The Measurement of Fluorescence Lifetimes

Once the fluorescence quantum yield has been determined, all that is required to calculate the fluorescence rate constant kf is the fluorescence lifetime Tf. Direct measurement of this quantity, like the measurement of the fluorescence quantum yield, is difficult, in this case because of the short lifetime of the fluorescent state (shorter than the normal flash from a flash lamp ). There are, however, several methods which have been developed to determine fluorescence lifetimes and these will be the subject of this section. 

The variable delay can be as simple as an RC network. Often the variable dday line is calibrated directly in terms of lifetime units (nanoseconds). When the reference and comparison signals are in phase the fluorescence lifetimes can simply be read off the calibrated variable delay. 

There are, however, problems associated with this method of determining fluorescence lifetimes. First, the phase method is not generally applicable for nonexponential signals and, as we shall see later, there are many cases where the observed fluorescence decay is indeed nonexponential. Second, the method 

The principal advantages of this technique are its very good time resolution, allowing the determination of lifetimes ranging from 10 to sec, and the fact that single photons are counted. Thus good results can be obtained even with very weakly fluorescent materials. 

We have now seen how fluorescence quantum yields and lifetimes are experimentally determined, along with some of the strengths and weaknesses of the methods used. For anthracene these constants have been determined to be 

---

At present, two main streams of techniques exist for the measurement of fluorescence lifetimes, time domain based methods, and frequency domain methods. In the frequency domain, the fluorescence lifetime is derived from the phase shift and demodulation of the fluorescent light with respect to the phase and the modulation depth of a modulated excitation source. Measurements in the time domain are generally performed by recording the fluorescence intensity decay after exciting the specimen with a short excitation pulse. 

The measurement of fluorescence lifetimes is an integral part of the anisotropy, energy transfer, and quenching experiment. Also, the fluorescence lifetime provides potentially useful information on the fluorophore environment and therefore provides useful information on membrane properties. An example is the investigation of lateral phase separations. Recently, interest in the fluorescence lifetime itself has increased due to the introduction of the lifetime distribution model as an alternative to the discrete multiexponential approach which has been prevalent in the past. 

Overview. The proposed near-term research is divided into three areas, which will be pursued simultaneously. The first is the complete characterization of the current mercury-responsive fluorescent chemosensor system, including the measurement of fluorescence lifetimes, to discern the origin of the conformational control of fluorescence. The second is to develop two classes of substituted biaryl acetylenic fluorescent chemosensors, to move the observed fluorescence into the visible region and increase the magnitude of the fluorescence signal, which occurs upon conformational restriction. 

A simplified instrument for the measurement of fluorescent lifetimes using the stroboscopic method has been described by Brown (67). The major virtue of this system is that it makes use of a Tektronix oscilloscope to obtain all the necessary trigger pulses, including a trigger of continuously variable delay. Since most laboratories are equipped with a good oscilloscope, the need to purchase expensive trigger-delay apparatus is thus eliminated. 

Fig. 8. Block diagram of a typical single-photon counting apparatus for the measurement of fluorescence lifetimes.

As a summary to this section dealing with the measurement of fluorescence lifetimes, Table 1 gives some examples of lifetime determinations for some diatomic molecules. The majority of the studies cited in Table 1 have been made since 1975. The quantity of data available, even just for diatomic molecules, is large and it is impossible to include all lifetime measurements in such a brief table. However, the examples in Table 1 do give some impression of the wide range of species to which the techniques discussed above can, and have, been applied. 

Over a substantial number of years the phase-shift or frequency-domain method has been employed for the measurement of fluorescence lifetimes. The technique requires the continuous excitation of a fluorescent sample with a source of varying intensity. The fluorescence response would normally be expected to increase and decrease to reflect the changes in excitation intensity. However, in a frequency-domain experiment the excitation beam is modulated at a high frequency, (o = 2nf, to produce a sinusoidally changing intensity given by ... 

R.D. Spencer, G. Weber, Influence of brownian rotations and energy transfer upon the measurement of fluorescence lifetime, J. Chem. Phys. 52 1654-1663 (1970)... 

Some information on the microviscosity can also be obtained by steady-state anisotropy measurements. A comparison of results for different media (e.g., for a series of mixed solvents differing in composition) is tricky and requires the measurement of fluorescence lifetimes. The steady-state anisotropy, (r), is a time-average weighted by the fluorescence intensity decay, I t) ... 

In our discussion of instrumentation foctors, we will stress their effects on excitation and emission spectra. However, similar concerns are inqx>rtant in the measurement of fluorescence lifetimes and anisotropies, which will be described in Chapters 4, 5, and 10. Additionally, the optical prc rties of the samples, such as optical density and turbidity, can also affect the spectra. Specific examples are given to clarify these effects and the means to avoid them. 

The idea to use frequency-domain detection for the measurement of fluorescent lifetimes dates back to 1921 [4], although the idea to measure small phase changes to determine short time intervals is much older [5]. The first instruments to measure non-spatially-resolved lifetimes in the 1920s were all based on frequency-domain detection. 

---