PublicationsUV Raman
Background on UV laser induced resonance Raman and resonance fluorescence
High levels of chemical specificity can be obtained using Raman spectroscopy without sample preparation, contact, or destruction Raman scattering is, in general, a very inefficient process. Normal Raman scatter cross-sections are about 10-26 cm² for a major Raman line (1615 cm-1) of a typical microorganism. Normal Raman occurs when the excitation wavelength is far from an electronic absorption band of the material. If the excitation wavelength is within a major electronic absorption band associated with the 1615 cm-1 Raman band, the scattering signal is “resonance” enhanced by as much as eight (8) orders or magnitude, such that the scatter cross-section improves to about 10-18 cm². In contrast, maximum resonance (native) fluorescence cross-sections for the same microorganism, measured over a 30nm wide bandwidth near the peak of fluorescence, are about 10-11 cm². This is a factor of 107 improvement over resonance Raman and clearly demonstrates the sensitivity of resonance fluorescence compared to resonance Raman. However, much higher levels of specificity can be obtained with Raman.
Resonance bands for nucleic and aromatic amino acids occur in the deep UV between about 220nm and 280nm. When excited at wavelengths less than 250nm, Raman scattering occurs within about 20nm to 30nm above the excitation wavelength, corresponding to about 4000 cm-1. Fluorescence occurs only above about 280nm, independent of excitation wavelength. Between the excitation wavelength at about 280nm, there exists a fluorescence-free region in which to observe the weak Raman scattering signal. A Raman shift of 4000 cm-1 corresponds to a wavelength of 247nm when excited at 225nm, 278nm when excited at 250nm and 298nm when excited at 266nm. It is therefore ideal to combine UV resonance fluorescence and resonance Raman spectroscopy to form an integrated tool for both detection and identification of biological agents since they offer a great combination of sensitivity and specificity that do not share overlapping observation wavebands.
Although resonance fluorescence is not the specific subject of this program, it is an integral part of the overall detection method and is closely tied to the excitation wavelengths used for resonance Raman excitation. Therefore we will include a brief discussion of resonance fluorescence here also.
UV Resonance Fluorescence Detection of Biological Agents
All microorganisms require continual input of free energy through cellular metabolism. The source of this energy input is electrochemical potential between electron donors and acceptors. The primary carrier of free energy is adenosine triphosphate (ATP), which is derived from the oxidation of fuel molecules such as carbohydrates and fatty acids. Typical molecules responsible for the transport of energy within cells are porphyrins, quinones, flavins, NADH, etc.. Other essential building blocks of living organisms are nucleic acids, amino acids and peptides, sugars and lipids, and polysaccharides. Most of these organic molecules contain chromophores which, when excited at an appropriate wavelength, will provide a signature of the material and give a good indication of the general class to which the microorganism or organic material belongs. Many other organic and inorganic materials also fluoresce that are not harmful to humans. However, when the excitation and emission wavebands are carefully chosen, these can be discriminated against with high reliability.
Optimum Fluorescence Excitation and Observation Wavelengths
Resonance fluorescence of biological agents is likely the only technique sufficiently sensitive to discover, in situ without any sample preparation, the presence and rough classification of a single or few numbers of microorganisms. It is the only viable method of performing non-contact biological classification of aerosols in situ because of the small dwell time for observation in an aerosol stream.
Figure 1 shows the molar absorptivity of the major aromatic amino acids. Note that the molar absorptivity peaks for Trp about 225nm and for Tyr about 230nm. Trp absorption at 225nm is 5 to 10 times stronger than in the traditional excitation wavelength at 280nm. The fluorescence cross-section and related efficiency is similarly higher when excited near their optima. It is a common notion that excitation at shorter wavelengths causes more interference with background materials. This is incorrect as will be shown below. The fluorescence cross-section and subsequent emission intensity is a function of both excitation and emission wavelength. This is illustrated in the following Excitation-Emission-Matrix (EEM) diagrams. EEM diagrams display the fluorescence intensity or cross-section as a function of both excitation and emission wavelength with the iso-intensity shown as contour lines, as illustrated below in Fig. 2 for Bacillus subtilis in both the vegetative and spore form.
It is important to note that both the spores the vegetative cells have two optimum excitation wavelengths, one near 230nm and one near 280nm. Emission maxima vegetative cells at both excitation wavelengths are the same, at about 330nm, as expected. The EEM diagram for Bacillus subtilis in spore form (@104 per ml) is also shown in Fig. 2. Optimum excitation wavelengths are essentially the same for B .subtilis spores and vegetative cell. However, the optimum emission wavelength for spores is close to 305nm compared to 340nm for vegetative cells. This is a clearly distinguishable marker feature of spores.
Cary, P.R. (1982) Biological applications of Raman and resonance Raman spectroscopies, Academic Press, New York.
Wilfred Nelson, U.Rhode Island, private communications.
Faris, G.W., R.A. Copeland, K. Mortelmans, and B.V.Bronk, “Spectrally resolved absolute fluorescence cross sections for bacillus spores”, App.Opt.,Vol.36, No.4, pp.958-967, 1 February 1997.
Thomas E. Creighton, Proteins, Structures and Molecular Properties, (W.H. Freeman and Company, New York, 1993)