caliste and its applications

1 CEA Saclay IRFU/DAp, 2 CEA Saclay IRFU/DEDIP,3 CEA Saclay LIST/LCAE, 4 3D plus

SEPT, 26th 2017, CEA, Saclay, France

CALISTE and its applications

Daniel Maier1, P.A. Bausson2, C. Blondel1, F. Carrel3,

G. Daniel1, C. Force3, O. Gevin2, H. Lemaire3,

O. Limousin1, D. Renaud1, J. Martignac1, A. Meuris1,

V. Schoepff3, F. Soufflet4, M.C. Vassal4, F. Visticot1

SUMMARY

Daniel Maier | SEPT 2017 | PAGE 2

ORIGAMIX: a portable

X- and gamma-ray

spectro-imaging camera

fields of application

history of the project

physical concepts

actual status

next steps

SATBOT: realtime

dosimetry

for radiotherapy

NP + radiotherapy

XRF dosimetry

physical concepts

actual status

next steps

CALISTE: a compact

X-ray spectro-imager

module

history

IDeF-X family

CALISTE family

PROJECT HISTORY


CALISTE is based on a long line of experience

but also aims for challenging new developments

based on IDeF-X readout ASICs- started in 2003; now in 7th generation

- properties: ultra-low noise, low power consumption,

channel individual triggered readout

integrated into a CALISTE module- combining several ASICs

- 3D electronics by 3D+

different combinations of ASICs + housing

CALISTE-64 CALISTE-256 CALISTE-HD CALISTE-SO D2R1 CALISTE-HD-BD

2007 2009 2011 2013 2016

more pixel less power better spec. & spat. resolution larger energy range

64 → 256 200mW→200mW 900μm → 300μm 250 keV → 1 MeV

900 eV → 580 eV FWHM @ 60keV

PROJECT HISTORY

ORIGAMIX uses a CALISTE module (HD or O) and

builds a portable device

a portable device

stable & safe housing

ensure vacuum tightness

cooling?

compact, low-power electronics

power supply: batteries, HV?

softwarereliability

easy to use and to understand → alert

control of

operation

user-friendly

mostly autonomous

parameters

calibration

analysis

0.5 mm Al

entrance

window

CALISTE HD

2x TEC modules

Flex PCB connexion

Carrier

motherboard

Power

management

Eth

Zynq SoCMicroZed

Interface board

CALISTE SC & DAQ

Wix

HV crystal bias

House keepingHV cable Daniel Maier | SEPT 2017 | PAGE 4

design:

CURRENT STATUS


detector: Caliste-HD

10 x 10 x 1 mm³ CdTe

camera: system-on-chip aq. system

thermo-electrical coolers

dimension: 7.5 x 7.5 x 23 cm³

mass: m < 1 kg

power: p < 10 W

TYPICAL FIELDS OF APPLICATION


we have a few ideas on that...

inspection after nuclear accidents

TYPICAL FIELDS OF APPLICATION


we have a few ideas on that...

monitoring areasinspection after nuclear accidents

medical imaging / radiotherapy safety inspections

... but there might be others...

PHYSICAL CONCEPTS


CdTe is becomming inefficient for E > 100-300 keV.

Then, Compton scattering gets the dominant effect.

photon interactions detection efficiency

photon interaction with CdTe (ZCd = 48, ZTe = 52)

1 10 100 1000 10 000

coherent scattering (a)incoherent scattering (b)

photoelectirc absorption (c)nuclear pair production (d)

electronic pair production (e)total ( f )

photon energy E [keV]

cro

ss s

ectio

n [b

arn

/ato

m]

10-4

10-3

10-2

10-1

100

101

102

103

104

105

(a)

(b)(c)

(f)

(d)

(e)

1

10

100

10 100 1000

photon energy E [keV]

inte

ractio

n p

rob

ab

ility

[%

]

d = 0.5 mm

d = 1.0 mm

d = 2.0 mm

d = 3.0 mm

d = 5.0 mm

P.E. only

P.E. + I.S.

CdTe

= photo electric absorption

= incoherent scattering ("Compton")

P.E.

I.S.

50 500

5

50

PHYSICAL CONCEPTS


IMAGING = source location

most promising is a two fold approach: mask + Compton

coded mask

+ high spatial resolution

+ big field-of-view

~ about 50% active area

→ easy solution for low energies

Compton imaging

- low spatial resolution

+ nearly 4π field-of-view

+ 100% "active" area

→ extension to higher energies

x

y

z

x'

y'

z'

photon origin

LED interaction

HED interaction

α

β

source plane S

LED

HED

D < D0 D = D0 D > D0

PHYSICAL CONCEPTS


SPECTROSCOPY = source identification

source separation

flux estimation

what kind of sources are

in the field-of-view?

what is the flux of the

different sources?

which source is where?

→ imaging

are there multiple sources?

→ subtracting the

strongest source

E [keV]

400

300

200

100

0

0 10 20 30 40 50 60 70 80 90

Gd Kα1,2

Gd

Kβ 1,2,

3

W Kα 1

Au

Kα 2 +

W Kβ 1,

3

Bi Kα 1

+ A

u Kβ 1,

3

Bi Kβ 1,

3

Sn Kα 1,

2

Sn Kβ 1,

2,3

Mo

Kα 1,2

Au

Kβ 2

Mo

Kβ 1,2,

3

W Kα 2

Au

Kα 1

Bi Kα 2

Bi Kβ 2

Bi Lβ1,2

kcnts []

CURRENT STATUS


several test campaings gave us useful feedback and new ideas

Tacq = 0.4 s

test with coded mask (MURA) - 74 MBq Am-241 in 1 m (290 nS/h):

source identification within 400 ms

- 10 MBq Eu-152 in 1 m (1.2 μS/h):

source localization: 7° ang. res.

Eu-152 Am-241

CURRENT STATUS


several test campaings gave us useful feedback and new ideas

test with coded mask (MURA) - 74 MBq Am-241 in 1 m (290 nS/h):

source identification within 400 ms

- 10 MBq Eu-152 in 1 m (1.2 μS/h):

source localization: 7° ang. res.

Eu-152

CURRENT STATUS


WIX-HD → WIX-O

lots of improvements in electronics and mechanics:

- improved temperature sensing

- enhanced vacuum capacity

- pressure sensor added

- embedded HV generation

- embedded TEC power & control

- batteries with charger and power ctrl.

from space to industry standard:

- 16x16 pixels, Schottky CdTe

- same power: 200 mW (0.8 mW/ch)

- same energy range: 2 keV to 1 MeV

- 1cm² x 1 mm → 2 cm² x 2 mm

- 625 μm → 800 μm pixel pitch

CALISTE-HD → CALISTE-O

CALISTE-HD

CALISTE-O

CURRENT STATUS


CALISTE-O

first tests show that all

pixel are working

spectral resolution: - best pixel: 927 eV FWHM @ 60 keV

- all pixel: 1.3 keV FWHM @ 60 keV

5 10 15 20 25 30 35 40 45 50 55 60 650

counts [ ]

5

10

15

20

25

30

35

40

20170626 122229, CALISTE-O, Am-241, 500V, tp2, LT15, singles

0

energy [keV]

counts [103]

0 2 4 6 8 10 12 14

0

2

4

6

8

10

12

14

x

y

20170626 122229, Am-241, 500V, tp2, LT15, all

2000 4000 6000 80000

CURRENT STATUS


CALISTE-O aims to detect gamma rays

Cs-137 source spectral resolution: - all pixel: 6.7 keV FWHM @ 662 keV

480

188

36.6

32.1

8.8

20170630_144111 Cs-137 only singles

100 200 300 400 500 600E [keV]

10

1000

100

1000

0

cnts []spectral features: - Cu fluorescence (8.8 keV)

- Ba Ka and Kb lines (32.1 & 36.6 keV)

- Ba* after Cs decay (662 keV)

- Compton scattering in the detector

(E < 480 keV)

- Compton scattering outside

(E > 188 keV)

- multiple (>2x) Compton sc.

det

662 keV

det

662 keV

det

662 keV

662

NEXT STEPS


we are not at the end...

test Compton imaging

autonomous control

autonomous analysis

point and interval estimations in x-, y-, and z-directions

combining mask and Compton

HV cycling to prevent CdTe instability

pixel individual trigger thresholds

optimal peaking time

energy calibration

source identification

flux / dose estimation

APPLICATION: medical physics

SATBOT:

radiotherapy + nano particles


biology

accumulation of NP

degradation of NP

5 Rs:

physics

enhanced absorption

because of high Z

generation of

generation of

fluorescence radiation

enhanced attenuation

-> dose enhancement

Auger e-

photo e-

Compton e-

chemistry

oxidative stress by

reactive oxigen species

(ROS):

H2O2

e- + H20 OH -

O2-

surface effec:

catalysator

coating

chemical enhancement

multidisciplinary field

Au

θX-ray tube

detector

repair,

redistribution

reoxygeneration

repopulation

radiosensitivity

BASIC PHYSICS

XRF detection

self absorptoin in humain tissue


depth

in s

oft tis

sue [m

m]

atomic number Z []

1

10

100

0 10 20 30 40 50 60 70 80 90

0

10

20

30

40

50

60

70

80

90

100

Kα1

absorp

tion

[%

]

100% 90%

80%

70%

60%

50%

40%

30%

20%

10%

BASIC PHYSICS

interaction photons ↔ humain


10-5

10-4

10-3

10-2

10-1

100

101

102

103

104

1 10 100 1000 10000

abs. coeff. μ

[1/c

m]

energy [keV]

ICRU4 coherent scattering

ICRU4 incoherent scattering

ICRU4 photo electric absorption.

ICRU4 pair creation (nucleus)

ICRU4 pair creation (atom)

bones

breast

blood

fat

ICRU 4 components

BASIC PHYSICS

XRF dosimetry

goal:- determine the concentration of the NP

- determine the dose at the

tumor level


~6

9 keV

t 1

d1

d2

t 2

human

tissue with

att. coef. μ1

tumor

+NP

human

tissue with

att. coef. μ2

detector

X-ray tube

F0

FILTER

F1 F2 F2 F3

F4

F5

F5

~7

8 keV

approach:

- prior knowledge of the tube flux

- determine the transmittivity t1 and t2

- tumor flux = tube flux * t1 →dose

- NP concentration ~ meas. flux / tumor flux

BASIC PHYSICS


scattering makes it difficult...

coherent scattering:

for low energies

scattering angle > 90°

0

1

2

3

4

5

6

7

0 20 40 60 80 100 120 140 160

dσ

/(s

in(θ

) dθ)

[b

/sr/

ato

m ]

θ [°]

incoherent scattering

Si, k0 = 200 keV

Si, k0 = 150 keV

Si, k0 = 80 keV

10-7

10-6

10-5

10-4

0.001

0.01

0.1

1

10

100

1000

θ [°]

dσ

/(s

in(θ

) dθ)

[b

/sr/

ato

m ]

coherent scattering

Si, k0 = 50 keV

Si, k0 = 662 keV

Au, k0 = 50 keV

Au, k0 = 662 keV

0 20 40 60 80 100 120 140 160

small effect

incoherent scattering:

for all energies of interest

scattering angle ≈ 90°

big effect

SATBOT results


first SATBOT campaign FEB 2017:

industrial robots for

alignment of X-ray tube

alignment of detector

SATBOT results


first SATBOT campaign FEB 2017:

industrial robots for

alignment of X-ray tube

alignment of detector

result

Au identification

independent of NP size

sensitivity: 408 ug Au Au / Kα

Au / Kβ

Nano en solution:

• 31 nm

• 45 nM

• 50 µL

22/2/2017

W / Kα

/ KβW

SATBOT results


second SATBOT campaign APRIL 2017:

filter for X-ray tube

shape the tube spectrum

enhance the high energetic part

result

Au identification for different mAu

lin. relation between signal & mAu

sensitivity: 380 ug Au

50 55 60 65 70 75 80

40000

60000

80000

100000

ENERGY (keV)

Counts, normal i zed

blanc

0,56 mg Au

1,12 mg Au

2,25 mg Au

problem:

detect a peak on top of a peak!

SATBOT results


third SATBOT campaign AUG 2017:

filter for X-ray tube

advaced filter

result

XRF peak in valley

clear Ka and Kb

detection

sensitivity:

Ka: 50 ug Au

Kb: 300 ug Au

Au Kα2 + Kα1

fit: SKα = 0.0949 cnt/(s cm2 μg) · m

Au Kβ3 + Kβ1

fit: SKβ = 0.0133 cnt/(s cm2 μg) · m

Au mass m [μg]

0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 2400

Au s

igna

l S

[c

nt/s]

0

50

100

150

200

250

2310 μg Au

1150 μg Au

577 μg Au

288 μg Au

144 μg Au

0 μg Au

20 40 60 80 100 56 60 64 68 72 76 80 84

flu

x [1

/(s k

eV

cm

2 )

]

E [keV] E [keV]0

0

20

40

60

80

100

60

50

70

80

90

detection limit

50 300

SATBOT new ideas: advanced X-ray filter


EXRF < K-edge

only E > K-edge causes XRF

background photons EXRF < E < K-edge → noise

X-ray filter without absorption edges

energy

transm

itta

nce

1

0

X-ray filter near an absorption edge

energy

transm

itta

nce

1

0

ideal X-ray filter

energy

transm

itta

nce

1

0

thin filter

thick filter

K-edge

unnecessary photons E < EXRF → unnecessary dose



EXRF < K-edge

only E > K-edge causes XRF

background photons EXRF < E < K-edge → noise

energytransmittance

1

0

usable

photons

for XRF

usable

photons

for XRF

n > 0

energy

transmittance

1

0Kedge, filter

Kedge, targetXRF

usable

photons

for XRF

n ≤ 0

energy

transmittance

1

0Kedge, filter

Kedge, targetXRF

usable

photons

for XRF

n << 0

XRF Kedge, targetKedge, filter

unnecessary photons E < EXRF → unnecessary dose

Zfilter = Ztarget + n

Combine filter with incoherent scattering

X-ray detector

(scintillator)

for transmitted radiation

ortho X-ray tube

(140 kV peak)

test sample

WIX X-ray camera

(CdTe) for fluorescence

and scattered radiation

X-ray

beam

scattered radiation loses energyTARGET

F

I

L

T

E

R

θX-ray beam

det.



0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90

Z []

Z []

n<< 0

n ≤ 0

n > 0

example: for Au: n≪0 for Z≤73

n<0 for 74 ≤ Z ≤79

n>0 for 80 ≤ Z ≤ 84



45

50

55

60

65

70

75

80

85

90

95

100

105

110

115

65 70 75 80 85 90

en

erg

y [ke

V]

atomic number Z []

20°30°40°

50°

60°

70°

80°

90°

100°

110°120°130°140°150°160°180°

Kedge

Kα1

Compton sc. Kedge color

θ

n=

1 n=

2

n=

-1

n=

3 n=

4 n=

5

n=

0

n=

-2

n=

-3

n=

-4

n=

-5

Which filter and which observation angle should we choose?



How thick should the filter be?

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

10 100 1000 10 000

cum

ula

tive p

hoto

ele

ctr

ic d

istr

ibution []

10-5

10-4

10-3

10-2

10-1

1.0

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10 000

rela

tive

photo

ele

ctr

ic c

ross s

ection []

E [keV] E [keV]

50050 5000

signal analysis:

flux

after

filter

flux

before

filter




noise analysis:

0.001

0.01

0.1

1

0 20 40 60 80 100 120 140 160

flu

x F

[c

nt/(s

ke

V c

m2 ]

energy E [keV]

10-6

10-5

10-4

10-3

10-2

10-1

1.0

energy E [keV]

flu

x F

[c

nt/(s

ke

V c

m2 ]

0 20 40 60 80 100 120 140 160

flu

x F

[c

nt/(s

ke

V c

m2 ]

0 20 40 60 80 100 120 140 160

10-6

10-5

10-4

10-3

10-2

10-1

1.0

energy E [keV]

Em

ax =

15

0 k

eV

filtered flux F2(E), Au-filter: 150 µm, 300 µm, 600 µm filtered flux F2 res(E), Au-filter: 150 µm, 300 µm, 600 µmunfiltered flux F1(E)

0 100 200 300 400 500 600 700

0.001

0.01

0.1

1

flu

x F

[c

nt/(s

ke

V c

m2 ]

energy E [keV]

0 100 200 300 400 500 600 700

flu

x F

[c

nt/(s

ke

V c

m2 ]

energy E [keV]

10-6

10-5

10-4

10-3

10-2

10-1

1.0

energy E [keV]

flu

x F

[c

nt/(s

ke

V c

m2 ]

10-6

10-5

10-4

10-3

10-2

10-1

1.0

0 100 200 300 400 500 600 700

Em

ax =

64

0 k

eV

filtered flux F2(E), Au-filter: 150 µm, 300 µm, 600 µm filtered flux F2 res(E), Au-filter: 150 µm, 300 µm, 600 µmunfiltered flux F1(E)




maximize signal-to-noise ratio

0.00000

0.00025

0.00050

0.00100

0.00200

0.00400

0.01000

0.02000

0.04000

0.08000 0.1

1

10

100

0 0.2 0.4 0.6 0.8 1

SN

R []

filter thk [mm]

Gold (Z = 80; n = 0; θ = 130°)add a constant

background flux




a filter can make it worse

a higher voltage of the X-ray tube makes the SNR always better ;-)

the optimal filter thickness is independent of the chosen voltage

SATBOT conclusions


SATBOT is a very dynamic project

the work is very interdisciplinary

next steps:

new filter: sensitivity < 20 ug

XRF tomography

SATBOT conclusions


the SATBOT team is very nice

caliste and its applications

Documents