Transcript Slide 1

Measurement of the Casimir force with a ferrule-top sensor
Paul Zuurbier
Supervisors:
Sven de Man
Davide Iannuzzi
Technical support:
Kier Heeck
Associated group members:
Grzegorz Gruca
Dhwajal Chavan
A phenomenon described in 1836
P.C.Caussée
L’Album du Marin
Two parallel ships are driven
to each other by a mysterious
attractive force
A likely explanation:
The two ships act like barriers
They are pushed one
against the other by the
waves outside “the gap”
The Casimir effect
e.m. wave = harmonic oscillator

1
E    d
2
0
in vacuum
1948: In the presence of two parallel plates (conductors)
The energy
Between the
Plates is lower
FCasimir
d
H.B.G.Casimir
(1909-2000)
Closely related
to van der
Waals force
The need of ferrule-top Casimir measurement
Increasing interest in studying the Casimir force in various
environments, for instance in liquids and with varying temperature.
Our group designed and manufactured the ferrule-top
sensor, which is versatile, adaptive and cost effective:
Measuring Casimir force is difficult,
so it is a good benchmark.
My job:
Test the new sensor by performing the first
ferrule-top Casimir force measurement.
Sphere and plate Casimir force

If
solution
too small → F too small
Radius ≈ 100 µm
d ≈ 40 – 200 nm
F < ~4000 pN
macroscopic objects
at microscopic distance
diameter ≈ 5000·dmin
Ferrule-top force sensor fabrication
Borosilicate ferrule
2.5 x 2.5 x 7.0 mm
Laser ablation:
200 x 200 µm ridge
100 µm gap
sphere is glued on
optical fiber is inserted
and fixed with glue
hole in cantilever is closed
gold layer is sputtered
on the sensor
Interferometer
Ferrule-top
not in
use
Table-top setup design
Left: Piezo translator with gold plate (varying d)
Right: Mechanical translator with sensor + sphere
Temperature stabilized Al cylinder
Al cover (dust and convection)
Dampers
Anechoic chamber
problems and solutions: Calibration
How does one calibrate a ferrule-top force sensor?
We calibrate continuously by applying a well known electrostatic force.
We apply an AC voltage to the sphere
We measure the signal due to this force
at double the frequency
We calculate the sensitivity
problems and solutions: Distance
How does one measure a distance
< 100 nm with ~1 nm accuracy?
With an second interferometer we
measure Δx.
At this stage we know d = Δx + d0,
but d0 is unknown.
From the electrostatic Coulomb
force
we get a signal S proportional to 1/d.
From this we can fit d0.
Δx
problems and solutions: Noise and drift
Since k~7 N/m and F<4 nN the cantilever bends only half a nanometer!
In this situation the drift of the interferometer intensity is overwhelming.
Therefore we vibrate the plate and measure ΔF:
F F  500 pN

d
d  7 nm
Because we are modulating the Casimir force we can use a
lock-in amplifier with superior noise suppression (AM).
problems and solutions: Hydrodynamics
Plate vibration
airflow
hydrodynamic force on sphere.
How does one distinguish between hydrodynamic and Casimir force?
The Casimir force depends on
The hydrodynamic force depends on
d ~ cos(ωt)
d
t
~ -sin(ωt)
Both signals are 90° out of phase (orthogonal).
The signal is measured with a lock-in amplifier and we can get
the Casimir force from channel X (in phase) and the hydrodynamic force
from channel Y (quadrature).
Hydrodynamic results
Final results
40.000 points
no free parameters
close agreement
with theory and earlier
experiments
conclusion:
the sensor is capable
of measuring Casimir
force,
article published NJP
End