MEASUREMENTS OF SUBNANOSECOND FLUORESCENCE LIFETIMES WITH A CROSS‐CORRELATION PHASE FLUOROMETER *

The methods for the direct determination of fluorescence lifetimes may be discussed by considering each single fluorescent species as a detector in which the response to an infinitely narrow pulse of light decays in time, t, according to the exponential law I ( t ) = I(0) x exp (kt) . From the relations of the exciting light with the light emitted by such exponential detector, the rate constant k or its reciprocal T, the lifetime of the excited state may be determined. FIGURE 1 shows the Fourier spectrum characteristic of an exponential detector (cf. McLachlan, 1963). The detector responds without attenuation or change of phase to all frequencies from zero to those approaching k, while frequencies of this order or exceeding k are progressively more attenuated. The phase of the emitted light lags behind that of the excitation by 4.5" when o = k, and tends to 90" lag towards complete attenuation. It follows that, for the determination of k, the fluorescence must be excited by light of one or more frequencies in the neighborhood of k (hatched area of FIGURE 1 ) . The former is practically realized by excitation with sinusoidally modulated light, and the latter by excitation with light pulses the frequency spectrum of which is made up of a narrow band of frequencies. Classical electrodynamics shows that, for emission in the visible and ultraviolet regions of the spectrum, k is of the order of los sec-1 (or T of order 10-8 sec). Therefore, the frequencies to be employed in the excitation must be of the order of l O s / 2 ~ or in the region of 1-50 MHz. Light modulated at a single frequency in this range may be produced by various electro-optical methods, (Gaviola, 1927; Maerks, 1938; Bauer & Rozwadowski, 1959; Miiller et al., 1965), whereas techniques for the production of approximately Gaussian pulses of half-width 0.3 to 3 nsec (D'Alessio et al. 1964, Hundley et al. 1967) have been described. Each type of method has its own advantages and disadvantages. The pulse method is particularly useful when several modes of decay are possible, although the methods employing a single frequency are undoubtedly the more accurate because of the extreme frequency selection attainable by electronic techniques. They are also better adapted to the measurement of lifetimes in the subnanosecond region, and because we had this object in mind we have chosen to employ sinusoidally modulated light of fixed frequency in our measurements. The theory of the fluorometer, so named by Gaviola who constructed the first apparatus of this kind in 1926, has been given in detail by Dushinsky (1933). If a fluorescent species is illuminated with light modulated with frequency f, describable by the expression E(t) = A + B cos 27rft * This research was supported by United States Public Health Service Grant GM 11223.