Fluorescence spectroscopy has been used to measure fluorescence quantum efficiency (QE) of dried Bacillus spores (washed and unwashed) fixed to a quartz substrate. Fluorescence spectra and QE of anthracene in ethanol was used as the standard. We measured the absorption and fluorescence signal of the spores as a function of the number of spores. The absorption was measured from 600 nm to 250 nm using the reflectance in an integrating sphere. The fluorescence spectra were measured using excitation wavelengths at 280, 360 and 400 nm at room temperature. The absorption cross sections for the unwashed spores were 1.3 × 10-8, 8 × 10-9, and 5 × 10-9 mm2/spore at 280, 360 and 400 nm, respectively. The fluorescence QE was 0.13 ± 0.03, 0.33 ± 0.12 and 0.43 ± 0.26 at 280, 360, and 400 nm, respectively. The QE decreased by a factor of 2, 4 and 4 at these same wavelengths after washing and redrying the spores.
©2005 Optical Society of America
A variety of methods have been developed for the detection and identification of bacteria. These methods range from cell cultures to the physio-chemical techniques. Cultures, because the time required for the bacteria to grow, tend to be slow. Several physical methods have been tried, with varying degree of success. Examples of these techniques include, flow cytometry, chemoluminescence, fluorescence spectroscopy, and resonance Raman spectroscopy. The fluorescence of bacteria and bacterial spores has been considered for the rapid detection and identification of the bacteria [1–13]. We believe that these early studies have been encouraging and fluorescence spectroscopy has a strong potential to effectively detect and identify bacteria and bacterial spores in many different settings. Spores of various Bacillus species are formed in the process of sporulation, and these spores are adapted for long-term survival because they are resistant to many environmental stresses and are metabolically dormant [14–17].
The ability to detect or identify spores with fluorescence spectroscopy depends upon the number of spores and the fluorescence quantum efficiency (QE) of the spores. An ideal system would be designed to detect and identify a sample with the fewest numbers of spores. Thus, to determine this minimum level of detection, the fluorescence QE must be known.
The fluorescence QE is the ratio of photons absorbed to photons emitted through fluorescence. Using the Beer-Lambert law  one can express fluorescence intensity of an optically dilute, homogeneous medium (sum of the optical densities at the excitation and emission wavelength, everywhere is less than unity)  as a linear function of the concentration of the fluorophores in that medium. The number of quanta emitted per unit time is proportional to the number of quanta absorbed per unit time multiplied times the fluorescence QE. We can write this as:
where I0 is the intensity of incident light; D is the absorption of the solution, which is the product of the extinction coefficient of the molecule ε, the concentration of the molecules c, and the optical path length l(D = εcl), If concentration c is low and hence D is small, then the Eq. (1) can be approximated as
Although fluorescence spectroscopy of optically dilute, homogeneous media is well understood, fluorescence spectroscopy of biological samples is complicated by their highly absorbing and scattering properties .
The most reliable method for recording QE is the comparative method that involves the use of well-characterized standard samples with known QE values [19,21,22]. A simple ratio of the integrated fluorescence intensities/optical densities of the two samples will yield the ratio of the QE values. Since QE for the standard sample is known, it is trivial to calculate the QE for the test samples using the formula
where the term Slope is the slope of the plot of the integrated fluorescence at a selected excitation wavelength versus the absorption of the test sample as well as the standard sample, n1 and n2 are the refractive indexes of the sample and the standard sample respectively, and g is the spectral response correction for the fluorometer between the wavelength the standard is measured (λ1) and the wavelength the sample is measured (λ2).
Fluorescence cross sections of aerosolized B. globigii have been measured by Faris et al. in a flowing aerosol stream . Fluorescence cross sections of more Bacillus samples were measured by Stephens  on a cloud of particles suspended in an electrodynamic particle trap. The most recent report of the QE measurements of biological samples also appeared in the literature .
Our proposed system of spore collection on the sticky side of tape or with filters and then analysizing the fluorescence of the tape or filter [26–28] might lead to spore QE’s that differ from true aerosols where the particles are free to spin or rotate. The goal of this study is to measure the QE of Bacillus spores in our proposed data collection geometry.
If you want to identify spores from their fluorescence, you need to know the fluorescence QE to determine the minimum number of spores necessary to identify a sample. We would also like to see how the QE values changed with different preparations and washings of the spores. We can use these data to determine the limits of detection and the limits of identification for any system design.
2. Materials and methods
We used three preparations of spores in this work. We term these three preparations as simply (i), (ii), and (iii). The spores of B. globigii (i), B. globigii (ii), B. globigii (iii) used for this experiment were furnished as a generous gift from the US Army SBCCOM. The three different samples are different preparations of the spores made in different laboratories and different times. The use of multiple samples yields information on the amount of variability in the QE for B. globigii. None of the details of spore production methods were made available to us, except that none of the spores were treated with any special flow agent or anti-clumping compounds. All the spore samples were measured as they were provided and in a dry state. We have performed a series of fluorescence measurements of B. globigii samples in an unwashed and a washed and redried form. All of our fluorescence and absorption measurements were done at room temperature. The dried spores were fixed to a quartz substrate (Esco, New Jersey, USA). The quartz was cut about 0.9 cm wide and 3 cm high. The quartz with dry spores was masked with aluminium foil with a 2 mm hole to view the spores.
To determine the number of spores within the 2 mm hole, images of the spores were taken in situ using an Olympus system microscope model CX41 (Melbourne, Australia) and a digital camera (Coolpix, Nikon, Mehlville, NY, USA). The micrograph images were transferred to a computer. Using Photoshop (Adobe Systems, San Jose, CA USA), the number of pixels for one spore was determined. The spores appeared as dark specks on the image. So, the total number of dark pixels for the entire image was found. If the number of dark pixels on the entire image is divided by the dark pixel per spore, we obtain an estimation of the total number of spores used as a sample.
The quartz with spores was placed diagonally in the quartz cuvet and placed in the spectrofluorometer (SLM 8000C), so that the front face of the quartz slide (with the spores) was positioned at a 30° angle to the excitation optical axis and 60° to the emission optical axes as shown in Fig. 1. The quartz plate was not placed at 45° to decrease the amount of light directly reflected into the detection system. The illumination spot for this fluorometer is 1.7 ± 0.3 mm, determined from direct observation. The slits on the monochromators were set at a 4.0 nm bandwidth. The fluorescence spectrum was measured with excitation wavelengths of 280, 360 and 400 nm. The emission spectrum was measured from 10 nm longer than the excitation wavelength to 10 shorter than twice the excitation wavelength. When integrating the total fluorescence signal, the spectra were cut to exclude the light that was reflected off the quartz plate. All flurometer parameters, including the photomultiplier high voltage (1000 V), step size (1.0 nm) and the data collection time (0.1 s per point) was kept the same for all measurements.
This quartz and the particular glue used have a very small fluorescent signal that does not overlap very strongly with the spore fluorescence (data not shown). The glue was from touching the quartz with a piece of 3M Super Strength Mailing Tape (3M, St. Paul, Minn, USA). The fluorescence of at least 10 slides, each containing a different number of spores was measured.
Absorption of the dried samples was measured using a GBC UV/VIS Cintra 40 spectrophotometer (Victoria, Australia). The spectrophotometer was equipped with an integrating sphere to measure highly scattering or reflecting samples. The absorption was measured from 200 to 600 nm. The slits were set with a 2 nm bandwidth.
For the washed and redried spores, the samples of B. globigii were prepared by washing the spores in cold distilled water and then centrifuging at 10000 rpm for five minutes. The supernatant was removed and the pellet was kept. Samples were all washed twice, and the spores were dried on a quartz surface at room temperature for several days. In the washed redried condition, we were not able to count the number of spores, because we could not image the individual spores.
The fluorescence standard used for this work was anthracene (BDH Chemicals, Ltd., Poole, England). Anthracene has a molar extinction coefficient of 9,700 M-1cm-1 at 356.25 nm  and fluorescence QE of 0.27 in ethanol [30,31]. The anthracene absorption depends on the wavelength of exciting light. It is important that the absorption of the solution is not more than about 0.2 OD so that the fluorescence intensity is approximately uniform throughout the solution. We chose to use concentrations 0.9 to 27 μM anthracene. Solutions of anthracene were prepared by weighing dry anthracene powder in a measured volume of spectrograde ethanol (Scharlau Chemie S.A., Barcelona, Spain)
The absorption spectra of the standard sample anthracene in ethanol were measured with varying concentrations in a scanning range 200–600 nm using the spectrometer in a conventional method and a quartz 1 cm pathlength cuvette. Here also the slit was set with a 2 nm width. We measured the fluorescence of ethanol and fluorescence intensities of the prepared anthracene solutions with varying concentrations. We also measured the absorption of the ethanol and used this as a baseline that was subtracted from the absorption of the anthracene solutions. The index of refraction for ethanol was taken to be 1.36 (CRC Handbook of Chemistry and Physics, Cleveland, Ohio, USA).
The fluorescence spectrum was measured with excitation wavelengths of 320, 330, 340, 350 and 360 nm. The emission spectrum was measured from 10 nm longer than the excitation wavelength to 10 nm shorter than twice the excitation wavelength. The integrated fluorescence was measured as a function of the concentration. The analogue to digital converter in the fluorometer changed the photomultiplier output voltage to “counts” and we summed all the counts over the emission wavelength range. For the spore samples, we needed to decrease the emission range. The spore samples reflected a lot of light into the emission monochromator, therefore scattered light dominated the shortest wavelengths and the second order transmission of the scatter light through the monochromator dominated the longer wavelengths. We found that we could exclude most of the scattered light by using emission ranges of 300 to 550 nm for 280 nm excitation; 380 to 700 nm for 360 nm excitation; and 420 to 700 nm for 400 nm excitation.
3.1 Absorption and fluorescence of anthracene in ethanol
The absorption and fluorescence spectra of seven different concentrations of anthracene were measured. In Fig. 2(a) we show representative absorption and in Fig. 2(b) we show representative fluorescence spectrum of anthracene in ethanol. The fluorescence was excited at 360nm. The classical mirror image representation of absorption and fluorescence is clearly seen.
In Fig. 3 we plot the integrated fluorescence intensity versus the absorption for 320 nm and 360 nm. The points are shifted on the horizontal axis for these two wavelengths because the absorption of anthracene changes from 360 nm to 320 nm. The lines are linear least squares fits to the data point and are restricted to pass through the origin.
Similar plots were made with excitation wavelengths at 330, 340, and 350 nm (data not shown). Since the QE of anthracene is not wavelength dependent, the changes in slope of the lines is the system spectral response of the fluorometer. In Fig. 4 we show a plot of the slopes versus wavelength for each of the excitations. We extrapolate the system response to be constant beyond the range measured.
3.2 Quantum efficiency of B. globigii spores
We measured QE of unwashed and washed B. globigii spores from three different preparations: B. globigii (i), B. globigii (ii), B. globigii (iii); at excitation wavelengths 280, 360 and 400 nm. In Fig. 5(a) we show a representative absorption spectrum from a spore sample. Since we are measuring the absorption from the sample reflectivity, there is no scattering background that needs to be subtracted. However, the absorption of the spores is arbitrarily set to zero at 600 nm.
In Fig. 5(b) the fluorescence spectrum of unwashed spores of B. globigii (i) at excitation wavelengths 280, 360 and 400 nm are shown. We show the emission range that is used to integrate the fluorescence for these samples.
We plotted the absorption (O.D) at 280, 360 and 400 nm of B. globigii (i) versus the number of spores for each sample in Fig. 6(a). The straight line fits show a very good correlation between the absorption and the number of spores. The absorption cross sections for B. globigii (i) as obtained from the slope of the plots are 14.5E-9, 10E-9 and 4.2E-9 mm2/spore at 280, 360, and 400 nm, respectively. The absorption cross sections for all the samples are given in Table I.
The integrated fluorescence intensity (counts) of B. globigii (i) spores is shown versus absorption (O.D) in Fig. 6(b). The line is linear least squares fit fixed through the origin. The slopes are proportional to the QE of B. globigii (i) spores (unwashed) at the selected excitation wavelengths.
We also measured the absorption and fluorescence of B. globigii (i) after being washed and redried. In Fig. 7(a) we show representative absorption spectrum from the spores. The dashed line on Fig. 7(a) shows how the sample absorption would have been before washing. The absorption spectrum from Fig. 5(a) was arbitrarily scaled to be similar to Fig.7(a). This comparison suggests that the washing process primarily decreased the sample absorption below 300 nm.
In Fig. 7(b) we show fluorescence spectrum of washed redried spores of B. globigii (i) at excitation wavelengths 280, 360 and 400 nm. There are only small differences in the shape of the fluorescence peaks. However, the relative strength of the fluorescence at all the wavelengths is reduced after washing. In particular, the fluorescence from the 360 nm excitation is reduced relative to the fluorescence from the 400 nm excitation after washing.
In Fig. 8 the integrated fluorescence intensity (counts) of washed and redried spores of B. globigii (i) is shown versus absorption (O.D) at excitation wavelengths 280, 360 and 400 nm. The line is linear least squares fit. The slopes are proportional to the QE of B. globigii (i) spores at the selected excitation wavelengths.
For a comparison of the spore fluorescence before and after washing, the integrated fluorescence versus absorption plot at excitation wavelength 400 nm is given in Fig. 9 for B. globigii (i) before and after washing. Similar plots were obtained with other B. globigii spores (not shown).
In Table 1 we show the summary of the measurements of the Bacillus spores investigated. Results of both unwashed and washed samples measurements are shown. The absorption cross sections of unwashed samples were calculated from the slope of the plots of absorption versus the number of spores. The QE of the samples were calculated relative to the standard, the known QE value of anthracene in ethanol.
The standard solution of anthracene in ethanol produced well-defined absorption spectrum in the range of 300–400 nm excitations light. We were not able to measure any fluorescence from the ethanol in this range. At shorter wavelengths, we could observe the Raman scattering from the ethanol. By measuring the anthracene a several wavelengths, we could determine our system response. It would have been better if we could have measured the anthracene at wavelengths as short as 280 nm and as long as 400 nm, to correspond with the range of wavelengths used with the spores. However, the absorption of the anthracene is too small over this extended range and measurements with reasonable signal to noise are not possible.
The extrapolation of the system response beyond the range measured is perhaps the most questionable aspect of this study. As we show in Fig. 4, the extrapolation was a constant for flat spectral response beyond the range measured. We note that the fluorometer was designed for a relatively flat system response. To check that the extrapolation was reasonable, we measured the QE of tryptophan (Fisher Scientific, Fairlawn, NJ, USA) solutions at 280 nm. We found a value of 0.12, which is very close to the reported values of 0.13 , 0.18 , and 0.17  as reported in the literature. Clearly, we are justified in not following the straight line system response, which would have yielded a QE of 0.36 for tryptophan.
The spores, unlike the anthracene, contain many fluorophores. Therefore, we would not expect the QE to be wavelength independent. Excitation near 280 nm probably excites primarily the tryptophan in the spores. The excitation at longer wavelengths excites other flurophores, such as NADH or dipicolinic acid. The composition of the Bacillus globigii fluorescence spectrum has not be completely resolved . So, we cannot unambiguously assign a fluorophore to each wavelength measured. We measured over the 280 to 400 nm range because the fluorescence is significant over this range.
We can assume the QE is due to the tryptophan absorption with excitation near 280 nm. This assumption is supported by Saavedra et al. . In a structural study of the proteins of the Bacillus stearothermophilus based on intrinsic tryptophan fluorescence, they reported that the fluorescence QE were independent of the excitation wavelength in the range 280–310 nm. Overall, the fluorescence QE did not vary with the samples (i, ii, and iii) at 280 nm. The unwashed spores showed a QE similar to a tryptophan solution. This is probably from cell debris or broth on the exterior of the spore. Once this is washed away, the QE is reduced by 50%.
The fluorescence QE measured at 360 and 400 nm differs by up to a factor of 3 with the different samples (i, ii, or iii). Thus, at these longer wavelengths, the spore preparation is much more important. This is further evidenced by the large change in the fluorescence QE at these wavelengths after the spores are washed.
Although we expected changes in the QE of the spores to change upon being washed, we did not expect all the QE to decrease so dramatically. The decrease might be due to the spores clumping together after being washed. Before washing, most of the spores were isolated from one another or in small clumps of a few spores. This was evident from the microscopic images (not shown). After the washing, the spot of dried spores was so tightly clumped, that we could not count individual spores. We could still measure the absorption of the sample and determine the fluorescence QE.
Steven et al. reported that the total fluorescence for aggregates of bacteria increases with particle size . Yet, we see decreased fluorescence from the washed and clumped bacteria. We question if the spores are porous enough to change composition upon washing. We have reported earlier that the room temperature multi-wavelength autofluorescence signature of wet bacterial spores is considerably different from that of the dry spores . There are also small, but indistinct, differences are observed between the dry and the re-dried spores . Additionally, Marco et al.  reported that the dimension of individual B. globigii spores decrease reversibly by 12 % in response to a change in the environment from fully hydrated to air-dried state. Westphal et al.  reported on the kinetics of these changes with B. thuringiensis spores. These studies all support the model of reversible water migration between inner spore compartments and the environment. If this is true, we should expect no significant changes in the composition of a spore due to being washed.
More work need to be done to understand the environmental changes to the Bacillus spores fully. Although, the reader is cautioned against using these results to assume that bacterial spores cannot be identified with fluorescence spectroscopy. The identification of spores with fluorescence depends upon the stability of the fluorescence spectrum. The QE depends on the ratio of the fluorescence to absorption. Thus, the QE could be affected only by absorption changes.
We do note that our values for the fluorescence QE can be converted into fluorescence cross sections for the unwashed spores, where the absorption per spore was measured. We find the unwashed spores have an average fluorescence cross section of 5 × 10-15 cm2/nm-sr-spore with 280 nm excitation and Faris et al.  found a value of 3.5 × 10-15 cm2/nm-sr-spore with 270 nm excitation for dry spores. Although Sivaprakasam et al.  claim to be in agreement, on average, with Faris et al, we are not in good agreement with their particular values. Exciting at 266 nm, Sivaprakasam et al. find a fluorescence cross section of 1.3 × 10-13 cm2/nm-sr-spore, more than an order of magnitude larger than what we found for the cross section at 280 nm. At 355 nm, they find a fluorescence cross section of 1.9 × 10-15 cm2/nm-sr-spore, about a factor of four less than our value of 7.8 × 10-15 cm2/nm-sr-spore that we measured at 360 nm. The shorter 266 nm excitation wavelength might explain some of the difference between these values. It is also possible that the optical filters used by Sivaprakasam et al. were not flat, as we had to assume in making the comparison.
We measured the fluorescence quantum yields of the test samples of Bacillus spores in comparison with a standard solution of anthracene in ethanol. We reported here that the quantum yield of Bacillus spores are wavelength dependent and have relative values range from 0.10 – 0.06 for the mean values of the washed spores. It was particular interest to notice that the QE of Bacillus globigii at 400 nm (fluorescence in the visible region) possess the highest QE. Our results are comparable with values reported by Faris et al. and Sivaprakasam et al. [23,25]. Further study will be necessary to find why the QE is always decreased upon washing the spores.
We are grateful to NovaSol, Honolulu, Hawaii for funding this work. SS acknowledges the support of a University of Canterbury Doctoral Scholarship.
References and links
1 . J. T. Coburn , F. E. Lytle , and D. M. Huber , “ Identification of Bacterial Pathogens by Laser Excited Fluorescence ,” Anal. Chem. 57 , 1669 – 1673 ( 1985 ). [CrossRef]
2 . T. M. Rossi and I. M. Warner , “ Bacterial Identification using fluorescence spectroscopy ,” in Rapid Detection and Identification of Microorganisms, W. H. Nelson , ed. ( Verlag Chemie, 1985 ), pp. 1 – 50 .
3 . L. Reinisch , W. P. Van de Merwe , and B. V. Bronk , “ Comparative fluorescence spectra from bacteria and spores in different stages of growth ,” Biophys. J. 59 , 161a ( 1991 ).
4 . B. V. Bronk and L. Reinisch , “ Variability of Steady State Bacterial Fluorescence with Respect to Growth Conditions ,” Appl. Spectrosc. 47 , 436 – 440 ( 1993 ). [CrossRef]
5 . R. Manoharan , E. Ghiamati , S. Chadna , W. H. Nelson , and J. F. Sperry ” Effect of cultural conditions of deep UV resonance Raman spectra of bacteria ,” Appl. Spectrosc. 47 , 2145 – 2150 ( 1993 ). [CrossRef]
6 . J. A. Werkhaven , L. Reinisch , M. Sorrel , J. Tribble , and R. H. Ossoff , “ Non-Invasive Optical Diagnosis of Bacteria Causing Otitis Media ,” Laryngoscope 104 , 264 – 268 ( 1994 ). [CrossRef] [PubMed]
7 . M. J. Sorrel , J. Tribble , L. Reinisch , J. A. Werkhaven , and R. H. Ossoff , “ Bacteria Identification of Otitis Media with Fluorescence Spectroscopy. ,” Lasers Surg. Med. 14 , 155 – 163 ( 1994 ). [CrossRef]
8 . S. A. Glazier and H. H. Weetall , “ Autofluorescence detection of Escherichia coli on silver membrane filters ,” J. Microbiol. Methods 20 , 23 – 27 ( 1994 ). [CrossRef]
9 . F. S. Ligler , G. P. Anderson , P. T. Davidson , R. J. Foch , J. T. Ives , K. D. King , G. Page , D. A. Stenger , and J. P. Whelan , “ Remote sensing using an airborne biosensor ,” Environ. Sci. Technol. 32 , 2461 ( 1998 ). [CrossRef]
10 . G. A. Luoma , P. P. Cherrier , and L. A. Retfalvi , “ Real-time warning of biological-agent attacks with the Canadian Integrated Biochemical Agent Detection System II (CIBADS II) ,” Field Anal. Chem. Technol. 3 , 260 – 273 ( 1999 ). [CrossRef]
11 . B. C. Spector , L. Reinisch , D. Smith , and J. A. Werkhaven , “ Noninvasive fluorescent identification of bacteria causing acute otitis media in a chinchilla model. ,” Laryngoscope 110 , 1119 – 1123 ( 2000 ). [CrossRef] [PubMed]
12 . J. Ho , “ Future of biological aerosol detection ,” Anal. Chim. Acta 457 , 125 – 148 ( 2002 ). [CrossRef]
13 . E. Yacoub-George , L. Meixner , W. Scheithauer , A. Koppi , S. Drost , H. Wolf , C. Danapel , and K. A. Feller , “ Chemiluminescence multichannel immunosensor for biodetection ,” Anal. Chim. Acta 457 , S 3 – 12 ( 2002 ). [CrossRef]
14 . A. Driks and P. Setlow , “ Morphogenesis and properties of the bacterial spore ,” in Prokaryotic Development, Y. V. Brun and L. J. Shimkets , eds. ( American Society for Microbiology, Washington D.C, 1999 ), pp. 191 – 218 .
15 . M. Paidhungat and P. Setlow , “ Spore germination and outgrowth ,” in Bacillus subtilis and Its Relatives: from Genes to Cells, A. Hoch , R. Losick , and A. L. Shoneshein , eds. ( J. Am. Soc. Microbiol., Washington DC, 2002 ), pp. 537 – 548 .
16 . P. Setlow , “ Mechanisms which contribute to the long-term survival of spores of Bacillus species ,” J. Appl. Bacteriol. 76 , 49S – 60S ( 1994 ). [CrossRef]
17 . P. Setlow , in Bacterial Stress Responses, G. Storz and R. Hengee - Aronis , eds. ( Am. Soc. Micorbiol., Washington, DC, 2000 ), pp. 217 – 230 .
18 . I. D. Campbell and R. A. Dwek , Biological Spectroscopy ( Benjamin/Cummings Publishing Co, Menlo Park, CA, 1984 ).
19 . J. R. Lakowicz , Principles of Fluorescence Spectroscopy ( Plenum Press, New York, 1999 ).
20 . W. F. Cheong and A. J. Welch , “ A Review of the Optical Properties of Tissues ,” IEEE. J. Quantum Electron. 26 , 2166 – 2185 ( 1990 ). [CrossRef]
21 . S. Dhami , A. J. de Mello , G. Rumples , S. M. Bisshop , D. Philips , and A. Beeby , “ Phthalocymine fluorescence at high concentration: dimmers or reabsorption effect? ” Photochem. Photobiol. 61 , 341 – 346 ( 1995 ). [CrossRef]
22 . A. T. R. Williams , S. A. Winfield , and J. N. Miller , “ Relative fluorescence quantum yields using a computer controlled fluorescence spectrometer ,” Analyst 108 , 1067 ( 1983 ). [CrossRef]
23 . G. W. Faris , R. A. Copeland , K. Mortelmans , and B. V. Bronk “ Spectrally resolved absolute fluorescence cross sections for Bacillus spores ,” Appl. Opt. 36 , 959 – 967 ( 1997 ). [CrossRef]
24 . J. R. Stephens , “ Measurements of the ultraviolet fluorescence cross sections and spectra of Bacillus anthracis simulants ,” ( Los Alamos National Laboratory, Los Alamos, NM, 1999 ).
25 . V. Sivaprakasam , A. Huston , C. Scotto , and J. D. Eversole , “ Multiple UV wavelength excitation and fluorescence of bioaerosols ,” Opt. Express 12 , 4457 – 4466 ( 2004 ), http://www.opticsexpress.org/abstract.cfm?URI=OPEX-12-19-4457 [CrossRef] [PubMed]
27 . J. Kunnil , S. Sarasanandarajah , E. Chacko , B. Swartz , and L. Reinisch , “ Identification of Bacillus Spores Using Clustering of Principal Components of Fluorescence Data ,” ( 2005 ).
28 . S. Sarasanandarajah , J. Kunnil , E. Chacko , B. V. Bronk , and L. Reinisch , “ Reversible changes in fluorescence of bacterial endospores found in aerosols due to hydration/drying ,” J. Aerosol Sci. 36 , 689 – 699 ( 2005 ). [CrossRef]
29 . I. B. Berlman , Handbook of Fluorescence Spectra of Aromatic molecules, 2nd ed. ( Academic Press, New York and London, 1971 ).
30 . W. R. Dawson and M. W. Windsor , “ Fluorescence yields of aromatic compounds ,” J. Phys. Chem. 72 , 3251 – 3260 ( 1968 ). [CrossRef]
31 . D. F. Eaton , “ Reference materials for fluorescence measurement ,” Pure Appl. Chem. 60 , 1107 – 1114 ( 1988 ). [CrossRef]
32 . R. F. Chen , “ Fluorescence Quantum Yields of Tryptophan and Tyrosine ,” Anal. Lett. 1 , 35 – 42 ( 1967 ). [CrossRef]
33 . G. Weber and F. W. J. Teale , “ Determination of the absolute quantum yield of fluorescent solutions ,” Trans. Far. Trans. , 53 , 646 – 655 ( 1957 ). [CrossRef]
34 . C. Saavedra , C. Vasquez , and M. V. Encinas , “ Structural studies of the Bst V1 restriction-modification proteins by fluorescence spectroscopy ,” Eur. J. Biochem. 263 , 65 – 70 ( 1999 ). [CrossRef] [PubMed]
35 . P. Marco , J. L. Terrance , E. W. Katherine , and J. M. Alexander , “ The high-resolution architecture and structural dynamics of Bacillus spores ,” Biophys. J. 88 , 603 – 608 ( 2004 ).
36 . A. J. Westphal , P. B. Price , T. J. Leighton , and K. E. Wheeler , “ Kinetics of size changes of individual Bacillus thuringiensis spores in response to changes in relative humidity ,” PNAS 100 , 3461 – 3466 ( 2003 ). [CrossRef] [PubMed]