CARBON ISOTOPE SELECTIVE SEPARATION FROM MPD OF CF3Br WITH A TEA COz LASER

monoatomic homonuclear diatomic (H2, D2, N2), polyatomic gases (CF4) and NO have been irradiated with a TEA CO2 laser. Carbon-13 enrichment factor and carbon-13 selectivity have been estimated from the mass spectrometric analysis of the undissociated CF3Br as a function of excitation wavelength, fluence, temperature and pressure. The irradiating wavelength is a crucial parameter to achieve enrichment. The nature of the added gas also affects considerably selectivity. Best enrichment is achieved in mixtures with NO due to its radical-like character.


INTRODUCTION
The process of isotopic selective multiphoton dissociation (MPD) of polyatomic molecules in the gas phase has been amply studied by several authors [1][2][3][4][5][6][7][8][9]. Pure CF3Br and mixtures with different gases have been also studied by MPD [10][11][12][13][14][15]. These studies reveal that the wavelength is the most relevant parameter to attain selectivity. They also show that when pressure is increased, the enrichment disappears and the presence of added gases does not affect substantially the selectivity. *Corresponding author. Fax: 3974187; e-mail: ptablas@bosque.sdi.uam.es We have also studied the MPD of pure CF3Br and mixtures with monoatomic, diatomic and polyatomic gases [16][17][18]. Our results corroborate the effect of wavelength and pressure observed in previous studies, although most of these experiments have been carried out with static cells and sharp focusing, while ours and some others [19,20] have been performed with a long and narrow cell as a waveguide reactor, and constant fluence. Due to these different experimental conditions, the results cannot be compared in detail.
Abdushelishvili et al. [21] have studied the effect of adding NO on the MPD of CF3I. A metallic support bears the cell and allows it to cool to different bath temperatures. A slush bath of heptane/decane/liquid nitrogen allows us to reach 188 K.
Two Gentec detectors mod. 200 are placed before and after the gas cell to measure the input and output energies. The ratio of these measurements reaches a value of 60% in the worst cases. The spot size diameter at the entrance of the cell is approximately 0.45 mm, and has been determined by using the areas of the burn patterns on heat sensitive paper. The cavity is evacuated at 10 -6 Torr. The windows and valves for the gas supply are mounted as close as possible to the ends of the irradiation cell to decrease the non irradiated volume. The relation of both volumes is approximately to 1.
A MKS Baratron which has an output of mV per mTorr is used at the entrance of the cell to control the pressure until Torr. For higher pressures another Baratron is used.
The dissociation products are analyzed with a SXP Elite mass spectrometer from Fisons capable of detecting ppm of masses up to 300 amu. The electrons are accelerated with an energy of 70 eV and the generated ions are positive. The mass spectrometer region is kept at a pressure of 10 -9 Torr. CF3Br (Praxair) and NO (Air Liquide) have been used without further purification.

RESULTS AND DISCUSSION
To .measure the enrichment of the process we use the enrichment factor of the residual CF3Br,/3, defined as: (/149//148) after irradiation (1149/I148 before irradiation where I149 and I148 are, respectively, the mass peak intensities of 13CF9Br and 12CFgBr. Under all the experimental conditions studied, the observed/3 values are smaller than 1, what shows that carbon-13 bearing molecules are preferentially dissociated.
When possible, we have also estimated the selectivity factor, c13, that is a direct measurement of the reaction yield ratio for the two isotopic species of the reactant, and is defined as [15] The results have been obtained as the averaging of at least 3 measurements under the same experimental conditions, and the error bars reflect their statistical deviation.
A. Wavelength Effect Figure 2 shows the enrichment factor, fl, for different irradiation wavenumbers ranging from 1035.5cm -1 to 1053.9cm-1, 9P32 and 9P12 CO2 laser lines respectively, for pure CF3Br. A minimum fl value, maximum enrichment, is reached around 1046.9 cm-1, 9P20 CO2 laser line. The enrichment quickly disappears when the laser is changed to shorter or longer wavelengths. We attribute the carbon-13 selectivity observed to the near resonance of the Ul mode of the 13CF3Br molecule, at 1058cm -1 [22]. There is a difference between the absorption and dissociation maxima of 11 cm -1 towards the red, as usual [1].
As it can be inferred from the absorption spectrum of CF3Br that appears in Figure 3, the absorption coefficient of the Ul mode of 13CF3Br is considerably larger than that of the 3u3 mode of 12CF3Br, what explains the observed selective dissociation of the first isotopic species.
Beta values for some 1:3 (CF3Br + X) representative mixtures are also shown in Figure 2. As it can be seen, the dependence of the observed in the enrichment because of the presence of these monoatomic gases. Therefore, we conclude that they act as spectators, due to the lack of vibro-rotational structure. Figure 4 shows the enrichment factor for the pure substance and 1:3 mixtures with homonuclear diatomic gases: H2, D2 and N2. A remarkable improvement of the enrichment is observed in these mixtures, being most noticeable in those prepared with H2. It seems that diatomic gases relax more efficiently 12CF3Br molecules through collisions, allowing the 13CF3Br species to be preferentially dissociated.
In Table I it is shown the enrichment factor for pure CF3Br and 1:3 mixtures with CF4. No effect is observed on the enrichment when this gas is added, what can be explained by considering that, due to its near resonance with all the isotopic species of CF3Br, it relaxes with similar efficiency both isotopic species: 12CF3Br and 13CF3Br.
C. Effect of NO Addition Figure 5 shows that adding NO to CF3Br instead of H2 improves the enrichment except at the lowest pressure, 0.5 Torr. This improvement is remarkable for total pressures between and 4 Tort.   The different behaviour of NO compared with the other gases studied could be based on its radical-like structure that makes it an excellent trapper of the radical photodissociation fragments.
It is believed that in the presence of NO the following reactions take place [21]: CF3Br + hu --CF3" +Br.
CF3" + NO -CF3NO Br. + NO --BrNO These reactions avoid the following recombination of the fragments: CF3-+Br. CFaBr which contributes to the loss of enrichment in the other mixtures studied. Figure 6 shows the enrichment factor for l: enrichment is achieved for the most diluted sample and for total pressures between 1 and 4 Torr. We believe that the enrichment will improve still further with more diluted samples, but we cannot corroborate it because we are in the detection limit of our mass spectrometer with 1:7.3 mixtures, as it is inferred from the large error bars. In any case, at higher dilutions the number of CF3Br molecules is so small that the efficiency of the process would be rather poor.
We have also estimated the selectivity factor [16], a13, which also reaches a maximum in the pressure range from Torr to 4 Torr, as it is shown in Figure 7. For higher pressures a decreases abruptly and, as usual, the process is not further selective.

D. Effect of Other Variables
We have estimated the enrichment factor for pure CF3Br at different fluences, whose values are affected of a 20% error: 0.94, 1.4 and 1.7 J/cm 2 [17]. It can be seen in Figure 8  decrease when fluence is increased, what means that the enrichment of the process augments with fluence. This trend is observed for most wavelengths, being specially noticeable at those that give higher selectivity values. It seems that, with fluences between 0.94 and 1.7 J/ cm2, maximum possible enrichment is not achieved. The behaviour of the enrichment factor with fluence is similar for all the mixtures studied [18].
The dissociation probability, DP, for the isotopic species 12CF3Br and 13CF3Br at the fluences indicated above has been estimated for nitrogen mixtures. DP increases with fluence for the majority of the wavelengths studied, but the increase observed for 13CF3Br is larger than the one observed for 12CF3Br, what confirms the improvement achieved in the enrichment at higher fluences.
The enrichment factor, /3, at different temperatures has been estimated for pure CF3Br [17], and mixtures [17,18]. In both cases it is observed that the enrichment improves when decreasing the temperature from 295-188 K for the most selective wavelengths, 1046.9 and 1048.7 cm-1, and remains practically unchanged for the other ones studied.
In the case of NO mixtures, we have been able to estimate the selectivity factor, c13, which augments appreciably when temperature decreases [16].
A decrease in the enrichment is observed when total pressure increases from 0.5 Torr to 10 Torr [16][17][18]. This fact can be explained by the enlargement of the number of collisions during the laser pulse when total pressure augments. This increase of collisions favours the ergodic distribution of the absorbed energy and, consequently, entails a loss of selectivity.

E. Efficiency of the Carbon Isotopic Separation Process
In order to estimate the efficiency of the separation process in NO mixtures, we define the following function: where the first factor takes account of the efficiency of the selective dissociation, and the second, of the number of particles susceptible of dissociation.
We consider that the maximum values of CR give the optimum pressure conditions for carbon isotopic separation. Table II shows   2. The addition of a radical trapper like NO improves significantly the enrichment.
3. The MPD process is more selective than it could be inferred from our previous results. The recombination of the radicals formed in the dissociation masks the selectivity of the initial MPD process. The presence of NO avoids this recombination preserving the selectivity of the multiple photon dissociation.
4. Maximum enrichment is achieved in the most diluted CF3Br + NO mixtures (1:7.3). 5. Although lowering the temperature does not influence noticeably the enrichment factor, fl, its effect on the c values is significant.
6. An efficient experimental method for obtaining carbon isotopes has been established.