Detectors for BLISS

1
Detectors for BLISS

• Jamie Bock (JPL)
• Kent Irwin (NIST)
• Erick Young (Steward/UA)
• with contributions from
• Rick LeDuc (JPL)
• Peter Day (JPL)
• Chao-Lin Kuo (JPL)

2

• Bolometers
• Photoconductors
• New Technologies

3
Sensitivity Requirements for FIR Spectroscopy

• Background at R 1000
• exceeds state-of-the art NEP
• FIR spectrometers are large
• Extreme stray-light sensitivity

4 K instrument AW 3 mm2, Dl/l 50 , e 10
4
Bolometers Principle of Operation
Infrared radiation
Fundamental noise limit NEP (4kT2G)1/2
T0
Speed of response t f(a) (C/G)
Heat sink
T
NEP2photon gt NEP2phonon NEP2Johnson NEP2amp
Thermal link G(T)
Transition-Edge Superconducting (TES) Bolometers
Electro-Thermal Feedback a dln(R)/dln(T)
1000 vs. 5 for NTD Ge Linear Response P Q
k ?TcTo G(T)dT P electrical power Q
optical power Response speed-up f(a) ? (1/a)
10-2 Saturation max(Q) ?TcTo G(T)dT
fsQ0 Sensitivity NEPbolo (8kT0fsQ0)1/2
scales with fs1/2 ( weakly with b) NEPphoton
(2hnQ0)1/2
5
Tradeoffs with Bolometers
Advantages Wavelength coverage High quantum
efficiency (typically gt 85 QE) Well
understood noise no photometric funnies
NEP ? 1e-19 W/?Hz possible (1e-18 W/?Hz
demonstrated) SQUID multiplexing
Demonstrations of SQUIDs, NTD Ge bolometers
mature

• Disadvantages
• - Low operating temperature
• (50 100 mK for BLISS)
• - Systems issues
• (EMI, B-fields, T stability, stray light)
• Dynamic range trades with sensitivity
• (this is not a strong constraint)
• - Low technology readiness of ultra-low
• background TES SQUID readout,
• plus an active cooler at system level

6
Spider Web Si3N4 NTD Ge Bolometer Arrays

• Characteristics
• Far-IR to mm-wave 30 mm lt l lt 3 mm
• High sensitivity NEP ? 1.5e-18
• Excellent noise stability 1/f lt 30 mHz
• High optical efficiency h ? 85
• 100s of detectors (SPIRE)
  330
• Low temperature T ? 300 mK

Electrical Lead
Support Leg
Absorber
NTD Germanium
7
SPIRE Bolometer Detector Arrays (1)
Feedhorn array
IR filter
300 mK attach
Kevlar suspension
2 K flange
Detector array (not visible)
Kapton Cu/Ni cables
Output connector
Exploded view
8
SPIRE Bolometer Detector Arrays (2)
300 mK assembly
Focal plane assembly
Kevlar suspension
9
SPIRE Bolometer Detector Arrays (3)
Focal Plane Array (l250 mm, 144-Elements, 99
yield)
10
Bolometer Array Performance Summary
Perfect bolometer is ideal model _at_ 81
end-to-end yield, goal/guidelines are demanding
11
Low-Power JFET Readouts

• Differential source-follower readout
• Si3N4 isolation ? JFETs operate at T 130 K
• lt 7 nV/?Hz noise at lt 200 mW per channel
• Dissipation dominated by Si3N4 conductance
• Perforated membranes show 2x lower G
• Can be placed on 20 K stage

Developed for Herschel/SPIRE at JPL
JFET Module
30-Channel Membrane
12
Ultra-Low NEP TES Bolometers (1)
FT Spectroscopy
G T2.78
Dispersive Spectroscopy
G T1.24
Phonon limited NEP2x10-20W Hz-1/2 _at_20mK !!!
Data P.Day, Fabrication R.LeDuc, T. Jun
13
Ultra-Low NEP TES Bolometers (2)
Lower G under development How does G may
scale? maybe as exp(-l)!
Data P.Day, Fabrication R.LeDuc, T. Jun
14
Testing of MoAu TES Films
1 square
1/2 square
1/5 square
o, o, o bias point
---, ---, --- Phonon-limited NEP

• Need low-impedance recipe for 100 mK 300 mK
• Need to understand noise behavior

15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
Noise Testing with SQUID MUX
Results from First integration
muxed
muxed differenced
Systematic Study of Mux Noise
unmuxed
TES Bolometer
21
(No Transcript)
22
2 x 5120-element arrays
23

• Bolometers
• Photoconductors
• New Technologies

24
(No Transcript)
25
(No Transcript)
26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
(No Transcript)
32
(No Transcript)
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36

• Bolometers
• Photoconductors
• New Technologies

37
(No Transcript)
38
(No Transcript)
39
Kinetic Inductance Detectors

• Bandgap detector
• Elegantly multiplexed readout
• uses room-temperature GHz electronics!

Peter Day, Rick LeDuc / JPL Jonas Zmuidzinas /
Caltech
40
(No Transcript)
41
Conclusions

• TES Bolometers
• NEP e-18 W/?Hz demonstrated
• NEP e-19 W/?Hz within reach at 50 mK
• NEP e-20 W/?Hz requires lower G
• Covers the whole wavelength range
• High QE, excellent photometric behavior
• SQUID multiplexing
• Systems Issues
• Photoconductors
• MIPS 32 x 32, NEP 1e-18 W/Hz
• AsSi SbSi arrays working well on Spitzer
• Implementation easier than bolometers
• Lower NEP? Requires lower amplifier noise
• Cannot cover long wavelengths
• Do we use more than one detector technology?
• New technologies - depends on SPICA schedule
• GaAs, Ge IBC, KIDS, NIS Coolers

42
Spectral Transmission of Silicon
No measureable loss in silicon Room temperature
sample, 1cm thick
Data courtesy of Peter Ade, U. Cardiff
43
Sensitivity Trade with Dynamic Range
NEP2tot NEP2photon NEP2bolo NEPbolo
(8kT0fsQ0)1/2 NEPphoton (2hnQ0)1/2
An Example Qsat/Q0 fs G0 selectable b
2 T0 100 mK Tc/T0 2 No amplifier noise
NEPbolo NEPphoton
Generally not too bad! More of an issue at long
l Lower T helps
44
(No Transcript)
45
(No Transcript)