Radiation Monitoring - Laboratori Nazionali di Legnaro

1

Radiation Monitoring

Vashek VyletDuke University

2Duke University

Contents

• Introduction• General Aspects• Types of Detectors• Neutron spectrometrywith Bonner spheres• Examples of measured spectra with BS

3Duke University

Introduction

• Goal: Measure/estimate ambient radiation fields generated by accelerators (induced activity - conventional techniques)

• What: Operational (ambient dose equivalent) or physical (Φ(E)) quantities, the latter can be translated to operational quantities using appropriate conversion factors

4Duke University

Introduction

• Purpose 1: Insure radiation safety of workers & public; evaluate and monitor efficacy of specific protection measures; backup information for personnel dose investigations

• Purpose 2: Insure and demonstrate regulatory compliance (keep records!)

5Duke University

Introduction

• Special to accelerators: great variations in type, energy & spatial distribution; time structure (pulsed fields) of prompt radiation

• Radiation fields depend on great number of parameters need to record details: date & time, average current, E, mode of operation, special machine setup, …etc

6Duke University

Sources

• Mostly photons, neutrons and muons• Beam losses, special operation (conditioning),

auxiliary devices (klystrons)• Direct & scattered radiation near sources,

skyshine from far away• Spatial aspects: uniform field or hot spots,

leakage through cracks, ducts and sharp forward sources (brems., muons)

7Duke University

Type of Measurements: Surveys

• Short/simple survey: field possibly known, looking for “ballpark” values or yes/no information - routinely done during normal operations

• Extensive surveys: at startup of new facility, after long shutdown periods or changes in configuration; normal operation and/or mis-steering and accident scenarios

8Duke University

Type: Long-Term Monitoring

• Continuous long-term monitoring, using active (with recording capability) or passivedevices. Recording of sampled or integrated (preferred) signal

• Stationary active monitors may be interlocked as part of safety system

9Duke University

Type: Special Measurements

• Detailed characterization of radiation fields: components, energy spectra, spatial distribution …

• Detailed characterization of detector behavior - requires additional instrumentation

• Complex measurements under special operating conditions (special measures)

10Duke University

General Aspects - Calibration

• S(E) = response function; ks = spectrum weighted response per unit fluence

tots

tot

s

RdEE

dEEESk

dEEkdEEESR

Φ=

Φ

Φ=

Φ=Φ=

∫∫

∫∫Φ

).(

).().(

).().().(43421

11Duke University

Calibration

• Cs = calibration factor for spectrum S, (hφ)s = spectrum weighted average fluence-to-dose equivalent conversion factor

ssss

s

khCRCH

dEEh

dEEEhh

)( .

).(

).().()(

φ

φ

φφ

=⎯→⎯=

Φ=

∫∫

12Duke University

Dead Time

• Dead time τ = difference in pulse arrival times below which two incoming pulses will not be resolved

• May be determined/dominated by intrinsic properties of detector or counting electronics

13Duke University

Dead Timen = true interaction ratem = recorded count rate

= system dead time

τmmn

−=

1Nonparalysable:

Paralysable: τnnem −=needs iterative approach

14Duke University

Dead TimeParalysable: two solutions possible! Solution: change true rateand observe if count rate increases or decreases.

15Duke University

Dead time with pulsed sources

• << T no problem• < T difficult problem

• T < < 1/f -T n = f .ln(f/(f- m)) does not deped on

m

n

16Duke University

Dead time with pulsed sources

• Rule of thumb: OK if count rate is less than a tenth of rep rate - then loss is ~5%

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 0.2 0.4 0.6 0.8 1

measured rate/machine pulse rate

mea

sure

d/re

al c

ount

rat

e

17Duke University

General Aspects - RF and mag. fields

• RF generates signal in anything acting as antena: unshielded or badly shielded cables & components attached to sensitive preamp.

high level of noise (false signal, distorted MCA spectra, …)

• Magnetic field affect moving charge: PMTs and gas detectors especially sensitive

Liu et aSLAC-PUB

19Duke University

Choice of Instrument & Method

• Adapted to application: en. response, particle type, temperature sensitivity, rf, …

• Care to avoid dead time problems, paralysis, saturation & recombination in intense fields

• Interlocked monitors should be failsafe -use check source or natural bkg, although the latter may pose problems (low rates)

20Duke University

Interlocked Area Monitors

Photon IC

Moderated BF3

21Duke University

Choice of Instrument & Method

• Unknown fields have to be characterized (energy or LET spectra) and proper calibration factors established

• Take into account directionality of field & detector response

• Take precautions during special (mis-steer, accident simulation) measurements requiring interlock bypass, roping of areas...

22Duke University

Gas Detectors

• GM, IC, PC: primary application in photon survey meters; IC & PC also for neutrons, with propper wall and gas choice

• Problems: pulsed fields, mag. & RF fields• Large ICs: stationary (interlocked) monitors

(BSOIC at SLAC, Chipmunk at Fermilab ..)• ICs operated in current mode impervious to

pulsed fields (except higher recombination)

23Duke University

Gas Detectors

• GMs - bad dead time problem (paralysable)• GMs not suitable for neutrons (equally

sensitive to photons), but …• Albatros: Moderated GM wrapped in silver

- activated by neutrons (saturation within minutes) - no trouble with pulsed fields

• PCs filled with BF3 or 3He used for thermal neutron detection

24Duke University

Special Gas Detectors: TEPC

• Tissue-Equivalent Proportinal Counter: TE walls and TE gas - from PHA one can determine D(y) ~D(LET) average Q

• HANDI used at CERN• Problems in pulsed fields• Not very sensitive (for 1-micron sphere)

25Duke University

Recombination IC

• Takes advantage of dependence of columnar recombination on LET

• I = kVn, where n=f(Q), or• Double chamber method (one at high and

one at low voltage)• Tricky, not widely used around accelerators

(earlier used at CERN and Fermilab)

26Duke University

Proton recoil PC (neutrons)

• Filled with hydrogen or hydrogenous gas, neutrons detected by proton recoils

• Neutron spectrum can be obtained by unfolding recoil proton spectrum

• Problems: sensitivity, pulsed fields• Narrow energy range for given volume and

pressure; 10 keV - few MeV requires use of several detectors in parallel (ROSPEC)

27Duke University

Scintillators

• Used in photon survey meters and monitors• For muons: photon bkg eliminated by

coincidence between two detectors forming a directional telescope (with absorber)

• Organic scintillators used for neutrons (proton recoil); photon bkg can be removed by pulse-shape analysis

28Duke University

Fermilab’s MERL

• Muon telescope in MERL

• Also portable version (muon “gun”)

29Duke University

Passive Detectors

• Nuclear Emulsions: disappearing art• TLD, OSL and track-etch: photons &

neutrons, mostly for personnel dosimetry, but also area monitoring

• Activation detectors: neutrons• Optichromic and radiochromic detectors:

high level doses (near beamline)

30Duke University

Bubble Detectors

• Superheated droplets suspended in gel• Neutron and gamma• Active or passive• Limited use life• Limited dynamic range

31Duke University

Moderated Neutron Detectors

• With active or passive th. n. detectors• Fast neutrons moderated to thermal

energies detected with greater efficiency, using high x-sections for thermal neutrons (10B, 3He, 6Li)

• Surveys, prolonged monitoring, and spectrometry

32Duke University

Moderated Neutron Detectors

• 5” PE cylinder with Panasonic TLD card used for area monitoring and SLAC and Duke

33Duke University

Rem-Meter• Response ~ Fluence-to-Dose-Equivalent

conversion factor hφ R~H, independent of neutrons energy counts ⇔ rems

10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102 1031.0×10-12

1.0×10-11

1.0×10-10

1.0×10-09

E [MeV]

h φ [S

v.cm

2 ]

34Duke University

Extension of E-range

• Energy response of ordinary rem-meter falls off beyond ~10 MeV

• Lead layer: (n,2n), (n,3n), …conversion to neutrons with lower energies (Linus:Birattari et al., 1990)

• Used in astrophysics for decades to detect cosmic ray neutrons (BF3 counters, PE, Pb)

35Duke University

Detection time in moderators

• Moderation, and especially thermal walk take time detection delayed from neutron entry into moderator volume

• Slow neutrons delayed in shielding• This leads to spread of detection time in

pulsed machines (to tens, even hundreds μs) and greatly alleviates dead time problems

36Duke University

Arrival time distribution - 3” ball

Liu et alRPD

37Duke University

Arrival time distribution - 8” ball

38Duke University

Arrival time distributions

• ANL, room return (20 MeV e- linac)

39Duke University

Neutron Spectrometry

• Proton recoil - PCs & scintillators - good E resolution, limited energy range, problems in pulsed fields

• Bubbles: dynamic range; statistics; cost• TOF - great resolution; complex, very low

sensitivity (not suitable at RP levels)• Activation threshold detectors: low

sensitivity, laborious readout;

41Duke University

Neutron Spectrometry

• Bonner spheres: set of spherical moderators of different sizes, used with th. n. detector

• Wide energy range, high sensitivity, low E resolution (sufficient for RP)

• Energy response can be further extended by use of (n,xn) in heavy materials (Pb, W)

42Duke University

Spectrometry with BSS

• Principle: measure n count rates, then unfold Φ(E):for i=1, …,m

• Typically m=5-8, n=30-100

• Under-determined system; additional info provided through “1st guess”, smoothing, and other constraints on the spectrum

∑=

Φ=n

gggii SR

1, .

43Duke University

Spectrometry with BSS

• Unfolding codes - same/similar as for spectrometry with activation foils

• BUNKI, LOUHI, SAND2, SWIFT (MC), MAXED, ...

• Most important: well tuned response matrix

44Duke University

Extension of E-range - Pb

10010-8 10-6 10-210-4 102 104

ENERGY [MeV]

SE

NS

ITIV

ITY

[a.u

.]

with leadno lead

1

10-1

10-2

10-3

12"

45Duke University

Extension of E-range - 11C

• 11C activation (threshold ~20 MeV) in 5x5”plastic scintillator (J. B. McCaslin, 1960)

• (n,2n) cross-section added to response matrix

• Saturated activity entered in lieu of sphere count rate

12C(n,2n)11C

00.5

11.5

22.5

3

10 100 1000

E [ MeV]

N. σ

46Duke University

Extension of E-range - 11C

• Typical irradiation time: ~40 min (2.T1/2)

C … net count rate ε … counting efficiency

)e)(1eε.(eC.λA

irrcdd λ.t)tλ(tλ.tsat −+−− −−=

Irradiation tirr td counting tc

11C : competing photo-activation 12C(γ,n)11C at electron machines

contamination ( plastic bags)

48Duke University

Extension of E-range

0.0E+0

5.0E-2

1.0E-1

1.5E-1

2.0E-1

2.5E-1

1.E-1 1.E+0 1.E+1 1.E+2 1.E+3

E [MeV]

Sens

itivi

ty [c

m^2

] 8”

12”

12Pb

C-11

49Duke University

Duke BS System

30.48

PA PortableMCAHV

A

25.420.32

14.2 8 cm

He-3counter

12.7x12.7cmscintillator (C11)

Optimized configuration of lead layer

50Duke University

14.2 cm sphere with inserts

51Duke University

Response Matrix (Duke system)

0.E+0

1.E-2

2.E-2

3.E-2

4.E-2

5.E-2

6.E-2

7.E-2

8.E-2

9.E-2

1.E-1

1.E-7 1.E-6 1.E-5 1.E-4 1.E-3 1.E-2 1.E-1 1.E+0 1.E+1 1.E+2 1.E+3

E [MeV]

Sens

itivi

ty [a

.u.]

bare

3.15"PE

5.6" PE

5.6"Pb

8" PE

8"Pb

10" PE

10" Pb

12" PE

12" Pb

Measured Neutron Spectra and Practical Experience with Bonner

Spheres

10” at Riken

53Duke University

What to expect ...

• Proton machines:– hadronic “hard” shower– neutrons:

• “rapid” component (> 20 MeV), part of cascade• evaporation neutrons from residual nuclei (~MeV)

– most penetrating part: neutrons >150 MeV (for E>150 MeV σin reaches a constant minimum: σin = 43.1 A0.70 )

54Duke University

What to expect ...

• Electron machines:– Elmag. “soft” shower– photoneutrons from

three production processes

– behind thick shields spectrum determined by high E neutrons

100

10-1

10-2

100 102 103

PHOTON ENERGY [MeV]

Giant Resonance( ,n)γ

Pions

Pseudodeuteron

/A(m

b/N

ucle

on)

σ

Photonuclear Cross Section

55Duke University

What to expect ...

• Common aspect of e and p machines:– thick shields neutrons with E>150 MeV are

the determining factor– “Equilibrium” spectra: High E portion +

products of hadronic cascade emerging from outer shielding layers

56Duke University

Specific Aspects

• Dead time in pulsed fields– SLAC: e- and e+ pulses from 10 ns to 1.6 μs– Moderation & thermal walk save the situation

(neutron signal spread over 100s of μs)• Variable beam intensity, beam interruptions

need a reference monitor ; more of a problem for 11C activation

57Duke University

Non-equilibrium spectra - SLAC SSRL

0 1 m leadconcrete

SPEAR

Faraday cup

dipole magnetFARC

135 90 45ReferenceBF3 counter

58Duke University

SLAC SSRL

• MCA Pulse-height spectra (with γ-flash)

0

50

100

150

200

250

0 200 400 600 800 1000

Channel

Cou

nts

gate offgate on

59Duke University

SLAC SSRL

• Gamma flash with “non-equilibrium”spectra

60Duke University

SLAC SSRL - SPEAR 135º

Energy (MeV)10 -6 10 -3 10 0 10 30

20000

40000

60000

80000Fl

uenc

e/le

thar

gy

with C-11without C-11

61Duke University

SLAC SSRL - SPEAR 90º

10 -6 10 -3 10 0 10 30

100000

200000

300000

Flue

nce/

leth

argy

Energy (MeV)

with C-11without C-11

62Duke University

SLAC SSRL - SPEAR 45º

10 -6 10 -3 10 0 10 30

50000

100000

150000

200000

with C-11without C-11

Flue

nce/

leth

argy

Energy (MeV)

63Duke University

Measured spectra - CERN

E.Φ

(E)

ENERGY [MeV]100 10310-310-6

105

106

107

concreteiron

64Duke University

Measured spectra - CERN

Iron Shielding Concrete Shielding

Φ<20 Φ>20 Φ total Φ<20 Φ>20 Φ totalFLUKA 11.44 0.66 12.10 0.83 0.62 1.45BS 16.51 0.50 17.01 1.08 0.49 1.57Ratio 1.44 0. 78 1.41 1.30 0.79 1.08

(normalized to 1E+6 counts of monitor IC)

• Comparison with FLUKA calculations:

65Duke University

Measured spectra - SLAC FFTB

• Final Focus Test Beam (FFTB)

Beam Dump

Beam Line

2147 GeV e-

Muonshielding

ConcreteIron0 1 2 3 m

66Duke University

Measured spectra - SLAC FFTB

ENERGY [MeV]

E.Φ

( E)

100 10310-310-6