Large Arrays for MM Waves

1
Large Arrays for MM Waves

• Paul L. Richards
• Department of Physics
• U. C. Berekeley
• Rencontres de Moriond
• March 2006

2
Talk Outline

• Berkeley group
• Bolometric detectors
• Horn coupled TES arrays
• APEXSZ and SPT
• Antenna coupled polarization sensitive TES arrays
• Polar Bear
• Frequency domain output multiplexer
• Other detectors and multiplexers

3
Fall 2005
4
Continuous Revolution in Bolometer Sensitivity

• From NEP 10-9 to 10-18 WHz-1/2 in 40 years
• Speed of measurement doubled every year
• Now at photon noise limit
• Next step is large format arrays
• Output multiplexing

5
5 mm
1 cm
10 cm
10 cm
1980s
1990s
2000s
6
Horn Coupled Arraysof Spiderweb TES Bolometers

• SZ Cluster Searches with
• APEXSZ 2006
• And SPT 2007 - 8

7
dN/dz for 4000 deg2 Survey
SZ as a probe of Structure formation
Springel, White, Hernquist astro-ph/0008133
1 deg2 of sky _at_ 150 GHz
With sufficient resolution and sensitivity, it
may be possible to place interesting constraints
on w, the DE equation of state
8
Experimental Cluster Yields
Name Type Beam (arcmin) Cluster Yield
ACBAR Bolo 4 Few
Bolocam Bolo 1 10s
SZA HEMT 1 100s
CBI HEMT 4 100s
AMI HEMT 1 100s
Amiba HEMT 1 100s
APEX Bolo 0.75 1000
SPT Bolo 1 20,000
Planck Bolo 5 10,000
ALMA HEMT -- --
9
TES Bolometer Array DevelopmentAPEX-SZ, SPT, EBEX

• Lithographic techniques allow large arrays of
  TESs on single wafer
• Prototype 55-element arrays on single 4 inch
  wafers have been fabricated

TES
10
APEX-SZ Single WedgeSix Wedges for 330 Detectors
11
APEX-SZ Horn plate and Feed Optics
APEX-SZ UCB, MPIfR, LBNL(Chile)
12
APEX-SZ Receiver Layout
120 cm
Pulse tube cooler
Simon-Chase 4He/3He/3He sorption fridge
Silicon Lens
SQUID readout modules (4 K)
Lyot stop (4 K) ? 10 cm
Foam vacuum window ? 24 cm
Horn coupled TES bolometer array (250 mK)
13
APEX Telescope

• 12 meter on-axis Cassegrain
• At 16,500 ft elevation on Atacama plateau
• 0.4? maximum field of view
• 40 resolution at 150 GHz
• 18 ?m surface accuracy goal
• APEX-SZ assigned 5 months over two years

14
APEX Beam Maps
Apex Beam Maps
15

• South Pole Telescope
• Sub-mm quality 10m off-axis telescope.
• Large stationary ground
• shield
• 1 deg FOV _at_ 2mm
• Initially dedicated to SZ

16
1000 Element TES Frequency Domain Multiplexed
Bolometer Array with cooled Secondary Mirror
To primary
Cold Secondary mirror
17
Antenna-Coupled Arrays

• Polar Bear
• Ground based CMB Polarization Anisotropy
  Experiment
• CMBPOL
• Space Mission

18
Science Motivation

• CMB Polarization gt signal from Inflation
• Confirm Inflation is correct model
• Test models (scalar potentials)
• First probe of physics at 1016 GeV GUT scale
• Large discovery potential
• Necessarily beyond the current Standard Models of
  physics cosmology
• CMB polarization lensing
• Probe of Dark Energy at early time
• Complementary to SNIa
• Measure sum of neutrino masses

19
POLARBEAR Sensitivity
Statistical Uncertainty (no lensing)

• Large arrays 3 years integration
• gt 105 effective integration time compared to
  MAXIMA
• Conservative estimate using achieved sensitivity

20
CMB Data Set Size
Samples Pixels
COBE (1989) 2x109 6x103
Boomerang (1998) 3x108 5x105
WMAP (2001) 7x1010 4x107
Planck (2007) 5x1011
6x108
Polar Bear (2007) 8x1012 6x106
CMBPol (2020) 1x1014 1x109
21
Radiometer on Chip
filters
superconducting transmission lines
antenna
bolometers
22
Monolithic Array with Lenslets

• Well developed for e.g. SIS mixer work
• High antenna gain
• Forward radiation pattern
• Efficient coupling to telescope
• 80-90 achieved Similar to scalar horn

23
Antenna Coupled Arrays
3 cm
5 mm

• 4 wafers 330 bolos
• Single-color, dual-pol at 90, 150, 220 GHz

24
POLARBEAR (LBNL/UCB, UCSD, Colorado, McGill)

• Deployment in California, then Chile
• Unique 3-layer ground shield
• Antenna coupled TES detectors being tested
• Cryostat construction started

3.5m Telescope Design
25
POLARBEAR (LBNL/UCB, UCSD, Colorado, McGill)

• Cryostat Construction Started

26
Frequency Domain Multiplexer
27
Motivation for Multiplexed Readout

• Multiplexers enable breakthrough in array size
• Multiplexers
• Reduce number of wires
• Reduce heat load to detector temp and 4K
• Reduce complexity e.g. connectors/bonds
• Reduce cold active devices
• Reduce power dissipation at detector temp and/or
  4K
• Reduce cost
• Reduce warm electronics
• Reduce power dissipation
• Reduce cost
• Different applications require different features
• Multiplexers part of scalable array technology

28
Readout System
Initial Work NSF Development NASA,
DOE Exporting Technology (U. Chicago, CWRU, U.
Minn)
29
Eight-Channel Demonstration

• Sensor noise white above 200mHz

30
Future MUX Work

• Observational Test of MUX System on APEXSZ
• Increase MUX Factor 8 gt 32
• Digital Room Temperature Electronics
• Cold Amplifier
• Dramatic Increase in Number of MUXed Pixels

31
Digital MUX

• bias comb synthesized digitally with FPGA at 25
  MHz

• sky-signal modulated comb is digitized directly
  at 25MHz
• demodulation and filtering occurs digitally in
  FPGA

Second prototype board being Fabricated now

• foundation for eventual migration to radiation
  hard ASIC technology.

32
SQUID Amplifier linearized by cold feedback

• COLD FEEDBACK LOOP
• Motivation
• Reduce wire length to lt 5cm
• Closed loop bandwidth gt 10 MHz
• Increase fMUX channel count

33
Berkeley FDM Experiments
SZ Galaxy Clusters
APEX-SZ 320 Channels 2005, Chile
South Pole Telescope 1000 Channels 2007
CMB-POL
EBEX 1000s Channels LD Balloon (2008)
POLARBEAR 300-1000 Channels (2007?)
34
Coordinated Development of Detector Arrays for
CMBPol
35
Requirements for CMBPOL

• Accurate Polarimetry
• Control of Systematic Errors
• 30 Times as many Detectors as Planck
• -Background-limited Arrays with N gt1000 pixels
• Polarization-Specific Array Architectures
• Mature Readout Multiplexer
• Wide 30-300 GHz Frequency Coverage to Evaluate
  Foregrounds
• -Mature Polarization Modulator
• System Level Tests
• -Demonstrated in Real Sub-orbital Experiments

36
Coordinated Development of Bolometric Array
Polarimeters for CMBPOL
Caltech/JPL Antenna Coupled
UCB/LBL Lens Antenna coupled
GSFC Horn Antenna Coupled
NIST TES SQUIDs
Frequency Domain MUX
Time Domain MUX
CMBPol Demo 2011

• This consortium can make space flight hardware
• Explores 3 basic coupling and 2 basic MUXing
  schemes
• Structured to arrive at CMBPol Demo by 2011

37
Goddard Space Flight Center
H. Moseley, T. Stevenson, A. Kogut, E. Wollack

• Beyond Einstein Inflation Probe requires
• large arrays of sensitive detectors
• control over systematic errors at -80 dB

• GSFC research and capabilities
• advanced light coupling schemes
• superconducting planar microwave circuits
• large superconducting detector arrays
• on-chip modulation
• laboratory and balloon flight system tests

38
Caltecha and Jet Propulsion Labb
J.J. Bockb,a, A.E. Langea, R. LeDucb, J.
Zmuidzinasa
16x16 planar array antenna
25 BW filter
8 x 8 TES Polarimeter Array Prototype for 150 GHz

• Planar antenna defines beam
• Stripline filter defines passband
• S/C switch modulates polarization
• Small detector volumes
• SQUIDs mount behind array
• Tiles to large focal planes

Polarization switch
TES bolometer
39
U.C. Berkeleya and LBNLb
W. Holzapfela, A.T. Leea,b, P.L. Richardsa, H.G.
Spielerb
Systems! APEX-SZ, SPT
Dual-Pol Antenna-Coupled Bolometer (tested!
Applied Physics Letters 2005)
Frequency Domain SQUID MUX
40
NIST
K.D. Irwin, W.D.Duncan, G.C. Hilton, J.N. Ullom

1,280-pixel TES bolometer array
1,280-pixel time-domain SQUID MUX
Hybridized 1,280-pixel SCUBA-2 array
SQUID capabilities

• TES capabilities
• Reliable, uniform full-wafer TES fab
• Engineered control of noise
• Extensive cryogenic characterization facilities

32-SQUID chip for time-domain MUX
100-SQUID series array for frequency-domain MUX
Time-domain MUX room-temperature electronics
41
Summary

• TES Bolometer Arrays Maturing
• APEX SZ Becoming Operational
• SPT Receiver Under Construction
• Polar Bear Under Construction
• F-MUX Working Well with
• Second Generation Development
• Coordinated Development for CMBPOL

42
Voltage-Biased Transition-Edge Sensor TES
Superconducting Bolometer

• Ptotal Popt Pbias

Pbias Vbias2/R
43
Berkeley TES Bolometer Experience

• 1996, A.T. Lee et al., APL
• Proof of principle Hot electron TES Bolometer _at_
  100 mK
• 1998, S.F. Lee et al., Applied Optics
• Suspended TES bolometer
• 1999, Gildemeister et al., APL
• Spiderweb TES bolometer
• 2000, Gildemeister et al., APL
• Monolithic filled array
• 2000, Yoon et al., APL
• Frequency domain SQUID multiplexer
• 2005, Myers et al., APL
• Antenna Coupled Bolometer with Bandpass Filter
• 2005, Lanting et al., APL
• 8-ch TES Multiplexer

44
Observing Strategy
Single pixel noise

• Drift scan, single pixel noise dominated by
  atmospheric fluctuations
• Slewing or chopping increases signal frequency,
  reducing atmospheric noise
• Common mode rejection using multiple pixels may
  also be possible

45
Getting at 1016 GeV
Inflation Window 1 (primordial power, e-folds,
exit) CMB LSS (redshift survey, weak
lensing, Ly? forest) Inflation Window 2 (energy
scale, extra physics) CMB Polarization
(gravitational wave induced)
160 ?K ? 8 ?K ? lt0.3 ?K
46
CMB and Neutrino Mass
That same polarization lensing gives a window on
absolute neutrino mass and possibly hierarchy.
On those same small scales, neutrino free
streaming suppresses matter growth. Suppression
depends on mfp m?-1 and the time when ? goes
non-relativistic. So can test for inverted mass
hierarchy. Directly related to DOE calls for ?
CD-0.
47
Antenna w/Contacting Lens

• Great flexibility in antenna design
• Dual-polarizations
• Broadband for multichroic pixels
• Antenna is small compared to pixel
• Use area under lens for filter, switches, mux
• Challenges Lens and anti-reflection coating
  require care

48
Microwave Circuit Characterizaton

• Test Chip gt Quantitative Test of Circuit
  Parameters

49
Future Antenna Coupled Development

• Characterization of Microwave Circuits
• 300 Bolometer Array
• Frequency RF Channelizer gt multicolor pixels
• Wideband Planar Antennas
• Polarization Modulators based on MEMs switches

ROSES proposal accepted this year
50
Multiplexer Techniques

• TES/SQUID
• Time-Domain
• NIST, GSFC
• Jena/MPIfR
• Frequency-domain
• Berkeley/LBNL (LLNL)
• ISAS
• SRON/VTT
• Moscow-RAS
• NIST (microwave FDM)

• KID/HEMT
• CIT/JPL
• STJ/HEMT
• RF-SET
• Yale
• STJ/RF-STJ
• Yale
• NIS/HEMT
• NIST
• NTD/HEMT/FET
• CNRS(CRTBT/LPN)

(A viable sensor is muxable)
51
Frequency Domain MUX Features

• Zero power dissipation at array stage
• Only passive components
• MUX inductors/capacitors can be integrated in
  focal plane
• Intrinsically ac biased
• Not dependent on low-f noise of SQUID
• Insensitive to microphonics
• Challenges
• Reduce power dissipation at 300K
• gt All digital circuit w/FPGAs (EBEX) gt ASICs
• Remove cold wire length limit
• gt cold SQUID feedback

52
Funded and Future Experiments
53
Conclusions

• Berkeley/LBNL Group
• Significant contributions over last decade
• Will contribute strongly to technology
  development for CMBPOL
• Capable of building flight hardware given
  LBNL/SSL
• Collaborate with consortium and ROSES groups
• NASA Support Important for NASA-specific Goals
• Coordination with DOE, NSF

54
CMBPol is a unique probe of Inflation(see Weiss
Report)
CMBPOL 1000 detectors 1.0 deg. Beam
CMBPOL 2000 detectors 0.1 deg. Beam
2 lines span foreground uncertainty
55
Technical Requirements

• Background-limited Arrays with N gt1000 pixels
• Polarization-Specific Array Architectures
• Mature Readout Multiplexer
• Wide 30-300 GHz Frequency Coverage to Evaluate
  Foregrounds
• Mature Polarization Modulator
• Development -gt System Level Test
• Demonstrated in Real Sub-orbital Experiments

56
Crosstalk

• Measurement upper limit lt1, Theory 10-3

57
Digital MUX

• Primary power consumer of system is analog
  components for modulator/demodulator board.
• New Digital MUX system being developed in
  collaboration with McGill (M. Dobbs)
• much lower power consumption (factor 10),
  amenable for Balloon flights and satellites.
• baseline system for EBEX CMB Polarization Balloon
• electronics scale with number of multiplexer
  modules instead of number of bolometers
• able to scale to much higher channel counts

58
Cosmology with SZ Surveys
M. White
Holder et al. 2000
APEX-SZ PLANK
59
Simulation
M. White
After applying 10mK noise and filtering Recover
25 clusters with S/Ngt4 in 1 square degree
60
Horn Coupled Detectors The South Pole Telescope
Array

• SPT 918 detectors, six 4wedges
• Wedges at different frequecies 90, 150, 220, 270
  GHz.

3.5 inches
61
South-Pole Telescope
U. Chicago, CfA, CWRU, U.C. Berkeley
Horns
Bolometer array

• 8m, off-axis
• Sub-mm capable

20 cm
Cutaway of 1000 element array
62
Berkeley Bolometric CMB Measurements

• Balloon CMB Spectrum
• Woody, Mather, Richards
• Nagoya Berkeley Rocket Spectrum
• Matsumoto, Lange Richards
• MAX Balloon Anisotropy
• Lange, Richards Lubin
• MAXIMA Balloon Anisotropy
• Richards, Lee, Hanany, Lange, De Bernardis
• MAXIPOL Balloon Polarization Anisotropy
• Hanany, Lee, Richards, Tran
• APEX and SPT Cluster Searches with SZ effect
• Lee, Holzapfel, Richards, Halverson
• Polar Bear Polarization Anisotropy

63
The APEX-SZ Focal Plane
64
2-SQUID Array LISA

• COLD FEEDBACK LOOP
• Performance
• Demonstrated at 10 MHz
• Preparing test with TES bolometers

100-element series array
Niobium film
65
Paul Richards SIS Mixer, stressed
photoconductor, 4He bolometer, CMB Spectrum, MAX,
MAXIMA, MAXIPOL, APEX-SZ, PolarBeaR
Helmuth Spieler High-Energy Physics, Vertex
Detectors, High-resistivity CCD, Multiplexed TES
readout, APEX-SZ, SPT, PolarBeaR
Adrian Lee TES Bolometers, MAXIMA, MAXIPOL,
APEX-SZ (US-PI), SPT, PolarBeaR (PI)
Bill Holzapfel Bolometers, SUZiE, BIMA, CBI,
DASI, ACBAR, APEX-SZ, SPT, BICEP, PolarBeaR
66
POLARBEAR (LBNL/UCB, UCSD, Colorado, McGill)

• Ground-based 3 meter telescope in Chile
• Optimized for CMB polarization
• Characterize E-modes
• Search for B-modes
• Novel technology
• Ready for construction
• RD under LBL LDRD funding (2001-2003)
• Positive review from SAGENAP
• Most ambitious experiment proposed yet
• Covers most of theory parameter space
• Complementary to future Space Mission

67
APEX SZ First LightUCB, MPIfR, LBNL
68
Dark Energy Parameters
Slope of w
POLARBEAR
w