Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 319, 331-339 (1997)

Previous Section Next Section Title Page Table of Contents

2. Experimental method

Our experimental set-up is sketched on Fig. 1. It is a modified version of that described in Moreels et al. (1994). The sample molecules are born by a molecular jet which is intersected with two parallel laser beams in opposite directions. The first laser ionizes (and/or fragments) the parent molecules and the second one probes the products for induced fluorescence after an appropriate time delay. We operate in a chamber with pressure of about 0.1 mbar. The gas in the jet is a mixture of an atomic carrier gas (helium or argon) and of PAHs vaporized from a powder sample. The backing pressure of the carrier gas is high (up to 5 bars); it is supersonically expanded in the low pressure chamber by passing through a heated nozzle of diameter 0.5 mm. The nozzle is closed by a valve pulsed at a frequency of 15 Hz and kept opened during ca 2 ms.

[FIGURE] Fig. 1. A partial sketch of our experimental set-up. The pulsed jet supersonically expands into the vacuum chamber after mixing of the sample molecules with helium or argon carrier gas. It is first crossed with the excimer laser beam and, 1 cm downstream, with the dye laser one. The fluorescence signal induced by the dye laser is collected by a lens whose axis is perpendicular to the laser beams. It is then dispersed by monochromators and sent onto a photomultiplier or a CCD array.

The density in the jet rapidly decreases after the nozzle as the square of the distance to the nozzle. The jet is first crossed with the focussed ([FORMULA] 1mm spot size) beam of an excimer laser close to the nozzle (a couple of mm below) to maximize excitation efficiency. The excimer laser pulsed at a repetition rate of 15 Hz, delivers a radiant energy of ca 150 mJ in each pulse of 15 ns duration. As neutral perylene exhibits a strong absorption band at 250 nm (the [FORMULA] -band, see Clar 1964), we selected a gas mixture to get the KrF excimer emission at 248 nm. The ionization potential (IP) of perylene being 7 eV (Leach 1987), two 248 nm-photons sequentially absorbed can ionize perylene.

The products of this first interaction then travel 1 cm downstream before encountering the beam of a Quantel dye laser, pumped by a Nd [FORMULA]:YAG laser, whose wavelength is tunable from 560 to 750 nm and extended to [FORMULA] 1 µm by stimulated Raman scattering in a high pressure H2 cell (see Bréchignac et al. 1986). The dye laser is triggered at 15 Hz and with the adequate time delay to ensure coincidence of the species earlier hit by the excimer laser. The delay depends on the carrier gas mass, it is of the order of 5 µsec in helium and 15 µsec in argon. The pulsed valve is synchronized with the laser pulses.

On their pathway through the 2 laser excitations, molecules are cooled very efficiently by collisions with the carrier gas. Possibly formed electronically excited states will also radiate to the electronic ground state during this delay. Typical rotation (vibration) temperatures currently achieved in such supersonic jets are 1-10 K (20-100 K). Moreover, the jet is very nearly monocinetic with a translational temperature down to 1 K so that there are no mutual collisions between molecules at the probing laser location. Argon has been recognized to be a more efficient jet coolant than helium (Smith et al. 1987 and this experiment): we therefore used it as carrier gas. In argon, however, clusters form more easily than in helium. We checked for the presence of van der Waals clusters by comparing excitation spectra (see below) obtained with helium and argon as carrier gases. As we found no significant spectral changes, we are confident that our results are not contaminated by the presence of clusters. Indeed, even if such weakly-bound species were formed in small concentration they would have almost no chance to survive after absorption of two (or more) 5 eV photons.
The supersonic jet producing cold and isolated molecules represents a very close simulation of interstellar conditions. Another advantage of the free jet is that the cations or radicals produced have little chance to recombine in the vacuum chamber where the density is decreasing very sharply away from the nozzle. Again, in contrast to rare-gas matrices where loosely trapped electrons and small fragments can diffuse to recombine, we can preserve most ions and radicals produced in the jet.

Collecting on the detector the fluorescence of the species excited by the dye laser, two kinds of measurements can be made. Monitoring the fluorescence signal while scanning the dye laser frequency provides an excitation spectrum, that gives the structure of the excited state (Fig.  2a): indeed, assuming a constant fluorescence quantum yield as a function of excitation wavelength, the excitation spectrum is expected to be proportional to the absotpyion spectrum. We can also, at fixed laser excitation wavelength, disperse the fluorescence and obtain an emission spectrum, which provides the ground state structure of the emitter(Fig. 2b).

[FIGURE] Fig. 2. The meaning of LIF spectrocopy is outlined. a The excited state of the emitting species is probed when the total fluorescence signal (monochromator's slits widely opened) is sent to the photomultiplier (excitation, or absorption, spectrum). b Conversely, the species' ground state is probed while the fluorescence is dispersed in the monochromator (emission spectrum).

The laser induced fluorescence photons go through a lens whose axis is perpendicular to the laser beams to reach the entrance slit of a monochromator. Two apparatus are then available to detect the signal. The first one is well-suited for absorption spectroscopy: we use a Jobin-Yvon monochromator with 1200 grooves mm-1 and wide entrance slit ([FORMULA] 1 mm). It is dispersive enough, however, to reject scattered laser light. At the exit slit the photons are collected on a Hamamatsu R943-02 photomultiplier. The signal is then gate-processed by a Boxcar integrator after proper amplification, and averaged over a selectable number (10 to 50) of laser shots before entering the data acquisition system. The gate is chosen so that the residual scattered light and eventual electronic noise are rejected. We obtain the structure of the absorbing species excited state. The laser linewidth is typically 0.1 cm-1.

The second channel of detection is available for acquiring emission spectra. It is a second monochromator with two gratings of low and high resolution. The size of the entrance slit is tunable down to 25 µm, and its image is formed on a CCD chip 256*1024 pixels. This allows, when the excitation wavelength is fixed, to measure in one exposure the whole emitted spectrum with the lowest resolution grating (150 grooves.mm-1), or to resolve one emission band with the 1200 grooves.mm-1 grating and a narrow slit. The highest resolving power obtained with this apparatus is [FORMULA] 3000. The CCD image is transfered to a PC computer, where the spectrum is averaged over the chip height (256 elements), which increases the S/N ratio. This detection is dedicated to probe the structure of the fluorescent species ground state, and therefore offers spectroscopical information complementary to the first channel described above.

To validate our experiment, we first studied the furan molecule photoexcited at 193 nm (ArF band) and reproduced quite well existing results (Smith et al. 1987). This experiment, contrarily to what was first announced, involved a neutral species (the C3 radical) as the fluorescence carrier as has been realized later on (Smith et al. 1988). We used this test case to optimize the adjustments of the whole set-up.
We also reproduced the results obtained on neutral perylene in free jet (Schwartz and Topp 1984).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998