The term excimer laser covers a family of laser systems that uses a gaseous mixture excited by electrical discharges as active medium. Excimer means excited-molecules bound only in the excited state and unbound (or weakly bound) in the ground state. This guarantees automatic lower-level depopulation, thereby allowing population inversion, as long as the excited molecules are present, and hence high laser output power. Among the excimer molecules, rare gas halides (RGHs) are the most efficient active media for emitting powerful ultraviolet (UV) laser light. One of the most effective RGH lasers is the XeCl, emitting laser photons at a wavelength l=308 nm. The wall-plug efficiency of commercial XeCl lasers ranges between 1% and 4%, one of the highest among UV lasers. This fact has promoted the widespread use of the XeCl and other excimer lasers in many applications requiring intense UV light, e.g., microelectronics, photo-chemistry, material processing, micromechanics, remote sensing and medicine. However, commercial excimer lasers are not suitable for important applications such as large-area material processing and propagation of high-energy and low-peak power laser pulses in optical fibres. This is due, respectively, to the limited (<5 cm2 ) beam size (which in turn limits the output energy) and to the short (< 30 ns) laser pulsewidth typical of commercial laser systems. These problems were addressed at ENEA Frascati back in the late seventies, when work started on designing and constructing XeCl lasers with a large active medium, high output energy and long pulsewidth.
During 1998, the work at Frascati was mainly focused on developing
and optimising the XeCl laser Ianus and on using the XeCl laser-
facility Hercules to process materials and to develop a plasma x-ray
source, in collaboration with universities and industries. In the
framework of the European FOTO Project,
the Frascati group designed the mechanical and electrical parts of an
XeC1 laser called “Hercules L”, to be used for producing flat-panel
with thin- film transistors by laser annealing of 10 cm x l0 cm a-Si.
laser will be assembled by the EL.EN. company and installed at the end
1999 at ENEA Portici.
The innovative feature of Ianus (fig. 1) is its double-discharge (three-electrode) structure, which makes it suitable for oscillator-amplifier configurations. The two laser discharges are geometrically in parallel so they can use the same gas-flow loop, and electrically in series so that they are simultaneously pumped. This guarantees their automatic synchronisation. The Ianus design is an ENEA patent.
Fig. 1: The three-electrode XeCl laser system
Ianus was used to generate a laser beam with bow-divergence, a
requirement in many applications. The smallest of the two discharges
(called “oscillator”) was equipped with a generalised self-filtering
unstable resonator (GSFUR), a special kind of non-confocal
negative-branch unstable resonator with magnification
M=8, which enabled a laser output energy of 10 mJ with a
(FWHM) pulse duration of 90 ns. When injecting the oscillator beam in
amplifier, the energy of the amplified beam was nine times that of the
To test the laser-beam optical quality, the experimental spatial energy distribution in the far field was compared with the diffraction-limited energy distribution, according to the most advanced definition of the “times diffraction limit” (TDL) number of highly diffracted laser beams. Figure 2 shows the interpolating function of the experimental far-field and diffraction-limited energy distributions calculated from the experimental near-field energy distribution. The oscillator laser beam was only 1000 over its diffraction limit (TDL= 1.1). This is an excellent result, considering the superradiant nature of high-gain excimer media and the small value of the diffraction-limited divergence due to the short wavelength emitted by excimer lasers.
Fig. 2: Diffraction-limited (dashed) and
experimental (continuous) far-field energy distributions of Ianus.
The quality of the amplified beam, analysed with the same methods used for the oscillator beam, had a TDL=1 .6. The discharge deterioration with time and the lack of suitable optical insulation between oscillator and amplifier account for the slight increment in the TDL. In any case, the most significant laser beam parameter, the beam radiance (defined as the laser peak-power/spot size/solid angle), increased from 5x1013 W/cm2/ster (oscillator) to 3x1014 W/cm2/ster (amplifier), one of the highest values reported in the literature for long-pulse UV lasers.
The foregoing measurements were repeated when operating the GSFUR in
burst mode up to a repetition rate of 50 Hz. A nearly
diffraction-limited divergence was achieved from the beginning of the
laser pulse and, most important,
the values of the TDL, M2 parameter and beam angular
(BAS) were maintained, independently of the repetition rate. The BAS
in fluctuations smaller than one third of the beam divergence.
Hercules, the oldest laser system operating at Frascati, was completed and characterised in the period 1987-1992. Since 1993, it has been available to users from universities as well as private and public companies interested in finding the optimum working point while irradiating and processing metals, semiconductors, glass, plastic and exotic materials and in driving soft-x-ray plasma sources.
Hercules is pumped by a discharge ignited by an X-ray pulse injected
the active medium when the desired voltage value across the electrodes
reached. This technique, called “phototriggered discharge”, allows
reliable and almost jitter-free operation in the repetition rate mode.
discharge geometry, following a numerical study of the electric field
in the discharge chamber which takes into account the nearby current
parts, the electrodes were shaped adopting a “mixed” solution
cathode and Ernst anode). In this way, a uniform pumping discharge and
higher voltage breakdown level were achieved: no surface discharges on
insulator are detected even at the maximum repetition rate of 10 Hz.
Today Hercules is a XeCl laser system (fig. 3) operating in the repetition rate mode and emitting one of the highest energies per pulse in Europe, with long lifetime components and reliable performance (summarised in table 1).
When required, Hercules can also be used to amp1ify a 10-ns and 0.05-J XeCI laser pulse emitted by a commercial system; the amplified output laser pulse has a 2-J energy, I0-ns duration and (0.1x0.1) mrad2 divergence.
The EU FOTO project up
The tasks of the excimer group at Frascati in the framework of the
EU-funded FOTO Project are to develop the electromechanical design of a
large-volume, high-output energy, industrial excimer laser called
Hercules L and to use the actual Hercules to test the performance of a
line-step beam homogeniser when annealing a-Si samples. Figure 4 shows the
layout of the industrial prototype laser Hercules L.
Fig. 4: Layout of the industrial prototype
Basically, Hercules L reproduces the scheme of Hercules, with some modifications to further enhance both laser performance and reliability (e.g., higher gas pressure, ceramic materials covering the insulator parts, state-of-the-art heat exchanger and easy access to optics and consumable parts). In particular, the preioniser is a new X-ray diode which uses commercial spark-plugs as the plasma cathode. Results show that this type of cathode is a low-cost, reliable, rugged and long-lifetime electron gun. After more than 106 shots, interrupted without any faults being found, the dose/shot was 7 mrd, the X-ray spatial distribution over a 100-cm length was uniform within 93% and the ionisation rate was greater than 1014 electrons/s/bar. These values guarantee are effective preionisation of XeCl excimer laser discharges. Due both to the long lifetime and to the substantial absence of maintenance, this x-ray diode seems suitable for preionising commercial gas lasers such as the excimer and the TEA CO2.
The beam homogeniser was designed to transform the Hercules laser
beam from 50 mm x 100 mm to 11 mm x 130 mm with a spatial uniformity
95%. The homogeniser consists of two arrays of cylindrical lenses, two
condensers, and a 45° mirror to bend the laser beam to a vertical
(fig. 3). Figure 5 shows the intensity
of the laser beam in the overlap plane of the line-step homogeniser
the horizontal and vertical directions. Note that in the central
region the homogeneity has a 3% root mean square (rms).
Preliminary annealing tests showed the formation of crystal grains with a size of about 1 mm, suitable for the production of high-mobility thin-film transistors. These results were obtained on a 500-Å a-Si film deposited on glass substrate, placed in a vacuum chamber and heated to 600 °C. The homogenised laser energy density on the sample was 0.4 J/cm2. Figure 6 shows a scanning electron microscopy (SEM) image of the grains in the irradiated Si.
Fig. 5: Intensity profiles of the Hercules
laser beam in the overlap-plane of the line-step homogeneiser along
the horizontal a) and the vertical b) directions. Note the different size of the beam along the two directions (units are mm).
Fig. 6: SEM image of Si sample after
irradiation with the homogeneised laser beam of fig. 5.
Note the 1-mm-wide grains created by the so-called "super-lateral-growth" effect.
Soft-X-ray generation up
An interesting application of high-power excimer lasers is the generation of high-radiance soft-x-ray sources through creating a small-size, high-density plasma on the surface of a target. When the target is heated by a focussed pulsed laser beam with an energy per shot of 1-10 J, pulse duration of 1-100 ns and low divergence, this forms a plasma with a temperature of ~106 K which emits radiation in the soft-x-ray spectrum, say 10-100 Å. The shorter the laser wavelength, the deeper the penetration of the laser radiation inside the plasma and the higher the conversion efficiency from Laser energy to soft-x-ray energy. As a result, excimer lasers are the best candidates for generating x-ray plasma sources. Hercules, when equipped with the positive-branch unstable resonator (PBUR) (see table 1), has the right energy, pulse width, divergence and wavelength for generating a high-radiance soft-x-ray source. Experiments based on focusing the Hercules laser beam into a 30-mm-diam circle, where the laser intensity reaches a value of 1013 W/cm2, were carried out in the framework of a collaboration between ENEA, L’Aquila University, Milan University and the Italian Institute of Health (Rome). The measured sizes of the x-ray source and laser spot are comparable (30 mm) and the energy conversion efficiency from laser to X-rays exceeds 20%. Hence, this x-ray source has a radiance value comparable to that of a synchrotron storage ring, especially in the spectral region below 1 keV.
The characteristics of this plasma source made it possible to work
following experiments during 1998:
a) atmospheric pressure soft-X-ray microscopy;
b) propagation of X-rays in capillary tubes;
c) reduction of particulates (debris) emitted by the plasma;
d) high-resolution X-ray spectroscopy;
e) construction of an extreme ultraviolet (XUV) laser.
a) A new x-ray microscopy technique based on the propagation of soft x-rays in He gas at atmospheric pressure was developed. In this way, the biological specimen can be quickly inserted into the laser- target interaction chamber; hence, even delicate and short-lifetime specimens can be imaged and the quality of the images is improved by the reduction in debris bombardment (He slows down the debris). Figure 7 shows x-ray microscopy images of ciano-bacteria and of an intracellular element (a mythocondrius). The resolution is better than 100 nm (fig. 7b), far beyond the resolution of a light microscope.
Fig. 7: Soft X-ray microscopy images obtained with a tantalum target, 2-J, 10-ns laser pulse. a) Ciano-bacteria Leptolyngbya in the vacuum chamber; b) mythocondrius in a chamber at atmospheric pressure of He.
b)Preliminary results of propagation of soft X-rays (at 70 and 1200 eV) in capillary tubes (see fig. 8) were obtained. These experiments are important as capillary tubes could be a powerful tool to transport the x-ray beam out of the laser-target interaction chamber and could potentially allow new applications such as near-field scanning microscopy: a transmission map of a biological specimen is obtained point by point with a spatial resolution better than 50 nm just by illuminating the sample with an X-ray microbeam coming out from a tapered glass microtube.
Fig. 8: pattern of X rays transmitted through
tapered glass capillary (Fin =
mm, Fout = 100 mm, length = 17 cm)
imaged on a sensitive X-ray film. Single-shot exposure. Film distance from capillary output = 6.5 cm.
c) Reduction of particulate emission from the plasma is very important for specific applications like X-ray microlithography where fragile masks and optics must be placed dose to the plasma source. Work was done on the characterisation and minimisation of both the amount and the speed of the debris emitted by the plasma for different laser parameters (pulse energy, spot diameter, etc.). A Faraday cup and an electrostatic analyser were used to measure the charge state and the speed of ions emitted by tantalum, copper and aluminium targets (fig. 9).
Fig. 9: Ion charge distribution a) and ion current density vs. time since plasma generation b) from a 100-mm-thick Ta tape target irradiated by 120-ns laser pulses. Distance between Farday cup and plasma = 1m. Residual air pressure = 10-5 mbar.
A gated charge coupled device (CCD) camera allowed observation of hot debris emitted by a tantalum target (fig. 10). This experimental result is very interesting for projection microlithography applications: it was found that the laser parameters which optimise the emission at 70 eV (300-mm laser spot diameter and 120-ns pulse duration) significantly reduce both amount and speed of debris with respect to the laser parameters typically used in microlithography (less than 100-mm spot diameter and 10-ns pulse duration).
Fig. 10: Hot debris emitted by a Ta terget.
Laser pulse 4 J, 120 ns, delay 0.5 ms, exposure time 50 ms.
d) The interaction between the plasma and a cold gas (helium al atmospheric pressure) was studied, and the presence of hollow atoms (atoms with the first electron shell, n = 1, empty) in the plasma obtained with a magnesium target was successfully investigated. These important results of basic plasma physics were obtained thanks to special high-resolution X-ray spectrometers manufactured by the Institute NPO “VNIIFTRI” of Moscow, in the framework of a bilateral co-operation between the governments of Italy and the Federal Republic of Russia.
e) In the framework of a collaboration between ENEA, L’Aquila
University and Pecs University (Hungary), Frascati contributed to the
design and construction of a laser system based on XUV emission from an
argon gas pumped by a capillary electric discharge. The principle of
operation is to discharge a 30-kA electric
current through a capillary filled with 1-mbar argon; the magnetic
squeezes the discharge into a narrow column (F<
0.5 mm), thus heating the argon at a temperature dose to 106
and creating a plasma with ions in the charge state Ar+8
(Ne-like). The construction of the laser was completed during 1998 at
the Physics Department
of L’Aquila University, and experiments are going on to achieve laser
at l = 46.9 nm (26 eV).
S. Bollanti, P. Di Lazzaro,
Flora, G. Giordano, T. Letardi, G. Schina, C. E. Zheng: "Ianus, the
laser" Appl. Phys. B 66 (1998) 401.
S. Bollanti, P. Albertano,
Belli, P. Di Lazzaro, A. Faenov, F. Flora, G. Giordano, A. Grilli, F.
Kukhlevsky, T. Letardi, A. Nottola, L. Palladino, T. Pikuz, A. Reale, L. Reale, A. Scafati, M. Tabocchini, E. Turcu, K.
Papadaki, G. Schina: "Soft X-ray plasma source for atmospheric pressure microscopy, radiobiology and other
applications" Il Nuovo Cimento D 20 (1998) 1685.
F. Rosmej, A. Faenov, T.
Pikuz, F. Flora, P. Di Lazzaro, T. Letardi, A. Grilli, A. Reale, L.
Scafati, L. Reale: "Line formation of high intensity Heb-Rydberg dielectronic satellite 1s3lnl’ in dense laser produced
plasma" J. Phys. B: At. Mol. Opt. Phys. 31 (1998) L921.
A. Magunov, A. Faenov, I.
Skobelev, T. Pikuz, D. Batani, M. Milani, M. Costato, A. Pozzi, E.
Turcu, R. Allot, M.
Koenig, A. Benuzzi, F. Flora, A. Reale: "Formation of X-ray spectrum of excimer laser-produced plasmas", Laser and
Particles Beams 16 (1998) 61.
B. Young, A. Osterheld, D.
R. Stepherd, R. Stewart, A. Faenov, A. Magunov, T. Pikuz, I. Skobelev,
Bollanti, P. Di Lazzaro, T. Letardi, A. Grilli, L. Palladino, A. Reale, A. Scafati, L. Reale: " High resolution X-ray
spectrometer based on spherically bent crystals for investigations of femtosecond laser plasmas", Review of Scientific
Instrument 69 (1998) 4049.
P. Di Lazzaro: "Hercules,
laser facility for high-intensity irradiation experiments" (Invited
Proc. 2nd GR-I
Int. Conf. on New Lasers, Technologies and Applications, A. Carabelas, P. Di Lazzaro, A. Torre, G. Baldacchini Eds.
SPIE vol. 3423 (1998) pp. 35-43.
S. Bollanti, P. Di Lazzaro,
Murra: "Definition and measurement of the times-diffraction-limit
laser beam" Proc. 2nd GR-I Int. Conf. on New Lasers, Technologies and Applications, A. Carabelas, P. Di Lazzaro, A.
Torre, G. Baldacchini Eds. SPIE vol. 3423 (1998) pp. 145-150.
P. Di Lazzaro, F. Flora, N.
"Measurement of the time-resolved divergence of pulsed laser beams"
Int. Conf. on New Lasers, Technologies and Applications, A. Carabelas, P. Di Lazzaro, A. Torre, G. Baldacchini Eds.
SPIE vol. 3423 (1998) pp. 155-159.
S. Bollanti, R. Cotton, P.
Lazzaro, F. Flora, G. Giordano, G. Schina, T. Letardi, N. Lisi, A.
Nottola, K. Papadaki, A.
Reale, L. Palladino, L. Reale, M. Belli, F. Ianzini, A. Tabocchini, A. Scafati, A. Faenov, T. Pikuz, P. Albertano:
"Applications of an X-ray plasma source driven by an excimer laser: in-vivo microscopy, microradiography, and
radiobiology" Proc. 2nd GR-I Int. Conf. on New Lasers, Technologies and Applications, A. Carabelas, P. Di Lazzaro,
A. Torre, G. Baldacchini Eds. SPIE vol. 3423 (1998) pp. 406-410.
S. Bollanti, P. Di Lazzaro,
Flora, G. Giordano, T. Letardi, D. Murra, G. Schina, C.E. Zheng: "X-ray
spark-plug plasma cathode", XII Int. Symp. on Gas Flow and Chemical Lasers and High Power Laser Conf., A.
Boreisho, G. Baranov Eds. SPIE vol. 3574 (1998) pp. 84–92.