The CNRS
Home IN2P3
Other CNRS sites
 
    Home > Research Activities > Low Loss Coatings
 
The subject of this reasearch is linked to the franco-italian program VIRGO for the detection of gravitational waves. VIRGO is financed, for the french part, by the CNRS (IN2P3). Four french laboratories (LAL-Orsay, LAPP-Annecy, ARTEMIS-Nice, LMA, ESPCI-Paris) and six italian laboratories (INFN Perugia, INFN Pise, INFN Naples, LNF Frascati, INFN Florence, INFN Rome) are involved in this program. For a long time, scientifics are trying to check and confirm the Einstein general theory of relativity which supposes the existence of gravitational waves. A lot of experiments have been done (suspended bars whose length variation is measured) but without a significant success. In 1960-1970, the principle of the gravitational waves interferometric detection has been defined and this leads to the VIRGO program. Other similar antennas will exist in the world : the american project LIGO including two interferometers, the german and british project GEO 600, the japanese one TAMA and the australian one ACIGA. To validate a detection, it is essential that several interferometers see the same phenomena at the same time.

The VIRGO principle is briefly explained below.

Space is not an absolute invariant entity, but is influenced by the distribution of mass and energy in the Universe. The gravitational waves are perturbations in the curvature of spacetime propagating with the velocity of light.
The gravitational waves, which are emitted by strongly accelerated large masses (for example binary systems made up of heavy compact stars or black holes) create a deformation of  the spacetime geometry and so a modification of the relative distance between two free masses. This phenomena can be measured with a Michelson interferometer adjusted on the dark fringe and illuminated with an ultra-stable source. A gravitational wave leads to a differential change Dl/l in the length of the two interferometer arms and then a change in the interference conditions. Thus, the detector placed at the output of the inteferometer will received a signal which is a signature of the gravitational wave. This Michelson type interferometer (Figure below) has extraordinary dimensions (3 km arms) and will have also extraordinary performances (measurement of length variations Dl/l about 10-21, which corresponds to the size of a hair in comparison with the earth-sun distance).


Optical drawing of the VIRGO interferometer ( R = Reflection )

A Fabry-Pérot cavity (M1-M2, M3-M4) is put in each arm. Thus, the arms are only 3 km long instead of 150 km. All the components are installed in the towers and are suspended by very powerful seismic isolation systems. The interferometer work under ultra high vacuum (10-8 à 10-9 Torr). It is located in Cascina near Pisa (Italy). This project is a real technological challenge due to its size and to the optical performances required for the mirrors. Indeed, the mirrors needed for VIRGO are the best optical components ever fabricated at the present moment. The losses at 1064 nm (Absorption, Scattering) must in the one part per million range (1 ppm) ; the wavefront of the coatings must be lower than 8 nm R.M.S on 150 mm diameter at 1064 nm. The technology for such optical components already exists (Dual Ion Beam Sputtering), but for small size about 1 inch (gyrolaser mirrors). This technology has to be improved to be able to coat 350 mm diameter components.

The LMA is involved in the VIRGO program since 1992. We have to coat the VIRGO large mirrors but also to characterize their optical performances (absorption and scattering losses at 1064 nm, wavefront homogeneity, microroughness).

The results obtained on the VIRGO mirrors are detailed more precisely at this link.

The table just below summarizes the optical performances evolution of the low loss coatings as a function of time. The result is remarkable.



Le tableau ci-dessous résume l'évolution des performances des couches minces faibles pertes au cours du temps. Le résultat est saisissant.




 

1992

1994

2006

Absorption moyenne à 633 nm

20 ppm 10 ppm < 5 ppm < 5 ppm
Absorption moyenne à 1064 nm 2 - 3 ppm 0,5 ppm 0,6 ppm
Diffusion moyenne à 633 nm 50 ppm 5 ppm 1,2 ppm X
Diffusion moyenne à 1064 nm X 2 ppm 0,6 ppm 4 ppm
Æ 200 mm
Front d'onde à 1064 nm X X X 3 nm RMS
Æ 150 mm  
Diamètre des composants  25 mm 50 mm 25 mm 400 mm

 

 

  Français

Laboratoire des Matériaux Avancés
(L.M.A.)