Reaction Dynamics of Electronically Excited Calcium Atom

Absolute values of the total chemiluminescence cross-section for the beam-gas Ca(3P, 1D) + Cl4C → CaCl(A, B) + Cl3C and Ca(3P, 1D) + SF6 → CaF(A, B) + SF5 reactions have been measured at low collision energy, E T = 0.15 and 0.14eV, respectively. Both metastable atomic calcium states Ca(3P, 1D) were produced under low voltage dc-discharge conditions. By changing the discharge conditions, different metastable concentrations were produced to measure the state-to-state cross-section for both 3P and 1D reactions. The following values for the total chemiluminescence cross-sections were obtained: σ D 1 = 1.77 A and σ P 3 = 0.25 A for the Ca(3P, 1D) + Cl4C → CaCl(A, B) + Cl3C reaction. σ D 1 = 0.59 A2 and σ P 3 = 0.56 A2 for the Ca(3P, 1D) + SF6 → CaF(A) + SF5 reaction. σ D 1 = 0.04 A2 and σ P 3 = 0.12 A for the Ca(3P, 1D) + SF6 → CaF(B) + SF5 reaction.In addition, beam-beam experiments were carried out at the same average low collision energy that of the beam-gas, and therefore, normalization between both experiments was possible. This procedure allowed us to obtain the excitation function of the Ca(1D) + SF6 reaction in absolute values over the 0.15–0.60eV collision energy range.On the other hand, by simulation, the ratio of CaCl(B-X/A-X) emissions intensities was found to be of 0.15. The variation of this ratio with the relative concentration of 1D/3P in a Broida oven leads to the conclusion that this state favours the formation of the B state in the chemiluminescent Ca(3P, 1D) + CH3CHCl2 → CaCl(A, B) + CH3CHCl reaction.

In particular, electronic selectivity of the Ca(3p, 1D)reactions has been studied with different molecules as SF 6 10and N20. 2 An important question arises in those reactions where one may have different electronically excited reactants as responsible for the observed chemiluminescence. Two methods can be applied to solve this question: (a) the analysis of the pressure dependence of the chemiluminescence cross-section 9 or (b) the analysis of the chemiluminescence emission as a function of the metastable concentration which for the two metastable calcium atom states (3p, D) has been termed the two state model analysis.
The present paper is an extension of this work to the Ca* + RX CaX* + R(X C1, F; R C13C SF5, CHaCHCI reactions, where both metastable Ca(3p, D) reactants concentrations were varied by using a low voltage discharge method. 3 We anticipate that one of the main results of the present work is the absolute determination of the reaction cross-section for both electronically excited reactions which indicates a clear selectivity favouring the D reaction with respect to the Ca(3p) one. In addition, normalization of the chemiluminescence yields between beam-gas and beam-beam experiments was carried out to obtain these excitation functions in absolute values. EXPERIMENTAL Part of present experiments were carried out in a beam-gas or beam-beam arrangements using our molecular beam apparatus described elsewhere. 9-1 Therefore, only a brief description is given here. The atomic calcium oven consists of a stainless steel heated oven where ground state Ca atoms can be excited to metastable 3p and D by a low voltage discharge, whose design can be found elsewhere. 2 A typical oven charge is of the order of 5 g. A 300 A and 3 V current was used to vaporize the Ca metal to the desired temperature. The Ca charge is typically maintained at 1150 K corresponding to a Ca vapor pressure of 5 Torr, while the source oriffice (0.5 mm in diameter) is at 1350 K (corrected in pyrometer readings). This temperature gradient was achieved by reducing the thickness of the upper part of the heater from 0.25 mm (lower half of the heater) to 0.18 mm (upper half of the heater). Under these conditions, runs of 12 hours can be easily achieved. Moreover, a discharge between the heater and the crucible was pulsed using a few microseconds pulse at ca. 2500 Hz from a Lyons Instruments pulse generator (mod. PG 75 A).
The calcium beam passed into a scattering cell of 24 mm total length through a collimating hole (7 mm in diameter), located 80 mm from the viewed source, containing flowing C14C or SF 6 vapor and chemiluminescence was observed perpendicular to the beam at a distance of 90 mm from the source. Typical background pressures in the source chamber were 10 -6 Torr and pressure never exceded 2.10-5 Torr when the gas was present in the reaction chamber. (2) m g inca # (4) mg + mca and 0g(g C14C or SF6) and 0tc the most probable velocities of the gas cell and the calcium beam, respectively, given by 0q= (i CI,, SF6, Ca (5) \m/ For the present experiment, a maxwellian (thermal) Ca beam was assumed. Previous studies 12 using the same oven and experimental conditions show by time-of-flight determinations that the metastable calcium beam we used has a near maxwelliam velocity distribution. The average collision energy under the present experimental conditions was found to be 0.15 and 0.14eV for the Ca* +C14C and Ca*+ SF 6 collisions, respectively.
In a second experimental procedure, we used a Broida oven to produce calcium vapor. In the apparatus already described, 14 calcium is heated in a graphite crucible by a tungsten wire; a cw discharge of 150 V between the crucible and the ground produces metastable 3p and 1D states which are carried together with ground state Ca by flowing helium at a pressure of [1][2][3][4][5] Torr with a fast pumping (roots 350m3/h).
The atoms are mixed in the reactive zone, which is 40cm above the crucible, with the halogenated compound. microcomputer. In these experiments, the average speed is almost thermal (T 400 K, measured by a thermometer), slightly increased by the gas velocity, 25 m/s. Table 1 lists the most relevant and typical conditions of both experiments.

RESULTS AND DISCUSSION
Ca* + Cl4C and CH3CHCI2 Systems Figure shows the energy levels of the products and reactants for the Ca* + C14C reaction. Note that the two chemiluminescent channels (i.e. the A and B states of the CaC1 product) are exoergic only if metastable atoms are used, i.e. the process in endoergic for ground state cacium atoms.
In addition, the adiabatic correlation diagramm of the Ca + C14C CaC1 + C13C system using the C group of symmetry is also shown in the same figure. It is interesting Energy levels for the Ca + CI,CaCI + C13C reaction. Energy units in eV. The adiabatic correlation between reactants and products is also shown using the Cs group of symetry. REACTION DYNAMICS OF Ca* 127 to point out that whereas the Ca(1D) state correlates with both CaC1 A and B states, the Ca(3p) only does with the CaCI(A2I-I) state. Figure 2 shows the (low-resolution) chemiluminescence spectrum together with background spectrum when the discharge was turned off. Emission from the two electronic states A and B of CaC1 corresponding to transitions B2E + XZE + and A2H xzE + are observed. The spectral shape was found to be independent of C14C pressure up to 8.10-a Torr, showing that single-collision conditions applied for the present experiment. Figure 3 shows the simulation of the A-X and B-X allowing the deconvolution of the experimental spectrum and giving a ratio of B-X/A-X intensities, R 0.15. Figure 4 shows the chemiluminescence spectrum obtained with the Broida oven for the Ca* CH3CHC12 reaction. Besides the A-X and B-X(Av 0) emission which are separated, one observes atomic Ca transitions coming both from energy pooling and reabsorption from the discharge in the reaction zone. In the same figure is shown the A-X transition deconvoluted from the 53S-43p line. This deconvolution was used to measure the variation of the ratio R with the discharge current.
As mentioned before, an important question arises when one considers which state i.e. Ca(1D) or Ca(3p) is responsible for the observed chemiluminescence. To solve this question two methods can be applied: (a) the analysis of the pressure dependence of the chemiluminescence cross-section 9 or (b) the analysis of the chemiluminescence emission as a function of the metastable concentration; the so-called two-state analysis model. In the present work, we have applied the later procedure. Essentially, it consists in measuring the product chemiluminescence as a function of both metastable emission intensities. Under our experimental conditions, it was very difficult to isolate the chemiluminescence yield from both A and B states mainly because the low resolution and the spectral overlapping of the bands. Therefore, we applied the two state model analysis to the entire chemiluminescence yield i.e. the B and A states. following reference, t one can write for the total CaCl emission intensity, lca, the following relation Icacr rIcI'cVR A3p" Fa= A'D where Ii, ai, A are the particular (i ap, 1D) metastable emission intensity, relative cross-section and atomic transition probability, respectively, and noc and ' stand for the number density of the CI,C and relative velocity of the reactants. Fs= is the fraction of ap in the J 1 level, found experimentally equal to 0.3.

Equation (6) can be rearranged as follows"
Icacl" A,Dare known to be 0.44 ms and 25 ms, respectively. 1 Since we found from laser induced fluorescence measurements that F s= 0.3, the following values for the total chemiluminescence reaction Ca* + C14C CaCI(A, B) + C13C were obtained tr, i 1.77 A 2 and trp 0.25 A 2 from the linear plot of Figure 5. The difference in the reaction cross-section indicates a clear electronic selectivity in the chemiluminescence channel although more beam-beam experiments changing the collision energy, as well as the electronic excitation would be necessary to confirm this feature of the reaction dynamics.
A similar procedure was used with the Broida oven to check the electronic selectivity observed in the beam-gas experiment, but now for the Ca* + CHaCHCI2 reaction.
First, we measured the intensities ratio of the 4pP-31D/53S-4aP lines observed in the reaction zone at 6717A, and 6102/, 6122A and 6162A respectively. The same transitions were also recorded in the discharge itself with a small monochromator and the response of the two monochromators adjusted by using a commercial Calcium-Neon hollow cathode. Assuming that the upper states of the transitions are formed by absorption of resonance photons coming from the discharge and taking into account the radiative lifetime of both 1D and 3p states, after correction of the photomultiplier responses, the ratio (1D/aP) is found to be p =0.31 at maximum current in the cw discharge (150 ma).
In a second procedure, we have measured the intensities ratio of the two forbidden transitions 1D-1S and 3p-1s at 4575 A and 6572 A respectively. Taking into account the radiative lifetimes again, and after photomultiplier response correction, we found 13p / I o / 10 2 Figure 5 Product chemiluminescence IQ reduced by/1c14c 'R III as a function of I3p/Itt for the Ca* + C14C reaction. This is equation (7) type of plot from which slope and intercept both reaction cross-sections were obtained (see text for comments).  Figure 6 CaCI* B-X/A-X intensity ratio, R, as a function of the 1D/aP intensity ratio obtained by changing the discharge current. The data is obtained from the Ca* + CHaCHC12 reaction. p 0.10 at maximum current. This is not the value obtained by the first procedure, indicating that we have to take into account other mechanisms like energy pooling for the formation of excited species.
We then measured the variation of p with the intensity of the discharge current, and the variation of R with the current. The results are shown on the Figure 6. One can clearly see that the B-X intensity increases with 1D concentration, which confirms the electronic selectivity found in the beam-gas experiments.
Ca* + SF 6 System Figure 7 shows the low resolution spectrum together with background spectrum when the discharge was turned off. The low-lying electronic states of CaF and available reaction energies in the Ca* + SF 6 can be found in Ref. 9 and 10. Emission from the two electronic states A and B of CaF corresponding to transitions B2E /and A2H X2E + are observed. No emission from the C-X band (/max "--330.9nm) was observed under the present experimental conditions. As in the previous case, the spectral shape was found independent of SF 6 pressure over the 0.8-8 mTorr range, showing that single-collision conditions applied for the present experiment.
Absolute chemiluminescence cross-section were obtained by the same procedure outlined above, e.g. changing the metastable density ratio. Linear plots of the form described by equation (7) are shown in Figure 8 for both A and B bands of CaF.
Indeed, Table 2 summarizes the present absolute determinations as well as a previously reported data on similar reactions.   (11) h D where the prime stands for the cross-sections value o"i (12) i.e. the formula used in Ref. 9 where the pressure dependence method was used. After inspection of the above equations, one can realize that in those situations when n,o/n3p << 1 both the new (present results) and the old value of the reaction cross,section are almost identical (very different) for the Ca(3p) reaction (Ca(1D) reaction).
Furthermore, in the case when n3r,/nt >> 1, one can expect a much higher uncertainty in extracting tr, D from the pressure dependence method. These considerations could justify the conflict between the present and old values while they indicate that the present values based on a direct change of both metastable concentration should give more reliable data.

CONCLUDING REMARKS
This paper centers its attention on the reaction dynamics of electronically excited alkaline earth atoms using two different but complementary techniques, e.g. the beam-gas or beam-beam method and the Broida (bulk) method.
An important common feature of both approaches is the experimental method to change the metastable ratio of the atomic calcium. By changing the discharge conditions, one is able to modify the 3P/1D metastable ratio and therefore to isolate the electronic selectivity associated to each metastable reactants. This procedure has been carried out in both techniques giving consistent results about the electronic energy dependence (el.ectronic selectivity) of the total reaction cross-section in the Ca(3p, 1D) + RCI(R CIaC', CHaCHCI') CaC1 + R. chemiluminescent reactions.
On the other hand, another interesting aspect of the present work is the use of beam-bas experiments, where the density gas is known, to normalize beam-beam reaction cross-section data. This procedure originally provides excitation function data of electronically excited species in absolute values, as shown for the Ca( D) + SF 6 CaF* + SF reaction. 136 E. VERDASCO ETAL.