IN SITU RADIOLOGICAL ASSESSMENT of SOILS in CEA AIEA - … · 2010. 9. 20. · DSV/FAR/USLT/SPRE...

1

IN SITU RADIOLOGICALASSESSMENT of SOILS in CEA

AIEA - MeetingDidier Dubot - Julien Attiogbe

Emilie AUBONNET, Patrick De Moura, Yvon DESNOYERS, Julie FARGIN, Léa PILLETTE

DSV/FAR/USLT/SPRE Section Assainissement du Site 5-9 JULY 2010

☛☛☛☛☛☛☛☛ [email protected]@cea.fr -- ✉✉✉✉✉✉✉✉ 06.7706.77--7373--4141--1414

2DSV/FAR/USLT/SPRE Section Assainissement du SiteSection Assainissement du Site 25 JUIN 2010

❘❘❘❘❘❘❘❘ Historical of CEA/Fontenay Centre❘❘❘❘❘❘❘❘ Rules and clean up methodology❘❘❘❘❘❘❘❘ Tools dedicated to the radiological characterization of contaminated sites❘❘❘❘❘❘❘❘ Vehicle laboratory

❘❘❘❘❘❘❘❘ Characterization with DSP and Spectrometry gamma in situ❘❘❘❘❘❘❘❘ Software developed by CEA Clean up Section❘❘❘❘❘❘❘❘ Sampling process and data processing by Geostatistical method❘❘❘❘❘❘❘❘ Optimized remediation and examples❘❘❘❘❘❘❘❘ Conclusions

SUMMARY


CEA/FAR: 1st CEA Centre, created in 19451999: Decision to clean up the site � Remediation Service: SAS

• Clean up of hazardous places done by 2012-2013• Tools development dedicated to sites and soils remediation• Expertise in French and foreign sites

Exploitation of the 2nd generation facilities Progressive stop of

nuclear activities and 2ndgeneration facilities decommissioning

Exploitation of the 1st generation facilities

Facilities decommissioningNew facilities for medical

research

The CEA of Fontenay-aux-Rosesand the «CLEAN UP SECTION»


❘❘❘❘❘❘❘❘ No remediation level in France for polluted site and sols

❘❘❘❘❘❘❘❘ No regulation for soil under and around nuclear facilities

❘❘❘❘❘❘❘❘ Optimizing the sanitary impact: 0,01 mSv/yr < Sanitary impact < 0,3mSv/yr

❘❘❘❘❘❘❘❘ Necessity to have methods, procedures and reliable tools dedicated to sampling

Waste production

150 t LA (500GBq)

10000 t VLA(30GBq)

20000 t industrial

waste

Clean up Clean up ProcessProcess of of CEA/FAR Site CEA/FAR Site �� 45 M45 M€€

Radiological surface evaluation

Area functional analysisData gathering

Radiological in-depth evaluationChemical measures since 1998

Cost/advantages study

Optimized remediation

Historical investigations

Final characterization

METHODOLOGY


Atmospheric measurementsBeacons to measure aerosols and ambient radiation.

Weather measurementsWind vane, anemometer, temperature sensor, pluviometer, relative humidity

Differential GPS with submetric accuracy

Gamma spectrometryin situ

Samples treatment cell

Centralization of the measures in the front

of the vehicle

Map makingOptimization of the

sampling planDiffusion gradients

Data processingthrough geostatistics

Monitoring of radiological incidentsWind scatteringGas and aerosols activity measurements

Containment Room Follow-up of the gamma environment.Centralization and management of equipment and measurement operation

LABORATORY VEHICLE: LAMAS


Ground samples activity measured by in situ alpha spectrometryGround samples activity measured by in situ alpha spectrometryMeasurements are taken by a PERALS system, inside the convey.It needs ventilation, a hotplate and centrifuge machine.

This alpha spectrometer has 3 characteristics:The sample remains in liquid phase. No electrodeposition is necessary,It uses a selective scintillating-extractant, reacting only with the targeted radioelement. There are different scintillating, according to the alpha transmitter to be measured, Its output (> 99%) and the low background noise make it a good quality instrument for the search of actinide traces on thesite.

Protocols of extraction (alpha totals, U+Pu, U or Pu), developed by BIII/DASE, allow to obtain, with sand or ground, detection limits < 1 Bq/kg within < 24 hours.

Time needed for extraction:………………………………………….5 hCounting time:…………………………………………………………8 h

ACTINIDES MEASUREMENTS

7DSV/FAR/USLT/SPRE Section Assainissement du SiteSection Assainissement du Site 5-9 JULY 2010

The Activity of samples of ground are measured by liquid scintilThe Activity of samples of ground are measured by liquid scintillationlationStrontium-90 is measured thanks to his descendant: yttrium 90, by liquid organically extraction. It requires a ventilated cupboard, a hotplate, a balance, a centrifuge and the Triathler (liquid scintillation measuring device).

The Sample is decontaminated beforehand by precipitating hydroxides with ammonia in chlorydric milieu. After regrowing from purified strontium-90, yttrium-90 is extracted in organical phase, then measured liquid scintillation.

The extraction protocol, for sand or dirt, gives detection limit of 30Bq/kg.Manipulation time:…………………………………. 5 hTime between the 2 extractions:………………... 24 hCounting time:……………………………………… 6 h

MESURING STRONTIUM-90


EXPERTISE VEHICLE:VEgAS

VEgASVEgAS is an expertise and investigation vehicle dedicated to radiological characterization of contaminated places

Vehicle• 4WD vehicle • Speed control system 2,6 ; 5 and 10km/h.

• Differential GPS with submetricaccuracy.

• Dopler SystemInstruments• Hyper pure germanium detector, efficiency 30%

• NaI detector (crystal 2,4L)• Two plastic scintillation detectors

• Centralization and supervision of the measures in the vehicle


REAL TIME ACQUISITION: COMPLETE CONFIGURATION

Choice of the instruments and measurement time.Initialization, reset…

Choice of the features for the data file

Displaying the data, each second

Displaying the cps of the selected isotopes measured by gamma spectrometry


REAL TIME ACQUISITION:FAST CONFIGURATION

Choice of the instruments and measurement time.Initialization, reset…

Automated Configuration

Validity of the devicesData reception

Choice of the file’s nameStart the acquisition

Display of functioning messages

Choice of the radionuclides to be measured


REAL TIME ACQUISITION:IDENTIFYING theRADIONUCLIDES

DSP 2DSP 1

Average DSPSpectro Gamma Total

NaI Total

For each n-second acquisition and for each radionuclide, we can observe how many cps were detected by the gamma spectrometry device

This table indicates the min & max values of each radionuclide


VEgAS:ACQUISITION, TREATMENT

and CARTOGRAPHYGPS

2 DSP NaI – Gamma SpectroKrigéo

Kartotrak RT Kartotrak RT

~ 4h acquisition (1s measure)


VEgAS:DSP DETECTIONPERFORMANCES

Detection performances of a punctual source at 2.5km/h

0,30 kBq2,9 kBq1,8 kBq37 kBq2,6 km/hPUNCTUAL SOURCE

5 km/h

4,4 kBq57Co E=124 keV226Raeq=60%Multi γ

137Cs E=662 keV241Am E=60 keV

LD

0,47 kBq

2,8 kBq58 kBq

Factors and detection limits for 1 m² ground /tar pollutions

2,6 km/h 5 km/h 10 km/h 2,6 km/h 5 km/h 10 km/h13 7Cs 0,309 0,369 0,513 57,2 68,3 94,96 0Co 0,162 0,193 0,268 30,1 35,9 49,9152 Eu 0,210 0,251 0,349 38,5 45,9 63,8

2 2 6Ra alone (FE=100 %) 0,0187 0,0223 0,0310 3,41 4,08 5,66

2 2 6Ra alone (FE=60 %) 0,0312 0,0372 0,0517 5,69 6,79 9,43

30,7 36,7 50,9

Radionuclides Detection Limits (Bq/kg)TAR ROADS GROUNDS

2 2 6Ra + descendants (FE=100 %)

Detection Limits (Bq/kg)

0,168 0,201 0,279

When driving at 2.6 km/h, the detection limit for 137Cs is approximately 1 Bq/g

(only with spectrometry gamma in situ)


IN SITU GAMMASPECTROMETRYMEASUREMENTSHyper Pure Germanium Detector, 30% effectivenessHyper Pure Germanium Detector, 30% effectiveness

-- Modeling 5 mModeling 5 m²²-- Measured thicknesses from 1mm to 100 cmMeasured thicknesses from 1mm to 100 cm

When driving at 2.6 km/h, the detectionWhen driving at 2.6 km/h, the detectionlimit for limit for 137137Cs is approximately 1 Cs is approximately 1 Bq/gBq/g

Detection limits obtained with a GeHP detector (Germanium Hyper Pure) 30 % effectiveness, 1 hour of acquisition

RadionuclideSurface

measurement (Bq/m²)

Measure thickness 1 cm

(Bq/kg)Measure

thickness 5 cm (Bq/kg)

Mass measurement thickness 10 cm (Bq/kg)

241Am 66 9,7 5,1 4,4228Ac 75 7,9 2,9 2,2137Cs 24 2,8 1 0,860Co 21 2 0,8 0,540K 151 15 5,4 3,9


GEOPHYSICSINVESTIGATIONS

Confirm the position of drill holes by checking the absence of metallic parts or underground network near the drill holes.

anomaly


SAMPLING TECHNIQUE

Drillings with auger or Drillings with auger or GeoprobeGeoprobe

• Depends on the field, maximal depth: 2m• Simple and relatively cheap• Unsuitable for chemical measures• Unsuitable for containment

Drillings with ultrasonic deviceDrillings with ultrasonic device• Every type of soils• Maximal depth 50 to 100m• Suitable for containment• Expensive technique


SOFTWARE DEVELOPMENT

RADIOLOGICAL CHARACTERIZATION for CONTAMINATED SITES through DATA PROCESSING ((KrigKrigééoo)) ASSOCIATED to a GIS((KartotraKKartotraK))OPTIMIZATION & DECISION making AID module

(STRATEGE)(STRATEGE)


SOFTWARES dedicated to the GEOGRAPHIC representation of the MEASURES in the

DIFFERENT STAGES of the CLEAN-UP process

Preparing the sampling plan

Observing the extension of the pollution

Establishing the theoretical mesh

Validating the clean-up objective

Calculation of the projected impact

Interaction of the different softwares

Real-time data acquisition

Data processingthrough Geostatistics


KARTOTRAK:GEOGRAPHIC INFORMATION

SYSTEM

Vectorial Maps� No pixellisation effect when zooming in Different layers: vectorial + raster

Path of the vehicle

Georeferenced maps� Used with a differential GPS + Doppler radar

Grid Editing


KARTOTRAK:GEOGRAPHIC INFORMATION

SYSTEMDifferent working processes: systematic mesh� Specify areas to be/not to be measured� Specify areas with an historical interest (known location of contamination)

Area of interest- 730 m²

245 245 georeferencedgeoreferenced datadataArea to be measured8400 m²

Excluded area(building) 2150 m²


EVALUATION OBJECTIVESand SAMPLING:STRATEGE

Different sampling plans for different evaluationsDifferent sampling plans for different evaluations

Random sampling Specific search sampling Circular grid sampling

BRGM data

Regular samplingProfile sampling Appraisal sampling


STRATEGE:SAMPLING OPTIMIZATION

5000 m²400 measures

13000 m²270 measures

Distance (m) Distance (m)

Variability Variability

Experimental variogram

Model

Too much ? Not enough ?


DECISION MAKING AID MODULE:STRATEGE

“STRATEGies d’Echantillonnage”Evaluation objectiveEvaluation objective• Removal of doubt / Initial cartography• Environmental diagnosis

• Radiological characterization• End of remediation controls

Determine and optimize the sampling planDetermine and optimize the sampling plan• Optimized grid• Circular grids• Drillings distribution• PESCAR, Wilks method

Forecast measures performancesForecast measures performances

Estimating the projected budgetEstimating the projected budget

Editing a report of the evaluation preparationEditing a report of the evaluation preparation68%

32%


STRATEGE:SAMPLING OPTIMIZATION

Analyzing the experience feedback

Optimizing the systematic gridin order to use geostatistics

(highlight the spatial structure of the data)

Considering the informations of the area ( historical, geology, topography) and the evaluation constraints

(budget, performances…)

Graphic indicators:- Probability of hitting a target- Impact/relevance of extra measures

0

20

40

60

80

100

120

0 5 10 15

Dimension de la cible (m)Pr

obab

ilité d

e tou

cher

(%)

Normal Intérêt

1645

1399

1806

1645

17551586

271

871

0

0,2

0,4

0,6

0,8

1

1,2

0 5 10 15 20 25Distance h

Vario

gramm

e (h)


RADIOLOGICAL ASSESSMENTAFTER CLEAN-UP PROCESS

FINAL CONTROLSPESCAR: Assessment of the average remaining activity, in order to determine the sanitary impact of the cleanup.

M1 WeightSamplesM2 WeightSamples

Random Sampling of 2x10 samples •• Computing the homogeneity constant AComputing the homogeneity constant A•• Computing the segregation constant BComputing the segregation constant B��Computing the optimum mass MComputing the optimum mass M��Computing the optimum amount of Computing the optimum amount of

samples for the last stagesamples for the last stage( ) 22 5.0 abaD Ζ+

Ζ+Ζ≥Ν

12

22

2121 )(MM

SSMMA−

−=

2

22

1

21 M

ASMASB −=−=

BAM optimale =

M1 Weight Samples

M2 Weight Samples

N Samples, MoptimalWeight

Random Sampling of N samples


RADIOLOGICAL ASSESSMENTAFTER CLEAN-UP PROCESS:

FINAL CONTROLSWilks Formula: using statistics with a few samples

( )[ ] βα ≥≥≤ maxXXPP

βα =− N1 α, proportion of the studied variableβ, confidence levelN, amount of measurement


GEOSTATISTICS: BASICS

Kriging

Histogram

Data Analysis

Experimental variogram and model

Known Data

2D Interpolation (initial cartography)

3D Interpolation (with drillings)

2D 3D


EXAMPLE of INITIALCARTOGRAPHY KRIGING and

UNCERTAINTY MAP55 DrillholesResolution: 30 cmKriged area: ≈ 5000 m²

Applicable to small and big surfaces

High variability

High uncertainty can be «solved» by collecting more measures if it’s not due to a pure nugget effect.


RISK ANALYSISPROBABILITY MAPS

Allows to estimate the local probability of exceeding

a given value

High probability, very likelyIntermediary probability, uncertaintyLow probability, very likely

� Complementary to the contaminated surface estimation (through simulation)� Helps to establish a relevant drilling positioning


MIGRATION PROFILESEXAMPLES (α, β, γ)

Profile A

0,020,010,01

0,010,05

0,30,1

0,32,12,0

0,91,3

2,30,5

1,11,1

-227

-207

-187

-167

-147

-127

-107

-87

-67

-47

-27

-7

0,001 0,01 0,1 1 10 100Activity (Bq/g)

Dept

h (cm

)

somme betasomme alphaActivité totalecésiumα/β

t

Profile B

0,02

0,52

0,87

0,61

2,63

0,75

-197

-167

-137

-107

-77

-47

-17

0,01 0,1 1 10 100Activity (Bq/g)

Depth

(cm)

somme betasomme alphaActivité totale α/β

Profile C

-1500

-1400

-1300

-1200

-1100

-1000

-900

-800

-700

-600

-500

-400

-300

-2000,001 0,01 0,1 1 10 100

Activity (Bq/g)

Depth

(cm)

SrPu239césium

Sampling step: 10 cm (2 m 30 core)

Sampling step: 30 cm(2 m core)

Sampling step: 1m(15 m core)


FROM the 2D to the3D VIEWING

The preliminary 2D mapping leads to a drill hole campaign processed in 3D

Drill holes

View from below


DEPTH EXCAVATIONOPTIMIZATION

Drillings and samplings

Analyses results

Typical spectrum determination per zone

In depth profile plotting

Excavation depth choice

Optimization graph functions of a remediation scenario

Cost/advantage study

Optimisation impact/coût (profil A)

1081

2109

1512

617

65 7 4,8 6,7 6,0 3,3 3,63,824

3342

6069

78

98107

116

155

87

51

0

500

1000

1500

2000

017-027027-037037-047047-057057-067067-077077-087087-097097-110110-120120-130130-140

Profondeur (cm)

Impa

ct µS

v/an

0,015,030,045,060,075,090,0105,0120,0135,0150,0165,0180,0195,0

Coût

(k€)

construction batimentcoût

In depth profile

0,020,010,01

0,010,05

0,30,1

0,32,12,0

0,91,3

2,30,5

1,11,1

-227

-207

-187

-167

-147

-127

-107

-87

-67

-47

-27

-7

0,00 0,01 0,10 1,00 10,00 100,00Activity(Bq/g)

Profo

ndeu

r (cm)

somme betasomme alphaActivité totalecésiumα/β


DEPTH EXCAVATIONOPTIMIZATION

Drillings and samplings

Analyses results

Typical spectrum determination per zone

In depth profile plotting

Excavation depth choice

Optimization graph functions of a remediation scenario. Cost/advantage study

Remediation scenarios come from IRSN methodology guide: “ Industrial sites management potentially

contaminated by radioactive substances ”• Residence and playgrounds• Primary school• Market gardening• Offices• Public car park• Waste land• Building site• Car park under construction


CASE 1: SURFACECARTOGRAPHY

Highlighting 2800 m² of interest out of a 7050 m² zone � one day after

Real Time Measuring Identifyingthe areas of interest


PROCESSING by 2D/3D LAYERS

585040

585040

585050

585050

585060

585060

585070

585070

X (m)

X (m)

2410700 2410700

2410710 2410710

2410720 2410720

2410730 2410730

2410740 2410740

2410750 2410750

2410760 2410760

2410770 2410770

2410780 2410780

2410790 2410790

Y (m) Y (m)

Ra_226 (Bq/kg)43851094 790 605 484 381 303 243 203 173 147 126 110 96 85 72 31

Activité (Bq/kg)

Drill holes

2D

Layer 16Layer 16--32cm32cm Layer 32Layer 32--50cm50cmLayer 0Layer 0--16cm16cm

Probability to exceed a given value

3D

Activity > 500 Bq/kg


SUMMARY

Underground evaluation

Roads monitoring

Radiological evaluation of walls and floors


ConclusionsConclusions

� Sampling takes a critical place in our project management� More than 120 sites characterized and permanent feedback� Current industrialization of the software platform Kartotrak� Cost/advantage study allows to optimize the remediation in function of the rehabilitation, to justify the remediation choices before the safety authority and to control cost and delays.

HIGH

Total Cost

Rehabilitation

Investigations

Rela

tive

Cos

t

BASIC APPROACH

DETAILED APPROACH

HIGH LEVEL APPROACH

Edited by BRGM, D. Hubé

LOW

The better you characterize upstream with surface and in-depth measures, the better you manage the remediation operations, respecting the projected cost and deadlines.

Scientific promotions, publications, seminars and collaborations :Decommissioning Avignon 2008; SIEN Bucharest 2009; Statgis 2009, SFEN 2009,

Intersol Paris 2010, DDR Idaho 2010, Environet program (AIEA), Collaboration with China, Russia and South Korea about contaminated sites and soils

IN SITU RADIOLOGICAL ASSESSMENT of SOILS in CEA AIEA - … · 2010. 9. 20. · DSV/FAR/USLT/SPRE...

Documents

Transcript of IN SITU RADIOLOGICAL ASSESSMENT of SOILS in CEA AIEA - … · 2010. 9. 20. · DSV/FAR/USLT/SPRE...