2.24.2 LUMINESCENT EXCITED STATES OF COORDINATION COMPOUNDS 318
2.24.3 LUMINESCENCE SPECTROSCOPIC METHODS 319
2.24.3.1 First Row Complexes 320
2.24.3.2 Second and Third Row Complexes321
2.24.3.3 Lanthanide Complexes 323
2.24.4 REFERENCES 323
2.24.1 GENERAL OVERVIEW AND BASIC CONCEPTS
Luminescence from discrete transition metal complexes was first observed for classical Werner complexesof CrIIImany years ago and early studies of metal complex luminescence provided the basis for characterizing the spin multiplicity and relaxation processes of excited states of com- plexes in solution and the solid state. This early work has been thoroughly documented in a number of books, monographs, and reviews on inorganic photochemistry.1–6
Emission from electronic excited states of transition metal complexes is generally observed from the ultraviolet to the near-infrared. Characterization of emission of molecular species involves measurement of three principal parameters: excited state energies, emission quantum yields, and lifetimes. Figure 1 shows a simple state diagram typical of a metal complex having a singlet ground state (i.e., low-spin d6). Electronic excitation leadsto population of a state having the same spin multiplicity as the ground state (S0!S1); excitation into higher energy excited states of the same spin multiplicity (i.e., S0!S2) also occurs. Relaxation of the Franck–Condon excited states to the thermally equilibrated excited state (state 00of S1) generally occurson the sub-pstime scale. In rare cases excitation into states higher in energy than S1results in direct relaxation to the ground state via emission of light (vide infra). Relaxation from the thermally equilibrated excited state can be via emission (kr), nonradiative decay (knr), or intersystem crossing to the triplet (T1) state. In addition, the excited state may react to yield net chemical products; for example, the excited molecule may participate in energy or electron transfer reactions with other chemical species. The T1state can also decay by radiative, nonradiative, and photoreaction paths.Figure 1b shows representative excitation and emission spectra that correspond to absorption into and emission from the S1 state of the energy level diagram. The excited state energy is defined as the energy difference between the zeroeth vibrational level of the ground and excited states (E0).
The difference in the maxima for the absorption and emission transitions between the zeroeth vibrational levels is a consequence of the fact that the equilibrium nuclear displacements of the
315
ground and excited states differ (the excited state potential surface is distorted relative to the ground state).
Emission spectra provide information on the relative excited state distortion (the Huang–Rhys factor, S), the medium- and low-frequency vibrational modes associated with excited state relaxation (h!) and the zero–zero energy (E0). An approach for fitting emission spectra has been discussed in detail in the work of Meyer and co-workers, and one expression for calculating the normalized luminescence intensity, I, asa function of frequencyisshown inEquation (1).7,8 The expression shown employs a single, medium-frequency, acceptor vibrational mode, h!m, the corresponding electron-vibrational coupling constant, Sm, and a term for the full width half maximum of the individual vibronic components, vh, which providesa measure of the average solvent (or medium) reorganizational energy associated with relaxation of the excited state.Figure 2 shows a typical emission spectrum of a RuIIdiimine complex obtained in a glass matrix at 77 K;
the average medium frequency vibrational spacing is clearly evident and the degree of excited state distortion is experimentally related to the relative intensities of the two observed vibronic components.
IẳX
m
Eomh!m Eo
3
" # Smm
mexp4 ln2 vEoþmh!m vh
2
" #
ð1ị
The observation of luminescence requires that the radiative decay rate constant be within a few orders of magnitude of the generally larger nonradiative decay constant. Excellent discussions of the factors influencing excited state relaxation of transition metal complexes have been published recently by Meyer9,10, Endicott11, Gudel12 and others13,14 Radiative decay rate constants are a function of the magnitude of the transition dipole moment and the emission frequency, increasing with increasing oscillator strength of the transition (kr/|M|2) and increasing emission frequency (kr/3). Nonradiative relaxation rate constants are a function of a variety of factors including the emission energy (knr/exp(Eem); the Energy Gap Law), the degree of distortion of the excited state, the number and frequency of vibrational modes on the chromophore, and the degree of coupling with vibrational modesof the surrounding medium. A few examplesfrom the literature are given below.
In the absence of reactions between two excited states or environmental effects resulting in significant inhomogeneities in the distribution of chromophores, emissive decay from a single excited state (i.e., either S1or T1ofFigure 1) will follow first order kinetics and the observed rate E
S0 S1
T1
Wavelength 0-0′ 0′-0
0-1′
0-2′
0′-1 0′-2
2 1 0 2′
1′
0′
(a) (b)
kf + knf
kp+ knp I0
kisc
Figure 1 (a) Yablonski diagram showing electronic and vibrational levels as well as transitions between states. (b) S0S1absorption and emission from the S1electronic state structure.
constant will equal the sum of the radiative and nonradiative decay rate constants. The excited state lifetime in such cases is simply the inverse of the observed decay rate constant (Equation (2)):
ẵEx:Sttẳ ẵEx:St:0expðkobstịwhereðkobsị1ẳẳ ðkrỵknị1 ð2ị Emission quantum yields (em= fraction of excited states that emit) and lifetimes are related and experimental determination of both often allowsunambiguousdetermination of radiative and nonradiative rate constants. In the case of exclusive emission from the initially formed excited state (S1inFigure 1), the emission quantum yield and lifetime are related byEquation (3)
emẳkf=ðkfỵknfỵkiscị ẳkff ð3ị In this case the radiative decay rate constant is obtained from the ratio of the emission yield and lifetime. However, even in this simple case the observed nonradiative decay is the sum of the rate constants for nonradiative relaxation of the singlet state and intersystem crossing. If emission is observed exclusively from a state with a spin multiplicity differing from the ground state (i.e., T1inFigure 1), the observed emission quantum yield will reflect the efficiency for populating the emissive excited state (Equation (4)).
emẳ kisc krþknfþkisc
kp
kpþknp
ẳisckpp ð4ị
The ratio of the emission quantum yield and lifetime of the triplet state emission (phosphor- escence) yields the product of the intersystem crossing efficiency and radiative decay rate con- stant. Determination of intersystem crossing efficiencies is generally not straightforward and often techniques other than emission spectroscopy, such as time-resolved photoacoustic calorimetry,15 are used.
h
αI2/I1 λ αvh hν
I1
I2 v
500 600 700 800
Wavelength, nm E0
Intensity
S
Figure 2 Emission spectrum fit with Equation (1). The value of E0, the average of medium frequency vibrational modescontributing to nonradiative relaxation, h, the degree of excited state distortion, proportional to the Huang–Rhysfactor S, and the degree of inhomogeneousbroadening of vibronic
components, vh, can be obtained.
Electronic Emission Spectroscopy 317
Often more complicated excited state decay behavior is observed. For example, if the S1T1
energy gap issmall (Figure 1), back intersystem crossing may occur and the fluorescence (S1!S0) will consist of a prompt decay and a much longer-lived decay that is associated with the repopulation of the S1state from the T1 state. An excellent discussion of such cases is given in booksby Ferraudi6and Demas.16