118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure.This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point.These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order. The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility.The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13.118 ELECTRICAL THEORY OF ADBORPTTON The writer considers the double layer as consisting of a swface of rigidly fixed atoms under continuous bombardment of positively and negatively charged ions, any particular point on the rigid surface becoming in turn negative, neutral and positive, these conditions arisdg in any order.The observed contact difference is the average effect of these conditions. Where several kinds of atoms are present in the solution the average number of any one of them at the surface will depend on their concentbration, valency and mobility. The variation of contact Werence from negative to neutral and positive was observed with cotton and aluminium sulphate near the neutral point. These variations occurred during the same experiment, the readings being direct measurements of E.1I.F.s developed by filtration under pressure. This point would be covered by putting n2 = 1 and = 2 or 3 in Mukherjee’s equation No. 13. Absorption Spectrum of Triplet Benzene BY T. S. GODFREY AND G.PORTER Dept. of Chemistry, The University, Sheffield 10 Receiced 12th August, 1965 The absorption spectrum of triplet benzene has been observed by flash photolysis studies of benzene in rigid hydrocarbon glasses at 77°K. Although the triplet-triplet absorption spectra of many organic molecules have been observed by flash photolysis methods, attempts to detect the corresponding ab- sorption in benzene have hitherto been unsuccessful.1 The absorption of these triplet states has been invoked to interpret biphotonic processes which occur with benzene and its derivatives, particularly the sensitized decomposition of solvent at 77°K reported by Voevodsky and collaborators 2 and also the photo-ionization and thermoluminescence of benzene derivatives in hydrocarbon glasses2 These observations suggested that benzene should have an observable absorption in the region of 2537& in accordance with the theoretical prediction of Pariser.4 We have therefore re-investigated the flash photolysis of benzene, using low-temperature glasses in which the triplet lifetimes were known to be several seconds, and we have successfully recorded the benzene triplet absorption spectrum in the region expected. EXPERIMENTAL The general technique of low-temperature flash photolysis has been described pre- viously 5 but some modifications have recently been made. One of the main difficulties in previous work was the poor transparency of hydrocarbons at low wavelengths when frozen to a rigid glass.However, the following method of purification and drying of isopentane and 3-methylpentane 69 7 resulted in glasses which were transparent down to 225 mp in a 23 cm quartz cell.The hydrocarbon was treated with concentrated H2SO4 and con- centrated HNO:, for at least a week. (Small quantities should be used, especially if the starting material is not pure, and precautions should be taken against the possible risk of explosion.) It was washed with concentrated H2SO4 and then with water, dried with CaC12 followed by CaS04, refluxed over sodium overnight, passed through about 9 ft of activated silica gel and stored under vacuum. Immediately before use it was distilled on to a fresh sodium mirror and, from there, into a reservoir attached to the reaction cell. Benzene (B.D.H. “ for molecular weight determinations ” grade) was purified by vapour phase chromatography, weighed, dissolved in a little isopentane and then also distilled on to the sodium mirror.The solution was degassed by the usual freeze, pump, thaw, shake technique. When flashing, the reaction cell was held rigidly in the bottom of the Dewar by a seal of 50 % isopentane and 50 % 3-methylpentane. One of the benzene solutions flashed, containing 100 % 3-methylpentane as solvent, had to be cooled very slowly to prevent the hydrocarbon glass from cracking. This same solution also tended to crack when flashed with 900 J but was stable when only 225 J were used. However, most of this work was done with solvents containing about 65 ”/, isopentane and 35 % 3-methylpentane which could be cooled rapidly and flashed with 1000 J without cracking.All solutions were cooled for at least 30 min before flashing. The spectrum of triplet benzene was recorded on a photographic flash apparatus.8 This had two vertical flash lamps connected in series to condensers of 9pF capacity and 78 ABSORPTION SPECTRUM OF TRIPLET BENZENE charged to 15 kV. The analysis flash came from a horizontal capillary lamp and a shutter, opening 2 msec after the photolysis flash, cut out scattered light and fluorescence. A Joyce Loebl microdensitometer was used to trace the spectrum. T he kinetics of absorption and phosphorescence decay were measured on a photoelectric flash apparatus.9 Recently the zirconium arc monitoring source in this has been replaced by an iodine quartz lamp (Atlas A1/125, 12 V, 100 W) which emits down to about 220 mp, mwh lower than the zirconium arc.The apparatus also contained a grating mono- chromator and a R.C.A. 1P 28 photomultiplier. An Aminco-Kiers spectrophosphorimeter was used for some phosphorescence measurements. RESULTS ABSORPTION SPECTRUM OF TRIPLET BENZENE The spectrum was recorded photographically 5 msec after a 1.7 x 10-4 M solu- tion of benzene in 65 % isopentane and 35 % 3-methylpentane at 77°K had been flashed with lOOOJ, and was measured by comparison with the spectrum before flashing. In this way the hexatriene derivative, which is one of the products of benzene photolysis and absorbs in the same region as the triplet,lo had a negligible effect on the result. Neutral optical density rhodium filters were used for calibration of plate densities.I I I 220 2 be) 3 0 0 340 0.0 I wavelength (mp) FIG. 1 .-Triplet benzene absorption spectrum. The spectrum which is shown in fig. 1 has no structure and increases regularly in extinction from 340 mp to shorter wavelengths. Below 240 mp the photographic plate density is small and the optical density calibration is rather inaccurate so that the maximum shown at 240 mp cannot be regarded as established. No absorption could be detected when a solution of pure solvent (i.e., containing no benzene) was flashed. KINETICS OF ABSORPTION AND PHOSPHORESCENCE DECAY The kinetics of absorption decay were measured near 310 mp. Although the absorption here is not very intense, this wavelength was chosen to avoid the region of benzene phosphorescence (above 340 mp 11) and that of the hexatriene derivativeT .S . GODFREY A N D G . PORTER 9 absorption (below 290 mp 12). For this reason relatively high benzene concentra- tions (10-2 M) were used. Small changes in the wavelength around 310 mp had no effect on the decay rates, nor did changes in flash energy (1 10,225,900 or 1000 J). Good first-order plots were obtained from the decay curves and the half-lives, z, are given below. Phosphorescence decay rates were measured at 375 mp in the flash apparatus and also on a spectrophosphoriineter. The values obtained were identical within experimental error. Solution A : 8 x 10-3 M benzene in 58 % isopentane+42 % 3-methylpentane absorption z = 1.7+0.2 sec (average of 7 values), phosphorescence z = 2.0 0.2 sec (average of 12 values). Solution B : 10-2 M benzene in 100 % 3-methylpentane absorption z = 3.5 0-4 sec (average of 9 values), phosphorescence z = 3.9 & 0.3 sec (average of 9 values).The errors quoted are mean deviations from the mean. The reported lifetime of benzene phosphorescence in EPA glass is 7 sec 11 and we obtained the same value in this medium at 77°K. The more rapid decay in hydrocarbon glasses is attributed to their lower viscosity, the viscosity of solution A being 2.5 x 108 poise and that of solution €3 about 1012 poise.13 The reasons for the lifetime dependence on viscosity have been discussed 5 and will not be considered further since the sole purpose of our lifetime measurements was to establish the identity of the absorbing species with the phosphorescence emitter. ROOM-TEMPERATURE FLASH SPECTROSCOPY A solution containing lO-3M benzene in 100 % isopentane was flashed with 1000 J at room temperature and the spectrum recorded 12 psec and 55 psec after flashing.No transient absorption was observed below 290 mp, the region of triplet- triplet absorption, but there was a transient absorption at longer wavelengths which had a maximum at 324mp. This may have been a radical formed in some secondary reaction. The results were complicated by the presence of considerable fluorescence which could not, at the short delay times used, be eliminated by the use of a shutter. ESTIMATION OF EXTINCTION COEFFICIENT It is possible to estimate at least a lower limit for the extinction coefficient of the triplet absorption at various wavelengths and the absorption spectrum of the 1-7 x 10-4 M solution was used for this purpose. The optical density at any wave- length was obtained directly from the spectrum and the concentration was deduced from the singlet depletion.In benzene the three sharp singlet absorption bands (249, 255 and 261 mp) appear in the middle of the triplet absorption but a rough idea of the amount of depletion was obtained in the following way. At 249 mp, the peak of a singlet band, the optical density of the solution was 0-40. From the smoothed triplet spectrum, the optical density of the solution if it had contained the original concentration of benzene would have been 0.45 at this wavelength. There was therefore an apparent depletion corresponding to an optical density change of 0-05.However, as optical densities could only be measured to k0.02 and we were using the difference between two values, we can only conclude that the change in optical density was &0.05. The extinction coefficient of the 249 mp singlet band at 77°K was estimated to be 162 and allowing for a 20 % decrease in volume when the solution was cooled, the concentration change corresponding10 ABSORPTION SPECTRUM OF TRIPLET BENZENE to an optical density change of 0.05 is 1-3 x 10-5 M (i.e., 6+ 7:). At 240 mp the optical density of the triplet absorption was 0.5, the length of the reaction cell was 23.5 cm and therefore the extinction coefficient of the triplet is >1600. It is also interesting to calculate the triplet extinction coefficient at 2537& the wavelength of the mercury radiation frequently used for photolysis.A value of ~ > 1 3 5 0 was obtained whereas the singlet extinction coefficient at 77°K at this wavelength is only 53. DISCUSSION Our observation of triplet benzene absorption with a high extinction coefficient in the region of 2537A removes what has been a serious objection to the triplet absorption biphotonic mechanism. It is clear that in studies of benzene with mercury resonance radiation at low temperatures, when quite high stationary triplet con- centrations are possible, absorption by triplet state benzene will generally be a sig- nificant factor since the triplet extinction coefficient exceeds that of the singlet at this wavelength by a factor >20. As the triplet absorption band is such a wide one, and we may not yet have found its maximum, our limiting values for the extinction coefficient show that the transition we are observing is probably an allowed one.This agrees with Pariser’s theoretical prediction 4 that there will be a triplet-triplet absorption around 260 mp with an oscillator strength of 0.8. The failure of previous attempts to detect this spectrum is attributable to the experimental difficulties of flash photolysis in rigid glasses, the overlap of singlet and triplet absorptions and the uncharacteristic diffuse nature of the triplet spectrum. This diffuseness of the spectrum is not, however, unexpected since the triplet absorp- tion spectra even of naphthalene and anthracene are predissociated 14 and the total energy of the excited benzene triplet is greater still, being 8.5 eV for 2537 A absorp- tion, and 8-8 eV for 2400 A absorption, compared with a vapour phase ionization potential for benzene of 9.25 eV. We thank Dr. M. I. Savadatti for valuable discussions and help with experi- One of us (T. S. G.) also thanks I.C.I. for the award of a research mental technique. fellowship. 1 Porter and Windsor, Proc. Roy. Soc. A , 1958, 245,238. 2 Voevodsky, Vinogradova, Shelimov and Fok, Doklady Akad. Nauk., S.S.S.R., 1964,154, 188. 3 Gibbons, Porter and Savadatti, Nature, 1965, 206, 1355. 4 Pariser, J. Chem. Physics, 1956, 24, 250. 5 Hilpern, Porter and Stief, Proc. Roy. SOC. A , 1964, 277,437. 6 Potts, Jr., J. Chem. Physics, 1952, 20, 809. 7 Vinogradov, Can. J. Chem., 1962,40,2170. 8 Jackson and Porter, Proc. Roy. SOC. A , 1961, 260, 13. 9 Bridge and Porter, Proc. Roy. SOC. A , 1958, 244,276. 10 Anderton, Chilton and Porter, Proc. Chem. SOC., 1960, 352. 11 McClure, J. Chem. Physics, 1949, 17, 905. 12 Migirdicyan and Leach, J. Chim. Physique, 1961, 58, 409. 13 Lombardi, Raymonda and Albrecht, J. Chem. Physics, 1964, 40, 1153, 14 Porter and Wright, Trans. Faruduy SOC., 1955, 51, 1205.