exams!) STATE INTRAMOLECULAR seem-27mm RELAXATtON POTENTIAL use In menoewmm meme Thesis for the Degree of M. S. MICHIGAN STATE UNIVERSlTY JOSEPH KORDAS 1974 ABSTRACT EXCITED STATE INTRAHOLECULAR GEONETRICAL RELAXATION POTENTIAL USE IN HICROFLUIOITY PROBING BY JOSEPH KOROAS Trans-l,l,h,h-tetraphenyl-2-methyl butadiene (TPNB) undergoes an lntramoiecular twisting relaxation after excitation. The extent of this relaxation depends on the viscosity of the medium. The fluorescence energy maximum exhibits a blue shift and the fluorescence intensity is greatly enhanced as the medium becomes rigid. we have focused our attention on the quantitative aspects of fluorescence energy and intensity dependences on viscosity and tem- perature of the medium with the purpose of explaining the effect of each of these parameters on radiationless processes of the excited state of TPNB and similar molecules. Of particular interest is the separation of viscosity and temperature effects. thle the fluorescence energy depends only on the viscosity of the medium, the quantum yield depends both on viscosity and temperature. Fluorescence intensity variation as the viscosity is changed by lowering temperature is interpreted in terms of solvent as well as solute activation energies. Solvent activation energies correspond to viscosity activation energies obtained from the macroscopic viscosity dependence on temperature. Solute activation energies correspond to torsional frequencies in the excited state of TPHB. Joseph Kordas A selective red-quenching mechanism is proposed to account for anomalous shifts that are dependent on the temperature and polarity of the medium. Selective red quenching appears to be a general phe- nomenon occurring in situations where excited molecules undergo geometrical relaxation during their lifetime. EXCITED STATE INTRANOLECULAR GEONETRICAL RELAXATION POTENTIAL USE IN NICROFLUIOITV PROBINC By mo Joseph Kordas o , ‘ (~-‘*( A TNESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Biophysics i9?“ To my wife ACKNOHLEOGENENT i would like to thank my research adviser, N. Ashraf El-Bayoumi. for his enthusiasm and guidance. I would also like to express my gratitude to my thesis cormlttee mbers Dr. Alfred Hang, Dr. Ahsan Khan and Dr. Peter Wagner. Special thanks are due Dr. Kenneth lngham and Dr. Fred Hatson for their experimental assistance and invaluable discussions. 4 This research was supported by the National institute of Health Training Gran GHOthZ and by funds from the College of Osteopathic Medicine and the College of Human Medicine of Nichlgan State University. TABLE OF CONTENTS RAGE LIST OF TABLES.. ........ ........................................ v LIST OF FIGURES.................................................vi CHAPTER I GENERAL INTRODUCTION....................................... l Polarization.......................................... 3 Time Dependent Polarization........................... 6 Excimer Formetion..................................... 3 Geometric Relaxation..................................ll CHAPTER 2 EXCITED STATE INTRAMOLECULAR TORSIONAL RELAXATION: VIS- COSITY, TEMPERATURE AND MEDIUM EFFECTS ON THE FLOORESCEMCE CHARACTERISTICS OF A STERICALLY CRONDED NOLECULE...........I3 introduction..........................................l3 Experimental..........................................l8 Viscosity Dependence of Fluorescence Energy...........2l Fluorescence Intensity................................25 Selective Red-Quenching...............................30 Medium Polarity Effects...............................32 Discussion............................................33 CHAPTER 3 CONCLUSIONS................................................39 n'u'mumYOQOQOOOOO0.00000000000000000....OOOOOOOOOOOOOOOOOOOO“ ' Iv LIST OF TABLES PACE TABLE l. Fluorescence Energy Maximum 3F and Half Band Nidths I} of TPMB ln 3MP, Paraffin Oil (P.0.) and a Plastic Matrix at var‘ous Tmporaturesfi00.0.00.O...OOOOOOOOOOOOOOOOOOOOOOZS FIGURE I. 2. LIST OF FIGURES PAGE Selective Excited State Processes........................ A Emission spectra of TPMB in 3-methylgentane at room temperature (fluid medium) and at 77 K (rigid glass). Emission at room temperature is much less Intense and Is recorded at a higher sensitivity.........................IS Qualitative potential-energy curves for the ground and excited states of TPMB. Emissions from a relaxed state (fluid medium), a Franck-Condon state (rigid medium) and an intermediate state are demonstrated...............l6 Viscosity-temperature data for triacetin, 2fpropanol and methylcyclohexane/isopentane mixture 3:l. These data are taken from the references cited In the text.....l7 The variation of the energy of the fluorescence maximum (UP) in three different solvents as a function of viscosity. The viscosity was varied by lowering the temperature.........................................22 The variation of the energy of the fluorescence maximum (vF) as a function of temperature in four different media. Arrows indicate the tsmperature at which the viscosity of the medium is ID cp. The dashed line indicates the values of vF in polystyrene film at different temperatures..................................23 The change of fluorescence Intensity of TPMB in ‘gypropanol and in MCH/IP (3:l) as a function of viscosity. The intensity is measured relative to the value at 298 K in‘nrpropanol........................27 The change of fluorescence Intensity of TPMB in gfpropanol, MCM/IP (3:l) and IP as a function of temperature. The intensity is measured relative to the value at 298 K in‘grpropanol.....................28 Emission spectra in MCMéIP (3:l) at three different temperatures 298°x, I60 K and 77°K. The relative sensitivities at which the spectra were recorded are XIBO, X7. XI respectively...............................3“ vi FIGURE IO. RAGE Emission spectra at 298°K in three solvents, MCH/IP (3:I),‘grpropanol and ethanol. The emission in ‘nypropanol and ethanol were recorded at the same sensitivity (XI) while that in MCH/IP (3:l) was recorded at a lower sensitivity (X.5)..................35 A plot of log (l -I) vs. T“ for TPMB in three solvents MCH/I 3:I),‘g¢propanol and iP were F is the absolute quantum yield at temperature T. The actlvgtion energies determin d from the plots are 890cm , l530cm and 7l0cm respectively.........38 le CHAPTER I GENERAL INTRODUCTION Many membrane processes are dependent on the fluidity of the membrane (I). Membrane fluidity determines the passive permeability of both Ionic and neutral species. It influences active transport-- the movement of carrier species across the membrane or the rotation of carrier enzymes. The ability of a membrane to undergo a confor- mational change also depends on its fluidity as does cellular develop- ment. In fact, any enzymatic process that occurs at the membrane surface may be affected by the membrane's fluidity. Many studies have been done to demonstrate both the fluidity of membranes and the liquid crystal nature of membranes, selective examples follow: I. Spin Label Studies. a.) McConneI et ai (2,3,h) found the rate of rotation of a nitroxide radical depends on its attachment site on the fatty acid. The closer It Is to the polar and the less it spins. b.) when the radical is attached near the hydrocarbon and, the medium is found to be more fluid. 2. Rhodospin Studies. Cone et al have (5) demonstrated that rhodospin is able to rotate In the plane of the rod outer segment disk. Also Blasie (6) found that rhodospin sinks Into the hydrocarbon region of the membrane upon bleaching. 3. Cell Fusion Studies. Frye and Edidis found (7) that after fusing tissue culture cells of mouse and human origin by using Lendai virus the antigens present on the surface of the two cells exchange and result In a mosaic. Their conclusion is that the antigens are free to diffuse along the surface of the cells (lateral diffusion). 4. Model Membranes Studies. Weber and Shinitzky (8,9) have used fluorescene depolari- ation (steady state) of perylene and 2-methylanthracene to investigate the fluidity of the interior of micelles. They found that the depolarization of these molecules In micelles is the same as that in hydrocarbons with mascroscoplc visclsities of between l7-SOcp. They found also that cholesterol Increased the rigidity of the model membranes. Cilter (IO) using the same technique, found that the depolarization of these molecules In hemoglobin free erythrocytes is equivalent to that in a hydrocarbon with macroscopic viscosity of l00-200cp. These studies have shown that: I. Membrane Interiors and fluid and that this fluidity depends on the composItIOn of the membrane. 2. Many biological processes are very dependent on the fluidity of membranes. It Is therefore worthwhile to Investigate possible methods of determing membrane fluidity. Many people (ll,I2,I3,lh,l5) have Investi- gated membrane and model membrane structure using ESR and NMR. Both 3 of these techniques are widely used. However optical methods also have great potential for such use. I will discuss some of the possible fluorescent methods and then in the next chapter characterize a sterically hindered molecule which might be a possible optical probe for environmental fluidity. There are several excited state processes that maybe useful for Investigating medium rigidity. Some of these are listed in figure I. Polarization I The most popular fluorescence technique for investigating membrane structure is steady state fluorescence depolarization. Molecules absorb light only when the electric vector of the light is parallel to the transition moment of the molecule. They also emit light with the electric vector parallel to the transition moment of molecule. Therefore if one excites a random rigid sample with polarized light, the fluorescence will also be polarized parallel to the exciting light. However, if the molecules are free to rotate and rotate rapidly com- pared to the fluorescence lifetime the fluorescence will be depolarized- molecules are able to randomize before emission. Perrln (l6,l7) derived the following relationship: (l/p - l/3)- (I/Po - 1/3) 0+6er where p-(IH - '1.) / (lII + l_L ), Po - limiting polarization, 1' - fluorescence lifetime and R - rate of rotation. Einstein (l8) found that If the molecule is spherical and in an Isotropic medium R-l/6(kT/17V)‘where VI volume cf the molecule. mummuuocm uhdfim nub-uxu u>.huu4mm . mxau.m - - $895 8?on ~58 9598 598 98% 55% 8:55 4 . _ . _ . _ _ . . u . . . . _ . . _ . . . . _ . — — - . . _ u _ _ n 2035942 _ . Emfiom . . 203.335 . 35.888 . . ~3sz _ t - m 555 _ s , I 28258 «5sz . \. to goxm 28.8: p \ \ . \ newum ommnoxm \ \ A \ - 23.8: s 323952 5 Hence by determining the fluorescence polarization p, one can get an approximation of n . In order to get p0, p must be studied as a function of TI?) and extrapolated to Tim -0. This type of measurement cannot be made with biological systems. Typically one determines p In the membrane environment and finds a hydrocarbon In which the polarization is the same. One can then say that the molecule experiences a rotational resistance similar to that In a hydrocarbon of a certain macroscopic viscosity. Weber and Shinitsky (8) using perylene and 2-methyl anthracene, found that for micelles made from the series of detergents lauryltrl- methyl-ammonuum bromide, myristyltrimethylammonium bromide, cetyitri- methylammonium bromide and stearyldlmethylbenzylammonlum bromide, the polarization at 27°C is equivalent to that In a hydrocarbon*wIth a macroscopic viscosity in the range of l7-50cp. The change in fluidity with temperature was found to follow a single exponential with an activation energy In the range of 6.I-9.6 kcal/mole-'. Also they found that the mlcrofluldlty Increases rapidly with the addition of cholesterol. Cogan et al (9) has used this same technique to Investigate the mlcroenvlronment of lecithin dispersions and observes a phase transition for the dlpalmltoyllecithln dispersions. Perrln's relationship only holds for spherical molecules. Most flurescent aromatic molecules are planar rather than spherical and therefore have two axes of rotation, In-plene rotation and out-of-plane rotation. Heber_(8) had derived the following relationship for this 6 ro/r - went (RP(2cosza -l) + no cosza ) / (3cosza -I) p where r is the anisotropy defind as r- ('H '{L ) l ('H +2IJ ). Rp is the rate of rotation about an axis normal to the ring plane (in-plane rate of rotation), Rop is the rate of rotation about an axis contained in the ring at right angle to the absorption oscillator, A is the rate of fluorescence of the molecule and a is the angle between absorption and emission asclllators. This relation reduces down to an equation In only one rate of rotation for particular values of po: when po -— l/7 rO/r-l+6R°P/)( when po - ~l/3 rolr-l+6RP/)( Therefore from studying the polarization at wavelengths of excitation at which pd-l/7 and p°--l/3. one can obtain the ln-plane and out-of- plane rates of rotation and learn both about the mIcrofluidIty of the environment and the degree of anisotropy of the medium. Thomas studied (l9,20,2l) the steady state polarization of pyrene and 2-methyl anthrancene to determine the mlcrovlscosity of micellar Interiors and Escherichia cell membrane vesicles. He finds that the polarization date agrees wIth fluorescence decay quenching data. Time Dependent Polarization Jablonskl (22) In l96l derived a relationship for the time dependent anisotropy which he defined as A(t)-(I” (t) - '1 (t))/(I ” (t)+2 ll (t)). He found that for rigid spherical molecules In an isotropic medium A(t) IAoexp(-t/¢ ) where di-VWi/kTiI/6D, Is the rotational correlation time, A°-.h(3cos%a -l)/2. a Is the angle between emission and absorption moment, and D Is the rotational diffusion coefficient. For ellipsoids the expression becomes more complicated (23,2h,25,26,27). 3 A(t)-A°:E:f'exp(t/¢>) i=1 Here (b's are related to the rotational diffusion coefficients of the ellipsoid. This simplifies to Aoexp(t/¢ ) If the major axis Is parallel to the emission moment. For completely asymmetric molecules A(t) Is a an of five exponentials (23,210. By studying the time dependent anisotropy, one can determine If the molecule which Is rotating Is spherical and If the medium is isotropic. If this is the case, one gets a single exponential decay for A(t) and can then determine 4) and from 4) get 17 . Until recently, time dependent anisotropy measurements have only been made on large macromolecular systems (23). For Instance, Stryer (2h) studies the time dependent anisotropy of dansyl-lyslne bound to an antibody. Also Hahl and TImasheff (28) used fluorescence time dependent anisotropy to study the aggregation of fii-Iactoglobulin A. From the rotational correlation times obtained, they were able to estimate the size of the various aggregates. Recently Vanderkool and Cehelnlk (29,30) have begun to use time dependent depolarization of small molecules to probe liquid crystal structure (29.30). Cehelnlk studies the time dependent anisotropy of 8 ’ all-trans-l,6-diphenylhexa-I,3,5-triene in several solvent systems. In methylcyclohexan at 25°C, he found rotational diffusion to be complete within l nano-second after excitation, while In paraffin oIl the rotation correlation time -h.83 X l0'9sec. He also found that for a nematlc liquid crystal medium (cholesteryl laurate/ cholesteryl chloride) orienated by an electric field A(t) remains constant for a period of A) lifetimes ( T nl3.0nsec in methylcydohexane). Vanderkool uses l2-(9-anthroyl) stearic acid to investigate the fluidity of phospholipld dispersions and red blood cell membranes. She found that the rotational correlation time, for l2-(9-anthroyl) stearic acid 9sec In normal red blood fluorescene anisotropy at 37°C is 7.8 X lo. cells and 8.5 X IO-9sec in blood cells containing twice the normal complement of cholesterol. Time dependent depolarization measurements may be a very fruitful method of investigating membrane environment; however, the ratio of iblr llmItes this technique. If the lifetime ls long compared to the rotation correlation time, fluorescence will be depolarized at all observable times after excitation. This Is the case for the simple situation of an aromatic chromophore In a fluid solvent at 25°C. On the otherhand, If the lifetime Is short compared to the correlation time, emission will be completely polarized at all observable times after excitation. The latter is the case for dlphenyl- hexatrlene in the liquid crystal medium. I Exgimer Formation Another excited state process which is a potential monitor of environmental fluidity or "mlcrovlscosity" Is excimer formation. 9 There are two types of excimers, Intermolecular and Intramolecular. In the first case the excited complex consists of two separate Identical monomers, one Initially In an excited state and the other In the ground state. In the second case, the excimer is formed by two non- Interactlng chromophores which are segments of a single molecule. Intermolecular excimer formation ls concentration dependent, while Intramolecular excimer formation is concentration Independent. The formation of intermolecular excimers in fluid membranes or any other medium Is a diffusion controlled process (30,3l,32). The value of the coefficient of lateral diffusion, Ddiff, Is obtained from the second order rate constant of excimer formation which can be obtained from the ratio of the excimer to the monomer fluorescencd yields (3|). A value for the mlcrofluldlty of the medium may be obtained directly from the Einstein-Scholuchowski diffusion theory (32). It relates the rate of diffusion to the viscosity of the medium by the expression: It. year/300012 ) (Pa/b) iniwhlch pa/b-l for pyrene. The relationship between the half intensity concentration, C'h, to the viscosity follows: MK max C'h-kf-kr) (300077/8RT) (I. lie ) where lolim-C/Cf, . C'h Is the concentration at which the monomer fluorescene yield is equal to the excimer yield, kf- rate of fluorescence of monomer, krl- rate of radiationless decay of monomer and I. 'NICX O and represent the maxium intrinsic fluourscence. Therefore the ID value of C'h can be used as an Index of viscosity. Pownall and Smith (32) plotted 'el'm versus concentration of pyrene in propanol and obtained C'h from the reciprocal of the slope. They then plotted C'h for several solvents versus the kinematic viscosity of the solvents. They found, as expected, that the relationship between C'h and ‘n Is linear and from this they obtained a value for (kf +krl) (3000/8RT) ). .73 x IO-3. Therefore by obtaining C'h. they determined ('mmaxl'omax the viscosity of the medium. Pownall and Smith used this technique to study the fluidity of the hydrocarbon region of micelles. They found that n varies from lSO-I90cp for micelles formed with various detergents. The pointed out that these values are consistently higher than those found by fluorescence depolarization; however, they offered little Insight into this descrepancy. The advantages of this technique over fluorescence depolarization measurements are: l. Excimer fluorescence is not affected by the depolarizing effect of micellar rotation. 2. Corrections necessary to account for turbid solutions which scatter and depolarlze fluorescence are not necessary. and 3. The concentration of surfactant does not appear to be Important In the dynamics of excimer fluorescence as long as the concentration ls greater than the critical micellar concentration. one great problem with using Intermolecular excimers to Investigate medium fluidity is that the measurement made Is the concentration value where 'e.'m‘ It is very difficult to measure the concentration of an aromatic molecule within such systems as micelles. Many assumptions must be made as to the effective volume of the micelle. Also one must assume that all of the fluorophore goes Into the micelle. If one ls studying 17 as a function of temperature, one must assume that the concentration remains unchanged. The concentration problem can be eleminated by using Intra- molecular excimers. This type of excimer formation ls concentration dependent. However, it Is not strictly diffusion controlled. One might say that it Is a quasi-diffusion controlled reaction. Consider a molecule like dinapthylpropane. The two phenyl rings must diffuse together to form the excimer, but the diffusion is restricted by the methylene chain (33). One way to use such a molecule would be to calibrate the le/lmversus fl , and then use this calibration curve to determine‘n . We plan to do this in the future. Unfortunately this molecule's sensitivity to oxygen complicates the measurement. Oxygen quenching affects the I’ll” ratio. This is, oxygen quenches the excimer more efficiently than the monomer. geometric Relaxation The last excited state process that I will talk about In this Introduction Is geometric relaxation. I will only talk very briefly about It here. This Is the excited state process‘whlch I have chosen to Investigate as a possible method of measuring medium fluidity. In the next chapter I will talk about the characterization of the fluorescence properties of a sterically hindered molecular which might be appropriate as a fluidity monitor. 2 Many molecules‘which normally would be planar In their ground state are forced Into nonplanar geometries due to sterlc hindrance. These molecules In the excited state tend to become more planar. As an example consider tetraphenylmethylbutadiene (TPMD). It is l2 believed that in its ground state one phenyl ring is forced out of the plane of the rest of the molecule by the steric hindrance between the phenyl ring and methyl group (see structure in chapter 2). The out of the plane rotation in the ground state is about an essential single bond. In the excited state this bond has more double bond character; therefore, the excited state equilibrium configuration Is more planar than that of the ground state. The Franck-Condon principle states that nuclear movement Is much slower than electronic movement. Therefore on excitation the molecule Is not promoted Into Its excited state equilibrium configuration but into Its Franck-Condon state. JThe molecule then relaxes into Its equilibrium excited state geometric configuration. One expects the rate of this relaxation to depend on the viscosity of the medium. I characterize the fluorescent pro- perties of TPMB in the next chapteriwith the purpose of exploring the .possibillty of using this type of excited state relaxation to probe fluldlty. l3 CHAPTER 2 EXCITED STATE INTRANOLECULAR TORSIONAL RELAXATION: VISCOSITY, TEMPERATURE AND MEDIUM EFFECTS ON THE FLUORESCENCE CHARACTERISTICS OF A STERICALLY CROVDED MOLECULE Introduction Flexible molecules that undergo lntramolecular torsional vibration around a bond linking two Interacting mioetles of the chromophore are weakly or non-fluorescent. The low frequency oscillations lead to efficient radiationless deactivation of the excited molecule In fluid media. Upon "freezing" these modes or limiting their amplitudes, the molecule becomes strongly fluorescent. This may be accomplished by using media of high viscosities or by rendering the molecule rIgId vIa a chemical bond. In addition to the fluorescent enhancement effect, flexible molecules may undergo a change In their equilibrium geometric con- figuration upon excitation, such as In sterically crowded molecules where the steric strain can be relieved by Internal rotation. In these cases a large Stokes shift of the fluorescence maximum is observed In fluid media. The fluorescence emission may originate from the equilibrium excited state, Franck-Condon state or an Inter- mediate geometric configuration depending on the relative magnitudes of the rate constants of the rotational relaxation process (kr) and fluorescence (kf). lf kr is viscosity dependent one may expect the fluorescence maximum to be a sensitive function of the viscosity pro- lh vlded that kf<:kr. As the medium becomes more viscous the fluorescence maximum shifts progressively to higher energies and a maximum blue shift Is observed in rigid media. In an earlier study (3h) we have shown that the fluorescence maximum of a sterically crowded molecule, namely trans-l,l,h,h- tetraphenyl-Z-methyl butadiene (TPMB), lies 3700 cm-' at higher energies in rigid glass at 77°k compared to Its maximum In fluid medium at room temperature as shown in Figure 2. The results were interpreted in terms of an Intramolecular twisting relaxation process that Is fast in fluid medium. In rigid glass however, the twisting relaxation is slowed down due to the high viscosity of the medium and the emission originates from an excited molecule which has a geometric configuration similar to that of the ground state, i.e. from the Franck-Condon state which lies at higher energies relative to the equilibrium excited state. Potential-energy curves are drawn qualitatively by considering both the variation of the resonance energy and the sterlc energy as a function of the angle of twist about an essential single bond (3“). These curves for the ground and excited state are shown in Figure 3. The curve In the upper state is a steeper function of the angle of twist reflecting the enhanced double bond character of the bond In question. The curves demonstrate the large Stokes shift In fluid medium, compared to that In rigid medium. Emissions from relaxed, intermediate, and Franck-Condon states (shown In Figure 3) demonstrate the blue shift of fluorescence as the medium becomes more rigid. l5 .xu.>_u.aeeu teem.g e um eeecouec a. one 3:3... and. gun... a. eeaueceoaeu .89. on 5.3.5 Anne—m 33... no: em vee A538. 33: 8322.53 soot ue 1.3.322an 5 ate» ea 9.32.. .3325 .N 552... A85 IkOZm4m><>> one 000 own oom 0m;e com a e _ _ . xeoem AllSNEIlNI . Konn l6 ENERGY --— \ 0° _ 9+- \J. FIGURE 3. Qualitative potential-energy curves for the ground and excited states of TPMD. Emissions from a relaxed state (fluid medlimi), a Franck-Condon state (rigid medium) and an Intermediate state are demonstrated. 17 T (°K) 250 200 I67 I43 I25 III l I I l l I 7 - TRIACETIN n- PROPANOL MCH/ISOP 3:1 LOG?) (cpl O l i l I TT l 3.0 5.0 1 7.0 9.0 /T x 103°K" FIGURE h. Viscosity-temperature data for triacetin, gfpropanol and methylcyclohexane/lsopentane mixture 3:l. These data are taken from the references cited In the text. l8 In addition to the blue shift phenomenon, we presently show that TPMB exhibits also a remarkable fluorescence enhancement effect as the viscosity of the medium is increased. We have focused our attention on the quantitative aSpects of fluorescence energy and Intensity dependences on viscosity and temperature of the medium with the purpose of explaining the role of each of these parameters on the radiationless processes of the excited state of TPMB and similar molecules. Of particular Interest Is the separation of viscosity and temperature effects. To accomplish this, we used various approaches namely: I. Use of mixtures of solvents with different viscosities at constant temperature. 2. Use of solvents whose viscosity dependences on temperature are different. 3. Studying temperature effects on luminescence of a sample in a plastic matrix. Exggrlmental Materials. Phillips pure grade 3-methylpentane (3MP) was purified by distillation and by passing through an activated silica gel column. Spectroquality Isopentane (IP), methylcyclohexane (MCN) and glycerol (obtained from Matheson Coleman and Bell Company) were used ‘wlthout further purification. Ethanol, gypropanol and butanol were distilled while octenol and triactln were vacuum distilled. Paraffin oll (obtained from Baker)‘was usediwithout further purification. All solventsiwere spectrally transparent above 350 nm and exhibited no l9 fluorescence when excited above 350 nm. In Figure A the viscosities of three solvents are ploted as log viscosity versus l/T. The data for grpropanol are taken from Denny (35) while that for triacetin and methylcyclohexane/isopentane (3:l) are taken from VonSalls (36). A pure sample of TPMB was further purified by successive re- crystallizations from ethanol. The plastic samples were formed by dissolving both TPMB and the plastic (either polystyrene or poly vinyl acetate) In spectroquality CHCI3 and evaporating off the solvent. A thin transparent plastic film was formed with the TPMB embedded In the matrix. Methods. The Aminco-Keirs Spectrophosphorlmeter (IP28 Phototube) was utilized to obtain the emission Intensity as a function of tem- perature. Relative quantum yields were obtained by measuring areas under emission curves and are uncorrected for phototube and monochromator response. In a limited temperature range, where only little shift of emission occurs, these correction factors are identical. A quartz dewar'wlth a flat quartz excitation'window and a l cm square suprasll cuvette were used. The temperature of the sample was controlled by boiling liquid nitrogen using a power resistor and allowing the N2 gas to flow Into the sample dewar. The temperature of the sample was monitored through a thermocouple (copper, constantan) attached to the outside of the cuvette Immediately above the point of excitation. Fifteen minutes were allowed for equilibration at every point after the thermocouple reached the appropriate temperature. Comparing readings of thermocouple on the outside and Inside of a cuvette containing 20 solvent, one finds not more than IOC difference between thermo- couples over wide ranges of temperature (room temperature down to -i5i°c). . A 750 mm Czerny-Turner Spectrometer (Spex l700-ll) In conjunction ‘with a PAR lock-In amplifier (HR-8) and an EHI 9558 QA phototube was used to obtain the energy of the fluorescence maximum ;.F° These spectra also are presented uncorrected for phototube and monochromator response. Absolute quantum yields for TPMB In various solvents at room temperature were obtained by utilizing a double beam quantum yield Instrument interfaced with a PDP II computer (37). This Instrument measures both emission and absorption and corrects for both monochromator and phototube response. It Is programmed to correct for Innerfllter effects for solutions with optical densities up to l.0. All solutions used here were IO.“ M. The excitation wavelength was again 360 nm. Qulnine sulfate (io'6 M) in I N H280“ was used as the quantum yield standard with a value of 0.5“. Because this apparatus uses phototubes ‘with a weak red response and because the measured quantum yields are very small, these absolute quantum yields are only approximate (:_201). It was noticed that a slow photochemical process occurs as a result of excitation at short wavelengths. This leads to changes In the Intensity of absorption and emission spectra. The excitation wavelength of 360 nmiwas used throughout our study; at thls‘wavelength, minimal changes occurred upon excitation. Fresh undegassed solutions were used. Degassing the samples produced no change in quantum yields. 2l Viscositygpependency of Fluorescence Energy The fluorescence frequency maximum B'F was measured in three different media as a function of temperature. The viscosities of the solvents used, namely gfpropanol, triacetin, MCH/IP mixture (3:l), are plotted as a function of temperature in Figure A. The differences In viscosities of these media at a given temperature enable us to distinguish between temperature and viscosity effects. The fluore- scence energy is plotted vs. the logarithm of viscosity In Figure 5. At fl 2;l00 cp the fluorescence maximum exhibits a blue shift, the magnitude of which Increases continuously in all three media as the I viscosity Is Increased, reaching a maximum value of about 23,200 cm- at very high viscosities. Plots of fluorescence energy maximum, 3} vs temperature In the three media are shown in Figure 6. In general 3} increases rapidly In a narrow temperature region which differs from one medium to the other. In contrast these large changes of 5} occur at nearly the same viscosity. The Inflectlon point of 5% vs. temperature plots occurs at approximately the same viscosity, (the arrows in Figure 6 mark a viscosity of IO“ cp) in spite of the fact that such viscosity is reach at widely different temperatures varying from l20°K up to 2500K, depending on the medium. This clearly demonstrates that 5% depends primarily on viscosity and not on temperature. In mixtures of 3MP and paraffin oil, the fluorescence energy maximum of TPMB remains essentially the some up to viscosities of about 30 cp. In pure paraffin oil at 298°K ( 17-77 cp) the observed 22 .ecaueceoseu ecu mc.eezo_ >4 ee.ee> was >u.eou.> ash .>u.eooa.s we co.»uc:e a me eueo>.0m ueoeoee_e omega c. filmy eel—xaa eueeumeeoe.u ecu uo >meoco ecu mo co_ue_ee> as» .m mama-u 33 e .00.. 2 3 op o m i n o n v m m p o a _ _ u q fl _ q _ _ . 1 0— e i on . e \ e i 8 Wm 0. ‘5 38 w. I .. l eozéomeic a .. mm 3 e152 0 2.591% m e .u .. mm 23 .eeceueeaeeau ueeeeee.e ea 5.: 2.9.31.2. e. uh mo nee-es ego neoeuzc. ac: magnet as... .3 a2 a. 5:33 ago me >u.noun.> ego gu.zs ue ecaueceoeau as» aueu.ee. wrece< .e.eea aceceue.e Laos e. oeeuecooeou mo eo_uueee e no “ADV ase_xae ouceueeeoa_e ecu mo >meace ago do eo.ue.ce> ask .0 u¢30.m XOF oom 0mm oom owp oop fl _ _ _ 4 7m .35: e eoz. 1024 n-PROPANOL t: O) I: w 1.. Z Lu 2 '- l 4 IO m ....l m C! 1001 l I l l i l ..1 0 2 4 6 8 LOG 1) .(cp) FIGURE 7. The change of fluorescence Intensity of TPMB ln‘nfpropanol and in MCH/IP (3:l) as a function of viscosity. The In- tensity Is measured relative to the value at 298°K In gypropanol. 28 lo a 5 ”3 >- l— . '1 — ‘2 0) l0“ E |-— e Z DIP LIJ ' " o MCH/IP 3=l > ' a en-PROPANOL :2 < J " ..J lo' " UJ s 0: I lo° - ' . . . 300 250 200 ISO 100 T°K FIGURE 8. The change of fluorescence Intensity of TPMB in n-propanol, MCH/IP (3:l) and IP as a function of temperature. The. In- tensity is measured relative to the value at 298°K In 2-propanol . .‘ki; “11.3: it fli- 5'- L l ' eel-LT I. A‘I'-“" e...e.._ __ -u—n—"R IE'I.‘ .« -. .1": ‘ ‘~' ‘-' __WLs-_ rI‘ .. _s ’ 'E'r . 44m.“ 1.. lent-.41: -J- - 29 approximate with an error of almost 20%. The fluorescence Intensity (corrected for changes of optical density with temperature) Increases by a factor of 270 In‘gfpropanol In the temperature range (298-I350K) and by a factor of 75 In MCM/IP mixture in the temperature range (298-II7OK). In plastic, the corrected fluorescence Intensity remains nearly constant In the same temperature rante. However, one must note that the frequency maximum In plastic corresponds to macroscopic viscosities where the quantum yield has already achieved It maximum value In the two solvent systems. In order to separate temperature from viscosity effects, the fluorescence intensities were measured In mixtures of paraffin oil and 3MP at room temperature, where the viscosity range varied from 0.5 in 3MP to h5cp in paraffin oil/3MP mixture (9:l). The fluorescence Intensity Increases by a factor of 7.2 over this range In these mixtures. This change Is attributed soley to viscosity. In MCH/IP mixture the fluorescence Intensity Increases by a factor of 33 over the same viscosity range when the temperature is lowered from 298°K down to l60°K: and therefore, one may attribute a factor of “.5 as being due to temperature effects on luminescence yield. This value Is obtained assuming a linear variation of logilwith percent volume of paraffin oil In the mixture. In ethanol/glycerol mixtures where the viscosity at room temperature varies from l.2cp for ethanol to 63cp for ethanol/glycerol mixture (“:6), the Intensity Increases by a factor of 8.7. In an experiment where the viscosity of propanol was Increased from 2.3cp 30 at room temperature to 63cp at 2050K, the intensity has Increased by a factor of l8. Thus a factor of approximately 2 may be attributed to lowering temperature from 290 to 2050K. Here we assume that the polarity of ethanol/glycerol mixtures is Invariant and is the same as that of propanol. Selective Red-Quenching As mentioned before the fluorescence energy maximum 5} decreases as the viscosity is decreased until it reaches a minimum value at viscosities around l00cp where complete relaxation occurs during the lifetime of the excited state. Further lowering of the viscosity should not have any effect on 5%. Indeed If the viscosity Is varied In the range of l-30cp by mixing 3MP and paraffin oil in various proportions at room temperature, 5% remains Invariant. However, In Figures 5 and 6, it Is clear that as the viscosity Is lowered 3} de- creases reachlng a minimum value near viscosities of about IOOcp, but then It begins to Increase upon further warming of the sample. This clearly demonstrates that the increase in 5} In a fluid medium is due to temperature and does not reflect a real change In transition energies. This behavior was also observed for arylethylenes (38) but no Interpretation was given. is Interpret this behavior one should make the following points: I.) Emission Is composite in nature, i.e. It originates from excited molecules which have different geometric configurations depending on the extent of relaxation at the time of the transition. Although in 3i a fluid medium most of the molecules relax to the equilibrium excited state geometric configuration before emission occurs, a few molecules still emit from partially relaxed geometric configurations. This ex- plains in part the fact that the emission In a fluid medium ls broad (see Table I), corresponding to various degrees of relaxation. In a rigid medium emission arises from one geometric configuration which corresponds approximately to that of the ground state. This accounts for the relatively small half width of the emission In a rigid medium at 770K. 2.) The paths of energy degradation (i.e. radiative vs. radiationless transitions) will depend on the excited state geometric configuration. A Franck-Condon (unrelaxed) state ls expected to have a small radiationless rate constant since the overlap factor* of the vibrationiess level of the upper state‘with the isoenergetlc levels of the ground state Is small; the radiative processes will therefore dominate. Therefore In a rigid medium (77>i08cp) the fluorescence ls Intense, the band Is relatively narrow and the maximum occurs at higher energies. The equilibrium excited state geometric configuration ls expected to have a large radiationless rate constant due to a large overlap factor arising from the displacement of the excited state potential surface relative to that of the ground state. There- fore In a fluid medium the fluorescence yield is low, the band is broad and the fluorescence maximum lies at lower energies. Since the fluorescence yield of completely relaxed molecules is very small comp pared to non-relaxed molecules, emission originating from the latter *The overlap factor In the rate'equetlon for radiationless transitions depends on the extent of displacement of the potential energy surfaces of the upper and lower states relative to each other and is expected to be minimum for the Franck-Condon state configuration. 32 will contribute more significantly than their proportions. 3.) In- creasing the temperature will change the Boltzmann distribution of excited state. If radiationless decay from upper torsional modes In the excited state is more efficient (larger rate constant), one would expect fluorescence quenching of molecules emitting at longer wavelengths as the temperature Is increased. Therefore the apparent blue shift which occurs at low viscosities as a result of warming the sample is really a selective-red quenching phenomenon. Increasing the temperature leads to a preferential quenching of fluorescence of completely relaxed molecules. Partially relaxed molecules will progressively contribute more to the emission intensity as the temperature Is increased. Since their emission occurs at higher energies, an apparent blue shift of the fluorescence maximum Is observed as the yield decreases with temperature, giving rise to a selective red-quenching effect. This is demonstrated in Figure 9, where the fluorescence spectra In MCH/IP (3:l) at different temperatures are shown. ‘Mgglum Polarity Effects. Comparing the fluorescence spectra In g-propanol, MCH/IP (3:l) and ethanol at room temperature one notices a decrease in intensity and a blue shift of the 3% as the polarity of the medium is increased. As shown In Figure ID, the half width of the fluorescence band is larger in hydrocarbon medium. One should note that TPMB is practically non-polar and therefore the observed blue shifts In the polar media could not be attributed to energy changes of the states Involved In emission. 33 The selective-red quenching observed when the medium is made polar Is interpreted in terms of a larger solvent-solute coupling (mainly due to dipole-induced dipole Interactions) in polar media. One may therefore expect a more efficient radiationless decay in polar medium particularly for completely relaxed molecules where radiationless processes are dominant. We believe that the selective-red quenching phenomenon observed when temperature is Increased or when the medium is made polar Is a general one and will occur in situations where excited molecule undergo geometric relaxation during their lifetime. Discussion From the previous sections, It Is apparent that the fluorescence Intensity depends on both the temperature and the viscosity of the medium. BY lowering the temperature the upper torsional modes of the equilibrium excited state become less populated and, If one makes the reasonable assumption that radiationless decay from these levels is more efficient compared to the lowest level in the excited state, one would expect the fluorescence intensity to Increase as temperature is lowered. In the viscosity range where no shifts are observed the increase in viscosity will have no effect on the excited state geometric configuration. However, the torsional oscillations will be damped and their maximum amplitudes will be smaller at higher viscosities, I.e. the shape of the potential energy surface will be dependent on the viscosity. In the excited state this will affect the overlap Integrals that govern the rates of radiationless decay. At low 3h .>.e>.uoeenee ._x .sx .om_x ace eeecooec aces ecuuoeu ego su.cz.ue ee.u.>_u.weeo e>.ua.ee on» .xonn ece xoom. .xommN weceueeeoeeu oceeeee_e eats» «a A.unv e.\:ux c. eeuueeu co_ea_Eu .m uc:u.u :5: IHOZml—m><>> com 000 own onv Gov _ fl xann AllSNEllNl 35 .35 5.3323 .332 e as 3.888 as. :5 e582 5 as: 2...: 25 5.329.: as: a...» an tattoo! ages .82.: ace .oceooceifl c. c2323 as... ._oce5e ec- _oeeeP.ei..... .Cunv fixtu: .3530. out.» 5 zooms um 93qu c2325 .2. was: A85 IHOZw4m><>> one 000 own oon omv com a . r. _ _ _ ._OZIO-6M) however not so soluble that It favors the water over the hydrocarbon. #2 2. The rate of relaxation must be sensitive to the packing of the medium. Hopefully, it will be sensitive In a fluidity range corresponding to a macrospic viscosity of 0-l00cp. All work in biological systems has fallen Into this region Investigated by steady state depolarization. 3. The '5} of emission should not depend on temperature so that It may be used to study such phenomena as membrane phase tran- sitlons. h. The quantum yield should be great enough for easy detection. 5. It would be nice if the probe's fluorescent lifetime was long enough so that the actual rate of geometric relaxation could be used to study medium rigidity; however, this Is probably impossible. B I BL IOGRAPHY 9. IO. I3. I“. .l5. l6. I7. BIOLIOGRAPHY Gitler, C., (I972) Annual Reviews of Biophysics and Bioengineering '1; SI. Jost, J., Waggoner, A.S., and Griffith, O.H. (l97l) "Structure and Function of Biological Membranes." Hubbell, H.L. and McConnell, H.M., (l97l) J. Amer. Chem. Soc. 2}, 3i“. Scandelle, C.J., Devaux, P. and McConnell, H.M., (l972) Proc. Nat. Acad. Scl. USA 92, 2056. Cone, R.A,, (I972) Nature M.B. 3}_6_. 39. Blasle, J.K., (I972) Biophysical J. 13, 205. Frye, L.D. and Edidln, M., (I970) J. Cell Sci. 1, 3l9. Shinitzky, M., Dianoux, A.C., Gitler, C. and Haber, G., (l97l) Biochem. 12, 2l06. Cogan, U., Shinitzky, M., Hover, G. and Nishlde, T., (I973) Biochem. 12, 52l. Ruby, B. and Gitler, C., (I972) Biochlm, Biophys. Acte 288, 23l. Sealing, J. and Nlederberg, U., (l97h) Biochem.‘12, l585. Chan, S.I., Salter, C.H.A., and Felgenson, G.U., (I972) Blochlm. Biophys. Res. Comm. 39, INBB. Hsu, M., and Chen, S.I., (i973) Biochem. 13, 3872. Lee, A.G., Birdsell, J.M. and Metcalfe, J.C., (I973) Biochem.‘lg, I650. Chapman, 0. and Selsbury, N.J., (I970) "Recent Progress In Surface Science,” 3, l2l. Parrin, F., (l93h) J. Phys. 5, #97. Parrln, F..(l936) J. Phys..z, I. I8. '9. 20. ll. -22. 23. 2h. 25. 26. 27a 28. 29. 30. 3i. 32. 33. 3h. 35. 36. 37. 38. “5 Einstein, A., (I956) "Investigations on the Theory of the Brownian Movement." Gratzel, M. and Thomas, J.K., (I973) J. Amer. Chem. Soc. 22, 6885. Chen, M., Gratzel, M. and Thomas, J.K., (I974) Chem. Phys. Lott. 35, as. . Chang, S., Thomas, J.M. and Kulpa, C.F., (l97h) Biochem. 13, Il35. Yguerablde, J., Epstein, J.F. and Stryer, L., (l970) J. Mol. Biol. 21. S73. Tao, T., (I969) "Biopolymers" 8, 609. Yguerablde, J., Epstein, J.F. and Stryer, L., (I970) J. Mol. Biol. 51, 573. Yguerabide, J., (I972) ”Methods in Enzymology,“ 16, “98. Weber, 6.. (I973) “Fluorescence Techniques in Cell Biology," P. 5. Rigler, R. and Ehrenberg, M., (I973) Quart. Rev. of Biophys. 6, I39. wahl, P. and TImasheff, S.M., (l969) Biochem. g, 29us. Vanderkool, J., Fischkoff, S., Chance, B. and Cooper, R.A., (I979) Biochem. 12, l58. Cehelnlk, E.D., Cundall, R.B., Lockwood, J.R. and Palmer, T.F., (i97h) J. Chem. Soc. Fer. Trans. 3, 2h“. Galla, H.J. and Sackmann, E., (l97h) Biochlm. Biophys. Acta 332, I03. Pownall, H.J. and Smlth, L.C., (l973) J. Amer. Chem. Soc. ' 22, 3l36. Avourls, P., Kordas, J. and El-Bayouml, M.A., (l97‘i) Chem. Phys. Lett. (In press) El-Bayouml, M.A. and NaIIm. F.M.A., (I968) J. Chem. Phys. fig, 2536. Danny, B. J., (I959) J. Chem. Phys. 29, I59. Von Salis, G.A. and Lebhert, M., (i968) J. Phys. Chem.‘z;, 752. Holland, J.F., Teets, R.E. and TImnlck, A., (l973) Anal. Chem. 35, IHS. Stegemeyer, H., (l968) Ber. Bunsenges, Phys. Chem. 12, 335. Ill||||||||||l||j|||||||||'||IIIIllllllllllllllllllilllIll