Stern–volmer plots in the photolysis of perfluoroazoethane
INTERNATIONAL JOURNAL OF CHEMICAL KINETICS, VOL. 11, 191-197 (1970) Stern-Volmer Plots in the Photolysis of Perfluoroazoethane G. 0. PRITCHARD AND W. A. MATTINEN Department of Chemistry, University of California, Santa Barbara, California 931 06 AND J. R. DACEY Department of Chemistry, Royal Military College of Canada, Kingston, Ontario Abstract Nitrogen quantum yields are reported for the photolysis of C,F,N=NC,F, at 3660 A over the pressure range 2-10 cm from 25' to 150'~. The Stern-Volmer plots obtained are discussed and compared with those obtained with azoethane and azoisopropane. 1. Introduction In previous work we have studied the photolysis of hexafluoroazomethane (HFAM) [I ] and perfluoroazoethane (PFAE)  at room temperature. Herein, we report studies extended to other temperatures in examining the effect of pres- sure (@) upon the nitrogen quantum yield (@) at 3660 A in the photodecomposi- tion of PFAE: C,F,N=NC,F, + hv ---f N, + 2C,F, PFAE was chosen as the source molecule because the larger number of available degrees of freedom in it results in a more pronounced dependence of Q, on p , and a re-investigation of the HFAM system has recently been made 131. Such Q, studies have also been made with azo compounds [4, 5 , 6, 71. Perfluoroazo compounds are to be preferred for some aspects of these investigations, since the photolysis mechanisms are less complicated in that radicals do not abstract from the source compound. In azo systems, these reactions, e.g., C,H5 + C,H,N=NC,H, -+ C,H, + C,H,N=NC,H5 lead to considerations concerning the dissociative fate of the new radical formed by abstraction [ 5 , 7, 81 and other complex secondary reactions , For any simple aliphatic azo compound we may write the general photolytic activation-collisional deactivation mechanism, as originally proposed by Weininger 191 0 1970 by John Wiley & Sons, Inc. 192 G. 0. PRITCHARD, W. A. MATTINEN, AND J. R. DACEY and Rice : (1) A + hv+A* (2) A * - + N , + 2R (3) A* + A-+A + A where, in the steady state, (4) W1 = 1 + (k,/kJ[A] From the slope of the Stern-Volmer plot of W1 against p, we obtain k,/k,, and the Arrhenius plot for values at different temperatures may be related [ 5 , 71 to the activation energy, E, , for reaction (2). Also, intercepts greater than unity have been indicated for azoethane (AE) [5, 61 and azoisopropane (AIP) . We discuss the data on PFAE in terms of Equation (4) in the present study. 2. Experimental and Results Experimental details have been given previously [2, 91, except that in these experiments the solution filters were replaced by a Farrand interference filter having a peak wavelength at 3700 A and a half band-width of 85 A. The trans- mitted intensity with the reaction cell under a vacuum ( I : ) , was 8 x 1015 or 2 x 10ls quanta sec-l depending on whether at BTH or a PEK high-pressure Hg-arc was used; values were established with solution actinometry . It was checked in each run, and blank experiments were conducted to check for possible extraneous light in the darkened laboratory. The transmitted intensity (It) was measured in each experiment by solution actinometry when the reaction cell was filled with varying pressures of PFAE. The (D determinations were inde- pendent of changes in the incident intensity (I,,). The experimental results are presented in Figure 1, and it is seen that inter- cepts of greater than unity are suggested, which tend to decrease in value with increasing temperature. The intercepts and slopes (k,/k,) obtained by least squares are 1.33 and 0.393 cm-l at 25"c, and 1.20 and 0.0846 cm' at 15O"c. The limited data at 70 and 105"c give slopes of 0.254 cm-l and 0.202 cm-l with intercepts 1.04 and 1.12, respectively. We have arbitrarily shifted the lines slightly so that the intercepts lie between the 25"c and 150"c values. Converting units gives as the least squares slopes: 7.30 x lo5 (25"), 5.43 x lo5 (70), 4.76 x lo5 (105") and 2.23 x lo5 (150"), all in cc mole-l. Although we assumed an intercept of unity, a least-squares treatment of our previous data  gives an intercept of 1.27 and a slightly smaller slope of 5.23 x lo5 cc mole-l. A possible explanation lies in the fact that the formula used for the absorbed intensity, I, = It - I t , does not allow for the fraction LX of the incident intensity which is STERN-VOLMER PLOTS 193 5 4 @-I 3 I I I I I 0 2 4 6 8 10 Pcm Figure 1. Plot of reciprocal nitrogen quantum yield versus pressure of PFAE. c , 25'~; 0, 70'c; 0, 105'c; 8, 150'~. - - _ , least squares line for data in Bibliography [Z]. a, intercept from CO, data. reflected and scattered back into the reaction cell (presumably mainly from the rear window) and is absorbed, so that I , = I t - It + do . In our original work [ 2 ] we used a Pyrex reaction cell, while in this work we employed a quartz vessel of similar dimensions; different reflectivities and scattering effects for the two experimental set-ups could cause the curves not to be collinear. The intercepts are unaffected and, despite the experimental scatter, the agreement gives some basis for the result that the deviation from unity is real, Cerfontain and Kutschke  estimate an uncertainty of f 10 % in the W1 scale, and this appears to be a minimum degree of uncertainty in this type of experiment. We carried out a number of experiments at constant PFAE pressure ( 3 f 0.05 cm) with varying pressures (2-20 cm) of CO, . Including the additional de- activation step : (5) A* + CO,+A + G O , 194 G. 0. PRITCHARD, W. A. MATTINEN, AND J. R . DACEY Equation (4) becomes : ( 6 ) @-' = + (k3/k2)[A1 + (k6/k2)[C021 The results were scattered, but a least-squares treatment gave an intercept of 2.59 (whichtis included in Figure I ) with a slope k,/k, = 0.226 cm-l or 4.20 x lo5 cc mole-l a t 25"c. This result demonstrates dissimilar efficiencies for de- activation of PFAE* by PFAE and CO, at room temperature. Writing k , = Q3Z3 and k, = Q J 5 , where Q is a quenching efficiency which represents the probability of complete deactivation on collision, and 2 the collision number, we may compute relative values of Q on a collision-for-collision basis. If collision diameters of 8.0 A are assumed for both PFAE and PFAE*, and 4.5 A for C 0 2 , 2, and 2, are almost identical, so that Q , = 0.58 relative to Q3 = 1 .O. PFAE and PFAE* are regarded as unlike species in the calculation of 2, [l]. These relative Q's are, of course, sensitive to the values chosen for the collision diameters. 6.2 6.0 (u Y \ Y" 5.8 2 b, 0 - 5.6 5.4 -d r I I I I 2.4 2.6 2.0 3.0 3.2 lo3/ T o K Figure 2. Arrhenius plots for k,/k,. 0, AIP (Bibliography ) (k3/k2 values X 2) . 8, AE (Bibliography ); 0 (Bibliography [S]) (k3/k2 values X 10). 0 , PFAE (this work); (Bibliography "21). STERN-VOLMER PLOTS 195 Experiments at 150" yielded 0 = 0.475 f 0.25, and no useful conclusion was obtained concerning the slope. An Arrhenius plot of log,, (k,/k,) (units: cc mole-l) against T-~("K) results in E , - E, P 2.0 kcal mole-l, and a similar result was obtained in both the AE  and the AIP  system. However, in all three cases the curves show distinct and similar curvature (see Figure 2). 3. Discussion A. Intercepts Much of the work on azo-photolysis has lead to reports of intercepts that are greater than unity [5-7, this work] although in most cases it is possible to draw quite reasonable "eye-ball" lines through unity to represent the data, particularly if a f 10 % variation in the 0-l scale is assumed (note also the differences in the ordinate scales). There is no way, however, that the data on AIP at 376" and 4 0 0 " ~  can be taken to have a unity intercept (values are 1.72 and 1.75, re- spectively), and, on balance, we conclude that Equation (4) is not obeyed exactly in all cases. For instance, although in their recent report on HFAM photolysis Wu and Rice [ 3 ] arbitrarily assume an intercept of unity, a least-squares treatment of their data up to 4 and 10 cm, between which pressures the onset of curvature occurs as indicated by the >10 cm data, yields intercepts of 1.09 and 1.13, respectively. The data on AE do not show curvature up to 20 cm [5 , 61 but, on the basis of the HFAM study, Wu and Rice pertinently suggest that curvature may well occur at higher pressures, or that the intercepts suggest the possibility of curvature at very low pressures, as is the case with hexafluoroacetone [lo]. Similar considerations apply to the present data on PFAE. Possible curvature due to the breakdown of the strong-collision assumption is presumed not to occur [lo, 111, and in any case multi-stage collisional deactivation would lead to zero slope at zero pressure [l 1, 121. The simplest explanation of the intercepts was originally adopted by Kutschke and coworkers [ 5 , 71, who suggested an internal conversion process : (7) A * + A where A is incapable of dissociation. This reaction can also presumably represent an intersystem crossing, provided that it is not collision-induced . There is no possibility of wall deactivation at the pressures employed . Such a non-dissociative process is included in Porter's generalized scheme [ 121 and, neglecting dissociation of the triplet state, we have : which adequately explains the linearity of the Stern-Volmer plots and the inter- cepts obtained. Data on ketene also conform to Equation (8) [12, 131. The 196 0. 0. PRITCHARD, W. A. MATTINEN, AND J. R. DACEY values of the intercepts are very similar for AIP and PFAE at room temperature, being 1.35 and 1.30, respectively. Riem and Kutschke  interpret the AIP system as an activation-energy difference between the two processes, (7) and (2) with E, - E, = 2 kcal mole-l. Our data on PFAE suggest a lesser and opposite temperature-dependence. The magnitude of the intercepts is important in view of Bowers analysis of ketene photodissociation , which he has recently extended to AE , using the Rice-Ramsperger-Kassel-Marcus formulation. The dissociating molecules have a vibrational-energy distribution which reflects the thermal-energy distribution of the ground state and the energy spread in the absorbed light. A slight curvature results in the Stern-Volmer plots, which would be difficult to determine experimentally. In the AE calculations the limiting low- and high- pressure slopes differ by only 10 %, and an intercept of unity is indicated . A similar conclusion was reached by Worsham and Rice [S]. B. Arrhenius curvature If k, is calculated from collision theory the temperature-dependence is slight, and the activation-energy difference, E, - E , , has been regarded as an activa- tion energy for the decomposition of the excited molecule, A* . Bowers and Porter [lo] have pointed out that the curvature is due to the fact that k, reaches a limiting value at low temperatures determined by the wavelength, and a minimum value of E, is obtained by taking the slope in the high-temperature limit where thermal effects are more dominant. From Figure 2 it is seen that E,(min) > 5 kcal mole-l. Steel and coworkers  have summarized the possible mechanisms of aliphatic azo photodecomposition and commented on the varying role of multi- plicity in azo photochemistry. For the compounds discussed in this paper, predissociation occurs from the vibrationally excited upper singlet state. t Wor- sham and Rices calculations  substantiate this, and Bowers more rigorous treatment indicates a critical dissociation energy of about 10 kcal mole-l for AE, which is not in disaccord with the trend in Figure 2 ; in fact Bowers calculations reproduce the temperature-dependence fairly well [ 141. Acknowledgements We gratefully acknowledge support of this research by the National Science Foundation under Grant GP-8069 and by the Defence Research Board of Canada, under Grant 9530/13. We thank Professor P. G. Bowers for a prepublication copy of his paper on azoethane, and for helpful discussion. t I t should also be noted that the decomposition of aliphatic azo compounds has been induced by quenching of the triplet states of acetone and biacetyl . STERN-VOLMER PLOTS 197 Bibliography [ 11 J. R. Dacey, R. F. Mann, and G. 0. Pritchard, Can. J. Chem., 43, 32 15 (1965).  J. R. Dacey, W. C. Kent, and G. 0. Pritchard, Can. J. Chem., 44,969 (1966).  E.-C. Wu and 0. K. Rice, J. Phys. Chem., 72, 542 (1968).  J. L. Weininger and 0. K. Rice, J. Am. Chem. SOC., 74,6216 (1952).  H. Cerfontain and K. 0. Kutschke, Can. J. Chem., 36, 344 (1958).  W. C. Uorsham and 0. K. Rice, J. Chem. Phys., 46, 2021 (1967).  R. H. Riem and K. 0. Kutschke, Can. J. Chem., 38,2332 (1960).  0. P. Strausz, R. E. Berkley, and H. E. Gunning, Can. J. Chem., 47, 3470 (1969) ; see also  G. 0. Pritchard, J. R . Dacey, W. C. Kent, and C. R. Simonds, Can. J. Chem., 44,171 (1966). H. S. Sandhu, J. Phys. Chem., 72, 1857 (1968). [lo] P. G. Bowers and G . B. Porter, J. Phys. Chem., 70, 1622 (1966). [ 111 G. B. Porter and K. Uchida, J. Phys. Chem., 70, 4079 (1966).  G. B. Porter and B. T. Connelly, J. Chem. Phys., 33, 81 (1960).  P. G. Bowers, J. Chem. SOC., A, 466 (1967).  P. G. Bowers, J. Phys. Chem., 74, 952 (1970).  B. S. Solomon, T. F. Thomas, and C. Steel, J. Am. Chem. SOC., 90,2249 (1968).  R. E. Rebbert and P. Ausloos, J. Am. Chem. SOC., 87, 1847 (1965). Received March 12, 1970.