CMB Polarization, QUIET, and an Inside View of Gravity Wave Searches in the CMB B' Winstein, U of Ch

1
CMB Polarization, QUIET, and an Inside View of
Gravity Wave Searches in the CMBB. Winstein, U
of Chicago

• What weve learned from the CMB
• What we can learn from its Polarization
• The QUIET Experiment
• Perspectives Reflections
• ARNPS articles Kamionkowski and Kosowski, 1999
  
• Samtleben, Staggs,
  BW, 2007
• Text book Modern Cosmology, Dodelson

My Goal impart the flavor, excitement, and
relevance to EPP of the field
2
The CMB

• Electrons and protons form first atoms Universe
  becomes transparent
• Photons from last scattering surface provide
  snapshot of infant universe
• Still present but cooled (shifted to microwaves)
  from the expansion

Z1000
3
(No Transcript)
4
Why a Black Body?

• Thomson scattering dominates just before
  recombination
• Changes photon directions but not energies
• Hence cannot establish thermal equilibrium
• Non-Planckian injection thermalized via
  Bremstrahhlung and Compton scattering
• e.g. ee- annihilation at z 109
• Arbitrary spectrum at z ? 106 is thermalized
• Roughly 2 months after the big bang (KeV scale)

5
Dark Matter ImplicationsSuperwimps (J. Feng et
al.)

• NLSP freezes out, with right relic density
• Later it decays to gravitinos
• Between BBN and decoupling
• Natural lifetimes 104-108 sec
• CMB rules out about half the parameter space

Phys.Rev.D68063504, 2003
6
WMAP MAP of the CMB Anisotropies
7
TT Power Spectrum Data with 6 Parameter Model
8
Gaussian Fluctuations(Monteserin et al.,
astro-ph 0706.4289)
wmap
1000 simulations
9
Parameter Estimation Results(from acoustic
oscillations between photons, electrons,
protons, and DM)
Confirms Dark Energy, Dark Matter, Baryon
density
Major Discovery reionization of the Universe-
important for Polarization
Major Hope reveal conflicts with Standard Model
10
CMB Collider DM Constraints
EPP2010
11
Just after decoupling photons have last
scattered and are starting to free stream
We are here
12
Particle Horizon
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
CMB Polarization

• Created by scattering
• Of quadrupole
• Really the cessation of scattering
• Measured in ?K
• Power difference in 2 orthogonal directions
• Expect 2 different spatial patterns (later)

22
Radiometer Sensitivity

Where does the radiometer equation come from?
Pulse counting!
23
(No Transcript)
24
8 sec of data 100kHz sampling
Temperature vs. time, 10 ms bins
unswitched
switched
25
Experimental Considerations

• Amplifier Drifts
• Electrical Grounding
• Mechanical pickup
• Optics/ground pickup
• Thermal regulation

• CMB is 2 x 10-13 Watts
• Need gain of 108
• ?K signals are a few nV

26
Temp anisotropy peaks at 75 uK
Expected Polarization Anisotropy
27
CAPMAP
Princeton, Chicago, Miami, JPL Collaboration
Crawford Hill, NJ 16 Correlation
Polarimeters 12 W-Band (84-100 GHz) 4
Q-Band (35-45 GHz)
28
Results
29
Results
11 ? detection best in the field (but surpassed
in 4 months)
30
Allowed E-mode power spectrum (MCMC results)
31
E Polarization Modes B
B only from Gravity Waves
E from Density Perturbations
32
B-mode Science (I)

• Late-Universe Gravitational Lensing
• Sure to exist, at scales around 0.1 degrees
• Factor of 30 sensitivity boost needed
• (only to see the effect)
• Sensitive to neutrino masses
• Early Universe metric perturbations
• Depends on inflationary model
• Sensitive to the energy scale of inflation
• At least another factor of 10 improvement needed
• If inflation occurs at the GUT scale
• Parameterized by T/S tensor to scaler ratio
  for perturbations

33
Lensing Simulation (W. Hu et al.)
34
Lensing BB Power Spectrum vs. Neutrino Mass
35
B-mode Science (II)

• Early Universe metric perturbations
• Depends on inflationary model
• Sensitive to the energy scale of inflation
• At least another factor of 10 improvement needed
• If inflation occurs at the GUT scale
• Parameterized by T/S tensor to scalar ratio
  for perturbations

36
(No Transcript)
37
Caldwell, Kamionkowski, Wadley, astro-ph/9807319
38
Optimism for Gravity Waves ?(Pagano et al.,
astro-ph 0707.2560)
WMAP 1(2??? bounds
39
V 1/4 3.3 x 1016 (T/S)1/4GeV
40
Simulated sky with T/S0.210 deg by 10 deg field
41
Simulated sky with T/S0.010 deg by 10 deg field
42
Simulated sky with T/S0.2B-modes only
500 Times Lower Power
43
Other obviously impossible experiments

• 10-21 strain measurements from G. Waves
• Measuring difference in particle and
  anti-particle branching rations to 10-7
• Measuring a 10-21 change/year in

44
Detectors for Polarization

• Bolometers (3 exp with polarization results)
• Incoherent
• Best sensitivity in space
• SPT, ACT, Polarbear, SPIDER, Planck, .
• HEMT Amplifiers (4 exp with polarization results)
• Coherent
• Some systematic advantages
• CAPMAP, QUIET, WMAP, DASI
• QUIET/Polarbear alliance is unique

45
Future CMB experiments(D. Samtleben)
46
Q/U Imaging ExperimenT Collaboration
Manchester Oxford
Chicago (KICP)
Oslo
MPIfR-Bonn
Stanford (KIPAC)
CAPMAP Observational Site
KEK
Caltech JPL
Columbia Princeton
Miami
Atacama Observational Site Chile (CBI
site)
5 countries, 12 institutes, 30 people
47
Radiometer on a Chip
Q-Band (44 GHz)
W-Band (90 GHz)
1 inch
48
QUIETs Radiometer on a Chip

• Only ground-based effort using coherent
  detectors
• Measuring Q/U simultaneously in each pixel
• Complementing frequencies from other experiments

1 inch

• Automated assembly
• and optimization
• Large array of
• correlation polarimeters

49
High Speed Sampling18 bits _at_ 800 kHz

• Q/U measurement every 250 ?s
• Monitors high-frequency noise
• Permits Quadrature Samples
• TOD noise with no signal

0.5ms
50
W-band Receiver Integration
51
Q-band Receiver
52
QUIET Array Polarization Sensitivities

• Q band lab
• Extrapolate to Chile
• 60 all 19 modules
• Same on sky (10)

• W band lab
• Extrapolate to Chile
• 149 19 modules, 8/21
• 127 26 modules, 9/12
• 89 53 modules, 11/9
• 70 84 modules, Dec. 15

Using optimization scheme (adjusts 10 bias values
for 7 modules, finding optimum setting for
polarization in 3 hrs)
53
Telescope / Optics
Receiver
Focal Plane (Receiver)
Secondary Mirror

• 1.4m primary mirror
• Resolution in FWHM
• 13 arcmin (W-band)
• 28 arcmin (Q-band)
• Inherit CBI mount

Electronics Box
Primary Mirror
Mount
CBI mount
40cm
Mirrors Support Structure
54
(No Transcript)
55
Observed Patches
Map precision on 1x1 degree pixel Planck
1 ?K (100 GHz) QUIET 10-1 ?K (90 GHz)
56
Scanning the Sky
57
Calibration

• Polarized laboratory sources
• We are doing 2 level measurements
• Assuming linearity
• In the field
• Planets
• pointing
• beam widths
• Polarized point sources
• The Moon
• Also polarized

58
FNAL Calibrator (H. Nguyen)

• Likely Saturation in lab
• FNAL 20K calibrator
• Should reveal compression
• Will later be polarized

59
Sky Coverage, Noise Filter
A. Kusaka
CMB Signal
Observed Sky
After Filtering
Detector Noise
Huge contamination by detector 1/f noise
1/f noise is removed by high-pass filtering
CMB power extraction
60
CMB Power, QUIET Sensitivity
A. Kusaka
E-mode power
Observed Sky
Auto Correlation
B-mode power

• QUIET sensitivity
• (10 months, 50 duty)
• 2? B-mode signal for r0.3
• Precise measurement of E-mode

r0.3 assumed
61
Data transfer from site
Observation at Chile
KEK, Japan (Mirror)
Important calibration, Digest (Internet)
2GB/day
Oslo, Norway (Mirror)
U Chicago (Primary)
Full data (DVD, snail) 10GB/day
U.S. Institutes
62
Data being looked at
Polarization
Time Stream (Moon)
Total power
?19 pix
Total Power
Receiver Response
Polarization
Time (sec)

• Receiver is working!
• Optics is OK
• Reasonable beam size
• Rough pointing understood

Jupiter observed by ?T module
63
Systematic/Foreground Studies
64
QUIET Summary

• Wasnt sure this would work
• Proposed just 100 detectors
• Good prospects for understanding and improving
• 19 element Q band array now taking data
• Array sensitivity now same as BICEP (state of the
  art today)
• Will operate till end of 2009
• 91 element W band array will be shipped in late
  February
• Phase II proposal Sept. 09
• Phase II 2011-15
• Unique technology
• Unique foreground lever arm
• Alliance with Polarbear
• Hope FNAL will play a role

65
Telescopes
1000 detectors Should reach r0.02
66
(No Transcript)
67
Some Observations about the CMB Field

• Receivers often deployed prematurely
• Not yet really serious about systematics
• New approaches (analysis) with every experiment
• No common platforms
• Funding/Management Issues
• Investigators on multiple projects
• No one body looking after the field, deciding on
  approvals
• Secrecy
• Little information until final results
• Proposals are not public
• C.f. FNAL
• Excitement of the Science compensates .

68
Concluding Thoughts

• Polarization is an exciting frontier
• May probe GUT scale physics (e.g. p decay)
• But similarly no real natural target scale
• Eventual reconstruction of V(?) (?)
• Very challenging experiments!
• Need to detect GWB at multiple frequencies and
  from space and the ground

69
EXTRAS
70
WMAP Vs Planck (TT)
71
WMAP vs Planck (EE)
72
Expected resultsfor phase I (phase II)
EE spectrum
BB spectrum (r0.2)
50 (500) W-band receivers with 340 mK sqrt(s)
per element 10 (20) months observing with 50
efficiency reaching r10-2 in phase
II
73
Satellite Vs. Ground-based

• Satellite
• All sky maps
• Access to lowest multipoles
• But foregrounds formidable!!
• Any frequency possible
• Sensitivity diluted
• Flight-ready equipment needed
• Expensive

• Ground-based
• Larger mirrors possible higher multipoles
• Atmosphere gives frequency restrictions
• Better S/N
• Can frequently swap out equipment
• Relatively inexpensive

74
Some idealism ..
Fractional error on power for a signal with
?ll, Optimized scanning.
Conclude factor of 200 needed, 240 designed.
75
Some realism ..

• Were not meeting goals
• Nobody meets their goals!
• excepting WMAP
• Particle Physics
• ? was about 0.5
• later came close to 1.0
• Or even higher!
• No one remembers
• science is so exciting

• Systematic errors
• based on experience
• determined insitu
• Foregrounds
• claims for clean patches
• need insitu study

76
Getting to Interesting Levels of T/S

• Experimental noise vs. multipole
• Ground-based best for lgt50
• Frequency restrictions
• Satellites for lower ls
• Foregrounds!

rT/S
rT/S
0.1
0.1
0.03
0.03
0.01
0.01
77
Can we get to interesting levels of T/S ?
White noise level of the different experiments in
comparison to the CMB power spectrum (using most
sensitive CMB frequency) Statistical uncertainty
only, No impact of foregrounds/systematics taken
into account! (but only one frequency used
from each experiment, others can be used for
foreground removal)
r0.3
r0.05
r0.01
V 1/4 3.3 x 1016 (T/S)1/4 GeV
Better S/N for the ground-based efforts (compared
to Planck)
78
Observation regions
Preliminary choice of 4x400 square degree
patches - low foreground regions
(overlap with Quad/EBEX/Polarbear planned)
- elevation above 40 degrees - distribution
to allow continuous scanning
K-band (23 GHz) WMAP 5 years, polarization
intensity
Map precision on 1x1 degree pixel Planck
1 mK (100 GHz) QUIET 10-1 mK (90 GHz)
CAPMAP field
79
Observing from the Ground Atmospheric opacity
O2
H2O
Brightness at zenith at 5 km height
CMB Intensity
80
Building up the Multipoles (by Clem Pryke)
81
(No Transcript)
82
(No Transcript)
83
(No Transcript)
84
(No Transcript)
85
(No Transcript)
86
Gravitational Lensing
87
Thermal Spectra

• Planck
• Bose-Einstein
• ???is the chemical potential)

Thermal equilibrium
Kinetic equilibrium- No processes to
change Photon number
88
Detectors for Polarization

• Bolometers
• Incoherent
• Best space sensitivity

• HEMT Amplifiers
• Coherent
• Systematic advantages

• Teams
• Chicago/Berkeley SPT
• Chicago-Columbia/JPL QUIET
• Berkeley Polarbear
• CIT-Chicago/JPL QuAD, BICEP, SPIDER, Planck
• Princeton/Goddard-NIST ACT

89
How good a BB?

• Residuals lt 50 ppm
• Limits on distortions (95)
• y lt 15 x 10-6 (spectral shift)
• ?? lt 9 x 10-5
• Planck spectrum becomes Bose-Einstein with energy
  release between 105 lt z lt 3x106
• Energy release with z lt 105

90
Scanning Pattern
A. Kusaka
91
Calibration/Characterization (lab)

• Polarized gainsPolarization sensitivity
• I to Q/U leakages

92
QUIET Calibration/Optimization
??????mK (Al)
93
Module Calibration Sensitivity(using QUIET
Optimizer)
94
Polarized Gain/Sensitivity
Ti
4 diodes from one module
SS
Al
95
Foregrounds

• Multi-frequencies critical
• Studying natural band differences among detectors
• Immanuel Buder 334 project
• Signal is 0.02 ?K2 (r0.1)
• WMAP FG0.09 ?K2
• All sky, l100, ?90 GHz
• Careful selection of patches to scan
• Alliance with POLARBEAR
• Will observe from the same location
• Will choose patches in concert
• Provides 45, 90, 150, 225 GHz

96
BB limits
lensing
97
Chicago Quiet Group

• 3 graduate students
• 1 fellow
• Long term visitor from KEK
• Undergraduates
• KICP support

W-band integration, Electronics, Quick
evaluation software, Offline pipeline management
98
Future CMB experiments(D. Samtleben)
SZ
SZ
X SZ
SZ
Balloon X Funding pending
Very few low frequency experiments underway!
Beware sqrt(2)s Plotted NEQ with NETNEQ
(Q(Tx-Ty)/2)
These are goals/current plans BUT some
experiments are not yet (completely) funded and
future technologies are not yet fully
established, dont take the numbers too
literally! Also SYSTEMATICS/FOREGROUNDS not
taken into account in these numbers
99
L
R
90 GHz Module Automatic Assembly
Simultaneous Q/U detections
Q
-Q
U
-U
100
(No Transcript)
101
QUIET L/R Correlator Simultaneous Q/U
measurements
102
(No Transcript)
103
QUIET Schedule

• Q receiver (19 elements)
• Now taking data
• W receiver (91 elements)
• Shipping to site in February
• Phase II 2011

104
Acoustic Oscillations

• Standing density waves on the sky
• Dark matter, protons, electrons, photons
• Gravitational contraction begins when entering
  horizon
• Resisted reversed by photon pressure
• Relativistic fluid of photons coupled to
  electrons, via Thomson scattering
• Electrons coupled to protons (em)
• Oscillations frozen at era of recombination

105
A Primer on (single-field) Inflation