Real-time measurement of dual-wavelength laser-induced fluorescence spectra of individual aerosol particles

We report the development of an in-situ aerosol detection system capable of rapidly measuring dual-wavelength laser-induced fluorescence spectra of single particles on the fly using a single spectrometer and a single 32-anode photomultiplier array. We demonstrate the capability of this system with both reference samples and outdoor air. We present spectra from separate excitation wavelengths from the same particle that demonstrate improved discrimination capability compared with only using one excitation wavelength. ©2008 Optical Society of America OCIS codes: (300.2530) Fluorescence, laser-induced; (010.1100) Aerosol Detection; (120.6200) Spectrometers and spectroscopic instrumentation; (170.6280) Spectroscopy, fluorescence and luminescence. References and links 1. V. Ramanathan, C. Chung, D. Kim, T. Bettge, L. Buja, J. T. Kiehl, W. M. Washington, Q. Fu, D. R. Sikka, and M. Wild, "Atmospheric brown clouds: Impacts on South Asian climate and hydrological cycle," Proc. Nat. Acad. Sciences (USA) 102, 5326-5333 (2005). 2. J. H. Seinfeld and S. N. Pandis, Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, 2 ed., (Wiley, New York: 2006), Chap. 24, pp. 1054-1091. 3. C. A. Pope and D. W. Dockery, "Health Effects of Fine Particulate Air Pollution: Lines that Connect," J. Air Waste Manage. Assoc. 56, 709-742 (2006). 4. D. W. Griffin, "Atmsospheric Movement of Microorganisms in Clouds of Desert Dust and Implications for Human Health," Clin. Microbiol. Rev. 20, 459-477 (2007). 5. C. A. Pope, "Respiratory Hospital Admissions Associated with PM10 Pollution in Utah, Salt Lake, and Cache Valleys," Arch Environ. Health 46, 90-97 (1991). 6. R. G. Pinnick, S. C. Hill, P. Nachman, J. D. Pendleton, G. L. Fernandez, M. W. Mayo, J. G. Bruno, "Fluorescence Particle Counter for Detecting Airborne Bacteria and other Biological Particles," Aerosol Sci. Technol. 23, 653-664 (1995). 7. P. P. Hairston, J. Ho, and F. R. Quant, “Design of an Instrument for Real-Time Detection of Bioaerosols using Simultaneous Measurement of Particle Aerodynamic Size and Intrinsic Fluorescence,” Aerosol Sci. Technol. 28, 471-482 (1997). 8. F. L. Reyes, T. H. Jeys, N. R. Newbury, C. A. Primmerman, G. S. Rowe, A. Sanches, "Bio-aerosol Fluorescence Sensor," Field Anal. Chem. Technol. 3, 240-248 (1999). 9. R. G. Pinnick, S. C. Hill, Y. L. Pan, R. K. Chang, "Fluorescence Spectra of Atmospheric Aerosol at Adelphi, Maryland, USA: Measurement and Classifcation of Single Particles Containing Organic Carbon," Atmos. Environ. 38, 1657-1672 (2004). 10. Y. L. Pan, R. G. Pinnick, S. C. Hill, J. M. Rosen, R. K. Chang, "Single-Particle Laser-Induced-Fluorescence Spectra of Biological and Other Organic-Carbon Aerosols in the Atmosphere: Measurements at New Haven, Connecticut, and Las Cruces, New Mexico," J. Geophys. Res. 112, D24S19 (2007). 11. S. C. Hill, R. G. Pinnick, S. Niles, Y. L. Pan, S. Holler, R. K. Chang, J. Bottiger, B. T. Chen, C. S. Orr, G. Feather, "Real-time Measurement of Fluorescence Spectra from Single Airborne Biological Particles," Field Anal. Chem. Technol. 3, 221–239 (1999). 12. V. Sivaprakasam, A. L. Huston, C. Scotto, J. D. Eversole, "Multiple UV Wavelength Excitation and Fluorescence of Bioaerosols," Opt. Express 12, 4457-4466 (2004). 13. P. Kaye, W. R. Stanley, E. Hirst, E. V. Foote, K. L. Baxter, and S. J. Barrington, "Single particle multichannel bio-aerosol fluorescence sensor," Opt. Express 13, 3583-3593 (2005). 14. J. R. Bottiger, P. J. Deluca, E. W. Stuebing, D. R. Vanreenen, "An Ink Jet Aerosol Generator," J. Aerosol Sci. t 29, S965-S966 (1998). #101416 $15.00 USD Received 18 Sep 2008; revised 26 Sep 2008; accepted 26 Sep 2008; published 1 Oct 2008 (C) 2008 OSA 13 October 2008 / Vol. 16, No. 21 / OPTICS EXPRESS 16523 15. Y. L. Pan, V. Boutou, J. R. Bottiger, S. S. Zhang, J. P. Wolf, R. K. Chang, "A Puff of Air Sorts Bioaerosols for Pathogen Identification," Aerosol Sci. Technol. 38, 598-602 (2004).


Introduction
Atmospheric aerosols have major effects on earth climate [1,2], atmospheric chemistry, and human health [3][4][5].The occurrence of airborne particles has been correlated with asthma, cardiopulmonary diseases, lung cancer and mortality [3,5].Some airborne bacteria and viruses transmit infectious diseases.Some of these have been used as biowarfare agents.There is a need for improved methods for rapid detection and characterization of aerosols, particularly organic carbon and biological aerosols, in both indoor and atmospheric environments.
Single-particle laser-induced fluorescence (LIF) can be used to differentiate bioaerosols from non-bioaerosols [6][7][8][9][10][11][12][13].Single-particle LIF spectra of atmospheric aerosols provide a basis for sorting particles into different clusters [9,10].Some materials which have almost identical LIF spectra using 266-nm excitation can have different LIF spectra with 355-nm excitation [11].However, until now a technique to measure two dispersed LIF spectra using dual-wavelength excitation from the same particle in a flow-through sampling system has not been demonstrated.One major challenge in building such a system is in preventing light that is scattered elastically from the particle from saturating and possibly damaging the detector.Previously, Sivaprakasam et al. [12] demonstrated a particle detector wherein rapidly gated photomultipliers (PMTs) detect LIF in three emission bands using two excitation wavelengths.Straightforward extension of that approach to measure dispersed spectra, however, might require two spectrometers, each with its own detector, where each detector has pixels which can be gated rapidly.Kaye et al. [13] developed a dual excitation wavelength single particle fluorescence detector using Xe lamps, but fluorescence was measured in only 2 bands.
We demonstrate an aerosol detection system which measures LIF spectra from the same particle using two excitation wavelengths.The approach uses a single spectrometer for spectral dispersion and a single 32-anode PMT for spectral detection.The elastic scattering is attenuated by using the imaging capability of the collection optics to focus the emission from flowing particles (moving at about 10 m/s) onto different regions of a filter when different lasers illuminate the particle.This system provides a means to detect bacterial aerosols and study atmospheric organic carbon aerosols with improved discrimination ability.

Experimental setup
The system is illustrated in Fig. 1, where an air-to-air particle concentrator (MSP M4220) is used to increase the particle sample rate.Ambient air is drawn into the concentrator at 300L/min, and the minority outlet flow (1 L/min; in which 2 to 10 μm diameter particles have been concentrated) is drawn into an optical chamber.As the aerosol particles enter the optical chamber they are focused and collimated into a particle stream of approximately 300 μm diameter by a sheath nozzle flow.Particles that flow through a trigger volume defined by the intersection of two diode lasers (wavelengths 650 nm and 685 nm; beam diameter of each is about 300 μm) scatter light at both wavelengths, which is detected by two photomultipliers with appropriate interference filters.A simultaneous scattering event triggers two successive UV lasers (fourth, then third harmonic Nd:YLF lasers; Photonics Industries) which emit 10 nsec pulses of light at 263 nm and 351 nm (26 μJ/pulse and 15 μJ/pulse).The pulses are separated by 12 μsec (the fastest time response of our data acquisition system (PhotoniQ from Vertilon)) so that the particle is illuminated by the two pulses at points separated vertically in the aerosol stream by about 120 μm.The scattering and fluorescence is collected by a Schwartzchild reflecting objective (Newport, 24x, 0.4 NA) and focused to the entrance slit of the spectrometer, where the light from the particle at the two illumination points in its trajectory is separated vertically by about 3 mm at the slit.A split filter (a combination of two glass long-pass filters with cutoff wavelengths at 295 nm and 380 nm to block the 263-nm and 351-nm scattered light, respectively) is positioned at the entrance slit so that the emission excited by each laser at different points along the trajectory of the particle is imaged to the corresponding filter to block the scattered laser light (see Fig. 1).The filters were chosen so that a small fraction of the scattered light leaks through each to provide a rough estimate of the particle size.The filters were cleaved and then polished with a succession of fine grit sandpapers and then mounted with these highly polished edges in contact.Because the filters are mounted in front of the entrance slit and have a finite thickness, a small portion of the light from each laser goes through the wrong filter, leading to higher leakage of the 351-nm scattered light.Fig. 1.Schematic of the detection system.Particles are drawn through the concentrator, and then through a nozzle that focuses the particles to a jet (approximately 300 μm in diameter).When two crossed diode lasers and detectors (not shown) detect a particle in the trigger volume, the two UV lasers fire in rapid succession.An image-preserving Schwartzchild objective focuses the scattering and emission from the particle to a split filter positioned to block the appropriate scattered light from the successive laser pulses at the input slit of the spectrograph.A 32-anode PMT records the two spectra dispersed by the spectrometer in rapid succession.
The light that enters the spectrometer is dispersed via a reflective grating onto a 32-anode photomultiplier tube (Hamamatsu H7260).The signal is read from the PMT via a PhotoniQ OEM high speed data acquisition system, and the spectra are recorded in sequence into a data file.The PhotoniQ OEM system is also capable of onboard real-time processing of the spectra.After data collection is complete, a computer program is used to confirm whether the two spectra are from the same particle by looking for scattering from the first excitation wavelength in a spectrum, and then the scattering from the other excitation wavelength in the following spectrum.If the two elastic scattering peaks are not in order the spectra are discarded.Due to nonlinearities which distort the spectral profile when the intensity is near saturation, spectra having intensities above 400 on the 0-600 scale of the photomultiplier were also discarded.
To test the system, aerosol particles were generated by an inkjet aerosol generator (IJAG) [14].The IJAG generates liquid droplets of approximately 50-μm diameter, which contain a solution or suspension of the material being aerosolized.The droplets pass through a heated drying column to form the residual dry aerosol.To sample ambient air (on the Yale University campus in New Haven, CT) the MSP aerosol concentrator was employed, and air was drawn from a first floor window.In our tests, the particle trigger rate never went above a few hundred Hz, and was usually in the range of a few 10's of Hz.At these count rates the probability of the two laser shots recording spectra from different particles is small.

Results
A series of E. coli (ATCC 11303) spectra recorded with the system is shown in Fig. 2. The particles were generated with the IJAG. Figure 2 illustrates the repeatability in the spectra that can be recorded, and provides confidence that measurements with atmospheric and other unknown samples can be reliable.Fig. 2. Fifteen sequential normalized single particle spectra of E. coli (thin solid lines) and the average spectrum (thick dotted lines) using (A) 263-nm excitation and (B) 351-nm excitation.Approximate particle size was 6 μm.
Among the reference samples we measured, two pairs of samples serve to illustrate how using two excitation wavelengths improves the discrimination ability of the system (Fig. 3).The spectra were normalized by reducing the area under the fluorescent portion (the scattering peak not included) to one, and then averaged to generate the spectra used in the graph.We are specifying that the tryptophan is "aged" because the bottle of tryptophan (Sigma) used is old, and the spectral peak has shifted to longer wavelength than it was when this tryptophan was measured previously (in 1999) [11].This peak shift is similar to one we have observed in the past when a tryptophan sample dissolved in water sits for a few days.However, for the spectra shown here, only a few hours elapsed between when the tryptophan was first added to water and when the spectra were measured.Using 263-nm excitation, the bovine albumin and aged tryptophan samples are indistinguishable (Fig. 3(A)).On the other hand, using 351-nm excitation (Fig. 3(B)), the aged tryptophan has a more pronounced fluorescence peak at 427nm, compared with the rather broad fluorescence of bovine albumin.Conversely, aged tryptophan and E. coli are indistinguishable using 351-nm excitation (Fig. 3(D)), but the two spectra have distinctly different peaks using 263-nm excitation (Fig. 3(C)).When used with the MSP concentrator, the system is able to measure ambient aerosol with acceptable sample rates quasi-continuously.An example of fifty consecutive particle spectra is shown in Fig. 4. Data collection was only stopped to clear out partial blockages in the inlet nozzle using compressed air and to change data files so that no one file became too big to easily process afterwards.Both of these procedures can be automated if a fully autonomous aerosol detector is desired.From the ambient data collected, we have selected two pairs of single particle spectra (Fig. 5) which also demonstrate the increased discrimination ability provided by using two excitation wavelengths.In Figs.5(A) and 5(B), the spectra from A1 and A2 are indistinguishable using 263-nm excitation, but have different peaks using 351-nm excitation.On the other hand, in Figs.5(C) and 5(D) the spectra from B1 and B2 are similar using 351-nm excitation but have markedly different peaks using 263-nm excitation.The composition of these ambient particles is unknown.However, in the future we would like to incorporate an air puffer [15] into the system which would allow us to collect specific particles for further analysis.

Conclusions
Here we have demonstrated our system's capability of measuring dual-wavelength excited fluorescence spectra of single flowing aerosol particles (using only one spectrometer and one 32-anode photomultiplier array), and have shown the advantage afforded by using both spectra.Some potential applications for a dual-wavelength fluorescence-based aerosol detector include: monitoring for particles containing bacteria, fungal spores, or pollen; bacterial detection in hazardous environments such as hospitals or livestock processing facilities; and studying carbonaceous particles in the atmosphere, the complex processing that occurs within such particles, and the effects of such particles on climate.

Fig. 3 .
Fig. 3. Dual-wavelength laser excitation spectra from laboratory samples: (A) and (B) show how bovine albumin and aged tryptophan are easier to discriminate using spectra from 351-nm excitation; (C) and (D) show how aged tryptophan and E. coli are easier to discriminate using spectra from 263-nm excitation.

Fig. 4 .
Fig. 4. Single particle fluorescence spectra of 50 consecutive aerosol particles sampled from ambient air.The second order of the grating starts at 527 nm for the 263-nm excitation spectra.

Fig. 5 .
Fig. 5. Normalized single particle fluorescence spectra from aerosol particles from ambient air (A) and (B) show two particles where 351-nm excitation offers improved discrimination (C) and (D) show another two particles where 263-nm excitation offers improved discrimination.The 527 nm peak in (A) is the 263-nm scattered light on the second order of the grating.