• Introduction
• Experimental details
• Results and discussion
• Conclusion
• Acknowledgments
• References

FIGURES

• Scheme 1
• Figure 1
• Figure 2
• Figure 3
• Figure 4

Due to the replacement of CFCs by the less harmful HFCs, a huge body of information about the chemistry of these fluorinated compounds has been published and is currently of paramount importance. The radicals produced by these molecules involve mainly the families known as "CF₃O_x" and "FCO_x". The FCO radical is thus involved in many degradation reactions.

CF₃ and FCO radicals are obtained when CF₃COF is photolyzed. Both radicals react to give C₂F₆, CO, CF₂O and also regenerate the precursor molecule thus giving a quantum yield of CF₃COF disappearance lower than unity (f =0.4).

CF₃CF₂C(O)F is a suitable source to generate the title radicals allowing to determine whether a longer chain can modify either the photolysis mechanism or the rate constant with respect to the CF₃COF.

We performed the photolysis of CF₃CF₂C(O)F either pure or in the presence of c- C₆H₁₂ and (FCO)₂ (oxalyl fluoride), following the concentration of the different species by FTIR spectroscopy.

In the photolysis of the pure precursor, the products formed were C₄F₁₀, CO and CF₂O in accordance with the case of CF₃C(O)F.

When enough ciclohexane is added, the radicals formed in the initial step (C₂F₅ and FCO) are trapped and the recombination reactions between them are suppressed. It is observed that the rate of reactant consumption is faster than without c-C₆H₁₂. This suggests that in the photolysis of pure CF₃CF₂COF, the reaction between C₂F₅ and FCO radicals plays a role. This was corroborated by the photolysis of CF₃CF₂C(O)F in the presence of (FCO)₂ (i.e. in presence of excess of FCO radicals). Here again, we observed the variation in the rate of disappearance of CF₃CF₂C(O)F.

The rate constant of the reaction C₂F₅ + FCO ® CF₃CF₂COF was obtained through a simulation using both, our experimental data and bibliography data available. The value found is similar, within experimental error, to the rate constant obtained for CF₃ and FCO radicals; therefore the length of the chain does not affect the mechanism.

Introduction

In the last teen years, it has been of great interest to know the impact of the substitutes of CFCs and their degradation products on atmospheric chemistry. In fact, the radicals CF₃O_x and FCO_x originated from the decomposition of HFCs are involved in many reactions in the higher atmosphere.

In previous works, the reaction between CF₃ and FCO radicals,^1,2 obtained when CF₃COF is photolyzed was studied.. Both radicals react to give C₂F₆, CO, CF₂O and also to regenerate the precursor molecule, thus giving a quantum yield of CF₃COF disappearance lower than unity (f =0.4). Similarly, the quantum yield of CF₃CF₂COF photolysis gave the same value as that for the CF₃COF,³ but there were no indication about the kinetics of the reaction.

Taking into account the above results, we performed the photolysis of CF₃CF₂COF alone and in the presence of either c-C₆H₁₂ or (FCO)₂ in order to know the mechanism which is taking place and to determine the C₂F₅ and FCO radical recombination rate.

Experimental details

1. Materials

Commercially available samples of perfluoropropyl fluoride (CF₃CF₂C(O)F) (P.C.R. Research Chemicals) were trap-to-trap distilled in vacuum before use and their purity was checked by IR spectroscopy. Cyclohexane (98%) was used as received. Oxygen was condensed by flowing O₂ at atmospheric pressure through a trap immersed in liquid air. It was then pumped under vacuum several times and transferred to a glass bulb whilst the trap was still immersed in liquid air. Oxalylfluoride (FCO)₂ was synthetized by fluorination of oxalylchloride with NaF in sulfolane as solvent and purified by trap-to-trap distillation⁴.

1. Procedure

Reactants and products were manipulated in a conventional high-vacuum system. The photolyses were carried out using low pressure Hg lamps surrounding a quartz cell fitted with KCl windows which was located in the optical path of a Fourier transform IR (FTIR) spectrometer which was used to follow the evolution of the reaction with time ( see Scheme 1).

Scheme 1

Results and discussion

The comparison between the photolysis rate of CF₃CF₂C(O)F alone and with either c-C₆H₁₂ or (FCO)₂ was performed using the setup described above. When CF₃CF₂C(O)F alone was photolized, the products formed were C₄F₁₀, CO and CF₂O (Fig. 1), in agreement with the results of Weibel et al.³ These products could be explained taking into account that the C-C bond breaking gives CF₃CF₂ and FCO radicals:

The CF₃CF₂C(O)F and CF₂O quantification wereperformed using calibration curve for each pure substance.

Fig. 1: IR spectra obtained for the photolysis of pure CF₃CF₂COF at 254 nm.

The experiments carried out in the presence of c-C₆H₁₂ gave CF₃CF₂H andHFCO as products. In these experimental runs it was not observed C₄F₁₀ or CF₂O formation.This fact indicates that all the CF₃CF₂ and FCO radicals formed reacted with the c-C₆H₁₂ molecules:

CF₃CF₂ + c-C₆H₁₂ ® CF₃CF₂H + c-C₆H₁₁ (2)

FCO + c-C₆H₁₂ ® HFCO + c-C₆H₁₁ (3)

When comparing the CF₃CF₂C(O)F photolysis rate alone and in the presence of c-C₆H₁₂(Fig. 2), it can be seen that the CF₃CF₂C(O)F amount reacted is greater in the second case. This observation suggests the CF₃CF₂ and FCO radicals react among them in a recombination reaction when there is no other gas added:

CF₃CF₂ + FCO ® CF₃CF₂C(O)F (-1)

Fig. 2: Plot of ln(moles CF₃CF₂COF) vs. photolysis time. (", with (FCO)₂ added; $, alone; %, with c-C₆H₁₂ added). The lines correspond to the least squares fitting to the experimental points

To corroborate this possibility we carried out a new CF₃CF₂C(O)F photolysis in the FCO radicals presence using a good FCO source such as oxalyl fluoride, (FCO)₂. In this case, the C₄F₁₀ formed as well as the reactant consumed are smaller than when CF₃CF₂C(O)F is photolysed alone. In Fig. 2 it can be seen the photolysis rate and in Fig. 3 are shown the spectra obtained in this new condition.

Fig. 3: IR spectra obtained for the photolysis of CF₃CF₂COF in the presence of (FCO)₂ at 254 nm.

In accordance with the results presented previously it is clear that the presence of an effective radical scavenger, able to trap radicals that participate in reactions (2) and (3), allows the reaction photolysis rate determination (1).

Two photolytic experiments were performed with initial CF₃CF₂C(O)F concentration of 4 x 10¹⁶ molecule cm^-3 in absence and in presence of c-C₆H₁₂ (90 torr). Photolysis time was 1200 s, to maintain low conversions. A comparison of the CF₃CF₂C(O)F concentration variation between the two runs is presented in Fig. 2. Note that the CF₃CF₂C(O)F disappearance is faster in the last case. From its slope (dotted line in Fig. 2), we obtain the CF₃CF₂C(O)F photolysis rate value of (1.4 ± 0.1)x 10^-4 s^-1.

The mechanism proposed after the C-C bond rupture, when the photolysis is carried out in c-C₆H₁₂ absence, is:

CF₃CF₂ + FCO ® CF₃CF₂C(O)F (-1)

CF₃CF₂ + CF₃CF₂ ® C₄F₁₀ (3)

FCO + FCO ® CF₂O +CO (4)

The k₃ and k₄ rate constants were obtained from literature. In the case of k₄, there are four values reported but only the work of Behr et al.⁵ was carried out at pressures comparable to those used in our experiments, (although independent) so we took the value (1.7 ± 0.3) x 10^-11 cm³ molecule^-1 s^-1. For k₃the value informed is 2.85 x 10^-12 cm³ molecule^-1 s^-1.⁶

With the mechanism just proposed, we simulated the time variation of the CF₃CF₂C(O)F and CF₂Oconcentration. The experimental and simulation results are shown in Fig. 4. The best fiitting of the k_-1 value gives (6.8 ± 0.8)x 10^-12 cm³ molecule^-1 s^-1. This value of the recombination rate constant also explains why, in the CF₃CF₂C(O)F photolysis in presence of oxalylfluoride, the C₄F₁₀ concentration drops so efficiently and the CF₃CF₂C(O)F concentration decreases much less than when it is photolysed alone.

Fig. 4: Fitting to the actual concentration of species ($, CF₃CF₂COF; Å, CF₂O) when the value assumed by k_-1 in the simulation is 6.8 ± 0.8)x 10^-12 cm³ molecule^-1 s^-1.

Conclusion

In the study of the CF₃CF₂C(O)F photolysisin the presence of c-C₆H₁₂, an increase in the photolysis rate of the reactant is observed. In our experimental conditions k₁ was (1.4 ± 0.1)x 10^-4 s^-1.

When (FCO)₂ is added, an excess of FCO produces a reduction of the CF₃CF₂C(O)F photolysis rate. The two former conclusions are compatible with a mechanism mediated by radicals.

The low quatum yield for the CF₃CF₂C(O)F disappearance and the results previously presented suggest that CF₃CF₂ and FCOradical recombination reaction is taking place and the calculated rate constant obtained for this recombination was (6.8 ± 0.8)x 10^-12 cm³ molecule^-1 s^-1.

The value found is similar, within experimental error, to the rate constant obtained for CF₃ and FCO radicals. This indicates that the length of the chain of the perfluoroalkyl radical does not affect the rate constant nor the mechanism.

Acknowledgments

We are indebted to CONICOR, CONICET and SECyT-UNC for financial support.

References

1-D.E. Weibel, C.M. Vöhringer, E.R. Staricco, E.H. Staricco, J. Photochem. Photobiol. A: Chem. 63 (1992) 1.

2-K.L. Bierbrauer, M.S. Chiappero, F.E. Malanca, G.A. Argüello, J. Photochem. Photobiol. A: Chem. 122 (1999) 73.

3-S.A. Cariati, D.E. Weibel, E.H. Staricco, J. Photochem. Photobiol. A: Chem. 123 (1999) 1.

4-C.W. Tullock, D.O. Coffman, J. Org. Chem. 25 (1960) 2016.

5-P. Behr, H. Heydtmann, Phys. Chem. Chem. Phys. 100 (1996) 553.

6-G.A. Skorobogatov, V.G. Seleznev, O.N. Slezar, Dokl. Acad. Nauk. SSSR 231 (1976) 1407.