Friday, March 21, 2014

Gravitational Waves Seen in the Polarization of Light From the Big Bang


The oldest light we can see in the Universe is called the cosmic microwave background (CMB) and it is the relic light from the Big Bang.  While this light is old, it isn't quite as old as our Universe.  Before an event called recombination, the Universe was not transparent to light, so the light couldn't propagate very far before being absorbed.  Recombination happened about 380,000 years after the Big Bang and the light from this time is what we observe in the CMB.

Everywhere we look on the sky, the frequency of this microwave light is very nearly the same.  Since heat can be transmitted through radiation (such as microwaves), we can characterize this light to have a temperature of about 3 K (or about 3 oC or 5.4 oF above absolute zero - the coldest anything in the Universe can be).  Why this temperature is the same everywhere on the sky doesn't immediately make sense since the heat hasn't had enough time to be transferred across the Universe.

The slight variations in the CMB temperature from opposite sides of the sky as measured by 9 years of data from the WMAP mission.  The fluctuation in the CMB temperature is measured to be ± 0.0002 oC (0.00036 oF).  [Source: Wikipedia]

The CMB is almost 14 billion light years away from us.  This is approximately the age of the Universe.  But there is no way for light to transfer heat from one side of the Universe (14 billion light years from us) and reach the other side (14 billion light years away from us in the opposite direction) since this would take about 28 billion years of travel time or twice the age of the Universe!  So there is no way that these widely separated parts of the Universe should have the same temperature if the Universe has expanded in a continuous way since the Big Bang.

To explain why the CMB is essentially in thermal equilibrium in every part of the Universe, something extraordinary needed to happen...


Almost immediately after the Big Bang, it is believed that the Universe entered a period of extremely rapid expansion called inflation.  This began at about 10-35 seconds after the Big Bang and the Universe proceeded to expand its volume by about 80 orders of magnitude (that's a 1 followed by 80 zeroes) in a fraction of a second.  During this time, gravitational waves would have originally been produced on a quantum mechanical scale and then blown up to cosmological scales during inflation.  The gravitational waves from the Big Bang are exactly these fluctuations in space-time that are still vibrating from the period of inflation.  (The wavelength now is its original wavelength, i.e. about 1% of the size of the Universe as it was then, stretched by the amount the Universe has expanded since then.)


Since gravitational waves were able to propagate through the early Universe long before light was, it is expected that there is evidence of these gravitational waves contained within the CMB.  We expect to see this in a special kind of polarization of the CMB (where polarization refers to the rotational orientation of the light waves).  There should be 2 kinds of polarization in the CMB, E-mode and B-mode.

A graphical history of the Universe showing when gravitational waves would have been created and how they affect matter along with density waves and their affect.  The effects that gravitational waves have on mattert cause B-mode polarization in the CMB while density waves are the primary contributors of E-mode polarization.  [Source: Wikipedia]

E-mode polarization means that the orientation of the polarization should not change as you move in a straight line.  B-mode polarization means that the rotation of the polarization changes or "curls" around itself.  The E and B in these mode names refer to how electric (E)  and magnetic (B) fields behave: a single charge will have an electric field pointing radially away from a single change while a magnet always have 2 poles causing the magnetic field to always curl back to the magnet.  The E-mode polarization in the CMB provides information about the fluctuation of density in the early Universe.  Because gravitational waves alternate, compressing space in one direction and expanding it in the orthogonal (at a right angle) direction, they caused the "curling" B-mode polarization.

Graphical illustration of the polarization patterns for E-modes and B-modes.  Note that B-mode patterns can be characterized by "rotating" clockwise or counter-clockwise while the E-modes cannot.  [Source: Press conference screen grab]

An experiment called BICEP2 (Background Imaging of Cosmic Extragalactic Polarization) announced to the world a few days before this post (17 March 2014) that they did indeed detect the B-mode polarization in the CMB.  The results are cataloged here.

The above image is the polarization of different points in the sky they observed from the South Pole.  The red colored areas are where the B-modes can be classified as clockwise and the blue colored areas are where they can be classified as counter-clockwise.


Any time scientists think they found something that they wanted to find, we immediately set to trying to disprove what we found.  (This is discussed on this blog in regard to LIGO with the blind injections known as "The Big Dog".)  After thorough vetting and analysis of this work, it has been determined that the chance of this B-mode signal has a chance of 1 in 3.5 million of being a false detection. 


This discovery of the imprint of gravitational waves on the CMB further hints at the promise that gravitational-wave astronomy with detectors like LIGO will have in the future.  Their discovery in no way diminishes the potential of LIGO and gravitational-wave astronomy - instead it increases its promise.

LIGO seeks to work like a gravitational-wave radio and record the gravitational-wave signals directly.  (This analogy is discussed in more depth on this blog here.)  For this analogy, the information about what made the gravitational wave is the music being carried on the radio wave (the gravitational wave in this analogy).  In this sense, LIGO will be making a distinctly different kind of detection than BICEP2 did.  We will be directly recording a gravitational wave as it passes by Earth and BICEP2 detected the imprint of gravitational waves on the CMB.

Also, LIGO looks for a wider range of gravitational waves.  While we also look for the relic gravitational waves from the Big Bang which we call stochastic gravitational waves, we search for three other kinds: continuous, inspiral, and burst.  (These are described in more detail on this blog here.)  This broad range of gravitational waves that detectors like LIGO will be able to "see" will allow gravitational waves to tell their own story of how they were made; perhaps from the collapse of a star into a black hole or the merging of two stars into one, or the echoes of the birth of the Universe.  We will not be seeing the evidence of gravitational waves that is imprinted onto light, but collecting information from the gravitational waves themselves.

As a side note:  Kip Thorne, a physicist who has pioneered work in general relativity and gravitational waves, made a prediction in 2006 of what detections will be made with gravitational waves in the next 50 years:
"Over the next 50 years, gravitational waves from the big bang will be detected, first indirectly by the imprint they leave on the cosmic microwave radiation and then directly, by space-based gravitational wave observatories."
 You can read the rest of his prediction on

Read LIGO's official congratulatory statement on the BICEP2 results to the web page.


The BICEP2 results do much more than suggest or support that inflation happened: it gives us some information about what happened during inflation.  The strength of the signals observed here informs us on the energy involved in inflation.  The ratio of the strength of the E-modes to the B-modes (a value referred to as r and measured here to be r = 0.2) is proportional to the energy density of the Universe at the time of inflation and this is consistent with energies needed in some of the grand unified theories (GUTs) (this is where the strong, weak, and electromagnetic forces become indistinguishable).

The BICEP2 results also serve to constrain the theories of what happened during inflation.  Several of these have been ruled out (e.g. large field inflation models are now highly unlikely).

Ultimately, these results need to be reproduced and refined by coming experiments.  This doesn't mean that the scientific community isn't confident in BICEP2's results, but science needs to be reproducible.  And in reproducing results, they are often refined and expanded upon.

This truly is an exciting time to be a scientist!

See Also:


  1. Thanks for making the BICEP2 announcement this past Monday more understandable! The background info (i.e. recombination and visuals to describe E/B-mode polarization) you provided was very valuable to a lay person like myself, though I will still need to read and reread your blog to allow all this info to sink in. I totally appreciate your including the Kip Thorne predictions (in 2006!) regarding future Gravitational Waves detections. Very impressed that he was right on with the CMB (indirectly), hopefully earth-based gravitational-wave observatories (LIGO, VIRGO, GEO, etc...) will make direct detections before the space-based observatories ones as plans for those seem to be put on hold for the moment!

    1. Thank you for your kind encouragement! It really does mean to world to me. And it took me many reads through the published results, watching the press conference more than once, and a day and a half (in between my LIGO work and teaching at university) to write this single post. This isn't what I specialize in so I want to make sure that what I share with you is as correct as possible.

      Like I said in the post, this is a wonderful time to be a scientist! There are so many new things being learned from experiments other than what I specialize in that I feel that I am forever a student keeping up with new discoveries and how they relate to my work.

      I never want to stop learning - that's when life get boring!

  2. Awesome write up Amber! I am really excited for all of you working so hard on this project!

    1. Thank you! While I wish I had something to do with BICEP2, it does make looking for gravitational waves more directly with LIGO a even more exciting. We had evidence that they were there before, but this puts a little spring in your step as far as getting back to the nitty-gritty of LIGO.

  3. Thanks for this, Amber! That was the perfect level of detail for my brain to process the news.

    1. I'm glad this was understandable! I never know for sure when I write a post if my explanations will make sense to anyone other than myself.

      And I'm even more happy that I haven't gotten any constructive/irate emails from experts in this field telling me how off the mark my explanations are!