Francesco Cottone

1
Thermomechanical properties of silicon fibers
for 3rd GW detectors

• ILIAS GW Meeting October 25th, 2005
• Palma de Mallorca, Spain

• Francesco Cottone
• INFN Physics Departments of
• Perugia, Pisa, Florence
• (Collaboration Work under VIRGO Project)

2
Outlines

• Requirements for low thermal noise Suspension
  wire
• Silicon thermo-mechanical properties
• Experimental setup
• Measurements of thermo-mechanical parameters at
  Room and low Temperature
• Conclusions

3
Sources of Thermal Noise in VIRGO Interferometer

• Mirror internal vibration
• Suspension wires fluctuation

4
Sources of Thermal Noise in VIRGO Interferometer
Suspension thermal noise
5
Requirements for low thermal noise Suspension wire
LOSS ANGLE
Horizontal displacement power spectrum
Tensile strength
Mirror mass
Wire length
Number of wires

• Low loss angle f
• High tensile strength TB
• High thermal conductivity
• Low thermal expansion coefficient

6
Materials Loss Angle
7
Thermo-elastic Loss AngleWhy crystalline Si for
future GW detectors?
8
Thermomechanical Properties of Crystalline Silicon
Low thermal expansion coefficient a That expected
to vanish at about 17 K and 123 K
Crystalline silicon is a good candidate thanks to
its high thermal conductivity (at 300K) 1.48
102Wm-1 K-1)
9
Thermomechanical Properties of Crystalline
Silicon
Amplitude of the linear thermoelastic loss angle
Vs Temperature
Expected temperature dependence of the
thermoelastic peak frequency
10
Production of Silicon Fibers (Pisa group)
µ-pulling down Technique
11
Cryogenic Experimental Setup (Perugia Lab)
Clamping system
steel alloy spring
Fiber
Copper block and conduction plate to realize
thermal link
Shadow meter
12
Experimental results at Room and Low Temperature
Resonance frequencies modes
The fiber has roughtly elliptical section
apporximated with two circular sections sections
with 2 diameters d1 and d2that can be deduced
from each mode
Free lenght111.5 mm diameter 242 mm
13
Experimental results at Room and Low Temperature
Young Modulus Vs Temperature
Relative resonance frequency variation
Wachtman et al., Phys. Rev. 122 (1961)
14
Experimental results at Room and Low Temperature
Loss Angle at Room Temperature Vs Frequency
(Firenze group)
Free lenght278 mm diameter 574 mm
Free lenght111.5 mm diameter 242 mm
15
Experimental results at Room and Low Temperature
Loss Angle at low Temperature (Perugia group)
Excess losses?
Loss angle vs temperature
Loss angle vs frequency
16
Conclusions

• The µ-pulling down technique and the etching
  procedure permit to realize fibers having the
  appropriate size
• At room temperature, the fibres confirm the
  expected properties
• the high thermal conductivity of the Silicon
  pushes the thermoelastic dissipation peak at
  high frequency
• At low temperature (about 120 K) the reduction of
  the thermoelastic dissipation, due to the
  thermal expansion coefficient behavior, is hidden
  by the presence of other dissipation mechanisms,
  probably related to bulk and surface defects.

17
Next steps

• Better control of the diameter regularity and
  crystal orientation must be developed
• Improving of the clamping system to realize more
  robust blocking to reduce extra losses (in
  progress)
• Necessity to investigate the behavior of a
  crystalline Silicon fibres suspension at really
  cryogenic temperature
• At about 520 K the thermoelastic dissipation
  will be negligible thanks both to the thermal
  expansion coefficient vanishing and to the direct
  temperature dependence of the thermoelastic
  strength ?.