Measuring the cosmic microwave background

27 October 2010

Measuring the cosmic microwave background

Sofia Sivertsson

Outline

✤

Theory: why we want to measure the CMB.

✤

Early history and detection.

✤

How we measure it, COBE and WMAP.

What is the CMB?

✤ The plasma of the early universe was not transparent to photons.

✤ After 380 000 years the Universe had

cooled enough for electrons and protons to form atoms, making it transparent to light.

✤ This gave a cosmic background radiation from the surface of last scattering.

✤ The CMB has a black body spectrum that cooled with the expansion of the Universe from 2970 K to 2.725 K.

✤ Photons leaving over-dense regions will be redshifted, making the density fluctuations leave an imprint in the CMB as temperature fluctuations.

Mathematical treatment

✤ The temperature fluctuations on the sky can be expanded in spherical harmonics:

✤ Observations so far support that are independent Gaussian random variables (follows l-dependent Gaussian distribution).

δT (θ, φ) = �

lm

a_lmY_lm(θ, φ)

a

✤ The statistical properties of the

temperature anisotropies can then be expressed as a function of l only.

✤ For every l there are (2l+1) independent giving a statistical uncertainty for small l, called the cosmic variance.

a

Information in the CMB

✤ The fraction of (and need for) dark matter.

✤ The baryon to photon ratio.

✤ Expansion and curvature of the Universe. Especially when combined with supernova and baryon acoustic oscillation data.

✤ Constrains the mass of neutrinos.

✤ Primordial power spectrum after inflation.

✤ Potential future exotics: gravitational waves and primordial non-gaussianities.

✤ This is what we know now, but let’s start from the beginning.

CMB history

✤ In 1946 Gamow introduced a hot big bang to account for the creation of the elements, it was then realized that this would give a temperature today of around 5 K.

✤ In 1941 absorption lines from first excited rotational state of interstellar CN molecules were observed. This led to a background radiation of temperature 2.3 K.

✤ No one cared.

✤ In 1964 Penzias and Wilson at Bell Telephone

Laboratories measured a 3.5 K background signal (for the second time).

✤ At the same time Dicke and Princeton colleagues were reinventing the hot big bang (in a bouncing universe) and designing a detector to measure the CMB.

Importance

✤

Measurements at more frequencies made it more and more clear that the CMB truly had a black body spectrum.

✤

Making it more difficult to explain in a steady state universe (for example by starlight).

✤

Next important step: to measure the CMB anisotropies, i.e. how the temperature varies over the sky.

✤

Requires much higher precision in the measurements.

Anisotropy difficulties

✤

The CMB anisotropies requires very precise measurements.

✤

The atmosphere radiates thermally, especially water vapor content varies in space and time which gives non-uniform atmospheric emission.

✤

Thermal radiation from the Earth sneaks in.

✤

Needs to go above the atmosphere!

✤

Can not get high enough precision by measuring the absolute temperature.

✤

Differential measurements: measuring the temperature difference in

different directions.

CMB dipole

✤

First reliable detection of dipole anisotropy in 1978 from a balloon borne experiment.

✤

Dipole due to our motion towards the ”great attractor”.

✤

Also the dipole anisotropy small:

✤

Differential measurements limits systematic errors.

✤

Galactic foreground.

✤

Sun, Moon and Earth radiates thermally.

T

/T

= 1.2 × 10

Satellite experiments

✤

No atmospheric emission, cryogenic cooling without windows, longer observation times and more freedom in choosing observation wavelengths.

✤

Relikt experiment on the Soviet Prognoz 9 satellite, launched 1 July 1983. Measured galactic flux, CMB dipole (l=1) and quadrupole (l=2).

✤

Highly eccentric orbit.

✤

Will discuss experimental techniques in the context of COBE, talk

some about WMAP and mention Planck.

COsmic Background Explorer (COBE)

✤

Orbit around Earth.

✤

Thermal shield for Sun, Earth & Moon.

✤

Measured dipole moment to:

and quadrupole moment to:

✤

Will discuss the FIRAS and DMR instruments in more detail.

✤

DIRBE (Diffuse Infrared Background Experiment) measures the cosmic infrared background, wavelength 1.25 to 240 microns. Not primarily a CMB measurement.

T

/T

= (5 − 6) × 10

T

/T

= 1.2 × 10

Heterodyne receivers

✤

Signal is mixed with a constant frequency signal, a low pass filter then picks out the source signal close to

✤

Need significant amplification, hard to avoid gain variations.

✤

Dicke switching: Rapidly switch between the antenna and a fixed temperature source,

optimally with temperature close to the antenna temperature of the source.

✤

Measures the temperature difference.

Mixer

Filter

Amplifier Oscillator

Source

Output

ν

ν

|ν − ν

|

ν

COBE Differential Microwave Radiometers (DMR)

✤

Heterodyne detector.

✤

Measures the temperature difference on the sky 60˚ apart (7˚resolution).

✤

The satellite rotates with 0.8 rpm, reducing instrumental error.

✤

4 year observation time, covering full sky every 6 months.

✤

Substantial data processing required to

generate temperature anisotropy map.

Galactic background

✤

CMB anisotropy and galactic

foreground emission at high galactic latitudes (WMAP observational

bands).

✤

COBE DMR units observed at the

frequencies: 31.5, 53 and 90 GHz

(31.3 - 31.8, 51.4 - 54.25 and 86 - 92

GHz protected for radio astronomy).

Bolometric detectors

✤

In heterodyne detectors the errors rise with frequency, use bolometers at frequencies above ~100 GHz (3 mm wavelength).

✤

Measures the energy of the incident radiation through the temperature increase.

✤

Detects radiation at all wavelengths, so filters are needed.

✤

Any material tend to emit radiation in the same frequency as it absorbs. Need to keep the closest filters cold.

✤

Need cryogenic temperatures.

Fourier transform spectrometer

✤

Michelson interferometer with a movable mirror.

✤

Scanning with the movable mirror

gives an interference pattern that is the Fourier transform of the source.

✤

Allows wide spectral range with good resolution.

✤

Can get distortions from non-linearity in moving the reflector or

time variation of the total flux.

COBE Far Infrared Absolute Spectrophotometer (FIRAS)

✤ A scanning Fourier interferometer.

✤ Output is proportional to the Fourier transform of the inputs difference in spectral power.

✤ Wavelength range about 0.1 mm to 10 mm.

✤ The reference temperature can be adjusted to obtain a nearly null output interferogram.

✤ Bolometric detector.

✤ The entire instrument is cryogenically cooled.

COBE Cryogenic system

✤

COBE carried 660 liter liquid helium at launch.

✤

The slow helium evaporation provided a stable 1.4 K environment for the 306 days the helium lasted.

✤

This ended the FIRAS experiment but

the DMR continued taking data.

Wilkinson Microwave Anisotropy Probe (WMAP)

✤

Two back-to-back off-axis Gregorian telescopes.

✤

Measures the anisotropies, i.e. the temperature difference.

✤

Similar to the COBE DMR instrument.

✤

WMAP gives 45 times better sensitivity and 33 time angular resolution (13 arcmin FWHM) of the COBE DMR instrument.

✤

Does not orbit the Earth but the second

Lagrange point.

WMAP

✤ Signals divided into orthogonal polarizations.

✤ Signal differenced against orthogonal polarization of the other side.

✤ Signals to be differentiated are

amplified by both amplification chain to cancel gain variations

✤ The phase switches gives a 180˚ phase shift between the two signal paths, interchanging which signal is fed to which detector, reducing systematic effects.

✤ WMAP has 10 of these devices (1 @ 22 GHz, 1 @ 30, 2 @ 60 and 4 @ 90 GHz). Chosen to allow off-shelf components where possible.

✤ WMAP’s orbit not around the Earth

Lagrange point L2

✤ Points which have the same orbital period as the Earth (i.e. stationary)

✤ Only L4 and L5 are stable

✤ L1: Solar telescopes, L3: ”Hidden planet”, L4 & L5:

Trojan asteroids

✤ L2 is 1.5 million km away (= 0.01 au = 235 Earth radii)

✤ Orbit around L2 need reoccurring orbit corrections.

✤ Sun and Earth always in the same direction, satellite outside Earth’s magnetic field.

WMAP scanning the sky

✤

WMAP telescope rotates fast around its own axis combined with a slow precession.

✤

Covers the whole sky as it moves with the Earth around the Sun.

✤

WMAP launched in 2001 and sent

into a graveyard orbit in October

2010.

Results

✤

WMAP 5 year results:

✤

COBE results:

More results

✤ Higher l measured by the ACBAR (Arcminute Cosmology Bolometer Array Receiver) and BOOMERANG (Balloon Observations Of Millimetric Extragalactic Radiation ANd Geophysics).

✤ Both located at the South Pole.

✤ Enough to measure a small fraction of the sky.

The future: Planck satellite

✤

Launched successfully by ESA on

May 14, 2009 together with Herschel.

✤

Stationed at L2.

✤

Started its first all sky survey on August 13, 2009.

✤

CMB and galactic emission sky maps will be released in 2012.

✤

Much more to say about Planck,

some other time...

References

✤ Books:

R. B. Partridge ”3K: The Cosmic Microwave Background Radiation”, Cambridge Astrophysics Series, 1995.

E. W. Kolb and M. S. Turner ”The Early Universe”, Addison-Wesley Publishing Company

B. F. Burke and F. Graham-Smith ”An Introduction to Radio Astronomy”, Cambridge University Press, Second Edition.

✤ Papers:

Smoot et. al. ”COBE Differential Microwave Radiometers: Instrument Design and Implementation”, ApJ 360: 685-695, 1990 Bennett et. al. ”The Microwave Anisotropy Probe Mission”, ApJ 583: 1-23, 2003

Rubakov and Vlasov ”What do we learn from the CMB observations?”, arXiv:1008.1704 (2010).

Wilkinson Microwave Anisotropy Probe (WMAP): Five-Year Explanatory Supplement, editor M. Limon, et. al. (Greenbelt, MD: NASA/GSFC). Available in electronic form at http://lambda.gsfc.nasa.gov

✤ Internet:

http://map.gsfc.nasa.gov/mission/observatory_l2.html http://lambda.gsfc.nasa.gov/product/relikt/

Image from: http://map.gsfc.nasa.gov/mission/observatory_freq.html

http://scienceworld.wolfram.com/physics/FourierTransformSpectrometer.html http://aether.lbl.gov/www/projects/cobe/cobe_pics.html

http://news.discovery.com/space/mission-complete-wmap-fires-its-thrusters-for-the-last-time.html http://www.rssd.esa.int/index.php?project=PLANCK&page=dev_news

http://cosmology.berkeley.edu/group/swlh/acbar/