Home / Columns / Gravitational waves: a new era for astronomy
Paul Ramond modifiée
π Space π Science and technology

Gravitational waves: a new era for astronomy

Paul Ramond, PhD student at Observatoire de Paris and ENSTA Paris (IP Paris)

The first ever direct obser­va­tion of grav­i­ta­tion­al waves in 2015 by the LIGO Sci­en­tif­ic Col­lab­o­ra­tion in the US is undoubt­ed­ly one of the biggest sci­en­tif­ic dis­cov­er­ies of the last decade, or even this cen­tu­ry. Six years lat­er, what can we say about these waves, and why is it so impor­tant to study them?

Grav­i­ta­tion­al waves (GWs), first pre­dict­ed to exist by Albert Ein­stein in 1916, allow for a brand-new way of look­ing at the uni­verse. Before their detec­tion, astronomers could only observe the sky using vis­i­ble light, and oth­er types of elec­tro­mag­net­ic radi­a­tion (includ­ing infrared, ultra­vi­o­let and gam­ma rays). 

While light is the prop­a­ga­tion of elec­tro­mag­net­ic fields vibrat­ing in space and time, GWs are com­plete­ly dif­fer­ent: they are rip­ples in the very fab­ric of space-time itself. They can thus be emit­ted by non-lumi­nous objects.

Studying black hole mergers

Cre­at­ing such rip­ples in the (rather rigid) fab­ric of space-time is not easy though. Indeed, GWs can only be pro­duced by accel­er­at­ing very small and extreme­ly mas­sive objects close to the speed of light. The best can­di­dates are there­fore black holes (which are the most com­pact objects in the uni­verse), and cer­tain very dense stars known as neu­tron stars (which are between 1.4 and 2.4 solar mass­es with a diam­e­ter of less than 20 km). To com­pare, the Sun has a diam­e­ter of 1.39 mil­lion kilometres.

In gen­er­al, an iso­lat­ed black hole does not pro­duce GWs. It needs a com­pan­ion to which it remains bound for a long time (much like Earth is bound to the Moon) to form what is called a bina­ry sys­tem. As they are extreme­ly dense, the black holes deform space-time in their vicin­i­ty as they orbit each oth­er, gen­er­at­ing GW rip­ples that prop­a­gate across the uni­verse at the speed of light.

As it emits these GWs, the bina­ry los­es some of the ener­gy that binds the black holes, which end up spi­ralling ever clos­er to each oth­er. This infer­nal waltz pro­duces more and more intense GWs (that can trav­el bil­lions of light-years across the uni­verse) until the black holes even­tu­al­ly merge. From time to time, one of these bina­ries pro­duces GWs with an ampli­tude that is just large enough to be detect­ed when it reach­es Earth, even though the sig­nal is extreme­ly weak.

The LIGO detection

The first sig­nal from such a bina­ry black-hole coa­les­cence, which occurred approx­i­mate­ly 1.3 bil­lion light-years from Earth, was detect­ed in Sep­tem­ber 2015 by an instru­ment called LIGO (for Laser Inter­fer­om­e­ter Grav­i­ta­tion­al-Wave Obser­va­to­ry)12. The coa­les­cence includes the “inspi­ral” (when the black holes become clos­er), the “merg­er” (when they touch) and the “ring­down” (when the new­ly formed, big­ger black hole relax­es into a steady state). 

The two black holes in ques­tion, of about 36 and 29 solar mass­es, even­tu­al­ly merged to form a sin­gle black hole of 62 solar mass­es. The 3 solar mass dif­fer­ence was entire­ly con­vert­ed into grav­i­ta­tion­al ener­gy car­ried by the GWs.

LIGO is a col­lab­o­ra­tive project with over 1000 sci­en­tists and engi­neers from more than 20 coun­tries, and three of its mem­bers were award­ed the 2017 Nobel Prize in Physics3. It took near­ly 50 years of intense research to build the GW detec­tors, and in June 2016 the researchers announced that they had observed a sec­ondary bina­ry black hole coa­les­cence4. The obser­va­tion was made on 26 Decem­ber 2015, and this time, the black holes were about 1.4 bil­lion light-years away. Rough­ly 50 such merg­er events have been detect­ed since this time. All these dis­cov­er­ies great­ly advanced many research fields and kicked off the era of grav­i­ta­tion­al-wave astronomy. 

Tiny length changes

Instru­ments like LIGO and oth­er ground-based GW detec­tors, such as Vir­go in Italy and Kagra in Japan, rely on an advanced sens­ing method called laser inter­fer­om­e­try. This tech­nique has long been used to detect dif­fer­ent sorts of sig­nals, but it had nev­er been pushed to the lim­it need­ed to detect the very weak sig­nals that GWs produce. 

The LIGO facil­i­ty basi­cal­ly works by send­ing twin laser beams down two 4 km-long “arms” arranged in an L‑shape and kept under a near-per­fect vac­u­um. The beams are reflect­ed by mir­rors pre­cise­ly posi­tioned at the ends of each arm.  As a GW pass­es through the obser­va­to­ry, it caus­es extreme­ly tiny dis­tor­tions in the dis­tance trav­elled by each laser beam. The instru­ment is thus able to mea­sure the local con­trac­tion and expan­sion of space-time caused by the GW.

The extreme sen­si­tiv­i­ty of the instru­ment means that it is prey to all sorts of exter­nal vibra­tions (such as those from planes fly­ing by and waves on a dis­tant shore). LIGO engi­neers there­fore had to design sev­er­al inge­nious noise-reduc­tion sys­tems that not only great­ly enhance the pre­ci­sion of the detec­tors, but also allow them to dif­fer­en­ti­ate between ter­res­tri­al arte­facts and the pre­cious GW signals.

By mea­sur­ing how long it takes for the laser beams to trav­el along an arm, researchers can extract infor­ma­tion such as the fre­quen­cy and ampli­tude of the GW from the sig­nal. These quan­ti­ties are of major impor­tance since they con­tain key phys­i­cal infor­ma­tion about the source of the wave, such as its dis­tance from Earth and its posi­tion in the sky, as well its mass and whether it is a black hole or a neu­tron star.

Future detectors

The char­ac­ter­is­tics of inter­fer­om­e­ters like LIGO makes them sen­si­tive only to grav­i­ta­tion­al waves with­in a cer­tain fre­quen­cy band, from about 10 Hz to 10 kHz, which cor­re­sponds to black holes of about 10 to 100 solar masses. 

To extend this fre­quen­cy range, the most promis­ing future project is the Laser Inter­fer­om­e­ter Space Anten­na (LISA)5. This Euro­pean space-based obser­va­to­ry, due to come online in 2034, will tar­get fre­quen­cies in the low­er mil­li­hertz range to detect waves from the merg­er of much big­ger black holes. These “super­mas­sive” objects are found at the cen­tres of most galax­ies – includ­ing our Milky Way – and have mass­es that are mil­lions or even bil­lions of times that of the Sun. 

LISA should also be able to observe “asym­met­ric” pairs, such as a neu­tron star orbit­ing a super­mas­sive black hole, and even the so-called “cos­mic grav­i­ta­tion­al-wave back­ground”, which is very impor­tant for cos­mol­o­gy since it con­tains infor­ma­tion about the pri­mor­dial GW cre­at­ed right after the Big-Bang6. Far more pre­cise than its ter­res­tri­al cousins, LISA will be a mil­lions-of-kilo­me­tre-long instru­ment con­sist­ing of three tiny robots posi­tioned in an equi­lat­er­al tri­an­gle pat­tern in solar orbit just behind Earth.

Interview by Isabelle Dumé
1https://​jour​nals​.aps​.org/​p​r​l​/​a​b​s​t​r​a​c​t​/​1​0​.​1​1​0​3​/​P​h​y​s​R​e​v​L​e​t​t​.​1​1​6​.​0​61102
2https://​www​.ligo​.cal​tech​.edu
3https://​www​.nobel​prize​.org/​p​r​i​z​e​s​/​p​h​y​s​i​c​s​/​2​0​1​7​/​p​r​e​s​s​-​r​e​l​ease/
4https://​jour​nals​.aps​.org/​p​r​l​/​a​b​s​t​r​a​c​t​/​1​0​.​1​1​0​3​/​P​h​y​s​R​e​v​L​e​t​t​.​1​1​6​.​2​41103
5https://​sci​.esa​.int/​w​e​b​/​lisa/
6https://link.springer.com/article/10.1007%2Fs10714-020–02691‑1

Contributors

Paul Ramond modifiée
Paul Ramond
PhD student at Observatoire de Paris and ENSTA Paris (IP Paris)

Paul Ramond's research topics concern various theoretical aspects of gravitational systems. His PhD research is carried out at the “Laboratoire Univers et Théories” in the Relativity and Compact Objects group. He works on the relativistic mechanics of black holes and neutron stars, which are the main astrophysical sources of gravitational waves. At ENSTA Paris, he is associated with the Laboratoire de Mathématiques Appliquées, where he works on classical gravitational dynamic systems.