Raman spectroscopy is often hampered by high fluorescence backgrounds that can easily swamp the much weaker Raman signals. A common solution to avoid this is to use excitation wavelengths that do not excite fluorescence, for example, using near-infrared (NIR) sources that operate in a region where few molecular systems have electronic absorption bands. However, while this circumvents the fluorescence problem very effectively,1–3 it fails to take advantage of the unique features offered by the resonance Raman technique in terms of tremendous sensitivity and selectivity that are so beneficial for investigating complex heterogeneous samples such as those encountered in many biomedical applications. One of the most widely used approaches for obtaining Raman data from molecular systems that give large fluorescence backgrounds is shifted excitation Raman difference spectroscopy4–6 (SERDS), a technique that is popular for its technical simplicity. The key benefit of SERDS is that it can effectively eliminate the large fluorescence backgrounds as well as other sources of random or systematic noise generated by, for example, the detector where the different sensitivity of individual pixels of the detector can produce systematic effects.5 Application of SERDS has permitted great sensitivity down to true photon shot levels.4 In its basic form, the SERDS approach relies on obtaining two Raman spectra using excitation wavelengths shifted by an amount comparable to the bandwidth of the measured Raman bands (typically 5–10 cm21). A similar but less effective result can also be achieved by shifting the spectrograph spectral window by this amount, and this can be preferred because of its instrumental simplicity.5,6 Having obtained two Raman spectra, S9l and S9l2d, these are subtracted from each other to produce a difference spectrum, D 5 S9l 2 S9l2d. This step results in the elimination of the invariant fluorescence signal that for the small change in excitation wavelength is unchanged. The invariance is a direct result of the fact that the majority of fluorescence is emitted from vibrationally relaxed states (Kasha’s rule) and a small shift in the excitation wavelength does not perturb its spectral profile. In
[1]
Hiro-o Hamaguchi,et al.
Measurements of Spontaneous Raman Scattering with Nd:YAG 1064-nm Laser Light
,
1986
.
[2]
D. Chase,et al.
Fourier transform Raman spectroscopy
,
1986,
Journal of the American Chemical Society.
[3]
Andreas Hoffmann,et al.
Near-infrared Fourier transform Raman spectroscopy : facing absorption and background
,
1991
.
[4]
Richard A. Mathies,et al.
Effective Rejection of Fluorescence Interference in Raman Spectroscopy Using a Shifted Excitation Difference Technique
,
1992
.
[5]
S. Lieberman,et al.
Fluorescence Rejection in Raman Spectroscopy by Shifted-Spectra, Edge Detection, and FFT Filtering Techniques
,
1995
.
[6]
Steven E. J. Bell,et al.
Analysis of luminescent samples using subtracted shifted Raman spectroscopy
,
1998
.
[7]
P. Matousek,et al.
Tunable picosecond optical parametric generator-amplifier system for time resolved Raman spectroscopy
,
1998
.
[8]
Pavel Matousek,et al.
Efficient Rejection of Fluorescence from Raman Spectra Using Picosecond Kerr Gating
,
1999
.
[9]
Pavel Matousek,et al.
Fluorescence suppression in resonance Raman spectroscopy using a high-performance picosecond Kerr gate
,
2001
.
[10]
Fritz S. Allen,et al.
Automated Fluorescence Rejection Using Shifted Excitation Raman Difference Spectroscopy
,
2002
.
[11]
Pavel Matousek,et al.
Fluorescence background suppression in Raman spectroscopy using combined Kerr gated and shifted excitation Raman difference techniques
,
2002
.