T H E N E A R U L T R A .V I O L E T S P E C T R A O F T O L U E N E P A R T II. T H E E M ISSIO N B A N D S
By R. K . ASUNDI and M. R. P A D H Y B IReceiVed for l>ubHeaUon, Nov. 8, ig48)
Plate X V
ABSTRACT An emission spectnini of toluene in the near ultraviolet is excited bv a transformer di.scharge through fl'wing vapour. Tin’s spectrum, which was previously observed by Stewart et al in Tesla di.scharge and mentioned by Kistiakow'ky and co-workers in re.sonance fluorescence, has been considerably extended and studied in greater detail.
Most of the bands are those observed already in absorption but there are a few which have not been recorded in the absorption spectrum. A complete analysi.s of the emission bands 'vhich is in agreement with the one for absorption band.s is proposed. The general appearance of bands and their intensities are compared \ ith those of the corre.sponding emi.ssion bunds of benzene. A fairly strong continuous emission band with a sharp short wavelength limit coinciding with the origin of the emission band system has also been recorded.
I N T R O D U C T I O N
The spectra of aromatic molecules have attained significance in recent years ovt'ing to the theoretical advances that have been successfully made in the interpretation of the spectra of benzene and some of its mono- derivatives. Recently we reported on the emission spectrum of benzene lAsundi and Padhye, 1945 and 1949). In this paper we communicate results on the emission .spectrum of one of its mono-derivatives—-toluene whose absorption .spectrum has been recently analysed (Ginsberg, Matsen and Robertson, 19+6) and also extended (Padhye, 1949) and whose Raman and mfra-red spectra (Pitzer and Scott, 1943^ are fairly web understood.
Stewart and collaborators (1930) were the first to excite the emission spectrum of toluene by the Tesla discharge through flowing vapour. They have observed about twenty-three bands, whose wavenumbers are recorded correct to four places only. It appears that they have observed also a fairly strong continuous background, Cuthbertson and Kistiakowsky (1936J have attempted to excite resonance fluorescence of toluene which even at o.oi mm.
pressure emit spectra which, by using a spectrograph with a resolving power of better than 25000, appear nearly continuous. At ordinary pressures and with illumination by the 2536A line, toluene emits a strong fluorescence consisting of mainly a continuum with a few faint and indistinct bands which become even fainter upon lowering the pressure. They do not record any numerical data.
• The emissiou spectrum of toluene, which has been excited in the present investigation by high frequency as well as ordinary uncondensed transformer
2— J7i*P— 8
37
340 R. K . Asundi and M. R . Padhye
discharge iii flowing vapour, shows that the bands of toluene are indeed super
posed by a continuous spectrum but the intensities of the two spectra being nearly of the same order with that of the continuum being uniformly slightly less than that of the bands, it Iras been possible to measure the heads of the emission bands. They are, however, not sharp like the benzene band heads but look rather diffused, probably on account of the continuum in which they are situated.
E X r K R I M K N T M / P R C’) C E D U R E A N D R E S U L T S
The experimental methods both for h. f. discharge and trrnsformer discharge arc exactly similar to those employed in the case of benzene (Asundi and Padhye, 1945 and 1949). Since the spectrum was developed with grpater intensity in transformer discharge like in benzene, it was photographed iJiider this discharge only using both the medium Hilger and K i Hilger quartz Littrow spe trographs described elsewhere. Iron and copper arc lines were used for standard wavelengths. Three plates, Kodak Special Rapid, ta^en on the E l Littrow instrument were selected for measurement with a Hilger Comparator reading upto o.oooi mm. The bands were measured by focusing the cross wires on the visually estimated maximum intensity part near the heads. The entire region, in which the bunds lie, is occupied also by a continuum which is mentioned above. This circumstance renders the weaker bands liable to be less accurately measured than the stronger ones. The three values meaiiired did not differ by more than 3 cm "' and »7 cm "' for the stronger and weaker ones respectively. The intensities represent visual estimates taking the most inten.se band at 37474* to be 5. The toluene employed was the one supplied by the Fine Chemicals Section of this University (b-p.iio.8'’C).
The bauds extend from 2667^ to 2900 X and are in the body of a con
tinuum which also starts practically with its maximum intensity from 2667A and extends towards the longer waves. It appears to have a sharp short wavelength limit at 2667-^. Also in a few spectrograms there could be seen a slight trace of its extention with weak intensity falling towards the shorter waves upto about 2603S.. No .sharp limit has been observed on the long wave side, although the intensity shows a drop at about 3S00A. The data on the baud heads reduced to wavenumbers in vacuo by the usc'bf K ayser’s Tabellen are given in Table T, along with the possible corresponding bands recorded by vStewart (1930). It also contains the visually estimated intensity values and the a.ssignments in terms of the analysis proposed. Corresponding values, wherever observed in the absorption spectrum, are also included.
These prove that the emission bands are the counterparts of the known absorption bands. T he bands are reproduced in Plate XV.
Unless otherwise stoted, the pn.sitinn of the bands is always given in this paper in wavenumbers per cm.
Near Ultra-violei Spectra oj Toluene 341
A N A Y S I S
It is Seen that the electronic transition resijousible for the ultraviolet bands of benzene is a forbidden one. Tlie symmetry character of this transition, is calculated on the assumption of a sijr-fold Dqu symmetry for benzene molecule. Mono substitution by any atom or radical destroys this six-fold symmetry and reduces it to one of Cg, type to which therefore the toluene molecule belongs. This symmetry has one two-fold axis in the molecular plane and a plane of reflection perpendicular to it.
Considerations of the energy states of the molecule closely follow that for benzene. A s a mailer of fact, in calculating the benzene levels a mean value of resonance energy given by thermo-chemical data on benzene, toluene, ethyl benzene and propylbenzene have been utilised (Pauling and Sherman, 193 3^ to evaluate the energy levels. T Ij u s not only are the characters of lh|e electronic states of toluene similar to those of benzene but to a first approximation even the energy values are the same. This is corro
borated by the fact that the bands, those in the near ultraviolet, lie in nearly the same region of the spectrum. On these considerations, the lowest state of toluene, as in benzene, is one of total symmetry viz.^ Ai^ corres
ponding to in benzene The excited stale of benzene is anti-syinmetrical to thew wo-fold axis and to the plane of reflection wdiich in the case of C'g,, is the B i, Hence the transition now is t h e w h i c h is an allowed one with the transition moment lying in the ]»lane of tlie molecule in the :v direction, i-e., the direction perpendicular to the C -C H g bond.
The absorption s])eclruni of toluene ivS thus difiereiit from that of benzene in so far as the latter is due to a forbidden transition wdiile the former to an allowed one. This circumstance renders particularly the vibration structures in two cases to be very different- Thus the o “ o band which is alxsent in the case of benzene at ordinary temperatures should he present in the case of toluene with considerable intensity. The totally symmetrical ai vibrations should be directly superposed over the o - o transition and should be very intense*
The components of the electric moment in two other directions (other than those due t o 1 ~ C l ) do not occur in this transition. But the product table shows that 6, and a. vibrations, where superposed, should give rise to weak bands with transition moment in these tw^o directions and that tlie vibration should not occur in the spectrum. An important class of transi
tions amongst this type found in all mono-substitiited benzenes is the one due to the lion-tolally symmetrical b, part of the split uj) 606 r j frequency which makes the benzene transition a l l ow e d . This non-lotally symmetrical vibration is found to be very prominent. Many of the previous investigators on the absorption spectrum of substituted benzenes have, after Sponer (1942), called the first part due to Ui vibration the allowed part, and the latter coming in due to superposition of the non-totally symmetrical vibrations the forbidden
part. Thus the spectrum due to a molecule with C ,„ symmetry should consist of these two parts. The third possible transition that can be allowed is the I - X transition of any vibration even including f>2. The bands due to the I - 1 transition of non-totally symmetrical vibrations will be very weak and scarcely observable.
As the symmetry table shows there can be only four types of vibrations possible in the ta t symmetry, via., totally symmetrical and bi and ba non*totalIy symraefrical, all non-degenerate. Hence the degenerate vibrations of benzene will split into two, most of them giving one totally symmetric
T a b i,e I
Emission Bands of Toluene
Asandi and M. R . Padkye
Pem'^ (Auiliors) h i t . A ssignm ent
Abvsor^tion banids
\ vStewart) Crinsbnr'g ei al
______ ____
37475 5 0,0 37477
.37452 3 u4-ij89 —I2I.3
37416 1.5 0 -5 9 3741S
37379 1 0 + 528-*6^,0 x3738i
.37354 0 -2 X 5 9 ^ 37356
3732W 1 0 + 456—620 37319
3728 37299 2-5 0 -17 8 37299
372J 37244 2 0 - 5 9 - 1 7 8 3724.3
37199 1*5 0 — 2X620+964, 0+932—1212 37191.
37206
3713 37130 0 — 340 ib\) 0 -2 X 178 ( ? ) 37121
3697 36960 - 0 -514 36963
3686 36888 '5
36851 2.5 0 —620 36857
3679 36785 2, 0 -5 1 4 -1 7 8 36784
3668 36688 4 0-785 36690
36657 0 0 — 2 X 1003 + J189
3662 3662s -5 0 -78 5 -5 9 1 0-842 (a2) 36631
36529 0
.36514 () 0 -7 8 5 -17 8 36517
1 36494 0 0-985 (a2) 36493
* T h e sym m e tric.^ o f v ib r a t io n s o t h e r t h a n 0i a r e i n d ic a t e d in b r a c k e t s .
5 UJ ,)I&PADHYE
plate XV
-2618-39A
(Cu)
-37475
(0,0)
2C67M)\
2 8 2 4 ’4 A (Cu)
Neat' Vltra-violet Spectra of Toluene Tabi,E I {contd.)
( S te w a r t ) (A u th o rs) Tut A s s ig m u e iit
A b s o r p t io n b a n d s G i n s b a r g ct al
3<'47 3
<^4'57
2 0 - TOO3 364653641 3843"
0
0 - 2 X 1003 -f 964 - 985 - 59364 J
5
() n —1003 — 59,0—T060 (bj) 36115 36403 (* 0-2 X 3003+932,0—1070 (b()36307 I 0—1176 36301
3<>98
36393 0 0-1003 — 17836281 u
3633 36263 1 0 -1212 36268
36205 0 0—1212
-59
359
" 35843 .3 u —1UJ2 —620,0—1630 (bj)3583
.35733
03567 35681 1 0 - 620 - 1176,0 — 1003 — 7H5 35687
3548 35469 f1 0 — 2X1003
I !
35421
n 1 11
!3538 35396 ; 0 , 0 — 1003 — 1176
3
S2
«f<3
S"7
35068 n i 0 —2X620—11763486 34842 1 n 1
1 0 —2X1003—620
3465 34680
i
’ 0 — 2 X 1003-78534*5
34265 !134202 1
11
and one non-totally symmetric vibrations. Further calculations show that toluene should have 14 of ai ty p e ; 4 of at, 13 of bi, and 8 of f»a type vibrations.
In the table given by Fitter and Scott (1943). one at vibration has been listed in bj making the former 3 and latter 13. Table II gives a list of the totally symmetric vibrations as given by these authors. These are all Raman lines active and polarised.
The above mentioned theoretical considerations are well borne Out in the analysis of the absorption spectrum of toluene- Some strong absorption bands were analysed by Sponer (1942) identifying some totally symmetric vibration frequencies in the upper state- The symmetry proposed was C» . But later analysis of the absorption spectrum and theoretical considerations
" prdVa that ttife syriutietry should be the same as other rtlono-Suhstitvited
344
Ta b l k H .
Totally symmetrical a, vibratioas of Toluene
H. K . Asundi and M, R . R ajkye
Raman speetrum
100?
521 I :nr»
1603 ] 175
7S5 JD30
13H0
2050
Alnsorptteti spectrum
Normal state
T(^()3
l'U2
K x c ile d state
1176
7 « 5
T012
933 456
I iSg
756 06^
N . H. V ib ratio n s in th is T ab le a s hi the rest o f the p ap e r are given in h av en n m b ers
per cm . j
benzenes, namely, (-^2^-- 'I'he absorption spectniiii of toluene has'been inves
tigated by Ginsberg, Matsen and Robeitson (1946). They have analysed the bands as due to superposition of a, vibrations on the 0 “ 0 transition and also the X“ X transition of these vibrations. Some Ilands are also analysed as due to superposition of tlie totally symmetrical vibrations on the non^
totally symmetrical part corresponding to the e't, frequency of benzene. A more detailed analysis shows that other cliaracteristics of the transition in question are also satisfied- An extension of the absorption bands towards longer waves, described elsew here, gives further evidence in support of these considerations.
In the case of emission, the most characteristic features have already been pul down. The continuum spreads over the whole of the region Starting with 0 - 0 band. The first and the strongest band at 37475 is the 0 - 0 band. A number of totally symmetric a, vibrations are superposed over this transition, the strongest being the 785 at 36688. The other vibrations observed are 5x4, 1x76,^x003 and 1212. The corresponding bands are quite intense. The fiequency of the ai vibration, classified by Pitzer and Scott as X030, has been found to be 1012 in the absorption by Ginsberg, Matsen, , and Robertson (1946). In the present experiments also a band corresponding, to 1030, could not be found and it is likely that , the band at 36467 whose distance from the 0 - 0 band is 1008 is due to the blending of two bands due
Near Vltra^oiolet Spectra of Toluene 345
to 1003 and 1013. The frequency 1003 is also excited by a quanta and band at 3546$ is classified as such. There are. also some differences in other frequencies considered e.g., the frequency tabulated by Pitzer and Scott as S2I is found to be 514 both in absorption and emission. These frequencies are tabulated in Table II together with the excited state frequencies found in absorption spectrum.
The other part which is called the forbidden part of the spectrum is also developed in emission- The band at 36851 is analysed as 0-620 where 620 i.s the non-totally symmetrical b, i^art of the e i fretpiency 606. The intensity of this band is rather large for the non-totally symmetrical vibration. It appears thei efore to indicate the importance of the mechanism which makes the spectrum of benzene allowed, and hence the weakness of the substituent in destroying the six-fold symmetry of benzene. There are some totally symmetrical vibrations superposed over this transition. There are also bands which can be classified as possibly due to the excitation of h, vibrations or at vibrations but there are in most cases other alternatives also possible for them and it is not possible to make a conclusive choice.
The emission spectrum differs rather widely, as it- should, from the absorp
tion spectrum even in appearance. The absorption spectrum has a number of bands lying to the short wavelength side of the 0 - o band, whereas in emission there are none But on the longer wavelength side of the 0 -0 band both in emission and absorption there are about 18 bands iti common which are similarly classified. These are all due to excitation of the totally symmetric ground state frequencies and those due to the further superposition of difference frequencies of 59 and 178 on them. The anals^is of the absorption spectrum also gives 59 and 178 as difference frequencies involved in the spectrum. Ihe same frequencies are also found in emission associated with the transitions involving a, vibrations. These low frequencies must be due to i - r transi- tion of a frequency which is quite low as the intensity of the band suggests.
In benzene such a difference frequency of 160, which was due to the ground state et 404 frequency, dropping to 240 in the upper state. The el frequency is split into 405 (7 and 467 ha in the present case. The assignment to either is not possible since is not excited in the ground state or in the excited state in either spectrum and the other single quantum excitation is not allowed.
It will be interesting to compare the spectrum with the emission spectrum of benzene. The transition involved in the case of benzene is a forbidden one. The spectrum of benzene even then shows a large number of bands which are quite intense and quite sharp involving a number of vibrations in keeping with the selection rules for the forbidden transition. But the transition in toluene is an allowed one. One would expect a spectrum consisting of a large number of bands involving the totally symmetrical frequencies which in number are many more than in ^nzene. It would have also been expected that the spectrum should be more irrtense than the benzene spectrum. These conditions are partly fulfilled in the ca.se of abs<tfpition
^46 R , K . Asundi and M, R , Padhye
spectrum but iu emission the expert nieotal fa d s are quite the opposite, r In all there are very few bands and these have no sharp heads- The total intensity of the spectrum as a whole including: the continuum is also less than in benzene as far as could be judged. This can be understood on the
following considerations. '
A comparative study of the total intensity of absorption due to different substituents has been done by Sklar (1939) and Mulliken (1939). Sktar has treated toluene as one substitution. Theoretically it is found that large intensification is closely associated with large directing power which in its turn is associated with the induction effect or migration effect between the ring and the substituent. The induction effect has very small influence and much less when the substituents arc not of a polar character. The migration effect is attributed to the non-bonding type of the />, electrons W the substituent. Thus large intensification ought to be expected in those cases where the substituent has low ionisation potential, a pair of unbound pe electrons and small ting-radical distance. Since these are not favourable in the case of the CH3 radicle, the intensification expected was small and actually the absorption spectrum of toluene is found to be only twice as intense as benzene. If the same considerations hold in the case of emission and if these are the only corresponding emission spectra, then the intensity of emission, which ought to have been larger in toluene than in benzene, is actually smaller. Probably an ex]jlanation of this anoinally is to be sought in the Oatuie of the excited states in the two molecules. *
A
C
KN
OW
LE
D OM
KN
T SWe desire to record our thanks to Prof. Sir K . S. Krishiian and Prof, Ur- N. R. Dhar of Allahabad University for permission to one of ns (M- R. P.) to work in Iheir laboratories and to use the E i quartz Littrow spectrograph -
He n a f k s Hi n d u ITn t v b r s iIy
R E F E R E N C E S Asundi and Padhve, 1945, Nature,185, 368 _
„ „ „ , xg^g.Ind. Jour. P h y s.,n , ig<) Cuthbertson and Kistiakowsky, 1936, /. Chem, Phy.,4, 9 Ginsberg, Matseii and Robertson, 1946, J. Chem. Phy., 1 4, 507 Mulliken, i9 3 9ifJ- Chem. Phy,,7, 353
Padhye, 1949, Jnd. Jour, Phys.,23, 331
Pauling and Shernian, 1933, J. Chem. Phy., 1, 679.
Pitzer and Scott, 1943,/./Im. Chew. Soc., 66, 803.
Sklar, 1939, J. Chem. Phys., 7,984.
Sponer, 194a, Rev. Mod. Phy., 1 4, 224.
, 1942, /. Chem. Phy., 10, 672.
■ Stewart, 1930, Recent Advances in Physical and Inargatilic Chemistry, p. 350. 'jC
