PVLAS Developments on Fabry-Perot Resonators
Locked to CW Lasers and Suitable for Laser Assisted
Lorentz Stripping of H" Beams
G. Cantatore*, M. Bregant*, F. Delia Valle*, G. Ruoso*, G. Zavattini@
^University and INFN - Trieste, Italy
INFN Legnaro National Laboratory, Legnaro, Padova, Italy
@
University and INFN - Ferrara, Italy
#
Abstract. The PVLAS experiment, running at the Legnaro National Laboratory of INFN (Padova, Italy), is operating a
6.4 m long, 105 finesse Fabry-Perot (FP) cavity frequency locked to a 1064 nm, 100 mW output power Nd:YAG infrared
laser. This cavity has a quality factor greater than 1012, and is kept at resonance for periods of time in excess of several
hours. A 100 mW, 532 nm, laser has been recently locked to an 87 cm long FP cavity with a finesse of 45000. Light
power density at the center of this resonator was 1200 W/cm2. Short term plans of the PVLAS experiment foresee the
installation of a 6.4 m long, 105 finesse, Fabry-Perot resonator operating at 532 nm. Dedicated developments with a 532
nm, 5 W, input laser would provide a 10 kW/cm2 power density, over a 1 cm2 cross section, useful for laser assisted H~
full stripping.
INTRODUCTION
THE PVLAS EXPERIMENT
Fabry-Perot interferometers are becoming basic
experimental tools in fields ranging from gravitational
wave detection, to metrology, to the study of quantum
vacuum[l]. The use of Fabry-Perot (FP) optical
cavities has recently been suggested within a laserassisted H~ double stripping scheme for the injection
of high-intensity proton drivers [2]. The PVLAS
experiment[3], running at the Legnaro National
Laboratory of INFN, near Padova, Italy, is conducting
an experimental study of the structure of quantum
vacuum and is operating a high-finesse, high-quality
factor, FP as a means to store photons in an interaction
region, where an intense magnetic field is used to
perturb the vacuum, and hence explore its structure. A
similar cavity could, under certain conditions, be
inserted without modifications in a laser-assisted
stripping apparatus. The PVLAS collaboration has also
successfully demonstrated the operation of a test FP
optical cavity coupled to a 532 nm laser, based on the
frequency-doubled emission from a Nd:YAG infrared
laser [4]. Using existing technology, this device could
be modified to accommodate the needs of the laser
assisted stripping schemes being currently proposed.
The first section below provides a short introduction to
the PVLAS experiment, while the following section
will present the current status of the optics in PVLAS.
The third section will briefly illustrate the original
techniques being used along with PVLAS driven
developments. Finally, the last section will discuss the
perspectives with respect to the H~ laser-assisted
double-stripping technique.
The basic idea underlying the PVLAS experiment
is to investigate the "material" properties of the
Quantum Vacuum with a specific reference to its
optical properties. It is well known that a gas-filled
region of space where a magnetic field is present
behaves as a birefringent crystal (Cotton-Mouton
effect). This effect depends, among other things, on
the gas pressure. In the zero pressure limit, contrary to
the classical expectation, a residual effect must be
present as evidence of the structure of the quantum
vacuum: Quantum Electrodynamics gives an
interpretation based on a photon-photon scattering
process[l]. In the PVLAS experimental scheme, the
vacuum is thought of as a target medium where an
external magnetic field establishes an anisotropy. A
linearly polarized light beam is then used as a probe to
sample this target by measuring the polarization
changes induced on the probe beam. In particular, one
wishes to measure the ellipticity, that is the ratio of the
semi-minor to the semi-major axis of the polarization
ellipse. The main elements of the apparatus are an
optical ellipsometer, a Fabry-Perot resonant optical
cavity, and a superconducting dipole magnet housed in
a rotating cryostat. The ellipsometer consists of two
crossed polarizing prisms and of an ellipticity
modulator, necessary to implement a heterodyne
detection scheme capable of detecting small, timevarying, signals over a large background (index of
refraction differences of the order of 10~22 must be
detected). To provide the time variation of the effect, a
1 m long, 6 T dipole magnet, operating at 4.2 K, is
CP642, High Intensity and High Brightness Hadron Beams: 20th ICFA Advanced Beam Dynamics Workshop on
High Intensity and High Brightness Hadron Beams, edited by W. Chou, Y. Mori, D. Neuffer, and J.-F. Ostiguy
© 2002 American Institute of Physics 0-7354-0097-0/02/$ 19.00
343
rotated, along with its enclosing cryostat, around a
vertical axis by means of a turntable. Finally, to
increase the optical path in the magnetic field region
up to 75 km, a 6.4 m long, 100000 finesse, FP cavity is
continuously kept at the resonance condition. This last
device will be illustrated in grater detail in the
following section.
By exploiting the fact that the Nd:YAG laser is based
on a Non Planar Ring Oscillator (NPRO) crystal,
which can be acted upon to change the beam
wavelength, the external frequency modulator is
eliminated from the feedback loop, providing
enhanced noise and stability characteristics. The
locking between laser and cavity can be maintained for
extended periods of time and the frequency difference
between laser and cavity can be kept at about 1
mHz/>/Hz around 2 Hz. This translates, for instance,
into the ability to sense changes in mirror separation
down to a level of AL = 6xlO~18
STATUS OF THE OPTICS
A Fabry-Perot optical resonator (or "cavity")
consists of two facing mirrors set at a given distance L
apart. In this device, light of wavelength A, can be
accumulated by constructive interference when
L = n(A/2), where n is an integer. For an ideal, lossless,
FP the input light intensity equals the output light
intensity, The performance of a FP is characterized by
its finesse F: the larger F, the longer the optical path,
the longer the average time T a photon will spend
inside the cavity. F is determined by mirror
reflectivity: for identical mirrors F = (7O/R)/(1-R), and
T = (FL)/(7Cc), where c is the speed of light. T, and
hence F, can be measured by observing the decay of
the light intensity transmitted by the cavity once the
input light is switched off. Such a decay curve,
measured for the PVLAS 6.4 m long FP cavity at
A=1064 nm yields T = 860 |is, that is F = 1.4 xlO5, for
a total path length within the magnetic field region of
(1 m)(2F/7i) = 75 km. During normal operation of the
experiment, the PVLAS cavity stays in the resonance
condition for a basically unlimited time while the
magnet-cryostat assembly rotates around it.
PVLAS Driven Developments
The present wavelength of the PVLAS laser beam
is 1064 nm. Operating at shorter wavelengths would
be advantageous from several points of view, among
which: the vacuum magnetic birefringence signal
increases with photon energy, a visible beam is easier
to align and operate, smaller beam radial dimensions
result in lower noise from optical components. For
these reasons a frequency doubled, continuous wave
(CW), 532 nm laser has been tested and locked to a
trial FP cavity, and is now ready to be installed on the
main PVLAS apparatus [6]. The tests have been
conducted on a Prometheus model laser, made by
Innolight,
Germany,
having
the
following
characteristics: infrared beam from a diode-pumped
Nd:YAG NPRO crystal doubled to 532 nm by a single
pass through a PPKTP non-linear crystal, visible 100
mW CW emission at 532 nm, secondary 1 W CW
emission at 1064 nm, frequency tunable. An 89 cm
long, 45000 finesse FP cavity was used for the locking
tests following the modified PDH locking scheme
outlined above. Stable locking was successfully
achieved and overall performance proved similar to
the performance of the laser-cavity system currently
installed on the main PVLAS apparatus.
ORIGINAL TECHNIQUES AND PVLAS
DRIVEN DEVELOPMENTS
Locking Technique
To operate properly, a FP cavity must be at
resonance. In practice, for mirror separations larger
than a few cm, resonance can be kept only by a
feedback scheme. One such scheme is the PoundDrever-Hall (PDH) method[5], which exploits the fact
that real FP cavities (with non-zero losses) reflect part
of the incoming light even at resonance. This reflected
light is used to sense the instantaneous frequency
difference between input laser light and cavity: the
laser wavelength is then continuously adjusted to
follow changes in mirror separation. In the original
PDH technique, the frequency difference between
laser and cavity is converted into an error signal,
which is then fed to a frequency modulator to close the
feedback loop. An original variation of this method
has been developed by the PVLAS collaboration [6].
PERSPECTIVES
It has recently been suggested that the high power
density which can be established inside a resonating
FP could be exploited to achieve the excitation of H°
atoms required by laser assisted stripping [2]. For a
Fabry-Perot cavity at resonance, one can construct a
power budget equation as follows:
wL=wout+wR+wloss+wHQ
(1)
where WL is the light power output by the laser
entering the FP cavity, Wout is the power in the
transmitted beam, WR is the reflected beam power,
Wioss is the overall power loss due to diffraction and
defects in the optical components, and WH° is a further
344
loss due to photons "lost" to H° excitation. At
resonance, light power within the cavity can be written
as Wc = (F/7c)Wout, where F is the cavity finesse, and
the
corresponding
power
density
is
DWC = (F/7c)(W0ut/S) = (r|FWL)/(7i;S), where S is the
beam cross section and r| = Wout/WL is the cavity
transmission ratio. If one assumes that 1016 H°/s must
be stripped[2], then 1016 photons/s are lost from the
cavity, leading (for A, = 532 nm), to a "stripping" loss
given by:
m long FP cavity, the shorter test FP cavity, the 6.4 m
long cavity upgraded to the 532 nm wavelength, and
for a projected cavity adjusted according to the
requirements of laser assisted stripping. In this last
case, only two of the parameters present technical
challenges which have not already been met: light
power input to the cavity and mirror radius of
curvature. Mirror radius of curvature depends on the
ability of manufacturers to prepare suitable mirror
substrates, and it can safely be assumed that the only
limiting factor would probably be cost. Assuming that,
for ease of operation, a Nd:YAG frequency-doubled
laser is chosen, laser power requirements could be met
by both increasing pump power to the primary NPRO
crystal and by increasing doubling efficiency. This
could be achieved, for instance, by using a larger nonlinear crystal and switching from a single pass
doubling scheme to a multiple pass one. In conclusion,
from the point of view of the construction of a FabryPerot cavity suitable for laser assisted stripping, the
technology is either already available or it appears to
be within reasonable reach.
ph
This loss is negligible compared to typical values
of the laser light power within the FP cavities tested by
PVLAS. The minimum power density required for
laser assisted stripping is 10 Kw/cm2, to be established
over a 1 cm2 beam cross section, that is a beam radius
(waist) of 0.56 cm. Using Gaussian beam propagation
equations, one can then estimate the FP cavity mirror
radius of curvature required for proper matching of the
beam to the cavity itself. Table 1 presents a summary
of the relevant parameters for the current PVLAS 6.4
Laser power
WL
A
ri
F
Mirror radius of curv.
Beam waist
L
Optical losses
DWC
WH°
WH7WC
TABLE 1. Summary of relevant Fabry-Perot parameters (see text)
PVLAS current
PVLAS test
PVLAS upgrade
100 mW
100 mW
100 mW
45 mW
20 mW
20 mW
1064 nm
532 nm
532 nm
0.13
0.07
0.13
100000
45000
100000
llm
llm
llm
1.3mm
0.7mm
0.92mm
6.4m
0.89m
6.4m
1.6xlO'5
1.6xlO'5
1.6xlO'5
3600 W/cm2
700 W/cm2
1200 W/cm2
3mW
3mW
3mW
2xlO'4
2xlO'4
2xlO'4
Projected
5-10 W
1.6 W
532 nm
0.2
100000
110m
5.6mm
6.4m
1.6xlO'5
10 kW/cm2
3mW
3xlO'7
5. Pound, R. V., Rev. Sci. Instr. 17, 490 (1946); Pound, R.
V., et al., Appl. Phys. B 31, 97 (1983).
REFERENCES
6. Cantatore, G., et al., Rev. Sci. Instr. 66, 2785 (1995).
1. Zavattini, E., et al., "The PVLAS Collaboration:
Experimental Search For Anisotropy Of The Phase
Velocity Of Light In Vacuum Due to a Static Magnetic
Field" in QED 2000, edited by G. Cantatore, AIP
Conference Proceedings 564, New York: American
Institute of Physics, 2001, p. 77.
2. Gastaldi, U., and Placentino, M., Nucl Instr. Meth. A
451, 318 (2000).
3. http://www.ts.infn.it/experiments/pvlas
4. Bregant, M., Cantatore, G., Delia Valle, F., Ruoso, G.,
and Zavattini, G., hep-ex/0202046, submitted to Rev. Sci.
Instr.
345