omhalation of reden g:s frsa a rectaland taillags pile.
i i
hre competent authorities require work-practice standards during the operational phase, thest any prove to generate as many problems as they do solutions. For instance, see of werk practice criteria as Lupervious liners, asLatained saturation or near saturation of below-grade tailings Layoundments.
'l 0485y i
-__,___.__m,
l l
I
,\\
,a.n.r.t..
1.,.
.....e p1... _ t.
1ff. +..,..tt1
.t.
- 1. t.
long-ters (thereby causing cracking and loss of uniform cover) and other difficulties. 'Ehese approaches need to balance reduction of releases over the short tars trith the reduction of potential releases over the long term.
Stariifed anthods for controlling and estimating the radon released f rom the surface of tailings piles are described in sub-section 5.3.2.2 and Appendix 3.
Redon exhalation rates are governed by the amount of parent radium and the percentage of moisture in the tallings. Therefore in areas where the moisture content is normally high (Canada for example) redon exhalation rates are low. Under these conditions little or no remedial action is required for redon control (although cover asterial night be required for other purposes). Typically, as well, northern areas are frosen for some months of the year, thus reducing further the total annual release of reden.
9.6.2 control of samma radiation at the impoundment surf ace Some sessas radiation attantuation from tailings is achieved if water is standing in the pond. Newever, control of gaens rediation during the operation is best schieved by progressive er latwin reclamation of exposed tailings beaches. The greater percentose of radioactivtty remains in the slines. >vt general levelsCeT10-35 R/a (is this correct ???) bys been measured at dry uranium mill tellingi sites in the westets United States of
- America, i
d 5 L % )-Q,;>
d(
6vo<c 3412/y Dss osssy f@
hvMG-c5%)
1 1
I
l l l
The sensaa radiation level can be significantly reduced either by covering tailings tapoundment areas as they dry durin6 the operational phase, or by using a soll cover at the end of the drying period, following the end of the operational phase.
Generally, a cover thickness in dry areas that minimizes redon exhalation to near background levels will be more than adequate to reduce the 1
extemal gasma-exposure rate on top of the pile to well below background levels. Procedures for calculating gasma attenuation through a shielding asterial are available from numerous reference sources, for example reference 11.
9.6.3 Erosion control bieter and wind are the primary causes of erosion of tailings or tai 1Lngs cover meterials (Section 5.1.2).
Steps should be taken to preclude the flow of water directly over tailings. Good surf ace drainage of the tallings cover con be effective in minLaising water infiltration t.o N tallings without the need for concentratLng water flows that could contribute to erosion. The two most important facters in contret of eroeLon are sitLng and design. Locating a tailings disposal in en area, to the extent permissible, which minimizes the throat and Lapact of significant er repetitive lesser erosion events is the ideal. However, other facters play a rete such as location near a source of uranium ers, amor a transportation eyeten and near a source for process woter. Likewise optimising a dealan to mint =tse the ef fects of signLficant retafall ovents mey result la Laeresse in other undesirable consequences. For example, very fist slopes for t'
,11 eever (en the order of 100 on 1) may sianificantly reduce the orteis.: impact of a retafall/fleeding event, but such a gradual sloce any incessee the infiltration of water through the essur into the teillags themselves.
In short acy strategy used to mitigate a sLagle aspect of the tallings stabilisation question, needs aise to address the perfomance of the stabi1Lsation/ isolation strategy as a whole. In the previous example of the gentle-stoped soil cover, use of a drainage layer of one11 to medium sized rock or cebble followed by & relatively impervious clay cover layer could redeen the seatle soil cover slope design in that the int 11tration may not be C645y
o a significant problem and, in addition, the redon attenuation capacity is improved. The clay cap provides a barrier against both fresh-water inflow and redon osanation to the atmosphere.
While the pri. mary means of isolating mill tallings must be physical barriers, it would be prudent to have some continued surveillance as well as control of land uses at tailings impoundment sites. These ac.tione should confirm that thece is no disevption by either natural erosion or by human or animal related activities.' The added benefit to this continued "presence" could be to acquire data on the actual performance of the stabilisation/ isolation strategy in light of the predicted perforsance.
Certainly, this continued surveillance and monitoring should be performed by an organization with some expectation of continuous existance. Ownership and custody by a governmental agency, as opposed to private companies, appears to be an effective way to provide this supplemet.ary long-term centrel.
10.
Application of Radiological protection principles LL.-
yequirementsforprotectionfromradiationasrecommendedbytheIAEA 113)gIcap (14) and applied by national authorities play en important role in hW the choice of mining and milling mothede, in the selection of sites for weste retention systeme, and the choice of designs for the tallings impoundment and
)
stabilisation for the long term. As given in Sectiers 4.*J. W besic 1
requirements are that radiation deoes esoulting from W rettese of i
radioactivity to the environment be justified, be kept as toer es reasonably achievable, economic and wecial factore belas taken inte account, and that the dose equivalents to individuels be maintained within suWrised 1Laita.
l l
To emeure that these requirements are met radio 1+sical protection should be optimised at the planning and design stage. pres this information, suitable choices of engineering solutions are effected. However, the optimisation of radiological protection is not yet performed in a formal or rigorvus way because the methodologies are currently in the early stages of development (44). This section introduces W concepts that are needed for optimisation of radiation protection.
The principal difficulty in optimising radiological protection for 0485y l
_____-.___-.,._...__,__._-.__-.__._._____-.______._______,______.________.,-,___,___m,...___.___
l I
l Comment on abbrevi31tng__the text Quite frankly, I would eliminate chapters 10 arid 11 as they are now. I would write a paragraph or two to summari:e t.9e headings and then reference the IAEA documents from which these chapters have been taken. I would try to keep enough text in order to keep Appendix C, because this would truly be useful to those 50*4 developing country readers.
I
. management of mill tailings lies in estimating the potentLal radiologica'i impact in the long tem af ter active operations have ceased at a tailings site.
yirst. this section briefly discusses the Lay 11 cations of current practices in providing for radiological protection associated with tailings annagement. Second, some discussion is presented on W fundamentals of optlaisetton of radiolosteel protection and the various dif ficulties that are involved. Optimisation methodology can be applied to the choice of methods for milling, the seier. tion of sites for weste retention systems, and the choice of designs for tallings Layoundsent systems and thele stabtLisation.
The optLaisetton of radiolegical protection constitutes only a portion of the plannLng and design procesess, and numerous other consideration necessary to arrive at preferr M allt tallings annagement schemes. When a cosmuretal operator is responsible for the closoout of a tallings impoundment, the j
coegetent auWrity should ensure Wt W operator can guarantee that sufficient funds will be avallable for W closecut and subsequent monitoring i
activities (1).
This section does wt attempt te provide information on the determination of the long-tecia radiological sensequences. This has been 4
coegrohensively reviewed by (& ly.
g a3L 10.1 Engineering practice and dose limitations The techoslogical features described La the aer11er sections include j
l some of W best tedutelegy ft c utenium mill tallings management. A 1
well-designes had managoc tai 1 Lass a_ 7_-r y it system will W refore implement i
an appropriate seImbination of these features and anchantees. Nevever, the perticular sLtos: and designs that arv seceptable and b asseures included in stahilistas the tailir.as ig:r? - te depend critically on sits charsetoristics and the leest envireamental tenditions.
For esariple, processing very high-grade ores (>2.5% U 0, equivalent) 3 Ln the ug opulated areas of northern saskatchewan. Canada, which have a cold c1Laste and high precipitaties tenditions, results in veey dif ferent tallings i
i sensgenote requirements free these appropciate for tallings that arise f rom t
the low-grade ores (<0.02% U 0, equhelenu Wt an mined in goWurah 3
}
mines situated in the densely populated industria1Lsed Witwatersrand area of South Mrice.
0485y i
o If appropriate control mechaniens and technological features are included in the design, operational snd stab 111sstion phases of a tailings impoundment system, it een be espected that compliancs with the appropriate national limits for radionuclide releases to the environment can be achieved during the operational phase of the layoundsent system 185]. Because national l
authorities generally base their 11alts on Icap recommendations, compliance with national authorised limits, ohich are sometimes more restrictive, will meet the Icap reconnended dese-equivalent limits.
In the post-eleeurs phase the tapeundment system can be designed to remain intact against the nomal envirorueental influences that cause e gradation ever time periods of the order of 200-1000 years (86).
Consequently, the release of radionuclides een be espected to remain within the appropriate national authorised limits during this time period.
Furthemore. Laplementation of continued surveillance, and maintenance as necessary, necess sentret and Ivnd use restrictions by W eppropriate regulatory bodies will increase confidence in the sentinued integrity ever i,hese time scales. getablishment of a long-tors maintenance fund independent rf W regulatory authority's regular budget will ensure that care and surveillance of N stabilised site will seatinue without disivptions. euch as Wee caused by eyelee of departmental fiscal oesterity. Arrangements for funding long-tern care should be established as early as possible in the operational phase of W urentum mill.
Beyond the 200-1000 yeero time scale, h difficulty of quantitatively l
prodActing the netwee and eartent of both sivilisation and er.vironmental (euch l
se stiantologisal and seemorpheleginal) ebanges askee it layeesible to be sentident that redtemustide rolessee, will met emened W au h rised Limits.
purthermore, se these time ocales asjer asturel wpheavels, such se emethquakes and fleeds, of angnitude esseedias the design beoes for such events any occur.
Htish eeu14 result in partial or subetanital damese to the !;rr'nt system and sensequent incrossed radionuclide rolesses.
The radielegical consequences of such a failure can be assessed by the postulation of failure and consequence models. These models are generally somewhat arbitrary or conservettre in the saetaptions on the angnitude of the radionuclide reiseses, the t.redtepert nochenisme ed other such mechantees that l
0485y
. influence the exposure of the human being. However, due to the relatively low radioactivity level of tallings, the radiological consequence of a serious f ailure sould be of chronic rather than en acute nature. Thus, enly long l
continued exposure would produce detectable adverse ef fects [section 5.2.4 of reference 11). Stailarly, prediction of the eventual increased radionuclide rolesses due to natural deterioration of en Layoundment systee at times escoeding tens of thousands of years after stabilisation, although theoretically possible, mast be af fected by the Laprecision Laplicit in the prediction of important parameters.
10.2 Optialzation of radiation protection To keep radiation doses to humans as low as reasonably achievable.
l taking account of social and economic considerstions, optiaination should be l
applied. This means that when further reduction in dose attribiltable to a given practice would not justify the "tacreaantal cost required to accomplish this reduction W proposeJ reduction of radiation dose is not justifLed.
Ysn a He esther To e! Tie 1st AT/ww Met Ac4iOtM9 ism etmyMg ear 71&tTn l NM 15If tw e M %T!=ATTJr!$rIT W/1 L yfe'f 10.2.1 Differential cost-benefit analysis This approach to W optisisatim of radiation protection consists of an analysis in which h level of protection Le increased increaantally, thus incrementally reducing the radiatten detriment, to a value such that further reductions in the detriment are less significent than the additlenal efforts required to eehieve such reductione (1). Empressed mathematically tMg involves defining and estimatias the seets of a rense of protection uptions, such that the latesyley betiveen the seet of protection and cost of W aseeeisted detriment sanglies with the relation I(w) + T(w) e alnim a shore I is the cost of protection, and Y is W cost of the radiation detriment, both at a level of protection represented by w (e.g. tailings cover thickness, alternative options for protection of groundwater, etc.).
It should be noted Wt w,1(w) and Y(w) son in some cases be continuous variables, while la other cases they take only discrete values.
The detrimert Y is supeessed in cellective effective dose equivalent coenitment S in units of men.Sv (or men.res). It should be recognised, that 040Sy
- 100 -
practical problems may arise when evaluating collective effective dose equivalent cousaitseents. The IAEA has published specist guidance for this purpose [84): noreover, several international organizations are carrying out specific prograsenes to develop further guidelines.
Some additional difficulties sesociated with optimisation of tailings management are still not resolved. For instance, since tailings contain such 0
long-lived radiocuelides as U.
Th and
'Ra, in assessing collectiva dose equivalent comunitments a relevant factor to be taken into account is a realistic period of time over which the integral should be autended. This is particularly of coacern when estinating the total detriment I
caused by an operation. For instance, estination of the contribution to the collective does coesaltaent made by radon release from the tailings pile remaining after the active life of the alli antal's sophisticated guesswork.
Radon emanation continues for millions of years be.euse of the long half-life 2M of the remainine Rh and
'u precursors. However, for such a long time,
perind. the actual physical state of the pile and the population considerations are only a guess.
1*
I However, in the optimisation precedure, the question should be limited l
l to assessing the detriment saving by introducing a epecific rather than another engineering solution during the time that these solutions will remain l
in operation. la optimising tailings amnegement, the change from one strategy to onether aan only be achtered La discrete insrements of hth settt I of protection and seet Y of detriment. If one stretegy has coets of protection and detriment Ig+Yg and onMm strategy I, + Y, Weh is assumed here to be more eseensive in protectise and sentrol), optimisation en a techmial basis of saias from A to a would be if I, + Y, < Ig+Yg or if I
~3 l
A
< 1 Y - Y, g
0485y
- 101 -
As long as this expression is less than one (<1) the change to strategy 5 is
~
cost-ef fective er.d W optimization is achieved when the expression is equal to 1.
As satt. from this formula it is not the absolute values of I or Y which are of interest but W dif ference between them.
These optimisation assessments requirs dimensional competability between cost of protection and cost of detriment. Usually for solving this conestability problem both the cost of protection and W cost of radiological detriment are empressed in monetary units. As the radiological detriment is proportional to the collective ef fective dose equivalent commitment (S.C 'I E
this proportionality any then be represented by a dim u sional term e and the problem reduces to the assignment of a monetary value to this ters. It aust be realised that this assignment is a social value' judgement, re W e then a scientific determinatice, reflecting how much a society is artlling to pay to, prevent a statistical deleterious ef fect. Generally this value will be established by the competent authority or some other governmental body. In the abeence e-en established value, a value, or range of values, any be assumed f,r che purpose of carrying out the analysis.
DistrNtional prehlems occur when people receiving ths detriment from the practice would met receive W bor.efit is equal degree. This problem of cost detriments, and benefits distribJted over dif ferent peyulations at different timme is souplex, at least from en ethicai viewpoint. It can also involve petitisel and legal templexitieJ. New this problen affects the value of a ten best be judged on a nattenal basis., Newever, a uniform approach requirse N t the value of a weed La optimisation should be applied to all sellostive deses. The volve f u n 11ective deoes appearing La the distant futaere mer met be e problem La the seee of optimisation, because in practice the lea 6 term centr N tions te se11estive dose commitments free W various optises under sensiderettee any be identical and concel out in the coopertsons (87).
In cases where dispersed radionuclides cross national Soundaries, the value of a used should not be lower, when applied to the owr countries, than the value applied within the eeurce country and should not be lower than an internationally agreed value (SS).
In many practical cases of optimization, the changes in proteutton 0485y
- =.-
- IUT -
levels are acfileved in finite increments, both I and S.C being discrete t
instead of continuous vettables. The decision to 30 from a level of protection A. represented by a certain desi5n option, to a more orpensive level of control B represented by a different design option, would be taken if b~A 1
e
~
E.C5 8,CA The optimisation assessment in this casa consists of a step by step 1
procedure that has to be verified to ensut a that at a giwen apparent optimum no seditional stop would make the ratio - A1/AS
- "*I E.C the value of e.
If a change of strategy is cost-effectiva, it would be suf ficient to, eseoes the collective effective dose equivalent commitment for each of the strategies during the period of time which equals the longest espected life-time of alternative solutions. As an example, assume Wt altamative A
- is expected to last for 100 years and alternative 2 fe 1000 years: also assume that tu both esses the sensequer. css at W and of N respective i
lifetimes are assumed to be identieel. The sumulative collective doses to be estimated are, fee option & N sum of ht portion caueM during the design Life of 100 - re.a that e.used be - 100.ad 1000 - re, for e, tion.
that portion caused over W 1000 year design life, i
To be able to sempere seet of protection ami detriment it is necessary t
te mes saeperetive units. This seena thet W sellestive effestive deee equivalent seemitamat should be empressed ta monetary values. Conversion of this sommitaaet to aseetery un!.te is v'ry arsweble. Newever, in the radiation protection f1 eld Og 8 10 - 10' is generally sensidered to be the rsesonable teet to avoid a tellective does equivelant of one men.sv. Se the j
4 5
monetary value would be US $ 10 -10 per men.sv (or US 8 100-1000 pg?
l men.ren).
In this case for a given level of protection, w, and the proportionality i
value e, any detriment T(w) would in renetary terne be represented as:
l i
Y(w) = m3,C 5
I l
1
l
- 103 -
The optimisation amans a minimization of the expre.;;sion 1(w) + estC Se in evolusting two levels of control A and 3, the supression I (w) - I (w)
, gg g
g "E. C I.Cg
~ E.C E
E A would be eve 1usted against W value e.
If i s e, then the decision to move free' control level A to, say a E.C more sapensive one la 5, enn be supported.
la eunnery, W fellevig steps may be used for W purposa of optimistas rodiet.se protectient (1) sempile all h eve 11eble senagement strategies Wt are tall 3 red to a perticular esse (2) ovelvete the seet of Laptementing seek strategy; (3) evoluete the sollective offestive deee equivalent commitment result;ng from each strategy:
(O detassimo b quettent of the differentist seet resulting from thenging from one stratasy to amether, and the ditforence of eatteettve dose that would remalt free t,he ehenge (3) eenpers the above quettent with the aseetorr welue per unit ef ee11esLive dose (e) oklah to essertable to h appropriate regulatory motherity if. La obenglas from one strategy to W neet, the value of the quettent is found te be lower them e, the nort strategy should be proferrrkdI (6) samfiru W t W selecte4 strategy will temply with h individual dose limit <.
This same approach any be applied to subeystems within the total system, provided W subersteme ere independent in the sense W t the protection in one of them dess met influence W sollective ef fective dose equivalent sesaltaent frua the others. Thus, in h case of independent subsystems.
0485y
- 104 -
optimization of protection can be obtained for the combined exposures from several installations at a given site can be obtained by optimizing separately the protection at each installation, provided the condition of independence applies.
purther discussion and illustration of the application of the dif ferential cost-knefit technique are available s.. the following sources:
Sesic safety standards. Annex IV, sections 200 and 204-216 provides a asore detailed mathematical development of the technique 113).
The WEA report on Long-Tern Radiological Aspects of the Manap,eaent of Wastes free Uranius Mining and Milling [89), in which the cost-benefit approach is applied on a case study beels to actuel or model situations i
in Australia. Canada and the United States of America. These case l
studies 111ustrate the problems encountered in the application of the
- approach, and some of the stay 11fying steps which can be taken to keep the scope of the analysis within practical and manageable 1 baits.
The WEA report on the Long-Tera Radiation protection Objectives for Radioactive Weste Disposal (13] provides further guldence on the estimation and use of tellective dose comualtaant. In particular, the i
report addresses the issue of trancatica of W integretten of i
se11ostive dose in time es a seems of addreestas N problems inherent in attempting to integrate to infinity, ceneidering N increasing
(
i uncertalaties in the fee future, periods of a few thousands of years are l
thought to be the y per limit for any realistis senaideretten in
==== tion with optimisatten.
10.2.2 htti-ettrikte snelytis This testatique may be woeful if the differential eset-benefit analysis I
does met leed to a 61 ear stoise of optione. Aloe by petuitting N l
t consideration of certata intangibles which are met treated La W cost-benefit enslysis, it represents en approach that any be useful and advantageous in its osan right. This technique has been described in references 90 to 92.
The following is an outreet from reference 15 which suunnerises the technique.
j I
j 04857 l
l t
_. _. ~ _.,. _..,. _.. - _
E
- 105 -
"The objective of this technique is to constmet a scoring system for alternatives, with the property that the alternative with the highest score is preferred. The first step is to st m eture the problem. It is necessary to identify all the factors which distinguish betwen the altamatives under tonsideration. However, only the most significant ones need be retained in the further steps wf the decision-making process. The outcome of stmeturing the ppblem is a hierarchy of attributas, in which some factors are tre9ted separately and seas are grouped together-(An example is shown in Figure 15). Radiological Lapact has a number of attributes which can be separately valued, but it side comparisons between alternatives to group them together.
This stmeturing exercise is veluable in itself in promoting openness in decision-making and in providing a framework for the assessment of alternatives.
The second stage is to st m eture the preferences for each factor in a' formel way. It is necessary to constmet a asesurement scals such that th'a preferences between different values of an attribute are represented by values on a et w scale. preferences are usually these of a group of individuals who have the roepensibility fee taking decisions in the public interest. When independent f acters are considered, autuel preference independence ensures that the weight given to one fetter does not very when the levels of other attributes ars changed. It is in addition necessary that a given increment in score for each attribute is equivalent me matter at what peint en the scale the increment arises; for emangle an increment from 40 to 50 points in the valuation of radiation protection preference maat seen the seen as an increment from 70 La 30 points. Once a asseursent procedure for each attribute is ateSed, it is meseseary te ensure that the weights applied to l
these volve fractions reflect the judgemer.ts en balances of preference between attributes. This will inevitably be subjective and approulaste, but use of unalti-attribute analysis can aske a velvable contribution by focussing discussion en real dif ferences between opinions and by identifying central issues. SLace it does not involve empression of all the facters in terms of 4 cossmen denominster (e.g. menetary value), insiti-attrikite analysis, in principle, is capable of dealing with all relevant factors."
0485y
I
'l l
l l
. 1_,
l l
l hbh h-
[d 1
. d.,
r l
l 1
l l
1 E*.
g en-e.
a, m.Y
~
l l
l l
~
r m
.w.
=
=
8.ne.e4 m
..e fit 0 4 C lt'
[ts']
M W af W One*1 > M af. ant ammeen.*w onn e by onnoser andym M i
D 9
- 106 -
In this section the application of radiological protection considerations that necessarily play an important role in the design, constnJction, operation and stabilisation of tailings bepoundment systems have been discussed. However, it should be noted that this discussion presents only a brief treatment of these considerations includes only an introduction to the principles and concepts of optimisation of radiological protection. It is anticipated that the application of these principles and concepts will undergo considerable development in the nort few years.
A simp 1Lited example of application of some of these principles is given in Appendix c.
Although this type of optimization has generally been used for radiological impacts, the process can and should be used for the hasardous non-radiological substances which are associated with uranius mill tallings.'
11.
Assessing W long-term safety of closecut options 11.1 Introduction Although it is relatively easy to design engineering schemes to be used for clooecut of an ispeundment facility and to amenitor W etfactiveness of the design over W short-tem, it is omtreenly difficult to estimate the long-terla layects. G6nerally the safety analysis of N long-torn bet.aviour is predicted usins mathematical andels la whlek the estual physical, thendM and bielegiset processes are replaced by anthematical approutsetions.
In this section, a brief everview of N general methodological approach to eefety assessment is first given to show W Individual steps in the analyses in content of their evere11 ebjectives. pellowing that, the modelling and analytteally techniques which are used are reviewed. The reader is referred to references such es 5, 93 and.. for further infetuation on W ee topics.
Appendix D gives an omaaple of W use of one type of mathematical coaguter code in the assesstant of closeeut options.
0485y l
107 _
11.2 overview of safety assessment 11.2.1 Methodological approach In the safety analysis, the weste, repository and site should be considered as en integrated system. The performance of the disposal erstem as a whole mast satisfy all.the regulatory or desired environmental pro'tection requirements 1931.
Safety analysis are carried out iteratively with the methodology, models and/or data being updated between iterations. Sees iterations are also made to obtain information about the uncertainty of results and means of cuecensatLng for this uncertainty.
The level of detall and complexity of any analysis depends to a 1erse.
estent on the vers in
- Leh the results are to be used. For example, e
relatively staple analyses any be suitable for secessing the f, sibility of l
disposal options in a generic menner, while more detailed enslyses are likely j
to be required for LLeensing putyoses. Newever, all safety analysei should be comprehensive in the sense that all the phenomena which any lead to release and transport of radionuclides into the envirorusent er Lafluence the rate of release er tronoport through the savironment need to be taken into account as the potential rediatten deoes are seleuteted. safety analyses smerefore have the fa11 ewing three hosis seepenents (931:
(1) identifisetten of the 7 ----.: which seu14 lead te e release of redleesselides or Lafluence the rates et ot.ich releases escur er infimense'the rete of transport of rodienuclides through the environment; (2) estimattoa of the probabilities of eesurrence of these phenomena end quantifisetion of their effects en the disposal systems (3) selculattee of the redtelogiset sensequences of rolesses (i.e. deses to individuals and populations and, if requirsd, eatiastin of subsequent health ef fects).
twec in relatively staple safety analyses, it is not sufficient to consider one particular release situation er to calculate release consequences veLag one set of perseters. Methodological approaches which neglect the prebebilities of escurrence of releases and the possible ranges of 0485y
los -
~
consequences are open to criticism, because it usually is possible to define worse or better cases, and thus to call into question the validity of using analysis results in softy assessments. Idestly, a safety analysis should produce estimates of the overall risks to humans from disposal and should l
therefore explicitly include estimates of the probabilities of occurrence of I
phenomena and probabilitty distributions of deses. Newever. in staple analyses hoe probabilistic aspects any receive less attention than in a full comprehensive analysis, for smaaple by exercising qualitative judgements as to I
the most probable release situations. Nevertheless, any such judgements should be clearly explained, and it should be recognised that subsoguent. acre detailed analyses will usualb/ need to treat the probabilistic components of I
risks quantitatively.
i The most frequently used methodological approach to safety analysis l
1 consists of carrying out scenario analysis (essentially items 1 and 2 above)*
end consequence analysis (item 3 above). pigure shows how these two 4
l components interact and how they fit into the safety assesseont and also shows l
the iterative -ture of safety analyses.nd a..ess.e,.ts.
i j
11 2.2 seenarie analysis I
scenarie analysis involves the identifiestion and quantitative definition of ;' ---
which eeu14 Laitiate N rolesse of radienuclides from a mill tailings repository and/or influense the rates at which releases and i
i treneport ecour. Quentitative definition eenstets of predicting the effects i
of W. imeludies the rates of relevant processes, and estimating the I
probabilities es esaurranse.
I these ;'---- _.c f all into three breed t stesortees (1) temen activities i
(e.g. sonstructies, farmins, det111ms for minec:1 resouroes): (2) natural i
processes and events (e.g. ereelse, fleedias, groundwater anvenant): and (3) weste and repository processes (seid generation, anchenical disturbance of i
l soil or rocks at the repository site).
l The approach to analysing scenarios involving phenseena La each of these l
categories requires the wee of teetmiques to determine which phenomena and coehinations of phencemus are most relevant and hence to previde a logiest j
framework for estimetion of the probebilities of escurrence of scenarios an-9
)
04857 I
Def6attiea of tastial state of spetoe A
I m t date y
- .se...e e,wi,..
no,ete.
e. 6, seIeeoe ies,3,ed ens eseels arW tremosert eete e#W U
seenaries seeme tig a
fee Geelga e-o Ceaoso,eaee see13eie (reteese, tramesert, eees selowlettoas)
J A
tastetten estes ecommentattee e
eefet, eseesseeat I
(Cassertsen of results e
sign esteetettitty er6teele) i 4
eeeeetee6 tits seelstea l'Me M /1 M hemar e/shry esmesseet e/o shadee greend dueemt syrwas.
lefeat eestrene Seemsele ensItoLe Cemeegseese seei,eae I
I
.....J...
I....L...S PTeam6416eele
' estesono6 46e estessestetLe Peeeeenntetse 8
sesamtesee n
' ioe nesen.
isoneio.e.
iesemie e.
I 4.... w s.. w' 0
8 IMIS OMe estes tefle {
evene stee emete Cetie l
,.... s
.....I testeesee shot she embetone to posseente e ees seer,==. to gi se stee 6e seetties to one eene sesensen.
'NM 8f M Coseeereteand3D M esee[ahfyemefydis.
W e
7,
,;7
- 109 -
for the quantitative definition of release and transport parameters. These techniques are essentially probabilistic: W main examples discussed here are fault trees, event trees, and Monte Carlo methods (section 11.3) 4 The results of scenario analyses coesast of the definition of scenarios (i.e. the release and transport parameters required for consequence analyses) and the estimates of the probabilities of occurrence of these scenarios as a 4
function of time. When scenario analyses are carried out for specific disposal systems and sites, it will usually be possible to eliminate some l
potentially relevant phenomena from detailed consideration, either because they have very low probabilities of occurrence or because their effects can be shown to be insignificant. The bases for eliminetton of phenomena sho.i: se i
l explicity stated and will fotia another type of result of scenario analyses.
11.2.3 consequence analysis After the impoundment f acility relsese scenarios have been defined. the consequences to teamens should be estlasted. Caeviation of consequences r, quires the development of a computational system which models the transport of radionuclides through the environment to humans and calculates radiological doses.
i The process called sensequence analysis censists of several steps. As a first step, a prediction of redienue1 Lee estesse rates free the repository should 1,e mode, felleued by estimet w of the redienuclide concentrations in the various sempartments of the envitesueent. Its second step consists of a prediction of transpect rates et released rodienuelides between various se g ertamats and men. The third step involves a prediction of radionuclide internetten with asa, resulting in saleviation of deses to individuals and to the population for each scenario identified during scenario analysis the 1
tellective dose commitment is slee estimated to indicate the total impact of the repository.
l i
I Cons % ance analysis involves fellowing the progress of the released radiofo6c11 des via different pathways (Figure 6).
Direct pathways include j
drinkins water, consumption of aquatic foods, bathing and swimming; indirect f
pathways include transport by plants and domestic animals (plants can accrue radionculides via irrigation or by transport through topsoil).
i 0485y I
- 110 -
To predict the rates of transport and of intake of radionuclides by humans, mathematical models are necessary. These phenomena can be described by a set of first-order differential equations which can resdily be derived and solved. The results obtained depend largely on the quality of input data and in thAs field some uncertainties exist.
The fier,t source of uncertainties is our limited knowledge of the human population, its habits and activities and various environmental factors in the future.
Ir. practice, little confidence is placed in predictions of this type and usually a base case is analysed with the assumption that present crnditions will continue. Variations of this base case can be explored later.
Another source of uncertainties lies in the variability of values which
{
are highly site-specific in nature, assigned to the various transfer rates A
between the components as well as between the individual f the pathways. Therefore, an extensive programme for development of appropriate pathway models should be carried out, the objective of this programme is to developundersthdingofallpotentialtranspcetprocessesinvolvedsothat this understanding may be extrapolated to other situations at other times, given an adequate description of all relevant physical, chemical and biological characteristics on another site.
Consequence analyses are primarily deterministic in nature; however, whan distribution functions of envirosamental transport parameters are considored, consequence analyses aise have probabilistic e m ponents. If submedels of portteular system sempenents are used, these smast be able to be readily Laterfaced to produce a model of the whole diepenal system 11.2.4 Evaluaties and application of results Acceptability criteria The basic requirement for all radiatin protection is that radiation exposure of men be kept at en acceptable level (section 10).
When applying the ICRP recommendations to the disposal of alli tailings, some special problems arise. One is to estimate future doses and their 04851
- 111 -
probability of occurrence; another is to figure the probability of initiating events in calculating collective doses for optimisation purposes. An important factor also is the time period over which the collective dose should be calculated. These and other esisted problems have en important bearing on the methods chosen for safety analyses, purther guidance can be found in Chapter 4 of the IAEA document on Criteria for Underground Disposal of Solid Radioattive Wastes L&C.
r N
Safety et.alyses should take into account all the phenomena iMich r.ay lead to a esisase of radionuclides from a repository or influence the este at which retsase occurs and should include quantitative predictions of the probabilities of occurrence and the radiological consequences of releases.
The weight to be placed on the varius types of analysis results (e.g.
probabilities, individuti doses, collective dose commitments) will very according to the acceptability criteria adopted, but this does not detract from the beric requirement to carry out comprehensive analyses. The criteria can be divided in different ways here they are divided into the two broad categories of does and risk criteria.
In the formulation of does criteria it is generally assunned that the guidance given by the Icap should be followed. This approach assumes that dose is en adequate asseure of the radiological impact of disposal, and it is Laplicit in the approach that the scenarios which are likely to be major contributors to radiological impact obsuid be detined. Tleas in conducting safety analyses when the results are ta be sempered with dose criteria, it is only necessary to define the most probably release scenarios and to calculate j
thste radiological eensequencea La terms of deees to individuals and collective does commitments, i
I In sentrast, if criteria are framed in terms of risks (where risk is t
defined as the probability of a deleterius effect and is esiculated by combining the probability of occurrence of a release and the probabilty that subsequent radiation doses will give rise to deleterious effects), it is necessary for the safety analysis to provide quantitative information about the probabilities and the radiological consequences of releases. Again i
individual and total risk must be evaluated. Both dose and risk criteria l
require the same type cf safety analysis, but in the former the probabilistic I
components tend to be a less important part of the analysis since these aspects are treated in a more qualitative way.
l
- 112 -
I 11.3 Modelling and analysis techniques l
To perfom safety analyses it is necessary to use a number of mathematical models and analysis techniques. For the purposes of this report, models and techniques will be divided into the following categories:
(1) probabillatic analysis
- fault tree
- event tree
- Monte Carlo methods (2)
Deterministic analysis Figure shows the components of safety analysis and the techniques which may be used at each stage.
=
It is important to recognise that probabilistic analysis and deteministic analysis are complementary techniques and that, as a general rule, both should be used in scenario end consequence analyses. Scenario analysis is necessarily probabilistic because the probabilities of occurrence of various events and processes met be taken into account. Newever, it aise
(
ham deterministic soeponents, since submedels met be used to predict the effects of various events and processes en the disposal system and hence to
]
provide the input data for release and transport calculattens. Consequence l
enalysis is primarily deteriministie because its asia objective is the
]
calculatise of rotes of redienue11de release and transport and subsequent doses to tassens, and the models used in these calculations are usually deterministis enes. Menever, when there are uncertainties in release and tremeyert parameters (ehich is usually the case) the distributten functions of thase peranaters will have to be used La the calculations te obtain ranges or distributions of deses. In this situation it is mesessary to use probabilistic (statistical) technique la sensequence analysis in addition to i
deteminietit enes. It een therefore be sensa that a comprehensive eaf aty l
analysis in which all the relevant phenomena are considered and the uncertainties in parameter values are explicitely included, requires both j
probabilistic and deteriministic techniques. For this reason the methodologies used in comprehensive analyses are of ten referred to as fully probabilistie.
l I
04557
1
- 113 -
11.3.1 ProbabL11stic analysis techniques probabilistic enalysis involves a set of statistLeal techniques for studying effects, parameters whose values are uncertain, events whose occurrences are'randes, and features which may or may not be present can be i
treated statistically. For example, the probabiltty or frequency of occurrence of an aeropiene crash at a specific repository site may 'a i
estimated. Various methods are availeble for considering how the I
j probabilistic variations in components of a system act together to cause reriatLon in the system as a whole. These include fault / event trge and Monte I
Carlo analyses.
l Fault / event tree analysis i
l The fault tree and event tree analysis techniques are the conventional methods for system reliability and probabilistic safety analysis. In this t
approach, system failure logic is graphleally displayed in tree-1Lhe i
structures. Coeyuter-aided methods are available for analysing these
[
structurse to determine both qualitative and quantitative aspects of system reliability and probabilistic safety performance. The precise application of feutt/ event tree analysis requires thorough knowledge of the system under study and esplicit laterrelattenships among the various components which comprise the systan.
Fault tree analysis is a deductive probabilistic technique by which the l
compensat f ailures leading to system failun can be logically deduced.
Applicaties of the technique yields sembLnations of basic events whose
(
oceurtense causes system f ailure events. Fault tree analyste starts arith careful definition of the failure events and systemattcally disgrens bachweeds to identify the events er combinations of events that could cause the f ailure event to occur. The logic is displayed graphically by Boolean techniques in a j
tree-like structure. The process stops when the analysts define events that are eLther not ammnable to further resolution or when it appears there is no need for additional definition.
l t
Event tree analysis is an inductive probabilistic technique which reversos the f ault approach by starting with the basic initiating events and 0685y
- 114 -
working forverd in time to display their logical propagation to systen failure events. Event trees diagrassaatier.117 illustrate the alternative outcomes or consequences of specified initiating events. The trees provide a pictorial representation of outcomes sad provide the basis for quantitative risk assessment of the variuos sequences s ich arise from the initiating event.
Isonte Carlo analysis:
The Monte Carlo tech ique is a stochastic one. It is often used for a mystem that cannot be described in a deterministic manner, either because it is too complex to be describ d in abstract anthematical form or because the mathematica). equations cannot be solved by analytical or simple runerical techniques. It is also used in sensitivity and variability studies den the distribution functions of parameters are known but een the system Le too complex to permit the use of analytical asthods. In safety analysis, the leonte Carlo tec k ique can be applied in both scenario and conseguence analyges.
When the leonte Carlo techique is used in scenario analysis, the starting point is a list of besic systes events and estimates of their probabilities of occurrence. The technique is then used by stopping through time, assuming occurrence cf these events according to their estimated probability distributions until a system failure occurs. Af ter the sisuistion is performed a large number of timme, a probablitty distribution of f ailures can be obtained. The Monte Carle technique is well suited to analysing discrete events quantiatively and because of its stochastic nature it is also able to analree complex interactions between these events. Its main disadvantage is the large number of simulations espected to be necessary to obtain adequate failure distributions fee law probability events. The technique een be generalised to consider the continuous pro eases occurring at underground repositories by updating the state of the system for their effect at each time step.
To use Monte Carle techniques in conseguence analysis, the procedure consists of running a deterministic model a large number of times with persanter values selected from the bown distributions or ranges. In this way the distribution of consequences can be determined. Again the technique has the disadvantage that a large numbst of sinalations are required. Its advantage is that it een be applied to complex systema for With the use of enalytical statistical techiques is not possible.
0685y
(
- 115 -
11.3.2 Deterministic analysis techniques Deterministic analysis techniques are the classical methods used in predictive mathe:.atical modelling of system behaviour. To use these techniques it is necessary to have sufficient understanding of the processes at work on and within a systess to be able to form:1ste the mathematical equations describing the principal processes. Detailed knowledge of each process is not necessarily required, since it will often be possible to develop a mathematical model which odequately predicts the behaviour of a
{
system or component from a general understading of the basic processes involved.
(For smaaple, rates of transport of radiersuclides thorush the environment can, for many purposes, be predicted by the use of comparaent models. The development of these models does not require detailed knowledge of each transport process; it only needs sufficient understanding and data to be able to predict rates of transfer of each radionuclide between compartments. ) Deterministic techniques are particularly useful in modelling the effects of continuous processes and can be applied to both steady-state conditions and dynamic conditions which change with time.
J e
l e
0g857 i
h t,
.g.
13.
References (1) INTERNATI0WAL ATOMIC WERCY ActWCY Safe Mana6ement of Wastes from the Mining and Milling of Uraniva and Therium Ores - Code of Practice and Guide to the Code. Safety series No. 85. IAEA Vienna (1987).
(2) IrttRNAT10WAL ATott!C WERCY ActWCY Management of Wastes from Uranium Mining and Milling. Proceedings of a rymposius Albuquerqueg10-14 May Jointly organ!. sed 'r the IAEA and WEA(CECD)jVienna (1982).
1942.
e (3) WUCLEAR WERCY ActNCY (OSCO). Long-tern Radiological Aspects of the Management of Wastes free Uranius Pining and Ellling. WEA/CICD, Paris (1984).
[4]
C. OnugnoLL: Can you provide the title of a sumary t oport which reviews the U.S. progress?
A. M e (5) RottRT305 g 6 et al., Canadia Uranium Mill Weste DiaPosal X
Technology. Propered for the Canadian Estional Uranium Tallings Prose a, (1987).
(6) INTERNAT105AL AT00lIC ENESCT ACDCY. Current Practices er.d Options for '
Confinessent of Uranium MLL1 Tallings. Technical Reports series No. 209 IAEA Vienna (1981).
(7) MotJets et al.. reference title, etc. required.
IS) INTERNATIONAL ATottIC ENERCY AC WCY. Radiation Protection Procedures.
Safety Series 50. 38. IARA. Vienna (1973).
[9] INTERNAT105AL C000tISS105 05 BAD 10 LOGICAL I g g et of Committee 11 on Permissible Dese for Internal Radiation.gm (1959).
l (10) Reference required.
l (11) UNITED STATES WUC:.RAR REGULATORY C00 BLISS 105. Final Generic Environmental l
Impact Statement se Urania Billing Industry. NUREC-0704 (1980) 10setilagten DC.
[12) EURBS-CE 4403 113) INTEMATICEAL AT00lIC EN33CY ACEBCY. Basic Safety Standards for Radiation Protecties: 1942 Editis.3. Safety Series Be. 9. IAEA. Vienna (1982).
[14) INTDMAT105AL C000tI58105 05 RADIOLOGICAL P90TECT105. Recommendations of the International Ceumission on Radiological Protection. Publication 26 Pergamon Press. London (1977).
[15) WUCLEAR ENESCY AC WCY OF THE OSCD. Long-Ters Radiation Protection Objectives for Radioactive Weste DirPosal. OECD/WEA. Paris (1984).
l l
1 04S51 l
l l
Oncventinna A Ravininna en Rik14navaphy p.
116 (4)
I am not exactly sure of what progress you are referring to. The best encyclopedic document on the operations of uranium mills, in all of their aspects, is HUREG-0706, which is reference 11 in the reference list. Another reference is the UNITED STATES ENVIRONMENTAI. PROTECTION AGENCY, Final Environmental Impact Statement for Standards for the Control of Byproduct Materials from Uranium Ore Processing (40 CFR 192). EPA 520/1-83-008-1&2.
(September 1983), Washington, D.C.
[12] UNITED STATES NUCLEAR REGULATORY COMMISSION, Summary of Waste Management. Programs at Uranium Recovery Facilities as They Relate to the 40 CFR Part 192 Standards, HUREG/CR-4403 (1985),
Washington, D. C.
p.117 (30] I don't have a copy of this, but usually a contractor report bears both the contract organization's identification code, as well as the NRC's designation. Also his name is spelled: G.
A. Eehmel, p.
118 (33] Refer to reference (12] listed above.
[41] U. S. Nuclear Regulatory Commission, Evaluation of Field-Tested Fugitive Dust Control Techniques for Uranium Mill Tailings Plies, NUREG/CR-4089, (January, 1985), Washington, D. C.
p.
119 (45] The title of RG 3.11 is: Design, Construction, and Inspection of Embankment Retention Systems for Uranium Mills.
(48] Reference 11 (NUREG-0706) describes the Union Carbide Uravan Mill, which uses an SX process to extract vanadium. Specifically, Volume III, Appendix T. Section T.2.2.
(50) RG 3.8 is titled: Preparation of Environmental Reports for Uranium Mills.
[53] The lead author's name is misspelled; it should be Searsno.
[54] USDOE. Remedial Action Plan for Stabilization of the Inactive Uranium Mill Tailings Site at Canonsburg, Pennsylvania. (October 198.
UMTRA-D05/AL-140.
(83] USNRC Regulatory Guide 3.64. Calculation of Radon Flux Attenuation by Earthen Uranium Mill Tailings Covers. To be issued. (Available as Draft Regulatory Guide Task WM 503-4)
1 r Ln Al no t d h m & ceA..vA-< " du "~ "" -a aM3 =% y
, PAJ M4 Aa, red.4p,' tw% ?NH 7&)
g - t'n rc D. C.
/ (16] INTERNATIONAL C000t!38105 05 BAD 10 LOGICAL PROTECTION. Radiation Protection PrinsiPlos for the DioPosal of Setid Radioactive Weste. Fab 11 cation 46 Portamon Press. Owford (1985).
[17) UNITED STAfts ENV!BCMEINTAL P90TICT10W ACENCY. It ndards for Remedial Acti me at Inactive Uranium Processing Sites. 40 grt '.92 part II (1983).
(181 Reference required Section 3.1.3.
lit) 013CRICH. O.E. et al., Benitoring reden around uranius mine and mill hd ettee with passive integrating detectors. Ibid reference 2.
(20) BUDY. C.R. et al., Outdoor Reden Monitoring at Connensburg. PA.
1980-1983 Empert IEJ6-32S4 Mout.d. tianisburt. Chie (1985).
[21) Eeefers reference requird section 5.1.4.
[22] UNITED STATES WUCLEAR RBCULAfotY C000tI5100. Final Generic Environmental l
lapact Statement en Uranium Milling Industry. WUU C-4704 (1980) Volume
!!! Appendix C.
123] 8785308 D.L; and BANDtt. T.J., EILDOS A Computer Program for i
Calculating Envirorumental Radiation Dooes from Uranius R cevery Operations. Report EUREC/CR-2011 (PWL-3767) prepared for the UsW3C by
- Petifit Berttopost Laboratory, tiehlend. M (1981) 124) WOODEUFF. 3.P., 31100MY, F.M., A wind erosten equation. Seil Sci. Sec.
Am.. Proc. 29 5 (1943) 123) UTAM M TER RESEABCM LABORATORY. Brosten Con 1 tharing Nighway g
Construction. Reports 1. 11 and III. Utah University. Logan, g4 500EP Project 16 3 for the Wetten41 Co-operative Misterey Research M7E' l Programme (1976).
[26) WETIER. W.C. et al., survey of mediological Distributions moeuttins free the Churek Boek. 3.E., prenim Bill Teilings Pond Den Failure. Report putaC/CE-2449 (PEL A122) Propered for the UsenC by Peelfis pertlerest Laborotary. Sieh1and. M (1991),
i 1271 Reference required Seatten 5.1.6 I
lit) anfarense required testion 5.1.6 l
129) W 1TED STAftB DEPARTMENT OF ENEBff. Radiattee Pethwe P
1 l
Bee 1th Impetta feen inestive pronium 3111 T.tiliass. ".
M-M ington. DC M D08-4JT 22, %
[ M 8' liN l
130)
- 6. A., Airter%e Particuleta Coneantrations and Fluzes at an Active
..-.ium Will Te111a6s Site. Sett411e Petific perthwest Laboratories.
gp Richland. M. Rep. FWL-SA-6972 and Report for the US suelear Regulatory Commission. WUBBC/CE-4250 (1970).
0485y 5 ettt C TAGE~ Ac tupU.'l 'T4.se esPest T3. of J u s 7 Do w $4 No!14Gt/N4
-~ ~~-
e HY
- W.
[31) SCHIACER, E.J., Analysis of radLation exposures on or near uranium mill
[
tailings piles, RadLat. Data Rep. 13 (1974) 411.
/'l32) UVITED NATIONS, Sources and Ef fects of Ioniting Radiation, Annex D.
Radioactive Contamination due to Wuclear Power ProductLon, United Nations Scientific Connittee on the Effects of Atomic Radiation 1977 seport to the General Assembly, with annexes, United Nations sales publication I.77. II 1, New York (1977).
I 133) Reference required Section S.3.1./
A (34) 308tRTS05 A.M., RAMB'tBC, S.A., LANCE, G., "Current uranium mill waste disposal concepts: a saltinational viewpoint". Uranius mill Tallings Disposal (proc. Symp. Fort Co11Las,1978), Colorado State University (1979).
/135) EcWORTMER, D.B., WELS0W, J.D., "Dralnage of earth-lined tallings impoundments", Uranium Mill tallings Management (Proc. Symp. Fort Collina, 1978), Colorado state Univeristy (1979).
/ [36) WARDWELL, R.E. et al, in-situ Dewatering TecMLques for Urenium Mill Tallings, Report WUREG/CR-3203 (ORNL/TM-8689) prepared for the USWRC by OtNL, Oak Ridge. TN (1933).
y (37) TAYLOR, M.J., ANTOWMARIA, F.E.,
amobilisation of radionuclides at uraniumtallingsdisposalsite[UteniumMillfailingsDisposal(Proc.
Symp. Fort Colllins, 1978), Colorado St. ate University (1979).
V (38) WILLIAKS, R.E., dentrol and prevention of seepage from uranium alli Y
weste disposal f acilitio[ management, stabL11sation and Environmental A I i
impact of Uranium Mill TaL11 ass (proc. Seminar Albuquerque,1978),
CBCD/WEA, Paris (1978).
/(39) CHIPIL, W.S., SeL1 Conditions that influence Wind Erosion USDA Agricultural Research Servise La so-operation with the Kansas Agricultural Emperleental Station. Tech. Bull. Se. 1185 (1958).
[40) INITED STAfta EUCLEAR REGULATORY CCIBEISSICS, Betulatory Guide 3.39, Methods for EatLastLas Redleastive and for.is Seurte Terms for Uranium millins Operations. Weakiaston, DC, (Dete t w ired).
/ [41) uvass/CR-ar46: Title, ets.. required.
/142) 18T1W000, p., Caleul de la quantita de redon dais par un tas radif&re, Internal Report CIA /5FT 180 (1979).
/ [43] ROBIESKI E.1., "Tailings disposal by the thickened discharge method for Laproved economy and envirorumental control". Proc. 2nd Int. Tailings Symp. Denver 1978.
06857 i
, - _ - _. -, _ _ - - _ _ _ _, _ - _. - - - - - _ _ _ ~ - - - - _
--.,__,--___.-,_..n.--,_-.--,-,
ooo
-pr-y 144] EITTAL. H.E., WOSCENSTERN 5.3..
Seepage Control in Tailings Dams. Can.
Geotech. J. 13 3 (1976) 277.
Vlas] UWITED 3TATES WUCLEAR RSCUufotY C00011SS105, tegulatory Guide 3.11. [7"/ y g,g,7 )
Washington, DC (1977).
/ 146) BAT 108AL LEAD CO. 15C.. The Centret of Radium and Thorium in the Urenius Willing Industry. Rep. WIN-112 (1960).
V (4 7) IT1080VITCH.1.J., RITCET. G.M.. Resevel of Radionuc1Wes from Process Streams - a Review. Canada Centre for Mineral and Energy TecMology.
Departasnt of Energy. Eines and Sesources. Ottawa Ontario, CANMET Rep.
79-21 (1979).
Y (48] Reference required Section 4.2.2 Y (49) CHIN, S., Limestone and Magnesia Precipitation of a Sulphuric Acid slu.se, nio Alge.1.tt atti.s t s.. Ontario, ne, 37 -05 (lun.
tro V 150) UNITED STAft$ WUCLEAR REGU M T0tY CCest138105 Regulatory Guide 3.8. [ y,7 Washington, DC (1978).
1
/151)CABADACETTREFORMINESALANDENESCTT3QG10 LOGY,PitSlopeBonus1. Chap.
9 - Weste Embankaants. Department of mergy. Mines and Resources. Ottawe.
[gif tario, CAMET top. 77-01 (1977).
~$
bet) SCAR 30. R., LINEMAN. J.J., Metent US Butlear tesMatory cosumission N
Licensing Regior Proces Uranium Bill Teilings Management (Proc. Cyup.
Fort Ce11Lns. 1978) Celerede State University (1979).
53 LS*1 Reference required section 7.2.1.3.
g t/tS4) Reference requi, red USDOS.1983 Cenonshu namedial iction Plan (section 7.2.2.1) fr-(ISS)WEL305.J.D.etal.EsthedelegiesfortvaluatingLerg-TorsStabilisat t
Designs of Uranium R$ 11 Taillage M----f
--.ts. Report FJRSG-CR-4620 (0ert/TE-10047) prepared ty Celerede state University for USWRC (1984).
g
' /ISS) Us1 Tap starts per&atutrf Or isrtasT. Tectmical sissanry of the UWTRA y*
Projoet Techselegy Devoleyment Program (It00-1944). Report i
I natrtA-DOS /AL-390125 (1985).
/ IS7) MAMER OF MINES, J0M&5WES9USG. Optimas Conditions for the Establishment of Togetaties en 31 Laos and Demo and sand Dumps, circular so.11/67 (1974).
I13e) CNAMBER Or MINES JONABWESDUtc. The Vesetotten of teaidue Dopoeits Against Water and Wtad Brecies. Randbook of thaidelines for Environmental Pretc,cties 2 (1979).
(59] wasAY. D.R., Togetation of Eine Weste Embenkaants in Canada. Department of Energy. Eines and Resources. Eines trench. Ottawe, Eir4s Branch Information circular IC *01 (1973).
048Sy up =
,#;~<@.M *'M-c
_. cs,g w me., w.<w ret %++d W =@t,%eut?m a&V=-%
M st4,, f NQeJ M 74 HLL4xcw.m.
.a.. o w>
L.,, m >
}M
. pe.
/(66)SITTtt.R.,FOURCADE.5..IETTwo00.P.,
amertues ou sujet de la i
stabilisation dee steriles de l'utenium per iglantation de ves6tation Eenagement, Stabilisation and Envirefusental Zapact of Uranium Mill Te111nss (Proc. Seminar Albuquerque,1978). OECD/ftA. Paris (1979).
/161) 00#0 VAN. R P., et al., Vegetative Stab l11setten of Mineral Weste Neaps.
Research Triangle Institute (RTF) Berth Caroline, under a Centract for the US Onvironmental Protection Agency. EFA-600/1-76 087 (1976).
d62)ReferencerequiredSection7.2.2.5(Cenede).
A43] EAYS, W.S., Cenettvetten of Linias for asservoirs. Tenks ers' Pollutten Centrol Feettities. J. Wiley a Sene. Bear Tork (1977).
sd 6el RAYS, W.G.,
ining systems for seepage sentrol in uranium mila tallings if holding ponds Uranium till Tellings Dieposal (Pree. Syup. Fort Collins.
,e 1978) Celerede Atate University (1979).
/ (65] Reference required Section 7.2.3.1 (Rays. Gelder a stoffen).
/ 166) Reference required b. tion 7.2.3.1 (seleer)
V'(671 Reference required Section 7.2.4 (Enight)
/(68) SLIGHT.0.5..STEFFEN.0.E.M.,"Seetoshniseofgoldminewestedispose1*.*
Current Geetechnical Practise in Eine Weste Dieposal. Geetech. ha. Div',
ASCE (1979).
/ (69) FRANEFORT. J.M.,
"A milling p-etees providing floribility to tailings disposet anthods". Ure, sus Bill Tt 511nss sensgement (Proc. Syup. Fort Colline 1978), Celt? ode state University
!!70)EDFFETT,D.,TheDisposalofSolidWestesandLiquidEffluentsfromthe j
E111 Lag of Uranina Orte Canada Centre for Eineral and Beergy Tottuselegy.
f i
Department of hergy. Rias and Resources. Ottawe. Ontario. CAMIET Beport (1976).
/[71) tamIER. M., unthod of Separation of Radine. Potent granted to j
Cennieseriet 4 1'amorgio Atomique. Paris Canadian Petant Be. 778.031.
j Petant Offles (1944).
/172) awltas. J.F. LTD. hvironmental Assosoment for the Propeeed tillet Lake premiun Einee Empension, vei. 4 (1978).
V 173) uBFFFTT, D., Charseterisetten and disposal of radioactive ef fluents from wrentum mining. CIM asil, 12 006 (1979) 152.
j V (741 Ref rence required Section 0.2 (fluidised bed).
)
v175) Uu1TED *TAfts trVIsoeurrTAL PacTtCTIos AcusCT anvironmental I
madioactivity surveillance maide. Office of Radiation Progreamse, neport oaf /stG-72-2 (1972).
1 j
0483y i
i
o j3l
. pr =
g/ 176) INTERNAT108AL ATott!C ENTICY ACDCY, Objectives and Design of h virorueental Monitoring Protresses for Radioactive Contaminants, Safetr Series 50. 41, IAEA, Vienna (1973).
/177) INTERNAT!OWAL ATott1C ENERCY ACDCT, Manuel en Radiological Saf ety in p
Uranium and Thorius Mines and Mills, Safety Series No. 43, IAEA, Vienna (1976).
]
VlF8) Reference required Section 8.3.2 (Analy. Methods & Tails SenFles?)
V lft) Beforence required Section 9.1 ( h ek) l v 180) Reference requited Section 9.4 (Ussa 1943)
P tut) 34ference required Section 9.5 (ralders 1984)
V 182] Reference required Sectie., 9.3 (Clif ten 1986)
!!83] United States suelear Regulatory Ceaunission, Budlear Regulatory Guide 3.44 (Title, etc. requirlt).
VISA) IFfERNAT105&L A..,;1C WDCT ACENCT, Prl)c1Plas for Establishing Limite for the Release of Radleettive Raterials into the Environment Safety Series Be. 45, IAEA, Vienna (1978).
V'l 85 ),
NUCLEAR ENERCY AC Banagement. Stabilisation and Envirormental Zaract of Uranium Mill TaL11 ass (Proc. Seminar Albuquerque, 1978), OtCD/tta, Paris (1978).
y 186) UWITED STATES ENVIROWNENTAL P90TECT105 AGENCT, Pinal Environmental Impact Statement for Benedial Action for Inactive Uranium Processing Sites (40 cft 192) Report WA 320/4-82-013-1 (Secties 3.1.1) Weshington, 1982, 187) INTERNAT105AL C0008188105 0F RAD 10tDCICAL P90TECT105, Cost.Senefit l
Analysis la the OPtimisatten of Radiation Protection. Publication 37, Persamen Press, omford (1983).
/ ISS) IFTEENATIOEAL ATOIIIC ENESCT AGENCY, Assigning e Value to Transboundary i
Radiation Exposure, Safety Series Be. 67, IAEA, Vienna (1983).
/159] WUC1. EAR ENESCT ASWCT Long-form Radieleginal Aspects of the
)(
Monasament ef Weetes fres Drentum Eining and utiling, cec 5/vsA, Peeie I
(1984).
gg ys e A 7'"
/ 90) MT905, 8.R.
TERD, S.M., A Seview of Some Fernal unthods for Dosiston jq 1
unking.
. CTEDIF CAR $/TR 28. Cash. Univ. Eng. Dort., UK (1981).
/191) MT905, 8.1., The tweluatise of Possible Butlear Weste Management Systems: Structuring the Issues, lateria Report 54. D04/EW83.049, UE Department of the Snvirorment London (1942).
0483y
l
. pc.
192) FOURCADE.
E., RIT N000 P., "Evolustion de different steearles de gestion d'un stockage de residus de trettement de mineral d'utentus", Manageunt of Wsates f ree Utentum Mining and Milling (Proc symy. Albuquerque, utt.,
1982), IAEA. Vienna (1982) 149.
j 1931
^1 wrondA T1ednL ATwevc Acatty AtmCh
$ M'TY 'fMALYllf ser A T ete s e Le ti d fe g, KA4/sMnd Wi+tfr 14Yef/184IIT ta S n LLo et CdeVMo.
swdn tot ses xs g <;,
(M fY) 169 von
[9Y) Tntrnewne Arvre awax(f AtacY, C4 i TYst s on
?st. nre now. csen vuo D rs NM, or /+ois se nur Wk m, lef a Y /kxist
//s 40 (19 t ?)
i G
i i
e i
'l l
l osesy l
i l
t i
l17
- W-Appendix A AssarVIAftD CMact-L!st OF Cous!DftAT1098 pot R&sACEMENT OF E!LL TAILIDGs i
a-1 vent. - ut and.he t.et i.pe.te Masardous meterial identiftsatten Ceneentrations and inventeries j
Bebilities L
Beegent attack on liners (seelelital and synthetic) pathwer analysis operational seneideratione
' Lens-tecu' sensiderations f
I A-2 Land use puture rosaurce ovellentlity Demography and projoeted Growth pett*cne sette-economise of the eroe Impoundeant ette pre 1Lforetion positive or adverse impette free propeeed use l
)
Limited future wee auffor estee provision i
Mistertest and ershoeelosital secos i
J A-3 Laag-term seasideretiene i
1 i
stabilisatten and rehabilisatten versus metatenance and monitoring j
meintemenee of sever j
Redes enhelaties l
l Samos shieldtas l
1 Tesetative protection and sentrol i
Surrowing antaat sentrol f
Financial guerentesc l
stant11:stion and rebettlitatten monitoring. surveillance and asintenance i
l
)
i cessy
lh
.w.
Govetweental and Laplementina organizatirns VisblLity Agency tronafers Record evettability and use FLle review and update Guardianship of property A-4 Legal requirements and Limitattens Areas esvered by law and regulettens Areas not severed by law and retulattens A-S Aesthetics Manneny with envirorument h Al Y o rWS
/
7 A-4 Other (??)
Atasepheric senditions e
0485y
) J D*
Appendit B APPROXNATING THE T.ADON SOURCE STRENGTH OF A TAILINGS IMPOUNDMENT
- 5 l. UNCOVERED Tall.INGS h
A sunphfied procedure is desenbed here for estunating the todon releau rate X
from n,i uncovered tailmes pile erven the quantity elthe 8"U meed and the estunated height of the tadtngs pde, (1) The total radioacttvity lowl 5 (l's curus) of the "*Ra preunt in the pine is j'
obtained by dividmg the quantity of estracted "U (in tonnes)by a factor 8
of 3.
(2) Esthriste the value for the effective reisist6on length H,n f the pile, in the o
@ p[
absence of data obtamed from a specific test made on the actual tailmes, ora may choose.
\\
(a) H,n m the ranse of 10 to 50 con if the matenalis wet and contams vnall gram staes and clay partkles (b) H,g m the rense of 30 te !!0 cm if the matenalis dry and mamty constituted of sand and relatnety coane particles.
(3) From the heir,ht of the pile H calculate HM,r, and from the curve versu H/H,n in Fis.3 determme the value of the alf confinement factor e.
(4) Estunate the value of the emanation factor r. In the abance of data obtamed from a specific test inade on taahass, one may answne a typical medrwn vahae h
of e = 0,1 (which is somewhat sonarvettve if the tashass are kept wet).
/
(5) The redon nisane rate is approsumated by Sefk(Cl s") uans the decay constaat 1 for 88Rn = 2.1 X 10" s", or by Sef, with 5 empromed in 8
t becquerels and the rados releans la atoms per econd, (6) The raden reinase rate may therefore he estumated by the forimula et X 2,1 X 10"(Cl s")= W3"Uer X 0.7 X 10 (Cl s")
3 M
where WS"U is the weight of '"U is the peseemed ore, of by et X 3.7 X 10'*(atosas. s")
i la carrying out th esortees it should be noted that the results are very 4
esemerg to the product of obh may very by more than an order of maerutudt I
/
The venne of this product le sentroted by the phyeneal propertse caraan ese and shape) and the amerensemeal propertme (esistense of clay pertasies for i Y g.
knesanes) of the taabnes to the mourure and the stockytte shape.
I g)
An uppee hamit for er is 10-8. Umes the vahas la the abow formula l
the reden rensees rete freen a stockpde of proceeend ore contamuis 10000 tonnes I
4 of U =tB be approtrastely 700 wC1.s", if the stockpda is umaowred and M
dry, data in the hterscate susseet values la the tones for et of 10*' to 4 x 10".
[
h p' Umns these vehmen, the redon release rete for taanes from 10 000 tonnes of v
'"U wiu be 70 to 280 wCl.s". If the mowtun content is kept lush enouah (i.e. nearty estuteted), et cza be la the order of 10 s, thus reduces the redon j
roleser to about 7 wCl. *8 The mean atmosphenc transfer oosfiloent is usually lees than 10" s.m's. Therefore, with the upper bruit fes er at 10", there is a]=sys less than 1000 PC4 m*8 in the se above a taAng pde if the uraasura content of the processed ore is less than 15 000 tonnes,
o e
/2. C- /1 I
(7) The rede flut is grvert either by kS' e H r (Ci m*8 s)
with 5' the radioactreity level (in cunes)of the 88'F.a premnt in one cubic metre of the pile, or by S' e H r (atoms m*8.s-8) with 5' ben the radmoettvity level (in becquerele) of t'w "*Ra present m one cub 4 metre of the pile.
Iri teost practical cams, eH = H,,r. With an ore contamins 2 r% U,0..
a 14&nst, specific gravity equal to 2, and f equal to 0.I, the redon fka from the stove formula will be I stom m*8 s*8 (28 pCl m*8 s'8)for He = 0.1 m 10 stoms m*8 s*8 (280 pCi m*8 s)for He = 1 m i
- 52. COVERED TAILING 5 la some countnes it has been recevamended to lunit me redom fka to 2 r<! m*8 s*'. Confinement facton la our esemple of I/14 and I/140 an reqwred for H.tr qual to 0.1 and I m, respectively, with the fkaee beina e
28 pCl m*8 s*' and 200 pCl m*8 s, respectmty. From shet6en 5J.2.3, S
I \\
e cover with a Nianeten lensth He = 0.1 m and a thkdiness of 0J m em reduce the Na to 1.$ PCl m*8.s*8 fee the first ease. A sever with a relanatma hasth s
He = 0.5 m and a 26chmees of $ a =4 redues the fka to 2 70 m*8 s for
.e.- __ es. et,.e es.e, a fe.se, of -
.[
saportance for assuolhne the redoc Ns.
I
- i..e e.e.f a e e, me.e of es.a.ve won of v-ierna m and taaskmeanne (h ), the total conAnsaneet festee le the product of the scannement i
fueerofeukinyee G am m esp -
1 3
As % eassaple, the reden the frose uneovered ladings la ese ease =ee 209 pC1/m i, 1
A "at eenemas of 61 as ciny, lll en esorturdes and 30 un topeed reduend i
ades fka to 2 pC1/m8 s.
l l
r I
i 3
i
- 126 -
5-3 Altemative method for uncovered end covered tailings (b.s. A.)
The techniques discussed in Appendices b l and 5-2 are still applicablet however, efforts in the U.S.A. in the areas of modelling (RAIOCM.
RADOW tedes ). laboratory measursewnt, field testing and field verification have it.e4 to a slightly different approach.
0/ bd 6-3.1 surf ace reden flus free uncovered (sang-tailings d
The general wquetion for W retsase of redon from a uranius tailings pile ist t
cz the &
l' Jg = 10 E a tg g g /( AD )i jne.(N tanh ( /(k/D f)
L a
g g
r where J is the redon flus from the surface of the tailings pile g
2 f 7,,, t, (PCL/s
,,)
y, 10" converts the flus unit free ca-2 g,,-2 E is W concentration of ta in the tailing solids (pci/s) c.ed}/g.
g p is W density of the tailings pile (t/cm )
g E is the reden emanation teefficient of the tailings j
g (dimenstenless)2 1 is the decay constant for 222,,,,, g,,q,,g g,,,g,g,-6,-1 D is W redon diffusion coeffisten'. of N tailings (ca.s*I)3 g
k is the thithness ur depth of the tailings layer being tensidered g
@/D8 Ap y
The fetter tanh( u(V9 )) se w atos for the bare tailings flus when i
g the tellings layer is relatively thia (less thee 0.2 m).
Otherwise, the l
faster is very aloes te unity and can be disregarded.
At present, the UsstC vees the RADos sede which is a seeputer progree l
to seeyuts reden flusee. mecoesary thithneados ot' covers and other estia.stes
- m m{V{p[ p i
\\
~
l wt,f 1 Usvac.UURSO/CR-3333, 1984 end US A tagulatory Guide 3.64 1988
[4 i
l 2 _s t s equivalent te e in appendiu S-1 3 Dg rectors in the paresity of the specific layer or'tattings being
'/ N i
considered. It is otherwise represented in the literature as D/p where D %d? l h,
is the ef fective bulk dif fusion coef fittent end y is N porosity of the h
1 tallings.
c(4
}4 WVtsC/CR-3333. "Radon Attenuation Handbook for Cranius Mill Tallings Cover Oesign." USNRC. April, 1934.
l 0803y 71%), ($ l Q.g* d4y l
- 121 -
t.
i 4
which are useful for decisions relating to recismation planning, m gAtec+t code is the parents
' ') code to RADOW.7:2' _____. a
^
" 't;:.C....ati-apM These codes af ford the user the flexib13 tty of assuming different radium concentration profiles in the tailings, as well as redon contributions from the soil cover materials themselves. The above equation for bare tallings radon flux asevnes homogeneity in the parameter values throughout the thickness of tellings.
This is a reasonable aseveption for a number of reasons:
reguaistions in the U.S. A. rufer to everage surf ace flux on an annual average bestal when tailings are prepared for cover placement, there is a degree of mixing and recontouring:
when tailings are relocated en even greater mixing occurs to provide for a more unifora mixture of tailings.
5-3.2 An approwlaste method for computing the redon flux from a tailings yile having a unifors soil esvor In W case that a single notorial is being used as a reden tallings cover, en apprestantion is evetlable for eettantes of the dealan release of redon gas from the statt1Lsed pile. Other layers such as rock cover, bedding layers, geof abric f Liters and revegetation een directly and indirectly (by enhancing meisture retention) igreve W reden sitenuetten capability of the cover systes as a whole. loowever, estimates of W necessary parameters for use in the predictive model are genere117 not known er at least ditficult to address quantitatively. These eWr features of a stant11 stion systen do contributo to the berrier's effectiveness and as such provide at least a level of sentidence la the protective mandre of W design.
The a;9roximetLos consists of estimatigneasuring er calculating a series of parameters reisting to epocific senditions representing etWr the tailings er the cover.
3-3.2.1 The bare and resultant fluxes In section 3 3.1 en equation is provided to calculate the bare redon flux. J. Cortsspondingly, the ultlante interest lies with J, the g
0485y
0 5g
- 123 -
5*
jj reeuttant f tur from the covered is,ounement, reequently, the usse hegins with c k s.
a bare flux and a required resultant flux criterion:
e.g., 20 PCi/a -set oem 2
73$
(0.74 sq/a -set).
In that case, the thickness of the soil cover, k. is aa e
tM J the desired estimate.
E jg 5-3.2.1 The reden diffusion coefficient A2~ n. TyM These parameters D for teilings and D, for the sever are the most g g?! T[me.n.ttive in d.termining mmmence of_themer erstasured s These een de g8 g l there are tecimiques for settmating the reden diffusion coefficient for either I ,u [ } g tailings er soil covers (Bef. USWBC get. Chalde 3.44). The correlation {g function ist 7eo aow 2 3 Rg3 D e 0.07 eny l-4(n-sm + a )) 0$S aoj S* where aistheasisturesaturationfraction,whichisthevolvestricfractfon of saturstion pere space (dimensionless)) g g n is W arterial peroetty (dimension 1oss): C' D is W reden ditfusion seef(Leient se defined in section 5-3.1 o ' 2 e1; (en.s 1) 8"S s33-Likewise, estimates fee a and a are avallable: T5I 335jl loi w a. %.h
- a 9, a h '.
p,La the asse denalty of water (3 em'I) j {, where }uI W is the lent tem everage asisture content (dry-weight percent) v 8,8 0 m is m material peroetty l ' > - ~ 'a $ $ p is the material density (3 en 3) 2:* ?2I 1 n. ue a *w l ?? where p is the dry bulk asse density of the materisi (3 em 3) ~ a,is defined as above i G is the specific gravity of the material (dtnensionisse) g 1 A 04857
o e - 129 - In the U.S. A. the regulatory authorities require a "reasonable assurance" t?.et the redon flux does not exceed 20 PCL/a -sec (0.74 94/a -sec) on en annual and surf ace area average. Since the diffusion coefficientisthemostsensitivepersakerforthecoversystemandsince moisture is the most significant factor in detersining the diffusion coef fittent, the Ost1tc builds the safety f actor into the choice of the value for a. 5-3.2.(
- t. ens-tors everage moistures An sayirical relationship for a is used by the Usuc's a = 0.026 + 0.005: + 0,0150r n
where a is the percent content of clay in the cover material is the percent of ortenis matter in the sever material -ene n is the cover material porosity
- t.LiW.
P This relationship relates the saturation fraction to the 134 erd soll moisture tension usually essociated with the witting point of plants in soil, i.e., that p*Lat where a reet system sannet drew any further moisture f rom the soil. Use of the wilting point approach is acceptable to the U.S. A. regulatory authority with regard to settsfying the "reasonable soeurence" j triterien stipulated by present standards. l t-3.2.3 Se es flus estimate De apprestante solutten to the steady-state dif ferential equation for the en fl det. in. tie..an h..ritta. e.i A y /)( s C d o 2Jeay(-th g J* t. eggsenh(b6' ' ' * - "F4'*"*E" **"**="2 5 Us n c Regulatory Guide 3.6a L} DC 0,,,,
- 130 - J, Jg have stready been defined e /)(g)t is two twickness of two ta111nss layer (ta) g A y, is ts. sat.iness of two.ever 1.rer ( > v, and e and b are defined as fellows: a is two interface tenstant between layers for seta asterial and La eseyuted ass 2 g = n D li - (1 - k) o )2 o gg g l .. n,n - n.,>.,i> Ik it, two equilibrium distributten seefficient for reden in water and air pCi.ca' water per PCL.ca' air and is settaated as k = 0.26) l The remaining vet tables above n.D,a have been previously defined. I g the relaustLon toe.:th dissveevd in Seetten S.3.2.2 Le represented asi ca( ; non l Qt 6 5T y two,.re t.r i to sus t. oree rela - tion 1 stai Y* sbG ) c h f ' L\\ 1 ca tww p cd-f-P Par ta(11 ass and the sever l l b,. /tu op and t,. /(vog' g g.u u 'k ? / l ] Under sectata sendittenal e.g when k >0.2 a and if it can be g .soumed a n,310, then J, een further be approntasted by: g osesy i l
e e - 131 - 2J esp (-b,y J t e 1+((a/aj g The above equat on en be used when a thich soli cover is being plac for asi ar a ~ < (g.7i reasons a*% reden attenuation. j 5-3.3 Einimum necessary thickness If a specified resultant fluw criterion has been stipulated by the competent authority or by other conditions, the equatten in Section 5-3.2.3 can be transformed to solve for the necessary thickness of soil, 1,. he. J,/J (1 - / Q tenh (b /g)) g g g . _ in 1 - 11. (J,/J ) (1 + /(a /s,ltenh(b f))(1 - / tanh(b/g)i)/2 a g g g A 4 s/ v whorstheaboveparesetershavebeenpreviouslydefined,andthegalueof J,, the performance criterion, has been stipulated. In this case, if the tailings layer is adequately thick, i.e. kg > 0.2 a and the specified flux limit is ten percent er less of the be flus, i.e. J /J, a 10 then 1, aan be appreuimated by f 2J /J, 4 g 1, = /[3A,h.In L 1. <(e,,. y. i i l 3-3.4 moittple layers of tailinse and severs l l l The saee of dieerste smaltiple layers of tallinse and seil sever meterials beeomme very semples when seing the opprestante techniques in the l AmothedelegyisprovidedinReference[.fN pretteus e ettene of appendix s-3. ] (USstC Betulatory Guide 3.64). I hea e multiple layer situation is appropriate, the user is advised to use e computer modeling approach such as the RAD 05 or RAICCet codes (Ref. U3ftc.WV18C/CR-3533, 1984 and U5WRC Re8ulatory Guide 3.64, 1988) 1 048Sy
I III. .y. Appendix C l EEAMPLE OF A 31NFLIFIED rsOGl!QUE FOR ASSES $liiC SITE AND a AILF DtEWT SYSTEMS A staylified technique for a prel) v: ,sessment of sites and systems for mill tellings managesent le s.. ere. This technique is effered to illustrate seen tonalderstlens that should be taken inte secount in eclecting the preferred alternatLve from a number of potentle11y outtable l management techniques and sites. This methodology is not intended to replace that required for optLatsatlen of altamatives for radietten protection as i discussed in SectLon 10. t (1) Identtry the possible alternative sites and design options. Sees of these eres (a) Siteet voller dans ring dykes, mined-out pits, spostal pits and L underground workings: (b) Liner optiones geolof tsal liners, slay liners and systhette liners; (c) management systen saturated, wet. seal-dry and dry (d) Dietharget sore discharge and treated dischatte. i i Select the senttaations that any be approprLite. These are the altamatives to be eyelvated. l (2) Identify eeurses of polluttes that mast be sentrolled, same examples ares (a) pollution due te surface tremeyerts spl11ege euring senvoyance. sttwsture1 f aLture ef embankmants, evertopping e! embankmenta. ereeion f ese te sensentrated ester flow, and sheet orecies: { (b) pollutist ese to oeepage sad teething (as4 m tret mothenises): seepage through liner, mistettee period (retardation of pollutants), f seit chemistry (retardation of pollutants) and potentist for teaching. (c) Air tellutices reden emanation and wind oreston. t (d) uteuse. j i 04SSy
o 01] .W. (3) For sesh option en siting and design being seneidered, evoluste and seeign a reting to oesh of the pettutten seursos. An arbitrary rating schese for esemple purposes any bei Prebebility Bli Very slight 811thL underate Minh Alaset sortsintF Reting (E ) 0 1 2 3 4 S g l (4) testue the severity of mash of the pollutten escurronese identified above. An orbitrary reting schase is os follows: severity miner moderate major noting (R ) 1 2 3 (5) Breluate the risk of pollutten from eesh eeurte (a)) soeumed here to be the product of probetility tLaos severity Bj=3 2 xR (6) m the pollution risk from eeth alteemetive to obtain en evere11 risk reting (I a ). It desirstle, rettage for dif ferent seursos sound bs weighted to rettest lesel espeoems. Alteemetives with the *:ssor risk reting would be seneidered the more favourette. memover la some seees the welshting of the senswee and probabilittee mer very from the lineer functLon stven here. l l C.1 ERAurta m The results of an asesyte eseosoment are given la Tobles C1 to C111. For the saae thesen, and W festers weed it is seen La Totte C131 that the La. pit eelutione have the lemmet risk retLas euttag N operating life of W mine. However, nessues of lens.tets surfose ereelon by water and a geseter potentist for groundweter teething, their post-operating retings (Leyeste) are Breeter then these for the ring dyte. Ostly
13V .g. The above evaluation dose not instude a seet-benefit onetysle. In thle ease, test estLastes would be devoleted and tensidered in concert with the present eve 19etion to arrive et deciatens resording eftions. It enould slee be pointed out that nattenal practlees any Lnvolve tensideretten of either peselbte tensequences or probabilittee for tellution without sonaldering the other. In these secos, the consideretiene preJented here eeu14 he modif104 estee4Lagly. l l e i l* ] i j l l J i I l e 9 04657 f ) ) I i
J f3 7 *E TABLE Cl. EXAMPLE OF $D4PLIFIED POLLUTION RISK ASSESSMENT FOR TAILINGS IMPOUNDMENT folkston due to msungt and crosson h 8messehwue tweentes oseristems et tsomee sw se see, meet A ense eme*WF esse tatse enteatmoets seteetod enser fnew eemse 3*)e4*$*6 5 OL P1 OL PA OL P1 OL P1 01 P1 OL PJ A. Yacar eso, see 1 (d'ee esam.poer eseheasmi (3 x 2) (3x3) O x H (4 KH (2x 3) ($ x H (t x I) (I x I) haar 6 0 6 3 12 0 6 18 8 1 4.3 3.4 (a) *n m. gesares how 0 x 1) (2 a 3) (1 x 3) (4 x H (2 x 3) (I I 31 6 0 6 3 Il e 6 18 I I 4.2 la
- e. ame N:. een 3 (0 8sawv seed eseasemmi 0ED U M 2) HEH HxH nam (t x )
1.4 1e hun 4 0 .I I e e 3 6 I .I C Ines, ase 3 (0 Wes weems. sreehren haar OMH (e X 3) (1 X H OXH U x () (3 X H 6 e e e e e a e 1 6 14 3e (3 s H (4 x 0 0sH (s) W srasn'. pene pselme.sai (3 a O ,e 0 e e e 1 6 1.2 3A herr S e e posens Awese=e M0 Wes dessel si tedhues a e veery eam hoeseesmant wat a sebesselr peer earwel eMesmal beer and a Isos ansehneet en Mel Wee empenni et seemes e a weitet eam empeneemmes a nesse west a sroeses hast se esseel sepass. Stil Rae evneee seed smonsennel how a tonesse tant vos eemmed Isee end and ever hem esser aswes. G0 le,s esposeemeat ce latade etch alspus resimused erf ass. Peer smetuesnel hast requeres asassemi sroestas heer enroe everstase. Qa) Inge espousesses a nesse W sua dry easpeaf messed et erschstes hear. Ear OL = operesas we et empensement. Pie Poseshamasse. e TABLE Cil. EX AMPLE OF SIMPLIFIED POLLUTION RISK ASSESSMENT FOR TAILINOS DdPOUNDMENT Pulkinon due to utper and leeckung f Poe Omspegs ereag Idagramme parend 8ed ehmestry PeasaJami for Aueress rus amar (pseeressame) (reenremessa) beshoe 8*I*10*11 4 OL PA OL PA 01 P1 OL F1 OL P1 A. Ysort ens. see I (O 'et #8mei poor gnehmani UXH U X 3) U N 3) Q X 3) UMH ($ N H () X H ($ N 31 haar il la is 13 18 18 9 18 83.3 11.0 00 Tes geme. speesus hmur Q X f) UXH UXH UMM (5 t l} ($ x H OXH ($ x H 4 H 8 18 f If 3 Il 4J li e S. Sme fyte, uns 1
- 0) 8mmeern seed gemenciami tsar (3 X 1) OXH OED 0MH HMH (I X H (2 x 1) (3 x H 4
3 6 8 8 8 4 4 de 3J C lege ame 3 (0 Des game, yeases amar 0XH U R 3) (I X 2) H X 3) UMH H X 3) (3 x H u x 2) h 18 3 3 3 3 4 le 3.3 ?.4 G3 Dr7 womm'i peer samengeams (4 x 3) O M 1) HIH H E 1) UMH (t I 8) (3 I 3) ($ x 1) tue 8 8 I I i 8 4 le 40 4J Sneen AJusasesse AU) Tee eseeans of teemos as e seest den empensammes ces a rehessed, peer aerwed enehemed tuer see a hres usansaree um Ma) Wee esposed to tatues e 4 veest es.s esposeement a stese sich a sreshoes amor es seesel empeos, Dio B.me tyne se ased smesmemmf huur e imemasse test wet eserved Irem eind and ever free vesar esere. On to,e unwomesses se hame enn es,ess e-* eevese. Peer pon ed mne noe oe enemment sroests aan e==e esereet Os) le,e anwesement as enose tot ese dry emessel usemed et gestsus tour. Ear 0L = Oversans we et se,senenses PA = Pen-
v - Is6 - TABLE Cill. EXAMPLE OF SIMPLIFIED POLLUTION RISK ASSESSMENT FOR TAILINGS NPOUNDMENT Air pollution and totalruk b REee emekstase Wedereses Aserage r6sa Sum of egrep nate 13 f4 (tereJ rus) 2 ?
- 12 + 15 01 PA 01 PA 01 PJ.
01 PL A. Yeary dem, se 1 (3 X 3) (2 X 1) (2 X 2)
- 0) We gems, peer seenismal Inner (1 X 2) 2 is 2
4 2.0 9.3 ?.2 94 (d)We emme, erstastu naar (1 X 2) (3X 3) (3 x l} (3 X 3) 2 l5 2 4 2.0 9.5 4J 94
- 8. Aas tyte. eu 2
- 0) 8sas4ry. amed suoisesel laser (4X2)
(3
- 3)
(4 X 1) (4 X 2) 8 9 4 8 6.0 8.3 41 4J C. Intet, ese 3
- 0) Wee srums, esthets imov (1X2)
(4 X 3) (2XI) (2 X 2) 2 Il 2 4 2.0 8.0 , 2J 6J (h)' Dry erema'; peer smelmosel liner ($X2) (4 X 3) (2 X 1) (2X2) le il 1 4 ke 8A 3.1 5.1 leoems Ahernaue* MI) Wet depoemd of tuhmes e e eeury den ammad-w wat a toisowPy pose senersi seeleasellamer ud a large cetrament eru Ma) Wet dasposal of tanhase a e einer dan unpenedemet a 6teve wun a estastas leer le seeW espees.
- 30) Rang dybe es seed godesel beer a leesues festy we8 sheltered free med and ever free esw seurus.
Ql) lesse sepowedssees se hande ett esoping reslammed erface. Peer eseemeneel haar requer,e ed&usaal syst4sts lumer dunas opetsboa. Q4) legat toposadmeet as above but wat dry despeani esased of prettets hast. Eay 01 = 09ersome are W amposammena. PA = Possmatuisatme. l \\ l I i
~ I31 - pr - Appendix D AN EXAMPLE OF THE USE OF MATHEMATICAL MODELS TO ASSESS THE LONC-TERM SAFETY OF MILL TAILINCS CLost0UT OPTICMS bl Introduction A peobabilistic predictive computer code called UTAF (The Uranium Tailings Assessment programme) has been developed (5) and applied to two mill tailings sites as a functional test. One site (Laenor) was selected because it was representative of the low grade pyritic ore of eastem Canada. The c.ther site (Ratbit Lake) contained a tigher sesJe ore with a more complex minerology but little pyrite. This ennex briefly describes the UTAP code and shows the type of input data, exposure pathways and output data from the assessment of the Lacnor site. Assessment runs took about one day on a desk top computer or considerably less time on a main frame computer. D-2 The UTAp programme The assessment of the closecut options for the Lacnor and Rabbit Lake sites were carried out with a computer programme called UTAP which links together a set of sub-models which are used to represent the disposal facility. The sub-models represent a particular aspect of the whole system for example the t4L11 ass, impoundment facility, groundwater transport and terrestrial environment. Each sub-model may contain different compartments. For example the terrestrial environment could be divided into soil, pisnts and animals. i The UTAp code linking prostsame was made as generic as possible. The user saa modify or replace say of the sub-models to model site specific featuras. The sede saa accept data is any of ten probability distributions (constant, notieel, log-normal, etc.). The values can be changed easily to determine the effect of change one parameter in the design, for example the design of the cop. 0485y i
Ist -p. D-3 Input data collection and preparation The input data required to model a specific site include: l 1 the exposure pathnys expected at the site. Figure D-1 shows the exposure pathways used in analysing the Lacnor mill tailings site details of the tite and the design of the various components of the impoundment facility (Section 7) details on the surrounding geological and hydrogeological environment especially h soil and rock adjacent to the facility local and regional citaatic conditions the occurrence of catastrophic environmental changes such as earthquakes, glaciation local, regional and global population distributions biological / bacterial assessment of N site essumed or measured liquid offluent flow rates chemical and physical characteristics of the tailings analysis of pors water, surface water and runoff radionuclides and non-radioactive pollutants in N tailings. The 226 radionuclides normally instuded are aa, natural uranius, natural O thorium and pb. Other espected ionic species such as sulphates, calcium and iron should aise be included red.m exhalation rates a social study to define h sharacteristics oi likely receptors a quality assurance plan designed to define the data limits. l These data wi11 be sabred from a variety of sourses such as l laboratory analyses, field measurements, governoont records and literature. The type and depth of data selected will depend on whether N assassment relates to a new repository / site system er W sleseeut of an established l repository. la the first sese, the assessment is being done to select that combination of design / siting features which will give an optimised facility to meet h established criteria. In the latter case the assessment is being done to select the best remedial actions to closecut the facility or to check if the established closecut procedure is safe. 0485y 7 I
1 ~ IM - W. D-4 Assessment of the Lacnor repository
- 13. John please provide brief history of Lacnor site.)
The first step in the assessment was the selection of the exposure pathways relevent to the Lacnor site (Figure D-1) and the modification of the sub-models of the UTAP code to represent W site, repository and the surrounding teerestrial environment. The site specific input data was collected and introduced into the data base in the proper format. Trial runs were then nede with N different design options so that W options which were most effective in minit: sing the envirortentsi impact could be selected. The final selection of the repository design would also include other f actors such as cost, regulations, eriteria and socist f actors (5). Four options were esamined at Laener e) [R. John to provide) b) c) d) The results of the coguter runs are in the form of a histograms of dose to a receptar at one point la time (Fi4ure D-2), a plot of the geometric noen of a number of these histotrees, showing the change of dose with time (Figure D-3). The output saa be presented La other ways so that greater inforwetion is available for the decision making process. For example Figure D-4 shows the contribution ti eeee by seek pathway. Clearly at the Canadian Lacner site swrfees water pathways ps&!=mte. Redon contributes an order of mesmitande less them the pathways and is less of sensern than in the more arid U.8. asaias stees. Figure D-3 shows the sentributies by radionue1 Lee. It shows that after en initial dominance by leed-210 the meet important radionuclide is es41ue. The model can be used to predict other parameters.ap Figure D-6 sPm the pH values at three loestions, at various distances, from the tailings 0485y
v ly0 .y-site. The pH depression is caused by the bacteria 11y-mediated oxidation of pyritic minerals. Figure D 6 represents the recults for one choice of disposal technology; in this case to take no remedial action. Figure D-7 shows the output for a trial with the smee input parameters but a dif ferent technology option; the addition of 2.0 m of depreitized tailings plus a calcite buffer. This has improved the pH values to a more acceptable level. The run could be repeated with thicker additions of cover material if desired. Figure D-B shows the output for all five technology options considered for the Lacnor site. It should be noted that although option 3 reduced the acidic drainage considerably (Figures D-6 and D-7) it made only a small change in the total dose. These results show that mathematical uodels can be used to predict the future impacts of a closed-out tailings site. These models can also be used to evaluate the effectiveness of the disposal tec hology options available. probe 1Latic models in particular can give a wealth of inforination that can be ' used by all partied in asking the decisions needed. D-5 18odel validation The results presented above are useful providing that the code is giving the essi answJr. However establishing how well the model will predict the correct answer (in model relidation) is very difficult. In UTAp's case this was partially achieved in three m ys. First, data from the original alli records at Lacnor were used to predict current values. The predicted values compared favourably with the measured values taken during the field programme. This represents a test for prediction over 30 years (compared to the 200 to 1000 years used in tailings forecests). Second, the UTAp model was part of en intercomparison of probaballstic models involving a besic theoretical exercise. Again the code was in favourable agreement with the other codes. The final validation effort was to investigate the ingset of a young uranium surficial deposit on the surroundings. The age of this deposit has been estimated geologically to be 10-15000 years. While the data collected cannot be compared directly with the results of the model runs, the fact that the deposit had had very little effect is consistent with the general conclusions of the two trials. None of these efforts validates the 0485y
~ l /41 ~V model in a real sense but it does sive a level of cofidence that it is not substenitally in error. D-6 IIonitoring After close out, the tailings site will have to be monitored to show that the chosen Wehnology is effective. The parameters to be twitored are chosen to indicate the status of the key components (for example radium-226 in surface water run-off would be one obvious choice for the Lacnor site). The monitoring prograses should last for several years (from say 5 to 15 years) so that conf! nce in the system is maintained. With time, it will be possible to reduce the amounc. and frequency of monitoring until the area is accepted as fully closad. It is posible that the regulatory body will do or require infrequent monitoring (say e.:: in five years) for some years after final closecut. 4 l 06857 l l t l
9 11 I i I t / \\ g ~ ~~ m u 1 m. g: n, 1, Rl-3 u i n 1 !5 I 1 M I d W E j.."." - = s y \\ GIN $ Is W124W
- b 3=w25 g
k' E' e o latef*8%W e --!{ts. 1 N 3 g 'I I C I 1 1 ] lI k I I a: 2I l 5 u 51 1 5 1 dltl l! I i N 1 8 L 4 m l.ir~ I y o. I y I ouC 1$ l!I 1i=1 & =3<w ~ ii. eer 1N O % *g sv i D$ )14 1 +ww .g.._.____ .a
4 e. D er Figure N. nose.e viu a f*Cm SC P.* A min 6. S Y R A filO%'Ct..DC Or.oht r.1141C M r. AN or Se va n g Ag,3 AT t t)Ds:N st AY Mi yceton se. h 22 Sc 224 s 4 21
- . 2 3 '
/ u 234 r g I . e. e #et e e 4,- A Mel a ese so. e em een se me se see seen 793 eM a e st Freear e.g . se i-,m osi a e icn=ce enous e==
- * = = = - = + = * =-* -=====..
Desos i ese essee. w g .e. e-o.. e ri 4, mees tese e ..a q, 49ead $d8ee ge I ,,e e
- t8 e.. q '
e mesen - ao e as aus em em m emme se e e g . e e. e-eamm e e as se tem 6* 9 4.e4 se+4 sem see se es eenet # WG 'tes 6}}