ACTIVITY USING MOLECULAR BEAM AND LASER TECHNIQUES : A PROGRESS REPORT

We shall briefly describe the main research areas, now in progress in our molecular beam and laser group. The current activity includes our traditional molecular beam work as well as other experiments carried out in the new labs of our Institute. Essentially we present here the current progress on Crossed-beam and Beam-gas Chemilumine scence, Laser-induced Beam Surface Reaction, Laser Analytical Chemistry, Laser-initiated Reactions Studies by Chemiluminescence and Laser-induced Fluorescence, Spectroscopy and Structure of van der Waals molecules. For each section recent results are presented. Due to the basic scope of the present progress report the emphasis is made on the new experimental developments as well as on the capabilities of both basic and applied research techniques now operative in our group.

calcium atoms in a beam collide with molecules in a supersonic beam (typically, HBr, HC1, N20, etc).By using a pulsed calcium beam any spontaneous product fluorescence can be detected in coincidence with the reagent pulse.This allows us to measure high resolution excitation functions like the one shown in Figure 2. The collision energy dependence of the chemiluminescent reaction cross-section as well as the polarization of the chemiluminescence were studied to gain insight into the dynamics and stereodynamics of these prototype heavy + heavy-light reactions.Significant features now under study include the Eco / kJ-mol ' FIGURE 2 Collision energy dependence of the cross-section for the Ca(1D) + HBr ---}CaBr(A2H)+ H reaction.Notice the energy threshold around Ea=0.2eV even though the reaction is exoergic in ADo 46.50 kJ.mol-1.
J. CASTANO et al.
possibility of quantal effects, which could manifest themselves as reso- nances in their excitation functions and a correlation between the L-L' vectors (i.e. between the reagent-product orbital angular momen- ta).As can be seen in Figure 2 an energy threshold results around E r_ 0.2 eV for the chemiluminiscent Ca(D2) + HBrCaBr(A2H) + H reaction.The appearance of such a threshold in spite of a reac- tion exothermicity of ADo 46.50 kJ.mol-is a matter now under study.A more complete discussion can be found in Ref. 4 and 16.   Figure 3 shows the excitation function for the Ca(1g)2)+ HCI CaCI(A) + H reaction (top Fig.) together with different model calcula- tions using a harpooning model based on succesive and independent crossings through the relevant covalent and ionic potentials of the displayed potential energy surfaces [12].The manifold of these cross- ings is shown in Figure 4.The major significance of these results obtained with the harpooning dynamics is that the observed transla- tional features of the Ca(1D2) + HC1 --.CaC1 + H chemiluminiscent re- action can be explained by the combination of two reaction channels.
One is open at all energies above 80 meV and the other is operative only above 200 meV.Whereas the former deals with crossings number and 2, see Figure 4, the latter is clearly associated to the inner crossing between the covalent Ca(1P)+HC1 and ionic Ca + (2D)+ HC1-potentials, and it has been shown [12] that is res- ponsible for the rising of an, the reaction cross-section, and its maxi- mum at the collision energy, E r 290 meV.
The onset of significant threshold for these highly exothermic reac- tions is of some interest by itself since it not only provides relevant data to understand the dynamics of excited reactions, but it also f-arnishes a good example on that the symmetry and topology of the excited potential, and not only the total energy, is one of the key factors controlling the reactivity of such excited species.
The product chemiluminescence spectrum as well as the polari- zation of the emission can be measured when the beam-gas arrange- ment is used.A schematic view of the set-up is shown in Figure 5.A detailed spectrum of the CaI* (A,B) formed in the Ca(D)+ CH3I--CaI (A2H, B2Z)+ CH 3 reaction is displayed in Figure 6.Notice the presence of both A and B states as well as the splitting of the 21-I state.
As mentioned above, looking at the polarization of the emission pro- vides insight into the stereodynamics of the elementary chemical E T / meV FIGURE 3 Three sample calculations of the cross-section for production of CaCI (A, 2I-I) corresponding to three different relative contributions due to the channel which involves crossing 2 (see Fig. 4).For a better illustration and comparison the experimen- tal c R (E), is also displayed at the top of the Figure.4 Plots of potential energy curves relevant for the description of the entrance channels for the reactions of excited calcium atoms and HC1, as estimated neglecting dependence from HC1 orientation (this allows the E, H, A labelings at long range, see text).Ionic curves are also shown as dotted lines and features of crossings as used in Eqs. 12, 13 are: R 4.58, R 3.92, R 2.89/, E 84, E 8 meV, E 201 meV, A --40, A --964 and A 2862 meV.-1.The lower curve, corres-ponding to the ground state, is discussed in the text.Adapted from Ref. x/nm In a series of experiments carried out under beam-gas conditions both chemiluminescence spectra and absolute values of the reaction cross-sections were measured for the reaction system Ca(1D2, aPs) + RX CaX + R in which R CH3, C2H and n-Call7 and X I, OH.One of the basic objectives in undertaking this systematic study was the study of the radical effect in these reactive collisions.Another, perhaps, more interesting objective was to measure the reactivity in absolute values.The reader is addressed to Ref. 1 for further details about the methods to estimate aR in absolute values, e.g./2.For a better illustration Figure 7 displays the evolution of the total reaction cross-section, normalized to the methyl reaction, as a function of the size of the radical group.For a better illustration data from the well known K + IR KI + R (R CH 3, C2H5, n-C3HT) reaction family, also mea- sured in our laboratory, is enclosed.After inspection of Figure 7a negative dependence on the aR value is clearly noticeable, as the radi- cal group increases.This body of data together with additional results from our and other laboratories, allowed us to suggest two distinct mechanisms for the Ca(1D2) and Ca(3pj) attack to the alkyliodide target.Briefly, whereas the Ca(3pj) reaction proceeds via a collinear attact to the I-C bond, the Ca(1D2) was suggested to react via an insertion into the same I-C bond.These two suggested mechanisms are pictorially depicted in Figure 8.The reader is addressed to Ref. 13  for further details.

LASER-INDUCED BEAM SURFACE REACTIONS [17-20]
A new beam-surface, ultrahigh-vacuum apparatus has been construc- ted in our laboratory.Its main details are schematically shown in  In this new set-up, a molecular beam of SF 6 is directed onto a Ba metal sample.A tunable infrared CO2 laser is focussed on the point where the molecular beam impinges on the barium surface.The typical experiment essentially consists of measuring either the positive or negative ion yield as a function of the laser wavelength.Our basic motivation is to look for beam-surface ionization processes induced by laser radiation.If we find this, we then look for thermally or vibration- ally enhanced effects.These studies are important not only from a fundamental viewpoint, as they may be important in laser catalysis or laser analytical chemistry of polyatomic adsorbed molecules.The reader is addressed to the Faraday Discussion on Dynamics at the Gas Solid Surface (Faraday Discussion no.96, 1993) for a recent account of the research activity in this field.
Figure 10 displays laser ionization time-of-flight (T.O.F.) spectra obtained using the above mentioned technique.As indicated, the solid line represents the negative ion yield (mostly SF-6) obtained with laser excitation at the resonant frequency of 10.6 tm.The broken line shows the same spectrum but using the non-resonant line at 9.6 tm.Note the significant signal enhancement at resonant excitation of the vibrational mode (i.e.10.6 tm) even though the laser Energy Fluence is smaller than that of the 9.6 tm line.See Figure and References 17-20 for further details.The vibrational enhancement of the negative ion yield is now being studied with higher resolution under time-of- flight mass spectrometric conditions.FIGURE 10 Time-of-flight spectra obtained by laser ionization of SF6/Barium inter- face.Solid line represents the negative ion yield (mostly SF6) obtained with laser excita- tion at the resonant frequency of 10.6 lam.Broken line shows the same spectrum but using the non-resonant line at 9.6 gm.The important enhancement at the resonant excitation, even carried out at lower laser energy fluence, should be noticed.
3. LASER ANALYTICAL CHEMISTRY [21][22] In this laboratory two different set-ups are now operating giving the first results.The first of these is a conventional laser ablation spectro- scopy set-up.A dye laser, pumped by the frequency-doubled 532 nm output of a Q-switched Nd:YAG laser, is focussed onto the solid sample to be analyzed.The microplasma emission produced after laser ablation is collected by an optical multichannel analyzer (OMA).
Novel features include the capability of changing a whole set of para- meters to optimize the data analysis.These are laser pulse energy, wavelength and delay time from the trigger pulse until the emission recording starts.In addition, the use of an OMA allows a broad range of emission wavelengths to be monitored simultaneously.This experi- mental set-up already operating in our laboratory has clear advantages with respect to more conventional methods.Some of these advantages are: faster analysis, high resolution and versatility.Preliminary results on A1 samples have proved to be useful in the analysis of trace elements such as Fe, Cu, Mn, Pb, Ti, etc, down to the low ppm range.Figure 11 shows a typical spectrum of the simultaneous detection of Mg, Mn, Fe and Pb in A1 samples using the above mentioned laser-breakdown spectroscopy.Using optimal conditions we were able to determine minor components in the 0.01% concentration range and so, quanti- tative analysis of metallic samples was accomplished with this laser analytical setup.Further details can be found in Refs. 21and 22.
The same laser is now being used in a new two-photon technique dedicated to environmental analysis of (solid) organic pollutants.Essen- tially the technique consists of using a first low energy photon to softly desorb the organic compounds avoiding (molecular) photofragmenta- tion.This vapour ;s then probed using a nitrogen laser, optimizing the time delay between both lasers.The laser induced fluorescence is col- lected and led to an Optical Multichannel Analyzer (OMA) by means of an optical fiber.The experimental arrangement is shown in Figure 12.Some preliminary desorption results for anthracene are shown in Figure 13.It is hoped that this technique will provide a convenient method for analysis of fluorescent pollutants in the environment.Wavelength (nm) FIGURE 13 Laser induced fluorescence spectrum of Anthracene desorbed by a pulse from a Nd: YAG laser.

LASER INITIATED REACTIONS STUDIED BY CHEMILUMINESCENCE
AND LASER-INDUCED FLUORESCENCE (LIF) [23]   The study of the fundamental reactions of the refractive materials is difficult due to the technical problems associated with producing such materials in the gas phase.In this experiment the relatively new tech- nique of laser ablation is applied for this purpose.The basic idea is to use a short intense laser pulse to vaporize the sample in a reactive atmosphere.The reaction of this sample can then be followed by chemiluminescence, or in the case where there is no chemilumines- cence, by laser induced fluorescence.
As an example of such a possibility we present some results obtained for the reaction A1 + O.-products.The aluminum atoms are ablated into the oxygen atmosphere and rapidly oxidized to A10.With high oxygen concentrations a sequence of reactions is possible which produces A10 in an excited state allowing chemiluminescence to be ob- served (see Fig. 14).In order to further study the reaction of this excited A10 with oxygen, we pumped this molecule with an Ar + c.w. laser and measured the fluorescence at different mixtures of 0 2 and N 2, keeping the total pressure at a constant value.This fluorescence is shown in Figure 15.The c.w. laser excites high J rotational transitions.Low J transitions are observed which are not pumped by the excitation laser, indicating that there is rotational relaxation.Evidence for the excited state reaction can be inferred from the decrease in emission from low J levels with an increase in oxygen pressure.This observation can only be satisfactorily explained by invoking an excited state reaction which removes the A10 before full rotational relaxation can occur.Analysis of this body of data has allowed us to derive rate constants for excited A10 reaction with oxygen (See Ref. 23) from different vibrational states.

SPECTROSCOPY AND REACTIVITY
OF VAN DER WAALS MOLECULES [24][25][26][27] This new research project is concentrated on the experimental investi- gation of the transition region of metal-molecule reactions.The rea- gents are condensed within a weakly bound species and the reaction is Wavelength (nm) FIGURE 14 Chemiluminescence spectrum corresponding to the A10 (B X) transi- tion obtained when aluminium atoms are ablated into an oxygen atmosphere.
initiated by optical means.The experimental setup incorporates previ- ous beam and laser spectroscopic developments described elsewhere [24][25][26].The structure and reactivity of van der Waals molecules is a very active field of research.The reader is addressed to Ref. 27 and references cited therein for detailed information about this subject.
van der Waals (vdW) molecules are prepared and analysed spec- troscopically in the experimental arrangement shown schematically in Figure 16.The focalized fundamental frequency of a pulsed Nd-YAG laser running at 10 Hz is used to vaporize atoms from a metal disc.The vapor of the metal is carried by a mixture of helium gas and a solvent, van der Waals molecules between the metal and the solvent are formed as a result of the supersonic expansion that takes place in AOC (_, -> AIO (X) AV-1 the first vacuum chamber.The supersonic jet phases through a skimmer and the molecular beam produced can then be investigated in different ways.First, in previous experiments, a quadrupole mass spectrometer was used for detection.With this scheme, time of flight spectra for various masses, demonstrating unambiguously the presence of van der Waals molecules for the Na expanded with CO2 beams, have been reported [26].Second, the molecular beam can also be probed inside the extraction region of a time of flight mass spectrometer equipped with microchannel plate detection.In this arrangement, the output of a Nd-YAG pumped dye laser is used to excite/ionize the species of interest.The three events, vaporization pulse, gas pulse and excita- tion/ionization pulse are synchronized using a home-made pulse gene- rator.With this setup it is possible to measure mass spectra as a function of the flight time and to perform one or two-color two- photon REMPI (resonance enhanced multiphoton ionization) scans monitoring a selected mass.Photodepletion experiments can also be carried out (see below).
Figure 17 shows the mass spectrum obtained from the ionization with the 4th harmonic of a Nd-YAG laser a molecular beam of ba- rium, CH3C1 and helium optimized for the Ba(CH3C1)2 vdW mol- ecule.The five most abundant isotopes of barium and their van der Mass (amu) FIGURE 17 Time of flight mass spectrum of a beam of Ba, (CH3C1) and helium obtained after ionization with the 4th harmonic of a Nd-YAG laser.
Waals complexes with CH3C1 are detected.BaO +, BaC1 + and BaC12 + are also present in the beam.The barium oxide is coming from the oxidized metal present in the disc and the BaC1 and BaC1/ are likely the reaction products of some of the van der Waals molecules after laser excitation.
A second experiment was also carried out to study the photodeple- tion of vdW molecules as a result of the chemical reaction photoinitiated by laser excitation (Fig. 18).In this case, the second harmonic output of the Nd-YAG laser is split into two beams, one is used to pump the dye laser and the other is doubled to give 266 nm.Both beams are optically synchronized to make sure that the visible light arrives a few nanoseconds earlier to the sample.When the dye laser is tuned to one of the electronic transitions of the van der Waals com- plex the excited state may react leading to different products.Consequently, the Ba(CH3C1), concentration decreases while new species are ionized and detected.This is the case shown in Figure 18.The upper trace corresponds to the mass spectrum recorded when the ultraviolet radiation is interacting with the beam and the dye laser is blocked.In this case, the Ba(CH3C1)2 signal is the most intense.When the dye 28 Time of flight (ts) FIGURE 18 Time of flight mass spectra of a beam of Ba, CH3C1 and helium obtained when excitation is produced with the 4th harmonic of a Nd-YAG laser (upper trace) and when a dye laser tuned to 553.5 nm is used together with the 266 nm beam (lower trace).The photodepletion of the van der Waals complex is clearly noticed.See text for details.
laser, tuned to the barium atomic transition (1S--1P)2 553.3 nm, en- + is ters the detection chamber a decrease in intensity of the Ba(CH3C1)2 clearly observed while the BaC1 + and Ba + increase (lower trace).The Ba ion signal increase is attributed to the resonantly enhanced two- color two-photon ionization whereas the BaC1 + is probably due to the ionization of the BaC1 produced from the photoinitiated reaction within the van der Waals molecule.
Work is now in progress to measure photodepletion spectra of van der Waals complexes in order to get insight into the transition state spectros- copy.In this case, the dye laser will be scanned over a wide range of frequencies while the 266 nm beam is kept for ionization.By varying the dye laser frequency we can access the reactive potential energy surface in various configurations and this will give us information about the reac- tion dynamics as a function of the transition state configuration.
Also, the photoionization threshold of these molecules can be measured.In this experiment a tunable laser is used to photoionize the van der Waals molecules.Starting at a wavelength where ion signal is observed, the wavelength is increased until the point is reached at which the signal disappears.At this point the photons no longer have sufficient energy to ionize the van der Waals molecules.This wavelength represents the adiabatic ionization potential, whose measurement can provide information about the structure of both the neutral and ionized van der Waals molecule.

FIGURE 7
FIGURE7 Normalized reaction cross-section data as a function of the radical size for several chemiluminiscent reactions as indicated.All the data is normalized to for the methyl group.Adapted from Refs. 9 and 13.

Figure 9 .
Figure 9.In this new set-up, a molecular beam of SF 6 is directed onto

FIGURE 8 FIGURE 9
FIGURE 8 Pictorial description of the abstraction (top) and insetion mechanism (bottom) for the Ca(3Ps), Ca(1D2) attack to the RI target.

FIGURE 11
FIGURE 11 Emission spectrum of trace elements in an Al-sample produced by laser ablation spectroscopy.Notice the Mg, Mn, Fe and Pb peaks.See references 21 and 22 for further information.
12..See for exampleRef. 16or a study of the stereodynamics of the Ca* + HBr CaBr* + H reaction based on measurements of the polarization of the chemiluminescence. reaction

Table I
lists ag values for several reactions mostly measured in our laboratory.

TABLE
Absolute values of a R obtained for several chemilumines- Laser induced fluorescence spectrum of A10 for varying partial pressures of O z.The wavelength of the excitation laser is 488.0 nm, which pumps high rotation levels.The presence of emission from low J levels is due to rotational relaxation.