2560
GUSNARWETTERMARK
ception similar to that of model B. Note that the measurements of relative fluorescence efficiency had been carried out under non-photostationary conditions. The conclusions given in our paper, however, concern the photoequilibrium of isomers where d[St*]/'dt = 0. Further, there might occur a thermal quenching St* + St by solvent molecules competing with fluorescence emission which had not been taken into account by Dyck and McClure. The evidence for this process may be seen from the dependence of fluorescence efficiency on solvents.' We feel that only measurements of the absolute quantum yield of fluorescence as a function of temperature performed at photoequilibrium would help to decide if any thermally activated S + T transition takes place in photoisomerization. ,is to model B, quantum yields should increase a t lontemperatures with decreasing wave length of irradiating light but this effect never has been found.3 Conclusions The experimental results on photochemical c i s F? trans isomerization of stilbene available hitherto seem to be satisfied better by model A than by model B. Even the results of Malkin and Fischer3 on azo compounds are in agreement with the mechanism of model A, as may be seen, for example, in 1-phenyla~onaphthalene.~In azo compounds, the activation energy for Tc+ Tt conversion seems to be very small. No special conception is necessary in model A
1-01. G(i
about the type of electronic excited intermediate state. Xote that Zimmerman a,nd co-workersy suggesting a thermal isomerization in an excited singlet state already had pointed out qualitatively the main features involved in model A. However, in order to decide exactly which type of mechanism is involved in photoisomerization, a conclusive proof might be obtained only from more ext'ensive information about the triplet state. Possibly, measurements of the phosphorescent lifetime of cis-stmilbenelymay indicate the order of magnitude of lil. Acknowledgments.-The author gratefully acknowledges the support of the U. S. Army Research Office (Durham) for his participation in the International Symposium on '' Reversible Photochemical Processes" held a t Duke University, Durham, N.C., April 16-18, 1962. The helpful criticism of Professor Dr. E. Kuss also is greatly appreciated, as well as the interesting discussions of Dr. E. Fischer and Professor Dr. G. Zimmerman. (19) T h e phosphorescence of cis-stilbene reported by Lewis a n d Kasha20 and, in more detail, by Nauman21 really is due t o phenanthrene.?% Recently, a e observed a strong blue luminescence of cisstilbene in rigid solvents a t 194' mainly in t h e region of about 4700 A. This effect is not exactly due t o phenanthrene or other impurities a s shown by special measurements. As t h e emission vanishes in fluid solvents even a t -78' a phosphorescence emission is supposed. (20) G. N. Lewis and ill. Kasha, J . Am. Chem. Sac., 66, 2100 (1944). (21) R. T'. Nauman, Dissertation, University of California, 1947. (22) The author wishes t o t h a n k Dr. Zander, Castrop-Rauxel, for referring him t o t h e paper of Nauman.21
-
LIGHT-INDUCED ISOMERIZATION OF o-NITROTOLVEXE I N WATER SOLUTION BY GUNNAR WETTERMARK' Pioneering Research Division, Quartermaster Research and Engineering Center, Natick, Mass. Receiaed M a y 26$ 1062
If o-nitrotoluene in water is exposed to ultraviolet light, a short-lived species is formed which has a strong absorption band in the blue and near ultraviolet regions of the spectrum. Two different absorption spectra of the species are given, interpreted as the absorption spectrum of the acid form and the anionic form of the species, respectively. The decay follows that of a first-order reaction with a rate constant of 0.76 X 102 set.? for the acid and 1.0 sec.-l for the anion.
Introduction
exposed to light and revert to the original colorless state in the dark. The solid has been extensively nitrotoluene in solution on exposure to light under- investigated over the last two decades4-' but it has goes a transformation to a short-lived colored statee2 been found only recently by employing low temThe decay of this colored state was found t'o follon- peratures that the color change can be brought the kinetics of a first-order reaction. In ethanol at about in solution.* Also by employing low temroom temperature (24 l o )the half-life time was peratures it has been found that other compounds closely related structurally, namely, 4-(2',4'-dievaluated to be 5 msec. Photochromism has been detected in compounds nitrobenzyl) -pyridine and 2- (2'-nitro-4'-cyanowhich are derivatives of o-nitrotoluene. Tschit(4) (a) K. Schofield, J . Chem. Sac., 2408 (1949); (b) A. J. N u n n schibabin and co-workers3 discovered that crystals ibid., 583 (1952). of 2-(2',4'-dinitrobenzy1)-pyridine turn blue when and(5)K .WSchofield, . C. Clark a n d G. F. Lothian, Trans. Faraday Soc., 64, 1790
It recently was found in this Laboratory that
o-
*
(1) National Academy of Sciences-National Research Council Visiting Scientists Research Associate a n d Guest of t h e Massachusetts Institute of Technology associated with Prof. L. J. Heidt of t h e Department of Chemistry. (2) G. Wettermark, Nature, 194, 677 (1962). (3) A. E. Tschitschibabin, R. M. Kuindshi, and S. \V. Renewolonnknja, Rer., 68, I580 (1925).
(1958). (6) H. S. Gutowsky and R. L. Rutledge, J . Cliem. Phys., 29, 1183 (1958). (7) B. A f . Kuindshi. L. A. Igonin, Z. P. Gribox.it, and A. N. Shabadash, Optics and Spectroscoptj, 12, No. 2 (1962). (8) R. Hardwick, H. S. >Insher, and E'. Passailaigue. Trans. F'aradau ,Tor., 56, 44 (1960).
beneyl)-pyridinel" are phototropic in solut'ion as well. Preliminary experiment'sin this Laboratory using flash photolysis have indicated that a variety of compounds, all being derivatives of o-nitrotoluene, are photochromic.' A closer study of the color change of o-nitrot'oluene therefore has been undertaken and is report,ed in this article.
-2
- 1.5
Experimental Flash Apparatus.-A description of the flash apparatus urill be given shortly by L. Lindqvist, Institute of Physical Chemistry, Cppsala University. Only some data pertinent to the actual experiments are given here. Briefly an electrical energy of 1800 joules (9 pf., 20 ltv.) was discharged through four straight quart,z lamps symmetrically arranged around a cylindrical reaction vessel giving a flash duration of 5 Msec. ( l / e time). The reaction vessel had an optical path length of 20 cm. and was provided with an outer filter jacket which contained a 0.5 cm. layer of 507; aqueous acetic acid. This filter absorbed all light of wave lengths shorter than approximat,ely 250 mp. The light absorption changes in the system were recorded by means of a single beam absorption spectrophotometer. The monitoring lamp was a zirconium lamp (Sylvania K 100 Q) powered by a large bank of storage batteries. Other essential parts of the spectrophotometer consisted of a Bausch & Lomb small grating monochromator, an E.M.I. 9552B photomultiplier tube, and a Tektronix 535A oscilloscope. The slit width of the monochromator was kept constant (0.5 mm.), corresponding to a spectral band width of 3 mp. Various Corning color filters were used to cut off the secondary pass band from the grating of the monochromator. The anode current from the photomultiplier was amplified in a cathode follower mounted directly on the photomultiplier tube. The over-all circuit had a time resolution of about 5 psec. in the present experiment,s. A11 experiments were carried out a t room temperature (24 z.t 10). Solutions.-o-Nitrotoluene ( Matheson Coleman 6: Bell) was recryst,allized from absolute ethanol seven times. simple apparatusll for low temperature recrystallization was used with Dry Ice and alcohol in the cooling bath. Further purification was obtained by passing an ethanol solution of the sample through a 60 em. long column filled with aluminum oxide and collecting the middle fraction. The concentration of o-nitrotoluene in the test solutions J f . o-Nitrotoluene was mixed with absolute was 2 X ethanol and diluted with aqueous 0.1 S NaOH and 1 -3HCl, respectively. The final alcohol concentration amounted t o 0.57; by volume. Deaerating of Solutions.-Before each experiment the solution was carefully deaerated. The technique employed was to freeze, evacuate, and melt three times and to shake the solution in the presence of argon (hirco for aluminum welding), then to freeze, evacuate, and melt three more times. The flask used for the deaeration was connected to t,he reaction vessel through a ground standard taper joint in such a w:tp that the solution could be transferred without coming in cimtact with the stopcock grease.
- I
- 0.5
0
0
I
2
3
sec
IO
20
30
msec.
- 1.5
- I
-0.5
"
1 I
0
Fig. 1.-Plots of loglog (Im/l)as afunction of time for various wave lengths. ( I is the intensity of the light transmitted through the solution during the fading of the short-lived species and I , the intensity after completion of the reaction.) ( a ) o-Xitrotoluene in 0.1 N NaOH; ( b ) o-nitrotoluene in 1 HC1.
to a flash was less than 4% of the reversible change. When the same solution was flashed several times the response gradually weakened due to the acResults and Discussion cumulation of photo product. No effect on the Flashing a solution of o-nitrotoluene revealed rate of decay of A* was observed even after ext'hat short-lived species (A*) are formed which tensive flashing (20-30 flashes). ,4fter about 20 absorb strongly in t'he blue and near ultraviolet flashes the degree of conversion of o-nitrotoluene regions of the spectrum. A st'able photo product, into A* was about half of that which was observed (B) also is formed, a$ could be seen by a permanent a t the first flash. The portion of o-nitrotoluene conchange in the optical transmission of the sohition. verted to A* in a flash could not be determined Scanning the spectrum over the region 300 t'o 600 from these experiments. The question arises whether the decay reaction mp showed that' the photo product (B) had an absorption band in t'he same region as A*. Over the of A* constitutes the transformation of A* into the wave length region investigated the permanent new species B or the decay back to the original o(irreversible) inrrease in optical density on exposure nitrotoluene (A) (or both). As the photo product (B) has an absorption band (9) H. S. Xlosher, C. Souein, a n d R. Ilardn-ick, J . Chem. Phus., 32, in the same region as A* it was impossible to 1888 (1960). measure the rate of formation of B by isolating a (IO) .J. .L. Fousn tind .J. Weinstein, J . Org. Chem., 27, 3 1 5 5 (IW'2). 15 nvc length viherr only R nlmrhed the light. (11) A. L. I h h m , .7. C'hem. Educ., 36, 200 (1958).
GUNNARWETTERMARK
2562
Vol. 66
-CH2- group to the 2'-nitro group together with a change in the structure of the aromatic ring to a quinoid configuration. It seems reasonable, therefore, to propose that the formation of A* in o-nitroi
toliiene is described by the reaction colorless isomer
$==J
CI13-
colored isomer
--
" 04
N HO'
'0
A
'0
A"
The colored isomer is probably a relatively strong acid. It is, however, reasonable to believe that in 1 N HC1 mostly the acid is present. In 0.1 N NaOH the colored isomer should exist mostly as the anion. 1I
,
300
400
500 rnp.
Fig. 2.--hbsorption for o-nitrotoluene in 0.1 N NaOH (9) and 1 N HCI ( 0 ) after exposure to ultraviolet light. The ordinate is proportional to optical density with different proportionality constants for the two solutions.
CHzQ
-.
A*
CHzQ
.c---
N
-o/
'0 0.1 N NaOH
N
HO'
'0 I N HCl
The available information on the system might support the assumption that A* decays pre- One might thus conclude that the two spectra given dominantly to A, that is, the over-all process is es- in Fig. 2 are the spectrum of the anion (0)and the spectrum of the acid ( 0 ) . sentially reversible. Table I gives the rate constants for the fading of If A* decays back to A in a first-order reaction A* in 0.1 N NaOH and in 1 N HCl visualized as the a straight line should be obtained if log log ( I m / I )is plotted as a function of time. (I symbolizes the in- fading of the anion and the acid, respectively. tensity of the light transmitted through the solution The data represent mean values obtained from all during the decay of A* and I , the intensity after the plots of log log (I,/I) as a function of time for completion of the reaction.) I n fact when such the different wave lengths (compare Fig. 1). plots are made straight lines are obtained. All plots TABLE I were made from the first or the second flashing RATE CONSTANT FOR THE FIRST-ORDER FADING REACTION of a fresh solution. Figure 1 gives some of these O F A* AT R O O M TEMPERATURE (24 i 1") plots for different monitoring wave lengths. Rate constant From the plots of log log (Im/])vs. time at difSolution (sec.-l) ferent wave lengths the absorption spectra of A* o-Nitrotoluene in 0.1 N XaOH (anion) 1.0 in 0.1 N NaOH and in 1 N HC1 could be calculated o-Nitrotoluene in 1 N HCl (acid) 0.76 X 102 and are given in Fig. 2 . It has been visualized3-12 that the photochromism of 2-(2',4'-dinitrobenzy1)Acknowledgment.-The author wishes to express pyridine is due to a proton transfer from the his sincere thanks to Dr. S. D. Bailey, Chief of this Laboratory, who helped make this work possible. (12) G. Wettermark, J . Am. Chem. SOC.,84, 3658 (1962).
