activation of targets and accelerator components at psi

42nd ICFA, Advanced Beam DynamicsWorkshop on High-Intensity, High-Brightness Hadron Beams, Nashville, 25.8.-29.8.2008

Activation of Targets and Accelerator Components at PSI

- a Comparison of Simulation and Measurement

Daniela Kiselev, Sabine Teichmann, Michael Wohlmuther

Contents:• Overview of PSI• Nuclear reactions• Codes used at PSI to calculate nuclide inventories, dose rates• Applications: 1) Nuclide inventory in samples around SINQ target 3 2) Dose rates along SINQ target 4 3) Dose rates at the PIREX beam line 4) Nuclide inventory in samples from the E4 beam line• Comparison of the 2 calculational methods• Summary


Proton accelerator facilities + Experimental hall

Birds view of the Paul Scherrer Institut (PSI)


Injector 2Ring cyclotron590 MeV

Target M

Target E

SINQ

UCN Comet: 250 MeVGantry 1

Gantry 2Upgrade 2 mA to 3.0 mA in progress PROSCAN

OPTIS

OPTIS 2

Injector 1

PSI Proton Accelerator Facilities


All beam line components are surrounded by iron shielding.

Beam line from Target E to beam dump (side view)2.

5 m

Concrete shielding

Working platform

Target E Collimators Diagnostics

Beam dump

5 m

movable slitsinflatable seals

to SINQ


Components around beam line get activated... need to be removed eventually

• for replacement waste Nuclide inventory needed before disposal as required by swiss authorities

• for repair/dismantling planning of work procedures dose rate estimation needed

in addition:for future (planned) installations/facilities:• estimation of the amount of total waste after operation• dose rates needed for construction of shielding


excitationn

AA+1

1) Direct irradiation (590 MeV p): at loss points (target, collimators)

main process: spallation

2) Secondary particle fields:

after penetration of shielding: mainly neutrons,

thermalized in shielding material

equilibrated energy spectrum

important process: neutron capture at thermal energies,

e.g 59Co(n,)60Co, 107Ag(n,)108mAg

Activation due to

Evaporationof protons,neutrons,also2H, 3H .. 4He

pprimary particle

highly excited nucleus spallation


case 1: mainly due to primary irradiation

Monte Carlo particle transport program : n,p,d,3H....Input: dedicated geometry, material compositions, cross sections: for n<20MeV ENDF-B-VI.6, otherwise modelsOutput: n-fluxes (E<20 MeV), residual nuclei production rates

via Perl script

Cinder’90 Orihet3SP-Fispact or or

Input: irradiation and cooling history cross sections: Cinder library, EAF 2003 decay properties of isotopes Output: nuclide inventory, photon rate

MCNPX

used as source in MCNPX

via Perl script

dose rate

Calculation of the Activation

decaycodes:


- book-keeping prg. to calculate nuclide inventory- contains PSIMECX cross section library (developed at PSI)

nuclide inventory, scaled to measured dose rate.

Input:

choice of the representativen-flux spectrum

Production rates by folding with cross sections(PSIMECX)

choice of11 materialcompositions

Buildup/decay (Orihet)

case 2: mainly due to secondary particles PWWMBS

weightmaterial

surface dose +irradiation history

location (58 zones)

Output:


1. Activation in the close vicinity of the SINQ-Target 3

irradiation: 6.77 Ah for 2 years

p-beam

geom. model in MCNPX:3 samples:

Zrscrew

AlMg3hull

shielding316L

Zr tubes


1.0

4

1.0

0

0.3

0

1.2

6

0.9

5

8.0

7

1.2

9

1.0

0

0.9

9

0.3

0

1.2

6 1.5

4

7.7

2

1.2

9

0.6

4

0.6

0

0.6

4

0.5

9

0.1

1.0

10.0

Cl36 Mn53 Mn54 Fe55 Co57 Ni59 Co60 Fe60 Ni63

‘s in Ge-detector chemical + quantitative Analysis + measurement in Liquid Scintillator Accelerator mass spectrometry (AMS)

measurement done byradiochemist group at PSIin collaboration with AMS groupsin Zürich + Munich

Results for steel shielding: exp. / calc. activity

Nuclide inventory

2 different material compositions

simulation by MCNPX + Cinder‘90


0.69 0.

91

0.71

1.48

5.44

1.40

0.82

0.86

0.79

1.63

0.72

0.65

0.61

0.10

1.00

10.00

Na22 Cl36 Ti44 Mn54 Co57 Co60 Zn65 Ag108m

0.1

0.1

Results for safety hull: exp. / calc. activity

remarks:• overall good agreement with MCNPX+Cinder‘90• Na22: Data/SP-Fispact = 0.72• Fe60: deviation of factor 8 seen in many other samples• the right material composition (including impurities) is important!

2 different material compositions

not obtained in calculation

simulation by MCNPX + Cinder‘90


2. Planned: New Hot cell at PSI

What is the required wall thickness (shielding) ?

Most activated piece at PSI: SINQ target (Pb in Zr tubes) calculation of the nuclide inventory with MCNPX + Cinder’90 calculation of the thickness of the shielding with Microshield

How good are the predictions?5 SINQ targets were irradiated and dismantled

Compare calculated dose rates with measured data


0.1

1.0

10.0

M1 M2 M3 M4 M5 M6 M7 M8M1

M8

M7

M6

M5

M4M3M2

target:steel tubesfilled with Pb

stee

l shi

eldi

ng

Dose rates at SINQ Target 4

irradiation: 2 years with 10 Ah, cooling: 10 monthsdose rates obtained at 3cm, 30cm, 100cmdose rates calculated with MCNPX + Cinder‘90

Dose rates in Sv/h at 100 cm distance

calculation

data


0.1

1.0

10.0

100.0

1000.0

M1 M2 M3 M4 M5 M6 M7 M8

calculation

data

Dose rates in Sv/h at 3 cm distance

discrepancy: • factor 2 in the target region• factor 10 for small distance at the shielding reason: in calculation contribution also from target region Geiger Müller tube: smaller solid angle

measurement:

GeigerMüllertube

simulation:

x 10


purpose (1988 - 2004):material irradiation and degradation for Gantry 12006:dismantling of the beam line calculation of dose rates with MCNPX + comparison with measured doses

1 2 3

model in MCNPX of vacuum chamber

Beam current monitor

Degrader (Cu + C)

Beam Dump (Cu)

1

2

3

Assumption (made in simulation):

Degrader 2 and Beam dump 3are always in beam Irradiation history (simplified):16.6 y @ 8.3 μA 2.3 y cooling

3. Dose rates at the Pirex beam line at PSI


Comparison with data:

Overall agreement good

Average ratio 1.13

Two points with ratio > 2:

smaller dose as measured in

beam dump:

degrader was not always in

0.350(0.434)

0.020(0.057)

0.700(0.501)

0.660(0.315)

0.300(0.279)

0.380(0.295)

0.230(0.255)

0.115(0.181)

0.700(0.529)

Sv/h

Results of dose rate comparison:

calculated values by M. Wohlmutherin brackets


E4

Target E

ASK61

sample from beam tube

4. Activation of samples in the E4 beam line

Several samples were taken from• bending magnet ASK61• shutter behind ASK61• shielding around shutter

components are not directly irradiated calculation of the activity with PWWMBS use representative n-flux spectrum in vicinity of Target E

bending magnet ASK61(stainlesssteel) irradiated

for13 years

beam entry

2004:


1.4

3.6

1.4

3.5

1.6

3.1

1.4

1.2 1.4 1

.7

1.3

4.2

3.7 4

.9 6.2

0.2

8

0.0

16

0.6

25

0.01

0.1

1

10

Na2

2

Al2

6

Cl3

6

Sc4

6

V48

Cr5

1

Mn5

4

Fe5

9

Fe5

5

Co5

6

Co5

7

Co5

8

Co6

0

Ni6

3

Zr8

8

Y88

Nb9

5

Sb1

24

Results for beam entry at ASK61

Measured activities via• detection• Accelerator mass spectromety for Al26, Cl36, Ni59• Liquid Scintillation: Ni63

exp./calc. activity

calculationwithPWWMBS+ typicalstainless steelcomposition

calc. values from F. Atchison & M. Wohlmuther


• Electrical installations• Targets• Vacuum Chambers

• Collimators• Beam Diagnostic Monitors• Magnets• Shielding

Radioactive waste filled into Containers

Total activity per container: 1010 – 1012 Bq (4.5 t)

0.01 0.1 1 10 100

other metals

tungsten

PVC

insulation materials

lead

aluminum

copper

cast iron

concrete

stainless steel

normal steel

MA

TE

RIA

LS

CONTRIBUTION IN % OF TOTAL

Material compositon:

for most of thecomponents:nuclide inventoryobtained withPWWMBS


Comparison of the calculational methods

• PWWMBS: activation due to secondary particles main purpose: nuclide activity calculation of radioactive waste+ fast throughput, user friendly+ minimizing of personal dose (no detailed measurements or samples)- no predictions possible (dose rate needed)- accuracy: factor 10 due to simplified assumptions, sufficient for authorities

• MCNPX: activation due to primary irradiated fields main purpose: nuclide inventory calculation dose rate calculation energy deposit (input for ANSYS temperature distribution)+ predictions possible- elaborate geometry input- large cpu time (~1day on 256 - 512 nodes), statistical limitations+ accuracy: factor ~ 3 (except some outliers)

Comparison between simulation and measurement important


Summary:Predictions/calculations of the nuclide inventory and dose rates important for- disposal- personnel work at activated components

Solution at PSI:direct irradiation:MCNPX + Cinder’90 (or SP-Fispact, Orihet3)secondary particle fields:PWWMBS (nuclide inventory only)

Examples:1) Nuclide inventory of shielding + safety hull of SINQ target 3: good agreement2) Dose rates along the SINQ target 4 insertion: agreement within factor 23) Dose rates in vacuum chamber of PIREX: agreement within factor 24) Nuclide inventory of ASK61: exp/calc. < 10, except for Cl36

UC

N PIF

G

DirektSekundäres Strahlungsfeld72 MeV Neutronen (noch nicht spezifiziert)

1) direct irradiation2) secondary particles: n-flux spectra calculated for 72 and 590 MeV area, dependent on - shielding material - depth in shielding3) thermal neutrons

Division of the area in regions

Pirex (1988-2004)

SINQ

Gantry 1

Comet Optis 2

Gantry 2

Gantry 1


Meson Production Area

Access to working platform after removing 3-4 m of concrete roof shielding

Use of remotely controlled shielded exchange flasks to remove defective components.

Repair work done in a hot cell.

activation of targets and accelerator components at psi

Documents