Many companies claim high sensitivity for their spectrofluorometers. In their specifications, instrument vendors often cite excitation of de-ionized water at 350 nm, which results in Raman emission at 397 nm (an energy loss of 3380 cm1). This Raman-emission band of water is used, because the signal is red-shifted and weak, thus mimicking weak fluorescence. De-ionized water is easy to prepare, making sample-to-sample reproducibility straightforward.
Other wavelengths, however, also should be considered, because a spectrofluorometer is used across a wide range of wavelengths. Optical design of less-sensitive instruments usually involves lenses and concave gratings, which lose focus at wavelengths other than the design wavelength, degrading instrument performance away from the optimum wavelength.
SPEX spectrofluorometers, however, are free from these artifacts, because their design uses only mirrors and plane gratings. Mirrors and plane gratings remain in focus at all wavelengths, maintaining high performance from the deep-UV to the near-IR. The Raman scatter of water becomes less intense as the excitation wavelength increases. Measurement of a series of Raman spectra across the instruments usable wavelength range is much more relevant as a figure of merit of instrument sensitivity, as the following experiment demonstrates.
Measurements were performed on a FLUOROLOG-3 spectrofluorometer with doublegrating monochromators, a 450-W xenon lamp, and a cooled R928 photomultiplier operated at 950 V in the photon-counting mode. The bandpass was set to 5.0 nm on both excitation and emission spectrometers. The same grating (330-nm blaze, optimized for 500 nm) was used for all scans. Only the integration time varied (see Table 1). The sample was deionized water puri