A study of the solution photophysics of the isomeric aminobenzoic acids, N,N-dimethylaminobenzoic acids and their methyl esters

L 

A STUDY OF THE SOLUTION PHOTOPHYSICS OF THE ISOMERIC 

AM]INOBENZOIC ACIDS, N,N-DIMETHYLAMINOBENZOIC ACIDS AND 

THEIR METHYL ESTERS 

John Austin Timothy Revill 

being a Thesis submitted in partial fulfilment of the requirements for the degree 

of 

DOCTOR OF PHILOSOPHY 

awarded by the 

UNIVERSITY OF CENTRAL LANCASHIRE 

Sponsoring Establishment: 

Collaborating Establishment: 

University of Central Lancashire 

Department of Applied Physics, 

University of Strathclyde, 

Glasgow 

Department of Chemistry, 

University of Central Lancashire, 

Preston. 

February 1993

Acknowledgements 

I would like to thank Professor J.D.Hepworth and especially Professor 

KG.Brown, my supervisors, for their always available help and guidance during 

my years at the Polytechnic and later the University. I would also like to thank 

Dr. D.Birch of Strathclyde University and Prof. W.Rettig of the Technical 

University of Berlin for invaluable discussions and advice. Also I would like to 

thank The Science and Engineering Research Council, without whose financial 

support this Ph.D. would have been impossible. 

I would like to say a special thankyou to my parents for moral support 

and helping to make life a little more pleasant.

Chapter 1 

Chapter 2 

Introduction 

1.1 The nature of light 1 

1.2 Molecular orbital theory 2 

1.3 Absorption of light 3 

1.4 Transition moments 6 

1.4.1 Vibrational overlap integral 8 

1.4.2 Spin overlap integral 12 

1.4.3 Transition moment 14 

1.5 Extinction coefficients 15 

1.6 Dissipative pathways 19 

1.6.1 Vibrational relaxation 20 

1.6.2 Radiative transitions 20 

1.6.3 Non-radiative transitions 22 

1.6.4 Chemical reactions 23 

1.7 Quenching 23 

1.7.1 Photophysical quenching 23 

1.8 Kinetics of fluorescence 27 

1.8.1 Stern-Volmer equation 29 

1.9 Acid Base properties 30 

1.10 Literature background to work 31 

1.10.1 4-Dimethylaminobenzonitrile 35 

1.10.2 4-Dimethylamino and 4-Diethylamino 41 

benzoic acid and theft methyl or ethyl ester 

1.10.3 Aims of project 42 

Experimental 

2.1 Materials 43 

2.1.1 Solvents 43 

2.1.2 Chemicals 43 

2.2 Instrumentation (techniques) 44 

2.2.1 Absorption spectra 44 

2.2.2 Ground and excited state pK a and pKb values 46 

2.2.3 Fluorescence spectra 49

2.2.4 Fluorescence decay profiles 51 

Chapter 3 

Solution Photophysics of 3-aminobenzoic acid and 

3-N,N-dimethylaminobenzoic acid and their methyl 

esters 

3.1 Ground state absorption spectra as a function of 55 

solvent 

3.2 Ground state absorption and fluorescence emission 60 

data as a function of pH 

3.3 Fluorescence emission properties as a function of 66 

solvent 

3.4 Fluorescence lifetime data as a function of solvent and 69 

pH 


temperature 

3.6 Summary of conclusions 76 

Chapter 4 

Solution Photophysics of 2-aminbbenzoic acid and 




solvent 


properties as a function of pH 


solvent 


pH 


temperature 


E

Chapter 5 

Solution Photophysics of 4-aminobenzoic acid and 




solvent 


properties as a function of pH 


solvent 

5.4 Effect of excitation wavelength on the fluorescence 109 

properties of 4DMABA and M4DMAB 

5.5 Variation of fluorescence intensity of 4DMABA and 111 

M4DMAB as a function of concentration 


pH 

5.7 Fluorescence properties of 4DMABA and M4DMAB 124 

in mixed solvents 


temperature 


Chapter 6 References 151 

Appendix A 159 

Appendix B 167

Abbreviations 

Solvents 

BCL I ip Chlorobutane / isopentane (4:1) 

BCN / iBCN Butyronitrile I isobutyronitrile (9:1) 

MCH I ip Methylcyclohexane / isopentane (4:1) 

Compounds 

DMABN 4-Dimethylaminobenzonitrile 

XABA 

(x=2, 3 or 4)-Aminobenzoic acid 

MxAB 

Methyl (x=2, 3 or 4)-aminobenzoate 

xDMABA (x=2, 3 or 4)-N,N-dimethylaminobenzoic acid 

MXDMAB Methyl (x=2, 3 or 4)-N,N-dimethylaminobenzoate 

E4DEAB Ethyl 4-diethylaminobenzoate 

Miscellaneous 

TICT 

Twisted intra-molecular charge transfer

ABSTRACT 

The solution photophysics of the isomeric aminobenzoic acids and N,Ndimethylaminobenzoic 

acids and their methyl esters were studied under a 

variety of conditions of solvent, pH, concentration and temperature. 

The most straightforward behaviour was obtained when the amino or 

N,N-dimethylamino substituent was in the meta- position relative to the 

carboxylic add or ester function. The amino compounds absorb with a peak 

between 300 and 325 nm depending on solvent, their fluorescence peaks lie 

between 350 and 410 rim. The corresponding N,N-dimethylamino compounds 

usually absorb and fluoresce some 20 - 30 rim to the red of these wavelengths. 

Both protonation and deprotonation of the ground state molecules takes place 

under weakly acidic conditions with pKi, values of approximately 3.0 and pK a 

values of approximately 5.0 being obtained. The observed variation in the 

fluorescence of the meta isomers with pH is largely caused by ground state 

equilibria. Förster cycle calculations suggest that the excited state pK a is little 

changed from the ground state whereas pKC is much smaller. This is in good 

agreement with the observed fluorescence data and the nature of the excited 

state. Measurements of fluorescence decay profiles reveal a longer lived excited 

singlet state (usually some 10 -20 ns) than is seen for the other isomers. The 

fluorescence intensity of the meta- isomers was observed to rise with decreasing 

temperature, although this could possibly be attributed to absorbance changes. 

The photophysics of the corresponding ortho compounds is complicated 

by steric and hydrogen-bonding interactions especially in the compounds 

containing an N,N-dimethylamino group. The increased interaction between 

the two substituent groups leads to an increase in extinction coefficient and 

fluorescence quantum yield and a decrease in fluorescence lifetime for 2- 

aminobenzoic acid and its methyl ester compared to the meta derivatives. 

Spectral band positions, pK values and the effect of temperature on the 

fluorescence are similar in the two sets of compounds. The 2-N,Ndimethylamino 

compounds do not parallel the corresponding meta- derivatives 

due to steric hindrance and intra-molecular hydrogen bonding. These cause 

weak, blue shifted absorptions and low fluorescence quantum yields. There 

was no evidence found in this work for the presence of a zwitterionic species in 

the absorption properties of 2-N,N-dimethylaminobenzoic acid. 

The photophysical properties of 4-aminobenzoic acid and its methyl 

ester are in many ways very similar to those of the corresponding orthosubstituted 

compounds. However the same certainly cannot be said for 4-N,Ndimethylaminobenzoic 

add and its methyl ester. These two compounds exhibit 

anomalous fluorescence in the form of two fluorescence bands in some 

solvents. The appearance of this anomalous fluorescence is both concentration 

and solvent composition dependent. Various mechanisms to explain these 

observations were tested; excimer, exciplex, TICT state and ground state dimer 

formation. The latter was found to be in best accord with the properties of these 

molecules, but it was clear that in binary solvent mixtures such as hexane / 

acetonitrile and acetonitrile / water other mechanisms could also be involved.

Chapter 1

Chapter 1 Introduction 

1.1 The nature of light 

Photochemistry is concerned with the chemical and physical effects of 

electronic excitation produced by the interaction of electromagnetic radiation 

with matter. There are two physical models which may be used to describe the 

properties of electromagnetic radiation (emr). The wave model takes into 

account the electric and magnetic components of the radiation and envisages 

electromagnetic radiation in terms of oscillating electric and magnetic fields 

in planes which are perpendicular to each other and to the direction of 

propagation 1 (figure 1.1). The z co-ordinate is in the direction of propagation, 

so that the electric vector is contained in the zy plane and the magnetic vector 

in the xz plane. 

I 

Figure 1.1 The instantaneous electric and magnetic field strength vectors as 

a function of position along axis of propagation of emr 

The quantum model considers light to have a particle nature (Planck 

1901) and to exist as discrete quanta called photons, whose energy E is given by 

E=hv= 1.1 

x 

where h = Plancks constant, c is the speed of light, V is the frequency of the 

radiation and Xis its wavelength. Table 1.1 gives details of selected values of 

these parameters. 

1

Radiation Wavelength Frequency Energy 

/ rim v/s-1x10 15 E / kJ mo!-' 

Far UV 100 3.00 1200 

Near UV 250 1.20 480 

U V 350 0.86 343 

Blue-green 500 0.60 240 

Red 700 0.41 171 

Infrared 1000 0.30 120 

Table 1.1 Colour, frequency, wavelength and energy of light 

1.2 Molecular orbital theory 

To understand the interaction of matter with electromagnetic 

radiation, a knowledge of molecular orbital theory is required. Molecular 

orbitals are formulated as linear combinations of the valence shell atomic 

orbitals. It is assumed that the inner electrons remain in their original atomic 

orbitals. For example the interaction of two identical atomic orbitals 23a and 

Ob give rise to two molecular orbitals of the form: 

411= 0a+ 13b 1.2 

2=t a -lb 1.3 

One molecular orbital ljf I is bonding (more stable than the original 

atomic orbitals) and the other W2 is antibonding (less stable) 2. Those which 

are completely symmetrical about the internuclear axis are designated sigma 

orbitals and arise when the interacting atomic orbitals 13a and ôb are s 

orbitals or p orbitals. Molecular orbitals derived from the interaction of two 

parallel p orbitals are called a orbitals and again bonding and antibonding 

molecular orbitals are formed. If, prior to formation of the molecular orbital, 

the atomic orbitals are each occupied by a single electron or if one is occupied 

by two electrons and the other is vacant then the electrons in the molecular 

system both occupy the lower energy bonding molecular orbital. This leads to 

a gain in stability over the isolated atoms and is the basis of the molecular 

orbital description of electron pair covalent bonding. Certain compounds

contain non-bonding valence shell electrons (n). These electrons are not 

involved in bonding relationships and can be regarded as being localised on 

their atomic nuclei. In terms of their energy, the molecular orbitals generally 

take the following order (figure 1.2). 

Antibonding * 

Antibonding .Jt * 

Energy 

Non-bonding 

Bonding 

It 

- Si 

Figure 1.2 Relative energies of molecular orbitals 

1.3 Absorption of light 

In a normal molecule, two electrons are assigned to each molecular 

orbital where the electrons fill the orbitals from the lowest energy orbital 

upwards in such a manner that their spins are paired, in order to have the 

most stable energy configuration. In order for electronic excitation to occur 

the energy of the absorbed photon (E) must equal the energy difference 

between an occupied molecular orbital and an unoccupied molecular orbital. 

This electronic excitation of a molecule results in the promotion of an 

electron from one molecular orbital to another of higher energy e.g. a-as, 

n_a* , it-it * or n-i? and retains the same electron spin. Figure 1.3 gives an 

example of a n-71* transition for methanal (formaldehyde). 

3

7t* 

HQ 

Op 

H'6I 

Ill 

hu - 

II 

n(p) 

n(sp) 

Figure 1.3 Electronic excitation of methanal 

11 

II 

In general terms the acY* transition requires the greatest energy and 

the wavelength of absorption associated with the transition will generally be 

in the far UV. Both the cy-a 

and the nG* transitions will be of little 

significance when discussing the photochemistry of compounds containing Jr 

electrons since these transitions involving it electrons are considerably more 

intense and of lower energy. In ethene, the lowest energy transition is the 

singlet ic-ic * which gives rise to an intense absorption in the ultraviolet 3. The 

triplet ic-ice transition (spin inversion of promoted electron) is so weak that a 

path length of 14m of liquid ethene would be required for its detection 4 . 

Conjugation shifts it-ice transitions to longer wavelengths. This is because the 

energy gap between the highest occupied molecular orbital (HOMO) and the 

lowest unoccupied molecular orbital (LUMO) becomes progressively smaller 

as the number of conjugated double bonds increases. Consequently the 

wavelength of maximum absorption increases. An example of this is found 

in butadiene which involves the interaction of two C=C double bonds, 

4

leading to a red shift in the absorption maximum relative to that found for 

ethene. 

The n_it* transition is characteristic of molecules possessing 

chromophores with multiple bonded hetero atoms. It is of low intensity 

because it is symmetry and/or overlap forbidden and is normally the band 

occurring at longest wavelength. Acetone exhibits an example of a n ir* 

transition. In the ground state the non-bonding and it-bonding orbitals are 

doubly occupied, leaving the ir orbital empty, figure 1.4. In the lowest excited 

singlet state the non-bonding and ic orbitals are both singly occupied, leaving 

their-bonding orbital doubly occupied. This nir* transition has a wavelength 

of maximum absorption of 280 nm (E=18 dm- 3 mol-1 cm -1). However in the 

region of 190 nm there is an intense band which may be due to an nG* or G- 

transition5. 

Plan 

Elevation 

Me\ Me.e Me.Q 

/CO 

Me Me / 0 Me43 

n-orbital n-orbital n *_orbita l 

Figure 1.4 Molecular orbitals of acetone 

Mixtures of electron donors and electron acceptors in solution often 

exhibit a new absorption band which is shown by neither component 

separately. It is attributed to the presence in such mixtures of a donor-acceptor 

complex 6. Typical acceptors are picric acid and other polynitro aromatic 

compounds; typical donors include aromatic hydrocarbons. The transition is 

referred to as a charge-transfer transition and is generally broad and 

structureless. The spectroscopic transition corresponds approximately to the 

light induced transfer of an electron from the donor to the acceptor. An 

5

example of such a complex is between the acceptor 4-benzoquinone and 

donor 2,3-dimethylbutadiene. In methylcyclohexane solution the mixture 

produces two bands at 450 and 300 nm, but in pure 2,3-dimethylbutadiene 

only a single peak is observed at 350 nm 7. A single molecule can include both 

donor, and acceptor groups such that a charge transfer (CT) transition can be 

observed in solutions of pure compounds. 

1.4 The transition moment and probability of transitions 

The probability of an optical transition between two states is given by 

the value of the transition moment (TM) integral 8 

TM =J tI'.4L'P 1 d'r 1.4 

where 'F. is the total wavefunction of the initial state, T f 

is the total 

wavefunction of the final state and j.t is the dipole moment operator. The 

transition moment in principle can be calculated from quantum mechanics 

but it can also be obtained from absorption spectra through its relation to the 

oscillator strength f (see later). Although this integral cannot be precisely 

evaluated because the exact forms of the wave functions are not known, 

approximations can be made. 

Nuclear motion is considered to be very slow in comparison to that of 

electrons because of the large mass differences. The motion of the electrons 

can therefore be considered as being independent of that of the nucleus. This 

approximation is known as the Born-Oppenheimer approximation, which 

bears up extremely well until degenerate orbitals are found where their 

potential energy surfaces cross. Thus the total wavefunction can be factorised 

into a nuclear (vibrational) wavefunction 0 and an electronic wavefunction 

w, ('PrOw,). The transition moment is then given by

TM = I OI.WLJLOLWf di . 1.5 

Since i' operates only on the electrons, equation 1.5 can be written as 

TM =5 Oj.O f dt I W.J.LW1 d; 1.6 

where n and e refer to nuclei and electrons respectively. Further 

approximations have to be made. It is assumed 

i) that w is the product of one electron wavefunctions (orbitals which are 

themselves linear combinations of atomic orbitals) and that the same orbitals 

may be used to describe both ground and excited state and 

ii) 

that only one electron is promoted during a transition. The transition 

moment can then be written 

TM = JOi O f d'Tn J4aI.L4idte 1.7 

where O i and of are the wavefunctions of the initial and final orbitals of the 

electron. 

The final approximation is that the orbitals can be factorised into a 

product of space and spin orbitals (Q and S respectively) 

1.8 

TM =fO i.Of dCn JSS1dt JQiIiQfdt e 1.9 

It should be noted that R only operates on the space co-ordinate. 

The first term in the transition moment expression (equation 1.9) is the 

overlap integral of the wavefunctions for nuclear vibrations. The second 

7

term is a spin overlap integral and its value depends on the initial and final 

spin states of the promoted electron. The third term is called the electronic 

transition moment and its value depends on the symmetries and the amount 

of spatial overlap of the initial and final orbitals. 

This application of quantum mechanical theory to electronic processes 

has led to a set of selection rules 9 which allow transitions to be classified as 

allowed or forbidden. The probability of occurrence of an electronic transition 

and hence the intensity of the absorption band is dependent upon various 

factors. These factors are included in the selection rules which govern 

whether or not a transition will be allowed or forbidden. Transitions which 

obey the selection rules can give rise to very intense absorption bands. 

Conversely, transitions which do not conform to the selection rules either do 

not occur or else the probability of occurrence is so low that only very weak 

bands are observed in the spectrum. 

The transition moment in equation 1.9 is a product of three separate 

integrals and its value will be zero if any one of the integrals is zero. When 

this happens the transition has a zero probability of occurrence and is said to 

be forbidden. When the transition moment is non-zero, the transition is said 

to be allowed. 

1.4.1 The vibrational overlap integral 10 JO1 O f dTn may be evaluated from a 

knowledge of the vibrational wavefunctions of the two states involved in the 

transition (obtained by solving the appropriate Schrodinger equation) and the 

application of the Franck-Condon principle 11 . The principle states that an 

electronic transition occurs so rapidly in comparison with vibrational 

frequencies that no change in internuclear separation occurs during the 

course of the transition. This implies that the transition may be represented 

by a vertical line connecting the two potential energy surfaces. The most 

probable transition from this vibrational level at the same internuclear 

F1

distance will be the turning point of the oscillation. In order to evaluate the 

f0 10 1d; integral, it is necessary to know the vibrational wavefunctions, which 

for simplicity will be treated as simple harmonic oscillators. 

-- awawaw 

R WV 

El an. 

111 way 

In 

I- - 

e 

a) 

Internuclear seporQtion Cr) 

Figure 1.5 Ground state vibrational wavefunctions and energy levels for a 

harmonic oscillator 

For such systems, the potential energy varies with internuclear separation as 

described by the parabolic curve of figure 1.5. The equilibrium separation re is 

given by the point a, where the potential energy is a minimum. Solution of 

the appropriate Schrodinger equation leads to the result that the vibrational 

energy (Ev) is quantized with values 

Ev=hU(V+ >) 1.10 

where 1) is the fundamental vibrational frequency and V the vibrational 

quantum number, which is an integer 0,1,2,.. .n. These values are represented 

by the equally spaced horizontal lines in the figure. The oscillator curves 

represent the wavefunctions associated with each level. Since the square of 

the amplitude (0) of the function gives the probability of a particular nuclear 

configuration, we can see from the bottom vibrational level (V"=O) that the

nuclei are most likely to be found at the equilibrium point re and as V" 

increases it becomes more probable that the molecule will come close to the 

configuration corresponding to the intersection of the horizontal lines and 

parabolic curve, (the turning point of the vibration, where the total energy 

equals the potential energy, the kinetic energy is zero, and the nuclei are 

static). If upon excitation there is no change in geometry, then figure 1.6A is a 

schematic representation of the most likely transition. From the figure it can 

be seen that the most probable transition is V"=O to V'=O i.e. the O—*O 

transition. 

Excited 

state 

0' 

a; 

C 

Li 

:1 

C) 

=0 

Le 

Ground 

state 

Internuclear 

Internuclear separation (r separation (-) 

Figure 1.6 Ground state and excited state potential energy curves. (A) with 

no change in geometry upon excitation and (B) with change in 

geometry upon excitation, the most common scenario 

Figure 1.6B represents the more usual case where stretching of bonds 

upon excitation leads to a change in geometry (from exciting the electron into 

an antibonding orbital thus weakening the bond). In figure 1.68, the V"=O to 

V'=2 transition will be the most likely vibrational transition. Very similar 

considerations apply to emission spectra, with the proviso that the excited 

state will normally become thermally equilibrated before emission, which 

i-El]

therefore takes place almost exdusively from the bottom vibrational level of 

the excited state (V'=O); see figure 1.7. 

a. 

I- 

a) 

C 

Ui 

p 

Figure 1.7 Representation of the most likely emission and absorption 

transitions 

The Franck Condon principle states that electronic transitions occur in 

an exceedingly brief interval of time so that no change in nuclear position or 

nuclear kinetic energy occurs during the transition. This implies that the 

transition may be represented by a vertical line connecting the two potential 

energy surfaces, and the most probable transition will be to that vibrational 

level with the same internuclear distance at the turning point of the 

oscillation, eg lines AY and ZB in Figure 1.7. A transition represented by line 

AX is extremely improbable because the molecule, in arriving at point X, 

would have suddenly acquired an excess kinetic energy given by XI. 

There are however various assumptions inherent in this treatment, which 

are questionable given that 

11

i) real molecules are anharmonic oscillators. For a diatomic molecule the 

potential energy diagram is approximated by a Morse curve 12, the higher 

vibrational levels of which become progressively closer together until they 

merge into a continuum at the dissociation limit. 

ii) For any molecule containing more than two atoms the potential 

energy curve becomes a surface in many dimensions, due to the large number 

of possible vibrational modes. 

iii) For a molecule which undergoes a change in geometry upon excitation, 

the vibrational and rotational transitions are so closely spaced that they 

overlap giving rise to a smooth broad absorption. Hence in the majority of 

cases little vibrational structure may be seen in visible or ultraviolet 

absorption bands. 

1.4.2 The spin overlap integral 

The effect of electron spin upon transition intensities is given by the 

spin overlap integral JSS1d The electrons involved in the transition can only 

have two spin states (designated a and 3) and there are three common 

situations 

12

(a) Singlet (--* Singlet Transitions 

In singlet e-* singlet transitions no change occurs in the spin state of the 

promoted electron and the spin overlap integral is unity 

Jaad=JJ3j3 d5= 1 

Lii 

because the spin wave functions are normalised. There are no spin 

restrictions on such transitions which are therefore fully allowed. 

(b) Triplet H Triplet Transitions 

Since the transition occurs with no change in multiplicity, the spin 

overlap integral is again unity and the transition is fully allowed. 

(c) Singlet *-* Triplet Transition 

Since the promoted electron changes its spin state the spin overlap 

integral is 

1j3j3d5=Jctccd5=O 1.12 

because the cx and ji spin wavefunctions are orthogonal. The transition 

moment is thus zero and the transition is strongly forbidden. 

Although singlet 4-> triplet transitions are strongly forbidden, they do 

occur through spin-orbit coupling. Since the electron is charged and spinning, 

it is expected to have not only spin angular momentum but also a magnetic 

moment. In a singlet

that although total angular momentum is conserved, neither orbital nor spin 

angular momentum are individually conserved. This means that the concept 

of a pure spin state is no longer tenable but that intermediate situations must 

be contemplated in which a normal singlet state has a certain degree of triplet 

character and vice versa. In conclusion, although singlet e-* triplet transitions 

are strictly forbidden, weak bands may be observed 4 . 

1.4.3 The electronic transition moment 

The electronic transition moment is intimately related to the 

symmetries of the orbitals. Transitions may be space and/or overlap 

forbidden. For example, n1c* transitions involve two orbitals located in 

different regions of space (i.e. no overlap occurs) and therefore these 

transitions are overlap forbidden. However, the two orbitals involved in the 

ic-ir transition have large amplitudes of their wavefunctions in the same 

region of space and are therefore not overlap forbidden. 

Symmetry forbidden transitions should have zero intensity and be 

unobservable, although they are found to have small but finite intensities. 

This feature arises because the vibrational (nuclear) and electronic motions 

are not completely independent of each other but are weakly coupled. The 

coupling, called vibronic coupling, is responsible for the non zero intensity of 

symmetry forbidden transitions such as the benzene absorption at 254 nm. 

This transition is forbidden if the molecule is in the form of a regular 

hexagon but there are vibrations which distort the hexagon and reduce its 

symmetry, so that the transition becomes weakly allowed. A further type of 

transition which involves symmetry is the parity forbidden transition. When 

the wave function of a molecule changes sign on inversion through a centre 

of symmetry it is called ungerade, while those not changing sign on reflection 

are called gerade. Selection rules state that ungerade -* gerade transitions will 

14

e allowed but gerade t4 gerade and ungerade #4 ungerade transitions are 

forbidden. 

The selection rules can be summarised as follows. If a completely 

allowed transition has an oscillator strength fa of the order of unity, then 

other transitions have oscillator strengths f given by 

f=fo *fs *fsym *fa 1.13 

where f0 is the overlap factor and fs is the electron spin factor and fsym is the 

symmetry or parity factor 14. 

1.5 Extinction coefficients 

Experimental measurements of absorption transitions usually use 

extinction coefficients (molar absorptivities) to quantify the transition 

intensity. Under normal circumstances absorption properties can be 

represented by the Beer-Lambert Law: 

I 

1.14 

= Ecl = Absorbance 

(L. ) 

where Jo is the intensity of the incident monochromatic radiation, I t is the 

intensity of the transmitted radiation, c is the concentration of the sample, 1 is 

the path length and S is the molar extinction coefficient. The molar extinction 

coefficient, or more correctly the decadic molar extinction coefficient, is 

characteristic of a particular compound at a particular wavelength of 

radiation. This empirical law is valid except when very high intensities of 

radiation are employed and a significant proportion of the molecules in a 

given region are in the excited state rather than the ground state at an one 

time. 

15

An alternative measure of absorption intensity, which can be related 

more readily to theoretical principles is the oscillator strength f given by the 

equation 1.15 

= 12303mc 2 

,tNe2 ]Fjcdv = 4.315x1crFJ cdv ..........1.15 

[ 

where the integration extends over the entire band related to the transition 

from state S-'S; N is Avogadros's number, and F constitutes a correction 

factor related to the refractive index of the medium in which the absorbing 

molecule is dissolved. In most cases this value is taken as 1. The major 

difference between the oscillator strength and the extinction coefficient is that 

the former is a measure of the integrated intensity of absorption over a whole 

band, whereas £ is a measure of the intensity of the absorption at a single 

wavelength. 

The lowest energy absorption transition in benzene, a 'Mg 3 1 132 u 

transition ( 1 A -* 'Lb in Platt notation), is a symmetry forbidden g -3 u 

transition which becomes partially allowed through vibronic coupling 15. The 

wavelength of maximum absorption lies around 255 nm with an extinction 

coefficient of 220 dm3 mol-1 cm-1 . The S0 -* S2 absorption is also forbidden, 

being a 1 Ai g -* 1 13 1 transition ('A -3 'L a), but exhibits a much higher 

extinction coefficient (see Table 1.2). The first allowed transition, 1 A1g -+ 'Ei 

('A -4 'B), is much more intense that either of the first two transitions but 

lies in the vacuum ultraviolet. 

On substitution, the high symmetry of the benzene molecule (D6h) is 

removed and the So - transitions become more allowed. The addition of 

both electron donating and electron withdrawing groups tends to cause a 

bathochromic shift in the lowest energy absorption band 16. On adding a 

carboxylic acid group, a new absorption band is observed around 280 nm (e 

16

1,000 dm3 mo!-1 cm-I)16 and there is an increase in the absorption intensity at 

230 run, giving a peak with an extinction coefficient of 11,600 dm 3 cm-1 moF' 

16 On the addition of an amino group to the benzene ring, to form ani!ine, 

the new absorption peak is also around 280 nm 16, but the extinction 

coefficient is approximately twice that of benzoic acid (E 1,850 dm3 cm-1mo! -1 ); 

Tab!e 1.2. When a N,N-dimethylamino group is substituted into the ring the 

absorption spectrum is further red shifted to 300 nm and the extinction 

coefficient becomes even larger (8 2,400 mo! -1 dm3 cm-1). One point of interest 

for aniline is that in cyclohexane the absorption spectrum is structured whilst 

in ethanol a c!ean structureless absorption band is seen 16 

Molecule / Species Solvent Transition ?max 

chromophore 

(nm) 

Benzene neutra! none 256 220 

200 6,300 

180 100,000 

Benzoic acid neutral Inone 280 1,000 

230 11,600 

Ani!ine neutra! water —.g* 280 1,850 

230 8,600 

cation acid (aq) t—nr* 254 160 

203 7,500 

8 

dm 3 moicnr1 

Table 1.2 

Examples of transitions for various organic molecules and their 

extinction coefficients 

Wepster 17 showed that in the case of both aniline and N,Ndimethy!ani!ine 

that the !one pair of e!ectrons were conjugated with the 

benzene it-system, but that the H or Me groups were "tilled', at an angle of 

approximately 30 0 re!ative to the plane of the ring, Figure 1.8. 

17

O ;R 

Figure 1.8 

Diagram representing the "tilt" angle between substituents on 

the aniline nitrogen atom and the plane of the n-system 

The presence of two substituents in the benzene ring raises the twin 

possibilities of electronic and steric interactions between the two substituents. 

Further work by Wepster 18 in the late 50's dealt with the steric effects 

observed in di- and higher- substituted benzene derivatives. Although much 

of his work was involved with orthosubstituted derivatives he also worked 

on meta- and para- disubstituted benzenes. The potential steric effect of 

substituents in the 2- or 2,6- positions of molecules such as aniline is to cause 

the -NR2 group (Figure 1.8) to rotate about its bond to the ring and thereby 

twist the nitrogen lone pair out of conjugation with the n-system of the ring. 

Wepster 19 found that even for aniline, with bulky t-butyl groups attached to 

the 2,6- positions, the NH2 group was only twisted slightly out of conjugation 

with the ring n-system (approximately 30 0 as measured by a 50% decrease in 

the extinction coefficient of the 1 A - 1Lb transition). By contrast, the 

extinction coefficient of the corresponding N,N-dimethyl compound was 

reduced to approximately 1/20th of that of the N,N-dimethylaniline 

indicating, on the basis of equation 1.15a (where c, is the extinction coefficient 

of the unsubstituted aniline and 8 the extinction coefficient of the ring 

substituted compound), a twist angle (p) of> 750 

2/60 = cos 2 ( l.15a 

The presence of an electron withdrawing group in the 4-position 

relative to the amine helps to reduce the twist angle caused by the presence of 

a particular 2- or 2,6- substituent 20, so that p for amine > (p for 4-ester > (p for 

iLJ

4-NO2. The absorption properties of the aminobenzoic acids are discussed in 

more detail in section 1.10. 

1.6 Dissipative pathways 

Having excited an electron into an orbital of higher energy, producing 

an excited state, the molecule has a number of possibilities open to it to 

dissipate the extra energy. These include vibrational relaxation, radiative 

transitions (fluorescence, delayed fluorescence and phosphorescence), 

radiationless transitions (internal conversion and intersystem crossing), 

energy transfer and chemical reaction. 

These processes are not mutually exclusive and they compete with 

each other for the deactivation of an excited state. Vibrational relaxation often 

precedes the others. They are portrayed in the Jablonski diagram shown in 

Figure 1.8. The radiative transitions are portrayed by straight lines and the 

non-radiative transitions by jagged lines. S n and Tn refer to the nth excited 

singlet and triplet states, where n is an integer, except for % which is the 

ground state. 

S3 

H 

Internal 

conversion 

Internal 

conversion 

SI 

Fluorescenc' 

T 1 

Absorption 

Ii 

coi 

S o 

Figure 1.8 A simplified Jablouski diagram. 

19

1.6.1 Vibrational Relaxation 

The energy of the absorbed photon will not often match exactly the 

energy difference between the lowest vibrational level of the ground and 

excited states of the absorbing molecule. Therefore the state initially produced 

is an upper vibrational / rotational state of the electronically excited 

molecule. The molecule loses excess vibrational energy through vibrational 

relaxation until the zeroth vibrational level of the excited state is reached. 

The loss of this excess vibrational (and/or rotational) energy is largely 

dependent on collisions, as a result of which the vibrational energy is 

converted into kinetic energy distributed between the partners in the 

collision. 

1.6.2 Radiative transitions 

In the radiative transitions represented by straight lines on the 

Jablonski diagram, an excited species passes from a higher energy state to one 

of lower energy with emission of a photon. 

a) Fluorescence is caused by a radiative transition between states of the 

same multiplicity, although for polyatomic molecules the transition is 

usually S1—+S0. S2—*S0 fluorescence is occasionally observed but it is usually 

hidden by the far more intense Si—*So transition. Azulene fluoresces from 52 

and only negligibly from S 1 21 . In many cases the absorption and fluorescence 

spectra are mirror images when plotted on a frequency scale. The existence of 

this mirror image relationship implies that only minor changes occur in the 

molecular structure of the compound during excitation. However the solvent 

and substituent groups affect the position and structure of the spectrum. 

20

) Phosphorescence is the emission of radiation accompanying the 

deactivation of an excited species to a lower state of different multiplicity e.g. 

T1—*S0. Rate constants are always smaller than those of the corresponding 

fluorescence transitions because the transition is spin forbidden. 

c) Delayed fluorescence, in which the luminescence decays more slowly 

than expected, can arise by several mechanisms of which triplet-triplet 

annihilation22 and thermal activation 23 have been studied most closely. 

A measure of the efficiency of a photochemical process such as 

fluorescence is of fundamental value. The concept of a quantum yield (4)) 

where 

- Number of molecules undergoing a process 

- Number of photons absorbed 

1.16 

can give valuable information on the nature of the process. Information can 

be gathered from the magnitude of this quantity and the effects that 

temperature, pressure, concentration, substituent groups and solvents have 

on it. The quantum yield for formation of a product may not be the same as 

the quantum yield for the decomposition of starting material, depending on 

the complexity of the reaction. 

The fluorescence intensity F per unit volume is proportional to the 

fractional light absorption Taper unit volume per unit time 

F41 1a 1.17 

where Ia=I0(1-e 23 ') 

1.18 

21

For dilute solutions the term inside the brackets can be expanded and if 

identical experimental conditions are maintained then fluorescence 

intensities and concentrations are linearly related according to the expression 

1.19 

F=4f 2.30310 Ed 1.19 

On the other hand, at high concentrations where all the incident light 

is completely absorbed, the exponential term in equation 1.18 becomes 

negligible and consequently: 

F=4 1 10 1.20 

The fluorescence intensity should remain constant for any further increase in 

concentration, but for many compounds the intensity starts decreasing . This 

decrease in fluorescence is due to the quenching of fluorescence by molecules 

of the same kind (concentration quenching) 14 . 

1.6.3 Non-radiative transitions 

Non-radiative transitions occur between degenerate vibrationalrotational 

levels of different electronic states. There is no change in the 

energy of the system and consequently no photon is emitted. 

a) Internal conversion is a non-radiative transition between degenerate 

states of the same multiplicity. These transitions are often extremely fast. 

b) Intersystem crossing is a non-radiative transition between two states 

of different multiplicity which are not necessarily degenerate. The 

intersystem crossing S1—,T1, which competes with fluorescence, is the process 

by which triplet states are normally populated. One mechanism of delayed 

22

fluorescence is when the electron undergoes a T1—*S1 transition with the help 

of thermal activation. 

1.6.4 Chemical reaction. 

The chemistry of an excited species usually differs from that of the 

ground state species as a result of the excess energy carried by the excited 

species and also as a result of the different electronic arrangement of the 

excited state. Chemical reactions can occur to yield a product or products of 

lower energy. 

1.7 Quenching 

A substance which accelerates the decay of an electronically excited state 

is described as a quencher and is said to quench the state. Some possible 

quenching routes are shown in figure 1.10. 

M 

. 

Photochemical 

quenching 

Photophysical 

quenching 

Products 

Figure 1.9 Routes for possible quenching mechanisms 

M 

Self quenching 

Q Impurity ____ 

guenching 

Electron 

transfer 

quenching 

quenching 

Energy 

transfer 

1.7.1 Photophysical quenching 

Unlike photochemical quenching, photophysical quenching does not 

lead to new ground state products. A number of different cases can be 

distinguished. 

i) Excimer or Exciplex formation. (Excited Dimer), (Excited complex) 

23

On increasing the concentration of a solute, the fluorescence intensity is 

decreased. This is caused by concentration quenching which also usually 

results in a new emission at a longer wavelength and can be explained in 

terms of excimer or exciplex formation. Such complexes exist only in the 

excited state, being fully dissociated, and therefore undetectable, in the ground 

state. Excimers are formed between identical molecules one of which is in the 

excited state and one in the ground state. Excimers are non-polar, while 

exciplexes, which are formed between an excited species and one or more 

different ground state molecules, are polar entities. They have large dipole 

moments as a result of some degree of electron transfer between the two 

components of the exciplex. In the case of organic molecules, excimer 

formation was first identified by Forster and Kasper 24 for pyrene in solution. 

Excimer and exciplex formation are often rapid diffusion controlled reactions 

and as a consequence of being diffusion controlled are concentration 

dependent. Care must be taken not to confuse these two species with excited 

dimers or excited complexes which are formed in the ground state and then 

absorb a photon of light. 

ii) Iwisted Intra-molecular Charge Iransfer (TJCT) states. 

Stable excited states can occur in molecules in which two 

chromophores, one a donor and one an acceptor, are separated by one 

(twisted) single bond. These have been used to explain anomalous 

fluorescence observed in various compounds. The basic theory was 

developed by Lippert 25 and later refined by Grabowski and Rettig and coworkers26' 

27 In compounds with a near linear planar ground state 

conformation, the Franck-Condon excited state already possesses considerable 

charge transfer characteristics because of the mesomeric interaction of donor 

and acceptor groups. Rettig 28 called this state the delocalised excited state (DE). 

In the perpendicular conformation, this mesomeric interaction can no longer 

24

take place and the available excited states are either different TICT states with 

electron transfer from the various donor orbitals to the various 1t*orbitals of 

the acceptor, or locally excited (LE) states within the donor and acceptor subunits. 

According to Rettig27, the TICT state can be populated spontaneously if 

the transition DE -, TICT is exothermic (or only slightly endothermic) or the 

activation barrier is low enough such that this transition can occur within 

the lifetime of the excited state. Freezing the solution increases the effective 

activation barrier such that in many cases TICT state formation is stopped. In 

the case of DMABN (the first molecule to be found to exhibit TICT behaviour) 

Rettig 27 assumed that the molecule was largely rigid, with only the 

dimethylamino group twisting out of plane, and that the local viscosity in the 

solvent shell surrounding the solute is not drastically altered by dipolar 

interactions. Rettig drew his conclusions for this from experimental 

evidence 29 where it was found that polar molecules exhibit the same 

reorientation relaxation time in polar or non-polar solvents. Work is 

currently underway to try and quantify the effects of excitation on molecular 

dipole moments30 . 

iii) Heavy atom quenching 

Molecular fluorescence is quenched by the presence of species 

containing heavy atoms. The phenomenon often arises through the 

formation of a singlet exciplex which, because of the heavy atom effect, 

undergoes erthanced intersystem crossing to the triplet exciplex, which then 

dissociates into its component parts. Thus solutions of anthracene and some 

of its derivatives become less fluorescent on addition of bromobenzene, while 

the triplet-triplet absorption intensity increases as a result of enhanced singlet 

to triplet inter-system crossing 31 . 

25

iv) Electronic energy transfer 

In this process, an excited donor molecule D* collapses to its ground 

state with the simultaneous transfer of its electronic excitation energy to an 

acceptor molecule A which is promoted to an excited state. 

D * +A—*A * +D 1.21 

The emission of D* is therefore quenched and replaced by the 

characteristics of A*. The process may involve the emission of a photon by D* 

which is re-absorbed by A (radiative transfer) or a non-radiative process. 

Radiative transitions require the acceptor to absorb the photons 

emitted by the donor. The probability (rate) of transfer will depend upon the 

quantum efficiency of emission of the donor, the number of acceptor 

molecules in the path of the emitted photon and the extent of overlap 

between the emission spectrum of the donor and the absorption spectrum of 

the acceptor. Radiative energy transfer can occur over extremely long ranges 

as for instance photochemical changes occurring on Earth under the 

influence of sunlight. 

Non-radiative energy transfer occurs through the interaction between 

the donor and acceptor molecules (no matter how small). Two theories have 

been put forward. The coulombic 32 interaction can be expressed as ahumber 

of terms, dipole-dipole, dipole-quadrupole, multipole-multipole of decreasing 

significance. The contribution of the normally dominant dipole-dipole 

interaction to energy transfer has been studied in depth by F6rster 24. The 

distance between the excited donor molecule and the ground state acceptor 

molecule at which energy transfer to A and internal deactivation of D* are 

equally probable is known as the critical transfer distance. Forster has 

calculated that energy transfer by dipole-dipole interaction can be significant 

even at distances of the order of 10 nm. The only transfer processes allowed by 

the coulombic interaction are those in which there is no change in spin. The 

26

ate constants for multipole-multipole and dipole-quadrupole interactions 

fall off with distance much more rapidly than that for dipole-dipole transfer 

and they are therefore only likely to be important at distances less than 4 run. 

The electron exchange 33 mechanism is an extremely short range 

process, as it involves the overlap of the orbitals to enable electrons to 

transfer. The energy transfer can be formulated thus: 

122 

Due to the intervention of a bimolecular intermediate, which could be 

an exciplex or a collision complex, the energy transfer will be subject to 

conservation of electron spin. Under Wigner's spin rules 34, both the 

following transitions are occasionally allowed. 

3D* + 1A_* 1D + 3A* 1.23 

1D* + IA.....> 1D + 1A* 1.24 

The former (triplet-triplet energy transfer), though allowed under the 

exchange interaction, is doubly forbidden under the coulombic dipole-dipole 

interaction, whilst singlet-singlet energy transfer is allowed under both 

interactions. 

1.8 Kinetics of fluorescence 

Consider the following scheme (Figure 1.10) where I is the rate of 

absorption of photons, K 1 and K are the rate constants for fluorescence and 

phosphorescence respectively and K ic and K1 are the rate constants for 

internal conversion and inter-system crossing. 

27

so 

hij(I) 

rS i 

____ lKiw 

Kic 

4 

S0 + hty' 

Figure 1.10 Diagram representing the processes involved in populating and 

depopulating excited states 

Under continuous irradiation, the steady state approximation is 

applicable to excited states and it follows that the rates of formation and 

destruction of Si are equal, i.e 

where 

1= [Si] E1K 

= Kk + Kf + isc 

1.25 

1.26 

The quantum yield of fluorescence is given by 

Of 

- Rate of emission by S 1 - K 1 [S 1] - K 1 

- Rate of absorption of photons by S - I - 

1.27 

If tf is the actual measured lifetime (t1 = 1/ 1K) then 

4f = K't 1.28 

Given the definition of radiative lifetime; T o = 1/K1 where 'C 0 is the time 

taken for the population to diminish to lie of its initial concentration 

assuming that no other non-radiative processes are occurring, the 

relationship between radiative lifetime and the actual measured lifetime of Si 

are given by the expression 

1f t4 1.29 

13

1.8.1 The Stern-Volmer equation 

The kinetics of fluorescence quenching may be treated by the use of the 

Stern-Volmer equation. Consider a fluorescing excited species which also 

decays through a non-radiative route or quenching with a rate constant K q. 

Applying the steady state approximation to the concentration of S 

gives 

I = [Si].(K1 + IC4 + K q[QI) 1.30 

where Kd = K1 + 1Kisc and [Q] is the concentration of the quencher. 

Therefore 

Of = K1[Si]/I = K f / (Kf+Kd+Kq[Q]) 1.31 

The quantum yield of fluorescence in the absence of quencher (4) is given by 

Hence 

= K1 /(K1+K4) 1.32 

lKf +Kd +Kq [Q] Kq[Q] =1+K q +t[Q] .......... 1 33 

K1+K4 

Of 

This is the Stern-Volmer relationship, where tis the fluorescence lifetime 

in the absence of quencher. 

Under ideal conditions, the plot of $7 / o f against 

quencher 

concentration gives a straight line of gradient K 'tv. i t is known, the 

quenching rate constant may be obtained. 

For many systems, Kq in mobile solvents is of the order of 10 9_10 1 dm3 

mol -1 sec-1 which is close to the diffusion controlled rate constant Kdiff. This 

suggests that in these cases quenching is so rapid that the rate determining 

step is the diffusion of the quencher to within the active sphere of S1.

1.9 Acid-base properties 

Both ground state absorption spectra and excited state emission spectra 

are very much pH dependent. For the aromatic aminoacids, there is a 

possibility of absorption I emission from the anion, neutral and cation 

species. The ground state pK a and pKi, values can be calculated easily by 

varying the pH of the solution and measuring the absorption spectra. 

Measurement of the excited state pK a values is somewhat more complicated. 

Excited state pK values are usually very different from those of the ground 

state 35. This phenomenon was explained by Förster who postulated that 

proton exchanges are so rapid that an acid-base equilibrium is established 

between the excited molecule and its conjugate base during the lifetime of 

the excited state species. 

HA 

ho) HA (_) 

1- 

HA+hi 

A+htf' 

Figure 1.11 Protonation and deprotonation reaction scheme 

In general, excited state molecules are stronger acids than the 

corresponding ground state molecules. Two methods have been employed to 

measure the excited state pKa values of excited singlet states. Both methods 

assume proton equilibrium during the lifetime of the excited state. 

1) The p1Ca* is that pH at which the fluorescence of HA* drops to one half 

of the intensity observed in solutions so strongly acid that only HA* emits. It 

is also the pH at which the intensity of the fluorescence of A* drops to one 

half of the value observed in solutions so alkaline that only A* emits. 

GTXI

2) This method depends on the Forster-Weller cycle. 

* 

-I 

(S1 )i- 

(SD) 

S0 ) 

Figure 1.12 Forster-Weller cycle for deriving pK values for excited states 

From the diagram it can be seen that 

AEHA-AEA.. = AH - = AG - AG* 1.34 

assuming that the entropy of dissociation is the same in both ground and 

excited state. It follows that 

- AE A.. = 

LnI.-') = 

LK) RT kT 

1.35 

where K* and K are the dissociation constants for IIA* and HA, and bsv is the 

difference in the frequency of absorption or emission of HA and A - . 

Knowledge of Av and K therefore allows calculation of K* 

1.10. Literature background to the project 

There have been several sporadic studies of the absorption and 

fluorescence properties of the isomeric aminobenzoic acids, with the earliest 

reported experiments those of Doub and Vandenbelt 36 in 1947, who studied 

the ultraviolet absorption characteristics of monosubstituted and 

p-disubstituted benzene derivatives. Included in their studies were benzene, 

31

enzoic add, aniline and the protic species of 2-aminobenzoic acid. Doub and 

Vandenbelt36 were initially interested in the effect of substituent groups on 

the wavelength of maximum absorption and the appearance of a secondary, 

and in some cases, a tertiary absorption band. They found that in the case of 

ortho- or pa Ta- directing substituent groups that the absorption peak was red 

shifted with the increasing ability of the substituent group to donate electrons. 

In 1949, they extended their studies to include a more extensive examination 

of the ortho- and meta- disubstituted benzenes 37 and studied the ground state 

absorption properties of these molecules as a function of pH and solvent as 

well as considering substituent effects. In 1955, Doub and Vandenbelt 38 

extended the range of solvents and substituent groups used, as well as 

investigating the effects of a third substituent group on the ground state 

absorption spectra. The first reported fluorescence spectra for these 

compounds came from Melhuish 39, who reported the fluorescence quantum 

yields of 2-aminobenzoic acid and methyl 2-aminobenzoate in ethanol and 

investigated oxygen quenching effects. The quantum yield of 2-aminobenzoic 

acid was reported as 0.56 I 0.59 and for methyl 2-aminobenzoate 0.66 I 0.69 in 

the presence and absence of oxygen respectively. However the concentrations 

used for this study were in the order of 10.2 mol dm-3, leading to problems 

with inner filter effects and dimerisation. In 1963, Mataga 40 extended the 

studies of Doub and Vandenbelt 36 by measuring both the absorption and 

emission properties of the isomeric aminobenzoic acids as a function of 

solvent, finding that the solvent effects upon the absorption spectra differed 

markedly from those on the fluorescence spectra. He plotted Stokes shifts 

against the solvent parameter F(D,n) (Equation 1.36), concluding that 

deviations from this relationship were due to strong intra- or inter-molecular 

hydrogen bonding which was dependent on solvent. 

( D+l n 2 +1 

F(D,n)=i + I 1.36 

2D+1 2n2 +1) 

where n is the refractive index and D is the dielectric constant of the solvent. 

32

COOH 

NH2 

COOH 

I 

NH3 

Kz 

NK, 

/Ka 

6~1~l r 

z 

Figure 1.13 Reaction scheme proposed by Leggate and Dunn 41 utilising 

nomenclature proposed by Tramer37 

In 1964, Leggate and Dunn 41 reported an extensive study of orthoaminobenzoic 

acids, and used the Hammett equation to calculate the 

equilibrium constant between the neutral form of 2-aminobenzoic acid and 

its zwitterion structure, utilising the above scheme. 

Leggate and Dunn 41 concluded that the ionisation constants of 2- 

aminobenzoic acid gave an excellent fit to the two parameter Hammett 

equation, suggesting that this acid at its isoelectric point is better represented 

as a mixture of zwitterion and neutral species, than as a single hybrid species. 

Kuhn and Geider 43 calculated K z for 2-aminobenzoic acid (2ABA), 2-Nmethylaminobenzoic 

acid (2MABA) and 2-N,N-dimethylaminobenzoic acid 

(2DMABA). Complementing this earlier work, Tramer 42 utilised both ground 

state absorption and JR studies to deduce more information about the 

zwitterion / neutral equilibrium, finding startling differences between 2ABA 

33

and 2DMABA. He showed that 2MABA dissolved in inert solvents behaves 

in an analogous manner to 2ABA and exists in the neutral molecular form 

(M), and that there is no evidence for an intra-molecular hydrogen bond. The 

reverse is the case in polar solvents, in that the predominant form is the 

zwitterion structure, assumed to be stabilised by solute/ solvent interactions. 

When the strong intra-molecular hydrogen bond is present in 2DMABA, the 

conjugation of the nitrogen lone pair electrons with the it electron system of 

the benzene ring is completely removed as shown by the extremely weak 

fluorescence spectrum observed in hexane. The difference between 2MABA 

and 2DMABA in terms of ability to form a hydrogen bonded complex was 

thought to be due to the enhancement of the basic properties of the amino 

group by the inductive release of the two methyl groups thus increasing the 

strength of the hydrogen bond. Tramer appears not to have looked into the 

major difference between 2DMABA and 2MABA, namely that 2DMABA will 

involve considerably more steric hindrance between the COOH and N(CH3)2 

substituents because of the bulky nature of the N,N-dimethylamino groups. 

Tramer 44 extended his studies of these three compounds by measuring their 

fluorescence and phosphorescence emission spectra as a function of solvent 

and pH. Although there is little new information in this 44 report, dimer 

formation was observed in more concentrated solutions of 2MABA in polar 

solutions. Tramer's calculated pKa values are in accordance with those 

reported by other authors. Tramer 42 and Kuhn and Geider 43 have calculated 

K Z constants in reasonably concentrated solutions (approximately 10.2 mol 

dm -3) a concentration which in most solvents will lead to dimerisation. All 

their calculations are based on the appearance of a peak at around 320 nm 

which has an extinction coefficient of 10 dm 3 mol -' cm', so it is not 

surprising that a similar study by Jian et al. 45 in 1986 made no mention of 

this Z-M equilibrium. In a similar manner to Tramer, Jian et al.45 measured 

the absorption and emission spectra of the isomeric aminobenzoic acids in

different solvents and as a function of pH. This is the first reported study of 3- 

aminobenzoic acid. They also studied the acid base equlibria in the ground 

and excited states. Jian et al.45 appear not to have taken into account the 

potential for hydrogen bonding. They do however include the results of 

CNDO calculations, calculating the wavelength of maximum absorption and 

the oscillator strength of the various types of transitions for all three isomeric 

aminobenzoic acids, which appear to correspond well with the experimental 

values. In conclusion they reported that in all three isomeric aminobenzoic 

acids the lone pair on the nitrogen is in conjugation with the it electrons of 

the aromatic ring system and this may lead to the energy of the ic—its 

transition being nearly equal to or slightly lower than the n1t* transition 

energy. That the observed absorption band is a ic-it" transition is borne out by 

the high extinction coefficients and the observed solvent effect, namely that 

increasing solvent polarity and / or hydrogen bonding ability leads to a red 

shift in both the absorption and emission spectra. 

There have only been three publications regarding the fluorescence of 

4DMABA, two by Cowley et al. 46,47 and the third by Brown and Revill 48 with 

only a handful of papers on the methyl esters of 4DMABA (M4DMAB and 

E4DEAB) 47' The fluorescence properties of these three compounds are 

anomalous in that they show dual fluorescence, the source of which is still a 

matter of debate. 

1.10.1 4-Dimethylaminobenzonitrile 

(During this discussion and throughout the Results we will utilise the 

accepted terminology for these systems where the "normal" blue fluorescence 

is attributed to an excited state labelled b* and the red "anomalous" 

fluorescence to a state labelled a".) 

Since the discovery of the dual fluorescence of 4-(N, N - 

dimethylamino)benzonitrile (DMABN) in 1962 by Lippert et al. 25, several 

"lii

theories have been proposed to explain the source of the anomalous 

fluorescence. Kosower and Dodiuk 55 proposed that the dual fluorescence was 

due to a species in which the nitrogen on the amino group was protonated, 

the proton being donated by the solvent. However this theory can be almost 

immediately ruled out since in this latter paper it is shown that dual 

fluorescence occurs in both acetonitrile and butyronitrile, both of which are 

normally incapable of proton donation. Although there has been no more 

evidence to support this theory, Kosower and Dodiuk also identified other 

emissions as being due to a dimer and perpendicular and planar monomers. 

The idea of a perpendicular monomer (conformational isomer) was also 

proposed by Lippert 56 and Grabowski et al. 26 . Grabowski et al. 26 . used the 

range of compounds shown in Figure 1.14 to argue for the acceptance of the 

idea of conformational isomers being responsible for the anomalous 

fluorescence. This was the initial proposal of the idea of a Twisted Intramolecular 

Charge Transfer or TICT state. 

Me\ /Me 

N 

Me 

/ 

N 

- - 

Me I- 

-fl 

• I 

• I 

• I 

I 

I 

I 

I 

C 

It' 

N 

C 

I'' 

N 

C 

It' 

N 

• 

C 

• • 

N 

. I 

I 

ED 

Wi 

Iv 

Figure 1.14 Compounds utilised by Grabowski et a126 to argue the existence 

of the TICT phenomenon

Structure IV is a representation of the TICT state of DMABN 

(compound I) which shows two fluorescence bands in polar solvents, while II 

shows only a single fluorescence band, which closely resembles a typical b* 

emission. Compound III, which one might expect to show a similar 

fluorescence spectrum, actually shows a fluorescence band representative of 

the a* emission. There is however one major difference between compounds 

II and Ill, the orientation of the lone pair of the nitrogen. In II the orientation 

of the lone pair is parallel to the plane of the ring system and in III orthogonal 

to the plane of the ring system. Thus the structural evidence seems to suggest 

that the anomalous emitting state in I corresponds to a twisted rotamer lv. 

Although this appears to be evidence of TICT formation, it is only based upon 

data acquired by static experiments and dynamic evidence was absent27' 7. 

In 1985, Heisel et al. 58 reported that using fast laser spectroscopy, they 

had found the kinetic relationship between a* and b* which was first 

predicted by Grabowski et al. 54. Their work agrees with the findings of 

Huppert et al.60 who found pulse limited rise times in the emission of the a* 

state which feeds directly the emission for the b* state. The fluorescence 

kinetics of the a* and b* states were found to vary with solvent and 

temperature, with the decay of the a* showing single exponential kinetics and 

b* dual exponential decay kinetics. These findings were interpreted as 

suggesting either thermally assisted intersystem crossing to a solvated triplet 

or thermally activated internal conversion to the ground state. 

Heisel et al.59 . also found a rise time of approximately 40 ps for the a* 

emission and the kinetics of the a* peak were elucidated by analysing the 

kinetics of the b* emission. They found that one of the components of the 

decay of the b* fluorescence was the same as the rise time of the a* peak, 

lending more proof to the theory that a* and b* are formed sequentially. They 

also' found that the b* fluorescing state was formed within 5 ps followed by 

the a* state within 40 ps. Their data agree with the proposal of Rotkiewicz et 

37

al.61 that the b state is the kinetic precursor of the a* state with kinetics 

which are found to be very viscosity and temperature dependent. Later in 

1985 Heisel et al. 59 calculated the rates of TICT formation. Their experimental 

evidence shows that the apparent activation energy for TICT formation is 

very solvent dependent, which is in agreement with the ideas of Rettig and 

co-workers 2527. Kobayashi et al.62 and August et al.63 have investigated the 

solution photophysics of DMABN using laser induced fluorescence (LIF) and 

proposed that the a* fluorescence was produced by a Van der Waals complex 

with the resulting spectra being attributed to the internal rotation of the N,Ndimethylamino 

group. They found no evidence for dimer or higher aggregate 

formation. August et al.&i also investigated the jet cooled LW fluorescence of 

DMABN in the gas phase. They interpreted their results in a similar manner 

to Kobayashi et al.62' finding excellent agreement with the expected dynamics 

of TICT state formation. 

However there still appears to be some confusion in the literature, 

with Mataga and Nakashima 64 in 1973 publishing results which indicated 

that the anomalous fluorescence was concentration dependent. They found 

that in toluene and diethyl ether solutions, the ratio of the a* / b* 

fluorescence intensities for DMABN increased over a concentration range of 

1x1O -S to 0.1 mol dm-3 and that the a* band showed a red shift at very high 

concentrations. Mataga and Nakashima 64 do not propose any mechanism for 

this apparent concentration dependence. However in polar solvents like 

methanol and acetonitrile, the intensity ratio decreased with the a* band 

showing a slight blue shift at higher concentrations( i.e. dimer / excimer 

formation). This shift in the wavelength of maximum emission was 

concluded to result from a change in the dipolar interaction between the 

excited state and the ground state, which is facilitated by higher 

concentrations. In all solvents studied no observable change could be seen in 

the ground state absorption spectra. The only solvent in which there was no 

RN

change in the observed fluorescence intensity was propylene glycol, from 

concentrations of 1x10 -5 up to 5x10 -3 mol dm-3' which is in complete 

contradiction to the findings of Khalil et al. 65 However, Khalil's findings 

agree with the rest of Mataga's results. An additional peak observed by Khalil 

et al. at 340 nm was ruled out by Mataga who found this was caused by an 

impurity which disappeared on repeated recrystallisation from n-hexane. 

In 1981 Lippert et al. 56 reported that the dual fluorescence of DMABN 

at sufficiently low (but unstated) concentrations was not concentration 

dependent, but in 1-chlorobutane was found to be temperature dependent. 

Chandros 66 has proposed that the anomalous fluorescence is due to a 1:1 

complex comprising an excited state monomer with a solvent molecule (i.e. 

an exciplex), being stabilised by more conventional specific solvent - solute 

interactions. Chandros 66 observed that the "normal" fluorescence band 

observed in methylcyclohexane followed a linear Stern-Volmer plot on being 

quenched by additions of triethylamine, and that, on addition of propionitrile 

to the methylcyclohexane solution, the b* fluorescence diminished and the a* 

fluorescence appeared. Utilising extremely similar experiments to Chandros, 

Varma et al. 50 proposed that the source of the exciplex must involve a a-type 

bond between the lone pair electrons of the amino nitrogen atom on the one 

hand and lone pair electrons of the polar solvent on the other. This has been 

challenged by Suppan 67, who has reported experimental evidence that the 

dual fluorescence observed for DMABN comes from a dielectric-stabilised 

highly polar twisted intra-molecular charge transfer state of DMABN and not 

from solute-solvent exciplexes. He also found that DMABN in 

perfluorohexane (which has 14 lone pair donor F atoms) shows only single 

fluorescence whilst in 1-fluoropentane, which has one F atom, it shows dual 

fluorescence, concluding that the solvent lone pair of electrons are not 

involved in the a* emission. 

39

In 1988 Varma et al.68 69 extended their work to inyolve time resolved 

measurements and found that the normal fluorescence required the sum of 

three exponentials to adequately fit the data. The three species were attributed 

to excited solute - solvent complexes, excited bare solutes and ground state 

solute-solvent complexes. Previous to this work, only dual exponential 

decays had been observed. Nag et al.51 studied the solution chemistry of 

DMABN and in a similar experiment to Varmaet al.70 and Chandros 66 added 

propionitrile to a solution of DMABN in an inert solvent. They observed that 

as the percentage of added propionitrile was increased the b* emission 

decreased with increasing a* emission (as observed by Chandros 66 and Varma 

et al.70). This was attributed to a reduction in the activation energy of TICT 

state formation by the added propionitrile. This trend was only found to be 

true up to 15% v/v, when the addition of more propionitrile caused the a* 

fluorescence to decrease in intensity. This was explained in terms of nonradiative 

processes becoming more favourable and it was concluded that the 

ratio of b*/a* emission vs % propionitrile is basically linear. Weisenborn et 

al.68 69 state that both the quenching of the a* and the growth of the b* exhibit 

linear Stern-Volmer behaviour as a function of the concentration of the polar 

compound P in the solvent mixture with cyclohexane and consider this to be 

strong evidence for exciplex formation between the excited DMABN 

molecule and P. 

As can be seen from the above discussion, there is still much confusion 

about the source of the anomalous fluorescence shown by the widely studied 

DMABN system. There are also several closely related molecules which 

exhibit this dual fluorescence. In particular, the methyl and ethyl esters of 4- 

N,N-dimethylaminobenzoic acid have received a certain amount of 

attention46' 47, 70 

40

1.10.2 4-Dimethylamino and 4-Diethylaminobenzoic acid and their 

methyl or ethyl esters 

There is one important difference between the energy levels of 

4DMABA, and M4DMAB (or their ethyl counterparts) as apposed to DMABN, 

in that the 'La and 'Lb states are inverted in energy 49. In similar experiments 

to those of Varma 50 and Nag et aL 51 , Cowley et a147 added small amounts of 

polar solvent to an inert solution of M4DMAB in cyclohexane and noted the 

appearance of dual fluorescence. Varma 50 attributed this to exciplex 

formation through the interaction of M4DMAB with a polar molecule. 

Cowley 47 however attributes this to solvent relaxation allowing the 

formation of the TICT state i.e. lowering of the energy of activation for TICT 

state formation. Cowley 47 concluded that ground state dimer formation 

seemed unlikely, since the ratio of the two bands was independent of 

concentration and that excimer formation could be ruled out because of the 

low concentration used (10 mol dm -3). However our results disagree with 

this statement as we find that the ratio of the two bands are concentration 

dependent48' 52• Quantum chemical calculations by Rettig, taking into account 

a-electrons and solvent effects, suggest that for the related molecule ethyl 4- 

diethylaminobenzoate (E4DEAB) dual fluorescence will be observed in both 

polar and non-polar solvents due to formation of a TICT state. Varma and coworkers 

53 believe that 1:1 exciplex formation is responsible for the dual 

fluorescence, whilst Phillips' group 49 disagree with Rettig but are in partial 

agreement with Varma 53 and ourselves in that a ground state species or and 

excimer / exciplex is the cause of the anomalous fluorescence. 

In conclusion, the recent studies on both 4-dimethylaminobenzonitrile 

and 4-dimethylaminobenzoic acid and its esters suggest that the origin of the 

observed dual fluorescence is not fully understood and may be due to some 

form of exciplex formation 66 or to ground sate dimer / aggregate formation 71 

or to a TICT state72 . 

41

1.10.3 Aims of this project 

By studying both the ground and excited states of the 2, 3 and 4- 

aminobenzoic acids, N,N-dimethylaminobenzoic acids and their methyl 

esters, it was hoped to understand how solvent, steric and substituent effects 

alter the photophysics of the molecules. In the case of the 4-isomers it was 

confirmed that when the substituent was a N,N-dimethylamino group, dual 

fluorescence was observed, so through various experiments utilising both 

steady state and time-resolved spectroscopy it was hoped to elucidate the 

origin of this anomaly. 

42

Chapter 2

2.1 Materials 

2.1.1 Solvents. 

The pure solvents (purchased from the Aldrich Chemical Company) 

and binary mixtures used in this study, which are given in Table 2.1 below, 

were either spectrophotometric or spectrofluorometric grade and were used 

as received except for acetonitrile, butyronitrile and water. The acetonitrile 

and butyronitrile were distilled over phosphorous pentoxide, collected 

under an inert atmosphere and then distilled again over potassium 

carbonate. Even then some fluorescence from the butyronitrile remained. 

Water was doubly distilled and then passed through an ion exchange 

column. Hexanes refers to a mixture of non-cyclic six carbon compounds. 

Relative 

dielectric 

constant 

Dipole 

moment 

Viscosity 

coefficient 

Refractive 

Index 

Solvent or Binary mixture eie, 11 

acetonitrile 37.5 3.37 0.375 1.3441 

butyronitrile / isobutyronitrile, (9:1) 20.3 3.57 0.624 1.3860 

chlorobutane / isopentane (9:1) 7.09 2.11 1 0.469 1.4021 

ethanol 24.3 1.68 1 1.078 1.3613 

n-hexane 1.89 0.08 1 0.324 1.3748 

hexanes 

assumed similar to n-hexane 

methylcyclohexane / isopentane (4:1) 1.95 1 0.00 0.324 1.3748 

water 81.7 1.85 0.891 1 1.3404 

Table 2.1 

Solvent properties 

The four pure solvents used in this work were chosen for their range 

of properties, varying from hydrogen bonding to inert and non-polar 

solvents. The three binary mixtures were chosen for the same reason but 

also for their ability to form good glasses at low temperature (77K). Some 

useful solvent properties are reported in table 2.1. 

2.1.2 Chemicals 

Quinine sulphate monohydrate and 9,10-diphenylanthracene which 

were used as quantum yield standards were purchased from the Aldrich 

43

Chemical company (Gold star) and were used as received. The 

aminobenzoic acids were purchased from Sigma I Aldrich or from Lancaster 

Synthesis and were recrystallised before use. Details of their properties are 

given in Table 2.2. The methyl esters were prepared by one of two methods: 

Method a:73 A solution containing methyl iodide (0.03 mol) in a small 

quantity of benzene was added to a solution of 2-aminobenzoic acid (0.01 

mol) in benzene and 1,8-diazobicyclo[5, 4,0]undec -7-ene (0.03 mol). The 

reaction mixture was stirred at room temperature for four hours and then 

washed with sodium hydrogen carbonate solution. Extraction with ethyl 

acetate and removal of the dried solvent gave methyl 2-aminobenzoate 

which was distilled twice. 

Method b: 74 A solution of 4-aminobenzoic acid (0.01 mol) in methanol (ca 

50 cm3) was added to boron trifluoride-methanol complex (0.03 mol) and the 

mixture was boiled under reflux for four hours. After cooling, the mixture 

was poured into saturated sodium hydrogen carbonate solution. The ester 

was extracted using ethyl acetate and the solution dried with magnesium 

sulphate. The ethyl acetate was removed under reduced pressure and the 

resulting methyl 4-aminobenzoate was twice recrystallised from ethanol. 

2.2 Instrumentation and Instrumental techniques 

22.1 Absorption spectra 

Absorption spectra were measured in the seven different solvent or 

binary mixtures on a Perkin-Elmer Lambda 3 spectrophotometer or a 

Hewlett-Packard HP8451A diode array spectrometer using matched quartz 

cells of variable pathlengths (1mm to 10mm). The accuracy of the peak 

wavelengths is estimated at +1- 2nrn and the extinction coefficients at +1- 

10%. Unless otherwise stated the concentration of the solutions used was 

around 1*10.4 mol dm 3 . 

44

— — — — 

- 

VP 

'0 

ret 

Cl 

LI) 

'0 

— — — — — — — 

cek 

'.o 

£ 

-Scnt ... 

'c -' 

as 

en reS cu 

0 .CIJ Cl 'a 

trt EEeut €E E 

—'-- 

?-t O ' 

.c-jt' en '0 tn in Cl 

rZ€'a j'at' "'.dt rC 

-toc't 

Gs 

2-S.2-2-a!S.S..2.aS.a 

- 

'a 

C 

IZ 

V 

a) 

-t 

I-. 

a) 

-a 

a) 

V 

. xx 

a 

U, 

5 

000 

II 

- 

II 

II 

— - 

—---- 

00 00 00 

I I I II 

ZZZ .0 

II 

u U II II II 

0 00 

- N- 

CICI)C) 

enenen r ' -- 

xxz z xx xx 0 00 

66 6 ±± . 6 66 6 66 

'00 

cQN 

Ot-' 

(I)(1) - —- 

I-. 

a) 

-E 

C 

ft 

m 

-a 

-ø 

ft 

U 

0 

a) 

0 

E 

cc 

C, 

'C 

a) 

S 

0 

a) 

V 

I 1 CI)U) C'. 1 1 

0 U)t 

tC t-- 

Nb'. 

qa- - 

a- ento 

a — N aCflC'1 Cl 

u - I - I - - '- - Cl O.CN 0C1 

cq 

LrI Cl Cl - 

a-. 

e t' cno 

oc.-r'? rtl 

!7b? 

cn'd' 

0' 

ON 

a) cc' tact 

O cctct' .s<

2.2.2 Ground and excited state pK values 

a) The pK values for ground state protonation and deprotonation 

were calculated by measuring the ground state absorption spectra as a 

function of pH75. The method depends upon the direct determination of the 

ratio of molecular species (neutral molecule) to ionised species in a series of 

non- absorbing buffer solutions with a span of at least half a pH unit. For 

this purpose, the spectrum of the molecular species was first obtained in a 

buffer solution whose pH was chosen such that the compound studied was 

present wholly as this species. This spectrum was then compared with that 

of the pure ionised species similarly isolated at another suitable pH. A 

wavelength was then chosen at which the greatest difference between the 

absorbances of the species was observed. If it is assumed that Beer's law is 

obeyed for both species, then the observed absorbance (A) at the chosen 

wavelength is due to the sum of the optical densities of the ionised species, 

A 1, and the molecular species Am. The absorbance is related to the 

concentration of the relevant species by A=Ecl. The concentration of the 

ionised species in the mixture is ac where a is the fraction of molecule 

ionised and hence its contribution to the observed absorbance is c 1acl and 

similarly the contribution from the molecule is Em (1-(X)cl where a is given 

by 

i-a = Ka /([H+]+Ka); a =[H.i-]/([H+]-i.Iç) for acids ..........2.1 

i-a= [H+]/([H+]+K1,); a= Kb /([H+]+Kb) for bases ..........2.2 

If the same cell pathlength is used throughout then 

A=(ej(X+em (1-cx))c 

and therefore, since E =A/c and a is defined above. 

46

EjK. 

= + €m[H] foracids . 2.3 

[H]+K0 [W]+K0 

£ = s_K0 + for bases 2.4 

[H]+K0 [H]+K0 

Provided that the same total concentration is used for all measurements, 

equations 2.3 and 2.4 may be written with the appropriate absorbances 

replacing the extinction coefficients. Each equation can be rearranged into 

two forms to suit different needs. When the functional group being 

determined is an acid, equation 2.5a is used if A 1 is greater than A. and 2.5b 

if the reverse is the case. 

pKa=pH+Iog A 2.5a 

..........2.Sb 

When the group being determined is a base, equation 2.6a is used if A is 

greater than Am, and 2.6b if the reverse is the case. 

pKb = pH + log t AJ 2.6a 

pK=pH+logAmA 2.6b 

A—A 1 

The pK a or p1

pH A ArnA 

Log(m) pKa 

1.50 0.1480 0.5948 0.1028 0.7624 2.26 

1.67 0.2002 0.5426 0.1550 0.5442 2.21 

1.82 0.2529 0.4899 0.2077 0.3727 2.19 

2.10 0.3491 0.3937 0.3039 0.1124 2.21 

2.26 0.4137 0.3291 0.3685 -0.0491 2.21 

2.54 0.5009 0.2419 0.4557 -0.2750 2.26 

2.73 0.5659 0.1769 0.5137 -0.4630 2.26 

Table 2.3 Determination of the ionisation constant for protonation of 2- 

aminobenzoic acid (1.92*104 mol dm-3). Temperature 20 0C, 

analytical wavelength 326 nm 

Once the pKa is known, it is possible to calculate the percentage of ionisation 

given the pH value using equations 2.7 and 2.8. 

100 

% Ionised = for acids 2.7 

1 + anti log(pK - pH) 

%Ionised = 

100 

1 + anti log(pH - pK) 

for bases 2.8 

b) Excited state pK* values may be determined in a similar 

manner to the ground state pK values detailed above 35 Variation of 

fluorescence intensity with pH yields a values for the excited states 

involved in the photochemistry and these data in turn may be used to 

determine pK* values via the Henderson-Hasselbalch relationship 80 . In 

experiments of this type, it is essential to excite the fluorescence at the 

isobestic point for any ground state equilibrium which is taking place in the 

pH range of interest. This ensures that changes in the fluorescence spectra 

due to (de)protonation processes are isolated from changes in the spectra 

caused by variations in the absorbed light intensity. 

B1

The effect of pH on the fluorescence spectra of all twelve compounds 

in aqueous buffers was studied in the pH range 11.0 - 2.00 and the results are 

reported in Chapters 3 - 5. 

c) The excited state pK values were calculated using equation 2.9 

based on the FOrster cycle (see figure 1.12) 

pK* - pK= 2.1 x 1ft 3(v -VA) 2.9 

where V 

base respectively. 

and VA are the band maxima (in cm 4) of the acid and conjugate 

2.2.3 fluorescence spectra 

Fluorescence spectra were measured using either a Perkin-Elmer LS5 

fluorescence spectrometer in the fully corrected mode or a Perkin-Elmer 

LS50. Unless otherwise stated a concentration of around 1*106 mol dm 3 

was used. The low temperature fluorescence spectra were recorded using 

equipment at the SERC Daresbury Laboratory (see section 2.2.4), utilising a 

xenon light source in place of the synchrotron radiation source, and an 

Oxford Instruments CF204 cryostat. The solution was placed into the sample 

chamber and the sample jacket was evacuated. Liquid nitrogen was 

transferred to the cryostat. The temperature was modulated by either 

restricting or increasing the flow of the liquid nitrogen. Measurements were 

taken at decreasing temperature from ambient rather than on warming up 

of the solution from 77K in order to reduce adsorption of water into the 

solution. Samples were not used more than once for this same reason. 

Spectra were only corrected for blank solvent fluorescence, which in the case 

of the butyronitrile / iso-butyronitrile mixture at temperatures of 133K and 

less was quite intense. 

49

Quantum yields were measured on optically dilute samples 

(absorbance .c 0.05) to eliminate the effect of self quenching and inner filter 

effects and so that the fluorescence intensity was proportional to 

concentration / absorbance (see introduction). The samples were degassed by 

bubbling with oxygen-free nitrogen for twenty minutes. 

Quinine sulphate (1.28 x 10-6 mol dm -3) dissolved in perchloric acid 

(0.01 mol dm4) was used as the quantum yield standard76' 77. A stock 

solution (0.00128 mol dm- 3) of quinine sulphate (0.0979g) in perchloric acid 

(1 dm3, 0.01 mol dm -3) was prepared. Dilutions were freshly made up on a 

daily basis. The lifetime of the stock solution was around one month. 

A BBC microcomputer (and later an IBM 386 personal computer) was 

interfaced through the chart output of the Perkin-Elmer LS5 fluorescence 

spectrometer and spectra were recorded on the computer. This enabled the 

area under the curve to be calculated. By using the following equation, 

unknown quantum yields were calculated (see Appendix A for full listing of 

program): 

• 1 (unknown) = Fluor. int. (unknown) x Abs(standard) 

4 1 (standard) Fluor. int. (s tan dard) x Abs(unknown) 

2.10 

where o f = fluorescence quantum yield, Fluor.int. = integrated area under 

the corrected fluorescence spectrum and Abs is the absorbance of the sample 

at the excitation wavelength. 

Phosphorescence spectra (77K) were measured on the Perkin-Elmer 

LSS spectrofluorimeter in phosphorescence mode with background 

correction using a cryostat attachment. An attempt was made to determine 

the phosphorescence lifetime by keeping the time gate (tg) constant and 

measuring the intensity at different delay times ('Cd). This was unsuccessful. 

50

2.2.4 fluorescence decay profiles 

Time-resolved fluorescence spectroscopy was performed using the 

single photon counting technique 78, using a variety of excitation sources; 

the Synchrotron Radiation Sources (SRS) at Daresbury, UK and BESSY, 

Berlin, Germany, a hydrogen flash lamp at the University of Strathclyde, 

Glasgow and a frequency doubled dye laser at the Rutherford Appleton 

Laboratory (R.A.L). The main work was undertaken at the SRS at Daresbury. 

Operating in single bunch mode the SRS provides pulses of light :5 200 ps 

duration of frequency of = 3.0 Ml-Iz. Wavelength selection is achieved with a 

Spex 15005P Czerny-Turner monochromator with continuously variable 

slits. Sample fluorescence was detected at right angles to the excitation beam 

with a Peltier-cooled Mullard PM2254 photomultiplier. Wavelength 

selection on the emission side was achieved with Barr and Stroud 

interference filters (bandwidth 10 nm f.w.h.m). Pulses from the 

photomultiplier were fed into an Ortec 583 constant fraction discriminator 

and then into the START channel of an Ortec 457 time-to-amplitude 

converter (TAC). 

Stop pulses for the TAC were generated either from a strip line, 

essentially an aerial close to the synchrotron ring which picks up a signal as 

the electrons pass by and an Ortec 473A constant-fraction discriminator, or 

from the 500 MHz signal used to drive the SRS klystron, divided down by a 

factor of 160 and passed through an Ortec 583 constant fraction 

discriminator. The TAC output was accumulated in 1024 channels of an mo- 

Tech 5400 multichannel analyser and then stored on a winchester hard disk 

on a POP-il mini-computer before being passed to a CONVEX mainframe 

for processing and permanent storage. The fluorescence decay profiles were 

evaluated by computer deconvolution using a kinetic model of the form in 

equation 2.11 

51

Fluorescence intensity(t) = I i a1e'1 . 2.11 

where a i is the amplitude of a component lifetime, t1 (where i = 1, 2 or 3). 

The data were analysed using an iterative convolution method using 

weighted, non-linear least-squares curve fitting routines. The computed fits 

to the experimental data were evaluated on the basis of the X2 values and 

the distribution of residuals. Care had to be taken because although x2 the 

values were a good indication of the suitability of a fit, they could not be 

completely relied upon. Consequently the distribution of the residuals, the 

percentage of each component present and the percentage error on the 

lifetimes were also used to choose the complexity of fit which most 

adequately described the decay kinetics of a particular species. Figure 2.1a 

shows a poor single exponential fit to the decay profile of 2ABA in water 

and figure 2.1b shows the good double exponential fit to the same data. 

a

2—aminobenzoic acid (H20) 

to 

4-I 

io2 

I. 

\4.. 

' I. .. 

lo t 

io o 

0 

2 4 6 0 10 12 

Fit I 

12 

Time/iO sec xso- 3.49 

C 

It 

a-, 

U) -12 

Figure 2.1a Single exponential fit to the fluorescence decay profile of 2ABA 

in water 

53

2-amirrnbenzoic acid (H20) 

In 

= 

0 to 

10 1 

I,, :. .. • 

10 0 - .aI 

Mb 

., 

Ga 2 

Cm 

0 

0. 2 4 B 8 10 12 

Fit 2 Time/10 ° sec X59 0.95 

Figure 2.lb Double exponential fit to the fluorescence decay profile of 

2ABA in water

Chapter 3

.. flu..nsnm 

3.1 Ground state absorption spectra as a function of solvent. 

For all the compounds investigated during the course of this study, the 

meta- isomers would be expected to form the simplest case due to their 

inability to form a resonance stabilised charge transfer state. There is also no 

possibility of steric interaction between the carboxylic acid or ester and the 

amino or N,N-dimethylamino substituent groups, although both substituent 

groups are capable of forming inter-molecular hydrogen bonds with suitable 

solvents. The importance of this steric interaction will be apparent in chapter 

four where the ortho- isomers are discussed. In all of the seven solvents and 

binary mixtures studied, 3-aminobenzoic acid (3ABA), methyl 3- 

aminobenzoate (M3AB), 3-N,N-dimethylaminobenzoic acid (3DMABA) and 

methyl 3-N,N-dimethylamino-benzoate (M3DMAB) all exhibit generally 

weak, UV absorption (extinction coefficients of the order of 1-3,000 dm 3 mol-' 

cm 1). 3ABA (Figure 3.1) and M3AB exhibit a single absorption band above 250 

nm with a wavelength of maximum absorption varying between 300 - 325 

nm (Table 3.1). 

0.16 

0.14 

0.12 

v 0.10 

C 

C 

0.08 

0 

0.06 

0.04 

0.02 

0.00 

200 220 240 260 280 300 320 340 360 380 400 

Wavelength (nm) 

Figure 3.1 Absorption spectrum of 3ABA in hexane 

55

The N,N-dimethylamino compounds have a wavelength of 

maximum absorption varying between 320 - 346 nm (Figure 3.2 and Table 3.1). 

Two other absorption bands are observed around 230 and 200 nm. The 

exceptions to these general comments are 3ABA in hexane and 

methylcyclohexane / isopentane (4:1), where a shoulder is observed on the 

main peak at 285 rim, some 20 rim blue shifted from the main peak (Figure 

3.1). 

The absorption spectra of 3ABA and 3DMABA (Figure 3.2) in 

unbuffered water shows unexpectedly low wavelengths of maximum 

absorption. However, given the pK a values reported later in this thesis (Table 

3.2), it is clear that this apparent anomaly is due to ionisation of the carboxylic 

acid group to form the anion which, in both cases, absorbs to the "blue" of the 

neutral molecule. In ethanol there is a similar though less noticeable effect 

with the wavelengths of maximum absorption being around 317 and 336 nm 

respectively (Figure 3.2 and Table 3.1). 

C, 

1 

0.9 

Acetonitrile 

0.8 Ethanol 

0.7 Water 

0.6 

' 0.5 

C 

. 0.4 

IC 

0.3 

0.2 

0.1 

0 

250 270 290 310 330 350 370 390 


Figure 3.2 Absorption spectra of 3DMABA in water, ethanol and 

acetonitrile 

As expected, replacement of the amino substituent with a N,Ndimethylamino 

group results in a red shift of the absorption band of some 20 

56

nm. It might be expected that there would be a simultaneous increase in 

extinction coefficient, but these seem to remain constant. 

Since the meta- isomer is incapable of forming a resonance stabilised 

charge transfer state, the only interaction between the amino and carboxyl 

groups in the excited state can be through the inductive effect. As can be seen 

from the data in Table 3.1, the substituent group has an important effect on 

both the intensity and wavelength of maximum absorption. 

Solvent Measured 3ABA M3AB MABA M3DMAB 

quantity 

______ _ 

Acetonitrile 323 323 343 

___ S 7dm3 cnr1 mot1 2,400 2,300 0 2,200 

BuCN/IBuCN 322 323 341 

S /dm3cm4mo1' 2,700 3,300 0 2,100 

BCl/iP 320 321 342 

S 7dm3 cur 1 mol1 n/s 2,344 5 2,421 

Ethanol 317 324 

344 

S /dm3cnr1mol4 2,000 2,100 0 2,100 

Hexane 306 312 336 

S 7dm3 cm4 mol4 n/s 2,100 0 2,500 

MCH/iP 316 319 

£ /dm3cnr1mol1 n/s 2,400 0 2,600 

Water 302 317 321 

S /dm3cnr1mol4 1,100 2,500 0 1,400 

Neutral 310 312 E326322 322 

£ /dm3cm4mot1 1,000 2,000 1,000 1,400 

Anion 300 - 308 - 

C /dm3 cm 1 mo11 11,700 1 - 1,400 - 

Cation ?,max Inm 1 272 273 272 274 

C /dr&cm 1 mol4 1 1,000 1,000 900 850 

Table 3.1 Primary ground state absorption characteristics of 3- 

aminobenzoic and 3-N,N-dimethylaminobenzoic acids and their 

methyl esters as a function of pH and solvent 

Alkyl substituents on the aromatic ring generally produce a red shift, 

with a methyl group having the largest effect. Both COOR and NR2 groups 

have the ability to interact with the electrons of the aromatic ring system, 

thereby increasing the wavelength of maximum absorption. The N,Ndimethylamino 

group is a better electron donor than an amino group, whilst 

57

acid groups are better electron acceptors than their corresponding esters. 

However this latter difference is not as significant as the former. Figure 3.3 

shows the possible resonance structures available for aniline and benzoic add. 

HO 0 

cs 

O /0 

HO /0 HO 

C 

/O) 

C 

+af)LJ 

I 

HO 0 

o 

It 

S+ 

'II 

I I 

II 

II 

I I 

I. .7 

INn2 

+ 

NH2 

-0 

00 

+ + 

NH2 

NH2 

NH2 

- 

Figure 3.3 Mesomeric structures of Benzoic acid and Aniline 

Since the two substituent groups are meta - disposed they cannot 

interact mesomerically through the ic-system. Any interaction must be 

through o-bonds, an effect which dies off rapidly with distance. This therefore 

gives rise to the following trend in the amount of charge transfer ability of the 

meta - compounds. 

3DMABA ~! M3DMAB> 3DABA 2: M3AB. 

By increasing the amount of delocalisation of the system the energy of 

the 7t_it* transition will be lower, which will shift the wavelength of 

maximum absorption to higher wavelengths. It is therefore predicted that the 

absorption maxima in a given solvent would follow a similar trend and this 

is generally observed (Table 3.1).

There have been few reports of the absorption properties of these 

compounds in the literature. In early literature (pre 1985), a passing study of 

3DMABA, by Doub and Leanord37 can be found. They observed that at a pH of 

3.75, three absorption bands were observed at 218.5 am (C 14,000 dm 3 mol-' 

cm-1), 250 nm (8 4,400 dm3 mo!- 1 cm-') and 310 nm (S 650 dm3 mol-' cm-1 )37 

However at a pH of 3.75 the species being investigated will be a mixture of 

cation and neutral species (30 / 70). Our results reproduce the tertiary 

absorption band observed around 313 am, which has been attributed to a 1t-t 

transition. 

In a more recent study, Jian measured the absorption spectra of 3ABA 

as a function of solvent and pH and our results agree with his data within 

experimental error (see Table 3.1a). 

So!vent (absorption maxima in nm and respective oscifiator strength) 

MCH Acetonitri!e Methanol iO M HC1 IM HC! Alkaline 

Xmax Amt Xrnax f Amax f Xrnax I Xmax I 

340 320 1 314 0.080 312 0.013 271 0.014 301 0.078 

248 246 1 245 0.166 271 0.008 227 0.199 233 0.456 

208 228 6 216 0.675 221 intense 208 0.032 212 0.636 

Table 3.1a 

Absorption maxima of 3-aminobenzoic acid as a function of 

solvent and pH, reported by Jian et al 

For both acids it was found that the cation absorbed at 272 am (C 1,000 

dm3 mol-' cm-'), with the neutral species absorbing at 326 nm (pH

3.2 Ground state absorption data as a function of pH. 

In water all the meta- substituted compounds are capable of 

protonation and the two acids can also be deprotonated to form the anion. 

The latter species absorbs more strongly than the corresponding neutral 

molecule and its absorption maximum is blue shifted by some 15 - 30 nm 

(Table 3.1). 

.1 

0.35 .4.\ 

0.304. \ 

I 

' 

• ' I 

0.25 j\ % •u, 

020 '. •) 

. oistN...S' 

0.10 1 

INZ 

pH 9.23 

pH 5.93 

pHt9S 

pH 4.37 

----pH 3.94 

pH 3.67 

pH 3.39 

0.05 

0.00 

250 270 

290 310 330 350 


370 390 

Figure 3.4 Absorption spectra of 3DMABA as a function of pH 

Figure 3.4 shows the ground state absorption properties of 3DMABA as 

a function of pH. It can be clearly seen that as the pH is reduced from a value 

of nine to a pH value of four, there is a two fold reduction in the absorption 

intensity. At pH 4 the distribution of species is approximately 60% neutral 

molecule and 40% anion. The absorption maximum has changed from 313 

nm at high pH to 326 nm. As the pH is reduced to lower values the peak at 

326 run continues to decrease in intensity and the intense peak observed at 

270 nm slowly levels out until at a pH of less than two no further change is 

observed in intensity until an acidity is reached which will produce the bi-

cation. This however requires the very strongly acidic conditions of a 

concentrated acid 45. These data are presented in Table 3.2. Also included in 

this table are the excited state pK values (pK*) which have been either 

determined from fluorescence spectral measurements or calculated from the 

Forster-Weller relationship 24. The latter have been calculated using both 

absorption and fluorescence maxima and both sets of results are presented. 

Ground 

Excited state S1 

state S0 

Compound pKb pKa pKb* plc* 

Experimental Calculated Experimental Calculated 

Abs Fluor Abs Fluor 

3ABA 2.75 4.49 3.03 -6.75 - 4.76 6.75 5.48 

(3.12) (4.74) 

(-7.1) ___ _____________ (7.2) (4.4) 

M3AB 3.52 - 5.00 -6.38 - - - - 

3DMABA 1 3.35 5.23 3.38 -9.44 - 5.24 9.00 5.78 

M3DMAB 3.79 - 4.88 -8.23 - - - - 

Table 3.2 

Ground and excited pK values for both protonation and 

deprotonation. Data in bold are values reported by Jian et al 

For the esters at a pH > 5, the ground state consists of the pure neutral 

species only. On the basis of the calculated pK b*, one would not expect any 

protonation of the excited neutral species so this should lead to a constant 

fluorescence spectrum and intensity. Within the limits of experimental error 

this is observed. The converse is also probably true in that at pHs c 1.5 the 

ground state will be composed of pure cation. Since the cation is nonfluorescent 

the amount of fluorescence observed will depend on the fraction 

of excited cation which loses a proton to form the excited neutral molecule. In 

the absence of protonation of the excited neutral species, it would again be 

expected that the fluorescence intensity would remain constant and this is 

also observed. The ratio of these two constant intensities (i.e the fluorescence 

intensity at a pH c 1.5 against the fluorescence intensity of a pH> 5) gives the 

proportion of excited cation which loses a proton relative to the population 

which decays non-radiatively. This ratio is 1/2.97 and 1/5.39 for M3AB and 

M3DMAB respectively. 

ril

At intermediate pHs i.e. 1.5 c pH .c 5.0, the fluorescence intensity is 

principally determined by the ground state equilibrium between the cation 

and neutral ester. The fluorescence intensity of both M3AB and M3DMAB is 

independent of pH in alkaline and neutral solutions but begins to decrease as 

pH is decreased, from approximately 6.0 until at a pH of approximately 3.0, a 

constant (relatively low) level of fluorescence is observed. The pKb* values 

determined from these spectral changes are 4.27 and 4.04 for M3AB and 

M3DMAB respectively if the above feedback step from the excited cation is 

neglected, and 5.00 and 4.88 respectively if it is included. 

It is noticeable that these values are vastly different from the pK b * 

values calculated via a Forster-Weller cycle which are also included in the 

Table 3.2. The latter suggest that the amino- substituents are much weaker 

bases in the excited state than in the ground state, which would be in accord 

with a first excited singlet state where there is a degree of charge transfer from 

the nitrogen atom to the carbonyl functional. Over the pH range used here 

the excited cation would be expected to loose a proton to form the excited 

neutral ester and thus the observed fluorescence intensity should remain 

constant when excited at the isobestic point. That this is not observed 

indicates that the lifetime of the excited cation is extremely short such that 

loss of a proton can only partially compete with the other decay processes. The 

fluorescence properties are therefore largely decided by the ground state cation 

4-4 neutral equilibrium and the pKb* determined from the fluorescence 

spectra should be approximately the same as the ground state pKb. The 

discrepancy between the pKj, values calculated from the absorption and 

fluorescence spectra is not insignificant: 3.52 versus 5.00 for M3AB and 3.79 

versus 4.88 for M3DMAB. The reason for this is not yet known. 

On decreasing the pH for the two acids, it is found that the fluorescence 

intensity is reduced in a similar manner for that of the two benzoates (Figure 

3.5). However, the situation is more complex here because the two acids can 

62

also deprotonate in addition to the protonation observed for their 

corresponding esters. As may be seen from Table 3.2 the two protic equilibria 

overlap in the ground state with pK a and pKb values which differ by (just) 

less than two pH units. Application of the Forster-Weller cycle to the excited 

state equilibria once again suggests that he two acids, like the two esters, are 

much weaker bases in the excited state, but that their acidity is only slightly 

less than in the ground state. The comments made previously with respect to 

the excited cation therefore also apply here i.e. that it should loose a proton to 

form the excited neutral species if this process is able to complete with the 

other excited state decay processes open to the cation. The consequence of the 

weakening of acidity in the excited state should bias the excited state 

distribution in favour of the excited neutral species at the expense of the 

excited anion. However, over a pH range of relevance / the [H+] available is 

insufficient to protonate the excited anion within its lifetime and it is no 

surprise to find that the pKa* values for 3ABA and 3DMABA determined 

from variations in fluorescence properties as. a function of pH (Table 3.2) are 

very similar to the ground state pK as. The same is also true of the pKb* values 

determined experimentally. There is some discrepancy between the pK a * 

values calculated using the Forster-Weller cycle and the absorption or 

fluorescence spectral maxima, but this is probably just a reflection of the 

several assumptions used in deriving the cycle. 

[*1

9 

pH8.45 

g pH 639 

pH5.6.5 

f. 

6. - - - - - 

pH 536 

ñ 

U pH6 

2 

305 355 405 455 505 

Wavelength (nnt) 

555 

Figure 3.5 fluorescence spectra of 3DMABA as a function of pH 

Along with Jian et alA 5, we have been unable to observe any evidence 

for zwitterionic species in 3ABA or 3DMABA. Our ground state absorption 

data as a function of solvent and our data collected for the neutral species in 

water are in good agreement with those reported by the latter authors. There 

are two obvious reasons for the apparent disagreement with the prediction of 

Kuhn 43 and Tramer42. Firstly, the concentrated solutions used in the initial 

studies 42 ' 43 will favour dimer formation, because of the very low extinction 

coefficient for the zwitterion (see introduction). Secondly, the complex 

protonation / deprotonation properties of the acids make it impossible to 

obtain a solution of just the pure neutral or pure zwitterion species. 

Whenever the neutral acid is present, some proportion is also protonated and 

I or deprotonated to complicate observation of any zwitterion. 

Estimates of the pKa* and pKb* values for the excited state equlibria 

from the fluorescence emission and absorption spectra using the Forster 

equation24 vary considerably, especially for the 3DMABA. The differences 

between the pKa*.values calculated from the absorption and fluorescence 

ErA'

spectra can be attributed to the differences in energy resulting from solvent 

reorientation in the ground and excited state (Figure 3.6). 

Excited molecule 

in ground state 

solvent environment 

Excited molecules 

state 

solvent environment 

Solvent 

adjust to accommodate 

excited molecules 

Energy 

Absorption 

TFluorescence 

Ground siate molecule 

in ground state solvent 

environment 

Solvent molecules 

adjust to minimise 

energy Ground state molecule SE 

in excited state solvent 

environment 

Figure 3.6 Energy diagram representing change in solvent orientation 

around ground and excited state molecules 

The energy of a solute / solvent system is determined to a large extent 

by electrostatic forces and so the interaction is greatest when a polar solute is 

used in a polar solvent. This leads to the greatest difference between the most 

stable and least stable configurations. When a molecule undergoes electronic 

excitation there is usually a significant change in its polarity. Consequently, 

the distribution of solvent molecules which gives the most stable 

configuration for the ground state does not necessarily minimise the energy 

for the same molecule in the excited state. The absorption process is virtually 

instantanepus compared with molecular motion and the solvent molecules 

subsequently adjust to achieve the most stable configuration for the excited 

65

state with a slight reduction in the energy. It is from this state that the 

molecule returns to the ground state where again the energy of the system is 

above the minimum because the distribution of the solvent molecules 

corresponds to the excited state situation. Subsequent adjustment then 

reduces the energy to the value from which the excitation originally occurred 

and the ground state environment is restored. Consequently the emitted 

photon is of lower energy than that of the corresponding exciting photon and 

therefore the emission wavelength appears at longer wavelengths than that 

of the absorption spectrum. 

The differences between pK a and pKa* and pKb and pKb* are worthy of 

comment. The large decrease in pKb upon excitation is understandable given 

that the excited state involves charge transfer from the amino nitrogen to the 

carbonyl function. The electron density on the nitrogen is therefore reduced 

and it will be less basic. One might anticipate that the acid function would be 

stronger in the excited state 81 i.e. pKa* < pKa, but the reverse appears to be the 

case. Once again the charge transfer from the nitrogen atom increases the 

electron density on the carbonyl carbon and thereby reduces the acidity of the 

acidic OH. 

3.3 fluorescence emission properties as a function of solvent. 

The fluorescence spectra for these compounds are generally 

structureless and red shifted from the wavelength of maximum absorption by 

approximately 4 - 5,000 cm -1. Unlike the ground state absorption spectra, the 

fluorescence spectra show a distinct trend with solvent polarity and hydrogen 

bonding ability. In strong hydrogen bonding or polar solvents, the 

fluorescence emission maximum is red shifted. In such solvents, 3ABA and 

M3AB emit in the region of 400 nm (Table 3.3), whilst the dimethylamino 

derivatives emit some 30 nm to the red. All four compounds fluoresce quite 

za

strongly (4 ~! 0.1) with only moderate variation from compound to 

compound and from solvent to solvent. Examples of the spectra are given in 

Figures 3.7 and 3.8. 

Solvent 

3ABA M3AB 3DMABA M3DMAB 

Quantity 

Species 

Acetonitrile Fluor Xrnax /nm 400 403 438 434 

0.27 0.35 0.25 0.25 

Of 

BuCN / iBuCN Huor, 390 392 425 420 

Of 

0.28 0.31 0.24 0.23 

* * * 

PhosXmax/nm * 

BCI 'jP Fluorkmax/rim 380 377 407 406 

0.33 0.35 0.33 0.36 

Of 

PhosXmax/nm 485 495 * 521 

Ethanol FluorXmax/nni 431 435 436 459 

0.24 0.26 0.19 0.16 

Of 

Hexanes FluorXmax/nm 370 354 378 378 

0.16 0.10 0.21 0.20 

Of 

MCH / iP FluorXmax/nm 380 362 395 388 

0.29 0.12 0.28 0.26 

Of 

PhosXmax/nm 425 509 * * 

Water Fluorlmax/nm 407 414 438 435 

0.40 0.32 0.26 0.26 

Of 

Neutral FluorXrnax/nm 407 355 440 435 

Anion Fluor kmax /nrn 415 - 

0.42 - 

Of 

* No measurable phosphorescence 

0.31 0.41 0.21 0.30 

Table 3.3 Excited state emission characteristics of 3-aminobenzoic and 3- 

N,N-dimethylaminobenzoic acids and their methyl esters as a 

function of solvent and pH 

435 

0.23 

- 

- 

The only reported emission study of any of the meta isomers is by Jian 

et al. 45 (Table 3.3a), who measured the emission spectra of 3ABA as a 

function of solvent and pH, but did not measure the quantum yields. In the 

solvents acetonitrile, ethanol and methylcyclohexane, the results are in 

67

agreement with those obtained in the present work within the limits of 

experimental error. 

Solvent (Fluorescence maxima in nm) 

Cyclohexane MCH Acetonitrile Ethanol 10-3 M HCI 10.2 M KOH 

367 385 396 428 400 1 403 

Table 3.3a fluorescence maxima of 3ABA as a function of solvent and pH, 

data reported by Jian et al .45 

There is a discrepancy with the anion and neutral species; as Jian et al. 

observed that the anionic species of 3ABA has a wavelength of maximum 

emission of 403 nm against our value of 415 nm. However the ground state 

pKs are in close agreement with those reported by Jian et al.45 (see table 3.2). 

9.00 

8.00 

6.00 . 

r 

I 

/ 

3ABA 

M3AB 

3DMABA 

0) 

U 

5.00. 

'I 

- -M3DMAB 

0) 

3.00 .. 

2.00 / ' 

1.00 . . . 

if 

N 

330 380 430 480 


530 580 

Figure 3.7 Fluorescence emission spectra of the four meta compounds in 

acetonitrile 

It can be seen from Figures 3.7 and 3.8 that the difference in 

fluorescence emission maxima between the four compounds is very much 

dependent upon solvent, and that the methyl isomers behave in a very 

similar manner to each other, as do the two dimethylamino isomers. The 

M.

effect of esterification of the acid upon the fluorescence is minimal, but 

methylation of the nitrogen causes a dramatic change. 

2 

a, 

C - 

C 

a, 

S. 

0 

300 350 400 450 500 550 


Figure 3.8 Fluorescence emission spectra of the four meta compounds in 

methylcyclohexane I isopentane (4:1) 

3.4 fluorescence lifetime data as a function of solvent and pH 

Fluorescence decay profiles were measured for all four compounds in 

all of the seven solvents and solvent mixtures. It was found that the decay 

profiles varied in complexity with compound, solvent and pH. These results 

are collated in Table 3.4. The vast majority of the measured decays are clean 

single exponentials, with the exception of 3ABA and 3DMABA in the nonpolar 

solvents, where double exponential decay profiles are found. Lifetime 

values are in the region of 10 ns, rising to between 15 - 20 ns upon degassing. 

However the value observed for M3AB in hexane of 1.85 ns is unexpectedly 

low in comparison to that observed in a methylcyclohexane / isopentane (4:1) 

mixture.

CMD Solvent V 

ti(ns) a1 

T2(') 

a2 'r. 

me an 

3ABA Acetonitrile 1.25 14.21 

BuCN/IBuCN 1.11 18.36 

BC1 / iF 1.02 16.87 

Ethanol 1.51 14.02 

Hexane 1.31 1.75 0.7061 4.35 0.2939 2.51 

MCH I iF 1.09 3.72 0.7973 16.66 0.2027 6.34 

Water 1.67 21.59 

Anion 1.12 23.78 

____ Neutral 1.12 23.07 

M3AB Acetonitrile 1.08 9.97 

BuCN / IBuCN 0.98 9.11 

BCI / iF 1.13 9.71 

Ethanol 1.22 13.76 

Hexane 1.35 1.85 

MCH / iF 1.07 9.47 

_ Water 1.29 23.82 

3DMABA Acetonitrile 1.19 13.13 

BuCN / IBuCN 1.02 19.19 

BC1 / i-P 1.05 19.42 

Ethanol 1.35 14.07 

Hexane 0.88 2.84 0.2046 6.44 0.7954 5.70 

MCH / iF 1.16 5.40 0.2446 9.49 0.7554 8.49 

Water 1.12 22.20 

Anion 1.11 20.31 

Neutral 1.10 11.88 

M3DMAB Acetonitrile 1.08 13.22 

BuCN / iBuCN 1.83 14.54 

BCI / iP 1.08 9.40 

Ethanol 1.14 14.29 

Hexane 1.40 4.95 

MCH / iF 1.04 7.66 

Water 11.27 22.45 

Table 3.4 

Fluorescence decay profiles as a function of solvent and pH for 

the four ineta compounds in all solvents at room temperature 

The non-polar solvents clearly provide a less than ideal solvent 

environment for the two acids as there are not only solubility problems but 

also these are the only cases where a single exponential decay is not observed 

for these compounds. Aggregation is well known in these solvents 67 and this 

may explain not only the double exponential fits required for 3ABA and 

3DMABA in hexane and methylcyclohexane I isopentane, but also the 

generally low lifetimes for M3AB and M3DMAB in these solvents. 

70

"Water, a singularly effective ionising solvent on account of its high 

dielectric constant and its ion-solvating ability, is able to stabilise the anionic 

state of the aromatic aminobenzoic acids through hydrogen bonding to the 

nitrogen groups as well as possibly through the carboxylic group" 82. In the 

light of this statement it is perhaps surprising that the decay profiles observed 

in the polar and hydrogen bonding solvents are not also more complex. It is 

therefore probable that there is one dominant solute / solvent arrangement 

and this is the one which is observed in the fluorescence. On the basis of the 

pKa values calculated earlier and the lifetime data presented in Table 3.4, it is 

clear that both 3ABA and 3DMABA exist as the anion in unbuffered water. 

The absence of any complexity in the aqueous decay profiles backs up the 

earlier conclusions about the pKa* values and the ability of the excited anion 

to gain a proton within its lifetime. 

The lifetime and quantum yield data has been combined to calculate 

rate constants for the radiative (fluorescence K1) and non-radiative (K nr) decay 

processes in the four molecules as a function of solvent (Table 3.5). The 

values observed for both rate constants are quite remarkably similar for all 

the solvents employed here if the K nr values in hexane and 

methylcyclohexane / isopentane (4:1) are omitted. This perhaps reflects the 

relative lack of interaction between the two substituents on the benzene ring 

when they are meta- orientated. 

71

— — 

— 

I- 

C) — l LI) N en 'oa' 

c4r-enen 

'r in 

0 N0' 

— LI) C'I ('I l Q en'o 

rCLtS2eliO 

x 

_ - - 

cq cn 

1! 

o C)'C 'ON 

r-J z 

a) 

-'-a 

Ct 

I- ----- 

C) 

01C\LI)'0C'IN 

Ct NCflflNri 

.Ii,-LI)rLLi 

Lci 

ci) 

• 

CCI — 

— 

.Ct 

CtC5 

I U) 

— 

a) 

C-) 

C) 

0-. 

U) 

I-i 

C) 

C') 

U) 

C) 

C, 

C 

I-. 

C) 

'C 

I 

Ct 

I-i 

C % ON 0' N l 

ca 

Qreii et'6 4 vi ceá 

Ct 

t 

cqZ 

c) 

• eflC'1 ON LO'0 00 O 

a) M aN qcr LO t40' Lnen 

> enenNentI) 

Ii 

o 

0 

C) 

hi 

.• O ,t —('4 O. 0\N 

c Q'ri M LO ON 00 ' 

Lei m'o V-itt) 

a., 

1.) 

o -, 

en 

en 

• 

.< 

_ 

I-) 

en 

I- — C 

-C 

Cc 

C) 

U) L. 

0_C 

U -s 

•d-aCt 

ci 

H 

—I 

Ct 

-a 

Ct 

I- 

C) 

'4- 

C 

C') 

C 

CC 

C-, 

0 C) 

'a. E c; 

C-C 

hi 

Ct —() 

-C 

tt_C a) 

Ct 

-:1 

Ct•_ 

r'a 

0 

c'S 

•. 

U) 

•2 

a 

V–li 

U 2 gr 

hi 

Ct 

CC 

hi 

a) 

-c 

72

3.5 fluorescence emission properties as a function of temperature 

For variable temperature measurements only the three binary solvent 

mixtures were used, and these were chosen for their ability to form good 

glasses at low temperature. The results are presented in Table 3.7. Although 

no new information was expected to be gained from this experiment, it was 

undertaken to enable a comparison to be made with the results obtained for 

the para- isomers where temperature effects on the normal and anomalous 

fluorescence were investigated. 

Cmd Solvent Quantumyield as a function of temperature 

293K 230K 193K 153K 133K 113K 

3ABA BuCN / iBuCN 0.28 0.50 0.55 0.58 0.58 0.58 

BC! / i-P 0.33 0.42 0.30 0.54 0.61 0.65 

MCH / iP 0.29 0.58 0.64 0.22 0.64 0.64 

M3AB BuCN / iBuCN 0.31 0.47 0.56 0.60 0.62 0.62 

BCI / iF 0.35 0.50 0.71 0.87 0.99 1.00 

MCH / iF 0.12 0.16 0.26 0.31 0.27 0.47 

3DMABA BuCN / IBuCN 0.24 0.41 0.43 0.42 0.39 0.39 

BCI / i-P 0.33 0.38 0.24 0.25 0.35 0.38 

MCH / iP 0.28 0.43 0.38 0.34 0.20 0.19 

M3DMAB BuCN I 1BuCN 0.23 0.47 0.56 0.53 0.54 0.59 

BCI / iF 0.36 0.69 0.82 0.84 0.86 0.86 

MCH / iF 0.26 0.31 0.39 0.43 0.43 0.44 

Table 3.7 

Excited state emission characteristics of the four meta 

compounds as a function of solvent and temperature 

In general, the fluorescence spectra of all the compounds broaden with 

lowering of temperature with an increase in the fluorescence quantum yield, 

which in some cases reaches nearly unity. These data have to be viewed with 

some caution however as there has been no attempt to correct for absorption 

changes as a function of temperature. This increase in quantum yie!d imp!ies 

the presence of an activation barrier of some sort. Since fluorescence rate 

constants are usually temperature independent, the energy barrier is 

presumably associated with a competing radiationless process. It appears that 

this process is intersystem crossing 83 from Si to a higher triplet T n which is 

73

approximately degenerate with S1. The effect of temperature is to increase the 

population of higher vibrational and rotational sub-levels of S1 from which 

faster rates of intersystem crossing may occur. Hence as the temperature rises 

the rate of intersystem crossing increases and consequently a smaller 

proportion of s1 molecules have an opportunity to fluoresce, and the 

quantum yield of fluorescence will decrease accordingly. The wavelength of 

maximum emission also red shifts. This shift is very dependent on solvent 

and substituent group. In chiorobutane and butyronitrile, the shift is of the 

order of 5 rim, whilst in the inert solvent methylcyclohexane the shift is as 

much as 15 nm. Unfortunately the accuracy of the wavelength measurements 

is +1- 5 nm which may somewhat disguise the real wavelength of maximum 

emission (see for example Figure 3.9). 

250000 

293K 

> 200000 

-a 

C 

4, 

U 

150000 

2 

0 

315 335 355 375 395 415 435 455 475 495 515 


Figure 3.9 fluorescence spectra of M3AB in chiorobutane I isopentane (9:1) 

as a function of temperature 

1 

In the non-polar methylcyclohexane, 3ABA and 3DMABA show the 

appearance of a secondary peak at 400 rim which is not observed in the other 

solvents used. This second peak is observed at temperatures as high as 230K 

(Figure 3.10 and 3.11) and may indicate the onset of aggregation. 

n 

74

— 

7000.00 

6000.00 

• 5000.00 

C 

0? 

.E 4000.00 

C, 

o 

I 

+ 

i 

+ 

30000 4 . 

U, 

(I 

I- al/I 

2000.00 

1 000.00 

Nt" 

293K 

-----230K 

193K 

------153 K 

133K 

-------113K 

77K 

0.00 

280.00 330.00 380.00 430.00 

Wavelength (mu) 

480.00 530.00 

Figure 3.10 Fluorescence spectra of 3ABA in methylcyclohexane / 

isopentane (4:1) as a function of temperature. 

700000 

293K 

230K 

C) 

400000 

193K 

153K 

—- 133K 

113K 

g200000 

77K 

W. 

100000 

0* 

330 

380 430 480 530 

Wavelength (urn) 

Figure 3.11 Fluorescence spectra of 3DMABA in methylcyclohexane / 


The largest change in wavelength of maximum emission is observed 

for M3AB in methylcyclohexane. Although phosphorescence might be 

expected at 77K, none is seen. This feature fits well with the observed increase 

75

in o f as the temperature is decreased, particularly if the interpretation 

advanced a little earlier regarding decreased inter-system crossing is the 

correct one. 

3.6 . Summary of conclusions 

From absorption spectroscopic studies the pK a and pKb values have 

been found for the four compounds, anlong with their ground state 

absorption properties as a function of solvent. Although there is evidence 

from the spectra of 3ABA and 3DMABA that two species may be present in 

hexane and methylcyclohexane / isopentane (4:1). This is a possible indication 

of ground state dimerisation but concentration measurements have not been 

undertaken to check this. If this is the case, then some evidence for this might 

be expected in the steady state room temperature emission spectra. The 

fluorescence emission spectra of 3ABA and 3DMABA at lower temperatures 

do however appear to show the appearance of another peak at 400 nm. The 

lowering of temperature may well favour formation of dimers or higher 

polymers 69. There are several other excited state possibilities such as excimer 

or exciplex formation which could account for the new peak, but there is no 

emission evidence for these except at lower temperatures. Further data from a 

study of the concentration dependence of these low temperature spectra (both 

absorption and fluorescence) coupled with low temperature fluorescence 

decay measurements will help to resolve the origin of the new bands. 

76

Chapter 4

Chapter 4 Photophysical properties of 2-aminobenzoic acid, and 2-N,N-dimethylaminobenzoic 

acid and their methyl esters 


Compared to the meta- isomers discussed in chapter 3, the absorption 

properties of the ortho- isomers are much more complex. However they have 

been studied in greater detail by several authors notably Kuhn and Geider 43, 

Tramer 42' 44, Mataga and Ottolenghi 40, Doub and Vandenbelt 38, Leggate and 

Dunn 41 and Jian et al. 45 . The ortho- and para- isomeric amino and N,Ndimethylaminobenzoic 

acids and their methyl esters differ from their metacounterparts 

in that the lone pair of electrons on the nitrogen are able to 

interact with the carbonyl group via the aromatic ring system and 

consequently the excited states can involve some form of resonance 

interaction (stabilisation). 

As well as the more usual inter-molecular hydrogen bonds, the orthoisomers 

can also form intra-molecular hydrogen bonds. The close proximity 

of the two substituent groups leads to steric hindrance, so that each group will 

have an influence on the other and so may be twisted out of the plane of the 

aromatic it system. This may well affect the amount of conjugation between 

the nitrogen lone pair and the carbonyl function and consequently affect the 

amount of charge transfer between the two substituents. 

In all of the pure solvents and binary mixtures used in this study, 2- 

aminobenzoic acid (2ABA), methyl 2-aminobenzoate (M2AB), 2-N,Ndimethylaminobenzoic 

acid (2DMABA) and methyl 2-N,N-dimethylaminobenzoate 

(M2DMAB), generally show structureless absorption spectra with a 

wavelength of maximum absorption ranging from 280 to 345 nm, depending 

upon the polarity of the solvent, its hydrogen bonding ability and on the 

substituents (Figures 4.1 and 4.2 and Table 4.1). 

ViA

AVA 

0.6 

03 

L 

0 

0.4 

I 

o 

I • 

20.31- 

< I 

0.24- 

'- 

01 

of 

Pt' 

I 

•7•- N. 

Acetonitrile 

-----Ethanol 

Hexane 

Water 

235 255 275 295 315 335 355 375 395 


Figure 4.1 Absorption spectrum of 2ABA in all four pure solvents 

Solvent 

Measured 2ABA M2AB 2DMABA M2DMAB 

Species 

quantity 

Acetonitrile Am /run 336 337 281 336 

El dm3 moF 1 cm4 4,700 5,200 800 2,600 

BuCN / IBuCN 338 337 283 339 

8/ dm3 mol4 car1 4,100 6,000 1,200 3,200 

BCI / iF Xmax 337 335 289 335 

________________ 8/ dm3 mol-1 y 1 5,000 5,300 1,000 3,300 

Ethanol A.max /p'n 335 339 271 339 

6/ dm3 mo!4 y1 4,400 4,900 800 2,700 

Hexane kmax /nm 338 333 281 336 

8/ dm3mol'cnr' 4,900 5,000 800 3,100 

MCH / i1' Xmax /nrn 336 333 287 336 

________________ 8/ dm3 mol-1 3,500 6,600 1.000 2,800 

Water Xmax /rm 314 327 270 327 

1,900 3,600 700 1,400 

Neutral c/ dm3 mo!-1 car1 326 326 300 328 

Xmax /nrrt 2,100 3,900 700 1,400 

Anion Xmax 310 - 260 - 

8/ dm3 mo!4 cm-1 3,000 900 - 

Cation I Xmax /nm 272 272 272 272 

2/ dm3 mo!4 -1 940 840 500 960 

in

Table 4.1 Primary ground state absorption characteristics of 2- 

aminobenzoic, 2-N,N-dimethylaminobenzoic acid and their 

methyl esters as a function of pH and solvent 

'Ii 

DIIQ 

0.07 AcetonUrile 

0.06 -----Ethanol 

C 0.05 Is 

/ Hexane 

1 

-- 

Water 

'C 

0.04 

0.03 

V. 

0.02 S. 

I-. . 

0.01 

------. 

-. -. -- 

0 I I I I I I 

250 260 270 280 290 300 310 320 330 340 350 


Figure 4.2 Ground state absorption spectra of 2DMABA in all four pure 

solvents 

Compared to the corresponding meta isomers, 2ABA and M2AB 

absorption bands are red shifted by some 15 - 30 nm and exhibit enhanced 

extinction coefficients usually by a factor of 2 - 3. However, the absorption 

bands in the ortho- isomers are not as intense as those for the para-isomers, 

probably reflecting the steric crowding in the former. The 2-N,Ndimethylamino 

compounds are clearly very different to both their rneta- and 

para- counterparts. The usual red shift of some 20 nm when the amino group 

is methylated is absent in the orthocompounds and, in the case of 2DMABA, 

is replaced by a 50 nm blue shift. Extinction coefficients in 2DMABA and 

M2DMAB are much lower than might be expected, reflecting not only steric 

crowding but also apparently intra-molecular hydrogen bonding in the case of 

2DMABA. 

79

In all of the solvents and binary mixtures studied, 2DMABA exhibits a 

more structured spectrum, with two peaks in the 265 - 280 nm region, 

generally split by 10 nm (see for example Figure 4.2). In water and ethanol 

another peak is observed around 300 and 315 nm respectively. In hexane and 

acetonitrile it is also possible to identify another peak around 350 nm (c

secondary absorption band in the 220-250 nm range, which varies 

considerably in position and strength, depending on compound and solvent. 

The absorption properties of M2AB are very similar to those of its parent add 

(Table 4.1). 2ABA and M2AB have fairly strong extinction coefficients of 

around 4-5,000 dm 3 mol-' cm -', whilst 2DMABA and M2DMAB have much 

smaller extinction coefficients, of the order of 2-3,000 dm 3 mol-' cm -1 for the 

benzoate and 700 - 1,200 dm 3 mol-1 cm -1 for the acid. This difference between 

the amino and N,N-dimethylamino substituent group can be explained in 

terms of the amount of steric hindrance between the substituent groups in 

each compound. Depending on the orientation of the carboxylic acid 

substituent, esterification will have little or no effect, but replacement of the 

hydrogen atoms on the nitrogen by methyl groups will make the nitrogen 

lone pair more available for bonding by increasing the basicity of the amino 

group, but also will potentially result in the N,N-dimethylamino group being 

twisted out of plane, thus decreasing the extent of conjugation and 

consequently the amount of charge transfer 17-20. For 2DMABA there is the 

added complication of the hydroxy group which creates a potential for two 

forms of hydrogen bonding (see Figure 44). 

10 o 

H 

0% /°H 0% /°H 

c 

I 

H 

Me 

C/ "H or \ H but only Me for ZDMABA 

Figure 4.4 Possible hydrogen-bonded structures for the ortho acids in inert 

solvents 

In terms of the variation of the wavelength of maximum absorption as 

a function of solvent, 2DMABA is a special case, being blue shifted some 40 - 

50 nm from 2ABA, M2AB and M2DMAB, which all show similar 

wavelengths of maximum absorption in a given solvent. M2DMAB has 

EjI

extinction coefficients approximately one half of those observed for 2ABA 

and M2AB. This as previously described can be attributed to steric hindrance 

between the carboxyl and N,N-dimethylamino groups which twists one or 

both of them out of plane. Consequently a lower extinction coefficient would 

be expected as a result of loss of conjugation. 2DMABA however appears to 

exhibit anomalous properties having very low extinction coefficients of the 

order of 1,000 dm3 mol-' cm-1 and a structured absorption spectrum. 2DMABA 

is behaving in a similar manner to benzene which has a primary absorption 

band at 230 nm 16 and a secondary structured absorption band around 280 nm 

(C = 1,045 dm3 mol- 1 cm-1). In 1969, Tramer42,44 investigated the absorption 

properties of 2-aminobenzoic acid and two of its derivatives namely, N- 

methyl 2-aminobenzoic add and N,N-dimethylaminobenzoic acid by infrared 

and ultraviolet spectroscopy 42 and later by fluorescence emission 

spectroscopy 44. He showed that in the molecular form (Figure 1.13) none of 

the three compounds form an intra-molecular hydrogen bond between the 

carboxyl and amino groups. In the quasi-zwitterionic form (Figure 1.13) a 

strong intra-molecular hydrogen bond is formed such that it changes the 

conformation of the amino group and removes the conjugation between the 

nitrogen lone pair and the it-electron system. 

Literature data on 2ABA and 2DMABA absorption properties are 

reproduced in Table 4.1a. As can be seen by comparing the literature values 

with those in Table 4.1, the results are in excellent agreement within 

experimental error. However, Tramer 44 has reported a peak in the region of 

345 nm with a quoted extinction coefficient of around 85 dm 3'mol- 1 cm 1 in 

acetonitrile. We were unable to observe this band, but this is not surprising 

given its very weak intensity. At the concentrations which would be 

necessary to observe this band (~ 1 mmol dm 3), the probability of aggregation 

and dimer formation must be high. 

IM

Solvent 

Measured 

quantity 

Compound 

2ABA 

2DMABA 

17/ 2nd Band 

Acetonitrile )Ur.ax (rim) 334 346 / 280 

£ moF 1 cm 1 dm - 857 11000 " 

Cydohexane ?tjnax (nm) 349 / 281" 

8 mo1 1 cm 1 dm3 14 / 950 

Ethanol )m,< (rim) 336 

2 mo!-1 cm-1 dr& - 

Water Lnax flI 

327 322 / 270 IT 

£ mo1 1 cm4 dm3 1,900 40 15 / 1,000 

Table 4.1a Absorption maxima of 2ABA and 2DMABA as a function of 

solvent and pH, as reported by various authors 

4.2 Ground state absorption and fluorescence emission properties as a 

function of pH. 

Measurement of absorption spectra in water as a function of pH for the 

four orthocompounds allows the calculation of pKb and pICa values for the 

two acids. Sample spectra are shown in Figure 4.5 for 2ABA. 

0.8 p1-I 7.25 

0.7 — —pH 6.00 

0.6 pH 5.00 

0.5 — - — - — - pH400 

250 270 290 310 330 350 370 390 

Wavelength (rim) 

410 

Figure 4.5 Absorption spectra of 2ABA as a function of pH 

As can be seen from Figure 4.5, as the pH decreases from around 6.0 the 

absorption begins to diminish and becomes red shifted, until at pH 4 the 

XV

absorption maximises at approximately 326 nm and another new band begins 

to appear at 272 nm, which is associated with the cation. At this pH it would 

be expected to have 98.5% of the neutral species and 1.5% of the anion. On 

lowering the pH to a value of

ut it is clear that the value quoted by Jian et aL 45 for the pKb of 2.55 is clearly 

the "odd man out" There is apparently no report in the literature regarding 

the pKb values for M2AB and M2DMAB, but the values given in Table 4.2 for 

these two quantities are similar to the pKb values for the parent acids. 

Comparison of our pK data for 2DMABA with the works of Tramer 44 

and Kuhn43 is difficult because of our inability to distinguish separate 

absorption features for the zwitterion and neutral species which they believe 

to exist. However, Tramer's results 44 are shown in Table 4.2a 

equilibrium pK pK* 

pKBcPKBC* 

- 1.5 

PKCM 4 -11.5 

PKcz 1.4 2.7 

pKMA 6 10.0 

pKm 8.58 0.3 

pK1 -2.6 9.8 

Table 4.2a pK values obtained by Tramer 42' 44 calculated from Figure 1.13 

Tramer 44 states that the zwitterionic form of 2DMABA absorbs at 270 

run and that the neutral species absorbs at 322 nm. This argument is based on 

the fact that the ground state absorption spectrum of 2DMABA is almost 

identical with that of benzoic acid, which is considered to be a strong 

argument in favour of a zwitterionic structure, because the influence of the 

quaternary ammonium group on the spectrum of benzene derivatives is 

insignificant, compared to that of an alkyl group 44. Tramer appears to have 

missed the fact that if the lone pair on the nitrogen were in fact to be removed 

from interaction with the plane of the aromatic ring system then a similar 

effect would be observed. This could easily happen if there was either an 

intra- or inter-molecular hydrogen bond formed or as a result of steric 

interactions. For all the compounds studied in water as a function of pH the 

peak observed in the 265 nm region decreases along with the 330 nm peak as 

the pH is lowered. Both 2ABA and M2AB show an isosbestic point at around 

E:l

290 nm whilst the N,N-dimethylamino derivatives do not. Since neither 

M2AB or M2DMAB can form a zwitterion, the 265 nm peak must be due to 

another transition within the molecule to form the second excited singlet 

state. This may invalidate the conclusions of Tramer 44 and other authors 

who have used his results. Our spectra of 2DMABA closely resemble those 

reported by Tramer 44, although he does not appear to have studied the ester. 

It appears more likely that the two amino substituted compounds 2ABA and 

M2AB, if they are hydrogen bonded at all, form inter-molecular hydrogen 

bonds involving the carbonyl group and the N-H functions. The absorption 

spectra of 2ABA and M2AB suggest that there is good conjugation between 

the NH2 and COOR groups and that there are negligible differences between 

the acid (R=H) and the methyl ester (R=CH3). The latter would probably not 

be the case if the acid OH in 2ABA were hydrogen bonded to the nitrogen 

lone pair. 

The situation is not as straightforward for the N,N-dimethylamino 

compounds. The addition of the methyl groups to the nitrogen will enhance 

its basic properties and consequently the lone pair of electrons on the nitrogen 

will be more available for either intra- or inter-molecular hydrogen bonding. 

With M2DMAB this hydrogen bond must be with the solvent because the 

carboxylate ester group has no acidic proton. This could explain why 

M2DMAB shows larger extinction coefficients than those observed for 

2DMABA which may preferentially form an intra-molecular hydrogen bond 

involving the lone pair of electrons on the nitrogen, consequently losing 

conjugation with the aromatic ring system. Another consequence of this 

intra-molecular hydrogen bond is that the pK a for 2DMABA is considerably 

greater than the pK a for 2ABA. The observed increase in pK a of nearly 3.8 pH 

units on di-N,N-methylation is much greater than that seen for the metaisomers 

(0.7 - 0.8 units) and the para- isomers (no change). The presence of the 

intra-molecular hydrogen bond makes the COOH proton considerably more 

ET

difficult to remove from the 2DMABA molecule and a slightly basic pH 

rather than a weakly acidic one is now required. 

Since it is not possible to assign any particular absorption to a 

zwitterion structure, our results are calculated from what we believe to be 

neutral molecular form (m). For the ground state protonation, our pKj, value 

of 2.93 is between Tramer's 44 pK values of 4 and 1.4 for cation to molecule 

pKcm and cation to zwitterion pK cz respectively, whilst our deprotonation 

value of 8.46 corresponds well to Tramer's 44 value calculated for the 

zwitterion to anion equilibrium pK za. Similar comments apply to the excited 

state deprotonation reaction. However our observations are in good 

agreement with those of Jian et al. 45 who were also unable to observe the 

zwitterion. 

V 

C 

3.5 

pH 6.27 

3.. , 

/ pH4.90 

2.5 /1 

2 

1.5 

o i. I, 

0.5 

0 

I 

•' 

- 

I 

'Is I 

/r 

• 

s. 

J 

/ 

St. 

p1-13.97 

--—"pH2.99 

-----"pHl.14 

4 1 

ts — - . ----.--- - 

300 350 400 450 500 


550 

Figure 4.6 Fluorescence spectra of 2ABA as a function of pH 

Figure 4.6 shows typical fluorescence emission spectra for the orthocompounds 

at different pH values. As can be seen from Figure 4.6, on 

lowering the pH the emission becomes red shifted and the intensity is 

reduced. This can be explained in a similar manner as that described for the 

meta compounds in chapter. three.

The fluorescence intensity of the four ortho compounds varies in a 

similar manner to that of the ground state absorption spectra. 2ABA has the 

largest of all the quantum yields (Table 4.3) with the pure anion having a 

quantum yield of around 0.43 compared to the neutral species which has a 

quantum yield of 0.30. Its methyl ester, M2AB shows a quantum yield of 0.13. 

2DMABA has the lowest of all the quantum yields with a value of around 

0.001 in water; this is further strong evidence for the fact that the lone pair of 

electrons on the nitrogen is no longer conjugated to the aromatic it-system. 

M2DMAB has a quantum yield which is greater by a factor of more than 10. 

However, it is noticeable that the quantum yield for the 2DMABA anion is 

also very small. Here the intra-molecular hydrogen bond no longer exists yet 

both the absorption intensity and fluorescence quantum yield for the 

2DMABA anion are still low. The negative charge on the carboxylate anion 

does not appear to reduce these properties for 2ABA. Steric effects may reduce 

the absorption intensity and quantum yield for M2DAB but not to the low 

levels observed for the anion. It is not clear at this juncture as to the reasons 

for these anomalous anion properties. 

4.3 fluorescence and emission properties as a function of solvent. 

Fluorescence emission spectra and quantum yields were measured for 

the four orthocompounds in the solvent range used in this study. 2ABA and 

its methyl ester have high fluorescence quantum yields, ranging from 0.54 in 

acetonitrile to 0.20 in a mixture of hexanes (Table 4.3). However there is a 

dramatic difference between these two compounds and their N,Ndimethylamino 

analogues. As mentioned earlier, there is likely to be steric 

hindrance due to the close proximity of the carboxylic acid / carboxylate ester 

function and N,N-dimethylamino substituent leading to one or both of the 

groups being twisted out of the plane of the aromatic ring system. In the case 

of 2DMABA the lone pair on the nitrogen may also be involved in a strong 

[$3

intra-molecular hydrogen bond. Both effects result in a loss of conjugation 

and consequently a loss of charge transfer efficiency in the system. This is 

supported by the low fluorescence quantum yields observed for M2DMAB, 

but even these are an order of magnitude greater than those found for the 

2DMABA. This again is further evidence that the lone pair of electrons on the 

nitrogen are used to form a strong hydrogen bond. 

Solvent Measured 2ABA M2AB 2DMABA M2DMAB 

Species 

quantity 

Acetonitrile FluorXmax/nm 394 399 420 429 

0.54 0.50 0.002 0.03 

BuCN / IBuCN Puor 390 392 436 412 

Of 

0.42 0.36 0.007 0.05 

PhosXmax/nm 451 453 415 480 

UC1/iP jquor ?.atnj /nan 389 388 425 415 

0.47 0.51 0.004 0.16 

PhosXmax/nm 454 452 480 500 

Ethanol Fluor Xinax /run 406 413 401 425 

0.32 0.42 0.002 0.04 

Hexane's FluorXrrtax/nm 380 377 400 395 

Of 0.20 0.30 0.008 0.05 

MCH / iF Fluor " /nm 386 378 400 398 

0.27 0.37 0.003 0.07 

PhosXmax/nm 453 451 478 474 

Water FluorXmax/nm 405 427 430 438 

0.38 0.10 0.003 0.03 

Neutral Fluor Xmax /nm 415 421 423 427 

0.30 0.13 0.001 0.04 

Anion Fluor 400 - 416 - 

0.43 - 0.001 - 

Table 4.3 Excited state emission characteristics of the 2-aminobenzoic acid, 

2-N,N-dimethylaminobenzoic acid and their methyl esters as a 


In general the quantum yields of all the orthocompounds increase 

with solvent polarity and hydrogen bonding ability. There is a similar trend 

for the wavelength of maximum emission in that in the non-polar solvents 

the wavelength of maximum emission is blue shifted in comparison to the

more polar solutions. However in all cases, the longest fluorescence 

wavelengths are seen in the hydrogen bonding solvents ethanol and water. 

Furthermore the fluorescence from the N,N-dimethylamino compound is 

red shifted some 20 nm from their corresponding amino compounds. There 

is one exception to this and that is 2DMABA in butyronitrile / isobutyronitrile 

(9:1) where the wavelength of maximum emission is red shifted 

a further 15 nm (Figure 4.7). Typical spectra of all the compounds in hexane 

are shown in Figure 4.8. 

4500 

a. • 

2ABA 

3500 

3000 

2500 

'C 

) 

I 

I 

I 

I 

2DMABA 

-- M2DMAB 

2000 

I 

I 

C 

1500 

1000 

04- 

300 350 400 450 500 550 600 


Figure 4.7 Fluorescence emission spectra of 2ABA, M2AB, 2DMABA and 

M2DMAB in butyronitrile I isobutyronitrile (9:1)

160000 2ABA 

140000 a-' 

.C' \\ 

------M2AB 

120000. 'I 

80000 . 

01 I 

II 

I 

Is 

I: 

I. 

A 

, 

/ /•. \ 

40000 us S 

20000- 

- 2DMABA 

-----M2DMAB 

51 - 

0:" I I 

310 360 410 460 510 


Figure 4.8 fluorescence emission spectra of 2ABA, M2AB, 2DMABA and 

M2DMAB in hexane 

I . 

There have been a number of reports of the fluorescence properties of 

the four ortho compounds. These are summarised in Table 4.3a. Our results 

are in excellent agreement with those found in the literature. 

Solvent 2ABA 

Species 

?unaxnrn 

Acetonitrile 395 -' 

400 40 

Ethanol 386 

413 40 

Anion 394 45 

39043 

2DMABA 

naxnm 

422 

410 40 

422 44 

Neutral 42043 4407T 

MCH 389 

387 40 

Water 40043 

408'" 

Table 4.3a fluorescence emission properties which have been previously 

reported in the literature 

91

4.4 fluorescence lifetime data as a function of solvent, pH and 

temperature 

Fluorescence decay profiles were measured for the four compounds in 

all seven solvents and binary solvent mixtures. It was found that the decay 

profiles varied in complexity with compound and solvent / pH. The results 

are tabulated in Table 4.4. There have been no other reported lifetime 

measurements on the 2-aminoben.zoic acids. As mentioned in chapter 3, 

water is a singularly effective ionising solvent which is able to stabilise the 

anionic state of the aromatic aminobenzoic acids through hydrogen bonding 

to the nitrogen groups as well as possibly through the carboxylic acid group. 

Thus for 2ABA and M2AB all the decays are mono exponential apart from 

those in water where a sum of two exponentials is required to adequately fit 

the data with values of 2.63 / 5.58 ns for 2ABA and 1.77 / 8.51 ns for M2AB. In 

the case of 2ABA, the neutral molecular species show a dual exponential 

decay with lifetimes of 3.01 and 5.70 ns while the anion shows a single 

exponential decay with a lifetime of 9.18 ns. In view of the range of different 

intra- and inter-molecular hydrogen bonds which could possibly exist for 

these compounds in water , the observation of dual exponential fluorescence 

kinetics is understandable although this was not seen for the meta- isomer. 

Whether the decays are truly dual exponential or actually represent a range of 

fluorescent solute / solvent species which are best fitted by a dual exponential 

is impossible to judge. 

In the other solvents studied for 2ABA, the lifetimes vary from 4.63 ns 

in the non-polar hexane to 7.49 ns in the polar hydrogen bonding solvent 

ethanol. There is no evidence (i.e. multi-component decays) for multiple 

hydrogen bonded species in ethanol. M2AB shows very similar properties 

with the shortest lifetime of 3.77 ns in hexanes and the longest lifetime 

observed in ethanol of 7.99 ns. It was expected that the next simplest case 

would be M2DMAB, but this does not appear to be so as it appears to show

dual exponential decays in all solvents studied. There is a need for further 

lifetime studies because of the conflicting data which have been obtained. The 

most recent study using laser excitation suggests that the decays may actually 

be single exponential. Unfortunately, attempts to obtain complementary data 

using, the synchrotron radiation source at BESSY have proved inconclusive 

because of instrumental artefacts. In a trend similar to that found for 2ABA 

and M2AB, the observed lifetime values are shortest in the non-polar 

solvents hexane and methylcyclohexane / isopentane (4:1) and longest in the 

more polar and hydrogen bonding solvents. It is difficult to envisage how 

there might be two species of M2DMAB which might fluoresce other than 

through different solvation arrangements. In some solvents it is possible that 

the lone pair on the nitrogen is involved in some form of bonding with the 

solvent. This could involve the formation of an exciplex with a C- type bond 

between the solvent and the molecule as proposed by Varma et al. for 4- 

dimethylaminobenzonitrile (DMABN) 50' 6869. This however does not explain 

why dual exponential decays are observed in the non-polar, non-hydrogen 

bonding solvents methylcyclohexane / isopentane (4:1) and hexane unless 

there is perhaps some form of dimer formation. Further data are clearly 

needed. 

2DMABA shows the same complex behaviour, but this is perhaps 

more easily explained. In all but the non-polar, non-hydrogen bonding 

solvents 2DMABA shows dual exponential decays, with very short lifetimes 

of the order of 1 - 1.5 ns and a longer lifetime of the order of 6-9 ns. In hexane 

and methylcyclohexane / isopentane (4:1), the single species observed has a 

lifetime of around 3.5 ns and, because of the inability of the solvent to interact 

with the molecule this is probably due to a molecular species with a strong 

intra-molecular hydrogen bond. However, in the more polar solutions there 

must be an interaction or a series of different interactions between the solvent 

and the molecule. This leads to the possibility of competition between the 

93

intra-molecular hydrogen bonded species and a solvated complex of a specific 

solute/solvent type. The percentage of the shorter lifetime species appears to 

be around 35 % and so is a substantial proportion. 

Compound Solvent 

a1 t2(ns) a2 

Species 

mean 

2ABA Acetonitrile 1.17 6.68 

BuCN / IBuCN 1.33 6.06 

BC! / iF 1.37 5.43 

Ethanol 1.14 7.49 

Hexane 1.21 4.63 1 

MCH / iF 1.37 5.21 

Water 0.86 2.63 0.4047 5.58 0.5953 4.39 

Anion 1.19 9.18 __ 

Neutral 0.97 3.01 0.7398 5.70 0.2602 3.71 

M2AB Acetonitrile 0.99 6.57 

BuCN / iBuCN 1.45 7.54 

BC! / i-F 1.30 4.72 

Ethano! 0.94 7.99 

Hexane 0.87 3.77 

Water 0.64 1.77 0.8982 8.51 0.1018 2.45 

2DMABA Acetonitrile 1.14 0.83 0.8482 7.91 0.1518 3.25 

BuCN / iBuCN 0.99 1.51 0.8637 8.90 0.1362 2.51 

BC1 / iF 1.14 0.81 0.8482 7.91 0.1518 1.89 

Ethanol 1.71 1.73 0.8574 11.08 0.1426 3.06 

Hexane 0.83 4.10 

MCH / iF 1.46 2.90 

Water 1.00 0.93 0.8202 13.51 0.1798 3.19 

Anion 1.90 0.79 0.8585 6.50 0.1415 1.59 

Neutral 1.50 1.42 

M2DMAB Acetonitrile 1.31 5.75 0.6987 11.37 0.3013 7.44 

BuCN / IBuCN 1.90 5.51 0.1737 11.80 0.8263 10.71 

BC! / iF 1.25 5.32 0.8032 12.36 0.1968 6.71 

Ethanol 1.01 6.80 0.2209 13.50 1 0.7791 1 12.01 

Hexane 0.98 2.67 0.2894 4.33 0.7106 

1 

3.85 

MCH / iF 1.70 3.51 0.1465 5.38 0.8535 5.11 

Water 1.51 5.72 0.8952 13.62 0.1048 6.55 

Table 4.4 

Fluorescence decay profiles as a function of solvent pH for the 

ortho compounds 

The lifetime and quantum yie!d data has been combined to calculate 

rate constants for the radiative (Kf) and non-radiative (K nr) decay processes in 

the four molecules as a function of solvent (Table 4.5). For the two amino 

94

substituted compounds K1 does not vary very much with solvent. However 

the value of K I, for 2ABA in hexane appears to rise steeply, this could again 

be due to aggregation. This steep rise is not observed for M2AB. In polar / 

hydrogen-bonding solvents both the N,N-dimethylamino substituted 

compounds show low K1 values, with 2DMABA being much less than 

M2DMAB, while the Knr values are similar to those found for the non 

methylated compounds. 

ii 

OY

a) * * * 

Lf) e"t 

'N o-tn 

'4 

Ct¼R,t LI). q o• . 

OC'4 tflr.4 

06- 'enomo- 

V 

m r-tf 

* * 

NC 

( 

CD 

t. 

4-, 

cc 

cz 

.t 

Cu 

I 

C 

a * * * * 

CNN 

Cue1oc.4c4ot%'tm 

t LI)OctJflrO0t% 

Cu 

c) 

* * * 

- -'qt'. o. q N en'R 

'ON 

d& 

C .J Ct) N 00 Q\ LI) 

V CLI) N''J ('IL 

crL) 

0) 

• o - - 

0) 

LI 

* * 

* 

N 

* * 

'o 

q oP 

'cq 

cO'ONN CRcno 

0 - 

C 

Ct 

U 

Cu 

U 

C 

N 

C 

.0 

C 

C 

-- as 

Ec 

(-'I- 

I-. 

'to 

a-a 

.45 

as 

cr c-I 

Cc 

U 

to 

I-ct 

-4 41 

— — 

Ctca 

41 

to 

cE 'c 

Ccl - 

CO 

p 

CU 

Cu 

Cl) 

.4-Jo 

a- - 

Iq 

V 

4-a 

U 

C) 

(I) 

a) 

'4 

Cl) 

C) 

Cl) 

Cl) 

C) 

U 

0 

04 

C) 

• -I 

as 

Cu 

• -4 

Cu 

C 

0 

C 

C 

Cu 

C) 

Cu 

Cu 

'-4 

C) 

ca 

aj 

cz 

t 

0 

u 

< 

'e Cu 

C) V 

— Eu 

F- *0 

I 

— 

- ('I r

4.5 Effect of temperature on the fluorescence emission properties of the 2- 

aminobenzoic acid, 2-N,N-dimethylaminobenzoic acid and their 

methyl esters. 

Fluorescence spectra of the four ortho-isomers were measured in the 

three binary solvent mixtures butyronitrile / isobutyronitrile (9:1), 

chiorobutane / isopentane (9:1) and methylcyclohexane / isopentane (4:1) in 

the temperature range 77 - 298K. Unlike the corresponding meta- isomers, 

2ABA, M2AB and M2DMAB exhibit no obvious change in the shape of their 

spectra as the temperature is changed. However, in all three solvents 

2DMABA phosphoresces much more strongly than it fluoresces at low 

temperature. This observation may provide a clue to the generally 

anomalous properties of 2DMABA. 

As Table 4.6 shows, the general trend, although not as marked as for 

the meta- isomers, is for an increase in the fluorescence quantum yield as the 

temperature decreases. 

Cmd Solvent Quantumyield as a function of temperature 

293K 230K 193K 153K 133K 113K 

2ABA BuCN I IBuCN 0.42 0.51 0.70 0.79 0.80 0.79 

BC1 uP 0.47 0.55 0.59 0.70 0.71 0.70 

MCH / iP 0.27 0.29 0.35 0.39 0.22 0.29 

M2AB BuCN / iBuCN 0.36 0.49 0.53 0.54 0.56 0.57 

BC1 / iP 0.51 0.67 0.73 0.77 0.80 0.83 

MCH / iP 0.37 0.41 0.48 0.51 0.53 0.54 

2DMABA BuCN / iBuCN 0.007 0.007 0.004 0.005 10.018 0.029 

BC! / iF 0.004 0.003 0.002 0.001 0.001 0.002 

MCI-I / iF 0.003 0.003 0.002 0.001 0.001 0.001 

M2DMAB BuCN / iBuCNJ 0.05 0.06 0.10 0.12 0.12 0.14 

____________ BC1 uP 0.16 0.20 0.22 0.19 0.16 0.16 

MCH / iF 0.07 0.09 0.13 0.15 0.16 0.14 

Table 4.6 

Excited state emission characteristics of the ortho -compounds as 

a function of solvent and temperature 

97

4.6 Summary of conclusions 

In conclusion, 2ABA and M2AB in all solvents except water exist in the 

neutral molecular form. In water both compounds form hydrogen bonded 

complexes which are in equilibrium with the neutral molecular form. 

2DMABA in inert solvents exists as an intra-molecular hydrogen bonded 

complex, but in polar or hydrogen bonded solvents, in a similar manner to 

M2DMAB, it exists as at least two species, the composition and structure of 

which can only be conjectured. The data could be explained by the formation 

of an exciplex which involves the formation of a bond utilising the lone pair 

of electrons on the nitrogen. The other possibility is of course a TICT state, but 

further experiments are required along the lines of those described in more 

detail in chapter 5 to try to elucidate the structures, and indeed to verify that 

M2DMAB does show dual exponential decays in all the solvents studied, 

especially the inert solvents. As can be seen from Table 4.5, 2DMABA is 

clearly anomalous, but the ester behaviour suggests that there are two 

different situations, corresponding to non-polar / medium polar and in 

highly polar solvents. 

02

Chapter 5

Chapter 5 


Unlike the ortho-isomers described in chapter four the 4-amino 

(4ABA) and 4-N,N-dimethylaminobenzoic acid (4DMABA)and their methyl 

esters (M4AB and M4DMAB) cannot form intra-molecular hydrogen bonds. 

In the case of the two acids in inert solvents and only slightly polar solvents, 

dimer formation is feasible 54. Both acids are partially soluble in the nonpolar 

solvents used (hexane and methylcyclohexane I isopentane (4:1)) and 

4DMABA is also only partially soluble in water. In all other solvents 

studied, the four compounds are completely soluble. 

These four compounds exhibit the strongest UV absorption bands of 

all the compounds studied and in all the solvents used (Table 5.1). The 

absorption properties of the methyl esters are very similar to those of the 

parent acids in terms of their wavelength of maximum absorption, but the 

extinction coefficients are more varied with no obvious overall pattern as a 

function of solvent. As would be expected, replacement of the amino 

substituent by an N,N-dimethylamino group results in a red shift of the 

absorption band of some 2 - 4,000 cm -1 and a general increase in the strength 

of absorption (Figure 5.1). The same effect was observed with both the orthoand 

meta-isomers. 

4ABA and M4AB show a strong absorption band in the 280 rim 

region, whilst 4DMABA and M4DMAB show generally stronger absorbances 

in the 310 nm region. The absorption spectra for all four compounds in 

hexane and methylcyclohexane / isopentane (4:1) are more structured, and 

4DMABA also exhibits this structural detail in acetonitrile and ethanol. The 

occurrence and position of this structured absorption band varies from 

solvent to solvent and between compounds.

Solvent 

Measured 4ABA M4AB 4DMABA M4DMAB 

Species 

quantity 

Acetonitrile Am /nin 287 286 310 308 

c/ dm3 mol4 cm-1 11,200 17,500 27,100 19,700 

BuCN / iBuCN Xmax trim 286 288 307 309 

ci din3 mol4 15,300 25,300 24,400 29,700 

BC! / iF Xm /rim 278 275 310 305 

c/ din3 mol4 cnr1 17,300 16,000 19,400 20,600 

Ethanol Ax ,m 289 295 299 299 

8/ dm3 mo!-1 cm-1 16,200 17,900 24,400 19,800 

Hexane's Xmax /jjn 275 271 308 272 

________________ e/ din3 mo!-1 cm-1 Is 17,000 n/s 17,000 

MCH / iF )m /nm 276 271 307 304 

E/ dm3 mo!4 am1 n/s 22,400 n/s 25,700 

Water Amax /nm 278 285 311 316 

c/ dm3 mo!4 cm-1 12,800 14,200 n/s 16,600 

Neutral Xmax/nm 284 284 314 314 

E/ din3 mol -1 cm-i 11,500 16,000 n/s 16,000 

Anion Xmax/nm 266 - 284 - 

e/ dm3 moF1 cm-1 12,600 - n/s - 

Cation Xmax /nm 272 270 272 276 

c/ dm3 mo!1 cm-1 950 1,200 n/s 1,100 

Table 5.1 

Ground state absorption characteristics of 4-aminobenzoic, 

4-N,N-dimethylaminobenzoic acids and their methyl esters as 

a function of pH and solvent 

030 

0.25 

0-20 

C 

I. 

- 0.15 

-C 

0.10 

0.05 

0.00 

220 240 260 280 300 320 340 360 

Wavelength (nn.) 

Figure 5.1 Absorption spectra of the para compounds in acetonitrile 

100

It is worth noting that 4ABA in hexane shows another peak at 310 

nm, which could be an indication of the formation of either a dimer or a 

zwitterionic species, although after consideration of the corresponding 

spectra of the orthocompounds the latter does seem unlikely. There is no 

reason to doubt the conclusions of Jian et al. 45 that the lowest energy 

absorption band in these compounds is due to a it-iC * charge transfer 

transition. The slight red shifts in the absorption maximum of all four 

compounds in polar solvents and the high extinction coefficients which are 

observed provide further confirmation. 

A second absorption band which varies considerably in strength and 

position is observed in the 220 - 250 nm region, which is in accordance with 

those observations made for both the ortho- and meta-isomers. The values 

presented in Table 5.1 are in excellent agreement with those reported in the 

literature (Table 5.1a) 45,46,47,64'84• 

Solvent measured 

quantity 

4ABA 

Acetonitrile x.ax /nm 286 

MCH Amax /nxn 275 

Neutral Xrnax /rJn 286 

Cation A.max /nm 266 

Anion Xmax/nm 265 

Table 5.1a Ground state absorption characteristics of 4-aminobenzoic acid 

reported from Jiterature 447 

It was also found that the absorption spectra of 4DMABA and 

M4DMAB in acetonitrile over a ten fold concentration range (10 - - 10 5 mol 

dm 3) did not vary either in terms of shape or extinction coefficient within 

experimental error (Table 5.2) 

101

Concentration 

mol dm 

Pathlength 

111111 

4DMABA 

c 

dm3 mo!-1 cm-1 

M4DMAB 

c 

dm3 mo!-1 cm4 

1.09x10-4 1.00 24,659 20,329 

5.08x10-5 2.00 22,126 22,993 

2.32110-5 5.00 23,181 19,666 

1.02x10-5 10.0 1 25,643 1 19,275 

Table 5.2 

Ground state absorption data for 4DMABA and M4DMAB as a 

function of concentration 

These experiments were undertaken in view of the conclusion (see 

later) that the anomalous fluorescence in these compounds is at least 

partially due to ground state dimer formation. 

5.2 Ground state absorption and fluorescence emission spectra as a 

function of pH. 

In water, the four compounds are capable of protonation and both 

acids are capable of deprotonation to form the anion. The latter species 

absorbs more strongly but to the blue of the neutral molecule. The cations 

also show similar properties to those seen for the ortho- and ineta-isomers 

and absorb much more weakly and are strongly blue shifted with respect to 

the neutral acid. Absorption and fluorescence spectral measurements as a 

function of pH a!low the calculation of pK values for ground and excited 

protonation and deprotonation and these are presented in Tab!e 5.3. A 

typical example of the variation of the absorption spectra with pH is given 

in Figure 5.2. Comparison with the literature values for the only compound 

where these are avai!able (4ABA) is very good. 

102

1.6 

1.4 

'••7C 

0.6 

- - I 

/ . F '•; 

I 

I' / 

/ I 

. I .• 

I / 

I I - 

0-4 .'\\\\\ / 1/ •'-- 

•• 

------. 

a2T 

01. 

220 240 260 280 320 

pH - 1.61 

pH-Z09 

pH-3.W 

pH-3.91 

pH-t77 

340 


Figure 5.2 Absorption spectra of 4ABA as a function of pH 

Ground state So 

Excited state Si 

Compound plC1, plc p K1,* pK* 

Abs Fluor Abs Fluor 

4ABA 2.75 4.76 -6.72 - 9.76 6.17 

(2.14) (4.85) (-3.4) (10.8) (6.9) 

M4AB 2.33 - -7.23 - - - 

4DMABA 2.34 4.79 -7.99 - 11.85 7.02 

M4DMAB 2.41 1 - -7.98 - - - 

Table 5.3 

Ground and excited pK values for both protonation and 

deprotonation (values in brackets are taken from literature) 45 

There is only a difference of 2.0 - 2.5 pK units between the PK a and 

pKb values for the two acids, and it is therefore impossible to measure an 

absorption spectrum for the pure neutral acid in water, since it will always 

be contaminated by a few percent of the cation and / or anion. In unbuffered 

water, both 4ABA and 4DMABA will be partially ionised to their anion and 

this ionisation is observed in the absorption spectra of the two acids in 

unbuffered water. The fluorescence spectra of the four pa Ta- compounds in 

water vary with pH. Naturally, the behaviour is more complex for the two 

acids, which involve two protic equilibria, than for the two esters, which 

have only one. Förster cycle calculations of pK a* and pKb* predict that the 

ilulci

carboxylic acid groups will be weaker acids in the excited state than the 

ground state and that the amino and dimethylamino substituents will be 

weaker bases in the excited state. These predictions are identical to those for 

the ortho- and ,neta-isomers 

For 4ABA and 4DMABA at pH values which are neutral or basic, the 

observed fluorescence spectra are independent of pH. The principal species 

in absorption is the anion and the same is true in emission even though the 

pKa* appears to lie in the neutral / weakly basic pH region (Table 5.3). The 

absence of any fluorescence from the neutral species reflects the inability of 

the excited anion to gain a proton within its excited state lifetime, even 

though PK a* may apparently favour formation of the excited neutral 

molecule. At proton concentrations of 10.6 mol dm-3 and less, the second 

order reaction between a proton and an excited anion simply cannot 

compete with the radiative and non-radiative processes which give the 

excited anion a sub-nanosecond lifetime (see later). As the pH is decreased 

from 7.0 to a value of approximately 3.5, the fluorescence spectra of both 

4ABA and 4DMABA in water red shift and decrease in intensity. The spectra 

consist of a mixture of fluorescence from the anion and the neutral species 

(Figure 5.3). Excitation at the isosbestic point and analysis of the resulting 

spectra as a function of pH yields pK a values of 4.73 and 4.93 for 4ABA and 

4DMABA respectively. (The fluorescence spectra were analysed in a similar 

maimer to that adopted for the ground state absorption spectra; see chapter 

two). 

These values are in excellent agreement with the values obtained from 

the absorption spectra of the two acids. The behaviour of the fluorescence 

spectra in this pH region is therefore governed by the ground state equilibrium. 

Once again the proton concentrations are too low for protonation to compete 

with other excited state processes and the excited anion / neutral ratio produced, 

by absorption remains unchanged. In the pH range 3.5 - 10.0, 4DMABA also 

104

exhibits anomalous fluorescence in water (Figure 5.3), which will be described 

more fully in the following section. Further reduction of the pH from 

approximately 3.5 to a proton concentration in the region of 1.0 mol dm 3 

decreases the intensity of fluorescence from the neutral species, but does not 

eradicate it. On the basis of the p1

position of the a* band is very solvent dependent. The origin of this 

anomalous fluorescence is controversial, with TICT state formation 28 ' 

exciplexes7° and ground or excited state dimers 484952 havhig been postulated. 

Solvent 

Species 

Measured 

4ABA M4AB 4DMABA 

___ 

M4DMAB 

______ ________ 

b* a* b* 

quantity 

Acetonitrile FluorXmax 331 333 350 480 350 480 

0.27 0.34 0.007* 0.009* 0.006* 0.011* 

BuCN / IBuCN Fluor Xmax ,'nm 338 333 340 460 342 460 

Of 

0.36 0.38 0.003* 0.031* 0.005* 0.022* 

Phos Xmax /nm 430 420 440 - 440 - 

BC1 / iP FluorAmax /nrn 325 325 350 425 352 415 

0.28 0.40 0.040* 0.12* 0.086* 0.17* 

Of 

PhosA.max/nm 448 430 450 - 450 - 

Ethanol FluorXmax /ran 345 456 351 420 347 - 

0.12 0.10 0.014 0.006* 0.060* - 

Hexane FluorXmax/nm 319 316 334 - 348 - 

4)f 0.09 0.14 0.14 - 0.30 - 

MCH / iF Fluor Xmax mm 322 320 340 - 335 - 

0.03 0.12 0.16 - 0.10 - 

Of 

PhosXmax/nm 440 421 451 - 445 - 

Water FluorXmax 353 366 500 346 - 

/TI 

0.07 0.02 0.015* 0.009* 0.022* - 

Of 

Neutral FluorXmax 349 354 370 - 365 - 

0.05 0.02 0.009 - 0.025* - 

Anion Fluor Xmax /nin 341 - 356 490 

Of 

0.102 - 0.018 0.011 

-. 

- 

- 

- 

Note on table b.4 - denotes a quantity wducn varies witn concentration 

Table 5.4 Excited state emission characteristics of the 4-aminobenzoic, 4- 

N,N-dimethylaminobenzoic acids and their methyl esters as a 


51 

Unlike the ground state absorption spectra, the excited emission 

spectra of the four pa Ta- compounds vary in an expected manner. In all the 

solvents used, the wavelength of maximum emission for the b* band is red 

106

shifted with increasing solvent polarity. It increases even more with the 

ability of the solvent to form hydrogen bonds. This trend is also seen for the 

a* fluorescence observed for 4DMABA and its methyl ester. The 

fluorescence band of 4DMABA and M4DMAB in hexane is similar to that 

found in all other solutions where the anomalous fluorescence occurs and 

can therefore be attributed to the normal b* fluorescence. The fluorescence 

spectral and quantum yield data as a function of solvent are presented in 

Table 5.4 and the data previously reported in the literature are shown in 

Table 5.4a. Literature data is surprisingly lacking for these compounds, but, 

where available it is in reasonable agreement with our results. 

Solvent Measured 4ABA 4DMABA M4DMAB 

quantity b* a* 

Acetonitrile Fluor Xmax /nm 346 46 480 46 480 46 

0.022 46 0.001346 

0.0029 

Of 

46 

Ethanol Fluor Xmax mm 338 

MCH / FluorXmax ,/rim 331 45 32346 r3324 

0.21 46 0.24 

Neutral Fluor 350 45 

Anion Fluor Xmax 11rim 338 

Table 5.4a Excited state emission characteristics of the four para 

compounds which have been reported in the literature 

The anomalous fluorescence of 4DMABA and M41)MAB is observed in 

acetonitrile, butyronitrile / isobutyronitrile (9:1) and chlorobutane / isopentane 

(9:1) with the acid also showing this anomalous fluorescence in 

water and ethanol. However as shown in Figure 5.3 the anomalous 

fluorescence observed in water is very pH dependent usually only showing 

very weak intensities in the a* band. In our hands, the fluorescence spectra 

and quantum yields for 4DMABA and M41)MAB in the solvents where the 

anomalous fluorescence is observed are concentration dependent, and 

consequently quantum yields for these solute / solvent combinations are 

107

quoted in Table 5.4 at a standard concentration of 1x10 4 mol dm7 3 to provide 

an indication of the relative strength of the fluorescence in these systems. 

We have also attempted to calculate individual quantum yields for the a * 

and b* fluorescence bands. This is relatively straightforward in solvents 

such as nitriles, where there is little overlap of the spectra. In chiorobutane 

and ethanol, the two individual quantum yields have to be viewed with 

caution as the two bands are rather closer together and separation of the 

individual spectra to produce the calculated quantum yields is more 

difficult. Cowley and co-workers 46'47 also quote some quantum yields for 

4DMABA and M4DMAB (Table 5.4a), which, given the concentration 

dependence of these values, are in reasonable agreement with our values in 

acetonitrile. 

C 

oj6 

.4S 

C 

1 

C 

@14 

U 

@1 

0 

- 

IL. 

I 

320 340 360 380 400 420 440 


Figure 5.4 fluorescence emission spectra of 4ABA and M4AB in 

acetonitrile 

The anomalous fluorescence observed for 4DMABA and M41DMAB is 

shown in Figures 5.6, 5.7, 5.11, 5.14 and 5.15, in much more detail. By way of 

contrast, the emission properties of 4ABA and M4AB are completely 

normal. A single emission band is observed peaking between 330 and 355 

nm (depending on solvent) and there is absolutely no evidence of any

anomalous emission. Typical fluorescence spectra for 4ABA and M4AB are 

shown in Figures 5.4 and 5.5 and the emission data for these compounds is 

summarised in Table 5.4. 

6 

I 

C 

1 

0 

290 310 330 350 370 390 410 430 450 


Figure 5.5 

fluorescence emission spectra of 4ABA and M4AB in hexane 

5.4 Effect of excitation wavelength on the fluorescence properties of 

4DMABA and M4DMAB 

The fluorescence observed from 4DMABA and M4DMAB in hexane 

and M4DMAB in ethanol and water is independent of excitation 

wavelength, apart from a change in the overall intensity. These, of course, 

are the solvents in which no anomalous fluorescence is observed for these 

compounds. In all the solvents for which anomalous fluorescence is 

observed for 4DMABA and M4DMAB, the intensity distribution between 

the a* and b* bands is dependent upon excitation wavelength. The results of 

measuring fluorescence spectra at different excitation wavelengths for these 

samples are given in Table 5.5. 

109

Xex 

(nm) 

Intensity 

Acetonitrile _ Ethanol __ __ Water ______ 

Intensity 

a * 

Ratio 

Intensity 

b 

Intensity 

a 

Ratio 

Intensity 

if 

Intensity 

a* 

Ratio 

b/a 

4-N,N-Dimethylaminobenzoic_acid 

230 56 - - 412 48 8.58 6.2 4.1 1.51 

240 56 - 244 48 5.08 5.8 4.8 1.12 

250 104 - 214 72 4.45 8.2 5.0 1.64 

260 180 - 228 100 2.28 14.0 3.8 3.68 

270 252 12 21.0 272 140 1.94 17.5 5.0 3.50 

280 252 16 15.8 308 156 1.93 17.4 5.6 3.10 

290 200 24 8.33 276 140 1.97 15.8 5.8 2.72 

300 128 28 4.60 169 100 1.69 11.3 5.5 2.10 

310 68 28 2.40 - - - 7.8 1 4.5 1.73 

320 22 16 138 - - - - - - 

__ Methyl_4-N,N-dimethylaminobenzoate 

240 - - - 3 - - 40 - - 

250 - - - 100 - - 44 - - 

260 22.5 - - 300 - - 46 - - 

270 62.5 40 1.56 312 - - 105 - - 

280 100 60 1.66 256 - - 75 - - 

290 130 120 1.08 310 - - 25.7 - - 

300 135 156 0.86 260 - - 18.0 - - 

310 105 170 0.62 190 - - 12.3 - - 

320 50 100 0.50 - - - 14.3 - - 

Table 5.5 

fluorescence intensity of both emission peaks observed for 

4DMABA and M4DMAB as a function of excitation 

wavelength 

2 

7.00 

6.00 

5.00 

4.00 

' 

La 27Onm 

-----La2SOnm 

Lex29Onm 

Lex300nm. 

La3l0nm 

3.00 

2.00 . 

/ : .----- --- .- 

1.00 :_. -- ----- 

0.00! I I I I I 

310 360 410 460 510 560 


Figure 5.6 fluorescence emission spectrum of 4DMABA in acetonitrile as 

a function of excitation wavelength 

110

An example of the observed variations is given in Figure 5.6 where 

the b* intensity has been normalised to the same value for all the spectra. 

For 4DMABA in acetonitrile, the excitation spectrum for the b* 

fluorescence peaks at 275 nm and that for the a* fluorescence at 

approximately 305 nm. The latter corresponds well to the ground state 

absorption spectrum. The values observed for the benzoate are 295 nm (b*) 

and 305 nm (a*). For the acid in acetonitrile, if the excitation wavelength is 

moved to a shorter value (270 nm), the intensity of the b* fluorescence 

increases approximately fourfold whilst the a* fluorescence almost 

disappears (Figure 5.6). In all cases, the a* fluorescence appears to be excited 

preferentially by longer wavelengths than the b* fluorescence (Table 5.5). 

For 4DMABA in ethanol and water a similar effect is observed, 

although in ethanol the two peaks appear to have very similar maximum 

excitation wavelengths. In water the maximum excitation wavelengths are 

280 rim (b*) and 295 nm (a). From these data, it seems likely that the species 

which is forming the anomalous fluorescence is already formed in the 

ground state, a feature which has not been reported before. A study of the 

effect of concentration on the fluorescence spectra was therefore thought to 

be appropriate. 

5.5 Variation of fluorescence intensity of 4DMABA and M4DMAB as a 

function of concentration 

The fluorescence emission spectra of both 4DMABA and M4DMAB 

in Ocetonitrile vary with concentration as depicted in Table 5.7. 

111

Table 5.6 

Concentration Intensity Intensity Ratio 

mold& b 

a* 

4-N,N-Dimethylaminobenzoicacid 

1.09x10 5 0.261 0.671 0.38 

5.42x10 6 0.205 0.515 0.39 

1.09x10 6 0.063 0.133 0.46 

5.42x10 7 0.042 0.082 0.50 

1.09x10 7 0.032 0.052 0.60 

1.09x10 0.024 0.011 1.00 

Methyl 4-N,N-dimeth 'laminobentoate 

5.28x10 6 0.602 1.181 0.51 

1.05x10 6 0.171 0.304 0.55 

5.28x10 7 0.082 0.149 0.55 

1.05x10 7 0.057 0.037 1.53 

5.28x10 8 0.066 0.011 6.01 

1.09x10 8 0.047 0.008 6.00 

fluorescence ethission ratios of 4DMABA and M4DMAI3 as a 

function of concentration in acetonitrile 

Figure 5.7, where the intensity of the b* band has once again been 

normalised, illustrates how the fluorescence of 4DMABA in acetonitrile 

varies with concentration. As the concentration is reduced the fluorescence 

intensity of the a* band becomes less and less until it is debatable whether it 

is still there or not (although this minimum intensity is not depicted in 

Figure 5.7). 

0.9 

0.8 

0.7 

'-5 

6 

>26 

- 0.6 

C 

3 

.E o.s 

4J 

U 

C 

0.4 

D.8 

D-8 

0.3 

0.2 

0.1 

0 

310 360 410 460 510 560 610 


Figure 5.7 fluorescence emission spectra of 4DMABA as a function of 

concentration (mol dm 4) in acetonitrile 

112

There are a substantial number of mechanisms which could be 

invoked to account for the occurrence of this anomalous fluorescence: 

1) A TICT mechanism in which there is no ground state molecule 

which is already twisted and in which the excited TICT molecule 

decays into a non-twisted ground state. 

2) A TICT mechanism with an equilibrium being established between 

TICT and planar molecule in the excited state, but still no ground 

state twist. 

3) A TICT mechanism with equilibrium in the ground state between 

planar and twisted species. 

4) Ground state dimer formation, with decay of excited dimer to ground 

state dimer or with decay of excited dimer into an excited monomer 

and a ground state molecule. 

5) Reversible or non-reversible excimer formation 

6) Reversible or non-reversible exciplex formation 

7) Two or more of the above happening simultaneously such as a nonreversible 

excimer formation, and exciplex formation. 

Of these mechanisms, only 4, 5 and 7 should show a concentration 

dependence such that a variation of the fluorescence properties with 

concentration would be expected. We have, however, attempted to gain data 

to investigate all the mechanisms and this is reported later in the thesis. 

113

Exploration of the possibility of excimer formation is based on the 

following mechanisms 

Non reversible excimer formation 

M 

h, 

+ M_ K, )'Excimer 

K1 

)M+ht1 

K 

)3M(isC)+M(1C) 

'Excimer 

1Excimer 

K 'ht + 2M 

4 3Excimer + isc or 2M(ic) 

Reversible excimer formation 

M 

h 

+ M_ K,, )'Excimer 

x 1 

K 

)3M(isc)+M(iC) 

1Excimer' 

'Excimer' 

K,,., M+'M' 

K ht4 + 2M 

1Excimer'_ ' ) 3Excimer + isc or 2M(ic) 

Applying the steady state approximation to these two mechanisms, 

equations 5.1 and 5.2 

1 & + Kex[M] 

Of sr — 1(1 K 1 (5.1) 

1 - (K +K)(K 1 +Knr) K +K, (5.2) 

Lx - 

lI KK CX [M] K lf 

are obtained for the variation of the fluorescence quantum yields for the 

mechanism involving non-reversible excimer formation. 

114

If excimer formation is reversible, the relationships are more 

complicated but their basic form is unchanged and equations 5.3 and 5.4 are 

obtained 

1 1 K nr (K+K ' nil 

\K rev 

[M] 

K f (5.3) 

1 - (K + + K )(K 1 + K s,) + K + (5.4) 

•Ex 

91 ICK eX EM] 

In both cases an excimer mechanism (reversible or non-reversible) predicts 

that plots of 1 / qM vs [M] and 1 / 4" vs 1/ [M] will be linear. The first of 

these plots is shown in Figure 5.8 for 4DMABA and M4DMABA in 

acetonitrile, whilst the latter is shown in Figure 5.9. 

45 

40 

35 

C 

0 

25 

t 20 

I. 

S 

01 I I I 

Q.QQ+O0 2.000-06 4000.06 6.000-06 8.00E-06 1000.05 1.200-05 

Concentration (mol dm-3) 

Figure 5.8 Plot of 1/4 vs [M] for both 4DMABA and M4DMAB in 

acetonitrile 

115

1.20E.02 

11K3E+02 

a 

8.00E+0I 

2 4.00E+0I 

C 

0 2.00E+O1 

O,00E.W 2E.W 4.E.07 6,E.W 8E,W 

I / concnetration (1/mat dm-3) 

Figure 5.9 Plot of 1 / 4" vs 1 / [M] for both compounds in acetonitrile 

Unfortunately, the possibility that the anomalous fluorescence arises 

from dimer formation in the ground state is difficult to test. The absence of 

any obvious change in the absorption spectra of 4DMABA and M4DMAB in 

acetonitrile with concentration makes it impossible to quantify any ground 

state equilibrium that is in existence. The effect of excitation wavelength on 

the fluorescence clearly suggests the presence of two ground state species 

and it is interesting to note that a plot of the quantum yield of the "normal" 

fluorescence against concentration for both 4DMABA and M4DMAB gives a 

near linear graph as shown in Figure 5.10. This may be simply fortuitous. 

0.7 

S. 

0.6 

03 

0.4 

0.3 

t 0.2 

& 

0.I 

0 I 

0.00E+00 2.00E06 4.00E-06 6000.06 6006.06 2.006-05 2.206-05 

Concentration (pm' dm.3) 

Figure 5.10 Plot of 4 

vs EM] for both compounds in acetonitrile 

5.6 fluorescence lifetime data as a function of solvent and pH 

1111

Fluorescence decay profiles may provide further evidence to assist in 

assigning the origin of the anomalous fluorescence in 4DMABA and 

M4DMAB. The fluorescence kinetics of both the b* and the a* bands have 

been measured in a variety of solvents at room temperature and analysed in 

terms of a sum of up to three exponential components. Parallel 

measurements have also been made for 4ABA and M4AB and these results 

are presented in Tables 5.7, 5,8 and 5.9. 

Solvent 

Species 

Xem 

(nm) 

X2 Ti (ns) a1 t2(ns) a2 

4-Aminobenzoic acid 

Acetonitrile 331 1.08 1.14 

BC! / iP 315 1.22 0.49 0.7995 4.08 0.2005 

Ethanol 345 1.09 0.28 0.9673 1.45 0.0327 

Hexane 319 1.02 1.01 0.9174 2.49 0.0826 

Water 349 0.63 0.79 

Anion 349 1.24 0.72 

Neutral 341 1.33 0.82 

Methyl &aminobenzoate 

Acetonitrile 333 1 1.03 11.13 1 

BC1/iP 330 1 1.19 11.27 1 

Ethanol 346 1.11 0.14 0.9959 1.48 0.0041 

Hexane 318 1.09 2.06 

Water 353 1.07 11.34 

Table 5.7 fluorescence decay data for 4ABA and M4AB as a function of 

solvent, pH and temperature 

In every case the observed profiles rise promptly and there is no 

evidence for an excited precursor to the fluorescent state within the 

experimental limits (say 100 ps) of the apparatus used. Although this in 

itself does not rule out the possibility of exciplex and excimer mechanisms 

which involve a very rapid (>1010 s 1 ) formation step, one would expect to 

see this reflected in the decay of the b* state. In virtually every case, the b* 

decay times are at least an order of magnitude longer lived (i.e. > 1.0 ns) 

than this. Additionally, for an excimer mechanism operating at 

concentrations of the order of 10 mol dm 3 and less, Kex would have to be 

>1015 dm3 mol1 1; five orders of magnitude greater than the diffusion 

controlled rate. We therefore believe that the fluorescence decay data back 

117

up the earlier argument that the anomalous fluorescence in 4DMABA and 

M4DMAB does not originate from an excimer. 

Solvent 

Xern V ti(ns) a1 t2(ns) a2 t3(ns) a3 

Species 

4-N,N-dimethylaminobenzoic acid 

Acetonitrile 350 1.29 1.72 

480 1.01 2.23 

BuCN / iBuCN 350 1.11 1.51 0.9590 5.86 0.0410 

485 11.54 2.39 

BC1 / IP 355 1.24 1.15 0.4470 3.78 0.5510 

420 1.30 3.25 0.2218 4.82 0.1814 0.04 0.5968 

Ethanol 351 1.01 2.86 0.8603 17.61 0.1397 

MCH / iP 334 1.21 1.25 0.7275 1 6.96 0.0333 0.45 0.2392 

370 1.43 1.29 0.9919 1 6.03 0.0081 

Hexane 334 1.01 1.28 

Water 345 1.05 0.77 0.9411 5.07 0.0589 

480 1.27 0.17 0.9580 2.39 0.0420 

Neutral 370 1.89 0.85 

480 1.27 0.17 0.9437 2.34 0.0563 

Anion 356 1.44 1 0.76 

480 1 1.19 1 0.25 0.9659 0.82 1 0.0341 

Table 5.8 

Lifetime data for 4DMABA as a function of pH and solvent 

Solvent 

Aern V 'tj(ns) a1 T2ns) a2 

Methyl_4-N,N-dimethylaminobenzoate 

Acetonitrile 350 1.08 1.65 0.8415 2.45 0.1585 

480 1.04 2.38 

BuCN / 1BuCN 340 1.23 1.07 0.9930 3.54 0.0070 

480 0.93 0.40 0.1721 3.00 0.8279 

BC1 / iP 345 1.11 1.12 0.3416 3.65 0.6584 

420 1.06 3.52 0.8429 5.27 0.1571 

Ethanol 347 1.94 0.26 0.9972 6.83 0.0028 

MCH / iP 334 1.19 0.95 0.9986 7.61 0.0014 

370 1.40 0.98 0.9525 6.98 0.0475 

Hexane 348 1.30 1.32 0.5000 2.12 0.5000 

Water 345 1.40 0.11 0.9915 5.30 0.0085 

Table 5.9 

Lifetime data for M4DMAB as a function of pH and solvent 

The fluorescence decay profiles for 4ABA and M4AB are for the most 

part adequately fitted by a single exponential. Where this is not the case, a I 

mean value (= E1a1t/E1a1 ) is also presented (Tables 5.13) to allow 

comparisons to be drawn amongst the data. In the absence of any 

118

comparable literature data on these two compounds and having no kinetic 

reasons to expect multi-exponential fluorescence kinetics from them, it is 

difficult to decide whether the double exponential fits are real or not. 

Further measurements are needed to clarify this question. Assuming that 

the fluorescence kinetics are uniformly first order and so single exponential 

decays should be observed, fluorescence lifetimes for 4ABA and M4AB of 1 - 

2 ns are observed in virtually all solvents. Combination of these lifetimes 

with the quantum yields given in Table 5.4 allows the calculation of 

radiative and non-radiative rate constants. These are presented in Table 

5.10. 

Solvent O f 'Cf Kf 

x10 8 

Knr 

xlO-8 


Acetonitrile 0.27 1.20 1.88 6.08 

BC1 / ip 0.28 1.21* 2.31 5.95 

Ethanol 0.12 0.32* 3.75 27.5 

Hexane 0.09 1.13* 0.80 8.05 

Water 0.08 082 0.61 11.61 

Methyl_4-aminobenzoate 

Acetonitrile 0.34 1.05 3.24 6.29 

BCl / ip 0.40 1.27 3.15 4.72 

Ethanol 0.10 0.15* 6.67 60.00 

Hexane 0.14 2.3 0.61 3.74 

Water 0.02 1.34 0.15 7.31 

* t mean calculated 

Table 5.10 Calculated radiative and non-radiative rate constants for 4ABA 

and M4AB 

It is clear from Tables 5.7 and 5.10 that, in terms of the interaction of 

4ABA and M4AB with the various solvents used in this study, there 

appears to be an anomaly with ethanol. This is the one solvent which leads 

to fluorescence lifetimes substantially below the 1 - 2 ns range and this 

appears to be due to an enhancement of the non-radiative decay process in 

119

the two molecules. It is possible to speculate about the nature of this 

interaction, but it is not possible to be sure as to its exact nature without 

further data. Herbich et al.85 have recently reported fluorescence quenching 

of 2-(2'-pyridyl)indoles by alcohols, speculating that this arises through 

enhanced internal conversion. The importance of cyclic inter-molecular, 

hydrogen-bonded structures in this enhancement is stressed (acyclic 

structures are found not to quench the fluorescence). It is difficult to see 

how such cyclic structures could be involved here, without involvement of 

several solvent molecules, but the idea of enhanced internal conversion via 

a specific solute I solvent interaction would seem a logical way to explain 

our data. The alternative to internal conversion is intersystem crossing, but 

it is hard to envisage how a molecule such as ethanol would increase spinorbit 

coupling between S1 and T n and thus increase ISC. In the case of the 

dimethylamino compounds 4DMABA and M4DMAB, a small amount of 

literature data is available and the fluorescence lifetime obtained for 

M4DMAB in hexane agrees excellently with values of 1.1 - 1.3 as for a range 

of non-polar solvents quoted by Howell et al.49 . 

The fluorescence decay profile of the anomalous fluorescence 

observed in ethanol for 4DMABA was not measured. The a* and b* peaks 

could not be adequately resolved and so the proximity of the two 

fluorescence bands leads to a complex decay profile requiring the sum of 

three exponentials to adequately fit the data. We are also able to confirm 

results reported by Howell et aL. 49 who observed dual exponential decays for 

M4DMAB in mixed alkane solvents. However in none of the cases have we 

observed a negative pre-exponential factor which would be expected if the 

a* fluorescence were to originate from the b* state as would be the case for 

both TICT and excimer formation. This points to the anomalous 

fluorescence being due to a ground state dimer. 

120

In those solvents where anomalous fluorescence is not observed, the 

quantum yield and lifetime data for 4DMABA and M4DMAB has been 

combined to produce rate constants (Table 5.1). 

The values obtained are very similar to those for 4ABA and M4AB. 

Once again there appears to be enhancement of the non-radiative decay 

route in ethanol as well as in water. There was some suggestion of the latter 

for 4ABA in water (Table 5.10), but not for M4AB even though it has a low 

fluorescence quantum yield. This leads to the suggestion that the lifetime 

for M4AB in water may be shorter than the quoted value and clearly points 

to the need for further data. 

Solvent O f tf Kf 

Knr 

x10 8 

x10 8 

4-N,N-Dimethylaminobenzoic acid 

Hexane 0.14 1.28 1.10 6.72 

MCH I ip . 0.16 1.25* 1.28 6.72 

Methyl 4-N,N-dimethylaminobenzoate 

MCH / ip 0.10 0.96* 1.04 9.38 

Ethanol 0.06 0.28* 2.14 33.57 

Hexane 0.30 1.72* 1.74 4.07 

Water 0.03 0.15* 1.67 65.00 

* tmean calculated 

Table 5.11 Calculated radiative and non-radiative rate constants for 

M4DMAB and 4DMABA in various solvents 

In view of the changes in the fluorescence spectra of 4DMABA and 

M4DMAB with excitation wavelength, concentration and pH discussed 

earlier in this chapter, fluorescence decay measurements were made for the 

two compounds in acetonitrile as a function of excitation wavelength and 

concentration (Tables 5.12 and 5.13). 

121

XExI XEni I x I115I a12i a2 


280 355 1.035 2.13 

1. 

480 1.243 2.19 

310 355 1.170 2.12 

480 1.356 2.15 

340 355 1.566 2.27 

480 1.309 2.18 

Methyl 4-N,N-dimethylaminobenzoate 

270 355 1.156 1.27 0.8947 2.31 0.1053 

270 480 1.439 2.68 

305 355 1.159 1.26 0.8140 2.49 0.1860 

305 480 1.467 2.64 

340 355 1.044 1.38 0.3072 2.77 0.6928 

340 480 1.340 2.67 

Table 5.12 fluorescence decay lifetimes of 4DMABA and M4DMAB in 

acetonitrile as a function of excitation wavelength 

Concentration XEm V 'C1 ins a1 t2 ins a2 

moldnr3 

mean 


5x10 5 350 1.61 1.66 

1. 

480 1.48 2.30 

5x10 6 350 1.26 1.29 0.8420 2.74 0.1580 1.52 

480 1.01 2.23 

5x10 7 350 1.23 1.20 0.6869 1.85 0.3131 1.98 

480 1.14 2.27 

Methyl_4-.N,N-dimethylaininobenzoate 

5x10 6 355 1.11 1.70 0.9926 2.21 0.0074 1.70 

_____ 480 1.04 2.36 

5x10 7 355 1.08 1.65 0.9826 2.45 0.0174 1.66 

_____ 480 0.97 2.49 

5x10 355 1.00 1.69 0.9598 2.49 0.0402 1.72 

480 0.96 2.71 

Table 5.13 Fluorescence decay lifetimes of 4DMABA and M4DMAB as a 

function of concentration in acetonitrile (excitation 

wavelength 305 nm) 

For 4DMABA, the lifetime of the a* band is independent of 

concentration and excitation wavelength and the same is true for the 

benzoate. A lifetime in the range 2.1 - 2.3 ns for the acid and 2.3 - 2.7 ns for 

the benzoate is obtained with neither showing evidence of a rise time which 

would be indicative of the excited state precursor which would be expected 

for TICT or excimer formation. These observations confirm our earlier 

122

conclusion about the presence of a ground state species producing the a * 

fluorescence. They also imply that this is the only source of the a* species. 

The b* excited state cannot be the precursor of a* although the reverse may 

possibly be true given the more complex decay profiles observed for the b* 

fluorescence band. The decay of the b* fluorescence is not so simple with the 

ben.zoate showing dual lifetimes of =1.7 ns and =2.4 ns, the latter values 

being in a similar range to those found for the a* emission although it only 

contributes a small percentage of the total emission. The acid on the other 

hand shows single exponential decays at a concentration of 1x10- 5 mol dm'3 

and this is independent of excitation wavelength. However at 

concentrations of lx10" 6 mol dm-3 or less, this single exponential changes to 

a dual exponential with lifetimes of 1.3 and 2.8 ns, the latter figure being 

similar to the value of 2.3 ns observed for the a* fluorescence. 

Multi-exponential decays are also observed in ethanol and water. 

Given the hydrogen bonding ability of both these solvents, it is not 

surprising that, more than one excited state solute / solvent arrangement 

may be formed and so the complex kinetics may be real. Indeed, the multiexponential 

fits may be due to a distribution of lifetimes from an array of 

slightly different solute I solvent interactions, as has been proposed by 

Howell et al.49 for M4DMAB in mixtures of non-polar solvents. The 

possibility of a non-fluorescent TICT state in these solvents is not ruled out. 

Unless there exists a very short lived precursor whose lifetime is so short 

that we are unable to separate it from the instrumental response function 

when we model the fluorescence decays of the a* and b* states, both a* and 

b* are excited directly from the ground state. Once the two excited states 

have been formed, it is possible that a* converts to b*, but there is no 

evidence for the reverse process. 

Varma and co-workers 50 have proposed that the anomalous 

fluorescence of ethyl 4-N,N-dimethylaminobenzoate and 4-NM- 

123

dimethylamino-benzonitrile in acetonitrile is due to exciplex formation 45 . 

Our conclusion from these data is that the fluorescence spectrum and decay 

profile of the b* state is comprised of contributions from a distribution of 

excited solute / solvent arrangements, the decay kinetics of which are 

adequately described by a sum of two exponentials. Unlike Howell et al.,49 

we have no evidence to suggest that a* derives from b*. The additional 

observation of the a* fluorescence band leads us to conclude that the a* is an 

excited dimer or possibly a higher aggregate which is populated directly 

from the ground state by absorption of a photon of light. Given the apparent 

lack of concentration dependence of the ground state absorption spectrum, 

the ground state dimer may not involve a strong interaction between 

molecules. However a loose association is sufficient to organise the 

molecules such that they are in approximately the correct position to form 

an excited dimer when a photon is absorbed by one of the molecules. We are 

conscious that the time resolution of our time-correlated, single photon 

counting system would not allow us to measure a rapid rise time (perhaps 

fluorescence properties of the two molecules is summarised in Table 5.14 

and typical spectra (for 4DMABA) are shown in figure 5.11. 

In both cases, although there is no obvious change in the absorption 

spectrum, the b* fluorescence in hexane is quenched by low concentrations 

of acetonitrile, but at slightly higher concentrations it increases in intensity 

and red shifts some 10 - 20 nm. At the same time there is a steady increase in 

a* fluorescence as the acetonitrile concentration is increased (Figure 5.11 and 

5.12). It would appear that there are some specific solvation effects taking 

place in addition to the mechanism producing the a* fluorescence. The red 

shift in the b* fluorescence probably results from the formation of a specific 

4DMABA / acetonitrile solute / solvent interaction, but could also arise 

because of the increase in the polarity of the overall solvent mixture as a 

result of the addition of the polar acetonitrile molecule. 

4DMABA 

M4DMAB 

%CH3CN mt b* Tnt a * a* / b* mt b* Tnt a* a* / b* 

0.0 3030 0.00 0.000 571 64.70 0.113 

0.2 29.00 5.00 0.172 570 127 0.222 

0.4 20.00 8.50 0.425 497 176 0.354 

0.6 8.00 7.00 0.875 433 212 0.490 

0.8 3.50 4.25 1.214 362 235 0.649 

1.0 9.20 10.10 1.097 341 252 0.739 

1.2 5.95 9.80 1.647 282 235 0.833 

1.4 14.00 22.40 1.600 280 247 0.882 

1.8 17.50 35.40 2.000 206 218 1.058 

2.0 14.00 29.30 2.092 194 229 1.180 

Table 5.14 Fluorescence intensities (Int) of both the normal and 

anomalous fluorescence peaks of both 4DMABA and 

M4DMAB in hexane (=1x104 mol dm 3) as a function of added 

acetonitrile 

125

45.00 

40.00.. 

i I 

300 

O.M. CH3CN 

-----0A%CH3CN 

0.6% C}-I3CN 

350 400 450 500 550 

Wavetength (nm) 

Figure 5.11 Fluorescence emission spectra of 4DMABA in hexane as a 

function of added acetonitrile 

The new band at 440 nm may be an exciplex or may be the same 

species as is seen in pure acetonitrile but with a blue shifted fluorescence 

due to the non-polar solvent environment in which it is situated. 

Whatever the nature of the species producing the 440 nm fluorescence, it is 

dear that the acetonitrile plays a key role in its formation. 

This effect of adding a polar solvent to a solvent of lower polarity was 

first noted by Chandros 66 who observed similar results to those above for 

DMABN in methylcyclohexane I acetonitrile mixtures. He felt that the 

addition of the polar solvent facilitated the formation of an exciplex. Varma 

et al.68 invoking a similar mechanism, considered that the anomalously 

emitting species is either formed from the reaction of a single polar 

molecule (P) with a molecule of DMABN in its first excited singlet state or 

arises from an already existing ground state complex between solute and 

solvent. If the latter was the case however, it would be expected that the 

ratio of the two peaks would depend on the initial concentration of 

DMABN, which has not been observed. In the case of 4DMABA and 

M4DMAB we have observed a concentration dependency in pure solvents 

iPLi

ut have not yet repeated this experiment in solvent mixtures at different 

solute concentrations. Rettig et al. 57 have also postulated that the affect of 

the addition of a polar solvent to an inert solvent helps to relax the 

activation energy of the TICT state and so facilitates TICT state formation. 

-2.50 

2.00 

150 

Cd 

1.00 

0.50 

i I I I I 

0.00 0.20 0.40 0.60 0.80 1.00 120 1.40 1.60 1.80 2.00 

Concentration of added acetonitrile % 

Figure 5.12 Plot of the ratio of a* / b* for both 4DMABA and M4DMAB in 

hexane as a function of added acetonitrile 

0% --- -- 0.25T---------0.40% - - -- -- 0.50% - ---- 75% 

1.00% ----- 2.0Ot- -------- 3.00% ------4.00% 

0.60 

0.50 

a, 

v 0.40 

e 0.30 

0 

0.20 

0.10 

0.00 

220 240 260 280 300 320 340 


Figure 5.13 Cround state absorption spectra of 4DMABA in acetonitrile 

(5.03x1O mol dma) as a function of added water 

127

Jouvet 63 has presented evidence that the anomalous fluorescence in 

this series of compounds derives from the formation of solute / water 

complexes. We have therefore undertaken some preliminary experiments 

on the effect of water concentration on the absorption and fluorescence 

properties of 4DMABA and M4DMAB in acetonitrile. Although the 

addition of acetonitrile to a solution of either 4DMABA or M4DMAB in 

hexane causes no change in the ground state absorption spectra, addition of 

water to a solution of 4DMABA in acetonitrile (Figure 5.13 and Table 5.14) 

does result in a dramatic change in the ground state absorption spectra. This 

is not the case for the corresponding benzoate. 

1.07x10 7 mol dm 3 5.03x10 6 mol din-3 

%H20 hit hit hit Mt 

308nm 282nm 308nm 282nin 

0 0.5510 0.2973 02662 0.0967 

0.25 0.4555 0.3319 0.2408 0.2621 

0.40 0.4164 0.3451 0.1517 0.1720 

0.50 0.4077 0.3610 1 0.1494 0.1743 

0.75 0.3950 0.3781 0.1374 0.1814 

1.0 0.3723 0.4204 0.1025 0.2051 

2.0 0.2307 0.4749 0.0776 0.2264 

3.0 0.2196 0.4959 0.0757 0.2304 

5.0 0.1939 0.4964 - - 

Table 5.15 Ground state absorption data for 4DMARA in acetonitrile as a 

function of added water at two concentrations 

On addition of > 5% water to the above concentrations of solution, 

the absorption spectrum becomes unreadable because of the turbidity of the 

solution and so only values up to 5% were taken. The spectra show that at 

concentrations of 2% water, the ground state species absorbing at 308 nm 

becomes the minor component and the 282 nm species becomes the major. 

This corresponds well to the fluorescence excitation spectrum of 4DMABA 

in acetonitrile which shows the anomalous fluorescence having a 

wavelength of maximum absorption around 305 nm and the normal 

fluorescence a value of 275 nm. This implies that the addition of water 

128

destroys the ground state species which is potentially forming the 

anomalous fluorescence. There was no observable change in the ground 

state absorption spectra of M4DMAB as a function of water added to a 

solution of acetonitrile. 

Fluorescence spectra were also measured for these solutions. The 

fluorescence spectra for the 4DMABA solutions were excited at the isosbestic 

point but as no change in the ground state absorption spectrum could be 

observed for M41DMAB an excitation wavelength of 306 nm was used. As 

water was added to a solution of M41DMAB in acetonitrile, the anomalous 

fluorescence decreased whilst the normal fluorescence increased (Figure 5.14 

and Table 5.16). 

6 

5 

C 

0)4 

C 

0) 

u3 

C 

0) 

V 

0) 

Li 

0 

1z 1 

f 

I 

( \\ 

0% Water 

------1.0% Water 

2.0% Water 

-----3.0% Water 

- 5% Water 

10% Water 

15% Water 

0 4- 

-S 

310 360 410 460 510 560 610 


Figure 5.14 Fluorescence spectra of M4DMAB in acetonitrile as a function 

of percentage added water 

129

%H20 mt a * hit b* Ratio 

0 61.0 52 1.17 

1.0 33.0 49 0.67 

2.0 23.0 42 0.54 

3.0 152 39 0.39 

4.0 14.5 36 0.40 

5.0 9.0 23 0.39 

10.0 5.3 15 0.35 

20.0 3.0 12 0.25 

Table 5.16 Variation of fluorescence intensity of a* and b* peaks of 

M4DMAB in acetonitrile as a function of added water 

Similar data are presented in Figure 5.15 and Table 5.17 for 4DMABA. 

7 

0% Water 

6 

U, 5 

0) 

I . 

I 

• 

-----0.50% Water 

1.00% Water 

------1.50% Water 

.4. 

01 

U 

C 

013, 

_ 

rn 

I' 

I 

-- - 3.0% Water 

6.0% Water 

I- 2 

iz 

1 

. 

"S 

-----10.0% Water 

o i --- I I 

310 360 410 460 510 


560 610 

Figure 5.15 fluorescence spectra of 4DMABA in acetonitrile as a function 

percentage added water 

ilkis]

%H 20 Ab mt b* Xmaxa mt a * 

0 350 160 488 117 

0.25 354 218 487 84 

0.50 354 270 490 64 

0.80 353 295 490 53 

1.00 355 343 490 47 

1.50 354 429 490 44 

2.00 356 538 490 20 

3.00 358 591 430 170 

4.00 358 473 438 212 

6.00 360 263 450 211 

10.0 360 114 456 163 

15.0 358 81 460 121 

20.0 355 62 462 95 

30.0 352 49 470 61 

50.0 352 46 475 34 

75.0 353 33 500 10 

90.0 354 31 500 8 

Table 5.17 fluorescence spectra of 4DMABA in acetonitrile as a function 

of added %water 

The ratio of decrease of the a* over increase of the b* intensity is 1 2, 

implying that the anomalous fluorescence is due to an interaction between 

at least two molecules of 4DMABA rather than interaction with a polar 

acetonitrile molecule. The anomalous fluorescence disappears at =2.0% v/v 

water in acetonitrile (Figure 5.16). A new fluorescence peak is then observed 

at approximately 430 nm . With further addition of water, the a * 

fluorescence starts to decrease as the intensity of the new peak grows, 

starting at 430 nm and becoming substantially red shifted to 500 run when 

the solvent composition is biased in favour of water. This could be due to 

the change in the dielectric constant of the water / acetonitrile mixture. For 

a concentration of 1.06x10 -5 mol dm 3, a plot of the %H 20 up to 2% against 

both the a* and b* intensity yields a straight line plot (Figure 5.16). It is a 

point worthy of note that although the absorption spectra were unreadable 

at water concentration> 5%, the fluorescence spectra were satisfactory. 

131

.b500 

'0 

C 

400 

1 300 

200 

100 

0 

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 

Percentage added water 

Figure 5.16 Plot of b* and a* emission from 4DMABA in acetonitrile as a 

function of added water 

We infer that if the anomalous fluorescence of 4DMABA in 

acetonitrile deriyes from a dimer, the addition of water preferentially forms 

the anion and consequently the emergence of the new peak is probably a 

solvent I solute complex which changes position due to the change in the 

dielectric constant of the mixture. Although this evidence is not conclusive 

it does appear that the addition of water to acetonitrile destroys the dimer. if 

the anomalous fluorescence observed in water does not originate from a 

ground state dimer, the proposal of Jouvet 63 may be supported by these data. 

Finally, a few fluorescence decay profiles have been measured for 

4DMABA in acetonitrile as a function of added water content (Table 5.18) 

132

%H20 

XEx XEm 

Intensity 

x 2 

Ir l 

al 'C2 a2 'C 

mean 

o 308 350 24.13 1.04 1.63 

490 24.00 1.13 2.09 

0.5 307 350 20.04 1.07 1.49 

490 15.03 1.19 1.36 

1.0 283 350 130.0 1.19 1.70 0.5513 3.03 0.4487 2.30 

J18 

3.0 279 350 115.0 1.09 0.35 0.4259 3.02 0.5741 

430 40.00 1.18 3.08 

480 25.00 2.29 3.13 

10.0 281.6 350 22.00 1.71 2.02 

_ 

450 36.00 1.09 0.33 0.1055 2.09 0.8945 

Table 5.18 Variation of fluorescence intensity of a* and b* peaks of 

M4DMAB in acetonitrile as a function of added water together 

with fluorescence decay data 

The lifetime of the b* fluorescence remains roughly constant in the 

1.49 - 2.02 ns range. The lifetime of the a* band initially decreases from 2.09 

to 1.36 ns on the addition of 0.5% v/v water, but increases to approximately 

3.1 ns in the presence of 3% v/v water. The latter correlates with the 

observed shift in the a* peak from 490 to 430 nm. The data are admittedly 

sparse and incomplete, but tend to suggest an initial quenching of the a* 

state on the addition of water, followed by the formation of a new a* species 

(at water concentrations > 2% v/v) with a 3.1 ns lifetime. From these data, it 

is suggested that fluorescence decay profiles for these mixed solvent systems 

could be a useful source of data to help explain the observed fluorescence 

changes. 

5.8 Effect of temperature on the emission properties of 4-aminobenzoic 

acid, 4-N,N-dimethylaminobenzoic acid and their methyl esters 

Rettig and co-workers 86 have highlighted the useful information 

which may be obtained from variable temperature fluorescence spectral and 

kinetic measurements, particularly when the photochemistry involves 

TICT states. Measurements of the fluorescence properties of 4DMABA and 

133

M4DMAB in the temperature range 77-298K have therefore been 

undertaken. Some data have also been recorded for 4ABA and M4AB which 

do not exhibit the anomalous fluorescence since they may provide a 

baseline from which to judge the 4DMABA and M4DMAB data. 

Fluorescence spectra for the four compounds in the solvent mixtures 

methylcyclohexane / isopentane (4:1), chlorobutane / isopentane (9:1) and 

butyronitrile I isobutyronitrile (9:1) were measured in the temperature 

range 77 - 298K. Based on the quantum yield at ambient temperature, 

fluorescence quantum yields were calculated at temperatures down to 77K. 

However, no allowance has been made for absorption changes as the 

temperature is reduced 87 and the quantum yield changes which are 

recorded in Table 5.17 must be viewed with this in mind. 

CMD Solvent Quantum yield as a function of temperature 

293K 230K 193K 153K 133K 113K 

4ABA BuCN / iBuCN 0.36 0.62 0.94 1.00 1.00 1.00 

BC! / iP 0.28 0.28 0.35 0.80 1.00 1.00 

MCH / 11' 0.03 0.05 0.08 0.09 0.09 0.08 

M4AB BuCN / IBuCN 0.38 0.63 0.93 0.99 1.00 1.00 

BC! / iF 0.40 0.54 0.75 0.90 1.00 1.00 

MCH / iF 0.12 0.18 0.31 0.43 0.35 0.12 

4DMABA BuCN / iBuCN 0.003 0.003 0.004 0.004 0.006 0.02 

BC! / iP 0.04 0.05 0.08 0.12 0.18 0.25 

MCH / iF 0.16 0.15 0.026 0.019 0.013 0.03 

480nm BuCN / 1BuCN 0.031 0.022 0.015 0.08 0.08 0.04 

BC! I iF 0.12 0.07 0.04 0.04 0.03 0.03 

M4DMAB BuCN / IBuCN 0.005 0.007 0.008 0.010 0.012 0.008 

BC1 / iP 0.09 0.09 0.10 0.10 0.14 0.17 

MCH / iF 0.10 10.14 10.14 0.12 0.08 10.16 

480nm BuCN / 1BuCN 0.022 10.014 10.008 0.004 10.004 1 P 

BC! / iF 0.17 10.11 10.09 0.06 10.04 1 0.04 

Table 5.19 

fluorescence quantum yields of 4-aminobenzoic acid, 4-N,Ndimethylaminobenzoic 

acid and their methyl esters as a 

function of solvent and temperature 

For 4ABA and M4AB, the shape and position of the fluorescence 

spectra are largely invariant with temperature in the solvents used. There is 

some evidence for a slight red shift as the temperature is decreased, as 

134

typified by M4AB in methylcyclohexane / isopentane (4:1) (Figure 5.17), but 

the shift is only of the order of 10-20 nm. The apparent quantum yield 

changes reported in Table 5.19 are much more significant and there is a clear 

general trend for the quantum yields to increase as the temperature is 

decreased. One would therefore anticipate that the fluorescence lifetime 

values would perhaps exhibit a similar trend, the quantum yield increase 

possibly being a consequence of a decrease in the non-radiative decay rate. 

40000 

35000 

30000 

C 

25000 

20000 

15000 

,a10000 

5000 

¼' 

293K 

-- - --230K 

193K 

------153K 

133K 

\ 

113K 

-----77K 

0 

280 300 320 340 360 380 400 420 440 


Figure 5.17 Fluorescence spectra of M4AB in methylcyclohexáne / 


The data that are available at this point is conflicting. The 

fluorescence decays at reduced temperature have been recorded at the 

BESS? synchrotron source in Berlin and restricted access to the source 

together with instrumental problems have limited the quantity and quality 

of data which have been gathered. The most complete dataset is for 4ABA in 

chiorobutane / isopentane (9:1) (Table 5.20) which suggests that the 

fluorescence lifetime is effectively constant in the temperature range 77 - 298 

K. 

135

Solvent 

Species 

Temp 

K 

I 

Xem 

(nm) 

%2 

I 

t1(ns) 

I 

a1 

I 

t2(ns) 

mean 


Acetonitrile 293 331 1.08 1.14 

BC! / iP 293 315 1.22 0.49 0.7995 4.08 0.2005 1.21 

193 315 Poor 0.52 0.8404 3.94 0.1596 1.07 

153 315 1.26 1.36 0.9805. 3.04 0.0195 1.39 

113 315 1.66 0.93 0.7009 1.78 0.2991 1.18 

77 315 1.11 0.76 0.6495 1.58 0.3505 1.05 

Ethanol 293 345 1.09 0.28 0.9673 1.45 0.0327 0.32 

Hexane 293 31.02 1.01 0.9174 2.49 0.0826 1.13 

Water 293 0.63 0.79 

Anion 293 1.24 0.72 

Neutral 293 0.82 1 

Methyl 4-aminobenzoate 

Acetonitrile 293 333 1.03 1.13 

BCl/iP 293 330 1.19 1.27 

193 330 Poor 2.43 2.43 

153 330 Poor 2.38 2.32 

113 330 1.25 2.91 2.68 

77 330 1.12 0.75 0.7998 2.39 0.2002 0.97 

Ethanol 293 346 1.11 0.14 0.9959 1.48 0.0041 1.02 

Hexane 293 318 1.09 2.06 

Water 293 1353 1 1.07 1 1.34 1 

Table 5.20 fluorescence decay data for 4ABA and M4AB as a function of 

solvent, pH and temperature 

I 

a2 

Although most of the decay data presented appears to be double 

exponential, the decays are usually dominated by a component with a 

lifetime of approximately 1.0 ns. This may be confirmed by comparing 

with t mean when it is noticeable that there is only a slight difference 

between the two values. The decays are largely mono-exponential with a 

small, longer lived tail. A similar conclusion regarding the invariance of 

the fluorescence lifetime with temperature can be drawn for 4ABA in 

methylcyclohexane I isopentane (4:1) although the data are sparse. The 

methyl ester, on the other hand, does appear to have an increased 

fluorescence lifetime at reduced temperature in both chiorobutane I 

isopentane (9:1) and methylcyclohexane I isopentane (4:1) (Table 5.20). It is 

136

possible that 4ABA and M4AB behave differently in terms of their 

fluorescence properties as a function of temperature, but in the absence of 

the absorption data highlighted earlier, it is impossible to say, or to confirm 

that the apparent quantum yield increases are real. This is particularly so for 

4ABA. 

The corresponding N,N-dimethylamino compounds, 4DMABA and 

M4DMAB, exhibit anomalous fluorescence in polar solvents. The 

fluorescence properties of both compounds were studied in the three 

solvent mixtures methylcyclohexane I isopentane (4:1), chlorobutane / 

isopentane (9:1) and butyronitrile / isobutyronitrile (9:1); three mixtures 

which represent low, intermediate and high polarity respectively and which 

form good quality glasses at low temperature. Fluorescence spectra and 

quantum yields were measured for both compounds in all three solvents in 

the temperature range 113-298K and the quantum yield values are given in 

Table 5.19. Both compounds produce anomalous fluorescence in 

butyronitrile / isobutyronitrile (9:1) and chlorobutane / isopentane (9:1), but 

in the low polarity alkane mixture only normal fluorescence is observed at 

all temperatures. This contrasts with the reports of anomalous fluorescence 

from ethyl 4-N,N-diethylaminobenzoate (E4DEAB) in low polarity 

solvents 51 . It is perhaps surprising that the replacement of the three methyl 

groups in M4DMAB with ethyl groups should have such a substantial effect. 

Like their amino counterparts, 4DMABA and M4DMAB appear to 

exhibit an increasing quantum yield of normal fluorescence as the 

temperature decreases. Once again, some or all of this increase may be 

attributed to increased sample absorbance at lower temperatures. However, 

it is clear that the steady decrease in the quantum yield of the anomalous 

fluorescence with decreasing temperature cannot be explained in a similar 

manner. Indeed, if the quantum yield values are being enhanced by 

increased sample absorbance, then correcting for this will cause the 

137

plots are found to be reasonably linear (Figure 5.19 and 5.20) and the slopes 

of the plots yield activation energies in the range 2.5 - 6.0 kJ mol 1 (table 

5.22). Rettig 88 has also derived similar linear plots for 4DMABN in 

propionitrile and capronitrile solvents giving values of 7.5 and 11.0 kj mo1 1 

respectivly. 

4DMABA 

M4DMAB 

Temp lIT BuCN / 1BuCN BC! lip BuCN / iBuCN BCI lip 

Lfl(4 a /4b) Ln(4a/4b) Ln®a/Gb) Ln(a/4b) 

293 0.003413 2.335 1.100 1.482 0.636 

230 0.004348 1.992 0.336 0.693 0.201 

193 0.005183 1322 -0.693 0.000 -0.105 

153 0.006536 2.996 

1 

-1.100 

1 

-0.916 -0.511 

133 10.007519 2.590 1 -1.790 1 -1.099 -1.253 

113 10.008849 0.069 1 -2.120 1 - -1.447 

Table 5.21 Variance of the natural log of the ratio of the two fluorescence 

peaks of 4DMABA and M4DMAB in chlorobutane / 

isopentane (9:1) and butyronitrile / isobutyronitrile (9:1) 

Solvent 

Compound BuCN / IP 

kJ mol-1 

BC1 / ip 

kJ mo1 1 

4DABA 4.15 5.91 

M4DMAB 6.47 3.96 

Table 5.22 Activation energies calculated from Figure 5.18 for 4DMABA 

and figure 5.19 M4DMAB in chlorobutane / isopentane (9:1) 

and butyronitrile / isobutyronitrile (9:1) 

139

3 — 

. 

2 

CY 

•: 

.5 • 

0.( 3 0.004 0.00..._ 

- 

0.006 0.007 0.008 0.009 

- BCN/IBCN -- 

-2 BestELBCN/1BCN 

• SCl/Ip 

-.-- . 

--ScstfitBCl/ip --- 

1/ T (IC-i) 

Figure 5.18 Arrhenius type plots of Ln(4/$b) versus liT for 4DMABA in 

chiorobutane / iso pentane (9:1) and butyronitrile / 

isobutyronitrile (9:1) 

-o 

a' 

a' 

C 

1.5 

1 

0.5 

0 

.t -0.5 

-1 

-1.5 

-2 

-2.5 

N 

e 

0.004 ~ 

• BCN/ IBCN 

Best fit BCN / IBCN 

• BCL/il' 

- 

Best fit BCI / ip 

0.007 0.008 0.009 

_Szczzz 

- 

ii T (K-i) 

Figure 5.19 Arrhenius type plots of Ln(4a/$b) versus 1ff for M4DMAB in 

chlorobutane I iso pentane (9:1) and butyronitrile I 

isobutyronitrile (9:1) 

140

The activation energies calculated by Rettig 71 are quite close to the 

activation energies for the viscous flow of these two solvents, allowing the 

conclusion that the rate determining step, and hence the main determinant 

of the temperature dependence, is the viscosity-dependent intramolecular 

twisting step (rate constant K1) which produces the a* state. The activation 

energies derived from our plots are generally considerably lower than the 

activation energies for viscous flow of chlorobutane (7.1 kJ mol 1 ) and 

butyronitrile (9.3 kJ mol 1). In view of our proposal that the anomalous 

fluorescence derives from ground state dimer formation, the activation 

energies derived from our plots of ln®a / tb) versus lIT perhaps relate to 

the monomer 4—> dimer equilibrium in the ground state 

An alternative interpretation of our observations and many "TICT - 

like" phenomena is that of Cazeau-Dubroca et al. 87, who proposed that the 

anomalous fluorescence from molecules such as 4DMABN derives from a 

twisted ground state, the formation of which is promoted by the presence of 

small amounts of water in the solvent. Absorption and fluorescence spectral 

changes at ambient and reduced temperature are cited to back up the 

proposal. We have been conscious throughout the work reported in this 

thesis that water contamination of the solvents may lead to spurious 

results, but we have not observed any obvious differences between samples 

where the solvent has been scrupulously dried and those where it has not. 

In addition, a mechanism involving a twisted ground state would not 

account for the concentration dependence that is observed for the 

anomalous fluorescence of 4DMABA and M4DMAB. We therefore believe 

that this mechanism is not operating in our systems. 

Fluorescence decay profiles have been measured for 4DMABA and 

M4DMAB in the three solvent systems in the temperature range 77 - 298K. 

The results of'these measurements agree given in Tables 5.23 - 5.25. 

141

Solvent Tern Aem V t1(ns) a1 12(ns) a2 t(ns) a3 


BuCN / iBuCN 293 350 1.11 1.51 0.9590 5.86 0.0410 __ 

230 345 1.99 0.021 0.9456 1.65 0.0334 6.67 0.0162 

193 345 1.17 0.095 0.9605 1.71 0.0382 9.76 0.0013 

153 345 1.32 0.39 0.5289 1.38 0.4641 4.71 0.0007 

113 350 1.29 1.71 0.9806 6.98 0.0194 

77 360 1.90 1.82 0.7479 11.11 0.2521 

293 485 1.54 2.39 

193 490 1.49 1.05 0.4477 5.78 0.5523 

153 495 1.45 0.25 0.7807 1.88 0.2193 

133 490 0.97 2.30 0.5869 4.58 0.4131 

77 440 1phos 

BC! / iP 293 355 1.24 1.15 0.4470 3.78 0.5510 

193 365 n/c 

153 365 1.48 0.06 0.9731 1.58 0.0268 9.63 0.0001 

133 350 1.12 0.36 0.7274 1.40 0.2717 4.35 0.0009 

77 355 n/c 

293 420 1.30 3.25 0.2218 4.82 0.1814 0.04 0.5968 

193 420 1.17 2.55 0.1648 15.24 0.0967 0.19 0.7385 

153 420 1.05 1.77 0.5096 23.82 0.4904 

133 420 1.19 1.05 0.5522 5.18 0.3444 70.6 0.1042 

77 420 0.99 0.26 0.7652 5.06 0.0841 53.8 0.1507 

MCH / iP 293 334 1.74 1.18 0.9923 4.93 0.0077 

230 334 2.46 0.38 0.6751 2.56 0.3249 

193 334 1.11 1.04 0.8276 2.52 0.1724 

153 334 Ct> 0.93 

133 334 1.17 0.82 0.6275 3.08 0.3725 

113 334 1.48 1.06 0.6543 6.78 0.4566 

77 334 1.37 1.38 0.4210 4.97 0.0053 1.37 0.5737 

293 370 1.43 1.29 0.9919 6.03 0.0081 

193 370 1.22 1.68 0.9581 3.67 0.0419 

153 370 1.34 2.10 0.1909 4.48 0.0680 0.17 0.7411 

133 370 n/c 

113 370 1.48 1.06 0.2134 6.78 0.0161 0.31 0.7706 

77 370 1.15 1.53 0.6904 7.46 0.0075 0.34 1 0.3021 

Phos = Phosphorescence occurring and fluorescence not measurable, and 

n/c means that the decay profile could not be fitted 

Table 5.23 Lifetime data of 4DMABA as a function of pH, solvent and 

temperature 

142

Solvent 

I Tern Acm 

pK 

(Nfl) 

XI I 11(ns) a1 t2(ns) a2 t3(ns) a3 

ethyl 4-N,N-dimethvlaminobenzoate 

BuCN / IBuCN 293 340 1.23 1.07 0.9930 3.54 0.0070 

193 340 1.37 1.75 0.4089 3.51 0.5911 

153 340 1.05 1.68 0.9956 9.94 0.0044 

133 340 1.33 1.71 0.9802 8.62 0.0198 

113 340 1.29 1.85 0.9644 6.40 0.0356 

77 340 1.28 1.81 0.9901 8.02 0.0099 

293 480 0.93 0.40 0.1721 3.00 0.8279 

193 480 2.26 3.96 0.5239 7.32 0.4761 

153 480 1.03 2.41 0.9464 6.01 0.0536 

133 480 1.04 3.88 0.9295 6.90 0.0705 

113 480 1.07 4.25 0.7604 51.75 0.2396 

77 480 n/c 

BC! / iP 293 345 1.11 1.12 0.3416 3.65 0.6584 

193 345 1.18 1.68 0.9895 5.81 0.0105 

153 345 1.41 0.10 0.9137 1.76 0.0863 

133 345 1.59 1.44 0.9840 5.55 0.0160 

77 345 1.29 1.57 0.8808 5.03 0.1182 

293 420 1.06 3.52 0.8429 5.27 0.1571 

193 420 1.42 4.22 0.3333 8.18 0.0944 0.35 0.5723 

153 420 1.07 2.92 0.6511 8.75 0.3489 

133 420 1.03 2.37 0.7568 7.83 0.2432 

77 420 1.00 2.32 0.3000 256.9 0.2619 0.04 0.4381 

MCH / iP 293 334 1.19 0.95 0.9986 7.61 0.0014 

193 334 1.06 2.06 0.9955 9.09 0.0045 

153 334 Poor 2.47 0.8726 4.43 0.1274 

133 334 1.35 1.12 0.0496 3.25 0.0270 0.23 0.9234 

113 334 1.14 0.98 0.3361 5.42 0.0045 0.35 0.6594 

77 334 1.60 1.37 0.9852 4.46 0.0148 

293 370 1.40 0.98 0.9525 6.98 0.0475 

193 370 1.28 3.99 0.8643 9.75 0.1337 

153 370 1.20 2.30 0.3625 9.53 0.6375 

133 370 1.18 1.52 0.3312 3.54 0.3364 0.32 0.3324 

113 370 1.42 0.78 0.5023 2.11 0.4941 12.7 0.0036 

___ ____ 

77 370 1.09 1.40 0.2210 1 2.77 0.7748 9.07 0.0042 

Phos = Phosphorescence occurring and fluorescence not measurable, and 

n/c means that the decay profile could not be fitted 

Table 5.24 Lifetime data of M4DMAB as a function of pH, solvent and 

temperature 

143

Solvent Temp Xem 4DMABA M4DMAB 

K tmean 'U mean 

BuCN / iBuCN 293 340 1.69 1.09 

193 340 1.97 1.45 

153 340 1.85 1.72 

133 340 1.81 1.85 

77 340 4.16 1.87 

293 480 2.39 1.27 

193 480 3.66 2.28 

153 480 0.61 1.30 

133 480 3.24 2.05 

BC1 / iF 293 345 2.60 2.78 

193 345 - 1.78 

153 345 1.61 1.51 

133 345 1.41 1.98 

77 345 - 1.52 

293 420 1.62 3.79 

193 420 2.03 2.38 

153 420 12.58 4.95 

133 420 9.72 3.70 

77 420 8.73 6.80 

MCH / iP 293 334 1.25 0.96 

193 334 0.08 2.09 

153 334 2.45 2.72 

133 334 3.30 0.36 

113 334 3.15 0.58 

77 370 1.39 1.42 

Table 5.25 Mean lifetimes of 4DMABA and M4DMAB as a function of 

solvent and temperature 

All the profiles rise promptly and there is no evidence for precursors 

to the emitting state in any of the recorded profiles. The decay profiles are 

largely mono-exponential at room temperature, but become more complex 

as the temperature is reduced. As the fluorescence spectra of the compounds 

only effectively change in terms of their overall intensity as the temperature 

is reduced, the reason for the increased complexity of the decay profiles is 

not obvious. In many cases the profiles are still largely mono-exponential, 

but in all cases where more than one exponential component is needed to 

adequately fit the data, the mean lifetime has also been calculated to aid 

comparison of data at different temperatures. 

144

The fluorescence lifetime ('C or 'C mean) of 4DMABA in 

methylcyclohexane / isopentane (4:1) appears to be effectively independent 

of temperature. There is one anomalous dataset at 113K, but virtually all the 

other values lie in the region 0.5 - 2.5 ns. It may be possible to draw the same 

conclusion for M4DMAB in this solvent, but the data are much more 

scattered. Figure 5.21 gives a picture of the variation of the lifetime values ('t 

or I mean) for M4DMAB with temperature in methylcyclohexane / 

isopentane(4:1) and lends some credence to the possibility that the lifetime 

is independent of temperature. This conclusion would certainly be in accord 

with the M4DMAB quantum yield which also appears to be temperatureindependent. 

Given the constant lifetime for 4DMABA in this low polarity 

solvent mixture, the large decrease in quantum yield (Table 5.19) appears 

out of place, especially when compared with the behaviour of the ester. The 

decrease in the 4DMABA quantum yield could reflect the lack of solubility 

of this compound in non-polar solvents - a point which has been made 

earlier about the room temperature absorption measurements. Lowering 

the temperature would be expected to exacerbate these solubility problems. 

The behaviour of the two compounds in the butyronitrile mixture appears 

to be quite similar. As Figures 5.21 - 5.24 illustrate, the lifetime of the 

normal fluorescence appears to be largely independent of temperature 

whilst that of the anomalous fluorescence increases somewhat as the 

temperature is reduced. 

145

4.50 

MKII 

2.50 

ii : 

"S 

77 113 133 153 193 233 293 

Temperature (K) 

Figure 5.20 Variation of lifetime values observed for M4DMAB in 

methylcyclohexane / isopentane (4:1) 

Rotkiewicz et al. 71 have found that 4DMABN and some related 

compounds form dimers and higher aggregates in non-polar solvents at 

reduced temperatures. These cause changes in the fluorescence spectra and 

lead to multi-exponential decay profiles. In view of the fact that we also 

observe the latter effect, the formation of these polymeric species in our 

systems cannot be ruled out. However, these dimers or higher aggregates 

either do not fluoresce (unlike the species reported by Rotkiewicz et al.) or 

they fluoresce weakly in the same region of the spectrum as the monomers. 

The large Stokes shift for the anomalous fluorescence attributed to dimer 

formation in more polar solvents is not repeated here. A substantial 

amount of further work is clearly needed to clarify the behaviour of 

4DMABA in non-polar solvents. 

146

The comments made above about the complexity of the fluorescence 

decays for 4DMABA and M4DMAB in non-polar solvents also apply to the 

other two solvent mixtures used here (Tables 5.23 - 5.25). 

too 

• a' -----Best fin' • V ------Bestfltb' 

£3.00 

2.50 

U 

2.00 

1.50 

g 1.00 

0.50 

0.00 

70 120 170 220 270 


Figure 5.21 The variation of the fluorescence lifetime with temperature for 

normal and anomalous fluorescence of 4DMABA in 

butyronitrile / isobutyronitrile (9:1) 

'I 

C' 

C 

I) 

U 

3.50 E-02 

3.00E-02 

' 2,50E-02 

2.00E.02 

1.50&02 

2 1.00E-02 

0 

• a' 

-----Best fit a' 

- 

• V 

- - - - 

- - - - -- Best fit V - --- 

1 

- - - 

--- • 

-- 

• 

W. 

5.00E-03 

• ----- 

O.00E+® 

70 120 170 220 270 


Figure 5.22 The variation of the fluorescence quantum yield with 

temperature for normal and anomalous fluorescence of 

4DMABA in butyronitrile I isobutyronitrile (9:1) 

147

4.50 

4.00 

3.50 

S 

• a• 

Bnt (It . 

• b' 

— Bt(Itb' 

.5 3.00 

E 

2.50 

—1 

4, 

2.00 

e 

- - — 

S 

—a 

a 1.50 

EC 

1.00 

0.50 

0.00 

70 

120 170 220 270 


Figure 5.23 The variation of the fluorescence lifetime with temperature for 

normal and anomalous fluorescence of M4DMAB in 

butyronitrile / isobutyronitrile (9:1) 

0.025 

•o 0.02 

• a 

----Best fit a 

6 

- 

4, • b' 

- 

E 

------ Bestfitb 

0.015 

.-- 

0.01 

4 

0.005 

. 

-- 

o I I 

70 120 170 220 270 


Figure 5.24 The variation of the fluorescence quantum yield with 

temperature for normal and anomalous fluorescence of 

M4DMAB in butyronitrile / isobutyronitrile (9:1) 

The latter effect is not large and the scatter in the data for the ester 

certainly invites scepticism as to the validity of this observation. However, 

148

an increasing lifetime for the a* fluorescence would certainly fit in with the 

observed decrease in the fluorescence yield and would point to a 

temperature dependence for the radiative rate constant for the a* excited 

state. In view of the low fluorescence quantum yields which are observed in 

the mixed butyronitrile solvent, the vast majority of the excitation energy is 

dissipated non-radiatively. Therefore it is possible to get large changes in 4, 

yet much smaller changes in the lifetime, since K, (or K) >> K 1 (or K). 

There will presumably also be a change in the monomer

concentration effects. Taken all together, the measurements made in this 

project and reported in this thesis point to a ground state origin for the 

anomalous fluorescence from M4DMAB and 4DMABA. 

5.9 Summary of conclusions 

The absorption and fluorescence spectra of 4ABA and M4AB do not exhibit 

any obviously anomalous properties. They are always some 20 nm blue 

shifted from their N, N- dime thylamino counterparts. The latter, however, 

exhibit anomalous fluorescence properties in some solvents. Despite the 

lack of concentration effects on the absorption spectra of 4DMABA and 

M4DMAB in acetonitrile in the concentration range 10 - iO mol dm 3, the 

anomalous fluorescence is attributed to ground state dimer formation. 

Evidence for this was derived from fluorescence excitation spectra, 

fluorescence decay profiles and the effect of temperature on the fluorescence 

properties of these molecules. Other mechanisms such as excimer, exciplex 

or TICT state formation were clearly at odds with the experimental data. 

However, some preliminary experiments on the effects of adding either 

small quantities of acetonitrile to hexane solutions of 4DMABA and 

M4DMAB or water to acetonitrile solutions of these two compounds suggest 

that more than one route to the anomalous fluorescence may be operating. 

150

Chavter 6

References 

1 R.B.Cundall and A.Gilbert, "Studies in modern photochemistry", 

p.2, Nelson, London, (1970) 

2 A.Streitweiser, "Molecular Orbital Theory for Organic Chemists", 

John Wiley and Sons, London, (1961) 

3 L.C.Jones and L.W.Taylor, Anal., Chem., 2Z, 228, (1955) 

4 J.A.Barltrop and J.D.Coyle, "Principles of Photochemistry", p.30, 

John Wiley and Sons, London, (1978) 

5 H.L.McMurry, J.Chem.Phys., 9, 231, (1941) 

6 D.F.Evans, J.Chem.Phys.. fl 1424, (1955) 

7 G. Briegleb, "Eletronen-Donator-Komplexe", Springer, Berlin, (1961) 

8 J.D.Coyle, "Introduction to Organic Photochemistry", John Wiley 

and Sons, Bristol, (1986) 

9 J.G.Calvert and J.N.Pitts, "Photochemistry", John Wiley and Sons, 

London, (1976) 

10 K.W.Hodgson, PhD thesis C.N.A.A., (1986) 

151

11 J.Franck, Trans.Faraday Soc., a 536, (1926) 

12 P.M.Morse, Phys.Rev., L4. 57, (1929) 

13 K.K.Rohatgi-Mukherjee, "Fundamentals of Photochemistrv", John 

Wiley and Sons, India, (1977) 

14 R.S.Mulliken, T.Chem.Phys., 1 14, (1939) 

15 J.B.Birks, "Photophysics of aromatic molecules", John Wiley and 

Sons, London, (1969) 

16 I.B.Berlman, "Handbook of Fluorescence Spectra of Aromatic 

Molecules", Academic Press, New York, (1971) 

17 B.M.Wepster, Rec.Trav.chim.Pays-Bas. Zñ 335, (1957) 

18 B.M.Wepster, Rec.Trav.chim.Pays-Bas. a 357, (1957) 

19 B.M.Wepster, Rec.Trav.chim.Pays-Bas. ZZ, 491, (1958) 

20 J.Burgers, M.A.Hoefnagel, P.E.Verkade, H.Visser and B.M.Wepster, 

Rec.Trav.chim.Pays-Bai, fl 269, (1958) 

21 E.Drent, Chem.Phys.Lett., 2. 526, (1968) 

22 C.A.Parker and C.G.Hatchard, Proc.Roy.Soc., A269. 574, (1962) 

152

23 C.A.Parker, Adv.Photochem., 9, 1, (1974) 

24 Th.Forster and K.Kasper, Z.Phys.Chem. N.F.), 1, 19, (1954), and 

Bunsenges.. Phys.Chem., a 977, (1955). 

25 E.Lippert, W.Luder, H.Boos,: "Advances in Molecular 

Spectroscopy", Ed. A.Mangini, Pergamon Press, Oxford, (1962) 

26 Z.R.Grabowski, K.Rotkiewicz and K.H.Grellman, Chem.Phys.Lett, 

19 315, (1973) 

27 W.Rettig, J.Phys.Chem.. , 1970, (1982) 

28 W.Rettig, "EPA Newsletter" No 41, p.3, (1991) 

29 G.R.Alms, D.R.Bauer, J.LBrauman and R.Pecora, T.Chem.Phys., a 

5310, (1973) 

30 W.Rettig, Angew.Chem.lnt.Ed.Engl. fl 971 (1986) 

31 R.P.Wayne, "Principles and Applications of Photochemistry", 

Oxford Science Publications, (1988) 

32 D.L.Dexter, T.Chem.Phys., a, 838, (1953) 

33 F.Wilkinson, Adv.Photochem.. 3. 241, (1965) 

153

34 J.G.Calvert and J.N.Pitts, 'Photochemistry", p.88, JohnWiley and 

Sons, London, (1966) 

35 E.Vander Donckt, Progr.Reaction Kinetics 5. 273, (1970) 

36 L.Doub and J.M.Vandenbelt, T.Am.Chem.Soc., 59, 2714, (1947) 

37 L.Doub and J.M.Vandenbelt, T.Am.Chem.Soc., 71, 2414, (1949) 

38 L.Doub and J.M.Vandenbelt, Organic and Biological Chemistry, ZZ, 

4535, (1955) 

39 W.H.Melhuish, T.Phys.Chem. 55, 229, (1961) 

40 N.Mataga and M.Ottolenghi, "Photochemical Aspects of 

Exciplexes", 

41 P.Leggate and G.E.Dunn, Can.T.Chem., 41 1158,(1965) 

42 A.Tramer, J.Mol.Structure., £ 313, (1969) 

43 R.Kuhn and K.Geider, Chem.Ber., ]&j, 3597, (1968) 

44 A.Tramer, T.Phys.Chem., 74 887, (1970) 

45 D.V.S.Jian, F.S.Nandel, P.Singla and D.J.Keiv, Indian I.Chem. 25A. 

15,(1986) 

154

46 D.J.Cowley and P.J.Healy, Proc.IR.Acad.. 77B. 397, (1977) 

47 D.J.Cowley, P.J.Healy and A.H.Peoples, T.Photochem., 9, 240, (1978) 

48 J.A.T.Revill and R.G.Brown, Chem.Phys.Lett.. it 433, (1992) 

49 R.Howell, A.C.Jones, A.G.Taylor and D.Phillips, Chem.Phys.Lett., 

163, 282,(1989) 

50 R.J.Visser, P.C.M.Weisenborn and C.A.G.O.Varma, T.Chem.Phys. 

113,330, (1984) 

51 A.Nag, TdCundu and K.Bhattacharya, Chem.Phys.Lett.. lift. 257, 

(1989) 

52 J.A.T.Revill and R.G.Brown, 1.Fluorescence. L 107, (1992) 

53 P.C.M.Weisenborn, A.H.Huizer and C.A.G.O.Varma, Chemical 

Physics. 1Z51 425, (188) 

54 Z.R.Grabowski, K.Rotkiewicz, A.Siemiarczuk, D.J.Cowley and 

W.Baumann. Nouveau T.Chimie. 3., 443, (1979) 

55 E.M.Kosower and H.Dodiuk, J.Am.Chem.Soc., 98. 924, (1976) 

56 G.Wermuth, W.Rettig and E.Lippert, Ber.Bunsenges.Phys.Chem.. 

~85 64, (1981) 

155

57 W.Rettig and R.Gleiter, J.Am.Chem.Soc. 9, 4676, (1985) 

58 F.Heisel, J.A.Miehe and J.M.G.Martinho, Chemical Physics., 243, 

(1985) 

59 F.Heisel and J.A.Miehe, Chemical Physics, 9, 233, (1985) 

60 D.Huppert, S.D.Rand, P.M.Rentzepis, P.F.Barbara, W.S.Struve and 

Z.R.Grabowski, T.Photochem., fl 89, (1982) 

61 K.Rotkiewicz, Z.R.Grabowski and J.Jasny, T.Chem.Phys., 5.0 55, 

(1975) 

62 T.Kobayashi, M.Futakami and O.Kajimoto, Chem.Phys.Lett.. 

63,(1986) 

63 J.Augüst, T.F.Palmer, J.P.Simons, C.Jouvet and W.Rettig, 

Chem.Phys.LetL 145,273,(1988) 

64 N.Nakashima and N.Mataga, Bull.Chem.Soc.Tapan. 4.5, 3016, (1973) 

65 O.S.Khahil, R.H.Hofeldt and S.P.McGlynn, Chem.Phys.Lett., U, 479, 

(1972) 

66 E.A.Chandross, "The Exciplex. Conf. Proc.". New York Academic 

Press, (1975) 

67 P.Suppan, Chem.Phys.Lett. 1 28,160,(1986) 

156


Physics. 126, 425,(1988) 


Physics. ifl 437, (1989) 

70 R.J.Visser, P.C.M.Weisenborn and C.A.G.O.Varma, 

Chem.Phys.Lett. ID, 330, (1985) 

71 K.Rotkiewicz, H.Leismann and W.Rettig, T.Photochem. and 

Photobiol., A. 49, 347, (1989) 

72 W.Rettig, Ber.Bunsenges.Phys.Chem.. 25, 259, (1991) 

73 O.Noboru, Bull.Chem.Soc.Tapan. a 2401, (1978) 

74 G.Hallas, i.Am.Chem.Soc., 5, 271, (1936) 

75 A.Albert and E.P.Serjeant, "The Determination of Jonisation 

Constants", p94, Chapman and Hall Ltd, Edinburgh, (1971) 

76 R.A.Velapoldi and K.D.Mielenz, NBS Special Publication. 260-64, 

(1980) 

77 J.N.Demas, T.Phys.Chem., fl, 991, (1971) 

157

78 D.Phillips and D.V.O'Connor, "Single Photon Counting", 

Academic Press, New York, (1984) 

79 R.Sparrow, R.G.Brown, E.H.Evans and D.Shaw, J.Chem.Soc., 

Faraday Trans II., fl 2249, (1986) 

80 H.H.Jaffe and M.Orchin, "Theory and Applications of Ultraviolet 

Spectroscopy", Wiley, New York, (1963) 

81 A.Weller, Progr. Reaction Kinetics.!, 189, (1961) 

82 P.Sykes, "Mechanisms in Organic Chemistry", p56, Longman, 

London, (1981) 

83 W.T.Stacey and C.E.Swenberg. T.Chem.Phys.,, 1962, (1970) 

84 T.P.Carsey, G.L.Findley and S.P.McGlynn, J.Am.Chem.Soc., J&IJ 

4502, (1979) 

85 J.Herbich, W.Rettig, R.P.Thummel and J.Waluk, Chem.Phys.LflL 

195,556,(1992) 

86 W.Rettig, W.Majenz, R.Lapouyade and M.Vogel, T.Photochem. and 

Photobiol., A,4, 49, (1992) 

87 C.Cazeau-Dubroca, S.A.Lyazidi, P.Caubou, A.Perigina, Ph.Cazeau 

and M.Pesquer, LPhys.Chem., fl 2347, (1989) 

LM;]

88 W.Rettig, J.Lumim, a 21, (1981) 

159

Appendix A

2 msc = 3100 REM value = to 1 volt 

4 pfin=28 

6 gyfin=350 

20 DIM 1(700) 

30 DIM WAVENUMBER(400) 

40 DIM FLUOR(400) 

50 MARK=0 

60 PRT=0 

90 GOSUB ten: 

120 CLSO: 

125 PRINT 

130 PRINT '* MENU 6/9/92 

140 PRINT "* (1) LOAD SAMPLE DATA " 

150 PRINT "* (2) CALCULATE AREA 

160 PRINT "* (3) RUN SAMPLE 

170 PRINT '* (4) QUANTUM YIELD 

180 PRINT "* (5) PRINT PARAMETERS 

215 PRINT "* e(6) DIRECTORY *" 

217 PRINT "* (7) QUIT * 

220 PRINT 

240 INPUT "YOUR CHOICE PLEASE"; C$ 

250 if C$ - "1" THEN GOSUB 2180 

260 if C$ - "2" THEN GOSUB 2720 

270 if C$ - "3" THEN GOSUB 1470 

280 if C$ - "4" THEN GOSUB 3700 

290 if C$ - "5" THEN PRT - 0 GOSUB 920 

325 if C$ - "6" TI-lEN GOSUB 3850 

327 if C$ "7" THEN END 

330 GOTO 120 

410 REM SUB GRAPH 

420 if MARK = 0 THEN PRINT "SORRY NO DATA LOADED ": CLS: 

GOTO 120 

430 CLS GOSUB ten 

440 PSET (50, 0) 

450 LINE (50, 5)-(610, 295), 1, B 

451 LOCATE 6, 1 PRINT "I" 

452 LOCATE 7, F PRINT 'n" 

453 LOCATE 8, 1: PRINT "t" 

454 LOCATE 9, 1: PRINT "e" 

455 LOCATE 10, 1: PRINT "n' 

456 LOCATE 11, F PRINT "s' 

457 LOCATE 12, 1: PRINT "i' 

458 LOCATE 13, 1: PRINT "F' 

459 LOCATE 14, 1: PRINT "y" 

465 LOCATE 21, 30: PRINT "Wavelength (nm)' 470 FOR x = 50 TO 610 

STEP 40 

480 LINE (x, 295)-(x, 290), 1 

1

485 LINE (x, 5)-(x, 10), 1 

490 NEXT x 

500 FORy=3OTO28OSTEP2S 

510 LINE (50, y)-(55, y), 1: LINE (605, y)-(610, y), 1 520 NEXT y 

530 x=0 

550 FORI=0T07 

600 LQCATE20,6+(I*10) : PR1NTINT(START+x+.5)610x=x+ 

(FINISH - START) / 7 

620 NEXT I 

625 x=0 

630 FORI=9TOOSTEP4 

640 LOCATE(I2)+1,2:PR1NTINT(x*10+.5)/ 10 

650 x=x+sen/9 

660 NEXT 

720 LOCATE pfin - 6, 1: PRINT "Date "; DT,, , "File "; title$ 

730 LOCATE pfin - 4, 1: PRINT "Name "; NAM$ 

750 LOCATE pfin - 5, 1: PRINT "Solvent "; SOLVENTs,,, 

"Concentration "; CONC 770 RETURN 

920 REM SUB PARA 

930 IF MARK = 1 THEN 990 

940 SLIT = 2.5: START = 380 

950 SCANSPEED = 240 

960 EXT = 2.5: WEXT = 340 

970 FINISH = 650 

980 sen=1 

985 errr = .935 

990 CLS 

1000 LOCATE 3, 5: PRINT" Parameter settings "1010 LOCATE 4, 5: 

PRINT 

1020 LOCATE 6, 1: PRINT" All following must be in order" 

1030 LOCATE 8, 1: PRINT "Emission slit ----------------- "; SLIT;" nm' 

1040 LOCATE 9, 1: PRINT "Emission wavelength start ----- "; START;" 

run" 1050 LOCATE 10, 1: PRINT "Emission wavelength finishing - 

FINISH;" nm' 

1060 nreadings = FINISH - START 

1070 LOCATE 11, 1: PRINT "Excitation slit --------------- "; EXT; " nm" 

1080 LOCATE 12, 1: PRINT "Excitation wavelength --------- "; WEXT;" 

nm" 

1090 LOCATE 13, 1: PRINT "Scan speed -------- ------"; SCANSPEED;" 

sec" 

1100 LOCATE 14, 1: PRINT "Sensitivity (Scale) ----------- "; sen 

1120 if PRT = 1 THEN GOTO 1310 

1130 MARK = 10 

1140 LOCATE 16, 1: PRINT "Do you wish to alter any parameters? 

(YIN)" 

1141 A$ = INKEYS 

1142 IF A$ = "Y" OR A$ = "y" THEN 1210

1143 IF A$ = "N" OR A$ = 'n" THEN 1280 

1144 GOTO 1141 

1210 LOCATE 8,32: INPUT SL: if SL c> 0 THEN SLiT = SL 

1220 LOCATE 9,32: INPUT ST: IF ST c> 0 THEN START = ST 

1230 LOCATE 10, 32: INPUT Fl: IF Fl 0 THEN FINISH = FT 

1240 LOCATE 11,32: INPUT EX: IF EX 0 THEN EXT = EX 

1250 LOCATE 12, 32: INPUT WE: IF WE 0 THEN WEXT = WE 

1260 LOCATE 13, 32: INPUT sc: if sc 0 0 THEN SCANSPEED = sc 

1270 LOCATE 14, 32: INPUT SE: IF SE o 0 THEN sen = SE 

1280 GOSUB 2060: 

1285 LOCATE 18, 1: PRINT "DO YOU REQUIRE A PRINT OUT 

1286 A$ = INTCEY$ 

1287 IF A$ - "1" OR A$ - "y" THEN GOSUB 5000 

1288 IF A$ = "N" OR A$ = "n" THEN 1310 

1290 GOTO 1286 

1300 C1.S 

1310 MARK = 1 

1320 RETURN 

1470 REM SUB DATA 

1480 GOSUB 3210 

1490 GOSUB 920 

1500 GOSUB 410 


1520 YSCALE = 290 / (msc - 2048) 

1530 XSCALE = (610 - 50) I nreadings 

1540 DLAY = SCANSPEED I 60 

1545 GOSUB 2720 

1550 LOCATE pfin - 1, 10: PRINT "Align parameters on Fluorimeter 

with computer" 

1560 LOCATE pfin, 13: PRINT "When ready start Fluorimeter" : A = 0 

1564 than = arrayl(1) 

1570 OUT (&H303), 1 

1572 OUT (&H302), 0 

1574 FORI=1TO100 

1576 NEXT I 

1578 10 = INP(&H300) AND &HFF 

1580 hi = INP(&H301) AND &HF 

1582 SPIKE = ((hi * 256) + 10) AND &HFFF 

1584 F SPIKE> 3000 THEN 1564 

1586 IF SPIKE > 2000 THEN A = A + 1 

1588 IFSPIKEc2000ANDA

1622 OUT (&H303), 0 

1623 OUT (&H302), 0 

1624 FORI=1TO100 

1625 NEXT I 

1626 10 = INP(&H300) AND &HFF 

1627 hi = INP(&H301) AND &HF 

1628 soft = ((hi * 256) + lo) AND &HFFF 

1630 I(count) = (soft - 2048) / (msc * sen) 

1640 LINE -(50 + count * XSCALE, 295 - (soft - 2048) * YSCALE), 

1645 NEXT count 

1650 GOSUB 1830 

1660 GOSUB 3600 

1670 RETURN 

1830 REM SUB FILEIN 

1840 LOCATE pfin -1,2: PRINT "DATA COLLECTION COMPLETE 

LOCATE pfin, 2: PRINT "SAVING DATA" 1850 OPEN "0', #1, 

title$ + 

1860 PRINT #1, NAM$ 

1861 PRINT #1, SOLVENT$ 

1862 PRINT #1, CONC 

1863 PRINT #1, SLIT 

1866 PRINT #1, EXT 

1864 PRINT #1, START 

1865 PRINT #1, FINISH 

1867 PRINT #1, WEXT 

1868 PRINT #1, SCANSPEED 

1869 PRINT #1, sen 

1870 FOR count = 0 TO nreadings 

1880 PRINT #1, count + START; CHR$(9); I(count) 1890 NEXT count 

1900 CLOSE #1 

1905 LOCATE pfin - 1, 2: PRINT "Do you wish to save a shorter data set 

(YIN)"; 1907 INPUT ans$ 

1908 IF ans$ = "1" OR ans$ = "y" THEN GOTO parse: 

1909 IF ans$ = "N" OR ans$ = 'n" THEN 1911 

1910 GOTO 1907 

1911 RETURN 

2060 REM SUB BLANK 

2080 LOCATE pfin - 3, 1: PRINT" 

"2090 LOCATE pfin - 2, 1: PRINT" 

"2091 LOCATE pfin - 1, 1: PRINT" 

"2095 LOCATE pfin, 1: PRINT" 

"2100 RETURN 

2180 REM SUB FILERET 

2200 CLS : LOCATE 1,4: PRINT "ENTER NAME OF FILE TO BE 

RETRIEVED": INPUT title$ 

2210 OPEN "I", #1, titleS + ".XLB" 

2211 INPUT #1, NAM$ 

El

2212 INPUT #1, SOLVENT$ 

2213 INPUT #1,CONC: 

2214 INPUT #1, SLIT 

2215 INPUT #1, EXT 

2216 INPUT #1, START 

2217 INPUT #1, FINISH 

2218 INPUT #1, WEXT 

2219 INPUT #1, SCANSPEED 

2220 INPUT #1, sen 


2230 FOR count = 1 TO nreadings 

2240 INPUT #1, A$ 

2242 temp$ = LEFT$(A$, 4) 

2243 A = LEN(A$) 

2244 I(count) = VAL(fflGHT$(A$, A - 5)) 

2250 NEXT 

2260 CLOSE #1 

2270 MARK = 1 

2280 CLS 

2290 LOCATE 1, 4: PRINT "data now loaded" 

2300 FOR x = 1 TO 5000: NEXT 

2310 CLS 

2320 RETURN 

2480 REM SUB DRAW 

2490 GOSUB ten: PSET (150, 400), 0 

2500 IF MARK = 0 THEN LOCATE 24,2: PRiNT "SORRY NO DATA 

STORED": RETURN 2515 YSCALE = 290 I (msc - 2048) 

2517 XSCALE = (610-50) / nreadings 

2525 PSET (50, 295) 

2530 FOR count = 0 TO (FINISH - START) 

2535 soft = I(count) * msc * sen 

2540 LINE -(50 + count * XSCALE, 295 - (soft) * YSCALE), 12550 NEXT 

2560 RETURN 

2720 GOSUB ten 

2730 GOSUB 410 

2731 GOSUB 2490 

2759 GOSUB 2060 

2760 LOCATE pfin - 2, 2: INPUT "IS THIS THE CORRECT CHART 

(YIN)"; 

2780 IF p$ = "1" OR p$ = "y" THEN 2800 

2790 LOCATE 23,2: PRINT "PLEASE LOAD THE REQUIRED DATA 

FIRST": FOR x = 1 TO 500: NEXT: GOTO 120 

2800 GOSUB 2060: LOCATE pfin - 1, 1: INPUT "INPUT INITIAL 

WAVELENGTH "; IN1T 

2805 LOCATE pfin, 1: INPUT "FINAL WAVELENGTH "; FINI 

2810 IFINITCSTARTORINIT>FINISHORFINI>FINISHORFINI< 

INIT THEN GOSUB 2060 flSE 2820 GOSUB 2060: LOCATE pfin, 1: 

5

PRINT "SORRY RANGE IS UNACCEPTABLE": FOR x = 1 TO 5000: 

NEXr: GOTO 2800 

2820 REM INTENSITY AT INIT=INT (VAR) 

2830 FORN=OTOFINTSH - START 

2840 FLUOR(N) = 1(N) 

2850 WAVENTJMBER(N) = 1E+07 / (START + N) 2860 NEXT N 

2870 Area = 0 

2880 FOR WAVENO = 40000 TO 13000 STEP -500 

2890 if WAVENO c 1E+07 I (FINI + 1) THEN FINTENSITY = 0: GOTO 

2970 2900 IF WAVENO> 1E+07 / (IN1T - 1) THEN FINTENSITY = 0: 

GOTO 2970 2910 FOR N = 1 TO FINISH - START 

2920 if WAVENTJMBER(N) c WAVENO THEN 2950 

2930 NEXT N 

2940 GOTO 3000 

2950 FINTENSITY = (FLUOR(N -1) + (FLUOR(N) - FLUOR(N - 1)) * 

(WAVENIJMBER(N - 1) - WAVENO) / 500) 2970 if WAVENO = 

40000 THEN Area = Area + FINTENSITY / 2: GOTO 2990 

2980 if WAVENO = 13000 THEN Area = Area + FINTENS1TY / 2 ELSE 

Area = Area + FINTENSITY 

2990 NEXT WAVENO 

3000 GOSUB 2060 

3010 LOCATE pfin - 1,2: PRINT" AREA BETWEEN"; INIT; 

"nm AND"; FINI; "nm IS"; INT(Area * 1000) I 1000 3011 LOCATE pfin, 

2: PRINT " Press Shift PRTSCR for Hard Copy any other key to return 

to menu" 

3012 A$ = INKEYS 

3013 if A$ - " THEN 3012 

3014 if A$ = 'Y" OR A$ = "y" THEN GOSUB sdump ELSE 3015 

3015 GOSUB 3600 

3017 RETURN 

3210 REM SUB INFO 

3220 PRINT : PRINT : PRINT: PRINT 

3230 if DT$ = " THEN 3240 ELSE 3250 

3240 INPUT "ENTER TODAYS DATE "; DT 

3250 INPUT "ENTER NAME OF FILE "; titleS 

3260 INPUT "ENTER NAME OF COMPOUND"; NAM$ 

3270 INPUT "ENTER SOLVENT USED"; SOLVENTS 

3280 INPUT "ENTER CONCENTRATION "; CONC 

3290 RETURN 

3600 REM SUB OPTIONS 

3610 GOSUB 2060 

3620 LOCATE pfin - 3, 15: PRINT "OPTION (1) Return to Menu" 

3630 LOCATE pfin - 2, 15: PRINT" (2) Calculate Area" 

3650 LOCATE pfin, 15: INPUT "PLEASE ENTER YOUR CHOICE"; R$ 

3660 IF R$ = "1" THEN CLS: GOTO 120 

3680 IF ES = "2" THEN GOSUB 2060: GOSUB 2760 

3690 RETURN

3700 REM SUB YIELD 

3710 CLS 

3720 PRINT: PRINT: PRINT 

3730 INPUT "ENTER ABSORBANCE OF QS"; ABSQ 

3740 INPUT "ENTER ABSORBANCE OF SAMPLE."; ABSS 

3750 INPUT "ENTER AREA UNDER QS"; ARQ 

3760 INPUT "ENTER AREA UNDER SAMPLE"; ARS 

3770 QY = ((ARS * ABSQ) / (ARQ * ABSS)) * .55 3780 CLS 

3800 PRINT 

3810 PRINT "* QUANTUM YIELD="; INT((QY * 100) + .5) / 100; " 

3820 PRINT 

3825 INPUT "PRESS I TO CONTINUE"; A$ 

3826 IF A$ = " THEN 3826 ELSE 3830 

3830 GOSUB ten 

3840 RETURN 

3850 CLS 

3870 FILES A$ 

3880 INPUT "PRESS Y TO CONTINUE"; K$ 

3890 IF K$ = " THEN 3890 

3900 RETURN 

4000 NEXT 

5000 LOCATE 1, 1: LPRINT" PARAMETER SEflINGS" 

5001 LOCATE 2, 1: LPRINT " 

5002 LOCATE 3, 1: LPRTNT "EMISSION SLiT "; SLIT; 

n m" 

5003 LOCATE 4, 1: LPRINT "EMISSION WAVELENGTH START 

START; "nm" 

5004 LOCATE 5, 1: LPRINT "EMISSION WAVELENGTH FINISHING"; 

FINISH; "nm' 

5005 LOCATE 6, 1: LPRINT "EXCITATION SUT "; EXT; 'nm" 

5006 LOCATE 7, 1: LPRINT "EXCITATION WAVELENGTH 

WEXT; 

"nm' 

5007 LOCATE 8, 1: LPRINT "SCANSPEED 

SCANSPEED; "sec" 

5008 LOCATE 9, 1: LPRINT "SENSITIVlTY "; sen 

5009 RETURN 

5010 ten: 

5020 SCREEN 12 

5030 RETURN 

5040 sdump: 

5050 LOCATE pfin, 1: PRINT" 

7000 'SCREEN DUMP ROUTINE for any 9 pinGRAPHICS 

printer 

7010 PRT: 

7020 LPRINT CHR$(13) 'clears print 

buffer 

7

7050 LPRINT CHR$(27); "3"; CHR$(12) 'sets line spacing to 12/216" 

7060 FOR y = 1 TO 350 STEP 8 'sets y scanning range (8 bit 

mode) 

7070 GOSUB scan: 'calls subroutine to scan x & print 

7080 NEXTy 

7090 RETURN 

7100 scan: 

7104 w = 3 

7105 . LPRINT CHR$(13) 'CR printer code 

7110 x = 300 'width of paper in head 

movements 

7120 nl = x MOD 256 'number of bit image data 

7130 n2 = INT(x / 256) 

7140 LPRINT CHR$(27); "L"; 

7145 CHR$(nl); CHR$(n2); 'sets high density bit image 

mode 7150 EORJ = 1 TO x 

7160 sc = 640 / 6 'scales the screen to printer 

7170 sum=0 

7180 IFPOINT(j*sc, y) w THEN sum = sum +4 

7240 IFPOINT(j*sc, y +6) w THEN sum = sum +2 

7250 IFPOINT(j* sc, y +7)o w THEN sum = sum +1 

7260 

7270 

LPRINT CHR$(sum); 

NEXT j. 

7380 RETURN 

7400 parse: 

7410 LOCATE pfin - 1, 1: PRINT" "LOCATE pfin, 1: PRINT" 

7420 LOCATE pfin - 1, 1: INPUT "Enter number of data points required"; 

ans 

j = nreadings / ans 

7430 jump=INT(j*10+.5)/10 

7450 title$ = titleS + 

8850 OPEN "0", #1, titleS 

8860 PRINT #1, "Name - "; CHR$(9); NAM$ 

8861 PRINT #1, "Solvent - "; CHR$(9); SOLVENTS 

8862 PRINT #1, "[Conc] - "; CHR$(9); CONC 

8863 PRINT #1, "Em slit - "; CHR$(9); SLIT 

8866 PRINT #1, "Ex slit - "; CHR$(9); EXT 

8864 PRINT #1, "Start - "; CHR$(9); START 

8865 PRINT #1, "Finish - "; CHR$(9); FINISH 

8867 PRINT #1, "Ex Lmax - "; CHR$(9); WEXT 

8868 PRINT #1, "Scanspeed "; CHR$(9); SCANSPEED 

E1

8869 PRINT #1, "Sensitivity"; CHR$(9); sen 

8870 FOR count = 0 TO nreadings STEP jump 

8880 PRINT #1, count + START; CHR$(9); I(count) 

8890 NEXT count 

8900 CLOSE #1 

8905 LOCATE pfin - 1, 1: PRINT "You now have two data sets one"; 

title$; 

".XLB containing all data" 

8906 LOCATE pfin, 1: PRINT "and another "; title$;" containing parsed 

data' 

8907 FOR x = 0 TO 5000: NEXT x 

8911 GOTO 1911

Appendix B

Volume laS, number 5,6 CHEMICAL PHYSICS LZfltRS 17 Januaty 1992 

Excimer venus TICT state formation in polar solutions 

of methyl 4- (N',N-dimethylamino )benzoate 

John A.T. Revill • and Robert G. Drown 

• DePartment QIChemisay. Lancashire Po&technic, hrswn. Lancashire FRI 2712 IlK 

• SERCOwesbury Laruio,y, Wanington, Cbnhfre WAI lAD. (iN 

Ralved 29 September 1991; in final (cnn 2$ October 1991 

The fluorescence properties of methyl 4-(N,N4imeth4amlno)beno.te In a nricty of solvents are reported. No.n,sr nunmnce 

from a distribution of excited wiute/wivent arrausements Is observed In most solvent Additionally, in acetonitrile, a 

Iongr wavelength fluorescence band Is observed which is auributed to excited dieter fluorescence. Itis proposed that the excited 

dimer is formed by dbt excitation from the pound state. 

1. Introdactlon 

The concept of the twisted intramolecular charge 

transfer ('ncr) state as an explanation for anomalous 

dual fluorescence has met with wide acceptance. 

In donor-acceptor molecules (typified by the 

4-(N,N-dialkylamino)benzcnitriles), the different 

relative orientations which can be adopted by the 

donor and acceptor parts of the molecule give risc to 

excited states which can be identified with "normal" 

and TIC' fluorescence (1-3 ]. It is now recognised 

that TICT slates may play a role in the photochemistry 

of a wide range of molecules and molecular systents 

( 4 1 

TICT state formation in the 4-(N,N.dialkylamino)benzonitriles 

has been investigated for a wide 

range of donor groups (act eg. ref. (] and references 

therein), but variation in the acceptor system 

is more difficult. Dual fluorescence has been obaerved 

for the methyl and ethyl estcn of 4-(N,N-diethyLamino) 

and 4-(N,N-dimethylamin0benioic 

acid. These observations have been ascribed to TICT 

state formation 11,6-81. This assignment has been 

challenged by Visser et al. (9] who believe that salutc-solvent 

exciplexea arc the causc of the dual fluorescence 

and by Howell ci aL (10) who have evidence 

for excimer formation in solution and for 

dimer (or lilgber aggregate) in a supersonic jet. Excimer 

fluorescence has also recently been reported in 

polar solutions of hcxadccyl 4-(N,N-dimethylaxnino)-benzoate 

[II). However, TICT state for' 

mation has also been claimed in the related system 

of methyl 2-meihoxy-4- (N.N-disnethylamino) henzoate 

[12,13], We arc investigating the photophysica 

of methyl 4-(N,N-dimethylamino)-benroate as 

part of a comprehensive study of the photochemistry 

of the amino- and (N,N-dimcthylamino )benzoic 

acids and their methyl esters. We report here the results 

of our measurements on this compound under 

various room temperature conditions. 

2. ExperImental 

Methyl 4-(N,N'dimethylamino)bcnzoate was prepared 

by the method of HaJIas (14]. A solution of 

4-dimethylaminobcnzoic acid (0.06 mol) in methanol 

(os cnr 3 ) was added to boron trifluoridemethanol 

complex (0J8 mol) and the mixture boiled 

under reflux for 4 h. After cooling, the mixture was 

poured into saturated sodium hydrogen carbonate 

solution and the ester extracted with ethyl acetate. 

The ethyl acetate extract was dried and the ethyl acetate 

removed to yield the crude ester which was twice 

recrystallised from ethanol Yield 10.02 g (79%), 

m.p. ll0-lll°C (literaturem.p. 111-113°C (IS)). 

The solvents used in these studies were either 

spectrophotometric or spectrofluorirnetric grade and 

00094614/9215 CL® 0 1992 Elsevler Science ?ublishcn B.V. All riJtts reserved. 433 

10

Volume IU. number 5,6 - CHEMICAL PHYSICS LETFEPS 17 January 1992 

were used as received except for acetonitrile. For the 

fluorescence measurements (spectra and decays), the 

acetonitrile was distilled over phosphorus pentoxide 

and collected under an inert atptospbert The pure 

solvent and all relevant samples were stored under 

argon to inhibit the absorption of water oroxygen. 

Water was doubly distilled and then passed through 

an ion exchange system. 

Absorption spectra were measured on a Perkin- 

Elmer Lambda 3 spectrophotometer using quartz 

cells of varying pathlenghts. *11 working solutions 

were prepared from a stock solution on a daily basis 

and the integrity of the stock solution was continuously 

monitored. The accuracy of the measurements 

is estimated at ±2 am for wavelengths and ±50% 

for extinction coefficients. 

Fluorescence spectra were measured on a Perkin- 

Elmer LSS spectrofluorimeter in fully corrected 

mode. The fluorescence spectra were collected on a 

Rae B microcomputer which was capable of calculating 

the area under the spectnaan for quantum 

yield determinations. Quantum yields were measured 

on optically dilute samples (absothance 

Volume ItS. number 3,6 CHEMICAL PHYSIC LZFItRS 37 January 1992 

I 

1' 

I 

Wavelength (am) 

Fig I.Absorptioa aadfluorncencespectno(M4DMABIn seetonithle (1.63x 10'mol dnr'). 

Tabk I 

Absoipsion and fluorescence properties of MIDMAB us function of solvent 

Solvent Absorption Fluorescence 

A. (cm) extinction coefficient 2,.,, (cm) 4 t, (us) 

(dm'molcm") 

hexane 212 17000 348 0.300 1.30 

ethanol 299 19800 341 0.067 0.23, 1.16" 

acntonithle 292 19100 335,480 -' - fl 

"titer 316 16600 346 0.022 0.11,3.30" 

"The quantum yield In acetonluilt niles with eoncentntlon. 

"Double exponential decays observed. "Complex behaviour - see text. 

different solute/solvent interactions 1211 as has been 

proposed by Howell et al. for M4DMAB in mixtures 

ofnon-polarsolvents 1101. The possibitityofa nonfluorescent 

TICT state in these solvents is not ruled 

out, however, and we are undertaking further measurements 

in hydrogen-bonding solvents in an attempt 

to clarify the position. 

The fluorescence properties of M4DMAB in acetonitrite 

are anomalous in that a second fluorescence 

peak is observed at 480 nm. This confirms the observations 

of other worken [l,6—t0J. Although mere 

is no obvious change in the absorption spectrum as 

concentration is varied, the fluorescence spectra do 

change and the intensity of the long wavelength peak 

decrcascs with decirasing concentration (fig. 2). The 

anomalous ("red") fluorescence peak is therefore 

outwardly exhibiting the typical characteristics of 

excimer formation 1221. However, the fluorescence 

quantum yields for the two fluorcscence bands do 

not vary with concentration in the manner which 

would be expected for excimer formation and the 

fluorescence decay profiles do not show the idationship 

that would be expected on the basis of cxcinier 

formation. Indeed, it is the "red" fluorescence 

435 

13

Volume ISt rnunbes 5,6 ! ;b'j.i (04 E u In 17januvy 1992 

I 

I LU 

I 

I 

0 

7 -.-- 

loaD 

p 

F1. 2. Fluoresccncc .pectz. of M4DMAB in aeetonlirile ass function ofvonautration: (—) 5.Ox 10-' rnol dnr", () 2.5x IV'mol 

dnr', (.--) I.OX 10 6 mol dir', (---) LOX I04moldnr1, (--) 2.5x lr'noldzr'. 

which exhibits a single exponential decay whilst the 

"blue" peak is double exponential! 

Over a range of concentrations (5x 10'-Sx 10 

inol dm') and excitation wavelengths (270-340 

nm), the "red" fluorescence band gives excellent 

single exponential fits to the expciimcntal decay profile 

with lifetimes in the range 2.36-2.71 iii. For the 

same solutions, the "blue" fluorescence is clearly not 

a single exponential but is adequately explained by 

a sum of two exponentials with lifetimes in the range 

1.27-1.70 and 2.21-2.77 as (table 2). At present we 

are unsure as to whether there is a correlation between 

the "red" lifetime and the longer of the two 

"blue" lifetimes. It is possible that the latter represents 

a process whereby the 'S" excited state is 

convened to the "blue" excited state as proposed by 

Howell et al. (10]. However, the discrepancy between 

the two lifetimes and the fact that the a2 values 

in the double exponential fits are positive when 

a risetime should have a negative pre-exponential 

Table 2 

fluorescence decay pm'pertics of M4DMAD in acctonithlc as a function ofonczntntlon 

Conccntntion 'Eftlltton a, r, (os) a, 1, (ns) 

(moldnr') wavelength(nm) 

3.0x10' 355 0.0403 1.70 0.0003 2.21 

480 0.0365 2.36 

LOX 10' 355 0.0395 1.65 0.0007 2.45 

480 0.0372 2.49 

Lox 10-' 355 0.0191 1.69 0.0008 2.49 

480 0.0375 2.71 

436 

iV

Vohuna 151, number 3.6 OIEMICAL PHYSICS LEfltRS *7 January 1992 

factor suests that the similarity in these lifetimes 

is purely coincidentaL 

Adapting the accepted terminology for these systems 

where the "normal" blue fluorescence is attributed 

to an ached state labelled be and the red fluorescencc 

to a state labelled C we can identify a 

number of difreitnt mechanisms for this fluoracence 

behaviour. Unless there exists a very thort-lived 

precursor whose lifetime is to short that we are unable 

to separate it from the instrumental response 

function when we model the fluorescence decays of 

as and b states, both 0 and b O are excited directly 

from the ground stata Once the two excited states 

have been formed, it h possible that a' converts to 

b' but there is no evidence for the reverse process. 

Our conclusion from this data Is that the fluorescence 

spectrum and dccay profile of the V state Is 

comprised of contributions from a distribution of 

excited solute/solvent arrangements, the decay kinetics 

of which are adequately described by a sum of 

two exponentials [2 3,24 ) Proving that the decays 

are multicxponcntial will require high quality data 

with large numbers of counts in the decay profiles 

(21) but we intend to undertake the required experiments 

to acquire this data in the near future. 

Unlike Howell a al. [10] we have no evidence to 

suggest that a' derives from b'. The additional obtervation 

of the concentration dependence of the a' 

fluorescence band leads usto conclude that as is an 

excited dirtier (or possibly a higher aggregate) which 

is populated directly from the ground state by absorption 

of a photon of light Given the lack of concentration 

dependence of the ground state absorption 

spectrum, the ground state "dimer" may not 

involve a strong interaction between molecules. 

However, a loose association is sufficient to organize 

the molecules such that they are in approximately 

the con'ect position to form an excited dimcr when 

a photon is absorbed by one of the molecules. We are 

conscious that the time resolution of our time-correlated, 

single photon counting system would not allow 

us to measure a rapid risetime (perhaps

Volume III. number 5.6 CHDIICAL rHYsia LEflS 17 J.nuaq £992 

(191DJ.S. BITSI, A. Duteb R.E. Imhof, B. Nadohki and I. 

Soutar.J.Pbctachenj8 (1997)239. 

[201 R. Snow, R.O. Bgown. Efl Ev.na and D. Sbaw.S. C,cm. 

Soc. Faraday Trans 1182 (1986) 2249. 

[21) D.R. Saint, and Wi. Ware, Gate.. Pitys. Letter, 126 

(1986)7. 

[22)S.t BirkL Photophyaa at aromaik molecule, (Wiley- 

Intendtnee, New Von, 1970). 

1231 5.9. Mad. and D. Pbiflip.. J. Otca Soc. Faraday Trw. 

1183 (1987) 1941. 

(24)W. RettiL P4. Vo,eI. E. Lippat and H. Otto, O,em. Pby. 

Letter, 103 (1986) 381. 

438 

16

Jaw'.aiefflanwa VOL 1 No, 2 1992 

Anomalous fluorescence Properties of 4-N,N- 

Dimethylaminobenzoic Acid in Polar Solvents 

John A. T. RievIll' and Robert C. BrownW 

ReedS Much 17. IPPt n,Ud laTh 199Z• .ecqtS Is 24 1992 

4.?6N-Dimcthylsmlnotcnzolc acid eth1bfts anomalous fluorescence In polar and hydro;enbonding 

solvents. The fluorescence spectra and kinella naggest that this arises due to the 

formation of a ground-sat; dimer or blgbcr polymer- Preliminary measurements in bent. 

containing small amounts of polar acstonitrila do not S. out the possibility of exnipiex fornation 

also ocarring. 

KEY WORDS; Bwawa anomalous fluotesesna; 444N-dlmttbytsStolc acM dImasSn. 

INTRODUCTION 

Derivatives of 4-a%N.4lmethybm1nobenzoic add 

such asS nitrile and various eaten have been known 

for • number of yen to show anomalous fluorescence 

properties (1-3]. Two fluorescence bands are observed 

for these compounds In some solvents. The shorter. 

wavelength emission, with a Stokes shift of a few thousand 

wavcnwnbcra, Is Identified as "normal" fluorescent 

from the knily exefted Lc., Franck-Coodco (It) 

slngjet state to the K ground state [WC)]). In current 

termjnolov this is also referred to as a b' state. The 

anomnious. longer-wavelength emission (from a state Ishelen 

a) has a number of possible origins. 

The suggestion that the a' emission results from a 

twisted intramolecular change transfer (flC) state has 

met with quite wide acceptance (4-4). ThIs transition 

would be labeled S, (CT) —. $ (PC) in the notation 

adopted by HeidI ci al. (7). However, Vatma and w- 

'G.a,thoy Dspanment, Univmi?y of Cesesi Uncasbire. Pretco. 

Lanhire, Pit 2H1 UK. 

'sac Dsresbwy Ubotstmy, Waningion. aesbfte. WA4 4AD UK. 

To whom amtjposexz slould bc addteex& 

workers, at an early stage in the studies of these mote-- 

cults, proposed that the a' emission was due to a solutel 

solvent exeiplex (8) and have maintained this belief In 

a series of later papers [see 9 and references therein). 

More recently, Phillips' group has presented evidence 

that the a' emission of 4-N N-dlmethylsn.invbeoaonlnil; 

and the methyl esters of 4491-dhnethylaznino- and 

4-N.N-dlcthylnmlnobcnzolc acid In a supersonic Jet Is 

due to dimer formatIon (10,11) We have reported a 

concentration dependence for the anomalous fluorescence 

of methyl 4-N,N41imethylsrnlnobenzo3te in soltition, 

which also points to dirner formation [121. Although 

there was no evidence from the absorption spectra of 

ground-state dimer formation, the fluorescence properties 

of this ester did not conform to those expected for 

excirncr formation and we concluded that at least a loose 

ground-state a.saociadon exists. 

We have also studied the absorption and fluorescues 

properties of the parent 4-N,N4imethylaminobenzoic 

acid. Cowley or aL (3) have reported that this 

molecule also exhibhs a fluorescence (in acetonitrile), 

but this appears to be the only recent study on the parent 

acid. We therefore present our findings on this compound 

to date in this paper. 

107 

lwafl.iqta. C nez ruaaa r.,ss.1 

17

log 

EXPERIMENTAL 

4-N,N-Dimetbylaminobenzoic acid was purchased 

from Lancaster Synthesis Ltd. and was reclystattlzed 

twice from ethanol prior to use; m.p., 244- 

245*C (literature value, 242-243t 113)). Infrared and 

proton and "C NMR spectra were also acceptable. The 

solvents used In this work were either spectrophotometric, 

spectrofluorimetrie. or HPLC giadeand were 

used as received with the exception of acetonitrile. 

This was distilled over phosphorus pentoxide and collected 

and stored under an inert atmosphere until it 

was used. The water was doubly distilled and then 

passed through an ion-exchange resin prior to use. The 

p11 of the unbuffered water was 615. 

Absorption spectra were measured on a Ferkin-Elmet 

lambda 3 speetrophotometer or & Hewlett-Packard 

HPS451A diode-tiny spectrometer using matched quart 

cuvettes of varying path lengths. All working solutions 

were prepared fresh on the day of use from a stock solution 

which was checked at regular intervals for pudty. 

The accuracy of the measurements made under these 

conditions is estimated as ± 2 nm for wavelengths and 

± 10% for extinction coefficients. 

fluorescence spectra were measured on Perkin-Elmet 

LSS or LS50 speetrofluorimeters in the fully corrected 

mode. Excitation and emission slit widths of S 

nm were used unless otherwise indicated. For the LSS 

the spectra were transferred to a BBC microcomputer to 

undertake quantum yield calculations. Quantum yields 

were measured on optically dilute samples (absorbance 

Anomalous fluorncence Properties of 4.N.N-Dimethylaminobeozolc Add 109 

So4nm 

Table I. AAwqxion and Flcsccna Ytupatin of 40MADA b Vattes Sent, it Bas Tcnwcnwtc 

Absorption 

maxims (am) 

Eminedno 

coefficient 

(din' md-' or') 

Thjogcsctace 

msxüna (nm) 

fluaracence 

,xcitaIoa 

maxima (vim) 

Hmnc 302 a 340 300 

225 a 

EThanol 295(1liy - 350 vs 

282 20,300 420 280 

Alaaluile 306 22,500 350 275 

295(a) - 480 308 

Wamr 

(nbgffcnd) 288 a 365 270 

222 a 500 290 

Water 

AnIon 284 a 355 274 

225 a 490 306 

Ncun.t 314 • 370 290(th).300 

230 a 480.490 300 

Cdoc 272 a C - 

230 

• amonnd a paitlsily z1ilg In this tolvem. 

'sh, s)oidc. 

t4ofluwtscencc. 

C 

a 

t 

in 

C 

S 

S 

S 

I' 

'a 

g 

Wautlenoth (vim) 

F!p t. fluorcscncc spews of 4DMABA (1.08 x 10 mol•dnr') In .cetooiuilc at different excitation wsvcicn8ths. 

19

no 

Revill and Brown 

U.. 

251 255 311 351 

Wauelangth (rim) 

FIg. 2. Absorption spectrum of 4DMABA (53 x 10" mol-dr') in aettoniufle. 

that this band corresponds to the normal V (5 1 (It) -, 

80 (FC) fluorescence) for 4DMABA and the longer 

wavelength band to anomalous (a) fluorescence. In 

henna, the fluorescence spectrum of 4DMABA is independent 

of excitation wavelength (except for a change 

in overall intensity) and the excitation spectrum is independent 

of emission wavelength. This is not the case 

in all the other solvent systems used. The measured 

quantum yield of 0.18 in hexane Is in good agreement 

with the value of 0.21 in cyclohexane reported by Cow. 

leyetaL (31- 

In acetonitrile, excitation of a I x 10" mol'dm' 

solution at the absorption maximum of 306 rim produces 

a relatively weak fluorescence ape ctrum (rig. 1) but with 

the two bands cicarly visible. If the excitation is moved 

to a shorter wavelength (e.g.. 270 rim), the intensity of 

the b fluorescence increases approximately fourfold, 

while the a fluorescence almost disappears. The spectra 

at intermediate wavelengths tie between these two cxtremes. 

It is not surprising therefore to find that the cx- 

citation spectra for the two fluorescence bands are very 

different, with that for the a' fluorescence peaking at 

308 nm and that for the b' fluorescence at 275 rim. The 

latter may correspond to the asymmetry on the absorpdon 

spectrum of 4DMABA in acetonitrile (Fig. 2). 

Emission spectra of 4DMABA in water exhibit an 

identical trend to thou in acetonitrile but with overall 

smaller changes; the V fluorescence Intensity increases 

by only a factor of two as the excitation wavelength is 

changed from 300 to 210 nfl and the a intensity de. 

ceases by a similar amount (Fig. 3). The excitation spectra 

peak at approximately 290 urn (Cf. absorption maximum 

at 288 urn) and 270 not for the a' and V fluorescence, 

respectively. In water there is no obvious absorption 

spectral feature at 270 nm to correspond to the latter 

peak. The fluorescence spectra of 4DMABA in water 

are also pl4 dependent (Pig. 4). At alkaline pH values 

both anomalous and normal fluorescence are observed, 

with the tatter peaking at 355 nut This must therefore 

correspond to normal fluorescence from the 4DMABA 

20

Anomalous Fluorescence Properties of 4-N,N-Dimctbylaminobenzoic Acid 111 

21 

- ZGDM 

• fl*mi 

a Sitvn 

P. 

ls 

2' 

a 

S 

N 

I1 

I 

Si, l'S Si, Sit 

Waoilanth lam) 

tlg.3. Muoresanee spectra of 4DMABA In watu at diiktcnt cxcitation wavelengths. 

carboxylate anion, as the ground-state pK. is 4.79 and 

there Is insufficient acidity present to convert the excited 

anion to the excited neutral species despite a calculated 

1201 pK,' of 11.95 [19]. As the pH is reduced, the b 

fluorescence band shifts to 375 am and the a fluorescence 

decreases in intensity until it is debatable whether 

it is still present. In strong acId the fluorescence disappean 

completely—the 4DMAEA cation Is nonfluorescent. 

In ethanol, the anomalous fluorescence is shifted 

less to the red than in acetonitrile and water and appears 

as a shoulder on the long-wavelength side of the emission 

band (Fig. 5). Once again, the spectrum varies with 

excitation wavelength but the observed changes are the 

smallest for the three solvents where the anomalous fluorescence 

is observed. These observations dearly suggest 

that there is more than one ground-state species 

present in the polar solvents used in these studies. Our 

measurements to date have not shown any change in 

absorption spectra with concentration but this may be 

due to the fact that any changes are taking place at much 

lower concentrations than we have studied to date - The 

evidence to 5upport this Is that the fluorescence spectra 

for 4DMABA In acetonitrile are concentration dependent. 

This has also been observed for M4DMAB [12). 

As the concentration of 4DMASA is varied between 

10' and 10' mol-dnr 3 (Fig. 6). the amount of a 

fluorescence decreases until at 10' mol-dm 3 there is 

vexy little remaining, although the overall weakness of 

the fluorescence from 4DMAEA makes measurements 

difficult at these low concentrations. It is possible that 

absorption spectral changes may be visible at these concentrations 

but the 4DMABA extinction coefficients preelude 

acetirate absorption measurements at concentratiotis 

below approximately 10' mol-dm'. 

The data presented so far suggest quite clearly that 

the anomalous fluorescence from 4DMABA arises from 

a ground-state species. The concentration dependence of 

the fluorescence in acetonitrile points to this species being 

a dimer or higher polymer, although we have no ab- 

21

112 Haiti and Brown 

a 

C 

ii 

I 

Wavelength tarn) 

Pig. & fluorescence spears of 4DMAKA in water at different pH values. 

sorption evidence for this. The variation of the fluorescence 

spectra with concentration certainly does not agree 

with a mechanism Involving the formation of a TICT 

state 14], and our attempts to analyze the data using 

standard exchner kinetics 1211 have not been successful—the 

expected dependence of fluorescence intensity 

and kinetics on concentration is not observed. Unfortunately, 

the absence of any sgnifkant changes in the 

absorption spectra with concentration means that we are 

unable to verify the presence of ground state equilibrium. 

We have, however, studicd the fluorescence decay 

profiles for the two fluorescence bands in some of the 

solvents used here. In hexane the decay is single exponential 

with a lifetime of 1.22 ns, very similar to the 

value of 1.30 ns found for M4DMAB in hexanc [12]. 

In ethanol, the proximity of the C and b fluorescence 

bands leads to complex decay profiles which require threeexponential 

components to fit the data adequately. We 

are making further measurements at a variety of cxci- 

tation and emission wavelengths to try to resolve this 

complexity, but in view of the various solute/solvent 

interactions which are poasible in hydrogen-bonding solvents 

(In addition to any other processes taking place), 

decay profiles which are not single exponential are not 

surprising. In water and acetonitrile, the as fluorescence 

decays by a single exponential, whereas the b' fluorescence 

requires two exponentials for an adeuqate fit (Table 

U). In aectonithle, where we have accumulated a 

large amount of decay data, the lifetime of the C band 

is independent of conantration and excitation wavelength. 

A lifetime in a range of 2.1-13 ns is always 

obtained and there is no evidence of a rise time indicative 

of an excited-state precursor. As Ilasselbacher eta?. [22] 

have shown, this does not absolutely rule out this mccl,- 

anism but it does indicate that at least one alternative 

mechanism of populating the a' state is operative. We 

interpret these observations as confirming our earlier 

conclusion about the presence of a ground-state species 

producing the a' fluorescence. They also imply that this 

22

Anomalous fluorescence Properties of 4-N,N-Dlmethylamlnobenzolc Add 113 

I 

$ 

S 

S 

.1 

5 

I 

Waslelenath Inmi 

Figs. Flisoruceac. aped,. of 4DMABA (1.14 x 10-' mot-dnr') In ethanol.' dirftrent excitation wavelengths. 

is the only source of the C species—the b' excited state 

does not appear to be a precinct of a', although the 

reverse may possibly be true given the more complex 

decay profiles observed for the b* fluorescence band. 

Vanna and co-workers (23,24J have proposed that 

the anomalous fluorescence of ethyl 4-N,N-dimethylamlnobcnzoate 

123] and 4-N,N4methylaminobenzontrile 

[24J in acetonitrile is due to exciplex formation. We have 

therefore added small amounts of acetonitrile to a solution 

of 4DMABA in hexane up to the miscibility limis 

(approximately 2.2% by volume) and have found that a 

new fluorescence band at approximately 440 nm results 

(Fig. 7). The observed variations in the emission spectra 

do not appeer to fit a standard cxciplcx mechanism. The 

b' fluorescence In hexane Is quenched by low concentrations 

of scetonitrile. but at slightly higher concentralions 

it increases in intensity and red-shifts by some 10- 

20 nm. At the same time there is a steady increase in a' 

fluorescence as the .cetonitrile concentration is in. 

creased. It would appear that there are some specific 

salvation effects taking place in addition to the mccliinism 

producing the C fluorescence. The red shift in the 

b fluorescence probably results from the formation of 

a specific 4DMABA/.cetonitrile solute/solvent interac. 

don. The new band at 440 urn may be an exciplex or 

may be the some species as Is seen in pure acetonitrile 

but with a blue-shifted fluorescence due to the nonpolar 

solvent environment in which it is situated. Whatever 

the nature of the species producing the 440-nm fluorescence, 

it is clear that acetonitrile plays a key role in its 

formstinn and we are undertaking further work to attempt 

to clarify the nature of this species. 

ACKNOWLEDGMtNrS 

We thank the Science and Engineering Research 

Council for a studentship (SAT?.) and for access to the 

Daresbwy Laboratory and University of Cenual Lanashire, 

the Ciba-Oci' Tnsst, NATO, and the British Coun- 

23

114 Revill and Brown 

Tabis 11. flootnaia Decay Peopcnie. of 4DMABA in Vuioua Solvezi, 

k.. Cornntjon Ti TI 

Solveat (na.) (net) (moltr') a, (ill) 

306 340 1.01 1.000 1.28 - - 

Ethuol 280 350 1.01 0.859 2.86 0.141 17.61 

Went 290 365 1.05 0.994 0.77 0.X6 5.07 

500 1.27 0.973 0.17 0.025 2.39 

Aato&vile 280 350 LOx 10a 1.04 I.000 2.13 - - 

233 500 LOX 10-0 1.74 LOOD 116 - - 

303 350 1.OxlO-' 3.61 LOOD 1.66 - - 

303 480 1.OxIO-' 3.48 1103 2.30 - - 

300 350 1.OxlD-' 1.26 0.842 1.29 0.158 2.74 

303 480 LOX10- 6 3.01 LODO 223 - - 

303 350 1.0x10' 1.23 0.687 0.20 0.313 1.85 

303 480 1.0x 10- 1.14 I. 127 - - 

310 350 1.OxlOa 1.17 1.000 2.1* - - 

310 503 1.OxlO" 1.36 IMD 2.15 - - 

340 350 I.OxlO-' 1.57 2.000 2.21 - - 

340 500 1.OxlO-' 1.31 1.000 2.17 - - 

1.3 

- lI.IjilI-1 ri 

SSSii,-7 H 

- 1.1I0I-1H 

4 I.1IeII-4M 

- l.lIaIU-5P4 

•1 

t 

5.2 

S 

a 

a 

I.' 

6 

Is 

525 374 42* 452 535 

Waustn9th (em) 

39, 

F1g.6. Flvon=m lpCtn ci 4DMADA in zcnoolnik at diffnnt COAMOUlliM. 

24

Anomalous Fluortscencc Properties of 4.N,N-Dimethyiamlnobenzoic Acid 115 

S 

a 

if 

t 

S 

a 

C 

S 

S 

E 

S 

a 

Wautlanath Inn) 

Elg.1. Fluonscence .pewa of 4DMABA in henna with different amounts of added acetooiufle; pure haanc (%. Wv). 

cii for financial suppoit. We are grateful to Professor 

Wolfgang Pettig of the TechniCal University, of Berlin for 

his comments and fn4tM discussion and to the Photophysic 

Research Group at the University of Strathclyde 

for allowing us access to their fluorescence lifetime spectrometers. 

REVERENCES 

I. 2. R. Crabowüi. K. Roakewia. D. J. Ssmiaruk. 0.1. Cow. 

Icy, and W. Banmann (1979) Now. J. CAL.,. 3, 443-lU. 

2. 6. Wermuth. W. Renig, and K Lippert (1931) Ba'. BUAIE$9EI. 

Pity,. Chant 85. 64-70. 

3. 0.1. Co-ky. P.i. Italy. iSA. H. Peoplca(1978)J. fob. 

them. P, 240-242. 

4. W. Rflhig (1I86) AMw. Chn. tar. Ed. Lag. 25, 971-916. 

S. W. ReaaI1 (1991) EPA New4eu., No. 41. 3-19. 

6. lii. Van dat Anwarser, Z. 2. GnbowakL and W. Reni1 (1991) 

J. flo'& Cheat 95. 2083-2092. 

7. J. HeIdi, D. Goimin. and M. (nba (1939) Chant P,s 136. 

321-344. 

& K. I. Ylsscr, and C. A. 0. 0. Yarmi (1980) 1. Chem, Soc. 

Fonda3, Tn,,,. 2 76, 433-476. 

9. P. C. M. Wcisenborn, A. H. Huizer, and C. A. G. 0. Var,,,. 

(1989) China. Thys. 133. 437-432. 

10. R. Howell, H. Petek, D. Phillips, and K. YoAiban (1991)Chem. 

FSn. L.M. 183, 249-253. 

11. R. Howell. A. C. jones, A. 0. Taylor. and D. Phillips (1989) 

CAn Ps In. 10.2*2-290. 

12. J. A. T. RaIl, and 2.0. Brown (1992) Cheat. n,& Leer. lB. 

433-43*. 

13. Lancaster, Synlhesis Lid. 1991. Orgsnie Resesrch Chemicals Catalogue. 

Lancaster Synthesis. 

14. R. A. Vetapoldi, and K. D. Mittens (1980) PJBS Special Pubti. 

catIon 260-6-4. 

15. D. P. Eaton (1981)1. han tWL CAr,,,. 60, 1107"I1I4, 

16. D. V. O'Connor. and D. PWllipa (1984) rn Carnioted Sbag!e 

Phcia.a Cowuint Andemie Press. New York. - 

II. D. I. S. Birch. A. Dutch, 2. E. I.nlaof. 8. NS4OI*IEI. and t. Sown 

(19*7)1. P*oroch.aa. II, 239-254. 

IS. R. Sparrow. HG. Dro.na, E. H. Evans, and D. flaw (lQS)J. 

Claim. Soc. Fsnhry twit 2 82, 2249-2262. 

19. J. A. T. RcviIl, and R. C. Brown, Unpublished work. 

20. Th. Farster (1950)1 Elacemehant 54, 531-539. 

21. J. B. SitU (1970) Fhmçtysia of Anmalic Molecules. Wiley 

Inrencienca, London. 

22. C. A. Hamlbacher, E. Waxy,..,,, L T. G.haü. P. B. Cooaino. 3. 

B. A. lWss, and W. R. Laws (1991)/. PhF. Chem. 93, 2993- 

3005. 

23. R. J. Visser, P. C. SI. Wthcnborn, and C. A. G. 0. Vanna 

(1985) Cheat Pity,. 1,14 113, 330-336. 

24. P. C. M. Weixobona. A. H. Huize,, and C. A. G. 0. Vanna 

(1988) Cheat PAys. 116.425-433. 

25

A study of the solution photophysics of the isomeric aminobenzoic acids, N,N-dimethylaminobenzoic acids and their methyl esters

Create successful ePaper yourself

Delete template?

Save as template?