On the Polarised Fluorescence of Organic Compounds

ao

“ O N T H E P O L A R I S E D F L U O R E S C E N C E O F O R G A N IC C O M P O U N D S ”

By

SACHINDRAMOHAN MITRA

^Received jor publication^ October 6, u;j9j

a b s t r a c t. The variation of the p(.)larisatioii of fluorcscenci: of the dyestuffs with the diange of (a) the viscosity of the solution, (b) the temperature of the solution, and (f) the con- oeiitratiou of the dvestuffs has been investigated expeiinientally. In all the cases it has been observed that the degree of polarisation tends to vanish at low viscosities or high temperatures or at high ronceiifratioiis of the dyestuffs while at very high viscosities or low tem])cratnre or at very low^ concentrations of the dyestuffs, the polarisation tends to reach asymptotically t\ certain limiting value which is dependent on (a) the nature of the dyestuffs, (b) the wave-length of the exciting radiation, and fc) the naluic of the solvent. These results have been discussed on tlie Ijrisis t)f Perrin’s theory. Tlic varialioli of the polarisation with the wave-lengths of the exciting radiations has also been investigated in detail. Tt was found tJiat tlie pfjlarisation first decreases with the increase of the w'ave-leiigtli of the exciting rarlialicm and increases again after reaching a minimum value which is negative and occurs at the exciting vvave-leiigth.s which are (Imracteristic for the molecules of the dyestuff.

Besides the investigations mentioned alxjve, the average life cd‘ the dyestuff molmdes in I lie excited states, the absorption sjjectra of a large number of dyestuffs in glycerine .solulitm in the ultra-violet region, tlie influence of one dyestuff molecule on the polarisaiion of fluorescence of another dyestuff molecule ami the ((uenching of the fluorescence nf the dyestuhs in solution by the addition of foreign substaiu'c have been investigated.

G K N K R A T I N T R 0 1) U C T 1 0 N

A large amount of experimental work lias been done, especially during recent years, on the fluorescence of dyestuffs in solution. ' Much of the work, however, has been more or less of a qualitative nature ; even where quantitative measurements have been attempted, the results obtained by the various experi­

menters differ so widely from one another that they can be considered only as giving the order of magnitude of the quantities measured. This is especially so in the case of measurements on the polarisation of fluoicscence. It is now definitely known from the work of Weigert, ^ Wawilow, ^ .Schmidt, * Perrin, “ Gaviola and Priugsheim and others that the degree of polarisation of the fluorescent radiations emitted by solutions of dycstulfs depends among others on the following factors the temperature and the viscosity of the solution, the concentration of the dyestuff and the wave-length of the exciting radiations.

For most of the measurements on polarisation, however, sufficient data regarding

3 5 0 S . M . M itra

these factors arc not available to enable us to know the exact conditions to which the measurements refer.

At the same time it is generally recognised that accurate measurements on the polarisation are likely to throw considerable light on the nature of fluorescence, which would be highly welcome in view of the unsatisfactory nature of the various theories that have been proposed from time to time—not one of them being capable of explaining satisfactorily even the essential facts of the phenomenon.

A few years ago it was, therefore, felt desirable to make systematic measurements on the polarisation of the fluorescence of some typical organic compounds for different wave-lengths of monochromatic excitation, at varioui^

temperatures, for different concentrations of the compound and also the influence of foreign substances. 'I'he present report gives the results of our further measurements with a crdticril discussion of the results obtained so far.

T H It o P T 1 C > L A R R A N G K M R N T

'I'he light from a quartz mercury lamp, automatic copper, cadmium and zinc arcs, rendered monochromatic by passage through a Hilger constant deviation quartz monochromator, was used, iu general, as the source of excitation. In ordci to eliminate the uncertain polarisation introduced by the crystal-qu^lz parts of the monochromator, the light issuing from the monochromator was allowed to pass through a polarising prism before incidence on the fluorescent solution. ^ Tlu' polarising prism was of the Gian type, and was transparent to the ultra-violet. The fluorescent solution was contained in a rectangular cell with windows of fused silica. By keeping the exit slit of the monochromator very wide, and shiftiiig the telescope lens of the instrument suitably towards the slit, it was possible to focus the radiations of any required wave-length at the centre of the cell containing the fluorescent solution. The light so focussed was naturally in the form of a vertical thin sheet, which, when viewed from above, looked, especially iu the neighbourhood of the focus, like a thin parallel pencil of light. This direction of observation was, therefore, very convenient for measurements on

“ transverse ” fluorescent radiations. -

The partial polarisation of the fluorescent light was measured in the usual manner by the Cornu method. Since the fluorescence was in the visible region for the dyestuffs studied here, the measurements could be made visually : where the intensity was very feeble,*and sometimes also in other cases for a corrobora­

tion of the visual measurements, the measurements were made by photographing the tracks by their fluorescent light. Since the technique of these measurements is well known, it is unnecessary to describe here the details. In some cases the measurements of the polarisation were made by a Savart plate in the usual way.

P ola rised F lu o re sce n ce o f O rg a n ic C o m p o u n d s 3 5 1

M E P E N D E N C E O F P O L A R T vS A 1' T O N 0N T IT E W A V E ’ L E N O T H O F T H E E X C T1_' 1 N G R A D I A T I O N vS

Before proceeding to the results obtained, it is necessary to explain the notation adopted by us. The incident light is linearly polarised, and the measurements refer to the fluorescent radiations along a direction normal to the plane containing the direction of vibration and the direction of propagation of

!be incident light. The fluorescent light is in general partially polarised, its

\'ibrations along the direction of propagation being usually less intense than the vibrations in the perpendicular direction. The polarisation is then taken as positive and is measured as usual by the ratio of the difference in the intensities n[ the two vibrations to the sum of the two intensities* the ratio being expressed a> a percentage. When the vibrations along the direction of propagation are more intense than those along the pen)cudicular direction, as happens in some (‘:Lses, the polarisation is taken, consistently witli the above notation, as

negative.

The results of our iiicasurements of the polarisation of fluoreseence excited by different wave-lengths after correcting for the ]solarised lluorescenee of glycerine, are given in the following tables and are gra[)hically shown by the accompanying curves. The tables and the figures conclusively show that the pularisatioii generally decreases to a minimum value as the wave-lengths of the cxeiling radiations decrease and then increases on the further decrease of the wave-lengths. This niininiiim value is negative and is reached when the exciting lip hi is near about A 3 13 1 A.

It may be mentioned here in passing that Jablouski ^ reported recently that in the cases of various colouring matters in glycerine solutions, the polarisation of the fluorescence progressively decreases with the decrease of the wave-length of the exciting radiation and under the ultra-violet excitation the polarisation practically vanishes. Later on Griseback ^ observed that the polarisation attains the maximum negative value at two wave-lengths* of tlic -xcitiug radiation in the case of eosiii instead of at one which \vc observe at about V 312 771/u. He also claims that the polarisation is zero when the exciting wave-length is at about A 365 mfi and 326 But our measurements of the jHilarisation for various exciting wave-lengths do not agree with those of the aforesaid workers.

^ Wc shall further return to this point at a later part of the paper.

352 S. M . Mitra

FlGtJKE

Wave-length

F

a

P o la r is e d F lu o r e s p m c e o f O rg a n ic C o m p o u n d s 3 5 3

Wave-leogth Figure

3

Tabee I

Percentage of Polarisation of Fluorescence in Glycerine.

Concentration of the conipound=ti x io “ ’ gm, c.c.

Temperature = 28°— 30®C

Wave­

length in mp

Fluor­

escein Succinyl-

fluorescein. Kosin. Succinyl

cosin. Magdala

red. Rhoda-

mine. ICryth- rosin.

M'S 43 40 4T 34 SO

S35 ^{— —} 44 42 40 33 —

SI 7 ₄₂ ^— 46 43 38 32 —

509 43 40 47 44 3<5 ₃₁ 49

480 ^— 48 ^— 34 27

466 ^{___ —} 48 — ^— 25 47

436 45 45 47 45 ^{26 20} 45

405 45 45 42 — 13

365 42 40 27 25 10 0 270

336 4 ⁰ -7 - 8 —6 — 12 — 0

313 — 10 - 8 - 9 - 9 ^{—6 — 10} '■ 9

298 - 1 4 " 9 -7 ^{—6 0 — 6} *“ 4

278 —5 6 0 7 II 8 14

265 6 20 9 — ^{— 20} 25

^54 29 ¹⁶ 34 25 27 38

»33 35 ^{38 3^ 29} 45

3 5 4 S . M . M ttra Tablb I {contd.)

Percentage of Polarisation of Fluorescence in Glycerine.

Concentration of the compound = 4 x io"*gni/c.c.

Temperature = 28® —30®C Wave-length

in mu flavine.^AcrO“ Aesculin. Naphthyl-

amine. salicylate.^Sodium

f?‘Amiun

benzoic acid.

436A05 365336 313298 278265 254233

3835 33o

“ 10-12 O

8 2815

34 37 3329 o

“ 9o 3136

22 1816 15O -8

10o

20

42 37 3530 o -7

195 33

14 16

7 ; 12 P O L A R I S E D P I / U O R E S C E N C E I N O T H E R S O L V E N T S

In order to find out whether this change of the percentage of polarisation, especially of the negative value, is due to the influence of the nature of the solvent, we undertook a detailed investigation on the effect of the exciting wave-lengths on the polarisation of the fluorescence of the compounds^ in various solvents. The solvents used were sugar solution, castor oil, glycerine-watei mixture, collodion-ether mixture. The results of our ineasui'einents are given below.

Tabue II

Percentage of Polarisatioji of Fluorescence in Castor oil.

Concentration of the compound = 4 x lo-^gni/c-c.

Temperature = 2 8 °“ 3o''C Wave-lengths

in fUfA. Fluorescein. Eosin. Asculin. Rhodaniiiie. Magdala red.

45 ■ _{40 43}

535 ^— 45 — — 4X

517 — ^— — — 33

509 — — ^— 31 3*

480 ^— — ^— 28 39

466 — 46 ^{— —}

436 45 • 47 30 2(5 _a;

, 45s 44 40 — 12 —

365 40 28 37 0 IS

336 — 10 -6 35 - 1 2 r-8

313 - 1 5 — 10 33 — 12 - 8

298 “ 7 _^ 0 — 6 - 6

378 0 a ? - 6 7 ^{. 0}

365 9 10 0 V JS

»34 31 20 33 12

333 ²⁶ 24 35 3b

P o la rised F lu o r e s c e n c e o f O rg a n ic C o m p o u n d s

355

Tablb I I I

Percentage of Polarisation of Fluorescence in Collodion-ether mixture.

Concentration of the compound = 4X io~*gm/c.c.

Temperature = 28“ —30®C

W a v e -l e n g t h s in

iHju.

Fluorescein, Kosin Asculin. Rhodamine. Magdala

red.

546 —

...1

42 40

535

—

— 41 —

517

—

5^^9

—

—

38

480 — — — 36

436 46 4a

28 31

44

40

— a6

365 36 26 32 ⁹ 18

326 0

— 10

313 — 10 28 — 12

398

—

-6

f —6

378

—

365

11

18

354 J5

20 20 27 10

333 32 23 ^{2CJ 34}

19

356 S. M. Mitra

Fig t o e 5

#

T he tables and the curves conclusively exhibit that though the value of P is different for different solvents, the general nature o f the graphs practically remains the same and the negative values persist in all the cases-

r O L A B I S E D F L U O R E S C E N C E I N T H E S O L I D S T A T E

W e have also measured the polarisation of fluorescence in the solid solution.

A measured quantity of gelatine was added with a small quantity o f water in a test-tube. T hey are then gently warmed till there was a very thick emulsion.

T o this was added a measured amount o f stock solution o f the fluorescent organic com pound. The whole was stirred vigorously so as~to make a uniform solution of the organic com pound. This emulsion of gelatine was poured on a flat glass plate. On being dried the film exhibited an intense flu oresc^ ce.

Measurements o f the ‘ polarisation w'ere made in the forward direction with the plane o f the film perpendicular to the path o f the exciting light, in order to avoid the influence of the polarisation due to the reflection at the surfaces, which would affect the measurements along other directions; complementary colour filters w'ere used to cut off this incident exciting light. T h e results are given in the Table IV .

Polarised Fluorescence of Organic Compounds

Table IV

Percentage of Polarisation of the Dyestuffs in solid Solution- Concentration = 4 X lo ’ ^gm/c.c.

Temperature = a8°C

3 5 7

Wave-length in

mjA. Fluorescein.

546 43^ 405 365 326 313 298 278 265 254

24 20 12 -4 -8 - 6 5 13 IQ

Fosin.

22 10 S

5

"2 ? -3 - 2 ?

15

'R h o d a n i i i i e R . M a g d a la R e d .

12

lu

8

o

“ 5 - 5

o 4 7 10

21 12 16 7

“ 3 -3

o 3 5 8

T N r Iv Tj K N C R O F T TT K V T vS C O vS I T Y OF' T H R S O L U T I O N O N T H B P O L A R I vS A T I O N

In a well-known paper Perrin has investigated the dependence of polarisation on the viscosity of the fluorescing solution, and has deduced on the basis of the theory of the Brownian rotation of molecules an expression for the above dependence. The essential principle of the calculation is as follow s:—

Assuming that the molecules are rigidly fixed in space, if an incident polarised light excites fluorescence in the mcdiuin, the polarisation of fluorescence will have a certain value jf>o which is characteristic of the substance. If on the other hand, as in the actual experiments, the molecules are rotating, the mean square of the angle of rotation per second can be calculated in terms of the viscosity of the solution and its temperature from Einstein's theory of the Brownian motion; hence the "e x p e c ta tio n ” of rotation for a time t can be calculated, where r is the mean duration of the molecules in the excited states. The smaller this angle of rotation, the more closely would the actually observed value of the polarisation approximate to the ideal value, viz., po- Thus for very large viscosity or very low temperatures p should reach asymptotically the limiting value Po, while at high temperatures or low viscosities the polarisation ought to reach again asymptotically zero value.

We win quote here only the final expression obtained t y Perrin,” viz.

358 S. M. Mitra

p — P o ~

I + ( l —

where p is the degree of polarisation observed under the actual condition of the experiment and po the value for the same molecules when they are not allowed to rotate from their initial orientations; p„ would evidently be limiting value oi p when either the viscosity is very large or the temperature very low. R is the gas constant for gram molecule and V the gram-molecular volume of the dyestuff, T is the absolute temperature and »? is the coefficient of viscosity, t gives the mean duration of the fluorescing molecules in the excited state (defined as usn^l by the relation I = ) under the actual conditions of the expciimeni-, the excited molecules are in an isolated state, i.e., free from collisional and othcii

influences of the neighbouring molecules, the duration will be that due to the.

radiational resistance alone, and can be readily calculated. For example, when the fluorescent band is in the green at about 5oooA°, calculation gives for r the value i i - i X lo” * secs, in vacuum, or in glycerine solution for which the refractive index is about 1*47 the value of t would be equal to 7'5 x lo"* sec.

T^able V

Polarisation of Fluorescence in Succinyl Fluorescence excited by A 4358 A Percentage by wt.

of glycerine. Viscosity in Poises.

Percentage of polarisation [P =p X 100]

Concentration

c = -oS.io'^gm/c.c. Concentration c —■ 8.io”S gm/c.c.

Concentration c==ri.io“6 gm/c.c.

99 57 50 47 45

927 i ‘8 46 42 37

86-3 74 40 35 29

8o-8 ■39 34 27 21

75*9 •21 28 21 12

yr6 •17 23 16 _ 7

67-8 •13 18 T2 5

64.9 ^{•10 15 9} ...

6i‘a .086 12 5 ...

58-3 •071 10 ... ...

55-8 *052 7 ... ...

S3‘4 ^*052 3 ... ...

*047 0 ...

P o la rised F lu o re sce n ce o f O rg a n ic C o m p o u n d s

559

But actually there may be other channels through which the energy of the fluorescent molecule can be dissipated—other than radiation—as for example collisions of the second kind, etc., all of which will effectively tend to diminish the mean duration in the excited state.

In order to be able to get some information regarding the duration of the excited state measurements were made on the polarisation of fluorescence of fluorescein solutions in glycerine-water mixtures whose relative proportions and hence the viscosity could be varied over a wide range. The concentration of the dyestuff per c-c. of the solution was kept the same in all cases. The results of the measurements are tabulated in the following tables.

The values for the viscosity given in column two of the table are calculated from the recent extensive measurements by Muller (International Critical tables) oil viscosity at different temperatures and various concentrations, by graphical

interpolation.

i _ c = 8 x i o '® gm./c.c.

2—c=4 •» ••

T^able VI

Polarisation of Fluorescence in Succinyl Fluorescein excited by \ 3650 A

3 6 0 S. M , M itra

Percentage by wt. of glycerine.

99927 86*38o-8

75’971*6 64.967-8

Percentage of Polarisation “ P Viscosity in Poises.

Concentration

4.io"6 gm./c.c. Concentration gm/c.c*

57 40 43

r8 33 38

74 20 _{30 ; 1}

‘39 10 _{13 \}

3 8 '

'13 0 5

• 10 0

The values of P for the two concentrations are plotted in the accompanying figures. It will be seen from the curves that at high dilutions, i.c., at low viscosi­

ties the values of P tend to zero and at high viscosities to the limiting value P^, as 'we should expect from Perrin's expression.

In order to have some idea about the mean life of the fluorescent molecules in the excited state, r was calculated from the relation —

^ _ ___ P o - P ____

/ , , R T ' (j iPo) Ytj the following gives the results of our calculation.

Table V II

r. 10" sec. for Succ. Fluorescein in Glycerine water.

Fxciting wave length A 4358 A

Viscosity in Poises.

T-10* in sec.

Concentration

•08 X io^6gni/c.c. Concentration

'8 X io"6gm./c.c. Concentration i-io“® gm./c.c.

74 3 5 9

■39 • 3 5 9

•ax 3 5 ⁸

•17 3 5 9

'13 3 5

•10 4

P o la rised F lu o r e s c e n c e o f O rg a n ic C o m p o u n d s 3 6 /

TABI.K V I I I

T. lo* sec. for Sue, Fluorescein in Glycerine water, Exciting-A 3650 A

Viscosity in Poises.

T-io^ in iSec.

Concentration

4-10'® gui./c.c. Concentration

■8 X 10"® gill,/c.c.

i ‘8- 15

-78 15 5

•39 ■ ⁶

•21 ^... 5

TABLtt I X

Values of r. 10 ” in seconds for Fluorescein in Glycerine water. Kxciting-A. 4358 A T.io® in sec.

Viscosity in Poises.

Concentration Concentration

i-io~5 gm /c.c. •o8 io"5 _gm./c.c.

5-7 9 3

1-8 9 4

•78 8 3

»2I 8 3

'I7 9 3

... 3

■ 10 ^... 3

-086 ^... 4

Table X

Values of T, 10* in seconds for Fluorescein in Glycerine water mixture.

Exciting wave-length A 3650 A __________________

Viscosity in Poises.

T-io* in sec.

1-8 -39

•3X-17

Concentration i io’ ® gm./c.c.

Concentration

•o8*io"® gm./c.c.

9 3

9 3

8 3

...

Thus u e find that within the limit of the experimental errors the values of r IS practically the same for all the viscosities, though the values are widely different for the different concentrations of the fluorescent compounds. Moreover, the values of r as calculated from the Perrin's equation are independent of the exciting Wave-length. ,

a

362 S. M . Mitra

D E P E N I) K N C H o r T H E r O E A R 1 S A T I 0 N O N T H E T E M r E R A T U R E

The influence of temperature on p will be two-fold ; directly owing to the greater thermal agitation at higher temperatures the ‘ expectation ’ of rotation of the molecules after excitation, from their initial positions, will be the greater and hence the value of p correspondingly smaller. It also affects indirectly by changing the viscosity of the solution, the effect of which is also in the same direction as the previous effect, viz,, to diminish as higher temperatures. Both of these influences are taken into consideration in Perrin’s theory. If these are the effects of temperature, p calculated from Perrin’s expression ought to be inde­

pendent of temperature. Conversely this independence, if established experi­

mentally, may be taken as an indirect proof that there are no other effects of temperature, e.g., through increased collisions between molecules, etc.

lixperimentally the value of p was measured for temperatures from o*C to about loo^C. The teclinique of the measurements is very simple and need not be described here. The solution was kept in a glass bulb inside a water tank whose temperature could he regulated as desired. The entrance as well as the observation sides were double walled and the space between the two walls was maintained dry by pieces of calcium chloride kept inside. This prevented the condensation at low temperatures of moisture on the walls of the vessel, which otherwise would diffuse the light and render accurate measurements impossible. A small heating coil in these two chambers served to maintain the outer wall at the room temperature.

The results of the measurements aie tabulated below, and the values of p are plotted in the accojnpauyiug curves (Fig. 7).

Fi g o t e 7

i--c=i X10*® gm./e.c. ?—c—4X gm./c.c. 3—c«»>8x io"S gni./c,c.. 4—c=ij x io~* gtn./f.c.

Values of p. lOo for Fiuore.sceiu in (llyceriiie. Kxcitatioii by A 4358 A P ola rised F lu o rescen ce o f

Organic

Takle X I

363

reniperature ®C.

Percentage of Polarisation.

c — gm./c.Q.

0

10

20

30 40 50

60

70 80

go 98

41 34 26 17

8

5 3 Almost zero

8 1 0

50 50

37 32 25

20

T2 5 Almost 7xro

1

4-10 50 50 48 48 36 29 26 20 13 5

I'JU

50 50 50 45 42 3Q 52 26 20 9 7

50 50 49 45 39 34 26 17 13 I'AUtK X II

Values of p. 100 for Fluorescein in Glycerine. Fxcilation by A 3650 A

Temperature ®C.

Concentration of dyestuff in lo gni. per c.c.

8 4

10 45 45

20 11

1 45

30 i1 41

40 ³⁴

SO 27 ²⁶

60 ^{20 18}

70 ^{12 .. .}

«o 5 ^{.. .}

90 ^{4 ••}

98 Almost zero

T^able X I I I

Values of p.ioo for Succinyl Fluorescein in Glycerine Excitation by A 4358 A

364 S. M . Mitra

Temperature in *C.

Concentration of dyestuff in 10 ® gm./c cj

II 8 4 ^I

0 50 50 50 50

10 48 so 50 50

ao 43 45 50 50 :

30 ²⁸ 37 45 ^{48 \}

40 19 33 36 40 ^'i

50 8 ₃₅ 30 37

6q 5 ²⁰ 28 33

70 3 12 22 34

80 Almost zero 5 ^5 ¹⁸

90 1* Almost zero 5 ^5

98 ^{... ...} * 10

Table X IV

Values of p 'lOO for Succinyl Fluorescein in Glycerine. Hxcitation by A 3650 A Temperature m "C

Concentration of dyestuff in 10'® gm/c.c.

II 8 4 1

0 ₄₂ 42 43 43

10 _{42 42} 43 43

20 40 43 43 ^ 43

30 28 ₃₂ 37 40

40 ^7 23 28 33

50 17 21 26

60 4 12 ₁₅ 20

70 Almbs ^zero ₇ II 14

80 ⁰ 6 > ^' 10

90 ^... 0 10

Polarised Fluorescence of

Values of j? ^ 100 for Rhodaiuine B in Glycerine. Kxcitatiou by X 5461 A

Temperature in "C.

Concentration of dyestuff in lo"® gm./c.c.

o xo

ao 30 40 50

60

70

80 90

97

39 37 33 37

31

14 9 7 5 o

41 41 41 34

29 26

21 17 15

13 s

41

41 41 39 36 3Jt

27

23 19 14 7

41 41 41 41 37 34

29

24 17 M

T

XVI

Value oi p X 100 for Rliodainine B in Glycerine. Excitation by A 4358 A

Temperature in * ’ C.

Concentration of dyestuff in lo"® gni./c.c.

8 4 2 _'5

0 22 22 22 22

10

22 22 22

30 22

22 22

30 1 9

20 22

1 5 1 7 1 9

20

5 0 XO 1 3 15 *7

60 7 II _{1 3 1 5}

70 4 ^{8 10 *3}

8 0

0

90 ^{• •a 8}

J66 S. M. Mitra

The general form of the curves is the same as for the curves representing the variation P with viscosity. At low temperatures p tends to reach asymihotically limiting value po as for very high viscosities, while at hjgh temperatures yj tends to vanish. On calculating the values of t on the basis of Perrin’s expression we gel the following results.

It is obvious from the nature of the expression that for values of p very near po as also very near zero, the calculated value of r cannot be reliable. Hence P has been calculated only for values other than these.

Table X V II

Values of r X lo ” for Fluorescein in (Jlycerine in secs.

Excitation by A, 4358 A

Concentration of dyestuffin

gtn/c.

T eiiiD erfttu rc

in "C.

t,

12 1

1 S

30

83

40 75 ^9

8

#

50 ^{72 18}

8

60

17

70 6

80

Table X V III

Values of r X 10® sec. for Fluorescein in Glycerine.

Temperature in “C.

ConcentratSon of dyestuff in lo'* gm/c.c.

8 4

30 1 ^—

46 1 18 IS

so ! 17 16

60 1 16

70 I ^{1 12}

Values of r X lo ” sec. for Succiuyl Fluorescein in (Mycenne.

Excitation by A. 4358 A

P ola rised F lu o re sce n ce o f O rg a n ic C o m p o u n d s T^a13I,K X IX

367

Temperature in “C. Concentration of dyestuff in if)“ ® gm/c,c.

10 8 4 1

30 8x 18 _

-10 7.S 15 9

so 70 18 i6 9

60 _{— 10 16} 8

70 iS 16 9

--- --- _{________ '} . ... . .

Table X X

Values of r X 10® in sec. for Succinyl Fluorescein 111 Glycerine.

Excitation by A 3650 A

Temperature in *C.

30 40 .so 60

Concentration of dyestuff in iO“ * gm/c.c.

18

^7 18 18

15 16 15

It is interesting to note that though the value of t shows a very striking dependence on the concentration of tlie dyestuff, it is practically independent of temperature except in the case of the strongest solution, v iz ., of concentration 12 X io~®gni. per c.c. This shows definitely that only effects of raising the temperature are through increased thermal agitation and the consequent larger value of expectation of Brownian rotation of the dye-molecule per second and through diminished viscosity.

In the case of the very strong solutions, however, the value of r tends to diminish with rise of temperature (sec Table X V II, Col, 2).

368

Measurements have also been made with other solvents excited by different wave-lengths. The results are included in the following tables and the changes with temperatures are graphically shown in the accompanying curves (Figs, 8f q).

F'.iJjriEJCEtH IN Castor oil

Figure 8 I—c»4Kio *gm/c.c.

3—C —8x10*® „

Fluorescein

IH

FIG0RB 9

Polarised

of Organic Compounds TABtB XXI

V'alues of P for Fluorescein in Castor oil.

Excitation by A 4358 A

3 69

Temperature ia "C.

o 10 20 30

40

5o 60 70

80

90

Concentration of dyestuff in to" “ gin/c .c.

45 45 44 36 27 20 15 11

8

5

45 45 45 43 32

26 22 IQ

i6 14

45 45 45 43

34 30 27 25 23 TABtE X X I I

Values of P for Fluorescein in Sugar mixture.

Excitation by A 4358 A

Temperature in

Concentration of dycvstuff in io“® gm/c.c.

8 4 ²

0 45 45 45

10 45 45 45

20 42 45 45

$0 34 ^{38 41}

40 25 29 31

50 20 24 27

60 17 20 23

70 H ^{17 20}

80 IX 15 18

. 90 9 ¹²

m

Values'of r- lo ” for Fluorescein in Castor oil.

Excitation by A, 4358 A

/ S. M.

TAAI.E X X III

Temperature in "Q Concentration of dyestuff in lo-'^ gm /c c.

30

40 60

14 15 M

9 II 10

We have also calculated the value of t in the case of fluorescein in castor oil, collodion ether and gelatine water mixtures. The following table gives the values of r for fluorescein in different solvents when excited by A 4358, the concentration of the dye and the temperature being same in all the cases

(c = 8 X lo**'''" gm./c.c.).

Tabt.k X X I V

0 Values of r. lo^’ Sec. for Fluorescein in various solvents.

Excitatibu by AI4358 A

Solvent. . . . T. io9 _Sec.

T, Olyccrlne 18

2. Cfistor oil 14

3. nther collodion 16

' 4. Qelatine water ²⁰

— ---L_____ _________ __________ ________

This shows that the axerage life depenc|s on the solvente.

1

N F L II K N C E O F CON C;,E N '

' R A T I O N O F T HE F L U O R E S C T N C C O M P O U N D ON THE POLARISATION.

We have already seen that the polarisation of fluorescence shows a conspi­

cuous change witl^ the change of the concentration of the fluorescing compound.

The results of our measurement are given in the following tables and are graphi­

cally shown by the accompanying curves ( F i g s . i o . i t ) , ^

Polarised Fluorescence oj

371

Figure io

Figure n Takee X X V Fluorescein at 30® C

C X

Io

gUl/c.C.

8

_ ■

1

^{'25 !}

^{■14 .'07}

Percentage of 4358

'42

46

« 50

polarisation for 3650 36

'

41 42

^ 1 2 S .

Af.

T^able X X V I Fluorescein at 50' C c X ic)5 gni/c.c. 1

1 8 4 ^I •08

Percentage of polarisation

for 4358 25 29 39 45

Table X X V I I Rhodainine B at 30° C

c X io5 _{gni/c.c. 8} 4 ² ’5

Percentage of polarisation

for S4_C’J 32 34 39 4_^

Table X X V III lioshi at so” C c X loBgm/c.c.

^ 1 ^{2 J '5 ■ 25 ■ 16 '08}

Percentage of polari­ 5461 37 41 44 44_'S 45 ⁴⁶ 47

sation for 4358 41 44 45 ⁴⁶ 47 50 50

3650 22 22 24 25 27 29 —

T^able X X I X Magdala red at 3o°C

c X lo® gm/c.c. 4 ² I ■5 2 5 08

Percentage of polarisa­ 5461 39 41 ^{4 2 44 47} 50

tion for 4358 ^{21 24} 27 31

3650 ₇ ro ts 17 ^{— —}

Polarised

of Organic Compounds 373

T^able X X X

Sucdnyl Fluorescein at 30° C

c ^ 10® gin/c.c. 11 8 4 1 1 8

1^ _____ •08 Percentage of po­

larisation for 4358 ²⁸ 37

...¹

1

1

^ 7 50 50

Ta b l e X X X I

Succinyl Fluorescein at 70° C

c X TO® gin/c.c.

Percentage of polari­

sation for 435^^

11 8 4 1

3 12 3= 24

It will be seen from the graphs and the tables that witli increase of concen­

tration the value p tends to vanish, while at very small concentrations it tends to reach asymptoticaly the same limiting value as is reached by p at large viscosi­

ties or low temperatures.

This result is very surprising. We would naturally expect that when we increase the concentration of the fluorescent dyestuff, either the effective value of 7* remains unaltered or probably due to increase in the number of collisions between excited molecules there will be a small tendency for the excited molecules to dissipate their energy in the form of kinetic energy of translation of the molecules. This would be equivalent to diminishing the mean period of duration of the molecules in the excited state. This is indeed the case as is shown by the direct measurements on the duration of the excited state.

This diminution in the mean life of the molecules in the excited state at high concentration of dyestuff w^ould conduce on the basis of Perrin‘.s theory to an increase in the value of the polarisation, since the molecule would not have sufficient time to rotate far from its original orientation where it was excited ; and thus the polarisation will approximate closer to its upper limitiug value. But our experimental results show quite the opposite effect.

TAPI.E X X X I I

Variation of r with the concentration of fluorescein in glycerine at 40^^ C

Concentration x 10 Sgm/c.c r, 10® RCC A. 4358 A

12

8

4X

’o8

72 18

17

7 3

374

Variation of

with concentration of Succinyl.

Fluorescein in Glycerine water at 3o*C.

S. M. Mitra T

« XXXIII

T X 10® 1sec.

Concentration X lo"* gm/c.c.

A 4358A A 3658A

1 , /

•o8 3 \

*8 5 6 \

i

1 ₉ 15

T

XXXIV

Variation of

with the concentration of Succinyl Fluorescein in Glycerine at 4o°C.

Concentration

II 8

4 I

T X U)’* sec.

A 43.sS A

75

19 15

9

ON T IIIC M A X I M U M V A I, U E O F T H E P O E A R I S A T I O N “

We have already mentioned that the value of fo is the value of the polarisa­

tion when the molecule is not allowed to rotate from its initial orientation, which is also evidently the limiting value of

when the viscosity is very large or the temperature very low. The following table shows the values of for the various dyestuffs in various solvents when excited by radiations of various wave­

lengths as obtained experimentally;—

The values of “ Po” (f>o x lOo)

Polarised Fluorescence of Organic Compounds T^abi,^e

XXXV

375

Dyestuffs Solvents

Fxcitiiig Wave-leiigih

5461 435S 3650A.

Fluorescein Glycerine ^ 50 45

CuvStor oil — 45 41

Sugar k water — 45

Gelatine ^k water 45 3«

Collodion & ether — 40 35

Succiiiyl fluorescein (Tlycerine — 50 42

Collodion k ether — 35 30

Rhodamiiie Glycerine 35 15

Collodion k elhcM' ₄₄ 3^^ 10

Acsculin Glycerine — 40 45

Collodion & ether — 35 41

Hosi n Glycerine 47 50 4a

Castor oil 45 48 30

Sugar & water 45 48 30

Collodion k ether 40 45 V

The foregoing table conclusively shows that this limiting value of the polari­

sation " p o ” depends on the following factors, viz.:—

(a) On the wave length of the exciting radiations ; {/)) On the nature of the fluorescent molecules ; (c) On the irature of the solvents,

U L T R A-V I O Iv n T a b s o r p t i o n OF T H E D Y E S T tj F F S I N S O L U T I O N A N D T H E N E G A T I V E P O L A R I v S A T l O N The literature on the absorption spectra of the dyestuffs in solution reveals that the great majority of the investigators in that particular dominion confined their investigation within the visible region of the spectrum in order to establish (jorrelatioQ between the absorption, colour and the structure of the molecules,

5 7 6 5 . M . M itra

Very few systematic (juatititalive measurements have been made in the ultra­

violet region. It might not be out of place to mention here in passing that Gnsebach''^ measured the absorption coefficients in the ultra-violet region of u few dyestuffs in solution and tried to establish a relation between the absorption curves and the polarisation curves of the fluorescence emitted by them in solution. In view of the scanty data as well as to find out whether there is actually any relation between the absorption and the polarisation curves as reported by Grisebach, we carried out detailed quantitative measurements on the absorption of the dyestuffs in glycerine solution, the following gives graphically the results of our measurements.

The absorption coefficients were determined by a calibrated rotating sectoy photometer used in conjunction with an Adam Hilger quartz spectrograph. The, results were, however, further verified by the measurements of absorption with a .sensitive Moll thermoihle and a galvanometer system. A parallel beam of light from a point source of mercury-quart/, lamp was allowed to fall on a cell containing dye solution through a combination of filters, which allowed a mono­

chromatic radiation to pass. The incident and transmitted radiations were measured with the help of a Moll thermopile and galvanometer. The absorp­

tion coefficients, <• were calculated from the relation.

0

Itruiimitted — Iinoid«nt where d is the thickness of the cell.

The results from thermoelectric measurements and photographic method are given below for comparison- The agreement, as will be evident from the follow­

ing tables, is within the experimental errors:—

Table X X X V l

I^vestuffs.

j. FluorevSccin 2 Kosin 3. Aesculin 4. Rhodiilin orange 5. Acxiflavin 6. Magdala red

Absorption co efficient for A 313 nifi

I'hermoelectric method. Photographic method

■ 657

■ 561

MI

■ 139

■579

M9*

*659 Mi6

’13

■58

Polarised Fluorescence of Organic Compounds 377

In the accompanying graphs are drawn the absorption coefficient curves as well as the polarisation curves. (Figs. 12-18.) In the case of fluorescein, we find that the position of the maximum negative polarisation corresponds to the junction of the two adsorption bands in the ultra-violet, as was observed by Grisebach. (See Pig. 14.) But in the case of eosin and magdala red it corres­

ponds to the first adsorption maximum in the near ultra-violet. (See Figs. la, 13.)

Figure 12

FiGURB 13 fO

378 S . M , M itra

40

KO

20

3s

k.O

FlGUMt 15

Palarised Fluorescence of Organic Compounds

F GDKH 17

3 8 0

S. M. Mitrd

O

OUi O<

Fi g u r e i8

R M L A T I O N B E T W E E N F L U O R E S C E N T B A N D . A B S O R P ­ T I O N B A N D A N D M A X I M U M N E G A T I V E

P O L A R I S A T I O N

We have seen that the value of the polarisation decreases as the wave-length of the incident exciting radiation is decreased and then increases with the further decrease of the wave-length of the exciting light. This minimum value is nega­

tive and the maximum negative value is reached for wave-lengths which are different for different dyestuffs as will be evident from the following table: —

Ta b l e X X X V I I

Dyestuffs in Glycerine solu­

tion ^ c —4 X io"6 gm/c.c.

Wave-length for the excitation of the fluorescence of the max.

negative polarisation in tn/*

Band maxima in Fluorescent. Absorption.

Aesculln _{^ 7 5} 459

Fluorescein 300 SI2 sou

Eosin 31a ₅₃₅ 52s

Riiodainine B 325 560 535