Low thermal noise suspensions
for Virgo
The Virgo Perugia Group
•Luca Gammaitoni
•Fabio Marchesoni
•Michele Punturo
•Ciro Cattuto
•Helios Vocca
•Paolo Amico
•Flavio Travasso
•Leone Bosi
Perugia University and INFN
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
The expected Virgo Sensitivity
-15
10
Total sensitivity
Total thermal noise
Thermal noise: Mirror
Thermal noise: Pendulum
Shot noise
Radiation Pressure
Seismic
Creep
Magnetic
Laser Power
Acoustic
Quantum Limit
Newtonian (Cella)
Newtonian (Thorne)
-16
10
-17
10
h(f) [1/sqrt(Hz)]
-18
10
-19
10
-20
10
-21
10
-22
10
-23
10
-24
10
-25
10
0
10
1
10
2
3
10
10
4
10
Frequency [Hz]
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Pendulum Thermal Noise
X 2 (ω) ≈ 4 kBT
 1 Eg
2 E  φwire
+
θ
 2

g
mLw  TbCS
 Lw 4π nm
g
ω5
Wire loss angle:
φwire = φ0 + φth + φe
φ0
: Material loss angle
φth
: Thermoelastic loss angle
φe
: Excess loss angle
Tb : Breaking strength
CS : Confidence factor
•
Thermoelastic loss
angle
ωτ
φth = ∆
1 + (ωτ )2
∆=
Eα2T
c
τ=
d w2
π 2D
•
Excess loss angle
– Frictional
processes in the
marionetta-wire
clamps
– Frictional losses
in the wiremirror contact
point
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Present solution
The Virgo present solution for the suspension last stage
uses four 200 µm diameter C85 steel wires.
In these conditions:
- φ0 ˜ 10 – 4
- φth (max) ˜ 10 – 3 (at 680 Hz)
- CS = 0.6
- Tb ˜ 2.9 GPa
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
The next step will be
the monolithic solution
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Measured losses of materials
φ≈10-4
φ≈10-6
-3
10
A
factor
10
on
“h”
sensitivity
-4
10
φw (f)
C85 steel
Sapphire
Fused Silica
New F.S. with bob
Y a g + A l2O 3 +ZrO2
-5
10
-6
10
-7
10
Synthetic FS with bob
1
10
100
1000
frequency (Hz)
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
FS Production facility
•
•
•
•
Pure O2 -H2 flame
No fiber-support
contact
Direct
measurement of
the fiber diameter
Directly tested
the fiber breaking
strength using a
new automatic
device
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
FS Breaking strength (1)
7
The breaking strength of FS fibers
depends on material purity,
production and handling procedure.
6
5
4
3
TbSiO2 = 4.05 ± 0.55 GPa
TbC 85 = 2.90 ± 0.02 GPa
2
70
Br
00
60
ea
00
kin
50
00
Dir
0
00
gS 4
tre
ng
30
th
[M
Alc
00
20
Pa
]
00
10
00
Ac
oh
1
φw/Tb is a factor 200 lower
0
for FS fibers, but Cs is
lowered by the large Tb
fluctuation.
ty
ol
id
H.Vocca et al., Proc. of the3th Lisa symposium (2000, sub. to Class. Quant. Grav)
M.Punturo et al, talk to the GREX meeting (Grasse, 2000)
P.Amico et al, Nuclear Inst. and Methods in Physics Research A, pp 297-299 Apr.2001
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
FS Breaking strength (2)
• Very strong under elongation (and compression)
• Very fragile for shocks
• The Breaking strength is dramatically reduced by:
– Cracks on the fiber surface
• contact with metals or other hard surfaces
– Humidity
O
H2O
O Si O
H+
O
OHO Si O
O
• some worry for ageing caused by humidity
O
O Si O
OH
OH
O Si O
O
– no statistical evidence for this effect in our labs, but some “alarms”
– this effect is reported by optical fiber producers (closest to the theoretical FS
breaking strength)
– good news from Glasgow for fibers under vacuum
– more tests needed
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Coating of the wire surface
•Surface coating could
prevent FS ageing
•Effect of a Carbon
coating (with Nitrogen
contaminations) on
loss angle has been
tested
•An evident increase
of the wire loss angle
has been measured
•More tests are needed
with different coatings
and better thickness
control.
•It could be useful to
damp the violin modes
of the wires.
10
φw (f) 10
10
-4
F.S. with C coating
F.S. with C plus N=30%
F.S. with C plus N=60%
F.S. without coating
-5
-6
100
1000
frequency (Hz)
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Potassium silicate bonding
Silicate bonding is a Hydroxide-Catalysis chemical process.
Test are made in a class 10000 clean room, under a class 100 laminar
flux.
For λ/10 samples we are
dominated by the clamping
•Mean λ/4 =2.5±0.6 MPa
•Mean λ/7 =3.3±0.8 MPa
•Mean λ/10 =3.9±0.5 MPa
2
3
e
4
[ w e e k ]
5
6
7
8
9
1 0
6
4 6 . 9
1 0
l o a d
p a ]
1
[ M
S t r e s s
In this conditions the
breaking strength vs time of
samples with different
flatness quality has been
investigated (λ/4, λ/7 and
λ/10).
B r e a k i n g
–control of the purity of H2O
and KOH
–accurate cleaning of the
SiO 2 samples
0
[ k g ]
T im
It is important a:
1 0
4
3 1 . 3
7
4
1 0
1 0
7
1 0
4
1 0
2
0
7
1 4
2 1
2 8
3 5
T im
e
4 2
4 9
1 5 . 6
5 6
6 3
B r e a k i n g
•
•
7 0
[ D a y ]
λ=633 nm
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Silicate bonding as current
technology
• Silicate bonding technology is used in the
current Virgo assembling:
– Spacers will be attached to the lateral surface of the
mirror (polished @ λ/10 in a 40mm height strip) to
improve the pendulum Q
– Camera targets will be attached to the mirror
– Magnets will be bonded through an intermediate
fused silica disk to the face of the mirror directly on
the reflective coating.
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Future Implementation
Marionetta
clamp
FS λ/10 polished ear
•2 flat strips
•4 40×40mm λ/10 polished region
•Because of the complexity of the Virgo SA the hanging procedure
will be very crucial (and difficult)
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Monolithic solution
• With FS fibers we can gain in
– φ0
– thermoelastic damping because of
α SiO 2 ≈ 0. 5 ⋅10 −6 K −1 < α C85 ≈ 17 ⋅ 10 −6 K −1 ⇒ ∆ SiO 2 < ∆ C85
– clamping losses
– breaking strength
• On the other hand we lose in CS due to the dispersion of
the breaking strength
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Virgo sensitivity with FS suspension
-18
10
C85
Pendulum thermal noise ( φ=4*10 )
-7
6
Mirror Thermal noise (Q=10 )
-19
h(f) [1/sqrt(Hz)]
10
7
Mirror Thermal noise (Q=10 )
-20
10
-21
10
-22
10
-23
10
1
10
100
1000
Frequency [Hz]
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
All Fused Silica Suspension
November 1998, Perugia, quartz suspension system,
in collaboration with GEO Glasgow
•
•
•
•
Mass 2.8 kg
Length 30 cm
Frequency 0.93 Hz
2 fused silica wires 400
µm diameter
Very high Q measurements on a fused silica
monolithic pendulum for use in enhanced gravitational
wave detector.
G. Cagnoli, L. Gammaitoni, J. Hough, J. Kovalik, S.
McIntosh, M. Punturo, S. Rowan.
Phys. Rev. Lett
Lett.. 85, 2442-2445,
2000
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Pendulum Q for Fused Silica Suspension
0.01
7
amplitude in m
Q=2.5X10
1E-3
Q=2.1X10
7
Q=2.4X10
7
1E-4
0
5
10
15
20
25
30
35
40
45
time in days
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
What can we
do best?
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Thermoelastic loss angle (T=300°K)
10
10
φ( f )
10
10
10
10
C85
SiO 2
Si
GSGG
SiO 2 cris.
YAG
-3
-4
10
10
-5
10
-6
10
-7
10
-8
10
1
10
100
1000
-3
-4
-5
-6
-7
-8
10000
Frequency f (Hz)
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Cryogenic approach
-18
10
φSiO2 =4*10
-7
φSi =2.8*10
-19
10
hth (f)
hth (f)
hth (f)
-8
SiO 2
Si
Si
T=300K
T=300K
T= 70K
h(f) [1/sqrt(Hz)]
-20
10
-21
10
-22
10
-23
10
-24
10
-25
10
1
10
100
Frequency [Hz]
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Localized cooling
(In collaboration with M. Tonelli, Dipartimento di Fisica of Pisa)
•
Mono-crystal fibers are more
robust with respect to ageing
and OH contamination then
amorphous ones
• With the addition of
impurities of Yb3+ , it is
possible to cool the fiber
I.R.laser
optically with an anti-Stokes
mechanism
• In the case of fibers, the
optical path of the pumping
radiation can be increased by
using Total Internal Reflection
to raise the intensity and
absorption of the radiation.
• With this technique, it is possible to decrease the temperature by a few tens of
Degrees Kelvin
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002
Conclusions
• Thermal noise is and probably will be the main limit to the GW
interferometers
– Radiation pressure will limit the sensitivity if the laser power will
increase dramatically
• Fused Silica suspension is the right solution if the ITF operates
at room temperature
• Si fibers are, from the thermoelastic point of view, a possible
alternative at RT.
• Cryogenic ITFs are the most promising evolution for the GW
detectors
– Mono-crystal fibers/rods (Si, or…) must be used to suspend and to
extract the heat from the mirror
– Localized cooling could be a useful technology to further reduce the
temperature in the bending points of the suspension fibers
• We have enough work for the next years!!!
H.Vocca – GWADV Isola d’Elba, Italy, May 19-26 2002