Gamma-ray microcalorimeter arrays for nuclearmaterials analysis
J. Ullom, NIST Boulder
NIST Boulder:
James BeallRandy DorieseWilliam DuncanLisa FerreiraGene HiltonRob HoranskyKent IrwinBen MatesNathan MillerGalen O’NeilCarl Reintsema
Dan Schmidt
Leila ValeYizi Xu
Denver UniversityBarry Zink
Los Alamos National LaboratoryMike RabinAndrew HooverSteve LamontCliff RudyDuc Vo
OutlineNIST Boulder:
1. Early single pixel results
2. Applications
3. Scaling to arrays
4. Pixel analysis
5. Higher Z absorbers
6. Spectrometer construction
�-ray microcalorimeters
1 mm
160
120
80
40
0
103.30103.25103.20103.15103.10103.05
42 eV FWHM
140
120
100
80
60
40
20
0
Counts
/ 5
eV
bin
1041021009896
Energy (keV)
LANL/NIST Calorimeter #1Gd-153 source with steel attenuator
97 keV and 103 keV gamma-rays
Sn absorber
Mo/Cu thermometer
first detector: spring 2005
record result:42 eV FWHM at 103 keV
current nondestructive analysismethods rely on knowledge ofreactor operating history
ucals may provide Pu measurementwhen operating history is lost orobscured
about 96 % (>600 tons) of the Pu under international safeguards is inspent fuel
Pu mass is a measure of reactor purpose: power vs weapons
Applications: measurement of Pu in spent reactor fuel
fuel in N. Korean cooling pond
Nondestructive assay for plutonium mass in reactor fuel
Spent Fuel XRF Spectrum 12889004
2000
2500
3000
3500
4000
4500
5000
5500
6000
6500
170 180 190 200 210 220 230 240 250 260
Channel
Co
un
ts
U K�1
98.43 keV
U K�2
95.65 keV
U K�3110.42 keV
U K�1111.30 keV
U K�2
114.44 keVPu K�1
103.73 keV
Eu-155
105.31 keV
100
2
3
4
5
6
7
8
91000
2
Counts
/ 2
0 e
V b
in
10610410210098
Energy (keV)
155Eu �-ray
Pu K�1 x-ray
U K�1 x-ray
HPGe spectrum of spent fuel from FSU(courtesy LANL)
Simulation of same spectrum with µcal
• ratio of U/Pu x-ray fluorescence measure of Pu mass• resolution of µcal minimizes interference from background, 155Eu
A 1,000 pixel �-ray spectrometer is feasible
existence proof: 1,280 pixel SCUBA 2
sub-array (for submm)
75 mm
area
count rate
energy resolution
quantum efficiency
1,000 pixels � 1 mm2/pixel
3.2 cm x 3.2 cm
1000 pixels � 100 Hz/pixel
100 kHz
< 42 eV at 100 keV
> 20% at 100 keV
gamma-ray arrays
NIST-LANL �-ray arrays
testbed with all readout circuitry andwiring for 128 pixels
routine fabrication of �-ray arrays
highly parallel micromachiningtechniques
6.25 mm
6.25 mm
mux chip
interface
chips
gamma-ray
arrays
Multiplexed array testing
zoomed view of 128 pixel testbed
6.25 mm
- interface chips for L/R filtering- 4x 32:1 multiplexer chips
muxed spectrum of 153Gd �-ray source
600
500
400
300
200
100
0
counts/5eVbin
103.4103.2103.0
energy (keV)
50.6 eV FWHM at 103.18 keV13 coadded pixels
spread inresolutions
energy resolutionhistogram
13 pixels total
4
3
2
1
0
pix
els
/ 5
eV
bin
120100806040200
energy resolution at 103 keV (eV)
chip 1 chip 2
Homogeneity
new performance record:25 eV FWHM at 103 keV
160
140
120
100
80
60
40
20
0
103102101100999897
energy (keV)
140
120
100
80
60
40
20
0
103.3103.2103.1
More on homogeneity
24
22
20
18
16
14
opera
ting r
esis
tance (
%R
no
rma
l)
43210-1-2
deviation from average Tc (mK)
chip 1 chip 2
why spread in �E ?
common bias and spread in Tc
“ “ complex transition shapes
energy resolution at 103 keV
pix
els
per
5eV
bin
bias chosen for more homogenous �E
14 pixels total
analysis to appear in Appl. Phys. Lett. (B. Zink et al, 2006)
Detailed analysis of device behavior
complex Z reveals absorber and(probably) frame
fitting noise requiresassumption that PN at ends oflink to absorber is uncorrelated
Higher efficiency possible with higher Z absorbers
SnTaMo
82
75
42
73
50
Z
20%250195Sn
94387258Ta
25
100
88
Efficiency100keV
43.8105Pb
1460430Re
1890460Mo
d (µm)�D (K)Material
some insulators andsemiconductors alsointeresting
But … complex behavior in some materials
thermal Decay
athermal
decay
Sn
Time Constant =
1.5 and 12.1 ms
Ta
Time Constant =
1.5 and 80-100 ms
Both Sn and Ta have second, athermal decay
? Ta has anomalous high heat capacity ?
: 6 keV x-ray
: 100 keV �-ray
How can we eliminate athermal decay in Sn ?
mechanical working leads to
smaller grains
cold rolled from supplier
has large grains
200 µm200 µm
• can control (and measure) Sn grain size
• will measure dependence of athermal
decay on grain size
• eliminating athermal decay will allow higher
count rates
!
NIST-designed cryogen-free refrigeratorassembled and tested• push-button operation• 13 hrs to cool from 300K• 7.5 days at 100 mK• < 1 hr to recycle
now available commercially fromHigh Precision Devices, Inc.
Spectrometer under construction for LANL
1 m
60
40
20
0C
ounts
per
8eV
bin
103.30103.25103.20103.15103.10103.05
Energy (keV)
cooler on: 45 eV FWHM at 103.18 keV
cooler off: 48 eV
330 mm
undegraded operation of gamma-ray ucalin electrical environment of pulse tube
TES operation in electrical environment of cryocooler
Conclusions
• there are applications for ucal arrays operating at ~100 keV
- Pu isotopes, spent fuel, HEU, …
• we have developed array-compatible techniques for fabricating �-ray ucals
• under muxed operation
- 50 eV at 103 keV in 13 coadded pixels
- best pixel = 25 eV
• work to understand absorber physics underway
• full spectrometer under development with Los Alamos
- cryogen-free dewar complete
- undegraded ucal operation
- testbed with wiring for 128 pixels complete
Ancient history (LTD11) : TES resolving power vs energy
5
6
7
89
103
2
3
4
5
6
7
89
104
reso
lvin
g p
ow
er
(E/�E
FW
HM
)
103
2 3 4 5 6 7 8 9
104
2 3 4 5 6 7 8 9
105
energy (eV)
STJ
Si thermistormag cal
TES, previous work
TES, this work
?
?
?
? LTD-12
Tc = 100 mK, M = 0
Tc = 100 mK, M = 1
TES resolving power vs energy
5
6
7
89
103
2
3
4
5
6
7
89
104
reso
lvin
g p
ow
er
(E/�E
FW
HM
)
103
2 3 4 5 6 7 8 9
104
2 3 4 5 6 7 8 9
105
energy (eV)
STJ
Si thermistormag cal
TES, previous work
TES, this work
??
??
??
?? LTD-12
Tc = 100 mK, M = 0
Tc = 100 mK, M = 1
Conclusions
• there are applications for ucal arrays operating at ~100 keV
- Pu isotopes, spent fuel, HEU, …
• we have developed array-compatible techniques for fabricating �-ray ucals
• under muxed operation
- 50 eV at 103 keV in 13 coadded pixels
- best pixel = 25 eV
• work to understand absorber physics underway
• full spectrometer under development with Los Alamos
- cryogen-free dewar complete
- undegraded ucal operation
- testbed with wiring for 128 pixels complete