Module 7: The cosmic microwave background · Lesson 7.2 · about 15 min
The most perfect blackbody: COBE
In 1990 the COBE satellite showed that the cosmic microwave background is a blackbody at 2.7255 K, matching theory to better than 50 parts per million. Why that perfect curve proves a hot, dense past, and why expansion keeps it perfect.
Module 7 · about 15 minutes. Read first: 7.1 Penzias and Wilson hear the Big Bang, and the astronomy course’s Blackbody radiation for what a blackbody curve is. Here we ask what the CMB’s spectrum tells us about the universe. By the end you will be able to say why its perfect blackbody shape is evidence for a hot, dense past, and work out what it looked like at any redshift.
1. The curve that got a standing ovation
Penzias and Wilson measured the glow at one wavelength. To be sure it was the afterglow of a hot universe, you need the whole spectrum: the brightness at every frequency. A hot, dense fog in equilibrium should give one particular shape, the blackbody (Planck) curve, which depends on nothing but the temperature. Earth’s air glows and absorbs at exactly the wavelengths near the peak, so the measurement had to go to space.
NASA’s COBE (Cosmic Background Explorer) was launched on 18 November 1989. Its FIRAS instrument, led by John Mather, compared the sky with a built-in blackbody of known temperature. In January 1990, after only nine minutes of sky data, Mather showed the spectrum at a meeting of the American Astronomical Society: the points lay exactly on a blackbody curve. The audience gave it a standing ovation.
The final FIRAS result: the CMB matches a blackbody to better than 50 parts per million of its peak brightness. Its temperature today, combining FIRAS with other measurements, is
From that one number follow the rest: it is brightest per unit frequency at 160.2 GHz, its Wien peak wavelength is 1.06 mm, and there are about 411 CMB photons in every cubic centimetre of space, more than a billion for every atom.
2. Why a perfect blackbody means a hot, dense past
A blackbody is what you get when light and matter have exchanged energy back and forth so many times that they share one temperature: thermal equilibrium. Think of a pot of soup simmered for hours. Every spoonful is the same temperature; you could not tell where any bit of heat came from.
Today’s universe could never do that. It is almost empty and transparent: CMB photons have crossed it for 13.8 billion years without hitting anything. Any mix of sources, such as stars, galaxies and warm dust at different temperatures, would add up to a lumpy, non-blackbody spectrum. To get a perfect one you need the whole universe to have been dense and opaque, a fog where light could not travel far before being absorbed or scattered. Before the first few months, processes that make and destroy photons wiped out any distortion. Later, until recombination, scattering off electrons kept sharing energy around. FIRAS found no trace of distortion at all, which also rules out any big injection of energy since those early months.
3. Why expansion keeps it a blackbody
The CMB has not been in contact with matter for 13.8 billion years, yet it is still a blackbody. This is a small miracle of geometry. As space stretches by a factor 1 + z, two things happen:
- ν
- the frequency of any one photon: stretched by the expansion, so it falls by 1 + z
- n
- the number of photons per cubic centimetre: the same photons spread through (1 + z)³ more volume
- Physical meaning
- A blackbody's photons are spread over frequency in a shape set by hν/kT. Lower every frequency and every temperature by the same factor and hν/kT is unchanged; thin the photons by the cube of that factor and the count at every frequency is exactly what a blackbody at T/(1 + z) needs. So a stretched blackbody is still a blackbody: T = T0(1 + z) holds for its whole spectrum, not just a single wavelength.
So the light that set off at about 3,000 K, from a fog as hot as the surface of a cool red star, arrives today as a 2.7 K microwave glow with exactly the same shape. Slide the figure to z = 1,089 to see it.
A rocket, then a satellite. In 1988 a rocket-borne experiment reported extra brightness at short wavelengths, beyond the blackbody curve, and theorists rushed to explain it. Two years later FIRAS showed that it was not there. COBE’s other instrument, DMR, found the tiny ripples in the CMB in 1992 (Lesson 7.4).
- 1974 NASA invites proposals; Mather's team proposes the mission that becomes COBE
- 1989 COBE launched on 18 November
- 1990 FIRAS spectrum shown in January: a perfect blackbody, T = 2.735 ± 0.06 K
- 1996 Full FIRAS data: deviations below 50 parts per million
- 2006 John Mather (FIRAS) and George Smoot (DMR) share the Nobel Prize in Physics
- 2009 Fixsen combines measurements: T0 = 2.7255 K
“The CMB peaks at 1.06 mm, so it peaks at c ÷ 1.06 mm = 283 GHz.” Wien’s wavelength peak (1.06 mm) and the frequency peak (160.2 GHz, which is c ÷ 1.87 mm) are both right: plotting the same light per unit wavelength or per unit frequency gives curves that peak in different places. What is the same in both is the temperature, 2.7255 K.
- COBE FIRAS (1990, final 1996) showed the CMB is a blackbody to better than 50 parts per million: T0 = 2.7255 K, frequency peak 160.2 GHz, Wien peak 1.06 mm, about 411 photons per cm³.
- A perfect blackbody needs thermal equilibrium, which only a hot, dense, opaque early universe could provide.
- Expansion lowers every photon’s frequency by 1 + z and the photon density by (1 + z)³: the result is exactly a cooler blackbody, T = T0(1 + z).
- Nobel Prize 2006 to John Mather and George Smoot.
Check yourself
Sources
- J. C. Mather et al. (1990), A preliminary measurement of the cosmic microwave background spectrum by the Cosmic Background Explorer (COBE) satellite, ApJ 354, L37 https://doi.org/10.1086/185717
- D. J. Fixsen et al. (1996), The cosmic microwave background spectrum from the full COBE FIRAS data set, ApJ 473, 576 https://doi.org/10.1086/178173
- D. J. Fixsen (2009), The temperature of the cosmic microwave background, ApJ 707, 916 https://doi.org/10.1088/0004-637X/707/2/916
- J. C. Mather (2007), Nobel lecture: From the Big Bang to the Nobel Prize and beyond, Reviews of Modern Physics 79, 1331 https://doi.org/10.1103/RevModPhys.79.1331
- NASA Goddard, COBE: the Cosmic Background Explorer https://lambda.gsfc.nasa.gov/product/cobe/
Your ticks stay in this browser. Save them as a file on the course home.
Updated 2026-11-22. Numbers are rounded for teaching; measured values come from the sources listed in each lesson. Simulations run in your browser; nothing you type is sent anywhere.